Biotechnology Advances 42 (2020) 107582

Contents lists available at ScienceDirect

Biotechnology Advances
journal homepage: www.elsevier.com/locate/biotechadv

Research review paper

Current trends in the production of biodegradable bioplastics: The case of T

polyhydroxyalkanoates
João Medeiros Garcia Alcântaraa, Francesco Distanteb, Giuseppe Stortib, Davide Moscatellia,
Massimo Morbidellia, Mattia Sponchionia,
⁎

a
Department of Chemistry, Materials and Chemical Engineering “Giulio Natta”, Politecnico di Milano, via Mancinelli 7, 20131 Milano, Italy
b
Institute for Chemical and Bioengineering, Department of Chemistry and Applied Biosciences, ETH Zurich, Vladimir-Prelog-Weg 1, 8093 Zurich, Switzerland

ARTICLE INFO ABSTRACT

Keywords: The global pollution caused by plastics and microplastics is stimulating intense research towards more en­
Polyhydroxyalkanoates vironmentally friendly materials, preserving the remarkable application characteristics of the currently available
Bioplastics polymers. Among these, polyhydroxyalkanoates (PHAs) have been hailed as the solution to replace conventional,
Microbiological oil-based plastics. Given their biodegradable nature and mechanical properties, their use can be envisioned in a
Fermentation
wide range of applications reducing the environmental footprint. Several types of processes have been proposed
Enzymatic
Chemical
for their production, which can be grouped in three main classes: (i) microbiological, (ii) enzymatic and (iii)
Ring-opening polymerization chemical processes. Given the significant amount of literature available on this topic, this review aims to cri­
Production tically analyse what has been proposed so far in each of these classes, with specific reference to their potential to
Biodegradable provide bioplastics that can actually replace the currently available materials. A comparison is made, based on
the following aspects: achievable molecular structures (such as molecular weight and composition distributions),
raw-material and production costs and availability of large-scale production technologies. Finally, some con­
siderations and ideas on what should be further investigated and implemented to realize the economically
sustainable production of PHA are brought forward.

1. Introduction been identified as having potential as substrates for PHA synthases and
hence for polymerization. However, most of these have only been used
Amidst concerns derived from the overuse of fossil-based plastics, in laboratory settings and solely a handful have been adopted by in­
there is their resistance to degradation, which leads to large accumu­ dustry. Interestingly, most building blocks are chiral and since PHA
lations of waste materials in the environment. This has stimulated sci­ synthases are stereospecific only R-configured PHAs are found in nature
entific and financial efforts on finding green alternatives to the tradi­ (Koller et al., 2013b).
tional plastics. Among these, bio-based polymers have the potential to Each HA monomer has a side chain (R), often a saturated alkyl
substitute traditional polymers, ensuring similar properties, while being group. However, this chain can also be constituted of unsaturated alkyl,
biodegradable and thus reducing the environmental footprint related to branched alkyl and substituted alkyl groups (Tan et al., 2017). PHAs are
anthropogenic activities (Raza et al., 2018). classified according to the number of carbons in this side chain. Less
A class of biopolymers that is currently attracting much attention, than 5 carbons correspond to short-chain length PHAs (scl-PHAs), be­
due to the wide range of mechanical and thermal properties it can ac­ tween 5 and 14 to medium-chain length (mcl-PHAs) and with more
cess, is the one of polyhydroxyalkanoates (PHAs). PHAs are linear than 14 carbons, to long-chain length PHAs (lcl-PHAs). However, lcl-
polyesters, whose general structure is schematized in Fig. 1D. The PHAs are not very common in nature (Raza et al., 2018).
possibility of tuning their properties arises from the side chain R and the In general, PHAs are biodegradable, biocompatible, non-toxic, in­
length of the alkyl chain n, which are characteristic of the monomer soluble in water and soluble in chloroform and other chlorinated sol­
used for the synthesis, besides their chain length, x. vents. Their glass transition temperature varies from −50 to 4 °C and
PHAs are naturally occurring polymers, mainly produced by bac­ the melting temperature from 40 to 180 °C, according to their chemical
teria for carbon and energy storage (Możejko-Ciesielska and Kiewisz, composition and chain length. However, depending on the type of PHA,
2016). Specifically, more than 150 monomeric building blocks have composition and molecular weight, these properties vary greatly, even

⁎
Corresponding author.
E-mail address: [email protected] (M. Sponchioni).

https://doi.org/10.1016/j.biotechadv.2020.107582
Received 16 April 2020; Received in revised form 8 June 2020; Accepted 18 June 2020
Available online 02 July 2020
0734-9750/ © 2020 Published by Elsevier Inc.
J. Medeiros Garcia Alcântara, et al. Biotechnology Advances 42 (2020) 107582

Fig. 1. Main processes for the production of PHA: micro­

biological, enzymatic and chemical. Light green boxes in­
dicate the factors which most affect each of the classes of
processes. Dark green boxes indicate the three main topics for
the critical comparison present in this work. A) Transmission
electron micrograph of bacterial cells accumulating very large
quantities of PHA as inclusion granules (white bodies inside
the cellular membrane). Bar represents 0.5 μm. Reproduced
with permission from Sudesh et al., 2000 (Sudesh et al.,
2000). B) Crystal structure of the PHA synthase (PhaC) from
Chromobacterium sp. USM2 bound to Coenzyme A. Image of
6K3C (rcsb.org) (Chek et al., 2020) (“Jmol: an open-source
Java viewer for chemical structures in 3D, 2020.”) C) Ex­
amples of lactones which can be polymerized into PHAs by
ring-opening polymerization (ROP) (from left to right: β-bu­
tyrolactone, γ-butyrolactone, δ-valerolactone). D) General
chemical structure of PHAs: R represents the side chain, n the
length of the alkyl chain and x the length of the polymeric
chain. (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this
article.)

affecting the degradation kinetics of the material (Raza et al., 2018). hindrances towards commercialization are discussed. These production
Therefore, the modulation of the chemical microstructure of these routes are then compared side-by-side in terms of
polymers can be viewed as a powerful tool for the tuning of their me­
chanical and thermal properties, thus accessing materials suitable to a i) molecular characteristics, such as the molecular weight (MW) and
wide variety of applications, possibly replacing conventional plastics. composition distribution of the products
These polymers were observed for the first time in 1888. However, ii) raw-materials and production costs
at that time, their composition and biological role could not be properly iii) availability of industrial technologies for large-scale production
defined. As early as 1926, poly-3-hydroxybutyric acid (P(3HB)) was
obtained from Bacillus megaterium by a French scientist, even if its These considerations will then highlight the issues that need to be
function as carbon and energy storage was established only in 1959. addressed to facilitate the affirmation of PHAs in the plastic market.
One year later, the commercialization of PHA started (Możejko-
Ciesielska and Kiewisz, 2016). Typically, commercial PHAs have been 2. The microbiological route
produced through a microbiological route (Tan et al., 2017). However,
both chemical and enzymatic processes have been widely investigated The production of PHAs is intrinsically connected to natural stra­
as a strategy to improve the control of the physicochemical properties tegies for survival deployed under environmental stress by bacteria,
of the product (Tang and Chen, 2018; Thomson et al., 2009). which form inclusion bodies as carbon and energy storage (Możejko-
Due to the increasing interest towards this new class of bioplastics Ciesielska and Kiewisz, 2016). The insoluble granules can take up 90%
and the growing literature on the different strategies for its production, of the dry weight of the cell mass (see Fig. 1A). This strategy is widely
the present work aims at reviewing the recent advances in the pro­ employed by gram-positive and gram-negative bacteria and, also, in
duction processes of PHAs. The three main routes, namely micro­ archaea. Representative examples include Cupriavidus necator (pre­
biological, enzymatic and chemical, schematically illustrated in Fig. 1, viously known as Ralstonia eutropha) and Pseudomonas sp. (Sudesh et al.,
are considered and their main features, potential scalability and current 2000). Besides the storage of energy, this process allows for increased

