©2021 The Author(s)

This is an Open Access book distributed under the terms of the Creative Commons
Attribution-Non Commercial-No Derivatives Licence (CC BY-NC-ND 4.0), which
permits copying and redistribution in the original format for non-commercial
purposes, provided the original work is properly cited.
(http://creativecommons.org/licenses/by-nc-nd/4.0/). This does not affect the rights
licensed or assigned from any third party in this book.

This title was made available Open Access through a

partnership with Knowledge Unlatched.

IWA Publishing would like to thank all of the libraries for

pledging to support the transition of this title to Open Access
through the 2020 KU Partner Package program.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Bioanalytical
Tools in Water
Quality Assessment
SECOND EDITION

Beate Escher, Peta Neale and Frederic Leusch

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Bioanalytical Tools in Water
Quality Assessment

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Bioanalytical Tools in Water
Quality Assessment
Second Edition

Beate Escher, Peta Neale and

Frederic Leusch

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Published by IWA Publishing
Republic – Export Building, 1st Floor
2 Clove Crescent
London E14 2BE, UK
Telephone: +44 (0)20 7654 5500
Fax: +44 (0)20 7654 5555
Email: [email protected]
Web: www.iwapublishing.com
First published 2021
© 2021 IWA Publishing

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as
permitted under the UK Copyright, Designs and Patents Act (1998), no part of this publication
may be reproduced, stored or transmitted in any form or by any means, without the prior
permission in writing of the publisher, or, in the case of photographic reproduction, in
accordance with the terms of licenses issued by the Copyright Licensing Agency in the UK, or
in accordance with the terms of licenses issued by the appropriate reproduction rights
organization outside the UK. Enquiries concerning reproduction outside the terms stated here
should be sent to IWA Publishing at the address printed above.

The publisher makes no representation, express or implied, with regard to the accuracy of the
information contained in this book and cannot accept any legal responsibility or liability for
errors or omissions that may be made.

Disclaimer
The information provided and the opinions given in this publication are not necessarily those of
IWA and should not be acted upon without independent consideration and professional advice.
IWA and the Editors and Authors will not accept responsibility for any loss or damage suffered by
any person acting or refraining from acting upon any material contained in this publication.

British Library Cataloguing in Publication Data

A CIP catalogue record for this book is available from the British Library

ISBN: 9781789061970 (paperback)

ISBN: 9781789061987 (eBook)

This eBook was made Open Access in June 2021.

© 2021 The Authors.

This is an Open Access eBook distributed under the terms of the Creative Commons Attribution
Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial
purposes with no derivatives, provided the original work is properly cited (https://
creativecommons.org/licenses/by-nc-nd/4.0/). This does not affect the rights licensed or
assigned from any third party in this book.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Contents

About the Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi
Acknowledgements (First Edition 2012) . . . . . . . . . . . . . . . . . xxiii
Acknowledgements (Second Edition 2021) . . . . . . . . . . . . . . . xxv

Chapter 1
Introduction to bioanalytical tools in water quality
assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Organic micropollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 Defining the issue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.2 Transformation products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.3 Low concentrations and mixtures . . . . . . . . . . . . . . . . . . . . . 6
1.3 Environmental toxicology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Environmental risk assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5 Bioanalytical tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.5.1 In vivo and in vitro bioassays . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.5.2 Cell-based bioassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.5.3 Modes of action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 vi Bioanalytical Tools in Water Quality Assessment

1.6 Bioassay selection and design of test batteries . . . . . . . . . . . . . . . 14

1.6.1 Design of test batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.6.2 Protection-goal-motivated test battery design . . . . . . . . . . 16
1.6.3 Chemical-group-motivated test battery design . . . . . . . . . 17
1.7 Chemical analysis and bioanalytical tools are complementary
monitoring tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.8 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Chapter 2
Risk assessment of chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.2 Current risk assessment of chemicals . . . . . . . . . . . . . . . . . . . . . . . 26
2.2.1 Hazard identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2.2 Effect assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.2.3 Exposure assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.2.4 Risk characterisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.2.5 Uncertainty analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.2.6 Risk management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.3 Application of bioanalytical tools in chemical risk
assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Chapter 3
Water quality assessment and whole effluent toxicity
testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Derivation of guideline values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.3 Human use of water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.3.1 Drinking water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.3.2 Recycled water, stormwater and managed aquifer
recharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.3.3 Dealing with unregulated chemicals in water . . . . . . . . . . . 40
3.4 Aquatic ecosystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.5 Comparison of environmental and drinking water
guideline values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.6 Whole effluent toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.6.1 Test systems in aquatic ecotoxicology commonly
applied to WET testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.6.2 In situ WET testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.6.3 Biomarkers in WET testing . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.6.4 ‘WET testing’ using bioanalytical tools . . . . . . . . . . . . . . . . 47
3.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Contents vii

Chapter 4
Modes of action and toxicity pathways . . . . . . . . . . . . . . . . . . . 51
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.2 Toxicokinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.2.1 Uptake, distribution and elimination . . . . . . . . . . . . . . . . . . . 52
4.2.2 Xenobiotic metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.2.3 Toxicokinetic indicators of chemical exposure . . . . . . . . . . 53
4.3 Toxicodynamic processes: toxicity pathways . . . . . . . . . . . . . . . . . 55
4.4 Mode of action classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.4.1 Non-specific toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.4.2 Specific modes of action . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.4.3 Reactive toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.5 Keeping the right balance: adaptive stress response
pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Chapter 5
Toxicity pathways of chemicals in humans . . . . . . . . . . . . . . . . 73
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2 Route of exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.3 Basal cytoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.4 Target organ toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.4.1 Hepatotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.4.2 Nephrotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.4.3 Cardiovascular toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.5 Non-organ-directed toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.5.1 Carcinogenicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.5.2 Developmental toxicology . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.6 System toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.6.1 Haematotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.6.2 Immunotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.6.3 Neurotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.6.4 Endocrine toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.6.5 Reproductive toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

Chapter 6
Adverse outcome pathways of chemicals in
aquatic organisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6.2 From the cellular level to the ecosystem . . . . . . . . . . . . . . . . . . . . . 92
6.3 Adverse outcome pathways for aquatic organisms . . . . . . . . . . . . 93

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 viii Bioanalytical Tools in Water Quality Assessment

6.3.1 Adverse outcome pathways for algae . . . . . . . . . . . . . . . . . 94

6.3.2 Adverse outcome pathways for invertebrates . . . . . . . . . . 95
6.3.3 Adverse outcome pathways for fish . . . . . . . . . . . . . . . . . . . 97
6.4 Using in vitro assays to understand toxicity pathways in
aquatic life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Chapter 7
Dose–response assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.2 Dose–response assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
7.2.1 Dose–response curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
7.2.2 Dose benchmark values . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
7.2.3 Continuum of toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.3 Concentration–response assessment . . . . . . . . . . . . . . . . . . . . . . 105
7.3.1 ‘Concentration’ versus ‘dose’ . . . . . . . . . . . . . . . . . . . . . . 105
7.3.2 ‘Response’ can mean toxicity or effect . . . . . . . . . . . . . . . 105
7.3.3 Concentration–response modelling . . . . . . . . . . . . . . . . . . 105
7.3.4 Concentration benchmark values . . . . . . . . . . . . . . . . . . . 107
7.3.5 Simultaneous effect and cytotoxicity in a
cell-based assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
7.3.6 Evaluating the linear portion of concentration–effect
curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
7.3.7 Antagonistic effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
7.4 Concentration-response curves of water samples . . . . . . . . . . . . 113
7.5 Bioanalytical equivalency concept . . . . . . . . . . . . . . . . . . . . . . . . . 115
7.5.1 Relative effect potency . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
7.5.2 Toxic units and toxic equivalent concentration . . . . . . . . . 116
7.5.3 Effect units and bioanalytical equivalent
concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Chapter 8
Mixtures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
8.2 Toxicity/effects of defined mixtures . . . . . . . . . . . . . . . . . . . . . . . . 121
8.2.1 Independent action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
8.2.2 Concentration or dose addition . . . . . . . . . . . . . . . . . . . . . 122
8.2.3 Synergistic and antagonistic effects . . . . . . . . . . . . . . . . . 123
8.2.4 Grouping of chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
8.2.5 Something from nothing? . . . . . . . . . . . . . . . . . . . . . . . . . . 126

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Contents ix

8.3 Assessment of concentration-additive effects using the

toxic equivalency concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
8.4 Mixtures in risk assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
8.4.1 Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
8.4.2 Do we have account for mixture effects in
risk assessment? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
8.4.3 Mixtures in chemicals regulations . . . . . . . . . . . . . . . . . . . 133
8.5 Mixtures and water quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
8.5.1 What type of mixture effects occur
in water samples? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
8.5.2 How much of the measured effects in water sample
can be explained by known and detected chemicals? . . 138
8.5.3 Mixture effects at very low effect levels (,10%) . . . . . . . 139
8.5.4 Component-based prediction of mixture toxicity
in water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
8.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

Chapter 9
In vitro assays for the risk assessment of chemicals . . . . . 143
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
9.2 Application of new approach methods in regulation . . . . . . . . . . 144
9.2.1 Alternatives to animal testing methods . . . . . . . . . . . . . . . 144
9.2.2 Integrated testing strategy in the
European Union . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
9.2.3 Toxicity testing in the 21st century (Tox21) strategy
in the United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
9.3 Application of in vitro assays in risk assessment . . . . . . . . . . . . . 148
9.3.1 A paradigm shift in human health risk assessment . . . . . 148
9.3.2 Quantitative adverse outcome pathways . . . . . . . . . . . . . 149
9.3.3 Quantitative in vitro to in vivo extrapolation . . . . . . . . . . . 150
9.3.4 Next-generation risk assessment . . . . . . . . . . . . . . . . . . . . 151
9.3.5 Applications of new approach methods for
environmental risk assessment . . . . . . . . . . . . . . . . . . . . . 154
9.4 Exposure in in vitro bioassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
9.4.1 Dose-metrics in cell assays . . . . . . . . . . . . . . . . . . . . . . . . 158
9.4.2 Serum-mediated passive dosing . . . . . . . . . . . . . . . . . . . . 161
9.4.3 Metabolism in cell-based bioassays . . . . . . . . . . . . . . . . . 163
9.5 Baseline toxicity and specificity of response . . . . . . . . . . . . . . . . . 163
9.6 Practical considerations for dosing of chemicals . . . . . . . . . . . . . 166
9.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 x Bioanalytical Tools in Water Quality Assessment

Chapter 10
Current bioanalytical tools for water quality
assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
10.2 Principles of cell-based bioassays . . . . . . . . . . . . . . . . . . . . . . . . 170
10.3 Bioassays indicative of xenobiotic metabolism . . . . . . . . . . . . . 173
10.3.1 Aryl hydrocarbon receptor . . . . . . . . . . . . . . . . . . . . . . . . 173
10.3.2 Peroxisome proliferator-activated receptor γ . . . . . . . . 175
10.3.3 Pregnane X receptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
10.4 Bioassays indicative of hormone receptor-mediated effects . . 179
10.4.1 Estrogen receptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
10.4.2 Androgen receptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
10.4.3 Glucocorticoid receptor . . . . . . . . . . . . . . . . . . . . . . . . . . 188
10.4.4 Progesterone receptor . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
10.4.5 Thyroid receptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
10.4.6 Mineralocorticoid receptor . . . . . . . . . . . . . . . . . . . . . . . . 195
10.4.7 Retinoic acid receptor and retinoid X receptor . . . . . . . 197
10.5 Bioassays indicative of other receptor-mediated
Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
10.5.1 Phytotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
10.5.2 Neurotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
10.5.3 Other assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
10.6 Bioassays indicative of reactive toxicity . . . . . . . . . . . . . . . . . . . . 202
10.6.1 Genotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
10.6.2 Mutagenicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
10.6.3 Non-genotoxic electrophilic mechanisms . . . . . . . . . . . 209
10.6.4 Oxidative stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
10.7 Bioassays indicative of adaptive stress responses . . . . . . . . . . 212
10.7.1 Oxidative stress response . . . . . . . . . . . . . . . . . . . . . . . 212
10.7.2 p53 response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
10.7.3 NF-κB response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
10.8 Bioassays indicative of apical effects . . . . . . . . . . . . . . . . . . . . . . 215
10.8.1 Cytotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
10.8.2 Algal growth inhibition . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
10.8.3 Fish embryo toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
10.9 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222

Chapter 11
Quality assurance and quality control (QA/QC) . . . . . . . . . . 225
11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
11.2 Method validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
11.2.1 Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Contents xi

11.2.2 Precision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

11.2.3 Robustness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
11.2.4 Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
11.2.5 Matrix interference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
11.2.6 Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
11.3 QA/QC in the laboratory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
11.3.1 Practical considerations . . . . . . . . . . . . . . . . . . . . . . . . . . 232
11.3.2 Replication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
11.3.3 Quality control samples . . . . . . . . . . . . . . . . . . . . . . . . . . 235
11.3.4 Control charts and fixed control criteria . . . . . . . . . . . . . 240
11.3.5 Standardisation and documentation . . . . . . . . . . . . . . . 242
11.3.6 Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
11.3.7 High-throughput screening . . . . . . . . . . . . . . . . . . . . . . . 243
11.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

Chapter 12
Sampling, sample preparation and dosing . . . . . . . . . . . . . . . 245
12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
12.2 Water sampling strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
12.3 Sample pre-treatment options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
12.3.1 Water sample preservation and storage . . . . . . . . . . . . 249
12.3.2 Water sample filtration . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
12.4 Extraction of water samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
12.4.1 Extraction versus testing the entire water sample . . . . 252
12.4.2 Solid-phase extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
12.4.3 Passive sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
12.4.4 Liquid–liquid extraction . . . . . . . . . . . . . . . . . . . . . . . . . . 254
12.4.5 Capturing volatile chemicals . . . . . . . . . . . . . . . . . . . . . . 255
12.5 Solid-phase extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
12.5.1 Solid-phase extraction sorbents . . . . . . . . . . . . . . . . . . . 255
12.5.2 Solid-phase extraction procedure . . . . . . . . . . . . . . . . . 256
12.5.3 Effect recovery by solid-phase extraction . . . . . . . . . . . 257
12.6 Sample collection and sample processing flow chart . . . . . . . . 260
12.7 Dosing into bioassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
12.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263

Chapter 13
Design of test batteries and interpretation of bioassay
results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
13.2 Test batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
13.2.1 Test battery design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 xii Bioanalytical Tools in Water Quality Assessment

13.2.2 Multiplex bioassays serving as test batteries . . . . . . . . 269

13.2.3 Routine test batteries for monitoring
applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
13.3 Linking bioassay results with chemical analysis: iceberg
modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
13.3.1 Iceberg modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
13.3.2 Effect-directed analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 275
13.4 Category 1 and category 2 bioassays . . . . . . . . . . . . . . . . . . . . . 278
13.5 Effect-based trigger values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
13.5.1 Approaches to derive effect-based trigger
values for category 1 bioassays . . . . . . . . . . . . . . . . . . 280
13.5.2 Approaches to derive effect-based trigger (EBT)
values for category 2 bioassays . . . . . . . . . . . . . . . . . . 286
13.5.3 Approaches to derive effect-based trigger (EBT)
values from read-across of in vivo data . . . . . . . . . . . . 291
13.6 What to do if a water extract exceeds the effect-based
trigger value? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
13.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296

Chapter 14
Case studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
14.2 Case Study 1: treatment of drinking water . . . . . . . . . . . . . . . . . 305
14.3 Case Study 2: quality of recycled water . . . . . . . . . . . . . . . . . . . 307
14.4 Case Study 3: wastewater treatment . . . . . . . . . . . . . . . . . . . . . . 311
14.5 Case Study 4: surface water impacted by wastewater
treatment plant effluent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
14.6 Case Study 5: benchmarking surface water quality
across the USA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
14.7 Case Study 6: benchmarking surface water quality in
small streams during rain events . . . . . . . . . . . . . . . . . . . . . . . . . 320
14.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324

Chapter 15
Application of bioanalytical tools beyond water:
Sediment and biota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
15.2 Suspended particulate matter, sediment and soil . . . . . . . . . . . 326
15.2.1 Suspended particulate matter . . . . . . . . . . . . . . . . . . . . 326
15.2.2 Sediments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
15.2.3 Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
15.3 Particles in air and dust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Contents xiii

15.4 Biota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334

15.4.1 Blood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
15.4.2 Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
15.5 Human biomonitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
15.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336

Chapter 16
A promising future for bioanalytical tools . . . . . . . . . . . . . . . . 339
16.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
16.2 Achievements so far . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
16.2.1 A sound guidance for selection of bioassays based
on the conceptual framework of toxicity pathways . . . 340
16.2.2 A more comprehensive measure of the realm of
organic pollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
16.2.3 Effect-based trigger values . . . . . . . . . . . . . . . . . . . . . . . 341
16.3 Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
16.3.1 Matrix effects and extraction methods . . . . . . . . . . . . . 342
16.3.2 Dosing into cell-based bioassays . . . . . . . . . . . . . . . . . 343
16.3.3 Linking bioanalysis with chemical analysis . . . . . . . . . 344
16.3.4 Linking bioanalysis with whole-animal testing . . . . . . . 345
16.3.5 Bioassays that require further development . . . . . . . . . 346
16.4 Future opportunities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
16.4.1 The ‘omics’ revolution . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
16.4.2 Three-dimensional cell models and organ- and
animal-on-a-chip systems to better model whole
organism response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
16.4.3 Moving from offline to online monitoring . . . . . . . . . . . . 348
16.4.4 Towards ultra-high-throughput testing, multiplex
assays and artificial intelligence-assisted
bioinformatics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
16.5 The road to regulatory acceptance . . . . . . . . . . . . . . . . . . . . . . . . 350
16.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 About the Authors

Beate Escher is internationally recognised for her work on chemical pollution in the
environment. She pioneered the field of water quality assessment by addressing
complex mixtures of chemical pollutants using in vitro bioassays. Beate Escher
obtained her PhD in 1995 and her Habilitation in 2002 at the Swiss Federal
Institute of Technology ETH, Zürich, Switzerland and is head of the Department
of Cell Toxicology at the Helmholtz Centre for Environmental Research in
Leipzig, Germany and professor at the Eberhard Karls University Tübingen,
Germany. She is also lecturer at ETHZ, Switzerland, holds an honorary
professorship at the University of Queensland and an adjunct professorship at
Griffith University, Australia. She was Associate Editor for Environmental
Science and Technology from 2012 to 2020 and is member of the Board of
Reviewing Editors at Science. In 2020, she was among the ‘Highly Cited
Researchers’. From 2011 to 2014 she held an Australian Research Council Future
Fellowship. In 2013 she won the Australian Water Association AWA National
Research Innovation Award for her work on cell-based bioassays in water
quality assessment.

Peta Neale is a research fellow at Griffith University and the central theme of her
research is to understand the fate and effect of emerging contaminants in the

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 xvi Bioanalytical Tools in Water Quality Assessment

aquatic environment and engineered systems. She completed her PhD at The
University of Edinburgh, United Kingdom, in 2009. She is currently an Associate
Editor for Water Research. She was named as Australia’s leading environmental
sciences researcher by The Australian’s 2018 research magazine. Her research
covers a range of areas relevant to environmental toxicology and environmental
chemistry and includes the application of bioanalytical tools to water, bioassay
validation, iceberg mixture modelling and chemical fate and partitioning.

Frederic Leusch is professor and Deputy Head (Research) in the School of

Environment and Science at Griffith University, Australia, where he teaches
biology and environmental toxicology. Fred also leads the Toxicology Research
Group (ARI-TOX) at the Australian Rivers Institute on the Gold Coast, which
focuses on assessing the impact of environmental contaminants on humans and
aquatic ecosystems. He completed his PhD at Lincoln University, New Zealand,
in 2005. His current research focuses on endocrine disruption, developing novel
bioassays for water quality assessment, validating ethical alternatives to animal
toxicity testing, and the application of systems biology methods to assess
exposure to environmental pollutants. He is Executive Editor for Environmental
Science and Technology and serves on various national and international
committees on issues related to trace organic pollutants and drinking and recycled
water quality. He currently chairs the Water Quality Advisory Committee of the
National Health and Medical Research Council (NHMRC), which provides
advice to the NHMRC on the on-going revisions of the Australian Drinking
Water Guidelines.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Preface

The last decades have seen an increase in research activities in the evaluation
of organic chemicals that may pollute the aquatic environment and our drinking
waters. While the majority of existing research has focused on identification
and quantification of individual chemicals by chemical analysis techniques,
effect-based methods have emerged in recent years to complement exposure-
based measures of chemical contamination that are obtained by chemical
analysis. These new effect-based methods include in vitro bioassays, and there
are an ever-increasing number of bioanalytical tools that hold great promise for
applications to water quality assessment.
The objective of this book is to summarise the scientific background underlying
the application of bioanalytical tools in water quality assessment for both a specialist
and non-specialist audience and to review the state-of-the-science. There is a focus
on drinking water, but other water sources such as surface waters (both freshwater
and marine), water from the urban water cycle (including wastewater and sewage),
industrial effluents and storm water that may be available for beneficial reuse
are also included, and we touch on applications of bioanalytical tools in other
areas of research and monitoring.
Chapter 1 gives a general overview of the field and provides some background
information on the type of chemicals that we are dealing with. The focus is on
organic chemicals, such as pesticides, pharmaceuticals, personal care products,

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_xvii

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 xviii Bioanalytical Tools in Water Quality Assessment

consumer products and industrial chemicals, as well as their transformation products

in the environment and in engineered systems.
Chapter 2 provides an introduction to risk assessment and international
regulations of chemicals.
Chapter 3 introduces standards and guideline values defined for various types of
water which are presented in a risk-based context. Applications of whole effluent
toxicity (WET) assessment, also termed direct toxicity assessment (DTA), are
also discussed.
The following chapters provide the scientific basis for bioanalytical tools.
Chapter 4 takes the reader to a cellular mechanistic level, introducing mode of
action classification and toxicity pathways that are crucial for the design and
application of bioanalytical tools. These cellular-level effects are the common
root for effects on human health and ecosystems. Chapter 5 summarises the
potential human health effects from chemical exposure that are triggered by
cellular-level effects and introduces related assessment endpoints. Chapter 6
expands the idea of toxicity pathways, introduced in Chapter 4, to adverse
outcome pathways that connect the dots to effects in environmental organisms,
populations and ecosystems.
Chapter 7 describes dose−response assessments, data reporting and derivation
of benchmark values. It also gives the mathematical background for calculating
toxic units (TU) and bioanalytical equivalent concentrations (BEQ). Chapter 8
provides an overview of mixture toxicity concepts, summarising the way
chemicals can interact as mixtures and delving in depth into the concept of toxic
equivalency, which is a means of reporting mixture toxicity using a simple metric.
Chapter 9 presents a brief overview of the application of high-throughput
in vitro bioassay testing for chemical risk assessment. Huge databases of in vitro
single chemical data have become available in the last 10 years and they serve
well for inspiration on the selection of bioassays for water quality testing but also
provides input data for the mixture models.
Chapter 10 provides a systematic review of bioassays currently applied for water
quality assessment, demonstrating the breadth and depth of the types of endpoints
that can now be accessed through effect-based monitoring.
The next chapters delve into practical aspects of bioassay application. Specific
QA/QC steps required to ensure that bioassay data are reliable, repeatable and
comparable across laboratories and demonstrate good assay practices are
discussed in Chapter 11, while considerations on sampling, sample preparation
and dosing are presented in Chapter 12.
Chapter 13 outlines how to develop a bioassay battery for water quality
testing, which assays to include, and how to interpret bioassay results,
introducing the concepts of iceberg mixture modelling and effect-based trigger
(EBT) values.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Preface xix

Chapter 14 illustrates the benefits of including bioanalytical tools for water

quality monitoring through selected case studies in surface water quality
assessment, wastewater, advanced water treatment and drinking water.
Chapter 15 briefly introduces some applications of bioanalytical tools beyond
water quality monitoring, providing some examples with sediment and biota
assessment as well as the potential of these tools for human biomonitoring.
The final chapter, Chapter 16, provides a synthesis and an outlook to future
developments in the field.
In addition, we have created a series of online resources and tools to apply some
of the principles and data methods explained in this book. This supplementary
information is available at www.ufz.de/bioanalytical-tools.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Foreword

The Global Water Research Coalition (GWRC) is an international organisation that

is dedicated to the exchange and generation of knowledge to support sustainable
development and management of the urban water cycle. The research agenda is
developed by the member organisations of the GWRC and reflects their priorities
and recognises global trends and drivers that affect the urban water cycle. The
agenda of the GWRC always includes the monitoring contaminants of emerging
concerns as one of the priority areas.
As we become more and more aware of the large number of pollutants
(particularly organic micropollutants) in the aquatic environment, it is no longer
possible to evaluate the elimination of the individual pollutants in water treatment
plants or to guarantee the absence of their transformation products, including
disinfection by-products. It is also difficult to evaluate which mixtures may
induce adverse health effects at a later date, given that very low concentrations
may already cause adverse effects, for example, endocrine disrupting effects.
While these low concentrations are unlikely to pose a significant health concern,
there is a scarcity of toxicity information on many of the chemicals currently in
commercial use, and in most cases, it is impossible to conduct a proper risk
assessment for all organic micropollutants.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 xxii Bioanalytical Tools in Water Quality Assessment

The possible health impacts of these substances are of major interest to water
operators and public consumers alike. As these concerns are widely debated
today, they require a scientific, objective and rigorous assessment of consumer
exposures and toxicity. There is an increased requirement to assess the level of
risks to human health under high-throughput, cost-effective and predictive
monitoring frameworks to better ensure that we limit our exposure to toxic
chemicals and avoid early biological effects.
In 2019, the GWRC commenced a multinational research project on ‘Effect
Based Monitoring for Water Safety Planning’. The project builds upon the
knowledge gained during the GWRC EDCI and EDCII Toolbox projects to
develop new practices that support the application of bioanalytical tools within an
internationally accepted water management framework, such as Water Safety
Plans. The main added value of this project is to combine substance-based and
effect-based monitoring tools to capture any adverse toxic chemicals missing
from current conventional substance-based targeting and demonstrate application
of this framework to assess the water quality profiles of different water from
resource to tap using key case studies.
One critical barrier to wider implementation of effect-based monitoring methods
is the lack of broader understanding of their usefulness in quantifying unknown
pollutants and their rich potential applications. This is a key area that the GWRC
project and this book aim to address. This book thoroughly lays the foundation
behind the science and carefully guides the reader through the concepts needed to
develop and successfully apply a battery of in vitro bioassays for enhanced water
quality assessment. The GWRC is thus pleased to endorse this book and hopes
that our joint effort and future reports will be useful to all who are active in the
field of understanding and venturing into ‘Effect Based Monitoring for Water
Safety Planning’.

Stéphanie Rinck-Pfeiffer
December 2020

(Managing Director GWRC)

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Acknowledgements
(First Edition 2012)

This book was made possible thanks to the generous financial support of the Urban
Water Security Research Alliance (UWSRA), a partnership between the
Queensland Government, CSIRO’s Water for a Healthy Country Flagship,
Griffith University and The University of Queensland, Australia. The Alliance
was formed to address South East Queensland’s emerging urban water issues
with a focus on water security and recycling. Particular thanks go to the Director
of the Alliance, Don Begbie, for being a great advocate of bioanalytical tools and
providing not only research opportunities but also his moral support and advice
throughout the review process of this book.
We are indebted to the project reference panel of our UWSRA project
‘Bioanalytical tools and risk communication’ lead by Greg Jackson, who
relentlessly motivated us and opened our eyes to the needs of the regulators and
practitioners. The reference panel is composed of Michael Bartkow, Kelly
Fielding, David Halliwell, Michael Lawrence, Richard Lim, Julia Playford,
Annalie Roux, Louis Tremblay, Heather Uwins and Christine Yeates. We are
especially grateful to Greg Jackson, Kelly Fielding, Annalie Roux and David
Halliwell for reviewing various draft chapters of this book.
We would especially like to thank our co-authors: Heather Chapman for her
significant contributions to Chapters 2 and 3, and Anita Poulsen for helping out

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 xxiv Bioanalytical Tools in Water Quality Assessment

with several chapters and scouring the literature for us. We are very grateful to have
had the both of you working with us on this book!
We also thank the team of researchers at our home and associated institutions,
namely A’edah Abu Bakar, Caroline Gaus, Eva Glenn, Marita Goodwin, Nat
Ling Jin, Matti Lang, Miroslava Macova, Erin Maylin, Ben Mewburn, Jochen
Mueller, Peta Neale, Janet Tang, Hanne Thoen and Wasa Wickramasinghe at the
National Research Centre for Environmental Toxicology (Entox), Julien
Reungoat, Maria-José Farré, Kristell LeCorré, Wolfgang Gernjak at the
Advanced Water Management Centre of the University of Queensland, and Erik
Prochazka, Vicky Ross and Phil Scott at the Smart Water Research Centre
(Griffith University) for the enthusiasm they bring to their work every day and
the impact they make on the development and application of bioanalytical tools
in a both a strict and wider sense, and who not only helped with critical reviews
of various draft chapters but also put up with sleep-deprived supervisors and
colleagues during the book writing process. You are the ones who make our
efforts worthwhile and we owe you!
The friends and colleagues who served as guinea pigs of our target audience (we
are afraid that in this particular case in vivo testing could not be replaced by in vitro
tools) are gratefully acknowledged: Rolf Altenburger, Janet Cumming, Meg Sedlak
and Michael Warne.
Paul Knittel helped us with the graphic design. Maggie Smith of IWA was a great
support during the publishing process.
There are many more people we want to thank as they have been important pieces
of the puzzle and have supported and inspired us even if they have not laid a hand on
the book itself. We won’t list any more names, you know who you are and it is great
to work and play with you.
Last but not least, we are grateful to our families for allowing us to vanish into
this book for so many weekends and giving us, from time to time, a reality check
that there is more to life than writing a book – although we had a lot of fun
climbing the treacherous learning curves of the book writing process. We are
looking forward to future editions as the field progresses.

Beate Escher
Frederic Leusch
August 2011

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Acknowledgements
(Second Edition 2021)

Ten years have passed. Our ‘little book’ proved to be much more useful than
anticipated, for teaching, to train new students and staff, but also as a resource for
this ever-growing field. What was initially a niche topic has exploded in the last
10 years: many barriers have been broken down and concepts have been steadily
improved. Effect-based trigger values, a mere concept a decade ago, have been
fine-tuned and now receive broader acceptance, although they are still only
slowly being taken up by regulatory bodies. The first edition of the book has also
been translated into Chinese, opening it up to a much wider audience than we
had initially hoped.
Nevertheless, we had not planned on writing a second edition yet. Then came
2020 – the COVID pandemic turned the world upside down and Beate was stuck
during lockdown in Brisbane. So close to Peta and Fred, and yet so far. What to
do? Well … write a book!
We wish to thank the Global Water Research Coalition and especially Stéphanie
Rinck-Pfeiffer for supporting this book by making material from the ongoing project
‘Effect Based Monitoring in Water Safety Planning’ available for this edition.
We would like to acknowledge the project team (Geertje Pronk, Stefan Kools, Milou
Dingemans, Jerome Enault, Jean-Francois Loret, Gaelle Meheut, Magali Dechesne,
Charlotte Arnal) as well as the project technical advisory committee (David

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 xxvi Bioanalytical Tools in Water Quality Assessment

Cunliffe, Dan Deere, Henner Hollert, Yves Levy, Alvine Mehinto, Leo Posthuma,
Shane Snyder, Suzanne van Drunick, Etienne Vermeirssen and Viviane Yargeau).
We thank many of our colleagues for their inspiration and collaboration during
the last 10 years – there are simply too many to name them all, but you know
who you are (!).
We thank the Zelltox team at UFZ for ‘testing’ the chapters and providing many
practical examples for the online resources and tools at www.ufz.
de/bioanalytical-tools. We are particularly indebted to Rita Schlichting, Luise
Henneberger, Maria König, Christin Kühnert and Lisa Glauch who were the
driving forces of many of the innovations in our CITEPro facility at UFZ where
we run in vitro bioassays now in a fairly high-throughput format. We also thank
the ARI-TOX team at Griffith University for their never-ending curiosity, forcing
us to constantly rethink our assumptions.
And, of course, we thank those who attended our many courses – starting in
2012 with a bioassay training course in California, becoming more formal within the
UFZ Graduate School ‘Higrade’ and other venues, including courses in Australia,
China, Germany and Switzerland – providing much useful input to the second edition.
A group of international students from Bachelor level to PhD level and some
fellow researchers not only tested the effectiveness of the teaching material
(slides and videos), questions and exercises on www.ufz.de/bioanalytical-tools
but also helped to improve these supplementary resources. We thank Maricor
Arlos, Kia Barrow, Victor Castaneda, Michelle Engelhard, Laura Fröhlich,
Andrea Gärtner, Fritz Kramer, Yuyuan Liu, Yue Meng, Lili Niu, Audrey
Ogefere, Pavel Sauer, Stephanie Spahr, Weiping Qin and Nelly Wang for their
input and enthusiasm during online classes. Sandy Schöne helped with setting up
the webpage www.ufz.de/bioanalytical-tool.
We thank Fabian Fischer for providing the photo of the cell lines on the
cover page – they are MCF7 cells stained with Hoechst (33342) and Alexa
(Flour 488) and photographed with a A Zeiss PALM CombiSystem enlargement
20× combining bright-field and fluorescence images.
The lake on the cover is the Haubitzer See, a former brown coal mine pit close to
Leipzig, Germany, with the coal-fired Lippendorf power station in the background
of the front cover and wind turbines on the back cover. From Lake Somerset dam in
Southeast Queensland, Australia, on the cover of the first edition to the rehabilitated
landscape of the German lake on the second edition – these pictures remind us of the
global importance of water but also the threats to our water resources by
human activities.
And last but not least, a huge thank you to our families, for putting up with our
extended periods of work on the book, including on Christmas Day! They are a deep
source of joy and a reminder that there is more to life than ‘The Book’.
Beate Escher
Peta Neale
Frederic Leusch
December 2020

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Chapter 1
Introduction to bioanalytical tools
in water quality assessment

1.1 BACKGROUND
Chemical monitoring provides a quantitative assessment of individual organic
contaminant concentrations in a water sample but does not account for the
presence of unknown compounds such as transformation products, untargeted
chemicals (i.e., not previously known to be present) or for interactions among
chemicals. Bioanalytical monitoring, also called effect-based monitoring (EBM),
is complementary to chemical analysis and provides information on all bioactive
micropollutants present in a sample ranked according to potency, that is, more
toxic chemicals are weighted higher than less toxic chemicals.
Classical aquatic toxicity tests used in water quality assessment include in vivo
assays with fish and aquatic invertebrates that measure endpoints such as
mortality, growth, reproduction, behavioural and feeding responses. In vitro
molecular and cell-based assays offer a sensitive, cost- and time-efficient ethical
alternative to classical whole animal testing. The implementation of human and
other living organism cell lines in water testing has facilitated high-throughput
evaluation of toxicological endpoints relevant for assessing the potential for
deleterious human and ecological health effects.
For the purpose of this book, we define ‘bioanalytical tools’ as cell-based in vitro
and in vivo bioassays that can be run in well-plate formats and that are indicative of
specific endpoints relevant for human and/or ecological health. These tools include
whole cell assays and assays with genetically modified cells, where natural features
have been over-expressed to enable more sensitive detection and/or where foreign

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0001

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 2 Bioanalytical Tools in Water Quality Assessment

features have been added for visualisation of effects. The cell membrane is an
important barrier and target site, and cell-free assays (such as immunoassays and
direct receptor binding assays) are generally excluded from this definition, with
the exception of some key enzyme assays. Assays with unicellular organisms,
such as algae, yeast or bacteria, and some high-throughput whole organisms such
as crustacea and fish embryo assays are also included in our definition of
‘bioanalytical tools’.
A major advantage of bioanalytical tools is the ability to detect the toxicity of
mixtures of known and unknown compounds, whereas chemical analysis can
only quantify the concentration of known, targeted chemicals irrespective of
toxicity. By measuring the mixture toxicity of a water sample, the bioassay
approach includes a risk perspective as it explicitly accounts for the differences in
toxicity across different chemicals and for interactions among chemicals in a
mixture. Many bioassays yield specific information on a given mode of action
rather than merely answering whether or not the cells are dead or alive after
exposure to the sample. This mechanistic information can be exploited by
running a series of bioassays indicative of a range of different modes of action in
parallel. In this way, a comprehensive bioanalytical test battery provides an
integrated measure of the toxicity of the biologically active substances in a water
sample. A bioassay can also be selected to target a specific protection goal such
as the maintenance of hormone balance or photosynthesis.
This book aims to provide a comprehensive understanding of the key concepts and
practical issues in the application of bioanalytical tools for water quality monitoring.
The focus is exclusively on organic chemicals. It is also possible to target metals
and inorganic pollutants with bioassays, however, sample treatment and data
interpretation differs between organics and inorganics. In addition, while there are
millions of organic chemicals, many of which may never be identified by chemical
analysis, the limited number of inorganic elements allows comprehensive chemical
analysis of metals and inorganics, reducing the need for effect-based analysis.

1.2 ORGANIC MICROPOLLUTANTS

1.2.1 Defining the issue
Organic micropollutants are a group of man-made chemicals such as pesticides,
industrial chemicals, consumer products and pharmaceuticals (Schwarzenbach
et al., 2006) (Table 1.1). As the name implies, micropollutants occur in
water and the environment in the microgram per litre concentration range
(1 μg/L = 10−6 g/L = 0.000001 g/L) or even lower, in the nanogram to
picogram range (1 ng/L = 0.000000001 g/L; 1 pg/L = 0.000000000001 g/L).
Not all water pollutants are man-made, natural compounds such as human
hormones and phytosterols can have adverse effects on aquatic life and natural

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 3

Table 1.1 Examples of ubiquitous organic water pollutants.

Origin//Usage Class Selected Examples

Industrial Solvents Tetrachloromethane

chemicals Intermediates Methyl-t-butylether
Petrochemicals BTEX (benzene, toluene,
ethylbenzene, xylene)
Plasticisers Phthalates
Lubricants Polychlorinated biphenyls (PCB)
Flame retardants Polybrominated diphenylethers
(PBDE), organophosphates
Consumer Detergents Nonylphenol ethoxylates
products Pharmaceuticals Antibiotics, painkillers
Hormones 17α-Ethinylestradiol
Personal care UV filters, hair dye, hydrotropes
products

Biocides Pesticides DDT, tributyltin, atrazine

Non-agricultural Triclosan
biocides
Disinfectants Bactericide, Isopropanol,
virucide alkyldimethyl-benzylammonium
chloride
Natural Taste and odour Geosmin, methylisoborneol
chemicals compounds
Natural toxins Microcystins, mycotoxins
Human hormones Estradiol, estrone, testosterone
Phytosterols Genistein, daidzein
Transformation Formed from all the Further details in Table 1.2
products above
Adapted from Schwarzenbach et al. (2006).

toxins such as those produced by plants, cyanobacteria or fungi may be extremely

harmful to aquatic life and human health.
In contrast to micropollutants, macropollutants are naturally occurring
compounds that exist locally in excess concentration, for example, phosphate and
nitrogen compounds, which can lead to eutrophication of surface waters
(Schwarzenbach et al., 2006). Macropollutants were the big environmental
problem of the 1960s and 1970s. Today macropollutants are usually properly
managed with the introduction of source controls and additional wastewater
treatment requirements. As a result, attention has shifted to micropollutants,
including inorganic and organic chemicals.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 4 Bioanalytical Tools in Water Quality Assessment

The widespread distribution of organic micropollutants in our waterways

presents a hazard to aquatic life. Hazards to humans can occur through the
consumption of food and drinking water and through other exposure routes such
as inhalation and dermal contact. Micropollutants enter the aquatic environment
via direct sources, such as industry and sewage effluent discharge, and via
non-point sources such as urban runoff and agriculture. Due to the complex
nature of the chemical mixtures occurring in domestic wastewater used for water
recycling schemes, conventional treatment is not always sufficient to remove
the entire contaminant load. Additional treatment steps such as ozonation and
sorption to activated carbon have been introduced to wastewater and recycled
water treatment to reduce more recalcitrant micropollutants. Membrane
processes such as reverse osmosis can also reduce a wide range of
micropollutants, but removal efficiency is related to the chemical structure and
size, and some compounds may not be fully removed by reverse osmosis alone.
Disinfection, such as chlorination and advanced oxidation processes, control
human pathogens (microorganisms that cause disease). While conventional
biological treatment and advanced treatment processes are very effective in
eliminating most unwanted pathogens and many micropollutants, they also
introduce other potentially harmful substances such as disinfection by-products
and transformation products.

1.2.2 Transformation products

Transformation products are micropollutants that have undergone chemical
reaction(s). It is currently unclear, which and how many transformation
products are formed, in what quantities and what level of harm they may cause.
Transformation products can arise from a variety of sources and can be formed in
the environment as well as in engineered systems (Table 1.2). Pharmaceuticals are
extensively metabolised in humans and animals and, hence, are typically not
excreted in the same form as they were ingested but as a variety of metabolites
(Lienert et al., 2007). Many pharmaceuticals are activated inside the body to the
pharmacologically active form, which may also be more potent than the precursor
with respect to its adverse effect. Most pesticides and other micropollutants
undergo biotic and abiotic transformation reactions in the environment. In surface
water, for example, exposure to sunlight can cause direct photodegradation or
indirect oxidation of micropollutants via formation of reactive oxygen species.
Biodegradation is extensive during biological wastewater treatment, yet full
mineralisation (complete degradation to carbon dioxide and water) is incomplete
for many chemicals, allowing biotransformation products to be formed.
Hydrophobic micropollutants are also removed from water by adsorption to the
sewage sludge without any transformation. Existing micropollutants in water can

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 5

Table 1.2 Transformation products of organic micropollutants and natural organic

matter that are more problematic than the original chemical (for more details see
Escher and Fenner, 2011).

Chemical Group Transformation Issues

Process

Transformation products introduced into the water

Pharmaceuticals Metabolism of Conjugates can later be cleaved
pharmaceuticals in back to the original compound
humans and animals Some pharmaceutical drugs are
followed by excretion more potent after metabolic
with urine and faeces activation
to wastewater
Transformation products formed in the environment
Pesticides Abiotic and biotic Pesticides (e.g.,
reactions in the organophosphates) can be
environment from activated by oxidation or other
direct runoff of chemical reactions
pesticides
All micropollutants Direct and indirect Can lead to a product more
photodegradation in persistent and toxic than the
surface water precursor (e.g., photochemical
condensation of triclosan
to a dioxin-like structure)
Transformation products formed in engineered systems
All micropollutants Biodegradation during Some transformation products are
wastewater treatment more persistent than the precursor
(e.g., 4-nonylphenol as breakdown
product of nonylphenol
polyethoxylate)
All micropollutants Advanced oxidation Disinfection and oxidation
and natural and disinfection during by-products (e.g., trihalomethanes
organic matter water treatment and haloacetic acids) from natural
organic matter are putative
carcinogens

be transformed during advanced oxidation and disinfection processes to more

persistent and/or toxic disinfection by-products (Table 1.2).
Disinfection by-products are important contaminants of drinking water and are
formed during disinfection from natural organic matter present even in the purest
water used as a source for drinking water (Table 1.2). During chlorination of
drinking water, a wide range of chlorinated chemicals are formed, for example,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 6 Bioanalytical Tools in Water Quality Assessment

trihalomethanes and haloacetic acids. Nitrosamines are further formed during

chloramination. Bromate can be formed after ozonation. As some disinfection
by-products are known to cause cancer and other adverse effects (Hrudey and
Fawell, 2015), their regulation and management is important while keeping in
mind that protection from pathogens is of the high priority to safeguard health in
the short term.
While most transformation products are less persistent, bioaccumulative and
toxic than the original compounds (Boxall et al., 2004), there are a number of
prominent exceptions. Some transformation products are more persistent than
their compounds of origin and thus accumulate in higher concentrations in the
environment. Other transformation products are more toxic than the original
chemicals (Escher and Fenner, 2011). An example is nonylphenol, which is a
degradation product of the industrial surfactant nonylphenol polyethoxylate
(NPE). Nonylphenol is highly persistent, bioaccumulative and in addition to
being more toxic than NPE in terms of acute toxicity, it exhibits weak estrogenic
effects (Fenner et al., 2002).

1.2.3 Low concentrations and mixtures

Regulators are faced with vast numbers of largely unknown micropollutants and
transformation products in water. Individual contaminants may be present at very
low concentrations, most far below any concentration expected to cause adverse
effects on their own but acting together in mixtures their biological activity may
lead to detectable effects.
All chemical analysis is limited to the lowest level of resolution of each analytical
method. In most analytical laboratories today, routine analysis is limited to the
microgram per litre (μg/L) range, while specialised methods may resolve
individual chemicals down to the nanogram (ng/L) or picogram per litre (pg/L)
level. Chemical analysis can only identify the tip of the iceberg with no
quantitative measure of the fraction that remains unaccounted for (Figure 1.1).
Although bioanalytical tools do not quantify the individual components in the
submerged part of the iceberg, they contribute a more complete picture of its total
size, thereby improving our ability to predict the possible health significance of
micropollutants. Bioassays indicative of specific modes of action, such as
estrogenicity and genotoxicity, may help refine this picture further by pointing to
specific groups of micropollutants with common modes of action, which often,
although not always, comprise structurally similar chemicals.

1.3 ENVIRONMENTAL TOXICOLOGY

Environmental toxicology has evolved over the last few decades from an
amalgamation of various scientific disciplines including biology, toxicology,
environmental chemistry, biochemistry, pharmacology, medicine and ecology.
The overall objective of environmental toxicology is to understand the impact of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 7

Figure 1.1 Organic micropollutants detected in water samples are only the tip of the
iceberg − many other micropollutants may be present, including transformation
products and known and unknown chemicals, which act together in mixtures and
these mixture effects can be captured by bioanalytical tools.

environmental pollutants on humans and ecosystems, encompassing all levels

of biological organisation. Effects range from biochemical interactions within
organisms to whole animals, populations and ecosystems. In human toxicology,
this sequence is paralleled starting from the toxicity pathways occurring at the
cellular level to failure or malfunction at organ level to population effects and
epidemiological studies (e.g., cancer clusters).
Traditionally, environmental toxicology has been divided into ecological
toxicology (or ecotoxicology) and human health toxicology. While the former
discipline was generally associated with environmental sciences and biology, the
latter is rooted in pharmacology and medicine. As these disciplines have become
more focused at the molecular level, it has been recognised that mechanistic
toxicity pathways have many common pathways and modes of action in all biota,
and the fields have again grown closer.
Environmental toxicology comprises the following sub-disciplines:
• Environmental science, an interdisciplinary science that studies the earth, air,
water, living environments and social components.
• Environmental chemistry and chemo-dynamics, the study of sources,
reactions and fate and transport of chemicals in the environment.
• Classical toxicology, which aims to protect human health.
• Epidemiology to understand effects on human populations and communities.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 8 Bioanalytical Tools in Water Quality Assessment

• Ecotoxicology (ecology + toxicology), which seeks to evaluate effects on

environmental organisms, populations, communities and ecosystems.
Aquatic toxicology is a subset of ecotoxicology, where exposure occurs via
aquatic ecosystems including saline, brackish and freshwater systems.

1.4 ENVIRONMENTAL RISK ASSESSMENT

Environmental toxicology, as a science, plays a central role in the development of
robust methods for environmental risk assessment of chemicals. Environmental
risk assessment consists of four steps (Figure 1.2). Hazard identification
includes the collection and evaluation of all available information for the given
chemical to assess its potential adverse effect and classification according to the
globally harmonized system of classification and labelling (GHS). The
assessment of the potential of a chemical to be categorised as persistent,
bioaccumulative and toxic (PBT) or carcinogenic, mutagenic or reproduction
toxic (CMR) is an important step in the European Union’s chemical regulation
to identify substances of very high concern. Effect assessment involves
dose−response characterisation and extrapolation, which yield predicted
‘derived no effect levels’ (DNEL) and ‘derived minimal effect levels’ (DMEL)
for humans and ‘predicted no effect concentrations’ (PNEC) for the
environment. The exposure assessment involves evaluation of the expected
exposure levels relevant for a given situation. For each exposure scenario, the
risk is then characterised by comparing the expected exposure level with the
DNEL/DMEL or PNEC, which should correspond to a safe dose over the entire
lifetime of a human or environmental organism.

Figure 1.2 Environmental risk assessment of chemicals, simplified from the REACH
Guidance Document for Chemical Safety Assessment (European Chemicals Agency,
2011).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 9

In vitro assays are used in an early screening stage of hazard assessment as part of
an integrated test strategy but if there is any indication that a chemical is of concern,
risk assessment in the European Union legislation REACH needs to be based on
in vivo information (EP&EC, 2006a). The U.S. risk assessment paradigm is
different. The U.S. EPA is committed to reduce animal testing for risk assessment
and to incorporate pathway-based in vitro toxicology in toxicity and risk
assessment (NRC, 2017).

1.5 BIOANALYTICAL TOOLS

Bioanalytical tools are defined as in vitro cell-based and in vivo bioassays indicative
of modes of action that are relevant for human and/or ecosystem health. These
include whole cell and reporter gene assays, tests with unicellular and small
organisms (Daphnia, fish embryo) as well as some enzyme and receptor-binding
assays. Previous reviews had wider or narrower definition. Behnisch et al. (2001)
included, for example, biomarkers and enzyme immunoassays, while Eggen and
Segner (2003) only included assays describing a defined chemical−biological
interaction excluding general cytotoxicity assays. The European technical report
on aquatic effect-based tools under the Water Framework Directive compiled
numerous in vitro and in vivo bioassays including biomarkers that have been
applied for water quality monitoring (Wernersson et al., 2015).

1.5.1 In vivo and in vitro bioassays

Toxicity testing can be performed at several levels of organisation. Epidemiological
studies attempt to link observed clusters of disease with human exposure to
chemicals. In vivo studies on individuals utilise historic case studies of human
poisoning or perform animal tests (e.g., using rodents) in order to obtain
toxicological information at the whole organism or organ level (Figure 1.3). As
with human toxicology, the in vivo scope of ecotoxicology may range across
organisms, populations, ecosystems and model ecosystems.
In vivo assays are whole organism exposure tests used to determine the toxicity of
a chemical, effluent or other mixture of interest to a target organism (Figure 1.4). In
addition to survival and reproduction, sub-lethal and behavioural effects can be
assessed. Sub-lethal effects are often quantified via biomarkers, which are
molecular characteristics that are objectively measured as indicators of a normal
biological process or in response to harm (Atkinson et al., 2001).
In vitro technology merges human toxicology and ecotoxicology. In vitro assays
are in a strict sense all assays that are performed in a controlled environment of a test
tube or microtitre plate (Figure 1.4). In practice, the term is often used
synonymously with ‘alternative test methods’ or ‘new approach methods’ (NAM)
that do not make use of test animals. While most in vitro assays are cell-based,
these also include isolated tissue (e.g., metabolically active liver homogenate) and
enzyme extracts. As cell lines (e.g., mammalian, fish, yeast and bacteria) can be

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 10 Bioanalytical Tools in Water Quality Assessment

Figure 1.3 Different levels of toxicological testing in human toxicology and

ecotoxicology and their agreed attributes.

obtained and grown without sacrificing test animals, molecular and cell-based
assays have the advantage of being of low ethical impact compared to in vivo
assays (Blaauboer, 2002; Hartung, 2010). Some mammalian cells cannot be
maintained in culture for a long time and have to be isolated from tissue (primary
cells), but other cell cultures, especially mammalian cancer cells and fish cells,
are immortal, that is, they can be cultured and reproduced indefinitely.
In vitro assays generally require less space (lower volumes) and are often more
practical for assessment of environmental samples with low levels of
micropollutants, which need to be enriched prior to toxicity testing. Cell-based
assays allow automation and high-throughput screening resulting in time- and
cost-effectiveness. Increased sensitivity can also be achieved through genetically
modified cell lines with amplified response (Figure 1.4).
Some in vivo assays share some of the advantages of in vitro assays. The fish
embryo test (FET), for instance, is a recommended alternative to traditional
ecotoxicological protocols. The FET is used in Germany to assess the quality of
wastewater before introduction into environmental waters (Embry et al., 2010).
In vivo biomarkers such as vitellogenin, a marker for estrogenicity, are also
sensitive and informative indicators of endocrine disruption (Purdom et al., 1994)
(Figure 1.4).
Yet, while in vivo bioassays are valuable for ecotoxicological assessment of pure
chemicals, applications for monitoring of water quality are generally limited to
whole effluent testing and low-complexity assays including those based on
biomarker responses (Figure 1.4). Reproductive and developmental effects are

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 11

Figure 1.4 Principles of in vivo and in vitro bioassays used in water quality
monitoring.

rarely assessed (with the exception of the FET). As in vivo bioassays are typically
performed with whole water samples, either directly or in diluted form, only
relatively polluted water can be tested using these methods. In contrast, water is
typically extracted and enriched before administration to in vitro bioassays, thus
allowing a much wider range of sample matrices (e.g., from wastewater to
drinking water) to be tested.
Additional very promising tools for hazard assessment of chemicals arise from
the emerging field of toxicogenomics. Toxicogenomics is the science of applying
genomic technologies to elucidate the toxicity pathways and modes of action
triggered by a micropollutant (Nuwaysir et al., 1999). Technologies applied in
toxicogenomics include profiling at the gene (transcriptomics) and protein
(proteomics) expression levels as well as profiling of the metabolic products
arising from biological reactions (metabolomics). Ecotoxicogenomics takes this
approach one step further by linking these cellular level effects with adverse
outcomes for whole organisms, populations and ecosystems (Ankley et al., 2006;
Fedorenkova et al., 2010). However, despite significant progress over the last
decade in pathway analysis and data curation (including the Comparative
Toxicogenomics Database), it is still difficult to link a specific gene, protein or
metabolic change to organism, population and ecosystem health outcomes
(Bahamonde et al., 2016), a process rendered even more difficult when applied to
mixtures of contaminants in environmental samples (Altenburger et al., 2012).
Thus, while omics profiling can offer novel insights during hazard identification,
toxicogenomic techniques are yet to be validated for use in regulatory risk

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 12 Bioanalytical Tools in Water Quality Assessment

assessment of chemicals or environmental monitoring programmes (Sauer et al.,

2017).
The use of in vitro bioassays in risk assessment has been limited due to
difficulty in determining their relevance to well-established in vivo toxicity
tests and predicting effects in whole organisms. Recent advances in molecular
toxicology and system biology, including those achieved by the Tox21
program of the National Institute of Health jointly with the United States
Environmental Protection Agency (U.S. EPA, Gibb, 2008), have led to a
paradigm shift (Hartung, 2010). In vitro bioassays are now gaining acceptance
provided an (ideally mechanistic) in vitro to in vivo extrapolation model exists
(Wetmore, 2015).

1.5.2 Cell-based bioassays

Cell-based bioassays target particular endpoints or mechanisms of toxicity and can
be divided into two groups:
• Bioassays with native cells (primary cells and immortal cell lines)
• Bioassays with recombinant cell lines

1.5.2.1 Native cells

Native cells are cells that have not been genetically modified. Primary cells can be
sourced directly from tissue samples but have limited life span in vitro. Immortal cell
lines are mutated cell lines that can proliferate indefinitely. Immortal cells are
preferable due to their high reproducibility and improved animal ethics and cost.
In mammals, only cancer and stem cells are immortal. More recently, methods
have become available to immortalise cells but so far, they have not been widely
used in practical applications for water quality assessment. In fish, any cell type
can theoretically be cultured and transformed from a primary to an immortal cell
line (Schirmer, 2006).
Native cells typically respond to all bioactive substances in a given sample and
are suitable for assessment of non-specific toxicity. Non-specific toxicity is typically
measured in bioassays that quantify cell growth/viability (cytotoxicity)
(Figure 1.5). Cytotoxicity assays can be more specific if cells are derived from
particular tissues such as pulmonary epithelial cells or liver cells. Growth of
neuronal cell lines can be used to assess not only cytotoxicity but also neurite
development as a more specific endpoint. The differential toxicity between
different cell types can further give an indication of the mode of action of the
chemicals in the sample. Some cells react specifically to groups of chemicals
with common modes of action by expressing a specific physiological response
such as direct inhibition of photosynthesis in algae or the proliferation of breast
cancer cells in the presence of estrogen.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 13

Figure 1.5 Design of cell-based bioassays. Receptor refers to a structural element in

the cell to which a chemical can bind and, subsequently, induce processes that may
cause toxicity. Reporters are gene products that are not naturally present in the cell
but have been introduced through genetic modification to allow visualisation of the
receptor binding.

1.5.2.2 Genetically modified cells

Recombinant cell bioassays use genetically modified cell lines and have emerged in
the last decade to detect and amplify specific toxic responses (Figure 1.5). Examples
include hormone-mimetic activity and induction of the arylhydrocarbon receptor
(AhR). The general design of recombinant cell bioassays is the integration of a
reporter plasmid into a cell (e.g., human or mammalian immortal cell line). A
plasmid is a circular DNA molecule, which carries a responsive element for the
receptor of interest, followed by a reporter gene that encodes a measurable feature
such as an enzyme (e.g., β-galactosidase or luciferase) or an easily measured
fluorescent protein. The amount of response quantified via the enzyme activity or
the fluorescence intensity of the fluorescent protein is proportional to the amount
of chemical bound to the receptor.

1.5.3 Modes of action

Modes of action (MOA) can be classified into three major groups: non-specific,
specific and reactive toxicity (see Chapter 4 for more details). Non-specific
toxicity refers to baseline toxicity and is the minimum toxicity that any
compound exerts without evoking specific effect. This minimum toxicity occurs

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 14 Bioanalytical Tools in Water Quality Assessment

at the same critical membrane concentrations irrespective of the cell type or

organism (Escher et al., 2019). Cytotoxicity assays will mainly be indicative of
non-specific toxicity. Specific toxicity refers to all mechanisms that involve
binding to a receptor or interference with an enzyme function. For reactive
toxicity, chemical reactions occur between the chemical and cell components,
such as DNA, proteins and phospholipids.
Bioassays can also be categorised in accordance with their potential to detect and
quantify the different MOAs within the above three classes. Non-specific toxicity
assays are crucial in providing an estimate of the overall toxic burden of all
chemicals within a mixture and include all cell viability/proliferation assays.
Specific toxicity assays target particular toxicant groups through detection of
specific endpoints. Typical bioassays applied for monitoring of specific toxicity
include recombinant cell bioassays capable of detecting the induction of nuclear
receptors, such as the estrogen, androgen, thyroid, aryl-hydrocarbon and retinoic
acid receptors. Bioanalytical capabilities are improving rapidly worldwide by
covering an increasing number of receptor-mediated toxicity endpoints. Reactive
toxicity includes any MOA that involves the chemical reaction between
chemicals and biological molecules, including DNA damage (genotoxicity,
mutagenicity), reactivity towards proteins, peptides and lipids, as well as
oxidative stress. The main focus has been on detection of mutagenicity and
genotoxicity using the classic Ames test (for mutagenicity) and assays indicative
of DNA repair. Genotoxicity tests based on mammalian cell lines with detection
of DNA damage using the Comet assay and/or the micronucleus assay have also
been introduced to water quality testing. Test battery-type approaches combine a
number of assays within and/or across the above categories enabling a more
comprehensive characterisation of various aspects of toxicity.

1.6 BIOASSAY SELECTION AND DESIGN OF TEST

BATTERIES
Cell-based assays have been applied for monitoring of water quality worldwide
since the 1960s (Figure 1.6). Early work mainly focused on assays indicative of
carcinogenicity (reactive toxicity) and general (non-specific) toxicity. The Ames
test (Ames et al., 1975), for example, has been used for water monitoring since
the 1970s (Simmon and Tardiff, 1976) and is still widely used. The Microtox
assay, which measures bioluminescence inhibition in the marine luminescent
bacteria Aliivibrio fischeri (formerly named Vibrio fischeri) as an indicator of
cytotoxicity, was first applied for water samples in the early 1980s (Chang et al.,
1981).
Researchers have applied single and multiple assays for water quality assessment
for decades, however, since testing was expanded from contaminated sites and
effluents to surface waters and highly treated waters, battery applications have
dramatically increased particularly in the last decade (Figure 1.6). Sanchez et al.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 15

Figure 1.6 Increase of studies that have applied bioanalytical tools in water quality
assessment since 1970. Search in Web of Science with the keywords ‘(in vitro or
vitro or bioanalytical) and battery and bioassay* and water and quality’ on 18
November 2020.

(1988) were among the first to employ an assay battery to evaluate the toxicity of
industrial effluents. The battery included five acute toxicity assays (three
bacterial, one in vivo and one molecular) and three mutagenicity assays (the
Ames test, and the Escherichia coli and Saccharomyces cerevisiae (yeast) reverse
mutagenicity assays).
Specific toxicity assays emerged in the 1990s to monitor endocrine disrupting
compounds (EDCs), which raised much concern due to the potential adverse
effects of these xenobiotics to wildlife. In particular, the observation of reduced
fish reproduction at environmentally relevant exposure concentrations sparked
much attention (Jobling et al., 1998).
The field really started to explode in the early 2000s (Figure 1.6). Mammalian
cell-based assays became more abundant in water quality testing as the focus
expanded from ecosystem health targeting surface water quality and wastewater
treatment to also include human health by considering advanced and drinking
water treatment and associated water quality (e.g., Brand et al., 2013; Leusch
et al., 2014a, 2014b; Hebert et al., 2018).
Most research has focused on surface waters and domestic and industrial
wastewaters. Scattered studies on pulp and paper mill effluents as well as oil
field-produced effluents are also found in the literature. In recent years, screening
of wastewater and advanced water treatment processes, disinfected drinking water
and recreational waters have emerged. Improved sample preparation and sample
enrichment methods as well as the introduction of more sensitive bioassay

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 16 Bioanalytical Tools in Water Quality Assessment

endpoints have enabled progression from highly contaminated water samples to

high-quality water such as purified recycled water and drinking water.
Apart from testing quality of different water types, an important application for
bioassays has been the assessment of treatment efficacy of a certain technical or
natural process, to evaluate trends in effect over time and to benchmark the
quality of water from different origins. Hence, the effects are typically compared
within a process, along a time axis or across different locations. This allows
calculation of treatment efficacies and can help to evaluate natural and engineered
treatment processes.

1.6.1 Design of test batteries

Comprehensive risk assessment requires a battery of bioassays to cover a range of
MOAs and/or recipients relevant for the water sample to be tested. Two distinct
approaches can be applied to design a test battery; one is driven by the protection
goal, while the other is driven by the chemical groups of concern and their modes
of action (Figure 1.7; see Chapter 13 for more details).

1.6.2 Protection-goal-motivated test battery design

A protection goal targets a health endpoint, organism or ecosystem process that it is
desired to be protected. A goal could be to minimise cancer occurrences in humans
or to ensure healthy fish reproduction in an aquatic ecosystem. Protection goals set
the context for all chemical risk assessment legislation and are often translated into

Figure 1.7 Design of test batteries for water quality assessment.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 17

specific assessment endpoints. Depending on the protection goal, a test battery

needs to include the relevant assessment endpoints. When selecting a battery of
bioassays, it is important to consider what we seek to protect (e.g., human health
versus aquatic ecosystem health, marine species versus freshwater species) as
well as the suite of tools necessary to conduct the assessment. The most
appropriate exposure route (and recipient tissue) must also be carefully evaluated.
If exposure to humans is via drinking water, for instance, the oral route is the
most important exposure pathway. If, on the other hand, exposure is to
recreational water via swimming, dermal contact is likely of higher significance
than ingestion. When the relevant organism(s), exposure route(s) and potential
risks posed to that organism(s) have been established, the relevant bioassays can
be selected. Assays relevant for inclusion will be specific to the protection goals
identified as part of hazard identification within a risk assessment framework.

1.6.3 Chemical-group-motivated test battery design

Chemicals with a common MOA and existing together in mixtures act according to
the concept of concentration addition (see Chapter 8 for more details). Chemicals
with a common MOA are also often (but not always) structurally similar.
MOA-specific bioassays can thus be used to identify relevant toxicant groups
present in a sample.
As all known and unknown chemicals with a common MOA will contribute to
the mixture toxicity in an associated bioassay, application of a MOA-based test
battery can help generate a more comprehensive picture of the toxic potency of a
water sample than chemical analysis alone. If the sample is suspected to contain
hormones (e.g., wastewater), it is sensible to include a test indicative of endocrine
disruption, particularly estrogenic and glucocorticoid activity. If herbicides are
likely to be present (e.g., agricultural runoff), a phytotoxicity assay will be suitable.
Effect-based batteries can be advanced further by including several assays for a
given toxic MOA. Estrogenic effects, for example, can be activated by direct
binding of the estrogen receptor (ER) by estrogenic compounds but also by
indirect mechanisms such as activation of the AhR by polyaromatic hydrocarbons
(PAHs) or inhibition of the cytochrome P450 19A aromatase that transforms
testosterone to 17β-estradiol.
Application of broad test batteries covering both non-specific cytotoxicity and
several specific endpoints allows the assessor to account for unexpected toxicant
groups that may otherwise go undetected. In the chemical-oriented design,
quantification of the risks posed by relevant groups of chemicals is prioritised.
Bioassays of high sensitivity towards the toxicant group of interest may therefore
be selected irrespective of their (lack of) direct relevance to the protection goal.
In order to assess drinking water for the presence of herbicides, for example, it
may be appropriate to include an algal assay, even if the water tested is destined
for human consumption and the protection goal is to achieve good human health.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 18 Bioanalytical Tools in Water Quality Assessment

Photosynthetic organisms are, however, particularly sensitive to herbicide exposure

and the test indicates exposure even in the absence of information on effects
in humans.
Both test battery approaches may lead to comparable and often overlapping sets
of bioanalytical tools as it is not possible to view chemicals independently of their
MOA. When researchers design test batteries, they will often consider both
approaches. It must further be noted that not all bioassays are fully selective and
100% indicative of a given MOA. All cell-based bioassays will be influenced by
a combination of non-specific and specific toxicity. In a water sample, there may
be a multitude of chemicals, of which only a fraction will respond specifically to
the endpoint featured in the applied assay. Within a range of concentrations, a
window typically exists where the specific effect sets in but is not yet suppressed
by cytotoxicity. The wider this window is, the more useful a given bioassay is for
application with complex water samples.

1.7 CHEMICAL ANALYSIS AND BIOANALYTICAL TOOLS

ARE COMPLEMENTARY MONITORING TOOLS
Bioanalytical tools do not replace chemical analytical monitoring. Both approaches
provide complementary information (Figure 1.8) and, if combined appropriately,
will allow for comprehensive assessment of organic micropollutants in a water
sample. What we can detect with chemical analysis is dependent on the sample
preparation, the chromatographic method (typically high-performance liquid
chromatography (HLPC) for waterborne pollutants but also gas chromatography
(GC)) and the detection method (mostly mass spectrometry (MS)). Precise
identification and quantification of chemicals in water samples can only be done
with target analysis using standards but it is possible to extend the list of target
analytes with suspect screening methods that apply a smaller range of standards
but can quantify larger sets of chemicals by a combination of retention times and
exact mass filtering (Krauss et al., 2010; Schymanski et al., 2014; Alyzakis et al.,
2018). Non-target analysis, that is, the identification of chemicals by
high-resolution MS, also shows enormous promises (Hollender et al., 2017) and
has already been applied to better understand pollution patterns in rivers
(Albergamo et al., 2019; Beckers et al., 2020). While there are clear promises of
non-target screening, it does not allow accurate quantification of micropollutants,
and does not provide any indication of the toxicity of the newly detected pollutants.
Typical monitoring programs routinely assess from 10 to 100 individual
chemicals, with more recent research publications including as much as 500
individual chemicals (Malaj et al., 2014; Bradley et al., 2017; Kandie et al., 2020).
Despite these amazing advances in analytical chemistry, it will never be possible
to quantify all chemicals. 175 million organic and inorganic chemical substances
were registered in the CAS (Chemical Abstract Services) Registry at the end of
2020. Several million of these chemicals are commercially available with more

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
Figure 1.8 The universe of organic micropollutants and the targets for chemical analysis (a) and bioanalytical tools (b). The spotlights
are turned onto a small fraction of all organic micropollutants present in a water sample. Adapted from Escher et al. (2020d).
Introduction to bioanalytical tools in water quality assessment
19
 20 Bioanalytical Tools in Water Quality Assessment

than 350,000 estimated to be in commercial use presently (Wang et al., 2020), not
counting transformation products. There are more than 1000 registered active
ingredients of pesticides in the USA alone, and up to 4000 for pharmaceutical
use. In the European Union .100,000 chemicals have been registered in REACH,
the European Union industrial chemicals’ regulation as of 2020.
Thus, only a small fraction of micropollutants potentially present in water can be
monitored by chemical methods. Whether or not the monitored chemicals are
relevant from a risk perspective, and cover the majority of the overall toxicant
burden, can only be ascertained via mixture toxicity assessment using bioassays.
As Figure 1.8a illustrates, target analysis can shed light only on a very small
fraction of chemicals. With suspect and non-target screening, many more
chemicals can be identified (albeit not quantified), but an unknown fraction still
lurks in the dark. With apical endpoints such as cytotoxicity, we can capture and
quantify the toxicity caused by all chemicals in a water sample acting together,
but we cannot identify the causative agents (Figure 1.8b). By applying bioassays
with specific modes of action, we can further narrow in groups of chemicals that
act according to the same mode of action, and better characterise the
toxicological profile. The results from the different methods can be connected in
a quantitative way through mixture modelling, which is outlined in Chapter 13.
The capabilities and limitations of chemical analysis and bioanalytical tools are
compared in Table 1.3.
Bioassays provide a comprehensive picture of biologically active chemicals
present in a sample. It is, however, not possible to elucidate which chemical(s) is
(are) responsible for the observed toxicity. To address this, a sample can be
fractionated, and the individual fractions tested for biological activity. The active
fractions may need to be fractionated further before the chemical(s) that caused
the effect can be identified. This so-called bioassay-directed fractionation
technique or effect directed analysis (EDA, Brack et al., 2008) is very promising
for water samples, where single contaminants dominate the overall toxicity such
as, for example, after an accidental release or illegal dumping of a chemical.
Generally, and for most applications in water quality monitoring, including
wastewater treatment, water recycling and drinking water treatment, there will be
no individual component(s) dominating the toxicity. The observed effect will
more likely reflect the combination of a large number of chemicals and their
transformation products, most likely present at individual concentrations below
the thresholds necessary to cause any observable single-chemical effect. Indeed,
when toxic chemicals act together in mixtures, concentrations below individual
effect thresholds may add up to measurable effects (Silva et al., 2002). Such
mixture effects cannot be accounted for using chemical analysis alone.
Bioassays that are selective for specific endpoints, such as binding to ER, will
respond to subgroups of chemicals that exhibit common MOAs and act together
in mixtures via concentration addition. Non-selective bioassays that detect
non-specific indicators, such as cytotoxicity or growth inhibition, are true sum

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 21

Table 1.3 Comparison of capabilities and limitations of chemical analysis and

bioanalytical tools.

Chemical Analysis Bioanalytical Tools

+ Capable of quantifying individual – Cannot resolve individual

chemicals (typically 10−500). compounds.
......................................................................................................................................
+ Selective for specific endpoints.
......................................................................................................................................
+ Sum parameter for chemicals with
the same MOA.
......................................................................................................................................
– Interactive mixture effects cannot + Detects mixtures of known and
be assessed. unknown bioactive chemicals
because all will respond to various
degree (weighted by potency).
......................................................................................................................................
+ Mixture toxicity measured but
interactive effects cannot be
differentiated from simple additive
effects.
......................................................................................................................................
– Full extent of chemical burden is + Non-specific endpoints account
unknown. for the sum of bioactive chemicals
present in a sample.
......................................................................................................................................
– In ‘clean’ water samples, + Bioassays are often less sensitive
individual components fall below than chemical analysis for
the limit of detection although individual chemicals, in complex
they may still contribute to mixtures their detection limit is
cumulative mixture toxicity. much better.
......................................................................................................................................
– Transformation products need to + Transformation products are
be identified before they can be accounted for by measurement of
quantified. mixture toxicity, but individual
contributions cannot be resolved.

parameters of the entire burden of micropollutants in a given water sample

(Figure 1.8). Effect-based sum parameters are weighted according to the toxic
potency of each individual mixture component and are superior to chemical sum
parameters, such as dissolved organic matter, where each component is weighted
according to its amount contribution to the mixture regardless of differing
individual toxicity.
Biodegradation and advanced oxidation processes will lead to substantial
formation of transformation products. Chemical analysis can only quantify

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 22 Bioanalytical Tools in Water Quality Assessment

transformation products that are known beforehand and/or are present in relatively
high concentrations. The identification of unknown transformation products is only
possible with highly sophisticated approaches and instrumentation (Kern et al.,
2009). As transformation products will contribute to the mixture toxicity, their
effect can be quantified with bioanalytical tools although their quantitative
contribution cannot be resolved.

1.8 APPLICATIONS
Bioanalytical tools have been applied widely to assess treatment efficiency of
technical processes, particularly primary and secondary wastewater treatment
(Prasse et al., 2015) (see Chapters 10 and 14 for more details). Advanced
treatment using flocculation and oxidation by ozone followed by biological
treatment has been shown to reduce specific toxicity (e.g., estrogenicity) to below
the limit of quantification, and to substantially reduce non-specific toxicity (Kim
et al., 2007; Tsuno et al., 2008; Escher et al., 2009; Stalter et al., 2011). In 2019,
46 comprehensive studies were available that applied bioassays to assess the
treatment efficacy of ozonation and activated carbon treatment (Volker et al.,
2019). Despite the observed reduction in toxicity, micropollutant concentrations
were still found to be sufficiently high to elicit distinct responses in some of the
selected bioassays. In this way, the bioassays enabled evaluation of the different
steps of the treatment process. In a comparison of ozonation, ultrafiltration and
reverse osmosis (RO), Cao et al. (2009) found RO to be the most efficient
technology for removing genotoxicity, mortality of the water flea Daphnia
magna, and effects caused by binding to the retinoic acid receptor (RAR). In
another study, the final steps of treatment following RO removed residual effects
even further, although many endpoints at this advanced stage of treatment fell
below detection limits (Escher et al., 2011).
The other main application of bioanalytical tools is to benchmark water quality.
This can be done by comparison between sites and time of sampling. In the last
years, effect-based trigger (EBTs) values have been developed that serve to
differentiate between acceptable and poor water quality (see Chapter 13 for more
details). Although EBTs have not yet been implemented in regulation, they are
widely used for research purposes and various methods have been developed for
their derivation. Most of the EBTs for drinking water have been developed by
read across from drinking water guideline values, in some cases using
toxicokinetic corrections. EBTs for surface water are also mainly read-across
methods from guideline values and environmental quality standards but several
methods specifically account for mixture effects.

1.9 CONCLUSION
Water samples contain an innumerable variety of contaminants from human
activities, such as pharmaceuticals, personal care products, industrial compounds,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Introduction to bioanalytical tools in water quality assessment 23

pesticides and others, as well as their many transformation products. It is thus

impossible to analyse all contaminants in a water sample using chemical analysis
techniques. Bioanalytical tools can detect both known and non-target chemicals,
and also provide a measure of the potency of these chemicals to interact with
biological targets, and ultimately produce an adverse effect in exposed organisms.
Bioanalytical tools allow us to view the whole iceberg of micropollutants, not
just the top part that we see through a conventional chemical analysis prism.
Chemicals can affect living organisms via a range of modes of action. Some
affect cause non-specific effects, which leads to cytotoxicity. Others will cause
specific toxic effects, such as binding to a receptor or interference with enzyme
function. Others yet can induce reactive toxicity, by interacting with cell
components such as DNA, proteins and phospholipids. It is therefore important
to consider and include multiple modes of action when designing a test battery.
A test battery design can be motivated either by a specific protection goal, or
focused on a specific mode of action, and there are a range of bioassays to
choose from, from native cells to genetically engineered platforms.
Yet bioanalytical tools also have their own limitations, including the fact that
they do not resolve individual compounds present in the sample. Bioanalytical
tools complement chemical analytical monitoring, greatly improving the overall
assessment of quality of any water type, including wastewater, surface, drinking
and reclaimed water.
The aim of this book is to provide the reader with an in-depth perspective on
the concepts and ideas in the application of bioanalytical tools to water quality
monitoring, including practical advice on sampling, analysis and interpretation of
effect-based monitoring. Ultimately, we want to enable the reader to apply
bioanalytical tools for water quality monitoring.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 2
Risk assessment of chemicals

2.1 INTRODUCTION
There exist more than 175 million chemicals and over 350,000 chemicals and their
mixtures have been registered in chemical inventories of 19 countries worldwide
(Wang et al., 2020). We know still too little about the risk they may pose to
humans and ecosystems.
Risk in the context of chemicals is the probability of an adverse effect on humans
or the environment occurring from exposure to chemicals. Risk assessment consists
of an objective evaluation of risk, in which assumptions and uncertainties are clearly
considered and presented. All activities, processes and products have some degree
of risk. The ultimate aim of chemical risk assessment is to provide the scientific,
social and practical information so that decisions can be made on the best way to
manage chemicals. The use of quantitative risk assessment in decision making is
becoming increasingly important as situations cannot be judged simply
binomially as ‘safe’ or ‘unsafe’. Risk assessment of chemicals is based on
scientific evidence, while risk management explores regulatory options by
weighing risk assessment with political and socio-economic factors.
The terms hazard and risk are frequently misunderstood and often incorrectly
used interchangeably. A hazard is a substance or event that has the potential to
cause harm. Risk is the probability or likelihood that this harm will occur. If
exposure is low or absent, then the risk is correspondingly low or absent,
irrespective of its potential to cause harm. In addition, if exposure is likely but
the effects are low or absent, the risk is low. The concentration of a chemical

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0025

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 26 Bioanalytical Tools in Water Quality Assessment

does not need to be nil necessarily, but it does need to be below a certain threshold
level of toxicity. This is based on the premise that at very low doses, below that
threshold, the chemical is considered to be safe over a lifetime of exposure.
Carcinogenic chemicals are treated differently in risk assessment and are called
‘non-threshold chemicals’ because it is assumed that there is no safe concentration.
Most countries have traditionally introduced separate legislations for risk
assessment of chemicals related to the environment – ecological (or
environmental) risk assessment (ERA) – and those related to human health –
human health risk assessment (HHRA). While terminology often differs, the
essential steps are the same in ERA and HHRA. The borders between ERA and
HHRA were broken down with the implementation of the European legislation,
REACH (Registration, Evaluation, Authorisation and Restriction of Chemical
substances), where a coherent assessment strategy applies for both fields
(EP&EC, 2006a), and also the Australian Guidelines for Environmental Health
Risk Assessment (enHealth, 2012). In keeping with this logical development, we
have attempted to integrate ERA and HHRA in the following overview. There
are differences, nevertheless. In HHRA we want to protect everyone, and
especially the most vulnerable members of the human population, unborn
children, mothers and the elderly from harm, while in ERA we want to protect
most species and the ecosystem in its structure and functioning, but not each and
every member of the ecosystem.

2.2 CURRENT RISK ASSESSMENT OF CHEMICALS

Risk assessment of chemicals encompasses the evaluation of impacts on human or
environment health arising from exposure to those chemicals. In the estimation of
risk, a number of steps are required, involving inputs from various disciplines.
Regulatory risk assessment of chemicals in most jurisdictions follows the
framework developed by the United States Environmental Protection Agency
(U.S. EPA, 1976) in the late 1970s (Figure 2.1). This process was reaffirmed by
the US National Research Council (NRC, 1983) and has since been adopted in
many national regulations among them the European Union (European Chemicals
Agency, 2011) and Australia (enHealth, 2012).
Hazard identification sets the scene and is followed by parallel effect and
exposure assessment (Figure 2.1). In the risk characterisation step, the probability
of exposure is evaluated against the severity of effect and conclusions are drawn
that inform risk management. The process is not linear, and includes feedback
loops that engage stakeholders, risk assessors, scientists, risk communicators and
communities (Figure 2.1).
The various published versions of the four-step framework for risk assessment
can be focused on environmental or on human health impacts, either from direct
exposure (e.g., food and water consumption) or indirect exposure (e.g., air toxins,
recreational exposure). Slight variations in the methods and terminology from this

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Risk assessment of chemicals 27

Figure 2.1 Generic risk assessment framework including feedback loops. Adapted
from enHealth (2012).

basic framework have developed in different documents (e.g., ‘hazard

identification’ is sometimes called ‘issue identification’, ‘effect assessment’ is
sometimes called ‘hazard assessment’ and ‘risk assessment’ is sometimes called
‘safety assessment’) but the principles remain similar.

2.2.1 Hazard identification

Hazard identification involves an issue identification step to define why a risk
assessment is required and the concerns that the assessment is to address. Then,
information on the inherent potential of chemicals to cause adverse effects is
collected, including physicochemical properties, information on modes or
mechanisms of toxic action, and human health and ecosystem effects.
A crucial outcome of hazard identification is classification and labelling of
chemicals and products. In terms of international trade, it is vital to implement an
internationally accepted way of labelling. The GHS – ‘Globally Harmonized
System of Classification and Labelling’ (United Nations, 2019) – is widely
accepted and has been implemented in many national legislations, for example in
2009 in the EU in the form of the Directive for Classification, Labelling and
Packaging as an important complement to the European Chemicals policy
REACH. In the GHS, criteria developed to assess the physical, health and
environmental hazard of chemicals or products lead to a GHS label in the form of
a pictogram (a stylised picture), a signal word and a hazard statement. A
pictogram is printed on the packaging and provides an indication of the type of
hazard the contents pose, for example, a dead fish with a dead tree indicates
‘dangerous for the environment’. Examples of signal words are ‘danger’ for
severe hazard categories or ‘warning’ for less severe hazard. A standard hazard

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 28 Bioanalytical Tools in Water Quality Assessment

statement would be ‘toxic in contact with skin’ (hazard phrase H311) or ‘harmful to
aquatic life’ (hazard phrase H402).
An important component of hazard identification is also to assess the inherent
properties of persistence (P), bioaccumulation (B) and toxicity (T) (EP&EC,
2006b). Chemicals with a half-life in water of .40 days (EP&EC, 2006b), .60
days (U.S. EPA, 1976; United Nations, 2009) or .180 days (Environment and
Climate Change Canada, 2020) are considered to be persistent. The main
criterion for bioaccumulation is that the aquatic bioconcentration factor (BCF) of
.2000 (EU) or .5000 (all other above-mentioned regulations, U.S. 1000–5000)
and ≥5000 for ‘very bioaccumulative’ in the EU. A no observed effect
concentration (NOEC) ,0.1 mg/L for aquatic toxicity or evidence of
carcinogenicity, mutagenicity or reproductive toxicity (CMR) classifies chemicals
as toxic in REACH.
If all three criteria are fulfilled, chemicals are considered PBT chemicals and a
full chemical safety assessment (EU term for risk assessment) must be conducted
(European Chemicals Agency, 2011). PBT and CMR chemicals are also added to
the candidate list of substances of very high concern (SVHC) in the EU. SVHCs
are intended to be phased out eventually, either by removing them from the
market entirely (restriction) or allowing them for specific uses only (authorisation).
PBT assessment is also essential for the international Stockholm Convention
(United Nations, 2009), which has the goal of protecting humans and the
environment from persistent organic pollutants (POP). In addition to being
persistent, bioaccumulative and toxic, a POP must also have long-range transport
potential, which means it can be found far away from its source. POPs such as
organochlorines have been found in the polar regions due to their combination of
longevity and physicochemical properties that makes them semi-volatile and
hydrophobic (Wania, 2003).

2.2.2 Effect assessment

Dose−response assessment is the term normally used in HHRA and characterises
safe levels of exposure to a variety of populations including children and the
elderly. To define a safe level, experimental (animal) toxicity data are collected
and extrapolated to a ‘derived no-effect level’ (DNEL) or ‘derived minimal effect
level’ (DMEL) for cancer. Ecological risk assessment on the other hand seeks to
protect a percentage (usually 95%) of species from adverse effects by deriving a
‘predicted no effect concentration’ (PNEC) from experimental ecotoxicity data on
selected species that are representative for the ecosystem.
DNELs are derived from the lowest effect level identified in a large set of
experimental acute and chronic animal toxicity studies (i.e., the lowest ‘no
observed adverse effect level’ (NOAEL) or ‘benchmark dose’ (BMD)) and
applying relevant uncertainty factors (also called extrapolation, safety or
assessment factors, Figure 2.2). The uncertainty factors can range from 10 to

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Risk assessment of chemicals 29

Figure 2.2 Derivation of no-effect levels (NEL) in HHRA and PNEC in ERA.

10,000 and account for extrapolation from animal studies to humans, from individual
people to a whole population, for differences in exposure duration, quality and
comprehensiveness of the database and for all other uncertainties related to
extrapolation from a model system to a human population (Ritter et al., 2007). In
HHRA, DNELs, also often referred to as ‘acceptable daily intake’ (ADI),
‘tolerable daily intake’ (TDI) or ‘reference dose’ (RfD), are defined as intakes of
chemicals in mg per kg body weight over a set window of time (usually a day, but
sometimes a week in the case of tolerable weekly intake (TWI)) that do not pose
an appreciable risk over the lifetime of a person (typically 70 years).
For the so-called ‘non-threshold effects’ (e.g., caused by carcinogens), the effect
expected is based on the assumption that the risk is proportional to the dose at all
low-dose levels. In other words, the sum of a number of small exposures has the
same effect as one larger exposure. While the threshold model assumes very
minor exposures are likely to have a negligible effect, it is believed that there
is no safe level for carcinogens. A linear extrapolation from the BMD10
indicative of 10% tumour incidences to zero is therefore used to calculate a
‘cancer slope factor’ (CSF) for non-threshold effects, such as chemically induced
carcinogenesis. Cancer risk is usually reported as an additional number of people
affected out of a million people per year, and ‘acceptable risk’ is determined by
regulation as a likelihood of a deleterious health outcome of 10−6 (one in a
million) (Australia, Europe for consumers) or 10−5 (one in hundred thousand)
(WHO, U.S., Europe for workers).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 30 Bioanalytical Tools in Water Quality Assessment

In ERA, the PNEC is commonly used as an effect measure. There are different
approaches to calculating the PNEC, depending on the availability of toxicity
data. When sufficient toxicity data are available (i.e., toxicity data of preferably
more than 15 species from different taxonomic groups), ‘species sensitivity
distributions’ (SSD) can be used to derive the concentration that protects 95% of
the species. However, if only acute toxicity data (e.g., effect concentration (EC)
such as lethal concentration for 50% of the test species LC50) are available, then
usually the lowest EC value from a minimum of three acute toxicity tests at
different trophic levels (typically algae, water flea and fish) is used to extrapolate
the PNEC with an uncertainty factor of 1000 to account for acute to chronic
extrapolation, differences in species sensitivity, lab to field and single organism
to ecosystem extrapolation, as necessary.

2.2.3 Exposure assessment

Exposure assessment determines the magnitude, frequency, character, extent and
duration of exposure to a hazard for an exposed population. An initial
requirement for exposure assessment is an understanding of the presence (or
absence) of a chemical and its concentration and distribution in different
environmental compartments (air, water, soil, sediment). In the absence of actual
exposure data, mathematical models can be used to predict exposure. In these
models the emissions are typically estimated from production volumes and
knowledge on application and use of the chemicals, and multimedia fate models
are used to evaluate the partitioning among different environmental compartments
and the degradation processes in each compartment. From these models predicted
environmental concentrations (PEC) are derived for multiple compartments.
For HHRA, the uptake of chemicals from various sources (air, water, food) and
via various exposure routes (inhalation, ingestion, dermal uptake) are combined to
derive a total daily intake.

2.2.4 Risk characterisation

Risk characterisation is the final step in risk assessment and determines whether
adverse health effects could occur at a particular exposure concentration. Risk
quotients (RQ) as defined in Equation (2.1) are often used and are expressed as
the ratio of the exposure to an acceptable effect level. In ERA the exposure level
is the PEC and the acceptable effect level is the PNEC.
exposure level
RQ = (2.1)
acceptable effect level
Different terminologies for RQ are used in different legislation (e.g., hazard
quotient (HQ) is used by some), but their meaning is essentially the same and the
only requirement is that both terms used in their calculation, that is, exposure
level and acceptable effect level, must have the same units, for example, aqueous

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Risk assessment of chemicals 31

concentrations for risk to aquatic organisms or oral dose for HHRA. If RQ ,1 no

risk is expected. The margin of safety (MOS) or margin of exposure (MOE) is
calculated as the inverse of the RQ if RQ ,1. It gives a measure of how much
difference there is between exposure and effect and is typically applied in
HHRA: the larger the MOS, the less concern there is that a chemical might
exceed the acceptable effect level.
An RQ ≥1 indicates a possibility of harm. Because exposure is likely to exceed
the acceptable effect levels, there is a requirement for a more in-depth risk
assessment and/or implementation of risk reduction measures. A chemical can
have multiple RQs for different protection goals (e.g., human health or
occupational health) and for different environmental compartment (e.g., water,
air, soil, sediment).
More recently, probabilistic methods have been introduced to better describe
variability and uncertainty of the many factors influencing exposure and effects.
Series of measured environmental concentrations or a Monte Carlo simulation of
various predicted environmental concentrations from exposure models can be
used to construct distributions of exposure levels. Distributions of effect data
such as ECs and NOECs capture variability in species sensitivity. If the
distributions of exposure and effect data overlap, there is a risk (Figure 2.3a).
Conversely, if the upper fifth percentile of the distribution of exposure
concentrations and lower fifth percentile of distribution of effect data do not
overlap, the chemical should be safe by the indicated MOS (Figure 2.3b).
The process of risk characterisation integrates information from effect assessment
and exposure assessment, provides an overview of the quality of the process and
describes the risks to individuals, communities and populations. This is the
information that is communicated to the risk managers. The summary should

Figure 2.3 Probabilistic risk assessment. (a) Distributions of exposure and effect
concentrations overlap, indicating risk; (b) upper 5% percentile of distribution of
exposure and lower fifth percentile of distribution of effect data do not overlap and
a MOS can be derived. MOS = margin of safety.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 32 Bioanalytical Tools in Water Quality Assessment

include a description of the key issues, and the overall strengths and limitations
(including uncertainties) of the conclusions. This may well result in a requirement
for additional information to improve the risk characterisation or may determine
that no further actions are necessary with the available information. Thus, risk
assessment is an iterative process where screening information is used to derive a
precautionary initial assessment. If this step identifies a problem, a more refined
assessment is carried out to reduce the uncertainty.

2.2.5 Uncertainty analysis

Given that all steps leading to risk characterisation apply simplified assumptions and
generalisation, an uncertainty analysis is vital for risk assessment. Uncertainty may
relate to lack of or limited knowledge of the true value of any parameters or
relationships among parameters. Uncertainty can be caused by indeterminacy,
when the true value of a parameter is not known, and variability, when the
parameters cover a range, such as temperature, system homogeneity and species’
and organisms’ sensitivity. In response to the need for uncertainty analysis, the
European Directive REACH has implemented a specific guidance document for
uncertainty analysis (European Chemicals Agency, 2012). This is especially
important when considering that the current international risk assessment have an
asymmetric perspective because they minimise the likelihood of positive results
by only continuing the process of the RQ ≥1 but stopping the process if RQ ,1.
Hence there could be a false-negative outcome and the reliability of the statement
‘no risk’ remains unknown.
An answer to uncertainty is the Precautionary Principle, which in its form as
Principle 15 of the Rio Declaration (United Nations, 1992) reads as follows: ‘In
order to protect the environment, the precautionary approach shall be widely
applied by States according to their capabilities. Where there are threats of serious
or irreversible damage, lack of full scientific certainty shall not be used as a reason
for postponing cost-effective measures to prevent environmental degradation’. The
Precautionary Principle goes back as far as 1974 as ‘Vorsorgeprinzip’ in the Clean
Water Act of West Germany and was declared as basis of the European Union’s
environmental policy by the Maastricht Treaty in 1992. History has shown how
damaging misuse or neglect of the Precautionary Principle can be when one
chemical is banned but many others that are similarly acting are replacing the
banned chemical (European Environment Agency, 2013). The Precautionary
Principle was invoked for the first time formally in risk assessment in the
European Union to ban the use of pentabromodiphenylether due to the high
uncertainty concerning exposure of infants via mothers’ milk as early as 2001.

2.2.6 Risk management

Risk management is a broader evaluation of the results of the risk assessment and
takes into account not only the scientific data but also social, economic and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Risk assessment of chemicals 33

Figure 2.4 Risk assessment/risk management paradigm. Risk assessment and risk
management are informed by each other and work towards ongoing improvement in
the process.

political considerations (Figure 2.4). Risk reduction measures can aim to replace a
chemical or implement control strategies to minimise exposure. Safety standards are
set by regulatory action and define safe levels of certain chemicals in various
environmental compartments (see Chapter 3).
Risk communication should be seen as a process to enable all stakeholders to
make an informed judgement about a risk and its management. There are
different perspectives to risk including actual risk, estimated risk and perceived
risk. Thorough risk assessment and risk communication minimise the mismatch
between the different perspectives. Risk management also monitors and evaluates
the effectiveness of the actions.

2.3 APPLICATION OF BIOANALYTICAL TOOLS IN

CHEMICAL RISK ASSESSMENT
In vitro methods can support risk assessment at all three main steps. During hazard
identification, in vitro tools can provide an initial screening of potential toxic effects
and classification of modes of toxic action.
During effect assessment, in vitro methods can provide additional evidence to (1)
identify mechanisms of chemically induced biological activity, (2) prioritise
chemicals for more extensive toxicological evaluation and (3) develop predictive
models of in vivo biological response (Shukla et al., 2010).
During exposure assessment, in vitro methods can serve as markers of exposure,
similar to biomarkers, if the relationship between chemical exposure and
magnitude of effect in an in vitro assay is established. Under certain
circumstances, in vitro methods may also serve as surrogates of chemical
analysis, especially for highly specific bioassays that are affected by relatively
limited numbers of contaminants.
More recently, there have been first attempts to base the screening-level risk
assessment solely on in vitro data. This is discussed in more detail in Chapter 9.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 3
Water quality assessment and
whole effluent toxicity testing

3.1 BACKGROUND
Water quality refers to the physical, chemical and biological characteristics of water.
For the purpose of this book the focus is on chemical water quality although it
should be emphasised that protection from pathogens is a key concern in drinking
water quality guidelines. Chemical quality encompasses salts, metals and organic
compounds, and our focus is on organic micropollutants.
The classical approach used for chemical water quality monitoring is to compare
detected individual chemical concentrations measured by targeted chemical analysis
to chemical guideline values (GVs) or standards. Safety standards are defined to
protect humans and the environment from unwanted chemicals and are the
foundation of water quality-based pollution control. There are several levels of
control (van Leeuwen and Vermeire, 2007):
• Water quality criteria are based on data, scientific judgement,
environmental and human health effects and provide guidance for
regulators when they are setting the standards, but they are not laid down
in any legislation. Despite this they provide a valuable tool in the
management of water pollution.
• Water quality guidelines provide recommendations on safe levels, but they
are not legally enforceable. They provide targets but exceeding them does not
necessarily result in clean up or enforcement actions.
• Water quality standards (QS) are upper exposure limits that are enshrined
in legislation. They are based on water quality guidelines or derived from

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0035

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 36 Bioanalytical Tools in Water Quality Assessment

scientifically based water quality criteria both by applying safety factors and
by political decision making.
The U.S. and European Countries have defined national standards for drinking
water quality: the National Primary and Secondary Drinking Water Regulations
as part of the Safe Drinking Water Act in the U.S. and the Drinking Water
Directive 2020/2184 of the European Parliament and of the Council (EP&EC,
2020). Australia (NHMRC, 2011) and Canada (Health Canada, 2020) on the other
hand rely on the guideline approach at the national level, which some
states/provinces have adopted, upon which they become legally binding
standards. The World Health Organisation has also defined drinking water
guidelines (WHO, 2017b). While these are evidently not legally binding standards,
they are meant to assist policy makers in the development of national standards.
There are guidelines for recycled water in some parts of the world. For example,
Australia has guidelines for potable water reclaimed from sewage (NRMMC/
EPHC/NHMRC, 2008), stormwater harvested for reuse (NRMMC/EPHC/
NHMRC, 2009b) and managed aquifer recharge (NRMMC/EPHC/NHMRC,
2009a). The WHO has the ‘Potable Reuse: Guidance for Producing Safe
Drinking-water’ (WHO, 2017c). Some U.S. states also have their own refined
guidance documents, for example, the ‘Water quality control policy for recycled
water’ for the State of California (State Water Resources Control Board, 2019).
Surface water guidelines and/or standards are intended to protect aquatic
ecosystems. They can have the character of standards, such as in the Water
Framework Directive (WFD) of the European Union (EP&EC, 2000) or
guidelines, such as the Australia and New Zealand Guidelines for Fresh and
Marine Water Quality (Australian Government, 2018a).
These documents provide frameworks for managing water quality, including by
setting chemical guideline/standard values for a range of chemicals. The WFD
contains environmental quality standards (EQS) for (groups of) 45 priority
substances (EP&EC, 2013), while the Australian Guidelines for Water Recycling
for Augmentation of Drinking Water Supplies (NRMMC/EPHC/NHMRC,
2008) provide guidelines for over 200 chemicals. Chemical guidelines cannot
possibly capture all chemicals potentially present in water, including
contaminants of emerging concern. Consequently, the recent revision of the EU
Drinking Water Directive allows risk-based monitoring approaches, provided that
they ensure full protection of public health (EP&EC, 2020). This revision allows
monitoring programmes to focus on chemicals that are relevant for a specific
water system.

3.2 DERIVATION OF GUIDELINE VALUES

There is similarity in the approaches for drinking water and surface water despite the
difference in protection goals (Figure 3.1). Drinking water GVs are typically derived

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Water quality assessment and whole effluent toxicity testing 37

Figure 3.1 Guideline values for single chemicals for drinking water and surface water
and how they are derived from in vivo toxicity data. NOAEL = no observed adverse
effect level, NOEC = no observed effect concentration, SSD = species sensitivity
distribution.

from an acceptable daily intake (ADI–also sometimes referred to as tolerable daily

intake (TDI) or reference dose (RfD)), which itself is usually derived from a no
observed adverse effect level (NOAEL) established from animal toxicity testing
and extrapolated to a human context (WHO, 2017b, Baken et al., 2018). The
ADI represents a daily intake that can be ingested over a lifetime without adverse
health effects. The ADI is multiplied by the average human body weight
(typically 60–70 kg) and divided by water consumption (typically 2 L/day) to
derive a safe concentration in water. This safe concentration is also often
multiplied by a relative source contribution factor to account for other sources of
exposure (not shown in Figure 3.1).
Surface water GVs are derived from no observed effect concentrations (NOECs)
using species sensitivity distributions (SSD) or extrapolation methods (European
Commission, 2011; Australian Government, 2018a) (Figure 3.1).

3.3 HUMAN USE OF WATER

Many chemical guidelines have been developed from risk assessment processes
based on typical exposure scenarios. For humans, the exposure is usually based

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 38 Bioanalytical Tools in Water Quality Assessment

on a daily consumption of 2 L of water per day over a lifetime, typically 70 years.

This information is used on the assumption that there will be no toxicological effects
despite continuous exposure over the entire lifetime. Human health standards in
food, water and air are based on the fact that for the vast majority of chemicals
there is a safe level of exposure, below which no adverse health effects occur.
Chemicals can sometimes be grouped with a typical compound used to represent
the whole group, such as, for example, benzo[a]pyrene for polycyclic aromatic
hydrocarbons (PAHs).

3.3.1 Drinking water

Drinking water can come from a range of sources including surface water (streams,
rivers and/or lakes), groundwater or rainwater directly. It can also be sourced from
seawater or wastewater treated and purified for human consumption. There are
growing numbers of chemical GVs and standards emerging, but the list of
regulated chemicals is often based on chemicals found in pristine water sources,
which are not applicable to less conventional water sources. Not all chemicals
have standards or guidelines if they are reasonably expected not to occur in a
drinking water source. Increasing global human population now encroach on
water catchments, with impacts from urban development, agriculture and forestry,
and seepage from landfill and runoff from mining to name just a few. In
many low- and middle-income countries there is simply insufficient access to safe
water sources causing local human populations to use water of impaired quality.
It is important to note that some chemicals do not have water quality criteria
because we do not yet have sufficient toxicological data to establish them. The
current World Health Organisation guidelines (WHO, 2017b) emphasise
preventative management of drinking water quality and the use of multiple
treatment barriers.
The Safe Drinking Water Act implemented by the United States Environmental
Protection Agency (U.S. EPA) sets legal limits on certain contaminants in drinking
water. They define the drinking water equivalent level (DWEL) as the concentration
in drinking water that over a lifetime exposure is protective of human health, at least
for threshold chemicals (i.e., non-carcinogenic chemicals). The DWEL reflect the
best available technology at the time and are subject to ongoing review. In
addition to the legal limits, the U.S. EPA determines water testing schedules and
methods that water providers must follow. Updated drinking water standards and
health advisories have recently been published by the U.S. EPA (2018).
In Europe, EU member states must comply with the EU Council Directive on the
quality of water intended for human consumption (EP&EC, 2020) but they can have
separate national regulations, provided they comply with the overarching Directive.
The Directive requires a regular monitoring programme using the analytical
methods specified therein, or equivalent methods. In the previous version, only
few organic contaminants were specifically regulated in the EU. The broad

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Water quality assessment and whole effluent toxicity testing 39

standard for individual pesticides was initially set in 1998 at 0.1 μg/L with all
pesticides combined not exceeding 0.5 μg/L (EP&EC, 1998). The 2020 revision
of the Directive laid down the essential QS at EU level (EP&EC, 2020) that were
derived from the WHO drinking water guidelines requiring monitoring and
regular testing of 48 microbiological, chemical and indicator parameters.
The Australian Drinking Water Guidelines (ADWG) are built around a 12-point
framework for the management of drinking water quality (NHMRC, 2011) and
provide the Australian community and water industry with guidance on the
provision of safe drinking water. The ADWG are part of the National Water
Quality Management Strategy (NWQMS, Australian Government, 2018b), a
nationally coordinated framework to facilitate consistent water quality
management across different types of waters (fresh water, marine water,
groundwater, estuarine water and recycled water) intended for a variety of uses
(for drinking, the environment, primary industry, recreation, industry and cultural
and spiritual values). The ADWG are subject to a rolling revision process with
regular amendment, the latest being from 2018, and the GVs have been
risk-based all along (NHMRC, 2011). The ADWG are intended to provide a
framework for good management of drinking water supplies that, if implemented,
will assure safety at the point of use. The ADWG does not provide mandatory
standards but gives guidance to agencies that have responsibilities associated with
the supply of drinking water, including catchment and water resource managers,
water regulators and health authorities in the states and territories of Australia.

3.3.2 Recycled water, stormwater and managed aquifer

recharge
Similar to the WHO and ADWG, the Australian Guidelines for Water Recycling
(AGWR, phases 1 and 2, NRMMC/EPHC/AHMC 2006; NRMMC/EPHC/
NHMRC 2008, 2009a, 2009b) are based on a preventative approach to water
safety management. Phase 1 of the AGWR provides a generic framework for
management of recycled water quality that applies to all combinations of recycled
water and end uses. These guidelines provide specific advice on the reuse of
treated sewage and grey water for purposes other than drinking and
environmental flows. Phase 2 extends the guidance in phase 1 on the planned use
of reclaimed water from sewage and stormwater to augment drinking water
supplies. The document focuses on the source of the water, initial treatment
processes and the blending of the water with drinking water sources
(NRMMC/EPHC/NHMRC, 2008). There is an increasing emphasis on the use
of a multi-barrier approach in preventing water quality incidents, rather than a
response when one occurs.
The AGWR phase 2 (NRMMC/EPHC/NHMRC, 2008) provide significantly
more GVs than the existing ADWG. This is because the source waters, in this
case sewage and stormwater, are expected to contain a broader range of chemical

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 40 Bioanalytical Tools in Water Quality Assessment

contaminants than conventional drinking water sources such as protected surface

water catchments. The process for setting these guidelines is a hierarchical
decision tree involving a number of steps including determining a list of
chemicals of interest, if there is an existing guideline, if the chemical is a
pharmaceutical and if there is health and toxicological information on which to
base the setting of a guideline. In the absence of these data it is then determined if
the chemical is likely to cause cancer, in which case it would be classified as
having no threshold of effect. If it is not causing cancer, then a threshold of
toxicological concern can be calculated as a conservative estimate of safe
concentrations and used together with exposure assessment data as a basis for risk
characterisation. This is a precautionary approach and protective of public health.

3.3.3 Dealing with unregulated chemicals in water

A serious drawback of most drinking water regulations is that they cannot react
promptly to contaminants of emerging concern. The AGWR (NRMMC/
EPHC/NHMRC, 2008) and Schriks et al. (2010a) have proposed an pragmatic
approach to derive provisional drinking water GVs for unregulated chemicals
once they have been detected in drinking water in the absence of statutory GVs
and this work was updated recently (Baken et al., 2018). The approach prioritises
available ADI, RfD or TDI values, and if those are not available, ADIs are
calculated from available toxicity data or, if needed, thresholds of toxicological
concern (TTC). When comparing GVs derived via this approach with measured
water concentrations of emerging pollutants, there is often a substantial margin of
safety between the measured water concentration and the provisional GVs, so that
there is no immediate action necessary (Schriks et al., 2010a, Baken et al., 2018).
As mixtures become more and more complex and as derivation of health-based
GV entail a risk-based approach, Dingemans et al. (2019) also suggested that
effect-based methods might be implemented in future drinking water legislations.

3.4 AQUATIC ECOSYSTEMS

The overarching goal of water quality risk assessment for ecosystems is to protect
biodiversity. A very important component of this is a thorough characterisation of
the ecosystem at risk and the environmental value placed on that ecosystem.
Environmental values are defined as values of the environment used for a healthy
ecosystem or for public benefit, welfare and safety and that require protection.
Important environmental values include aquatic ecosystems, primary industries,
recreation and aesthetics, and cultural and spiritual values (Australian
Government, 2018a). All water resources are subject to at least one environmental
value and in most cases several apply.
In the U.S. the Federal Water Pollution Control Act (U.S. EPA, 1976) from 1948
with amendments through to 1987 (now the Clean Water Act CWA) employs a
variety of regulatory and non-regulatory tools to reduce direct pollutant discharges

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Water quality assessment and whole effluent toxicity testing 41

into waterways and to manage pollutant runoff. The intention is to provide a range of
tools to achieve the broader goal of restoring and maintaining the chemical, physical
and biological integrity of waters that can support aquatic life. In the earlier years of
the legislation there was a focus on regulating discharges from point sources such as
municipal sewage treatment plants and industrial facilities. Since the 1980s efforts to
reduce non-point source pollution (e.g., from runoff) have been introduced including
cost sharing with landowners as a key tool. Under the CWA, the U.S. EPA has
implemented control programmes such as setting water QS (U.S. EPA, 2018).
This includes numeric and narrative water quality criteria, for example, ‘waters
shall be free from toxic pollutants in toxic amounts’. Whole effluent toxicity
(WET) testing, which is described in more detail in Section 3.6, is an important
component of the National Pollutant Discharge Elimination System. The most
recent update (2015) includes 50 chemical CMCs (criterion maximum
concentration) for the protection of aquatic life.
In Europe, the WFD (EP&EC, 2000) has set a goal to achieve ‘good ecological
status’ and ‘good chemical status’. Good chemical status equates to achieving the
QS established for chemical substances at the European level. 45 priority
substances have been assigned EQS (EP&EC, 2013), which have been derived
according to a technical guidance document for the derivation of EQS (European
Commission, 2011). In addition, there is a surface water ‘watch list’ of potential
water pollutants that are monitored to determine the risk they pose to the aquatic
environment and whether they should be included in the priority list. The EQS
are linked with emission limit values and discharge permits to ensure compliance
with the WFD. In the technical guidance document, QS are defined for the three
environmental compartments water, sediment and biota, considering the various
receptors at risk (humans, benthic biota, pelagic biota and top predators (birds
and mammals)). QS for biota refer the consumption of fish by humans or
secondary poisoning of aquatic organisms. Not all combinations of compartment
and receptor require the definition of QS for a given chemical in relation to the
physicochemical properties that define its environmental fate. However, if several
combinations are relevant, for example, for a hydrophobic and bioaccumulative
chemical, the QS are derived for all compartments and QSbiota and QSsediment are
translated to water concentrations. The lowest of these values is adopted as the
overall EQS. The effect assessment conducted in REACH (EP&EC, 2006a) and
the approach to estimate the QS share many principles of derivation. There are
two types of EQS defined in the WFD:
• the annual average EQS (AA-EQS) refer to the annual average concentrations
and are derived from chronic toxicity data, and
• the maximum acceptable concentrations EQS (MAC-EQS) refer to the
maximum concentration measured and are derived from acute toxicity data.
Von der Ohe et al. (2011) evaluated and prioritised 500 existing and emerging
micropollutants with this method and found in a monitoring study covering four

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 42 Bioanalytical Tools in Water Quality Assessment

European river basins that 44 of these 500 micropollutants exceeded the tentative
EQS, most of them pesticides. Monitoring of 223 pollutants at 4000 sites
confirmed that European freshwater is under pressure and that poor chemical
status was associated with poor ecological status (Malaj et al., 2014).
In Australia, all water quality is managed within the NWQMS (Australian
Government, 2018b), as noted above. The Australian and New Zealand
Guidelines for Fresh and Marine Water Quality (Australian Government, 2018a)
provide specific guidance to manage fresh and marine water quality for a variety
of uses, including the protection of aquatic ecosystems. There are default GV for
various stressors but site-specific GV are recommended that are relevant to local
conditions. Where possible, default GVs are derived using the SSD approach; for
chemicals with insufficient toxicity data for an SSD approach, GVs are derived
from predicted no effect concentration (PNEC) using an assessment factor
approach (ANZECC/ARMCANZ, 2000). The national guidelines are not
mandatory. Their enforcement is a state or territory responsibility through their
legislation. This extends to the requirement for WET testing (referred to locally
as direct toxicity assessment (DTA); see Section 3.6), which has now been
incorporated into discharge licences in various parts of Australia under state and
territory legislation for discharge to aquatic ecosystems. The most recent update
(2018) includes 138 default GVs for chemicals or chemical groups.

3.5 COMPARISON OF ENVIRONMENTAL AND DRINKING

WATER GUIDELINE VALUES
It might come as a surprise but for many chemicals the drinking water GV are higher
than their equivalent GV for the protection of aquatic ecosystems. For example,
Figure 3.2 compares WHO drinking water guidelines (WHO, 2017a, 2017b,
2017c) with the relevant European EQS values. This difference can be
rationalised by the fact that environmental GVs need to protect the most sensitive
aquatic species and that aquatic organisms are continuously exposed to the water
in which they live, while drinking water GVs protect just one species (humans)
only intermittently exposed to the water when they drink (an average of 2 L/day).

3.6 WHOLE EFFLUENT TOXICITY

Whole effluent toxicity (WET), whole effluent assessment (WEA) and direct
toxicity assessment (DTA) all refer to the assessment of the combined toxicity of
the mixture of all micropollutants in an effluent sample to aquatic organisms
using a suite of standardised aquatic toxicology assays (Gruiz et al., 2016). WET
testing has become an important component of the municipal and industrial
National Pollutant Discharge Elimination System (NPDES) in the U.S. (Grothe
et al., 1995) and as WEA in the European Union. The Australian and New
Zealand Guidelines for Fresh and Marine Water Quality (Australian Government,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Water quality assessment and whole effluent toxicity testing 43

Figure 3.2 Comparison of environmental quality standards (EQS) from the EU’s Water
Framework Directive or Swiss law with drinking water guideline values (DWGV; WHO
2017b). Data compiled in the Supplementary Information of Escher et al. (2018).

2018a) recommend the use of DTA both to monitor the impact of environmental
discharges as well to derive site-specific guidelines together with measurements
of single chemicals and biological monitoring (van Dam and Chapman, 2001).
The primary purpose of WET testing is to confirm that effluents discharged into
receiving waters do not adversely affect aquatic life. An advantage of testing whole
effluents is that it integrates the effect of all of the constituents in discharge water.
Typical aquatic toxicology tests are presented below. These test systems are
applied both for environmental risk assessment of chemicals (Chapter 2) and for
derivation of water quality criteria/standards, as well as for WET testing using
effluents or complex mixtures, but the focus of the sections below is on their
application in WET.

3.6.1 Test systems in aquatic ecotoxicology commonly

applied to WET testing
Acute and chronic WET testing methods had their beginnings in the 1950s and
1980s, respectively (Grothe et al., 1995) and were used to estimate the toxicity of
wastewaters. WET testing evaluates the adverse effects or toxicity to a population
of aquatic organisms determined experimentally in the laboratory with surrogate
organisms believed to be representative of those in the environment exposed to
the effluent discharge. This enables a situation-specific assessment. The method
can be used, for example, to derive guidance on the amount of dilution required

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 44 Bioanalytical Tools in Water Quality Assessment

to safely discharge an effluent to an aquatic environment or monitoring the

effectiveness of an effluent discharge management programme. The method can
also be used as a monitoring tool, testing ambient water that has or is suspected
of receiving a chemical pollutant discharge. More recently there had been a move
towards application of alternative test methods to WET to reduce animal testing
(Norberg-King et al., 2018).
Acute toxicity of effluents is generally measured using the original sample and a
minimum of five dilution concentrations. The tests are designed to produce
concentration–effect data expressed as a per cent dilution that is lethal (or causes
the defined effect) to 50% of the test organisms within a specified time interval
(24–96 h) or the highest concentration that is not statistically different from the
control (NOEC). A negative result in a single acute test does not preclude the
possibility of chronic toxicity or the possibility of temporal variability in an
effluent discharge. It also does not preclude the possibility of effects with some
taxonomic groups (e.g., plants) but not others (e.g., fish or crustaceans).
If toxicity tests are performed with single species, they should be representative
of the different trophic levels, that is, the position in the aquatic food chain. The three
most common taxa considered in aquatic toxicology are green algae as
representatives of primary producers, aquatic invertebrates such as water fleas as
primary consumers, and fish as aquatic vertebrates and secondary consumers
(Figure 3.3).
Typical test species in the U.S. EPA ‘Methods for Measuring Acute Toxicity to
Freshwater and Marine Organisms’ (U.S. EPA, 2002a) include freshwater species
such as water fleas (Ceriodaphnia dubia and Daphnia ssp.) and fish species
including fathead minnow (Pimephales promelas) and rainbow trout

Figure 3.3 Representative test organisms in aquatic toxicology, representing

different trophic levels: algae, water flea (Daphnia) and fish also simulate a
simplified food chain.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Water quality assessment and whole effluent toxicity testing 45

(Oncorhynchus mykiss). Marine species include fish such as the sheepshead

minnow (Cyprinodon variegates) and silverside (Menidia spp.), and the mysid
shrimp (Americamysis bahia). Chronic toxicity testing is also conducted using
similar species but also including the freshwater algae (Selanastrum
capricornutum) for growth, and the sea urchin (Arbacia punctulata) fertilisation
test (U.S. EPA, 2002b). Correct endpoint selection is critical to estimate the
sample effect concentration and to enable management strategies to be developed.
The use of standardised species and testing methods is advantageous in that the
results can enable comparison between effluents from different industries and will
lead to sound scientific data, however, they may be less relevant to specific sites
of interest in that endemic species may differ in sensitivity. The methods do,
however, allow for prediction of a safe concentration for similar species so is still
useful as a screening approach.
There are several standardised guidelines for toxicity testing of aquatic species
(OECD, 2006) and the International Organisation for Standardization (ISO) has
provided a number of guideline documents on water quality testing that have
been either directly adopted or further adapted by national bodies (e.g., DIN in
Germany, ASTM International in the U.S.). For algae, an exposure period of 72 h
followed by assessment of the inhibition of growth rate or biomass yields an
EC50 value (effect concentration for effect on 50% of the population; for
definition see Chapter 7) that is considered to represent the acute toxicity towards
algae, and the NOEC of the same concentration–effect curve is considered as the
chronic endpoint (ISO8692, 2004; OECD, 2011). For the water flea Daphnia
magna, the most popular invertebrate in ecotoxicity testing, the EC50 for
immobilisation after exposure of 24 hours to the chemical is taken as indicator of
acute toxicity (ISO6341, 1996), and the NOEC for the reproductive success
(number of progeny per adult) after 21 days of exposure is considered
representative of chronic toxicity (ISO10706, 2000).
The acute toxicity test for adult fish generally lasts 96 hours and allows the
derivation of the LC50, that is the concentration that is lethal for 50% of the test
fish (OECD, 1992; ISO7346-3, 1996). Warm-adapted fish species are often used,
such as fathead minnow or guppy, but many national regulations prefer the use of
more representative native species. Toxicity testing with fish should ideally be
performed in flow-through aquaria to ensure that the chemical exposure is
constant during the entire experiment. Since the life cycle of a fish may be
several years, chronic toxicity testing is limited only to sensitive life stages,
usually early life stages. Ethical issues with vertebrate testing have put more
pressure on replacing standard in vivo fish tests with alternative test methods. The
‘early life stage test’ evaluates the embryo and egg yolk larvae stage for
mortality, growth and deformation (ISO12890, 1999). The early part of this test,
the ‘fish embryo test’ (FET), which uses fish embryo up to hatching, is
considered an in vitro method in most legislations (OECD, 2013). This is
discussed further in Section 3.6.4.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 46 Bioanalytical Tools in Water Quality Assessment

A WET testing study of effluent from various wastewater treatment plants

(WWTPs) in Sydney, Australia, serves as an illustrative example (Bailey et al.,
2005). Effluent samples were tested with the acute toxicity test over 48 hours
with the water flea C. dubia. An effluent sample collected from one WWTP
exhibited acute toxicity to C. dubia as part of a routine screening programme.
The effluent sample tested was a composite of surface water grab samples
collected from the discharge stream. The 48 h-LC50 for C. dubia test was 31.9%
effluent, thus the pure effluent would be almost fully lethal for this species. A
toxicity identification evaluation (TIE) identified chlorfenvinphos, an
organophosphorous insecticide from a pet grooming business, as the source of the
high toxicity (Bailey et al., 2005).

3.6.2 In situ WET testing

Standardised laboratory tests combined with chemical characterisation of effluents
can be used to predict safe discharge concentrations to a receiving environment
based on the amount of dilution expected to occur on discharge. Field validation
of laboratory results is required to gain confidence in the ability of the laboratory
methods to extrapolate to field effects. For a controlled discharge of effluents, the
amount of dilution required can be calculated from the laboratory WET testing
combined with field studies. This can include the use of caging animals in the
field (Lazaro-Cote et al., 2018), and would typically involve having a series of
cages from the point of discharge to a predicted safe distance downstream in a
river for example, or beyond the mixing zone in the case of lakes and ocean
outfalls, to test the accuracy of the dilution prediction for the concentration of
contaminants to become acceptable. After an appropriate period of exposure
(determined by site and species) the animals are brought back to the laboratory
and examined for effects and biomarkers.
Meso- and macrocosms, enclosed experimental environments that replicate
larger ecosystems, are viable at least over one growth period (6–8 months),
although they rarely include higher vertebrates such as fish and reptiles. Finally,
outdoor bypass systems and field studies encompass the interactions in the
community and indirect effects such as predation, but the drawbacks are the low
number of possible replicates and high costs.

3.6.3 Biomarkers in WET testing

WET testing is based on standard testing methods that can consist of measuring a
range of endpoints including biomarkers of exposure and of effect (or sometimes
both) such as vitellogenin (egg protein) induction in male fish (Sumpter and
Jobling, 1995), or biochemical markers in liver and kidneys in fish (Petala et al.,
2009). This may require sacrificing test animals or taking of fluid samples such
as blood, however, this is seen as more acceptable than conducting experiments
on live animals. Biomarkers are a popular means of quantifying exposure to

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Water quality assessment and whole effluent toxicity testing 47

chemicals. For example, the induction of the egg yolk protein vitellogenin in male
fish is an indicator of the presence of estrogens and xenobiotic estrogenic
compounds in water (Jobling and Tyler, 2003).

3.6.4 ‘WET testing’ using bioanalytical tools

In vitro bioassays normally require sample extraction and enrichment (e.g.,
solid-phase or liquid–liquid extraction, see Chapter 12) to compensate for the
dosing factor in the assay medium. This has drawn some criticism because some
compounds (e.g., inorganics and metals) are lost during the extraction step. As
discussed earlier, bioanalytical tools are particularly useful to assess organic
chemicals in water because of the sheer number of chemicals that may be present.
Other inorganic water pollutants, such as metals, are more limited in number and
can be analysed by already exquisitely sensitive methods. Nevertheless, several
bioanalytical tools can and have been used in WET testing.
A small number of cell-based bioassays are classified as whole organism tests
(e.g., the Microtox assay based on bacteria, and chlorophyll fluorescence assays
based on green algae), and several studies have applied those alongside with
more conventional WET tests to a variety of whole effluents (Chang et al., 1981;
Dizer et al., 2002; Latif and Licek, 2004; Zurita et al., 2019).
A few studies have even adapted other bioanalytical tools to a WET format,
although some minimal sample preparation is often still required, for example,
filtration, pH adjustment, addition of powdered medium (Wagner and Oehlmann,
2009; Zegura et al., 2009; Niss et al., 2018). The introduction of whole effluent
into in vitro assays can, however, result in unpredictable effects that are not
necessarily associated with actual toxicity, but side effects caused by the matrix.
This requires thorough testing and validation of robustness to matrix interference.
The FET is a short-term toxicity test on embryo and sac fry stages of fish that may
serve as an ethical alternative in WET testing (Norberg-King et al., 2018). In this test
the embryos of the zebrafish (Danio rerio) or another fish species are exposed in
24-well plates to a range of dilutions of wastewater. A variety of parameters are
observed frequently during the course of the exposure. These parameters include
survival/mortality at the different stages, time to hatching, length, morphological,
physiological (e.g., heart rate) and behavioural abnormalities. For wastewater, a
shortened 48 h exposure standard test on the egg stage alone has been established
by the ISO (ISO15088, 2007). The zebrafish FET has been mandatory in
Germany for testing wastewater discharges since 2005 and has fully replaced the
96-h acute toxicity test on adult fish.
Lahnsteiner (2008) applied the zebrafish FET to screen wastewater quality and
compared the obtained results with acute toxicity testing with adult fish. Six types
of wastewater were sampled from Austrian factories involved in industrial
processes from the internal sewage collection point in each factory and from the
receiving environment from the sewage treatment plant. For dilution of the

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 48 Bioanalytical Tools in Water Quality Assessment

wastewater samples, groundwater was used. Acute toxicity tests for the fish eggs
were conducted using exposure for 48 hours and the results reported as EC50
values. Embryos were defined as ‘dead’ when they showed no heartbeat, no
somites were differentiated, egg yolk material was coagulated, or the tail was not
detached from the yolk sac. Most investigated wastewaters did not affect
zebrafish embryo viability. Only undiluted or marginally diluted wastewater from
hide tanning and galvanising metal industries induced effects in the FET. The
FET and acute adult fish toxicity agreed fairly well in this study. Gartiser et al.
(2009) applied the FET with zebrafish to a wider range of industrial effluents and
compared with other in vivo tests such as toxicity towards algae and water flea.
Algae turned out to be the most sensitive endpoint but as they cover a different
spectrum of pollutant and effects by colour cannot be excluded, the authors
recommended the use of a comprehensive test battery.
The FET has also been applied to investigate the success of advanced water
treatment. Cao et al. (2009) investigated the WET of secondary effluents treated
with chlorination, ozonation and UV irradiation using the FET with Japanese
medaka. While the controls and reverse osmosis permeate had .90% hatching
success, this was reduced to less than 40% in the secondary effluent. All
oxidative treatment steps reduced the toxicity towards embryos with a hatching
success increasing from 45 to 65% after treatment. Parallel to the decrease
hatching success, the percentage of dead and abnormal embryos was increased as
compared to the controls. Another study applied the FET with rainbow trout
(Oncorhynchus mykiss) eggs on a full-scale wastewater treatment plant with an
additional ozonation and sand filtration step (Stalter et al., 2010a, 2010b). All
waters had to be filtered as the raw water caused severe effects due to microbial
contamination. The membrane-filtered water did cause a slight time delay in
hatching, especially for ozonated wastewater but still between 70 and 80% of
larvae hatched as compared to 90% in the control. Only after the larvae
transitioned to the juvenile stage and started to feed were more significant effects
after ozonation observed, although those effects disappeared again after the
subsequent sand filtration. In both of these studies, water concentrated by
solid-phase extraction was tested with cell-based assays in parallel to the FET
assay. In vitro tools and the FET gave complementary information on groups of
chemicals being reduced (or not) with the different treatment and the overall
effect of the treated water, which can also be caused by mobilised organic matter
and transiently formed polar and reactive metabolites.

3.7 CONCLUSIONS
It is likely that chemical-by-chemical risk assessment for new and emerging
chemicals using whole animals will continue for some time yet for registration
purposes and where there are a single or limited number of chemicals being
discharged from a specific point source. Likewise, chemical GV will continue to

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Water quality assessment and whole effluent toxicity testing 49

be an important regulatory tool to assure good water quality for different uses.
However, as we realise that we are not exposed to single chemicals but rather
complex mixtures in real life and as alternatives to animal testing gain
momentum, this is likely to change (Norberg-King et al., 2018). Already, some
guidelines recommend the use of both in vivo and in vitro bioassay methods for
monitoring purposes. The concept of WET testing relies on testing
non-concentrated water and can be applied to a variety of test systems. It is the
test media that separates WET from other methods rather than the actual
endpoints themselves. WET testing has some parallels with bioanalytical methods
in that it can measure the aggregate effect of a range of chemicals in a mixture.
The limitation, however, is that when chemicals occur at trace concentrations
(e.g., pg/L or ng/L) whole-organism tests may not be sensitive enough to detect
those minor changes in water quality, which is dominated by bulk properties such
as salinity, pH or organic matter. A combination of WET testing methods with
the bioanalytical methods presented in this book may provide a very powerful
approach as neither one can replace the other but they provide complementary
information.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 4
Modes of action and toxicity
pathways

4.1 INTRODUCTION
When humans or wildlife are exposed to chemicals, several barriers must
be overcome before a chemical can elicit an adverse effect. The processes that
occur between exposure to that chemical and the adverse cellular effect can be
broken down into two phases: the toxicokinetic and toxicodynamic phases
(Figure 4.1).
The toxicokinetic phase describes all processes that link the external exposure
(e.g., via drinking water) to the biologically effective concentration within the
cell. Toxicokinetics encompass absorption and excretion, and internal distribution
and metabolism of a chemical within the whole body and within cells.
The toxicodynamic phase describes the cellular toxicity pathways taking place
inside the cell starting with the initial molecular interaction of the chemical and
its biological target. These interactions can induce cellular defence mechanisms
and other cellular responses that ultimately lead to observable toxic effect(s).
For the application of bioanalytical tools to be meaningful, the selected
assays must cover not only well-defined toxic mechanisms but also relevant
toxicokinetic steps. Cells can be thought of as simple models of organisms that
simulate many crucial processes. The lipid bilayer of the cell membrane is a major
barrier to chemical exposure. This is the main reason for advocating for the use
of whole-cell bioassays for the assessment of environmental samples and for
advising against molecular-based cell-free bioassays such as enzyme or receptor-
binding assays, which do not include a toxicokinetic component.

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0051

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 52 Bioanalytical Tools in Water Quality Assessment

Figure 4.1 Pathway from exposure to effect. Adapted from Escher and Hermens
(2002).

