227

REVIEW

Unique multifunctional HSD17B4 gene product:

17â-hydroxysteroid dehydrogenase 4 and D-3-hydroxyacyl-
coenzyme A dehydrogenase/hydratase involved in
Zellweger syndrome
Y de Launoit1 and J Adamski2
1
Virology Unit, Faculty of Medicine, CP 614, Free University Brussels, 808 route de Lennik,
1070 Brussels, Belgium
2
GSF National Research Center for Environment and Health, Institute of Mammalian Genetics,
Genome Analysis Center, 85764 Neuherberg, Germany
(Requests for offprints should be addressed to Y de Launoit)

ABSTRACT
Six types of human 17â-hydroxysteroid dehydro- able to perform the dehydrogenase reaction not only
genases catalyzing the conversion of estrogens and with steroids at the C17 position but also with
androgens at position C17 have been identified so -3-hydroxyacyl-coenzyme A (CoA). The enzyme
far. The peroxisomal 17â-hydroxysteroid dehydro- is not active with -stereoisomers. The central part
genase type 4 (17â-HSD 4, gene name HSD17B4) of the 80 kDa protein (amino acids 324–596)
catalyzes the oxidation of estradiol with high catalyzes the 2-enoyl-acyl-CoA hydratase reaction
preference over the reduction of estrone. The with high efficiency. The C-terminal part of the
highest levels of 17â-HSD 4 mRNA transcription 80 kDa protein (amino acids 597–737) facilitates the
and specific activity are found in liver and kidney transfer of 7-dehydrocholesterol and phosphatidyl-
followed by ovary and testes. A 3 kb mRNA codes choline between membranes in vitro. The
for an 80 kDa (737 amino acids) protein featuring HSD17B4 gene is stimulated by progesterone, and
domains which are not present in the other ligands of PPARá (peroxisomal proliferator acti-
17â-HSDs. The N-terminal domain of 17â-HSD 4 vated receptor alpha) such as clofibrate, and is
reveals only 25% amino acid similarity with the down-regulated by phorbol esters. Mutations in the
other types of 17â-HSDs. The 80 kDa protein is HSD17B4 lead to a fatal form of Zellweger
N-terminally cleaved to a 32 kDa enzymatically syndrome.
active fragment. Both the 80 kDa and the Journal of Molecular Endocrinology (1999) 22, 227–240
N-terminal 32 kDa (amino acids 1–323) protein are

INTRODUCTION names are used frequently. The complexity of

related enzymes was reviewed recently elsewhere
The genetics and biochemistry of steroid conversion (Reed 1991, Andersson 1995, Poutanen et al. 1995,
have recently been developing very fast. There is a Penning 1997).
growing body of reports showing the involvement of
steroid converting enzymes in several human
disorders. In this review we will focus on the most MULTIPLE DEHYDROGENASES CONVERT
unique protein among the 17â-hydroxysteroid STEROIDS AT POSITION C17
dehydrogenases, the type 4 enzyme. For the sake of
clarity we will use the abbreviation 17â-HSD 4, Biological potency of estrogens and androgens is
although, as will be discussed later, alternative regulated by conversions at position C17 by

Journal of Molecular Endocrinology (1999) 22, 227–240 Online version via http://www.endocrinology.org
0952–5041/99/022–227  1999 Society for Endocrinology Printed in Great Britain
228    and   · 17â-HSD 4

 1. Human 17â-hydroxysteroid dehydrogenases

Subcellular Mass Km nmol/mg Catalysis

Tissue localization (kDa) Best substrate (ìM) protein in vivo
Enzyme
17â-HSD 1 Placenta Soluble 34 Estrone 8·6 0·02 Reduction
(Peltoketo et al. 1988) Ovary Estradiol 5·9
17â-HSD 2 Placenta Microsomal 43 Testosterone 0·4 0·22 Oxidation
(Wu et al. 1993) Liver Estradiol 0·2
17â-HSD 3 Testis Microsomal 35 Androstenedione 0·5 0·09 Reduction
(Geissler et al. 1994) Estrone 0·5
17â-HSD 4 Liver Peroxisomal 80 Estradiol 0·2 0·14 Oxidation
(Adamski et al. 1995) Kidney Ä5-Androstenediol 0·4
17â-HSD 5 Liver Microsomal 34 5á-Dihydrotestosterone 19·0 2·45 Reduction
(Lin et al. 1997; Zhang et al. 1995) Testis 3á-Androstanediol 17·1 4·83 Oxidation

17â-hydroxysteroid dehydrogenases (17â-HSDs). et al. 1997). Later, we will discuss the issue of
Several enzymes with close substrate specificity multifunctionality in more detail.
participate in that process. The identification and The oxidative 17â-HSD activity found in human
characterization of individual human 17â-HSDs uterus endometrium could not be unequivocally
was limited by the minute amounts of tissue ascribed to the known enzymes (Tseng & Mazella
available for purification. However, analyses per- 1981). Attempts to isolate the endometrial 17â-
formed with homogenates or with subcellular HSD from the particulate fraction of homogenates
fractions allowed the kinetical differentiation of (Pollow et al. 1976) resulted in a 40-fold enrich-
several enzymes such as the soluble 17â- ment. However, because of difficulties in collecting
hydroxysteroid oxidoreductase of placenta or the and the paucity of starting material, the enriched
structure-associated 17â-estradiol dehydrogenase fractions were not applied to amino acid sequencing
of uterus epithelium (Engel & Groman 1974, Tseng and antibody production. Entenmann et al. (1980)
& Gurpide 1974, Tseng & Mazella 1981). Before discovered oxidative activity for 17â-estradiol in
molecular biology techniques became widespread, porcine endometrium. This microsomal activity
the readily available human placenta allowed the revealed comparable kinetical parameters (NAD+-
purification of the first human 34 kDa 17â- dependency, Km less than 1 µM). The parameters
hydroxysteroid dehydrogenase (17â-HSD 1), which suggested a role in the inactivation of hormones.
became a model for studies of steroid converting
enzymes. After cloning and elucidation of the gene
structure (Peltoketo et al. 1988, Luu-The et al. PURIFICATION OF PORCINE 17â-ESTRADIOL
1989), it represents the best characterized human DEHYDROGENASE 4
steroid dehydrogenase. In vivo this enzyme
participates in the synthesis of steroids via a To accomplish identification of a novel 17â-HSD
reductive pathway. However, detailed kinetical we have used a pig (Sus scrofa) model. The
studies (Blomquist et al. 1985) have shown that the epithelial layer of the porcine uterus could easily
placenta expresses additional HSDs and one of be collected at a preparative scale by curettage
them, namely 17â-HSD 2, was cloned (Wu et al. (Sierralta et al. 1978). The 17â-estradiol dehydro-
1993) (Table 1). This enzyme is a 43 kDa genase activity had to be solubilized from the
microsomal dehydrogenase revealing a twofold particulate fraction (Adamski et al. 1992a). Extracts
higher rate of oxidation than reduction for both were applied to DEAE-Sepharose, depleted of free
estrogens and androgens. A further enzyme, detergent on Amberlite XAD-2 and further purified
17â-HSD 3, is most abundant in testes and by affinity chromatography on blue Sepharose
represents a transmembrane microsomal 35 kDa (Adamski et al. 1992b). Further purification on
protein with a strong preference for the reduction of butyl Sepharose resulted in two products rich in
androgens (Geissler et al. 1994). Mouse and human 17â-estradiol dehydrogenase activity: a major mod-
17â-HSD type 5 has not yet been fully character- erately hydrophobic fraction (EDH) and a minor
ized (Deyashiki et al. 1995, Zhang et al. 1995). very hydrophobic fraction (VHF). They were
However, this enzyme has wide tissue distri- processed in parallel by gel filtration and ion-
bution and has also been identified as 3á-HSD type exchange chromatography on Mono S. The EDH
2 (Lin et al. 1997) (for a review see Penning fraction was purified to homogeneity and revealed a
Journal of Molecular Endocrinology (1999) 22, 227–240
 17â-HSD 4 ·    and   229