2
J. Medeiros Garcia Alcântara, et al. Biotechnology Advances 42 (2020) 107582

Fig. 2. Simplified metabolic pathway for the production of P(3HB). PhaA: β-ketothiolase. PhaB: NADPH-dependent acetoacetyl-CoA reductase, PhaC: PHA synthase,
PhaZ: PHA depolymerase.

stress tolerance to transient environmental conditions, such as heat, metabolic flux of acetyl-CoA. When there is a surplus of carbon and
osmotic shock and ultraviolet radiation (Tan et al., 2014). Although this there is no other growth-related limitation, the synthesis of consider­
strategy is naturally occurring in a wide variety of microorganisms, able amounts of CoA from the Krebs cycle blocks PHA synthesis by
alike many other bioproducts, it is also possible and advantageous to inhibiting PhaA. Acetyl-CoA is in fact assimilated into the Krebs cycle
genetically alter non-PHA producers (e.g. Escherichia coli) to obtain very for energy production and cell growth. However, if the carbon surplus
efficient PHA producers, as discussed in details later in this work conditions are maintained and there is a limitation of an essential nu­
(Możejko-Ciesielska and Kiewisz, 2016). trient, the CoA concentration becomes non-inhibitory, leading to the
supply of acetyl-CoA to the PHA biosynthetic pathways (Tan et al.,
2014). Based on this metabolic causality network, these conditions are
2.1. Metabolic pathways mimicked in bioreactors in order to produce PHA. Following the normal
procedure for producing secondary metabolites (i.e. metabolites not
The carbon source from which the PHA is produced can be cata­ associated with the cellular growth stage), most PHA producing pro­
bolized through a number of metabolic pathways (Meng et al., 2014). cesses consist of two steps. The first one is the growth stage, which aims
As a representative example of the PHA class, this section will deal with to increase the cell concentration inside the reactor, where there is no
the specific case of P(3HB), which is currently the most synthetized nutrient limitation. This is followed by the production stage, where a
PHA worldwide (Vandi et al., 2018). For this biopolymer, the final steps limitation is introduced (e.g. lack of nitrogen source), in order to direct
are invariably driven by three different enzymes: PhaA, PhaB and PhaC the metabolic pathways to the accumulation of PHA and to stop growth
(see Fig. 2). In this case, the carbon source is transformed into acetyl- and cell multiplication processes (Koller et al., 2018).
coenzymeA (CoA), 2 molecules of which are converted into acetoacetyl- It should be pointed out that, after PHA accumulation, if these
CoA by a thiolase (β-ketothiolase, PhaA). The precursor for this type of limitations are removed, the PHA can be depolymerised and the normal
PHA, (R)-3-Hydroxybutyryl-CoA is then formed by the NADPH-depen­ metabolism of the bacteria for energy production and growth is re-es­
dent acetoacetyl-CoA reductase (PhaB), leading to the polymerization tablished. Under this change in conditions, a PHA depolymerase (PhaZ)
of P(3HB) by the PHA synthase (PhaC) (Sudesh et al., 2000). breaks down the PHA into easily assimilated energy compounds
This type of metabolite production is deeply linked to the central (Sudesh et al., 2000).
metabolic pathways of the bacteria, including, but not limited to: gly­ There is a wide variety of possible substrates and ensuing metabolic
colysis, Krebs cycle, β-oxidation, de novo fatty acid synthesis, amino pathways for the production of PHAs, which cannot be discussed here
acid catabolism, Calvin cycle and serine pathway. For example, Acetyl- in detail. For this, the interested reader is directed to the comprehensive
CoA is an intermediate which is shared by these pathways and the review by Steinbüchel and Lütke-Eversloh (Steinbüchel and Lütke-
production of PHA. In fact, in most native PHA producers, the forma­ Eversloh, 2003).
tion of PHA only occurs in certain environmental conditions due to the