Cell-based bioassays can yield information on both the general toxicity to cells
(cytotoxicity) and on specific modes of action (MOA). This is important because
groups of chemicals with common MOA act together in mixtures by
concentration addition (Chapter 8). Using a suite of bioassays that covers various
MOA enables the generation of mechanistic information relevant for predicting
adverse health outcomes.
In this chapter, the structuring principles of toxicity pathways are summarised to
provide a better understanding of the processes that occur in cells and to introduce a
mode of action classification that serves as a basis for the selection of bioassays
discussed in Chapter 10.

4.2 TOXICOKINETICS
4.2.1 Uptake, distribution and elimination
Uptake and elimination can be a passive or an active process. Passive uptake is the
concentration-dependent diffusion of chemicals over cellular barriers (e.g.,
epithelial cells or biological membranes) and depends on the physicochemical
properties of the chemical. Hydrophobic chemicals accumulate in biota to a
higher extent but via slower uptake kinetics than more hydrophilic chemicals.
Active transport processes require energy and are capable of moving chemicals
even against a concentration gradient. Active transport is generally more
important for metals than for organics but one group, the ATP-binding cassette
(ABC) family of drug transporters, is also of importance for organic chemicals.
Uptake and elimination steps in cell-based bioassays are governed by the same
processes as in the whole body. There are, however, quantitative differences and
the most important step for in vitro to in vivo extrapolation is to account for the
higher complexity of uptake, distribution and elimination processes that occur in
a whole organism.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 53

Once taken up by the organism, chemicals are distributed via the lymph and
blood stream to organs and various tissues ultimately reaching the target cells.
Distribution is also relevant on the cellular level as it determines if a chemical
can reach its target site. As a hydrophobic and reactive chemical will more likely
be accumulated in biological membranes, for example, it is prevented from
reaching and reacting with DNA.

4.2.2 Xenobiotic metabolism

Xenobiotics are chemicals that are not native to an organism’s normal biochemistry.
When cells absorb xenobiotic chemicals, they are metabolised, a process called
biotransformation. This typically proceeds in three phases (Figure 4.2). Phase I
enzymes, such as the cytochrome P450 monooxygenases (CYP), one of the most
important families of metabolic enzymes, oxidise chemicals by adding functional
groups such as hydroxides to the molecules. In phase II reactions these functional
groups can conjugate with molecular entities such as sulphate and glucuronic acid
to yield larger and highly water-soluble metabolites, which are more easily
excreted from the body (Omiecinski et al., 2011). Phase III refers to the active
transport of chemicals across cell membranes by the ABC transporters mentioned
above. Phase III processes are not strictly metabolic; however, they do contribute
to increased elimination of chemicals from the cell and, therefore, are often
presented alongside phase I and II metabolic processes.
While the role of metabolism is primarily to detoxify chemicals, it can in some
cases produce more toxic metabolites, particularly in the oxidation reactions of
phase I. One prominent example is the bioactivation (i.e., oxidation) of polycyclic
aromatic hydrocarbons (PAH) to reactive epoxides, which may cause DNA
damage and subsequently carcinogenesis.

4.2.3 Toxicokinetic indicators of chemical exposure

Many xenobiotic chemicals trigger metabolic pathways, whereby they activate
and/or increase the metabolic activity within a given cell. Most cell types exhibit

Figure 4.2 Three phases of xenobiotic metabolism in a cell in relation to the other
toxicokinetic processes of absorption and excretion.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 54 Bioanalytical Tools in Water Quality Assessment

some degree of metabolic capacity. Liver cells (hepatocytes) have a particularly

large capacity for biotransformation.
Metabolic pathways themselves can be used to indicate the presence of
chemicals. Cellular pathways related to metabolism are regulated by the so-called
xenobiotic receptors (Omiecinski et al., 2011). The most prominent member of
this family of nuclear receptors is the aryl hydrocarbon receptor (AhR), a nuclear
receptor that is responsive to dioxin-like chemicals and other ligands.
Specific chemicals bind to these xenobiotic receptors and induce the transcription
of genes that encode metabolic enzymes. The binding of a nuclear receptor to its
nuclear binding site is not a toxic process in itself; however, it indicates the
presence of xenobiotic chemicals. Furthermore, the metabolic machinery set off
by binding to the receptor will change the structure of the molecule.
All xenobiotic nuclear receptors function in a similar way. In principle, a
chemical (or ligand) binds to the receptor (e.g., dioxin binding to AhR), which
causes bound proteins (e.g., heat-shock protein and other subunits in the case of
AhR) to dissociate from the receptor. The ligand–receptor complex can then
translocate into the nucleus, where in the case of AhR it associates with AhR
nuclear translocator (ARNT) to facilitate binding to a receptor-specific response
element on the DNA (e.g., dioxin response element DRE in the case of the AhR),
thus triggering the expression of the associated gene (e.g., CYP1A1 in the case of
the AhR) and the production of associated metabolic enzymes (Figure 4.3).

Figure 4.3 Activation of xenobiotic receptors using the example of the aryl
hydrocarbon receptor (AhR; ARNT = AhR nuclear translocator, DRE = dioxin-
responsive element). Persistent activators of AhR such as dioxin-like chemicals
cannot be metabolised by the produced enzymes and lead to a range of
AhR-related toxic effects (Denison et al., 2011), while those AhR ligands that can
be metabolised after activating the AhR, such as polycyclic aromatic hydrocarbons
(PAH), cause a different spectrum of effects.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 55

Table 4.1 Functions of the currently known nuclear xenobiotic receptors related to
metabolism and examples of chemicals that induce them.
Nuclear Receptor Function Inducing Chemicals
Pregnane X Induction of various phase I Steroids
receptor (PXR) enzymes (CYP)
Constitutive Protective role against Indirectly activated by
androstane toxicity induced by bile acid, phenobarbital, various
receptor (CAR) regulation of physiological pharmaceuticals
functions
Peroxisome Glucose, lipid and fatty acid Phthalates, fibrate
proliferator receptor metabolism pharmaceuticals
(PPAR)
Aryl hydrocarbon Induction of cytochrome PAH, PCDD
receptor (AhR) P450 (CYP1A1)
CYP = cytochrome P450 monooxygenase; PAH = polycyclic aromatic hydrocarbons; PCDD =
polychlorinated dibenzodioxins.

Table 4.1 lists the currently known nuclear xenobiotic receptors that are involved
in regulation of metabolism and which are all relevant for water quality testing.
Each receptor has several functions taking part in various metabolic processes
and in cell homeostasis. The AhR is the most pertinent receptor for toxicological
investigations. While the full physiological role of the AhR remains unclear,
activation of this receptor contributes to carcinogenicity via the CYP enzymes,
which can convert many of its ligands to reactive intermediates, consequently
causing DNA damage. Persistent AhR activators, such as dioxin-like chemicals,
cannot be metabolised by the produced enzymes and lead to a range of AhR-
related toxic effects (Denison et al., 2011), while those AhR ligands that can be
metabolised after activating the AhR, such as PAH, cause a different spectrum
of effects.

4.3 TOXICODYNAMIC PROCESSES: TOXICITY PATHWAYS

Toxicity pathways are defined as the cellular response pathways induced after
chemical exposure that are expected to result in adverse health effects (Collins
et al., 2008) (Figure 4.4). The starting point is the molecular interaction between
the xenobiotic chemical and the receptor or other biomolecules. This is called the
molecular initiating event (MIE). The chemical–biomolecule interaction triggers a
cellular response (e.g., translocation of the complex from the cytoplasm to the
nucleus, activation of genes, production or depletion of proteins or altered protein
signalling) that ultimately leads to observable endpoints or disease. We can
capture either critical steps of these cellular responses, the so-called key events

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 56 Bioanalytical Tools in Water Quality Assessment

Figure 4.4 Principle of cellular toxicity pathways. Adapted from Collins et al. (2008);
Ankley et al. (2010).

(KE), or the more general cellular stress responses, also called adaptive stress
responses.
As cellular responses occur via multiple steps, there are many points of
cross-over and branching both within and between toxicity pathways. Some of
the pathways induced by chemicals are natural endogenous pathways, whereby
the xenobiotic chemical simply replaces a natural ligand. Some authors thus
advocate the use of the term biological pathway instead of toxicity pathway.
Biological pathways may not directly cause an adverse effect but changed levels
of activity are still indicative of the presence of xenobiotic chemicals.
In ecotoxicology, the concept of toxicity pathways has been expanded to the
so-called ‘adverse outcome pathways’ (AOP, Ankley et al., 2010). An AOP links
the toxicity pathway (at the cellular level) with the response at the organ level,
followed by the response of the organism and finally the effect on the population
(Figure 4.5). Organ-level responses include altered physiology of the organ,
disruption of homeostasis, altered tissue development and/or disruption of organ
function. On the organism level, these effects translate to impaired development,
reproduction and/or death. These responses may then be observed across a
population and with potential implications for population and ecosystem
health. Organ- and organism-level responses are discussed in relation to human
health in Chapter 5. The AOP principles integrate human health and
environmental/ecological risk assessment. Representative test organisms and
population-level endpoints typically applied in environmental risk assessment are
discussed in more detail in Chapter 6.
In vitro cell-based assays can be used to indicate toxicity pathways at the cellular
level. Cellular responses do not necessarily imply higher-level effects in an

Figure 4.5 Principle of adverse outcome pathways (AOP). Adapted from Ankley
et al. (2010).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 57

organism, but they are a prerequisite. Species-specific factors and the inherent
sensitivity of individuals due to genetic polymorphism will further modulate the
causal chain. In addition, environmental factors have the potential to affect
individual and population health (Gohlke and Portier, 2007).
Examples of the processes taking place in a cell exposed to xenobiotic chemicals
are depicted in Figure 4.6. Chemicals can disturb membrane integrity and thus
membrane function by non-specific partitioning into the membranes of cells and
organelles. Further, xenobiotic chemicals can bind non-specifically and
specifically to proteins. Non-specific interaction with proteins can lead to protein
depletion, which ultimately causes oxidative stress. Specific binding to proteins
(e.g., receptors and enzymes) can result in inhibition or stimulation of
endogenous processes. Most often, binding to enzymes causes a blockage of the
active site thus inhibiting enzyme activity. Receptor binding can induce
endogenous processes. The (weak) binding of nonylphenol to the estrogen
receptor is one example of such agonistic effect on a receptor. Xenobiotic
chemicals can also block access of the endogenous agonist to the receptor, hence
decreasing normal activity – this is referred to as antagonistic activity. Finally,
the interaction (intercalation or covalent binding) of a chemical with DNA can
result in errors during replication and transcription. Repair and defence
mechanisms are in place to protect the cell from DNA damage up to a certain
threshold, above which, the damage becomes permanent.
Direct measurement of the interactions between a chemical and its cellular
target is difficult. The associated cell responses (e.g., gene activation after a

Figure 4.6 Possible toxicity pathways in a cell-based in vitro bioassay, ▴ = chemical.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 58 Bioanalytical Tools in Water Quality Assessment

receptor–ligand has been formed or the induction of DNA repair) are, however,
useful surrogate measures of these primary interactions (Figure 4.6). When the
capacity of repair and defence mechanisms is exceeded, all mechanisms
ultimately lead to cytotoxicity, that is, cell death.
There are two types of cell death; (i) apoptosis or programmed cell death and (ii)
necrosis, which occurs following irreversible inhibition of vital cell function.
Apoptosis is initiated to remove damaged cells and plays an important role in the
elimination of pre-cancerous cells. Both types of cell death can occur in relation
to non-specific and specific toxicity.

4.4 MODE OF ACTION CLASSIFICATION

With only a limited number of biomolecule types and a myriad of different receptors
and molecule–receptor permutations possible, some structuring principles can help
classify chemicals in groups that act together in a similar way.
A mode of action (MOA) is a common set of physiological and behavioural signs
that characterise a particular type of adverse biological response (Rand, 1995).
These responses can be caused by a range of molecular (toxic) mechanisms.
Molecular (toxic) mechanisms represent the crucial biochemical processes and/or
xenobiotic–biological interactions underlying a given mode of action. It must be
noted that the MOA is not a universal property of a chemical but is related to the
target organism, organ and/or tissue. As a given chemical can exhibit multiple
mechanisms of toxicity, the MOAs displayed may vary with exposure duration
(acute vs. chronic) and organism (a human will respond differently to a shrimp)
(Escher and Hermens, 2002).
The terminology for MOA and mechanism of action is not consistent in the
literature. For the purposes of this book and to be consistent with the framework
of AOP (Ankley et al., 2010), a toxic mechanism is defined as the initial
chemical–target molecule interaction (the MIE) and the resulting cellular
response. MOA refers to ‘biologically plausible series of key events leading to an
effect’ (Meek et al., 2014).
MIEs can be classified according to the type and degree of interaction taking
place between a chemical pollutant and its target molecule or target site (Escher
and Hermens, 2002). The main target classes for environmental pollutants are
(membrane) lipids, proteins and peptides and DNA (Table 4.2).
Depending on the type of interaction of the xenobiotic chemical with the target,
one can differentiate between non-specific, specific and reactive toxicity
(Table 4.3). Non-specific toxicity involves partitioning to the target site only,
whereas specific effects are the results of three-dimensional interactions including
specific H-donor/acceptor interactions and ionic interactions between the
chemical and target molecules. MOAs are classified as reactive when covalent
bonds are formed between the chemical and its target or when chemical reactions
are involved (e.g., oxidative stress) (Escher and Hermens, 2002). This generic

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 59

Table 4.2 Target molecules and target sites where the molecular initiating event
(MIE) takes place.
Target Class Target Molecules Target Site
Lipid Phospholipid Biological membrane
(phospholipid bilayer)
Lipid Triolein and other glycerols Storage lipid
Protein Structural protein, e.g., collagen Tissue
Protein Enzyme All cell types
Protein Nuclear receptor All cell types
DNA DNA bases (nucleic acids) Nucleus
DNA DNA backbone Nucleus

classification scheme can be further refined by differentiation between more

specialised target sites such as specific enzymes and receptors. Particularly
prominent is the nuclear receptor super family, a class of proteins that sense
hormones and regulate gene expression. Evaluation of hormone-induced
responses is more complicated than evaluating those of enzyme inhibition
because receptor binding effects can lead to complex feedback loops making
causal relationships difficult to elucidate. The same holds true for DNA damage,
for which numerous repair mechanisms have evolved. The induction of repair
processes is, however, still a valuable surrogate for the damage that has occurred.
The MOA classification is universal for all species and provides a common link
between human health and ecological risk assessment. Features exist, however, that
are unique to certain species, organ and tissue types. Only plants, algae and certain
bacteria have the capability to perform photosynthesis. The effect of herbicides that
specifically bind to and block the photosystem will therefore only be observable in
photosynthetically active cells. Chemically induced immunosuppression will only
be relevant for species that have developed an immune system. Yet, there are
many highly conserved features in all cells, and many are similar between
eukaryotic and prokaryotic (bacterial) cells.
The three major classes of MOAs are reviewed further in the following
subchapters to provide a basic understanding of the concept. For a detailed
treatise the reader is referred to Timbrell’s Principles of Biochemical Toxicology
(Timbrell, 2009).

4.4.1 Non-specific toxicity

Non-specific toxicity encompasses all cytotoxic responses that are not mediated by
specific or reactive mechanisms. Non-specific toxicity is often termed ‘narcosis’ or
‘baseline toxicity’ in ecotoxicology and ‘basal toxicity’ in human toxicology.
Chemicals disrupt membrane function by merely accumulating in biological

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 4.3 Mode of action classification scheme
60
MOA Class Target Molecules or Site(s) Molecular Mechanism(s) Mode of Action
Non-specific All membranes Non-specific disturbance of membrane Baseline toxicity
structure and functioning
Specific Energy transducing membranes Ionophoric shuttle mechanisms Uncoupling/depletion of ATP
(mitochondria)
Energy transducing membranes Blocking of quinone and other binding Inhibition of the electron transport chain
(mitochondria) sites, etc.
Energy transducing membranes Blocking of proton channels and other Inhibition of ATP synthesis/depletion of
(mitochondria) transport channels ATP
Photosynthetic membranes Blocking of photosynthetic electron transport Photosynthesis inhibition
Nerve cell membranes Interference with signal transduction Neurotoxicity
Specific enzymes Binding to enzymes Enzyme inhibition, e.g., AChE
(Nuclear) receptors Binding to (nuclear) receptors Inhibition or induction of (nuclear)
receptors, e.g., AhR, ER

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Reactive DNA, RNA Base modification and damage: electrophilic Direct genotoxicity (frameshift, cross-links,
(alkylation) and oxidative damage, bulky strand breaks, deletion, etc.)
adducts
Enzymes and receptors Non-covalent or covalent binding to enzymes Indirect genotoxicity (DNA repair,
of the nucleic acid metabolism, effect on recombination, regulation)
replication or repair
All proteins, peptides Electrophilic reactivity, alkylation and Damage and depletion of biomolecules
oxidation of proteins and GSH
Bioanalytical Tools in Water Quality Assessment

All membranes Formation of reactive intermediates (e.g., Degradation of membrane lipids and
ROS) causing peroxidation of membrane membrane proteins
lipids and membrane proteins
This classification is illustrative although not comprehensive.
AChE = acetylcholinesterase, AhR = arylhydrocarbon receptor, ATP = adenosine-5’-triphosphate, DNA = deoxyribonucleic acid, ER = estrogen
receptor, GSH = glutathione, RNA = ribonucleic acid, ROS = reactive oxygen species.
Adapted from Escher and Hermens (2002).
 Modes of action and toxicity pathways 61

membranes and at the interfaces of membrane proteins (van Wezel and

Opperhuizen, 1995). As cells lose their integrity, ion and proton gradients cannot
be maintained across membranes and ATP is depleted, consequently impairing
active transport and other ATP-dependent processes.
The AOP for baseline toxicity encompasses three possible MIEs: narcosis, direct
mitochondrial inhibition and decompartmentalisation (Vinken and Blaauboer,
2017). All lead to the key event of mitochondrial dysfunction followed by cell
death, both due to apoptosis and necrosis (Figure 4.7). In this scheme ‘narcosis’
refers to the above-mentioned intercalation of chemicals in biological membranes
and ‘decompartmentalisation’ to the disturbance of cellular organelles’ structure
and functioning. Mitochondria can be shut down by the specific mechanisms of
uncoupling, inhibition of the electron transport chain and the ATP synthase but
partitioning of chemicals into energy-transducing membranes also leads to
mitochondrial dysfunction (Escher et al., 2002).
Baseline toxicity is the minimum toxicity any organic chemical can exhibit. All
chemicals cause baseline toxicity with the same intrinsic potency, which means that
there are constant critical membrane concentrations independent of chemical
structure. All chemicals are equipotent when the effect is related to concentrations
in the biological membranes, the target concentration, but due to differences in
cellular uptake and makeup of the bioassay the nominal effect concentrations for
baseline toxicity in cell assays differ between chemicals and assays (Escher et al.,
2019). Even specifically acting and reactive toxicants induce baseline toxicity as
the underlying toxic mechanism, but the concentration necessary to induce
baseline toxicity is typically much higher than that required to induce specific
effects, so baseline toxicity does not play a role. For reporter gene assays, the
cytotoxicity is often caused by baseline toxicity and can be recorded
independently from the reporter gene activation. More details in Chapter 9.

Figure 4.7 Cellular toxicity pathway for baseline or basal toxicity. Adapted from
Vinken and Blaauboer (2017).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 62 Bioanalytical Tools in Water Quality Assessment

4.4.2 Specific modes of action

The common mechanistic basis for most specific MOAs is the selective binding of a
chemical to a protein (enzyme or receptor) target site (Table 4.3). The following
sub-sections discuss the molecular and cellular basis for a selection of relevant
MOAs for toxicological assessment. Integrative effects on the human organ and
organ system level are detailed in Chapter 5 and ecotoxicological endpoints are
reviewed in Chapter 6.

4.4.2.1 Enzyme inhibition

The most direct pathway for inactivation of enzymes is binding of the chemical to
the active site of the enzyme. Organophosphates, for instance, are inhibitors of
acetylcholinesterase (further discussed in Section 4.4.2.3).
Enzyme function can also be adversely affected via indirect toxicity pathways.
Haloacetic acids have been used as pesticides in the past and are also disinfection
by-products formed during chlorination in drinking water treatment. Haloacetic
acids are capable of replacing acetate in the mitochondrial tricarboxylic acid (TCA)
cycle, which is important for cell energy metabolism (Landis and Yu, 2004).
Fluoroacetate is formed from haloacetic acids in this process and undergoes all steps
of the TCA cycle leading to formation of fluorocitrate, which is a potent inhibitor
of the aconitase, the enzyme that converts citrate to isocitrate. In this example, the
toxicity is caused by the metabolite rather than by the haloacetic acid itself.
Many enzymes require cofactors (various metal ions, e.g., Fe3+, Ca2+) or organic
coenzymes (e.g., nicotinamide adenine dinucleotide phosphate (NADPH)) for their
catalytic function. Chemicals that destroy or deplete these cofactors will also
degrade the enzyme’s catalytic function. For example, fluoride complexes with
the cofactors Ca2+ and Mg2+ and inhibits the activity of important enzymes that
require these cofactors.

4.4.2.2 Disturbance of energy production

Mitochondria are the power plants of all cells. Interference with the mitochondrial
electron transport chain and oxidative phosphorylation leads to inhibition of ATP
synthesis, resulting in depletion of energy (Nicholls, 2013). Energy depletion
affects all cells with acute cell death as the outcome. In addition to non-specific
toxicity, disturbance of energy transduction mainly occurs through binding to
proteins and disruption of ion gradients across membranes.
Some chemicals, the so-called ‘uncouplers’, can shuttle ions and protons across
membranes and are thus more toxic than baseline toxicants without binding to
specific receptors (Terada, 1990). Uncouplers are typically weak organic acids
that form lipid-soluble conjugated bases, whose diffusion over the membrane
results in a net proton transfer (Spycher et al., 2008).
Chemicals such as cyanide, strobins and rotenone bind to the quinone binding
sites in the mitochondrial electron transfer chain and inhibit electron transport and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 63

thus ultimately energy production, while organotin compounds (e.g., tributyltin) and
N,N′ -dicyclohexylcarbodiimide directly inhibit ATP synthase.
In analogy, for photosynthetic organisms, energy production through
photosynthesis is inhibited by chemicals that block the photosystem or the
electron transport chain of chloroplasts (Moreland, 1980). Many herbicides such
as triazines (e.g., atrazine) or phenylureas (e.g., diuron) are direct inhibitors of
photosystem II. While herbicides exhibit low toxicity to mammals and most
vertebrates, some are suspected of possessing additional MOAs, for example,
atrazine is considered to be a modulator of aromatase (see Section 4.4.2.4).

4.4.2.3 Neurotoxicity
Many insecticides are neurotoxicants that act through interference with electrical
signal transduction or by inhibition of chemical signal transduction at the
synapse. At the molecular level, natural and synthetic pyrethroids (e.g., pyrethrin,
permethrin) inhibit sodium channels, which are responsible for transmission of
electrical signals through cells. By slowing down the re-closure of the sodium
channels, pyrethroids cause over-excitation. Organophosphate pesticides bind to
the enzyme acetylcholinesterase, inhibiting the cleavage of acetylcholine and
hence interfering with chemical signal transduction. Similarly, the neonicotinoid
imidacloprid acts as an antagonist on the nicotinic acetylcholine receptor.
The γ-aminobutyric acid (GABA) receptor is another target in nerve cells. The
GABA receptor acts as a gate for chloride channels and as an inhibitory
neurotransmitter by reducing the flow of chloride ions across chloride channels.
Some pesticides such as dieldrin, lindane (γ-hexachlorocyclohexane) and
avermectins are GABA agonists.
Insecticides have a lower toxicity to humans than to insects for a variety of
reasons. The organophosphates, for example, are better detoxified (metabolised)
by mammals than by insects. For some insecticides, the relevant receptors simply
play a different role in mammals and insects. The GABA receptor is important
for the peripheral nervous system of invertebrates, in which agonistic activity will
lead to paralysis. Conversely in mammals, the GABA receptors are only
important for the central nervous system and as many of the GABA agonistic
insecticides such as the macrocyclic lactones are incapable of crossing the blood–
brain barrier, mammals are not affected. This example demonstrates that even
highly conserved molecular targets can lead to very different adverse outcomes
depending on the organism of interest. This needs to be considered when using
bioassays as a tool for tracking specific group of chemicals.

4.4.2.4 Modulation of endocrine functions

Hormones are chemical signalling agents. When hormones bind to receptors, the
receptor–ligand complex triggers a series of effects through cell surface or
internal (cytosolic) receptors. The level of hormones is modulated by a negative

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 64 Bioanalytical Tools in Water Quality Assessment

Figure 4.8 Genomic pathway of hormone action.

feedback system. In the case of cytosolic receptors, the receptor–ligand complex

crosses the nuclear envelope into the nucleus, where the complex binds to a
specific promoter region on the DNA triggering the transcription and translation
of specific gene products (genomic pathway of hormone action, Figure 4.8).
Various components of the endocrine system are modulated by various
micropollutants as further described in Chapter 5.
Chemicals can interfere with hormone receptors by agonism and antagonism.
Agonists mimic the natural function of hormones, whereas antagonists block the
hormone receptor (Figure 4.9). Both functions can result in adverse outcomes.
Endocrine disruption is relevant for both humans and wildlife. A well-known
example of endocrine disruption in wildlife is the observed feminisation of male
fish caused by natural and synthetic human hormones as well as some industrial
chemicals in wastewater (Sumpter, 2002).
Chemicals can also interfere with the endocrine system via non-receptor-
mediated pathways such as inhibition of enzymes relevant for hormone
production. Hydroxylated polychlorinated biphenyls (OH-PCBs), for example,
inhibit the estrogen sulphotransferase causing increased estrogen levels in blood.
The enzyme aromatase is responsible for transforming testosterone to estradiol.
This important process can be induced by, for example, atrazine, or inhibited by,
for example, triorganotin compounds.

4.4.3 Reactive toxicity

The toxicity of reactive chemicals is caused by their reaction with endogenous
molecules. Examples of biological molecules (nucleophiles) that are attacked by
reactive chemicals (electrophiles) are the amino acid cysteine in peptides and
proteins, the bases in DNA, and the double bonds in phospholipids.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 65

Figure 4.9 Agonistic and antagonistic effects of chemicals.

4.4.3.1 Direct genotoxicity

DNA can be damaged by direct reaction with chemicals, reactive oxygen species
(ROS) or stressors such as ultraviolet light (UV). Alkylating agents, for example,
fluorouracil or methyl iodide, can covalently bind to DNA creating methyl
adducts particularly via the nitrogen atoms of guanine and adenine bases in DNA
(Figure 4.10). Electrophilic reactions of larger multifunctional molecules can
produce cross-links within or between DNA strands and large adducts can create
errors in translation or replication. Furthermore, large planar molecules can
intercalate into the DNA thus distorting its structure without directly reacting and
modifying DNA. Such distortion can nevertheless lead to mutations and other
errors during replication.
Enzymes are able to recognise damaged DNA and trigger repair mechanisms.
Sensors of the p53 pathway (see Section 4.5), for example, can detect strand
breaks and trigger DNA repair.
Methyl adducts that have been formed by alkylation of DNA can be
de-methylated by the enzyme alkyl transferase. Small lesions are repaired by base
excision repair and larger adducts are repaired by nucleotide-excision repair
(Figure 4.10). Repair mechanisms are, however, prone to error. Failure to repair
DNA generally triggers cell death via apoptosis.
DNA damage can (but does not necessarily) lead to loss of bases or strand breaks
into which incorrect bases can be inserted, resulting in irreversible mutations.
Mutations can cause errors in protein synthesis and are a major cause of cancer.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 66 Bioanalytical Tools in Water Quality Assessment

Figure 4.10 Reactive mechanisms leading to DNA damage and repair mechanisms.

4.4.3.2 Non-specific reactivity towards proteins

Biocides such as the antifoulant Sea-Nine, electrophilic chemicals (e.g., acrylates)
and dithiocarbamate pesticides can directly react with the thiol group in the
amino acid cysteine. Heavy metals such as mercury (Hg2+) and cadmium (Cd2+)
can also form complexes with thiol groups. These complexes can cause structural
damage to proteins and if this damage affects an enzymatic site of a protein,
non-specific enzyme inhibition may also occur.
Glutathione (GSH) is a peptide that contains cysteine and plays an important role
in the defence against reactive chemicals and internal reactive oxygen species
(ROS). Exposure to micropollutants and subsequent defence mechanisms can
lead to GSH depletion, which can cause proteins to lose their protection against
oxidative stress resulting in direct protein damage.

4.4.3.3 Oxidative stress

ROS such as the superoxide radical (O2 · −), hydrogen peroxide (H2O2) and the
hydroxyl radical (OH·) are formed during normal cell processes, particularly in
mitochondria during electron transport and NADPH-dependent enzyme processes
(Figure 4.11). ROS can also be formed by certain radical chemicals (e.g., paraquat)
and redox cyclers (e.g., quinones). Inhibition of the mitochondrial electron transfer
chain will also lead to the formation of ROS. In the presence of divalent iron
(Fe2+), reactive hydroxyl radicals will be formed. ROS can cause lipid
peroxidation, DNA damage and oxidation of proteins followed by loss of
enzymatic activity.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 67

Figure 4.11 Formation and deactivation of reactive oxygen species (ROS).

GSH = glutathione GSSG = glutathione disulphide, O2 = molecular oxygen,
O2 · − = superoxide, H2O2 = hydrogen peroxide, OH· = hydroxyl radicals,
NADP+ = nicotinamide dinucleotide phosphate, NADPH = reduced NADP+.

Cells have developed sophisticated systems to detoxify ROS and keep the redox
balance in the cell stable. Chemical stressors can, however, put more pressure on the
cellular redox balance overcoming the natural compensation mechanisms.
During detoxification of ROS, GSH is oxidised to the dimer glutathione
disulphide (GSSG) (Figure 4.11). A change in the ratio of GSH to GSSG is an
indicator of oxidative stress and ultimately leads to a disturbance of the cellular
redox homeostasis. Such imbalance will also impact other redox systems in the
cell. The hydrogen transferring coenzymes NADP+/NADPH, for example, will
be affected by a change in GSSG/GSH because NADPH is needed to reduce
GSSG back to GSH. Oxidative stress can in this way reduce the amount of
NADPH available for other vital functions, such as acting as coenzyme for the
phase I metabolic enzyme cytochrome P450.

4.4.3.4 Lipid peroxidation

ROS can not only damage DNA and proteins but also play a role in lipid
peroxidation. Polyunsaturated phospholipids are particularly vulnerable to this
attack, which leads to a chain reaction breakdown of fatty acids, which are
important components of membrane lipids. The degradation of fatty acids leads to
a change in (un)saturation, which causes alteration in fluidity of membranes and
structural damage of membranes. Lysosomes may lose their hydrolytic content
and the function of membrane-bound enzymes in the mitochondria and
endoplasmic reticulum can be disturbed.

4.5 KEEPING THE RIGHT BALANCE: ADAPTIVE STRESS

RESPONSE PATHWAYS
Damage to cellular macromolecules and cellular structures including nucleus,
mitochondria, endoplasmic reticulum and lysosomes triggers one or more cellular
stress pathways crucial for maintaining the balance in the cell (cell homeostasis)

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 68 Bioanalytical Tools in Water Quality Assessment

Figure 4.12 Activation of an adaptive stress response pathway. Adapted from

Simmons et al. (2009). TF = transcription factor.

and/or for repairing damage by transcriptional activation of genes that protect the
cells (Simmons et al., 2009). These stress responses are only induced by
chemicals or other stressors and thus referred to as adaptive. Adaptive stress
response pathways are activated and measurable at much lower concentrations
than cytotoxicity and can therefore serve as early warning signals of exposure to
chemicals or other stressors.
The principle of an adaptive stress response pathway is depicted in Figure 4.12.
On the left, the cell is shown under normal conditions. The transcription factor (TF),
which is the protein responsible for triggering the adaptive response, is silenced by a
sensor molecule. In this state, the sensor–TF complex cannot enter the nucleus.
When cells are under stress, the transducers break the sensor–TF complex, setting
the TF free. The TF then enters the nucleus, where it binds to specific sites on
the DNA (response elements), which in turn trigger the expression of the
associated genes.
This general adaptive stress response pathway shares similarities with both the
xenobiotic metabolism and the hormone response pathways in that they all
involve some mediating proteins, either nuclear receptors or TF. An important
difference is that the adaptive stress pathways occur in all cells, while other
toxicity pathways are specific to certain tissues and organs, for example, the liver
or reproductive organs.
The heat-shock response was the first stress response pathway discovered and is
important for the adaptation to hyperthermia (Table 4.4). The resulting gene
products help prevent heat denaturation of proteins. Chemicals that denature
proteins also trigger this protective pathway.
Exposure to metals and carbon monoxide can cause cell oxygen levels to be
depleted, which activates the hypoxia stress response pathway, triggering for
example transcription of proteins that increase transport of oxygen and iron
(Table 4.4). The metal response pathway differs from the other stress response

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 69

Table 4.4 Relevant adaptive cellular stress response pathways (Simmons et al.,
2009).
Pathway Sensor Transcription Inducing Chemicals and
Factor (TF) Stressors
Heat-shock Hsp90 HSF-1 Temperature, metals
response
Hypoxia VHL HIF-1 Oxygen depletion
Metal stress VHL MTF-1 Metals
Endoplasmic BiP XBP-1 Norephedrine,
reticulum stress diphenylcyclopropenea
Osmotic stress None MTF-1 High salt, glycol
Inflammation IkB NF-κB Metals, PCBs, smoke,
particles
Oxidative stress Keap1 Nrf2 Chemicals that produce ROS
DNA damage MDM2 p53 Electrophilic chemicals, UV
radiation
a
Hirota et al. (2010); Yang et al. (2011).

pathways in being constitutive (always expressed) as opposed to adaptive (only

expressed after activation). Its activation induces increased synthesis of
metallothionein proteins, which are small cysteine-rich proteins that chelate metals.
The endoplasmic reticulum plays a central role in lipid synthesis and folding and
maturation of proteins. The endoplasmic reticulum stress response pathway induces
genes that help refold proteins and remove damaged ones.
Osmotic stress triggers a pathway that ultimately leads to increased solute
transport across membranes. The inflammatory stress response is mediated by
the nuclear factor kappa B (NF-κB), which is closely related to immune
responses and causes induction of cytokines, cytochromes P450 and regulators of
apoptosis.
The mammalian cellular defence mechanism against oxidative stress is primarily
mediated at the transcriptional level by Nrf2 (NF-E2-related factor 2), which is
responsible for the induction of detoxification and antioxidant genes (Nguyen
et al., 2009) (Table 4.4). Nrf2 activates the transcription of sequences containing
the antioxidant response element (ARE). ARE is found in the promoter region of
genes encoding the major detoxification enzymes including glutathione
S-transferase A2 (GSTA2) and NADPH:quinone oxidoreductase 1 (NQO1),
which are the two major contributors to cellular protection. These enzymes serve
to neutralise ROS and reactive chemicals, biosynthesise GSH, direct xenobiotic
efflux and remove oxidised proteins. The net result is to limit oxidative damage
and to detoxify cells.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 70 Bioanalytical Tools in Water Quality Assessment

The most important response to DNA damage is regulated by the p53 family of
TFs. Under normal conditions, p53 is negatively regulated by the sensor MDM2.
Upon DNA damage, the p53 is stabilised and triggers a series of DNA repair
mechanisms. p53 is also referred to as ‘the tumour suppressor gene’ and regulates
cell cycle arrest and apoptosis.
As discussed, stimulation of adaptive stress response pathways is not a direct
indicator of toxicity but an early indicator of the presence of stressors. Since
activation occurs at concentrations of micropollutants lower than those required
to elicit an observable adverse effect, these pathways are useful early warning
signs with potential for application in water quality assessment.

4.6 CONCLUSIONS
All the different toxicity pathways discussed in this chapter are highly
interconnected. The complete picture is thus much more complex than presented
through independent view of the individual processes. Figure 4.13 expands the
simplistic picture drawn in Figure 4.6 and interconnects all these different
processes. First, metabolism can lead to both toxification and detoxification, and
the reactive metabolites of phase I in particular may cause direct reactive toxicity
and oxidative stress. Second, GSH acts as scavenger of reactive intermediates but

Figure 4.13 Interplay between various toxicity pathways and the effects induced if
injury is beyond repair.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Modes of action and toxicity pathways 71

if it is depleted, it cannot continue its role in keeping the redox homeostasis. The cell
then invokes a second line of damage control, represented by the adaptive stress
response pathways discussed in Section 4.5.
The lower part of Figure 4.13 highlights how the defence and repair mechanisms
initially serve to protect the cell but can become overwhelmed if the damage is too
great. As a final resort, the cells can invoke programmed cell death (apoptosis), but if
the damage is too severe, necrotic cell death will occur.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 5
Toxicity pathways of chemicals
in humans

5.1 INTRODUCTION
This chapter explores the different types of toxicity that may be associated
with chemical contaminants in drinking water, including toxicity to specific
organs (e.g., liver, kidneys, heart), organ systems (e.g., blood formation,
immune, nervous, endocrine systems) and integrated organism effects (e.g.,
developmental and reproductive effects, carcinogenicity). Molecular and/or
cellular level toxicity that is described in Chapter 4 can translate into effects at
the tissue, organ, organ system and eventually at the organism and population
level (Figure 5.1).
Within the framework of the adverse outcome pathway (AOP) (Chapter 4), once
a toxicant has reached its target site (toxicokinetic phase) and affected its biological
target (toxicodynamic phase), the resulting cellular-level effect(s) can lead to
dysfunctions at higher levels depending on its severity and the capacity of repair
and compensation mechanisms (Figure 5.2). It should be noted that on the
organism scale the toxicokinetic processes are broader than at the cellular level,
and are referred to as absorption, distribution, metabolism and excretion
(ADME). These processes deliver the chemical to its cellular target site, where
the AOP is initiated.
Exposure to environmental pollutants can result in a variety of effects in the
exposed organism. In broad terms, toxicity is an adverse effect on the production,
function and/or survival of cells. Some of these toxic effects are very general and
can potentially affect all types of cells, while others are specific to certain tissues

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0073

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 74 Bioanalytical Tools in Water Quality Assessment

Figure 5.1 Simplified scheme of biological organisation illustrating, where in vitro and
in vivo investigations fit in the sequence of effects that take place from the molecular to
the ecosystem level.

due to their unique structure and/or function. Some biological functions are fulfilled
by multiple organs (e.g., the endocrine system), and toxicity to any of the organs
involved may result in failure of the whole system. A thorough understanding of
these concepts for each potential site of toxicity is critical when developing a
comprehensive screening battery for risk assessment.
Unless otherwise indicated, the information in this chapter is based on classical
toxicology textbooks, in particular, Ballantyne et al. (1995), Fox (1991) and
Klaassen (2013).

5.2 ROUTE OF EXPOSURE

The route of exposure is significant because chemicals can reach different targets
depending on the point of entry into an organism (Figure 5.3). For drinking
water, oral ingestion is the main route of exposure and the digestive system will
be the focal point of entry into the organism.
The majority of ingested toxicants will be absorbed in the small intestine, which
has a very large, specialised surface area making it very efficient at absorbing
nutrients but also unfortunately toxicants from food and water. Absorption can
take place via active transporters but is usually a passive process, where toxicants
traverse the epithelial barriers and reach blood capillaries by diffusing through

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Toxicity pathways of chemicals in humans 75

Figure 5.2 Toxicokinetic processes on the organism level (ADME – absorption,

distribution, metabolism and excretion) connected to the adverse outcome pathway
(AOP) that links cellular-level events to whole organism effects.

cell membranes. Lipid solubility is usually the most important property influencing
absorption with lipophilic (fat-loving) chemicals more readily absorbed than
hydrophilic (more water-soluble) substances, but size and charge also influence
diffusion with small and uncharged substances more easily absorbed. Xenobiotics
absorbed in the gastrointestinal tract are transported to the liver via the hepatic
portal vein.
Once in the liver, the absorbed chemicals will undergo ‘first-pass metabolism’:
biotransformation by cytochrome P450 enzymes and conjugation with large
hydrophilic molecules (such as glucuronide or sulphate). The biological purpose
of first-pass metabolism is to make lipophilic toxicants more water soluble, thus
facilitating their excretion. After biotransformation, xenobiotics can travel via
either of two routes: large water-soluble chemicals are excreted back into the
small intestine via the bile duct and eventually excreted from the body in faeces.
Those chemicals no longer pose a risk to other organs because no further
contact will occur (unless they are susceptible to ‘enterohepatic recycling’,
where the conjugate is cleaved off by microbial activity in the intestinal tract
and the chemical can be reabsorbed and sent back to the liver). Alternatively,
xenobiotics that are still sufficiently small and lipophilic after first-pass
metabolism can instead enter the systemic blood circulation, where they can
reach and affect any tissue perfused by blood – in other words, all tissues –
particularly if they are lipid soluble. Hydrophilic toxicants will eventually be
excreted into bile by the liver or into urine by the kidneys. Highly lipophilic
chemicals that are resistant to biotransformation (such as polyhalogenated
biphenyls and chlorinated hydrocarbons) are very hard to eliminate and tend
to accumulate in the body upon repeated exposure – a process called
‘bioaccumulation’.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 76 Bioanalytical Tools in Water Quality Assessment

Figure 5.3 Simplified absorption, distribution, metabolism and excretion (ADME)

diagram showing sites of intake (large green arrows: oral, respiratory and dermal
intake), excretion (large red arrows: faecal, urinary and respiratory excretion) and
internal barriers for xenobiotics (dashed blue lines: lining of the GI tract, skin,
blood–brain barrier, blood–testis barrier in males, placental barrier in pregnant
females and liver metabolism). Oral ingestion is the main route of exposure for
contaminants in drinking water. GI = gastrointestinal.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Toxicity pathways of chemicals in humans 77

5.3 BASAL CYTOXICITY

Basal cytotoxicity refers to the ability of a chemical substance to damage living cells
and may be described as any general cellular-level effect that leads to dysregulation
of on-going cellular activity (e.g., disruption of ATP production, protein synthesis),
impairment of cellular maintenance, dysregulation of gene expression and/or
physical damage to cell structures (e.g., proteins, plasma membranes) (Vinken
and Blaauboer, 2017). Basal cytotoxicity is not cell-type specific – it can affect
any tissue type and occurs at similar critical membrane concentrations in all cell
types (Escher et al., 2019). It is therefore also often called ‘baseline toxicity’ (see
Chapter 4). This impaired cell structure and function can lead to cell and tissue
death, either planned (apoptosis) or unplanned (necrosis) cell death.

5.4 TARGET ORGAN TOXICITY

In the context of drinking water, toxicity to three organs is particularly relevant: the
liver (hepatotoxicity), the kidneys (nephrotoxicity) and the heart/blood system
(cardiovascular toxicity). Gastrointestinal tract and bladder toxicity are of course
also important, but the main mechanisms of toxicity to these targets are basal
cytotoxicity (Section 5.3) and carcinogenicity (Section 5.5.1).

5.4.1 Hepatotoxicity
The liver is the main organ where exogenous chemicals are metabolised to make
them more hydrophilic for excretion. Consequently, liver cells can be exposed to
high concentrations of toxicants. Thankfully, a healthy liver has an immense
capacity for self-repair and once the toxicant is removed recovery is usually possible.
After absorption by the small intestine, ingested nutrients, vitamins, metals,
drugs and environmental toxicants are all distributed to the liver via the hepatic
portal vein (Figure 5.3). Efficient scavenging or uptake processes extract these
absorbed materials from the blood for catabolism, storage and/or excretion into
bile. Hepatocytes (liver cells) are rich in mitochondria to provide for their high
energy needs and cytochrome P450 enzymes, which conduct the liver’s main
function of metabolism and detoxification. Hepatocytes also have a significant
role in protein synthesis by recycling all major plasma proteins, carbohydrate and
lipid metabolism, cholesterol production, and bile secretion, which can be an
important detoxification mechanism.
There are several key factors that modulate hepatotoxicity:
• Uptake and concentration: the liver is immediately ‘downstream’ of the
gastrointestinal tract, and as such receives the highest concentration of
lipophilic drugs and environmental pollutants from the oral route. Other
toxins are rapidly extracted from the blood into hepatocytes via active
transport mechanisms.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 78 Bioanalytical Tools in Water Quality Assessment

Figure 5.4 Spectrum of hepatotoxicity from the molecular to the organism level.

• Bioactivation and detoxification: one of the vital functions of the liver is to

eliminate exogenous chemicals and toxic endogenous intermediates.
Biotransformation, and in particular phase I metabolism, can, however,
generate reactive metabolites, which can interact with proteins and other
biological molecules.
• Regeneration: the liver has a high capacity to restore lost tissue and function
by regeneration. Loss of hepatocytes triggers proliferation of mature
hepatocytes to replace the lost tissue, which is initiated by cytokines and
growth factors. Nevertheless, chemicals that can interfere with the cell
cycle (e.g., colchicine) can block that regenerative ability.

Hepatotoxicity results in impaired liver function and the potential build-up of toxic
by-products of cellular metabolism (Figure 5.4). There are several known
mechanisms of toxicity to liver cells, including direct cytotoxicity to hepatocytes
(e.g., acetaminophen, carbon tetrachloride, microcystin), damage to epithelial
cells of liver capillaries (e.g., after excessive dose of acetaminophen, endotoxin,
microcystin), impaired bile excretion (usually from interference of bile salt export
pumps by toxicants such as pharmaceuticals, hormones and metals) and excessive
cell proliferation to replace dead cells (hyperplasia; e.g., after chronic exposure to
excess androgens, alcohol and aflatoxin).

5.4.2 Nephrotoxicity
The principal role of the kidneys is to filter blood and maintain total body
homeostasis. The kidneys play a central role in excretion of metabolic wastes

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Toxicity pathways of chemicals in humans 79

(such as urea) and in the regulation of extracellular fluid volume, electrolyte

composition and pH of the blood. The kidneys also produce hormones (renin and
erythropoietin) that regulate extracellular volume and red blood cell production
and metabolise vitamin D3 to its active form. Similar to the liver, the kidneys are
equipped with a variety of detoxification mechanisms and have considerable
functional reserve and regenerative capacities.
The kidneys are particularly sensitive to blood-borne toxicants as they receive
about a quarter of the cardiac output (Figure 5.3). The processes involved in the
production of urine may also concentrate potential toxicants in the tubular fluid.
A wide variety of pharmaceuticals (e.g., antibiotics, analgesics, radiocontrast
media, anti-cancer agents and angiotensin inhibitors and blockers),
environmental chemicals and metals can cause nephrotoxicity via structural
and/or functional damage. Proper kidney function is highly dependent on
passive and active (ATP-driven) transport mechanisms. Toxicant-induced
interruptions in energy production for any of these active transport mechanisms
or interference with critical membrane-bound enzymes and/or transporters can
thus seriously impact kidney function (Figure 5.5). The efficiency of the
kidneys is also dependent on tight control of capillary pressure, and the
kidneys are particularly sensitive to vasoactive substances, that is, substances
that modulate blood pressure.

5.4.3 Cardiovascular toxicity

Cardiovascular toxicology focuses on adverse effects on the heart and the vascular
system. Exposure to toxic chemicals can result in alterations of biochemical
pathways, defects in cellular structure and function, and pathogenesis of the
affected cardiovascular system.

Figure 5.5 Spectrum of nephrotoxicity from the molecular to the organism level.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 80 Bioanalytical Tools in Water Quality Assessment

5.4.3.1 Cardiotoxicity
Heartbeat is controlled by specialised pacemaker cells, and cardiac
electrophysiology and function are under neuro-hormonal regulation. The
primary contractile unit in the heart is the cardiac muscle cell, or cardiomyocyte.
Stimulation of cardiomyocytes through bioelectricity is due to carefully
orchestrated transport of three positively charged ions: calcium, sodium and
potassium. Each of the ions has specific channels and pumps on the membrane of
cardiac myocytes.
The very high energy requirements of the heart muscle (continuous synthesis of
ATP via mitochondrial oxidative phosphorylation is required for cardiomyocyte
function) and heavy reliance on ion channels and pumps are particularly relevant
for cardiotoxicity. Not surprisingly, many substances can cause cardiac toxic
responses, mostly by affecting ion channels (e.g., the anti-arrhythmic drugs
verapamil and quinidine), calcium ion homeostasis (e.g., pharmaceuticals such as
ouabain, some antimicrobial and antiviral agents, aldehydes, halogenated alkanes
and metals), and electrical excitability and action potential generation (e.g., local
anaesthetics like benzocaine or procainamide) (Figure 5.6).

5.4.3.2 Vascular toxicity

Toxic responses of the vascular system include changes in blood pressure and
damage to blood vessels (Figure 5.6). The main function of the vascular system is
to provide oxygen and nutrients and to remove carbon dioxide and metabolic
products to/from organ systems. The vascular system also delivers hormones and
cytokines to target organs (Figure 5.3). Dilation and constriction of blood vessels
are controlled remotely by neurons and hormones, such as epinephrine,
norepinephrine and angiotensin, and locally by oxygen supply and

Figure 5.6 Spectrum of cardiovascular toxicity from the molecular to the organism
level.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Toxicity pathways of chemicals in humans 81

endothelium-derived relaxing factor (EDRF). Thus, neurotoxicity or endocrine

disruption may also affect vascular function.
Blood vessels are mostly composed of epithelial cells enveloped by smooth
muscle cells. After absorption, all chemicals come into contact with the vascular
system. Specific vascular toxicity can occur from damage to either epithelial cells
(e.g., aspirin, endotoxins, carbon monoxide) or smooth muscle cells (e.g., metals
interfering with calcium homeostasis) or from exposure to vasoactive chemicals
(e.g., cocaine, nicotine, metals). It is not entirely clear how toxic responses of the
vascular system affect physiological function and/or cause toxicity to other
organs, but damage to vascular epithelial cells could produce reactive oxygen
species (ROS) and subsequent oxidative injury.

5.5 NON-ORGAN-DIRECTED TOXICITY

Non-organ-directed toxicity includes carcinogenicity and developmental toxicity.

5.5.1 Carcinogenicity
Chemicals that induce cancer have been broadly classified in two categories: (i)
genotoxic carcinogens (e.g., PAHs) that interact physically with DNA to alter or
damage its structure, and (ii) epigenetic carcinogens that impact DNA expression
through DNA methylation, protein phosphorylation and receptor-mediated
effects, without directly affecting DNA structure (Figure 5.7). Either case can
eventually lead to aberrant cell cycle kinetics and unregulated cell growth.
Carcinogenesis develops over three stages:
• Initiation is the introduction of a ‘mistake’ (mutation) in the DNA sequence.
Initiation can be caused by genotoxic carcinogens binding to DNA and
causing errors during DNA synthesis. Initiation on its own is not sufficient
to cause abnormal cell growth because DNA damage can sometimes be
repaired or because the cell can lose its viability due to the mutation.
• Promotion is the selective expansion of initiated cells.

Figure 5.7 Spectrum of carcinogenicity from the molecular to the organism level.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 82 Bioanalytical Tools in Water Quality Assessment

• Progression involves the conversion of unstable promoted cells into stable

malignant tumours. Due to the resulting increased DNA synthesis,
additional genotoxic events may occur at this stage, resulting in additional
DNA damage including chromosomal aberrations and translocations.
Complete carcinogens have the ability to function at all levels, that is, initiation,
promotion and progression. Many carcinogens are not intrinsically carcinogenic
but require metabolic activation to become carcinogenic. This may also result in
tissue-specific effects, as different tissues can have different levels of
enzyme expression.

5.5.2 Developmental toxicology

Developmental toxicity focuses on adverse effects on development caused by
exposure to toxicants. Development is characterised by changes that are
orchestrated by a cascade of factors regulating gene transcription (Figure 5.8). A
particularity of developmental toxicology is that the sensitivity of the organism to
toxicants can vary depending on its developmental stage.
Embryotoxic chemicals affect the conceptus prior to the foetal stage (usually up
to 8 weeks in humans). Imprinting, implantation, gastrulation and organogenesis all
occur during embryo development, and toxicants that interfere with cell
proliferation, differentiation and/or apoptosis may lead to embryotoxicity (e.g.,
cyclophosphamide).
Foetotoxic chemicals affect the conceptus from the foetal stage onwards (usually
after 8 weeks in humans).
Teratogens are toxic chemicals that cause birth defects and can lead to pre- and
postnatal mortality. Gastrulation and organogenesis during embryo development
and the subsequent tissue differentiation and growth during foetal development
are particularly sensitive to teratogens. Toxicants that can affect cell migration,

Figure 5.8 Spectrum of developmental toxicity from the molecular to the organism
level.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Toxicity pathways of chemicals in humans 83

cell–cell interactions, differentiation, morphogenesis and energy metabolism are

often teratogens (Figure 5.8).
Exposure to developmental toxicants can result in death of the embryo, death of
the foetus or teratogenesis. Endocrine-disrupting compounds (EDCs, e.g.,
diethylstilbestrol) may also negatively affect development of the conceptus (see
Section 5.6.4 for more details on endocrine disruption).

5.6 SYSTEM TOXICITY

Some biological functions are fulfilled by systems composed of multiple organs
(e.g., the immune system), and toxicity to any of the organs involved may result
in failure of the whole system. This section discusses toxicity towards the blood
system (haematotoxicity), immune system (immunotoxicity), nervous system
(neurotoxicity), endocrine system (endocrine disruption) and reproduction
(reproductive toxicity).
Some health outcomes of relevance for risk assessment of new chemicals have
been excluded from this review because they were deemed unlikely to result from
exposure to chemical contaminants in drinking water (although they may occur
from different exposure routes to that same water, e.g., showering). These include:
• Sensory organ toxicity, including ocular toxicity,
• Respiratory toxicity,
• Cutaneous toxicity,
• Musculoskeletal toxicity (e.g., myotoxicity).

5.6.1 Haematotoxicity
The production of blood cells (haematopoiesis) is a highly regulated sequence of
events by which blood cell precursors proliferate and differentiate to meet the
body’s relentless needs for oxygen transport, host defence and repair and blood
homeostasis. The main organs involved in haematopoiesis are the bone marrow
and the spleen. A haematotoxicant is a toxicant that either interferes with
haematopoiesis or affects the viability of red blood cells, which can result in
anaemia and hypoxia (lack of oxygen). Effects on white blood cell viability are
covered in the section on immunotoxicity below.
Haematopoiesis requires carefully orchestrated cell maturation and
differentiation and is particularly sensitive to cytoreductive or antimitotic drugs
used for cancer treatment and toxicants that can interfere with differentiation and
maturation of blood cell precursors.
The viability of red blood cells can be affected by oxidative damage, which can
interfere with the oxygen-carrying capacity of haemoglobin, or by modification of
cell surface proteins (e.g., mefenamic acid), which can lead to loss of ‘self’ antigens
(cell surface markers that identify the cell as part of the self, as opposed to foreign)
and subsequent destruction by white blood cells (Figure 5.9).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 84 Bioanalytical Tools in Water Quality Assessment

Figure 5.9 Spectrum of haematotoxicity from the molecular to the organism level.

5.6.2 Immunotoxicity
Broadly defined, immunotoxic agents adversely affect the immune system, which
protects the organism against pathogens and tumours. The immune system
comprises numerous lymphoid organs (e.g., bone marrow, thymus, spleen, lymph
nodes) and cell populations with a variety of functions. Antigen recognition is the
cornerstone of the immune system. Antigens, usually protein or polysaccharide
‘signatures’ of foreign material, are recognised by specific antibodies, which
subsequently initiates an immune response.
There are two types of immune response: innate and adaptive. The innate
immune system is non-specific and is the body’s primary defence mechanism. It
relies on a variety of proteins (called the complement system) and involves
several immune cells, such as natural killer cells, macrophages and neutrophils.
Natural killer cells release cytokines and cytolytic chemicals that destroy the
target cell. Macrophages and neutrophils are phagocytic cells that eliminate most
microorganisms through the release of ROS.
The adaptive (or ‘acquired’) immune system is an antigen-specific response
triggered by the innate immune system. In simple terms, immune cells learn
to recognise the invading pathogen and deploy a more sophisticated set of
specialised cells, such as helper T-cells and killer T-cells. Helper T-cells
secrete cytokines and help direct the immune response depending on the
nature of the threat. Killer T-cells bind to the target cell and release the
content of cytolytic granules, which contain cytokines, perforins and other
enzymes, on the target cell, a process called degranulation. Once degranulated,
the killer T-cell releases the dying target cell and moves on to kill other
target cells.
The immune system can also call upon other specialised cells when fighting
inflammation, such as basophils and mastocytes. When stimulated, these cells
degranulate to release histamine, proteoglycans, proteolytic enzymes, leukotrienes
and cytokines. These chemicals attract other immune cells.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Toxicity pathways of chemicals in humans 85

Figure 5.10 Spectrum of immunotoxicity from the molecular to the organism level.

The immune system must strike a delicate balance between excessive and
insufficient immune response. Toxicant exposure can result in dysfunctions of the
immune system:
• Immunosuppression results in reduced efficacy of the immune response (i.e.,
impaired resistance), while immunostimulation stimulates the immune
system, which may result in excessive immune response (Figure 5.10). A
very wide range of toxicants have been shown to suppress or stimulate the
immune system, including polychlorinated biphenyls (PCB), polycyclic
aromatic hydrocarbons (PAH), pesticides, metals, solvents, hormones,
pharmaceuticals and UV radiation. Some toxicants (e.g., sulphamethoxazole)
can stimulate immune cells directly by binding to their membrane receptors.
• Hypersensitivity reactions (allergies) result from the immune system
responding in an exaggerated or inappropriate manner (e.g., penicillin).
Hypersensitivity has been linked with exposure to industrial chemicals,
metals, solvents and pharmaceuticals.
Autoimmune disease occurs when the reactions of the immune system are
directed at the body’s own tissues. It is more difficult to establish a clear link
between toxicant exposure and autoimmunity. Some chemicals have, however,
been implicated in chemical-induced autoimmunity. These include some
pharmaceuticals, plastic monomers (vinyl chloride), mercury and some pesticides
(e.g., hexachlorobenzene). Interaction between toxicants and endogenous proteins
can sometimes result in the altered protein no longer being recognised as own
tissue (e.g., penicillin).

5.6.3 Neurotoxicity
A neurotoxicant is a toxic chemical that affects the development, function or
viability of neurons and the nervous system (Figure 5.11). The nervous system
coordinates numerous functions in the organism via neurons and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 86 Bioanalytical Tools in Water Quality Assessment

Figure 5.11 Spectrum of neurotoxicity from the molecular to the organism level.

neurotransmitters. There are two cell populations in nervous tissues: the neurons,
which specialise in generation, reception and transfer of information (transmitted
by neurotransmitters such as acetylcholine and epinephrine); and glial cells,
which provide support and nutrition to neurons.
The four most common targets of neurotoxicants are the neuron, the axon (the
neuron’s projection towards other neurons), the myelinating cell and the
neurotransmitter system.
• Neuronopathy: Although the neuron is similar to other types of cells in many
respects, some features of the neuron are unique, and provide distinctive
vulnerabilities. Some of those unique features are a high metabolic rate, a
long cellular process supported by the cell body (the axon) and an excitable
membrane that is rapidly depolarised and repolarised. A large number of
chemicals are known to result in toxic neuronopathy, including metals
(aluminium, arsenic, lead, manganese, mercury, methyl mercury), industrial
chemicals (trimethyltin), pharmaceuticals and solvents.
• Axonopathy: Some toxicants can physically damage the axon, resulting in
a degradation of neuron transmission. Many chemicals have been linked
to axonopathy, including metals (gold and platinum), alkaloids,
pharmaceuticals, industrial chemicals (acrylamide), solvents and pesticides.
• Myelinopathy: Myelin provides electrical insulation of neuronal processes,
and its absence leads to a slowing and/or aberrant conduction of electrical
impulses. Some toxicants can interfere with myelin maintenance or function.
• Neurotransmitter-associated toxicity: A wide range of naturally occurring
toxins as well as pesticides and pharmaceuticals can inhibit normal
neurotransmitter function. Organophosphate and carbamate pesticides, for
example, inhibit the enzyme acetylcholinesterase, which is responsible for
recycling the neurotransmitter acetylcholine.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Toxicity pathways of chemicals in humans 87

There are several morphological idiosyncrasies of the nervous system. Some, such a
mesh of endothelial cells called the ‘blood–brain barrier’ (Figure 5.3), provide an
additional barrier to toxicants reaching the central nervous system. On the other
hand, some make it more sensitive to toxicants. The unusual cell morphology of
neurons, for example, which are very elongated rather than small and spherical,
creates extraordinary demands on protein synthesis and transport of vesicles and
organelles. The myelin sheet, which is rich in lipids and dependent on the proper
function of a number of membrane-associated proteins, is also a sensitive target
site for toxicants. Finally, the high energy requirements of neurons make them
extremely sensitive to interruptions in the supply or oxygen or glucose that can
be caused by toxicants such as cyanide and carbon monoxide as well as very
sensitive to mitochondrial toxicants.
Astrocytes (a type of glial cell) play an important role in defence against
neurotoxicants. They are activated by hypoxia and inflammation and have far
greater antioxidant abilities than neurons, protecting axonal structure and
processes. Intercellular communication between astrocytes and neurons also
involves organelle exchange, including the transfer of healthy astrocytic
mitochondria to adjacent neurons to restore degraded mitochondrial function, in
exchange for damaged and defective neuronal mitochondria, which are then
broken down in the astrocytes (a process called ‘mitophagy’).

5.6.4 Endocrine toxicity

Endocrine glands are collections of specialised cells that synthesise, store and
release their secretions (hormones) directly into the bloodstream. As sensing and
signalling devices capable of responding to changes in the internal and external
environments, the hormone system coordinates a multiplicity of activities that
maintain homeostasis. Disruption of normal endocrine function can thus have a
wide range of effects, potentially affecting many different organ systems
(Figure 5.12). The main endocrine glands of the body are the following:
The pituitary gland is a small protrusion off the hypothalamus at the base of the
brain, which secretes hormones (including the so-called ‘trophic hormones’ to other
endocrine glands) under stimulation of the hypothalamus. The pituitary releases
hormones related to growth (growth hormone), lactation (prolactin), reproductive
function (gonadotropins and corticotropic hormones) and thyroid activity
(thyroid-stimulating hormone).
The adrenal glands are located above both kidneys and are mainly responsible for
regulating the stress response through the synthesis of corticosteroids (cortisol and
aldosterone) and catecholamines (epinephrine, norepinephrine and dopamine).
They affect glucose metabolism (glucocorticoids) and reproduction (androgens,
estrogens and progestins).
The pancreas produces digestive enzymes and hormones that regulate glucose
metabolism (insulin, glucagon and somatostatin).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 88 Bioanalytical Tools in Water Quality Assessment

Figure 5.12 Spectrum of endocrine effects from the molecular to the organism level.

The thyroid gland secretes the thyroid hormones thyroxine (T4) and
triiodothyronine (T3) under stimulation of the pituitary gland. Thyroid hormones
increase metabolic rate and glucose availability, stimulate new protein synthesis,
heart rate, cardiac output and blood flow, and increase neuronal development in
young animals. The thyroid gland also produces calcitonin, which is involved in
calcium homeostasis.
The parathyroid gland produces hormones involved in calcium homeostasis
(parathyroid hormone, calcitonin and vitamin D) under stimulation of
calcium-sensing receptors. This unique feedback system is sensitive to toxicants
similar to calcium ions (e.g., aluminium).
The gonads (testes in males, ovaries in females) produce sex hormones
(androgens, estrogens) and progestogens (progesterone). Hormone production
and gametogenesis in the gonads is under direct pituitary hormonal control.
Gonads are sensitive to toxic substances because gametogenesis relies on rapidly
dividing cells, which are often vulnerable to chemical destruction. The blood–
testes barrier controls the entry of large molecules and toxicants into the
seminiferous tubules, where gametogenesis occurs.
There are four main mechanisms of endocrine toxicity (Figure 5.12):
• Excessive stimulation can cause hyperplasia (excessive cellular development)
and hypertrophy (gross enlargement) of individual endocrine organs, and
eventually lead to tumour development.
• Interference with hormone synthesis or secretion. For example, some
pharmaceuticals (e.g., sulphonamides, 2,4-dihydroxybenzoic acid,
aminotriazole, antipyrine) and pesticides (e.g., amitrole) interfere with
thyroid hormone synthesis.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Toxicity pathways of chemicals in humans 89

• Increased hormone catabolism (destruction). Toxicants that induce liver

enzymes (e.g., phenobarbital, benzodiazepines, DDT, chlorinated
hydrocarbons) can increase the rate of conjugation and excretion of
hormones such as T3 and T4.
• Interference with hormone signalling (endocrine disruption). Hormones act
by binding to highly specific hormone receptors (e.g., estrogen receptor,
androgen receptor, progesterone receptor, thyroid receptor), which results
in a biochemical cascade that eventually triggers the intended effect.
Toxicant-induced interference with hormone signalling can result in
erroneous endocrine communication and/or interfere with the complex
system of hormonal feedback loops. Some toxicants can mimic hormone
activity (agonists) while others can inhibit normal hormonal function
(antagonists). In some instances, this effect is intentional (e.g.,
pharmaceuticals used for birth control such as ethinylestradiol or
levonorgestrel), but toxicants can also be unintentional endocrine disruptors
(e.g., industrial chemicals like bisphenol A, phthalates). Interference or
mimicry of sex steroids (estrogens and androgens) can also significantly
affect reproduction. Most of the activity of the endocrine system has
sensitive feedback loops, and toxicants that affect or mimic hormones often
affect multiple endocrine glands.

5.6.5 Reproductive toxicity

The purpose of the reproductive system is to produce good quality gametes, capable
of fertilisation and producing a viable offspring, which can in turn successfully
reproduce. This requires a large number of complex processes, orchestrated in a
precise order for optimal performance at different life stages. The fact that
chemicals can adversely affect reproduction in males and females is not a new
notion. One only has to look at the importance of drugs as contraceptives to realise
how sensitive the reproductive system can be to external chemical influences.
Endocrine communication is critical to proper reproductive function, and toxicants
that can adversely affect endocrine glands also generally result in reproductive
toxicity. A wide range of environmental chemicals are known to mimic or inhibit
androgens (e.g., trenbolone, vinclozolin, procymidone, linuron, p,p′ -DDE,
phthalates), estrogens (e.g., methoxychlor metabolites, ethinylestradiol, bisphenol
A, nonylphenol, DDT) or progestogens (e.g., levonorgestrel, norethindrone).
Exposure to these hormone mimics can adversely affect reproductive functions.
Sex hormones (androgens and estrogens) are particularly important in foetal
reproductive organ development, puberty and sexual maturation. These stages are
thus inherently susceptible to endocrine disruption. Toxicants such as PCBs,
DDT/DDE, brominated flame retardants, dioxins, hexachlorobenzene, personal
care products and heavy metals have been linked to reproductive abnormalities,
although their exact mechanism is often unknown.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 90 Bioanalytical Tools in Water Quality Assessment

The female reproductive cycle relies on subtle hormonal communication

between the pituitary and ovarian secretions of progesterone and estrogens. These
hormones determine ovulation and prepare the female accessory organs to receive
male sperm. Disruption of these hormonal cues can lead to infertility. Disruption
of the luteinising hormone (LH) surge, for example, by the pesticides
chlordimeform and N-methyldithiocarbamate prevents or delays ovulation, which
has been shown to cause infertility in laboratory animals.
Male reproductive processes also rely on carefully orchestrated hormonal
communication through the hypothalamic–pituitary–testes axis, and endocrine
disruption can also affect male reproduction. The vast majority of male
reproductive toxicants affect sperm production (spermatogenesis). Some of these
toxicants act via indirect routes, such as nutrient disruption after exposure to zinc,
or increased steroid clearance by the liver due to carbon tetrachloride exposure.
Most toxicants affecting spermatogenesis, however, do so via a direct effect on
testis or spermatogenesis itself by interfering with or damaging Sertoli cells (e.g.,
phthalate ester metabolites, dibromoacetic acid, m-dinitrobenzene) or interfering
with energy production in sperm cells (e.g., chlorosugars, epichlorohydrin).
Reproductive toxicants can also affect fertilisation and implantation as successful
pregnancy depends heavily on complex and subtle hormonal communication. These
processes are also susceptible to endocrine disruption (e.g., some pharmaceuticals
used to terminate pregnancies interfere with progesterone synthesis).

5.7 CONCLUSIONS
The purpose of this chapter was to give the reader an appreciation of the normal
function and significance of different organs and organ systems in the human
body, and to describe toxic effects and define mechanisms of toxicity. Humans
exposed to contaminated water can exhibit a wide variety of tissue-, organ- and
organ system-level responses, many of which can be traced back to the effect of
the toxicant at the molecular or cellular level, illustrating the concept of toxicity
pathways introduced in the previous chapter. Monitoring those molecular or
cellular events using in vitro bioassays may therefore provide a simple screening
method to detect toxicants in water, and Chapter 10 reviews in vitro methods
available to measure toxic effects discussed here.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Chapter 6
Adverse outcome pathways of
chemicals in aquatic organisms

6.1 INTRODUCTION
Ecotoxicology is the study of the effects of toxic substances and other stressors on
the structure and function of ecosystems. The discipline has evolved relatively
independently of human toxicology but with the concept of adverse outcome
pathways (AOPs) and the recognition that chemical stressors perturb cellular
functions in a similar manner, the two fields have come closer together.
Aquatic organisms can bioaccumulate pollutants via direct uptake from water
across skin or gills and from their diet. The level of bioaccumulation is dependent
on the physicochemical properties of the chemicals with lipophilic chemicals
accumulating to a greater extent than hydrophilic chemicals.
The aquatic component of the ecosystem is in perpetual contact with the other
environmental compartments including those of air, sediment and soil, and the
ecotoxicological principles for all compartments are similar.
A crucial driver of the discipline of ecotoxicology was Rachel Carson’s ‘Silent
Spring’, published in 1962 (Carson, 1962). It denounced the negative impacts of
organochlorine pesticides such as DDT on humans and the environment. A
comprehensive treatise on ecotoxicology can be found in various textbooks, with
recommendations for Newman’s ‘Fundamentals of Ecotoxicology’ (2019),
Walker’s ‘Principles of Ecotoxicology’ (2012) and Landis and Yu’s ‘Introduction
to Environmental Toxicology: Impacts of Chemicals Upon Ecological Systems’
(2004).

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0091

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 92 Bioanalytical Tools in Water Quality Assessment

In this chapter, we focus on some of the basics of ecotoxicology and mechanistic

approaches to aquatic toxicology and ecosystem health. This chapter builds on the
foundation laid in Chapter 4 and uses the AOP concept in a conceptual manner
rather than detailing individual AOPs. The purpose of this chapter is to
understand the toxicity pathways of chemicals in aquatic organisms sufficiently
to select appropriate in vitro tools for water quality assessment. Since the first
edition of this book, which appeared shortly after the introduction of the AOP
concept to this field of research, more than 350 AOPs have been developed
that contain .1000 key events and almost 2000 key event relationships
(aopwiki.org, accessed on 20 December 2020). Many of the available AOPs
describe adverse outcomes on fish health with only a few AOPs addressing algae
and invertebrates.

6.2 FROM THE CELLULAR LEVEL TO THE ECOSYSTEM

To understand the effects of chemicals on ecosystems, we need a mechanistic
understanding on how chemicals act on individual organisms as well as an
understanding of the complex interactions occurring in ecosystems.
Comprehensive field studies in natural ecosystems are limited to cases of
unintentional contamination because obviously planned exposure studies and the
intentional contamination at the ecosystem level are ethically unacceptable. The
majority of ecotoxicology research has therefore focused on laboratory studies
with single species that are representative of different trophic levels. An overview
of commonly used test species and endpoints is given in Chapter 3. While
integrated assessment in ecosystems may be more relevant than laboratory
studies, they are less controlled and can only detect overall apical responses that
are difficult to interpret. Studies with simplified model ecosystems (e.g., meso-
and macrocosm) are viable alternatives to real ecosystem assessments because
aspects of food chains and ecosystem interaction can be integrated (Chapter 3).
Further, controlled model systems have the benefit of being able to include
quality assurance/quality control (QA/QC) measures such as negative and
positive controls as well as replicate experiments (Chapter 11). Despite these
obvious advantages, model ecosystem studies are very expensive and typically
only conducted in higher tiers of pesticide risk assessment.
If single organisms are tested, apical endpoints such as developmental
dysfunction, reproductive failure and death are often measured (Chapter 3). Acute
toxicity tests are performed over a timeframe of several hours to several days. To
be protective of the entire ecosystem, one must extrapolate the information
gained from acute tests to safe levels for the entire population and from
short-term to long-term exposure (Figure 6.1, top panel). Such extrapolation can
be accomplished by applying extrapolation and uncertainty factors (Chapter 2).
This process must be completed for many aquatic species to obtain an idea about
the species sensitivity distributions (SSD). Generally, the lower fifth percentile
(HC5) of a distribution of chronic ‘no observed effect concentrations’ (NOEC) is

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Adverse outcome pathways of chemicals in aquatic organisms 93

Figure 6.1 Comparison of the traditional holistic approach for aquatic toxicity
assessment of chemicals and the novel mechanistic approach based on AOPs.

considered sufficiently low to protect ecosystem integrity and functioning

(Posthuma et al., 2002), although concentrations as low as one percentile may be
needed for chemicals known to bioaccumulate.
In the AOP approach the cellular level response pathway, which is often highly
conserved in different species, is used to obtain a mechanistic picture of the species
sensitivity, which facilitates extrapolation from cellular responses to expected
effects on ecosystems (Figure 6.1, bottom panel).
Although AOPs are central organising principles, the quantitative link to
individual organism responses must still be established to allow the
implementation of AOPs to population level models (Kramer et al., 2011).
Species sensitivity differences can be related to differences in toxicokinetics,
specifically the metabolic capacity of different species, and to differences in the
toxicity pathways. Species with shared ancestry will have similar AOPs due to
the close homology of genes (Celander et al., 2011). Read-across is thus possible
for highly homologous species. Read-across from human drug targets to adverse
effects in fish is even possible for evolutionarily well-conserved targets.

6.3 ADVERSE OUTCOME PATHWAYS FOR AQUATIC

ORGANISMS
Chapter 3 introduced the most fundamental members of aquatic food chains –
primary producers, invertebrates and vertebrates. Common apical endpoints
measured for these organism groups are growth inhibition, immobilisation and
mortality, respectively. In the following section, the concept of AOPs for aquatic
organisms is illustrated using examples from these three taxonomic groups. For
each organism group, the AOPs for non-specific toxicity are compared with one
selected specific mode of action (MOA), using specific toxicity of herbicides,
insecticides and estrogens as illustrative examples for algae, water fleas and
fish, respectively.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 94 Bioanalytical Tools in Water Quality Assessment

6.3.1 Adverse outcome pathways for algae

6.3.1.1 Baseline toxicity
We have learnt in Chapter 4.4.1 that baseline toxicity is the minimal toxicity that any
compound exhibits. This is of high relevance in the aquatic environment, where all
chemicals present act together in a concentration additive manner, potentially
resulting in an appreciable level of baseline toxicity. The molecular targets of
baseline toxicants are biological membranes, which can become leaky, disrupting
their structure and function (van Wezel and Opperhuizen, 1995) (Figure 6.2). In
algae, such membrane disruption will indirectly affect photosynthesis efficiency,
as the electron transfer chain of algal photosystems is embedded in the
chloroplast membrane.
The biologically effective concentration of baseline toxicants is the same for all
chemicals, that is, every chemical has the same baseline effect once it has crossed
the membrane. The apparent differences in potency between different chemicals
thus stem exclusively from differences in uptake and other toxicokinetic
processes. The more hydrophobic a chemical is, the higher its degree of
accumulation in algae and consequently the lower the dosed concentration
required to trigger the baseline-toxic effect.
A feature of baseline toxicity is that it is a reversible MOA. Once the algae are
transferred to clean water, the baseline toxicants can be depurated, enabling the
algae to recover from the toxic stress, provided that no irreversible secondary
effects have occurred. Furthermore, baseline toxicity-mediated photosynthesis
inhibition occurs at much higher chemical concentrations than specific inhibition
of photosynthesis. Nevertheless, at the organism level, reduced photosynthesis
leads to lowered energy for cellular growth causing cells to stop dividing. At the
population level, this effect manifests itself by a slower population growth rate or
decline in population size (Figure 6.2).

Figure 6.2 Adverse outcome pathway for baseline toxicants in algae. Adapted from
Ankley et al. (2010).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Adverse outcome pathways of chemicals in aquatic organisms 95

Figure 6.3 Adverse outcome pathway for herbicides in algae.

6.3.1.2 Inhibition of photosynthesis by herbicides

Herbicides are used for weed control but they also have the negative effect of
inhibiting photosynthesis in non-target plants and green algae. Important classes
of herbicides are the triazines, such as atrazine, simazine and irgarol, and
phenylurea herbicides, such as diuron and isoproturon. Both groups of herbicides
bind to a quinone-binding site on photosystem II, blocking photosynthetic
electron transfer (Figure 6.3). As a result, no energy in form of ATP is produced,
causing a loss of chlorophyll, slower growth, smaller cell size and ultimately
death of algal cells. The ultimate observable effects are the same as for baseline
toxicity, but these specific effects occur at chemical concentrations that are orders
of magnitude lower than baseline toxicity.
Vogs and Altenburger (2016) developed a combined toxicokinetic/
toxicodynamic model that could differentiate well between different AOPs with
effect progression, that is, time to progress from molecular initiating event
(MIE) to apical effect, being shortest for specific photosynthesis inhibitors
followed by chemicals that cause oxidative stress and those which interfere with
lipid synthesis.

6.3.2 Adverse outcome pathways for invertebrates

6.3.2.1 Baseline toxicity
Baseline toxicity occurs in aquatic invertebrates such as the water flea Daphnia
magna at the same internal concentration as in algae. As water fleas are
metabolically more active, the baseline toxicants can be better detoxified and thus
the apparent toxicity in relation to the external exposure concentration may be
lower. Apart from this one exception, the molecular interactions of baseline
toxicants in cell membranes of water fleas and algae are the same (Figure 6.4).
Mitochondrial dysfunction is also one of the characteristics of baseline toxicity
(Vinken and Blaauboer, 2017). Despite this similarity in MIE, the apical

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 96 Bioanalytical Tools in Water Quality Assessment

Figure 6.4 Adverse outcome pathway for baseline toxicants in the water flea
Daphnia magna. Adapted from Ankley et al. (2010) and Vinken and Blaauboer (2017).

endpoints observed in water fleas differ from algae. The effects in water fleas will
mainly manifest themselves as reduced ventilation/respiration rate, which equates
to immobilisation and eventually death at the organism level.

6.3.2.2 Activity of insecticides

Insecticides such as organophosphates or carbamates inhibit the enzyme
acetylcholinesterase (AChE). AChE is responsible for breaking down the
neurotransmitter acetylcholine. Inhibiting this breakdown process induces a
constant chemical signal firing across the synapse, leading to over-excitation of
nerves. This over-excitation is initially expressed as rapid movement,
consequently leading to excessive expense of energy before the ultimate
responses of immobilisation and death occur. Invertebrates such as insects and, in
our example, water fleas are particularly sensitive to organophosphates. First, in
terms of toxicokinetics, organothiophosphates such as diazinon must be
metabolically activated to their bioactive form, for example, diazoxon, in order to
induce their MIE of binding to AChE (Figure 6.5). This metabolic activation is
particularly efficient in water fleas with little detoxification but strong activation

Figure 6.5 Adverse outcome pathway for the inhibition of acetylcholinesterase

(AChE) in the water flea Daphnia magna. Adapted from Russom et al. (2014).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Adverse outcome pathways of chemicals in aquatic organisms 97

(Kretschmann et al., 2011). Second, in terms of toxicodynamics, the AChE of

Daphnia magna is particularly sensitive to organophosphates (Kretschmann
et al., 2012). Carbamates differ from organophosphates in that metabolic
activation is not required for the binding, which is reversible (Jeon et al., 2013).
Organophosphates bind covalently to the enzyme, causing it to hydrolyse further,
which makes the binding irreversible.
Russom et al. (2014) developed a formal AOP for inhibition of AChE leading to
acute mortality for various taxa, recognising the highly conserved MIE but also
species-specific differences in toxicokinetics and AChE binding domains.

6.3.3 Adverse outcome pathways for fish

6.3.3.1 Baseline toxicity and mitochondrial dysfunction
Baseline toxicity affects the same molecular and cellular mechanisms in fish as in
the invertebrate and algal species discussed above, but the consequences are
somewhat different (Figure 6.6). The initial effect of baseline toxicity is a loss of
equilibrium. Fish that normally swim in fast flowing rivers against the current can
no longer maintain their position. Non-targeted swimming may lead to increased
vulnerability to predators and difficulty in mating. At higher concentrations,
baseline toxicity leads to narcotic effects and eventually to death.
Mitochondrial dysfunction is also one of the key features of baseline toxicity but
there are also specific effects on mitochondria that lead to the same symptoms at
much lower concentrations of chemicals. Souders et al. (2018) demonstrated how
oxidative respiration in fish embryos can be used to identify specific MIEs such
as uncoupling of oxidative phosphorylation, specific inhibition of the electron
transfer chain and inhibition of ATP synthesis.

6.3.3.2 Reproductive toxicity

Many pathways can lead to reduced reproductive success or even reproductive
failure (Figure 6.7). Natural estrogens and xenoestrogens (e.g., nonylphenol and

Figure 6.6 Adverse outcome pathway for baseline toxicants in fish.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 98 Bioanalytical Tools in Water Quality Assessment

Figure 6.7 Adverse outcome pathway for reproductive effects in fish. COX =
cyclooxygenase. Adapted from Ankley et al. (2010), Martinovic-Weigelt et al.
(2017) and Martyniuk et al. (2020).

bisphenol A) occurring in environmental waters can bind to estrogen receptors in

fish, triggering a number of cellular responses that would normally be triggered
by endogenous hormones. Inducing such cellular responses at the wrong time can
cause both structural and functional problems in the fish. One major problem of
estrogenic effects is that they can induce feminisation of male fish. At the cellular
level, this translates to feminisation of the phenotype and production of
vitellogenin (an egg yolk protein precursor produced in liver cells) in both
females and males. In male fish, estrogenicity can further induce notable changes
in sexual development such as the formation of ovotestis (egg cells in the testis
tissue) and intersex. Such changes evidently lead to reproductive failure (or
reduced reproductive success) and can eventually wipe out entire populations
(Kidd et al., 2007). Anti-androgenic chemicals can similarly antagonistically
impact the androgen receptor, reducing testosterone levels and cause reproductive
failure due to damaged or immature sperm. Cyclooxygenase (COX) inhibitors
disrupt prostaglandin synthesis, leading also ultimately to reproductive toxicity in
female fish by preventing oocyte maturation.

6.4 USING IN VITRO ASSAYS TO UNDERSTAND TOXICITY

PATHWAYS IN AQUATIC LIFE
A general introduction as to how in vitro assays can be used for the elucidation of
toxicity pathways is given in Chapter 9 and applications for water quality
monitoring are discussed in Chapter 10. Specifically, for aquatic organisms, one
would choose cell lines and reporter gene assays derived from the tissues of
aquatic biota such as fish cell lines (Schirmer, 2006).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Adverse outcome pathways of chemicals in aquatic organisms 99

However, given the high degree of conservation of cellular pathways, it is often

acceptable to use a construct from different species. The yeast estrogen screen
(Routledge and Sumpter, 1996), for example, is based on a yeast cell line
transfected with a human estrogen receptor, but the assay is a widely applied
bioanalytical tool for assessing estrogenicity in water, where fish are the primary
target for EDCs. Likewise, reporter gene assays based on human and zebrafish
nuclear receptors yielded similar responses of water samples (Neale et al., 2020c).

6.5 CONCLUSIONS
The concept of AOPs bridges the initial cellular level effects of toxicants with the
organism level outcome. The application of this concept to aquatic organisms
demonstrates the ability of in vitro bioassays to detect subtle toxic effects in
aquatic organisms, thereby highlighting the potential of in vitro bioassays to
replace, reduce and/or refine whole organism testing.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 7
Dose–response assessment

7.1 INTRODUCTION
The central paradigm in toxicology is that ‘the dose makes the poison’. In fact, all
substances, even something as innocuous as table salt, are poisonous if taken in
sufficient quantity. Understanding the dose–response relationship can therefore,
quite literally, be a matter of life or death. A dose–response curve is the
mathematical display of this paradigm. From a dose–response curve we can
derive various descriptors of effect, for example, the lethal dose LD50 or the
highest concentration where no effect was observed, the no observed effect
concentration (NOEC).
A dose is the total quantity of a chemical delivered to a test animal or system.
We know doses from pharmaceuticals we take, for example, one tablet or 200 mg
of aspirin if we have a headache. In toxicity testing it makes more sense to report
a normalised dose, for example, dose per kg of body weight of the test animal.
For in vitro bioassays, the dose is difficult to quantify because it will depend on
the number of cells and the medium volume. Therefore, concentrations are the
typical dose-metrics in cell assays. A concentration is the mass or molar amount
of a chemical divided by the volume of the test system.
This chapter introduces dose–response assessment in general terms before we
focus on concentration–response curves (CRC) in in vitro assays. We
differentiate between the activation of a certain response on the level of
molecular initiating event or key event (Chapter 4) and cytotoxicity. We present
various options for modelling CRCs and deriving benchmark doses (BMD) or

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0101

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 102 Bioanalytical Tools in Water Quality Assessment

benchmark concentrations (BMC). Effect concentrations of water samples are

expressed in units of relative enrichment factors (REF) instead of concentrations
because water samples contain undefined complex mixtures of chemicals but can
be translated into bioanalytical equivalent concentrations (BEQ) for an easier
analogy to the effects caused by single chemicals.

7.2 DOSE–RESPONSE ASSESSMENT

7.2.1 Dose–response curves
A dose–response curve plots the response (e.g., death, or a more subtle sublethal
effect) of a population under study (e.g., rats, mice) against increasing dose of the
test chemical applied to the system (Figure 7.1). The dose–response curve looks
like a saturation curve on a linear dose scale but is typically sigmoidal on a
logarithmic scale.
On the lower left of the log dose–response curve, at very low doses, there is no
measurable response in the test system. It is important to note, however, that a close
look at the lower end of the dose–response curve on a linear dose scale reveals that
the relationship between dose and response is in fact linear at low doses when the
dose is not in logarithmic scale (Figure 7.1).
Figure 7.1 shows that as the dose is increased, there is initially no detectable
increase in the monitored response. Eventually, a detection threshold is reached
above which the response quickly increases with increasing dose in a near linear
manner (actually log-linear, as the x-axis is on a log-scale). The dose–response
curves then levels to a maximum of 100%. This dose–response curve is common
to all biological responses, although it can sometimes be affected when more than
one type of effect co-occur in the test system as is discussed in more detail for
cytotoxicity and activation of an effect.

Figure 7.1 A dose–response curve with the dose on the x-axis and the response
(in %) on the y-axis. If the dose is plotted on a logarithmic scale, the typical
sigmoidal form is visible. The figure on the right depicts the low-response level
linear portion of the dose–response curve.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Dose–response assessment 103

7.2.2 Dose benchmark values

Several important parameters can be described by the dose–response curve. The
most common descriptor is the LD50, the dose that is lethal to 50% of the
organisms in a test population (Figure 7.2). The highest dose tested that is not
statistically different from the negative control is referred to as the ‘no observed
adverse effect level’ (NOAEL), while the next dose above, that is, the lowest
dose tested that is statistically different from the negative control, is called the
‘lowest observed adverse effect level’ (LOAEL). In general term, these dose
values are called point of departure (POD) in risk assessment for extrapolation to
safe levels for sensitive human populations.
A major issue with LOAEL and NOAEL as expressions of toxicity is that they
are dependent on the experimental design (i.e., dose spacing) and variability of
the data, which could be affected by the number of replicates for each dose or the
inherent variability of the test system. Therefore, parameters that are calculated
from the whole dose–response relationship are preferred expressions of toxicity,
such as the BMD. The BMD10 for example is the dose calculated from the
dose–response curve required to produce 10% of the maximum response.
Another common parameter is the benchmark dose level (BMDL), which is the
lower confidence boundary of the BMD. The BMD is preferred over NOAEL
and LOAEL because it does not depend on the experimental doses and takes into
account the shape of the full dose–response curve (black line in Figure 7.2). A
limited number of data points or high variability in the controls can cause the
NOAEL to be much higher, as high as 40–50% of the response, while for the
same situation the BMDL becomes smaller due to the higher uncertainty (Haber
et al., 2018), which is more precautious from a risk assessment perspective.

Figure 7.2 A typical dose–response curve with the logarithm of the dose on the x-axis
and the response (in %) on the y-axis. (a) Individual response data points and
derivation of no observed adverse effect level (NOAEL) and lowest observed
adverse effect level (LOAEL). The grey area represents the variability of the
negative control. (b) The same data plotted as averages and described with a
dose–response model used to derive the lethal dose (LD50) and the benchmark
dose (BMD10).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 104 Bioanalytical Tools in Water Quality Assessment

7.2.3 Continuum of toxicity

As described in Chapter 4, in the case of chemical-mediated toxicity, effects at low
levels of biological complexity (e.g., molecular and cellular) can translate into
higher-order effects at the tissue, organ and eventually organism level, depending
on the severity of the effect and the ability of compensation and repair
mechanisms to cope with the damage. This continuum of toxicity can also be
plotted on a dose–response graph, which shows that as the dose increases so does
the level of biological complexity affected (Figure 7.3).
Let us delve into this continuum by using the example of a liver toxicant that
interferes with a liver enzyme. At low dose, the toxicant has no noticeable effect
on the liver enzyme. As the dose increases, increasing amounts of the enzyme is
inactivated by the toxicant. Initially, this effect has no further consequence
beyond the molecular level, but as more and more of the enzyme is affected, the
increasing molecular dysfunction starts to affect cellular health and finally,
cytotoxicity becomes evident. The organ, in this example the liver, can deal with
a certain amount of cellular damage and initially the death of a few individual
liver cells has no further impact. But above a certain dose, the cellular damage
exceeds the compensatory capacity of the organ, and organ damage occurs.
Eventually, sufficient organ damage occurs to negatively affect the whole
organism, ultimately resulting in death.
The continuum of toxicity is one of the reasons it is logical to attempt to detect
toxicants by their effect at the molecular and cellular level, as is done with in vitro
bioassays. Indeed, the molecular and cellular effect will occur at much lower doses
than those causing organ and/or organism-level effects detectable in in vivo tests.
Using a well-designed battery of in vitro bioassays therefore allows a sensitive
assessment of toxic potential. As previously mentioned, a response at the
molecular and cellular level does not necessarily translate into higher-order

Figure 7.3 Continuum of toxicity. As the dose increases, so does the level of
biological complexity that is affected by the toxic compound.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Dose–response assessment 105

effects, but in chemical-mediated toxicity, higher-order effects cannot occur without

preliminary molecular or cellular level effect.

7.3 CONCENTRATION–RESPONSE ASSESSMENT

7.3.1 ‘Concentration’ versus ‘dose’
In aquatic ecotoxicology and in cell-based bioassays, the exposure is usually
expressed as the concentration of chemical in the medium (e.g., ng test
compound per litre of water) instead of the dose (e.g., mg test compound per kg
body weight). The cellular concentration is the biologically effective
concentration (BEC), but we cannot easily measure this dose as is outlined in
more detail in Chapter 9. The concentration in the medium surrounding the cells
provides a better estimate of the BEC than any actual dose parameter (Kisitu
et al., 2019). The dose in an in vitro test would relate to the amount of chemical
added to the well of a microtitre plate. How much of that dose reaches the cells
depends on the number of cells and especially the volume of the medium. For the
same number of cells and the same dose added, there will be threefold increase in
concentration if the volume of the medium is reduced by a factor of 3. Therefore,
‘concentration’ is more accurate than ‘dose’ with aquatic test systems, and the
‘dose–response curve’ in cell-based bioassays is more commonly referred to as
the ‘concentration–response curve’ (CRC). More details on dose-metrics are
discussed in Chapter 9.4.1.

7.3.2 ‘Response’ can mean toxicity or effect

In order to avoid confusion, we are referring to any kind of biological effect as a
‘response’. All CRC can be fitted with the same mathematical models but,
depending on the bioassay, the response can refer to toxicity (lethality, inhibition
of growth, inhibition of cell viability) or to effects, which may include binding to
nuclear receptors, induction or inhibition of metabolic enzymes or activation of
adaptive stress responses. Effects do not necessary lead to toxicity and
manifestation of an effect in an in vitro bioassay such as a reporter gene assay
does not necessarily equate to effects on higher levels of biological organisation,
and it is inaccurate to use the term ‘toxicity’ when describing sublethal responses.
In the section below, we therefore describe the CRC equations in general terms
before discussing benchmark values for toxicity and effects independently.

7.3.3 Concentration–response modelling

There are many different approaches to fit the CRC to the raw data. Many parametric
models are available in the literature that have been successfully used to fit
sigmoidal logarithmic CRCs of various forms (Scholze et al., 2001). More
recently, Bayesian approaches have become popular for CRC modelling. While
they can fit even the noisiest data, such models might not be useful for the

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 106 Bioanalytical Tools in Water Quality Assessment

purpose of water quality monitoring. For water quality monitoring it is essential to

be able to compare between samples and to compare samples with individual
components’ CRCs as well as process large numbers of CRCs.
Among the many parametric models to choose from, the log-logistic equation
presented in Equation 7.1 (also called Verhulst or Hill equation) has the
advantage that it provides the median response concentration RC50 directly and
that the other parameters of the equation (min, max and slope) all have biological
relevance. Figure 7.4 depicts a CRC for bacterial cytotoxicity caused by the
pharmaceutical diclofenac. Note that because the response is inhibition of
bioluminescence, the RC50 is termed inhibitory concentration IC50 in this example.
max − min
Response (%) = min+ (7.1)
1 + 10slope · (log RC50 −log C)
where min is the minimum response (0% inhibition in the example in Figure 7.4),
max is the maximum response (100% inhibition in Figure 7.4), slope is the slope
of the curve (2.4 in Figure 7.4), log RC50 is the logarithmic value of the RC50
(logIC50 = −3.9 in this example, corresponding to an IC50 of 1.3 × 10−4 M in
Figure 7.4), and log C is the logarithmic concentration.
If the response ranges from 0 to 100%, the equation simplifies to
1
Response (%) = (7.2)
1 + 10slope · (log RC50 −log C)
The concentration associated with a y% response (RCy) can be calculated by
solving the equation for log RCy, as shown in Equation 7.3.


1 max − min
logRCy = logRC50 − · log −1 (7.3)
slope y − min
where logRCy is the logarithm of the response concentration RCy, logRC50 is the
logarithmic RC50 value from the curve equation, max is the maximal response

Figure 7.4 Typical concentration–response curve on the example of diclofenac in the

bioluminescence inhibition assays with Aliivibrio fischeri (Baumer et al., 2017).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Dose–response assessment 107

(usually 100%), min is the minimal response (usually 0%), y is the % response for
which you wish to determine the concentration (e.g., for RC10 the response is 10%),
and slope is the slope of the curve.

7.3.4 Concentration benchmark values

Concentration–response models allow the computation of the concentration
associated with any specific response level, where RCy is the effective
concentration required to cause y% of the response. The most commonly used
value is the RC50, the concentrations required to cause 50% of the response. In
assays for cytotoxicity, the response is usually referred to as lethal, and the
concentration referred to as lethal concentration (LCy). Likewise, inhibitory
concentrations (ICy) are used when the assay quantified an inhibitory response. In
the example in Figure 7.4, the IC50 identifies the concentration that causes 50%
inhibition of bioluminescence, while the IC10 is the concentration that causes
10% inhibition.
In parallel to the dose–response analysis (see Section 7.2.2), the ‘lowest observed
effect concentration’ (LOEC) is the lowest concentration tested that produce a
significant deviation from the control (unexposed bacteria in this example) and
the ‘no observed effect concentration’ (NOEC) is the highest tested concentration
that does not produce an effect significantly different from the control
(Figure 7.5). Just like the LOAEL and NOAEL, the LOEC and NOEC are
affected by the specifics of the experimental design (concentrations tested,
number of replicates and variability of the test system). They are therefore falling
out of favour, replaced by parameters calculated from the full CRC such as the
concentration required to produce 10% of the response (RC10) – commonly

Figure 7.5 Typical concentration–toxicity curve with diclofenac in the

bioluminescence inhibition assays with Aliivibrio fischeri as an example (Baumer
et al., 2017) with ‘no observed effects concentration’ (NOEC), ‘lowest observed
effects concentration’ (LOEC), and inhibitory concentrations IC10 and IC50
indicated. The grey bar depicts the variability of the controls (unexposed cells) with
a mean of 0.01 + 1.9% inhibition, which is enlarged in the inset.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 108 Bioanalytical Tools in Water Quality Assessment

Figure 7.6 Typical concentration–response curve (Equation 7.1) with EC10 and EC50
values indicated. The depicted data are typical of average + standard deviation of an
experiment with the ER CALUX bioassay (Leusch, unpublished).

referred to as EC10, LC10 or IC10 (‘effective’, ‘lethal’ and ‘inhibitory’

concentrations, respectively) or for reporter gene and transactivation assays
PC50, the concentration of a test chemical, at which the measured activity is
50% of the maximum activity induced by the positive control (PC). The big
advantage of synthesising the whole concentration–response relationship into a
few parameters is that the entire dataset is used rather than individual points (such
as LOEC and NOEC). This provides more confidence in the reliability of
the analysis.
Another example of a CRC with a reporter gene assay (Figure 7.6) identifies
the concentration of 17β-estradiol concentration required to produce 10% and
50% of the maximal luciferase induction in the ER-CALUX reporter gene
bioassay for estrogenicity, calculated as EC50 = 1.2 ng/L and EC10 = 0.06 ng/L
using Equation 7.1 with EC replacing RC.
Tox21 employs a specific data evaluation pipeline in R called ToxCast Analysis
Pipeline (tcpl) for high-throughput screening data, which includes three CRC
models. One is simply a constant effect over concentration (slope = 0), the
second is equivalent to Equation 7.1, and the third is a gain-loss model, which
can describe CRC that after saturation comes down again (Filer et al., 2016).
This inversion is typically caused by cytotoxicity masking the effect and
therefore, for the purpose of water quality assessment, we recommend using a
cut-off for cytotoxicity instead as described in Section 7.3.5, rather than using a
CRCs with a turning point.
The benchmark values used by Tox21 are indicated in Figure 7.7 using
the same data as Figure 7.6. The AC50 and AC10 are equivalent to EC50 and
EC10, respectively. The activity concentration at cut-off (ACC) is set at a
user-defined cut-off (usually 15%–20%). The activity concentration at baseline
(ACB) uses the two lowest concentration to derive the baseline band (BMAD)
accounting for the noise of the bioassay by using three times median absolute
deviation (MAD) over all responses of the two lowest concentrations around the
zero effect.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Dose–response assessment 109

Figure 7.7 Benchmark values applied in the ToxCast Analysis Pipeline (tcpl).
Modified on the example of the data from Figure 7.6 according to Filer et al. (2016).
ACC set to 20% cut-off.

7.3.5 Simultaneous effect and cytotoxicity in a

cell-based assay
In many cell-based assays there are two response types occurring at the same time –
the actual effect of interest, and cytotoxicity. Enzyme induction in a liver cell
line will, for example, increase with increasing concentration at first but decrease
at higher doses. This is a typical example of cytotoxic interference, where
measurable enzyme activity decreases at higher doses because the toxicant is
becoming cytotoxic and starts destroying the liver cells (Figure 7.8). The same
phenomenon can be observed in reporter gene assays and transactivation assays,
where the activation of the reporter gene or the transactivation decreases when
cells are dying. A related phenomenon is the cytotoxicity burst (Judson et al.,
2016), which occurs close to cell death and where all reporter genes can be
non-specifically induced in a last-ditch attempt to rescue the cell.
In bioassays where a specific response is normalised to the number of cells,
cytotoxicity (a decrease in cell number) will cause an increase in the response:cell

Figure 7.8 Example of a bioassay response with cytotoxicity interference. The dotted
line shows the theoretical effect but due to cytotoxicity (black line is cell viability), the
measured effect has an inverted U-shape. In this case, the effect can only be
evaluated up to the IC10 for cytotoxicity.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 110 Bioanalytical Tools in Water Quality Assessment

ratio, thus in some cases wrongly suggesting an increase in the specific response.
The CRC will in this case likely take an exponential shape because some of the
cells found dead (decreased cell count) at the end of the exposure period will
have produced a response in the earlier stages before succumbing to the cytotoxicity.
Figure 7.8 demonstrates how we can deal with CRCs that are compromised by
cytotoxicity interference. The effect data are removed above concentrations that
cause 10% cytotoxicity (IC10). The downside of the need to omit any cytotoxic
concentrations is that it often leaves an incomplete CRC, which means that the
classical models cannot be applied because there are not enough values above
50% effect included for curve fitting. This problem can be alleviated by focusing
on the linear low effect portion of the CRC.

7.3.6 Evaluating the linear portion of concentration–effect

curves
The higher effect levels are often compromised due to cytotoxicity in reporter gene
assays when environmental samples are tested. Therefore, it is sensible to evaluate
only the low-level linear range of the CRC. In addition, many water types such as
drinking water and recycled water as well as pristine surface water contain few
micropollutants and high effect levels are rarely reached.
CRCs are typically linear up to 30% effect or cytotoxicity (Escher et al., 2018)
and of the form of Equation 7.4, where y is the % effect or % cytotoxicity.
Variability of the negative controls in cell-based assays is usually much tighter
than it is in whole organism assays, and 10% effect is typically above the limit of
detection (LOD, see Chapter 11). It is thus possible to derive EC10 or IC10 values
for the linear CRCs using Equation 7.5 for ECy and Equation 7.6 for derivation
of the standard error SE of ECy.
y = slope · concentration (7.4)
y
ECy = (7.5)
slope
y
SE (ECy ) = · SE (slope) (7.6)
slope2
In practice, it is recommended to first derive the IC10 for cytotoxicity from the
linear portion of the CRC (% cytotoxicity = 100% – % cell viability). There are
various possibilities to measure cell viability in parallel to effect endpoints as
presented in Chapter 10. Then, only concentrations , IC10 are used for the
derivation of the EC10. In addition, the CRC is only linear up to about 30% effect
(Figure 7.1), and any concentrations that trigger .30% effect also have to be
omitted for CRC modelling (Figure 7.9).
Certain reporter gene assays, for example, those indicative of adaptive stress
responses or DNA repair, have no maximum for the CRC curves. For these types
of assays, the signal can be converted to an induction ratio (IR) by dividing the
signal by the signal of the negative control (Escher et al., 2014). The IR of the

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Dose–response assessment 111

Figure 7.9 Derivation of inhibitory concentration causing 10% cytotoxicity IC10 and
effect concentration causing 10% effect EC10 from the linear portion of the CRC.
Modified from Escher et al. (2018). Reprinted with permission by John Wiley and
Sons © 2018.

negative control is then by definition 1 and accordingly the linear portion of the CRC
has an intercept of 1 (Equation 7.7). The benchmark concentration is ECIRz, with an
IR of z, often z is 1.5, that is, 50% over the control and the EC is called ECIR1.5
(Equation 7.8 with standard error in Equation 7.9).
y = 1 + slope · concentration (7.7)
0.5
ECIR1.5 = (7.8)
slope
0.5
SE (ECIR1.5 ) = · SE (slope) (7.9)
slope2
As for bioassays where a maximum can be reached, we still have to deal with
cytotoxicity interferences as Figure 7.10 shows. Concentrations . IC10 for
toxicity must be removed and the CRC is typically only linear up to an IR of
approximately 3–4.
Depending on the bioassay, the maximum response can be at IR of 2 up to over
100. For a response where the maximum IR reaches 6, then the ECIR1.5 is equivalent
to the EC10, but if the maximum IR is 2, then the ECIR1.5 is equivalent to the EC50.
Hence, direct comparisons between EC values becomes cumbersome but the
bioanalytical equivalency concept outlined in Section 7.5 can help us to
overcome this difficulty.

Figure 7.10 Derivation of effect concentration triggering an IR of 1.5 ECIR1.5 from the
linear portion of the CRC.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 112 Bioanalytical Tools in Water Quality Assessment

7.3.7 Antagonistic effects

Effects on hormone receptors elicited by environmental chemicals in water samples
may not only be of activating nature (agonism) but also inhibitory (antagonism). As
a mirror to estrogenic effects, we can often observe anti-androgenic effects in
surface water impacted by wastewater treatment plant effluent. Antagonistic
effects are quantified in the presence of a constant concentration of an agonist,
with antagonists suppressing the signal of the agonist. Typically, a concentration
of agonist (e.g., estradiol) is chosen that has an effect of approximately 80% of
the maximum effect (Figure 7.11a).
A suppression ratio SPR can be calculated from Equation 7.10 from the effect
(e.g., activation of the estrogen receptor) of the sample run in antagonist mode
divided by the effect of the agonist alone. Note that previous work has
abbreviated suppression ratio as ‘SR’, but we are using ‘SPR’ to avoid any
confusion with the specificity ratio, which has the abbreviation SR (see Chapter 9).

activationsample
SPR = 1 − (7.10)
activationagonist

The SPR, often expressed as percentage (1 = 100%), is then also plotted against
the logarithm of the concentration of the antagonist (Figure 7.11b). As for agonism,
the CRC is also linear up to a SPR of 30% (Figure 7.11c). However, it is typically
not possible to derive an effect concentration for the SPR10 because this is often still
within the variability of the agonist (red bar in Figure 7.11). Therefore, it is common
practice to derive the ECSPR20 from the linear portion of the CRC with Equation 7.11
(Escher et al., 2014).

20%
ECSPR20 = (7.11)
slope

Figure 7.11 Approach to determine the antagonistic effect of a chemical or sample.

(a) CRC of the activation of the estrogen receptors in presence of a constant
concentration of agonist; (b) CRC for the data converted to suppression ratio (SPR)
and (c) derivation of the concentration causing a SPR of 20%. Data from Escher
et al. (2014).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Dose–response assessment 113

Figure 7.12 Example of the antagonistic effect of tamoxifen against different

constant concentrations of 17β-estradiol as agonist. Data from Neale and Leusch
(2015).
There are four major caveats when it comes to antagonism testing. First, the
ECSPR20 is heavily dependent on the concentration of the competing agonist. This
is illustrated in Figure 7.12 for 17β-estradiol as agonist at different starting
concentration with varying concentration of the antagonist tamoxifen (Neale and
Leusch, 2015). Most stable and reproducible CRCs of SPR are obtained with a
concentration of agonist that produces about 80% of the maximum effect (Neale
and Leusch, 2015).
Second, antagonism response curves are often more variable than their agonist
equivalents. This is in part because it can be difficult to hit exactly the correct
effect with the competing agonist, and even a small deviation from the
recommended 80% can produce large deviations in the response.
Third, antagonism testing is even more susceptible to cytotoxicity interference
because concomitant cytotoxicity produces the same suppression of the activation
signal as antagonism does. Therefore, the cytotoxicity cut-off has to be even
stricter than in the case of activation, where we recommended to exclude all
concentrations . IC10 from the evaluation of activation (see Section 7.3.6). In
antagonist mode, 10% cytotoxicity would cause a suppression of more than 10%,
hence it is recommended to use an even stricter cytotoxicity cut-off.
Finally, natural organic matter present in the sample after extraction can serve as
a binding site for the competing agonist. This causes an apparent decrease in the
activation signal that looks like antagonism but is really only an artefact due to loss
of the agonist available for interaction with the reporter gene (Neale et al., 2015a).

7.4 CONCENTRATION-RESPONSE CURVES OF WATER

SAMPLES
In applying in vitro bioassays to water quality assessment, it is imperative to be able
to express the bioassay results in quantitative terms. Being able to express a bioassay
result as a discrete value allows, for example, quantitative comparisons and ranking
among different water samples, or comparison between ‘before’ and ‘after’ samples

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 114 Bioanalytical Tools in Water Quality Assessment

to calculate treatment efficacy. Testing of water samples in in vitro bioassays is very

similar to testing of model compounds. Once the water sample has been collected,
extracted and concentrated, the extract is diluted to create a serial dilution series (i.e.,
a stepwise dilution that results in a range of concentrations) and each dilution is
tested in the assay, as one would test different concentrations of a test chemical.
In the case of water samples, each dilution will not have a typical concentration
unit (such as μg/L or mol/L). Instead, the ‘concentration’ can be expressed as a
relative extraction factor (REF), which is unitless.
The REF is calculated from Equation 7.12 as the product of the extraction factor
(EF) from the extraction and concentration step and the dosing factor of the sample
in the bioassay. If an undiluted water sample is used, the EF is set to 1. The EF can be
calculated with Equation 7.13 and the dosing factor with Equation 7.14.
mass or volume extracted
Relative extraction factor REF =
final volume in bioassay
 
kgmatrix or Lwater
= EF · DF (7.12)
Lbioassay
 
mass or volume extracted kgmatrix or Lwater
Extraction factor EF = (7.13)
final volume of extract Lextract
 
volume of extract dosed Lextract
Dosing factor DF = (7.14)
final volume in bioassay Lbioassay
These equations work for any matrix, not just water. For water, which is often
enriched by solid-phase extraction or liquid–liquid extraction, the EF is typically
called ‘enrichment factor’. Other matrices from contaminated sites such as
sediment and soil may have higher loads of contaminants and therefore are not
enriched by the extraction process. We have therefore used ‘extraction factor’ as
the more general term. Practical aspects of extraction and dosing are discussed in
more detail in Chapter 12.
The biological response can then be plotted against the REF of the sample for
a range of dilutions to generate typical CRCs for diverse water samples with log
REF as the concentration unit (Figure 7.13). Testing a water sample without
enrichment would only really be worthwhile at the inlet of a wastewater treatment
plant (WWTP), which generally produces a detectable response even at an REF 1
(log REF = 0). Even then, effects after wastewater treatment are typically below
10% (Figure 7.13a) and so it would be difficult to quantify treatment efficacy if
water was not extracted and enriched. If we enrich by up to an REF of 100 (log
REF = 2) and run full CRCs, then there is a very visible difference between
untreated and treated wastewater (Figure 7.13a). Even field and lab blanks can
show an effect (mostly due to impurities in extraction solvents), but normally
only above an REF 30-100. For advanced treatment, drinking water and bottled
water, it is even more important to enrich a lot: no effects would be visible in the

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Dose–response assessment 115

Figure 7.13 Concentration–inhibition curves of water samples from the Oxley Creek
WWTP (a) and from the Bundamba AWTP and Brisbane River water (b) in the
Microtox assay of bioluminescence inhibition in Aliivibrio fischeri. Concentrations
are expressed as relative enrichment factor (REF, Equation 7.12). WWTP =
wastewater treatment plant, AWTP = advanced water treatment plant, DWTP =
drinking water treatment plant. Data from Macova et al. (2011).

original water sample, while extracted samples have a spread of EC50 values. It is
important to check and assure that all CRCs show distinct differences from the
field and lab blanks (Figure 7.13b).

7.5 BIOANALYTICAL EQUIVALENCY CONCEPT

7.5.1 Relative effect potency
Being able to express the entire CRC in mathematical terms using as little as one or
two parameters (logEC50 and slope for sigmoidal CRCs and slope for linear CRCs)
allows comparisons of the potency (i.e., the strength) of different samples or test
compounds. To express the potency of one chemical i relative to another (usually
the reference compound used in the assay), the CRCs of both chemicals are
compared (Figure 7.14).
The relative effect potency (REPi) of the test compound i relative to the reference
compound can then be calculated by Equation 7.15 (Villeneuve et al., 2000) with the
standard error SE(REPi)) by Equation 7.16.
ECy (reference)
REPi = (7.15)
ECy (i)

1 ECy (reference)2
SE(REPi ) = 2
· SE(ECy (reference))2 + · SE(ECy (i))2
ECy (i) ECy (i)4
(7.16)

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 116 Bioanalytical Tools in Water Quality Assessment

Figure 7.14 Concentration–effect curves of a reference and a test compound. REPi

is the relative effect potency of the test compound i compared with the reference
compound. (a) REPi varies with effect level if the logarithmic CRCs are not parallel.
For logarithmic CRC with the same slope (b) and for linear CRCs (c), the REPi are
independent of the effect level. Reprinted with permission from Escher et al.
(2018). The advantages of linear concentration–response curves for in vitro
bioassays with environmental samples. Environmental Toxicology and Chemistry,
37(9): 2273–2280. © 2018. John Wiley and Sons.

As shown in Figure 7.14a and b, the REPi is only independent of the effect level if
the logarithmic CRCs are parallel to each other, that is, have the same slope. In
reality this is rarely the case and resulting issues are discussed in more detail in
Villeneuve et al. (2000). In contrast, if the linear CRCs are used to derive the
REPi (Figure 7.14c), the ratio is independent on the effect level and can also be
calculated from the inverse ratio of the slopes of the linear CRC (Escher et al., 2018).
The REPs derived from both in vitro and in vivo bioassays are used to determine
the more comprehensive ‘toxic equivalency factors’ (TEFs). The TEF concept is
used frequently in risk assessment, for example in the assessment of dioxin-like
effects by the World Health Organization (Van den Berg et al., 2006). The TEF
concept will be expanded upon in Chapter 8.

7.5.2 Toxic units and toxic equivalent concentration

Since small values of LCy or ICy equate to high toxicity and high values of LCy or
ICy to low toxicity, we often use the inverse of the LCy or ICy to visualise toxicity as
toxic units (TU, Equation 7.17). If the TU is directly derived from a bioassay
experiment, we refer to it as TUbio. The TU can also be calculated from the
chemical concentrations detected and their effect and are then termed TUchem,
which will be further discussed in Chapter 8 on mixtures.
1
TUbio = (7.17)
ICy (sample)
In the same way that REP was used to quantify the relative effect of individual
chemical compounds, the water sample can be quantified relative to the assay

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Dose–response assessment 117

reference compound using toxic equivalent concentrations (TEQ). The TEQ is the
concentration of the reference compound that would be required to produce the
equivalent effect in an assay as a given sample concentration. As highlighted in
the discussion on REP above, it is important to ensure equality of slopes if
sigmoidal logarithmic CRCs are used. Linear CRCs and linearly derived LC10
and IC10 are preferable for determination of TEQ, where possible. The TEQbio is
the ratio of the LCy or ICy of the reference compound and the LCy or ICy of the
water sample (Equation 7.18), where the former is expressed either as a molar
concentration (e.g., 3 pmol/L) or as a mass-based concentration (e.g., 0.88 ng/L),
while the latter is expressed in REF (i.e., unitless):
ICy (reference)
TEQbio = (7.18)
ICy (sample)

7.5.3 Effect units and bioanalytical equivalent

concentration
Formerly, TU and TEQ were also used to express effect concentrations. This is
misleading however, because, as discussed above, a measured effect in an in vitro
assay is not necessarily a toxic effect – it might be an adverse effect or even a
beneficial effect, such a defence mechanism. Therefore, we differentiate in this
book strictly between ‘toxicity’ and ‘effect’, using ‘response’ as the general term.
The analogous term to TU then is effect units (EU), and EUbio can be calculated
by Equation 7.19.
1
EUbio = (7.19)
ECy (sample)

The BEQ is the ratio of the ECy of the reference compound (expressed as a molar
or mass-based concentration) and the ECy of the water sample (expressed as REF,
i.e., unitless) (Equation 7.20) with the SE in Equation 7.21.

ECy (reference)
BEQbio = (7.20)
ECy (sample)


1 ECy (reference)2
SE(BEQbio ) = · SE(ECy (reference))2 + · SE(ECy (sample))2
ECy (sample) 2
ECy (sample)4
(7.21)

When using the BEQ concept, it is important to pay particular attention to the
choice of the reference compound. The optimal reference compound should be (i)
a chemical linked to the mode of action of the bioassay, (ii) one of the most

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 118 Bioanalytical Tools in Water Quality Assessment

potent compounds in the bioassay, and (iii) likely to be present in water samples.
17β-Estradiol is an example of a good reference compound for bioassays that
detect estrogenic endocrine disrupting compounds: it is a natural hormone and the
native ligand for the estrogen receptor (the keystone of the estrogenic response),
it is one of the most potent estrogenic compounds (second only to synthetic
estrogens such as ethinylestradiol) and is commonly found in sewage-impacted
waters.
As an example, if the EC10 for estradiol in a bioassay for estrogenicity was
0.5 ng/L and a water sample had an EC10 of 0.1 REF, the BEQbio expressed as
estradiol equivalent concentration (EEQ) for this sample would be calculated as
EEQ = 0.5 / 0.1 = 5 ng/L.
There are many other BEQs – some commonly used BEQs are
dihydrotestosterone equivalent (DHTEQ) used in bioassays for androgenic
endocrine activity, diuron equivalent (DEQ) used in bioassays for photosynthesis
inhibition, and chlorpyrifos equivalent (ChlEQ) used in bioassays for
acetylcholinesterase inhibition. More examples are given in Chapter 10, where
individual bioassays are discussed.

7.6 CONCLUSIONS
CRCs are the graphical representation of the response of a biological system to
toxicants, and one of the cornerstones of toxicology. The concentration–response
relationship can be expressed in simple mathematical terms using, for example, a
log-logistic equation and at low effect levels even by a linear regression. This
allows quantitative comparisons of different CRCs, determination of REPs of
different toxicants and calculation of TEQ and BEQ for water samples.
In the supplementary information to this book on www.ufz.de/bioanalytical-
tools we provide additional resources for concentration–response assessment
including example spreadsheets for the methods discussed in this chapter.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Chapter 8
Mixtures

8.1 INTRODUCTION
Chemicals rarely occur alone in environmental water samples and wastewater, but
instead occur as mixtures. It is the mixtures that threaten water quality (Kortenkamp
et al., 2019). While the concentrations of individual chemicals are often below any
toxicity threshold and below the limit of detection (LOD) of chemical analysis, the
effect of the mixture may still be a cause for concern.
A wealth of literature is available on mixture toxicity experiments with binary
(two components) and multiple component (more than two components) mixtures
(Kortenkamp et al., 2009). Systematic investigations of complex mixtures
comprising individual components at very low concentrations, such as found in
wastewater and recycled water, are still not very common but we review in this
chapter a number of emerging studies and introduce typical mixture designs.
Unfortunately, many of the available mixture studies provide only anecdotal
evidence and lack an explicit mechanistic understanding. In the last decades,
concepts from pharmacology have been adapted to toxicology (Kortenkamp
et al., 2009; Rider et al., 2018). With these came a conceptual breakthrough in
that mixture effects are now typically categorised in four classes (Table 8.1). Two
of these classes, called independent action (IA) and concentration addition/dose
addition (CA/DA), are more common and have underlying mathematical models.
IA applies when chemicals act according to different modes of toxic action. CA
(in aquatic ecotoxicology) and DA (in mammalian toxicology) apply when
chemicals trigger similar modes of toxic action. Again, the toxicity pathways give

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0119

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 120 Bioanalytical Tools in Water Quality Assessment

Table 8.1 Concepts of mixture toxicity.

Similar Mode of Dissimilar Mode of

Toxic Action Toxic Action

No interaction between Concentration addition Independent action (IA)

chemicals in mixture (CA) or dose addition (DA)
Interaction between Complex action (may lead Dependent action (may
chemicals in mixture to synergy or antagonism) lead to synergy or
antagonism)

us guidance as to what constitutes similarity – starting with mixtures affecting the

same target sites, triggering the same molecular initiating events (MIE) or key
events (KE) all the way through to similarity of adverse outcomes.
The concepts of CA/DA and IA provide quantitative and reliable predictions of
combined effects from existing information on the activity of individual
components of complex mixtures, assuming that the mixture components do not
interact. If the mixture components interact, one is faced with complex and
dependent action, which may lead to synergy (higher toxicity than expected from
CA/DA) or antagonism (lower toxicity than expected from IA).
In this chapter, the major mixture toxicity concepts are introduced and illustrated
with examples from carefully crafted mixture toxicity studies. These examples help
us appreciate the importance of considering chemicals not one-by-one but rather by
looking at their cumulative exposures and combined effect.
A burning issue of international discussion focuses on mixtures at very low
concentrations/doses, where individual chemicals are below their observable
effect level. Current opinions are split as to whether mixture effects in these cases
can be ignored or need to be addressed. Various experimental mixture studies
with multiple components using in vitro bioassays give us the confidence that
even mixture components below their individual detection limits contribute to the
mixture effects.
The evidence from studies with bioanalytical tools that allow testing of
large number of samples and mixtures at low effect levels seem to point
towards linearity of the concentration−response curves (CRCs) at low
concentrations/doses (see Figure 7.1). This is mathematically reasonable and
appears to fit the experimental evidence, although it should be acknowledged that
the detection limit of even an excellent bioassay will hardly ever be below 5% of
response (Chapter 12) and it is thus impossible to experimentally investigate
linearity below 5% of the response.
The toxic equivalency approach is a special case of CA/DA that has proven to be
very useful for the risk assessment of chemicals acting by the same mode of action
(MOA). In this chapter, we review the history of this approach and its application for
risk assessment and water quality assessment.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 121

The importance of mixtures is undeniable (Drakvik et al., 2020), but mixtures are
only slowly making their way into regulatory risk assessment despite very
encouraging recent developments (Bopp et al., 2018; Kortenkamp and Faust,
2018; Rotter et al., 2018). It is well recognised that the risk comes from mixtures
of chemicals not from individual chemicals but how we assess and regulate
mixtures remains a challenge.

8.2 TOXICITY//EFFECTS OF DEFINED MIXTURES

8.2.1 Independent action
Chemicals that act according to different modes of toxic action produce their
effects independently. Naturally, one cannot simply add up the effects. Two
chemicals that produce 60% effect each, for example, will not result in a total
effect of 120% as this is biologically impossible. Rather these chemicals will
produce the total sum effect minus the product of the individual effects according
to the statistical concept of independent random events (in this case 60% × 60%
equals 36%, which results in a combined effect of 120%−36% = 84%) (see
Equation 8.1). For a multi-component mixture of n chemicals i, Equation (8.1)
expands to Equation (8.2).

Effect (mixture) = effect(A) + effect(B)

(8.1)
− (effect(A) × effect(B))

n
Effect (mixture) = 1− (1 − effect(i)) (8.2)
i=1

IA implies that no mixture toxicity or effect will occur if the effects of individual
components are below detection limit. It also implies that subsequent exposure
events would not lead to effects. IA is based on strict stochasticity of toxicity,
which is why IA would in theory also apply to the same agent applied again and
again. In reality there is some stochastic component in toxicity, but individual
tolerance also needs to be taken into account (Ashauer et al., 2017), so the
concept does have some shortcomings for translation into realistic exposure
scenarios.
Despite this limited theoretical foundation of IA, the model of IA was
successfully applied to predict mixture effects of chemicals with strictly
dissimilar MOAs in Aliivibrio fischeri (Backhaus et al., 2000) and green algae
(Faust et al., 2003).
When applying IA, it is also important to consider what ‘no effect’ means. The
‘no observed adverse effect level’ (NOAEL) in mammalian toxicology can be as
high as 20% and the ‘no observed effect concentration’ (NOEC) in ecotoxicology
can be as high as 40% depending on the test design (sample size, replication and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 122 Bioanalytical Tools in Water Quality Assessment

dose spacing; see Chapter 7). There is therefore a high likelihood that chemicals at
the NOAEL or NOEC can still act together to elicit considerable mixture toxicity.

8.2.2 Concentration or dose addition

If chemicals affect the same target site or act through the same MOA, their combined
effect is expected to follow the concept of concentration or dose addition. Here,
instead of summing up the measured effects, mixture toxicity is determined
according to the dose (mammalian toxicology) or concentration (in vitro
toxicology and aquatic ecotoxicology) causing the effect. For a binary mixture,
this can be rationalised as follows: chemical A has an effective concentration
causing 50% of maximum effect (EC50) of 12 μg/L and chemical B has an EC50
of 20 μg/L. In this case, combination of half of the EC50 of A (CA = 6 μg/L) and
half of the EC50 of B (CB = 10 μg/L) will cause 50% of effect. Any other
combination of fractions of the EC50,A and the EC50,B, where the fractions add up
to 1, will also result in 50% effect (e.g., ¼ of EC50,A and ¾ of EC50,B).
A so-called isobologram illustrates the concept of CA for binary mixtures
(Figure 8.1). The concentrations are plotted in toxic units TU, which are
calculated from Equation (8.3) as the concentration of one mixture component
(Ci) divided by its LC50 (LC50,i). The same equation applies for effect units (EU),
which are calculated as the concentration of one mixture component (Ci) divided
by its EC50 (EC50,i).

Ci Ci
TUi = or EUi = (8.3)
LC50,i EC50,i

A straight line connecting EUA = 1 and EUB = 1 (i.e., ΣEUi = 1) corresponds to

CA (Figure 8.1). Any combination of EUs that causes 50% effect and ΣEUi , 1

Figure 8.1 Isobologram for a binary mixture.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 123

indicates that a lower concentration is needed to elicit the effect (i.e., synergy); any
combination that leads to 50% effect and ΣEUi . 1 indicates IA or antagonism.
For multi-component mixtures with n components i, each in a fraction ( pi) of the
total concentration, CA translates to Equation (8.4) for the ECy of the mixture
(ECy(mixture)).
1
ECy (mixture) = n (8.4)
i=1 ( p i /ECyi )

With the IA and CA/DA models, we have the reference cases for mixture toxicity
established. The relevance of CA/DA has been convincingly demonstrated for
estrogenic chemicals in a series of experiments ranging from in vitro and low
complexity bioassays all the way through to in vivo chronic endpoints (reviewed
in Kortenkamp, 2007). Both natural estrogens and xenoestrogens acted in a
CA/DA manner. While this line of evidence is less comprehensive for other
modes of toxic action, there is a large body of literature that substantiates the
concept of CA/DA (Kortenkamp et al., 2009).
Initially focused on ecotoxicological endpoints, these mixture concepts are now
also more frequently applied in human toxicology studies. All conceptual studies
have confirmed that mixture toxicity is clearly higher than the toxicity of
individual chemicals, whether CA/DA or IA applies.