 2. Kinetic parameters of purified porcine this fragment as a probe, a 3 kb cDNA was isolated
17â-estradiol dehydrogenase 4 from a porcine ëZAP kidney cDNA library. The
E2<E1 E2=E1
sequence was later confirmed in porcine uterus
(Leenders et al. 1994a). The cDNA coded for a
Parameter protein of 80 kDa consisting of 737 aa and was not
Optimal pH 7·8 6·6 similar to any known steroid dehydrogenases.
Km for steroid 0·22 ìM 1·10 mM
Best cofactor NAD + NADPH
About 70% of its amino acid sequence was already
Km for cofactor 44 mM 21 mM known from peptides of the 32 and 80 kDa proteins.
Screening of human and mouse cDNA ëgt11
E2, 17â-estradiol; E1, estrone. libraries of liver was performed with porcine
enzyme cDNA. Novel 3 kb cDNAs were identified
which coded for proteins of 735 and 736 aa
single band at 32 kDa in the denaturing SDS- representing the human and mouse counterparts
PAGE. The VHF was a mixture of proteins of 32, respectively of the porcine enzyme (Adamski et al.
45 and 80 kDa (Adamski et al. 1992b). 1995, Normand et al. 1995).

CHARACTERIZATION OF PURIFIED PARALLEL WAYS OF HSD17B4

PORCINE 17â-HSD 4 IDENTIFICATION

Both purification products, the EDH and the VHF, The product of the HSD17B4 gene, an 80 kDa
show the same Km for steroids and cofactors as protein, was detected almost in parallel by other
those measured with two-substrate kinetics in the groups. During studies on peroxisomal â-oxidation
particulate fraction of homogenates of porcine of pristanic acid and bile acid intermediates in rat
uterus epithelium (Table 2). They reveal an ordered and man, 80 kDa -specific hydroxyacyl-coenzyme
mechanism of reaction with the cofactor binding A (CoA) dehydrogenase/hydratase (also called
first (Adamski et al. 1992b, Marks 1992). EDH and multifunctional protein 2–MFP2) were purified
VHF also share the same substrate specificity, (Novikov et al. 1994, Qin et al. 1997b), character-
which is highest for 17â-estradiol and, unexpect- ized and cloned (Dieuaide-Noubhani et al. 1996a,
edly, comparably high for 5-androstene-3â,17â-diol 1997a,b, Jiang et al. 1996, 1997, Novikov et al.
(Km=0·2 µM). Other androgens or progestagens are 1997). The amino acid sequence of the human
not converted. enzyme was identical to that of 17â-HSD 4
The molecular mass was estimated in denaturing (Adamski et al. 1995).
SDS-PAGE and under native conditions by gel Guinea pig enzyme was cloned from a cDNA
filtration and density gradient centrifugation library while using rat 17â-HSD 4 as a probe (Caira
(Adamski et al. 1992b). The 32 kDa protein has a et al. 1998).
capability of forming dimers with an apparent Another study pursued the identification of
molecular weight of 75 kDa (Carstensen et al. 1996). cDNAs up-regulated in rat by peroxisomal prolif-
The interactions in the VHF are more complex. erators such as WY14,643 (Corton et al. 1996,
Both gel filtration analyses and density gradient 1997). One of the affected proteins turned out to be
centrifugation revealed the presence of a hetero- the rat ortholog of previously known porcine, mouse
geneous complex with at least two molecular mass and human 17â-HSD 4.
forms of 170 kDa and 240 kDa. Unusual was the way that chicken protein was
identified (Kobayashi et al. 1997). A monoclonal
antibody 3b5 was prepared against isolated retinal
CLONING OF THE PORCINE, HUMAN AND pigment epithelium cells. The antibody recognized
MOUSE 17â-ESTRADIOL DEHYDROGENASE a 75 kDa protein on western blots of retinal pigment
TYPE 4 epithelium. This antibody was then used to screen a
lambda expression library. This approach resulted
With degenerated PCR-primers, designed accord- in the identification of a full length cDNA clone
ing to partial amino acid (aa) sequence of the 32 kDa coding for the chicken 17â-HSD 4.
protein, a fragment of 405 base pairs (bp) was Amino acid identities between 17â-HSD 4 from
amplified from porcine endometrium cDNA. It had different species are very high, around 80% (Fig. 1).
a single open reading frame coding for an amino Several hydroxysteroid dehydrogenases could be
acid sequence which was identical to the 32 kDa traced to a common ancestor close to 3-ketoacyl-
protein as confirmed by Edman degradation. Using acyl carrier protein reductase of Eschericia coli
Journal of Molecular Endocrinology (1999) 22, 227–240
230    and   · 17â-HSD 4

 1. Alignment of amino acid sequences of 17â-HSD 4 from different species. Sequences were ordered to
obtain maximum similarity with ClustalX software. Identical amino acids are grey shaded, similar amino acids are
boxed. Accession numbers for protein sequences are: mouse, P51660; chicken, U77911; rat, U37486; human, P51659;
porcine, X78201; guinea pig, Y13623.

(Baker 1996). Phylogenetic studies suggest a GENE STRUCTURE

common ancestor for the 17â-HSD 1 and 2,
whereas the 17â-HSD 3 and 4 are examples of The chromosomal assignment and structure of the
convergence (Baker 1994, 1996). In particular human HSD17B4 gene were determined recently
17â-HSD 4 appears to be a very ancient protein (Leenders et al. 1996a, Novikov et al. 1997,
whose functionality and amino acid sequence are Leenders et al. 1999). The HSD17B4 gene localizes
strongly conserved. Similarities to proteins of to chromosome 5q2 and was found to be more than
primitive organisms will be discussed in the gene 100 kbp in length (Fig. 2). The size of the exons
structure section. ranges from 21 bp (exon 5) to 286 bp (exon 13). The
smallest intron with 100 bp was found between
exons 5 and 6.
NAMES USED The N-terminal domain of 17â-HSD 4 reveals
functional similarity and amino acid homology
Due to multifunctionality of the encoded protein, (54%) to the family of short chain alcohol
the HSD17B4 gene product was discovered by dehydrogenases (SCAD) (Persson et al. 1991,
many groups using different strategies. Because of Leenders et al. 1994b, Jörnvall et al. 1995) and the
that there are several names being used. Table 3 central domain to several fatty acid hydratases
gives a summary of them. The references given (HDE), especially to that of Candida tropicalis (40%)
should be understood as a source for the represen- (Baker 1996). However, neither sizes of the
tative description. Quite often different synonyms corresponding genes nor the exon/intron structures
are used in the same reports, depending on the are analogous. As discussed later, the 17â-HSD 4 is
functionality analyzed. an exception in the group of 17â-HSDs, since it
Journal of Molecular Endocrinology (1999) 22, 227–240
 17â-HSD 4 ·    and   231