3
J. Medeiros Garcia Alcântara, et al. Biotechnology Advances 42 (2020) 107582

2.2. PHA synthases of PhaC from Chromobacterium sp. USM2 (Chek et al., 2017) and Cu­
priavidus necator, the latter by two distinct research groups (Kim et al.,
PHA synthases catalyse the polymerization of CoA thioesters of HA 2017; Wittenborn et al., 2016). Specifically, they found that the dis­
into PHA polymers, releasing the CoA moiety (Zou et al., 2017). In tance between the catalytic cysteine residues present in each of the
general, this enzyme is denominated PhaC, since it is encoded by the subunits of PhaC was too large to allow for the ensuing chemical re­
gene phaC. This gene is normally clustered with other genes that are actions that should occur in the first model. Hence, they favour a cat­
necessary for the production of PHAs, namely phaA and phaB, which alytic mechanism with only one catalytic site. In spite of this, no
encode the β-ketothiolase and the NADPH-dependent acetoacetyl-CoA agreement has been found on the exact enzymatic mechanism that
reductase, respectively (Zou et al., 2017). occurs at the single catalytic site.
Generally, PHA synthases can be divided into 4 classes (Rehm, Two opposing models have been proposed. On the one hand, Kim
2003; Tsuge, 2016). Class I and class II include enzymes constituted by et al. postulated a single active site non-processive ping-pong me­
only one type of subunit (PhaC), with molecular masses between 61 and chanism (also called by other authors, as a processive mechanism with
73 kDa. Both classes utilize specific substrates. Class I synthases, present a single tunnel) (Kim et al., 2017), in which the ingress to and egress
e.g. in C. necator, preferentially use CoA thioesters of several (R)-3- from the active site is done via the same tunnel. On the other hand, in
hydroxy fatty acids with 3 to 5 carbon atoms. On the other hand, class II the processive mechanism (or processive mechanism with in-out tun­
synthases prefer CoA thioesters of various (R)-3-hydroxy fatty acids nels) (Chek et al., 2017; Teh et al., 2018; Wittenborn et al., 2016) a
with 6 to 14 carbon atoms (Rehm, 2003). through passage exists and there is the covalent bonding of the poly­
Both the remaining classes are constituted by two distinct types of meric chain to the cysteine residue present in the active site. A com­
subunits, one of which they have in common: the PhaC subunit. This prehensive overview of these putative mechanisms can be found in the
has an amino acid sequence with a similarity of 21–28% to class I and II work by Sagong et al. (Sagong et al., 2018).
synthases and molecular mass of about 40 kDa. The other subunit is Very recently, Chek et al. have brought forward the crystal structure
designated PhaE (approximately 40 kDa) and PhaR (approximately of the catalytic domain of PhaC of Chromobacterium sp. bound to CoA
20 kDa), for class III and IV, respectively. These synthases have a sub­ (Chek et al., 2020). This is significant since it allows further under­
strate specificity towards CoA thioesters of several (R)-3-hydroxy fatty standing into the dynamics of the dimerization process. Explicitly, it
acids with 3 to 5 carbon atoms (Rehm, 2003). was found that PhaC undergoes a conformational change when bound
As expected, based on the exhibited substrate specificity, synthases to CoA, forming an asymmetric dimer with a closed and an open con­
from classes I, III and IV produce scl-PHAs, while enzymes from class II formation for each of the subunits. In addition, this research group
produce mcl-PHAs. postulates that this conformation eases both the entrance of substrates
Representative species are as follows: Cupravidius necator (pre­ to the active site and the formation of an egress tunnel for the product,
viously Ralstonia eutropha), P. aeruginosa, Allochromatium vinosum and adding evidence to support the processive mechanism with in-out
Bacillus megaterium, for class I, II, III and IV, respectively (Tsuge, 2016). tunnels. However, it should be pointed out that this PhaC-CoA crystal
The nucleotide sequences of at least 59 PHA synthases have been ob­ structure has still not been determined for C. necator, which should
tained. As such, some of these do not conform to the aforementioned provide further insight into the discussed enzymatic mechanisms.
classes. Examples include the PHA synthases obtained from the fol­ In any case, the polymerization process will continue with the
lowing microorganisms: Thiocapsa pfennigii, Aeromonas punctate and propagation phase, in which, the polymeric chain grows by the addition
some Pseudomonas sp.(Rehm, 2003). Interestingly, a PHA synthase from of monomers. Finally, chain termination occurs when the PHA synthase
an extremely halophilic arcaebacterium was identified and character­ reaches the limit of its catalytic activity and becomes inactive (Tsuge,
ized, possibly constituting a new class of synthases. This enzyme had its 2016). With this regard, Tian et al. (Tian et al., 2005) claim that the
maximum activity at 40 °C, and was stable up to 60 °C, where it de­ synthase can “sense” the size of the outgoing chain and catalyse the
creased by only about 10% (Hezayen et al., 2002; Rehm, 2003). movement of the polymeric chain to another amino acid, while leaving
With regards to the molecular mass, the PHAs produced by class I a primed monomer in its place, ready to restart the polymerization of a
enzymes exhibit a higher molecular mass than those obtained by class new chain. In alternative, a chain transfer (CT) reaction can also occur.
II, normally, between 500 kDa to several millions and from 50 kDa to In this case, the polymer chain is transferred to a CT agent, which
500 kDa, respectively. Class III and IV synthases produce PHAs with covalently binds to the carboxy terminal of the polymer chain. Possible
intermediate molecular masses (Rehm, 2003). CT agents include water, 3HB and other hydroxy compounds (Tsuge,
2016).
2.3. Catalytic mechanism It should be noted that the polymerization carried out by the PHA
synthase can be regarded as a living polymerization. Su and co-workers
The model that the current literature suggests for the polymeriza­ (Su et al., 2000) investigated this system and found evidence that the
tion of PHAs is based on the PHA synthase-catalysed polymerization molecular weight of the final biopolymer is closely related to the
reaction that includes 3 steps: initiation, propagation and termination monomer to enzyme ratio. In particular, when reaching total conver­
(Tsuge, 2016). Fig. 3 describes these steps from a chemical standpoint sion, the number average molar mass was uniquely defined by this
using a class I synthase, as an example. This type of synthase has only ratio. In addition, by running two of such reactions, one after the other,
one type of subunit, PhaC, which after translation, exists as water-so­ they obtained a di-block copolymer - also a characteristic of living
luble single forms. In the initiation step, two PhaC subunits are di­ polymerizations. This is in spite of the MW distribution of the produced
merized, thus forming the active enzyme. There are two proposed polymers being somewhat broad. This is indeed unusual for living
mechanisms for this process, involving two thiol groups (Stubbe and polymerizations, which are often characterized by polydispersity
Tian, 2003; Tsuge, 2016). The difference lies in the origin of these smaller than about 1.5. However, such broadening can be explained by
groups: in one of the models the thiols are provided by the cysteine the fact that in these systems the rate of initiation is considerably slower
residues from each of the PhaC subunits in the dimer. In the other one, than that of propagation.
one thiol is provided by a cysteine residue and the other by the thioester
bond present in the monomer. 2.4. Microbiological strains
The exact catalytic mechanism for this step has been researched for
more than 30 years (Sagong et al., 2018), however no definitive answer The selection of the microbiological strain is one of the points of
has been found to date. Nonetheless, significant insight has been un­ prime importance for the successful production of the desired PHA. This
covered in the last years with the determination of the crystal structure choice directly impacts the type of substrate used, as well as the

4
J. Medeiros Garcia Alcântara, et al. Biotechnology Advances 42 (2020) 107582

Fig. 3. Model proposed by Tsuge (Tsuge, 2016) for the enzymatic polymerization of P(3HB). CoA-SH – free coenzyme A, E-SH – PHA synthase subunit, E2-SH –
dimerized PHA synthase (active), E2-SH – dimerized PHA synthase (inactive), HO-X – chain transfer agent, 3HB – 3-hydroxybutyryl monomer.

molecular weight and molecular structure of the final polymer, in ad­ producing PHA the reader is directed to the review by Tan et al. (Tan
dition to other physico-chemical properties (Wang et al., 2014). In et al., 2014).
addition, it can also affect the process for isolating and purifying the As mentioned above, non-native bacteria can be modified to pro­
produced PHA due to structural differences of the corresponding cells duce PHA. Here, Escherichia coli, the “work-horse” of biotechnology, is
(Koller et al., 2013a). indeed one of the most used. This organism possesses several ad­
The ability to produce PHA is a somewhat diffused attribute in vantages: the possibility of using a variety of cheap carbon sources, no
microorganisms, having been identified in more than 70 bacterial and intracellular PHA depolymerase and a very well-known genetic back­
archaeal genera. In particular, scl-PHAs are produced by a wide range ground. These characteristics establish this species as one of the most
of bacteria, e.g. Cupriavidus necator, while mcl-PHAs are mainly pro­ promising organisms for the large-scale production of PHA. Specifically,
duced by Pseudomonas sp. (Możejko-Ciesielska and Kiewisz, 2016). As the production of PHA in E. coli entails the transfer of a PHA synthase
mentioned earlier, the variety of PHA-producing strains also entails the structural gene from a native PHA producer, the assurance of the ex­
myriad of carbon catabolic pathways and PHA anabolic pathways, pression of the active form of the enzyme and the engineering of
generating a huge diversity of substrates from which PHA production pathways to provide sufficient concentration of the precursors (Li et al.,
can occur (Tan et al., 2014). 2007). This species has in fact been genetically engineered to produce
In general, for biotechnology purposes, an industrial production PHAs up to 90% of the CDW (Lee and Chang, 1995). Chen and Jiang
strain should possess the following characteristics: not be a pathogen have published a detailed review on the genetical manipulation of
and not produce toxins, have a clear genomic background, be of easy bacteria to increase PHA production (Chen and Jiang, 2017). In addi­
genetical manipulation, have a fast growth and a wide substrate utili­ tion to the increased PHA accumulation, non-native PHA producers as
zation. If possessing such characteristics, most likely the production E. coli have the potential to narrow the molecular weight distribution of
strain can be manipulated to increase the levels of PHA accumulation the produced polymer. Polydispersity indexes as low as 1.4 have been
and substrate conversion and even the cell size. All of this enables the achieved, very similar to the values obtained with an in vitro PHA
augmentation of the amount of polymer that each cell can accumulate polymerization (Leong et al., 2014; Tsuge, 2016).
and facilitates the rupture of the cellular membrane and hence the re­ In general, several factors related to the chosen strain can impact
moval of the PHA granules (Tan et al., 2017; Wang et al., 2014). the characteristics of the produced PHA, and in particular the following
Native PHA-producing bacteria can be genetically engineered to ones impact directly the molecular weight (Rehm, 2003; Tsuge, 2016):
increase their polymer production. Species like Cupriavidus necator can
accumulate PHAs in satisfactory amounts, and was, in fact, one of the i) The level of expression of active PHA synthase protein in the cells.
first ones used at an industrial level (Możejko-Ciesielska and Kiewisz, The higher the concentration, the lower the molecular mass.
2016). Specifically, accumulations have been reported until 89% of ii) The degradation of PHA during biosynthesis, such as the avail­
cellular dry weight (CDW) (Tan et al., 2014) and productivities of at ability of enzymes that hydrolyse PHAs, including PHA depoly­
least 3.1 g/(L h) (Koller et al., 2018). For a comprehensive comparison merases, but also of other unspecific esterases and lipases. If these
of the accumulation and productivities of the different microorganisms enzymes are not present, polymers with higher molecular masses