8.2.3 Synergistic and antagonistic effects

There are exceptions to the reference cases of CA/DA and IA, and these apply when
the mixture components interact. Such interactions often take place in the
toxicokinetic phase (Figure 8.2). If one component activates a metabolic enzyme,
for example, causing faster detoxification of the other mixture component and

Figure 8.2 Flow chart for the occurrence of synergistic, antagonistic, IA and CA/DA
mixture effects.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 124 Bioanalytical Tools in Water Quality Assessment

hence causing the resulting mixture effect to be less than predicted with IA, this is
called antagonism. If one compound facilitates uptake of another component or
suppresses detoxifying enzymes, interaction in the toxicokinetic phase may lead
to synergy, that is, effects are significantly stronger than predicted by the model
of CA/DA.
A classic example is piperonyl butoxide (PBO), which is an inhibitor of cytochrome
P450, the enzyme that catalyses the phase I oxidation step in metabolism. PBO
is often used in pesticide formulations, where it shows synergistic effects with
many pesticides, such as the herbicide atrazine. The presence of PBO prevents
the detoxification of these pesticides essentially causing them to become more
toxic than they would be in a metabolically active organism. There are also cases
where the inhibitory action of PBO causes antagonistic effects, for example, with
organophosphate insecticides, which require metabolic activation in order to
inhibit their target, the enzyme acetylcholinesterase. Finally, some organochlorine
pesticides and atrazine induce the activity of cytochrome P450 and thus can act
synergistically with organophosphates by enhancing their metabolic activation.
By analogy, in the toxicodynamic phase, interactive effects at the target site may
lead to antagonistic or synergistic mixture toxicity (Figure 8.2). CA/DA can
generally be expected when chemicals act at the same target site and according to
the same mechanism (toxicity pathway). CA/DA has been shown to apply even in
cases where only the MOA is similar, despite differences in the molecular pathway.
Interactive effects can also occur at the interface between toxicokinetics and
toxicodynamics. An aryl hydrocarbon agonist can, for example, inhibit the
activity of estrogenic chemicals by downregulating the expression of the estrogen
receptor and inducing the enzymes that metabolise estrogens. In these cases,
different target sites may not necessarily directly lead to IA (depicted in Figure 8.2).
While interactive effects are frequently observed for metals, they seem to be
rather rare or quantitatively less important for organic chemicals. The deviations
from the reference models of IA and CA become less and less frequent as the
number of components in a mixture increases. Few studies in the literature that
reported synergistic effects could be confirmed when the observed effects were
compared with the CA/DA predictions (Kortenkamp et al., 2009). Cedergreen
(2014) reviewed the literature on synergistic effects of pesticides, metals and
biocides and concluded that most synergistic effects occurred with mixtures of
low number of components at high concentrations but were otherwise rare and
rather an exception than the norm. Many of the apparently reported synergistic
effects in the literature were artefacts with only 6 out of 90 animal studies
showing true deviation from CA/DA by a factor of 1 to 3.5 (Boobis et al., 2011).
Where deviations from CA/DA were observed in both human and ecological
context, they were typically no larger than a factor of 4 in either direction,
towards synergy or antagonism. In addition, IA often gives predictions for
mixture effects that are only slightly smaller than CA/DA (also generally falling
within less than a factor of 3).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 125

Thus, while it is interesting to examine the subtle differences between the

different mixture toxicity scenarios to understand the mechanisms of interaction
for diagnostic purposes, CA/DA is a robust and ‘realistic worst-case scenario’ to
default on for the evaluation of environmental mixture toxicity for the purpose of
risk assessment. Another advantage of CA/DA is that benchmark toxicity values
(EC50, LD50), which are often already available or can be easily measured, are
sufficient to calculate mixture predictions. On the other hand, full dose−response
relationships need to be established for IA predictions, and complex
toxicokinetic/toxicodynamic models need to be developed for predictions of
synergy and antagonism. Hence, CA/DA provides a practical and easily
applicable model for mixture effects that appears to be robust and sufficiently
accurate for most environmental risk assessment applications.

8.2.4 Grouping of chemicals

But where do mixtures start and where do they end? The European Food Safety
Agency (EFSA) is favouring an approach that uses ‘(cumulative) assessment
groups’, which encompass ‘chemical substances that are treated as a group by
applying a common risk assessment principle (e.g., dose addition) because these
components have some characteristics in common (i.e., the grouping criteria)’
(More et al., 2019).
The mixture toxicity concept of CA/DA has been derived by grouping chemicals
with clearly identified and identical mechanism of toxicity, but practical experience
has demonstrated that even grouping according to the same mode of toxic action will
result in CA/DA responses. When looking at the adverse outcome pathways
introduced in Chapter 4, any commonality along the toxicity pathway may lead to
CA/DA effects (Figure 8.3). The U.S. EPA uses the additional term ‘sufficiently
similar’ for mixture effects elicited on the same target organ or with the same
symptoms (Teuschler, 2007).

Figure 8.3 Adverse outcome pathways in relation to the mixture toxicity concepts.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 126 Bioanalytical Tools in Water Quality Assessment

Kortenkamp (2020) also cautioned from narrowing the definition of cumulative

assessment groups to MIEs and recommended AOP thinking and accounting for
converging pathways that still might lead to mixture effects despite not sharing
MIEs. Thus, grouping has to go beyond the distinction of similar and dissimilar
action and account for joint action along the entire pathway.
For the application of bioanalytical tools, these considerations translate into the
expectation that CA/DA will occur in assays indicative of a clearly defined MOA or
mechanism of toxicity. Reporter gene assays only express one pathway, so in
principle all chemicals that are active in a reporter gene assay should act
according to CA/DA. For in vitro assays that are indicative of steps further down
the toxicity pathway such as genotoxicity assays, the definition of rather broad
cumulative assessment groups lead to predictions consistent with CA as
demonstrated with various genotoxicants, which differed in their MIE, but
showed near CA mixture effects in the micronucleus assay in Chinese hamster
ovary-K1 cells (Ermler et al., 2014).
Most importantly, CA/DA is also applicable even if individual chemicals are
present at concentrations that do not cause any observable effect, as discussed below.

8.2.5 Something from nothing?

One would intuitively assume that a mixture of compounds, each of which is
present at a concentration below that known to cause an effect individually, would
not cause any effect. This, however, is not the case when chemicals occur in mixtures.
As early as 1988, Deneer et al. (1988) mixed 50 industrial chemicals at
concentrations 20−400 times less than the EC50 of each compound and observed
mixture effects consistent with CA/DA (Table 8.2). Subsequently, researchers
systematically investigated and compared the mixture toxicity concepts of
CA/DA and IA. When similarly acting chemicals were mixed at their EC01 level
(i.e., the 1% effect level, which is not actually measurable in practice) for various
ecotoxicological endpoints, the observed effects were typically high in
accordance with the predictions for CA (Table 8.2).
Similar studies were performed with estrogenic chemicals (Silva et al., 2002).
Eight xenoestrogens (industrial chemicals that have a low estrogenic potency,
e.g., bisphenol A) were mixed at their 1% effect level and together caused an
appreciable effect in the yeast estrogen screen. The observed effect was
quantitatively consistent with the prediction of CA/DA (Table 8.2). Even when
the very potent natural estrogen, 17β-estradiol, was mixed with several
xenoestrogens, the CA/DA model remained valid at a mixing ratio of 1:50,000
(17β-estradiol:xenoestrogens) (Rajapakse et al., 2002). This in vitro observation
was confirmed with an in vivo endpoint of estrogenicity, the production of the
egg yolk precursor protein vitellogenin in male fish (Brian et al., 2005).
Even for chemicals that act according to different modes of toxic action,
mixtures of more than 10 compounds at low effect level have shown

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 8.2 The ‘something from nothing’ effect for compounds with the same mode of toxic action (MOA).

MOA Chemicals in Bioassay Ci Exp. Effect CA//DA Ref

Mixture Predict.

Non-specific toxicity 50 industrial Water flea Daphia 20−400 times 50% 50% 1
chemicals magna lower than EC50
Cytotoxicity 10 quinolones Microtox (biolumine EC01 54% 72% 2
scence inhibition in
Aliivibrio fischeri)
Respiratory 16 phenols Microtox EC01 82% 95% 3
uncouplers
Reproduction 18 triazine herbicides Reproduction of EC01 47% 44% 4
green algae
Binding to ER 8 xeno-estrogens Yeast estrogen EC01 ∼25% ∼25% 5

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Mixtures

screen
Estrogenicity 2 estrogens, Vitellogenin induction NOEC ∼58% ∼50% 6
3 xeno-estrogens (male fathead
minnow fish)
Androgen receptor 13 pesticides and MDA-kb2 assay, IC01, IC10, IC20 Diverse, CA provided best 7
antagonism 17 antioxidants, PCP, suppression of prediction
industrial pollutants dihydro-testosterone
effects
CA/DA = concentration addition/dose addition; ER = estrogen receptor. ‘Exp. effect’ = effect measured experimentally, ‘CA/DA predict.’ =
prediction of the CA/DA model, Ci = concentration of component i, EC50 = effect concentration causing 50% of maximum effect, EC01 = effect
concentration causing 1% of maximum effect, NOEC = no observed effect concentration, PCP = personal care products, Ref = literature reference.
1
Deneer et al. (1988), 2Backhaus and Grimme (1999), 3Altenburger et al. (2000), 4Faust et al. (2001), 5Silva et al. (2002), 6Brian et al. (2005),
7
Orton et al. (2014).
127
by guest
128

Table 8.3 The ‘something from nothing’ effect for chemicals with different modes of action (MOAs).

Chemicals in Bioassay Ci Experimental Comparison Ref

Mixture Effect with
Predictions

11 aquatic priority Reproduction of NOECb 64% ≈ IA (,CA) 1

pollutants green algae
16 dissimilarly Reproduction of 6.6–66% of NOECb 18% ≈ IA (,CA) 2
acting compounds green algae

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Four pesticides with E-SCREEN 25–100% of NOECb Significant = IA 3
different MOA (proliferation of
MCF7 cells)
Ci = Concentration of component i, NOEC = no observed effect concentration, CA = concentration addition, IA = independent action,
Ref = literature reference.
1
Walter et al. (2002), 2Faust et al. (2003), 3Payne et al. (2001).
Bioanalytical Tools in Water Quality Assessment
 Mixtures 129

measurable effects, which were consistent with the prediction obtained from IA
(Table 8.3). This is very important as the concept of IA actually implies that if an
individual chemical has no effect, the mixture will have no effect. An
experimentally determined ‘no effect’ is, however, not necessarily a true ‘zero
effect’, it is often merely a non-observable effect. If 100 chemicals were mixed at
concentration ratios where each would individually elicit a 1% effect, the joint IA
effect would be as high as 63%. If the 100 individual chemicals were present at
0.1% effect level, the mixture effect would still reach 9.3%. Thus, even for IA, we
can conclude that mixture toxicity has the potential to be of critical importance.
These findings imply that there are no ‘safe’ concentration levels at which
chemicals do not contribute to mixture effects (Kortenkamp et al., 2007).

8.3 ASSESSMENT OF CONCENTRATION-ADDITIVE

EFFECTS USING THE TOXIC EQUIVALENCY CONCEPT
The concept of toxic equivalency factors (TEF) and toxic equivalent concentrations
(TEQ) is an extension of the mixture concept of CA/DA and is only applicable for
groups of chemicals with a common MOA. This is a special case of CA/DA and the
condition of parallel dose−response curves must be fulfilled. Due to its ease of
application and communication it is now widely applied in the risk assessment of
chemical mixtures.
The TEF concept was initially developed for the binding of polychlorinated
dibenzodioxins (PCDD) to the aryl hydrocarbon receptor (AhR) and associated
toxicity endpoints. It was soon expanded to polychlorinated dibenzofurans
(PCDF) and coplanar polychlorinated biphenyls (PCB). The reference compound
for dioxin-like activity is 2,3,7,8-tetrachlorodibenzodioxin (TCDD), the most
potent activator of the AhR discovered to date. The effect of the mixture is then
expressed as the concentration or dose of TCDD that would elicit the same effect
as the mixture, rather than dealing with the many different toxicity measures for
the various chemicals that act according to the same MOA as TCDD.
In Chapter 7, we have learnt that the toxicity of chemicals with the same MOA
can be expressed in relation to a reference chemical i as relative effect potency REPi
(Equation 7.16). For the same pair of chemical i and reference chemical, there will
be different REPi values for different toxicity endpoints and test species or cells, as
well as for different exposure scenarios (duration, frequency). A variety of REPi
values is therefore available for each chemical i in a range of in vitro, in vivo and
epidemiological endpoints. For risk assessment a consensus toxic equivalent
factor (TEFi) was derived by a group of experts for a number of chemicals using
a large experimental REPi database and expert knowledge (Van den Berg et al.,
2006). The TEFTCDD of TCDD was set by definition as 1, and those of the other
dioxins vary from 1 (for penta-CDD) to 0.0003 for octa-CDD. The furans have
similar TEFs as the corresponding dioxins (e.g., 2,3,7,8-TCDD vs.
2,3,7,8-TCDF), while the planar PCBs all have TEFs of 0.00003.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 130 Bioanalytical Tools in Water Quality Assessment

According to Equation (8.5), the TEQ of a mixture of chemicals is the sum of the
product of the concentration of each component i in the mixture, Ci, and its TEFi
value.

n
TEQ = Ci × TEFi (8.5)
i=1

As research with dioxins has shown, the TEQ concept has proven to be
applicable to whole organism endpoints despite being strictly valid for
receptor-mediated toxicity only. The application of the TEQ concept has,
therefore, also been recommended for the chemical risk assessment of estrogenic
chemicals (Simon et al., 2007), polycyclic aromatic hydrocarbons (PAH) (Nisbet
and Lagoy, 1992) and neurotoxic PCBs (Simon et al., 2007). In principle there
seems to be no limitation for the application of this useful concept, provided that
the chemicals assessed act via concentration-addition and share a common slope
of the logarithmic dose−response curve. The toxic units (TU) introduced in
Section 8.5.4 rely on the very same assumptions as the TEQ approach with strict
CA/DA and similar shape of CRCs.

8.4 MIXTURES IN RISK ASSESSMENT

8.4.1 Concepts
To date, no common worldwide consensus has been reached for chemical risk
assessment of mixtures although mixtures are implicitly and/or explicitly
addressed in many regulatory documents.
The International Programme on Chemical Safety (IPCS) of the World Health
Organization has developed a framework for assessing cumulative risk (IPCS,
2009). This framework accounts for the mixture toxicity concepts discussed in
this chapter using the tiered approach depicted in Figure 8.4 (slightly modified

Figure 8.4 Steps for inclusion of mixtures into risk assessment (adapted from IPCS,
2009). CA = concentration addition; DA = dose addition; IA = independent action;
RA = risk assessment.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 131

from the report). As a first step (issue identification), it is necessary to assess if

combined exposure is likely to occur. The IPCS proposed in tier 1 to adopt a
mixture specific extrapolation factor for the derivation of acceptable effect
level before moving onto the higher tiers with explicit consideration of
mixture effects. In tier 2, if the likelihood of mixture effects has been identified
but is yet to be further characterised, CA/DA is used as the default concept. In
tier 3, more information is required on the MOAs of the mixture components in
order to select the appropriate mixture toxicity models for the risk assessment
(Figure 8.4).
If chemicals act by CA/DA, the risk quotients (RQi, see Chapter 2 for definition)
will add up to a cumulative risk index (RI) for n mixture components i (Equation
8.6). An alternative but synonymous nomenclature is the hazard index (HI) based
on hazard quotients (HQi). As in the case of RQ, a RI of 1 is the threshold
between acceptable (RI , 1) and unacceptable risk (RI . 1).

n 
n
exposure leveli
RI = RQi = (8.6)
i=1 i=1
acceptable effect leveli

Different measures for exposure and acceptable effect levels may apply for
ecological and human health risk assessment (e.g., dose or concentration). For
Equation (8.6) to be valid, the exposure and acceptable effect levels must have
the same units.
If TEFs are available, the RI can be derived from the TEQ of the mixture divided
by the acceptable effect level of the reference chemical using Equations (8.7)
or (8.8).

n
TEQ
RI = (8.7)
i=1
acceptable effect levelreference chemical
n
i=1 TEFi × exposure leveli
RI = (8.8)
acceptable effect levelreference chemical

8.4.2 Do we have account for mixture effects in risk

assessment?
A useful indicator of whether mixture effects play a role in risk assessment is the
maximum cumulative ratio (MCR), which is the ratio between the observed
cumulative toxicity and the maximum toxicity caused by one chemical
(Equation 8.9).
cumulative toxicity
MCR = (8.9)
maximum toxicity from one chemical
Mathematically, the MCR can be expressed as the ratio of the RI to the RQmax,
that is, the maximum RQ relates to the chemical with highest exposure and/or

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 132 Bioanalytical Tools in Water Quality Assessment

effect (Equation 8.10).

RI
MCR = (8.10)
RQmax

If the MCR equals to 1, the mixture toxicity is caused solely by one mixture
component. If all n mixture components contribute to the cumulative toxicity, the
MCR will reach n. In the former case (MCR = 1), cumulative risk assessment
would not be required, whereas in the latter case (MCR close to n), cumulative
risk assessment would be imperative.
The MCR is also a measure of the fraction of toxicity equivalents that derive from
the most toxic component in the mixture, for example, an MCR of 2 indicates that
the most toxic chemical is responsible for 50% of the mixture effect and an MCR of
1.1 indicates that the most toxic component causes approximately 90% of the
mixture effect.
Reality lies somewhere between these extremes. Price and Han (2011) calculated
the MCR values for the pesticide water concentrations found in more than 4000
samples collected by the U.S. Geological Survey (USGS) in the 1990s. Each
sample was analysed for up to 83 pesticides, of which up to 30 were actually
detected. The obtained MCR values for chronic human health ranged from 1 to
6. A similar result was found for chronic effects in fish (representing ecological
risk). For the water samples with RI , 1, the MCR values were generally higher
than for those with RI . 1. This result indicates that in many cases where
individual water samples exceeded a RI of 1, the mixture toxicity was driven by
only a few components. Conversely, the findings also imply that when low
hazard was associated with a water sample and MCR was higher, it was not
possible to identify individual culprits. Therefore, the joint effect of many
chemicals needs to be considered in order to evaluate the overall hazard.
A similar study performed on the mixture risk of 26 pharmaceuticals in
wastewater treatment plant effluents yielded MCRs between 1.2 and 4.2
(Backhaus and Karlsson, 2014). A study for agricultural contamination by
pesticides and veterinary pharmaceuticals (Holmes et al., 2018) predicted that in
less than 4% of the scenarios the MCR would be .2 for cases with RI . 1 but
that the MCR would increase with decreasing RI. The MCR also helped to
identify the small fraction of mixtures out of over 3000 surface water samples
where the single chemical RQ would have largely underestimated the mixture
risk (Vallotton and Price, 2016).
The MCR was also applied to evaluate the human health risk for
anti-androgenicity of phthalates in a retrospective study that analysed 27 years of
exposure data in 24-h urine samples from the German Environmental Specimen
Bank (Apel et al., 2020). While RI decreased from 1.8 to 0.2 from 1985 to 2015,
the MCR actually increased during this time from 1.5 to 1.9. This means that the

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 133

overall mixture hazard decreased but the role of mixtures has become more
important during that time.

8.4.3 Mixtures in chemicals regulations

The IPCS framework for the risk assessment of combined exposure to multiple
chemicals (Meek et al., 2011, 2014) provides the guiding principles for most
approaches on how to deal with mixtures in risk assessment.
The IPCS has been instrumental in developing and supporting harmonised
approaches for the risk assessment of mixtures of dioxin-like chemicals, where
the TEF concept is applied (Van den Berg et al., 2006). The TEFs defined by the
WHO working group are internationally accepted and have found their way into
many national regulations. Furthermore, TEFs are a cornerstone of the Stockholm
Convention, which is a global treaty to protect human health from exposure to
persistent organic pollutants (POPs). The WHO has also adopted the IPCS
framework for assessing cumulative risk for mixture of chemicals in drinking
water (WHO, 2017a).
The globally harmonised system for the classification and labelling of chemicals
(GHS) employs a summation method that classifies a mixture according to its
relative amounts of already classified chemicals or on the basis of a similar
mixture that is already classified (United Nations, 2009). Only components that
exceed 1% on a mass basis need to be included in the evaluation, unless there is
indication that a component present at a lower concentration is toxicologically
relevant.
The U.S. has implemented mixtures in its regulatory frameworks as early as 1986
(U.S. EPA, 1986). The RI approach has been used to evaluate the risk posed by
air pollutants and hazardous waste. The cumulative risk assessment of pesticides
by the U.S. EPA makes explicit use of the mixture toxicity concepts, employing
CA/DA for pesticides that share a common mode of toxic action and IA for
those that do not (U.S. EPA, 2002a).
Similarly, in Australia and New Zealand, the CA/DA model has been
recommended as the method for assessing whether or not the water quality
criteria are exceeded (Australian Government, 2018a).
In the European Union’s chemical regulation, REACH (EP&EC, 2006a), some
industrial mixtures are explicitly regulated under the name ‘substances’, which
refers to single chemicals and products that are mixtures of unknown and variable
composition (e.g., petroleum hydrocarbons, surfactants) and multi-component
substances. Such mixtures are treated as if they were individual chemicals in the
risk assessment process with toxicity testing on the mixtures. Furthermore, the
read-across approach of REACH (European Chemicals Agency, 2017) allows
implicit introduction of mixtures in risk assessment.
In the future, one can expect further development in mixture assessment and its
application to different environmental and health regulations in Europe, indicated by

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 134 Bioanalytical Tools in Water Quality Assessment

the ‘Chemicals Strategy for Sustainability towards a Toxic-free Environment’

(European Commission, 2020), which explicitly calls for the introduction of
‘mixture assessment factors’ in REACH and for accounting for mixture effects in
other chemicals regulations.
A mixture assessment factor (MAF) (Bopp et al., 2018) is an extrapolation factor
that would be used on a single chemical’s RQi to account for mixture effects by
multiplying the RQi by the MAF to obtain the mixture RI. The MCR analysis
(Price and Han, 2011) described above would yield a MAF of 2−17, which
implies that the mixture scenario is dependent on the exposure situation and is
probably site-specific (Bopp et al., 2018). Van Broekhuizen and Traas (2016)
argued that only 5−10 chemicals dominate the mixture effect under most
scenarios and therefore proposed a MAF of 10 for environmental risk. Although

Figure 8.5 Mapping of chemicals and their mixtures to the risks they pose for various
toxicological effects. For each nine chemicals the individual risk quotients RQi are
presented for different types of effect (e.g., hepatotoxicity, neurotoxicity, etc.).
Chemicals are grouped according to the legislative sector they are regulated under
(e.g., REACH, pesticides, cosmetics, food contaminants, etc.). The risk index RI,
that is, the sum of the risk quotients RQi is illustrated for mixtures within each
sector and in the last column for the cross-sectorial mixture. Reprinted from Bopp
et al. (2019). Regulatory assessment and risk management of chemical mixtures:
challenges and ways forward. Critical Reviews in Toxicology, 49(2): 174–189.
BY-NC-ND licence © 2019 European Union. Published by Informa UK Limited,
trading as Taylor & Francis Group.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 135

this is a very intuitive approach, in practice it will be more difficult to define MAFs
because they will not only depend on the number of chemicals included in their
derivation and their relative effect potency but also on the RQi of the single
chemical as well as on the exposure scenario. If the mixture effect can be
accounted for completely by the known chemicals and only few are mixture
effect drivers, a MAF of around 10 would be reasonable. However, the MAF
might be substantially higher for those exposure scenarios and biological
endpoints where the known chemicals only explain a small fraction of the effect.
Another challenge that remains for mixture risk assessment is that of regulatory
silos, that is, different legislative sectors producing different guidance for the same
chemicals or chemicals of a cumulative assessment group that are expected to act
together in mixtures (Kortenkamp and Faust, 2018; Bopp et al., 2019).
Regulation of mixture risk will have to include not only multiple types of effects
for multiple chemicals but also to bridge across different regulatory spaces as is
illustrated in Figure 8.5.

8.5 MIXTURES AND WATER QUALITY

8.5.1 What type of mixture effects occur in water samples?
While it has been established that the mixtures of chemicals at concentration levels
that are well below observable effect levels may produce substantial mixture toxicity
(Section 8.2.5), there are very few studies in the literature that tackle designed
mixtures with many components mixed in the concentrations as they typically
occur in water samples. Translation of the results from designed mixture toxicity
experiments to real water samples that may contain thousands of chemicals at
exceedingly low concentrations, therefore, remains uncertain, although existing
evidence points to chemicals acting together at low concentrations in a
concentration-additive manner even if they do not show any effect on their own.
Early mixture toxicity studies on guppy fish demonstrated that mixtures of
chemicals at very low concentrations could be described by the CA/DA model
for the underlying baseline toxicity, independent of the specific MOA these
chemicals produce at higher concentrations (Hermens et al., 1985). Warne and
Hawker (1995) took this concept a step further arguing that with an increasing
number of chemicals at equitoxic concentrations in a mixture, CA/DA was the
more likely outcome, and demonstrated that the results of a large number of
experimental studies confirmed this hypothesis.
Two types of mixture studies have been undertaken to evaluate how chemicals
act together in mixtures. The first is to combine water samples with single
reference chemicals for the given bioassays. This was done for a mixture of
wastewater and the herbicide reference chemical diuron as well as a baseline
toxicant in the combined algae test (Escher et al., 2008a). The isobolograms
derived from the TUs for photosynthesis inhibition for the mixture of a
wastewater treatment plant effluent and the herbicide diuron (Figure 8.6a) and for

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 136 Bioanalytical Tools in Water Quality Assessment

Figure 8.6 Isobologram of a ‘binary’ mixture of (a) wastewater treatment plant

effluent and the herbicide diuron and (b) wastewater treatment plant primary
effluent and the baseline toxicant 3-nitroaniline. Data from Escher et al. (2008a).

growth rate of a wastewater treatment plant primary effluent and the baseline
toxicant 3-nitroaniline (Figure 8.6b) confirmed that the water samples acted in a
concentration-additive manner with reference chemicals.
The second approach is to mix chemicals at the concentrations detected in water
samples and test for their interaction by comparing the predicted mixture effects
with the measured mixture effects. Junghans et al. (2006) mixed 25 pesticides in
typical exposure scenarios in field runoff and CA predicted the mixture effects
best, although IA predictions differed only by a factor of 1.3.
In a study that focused on the oxidative stress response detected with AREc32,
pharmaceuticals and pesticides were mixed in groups of 5 and 10 each and
together in equipotent concentration ratios and in concentration ratios of their
drinking water guideline values, making up a mixture of 5, 10, 15 and 20
components (Escher et al., 2013). All experimental mixtures showed very
similar effects as the predictions by CA in the AREc32 assay, confirming that
CA is a robust prediction model for chemicals acting according to the same
MOA. In a way, the similarity of MOA was forced in this example because
the AREc32 reporter gene assay only reacts to one MOA. Nevertheless, it is
still encouraging to see that mixture effects of environmentally realistic
mixtures of very diverse types of chemicals could be well described by simple
mixture models.
Tang et al. (2014) evaluated samples along the entire wastewater treatment train
from WWTP influent to advanced treated water with three bioassays, including
non-specific toxicity towards A. fischeri, photosynthesis inhibition in green algae
and oxidative stress response in AREc32. They detected four to 50 chemicals in
the different samples and tested them alone, as mixtures in six groups (endocrine
disrupting chemicals, iodinated contrast media, antibiotics, pharmaceuticals,
pesticides and others) and all mixed together. There was excellent agreement
between measured mixtures within groups and predictions from single chemicals

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 137

Figure 8.7 Measured BEQ of the designed mixtures in comparison with the
cumulative BEQs from component-based mixture prediction of the six chemical
groups: (a) non-specific toxicity towards Aliivibrio fischeri (baseline toxicity
equivalent concentrations, baseline-TEQ), (b) photosynthesis inhibition in green
algae (diuron equivalent concentrations, DEQ), (c) oxidative stress response in
AREc32 (t-butylhydroquinone equivalent concentrations, tBHQ EQ). WWTP =
Wastewater treatment plant; UF = ultrafiltration; RO = reverse osmosis; EDC =
endocrine disrupting chemicals; ICM = iodinated contrast media. Reprinted with
permission from Tang and Escher (2014). Which chemicals drive biological effects
in wastewater and recycled water? Water Research, 60: 289–299. © 2014 Elsevier.

(not shown) and between experimental mixtures of each of the six groups summed
up in comparison with experimental mixtures of all chemicals (Figure 8.7).
Although the chemicals in the groups were not all acting according to the same
MOA, the steps from individuals to groups to mixtures of all detected chemicals
increase our confidence that chemicals combined in mixtures as they occur in real
water samples act together to trigger measurable effects and that these effects
could be predicted by the summation of bioanalytical equivalent concentrations
(BEQ), which is effectively CA.
Note, however, that this strong agreement between experimental and predicted
mixture effect is limited to designed mixtures. If the effects of the designed
mixtures were compared with the effects of the entire water samples, then only
photosynthesis inhibition show a good agreement (Tang and Escher, 2014). For
the other two endpoints, the detected chemicals explained less than 3% of the
effect in the entire water sample due to the presence of a large number of
undetected bioactive chemicals (Escher et al., 2013; Tang and Escher, 2014).
This issue will be discussed further in Section 8.5.2 and Chapter 13.
Another example is a study that tested over 200 designed mixtures of two to 14
chemicals mixed in concentrations as they occurred in water samples taken during
rain events in small creeks that showed a very high diversity of chemical burden
and mixture compositions (Neale et al., 2020a). The activation of AhR in AhR
CALUX, the activation of peroxisome proliferator-activated receptor gamma
(PPARγ) in PPARγ GeneBLAzer and the oxidative stress response in AREc32
all followed very well the model of CA. CA even applied to the cytotoxicity of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 138 Bioanalytical Tools in Water Quality Assessment

the three cell lines (Escher et al., 2020b). This example is presented in more detail
as case study 6 in Chapter 14.7.
Altenburger et al. (2018) took a different approach where just two defined
12-component mixtures were tested in 19 in vitro bioassays. The idea was that
each bioassay would respond to one of more of the mixture components and the
challenge was to confirm the associated mixture prediction against the
background of inactive chemicals. This ambitious study confirmed CA in many
bioassays but also pointed out issues with respect to solubility or masking of
effects by cytotoxicity in reporter gene assays.
More of these types of studies with complex mixtures should be encouraged
to obtain a better picture of what really happens in water samples with
potentially thousands of chemicals present at very low concentrations. Currently,
evidence is merely anecdotal, but a better conceptual understanding of the
mixture effects under these scenarios will greatly enhance future water quality
assessment.

8.5.2 How much of the measured effects in water sample

can be explained by known and detected chemicals?
The answer to this question is not straightforward. It depends on the one hand on the
type of water sample and on the other, on the bioassay that is used to evaluate a given
water sample.
For some types of water samples (e.g., industrial effluents, contaminated sites,
accidents), a few – often well-known – chemicals explain the majority of the
effect. Here, effect-directed analysis (EDA, see Chapter 13, Section 13.3.2) might
identify the dominant risk drivers. In other water types (urban wastewater, surface
water), which are impacted by a large number of very diverse chemicals, the
identified chemicals often only explain a minute fraction of the observed effect.
In these cases, bioassays are particularly valuable to estimate the full extent of
toxicity or effect in a sample.
How much of the activity can be explained by the detected chemicals also
depends on the selected bioassay. In Chapter 13.4, we will introduce a
classification of bioassays into two categories based on the type of responsive
chemicals and their degree of specificity. From observation alone, we can already
say that some assays respond very strongly to a small number of potent
chemicals. For whole organism tests, photosynthesis inhibition in algae or
immobilisation of Daphnia magna by insecticides are good examples. Herbicides
are 10,000 to 100,000 times more potent than other chemicals for inhibition of
photosynthesis and hence the mixture effect is often completely dominated by
herbicides unless a sample does not contain any herbicides (Tang and Escher,
2014; de Baat et al., 2018; Glauch and Escher 2020). Herbicides are not only
used in agriculture but also in urban applications, so they are quite ubiquitous and
general mixture risk drivers. Insecticides show a similarly high specificity, but

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 139

their mixture dominance occurs mainly in agricultural streams because D. magna

reacts also quite sensitively to other toxicants.
For reporter gene assays, hormone receptor transactivation assays are another
good example. Natural hormones are up to a million or more times more potent
in reporter gene assays than most other chemicals and even xenostrogens are only
two to four orders magnitude more potent than non-specifically acting chemicals.
Any water sample with the slightest contribution of wastewater, treated or not, or
manure from animals will respond mainly to the natural estrogens (Könemann
et al., 2018) and mixture effects can typically be well explained by detected
estrogenic chemicals.
It is a very different case for reporter gene assays for receptors of lower
specificity, such as the pregnane X receptor (PXR) and PPARγ, and for
KE-based bioassays, such as those responding to adaptive stress responses, and
for apical endpoints, such as cytotoxicity and fish embryo toxicity. Here, the
mixture effects of the detected chemicals can only explain a small fraction of the
entire water sample’s toxicity or effects.

8.5.3 Mixture effects at very low effect levels (,10%)

With the exception of wastewater, most water samples do not cause an effect unless
the organic micropollutants are extracted and enriched prior to dosing in a bioassay
(Chapter 7.4). CRCs are linear at low effect levels up to 30% of effect and can be
described by one single parameter, the slope of the CRC (Chapter 7.3.6). At even
lower effect levels, below 10% of effect, the predictions for CA and IA overlap
perfectly and can be described by Equation (8.11), which is a joint CA/IA
mixture model (Escher et al., 2020b).


n
Effect (mixture) = pi × slopei × Ctot
i=1 (8.11)
= slopemixture × Ctot

Several case studies demonstrated the validity of this equation in designed

mixture studies for cytotoxicity and reporter gene activation (Escher et al.,
2020b). Even more importantly, most component-based predictions of mixture
effects in water from analytical data were within the applicability range of
this equation (,10% effect). This means that in the future, we do not need
to worry whether detected chemicals have the same MOA or not, we can
simply apply the joint CA/IA mixture model. This effectively means that
TU summation is valid for chemicals with similar and dissimilar MOAs
provided that resulting effect levels do not exceed 10% in theory and 30%
in practice.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 140 Bioanalytical Tools in Water Quality Assessment

8.5.4 Component-based prediction of mixture toxicity

in water
In a component-based risk assessment, the risk index RI (Equation 8.7) of combined
exposure to multiple chemicals is calculated from the exposure and effect data of the
individual components. A similar approach can also be taken for predicting the
mixture toxicity in a water sample by weighting the measured concentrations by
their effect potency and summing them up over all mixture components. This is
analogous to the TU/EU approach introduced as Equation (8.3) in Section 8.2.2,
only without the condition that the concentrations are a fraction of the
LC50/EC50; instead, they are the measured concentrations, and a TU/EU
exceeding 1 would imply that there is a risk.
The sum of the TUs is defined by Equation (8.12) and analogously for EU by
Equation (8.13). We give this component-based TU/EU the index ‘chem’
because the TU/EU can also be directly measured with a bioassay in a water
sample as TUbio (Equation 7.18) and EUbio (Equation 7.20) as outlined in
Chapter 7.5.

n n
Ci
TUchem = TUi = (8.12)
i=1 i=1
LC50,i

n n
Ci
EUchem = EUi = (8.13)
i=1 i=1
EC50,i

The component-based approach is the most traditional approach when evaluating

mixtures in water quality. The reader has probably wondered why this approach was
not introduced earlier in this chapter. There is a reason for it. The component-based
approach is often ‘sold’ as characterising true mixture effects but as we have learnt
above, it only constitutes the mixture effects of the detected chemicals that have
available toxicity/effect data. If we include 20 chemicals in the approach and find
that five dominate the TU, will this mean that these five are the true risk drivers
or could there be unknown chemicals that contribute to the mixture effects?
Comparison of TUchem with TUbio as outlined in Chapter 13 will shed light on
the mixture effects elicited by unidentified chemicals and chemicals below their
detection limit.
Also, we have learnt that the component-based approach only holds for CA, but it
has been used widely irrespective of the MOAs of the components. The most recent
insights that realistic mixtures in water mostly obey CA and that CA and IA merge to
one single model at low effect levels justify the application of component-based
approaches, but one must always be aware that there is no guarantee that the
entire mixture toxicity is captured.
One of the earliest mixture studies showed that the TUchem of 12 organics and 11
metals in fish samples were not a good predictor of the integrity of the fish
community (Dyer et al., 2000). Nevertheless, the component-based approach had

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Mixtures 141

a revival in the last decade, often without verification by comparison with

ecological quality.
Water quality monitoring concentrations are now often directly reported as
TUchem. While this is evidence of good intentions, the results are often obscured
because the choice of the effect data is not always transparent, often relying on
predicted effect data or mixed experimental/modelled effect data. If well
performed, component-based mixture predictions can help us understand the
importance of mixture effects and the contribution of known chemicals, but we
always have to keep in mind the bias due to the selection of analytes in the
chemical analysis. Malaj et al. (2014) demonstrated how the number of acute-risk
chemicals was related to the overall risk in a study that included over 200
chemicals measured at over 4000 sites in European rivers.
The TUchem of a diverse set of wastewater treatment plant effluent across Europe
derived from algae, daphnid and fish toxicity data for 26 pharmaceuticals were
dominated by less than 10 pharmaceuticals overall. In each sample as few as two
to four pharmaceuticals explained most of the mixture effect (Backhaus and
Karlsson, 2014). Even when evaluating the monitored concentrations of 110
chemicals in an Asian river basin, the top 10 mixture components explained more
than 80% of TUchem for acute and 95% of TUchem for chronic effects (Liu et al.,
2020). TUchem of 75 detected compounds in the Lake Victoria Basin in Kenya
indicated chronic risk with 16 chemicals being risk drivers (Kandie et al., 2020).
Gustavsson et al. (2017) considered chemicals below the LOD in TUchem
analysis of 141 pesticides in Swedish streams. They either left out chemicals
below the LOD, used LOD/2 or estimated likely concentrations below LOD.
Using LOD/2 most likely overestimated the TUchem by two orders of magnitude
and leaving out chemicals below their LOD probably came closer to a realistic
TUchem. Interestingly, while often only one to three pesticides dominated the
mixture risk in each sample, these pesticides varied from site to site, so that 83 of
141 pesticides would need to be included to capture 95% of the risk at all sites.
The concept of mixture toxic pressure assessment (Posthuma et al., 2008) relies
on multi-substance potentially affected fraction (msPAF) (Klepper et al., 1998;
Posthuma et al., 2008) and is an expansion of the component-based approach that
also integrates species-sensitivity distributions. The msPAF is the lower fifth
percentile of the cumulative density function of the log-normal distribution of
TUchem. Only three pharmaceuticals contributed to the acute mixture toxic
pressure while 8 (of totally 54) contributed to the chronic mixture toxic pressure
in seven Swedish rivers (Lindim et al., 2019). The mixture toxic pressure
approach has great potential, especially as quality-controlled SSDs are available
for more than 12 000 chemicals (Posthuma et al., 2019). Case studies with this
dataset illustrated the utility of this approach and identified new previously
untagged priority pollutants.
It is important to note that these concepts work well within the domain of
analysed chemicals (i.e., TUchem of one water sample compared to TUchem of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 142 Bioanalytical Tools in Water Quality Assessment

another water sample), but that for some water samples and/or bioassays the
measure response (TUbio) can be significantly greater than TUchem, as will be
discussed in more detail in Chapter 13.

8.6 CONCLUSION
Risk assessment focusing on single chemicals can only be justified if only one
chemical of a mixture is toxic, while all other compounds are inert, or their
mixture effect is no larger than the toxicity of the dominant component. This
condition is only met occasionally for contaminated sites and industrial effluents.
In contrast, wastewater and treated water usually contain thousands of
compounds including many unknown chemicals and transformation products.
While under these conditions, it is virtually impossible to fully understand and
evaluate the complex mixture interactions, bioanalytical tools may provide
valuable information about the overall mixture effect. It is futile to attempt to
resolve all chemicals in any given water sample, but a smart combination of
chemical and bioassay analysis combined with mixture toxicity modelling will
give us a wealth of information for risk characterisation.
Regulatory chemical risk assessment is ready for implementation of mixtures and
various approaches are under consideration in different regions of the world, with
the biggest challenge being to decide which chemicals to group together as mixtures.
Despite of all scientific progress described in Section 8.5, there has been no
formal uptake for mixtures in water quality legislation. The adoption of
effect-based methods would by definition be an acceptance of the importance of
mixture effects (Brack et al., 2019). The Water Framework Directive has been
exploring the use of effect-based methods with a detailed technical report
(Wernersson et al., 2015) but at present this approach to include mixtures in
water quality assessment has not reached legal status. A major hindrance might
be the lack of accepted effect-based trigger (EBT) values. As outlined in Chapter
13, much scientific progress has been made over the past years on the derivation
of EBTs for various water types in a scientific context – this hopefully will lead
to increased regulatory uptake in the near future.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Chapter 9
In vitro assays for the risk
assessment of chemicals

9.1 INTRODUCTION
The last two decades have seen unprecedented progress in transforming risk
assessment from animal testing to an approach that relies on alternative methods
comprised of in vitro assays and in vitro to in vivo extrapolation methods. This
paradigm shift was initiated by the U.S. National Research Council’s strategy to
modernise toxicity testing with high-throughput pathway-based methods (NRC,
2007) and the parallel implementation of an integrated testing strategy (ITS) in
the European Chemicals Regulation REACH (EP&EC, 2006a, b). Provided that
an in vitro method is fully validated and there is an (ideally mechanistic) in vitro
to in vivo extrapolation model available, in vitro data can now be used in
quantitative risk assessment (Blaauboer, 2008).
The field of water quality assessment has profited enormously from the progress
made in chemical risk assessment because large numbers of in vitro bioassays have
been developed and validated and effect data for single chemicals have become
publicly available that serve not only for the elucidation of toxicity and risk
assessment but also for interpreting water testing results and for linking bioassay
responses to chemical analysis.
In this chapter we give a brief overview of the developments in this area, starting
with an introduction of the regulatory background for the application of in vitro
assays and computational methods. In different regulatory environments these
novel tools that replace animal testing are referred to as ‘alternative test methods’
or ‘new approach methods’. We give an overview of international developments

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0143

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 144 Bioanalytical Tools in Water Quality Assessment

with a focus on the ‘Integrated Testing Strategy’ in the European Union and the
‘Toxicity Testing in the 21st Century’ strategy in the U.S. (National Academies
of Sciences, 2017).
Then we focus on the CompTox strategic initiative of the U.S. EPA that led to the
ToxCast program as well as the Tox21 initiative, which has provided (and continues
to provide) a huge set of easily accessible in vitro activity data for use by researchers
and regulators. We outline the main applications of these data for elucidation
of toxicity pathways and prioritisation of chemicals for in-depth risk assessment.
Further, quantitative in vitro to in vivo extrapolation (QIVIVE) models serve
for human health risk assessment but can also be used for ecological risk
assessment.
We conclude with a detailed description of exposure in cell-based bioassays
because this will become of high relevance for improving the existing QIVIVE
models and will give more confidence in the application of in vitro bioassays in
risk assessment but also for environmental and biomonitoring applications.

9.2 APPLICATION OF NEW APPROACH METHODS IN

REGULATION
9.2.1 Alternatives to animal testing methods
The 3Rs to Replace, Reduce and Refine laboratory animal testing have a long
tradition since the 1950s (Hartung, 2010). In vitro assays are not a direct
substitute for animal testing but are the basis of an alternative test method. An
alternative test method for the replacement of an animal test is the combination of
an in vitro test system and a predictive in silico model. The prediction model is
an unambiguous algorithm for converting in vitro data into predictions of
toxicological endpoints in animals or humans.
A response in an in vitro assay does not always result in harm to a whole
organism but can provide valuable information as to whether an organism-level
effect is possible or even likely. A clear advantage of in vitro systems is that they
can be derived from human cell lines making interspecies extrapolation obsolete
(Figure 9.1).

Figure 9.1 The in vitro−in vivo parallelogram. HHRA = human health risk
assessment; (Q)IVIVE = (quantitative) in vitro to in vivo extrapolation.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 145

It is imperative that in vitro to in vivo extrapolation models account for the

toxicokinetic processes that describe the pathway from external to internal dose,
which are influenced by bioavailability, uptake and excretion, and metabolism
of chemicals in the body (Blaauboer, 2010). In vitro models are available
for every one of these toxicokinetic and toxicodynamic processes. In the future
we can expect to see the development of whole ‘virtual organisms’ that are
comprehensive toxicokinetic−toxicodynamic models parameterised with in vitro
data.
In 1991, the European Centre for the Validation of Alternative Methods
(ECVAM) was founded to support the regulatory application of alternative
test methods. ECVAM’s important role is emphasised by the Directive on
protection of animals used for scientific purposes (EP&EC, 2010), which aims
to minimise the use of test animals and sets ECVAM up as the EU reference
laboratory.
The U.S. counterpart, the Interagency Coordinating Committee on the Validation
of Alternative Methods (ICCVAM), is an interagency committee of the U.S.
government composed of representatives from 15 federal regulatory and research
agencies. ICCVAM conducts technical evaluations of new, revised and
alternative test methods with regulatory applicability, and promotes the scientific
validation and regulatory acceptance of test methods that more accurately assess
the safety and hazards of chemicals and products and that refine, reduce or
replace animal use.
The U.S. EPA made a commitment in 2019 to reduce animal testing and
funding by one third by 2025 and to eliminate it entirely by 2035. To achieve
this goal, they support the development of ‘new approach methods’ (NAM),
which is a general term that refers to any non-animal-based approaches that can
be used to provide information in the context of chemical hazard and risk
assessment.
In 2009, the United States, Japan, the EU and Canada signed a Memorandum of
Cooperation establishing the ‘International Cooperation on Alternative Test
Methods’ (ICATM) to enhance international cooperation and coordination on the
validation of non-animal or reduced-animal toxicity testing methods. Since then,
South Korea, Brazil and China have joined in. A global harmonisation is believed
to help the regulatory acceptance. The Organisation for Economic Cooperation
and Development (OECD) published a guidance document on good in vitro
method practices (OECD, 2018) and is committed to the implementation of the
3R-principles.

9.2.2 Integrated testing strategy in the European Union

Reliance on animal in vivo toxicity data to extrapolate safe effect levels has made the
risk assessment process very slow and costly and has resulted in incomplete
evaluations and data gaps. Testing methods that use mammals or other

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 146 Bioanalytical Tools in Water Quality Assessment

Figure 9.2 Paradigm shift in human health risk assessment (HHRA) to integrate
alternative test methods. QSAR = quantitative structure−activity relationship.

vertebrates carry the additional issue of animal ethics. Even then, the relevance of
animal data introduces a level of uncertainty and requires interspecies
extrapolation to derive human safe levels. These issues have been recognised in
the EU chemical legislation REACH, which provides a tiered approach that uses
non-animal alternative test methods. In an integrated testing strategy (ITS),
in silico and in vitro approaches are applied for all assessment endpoints. Both of
these approaches can be based on animal and/or human cell models (van
Leeuwen and Vermeire, 2007). Figure 9.2 exemplifies ITS using the example of
effect assessment in human health risk assessment.
Read-across from similar chemicals is the first step in ITS, where in silico
methods (i.e., computer-based) are used to predict the toxicity of the test
chemical based on the toxicity of other chemicals that share structural or
physicochemical properties. For example, quantitative structure−activity
relationships (QSAR) provide one approach that can be used to predict the
toxicological effects of untested chemicals – provided their properties fall within
the applicability domain of the chosen QSAR model. The OECD has made
several QSAR models available to the scientific community in the form of a free
software suite (OECD QSAR Toolbox, van Leeuwen et al., 2009).
In the next step of ITS, in vitro bioassays are applied (Figure 9.2). In vitro
tests can provide a substantial advantage over in vivo testing, including the
possibility to test toxicity on cells derived from humans (not test animals),
lower variability, better experimental control, higher sensitivity, shorter
duration and lower financial and ethical cost than whole-animal tests. Despite
the great potential of in vitro methods, their practical application is presently
limited in the EU regulation to screening and priority setting, as well as
classification and labelling.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 147

9.2.3 Toxicity testing in the 21st century (Tox21) strategy

in the United States
The U.S. National Research Council’s strategy to modernise toxicity testing with
high-throughput pathway-based methods (NRC, 2007) was realised with two
major strategic programmes: ToxCast and Tox21. The U.S. EPA formed the
National Center for Computational Toxicology (NCCT) and developed the
Toxicity ForeCaster (ToxCast) project for advanced toxicity testing and
modelling. The Tox21 Initiative (Collins et al., 2008) is a collaboration between
the NCCT of the U.S. EPA, the National Toxicology Program (NTP) of the
National Institute of Environmental Health Science (NIEHS), the National Center
for Advancing Translational Sciences (NCATS) of the National Institutes of
Health (NIH) and the Food and Drug Administration (FDA).
It is the goal of Tox21 to (1) identify mechanisms of chemically induced biological
activity, (2) prioritise chemicals for more extensive toxicological evaluation and (3)
develop predictive models of in vivo biological response (Shukla et al., 2010). To
achieve this goal, Tox21 has set up a process to develop high-throughput
screening (HTS) assays, to test chemicals in these assays, and to make the data
continuously accessible in public databases (Figure 9.3). Nominated assays
underwent a review process before they were optimised for robotic quantitative
HTS (qHTS) format in 1536-well plates. After robotic validation, almost 10 000
chemicals (the so-called ‘Tox21 10 K library’) were run in this qHTS format. An
R data processing and evaluation pipeline ToxCast Analysis Pipeline (tcpl, more
information in Chapter 7.3.4) was used to derive toxicity benchmark
concentrations. After application of the data within the consortium, all results
including the raw data were made publicly available in databases that can easily be
accessed, for example, via the CompTox Chemistry Dashboard (U.S. EPA, 2020).
As of 2020, the Tox21 10 K library has been run in over 50 bioassays, mainly
assays on molecular initiating events (MIE), focussing on nuclear receptors, and
key events (KE), focussing on stress response pathways, generating over 85
million data points. ToxCast included only 300 chemicals in the first phase,
which were screened with 700 assay endpoints, and expanded in the second
phase to 1000 chemicals screened in approximately 1000 assay endpoints. These
data are now available for developing predictive models and running screening
level risk assessment – a few illustrative examples will be presented below. They
are also, of course, available for iceberg mixture modelling (Chapter 13),
enabling translation between chemical and bioassay results. Ongoing
experimental developments of Tox21 are HT transcriptomics (HTTr) and
high-content imaging of cultured cells with multiple fluorescent probes for HT
phenotyping (Thomas et al., 2019).
At present, in vitro data have to be validated with in vivo data. It is Tox21’s
expressed focus to better curate and characterise legacy in vivo toxicity studies so
they can be efficiently used for validation (Thomas et al., 2018).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 148 Bioanalytical Tools in Water Quality Assessment

Figure 9.3 The set-up of the Tox21 HTS bioassay profiling platform. NCATS =
National Center for Advancing Translational Sciences; qHTS = quantitative
high-throughput screening. Figure reprinted with permission from Sakamuru et al.
(2020). Profiling the Tox21 chemical library for environmental hazards: applications
in prioritisation, predictive modelling, and mechanism of toxicity characterisation.
In: Big Data in Predictive Toxicology, Editors Neagu and Richarz, pp. 242–263.
© 2020. The Royal Society of Chemistry.

Although the HTS assays are the core of Tox21 and ToxCast, the computational
models developed and refined are necessary to make best use of this wealth of
experimental data and apply them for mechanistic toxicology and risk assessment.

9.3 APPLICATION OF IN VITRO ASSAYS IN RISK

ASSESSMENT
9.3.1 A paradigm shift in human health risk assessment
Initial applications of NAM were geared at replacing individual elements in the risk
assessment process but did not question the risk assessment paradigm itself. Judson
and co-authors from the U.S. EPA initiated an entirely new risk assessment process,
which is based on the direct inclusion of in vitro information into quantitative risk
assessment (Judson et al., 2011). They termed the proposed framework ‘high
throughput risk assessment’ because it relies on HTS in vitro data targeting
specific toxicity pathways. This is now recognised as next-generation risk
assessment (NGRA) and is detailed in the next sections. Given the importance of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 149

Figure 9.4 Framework of high-throughput chemical risk assessment. Modified and

reprinted with permission from Villeneuve et al. (2019). High throughput screening
and environmental risk assessment – state of the science and emerging
applications. Environmental Toxicology and Chemistry, 38(1): 12–26. © 2019. John
Wiley and Sons.

the initial formulation of the revised risk assessment paradigm and that the essential
elements remain valid, we are starting with looking back in time and present the
earliest version before we move on to the state of the art.
In NGRA, a chemical is tested for activity in a large number of HTS in vitro
assays (Figure 9.4 left). The benchmark concentration derived from
concentration−response assessment in these assays is termed ‘biological pathway
altering concentration’ (BPAC). BPACs are typically log-normally distributed if
as many pathways as possible are investigated (Figure 9.4, middle). On the
exposure side, external doses are linked to concentrations in blood with
physiologically based toxicokinetic (PBTK) models.
The distribution of cellular concentrations can then be combined with
distributions of BPAC using probabilistic re-sampling techniques such as Monte
Carlo to derive a distribution of doses that perturb biological pathways
(Figure 9.4, right). A ‘lower limit of biological pathways altering dose’ (BPADL)
is derived from a low percentile of this distribution (e.g., 10% percentile) and
treated as the no effect level (NEL) in the effect assessment step of risk assessment.
This generic initial concept has been vastly expanded in the last 10 years
(Wetmore, 2015; Sipes et al., 2017; Bell et al., 2018). Many modifications and
improvements were introduced, and easily accessible computation tools have
been made available to the public. In the following we discuss the individual
steps before we outline the NGRA.

9.3.2 Quantitative adverse outcome pathways

HTS of perturbations in cellular pathways using a large test battery of in vitro assays
has allowed the identification of molecular targets and crucial biological pathways
that are linked to adverse effects in vivo, and subsequently can be used as molecular
biomarkers (Andersen et al., 2010; Martin et al., 2010; Knudsen et al., 2011). The
HTS data together with extensive data mining in existing literature have produced an

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 150 Bioanalytical Tools in Water Quality Assessment

Figure 9.5 Putative adverse outcome pathway (AOP) for uterotrophy elucidated
quantitatively with Tox21 bioassays. Modified after Bell et al. (2018). In vitro to in vivo
extrapolation for high throughput prioritization and decision making. Toxicology
In Vitro, 47: 213–227. © 2018. Elsevier. KE = key event; MIE = molecular initiating
event.

unbiased database of chemicals to be prioritised for in-depth toxicological

characterisation and stitching together of AOPs (Judson et al., 2009).
Figure 9.5 is an illustration of how the Tox21 HTS assays helped develop a
quantitative AOP using the example of uterotrophy, which is characterised in vivo
by increased uterine weight (Bell et al., 2018). The likely MIE is binding to the
estrogen receptor (assessed by Novascreen receptor binding assay NVS-NR-hER)
and intermediate effects can be quantified by the activation of the estrogen
receptor through reporter gene assay such as the Attagene ERa-trans and the
Tox21-ERa_BG1_Luc.
A quantitative association could be made by QIVIVE analysis, indicating that the
concentration needed to activate the HTS assays translated as an internal blood
concentration agreed very well with the estimated blood concentrations in the
in vivo assay on mice (Bell et al., 2018). The assumptions made for the QIVIVE
in this example is that the nominal active concentrations in the HTS assays are
equal to the plasma concentration in vivo.

9.3.3 Quantitative in vitro to in vivo extrapolation

QIVIVE is often directly integrated into the risk assessment but for didactive
purposes, we are introducing its underlying concept independently.
The active concentration in a HTS assay, for example, the activity concentration
at cut-off (ACC) or the 50% activity concentration AC50 (see Chapter 7 for
definition of these benchmark concentrations), is adjusted by a conversion factor
to obtain an equivalent administered dose (EAD) in units of mg/kg/day for
the particular biological endpoint (Figure 9.6). The conversion factor is derived
by reverse dosimetry (Bell et al., 2018). An arbitrary dose, for example,
1 mg/kg/day is divided by the associated predicted plasma concentration in
humans at this dose level. The predicted plasma concentration can be expressed

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 151

Figure 9.6 Principle of quantitative in vitro to in vivo extrapolation (QIVIVE). CSS =

steady-state plasma concentration; Cmax = maximum plasma concentrations.

either as steady-state concentration Css or maximum concentration Cmax and can be

predicted by classical PBTK-models that are well-established models in
pharmacology and toxicology. The new high-throughput toxicokinetic models
described in more detail in Section 9.3.4.2 may also serve for this purpose.
Using a range of bioassays yields a range of EADs (Figure 9.6). The EAD or the
distribution of EADs can be compared to doses in vivo or a lower percentile of the
EAD distribution used in the risk assessment process as outlined below.
Implementation of bioassay exposure considerations as described in Section 9.4
might further improve the QIVIVE models if the exposure in in vitro assays is
described by freely dissolved concentrations instead of nominal concentrations
(Honda et al., 2019).

9.3.4 Next-generation risk assessment

NGRA combines a systematic HTS-based tiered hazard assessment with
high-throughput toxicokinetics on the exposure side. The idea behind NGRA is
not to predict animal toxicity by in silico and in vitro assays but rather relies
on the fact that effects on a molecular and cellular level occur at lower
concentrations than would be expected from exposure of the population
investigated and would hence be protective.

9.3.4.1 Hazard assessment in next generation risk assessment

Thomas et al. (2019) have outlined a roadmap to implement HT hazard assessment
in NGRA. Existing HTS assays and new assay developments will be implemented
in a tiered hazard assessment strategy (Figure 9.7). The first tier of the hazard

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 152 Bioanalytical Tools in Water Quality Assessment

Figure 9.7 Tiered testing framework for hazard characterisation in Tox21 as a

component of next-generation risk assessment (NGRA). Simplified from Thomas
et al. (2019). POD = point of departure; AOP = adverse outcome pathway; KE =
key event; MIE = molecular initiating event.

assessment is comprised of the mentioned high content screening assays (HTTr and
HT cellular phenotyping), which are presently under development, and may in the
meantime be covered by existing multiplexed assays or larger interrogative test
batteries. As one outcome of this tier, new chemicals can be grouped with
chemicals of a known biological target or pathway.
One of the existing problems is that many chemicals are biologically
promiscuous, that is, trigger many pathways. This is especially true at
concentrations near cytotoxicity, where a cytotoxicity burst phenomenon can
often be observed, which is the non-specific activation of multiple pathways
(Judson et al., 2016).
Therefore, only the most sensitive pathways will move forward to a second tier,
where more in-depth and targeted in vitro assays will be performed to determine if
the tested chemical can be associated with an existing AOP. In this case, tier 3
testing can move forward to AOP modelling and probing associated KEs. If no
AOP exists, focus will be on organotypic assays and microphysiological
endpoints (Thomas et al., 2019). The eventual aim of the process outlined in
Figure 9.7 is to produce a point of departure (POD) for risk assessment.
The process produces PODs with a decreasing degree of confidence going from

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 153

left to right and will therefore require increasing extrapolation factors in risk
assessment.

9.3.4.2 Exposure assessment in next generation risk assessment:

high-throughput toxicokinetics
Toxicokinetic model predict blood and plasma concentrations from the administered
dose. Reverse toxicokinetic models are needed for QIVIVE to go from assay
concentrations to the administered dose. Numerous physiologically based
toxicokinetic models (PBTK) models are available, but few are amenable to high
throughput due to the large number of required input parameters. Moxon et al.
(2020) have recently proposed a tiered approach for implementing PBPK models
in higher throughput.
High-throughput toxicokinetic (HTTK) models use experimental in vitro or
predicted data on the hepatic clearance of the chemical of interest and its plasma
protein binding to predict the plasma concentration in humans using a simple
three-compartment toxicokinetic model composed of liver, gut and the rest of the
body (Sipes et al., 2017). Comparison with in vivo data has confirmed the
suitability of these HTTK models (Wambaugh et al., 2018). HTTK modelling is
thus able to provide exposure assessment in NGRA.

9.3.4.3 Case studies on next-generation risk assessment

There are already some early examples of NGRA emerging. A case study on dermal
risk of coumarin in personal care products relied on two assay panels covering nuclear
receptors and cell stress, an immunomodulary screening assay, genotoxicity
assessment with ToxTracker and HTTr (Baltazar et al., 2020). Using this
comprehensive first-tier assessment, POD distributions of nominal concentrations
were derived from the various HTS assay panels applied. Interestingly, despite
coumarin showing extensive metabolism, the PODs from metabolically competent
and incompetent cells did not show substantial differences. Different exposure
scenarios assuming application of face cream or body lotion that contained the
products using the PBPK model by Moxon et al. (2020) yielded Cmax plasma
concentrations that were two to three orders of magnitude lower than the most
sensitive HTS endpoints, the enzymatic assay of carbonic anhydrase type 1 and a
HT transcriptomics assay with a HepaRG cell line. The authors concluded that
NGRA was a valuable addition to existing risk assessment for this compound and
confirmed that there was a substantial margin of safety and hence minimal concern
considering the investigated exposure scenarios (Baltazar et al., 2020). They also
stated that ‘there is not yet agreement on how large a margin of safety derived in
an NGRA needs to be to assure human safety’, emphasising that these first case
studies are not an end but an encouraging start of new developments.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 154 Bioanalytical Tools in Water Quality Assessment

9.3.5 Applications of new approach methods

for environmental risk assessment
Although HTS assays of Tox21 and ToxCast are primarily targeting human health
endpoints, they have also great potential for applications in monitoring and
ecological risk assessment (Schroeder et al., 2016). Given that the cellular
toxicity pathways are quite conservative and applicable to many biological
organisms, it seems conceivable to apply HTS data in ecological risk assessment
in a similar way as for human health risk assessment.
Applications for environmental monitoring are described throughout this book.
The Attagene multiplex assays were actually applied for water quality monitoring
(Escher et al., 2014, Section 13.2.2) before being included as one of the core
assays in Tox21. The GeneBLAzer battery came into practical applications in
water quality monitoring at about the same time as it was implemented in Tox21.
In vitro bioassays have a longer tradition in water quality testing, but with the
parallel development of HTS assays, traditional assays have been adapted to
higher throughput. The available data from single chemical screening has helped
justify the choice of bioassays and anchored them better in toxicity pathways and
AOPs. The abundance of HTS data that is now publicly available can be used for
mixture modelling and iceberg modelling, linking chemical analysis to in vitro
bioassays as described in more detail in Chapter 13.
There is also a desire to replace animal testing in risk assessment. The HTS
assays may well serve this purpose. As outlined in Figure 9.8, we can use HTS
data to train predictive models and use the data and the models for prediction of in
vivo effect concentrations (Villeneuve et al., 2019). As for human health we need
to apply QIVIVE models and integrate not just one endpoint but a whole battery
of HTS assays to integrate multiple pathway perturbations and link them ideally to
established AOPs. Cross-species extrapolation can then draw on established
models, but species-specific QIVIVE models might also be able to shed light on
the mechanistic differences in species sensitivity of chemical stressors.
Similar to QIVIVE for humans, one can envision an approach that applies the
same principles to relate effect concentrations in vitro to critical concentrations in
fish plasma by assuming that the freely dissolved concentration in the in vitro
assay equals the unbound concentration in the fish plasma (Figure 9.9). Then
reverse bioconcentration modelling can translate the plasma concentration back to
external exposure concentration.
Finally, quantitative AOPs (qAOP) can be developed as is illustrated in
Figure 9.10 for an AOP in female fathead minnows that is triggered by the MIE
inhibition of the cytochrome P450 19A aromatase (Conolly et al., 2017). Three
quantitative prediction models were developed: the first describes the KE
happening on hypothalamic−pituitary−gonadal axis with aromatase inhibition
decreasing the transformation of testosterone to 17β-estradiol (E2), which leads to
a reduced synthesis of the egg yolk protein precursor vitellogenin; the second

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 155

Figure 9.8 Potential use of HTS data and integration for environmental risk
assessment. Reprinted with permission from Villeneuve et al. (2019). High
throughput screening and environmental risk assessment – state of the science
and emerging applications. Environmental Toxicology and Chemistry, 38(1): 12–26.
© 2019. John Wiley and Sons.

describes the resulting reduction in fecundity expressed as eggs produced per day;
which in the third eventually leads to a declining population.
This qAOP model is not chemical-specific and could be translated to other
untested aromatase inhibitors (Conolly et al., 2017). The strong quantitative links

Figure 9.9 Extrapolating in vitro concentration−response to equivalent in vivo effect

in fish analogously to the QIVIVE model for Judson et al. (2011) presented in
Figure 9.4. Plasma concentrations in fish estimated based on the free fraction of
the active chemical concentration in the assay test well. Reverse bioconcentration
modelling can then be used to estimate the water concentration that would yield
the equivalent internal dose. Modified and reprinted with permission from
Villeneuve et al. (2019). High throughput screening and environmental risk
assessment – state of the science and emerging applications. Environmental
Toxicology and Chemistry, 38(1): 12–26. © 2019. John Wiley and Sons.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
156

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Bioanalytical Tools in Water Quality Assessment

Figure 9.10 Development of a quantitative adverse outcome pathway (qAOP) in fathead minnow on the example of the anti-breast
cancer drug fadrozole. MIE = molecular initiating event; E2 = 17β-estradiol; KE = key event. Reprinted after modification with
permission from Conolly et al. (2017). Quantitative adverse outcome pathways and their application to predictive toxicology.
Environmental Science & Technology, 51(8): 4661–4672. © 2017. American Chemical Society.
 In vitro assays for the risk assessment of chemicals 157

between KE and adverse outcome give us confidence in using MIE- and KE-based
in vitro assays in risk assessment and regulation.

9.4 EXPOSURE IN IN VITRO BIOASSAYS

When using in vitro data for human health risk assessment, the limitations of in vitro
systems need to be clearly understood and stated to achieve a meaningful risk
assessment. One of the most critical issues is exposure.
Cell-based in vitro assays are mostly performed in 96- or 384-well plates
(Figure 9.11), although the ultra-HTS 1536-well plate format is now also used in
robotic platforms such as the Tox21 programme. Most cells adhere to the bottom
of the well, with a few cell lines floating freely in suspension. Medium containing
serum proteins and other nutrients is added to the well to ensure that sufficient
nutrition is available for sustained cellular growth. Only a portion of chemicals
added to the well is taken up by the cells. A large fraction is bound to medium
components such as serum proteins and lipids. Another fraction of chemicals can
be lost due to water−air transfer, and some may sorb to the plastic plate (Figure 9.11).
The chemical fraction that is taken up by cells can be metabolised and/or
distributed to target and non-target sites (Figure 9.11). Storage lipids are
non-target sites, while macromolecules such as membrane lipids, proteins
(enzymes and receptors) and DNA are target sites for the MIE to occur. Both the

Figure 9.11 Exposure and toxicokinetic processes relevant in a cell-based in vitro

assay.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 158 Bioanalytical Tools in Water Quality Assessment

initially dosed chemical and its metabolites can contribute to the toxic effect. In
some instances, chemicals are activated by metabolism and thereby become more
potent than the dosed chemical. More often though, metabolites are more water
soluble and hence more easily excreted than the parent chemical. To account for
both possibilities, assays with cells of low metabolic capacity are often run once
in the presence of an oxidising agent or an isolated liver enzyme mixture (e.g., S9
liver enzyme preparation) and once without this addition. If the observed effects
differ, one can conclude that metabolic activation and/or detoxification plays a
role. More on metabolism is discussed in Section 9.4.3.
There exist fairly simple mass balance models for stable chemicals that do not
degrade in the medium nor are metabolised (Armitage et al., 2014; Fischer et al.,
2017). Absorption and excretion process are relatively fast in cells (Fischer et al.,
2018a), hence, steady state can be assumed for many chemicals. For very
hydrophobic chemicals more than 99% will be sorbed to serum proteins in the
medium, whereas very hydrophilic chemicals will remain predominantly in the
water phase rather than accumulate in the cell. The dose-metric used to describe
exposure in cell assays therefore matters.

9.4.1 Dose-metrics in cell assays

The target concentration is the concentration of a chemical present at the target site,
which initiates a response (Table 9.1, Figure 9.11). It is also called the biologically
effective dose (BED) or biologically effective concentration. It is almost impossible
to measure the target site concentration but there are models to estimate it if the
target site is the membrane. If the target is in the cytoplasm, we can often assume
that the concentration in the cytoplasm is equal to the external freely dissolved
concentration unless the chemicals are ionic or active transport processes play a
role (Escher et al., 2020a). If a chemical is fluorescent, it is possible to locate it in
the cell by imaging methods.
The cellular concentration is the next best proxy for the target concentration
(Table 9.1, Figure 9.11). It can be quantified if the experiment is scaled up and
sufficient numbers of cells can be collected, separated from the medium and
extracted. It is not easily feasible to assess and quantify the cellular concentration
routinely in parallel to running plate-based experiments.
The freely dissolved concentration is the bioavailable concentration and the best
experimentally accessible dose-metric because it can be directly compared between
different assay types that use different formats and media (Table 9.1, Figure 9.11).
The freely dissolved concentrations has been quantified for more than a decade in
specific applications (Heringa and Hermens, 2003; Kramer et al., 2012) but has
only recently become available for 96-well plate assays thanks to versatile solid-
phase microextraction (SPME) fibres with low coating volume (Henneberger
et al., 2019; Huchthausen et al., 2020). SPME fibres coated with polymers and
C18 that sorb charged chemicals have very recently become available in

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 159

Table 9.1 Typical dose-metrics for cell-based bioassays.

Dose-Metric Definition Unit Measurement// Model

Target Concentration at mol/kgmembrane Only modelled,

concentration target site or mol/Lcytoplasm qualitatively with
/BED (membrane, imaging methods
cytoplasm,
proteins)
Cellular Total concentration mol/106 cells Measured after
concentration in the cell separation of cells and
extraction with solvent
Freely Concentration in the mol/Lmedium Measured with
dissolved surrounding solid-phase
concentration medium that is not microextraction
bound to proteins (SPME)
Total Concentration in mol/Lmedium Measured after total
concentration cells and medium (volume of cells extraction with solvent
negligible)
Nominal Total amount of mol/Lmedium Calculated from added
concentration chemical divided by amount
the volume of
exposure medium
Adapted from Groothuis et al. (2015). BED = biologically effective dose.

multiplexed plate format, allowing in the future more routine and HT quantification
of freely dissolved concentrations.
It is common practice to use the nominal concentration in cell-based bioassays,
although strictly speaking the total concentration would provide a better measure
because it corrects for irreversible loss processes to the air and to the plastic of
the plates and degradation in the medium (Table 9.1, Figure 9.11).
Loss to air is a problem in well-plate format that is often underestimated. This
process might not only lead to loss of chemicals but semi-volatile chemicals can
even cross over and contaminate neighbouring wells (Birch et al., 2019). Only
chemicals with a medium–air partition constant Kmedium/air .104 are fully
retained during 24 hours of exposure at 37°C (Escher et al., 2019). Chemicals
just below this cut-off are the ones likely to contaminate neighbouring wells
(Escher et al., 2019). The Kmedium/air is not only dependent on the Henry constant
KH or the air–water partition constant Kaw, which are in turn a function of vapour
pressure and solubility, but also dependent on the medium-water partition
constant Kmedium/water. Hydrophobic chemicals that bind strongly to proteins and
lipids of the medium are better retained in the bioassay well than hydrophilic
chemicals with the same Kaw. Up to 20% of the Tox21 chemicals might have
been partially lost during experiments even with protein-rich medium, for

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 160 Bioanalytical Tools in Water Quality Assessment

example, when 10% serum is supplemented (Escher et al., 2019), but not all Tox21
assays are run with such protein-rich conditions so losses might have been even
higher. Differences in medium composition therefore have an influence on the
loss processes in cell assays.
For water quality testing loss to the air is less of a problem as most water
pollutants are not volatile. In addition, the sample preparation process involves
several blow-down steps so most residual volatile chemicals would be removed
prior to bioassay testing in any case.
In contrast, sorption to the plastic plate materials (mostly polystyrene) is a
problem that is often overrated under standard cell assays conditions. Organic
chemicals do indeed have rather high polystyrene−water partition constants
KPS/w, but still about one to three orders of magnitude smaller than the
corresponding octanol−water partition constant Kow (Fischer et al., 2018b).
Given the large mass-to-volume ratio of polystyrene, the loss would be huge if
equilibrium were attained – but the diffusion coefficients of organic chemicals in
polystyrene are very small. Therefore, the chemicals hardly have the time to
partition into the polystyrene during the duration of a typical HTS cell assay, but
instead sorb to the surface and penetrate only a few micrometres. Nevertheless,
the loss due to binding to the plate material is very much dependent on the test
conditions. The fish embryo assay, for instance, is typically conducted in pure
water without supplements and the losses to a polystyrene 24-well plate are huge
for chemicals of high hydrophobicity (high Kow) and not negligible for chemicals
of medium hydrophobicity (3 , log Kow , 4) (Figure 9.12a). It is therefore not

Figure 9.12 Depletion of the medium concentration by sorption to the polystyrene of

a well-plate assay. (a) Substantial and chemical-dependent depletion in a fish embryo
toxicity assay during 96 h in a 24-well plate. (b) Negligible loss in a mammalian cell
assay in medium supplemented with 10% fetal bovine serum (FBS) during 24 h in
a 384-well plate. Kow = octanol−water partition constant. Reprinted with permission
from Fischer et al. (2018b). Application of experimental polystyrene partition
constants and diffusion coefficients to predict the sorption of organic chemicals to
well plates in in vitro bioassays. Environmental Science & Technology, 52: 13511–
13522. © 2018. American Chemical Society.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 161

recommended to run the fish embryo assay in polystyrene plates: glass or

glass-coated plates reduce the loss to the plate material. In contrast, a cell assay in
medium supplemented with 10% fetal bovine serum will hardly be affected by
sorption to the polystyrene because the medium proteins and lipids retain the
chemical and make is less available for sorption to the well plates (Figure 9.12b).
There is no hydrophobicity-dependence of the loss because the partitioning to
medium components and to polystyrene are proportional to each other. If
serum-free medium is used, the stabilising effect of the medium is much smaller.
Finally, the nominal concentration is the concentration calculated from the
amount of chemical added to the assay (Table 9.1, Figure 9.11) and is the one
reported in most publications and databases, including Tox21 and ToxCast. It is
reasonable to use nominal concentrations for relative comparisons as long as all
exposure conditions, number of cells, and medium composition remain constant.
Hence, appropriate QA/QC (Chapter 11) is vital when using nominal
concentrations as the dose-metric.

9.4.2 Serum-mediated passive dosing

The medium components stabilise the system and reduce possible loss processes.
The cellular uptake does not lead to depletion because the freely dissolved
concentration can be replenished by desorption from the medium proteins and
lipids. This process has also been termed serum-mediated passive dosing (SMPD,
Fischer et al., 2019). SMPD can be somewhat deceptive as it also decreases the
sensitivity of the bioassay because a lower fraction of total chemicals is available
for cellular uptake. The cell concentration at steady state cannot exceed the
partition constant between cell and water, and hence the cellular concentration is
much smaller than in serum-free medium.
The relative effect potency (REP) derived from nominal concentrations is
actually much closer to the REP for cellular concentrations than if derived from
freely dissolved concentrations (Fischer et al., 2017). This is intuitively easy to
understand because partitioning between cells that are made up of proteins and
lipids and medium that is made up of proteins and lipids is fairly uniform across
a wide range of chemicals (Jahnke et al., 2016). The freely dissolved fraction is
dependent on the hydrophobicity of the chemicals with very low freely dissolved
concentrations of hydrophobic chemicals versus much higher freely dissolved
concentrations of hydrophilic chemicals at the same nominal concentration
(Figure 9.13). For chemicals with log Kow . 3, the cellular concentration is lower
but directly proportional to the nominal concentration; even for chemicals with
log Kow below 3, the decrease of cellular concentration is not excessive.
Serum-mediated passive dosing works well in practice for water quality testing,
particularly considering that a water extract contains thousands of chemicals. Thus,
solubility of individual mixture components rarely poses a problem, especially

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 162 Bioanalytical Tools in Water Quality Assessment

Figure 9.13 Relative concentrations at concentration nominal concentration in

medium, cells and freely dissolved. Calculated with a mass balance model for
56 000 cells in 100 μL medium supplemented with 10% FBS (Armitage et al.,
2014). Reprinted with permission from Jahnke et al. (2016). Strategies for
transferring mixtures of organic contaminants from aquatic environments into
bioassays. Environmental Science & Technology, 50(11): 5424–5431. © 2016.
American Chemical Society. Kow = octanol–water partition constant.

given that a water extract contains few very hydrophobic but many hydrophilic and
many charged chemicals.
While media containing proteins and lipids stabilise the bioassay system, the
disadvantage is that if the composition of the medium changes and cell numbers
are variable, then the nominal concentration will not be directly related to the
biologically effective cellular concentration. Therefore, it is very important to
characterise the media used to ensure as little variability as possible and to use
the same assay components and protocols across all water sample types and
single chemicals.
If the serum is replaced by non-animal alternatives, the impact of lowering the
sorptive capacity of the medium needs to be accounted for. Likewise, for
serum-free set ups of cell assays, cell-free assays and whole organism assays with
invertebrates and fish embryos, dosing and exposure assessment needs to be
carefully set up to avoid any artefacts due to depletion of the sample. In these
circumstances, passive dosing from a polymer might be a viable alternative. In
passive dosing, the chemical or sample is partitioned into a polymer. The loaded
polymer, once inserted in the bioassay well, slowly releases the chemical to the
medium from which it can be taken up by the cells (Smith et al., 2015; Smith and
Schäfer, 2016). In case of adherent cells, the set-up might be modified, that the
cells do not touch the passive dosing device, for example, by growing the cells
on inserts hung into the dosing plate (Kramer et al., 2010). Passive dosing is
especially important for assays like the 96-h fish embryo toxicity assay, where
even in glass vials the uptake of chemical into the organism might lead to

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 163

depletion of the aqueous concentration (Seiler et al., 2014; Vergauwen et al., 2015).
Passive dosing may also overcome solubility limitations when testing hydrophobic
chemicals (Smith et al., 2010).

9.4.3 Metabolism in cell-based bioassays

Toxicokinetics are more complex in tissues, organs and whole organisms than in
cells. The liver has the most metabolically active cells. However, the principles
that govern an organism’s toxicokinetics and the processes of cellular uptake are
the same and the liver cell line HepG2 is often used to simulate liver metabolism.
The liver can generate circulating metabolites and this process can be simulated
in vitro. In addition, metabolites can be formed in the medium by adding a liver
enzyme mixture (e.g., S9 fraction) to the cell medium of in vitro assays to
simulate hepatic metabolism (Figure 9.14, extracellular approach). Alternatively,
if cells express metabolising enzymes, metabolites can also be generated inside
the cell (Figure 9.14, intracellular approach). Not all cells lines are metabolically
active. Cancer-cell derived reporter gene assays are often considered to have
limited metabolic capacity, but this really depends on the cell type. For instance,
the HepaRG cell line derived from a human hepatocellular carcinoma has very
active metabolic enzyme systems constitutively expressed (Guillouzo et al.,
2007). Exposure to chemicals might also induce metabolic enzymes in reporter
gene assays that are not constitutively expressed, which has to be considered in
these approaches (Fischer et al., 2020). It is also possible to retrofit cells with
metabolising enzymes which has potential to make in vitro assays more realistic
(Thomas et al., 2019).

9.5 BASELINE TOXICITY AND SPECIFICITY OF

RESPONSE
Baseline toxicity is the minimum effect any chemical elicits and can therefore serve
as an anchor to relate effects to (Chapter 4). Baseline toxicants were shown to have
an average constant membrane concentration of 69 mmol/Lmembrane (95%
confidence interval 49–89) across several cell lines (Escher et al., 2019). Mass
balance models can be used to translate this critical membrane concentration back
to the nominal inhibitory concentration that reduces cell viability and growth by
10% IC10,nom (Equation 9.1).


Vflipid,cell Vcell Kmedium/water
IC10,nom = 69 mM · · + (9.1)
fmembrane,cell Vmedium + Vcell Kcell/water

The only determinants going into this equation are the volume fraction of
the lipids in the cell Vflipid,cell, the fraction of chemicals in the membranes of
the cell fmembrane,cell, the volume of cell Vcell and medium Vmedium and the
partition constants between cell and water Kcell/water and medium and water

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 164 Bioanalytical Tools in Water Quality Assessment

Figure 9.14 Integrated strategy to model in vivo bioactivation and detoxification in a

diverse range of in vitro assays using either the extracellular or the intracellular
approach. Reprinted with permission from Thomas et al. (2019). The next
generation blueprint of computational toxicology at the U.S. Environmental
Protection Agency. Toxicological Sciences, 169(2): 317–332. Published by Oxford
University Press on behalf of the Society of Toxicology. This work is written by U.S.
Government employees and is in the public domain in the U.S.

Kmedium/water. Details of the derivation of this mass balance model and typical cell
parameters are given in Escher et al. (2019). The IC10,nom increases with
increasing hydrophobicity and depends on the serum content (Figure 9.15a). The
empirical relationship between log Kow and log(1/IC10,nom) was previously fitted
as a linear regression in the range 0.6 , log Kow , 4.3 (Escher et al., 2019).
Experimental data agreed rather well with the non-linear relationship predicted
by the mass balance model, which extends the applicability range to higher
hydrophobicity.
Any specifically acting chemical has cytotoxic concentrations below baseline
toxicity. The toxic ratio (TR) is a measure of how much more cytotoxic a
chemical is compared to its baseline toxicity (Equation 9.2, Figure 9.15b).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 165

Figure 9.15 (a) Relationship between hydrophobicity expressed as octanol−water

partition constant log Kow and nominal inhibitory concentration that reduced cell
viability and growth by 10% log IC10,nom predicted with the mass balance model
described in Escher et al. (2019). (b) Derivation of the specificity ratio SRcytotoxicity
(Equation 9.3) and SRbaseline (Equation 9.4) from the experimental cytotoxicity IC10
and the predicted baseline cytotoxicity IC10 and the measured effect concentration
for effect y, ECy. Effect y may refer to 10% or induction ratio IR 1.5.

A TR ≥ 10 is associated with specific or reactive toxicity (Maeder et al., 2004).

IC10,baseline
TR = (9.2)
IC10
Cytotoxic chemicals with a high TR are often reactive chemicals or chemicals
where the MIE or KE detected by the reporter gene assays cannot be
differentiated from the final adverse outcome at the cellular level because the
effect leads directly to the adverse outcome of reduced cell viability. Chemicals
with a high TR also often exhibit multiple modes of action. Hence a TR analysis,
although hardly ever performed, could be very important for the interpretation of
bioassay results. It makes a difference if a chemical is toxic because it is just
hydrophobic (baseline toxicity) or if it exhibits specific or reactive toxicity on the
apical endpoint of cytotoxicity.
We define the specificity ratio (SR) as the ratio between the experimental
cytotoxicity (IC10) and effect concentration (EC10 or ECIR1.5) (Escher et al.,
2020b). As is illustrated in Figure 9.15b, the specificity ratio can either relate to
the experimental IC10 (SRcytotoxicity, Equation 9.3) or to the predicted IC10,baseline
(SRbaseline, Equation 9.4).
IC10 IC10
SRcytotoxicity = or SRcytotoxicity = (9.3)
EC10 ECIR1.5
IC10,baseline IC10,baseline
SRbaseline = or SRbaseline = (9.4)
EC10 ECIR1.5

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 166 Bioanalytical Tools in Water Quality Assessment

SRbaseline is a measure of the specificity of a chemical acting in a given bioassay. The

higher SRbaseline the more specific the chemical. We will use the SRbaseline in Chapter
13 to classify bioassays and to derive effect-based trigger values.
SRcytotoxicity is a measure of the selectivity of a chemical in a given bioassay. If all
chemicals tested have a low selectivity, the assay has little value for evaluation of
chemicals and water samples. Reporter gene assays for nuclear receptors related
to endocrine effects often have very high selectivity for individual chemicals.
What kills the cell is not estrogenicity but genotoxicity, so the reporter gene and
the cause of cytotoxicity are not directly connected.
Assays for adaptive stress response are generally less selective because the
dysfunction and induction of the adaptive stress response is eventually what kills
the cell (e.g., due to excess reactive oxygen species). An extreme example of low
selectivity is a reporter gene assay indicative of the activation of the p53
pathway, which is an adaptive stress response to DNA damage that triggers
repair, cell cycle arrest and apoptosis. Chemicals that activate p53 are often
cytotoxic with a high TR. This also has an impact on activation of the reporter
gene. The gap between activation of p53 and cytotoxicity is small, even for
single reactive chemicals (Stalter et al., 2016a). When testing disinfected water
samples, the activation was in most cases masked by cytotoxicity (Yeh et al.,
2014) making this assay unsuitable for water quality testing.
If SRbaseline is high but SRcytotoxicity is low, the effect is mainly due to
cytotoxicity, not the specific endpoint. If both SRbaseline and SRcytotoxicity are low,
the chemical is a baseline toxicant with respect to cytotoxicity and not
specifically acting. In both cases, when the specific effect occurs at
concentrations close to cytotoxicity, the cytotoxicity burst phenomenon can lead
to expression of specific endpoints, which are then just an experimental artefact
(Judson et al., 2016; Fay et al., 2018; Escher et al., 2020c).
The combination of experiments and modelling helps to better understand the
responses that can be measured with in vitro assays and can scrutinise effect data
to be included in mixture models or other interpretation frameworks.

9.6 PRACTICAL CONSIDERATIONS FOR DOSING OF

CHEMICALS
The considerations on baseline toxicity and specificity can help us also in practice.
When planning an experiment with single chemicals or defined mixture of
chemicals, the information in this section can be synthesised to plan a successful
experiment. Our recommendation is to plan the highest concentration at
maximum solubility in the medium and to consider the implications of baseline
toxicity. The medium solubility Smedium can be calculated from Equation (9.5)
(Fischer et al., 2019).
Smedium = Swater · K medium/w (9.5)

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 In vitro assays for the risk assessment of chemicals 167

The aqueous solubility Swater can be retrieved from databases such as the
CompTox Chemistry Dashboard (U.S. EPA, 2020) and the medium–water
partition constant Kmedium/w (Equation 9.6) can be approximated from the
protein−water partition constant KBSA/w and the liposome−water partition
constant Klip/w and the fraction of FBS added to the medium βFBS. Evidently, the
composition of FBS is somewhat variable with respect to protein and lipid content
but to simplify the Equation (9.6) we have assumed an average composition
(Fischer et al., 2019). In turn, log KBSA/w (Equation 9.7) and log Klip/w
(Equation 9.8) can be simply predicted from the log Kow for neutral chemicals.
Kmedium/w = 0.046 · bFBS · K BSA/w + 0.0015 · bFBS · K lip/w
(9.6)
+ 0.9525 · bFBS + (1−bFBS )
log K BSA/w = 0.71 · log K ow + 0.42 (9.7)
log K lip/w = 1.01 · log K ow + 0.12 (9.8)
With these equations it is possible to build a fairly robust estimate of medium
solubility, which is much higher than the aqueous solubility for hydrophobic
chemicals (Fischer et al., 2019).
One must also consider the conditions under which the medium can act as a
chemical reservoir. The depletion of the freely dissolved concentration should be
,5% over the 24 hours duration of the assay. If we consider non-volatile and
stable chemicals and the main loss being binding to the polystyrene of the well
plate, then we need between 3% and 10% FBS as a minimum to assure
non-depletive conditions (Table 9.2).
We illustrate the concept on the example of serum-supplemented medium in
Table 9.2. There is a trend of replacing FBS with animal-free products to make
in vitro assays truly free of animal-derived products. If such alternative media are
used, they should be chemically defined to assure consistent outcomes and
synthetic proteins could be added to the desired level of retaining capacity.
A final consideration is to test up to baseline toxicity. Remember that the baseline
toxicity IC10 can be calculated using Equation (9.1) and estimated from the log Kow
in Figure 9.15a. As baseline toxicity is the minimum toxicity every chemical has,
baseline toxicity occurs at the highest possible concentration. Reporter gene

Table 9.2 Standard set-up of an HTS in vitro cell assays that assures that depletion of
the freely dissolved concentration is , 5% over the 24-h assay duration (Fischer et al.,
2019).

Plate format 96-well 384-well 1536-well

Medium volume (μL) 120 40 6

Cell number 10 000 5000 2000
Required % FBS ≥3 ≥5 ≥10

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 168 Bioanalytical Tools in Water Quality Assessment

activation is only valid if it occurs at a lower concentration than baseline toxicity and
any specific cytotoxic effect also occurs at lower concentrations (Escher et al.,
2020c). Hence, baseline toxicity should always be predicted for individual
chemicals, compared with medium solubility and ideally used as the highest test
concentration. More practical recommendations for dosing can be found in
Fischer et al. (2019).

9.7 CONCLUSIONS
In this section we have explored the exciting advances in HTS assays over the last
10 years – a new chapter that did not exist and could not have been told in the
first edition of this book.
The new European chemicals legislation REACH, ToxCast and Tox21 in the
U.S., and researchers and agencies worldwide have made it possible to do what
was considered impossible two or three decades ago: assessing the risk of
chemicals to the environment and to human health based on in vitro methods and
in silico models. This is only possible because of sound scientific work anchored
against classical animal-based methods for validation purposes. We have outlined
the first applications to give an idea of what lies ahead in the exciting field of NGRA.
We are not at the end of the story but still at the beginning. Despite breath-taking
advances there remain a lot of scientific questions to be solved. The first generation
of HTS assays, mainly receptor binding assays and reporter gene assays, have
limitations and new tools are emerging that improve individual assays but also go
in entirely new directions such as the assays using HTTr and HT phenotyping.
A downside of HTS relates to limitations of dosing and exposure assessment.
SPME methods to quantify the freely dissolved concentration as proxy of the
BED are at present limited to single chemicals and defined mixtures. There is
always a trade-off between exactly characterising the dosing in bioassays versus
running hundreds of samples in screening mode. Since we will never know the
concentrations of all chemicals in a water extract, it is futile to aim and
characterising exposure concentrations in cell-based assays for water extracts.
Instead, we have to ensure that there is a robust relationship between the nominal
and the cellular dose.
The progress in applying HTS for risk assessment also paved the way for
applications in monitoring. In vitro assays have long been used for water quality
monitoring, but the stronger scientific underpinning will help their acceptance in
regulation in the future.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Chapter 10
Current bioanalytical tools for
water quality assessment

10.1 INTRODUCTION
In vitro bioassays that respond to relevant initiating triggers in a toxicity pathway or
are linked to a known mode of toxic action with a defined health outcome have the
potential to be useful tools for water quality assessment. Some assays are, however,
not suitable for screening of water samples. Assay robustness and specificity in the
presence of water matrix components and other chemicals must be characterised and
validated prior to their implementation as monitoring tools. This chapter reviews
in vitro bioassays and well plate-based in vivo assays that have already been
applied for water quality monitoring in drinking water, surface water, wastewater
and recycled water (up to September 2020). Specifically, we focus on assays
adapted to high-throughput screening (e.g., run in a 96-well or 384-well plate)
because high throughput is essential for large numbers of samples in routine
water quality monitoring. Furthermore, we only considered studies where water
samples were extracted by solid-phase extraction (SPE), passive sampling or
liquid–liquid extraction (LLE), rather than testing unenriched or native water
samples, which is covered in Chapter 3 on whole effluent toxicity testing. Native
water samples may contain metals, salts and other inorganics, in addition to
micropollutants, meaning that the response in a native water sample cannot be
attributed to organic micropollutants alone. It should be noted that despite our
best efforts this list may not include all deployed assays, but it should
nevertheless provide a good perspective on the types of assays currently in use
for water quality monitoring. Assays that have so far been limited to chemical

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0169

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 170 Bioanalytical Tools in Water Quality Assessment

Figure 10.1 Toxicity pathway in a cell (modified from Chapter 4).

risk assessment of individual compounds or mixtures of known composition

were excluded.
As explained in Chapter 4, cells can be thought of as models of whole organisms
that simulate many crucial processes. Using a cell system ensures, for example, that
relevant toxicokinetic processes are accounted for in the bioassay. Cellular
responses can then be visualised along the three steps of the cellular toxicity
pathway, from the molecular interaction to the cellular effect (Figure 10.1).
Some bioassays are capable of detecting the initiating event, that is, the
interaction of the micropollutant with its biological target. The initiating event is
considered a measure of potential effect, as later repair and defence mechanisms
may be able to prevent an adverse outcome. From a precautionary risk
perspective, the potential to do harm is an important assessment endpoint. Often,
however, it is not possible to measure the initiating event directly. In such cases,
a later step in the cellular toxicity pathway can be quantified as an assay
endpoint, either using a natural biomarker or a reporter gene product. Finally,
cytotoxicity (cell death or decreased growth) can be quantified as an integrative
parameter for all toxicity pathways in a cell (Figure 10.1). In addition to in vitro
assays, well plate-based in vivo assays can capture effects from multiple cellular
toxicity pathways resulting in an adverse outcome measured by apical effects
such as mortality, growth and development impairment (Wernersson et al., 2015).
Bioassays based on different stages of cellular toxicity pathways including
induction of xenobiotic metabolism, receptor-mediated effects, adaptive stress
responses and apical effects have been widely applied to evaluate different water
extracts (e.g., Escher et al., 2014; Rosenmai et al., 2018; Alygizakis et al., 2019).
Many of the bioassays included have also been applied to support health risk
assessment of chemicals (Wetmore, 2015; Bell et al., 2018) and this information
is useful to estimate how the in vitro endpoints are connected to health
consequences. However, it must be stressed that the use of in vitro bioassays to
water samples does not, at this stage, allow prediction of likely health risk.

10.2 PRINCIPLES OF CELL-BASED BIOASSAYS

The cells used in bioanalytical tools can be primary cells or immortalised cell lines.
Primary cells are excised from living organisms (e.g., primary hepatocytes are
isolated from liver tissue) and are usually more representative of ‘natural’ cellular

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 171

function in the organism. Primary cells have a limited lifespan in vitro and
eventually stop dividing. Not only does this reduce the ethical benefit of using
bioanalytical tools, but it also increases the variability of the results as different
batches of cells may be sourced from different individuals. Immortalised cell
lines, on the other hand, have been either accidentally or deliberately mutated to
proliferate indefinitely. This means that a constant supply of identical cells can be
produced to ensure minimal variability between different experiments using the
same cell lines. Due to mutations, gene expression can change over time even in
immortalised cells. It is therefore important to keep the passage number (an
indicator of the age of the cell culture) within 40–80 passages from the source
cell, depending on the cell type.
A variety of cells and cell lines can be used in cell-based bioassays. Almost any
organism can be used, including humans, other mammals such as mice and
monkeys, simple eukaryotes such as plants and yeast, or prokaryotic organisms
such as bacteria. Higher forms of life are complex multi-cellular organisms with
many different cell types, many of which can be cultured in vitro and used as
models for cell-based bioassays.
Growth rate and cell viability can be determined in all cell-based bioassays as a
measure of non-specific cytotoxicity. The measurement is carried out either by
direct counting of cells using a specialised cell counter, a flow cytometer, a
haemocytometer, microscope imaging methods or through indirect measurement
of cellular activity such as metabolic and mitochondrial activity, active transport
mechanisms and/or cell membrane permeability.
In some instances, the initiating event causes a specific detectable cellular
response in native cells. This is called a biomarker and can be a change of state
or the production of a specific chemical or protein. The production of vitellogenin
in fish liver cells upon exposure to estrogenic compounds is a typical example of

Figure 10.2 Types of cell-based assays and assessment endpoints. MOA = mode of
action.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 172 Bioanalytical Tools in Water Quality Assessment

a biomarker. Often the native response is difficult to measure, and cells can be
genetically modified into recombinant cells that produce a more visible (and
hence measurable) effect in response to the initiating event (Figure 10.2).
The reporter gene assay is one example of genetic engineering used to enhance
our visualisation of the cellular response. In a reporter gene assay, a gene encoded
with an easily detectable product such as a fluorescent protein or enzyme (e.g.,
luciferase or β-galactosidase) is paired with a promoter region specific for the
mode of action to be investigated. When the initiating event triggers the cellular
response, the reporter gene is transcribed to messenger RNA, which is then
translated as a fluorescent protein or enzyme that can be measured by
fluorescence or enzymatic assays (Figure 10.3).
The production of the reporter product by the cell’s genetic machinery is
proportional to the induction of the receptor: the stronger the stimulation, the
more reporter is produced. In some recombinant cells, multiple copies of both
the promoter and the reporter are included to improve sensitivity. A typical
example of a reporter gene technique is the AhR CAFLUX assay (Nagy et al.,
2002). This assay uses a mouse hepatoma cell (Hepa1c1c7) stably transfected
with a plasmid containing enhanced green fluorescent protein (EGFP) as
reporter protein downstream of a promoter consisting of four dioxin-response
elements (DREs). These recombinant cells produce EGFP upon exposure to
dioxin-like compounds, and the amount of EGFP is dependent on the amount
of aryl hydrocarbon receptor (AhR) stimulation from the sample. Reporter gene

Figure 10.3 Principle of the reporter gene assay. A plasmid containing a reporter
gene downstream of the natural gene promoter is inserted into a recombinant cell,
and activation of the promoter results in production of a fluorescent reporter protein.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 173

assays indicative of different endpoints are commonly used for water quality
assessment; indeed, the majority of assays reviewed in this chapter are reporter
gene assays.

10.3 BIOASSAYS INDICATIVE OF XENOBIOTIC

METABOLISM
The presence of chemicals can induce biotransformation processes in cells to
metabolise, detoxify or in some cases bioactivate chemicals (Omiecinski et al.,
2011). Some important xenobiotic metabolism receptors include the AhR, the
peroxisome proliferator-activated receptor (PPAR), the pregnane X receptor
(PXR) and the constitutive androstane receptor (CAR). Assays indicative of
xenobiotic metabolism may not result in cell death, but instead can act as
sensitive indicators of the presence of chemicals, with many receptors considered
capable of binding a wide range of chemicals. To date, only a few studies have
applied assays indicative of CAR to environmental extracts (Escher et al., 2014),
with effects detected in wastewater, surface water and drinking water extracts
using a yeast-based CAR assay or the Attagene-multiplexed assay (Blackwell
et al., 2019). Therefore, this section will focus on assays indicative of the three
xenobiotic metabolism receptors commonly applied to water samples: AhR,
PPAR and PXR.

10.3.1 Aryl hydrocarbon receptor

The AhR is a ligand-dependent transcription factor that is necessary for virtually all
of the toxicity of halogenated aromatic hydrocarbons such as polychlorinated and
brominated dibenzo-p-dioxins and biphenyls as well as polycyclic aromatic
hydrocarbons (PAHs). It activates target genes encoding for the metabolic
enzymes CYP1A1, CYP1B1 and NADPH-quinone oxidoreductase (NQO1) but
there is also cross-talk with Nrf2, the master regulator of the antioxidant
response, and the hypoxia-inducible factor HIF1α. Its activation contributes to
carcinogenicity because cytochrome P450 monooxygenase (CYP) can convert
many of its ligands to reactive intermediates capable of interacting with and
causing DNA damage. Assays indicative of activation of AhR are traditionally
applied to detect the presence of dioxin-like chemicals, but recent studies have
shown that many environmental chemicals can also activate the AhR pathway
(Martin et al., 2010).
Some of the more common reporter gene assays that have been applied to
evaluate activation of AhR in water extracts are provided in Table 10.1, along
with the reported concentration causing 10% effect (EC10). Generally, AhR
CAFLUX (mouse H1.G1.1c3 and rat H4.G1.1c2), AhR CALUX (rat H4L1.1c4)
and H4IIE-luc are similarly sensitive to individual chemicals, with EC10 values in
the low ng/L range for reference compound 2,3,7,8-tetrachloro-p-dibenzodioxin

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 174 Bioanalytical Tools in Water Quality Assessment

Table 10.1 Common cell-based reporter gene assays applied to evaluate aryl
hydrocarbon receptor (AhR) activity in water extracts.
Assay Cell Line Detection TCDD TCDD EC
Method EC10 (M) EC10 Reference
(ng//L)
AhR H1.G1.1c3 Fluorescence 8.89×10−13a 0.29 König et al.
CAFLUX (2017)
AhR H4.G1.1c2 Fluorescence 5.22×10−13a 0.19 König et al.
CAFLUX (2017)
AhR H4L1.1c4 Luminescence 5.92×10−13 0.19 Nivala et al.
CALUX (2018)
AhR HepG2 Luminescence 6.22×10−11a 20 Rosenmai
HepG2 et al. (2018)
H4IIE-luc H4IIE Luminescence 1.60×10−13 0.05 Lee et al.
(2015)
a
EC10 value converted from EC50 value assuming a slope of the log-logistic concentration
response curve of 1.

(TCDD). It should be noted that there are other AhR CALUX cell lines used, for
example, H1L6.1c2 (Mehinto et al., 2017) and H4L1.1c2 (Daniels et al., 2018),
but these have not been as widely applied as the others to date. The human
HepG2-based AhR reporter gene assay is around two orders of magnitude less
sensitive than the other reporter gene assays (Rosenmai et al., 2018) and has only
been applied in a limited number of studies.
In addition to the assays in Table 10.1, the PAH CALUX assay, which uses
H4IIE rat hepatoma cells, has also been applied to wastewater effluent
(Alygizakis et al., 2019) and surface water (de Baat et al., 2019a) extracts. The
assay reference compound is benzo[a]pyrene (EC50 3.0 × 10−9 M (Pieterse et al.,
2013)), with all results expressed as benzo[a]pyrene equivalent concentrations
(B[a]P EQ). Other studies have also applied yeast-based activation of AhR
assays, such as the yeast dioxin screen (YDS). The reference compound for the
YDS is β-naphthoflavone, with an EC50 of 3.0 × 10−8 M (Stalter et al., 2011).
An assay indicative of activation of AhR, AhR_LUC, was also included in the
U.S. EPA ToxCast database. This is based on the human HepG2 cell line (He
et al., 2011). TCDD was not measured in ToxCast, but the EC10 values of
common chemicals run in both AhR_LUC and AhR CALUX were generally
within an order of magnitude (Neale et al., 2020a).
In addition to reporter gene assays, the native metabolic enzyme activation can
also be quantified in cell lines. The ethoxyresorufin-O-deethylase (EROD) assay
measures specific CYP enzyme activity as an indicator of specific CYP isoforms

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 175

induction, namely CYP1A. This assay was first developed in 1974 (Burke and
Mayer, 1974). Early applications of the EROD microplate assay for water quality
testing include the study by Huuskonen et al. (1998), who assessed the toxicity of
lake water receiving paper mill effluents. Louiz et al. (2008) demonstrated, using
a fish liver cell line (PLHC-1), that PAHs show higher potency after 4 h of
exposure, whereas dioxins caused similar responses whether exposed for 4 or
24 h. In this way, the assay becomes somewhat specific to chemical groups by
varying the exposure duration. Although the cell-based assay has been further
improved and applied to different cell lines (Heinrich et al., 2014), it is still too
variable for routine monitoring. Ongoing improvements in the method may in the
near future make this assay a more suitable one.
A summary of reported AhR activity in wastewater, surface water, recycled water
and drinking water is provided in Table 10.2. A number of studies have evaluated
activation of AhR in passive sampler extracts (Jarošová et al., 2012; Hamers
et al., 2018; de Baat et al., 2020), but only studies that have applied SPE are
included in Table 10.2. Based on the more sensitive reporter gene assays, the
TCDD EQ covered the range of 0.1–3.3 ngTCDD/L in wastewater influent, 0.007–
1.2 ngTCDD/L in treated wastewater, 0.004–0.36 ngTCDD/L in recycled water and
0.002–0.21 ngTCDD/L in surface water. The reported wastewater treatment plant
(WWTP) removal efficacy ranged from 13% to 90% (Jalova et al., 2013; Nivala
et al., 2018). TCDD EQ in drinking water ranged from ,0.004 to 0.17
ngTCDD/L. Based on EC10 values, effects were detected after 0.7–0.8 times
enrichment in wastewater influent, between 0.8 and 31 times enrichment in
wastewater effluent and 0.4 and 89 times enrichment in surface water.

10.3.2 Peroxisome proliferator-activated receptor γ

PPAR is also a transcription factor that belongs to the superfamily of nuclear
receptors and is involved in the regulation of glucose and lipid metabolism, not
so much in xenobiotic metabolism (Scarsi et al., 2007). As the name indicates,
the main function of PPAR is the delivery of peroxisomes, which are important
for fatty acid oxidation and thus relevant for lipid metabolism. There are three
isoforms of PPAR – PPARα, PPARβ (also called δ) and PPARγ, which are
encoded by different genes, show different tissue expressions and perform
slightly different functions. PPARα is expressed predominantly in metabolically
active tissues such as liver and kidney cells where its ligands include fatty acids,
hypolipidemic drugs and xenobiotics (Seimandi et al., 2005). PPARγ is the key
receptor in maintaining glucose and lipid homeostasis and its activation increases
the insulin resistance of the cell (Scarsi et al., 2007). To date, most studies have
applied assays indicative of binding to PPARγ to environmental water extracts,
with only a few studies applying assays indicative of PPARα (Escher et al.,
2014; Alygizakis et al., 2019). Therefore, this section will focus on PPARγ, with

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 10.2 Summary of reported aryl hydrocarbon receptor (AhR) activity as TCDD EQ in different environmental water extracts.
Matrix Assay Activation of AhR Reference
TCDD EQ (ngTCDD/ L) 176
Wastewater influent AhR CAFLUX (H1.G1.1c3) 1.1–1.8 1
AhR CALUX (H4L1.1c4) 0.25–0.27 2
AhR reporter gene assay 187–386 3
H4IIE-luc ,0.1–3.3 4–5
Wastewater effluent AhR CAFLUX (H1.G1.1c3) 0.007–1.2 1, 6–9
AhR CAFLUX (H4G1.1c2) 0.063–0.10 10
AhR CALUX (H4L1.1c4) 0.12–0.13 2, 11
AhR reporter gene assay 97–168 3
H4IIE-luc ,0.05–0.7 4, 8, 12, 13
PAH CALUX 52–242b 14
YDS 16–158c 15
Recycled water AhR CAFLUX (H1.G1.1c3) ,0.007a–0.36 6–9
H4IIE-luc ,0.004 5, 8
Surface water AhR CAFLUX (H1.G1.1c3) 0.01–0.19 1, 8, 16
AhR CAFLUX (H4G1.1c2) 0.002–0.16 10, 16, 17

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

AhR CALUX (H4L1.1c4) 0.002–0.21 11, 18, 19, 20
AhR reporter gene assay 53 21
H4IIE-luc 0.009a–0.18 8, 13
YDS ,320–602c 22
Drinking water AhR CAFLUX (H1.G1.1c3) 0.024a–0.17 1, 8
AhR reporter gene assay 45–52 21
H4IIE-luc ,0.004a 8
Bioanalytical Tools in Water Quality Assessment

NB: Only studies that have applied SPE or LLE are included.
a
TCDD EQ calculated using EC10 values from Table 10.1.
b
PAH CALUX reference compound is benzo[a]pyrene.
c
YDS reference compound is β-naphthoflavone.
References: 1(Macova et al., 2011); 2(Nivala et al., 2018); 3(Lundqvist et al., 2019b); 4(Jalova et al., 2013); 5(Lee et al., 2015); 6(Macova et al., 2010);
7
(Reungoat et al., 2010); 8(Escher et al., 2014); 9(Jia et al., 2015); 10(Neale et al., 2017c); 11(Mueller et al., 2021), 12(Loos et al., 2013), 13(Maier et al.,
2016); 14(Alygizakis et al., 2019); 15(Stalter et al., 2011); 16(König et al., 2017); 17(Neale et al., 2015b); 18(Müller et al., 2018); 19(Neale et al., 2018a);
20
(Neale et al., 2020a); 21(Lundqvist et al., 2019a); 22(Brettschneider et al., 2019).
 Current bioanalytical tools for water quality assessment 177

Table 10.3 Common cell-based reporter gene assays applied to evaluate

peroxisome proliferator-activated receptor (PPARγ) activity in water extracts.
Assay Cell Line Detection Rosiglitazone Rosiglitazone EC
Method EC10 (M) EC10 (ng// L) Reference

PPARγ U2OS Luminescence 1.00 × 10−8 3600 Gijsbers

CALUX et al. (2011)
PPARγ HEK293H Fluorescence 3.30 × 10−10 118 Jia et al.
GeneBLAzer (2015)

two assays frequently applied, PPARγ CALUX and PPARγ GeneBLAzer

(Table 10.3). The EC10 value for reference compound antidiabetic pharmaceutical
rosiglitazone was over an order of magnitude lower for PPARγ GeneBLAzer.
PPARγ activity has been detected in wastewater influent, wastewater effluent and
surface water, with activity in recycled water and drinking water below the limit
of detection (Table 10.4). Activity in wastewater influent ranged from 500 to
936 ngrosiglitazone/L rosiglitazone EQ, with wastewater effluent ranging from 83 to
1680 ngrosiglitazone/L rosiglitazone EQ (Table 10.4). The studies that have

Table 10.4 Summary of reported peroxisome proliferator-activated receptor (PPARγ)

activity reported as rosiglitazone EQ in different environmental water extracts.
Matrix Assay PPARγ Activity Reference
Rosiglitazone EQ
(ngrosiglitazone/ L)
Wastewater influent PPARγ CALUX 500–800 1
PPARγ GeneBLAzer 719–936 2
Wastewater effluent PPARγ CALUX ,32–640 1, 3
PPARγ GeneBLAzer ,59–1680 2, 4, 5
Recycled water PPARγ CALUX ,119 a
4
PPARγ GeneBLAzer ,59a 4
Surface water PPARγ CALUX ,119 a
4
PPARγ GeneBLAzer 0.59–1092 5–9
Drinking water PPARγ CALUX ,119 a
4
PPARγ GeneBLAzer ,1.2 to ,59
a a
10
NB: Only studies that have applied SPE or LLE are included.
a
EC10 converted to Rosiglitazone EQ using EC10 values provided in Table 10.3.
References: 1(Bain et al., 2014); 2(Nivala et al., 2018); 3(Alygizakis et al., 2019); 4(Escher et al.,
2014); 5 (Mueller et al., 2021); 6(König et al., 2017); 7(Müller et al., 2018); 8(Neale et al.,
2018a); 9(Neale et al., 2020a); 10(Albergamo et al., 2020).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 178 Bioanalytical Tools in Water Quality Assessment

evaluated the removal of PPARγ during wastewater treatment found between 69%
and .94% removal (Bain et al., 2014; Nivala et al., 2018). The rosiglitazone EQ in
surface water varied from 0.6 to 1092 ngrosiglitazone/L, with sites downstream of
WWTPs having the highest effect. Based on EC10 values, effects were detected
after 0.2–0.3 times enrichment in wastewater influent, between 0.09 and .30
times enrichment in wastewater effluent and 0.2 and 97 times enrichment in
surface water.

10.3.3 Pregnane X receptor

PXR is a promiscuous nuclear receptor with a large ligand binding pocket that can
help protect the cell by triggering detoxification pathways (Grimaldi et al., 2015).
PXR controls the transcription of a large array of genes encoding for phase I
metabolic enzymes, especially the CYP3A family, which plays an important
role in drug metabolism. Two reporter gene assays have been applied to
evaluate PXR activity in water extracts, HG5LN hPXR and PXR CALUX. The
reference compound for HG5LN hPXR is the pharmaceutical SR12813 (EC10:
1.58 × 10−8 M (Neale et al., 2015b)), while the reference compound for PXR
CALUX is the pharmaceutical nicardipine. Effect concentrations for the
industrial compound di(2-ethylhexyl)-phthalate (DEHP) for both assays were
presented in Escher et al. (2018), with a slightly lower value for HG5LN hPXR
(Table 10.5). PXR activity was detected in all water samples tested, except for
recycled water after reverse osmosis and advanced oxidation (RO/AO) (Escher
et al., 2014). The reported PXR activity in wastewater effluent was 3.8–4.7
µgSR12813/L SR12813 EQ or 20–240 µgnicardipine/L nicardipine EQ, while the
activity in surface water ranged from ,0.02 to 2.3 µgSR12813/L SR12813 EQ
(Table 10.6). Based on the EC10 values in the HG5LN hPXR assay, effects in
wastewater were detected after around two times enrichment in wastewater
effluent, between 3 and 30 times enrichment in surface water and 2.5 times
enrichment in drinking water.

Table 10.5 Common cell-based reporter gene assays applied to evaluate activation
of the pregnane X receptor (PXR) in water extracts.
Assay Cell Line Detection DEHPa DEHP EC
Method EC10 (M) EC10 Reference
(µg// L)
HG5LN HG5LN Luminescence 2.77 × 108 Escher et al.
hPXR (HeLa) 10−7 (2018)
PXR U2OS Luminescence 3.97 × 155 Escher et al.
CALUX 10−7 (2018)
a
DEHP = Di(2-ethylhexyl)-phthalate.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 179

Table 10.6 Summary of reported pregnane X receptor (PXR) activity in different

environmental water extracts.
Matrix Assay PXR Activity Reference
a
Wastewater HG5LN hPXR 3.8–4.7 µg/L SR12813 EQ 1
effluent PXR CALUX 20–240 µg/L nicardipine EQ 2
Recycled water HG5LN hPXR ,0.66–0.98 µg/L SR12813 EQ a
1
Surface water HG5LN hPXR ,0.02–2.3 µg/L SR12813 EQ 1, 3, 4
a
Drinking water PPARγ CALUX 3.2 µg/L SR12813 EQ 1
NB: Only studies that have applied SPE or LLE are included.
a
EC10 converted to SR12813 EQ using EC10 values provided in Table 10.5.
References: 1(Escher et al., 2014); 2(Alygizakis et al., 2019); 3(Neale et al., 2015b); 4(Neale et al.,
2018a).

10.4 BIOASSAYS INDICATIVE OF HORMONE

RECEPTOR-MEDIATED EFFECTS
Hormonal pathways are essential for processes related to growth, sexual
development, metabolism and homeostasis. Endocrine-disrupting chemicals,
including synthetic hormones, industrial chemicals and pesticides can interfere
with hormonal systems by interacting with hormone receptors (le Maire et al.,
2010). This includes activating or inhibiting hormone receptors. To date, most
of the research has focused on the estrogen receptor (ER), followed by the
androgen receptor (AR). However, other relevant nuclear receptors include the
glucocorticoid receptor (GR), progesterone receptor (PR), thyroid receptor (TR),
mineralocorticoid receptor (MR), retinoic acid receptor (RAR) and retinoid X
receptor (RXR). Further information about the sensitivity of assays indicative of
ER, AR, PR, GR and TR can be found in the review by Leusch et al. (2017). In
addition to the more commonly applied hormone receptors, this section will also
focus on assays indicative of MR, RAR and RXR. Due to the large number of
studies that have applied assays indicative of hormone receptor-mediated effects
we decided to focus on the more commonly applied assays.

10.4.1 Estrogen receptor

10.4.1.1 Estrogen receptor agonism
Nuclear estrogen receptors ERα and ERβ are important for the growth and
homeostasis of the uterus and mammary glands, as well as bones and the
cardiovascular system (le Maire et al., 2010). The majority of assays applied to
environmental water extracts focus on ERα (e.g., ERα CALUX and ERα
GeneBLAzer), although the T47D-KBluc assay uses the T47D cell line, which
expresses both ERα and ERβ (Wilson et al., 2004).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 180 Bioanalytical Tools in Water Quality Assessment

Estrogenic activity is the most commonly studied endpoint in water extracts. As

many different assays have been applied in the literature, we focused on assays that
have been applied to water samples in four or more studies. The exception was the
embryonic zebrafish assay EASZY, which has only been included in two studies,
but was included to represent a whole organism assay. A summary of the
included activation of ER assays is provided in Table 10.7, with similar
sensitivity across the mammalian reporter gene assays. The EC10 values of the
reference compound 17β-estradiol (E2) vary between 0.13 ng/L for T47D-KBluc
and 2.1 ng/L for HeLa-9903. The yeast estrogen screen (YES) and EASZY are
less sensitive. Further information about the sensitivity of many of these assays
can be found in Leusch et al. (2017).
Not only does the sensitivity differ between different bioassays for the reference
compound E2 but there are also differences in the relative effect potency (REP) of
different compounds. The REP relates the effect of an individual chemical to the
effect of the assay reference compound with REP values close to 1 indicating that
the individual chemical has a similar potency to the reference chemical and REP
,1 indicating the individual chemical is less potent than the reference chemical
(Chapter 7). The natural estrogen estrone (E1) has an REP of 0.02 in ERα
CALUX but an REP of 0.10 in ERα GeneBLAzer (Escher et al., 2018)
(Figure 10.4). Therefore, E1 is more potent relative to E2 in ERα GeneBLAzer
than ERα CALUX.
Estrogenic activity is reported as 17β-estradiol EQ (EEQ) in units of ngE2/L in
wastewater influent, wastewater effluent, recycled water, surface water and
drinking water in Table 10.8. Focusing on the mammalian reporter gene assays,
the estrogenic activity expressed as EEQ in wastewater influent ranged from 0.5
to 122 ngE2/L, while the EEQ were mostly between 0.1 and 10 ngE2/L in treated
effluent. Estrogenic activity is typically well removed during wastewater
treatment, with 80% to .99% removal efficacy reported in the literature for a
variety of WWTPs (Jugan et al., 2009; Jalova et al., 2013; Bain et al., 2014;
Hamilton et al., 2016; Houtman et al., 2018; Nivala et al., 2018). Estrogenic
activity was mostly below the limit of detection in recycled water. Estrogenicity
varied greatly in surface water, with EEQ from 0.005 up to 190 ngE2/L, with
factors such as proximity to wastewater effluent discharges impacting the
observed effects. Finally, low estrogenic activity (EEQ , 0.01 ngE2/L) was often
detected in treated drinking water, with one study from China finding EEQ of
5.3 ngE2/L in treated drinking water (Shi et al., 2018). A number of studies have
measured estrogenic activity in both source water and treated drinking water,
with 39%–99% removal efficacy observed (Escher et al., 2014; Huang et al.,
2016; Lv et al., 2016; Xiao et al., 2016; Conley et al., 2017b; Xiao et al., 2017;
Shi et al., 2018; Neale et al., 2020b). Based on the EC10 values, estrogenic
activity was detected after 0.1–6.4 times enrichment in wastewater effluent, 0.09–
145 times enrichment in surface water and 20–110 times enrichment in
drinking water.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 10.7 Common assays applied to evaluate estrogenic activity in water extracts.
Assay Cell Line// Test Detection Method 17β-Estradiol 17β-Estradiol EC Reference
System EC10 (M) EC10 (ng// L)
Yeast reporter gene
YES Yeast Absorbance 3.75 × 10−11a 10.2a Escher et al.
(2008b)
Mammalian reporter gene
ERα CALUX U2OS Luminescence 7.13 × 10−13 0.19 Jia et al. (2015)
ERα HEK293T Fluorescence 9.87 × 10−12 2.7 Nivala et al.
GeneBLAzer (2018)
HeLa-9903 HeLa Luminescence 7.78 × 10−12a 2.1a Valcarcel et al.
(2018)
MELN MCF-7 Luminescence 2.42 × 10−12 0.66 Neale et al.
(2015b)

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

MVLN MCF-7 Luminescence 3.16 × 10−12a 0.86a Shue et al.
(2009)
T47D-KBluc T47D Luminescence 4.63 × 10−13a 0.13a Liu et al. (2018)
Cell proliferation
E-SCREEN MCF7 Absorbance (cell 8.18 × 10−13a 0.22a Macova et al.
viability (2010)
measured using
CellTiter (MTS))
Current bioanalytical tools for water quality assessment

Whole organism
EASZY Embryonic Fluorescence EC50 6.20 × 10−10 EC50 168 Brion et al.
zebrafish (2019)
181

a
Presented EC10 value converted from EC50 value assuming a slope of the log-logistic concentration response curve of 1.
 182 Bioanalytical Tools in Water Quality Assessment

Figure 10.4 Relative effect potency (REP) of estrone (E1) and 17α-ethinylestradiol
(EE2) in different assays indicative of estrogenic activity. Data available in
Appendix A of Escher et al. (2018), for T47D-KBluc in Bermudez et al. (2012) and
for E-SCREEN in Soto et al. (1995).

10.4.1.2 Estrogen receptor antagonism

In contrast to estrogenic activity, anti-estrogenic activity is much less studied in
environmental water extracts. Three assays commonly applied to evaluate
anti-estrogenic activity include the yeast anti-estrogen screen (YAES) and the
mammalian reporter gene ERα CALUX and ERα GeneBLAzer (Table 10.9).
Based on the reference compound tamoxifen, the ERα CALUX was the most
sensitive assay, with an effect concentration causing a suppression ratio (SPR) of
20% (ECSPR20) value of 0.56 µg/L. Cytotoxicity masked anti-estrogenic activity
in wastewater influent, while many treated wastewater effluent samples were
below detection (Table 10.10). Using the YAES assay, three studies found
tamoxifen EQ (TMX EQ) ranging from 0.7 to 97 µgtamoxifen/L in treated effluent
(Conroy et al., 2007; Fang et al., 2012; Archer et al., 2020). Anti-estrogenic
activity was either low or below detection in surface water, while no
anti-estrogenic activity was detected in drinking water.

10.4.2 Androgen receptor

10.4.2.1 Androgen receptor agonism
The androgen receptor (AR) is expressed in a range of tissues and has implications
for the development and maintenance of a number of systems, including the
reproductive, immune, musculoskeletal and cardiovascular systems (Davey and
Grossmann, 2016). Although a number of assays have been applied to evaluate
androgenic activity, the assays most commonly applied to water extracts include
the yeast androgen screen (YAS) and the mammalian reporter gene assays AR
CALUX, AR GeneBLAzer and MDA-kb2 (Table 10.11). Based on the reference
compound dihydrotestosterone (DHT), the mammalian reporter gene assays were
more sensitive compared to YAS. Further information about assay sensitivity can
be found in Leusch et al. (2017).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 10.8 Summary of estrogenic activity reported as 17β-estradiol EQ (EEQ) in different environmental water extracts.
Matrix Assay Activation of ER Reference
EEQ (ngE2/ L)
Wastewater influent YES 4.4–86 1–5
ERα CALUX 0.5–122 3, 6–9
ERα GeneBLAzer 11–24 10
MELN 15–94 11, 12
MVLN 5.4–124 2, 13
E-SCREEN ,0.02–225 14–16
Wastewater effluent YES ,0.1–91 1–5, 17–25
ERα CALUX ,0.006–22.9 3, 6–9, 19, 22, 25–31
ERα GeneBLAzer 0.03–151 10, 19, 28, 32–36
hERα-HeLa-9903 0.03–24 19, 28, 37
MELN 0.04–24 11, 12, 28, 38, 39
MVLN 0.1–24 2, 13, 40, 41
T47D-KBluc 15 42

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

E-SCREEN ,0.02–34 14–16, 19, 37, 43–46
EASZY ,6.3–673 39, 47
a
Recycled water YES ,0.34 19
ERα CALUX ,0.008a–0.17 19, 27, 29
ERα GeneBLAzer ,0.14a–2.6 19, 34
hERα-HeLa-9903 ,0.08a 19
E-SCREEN ,0.007a to ,0.06 15, 16, 19, 45, 46
Current bioanalytical tools for water quality assessment

Surface water YES ,0.07–324 1, 3, 17–19, 21, 24, 48–55

ERα CALUX ,0.015–129 3, 8, 19, 28, 30, 31, 56–59
ERα GeneBLAzer 0.005–39 19, 28, 32, 33, 35, 36, 60–68
hERα-HeLa-9903 ,0.02–2.0 19, 28, 69
183

MELN 0.003–23 11, 12, 28, 38, 39, 65, 70–73

(Continued)
by guest
Table 10.8 Summary of estrogenic activity reported as 17β-estradiol EQ (EEQ) in different environmental water extracts
(Continued ).
184

Matrix Assay Activation of ER Reference

EEQ (ngE2/ L)
MVLN 1.2–32 40, 74
T47D-KBluc ,0.03–190 42, 75, 76, 77
E-SCREEN 0.005–85 15, 16, 19, 37, 43, 62, 77, 78
EASZY ,2.0–30 39, 47
Drinking water YES 0.02–1.4 19, 51, 53, 54, 79
ERα CALUX ,0.008a–5.3 19, 30, 56, 59
ERα GeneBLAzer ,0.03–0.04 19, 33, 66, 80
hERα-HeLa-9903 ,LOD to 0.35a 19, 81
MELN ,0.3 12
T47D-KBluc ,0.025–0.11 42, 76, 79

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

E-SCREEN ,0.007a 16, 19
NB: Only studies that have applied SPE or LLE are included.
a
EC10 converted to 17β-estradiol equivalent concentrations EEQ using EC10 values provided in Table 10.7; LOD: limit of detection.
References: 1(Archer et al., 2020); 2(Kusk et al., 2011); 3(Murk et al., 2002); 4(Stalter et al., 2011); 5(Zhang et al., 2011); 6(Bain et al., 2014); 7(Houtman
et al., 2018); 8(Roberts et al., 2015); 9(Välitalo et al., 2017); 10(Nivala et al., 2018); 11(Cargouet et al., 2004); 12(Jugan et al., 2009); 13(Jalova et al.,
2013); 14(Hamilton et al., 2016); 15(Leusch et al., 2014a); 16(Macova et al., 2011); 17(Aerni et al., 2004); 18(Escher et al., 2008b); 19(Escher et al., 2014);
20
(Fang et al., 2012); 21(French et al., 2015); 22(Gehrmann et al., 2018); 23(Huggett et al., 2003); 24(Pawlowski et al., 2003); 25(Zeng et al., 2016);
26
(Alygizakis et al., 2019); 27(Jia et al., 2015); 28(Könemann et al., 2018); 29(Leusch et al., 2014b); 30(van der Linden et al., 2008); 31(Houtman et al.,
Bioanalytical Tools in Water Quality Assessment

2020); 32(González et al., 2020); 33(Leusch et al., 2018b); 34(Mehinto et al., 2015); 35(Mehinto et al., 2016); 36(Mueller et al., 2021); 37(Henneberg et al.,
2014); 38(Miege et al., 2009); 39(Neale et al., 2017c); 40(Furuichi et al., 2004); 41(Jarošová et al., 2014b); 42(Medlock Kakaley et al., 2020); 43(Bicchi
et al., 2009); 44(Körner et al., 2001); 45(Macova et al., 2010); 46(Reungoat et al., 2010); 47(Brion et al., 2019); 48(Brettschneider et al., 2019); 49(Chen
et al., 2016); 50(Huang et al., 2016); 51(Lv et al., 2016); 52(Vermeirssen et al., 2005); 53(Xiao et al., 2016); 54(Xiao et al., 2017); 55(Zhao et al., 2011);
56
(Brand et al., 2013); 57(Jia et al., 2019); 58(Scott et al., 2014); 59(Shi et al., 2018); 60(Daniels et al., 2018); 61(Hashmi et al., 2018); 62(König et al.,
2017); 63(Mehinto et al., 2017); 64(Müller et al., 2018); 65(Neale et al., 2018a); 66(Neale et al., 2020b); 67(Neale et al., 2020a); 68(Scott et al., 2018);
69
(Prochazkova et al., 2018); 70(Mnif et al., 2012); 71(Neale et al., 2015b); 72(Serra et al., 2020); 73(Toušová et al., 2017); 74(Shue et al., 2009);
75
(Conley et al., 2017a); 76(Conley et al., 2017b); 77(Liu et al., 2018); 78(Oh et al., 2006); 79(Van Zijl et al., 2017); 80(Albergamo et al., 2020); 81(Valcarcel
et al., 2018).
 Current bioanalytical tools for water quality assessment 185

Table 10.9 Common assays applied to evaluate anti-estrogenic activity in water extracts.
Assay Cell Line// Test Detection Tamoxifen Tamoxifen EC
System Method ECSPR20a ECSPR20 Reference
(M) (µg// L)
Yeast reporter gene
YAES Yeast Absorbance 6.00 × 10−7 223 Conroy et al.
(2007)
Mammalian reporter gene
ERα U2OS Luminescence 1.50 × 10−9 0.56 Jia et al.
CALUX (2015)
ERα HEK293T Fluorescence 5.86 × 10−6 2177 Neale et al.
GeneBLAzer (2020b)
a
SPR = suppression ratio.