 3. Names used for the HSD17B4 gene product

Reference
Name
17â-hydroxysteroid dehydrogenase type 4 or 17â-HSD 4 Leenders et al. (1994)
Multifunctional protein 2 or MFP-2 Dieuaide-Noubhani et al. (1996)
Peroxisomal multifunctional enzyme 2 or perMFE2 Qin et al. (1997)
-specific multifunctional protein 2 Caira et al. (1998)
-3-hydroxyacyl-CoA dehydrogenase Dieuaide-Noubhani et al. (1997); Novikov et al. (1994)
2-trans-enoyl-CoA hydratase Dieuaide-Noubhani et al. (1997)
-bifunctional protein Jiang et al. (1997)
-3-hydroxyacyl-CoA dehydratase/ Jiang et al. (1996)
-3-hydroxyacyl dehydrogenase bifunctional protein

 2. Structure of 17â-HSD 4 protein and HSD17B4 gene. A schematic structure of

the human 17â-HSD IV protein is given above that of the gene. Arrows point to the
position of the G16S mutation, cleavage site and C-terminal peroxisomal targeting signal.
The black boxes representing exons are drawn to scale according to the indicated 10 kb
marker. The sizes of introns are given above: introns smaller than 7 kb are drawn to scale,
larger introns are represented by broken lines. The numbers of the exons are below the
boxes. Exons 1 and 24 consist of translated (black box) and untranslated (grey box) regions.
Accession numbers for the nucleotide sequence are AF057720-AF057740.

consists of 3 domains. Only the N-terminal domain functionality as the C-terminal domain (sterol
of 320 amino acids participates in steroid metab- carrier protein (SCP) 2) of the SCPX protein (Ohba
olism and is coded by 12 exons spanning about et al. 1994, Leenders et al. 1996b). Interestingly, the
40 kbp. In comparison to other 17â-hydroxysteroid gene structure of the last 3 exons is also similar.
dehydrogenases, the SCAD domain of the This observation supports the hypothesis that the
HSD17B4 gene is much bigger than the 17â-HSD 1 HSD17B4 gene is the result of a gene fusion.
(6 exons on 3·3 kbp) (Luu-The et al. 1990), close to
the 17â-HSD 2 (7 exons on 40 kbp) (Labrie et al.
1995) and smaller than the 17â-HSD 3 (11 exons on PROCESSING OF 80 kDa PROTEIN
60 kbp) (Geissler et al. 1994).
The amino acid sequence C-terminal domain of Some peroxisomal proteins (SCP2, 3-ketoacyl-CoA
17â-HSD 4 reveals 40% identity and the same thiolase) are cleaved from larger precursors in the
Journal of Molecular Endocrinology (1999) 22, 227–240
232    and   · 17â-HSD 4

 4. Distribution of 17â-estradiol dehydrogenase activity in porcine

tissues

E2<E1 E2=E1
(ìU/mg) (ìU/mg) Immunocytochemistry
Tissue
Liver 687·5 5·6 Hepatocytes
Kidney 336·4 0·7 Epithelium of proximal tubuli
Ovary 293·9 1·5 Granulosa cells
Lung 185·4 0·8 Bronchial epithelium
Testes 69·2 1·8 Leydig cells
Uterus 30·1 1·2 Luminal and glandular epithelium
Skeletal muscle 10·9 ND Myocytes
Prostate 0·8 0·1 Epithelium
Blood erythrocytes ND ND ND

E2, 17â-estradiol; E1, estrone; ND, not detected.

course of translocation (Swinkels et al. 1991). A Highest activities are found in liver and kidney
protease present in peroxisomes has been proposed followed by uterus, lung, ovary and testes (Table 4).
to recognize the sequence Ala-[AlaVal]-Pro (Mori Immunohistochemical analysis of tissues showing
et al. 1991). The 80 kDa full length HSD17B4 gene low 17â-HSD 4 mRNA expression, such as brain,
product is N-terminally cleaved, probably after lung and uterus, reveals that the enzyme is present
the sequence Ala320-Ala-Pro-Ser324, to a 32 kDa in specific cells within these organs. In the rat and
fragment representing a SCAD domain (Leenders mouse cerebellum 17â-HSD 4 is confined to
et al. 1994b). Such processing has also been Purkinje cells, and the expression is also seen in the
observed in mice and rats (Novikov et al. 1994, anterior pituitary (Normand et al. 1995). There is
Normand et al. 1995, Dieuaide-Noubhani et al. high expression of 17â-HSD 4 in chick retinal
1997b) although the protease recognition motif was pigment epithelium but not in other parts of the eye
not conserved. The meaning of this cleavage for (Kobayashi et al. 1997). In the lung the bronchial
17â-HSD 4 is not known. Interestingly, the extent epithelium expresses high levels of 17â-HSD 4
of processing of 80 kDa into 32 kDa varies among (Möller et al. 1999) and in the uterus the protein is
porcine tissues (Adamski et al. 1997). Western blot present in luminal and glandular epithelium and not
analyses of porcine organs revealed that target in stromal cells (Husen et al. 1994).
tissues (uterus and mammary glands) show high A slightly different expression pattern was seen in
processing. In these tissues the 32 kDa form human tissues. The highest mRNA level of human
dominates. On the other hand, non-target tissues 17â-HSD 4 was observed in liver, followed by
participating in â-oxidation of fatty acids (such as heart, prostate and testis. Moderate expression
liver or kidney) showed low processing. Neverthe- occurred in lung, skeletal muscle, kidney, pancreas,
less, in either tissue both proteins are present. The thymus, ovary, intestine and term placenta. Weak
differential processing raised the question whether signals were observed in brain, spleen, colon and
the release of the 32 kDa fragment from the 80 kDa lymphocytes. In all cases only a single band at 3 kb
protein is an activation step for 17â-hydroxysteroid was detectable (Adamski et al. 1995). The wide
dehydrogenase. However, both 80 and 32 kDa distribution of 17â-HSD 4 may in part explain
proteins had comparable kinetical parameters after oxidative activities measured in human tissues
transient expression in HEK 293 cells or puri- (Martel et al. 1992). The expression of 17â-HSD 4
fication of recombinant proteins from E. coli is in contrast with that of 17â-HSD 1 and 2 which
(Leenders et al. 1996b, Adamski et al. 1997). Also are predominantly seen in the placenta.
the preferred reaction direction, i.e. oxidation, Several human cancer cell lines also express
remained unchanged for purified, recombinant and 17â-HSD 4. The estrogen receptor positive mam-
transiently expressed proteins. mary cell line, T47D, expresses more 17â-HSD
4 mRNA transcript than BT-20, MDA-MB-453
and MDA-MB-231 cell lines which are estrogen
TISSUE DISTRIBUTION receptor negative. The megakaryotic cell line,
DAMI, reveals a very high level of 17â-HSD 4
In the porcine tissues studied, the oxidation of mRNA while less is present in hepatocellular
17â-estradiol predominates over the reduction. carcinoma HEP-G2 and early embryonic Tera-1
Journal of Molecular Endocrinology (1999) 22, 227–240
 17â-HSD 4 ·    and   233