5
J. Medeiros Garcia Alcântara, et al. Biotechnology Advances 42 (2020) 107582

can be produced. any other factor, it is the genetic background that affects the most the
iii) The catalytic activity of the active PHA synthase. characteristics of the produced polymer.
iv) Occurrence of chain transfer reactions (see section 3.2). In the above mentioned review of Leong et al. (Leong et al., 2014),
the production of ultra-high MW PHA using different E. coli strains was
Interestingly, new bacterial strains have been discovered which compared. They found large differences among the different strains and
produce PHAs extracellularly. These include Pseudomonas Corrugata that even the way the plasmid was constructed and its bacterial origin
and Pseudomonas Mediterranea (Licciardello et al., 2019). While, cur­ affected the MW and polydispersity of the polymer, as well as its
rently, the yields are still low, this could be an interesting option, since composition in the case of copolymers.
it would remove the extraction stage from the PHA production process, Currently, all the industrial PHA producers either adopt batch or
with significant cost reductions. fed-batch bioreactors. However, lately there has been some interest in
the utilization of continuous processes (Koller et al., 2018). This is
2.5. Feedstocks driven by the potential of such processes to increase productivities and
reduce the plant footprint for the same PHA production. Stand-alone
The choice of the substrate is of paramount importance for at least chemostats have been considered to produce PHA (Zinn et al., 2003).
two reasons. Firstly, it can represent up to 50% of the overall produc­ However, since PHA is a secondary metabolite, that is its formation
tion cost (Raza et al., 2018). Secondly, it is probably the most important does not occur concomitantly to cell growth, it is not possible to have in
factor determining the polymer/co-polymer composition. Since dif­ the same reactor optimal conditions for cell growth and limiting con­
ferent carbon sources are metabolised through distinct metabolic ditions for stimulating PHA production. For this, the use of cascades of
pathways, they produce different PHA monomers (Leong et al., 2014). continuous stirred tank reactors (CSTRs) appears better suited. For
In addition, there is also an ethical consideration: the selected feedstock example, Atlić et al. have realized a 5-stage CSTR cascade, where the
should not compete with food and feed related sources. This means that first one is used for cell growth, while the others operate under nitrogen
the use of the so-called 2nd-generation feedstocks is strongly en­ limitation, so as to induce PHA accumulation. Using this process a
couraged, to prevent the rise of prices of the global food supply (Jiang productivity of 1.85 g/(L h) and a PHA accumulation of 77% of CDW
et al., 2016; Koller and Braunegg, 2018). Also, for this reason, research (Atlić et al., 2011) were achieved. This aspect is well discussed in the
is focused on reducing the cost of production by using various waste comprehensive review by Koller (Koller et al., 2018).
streams as carbon source. These include: whey, starch, spent coffee Concluding, albeit centred around genetic makeup of the selected
grounds, wastewaters, wheat and rice straw (Raza et al., 2018), glycerol PHA producer, various factors can influence the productivity and the
(Koller and Marsalek, 2015; Phithakrotchanakoon et al., 2015), and characteristics of biosynthetic PHA, including molecular weight and
lignin (Kumar et al., 2019), among others. These research efforts also composition distributions. Although the amount of data is still limited,
are fully in line with the concept of a biorefinery, where the PHA it is clear that these factors strongly affect the thermo-mechanical
production is integrated into the manufacture of other products and properties and workability of the resulting polymer.
bioplastics are produced to valorise waste streams (Kumar et al., 2017).
However, one major concern in using these strategies for PHA pro­ 3. The enzymatic route
duction is the necessity to assure constant characteristics of the poly­
merization feedstocks. This appears, in fact, to be a prerequisite to In vitro PHA production has been hailed as having the potential to
prevent producing biopolymers with non-reproducible characteristics, increase control over the polymer characteristics, such as molecular
such as changing molecular mass and copolymer composition (Raza weight and co-polymer composition, compared to the biological route,
et al., 2018). Also, this may impact some end applications of the PHA. while reducing costs by separating the PHA synthesis from bacterial
For example, in the case of medical applications, the PHA produced growth (Thomson et al., 2009). However, this process closely depends
from waste materials may contain impurities (Raza et al., 2018) that on the supply of high-quality synthase, monomers and other adjuvants
could harm the biocompatibility of the produced plastics (Koller et al., for enzymatic reactions, which can have elevated costs (Rehm, 2010).
2013a). In particular, the production of substantial amounts of synthase has
been the main obstacle to the large-scale application of this process.
2.6. Process-related factors Specifically, extraction from the native host is hampered due to ag­
gregation and current production involves over-expression and pur­
Several factors related to the production process itself also affect the ification in heterologous recombinant organisms, such as E. coli
quality and the chemical composition of the polymer. For example, (Thomson et al., 2009).
Penloglou et al. (Penloglou et al., 2012) investigated the effect of key Also, monomer production can be highly laborious due to the costs
operating variables, such as nutritional and aeration conditions, on the and complexity of producing an enantiomerically pure compound and
molecular weight of the resulting polymer, through the development of in its active form, meaning that CoA needs to be chemically linked to
a metabolic model of the microorganism coupled with a kinetic model the produced monomer. Although PHA polymerase is absolutely en­
of the polymerization steps. They found that even the feeding policy has antio-selective, the supply of only the R enantiomer to the enzyme is
the potential to modify this polymer characteristic. critical. Tajima and co-workers (Tajima et al., 2009) found, in fact, that
Temperature is another process variable that influences the mole­ the molecular weight of the polymer was reduced when both en­
cular weight of the produced PHA. Cultures of E. coli JM109 (phaRCBSP) antiomers were in solution. It was postulated that the S monomer, al­
operated at 37 °C and 25 °C showed that the MW of the PHA decreases though not reactive, binds to the catalytic active site of the synthase,
with the culture time at the higher but not at the lower temperature. which then cannot further activate the addition of R monomers. There
This behaviour was explained by the occurrence of significant random are several chemical routes for the production of the monomer. How­
scission of the polymer chains at 37 °C. However, this conclusion was ever, they all lack the ability of producing only the R enantiomer.
not confirmed when synthases from other sources were heterologously Noyori and co-workers employed an asymmetric hydrogenation of keto
expressed in E. coli. The authors claimed that this behaviour was then esters, producing the monomer with an enantiomeric excess of 99.4%
unique to this synthase extracted from Bacillus sp. INT005 (Agus et al., (Noyori et al., 1987). Huang and co-workers (Huang and
2010; Leong et al., 2014). Actually, other related studies found that for Hollingsworth, 1998) also developed a method, starting from a lactone,
the production of ultra-high MW PHA, higher temperatures, combined for the production of 3-hydroxy acids with similar enantiomeric excess,
with a slightly acidic medium, contribute to the polymer chain elon­ which could be applied to the production of the 3HB monomer, how­
gation (Leong et al., 2014). This qualifies as evidence that more than ever it is rather complex. There have also been reports of a synthetic