Androgenic activity in wastewater, surface water and drinking water extracts is

summarised in Table 10.12. Between 30 and 350 ng/L DHT EQ was detected in
wastewater influent, with low or no androgenic activity typically present in
wastewater effluent. The reported WWTP removal efficacy ranged from 95% to
.99.9% (Jalova et al., 2013; Bain et al., 2014; Houtman et al., 2018), explaining

Table 10.10 Summary of anti-estrogenic activity reported as tamoxifen EQ (TMX EQ)

in different environmental water extracts.
Matrix Assay Anti-estrogenic Activity Reference
TMX EQ (µgtamoxifen/ L)
Wastewater influent ERα GeneBLAzer Cytotoxic 1
Wastewater effluent YAES 0.7–97 2–4
ERα CALUX ,0.07a–110 5–7
ERα GeneBLAzer ,2177 , cytotoxic a
1, 8
Recycled water ERα CALUX ,0.04a–4.4 5, 7, 9
Surface water YAES ,50 10
ERα CALUX ,0.04 –6a
5, 11–13
ERα GeneBLAzer ,1–2.7, cytotoxic 8, 14, 15
Drinking water ERα CALUX ,0.04a 5
ERα GeneBLAzer ,22a to ,109a 8, 14
NB: Only studies that have applied SPE or LLE are included.
a
ECSPR20 values converted to TMX EQ using ECSPR20 values provided in Table 10.9.
References: 1(Nivala et al., 2018); 2(Archer et al., 2020); 3(Conroy et al., 2007); 4(Fang et al.,
2012); 5(Escher et al., 2014); 6(Gehrmann et al., 2018); 7(Jia et al., 2015); 8(Leusch et al., 2018b);
9
(Leusch et al., 2014b); 10(Zhao et al., 2011); 11(Daniels et al., 2018); 12(Jia et al., 2019); 13(Scott
et al., 2014); 14(Neale et al., 2020b); 15(Scott et al., 2018).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 186 Bioanalytical Tools in Water Quality Assessment

Table 10.11 Common assays applied to evaluate androgenic activity in

water extracts.
Assay Cell Detection DHT DHT EC
Line// Test Method EC10 (M) EC10 Reference
System (ng// L)
Yeast reporter gene
YAS Yeast Absorbance 2.86 × 10−10 83 Sohoni and
Sumpter
(1998)
Mammalian reporter gene
AR CALUX U2OS Luminescence 1.00 × 10−10 29 Jia et al.
(2015)
AR HEK293MSR Fluorescence 1.40 × 10−10 41 Leusch
GeneBLAzer et al. (2017)
MDA-kB2 MDA-MB-453 Luminescence 3.12 × 10−11 9.1 Neale et al.
(2017c)
DHT = dihydrotestosterone.

the low activity in treated effluent. Based on the mammalian reporter gene assays,
only low androgenic activity was detected in surface water, with 0.25–12 ng/L
DHT EQ reported (Table 10.12). Based on the EC10 values of DHT in
Table 10.11, this means samples would need to be enriched between 1.8 and 140
times in the assay to detect an effect in surface water. No androgenic activity was
detected in recycled water, with only one study detecting androgenic activity in
drinking water at 0.13 ng/L DHT EQ (Brand et al., 2013). Based on the DHT
EC10 for AR CALUX in Table 10.11, this equates to an EC10 of relative
extraction factor (REF) 223, meaning the sample would need to be enriched over
200 times to elicit 10% of activation of AR.

10.4.2.2 Androgen receptor antagonism

The ECSPR20 values of the reference compound flutamide in Table 10.13 are similar
in the yeast and mammalian reporter gene anti-androgenic assays. Reported
anti-androgenic activity in environmental extracts is provided in Table 10.14.
Anti-androgenic activity was highest in wastewater effluent, with flutamide EQ
between 0.5 and 360 µgflutamide/L reported for the mammalian reporter gene
assays (Table 10.14). Based on the flutamide ECSPR20 values in Table 10.13, this
indicates between 0.8 and 600 times enrichment in the assay would be required to
detect an effect. Anti-androgenic activity in surface water expressed as flutamide
EQ ranged from 0.3 to 257 µgflutamide/L, while no anti-androgenic activity was
detected in drinking water or recycled water.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 10.12 Summary of androgenic activity reported as dihydrotestosterone (DHT) EQ in different environmental water extracts.
Matrix Assay Androgenic Activity Reference
DHT EQ (ngDHT/ L)
Wastewater influent AR CALUX ,0.62–350 1–6
AR GeneBLAzer 50–83a 7
Wastewater effluent YAS ,2.8a–138 8, 9
AR CALUX ,0.02–2.7 1–6, 8, 10–14
AR GeneBLAzer ,2, cytotoxic 7, 8, 15
MDA-kB2 ,0.77–12 8, 16, 17
Recycled water YAS ,2.8a 8
AR CALUX ,0.97a to ,30 3, 8, 12, 13
AR GeneBLAzer ,2.0a 8
MDA-kb2 ,0.34a 8
a
Surface water YAS ,2.8 –69 8, 9, 18, 19
AR CALUX 3–5, 8, 11, 14, 20, 21

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

,0.08–12
AR GeneBLAzer 0.29–4.9a 8, 15, 22–27
MDA-kb2 ,0.02–4.8 8, 16, 17, 23, 28, 29
a
Drinking water YAS ,2.8 8
AR CALUX ,0.08–0.13 8, 20
AR GeneBLAzer ,0.1 to ,2a 8, 15, 26, 30
MDA-kb2 ,0.34a to ,0.77 8, 16
NB: Only studies that have applied SPE or LLE are included.
a
EC10 converted to DHT EQ using EC10 values provided in Table 10.11.
Current bioanalytical tools for water quality assessment

References: 1(Bain et al., 2014); 2(Houtman et al., 2018); 3(Leusch et al., 2014a); 4(Roberts et al., 2015); 5(Šauer et al., 2018); 6(Välitalo et al., 2017);
7
(Nivala et al., 2018); 8(Escher et al., 2014); 9(French et al., 2015); 10(Gehrmann et al., 2018); 11(Houtman et al., 2020); 12(Jia et al., 2015); 13(Leusch
et al., 2014b); 14(van der Linden et al., 2008); 15(Leusch et al., 2018b); 16(Medlock Kakaley et al., 2020); 17(Neale et al., 2017c); 18(Huang et al., 2016);
19
(Zhao et al., 2011); 20(Brand et al., 2013); 21(Scott et al., 2014); 22(Hashmi et al., 2018); 23(König et al., 2017); 24(Müller et al., 2018); 25(Neale et al.,
187

2018a); 26(Neale et al., 2020b); 27(Scott et al., 2018); 28(Conley et al., 2017a); 29(Toušová et al., 2017); 30(Albergamo et al., 2020).
 188 Bioanalytical Tools in Water Quality Assessment

Table 10.13 Common assays applied to evaluate anti-androgenic activity in

water extracts.
Assay Cell Detection Flutamide Flutamide EC
Line// Test Method ECSPR20 ECSPR20 Reference
System (M) (µg// L)
Yeast reporter gene
YAAS Yeast Absorbance 7.50 × 10−7a 207a Stalter et al.
(2011)
Mammalian reporter gene
AR CALUX U2OS Luminescence 1.10 × 10−6 304 Jia et al.
(2015)
AR HEK293MSR Fluorescence 5.50 × 10−7a 152a Leusch
GeneBLAzer et al. (2017)
MDA-kB2 MDA- Luminescence 2.07 × 10−7 57 Neale et al.
MB-453 (2017b)

10.4.3 Glucocorticoid receptor

10.4.3.1 Glucocorticoid receptor agonism
The glucocorticoid receptor (GR) is a corticosteroid receptor that controls the
actions of glucocorticoids, and a wide range of environmental contaminants can
interfere with glucocorticoid activity (Zhang et al., 2019). Mammalian reporter
gene assays have been applied to evaluate glucocorticoid activity in
environmental water extracts (Table 10.15), with GR CALUX and GR
GeneBLAzer most commonly applied. The pharmaceutical dexamethasone serves
as the assay reference compound, with the lowest EC10 reported for GR
GeneBLAzer. The mammalian reporter gene CV-1 GR assay has also been
recently applied to detect glucocorticoid activity in different water samples
(Conley et al., 2017a; Medlock Kakaley et al., 2020), but no EC value for
dexamethasone was available, so this assay was not included in Table 10.15.
Glucocorticoid activity in wastewater influent expressed as dexamethasone
EQ ranged from 37 to 121 ngdexamethasone/L, with between 1.8 and 628
ngdexamethasone/L reported in wastewater effluent (Table 10.16). The studies that
evaluated WWTP treatment efficacy found between −7% and 66% removal
(Bain et al., 2014; Roberts et al., 2015; Houtman et al., 2018), indicating much
poorer removal for glucocorticoid activity compared to estrogenic activity and
androgenic activity. As a result, glucocorticoid activity was frequently detected in
surface water, with dexamethasone EQ between 9 and 170 ngdexamethasone/L in an
effluent impacted river (Daniels et al., 2018). Based on the GR GeneBLAzer
dexamethasone EC10 value in Table 10.15, this equates to 0.5–9 times

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 10.14 Summary of anti-androgenic activity reported as flutamide EQ in different environmental water extracts.
Matrix Assay Anti-androgenic Activity Reference
Flutamide EQ (µgflutamide/ L)
Wastewater influent AR CALUX 5.2–26 1
AR GeneBLAzer Cytotoxic 2
Wastewater effluent YAAS 16–3190 3–5
AR CALUX 0.48–360 1, 4, 6–10
AR GeneBLAzer ,15, cytotoxic 2, 11
a
MDA-kb2 ,4.4 7
a
Recycled water AR CALUX ,20 to , 300 7, 10
MDA-kb2 ,1.9a 7
Surface water YAAS 20–935 12

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

AR CALUX 3.3–257 1, 7, 9, 13, 14
AR GeneBLAzer ,0.51–90, cytotoxic 11, 15–18
MDA-kb2 ,0.12–9.2a 7, 15, 19
Drinking water AR CALUX ,20a 7
AR GeneBLAzer ,0.7 to ,1.5a 11, 17
a
MDA-kb2 ,1.9 7
NB: Only studies that have applied SPE or LLE are included.
Current bioanalytical tools for water quality assessment

a
ECSPR20 values converted to Flutamide EQ using ECSPR20 values provided in Table 10.13.
References: 1(Šauer et al., 2018); 2(Nivala et al., 2018); 3(Fang et al., 2012); 4(Gehrmann et al., 2018); 5(Stalter et al., 2011); 6(Alygizakis et al., 2019);
7
(Escher et al., 2014); 8(Houtman et al., 2018); 9(Houtman et al., 2020); 10(Jia et al., 2015); 11(Leusch et al., 2018b); 12(Zhao et al., 2011); 13(Jia et al.,
2019); 14(Scott et al., 2014); 15(König et al., 2017); 16(Müller et al., 2018); 17(Neale et al., 2020b); 18(Scott et al., 2018); 19(Toušová et al., 2017).
189
 190 Bioanalytical Tools in Water Quality Assessment

Table 10.15 Common assays applied to evaluate glucocorticoid activity in water extracts.
Assay Cell Detection Dexamethasone Dexamethasone EC
Line Method EC10 (M) EC10 (ng// L) Reference
GR CALUX U2OS Luminescence 8.00 × 10−10 314 Jia et al.
(2015)
GR HEK Fluorescence 2.08 × 10−10 82 Nivala et al.
GeneBLAzer 293T (2018)
GR HT1080 Luminescence 5.00 × 10−10 196 Jia et al.
Switchgear (2015)

enrichment in the assay. Glucocorticoid activity has not been reported in drinking
water (Table 10.16).

10.4.3.2 Glucocorticoid receptor antagonism

Only two assays, the GR CALUX and GR GeneBLAzer, have been applied in the
literature to evaluate anti-glucocorticoid activity, with the GR GeneBLAzer much
more sensitive compared to GR CALUX based on the reference compound
mifepristone ECSPR20 values (Table 10.17). Anti-glucocorticoid activity was only
detected in surface water in two studies (König et al., 2017; Jia et al., 2019), with
mifepristone EQ reported between 2.5 and 610 ngmifepristone/L. Anti-glucocorticoid
activity was not detected in wastewater effluent, recycled water or drinking water,
with cytotoxicity masking the effect in wastewater influent (Table 10.18).

10.4.4 Progesterone receptor

10.4.4.1 Progesterone receptor agonism
Two assays, the progesterone receptor (PR) CALUX and PR GeneBLAzer, have been
applied to evaluate progestogenic activity in environmental water extracts, with both
assays having similar EC10 values for the synthetic hormone levonorgestrel
(Table 10.19). Progestogenic activity in water extracts has been reported in
different equivalent concentrations, including progesterone EQ, levonorgestrel EQ,
promegestone EQ and Org2058 EQ. To assist with comparison, the results from
the literature were converted to levonorgestrel EQ based on published potency
data. Up to 3.2 ng/L levonorgestrel EQ was detected in wastewater influent, while
between 0.43 and 7.1 ng/L levonorgestrel EQ was detected in treated effluent
(Table 10.20). A number of studies have found increased progestogenic activity
after wastewater treatment (Roberts et al., 2015; Houtman et al., 2018), while Bain
et al. (2014) found between 12% and .93% removal efficacy in three WWTPs in
Australia. Up to 9.6 ng/L levonorgestrel EQ was detected in surface water from
the Netherlands, although progestogenic activity was often below the assay
detection limit or masked by cytotoxicity in surface water. No progestogenic
activity has been detected in recycled water or drinking water (Table 10.20).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 10.16 Summary of glucocorticoid activity reported as dexamethasone EQ in units of ngdexamethasone/L in different
environmental water extracts.
Matrix Assay Glucocorticoid Activity Reference
Dexamethasone EQ (ngdexamethasone/ L)
Wastewater influent GR CALUX 37–121 1–3
GR GeneBLAzer ,400, cytotoxic 4, 5
a
Wastewater effluent GR CALUX 11–628 1–3, 6–12
a
GR GeneBLAzer ,8.2 –392 5, 7, 8, 10, 13–16
GR Switchgear 19–24a 7, 8
CV-1 GR 1.8–21 17
Recycled water GR CALUX ,10a to ,310 7–9
a
GR GeneBLAzer ,4.1 –65, ,230 7, 8, 15
GR Switchgear ,9.8a–16 7, 8
Surface water GR CALUX 0.30–34, ,500 3, 7, 10–12, 18–20

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

GR GeneBLAzer ,1.1–170, cytotoxic 7, 10, 16, 21–27
a
GR Switchgear ,9.8 7
CV-1 GR ,1.2–43 17, 28
Drinking water GR CALUX ,2.0 to ,120 7, 10, 18
GR GeneBLAzer ,0.82 to ,5.8 7, 10, 27, 29
GR Switchgear ,9.8a 7
CV-1 GR ,1.2 17
Current bioanalytical tools for water quality assessment

NB: Only studies that have applied SPE or LLE are included.
a
EC10 converted to dexamethasone EQ using EC10 values provided in Table 10.15.
References: 1(Bain et al., 2014); 2(Houtman et al., 2018); 3(Roberts et al., 2015); 4(Lee et al., 2015); 5(Nivala et al., 2018); 6(Alygizakis et al., 2019);
7
(Escher et al., 2014); 8(Jia et al., 2015); 9(Leusch et al., 2014b); 10(Leusch et al., 2018b); 11(van der Linden et al., 2008); 12(Houtman et al., 2020);
13
(Chen et al., 2016); 14(Jia et al., 2016); 15(Mehinto et al., 2015); 16(Mehinto et al., 2016); 17(Medlock Kakaley et al., 2020); 18(Brand et al., 2013);
191

19
(Schriks et al., 2013); 20(Toušová et al., 2017); 21(Daniels et al., 2018); 22(Hashmi et al., 2020); 23(König et al., 2017); 24(Mehinto et al., 2017);
25
(Müller et al., 2018); 26(Neale et al., 2018a); 27(Neale et al., 2020b); 28(Conley et al., 2017a); 29(Albergamo et al., 2020).
 192 Bioanalytical Tools in Water Quality Assessment

Table 10.17 Common assays applied to evaluate anti-glucocorticoid activity in

water extracts.
Assay Cell Detection Mifepristone Mifepristone EC
Line Method ECSPR20 (M) ECSPR20 Reference
(ng// L)
GR CALUX U2OS Luminescence 2.90 × 10−9 1246 Jia et al.
(2015)
GR HEK Fluorescence 1.00 × 10−10 43 Jia et al.
GeneBLAzer 293T (2015)

10.4.4.2 Progesterone receptor antagonism

Two assays, the PR CALUX and PR GeneBLAzer, have been applied to evaluate
anti-progestogenic activity in environmental extracts. Based on the reference
compound mifepristone ECSPR20 value, the PR CALUX was more sensitive
compared to PR GeneBLAzer (Table 10.21). Most water extracts either had no

Table 10.18 Summary of anti-glucocorticoid activity reported as mifepristone EQ in

different environmental water extracts.
Matrix Assay Anti-glucocorticoid Reference
Activity
Mifepristone EQ
(ngmifepristone/ L)
Wastewater GR Cytotoxic 1
influent GeneBLAzer
Wastewater GR CALUX ,623a to ,1200 2, 3
effluent GR ,21 to ,60, cytotoxic
a
1–4
GeneBLAzer
Recycled water GR CALUX ,1200 to ,1246a 2, 3
GR ,40 to ,43 a
2, 3
GeneBLAzer
Surface water GR CALUX ,50–610, ,1246a 2, 5
GR ,0.2–2.5, ,43 , cytotoxic
a
2, 4, 6–8
GeneBLAzer
Drinking water GR CALUX ,1246a 2
GR ,0.49 to ,43 a
2, 4, 8
GeneBLAzer
NB: Only studies that have applied SPE or LLE are included.
a
ECSPR20 values converted to mifepristone EQ using ECSPR20 values provided in Table 10.17.
References: 1(Nivala et al., 2018); 2(Escher et al., 2014); 3(Jia et al., 2015); 4(Leusch et al.,
2018b); 5(Jia et al., 2019); 6(Daniels et al., 2018); 7(König et al., 2017); 8(Neale et al., 2020b).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 193

Table 10.19 Common assays applied to evaluate progestogenic activity in water extracts.
Assay Cell Detection Levonorgestrel Levonorgestrel EC
Line Method EC10 (M) EC10 (ng// L) Reference
PR CALUX U2OS Luminescence 3.44 × 10−11a 10.8 Scott et al.
(2014)
PR HEK Fluorescence 1.22 × 10−11a 3.8 Leusch
GeneBLAzer 293T et al.
(2018b)
a
Presented EC10 value converted from EC50 value assuming a slope of the log-logistic
concentration response curve of 1.

Table 10.20 Summary of progestogenic activity reported as levonorgestrel EQ in

different environmental water extracts.
Matrix Assay Progestogenic Activity Reference
Levonorgestrel EQ
(nglevonorgestrel/ L)
Wastewater PR CALUX ,0.32–3.2b 1–3
influent PR Cytotoxic 4
GeneBLAzer
Wastewater PR CALUX ,0.01–7.1b, ,90 1–3, 5–9
effluent PR ,0.38b–5.7, cytotoxicity 4, 5, 8, 10
GeneBLAzer
Recycled water PR CALUX ,0.01 to ,90 5–7
PR ,0.19b to ,1.4 5, 10
GeneBLAzer
Surface water PR CALUX ,0.11c–9.6c 3, 5, 8, 9,
11, 12
PR ,0.04a–1.1, cytotoxic 5, 8, 13–18
GeneBLAzer
Drinking water PR CALUX ,0.11c to ,2.0 5, 8, 11
PR ,0.04a to ,0.19a 5, 8, 17, 19
GeneBLAzer
NB: Only studies that have applied SPE or LLE are included.
a
EC10 values converted to levonorgestrel EQ using EC10 values provided in Table 10.19.
b
Results presented as progesterone EQ and converted to levonorgestrel EQ using an REP of
0.16 (Sonneveld et al., 2011).
c
Results presented as Org2058 EQ and converted to levonorgestrel EQ using an REP of 2.14
(Sonneveld et al., 2011).
References: 1(Bain et al., 2014); 2(Houtman et al., 2018); 3(Roberts et al., 2015); 4(Nivala et al.,
2018); 5(Escher et al., 2014); 6(Jia et al., 2015); 7(Leusch et al., 2014b); 8(Leusch et al., 2018b);
9
(van der Linden et al., 2008); 10(Mehinto et al., 2015); 11(Brand et al., 2013); 12(Scott et al.,
2014); 13(Hashmi et al., 2020); 14(König et al., 2017); 15(Müller et al., 2018); 16(Neale et al.,
2018a); 17(Neale et al., 2020b); 18(Scott et al., 2018); 19(Albergamo et al., 2020).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 194 Bioanalytical Tools in Water Quality Assessment

Table 10.21 Common assays applied to evaluate anti-progestogenic activity in

water extracts.
Assay Cell Detection Mifepristone Mifepristone EC
Line Method ECSPR20 ECSPR20 Reference
(M) (ng// L)
PR CALUX U2OS Luminescence 2.00 × 10−11 8.6 Jia et al.
(2015)
PR HEK Fluorescence 3.00 × 10−10 129 Nivala et al.
GeneBLAzer 293T (2018)

Table 10.22 Summary of anti-progestogenic activity reported as mifepristone EQ in

different environmental water extracts.
Matrix Assay Anti-progestogenic Reference
Activity Mifepristone
EQ (ngmifepristone/ L)
Wastewater influent PR GeneBLAzer Cytotoxic 1
Wastewater effluent PR CALUX ,0.72–17 2–4
PR GeneBLAzer ,2, cytotoxic 1, 5
Recycled water PR CALUX ,8.6 a
3, 4
Surface water PR CALUX ,8–32,000 3, 6
PR GeneBLAzer ,0.5–4.2 5–10
Drinking water PR CALUX ,8.6 a
3
PR GeneBLAzer ,0.1 to ,1.3 a
5, 9
NB: Only studies that have applied SPE or LLE are included.
a
ECSPR20 values converted to mifepristone EQ using ECSPR20 values provided in Table 10.21.
References: 1(Nivala et al., 2018); 2(Alygizakis et al., 2019); 3(Escher et al., 2014); 4(Jia et al.,
2015); 5(Leusch et al., 2018b); 6(Scott et al., 2014); 7(König et al., 2017); 8(Müller et al., 2018);
9
(Neale et al., 2020b); 10(Scott et al., 2018).

response or were masked by cytotoxicity in PR CALUX or PR GeneBLAzer when

run in antagonist mode (Table 10.22). Only one study reported anti-progestogenic
activity in wastewater effluent (Alygizakis et al., 2019), with 9 of 12 samples
active, while two studies detected anti-progestogenic activity in Australian
surface waters (Scott et al., 2014; Scott et al., 2018).

10.4.5 Thyroid receptor

10.4.5.1 Thyroid receptor agonism
A number of assays have been applied to evaluate thyroid receptor (TR) activity in
environmental water extracts including yeast reporter gene assays, mammalian

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 195

reporter gene assays, cell proliferation assays and a whole organism assay (Xenopus
eleutheroembryonic thyroid assay, XETA) using embryonic Xenopus laevis
(Table 10.23). Based on reference compound triiodothyronine (T3), the reporter
gene assays were the most sensitive. However, effects in mammalian reporter
gene assays have only been observed in wastewater influent, with T3 EQ of 25
ngT3/L reported in French wastewater using the PC-DR-LUC assay
(Table 10.24). All other reporter gene assays did not detect thyroid activity in
wastewater effluent, surface water or recycled water. In contrast, T3 EQ between
1100 and 1340 ngT3/L were detected in wastewater effluent using the XETA,
with 960 ngT3/L detected in surface water (Välitalo et al., 2017; Leusch et al.,
2018a). This suggests that the XETA, which incorporates toxicokinetic processes
and other non-TR-mediated thyroid effects, may be more suitable to evaluate
thyroid activity in water extracts than mammalian reporter gene assays.

10.4.5.2 Thyroid receptor antagonism

Three assays have been applied to evaluate anti-thyroid activity in environmental
water extracts, the yeast reporter gene yeast two-hybrid assay and the mammalian
reporter gene assays GH3.TRE-Luc and TRβ GeneBLAzer. The anti-thyroid
activity assay reference compound is the pharmaceutical amiodarone
hydrochloride (AH), with the EC50 values lower for the mammalian reporter gene
assays (Table 10.25). Anti-thyroid activity expressed as amiodarone
hydrochloride equivalent concentrations (AH EQ) was detected in wastewater
influent (60–422 µgAH/L), wastewater effluent (13–35 µgAH/L) and surface water
(3.3–16 µgAH/L) using the yeast two-hybrid assay (Table 10.26). In contrast,
only wastewater effluent had a response in TRβ GeneBLAzer, with none of the
samples having a response in GH3.TRE-Luc in antagonist mode.

10.4.6 Mineralocorticoid receptor

Similar to the GR, the mineralcorticoid receptor (MR) is a corticosteroid receptor
that controls the action of mineralocorticoids (Zhang et al., 2019). Currently, there
is only one assay used to assess mineralocorticoid activity in water extracts, the
HG5LN-hMR, which can be run in both agonist and antagonist modes. The
reference compound in the agonist mode is the hormone aldosterone with an
EC50 of 9.80 × 10−10 M (Leusch et al., 2018b), while the pharmaceutical
spironolactone is the antagonist reference compound with an EC50 of 2.89 ×
10−9 M (Bellet et al., 2012). To date, mineralocorticoid activity has not been
detected in wastewater, surface water or drinking water (Bellet et al., 2012;
Creusot et al., 2014; Leusch et al., 2018b). However, anti-mineralocorticoid
activity has been detected in wastewater influent, wastewater effluent and
surface water (Table 10.27), with spironolactone EQ between up to 0.9 and 2.3
µgspironolactone/L reported. Anti-mineralocorticoid activity was below the limit of
detection in drinking water.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
196
Table 10.23 Common assays applied to evaluate the thyroid receptor (TR) activity in water extracts.
Assay Cell Line// Test Detection Method Triiodothyronine Triiodothyronine EC Reference
System (T3) (T3)
EC10 (M) EC10 (ng// L)
Yeast reporter gene
Yeast Yeast Absorbance 2.60 × 10−8 17,000 Li et al. (2008)
two-hybrid
Mammalian reporter gene
TRβ CALUX U2OS Luminescence 8.60 × 10−12 5.6 Jia et al. (2015)
TRβ HEK293T Fluorescence 6.00 × 10−11 41 Leusch et al.
GeneBLAzer (2017)
GH3.TRE-Luc GH3 Luminescence 6.67 × 10−12a 4.3a Leusch et al.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

(2018a)
PC-DR-LUC PC12 Luminescence 2.00 × 10−11a 13a Jugan et al.
(2009)
Cell proliferation
T-Screen GH3 Fluorescence (cell 2.80 × 10−10 182 Jia et al. (2015)
viability measured using
Alamar Blue (resazurin))
Bioanalytical Tools in Water Quality Assessment

Whole organism
XETA Embryonic Fluorescence EC50 4.50 × 10−9 EC50 3000 Leusch et al.
Xenopus (2018a)
a
EC10 value converted from EC50 value assuming a slope of the log-logistic concentration response curve of 1.
 Current bioanalytical tools for water quality assessment 197

Table 10.24 Summary of thyroid receptor (TR) activity reported as triiodothyronine

(T3) EQ in different environmental water extracts.
Matrix Assay Thyroid Activity Reference
T3 EQ (ngT3/ L)
Wastewater influent PC-DR-LUC ,20–25 1
T-Screen 190–204 2
Wastewater effluent TRβ CALUX ,0.14 to ,21 a b
3–5
TRβ GeneBLAzer ,2.3 b
5
GH3.TRE-Luc ,10b to ,25 5, 6
PC-DR-LUC ,20 1
T-Screen ,5.5 to ,6.1a 3, 4
XETA ,LOD to 1340 5, 7
Recycled water TRβ CALUX ,0.14 to ,5.6 a
3, 4
T-Screen ,5.5 to ,6.1 a
3, 4
Surface water Yeast two-hybrid TR ,14–43 8
TRβ CALUX ,0.14 to ,3.5 a b
3, 5
TRβ GeneBLAzer ,1.2b 5
GH3.TRE-Luc ,5.3 to ,20
b
5, 6
PC-DR-LUC ,20 1
T-Screen ,6.1 a
3
b
XETA 960 5
Drinking water TRβ CALUX ,0.14 to ,3.5 a b
3, 5
TRβ GeneBLAzer ,1.2b 5
GH3.TRE-Luc ,0.05 to ,1.3 b
5, 6
PC-DR-LUC ,20 1
T-Screen ,6.1a 3
XETA ,300b 5
NB: Only studies that have applied SPE or LLE are included.
a
EC10 values converted to triiodothyronine (T3) EQ using EC10 values provided in Table 10.23.
b
Results presented as thyroxine (T4) EQ and converted to T3 EQ using the REP from Leusch
et al. (2018a). LOD: limit of detection.
References: 1(Jugan et al., 2009); 2(Kusk et al., 2011); 3(Escher et al., 2014); 4(Jia et al., 2015);
5
(Leusch et al., 2018a); 6(Leusch et al., 2018b); 7(Välitalo et al., 2017); 8(Chinathamby et al., 2013).

10.4.7 Retinoic acid receptor and retinoid X receptor

Only a handful of studies have applied assays indicative of the retinoic acid receptor
(RAR), where the effect is expressed as all-trans retinoic acid EQ (ATRA EQ), and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 198 Bioanalytical Tools in Water Quality Assessment

Table 10.25 Common assays applied to evaluate anti-thyroid activity in

water extracts.
Assay Cell Detection Amiodarone Amiodarone EC
Line// Test Method Hydrochloride Hydrochloride Reference
System EC50 (M) EC50 (µg// L)
Yeast reporter gene
Yeast Yeast Absorbance 3.10 × 10−5 21,000 Li et al.
two-hybrid (2008)
Mammalian reporter gene
TRβ HEK293T Fluorescence 7.30 × 10−6 5000 Leusch
GeneBLAzer et al.
(2018a)
GH3. GH3 Luminescence 8.40 × 10−6 5700 Leusch
TRE-Luc et al.
(2018a)

Table 10.26 Summary of anti-thyroid activity reported as amiodarone hydrochloride

EQ (AH EQ) in different environmental water extracts.
Matrix Assay Anti-thyroid Activity Reference
AH EQ (µgAH/ L)
Wastewater influent Yeast two-hybrid 60–422 1
Wastewater effluent Yeast two-hybrid 13–35 1
TRβ GeneBLAzer 350 2
GH3.TRE-Luc ,1700 2
Surface water Yeast two-hybrid 3.3–16 1
TRβ GeneBLAzer ,28 2
GH3.TRE-Luc ,870 2
Drinking water TRβ GeneBLAzer ,28 2
GH3.TRE-Luc ,87 2
NB: Only studies that have applied SPE or LLE are included.
References: 1(Li et al., 2011); 2(Leusch et al., 2018a).

the retinoid X receptor (RXR with 9-cis-retinoic acid EQ. For example, wastewater
effluent, surface water and drinking water had no response in the RXR CALUX and
HELN-RARa-RXR assays (Leusch et al., 2018b). In contrast, some surface water
extracts from Serbia had a response in the RAR GeneBLAzer with ATRA EQ
,0.02–0.15 ngATRA/L and in RXR GeneBLAzer with 9-cis-retinoic acid EQ of
7 ng9-cis-retinoic acid/L reported (König et al., 2017). This equates to an effect after
41–170 times enrichment in the assay for RAR GeneBLAzer and 240 times

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 199

Table 10.27 Summary of reported anti-mineralocorticoid activity reported as

spironolactone EQ in different environmental water extracts.
Matrix Assay Anti-mineralocorticoid Reference
Activity
Spironolactone EQ
(µgspironolactone/ L)
Wastewater HG5LN-hMR 1.3–2.3 1
influent
Wastewater HG5LN-hMR ,3.1–3.1 2
effluent
Surface water HG5LN-hMR ,0.66–0.91 2
Drinking water HG5LN-hMR ,0.16 2
NB: Only studies that have applied SPE or LLE are included.
References: 1(Bellet et al., 2012); 2(Leusch et al., 2018b).

enrichment in the assay for RAR GeneBLAzer. RAR activity was also detected in
surface water using the yeast two-hybrid RAR assay (ATRA EQ ,0.4–8
ngATRA/L) (Chinathamby et al., 2013) and an in vitro reporter gene bioassay
using P19/A15 cells (ATRA EQ ,10–29 ngATRA/L) (Javurek et al., 2015).
Escher et al. (2014) also applied the P19/A15 assay to wastewater, recycled
water, surface water and drinking water extracts, with only one wastewater
effluent sample inducing 10% effect after 25 times enrichment in the assay.

10.5 BIOASSAYS INDICATIVE OF OTHER

RECEPTOR-MEDIATED EFFECTS
In addition to hormone receptor-mediated effects, other relevant specific modes of
action include phytotoxicity and neurotoxicity.

10.5.1 Phytotoxicity
Although not directly relevant for human health, several studies have applied algal
assays to assess photosystem II (PSII) inhibition in a range of water matrices (Tang
and Escher, 2014; Hamers et al., 2018). Most studies have assessed PSII inhibition
using the combined algae test (CAT) with PSII inhibition measured after 2 h using
imaging pulse-amplitude modulated (PAM) fluorometry using green microalgae
Raphidocelis subcapitata (formerly named Selenastrum capricornutum and
Pseudokirchneriella subcapitata) (Escher et al., 2008a; Glauch and Escher 2020).
The assay reference compound is the herbicide diuron, with reported EC50 values
ranging from 1.40 to 4.3 µgdiuron/L (Jia et al., 2015; Allan et al., 2017). In
contrast, only two studies used Max-I-PAM with Chlorella vulgaris, with EC50
values of 16 µgdiuron/L reported (Macova et al., 2010; Leusch et al., 2014a).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 200 Bioanalytical Tools in Water Quality Assessment

Table 10.28 Summary of studies that have applied photosystem II (PSII) inhibition
assays to different environmental water extracts, where PSII inhibition is expressed
as diuron EQ.
Matrix Assay PSII Inhibition Reference
Diuron EQ (µgdiuron/ L)
Wastewater influent CAT (2 h IPAM) 0.07–2.2 1, 2, 3
a
Max-I-PAM 0.04–0.23 3
Wastewater effluent CAT (2 h IPAM) 0.03–1.3 1, 2, 5, 6, 7, 8, 9
Max-I-PAM a
,0.03–0.12 4, 10
Recycled water CAT (2 h IPAM) ,0.004–1.3 1, 2, 3, 8, 9
a
Max-I-PAM 0.02–,0.03 3, 10
Surface water CAT (2 h IPAM) 0.01–1.3 1, 2, 3, 6, 7, 11
Max-I-PAMa ,0.03–0.06 4
Drinking water CAT (2 h IPAM) 0.02–0.05 1, 2
NB: Only studies that have applied SPE or LLE are included. CAT =combined algae test.
a
The Max-I-PAM assay uses Chlorella vulgaris, while all other assays used Raphidocelis
subcapitata (formerly named Selenastrum capricornutum and Pseudokirchneriella
subcapitata).
References: 1(Macova et al., 2011); 2(Tang and Escher, 2014); 3(Glauch and Escher 2020);
4
(Leusch et al., 2014a); 5(Escher et al., 2008b); 6(Jia et al., 2015); 7(Neale et al., 2017c);
8
(Reungoat et al., 2010); 9(Tang et al., 2014); 10(Macova et al., 2010); 11(Allan et al., 2017).

Consequently, it appears that R. subcapitata is more sensitive. PSII inhibition is

commonly expressed in diuron EQ (DEQ), with 0.04–2.2 µgdiuron/L DEQ
detected in wastewater influent, 0.03–1.3 µgdiuron/L DEQ detected in wastewater
effluent and 0.01–1.3 µgdiuron/L DEQ detected in surface water (Table 10.28).
Based on the EC10 value, this translates to an effect being detected after 0.3–99
times enrichment. Low activity was reported in drinking water (0.02–0.05
µgdiuron/L DEQ) and reverse osmosis-treated recycled water (,0.004–0.05
µgdiuron/L DEQ) (Table 10.28). The observed effects in water extracts were
primarily explained by PSII-inhibiting herbicides, with other chemicals only
having a minor contribution to the observed effect (Tang and Escher, 2014; Neale
et al., 2017c; Glauch and Escher 2020). Therefore, the PSII inhibition assay
could be applied in cases where agricultural activities may potentially impact
source water quality.

10.5.2 Neurotoxicity
Neurotoxicity assays applied for ecotoxicology and water quality monitoring have
been recently reviewed (Legradi et al., 2018). Inhibition of acetylcholinesterase
(AChE) is the most common in vitro endpoint for specific neurotoxicity

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 201

measured in water quality assessment. AChE is the enzyme responsible for

recycling the neurotransmitter acetylcholine and its inhibition indicates
interference with neurotransmitter metabolism. The AChE inhibition assay was
first developed by Ellman et al. (1961) and is a useful tool for detection of
neurotoxic insecticides such as the organophosphates and carbamate insecticides.
The assay was later optimised for use with environmental samples by Hamers
et al. (2000), who validated the technique with rainwater samples. It has been
used to assess neurotoxicity in diverse types of environmental water extracts
(Macova et al., 2011; Toušová et al., 2017). Organophosphate insecticide
parathion is commonly used as the assay reference compound and the reported
parathion EC50 values range from 26.3 to 120 µg/L (Escher et al., 2008b;
Macova et al., 2010). AChE inhibition was detected in wastewater influent,
wastewater effluent, recycled water, surface water and drinking water extracts
(Table 10.29), but the assay does suffer from false-positives that limit its
application. For example, Neale and Escher (2013) found that dissolved organic
carbon (DOC) concentrations as low as 2 mg of carbon/L (mgC/L) caused
quenching in the assay. DOC can be co-extracted during sample processing, so
the AChE inhibition assay is not recommended for DOC-rich samples, such as
wastewater or surface water.
Additional in vitro neurotoxicity assays that are currently available, although
most are still to be validated for water quality assessment, include neuronal and
glial cell viability assays using SK-N-SH (and derivatives, such as SH-SY5Y
cells) and C6 cells. These assays further include precursor cell differentiation

Table 10.29 Summary of studies that have applied the enzymatic

acetylcholinesterase (AChE) inhibition assay to different environmental water
extracts, where the AChE inhibition is expressed as parathion EQ.
Matrix AChE Inhibition Reference
Parathion EQ (µgparathion/ L)a
Wastewater influent 4.4–6.0 1
Wastewater effluent ,0.04 –3.9
b
1–8
Recycled water ,0.06–1.2, ,2.6 a
1, 3–5, 8
Surface water 0.11–0.27, ,2.6a 1–3, 6, 7
Drinking water 0.28 to ,2.6 a
1, 3
NB: Only studies that have applied SPE or LLE are included.
a
Parathion EQ calculated using the parathion EC10 in Neale et al. (2017c); bChlorpyrifos EQ
converted to Parathion EQ based on relative effect potency (REP) of 4.05 ( Neale and Escher,
2013).
References: 1(Macova et al., 2011); 2(Escher et al., 2008b); 3(Escher et al., 2014); 4(Leusch
et al., 2014b); 5(Macova et al., 2010); 6(Neale and Escher, 2013); 7(Neale et al., 2017c);
8
(Reungoat et al., 2010).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 202 Bioanalytical Tools in Water Quality Assessment

and apoptosis assays using neuroblastoma cells, glial maturation (myelination)

in U-373MG human astrocytoma cells, neurotransmitter receptor profiles in
neuroblastoma cells and interference with neurotransmitter enzymes or
postsynaptic receptors (reviewed in Atterwill et al., 1994; Costa, 1998;
Tiffany-Castiglioni et al., 2006; Coecke et al., 2007). The neuroblastoma assay
was developed for detection of natural neurotoxins such as cyanotoxins and
paralytic shellfish toxins (Kogure et al., 1988; Jellett et al., 1992; Manger et al.,
1993; Manger et al., 1995). This assay is relevant for testing of recreational
waters such as lakes and drinking water withdrawn from surface water. Although
the neuroblastoma assay has been mostly tested with freeze-dried samples, it is
becoming an increasingly applied tool for testing of water quality (Wood et al.,
2006; Cetojevic-Simin et al., 2009; Campora et al., 2010; Kerbrat et al., 2010).
For a xenobiotic to be toxic to the central nervous system, it must first cross the
blood–brain barrier. This process can be modelled in vitro using immortalised
brain endothelial cells, for example, SV-HCEC, HBEC-51 or BB19 cells (Prieto
et al., 2004), but is yet to be trialled with water samples.

10.5.3 Other assays

In addition to the above studies, new bioassays have been developed specifically to
detect modes of action relevant to pharmaceuticals. For example, Bernhard et al.
(2017) have developed beta-blocker and non-steroidal anti-inflammatory drug
assays, with a limit of detection of 2 µg/L metoprolol EQ and 0.5 µg/L
diclofenac EQ, respectively. The assays were applied to SPE-enriched wastewater
effluent, with 3.2–4.2 µg/L metoprolol EQ and 3.5 µg/L diclofenac EQ detected.
Recent studies have applied a TGFα-shedding assay to assess the biological
activity of pharmaceuticals that bind to G protein-coupled receptors (Zhang et al.,
2018a; Ihara et al., 2020). SPE-enriched wastewater from Japan and the UK were
applied, with most samples found to be angiotensin (AT1), dopamine (D2),
adrenergic (β1), acetylcholine (M1) and histamine (H1) receptor antagonists
(Zhang et al., 2018a). These assays have not been widely applied to date.

10.6 BIOASSAYS INDICATIVE OF REACTIVE TOXICITY

Reactive toxicity occurs when chemicals form covalent bonds with DNA, proteins
and membrane lipids. The field of in vitro cancer research has been active for several
decades and a few bioassay tools applied for water quality assessment have been
adapted from the medical field. Genotoxicity assays include those detecting
cytogenetic damage (i.e., structural DNA damage, e.g., the Comet assay) or
mutagenicity (i.e., introduction of mutations, e.g., the Ames test). Surface water
and wastewater have received early attention in terms of bioanalytical screening
of genotoxicants. Disinfection by-products (DBPs), which form due to the
reaction of disinfectants such as chlorine and chloramine with organic matter in
drinking and swimming pool water, are under increasing scrutiny for their

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 203

contribution to genotoxicity and carcinogenicity. DBPs are reactive chemicals and

are responsive in a number of assays indicative of reactive toxicity (Stalter et al.,
2016a). To date, there have been .400 studies on genotoxicity of drinking water
(Cortes and Marcos, 2018; DeMarini, 2020).
Most genotoxicity and mutagenicity assays can be run either with or without rat
liver S9 fraction, which is used to simulate metabolic activation. Epigenetic
carcinogens are chemicals that are not DNA-reactive and can cause cancer via
non-genotoxic mechanisms such as alteration of gene expression. In chemical risk
assessment, cell transformation assays using BALB/c 3T3 mouse fibroblast cells
closely model the various stages of in vivo carcinogenesis and provide an
integrated measure of carcinogenicity via multiple mechanisms including both
genotoxic and non-genotoxic pathways (Combes et al., 1999). To our knowledge,
these tests have not yet been used in water quality assessment and in the
following detailed sections we focus on genotoxicity and mutagenicity and
briefly discuss reactivity towards proteins and oxidative stress.

10.6.1 Genotoxicity
The umuC assay, which is also known as umu or SOS/umu, is used to assess DNA
damage via the inducible SOS response (Table 10.30). SOS genes are repressed
under normal conditions but are released in the presence of DNA damage to help
repair any damage (Michel, 2005). A number of Salmonella typhimurium strains
including TA1535/pSK1002, NM2009, NM3009 and NM5004 and the
Escherichia coli SOS Chromotest have been applied to water extracts (Escher
et al., 2014; Han et al., 2016). However, most studies have used TA1535/
pSK1002 either with or without metabolic activation, so this section will focus on
the TA1535/pSK1002 strain. The reference compound without metabolic
activation is usually 4-nitroquinoline N-oxide (4NQO) (ECIR1.5: 9.47 × 10−8 M
(Macova et al., 2011)), while 2-aminoanthracene (2AA) (ECIR1.5: 2.42 × 10−7 M
(Tang et al., 2014)) is often used as the reference compound with metabolic
activation.
The SOS Chromotest (Quillardet et al., 1982) and SOS umu/umuC (Oda et al.,
1985) assays were developed in the 1980s and are commonly used tools to screen
environmental water for genotoxicity (Table 10.30). Both SOS techniques
respond to genotoxicants through colorimetric detection of the SOS response,
which is induced by DNA damage. In the early 1990s, the SOS umu and
Chromotest assays were optimised for high throughput screening of surface
waters (Reifferscheid et al., 1991; Langevin et al., 1992). More recently, the
Vitotox kit was developed for detection of SOS response by luminescence (van
der Lelie et al., 1997; Verschaeve et al., 1999). The Vitotox assay is also in use
for water quality screening (Pessala et al., 2004).
The Comet assay (also known as single cell gel electrophoresis (SCGE) assay) is
another popular technique for detection of reactive toxicity in polluted waters

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
204

Table 10.30 Selection of in vitro assays used for assessment of genotoxicity in water samples, which target chlorinated by-products,
aromatic amines, PAHs, haloacetic acids and other disinfection by-products (DBP).
Assay for DNA Damage Cell Type Endpoint Reference
SOS response assays: umuC Bacterium S. typhimurium TA Induction of the umu operon (SOS 1–8
assay (also called umu and 1535/pSK1002 response) activates β-galactosidase,
SOS/umu), umu microtest which can metabolise the substrate to a
and SOS Chromotest coloured product for colorimetric
measurement
Cytotoxicity in SOS defective Bacterium E. coli (several K12 AB and Colony formation 9, 10
E. coli KL strains)
Vitotox assay (kit for detection Bacteria genetically modified SOS response, which induces 11, 12
of SOS response) S. typhimurium (TA 104 recN2–4 luminescence (the TA 104 pr1 strain,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

strain) which constantly expresses lux genes
is used as positive control)
Comet assay (also known as A variety of mammalian (incl. human) Measures DNA double-strand breaks in 13–18
SCGE assay) cells, also fish liver cells (zebrafish single cells (single-strand breaks in
Danio rerio and rainbow trout some variants). Staining technique,
Oncorhynchus mykiss RTL-W1, fluorescence. Image analysis results in
RTH-149) an output image resembling a comet.
The body of the comet represents
Bioanalytical Tools in Water Quality Assessment

undamaged cells and the tail, the

damaged cells
Alkaline yeast comet/SCGE Yeast Saccharomyces cerevisiae Same as normal comet but appears 19
DLH3 to be more sensitive than mammalian
cell line
(Continued)
by guest
Table 10.30 Selection of in vitro assays used for assessment of genotoxicity in water samples, which target chlorinated by-products,
aromatic amines, PAHs, haloacetic acids and other disinfection by-products (DBP) (Continued ).

Assay for DNA Damage Cell Type Endpoint Reference

Micronucleus formation Non-secreting human lymphoblast Micronucleus formation, measured by 20,21
measured by flow cytometry (WIL2-NS) flow cytometry
(FCMN or FCMMN assay)
Propidium iodide (PI) staining Mammalian and human cell lines can PI is fluorogenic and binds 22,23
and flow cytometry be used stoichiometrically to nucleic acid. DNA
content can be quantified via
fluorescence
GreenScreen EM (yeast Yeast S. cerevisiae transfected with a DNA damage, or rather the resulting 24,25
reporter gene assay) plasmid incorporating γEGFP3 DNA repair, which induces the GFP

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

SCE induction Chinese hamster lung (CHL) cells SCE is measured by a fluorescence 26–27
staining technique
References: 1(ISO13829); 2(Oda et al., 1985); 3(Hu et al., 2007); 4(Cao et al., 2009); 5(Reifferscheid et al., 1991); 6(Langevin et al., 1992); 7(Quillardet
et al., 1982); 8(White et al., 1996); 9(Aleem and Malik, 2003); 10(Aleem and Malik, 2005); 11(van der Lelie et al., 1997); 12(Verschaeve et al., 1999);
13
(Ostling and Johanson, 1984); 14(Rydberg and Johanson, 1978); 15(Singh et al., 1988); 16(Schnurstein and Braunbeck, 2001); 17(Plewa et al., 2002);
18
(Wagner and Plewa, 2008); 19(Miloshev et al., 2002); 20(Laingam et al., 2008); 21(NWC, 2011); 22(Nicoletti et al., 1991); 23(Riccardi and Nicoletti,
2006); 24(Cahill et al., 2004); 25(Keenan et al., 2007); 26(Ohe et al., 2009); 27(Perry and Wolff, 1974).
Current bioanalytical tools for water quality assessment
205
 206 Bioanalytical Tools in Water Quality Assessment

(Table 10.30). This assay relies on the differences in migration behaviour between
intact and damaged DNA in an electric field (Rydberg and Johanson, 1978; Ostling
and Johanson, 1984). Initially, the Comet assay was only able to detect
double-strand breaks but was later optimised for detection of single-strand breaks
(Singh et al., 1988). The first applications of the Comet assay for water quality
analysis were carried out in 2001 to assess the genotoxicity of rivers in Germany
and China (Schnurstein and Braunbeck, 2001; Zhong et al., 2001). Plewa and
coworkers have developed both bacterial and mammalian assays to test for DBP
genotoxicity. The mammalian assay applies a CHO cell line in the SCGE Comet
assay run alongside the CHO microplate cytotoxicity assay (Plewa et al., 2002;
Plewa et al., 2004a). This CHO cell adaptation of the assay has been utilised for
assessment of recreational waters (Liviac et al., 2010; Plewa et al., 2011).
The GreenScreen is a microplate assay that measures DNA repair (as a
consequence of DNA damage) in yeast transfected with green fluorescent protein
(GFP) (Cahill et al., 2004). The GreenScreen has been applied for assessment of
genotoxicity in industrial effluents (Keenan et al., 2007).
The ability to detect micronucleus formation by flow cytometry (Laingam et al.,
2008) has been exploited to detect this response in human lymphocytes (WIL2-NS)
exposed to a variety of water matrices, including treated sewage, reclaimed water
and drinking water (Leusch et al., 2014b).
There are several additional in vitro assays to detect genotoxic carcinogens
(Table 10.30), including thymidine kinase and hypoxanthine guanine
phosphoribosyltransferase mutation assays, sister chromatid exchange (SCE)
assays and chromosomal aberration assays (reviewed in Combes et al., 1999;
Kowalski, 2001), and new assays and protocols are regularly developed (Corvi
and Madia, 2017) but these have generally seen less to no application in water
quality monitoring.
The umuC assay has been applied to wastewater, surface water, recycled water
and drinking water (Table 10.31). Wastewater influent and wastewater effluent
were the most responsive, with many samples not having a response up to the
maximum REF in surface water or highly treated recycled water (e.g., RO or
ozone and biological-activated carbon (O3/BAC)). Increased genotoxicity after
disinfection in an Australian drinking water treatment plant (DWTP) was
observed by Neale et al. (2012). Genotoxicity increased from 4NQO EQ (−S9)
of 0.05 µg4NQO/L and 2AA EQ (+S9) of 0.18 µg2AA/L at the inlet to 4NQO EQ
(−S9) 0.43 µg4NQO/L and 2AA EQ (+S9) of 1.29 µg2AA/L at the outlet. The
effect at the outlet was observed after 23–33 times enrichment, with genotoxicity
without metabolic activation more sensitive. This suggests that with sufficient
enrichment, the umuC assay can be used to assess DBP formation during
drinking water treatment. Tap water also induced a response in the umuC assay
without S9 after 29–65 times enrichment (Stalter et al., 2016b). Focusing on
single DBPs, the addition of S9 did not increase genotoxicity and in some cases

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 207

Table 10.31 Summary of studies that have applied the umuC assay indicative of
genotoxicity to environmental water extracts, where genotoxicity with metabolic
activation (+S9) is expressed as 2-aminoanthracene (2AA) EQ and genotoxicity
without metabolic activation (−S9) is expressed as 4-nitroquinoline N-oxide
(4NQO) EQ.
Matrix Strain Genotoxicity +S9 Reference
2AA EQ (µg2AA/ L)
Wastewater influent TA1535/pSK1002 2–19a 1–3
Wastewater effluent TA1535/pSK1002 ,0.2–4.6 a
2, 3, 4–8
Recycled water TA1535/pSK1002 ,0.2–4.3 2–4, 7, 8
Surface water TA1535/pSK1002 0.18–0.38a, ,1.6a 2, 4, 6, 9, 10
Drinking water TA1535/pSK1002 ,0.34b–1.29, ,1.6a 2, 4, 10
Matrix Strain Genotoxicity −S9 Reference
4NQO EQ (µg4NQO/ L)
Wastewater influent TA1535/pSK1002 0.48–6.9c 1–3
c
Wastewater effluent TA1535/pSK1002 0.12–5.8 2–4, 6–8, 11
Recycled water TA1535/pSK1002 ,0.05–0.72c 2–4, 7, 8, 10
Surface water TA1535/pSK1002 0.01–0.69 2, 4, 9, 10, 12, 13
Drinking water TA1535/pSK1002 ,0.05–0.61c 2, 4, 10, 14
NB: Only studies that have applied SPE or LLE are included.
a
2AA EQ calculated using 2AA ECIR1.5 of 46.7 ng/L (Tang et al., 2014).
b
Results presented as benzo[a]pyrene (B[a]P) EQ and converted to 2AA EQ using the REP from
Macova et al. (2011);
c
4NQO EQ calculated using 4NQO ECIR1.5 of 18 ng/L (Macova et al., 2011).
References: 1(Lee et al., 2015); 2(Macova et al., 2011); 3(Tang et al., 2014); 4(Escher et al.,
2014); 5(Escher et al., 2008b); 6(Fang et al., 2012); 7(Jia et al., 2015); 8(Macova et al., 2010);
9
(Farre et al., 2013); 10(Neale et al., 2012); 11(Reungoat et al., 2010); 12(Han et al., 2016);
13
(Sun et al., 2017); 14(Stalter et al., 2016b).

reduced the potency (Stalter et al., 2016a), suggesting many DBPs are direct
genotoxicants.

10.6.2 Mutagenicity
The bacterial Ames assay for detection of mutagens was employed for water quality
testing soon after its publication in 1975 (Ames et al., 1975) (Table 10.32). Even
then, the assay was often applied directly, without extraction, to a wide range of
water types including surface water (Pelon et al., 1977; Vankreijl et al., 1980),
ozonated recycled water (Gruener, 1978), coal gasification process water (Epler
et al., 1978), drinking water (Simmon and Tardiff, 1976; Nestmann et al., 1979;

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 208 Bioanalytical Tools in Water Quality Assessment

Table 10.32 Selection of in vitro assays used for assessment of mutagenicity in water
samples, which target chlorinated by-products, aromatic amines, PAHs, haloacetic
acids and other DBPs.
Assay for Cell Type Endpoint Reference
Mutagenicity
Ames test (and Bacterium S. Number of histidine 1–4
modified Ames typhimurium (many revertants
test) strains incl. TA98,
TA100 and 98NR)
Mutatox assay Bacterium Genotoxic damage such 5
Aliivibrio as frame-shift mutations
fischeri or base-substitution point
mutations and more,
which induce a dark
variant of A. fischeri to
regain its luminescence
Alternative Yeast S. cerevisiae Formation of 6
mutagenicity D7 diploid strain ‘mutagen-specific’
test colonies on selective
media
References: 1(Ames et al., 1975); 2(Maron and Ames, 1983); 3(Barrueco et al., 1991); 4(Kado
et al., 1986); 5(Ulitzur et al., 1980); 6(Zimmermann et al., 1975).

Cheh et al., 1980), marine water (Kurelec et al., 1979), pulp and paper mill effluents
(Bjorseth et al., 1979; Carlberg et al., 1980) and different wastewaters (Rappaport
et al., 1979; Saxena and Schwartz, 1979). The Salmonella preincubation assay is a
modified Ames mutagenicity assay (Kargalioglu et al., 2002; Plewa et al., 2004b),
which is run in conjunction with the Salmonella typhimurium microplate
cytotoxicity assay and was applied to assess DBPs.
Most of the studies on mutagenicity testing of water quality simply reported a
positive or negative response (Berninger et al., 2019; Albergamo et al., 2020),
without determining an effect concentration (EC) or bioanalytical equivalent
concentration but there is also guidance on how to interpret such data (Roubicek
et al., 2020). Table 10.33 includes only studies that have reported an EC value, with
studies reporting the effect concentration inducing a revertant ratio of 1.5 (ECRR1.5).
The S. typhimurium strains applied in Table 10.33 include TA98, which responds
to frameshift mutations, and TA100 and TAmix, which both respond to base pair
substitutions (Kamber et al., 2009). Escher et al. (2014) found increasing
mutagenicity (e.g., lower ECRR1.5 values) in treated drinking water compared to
source water for all strains tested, with little difference between strains or with or
without S9. Effects were detected after 3–14 times enrichment in drinking water,
5–25 times enrichment in surface water and 0.6–83 times enrichment in
wastewater effluent (Table 10.33). Furthermore, Ames strains TA98, TA100 and
YG7108, the latter of which is responsive to nitrosamines, were tested with and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 209

Table 10.33 Summary of studies that have applied the Ames assay indicative of
mutagenicity to different environmental water extracts, where mutagenicity is
expressed as an effect concentration inducing a revertant ratio RR of 1.5 (ECRR1.5)
in units of REF.
Matrix Strain ECRR1.5 (REF) +S9 Reference
Wastewater effluent TA98 3.5 to .100 1, 2
TAmix 2.9–66 1, 2
Recycled water TA98 12.5 to .100 1, 2
TAmix 13.7 to .100 1, 2
Surface water TA98 4.5 1
TAmix .30 1
Drinking water TA98 3.2 1
TAmix 13.8 1
Matrix Strain ECRR1.5 (REF) −S9 Reference
Wastewater effluent TA98 6.3 to .100 1, 2
TA100 0.6–16 1, 2
TAmix 6.9–83 1, 2
Recycled water TA98 .30 to .100 1, 2
TA100 0.5 to .30 1, 2
TAmix 35–69 1, 2
Surface water TA98 14 1
TA100 25 1
TAmix .30 1
Drinking water TA98 4.6 1
TA100 5.0 1
TAmix 4.9 1
References: (Escher et al., 2014); (Jia et al., 2015).
1 2

without metabolic activation in source water and treated drinking water extracts
from three French DWTPs (Neale et al., 2020b). However, none of the samples
had an effect up to the maximum REF of 200.

10.6.3 Non-genotoxic electrophilic mechanisms

Reactive toxicity can also take place via electrophilic attack by electron-deficient
chemicals (electrophiles, e.g., the pesticide atrazine), which can bind to nucleophilic
(electron donating) groups on endogenous molecules causing structural damage. An
example of a biological nucleophile is the amino acid cysteine in peptides and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 210 Bioanalytical Tools in Water Quality Assessment

proteins, and the bases of DNA. Although reactivity with DNA causes genotoxicity and
has been subject to intensive research, no bioassays indicative of reactive toxicity
caused by binding to cysteine have been reported for water quality assessment. This
is an important gap in knowledge as damage to proteins may also lead to adverse
effects. Glutathione (GSH) is a small cysteine that protects the cell as an antioxidant
and by detoxifying electrophiles. Xenobiotic exposure thus has the potential to cause
GSH depletion, which can ultimately lead to protein damage. The naturally high
concentration of GSH in cells allows its depletion to be quantified using chemical
methods or an enzymatic assay. Although depletion of cellular GSH has been
assessed in various cell types and water samples (Table 10.34), it is difficult to
interpret the significance of the results.
The role of GSH for cell viability can also be assessed with a differential bacterial
assay. The wild type E. coli strain MJF276 and its mutant strain MJF335 differ in
that, MJF355 lacks the enzymes to synthesise GSH. In a growth inhibition assay,
both strains show the same sensitivity, unless the tested chemical is a reactive
electrophile. In this case, the mutant strain cannot defend the cell and thus the
resulting EC50 will be lower than for the wild type (Harder et al., 2003; Richter
and Escher, 2005). This assay has been applied for fingerprinting the toxicity of
DBPs (Stalter et al., 2016a) but proved to be too insensitive for drinking water
testing.

Table 10.34 Selection of in vitro assays used for assessment of reactive toxicity
towards proteins (GSH depletion).
Assay Cell Type Endpoint Reference
GSH assay Human liver cells GSH depletion. 1, 2
(Hep-G2) Fluorimetric
quantification of GSH
concentration in cells
GSH reductase Rainbow trout GSH depletion. 3–5
enzymatic primary Colorimetric
recycling assay hepatocytes measurement of GSH
(modified) concentration in cells
Differential Bacterium E. coli Growth inhibition. 6, 7
bacterial growth (MJF276 and Comparison of the EC50s
inhibition assay MJF335 (mutant)) of two differential strains
of which, the mutant
lacks GSH for cellular
defence
References: 1(Hissin and Hilf, 1976); 2(Marabini et al., 2006); 3(Owens and Belcher, 1965);
4
(Baker et al., 1990); 5(Farmen et al., 2010); 6(Harder et al., 2003); 7(Richter and Escher, 2005).
GSH = glutathione.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 211

10.6.4 Oxidative stress

A selection of available assays for detection of oxidative stress in water is listed in
Table 10.35. The presence of reactive oxygen species (ROS) is used as a warning
sign of oxidative stress in water samples (Marabini et al., 2006). ROS can be
quantified using a colorimetric method, in which oxidation by ROS causes a
substrate added to the cells to fluoresce. This endpoint, however, lacks
specificity. In a study on disinfected drinking water, where additional indicators
of oxidative stress were measured (Table 10.35), it was shown that the products
of lipid peroxidation were present and that antioxidant enzymes were active
although no ROS could be quantified (Shi et al., 2009b). ROS can also be
measured as the production of free radicals (superoxide and hydroxyl radical) in
a cell line exposed to a water sample (Xie et al., 2010). Increased levels of the
oxidised form of GSH, the GSH disulphide GSSG, in relation to the
concentration of GSH is suggestive of oxidative stress but a study on surface
water found no clear correlation between the GSH/GSSG ratio and ROS
formation (Neale et al., 2017a).

Table 10.35 Selection of in vitro assays used or potentially useful for assessment of
oxidative stress in water samples.
Assay Cell Type Endpoint Reference
ROS assay (indirect Human liver Oxidation of a substrate 1–4
detection of ROS) cells (Hep-G2) (DCFH-DA) leads to a
fluorescent product
ROS assay (as Rainbow trout Oxidation of a substrate 5
above) primary (DCF-DA) leads to a
hepatocytes fluorescent product
GSH/GSSG-Glo Human liver GSH/GSSG ratio 4
assay cells (Hep-G2
Antioxidant Hep-G2 cells Enzyme activity of 3, 6, 7
response antioxidase, GSH
peroxidase (GSH-Px),
superoxide dismutase
(SOD)
Antioxidant Human liver Cellular concentration of 3, 8, 9
response cells (Hep-G2, lipid peroxidation
L-02) product malonaldehyde
(MDA)
References: 1(Wang and Joseph, 1999); 2(Marabini et al., 2006); 3(Shi et al., 2009b); 4(Neale
et al., 2017a); 5(Farmen et al., 2010); 6(Flohe and Gunzler, 1984); 7(Oberley and Spitz, 1984);
8
(Yagi, 1998); 9(Xie et al., 2010). DCF-DA or DCFH-DA = 2′ ,7′ -dichlorofluorescein diacetate.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 212 Bioanalytical Tools in Water Quality Assessment

Water samples that showed ROS formation also often activated the oxidative
stress response (Neale et al., 2017a). A complementary endpoint would therefore
be induction of the mammalian cellular defence mechanism against oxidative
stress, which is activated by multiple pathways (not only ROS) and is discussed
in detail in Section 10.7.1.

10.7 BIOASSAYS INDICATIVE OF ADAPTIVE STRESS

RESPONSES
Adaptive stress response pathways are activated to help restore cells back to
homeostasis after damage from stressors, including organic chemicals (Simmons
et al., 2009). This chapter will focus on three adaptive stress response pathways
commonly applied to environmental water extracts: oxidative stress response
(Nrf2), p53 response for genotoxicity and NF-κB response for inflammation.
Assays indicative of adaptive stress responses hypoxia and heat shock response
did not have a response in drinking water, surface water, wastewater or recycled
water extracts (Escher et al., 2014).

10.7.1 Oxidative stress response

Chemicals that produce ROS and electrophilic chemicals can induce the oxidative
stress response, in particular the antioxidant defence pathway Keap1-Nrf2
(Kobayashi et al., 2009). These chemicals release transcription factor Nrf2 from
the negative regulator Keap1, which then translocates to the nucleus and activates
the antioxidant response element (Zhang, 2006). Five mammalian reporter gene
assays have been applied to evaluate the oxidative stress response in
environmental water extracts (Table 10.36). tert-Butylhydroquinone (tBHQ) is
often used as the assay reference compound, with similar ECIR1.5 or lowest
observed effect concentration (LOEC) at an induction factor of 1.5 for AREc32,
ARE GeneBLAzer, Nrf2 CALUX and Nrf2 reporter gene assay. The tBHQ
ECIR1.5 value for Nrf2-MDA-MB was over an order of magnitude higher (i.e.,
less toxic) and the assay did not report any effect with drinking water, surface
water, wastewater effluent or recycled water extracts (Escher et al., 2014; Jia
et al., 2015). Consequently, Nrf2-MDA-MB does not appear to be suitable for
environmental water extracts.
Effects in drinking water have been detected after 3–102 times enrichment
(Table 10.37), with drinking water treated with membrane filtration often not
inducing the oxidative stress response up to the maximum tested REF of 100–150
(Albergamo et al., 2020; Neale et al., 2020b). In contrast, an increase in the
oxidative stress response has been observed after drinking water disinfection
(Neale et al., 2012; Escher et al., 2013; Hebert et al., 2018), indicating that
oxidative stress response assays can detect formed DBPs. The ECIR1.5 in surface
water varied from 0.6 to 93 REF, while effects were observed in wastewater

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 213

Table 10.36 Common assays applied to evaluate the oxidative stress response in
water extracts.
Assay Cell Line Detection tBHQ tBHQ EC
Method ECIR1.5 ECIR1.5 Reference
(M) (µg//L)
AREc32 MCF-7 Luminescence 1.32 × 219 Escher et al.
10−6 (2012)
ARE HepG2 Fluorescence 2.44 × 406 Neale et al.
GeneBLAzer 10−6 (2015b)
Nrf2 CALUX U2OS Luminescence 1.00 × 166 van der
10−6 a Linden et al.
(2014)
Nrf2 reporter HepG2 Luminescence 2.00 × 332 Lundqvist
gene assay 10−6 et al. (2019a)
Nrf2-MDA-MB MDA-MB- Luminescence 3.30 × 5490 Jia et al.
231-745 10−5 (2015)
a
LOEC at an induction factor of 1.5. tBHQ = t-butylhydroquinone.

influent and effluent from ECIR1.5 0.3 to 30 REF and ECIR1.5 2 to 47 REF,
respectively (Table 10.37). The reported WWTP removal efficacy ranged from
61% to 85% (Volker et al., 2017; Nivala et al., 2018). Effects in recycled water
ranged from ECIR1.5 4 to 94 REF.

10.7.2 p53 response

The p53 response responds to DNA damage by initiating repair proteins, changing
the cell cycle or by causing apoptosis (Knight et al., 2009). Two assays, the p53
GeneBLAzer and p53 CALUX, have been applied to environmental water
extracts (Table 10.38). The two assays use different reference compounds,
making it difficult to compare assay sensitivity, with p53 CALUX results
reported in both actinomycin D EQ and cyclophosphamide EQ. Most studies
report either no effect or cytotoxicity in p53 response assays (Table 10.39). The
p53 response in the Danube River was observed after 65 times enrichment in the
p53 GeneBLAzer, which gave a mitomycin EQ of 235 ngmitomycin/L (Neale et al.,
2015b). p53 activity was also detected in wastewater influent and effluent in
Finland using the p53 CALUX after the addition of S9 for metabolic activation
(Välitalo et al., 2017), with the results expressed as cyclophosphamide EQ.

10.7.3 NF-κB response

The NF-κB-mediated response to inflammation can be induced by a range of
compounds including metals, carcinogens and bacterial products (Ahn and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 214 Bioanalytical Tools in Water Quality Assessment

Table 10.37 Summary of studies that have applied assays indicative of the oxidative
stress response to different environmental water extracts, where the oxidative stress
response is expressed as the concentration causing an induction ratio of 1.5 (ECIR1.5)
in units of REF.
Matrix Assay ECIR1.5 Reference
(REF)
Wastewater influent AREc32 0.28–4.7 1–4
Nrf2 reporter gene assay 8.1–30 5
Wastewater effluent AREc32 1.5–22 1–4, 6–8
ARE GeneBLAzer 8.9–17 9
Nrf2-CALUX 4.8 10
Nrf2 reporter gene 47 to .50 5
assay
Nrf2-MDA-MB .10 7, 10
Recycled water AREc32 4.2–94 1, 3, 6, 7
Nrf2-CALUX 4.8 to .30 10
Nrf2-MDA-MB .10 to .20 7, 10
Surface water AREc32 0.6 to .100 1, 6, 8, 11–17
ARE GeneBLAzer 6.9 to .490 9, 18, 19
Nrf2-CALUX 6.9 10
Nrf2 reporter gene 22 20
assay
Nrf2-MDA-MB .20 10
Drinking water AREc32 2.5 to .150 1, 6, 14, 16, 21–23
Nrf2-CALUX 2.9 1
Nrf2 reporter gene 21–25 20
assay
Nrf2-MDA-MB .20 10
NB: Only studies that have applied SPE or LLE are included.
References: 1(Escher et al., 2012); 2(Nivala et al., 2018); 3(Tang et al., 2014); 4(Volker et al.,
2017); 5(Lundqvist et al., 2019b); 6(Escher et al., 2013); 7(Jia et al., 2015); 8(Mueller et al., 2021);
9
(Neale et al., 2017c); 10(Escher et al., 2014); 11(Farre et al., 2013); 12(Hashmi et al., 2018);
13
(Müller et al., 2018); 14(Neale et al., 2012); 15(Neale et al., 2018a); 16(Neale et al., 2020b);
17
(Neale et al., 2020a); 18(König et al., 2017); 19(Neale et al., 2015b); 20(Lundqvist et al., 2019a);
21
(Albergamo et al., 2020); 22(Hebert et al., 2018); 23(Stalter et al., 2016b).

Aggarwal, 2005). These stimuli destabilise the IkB–NF-κB complex, allowing

NF-κB to activate target genes in the nucleus (Gilmore, 2006). Three reporter
gene assays have been applied to evaluate the NF-κB response in environmental

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 215

Table 10.38 Common assays applied to evaluate the p53 response in water extracts.
Assay Cell Detection Reference EC10* or EC10* or
Line Method Compound EC†IR1.5 EC†IR1.5 (µg// L)
(M)
p53 CALUX U2OS Luminescence Actinomycin D 2.00 × 2.5*
(+/−S9) 10−9*
p53 HCT-116 Fluorescence Mitomycin C 4.53 × 15†
GeneBLAzer 10−8†

water extracts (Table 10.40). The tumour necrosis factor alpha (TNFα) ECIR1.5
value for NF-κB GeneBLAzer was 4.5 times lower than the NF-κB reporter gene
assay developed by Lundqvist et al. (2019b) (Table 10.40), while no reference
compound data were available for NF-κB CALUX. Extracts of drinking water,
surface water and wastewater were active in the NF-κB GeneBLAzer assay,
while only wastewater influent was active in the NF-κB reporter gene assay and
no samples were active in NF-κB CALUX (Table 10.41). Although effects are
often observed at low enrichment factors in NF-κB GeneBLAzer, the assay may
not be suitable to evaluate the effects of micropollutants. Endotoxins activate
NF-κB and a recent study showed that co-extracted endotoxins likely explained
most of the effects in surface water extracts in the NF-κB GeneBLAzer assay
(Neale et al., 2018b).

10.8 BIOASSAYS INDICATIVE OF APICAL EFFECTS

In addition to assays indicative of different stages of the cellular toxicity pathway,
cytotoxicity and whole organism assays indicative of apical effects are commonly
applied to water quality monitoring. Furthermore, some organisms, such as
zebrafish, are used as a model species for human health risk assessment
(Bambino and Chu, 2017). Although these assays are often used for direct
toxicity assessment, we have focused on assays applied to water extracts,
including cytotoxicity, algal growth inhibition and fish embryo toxicity (FET).
These assays can be run in well plates. The Daphnia immobilisation assay has
also been applied to passive sampler extracts (Hamers et al., 2018; de Baat et al.,
2019a; de Baat et al., 2020), but to our knowledge not to SPE extracts.

10.8.1 Cytotoxicity
Cytotoxicity assays can be used to monitor cytotoxicity on its own but are mostly
included as part of a test battery or for concurrent monitoring of acute toxicity as
a quality assurance step for specific toxicity assays.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
216
Table 10.39 Summary of studies that have applied assays indicative of the p53 response to different environmental water
extracts, where the p53 response is expressed as the concentration causing an induction ratio of 1.5 (ECIR1.5) or equivalent
concentration (EQ).
Matrix Assay ECIR1.5 (REF) Equivalent Concentration Reference
Wastewater influent p53 CALUX +S9 61–6,200 μg/L 1
Cyclophosphamide EQ
Wastewater effluent p53 CALUX .30 ,53–540 μg/L 1, 2
Cyclophosphamide EQ
p53 CALUX +S9 .30 2
p53 GeneBLAzer .10, cytotoxic 2, 3
Recycled water p53 CALUX .30 2
p53 CALUX +S9 .30 2

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

p53 GeneBLAzer .20 to .30 2
Surface water p53 CALUX .30 2
p53 CALUX +S9 .30 2
p53 GeneBLAzer 65 to .450, cytotoxic ,34–235 ng/L 2–6
Mitomycin EQ
Drinking water p53 CALUX .30 2
Bioanalytical Tools in Water Quality Assessment

p53 CALUX +S9 .30 2

p53 GeneBLAzer .150 2, 6, 7
NB: Only studies that have applied SPE or LLE are included.
References: 1(Välitalo et al., 2017); 2(Escher et al., 2014); 3(Neale et al., 2017c); 4(König et al., 2017); 5(Neale et al., 2015b); 6(Neale et al., 2020b);
7
(Hebert et al., 2018).
 Current bioanalytical tools for water quality assessment 217

Table 10.40 Common assays applied to evaluate the NF-κB response in

water extracts.
Assay Cell Detection TNFα EC Reference
Line Method ECIR1.5
(ng//L)
NF-κB THP-1 Fluorescence 20 Neale et al.
GeneBLAzer (2015b)
NF-κB reporter HepG2 Luminescence 90 Lundqvist et al.
gene assay (2019b)
NF-κB CALUX — Luminescence — —

Table 10.41 Summary of studies that have applied assays indicative of the NF-κB
response to different environmental water extracts, where the NF-κB response was
expressed as the concentration causing an induction ratio of 1.5 (ECIR1.5) in units of
REF.
Matrix Assay ECIR1.5 (REF) Reference
Wastewater influent NF-κB GeneBLAzer 0.05–0.36 1
NF-κB reporter gene assay 0.30 to .50 2
Wastewater effluent NF-κB GeneBLAzer 0.09 to .20 1, 3, 4
NF-κB reporter gene assay .50 2
NF-κB CALUX .30 3
Recycled water NF-κB GeneBLAzer .20 3
NF-κB CALUX .30 3
Surface water NF-κB GeneBLAzer 0.1 to .250 3–8
NF-κB CALUX .30 3
Drinking water NF-κB GeneBLAzer 10 to .500 3, 8, 9
NF-κB CALUX .30 3
NB: Only studies that have applied SPE or LLE are included.
References: 1(Nivala et al., 2018); 2(Lundqvist et al., 2019b); 3(Escher et al., 2014); 4(Neale et al.,
2017c); 5(König et al., 2017); 6(Neale et al., 2015b); 7(Neale et al., 2018b); 8(Neale et al., 2020b);
9
(Hebert et al., 2018).
REF= relative extraction factor.

10.8.1.1 Bacterial toxicity

Bacterial bioluminescence inhibition assays, such as the Microtox and
BLT-Screen assays, have been applied to drinking water, surface water and
wastewater extracts (Tang et al., 2013b; van de Merwe and Leusch, 2015).
This type of assay utilises the light emission in naturally bioluminescent

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 218 Bioanalytical Tools in Water Quality Assessment

bacteria, for example, Aliivibrio fischeri (formerly called Vibrio fischeri),

Photobacterium phosphoreum or P. leiognathi, as a measure of overall cellular
energy status and health. A decreased light output indicates interference with
energy metabolism and overall cellular health, reflecting the combined baseline
toxicity of all chemicals in the sample. Bacterial bioluminescence inhibition
assays are therefore suitable for screening of overall non-specific toxicity. In
addition to the low cost and simplicity of these assays, their applications in
water quality testing are wide ranging and well represented in the literature
providing a large volume of comparative information. The water types tested
with the Microtox, for example, span across effluents of coal gasification
(Timourian et al., 1982), oil refineries (Chang et al., 1981), pulp mills (Rosa
et al., 2010) and sewage treatment plants (Farré et al., 2002) as well as
environmental waters (Dizer et al., 2002) and drinking water (Guzzella et al.,
2004). Recent modifications have seen this type of assay adapted to 96-well
plate format (Macova et al., 2010).
Although bacterial toxicity assays are simple, they are fast (15–30 minutes)
and are responsive to a wide range of organic micropollutants. For example,
50% inhibition of bioluminescence was observed after 0.5–17 times enrichment
in wastewater influent, 2–27 times enrichment in wastewater effluent, 8–87
times enrichment in surface water and 3–40 times enrichment in drinking
water (Table 10.42). Bacterial toxicity assays can detect the formation of
DBPs, with increasing effect (i.e., decreasing EC50 values) reported throughout

Table 10.42 Summary of studies that have applied bacterial toxicity assays to
different environmental water extracts, where the effect is expressed as the
concentration causing 50% effect (EC50).
Matrix Assay EC50 (REF) Reference
Wastewater influent Microtox 0.48–17 1–4
Wastewater effluent Microtox 3.0–27 1–6
BLT-Screen 1.6 7
Recycled water Microtox 10–102 1–3, 5, 6
Surface water Microtox 8.2–87 1, 2, 6, 8, 9
BLT-Screen 12 7
Drinking water Microtox 3.2–40 1, 2, 6, 9, 10
BLT-Screen 5.8 7
NB: Only studies that have applied SPE or LLE are included.
References: 1(Escher et al., 2012); 2(Macova et al., 2011); 3(Tang et al., 2014); 4(Volker et al.,
2017); 5(Macova et al., 2010); 6(Tang et al., 2013b); 7(van de Merwe and Leusch, 2015); 8(Farre
et al., 2013); 9(Neale et al., 2012); 10(Stalter et al., 2016b).
REF= relative extraction factor.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 219

DWTPs (Escher et al., 2012; Neale et al., 2012). Although bacterial toxicity
assays only provide information about non-specific effects and should be
complemented with assays indicative of specific effects, their advantage is that
they can also be used in locations that only have access to microbiology
laboratory facilities.
In addition to bacterial bioluminescence inhibition assays, luminescent bacterial
biosensors indicative of DNA damage, oxidative stress and protein damage have
been developed in recent years (Woutersen et al., 2011). There is increasing
interest in applying these biosensors as online water quality monitoring tools,
although further research is required to improve sensitivity.

10.8.1.2 Mammalian cytotoxicity

Mammalian and human cell-based assays are typically applied to measure specific
toxicity in reporter gene assays. As discussed, however, it is crucial when running
such assays to also monitor baseline toxicity to ensure that the specific response is
not masked by cytotoxicity. The cytotoxicity endpoint is a valuable endpoint in its
own rights because it targets all chemicals in a water sample and can actually be
applied for specific cell types that are representative and relevant for the
particular case of concern. Different types of cells may exhibit differential
toxicity due to selective and cell-specific function toxicity (Seibert et al., 1996).
A project assessing chemical-induced toxicity from drinking water, for example,
should include a measure of cytotoxicity to gastro-intestinal and liver cells, which
are the cell types likely to be exposed to the highest doses following drinking
water intake.
Most cell viability assays rely on functional assays (Vinken and Blaauboer,
2017), such as mitochondrial activity quantified with the 3-[4,5-dimethylthiazol-
2-yl]-2,5-diphenyltetrazolium (MTT) assay, where tetrazolium is reduced to
formazan by the mitochondrial enzyme succinate dehydrogenase (Table 10.43). A
similar assay principle underlies the Alamar Blue assay, where resazurin is
reduced to resorufin by mitochondrial enzymes (Table 10.43). Another cell
viability assay principle is the membrane integrity as can be assessed by the
neutral red uptake (NRU) or the lactate dehydrogenase (LDH) assay. LDH is a
stable enzyme that leaks from the cell in relatively high amounts upon cell
plasma membrane damage and can be quantified colorimetrically (Table 10.43).
Cell growth inhibition can be measured by absorbance, after staining and with
bright-field microscopic imaging (Table 10.43).

10.8.1.3 Fish cell lines

Fish cell lines have been used for chemical and effluent assessment since the
early 1990s (Table 10.44). Cytotoxicity is usually measured in fish cells via
different staining techniques including Alamar Blue and NRU. The response in
fish cell lines was well correlated with acute fish toxicity (Tanneberger et al.,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 10.43 Selection of human and other mammalian in vitro bioassays used for assessment of cytotoxicity in water samples.
Assay Cell Type Endpoint Reference 220
MTT assay Various human (e.g., Hep-G2, MELN, Colorimetric measurement of 1–6
HG5LN-hPXR (transfected HeLa cells)) and live cells
other mammalian cells (e.g., mouse lymphoma
cells EL4.3)
Alamar Blue assay (also known Human kidney cells (HK2) Reduction of substrate 7–8
as resazurin reduction assay) (Alamar Blue) by live cells
yields fluorescent product
Caco2-NRU Human epithelial colorectal adenocarcinoma Cell viability (measured by 9–11
cells (Caco-2) NRU test)
NRU Human breast cancer cells (MCF-7) Cell viability (measured by 12, 2
NRU test)
LDH leakage, for example, Human liver cells (HepG2) Cell viability. Colorimetric 13–15
CytoTox 96® Non-radioactive measurement of LDH leakage

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

cytotoxicity assay from lysed cells
Mammalian cell microplate Chinese hamster ovary cells (CHO AS52) Cell growth inhibition 16–18
cytotoxicity assay (measured by absorbance at
595 nm)
SRB assay Mouse neuroblastoma cells (neuro-2A); Cell growth (measured by 19, 20
human foetal lung cells (MRC-5) protein staining)
Cell viability and growth Any adherent cell Confluency assessed via 21
Bioanalytical Tools in Water Quality Assessment

imaging analysis using an

Incucyte S3
References: 1(Mosmann, 1983); 2(Zegura et al., 2009); 3(Miege et al., 2009); 4(Shi et al., 2009b); 5(Creusot et al., 2010); 6(Delgado et al., 2011); 7(Page
et al., 1993); 8(Bunnell et al., 2007); 9(Borenfreund and Puerner, 1985); 10 (Konsoula and Barile, 2005);11(NWC, 2011); 12(Ma et al., 2005); 13(Nachlas
et al., 1960); 14(Promega, 2009); 15(Marabini et al., 2007); 16(Plewa et al., 2000); 17(Plewa et al., 2002);18(Plewa et al., 2004a);, 19(Skehan et al., 1990);
20
(Cetojevic-Simin et al., 2009); 21(Escher et al., 2019). LDH = lactate dehydrogenase, MTT = 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium,
NRU = neutral red uptake, SRB = sulphorhodamine B.
 Current bioanalytical tools for water quality assessment 221

Table 10.44 Selection of fish in vitro bioassays used for assessment of cytotoxicity
targeting all chemicals in water samples.
Assay Cell Type Endpoint Reference
Alamar Blue Rainbow trout (O. Reduction of substrate 1–5
assay (also mykiss) liver and gill (Alamar Blue) by live cells
known as cells (e.g., RTL-W1, yields fluorescent product
resazurin RTgill-W1), brown
reduction assay) bullhead (Ictalurus
nebulosus, BB-3
cell line)
CFDA-AM Rainbow trout (O. Membrane integrity. The 2–6
mykiss) liver and gill esterase substrate
cells (e.g., RTL-W1, (CFDA-AM) is converted
RTgill-W1), brown to a fluorescent product
bullhead (I. by esterases in intact
nebulosus, BB-3 plasma membranes
cell line)
NRU assay Rainbow trout (O. Cell viability measured by 7–10
mykiss) liver cells staining (retention of
(RTL-W1) neutral red)
PI staining and Rainbow trout (O. PI is fluorogenic and 11, 12
flow cytometry mykiss), liver cells binds stoichiometrically to
nucleic acid. DNA content
can be quantified via
fluorescence
References: 1(Page et al., 1993); 2(Schreer et al., 2005); 3(Schirmer et al., 2001); 4(Grung et al.,
2007); 5(Farmen et al., 2010); 6(O’Connor et al., 1991); 7(Borenfreund and Puerner, 1985);
8
(Klee et al., 2004); 9(Keiter et al., 2006); 10(Wölz et al., 2008); 11(Zucker et al., 1988); 12(Gagné
and Blaise, 1998). CFDA-AM = 5-carboxyfluorescein diacetate acetoxymethyl methyl ester,
NRU = neutral red uptake, PI = propidium iodide.

2013), in particular if internal concentrations are compared (Stadnicka-Michalak

et al., 2014).

10.8.2 Algal growth inhibition

Algal growth inhibition assays using green microalgae R. subcapitata have been
applied to wastewater effluent, recycled water and surface water extracts, with
only one study testing drinking water extracts (Table 10.45). Most studies applied
the CAT, with growth inhibition measured after 24 h based on Escher et al.
(2008a), while one study used 72 h algal growth inhibition based on the OECD
guideline 201 (OECD, 2011). The CAT was responsive to wastewater effluent
extracts, with 10% growth inhibition observed at an REF from 0.7 to 13, while

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 222 Bioanalytical Tools in Water Quality Assessment

Table 10.45 Summary of studies that have applied algal growth inhibition assays to
different environmental water extracts.
Matrix Assay EC10 (REF) Reference
Wastewater effluent CAT (24 h growth) 0.71–13 1–4
Recycled water CAT (24 h growth) 0.70 to .20 2, 3
Surface water CAT (24 h growth) 1.3 to .90 1, 2, 4
Algal growth inhibition 17 to .100a 5
Drinking water CAT (24 h growth) 14 2
The effect was expressed as the concentration causing 50% effect (EC50) in units of REF.
NB: Only studies that have applied SPE or LLE are included. CAT =combined algae test.
a
EC50 reported.
References: 1(Escher et al., 2008b); 2(Escher et al., 2014); 3(Jia et al., 2015); 4(Neale et al.,
2017c); 5(Toušová et al., 2017).

10% growth inhibition in surface water occurred at an REF from 1.3 to 51

(Table 10.45). No effects were observed in recycled water treated with RO/AO
or O3/BAC up to the maximum tested REF.

10.8.3 Fish embryo toxicity

The FET assay has only been applied for water quality monitoring in a limited
number of studies. Escher et al. (2014) applied the 48 h FET assay to drinking
water, surface water, wastewater effluent and drinking water, but only one
wastewater effluent extract had an effect, with 10% mortality observed at an REF
of 5. Furthermore, the 48 h FET assay was applied to surface water extracts from
the Danube River, with the EC50 value ranging from REF 111 to 665 (Neale
et al., 2015b). The EC50 values were around an order of magnitude lower in
surface water from four river basins in Europe in the 96 h FET assay (Toušová
et al., 2017). Both Toušová et al. (2017) and Neale et al. (2015b) applied large
volume SPE, with 50–500 L of water concentrated. The 48 h FET assay has also
been applied to surface water and wastewater passive sampler extracts, with
styrene divinylbenzene ‘Speedisk’ samplers proving more responsive than
silicone rubber samplers (Hamers et al., 2018).

10.9 CONCLUSIONS
A wide range of in vitro and well plate-based in vivo assays based on different stages
of the cellular toxicity pathway including induction of xenobiotic metabolism,
receptor-mediated effects, adaptive stress responses and apical effect have been
widely applied for water quality assessment. The majority of studies have
focussed on estrogenic activity, followed by other endocrine modes of action.
Assays indicative of activation of AhR, activation of PXR, estrogenic activity,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Current bioanalytical tools for water quality assessment 223

oxidative stress response and bacterial toxicity have consistently detected effects in
wastewater effluent, surface water and drinking water. Mammalian reporter gene
assays tend to be more sensitive than yeast reporter gene assays or whole
organism assays indicative of the same endpoint. Over the past decade, there
have been significant improvements in quantitative expression of bioassay results
moving away from the initially simpler binary ‘yes/no’ results. This has made it
easier to compare bioassay results between studies and across different types of
waters, providing a much finer understanding and meaningful interpretation of
bioanalytical results as a measure of water quality.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 11
Quality assurance and quality
control (QA/QC)

11.1 INTRODUCTION
Many of the criticisms of in vitro bioassays about their perceived lack of
reproducibility, reliability and standardisation across different laboratories can be
addressed by establishing quality assurance and quality control (QA/QC) checks
in all experiments undertaken in the laboratory. The ‘reproducibility crisis’ in
biomedical sciences can only be overcome in in vitro toxicology by strict and
transparent method validation and continuous QA/QC (Hirsch and Schildknecht,
2019).
In a creative research environment, the idea of creating an additional workload by
incorporating additional QA/QC steps in experimental protocols may appear
cumbersome. The additional degree of confidence in the data generated is,
however, undeniably worth the extra effort. In water quality monitoring, this
additional degree of confidence is absolutely necessary to ensure the accuracy
and reproducibility of the bioassay results, as these can have significant
implications for the utility or authority that manages the (waste)water treatment
plant or the receiving watershed where the samples came from.
In the early stages of development, bioanalytical tools were often performed in
individual vials, where only a few samples can be tested in each assay run.
Although low-throughput is not an issue in initial assay development, the
conversion of an assay to a high-throughput microplate format is a prerequisite
for adapting that assay to water quality monitoring. This chapter is therefore
written for 96- and 384-well plate assays, which are currently the most common

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0225

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 226 Bioanalytical Tools in Water Quality Assessment

microplate format. Although some details may differ, the concepts and
principles discussed are of course applicable to any assay format and can easily
be adapted.
The following section gives an introduction to method validation. Once a method
has been validated and a standard operating protocol (SOP) produced for routine
application, QA/QC procedures need to be put in place to monitor and ensure
consistent performance over time. QA/QC implementation is addressed in the
subsequent sections.

11.2 METHOD VALIDATION

Method validation is a crucial step in adapting a bioassay from a research tool or
application in chemical risk assessment into a bioanalytical tool for water quality
assessment. Validation of a bioanalytical method demonstrates that the assay is
fit-for-purpose and ensures that the analysis is reliable, consistent and that there is
confidence in the result. Determining the capabilities of an assay is also useful for
understanding how it performs compared to other methods, whether these are
other bioassays or chemical analysis methods. Bioanalytical method validation
involves determining a variety of assay performance characteristics such as
accuracy, precision, robustness, sensitivity and sample stability (CDER/FDA,
2001; EMEA, 2011). All of these parameters can be determined using relatively
simple (albeit time-consuming) test protocols, where a series of samples is
analysed repeatedly on different assay runs with different operators over a period
of several weeks. Specificity and selectivity are also important attributes for
method validation that have been covered in depth in Chapter 9.

11.2.1 Accuracy
Accuracy describes how close a bioassay result is to the true value (Figure 11.1).
Accuracy is determined by replicate analysis of samples including at least three
different concentrations of the analyte of interest. With in vitro bioassays, the
analyte is usually a model compound used as a reference toxicant (e.g.,
17β-estradiol for bioassays that measure estrogenic activity). The mean value
from repeated testing in an accurate in vitro assay should be within 15%–30% of
the actual value.

11.2.2 Precision
Precision describes the closeness of repeated individual measures of the same
sample. Figure 11.1 illustrates how accuracy and precision are connected.
Although high accuracy and high precision are the target, a test with high
accuracy but lower precision might still be acceptable. Precision of an accurate
test might be improved by method optimisation. The combination of high
precision with low accuracy is problematic because the bioassay results might

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 227

Figure 11.1 Definition of accuracy and precision.

look really good thanks to the high precision, but the values are systematically off
the target. Low accuracy combined with low precision are easy to identify and those
assays should be avoided.
Precision can be determined at different operating levels, including within-run
precision, between-run precision with the same operator (i.e., repeatability) and
between-run precision with different operators or even laboratories (i.e., intra-
and inter-laboratory reproducibility, respectively). Precision is expressed as the
coefficient of variation for repeatability (CVr) (Equation 11.1) and for
reproducibility (CVR) (Equation 11.2):

CVr = sr /mr (11.1)

CVR = sR /mR (11.2)

Here, μr is the mean and σr is the standard deviation of the validation sample results
repeated on multiple occasions by the same operator. μR is the mean and σR is the
standard deviation of the validation sample results repeated on multiple occasions
by different operators and/or laboratories.
Both of these parameters should be less than 15%–20% for an assay to be deemed
repeatable and reproducible.

11.2.3 Robustness
Robustness characterises the sensitivity of a method to operational variations and
is a measure of how transferrable the method is to other operators and/or

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 228 Bioanalytical Tools in Water Quality Assessment

laboratories. Robustness is usually calculated as the ratio (Equation 11.3) of the CVr
over the CVR:
CVr
Robustness index = (11.3)
CVR
This robustness index should be between 0.8 and 1.2, and preferably close to 1.
A method with low robustness may not yield consistent results between operators
and/or laboratories.

11.2.4 Quality
The Z-factor is a measure of assay quality (Zhang et al., 1999). It is an expression of
the separation band between positive and negative controls (Equation 11.4), and
thus a measure of the effective dynamic range:
3 × (sp + sn )
Z-factor = 1 − (11.4)
|mp − mn |

Here, μp is the mean and σp is the standard deviation of the positive control, and μn is
the mean and σn is the standard deviation of negative controls.
A good assay should have a Z-factor between 0.5 and 1. Any Z-factor less than
0.5 indicates a marginal assay, and a Z-factor less than 0 indicates that there is too
much overlap between the positive and negative controls for the assay to be useful.
The Z-factor is assay-specific and needs to be determined during method
development over several plates and independent repeats or if an existing assay is
established newly in a laboratory. A further QA/QC measure is to check the
Z-factor on every plate to assure that it remains consistently high. It might
deteriorate if the reagents expire, if there are issues with pipetting or if other
errors occur.
For example, the Z-factors of GeneBLAzer assays normally range between 0.7
and 0.9, and the ERα-GeneBLAzer illustrated in Figure 11.2 has a Z-factor of 0.88.

11.2.5 Matrix interference

An assay might work perfectly fine with single chemicals but that does not mean that
it is compatible with complex environmental samples. Most of the assays that are
used in water quality assessment have been developed and initially used to screen
pure chemicals. When adapting a bioassay from a chemical toxicity testing
context to water quality testing one must consider matrix interferences and
characterise how much the analysis is affected by the presence of other
components in a water sample. Although water is typically extracted prior to
dosing in an assay (Chapter 12) dissolved organic carbon (DOC) is partially
co-extracted. The co-extracted DOC has little impact on cell-based assays (Neale
and Escher, 2014), but can interfere with antagonism measurements (Neale and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 229

Figure 11.2 Z-factor (Equation 11.4) illustrated on the example of a concentration–

response curve (CRC) for 17β-estradiol in the ERα-GeneBLAzer assay.

Leusch, 2015) and has a strongly suppressive effect in cell-free assays, such as the
popular acetylcholinesterase inhibition assay (Neale and Escher, 2013).
Solvents (e.g., ethanol, methanol and dimethyl sulphoxide (DMSO)) used during
extraction might also interfere with the assay so great care has to be taken to reduce
the fraction of solvent or remove the solvent by evaporation prior to testing. For
more information on solvent effects see Section 11.3.3.4.

11.2.6 Sensitivity
Sensitivity determines how well an assay responds to varying amounts of
target compound(s) and includes determination of the limit of detection (LOD,
Equation 11.5), the limit of quantification (LOQ, Equation 11.6) and the
calibration curve:

LOD = mn + (3 × sn ) (11.5)

LOQ = mn + (10 × sn ) (11.6)

Here, μn is the average response of the negative control and σn is the standard
deviation of response with the negative control. Typically, the negative controls
are unexposed cells. For bioassays, the ‘calibration curve’ is the concentration–
effect curve for the reference and other relevant compounds.
As discussed in Chapter 7 bioassay results are often expressed in effect
concentrations ECy or ECIRz (e.g., EC10 or ECIR1.5). Therefore, the LOD or LOQ
are typically reported not as effects but as the EC of a reference compound, that
is, ECLOD. The LOD expressed as a concentration is defined as the concentration
of the reference chemical causing three times the standard deviation of the
response of the negative control, which corresponds to Equation (11.7) for
log-logistic concentration–effect curves and Equation (11.8) for linear
concentration–effect curves with % effect and Equation (11.9) with induction

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 230 Bioanalytical Tools in Water Quality Assessment

ratio (IR) as the effect measure:


1 max − min
logECLOD = logEC50 − log −1 (11.7)
slope LOD − min
LOD
ECLOD = (11.8)
slope
LOD − 1
ECIR,LOD = (11.9)
slope
Figure 11.3 illustrates the derivation of the LOQ and LOD for 17β-estradiol (E2)
in the ERα GeneBLAzer assay for activation of the estrogen receptor (ERα). The
LOD is 2.2% and the LOQ is 7.3%. The corresponding ECLOD is 0.9 ngE2/L and
the ECLOQ amounts to 3.1 ngE2/L. It is interesting to note that the LOQ is close
to the 10% effect in this assay, confirming the choice of 10% effect level as
statistically significantly different from the negative controls and the EC10 as an
appropriate effect descriptor.
When it comes to testing water samples, we cannot directly translate the ECLOD
and the ECLOQ into the associated bioanalytical equivalent concentrations (BEQ).
The BEQLOD and BEQLOQ are determined both by the assay’s inherent
sensitivity and variability, but also by the relative enrichment of a particular
sample. An ECLOQ of 3.1 ngE2/L does not mean that estrogenicity can be
detected in water samples only if they have 17β-estradiol equivalent
concentrations (EEQ) .3.1 ng/L. Water samples can be enriched so the BEQLOQ
for a particular water sample can be much lower. For example, a sample with a
relative enrichment factor (REF) of 100 in this assay would have an EEQLOQ of
0.031 ngE2/L. However, at this high REF, cytotoxicity might interfere and mask
the estrogenic effect, which would again decrease the practical detection limit.

Figure 11.3 Derivation of LOD and LOQ using the example of a concentration–
response curve (CRC) of 17β-estradiol in the ERα GeneBLAzer assay. On the left,
the LOQ and LOD effect levels are indicated in the logarithmic CRC and on the
right, the associated ECLOD of 0.9 ngE2/L and ECLOQ of 3.1 ngE2/L are shown in
the corresponding linear CRC. The EC10 is 4.2 ngE2/L.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 231

Figure 11.4 Illustration of the derivation of detection limits in water samples on the
example of a wastewater treatment plant effluent sample tested in the ERα
GeneBLAzer assay (matching the concentration–response curve (CRC) of
17β-estradiol in Figure 11.3). From the cell viability data and IC10 of REF 13 can be
derived. The LOD and LOQ remain the same at the effect level because they are
derived from the negative controls on the same plate as 17β-estradiol in
Figure 11.3 but the ECLOD and ECLOQ are related to relative enrichment factors
(REF) of the water sample and the detection window is between ECLOQ of REF 7.8
and the IC10 of REF 13.

If there is cytotoxicity in a water sample then, as discussed in Chapter 7, all data

points above the cytotoxicity threshold must be removed from the analysis. For
example, the wastewater treatment plant effluent sample in Figure 11.4 exhibits
an IC10 of REF 13, thus all response data at REF .13 should be disregarded.
Using the filtered data, we can calculate the ECLOD at REF 2.3 and ECLOQ at
REF 7.8 using Equations (11.5) and (11.6), respectively, followed by Equation
(11.8). The ECLOQ in turn can be translated to an EEQLOQ of 0.40 ngE2/L using
the ECLOQ of E2 of 3.1 ngE2/L divided by the ECLOQ of the sample in REF
(in this example, EEQLOQ = 3.1 ngE2/Lbioassay/7.8 Lwater/Lbioassay = 0.40 ngE2/
Lwater). The IC10 cut-off is at REF 13, so the detectable window is fairly narrow,
from REF 7.8 to 13. If a sample does not induce a detectable response even at the
highest valid REF tested (i.e., effect ,LOQ even at the highest valid REF tested),
then the limit of activity in the sample is usually calculated as EEQLOQ divided
by the highest valid REF tested – in the example below assuming no effect up to
the IC10 cut-off of REF 13, this would be ‘,0.24 ngE2/L’.
As discussed in Section 11.3.3.5, the extraction blank might pose a problem, and
even if it can be subtracted, it will interfere with the detectability of the assay.
Especially, if the extraction blank is highly cytotoxic, this will mask the activation of
the sample and lower the detectability to a point where no activation can be measured.

11.3 QA//QC IN THE LABORATORY

The concept of QA/QC was initially developed in the manufacturing industry as a set
of documents and procedures to ensure consistent product quality. In the laboratory,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 232 Bioanalytical Tools in Water Quality Assessment

the application of QA/QC principles ensures that bioassay results are accurate and
consistent. This also entails some practical considerations, such as setup, materials
and tracking of materials. As a minimum, the laboratory QA/QC steps should
consist of analysis replication, inclusion of adequate positive and negative control
samples, verification of assay performance with control charts and fixed control
criteria, good record-keeping and having appropriate SOPs in place. A series of
OECD documents details the principles of Good Laboratory Practice (OECD,
1998), including one specifically developed for in vitro studies (OECD, 2004).

11.3.1 Practical considerations

There are some practical considerations to guide how assays can be optimised
during development and validation, and how to ensure that QA/QC criteria are
consistently met. The following provides a short overview of parameters that can
affect the quality of bioassay results (Figure 11.5). More detailed information can
be found in relevant chapters of the Assay Guidance Manual (Auld et al., 2020)
and in a guidance document on reproducibility of in vitro assays (Hirsch and
Schildknecht, 2019).
First, a decision on the plate format and the cell model or assay to be used has to
be made. Note that assays in 96-well format do not always require the availability of
lab automation instrumentation, but these devices become practical for the 384-well
plate and indispensable for 1536-well plate formats. In principle, all materials that
come into contact with the cells and environmental factors can influence the
performance of an assay. Thus, it is worthwhile to plan the optimisation and
validation of an assay carefully.
Some important factors are summarised in Figure 11.5. They include the plate
material, which can have an influence on the quality of the read out measured
with a plate reader or microscope, but also on cell health directly, and in some
instances on the bioassay response itself (e.g., some plasticware contains
xeno-estrogenic compounds, which can interfere with sensitive reporter gene
assays). At the next level, the media used to grow and/or differentiate cells as
well to run a specific assay should be selected with care and adapted if necessary.
Media components can interfere with measurement of fluorescence (e.g., phenol
red) or influence the availability of chemicals to the cells (see also Chapter 9). In
addition, optimised seeding densities and incubation times should be determined
to ensure the best possible quality of measurements and cell health. Besides
these, additional measures such as the use of gas-permeable seals or
preincubation of cells at room temperature after seeding before transfer to an
incubator (Lundholt et al., 2003) can also be considered to reduce edge effects.
It is also recommended to track the lots of plates, media, supplements and other
materials that are used. A lot is a batch of consumables that were manufactured
together and have the same lot number. This practice facilitates troubleshooting.
Moreover, lot tests should be performed before running large batches of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 233

Figure 11.5 Practical considerations for setting up a quality-controlled bioassay. This

flowchart provides a summary of factors that influence the QA/QC of bioassays. In
practice, this is an iterative process that must balance competing factors such as
automation compatibility, cost, availability and performance (adapted from Auld
et al., 2020). ECM = extracellular matrix, QIVIVE = quantitative in vitro to in vivo
extrapolation.

bioassays. Based on the results of the lot tests, all materials tested should be reserved
in sufficient quantities to run a project or sampling campaign with one lot of the
respective materials. Lot tests are also recommended for all materials required
during the sample preparation process.

11.3.2 Replication
Replication of experiments is absolutely critical in any type of analysis, including
bioassay analysis. Several levels of replication are required for comprehensive
QA/QC: within plates, between plates and between runs (Figure 11.6).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 234 Bioanalytical Tools in Water Quality Assessment

Figure 11.6 Three different levels of replication are within plates (intra-plate),
between plates (inter-plate or intra-assay) and between runs (inter-assay).

11.3.2.1 Intra-plate replication

First, the sample is analysed in replicate (usually duplicate or triplicate, depending
on the inherent variability of the assay) side-by-side on the same plate (‘intra-plate
replication’ in Figure 11.6).
A phenomenon called ‘edge effect’ can severely hamper the quality of the
intra-plate replication in biochemical and cell-based assays. The edge effect is
caused by the design of the plate which results in increased evaporation in the
outer wells and also a temperature gradient from the outer to the inner wells.
Some options to deal with the edge effect are listed in Section 11.3.1.
The purpose of this first level replication is to determine the variability between
different wells. Such variability can be caused by a variety of operator factors (e.g.,
erratic liquid handling) or by environmental factors (e.g., humidity). The coefficient
of variation (CV = σ/μ) of these replicate analyses should not exceed a
pre-determined CV threshold (this can vary between different assays but is
generally set at CV , 10%–15%).

11.3.2.2 Inter-plate replication

Intra-assay replication or inter-plate replication (Figure 11.6) involves analysing the
same sample multiple times in the same assay run but at different stages of the run
(e.g., once at the beginning of a plate and once at the end or better yet on a different
plate altogether). Intra-assay replication verifies that there is no temporal drift during
the assay run due to either environmental factors (e.g., an increase in temperature) or
instrumental issues (e.g., a spectrophotometer that loses its sensitivity during the
experiment). During data analysis, the result for each replicate is compared to
ensure that the variability does not exceed a pre-determined level (usually CV ,
10%–15%). The average of the intra-assay replicates is then used, again as a

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 235

single number, as the variability here is still not indicative of the sample variability.
Intra-assay replication is only necessary for a small subset of randomly selected
samples, usually one sample per assay run.

11.3.2.3 Inter-assay replication

Inter-assay replication (Figure 11.6) involves analysing the same sample in
independent assay runs. This verifies that there is no assay drift or bias over time.
Samples should always be analysed on at least two independent runs performed
on different days. During data analysis, the independent values are compared. If
the CV is more than a pre-determined value, which varies between different
assays, but is usually no more than 15%–20%, the analysis is performed a third
time to obtain a more precise result. This way, the concentration–response curves
(CRCs) for all repeats can also be evaluated together. The final result is often
reported as the average of the two closest results. The variability of the results at
this stage is an indication of variability of the bioanalytical method assuming
there is no sample degradation. Although inter-assay variability can be reported
as an indication of the confidence in the analytical result, it is not a measure of
the true variability of the water sample.

11.3.2.4 True sample replicates

In an appropriately designed monitoring programme, truly independent replicate
samples should be taken and analysed. Truly independent replicates are for
example water samples taken in triplicate and independently run through the
entire extraction process (Chapter 12). The results of the independent samples can
be reported as the average + standard deviation as in this case the variability
does provide a measure of precision. Note, however, that this value includes a
variability component from both sample collection and sample processing (i.e.,
preservation and extraction) and analysis.

11.3.3 Quality control samples

In chemical analysis, it is important to service and maintain the physical detectors of
analytical instruments regularly. This ‘servicing and maintenance’ is even more
critical with the biological detectors (i.e., cells) used in bioanalytical tools as
these require carefully controlled environmental and culture conditions. Cell
culture conditions need to be carefully developed for each test system while
keeping in mind good cell culture practices (Bal-Price and Coecke, 2011).
Variations in cell culture or practice can cause abnormal behaviour of the test
system and produce inaccurate results. Always including and analysing control
samples allows detection of such abnormal events and ensures that only data
from valid assay runs are analysed and reported.
Quality control samples are a crucial component of QA/QC for bioanalytical
tools. A full CRC of the reference compound (‘standard curve’), positive and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 236 Bioanalytical Tools in Water Quality Assessment

negative control samples, blanks and inter-assay samples should be included in

every run. These quality measures are described in more detail below. The results
of the quality control samples are compared with control charts (see Section
11.3.4). If any control parameter falls outside the acceptable range, the entire
dataset should be rejected and all affected samples re-analysed.

11.3.3.1 Standard curve

A standard curve is a CRC (see Chapter 7) of the reference compound covering the
full range of effect from 0% to 100%. For cytotoxicity, calculated as 100% minus
the percent cell viability, the minimum effect is by definition 0% (all cells are
viable) and the maximum effect is 100% (all cells are dead). With reporter gene
assays, the effect is in most cases measured as a colour change (absorbance),
relative light units (RLU) or relative fluorescence units (RFU), or by a change of
fluorescence. In this case the minimum and maximum might have different
absolute values but the % effect of the standard curve must be repeatable. Two
typical standard curves for 2,3,7,8-tetrachlorodibenzodioxin (TCDD) in the AhR
CALUX assay from two different plates are presented in Figure 11.7. When the
RLU were plotted, there were differences in the maximum RLU reached in two
plates. However, when the maximum was fitted to the highest response for each
plate and the RLU converted to % activation, both curves then overlapped. Note
that some RLU data had to be removed due to cytotoxicity interferences (grey bar
in Figure 11.7). The same principle applies for assays that are based on RFU
measurements.
A standard curve must be included for every assay run and benchmarked against
previous curves obtained in the same assay. The standard curve in particular yields
four important parameters: the EC50, the slope and the minimum and maximum
effects (see Chapter 7).
The standard curve is also often used to derive the BEQ if the reference
compound for QA/QC is the same as the reference compound for the BEQ
derivation.

Figure 11.7 Typical standard curves obtained for 2,3,7,8-tetrachlorodibenzodioxin

(TCDD) in the AhR CALUX assay.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 237

11.3.3.2 Positive control sample

A positive control sample is a sample that contains a high bioactive test compound
other than the reference compound. The positive control should preferably
have different physico-chemical properties to the reference compound. For
estrogenicity, for example, a good reference compound (standard) would be the
natural hormone 17β-estradiol and a good positive control could be the industrial
compound 4-nonylphenol.

11.3.3.3 Negative control sample

There are two types of negative control samples: solvent and media negative
controls. A solvent negative control is a blank sample that contains an equivalent
amount of solvent as the actual water sample extracts, which are usually in
ethanol, methanol or DMSO. Even if a solvent extract is blown down prior to
dissolving in medium, the negative control should mimic that and should be the
equivalent solvent and the same volume blown down and treated exactly
the same way as the samples. The negative control provides verification that the
solvent itself is not responsible for any of the observed effect.
A medium negative control measures the minimal response in the assay media
without any additional influences. Often, the medium negative control is also
needed to calculate the response from the raw signals, for example for the
autofluorescence correction for fluorescent samples in the β-lactamase assay with
a FRET (Förster resonance energy transfer) reagent.
In reality, the response with the solvent and media negative controls should be
comparable, otherwise it is an indication that the solvent used to reconstitute the
water extract is interfering with the assay.

11.3.3.4 Solvent effects

It is critical to understand the effect of solvents on any bioassay. If solvent
interference is observed, it may be necessary to reduce the amount of solvent
used in the assay or use a different solvent that does not affect the assay. For
testing of water samples, it is recommended to use volatile solvents that can be
completely removed by evaporation prior to addition of assay media. It is a
common misconception that this process will also remove chemicals provided
that blow down is gentle, that is, by a stream of nitrogen gas and not by vacuum.
We need to keep in mind that volatile chemicals would have been lost already
during the solvent blow-down stage of liquid–liquid or solid-phase extraction
(SPE). In addition, in vitro assays are run on well plates that are not hermetically
sealed and volatile chemicals (or more precisely chemicals with an air–medium
partition constant over 10−4) will be partially lost during the 24–48 h incubation
at 37°C of a bioassay experiment (Escher et al., 2019). The vial weight of
extracts using volatile solvents has to be continuously tracked to assure that no
losses are occurring or, if they are, that the solvent is topped up.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 238 Bioanalytical Tools in Water Quality Assessment

Although blow down is recommended for water samples, it might not be

applicable for other sample types and pure chemicals and needs to be assessed for
each sample type prior to deciding on the sample treatment. For hydrophobic
chemicals and extracts from condensed phases such as sediment and tissue,
DMSO remains a solvent of choice. Because of its low volatility, it is not
necessary to track the vial weight of DMSO extracts as losses from evaporation
are negligible. DMSO extracts have a high stability, provided they are kept in the
dark because DMSO is subject to oxidation and photodegradation. Some of the
DMSO photoproducts are toxic and can produce blank effects. Therefore, DMSO
should only be used fresh from ampoules, kept in the dark and not stored in
larger solvent bottles on shelves.
The IC10 of DMSO depends on the cell type and ranges between 0.5% and 3% of
DMSO for common reporter gene assays used in water quality assessment
(Table 11.1). No corresponding list is available for other solvents as they partially
evaporate during the 24-h incubation at 37°C and no robust IC10 can be derived.
DMSO can, however, be problematic in some assays, even at concentration
below cytotoxic concentrations. DMSO can reduce the metabolic activity of cells
(Ferk and Daris, 2018), stimulate anti-inflammatory activity (Costa et al., 2017),
activate the oxidative stress response (Escher et al., 2012) and interfere with the
assessment of immunomodulatory effects (Timm et al., 2013). DMSO can also
interfere with gene expression at concentrations as low as 0.5% (Sumida et al.,
2011), and thus could subtly affect the results of gene expression assays at
noncytotoxic concentrations (Leusch et al., 2017). As a matter of fact, even at
concentrations where DMSO is typically used as vehicle (0.1%–1.5%)
alteration in proteins and DNA, including alteration of DNA topology, have been
observed and there was an antioxidant effect as low as 0.5% (Tuncer et al.,
2018). At 0.1% DMSO, .2000 genes were differentially expressed in cardiac

Table 11.1 Cytotoxicity of DMSO, quantified by 24 h growth/cell viability via

microscopic imaging as IC10 + standard error of mean, derived by error propagation
(Escher et al., 2019).

Cell line IC10 (mM) IC10 (%)

AREc32 92 + 4 0.7 + 0.1
AhR CALUX 149 + 19 1.1 + 0.1
PPARγ-BLA 303 + 50 2.2 + 0.4
AR-BLA 430 + 159 3.1 + 1.1
ERα-BLA 172 + 14 1.2 + 0.1
PR-BLA 128 + 8 0.9 + 0.1
GR-BLA 63 + 2 0.5 + 0.1
ARE-BLA 553 + 112 3.9 + 0.8

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 239

and hepatic microtissue and methylation pattern indicated epigenetic changes

(Verheijen et al., 2019).

11.3.3.5 Field and laboratory blanks

Blank samples are important to confirm that sampling and extraction are not
responsible for any of the observed activity. Although this is usually investigated
during method validation, it is important to include blanks nevertheless in case an
unknown parameter changes that may affect the assay results (e.g., a change in
the manufacturing process that introduces a chemical in a consumable that can
leach into the sample and cause accidental contamination during sampling and
extraction).
There are two types of blanks: field and laboratory blanks. A field blank is an
ultrapure water sample that is taken to the field and exposed to similar conditions
as the actual samples (e.g., temperature variation during travel) and is extracted,
concentrated and analysed at the same time and in the same way as the actual
samples. A laboratory blank is similar, except that it has not been taken into the field.
For water samples, blanks are expected to be fairly clean and not cause any
effects or cytotoxicity .10% up to an REF of 100. To achieve this, great care
has to be taken in the extraction process (Chapter 12), high purity solvents have
to be used and SPE material needs to be cleaned and conditioned prior to use. In
practice, only very clean water samples such as drinking water are enriched to
such high REFs, and for more contaminated water it will be sufficient that there
are no effects or cytotoxicity .10% up to the highest tested REF of the samples.
Blanks are often more problematic for extracts from condensed phases such as
accelerated solvent extracts from soil and sediment or biota. In this case, it might
be necessary to apply a blank subtraction for data evaluation. If necessary, this
should be performed at the level of effect units (EU = 1/EC, Equation 11.10) or
toxic units (TU = 1/IC, Equation 11.11) and a subtraction should only be
performed if the blank has an EUblank or TUblank that is less than 50% of the
EUsample or TUsample (Jahnke et al., 2018):

EUblank−corrected = EUsample − EUblank (11.10)

TUblank−corrected = TUsample − TUblank (11.11)

11.3.3.6 Matrix spike recovery

Sample preparation is discussed in detail in Chapter 12 but a relevant routine
QA/QC measure for sample preparation is matrix spike recovery (Denison et al.,
2020). A positive control that is also representative for the physico-chemical
properties of the targeted chemicals is spiked in a defined concentration to the
water sample and undergoes all extraction steps. The matrix spike recovery is

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 240 Bioanalytical Tools in Water Quality Assessment

calculated using Equation (11.12):

EUwater+spike − EUwater
Matrix spike recovery = (11.12)
EUspike

11.3.3.7 Inter-assay sample

An inter-assay sample is a previously analysed positive sample from a similar water
matrix as the one under investigation (e.g., wastewater and drinking water).
Analysis and quantification of the sample and comparison with previously
obtained results confirms that the quantification step is reliable and repeatable
and shows that the sample would be assigned the same BEQ (Chapter 7)
irrespective of when or by whom it was tested.

11.3.4 Control charts and fixed control criteria

11.3.4.1 Control charts
Control charts (also called Shewart charts) are a useful QA/QC tool to benchmark
the results of a bioassay run against those obtained from previous runs. Control
charts also help determine if there is a gradual shift in assay performance.
In a control chart, the value of a control parameter (e.g., EC50 and slope) is plotted
against time (or run number) (Figure 11.8a). The mean value of all previous runs as
well as warning and control (or action) limits are all indicated on the graph. The
control limits are set as three standard deviations away from the mean. Any value
that falls outside these control limits indicates abnormal bioassay behaviour in
which case the entire bioassay dataset should be discarded and the affected
samples re-tested. The warning limits are set as two standard deviations away
from the mean. Any value that falls between the warning and control limits
indicates a possible concern about the quality of the data that should be
investigated. The whole dataset and other control parameters should then be
checked to determine its validity.
Figure 11.8b shows an example of the log EC50 of 17β-estradiol in ERα
GeneBLAzer over three years. In 2018/2019, the data varied with larger error
bars per run and also larger variation between runs, with one plate even falling
below the lower warning limit (but still within control range). After optimisation
of the pipetting method in 2019, the EC50 had lower error bars and also varied
much less between the runs performed on different dates. A similar development
was seen for ECIR1.5 of t-butylhydroquinone (tBHQ) in AREc32 (Figure 11.8c).
tBHQ hydrolyses quickly, so the experiment is time-critical and methanol stocks
cannot be kept for a long time. Overall, the quality was good but rare individual
outliers can be easily identified using these plots.
If a consistent shift of the EC50 or ECIR1.5 can be observed between plates and
over more than five rounds of experiments performed on different dates, this may

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 241

Figure 11.8 (a) Makeup of a control chart showing the mean value, upper and lower
warning limits (mean + 2σ) and upper and lower control limits (mean + 3σ);
(b) example control chart for the log EC50 of 17β-estradiol in ERα GeneBLAzer and
(c) example control chart for the ECIR1.5 of t-butylhydroquinone in AREc32.

indicate a gradual shift in assay performance. This shift may be due to a variety of
factors but the most common are cell passage number, instrument age (e.g., light
bulb on a plate reader) and degrading reagents. The reason for the shift should be
investigated as it may be an early warning sign of a potentially larger problem.

11.3.4.2 Fixed control criteria

For some control parameters (typically minimum, maximum and induction ratio), it
is more appropriate to determine a minimum or maximum value, called a ‘fixed
control criteria’, rather than a range. The assay run is valid only if the control
parameter meets the fixed control criteria. Some parameters are universal: for
example, a minimum Z-factor of 0.5. Others depend on the assay: a minimum
induction of 6, for example, may be a valid control criterion for luciferase activity
measured in a reporter gene assay and an induction of 2 may be the minimum in
a cell proliferation assay.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 242 Bioanalytical Tools in Water Quality Assessment

The exact mix of control charts and fixed control criteria will depend on the assay
type. What is important is to determine acceptable values and ranges prior to
analysis and to use benchmarks to validate (or reject) a bioassay run.

11.3.5 Standardisation and documentation

Accountability is an important aspect of all laboratory work including of course
water quality testing. Standardisation of methodology will contribute to QA/QC
by limiting the likelihood of error and includes implementation of SOPs. The
SOP must include a detailed description of the experimental protocols, its field of
application, equipment operation and standardised data analysis methods (e.g., a
preformatted Excel template with the appropriate formulae that simply requires a
cut-and-paste of the raw data or an automated script like the ToxCast Analysis
Pipeline (tcpl) coded in R (github.com/USEPA/CompTox-ToxCast-tcpl)).
Good record-keeping based on the FAIR principles – Findability, Accessibility,
Interoperability, Reusability (Wilkinson et al., 2016) – will ensure that all necessary
information is available. Such records should include documents such as chain of
custody and field observation forms for sampling, information on sample origin
and manipulation in centralised sample tracking databases, well-maintained
laboratory books with details of experimental manipulations (e.g., cell passage
number and operator), and archives of raw and analysed data in a safe location.

11.3.6 Guidelines
There are a number of test guidelines available for in vitro assays and water
quality monitoring. The series of guidelines for water quality testing from the
International Organization for Standardization (ISO) includes mutagenicity
testing with the Ames test (ISO 11350, 2012), bioluminescence inhibition of
Aliivibrio fischeri (ISO11348-3, 1998), algal toxicity (ISO8692, 2004) and the
fish embryo toxicity test (ISO15088, 2007). All these guidelines apply direct
water and wastewater testing, and mere sample filtration is recommended for
sample pre-treatment. Here, the benchmark concentration is the lowest ineffective
dilution, which is the highest dilution of the original water sample that is below
the LOQ (Equation 11.5).
The newest addition to this series are reporter gene assays for estrogenicity
using two types of the yeast estrogen screen (YES, ISO/DIS19040-1, 2017;
ISO/DIS19040-2, 2017) and a human cell-based reporter gene assay
(ISO/DIS19040-3, 2017). In these guidelines, a choice is given between testing
original water samples and testing SPE extracts in DMSO.
There are also OECD technical guidelines for testing of chemicals. The
‘Performance-Based Test Guideline (PBTG)’ 455 (OECD, 2015) describes the
methodology of ‘stably transfected transactivation in vitro assays to detect
estrogen receptor agonists and antagonists (ER TA assays)’. These OECD
technical guidelines are intended for chemical testing, but their detailed quality

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Quality assurance and quality control (QA/QC) 243

control procedures and assay descriptions may serve as inspiration for water quality
testing applications.

11.3.7 High-throughput screening

Automation of cellular and other plate-based assays in high-throughput screening
(HTS) setups is particularly important for drug development. With the Tox21 and
ToxCast initiatives of the U.S. National Institute of Health and the U.S.
Environmental Protection Agency, HTS has also found its way into chemical risk
assessment (Chapter 9). 384- and even 1536-well plates and low-volume
pipetting platforms are common in HTS. At least partial automation is now also
emerging in water quality assessment. Automation poses specific challenges for
QA/QC and while a detailed coverage is beyond the scope of this chapter, we
refer the reader to a book chapter by Powell et al. (2016) for more details.

11.4 CONCLUSIONS
Implementation of QA/QC protocols increases the amount of time and consumables
needed to conduct bioassay testing, but the benefits of confidence in the accuracy,
reliability and consistency of the data are well worth the effort.
A good bioassay for water quality testing is one that is accurate and precise, both
within a run, but also between runs (repeatable) and among different operators and
laboratories (reproducible), robust, sensitive and is not affected by matrix
interference. For reporter gene assays, the window between the specific effect and
effects on cell viability should be sufficiently large to ensure that specific effects
are not masked by cytotoxicity.
In the Supplementary Information to this book on www.ufz.de/bioanalytical-
tools, we provide additional resources for QA/QC including example
spreadsheets for the approaches discussed in this chapter.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 12
Sampling, sample preparation
and dosing

12.1 INTRODUCTION
To ensure that bioassay results are meaningful, it is important to select an
appropriate sampling strategy and use suitable sample pre-treatment and
processing methods. This is because organic micropollutants are often present at
low concentrations in the range of nanogram per litre, particularly in drinking
water and clean source waters (Glassmeyer et al., 2017; Troger et al., 2018), so
water samples may need to be enriched up to 100 times in the bioassay before an
effect can be detected.
Although issues of sample preparation are not specific to bioanalytical tools,
good sample preparation is essential for bioanalysis in the same way as it is for
chemical analysis. Even the best bioassay cannot rectify problems created by
poor sample collection and preparation. High effects and cytotoxicity in the
blanks may jeopardise entire monitoring projects. Liquid–liquid extraction (LLE)
and solid-phase extraction (SPE) are currently the most common methods applied
for preparation of water samples for subsequent bioanalytical measurement.
Condensed phases such as sediments or biota samples are also typically extracted
with LLE but require additional clean-up of the extracts. These non-aqueous
sample types will be discussed in Chapter 15, while we focus in this chapter on
water samples.
It is necessary to validate the chosen extraction method using both chemical and
bioassay analysis before any serious monitoring is undertaken. Such validation is
necessary in order to establish that the extraction procedure provides good and

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0245

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 246 Bioanalytical Tools in Water Quality Assessment

Figure 12.1 From environment to laboratory and into the well-plate: overview of
sampling, sample preparation and dosing.

stable recoveries for a range of environmentally relevant analytes known to induce a

biological response in the assay. This prior validation is particularly critical because,
unlike chemical analysis where labelled standards can be added to calculate the
recovery efficiency of each sample, surrogate standards are generally not added
to water samples for bioassay analysis, which is unable to differentiate between a
spiked compound and a compound present in the original sample. Other quality
assurance/quality control (QA/QC) documentation (such as certificate of analysis
of the solvent and/or sorbent, chain of custody and field observation forms) and
procedures (such as a field blank and positive control) become very important to
compensate for the lack of a spiked standard in every sample. See also Chapter
11 for QA/QC of bioassays.
This chapter provides an overview of sampling strategies and sample
pre-treatment and extraction options (Figure 12.1) and includes a decision-
making flow chart to help users select appropriate sampling/extraction methods
for aqueous samples. Finally, independent of the sample type and sample
treatment, dosing into the bioassay is also an important practical consideration
(Figure 12.1).
Various case studies that describe several different sampling scenarios across the
entire water cycle in combination with diverse bioassay test batteries are described
in more detail in Chapter 14.