cell lines (Adamski et al. 1995). The prostate cancer a down-regulation (instead of up-regulation) of the
cell lines DU145, LNCaP and PC3 express expression of 17â-HSD 4 (Caira et al. 1998).
17â-HSD 4 but not 17â-HSD 1 or 3 (Carruba et al.
1997, Castagnetta et al. 1997). Interestingly,
17â-HSD 2, which appears to be the principal SUBCELLULAR DISTRIBUTION OF PORCINE
isozyme expressed in the prostate, is present only in 17â-HSD 4
PC3 cells (Delos et al. 1995, 1998, Elo et al. 1996,
Castagnetta et al. 1997). Most probably, 17â-HSD Immunocytochemical and immunofluorescence
2 is the main protective dehydrogenase in the studies in porcine uterus restricted porcine 17â-
prostate. HSD 4 to luminal and glandular epithelium (Husen
et al. 1994) similar to analyses of human endo-
metrium (Scublinsky et al. 1976, Mäentausta et al.
REGULATION OF HSD17B4 1991). However, the staining in the cytoplasm was
not diffuse but showed a punctuate appearance. The
Porcine 17â-hydroxysteroid dehydrogenase 4 is the intensity of the monoclonal antibody F1-peroxidase
first peroxisomal enzyme known to be stimulated by staining followed the changes in porcine 17â-
progesterone as checked by mRNA expression and hydroxysteroid dehydrogenase activity. It was
immunohistochemistry (Kaufmann et al. 1995). In raised fourfold after day 5 of the ovarian cycle and
the early 1970s the hormone was observed to rapidly decreased after day 17 in a manner similar to
increase estradiol oxidation in human tissues (Tseng the levels of progesterone. On day 4 faint spots of
& Gurpide 1975, 1979). Later it was shown to fluorescence appeared in the cytoplasm of the
up-regulate 17â-HSD 1 in human breast cancer glandular epithelium. The spots accumulated at
cells (Poutanen et al. 1990) and to increase mRNA the cell bases between days 11 and 17 (luteal phase)
expression of 17â-HSD 2 in human endometrium and disappeared within one day. The pattern of
(Casey et al. 1994). immunofluorescence staining suggested that porcine
Although 17â-HSD 1 is supposed to participate 17â-HSD 4 is localized in vesicles. The latter have
in the synthesis and 17â-HSD 4 in the inactivation been isolated from porcine uterus epithelium
of steroids there is a simultaneous expression of homogenates by sequential density gradients of
both enzymes in gonads (Luu-The et al. 1990, isopycnic 30% Percoll and linear 0·3–2 M sucrose in
Carstensen et al. 1996). However, the correspond- vertical rotors (Adamski et al. 1987, Adamski 1991,
ing regulatory pathways of protein kinase C are Adamski et al. 1993). The vesicles harboring the
not controlled in the same way. In vitamin 17â-HSD activity equilibrated at a density of
D-differentiated human leukemia THP 1 cells, 1·18 g/ml, were 120–200 nm in diameter, revealed
17â-HSD 4 mRNA was stimulated twofold by a moderate electron–dense matrix bounded by a
dexamethasone but it was completely down- single membrane and were morphologically and
regulated by phorbol esters (Jakob et al. 1995, enzymatically distinct from mitochondria, lyso-
1997). This is in contrast to the dose- and somes, fragments of plasma membrane, endoplas-
time-dependent increase in gene transcripts of mic reticulum and the Golgi apparatus. In
17â-HSD 1 under similar treatment (Tremblay & immunogold electron microscopy the labeling with
Beaudoin 1993). monoclonal antibody F1 (recognizing the 32 kDa
The recently purified and cloned rat 80 kDa and the 80 kDa protein) and W1 (reacting with
homolog of human 17â-HSD 4 is up-regulated by 32 kDa only) confirmed that all forms of the enzyme
peroxisomal proliferators such as clofibrate and WY are present in the same vesicles, both in tissue and
14,643 (Novikov et al. 1994, Corton et al. 1995). In in the isolated fraction (Adamski et al. 1993).
the PPARá (peroxisomal-proliferator activated re- Several clues pointed to the identity of the
ceptor alpha) knock-out mice the expression of 17â-HSD 4 containing vesicles as peroxisomes: (1)
17â-HSD 4 is low and does not change after the morphology and density is similar to that of
treatment with WY 14,643 (Aoyama et al. 1998). peroxisomes, (2) the 80 kDa primary transcript
The 80 kDa protein seems to be controlled by features the peroxisomal targeting signal Ala-Lys-
modulators of both steroid and fatty acid metab- Ile and a putative recognition sequence (Ala-Ala-
olism. Another example of PPARá-mediated regu- Pro) for a protease processing peroxisomal protein
lation is the activation of steroid metabolizing (Mori et al. 1991) and (3) the 80 kDa protein is
enzyme, NADP+-dependent 3á-HSD in human similar to enzymes participating in peroxisomal
liver by derivatives of clofibrate (Matsuura et al. â-oxidation of fatty acids. Indeed, typical peroxi-
1996). In contrast, the guinea pig ortholog is so far somal markers such as catalase and acyl-CoA
the only species in which the same treatment causes oxidase co-localized with porcine 17â-HSD 4 in
Journal of Molecular Endocrinology (1999) 22, 227–240
234    and   · 17â-HSD 4

immunogold labeling studies in uterus, kidney and

Recombinant 32 kDa and 80 kDa proteins were assayed in homogenates of transfected cells and were corrected for background conversion (Carstensen et al. 1996). The recombinant central domain
(ìmoll
1 min
1 mg prot.
1)
liver (Markus et al. 1995a,b).

Kinetic parameters of 17â-HSD 4 were assayed with 17â-estradiol, fatty acid-CoA hydratase with crotonyl-CoA and fatty acid-CoA dehydrogenase with acetoacetyl-CoA (Leenders et al. 1996).
In other species, such as the rat, 17â-HSD 4 was
indeed purified from isolated peroxisomes (Novikov
et al. 1994). However, at that time the amino acid
sequence of the rat enzyme was not known.
Peroxisomal localization of 17â-HSD 4 extends
our understanding on how steroid, sterol and bile
acid metabolism are interlinked. Different sub-
cellular distribution of various 17â-HSDs allows

dehydrogenase
Fatty acid-CoA
for local control of their activities. Peroxisomes

Vmax

3·31
2·91
1·25
ND
ND
were initially believed to play only a minor role in
mammalian metabolism. However, they play an

(ìM)
indispensable role in many metabolic pathways, like

34·8
35·3
21·9
ND
ND
Km
synthesis of cholesterol, â-oxidation of fatty acids,
biosynthesis of ether lipids and bile acids,

(ìmoll
1 min
1 mg prot.
1)

á-oxidation of phytanic acid and, as depicted in this
review, oxidation of steroids (Wanders et al. 1995,
Krisans 1996, Magalhaes & Magalhaes 1997,
Verhoeven et al. 1997, Seedorf et al. 1998).

MULTIFUNCTIONALITY OF 17â-HSD 4

In order to check for the presence of activities

Fatty acid-CoA
predicted by amino acid similarities of the 80 kDa

Vmax

4·44
3·92
1·36
protein (Fig. 2) its three domains were expressed

ND
ND
separately (Leenders et al. 1996b). The N-terminal hydratase
domain (aa 1–323) catalyzed both the 17â-
(ìM)

34·0
37·1
34·7
ND
ND
Km
hydroxysteroid and the 3-hydroxyacyl-CoA dehy-
drogenase reactions (Table 5). Kinetic parameters
(Km, Vmax) of the expressed full length 80 kDa
(nmoll
1 min
1 mg prot.
1)

protein were close to those observed for the single

was characterized after purification on glutathione-agarose and thrombin cleavage.