6
J. Medeiros Garcia Alcântara, et al. Biotechnology Advances 42 (2020) 107582

route to produce enantiomerically pure 3HB-CoA, however this in­ (Lecomte and Jérôme, 2012) to further deepen the chemistry of these
volves 9 steps and exhibits overall yields below 35% (Jia et al., 2016). systems.
Processes suitable for commercial applications should start from It should be noted that all the ROP processes mentioned so far either
widely available and less expensive substrate and perform an entirely make use of enantiomerically pure lactones or produce the atactic
enzymatic process. For example, Tokiwa et al. reported the conversion polymer. Since the challenge is to reproduce the naturally occurring
of acetoacetate into (R)-3-HB by the (R)-3-HB dehydrogenase with the polymer, while using a synthetic process with better economics, the
addition of NADH and H+ (Tokiwa and Ugwu, 2007). Nevertheless, the racemic mixture of the lactone should be used, whilst guaranteeing the
adjuvants still need to be furnished, possibly including, besides CoA, production of (R)-PHA. Following this concept, Tang & Chen (Tang and
ATP and NADPH (Thomson et al., 2009). Another downside is created Chen, 2018) proposed a different process, starting from the eight-
by the concomitant release of CoA as the polymerization progresses. As membered cyclic diolide, which can be obtained from bio-sourced di­
it was mentioned above, CoA can, in fact, have an inhibitory effect on methyl succinate, and using an yttrium racemic salen complexes-based
PhA and PhaC (Zou et al., 2017). Several processes have been devel­ stereoselective catalyst. The high molecular weight isotactic polymer
oped to counteract these disadvantages. One option is to reduce the described in the last row of Table 1 was obtained.
CoA consumption, by recycling it into the system several times. One Actually, the ROP of lactones can also be catalysed enzymatically.
example of this involves the use of an acyl-CoA synthase and has the Gorke et al. (Gorke et al., 2007) used the lipase B enzyme from Candida
advantage of allowing the strict control of monomer composition in the Antarctica and produced different PHAs with molecular weights up to
polymerization of PHA co-polymers, through the ratio of monomers in 8.1 kg/mol and yields in the order of 85%. However, this process suffers
solution. However, the disadvantage of this system is that free CoA from all the drawbacks related to the cost and availability of the en­
remains in solution, inhibiting the activity of the PhaA (Satoh et al., zyme discussed in the previous section about the enzymatic route.
2005). With a view to solve this problem, a two-phase reaction system
was developed which also included a mechanism for CoA recycling 5. Critical comparison of the different production processes and
(Han et al., 2009; Tajima et al., 2004). For this mechanism, the system future trends
relies on an ester exchange reaction between CoA in the water phase
and acetyl thioester of ethyl thioglycolate (AcETG), leading to the In order to properly compare the three different routes proposed so
formation of AcCoA in the water phase. Through the use of a propionyl far for the production of PHA, the following factors should be taken into
CoA transferase, the AcCoA reacts with the chemically synthesised account: (i) molecular characteristics of the polymer, (ii) raw-material
monomer (3HB), leading to the formation of the activated monomer, and production costs and (iii) availability of large scale production
3HB-CoA. This is, in turn, polymerized by a PHA synthase into P(3HB) technologies.
(Han et al., 2009). When considering the quality of PHAs, the most important para­
Another interesting possibility would be to immobilize the enzyme meters are the molecular weight distribution and the tacticity, since
enabling its reuse. Several established industrial processes make use of these determine the final mechanical properties of the biopolymer.
immobilized catalysts, such as the production of high fructose corn In terms of MW, the microbiological route has set the gold standard,
syrup (glucose isomerase) and the transesterification of food oils (li­ at least with respect to the order of magnitude of the maximum average
pase), among others (DiCosimo et al., 2013). However, despite several value of the MW, which is in the range of 104 to 105 g/mol for native
authors arguing that it should be possible to immobilize the PHA syn­ bacteria (Tan et al., 2014). Although this is already high with respect to
thase, to our knowledge this result has not yet been achieved (Thomson many applications, even higher values have been obtained through
et al., 2009). genetic engineering modifications, such as rearranging of metabolic
pathways or eliminating the presence of PHA depolymerases. Currently
4. The chemical route it is possible to achieve MW as high as 107 g/mol (Sudesh et al., 2000).
Similar values have been achieved also through enzymatic reactions.
The most used strategy to chemically synthetize PHA is the ring- This is not surprising since these reactions are more easily controlled
opening polymerization (ROP). This is a living polymerization em­ and exhibit an intrinsically living character, as discussed above. How­
ploying a metal catalyst and an initiator (usually an alcohol) for the ever, this does come at a cost: complex processes and expensive raw
polymerization of cyclic ester monomers, or lactones, into linear materials. While bacteria produce PHA with similar characteristics in a
polyesters. Given the living nature, the average chain length can be very efficient manner, without requiring elaborate upstream strategies
modulated through the ratio between the monomer and the initiator. In and using cheap carbon sources, enzymes need expensive raw materials
addition, the molecular weight distribution is usually narrow (poly­ such as the monomer and CoA, in addition to other chemicals needed to
dispersity < 1.5), provided that transesterification reactions are realize the chosen production system. It has been estimated that the
avoided during the process. This strategy enables, through the proper production cost of PHB through enzymatic reactions is in the order of
choice of the monomer, to synthesize a wide variety of homo and co- 286,000 USD per gram, against the 0.0025 USD per gram requested by
polymers (Hori et al., 1999, 1995; Tang and Chen, 2018). Table 1 shows the microbiological route (Rehm, 2010). Such a huge difference, al­
examples of various lactones that can be used to produce different kinds though it may be somehow overestimated, makes the enzymatic route
of PHA homopolymers. of interest only for purely scientific investigations. With this respect, the
The ROP processes described in the literature make use of different chemical production route is similar to the enzymatic process. On the
catalysts, such as β-diiminate zinc alkoxide complexes (Rieth et al., one hand, the raw materials are cheaper than the enzymatic ones, since
2002), lanthanide or yttrium complexes (Hong and Chen, 2016a; Tang only lactones are used. On the other hand, due to the occurrence of
and Chen, 2018), magnesium dibutyl catalysts (Hong and Chen, transesterification reactions, the polydispersity is often greater than 2
2016a), organic catalysts (Hong and Chen, 2016a; Lohmeijer et al., (see Table 1). Since the chemical route tries to mimic the PHA produced
2006; Makiguchi et al., 2011), tert-Bu-P4 base (Hong and Chen, 2016b), through fermentation methods (which is invariably (R) configured),
distannoexane catalysts (Hori et al., 1999, 1995) and magnesium and tacticity also comes into the play as an additional issue. The limited
zinc complexes (Liu et al., 2010). In addition to homopolymers, due to availability and the high cost of enantiomerically pure lactones, which
its the living nature, ROP is very convenient to produce also block co- are necessary to grant isotacticity, add on to the production costs.
polymers. With respect to the reaction mechanisms, we believe that all About the polydispersity, Ð, the situation is quite different. The
of the processes referred to in Table 1 occur through a coordination- chemical and enzymatic processes almost always produce very narrow
insertion mechanism (Hong and Chen, 2016a; Tang and Chen, 2018). molecular weight distributions, due to the fact that side reactions are
The interested reader can refer to the review by Lecomte and Jérôme almost completely absent and the process is highly controlled. This is

7
J. Medeiros Garcia Alcântara, et al. Biotechnology Advances 42 (2020) 107582

Table 1
Production of several types of polyhydroxyalkanoates via ROP of diverse lactones. Mn: number average MW, Ð: polydispersity. AAPip: (E)-2,6-Diisopropyl-N-(2-((2-
(Piperidin-1-yl)Ethylimino)Methyl)Phenyl)Aniline, n.d.: non determined.
Raw material Mn (kg/ Ð Conv. (%) Catalyst Initiator Prod. Poly. Ref.
R or S mol)

Type Stereochemistry Type Tact.