12.2 WATER SAMPLING STRATEGIES

The sampling strategy depends on the purpose, objectives and sample context
(Figure 12.2). If the purpose of a sampling campaign is to assess the product
quality of a water treatment plant, with the objective of comparing the effect in
the final water to an effect-based trigger value (Chapter 13), then only the
product water needs to be collected. In contrast, both source water and product
water are required if the purpose is to evaluate treatment process efficacy.
Samples can also be collected after intermediate steps throughout the treatment
train, such as after advanced oxidation or disinfection if the purpose of the
sampling campaign is to identify and understand critical processes.
Composite samples are recommended for wastewater to capture the diurnal
variation observed for some micropollutants (Nelson et al., 2011; Petrie et al.,
2017), with many studies collecting 24 h composite influent and effluent samples
(Körner et al., 2001; Bicchi et al., 2009; Macova et al., 2010; Jalova et al., 2013;
Bain et al., 2014; Roberts et al., 2015) and even some seven consecutive 24h

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Sampling, sample preparation and dosing 247

Figure 12.2 Examples of different sampling campaign purposes and objectives for
drinking water, wastewater, water reuse and surface water with the required
samples for each purpose indicated. WWTP = wastewater treatment plant; EBT =
effect-based trigger (Chapter 13).

samples to evaluate potential weekly variations of wastewater quality (Neale et al.,

2020d). Grab sampling is suitable for collecting drinking water or recycled water
samples if little difference in quality over time is demonstrated.
If temporal dynamics are to be expected, autosamplers allow subsequent
sampling of larger number of individual samples to describe diurnal changes in
water quality (Mueller et al., 2020), which will be important if photodegradation
plays a role or if very dynamic storm events need to be captured (Mueller et al.,
2021). Event-driven sampling can be accomplished by setting the autosampler off
with a trigger, for example, if the water levels rise by a certain value. In this way,
more than 120 samples were collected triggered by rain events in 44 small

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 248 Bioanalytical Tools in Water Quality Assessment

streams in one summer and they demonstrated a diversity of pollution patterns

(Neale et al., 2020a).
Water samples should be collected in solvent washed amber glass bottles, with
the bottles stored on ice and in the dark until returning the samples to the
laboratory for further processing.
The required volume of water to be collected will depend on the expected level of
chemical contamination and thus the need for concentration, with less sample
required for wastewater influent and a larger volume needed for drinking water.
Using the example of SPE, previous experience from the literature suggests that a
common 200 mg/6 cc SPE cartridge can enrich 0.5 L of wastewater influent, 1 L
of wastewater effluent or surface water and 2 L of drinking water, recycled water
or clean surface water. Double the volume can be applied to larger SPE sorbent
masses (e.g., 500 mg cartridges). Assuming the volumes above applied to a 200
mg SPE cartridge and a final extract volume of 0.5 mL, this equates to an
enrichment factor (EF) of 1000 for wastewater influent, 2000 for wastewater
effluent or surface water and 4000 for drinking water, recycled water or clean
surface water. Assuming a bioassay dosing factor (DF) of 0.01 (1% solvent) to
0.001 (0.1% solvent), this results in a maximum relative extraction factor (REF)
of 1–10 for wastewater influent, 2–20 for wastewater effluent or surface water
and 4–40 for drinking water, recycled water or clean surface water. It is worth
noting that some studies have enriched up to REF 500 (König et al., 2017;
Hebert et al., 2018) but in these cases the solvent was blown down prior to
dosing into the bioassays (Section 12.7).
The number of assays that can be run from a single extract will depend on the
number of repeats planned and how much extract is dosed, but typically only
small extract volumes (e.g., a few microlitres) are required for 96 and 384-well
plate assays so a number of different assays can often be run.
If higher REFs were required, samples can be extracted on multiple SPE
cartridges and combined into one final extract (Escher et al., 2014; Neale et al.,
2017c). An alternative approach is to use large volume SPE (LVSPE), which has
been applied to enrich between 6 (influent) to 500 L (surface water) (Neale et al.,
2015b; Välitalo et al., 2017). LVSPE has been applied in a smaller number of
studies than SPE and primarily to surface water (König et al., 2017; Toušová
et al., 2017). LVSPE allows onsite sampling, but also requires more equipment
than other sample extraction methods.
Simon et al. (2019) explored the impact of sample volume on effect recovery,
with 0.5 and 2 L of wastewater effluent and 1 and 4 L of surface water spiked
with a mixture of four estrogenic compounds and 11 pesticides and enriched with
SPE using 300 mg LiChrolut EN/RP-18 sorbent. The extracts were analysed in
assays indicative of estrogenic activity (ERα CALUX), photosystem II inhibition
and algal growth (combined algae test) and bacterial toxicity using Aliivibrio
fischeri. The average activity in the large volume extracts was between 79% and
104% of the activity in the lower volume extracts, showing that sample volume

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Sampling, sample preparation and dosing 249

did not have a significant impact on recovery. Therefore, the sample volumes should
not have a large impact on recovery, though it should be noted that some chemicals,
such as highly polar chemicals, may not be well recovered by conventional SPE.
This is discussed further in Section 12.5.1.

12.3 SAMPLE PRE-TREATMENT OPTIONS

Once an appropriate water sample has been collected, there are a several decisions to
be made regarding sample pre-treatment, including sample storage time and
sample filtration.

12.3.1 Water sample preservation and storage

If not enriching onsite, the pH of a water sample is commonly adjusted to an acidic
pH prior to sample enrichment. Microbial activity can potentially cause
micropollutant biodegradation or biotransformation and adjusting to an acidic pH
can reduce microbial activity in the sample. Reducing the pH can also improve
the extraction of weak acids, with improved recovery of a pharmaceutical
cocktail in a bioluminescence inhibition assay at pH 3 compared to pH 7
observed for three different SPE sorbents (Escher et al., 2005). In contrast, Šauer
et al. (2018) found no difference in androgenic or anti-androgenic activity in
wastewater influent in duplicate samples at pH 3 and pH 7.4, respectively, using
C18 SPE discs. If the pH of the sampled water had been adjusted to pH 3 or
lower, a few millilitres of ultrapure water should be passed over the SPE
cartridge after passing the sample to avoid that the elution extract becomes acidic,
which could disturb the bioassay.
In addition to pH adjustment, a small number of studies have used 10 mg/L
copper(II) sulphate pentahydrate (Conley et al., 2017a) or 1 g/L sodium azide
(Mehinto et al., 2015) for sample preservation.
In the case of chlorinated samples, the chlorine residual should be quenched
immediately after sampling. Although chlorine will not be extracted by SPE,
quenching is important to prevent the formation of additional disinfection
by-products (DBPs) and to prevent the chlorine from potentially reacting with the
SPE sorbent. Many studies used sodium thiosulphate to quench the chlorine
residual (Macova et al., 2011; Escher et al., 2014; Neale et al., 2020b), and
others used ascorbic acid (Conley et al., 2017b). Typically, 3.5 mg/L of sodium
thiosulphate quenches 1 mg/L free Cl2 and 5 mg/L ascorbic acid quenches
1 mg/L free Cl2 (Farre et al., 2013). Hebert et al. (2018) found no difference in
effects of glass-bottled Evian water controls with and without 20 mg/L sodium
thiosulphate in assays indicative of the oxidative stress response, p53 response
and NF-κB response. Similarly, sodium thiosulphate controls did not have any
effects in assays indicative of hormone receptor-mediated effects (Neale et al.,
2020b). As we are not aware of any studies that have assessed any potential
effects from quenching with ascorbic acid, we recommend using sodium

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 250 Bioanalytical Tools in Water Quality Assessment

thiosulphate for quenching as it has been shown not to induce a response in a number
of bioassays.
After collection, water samples are commonly stored for no longer than 48 h at
4°C before extraction (Aerni et al., 2004; Cargouet et al., 2004; Fang et al., 2012;
Daniels et al., 2018). Alternatively, some studies froze water samples to store for a
longer period prior to extraction (Könemann et al., 2018). Jarošová et al. (2014b)
investigated the impact of sample storage time prior to extraction on estrogenic
activity in wastewater effluent, with matching samples stored at 4°C and
extracted 48 h and 45 days after collection. Of the seven samples, the estrogenic
activity at least doubled in two of the samples, but overall, the difference in effect
was small (e.g., 0.7 ng/L after 48 h and 1.7 ng/L after 45 days).

12.3.2 Water sample filtration

Water samples are commonly filtered prior to SPE, with glass fibre filters most
frequently used. Glass fibre filters have previously shown to sorb only negligible
amounts of estrogens compared to cellulose acetate and nylon filters, with the
latter found to adsorb a significant fraction of estrogens from solution (Walker
and Watson, 2010). The filter pore size (i.e., the particle size retained with 98%
efficacy) used in the literature varies widely from 0.1 to 11 µm, with the majority
of studies using filters with a pore size of 0.7–2 µm (Figure 12.3).
The European Union Water Framework Directive (WFD) recommends sample
extraction methods that capture the whole water sample, which includes both the
dissolved and suspended particulate phases (European Commission, 2009). The
suspended particulate matter (SPM) captured by sample filtration is often
discarded, but several studies have shown considerable biological activity

Figure 12.3 Overview of different filter pore sizes in micrometres (µm) used for
sample filtration prior to solid-phase extraction (studies published from 2001 to
January 2020).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Sampling, sample preparation and dosing 251

associated with particulate matter (Legler et al., 2003; Hamers et al., 2015; Schulze
et al., 2015; Mueller et al., 2021). For example, Dagnino et al. (2010) evaluated the
estrogenic and aryl hydrocarbon receptor (AhR) activity in both the dissolved and
particulate phases from three wastewater treatment plants, with both phases
contributing to the estrogenic and AhR load discharged from the plants. A higher
fraction of AhR activity was found on SPM (Dagnino et al., 2010). Similarly,
Wölz et al. (2008) found that SPM from the Neckar River induced AhR-mediated
activity in the ethoxyresorufin-O-deethylase (EROD) assay while the
corresponding extracted water samples had no effect. These studies suggest that it
is important to consider SPM to gain a better understanding of the bioactivity, but
the decision to collect SPM will depend on the objectives of the sampling
campaign and the studied endpoints, with SPM likely to be important for
non-specific toxicity, activation of AhR and binding to the peroxisome
proliferator-activated receptor. During a heavy storm event in a small river, the
effect load coming from SPM was as high as or higher than in the aqueous phase
despite the SPM being present in mass (SPM) to water volume ratios lower than
0.001 during the rain event (Mueller et al., 2021).
If intending to capture the entire water sample, the options are to use LLE, not
filter before SPE, or to filter prior to SPE and extract the captured SPM
separately using solvents (Ademollo et al., 2012). Some studies only filtered
samples that were expected to block the SPE cartridge, such as wastewater
influent or river water, due to the high particle content (Gehrmann et al., 2018;
Xiao et al., 2016). Similarly, many studies evaluating drinking water did not
apply a filtration step (Van Zijl et al., 2017; Hebert et al., 2018; Valcarcel et al.,
2018). Könemann et al. (2018) found no significant difference in estrogenicity
for filtered and unfiltered surface water samples.
To provide guidance on whether to filter or not, U.S. EPA Method 1694
‘Pharmaceuticals and Personal Care Products in Water, Soil, Sediment, and
Biosolids by HPLC/MS/MS’ recommends that aqueous samples containing
visible particles should be filtered prior to SPE (U.S. EPA, 2007). Water samples
with a turbidity of 5 nephelometric turbidity units (NTU) will visually appear
slightly milky or cloudy, while crystal clear water usually has turbidity less than
1 NTU, with the turbidity only detected by instrumental analysis (NHMRC,
2011). Consequently, water samples with a turbidity of 5 NTU or greater should
be filtered prior to SPE. In general, drinking water is typically ,1 NTU, recycled
water can range from ,1 to 2 NTU depending on treatment processes and
secondary effluent is generally ,2 NTU but can increase when sludge is poorly
settled. The turbidity of river water can vary greatly, while the turbidity of lakes
tends to be more stable. As an example, the turbidity of treated drinking water
from Paris ranged from 0.02 to 0.04 NTU, while the water feeding these plants
ranged from 2 to 14.3 NTU (Neale et al., 2020b). For comparison, the turbidity
of Canadian surface waters ranged from 0.5 to 50 NTU (Cantwell and Hofmann,
2011).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 252 Bioanalytical Tools in Water Quality Assessment

Based on low sorption capacity, glass fibre filters are recommended for filtration
of samples with a turbidity of 5 NTU or greater. Although a wide range of filter pore
sizes are used within the literature, glass fibre filters between 0.7 and 1.5 µm were
recommended for filtration prior to SPE for chemical analysis (ISO11369, 1997;
U.S. EPA, 2007; Furlong et al., 2008).

12.4 EXTRACTION OF WATER SAMPLES

12.4.1 Extraction versus testing the entire water sample
In vitro assays typically target complex mixtures of organic micropollutants but not
inorganics and metals, which can be comprehensively analysed using chemical
methods. Therefore, extraction methods also serve to separate the organic
micropollutants from the matrix, inorganics and metals in a water sample
(Figure 12.4). This differs from whole effluent toxicity (WET), which considers
the whole water sample and stress from organic and inorganic chemicals as well
as ionic strength, DOC and pH (see Chapter 3).
Few studies have run unenriched or native water samples in yeast reporter gene
and mammalian reporter gene assays (Niss et al., 2018; Abbas et al., 2019;
Brettschneider et al., 2019). This is equivalent to WET testing and would
incorporate the effect from different components in water including salts, metals
and other inorganics, as well as organic micropollutants. Consequently, the effect
of organic micropollutants could not be differentiated from other components in

Figure 12.4 Enrichment of samples by solid-phase extraction (SPE), liquid-liquid

extraction (LLE) or passive sampling selects organic micropollutants and removes
matrix components, such as metals and inorganics.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Sampling, sample preparation and dosing 253

water. This approach was used for wastewater samples but is unlikely to be able to
detect an effect in cleaner water samples, such as highly treated wastewater, drinking
water and surface water, which usually need to be concentrated to detect effects. As
a minimum pre-treatment for testing native water samples, water must be filtered as
described above to remove any particles. For mammalian cell-based bioassays, it is
also important to sterile filter the sample using a filter with a 0.22 µm pore size (Niss
et al., 2018).

12.4.2 Solid-phase extraction

SPE dates back to the 1970s and relies on passing a water sample through a column
packed with a sorbent that will retain the analytes as the water passes through. SPE is
the most commonly used method to enrich micropollutants from water prior to
bioanalysis. SPE cartridges or discs contain sorbents that retain organic
micropollutants, while other components present in water, such as metals, salts
and other inorganics, pass through the cartridge or disc, thus simplifying the
matrix (Poole, 2003). After drying the cartridges to remove any residual water,
the sorbed analytes can then be eluted with solvents, creating a concentrated
extract that can be run in the bioassays. SPE has a number of advantages
including good recovery of a wide range of contaminants and ability to be
automated, although cartridges can clog with samples with a high particulate
content as discussed in Section 12.3.2.
Reversed-phase and ion-exchange columns are normally used for aqueous
applications. A variety of different sorbents have been fashioned for water
samples such as silica-based (e.g., C2, C8 and C18), hydrophilic lipophilic
balanced (HLB) co-polymers and graphitised carbon black (GCB). These
sorbents are specific for different classes of compounds, allowing selective
extraction of analytes from water for chemical analysis. For bioanalytical
measurement, sample preparation should be as non-selective as possible because
otherwise only a portion of the chemical contaminants in the water would be
retained and the mixture composition changed for detection in the bioassay.
Co-polymer cartridges such as HLB are therefore widely used in bioanalytical
monitoring due to their wide retention spectrum. SPE is commonly used in
bioanalysis because it can be easily automated, does not produce emulsions, uses
less solvent and usually produces better recoveries than LLE.
The main drawbacks of SPE are that the cartridges can become clogged with
samples rich in organic matter (Ademollo et al., 2012). It also has to be
considered that the sorbent has a saturable capacity, which corresponds to 1%–
5% of the sorbent mass, that is, a 200 mg sorbent bed can retain 2–10 mg of
analyte and can become saturated with highly polluted samples causing a
breakthrough of un-retained micropollutants.
Given the wide use of SPE, Section 12.5 will provide more information about
SPE sorbents and extraction procedures.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 254 Bioanalytical Tools in Water Quality Assessment

12.4.3 Passive sampling

Passive samplers collect micropollutants from the water environment over a longer
period of time and allows chemical and bioassay analyses of very low
concentrations of chemicals. To date, passive sampling has primarily been
applied to surface water (van der Oost et al., 2017a; Toušová et al., 2019).
However, uncertainties regarding the volume of water sampled need to be
considered and parameters such as temperature and flow velocity can affect the
uptake of chemicals into the sampler (Novák et al., 2018). Furthermore, the
composition of the chemical mixture taken up into the sampler may differ from
the chemical mixture in the water as different chemicals will have different
uptake rates into the passive sampler. A number of different types of passive
samplers have been applied in the literature, including silicone rubber, Empore
discs and polar organic chemical integrative samplers (POCIS), in order to target
chemicals with a wide range of hydrophobicity. For example, increased
biological activity was found in Empore disc extracts compared to silicone rubber
(Novák et al., 2018), and Hamers et al. (2018) observed greater effect in
Speedisk passive samples, which contain styrene divinylbenzene sorbent,
compared to silicone rubber. Based on the chemicals extracted by different
passive samplers, de Baat et al. (2019a) applied non-polar silicone rubber extracts
to assays indicative of activation of AhR, oxidative stress response and pregnane
X receptor (PXR), while polar POCIS extracts were tested in assays indicative of
hormone receptor-mediated effects.

12.4.4 Liquid–liquid extraction

LLE has been used since the early nineteenth century and relies on the partitioning
of the analytes from the water into a solvent that is immiscible with water, for
example, dichloromethane, hexane, ethyl acetate or methyl t-butyl ether (MTBE).
After vigorous shaking, the solvent and the water are allowed to separate into two
clearly distinct layers. The solvent layer is collected, and the procedure is
repeated twice with the water sample and fresh solvent to achieve the best
possible recovery of the analyte from the water. The three solvent batches are
combined and concentrated, usually by evaporation of the solvent. The advantage
of LLE is that it is relatively simple to perform and it extracts everything that is
in water, also SPM. On the downside, however, LLE requires large quantities of
solvents, can be time consuming and emulsions can occur between the two liquid
phases, impacting the reliability of the extraction process.
Only a few studies have applied LLE for sample enrichment prior to dosing into
in vitro bioassays (van der Linden et al., 2008; Brand et al., 2013). In these studies,
300 mL of solvent were required per litre of water, which is around 7.5–10 times
more than required for SPE (6 cc/500 mg cartridge). Therefore, great care has to
be taken and high purity solvents have to be used to assure that there are no
blank effects in the bioassays.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Sampling, sample preparation and dosing 255

Due to the high solvent use, time consuming nature and lack of case studies, LLE
is not recommended for sample enrichment of water samples, unless the focus is on
total extraction of water plus SPM to capture also particle-bound water pollutants.
Note that this can also be achieved with SPE by extracting the SPM retained on
the filter. The separate analysis also has the advantage that we can differentiate
between the aqueous and sorbed fraction of chemicals in a whole water sample.

12.4.5 Capturing volatile chemicals

Any solvent extraction, passive sampling and conventional SPE sample processing
involves a blow down step, meaning that volatile chemicals, such as some DBPs
(e.g., trihalomethanes), will not be retained in the final extract. Furthermore,
mammalian cell-based assays are incubated at 37°C for often 16–24 h, also
potentially resulting in the loss of volatile chemicals, although some bioassays
can be adapted to be run without a headspace to prevent the loss of volatile
chemicals (Stalter et al., 2013). Stalter et al. (2016c) developed a purge and
cold-trap method to capture and concentrate volatile DBPs from drinking water.
However, the method is tedious and requires extraction onsite or within a very
short period of time. Therefore, it is not recommended for routine monitoring, but
instead for research purposes. Importantly, volatile DBPs appeared to only have a
minor contribution to the overall effects in drinking water (Stalter et al., 2016c),
which suggests that we can capture the majority of DBP-associated toxicity with
simpler common SPE methods.

12.5 SOLID-PHASE EXTRACTION

As SPE is the most commonly applied extraction method, more detailed information
on SPE, including different SPE sorbent options and common SPE procedure, is
provided below.

12.5.1 Solid-phase extraction sorbents

A wide range of sorbents have been used for SPE, with Oasis HLB (Waters) the
most commonly used. Other common SPE sorbents include Chromabond HR-X
(Macherey-Nagel), StrataX (Phenomenex) and octadecyl silica C18. Most of the
commonly used sorbents contain a copolymer mix, such as poly
(divinylbenzene-co-N-vinylpyrrolidone), with a hydrophilic monomer to capture
polar chemicals and a lipophilic monomer to capture hydrophobic chemicals.
However, these sorbents tend to recover a lower fraction of charged chemicals
compared to neutral chemicals (Neale et al., 2018; Osorio et al., 2018).
Consequently, some studies have applied combinations of multiple sorbents, such
as reverse-phase sorbents with ion-exchange materials, in order to capture a wider
range of micropollutants, including very polar chemicals and charged chemicals
(Aerni et al., 2004; Toušová et al., 2017; Osorio et al., 2018). Other studies have

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 256 Bioanalytical Tools in Water Quality Assessment

applied multilayer SPE with Oasis HLB and coconut charcoal to improve the
recovery of highly polar compounds (Escher et al., 2014; Leusch et al., 2014b).
However, a multilayer SPE cartridge with Oasis HLB, Strata-X-CW,
Strata-X-AW and Isolute ENV+ (1:1:1.5) and Supelclean EnviCarb was found to
cause blank effects at a comparatively low REF of 20 (Neale et al., 2018).
Bioassays are not able to differentiate between effects from a sample and effects
due to impurities from sample processing, so it is important to select an
extraction method with minimal blank effects. In any case, it is important to
always include processing controls with ultrapure water or glass-bottled water
when enriching water samples to confirm that the observed effects are due to
micropollutants in the sample and not related to the SPE sorbent or solvents.
In addition, a number of studies have compared the influence of different SPE
sorbents on bioactivity. For example, Rosenmai et al. (2018) applied both Oasis
HLB (poly(divinylbenzene-co-N-vinylpyrrolidone) and BondElut ENV (modified
styrene divinylbenzene) sorbents to extract wastewater and drinking water
samples, with no consistent difference in effect observed. Abbas et al. (2019)
compared three SPE sorbents, Oasis HLB, Telos C18/ENV and Supelco
ENVI-Carb+, at pH 2.5 and 7 and found that Telos C18/ENV at pH 7 was the
most effective for wastewater effluent and groundwater, although considerable
cytotoxicity was observed, which can mask the effect.

12.5.2 Solid-phase extraction procedure

Prior to extracting the water sample by SPE, it is necessary to condition the SPE
cartridge or disc to wet and activate the sorbent bed. Water-miscible methanol,
followed by ultrapure water is commonly used for conditioning (Bain et al.,
2014; Alygizakis et al., 2019; Lundqvist et al., 2019a). However, if other less
polar solvents are used for eluting the cartridge, such as dichloromethane or ethyl
acetate, then these solvents must also be used for conditioning. After conditioning
is finished, the cartridge must not run dry and the water sample should be
immediately percolated through the SPE cartridge.
Once water has been percolated and the chemicals in water sample are sorbed on
the SPE sorbent, the cartridge must be completely dried in a vacuum or nitrogen
stream. The dried cartridge can be sealed with parafilm and aluminium foil and
stored at −20°C until elution (Tang et al., 2014). Other studies have stored dried
SPE cartridges for up to 2 weeks at 4°C (Scott et al., 2014). Dried cartridges can
also be sent to bioassay laboratories for elution, which is simpler, safer and
cheaper than sending litres of unenriched water or solvent extracts.
To elute a wider range of polar and non-polar chemicals, multiple solvents are
often used for elution, such as methanol and 1:1 hexane:acetone (Scott et al.,
2014; Jia et al., 2015) or methanol and ethyl acetate (Houtman et al., 2018;
Müller et al., 2018). Other solvents used in the literature in different
combinations include acetonitrile, dichloromethane and MTBE. It should be

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Sampling, sample preparation and dosing 257

noted that any impurity in the solvents can potentially cause blank effects in the
bioassays. Consequently, it is important to use high purity (e.g., HPLC grade)
solvents for conditioning and elution and to limit the volume of solvent used.
Based on 500 mg/6 cc Oasis HLB SPE cartridge, 10 mL of each solvent is often
used for conditioning and 10 mL of each solvent is used for elution (Scott et al.,
2014; Müller et al., 2018). Smaller solvent volumes can be used with smaller
SPE sorbent beds.
Several studies have compared the effect of conditioning and elution solvents on
bioactivity. For example, Leusch et al. (2014a) found no significant difference in the
bioanalytical results when using 1:1 hexane:acetone and methanol compared to
methanol alone for conditioning and elution. Furthermore, Prochazkova et al.
(2018) compared the effect of two different solvent conditioning and elution
combinations on estrogenic activity in surface water extracts. The first method
targeted estrogenic compounds by conditioning and eluting with methanol, while
the second method targeted less polar compounds and used ethyl acetate,
methanol and 20% 2-propanol for conditioning and ethyl acetate for elution.
The different solvents often resulted in different 17β-estradiol equivalent
concentration values for the matching samples, but no systematic difference in
estrogenicity was observed.
After elution, the solvent is blown to dryness under nitrogen gas and resuspended
in a final solvent, such as methanol, dimethyl sulphoxide (DMSO) or ethanol.
Murk et al. (2002) compared the effect of storage conditions on the same extract
dissolved in ethanol and DMSO in the ERα CALUX assay. Initially, there was
no difference in effect, but ethanol was found to evaporate quickly when stored at
room temperature or 4°C and even evaporated at −20°C within 6 weeks. In
contrast, the DMSO stock did not significantly change in activity over the 6-week
period when stored at 4°C and −20°C. In Chapter 7, we discussed the
cytotoxicity and effects of solvents. DMSO is most problematic. Hence, we still
recommend using methanol as the solvent but only if the weight of the vial is
continuously monitored during the chain of custody and the solvent is topped up
if losses are recorded.

12.5.3 Effect recovery by solid-phase extraction

The recovery of individual chemicals by SPE has been well studied (Osorio et al.,
2018; Schulze et al., 2017), but less is known about effect recovery for bioassays.
Furthermore, unlike chemical analysis where an internal standard can be added to
correct for chemical recovery by SPE, internal standards should not be used for
bioanalysis as they may induce an effect in the bioassay that cannot be
distinguished from the other micropollutants in the sample.
There are a number of approaches that have been applied in the literature to assess
recovery, with most involving spiking a cocktail of chemicals into the water matrix
prior to SPE enrichment (Figure 12.5).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 258 Bioanalytical Tools in Water Quality Assessment

Figure 12.5 Various approaches to evaluate solid-phase extraction (SPE) recovery

of chemicals and effects. BEQ = bioanalytical equivalent concentration (definition
and equations in Chapter 7), SPE = solid-phase extraction, REPi = relative effect
potency of chemical i. Figure adapted from Neale et al. (2018). Solid-phase
extraction as sample preparation of water samples for cell-based and other in vitro
bioassays. Environmental Science: Processes & Impacts, 20: 493–504.
Reproduced by permission of The Royal Society of Chemistry © 2018.

As it can be difficult to measure the effect of the water alone, many studies
compared the effect in the extract, often expressed as a bioanalytical equivalent
concentration from bioanalysis (BEQbio,extract) to the predicted effect based on the
concentration of chemicals detected in the extract (BEQchem,extract) (Kolkman
et al., 2013; Leusch et al., 2010) or the nominal concentration of spiked
chemicals (BEQchem,nominal) (Thorpe et al., 2006; Kunz et al., 2017). The ratios
BEQchem,extract/BEQbio,extract and BEQchem,nominal/BEQbio,extract can be used as
proxy for effect recovery (Figure 12.5). Furthermore, Abbas et al. (2019)
attempted to assess SPE recovery by comparing the effect of the native water and
SPE extracts in unspiked wastewater and groundwater. Since other components
in the native water sample, such as salts, metals and other inorganics, could also
have an effect in the bioassay in addition to organic micropollutants, this
comparison is difficult to interpret.
To more quantitatively evaluate effect recovery by SPE, it is necessary to
consider the effect of the spiked mixture alone, the effect of the extracted sample
and the effect of the unspiked water alone (Figure 12.5). Neale et al. (2018)
evaluated the effect recovery of a mixture of 579 micropollutants spiked into
pristine surface water using a suite of bioassays indicative of xenobiotic
metabolism, hormone receptor-mediated effects and adaptive stress responses.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
Figure 12.6 Comparison of the bioanalytical equivalent concentrations BEQchem,extract and BEQchem,modelled 100% recovery
for activation of PXR, activation of ER (both ER GeneBLAzer (ER-bla) and MELN) and oxidative stress response
Sampling, sample preparation and dosing

(both AREc32 and ARE-bla) in various studies. The dotted lines indicate a factor of 2 difference between BEQchem,extract and
BEQchem,modelled 100% recovery. Figure adapted from Neale et al. (2018). Solid-phase extraction as sample preparation of water
samples for cell-based and other in vitro bioassays. Environmental Science: Processes & Impacts, 20: 493–504. Reproduced by
permission of The Royal Society of Chemistry © 2018.
259
 260 Bioanalytical Tools in Water Quality Assessment

LVSPE with HR-X sorbent was used. Effect recovery was calculated using the
effect of the spiked water (BEQbio,extract(water + mix)) minus the effect of the
unspiked water (BEQbio,extract(water)) divided by the effect of the mixture stock
solution (BEQbio(mix)). Effect recovery was within a factor of 2 of the optimal
100% recovery for most assays, which suggests that LVSPE is suitable for
capturing the majority of active chemicals (Neale et al., 2018). Also, a
comparison of BEQchem,extract and BEQchem,modelled 100% recovery from various
studies showed that loss during SPE can be neglected when testing water samples
(Figure 12.6).

12.6 SAMPLE COLLECTION AND SAMPLE PROCESSING

FLOW CHART
There are a number of decisions to be made regarding sample collection,
pre-treatment and enrichment. A decision-making flow chart (Figure 12.7) guides
users through some of those key decisions. Once the final sample pre-treatment
and processing methods have been selected, it is important to use the same
approach for all samples that are to be compared. It is not possible to truly
compare changes over time or differences between sites if different sample
pre-treatment and processing methods are used as this can affect the chemical
mixture in the final extract. Furthermore, where possible, the same bioassay and
chemical analysis pre-treatment and sample processing methods should be used
to allow greater comparability between the results.
The information used to support sample pre-treatment and processing decisions
is often based on user experience, with few studies investigating the impact of
different sample processing options on the biological effect. The majority of these
studies focus on estrogenic activity, with little known about other endpoints.
Furthermore, some of the advice is based on chemical analysis protocols, rather
than being specific for bioassays.
One of the least standardised but very important pre-treatment steps is sample
filtration (Figure 12.7). A wide range of filter pore sizes are used in the literature,
which will affect the amount of SPM retained on the SPE cartridge. However, to
our knowledge, the decision regarding which filter size to select or whether to
filter or not has not been based on scientific studies but rather on user experience.
Therefore, we suggest a uniform approach of filtering samples with a turbidity of
greater than 5 NTU using glass fibre filters. The role of SPM as a ‘carrier’ of
toxicity in water is important and hence, a concise definition of what is
considered as particulate matter and what is dissolved, including chemicals bound
to dissolved organic matter, is vital if one wants to compare different studies and
also evaluate the contribution of dissolved versus bound chemicals in water.
Concerning SPE, the volumes that are given in Figure 12.7 are merely indicative
as was discussed in Section 12.2.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Sampling, sample preparation and dosing 261

Figure 12.7 Sample pre-treatment and processing decision-making flow chart. The
following volumes are based on solid-phase extraction (SPE) materials Oasis HLB,
StrataX, Chromabond HR-X cartridges: a0.5 L per 200 mg or 1 L per 500 mg SPE
cartridge; b1 L per 200 mg or 2 L per 500 mg SPE cartridge; c2 L per 200 mg or 4 L
per 500 mg SPE cartridge; d1–2 L per 200 mg or 2–4 L per 500 mg SPE cartridge;
e
sodium azide can also be added for preservation; fglass fibre filter with pore size
0.7–1.5 µm recommended. *Not recommended. ‡3.5 mg/L sodium thiosulphate or
5 mg/L ascorbic acid to quench 1 mg/L free Cl2.

12.7 DOSING INTO BIOASSAYS

The sample extracts can be dosed directly into the bioassay or solvent exchanged to
a less toxic solvent by blowing down the elution solvent and resuspending in a final

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 262 Bioanalytical Tools in Water Quality Assessment

solvent, such as DMSO or methanol. Methanol and DMSO are the most commonly
used solvents for bioassay dosing. DMSO can dissolve a wider range of compounds
compared to methanol, but it is non-volatile, meaning that the extract cannot be
further enriched by blowing down. In addition, as noted in Chapter 7, DMSO is
more toxic compared to methanol. In contrast, methanol is volatile, meaning it
can be blown down to further enrich the sample.
Figure 12.8 gives an example of the combination of extraction and dosing into
well plate-based assays. Typically, a dosing plate is first prepared and then a
fraction of the content of this plate is transferred to an assay plate that already
contains cells that have been incubated for up to 24 h to achieve adherence to the
plate in an appropriate volume of medium, for example, 4000–20,000 cells in
120 µL medium for 96-well plates or 1000–5000 cells in 30 µL medium for
384-well plates. The transfer from the dosing plate to the assay plate can be
accomplished manually using a multichannel pipette, which is the typical
approach for 96-well plates, or by copying the entire 96-wells of the dosing plate
over to a 384-well plate using robotic devices, ideally using a 96-tip pipetting
head. This allows quadruplicates on the 384-well plate. As the inter-plate
variability (Chapter 11) is often rather low, it is common practice to transfer each
dosing plate over only in duplicates by using two 96-well dosing plates per
384-well assay plate.
Using the volumes given in the example in Figure 12.8, the EF for the extraction
is 4000 and the dilution for the highest concentration tested in the bioassay is 160,
resulting in a dosing factor of 0.00625. The REF value for the highest
concentration tested then becomes: REF = 4000 × 0.00625 = 25. The second
most concentrated sample would have an REF of 25/2 = 12.5, while the third
most concentrated sample would have an REF of 12.5/2 = 6.25, and so forth.
In this example, the highest concentration of solvent is 1%. DMSO is more toxic
compared to methanol, with a final DMSO concentration of 0.1% recommended in
cell-based bioassays. In contrast, up to 1% of methanol can be added to some
mammalian reporter gene assays (Leusch et al., 2017). This equates to a dilution
in the assay of 1000 for DMSO extracts compared to 100 for methanolic extracts,
meaning extracts in DMSO need to be enriched 10 times more than methanolic
extracts to achieve the same REF.
The REF can be further increased by exchanging the methanolic extract (or any
other extract in a volatile solvent) with cell culture media. This is achieved by
adding a volume of methanol to a glass vial (i.e., 2 mL HPLC vial), blowing down
to dryness and resuspending in cell culture media, which can be directly
transferred to the cells. This allows one to increase the REF in the assay without
inducing any solvent effects. This approach has been applied recently to drinking
water extracts to help detect effects in relatively clean samples (Hebert et al., 2018;
Neale et al., 2020b). It is important to ensure that the sample is well dissolved in
the bioassay medium. This is rarely a problem for water extracts but may occur if
SPM extracts are tested. It is also important to include solvent controls in the assay

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Sampling, sample preparation and dosing 263

Figure 12.8 Typical sample treatment and dosing into bioassays. (a) Sample
extraction and concentration, starting with 2000 mL of water sample concentrated
down to 0.5 mL of water extract (i.e., an enrichment factor EF of 4000). (b) The
water extract is transferred to a plate and serially diluted. The serial dilution is then
transferred to the assay plate in replicates with the option to transfer to a 96-well
plate in this case in triplicate or use a 96-tip head to copy over two to four times to
a 384-well plate.

to ensure that the solvent itself does not induce a response in the assay and even if the
solvent is blown down, the solvent controls have to mimic that step entirely.

12.8 CONCLUSIONS
The whole workflow from sampling to dosing is important to achieve meaningful
bioassay responses. The sampling strategy will be dictated by the purpose and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 264 Bioanalytical Tools in Water Quality Assessment

objective of the sampling campaign, as well as the sample context. After sample
collection, important considerations include sample preservation, storage time and
whether to filter or not prior to enrichment. SPE is most commonly used for
sample enrichment, but it is important to select SPE sorbents and conditioning
and elution procedures with no blank effects. This can be an issue when using
enrichment methods adapted from chemical analysis workflows. Many of the
currently used methods are based on user experience, with further research
required to evaluate the influence of different sample pre-treatment and
processing steps on the biological effect.
Further dosing poses a challenge if DMSO is used as a solvent. Although DMSO
extracts are easier to handle, extracts in volatile solvents pose the advantage that the
solvent can be evaporated prior to dosing into cell assays.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Chapter 13
Design of test batteries and
interpretation of bioassay results

13.1 INTRODUCTION
Cell-based bioassays have been applied for water quality assessment for decades.
Initially, batteries of assays were put together rather arbitrarily, limited by the
availability of particular assays or prior experience. More recently, rational
designs of test batteries have been attempted and groups have collaborated to
assemble broad test batteries for evaluative purposes (Brack et al., 2019). After
years of application, practical monitoring and screening test batteries have
emerged that are widely applied. In this chapter, we outline design principles of
test batteries for water quality assessment for different purposes.
Due to the complex mixture of chemicals present in environmental water extracts,
a single bioassay cannot capture all the effects that may be triggered. Therefore, a test
battery of bioassays covering different modes of action is required to detect the effect
of as many active chemicals in a water sample as possible. Some studies have
proposed bioassay test batteries as a proxy for acute and chronic effects at an
ecosystem level (Schweigert et al., 2002; Di Paolo et al., 2016) or based on
potential human health effects (Leusch et al., 2014b), while other test batteries
have focused on covering relevant modes of action (Escher et al., 2008b; Neale
et al., 2017b) or focused on specific modes of action such as genotoxicity
(Pellacani et al., 2006) or endocrine disruption (Swart et al., 2011).
Significant attention has been given in the last decade to endocrine disrupting
effects, and particularly bioassays for estrogenicity. There was good reason for
this focus: estrogens are very potent and the concentrations detected in surface

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0265

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 266 Bioanalytical Tools in Water Quality Assessment

waters are in the range where developmental and reproductive effects on fish have
been observed. It is important, however, to include a variety of bioassays in any test
battery to ensure that the spectrum of evaluated endpoints is sufficiently broad to
provide a thorough assessment of overall water quality.
As discussed in previous chapters, bioassays can be selected to target either
particular groups of chemicals (e.g., photosynthesis inhibition is often very well
correlated with herbicide concentration) or toxicity endpoints relevant to the
protection goal of interest (e.g., bacterial cytotoxicity assays would be useful to
predict the impact of toxicants on microorganisms in wastewater treatment
plants). In many instances, both approaches can cross over and lead to similar
bioassay selections. A project evaluating the suitability of stormwater reuse for
irrigation, for example, would most likely include an assay targeting
photosynthesis inhibition, either because it can detect herbicides potentially
present in the samples (chemical group motivated) and/or because photosynthesis
inhibition is a relevant toxic endpoint for the crops to be irrigated (protection
goal motivated). A project considering the suitability of reclaimed water for
potable use would certainly include bioassays for genotoxicity, because this is a
relevant protection goal for human health, but also because some compounds of
concern potentially present in reclaimed water (e.g., disinfection by-products)
are genotoxic.
Bioassays that have been applied for water quality monitoring are reviewed in
Chapter 10. Here, we introduce how test batteries are set up. It is important that
each assay in the battery is validated and that issues that can affect the quality of
the data are well understood. Chapter 11 provides some advice on quality
assurance and quality control protocols that can help ensure the production of
reliable and meaningful data. Chapter 12 gives guidance of sample preparation,
which is relevant for the interpretation of bioassay results. It is possible to test the
entire water sample (see also Chapter 3 on whole effluent testing) but it is most
common practice to extract the organic micropollutants using solid-phase
extraction (SPE), liquid–liquid extraction (LLE) or passive samplers. The
extraction of complex mixtures of organic micropollutants, leaving behind
inorganic components and most of natural organic matter, becomes relevant when
interpreting the results of bioassays.
In this chapter, we also show how chemical analysis can be linked to the bioassay
results by iceberg modelling, a mixture modelling approach that relies on the concept
of concentration-additive effects (Chapter 8). Iceberg modelling leads to the
classification of bioassays into two categories. Category 1 assays are highly specific
bioassays that are mainly triggered by a limited number of known and generally
potent chemicals. For these category 1 bioassays, almost all of the detected effects
can typically be explained by known chemicals. A good example of a category 1
assay is a reporter gene assay for the estrogen receptor (ER). In most water
samples, .90% of the estrogenic activity will be caused by natural hormones (such
as estradiol, estrone and estriol), synthetic hormones (such as ethinylestradiol) and

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 267

a few industrial xenoestrogens (such as bisphenol A or nonylphenol). Effect-directed

analysis (EDA) has been applied successfully to identify unknown contributors to the
mixture effects in water samples for category 1 bioassays.
In category 2 bioassays, only a small fraction of the measured effect can typically
be explained by known and detected chemicals, which has important implications
for the interpretation of the bioassay results. Typical category 2 bioassays are
those indicative of adaptive stress responses or apical endpoints.
Many in vitro bioassays, particularly mammalian reporter gene assays, are highly
sensitive by design and can detect effects in relatively clean waters, such as drinking
water and recycled water, especially after sufficient enrichment (Jia et al., 2015;
Conley et al., 2017b; Neale et al., 2020b). However, just because an effect is
detected in a bioassay does not necessarily mean that the chemical water quality
is unacceptable. To help bioassay users differentiate between an acceptable and
unacceptable bioassay response, effect-based trigger values (EBTs) have been
proposed for both category 1 and category 2 bioassays. An EBT is comparable to
a water quality guideline value (GV) but provides a specific threshold for each
type and class of bioassay. What is acceptable or not depends on the water type
and its usage and should be related to safe concentrations of regulated chemicals
and be protective for the target organisms, for example, aquatic life in surface
water, humans for drinking water.

13.2 TEST BATTERIES

13.2.1 Test battery design
Ideally, a bioassay test battery for water quality assessment should focus on effects
commonly detected in water and include assays that cover a range of relevant
modes of action as well as apical effects in whole organisms (Figure 13.1) (Neale
et al., 2017b). This helps to ensure that expected effects are covered, while
safeguarding against missing any unexpected modes of action.
Test batteries indicative of different stages of cellular toxicity pathways, such as
induction of xenobiotic metabolism, receptor-mediated effects, adaptive stress
responses and cytotoxicity, have been applied to wastewater, surface water,
stormwater, recycled water and drinking water (Escher et al., 2014; Leusch et al.,
2014b; Nivala et al., 2018; de Baat et al., 2020). Although an effect at the cellular
level may not necessarily result in higher level adverse effects (Ankley et al.,
2010), assays indicative of different stages of cellular toxicity pathways can act
as sensitive indicators of chemicals. In this respect, they can be used as proxies
for chronic effects. An example is endocrine disruption, which would require
testing of the whole life cycle and adverse effect on reproduction, both of which
are far too time-consuming and expensive for routine testing of water samples.
We know from field studies that estrogenic compounds in wastewater can lead
to feminisation of male fish, and that a low nanogram per litre level of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 268 Bioanalytical Tools in Water Quality Assessment

Figure 13.1 Bioassays relevant for water quality monitoring (circles) anchored in
crucial steps of the adverse outcome pathway (boxes). ER = estrogen receptor,
AR = androgen receptor, GR = glucocorticoid receptor, PR = progesterone
receptor, TR = thyroid receptor, ARA = retinoic acid receptor, RXR retinoid X
receptor, p 53 =, Nrf2 = NF-E2 related factor 2, nfκB = nuclear factor kappa B,
HRE = hypoxia response element. Modified and reprinted with permission from
Neale et al. (2017b). Development of a bioanalytical test battery for water quality
monitoring: Fingerprinting identified micropollutants and their contribution to effects
in surface water. Water Research, 123, 734–750. © 2017 Elsevier.

17α-ethinylestradiol can collapse entire fish populations, hence, including an

estrogenicity test in a bioassay battery will be protective for this aspect of
reproductive chronic effects.
In addition, whole organism assays indicative of apical effects, such as algal
growth inhibition, invertebrate and fish embryo toxicity, are a useful complement
to cell-based assays as they can integrate effects from multiple toxicity pathways
and assure that all bioavailable chemicals are considered. For drinking water,
studies with mammals are prohibitive in a routine monitoring context and have
only rarely been performed (Narotsky et al., 2013) but the fish embryo toxicity
test with Danio rerio may serve as a proxy for acute toxicity and reporter gene
assays for chronic effect. Hence, drinking and surface waters can be monitored
with very similar test batteries, which can be applied across the entire water cycle
from wastewater to recycled water, from groundwater to drinking water and
beyond (Escher et al., 2014).
In all test batteries, the specific effect measured with a reporter gene assay or a
biomarker should be accompanied by cytotoxicity assessment (Chapter 7). This is
because cytotoxicity may cause false-negative results by masking the effect or
may elicit false-positive results if there is apparent activation due to the
‘cytotoxicity burst’ phenomenon (Judson et al., 2016; Escher et al., 2020a).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 269

13.2.2 Multiplex bioassays serving as test batteries

In addition to assays indicative of a single endpoint, some studies have
applied multiplex high-throughput screening assays such as the Attagene
trans-FACTORIALTM and cis-FACTORIALTM assays to drinking water, surface
water and wastewater (Escher et al., 2014; Blackwell et al., 2019; Medlock
Kakaley et al., 2020). These assays include many targeted endpoints covering
different stages of cellular toxicity pathways. For example, Escher et al. (2014)
found the pregnane X receptor (PXR), ERα, peroxisome proliferator-activated
receptor (PPARγ), aryl hydrocarbon receptor (AhR) and the antioxidant response
element ARE were the most responsive in wastewater, surface water and recycled
water extracts. Blackwell et al. (2019) evaluated the surface water of 38 rivers in
the USA using this multiplexed battery and also confirmed that the ERα and
GR were most prominently activated, together with AhR and PXR. Figure 13.2
shows an exemplary radar plot of blank-normalised area-under-the-curve (AUC)
values for one water sample of this study. Medlock Kakaley et al. (2020) found
that PXR and PXR signalling pathway (PXRE) were the only nuclear receptors
activated in the intake to a drinking water treatment plant (DWTP), with no effect
in treated water. These results support the findings from individual reporter gene
assays, with assays indicative of activation of PXR, AhR and ERα found to be
most responsive to a wide range of water types. They also confirm that the
iterative strategy over the last few years of implementing various reporter gene
assays has narrowed down the pertinent endpoints and has not overlooked
important endpoints relevant for water quality. However, compared to targeted
reporter gene assays, Blackwell et al. (2019) reported that the multiplex assays
were less sensitive when applied to the same surface water samples compared
to reporter gene assays. Thus, they might be less suitable for surveillance
monitoring, where reporter gene assays will remain the preferred choice. Since
these multiplexed assays cover a wider range of endpoints, they are especially
suitable for screening purposes to assure that new endpoints are not overlooked
when a new type of water is investigated.

13.2.3 Routine test batteries for monitoring applications

A practical test battery of at least three or four bioassays representative of effects
commonly detected in water samples and aligned with relevant steps of cellular
toxicity pathways is recommended for routine water quality monitoring.
Although it is possible that other relevant effects may be missed with only three
to four bioassays, such a routine test battery can still provide a good measure of
the overall water quality and these routine test batteries are useful for
benchmarking across different stages of the water cycle if those bioassays were
chosen because they are most responsive to the target type of water.
Assay selection will depend both on the context (e.g., water type and treatment
type) and the purpose of the sampling campaign (e.g., to assess product quality or

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 270 Bioanalytical Tools in Water Quality Assessment

Figure 13.2 Example radar plot showing commonly activated endpoints of the
Attagene cis-FACTORIALTM and trans-FACTORIALTM assay in surface water
extracts on the example of one site in South Platte River in Colorado. The effect
endpoint is the blank-normalised AUC between an REF of 0.04 and 10. The dark
circle marks the effect threshold of 1.5. Figure reprinted with permission from
Blackwell et al. (2019). Potential toxicity of complex mixtures in surface waters from
a nationwide survey of United States streams: Identifying in vitro bioactivities and
causative chemicals. Environmental Science & Technology, 53(2), 973–983.
© 2019 American Chemical Society.

treatment process efficacy). In the case of wastewater and water reuse for
non-potable use, a minimum test battery should include assays indicative of
activation of the AhR, activation of ERα and oxidative stress response. These
three endpoints are responsive to extracts of a range of water types, as
demonstrated by both individual and multiplexed assays, and represent different
stages of cellular toxicity pathways, that is, xenobiotic metabolism, receptor-
mediated effects and adaptive stress responses. Furthermore, proposed EBTs are
available for these endpoints as outlined below. This recommendation aligns with

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 271

recommendations for testing surface water quality (Brack et al., 2019), and such
harmonisation is important given that rivers are receiving effluent discharges and
are often at the same time the source water for DWTPs.
In the context of drinking water treatment or water reuse for potable use, a test
battery should include an assay indicative of genotoxicity or mutagenicity in
addition to activation of AhR, activation of ERα and oxidative stress response. It
is worth noting that oxidative stress response assays can detect increased effects
after drinking water disinfection. However, they cannot replace mutagenicity or
genotoxicity testing (with traditional bacterial assays such as the Ames test or
umuC assay) but are often also triggered by genotoxic chemicals, not only those
with direct reactive toxicity.
Although test batteries of three or four assays are recommended as the
minimum in most situations, more comprehensive test batteries could include
any assay previously found to have a response in water extracts plus whole
organism assays indicative of apical effects (e.g., an algal growth inhibition
assay or fish embryo toxicity assay). The selection of additional assays may be
related to specific water quality concerns. For example, a phytotoxicity assay
could be included if raw drinking water is collected from a catchment with
significant agricultural activity.
If advanced cell culture laboratories required to run mammalian reporter gene
assays are not accessible, simple bacterial toxicity assays, such as the Microtox or
BLT-Screen, could be applied as a simple screening tool. Both assays are
similarly sensitive and have been applied to wastewater, surface water and
drinking water. It should be noted that these assays only provide information
about non-specific effects and should be complemented with assays indicative of
specific effects when possible, but they can be powerful as indicators of relative
chemical water quality (e.g., to measure changes over time or to compare
different waters).

13.3 LINKING BIOASSAY RESULTS WITH CHEMICAL

ANALYSIS: ICEBERG MODELLING
13.3.1 Iceberg modelling
In Chapter 1 we introduced the spotlight analogy with different analytical and
bioanalytical tools shedding light on different parts of the universe of chemicals
(Figure 13.3). Cytotoxicity gives a measure of all chemicals that are bioavailable
and can be taken up into cells. Thus, each chemical in a water sample or extract
contributes to the mixture effects. Specific effects quantified with reporter gene
assays shed a bright light only on a small subset of the chemicals that can be
captured by cytotoxicity. There is overlap of the chemicals captured by bioassays
and the chemicals that can be quantified by target analysis and suspect screening
(Figure 13.3).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 272 Bioanalytical Tools in Water Quality Assessment

Figure 13.3 Chemical analysis and bioanalytical tools shed light on different parts of
the universe of organic micropollutants. This is a simplified version of Figure 1.7 in
Chapter 1, which was adapted from Escher et al. (2020b).

To quantify this overlap and to estimate how much of the observed cytotoxicity
and specific effects can be explained by the detected and quantified chemicals,
so-called ‘iceberg (mixture) modelling’ can be used. In Chapter 7, we have learnt
how effect concentrations ECy can be derived from concentration–response
curves describing effect y, which is typically 10% of maximum effect or an
induction ratio (IR) of 1.5. For better comparison between assays and samples we
have also introduced the so-called ‘bioanalytical equivalent concentrations’
(BEQbio) that translate the effect detected in a sample into concentration units of
a reference chemical for this bioassay by dividing the ECy of the reference
compound by the ECy of the water sample (Figure 13.4). The BEQbio gives the
overall effect that a sample can trigger expressed as concentration of a reference
compound that would cause exactly this effect.
In Chapter 8 we have learnt about mixture toxicity concepts and that at low effect
levels, as they occur in water samples, we can predict the combined effect of
mixtures with many components for both similarly acting and dissimilarly acting
compounds with a very simple model where the products of the relative effect
potencies (REPi) of each known mixture component i and its concentration (Ci)
can be summed up to calculate the predicted mixture BEQchem. In the iceberg
analogy the BEQchem constitutes the visible tip of the iceberg and the BEQbio the
entire iceberg. The part of the iceberg that is submerged represents the effects
from unknown chemicals or those that are present below their chemical limit of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 273

Figure 13.4 Iceberg modelling connecting mixture effects predicted from

concentration of n chemicals i (subscript chem) with measured effects (subscript
bio). The unknown effect (subscript unknown) can be calculated by subtraction of
the measured effect from that predicted from known chemicals. Figure adapted
from Villeneuve et al. (2019) and Neale et al. (2020b). BEQ = bioanalytical
equivalent concentrations; EU = effect unit; REP = relative effect potency; Ci =
concentration of chemical i; ECy = effect concentration triggering effect y.

detection. The BEQunknown can be deduced from subtracting BEQchem from BEQbio
(Figure 13.4). Similar calculations hold for effect units (EU; Figure 13.4).
Analogously to EU, we can define toxic units (TU) for cytotoxicity from
inhibitory concentration (ICy) for cytotoxicity or lethal concentrations for whole
organism toxicity (not shown in the figure).
If BEQbio = BEQchem, then the measured effect can be fully explained by known
chemicals. If BEQbio . BEQchem, then the detected chemicals cannot fully explain
the measured effect, either because their concentrations are below the chemical
detection limit or because there are additional chemicals that were not measured
but contribute to the mixture effect (Figure 13.5a).
Typical results of iceberg modelling are shown in Figure 13.5b. For example, a
study on surface water impacted by wastewater treatment plants (Könemann et al.,
2018) reported three potent estrogens (17β-estradiol, 17α-ethinylestradiol and
estrone) often explained almost all of the measured estrogenic activity EEQbio,
especially at higher EEQ levels. At lower EEQ levels, EEQchem were slightly
lower than EEQbio. This is likely due to one or more of the potent estrogens falling

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
274

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Figure 13.5 (a) Comparison between bioanalytical equivalent concentrations BEQbio from bioassay measurements and the
BEQchem predicted from the detected chemicals’ concentrations and their REP (see Figure 13.4 for equations). For category 1
bioassays, BEQbio ≅ BEQchem and the effect can be explained by the detected chemicals. For category 2 bioassays, BEQbio .
BEQchem and most of the effect stems from unknown chemicals. (b) Examples for estrogenicity (EEQ, green diamonds) detected
Bioanalytical Tools in Water Quality Assessment

with ERα GeneBLAzer in surface water impacted by WWTP (Könemann et al., 2018) and surface water during rain events (Neale
et al., 2020a): benzo[a]pyrene EQ (B[a]P EQ) in AhR CALUX (blue upward triangles), rosiglitazone EQ in PPARγ GeneBLAzer
(blue downward triangles) and dichlorvos EQ in AREc32 (red squares).
 Design of test batteries and interpretation of bioassay results 275

below the detection limit of the chemical analysis (usually around 0.1–1 ng/L).
Similar good agreement between BEQbio and BEQchem has also been observed for
other hormone receptor activation (Hashmi et al., 2018, 2020) and photosynthesis
inhibition in green algae (Tang and Escher, 2014; Glauch and Escher 2020).
Bioassays where BEQbio ≅ BEQchem have been termed ‘category 1 bioassays’
(Escher et al., 2018).
In contrast, bioassays where BEQbio . BEQchem have been termed ‘category 2
bioassays’ (Escher et al., 2018). For example, less than 1% of BEQbio can
typically be explained by BEQchem when measuring activation of the AhR in the
AhR CALUX, activation of PPARγ in the PPARγ GeneBLAzer or oxidative
stress response with the AREc32 assay (Figure 13.5b), even when many
chemicals are analysed. To illustrate, Neale et al. (2020a) detected 290 chemicals
in surface water samples, but almost a fifth of those chemicals (55–58) had
no experimental effect data, and thus it was not possible to calculate the
BEQchem contribution for those compounds. Of the remaining .200 chemicals,
40 were active in the AhR CALUX, 20 in the PPARγ GeneBLAzer and 52 in the
AREc32, but these still only explained ,5% of the BEQbio. The low fraction of
effect explained is not solely due to a lack of data but lays in the type of
chemicals and their potency in these bioassays, as will be detailed in Section 13.4.

13.3.2 Effect-directed analysis

So-called ‘effect-directed analysis’ (EDA) can be used in category 1 bioassays to
identify the bioactive components of a complex mixture in a water sample
(Houtman et al., 2011; Brack et al., 2016). In EDA, a water sample or extract
that is bioactive can be fractionated and the individual fractions re-tested for
bioactivity (Figure 13.6). The bioactive fractions then undergo chemical analysis
to identify the causative agents. These identified chemicals are also tested in the
bioassays, ideally in mixtures in concentration in which they were found to
confirm the extract/fraction effect. Several iterations of the process can be
performed to narrow in the causative toxicants. The whole process is similar to
the toxicity identification evaluation (TIE) protocol in WET testing (Chapter 3),
but in this case applied to a cell-based bioassay context.
For category 2 bioassays, EDA is not easily applicable because the bioactivity is
typically spread over many fractions and will disappear after fractionation as has
been demonstrated on the example of the AREc32 assay for oxidative stress
response (Hashmi et al., 2018).
Hashmi et al. (2018) identified the causative estrogens in surface water impacted
by untreated wastewater as mainly 17β-estradiol, estrone and 17α-ethinylestradiol
with some very minor contribution by genistein, daidzein and testosterone. They
also showed a larger spectrum of natural and man-made hormones to be responsible
for the androgenic effects. Mixtures of the detected chemicals confirmed that
mixture effects of the components were concentration-additive in most fractions.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 276 Bioanalytical Tools in Water Quality Assessment

Figure 13.6 Principle of EDA. Modified after and reprinted with permission from
Brack et al. (2016). Effect-directed analysis supporting monitoring of aquatic
environments – An in-depth overview. Science of the Total Environment, 544,
1073–1118. © 2016 Elsevier.

Fractionation can also ‘bring out’ the activity. For example, activation of
hormone receptors may be masked by cytotoxicity in unfractionated samples,
or agonistic and antagonistic effects may cancel each other out. The same
unfractionated water samples from Hashmi et al. (2018) did not show any
progestogenic and glucocorticoid activity but bioactivity was detected with
reporter gene assays for the progesterone receptor (PR) and glucocorticoid
receptor (GR) after fractionation (Hashmi et al., 2020). It proved to be much
more challenging to identify the causative chemicals for progestogenic and
glucocorticoid activity requiring non-target screening analysis before confirmation
of individual components’ activity with reporter gene assays (Hashmi et al., 2020).
Classical EDA is quite work-intensive and has been more commonly applied to
identify causative chemicals at contaminated sites or in research contexts than for
water quality monitoring. This might change in the future with the development of
high-throughput methods for EDA. Zwart et al. (2018, 2020) developed an
EDA platform that combines liquid chromatography (LC) with mass spectrometry
(MS) and parallel bioassay detection. The direct comparison of the MS
chromatograms with the bioassay responses revealed prominent MS peaks with no
bioactivity but also identified smaller peaks with high bioactivity. This method
was not only applied to estrogenic, androgenic and glucocorticoid activity (Zwart
et al., 2018, 2020) but also to the Ames assay for mutagenicity (Zwart et al.,
2020). Here, 1,2,3-benzotriazole, a highly abundant anticorrosion agent, and
2-formyl-1H-pyrrole could be identified as a mutagenic component of wastewater.
In a further optimisation step of the platform for even higher resolution,
the fractions were directly split after the LC separation, with an aliquot going to
mass spectrometry analysis and the other collected in 384-well plates that could
be used directly for the bioassay (Houtman et al., 2020). No concentration–
response curve could be run but almost 300 fractions per sample provided
something like a ‘bioassay chromatogram’ (Figure 13.7). The typical culprits

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
Figure 13.7 Bioassay chromatogram for ER and GR CALUX (left) as we run AR CALUX in agonistic and antagonistic modes
(right) with the bioactive components identified by mass spectrometry (MS). AR = androgen receptor, ER = estrogen receptor,
GR = glucocorticoid receptor, PR = progesterone receptor. Reprinted with permission from Houtman et al. (2020). High resolution
effect-directed analysis of steroid hormone (ant)agonists in surface and wastewater quality monitoring. Environmental Toxicology
and Pharmacology, 80, 103460. © 2020 Elsevier.
Design of test batteries and interpretation of bioassay results
277
 278 Bioanalytical Tools in Water Quality Assessment

(17β-estradiol, estrone and 17α-ethinylestradiol) could be identified with the

estrogen receptor (ER) CALUX assay but also less well-known compounds were
revealed as contributing strongly to bioactivity in GR CALUX and AR CALUX,
which was also run in antagonistic mode. These early results are quite
encouraging for future applications of EDA as a routine method in monitoring.

13.4 CATEGORY 1 AND CATEGORY 2 BIOASSAYS

According to iceberg modelling, category 1 bioassays target highly specific, mainly
receptor-mediated effects such as binding to hormone receptors or inhibition of
photosynthesis. Category 2 bioassays are those assays covering effect endpoints
that are triggered by many more and more diverse chemicals that still exhibit
specific effects but with a lower degree of specificity. Category 2 bioassays also
include oxidative stress response and reporter gene assays with more
promiscuous nuclear receptors, such as AhR and PPARγ.
The specificity ratio SRbaseline, introduced in Chapter 9 (Equation 9.4), provides
another more quantitative way to classify bioassays into these two categories.
SRbaseline is a quantitative measure of how much more potent the activation of a
specific response is in comparison with the baseline cytotoxicity of the same

Figure 13.8 (a) Conceptual figure describing the derivation of the specificity ratio
SRbaseline from experimental effect concentrations (EC) and the inhibitory
concentration causing 10% reduction in cell viability (IC10,baseline) that was predicted
from the quantitative structure activity relationship (QSAR) for baseline toxicity. (b)
Conceptual figure of distributions of specificity ratios log SRbaseline for highly specific
bioassays (category 1) and those that respond to many different chemicals (category
2) compared to distributions of log SRbaseline of baseline toxicants. Figure reprinted
with permission from Escher and Neale (2021). Effect-based trigger values for
mixtures of chemicals in surface water detected with in vitro bioassays. Environmental
Toxicology and Chemistry, 40, 487–499. © 2021. The Authors. Environmental
Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 279

compound (Figure 13.8a). Baseline toxicants have an SRbaseline of 1 and the data are
logarithmically distributed around log SRbaseline of 0 (Escher and Neale, 2021).
Chemicals activating category 1 bioassays have high log SRbaseline values, often
in the range of 4–8, while category 2 chemicals have log SRbaseline with medium
values and often very broad distributions (Figure 13.8b).
Chemicals active in category 2 assays follow a broad log-normal distribution
(over 4–5 log units wide) centred close to a value of log SRbaseline = 0, as
demonstrated in Figure 13.9a–c for the activation of the oxidative stress response,
PPARγ and AhR receptors. Category 1 bioassays, in contrast, have a low number
of very highly potent chemicals with medians of the log-normal distribution at
SRbaseline of 4 × 106 (log 6.6) for estrogenicity (Figure 13.9d) and 2 × 104 (log
4.3) for inhibition of the photosystem II in green algae (Figure 13.9e). What

Figure 13.9 Distributions of specificity ratio SRbaseline for category 1 and 2 bioassays
using experimental data from Escher and Neale (2021) for (a–d), and from Glauch
and Escher (2020) for (e). (a) Activation of oxidative stress response ARE, (b)
activation of the PPARγ receptor, (c) AhR, (d) estrogen receptor ERα and (e)
inhibition of photosynthesis in green algae after 24 h.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 280 Bioanalytical Tools in Water Quality Assessment

makes the picture more complex is that category 1 bioassays additionally respond to
a large number of chemicals with very low SRbaseline that show a distinctly separate
distribution from the high potency chemicals. Those low potency chemicals
generally do not play a significant role in iceberg modelling unless they are
present at significantly higher concentration (1000 to 1 million times higher)
compared to the high potency chemicals.
It must be noted, though, that the classification is dependent on the context. For
AhR, there are highly potent ligands, such as polychlorinated dibenzo-p-dioxins and
furans and if these were included, AhR would also be classified as a category 1
bioassay. However, those chemicals are very hydrophobic and do not dissolve in
water. The AhR assay is therefore classified as a category 2 bioassay for
water samples.
The classification criteria for category 1 and 2 bioassays are also relevant for the
derivation of EBT values discussed in Section 13.5.

13.5 EFFECT-BASED TRIGGER VALUES

As noted earlier, in vitro bioassays (particularly mammalian reporter gene assays)
are highly sensitive and can detect effects in relatively clean waters especially
after sufficient enrichment. It is critical therefore to provide a tool to water
utilities and regulators to determine a value below which a bioassay response is
unlikely to produce adverse effects. This threshold is called an ‘effect-based
trigger value’ (EBT), and various approaches have been devised to calculate
EBTs for category 1 and category 2 assays. There are principal differences for the
two categories because we know the mixture effect drivers in category 1
bioassays, while there are many unknown chemicals contributing to the effects of
category 2 bioassays. The two are thus presented next in two sections, one for
each category of bioassays. It must be noted that this category 1 and 2
classification is a recent concept, and therefore the reviewed papers did not
specifically provide this classification. The approaches were divided into category
1 and category 2 for the purpose of this chapter.
EBTs for surface water and wastewater need to be protective of ecosystem health
and are therefore typically derived from safe concentrations for aquatic organisms.
After translation to an in vitro bioassay, the cell line applied does not necessarily
need to be of the origin of an aquatic species because the in vitro effects are used
as bioanalytical measures and not as ecological effect measures. The same
considerations hold for drinking water quality.

13.5.1 Approaches to derive effect-based trigger

values for category 1 bioassays
Several different approaches have been applied to derive EBTs for category 1
bioassays, covering both drinking water and surface water (Figure 13.10). Most

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 281

Figure 13.10 Derivation of EBT values of category 1 bioassays for drinking water and
surface water by different approaches: (1) from an ADI of a RfD; (2) from an ADI/RfD
incorporating toxicokinetic parameters; (3) from an established GV and (4) from an
established GV incorporating relative potencies.

rely on chemical GVs as the point of departure. A summary of the currently

available EBTs for category 1 bioassays is provided in Table 13.1.
Bioassay-specific EBTs, rather than generic EBTs for an endpoint, are provided
in Table 13.1 because differences in assay sensitivity and chemical potency can
result in different EBTs for assays indicative of the same endpoint.
The simplest approach for drinking water EBTs is to translate the reference
dose (RfD) or acceptable daily intake (ADI) of a chemical that also serves as a
potent reference compound to water concentrations in a similar fashion as
World Health Organisation (WHO) drinking water guidelines are derived
(WHO, 2017b), using the average body weight (usually taken as 60–70 kg), the
amount of water typically consumed in a day (usually taken as 2 L/day)
and a specific allocation for drinking water (10%–20%) to account for other
sources of exposure. These are combined to calculate a concentration that is
equivalent to the GV for drinking water. The EBT is then simply the
bioanalytical equivalent of this concentration (approach 1 in Figure 13.10).
Genthe et al. (2009) derived an EBT for estrogenic activity from the ADI with
this method.
In addition, the concentrations of reference compounds considered safe in vivo
can be converted to concentrations in in vitro assays using differences in the
toxicokinetics of different compounds to adjust this direct read-across EBT

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 282 Bioanalytical Tools in Water Quality Assessment

Table 13.1 Summary of proposed effect-based trigger values for category 1 bioassays
covering human health and ecological health that are currently available in the literature.

Endpoint// Bioassay Human EBT (Drinking and Ecological EBT

Recycled Water for (Surface Water)
Indirect Potable Reuse)

Receptor-mediated effects
Estrogenic activity
–a 0.2 ng/L EEQ(5) 0.4 ng/L EEQ(1)
ERα CALUX 0.2 ng/L EEQ(2) 0.5 ng/L EEQ(3)
3.8 ng/L EEQ(6) 0.10 ng/L EEQ(4)
0.25 ng/L EEQ(7) 0.28 ng/L EEQ(7)
0.2–0.4 ng/L EEQb (8)
ERα GeneBLAzer 1.8 ng/L EEQ(2) 0.34 ng/L EEQ(4)
0.24 ng/L EEQ(7)
E-SCREEN 0.9 ng/L EEQ(2) 0.1–0.3 ng/L EEQb (8)
YES 12 ng/L EEQ (2) 0.2–0.4 ng/L EEQb (8)
HeLa-9903 0.6 ng/L EEQ (2) 1.0 ng/L EEQ(4)
0.18 ng/L EEQ(7)
MELN 0.37 ng/L EEQ(4)
0.56 ng/L EEQ(7)
0.2–0.3 ng/L EEQb (8)
MVLN 0.1–0.3 ng/L EEQb (8)
ERα-Luc-BG1 0.62 ng/L EEQ(4)
A-YES 0.56 ng/L EEQ(4)
3d YES 0.88 ng/L EEQ(4)
ISO-LYES (Sumpter) 0.97 ng/L EEQ(4)
ISO-LYES (McDonnell) 1.1 ng/L EEQ(4)
pYES 0.5 ng/L EEQ(7)
EASZY 2.2 ng/L EEQ(4)
(Cyp19a1b-GFP)
REACTIV (unspiked) 0.80 ng/L EEQ(4)
Androgenic activity
AR CALUX 11 ng/L DHT EQ(6)
4.5 ng/L DHT EQ(7)
AR GeneBLAzer 14 ng/L testosterone EQ(2)
AR CALUX 11 ng/L DHT EQ(6)
4.5 ng/L DHT EQ(7)
Anti-androgenic activity
Anti-AR CALUX 4.8 µg/L flutamide EQ(7) 25 µg/L flutamide EQ(3)
14 µg/L flutamide EQc (4)
Anti-AR GeneBLAzer 3.3 µg/L flutamide EQc (4)

(Continued )

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 283

Table 13.1 Summary of proposed effect-based trigger values for category 1

bioassays covering human health and ecological health that are currently available in
the literature (Continued).

Endpoint// Bioassay Human EBT (Drinking and Ecological EBT

Recycled Water for (Surface Water)
Indirect Potable Reuse)

Anti-MDA-kb2 3.5 µg/L flutamide EQc (4)

Anti-AR RADAR 3.6 µg/L flutamide EQc (4)
(spiked)
Glucocorticoid activity
GR CALUX 150 ng/L dexamethasone 100 ng/L dexamethasone
EQ(2) EQ(3)
21 ng/L dexamethasone
EQ(6)
Progestogenic activity
PR CALUX 724 ng/L levonorgestrel
EQd (6)
Anti-progestogenic activity
Anti-PR CALUX 1967 ng/L endosulphan
EQc (4)
Thyroid activity
TTR RLBA 0.06 µg/L thyroxine EQ(4)
TTR FITC-T4 0.49 µg/L thyroxine EQ(4)
XETA (unspiked) 0.62 ng/L triiodothyronine
EQ(4)
Anti-thyroid activity
Anti-TR-LUC-GH3 0.60 µg/L bisphenol A
EQc (4)
Photosynthesis inhibition
Combined algae test 0.6 µg/L diuron EQ(2) 0.07 µg/L diuron EQ(4)
(2 h-PSII)
Acetylcholinesterase inhibition
AChE assay 26 µg/L parathion EQ(2)
DHT = dihydrotestosterone; EEQ = 17β-estradiol equivalent concentration;
PSII = photosystem II.
a
No specific assay indicated.
b
EBT calculated specifically for wastewater effluent.
c
A mixture factor (MF) of 100 was applied (strictly speaking this would then apply as category 2
bioassay).
d
Converted from Org2058 equivalent concentration to levonorgestrel equivalent concentration
using REP in Brand et al. (2013).
References for EBT: 1(Kunz et al., 2015); 2(Escher et al., 2015); 3(van der Oost et al., 2017b);
4
(Escher et al., 2018); 5(Genthe et al., 2009); 6(Brand et al., 2013); 7(Brion et al., 2019);
8
(Jarošová et al., 2014a).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 284 Bioanalytical Tools in Water Quality Assessment

(approach 2 in Figure 13.10). Brand et al. (2013) derived EBTs using the ADI of a
potent reference compound and considered oral bioavailability and the fraction
unbound to plasma as indicators of adsorption and distribution to estimate the
safe internal concentration. Toxicokinetic data, including bioavailability and
fraction unbound to proteins, were also considered for other potent chemicals that
act by the same mode of action. This approach was applied to a battery of
CALUX assays and used REP specific to these assays.
Although in vitro effects can be extrapolated to in vivo for risk assessment by
so-called quantitative in vitro to in vivo extrapolations (Wetmore, 2015; Yoon
et al., 2015), the derivation of EBTs takes the inverse route, using safe
concentration in vivo and extrapolating them to the in vitro situation. This is
typically done for reference compounds, assuming that it is representative for all
chemicals with this effect in the bioassay. Both approaches 1 and 2 are applicable
for drinking water only.
Many other studies have translated existing chemical GVs, both drinking water
GVs and environmental quality standards (EQS), directly into in vitro BEQ or in
vitro bioassay effect thresholds (approaches 3 and 4 in Figure 13.10). These
approaches are suitable for, both, drinking water or surface water GVs.
In the simplest way, the GV is directly translated into the BEQ for a given
bioassay related to the bioassay’s reference compound (approach 3 in
Figure 13.10). For example, Kunz et al. (2015) proposed to use the 17β-estradiol
(E2) annual average environmental quality standard (AA-EQS) of 0.4 ng/L as the
EBT to assess whether the risk for adverse reproductive effects was tolerable
using in vitro assays indicative of estrogenic activity. The E2 AA-EQS was
selected rather than the estrone (E1) or 17α-ethinylestradiol (EE2) AA-EQS
values because E2 is commonly used to express the results of bioassay studies
and because in vitro and in vivo REP values for E1 and EE2 are expressed
relative to E2. Using a similar approach, Leusch et al. (2014a) proposed a
threshold of 0.1 ng/L EEQ for the E-Screen assay and 0.2 µg/L diuron EQ for
photosynthesis inhibition in the green algae Chlorella vulgaris based on
Australian and New Zealand Guidelines for Fresh and Marine Water Quality
(Australian Government, 2018a).
Approach 3 implies that this reference compound is representative for all
chemicals causing the specific effect of the bioassay. This is not necessarily the
case and therefore a number of studies have determined the in vitro effect at the
GV concentration using the different potencies of various bioactive chemicals
(approach 4 in Figure 13.10).
Jarošová et al. (2014a) derived ‘17β-estradiol equivalent concentration (EEQ)
Safe regarding Steroid Estrogens’ (EEQ-SSE) values based on the assumption
that four potent estrogens, E1, E2, estriol (E3) and EE2, explained most
estrogenic activity in wastewater. EEQ-SSE was defined as the EEQ at which no
adverse effects should be observed in municipal effluent based on the in vivo
predicted no effect concentration (PNEC). A literature review was conducted to

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 285

identify general concentrations of the four estrogens in municipal effluent, along

with their REP in common in vitro estrogenic activity assays. The long term
EEQ-SSE ranged from 0.1 to 0.4 ng/L EEQ. Although based on wastewater
effluent, the EEQ-SSE can be applied to surface water receiving effluent after
appropriate dilution factors are considered.
Applying approach 4, Escher et al. (2015) used the Australian Drinking Water
Guidelines (ADWG) (NHMRC, 2011) and the Australian Guidelines for Water
Recycling (AGWR) (NRMMC/EPHC/NHMRC, 2008) to derive EBTs for
assays indicative of receptor-mediated effects. EC values for individual
chemicals with GVs were collected from the literature for each bioassay, then
filtered with EC values over an order of magnitude smaller or larger than the
GV excluded. This was to prevent extremely more potent or less potent
chemicals from skewing and dominating the distribution. The REP of each
filtered chemical was calculated, which normalised the potency of the chemical
to the assay reference chemicals, then the GV was converted to a BEQ by
multiplying the REP and the guideline concentration. The EBT was derived
from the 5th percentile of a cumulative distribution of the BEQ values. The 5th
percentile was selected to be protective for the majority of chemicals, while still
accounting for mixture effects. This process was repeated for each bioassay,
with 11 of the 18 assays (mainly category 1 bioassays but also including some
category 2 bioassays) having sufficient data to derive a preliminary EBT. The
preliminary EBTs were able to differentiate between recycled water for indirect
potable reuse, which was below EBTs, and secondary treated effluent, which
was above.
Following on from this approach, Escher et al. (2018) used current and proposed
AA-EQS values to derive EBTs for 48 in vitro and in vivo assays, with sufficient
data available to derive 32 preliminary EBT. Similar to Escher et al. (2015),
single chemical data were collected from the literature to translate the EQS into a
BEQ value. However, rather than using one algorithm for all assays, different
approaches were used for different classes of bioassays. For category 1 assays
(receptor-mediated effects), the EBT was derived based on the average BEQ of
all chemicals, with a filtering step applied if low potency chemicals were
included to prevent these chemicals reducing the EBT to unrealistically low
levels. In the case of estrogenic activity, where the mixture composition of potent
estrogens often had a similar pattern in environmental samples, an
exposure-corrected approach was applied. The BEQ was multiplied by the
fraction of potent estrogens commonly found in wastewater and surface water of
approximately 11% E2, 9% EE2 and 80% E1 (Kase et al., 2018).
The EBTs in Table 13.1 are sorted according to endpoint and bioassay, with
different columns for drinking and recycled water (i.e., human health-relevant
EBTs) and surface waters (i.e., ecological health-relevant EBTs).
Separate EBTs for drinking water and environmental waters are provided
because the difference in the health target (humans consuming drinking water

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 286 Bioanalytical Tools in Water Quality Assessment

Figure 13.11 Overview of published EBT values for estrogenic activity in units of
17β-estradiol equivalent concentration (EEQ, ngE2/L) for drinking water (circles),
surface water (triangles) and wastewater effluent (squares). See Table 13.1 for
numerical values and more details.

versus wildlife living in and consuming surface water) causes differences in safe
concentrations even if the very same assays can be applied to evaluate both
drinking and surface water quality.
As an example, available EBTs for estrogenic activity are shown in Figure 13.11.
Surface water EBTs are often lower than drinking water EBTs. This is not
unexpected because GVs for surface water tend to be more protective for
specifically vulnerable species than the drinking water GVs, which ‘only’ need to
be protective of human health (see Chapter 3). For example, estrogenic chemicals
cause adverse effects at very low concentrations in aquatic organisms, whereas
terrestrial animals exposed primarily through dietary intake are less adversely
affected (Escher et al., 2018). However, differences between EBTs indicative of
the same assay and the same water type still exist due to differences in the
derivation methods applied.

13.5.2 Approaches to derive effect-based trigger (EBT)

values for category 2 bioassays
Directly reading across from chemical GV is only sensible for category 1 bioassays,
but for category 2 bioassays, where many chemicals can contribute to the observed
effect, the EBT derivation needs to account for mixture effects.
Initial attempts were made to derive the EBTs for category 2 bioassays with the
same approach 4 (Figure 13.12) as for category 1 bioassays (Figure 13.10).
However, the majority of the chemicals had to be filtered out due to low potency

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 287

Figure 13.12 Derivation of effect-based trigger values (EBT) of category 2 bioassays

for drinking water and surface water by different approaches: (4) from an established
GV after filtering out low potency chemicals and incorporating relative potencies (this
is the same approach 4 as for category 1 bioassays in Figure 13.10), (5) as approach
4 but adding a mixture factor (MF) and (6) starting from a generic effect-based trigger
(EBT) for cytotoxicity divided by median of the specificity ratio of the bioassay.

and the filtering step was not clearly defined. The EBTs derived with this method are
marked with ‘low potency filter’ in Table 13.2.
In the next step a mixture factor (MF) was introduced that depended on the
fraction of responsive chemicals in the category 2 bioassay (approach 5 in
Figure 13.12). This MF ranged from 10 to 10,000 as shown in Table 13.2 and is
discussed in detail further below.
Finally, an alternative approach was proposed that did not read across from GV
but took a different point of departure. An EBT for cytotoxicity was derived from
negligible disturbance of cell membranes and then adjusted by a measure that
accounts for the degree of specificity of the chemicals being active in the given
specific endpoints (approach 6 in Figure 13.12).
In line with approach 5, Tang et al. (2013b) derived EBTs for bacterial toxicity
based on bioluminescence inhibition in Aliivibrio fischeri (Microtox assay) by
reading across from the ADWG and AGWR. The EBT was derived based on the
predicted mixture effect of all chemicals in the guidelines using the model of
concentration addition divided by the sum molar concentration of all chemicals in
the guidelines. This EBT included an extrapolation factor to account for the
number of chemicals included in the derivation, model uncertainties and the

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 288 Bioanalytical Tools in Water Quality Assessment

Table 13.2 Summary of proposed effect-based trigger values for category 2

bioassays that are currently available in the literature.

Endpoint//Bioassay Mixture Water Effect-based Trigger

Approach Type EBT-BEQ

Xenobiotic metabolism
AhR activity
AhR-cis- Low potency DW, RW 18 µg/L carbaryl EQ(1)
FACTORIALTM filter
PAH CALUX Read-across SW 150 ng/L B[a]P EQ(2)
MF 100 SW 6.2 ng/L B[a]P EQ(3)
intermediate SW 62.1 ng/L B[a]P EQ(4)
DR CALUX MF 100 SW 0.05 ng/L TCDD EQ(2)
H4L1.1c4 AhR assay MF 100 SW 6.4 ng/L B[a]P EQ(3)
MF 100 SW 4.3 ng/L B[a]P EQ(6)
SR-method SW 250 ng/L B[a]P EQ(5)
PPARγ activity
PPARγ CALUX Read-across SW 10 ng/L rosiglitazone EQ(2)
PPARγ-GeneBLAzer MF 100 SW 36 ng/L rosiglitazone EQ(3)
MF 100 19 ng/L rosiglitazone EQ(6)
SR-method 1.2 µg/L rosiglitazone
EQ(5)
PXR activity
PXR-cis-
FACTORIALTM Low potency DW, RW 59 µg/L metolachlor EQ(1)
filter
PXR CALUX Read-across SW 3.0 µg/L nicardipine EQ(2)
MF 100 272 µg/L DEHP EQ(3)
intermediate corresponding to 54 µg/L
nicardipine EQ
5.4 µg/L nicardipine EQ(4)
HG5LN-hPXR SW 16 µg/L DEHP EQ(3)
Adaptive stress response
AREc32 Low potency DW, RW 284 µg/L dichlorvos EQa(7)
filter
MF 1000 SW 156 µg/L dichlorvos EQ(3)
MF 1000 SW 140 µg/L dichlorvos EQ(6)
SR-method SW 1400 µg/L dichlorvos EQ(5)
Nrf2 CALUX Read-across SW 10 µg/L curcumin EQ(2)
MF 1000 26 µg/L dichlorvos EQ(3)
ARE GeneBLAzer MF 1000 SW 392 µg/L dichlorvos EQ(3)

(Continued )

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 289

Table 13.2 Summary of proposed effect-based trigger values for category 2

bioassays that are currently available in the literature (Continued).

Endpoint//Bioassay Mixture Water Effect-based Trigger

Approach Type EBT-BEQ

Apical effects in well plate-based in vivo assays

Bacterial toxicity
Microtox MF 10,000 DW 4100–4392 µg/L baseline
TEQb(8)
SW 1264 µg/L baseline TEQ(3)
Algal toxicity
72 h algal growth MF 1 SW 0.12 µg/L diuron EQ(3)
inhibition
24 h synchronous MF 1 SW 0.11 µg/L diuron EQ(3)
algae reproduction
Combined algae MF 1 SW 0.13 µg/L diuron EQ(3)
assay (24 h growth)
Invertebrates
48 h Daphnia MF 10 SW 15 ng/L chlorpyrifos EQ(3)
immobilisation test
Fish embryo
Fish embryo toxicity MF 100 SW 276 µg/L bisphenol A EQ(3)
(48 h)
Fish embryo toxicity MF 100 SW 183 µg/L bisphenol A EQ(3)
(96–120 h)
Legend for column ‘Mixture approach’: Low potency filter (approach 4), MF = mixture factor
(approach 5), intermediate = value chosen between two other highly divergent EBT, SR-method
(approach 6 in Figure 13.12), Read-across = (partial) read-across from in vivo data. Water type:
DW = drinking water, RW = recycled water, SW = surface water.
a
Converted to dichlorvos equivalent concentration using dichlorvos EC value in Escher et al.
(2018).
b
Converted to baseline toxic equivalent concentration (baseline TEQ) using the virtual baseline
toxicant EC value in Escher et al. (2018). AhR, aryl hydrocarbon receptor; B[a]P = benzo[a]
pyrene; DEHP = di(2-ethylhexyl)-phthalate; PPARγ = peroxisome proliferator-activated
receptor gamma; PXR = pregnane X receptor; TCDD = 2,3,7,8-tetrachloro-p-dibenzodioxin.
References for EBT: 1(Escher et al., 2015); 2(van der Oost et al., 2017b); 3(Escher et al., 2018);
4
(de Baat et al., 2020); 5(Escher and Neale, 2021); 6(Neale et al., 2020a); 7(Escher et al., 2013);
8
Tang et al., 2013b).

acceptable fraction of chemicals present at their GV. This yielded an EBT-EC50 of 3

in units of relative enrichment factor (REF) for drinking water and REF 2.8 for
recycled water for indirect potable reuse. This means that a drinking water extract
would exceed the EBT if it induced 50% bacterial toxicity after less than three

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 290 Bioanalytical Tools in Water Quality Assessment

times enrichment. Converted to TU, the EBT would be 0.33 TU for drinking water
and 0.36 TU for recycled water. The EC50 values measured in actual recycled water
samples were much higher (i.e., less toxic) than the EBT-EC50 at REF 2.8.
A similar approach was applied by Escher et al. (2013) for the oxidative stress
response assay. A tentative EBT of an effect concentration causing an IR of 1.5
(ECIR1.5) at REF 6 was proposed for both drinking water and recycled water.
Based on the ECIR1.5 value of potent pesticide dichlorvos, this translates to a
dichlorvos equivalent concentration (dichlorvos EQ) of 284 µg/L (Table 13.2).
All EBTs for category 2 bioassays are listed in Table 13.2. They are expressed as
BEQ, although the reference compound used for a particular endpoint can vary
between studies. For example, the EBT for oxidative stress response assays is in
units of dichlorvos EQ in Escher et al. (2018) but in curcumin equivalent
concentrations (curcumin EQ) in van der Oost et al. (2017b). If one knows the
effect concentrations of the reference compounds in both bioassays (and thus the
relative potency of each compound in the assay), these EBTs can be easily
converted.
In approach 5, an MF was included to account for the many chemicals present
and acting together (remember that for category 2 assays, BEQchem ≪ BEQbio).
Initially, an MF of 100 was set for assays indicative of xenobiotic metabolism
and 1000 was used for adaptive stress responses (Escher et al., 2018). The MF
was multiplied by the average BEQ. The MF values were based on experience
with the fraction of effect explained in a particular assay by known chemicals
using iceberg modelling. For example, less than 0.1% of the effect was generally
explained in the oxidative stress response assay (Escher et al., 2013; Neale et al.,
2017b), hence an MF of 1000 was applied.
For apical endpoints and whole organism bioassays, an MF also needs to be
included to account for different degrees of susceptibility. Algae react very
specifically to herbicides that inhibit the photosystem II, and hence any algal
toxicity assay is close to a category 1 bioassay and the MF can be reduced to 1
for most applications. An MF of 10 or smaller was used for Daphnia magna
because they are especially susceptible to insecticides, but the selectivity is not as
high as for herbicides in algae. For the fish embryo toxicity bioassay an MF of
100 was proposed (Escher et al., 2018).
Recognising the subjectivity associated with an MF correction based mainly on
expert knowledge, Escher and Neale (2021) recently proposed a method to derive
the EBT for category 2 assays from experimental data specific for each bioassay
(approach 6 in Figure 13.12), starting with the constant critical membrane
concentration for baseline toxicity (minimum toxicity caused by narcosis, see
Chapter 4) common to all cells (Escher et al., 2019). By definition, mixtures of
baseline toxicants will have very similar cytotoxicity IC10 irrespective of the cell
line used. If 1% cytotoxicity, an effect that is in practical terms indistinguishable
from the negative controls, is accepted as a safe toxicity level, then the EBT-IC10
would calculate to an REF of 10 in all cell lines.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 291

For specific endpoints, the generic EBT-IC10 of REF 10 was then divided by the
median of the distributions of the specificity ratios log SRbaseline of the given
bioassay (Escher and Neale, 2021). The resulting effect concentration can then be
translated to an EBT-BEQ for mixtures in category 2 bioassays by applying the
bioassays-specific EC10 of the associated reference compound (Figure 13.13).
The EBT-BEQ derived using this approach performed much better than previous
ones derived with the MF in differentiating between wastewater, which was
generally above the EBT, and surface water, which was generally below the EBT
(Figure 13.14).

13.5.3 Approaches to derive effect-based trigger (EBT)

values from read-across of in vivo data
In addition to the approaches using GV and in vitro data alone to derive EBTs, in
vivo data can also help to strengthen the EBT derivation. Brion et al. (2019)
derived EBTs for five in vitro estrogenic activity assays using experimental in
vitro and in vivo results for 16 surface water and 17 wastewater extracts. The
response in the four mammalian reporter gene assays and one yeast reporter gene
assay were compared with the whole organism EASZY assay, with true negatives
(i.e., samples with in vitro activity below the EBT and no response in vivo) and
true positives (i.e., samples with in vitro activity above the EBT and an in vivo
response) used to determine specificity and sensitivity, respectively. Logistic
regression models were applied to determine the maximum sensitivity and
specificity cut-off for each assay, and this was used to estimate the EBT.
Although this is based on a limited number of samples, the EBTs are within a
similar range to other estrogenic in vitro assays (Table 13.1). However, this
approach is only valid for endpoints with both in vitro and in vivo assays available.
In an approach that combined in vivo effect data with experimental data, van der
Oost et al. (2017b) applied three different methods to determine BEQ, which were
used to derive EBTs for a battery of CALUX assays indicative of xenobiotic
metabolism (PAH CALUX, DR CALUX, PPARγ CALUX and PXR CALUX),
hormone receptor-mediated effects (ERα CALUX, Anti-AR CALUX and GR
CALUX) and adaptive stress responses (Nrf2 CALUX). A list of chemicals was
selected based on their available toxicity data and reported concentrations in
water, with chemicals with low REP values removed by filtering to prevent them
from biasing the EBT (similar to approach 4). In vivo toxicity data, including no
observed effect concentrations (NOEC), lowest observed effect concentrations
(LOEC) and predicted no effect concentration (PNEC) and effect concentration
causing 50% effect (EC50) were converted to BEQ using the REP, with an
acute-to-chronic ratio of 10 applied to acute data. Safe BEQ, which indicated no
risk to the ecosystem, were derived using the lowest BEQ and dividing by an
assessment factor based on the endpoint. The 5th percentile BEQ (HC5 BEQ),
which indicated low risk, was derived using a BEQ distribution, similar to a

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
292

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Figure 13.13 Derivation of EBT-BEQ with the ‘SR-method’ from the cytotoxicity EBT scaled by the specificity ratio SRbaseline.
Figure modified and reprinted with permission from Escher and Neale (2021). Effect-based trigger values for mixtures of
chemicals in surface water detected with in vitro bioassays. Environmental Toxicology and Chemistry, 40, 487–499. © 2021. The
Bioanalytical Tools in Water Quality Assessment

Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.
by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
Figure 13.14 New EBT-BEQ using the SR-method (red dashed lines) for category 2 bioassays perform much better than the
previously proposed EBT-BEQ derived with an MF of 100 and 1000 (black dotted lines (Escher et al., 2018)) in differentiating
between wastewater (above EBT) and surface water (below EBT). Numerical values of the EBT for (a) AREc32, (b) PPARγ
GeneBLAzer and (C) AhR CALUX are given in Table 13.2. Figure modified and reprinted with permission from Escher and Neale
(2021). Effect-based trigger values for mixtures of chemicals in surface water detected with in vitro bioassays. Environmental
Toxicology and Chemistry, 40, 487–499. © 2021. The Authors. Environmental Toxicology and Chemistry published by Wiley
Periodicals LLC on behalf of SETAC.
Design of test batteries and interpretation of bioassay results
293
 294 Bioanalytical Tools in Water Quality Assessment

species sensitivity distribution approach. Finally, bioanalysis of surface water with

good ecological status was used to determine background BEQ values. An EBT for
each assay was derived based on the evaluation of the safe BEQ, HC5 BEQ and
background BEQ values, with different multiplication factors applied based on
expert judgement. In the case of non-specific effects (e.g., non-specific toxicity to
bacteria and zooplankton), an average acute-to-chronic ratio of 10 and a safety
factor of 2 for assumed 50% recovery by SPE or passive sampling was used to
determine an EC at an REF of 20 or a TU of 0.05. REF takes into consideration
both sample enrichment and dilution in the bioassay, so samples that have a
non-specific effect after 20 times or more enrichment would be considered
acceptable. A similar approach was taken for genotoxicity (Ames, umuC and p53
CALUX), with EBTs derived for non-specific endpoints reduced by an
assessment factor of 10. This gave an EC REF of 200 or a TU of 0.005.

13.6 WHAT TO DO IF A WATER EXTRACT EXCEEDS THE

EFFECT-BASED TRIGGER VALUE?
To facilitate the implementation of EBTs for regulatory purposes, there is a need for
a framework to determine what steps to take if the measured effect in a water sample
exceeds the proposed EBT. Leusch and Snyder (2015) have proposed such a
framework with three tiers of screening, targeted analysis and exploration
recommended. If the measured effect exceeds the EBT during the screening tier,
re-testing is recommended to confirm the results and to determine if this is an
on-going issue. A second tier of targeted analysis of known potent chemicals
with GVs is then recommended if the re-tested sample still exceeds the EBT. If a
detected chemical’s concentration exceeds the GV then operators would need to
follow the well-established chemical GV exceedance response procedure. If the
measured effect is over 10 times higher than the EBT, a number of options are
suggested in a third tier including (1) full chemical analysis of all chemicals in
the relevant guidelines, (2) EDA to identify the causative chemicals or (3)
bench-scale experiments to identify effective treatment methods to reduce effect
detected by bioassay even if the identity of the causative chemicals is unknown.
It should be noted that options 1 and 2 would not be suitable for category 2
assays where many low potency chemicals can contribute to the effect, even if
non-targeted screening approaches can be used to investigate the potential
presence of unknown chemicals and transformation products (Brunner et al., 2019).
van der Oost et al. (2017b) have also suggested a two-tier system consisting of
hazard identification and risk analysis as part of the Smart Integrated Monitoring
approach. If the effect of a water sample exceeds the EBTs in the hazard
identification tier, chemical analysis and EDA are among the steps recommended
in the risk analysis tier.
It is possible to integrate both category 1 and 2 bioassays in an assessment
framework outlined in Figure 13.15. This framework follows essentially Leusch

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 295

Figure 13.15 Framework for application of EBTs in water quality assessment.

Modified from Leusch and Snyder (2015).

and Snyder (2015) for category 1 bioassays and attempts to also integrate category 2
bioassays in a common decision framework.
When an EBT-BEQ is exceeded, it is important to confirm that this result is
correct by a quality control check. If the exceedance of the EBT-BEQ is
confirmed, then it is important to consider the bioassay category. In the case of
category 1 bioassays, targeted chemical analysis of known potent chemicals (Ci)
with GVs should be performed to calculate the BEQchem. If the BEQbio is close to
the BEQchem, that is, most causative chemicals are known, then determination by
chemical analysis of the concentration Ci of individual target chemicals and
comparison with the relevant single chemical GVi will be sufficient to judge

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 296 Bioanalytical Tools in Water Quality Assessment

water quality. If Ci , GVi for all chemicals i, then the water is compliant with
current chemical guidelines, and no further action is warranted. If Ci . GVi, then
the conventional response to chemical GV exceedance is to be followed. Going
back a step, if BEQbio is not close to BEQchem, then this indicates that an
unexpected chemical may be causing the bioassay response. Efforts should be
focused on identifying the new mixture risk drivers – if this effort is successful,
then their concentrations can also be compared to GVs, if available.
If the applied assay is a category 2 assay, then it becomes important to determine
if IC10 is lower (i.e., sample is more toxic) than the cytotoxicity EBT (EBT-IC10).
Exceedance of the EBT but not of EBT-IC10 for non-specific toxicity means that
specific bioassays should be checked to determine if they are also exceeding their
respective EBTs.
In those cases of category 1 bioassays where the chemicals with GVs cannot
explain the BEQbio and mixture risk drivers cannot be identified, then either the
treatment processes need to be optimised to reduce the bioassay response if that
is possible, or the biological quality of the surface water should be assessed, and
the source of the pollutant should be identified. The same should happen of
IC10 , EBT-IC10 for category 2 bioassays.

13.7 CONCLUSIONS
A battery of bioassays indicative of different stages of cellular toxicity pathways
as well as apical effects in whole organisms is recommended to detect the effects
of all active chemicals in a water sample. For routine water quality monitoring, a
practical test battery of assays indicative of activation of AhR, activation of ERα
and oxidative stress response is recommended for wastewater and recycled water
for non-potable reuse, while an additional assay indicative of mutagenicity or
genotoxicity is suggested for drinking water or recycled water for potable water
reuse. The new multiplexed assays are good tools for screening purposes and
for selection of bioassays, but they are generally of too low sensitivity to be
used for routine monitoring purposes. In Chapter 14 several case studies are
outlined that illustrate the applications of the concepts introduced in the
present chapter.
There has been substantial harmonisation on how to interpret in vitro bioassay
results over the last few years. Linking bioassay data to analytically quantified
concentrations has become an important tool not only to understand and identify
mixture risk drivers but also to get a feeling for the relevance and meaning of
different bioassay endpoints. Iceberg modelling has clearly identified that there
are two categories of bioassays – category 1 assays where few chemicals
dominate the mixture effect (BEQchem ≈ BEQbio) and category 2 assays where
many low potency chemicals contribute to the effect (BEQchem , BEQbio).
EBTs are essential to understand the significance of bioassay results and for the
wider acceptance of effect-based monitoring because they can distinguish between

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Design of test batteries and interpretation of bioassay results 297

acceptable and unacceptable chemical water quality. A number of different

approaches have been applied to develop EBTs, including simple translation from
ADIs and chemical GVs, incorporation of chemical potency and mixtures,
comparison of in vitro and in vivo responses to determine maximum sensitivity
and specificity cut-offs (Brion et al., 2019) and using multiple lines of evidence
(van der Oost et al., 2017b). Some of these approaches are only valid for
category 1 assays, while other approaches have been developed specifically for
category 2 assays.
The majority of EBTs were derived for receptor-mediated effects, with
estrogenic activity the most commonly assessed endpoint. There are fewer EBTs
available for induction of xenobiotic metabolism, adaptive stress responses and
apical effects, with the majority of these EBTs derived for surface water rather
than drinking water. EBTs based on surface water GVs are typically lower than
drinking water GVs, so surface water EBTs are likely to be protective of
drinking water.
Despite a plethora of different approaches with differing requirements for expert
opinion being applied, it is remarkable that most of the current EBTs generally end
up within a log unit of one another. EBTs are increasingly applied in the literature to
benchmark water quality and to evaluate treatment efficacy, giving input for
practical frameworks proposed for steps to take should the effect in a sample
exceed the EBT. As of 2020, EBTs have only been implemented by the State
Water Resources Control Board of the State of California (2019), which proposed
monitoring trigger values of 3.5 ng/L EEQ in ERα assays, and 0.5 ng/L
TCCD-EQ for AhR assays for recycled water. There is to date no implementation
in any other legislation but with their maturation and increasing adoption in
research contexts, this will hopefully change in the future.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 14
Case studies

14.1 INTRODUCTION
The number of publications reporting on the application of bioanalytical tools for
water quality assessment is ever increasing. Bioassays have been applied to
evaluate the treatment efficacy of wastewater treatment plants (WWTPs),
advanced water reclamation plants and drinking water treatment plants (DWTPs).
Table 14.1 provides an overview of studies that have applied at least three
different assays, including at least one cell-based assay, to assess water quality in
different water types. Most of these studies were on wastewater and surface water
followed by drinking water (Figure 14.1) with some studies addressing different
water types and also entire treatment trains. Hormone receptor-mediated effects
were the most popular endpoints, often covering a whole suite of different
hormone receptors or different aspects of estrogenicity (Figure 14.1).
Genotoxicity assays and other assays covering reactive toxicity were most
popular with drinking water, due to the known modes of action of disinfection
by-products (DBPs) (Figure 14.1). Apical endpoints from whole organisms
testing and cytotoxicity often complemented the test batteries, although
cytotoxicity was not always measured despite it being recommended to always
assess cytotoxicity in parallel to reporter gene activation measurement.
From the wealth of available publications in Table 14.1, several case studies were
selected for more detailed review to highlight particular applications of in vitro
bioassays in water quality assessment. These case studies cover different water
types from wastewater to drinking water and provide examples of how bioassays
can be used to describe water quality, assess treatment efficacy and for critical

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0299

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 300 Bioanalytical Tools in Water Quality Assessment

Table 14.1 Overview of studies published (until September 2020) that have
applied test batteries of in vitro and in vivo assays indicative of at least three
different endpoints and at least one cell-based assay to a range of water types.

Water XM HRM RM RT ASR AE Ref.

WW, SW 3 6 Sanchez et al. (1988)

WW 2 1 Gagné and Blaise (1998)
WW 1 2 Castillo et al. (2001)
WW 3 Garcia-Reyero et al. (2001)
WW, SW 1 2 1 Dizer et al. (2002)
WW, SW 3 Murk et al. (2002)
WW 1 6 Manusadzianas et al. (2003)
WW, SW 5 Pawlowski et al. (2003)
WW 2 1 2 Aguayo et al. (2004)
SW, DW 2 3 Buschini et al. (2004)
SW, DW 4 2 Guzzella et al. (2004)
WW 1 2 Klee et al. (2004)
WW 1 1 1 4 Pessala et al. (2004)
WW 3 1 Rutishauser et al. (2004)
WW 1 2 Schiliró et al. (2004)
WW 3 Emmanuel et al. (2005)
DW 4 Lah et al. (2005)
WW 1 1 1 Ma et al. (2005)
SW 3 Matsuoka et al. (2005)
SW 1 2 Pillon et al. (2005)
SW 4 2 Zani et al. (2005)
WW, SW 4 Bandelj et al. (2006)
WW 3 4 Fatima and Ahmad (2006)
DW 3 2 Guzzella et al. (2006)
SW 2 2 Keiter et al. (2006)
WW 3 1 Leusch et al. (2006)
DW 2 1 Marabini et al. (2006)
SW 3 2 Pellacani et al. (2006)
WW 2 1 Allinson et al. (2007)
WW 1 1 1 Gustavsson et al. (2007)
WW 1 2 1 Isidori et al. (2007)
DW 3 3 Marabini et al. (2007)

(Continued )

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 301

Table 14.1 Overview of studies published (until September 2020) that have
applied test batteries of in vitro and in vivo assays indicative of at least three
different endpoints and at least one cell-based assay to a range of water types
(Continued).

Water XM HRM RM RT ASR AE Ref.

WW, SW 1 2 Escher et al. (2008a)

WW, SW 1 2 1 2 Escher et al. (2008b)
WW 4 Krishnamurthi et al. (2008)
WW, SW, DW 3 van der Linden et al. (2008)
WW, RW 1 3 Cao et al. (2009)
WW, RW 3 1 2 Escher et al. (2009)
WW 2 4 Gartiser et al. (2009)
SW 3 2 Inoue et al. (2009)
DW 3 3 Maffei et al. (2009)
WW 2 1 Mahjoub et al. (2009)
WW 1 6 Mendonca et al. (2009)
WW 3 Shi et al. (2009a)
WW, RW 2 4 1 Shi et al. (2009b)
WW, RW 2 1 Wu et al. (2009)
WW, SW, DW 2 1 Zegura et al. (2009)
DW 3 Ceretti et al. (2010)
WW 2 4 1 Creusot et al. (2010)
RW 2 1 Farmen et al. (2010)
WW 1 5 Gartiser et al. (2010)
WW 4 Leusch et al. (2010)
WW 4 Li et al. (2010)
WW, RW 1 1 2 1 1 Macova et al. (2010)
WW, RW 2 1 Mnif et al. (2010)
WW 2 1 Rodrigues et al. (2010)
WW 1 1 4 Stalter et al. (2010a)
DW 1 2 Xie et al. (2010)
WW 1 1 2 1 1 Macova et al. (2010)
WW 1 1 2 1 1 Reungoat et al. (2010)
RW 1 1 1 Escher et al. (2011)
WW 3 Kusk et al. (2011)
WW, RW, 1 1 2 1 1 Macova et al. (2011)
SW, DW
(Continued )

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 302 Bioanalytical Tools in Water Quality Assessment

Table 14.1 Overview of studies published (until September 2020) that have
applied test batteries of in vitro and in vivo assays indicative of at least three
different endpoints and at least one cell-based assay to a range of water types
(Continued).

Water XM HRM RM RT ASR AE Ref.

WW 1 4 Stalter et al. (2011)

SW 4 Zhao et al. (2011)
WW 8 Bellet et al. (2012)
WW 4 1 Fang et al. (2012)
SW 1 2 Jarošová et al. (2012)
SW 2 2 Mnif et al. (2012)
DW 1 1 1 Neale et al. (2012)
DW 4 Brand et al. (2013)
SW 1 3 Chinathamby et al. (2013)
SW 1 1 1 Farre et al. (2013)
WW, SW 1 2 Jalova et al. (2013)
Stormwater 1 1 1 1 1 2 Tang et al. (2013a)
WW 1 6 Bain et al. (2014)
SW 1 6 Creusot et al. (2014)
WW, RW, SW, 6 10 2 3 3 3 Escher et al. (2014)
DW, stormwater
WW, RW, SW 2 1 1 1 Leusch et al. (2014a)
WW, RW 7 1 1 Leusch et al. (2014b)
SW 6 Scott et al. (2014)
WW, RW 1 1 1 1 Tang et al. (2014)
WW, SW 1 2 Zounkova et al. (2014)
WW, RW, GW 3 9 1 2 1 2 Jia et al. (2015)
RW, GW 1 2 1 Lee et al. (2015)
WW, RW 3 Mehinto et al. (2015)
SW 2 1 3 1 Neale et al. (2015b)
WW, SW 1 5 Roberts et al. (2015)
DW 1 1 1 Stalter et al. (2016b)
SW 3 Conley et al. (2017a)
SW 3 9 3 König et al. (2017)
SW 1 2 Mehinto et al. (2017)
WW, SW 1 2 2 1 3 1 Neale et al. (2017c)
(Continued )

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 303

Table 14.1 Overview of studies published (until September 2020) that have
applied test batteries of in vitro and in vivo assays indicative of at least three
different endpoints and at least one cell-based assay to a range of water types
(Continued).

Water XM HRM RM RT ASR AE Ref.

SW 5 1 2 Toušová et al. (2017)

WW 3 1 1 1 Välitalo et al. (2017)
SW 3 3 2 3 van der Oost et al. (2017)
SW 1 4 1 Daniels et al. (2018)
SW 4 1 Gao et al. (2018)
WW 4 Gehrmann et al. (2018)
WW, SW 1 4 1 3 Hamers et al. (2018)
SW 2 1 Hashmi et al. (2018)
DW 3 Hebert et al. (2018)
WW 5 Houtman et al. (2018)
WW, SW, DW 13 Leusch et al. (2018b)
SW 2 6 1 Müller et al. (2018)
SW 3 4 1 Neale et al. (2018a)
WW 2 8 2 Nivala et al. (2018)
SW 2 3 3 Novák et al. (2018)
SW, DW 2 3 1 Rosenmai et al. (2018)
SW 6 Scott et al. (2018)
SW, DW 2 2 Shi et al. (2018)
DW 3 Valcarcel et al. (2018)
WW 4 4 1 Alygizakis et al. (2019)
SW 1 2 Brettschneider et al. (2019)
SW 3 3 2 3 de Baat et al. (2019a)
SW 6 Jia et al. (2019)
WW 1 4 2 Lundqvist et al. (2019b)
SW 1 3 Toušová et al. (2019)
DW 2 4 1 1 Albergamo et al. (2020)
SW 4 3 1 2 2 de Baat et al. (2020)
WW, SW 4 Houtman et al. (2020)
WW, SW, DW 3 Medlock Kakaley et al. (2020)
WW, SW 2 1 1 Mueller et al. (2020)
SW 2 1 1 Neale et al. (2020a)
(Continued )

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 304 Bioanalytical Tools in Water Quality Assessment

Table 14.1 Overview of studies published (until September 2020) that have
applied test batteries of in vitro and in vivo assays indicative of at least three
different endpoints and at least one cell-based assay to a range of water types
(Continued).

Water XM HRM RM RT ASR AE Ref.

SW, DW 8 1 3 Neale et al. (2020b)

WW, SW 9 Neale et al. (2020c)
Xenobiotic metabolism: aryl hydrocarbon receptor (AhR); peroxisome proliferator-activated
receptor (PPAR), pregnane X receptor (PXR). Hormone receptor-mediated effects: estrogen
receptor (ER) (agonist and antagonist), androgen receptor (AR) (agonist and antagonist),
glucocorticoid receptor (GR) (agonist and antagonist), progesterone receptor (PR) (agonist and
antagonist), thyroid receptor (TR) (agonist and antagonist), mineralocorticoid receptor (MR)
(agonist and antagonist), retinoic acid receptor (RAR), retinoid X receptor (RXR).
Other receptor-mediated effects: phytotoxicity, neurotoxicity.
Reactive toxicity: genotoxicity, mutagenicity.
Adaptive stress responses: oxidative stress response, p53 response, NF-κB response.
Apical effects: bacterial toxicity, algal growth inhibition, fish embryo toxicity.
Water: Water type; WW: wastewater; RW: recycled water (for direct or indirect drinking water
augmentation); SW: surface water, DW: drinking water; GW: groundwater.
Bioassay type: XM: Xenobiotic metabolism; HRM: Hormone receptor-mediated effects; RM;
Other receptor-mediated effects; RT: Reactive toxicity; ASR: Adaptive stress response; AE:
Apical effects; Ref.: Literature reference.

control point verification monitoring. Following on from Chapter 13, all studies
used test batteries that covered different stages of the cellular toxicity pathway,
with assays relevant for both human and ecosystem health. Furthermore, the
observed effects were compared with surface water and drinking water effect-
based trigger values (EBT) described in Chapter 13 (Section 13.5) in some
case studies.
The first case study used bioassays indicative of hormone receptor-mediated
effects, reactive toxicity and adaptive stress responses to evaluate treatment
efficacy and DBP formation in DWTPs (Neale et al., 2020b) (Section 14.2). The
second case study benchmarked different types of product water for potable and
non-potable reuse (Leusch et al., 2014b) (Section 14.3). This study methodically
applied assessment endpoints selected from a human health relevance
perspective. The third case study applied a bioassay test battery to evaluate
treatment efficacy of different constructed wetlands as well as a conventional
WWTP (Nivala et al., 2018) (Section 14.4). The fourth case study used bioassays
to evaluate the contribution of wastewater effluent to the chemical burden in
small streams and applied iceberg modelling to determine which detected
chemicals are contributing to the observed effect (Neale et al., 2017c) (Section
14.5). The fifth case study applied a multiplex test battery to surface water across
the USA and predicted the mixture effects of the detected chemicals by using the

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 305

Figure 14.1 Overview of studies published that have applied test batteries of in vitro
and in vivo assays until September 2020 (for details see Table 14.1).

exposure activity ratio (Section 14.6). Finally, our last case study on surface water
during rain events compared not only the detected bioassay responses with those
predicted by iceberg modelling from the detected chemicals, but also mixed the
detected chemicals in the detected concentration ratios to test how chemicals act
together in realistic mixtures (Section 14.7).
For additional case studies, the reader is referred to the European technical report
on aquatic effect-based tools under the Water Framework Directive, which includes
further 13 case studies where in vitro and in vivo bioassays have been applied in
water quality monitoring (Wernersson et al., 2015).

14.2 CASE STUDY 1: TREATMENT OF DRINKING WATER

Source waters that feed DWTPs may contain organic micropollutants, and DBPs
may form during treatment processes, such as chlorination. The first case study
aimed to evaluate the efficacy of treatment processes in three DWTPs in the Paris
area, France, using a test battery focusing on hormone receptor-mediated effects
to evaluate micropollutant removal, and reactive toxicity and adaptive stress
responses to assess DBP formation (Neale et al., 2020b). Two of the plants,
DWTP 1 and DWTP 2, applied pre-ozonation (DWTP 2 only), clarification, sand
filtration, ozonation, granular-activated carbon filtration, ultraviolet (UV) and
chlorination, with samples collected from the source water, after UV treatment
and after chlorination. The third plant, DWTP 3, treated 30% of the water using a
similar biological treatment process to the other DWTPs, with 70% of the water
treated using nanofiltration. Samples at DWTP 3 were collected from the source
water, after biological treatment, after nanofiltration and after chlorination.
Samples were collected from all DWTPs over four seasons in 2018.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 306 Bioanalytical Tools in Water Quality Assessment

The water samples were extracted using solid-phase extraction (SPE) and
measured in assays indicative of activation and inhibition of the estrogen receptor
(ERα), androgen receptor (AR), glucocorticoid receptor (GR) and progesterone
receptor (PR), as well as assays indicative of mutagenicity, oxidative stress
response, p53 response and NF-κB response. As low effects were expected in
the product water, the methanolic extracts were exchanged with cell culture media
using the approach described in Chapter 12 (Section 12.6). Concentration–
response curves (CRCs) for activation of ERα are shown in Figure 14.2 for source
and product water from DWTP 1, demonstrating the large potency differences and
the need to apply low-level linear CRCs because the effect threshold of 10% was
only reached after enriching 150-fold. Such a high enrichment means that the
sample is very clean and has no issues with respect to EBT values, but it is still
important to derive an EC10 in order to calculate the treatment efficacy of the DWTP.
Of the studied endpoints, the extracts were only active in assays indicative of
activation of ER, oxidative stress response and NF-κB response (Figure 14.3). No
other hormonal activity was detected in any of the samples, while cytotoxicity
often masked the p53 response in source water. Furthermore, none of the source
or product water samples were mutagenic in the Ames assay using strains
Salmonella typhimurium TA98, TA100 and YG7108 (both with and without
metabolic activation).
Although estrogenic activity was commonly detected in source water and
ranged from 0.17 to 3.98 ng/L 17β-estradiol equivalent concentrations (EEQ), the
treatment processes were able to remove the estrogenic activity, with the effect
in the product water not detectable at the highest relative enrichment factor (REF)
of 100–150 tested (with one exception where the EC10 was just above REF 100).
To determine whether the detected low estrogenic effects in the source water
were of any concern, the estrogenic effects were compared with the EBT-EEQ of
1.8 ngE2/L for drinking water (Escher et al., 2015), which translated into an EC10
of REF 2.4 (Figure 14.3). Since there was no estrogenic effect up to the highest

Figure 14.2 Example full concentration–effect curves for activation of the estrogen
receptor ERα (filled symbols) and cell viability (empty symbols) for source water
(diamond symbols) and product water (triangle symbols) in ERα GeneBLAzer
(agonist mode) (a), along with linear concentration–effect curves for activation of
ERα (b) and cytotoxicity (c). Data from Neale et al. (2020b).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 307

Figure 14.3 Effect concentrations for activation of ER (EC10, orange circles),

oxidative stress response (ECIR1.5, red squares) and NF-κB response (ECIR1.5, red
triangles) in units of relative enrichment factor (REF) for all DWTPs in May 2018.
Data from Neale et al. (2020b).

tested enrichment of REF 100 or 150, the estrogenic effects in the product water
must be at least 80 times lower than the effect at the EBT-EEQ.
The oxidative stress response was often highest in the source water, decreased
after UV treatment and then increased slightly in the product water after
chlorination (Figure 14.3). However, the extracts still needed to be significantly
enriched, between 78 and 136 times in the assay, before an effect could be
detected in both treated and product water. The effect before and after chlorination
was assessed to determine the contribution of formed DBPs to the observed
oxidative stress response based on the approach described in Hebert et al. (2018).
Between 25% and 32% of the oxidative stress response could be attributed to
formed DBPs, showing that most of the effect was due to chemicals already
present in the source water. The oxidative stress response in the product water was
13–17 times lower than the EBT expressed as ECIR1.5 of REF 6 (Escher et al.,
2013). Both EBTs were derived using the Australian Drinking Water Guidelines,
so are not specific for France. However, the large difference between the observed
effect and the proposed EBTs shows the high quality of the final treated water.

14.3 CASE STUDY 2: QUALITY OF RECYCLED WATER

The second case study is a project that applied a battery of bioanalytical tools to
provide a measure of human health risk associated with potable use of recycled
water (Leusch et al., 2014b). The bioassay selection process was started with a
thorough review of the relevant toxicological endpoints associated with negative
human health outcomes from drinking water, conducted in parallel with a review

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 308 Bioanalytical Tools in Water Quality Assessment

of available in vitro methods to monitor these toxicological endpoints (see Chapter

10). Based on this assessment, a battery of protection goal-oriented bioassays was
developed (Table 14.2).
The battery was first benchmarked against a list of 39 priority chemicals
selected based on their likelihood of occurrence in recycled water, the availability
of chemical and bioassay methods to detect them and the scientific and social
perception of their hazard. The priority chemical list covered a range of
chemicals, including natural and synthetic hormones, industrial chemicals,
pharmaceuticals and personal care products, veterinary drugs, pesticides and
DBPs. This benchmarking allowed chemical/group fingerprinting. Estrogen
hormones, for example, were highly estrogenic and anti-androgenic, while also
slightly cytotoxic and genotoxic. Pesticides were found to be slightly estrogenic
with two of the tested chemicals (chlorpyrifos and diazinon) also neurotoxic.
DBPs were slightly genotoxic, cytotoxic and immunosuppressive (Table 14.3).
Benchmarking bioassay effects with a wide range of chemicals is valuable when
applying an integrated testing strategy scheme, where bioanalytical tools are
used to screen samples and direct subsequent chemical analysis.
When benchmarking was complete, grab water samples were collected from
several water reclamation plants with a variety of treatment trains and end uses of
the reclaimed water as well as tap (drinking) water, bottled water and rainwater
collected from a tank in a suburban area. The water samples were extracted and
concentrated using SPE prior to chemical analysis (targeting the 39 priority
chemicals) and testing in the bioassay battery.
Biological activity was found to decrease with increasing level of treatment from
treated wastewater and class A recycled water intended for irrigation (secondary
wastewater that has been subjected to UV and/or chlorination, which remove
pathogens but do not greatly remove chemical contaminants) to reverse osmosis
(RO) treated water, drinking water, bottled water and rainwater (Table 14.4).
Treated wastewater and class A water in particular exhibited significant levels
of estrogenic and progesterone-like activity, most likely due to natural and
synthetic hormones. Overall, the results were in good agreement with chemical
analysis and the effect fingerprint matrix. Interestingly, slight estrogenicity and
anti-estrogenicity were detected in RO-treated water. No chemicals were
identified in the RO-treated samples that could explain this activity. The authors
hypothesised that plasticisers in RO membranes may be implicated as these are
well known to possess low estrogenic endocrine activity.
The results of the chemical analysis show that advanced water treatment could
effectively remove all chemicals tested. The bioassay results confirmed that no
toxic ‘unknown’ chemicals were present in either RO-treated water or drinking
water. This case study shows that bioanalytical tools, along with appropriate
risk assessment, management and communication, have the capacity to facilitate a
significant improvement to the current chemical-by-chemical risk assessment
approach and help to communicate risks to the community.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 14.2 Bioassay test battery applied for benchmarking the human health risks associated with potable use of recycled water
(Leusch et al., 2014b).

Mode of Action Assay Toxicological Endpoint Result Expression

Xenobiotic metabolism
Hepatotoxicity HepCYP1A2 (enzymatic Induction of the multi-function Benzo[a]pyrene EQ (B[a]P EQ)
activity determined by oxidase CYP1A2 in liver cells
luciferase precursor assay)
Hormone receptor-mediated effects
Estrogenicity and ERα CALUX (ERα luciferase ERα mediated transcriptional 17β-Estradiol EQ (EEQ) for
anti-estrogenicity reporter gene assay) estrogenic effect agonist; Tamoxifen EQ (TMXEQ)
for antagonist
Androgenic activity and AR CALUX (AR luciferase AR-mediated transcriptional DHT EQ (DHT EQ) for agonist;
anti-androgenic activity reporter gene assay) androgenic effect Flutamide EQ (Flu EQ) for
antagonist

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Glucocorticoid activityGR CALUX (GR luciferase GR-mediated transcriptional Dexamethasone EQ (Dexa EQ)
Case studies

reporter gene assay) glucocorticoid effect

Progestogenic activity PR CALUX (PR luciferase PR-mediated transcriptional Org2058 EQ (Org2058 EQ)
reporter gene assay) progesterone-like effect
Thyroid activity TRβ CALUX (TRβ luciferase TRβ-mediated transcriptional Thyroid hormone EQ (T3 EQ)
reporter gene assay) thyroid-like effect
Other receptor-mediated effects
Neurotoxicity AChE assay Inhibition of acetylcholinesterase Chlorpyrifos EQ (Chlorpy EQ)
(AChE)
Reactive toxicity
Mutagenicity Ames TA98 and TA100 test, Mutagenic potential Relative genotoxic unit (rGTU)
with and without S9 metabolic
309

activation

(Continued)
by guest
310

Table 14.2 Bioassay test battery applied for benchmarking the human health risks associated with potable use of recycled water
(Leusch et al., 2014b) (Continued).

Mode of Action Assay Toxicological Endpoint Result Expression

Genotoxicity WIL2NS FCMN (flow DNA damage leading to Relative genotoxic unit (rGTU)
cytometry micronucleus test) micronucleus formation
Adaptive stress response
Immunotoxicity: THP1-CPA (IL1β production Modulation of cytokine IL1β PMA EQ (PMA EQ) for stimulation;
stimulation and by THP1 cells, measured by production by monocytes Dexamethasone EQ (Dexa EQ) for
suppression ELISA) (stimulation or inhibition) suppression
Non-specific toxicity
Cytotoxicity Caco2 NRU (neutral red Basal cytotoxicity and decreased Relative toxic unit (rTU)

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

uptake assay with Caco-2 viability of gastrointestinal cells
cells)
Cytotoxicity WIL2NS TOX (flow cytometry Basal cytotoxicity and decreased Relative toxic unit (rTU)
with WIL2-NS cells) viability of white blood cells
Cytotoxicity HepaTOX (resazurin Basal cytotoxicity and decreased Relative toxic unit (rTU)
reduction assay with viability of liver cells
C3A cells)
Bioanalytical Tools in Water Quality Assessment

AR = androgen receptor; CALUX = Chemical Activated Luciferase gene eXpression; CPA = cytokine production assay;
DHT = dihydrotestosterone; ER = estrogen receptor; EQ = equivalent concentration; GR = glucocorticoid receptor;
Org2058 = 16α-ethyl-21-hydroxyl-19-norpregn-4-ene-3,20-dione; PMA = phorbol-12-myristate-13-acetate; PR = progesterone receptor.
 Case studies 311

Table 14.3 Effect fingerprinting for different chemical groups.

Endpoint Horm. Industr. PPCP Vet. Pestic. DBP

(Hepatotoxicity)

ERα (+)

ERα (−)

AR (+)

AR (−)

GR

PR

TRβ

(Neurotoxicity)

Genotoxicity

Immunotoxicity (+)

Immunotoxicity (−)

Cytotoxicity
Hepatotoxicity and neurotoxicity are in brackets to highlight that the associated assays are
incomplete and/or indirect indicators of toxicity, that is, the ‘hepatotoxicity’ assay measures liver
enzyme induction (not necessarily toxicity), and the ‘neurotoxicity’ assay measures
acetylcholinesterase (a relatively limited measure of total neurotoxicity). (+) indicates agonistic
effect. (−) indicates antagonistic effect.
Abbreviations: ER = estrogen receptor; AR = androgen receptor; GR = glucocorticoid receptor;
PR = progesterone receptor; TR = thyroid receptor; Horm. = hormones (natural and synthetic);
Industr. = industrial chemicals; PPCP = pharmaceutical and personal care products; Vet. =
veterinary drugs; Pestic. = pesticides; DBP = disinfection by-products.
The colour coding reflects the potency of the chemicals tested: white = no significant effect,
orange = low-to-moderate effect, red = strong effect (Leusch et al., 2014b).

14.4 CASE STUDY 3: WASTEWATER TREATMENT

In this next case study, a battery of bioassays indicative of xenobiotic metabolism,
hormone receptor-mediated effects and adaptive stress responses was applied to
evaluate effect removal by a conventional WWTP and different constructed
wetlands (Nivala et al., 2018). Constructed wetlands are an option for wastewater
treatment when centralised treatment is not possible, but few studies had
evaluated effect removal by constructed wetlands. Therefore, a test battery
covering different stages of cellular toxicity pathways was assembled

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 312 Bioanalytical Tools in Water Quality Assessment

Table 14.4 Summary of the toxicity results for the different water samples.

Endpoint WW Class A RO DW BW RW

(Hepatotoxicity)

ERα (+)

ERα (−)

AR (+)

AR (−)

GR

PR

TRβ

(Neurotoxicity)

Genotoxicity

Immunotoxicity (+)

Immunotoxicity (−)

Cytotoxicity
Hepatotoxicity and neurotoxicity are in brackets to highlight that the associated assays are
incomplete and/or indirect indicators of toxicity, that is, the ‘hepatotoxicity’ assay measures liver
enzyme induction (not necessarily toxicity), and the ‘neurotoxicity’ assay measures
acetylcholinesterase (a relatively limited measure of total neurotoxicity). (+) indicates agonistic
effect. (−) indicates antagonistic effect.
Abbreviations: ER = estrogen receptor; AR = androgen receptor; GR = glucocorticoid receptor;
PR = progesterone receptor; TR = thyroid receptor; WW = treated wastewater; Class A = Class
A recycled water intended for irrigation only;
RO = RO-treated water, potentially for potable use; DW = drinking water; BW = bottled water;
RW = rainwater.
The colour coding reflects the biological activity detected in the samples: white = no significant
activity, orange = low-to-moderate activity, red = strong activity.

(Table 14.5) and applied to influent and effluent samples from pilot-scale
conventional and intensified constructed wetlands, as well as a conventional
WWTP. In addition to the bioassays, a suite of indicator chemicals, including
pharmaceuticals and food additives, were measured to evaluate chemical removal
efficacy. The indicator chemicals were selected based on their biodegradability.
The majority of samples were active in assays indicative of activation of aryl
hydrocarbon receptor (AhR), binding to peroxisome proliferator-activated

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 313

Table 14.5 Overview of bioassays applied in Nivala et al. (2018).

Mode of Action Assay Positive Reference

Compound

Xenobiotic metabolism
Activation of aryl hydrocarbon AhR CALUX 2,3,7,8-
receptor (AhR) Tetrachlorodibenzo-p-
dioxin (TCDD)
Binding to peroxisome PPARγ GeneBLAzer Rosiglitazone
proliferator-activated receptor
gamma (PPARγ)
Hormone receptor-mediated effects
Activation of estrogen receptor ERα GeneBLAzer 17β-Estradiol
(ER)
Inhibition of estrogen receptor ERα GeneBLAzer Tamoxifen
(ER)
Activation of androgen receptor AR GeneBLAzer Metribolone (R1881)
(AR)
Inhibition of androgen receptor AR GeneBLAzer Cyproterone acetate
(AR)
Activation of glucocorticoid GR GeneBLAzer Dexamethasone
receptor (GR)
Inhibition of glucocorticoid GR GeneBLAzer Mifepristone (RU486)
receptor (GR)
Activation of progesterone PR GeneBLAzer Promegestone
receptor (PR)
Inhibition of progesterone PR GeneBLAzer Mifepristone (RU486)
receptor (PR)
Adaptive stress response
Oxidative stress response AREc32 tert-
Butylhydroquinone
(tBHQ)
NF-κB response NF-κB-GeneBLAzer Tumour necrosis
factor alpha
(TNFα)
Cytotoxicity was measured in parallel for all assays.

receptor gamma (PPARγ), activation of ER, oxidative stress response and NF-κB
response, with the NF-κB response the most responsive endpoint in the study. In
contrast, the effect was often masked by cytotoxicity or there was no effect up to
the maximum tested REF for the other endpoints. Consequently, effect removal
efficacy could only be calculated for five endpoints.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 314 Bioanalytical Tools in Water Quality Assessment

Figure 14.4 Effect removal efficacy (%) after treatment by a conventional wastewater
treatment plant and different constructed wetlands. Data from Nivala et al. (2018).

Effect removal by intensified wetlands was comparable to or better than the

conventional WWTP (Figure 14.4), ranging from 74% to 85% removal of AhR
activity to 98% to 99% removal of estrogenic activity. Between 46% and 69% of
AhR activity was removed after activated carbon filtration and ozonation in a
water reclamation plant (Reungoat et al., 2010), while 80% to .99% removal of
estrogenic activity by WWTPs has been reported (e.g., Jalova et al., 2013;
Houtman et al., 2018). This shows that intensified wetlands can remove
biological activity to a greater extent compared to some conventional WWTPs. In
contrast to the intensified wetlands, the conventional horizontal flow wetland
without aeration had much poorer removal of biological activity, particularly for
PPARγ activity, estrogenic activity and NF-κB activity.