 5. Kinetic parameters of domains of porcine 17â-HSD 4

expressed domains or for the native purified enzyme

(EDH or VHF) (Leenders et al. 1996b).
This was the first observation of an enzyme
performing dehydrogenase activity not only with
steroids but also with 3-hydroxyacyl-CoA derivates
of fatty acids. Furthermore, the central domain
17â-Hydroxysteroid

(aa 324–596) was responsible for the 2-enoyl-acyl-

CoA hydratase activity. Essentially the same data
dehydrogenase

concerning conversion of 3-hydroxyacyl-CoAs and

Vmax

0·15
0·14
0·19
0·11
ND

2-enoyl-acyl-CoAs were obtained simultaneously

with expressed or purified proteins of the rat
(ìM)

(Novikov et al. 1994, Qin et al. 1997a,b). Estimated

ND
Km

0·2
0·3
0·3
0·4

Km values for the hydratase and fatty acid

dehydrogenase are similar to those of other enzymes
Recombinant central domain

of â-oxidation of fatty acids (Steinman & Hill 1975,

Palosaari & Hiltunen 1990). Both Km for steroids
ND, activity not detected.

and fatty acids are in the physiological range.

Recombinant 32 kDa

Recombinant 80 kDa

Further studies on peroxisomal â-oxidation of

Purified 32 kDa

fatty acids and precursors of bile acids demon-

strated that the 17â-HSD 4 specifically forms and
dehydrogenates -3-hydroxyacyl-CoAs and not
Sample

their -stereoisomers (Dieuaide-Noubhani et al.

VHF

1997a, Jiang et al. 1997). This differentiates the

Journal of Molecular Endocrinology (1999) 22, 227–240
 17â-HSD 4 ·    and   235

enzyme from the multifuntional enzyme 1 (MFP1) The role of SCP2 in steroidogenesis and arterio-
(Osumi et al. 1985b) which is -specific. sclerosis has been extensively studied (Seedorf et al.
The velocity of 17â-estradiol oxidation by 1993, Krisans 1996, Magalhaes & Magalhaes
porcine, mouse and human 17â-HSD 4 is several 1997, Wanders et al. 1997). A synopsis of this
fold lower than that of fatty acyl-CoA (Leenders functionality is beyond the scope and limits of this
et al. 1994a, 1996b, Adamski et al. 1995, 1997, review.
Normand et al. 1995). The same observation was The role of the SCP2 domain for the functionality
made in rat (Dieuaide-Noubhani et al. 1996b, Qin of 17â-HSD 4 is not clear. The expressed por-
et al. 1997b). However, all other known 17â-HSDs cine SCP2 domain facilitates the transfer of
have the same conversion rates for steroids within 7-dehydrocholesterol and phosphatidylcholine be-
the range 0·1 to 2·45 nmol/min/mg protein (Table tween membranes in vitro. The activities of the
1). Because the enzymatic parameters (Vmax, Km) N-terminal domain towards steroids or fatty
of the 17â-HSD 4 for both fatty acyl-CoA and acyl-CoA are not changed if the SCP2 domain is
steroids are close to those known for other enzymes deleted (Leenders et al. 1996b).
of the SCAD gene family it remains to be settled Recently, gene targeting in mice was used to
which substrates are physiological. study the unknown function of SCP2 (Seedorf et al.
In comparison to other 17â-HSDs the type 4 1998). In the Scp2(-/-) mice with complete
enzyme accepts the most structurally diverse deficiency of SCP2 and SCPX, marked alterations
substrates. Human 17â-HSD 1–3 and mouse and in gene expression, peroxisome proliferation, hypo-
rat 17â-HSD 7 (Nokelainen et al. 1998) are lipidemia, impaired body weight control, and
practically only active with steroids at position 17. neuropathy were observed. Knock-out mice showed
However, 17â-HSD 5 and the recently cloned rat impaired catabolism of methyl-branched fatty
17â-HSD 6 (Biswas & Russell 1997) must be acyl-CoAs, especially of the tetramethyl-branched
considered multifunctional because of their 3á- fatty acid, phytanic acid. The gene disruption led to
hydroxysteroid dehydrogenase activity. Actually, inefficient import of phytanoyl-CoA into peroxi-
17â-HSD 5 was first identified in a row of different somes and to defective thiolytic cleavage of
3á-HSDs and assigned type 2 among them (for 3-ketopristanoyl-CoA.
reviews see Lin et al. 1997, Penning 1997). One
3á-HSD was even first identified as a bile acid
binding protein (Nanjo et al. 1995). MUTATIONS IN THE HSD17B4 GENE
Interesting is the novel 17â-hydroxysteroid
dehydrogenase previously known as Ke6 protein in Recent research on peroxisomal disorders revealed
mouse and human (Ando et al. 1996, Formicheva that 17â-HSD 4 is deficient in Zellweger syndrome
et al. 1998). This enzyme which might be termed (Novikov et al. 1997, Suzuki et al. 1997, van
17â-HSD 8 has a Vmax of 0·27 nmol/min/mg Grunsven et al. 1998). Peroxisomal organelle
protein with 17â-estradiol/NAD+ and reveals the deficiency results in disorders of lipid, fatty acid and
highest (37%) amino acid sequence identity among sterol metabolism such as Zellweger syndrome,
the 17â-HSDs to 17â-HSD 4. It remains to be adrenoleukodystrophy, infantile refsum disease and
verified if 17â-HSD 8 is able to metabolize bile hyperpipecolic acidemia (Lazarow & Moser 1995,
acids or fatty acids. Wanders et al. 1995). Patients with peroxisomal
deficiency reveal high plasma concentrations of long
chain fatty acids, bile acids and deficient synthesis
ROLE OF SCP2 DOMAIN of plasmalogens. This exerts pleiotropic influence
of renal functions impairment and neuronal
The most C-terminal part of the 80 kDa protein development (neuronal migration defects and
(amino acids 597–737) has 39% similarity to rat degeneration).
SCP2 (Leenders et al. 1994b). This non-specific One example of the molecular basis of the
lipid transfer protein is highly conserved, even in recently identified 17â-HSD 4 deficiency is the
evolutionarily distant species such as chicken and mutation G16S (van Grunsven et al. 1998). This
human (Yamamoto et al. 1991, Pfeifer et al. 1993). mutation is localized in the first exon (Fig. 2) and
This product of an SCPX gene, which actually disturbs the conformation of the Rossman fold
encodes two proteins, SCP2 and SCPX, is a fusion required for cofactor (NAD+) binding. The mutant
between SCP2 and thiolase (Ohba et al. 1994, is inactive with both steroids and bile acids (van
Seedorf et al. 1994). SCP2 is a 13 kDa basic Grunsven et al. 1998, Möller et al. 1999). The
protein believed to participate in the intracellular mutation is lethal, most probably because
movement of cholesterol and lipids (Wirtz 1997). the -specific pathways of pristanic acid and
Journal of Molecular Endocrinology (1999) 22, 227–240
236    and   · 17â-HSD 4