Pip
β-butyrolactone racemic 64.3 1.06 90 Complex (AA )ZnEt 9-AnOH P(3HB) (S/R) (Liu et al., 2010)
27.5 1.12 99
311 2.27 1-ethoxy-3- – (Hori et al., 1993)
(R) 178 2.38 chlorotetrabutyldistannoxane (R)
γ-butyrolactone racemic 30.2 2.40 29 La[N(SiMe3)2]3 C6H4(CH2OH)2 P(4HB) n.d. (Hong and Chen,
11.5 1.8 90 2016a)
36.8 1.82 13.9 KH – (Hong and Chen,
25.0 2.94 90 tert-Bu-P4 BnOH 2016b)
δ-valerolactone 27.5 1.08 90.7 Diphenyl Phosphate 3-Phenyl-1- P(5 HV) (Makiguchi et al., 2011)
5.2 1.09 95.0 propanol
ε-caprolactone 21.6 1.07 61.6 P(6HH) (Makiguchi et al., 2011)
5.9 1.07 97.1
Cyclic Diolide 154 1.01 71 Yttrium racemic salen complex BnOH P(3HB) (R) (Tang and Chen, 2018)

generally not achievable by bacterial fermentation. However, more Acknowledgements

recently, through metabolic engineering it was possible to reorganize
the metabolic pathways of the PHA producers, suppressing non-useful This work has received funding from the European Union's Horizon
pathways and assuring the correct supply of sufficient precursors, so as 2020 research and innovation program under the Marie Skłodowska-
to achieve very low polydispersity indexes (Tsuge, 2016). Thus, sum­ Curie grant agreement No 812909 CODOBIO, within the Marie
marizing, mainly due to the large differences in process costs mentioned Skłodowska-Curie International Training Networks framework.
before, it is clear that the microbiological route is the one of choice for
any possible large-scale production. References
The situation is quite different when considering the currently
available industrial technologies for large scale production, the third Agus, J., Kahar, P., Hyakutake, M., Tomizawa, S., Abe, H., Tsuge, T., Satoh, Y., Tajima, K.,
factor in our list above, where the microbiological and chemical routes 2010. Unusual change in molecular weight of polyhydroxyalkanoate (PHA) during
cultivation of PHA-accumulating Escherichia coli. Polym. Degrad. Stab. 95,
are both in a very favourable position. The technology to carry out, at 2250–2254. https://doi.org/10.1016/J.POLYMDEGRADSTAB. 2010.09.009.
the industrial scale, fermentation and polymerization exists and it is Atlić, A., Koller, M., Scherzer, D., Kutschera, C., Grillo-Fernandes, E., Horvat, P., Chiellini,
widespread, and building a new production facility from scratch would E., Braunegg, G., 2011. Continuous production of poly([R]-3-hydroxybutyrate) by
Cupriavidus necator in a multistage bioreactor cascade. Appl. Microbiol. Biotechnol.
require a similar investment, since similar equipment is required in the 91, 295–304. https://doi.org/10.1007/s00253-011-3260-0.
two cases. Moreover, it would be relatively easy to adapt existing in­ Chek, M.F., Kim, S.-Y.Y., Mori, T., Arsad, H., Samian, M.R., Sudesh, K., Hakoshima, T.,
dustrial facilities to the bacterial or chemical production of PHA. To 2017. Structure of polyhydroxyalkanoate (PHA) synthase PhaC from
Chromobacterium sp. USM2, producing biodegradable plastics. Sci. Rep. 7. https://
achieve the same level of maturity with the enzymatic route, it would
doi.org/10.1038/s41598-017-05509-4.
require a tremendous effort due to the complexity of these processes. Chek, M.F., Kim, S.-Y., Mori, T., Tan, H.T., Sudesh, K., Hakoshima, T., 2020. Asymmetric
However, one might ask, if the technology is already there, why open-closed dimer mechanism of polyhydroxyalkanoate synthase PhaC. iScience 23,
101084. https://doi.org/10.1016/j.isci.2020.101084.
industrial units for the production of PHA and the use of PHA-based
Chen, G.-Q., Jiang, X.-R., 2017. Engineering bacteria for enhanced polyhydroxyalk­
plastics are not widely spread in our society? The answer is of course in anoates (PHA) biosynthesis. Synth. Syst. Biotechnol. 2, 192–197. https://doi.org/10.
the associated cost that, even for the most convenient case of the mi­ 1016/J.SYNBIO.2017.09.001.
crobiological route, is still too high to compete with that of traditional DiCosimo, R., McAuliffe, J., Poulose, A.J., Bohlmann, G., 2013. Industrial use of im­
mobilized enzymes. Chem. Soc. Rev. 42, 6437. https://doi.org/10.1039/c3cs35506c.
plastics of fossil origin. This justifies the large research effort currently Dietrich, K., Dumont, M.J., Del Rio, L.F., Orsat, V., 2017. Producing PHAs in the bioec­
devoted both in industry and in academia directed to the identification onomy — towards a sustainable bioplastic. Sustain. Prod. Consum. 9, 58–70. https://
of suitable and cheap raw materials, such as waste streams coming from doi.org/10.1016/j.spc.2016.09.001.
Gorke, J.T., Okrasa, K., Louwagie, A., Kazlauskas, R.J., Srienc, F., 2007. Enzymatic
other industrial or civil activities. This requires the definition of sui­ synthesis of poly(hydroxyalkanoates) in ionic liquids. J. Biotechnol. 132, 306–313.
table microorganisms, possibly properly engineered, and the realisation https://doi.org/10.1016/J.JBIOTEC.2007.04.001.
of a robust reacting system able to produce macromolecules with the Han, X., Satoh, Y., Tajima, K., Matsushima, T., Munekata, M., 2009. Chemo-enzymatic
synthesis of polyhydroxyalkanoate by an improved two-phase reaction system
required molecular characteristics, in terms of molecular mass and (TPRS). J. Biosci. Bioeng. 108, 517–523. https://doi.org/10.1016/J.JBIOSC.2009.06.
composition distributions. However, this is only one component of a 004.
successful industrial technology. More work needs to be done in the Hezayen, F.F., Steinbüchel, A., Rehm, B.H., 2002. Biochemical and enzymological
properties of the polyhydroxybutyrate synthase from the extremely halophilic ar­
direction of process engineering, for example using intensification chaeon strain 56. Arch. Biochem. Biophys. 403, 284–291. https://doi.org/10.1016/
techniques, including integrated continuous production units. S0003-9861(02)00234-5.
The case of other industries can be brought forward. For example, Hong, M., Chen, E.Y.-X., 2016a. Completely recyclable biopolymers with linear and cyclic
topologies via ring-opening polymerization of γ-butyrolactone. Nat. Chem. 8, 42–49.
the (bio)pharmaceutical industry is now taking the example of oil re­
https://doi.org/10.1038/nchem.2391.
fineries and other mechanical and chemical industries by investing into Hong, M., Chen, E.Y.-X., 2016b. Towards truly sustainable polymers: a metal-free re­
fully integrated and continuous production lines, leading to amazing cyclable polyester from biorenewable non-strained γ-Butyrolactone. Angew. Chem.
improvement in production costs and product quality (Karst et al., Int. Ed. 55, 4188–4193. https://doi.org/10.1002/anie.201601092.
Hori, Y., Suzuki, M., Yamaguchi, A., Nishishita, T., 1993. Ring-opening polymerization of
2018; Pfister et al., 2018). The same can be envisioned for the pro­ optically active β-Butyrolactone using distannoxane catalysts: synthesis of high mo­
duction of bioplastics. All current PHA producers rely, at least partially, lecular weight poly(3-hydroxybutyrate). Macromolecules 26, 5533–5534. https://
on batch or fed-batch technologies (Koller et al., 2018), and the in­ doi.org/10.1021/ma00072a037.
Hori, Y., Yamaguchi, A., Hagiwara, T., 1995. Chemical synthesis of high molecular weight
troduction of such new technologies, possibly in the frame of bior­ poly(3-hydroxybutyrate-co-4-hydroxybutyrate). Polymer (Guildf). 36, 4703–4705.
efineries (Dietrich et al., 2017) can significantly boost these natural https://doi.org/10.1016/0032-3861(95)96838-Y.
polymers in the industrial scale scenario. Hori, Y., Hideyuki, H., Hagiwara, T., 1999. Ring-opening copolymerization of (R)-β-