14.5 CASE STUDY 4: SURFACE WATER IMPACTED BY

WASTEWATER TREATMENT PLANT EFFLUENT
This next case study applied a bioassay test battery indicative of cellular toxicity
pathways as well as apical effects in whole organisms to evaluate the contribution
of wastewater effluent to the chemical burden in small streams (Neale et al.,
2017c). Water samples were collected from three sites in Switzerland
(Birmensdorf, Muri and Reinach) with wastewater effluent and surface water
upstream and downstream of the WWTP collected under low flow conditions and
enriched using SPE. The extracts were run in a test battery of assays indicative of
13 endpoints (Figure 14.5), with chemical analysis for 405 chemicals conducted
in parallel. A mass balance approach was applied to determine the fraction of
wastewater effluent downstream of the WWTP, while iceberg modelling (see
Section 13.3 in Chapter 13) was used to link chemical analysis and bioassay results.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 315

Figure 14.5 Bioassay test battery applied in Neale et al. (2017c).

Of the 405 analysed chemicals, 191 were detected at least once. The lowest number
of chemicals and lowest sum of chemical concentrations were detected at the
upstream sites, with pesticides contributing 43%–90% of the total chemical
concentration. In contrast, wastewater effluent had the highest total chemical
concentrations, with corrosion inhibitors (e.g., benzotriazoles) and pharmaceuticals
the main chemical classes detected. A similar profile was observed downstream of
the WWTPs.
For most assays, the response was highest in the effluent samples, followed by
the downstream sites, with the lowest response at the upstream sites. This is
shown in Figure 14.6, where a small effect concentration (EC) indicates a greater

Figure 14.6 Effect concentrations (EC) in units of relative enrichment factor (REF) for
the Muri samples. Figure adapted with permission from Neale et al. (2017c).
Integrating chemical analysis and bioanalysis to evaluate the contribution of
wastewater effluent on the micropollutant burden in small streams. Science of the
Total Environment, 576, 785–795. © 2017 Elsevier.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 316 Bioanalytical Tools in Water Quality Assessment

response. Assays indicative of activation of ER and NF-κB response were among

the most responsive assays, while mutagenicity was the least responsive and the
p53 response was masked by cytotoxicity in all samples.
The mean fraction of effluent downstream of the WWTP (feff) was calculated
using both the detected concentration of individual chemicals (Ci) and the
biological effect, expressed as bioanalytical equivalent concentration (BEQbio).
This was calculated using the upstream, downstream and effluent samples
(Equation 14.1). feff should be the same for both bioanalysis and chemical
analysis based on the assumption of pure mixing of the effluent and no
micropollutant degradation given the small temporal and spatial scale. The
mean feff for chemical analysis ranged from 0.20 to 0.30 for the three sites, with a
mean feff of 0.23 shown by the solid line for the sampling site Muri in
Figure 14.7. This indicates a substantial influence of wastewater effluent on
the downstream site, with a mean dilution of only 4.3-fold. The feff for
bioanalysis ranged from 0.13 for activation of ER to 0.49 for activation of AhR
for Muri, with the feff for activation of AR and 2 h photosystem II (PSII)
inhibition similar to the feff for chemical analysis at 0.22 and 0.24, respectively
(Figure 14.7). The high uncertainty for activation of AhR was due to the small
differences between BEQbio,effluent and BEQbio,upstream. Overall, both chemical
analysis and bioanalysis gave similar feff values, showing that the applied mass
balance approach could be applied to evaluate the influence of wastewater

Figure 14.7 Fraction of wastewater effluent downstream of the wastewater treatment

plant (feff) based on both bioanalysis (coloured bars) and chemical analysis (solid
horizontal line) for Muri. The error bars indicate standard deviation and were
calculated using error propagation. Figure adapted with permission from Neale
et al. (2017c). Integrating chemical analysis and bioanalysis to evaluate the
contribution of wastewater effluent on the micropollutant burden in small streams.
Science of the Total Environment, 576, 785–795. © 2017 Elsevier.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 317

downstream:

Ci,downstream − Ci,upstream BEQbio,downstream − BEQbio,upstream

feff = or feff =
Ci,effluent − Ci,upstream BEQbio,effluent − BEQbio,upstream
(14.1)
The contribution of detected chemicals to the observed effect was assessed using
iceberg modelling. Between 45% and 108% of 2 h PSII inhibition could be
explained by the detected herbicides (Figure 14.8), with the majority of the effect
explained by PSII inhibitors diuron and terbuthylazine. A similarly high fraction
of the effect was also previously explained by detected chemicals for
photosynthesis inhibition (Bengtson Nash et al., 2006; Escher et al., 2011; Tang
and Escher, 2014). Up to 30% of the effect could be explained by four chemicals
in the activation of AhR assay, with most of the effect explained by fungicide
propiconazole. Of the 191 detected chemicals, 135 chemicals were tested in the
ToxCast database for the oxidative stress response assay, with 26 of the
chemicals reported to be active, that is, an EC could be derived. Despite a large
number of detected chemicals being active, only 0.06%–1.9% of the oxidative
stress response could be explained by the detected chemicals (Figure 14.8).

Figure 14.8 Comparison of bioanalytical equivalent concentrations from bioanalysis

(BEQbio) with bioanalytical equivalent concentrations from chemical analysis
(BEQchem) for 2 h PSII inhibition in algae, activation of AhR and oxidative stress
response. Figure adapted with permission from Neale et al. (2017c). Integrating
chemical analysis and bioanalysis to evaluate the contribution of wastewater
effluent on the micropollutant burden in small streams. Science of the Total
Environment, 576, 785–795. © 2017 Elsevier.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 318 Bioanalytical Tools in Water Quality Assessment

Typically, less than 1% of the oxidative stress response can be explained by detected
chemicals (Tang et al., 2014; Yeh et al., 2014; Neale et al., 2015b, 2020a).
The large difference in the fraction of effect explained for the three assays is
related to the assay category. The 2 h PSII inhibition assay is an example of a
category 1 assay, where a small number of highly potent chemicals are active,
PSII-inhibiting herbicides in the current example. Other examples of category 1
assays include assays indicative of receptor-mediated effects, such as activation
of ER (Murk et al., 2002; Rutishauser et al., 2004; Könemann et al., 2018) and
activation of GR (Schriks et al., 2010b; Jia et al., 2016).
In contrast, the oxidative stress response assay is an example of a category 2
assay, which detects more integrative effects and many low potency chemicals
can contribute to the effect. A large number of chemicals can have an effect in
the oxidative stress response assay, so measuring more chemicals to increase the
percentage of effect explained is not practically feasible. Other examples of
category 2 assays include some assays indicative of xenobiotic metabolism, such
as activation of pregnane X receptor (PXR) (Creusot et al., 2014), and assays
indicative of apical effects, such as fish embryo toxicity (Neale et al., 2015b) and
bacterial toxicity (Tang et al., 2013b). This case study helps to highlight the
importance of applying bioassays alongside chemical analysis to gain a better
understanding of the chemical burden in surface water and wastewater.

14.6 CASE STUDY 5: BENCHMARKING SURFACE

WATER QUALITY ACROSS THE USA
Blackwell et al. (2019) evaluated the bioactivity of 38 rivers in the USA with the
multiplex Attagene FACTORIALTM battery that included 96 different endpoints
(see Chapter 13.2.2 for more information on this assay and its typical response in
surface water). PXR- and AhR-related endpoints were most commonly activated
in these samples, followed by ERα- and PPARγ-related endpoints (Figure 14.9)
and the observed effects varied widely across sampling sites with sites impacted
by wastewater high in ERα- and GR-activity while AhR and PXR were instead
associated with urban activity but also with natural organics from swamps
(Blackwell et al., 2019).
The same samples were also measured with reporter gene assays for ERα, AR
and GR (Conley et al., 2017a). More samples were active in the reporter gene
assays than in the FACTORIALTM endpoints, pointing to a higher sensitivity of
the reporter gene assays. The active samples showed a good correlation between
the estrogenic response in the T47D-KBluc reporter gene assay and ERE_cis and
ERa_trans of the FACTORIALTM assay (Figure 14.10).
In addition, more than 700 micropollutants were quantified in these samples
(Bradley et al., 2017). The mixture effects of the detected chemicals in a water
sample Ci were evaluated with the exposure activity ratio EARmixture, which is

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 319

Figure 14.9 Heat map showing effects expressed as AUC for commonly activated
endpoints of the Attagene cis-FACTORIALTM and trans-FACTORIALTM assay in
surface water extracts. Data from Blackwell et al. (2019). AUC = area under the curve.

the ratio of the exposure concentrations and the ECs, in this case the activity
concentrations at cut-off (ACCi) from the ToxCast database (Equation 14.2).
The EARi of all detected bioactive chemicals were then summed up to obtain
the EARmixture. In analogy to the mixture risk quotients, that is, the risk index
(Chapter 8), if the EARmixture exceeds 1, then the detected chemicals are present
at concentrations where together they could elicit a bioassay effect. The EARi of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 320 Bioanalytical Tools in Water Quality Assessment

Figure 14.10 Comparison of the EEQ measured with a reporter gene assay
(T47D-KBluc) (Conley et al., 2017a) with the predicted exposure activity ratio
EARmix for the Attagene FACTORIALTM endpoints ERα-trans and ERE_cis and the
measured AUC of the samples in the Attagene FACTORIALTM endpoints
ERα-trans and ERE_cis (right y-axis). Data from Blackwell et al. (2019).

the individual compound can be evaluated to identify mixture risk drivers:

n
exposurei

n
chemical concentration Ci
EARmixture = = (14.2)
i=1
activityi i=1
ToxCastACCi

The EARmixture for the estrogenicity endpoints were above 1 for those sites where
a high effect (expressed as area under the curve, AUC) was also detected
(Figure 14.10).
A tiered approach would be useful in future studies: in a first step the relevant
endpoints could be identified with the multiplex Attagene FACTORIALTM
assays and then these relevant endpoints could be quantified using more sensitive
single-endpoint reporter gene assays.

14.7 CASE STUDY 6: BENCHMARKING SURFACE WATER

QUALITY IN SMALL STREAMS DURING RAIN EVENTS
The final case study describes water quality monitoring triggered by rain events of
44 small rivers all across Germany (Neale et al., 2020a). Despite the sites being
selected mainly based on anticipated agricultural impact, chemical markers of
untreated wastewater and estrogenic effects were high during some rain events,
likely indicating sewer overflow. Road run-off also impacted many of the rain
events with high concentrations of chemicals typical of tyre wear. The diversity
of the events was striking with some sites having two to four events that often
showed little similarity with respect to chemical composition, overall
concentration, effect magnitude and effect pattern (Neale et al., 2020a).
In this case study, in addition to a battery of bioassays (ERa GeneBLAzer, AhR
CALUX, PPARγ GeneBLAzer and AREc32) and chemical analysis of close to

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Case studies 321

Figure 14.11 Iceberg modelling in this case study was complemented by designed
mixture experiments of mixtures of the detected bioactive chemicals (figure
adapted from Figure 13.4 in Chapter 13). CA = concentration addition.

400 chemicals, designed mixture experiments were performed with the chemicals
that were detected in the samples. The bioanalytical equivalent concentration
measured in the designed mixtures BEQbio,tip was quantified by the defined
mixture experiments (Figure 14.11) in addition to just comparing the bioanalytical
equivalent concentration from the bioassays BEQbio with the bioanalytical
equivalent concentration from chemical analysis BEQchem (Chapter 13).
Despite the fact that almost 400 chemicals were included in the analytical
method, the predicted mixture effects of the detected chemicals (BEQchem) only
explained between 0.01% and 1% for the category 2 bioassays AhR CALUX,
PPARγ GeneBLAzer and AREc32, with only a few samples having a higher
fraction of effect explained (Figure 14.12a-c).
If the 2 to 14 chemicals that were, both, detected and active in the tested bioassays
were mixed in the concentrations as they occurred in the samples and tested in the
bioassays, there was an excellent agreement between mixture effects BEQchem,tip

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 322 Bioanalytical Tools in Water Quality Assessment

Figure 14.12 (a-c) Comparison of the measured bioanalytical equivalent

concentrations BEQbio and the predicted mixture effect of the detected chemicals
expressed as BEQchem: (a) B[a]P EQchem only explains a small fraction of B[a]P
EQbio in AhR CALUX. (b) Rosiglitazone EQchem only explains a small fraction of
Rosiglitazone EQbio in PPARγ GeneBLAzer. (c) Dichlorvos EQchem only explains a
small fraction of Dichlorvos EQbio in AREc32. (d-f) Comparison of the predicted
mixture effects BEQchem,tip and the mixture effect of the designed mixtures
BEQbio,tip for (d) AhR CALUX, (e) PPARγ GeneBLAzer and (f) AREc32. Reprinted
with permission from Neale et al. (2020a). Assessing the mixture effects in in vitro
bioassays of chemicals occurring in small agricultural streams during rain events.
Environmental Science & Technology, 54(13): 8280–8290. © 2020 American
Chemical Society.

predicted with the model of concentration addition of the mixture components and
their measured mixture effect BEQbio,tip (Figure 14.12d-f). These mixture
experiments confirmed that the BEQ concept is valid and can be applied to
environmental samples independent of the concentration ratio in the mixture and
the effect potencies of the individual components.
In this study, we also took a closer look at the known fraction of effects, the ‘tip of
the iceberg’. If we consider 15 chemicals that have experimental effect data for
single chemicals and were detected frequently (but not in all rain event samples),
their composition and overall concentrations varied substantially. The AhR
CALUX was highly dominated by the urban herbicide diuron at a few sites

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
Case studies

Figure 14.13 Contribution of 15 chemicals to the BEQchem in the AhR CALUX assay; % activation of AhR explained =
BEQchem/BEQbio. Data from and detailed description of the sampling sites in Neale et al. (2020a).
323
 324 Bioanalytical Tools in Water Quality Assessment

(Figure 14.13), which could explain as much as 1.1% and 2.8% of the mixture effect
of the known chemicals in two of the samples. At another site, 7-diethylamino-4-
methylcoumarin, which was consistently present at high concentration during
four rain events, alone explained 0.3%–0.8% of the biological effect of the
sample. Interestingly, 2-benzothiazolesulphonic acid, a rather inconspicuous
transformation product of mercaptobenzothiazole and chemical used on the
production of rubber, had a consistently relevant contribution to both activation
of the AhR and oxidative stress response. There was a much more variable
contribution of many chemicals of diverse origin at most sites, and most other
chemicals were fairly specific for one or the other assay (Neale et al., 2020a).
This means that although in each sample there were often few chemicals
dominating the mixture effects of the known chemicals, the dominant chemicals
were in many cases site-specific. Minor contributing chemicals, in contrast, were
often found across many samples.
This case study demonstrates the need to complement chemical analysis with
bioanalytical tools, especially during random events (such as rain events), where
unexpected chemicals might be released and where the bioassays provide a
reliable sum parameter for chemical pollution.

14.8 CONCLUSIONS
Bioanalytical tools have been widely applied in the literature, with studies focusing
on wastewater, surface, drinking and recycled water. The large number of case
studies in Table 14.1 is encouraging, especially the large number of additional
applications and diversity of applications since the first edition of this book 10
years ago. Every application demonstrates the progress but also the limitations of
the bioanalytical approach to water quality assessment and how computational
analyses and concurrent analytical quantification of micropollutants strengthen
the studies.
Several case studies covering different water types and different applications
were discussed in more detail in this chapter to show how bioassays can be
applied to monitor water quality, assess treatment efficacy and evaluate the
effectiveness of critical control points. Some of the case studies demonstrated
how water quality can be benchmarked against existing EBTs, or how iceberg
modelling can help understand drivers of toxicity. The case studies also highlight
the advantage of applying bioassays alongside chemical analysis. These case
studies can only demonstrate a small part of the potential bioanalytical tools have
for the application in water quality monitoring.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Chapter 15
Application of bioanalytical tools
beyond water: Sediment and biota

15.1 INTRODUCTION
This book focuses on the application of bioanalytical tools to water samples, but
bioanalytical tools are being applied to an ever-increasing range of environmental
samples. Essentially, any mixture of micropollutants can be characterised by
bioanalytical tools provided it can be extracted unaltered from the sample. As the
matrices tested become more varied, the separation of organic micropollutants
from the matrix becomes more and more important to ensure that matrix effects
do not affect the bioassay results.
Organic chemicals commonly detected in water are also, of course, detected in
biota, our food and our bodies (CDC, 2019). We have to look at the environment
as an interconnected system where chemicals move between the different
compartments of water, air and earth, become enriched along the food chain, and
eventually end up in people (Escher et al., 2020).
Condensed matrices (solid samples) such as soil, sediment, plants, biota and
tissue samples pose a particular challenge with respect to sampling and sample
preparation. While most waters can be tested directly (Chapter 3) or only undergo
fairly simple enrichment/extraction without any clean-up (Chapter 12), this is not
possible for condensed matrices.
Traditionally, chemicals in such matrices have been extracted using (accelerated)
solvent extraction on previously freeze-dried samples. Coextracted lipids and matrix
components require tedious clean-ups, including gel permeation chromatography
and acid digests, to remove the natural matrix and to isolate the micropollutants
of interest. Blood can be relatively easily extracted with a simple protein

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0325

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 326 Bioanalytical Tools in Water Quality Assessment

precipitation with acetonitrile aided by salting out followed by minimal clean-up,

the so-called QuEChERS (quick, easy, cheap, effective, rugged and safe)
extraction method (Baduel et al., 2015; Plassmann et al., 2015). QuEChERS has
also been applied to food items and fish tissue (Ramalhosa et al., 2009; Forsberg
et al., 2011; Norli et al., 2011; Jakimska et al., 2013; Cloutier et al., 2017),
which required more extensive clean-up.
Harsh clean-up results in the removal of a substantial fraction of chemicals and
potentially alters the composition of the samples. The challenge for condensed
matrices is to use sample preparation and clean-up methods that are not
discriminating and not destructive. Various methods have been optimised for
chemical analysis but most of them have low recoveries or are selective for
certain chemical groups and therefore are not suitable for sample preparation
for in vitro bioassays. Effect-directed analysis (EDA, Chapter 13) that is typically
used to fractionate samples to identify mixture effect drivers also has the
advantage of partially removing matrices and endogenous chemicals that can
mask the effect of pollutants.
In situ and ex situ equilibrium passive sampling techniques can be used as a
one-step sampling and sample preparation method for sediments and biota.
Silicone (polydimethylsiloxane, PDMS) in particular offers many advantages
because the partition coefficients are fairly independent of the hydrophobicity of
the chemicals for most non-aqueous matrices (Jahnke et al., 2016) and kinetics are
reasonably fast in sediments (Li et al., 2013), lipid-rich tissues (Jin et al., 2013)
and blood (Jin et al., 2015a). In addition, due to limited matrix interference,
extracts from silicone passive samplers can often be used directly in chemical
analysis and bioassays without additional or minimal clean-up. The polymer phase
may thus serve as a reference phase that is equilibrated with complex media to
extract a defined fraction of a sample without changing its composition.
In this chapter, we give a brief summary of applications of bioanalytical tools to
non-aqueous matrices, including air and dust, condensed environmental matrices
(suspended particulate matter (SPM), sediment and soil), biota (fish tissue, marine
mammal tissue) and in human biomonitoring. We focus on studies that have
attempted to measure complex mixtures of organic chemicals in an unbiased
fashion. In the past, in vitro bioassays were applied to quantify dioxin-like and
other persistent chemicals in biospecimen after solvent extraction and acid
digestion but the true mixture effects in a sample are not accessible with these
methods, and we therefore did not include those types of studies here.

15.2 SUSPENDED PARTICULATE MATTER, SEDIMENT

AND SOIL
15.2.1 Suspended particulate matter
SPM is probably the most natural follow-up phase after water to investigate with
bioanalytical tools because SPM is in intimate contact with the water phase,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Application of bioanalytical tools beyond water 327

continuously exchanging chemicals. The importance of sorbed chemicals can be

strikingly demonstrated by an example from a study where water and associated
particulate matter were sampled during a storm event (Mueller et al., 2021). With
SPM concentrations less than 1 g/L, there was about 1000 times less mass of
SPM than water, but the associated bioanalytical effect fluxes (BEF) were
approximately equal in water and SPM. This suggests that SPM carries a
significant chemical load associated with significant bioactivity (Figure 15.1). This
is, however, an extreme situation. During dry weather, the SPM concentration is
much lower, and the associated effect load of SPM is hardly measurable. But this
example shows the potential relevance of SPM. Despite of this, remarkably few
studies have addressed SPM and, as we noted in Chapter 12, not enough attention
has been given to filtration when developing water extraction methods.

15.2.2 Sediments
The hazard potential of sediments (Burton et al., 2000) can be assessed after total
extraction of sediments and submitting them to batteries of in vitro bioassays
(Boehler et al., 2017). Given the complexity of the matrix and mixture of
contaminants in sediment, EDA has been often applied to identify bioactive
components in sediments (Brack et al., 1999, 2005; Li et al., 2018, 2019).
Interestingly, the in vitro effects in a polluted river site were dominated by the
polar fractions in EDA (Luebcke-von Varel et al., 2011) but at similar sites
polycyclic aromatic hydrocarbons and legacy persistent organic pollutants mainly
contributed to the activation of AhR (Otte et al., 2013). These and other case

Figure 15.1 Equal role of the bioanalytical effect fluxes (BEF) in water and SPM
expressed as dichlorvos BEF measured with AREc32 for oxidative stress response
over the time course of a storm event despite the water flux Q and the SPM flux
differed by more than a factor of 1000. The BEF is calculated by multiplying the
water or SPM flux with the measured bioanalytical equivalent concentration (BEQ)
with permission from Mueller et al. (2021). Storm event-driven occurrence and
transport of dissolved and sorbed organic micropollutants and associated effects in
the Ammer River, Southwestern Germany. Environmental Toxicology and
Chemistry, 40(1): 88–99. © 2020 The Authors. Environmental Toxicology and
Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 328 Bioanalytical Tools in Water Quality Assessment

studies demonstrate the complexity and variability of contaminated sediments and

how useful bioanalytical tools can be for monitoring sediment contamination.
Bioavailability is decisive in both soils and sediments to determine if there is an
environmental risk (Reid et al., 2000; Semple et al., 2004; Mayer et al., 2014).
Therefore, biological assessment initially relies on direct contact testing with
sediment-dwelling organisms (Tuikka et al., 2011; de Baat et al., 2019a, b).
Application of bioanalytical tools is possible if the bioavailable fraction is
extracted with passive sampling methods (Jonker et al., 2020), and there is also
the option of performing EDA after extractions of the bioavailable fraction (You
and Li, 2017).
Numerous studies have performed bioanalytical assessment of total sediment
extracts and ex situ passive sampling with silicone and a selection of results for
three endpoints is given in Table 15.1. The concentrations per kg silicone can be
almost converted one-to-one to concentrations per kg organic carbon (OC) (Li
et al., 2013; Niu et al., 2020), hence it is possible to compare the total with the
bioavailable effects expressed as bioanalytical equivalent concentrations (BEQbio,
for definition see Chapter 7).
Li et al. (2013) compared the BEQ in total sediment extracts with the BEQ in
extracts from PDMS ex situ passive sampling of sediment. They demonstrated
that the ‘bioavailable effects’ matched quite well with the effects from total
extracts for herbicidal activity and oxidative stress response but were
substantially smaller for assays that are triggered by more hydrophobic chemicals
such as the activation of AhR. Based upon these experiences and additionally
accounting for binding to black carbon, Braunig et al. (2016) developed a
bioanalytical effect balance model that allowed the differentiation between total
effect burden and that associated with organic matter and black carbon
(Figure 15.2a). The model was applied to sediments from different sources from
urban to pristine river sites. In the AhR CAFLUX, the bioavailable fraction
(BEQOC) contributed only 1%−11% (BEQOC/(BEQOC + BEQBC)) to the total
effect burden (Figure 15.2b) at all sites, some of which had strong urban impact,
except at one pristine reference site where the contribution from bioavailable
TCDD equivalent concentration (TCCD EQOC) dominated the overall effect. The
situation was different for the AREc32, which responds to chemicals activating
the oxidative stress response (Figure 15.2b). Here the contribution from the
aqueous phase became important with 7%−26% of effect burden expressed as
tBHQ EQ (BEQw/(BEQw + BEQOC + BEQBC)). The contribution of OC to the
total sorbed chemical mixtures (BEQOC/(BEQOC + BEQBC)) was higher with
7%−24% compared to the 1%−11% for the AhR CAFLUX described above.
This model underpins the bioassay-specific fate of effects in sediment−water
systems, which is relevant for risk assessment.
Vethaak et al. (2017) later proposed a similar approach and applied total solvent
and PDMS extracts to five in vitro bioassays (DR-Luc, ERLuc, AR-EcoScreen, a
transthyretin (TTR) binding assay and Aliivibrio fischeri bioluminescence

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
by guest
Table 15.1 Global comparison of bioanalytical equivalent concentrations (BEQbio) of exhaustive sediments and PDMS extracts with
AhR CALUX, AREc32 and PPARγ GeneBLAzer.

AhR Assay

Sampling Site BEQ of Total Extract BEQ of PDMS Ref. Lit.

(ngref/ kgsed,dw) Extract
(ngref/ kgPDMS)
Raw Clean-up
Sample Sample
Sweden 83.5–143,640 755–90,000 TCDD 1
Coastal Svalbard and offshore deep sea 88.1–1877 316–1030 TCDD 1
Rivers/coastal areas in Queensland, 21.4–506 222–5450 TCDD 1
Australia
French−German river catchment 499–18,668 535–8500 TCDD 1
Brisbane River, Australia 10–927 122–1186 TCDD 2

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Gladstone Harbour, Australia 284–12,198 1880–125,641 TCDD 3
(wet basis)
Lake Tai, China 20–300 ,110–6000 TCDD 4
Haihe River, China 330–930 TCDD 5
Dagu River, China 1200–13,900 TCDD 5
Wenyu River, China 8.5–336 2.7–63.8 TCDD 6
Application of bioanalytical tools beyond water

Yellow Sea region 0–28 TCDD 7

Lake Shihwa and Masan Bay, Korea 14–868 TCDD 8
River Elbe, Germany 15.5–322 TCDD 9
Pohang Area, Korea 0–800 TCDD 10

(Continued)
329
by guest
Table 15.1 Global comparison of bioanalytical equivalent concentrations (BEQbio) of exhaustive sediments and PDMS extracts 330
with AhR CALUX, AREc32 and PPARγ GeneBLAzer (Continued ).

AhR Assay
Sampling Site BEQ of Total Extract BEQ of PDMS Ref. Lit.
(ngref/ kgsed,dw) Extract
(ngref/ kgPDMS)
Raw Sample Clean-up
Sample
UK estuaries 1100–177,000 1–106 TCDD 11
Tietê River, Brazil 160–24,170 TCDD 12
North Sea, south-western Baltic Sea and 20–3493 97–7257 TCDD 13
western Mediterranean
West coast of South Korea 0–57,000 B[a]P 14
Beijing-Hangzhou Grand Canal 25.2–208 13.4–118 TCDD 15

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Qiantang River 17.4–24.0 4.68–14.0 TCDD 15
Beijing-Hangzhou Grand Canal 66.3–545 35.3–311 B[a]P 15
Qiantang River 45.1–62.9 12.3–36.8 B[a]P 15
AREc32 Assay

Sampling Site BEQ of Total Extract BEQ of PDMS Ref. Lit.

(mgref/ kgsed,dw) Extract
Bioanalytical Tools in Water Quality Assessment

(mgref/ kgPDMS)

Sweden 6–1262 25–359 tBHQ 1

Coastal Svalbard and offshore deep sea 10–201 14.1–79.0 tBHQ 1
Rivers/coastal areas in Queensland, 3.1–11.3 19.0–93.5 tBHQ 1
Australia
by guest
French−German river catchment 26–283 33.5–257 tBHQ 1
Brisbane River, Australia 79–564 1666–4591 tBHQ 2
Gladstone Harbour, Australia 5.7–31.3 (ww) 59.0–205 tBHQ 3
Beijing-Hangzhou Grand Canal 7.04–18.0 8.9–21.0 tBHQ 15
Qiantang River 3.85–6.25 10.2–22.0 tBHQ 15
PPARγ GeneBLAzer Assay

Sampling Site BEQ of Total Extract BEQ of PDMS Ref. Lit.

(μgref/ kgsed,dw) Extract
(μgref/ kgPDMS)

Sweden 1.0–186 28.3–286 Rosiglitazone 1

Coastal Svalbard and offshore deep sea 2.3–7.1 9.9–91.0 Rosiglitazone 1
Rivers/coastal areas in Queensland, 1.7–5.1 20.2–29.0 Rosiglitazone 1
Australia

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

French−German river catchment 3.3–29.6 31.5–298 Rosiglitazone 1
Beijing−Hangzhou Grand Canal 2.72–9.60 5.08–28.4 Rosiglitazone 15
Qiantang River 0.94–4.34 1.70–10.5 Rosiglitazone 15
Reprinted with permission from Niu et al. (2020). Mixture risk drivers in freshwater sediments and their bioavailability determined using passive
equilibrium sampling. Environmental Science & Technology, 54(20): 13197–13206. © 2020 American Chemical Society.
Abbreviations: dw = dry weight; ww = wet weight; Ref. = reference chemical. Lit. = literature reference: 1Jahnke et al. (2018); 2Li et al. (2013);
3
Braunig et al. (2016); 4Li et al. (2016); 5Song et al. (2006); 6Luo et al. (2009); 7Hong et al. (2012); 8Yoo et al. (2006); 9Otte et al. (2013); 10Hong et al.
Application of bioanalytical tools beyond water

(2014); 11Hurst et al. (2004); 12Rocha et al. (2010); 13Vethaak et al. (2017); 14Jeon et al. (2017), 15Niu et al. (2020).
331
by guest
332

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

Figure 15.2 (a) Conceptual model describing the relationships and equilibria in a sediment system. Gray circles represent extracts
used in the study. Bold arrows refer to exhaustive extractions, two-way arrows refer to equilibrium partitioning, while dashed arrows
indicate limited exchange between phases. BEQ = bioanalytical equivalent concentration; SPE = solid-phase extraction; ASE =
accelerated solvent extraction; K = partition constant; PDMS = polydimethylsiloxane (silicone) extraction; BC = black carbon;
OC = organic carbon; IC = inorganic carbon, f = fraction. (b) Contribution of water, OC and BC to the total sediment
contamination quantified with AhR CALUX and AREc32. TCDD EQ=TCDD equivalent concentration, tBHQ EQ =
Bioanalytical Tools in Water Quality Assessment

t-butylhydroquinone equivalent concentration. The lines connect data from the same sampling site. Figure (a) reprinted with
permission from Braunig et al. (2016). Bioanalytical effect-balance model to determine the bioavailability of organic contaminants
in sediments affected by black and natural carbon. Chemosphere, 156: 181–190. © 2016 Elsevier.
 Application of bioanalytical tools beyond water 333

inhibition) and one in vivo assay on the embryo toxicity of the sea urchin. They
confirmed the earlier results qualitatively but did not convert the BEQPDMS in
BEQOC, so a retrospect effect balance was not possible. They compared the
bioassay results with chemical analysis and indicated that the detected chemicals
could not entirely explain the effects detected with the bioassays.
Two more recent studies evaluated Chinese sediments (Niu et al., 2020) and
world-wide sampled sediments (Jahnke et al., 2018; Muz et al., 2020) using both
total accelerated solvent extraction and PDMS and confirmed that typically less
than 10% of the effects and BEQ were bioavailable (Table 15.1).
The responses in the AhR CALUX, PPARγ GeneBLAzer and AREc32 assay
expressed as BEQbio and TUbio were compared with the predicted effect BEQchem
and TUchem (for definition see Chapter 8) from chemical analysis of up to 650
organic chemicals with up to 420 chemicals detected (Niu et al., 2020). Despite
of this massive analytical effort only a small fraction of effects (BEQchem/BEQbio
and TUchem/TUbio) could be explained by the detected chemicals: 0.1%–28% in
whole sediment and 0.009%–3.3% in the bioavailable fraction of the Chinese
sediments (Niu et al., 2020). Similarly, less than 10% effect in the bioavailable
fraction of the sediments sampled worldwide could be explained by the detected
chemicals with the exception of only one site in Sweden, where benzo[a]pyrene
and benzo[k]fluoranthene together explained more than 30% of the effect in
AREc32 (Muz et al., 2020).
These studies highlight how important it is to not only analyse chemicals in
sediments but also assess the mixture effects with bioanalytical tools for a
comprehensive understanding of the immediate risk (bioavailable effects from
PDMS extracts) but also the total hazard potential (mixture effects from
exhaustive extracts).

15.2.3 Soil
Soil has received far less attention with bioanalytical tools (Xiao et al., 2006). The
few studies available focus on soils in connection with the application of sewage
sludge to agricultural soils (Liu et al., 2014) or soil of e-waste recycling sites
(Shen et al., 2008) or other contaminated areas (Sidlova et al., 2009; Lam et al.,
2018).

15.3 PARTICLES IN AIR AND DUST

Several studies have applied bioanalytical tools to test particle extracts from air and
dust. Most studies initially focused on genotoxicity and mutagenicity of air
particulate material (Lewtas et al., 1990; Isidori et al., 2003; Marvin and Hewitt,
2007) and the activation of the aryl hydrocarbon receptor (AhR, Skarek et al.,
2007; Khedidji et al., 2017; Zhang et al., 2018a, b; McDonough et al., 2019) but
other studies have also expanded the range of endpoints to endocrine activity
(Clemons et al., 1998; Kennedy et al., 2009; Novak et al., 2020), oxidative stress

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 334 Bioanalytical Tools in Water Quality Assessment

and inflammation (Al Hanai et al., 2019; Chen et al., 2020). Particles in air are
usually collected using either passive or active air samplers, and chemicals bound
to particulates (such as PM2.5 and PM10) are extracted by (accelerated)
solvent extraction.
The studies have shown that chemicals bound to air particulate matter can induce
significant mutagenicity, AhR activity, estrogenicity, oxidative stress and
inflammation in cell-based bioassays, indicative of a risk to human health.
Dust is also a source of persistent organic pollutants, including flame retardants,
and dust has been used as proxy for human exposure. Dioxin-like chemicals and the
AhR activation were the focus of many studies on dust (Tue et al., 2013; Suzuki
et al., 2017) but also inflammation (Allermann and Poulsen, 2000) and endocrine
effects (Chou et al., 2015).
Microplastic in water and air poses an environmental threat due to the plastic as
such but also because plastic materials can carry micropollutants from production
but also from sorption processes in the environment (Koelmans et al., 2016). In
vitro bioassays have been used to assess the micropollutant burden in plastic and
microplastic (Rummel et al., 2019; Zimmermann et al., 2019, 2020).

15.4 BIOTA
Applications of in vitro bioassays in biota samples were reviewed by Jin et al.
(2015b) in marine wildlife. Seminal studies have not only used commercially
available cell lines, such as the ones typically used for water quality monitoring
but have tested extracts in cell lines that were derived from tissues of those or
similar test species.

15.4.1 Blood
Dogruer et al. (2018) applied a modified QuEChERS extraction method in a
preliminary study to blood samples from marine turtles and were able to quantify
cytotoxicity, activation of AhR and oxidative stress response and concluded that
turtles foraging closer to agricultural areas were associated with higher burdens of
mixture effects. Finlayson and co-workers developed several green turtle cell
lines (Finlayson et al., 2019a, 2019b) and measured cytotoxicity and oxidative
stress in green turtle cells associated with extracts from QuEChERS of green
turtle blood from those same sites (Finlayson et al., 2020), concluding that turtles
in coastal areas of the Great Barrier Reef were at risk from current concentrations
of organic contaminants.
Green turtle blood has also been extracted with silicone and 69–98% of the
detected response in the AhR CAFLUX assay could be explained by dioxin-like
compounds such as polychlorinated dibenzodioxins, furans and PCBs (Jin et al.,
2015a). The samples also activated oxidative stress response in AREc32 but the
causative chemicals could not be identified (Jin et al., 2015a). Blood of polar
bears also tested positively for AhR activation in the majority of tested samples

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Application of bioanalytical tools beyond water 335

while estrogenicity was agonistic or antagonistic in a smaller number of samples

(Erdmann et al., 2013).
Thyroid-disrupting compounds including various nonylphenol isomers and
hydroxylated chlorinated biphenyls were identified in plasma samples of polar
bears using EDA and a transthyretin (TTR)-binding assay (Simon et al., 2013).

15.4.2 Tissue
Most studies that have applied bioanalysis of tissue samples in wildlife have worked
with acid-treated tissue samples and some have applied passive sampling
techniques. Similar test batteries as for water quality testing have been applied
with more of a focus on the activation of the AhR and EROD activity because
many of the earlier studies focused on persistent organic pollutants and more
recently also hormone receptor activation (Jin et al., 2015b).
Suzuki et al. (2011) extracted liver and blubber from diverse marine mammals
and birds followed by acid-digest clean-up. The samples were run in a broad test
battery including AhR, the estrogen receptor (ER), the androgen receptor (AR),
the glucocorticoid receptor (GR) and the peroxisome proliferator-activated
receptor (PPARγ) and found no activity in these assays for most tissues but they
detected dioxin-like activity in AhR and AR antagonistic effects in blubber from
Baikal seals. Desforges et al. (2017) extracted blubber of killer whales and polar
bears and tested the extracts for several immunotoxicity endpoints, with
cytotoxicity and strong effects on T cell proliferation and phagocytosis detected.
Other groups have applied silicone to extract organic compounds from the
sample directly, without the need to purify the extract any further and thus ensure
that the sample composition did not change during the extraction process. Jin
et al. (2013) found a fairly good agreement between the BEQbio in the AhR
CAFLUX assay from directly analysing blubber extracts of dugongs and the
BEQchem predicted from the detected polychlorinated dibenzodioxins and furans
(for definitions of BEQbio and BEQchem, see Chapters 7 and 8). Similar extracts
tested in addition for their adaptive stress responses only showed activation of the
oxidative stress response and the detected dibenzodioxins and furans together
with some chlorinated pesticides could not explain the biological effect (Jin et al.,
2015c).
It must be noted, however, that even small fractions of coextracted lipids decrease
the sensitivity of the in vitro bioassays (Reiter et al., 2020). This is caused by
decreased bioavailability of the contaminants, as they are retained in the co-dosed
lipids in a similar manner as the serum-mediated passive dosing (SMPD)
described in Chapter 9. Unlike the effect of SMPD, which is constant, the effect
of lipids is diluted with dilution of the sample and eventually disappears. If the
chemical burden is high enough, this effect can eventually become negligible
with sufficiently large dilution, but caution must be applied for samples with
low contamination.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 336 Bioanalytical Tools in Water Quality Assessment

15.5 HUMAN BIOMONITORING

Applications of bioanalytical tools in human biomonitoring are only slowly
emerging. With the concept of the exposome (Wild, 2005), which is defined as
the totality of exposure to humans over a lifetime, came the awareness of
complex mixtures in our bodies and with it the idea to also apply bioanalytical
tools to capture mixture effects (Escher et al., 2017).
However, these tools are not widely applied yet. This can partly be explained out
of practical necessity that biomonitoring often needs to rely on more easily
accessible samples such as urine or hair. Metabolites of pollutants can be
analysed in urine and the detected concentrations of metabolites can be translated
to human exposure with toxicokinetic models but evidently mixture effects
cannot reasonably be measured in those types of samples. The preferred matrices
for quantification of mixture effects using bioanalytical tools are blood, as a
proxy for the immediate past exposure to chemicals, and lipid tissues, as a
measure of long-term storage of persistent organic pollutants.
Most studies applying in vitro bioassays have worked with blood or serum, often
as part of cohort studies, thereby interrogating association with certain health
outcomes. This is beyond the scope of the review in this chapter, for more details
the reader is referred to the excellent review by Vinggaard et al. (2020), who
collated and analysed 43 studies on human tissue that have applied
in vitro bioassays.
By far the most evaluated endpoint remains AhR activity. This is because many
persistent organic pollutants activate this receptor. It is also relevant for many
contaminants of emerging concerns, for which metabolic activation and
deactivation may also play a role. While AhR activation was mainly tested in
blood samples, the activation of hormone receptors has often been probed with
fat tissue or placenta extracts (as well as blood). Estrogenic effects and
anti-androgenic effects were most prominent and could in some cases be related
to adverse health outcomes. Applications with bioassays related to thyroid
disruption are scarce but important additions.
Fractionation techniques are generally used to remove endogenous chemicals, in
particular hormones, but some studies have also applied extensive EDA. Given the
complex mixtures of endogenous compounds present at higher concentrations than
the organic micropollutants, EDA seems most promising to ensure that neither false
positives nor masking effects impact the bioassay results.

15.6 CONCLUSION
Bioanalytical tools have great potential to capture mixture effects of organic
micropollutants in non-aqueous environmental and biota samples. The same
bioassays can be applied and a lot of the developments with respect of mapping
chemicals to modes of action and toxicity pathways and mixture effects

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Application of bioanalytical tools beyond water 337

modelling can be directly translated to these other matrices. Iceberg modelling, for
instance, has already been applied to sediments (Muz et al., 2020; Niu et al., 2020)
and to biota samples (Jin et al., 2015a).
However, sample preparation and sample clean-up are much more challenging
than for water samples, especially for biota samples and human specimen, where
endogenous chemicals may contribute to the effect. EDA has great potential to
not only identify causative chemicals but also help with separation and clean-up
without altering the sample itself. Equilibrium passive sampling with polymers
also seems very promising, however, samples with trace concentrations of
pollutants might be below detection limits as enrichment is limited with this method.
Triggered by the exposome concept and increasing awareness of the relevance of
mixture effects in the environment, we can expect tremendous growth in this field in
the near future. We fully expect that the third edition of this book may not be called
‘Bioanalytical tools in water quality assessment’ but ‘Bioanalytical tools for
environmental monitoring and biomonitoring’ (!)

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Chapter 16
A promising future for bioanalytical
tools

16.1 INTRODUCTION
In the almost 10 years since the publication of the first edition of this book, there
have been major developments and innovations in this and related fields that have
dramatically improved our ability to apply bioanalytical tools for water quality
monitoring and interpret their results.
In this last chapter we provide a summary of the achievements in the field so
far, followed by a knowledge gap analysis and an outlook into future research
needs and opportunities. We also discuss what work still needs to be carried
out for wider regulatory acceptance of bioanalytical tools in water quality
assessment.

16.2 ACHIEVEMENTS SO FAR

In the previous chapters we have learnt about the scientific background of
bioanalytical tools in the context of human and environmental health and how
they can be applied to water quality monitoring. There is now a wide array of
case studies in the literature where bioanalytical tools have delivered information
on mixtures of organic micropollutants in diverse water types – from pristine
water sources to contaminated sites and sewage. In addition to benchmarking
water quality, the overall treatment efficacy of a specific water treatment
technology or an entire treatment train can be assessed both in the validation and
verification stages.

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0339

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 340 Bioanalytical Tools in Water Quality Assessment

The combination of bioassays with targeted chemical analysis and suspect

screening is particularly powerful because they provide complementary
information and mixture modelling techniques can make a quantitative link
between the different techniques (Escher et al., 2020).

16.2.1 A sound guidance for selection of bioassays based

on the conceptual framework of toxicity pathways
The toxicity pathway and adverse outcome pathway concept presented in Chapter 4
and followed throughout the remainder of the book provides an excellent basis for
the categorisation and rationalisation of the choice of bioanalytical tools. In
addition, the toxicity pathway concept allows us to converge on the
commonalities between human health and ecological risk assessment by focusing
on cellular effects and responses.
With the diversity of toxicity pathways relevant for human and environmental
health, it becomes evident that there is no ‘one size fits all’ bioassay, but rather
that a test battery should include several bioassays with at least one from each
mode of action category (i.e., non-specific, reactive and specific toxicity).
Activation of hormone receptors and endocrine pathways may have adverse
effects on reproduction at sub-ng/L concentrations of pollutants that are difficult
to quantify with routine chemical analysis due to the low concentrations, but
reporter gene assays for hormone receptor activation are highly sensitive and
detect even very small quantities of bioactive chemicals.
In addition, adaptive cellular response pathways are important as they constitute
an aggregated response to the toxic insult and are more integrative than assessing the
individual modes of action. Finally, cytotoxicity provides a sum parameter of all
bioactive chemicals in complex mixtures and assures that ‘nothing has been
overlooked’. These principles of test battery design are covered in extensive
detail in Chapter 13, which relies on a tiered approach of indicator bioassays
rather than attempting to measure every receptor and molecular initiating event
we know of.

16.2.2 A more comprehensive measure of the realm of

organic pollutants
Bioanalytical tools bridge the gap between individual chemical analysis and direct
toxicity assessment of water samples. While direct toxicity assessment is very
comprehensive in that all components including salts and macropollutants such as
phosphate and nitrogen compounds (nitrate, nitrite) are evaluated together
(Figure 16.1), it has drawbacks with respect to sensitivity, assessing the causative
agents and avoiding artefacts (due to non-chemical factors, such as pH or
temperature). Macropollutants and metals can be comprehensively analysed by
chemical analysis, but organic micropollutants consist of such a wide range of
compounds and their transformation products that chemical analysis alone cannot

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 A promising future for bioanalytical tools 341

Figure 16.1 Bioanalytical tools as part of a comprehensive assessment of chemical

pollution in water. TOC = total organic carbon; AOC = assimilable organic carbon;
TOX = total organic halogens.

provide sufficient information and needs to be supplemented by alternative methods

that provide sum and surrogate parameters (Figure 16.1). Bioanalytical tools
provide a more refined sum parameter to the overall burden of organic
micropollutants than chemical surrogates such as total organic carbon (TOC),
assimilable organic carbon (AOC), or total organic halogen compounds (TOX).
Another advantage of bioanalytical tools is that bioassay responses are by nature
risk-scaled, that is, more toxic chemicals are given more weight in a bioassay than
less potent chemicals. A drawback is that the observed effect cannot be directly
related to a causative agent, and bioanalytical tools should therefore always be
used in conjunction with other techniques such as chemical analysis. ‘Iceberg’
mixture modelling provides a measure of how much of the measured effect is
caused by the detected chemicals and effect-directed analysis (EDA) can be used
to identify mixture risk drivers. These concepts are described in more detail in
Chapter 13.

16.2.3 Effect-based trigger values

A major change since the publication of the first edition of this book has been the
development of several approaches to derive effect-based trigger values (EBTs)
for a wide range of both category 1 and 2 assays (see Chapter 13 for details).
EBTs are specific for the water type, and might be different for drinking water,
surface water and recycled water from non-potable reuse. They all share a
common point of departure from safe concentrations for individual chemicals,
such as acceptable daily intake (ADI) for humans or environmental quality
standards (EQS) for the aquatic environment, but they go beyond single
chemicals and include mixtures in their derivation.
The lack of a threshold or trigger level that could be used to determine whether a
bioassay response was of concern or not has, understandably, previously been a

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 342 Bioanalytical Tools in Water Quality Assessment

major limitation to the wider adoption of bioanalytical tools for water quality
monitoring. With a larger set of EBTs now available to assess risk to both
humans and receiving ecosystems, and various frameworks to develop new EBTs
as needed, this step is no longer a limiting issue.
In addition, there are now also frameworks for application of EBTs in water
quality assessment (including our synthesised version in Figure 13.15) to help
practitioners determine how to adequately respond to exceedances of EBT
thresholds. Together, these new developments greatly enhance the usability of
bioanalytical tools in water quality assessment.

16.3 CHALLENGES
Despite major advances in the field, there are still knowledge gaps that need to be
addressed and require more research efforts. The most important ones in the
context of water quality assessment relate to the validity of the sample
preparation methods and experimental artefacts from the matrix itself. As
cell-based bioassays stand somewhere between chemical analysis and in vivo
toxicity testing, a quantitative connection between these different tools will help
to better understand the performance and limits of bioanalytical tools. However,
there is also a need for advances in basic science, with some endpoints and
bioassays requiring further development.

16.3.1 Matrix effects and extraction methods

Among the practical limitations are issues of sample preparation as discussed in
Chapter 12. This of course is not an inherent problem of the bioassays
themselves, but it nevertheless is a problem that has prevented regulatory
implementation in the past. Once a water sample is processed and not directly
tested, it is unclear how much of the overall toxic potential is actually retained
with common extraction and enrichment methods. Direct water testing is only
possible with highly contaminated effluents and wastewater, while recycled and
drinking water cannot be assessed without prior extraction and enrichment
because (a) concentrations of micropollutants are too low to be detected and (b)
matrix effects by salts and organic matter may interfere with the bioassays.
Quantifying the extraction yield is a problem that chemical analysis has
overcome with the addition of labelled standards, which allow correction for loss
during sample preparation. Internal standards should not be used with bioassays
because of the difficulty of differentiating between added standard and chemicals
originally present in the sample. Instead, bioassay studies have used spiked
mixtures and calculated recovery of the mix (‘effect recovery’) using mixture
modelling. While this approach cannot quantify individual sample recovery the
way deuterated standards are used for in chemical analysis, it can at least
establish the recovery efficiency for a particular sample extraction methodology.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 A promising future for bioanalytical tools 343

While methods for aqueous sample extraction such as liquid−liquid extraction

(LLE) or solid-phase extraction (SPE) have been used for decades, new
developments in sorbent materials for SPE and new extraction methodologies
need to be routinely evaluated to ensure optimal selection of the extraction
technique and parameters for bioassay analysis of a particular matrix. Co-polymer
SPE materials such as HLB or HRX can achieve remarkably good effect
recoveries. This can be rationalised by the fact that these polymers are almost
biomimetic – they efficiently extract chemicals of medium hydrophobicity and
charged chemicals that are bioavailable to the cells. They perform more poorly
for very water-soluble chemicals and those with multiple charges, but these often
show low bioavailability and have a low contribution to mixture toxicity.

16.3.2 Dosing into cell-based bioassays

Dosing into a bioassay might seem straightforward at first sight but can also pose
challenges with respect to loss of chemicals and dosing techniques, including the
choice of solvent in which the water extract is dosed and if the solvent is
removed prior to testing.
High-throughput screening (HTS) bioassays are only amenable to non-volatile
chemicals and the typical cut-off of a medium–air partition constant of 10,000
(Escher et al., 2019) comes as a surprise to many toxicologists unfamiliar with
concepts of multimedia fate modelling. Loss from the well of a plate is not just
related to the volatility of a chemical. One has to keep in mind that the medium
components such as lipids and proteins retain semi-volatile chemicals in the
assays to a certain extent and they also prevent too much loss due to binding to
the polystyrene of the well plates.
Even sealing the plates may not prevent loss of volatile chemicals, as plate
sealants must allow for gas exchange necessary for cellular respiration. Hence,
different extraction techniques and bioassay set-up need to be used for volatile
chemicals, for example, when testing the formation of disinfection by-products in
chlorinated drinking water (Stalter et al., 2016c).
Passive dosing techniques have been advocated in the past to overcome the issue
of loss processes (Smith and Schäfer, 2016). While passive dosing might be very
useful for testing of very hydrophobic chemicals to overcome solubility issues
and to obtain effect concentrations expressed as freely dissolved or cellular
concentrations, we do not recommend passive dosing for testing of water extracts
that also contain many hydrophilic and charged chemicals. Furthermore, in
cell-based assays that are run with protein-rich media, the medium itself serves as
a passive dosing device that counteracts the depletion due to cellular uptake and
loss processes (Fischer et al., 2019).
There has also been much speculation about solvent toxicity and what solvent to
use for dosing of water extracts. It has been widely reported that methanol and other
volatile solvents have a lower toxicity than DMSO. This is true, but most of this

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 344 Bioanalytical Tools in Water Quality Assessment

difference is likely because volatile solvents are lost during the incubation period,
especially for human cell assays that are incubated at 37°C. If solvents are lost
anyway – and not in a predictable and reproducible fashion – why not remove the
solvent right away prior to testing? Exchanging from a volatile solvent into
medium is fairly simple for water samples but might be more challenging for
extracts from tissues and sediments that contain more hydrophobic chemicals.
The only difficulty with using volatile solvents is that they evaporate – even
during storage at low temperatures. Therefore, vial weight needs to be
continuously tracked during the experimental workflow and volatile solvents
need to be topped up to initial weight, which seems laborious at first sight but
can be automated and is worth the effort, given the subtle and overt toxic effects
of DMSO as the most popular alternative. There are cases where it will be near
impossible to replace DMSO, for example, if realistic mixtures of many
chemicals are formulated or when hydrophobic chemicals are tested but it is
important for the reader to keep in mind that there might be alternative
approaches to ‘what we have been doing the last twenty years’.

16.3.3 Linking bioanalysis with chemical analysis

All of our current risk/safety assessment and guidelines are based on single
chemical testing, and all our regulations are written in this single chemical
language. To move forward towards a more realistic mixture assessment without
making our considerable single chemical toxicity information irrelevant, it is
important to ensure that we understand how to correlate bioanalytical measures
with chemical analysis. The scientific underpinning of mixture integration was
discussed in Chapter 8, and there has been significant progress in linking
chemical and bioassay results quantitatively over the past decade (see Chapter 13).
By definition, category 1 bioassays are activated by only a small number of
highly potent chemicals and most water samples tested with these assays show
good agreement between predicted and measured bioassay response, particularly
when applied to relatively well-understood matrices. For these category 1 assays,
EDA is also a way to identify causative chemicals. For those types of endpoints,
bioassays and chemical analysis could be used interchangeably as both can in
principle provide the same information but it remains beneficial to apply both
techniques in parallel. Especially very potent chemicals might fall below
individual detection limits but still contribute to the mixture effects and bioassays
can give direct responses without the need for extensive mixture modelling.
Further, applying both techniques gives more confidence that many chemicals
present at low concentrations act indeed according to the easily predictable
models of concentration addition and independent action (which actually yield
the same results at very low effect levels, typically ,10%).
It is an urban myth that water samples might contain highly synergistically acting
mixtures. Synergy is of academic interest and exploited for pesticide formulations –

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 A promising future for bioanalytical tools 345

mostly by inactivating detoxifying systems so that the synergy is rather a

toxicokinetic effect than a true synergistic toxicodynamic interaction). Even if a
small fraction of mixture components interacts synergistically, this will hardly
affect the mixture effect of the entire water sample that contains thousands,
potentially even hundreds of thousands of individual chemicals.
Category 2 bioassays are an entirely different story. There are many chemicals
that activate these assays, and it is not possible to explain the mixture effects by
the detected chemicals. EDA usually just results in every fraction being active or
loss of activity in all fractions but no clear mixture effect drivers, unless the
samples are from a contaminated site or an accident. While we do not know what
the unknown active chemicals are, we can at least quantify how much of the
effect is unknown: in a water sample the fraction of unexplained effect can be as
high as 90−99% of the total effect quantified with a category 2 bioassay. This
means that, for these endpoints, we need bioassays to complement the
information of chemical analysis and that component-based prediction of the
mixture effects from the identified chemicals will never close the ‘effect’ balance.
This observation has actually been misused as an argument against these
bioassays – some researchers and stakeholders reject category 2 bioassays
because their effects cannot be explained by the detected chemicals. The opposite
should be the case. If we can assure that a category 2 bioassay is describing a
molecular initiating event or a key event that is linked to an adverse outcome,
then the mixture effect is of biological relevance, whether we know what
chemicals trigger it or not – provided of course that the bioassays are run to
high QA/QC standards, that sample preparation is of high quality, and that
there are no matrix artefacts. Then ‘iceberg’ modelling of effects observed in
category 2 bioassays will provide a measure of chemical burden that chemical
analysis hitherto overlooked entirely but may contribute significantly to the risk
of mixtures.
Mathematical models have also been developed to meaningfully link category 2
bioassay responses to chemical guidelines. Additional studies will add to the
weight-of-evidence needed to convince risk assessors and regulators that
chemical and bioassay analysis complement each other.

16.3.4 Linking bioanalysis with whole-animal testing

Understanding how predictive in vitro bioassays are of whole organism effects is
crucial to answer the ‘so what?’ question. The adverse outcome pathway provides
the logical framework to connect the dots, but much research is still needed to
connect in vitro bioassays with adverse outcomes in a quantitative manner.
Adverse outcome pathways are not linear, but they are actually networks,
therefore there is no one-to-one relationship between one in vitro bioassay and
the adverse outcome but often several endpoints need to be combined to predict
the adverse outcome.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 346 Bioanalytical Tools in Water Quality Assessment

The development of alternative test methods for risk assessment of chemicals has
led to a paradigm shift from exclusive reliance on in vivo toxicity testing to a process
that relies on in vitro bioassays in conjunction with (quantitative) in vitro to in vivo
extrapolation (QIVIVE) models (discussed in more detail in Chapter 9).
The validation of these alternative test methods can be used as the basis for
application to water samples. Validated in vitro systems with a clear anchor and
effect thresholds for adversity through inverse QIVIVE modelling can be applied
for monitoring complex mixtures in the environment and for improving the
derivation of EBTs. To adapt single chemical QIVIVE models to mixture QIVIVE
models will most likely be possible in the future for defined mixture but might
remain out of reach for complex environmental samples of unknown composition.
Ultimately, it may be sufficient if we can make the connection for a large number
of individual chemicals to extrapolate eventually to environmental mixtures.

16.3.5 Bioassays that require further development

As identified in Chapter 5 some endpoints are currently poorly modelled using
in vitro methods, particularly if they rely on meta-cellular and integrated
organism responses. Reproductive and developmental toxicity are two classical
examples for which there is currently no comprehensive in vitro model. The
development of pluripotent stem cells from non-embryonic tissue removes the
ethical barrier of using embryonic stem cells, while advances in microfluidics
leading to ‘animal-on-a-chip’ platforms (see Section 16.3.6) may enable future
development in simulation of integrated organism responses at the in vitro level,
although much work is still needed before this occurs.
While there are several good high-throughput bacterial assays to assess
mutagenicity and genotoxicity, it is important to remember that there are
significant differences in the organisation and protection of DNA between
prokaryotes and eukaryotes. For example, DNA in eukaryotes is located inside the
nuclear envelope while prokaryotic DNA is located in the cytoplasm, or eukaryotic
DNA is bound to proteins and scaffolded into condensed chromatin while
prokaryotic DNA is naked and unwound. There may therefore be differences in
sensitivity of prokaryotes and eukaryotes to mutagenic and genotoxic chemicals.
While there are options for eukaryotic DNA toxicology such as the micronucleus
and Comet assays, they tend to be cumbersome and low throughput. With the
advent of automated microscope imaging suitable for well plate formats and
artificial intelligence-enabled image analysis, it is becoming possible to conduct
these generally low-throughput methods in significantly higher throughput
formats, which would make them suitable for application to water quality monitoring.
And there are a range of other endpoints where bioassays are currently limited,
including activation of the constitutive androstane receptor (CAR) and assays for
neurotoxicity, just to name a few. New developments would enable wider
application of these assays to water quality testing.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 A promising future for bioanalytical tools 347

We also need to keep in mind that most mammalian cell-based bioassay and
reporter gene assays are derived from undifferentiated cancer cells. This is
because primary cells survive only for a few cell cycles before they die off
in vitro. Primary cells need to be constantly sourced from new tissue, resulting in
ethical and reproducibility problems. In contrast, cancer cells are immortal and
can be kept in a stable state for a very long time but are by definition mutants and
can sometimes be very different from primary cells. Several innovations have
enabled the creation of immortal non-cancerous cell lines, though these have not
yet been widely used as platforms to develop reporter gene assays. Future work
should be directed towards combining these emerging technologies of cell
immortalisation with the development of novel bioassay.

16.4 FUTURE OPPORTUNITIES

High-throughput toxicogenomics, three-dimensional cell systems and organism-
on-a-chip technology, and adaptation to online monitoring format offer future
opportunities for adoption into water quality assessment once the immediate
research gaps are closed.

16.4.1 The ‘omics’ revolution

The ‘omics’ revolution, the exciting developments in genomics, transcriptomics,
proteomics and metabolomics have the potential to be brought into practical
application. Technological advances have made it possible to run some of these
in a high-throughput format using in vitro exposures in 96- and 384-well plate
format, and they are increasingly used in chemical screening and screening-level
risk assessment (Harrill et al., 2019). The field is being developed from
qualitative to quantitative assessment and from single compounds to mixtures
(Spurgeon et al., 2010), but some harmonisation and methodological refinements
are still necessary to reduce assay variability (Sauer et al., 2017). Once assay
protocols are refined to reduce variability and standardised methods are worked
out to assess and evaluate mixtures, it should be possible to also move on to
complex and unresolved mixtures as they occur in water samples. There are still
many questions to be solved − the most pertinent being ‘what does it mean if a
gene is transcribed x-fold more than in the control?’ or ‘how does a change in
metabolite x affect the cell and the organism?’, which relates to the issue of
in vitro to in vivo extrapolation debated in Section 16.3.4.

16.4.2 Three-dimensional cell models and organ- and

animal-on-a-chip systems to better model whole
organism response
Undifferentiated cells grown artificially in a two-dimensional monolayer on plastic
surfaces optimised for cell culture, as commonly used in in vitro toxicity testing, are

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 348 Bioanalytical Tools in Water Quality Assessment

clearly not fully representative of the in vivo environment. In particular, they lack
interaction with the extracellular matrix and more complex inter-cellular
communication, which is often vital in the function of organs. Therefore,
three-dimensional cell models have been developed to simulate a variety of organs
such as the small intestine, kidney tubules, bronchioles, liver, blood−brain barrier,
lung alveoli and bone marrow (reviewed in Jensen and Teng, 2020).
The colon cancer cell line Caco-2 is a widely used in vitro model for intestinal
uptake and is especially important in the pharmaceutical industry for testing the
oral availability of pharmaceuticals. If grown on a microporous membrane,
Caco-2 cells are capable of forming a three-dimensional epithelial barrier that
expresses tight junctions and can support active and passive uptake processes
(Cencic and Langerholc, 2010).
Tests for skin sensitisation by chemicals and permeation through skin are
typically performed in three-dimensional (3D) reconstructed human epidermal
models, which consist of several cell types and a dermal matrix. These 3D skin
models can simulate many skin functions including barrier and immunological
functions.
A disadvantage of many 3D cell models is that they are essentially not ‘true’
in vitro systems, but rather ex vivo, because they require primary cell lines or
tissues from animals that are isolated from a living organism and cannot be
maintained in culture for long periods. While they are becoming powerful tools
in the earlier tiers of drug development pipelines (Langhans, 2018), application of
3D cell culture to HTS required for environmental monitoring applications
remains a challenge.
By etching microscopic channels and microfluidic engineering, researchers have
developed organ-on-a-chip platforms that mimic the function of various organs
(e.g., spleen, lung, liver, kidney, heart, gut, blood−brain barrier), going as far as
developing body-on-a-chip (also known as animal-on-a-chip) platforms by
including cells from multiple organ systems to integrate organ interactions
(reviewed in Zheng et al., 2016). Microfluidics platforms can also be combined
with 3D cell models to produce advanced organ-on-a-chip models to investigate
organ-level function (Shoemaker et al., 2020). Chip-based methods are also
available for ecotoxicological test systems (Campana and Wlodkowic, 2018).
Like 3D cell models, these new developments greatly enhance the ability of
in vitro systems to mimic whole organism responses, but they are not yet
compatible with HTS application.

16.4.3 Moving from offline to online monitoring

Surveillance monitoring requires changes in chemical pollutants or accidents to be
detectable immediately, ideally using on site and online monitoring. Most
bioanalytical tools are presently applied only offline, that is, water samples are
taken and transported to the laboratory for processing and analysis. The bioassay

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 A promising future for bioanalytical tools 349

then requires cells to be exposed for several minutes to several days, with 24 hours
the most common exposure period. Since many commonly applied cell lines are
genetically modified, they can only be maintained in certified facilities with
secure quarantine and containment. The results are therefore obtained at best with
a delay of 24 hours, usually more. Automation using online extractions and
automatic HTS is technically possible as developments in analytical chemistry
and pharmaceutical drug discovery as well as in the Tox21 programme have
demonstrated. It should therefore be possible to shorten the interval between
collecting a sample and receiving the analysis results and ensure that
bioanalytical methods fulfil their potential in surveillance monitoring. There are
already some online monitoring systems available to monitor toxicity to bacteria
(e.g., iTOXcontrol, Microtox CTM and NitriTox systems) and algae (e.g.,
AquaSentinel, Algae Toximeter II) as well as whole-cell biochips for online
monitoring (Elad and Belkin, 2012), now also for mobile phone applications (Lu
et al., 2019). Vertebrate cells require surrounding media with serum (including
growth factors, amino acids and nutrients to feed and sustain the cells), and
online monitoring with vertebrate cells is therefore significantly more challenging.
It is worth noting, however, that offline and online monitoring is not as binary as
one might think. Indeed, in addition to ‘offline’ (sample taken manually, analysis in
off-site laboratory) and ‘online’ monitoring (fully automated sampling, analysis on
site), there is also ‘at-line’ monitoring (manual sampling, but analysis on site) and
‘inline’ monitoring (where a probe is placed directly in the process stream).
Continuous monitoring of water quality requires at least at-line monitoring – and
at-line bioanalytical monitoring is already possible. Using electric cell-substrate
impendence sensing (ECIS), U.S. Army researchers have established and
validated a field-portable drinking water toxicity sensor to monitor acute toxicity
to rainbow trout gill epithelial cells as a way to monitor drinking water quality
(Widder et al., 2015). A variety of other sensors are also now becoming
available, widening the field of the types of cells that can be used in at-line
monitoring applications (Tan and Schirmer, 2017).

16.4.4 Towards ultra-high-throughput testing, multiplex

assays and artificial intelligence-assisted bioinformatics
In their aim to test thousands of environmental chemicals in .1000 different assays,
programmes such as ToxCast and Tox21 have deployed testing platforms that
use 1536-well plates and low volume acoustic pipetting robots capable of
accurately dispensing volumes as small as 2.5 nL. These tools create incredible
opportunities for ultra-HTS of environmental samples.
Multiplex assays, where multiple effects are monitored simultaneously, would
also offer a way to increase testing throughput by creating assays that can
measure multiple responses at once. To a certain extent, most assays are already
multiplexed: indeed, combining cell viability and cell vitality assessment for

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 350 Bioanalytical Tools in Water Quality Assessment

cytotoxicity with a reporter gene assay already provides two streams of toxicity
information. But new developments in fluorophores, luciferase enzymes and
emission spectra deconvolution open up avenues for highly multiplexed assays
that retain the sensitivity of reporter gene assays.
Advancements in imaging technology have also made it possible to read out
more from existing bioassays: phenotypic profiling, which is already popular for
fish embryo toxicity assays can also be applied to cell-based assays. A study that
applied multiparameter phenotypic profiling in MCF-7 showed how the size and
structure of cells is related to biological processes like cell growth, death and
communication and applied these tools to testing environmental waters (Wang
et al., 2018).
Ultra-HTS, more reliance on imaging techniques and assay multiplexing would
dramatically increase our capacity to test the toxicity of chemicals, chemical
mixtures and environmental samples, including water samples – but they also
require increasingly complex bioinformatics workflows, often including a degree
of artificial intelligence support. The risk is that bioassay analysis becomes a
complex black box, which could lead to errors if improperly designed. The future
of cell-based toxicity testing is bright, but we need to stride forwards with open
eyes and critical minds.

16.5 THE ROAD TO REGULATORY ACCEPTANCE

At present, bioanalytical tools are popular research tools and while their
development and validation are by no means fully accomplished, we believe it is
time to move them forward to regulatory applications. Regulators and proponents
of the water industry all agree that we need to overcome the limitations of a
chemical-by-chemical approach. In their ‘Chemicals Strategy for Sustainability
towards a Toxic-free Environment’, the European Commission has clearly stated
that ‘scientific consensus is emerging that the effect of chemical mixtures needs
to be taken into account and integrated more generally into chemical risk
assessments’ (European Commission, 2020). What is preventing this crucial next
step in the implementation of bioanalytical tools for water quality assessment?
We did not wait for the development of advanced chemical tools such as
LC-MS/MS before we decided to define chemical-based water quality criteria.
The definition of guideline values has actually boosted rapid advances in the field
of chemical analysis. Likewise, we should not wait until we can produce the
‘perfect’ bioassay battery but instead consider its application today and start
gaining more practical experience about applicability and limitations. We should
trust in the fundamental principles of physical chemistry and biology that this
field of science is based on, and not require everything to be demonstrated
experimentally ad nauseam.
The most important benefit of bioanalytical tools is that they are risk-based
measures, that is, a more toxic or more potent chemical contributes more to a

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 A promising future for bioanalytical tools 351

bioassay response than a less potent chemical – just like it does in whole organisms.
Bioanalysis therefore has huge potential as a prioritisation and monitoring tool. In a
tiered approach bioanalytical tools could be used for initial screening and hazard
identification, where only samples that exceed a given threshold such as an EBT
go into more detailed evaluation (Figure 16.2). In a world of limited resources,
this would allow a risk-based prioritisation of samples for complex chemical
analysis.
The fundamental question then becomes ‘where do we set that threshold?’. There
has been significant innovation in this field in the past decade, and several
approaches to derive EBTs are discussed in Chapter 13 (Section 13.5). Once a
threshold has been set, the question becomes ‘what to do if the threshold is
exceeded?’. Again, the last decade has seen several approaches to respond to this
question, and we summarised our current thinking in Chapter 13 (Figure 13.15).
These two key questions have in the past been key to regulatory acceptance.
The State of California leads the world at the moment as the first to implement
bioassays in the Water Quality Control Policy for Recycled Water (State Water
Resources Control Board, 2019). The policy recommends a trigger level of 3.5
ng/L EEQ in an ERα reporter gene assay and 0.5 ng/L TCDD EQ in an AhR
reporter gene assay. Other guidance documents recommend the application of
bioassays for assessing the presence and possible risks associated with chemicals
in water, although they do not go as far as recommending a trigger value. These
include the Guidance for Producing Safe Drinking-Water (Section 3.3 in WHO,
2017c) and the Australian Guidelines for Water Recycling – Augmentation of
Drinking Water Supplies (Section 4.5.1 in NRMMC/EPHC/NHMRC, 2008).
Water Safety Plans (WSPs) were developed by the World Health Organisation
(WHO) and offer an internationally accepted approach to ensure the safety and
acceptability of drinking water supplies, with a focus on hazard prevention

Figure 16.2 A tiered approach for water quality assessment with respect to
chemicals. MIE = molecular initiating event; KE = key event.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 352 Bioanalytical Tools in Water Quality Assessment

Figure 16.3 Common purposes of effect-based monitoring (EBM) and how they fit
into monitoring within Water Safety Plans (WSP) (Bartram et al., 2009). EBT =
effect-based trigger value.

(Bartram et al., 2009). WSPs often provide the backbone, on which national
drinking water guidelines are built, and understanding how to integrate
bioanalytical tools in WSPs could help regulators appreciate how to incorporate
bioanalytical tools in their water quality management frameworks. WSPs consist
of 11 modules that systematically articulate a framework to characterise, monitor
and manage drinking water quality. Bioanalytical tools would logically fit into
seven of those modules (Figure 16.3), and a further two (modules 8 and 9) would
benefit from the development of bioanalytical standard operating protocols
(SOPs; see Chapter 11) and training programmes.
Now that we have (1) more than a decade of experience with the systematic
application of bioanalytical tools to water quality monitoring, (2) established EBT
values for a wide range of assays and developed protocols to derive EBTs for
new assays, (3) established a simple and realistic framework to respond to
exceedances of EBT values and (4) are seeing some initial uptake by regulatory
bodies such as the California EPA, it is likely that we will start to see greater
acceptance of these tools in regulatory contexts.

16.6 CONCLUSIONS
In this chapter we have outlined the potential and limitations of bioanalytical tools
and have dreamt about possible future development. We hope that the review of the
state-of-the-science and applications as presented in this book will help to encourage

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 A promising future for bioanalytical tools 353

fellow researchers to persevere and progress bioanalytical approaches from research

tools to practical application, give practitioners in (waste) water treatment
recommendations on how to apply in vitro bioassays in their daily business and
help regulators to implement them as regulatory tools.
Finally, we conclude on an optimistic notion, noting that we have seen
tremendous progress in this field since the publication of the first edition of this
book almost a decade ago, and we hope that the case studies that we are
providing in this book will be outdated very soon by new advances in this field!

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 Glossary

AA-EQS: annual average EQS (→ EQS).

ABC transporters: ATP binding cassette transporters, family of drug transporters
involved in active transport of organics (→ active transport).
Abiotic: a process that does not involve living organisms, for example, ‘abiotic
transformation reactions’ take place via physical and chemical processes
rather than via biological processes (→ biotic).
ACB: → activity concentration at baseline (benchmark value used by Tox21).
ACC: → activity concentration at cutoff (benchmark value used by Tox21).
Acceptable (or allowable) daily intake (ADI): the maximum amount of a
chemical that can be ingested daily over a lifetime with no appreciable health
risk (→ tolerable daily intake, → reference dose).
Acetylcholinesterase (AChE): an enzyme that catalyses the hydrolysis of the
neurotransmitter acetylcholine.
Active transport: transport of molecules over cell membranes that requires energy
(e.g., ATP) (→ ABC transporters, → passive transport).
Activity concentration at baseline (ACB): concentration in a → concentration
response curve that is associated with the three times the median absolute
deviation of the effects of the two lowest concentration tested (benchmark
value used by Tox21).
Activity concentration at cutoff (ACC): concentration in a → concentration
response curve that is associated with a defined response cutoff, for example,
15%.
Acute exposure: exposure over a short period of time (hours or days) (→ chronic
exposure).

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0355

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 356 Bioanalytical Tools in Water Quality Assessment

Adaptive stress response pathway: the cellular pathway for stress responses
induced by chemicals and other stressors.
ADI: → acceptable (or allowable) daily intake.
ADME: absorption, excretion, distribution and metabolism (→ toxicokinetics).
Adverse outcome pathway (AOP): conceptual framework that leads from the
initiating event of interaction between a toxicant and a receptor in an
organism from cellular and organ response to an adverse outcome at
organism- or population level (→ toxicity pathway).
ADWG: Australian Drinking Water Guidelines (NHMRC, 2011).
Aflatoxin: a type of mycotoxin (fungal toxin).
Agonist: chemical that mimics the action of a natural substrate, for example, binds
to a receptor of a cell and triggers the natural response of that cell
(→ antagonist).
AGWR: Australian Guidelines for Water Recycling (NRMMC/EPHC/AHMC
2006; NRMMC/EPHC/NHMRC 2008, 2009a, 2009b).
AhR: aryl hydrocarbon receptor (also termed dioxin receptor). The AhR is
involved with the induction of cytochrome P450 monooxygenase and
induced by dioxins and dioxin-like chemicals such as PAHs.
AhR CAFLUX: the AhR chemically activated fluorescence expression assay is a
reporter gene assay for detection of dioxin-like activity in water samples.
Aliivibrio fischeri: marine bioluminescent bacterium used in the Microtox assay
(→ Microtox) formerly called Vibrio fischeri.
Ames test: assay for mutagenicity that measures the ability of toxicants to mutate a
histidine-dependent strain of the bacterium Salmonella typhimurium to grow on
a histidine-deficient substrate.
Anaemia: blood deficiency.
Androgen: natural or synthetic hormones including testosterone that regulate
development and maintenance of masculine characteristics via the androgen
receptor (→ AR).
Antagonist: chemical that blocks the action of an agonist (→ agonist).
Antibody: protein used by the immune system to detect and neutralise a foreign
substance (e.g., microorganism).
Antigen: foreign agent (e.g., bacterium, virus) that triggers the production of an
antibody (→ antibody).
Antimitotic drug: a drug used to fight cancer that inhibits cell division by
interfering with mitosis.
AOC: assimilable organic carbon.
AOP: → adverse outcome pathway.
Apical endpoints: traditionally measured outcomes of toxicity in whole
organisms, for example, lethality or reproductive failure.
Apoptosis: programmed cell death (as opposed to unplanned cell death, →
necrosis).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 357

AR: androgen receptor, important for male sexual development and reproduction
and induced by natural and synthetic androgens (→ androgen).
ARE: antioxidant response element.
Aromatase: an enzyme in the → CYP family that is important for biosynthesis of
estrogens and hence, for sexual development.
Assay: procedure in toxicology for testing the activity of a chemical on a biological
system (cells, organisms, populations).
ATP: adenosine-5’-triphosphate, a multifunctional nucleotide that transports
energy within cells.
Autoimmune disorder: a disease whereby the immune system starts attacking an
organism’s own tissues.
AWTP: advanced water treatment plant.
Axon: the projection of a neuron that conducts electrical impulses.
B[a]P: benzo[a]pyrene.
Basal toxicity: term for baseline toxicity in human toxicology (→ baseline
toxicity).
Baseline toxicity: the minimal toxicity exhibited by any compound. Partitioning
of chemicals into biological cell membranes causing non-specific disturbance
of membrane integrity and functioning (→ narcosis, → basal toxicity).
BCF: bioconcentration factor.
BEC: → biologically effective concentration.
BED: → biologically effective dose.
Benchmark dose (BMD): dose associated with a specified level of response, the
→ benchmark response.
Benchmark response (BMR): certain response level from which → BMD are
derived, typically 10%.
BEQ: → bioanalytical equivalent concentration.
Bioaccumulation: accumulation of chemicals within organisms exposed via
the surrounding environment (e.g., air, water, soil, sediment, food)
(→ bioconcentration).
Bioactivation: biological activation of a chemical, that is, biological
transformation that produces a metabolite that is more toxic than its precursor.
Bioanalytical equivalent concentration (BEQ): the concentration of a reference
chemical that would elicit the same effect as the mixture of micropollutants in a
water sample. BEQ differ from → TEQ in that effects on the level of a → MIE or
→ KE are assessed, which may or may not result in an → AO, while TEQ refer
to toxicity, that is, an → AO.
Bioanalytical tool: cell-based or low-complexity in vitro bioassay indicative of
a specified endpoint relevant for human and/or environmental health.
Bioavailability: refers to fraction of a chemical that can be taken up into cells.
Bioconcentration: accumulation of chemicals in aquatic organisms from the
surrounding water via gills, skin and carapace. Bioconcentration does not include
dietary accumulation and/or other non-water sources (→ bioaccumulation).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 358 Bioanalytical Tools in Water Quality Assessment

Biodegradation: biological degradation of chemicals by microorganism (e.g.,

bacteria).
Biological pathway altering concentration (BPAC): effective concentration in
an in vitro assay indicative of a specific toxicity pathway.
Biologically effective concentration (BEC) or biologically effective dose
(BED): the BEC or BED is the amount of a toxicant that reaches cells, sites
or membranes where adverse effects occur. The BEC or BED may represent
only a fraction of the delivered concentration, but it is the most appropriate
measure of exposure for predicting adverse effects.
Biomarker: biochemical, physiological or histological response of a cell or
organism to a stressor that can be easily measured and used as a measure of
exposure or effect.
Biotic: a process involving live organisms, for example, a ‘biotic transformation
reaction’ takes place via biological processes as opposed to physical and
chemical processes (→ abiotic).
Biotransformation: metabolic transformation of chemicals within cells
and organisms.
BMD: → benchmark dose.
BMR: → benchmark response.
BPAC: → biological pathway altering concentration.
BTEX: benzene, toluene, ethylbenzene and xylene, a group of aromatic
hydrocarbons.
CA// DA: concentration addition/dose addition (→ concentration addition).
CA: → concentration addition.
CALUX: the chemically activated luciferase gene-expression group of assays
comprise cell lines for detection of receptor-mediated activity for several
nuclear receptors, for example, → ER (i.e., ER-CALUX), → AhR, → AR,
→ GR, → PR, → TR.
CAR: the constitutive androstane receptor is involved with protection of toxicity
induced by bile acid and regulation of physiological functions.
Carbamate pesticides: class of insecticide related to organophosphate
insecticides that likewise interfere with the enzyme → actylcholinesterase
(AChE) at neuronal synapses and neuromuscular junctions.
Carcinogenesis: formation of cancer cells, that is, the initiation of cancer.
Carcinogenicity: the mode of toxic action for cancer, which can be caused by
direct DNA damage (genotoxicity) or interference with gene regulation
(epigenetic) (→ epigenetic carcinogen, → genotoxicity).
Carcinoma: malignant tissue growth (→ tumour).
Cardiomyocyte: cardiac muscle cell.
Cardiovascular toxicity: toxicity to the heart and vascular system (i.e., blood
vessels).
Catabolism: metabolic breakdown of endogenous and exogenous molecules.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 359

Chemical-group motivated approach: test battery design approach that

prioritises the detection of a specific group(s) of similarly acting chemicals
rather than the protection goal (→ protection-goal motivated approach).
Chronic exposure: exposure over a longer time period compared to acute
exposure (weeks to years) (→ acute exposure).
Comet assay: assay that detects DNA strand breaks in different cell types
including human, other mammalian and fish cells. Also known as → single
cell gel electrophoresis (SCGE) assay.
Concentration addition (CA): quantitative model for mixture toxicity of
chemicals that act via the same target site and/or the same mode of toxic
action, that is, the individual toxicities (expressed as effect concentrations)
add up to the sum toxicity of the mixture. CA is applied in aquatic
ecotoxicology, whereas dose addition (DA) is used in mammalian toxicology
(→ independent action).
Concentration: the mass or molar amount of a chemical divided by the volume of
the test system.
Concentration−response curve (CRC): a plot of the degree (%) of
response observed in a test population against increasing exposure
concentration (also known as dose−response curve for in vivo
mammalian testing).
Confidence interval: the 95% confidence interval, for example, is the range of
values between which 95% of the data will fall into.
Conjugate: adduct between a foreign chemical and a small hydrophilic
biomolecule (e.g., glucoronide or sulphate), catalysed by a metabolic
enzyme, typically in the liver.
Conjugation: a phase II metabolic reaction that involves the addition of a
hydrophilic biomolecule to a xenobiotic in order to form a (more)
water-soluble product (conjugate) for excretion.
CRC: → concentration−response curve
CVr: repeatability coefficient of variation.
CVR: reproducibility coefficient of variation.
CWA: U.S. Clean Water Act.
Cyanotoxin: a type of toxin produced naturally by cyanobacteria (blue-green
algae), which can form toxic blooms in both fresh- and marine water.
CYP: → cytochrome.
CYP450: → cytochrome P450.
Cytochrome: redox active proteins bonded to a heme, involved as electron
transfer agents in many metabolic pathways, e.g., → cytochrome P450.
Cytochrome P450: a monooxygenase enzyme superfamily involved in the
metabolism of endogenous and xenobiotic compounds.
Cytokine: cell-signalling molecule.
Cytoreductive drug: a drug than can reduce cell number (e.g., cancer cells).
Cytotoxicity: toxicity of a chemical or other stressor to living cells.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 360 Bioanalytical Tools in Water Quality Assessment

DA: → dose addition.

DBP: → disinfection by-product.
DF: → dosing factor.
Dioxin: colloquial for → polychlorinated dibenzodioxin.
Direct toxicity assessment (DTA): term commonly used in Australia to describe
whole effluent toxicity (WET) testing (→ whole effluent toxicity testing).
Disinfection by-product (DBP): chemical formed reactions between organic and
inorganic matter during chemical disinfection of, for example, drinking water
and pool water, including trihalomethanes (THM) and haloacetic acids (HAA).
Dithiocarbamate pesticides: a type of carbamate pesticide (→ carbamate
pesticide).
Diuron: herbicide that acts on photosynthesis by inhibition of photosystem II
(→ photosystem II).
DMSO: dimethyl sulphoxide, a solvent.
DNA: deoxyribonucleic acid, the macromolecule that encodes the genome and
carries all the genetic information in living organisms.
DOC: dissolved organic carbon.
Dose addition (DA): same concept as → concentration addition, however, DA is
used in mammalian toxicology.
Dose−response assessment: relationship between dose or level of exposure to a
substance, and the incidence and severity of an effect (→ dose−response
curve).
Dose: total quantity of a chemical delivered to a test animal or test system.
Dose−response curve: a plot of the degree (%) of toxicological response observed
in a test population against increasing exposure dose (→ concentration−effect
curve for in vitro testing).
Dosing factor (DF): dosing factor in a bioassay, the ratio of the volume of an
extract dosed into a well of a microtitre plate divided by the total volume of
the medium in the well.
DTA: → direct toxicity assessment.
DWEL: drinking water equivalent level, a lifetime exposure concentration
protective of adverse, non-cancer health effects, assuming that all of the
exposure to a contaminant is from drinking water (based on → acceptable (or
allowable) daily intake).
DWTP: drinking water treatment plant.
EBT: → effect-based trigger value.
EC: effect concentration.
EC50: effect concentration causing 50% of maximum effect in cell-based assays.
Ecological Risk Assessment (ERA): process of estimating potential impact of a
chemical, biological or physical agent on a specific ecological species,
population or ecosystem (→ human health risk assessment).
ECVAM: European Centre for the Validation of Alternative Methods.
EDA: → effect-directed analysis.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 361

EEQ: 17β-estradiol equivalent concentration.

EF: enrichment factor or → extraction factor.
Effect unit (EU): Effect unit from chemical analysis (EUchem) is the ratio of
concentration of a chemical concentration divided by its effect concentration,
and effect unit from bioassays (EUbio) is the inverse of the effect
concentration of a mixture or environmental samples (1/EC) (→ EC).
Effect-based trigger value (EBT): acceptable effect level or → BEQ in a specific
water type (surface water, drinking water)
Effect-directed analysis (EDA): identification of mixture effect drivers by
fractionation of a complex water sample and testing the effect in the fraction
in several iterations until the causative agent is identified (→ toxicity
identification evaluation).
ELISA: enzyme-linked immunosorbent assay, an immunological assay
technique that relies on an enzyme bonded to a particular antibody
(→ immunoassay).
Endocrine disrupting compound (EDC): a chemical capable of modifying
natural hormone function (→ endocrine disruption).
Endocrine disruption: interference with the endocrine (hormone) system.
Endogenous: originating from within, that is, endogenous substances are
produced within an organism or cell. Endogenous processes occur within an
organism or cell and involve endogenous substances.
Endothelium: the inner cell layer of blood vessels or organs.
Endotoxin: bacterial toxin.
Endpoint: an observable or measurable biological event used as an indicator
of effect.
Enrichment factor: → extraction factor.
Environmental Risk Assessment: → Ecological Risk Assessment.
Enzyme: an endogenous protein that catalyses (i.e., increases the reaction rate of)
chemical reactions.
Epidemiology: the study of disease clusters in human populations and the attempt
to link these to human exposure to chemicals.
Epigenetic carcinogen: a chemical that induces cancer via interference with
gene regulation mechanisms, as opposed to genotoxic carcinogens that
induce cancer via direct damage to DNA structure (→ carcinogenicity,
→ genotoxicity).
Epithelial cell: skin cells or cells that form the outer layer of organs within an
organism.
EQS: Environmental Quality Standard.
Equitoxic concentration: chemical concentrations of different chemicals that
cause the same level of toxicity.
ER: estrogen receptor, important for female sexual development and reproduction.
The ER is modulated by natural estrogens and xenoestrogens (→ estrogen,
→ xenoestrogen).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 362 Bioanalytical Tools in Water Quality Assessment

ERA: → Ecological Risk Assessment or → Environmental Risk Assessment.

ER-CALUX: assay for detection of estrogenicity in water samples (→ CALUX).
E-SCREEN: assay for detection of estrogenicity, which is based on cell
proliferation in estrogen-dependent (MCF-7) cancer cells (→ MCF-7).
Estradiol (E2): a natural → estrogen (female sex hormone).
Estriol (E3): a natural → estrogen (female sex hormone).
Estrogen (or oestrogen): group of female sex hormones including three naturally
occurring steroidal hormones (estrone (E1), estradiol (E2) and estriol (E3))
and synthetic compounds including 17α-ethinylestradiol (EE2) (→ ER,
→ estradiol, → estriol, → estrone, → 17α-ethinylestradiol).
Estrogenicity: toxic mode of action caused by estrogenic activity, for example,
binding to the estrogen receptor (ER).
Estrone (E1): a natural → estrogen (female sex hormone).
17α-Ethinylestradiol (EE2): potent synthetic estrogen, active ingredient of birth
control pills.
EU: → effect unit.
Eukaryote: an organism that has its DNA contained within the nucleus, that is,
most living organisms except some bacteria (→ prokaryote).
Eutrophication: when a water body receives a surplus of nutrients, leading to
excess plant growth and algal blooms.
Ex vivo: experiments using tissue or cells isolated from a living organism and
performed outside the organism.
Exogenous: originating from outside, that is, a process taking place within a cell or
organism is referred to as ‘exogenous’ if it is caused by a (exogenous) substance
with origin outside that organism or cell.
Exposure assessment: the determination of the emissions, pathways and rates of
movement of a substance and its transformation or degradation in order
to estimate the concentrations/doses to which human populations or
environmental compartments are or may be exposed (→ risk assessment).
External exposure concentration: the concentration of a chemical in the
exposure medium (e.g., cell medium, water, sediment, food) as opposed to
the BEC (→ biologically effective concentration).
Extraction factor (EF): the ratio between the mass or volume of a sample
that is extracted to the volume of the final extract (synonym for →
enrichment factor).
FCMN: flow cytometry micronucleus test for detection of micronucleus formation
(DNA damage).
FDA: U.S. Food and Drug Administration.
FET: fish embryo toxicity test.
Frameshift mutation (DNA// RNA): a change in the genetic (three letter codon)
reading frame caused by insertion or deletion of a number of nucleotides
different to three from a DNA sequence (a type of direct genotoxicity).
FRET: Förster resonance energy transfer.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 363

Furan: → PCDF.
GAC: granular activated carbon.
β-Galactosidase: hydrolysing enzyme often inserted in recombinant cell lines as a
marker that can be measured by addition of a substrate that forms a coloured
product upon hydrolysis.
GCB: graphitised carbon black used as solid material in → solid-phase extraction.
Gene activation: activation of gene expression, for example, by binding of a
nuclear receptor−ligand complex to DNA.
Genetic polymorphism: the co-occurrence of two or more genetically different
traits (phenotypes, morphs) within a population.
Genetically modified cell: a cell in which natural features have been
over-expressed by genetic engineering to enable more sensitive detection
and/or in which foreign features have been added for visualisation of effects
(→ recombinant cell).
Genomics: the study of genomes (the total sum of genes in a cell or organism), that
is, transcriptomics, proteomics and metabolomics (→ toxicogenomics).
Genotoxicity: the mode of action for DNA damage, for example, by direct reaction
with chemicals and reactive oxygen species (→ epigenetic carcinogens,
→ carcinogenicity).
GFP: → green fluorescent protein.
GHS: Globally Harmonised System for the Classification and Labelling of
Chemicals.
GI (tract): gastrointestinal (tract).
Glial cell: a cell type of the nervous system that is important for homeostasis,
myelination and support of neurons (→ myelin, → neuron).
Glutathione (GSH): antioxidant tripeptide, important for cellular defence against
ROS and conjugation of xenobiotics.
GR: glucorticoid receptor, important for regulation of development, metabolism
and the immune system.
Granulosa cells: estrogen-secreting cells that form the lining around female
oocytes (eggs).
Green fluorescent protein (GFP): a reporter gene often introduced to
recombinant cell lines as an easily measurable marker (→ recombinant cell).
Grey water: water used for domestic purposes such as laundry, dishwashing
and showering.
GSH: → glutathione.
GV: guideline value.
Haematopoiesis: production of blood cells.
Haematotoxicity: toxicity to the blood system.
Haloacetic acids (HAA): group of disinfection by-products formed from natural
organic matter during chemical disinfection of drinking water and pool water
(→ disinfection by-product).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 364 Bioanalytical Tools in Water Quality Assessment

Hazard: the inherent capacity of a chemical or mixture to cause adverse effects in

humans or the environment under the conditions of exposure.
Hazard assessment: assessment of the possible adverse effects that could result
from exposure to a hazard (→ risk assessment).
Hazard identification: a process that involves determining the nature and context
of a risk management issue, sometimes referred to as issue identification
(→ risk assessment).
Hazard quotient (HQ): → risk quotient.
Hepatocyte: liver cell.
Hepatotoxicity: liver toxicity.
HHRA: → Human Health Risk Assessment.
HLB: hydrophilic lipophilic balanced co-polymer used for → solid-phase
extraction.
Homeostasis: maintenance of internal stability of conditions (redox and chemical
steady state) within a cell or other system.
HQ: hazard quotient (→ risk quotient).
HT: high throughput.
HTS: high-throughput screening.
HTTK: high-throughput toxicokinetics.
HTTr: high-throughput transcriptomics.
Human Health Risk Assessment (HHRA): process of estimating potential
impact of a chemical, biological or physical agent on a specific human
population (→ Ecological Risk Assessment).
Hydrophilicity: affinity for water. A hydrophilic chemical is more likely to stay in
the aqueous compartments than to be taken up by the lipids in cells.
Hydrophobicity: incompatibility with water. A hydrophobic chemical has high
affinity for the lipids in cells and is more likely to be taken up in the lipid
compartments than to stay in water.
Hyperplasia: excessive cell proliferation.
Hyperthermia: increased body temperature caused by dysfunctional
heat regulation.
Hypoxia: oxygen deficiency.
IA: → independent action.
ICATM: International Cooperation on Alternative Test Methods.
ICCVAM: U.S. Interagency Coordinating Committee on the Validation of
Alternative Methods.
Immortalised cell line: a cell line that has been either accidentally or deliberately
mutated to proliferate indefinitely as opposed to a primary cell line with limited
lifespan (→ primary cell line).
Immunoassay: a technique that allows detection of an antigen with affinity to bind
to a specific antibody (e.g., enzyme-linked immunosorbent assay ELISA,
radioimmunoassay RIAs).
Immunotoxicity: toxicity to the immune system.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 365

In silico: refers to predictive computer models.

In vitro: literally ‘in glass’, refers to tests conducted outside the organism, for
example, using immortal cell lines or tissue/enzymes isolated and performed
in a vial or dish (traditionally in glass but more recently rather in plastic well
plates).
In vivo: refers to tests performed with whole organisms.
Independent action (IA): quantitative model for mixture toxicity of chemicals
that act via different target sites and different modes of toxic action, that is,
the combined toxicity will be less than the sum of the individual effects
(→ concentration addition).
Induction ratio (IR): ratio of signal to signal of unexposed cells in a reporter gene
assays based on transcription factors.
Intercalation (DNA): the inclusion of a large planar molecule between two
DNA bases.
Ionophoric shuttle mechanism: transport (shuttling) of ions across the
cell membrane lipid bilayer by ionophores (lipid-soluble molecules, uncouplers).
IPCS: International Programme on Chemical Safety of the World Health
Organisation (WHO).
ISO: International Organisation for Standardisation.
ITS: Integrated testing strategy, which incorporates multiple lines of investigation
from predictive computer modelling (in silico) and in vitro testing to reduce,
refine and replace whole animal testing.
KE: → key event.
Key event (KE): an observable effect, which is critical to the induction of a
toxicological response following a molecular initiating event (→ molecular
initiation event, → Adverse Outcome Pathway).
LC-MS// MS: liquid chromatography with tandem mass spectrometry.
LD50: lethal dose for 50% of the test animals.
LDH: lactate dehydrogenase.
Leydig cells: testosterone-producing cells in the testis.
Ligand: a molecule with binding affinity for a specific biomolecule such as a
receptor (→ receptor).
Lipid peroxidation: oxidative breakdown of lipids.
LLE: liquid−liquid extraction, a method used in the laboratory to extract
chemicals from water using solvents (→ solid-phase extraction).
LOAEL: lowest observed adverse effect level.
LOD: limit of detection.
LOEC: lowest observed effect concentration.
LOEL: lowest observed effect level.
LOQ: limit of quantification.
Luciferase (Luc): a type of luminescent enzyme that is often utilised in
recombinant cell lines as an easily measurable biomarker.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 366 Bioanalytical Tools in Water Quality Assessment

MAC-EQS: Maximum Acceptable Concentration – Environmental Quality

Standard, a short-term EQS (→ EQS).
Macropollutant: toxicant, usually metal or salt, found in the mg/L to µg/L range
(→ micropollutant).
MAF: → mixture assessment factor.
Margin of safety (MOS): ratio of the acceptable over anticipated exposure
concentration. The larger the MOS, the lower the risk. Sometimes also
referred to as margin of exposure (MOE).
Matrix effects: when components of the sample matrix (water) interfere with
the bioassay.
MCF-7: human breast cancer cell line.
MCR: the maximum cumulative ratio is the ratio between the observed cumulative
toxicity of a mixture and the maximum toxicity caused by an individual
chemical in the mixture.
Mechanism of toxicity: crucial biochemical processes and/or
xenobiotic−biological interactions underlying a given mode of action.
Metabolic activation: metabolic transformation of a chemical that, rather than the
intended detoxification, produces a metabolite that is more toxic than its
precursor. (Also termed bioactivation or biological activation).
Metabolic pathway: cellular pathway involved with metabolism.
Metabolism: the biological production (anabolism) and breakdown (catabolism)
of organic molecules within living organisms. For xenobiotics, the primary
role of metabolism is to catabolise chemicals for excretion. In the case of
toxic chemicals, this can also be termed detoxification although in some
cases, metabolism leads to metabolites more toxic than the precursor
(→ metabolic activation, → phase I and II metabolism).
Metabolite: degradation product of metabolism (also called biotransformation
product, breakdown product).
Metabolomics: global analysis of presence and abundance of low molecular
weight metabolites in cells after exposure to a chemical stressor.
MF: membrane filtration.
Microcystin: a type of cyanotoxin that causes hepatotoxicity through inhibition of
protein phosphatases.
Micropollutant: man-made organic chemicals including pesticides, industrial
chemicals, consumer products and pharmaceuticals but also natural
compounds such as hormones, usually occurring in the sub µg/L range
(→ macropollutant).
Microtitre plate: also referred to as a microplate or microwell plate (e.g., 96- or
384-well plate). A flat plate with multiple small wells applied in cell-based
assays to hold dilution series of samples/standards.
Microtox: commercially available kit to measure bioluminescence inhibition in
naturally luminescent bacteria, Aliivibrio fischeri.
MIE: → molecular initiating event.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 367

Mixture assessment factor (MAF): extrapolation factor that is added to a single

chemical’s risk quotient to account for mixture effects.
Mode of (toxic) action (MOA): a common set of physiological and behavioural
signs that characterise a type of (adverse) biological response; in a more
recent definition related to AOP (→ adverse outcome pathway) is defined as
a biologically plausible series of key events leading to an effect.
MOE: margin of exposure (→ margin of safety).
Molecular initiating event (MIE): the molecular interaction between a xenobiotic
and a biomolecule that starts a cellular toxicity pathway (→ key event,
→ adverse outcome pathway).
Morphogenesis: development of shape.
MOS: → margin of safety.
msPAF: → multi-substance potentially affected fraction.
Multi-substance potentially affected fraction (msPAF): fraction of species
exposed to a mixture concentration above their toxicity threshold.
Mutagenicity: mode of toxic action for toxicants causing mutations
(→ genotoxicity).
Mutation: a change in the genomic sequence by, for example, insertion of
incorrect bases following base excision or strand breaks from DNA damage.
Myelin sheet: electrically insulating layer that forms around the axon of neurons
(nerve cells).
Myelinating cell: a cell forming the myelin sheet (→ myelin sheet).
NADPH: nicotinamide adenine dinucleotide phosphate, a metabolic coenzyme
and electron donor (reducing agent).
NAM: → new approach method.
Narcosis (mode of action): physiological and behavioural responses elicited
through exposure to baseline toxicants. Narcosis in this context refers to the
minimum toxicity exhibited by any compound and is not related to
narcosis/anaesthesia in clinical medicine (→ baseline toxicity).
Native cell: primary or immortalised cell line that has not been genetically
modified (→ primary cell line, → immortalised cell line).
NCATS: National Center for Advancing Translational Sciences.
NCCT: National Center for Computational Toxicology.
Necrosis: unplanned cell death following irreversible damage (as opposed to
programmed cell death, → apoptosis).
Negative control: a control sample of the test solvent or media to ensure these
have no effect on the assay result.
NEL: no effect level.
Neonicotinoids: a group of neurotoxic insecticides.
Nephrotoxicity: toxicity to the kidneys.
Neuron: type of nerve cell that is important for generation and transfer of
information (via neurotransmitters), neurons are supported by glial cells
(→ glial cell, → neurotransmitter, → axon).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 368 Bioanalytical Tools in Water Quality Assessment

Neuronopathy: destruction of neurons.

Neurotoxicity: toxicity to the nervous system.
Neurotransmitter: endogenous chemical (e.g., acetylcholine) that transmits
information from neurons to target cells.
New approach method (NAM): any non-animal-based approaches that can be
used to provide information in the context of chemical hazard and risk
assessment.
Next-generation risk assessment (NGRA): risk assessment strategy that relies on
→ HTS methods.
NGRA: → next-generation risk assessment.
NIEHS: National Institute of Environmental Health Science.
Nitrosamines: a group of chemical compounds (with the structure R1N
(-R2)-N=O) found in many common products such as rubbers, tobacco and
foods (e.g., formed in cured meats from amines in the meat and the additive
sodium nitrite). Nitrosamines are also → DBPs formed during
chloramination of drinking water. Many nitrosamines are carcinogenic.
NOAEL: no observed adverse effect level.
NOEC: no observed effect concentration.
NOEL: no observed effect level.
Non-specific (mode of action): physiological and behavioural responses elicited
through exposure to baseline toxicants, often used synonymous to ‘narcosis’
(→ non-specific toxicity, → baseline toxicity).
Non-specific toxicity: → baseline toxicity.
Non-threshold chemical: a chemical for which it is assumed that there is no safe
level of exposure at which there is no effect (e.g., carcinogens) (→ threshold
chemical).
NPDES: National Pollutant Discharge Elimination System (U.S.).
Nrf2: NF-E2-related factor 2, a transcription factor for defence against
oxidative stress.
NRU: Neutral Red uptake assay using dye to assess cell viability.
NTP: National Toxicology Program of the → NIEHS (U.S.).
Nuclear receptor: a protein receptor that is capably of sensing hormones and of
binding directly to DNA, thereby regulating gene expression (→ receptor).
Nuclear xenobiotic metabolism receptor: a protein receptor that senses
xenobiotics that induce the gene expression of metabolic enzymes
(→ nuclear receptor).
NWQMS: National Water Quality Management Strategy (Australia).
O3: ozone (refers to water treatment by ozonation).
OECD: Organisation for Economic Co-operation and Development.
Organogenesis: the development of organs.
Organophosphate pesticide: a type of insecticide that causes neurotoxicity via
inhibition of the enzyme AChE (→ acetylcholinesterase).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 369

Oxidative stress: imbalance in the level of reactive oxygen species and the
system’s capacity for detoxification (→ reactive oxygen species).
P450 enzymes: → CYP450.
P53: family of transcription factors for an important adaptive stress response
pathway for DNA damage (→ genotoxicity). (Also called ‘tumour suppressor
gene’).
PAH: polycyclic aromatic hydrocarbon.
Passive dosing: a technique used in tests with hydrophobic chemicals, where the
test compound is added via a solid phase to maintain a constant exposure
concentration in the cell medium (also termed partition-controlled dosing).
Passive sampling: time-integrated sampling of water through deployment of
passive sampling devices containing sorbent material with affinity for groups
of chemicals with similar physicochemical properties.
Passive transport: passive diffusion of molecules across cell membranes via
a concentration gradient from high to low concentration of the substance
(→ active transport).
Pathogen: microorganism(s) causing disease in plants, animals and humans.
Pathway of toxicity: essentially synonymous to → toxicity pathway; cellular
processes that mediate adverse outcomes of toxicants.
PBDE: polybrominated diphenyl ether. Group of structurally similar brominated
compounds also referred to as brominated flame retardants.
PBT: persistent, bioaccumulative and toxic.
PBTK: physiologically based toxicokinetic model (→ toxicokinetics).
PCB: polychlorinated biphenyl. Group of 209 structurally similar industrial
chemicals that were previously produced and used in large quantities in, for
example, electrical appliances. Although these compounds have been banned
as POPs/PBTs under the Stockholm Convention, traces of PCBs are still
found in the environment.
PCDD: → polychlorinated dibenzodioxin.
PCDF: → polychlorinated dibenzofuran.
Phagocytes: white blood cells capable of eliminating many microorganisms
by absorption.
Phase I metabolism: biotransformation of chemicals via oxidation, reduction and
hydrolysis (→ metabolism).
Phase II metabolism: conjugation of the functional groups added in phase I
metabolism with molecular entities such as sulphate and glucuronic acid to
yield highly water-soluble metabolites, which are more easily excreted from
the body (→ metabolism).
Phenotype: external characteristics of an organism that is the expression of its
genotype, that is, its DNA make-up.
Photodegradation: the breakdown of organic chemicals through absorption of
photons during sunlight exposure.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 370 Bioanalytical Tools in Water Quality Assessment

Photosynthesis: the conversion of carbon dioxide and water to sugars and oxygen
by plants, algae and some bacteria, using sunlight energy.
Photosystem II: a protein complex that delivers electrons for photosynthesis to
occur (→ photosynthesis).
Phytotoxicity: toxicity to plants.
Plasmid: circular DNA molecule, which carries a responsive element for a
receptor of interest, followed by a reporter gene that encodes a measurable
feature such as an enzyme or green fluorescence protein.
PNEC: predicted no effect concentration.
Point of departure (POD): point on a dose−response curve corresponding to an
estimated low effect level or no effect level.
Point of inflexion: the point on a curve at which the slope changes.
Polychlorinated dibenzodioxin (PCDD): a group of structurally similar
chlorinated compounds that are formed as by-products during the production
of other chlorinated compounds such as some pesticides. The best-known
example is 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), which was a major
contaminant in Agent Orange, the herbicide used in the Vietnam war.
Although these compounds have been banned as POPs/PBTs under the
→ Stockholm Convention, traces are still found in the environment.
Polychlorinated dibenzofuran (PCDF): a group of dioxin-like persistent organic
pollutants (→ Stockholm Convention) structurally similar to
→ polychlorinated dibenzodioxins.
Polyhalogenated biphenyl: brominated, chlorinated and fluorinated biphenyls,
for example, polychlorinated biphenyl (→ PCB).
POP: persistent organic pollutant (→ Stockholm Convention).
Positive control: a sample that contains a supra-maximal concentration of a test
compound that is not the reference compound (→ supra-maximal concentration).
PPAR: peroxisome proliferator receptor, involved with metabolism of glucose,
lipids and fatty acids.
PR: progesterone receptor, important for development and reproduction
(fertility) and induced by progestogens and progestogen-like chemicals
(→ progestogens).
Primary cell line: cell line isolated from living tissue. Most primary cell cultures
have a limited lifespan (i.e., have not been immortalised) (→ immortalised cell
line).
Primary mechanism, primary effects: the type and degree of interaction of a
toxicant with biomolecules at the target site.
Progestogens: steroid hormones such as progesterone that are important regulators
of, for example, pregnancy and menstruation (→ PR).
Prokaryote: a single-celled organism that has no distinct nucleus (e.g., bacteria,
cyanobacteria) (→ eukaryote).
Promoter: a region in DNA that regulates transcription of a specific gene.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 371

Protection-goal-motivated approach: a test battery design approach that targets

the protection goal (e.g., human health or ecosystem health) rather than specific
chemical groups (→ chemical-group-motivated approach).
Proteolytic enzyme: an enzyme that splits proteins into their constitutive building
blocks, that is, peptides or amino acids.
Proteomics: global analysis of presence and abundance of functional proteins in
cells after exposure to a chemical stressor.
PRW: purified recycled water.
PXR: pregnane X receptor, involved in the induction of various phase I enzymes
(→ phase I metabolism).
Pyrethroids: group of neurotoxic insecticides.
QA// QC: quality assurance and quality control.
qAOP: quantitative adverse outcome pathway (→adverse outcome pathway),
(→ AOP).
QIVIVE: → quantitative in vitro to in vivo extrapolation.
QS: quality standard.
QSAR: quantitative structure−activity relationship.
Quantitative in vitro to in vivo extrapolation (QIVIVE): quantitative model that
extrapolates from in vitro data to potential in vivo effects in humans.
Quinolones: group of synthetic antibiotics.
Radioimmunoassay (RIA): an immunoassay using radiolabelling to enable easy
detection of an antigen.
RAR: retinoic acid receptor, important for regulation of, for example,
development and homeostasis.
REACH: The European chemicals regulation titled Registration, Evaluation,
Authorisation and Restriction of Chemical Substances.
Reactive oxygen species (ROS): endogenously produced oxygen containing
reactive molecules including the superoxide radical (O2•−), hydrogen
peroxide (H2O2) and the hydroxyl radical (OH•). Chemical and other stress
can increase ROS formation to levels that may cause lipid peroxidation,
DNA damage and oxidation of proteins, followed by loss of enzymatic activity.
Reactive toxicity: mode of toxic action that is associated with chemical reactions
where covalent bonds are formed. Direct reactivity is the direct reaction of
electrophilic chemicals with biological nucleophiles such as DNA bases or
proteins. Indirect reactivity is the formation of reactive oxygen species
(ROS) that are potent oxidants.
Read-across: a technique for predicting the properties of one chemical by using
data from another chemical.
Receptor binding assay (RBA): an assay that measures the competitive binding
of a chemical or environmental sample and a native ligand molecule to a
receptor (→ receptor, → ligand).
Receptor: a protein to which certain ligands (e.g., hormones and
hormone-mimicking chemicals) can bind. Each receptor is specific to
binding of ligands with particular structure(s) (→ nuclear receptor).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 372 Bioanalytical Tools in Water Quality Assessment

Receptor-mediated toxicity: toxicity caused by induction of a specific receptor

(e.g., binding of xenoestrogen to the estrogen receptor) (→ receptor,
→ nuclear receptor).
Recombinant cell lines: cell lines that are created by insertion of a reporter
plasmid, which carries a responsive element for a particular receptor
followed by a reporter gene encoding a measurable marker (e.g., green
fluorescent protein).
REF: → relative enrichment factor, or relative extraction factor.
Reference dose (RfD): an estimate (with uncertainty factors) of the daily exposure
(mg/kg/day) to the general population that is without harm (→ acceptable daily
intake, → tolerable daily intake).
Relative enrichment factor (REF): is a measure of sample concentration that
takes into account the enrichment of the sample that occurs during extraction
and clean up, relative to the sample dilution that takes place during the assay.
Product of → extraction factor (EF) and → dosing factor (DFbioassay).
Relative extraction factor (REF): synonym for → relative enrichment factor.
Relative fluorescence units (RFU): fluorescence measurements with a plate
reader that are not calibrated but are instrument-specific.
Relative light units (RLU): luminescence measurements that are specific to the
luminometer and are therefore relative and can only be compared in the
sample instrument.
REP: the relative effect potency is a measure of the relative toxicity of one
chemical compared with another (often a reference compound) established
through comparison of the individual concentration−effect curves.
Reporter gene: a gene with a particular characteristic that can be utilised
through insertion into a gene that would not otherwise express this feature.
An example of a reporter gene is the green fluorescent protein (GFP), which
is introduced to cells in order to encode a measurable marker
(→ recombinant cell).
Reporter plasmid: a transferable form of DNA that is separate to and independent
of chromosomal DNA.
Responsive element: or hormone responsive element (HRE, or, for example, ERE
for the estrogen receptor); a short sequence of DNA capable of binding to
certain receptors (e.g., ER), thereby regulating transcription.
Reverse osmosis (RO): a process that applies water pressure to force water
movement over a membrane filter, which is impassable by many (but not
all) micropollutants.
RfD: → reference dose.
RFU: → relative fluorescence units
rGTU: relative genotoxic unit (similar to → toxic unit, but specifically for
genotoxicity).
RIA: → radioimmunoassay.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 373

Risk assessment: a process, which entails the following elements: hazard

identification, effects assessment, exposure assessment and risk
characterisation.
Risk characterisation: estimate of the incidence and severity of the adverse
effects likely to occur in a human population or environmental compartment
due to actual or predicted exposure to a substance, and may include ‘risk
estimation’, that is, the quantification of that likelihood (→ risk assessment).
Risk index (RI): cumulative risk measure, the sum of risk quotients (→ RQ).
Sometimes also called hazard index (HI) (→ risk quotient).
Risk management: a decision-making process that entails weighing political,
social, economic, and engineering information against risk-related
information to develop, analyse and compare regulatory options and select
the appropriate regulatory response to a potential health or environmental
hazard (→ risk assessment).
Risk quotient (RQ): ratio of the exposure estimate over an acceptable effect level
(also called hazard quotient HQ) (→ risk index, → hazard quotient).
Risk reduction: taking measures to protect humans and/or the environment from
the risks identified.
Risk: the probability of an adverse effect on humans or the environment occurring
as a result of a given exposure to a chemical or mixture.
RLU: → relative light units
RNA: ribonucleic acid, the macromolecule controlling, for example, gene
expression and protein synthesis. DNA messenger.
RO: → reverse osmosis.
ROS: → reactive oxygen species.
RQ: → risk quotient.
rTU: relative toxic unit (similar to → toxic unit).
RXR: retinoid X receptor, forms heterodimers with many other nuclear receptors
such as the RAR and thus has many regulatory functions (→ RAR).
S9: liver enzyme mix containing a wide range of metabolising enzymes including
CYP450s. S9 mix is used in bioassays to study the effect of metabolism
on xenobiotics, that is, some xenobiotics need metabolic activation
(→ metabolic activation).
SCGE: → single cell gel electrophoresis
Sertoli cells: nurse cells within the testis.
Single cell gel electrophoresis (SCGE): an assay that detects DNA strand breaks
in different cell types including human, other mammalian and fish cells. Also
known as the Comet assay (→ Comet assay).
SMPD: serum-mediated passive dosing.
SOP: standard operating protocol (or procedure).
SOS Chromo: assay for measuring the SOS response in bacterial cells (→ SOS
response).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 374 Bioanalytical Tools in Water Quality Assessment

SOS response: defence strategy of a cell in response to DNA damage. Cell assays
targeting the SOS response include the umuC, SOS/umu and SOS Chromo.
SPE: solid-phase extraction, a laboratory method that uses cartridges packed with
specialised sorbents to extract chemicals from water (→ LLE).
Specific mode of toxic action: a mode of toxic action caused by specific
interaction with receptors or enzymes that result in higher toxicity than
baseline toxicity.
Specificity ratio (SR): ratio between the inhibitory concentration predicted for
baseline toxicity and experimental effect concentration. Chemicals with an
SR close to 1 act as baseline toxicants, while chemicals with a high SR are
specifically active in this particular endpoint.
Spermatogenesis: production and development of mature sperm cells.
SPM: → suspended particulate matter.
SPR: → suppression ratio.
SR: → specificity ratio.
Stable transfection: stable transfer of genetic material into a cell as opposed to
transient transfection, which is unstable after reproduction.
Stockholm Convention: a global treaty to ban release and minimise exposure to
persistent organic pollutants.
Super-minimal concentration: the highest concentration causing an effect that is
not statistically different from the negative control (→ NOEC).
Suppression ratio (SPR): effect of an antagonist in a reporter gene assay run in
antagonism mode, in some studies also called SR (to be avoided, might be
confused with → suppression ratio).
Supra-maximal concentration: the lowest concentration causing 100% effect.
Suspended particulate matter (SPM): small particles composed of minerals and
organic matter that are suspended in surface water
Synapse: junction between two nerve cells.
Synergy or synergism: when the combined toxicity of two or more toxicants is
higher than the sum of the individual effects.
TCA: → tricarboxylic acid.
TCDD: 2,3,7,8-tetrachloro-dibenzodioxin.
tcpl: → ToxCast analysis pipeline.
TD: → toxicodynamics.
TDI: tolerable daily intake (→ acceptable daily intake, → reference dose).
TEF: toxic equivalency factor.
TEQ: → toxic equivalent concentration.
TEQbio: bioassay-derived TEQ (→ toxic equivalent concentration).
TEQchem: chemical analysis-derived TEQ (→ toxic equivalent concentration).
Teratogenesis: interference with embryonic or foetal development resulting in
pre-natal or birth defects.
TF: → transcription factor.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 375

THP1: human acute monocytic leukaemia cell line (monocyte cancer cell – the
monocyte is a precursor to macrophages, a type of white blood cell).
THP1-CPA: THP1 cytokine production assay (→ THP1, → cytokine), a measure
of immunotoxicity.
Threshold chemical: a chemical for which it is assumed that there is a safe dose or
concentration, below which there is no appreciable risk to exposed organisms
(→ non-threshold chemical).
Threshold of Toxicological Concern (TTC): level of human intake or exposure
that is considered to be of negligible risk to human health.
TIE: → toxicity identification evaluation.
TIF2: human co-activator (→ co-activator).
TK: → toxicokinetics.
TMX: tamoxifen, an anti-estrogenic drug used in treatment of hormone-related
cancer (→ estrogen).
TOC: total organic carbon.
TOX: total organic halogen compounds.
Tox21: high-throughput data initiative, collaboration between the National Center
for Computational Toxicology (NCCT) of the U.S. EPA, the National
Toxicology Program of the National Institute of Environmental Health
Science (NIEHS) and the National Center for Advancing Translational
Sciences (NCATS) and the U.S. Food and Drug Administration (FDA).
ToxCast analysis pipeline (tcpl): specific data evaluation pipeline in R for
concentration−response modelling of high-throughput screening data
developed by the → ToxCast project. (→ tcpl).
ToxCast: Toxicity ForeCaster project, launched by the National Center for
Computational Toxicology (NCCT) of the U.S. EPA.
Toxic equivalent concentration (TEQ): the concentration of a reference
chemical that would elicit the same effect toxicity as the mixture of
micropollutants in a water sample.
Toxic unit (TU): the toxic unit from chemical analysis (TUchem) is the ratio of
concentration of a chemical concentration divided by its toxic concentration,
e.g. → LC50, and the toxic unit from bioassays (TUbio) is the inverse of the
toxic concentration of a mixture or environmental samples (1/LC50)
(→ effect unit).
Toxicity identification evaluation (TIE): a procedure that combines multiple
fractionation and bioassay testing to isolate a toxic compound from a mixture
(→ effect-directed analysis).
Toxicity pathway: a cellular response pathway that, when sufficiently perturbed,
is expected to result in an adverse health outcome (→ pathway of toxicity,
→ adverse outcome pathway).
Toxicodynamics (TD): the actual toxicity pathways taking place inside the cell
including the initial molecular interaction of the chemical and its biological
target (→ toxicokinetics).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 376 Bioanalytical Tools in Water Quality Assessment

Toxicogenomics: the application of genomics in toxicological research.