di/tri-hydroxycholestanoic acid metabolism are (Grandien et al. 1995). 17â-HSD 4 inactivates the
disrupted. As mentioned above, the multifunctional conversion of Ä5-androstene-3â, 17â-diol to dehy-
protein 1 (Osumi et al. 1985a) is -specific and droepiandrosterone (DHEA), a known peroxisomal
cannot substitute the deficient -pathway proliferator (Prough et al. 1994). Although both
(Dieuaide-Noubhani et al. 1997a). Because 17â- DHEA and clofibrate induce peroxisomes they
HSD 4 has a ubiquitous distribution, any mutation have opposite effects on the concentrations of
would affect the whole organism. In addition, triglycerides and cholesterol in blood. DHEA
developmental studies have shown that this enzyme increases the levels of lipids while clofibrate acts as
is present as early as at least day 7 post coitus of a hyperlipidemic drug. Decreased expression of
embryonic development (Mustonen et al. 1997). enzymes which inactivate estradiol, including
The lack of observation of any steroid hormone Cyp2C11, and the reported increased expression of
related phenotype might be due to compensation by aromatase (converting testosterone to estradiol) may
other 17â-hydroxysteroid dehydrogenases. explain why male rats exposed to diverse peroxi-
somal proliferators have higher serum estradiol
levels. These higher estradiol levels in male rats
PHYSIOLOGICAL SIGNIFICANCE IN have been thought to be mechanistically linked to
STEROID METABOLISM Leydig cell hyperplasia and adenomas. Increased
conversion of estradiol to the less active estrone by
Table 1 compares human 17â-HSD 1–5 and depicts 17â-HSD 4 induction may explain how exposure to
differences in catalytic parameters, posttranslational the di-(2-ethylhexyl)-phthalate leads to decreases in
processing and subcellular localization. The cata- serum estradiol levels and suppression of ovulation
lytical property of 17â-HSD 4, revealing the in female rats (Srivastava & Srivastava 1991, Corton
virtually unidirectional oxidative activity, clearly et al. 1997).
defines it as a steroid inactivating enzyme (Gurpide
& Marks 1981), since it produces estrone which
shows little affinity to the estradiol receptor. The ACKNOWLEDGEMENTS
conversion of 17â-estradiol to estrone might be
complemented by hydroxylations in positions 6á or The authors did their best to review known data on
7á (Maschler et al. 1983) producing steroids devoid 17â-HSD 4 within this review. We apologize to our
of estradiol receptor affinity and permitting fast colleagues whose contributions were not included
release from cells after formation. The Vmax and Km here. This occurred solely because of limitation of
values for EDH are similar to those for estrone space.
hydroxylases (Adamski et al. 1994) and allows the This work has been carried out on the basis of
metabolic conversion 17â-estradiol<estrone<6á/ grants awarded in part by the Institut Pasteur de
7á-hydroxy-estrone without rate-limiting steps. Lille (France), the CNRS (France), the FNRS
Our discovery of 17â-HSD 4 in peroxisomes (Belgium) and the Deutsche Forschungsgemein-
stimulated discussions about the possible role of schaft (Germany) (grant 127/4–1 to J A).
peroxisomes in steroid metabolism (Markus et al.
1995b). The enzyme has an as yet unseen ability to
be stimulated by both (1) progestins (Kaufmann
et al. 1995), which is a common feature of other REFERENCES
17â-HSDs (Tseng & Gurpide 1975, Poutanen et al.
1990), and (2) peroxisomal activators such as Adamski J 1991 Isolation of vesicles mediating the conversion
of 17â-estradiol to estrone. European Journal of Cell Biology
WY14,643 (Corton et al. 1996, 1997) or clofibrate 54 166–170.
(M Markus & J Adamski, unpublished results). The Adamski J, Sierralta WD & Jungblut PW 1987 Harvesting and
stimulation produces a tissue-specific response: separation of two populations of lysosomes from porcine
progestins induce about a tenfold increase in endometrium. European Journal of Cell Biology 45 238–245.
17â-HSD 4 mRNA and protein synthesis in the Adamski J, Husen B, Marks F & Jungblut PW 1992a The
membrane-bound 17â-estradiol dehydrogenase of porcine
uterus epithelium but not in liver whereas clofibrate endometrial cells: purification, characterization and
treatment has no effects on uterus but results in a subcellular localization. Journal of Steroid Biochemistry and
threefold induction in the liver. It is not known Molecular Biology 43 1089–1093.
whether the differences are caused by local Adamski J, Husen B, Marks F & Jungblut PW 1992b
concentrations of receptors, modulation of receptor Purification and properties of oestradiol 17â-dehydrogenase
extracted from cytoplasmic vesicles of porcine endometrial
action by spatially restricted factors or alternated cells. Biochemical Journal 288 375–381.
promoter usage similar to that observed for the Adamski J, Husen B, Marks F & Jungblut PW 1993 The
estrogen receptor in osteoblasts and breast tissue 17â-oestradiol dehydrogenase of pig endometrial cells is