8
J. Medeiros Garcia Alcântara, et al. Biotechnology Advances 42 (2020) 107582

Butyrolactone with macrolide: a new series of poly(Hydroxyalkanoate)s. 2015. Production of polyhydroxyalkanoates from crude glycerol using recombinant
Macromolecules. https://doi.org/10.1021/MA981846E. 3537–2539. Escherichia coli. J. Polym. Environ. 23, 38–44. https://doi.org/10.1007/s10924-014-
Huang, G., Hollingsworth, R.I., 1998. An efficient synthesis of (R)-3-hydro­ 0681-8.
xytetradecanoic acid. Tetrahedron Asymmetry 9, 4113–4115. https://doi.org/10. Raza, Z.A., Abid, S., Banat, I.M., 2018. Polyhydroxyalkanoates: characteristics, produc­
1016/S0957-4166(98)00441-8. tion, recent developments and applications. Int. Biodeterior. Biodegradation 126,
Jia, K., Cao, R., Hua, D.H., Li, P., 2016. Study of class I and class III polyhydroxyalkanoate 45–56. https://doi.org/10.1016/J.IBIOD.2017.10.001.
(PHA) synthases with substrates containing a modified side chain. Rehm, B.H.A., 2003. Polyester synthases: natural catalysts for plastics. Biochem. J. 376,
Biomacromolecules 17, 1477–1485. https://doi.org/10.1021/acs.biomac.6b00082. 15–33. https://doi.org/10.1042/BJ20031254.
Jiang, G., Hill, D.J., Kowalczuk, M., Johnston, B., Adamus, G., Irorere, V., Radecka, I., Rehm, B.H.A., 2010. Bacterial polymers: biosynthesis, modifications and applications.
2016. Carbon sources for polyhydroxyalkanoates and an integrated biorefinery. Int. Nat. Rev. Microbiol. 8, 578–592. https://doi.org/10.1038/nrmicro2354.
J. Mol. Sci. https://doi.org/10.3390/ijms17071157. Rieth, L.R., Moore, D.R., Lobkovsky, E.B., Coates, G.W., 2002. Single-site β-diiminate zinc
Jmol: an open-source Java viewer for chemical structures in 3D, 2020. WWW Document. catalysts for the ring-opening polymerization of β-butyrolactone and β-valerolactone
n.d. URL. http://www.jmol.org/. to poly(3-hydroxyalkanoates). J. Am. Chem. Soc. 124, 15239–15248. https://doi.
Karst, D.J., Steinebach, F., Morbidelli, M., 2018. Continuous integrated manufacturing of org/10.1021/ja020978r.
therapeutic proteins. Curr. Opin. Biotechnol. 53, 76–84. https://doi.org/10.1016/j. Sagong, H.Y., Son, H.F., Choi, S.Y., Lee, S.Y., Kim, K.J., 2018. Structural insights into
copbio.2017.12.015. polyhydroxyalkanoates biosynthesis. Trends Biochem. Sci. https://doi.org/10.1016/
Kim, J., Kim, Y.-J., Choi, S.Y., Lee, S.Y., Kim, K.-J., 2017. Crystal structure of Ralstonia j.tibs.2018.08.005.
eutropha polyhydroxyalkanoate synthase C-terminal domain and reaction mechan­ Satoh, Y., Murakami, F., Tajima, K., Munekata, M., 2005. Enzymatic synthesis of poly(3-
isms. Biotechnol. J. 12, 1600648. https://doi.org/10.1002/biot.201600648. hydroxybutyrate-co-4-hydroxybutyrate) with CoA recycling using polyhydroxyalk­
Koller, M., Braunegg, G., 2018. Advanced approaches to produce polyhydroxyalkanoate anoate synthase and acyl-CoA synthetase. J. Biosci. Bioeng. 99, 508–511. https://doi.
(PHA) biopolyesters in a sustainable and economic fashion. EuroBiotech J. 2, 89–103. org/10.1263/JBB.99.508.
https://doi.org/10.2478/ebtj-2018-0013. Steinbüchel, A., Lütke-Eversloh, T., 2003. Metabolic engineering and pathway construc­
Koller, M., Marsalek, L., 2015. Principles of glycerol-based polyhydroxyalkanoate pro­ tion for biotechnological production of relevant polyhydroxyalkanoates in micro­
duction. Appl. Food Biotechnol. 2, 3–10 doi:10.22037/AFB.V2I4.8270. organisms. Biochem. Eng. J. https://doi.org/10.1016/S1369-703X(03)00036-6.
Koller, M., Niebelschütz, H., Braunegg, G., 2013a. Strategies for recovery and purification Stubbe, J., Tian, J., 2003. Polyhydroxyalkanoate (PHA) homeostasis: the role of the PHA
of poly[(R )-3-hydroxyalkanoates] (PHA) biopolyesters from surrounding biomass. synthase. Nat. Prod. Rep. 20, 445. https://doi.org/10.1039/b209687k.
Eng. Life Sci. 13, 549–562. https://doi.org/10.1002/elsc.201300021. Su, L., Lenz, R.W., Tajagi, Y., Zhang, S., Goodwin, S., Zhong, L., Martin, D.P., 2000.
Koller, M., Salerno, A., Braunegg, G., 2013b. Polyhydroxyalkanoates: basics, production Enzymatic polymerization of (R)-3-hydroxyalkanoates by a bacterial polymerase.
and applications of microbial biopolyesters. In: Bio-Based Plastics. John Wiley & Sons Macromolecules 229–231. https://doi.org/10.1021/MA990949+.
Ltd, Chichester, UK, pp. 137–170. https://doi.org/10.1002/9781118676646.ch7. Sudesh, K., Abe, H., Doi, Y., 2000. Synthesis, structure and properties of poly­
Koller, M., Koller, Martin, 2018. A review on established and emerging fermentation hydroxyalkanoates: biological polyesters. Prog. Polym. Sci. 25, 1503–1555. https://
schemes for microbial production of Polyhydroxyalkanoate (PHA) biopolyesters. doi.org/10.1016/S0079-6700(00)00035-6.
Fermentation 4, 30. https://doi.org/10.3390/fermentation4020030. Tajima, K., Satoh, Y., Nakazawa, K., Tannai, H., Erata, T., Munekata, M., Kamachi, M.,
Kumar, M., Singhal, A., Verma, P.K., Thakur, I.S., 2017. Production and characterization Lenz, R.W., 2004. Chemoenzymatic synthesis of poly(3-hydroxybutyrate) in a water-
of Polyhydroxyalkanoate from lignin derivatives by Pandoraea sp. ISTKB. ACS Omega organic solvent two-phase system. Macromolecules 4544–4546. https://doi.org/10.
2, 9156–9163. https://doi.org/10.1021/acsomega.7b01615. 1021/MA049828M.
Kumar, P., Maharjan, A., Jun, H., Kim, B.S., 2019. Bioconversion of lignin and its deri­ Tajima, K., Satoh, Y., Satoh, T., Itoh, R., Han, X., Taguchi, S., Kakuchi, T., Munekata, M.,
vatives into polyhydroxyalkanoates: challenges and opportunities. Biotechnol. Appl. 2009. Chemo-enzymatic synthesis of poly(lactate- co -(3-hydroxybutyrate)) by a
Biochem. 66, 153–162. https://doi.org/10.1002/bab.1720. lactate-polymerizing enzyme. Macromolecules 42, 1985–1989. https://doi.org/10.