Toxicokinetics (TK): the kinetic processes of uptake, distribution, metabolism
and elimination that link external exposure (e.g., via drinking water or eating
food) to biologically effective concentration within a cell. These processes
include absorption, excretion, internal distribution and metabolism
(→ ADME) of a chemical within the whole body and within cells.
TR: thyroid receptor, important for regulation of heart rate, metabolism
and development.
Transcription factor (TF): a protein responsible for transcription of an adaptive
stress response pathway.
Transcription: generation of a RNA copy of DNA, the first step in
gene expression.
Transcriptomics: expression profiling, that is, global analysis of RNA levels in
cells after exposure to a chemical stressor.
Tricarboxylic acid (TCA) cycle: series of biochemical reaction used by all
aerobic organisms to release stored chemical energy. Takes place in the
mitochondria. Also known as the citric acid cycle.
Trihalomethanes (THM): group of disinfection by-products formed from natural
organic matter during chemical disinfection of drinking water and pool water
(→ disinfection by-product).
Trophic level: position that an organism occupies in a food chain. The lowest
trophic level are primary producers (photosynthetic organisms), the highest
trophic levels are organisms that feed on other carnivores.
TTC: → threshold of toxicological concern.
TU: → toxic unit.
Tumour: abnormal mass of tissue. Malignant tumours can lead to cancer.
U.S. EPA: U.S. Environmental Protection Agency.
umuC: assay commonly used in water monitoring to measure SOS response in
bacterial cells, also called SOS/umu (→ SOS response).
Uncoupler: chemical that disturbs mitochondrial energy transduction by a
protonophoric shuttle mechanism.
Vasoactive agent: an agent capable of reducing or increasing blood pressure.
Vitellogenin (Vtg): a phospholipoprotein precursor of egg yolk, Vtg is stimulated
by exposure to endogenous and exogenous estrogens (estrogenic compounds).
Vtg is often measured in vivo but can be used in cell-based assays (→ endocrine
disruption).
WET: → whole effluent toxicity.
WFD: European Union Water Framework Directive.
WHO: World Health Organisation.
Whole effluent toxicity (WET) testing: a procedure where the toxicity of water is
tested exposing whole organisms to un-extracted water (→ direct toxicity
assessment).
WWTP: wastewater treatment plant.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Glossary 377

Xenobiotic receptor: a receptor that regulates metabolism of xenobiotics (e.g.,

AhR).
Xenobiotic: foreign substance.
Xenoestrogen: exogenous estrogen (→ estrogen).
Yeast estrogen screen (YES): common assay used to test water samples for
estrogenicity, which is measured via induction of human ER in recombinant
yeast (→ ER, → recombinant cell lines).
Yeast two-hybrid assay: applies a recombinant yeast cell line, which has been
transfected with two different reporter plasmids (→ reporter plasmid,
→ recombinant cell).
Zona radiata protein (Zrp): eggshell protein.
96- or 384-well plate: → microtitre plate.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf
by guest
 References

Abbas A., Schneider I., Bollmann A., Funke J., Oehlmann J., Prasse C., Schulte-Oehlmann
U., Seitz W., Ternes T., Weber M., Wesely H. and Wagner M. (2019). What you
extract is what you see: Optimising the preparation of water and wastewater samples
for in vitro bioassays. Water Research, 152, 47–60.
Ademollo N., Patrolecco L., Polesello S., Valsecchi S., Wollgast J., Mariani G. and Hanke G.
(2012). The analytical problem of measuring total concentrations of organic pollutants in
whole water. TrAC-Trends in Analytical Chemistry, 36, 71–81.
Aerni H. R., Kobler B., Rutishauser B. V., Wettstein F. E., Fischer R., Giger W.,
Hungerbuhler A., Marazuela M. D., Peter A., Schonenberger R., Vogeli A. C., Suter
M. J. F. and Eggen R. I. L. (2004). Combined biological and chemical assessment of
estrogenic activities in wastewater treatment plant effluents. Analytical and
Bioanalytical Chemistry, 378(3), 688–696.
Aguayo S., Muñoz M. J., de la Torre A., Roset J., de la Peña E. and Carballo M. (2004).
Identification of organic compounds and ecotoxicological assessment of sewage
treatment plants (STP) effluents. Science of the Total Environment, 328(1–3), 69–81.
Ahn K. S. and Aggarwal B. B. (2005). Transcription factor NF-κB – A sensor for smoke
and stress signals. In: Natural Products and Molecular Therapy G. J. Kotwal and
D. K. Lahiri, Wiley, Hoboken, NJ, USA, pp. 218–233.
Al Hanai A. H., Antkiewicz D. S., Hemming J. D. C., Shafer M. M., Lai A. M., Arhami M.,
Hosseini V. and Schauer J. J. (2019). Seasonal variations in the oxidative stress and
inflammatory potential of PM2.5 in Tehran using an alveolar macrophage model; The
role of chemical composition and sources. Environment International, 123, 417–427.
Albergamo V., Schollee J. E., Schymanski E. L., Helmus R., Timmer H., Hollender J. and
de Voogt P. (2019). Nontarget screening reveals time trends of polar micropollutants in
a riverbank filtration system. Environmental Science & Technology, 53(13), 7584–7594.
Albergamo V., Escher B. I., Schymanski E. L., Helmus R., Dingemans M. M. L., Cornelissen
E. R., Kraak M. H. S., Hollender J. and de Voogt P. (2020). Evaluation of reverse

© IWA Publishing 2021. Bioanalytical Tools in Water Quality Assessment

Authors: Beate Escher, Peta Neale and Frederic Leusch
doi: 10.2166/9781789061987_0379

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 380 Bioanalytical Tools in Water Quality Assessment

osmosis drinking water treatment of riverbank filtrate using bioanalytical tools and
non-target screening. Environmental Science-Water Research & Technology, 6(1),
103–116.
Aleem A. and Malik A. (2003). Genotoxicity of water extracts from the River Yamuna at
Mathura, India. Environmental Toxicology, 18(2), 69–77.
Aleem A. and Malik A. (2005). Genotoxicity of the Yamuna River water at Okhla (Delhi),
India. Ecotoxicology and Environmental Safety, 61(3), 404–412.
Allan H. L., van de Merwe J. P., Finlayson K. A., O’Brien J. W., Mueller J. F. and Leusch
F. D. L. (2017). Analysis of sugarcane herbicides in marine turtle nesting areas and
assessment of risk using in vitro toxicity assays. Chemosphere, 185, 656–664.
Allermann L. and Poulsen O. M. (2000). Inflammatory potential of dust from waste handling
facilities measured as IL-8 secretion from lung epithelial cells in vitro. Annals of
Occupational Hygiene, 44(4), 259–269.
Allinson G., Allinson M., Salzman S., Shiraishi F., Myers J., Theodoropoulos T., Hermon K.
and Wightwick A. (2007). Hormones in recycled water. Final report. DPI (Department
of Primary Industries), Queenscliff, Australia https://dro.deakin.edu.au/eserv/
DU:30010499/salzman-hormonesintreated-2007.pdf. (Accessed 10 April 2021).
Altenburger R., Backhaus T., Boedeker W., Faust M., Scholze M. and Grimme L. H. (2000).
Predictability of the toxicity of multiple chemical mixtures to Vibrio fischeri: mixtures
composed of similarly acting chemicals. Environmental Toxicology and Chemistry,
19(9), 2341–2347.
Altenburger R., Scholz S., Schmitt-Jansen M., Busch W. and Escher B. I. (2012). Mixture
toxicity revisited from a toxicogenomic perspective. Environmental Science &
Technology, 46(5), 2508–2522.
Altenburger R., Scholze M., Busch W., Escher B. I., Jakobs G., Krauss M., Krüger J., Neale
P. A., Aït-Aïssa S., Almeida A. C., Seiler T.-B., Brion F., Hilscherová K., Hollert H.,
Novák J., Schlichting R., Serra H., Shao Y., Tindall A. J., Tolefsen K. E., de Aragão
Umbuzeiro G., Williams T. D. and Kortenkamp A. (2018). Mixture effects in samples
of multiple contaminants – an inter-laboratory study with manifold bioassays.
Environment International, 114, 95–116.
Alygizakis N. A., Besselink H., Paulus G. K., Oswald P., Hornstra L. M., Oswaldova M.,
Medema G., Thomaidis N. S., Behnisch P. A. and Slobodnik J. (2019).
Characterization of wastewater effluents in the Danube River Basin with chemical
screening, in vitro bioassays and antibiotic resistant genes analysis. Environment
International, 127, 420–429.
Alyzakis N. A., Samanipour S., Hollender J., Ibanez M., Kaserzon S., Kokkali V., van
Leerdam J. A., Mueller J. F., Pijnappels M., Reid M. J., Schymanski E. L., Slobodnik
J., Thomaidis N. S. and Thomas K. V. (2018). Exploring the potential of a global
emerging contaminant early warning network through the use of retrospective suspect
screening with high-resolution mass spectrometry. Environmental Science &
Technology, 52(9), 5135–5144.
Ames B. N., McCann J. and Yamasaki E. (1975). Methods for detecting carcinogens and
mutagens with the salmonella/mammalian-microsome mutagenicity test. Mutation
Research/Environmental Mutagenesis and Related Subjects, 31(6), 347–363.
Andersen M. E., Al-Zoughool M., Croteau M., Westphal M. and Krewski D. (2010). The
future of toxicity testing. Journal of Toxicology and Environmental Health-Part
B-Critical Reviews, 13(2–4), 163–196.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 381

Ankley G. T., Daston G. P., Degitz S. J., Denslow N. D., Hoke R. A., Kennedy S. W., Miracle
A. L., Perkins E. J., Snape J., Tillitt D. E., Tyler C. R. and Versteeg D. (2006).
Toxicogenomics in regulatory ecotoxicology. Environmental Science & Technology,
40(13), 4055–4065.
Ankley G. T., Bennett R. S., Erickson R. J., Hoff D. J., Hornung M. W., Johnson R. D., Mount
D. R., Nichols J. W., Russom C. L., Schmieder P. K., Serrrano J. A., Tietge J. E. and
Villeneuve D. L. (2010). Adverse outcome pathways: a conceptual framework to
support ecotoxicology research and risk assessment. Environmental Toxicology and
Chemistry, 29(3), 730–741.
ANZECC/ARMCANZ (2000). Australian and New Zealand Guidelines for Fresh and Marine
Water Quality. Vol. 1, Australian and New Zealand Environment Conservation Council
and Agriculture and Resource Management Council of Australia and New Zealand,
Canberra. www.waterquality.gov.au/anz-guidelines/resources/previous-guidelines/
anzecc-armcanz-2000. (Accessed 18 February 2021).
Apel P., Kortenkamp A., Koch H. M., Vogel N., Ruther M., Kasper-Sonnenberg M., Conrad
A., Bruning T. and Kolossa-Gehring M. (2020). Time course of phthalate cumulative
risks to male developmental health over a 27-year period: Biomonitoring samples
of the German Environmental Specimen Bank. Environment International, 137,
105467.
Archer E., Wolfaardt G. M., van Wyk J. H. and van Blerk N. (2020). Investigating (anti)
estrogenic activities within South African wastewater and receiving surface waters:
Implication for reliable monitoring. Environmental Pollution, 263, 114424.
Armitage J. M., Wania F. and Arnot J. A. (2014). Application of mass balance models and the
chemical activity concept to facilitate the use of in vitro toxicity data for risk assessment.
Environmental Science & Technology, 48, 9770–9779.
Ashauer R., O’Connor I. and Escher B. I. (2017). Toxic mixtures in time – the sequence
makes the poison. Environmental Science & Technology, 51(5), 3084–3092.
Atkinson A. J., Colburn W. A., DeGruttola V. G., DeMets D. L., Downing G. J., Hoth D. F.,
Oates J. A., Peck C. C., Schooley R. T., Spilker B. A., Woodcock J. and Zeger S. L. and
Biomarkers Definitions Working G. (2001). Biomarkers and surrogate endpoints:
Preferred definitions and conceptual framework. Clinical Pharmacology &
Therapeutics, 69(3), 89–95.
Atterwill C. K., Bruinink A., Drejer J., Duarte E., Abdulla E. M., Meredith C., Nicotera P.,
Regan C., Rodriguezfarre E., Simpson M. G., Smith R., Veronesi B., Vijverberg H.,
Walum E. and Williams D. C. (1994). In vitro neurotoxicity testing – The report and
recommendations of ECVAM workshop-3. Atla-Alternatives to Laboratory Animals,
22(5), 350–362.
Auld D., Coassin P. A., Coussens N. P., Hensley P., Klumpp-Thomas C., Michael S.,
Sittampalam G. S., Trask O. J., Wagner B. K., Weidner J. R., Wildey M. J. and
Dahlin J. L. (2020). Microplate selection and recommended practices in
high-throughput screening and quantitative biology. In: Assay Guidance Manual
[Internet], S. Markossian, G. Sittampalam and A. Grossman (eds.), Eli Lilly &
Company and the National Center for Advancing Translational Sciences, Bethesda
(MD). www.ncbi.nlm.nih.gov/books/NBK558077/ (Accessed on 2 January 2021).
Australian Government (2018a). Australian and New Zealand Guidelines for Fresh and
Marine Water Quality. Water Quality Policy Sub Committee (WQPSC) and National

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 382 Bioanalytical Tools in Water Quality Assessment

Water Reform Committee (NWRC), Australian Government, Canberra, ACT. www.

waterquality.gov.au/anz-guidelines (Accessed 2 January 2021).
Australian Government (2018b). Charter: National Water Quality Management Strategy.
Department of Agriculture and Water Resources, Canberra. www.waterquality.gov.
au/about/charter (Accessed 2 January 2021).
Backhaus T. and Grimme L. H. (1999). The toxicity of antibiotic agents to the luminescent
bacterium Vibrio fischeri. Chemosphere, 38(14), 3291–3301.
Backhaus T. and Karlsson M. (2014). Screening level mixture risk assessment of
pharmaceuticals in STP effluents. Water Research, 49, 157–165.
Backhaus T., Altenburger R., Boedeker W., Faust M., Scholze M. and Grimme L. H. (2000).
Predictability of the toxicity of multiple mixtures of dissimilarly acting chemicals to
Vibrio fischeri. Environmental Toxicology and Chemistry, 19(9), 2348–2356.
Baduel C., Mueller J. F., Tsai H. H. and Ramos M. J. G. (2015). Development of sample
extraction and clean-up strategies for target and non-target analysis of environmental
contaminants in biological matrices. Journal of Chromatography A, 1426, 33–47.
Bahamonde P. A., Feswick A., Isaacs M. A., Munkittrick K. R. and Martyniuk C. J. (2016).
Defining the role of omics in assessing ecosystem health: perspectives from the
Canadian environmental monitoring program. Environmental Toxicology and
Chemistry, 35(1), 20–35.
Bailey H. C., Elphick J. R., Krassoi R., Mulhall A. M., Lovell A. J. and Slee D. J. (2005).
Identification of chlorfenvinphos toxicity in a municipal effluent in Sydney, New
South Wales, Australia. Environmental Toxicology and Chemistry 24(7), 1773–1778.
Bain P. A., Williams M. and Kumar A. (2014). Assessment of multiple hormonal activities in
wastewater at different stages of treatment. Environmental Toxicology and Chemistry,
33(10), 2297–2307.
Baken K. A., Sjerps R. M. A., Schriks M. and van Wezel A. P. (2018). Toxicological risk
assessment and prioritization of drinking water relevant contaminants of emerging
concern. Environment International, 118, 293–303.
Baker M. A., Cerniglia G. J. and Zaman A. (1990). Microtiter plate assay for the measurement
of glutathione and glutathione disulfide in large numbers of biological samples.
Analytical Biochemistry, 190(2), 360–365.
Ballantyne B., Marrs T. and Syversen T. (1995). General and Applied Toxicology – Abridged
edition. Macmillan Press, London, UK.
Bal-Price A. and Coecke S. (2011). Guidance on Good Cell Culture Practice (GCCP). In: Cell
Culture Techniques, M. Aschner, C. Sunol and A. Bal-Price (eds.), Springer,
Heidelberg, Germany, pp. 1–25.
Baltazar M. T., Cable S., Carmichael P. L., Cubberley R., Cull T., Delagrange M., Dent M. P.,
Hatherell S., Houghton J., Kukic P., Li H. Q., Lee M. Y., Malcomber S., Middleton A. M.,
Moxon T. E., Nathanail A. V., Nicol B., Pendlington R., Reynolds G., Reynolds J., White
A. and Westmoreland C. (2020). A next-generation risk assessment case study for
Coumarin in cosmetic products. Toxicological Sciences, 176(1), 236–252.
Bambino K. and Chu J. (2017). Zebrafish in Toxicology and Environmental Health. In:
Zebrafish at the Interface of Development and Disease Research, K. C. Sadler (ed.),
Elsevier, pp. 331–367.
Bandelj E., Van den Heuvel M. R., Leusch F. D. L., Shannon N., Taylor S. and McCarthy
L. H. (2006). Determination of the androgenic potency of whole effluents using
mosquitofish and trout bioassays. Aquatic Toxicology, 80(3), 237–248.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 383

Barrueco C., Herrera A. and de la Pea E. (1991). Mutagenic evaluation of trichlorfon using
different assay methods with Salmonella typhimurium. Mutagenesis, 6(1), 71–76.
Bartram J., Corrales L., Davison A., Deere D., Drury D., Gordon B., Howard G., Rinehold A.
and Stevens M. (2009). Water Safety Plan Manual: Step-by-Step Risk Management for
Drinking-Water Suppliers. World Health Organization, Geneva. www.preventionweb.
net/publications/view/8367 (Accessed 10 April 2021).
Baumer A., Bittermann K., Klüver N. and Escher B. (2017). Baseline toxicity and ion
trapping models to describe the pH-dependence of bacterial toxicity of
pharmaceuticals. Environmental Science: Processes & Impacts, 19, 901–916.
Beckers L. M., Brack W., Dann J. P., Krauss M., Müller E. and Schulze T. (2020). Unraveling
longitudinal pollution patterns of organic micropollutants in a river by non-target
screening and cluster analysis. Science of the Total Environment, 727, 138388.
Behnisch P. A., Hosoe K. and Sakai S.-i. (2001). Bioanalytical screening methods for dioxins
and dioxin-like compounds – a review of bioassay/biomarker technology. Environment
International, 27(5), 413–439.
Bell S. M., Chang X. Q., Wambaugh J. F., Allen D. G., Bartels M., Brouwer K. L. R., Casey
W. M., Choksi N., Ferguson S. S., Fraczkiewicz G., Jarabek A. M., Ke A., Lumen A.,
Lynn S. G., Paini A., Price P. S., Ring C., Simon T. W., Sipes N. S., Sprankle C. S.,
Strickland J., Troutman J., Wetmore B. A. and Kleinstreuer N. C. (2018). In vitro to
in vivo extrapolation for high throughput prioritization and decision making.
Toxicology in Vitro, 47, 213–227.
Bellet V., Hernandez-Raquet G., Dagnino S., Seree L., Pardon P., Bancon-Montiny C., Fenet
H., Creusot N., Ait-Aissa S., Cavailles V., Budzinski H., Antignac J.-P. and Balaguer P.
(2012). Occurrence of androgens in sewage treatment plants influents is associated with
antagonist activities on other steroid receptors. Water Research, 46(6), 1912–1922.
Bengtson Nash S. M., Goddard J. and Muller J. F. (2006). Phytotoxicity of surface waters of
the Thames and Brisbane River Estuaries: A combined chemical analysis and bioassay
approach for the comparison of two systems. Biosensors & Bioelectronics, 21(11),
2086–2093.
Bermudez D. S., Gray L. E. and Wilson V. S. (2012). Modelling defined mixtures of
environmental oestrogens found in domestic animal and sewage treatment effluents
using an in vitro oestrogen-mediated transcriptional activation assay (T47D-KBluc).
International Journal of Andrology, 35(3), 397–406.
Bernhard K., Stahl C., Martens R., Kohler H. R., Triebskorn R., Scheurer M. and Frey M.
(2017). Two novel real time cell-based assays quantify beta-blocker and NSAID
specific effects in effluents of municipal wastewater treatment plants. Water
Research, 115, 74–83.
Berninger J. P., DeMarini D. M., Warren S. H., Simmons J. E., Wilson V. S., Conley J. M.,
Armstrong M. D., Iwanowicz L. R., Kolpin D. W., Kuivila K. M., Reilly T. J., Romanok
K. M., Villeneuve D. L. and Bradley P. M. (2019). Predictive analysis using chemical-
gene interaction networks consistent with observed endocrine activity and mutagenicity
of US streams. Environmental Science & Technology, 53(15), 8611–8620.
Bicchi C., Schiliro T., Pignata C., Fea E., Cordero C., Canale F. and Gilli G. (2009). Analysis
of environmental endocrine disrupting chemicals using the E-SCREEN method and stir
bar sorptive extraction in wastewater treatment plant effluents. Science of the Total
Environment, 407(6), 1842–1851.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 384 Bioanalytical Tools in Water Quality Assessment

Birch H., Kramer N. I. and Mayer P. (2019). Time-resolved freely dissolved concentrations of
semivolatile and hydrophobic test chemicals in in vitro assays. Measuring high losses
and crossover by headspace solid-phase microextraction. Chemical Research in
Toxicology, 32(9), 1780–1790.
Bjorseth A., Carlberg G. E. and Moller M. (1979). Determination of halogenated
organic-compounds and mutagenicity testing of spent bleach liquors. Science of the
Total Environment, 11(2), 197–211.
Blaauboer B. J. (2002). The applicability of in vitro-derived data in hazard identification and
characterisation of chemicals. Environmental Toxicology and Pharmacology, 11(3–4),
213–225.
Blaauboer B. J. (2008). The contribution of in vitro toxicity data in hazard and risk
assessment: Current limitations and future perspectives. Toxicology Letters, 180(2),
81–84.
Blaauboer B. J. (2010). Biokinetic modeling and in vitro-in vivo extrapolations. Journal of
Toxicology and Environmental Health, Part B: Critical Reviews, 13(2–4), 242–252.
Blackwell B. R., Ankley G. T., Bradley P. M., Houck K. A., Makarov S. S., Medvedev A. V.,
Swintek J. and Villeneuve D. L. (2019). Potential toxicity of complex mixtures in
surface waters from a nationwide survey of United States streams: Identifying in vitro
bioactivities and causative chemicals. Environmental Science & Technology, 53(2),
973–983.
Boehler S., Strecker R., Heinrich P., Prochazka E., Northcott G. L., Ataria J. M., Leusch
F. D. L., Braunbeck T. and Tremblay L. A. (2017). Assessment of urban stream
sediment pollutants entering estuaries using chemical analysis and multiple bioassays
to characterise biological activities. Science of the Total Environment, 593–594,
498–507.
Boobis A., Budinsky R., Collie S., Crofton K., Embry M., Felter S., Hertzberg R., Kopp D.,
Mihlan G., Mumtaz M., Price P., Solomon K., Teuschler L., Yang R. and Zaleski R.
(2011). Critical analysis of literature on low-dose synergy for use in screening
chemical mixtures for risk assessment. Critical Reviews in Toxicology, 41(5), 369–383.
Bopp S. K., Barouki R., Brack W., Dalla Costa S., Dorne J., Drakvik P. E., Faust M.,
Karjalainen T. K., Kephalopoulos S., van Klaveren J., Kolossa-Gehring M.,
Kortenkamp A., Lebret E., Lettieri T., Norager S., Ruegg J., Tarazona J. V., Trier X.,
van de Water B., van Gils J. and Bergman A. (2018). Current EU research activities
on combined exposure to multiple chemicals. Environment International, 120, 544–562.
Bopp S. K., Kienzler A., Richarz A. N., van der Linden S. C., Paini A., Parissis N. and Worth
A. P. (2019). Regulatory assessment and risk management of chemical mixtures:
challenges and ways forward. Critical Reviews in Toxicology, 49(2), 174–189.
Borenfreund E. and Puerner J. A. (1985). Toxicity determined in vitro by morphological
alterations and neutral red absorption. Toxicology Letters, 24(2–3), 119–124.
Boxall A. B. A., Sinclair C. J., Fenner K., Kolpin D. and Maud S. J. (2004). When synthetic
chemicals degrade in the environment. Environmental Science & Technology, 38(19),
368A–375A.
Brack W., Altenburger R., Ensenbach U., Moder M., Segner H. and Schuurmann G. (1999).
Bioassay-directed identification of organic toxicants in river sediment in the industrial
region of Bitterfeld (Germany) – A contribution to hazard assessment. Archives of
Environmental Contamination and Toxicology, 37(2), 164–174.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 385

Brack W., Schirmer K., Erdinger L. and Hollert H. (2005). Effect-directed analysis of
mutagens and ethoxyresorufin-O-deethylase inducers in aquatic sediments.
Environmental Toxicology and Chemistry, 24(10), 2445–2458.
Brack W., Schmitt-Jansen M., Machala M., Brix R., Barcelo D., Schymanski E., Streck G.
and Schulze T. (2008). How to confirm identified toxicants in effect-directed analysis.
Analytical and Bioanalytical Chemistry, 390(8), 1959–1973.
Brack W., Ait-Aissa S., Burgess R. M., Busch W., Creusot N., Di Paolo C., Escher B. I.,
Hewitt L. M., Hilscherova K., Hollender J., Hollert H., Jonker W., Kool J., Lamoree
M., Muschket M., Neumann S., Rostkowski P., Ruttkies C., Schollee J., Schymanski
E. L., Schulze T., Seiler T.-B., Tindall A. J., Umbuzeiro G. D. A., Vrana B. and
Krauss M. (2016). Effect-directed analysis supporting monitoring of aquatic
environments – An in-depth overview. Science of the Total Environment, 544,
1073–1118.
Brack W., Ait Aissa S., Backhaus T., Dulio V., Escher B. I., Faust M., Hilscherova K.,
Hollender J., Hollert H., Muller C., Munthe J., Posthuma L., Seiler T. B., Slobodnik
J., Teodorovic I., Tindall A. J., Umbuzeiro G. D., Zhang X. W. and Altenburger R.
(2019). Effect-based methods are key. The European Collaborative Project
SOLUTIONS recommends integrating effect-based methods for diagnosis and
monitoring of water quality. Environmental Sciences Europe, 31, 10.
Bradley P. M., Journey C. A., Romanok K. M., Barber L. B., Buxton H. T., Foreman W. T.,
Furlong E. T., Glassmeyer S. T., Hladik M. L., Iwanowicz L. R., Jones D. K., Kolpin
D. W., Kuivila K. M., Loftin K. A., Mills M. A., Meyer M. T., Orlando J. L., Reilly
T. J., Smalling K. L. and Villeneuve D. L. (2017). Expanded target-chemical analysis
reveals extensive mixed organic-contaminant exposure in US streams. Environmental
Science & Technology, 51(9), 4792–4802.
Brand W., de Jongh C. M., van der Linden S. C., Mennes W., Puijker L. M., van Leeuwen
C. J., van Wezel A. P., Schriks M. and Heringa M. B. (2013). Trigger values for
investigation of hormonal activity in drinking water and its sources using CALUX
bioassays. Environment International, 55, 109–118.
Braunig J., Tang J. Y. M., Warne M. S. J. and Escher B. I. (2016). Bioanalytical effect-balance
model to determine the bioavailability of organic contaminants in sediments affected by
black and natural carbon. Chemosphere, 156, 181–190.
Brettschneider D. J., Misovic A., Schulte-Oehlmann U., Oetken M. and Oehlmann J. (2019).
Detection of chemically induced ecotoxicological effects in rivers of the Nidda
catchment (Hessen, Germany) and development of an ecotoxicological, Water
Framework Directive-compliant assessment system. Environmental Sciences Europe,
31, 7.
Brian J. V., Harris C. A., Scholze M., Backhaus T., Booy P., Lamoree M., Pojana G., Jonkers
N., Runnalls T., Bonfa A., Marcomini A. and Sumpter J. P. (2005). Accurate prediction
of the response of freshwater fish to a mixture of estrogenic chemicals. Environmental
Health Perspectives, 113(6), 721–728.
Brion F., De Gussem V., Buchinger S., Hollert H., Carere M., Porcher J. M., Piccini B., Feray
C., Dulio V., Konemann S., Simon E., Werner I., Kase R. and Ait-Aissa S. (2019).
Monitoring estrogenic activities of waste and surface waters using a novel in vivo
zebrafish embryonic (EASZY) assay: Comparison with in vitro cell-based assays and
determination of effect-based trigger values. Environment International, 130, 104896.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 386 Bioanalytical Tools in Water Quality Assessment

Brunner A. M., Dingemans M. M. L., Baken K. A. and van Wezel A. P. (2019). Prioritizing
anthropogenic chemicals in drinking water and sources through combined use of
mass spectrometry and ToxCast toxicity data. Journal of Hazardous Materials, 364,
332–338.
Bunnell J. E., Tatu C. A., Lerch H. E., Orem W. H. and Pavlovic N. (2007). Evaluating
nephrotoxicity of high-molecular-weight organic compounds in drinking water from
lignite aquifers. Journal of Toxicology and Environmental Health-Part A-Current
Issues, 70(24), 2089–2091.
Burke M. D. and Mayer R. T. (1974). Ethoxyresorufin – direct fluorimetric assay of a
microsomal o-dealkylation which is preferentially inducible by 3-methylcholanthrene.
Drug Metabolism and Disposition, 2(6), 583–588.
Burton G. A., Jr., Pitt R. and Clark S. (2000). The role of traditional and novel toxicity test
methods in assessing stormwater and sediment contamination. Critical Reviews in
Environmental Science and Technology, 30(4), 413–447.
Buschini A., Carboni P., Frigerio S., Furlini M., Marabini L., Monarca S., Poli P., Radice S.
and Rossi C. (2004). Genotoxicity and cytotoxicity assessment in lake drinking water
produced in a treatment plant. Mutagenesis, 19(5), 341–347.
Cahill P. A., Knight A. W., Billinton N., Barker M. G., Walsh L., Keenan P. O., Williams
C. V., Tweats D. J. and Walmsley R. M. (2004). The GreenScreen((R)) genotoxicity
assay: a screening validation programme. Mutagenesis, 19(2), 105–119.
Campana O. and Wlodkowic D. (2018). Ecotoxicology goes on a chip: embracing
miniaturized bioanalysis in aquatic risk assessment. Environmental Science &
Technology, 52(3), 932–946.
Campora C. E., Hokama Y., Tamaru C. S., Anderson B. and Vincent D. (2010). Evaluating
the risk of ciguatera fish poisoning from reef fish grown at marine aquaculture facilities
in Hawai’i. Journal of the World Aquaculture Society, 41(1), 61–70.
Cantwell R. E. and Hofmann R. (2011). Ultraviolet absorption properties of suspended
particulate matter in untreated surface waters. Water Research, 45(3), 1322–1328.
Cao N., Yang M., Zhang Y., Hu J., Ike M., Hirotsuji J., Matsui H., Inoue D. and Sei K. (2009).
Evaluation of wastewater reclamation technologies based on in vitro and in vivo
bioassays. Science of the Total Environment, 407(5), 1588–1597.
Cargouet M., Perdiz D., Mouatassim-Souali A., Tamisier-Karolak S. and Levi Y. (2004).
Assessment of river contamination by estrogenic compounds in Paris area (France).
Science of the Total Environment, 324(1–3), 55–66.
Carlberg G. E., Gjos N., Moller M., Gustavsen K. O., Tveten G. and Renberg L. (1980).
Chemical characterization and mutagenicity testing of chlorinated trihydroxybenzenes
identified in spent bleach liquors from a sulfite plant. Science of the Total
Environment, 15(1), 3–15.
Carson R. (1962). Silent Spring. Houghton Mifflin, New York, NY, USA.
Castillo M., Alonso M. C., Riu J., Reinke M., Kloter G., Dizer H., Fischer B., Hansen P. D.
and Barcelo D. (2001). Identification of cytotoxic compounds in European wastewaters
during a field experiment. Analytica Chimica Acta, 426(2), 265–277.
CDC (2019). Fourth National Report on Human Exposure to Environmental Chemicals.
Updated Tables, January 2019. Centers for Disease Control and Prevention, US
Department of Health and Human Services, Washington DC, USA. www.cdc.gov/
exposurereport/index.html (Accessed 18 February 2021).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 387

CDER/FDA (2001). Guidance for Industry: Bioanalytical Method Validation. Center for
Drug Evaluation (CDER) and US Food and Drug Administration (FDA). www.fda.
gov/files/drugs/published/Bioanalytical-Method-Validation-Guidance-for-Industry.pdf
(Accessed 18 February 2021).
Cedergreen N. (2014). Quantifying synergy: A systematic review of mixture toxicity studies
within environmental toxicology. Plos One, 9(5), e96580.
Celander M. C., Goldstone J. V., Denslow N. D., Iguchi T., Kille P., Meyerhoff R. D., Smith
B. A., Hutchinson T. H. and Wheeler J. R. (2011). Species extrapolation for the 21st
century. Environmental Toxicology and Chemistry, 30(1), 52–63.
Cencic A. and Langerholc T. (2010). Functional cell models of the gut and their applications
in food microbiology: A review. International Journal of Food Microbiology, 141,
S4–S14.
Ceretti E., Zani C., Zerbini I., Guzzella L., Scaglia M., Berna V., Donato F., Monarca S. and
Feretti D. (2010). Comparative assessment of genotoxicity of mineral water packed in
polyethylene terephthalate (PET) and glass bottles. Water Research, 44(5), 1462–1470.
Cetojevic-Simin D., Svircev Z. and Baltic V. V. (2009). In vitro cytotoxicity of cyanobacteria
from water ecosystems of Serbia. Journal of the Balkan Union of Oncology, 14(2),
289–294.
Chang J. C., Taylor P. B. and Leach F. R. (1981). Use of the Microtox® assay system for
environmental samples. Bulletin of Environmental Contamination and Toxicology,
26(2), 150–156.
Cheh A. M., Skochdopole J., Koski P. and Cole L. (1980). Non-volatile mutagens in drinking
water – production by chlorination and destruction by sulfite. Science, 207(4426),
90–92.
Chen Q. Y., Jia A., Snyder S. A., Gong Z. Y. and Lam S. H. (2016). Glucocorticoid activity
detected by in vivo zebrafish assay and in vitro glucocorticoid receptor bioassay at
environmental relevant concentrations. Chemosphere, 144, 1162–1169.
Chen S., Li D. C., Wu X. N., Chen L. P., Zhang B., Tan Y. F., Yu D. K., Niu Y., Duan H. W.,
Li Q., Chen R., Aschner M., Zheng Y. X. and Chen W. (2020). Application of cell-based
biological bioassays for health risk assessment of PM2.5 exposure in three megacities,
China. Environment International, 139, 105703.
Chinathamby K., Allinson M., Shiraishi F., Lopata A. L., Nugegoda D., Pettigrove V. and
Allinson G. (2013). Screening for potential effects of endocrine-disrupting chemicals
in peri-urban creeks and rivers in Melbourne, Australia using mosquitofish and
recombinant receptor-reporter gene assays. Environmental Science and Pollution
Research, 20(3), 1831–1841.
Chou P. H., Lee C. H., Ko F. C., Lin Y. J., Kawanishi M., Yagi T. and Li I. C. (2015).
Detection of hormone-like and genotoxic activities in indoor dust from Taiwan using
a battery of in vitro bioassays. Aerosol and Air Quality Research, 15(4), 1412–1421.
Clemons J. H., Allan L. M., Marvin C. H., Wu Z., McCarry B. E., Bryant D. W. and
Zacharewski T. R. (1998). Evidence of estrogen- and TCDD-like activities in crude
and fractionated extracts of PM10 air particulate material using in vitro gene
expression assays. Environmental Science & Technology, 32(12), 1853–1860.
Cloutier P. L., Fortin F., Groleau P. E., Brousseau P., Fournier M. and Desrosiers M. (2017).
QuEChERS extraction for multi-residue analysis of PCBs, PAHs, PBDEs and PCDD/Fs
in biological samples. Talanta, 165, 332–338.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 388 Bioanalytical Tools in Water Quality Assessment

Coecke S., Goldberg A. M., Allen S., Buzanska L., Calamandrei G., Crofton K., Hareng L.,
Hartung T., Knaut H., Honegger P., Jacobs M., Lein P., Li A., Mundy W., Owen D.,
Schneider S., Silbergeld E., Reum T., Trnovec T., Monnet-Tschudi F. and Bal-Price
A. (2007). Workgroup report: Incorporating in vitro alternative methods for
developmental neurotoxicity into international hazard and risk assessment strategies.
Environmental Health Perspectives, 115(6), 924–931.
Collins F., Gray G. N. and Bucher J. R. (2008). Transforming environmental health
protection. Science, 319, 906–907.
Combes R., Balls M., Curren R., Fischbach M., Fusenig N., Kirkland D., Lasne C., Landolph
J., LeBoeuf R., Marquardt H., McCormick J., Muller L., Rivedal E., Sabbioni E., Tanaka
N., Vasseur P. and Yamasaki H. (1999). Cell transformation assays as predictors of
human carcinogenicity – The report and recommendations of ECVAM Workshop 39.
ALTA-Alternatives to Laboratory Animals, 27(5), 745–767.
Conley J. M., Evans N., Cardon M. C., Rosenblum L., Iwanowicz L. R., Hartig P. C., Schenck
K. M., Bradley P. M. and Wilson V. S. (2017a). Occurrence and in vitro bioactivity of
estrogen, androgen, and glucocorticoid compounds in a nationwide screen of United
States stream waters. Environmental Science & Technology, 51(9), 4781–4791.
Conley J. M., Evans N., Mash H., Rosenblum L., Schenck K., Glassmeyer S., Furlong E. T.,
Kolpin D. W. and Wilson V. S. (2017b). Comparison of in vitro estrogenic activity and
estrogen concentrations in source and treated waters from 25 US drinking water
treatment plants. Science of the Total Environment, 579, 1610–1617.
Conolly R. B., Ankley G. T., Cheng W. Y., Mayo M. L., Miller D. H., Perkins E. J.,
Villeneuve D. L. and Watanabe K. H. (2017). Quantitative adverse outcome pathways
and their application to predictive toxicology. Environmental Science & Technology,
51(8), 4661–4672.
Conroy O., Sáez A. E., Quanrud D., Ela W. and Arnold R. G. (2007). Changes in
estrogen/anti-estrogen activities in ponded secondary effluent. Science of the Total
Environment, 382(2–3), 311–323.
Cortes C. and Marcos R. (2018). Genotoxicity of disinfection byproducts and disinfected
waters: A review of recent literature. Mutation Research-Genetic Toxicology and
Environmental Mutagenesis, 831, 1–12.
Corvi R. and Madia F. (2017). In vitro genotoxicity testing – Can the performance be
enhanced? Food and Chemical Toxicology, 106, 600–608.
Costa L. G. (1998). Neurotoxicity testing: A discussion of in vitro alternatives. Environmental
Health Perspectives, 106, 505–510.
Costa L. D., Ottoni M. H. F., dos Santos M. G., Meireles A. B., de Almeida V. G., Pereira
W. D., de Avelar-Freitas B. A. and Brito-Melo G. E. A. (2017). Dimethyl sulfoxide
(DMSO) decreases cell proliferation and TNF-alpha, IFN-gamma, and IL-2 cytokines
production in cultures of peripheral blood lymphocytes. Molecules, 22(11), 1789.
Creusot N., Kinani S., Balaguer P., Tapie N., LeMenach K., Maillot-Marechal E., Porcher
J. M., Budzinski H. and Ait-Aissa S. (2010). Evaluation of an hPXR reporter gene
assay for the detection of aquatic emerging pollutants: screening of chemicals and
application to water samples. Analytical and Bioanalytical Chemistry, 396(2), 569–583.
Creusot N., Ait-Aissa S., Tapie N., Pardon P., Brion F., Sanchez W., Thybaud E., Porcher
J. M. and Budzinski H. (2014). Identification of synthetic steroids in river water
downstream from pharmaceutical manufacture discharges based on a bioanalytical

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 389

approach and passive sampling. Environmental Science & Technology, 48(7),

3649–3657.
Dagnino S., Gomez E., Picot B., Cavailles V., Casellas C., Balaguer P. and Fenet H. (2010).
Estrogenic and AhR activities in dissolved phase and suspended solids from wastewater
treatment plants. Science of the Total Environment, 408(12), 2608–2615.
Daniels K. D., VanDervort D., Wu S. M., Leusch F. D. L., van de Merwe J. P., Jia A. and
Snyder S. A. (2018). Downstream trends of in vitro bioassay responses in a
wastewater effluent-dominated river. Chemosphere, 212, 182–192.
Davey R. A. and Grossmann M. (2016). Androgen receptor structure, function and biology:
From bench to bedside. The Clinical Biochemist Reviews, 37(1), 3–15.
de Baat M. L., Bas D. A., van Beusekom S. A. M., Droge S. T. J., van der Meer F., de Vries
M., Verdonschot P. F. M. and Kraak M. H. S. (2018). Nationwide screening of surface
water toxicity to algae. Science of the Total Environment, 645, 780–787.
de Baat M. L., Kraak M. H. S., Van der Oost R., De Voogt P. and Verdonschot P. F. M.
(2019a). Effect-based nationwide surface water quality assessment to identify
ecotoxicological risks. Water Research, 159, 434–443.
de Baat M. L., Wieringa N., Droge S. T. J., van Hall B. G., van der Meer F. and Kraak M. H. S.
(2019b). Smarter sediment screening: effect-based quality assessment, chemical
profiling, and risk identification. Environmental Science & Technology, 53(24),
14479–14488.
de Baat M. L., Van der Oost R., Van der Lee G. H., Wieringa N., Hamers T., Verdonschot
P. F. M., De Voogt P. and Kraak M. H. S. (2020). Advancements in effect-based
surface water quality assessment. Water Research, 183, 116017.
Delgado L. F., Faucet-Marquis V., Pfohl-Leszkowicz A., Dorandeu C., Marion B., Schetrite
S. and Albasi C. (2011). Cytotoxicity micropollutant removal in a crossflow membrane
bioreactor. Bioresource Technology, 102(6), 4395–4401.
DeMarini D. M. (2020). A review on the 40th anniversary of the first regulation of drinking
water disinfection by-products. Environmental and Molecular Mutagenesis, 61(6),
588–601.
Deneer J. W., Sinnige T., Seinen W. and Hermens J. (1988). The joint acute toxicity to
Daphnia magna of industrial organic chemicals at low concentration. Aquatic
Toxicology, 12, 33–38.
Denison M. S., Soshilov A. A., He G. C., DeGroot D. E. and Zhao B. (2011). Exactly the same
but different: promiscuity and diversity in the molecular mechanisms of action of the aryl
hydrocarbon (dioxin) receptor. Toxicological Sciences, 124(1), 1–22.
Denison M. S., Mehinto A., Olivieri A., Plumlee M., Schlenk D., Thompson S. and Waggoner
C. (2020). Bioanalytical tools for detection and quantification of estrogenic and
dioxin-like chemicals in water recycling and reuse. Guidance Document for
Developing a Standard Operating Procedure. Prepared for WateReuse California and
National Water Research Institute, www.cwea.org/news/nwri-releases-sop-guidance-
document-for-bioassays/ (Accessed 7 June 2021).
Desforges J. P., Levin M., Jasperse L., De Guise S., Eulaers I., Letcher R. J., Acquarone M.,
Nordoy E., Folkow L. P., Jensen T. H., Grondahl C., Bertelsen M. F., Leger J. S.,
Almunia J., Sonne C. and Dietz R. (2017). Effects of polar bear and killer whale
derived contaminant cocktails on marine mammal immunity. Environmental Science
& Technology, 51(19), 11431–11439.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 390 Bioanalytical Tools in Water Quality Assessment

Di Paolo C., Ottermanns R., Keiter S., Ait-Aissa S., Bluhm K., Brack W., Breitholtz M.,
Buchinger S., Carere M., Chalon C., Cousin X., Dulio V., Escher B. I., Hamers T.,
Hilscherova K., Jarque S., Jonas A., Maillot-Marechal E., Marneffe Y., Nguyen
M. T., Pandard P., Schifferli A., Schulze T., Seidensticker S., Seiler T. B., Tang J.,
van der Oost R., Vermeirssen E., Zounkova R., Zwart N. and Hollert H. (2016).
Bioassay battery interlaboratory investigation of emerging contaminants in spiked
water extracts – Towards the implementation of bioanalytical monitoring tools in
water quality assessment and monitoring. Water Research, 104, 473–484.
Dingemans M. M. L., Baken K. A., van der Oost R., Schriks M. and van Wezel A. P. (2019).
Risk-based approach in the revised European Union drinking water legislation:
Opportunities for bioanalytical tools. Integrated Environmental Assessment and
Management, 15(1), 126–134.
Dizer H., Wittekindt E., Fischer B. and Hansen P. D. (2002). The cytotoxic and genotoxic
potential of surface water and wastewater effluents as determined by
bioluminescence, umu-assays and selected biomarkers. Chemosphere, 46(2), 225–233.
Dogruer G., Weijs L., Tang J. Y.-M., Hollert H., Kock M., Bell I., Hof C. A. M. and Gaus C.
(2018). Effect-based approach for screening of chemical mixtures in whole blood of
green turtles from the Great Barrier Reef. Science of the Total Environment, 612,
321–329.
Drakvik E., Altenburger R., Aoki Y., Backhaus T., Bahadori T., Barouki R., Brack W.,
Cronin M. T. D., Demeneix B., Bennekou S. H., van Klaveren J., Kneuer C.,
Kolossa-Gehring M., Lebret E., Posthuma L., Reiber L., Rider C., Ruegg J., Testa G.,
van der Burg B., van der Voet H., Warhurst A. M., van de Water B., Yamazaki K.,
Oberg M. and Bergman A. (2020). Statement on advancing the assessment of
chemical mixtures and their risks for human health and the environment. Environment
International, 134, 105267.
Dyer S. D., White-Hull C. E. and Shephard B. K. (2000). Assessments of chemical mixtures
via toxicity reference values overpredict hazard to Ohio fish communities.
Environmental Science & Technology, 34(12), 2518–2524.
Eggen R. and Segner H. (2003). The potential of mechanism-based bioanalytical tools in
ecotoxicological exposure and effect assessment. Analytical and Bioanalytical
Chemistry, 377(3), 386–396.
Elad T. and Belkin S. (2012). Whole-cell biochips for online water monitoring.
Bioengineered, 3(2), 124–128.
Ellman G. L., Courtney K. D., Andres V. and Featherstone R. M. (1961). A new and rapid
colorimetric determination of acetylcholinesterase activity. Biochemical
Pharmacology, 7(2), 88–95.
Embry M. R., Belanger S. E., Braunbeck T. A., Galay-Burgos M., Halder M., Hinton D. E.,
Léonard M. A., Lillicrap A., Norberg-King T. and Whale G. (2010). The fish embryo
toxicity test as an animal alternative method in hazard and risk assessment and
scientific research. Aquatic Toxicology, 97(2), 79–87.
EMEA (2011). Guideline on validation of bioanalytical methods. EMEA/CHMP/EWP/
192217/2009 Rev. 1 Corr. 2** Committee for Medicinal Products for Human
Use (CHMP), European Medicines Agency, London, UK. www.ema.europa.eu/
en/documents/scientific-guideline/guideline-bioanalytical-method-validation_en.pdf
(Accessed 18 February 2021).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 391

Emmanuel E., Perrodin Y., Keck G., Blanchard J. M. and Vermande P. (2005).
Ecotoxicological risk assessment of hospital wastewater: A proposed framework for
raw effluents discharging into urban sewer network. Journal of Hazardous Materials,
117(1), 1–11.
enHealth (2012). Environmental Health Risk Assessment. Guidelines for Assessing Human
Health Risks from Environmental Hazards. Department of Health and Aging and
enHealth Council, Canberra, Australia. www1.health.gov.au/internet/main/
publishing.nsf/Content/health-pubhlth-publicat-environ.htm (Accessed 3 November
2020).
Environment and Climate Change Canada (2020). Evaluating Existing Substances. www.ec.
gc.ca/ese-ees/ (Accessed 15 October 2020).
EP&EC (1998). Council Directive 98/83/EC of 3 November 1998 on the quality of water
intended for human consumption. European Parliament and European Council,
Official Journal of the European Communities, L330/32.
EP&EC (2000). Directive 2000/60/EC of the European Parliament and the Council of 23
October 2000 establishing a framework for Community action in the field of water
policy (short: Water Framework Directive). European Parliament and European
Council, Official Journal of the European Communities, L327/1.
EP&EC (2006a). Regulation (EC) No 1907/2006 of the European Parliament and of the
Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation
and Restriction of Chemicals (REACH), establishing a European Chemicals Agency,
amending Directive 1999/45/EC and repealing Council Regulation (EEC) No
793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive
76/769/EEC and Commission Directives 91/155/EEC, 93/67/EEC, 93/105/EC and
2000/21/EC. European Parliament and European Council, Official Journal of the
European Communities, 396/1.
EP&EC (2006b). Regulation (EC) no 1907/2006 REACH, Criteria for the identification of
persistent, bioaccumulative and toxic substances, and very persistent and very
bioaccumulative substances, Annex XIII. European Parliament and European
Council, Official Journal of the European Communities, L69/7.
EP&EC (2010). Directive 2010/63/EU of 22 September 2010 on the protection of animals
used for scientific purposes. European Parliament and European Council, Official
Journal of the European Communities, L 276/233.
EP&EC (2013). Directive 2013/39/EU of the European Parliament and of the Council of of
12 August 2013 amending Directives 2000/60/EC and 2008/105/EC as regards
priority substances in the field of water policy. European Parliament and European
Council, Official Journal of the European Communities, L226/1.
EP&EC (2020). Directive of the European Parliament and of the Council of 16 December
2020 on the quality of water intended for human consumption (recast). European
Parliament and European Council, Official Journal of the European Communities,
L435/1.
Epler J. L., Larimer F. W., Rao T. K., Nix C. E. and Ho T. (1978). Energy-related pollutants in
the environment – use of short-term tests for mutagenicity in the isolation and
identification of biohazards. Environmental Health Perspectives, 27, 11–20.
Erdmann S. E., Dietz R., Sonne C., Bechshoft T. O., Vorkamp K., Letcher R. J., Long
M. H. and Bonefeld-Jorgensen E. C. (2013). Xenoestrogenic and dioxin-like

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 392 Bioanalytical Tools in Water Quality Assessment

activity in blood of East Greenland polar bears (Ursus maritimus). Chemosphere, 92

(5), 583–591.
Ermler S., Scholze M. and Kortenkamp A. (2014). Genotoxic mixtures and dissimilar action:
concepts for prediction and assessment. Archives of Toxicology, 88(3), 799–814.
Escher B. I. and Fenner K. (2011). Recent advances in the environmental risk assessment of
transformation products. Environmental Science & Technology, 45(9), 3835–3847.
Escher B. I. and Hermens J. L. M. (2002). Modes of action in ecotoxicology: their role in body
burdens, species sensitivity, QSARs, and mixture effects. Environmental Science &
Technology, 36, 4201–4217.
Escher B. I. and Neale P. A. (2021). Effect-based trigger values for mixtures of chemicals in
surface water detected with in vitro bioassays. Environmental Toxicology and
Chemistry, 40(2), 487–499.
Escher B. I., Eggen R. I. L., Schreiber U., Schreiber Z., Vye E., Wisner B. and Schwarzenbach
R. P. (2002). Baseline toxicity (narcosis) of organic chemicals determined by in vitro
membrane potential measurements in energy-transducing membranes. Environmental
Science & Technology, 36(9), 1971–1979.
Escher B. I., Bramaz N., Eggen R. I. L. and Richter M. (2005). In vitro assessment of modes of
toxic action of pharmaceuticals in aquatic life. Environmental Science & Technology,
39, 3090–3100.
Escher B. I., Bramaz N., Mueller J. F., Quayle P., Rutishauser S. and Vermeirssen E. L. M.
(2008a). Toxic equivalent concentrations (TEQs) for baseline toxicity and specific
modes of action as a tool to improve interpretation of ecotoxicity testing of
environmental samples. Journal of Environmental Monitoring, 10(5), 612–621.
Escher B. I., Bramaz N., Quayle P., Rutishauser S. and Vermeirssen E. L. M. (2008b).
Monitoring of the ecotoxicological hazard potential by polar organic micropollutants
in sewage treatment plants and surface waters using a mode-of-action based test
battery. Journal of Environmental Monitoring, 10(5), 622–631.
Escher B. I., Bramaz N. and Ort C. (2009). JEM Spotlight: Monitoring the treatment
efficiency of a full scale ozonation on a sewage treatment plant with a mode-of-action
based test battery. Journal of Environmental Monitoring, 11(10), 1836–1846.
Escher B. I., Lawrence M., Macova M., Mueller J. F., Poussade Y., Robillot C., Roux A. and
Gernjak W. (2011). Evaluation of contaminant removal of reverse osmosis and advanced
oxidation in full-scale operation by combining passive sampling with chemical analysis
and bioanalytical tools. Environmental Science & Technology, 45, 5387–5394.
Escher B. I., Dutt M., Maylin E., Tang J. Y. M., Toze S., Wolf C. R. and Lang M. (2012).
Water quality assessment using the AREc32 reporter gene assay indicative of the
oxidative stress response pathway. Journal of Environmental Monitoring, 14(11),
2877–2885.
Escher B. I., van Daele C., Dutt M., Tang J. Y. M. and Altenburger R. (2013). Most oxidative
stress response in water samples comes from unknown chemicals: The need for
effect-based water quality trigger values. Environmental Science & Technology, 47
(13), 7002–7011.
Escher B. I., Allinson M., Altenburger R., Bain P., Balaguer P., Busch W., Crago J., Humpage
A., Denslow N. D., Dopp E., Hilscherova K., Kumar A., Grimaldi M., Jayasinghe B. S.,
Jarosova B., Jia A., Makarov S., Maruya K. A., Medvedev A., Mehinto A. C., Mendez
J. E., Poulsen A., Prochazka E., Richard J., Schifferli A., Schlenk D., Scholz S., Shiraishi

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 393

F., Snyder S., Su G., Tang J., van der Burg B., van der Linden S., Werner I., Westerheide
S. D., Wong C. K. C., Yang M., Yeung B., Zhang X. and Leusch F. D. L.
(2014). Benchmarking organic micropollutants in wastewater, recycled water and
drinking water with in vitro bioassays. Environmental Science & Technology, 48,
1940–1956.
Escher B. I., Neale P. A. and Leusch F. D. L. (2015). Effect-based trigger values for in vitro
bioassays: Reading across from existing water quality guideline values. Water Research,
81, 137–148.
Escher B. I., Hackermüller J., Polte T., Scholz S., Aigner A., Altenburger R., Böhme A., Bopp
S. K., Brack W., Busch W., Chadeau-Hyam M., Covaci A., Eisenträger A., Galligan J. J.,
Garcia-Reyero N., Hartung T., Hein M., Herberth G., Jahnke A., Kleinjans J., Klüver N.,
Krauss M., Lamoree M., Lehmann I., Luckenbach T., Miller G. W., Müller A., Phillips
D. H., Reemtsma T., Rolle-Kampczyk U., Schüürmann G., Schwikowski B., Tan Y.-M.,
Trump S., Walter-Rohde S. and Wambaugh J. F. (2017). From the exposome to
mechanistic understanding of chemical-induced adverse effects. Environment
International, 99, 97–106.
Escher B., Neale P. A. and Villeneuve D. (2018a). The advantages of linear
concentration-response curves for in vitro bioassays with environmental samples.
Environmental Toxicology and Chemistry, 37(9), 2273–2280.
Escher B. I., Ait-Aissa S., Behnisch P. A., Brack W., Brion F., Brouwer A., Buchinger S.,
Crawford S. E., Du Pasquier D., Hamers T., Hettwer K., Hilscherova K., Hollert H.,
Kase R., Kienle C., Tindall A. J., Tuerk J., van der Oost R., Vermeirssen E. and
Neale P. A. (2018b). Effect-based trigger values for in vitro and in vivo bioassays
performed on surface water extracts supporting the environmental quality standards
(EQS) of the European Water Framework Directive. Science of the Total
Environment, 628–629, 748–765.
Escher B. I., Glauch L., Konig M., Mayer P. and Schlichting R. (2019). Baseline toxicity and
volatility cutoff in reporter gene assays used for high-throughput screening. Chemical
Research in Toxicology, 32(8), 1646–1655.
Escher B. I., Abagyan R., Embry M., Klüver N., Redman A. D., Zarfl C. and Parkerton T. F.
(2020a). Recommendations for improving methods and models for aquatic hazard
assessment of ionizable organic chemicals. Environmental Toxicology and Chemistry,
39(1), 269–286.
Escher B. I., Braun G. and Zarfl C. (2020b). Exploring the concepts of concentration addition
and independent action using a linear low-effect mixture model. Environmental
Toxicology and Chemistry, 39(12), 2552–2559.
Escher B. I., Henneberger L., Schlichting R. and Fischer F. C. (2020c). Cytotoxicity burst or
baseline toxicity? Differentiating specific from nonspecific effects in reporter gene
assays. Environmental Health Perspectives, 128(7), 077007.
Escher B. I., Stapleton H. M. and Schymanski E. L. (2020d). Tracking complex mixtures of
chemicals in our changing environment. Science, 367, 388–392.
European Chemicals Agency (2011). Guidance on information requirements and chemical
safety assessment. Part A: Introduction to the guidance document. echa.europa.
eu/guidance-documents/guidance-on-information-requirements-and-chemical-safety-
assessment (Accessed 15 October 2020).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 394 Bioanalytical Tools in Water Quality Assessment

European Chemicals Agency (2012). Guidance for the implementation of REACH, Guidance
on information requirements and chemical safety assessment. Chapter R.19: Uncertainty
analysis. echa.europa.eu/documents/10162/13632/information_requirements_r19_en.
pdf (Accessed 23 October 2020).
European Chemicals Agency (2017). Read-Across Assessment Framework (RAAF) –
considerations on multi-constituent substances and UVCBs. echa.europa.eu/
documents/10162/13630/raaf_uvcb_report_en.pdf/3f79684d-07a5-e439-16c3-d2c8da
96a316 (Accessed 20 December 2020).
European Commission (2009). Common Implementation Strategy for the WFD, Guidance
Document No. 19, Guidance on Surface Water Chemical Monitoring. https://ec.
europa.eu/environment/water/water-framework/facts_figures/guidance_docs_en.htm
(Accessed 16 February 2021).
European Commission (2011). Common implementation strategy for the water framework
directive (2000/60/EC). Guidance document No. 27. Technical guidance for deriving
environmental quality standards, European Communities. Version 2018. https://ec.
europa.eu/environment/water/water-framework/facts_figures/guidance_docs_en.htm
(Accessed 16 February 2021).
European Commission (2020). Communication from the Commission to the European
Parliament, the Council, the European Economic and Social Committee and the
Committee of the Regions (COM (2020) 667 final). Chemicals Strategy for
Sustainability towards a Toxic-free Environment, European Communities. eur-lex.
europa.eu (Accessed 10 April 2021).
European Environment Agency (2013). Late lessons from early warnings: science,
precaution, innovation. www.eea.europa.eu/publications/late-lessons-2 (Accessed 23
October 2020).
Fang Y.-X., Ying G.-G., Zhao J.-L., Chen F., Liu S., Zhang L.-J. and Yang B. (2012).
Assessment of hormonal activities and genotoxicity of industrial effluents using in
vitro bioassays combined with chemical analysis. Environmental Toxicology and
Chemistry, 31(6), 1273–1282.
Farmen E., Harman C., Hylland K. and Tollefsen K. E. (2010). Produced water extracts from
North Sea oil production platforms result in cellular oxidative stress in a rainbow trout in
vitro bioassay. Marine Pollution Bulletin, 60(7), 1092–1098.
Farré M., Klöter G., Petrovic M., Alonso M. C., de Alda M. J. L. and Barceló D. (2002).
Identification of toxic compounds in wastewater treatment plants during a field
experiment. Analytica Chimica Acta 456(1), 19–30.
Farre M. J., Day S., Neale P. A., Stalter D., Tang J. Y. M. and Escher B. I. (2013).
Bioanalytical and chemical assessment of the disinfection by-product formation
potential: Role of organic matter. Water Research, 47(14), 5409–5421.
Fatima R. A. and Ahmad M. (2006). Genotoxicity of industrial wastewaters obtained from two
different pollution sources in northern India: A comparison of three bioassays. Mutation
Research – Genetic Toxicology and Environmental Mutagenesis, 609(1), 81–91.
Faust M., Altenburger R. and Grimme L. H. (2001). Predicting the joint algal toxicity of
multicomponent s-triazine mixtures at low-effect concentrations of individual
toxicants. Aquatic Toxicology, 56(1), 13–32.
Faust M., Altenburger R., Backhaus T., Blanck H., Boedeker W., Gramatica P., Hamer V.,
Scholze M., Vighi M. and Grimme L. H. (2003). Joint algal toxicity of 16

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 395

dissimilarly acting chemicals is predictable by the concept of independent action.

Aquatic Toxicology, 63(1), 43–63.
Fay K. A., Villeneuve D. L., Swintek J., Edwards S. W., Nelms M. D., Blackwell B. R. and
Ankley G. T. (2018). Differentiating pathway-specific from nonspecific effects in
high-throughput toxicity data: A foundation for prioritizing adverse outcome pathway
development. Toxicological Sciences, 163(2), 500–515.
Fedorenkova A., Vonk J. A., Lenders H. J. R., Ouborg N. J., Breure A. M. and Hendriks A. J.
(2010). Ecotoxicogenomics: Bridging the gap between genes and populations.
Environmental Science & Technology, 44(11), 4328–4333.
Fenner K., Kooijman C., Scheringer M. and Hungerbuhler K. (2002). Including
transformation products into the risk assessment for chemicals: The case of
nonylphenol ethoxylate usage in Switzerland. Environmental Science & Technology,
36(6), 1147–1154.
Ferk P. and Daris B. (2018). The influence of dimethyl sulfoxide (DMSO) on metabolic
activity and morphology of melanoma cell line WM-266-4. Cellular and Molecular
Biology, 64(11), 41–43.
Filer D. L., Kothiya P., Setzer R. W., Judson R. S. and Martin M. T. (2016). tcpl: the ToxCast
pipeline for high-throughput screening data. Bioinformatics, 33(4), 618–620.
Finlayson K. A., Leusch F. D. L. and van de Merwe J. P. (2019a). Cytotoxicity of organic and
inorganic compounds to primary cell cultures established from internal tissues of
Chelonia mydas. Science of the Total Environment, 664, 958–967.
Finlayson K. A., Leusch F. D. L. and van de Merwe J. P. (2019b). Primary green turtle
(Chelonia mydas) skin fibroblasts as an in vitro model for assessing genotoxicity and
oxidative stress. Aquatic Toxicology, 207, 13–18.
Finlayson K. A., Madden Hof C. A. and van de Merwe J. P. (2020). Development and
application of species-specific cell-based bioassays to assess toxicity in green sea
turtles. Science of the Total Environment, 747, 142095.
Fischer F., Henneberger L., König M., Bittermann K., Linden L., Goss K.-U. and Escher B. I.
(2017). Modeling exposure in the Tox21 in vitro bioassays. Chemical Research in
Toxicology, 30, 1197–1208.
Fischer F., Abele C., Droge S. T. J., Henneberger L., König M., Schlichting R., Scholz S. and
Escher B. I. (2018a). Cellular uptake kinetics of neutral and charged chemicals in in vitro
assays measured by fluorescence microscopy. Chemical Research in Toxicology, 31,
646–657.
Fischer F. C., Cirpka O., Goss K. U., Henneberger L. and Escher B. I. (2018b). Application of
experimental polystyrene partition constants and diffusion coefficients to predict the
sorption of organic chemicals to well plates in in vitro bioassays. Environmental
Science & Technology, 52, 13511–13522.
Fischer F. C., Henneberger L., Schlichting R. and Escher B. I. (2019). How to improve the
dosing of chemicals in high-throughput in vitro mammalian cell assays. Chemical
Research in Toxicology, 32(8), 1462–1468.
Fischer F. C., Abele C., Henneberger L., Klüver N., König M., Mühlenbrink M., Schlichting
R. and Escher B. I. (2020). Characterizing cellular metabolism in high-throughput in
vitro reporter gene assays. Chemical Research in Toxicology, 33(7), 1770–1779.
Flohe L. and Gunzler W. A. (1984). Assays of glutathione-peroxidase. Methods in
Enzymology, 105, 114–121.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 396 Bioanalytical Tools in Water Quality Assessment

Forsberg N. D., Wilson G. R. and Anderson K. A. (2011). Determination of parent and

substituted polycyclic aromatic hydrocarbons in high-fat salmon using a modified
QuEChERS extraction, dispersive SPE and GC-MS. Journal of Agricultural and
Food Chemistry, 59(15), 8108–8116.
Fox S. I. (1991). Perspectives on Human Biology. William C. Brown Publishers, Dubuque,
IA, USA.
French V. A., King S. C., Kumar A., Northcott G., McGuinness K. and Parry D. (2015).
Characterisation of microcontaminants in Darwin Harbour, a tropical estuary of
northern Australia undergoing rapid development. Science of the Total Environment,
536, 639–647.
Furlong E. T., Werner S. L., Anderson B. D. and Cahill J. D. (2008). Determination of
human-health pharmaceuticals in filtered water by chemically modified styrene–
divinylbenzene resin-based solid-phase extraction and high-performance liquid
chromatography/mass spectrometry. In: US Geological Survey Techniques and
Methods (ed.), USGS, pubs.usgs.gov/tm/tm5b5/pdf/tm-5-b5_508.pdf (Accessed 18
February 2021).
Furuichi T., Kannan K., Glesy J. P. and Masunaga S. (2004). Contribution of known
endocrine disrupting substances to the estrogenic activity in Tama River water
samples from Japan using instrumental analysis and in vitro reporter gene assay.
Water Research, 38(20), 4491–4501.
Gagné F. and Blaise C. (1998). Estrogenic properties of municipal and industrial wastewaters
evaluated with a rapid and sensitive chemoluminescent in situ hybridization assay
(CISH) in rainbow trout hepatocytes. Aquatic Toxicology, 44(1–2), 83–91.
Gao X. Z., Huang C., Rao K. F., Xu Y. P., Huang Q. H., Wang F., Ma M. and Wang Z. J.
(2018). Occurrences, sources, and transport of hydrophobic organic contaminants in
the waters of Fildes Peninsula, Antarctica. Environmental Pollution, 241, 950–958.
Garcia-Reyero N., Grau E., Castillo M., De Alda M. J. L., Barcelo D. and Pina B. (2001).
Monitoring of endocrine disruptors in surface waters by the yeast recombinant assay.
Environmental Toxicology and Chemistry, 20(6), 1152–1158.
Gartiser S., Hafner C., Oeking S. and Paschke A. (2009). Results of a “Whole Effluent
Assessment” study from different industrial sectors in Germany according to
OSPAR’s WEA strategy. Journal of Environmental Monitoring, 11(2), 359–369.
Gartiser S., Hafner C., Hercher C., Kronenberger-Schafer K. and Paschke A. (2010). Whole
effluent assessment of industrial wastewater for determination of bat compliance. Part I:
paper manufacturing industry. Environmental Science and Pollution Research, 17(4),
856–865.
Gehrmann L., Bielak H., Behr M., Itzel F., Lyko S., Simon A., Kunze G., Dopp E., Wagner
M. and Tuerk J. (2018). (Anti-)estrogenic and (anti-)androgenic effects in wastewater
during advanced treatment: Comparison of three in vitro bioassays. Environmental
Science and Pollution Research, 25, 4094–4104.
Genthe B., Steyn M., Aneck-Hahn N., Van Zijl C. and De Jager C. (2009). The feasibility
of a health risk assessment framework to derive guidelines for oestrogen activity in
treated drinking water. Water Research Commission, Pretoria, South Africa WRC.
1749, p. 02.
Gibb S. (2008). Toxicity testing in the 21st century: A vision and a strategy. Reproductive
Toxicology, 25(1), 136–138.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 397

Gijsbers L., Man H. Y., Kloet S. K., de Haan L. H., Keijer J., Rietjens I. M., van der Burg B.
and Aarts J. M. (2011). Stable reporter cell lines for peroxisome proliferator-activated
receptor gamma (PPARgamma)-mediated modulation of gene expression. Analytical
Biochemistry, 414(1), 77–83.
Gilmore T. D. (2006). Introduction to NF-κB: players, pathways, perspectives. Oncogene, 25
(51), 6680–6684.
Glassmeyer S. T., Furlong E. T., Kolpin D. W., Batt A. L., Benson R., Boone J. S., Conerly
O., Donohue M. J., King D. N., Kostich M. S., Mash H. E., Pfaller S. L., Schenck K. M.,
Simmons J. E., Varughese E. A., Vesper S. J., Villegas E. N. and Wilson V. S.
(2017). Nationwide reconnaissance of contaminants of emerging concern in source
and treated drinking waters of the United States. Science of the Total Environment,
581, 909–922.
Glauch L. and Escher B. I. (2020). The combined algae test for the evaluation of mixture
toxicity in environmental samples. Environmental Toxicology and Chemistry, 39(12),
2496–2508.
Gohlke J. M. and Portier C. J. (2007). The forest for the trees: a systems approach to human
health research. Environmental Health Perspectives, 115, 1261–1263.
González A., Kroll K. J., Silva-Sanchez C., Carriquiriborde P., Fernandino J. I., Denslow
N. D. and Somoza G. M. (2020). Steroid hormones and estrogenic activity in the
wastewater outfall and receiving waters of the Chascomús chained shallow lakes
system (Argentina). Science of the Total Environment, 743, 140401.
Grimaldi M., Boulahtouf A., Delfosse V., Thouennon E., Bourguet W. and Balaguer P.
(2015). Reporter cell lines for the characterization of the interactions between human
nuclear receptors and endocrine disruptors. Frontiers in Endocrinology, 6, 62.
Groothuis F. A., Heringa M. B., Nicol B., Hermens J. L. M., Blaauboer B. J. and Kramer N. I.
(2015). Dose metric considerations in in vitro assays to improve quantitative in vitro–in
vivo dose extrapolations. Toxicology, 332, 30–40.
Grothe D. R., Dickson K. L. and Reed-Judkins D. K. (eds.) (1995). Whole Effluent Toxicity
Testing. An Evaluation of the Methods and Prediction of Receiving System Impacts.
SETAC Special Publication Series, SETAC Press, SETAC Press, Pensacola, FL, USA.
Gruener N. (1978). Mutagenicity of ozonated, recycled water. Bulletin of Environmental
Contamination and Toxicology, 20(4), 522–526.
Gruiz K., Fekete-Kertesz I., Kunglne-Nagy Z., Hajdu C., Feigl V., Vaszita E. and Molnar M.
(2016). Direct toxicity assessment – Methods, evaluation, interpretation. Science of the
Total Environment, 563, 803–812.
Grung M., Lichtenthaler R., Ahel M., Tollefsen K.-E., Langford K. and Thomas K. V. (2007).
Effects-directed analysis of organic toxicants in wastewater effluent from Zagreb,
Croatia. Chemosphere, 67(1), 108–120.
Guillouzo A., Corlu A., Aninat C., Glaise D., Morel F. and Guguen-Guillouzo C. (2007). The
human hepatoma HepaRG cells: A highly differentiated model for studies of liver
metabolism and toxicity of xenobiotics. Chemico-Biological Interactions, 168(1),
66–73.
Gustavsson L., Hollert H., Jonsson S., van Bavel B. and Engwall M. (2007). Reed beds
receiving industrial sludge containing nitroaromatic compounds – Effects of outgoing
water and bed material extracts in the umu-C genotoxicity assay, DR-CALUX assay

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 398 Bioanalytical Tools in Water Quality Assessment

and on early life stage development in Zebrafish (Danio rerio). Environmental Science
and Pollution Research, 14(3), 202–211.
Gustavsson M., Kreuger J., Bundschuh M. and Backhaus T. (2017). Pesticide mixtures in the
Swedish streams: Environmental risks, contributions of individual compounds and
consequences of single-substance oriented risk mitigation. Science of the Total
Environment, 598, 973–983.
Guzzella L., Di Caterino F., Monarca S., Zani C., Feretti D., Zerbini I., Nardi G., Buschini A.,
Poli P. and Rossi C. (2006). Detection of mutagens in water-distribution systems after
disinfection. Mutation Research – Genetic Toxicology and Environmental
Mutagenesis, 608(1), 72–81.
Guzzella L., Monarca S., Zani C., Feretti D., Zerbini I., Buschini A., Poli P., Rossi C. and
Richardson S. D. (2004). In vitro potential genotoxic effects of surface drinking water
treated with chlorine and alternative disinfectants. Mutation Research – Genetic
Toxicology and Environmental Mutagenesis, 564(2), 179–193.
Haber L. T., Dourson M. L., Allen B. C., Hertzberg R. C., Parker A., Vincent M. J., Maier A.
and Boobis A. R. (2018). Benchmark dose (BMD) modeling: current practice, issues,
and challenges. Critical Reviews in Toxicology, 48(5), 387–415.
Hamers T., Kamstra J. H., van Gils J., Kotte M. C. and van Hattum A. G. M. (2015). The
influence of extreme river discharge conditions on the quality of suspended
particulate matter in Rivers Meuse and Rhine (The Netherlands). Environmental
Research, 143, 241–255.
Hamers T., Legradi J., Zwart N., Smedes F., de Weert J., van den Brandhof E. J., van
de Meent D. and de Zwart D. (2018). Time-Integrative Passive sampling combined
with TOxicity Profiling (TIPTOP): an effect-based strategy for cost-effective
chemical water quality assessment. Environmental Toxicology and Pharmacology,
64, 48–59.
Hamers T., Molin K. R. J., Koeman J. H. and Murk A. J. (2000). A small-volume bioassay for
quantification of the esterase inhibiting potency of mixtures of organophosphate and
carbamate insecticides in rainwater: Development and optimization. Toxicological
Sciences, 58(1), 60–67.
Hamilton L. A., Tremblay L. A., Northcott G. L., Boake M. and Lim R. P. (2016). The impact
of variations of influent loading on the efficacy of an advanced tertiary sewage treatment
plant to remove endocrine disrupting chemicals. Science of the Total Environment, 560,
101–109.
Han Y. N., Li N., Oda Y., Ma M., Rao K. F., Wang Z. J., Jin W., Hong G., Li Z. G. and Luo Y.
(2016). Evaluation of genotoxic effects of surface waters using a battery of bioassays
indicating different mode of action. Ecotoxicology and Environmental Safety, 133,
448–456.
Harder A., Escher B. I., Landini P., Tobler N. B. and Schwarzenbach R. P. (2003).
Evaluation of bioanalytical tools for toxicity assessment and mode of toxic action
classification of reactive chemicals. Environmental Science & Technology, 37(21),
4962–4970.
Harrill J., Shah I., Setzer R. W., Haggard D., Auerbach S., Judson R. and Thomas R. S.
(2019). Considerations for strategic use of high-throughput transcriptomics chemical
screening data in regulatory decisions. Current Opinion in Toxicology, 15, 64–75.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 399

Hartung T. (2010). Lessons learned from alternative methods and their validation for a new
toxicology in the 21st century. Journal of Toxicology and Environmental Health, Part
B: Critical Reviews, 13(2), 277–290.
Hashmi M. A. K., Escher B. I., Krauss M., Teodorovic I. and Brack W. (2018).
Effect-directed analysis (EDA) of Danube River water sample receiving untreated
municipal wastewater from Novi Sad, Serbia. Science of the Total Environment, 624,
1072–1081.
Hashmi M. A. K., Krauss M., Escher B. I., Teodorovic I. and Brack W. (2020).
Effect-directed analysis of progestogens and glucocorticoids at trace concentrations in
river water. Environmental Toxicology and Chemistry, 39, 189–199.
He G. C., Tsutsumi T., Zhao B., Baston D. S., Zhao J., Heath-Pagliuso S. and Denison M. S.
(2011). Third-generation Ah receptor-responsive luciferase reporter plasmids:
Amplification of dioxin-responsive elements dramatically increases CALUX bioassay
sensitivity and responsiveness. Toxicological Sciences, 123(2), 511–522.
Health Canada (2020). Guidelines for Canadian Drinking Water Quality – Summary Table.
Water and Air Quality Bureau, Healthy Environments and Consumer Safety Branch,
Health Canada, Ottawa, Ontario, Canada.
Hebert A., Feliers C., Lecarpentier C., Neale P. A., Schlichting R., Thibert S. and Escher B. I.
(2018). Bioanalytical assessment of adaptive stress responses in drinking water as a tool
to differentiate between micropollutants and disinfection by-products. Water Research,
132, 340–349.
Heinrich P., Diehl U., Forster F. and Braunbeck T. (2014). Improving the in vitro
ethoxyresorufin-O-deethylase (EROD) assay with RTL-W1 by metabolic
normalization and use of beta-naphthofiavone as the reference substance.
Comparative Biochemistry and Physiology C-Toxicology & Pharmacology, 164, 27–34.
Henneberg A., Bender K., Blaha L., Giebner S., Kuch B., Kohler H. R., Maier D., Oehlmann
J., Richter D., Scheurer M., Schulte-Oehlmann U., Sieratowicz A., Ziebart S. and
Triebskorn R. (2014). Are in vitro methods for the detection of endocrine potentials
in the aquatic environment predictive for in vivo effects? Outcomes of the projects
SchussenAktiv and SchussenAktiv plus in the Lake Constance area, Germany. Plos
One, 9(6), e98307.
Henneberger L., Muhlenbrink M., Konig M., Schlichting R., Fischer F. C. and Escher B. I.
(2019). Quantification of freely dissolved effect concentrations in in vitro cell-based
bioassays. Archives of Toxicology, 93(8), 2295–2305.
Heringa M. B. and Hermens J. L. (2003). Measurement of free concentration using
negligible-depletion-solid phase microextraction (nd-SPME). Trends in Analytical
Chemistry, 22(1), 575–587.
Hermens J., Könemann H., Leeuwangh P. and Musch A. (1985). Quantitative
structure-activity relationships in aquatic toxicity studies of chemicals and complex
mixtures of chemicals. Environmental Toxicology and Chemistry, 4, 273–279.
Hirota M., Motoyama A., Suzuki M., Yanagi M., Kitagaki M., Kouzuki H., Hagino S., Itagaki
H., Sasa H., Kagatani S. and Aiba S. (2010). Changes of cell-surface thiols and
intracellular signaling in human monocytic cell line THP-1 treated with
diphenylcyclopropenone. Journal of Toxicological Sciences, 35(6), 871–879.
Hirsch C. and Schildknecht S. (2019). In vitro research reproducibility: keeping up high
standards. Frontiers in Pharmacology, 10, 1484.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 400 Bioanalytical Tools in Water Quality Assessment

Hissin P. J. and Hilf R. (1976). A fluorometric method for determination of oxidized and
reduced glutathione in tissues. Analytical Biochemistry, 74(1), 214–226.
Hollender J., Schymanski E. L., Singer H. P. and Ferguson P. L. (2017). Nontarget screening
with high resolution mass spectrometry in the environment: ready to go? Environmental
Science & Technology, 51(20), 11505–11512.
Holmes C. M., Brown C. D., Hamer M., Jones R., Maltby L., Posthuma L., Silberhorn E.,
Teeter J. S., Warne M. S. and Weltje L. (2018). Prospective aquatic risk assessment
for chemical mixtures in agricultural landscapes. Environmental Toxicology and
Chemistry, 37(3), 674–689.
Honda G. S., Pearce R. G., Pham L. L., Setzer R. W., Wetmore B. A., Sipes N. S., Gilbert J.,
Franz B., Thomas R. S. and Wambaugh J. F. (2019). Using the concordance of in vitro
and in vivo data to evaluate extrapolation assumptions. Plos One, 14(5), e0217564.
Hong S., Khim J. S., Naile J. E., Park J., Kwon B.-O., Wang T., Lu Y., Shim W. J., Jones P. D.
and Giesy J. R. (2012). AhR-mediated potency of sediments and soils in estuarine and
coastal areas of the Yellow Sea region: A comparison between Korea and China.
Environmental Pollution, 171, 216–225.
Hong S., Khim J. S., Park J., Kim S., Lee S., Choi K., Kim C. S., Choi S. D., Park J., Ryu J.,
Jones P. D. and Giesy J. P. (2014). Instrumental and bioanalytical measures of
dioxin-like compounds and activities in sediments of the Pohang Area, Korea.
Science of the Total Environment, 470–471, 1517–1525.
Houtman C. J., Legler J. and Thomas K. (2011). Effect-Directed Analysis of Endocrine
Disruptors in Aquatic Ecosystems. In: Effect-Directed Analysis of Complex
Environmental Contamination, W. Brack (ed.), Springer, Heidelberg, Germany,
pp. 237–265.
Houtman C. J., ten Broek R. and Brouwer A. (2018). Steroid hormonal bioactivities, culprit
natural and synthetic hormones and other emerging contaminants in waste water
measured using bioassays and UPLC-tQ-MS. Science of the Total Environment, 630,
1492–1501.
Houtman C. J., ten Broek R., van Oorschot Y., Kloes D., van der Oost R., Rosielle M. and
Lamoree M. H. (2020). High resolution effect-directed analysis of steroid hormone
(ant)agonists in surface and wastewater quality monitoring. Environmental Toxicology
and Pharmacology, 80, 103460.
Hrudey S. E. and Fawell J. (2015). 40 years on: what do we know about drinking water
disinfection by-products (DBPs) and human health? Water Science and
Technology-Water Supply, 15(4), 667–674.
Hu J., Wang W., Zhu Z., Chang H., Pan F. and Lin B. (2007). Quantitative structure-activity
relationship model for prediction of genotoxic potential for quinolone antibacterials.
Environmental Science & Technology, 41(13), 4806–4812.
Huang G. Y., Liu Y. S., Chen X. W., Liang Y. Q., Liu S. S., Yang Y. Y., Hu L. X., Shi W. J.,
Tian F., Zhao J. L., Chen J. and Ying G. G. (2016). Feminization and masculinization of
western mosquitofish (Gambusia affinis) observed in rivers impacted by municipal
wastewaters. Scientific Reports, 6, 20884.
Huchthausen J., Muhlenbrink M., Konig M., Escher B. I. and Henneberger L. (2020).
Experimental exposure assessment of ionizable organic chemicals in in vitro
cell-based bioassays. Chemical Research in Toxicology, 33(7), 1845–1854.
Huggett D. B., Foran C. M., Brooks B. W., Weston J., Peterson B., Marsh K. E., La Point
T. W. and Schlenk D. (2003). Comparison of in vitro and in vivo bioassays for

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 401

estrogenicity in effluent from North American municipal wastewater facilities.

Toxicological Sciences, 72(1), 77–83.
Hurst M. R., Balaam J., Chan-Man Y. L., Thain J. E. and Thomas K. V. (2004).
Determination of dioxin and dioxin-like compounds in sediments from UK estuaries
using a bio-analytical approach: chemical-activated luciferase expression (CALUX)
assay. Marine Pollution Bulletin, 49(7–8), 648–658.
Huuskonen S., Koponen K., Ritola O., Hahn M. and Lindstrom-Seppa P. (1998). Induction of
CYP1A and porphyrin accumulation in fish hepatoma cells (PLHC-1) exposed to
sediment or water from a PCB-contaminated lake (Lake Kernaala, Finland). Marine
Environmental Research, 46(1–5), 379–384.
Ihara M., Hanamoto S., Ihara M. O., Zhang H. and Tanaka H. (2020). Wastewater-derived
antagonistic activities of G protein-coupled receptor-acting pharmaceuticals in river
water. Journal of Applied Toxicology, 40(7), 908–917.
Inoue D., Nakama K., Matsui H., Sei K. and Ike M. (2009). Detection of agonistic activities
against five human nuclear receptors in river environments of Japan using a yeast
two-hybrid assay. Bulletin of Environmental Contamination and Toxicology, 82(4),
399–404.
IPCS (2009). Assessment of combined exposures to multiple chemicals: Report of a
WHO/IPCS international workshop. W. H. O. International Programme on Chemical
Safety, www.who.int/ipcs/methods/harmonization/areas/aggregate/en/index.html
(Accessed 2 January 2021).
Isidori M., Ferrara M., Lavorgna M., Nardelli A. and Parrella A. (2003). In situ monitoring of
urban air in Southern Italy with the tradescantia micronucleus bioassay and
semipermeable membrane devices (SPMDs). Chemosphere, 52(1), 121–126.
Isidori M., Lavorgna M., Palumbo M., Piccioli V. and Parrella A. (2007). Influence of
alkylphenols and trace elements in toxic, genotoxic, and endocrine disruption activity
of wastewater treatment plants. Environmental Toxicology and Chemistry, 26(8),
1686–1694.
ISO6341 (1996). Water quality – Determination of the inhibition of the mobility of Daphnia
magna Straus (Cladocera, Crustacea) – Acute toxicity test. International Organization
for Standardization (ISO), Geneva, Switzerland.
ISO7346-3 (1996). Water Quality – Determination of the Acute Lethal Toxicity of Substances
to a Freshwater Fish [Brachydanio rerio Hamilton-Buchanan (Teleostei, Cyprinidae)] –
Part 3: Flow-through Method. International Organization for Standardization (ISO),
Geneva, Switzerland.
ISO11369 (1997). Water quality – determination of selected plant treatment agents –
method using high-performance liquid chromatography with UV detection after
solid-liquid extraction. International Organization for Standardization (ISO), Geneva,
Switzerland.
ISO11348-3 (1998). Water Quality – Determination of the Inhibitory Effect of Water Samples
on the Light Emission of Vibrio fischeri (Luminescent Bacteria Test). International
Organization for Standardization (ISO), Geneva, Switzerland.
ISO12890 (1999). Water Quality – Determination of Toxicity to Embryos and Larvae of
Freshwater Fish – Semi-static Method. International Organization for Standardization
(ISO), Geneva, Switzerland.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 402 Bioanalytical Tools in Water Quality Assessment

ISO10706 (2000). Water Quality – Determination of Long Term Toxicity of Substances to

Daphnia magna Straus (Cladocera, Crustacea). International Organization for
Standardization (ISO), Geneva, Switzerland.
ISO 13829 (2000). Water quality – Determination of the genotoxicity of water and waste
water using the umu-test. International Organization for Standardization (ISO),
Geneva, Switzerland.
ISO8692 (2004). Water quality – Freshwater Algal Growth Inhibition Test with Unicellular
Green Algae. International Organization for Standardization (ISO), Geneva,
Switzerland.
ISO15088 (2007). Water Quality – Determination of the Acute Toxicity of Waste Water to
Zebrafish Eggs (Danio rerio). International Organization for Standardization (ISO),
Geneva, Switzerland.
ISO 11350 (2012). Water Quality – Determination of the Genotoxicity of Water and Waste
Water: Salmonella/Microsome Fluctuation Test (Ames Fluctuation Test).
International Organization for Standardization (ISO), Geneva, Switzerland.
ISO/DIS19040-1 (2017). Water Quality – Determination of the Estrogenic Potential of Water
and Waste Water – Part 1: Yeast Estrogen Screen (Saccharomyces cerevisiae).
International Organization for Standardization (ISO), Geneva, Switzerland.
ISO/DIS19040-2 (2017). Water Quality – Determination of the Estrogenic Potential of Water
and Waste Water – Part 2: Yeast Estrogen Screen (A-YES, Arxula adeninivorans).
International Organization for Standardization (ISO), Geneva, Switzerland.
ISO/DIS19040-3 (2017). Water Quality – Determination of the Estrogenic Potential of Water
and Waste Water – Part 3: In Vitro Human Cell-based Reporter Gene Assay.
International Organization for Standardization (ISO), Geneva, Switzerland.
Jahnke A., Mayer P., Schaefer S., Witt G., Haase N. and Escher B. I. (2016). Strategies for
transferring mixtures of organic contaminants from aquatic environments into
bioassays. Environmental Science & Technology, 50(11), 5424–5431.
Jahnke A., Sobek A., Bergmann M., Bräunig J., Landmann M., Schäfer S. and Escher B. I.
(2018). Effect-based characterization of mixtures of environmental pollutants in
diverse sediments. Environmental Science: Processes & Impacts, 20, 1667.
Jakimska A., Huerta B., Barganska Z., Kot-Wasik A., Rodriguez-Mozaz S. and Barcelo D.
(2013). Development of a liquid chromatography-tandem mass spectrometry
procedure for determination of endocrine disrupting compounds in fish from
Mediterranean rivers. Journal of Chromatography A, 1306, 44–58.
Jalova V., Jarosova B., Blaha L., Giesy J. P., Ocelka T., Grabic R., Jurcikova J., Vrana B. and
Hilscherova K. (2013). Estrogen-, androgen- and aryl hydrocarbon receptor mediated
activities in passive and composite samples from municipal waste and surface waters.
Environment International, 59, 372–383.
Jarošová B., Blaha L., Vrana B., Randak T., Grabic R., Giesy J. P. and Hilscherova K. (2012).
Changes in concentrations of hydrophilic organic contaminants and of
endocrine-disrupting potential downstream of small communities located adjacent to
headwaters. Environment International, 45, 22–31.
Jarošová B., Blaha L., Giesy J. P. and Hilscherova K. (2014a). What level of estrogenic
activity determined by in vitro assays in municipal wastewaters can be considered as
safe? Environment International, 64, 98–109.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 403

Jarošová B., Erseková A., Hilscherová K., Loos R., Gawlik B. M., Giesy J. P. and Bláha L.
(2014b). Europe-wide survey of estrogenicity in wastewater treatment plant effluents:
the need for the effect-based monitoring. Environmental Science and Pollution
Research, 21(18), 10970–10982.
Javurek J., Sychrova E., Smutna M., Bittner M., Kohoutek J., Adamovsky O., Novakova K.,
Smetanova S. and Hilscherova K. (2015). Retinoid compounds associated with water
blooms dominated by Microcystis species. Harmful Algae, 47, 116–125.
Jellett J. F., Marks L. J., Stewart J. E., Dorey M. L., Watson-Wright W. and Lawrence J. F.
(1992). Paralytic shellfish poison (saxitoxin family) bioassays: Automated endpoint
determination and standardization of the in vitro tissue culture bioassay, and
comparison with the standard mouse bioassay. Toxicon, 30(10), 1143–1156.
Jensen C. and Teng Y. (2020). Is it time to start transitioning from 2D to 3D cell culture?
Frontiers in Molecular Biosciences, 7, 33.
Jeon J., Kretschmann A., Escher B. I. and Hollender J. (2013). Characterization of
acetylcholinesterase inhibition and energy allocation in Daphnia magna exposed to
carbaryl. Ecotoxicology and Environmental Safety, 98, 28–35.
Jeon S., Hong S., Kwon B. O., Park J., Song S. J., Giesy J. P. and Khim J. S. (2017).
Assessment of potential biological activities and distributions of endocrine-
disrupting chemicals in sediments of the west coast of South Korea. Chemosphere,
168, 441–449.
Jia A., Escher B. I., Leusch F. D. L., Tang J. Y. M., Prochazka E., Dong B., Snyder E. M. and
Snyder S. A. (2015). In vitro bioassays to evaluate complex chemical mixtures in
recycled water. Water Research, 80, 1–11.
Jia A., Wu S., Daniels K. D. and Snyder S. A. (2016). Balancing the budget: Accounting for
glucocorticoid bioactivity and fate during water treatment. Environmental Science &
Technology, 50(6), 2870–2880.
Jia Y. L., Hammers-Wirtz M., Crawford S. E., Chen Q. Q., Seiler T. B., Schaffer A. and
Hollert H. (2019). Effect-based and chemical analyses of agonistic and antagonistic
endocrine disruptors in multiple matrices of eutrophic freshwaters. Science of the
Total Environment, 651, 1096–1104.
Jin L., van Mourik L., Gaus C. and Escher B. (2013). Applicability of passive sampling to
bioanalytical screening of bioaccumulative chemicals in marine wildlife.
Environmental Science & Technology, 47(14), 7982–7988.
Jin L., Escher B. I., Limpus C. and Gaus C. (2015a). Coupling passive sampling with in vitro
bioassays and chemical analysis to understand combined effects of bioaccumulative
chemicals in blood of marine turtles. Chemosphere, 138, 292–299.
Jin L., Gaus C. and Escher B. (2015b). Bioanalytical Approaches to Understanding
Toxicological Implications of Mixtures of Persistent Organic Pollutants in Marine
Wildlife. In: Comprehensive Analytical Chemistry. Persistent Organic Pollutants
(POPs): Analytical Techniques, Environmental Fate and Biological Effects, E. Zheng
(ed.), Elsevier, pp. 58–84.
Jin L., Gaus C. and Escher B. I. (2015c). Adaptive stress response pathways induced by
environmental mixtures of bioaccumulative chemicals in Dugongs. Environmental
Science & Technology, 49(11), 6963–6973.
Jobling S. and Tyler C. R. (2003). Endocrine disruption in wild freshwater fish. Pure and
Applied Chemistry, 75(11–12), 2219–2234.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 404 Bioanalytical Tools in Water Quality Assessment

Jobling S., Nolan M., Tyler C. R., Brighty G. and Sumpter J. P. (1998). Widespread sexual
disruption in wild fish. Environmental Science & Technology, 32(17), 2498–2506.
Jonker M. T. O., Burgess R. M., Ghosh U., Gschwend P. M., Hale S. E., Lohmann R., Lydy
M. J., Maruya K. A., Reible D. and Smedes F. (2020). Ex situ determination of freely
dissolved concentrations of hydrophobic organic chemicals in sediments and soils:
Basis for interpreting toxicity and assessing bioavailability, risks and remediation
necessity. Nature Protocols, 15(5), 1800–1828.
Judson R., Richard A., Dix D. J., Houck K., Martin M., Kavlock R., Dellarco V., Henry T.,
Holderman T., Sayre P., Tan S., Carpenter T. and Smith E. (2009). The toxicity data
landscape for environmental chemicals. Environmental Health Perspectives, 117(5),
685–695.
Judson R. S., Kavlock R. J., Setzer R. W., Hubal E. A. C., Martin M. T., Knudsen T. B.,
Houck K. A., Thomas R. S., Wetmore B. A. and Dix D. J. (2011). Estimating
toxicity-related biological pathway altering doses for high-throughput chemical risk
assessment. Chemical Research in Toxicology, 24(4), 451–462.
Judson R., Houck K., Martin M., Richard A. M., Knudsen T. B., Shah I., Little S., Wambaugh
J., Setzer R. W., Kothya P., Phuong J., Filer D., Smith D., Reif D., Rotroff D.,
Kleinstreuer N., Sipes N., Xia M. H., Huang R. L., Crofton K. and Thomas R. S.
(2016). Analysis of the effects of cell stress and cytotoxicity on in vitro assay
activity across a diverse chemical and assay space. Toxicological Sciences, 152(2),
323–339.
Jugan M. L., Oziol L., Bimbot M., Huteau V., Tamisier-Karolak S., Blondeau J. P. and Lévi
Y. (2009). In vitro assessment of thyroid and estrogenic endocrine disruptors in
wastewater treatment plants, rivers and drinking water supplies in the greater Paris
area (France). Science of the Total Environment, 407(11), 3579–3587.
Junghans M., Backhaus T., Faust M., Scholze M. and Grimme L. H. (2006). Application and
validation of approaches for the predictive hazard assessment of realistic pesticide
mixtures. Aquatic Toxicology, 76(2), 93–110.
Kado N. Y., Guirguis G. N., Flessel C. P., Chan R. C., Chang K. I. and Wesolowski J. J.
(1986). Mutagenicity of fine (,2.5 µm) airborne particles: Diurnal variation in
community air determined by a Salmonella micro preincubation (microsuspension)
procedure. Environmental Mutagenesis, 8(1), 53–66.
Kamber M., Fluckiger-Isler S., Engelhardt G., Jaeckh R. and Zeiger E. (2009). Comparison
of the Ames II and traditional Ames test responses with respect to mutagenicity, strain
specificities, need for metabolism and correlation with rodent carcinogenicity.
Mutagenesis, 24(4), 359–366.
Kandie F. J., Krauss M., Beckers L. M., Massei R., Fillinger U., Becker J., Liess M., Torto B.
and Brack W. (2020). Occurrence and risk assessment of organic micropollutants in
freshwater systems within the Lake Victoria South Basin, Kenya. Science of the Total
Environment, 714, 136748.
Kargalioglu Y., McMillan B. J., Minear R. A. and Plewa M. J. (2002). Analysis of the
cytotoxicity and mutagenicity of drinking water disinfection by-products in
Salmonella typhimurium. Teratogenesis, Carcinogenesis, and Mutagenesis, 22(2),
113–128.
Kase R., Javurkova B., Simon E., Swart K., Buchinger S., Konemann S., Escher B. I., Carere
M., Dulio V., Ait-Aissa S., Hollert H., Valsecchi S., Polesello S., Behnisch P., di Paolo

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 405

C., Olbrich D., Sychrova E., Gundlach M., Schlichting R., Leborgne L., Clara M.,
Scheffknecht C., Marneffe Y., Chalon C., Tusil P., Soldan P., von Danwitz B.,
Schwaiger J., Palao A. M., Bersani F., Perceval O., Kienle C., Vermeirssen E.,
Hilscherova K., Reifferscheid G. and Werner I. (2018). Screening and risk
management solutions for steroidal estrogens in surface and wastewater. Trends in
Analytical Chemistry, 102, 343–358.
Keenan P. O., Knight A. W., Billinton N., Cahill P. A., Dalrymple I. M., Hawkyard C. J.,
Stratton-Campbell D. and Walmsley R. M. (2007). Clear and present danger? The use
of a yeast biosensor to monitor changes in the toxicity of industrial effluents
subjected to oxidative colour removal treatments. Journal of Environmental
Monitoring, 9(12), 1394–1401.
Keiter S., Rastall A., Kosmehl T., Wurm K., Erdinger L., Braunbeck T. and Hollert H. (2006).
Ecotoxicological assessment of sediment, suspended matter and water samples in the
upper Danube River – A pilot study in search for the causes for the decline of fish
catches. Environmental Science and Pollution Research, 13(5), 308–319.
Kennedy K., Macova M., Leusch F., Bartkow M. E., Hawker D. W., Zhao B., Denison M. S.
and Mueller J. F. (2009). Assessing indoor air exposures using passive sampling with
bioanalytical methods for estrogenicity and aryl hydrocarbon receptor activity.
Analytical and Bioanalytical Chemistry, 394(5), 1413–1421.
Kerbrat A.-S., Darius H. T., Pauillac S., Chinain M. and Laurent D. (2010). Detection of
ciguatoxin-like and paralysing toxins in Trichodesmium spp. from New Caledonia
lagoon. Marine Pollution Bulletin, 61(7–12), 360–366.
Kern S., Fenner K., Singer H. P., Schwarzenbach R. P. and Hollender J. (2009). Identification
of transformation products of organic contaminants in natural waters by computer-aided
prediction and high-resolution mass spectrometry. Environmental Science &
Technology 43, 7039–7046.
Khedidji S., Croes K., Yassaa N., Ladji R., Denison M. S., Baeyens W. and Elskens M.
(2017). Assessment of dioxin-like activity in PM10 air samples from an industrial
location in Algeria, using the DRE-CALUX bioassay. Environmental Science and
Pollution Research, 24(13), 11868–11877.
Kidd K. A., Blanchfield P. J., Mills K. H., Palace V. P., Evans R. E., Lazorchak J. M. and
Flick R. W. (2007). Collapse of a fish population after exposure to a synthetic
estrogen. Proceedings of the National Academy of Sciences of the United States of
America, 104(21), 8897–8901.
Kim H.-S., Yamada H. and Tsuno H. (2007). The removal of estrogenic activity and control
of brominated by-products during ozonation of secondary effluents. Water Research,
41(7), 1441–1446.
Kisitu J., Bennekou S. H. and Leist M. (2019). Chemical concentrations in cell culture
compartments (C-5) concentration definitions. ALTEX-Alternatives to Animal
Experimentation, 36(1), 154–160.
Klaassen C. D. (2013). Casarett & Doull’s Toxicology: The Basic Science of Poisons, 8th
edn. McGraw-Hill.
Klee N., Gustavsson L., Kosmehl T., Engwall M., Erdinger L., Braunbeck T. and Hollert H.
(2004). Changes in toxicity and genotoxicity of industrial sewage sludge samples
containing nitro- and amino-aromatic compounds following treatment in bioreactors

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 406 Bioanalytical Tools in Water Quality Assessment

with different oxygen regimes. Environmental Science and Pollution Research, 11(5),
313–320.
Klepper O., Bakker J., Traas T. P. and van de Meent D. (1998). Mapping the potentially
affected fraction (PAF) of species as a basis for comparison of ecotoxicological risks
between substances and regions. Journal of Hazardous Materials, 61(1–3), 337–344.
Knight A. W., Little S., Houck K., Dix D., Judson R., Richard A., McCarroll N., Akerman G.,
Yang C. H., Birrell L. and Walmsley R. M. (2009). Evaluation of high-throughput
genotoxicity assays used in profiling the US EPA ToxCast (TM) chemicals.
Regulatory Toxicology and Pharmacology, 55(2), 188–199.
Knudsen T. B., Houck K. A., Sipes N. S., Singh A. V., Judson R. S., Martin M. T., Weissman
A., Kleinstreuer N. C., Mortensen H. M., Reif D. M., Rabinowitz J. R., Setzer R. W.,
Richard A. M., Dix D. J. and Kavlock R. (2011). Activity profiles of 309
ToxCast (TM) chemicals evaluated across 292 biochemical targets. Toxicology, 282
(1–2), 1–15.
Kobayashi M., Li L., Iwamoto N., Nakajima-Takagi Y., Kaneko H., Nakayama Y., Eguchi
M., Wada Y., Kumagai Y. and Yamamoto M. (2009). The antioxidant defense system
Keap1-Nrf2 comprises a multiple sensing mechanism for responding to a wide range
of chemical compounds. Molecular and Cellular Biology, 29(2), 493–502.
Koelmans A. A., Bakir A., Burton G. A. and Janssen C. R. (2016). Microplastic as a vector for
chemicals in the aquatic environment: critical review and model-supported
reinterpretation of empirical studies. Environmental Science & Technology, 50(7),
3315–3326.
Kogure K., Tamplin M. L., Simidu U. and Colwell R. R. (1988). A tissue culture assay for
tetrodotoxin, saxitoxin and related toxins. Toxicon, 26(2), 191–197.
Kolkman A., Schriks M., Brand W., Bauerlein P. S., van der Kooi M. M. E., van Doorn R. H.,
Emke E., Reus A. A., van der Linden S. C., de Voogt P. and Heringa M. B. (2013).
Sample preparation for combined chemical analysis and in vitro bioassay application
in water quality assessment. Environmental Toxicology and Pharmacology, 36(3),
1291–1303.
Könemann S., Kase R., Simon E., Swart K., Buchinger S., Schlüssener M., Hollert H., Escher
B. I., Werner I., Ait-Assa S., Vermeirssen E., Dulio V., Valsecchi S., Polesello S.,
Behnisch P., Javurkova B., Perceval O., di Paolo C., Olbrich D., Sychrova E.,
Schlichting R., Leborgne L., Clara M., Scheffknecht C., Marneffe Y., Chalon C.,
Tusil P., Soldan P., von Danwitz B., Schwaiger J., San Martin Becares M. I., Bersani
F., Hilscherova K., Reifferscheid G., Ternes T. and Carere M. (2018). Effect-based
and chemical analytical methods to monitor estrogens under the European Water
Framework Directive. Trends in Analytical Chemistry, 102, 225–235.
König M., Escher B. I., Neale P. A., Krauss M., Hilscherová K., Novák J., Teodorović I.,
Schulze T., Seidensticker S., Kamal Hashmi M. A., Ahlheim J. and Brack W. (2017).
Impact of untreated wastewater on a major European river evaluated with a
combination of in vitro bioassays and chemical analysis. Environmental Pollution,
220, 1220–1230.
Konsoula R. and Barile F. A. (2005). Correlation of in vitro cytotoxicity with paracellular
permeability in Caco-2 cells. Toxicology In Vitro, 19(5), 675–684.
Körner W., Spengler P., Bolz U., Schuller W., Hanf V. and Metzger J. W. (2001). Substances
with estrogenic activity in effluents of sewage treatment plants in southwestern

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 407

Germany. 2. Biological analysis. Environmental Toxicology and Chemistry, 20(10),

2142–2151.
Kortenkamp A. (2007). Ten years of mixing cocktails: a review of combination effects of
endocrine-disrupting chemicals. Environmental Health Perspectives, 115(Suppl. 1),
98–105.
Kortenkamp A. (2020). Which chemicals should be grouped together for mixture risk
assessments of male reproductive disorders? Molecular and Cellular Endocrinology,
499, 110581.
Kortenkamp A. and Faust M. (2018). Regulate to reduce chemical mixture risk. Science,
361(6399), 224–226.
Kortenkamp A., Faust M., Scholze M. and Backhaus T. (2007). Low-level exposure to
multiple chemicals: Reason for human health concerns? Environmental Health
Perspectives, 115, 106–114.
Kortenkamp A., Backhaus T. and Faust M. (2009). State of the art report on mixture toxicity.
Report, European Commission 070307/2007/485103/ETU/D.1, ec.europa.eu/
environment/chemicals/effects/pdf/report_mixture_toxicity.pdf (Accessed 14
February 2021).
Kortenkamp A., Faust M., Backhaus T., Altenburger R., Scholze M., Muller C., Ermler S.,
Posthuma L. and Brack W. (2019). Mixture risks threaten water quality: the European
Collaborative Project SOLUTIONS recommends changes to the WFD and better
coordination across all pieces of European chemicals legislation to improve protection
from exposure of the aquatic environment to multiple pollutants. Environmental
Sciences Europe, 31(1), 69.
Kowalski L. A. (2001). In vitro carcinogenicity testing: Present and future perspectives in
pharmaceutical development. Current Opinion in Drug Discovery & Development 4,
29–35.
Kramer N. I., Busser F. J. M., Oosterwijk M. T. T., Schirmer K., Escher B. I. and Hermens
J. L. M. (2010). Development of a partition-controlled dosing system for cell assays.
Chemical Research in Toxicology, 23(11), 1806–1814.
Kramer V. J., Etterson M. A., Hecker M., Murphy C. A., Roesijadi G., Spade D. J.,
Spromberg J. A., Wang M. and Ankley G. T. (2011). Adverse outcome pathways and
ecological risk assessment bridging to population-level effects. Environmental
Toxicology and Chemistry, 30(1), 64–76.
Kramer N. I., Krismartina M., Rico-Rico A., Blaauboer B. J. and Hermens J. L. M. (2012).
Quantifying processes determining the free concentration of phenanthrene in basal
cytotoxicity assays. Chemical Research in Toxicology, 25(2), 436–445.
Krauss M., Singer H. and Hollender J. (2010). LC-high resolution MS in environmental
analysis: from target screening to the identification of unknowns. Analytical and
Bioanalytical Chemistry, 397(3), 943–951.
Kretschmann A., Ashauer R., Hitzfeld K., Spaak P., Hollender J. and Escher B. I. (2011).
Mechanistic toxicodynamic model for receptor-mediated toxicity of diazoxon, the
active metabolite of Diazinon, in Daphnia magna. Environmental Science &
Technology, 45(11), 4980–4987.
Kretschmann A., Ashauer R., Hollender J. and Escher B. I. (2012). Toxicokinetic and
toxicodynamic model for diazinon toxicity-mechanistic explanation of differences in

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 408 Bioanalytical Tools in Water Quality Assessment

the sensitivity of Daphnia magna and Gammarus pulex. Environmental Toxicology and
Chemistry, 31(9), 2014–2022.
Krishnamurthi K., Devi S. S., Hengstler J. G., Hermes M., Kumar K., Dutta D., Vannan S. M.,
Subin T. S., Yadav R. R. and Chakrabarti T. (2008). Genotoxicity of sludges,
wastewater and effluents from three different industries. Archives of Toxicology, 82
(12), 965–971.
Kunz P. Y., Kienle C., Carere M., Homazava N. and Kase R. (2015). In vitro bioassays to
screen for endocrine active pharmaceuticals in surface and waste waters. Journal of
Pharmaceutical and Biomedical Analysis, 106, 107–115.
Kunz P. Y., Simon E., Creusot N., Jayasinghe B. S., Kienle C., Maletz S., Schifferli A.,
Schönlau C., Aït-Aïssa S., Denslow N. D., Hollert H., Werner I. and Vermeirssen
E. L. M. (2017). Effect-based tools for monitoring estrogenic mixtures: Evaluation of
five in vitro bioassays. Water Research, 110, 378–388.
Kurelec B., Matijasevic Z., Rijavec M., Alacevic M., Britvic S., Muller W. E. G. and Zahn
R. K. (1979). Induction of benzo(a)pyrene mono-oxygenase in fish and the
Salmonella test as a tool for detecting mutagenic-carcinogenic xenobiotics in the
aquatic environment. Bulletin of Environmental Contamination and Toxicology, 21
(6), 799–807.
Kusk K. O., Kruger T., Long M. H., Taxvig C., Lykkesfeldt A. E., Frederiksen H., Andersson
A. M., Andersen H. R., Hansen K. M. S., Nellemann C. and Bonefeld-Jorgensen E. C.
(2011). Endocrine potency of wastewater: contents of endocrine disrupting chemicals
and effects measured by in vivo and in vitro assays. Environmental Toxicology and
Chemistry, 30(2), 413–426.
Lah B., Zinko B., Narat M. and Marinsek-Logar R. (2005). Monitoring of genotoxicity in
drinking water using in vitro comet assay and Ames test. Food Technology and
Biotechnology, 43(2), 139–146.
Lahnsteiner F. (2008). The sensitivity and reproducibility of the zebrafish (Danio rerio)
embryo test for the screening of waste water quality and for testing the toxicity of
chemicals. ATLA-Alternatives to Laboratory Animals, 36(3), 299–311.
Laingam S., Froscio S. M. and Humpage A. R. (2008). Flow-cytometric analysis of in vitro
micronucleus formation: Comparative studies with WIL2-NS human lymphoblastoid
and L5178Y mouse lymphoma cell lines. Mutation Research – Genetic Toxicology
and Environmental Mutagenesis, 656(1–2), 19–26.
Lam M. M., Engwall M., Denison M. S. and Larsson M. (2018). Methylated polycyclic
aromatic hydrocarbons and/or their metabolites are important contributors to the
overall estrogenic activity of polycyclic aromatic hydrocarbon-contaminated soils.
Environmental Toxicology and Chemistry, 37(2), 385–397.
Landis W. G. and Yu M.-H. (2004). Introduction to Environmental Toxicology: Impacts of
Chemicals Upon Ecological Systems. 3rd edn. CRC Press.
Langevin R., Rasmussen J. B., Sloterduk H. and Blaise C. (1992). Genotoxicity in water and
sediment extracts from the St Lawrence river system, using the SOS chromotest. Water
Research, 26(4), 419–429.
Langhans S. A. (2018). Three-dimensional in vitro cell culture models in drug discovery and
drug repositioning. Frontiers in Pharmacology, 9, 6.
Latif M. and Licek E. (2004). Toxicity assessment of wastewaters, river waters, and sediments
in Austria using cost-effective microbiotests. Environmental Toxicology, 19(4),
302–309.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 409

Lazaro-Cote A., Sadoul B., Jackson L. J. and Vijayan M. M. (2018). Acute stress response of
fathead minnows caged downstream of municipal wastewater treatment plants in the
Bow River, Calgary. Plos One, 13(6), e0198177.
le Maire A., Bourguet W. and Balaguer P. (2010). A structural view of nuclear hormone
receptor: endocrine disruptor interactions. Cellular and Molecular Life Sciences, 67
(8), 1219–1237.
Lee D. G., Roehrdanz P. R., Feraud M., Ervin J., Anumol T., Jia A., Park M., Tamez C.,
Morelius E. W., Gardea-Torresdey J. L., Izbicki J., Means J. C., Snyder S. A. and
Holden P. A. (2015). Wastewater compounds in urban shallow groundwater
wells correspond to exfiltration probabilities of nearby sewers. Water Research, 85,
467–475.
Legler J., Leonards P., Spenkelink A. and Murk A. J. (2003). In vitro biomonitoring in polar
extracts of solid phase matrices reveals the presence of unknown compounds with
estrogenic activity. Ecotoxicology, 12(1–4), 239–249.
Legradi J. B., Di Paolo C., Kraak M. H. S., van der Geest H. G., Schymanski E. L., Williams
A. J., Dingemans M. M. L., Massei R., Brack W., Cousin X., Begout M. L., van der Oost
R., Carion A., Suarez-Ulloa V., Silvestre F., Escher B. I., Engwall M., Nilen G., Keiter
S. H., Pollet D., Waldmann P., Kienle C., Werner I., Haigis A. C., Knapen D.,
Vergauwen L., Spehr M., Schulz W., Busch W., Leuthold D., Scholz S., vom Berg
C. M., Basu N., Murphy C. A., Lampert A., Kuckelkorn J., Grummt T. and Hollert
H. (2018). An ecotoxicological view on neurotoxicity assessment. Environmental
Sciences Europe, 30, 46.
Leusch F. D. L. and Snyder S. A. (2015). Bioanalytical tools: half a century of application for
potable reuse. Environmental Science-Water Research & Technology, 1(5), 606–621.
Leusch F. D. L., Chapman H. F., van den Heuvel M. R., Tan B. L. L., Gooneratne S. R. and
Tremblay L. A. (2006). Bioassay-derived androgenic and estrogenic activity in
municipal sewage in Australia and New Zealand. Ecotoxicology and Environmental
Safety, 65(3), 403–411.
Leusch F. D. L., De Jager C., Levi Y., Lim R., Puijker L., Sacher F., Tremblay L. A., Wilson
V. S. and Chapman H. F. (2010). Comparison of five in vitro bioassays to measure
estrogenic activity in environmental waters. Environmental Science & Technology,
44(10), 3853–3860.
Leusch F. D. L., Khan S. J., Gagnon M. M., Quayle P., Trinh T., Coleman H., Rawson C.,
Chapman H. F., Blair P., Nice H. and Reitsema T. (2014a). Assessment of
wastewater and recycled water quality: A comparison of lines of evidence from in
vitro, in vivo and chemical analyses. Water Research 50, 420–431.
Leusch F. D. L., Khan S. J., Laingam S., Prochazka E., Froscio S., Trinh T., Chapman H. F.
and Humpage A. (2014b). Assessment of the application of bioanalytical tools as
surrogate measure of chemical contaminants in recycled water. Water Research, 49,
300–315.
Leusch F. D. L., Neale P. A., Hebert A., Scheurer M. and Schriks M. C. M. (2017). Analysis
of the sensitivity of in vitro bioassays for androgenic, progestagenic, glucocorticoid,
thyroid and estrogenic activity: Suitability for drinking and environmental waters.
Environment International, 99, 120–130.
Leusch F. D. L., Aneck-Hahn N. H., Cavanagh J.-A. E., Du Pasquier D., Hamers T., Hebert
A., Neale P. A., Scheurer M., Simmons S. O. and Schriks M. (2018a). Comparison of in

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 410 Bioanalytical Tools in Water Quality Assessment

vitro and in vivo bioassays to measure thyroid hormone disrupting activity in water
extracts. Chemosphere, 191(Suppl. C), 868–875.
Leusch F. D. L., Neale P. A., Arnal C., Aneck-Hahn N. H., Balaguer P., Bruchet A., Escher
B. I., Esperanza M., Grimaldi M., Leroy G., Scheurer M., Schlichting R., Schriks M. and
Hebert A. (2018b). Analysis of endocrine activity in drinking water, surface water and
treated wastewater from six countries. Water Research, 139, 10–18.
Lewtas J., Chuang J., Nishioka M. and Petersen B. (1990). Bioassay-directed fractionation of
the organic extract of SRM-1649 urban air particulate matter. International Journal of
Environmental Analytical Chemistry, 39(3), 245–256.
Li J., Ma M. and Wang Z. J. (2008). A two-hybrid yeast assay to quantify the effects of
xenobiotics on thyroid hormone-mediated gene expression. Environmental Toxicology
and Chemistry, 27(1), 159–167.
Li J., Wang Z., Ma M. and Peng X. (2010). Analysis of environmental endocrine disrupting
activities using recombinant yeast assay in wastewater treatment plant effluents. Bulletin
of Environmental Contamination and Toxicology, 84(5), 529–535.
Li N., Ma M., Rao K. and Wang Z. (2011). In vitro thyroid disrupting effects of
organic extracts from WWTPs in Beijing. Journal of Environmental Sciences, 23(4),
671–675.
Li J. Y., Tang J. Y. M., Jin L. and Escher B. I. (2013). Understanding bioavailability and
toxicity of sediment-associated contaminants by combining passive sampling with in
vitro bioassays in an urban river catchment. Environmental Toxicology and
Chemistry, 32(12), 2888–2896.
Li J. Y., Su L., Wei F. H., Yang J. H., Jin L. and Zhang X. W. (2016). Bioavailability-based
assessment of aryl hydrocarbon receptor-mediated activity in Lake Tai Basin from
Eastern China. Science of the Total Environment, 544, 987–994.
Li H. Z., Zhang J. and You J. (2018). Diagnosis of complex mixture toxicity in sediments:
Application of toxicity identification evaluation (TIE) and effect-directed analysis
(EDA). Environmental Pollution, 237, 944–954.
Li H. Z., Yi X. Y., Chen F., Tong Y. J., Mehler W. T. and You J. (2019). Identifying
organic toxicants in sediment using effect-directed analysis: A combination of
bioaccessibility-based extraction and high-throughput midge toxicity testing.
Environmental Science & Technology, 53(2), 996–1003.
Lienert J., Güdel K. and Escher B. I. (2007). Screening method for ecotoxicological hazard
assessment of 42 pharmaceuticals considering human metabolism and excretory
routes. Environmental Science & Technology, 41(12), 4471–4478.
Lindim C., de Zwart D., Cousins I. T., Kutsarova S., Kuhne R. and Schuurmann G. (2019).
Exposure and ecotoxicological risk assessment of mixtures of top prescribed
pharmaceuticals in Swedish freshwaters. Chemosphere, 220, 344–352.
Liu C., Xu Y., Ma M., Huang B., Wu J., Meng Q., Wang Z. and Gearheart R. A. (2014).
Evaluation of endocrine disruption and dioxin-like effects of organic extracts from
sewage sludge in autumn in Beijing, China. Frontiers of Environmental Science &
Engineering, 8(3), 433–440.
Liu Y. Y., Lin Y. S., Yen C. H., Miaw C. L., Chen T. C., Wu M. C. and Hsieh C. Y. (2018).
Identification, contribution, and estrogenic activity of potential EDCs in a river receiving
concentrated livestock effluent in Southern Taiwan. Science of the Total Environment,
636, 464–476.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 411

Liu J. C., Lu G. H., Yang H. H., Dang T. J. and Yan Z. H. (2020). Ecological impact
assessment of 110 micropollutants in the Yarlung Tsangpo River on the Tibetan
Plateau. Journal of Environmental Management, 262, 110291.
Liviac D., Wagner E. D., Mitch W. A., Altonji M. J. and Plewa M. J. (2010). Genotoxicity of
water concentrates from recreational pools after various disinfection methods.
Environmental Science & Technology, 44(9), 3527–3532.
Loos R., Carvalho R., Antonio D. C., Cornero S., Locoro G., Tavazzi S., Paracchini B.,
Ghiani M., Lettieri T., Blaha L., Jarosova B., Voorspoels S., Servaes K., Haglund P.,
Fick J., Lindberg R. H., Schwesig D. and Gawlik B. M. (2013). EU-wide monitoring
survey on emerging polar organic contaminants in wastewater treatment plant
effluents. Water Research, 47(17), 6475–6487.
Louiz I., Kinani S., Gouze M. E., Ben-Attia M., Menif D., Bouchonnet S., Porcher J. M.,
Ben-Hassine O. K. and Ait-Aissa S. (2008). Monitoring of dioxin-like, estrogenic
and anti-androgenic activities in sediments of the Bizerta lagoon (Tunisia) by
means of in vitro cell-based bioassays: Contribution of low concentrations of
polynuclear aromatic hydrocarbons (PAHs). Science of the Total Environment, 402
(2–3), 318–329.
Lu M. Y., Kao W. C., Belkin S. and Cheng J. Y. (2019). A smartphone-based whole-cell array
sensor for detection of antibiotics in milk. Sensors, 19(18), 3882.
Luebcke-von Varel U., Machala M., Ciganek M., Neca J., Pencikova K., Palkova L.,
Vondracek J., Loeffler I., Streck G., Reifferscheid G., Flueckiger-Isler S., Weiss
J. M., Lamoree M. and Brack W. (2011). Polar compounds dominate in vitro effects
of sediment extracts. Environmental Science & Technology, 45(6), 2384–2390.
Lundholt B. K., Scudder K. M. and Pagliaro L. (2003). A simple technique for
reducing edge effect in cell-based assays. Journal of Biomolecular Screening, 8(5),
566–570.
Lundqvist J., Andersson A., Johannisson A., Lavonen E., Mandava G., Kylin H., Bastviken
D. and Oskarsson A. (2019a). Innovative drinking water treatment techniques reduce the
disinfection-induced oxidative stress and genotoxic activity. Water Research, 155,
182–192.
Lundqvist J., Mandava G., Lungu-Mitea S., Lai F. Y. and Ahrens L. (2019b). In vitro
bioanalytical evaluation of removal efficiency for bioactive chemicals in Swedish
wastewater treatment plants. Scientific Reports, 9, 1–9.
Luo J., Ma M., Zha J. and Wang Z. (2009). Characterization of aryl hydrocarbon receptor
agonists in sediments of Wenyu River, Beijing, China. Water Research, 43(9),
2441–2448.
Lv X. M., Xiao S. H., Zhang G., Jiang P. and Tang F. (2016). Occurrence and removal of
phenolic endocrine disrupting chemicals in the water treatment processes. Scientific
Reports, 6, 1–10.
Ma M., Li J. and Wang Z. J. (2005). Assessing the detoxication efficiencies of wastewater
treatment processes using a battery of bioassays/biomarkers. Archives of
Environmental Contamination and Toxicology, 49(4), 480–487.
Macova M., Escher B. I., Reungoat J., Carswell S., Chue K. L., Keller J. and Mueller J. F.
(2010). Monitoring the biological activity of micropollutants during advanced
wastewater treatment with ozonation and activated carbon filtration. Water Research,
44(2), 477–492.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 412 Bioanalytical Tools in Water Quality Assessment

Macova M., Toze S., Hodgers L., Mueller J. F., Bartkow M. E. and Escher B. I. (2011).
Bioanalytical tools for the evaluation of organic micropollutants during sewage
treatment, water recycling and drinking water generation. Water Research, 45(14),
4238–4247.
Maeder V., Escher B. I., Scheringer M. and Hungerbuhler K. (2004). Toxic ratio as an
indicator of the intrinsic toxicity in the assessment of persistent, bioaccumulative, and
toxic chemicals. Environmental Science & Technology, 38(13), 3659–3666.
Maffei F., Carbone F., Forti G. C., Buschini A., Poli P., Rossi C., Marabini L., Radice S.,
Chiesara E. and Hrelia P. (2009). Drinking water quality: An in vitro approach for the
assessment of cytotoxic and genotoxic load in water sampled along distribution
system. Environment International, 35(7), 1053–1061.
Mahjoub O., Leclercq M., Bachelot M., Casellas C., Escande A., Balaguer P., Bahri A.,
Gomez E. and Fenet H. (2009). Estrogen, aryl hydrocarbon and pregnane X receptors
activities in reclaimed water and irrigated soils in Oued Souhil area (Nabeul, Tunisia).
Desalination, 246(1–3), 425–434.
Maier D., Benisek M., Blaha L., Dondero F., Giesy J. P., Kohler H. R., Richter D., Scheurer
M. and Triebskorn R. (2016). Reduction of dioxin-like toxicity in effluents by additional
wastewater treatment and related effects in fish. Ecotoxicology and Environmental
Safety, 132, 47–58.
Malaj E., von der Ohe P. C., Grote M., Kuehne R., Mondy C. P., Usseglio-Polatera P., Brack
W. and Schaefer R. B. (2014). Organic chemicals jeopardize the health of freshwater
ecosystems on the continental scale. Proceedings of the National Academy of
Sciences of the United States of America, 111(26), 9549–9554.
Manger R. L., Leja L. S., Lee S. Y., Hungerford J. M. and Wekell M. M. (1993).
Tetrazolium-based cell bioassay for neurotoxins active on voltage-sensitive
sodium-channels – semiautomated assay for saxitoxins, brevetoxins, and ciguatoxins.
Analytical Biochemistry, 214(1), 190–194.
Manger R. L., Leja L. S., Lee S. Y., Hungerford J. M., Hokama Y., Dickey R. W., Granade
H. R., Lewis R., Yasumoto T. and Wekell M. M. (1995). Detection of sodium-channel
toxins – directed cytotoxicity assays of purified ciguatoxins, brevetoxins, saxitoxins, and
seafood extracts. Journal of AOAC International, 78(2), 521–527.
Manusadzianas L., Balkelyte L., Sadauskas K., Blinova I., Põllumaa L. and Kahru A. (2003).
Ecotoxicological study of Lithuanian and Estonian wastewaters: Selection of the
biotests, and correspondence between toxicity and chemical-based indices. Aquatic
Toxicology, 63(1), 27–41.
Marabini L., Frigerio S., Chiesara E. and Radice S. (2006). Toxicity evaluation of surface
water treated with different disinfectants in HepG2 cells. Water Research, 40(2),
267–272.
Marabini L., Frigerio S., Chiesara E., Maffei F., Cantelli Forti G., Hrelia P., Buschini A.,
Martino A., Poli P., Rossi C. and Radice S. (2007). In vitro cytotoxicity and
genotoxicity of chlorinated drinking waters sampled along the distribution system of
two municipal networks. Mutation Research – Genetic Toxicology and Environmental
Mutagenesis, 634(1–2), 1–13.
Maron D. M. and Ames B. N. (1983). Revised methods for the Salmonella mutagenicity test.
Mutation Research, 113(3–4), 173–215.
Martin M. T., Dix D. J., Judson R. S., Kavlock R. J., Reif D. M., Richard A. M., Rotroff
D. M., Romanov S., Medvedev A., Poltoratskaya N., Gambarian M., Moeser M.,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 413

Makarov S. S. and Houck K. A. (2010). Impact of environmental chemicals on key

transcription regulators and correlation to toxicity end points within EPA’s ToxCast
program. Chemical Research in Toxicology, 23(3), 578–590.
Martinovic-Weigelt D., Mehinto A. C., Ankley G. T., Berninger J. P., Collette T. W., Davis
J. M., Denslow N. D., Durhan E. J., Eid E., Ekman D. R., Jensen K. M., Kahl M. D.,
LaLone C. A., Teng Q. and Villeneuve D. L. (2017). Derivation and evaluation of
putative adverse outcome pathways for the effects of cyclooxygenase inhibitors on
reproductive processes in female fish. Toxicological Sciences, 156(2), 344–361.
Martyniuk C. J., Mehinto A. C. and Denslow N. D. (2020). Organochlorine pesticides:
Agrochemicals with potent endocrine-disrupting properties in fish. Molecular and
Cellular Endocrinology, 507, 110764.
Marvin C. H. and Hewitt L. M. (2007). Analytical methods in bioassay-directed
investigations of mutagenicity of air particulate material. Mutation Research-Reviews
in Mutation Research, 636(1–3), 4–35.
Matsuoka S., Kikuchi M., Kimura S., Kurokawa Y. and Kawai S. (2005). Determination of
estrogenic substances in the water of Muko River using in vitro assays, and the
degradation of natural estrogens by aquatic bacteria. Journal of Health Science, 51(2),
178–184.
Mayer P., Parkerton T. F., Adams R. G., Cargill J. G., Gan J., Gouin T., Gschwend P. M.,
Hawthorne S. B., Helm P., Witt G., You J. and Escher B. I. (2014). Passive sampling
methods for contaminated sediments: Scientific rationale supporting use of freely
dissolved concentrations. Integrated Environmental Assessment and Management, 10
(2), 197–209.
McDonough C. A., Franks D. G., Hahn M. E. and Lohmann R. (2019). Aryl
hydrocarbon receptor-mediated activity of gas-phase ambient air derived from
passive sampling and an in vitro bioassay. Environmental Toxicology and
Chemistry, 38(4), 748–759.
Medlock Kakaley E. K., Blackwell B. R., Cardon M. C., Conley J. M., Evans N., Feifarek
D. J., Furlong E. T., Glassmeyer S. T., Gray L. E., Hartig P. C., Kolpin D. W., Mills
M. A., Rosenblum L., Villeneuve D. L. and Wilson V. S. (2020). De facto water
reuse: Bioassay suite approach delivers depth and breadth in endocrine active
compound detection. Science of the Total Environment, 699, 134297.
Meek M. E., Boobis A. R., Crofton K. M., Heinemeyer G., Van Raaij M. and Vickers C.
(2011). Risk assessment of combined exposure to multiple chemicals: A WHO/IPCS
framework. Regulatory Toxicology and Pharmacology, 60(2), S1–S14.
Meek M. E., Boobis A., Cote I., Dellarco V., Fotakis G., Munn S., Seed J. and Vickers C.
(2014). New developments in the evolution and application of the WHO/IPCS
framework on mode of action/species concordance analysis. Journal of Applied
Toxicology, 34(1), 1–18.
Mehinto A. C., Jia A., Snyder S. A., Jayasinghe B. S., Denslow N. D., Crago J., Schlenk D.,
Menzie C., Westerheide S. D., Leusch F. D. L. and Maruya K. A. (2015). Interlaboratory
comparison of in vitro bioassays for screening of endocrine active chemicals in recycled
water. Water Research, 83, 303–309.
Mehinto A. C., Jayasinghe B. S., Vandervort D. R., Denslow N. D. and Maruya K. A. (2016).
Screening for endocrine activity in water using commercially available in vitro
transactivation bioassays. JOVE-Journal of Visualized Experiments, 118, 54725.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 414 Bioanalytical Tools in Water Quality Assessment

Mehinto A. C., VanDervort D. R., Lao W., He G., Denison M. S., Vliet S. M., Volz D. C.,
Mazor R. D. and Maruya K. A. (2017). High throughput in vitro and in vivo
screening of inland waters of Southern California. Environmental Science-Processes
& Impacts, 19(9), 1142–1149.
Mendonca E., Picado A., Paixao S. M., Silva L., Cunha M. A., Leitao S., Moura I., Cortez C.
and Brito F. (2009). Ecotoxicity tests in the environmental analysis of wastewater
treatment plants: Case study in Portugal. Journal of Hazardous Materials, 163(2–3),
665–670.
Michel B. (2005). After 30 years of study, the bacterial SOS response still surprises us. Plos
Biology, 3(7), 1174–1176.
Miege C., Karolak S., Gabet V., Jugan M. L., Oziol L., Chevreuil M., Levi Y. and Coquery M.
(2009). Evaluation of estrogenic disrupting potency in aquatic environments and urban
wastewaters by combining chemical and biological analysis. TrAc-Trends in Analytical
Chemistry, 28(2), 186–195.
Miloshev G., Mihaylov I. and Anachkova B. (2002). Application of the single cell gel
electrophoresis on yeast cells. Mutation Research – Genetic Toxicology and
Environmental Mutagenesis, 513(1–2), 69–74.
Mnif W., Dagnino S., Escande A., Pillon A., Fenet H., Gomez E., Casellas C., Duchesne
M.-J., Hernandez-Raquet G., Cavaillès V., Balaguer P. and Bartegi A. (2010).
Biological analysis of endocrine-disrupting compounds in Tunisian sewage treatment
plants. Archives of Environmental Contamination and Toxicology, 59(1), 1–12.
Mnif W., Zidi I., Hassine A. I. H., Gomez E., Bartegi A., Roig B. and Balaguer P. (2012).
Monitoring endocrine disrupter compounds in the Tunisian Hamdoun River using in
vitro bioassays. Soil & Sediment Contamination, 21(7), 815–830.
More S. J., Hardy A., Bampidis V., Benford D., Bennekou S. H., Bragard C., Boesten
J., Halldorsson T. I., Hernandez-Jerez A. F., Jeger M. J., Knutsen H. K.,
Koutsoumanis K. P., Naegeli H., Noteborn H., Ockleford C., Ricci A., Rychen
G., Schlatter J. R., Silano V., Nielsen S. S., Schrenk D., Solecki R., Turck D.,
Younes M., Benfenati E., Castle L., Cedergreen N., Laskowski R., Leblanc
J. C., Kortenkamp A., Ragas A., Posthuma L., Svendsen C., Testai E., Dujardin
B., Kass G. E. N., Manini P., Jeddi M. Z., Dorne J., Hogstrand C. and Comm
E. S. (2019). Guidance on harmonised methodologies for human health, animal
health and ecological risk assessment of combined exposure to multiple
chemicals. EFSA Journal, 17(3), 5634.
Moreland D. E. (1980). Mechanism of action of herbicides. Annual Review Plant Physiology,
31, 597–638.
Mosmann T. (1983). Rapid colorimetric assay for cellular growth and survival: Application to
proliferation and cytotoxicity assays. Journal of Immunological Methods, 65(1–2),
55–63.
Moxon T. E., Li H. Q., Lee M. Y., Piechota P., Nicol B., Pickles J., Pendlington R., Sorrell I.
and Baltazar M. T. (2020). Application of physiologically based kinetic (PBK)
modelling in the next generation risk assessment of dermally applied consumer
products. Toxicology In Vitro, 63, 104746.
Mueller M. E., Werneburg M., Glaser C., Schwientek M., Zarfl C., Escher B. I. and Zwiener
C. (2020). Influence of emission sources and tributaries on the spatial and temporal

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 415

patterns of micropollutant mixtures and associated effects in a small river.