Journal of Molecular Endocrinology (1999) 22, 227–240

 17â-HSD 4 ·    and   237

localized in specialized vesicles. Biochemical Journal 290 Corton JC, Bocos C, Cattley RC & Gustafsson JA 1995 The
777–782. rat 17â-estradiol dehydrogenase type IV is a novel
Adamski J, Hohls E & Jungblut PW 1994 Characterization of peroxisome proliferator responsive gene. In Peroxisomes:
estrone hydroxylase activities in procine endometrial cells. Biology and Role in Toxicology and Disease. Aspen, Colorado,
Experimental and Clinical Endocrinology 102 388–393. USA: The Aspen Institute.
Adamski J, Normand T, Leenders F, Monte D, Begue A, Corton JC, Bocos C, Moreno ES, Merritt A, Marsman DS,
Stehelin D, Jungblut PW & de Launoit Y 1995 Molecular Sausen PJ, Cattley RC & Gustafsson JA 1996 The rat
cloning of a novel widely expressed human 80 kDa 17â-hydroxysteroid dehydrogenase type IV is a novel
17â-hydroxysteroid dehydrogenase IV. Biochemical Journal peroxisome-proliferator-inducible gene. Molecular
311 437–443. Pharmacology 50 1157–1166.
Adamski J, Leenders F, Carstensen JF, Kaufmann M, Markus Corton JC, Bocos C, Moreno ES, Merritt A, Cattley RC &
MM, Husen B, Tesdorpf JG, Seedorf U, de Launoit Y & Gustafsson JA 1997 Peroxisome proliferators alter the
Jakob F 1997 Steroids, fatty acyl-CoA, and sterols are expression of estrogen-metabolizing enzymes. Biochimie 79
substrates of 80-kDa multifunctional protein. Steroids 62 151–162.
159–163. Delos S, Carsol JL, Ghazarossian E, Raynaud JP & Martin PM
Andersson S 1995 17â-Hydroxysteroid dehydrogenases: 1995 Testosterone metabolism in primary cultures of human
isozymes and mutations. Journal of Endocrinology 146 prostate epithelial cells and fibroblasts. Journal of Steroid
197–200. Biochemistry and Molecular Biology 55 375–383.
Ando A, Kikuti YY, Shigenari A, Kawata H, Okamoto N, Delos S, Carsol JL, Fina F, Raynaud JP & Martin PM 1998
Shiina T, Chen L, Ikemura T, Abe K, Kimura M & Inoko 5á-Reductase and 17â-hydroxysteroid dehydrogenase
H 1996 cDNA cloning of the human homologues of the expression in epithelial cells from hyperplastic and
mouse Ke4 and Ke6 genes at the centromeric end of the malignant human prostate. International Journal of Cancer 75
human MHC region. Genomics 35 600–602. 840–846.
Aoyama T, Peters JM, Iritani N, Nakajima T, Furihata K, Deyashiki Y, Ohshima K, Nakanishi M, Sato K, Matsuura K
Hashimoto T & Gonzalez FJ 1998 Altered constitutive & Hara A 1995 Molecular cloning and characterization of
expression of fatty acid-metabolizing enzymes in mice mouse estradiol 17 beta-dehydrogenase (A-specific), a
lacking the peroxisome proliferator-activated receptor alpha member of the aldoketoreductase family. Journal of Biological
(PPARá). Journal of Biological Chemistry 273 5678–5684. Chemistry 270 10461–10467.
Baker ME 1994 Sequence analysis of steroid- and Dieuaide-Noubhani M, Novikov D, Baumgart E, Vanhooren
prostaglandin-metabolizing enzymes: application to JC, Fransen M, Goethals M, Vandekerckhove J,
understanding catalysis. Steroids 59 248–258. Van Veldhoven PP & Mannaerts GP 1996a Further
Baker ME 1996 Unusual evolution of 11beta- and 17beta- characterization of the peroxisomal 3-hydroxyacyl-CoA
hydroxysteroid and retinol dehydrogenases. Bioessays 18 dehydrogenases from rat liver. Relationship between the
63–70. different dehydrogenases and evidence that fatty acids and
Biswas MG & Russell DW 1997 Expression cloning and the C27 bile acids di- and tri-hydroxycoprostanic acids are
characterization of oxidative 17beta- and 3alpha- metabolized by separate multifunctional proteins. European
hydroxysteroid dehydrogenases from rat and human prostate. Journal of Biochemistry 240 660–666.
Journal of Biological Chemistry 272 15959–15966. Dieuaide-Noubhani M, Novikov D, Baumgart E, Vanhooren
Blomquist CH, Lindemann NJ & Hakanson EY 1985 JCT, Fransen M, Goethals M, Vandekerkhove J,
17â-Hydroxysteroid and 20á-hydroxysteroid dehydrogenase Van Veldhoven PP & Mannaerts GP 1996b Further
activities of human placental microsomes: kinetic evidence characterisation of the peroxisomal 3-hydroxyacyl-CoA
for two enzymes differing in substrate specificity. Archives of dehydrogenase from rat liver. Relationship between different
Biochemistry and Biophysics 239 206–215. dehydrogenases and evidence that fatty acids and the C27
Caira F, Clemencet MC, Cherkaoui-Malki M, Dieuaide- bile acids di- and tri-hydroxycoprostanic acids are
Noubhani M, Pacot C, Van Veldhoven PP & Latruffe N metabolized by separate multifunctional proteins. European
1998 Differential regulation by a peroxisome proliferator of Journal of Biochemistry 240 660–666.
the different multifunctional proteins in guinea pig: cDNA Dieuaide-Noubhani M, Asselberghs S, Mannaerts GP & Van
cloning of the guinea pig -specific multifunctional protein 2. Veldhoven PP 1997a Evidence that multifunctional protein
Biochemical Journal 330 1361–1368. 2, and not multifunctional protein 1, is involved in the
Carruba G, Adamski J, Calabrò M, Miceli MD, Cataliotti A, peroxisomal beta-oxidation of pristanic acid. Biochemical
Bellavia V, Bue AL, Polito L & Castagnetta L 1997 Molecular Journal 325 367–373.
expression of 17â-hydroxysteroid dehydrogenase types in Dieuaide-Noubhani M, Novikov D, Vandekerckhove J,
relation to their activity in intact human prostate cancer cells. Veldhoven PPV & Mannaerts GP 1997b Identification and
Molecular and Cellular Endocrinology 131 51–57. characterization of the 2-enoyl-CoA hydratases involved in
Carstensen JF, Tesdorpf JG, Kaufmann M, Markus MM, peroxisomal beta-oxidation in rat liver. Biochemical Journal
Husen B, Leenders F, Jakob F, de Launoit Y & Adamski J 321 253–259.
1996 Characterization of 17â-hydroxysteroid dehydrogenase Elo JP, Akinola LA, Poutanen M, Vihko P, Kyllonen AP,
IV. Journal of Endocrinology 150 3–12. Lukkarinen O & Vihko R 1996 Characterization of
Casey ML, MacDonald PC & Andersson S 1994 17â- 17â-hydroxysteroid dehydrogenase isoenzyme expression in
Hydroxysteroid dehydrogenase type 2: chromosomal benign and malignant human prostate. International Journal
assignment and progestin regulation of gene expression in of Cancer 66 37–41.
human endometrium. Journal of Clinical Investigation 94 Engel LL & Groman EV 1974 Human placental 17â-estradiol
2135–2141. dehydrogenase: characterization and structual studies. Recent
Castagnetta L, Carruba G, Traina A, Granata OM, Markus M, Progress in Hormone Research 30 139–169.
Pavone-Macaluso M, Blomquist CH & Adamski J 1997 Entenmann AH, Sierralta WD & Jungblut PW 1980 The
Expression of different 17â-hydroxysteroid dehydrogenase dehydrogenation of estradiol to estrone by porcine
types and their activity in human prostate cancer cells. endometrial lysosomes. Hoppe-Seyler’s Zeitschrift für
Endocrinology 138 4876–4882. Physiologische Chemie 361 959–968.