Lecomte, P., Jérôme, C., 2012. Recent developments in ring-opening polymerization of 1021/ma802579g.
lactones. Adv. Polym. Sci. https://doi.org/10.1007/12_2011_144. Tan, G.-Y.A., Chen, C.-L., Li, L., Ge, L., Wang, L., Razaad, I.M.N., Li, Y., Zhao, L., Mo, Y.,
Lee, S.Y., Chang, H.N., 1995. Production of poly(hydroxyalkanoic acid). Adv. Biochem. Wang, J.-Y., 2014. Start a research on biopolymer polyhydroxyalkanoate (PHA): a
Eng. Biotechnol. https://doi.org/10.1007/bfb0102315. review. Polymers (Basel). 6, 706–754. https://doi.org/10.3390/polym6030706.
Leong, Y.K., Show, P.L., Ooi, C.W., Ling, T.C., Lan, J.C.-W., 2014. Current trends in Tan, D., Yin, J., Chen, G.-Q., 2017. Production of polyhydroxyalkanoates. Curr. Dev.
polyhydroxyalkanoates (PHAs) biosynthesis: insights from the recombinant Biotechnol. Bioeng. 655–692. https://doi.org/10.1016/B978-0-444-63662-1.
Escherichia coli. J. Biotechnol. 180, 52–65. https://doi.org/10.1016/J.JBIOTEC. 00029-4.
2014.03.020. Tang, X., Chen, E.Y.-X., 2018. Chemical synthesis of perfectly isotactic and high melting
Li, R., Zhang, H., Qi, Q., 2007. The production of polyhydroxyalkanoates in recombinant bacterial poly(3-hydroxybutyrate) from bio-sourced racemic cyclic diolide. Nat.
Escherichia coli. Bioresour. Technol. 98, 2313–2320. https://doi.org/10.1016/J. Commun. 9, 2345. https://doi.org/10.1038/s41467-018-04734-3.
BIORTECH.2006.09.014. Teh, A.H., Chiam, N.C., Furusawa, G., Sudesh, K., 2018. Modelling of polyhydroxyalk­
Licciardello, G., Catara, A.F., Catara, V., 2019. Production of polyhydroxyalkanoates and anoate synthase from Aquitalea sp. USM4 suggests a novel mechanism for polymer
extracellular products using pseudomonas corrugata and P. mediterranea: a review. elongation. Int. J. Biol. Macromol. 119, 438–445. https://doi.org/10.1016/j.
Bioengineering 6, 105. https://doi.org/10.3390/bioengineering6040105. ijbiomac.2018.07.147.
Liu, Y.-C., Lin, C.-H., Ko, B.-T., Ho, R.-M., 2010. Ring-opening polymerization of β-bu­ Thomson, N., Roy, I., Summers, D., Sivaniah, E., 2009. In vitro production of poly­
tyrolactone catalyzed by efficient magnesium and zinc complexes derived from tri­ hydroxyalkanoates: achievements and applications. J. Chem. Technol. Biotechnol.
dentate anilido-aldimine ligand. J. Polym. Sci. Part A Polym. Chem. 48, 5339–5347. 85, 760–767. https://doi.org/10.1002/jctb.2299.
https://doi.org/10.1002/pola.24334. Tian, J., Sinskey, A.J., Stubbe, J., 2005. Class III polyhydroxybutyrate synthase: in­
Lohmeijer, B.G.G., Pratt, R., Leibfarth, F.W., Logan, J., Long, D.A.P., Dove, A., Nederberg, volvement in chain termination and Reinitiation. Biochemistry 44, 8369–8377.
F., Choi, J., Wade, C., Waymouth, R.M., Hedrick, J.L., 2006. Guanidine and Amidine https://doi.org/10.1021/BI050331U.
Organocatalysts for Ring-Opening Polymerization of Cyclic Esters. https://doi.org/ Tokiwa, Y., Ugwu, C.U., 2007. Biotechnological production of (R)-3-hydroxybutyric acid
10.1021/MA0619381. monomer. J. Biotechnol. 132, 264–272. https://doi.org/10.1016/J.JBIOTEC.2007.
Makiguchi, K., Satoh, T., Kakuchi, T., 2011. Diphenyl phosphate as an efficient cationic 03.015.
organocatalyst for controlled/living ring-opening polymerization of δ-valerolactone Tsuge, T., 2016. Fundamental factors determining the molecular weight of poly­
and ε-caprolactone. Macromolecules 44, 1999–2005. https://doi.org/10.1021/ hydroxyalkanoate during biosynthesis. Polym. J. 48, 1051–1057. https://doi.org/10.
ma200043x. 1038/pj.2016.78.
Meng, D.C., Shen, R., Yao, H., Chen, J.C., Wu, Q., Chen, G.Q., 2014. Engineering the Vandi, L., Chan, C.M., Werker, A., Richardson, D., Laycock, B., Pratt, S., 2018. Wood-PHA
diversity of polyesters. Curr. Opin. Biotechnol. https://doi.org/10.1016/j.copbio. Composites : Mapping Opportunities 1–15. https://doi.org/10.3390/
2014.02.013. polym10070751.
Możejko-Ciesielska, J., Kiewisz, R., 2016. Bacterial polyhydroxyalkanoates: still fabulous? Wang, Y., Yin, J., Chen, G.-Q., 2014. Polyhydroxyalkanoates, challenges and opportu­
Microbiol. Res. 192, 271–282. https://doi.org/10.1016/J.MICRES.2016.07.010. nities. Curr. Opin. Biotechnol. 30, 59–65. https://doi.org/10.1016/J.COPBIO.2014.
Noyori, R., Ohkuma, T., Kitamura, M., Takaya, H., Sayo, N., Kumobayashi, H., 06.001.
Akutagawa, S., 1987. Asymmetric hydrogenation of .beta.-keto carboxylic esters. a Wittenborn, E.C., Jost, M., Wei, Y., Stubbe, J.A., Drennan, C.L., 2016. Structure of the
practical, purely chemical access to .beta.-hydroxy esters in high enantiomeric purity. catalytic domain of the class I polyhydroxybutyrate synthase from Cupriavidus ne­
J. Am. Chem. Soc. 109, 5856–5858. https://doi.org/10.1021/ja00253a051. cator. J. Biol. Chem. 291, 25264–25277. https://doi.org/10.1074/jbc.M116.756833.
Penloglou, G., Chatzidoukas, C., Kiparissides, C., 2012. Microbial production of poly­ Zinn, M., Weilenmann, H.U., Hany, R., Schmid, M., Egli, T., 2003. Tailored synthesis of
hydroxybutyrate with tailor-made properties: an integrated modelling approach and poly([R]-3-hydroxybutyrateco-3-hydroxyvalerate) (PHB/HV) in Ralstonia eutropha
experimental validation. Biotechnol. Adv. 30, 329–337. https://doi.org/10.1016/J. DSM 428. Acta Biotechnol. 23, 309–316.
BIOTECHADV.2011.06.021. Zou, H., Shi, M., Zhang, T., Li, Lei, Li, Liangzhi, Xian, M., 2017. Natural and engineered
Pfister, D., Nicoud, L., Morbidelli, M., 2018. Continuous Biopharmaceutical Processes, polyhydroxyalkanoate (PHA) synthase: key enzyme in biopolyester production. Appl.
Continuous Biopharmaceutical Processes. Cambridge University Presshttps://doi. Microbiol. Biotechnol. 101, 7417–7426. https://doi.org/10.1007/s00253-017-
org/10.1017/9781108332897. 8485-0.
Phithakrotchanakoon, C., Champreda, V., Aiba, S., Pootanakit, K., Tanapongpipat, S.,