Environmental Toxicology and Chemistry, 39(7), 1382–1391.
Mueller M. E., Zwiener C. and Escher B. I. (2021). Storm event-driven occurrence and
transport of dissolved and sorbed organic micropollutants and associated effects in the
Ammer River, Southwestern Germany. Environmental Toxicology and Chemistry,
40(1), 88–99.
Müller M. E., Escher B. I., Schwientek M., Werneburg M., Zarfl C. and Zwiener C. (2018).
Combining in vitro reporter gene bioassays with chemical analysis to assess changes in
the water quality along the Ammer River, Southwestern Germany. Environmental
Science Europe, 30, 20.
Murk A. J., Legler J., van Lipzig M. M. H., Meerman J. H. N., Belfroid A. C., Spenkelink A.,
van der Burg B., Rijs G. B. J. and Vethaak D. (2002). Detection of estrogenic potency in
wastewater and surface water with three in vitro bioassays. Environmental Toxicology
and Chemistry, 21(1), 16–23.
Muz M., Escher B. I. and Jahnke A. (2020). Bioavailable environmental pollutant patterns
in sediments from passive equilibrium sampling. Environmental Science &
Technology, 54(24), 15861–15871.
Nachlas M. M., Margulies S. I., Goldberg J. D. and Seligman A. M. (1960). The
determination of lactic dehydrogenase with a tetrazolium salt. Analytical
Biochemistry, 1(4–5), 317–326.
Nagy S. R., Sanborn J. R., Hammock B. D. and Denison M. S. (2002). Development
of a green fluorescent protein-based cell bioassay for the rapid and inexpensive
detection and characterization of Ah receptor agonists. Toxicological Sciences, 65(2),
200–210.
Narotsky M. G., Klinefelter G. R., Goldman J. M., Best D. S., McDonald A., Strader L. F.,
Suarez J. D., Murr A. S., Thillainadarajah I., Hunter E. S., Richardson S. D., Speth T. F.,
Miltner R. J., Pressman J. G., Teuschler L. K., Rice G. E., Moser V. C., Luebke R. W.
and Simmons J. E. (2013). Comprehensive assessment of a chlorinated drinking water
concentrate in a rat multigenerational reproductive toxicity study. Environmental
Science & Technology, 47(18), 10653–10659.
National Academies of Sciences, Engineering and Medicine (2017). Using 21st Century
Science to Improve Risk-Related Evaluations. The National Academies Press,
Washington, DC. www.nap.edu/catalog/24635/using-21st-century-science-to-
improve-risk-related-evaluations (Accessed 18 February 2021).
Neale P. A. and Escher B. I. (2013). Coextracted dissolved organic carbon has a suppressive
effect on the acetylcholinesterase inhibition assay. Environmental Toxicology and
Chemistry, 32(7), 1526–1534.
Neale P. A. and Escher B. I. (2014). Does co-extracted dissolved organic carbon cause
artefacts in cell-based bioassays? Chemosphere, 108, 281–288.
Neale P. A. and Leusch F. D. L. (2015). Considerations when assessing antagonism in vitro:
Why standardizing the agonist concentration matters. Chemosphere, 135, 20–30.
Neale P. A., Antony A., Bartkow M. E., Farre M. J., Heitz A., Kristiana I., Tang J. Y. M. and
Escher B. I. (2012). Bioanalytical assessment of the formation of disinfection byproducts
in a drinking water treatment plant. Environmental Science & Technology, 46(18),
10317–10325.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 416 Bioanalytical Tools in Water Quality Assessment

Neale P.A., Escher B. I. and Leusch F. D. L. (2015a). Understanding the implications of

dissolved organic carbon when assessing antagonism in vitro: An example with an
estrogen receptor assay. Chemosphere, 135, 341–346.
Neale P. A., Ait-Aissa S., Brack W., Creusot N., Denison M. S., Deutschmann B.,
Hilscherová K., Hollert H., Krauss M., Novák J., Schulze T., Seiler T.-B., Serra H.,
Shao Y. and Escher B. I. (2015b). Linking in vitro effects and detected organic
micropollutants in surface water using mixture-toxicity modeling. Environmental
Science & Technology, 49(24), 14614–14624.
Neale P. A., Achard M., Escher B. I. and Leusch F. D. L. (2017a). Exploring the oxidative
stress response mechanism triggered by environmental water samples. Environmental
Science: Processes & Impacts, 19, 1126–1133.
Neale P. A., Altenburger R., Ait-Aissa S., Brion F., Busch W., de Aragão Umbuzeiro G.,
Denison M. S., Du Pasquier D., Hilscherova K., Hollert H., Morales D. A., Novac J.,
Schlichting R., Seiler T.-B., Serra H., Shao Y., Tindall A. J., Tollefsen K. E.,
Williams T. D. and Escher B. I. (2017b). Development of a bioanalytical test battery
for water quality monitoring: Fingerprinting identified micropollutants and their
contribution to effects in surface water. Water Research, 123, 734–750.
Neale P. A., Munz N. A., Ait-Aissa S., Altenburger R., Brion F., Busch W., Escher B. I.,
Hilscherova K., Kienle C., Novak J., Seiler T.-B., Shao Y., Stamm C. and Hollender
J. (2017c). Integrating chemical analysis and bioanalysis to evaluate the contribution
of wastewater effluent on the micropollutant burden in small streams. Science of the
Total Environment, 576, 785–795.
Neale P. A., Brack W., Ait-Aissa S., Busch W., Hollender J., Krauss M., Maillot-Maréchal E.,
Munz N. A., Schlichting R., Schulze T., Vogler B. and Escher B. I. (2018a). Solid-phase
extraction as sample preparation of water samples for cell-based and other in vitro
bioassays. Environmental Science: Processes & Impacts, 20, 493–504.
Neale P. A., Leusch F. D. L. and Escher B. I. (2018b). What is driving the NF-kappa B
response in environmental water extracts? Chemosphere, 210, 645–652.
Neale P. A., Braun G., Brack W., Carmona E., Gunold R., König M., Krauss M., Liebmann
L., Liess M., Link M., Schäfer R. B., Schlichting R., Schreiner V. C., Schulze T.,
Vormeier P., Weisner O. and Escher B. I. (2020a). Assessing the mixture effects in in
vitro bioassays of chemicals occurring in small agricultural streams during rain
events. Environmental Science & Technology, 54(13), 8280–8290.
Neale P. A., Feliers C., Glauch L., König M., Lecarpentier C., Schlichting R., Thibert S. and
Escher B. I. (2020b). Application of in vitro bioassays for water quality monitoring in
three drinking water treatment plants using different treatment processes including
biological treatment, nanofiltration and ozonation coupled with disinfection.
Environmental Science: Water Research & Technology, 6, 2444–2453.
Neale P. A., Grimaldi M., Boulahtouf A., Leusch F. D. L. and Balaguer P. (2020c). Assessing
species-specific differences for nuclear receptor activation for environmental water
extracts. Water Research, 185, 116247.
Neale P. A., O’Brien J. W., Glauch L., König M., Krauss M., Mueller J. F., Tscharke B. and
Escher B. I. (2020d). Wastewater treatment efficacy evaluated with in vitro bioassays.
Water Research X, 9, 100072.
Nelson E. D., Do H., Lewis R. S. and Carr S. A. (2011). Diurnal variability of
pharmaceutical, personal care product, estrogen and alkylphenol concentrations in

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 417

effluent from a tertiary wastewater treatment facility. Environmental Science &

Technology, 45(4), 1228–1234.
Nestmann E. R., Lebel G. L., Williams D. T. and Kowbel D. J. (1979). Mutagenicity
of organic extracts from Canadian drinking water in the Salmonella/
mammalian-microsome assay. Environmental Mutagenesis, 1(4), 337–345.
Newman M. C. (2019). Fundamentals of Ecotoxicology. 5th edn. CRC Press, Boca Raton,
FL, USA.
Nguyen T., Nioi P. and Pickett C. (2009). The Nrf2-antioxidant response element signaling
pathway and its activation by oxidative stress. Journal of Biological Chemistry, 284,
13291–13295.
NHMRC (2011). Australian drinking water guidelines 6. Version 3.5 Updated August
2018, National Health and Medical Research Council (NHMRC) and the
Natural Resource Management Ministerial Council, Canberra, Australia. www.nhmrc.
gov.au/about-us/publications/australian-drinking-water-guidelines. (Accessed 18
February 2021).
Nicholls D. G. (2013). Bioenergetics. 4th edn. Academic Press, New York, NY.
Nicoletti I., Migliorati G., Pagliacci M. C., Grignani F. and Riccardi C. (1991). A rapid and
simple method for measuring thymocyte apoptosis by propidium iodide staining and
flow-cytometry. Journal of Immunological Methods, 139(2), 271–279.
Nisbet I. C. T. and Lagoy P. K. (1992). Toxic equivalency factors (TEFs) for polycyclic
aromatic-hydrocarbons (PAHs). Regulatory Toxicology and Pharmacology, 16(3),
290–300.
Niss F., Rosenmai A. K., Mandava G., Orn S., Oskarsson A. and Lundqvist J. (2018).
Toxicity bioassays with concentrated cell culture media-A methodology to overcome
the chemical loss by conventional preparation of water samples. Environmental
Science and Pollution Research, 25(12), 12183–12188.
Niu L. L., Carmona E., Konig M., Krauss M., Muz M., Xu C., Zou D. L. and Escher B. I.
(2020). Mixture risk drivers in freshwater sediments and their bioavailability
determined using passive equilibrium sampling. Environmental Science &
Technology, 54(20), 13197–13206.
Nivala J., Neale P. A., Haasis T., Kahl S., König M., Müller R. A. M., Reemtsma T.,
Schlichting R. and Escher B. I. (2018). Application of cell-based bioassays to
evaluate treatment efficacy of conventional and intensified treatment wetlands.
Environmental Science: Water Research & Technology, 4, 206–217.
Norberg-King T. J., Embry M. R., Belanger S. E., Braunbeck T., Butler J. D., Dorn P. B., Farr
B., Guiney P. D., Hughes S. A., Jeffries M., Journel R., Leonard M., McMaster M., Oris
J. T., Ryder K., Segner H., Senac T., Van Der Kraak G., Whale G. and Wilson P. (2018).
An international perspective on the tools and concepts for effluent toxicity assessments
in the context of animal alternatives: reduction in vertebrate use. Environmental
Toxicology and Chemistry, 37(11), 2745–2757.
Norli H. R., Christiansen A. and Deribe E. (2011). Application of QuEChERS method for
extraction of selected persistent organic pollutants in fish tissue and analysis by gas
chromatography mass spectrometry. Journal of Chromatography A, 1218(41),
7234–7241.
Novák J., Vrana B., Rusina T., Okonski K., Grabic R., Neale P. A., Escher B. I., Macova M.,
Ait-Aissa S., Creusot N., Allan I. and Hilscherová K. (2018). Effect-based monitoring of

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 418 Bioanalytical Tools in Water Quality Assessment

the Danube River using mobile passive sampling. Science of the Total Environment,
636, 1608–1619.
Novak J., Vaculovic A., Klanova J., Giesy J. P. and Hilscherova K. (2020). Seasonal
variation of endocrine disrupting potentials of pollutant mixtures associated with
various size-fractions of inhalable air particulate matter. Environmental Pollution,
264, 114654.
NRC (1983). Risk assessment in the federal government. Managing the process. National
Research Council (NRC), National Academy Press, Washington, DC, USA.
NRC (2007). Toxicity Testing in the 21st Century: A Vision and a Strategy. National
Research Council (NRC), National Academies Press, Washington, DC. www.nap.
edu/catalog/11970/toxicity-testing-in-the-21st-century-a-vision-and-a (Accessed 18
February 2021).
NRC (2017). Incorporating 21st Century Science into Risk-Based Evaluations. National
Research Council (NRC), National Academies Press, Washington, DC. www.nap.
edu/catalog/24635/using-21st-century-science-to-improve-risk-related-evaluations
(Accessed 18 February 2021).
NRMMC/EPHC/AHMC (2006). Australian Guidelines for Water Recycling: Managing
Health and Environmental Risks (Phase 1). National Water Quality Management
Strategy (NWQMS), Natural Resource Management Ministerial Council,
Environment Protection and Heritage Council, and Australian Health Minister’s
Conference. Canberra, Australia.. www.waterquality.gov.au/media/87 (Accessed 2
January 2021).
NRMMC/EPHC/NHMRC (2008). Australian Guidelines for Water Recycling: Managing
Health and Environmental Risks (Phase 2). Augmentation of drinking water supplies,
National Water Quality Management Strategy (NWQMS), Natural Resource
Management Ministerial Council, Environment Protection and Heritage Council, and
National Health and Medical Research Council, Canberra, Australia. www.
waterquality.gov.au/media/85 (Accessed 2 January 2021).
NRMMC/EPHC/NHMRC (2009a). Australian Guidelines for Water Recycling: Managing
Health and Environmental Risks (Phase 2). Managed aquifer recharge, National
Water Quality Management Strategy (NWQMS), Natural Resource Management
Ministerial Council, Environment Protection and Heritage Council, and National
Health and Medical Research Council, Canberra, Australia. www.waterquality.gov.
au/media/88 (Accessed 2 January 2021).
NRMMC/EPHC/NHMRC (2009b). Australian Guidelines for Water Recycling: Managing
Health and Environmental Risks (Phase 2). Stormwater harvesting and reuse, National
Water Quality Management Strategy (NWQMS), Natural Resource Management
Ministerial Council, Environment Protection and Heritage Council, and National
Health and Medical Research Council, Canberra, Australia.. www.waterquality.gov.
au/media/90 (Accessed 2 January 2021).
Nuwaysir E. F., Bittner M., Trent J., Barrett J. C. and Afshari C. A. (1999). Microarrays
and toxicology: The advent of toxicogenomics. Molecular Carcinogenesis, 24(3),
153–159.
NWC (2011). A national approach to health risk assessment, risk communication and
management of chemical hazards from recycled water. Chapman HF, Leusch FDL,
Prochazka E, Cumming J, Ross V, Humpage AR, Froscio S, Laingam S, Khan SJ,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 419

Trinh T, McDonald JA. Waterlines report No 48. National Water Commission,

Canberra, Australia. www.academia.edu/11644319/A_national_approach_to_health_
risk_assessment_risk_communication_and_management_of_chemical_hazards_from_
recycled_water (Accessed 18 February 2021).
Oberley L. W. and Spitz D. R. (1984). Assay of superoxide-dismutase activity in
tumor-tissue. Methods in Enzymology, 105, 457–464.
O’Connor S., McNamara L., Swerdin M. and Van Buskirk R. G. (1991). Multifluorescent
assays reveal mechanisms underlying cytotoxicity – phase I-CTFA compounds. In
Vitro Toxicology, 4(3), 197–206.
Oda Y., Nakamura S.-I., Oki I., Kato T. and Shinagawa H. (1985). Evaluation of the new
system (umu-test) for the detection of environmental mutagens and carcinogens.
Mutation Research, 147(5), 219–229.
OECD (1992). Test Guidelinek 203. Fish, Acute Toxicity Test. Organisation for Economic
Co-operation and Development, Paris, France.
OECD (1998). OECD Series on Principles of GLP and Compliance Monitoring, Number 1:
OECD Principles on Good Laboratory Practice. Environmental Directorate,
Organisation for Economic Co-operation and Development, Paris, France.
OECD (2004). OECD Series on Principles of GLP and Compliance Monitoring, Number 14:
The Application of the Principles of GLP to In Vitro Studies. Environmental Directorate,
Organisation for Economic Co-operation and Development, Paris, France.
OECD (2006). Guidelines for the Testing of Chemicals. Organisation for Economic
Co-operation and Development, Paris, France.
OECD (2011). Test Guideline 201. Freshwater Alga and Cyanobacteria, Growth Inhibition
Test. Organisation for Economic Co-operation and Development, Paris, France.
OECD (2013). Test Guideline 236. Fish Embryo Acute Toxicity Test. Organisation for
Economic Co-operation and Development, Paris, France.
OECD (2015). Test Guideline. 455. Performance-Based Test Guideline for Stably
Transfected Transactivation In Vitro Assays to Detect Estrogen Receptor Agonists
and Antagonist. Organisation for Economic Co-operation and Development, Paris,
France.
OECD (2018). OECD Series on Testing and Assessment, Number 286: Guidance Document
on Good In Vitro Method Practices (GIVIMP). Environmental Directorate, Organisation
for Economic Co-operation and Development, Paris, France.
Oh S. M., Park K. and Chung K. H. (2006). Combination of in vitro bioassays encompassing
different mechanisms to determine the endocrine-disrupting effects of river water.
Science of the Total Environment, 354(2–3), 252–264.
Ohe T., Suzuki A., Watanabe T., Hasei T., Nukaya H., Totsuka Y. and Wakabayashi K.
(2009). Induction of SCEs in CHL cells by dichlorobiphenyl derivative water
pollutants, 2-phenylbenzotriazole (PBTA) congeners and river water concentrates.
Mutation Research – Genetic Toxicology and Environmental Mutagenesis, 678(1),
38–42.
Omiecinski C. J., Heuvel J. P. V., Perdew G. H. and Peters J. M. (2011). Xenobiotic
metabolism, disposition, and regulation by receptors: From biochemical phenomenon
to predictors of major toxicities. Toxicological Sciences, 120, S49–S75.
Orton F., Ermler S., Kugathas S., Rosivatz E., Scholze M. and Kortenkamp A. (2014).
Mixture effects at very low doses with combinations of anti-androgenic pesticides,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 420 Bioanalytical Tools in Water Quality Assessment

antioxidants, industrial pollutant and chemicals used in personal care products.

Toxicology and Applied Pharmacology, 278(3), 201–208.
Osorio V., Schriks M., Vughs D., de Voogt P. and Kolkman A. (2018). A novel sample
preparation procedure for effect-directed analysis of micro-contaminants of emerging
concern in surface waters. Talanta, 186, 527–537.
Ostling O. and Johanson K. J. (1984). Microelectrophoretic study of radiation-induced DNA
damages in individual mammalian cells. Biochemical and Biophysical Research
Communications, 123(1), 291–298.
Otte J. C., Keiter S., Fassbender C., Higley E. B., Rocha P. S., Brinkmann M., Wahrendorf
D. S., Manz W., Wetzel M. A., Braunbeck T., Giesy J. P., Hecker M. and Hollert H.
(2013). Contribution of priority PAHs and POPs to Ah receptor-mediated activities in
sediment samples from the River Elbe Estuary, Germany. Plos One, 8(10), e75596.
Owens C. W. I. and Belcher R. V. (1965). A colorimetric micro-method for determination of
glutathione. Biochemical Journal, 94(3), 705–711.
Page B., Page M. and Noel C. (1993). A new fluorometric assay for cytotoxicity
measurements in vitro. International Journal of Oncology, 3(3), 473–476.
Pawlowski S., Ternes T., Bonerz M., Kluczka T., van der Burg B., Nau H., Erdinger L. and
Braunbeck T. (2003). Combined in situ and in vitro assessment of the estrogenic activity
of sewage and surface water samples. Toxicological Sciences, 75(1), 57–65.
Payne J., Scholze M. and Kortenkamp A. (2001). Mixtures of four organochlorines enhance
human breast cancer cell proliferation. Environmental Health Perspectives, 109(4),
391–397.
Pellacani C., Buschini A., Furlini M., Poli P. and Rossi C. (2006). A battery of in vivo and in
vitro tests useful for genotoxic pollutant detection in surface waters. Aquatic Toxicology,
77(1), 1–10.
Pelon W., Whitman B. F. and Beasley T. W. (1977). Reversion of histidine-dependent mutant
strains of Salmonella-typhimurium by Mississippi River water samples. Environmental
Science & Technology, 11(6), 619–623.
Perry P. and Wolff S. (1974). New Giemsa method for differential staining of sister
chromatids. Nature, 251(5471), 156–158.
Pessala P., Schultz E., Nakari T., Joutti A. and Herve S. (2004). Evaluation of wastewater
effluents by small-scale biotests and a fractionation procedure. Ecotoxicology and
Environmental Safety, 59(2), 263–272.
Petala M., Kokokiris L., Samaras P., Papadopoulos A. and Zouboulis A. (2009).
Toxicological and ecotoxic impact of secondary and tertiary treated sewage effluents.
Water Research, 43(20), 5063–5074.
Petrie B., Proctor K., Youdan J., Barden R. and Kasprzyk-Hordern B. (2017). Critical
evaluation of monitoring strategy for the multi-residue determination of 90 chiral and
achiral micropollutants in effluent wastewater. Science of the Total Environment, 579,
569–578.
Pieterse B., Felzel E., Winter R., van der Burg B. and Brouwer A. (2013). PAH-CALUX, an
optimized bioassay for AhR-mediated hazard identification of polycyclic aromatic
hydrocarbons (PAHs) as individual compounds and in complex mixtures.
Environmental Science & Technology, 47(20), 11651–11659.
Pillon A., Boussioux A.-M., Escande A., Aït-Aïssa S., Gomez E., Fenet H., Ruff M., Moras
D., Vignon F., Duchesne M.-J., Casellas C., Nicolas J.-C. and Balaguer P. (2005).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 421

Binding of estrogenic compounds to recombinant estrogen receptor-α: Application to

environmental analysis. Environmental Health Perspectives, 113(3), 278–284.
Plassmann M. M., Schmidt M., Brack W. and Krauss M. (2015). Detecting a wide range of
environmental contaminants in human blood samples-combining QuEChERS with
LC-MS and GC-MS methods. Analytical and Bioanalytical Chemistry, 407(23),
7047–7054.
Plewa M. J., Kargalioglu Y., Vankerk D., Minear R. A. and Wagner E. D. (2000).
Development of a quantitative comparative cytotoxicity and genotoxicity assays for
environmental hazardous chemicals. Water Science and Technology, 42(7–8), 109–116.
Plewa M. J., Kargalioglu Y., Vankerk D., Minear R. A. and Wagner E. D. (2002). Mammalian
cell cytotoxicity and genotoxicity analysis of drinking water disinfection by-products.
Environmental and Molecular Mutagenesis, 40(2), 134–142.
Plewa M. J., Wagner E. D. and Mitch W. A. (2011). Comparative mammalian cell
cytotoxicity of water concentrates from disinfected recreational pools. Environmental
Science & Technology, 45(9), 4159–4165.
Plewa M. J., Wagner E. D., Jazwierska P., Richardson S. D., Chen P. H. and McKague A. B.
(2004a). Halonitromethane drinking water disinfection byproducts: Chemical
characterization and mammalian cell cytotoxicity and genotoxicity. Environmental
Science & Technology, 38(1), 62–68.
Plewa M. J., Wagner E. D., Richardson S. D., Thruston A. D., Woo Y.-T. and McKague A. B.
(2004b). Chemical and biological characterization of newly discovered iodoacid
drinking water disinfection byproducts. Environmental Science & Technology, 38
(18), 4713–4722.
Poole C. F. (2003). New trends in solid-phase extraction. TrAC-Trends in Analytical
Chemistry, 22(6), 362–373.
Posthuma L., Suter G. W. and Traas T. P. (2002). Species Sensitivity Distributions in
Ecotoxicology. Lewis, Boca Raton, FL, USA.
Posthuma L., Richards S., De Zwart D., Dyer S., Sibley P., Hickey C. and Altenberger R.
(2008). Mixture extrapolation approaches. In: Extrapolation practice for
ecotoxicological effect characterization of chemicals, K. Solomon, T. Brock,
D. de Zwart et al. (eds.), Taylor & Francis, Boca Raton, FL, USA, pp. 137–185.
Posthuma L., van Gils J., Zijp M. C., van de Meent D. and de Zwart D. (2019). Species
sensitivity distributions for use in environmental protection, assessment, and
management of aquatic ecosystems for 12 386 chemicals. Environmental Toxicology
and Chemistry, 38(4), 905–917.
Powell D. J., Hertzberg R. P. and Macarron R. (2016). Design and implementation of
high-throughput screening assays. In: High Throughput Screening: Methods and
Protocols, 3rd edn. W. P. Janzen, Humana Press, Totowa, NJ, USA, pp. 1–32.
Prasse C., Stalter D., Schulte-Oehlmann U. and Oehlmann J. (2015). Spoilt for choice: A
critical review on the chemical and biological assessment of current wastewater
treatment technologies. Water Research, 87, 237–270.
Price P. S. and Han X. (2011). Maximum cumulative ratio (MCR) as a tool for assessing the
value of performing a cumulative risk assessment. International Journal of
Environmental Research and Public Health, 8(6), 2212–2225.
Prieto P., Blaauboer B. J., de Boer A. G., Boveri M., Cecchelli R., Clemedson C., Coecke S.,
Forsby A., Galla H. J., Garberg P., Greenwood J., Price A. and Tahti H. (2004).
Blood-brain barrier in vitro models and their application in toxicology – The report

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 422 Bioanalytical Tools in Water Quality Assessment

and recommendations of ECVAM workshop 49. ALTA-Alternatives to Laboratory

Animals, 32(1), 37–50.
Prochazkova T., Sychrova E., Vecerkova J., Javurkova B., Otoupalikova A., Pernica M.,
Simek Z., Smutna M., Lepsova-Skacelova O. and Hilscherova K. (2018). Estrogenic
activity and contributing compounds in stagnant water bodies with massive
occurrence of phytoplankton. Water Research, 136, 12–21.
Promega (2009). CytoTox 96® Non-radioactive cytotoxicity assay. Instructions for
use of product G1780. www.promega.com/resources/protocols/technical-
bulletins/0/ cytotox-96-non-radioactive-cytotoxicity-assay-protocol/ (Accessed 14
February 2021).
Purdom C. E., Hardiman P. A., Bye V. V. J., Eno N. C., Tyler C. R. and Sumpter J. P. (1994).
Estrogenic effects of effluents from sewage treatment works. Chemistry and Ecology,
8(4), 275–285.
Quillardet P., Huisman O., Dari R. and Hofnung M. (1982). SOS Chromotest, a direct assay
of induction of an SOS function in Escherichia coli K-12 to measure genotoxicity.
Proceedings of the National Academy of Sciences of the United States of
America-Biological Sciences, 79(19), 5971–5975.
Rajapakse N., Silva E. and Kortenkamp A. (2002). Combining xenoestrogens at levels below
individual no-observed-effect concentrations dramatically enhances steroid hormone
action. Environmental Health Perspectives, 110(9), 917–921.
Ramalhosa M. J., Paiga P., Morais S., Delerue-Matos C. and Oliveira M. (2009). Analysis
of polycyclic aromatic hydrocarbons in fish: Evaluation of a quick, easy, cheap,
effective, rugged, and safe extraction method. Journal of Separation Science, 32(20),
3529–3538.
Rand G. (1995). Fundamentals of Aquatic Toxicology: Effects, Environmental Fate and Risk
Assessment. Taylor & Francis, Washington, DC.
Rappaport S. M., Richard M. G., Hollstein M. C. and Talcott R. E. (1979). Mutagenic activity
in organic wastewater concentrates. Environmental Science & Technology, 13(8),
957–961.
Reid B. J., Jones K. C. and Semple K. T. (2000). Bioavailability of persistent organic
pollutants in soils and sediments – A perspective on mechanisms, consequences and
assessment. Environmental Pollution, 108(1), 103–112.
Reifferscheid G., Heil J., Oda Y. and Zahn R. K. (1991). A microplate version of the
SOS/umu-test for rapid detection of genotoxins and genotoxic potentials of
environmental samples. Mutation Research – Environmental Mutagenesis and
Related Subjects, 253(3), 215–222.
Reiter E. B., Jahnke A., Konig M., Siebert U. and Escher B. I. (2020). Influence of co-dosed
lipids from biota extracts on the availability of chemicals in in vitro cell-based bioassays.
Environmental Science & Technology, 54(7), 4240–4247.
Reungoat J., Macova M., Escher B. I., Carswell S., Mueller J. F. and Keller J. (2010).
Removal of micropollutants and reduction of biological activity in a full-scale
reclamation plant using ozonation and activated carbon filtration. Water Research,
44(2), 625–637.
Riccardi C. and Nicoletti I. (2006). Analysis of apoptosis by propidium iodide staining and
flow cytometry. Nature Protocols, 1(3), 1458–1461.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 423

Richter M. and Escher B. I. (2005). Mixture toxicity of reactive chemicals by using two
bacterial growth assays as indicators of protein and DNA damage. Environmental
Science & Technology, 39, 8753–8761.
Rider C. V., Dinse G. E., Umbach D. M., Simmons J. E. and Hertzberg R. C. (2018). Mixture
toxicity with models of additivity. In: Chemical Mixtures and Combined Chemical and
Nonchemical Stressors, C. Rider and J. Simmons (eds.), Springer, Cham.
Ritter L., Totman C., Krishnan K., Carrier R., Vezina A. and Morisset V. (2007).
Deriving uncertainty factors for threshold chemical contaminants in drinking water.
Journal of Toxicology and Environmental Health-Part B-Critical Reviews, 10(7),
527–557.
Roberts J., Bain P. A., Kumar A., Hepplewhite C., Ellis D. J., Christy A. G. and Beavis S. G.
(2015). Tracking multiple modes of endocrine activity in Australia’s largest inland
sewage treatment plant and effluent-receiving environment using a panel of in vitro
bioassays. Environmental Toxicology and Chemistry, 34(10), 2271–2281.
Rocha P. S., Azab E., Schmidt B., Storch V., Hollert H. and Braunbeck T. (2010). Changes in
toxicity and dioxin-like activity of sediments from the Tiete River (Sao Paulo, Brazil).
Ecotoxicology and Environmental Safety, 73(4), 550–558.
Rodrigues F. P., Angeli J. P. F., Mantovani M. S., Guedes C. L. B. and Jordao B. Q. (2010).
Genotoxic evaluation of an industrial effluent from an oil refinery using plant and animal
bioassays. Genetics and Molecular Biology, 33(1), 169–175.
Rosa R., Moreira-Santos M., Lopes I., Silva L., Rebola J., Mendonca E., Picado A. and
Ribeiro R. (2010). Comparison of a test battery for assessing the toxicity of a
bleached-kraft pulp mill effluent before and after secondary treatment
implementation. Environmental Monitoring and Assessment, 161(1–4), 439–451.
Rosenmai A. K., Lundqvist J., le Godec T., Ohisson A., Troger R., Hellman B. and Oskarsson
A. (2018). In vitro bioanalysis of drinking water from source to tap. Water Research,
139, 272–280.
Rotter S., Beronius A., Boobis A. R., Hanberg A., van Klaveren J., Luijten M., Machera K.,
Nikolopoulou D., van der Voet H., Zilliacus J. and Solecki R. (2018). Overview on
legislation and scientific approaches for risk assessment of combined exposure to
multiple chemicals: The potential EuroMix contribution. Critical Reviews in
Toxicology, 48(9), 796–814.
Roubicek D. A., Rech C. M. and Umbuzeiro G. A. (2020). Mutagenicity as a parameter in
surface water monitoring programs-opportunity for water quality improvement.
Environmental and Molecular Mutagenesis, 61(1), 200–211.
Routledge E. J. and Sumpter J. P. (1996). Estrogenic activity of surfactants and some of their
degradation products assessed using a recombinant yeast screen. Environmental
Toxicology and Chemistry, 15(3), 241–248.
Rummel C. D., Escher B. I., Sandblom O., Plassmann M. M., Arp H. P. H., MacLeod M. and
Jahnke A. (2019). Effects of leachates from UV-weathered microplastic in cell-based
bioassays. Environmental Science & Technology, 53(15), 9214–9223.
Russom C. L., LaLone C. A., Villeneuve D. L. and Ankley G. T. (2014). Development of an
adverse outcome pathway for acetylcholinesterase inhibition leading to acute mortality.
Environmental Toxicology and Chemistry, 33(10), 2157–2169.
Rutishauser B. V., Pesonen M., Escher B. I., Ackermann G. E., Aerni H. R., Suter M. J. F. and
Eggen R. I. L. (2004). Comparative analysis of estrogenic activity in sewage treatment

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 424 Bioanalytical Tools in Water Quality Assessment

plant effluents involving three in vitro assays and chemical analysis of steroids.
Environmental Toxicology and Chemistry, 23(4), 857–864.
Rydberg B. and Johanson K. J. (1978). Estimation of DNA strand breaks in single
mammalian cells. In: DNA repair mechanisms, P. C. Hanawalt, E. C. Friedberg and
C. F. Fox (eds.), Academic Press, New York, NY, USA. pp. 465–468.
Sanchez P. S., Sato M. I. Z., Paschoal C., Alves M. N., Furlan E. V. and Martins M. T. (1988).
Toxicity assessment of industrial effluents from S. Paulo State, Brazil, using short-term
microbial assays. Toxicity Assessment, 3(1), 55–80.
Sakamuru S, Zhu H, Xia M, Simeonov A and Huang R. 2020. Profiling the Tox21
Chemical Library for Environmental Hazards: Applications in Prioritisation,
Predictive Modelling, and Mechanism of Toxicity Characterisation. In D Neagu and
AN Richarz, eds, Big Data in Predictive Toxicology. -Issues in Toxicology, 41,
242–263.
Sauer U. G., Deferme L., Gribaldo L., Hackermueller J., Tralau T., van Ravenzwaay B., Yauk
C., Poole A., Tong W. and Gant T. W. (2017). The challenge of the application of ’omics
technologies in chemicals risk assessment: Background and outlook. Regulatory
Toxicology and Pharmacology, 91, S14–S26.
Šauer P., Borik A., Golovko O., Grabic R., Stanova A. V., Valentova O., Stara A., Sandova
M. and Kroupova H. K. (2018). Do progestins contribute to (anti-)androgenic activities
in aquatic environments? Environmental Pollution, 242, 417–425.
Saxena J. and Schwartz D. J. (1979). Mutagens in wastewaters renovated by advanced
wastewater-treatment. Bulletin of Environmental Contamination and Toxicology,
22(3), 319–326.
Scarsi M., Podvinec M., Roth A., Hug H., Kersten S., Albrecht H., Schwede T., Meyer U. A.
and Rucker C. (2007). Sulfonylureas and glinides exhibit peroxisome
proliferator-activated receptor gamma activity: A combined virtual screening and
biological assay approach. Molecular Pharmacology, 71(2), 398–406.
Schiliró T., Pignata C., Fea E. and Gilli G. (2004). Toxicity and estrogenic activity of a
wastewater treatment plant in northern Italy. Archives of Environmental
Contamination and Toxicology, 47(4), 456–462.
Schirmer K. (2006). Proposal to improve vertebrate cell cultures to establish them as
substitutes for the regulatory testing of chemicals and effluents using fish. Toxicology,
224(3), 163–183.
Schirmer K., Tom D. J., Bols N. C. and Sherry J. P. (2001). Ability of fractionated petroleum
refinery effluent to elicit cyto- and photocytotoxic responses and to induce
7-ethoxyresorufin-O-deethylase activity in fish cell lines. Science of the Total
Environment, 271(1–3), 61–78.
Schnurstein A. and Braunbeck T. (2001). Tail moment versus tail length – Application of an
in vitro version of the comet assay in biomonitoring for genotoxicity in native surface
waters using primary hepatocytes and gill cells from zebrafish (Danio rerio).
Ecotoxicology and Environmental Safety, 49(2), 187–196.
Scholze M., Boedeker W., Faust M., Backhaus T., Altenburger R. and Grimme L. H. (2001).
A general best-fit method for concentration-response curves and the estimation of
low-effect concentrations. Environmental Toxicology and Chemistry, 20(2), 448–457.
Schreer A., Tinson C., Sherry J. P. and Schirmer K. (2005). Application of Alamar
blue/5-carboxyfluorescein diacetate acetoxymethyl ester as a noninvasive cell

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 425

viability assay in primary hepatocytes from rainbow trout. Analytical Biochemistry, 344
(1), 76–85.
Schriks M., Heringa M. B., van der Kooi M. M. E., de Voogt P. and van Wezel A. P. (2010a).
Toxicological relevance of emerging contaminants for drinking water quality. Water
Research, 44(2), 461–476.
Schriks M., van Leerdam J. A., van der Linden S. C., van der Burg B., van Wezel A. P. and de
Voogt P. (2010b). High-resolution mass spectrometric identification and quantification
of glucocorticoid compounds in various wastewaters in the Netherlands. Environmental
Science & Technology, 44(12), 4766–4774.
Schriks M., van der Linden S. C., Stoks P. G. M., van der Burg B., Puijker L., de Voogt P. and
Heringa M. B. (2013). Occurrence of glucocorticogenic activity in various surface
waters in the Netherlands. Chemosphere, 93(2), 450–454.
Schroeder A. L., Ankley G. T., Houck K. A. and Villeneuve D. L. (2016). Environmental
surveillance and monitoring: The next frontiers for high-throughput toxicology.
Environmental Toxicology and Chemistry, 35(3), 513–525.
Schulze T., Ulrich M., Maier D., Maier M., Terytze K., Braunbeck T. and Hollert H. (2015).
Evaluation of the hazard potentials of river suspended particulate matter and floodplain
soils in the Rhine basin using chemical analysis and in vitro bioassays. Environmental
Science and Pollution Research, 22(19), 14606–14620.
Schulze T., Ahel M., Ahlheim J., Ait-Aissa S., Brion F., Di Paolo C., Froment J., Hidasi A. O.,
Hollender J., Hollert H., Hu M., Klolss A., Koprivica S., Krauss M., Muz M., Oswald P.,
Petre M., Schollee J. E., Seiler T. B., Shao Y., Slobodnik J., Sonavane M., Suter M. J. F.,
Tollefsen K. E., Tousova Z., Walz K. H. and Brack W. (2017). Assessment of a novel
device for onsite integrative large-volume solid phase extraction of water samples to
enable a comprehensive chemical and effect-based analysis. Science of the Total
Environment, 581, 350–358.
Schwarzenbach R. P., Escher B. I., Fenner K., Hofstetter T. B., Johnson C. A., von Gunten U.
and Wehrli B. (2006). The challenge of micropollutants in aquatic systems. Science,
313(5790), 1072–1077.
Schweigert N., Eggen R. I. L., Escher B. I., Burkhardt-Holm P. and Behra R. (2002).
Ecotoxicological assessment of surface waters: A modular approach integrating in
vitro methods. ALTEX-Alternativen zu Tierexperimenten, 19(Suppl. 1), 30–37.
Schymanski E. L., Singer H. P., Longree P., Loos M., Ruff M., Stravs M. A., Vidal C. R. and
Hollender J. (2014). Strategies to characterize polar organic contamination in
wastewater: exploring the capability of high resolution mass spectrometry.
Environmental Science & Technology, 48(3), 1811–1818.
Scott P. D., Bartkow M., Blockwell S. J., Coleman H. M., Khan S. J., Lim R., McDonald J. A.,
Nice H., Nugegoda D., Pettigrove V., Tremblay L. A., Warne M. S. J. and Leusch F. D. L.
(2014). An assessment of endocrine activity in Australian rivers using chemical and in
vitro analyses. Environmental Science and Pollution Research, 21(22), 12951–12967.
Scott P. D., Coleman H. M., Khan S., Limc R., McDonald J. A., Mondon J., Neale P. A.,
Prochazka E., Tremblay L. A., Warne M. S. and Leusch F. D. L. (2018).
Histopathology, vitellogenin and chemical body burden in mosquitofish (Gambusia
holbrooki) sampled from six river sites receiving a gradient of stressors. Science of
the Total Environment, 616, 1638–1648.
Seibert H., Balls M., Fentem J. H., Bianchi V., Clothier R. H., Dierickx P. J., Ekwall B., Garle
M. J., GomezLechon M. J., Gribaldo L., Gulden M., Liebsch M., Rasmussen E., Roguet

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 426 Bioanalytical Tools in Water Quality Assessment

R., Shrivastava R. and Walum E. (1996). Acute toxicity testing in vitro and the
classification and labelling of chemicals – The report and recommendations of
ECVAM workshop 16. ATLA-Alternatives to Laboratory Animals, 24(4), 499–510.
Seiler T.-B., Best N., Fernqvist M. M., Hercht H., Smith K. E. C., Braunbeck T., Mayer P. and
Hollert H. (2014). PAH toxicity at aqueous solubility in the fish embryo test with Danio
rerio using passive dosing. Chemosphere, 112, 77–84.
Seimandi M., Lemaire G., Pillon A., Perrin A., Carlavan I., Voegel J. J., Vignon F., Nicolas
J. C. and Balaguer P. (2005). Differential responses of PPAR alpha, PPAR delta, and
PPAR gamma reporter cell lines to selective PPAR synthetic ligands. Analytical
Biochemistry, 344(1), 8–15.
Semple K. T., Doick K. J., Jones K. C., Burauel P., Craven A. and Harms H. (2004). Defining
bioavailability and bioaccessibility of contaminated soil and sediment is complicated.
Environmental Science and Technology, 38(12), 228A–231A.
Serra H., Brion F., Chardon C., Budzinski H., Schulze T., Brack W. and Ait-Aissa S.
(2020). Estrogenic activity of surface waters using zebrafish- and human-based in
vitro assays: The Danube as a case-study. Environmental Toxicology and
Pharmacology, 78, 103401.
Shen C., Huang S., Wang Z., Qiao M., Tang X., Yu C., Shi D., Zhu Y., Shi J., Chen X., Setty
K. and Chen Y. (2008). Identification of Ah receptor agonists in soil of E-waste
recycling sites from Taizhou area in China. Environmental Science & Technology,
42(1), 49–55.
Shi W., Wang X. Y., Hu W., Sun H., Shen O. X., Liu H. L., Wang X. R., Giesy J. P., Cheng
S. P. and Yu H. X. (2009a). Endocrine-disrupting equivalents in industrial effluents
discharged into Yangtze River. Ecotoxicology, 18(6), 685–692.
Shi Y., Cao X.-W., Tang F., Du H.-R., Wang Y.-Z., Qiu X.-D., Yu H.-P. and Lu B. (2009b).
In vitro toxicity of surface water disinfected by different sequential treatments. Water
Research, 43(1), 218–228.
Shi P., Zhou S. C., Xiao H. X., Qiu J. F., Li A. M., Zhou Q., Pan Y. and Hollert H. (2018).
Toxicological and chemical insights into representative source and drinking water in
eastern China. Environmental Pollution, 233, 35–44.
Shoemaker J. T., Zhang W., Atlas S. I., Bryan R. A., Inman S. W. and Vukasinovic J. (2020).
A 3D cell culture organ-on-a-chip platform with a breathable hemoglobin analogue
augments and extends primary human hepatocyte functions in vitro. Frontiers in
Molecular Biosciences, 7, 568777.
Shue M. F., Chen F. A., Kuo Y. T. and Chen T. C. (2009). Nonylphenol concentration and
estrogenic activity in Kaoping River and its tributaries, Taiwan. Water Science and
Technology, 59(10), 2055–2063.
Shukla S. J., Huang R. L., Austin C. P. and Xia M. H. (2010). The future of toxicity testing: A
focus on in vitro methods using a quantitative high-throughput screening platform. Drug
Discovery Today, 15(23–24), 997–1007.
Sidlova T., Novak J., Janosek J., Andel P., Giesy J. P. and Hilscherova K. (2009). Dioxin-like
and endocrine disruptive activity of traffic-contaminated soil samples. Archives of
Environmental Contamination and Toxicology, 57(4), 639–650.
Silva E., Rajapakse N. and Kortenkamp A. (2002). Something from “nothing” – eight weak
estrogenic chemicals combined at concentrations below NOECs produce significant
mixture effects. Environmental Science & Technology, 36, 1751–1756.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 427

Simmon V. F. and Tardiff R. G. (1976). Mutagenic activity of drinking-water concentrates.

Mutation Research, 38(6), 389–390.
Simmons S. O., Fan C. Y. and Ramabhadran R. (2009). Cellular stress response pathway
system as a sentinel ensemble in toxicological screening. Toxicological Sciences, 111
(2), 202–225.
Simon T., Britt J. K. and James R. C. (2007). Development of a neurotoxic equivalence
scheme of relative potency for assessing the risk of PCB mixtures. Regulatory
Toxicology and Pharmacology, 48(2), 148–170.
Simon E., van Velzen M., Brandsma S. H., Lie E., Loken K., de Boer J., Bytingsvik J.,
Jenssen B. M., Aars J., Hamers T. and Lamoree M. H. (2013). Effect-directed
analysis to explore the polar bear exposome: Identification of thyroid hormone
disrupting compounds in plasma. Environmental Science & Technology, 47(15),
8902–8912.
Simon E., Schifferli A., Bucher T. B., Olbrich D., Werner I. and Vermeirssen E. L. M. (2019).
Solid-phase extraction of estrogens and herbicides from environmental waters for
bioassay analysis – effects of sample volume on recoveries. Analytical and
Bioanalytical Chemistry, 411(10), 2057–2069.
Singh N. P., McCoy M. T., Tice R. R. and Schneider E. L. (1988). A simple technique for
quantitation of low levels of DNA damage in individual cells. Experimental Cell
Research, 175(1), 184–191.
Sipes N. S., Wambaugh J. F., Pearce R., Auerbach S. S., Wetmore B. A., Hsieh J. H., Shapiro
A. J., Svoboda D., DeVito M. J. and Ferguson S. S. (2017). An intuitive approach for
predicting potential human health risk with the Tox21 10k Library. Environmental
Science & Technology, 51(18), 10786–10796.
Skarek M., Janosek J., Cupr P., Kohoutek J., Novotna-Rychetska A. and Holoubek I. (2007).
Evaluation of genotoxic and non-genotoxic effects of organic air pollution using in vitro
bioassays. Environment International, 33(7), 859–866.
Skehan P., Storeng R., Scudiero D., Monks A., McMahon J., Vistica D., Warren J. T.,
Bokesch H., Kenney S. and Boyd M. R. (1990). New colorimetric cytotoxicity assay
for anticancer-drug screening. Journal of the National Cancer Institute, 82(13),
1107–1112.
Smith K. E. C. and Schäfer S. (2016). Defining and controlling exposure during in vitro
toxicity testing and the potential of passive dosing. Advances in Biochemical
Engineering / Biotechnology, 157, 263–292.
Smith K. E. C., Oostingh G. J. and Mayer P. (2010). Passive dosing for producing defined and
constant exposure of hydrophobic organic compounds during in vitro toxicity tests.
Chemical Research in Toxicology, 23(1), 55–65.
Smith K. E. C., Jeong Y. and Kim J. (2015). Passive dosing versus solvent spiking for
controlling and maintaining hydrophobic organic compound exposure in the Microtox
(R) assay. Chemosphere, 139, 174–180.
Sohoni P. and Sumpter J. P. (1998). Several environmental oestrogens are also
anti-androgens. Journal of Endocrinology, 158(3), 327–339.
Song M., Jiang Q., Xu Y., Liu H., Lam P. K. S., O’Toole D. K., Zhang Q., Giesy J. P. and
Jiang G. (2006). AhR-active compounds in sediments of the Haihe and Dagu Rivers,
China. Chemosphere, 63(7), 1222–1230.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 428 Bioanalytical Tools in Water Quality Assessment

Sonneveld E., Pieterse B., Schoonen W. G. and van der Burg B. (2011). Validation of in vitro
screening models for progestagenic activities: Inter-assay comparison and correlation
with in vivo activity in rabbits. Toxicology In Vitro, 25(2), 545–554.
Soto A. M., Sonnenschein C., Chung K. L., Fernandez M. F., Olea N. and Serrano F. O.
(1995). The E-SCREEN assay as a tool to identify estrogens: An update on estrogenic
environmental pollutants. Environmental Health Perspectives, 103(Suppl. 7), 113–122.
Souders C. L., Liang X. F., Wang X. H., Ector N., Zhao Y. H. and Martyniuk C. J. (2018).
High-throughput assessment of oxidative respiration in fish embryos: Advancing
adverse outcome pathways for mitochondrial dysfunction. Aquatic Toxicology, 199,
162–173.
Spurgeon D. J., Jones O. A. H., Dorne J., Svendsen C., Swain S. and Sturzenbaum S. R.
(2010). Systems toxicology approaches for understanding the joint effects of
environmental chemical mixtures. Science of the Total Environment, 408(18),
3725–3734.
Spycher S., Netzeva T. I., Worth A. and Escher B. I. (2008). Mode of action-based
classification and prediction of activity of uncouplers for the screening of chemical
inventories. SAR and QSAR in Environmental Research, 19, 433–463.
Stadnicka-Michalak J., Tanneberger K., Schirmer K. and Ashauer R. (2014). Measured and
modeled toxicokinetics in cultured fish cells and application to in vitro-in vivo toxicity
extrapolation. Plos One, 9(3), e92303.
Stalter D., Magdeburg A. and Oehlmann J. (2010a). Comparative toxicity assessment of
ozone and activated carbon treated sewage effluents using an in vivo test battery.
Water Research, 44(8), 2610–2620.
Stalter D., Magdeburg A., Weil M., Knacker T. and Oehlmann J. (2010b). Toxication or
detoxication? In vivo toxicity assessment of ozonation as advanced wastewater
treatment with the rainbow trout. Water Research, 44(2), 439–448.
Stalter D., Magdeburg A., Wagner M. and Oehlmann J. (2011). Ozonation and activated
carbon treatment of sewage effluents: Removal of endocrine activity and cytotoxicity.
Water Research, 45, 1015–1024.
Stalter D., Dutt M. and Escher B. I. (2013). Headspace-free setup of in vitro bioassays for the
evaluation of volatile disinfection by-products. Chemical Research in Toxicology, 26,
1605–1614.
Stalter D., O’Malley E., von Gunten U. and Escher B. I. (2016a). Fingerprinting the reactive
toxicity pathways of 50 drinking water disinfection by-products. Water Research, 91,
19–30.
Stalter D., O’Malley E., von Gunten U. and Escher B. I. (2016b). Point-of-use water filters can
effectively remove disinfection by-products and toxicity from chlorinated and
chloraminated tap water. Environmental Science: Water Research & Technology, 2
(5), 875–883.
Stalter D., Peters L. I., O’Malley E., Tang J. Y. M., Revalor M., Farre M. J., Watson K., von
Gunten U. and Escher B. I. (2016c). Sample enrichment for bioanalytical assessment of
disinfected drinking water: Concentrating the polar, the volatiles, and the unknowns.
Environmental Science & Technology, 50, 6495–6505.
State Water Resources Control Board (2019). Water quality control policy for recycled water.
California Environmental Protection Agency, USA. www.waterboards.ca.gov/

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 429

board_decisions/adopted_orders/resolutions/2018/121118_7_final_amendment_oal.pdf
(Accessed 16 November 2020).
Sumida K., Igarashi Y., Toritsuka N., Matsushita T., Abe-Tomizawa K., Aoki M., Urushidani
T., Yamada H. and Ohno Y. (2011). Effects of DMSO on gene expression in human and
rat hepatocytes. Human & Experimental Toxicology, 30(10), 1701–1709.
Sumpter J. P. (2002). Endocrine disruption and feminization in fish. Toxicology, 178(1), 39–40.
Sumpter J. P. and Jobling S. (1995). Vitellogenesis as a biomarker for estrogenic
contamination of the aquatic environment. Environmental Health Perspectives, 103,
173–178.
Sun J., Zhang R., Qin L., Zhu H. X., Huang Y., Xue Y. G., An S. Q., Xie X. C. and Li A. M.
(2017). Genotoxicity and cytotoxicity reduction of the polluted urban river after
ecological restoration: A field-scale study of Jialu River in northern China.
Environmental Science and Pollution Research, 24(7), 6715–6723.
Suzuki G., Tue N. M., van der Linden S., Brouwer A., van der Burg B., van Velzen M.,
Lamoree M., Someya M., Takahashi S., Isobe T., Tajima Y., Yamada T. K.,
Takigami H. and Tanabe S. (2011). Identification of major dioxin-like compounds
and androgen receptor antagonist in acid-treated tissue extracts of high trophic-level
animals. Environmental Science and Technology, 45(23), 10203–10211.
Suzuki G., Nakamura M., Michinaka C., Nguyen Minh T., Handa H. and Takigami H. (2017).
Separate screening of brominated and chlorinated dioxins in field samples using in vitro
reporter gene assays with rat and mouse hepatoma cell lines. Analytica Chimica Acta,
975, 86–95.
Swart J. C., Pool E. J. and van Wyk J. H. (2011). The implementation of a battery of in vivo
and in vitro bioassays to assess river water for estrogenic endocrine disrupting
chemicals. Ecotoxicology and Environmental Safety, 74(1), 138–143.
Tan L. and Schirmer K. (2017). Cell culture-based biosensing techniques for detecting
toxicity in water. Current Opinion in Biotechnology, 45, 59–68.
Tang J. Y. M. and Escher B. I. (2014). Realistic environmental mixtures of micropollutants in
wastewater, recycled water and surface water: herbicides dominate the mixture toxicity
towards algae. Environmental Toxicology and Chemistry, 33(6), 1427–1436.
Tang J. Y. M., Glenn E., Aryal R., Gernjak W. and Escher B. I. (2013a). Toxicity
characterization of urban stormwater with bioanalytical tools. Water Research 47(15),
5594–5606.
Tang J. Y. M., McCarty S., Glenn E., Neale P. A., Warne M. S. and Escher B. I. (2013b).
Mixture effects of organic micropollutants present in water: Towards the development
of effect-based water quality trigger values for baseline toxicity. Water Research 47
(10), 3300–3314.
Tang J. Y. M., Busetti F., Charrois J. W. A. and Escher B. I. (2014). Which chemicals drive
biological effects in wastewater and recycled water? Water Research, 60, 289–299.
Tanneberger K., Knöbel M., Busser F. J. M., Sinnige T. L., Hermens J. L. M. and Schirmer K.
(2013). Predicting fish acute toxicity using a fish gill cell line-based toxicity assay.
Environmental Science & Technology, 47(2), 1110–1119.
Terada H. (1990). Uncouplers of oxidative phosphorylation. Environmental Health
Perspectives, 87, 213–218.
Teuschler L. K. (2007). Deciding which chemical mixtures risk assessment methods work
best for what mixtures. Toxicology and Applied Pharmacology, 223(2), 139–147.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 430 Bioanalytical Tools in Water Quality Assessment

Thomas R. S., Paules R. S., Simeonov A., Fitzpatrick S. C., Crofton K. M., Casey W. M. and
Mendrick D. L. (2018). The US Federal Tox21 program: A strategic and operational plan
for continued leadership. ALTEX-Alternatives to Animal Experimentation, 35(2),
163–168.
Thomas R. S., Bahadori T., Buckley T. J., Cowden J., Deisenroth C., Dionisio K. L., Frithsen
J. B., Grulke C. M., Gwinn M. R., Harrill J. A., Higuchi M., Houck K. A., Hughes M. F.,
Hunter E. S., Isaacs K. K., Judson R. S., Knudsen T. B., Lambert J. C., Linnenbrink M.,
Martin T. M., Newton S. R., Padilla S., Patlewicz G., Paul-Friedman K., Phillips K. A.,
Richard A. M., Sams R., Shafer T. J., Setzer R. W., Shah I., Simmons J. E., Simmons
S. O., Singh A., Sobus J. R., Strynar M., Swank A., Tornero-Valez R., Ulrich E. M.,
Villeneuve D. L., Wambaugh J. F., Wetmore B. A. and Williams A. J. (2019). The
next generation blueprint of computational toxicology at the US environmental
protection agency. Toxicological Sciences, 169(2), 317–332.
Thorpe K. L., Gross-Sorokin M., Johnson I., Brighty G. and Tyler C. R. (2006). An
assessment of the model of concentration addition for predicting the estrogenic
activity of chemical mixtures in wastewater treatment works effluents. Environmental
Health Perspectives, 114, 90–97.
Tiffany-Castiglioni E., Hong S., Qian Y., Tang Y. and Donnelly K. C. (2006). In vitro models
for assessing neurotoxicity of mixtures. Neurotoxicology, 27(5), 835–839.
Timbrell J. (2009). Principles of Biochemical Toxicology, 4th edn. Informa Healthcare,
New York, US.
Timm M., Saaby L., Moesby L. and Hansen E. W. (2013). Considerations regarding use of
solvents in in vitro cell based assays. Cytotechnology, 65(5), 887–894.
Timourian H., Felton J. S., Stuermer D. H., Healy S., Berry P., Tompkins M., Battaglia G.,
Hatch F. T., Thompson L. H., Carrano A. V., Minkler J. and Salazar E. (1982).
Mutagenic and toxic activity of environmental effluents from underground coal
gasification experiments. Journal of Toxicology and Environmental Health, 9(5–6),
975–994.
Toušová Z., Oswald P., Slobodnik J., Blaha L., Muz M., Hu M., Brack W., Krauss M., Di
Paolo C., Tarcai Z., Seiler T. B., Hollert H., Koprivica S., Ahel M., Schollee J. E.,
Hollender J., Suter M. J. F., Hidasi A. O., Schirmer K., Sonavane M., Ait-Aissa S.,
Creusot N., Brion F., Froment J., Almeida A. C., Thomas K., Tollefsen K. E., Tufi
S., Ouyang X. Y., Leonards P., Lamoree M., Torrens V. O., Kolkman A., Schriks
M., Spirhanzlova P., Tindall A. and Schulze T. (2017). European demonstration
program on the effect-based and chemical identification and monitoring of organic
pollutants in European surface waters. Science of the Total Environment, 601,
1849–1868.
Toušová Z., Vrana B., Smutná M., Novák J., Klučárová V., Grabic R., Slobodník J., Giesy
J. P. and Hilscherová K. (2019). Analytical and bioanalytical assessments of organic
micropollutants in the Bosna River using a combination of passive sampling,
bioassays and multi-residue analysis. Science of the Total Environment, 650,
1599–1612.
Troger R., Klockner P., Ahrens L. and Wiberg K. (2018). Micropollutants in drinking water
from source to tap – Method development and application of a multiresidue screening
method. Science of the Total Environment, 627, 1404–1432.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 431

Tsuno H., Arakawa K., Kato Y. and Nagare H. (2008). Advanced sewage treatment with
ozone under excess sludge reduction, disinfection and removal of EDCs.
Ozone-Science & Engineering, 30(3), 238–245.
Tue N. M., Suzuki G., Takahashi S., Kannan K., Takigami H. and Tanabe S. (2013).
Dioxin-related compounds in house dust from New York State: Occurrence, in vitro
toxic evaluation and implications for indoor exposure. Environmental Pollution, 181,
75–80.
Tuikka A. I., Schmitt C., Hoss S., Bandow N., von der Ohe P. C., de Zwart D., de Deckere E.,
Streck G., Mothes S., van Hattum B., Kocan A., Brix R., Brack W., Barcelo D.,
Sormunen A. J. and Kukkonen J. V. (2011). Toxicity assessment of sediments from
three European river basins using a sediment contact test battery. Ecotoxicology and
Environmental Safety, 74(1), 123–131.
Tuncer S., Gurbanov R., Sheraj I., Solel E., Esenturk O. and Banerjee S. (2018). Low dose
dimethyl sulfoxide driven gross molecular changes have the potential to interfere with
various cellular processes. Scientific Reports, 8, 14828.
U.S. EPA (1976). Toxic Substances Control Act. Washington, DC, USA.
U.S. EPA (1986). Guidelines for the health risk assessment of chemical mixtures. Federal
Register, 51, 34014–34025.
U.S. EPA (2002a). Guidance on cumulative risk assessment of pesticide chemicals that
have a common mechanism of toxicity. Washington, DC. www.epa.gov/pesticide-
science-and-assessing-pesticide-risks/guidance-cumulative-risk-assessment-pesticide
(Accesssed on 18 February 2021).
U.S. EPA (2002b). Methods for Measuring the Acute Toxicity of Effluents and Receiving
Waters to Freshwater and Marine Organisms. 5th edn. EPA-821-R-02-012. www.
epa.gov/sites/production/files/2015-08/documents/acute-freshwater-and-marine-wet-
manual_2002.pdf (Accesssed on 18 February 2021).
U.S. EPA (2002c). Short-Term Methods for Estimating the Chronic Toxicity of Effluents
and Receiving Water to Marine and Estuarine Organisms. www.epa.gov/sites/
production/files/2015-08/documents/short-term-chronic-marine-and-estuarine-wet-
manual_2002.pdf (Accesssed on 18 February 2021).
U.S. EPA (2007). Method: 1694, Pharmaceuticals and personal care products in water, soil,
sediment, and biosolids by HPLC/MS/MS. In: EPA-821-R-08-002, Washington
DC. www.epa.gov/sites/production/files/2015-10/documents/method_1694_2007.pdf
(Accesssed on 18 February 2021).
U.S. EPA (2018). 2018 Edition of the drinking water standards and health advisories. www.
epa.gov/sites/production/files/2018-03/documents/dwtable2018.pdf (Accesssed on
18 February 2021).
U.S. EPA (2020). CompTox Chemistry Dashboard. comptox.epa.gov/dashboard/ (Accessed
20 December 2020).
Ulitzur S., Weiser I. and Yannai S. (1980). A new, sensitive and simple bioluminescence test
for mutagenic compounds. Mutation Research – Environmental Mutagenesis and
Related Subjects, 74(2), 113–124.
United Nations (1992). Report of the United Nations Conference on Environment
and Development. www.un.org/en/development/desa/population/migration/
generalassembly/docs/globalcompact/A_CONF.151_26_Vol.I_Declaration.pdf
(Accessed 23 October 2020).

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 432 Bioanalytical Tools in Water Quality Assessment

United Nations (2009). Stockholm Convention on Persistent Organic Pollutants. Geneva,

Switzerland, and New York, NY, USA.
United Nations (2019). Globally Harmonized System of Classification and Labelling of
Chemicals (GHS), 8th revised edition. Geneva, Switzerland, and New York, NY, USA.
Valcarcel Y., Valdehita A., Becerra E., de Alda M. L., Gil A., Gorga M., Petrovic M., Barcelo
D. and Navas J. M. (2018). Determining the presence of chemicals with suspected
endocrine activity in drinking water from the Madrid region (Spain) and assessment
of their estrogenic, androgenic and thyroidal activities. Chemosphere, 201, 388–398.
Välitalo P., Massei R., Heiskanen I., Behnisch P., Brack W., Tindall A. J., Du Pasquier D.,
Küster E., Mikola A., Schulze T. and Sillanpää M. (2017). Effect-based assessment of
toxicity removal during wastewater treatment. Water Research, 126, 153–163.
Vallotton N. and Price P. S. (2016). Use of the maximum cumulative ratio as an approach for
prioritizing aquatic coexposure to plant protection products: A case study of a large
surface water monitoring database. Environmental Science & Technology, 50(10),
5286–5293.
van Broekhuizen F. L. P. and Traas T. (2016). Addressing combined effects of chemicals in
environmental safety assessment under REACH – a thought starter. RIVM Letter report
2016-0162, National Institute for Public Health and the Environment, Bilthoven, The
Netherlands. www.rivm.nl/bibliotheek/rapporten/2016-0162.pdf (Accessed 10 April
2021).
van Dam R. A. and Chapman J. C. (2001). Direct toxicity assessment (DTA) for water quality
guidelines in Australia and New Zealand. Australasian Journal of Ecotoxicology, 7,
175–198.
van de Merwe J. P. and Leusch F. D. L. (2015). A sensitive and high throughput bacterial
luminescence assay for assessing aquatic toxicity – the BLT-screen. Environmental
Science-Processes & Impacts, 17(5), 947–955.
Van den Berg M., Birnbaum L. S., Denison M., De Vito M., Farland W., Feeley M., Fiedler
H., Hakansson H., Hanberg A., Haws L., Rose M., Safe S., Schrenk D., Tohyama C.,
Tritscher A., Tuomisto J., Tysklind M., Walker N. and Peterson R. E. (2006). The
2005 World Health Organization reevaluation of human and mammalian toxic
equivalency factors for dioxins and dioxin-like compounds. Toxicological Sciences,
93(2), 223–241.
van der Lelie D., Regniers L., Borremans B., Provoost A. and Verschaeve L. (1997). The
VITOTOX(R) test, an SOS bioluminescence Salmonella typhimurium test to measure
genotoxicity kinetics. Mutation Research-Genetic Toxicology and Environmental
Mutagenesis, 389(2–3), 279–290.
van der Linden S. C., Heringa M. B., Man H. Y., Sonneveld E., Puijker L. M., Brouwer A. and
Van der Burg B. (2008). Detection of multiple hormonal activities in wastewater
effluents and surface water, using a panel of steroid receptor CALUX bioassays.
Environmental Science & Technology, 42(15), 5814–5820.
van der Linden S. C., von Bergh A. R. M., van Vught-Lussenburg B. M. A., Jonker L. R. A.,
Teunis M., Krul C. A. M. and van der Burg B. (2014). Development of a panel of
high-throughput reporter-gene assays to detect genotoxicity and oxidative stress.
Mutation Research-Genetic Toxicology and Environmental Mutagenesis, 760, 23–32.
van der Oost R., Sileno G., Janse T., Nguyen M. T., Besselink H. and Brouwer A. (2017a).
SIMONI (Smart Integrated Monitoring) as a novel bioanalytical strategy for water

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 433

quality assessment: Part II–field feasibility survey. Environmental Toxicology and

Chemistry, 36(9), 2400–2416.
van der Oost R., Sileno G., Suarez-Munoz M., Nguyen M. T., Besselink H. and Brouwer A.
(2017b). SIMONI (Smart Integrated Monitoring) as a novel bioanalytical strategy for
water quality assessment: Part I–model design and effect-based trigger values.
Environmental Toxicology and Chemistry, 36(9), 2385–2399.
van Leeuwen C. J. and Vermeire T. G. (2007). Risk Assessment of Chemicals: An
Introduction. 2nd edn. Springer, Dordrecht, The Netherlands.
van Leeuwen K., Schultz T. W., Henry T., Diderich B. and Veith G. D. (2009). Using
chemical categories to fill data gaps in hazard assessment. SAR and QSAR in
Environmental Research, 20(3–4), 207–220.
van Wezel A. P. and Opperhuizen A. (1995). Narcosis due to environmental pollutants in
aquatic organisms: Residue-based toxicity, mechanisms, and membrane burdens.
Critical Reviews in Toxicology, 25, 255–279.
Van Zijl M. C., Aneck-Hahn N. H., Swart P., Hayward S., Genthe B. and De Jager C. (2017).
Estrogenic activity, chemical levels and health risk assessment of municipal distribution
point water from Pretoria and Cape Town, South Africa. Chemosphere, 186, 305–313.
Vankreijl C. F., Kool H. J., Devries M., Vankranen H. J. and Degreef E. (1980). Mutagenic
activity in the rivers Rhine and Meuse in the Netherlands. Science of the Total
Environment, 15(2), 137–147.
Vergauwen L., Schmidt S. N., Stinckens E., Maho W., Blust R., Mayer P., Covaci A. and
Knapen D. (2015). A high throughput passive dosing format for the fish embryo
acute toxicity test. Chemosphere, 139, 9–17.
Verheijen M., Lienhard M., Schrooders Y., Clayton O., Nudischer R., Boerno S.,
Timmermann B., Selevsek N., Schlapbach R., Gmuender H., Gotta S., Geraedts J.,
Herwig R., Kleinjans J. and Caiment F. (2019). DMSO induces drastic changes in
human cellular processes and epigenetic landscape in vitro. Scientific Reports, 9, 4641.
Vermeirssen E. L. M., Korner O., Schonenberger R., Suter M. J. F. and Burkhardt-Holm P.
(2005). Characterization of environmental estrogens in river water using a three pronged
approach: Active and passive water sampling and the analysis of accumulated estrogens
in the bile of caged fish. Environmental Science & Technology, 39(21), 8191–8198.
Verschaeve L., Van Gompel J., Thilemans L., Regniers L., Vanparys P. and van der Lelie D.
(1999). VITOTOX (R) bacterial genotoxicity and toxicity test for the rapid screening of
chemicals. Environmental and Molecular Mutagenesis, 33(3), 240–248.
Vethaak A. D., Hamers T., Martinez-Gomez C., Kamstra J. H., de Weert J., Leonards P. E. G.
and Smedes F. (2017). Toxicity profiling of marine surface sediments: A case study
using rapid screening bioassays of exhaustive total extracts, elutriates and passive
sampler extracts. Marine Environmental Research, 124, 81–91.
Villeneuve D. L., Blankenship A. L. and Giesy J. P. (2000). Derivation and application of
relative potency estimates based on in vitro bioassay results. Environmental
Toxicology and Chemistry, 19(11), 2835–2843.
Villeneuve D. L., Coady K., Escher B. I., Mihaich E., Murphy C., Schlekat T. and
Garcia-Reyero N. (2019). High throughput screening and environmental risk
assessment – state of the science and emerging applications. Environmental
Toxicology and Chemistry, 38(1), 12–26.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 434 Bioanalytical Tools in Water Quality Assessment

Vinggaard A. M., Bonefeld-Jorgensen E. C., Jensen T. K., Fernandez M. F., Rosenmai A. K.,
Taxvig C., Rodriguez-Carrillo A., Wielsoe M., Long M., Olea N., Antignac J.-P.,
Hamers T. and Lamoree M. (2020). Receptor-based in vitro activities to assess human
exposure to chemical mixtures and related health impacts. Environment International,
146, 106191–106191.
Vinken M. and Blaauboer B. J. (2017). In vitro testing of basal cytotoxicity: Establishment of
an adverse outcome pathway from chemical insult to cell death. Toxicology In Vitro, 39,
104–110.
Vogs C. and Altenburger R. (2016). Time-dependent effects in algae for chemicals with
different adverse outcome pathways: A novel approach. Environmental Science &
Technology, 50(14), 7770–7780.
Volker J., Vogt T., Castronovo S., Wick A., Ternes T. A., Joss A., Oehlmann J. and Wagner
M. (2017). Extended anaerobic conditions in the biological wastewater treatment:
Higher reduction of toxicity compared to target organic micropollutants. Water
Research, 116, 220–230.
Volker J., Stapf M., Miehe U. and Wagner M. (2019). Systematic review of toxicity removal
by advanced wastewater treatment technologies via ozonation and activated carbon.
Environmental Science & Technology, 53(13), 7215–7233.
von der Ohe P. C., Dulio V., Slobodnik J., De Deckere E., Kuhne R., Ebert R. U., Ginebreda
A., De Cooman W., Schuurmann G. and Brack W. (2011). A new risk assessment
approach for the prioritization of 500 classical and emerging organic
microcontaminants as potential river basin specific pollutants under the European
Water Framework Directive. Science of the Total Environment, 409(11), 2064–2077.
Wagner M. and Oehlmann J. (2009). Endocrine disruptors in bottled mineral water: Total
estrogenic burden and migration from plastic bottles. Environmental Science and
Pollution Research, 16(3), 278–286.
Wagner E. D. and Plewa M. J. (2008). Chapter 3 Microplate-based comet assay. In: The
Comet Assay in Toxicology, The Royal Society of Chemistry, pp. 79–97.
Walker C. W. and Watson J. E. (2010). Adsorption of estrogens on laboratory materials and
filters during sample preparation. Journal of Environmental Quality, 39(2), 744–748.
Walker C. H., Hopkin S. P., Sibly R. M. and Peakall D. B. (2012). Principles of
Ecotoxicology. 4th edn. Taylor & Francis, Milton Park, UK.
Walter H., Consolaro F., Gramatica P., Scholze M. and Altenburger R. (2002). Mixture
toxicity of priority pollutants at no observed effect concentrations (NOECs).
Ecotoxicology, 11, 299–310.
Wambaugh J. F., Hughes M. F., Ring C. L., MacMillan D. K., Ford J., Fennell T. R., Black
S. R., Snyder R. W., Sipes N. S., Wetmore B. A., Westerhout J., Setzer R. W., Pearce
R. G., Simmons J. E. and Thomas R. S. (2018). Evaluating in vitro-in vivo
extrapolation of toxicokinetics. Toxicological Sciences, 163(1), 152–169.
Wang H. and Joseph J. A. (1999). Quantifying cellular oxidative stress by dichlorofluorescein
assay using microplate reader. Free Radical Biology and Medicine, 27(5–6), 612–616.
Wang W. L., Tada M., Nakajima D., Sakai M., Yoneda M. and Sone H. (2018).
Multiparameter phenotypic profiling in MCF-7 cells for assessing the toxicity and
estrogenic activity of whole environmental water. Environmental Science &
Technology, 52(16), 9277–9284.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 435

Wang Z., Walker G., Muir D. and Nagatani-Yoshida K. (2020). Toward a global
understanding of chemical pollution: A first comprehensive analysis of national and
regional chemical inventories. Environmental Science & Technology, 54(5),
2575–2584.
Wania F. (2003). Assessing the potential of persistent organic chemicals for long-range
transport and accumulation in polar regions. Environmental Science & Technology,
37(7), 1344–1351.
Warne M. S. J. and Hawker D. W. (1995). The number of components in a mixture determines
whether synergistic and antagonistic or additive toxicity predominate – the funnel
hypothesis. Ecotoxicology and Environmental Safety, 31(1), 23–28.
Wernersson A.-S., Carere M., Maggi C., Tusil P., Soldan P., James A., Sanchez W., Dulio V.,
Broeg K., Reifferscheid G., Buchinger S., Maas H., Van Der Grinten E., O’Toole S.,
Ausili A., Manfra L., Marziali L., Polesello S., Lacchetti I., Mancini L., Lilja K.,
Linderoth M., Lundeberg T., Fjallborg B., Porsbring T., Larsson D. G. J.,
Bengtsson-Palme J., Forlin L., Kienle C., Kunz P., Vermeirssen E., Werner I.,
Robinson C. D., Lyons B., Katsiadaki I., Whalley C., den Haan K., Messiaen M.,
Clayton H., Lettieri T., Carvalho R. N., Gawlik B. M., Hollert H., Di Paolo C., Brack
W., Kammann U. and Kase R. (2015). The European technical report on aquatic
effect-based monitoring tools under the water framework directive. Environmental
Sciences Europe, 27, 1–11.
Wetmore B. A. (2015). Quantitative in vitro-in vivo extrapolation in a high-throughput
environment. Toxicology, 332, 94–101.
White P. A., Rasmussen J. B. and Blaise C. (1996). A semi-automated, microplate version of
the SOS Chromotest for the analysis of complex environmental extracts. Mutation
Research – Environmental Mutagenesis and Related Subjects, 360(1), 51–74.
WHO (2017a). Chemical mixtures in source water and drinking-water. World Health
Organization, www.who.int/water_sanitation_health/publications/chemical-mixtures-
in-water/en/ (Accessed 14 February 2021).
WHO (2017b). Guidelines for Drinking-Water Quality. 4th edn, incorporating the 1st
addendum. Geneva, Switzerland. www.who.int/water_sanitation_health/publications/
2011/dwq_guidelines/en/index.html. (Accessed 18 February 2021).
WHO (2017c). Potable Reuse: Guidance for Producing Safe Drinking-Water. World
Health Organization. www.who.int/water_sanitation_health/publications/potable-
reuse-guidelines/en/ (Accessed 18 February 2021).
Widder M. W., Brennan L. M., Hanft E. A., Schrock M. E., James R. R. and van der Schalie
W. H. (2015). Evaluation and refinement of a field-portable drinking water toxicity
sensor utilizing electric cell-substrate impedance sensing and a fluidic biochip.
Journal of Applied Toxicology, 35(7), 701–708.
Wild C. P. (2005). Complementing the genome with an “exposome”: The outstanding
challenge of environmental exposure measurement in molecular epidemiology.
Cancer Epidemiology Biomarkers & Prevention, 14(8), 1847–1850.
Wilkinson M. D., Dumontier M., Aalbersberg I. J., Appleton G., Axton M., Baak A.,
Blomberg N., Boiten J. W., Santos L. B. D., Bourne P. E., Bouwman J., Brookes
A. J., Clark T., Crosas M., Dillo I., Dumon O., Edmunds S., Evelo C. T., Finkers R.,
Gonzalez-Beltran A., Gray A. J. G., Groth P., Goble C., Grethe J. S., Heringa J.,
Hoen P. A. C., Hooft R., Kuhn T., Kok R., Kok J., Lusher S. J., Martone M. E.,

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 436 Bioanalytical Tools in Water Quality Assessment

Mons A., Packer A. L., Persson B., Rocca-Serra P., Roos M., van Schaik R., Sansone
S. A., Schultes E., Sengstag T., Slater T., Strawn G., Swertz M. A., Thompson M.,
van der Lei J., van Mulligen E., Velterop J., Waagmeester A., Wittenburg P.,
Wolstencroft K., Zhao J. and Mons B. (2016). The FAIR Guiding Principles for
scientific data management and stewardship. Scientific Data, 6, 160018.
Wilson V. S., Bobseine K. and Gray L. E. , Jr. (2004). Development and characterization of a
cell line that stably expresses an estrogen-responsive luciferase reporter for the detection
of estrogen receptor agonist and antagonists. Toxicological Sciences, 81(1), 69–77.
Wölz J., Engwall M., Maletz S., Takner H. O., van Bavel B., Kammann U., Klempt M.,
Weber R., Braunbeck T. and Hollert H. (2008). Changes in toxicity and Ah receptor
agonist activity of suspended particulate matter during flood events at the rivers
Neckar and Rhine – a mass balance approach using in vitro methods and chemical
analysis. Environmental Science and Pollution Research, 15(7), 536–553.
Wood S. A., Holland P. T., Stirling D. J., Briggs L. R., Sprosen J., Ruck J. G. and Wear R. G.
(2006). Survey of cyanotoxins in New Zealand water bodies between 2001 and 2004.
New Zealand Journal of Marine and Freshwater Research, 40(4), 585–597.
Woutersen M., Belkin S., Brouwer B., van Wezel A. P. and Heringa M. B. (2011). Are
luminescent bacteria suitable for online detection and monitoring of toxic compounds in
drinking water and its sources? Analytical and Bioanalytical Chemistry, 400(4), 915–929.
Wu Q.-Y., Hu H.-Y., Zhao X. and Sun Y.-X. (2009). Effect of chlorination on the
estrogenic/antiestrogenic activities of biologically treated wastewater. Environmental
Science & Technology, 43(13), 4940–4945.
Xiao R., Wang Z., Wang C. and Yu G. (2006). Soil screening for identifying ecological risk
stressors using a battery of in vitro cell bioassays. Chemosphere, 64(1), 71–78.
Xiao S. H., Lv X. M., Lu Y., Yang X. M., Dong X. R., Ma K. P., Zeng Y. F., Jin T. and Tang
F. (2016). Occurrence and change of estrogenic activity in the process of drinking water
treatment and distribution. Environmental Science and Pollution Research, 23(17),
16977–16986.
Xiao S. H., Lv X. M., Zeng Y. F., Jin T., Luo L., Zhang B. B., Zhang G., Wang Y. H., Feng L.,
Zhu Y. and Tang F. (2017). Mutagenicity and estrogenicity of raw water and
drinking water in an industrialized city in the Yangtze River Delta. Chemosphere,
185, 647–655.
Xie S. H., Liu A. L., Chen Y. Y., Zhang L., Zhang H. J., Jin B. X., Lu W. H., Li X. Y. and Lu
W. Q. (2010). DNA damage and oxidative stress in human liver cell L-02 caused by
surface water extracts during drinking water treatment in a waterworks in China.
Environmental and Molecular Mutagenesis, 51(3), 229–235.
Yagi K. (1998). Simple assay for the level of total lipid peroxides in serum or plasma.
Methods in Molecular Biology, 108(1), 101–106.
Yang C., Zhou J. Y., Zhong H. J., Wang H. Y., Yan J., Liu Q., Huang S. N. and Jiang J. X.
(2011). Exogenous norepinephrine correlates with macrophage endoplasmic reticulum
stress response in association with XBP-1. Journal of Surgical Research, 168(2),
262–271.
Yeh R. Y. L., Farré M. J., Stalter D., Tang J. Y. M., Molendijk J. and Escher B. I. (2014).
Bioanalytical and chemical evaluation of disinfection by-products in swimming pool
water. Water Research, 59, 172–184.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 References 437

Yoo H., Khim J. S. and Giesy J. P. (2006). Receptor-mediated in vitro bioassay for
characterization of Ah-R-active compounds and activities in sediment from Korea.
Chemosphere, 62(8), 1261–1271.
Yoon M., Blaauboer B. J. and Clewell H. J. (2015). Quantitative in vitro to in vivo
extrapolation (QIVIVE): An essential element for in vitro-based risk assessment.
Toxicology, 332, 1–3.
You J. and Li H. Z. (2017). Improving the accuracy of effect-directed analysis: The role of
bioavailability. Environmental Science-Processes & Impacts, 19(12), 1484–1498.
Zani C., Feretti D., Buschini A., Poli P., Rossi C., Guzzella L., Caterino F. D. and Monarca S.
(2005). Toxicity and genotoxicity of surface water before and after various
potabilization steps. Mutation Research – Genetic Toxicology and Environmental
Mutagenesis, 587(1–2), 26–37.
Zegura B., Heath E., Cernosa A. and Filipic M. (2009). Combination of in vitro bioassays for
the determination of cytotoxic and genotoxic potential of wastewater, surface water and
drinking water samples. Chemosphere, 75(11), 1453–1460.
Zeng S. Y., Huang Y. Q., Sun F., Li D. and He M. (2016). Probabilistic ecological risk
assessment of effluent toxicity of a wastewater reclamation plant based on process
modeling. Water Research, 100, 367–376.
Zhang D. D. (2006). Mechanistic studies of the Nrf2-Keap1 signaling pathway. Drug
Metabolism Reviews, 38(4), 769–789.
Zhang J. H., Chung T. D. Y. and Oldenburg K. R. (1999). A simple statistical parameter for
use in evaluation and validation of high throughput screening assays. Journal of
Biomolecular Screening, 4(2), 67–73.
Zhang Z. H., Feng Y. J., Gao P., Wang C. and Ren N. Q. (2011). Occurrence and removal
efficiencies of eight EDCs and estrogenicity in a STP. Journal of Environmental
Monitoring, 13(5), 1366–1373.
Zhang H., Ihara M., Hanamoto S., Nakada N., Jurgens M. D., Johnson A. C. and Tanaka H.
(2018a). Quantification of pharmaceutical related biological activity in effluents from
wastewater treatment plants in UK and Japan. Environmental Science & Technology,
52(20), 11848–11856.
Zhang S. Y., Li S. Z., Zhou Z. G., Fu H. L., Xu L., Xie H. and Zhao B. (2018b). Development
and application of a novel bioassay system for dioxin determination and aryl
hydrocarbon receptor activation evaluation in ambient-air samples. Environmental
Science & Technology, 52(5), 2926–2933.
Zhang J. Y., Yang Y., Liu W. P., Schlenk D. and Liu J. (2019). Glucocorticoid and
mineralocorticoid receptors and corticosteroid homeostasis are potential targets for
endocrine-disrupting chemicals. Environment International, 133, 105133.
Zhao J. L., Ying G. G., Yang B., Liu S., Zhou L. J., Chen Z. F. and Lai H. J. (2011). Screening
of multiple hormonal activities in surface water and sediment from the Pearl River
system, South China, using effect-directed in vitro bioassays. Environmental
Toxicology and Chemistry, 30(10), 2208–2215.
Zheng F., Fu F., Cheng Y., Wang C., Zhao Y. and Gu Z. (2016). Organ-on-a-chip systems:
Microengineering to biomimic living systems. Small, 12(17), 2253–2282.
Zhong Y., Feng S. L., Luo Y., Zhang G. D. and Kong Z. M. (2001). Evaluating the
genotoxicity of surface water of Yangzhong City using the Vicia faba micronucleus

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 438 Bioanalytical Tools in Water Quality Assessment

test and the comet assay. Bulletin of Environmental Contamination and Toxicology, 67
(2), 217–224.
Zimmermann F. K., Kern R. and Rasenberger H. (1975). A yeast strain for simultaneous
detection of induced mitotic crossing over, mitotic gene conversion and reverse
mutation. Mutation Research – Fundamental and Molecular Mechanisms of
Mutagenesis, 28(3), 381–388.
Zimmermann L., Dierkes G., Ternes T. A., Volker C. and Wagner M. (2019). Benchmarking
the in vitro toxicity and chemical composition of plastic consumer products.
Environmental Science & Technology, 53(19), 11467–11477.
Zimmermann L., Dombrowski A., Volker C. and Wagner M. (2020). Are bioplastics and
plant-based materials safer than conventional plastics? In vitro toxicity and chemical
composition. Environment International, 145, 106066.
Zounkova R., Jalova V., Janisova M., Ocelka T., Jurcikova J., Halirova J., Giesy J. P. and
Hilscherova K. (2014). In situ effects of urban river pollution on the mudsnail
Potamopyrgus antipodarum as part of an integrated assessment. Aquatic Toxicology,
150, 83–92.
Zucker R. M., Elstein K. H., Easterling R. E. and Massaro E. J. (1988). Flow
cytometric analysis of the cellular toxicity of tributyltin. Toxicology Letters, 43(1–3),
201–218.
Zurita J., del Peso A., Rojas R., Maisanaba S. and Repetto G. (2019). Integration of fish cell
cultures in the toxicological assessment of effluents. Ecotoxicology and Environmental
Safety, 176, 309–320.
Zwart N., Nio S. L., Houtman C. J., de Boer J., Kool J., Hamers T. and Lamoree M. H. (2018).
High-throughput effect-directed analysis using downscaled in vitro reporter gene assays
to identify endocrine disruptors in surface water. Environmental Science & Technology,
52(7), 4367–4377.
Zwart N., Jonker W., ten Broek R., de Boer J., Somsen G., Kool J., Hamers T., Houtman C. J.
and Lamoree M. H. (2020). Identification of mutagenic and endocrine disrupting
compounds in surface water and wastewater treatment plant effluents using
high-resolution effect-directed analysis. Water Research, 168, 15204.

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Index

Note: t = Table, f = Figure

A ADI. See Acceptable (or allowable) daily

AA-EQS (annual average environmental intake
quality standard), 41, 284, 285, 355 Advanced water treatment plant (AWTP),
ABC transporters, 53, 355. See also Active 115f, 357
transport Adverse outcome pathway (AOP), 56, 56f,
Abiotic, 4, 5t. See also Biotic 58, 61, 73, 75f, 91–99, 94f–98f, 125,
Absorption, 51, 53f, 73, 74, 75, 77, 81, 158 125f, 126, 152, 154, 268f, 340, 345, 356.
ACC (activity concentration at cutoff), 108, See also Toxicity pathway
109f, 150, 319, 355 ADWG (Australian drinking water
Acceptable (or allowable) daily intake (ADI), guideline), 39, 285, 287, 307, 356
29, 37, 40, 281, 284, 341, 355, 356 Aflatoxin, 78, 356
Accuracy, 46, 225–227, 227f, 243 Agonist, 57, 63–64, 65f, 89, 112–113, 112f,
Acetylcholinesterase (AChE), 60t, 62–63, 124, 195, 242, 276, 277f, 304t, 306f,
86, 96–97, 96f, 118, 124, 200–201, 201t, 309t, 311t, 312t, 335, 356. See also
229, 283t, 309t, 311t, 312t, 355, 358, 368 Antagonist
AChE. See Acetylcholinesterase AGWR (Australian guidelines for water
ACR. See acute-to-chronic ratio recycling), 39, 40, 285, 287, 351, 356
Active transport, 52–53, 61, 74, 77, 79, 158, AhR. See Aryl hydrocarbon receptor
171, 355, 369. See also ABC AhR CAFLUX, 172–173, 174t, 176t, 328,
transporters, Passive transport 334–335, 356
Acute exposure, 355, 359. See also Chronic Algae, 2, 12, 30, 44, 44f, 45, 47, 48, 59,
exposure 92–96, 94f, 95f, 121, 127t, 128t, 136,
Acute toxicity, 6, 15, 30, 41, 44–48, 92, 215, 137f, 138, 141, 199, 200t, 221, 222t, 275,
268, 349 279, 279f, 284, 289t, 290, 317f, 349
Acute-to-chronic ratio (ACR), 291, 294 Aliivibrio fischeri, 14, 106f, 107f, 115f, 121,
Adaptive stress response pathway, 67–70, 127t, 136, 137f, 208t, 218, 242, 248, 287,
68f, 212, 356, 369, 376 328, 356. See also Vibrio fischeri
Adenosine-5’-triphosphate (ATP), 52, 60t, American Society for Testing and Materials
61–63, 77, 79–80, 95, 97, 355, 357 (ASTM), 45

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 440 Bioanalytical Tools in Water Quality Assessment

Ames test, 14–15, 202, 208t, 242, 271, 356 ASTM. See American Society for Testing
Anaemia, 83, 356 and Materials
Androgen, 14, 78, 87–89, 118, 127t, 179, ATP. See Adenosine-5’-triphosphate
182–188, 185t, 186t, 187t, 249, 268, Attagene, 150, 154, 173, 269, 270f, 318,
275–277, 277f, 282t, 306, 309t–313t, 319f, 320, 320f
335, 356 Australian and New Zealand Environment
Androgen receptor (AR), 89, 98, 127t, 179, and Conservation Council
182–188, 306, 311t–313t, 316, 335, 357 (ANZECC), 42
Annual average EQS (AA-EQS), 41. Australian Drinking Water Guidelines
See also EQS (ADWG), 39, 285, 287, 357
Antagonism, 64, 112, 113, 120, 120t, Australian Guidelines for Water Recycling
123–125, 127t, 182, 186, 190, 192, (AGWR), 39–40, 285, 287, 357
195, 228 Autoimmune disease, 85
Antagonist, 57, 63–64, 65f, 89, 98, AWTP. See Advanced water treatment plant
112–113, 113f, 123–125, 194–195, 242, Axon, 86, 357
276–278, 277f, 304t, 311t, 312t, 335,
356. See also Agonist
Antibody, 84, 356 B
Antigen, 83–84, 356. See also Antibody Basal toxicity, 59–60, 61f, 357. See also
Antimitotic drug, 83, 356 Baseline toxicity
Antioxidant response element (ARE), 32, Baseline toxicity, 13, 59, 60t, 61, 77, 94–97,
69, 136, 137, 137f, 212, 213t, 214t, 238t, 135, 137, 163–168, 218–219, 278f,
240, 241f, 259f, 274f, 275, 288t, 293f, 290, 357
313t, 320, 321, 322f, 327f, 328, 329t, BEC. See Biologically effective
330t, 332f, 333, 334, 357 concentration
AOC. See Assimilable organic carbon BED. See Biologically effective dose
AOP. See Adverse Outcome Pathway Benchmark dose (BMD), 28, 101, 103,
Apical endpoints, 20, 92–93, 139, 267, 290, 103f, 357
299, 356 Benchmark response (BMR), 357
Apoptosis, 58, 61, 65, 69–71, 77, 82, 166, BEQ. See Bioanalytical equivalent
202, 213, 356 concentration
AR. See Androgen receptor BEQbio, 117–118, 258, 260, 272–275,
ARE. See Antioxidant response element 290, 295–296, 316–317, 321–323,
AREc32, 212, 213t, 214t, 274f, 275, 288, 329–330
293f BEQchem, 258–260, 259f, 272–275, 274f,
Aromatase, 17, 63–64, 154–155, 357 290, 295–296, 317, 321, 322f–323f,
Aryl hydrocarbon receptor (AhR), 13, 17, 333, 335
54–55, 54f, 55t, 60t, 129, 137, 172–175, Bioaccumulation, 28, 75, 91, 357
174t, 176t, 222, 236, 251, 254, 269–271, Bioactivation, 63, 178, 357
274f, 275, 278–280, 279f, 288t, 289t, Bioanalytical effect flux (BEF),
293f, 296–297, 304t, 312, 313t, 327, 327f
314–318, 317f, 320–324, 322f, 323f, Bioanalytical equivalent concentration
327–328, 329t, 330t, 332f, 333–336, 356 (BEQ), 102, 117–118, 137, 137f, 208,
Assay, 109–110, 114–117, 126–127, 136, 230, 236, 240, 258, 258f, 259f, 272,
144–250, 357 274f, 284, 285, 290, 291, 294, 316, 317f,
Assimilable organic carbon (AOC), 341, 321, 322, 322f, 327f, 328, 329t, 330t,
341f, 356 333, 357

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Index 441

Bioanalytical tool, 1–2, 6, 7f, 9–14, 15f, Carcinogenesis, 29, 53, 81, 203, 358
18–22, 21t, 23, 33, 47–48, 51, 99, 120, Carcinogenicity, 14, 28, 55, 73, 77, 81–82,
126, 142, 169–223, 225–226, 235, 245, 173, 203, 358. See also Epigenetic
271, 272f, 299, 307–308, 324, 325–337, carcinogen, genotoxicity
339–353, 357 Carcinoma, 163, 220t, 358. See also Tumour
Bioavailability, 145, 284, 328, 331f, 332f, Cardiomyocyte, 80, 358
335, 343, 357 Cardiovascular toxicity, 77, 79–81, 358
Bioconcentration, 28, 154, 155f, 357. Catabolism, 77, 89, 358
See also Bioaccumulation Category 1 bioassay, 266, 267, 274f, 275,
Biodegradation, 4, 5t, 21, 249, 358 278–286, 278f, 279f, 281f, 282t, 283t,
Biologically effective concentration (BEC), 287f, 290, 294–296, 344
51, 94, 105, 158, 358 Category 2 bioassay, 267, 274f, 275,
Biologically effective dose (BED), 51, 94, 278–280, 279f, 285, 286–291, 287f,
105, 158, 159t, 168, 357, 358 288t–289t, 293f, 294–296, 321, 345
Biological pathway altering concentration Chemical-group motivated approach, 359.
(BPAC), 149, 358 See also Protection-goal motivated
Biomarker, 9–10, 33, 46–47, 149, 170–172, approach
268, 358 Chronic exposure, 78, 359
Biota, 7, 41, 52, 98, 239, 245, 325–337 Chronic toxicity, 41, 44, 45
Biotic, 4, 5t, 358. See also Abiotic Clean Water Act (CWA), 32, 40, 41, 359
Biotransformation, 4, 53–54, 75, 78, 173, Cmax (maximum plasma concentration),
249, 358 151, 151f, 153
Blank, 114, 115, 231, 236–239, 245, 246, Coefficient of variation for repeatability
254, 256, 257, 264 (CVr), 227–228, 359
BMD. See Benchmark dose Coefficient of variation for reproducibility
BMDL (benchmark dose level), 103 (CVR), 227–228
BMR. See Benchmark response Combined algae test (CAT), 135, 199, 200t,
BPAC. See Biological pathway altering 221, 222t, 248, 283t
concentration Comet assay, 14, 202–203, 204t, 206, 346,
BPADL (lower limit of biological pathways 359. See also Single Cell Gel
altering dose), 149 Electrophoresis
BTEX (benzene, toluene, ethylbenzene, Complex mixture, 21t, 43, 49, 102, 119,
xylene), 3t, 358 120, 138, 142, 252, 265, 266, 270f, 275,
326, 336, 340, 346
C CompTox Chemistry Dashboard, 147, 167,
CA. See Concentration addition 101, 359
CA/DA. See Concentration addition/dose Concentration addition (CA), 119–133,
addition 136–140, 321f, 358, 359
CALUX, 108f, 108, 137, 173, 174f, Concentration addition/dose addition
176–199, 212, 213, 213f, 215, 215f, (CA/DA), 120–133, 127t, 135, 358
216f, 216f, 236, 236f, 238t, 248, 257, Concentration-effect curve, 45, 110–111,
274f, 275, 277f, 278, 282t, 283f, 284, 116f, 229, 306f
288, 291, 294, 308t, 313t, 320, 321, Concentration–response curve (CRC), 101,
322t, 323f, 329t, 330t, 333 105–108, 106f, 108f, 110–112, 111f,
CAR. See Constitutive androstane receptor 112f, 113–116, 116f, 120, 139, 174t,
Carbamate pesticides, 86, 358. See also 181t, 193t, 196t, 229f, 230f, 231f, 235,
Acetylcholinesterase 236, 272, 276, 306, 359

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 442 Bioanalytical Tools in Water Quality Assessment

Confidence interval, 163, 359 Dissolved organic carbon (DOC), 201, 228,
Conjugate, 53, 75, 359 252, 360
Conjugation, 75, 89, 359 Dithiocarbamate pesticides, 66, 360
Constitutive androstane receptor (CAR), Diuron, 63, 95, 118, 135–137, 199–200,
55t, 173, 346, 358 200t, 283, 289t, 317, 322, 360. See also
Control chart, 232, 236, 240–242, 241f Photosystem II
CRC. See Concentration−response curve DMSO. See Dimethyl sulfoxide
CVr. See Coefficient of variation for DNA, 13, 14, 23, 53–55, 57–59, 59t, 60t,
repeatability 64–68, 66f, 69t, 70, 81, 82, 110, 157,
CVR. See Coefficient of variation for 166, 173, 202, 203, 204t–205t, 206, 210,
reproducibility 213, 219, 221t, 238, 310t, 346, 360
CWA. See Clean Water Act DOC. See Dissolved organic carbon
Cyanotoxin, 202, 359 Dose, 101, 360
Cytochrome (CYP), 17, 53, 55t, 67, 69, 75, Dose addition (DA), 119, 120–133, 120t,
77, 124, 154, 173, 359 127t, 135, 358
Cytokine, 69, 78, 80, 84, 310t, 359 Dose-metric, 101, 105, 158–161, 159t
Cytoreductive drug, 83, 359 Dose−response assessment, 28, 101–118,
Cytotoxicity, 9, 12, 14, 17, 18, 20, 23, 52, 360. See also Dose–response curve
58, 61, 68, 77, 78, 101, 102, 104, Dose−response curve, 101, 102–103, 105,
106–111, 109f, 111f, 113, 127t, 130, 360
137–139, 152, 165, 165f, 166, 170, 171, Dosing factor (DF), 47, 114, 248, 262, 360
182, 190, 193t, 194, 204t, 206, 208, 213, Drinking water, 4, 5, 11, 15–17, 20, 22, 35,
215–221, 220t, 221t, 230, 231, 236, 238t, 36, 37f, 38–40, 42, 51, 62, 73, 74, 76f,
239, 243, 245, 256, 257, 266–268, 77, 83, 110, 114, 133, 136, 169, 173,
271–273, 276, 278, 287, 287f, 290, 292f, 175, 176t, 177, 177t, 178, 179t, 180, 182,
296, 299, 306, 306f, 310t–312t, 313, 184t, 185, 185t, 186, 187t, 189t, 190,
316, 334, 335, 340, 350, 359. See also 191t–194t, 195, 197t–201t, 198–203,
Basal toxicity 206–212, 207t, 209t, 214t, 215, 216t,
217t, 217–219, 218t, 221–223, 222t,
D 239, 240, 245, 247, 247f, 248, 251, 253,
DA (dose addition), 119, 120t, 122–126, 255, 256, 262, 267–269, 271, 280, 281,
123f, 127t, 129–131, 133, 135, 358, 360 281f, 284–286, 286f, 287f, 289–290,
Daphnia magna, 22, 45, 95, 96f, 97, 127t, 296, 297, 299, 304–308, 341–343, 349,
138, 139, 290 351, 352
DBP. See Disinfection by-product Drinking water equivalent level (DWEL),
Denaturing, 68 38, 360
DF (dosing factor), 47, 114, 248, Drinking water treatment plant (DWTP),
262, 360 115f, 206, 269, 305–306, 360
Dimethyl sulfoxide (DMSO), 229, 237–238, DTA. See Direct toxicity assessment
238t, 242, 257, 262, 264, 343–344, 360 DWEL. See Drinking water equivalent level
Dioxin, 55t, 129, 360 DWTP. See Drinking water treatment plant
Direct toxicity assessment (DTA), 42, 215,
340, 360. See also WET
Disinfection by-product (DBP), 4–6, 62, E
202, 203, 204t–205t, 206, 207, 208, 208t, 17α-Ethinylestradio (EE2), 3t, 182f, 268,
210, 212, 218, 249, 255, 266, 299, 304, 273, 275, 278, 284, 285, 362
305, 307, 308, 311t, 343, 360 EBT. See Effect-based trigger value

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Index 443

EC10, 108, 108f, 110, 111, 111f, 118, 165, Enzyme-linked immunosorbent assay
173–175, 174t, 177, 177t, 178, 178t, (ELISA), 310t, 361. See also
179t, 180, 181t, 184t, 186, 186t, 187t, Immunoassay
188, 190, 190t, 191t, 193t, 196t, 197t, EQS (environmental quality standard), 22,
200, 201t, 215t, 222t, 229, 230, 230f, 36, 43f, 284, 341, 361
291, 306, 307f ERA. See Ecological or environmental risk
EC50, 45, 48, 108, 108f, 111, 115, 122, assessment
125, 126, 127t, 140, 174, 181t, 195, ER (estrogen receptor), 17, 57, 89, 98, 99,
196t, 198t, 199, 201, 210, 218, 218t, 112, 112f, 118, 124, 150, 179–182, 242,
222, 230, 236, 240–241, 266, 278, 306, 313t, 335, 361
289–291, 360 ER-CALUX, 108, 108f, 179–180, 181t,
ECIR1.5, 111, 111f, 165, 203, 207t, 212, 213, 182, 183t–185t, 362
213t–215t, 215, 216t, 217t, 229, 240, EROD. See Ethoxyresorufin-O-deethylase
241f, 290, 307, 307f E-SCREEN, 128t, 181t, 182f, 183t–184t,
Ecological risk assessment (ERA), 8–9, 282t, 284, 362
8f, 26, 29f, 30, 43, 56, 125, 149f, Estriol (E3), 266, 284, 362
154–157, 155f, 360, 361, 362. See also Ethoxyresorufin-O-deethylase (EROD),
Human health risk assessment 174, 175, 251, 335
ECSPR20, 112, 113, 182, 185t, 186, 188t, EU (effect unit), 117–118, 122, 239, 273,
189t, 190, 192, 192t, 194t 361, 362
ECVAM, 145, 360 Eukaryote, 171, 362
EDA. See Effect-directed analysis Exposure activity ratio (EAR), 305,
EDC. See Endocrine disrupting compound 318, 320f
EE2. See 17α-Ethinylestradiol Extraction factor (EF), 114, 361, 362
EF. See Extraction factor or enrichment
factor F
Effect-based trigger value (EBT), 166, 246, FCMN (flow cytometry micronucleus),
267, 278f, 280–296, 281f, 282t–283t, 205t, 310t, 362
287f, 288t–289t, 292f, 293f, 304, FET. See Fish embryo toxicity
341–342, 360, 361 Filtration, 47, 48, 212, 242, 249–252, 260,
Effect concentration (EC), 30, 31f, 45, 305, 314, 327
61, 102, 111f, 112, 117, 154, 165, Fish, 1, 9, 10, 12, 15, 16, 27, 30, 41, 44–48,
165f, 178, 182, 208, 209t, 229, 272, 44f, 64, 92, 93, 97–99, 97f, 98f, 126,
278f, 290, 291, 307f, 315, 315f, 127t, 132, 135, 139–141, 154, 155f, 171,
343, 360 175, 204t, 219–221, 221t, 266–268, 326
Effect-directed analysis (EDA), 20, 138, Fish embryo, 2, 9, 45, 48, 97, 160,
267, 275–278, 276f, 277f, 294, 161, 162
326–328, 335–337, 341, 344, 345, Fish embryo toxicity (FET), 10, 45, 47–48,
360, 361 139, 160f, 162, 215, 222, 242, 268, 271,
Effect unit (EU), 117–118, 122, 239, 273, 289t, 290, 318, 350, 362
361, 362 Flow cytometry, 205t, 206, 221t, 310t
Enrichment factor (EF), 114, 215, 248, Food chain, 44, 44f, 92, 93, 325
263f, 361 Frameshift mutation (DNA/RNA), 208, 362
Environmental risk assessment, 8–9, 8f, Freely dissolved concentration, 151, 154,
26, 43, 56, 125, 149f, 154–157, 155f, 158, 159, 159t, 161, 167, 167t, 168
361, 362 Furans (PCDF), 129, 370

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 444 Bioanalytical Tools in Water Quality Assessment

G Haloacetic acids (HAAs), 5t, 6, 62, 204t,

β-Galactosidase, 13, 172, 204t, 363 205t, 208t, 363. See also Disinfection
GAC (Granular activated carbon), 305, 363 by-product
Gastrointestinal tract (GI tract), 75, Hazard quotient (HQ), 30, 131, 364. See also
76f, 77 Risk quotient
Gene activation, 57, 61, 139, 299, 363 Hazard assessment, 9, 11, 27, 151, 364.
GeneBLAzer, 137, 154, 177, 177t, 179, See also Risk assessment
180, 181t, 182, 183t–194t, 188, 190, Hazard identification, 8, 11, 17, 26, 27–28,
192, 194, 195, 196t–198t, 198, 199, 33, 294, 351, 364. See also Risk
212, 213, 213t–217t 215, 228, 230, assessment
230f, 231f, 240, 241f, 259f, 274f, Hazard, 4, 25, 27–28, 30, 132, 133, 145,
275, 282t, 288t, 293f, 306f, 313t, 152f, 308, 327, 333, 364
320, 321, 322f, 329t–331t, 333 Hepatocytes, 54, 77, 78, 170, 210t, 211t
Genetic polymorphism, 57, 363 Hepatotoxicity, 77–78, 78f, 309t, 311t,
Genetically modified cell, 1, 10, 312t, 364
13, 363 Herbicide, 17, 18, 59, 63, 93, 95, 95f, 124,
Genomics, 347, 363 127t, 135, 136f, 138, 199, 200, 266, 290,
Genotoxicity, 6, 14, 22, 60t, 65, 126, 153, 317, 318, 322
166, 202–207, 204t–205t, 207t, 210, HHRA. See Human health risk assessment
212, 265, 266, 271, 294, 296, 299, High throughput, 1, 2, 143, 148, 149f, 150f,
310t–312t, 333, 346, 363 151, 153, 169, 225, 276, 346, 347,
GFP. See Green fluorescent protein 349, 364
GHS (Globally Harmonised System), 8, 27, HLB (hydrophilic-lipophilic balance), 253,
133, 363 255–257, 261f, 343, 364
Glial cells, 86, 87, 201, 363. See also Myelin Homeostasis, 55, 56, 67, 71, 78, 80, 83, 87,
sheet, Neuron 88, 179, 212, 364
Glucocorticoid receptor (GR), 179, HTS (high throughput screening), 10, 108,
188–190, 190t, 191t, 192t, 276, 306, 147, 149f, 155f, 169, 203, 243, 343, 364
313, 313t, 335 HTTK (high-throughput toxicokinetics),
Glutathione (GSH), 60t, 66, 67, 69, 70, 210, 151, 153, 364
210t, 211, 211t, 363 HTTr (high-throughput transcriptomics),
GR. See Glucocorticoid receptor 147, 152, 153, 168, 364
Granulosa cells, 363 Human Health Risk Assessment (HHRA),
Green fluorescent protein (GFP), 205t, 206, 26, 28–31, 29f, 131, 144, 146, 146f,
363. See also Recombinant cell 148–149, 154, 157, 215, 364. See also
Grey water, 39, 363 Ecological risk assessment
Guideline, 36–40, 42, 43, 45, 49, 221, Hydrophilicity, 364
242–243, 285, 287, 294, 296, 344, Hydrophobicity, 160, 161, 164, 165f, 254,
345, 350 326, 364
Guideline value, 22, 35–37, 37f, 40, 42, 48, Hyperplasia, 78, 88, 364
267, 281, 281f, 284–287, 289, 291, 294, Hyperthermia, 68, 364
296, 350, 363 Hypoxia, 69t, 83, 87, 364
GV. See Guideline value Hypoxia-inducible factor (HIF), 173

H
Haematopoiesis, 83, 363 I
Haematotoxicity, 83, 84f, 363 IA. See Independent action

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Index 445

ICATM (International Cooperation on IPCS (International Programme on

Alternative Test Methods), 145, 364 Chemical Safety), 130, 130f, 131,
ICCVAM (US Interagency Coordinating 133, 365
Committee on the Validation of ISO, 45, 47, 242, 282t, 365
Alternative Methods), 145, 364 ITS, 143–146, 308, 365>
Iceberg modelling, 147, 154, 266, 271–278,
273f, 280, 290, 296, 304, 305, 314, 317, K
321f, 324, 337, 341, 345 Key event (KE), 55–56, 58, 61, 92, 101, 120,
Immortalised cell line, 170, 364. See also 147, 345, 365
Primary cell line
Immunoassays, 2, 9 L
Immunotoxicity, 83, 84–85, 310t, 311f, Lactate dehydrogenase (LDH), 219, 220t,
312t, 335, 364 365
Independent Action (IA), 119, 120, LC50 (lethal concentration for 50% of the
121–122, 124, 126, 128t, 130f, 136, 139, test species), 30, 45, 46, 122, 140
344, 365. See also Concentration LC-MS/MS (liquid chromatography with
addition tandem mass spectrometry), 350, 365
Inflammation, 69t, 84, 87, 212, 213, 334 LD50, 101, 103f, 125, 365
Insecticide, 46, 63, 93, 96–97, 124, 138, Legacy, 147, 327
201, 290 Leydig cells, 365
Integrated (or intelligent) testing strategy Ligand, 54, 56, 175, 178, 280, 365. See also
(ITS), 143, 145–146, 365 Receptor
Intercalation (DNA), 57, 365 Limit of detection (LOD), 21t, 110, 119,
Internal concentration, 95, 221, 284 177, 180, 184t, 195, 197t, 202, 229, 365
International Cooperation on Alternative Limit of quantification (LOQ), 22, 229, 365
Test Methods (ICATM), 145, 364 Lipid peroxidation, 66, 67, 211, 211t, 365
International Organization for Liquid-liquid extraction (LLE), 47, 114,
Standardization (ISO), 45, 47, 242, 169, 245, 252f, 254–255, 266, 343, 365.
282t, 365 See also Solid phase extraction
In silico, 144, 146, 151, 168, 365 LLE. See Liquid-liquid extraction
Invertebrate, 1, 44, 45, 63, 92, 93, 95–97, LOAEL (lowest observed adverse effect
162, 268, 289t level), 103, 103f, 107, 365
In vitro, 1, 9, 10, 11, 15, 33, 45, 47, , 52, 56, LOEC. See Lowest observed effect
57f, 74f, 90, 92, 98–99, 101, 104, 105, concentration
113, 114, 116f, 117, 120, 122, 123, 126, Lowest Observed Effect Concentration
129, 138, 143–168, 69–71, 199t, 200, (LOEC), 107, 107f, 108, 212, 291, 365
201, 202, 204t, 205, 208t, 210t, 211t, Luciferase (Luc), 13, 108, 172, 241,
221t, 225, 226, 232, 242, 252, 267, 278f, 309–310t, 350, 365
280, 284, 285, 291, 303t, 304t, 305f,
322f, 326, 328, 334, 348, 365 M
In vivo, 1, 9–12, 33, 37f, 45, 48, 49, MAC-EQS, 41, 366. See also EQS
52, 74f, 104, 116, 123, 126, 129, Macrocosm, 46, 92. See also Mesocosm
146, 150f, 151f, 155f, 203, 233f, Macropollutant, 3, 340, 366. See also
284, 285, 289, 291–4, 303–303t, Micropollutant
333, 342, 365 MAF (mixture assessment factor), 134, 135,
Ionophoric shuttle mechanism, 60t, 365 366, 367

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 446 Bioanalytical Tools in Water Quality Assessment

Margin of exposure (MOE), 31, 366. Monitoring, 1, 18–22, 90, 154, 169, 268f,
See also Margin of safety 269–271, 348–349
Margin of safety (MOS), 31, 31f, 40, 153, Morphogenesis, 83, 367
366 Multiplex, 152, 154, 159, 269, 270, 296,
Matrix, 47, 114, 169, 176t, 177t, 179t, 183t, 304, 349–350
184t, 185t, 187t, 189t, 192t, 193t, 194t, Multi-substance potentially affected fraction
197t, 198t, 199t, 200t, 201t, 207t, 209t, (msPAF), 141, 367
214t, 216t, 217t, 218t, 222t, 228–229, Mutagenicity, 14, 15, 28, 207–209, 242,
252f, 253, 325, 342–343, 366 271, 276, 296, 306, 346, 367. See also
Matrix effects, 325, 342–343, 366 Genotoxicity
Maximum cululative ratio (MCR), 131–132, Mutation, 65, 81, 367
134, 366 Myelinating cells, 86, 367. See also Myelin
MCF-7, 181t, 213t, 220t, 350, 366 sheet
MCR. See Maximum cumulative ratio Myelin sheet, 87, 367
Mechanism of toxicity, 125, 148f, 366
Membrane filtration (MF), 212, 366
Mesocosm, 46, 92 N
Metabolic activation, 5t, 82, 96, 97, 124, NADPH (reduced nicotinamide adenine
158, 203, 206, 207t, 213, 306, 336, 366 dinucleotide phosphate), 62, 66, 67f, 69,
Metabolic pathway, 54, 366 173, 367
Metabolism precursor, 5t, 53, 78, 153, 175, NADPH-quinoneoxidoreductase (NQO1),
218, 258. See also Metabolic activation, 69, 173
Phase I and II metabolism NAM. See New approach method
Metabolite, 4, 48, 53, 62, 70, 78, 89, 90, 158, Narcosis (mode of action), 59, 61, 290, 367.
163, 336, 347, 366 See also Baseline toxicity
Metabolomics, 11, 347, 366 National Center for Advancing
Microcystin, 78, 366 Translational Sciences (NCATS), 147,
Micropollutant, 42, 64, 66, 70, 110, 139, 148f, 367, 375
169, 170, 215, 246, 315f, 324, 366. National Center for Computational
See also Macropollutant Toxicology (NCCT), 147, 367, 375
Microtiter plate, 9, 105, 366 National Institute of Environmental Health
Microtox, 14, 47, 115, 127t, 218, 271, 289t, Science (NIEHS), 147, 368
366 National Pollutant Discharge Elimination
MIE. See Molecular initiating event System (NPDES), (USA), 41, 42, 368
Mineralocorticoid receptor (MR), 79, National Research Council (NRC), 26,
195–197, 304t 143, 147
Mixture factor (MF), 283t, 287, 287f, 289t National Water Quality Management
Mixture assessment factor (MAF), 134, 135, Strategy (NWQMS), (Australia), 39,
366, 367 42, 368
Mode of (toxic) action (MOA), 12, 20, Native cell, 12–13, 23, 171, 367. See also
58–67, 93, 120, 125, 127t, 133, 172, 284, Primary cell lines, Immortalised cell lines
309t, 310t, 313t,340, 367 Necrosis, 58, 61, 77, 367
Molecular initiating event (MIE), 55, 58, Negative control, 103, 110–111, 228–230,
59t, 95, 96, 97, 101, 120, 126, 147, 150f, 231f, 232, 236, 237, 290, 367
152f, 154, 156f, 157, 165, 340, 345, Neonicotinoid, 63, 367
351f, 367 Nephelometric turbidity units (NTU), 251

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Index 447

Nephrotoxicity, 78–79, 79f, 367 Organogenesis, 82, 368

Neuron, 80, 85–87, 367. See also Glial cell, Organophosphate pesticide, 63, 368.
Neurotransmitter, Axon See also Acetylcholinesterase
Neuronopathy, 86, 368 Oxidation, 4, 5t, 21, 53, 178, 211, 238
Neurotoxicity, 60t, 63, 81, 85–87, 86f, Oxidative stress, 14, 57, 66–67, 69t, 136,
200–202, 309t, 311t, 312t, 346, 368 137f, 203, 211–212, 333, 334, 369.
Neurotransmitter, 63, 86, 96, 201, 202, 368 See also ROS
Neutral red uptake (NRU), 219, 220t, 221t, Ozone (O3), 22, 368
368
New approach method (NAM), 9, 143, P
144–148, 154–157, 368 p53, 69t, 70, 166, 213, 215t, 216t, 306, 316,
Next-generation risk assessment (NGRA), 369. See also Genotoxicity, Tumour
148, 151–153, 368 suppressor gene
Nicotinamide adenine dinucleotide Partition constant, 159, 160, 161, 162f, 165f,
phosphate (NADPH), 62, 66, 67f, 69, 167, 237, 332f, 343
173, 367 Passive dosing, 162, 163, 343, 369
Nitrosamines, 6, 208, 368. See also Passive sampling, 169, 252f, 254, 255, 294,
Disinfection by-product 326, 328, 335, 337, 369
Nominal concentration, 151, 153, 159t, 161, Passive transport, 369. See also Active
162, 258 transport
Non-specific (mode of action), 368. Pathogen, 4, 6, 35, 84, 308, 369
See also Non-specific toxicity, Pathway of toxicity, 369
Baseline toxicity PBTK (physiologically based toxicokinetic
Non-specific toxicity, 12–14, 58–62, 93, model), 153, 369
127t, 136, 137f, 218, 251, 294, PBT. See persistent, bioaccumulative and
296, 310t, 368. See also Baseline toxicity toxic
Non-threshold chemicals, 26, 368. See also PCB. See Polychlorinated biphenyls
Threshold chemicals PCDD (polychlorinated dibenzodioxin),
No Effect Level (NEL), 29t, 149 55t, 129, 334, 335, 370
No Observable Adverse Effect Level PCDF (polychlorinated dibenzofuran),
(NOAEL), 28, 37, 103, 121, 122, 368 129, 370
No Observable Effect Concentration PDMS. See polydimethylsiloxane
(NOEC), 28, 31, 37, 44, 45, 92, 101, 107, persistent, bioaccumulative and toxic (PCT),
121, 122, 291, 368 6, 8, 28, 369
No Observable Effect Level (NOEL), 368 Persistent organic pollutants (POP), 28, 133,
Nrf2, 69, 69t, 173, 212, 213t, 214t, 288t, 368 327, 334–336, 370. See also Stockholm
NRU. See Neutral red uptake Convention
Nuclear factor kappa B (NF-κB), 69 Pesticide, 5t, 20, 23, 39, 42, 62, 66, 85,
Nuclear receptor, 14, 54, 55t, 59, 68, 99, 124, 132, 136, 141, 179, 209, 248,
105, 147, 153, 166, 175, 178, 179, 269, 290, 308, 315
278, 368. See also Receptor Phagocytes, 369
Phase I metabolism, 53, 70, 78, 178, 369.
See also Metabolic activation
O Phase II metabolism, 53, 369. See also
Organisation for Economic Co-operation Metabolic activation
and Development (OECD), 145, 146, Phenotype, 98, 369
221, 232, 242, 368 Photodegradation, 4, 5t, 238, 247, 369

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 448 Bioanalytical Tools in Water Quality Assessment

Photosynthesis, 2, 12, 59, 60t, 63, 94, 95, QIVIVE (quantitative in vitro to in vivo
118, 135–138, 137f, 266, 275, 278, 283t, extrapolation), 144, 150–151, 233, 284,
284, 317, 370 346, 371
Photosystem II, 63, 95, 199, 200t, 279, 290, Quality assurance and quality control
316, 370. See also Photosynthesis (QA/QC), 92, 161, 225–243, 246, 371
Phytotoxicity, 17, 199–200, 271, 370 Quality standards (QS), 41, 371
Plasmid, 13, 172, 172f, 370 Quantitative structure-activity relationship
Point of departure POD), 103, 152f, 281, (QSAR), 146, 278f, 371
287, 341, 370 QuEChERS (quick, easy, cheap, effective,
Point of inflexion, 370 rugged and safe), 326, 334
Polycyclic aromatic hydrocarbon (PAH), Quinolones, 127t, 371
38, 53, 54f, 85, 130, 173, 327, 369
Polydimethylsiloxane (PDMS), 326, 332f
Polychlorinated biphenyls, 75, 370. See also R
(PCBs) Radioimmunoassay (RIA), 371
POPs. See persistent organic pollutants RAR. See retinoic acid receptor
Positive control, 108, 228, 237, 239, 370 REACH (Registration, Evaluation,
PPAR, 55t, 173, 175, 370 Authorisation and Restriction of
PR. See Progesterone receptor Chemicals), 9, 20, 26–28, 32, 41, 133,
Precautionary principle, 32 134, 371
Precision, 226–227, 235 Reactive oxygen species (ROS), 60t, 65–67,
Predicted No Effect Concentration (PNEC), 67f, 69t, 81, 84, 166, 211–212, 371, 373
8, 28, 29f, 30, 42, 284, 291, 370 Reactive toxicity, 13, 14, 64–67, 70, 165,
Primary cell lines, 348, 370 202–212, 309t, 371
Primary mechanism, primary effects, 370 Read-across, 22, 93, 133, 146, 281,
Primary producer, 44, 93 291–294, 371
Progesterone receptor (PR), 89, 179, Receptor, 13f, 41, 55, 57–59, 63, 139, 172,
190–194, 268f, 276, 304t, 306, 311t, 336, 371. See also Nuclear receptor
312t, 313t, 370.. See Progestogens Receptor binding assay (RBA), 2, 9, 168,
Progestogens, 88, 89, 370. See also 371. See also Receptor, Ligand
Progesterone receptor Receptor-mediated toxicity, 14, 130, 372.
Prokaryote, 346, 370. See also Eukaryote See also Receptor, Nuclear receptor
Promoter, 64, 69, 172, 172f, 370 Recombinant cell lines, 12, 372
Protection-goal motivated approach, 16–17, Recovery, 77, 249, 257–260, 342
371. See also Chemical-group motivated REF. See Relative enrichment factor or
approach relative extraction factor
Proteolytic enzymes, 84, 371 Reference dose (RfD), 29, 37, 281, 372
Proteomics, 11, 347, 371 Relative effect potency (REP), 116, 117,
Purified recycled water (PRW), 16, 371 161, 180, 182f, 284, 285, 291, 372
PXR, 55t, 139, 173, 178, 178t, 179t, 254, Relative enrichment factor (REF), 102,
269, 318, 371. See also Phase I 114, 115f, 117, 118, 155f, 186, 230,
metabolism 231f, 248, 262, 289, 294, 306, 307f,
Pyrethroids, 63, 371 315f, 372
Relative extraction factor (REF), 114, 186,
Q 248, 372
qAOP (quantitative adverse outcome Relative fluorescence unit (RFU), 236, 372
pathway), 326, 334 Relative light units (RLU), 372

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Index 449

Relative genotoxic unit (rGTU), Sertoli cells, 90, 373

309t–310t, 372 Serum-mediated passive dosing (SMPD),
Relative toxic unit (rTU), 310t, 373 161–163, 335, 373
REP. See Relative effect potency Shewart chart, 240
Replication, 57, 60t, 65, 233–235 Single cell gel electrophoresis
Reporter gene, 9, 13, 61, 98, 99, 108–110, (SCGE), 203, 204t, 373. See also Comet
126, 139, 163, 166–168, 172–174, 172f, assay
178, 195, 236, 242, 269, 276, 318, 320f, SMPD. See Serum-mediated passive
347, 350, 372 dosing
Reporter plasmid, 13, 372 Soil, 91, 114, 239, 251, 326–333, 357
Responsive element, 13, 372 Solid phase extraction (SPE), 48,
Retinoic acid receptor (RAR), 22, 179, 114, 169, 237, 245, 250f, 252, 253,
197–199, 268f, 304, 371 255–260, 266, 306, 343, 374. See also
Retinoic X receptor (RXR), 179, 198, 268f, Liquid-liquid extraction
304f, 373 Solid phase microextraction (SPME), 158,
Reverse dosimetry, 151 159t, 168
Reverse osmosis (RO), 4, 22, 48, 178, Solvent, 229, 237–239, 253, 254, 257, 261,
308, 372 264, 333, 343, 344
Reverse toxicokinetics, 153 SOS Chromo, 203, 204t, 373. See also SOS
Revertant ratio (RR), 208, 209 response
RIA. See Radioimmunoassay SOS response, 203, 204t, 374
Risk, 2, 8, 25, 350, 373 SPE. See Solid phase extraction
Risk assessment, 9, 25–33, 130–135, Specificity, 138, 139, 163–166, 226, 287
142–168, 373 Specificity ratio (SR), 112, 165f, 165, 278,
Risk characterisation, 26, 30–32, 40, 142, 278f, 279f, 287f, 292f, 374
373. See also Risk assessment Specific mode of toxic action, 374
Risk index (RI), 131, 140, 373. See also Risk Spermatogenesis, 90, 374
quotient Species sensitivity distribution (SSD), 30,
Risk management, 25, 26, 32–33, 33f, 373 37, 92, 141
Risk quotient (RQ), 30, 131, 134f, 373. SPM. See Suspended particulate matter
See also Risk index, Hazard quotient SPME. See Solid phase microextraction
Risk reduction, 31, 33, 373 SPR. See Suppression ratio
RNA, 60t, 373 SR. See Specificity ratio
Robustness, 47, 169, 227–228 SSD. See also species sensitivity
ROS. See reactive oxygen species distributions
RQ. See Risk Quotient Stable transfection, 374
RXR. See Retinoic X receptor Standard operating protocol (or procedure)
(SOP), 226, 242, 352, 373
S Steady-state, 151f
S9, 158, 163, 203, 206, 208, 213, 373. Stockholm Convention, 28, 133, 374.
See also Metabolic activation See also POP
SCGE. See Single cell gel electrophoresis Substances of very high concern (SVHC),
Sediment, 30, 41, 114, 238, 325–337 8, 28
Selectivity, 166, 226 Super-minimal concentration, 374. See also
Sensitivity, 57, 82, 180, 227, 229–231, NOEC
335, 350 Superoxide dismutase (SOD), 211t

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 450 Bioanalytical Tools in Water Quality Assessment

Suppression ratio (SPR), 112, 112f, 182, Threshold of toxicological concern (TTC),
185t, 374 40, 375
Supra-maximal concentration, 374 Thyroid receptor (TR), 179, 194–195, 196f,
Surface water, 4, 5t, 15, 22, 36, 37f, 38, 197f, 268t, 304t, 311t, 312t, 376
41, 176t, 177t, 183t, 189t, 191t, TIE. See Toxicity identification evalutation
192t, 193t, 194t, 199t, 200t, 201t, TIF2, 375. See also Co-activator
208, 214, 222, 248, 269, 275, 282t, Tissue, 59t, 77, 78, 104, 175, 238, 335
283t, 286f, 292f, 293f, 304, 314–318, TMX. See Tamoxifen
320–324 TOC (total organic carbon), 341
Suspended particulate matter (SPM), 250, TOX (total organic halogens), 341
326–333, 374 Tox21, 12, 108, 144, 147–148, 150f, 154,
SVHC. See substance of very high concern, 157, 159, 161, 243, 349, 375
8, 28 ToxCast, 108, 109f, 144, 147, 148, 154, 168,
Synapse, 63, 96, 374. See also Axon 174, 242, 317, 319, 349, 375
Synergism, 120, 124, 125, 344–345, 374 Toxic equivalency factor (TEF), 116, 129,
133, 374
T Toxic equivalent concentration (TEQ),
Tamoxifen (TMX), 113f, 182, 185t, 309t, 116–117, 129–131, 375
313t, 375 Toxicity identification and evaluation (TIE),
Target concentration, 61, 158 46, 275, 375
Target site, 2, 53, 58, 59t, 62, 73, 87, 122, Toxicity pathway, 7, 11, 51–71, 73–90,
124, 158, 159t 98–99, 119, 124–126, 144, 154, 169,
TCDD, 129, 173–175, 174t, 236, 329t–330t, 170f, 340, 375. See also Adverse
374 outcome pathway
Tcpl (ToxCast Analysis Pipeline), 108, Toxicity Testing in the 21st Century, 144,
109f, 147, 242, 375 147–148
TD. See Toxicodynamics Toxicodynamics (TD), 97, 124, 375.
TDI, 29, 37, 374. See also Acceptable daily See also Toxicokinetics
intake, Reference dose Toxicogenomics, 11, 376
Test battery, 2, 14, 16–17, 149, 215, Toxicokinetics, 51–55, 96, 153, 163, 281,
267–268, 270, 296, 304, 305, 309t, 310t, 376. See also Toxicodynamics
314, 315f, 340 Toxic ratio (TR), 164
TEQ. See Toxic equivalent concentration Toxic unit (TU), 116, 117, 122, 130, 139,
TEQbio, 117, 374. See also Toxic equivalent 273, 290, 310t, 375
concentration Toxtracker, 153
TEQchem, 374. See also Toxic equivalent TR. See Thyroid receptor
concentration Transcription, 54, 57, 64, 68, 69, 82, 178
Teratogenesis, 83, 374 Transcription factor (TF), 68, 69t,
TF. See Transcription factor 175, 376
The International Programme on Chemical Transcriptomics, 11, 153, 347, 376
Safety (IPCS) of the World Health Transthyretin (TTR), 283t, 328, 335
Organization (WHO), 130, 365 Trihalomethanes (THM), 5–6, 255, 376.
THP1, 310t, 375 See also Disinfection by-product
THP1-CPA, 310t, 375. See also THP1, Trophic level, 30, 44, 92, 376
Cytokine TTC. See Threshold of toxicological concern
Threshold chemical, 38, 375. See also TU. See Toxic unit
Non-threshold chemical Tumour, 70, 82, 84, 376

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 Index 451

Tumour suppressor gene (p53), 65, 69t, 70, Well, 1, 47, 147, 157, 158, 159, 160, 160f,
166, 212, 213–215, 216t, 249, 294, 306, 161, 167, 169, 170, 215, 218, 222, 225,
316, 369 232, 237, 243, 246f, 248, 262, 263f, 276,
283t, 343, 346, 347, 349, 377
U WET, 41–49, 252, 376
umuC, 203, 204t, 206, 207t, 271, 294, 376.. WFD. See Water Framework Directive
See SOS response WHO, 29, 36–39, 42, 133, 281, 351, 376
Uncertainty analysis, 32 WWTP. See Wastewater treatment plant
Uncoupler, 62, 376
X
V Xenobiotic, 53, 54–57, 69, 75, 76f, 175, 202,
Validation, 145, 147, 168, 226–231, 239, 210, 377
246, 346, 350 Xenobiotic metabolism, 53, 53f, 68, 170,
Vasoactive agent, 376 173–179, 222, 258, 267, 270,
Vibrio fischeri, 14, 218. See also Aliivibrio 288t, 290, 291, 297, 304t, 309t, 311,
fischeri 313t, 318
Vitellogenin (Vtg), 10, 46, 47, 98, 126, 154, Xenobiotic receptor, 54, 55, 377
171, 376. See also Endocrine disruption Xenoestrogen, 97, 123, 126, 267, 377.
See also Estrogen
W
96-Well plate: microtiter plate, 157, 158, Y
169, 218, 225, 248, 262, 347 Yeast estrogen screen (YES), 99, 126, 180,
Water Framework Directive (WFD), 9, 36, 242, 377. See also Estrogen receptor,
43f, 142, 250, 305, 376 Recombinant cell lines
Water Safety Plan (WSP), 351, 352f Yeast two-hybrid assay, 195, 199
Wastewater treatment plant (WWTP), 46,
48, 112, 114, 115f, 132, 135, 136f, 137f, Z
141, 175, 231f, 247f, 266, 273, 299, Z-factor, 228, 229f, 241
314f, 314–318, 376 Zona radiata protein (Zrp), 377

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest
 The first edition of Bioanalytical Tools in Water Quality Assessment was
released in 2012. The field has exploded since and the second edition
updates and reviews the application of bioanalytical tools for water
quality assessment including surveillance monitoring. The book focuses
on applications to water quality assessment ranging from wastewater to
drinking water, including recycled water, as well as treatment processes and
advanced water treatment. Emerging applications for other environmental
matrices are also included. Bioanalytical Tools in Water Quality Assessment,
Second Edition, not only demonstrates applications but also fills in the
background knowledge in toxicology/ecotoxicology needed to appreciate
these applications. Each chapter summarises fundamental material in a
targeted way so that information can be applied to better understand the
use of bioanalytical tools in water quality assessment.

The book can be used by lecturers teaching academic and professional courses
and also by risk assessors, regulators, experts, consultants, researchers and
managers working in the water sector. It can also be a reference manual for
environmental engineers, analytical chemists and toxicologists.

iwapublishing.com
@IWAPublishing
ISBN: 9781789061970 (Paperback)
ISBN: 9781789061987 (eBook)

Downloaded from http://iwaponline.com/ebooks/book-pdf/898463/wio9781789061987.pdf

by guest