Journal of Molecular Endocrinology (1999) 22, 227–240

238    and   · 17â-HSD 4

Formicheva J, Baker M, Anderson E, Lee GY & Aziz N 1998 Lazarow PB & Moser HW 1995 Disorders of Peroxisome
Characterization of Ke6, a new 17â-hydroxysteroid Biogenesis. New York: McGraw-Hill.
dehydrogenase and its expression in gonadal tissues. Journal Leenders F, Adamski J, Husen B, Thole HH & Jungblut PW
of Biological Chemistry 273 22664–22671. 1994a Molecular cloning and amino acid sequence of the
Geissler WM, Davis DL, Wu L, Bradshaw KD, Patel S, porcine 17â-estradiol dehydrogenase. European Journal of
Mendonca BB, Elliston KO, Wilson JD, Russell DW & Biochemistry 222 221–227.
Andersson S 1994 Male pseudohermaphroditism caused by Leenders F, Husen B, Thole HH & Adamski J 1994b
mutations of testicular 17â-hydroxysteroid dehydrogenase 3. The sequence of porcine 80 kDa 17â-estradiol
Nature Genetics 7 34–39. dehydrogenase reveals similarities to the short chain alcohol
Grandien K, Backdahl M, Ljunggren O, Gustafsson JA & dehydrogenase family, to actin binding motifs and to sterol
Berkenstam A 1995 Estrogen target tissue determines carrier protein 2. Molecular and Cellular Endocrinology 104
alternative promoter utilization of the human estrogen 127–131.
receptor gene in osteoblasts and tumor cell lines. Leenders F, Prescher G, de Launoit Y & Adamski J 1996a
Endocrinology 136 2223–2229. Assignment of human 17â-hydroxysteroid dehydrogenase IV
van Grunsven EG, van Berkel E, Ijlst L, Vreken P, de Klerk to chromosome 5q2 by fluorescence in situ hybridization.
JBC, Adamski J, Lemonde H, Clayton PT, Cuebas DA & Genomics 37 403–404.
Wanders RJA 1998 Peroxisomal -hydroxyacyl-CoA Leenders F, Tesdorpf JG, Markus M, Engel T, Seedorf U &
dehydrogenase deficiency: resolution of the enzyme defect Adamski J 1996b Porcine 80 kDa protein reveals intrinsic
and its molecular basis in bifunctional protein deficiency. 17â-hydroxysteroid dehydrogenase, fatty acyl-CoA-
Proceedings of the National Academy of Sciences of the USA hydratase/dehydrogenase and sterol transfer activities.
95 2128–2133. Journal of Biological Chemistry 271 5438–5442.
Gurpide E & Marks C 1981 Influence of endometrial 17â- Leenders F, Dolez V, Begue A, Möller G, Gloeckner JC, de
hydroxysteroid dehydrogenase activity on the binding of Launoit Y & Adamski J 1999 Structure of the gene for the
estradiol to receptors. Journal of Clinical Endocrinology and human 17â-hydroxysteroid dehydrogenase type IV.
Metabolism 52 252–255. Mammalian Genome 9 1036–1041.
Husen B, Adamski J, Szendro PI & Jungblut PW 1994 Lin HK, Jez JM, Schlegel BP, Peehl DM, Pachter JA &
Alterations in the subcellular distribution of 17â-estradiol Penning TM 1997 Expression and characterization of
dehydrogenase in porcine endometrial cells over the recombinant type 2 3á-hydroxysteroid dehydrogenase (HSD)
course of the estrous cycle. Cell and Tissue Research 278 from human prostate: demonstration of bifunctional
227–233. 3á/17â-HSD activity and cellular distribution. Molecular
Jakob F, Homann D & Adamski J 1995 Expression and Endocrinology 11 1971–1984.
regulation of aromatase and 17â-hydroxysteroid Luu-The V, Lachance Y, Labrie C, Leblanc G, Thomas JL,
dehydrogenase type 4 in human THP 1 leukemia cells. Strickler RC & Labrie F 1989 Full length cDNA structure
Journal of Steroid Biochemistry and Molecular Biology 55 and deduced amino acid sequence of human 3â-hydroxy-
555–563. 5-ene steroid dehydrogenase. Molecular Endocrinology
Jakob F, Siggelkow H, Homann D, Kohrle J, Adamski J & 3 1310–1312.
Schutze N 1997 Local estradiol metabolism in osteoblast- Luu-The V, Labrie C, Simard J, Lachance Y, Zhao H-F,
and osteoclast-like cells. Journal of Steroid Biochemistry and Couet J, Leblanc G & Labrie F 1990 Structure of two in
Molecular Biology 61 167–174. tandem human 17â-hydroxysteroid dehydrogenase genes.
Jiang LL, Kobayashi A, Matsuura H, Fukishima H & Molecular Endocrinology 4 268–275.
Hashimoto T 1996 Purification and properties of human Mäentausta O, Sormunen R, Isomaa V, Lehto V-P, Jouppila P
-3-hydroxyacyl-CoA dehydratase: medium chain enoyl-CoA & Vihko R 1991 Immunohistochemical localization of
hydratase is -3-hydroxyacyl-CoA dehydratase. Journal of 17â-hydroxysteroid dehydrogenase in the human
Biochemistry 120 624–632. endometrium during the menstrual cycle. Laboratory
Jiang LL, Kurosawa T, Sato M, Suzuki Y & Hashimoto T Investigation 65 582–587.
1997 Physiological role of -3-hydroxyacyl-CoA dehydratase/ Magalhaes MM & Magalhaes MC 1997 Peroxisomes in adrenal
-3-hydroxyacyl-CoA dehydrogenase bifunctional protein. steroidogenesis. Microscopy Research and Technique 36
Journal of Biochemistry 121 506–513. 493–502.
Jörnvall H, Persson B, Krook M, Atrian S, Gozales-Duarte R, Marks F 1992 Isolierung und Charakterisierung der 17â-
Jeffery J & Ghosh D 1995 Short-chain dehydrogenases/ Estradioldehydrogenase des Schweineuterus. PhD Thesis,
reductases (SDR). Biochemistry 34 6003–6013. University of Hannover, Germany.
Kaufmann M, Carstensen J, Husen B & Adamski J 1995 Markus M, Husen B & Adamski J 1995a The subcellular
The tissue distribution of porcine 17â-estradiol localization of 17â-hydroxysteroid dehydrogenase type 4 and
dehydrogenase and its induction by progesterone. its interaction with actin. Journal of Steroid Biochemistry and
Journal of Steroid Biochemistry and Molecular Biology 55 Molecular Biology 55 617–621.
535–539. Markus M, Husen B, Leenders F, Jungblut PW, Hall PF &
Kobayashi K, Kobayashi H, Ueda M & Honda Y 1997 Adamski J 1995b The organelles containing porcine 17â-
Expression of 17â-hydroxysteroid dehydrogenase type IV in hydroxysteroid dehydrogenase are peroxisomes. European
chick retinal pigment epithelium. Experimental Eye Research Journal of Cell Biology 68 263–267.
64 719–726. Martel C, Rheaume E, Takahashi M, Trudel C, Couet J, Luu
Krisans SK 1996 Cell compartmentalization of cholesterol TV, Simard J & Labrie F 1992 Distribution of 17 beta-
biosynthesis. Annals of the New York Academy of Sciences hydroxysteroid dehydrogenase gene expression and activity
804 142–164. in rat and human tissues. Journal of Steroid Biochemistry and
Labrie Y, Durocher F, Lachance Y, Turgeon C, Simard J, Molecular Biology 41 597–603.
Labrie C & Labrie F 1995 The human type II 17â- Maschler I, Ball P, Bayerköhler G, Gaues J & Knuppen R
hydroxysteroid dehydrogenase gene encodes two 1983 Identification of 6á- and 7á-hydroxyestrone as major
alternatively spliced mRNA species. DNA and Cell Biology metabolites of estrone and estradiol in porcine uterus.
14 849–861. Steroids 41 597–607.

Journal of Molecular Endocrinology (1999) 22, 227–240

 17â-HSD 4 ·    and   239

Matsuura K, Tamada Y, Deyashiki Y, Miyabe Y, Nakanishi 2-enoyl-CoA hydratase, 3-hydroxy-CoA dehydrogenase and
M, Ohya I & Hara A 1996 Activation of human liver 3,2-enoyl-CoA isomerase activities. Journal of Biological
3á-hydroxysteroid dehydrogenase by sulphobromophthalein. Biochemistry 265 2446–2449.
Biochemical Journal 313 179–184. Peltoketo H, Isomaa V, Mäentausta O & Vihko R 1988
Möller G, Leenders F, van Grunsven E, Dolez V, Qualmann Complete amino acid sequence of human placental
B, Kessels MM, Markus M, Krazeisen A, Husen B, 17â-hydroxysteroid dehydrogenase deduced from cDNA.
Wanders RJA, de Launoit Y & Adamski J 1999 FEBS Letters 239 73–77.
Characterization of the HSD17B4 gene: -specific Penning TM 1997 Molecular endocrinology of hydroxysteroid
multifunctional protein 2/17â-hydroxysteroid dehydrogenase dehydrogenases. Endocrine Reviews 18 281–305.
IV. Journal of Steroid Biochemistry and Molecular Biology (In Penning TM, Bennett MJ, Smith-Hoog S, Schlegel BP, Jez
Press). JM & Lewis M 1997 Structure and function of 3á-
Mori T, Tsukamoto T, Mori H, Tashiro Y & Fujiki Y 1991 hydroxysteroid dehydrogenase. Steroids 62 101–111.
Molecular cloning and deduced amino acid sequence of Persson B, Krook M & Jörnvall H 1991 Characteristics of
nonspecific lipid transfer protein (sterol carrier protein 2) of short-chain alcohol dehydrogenases and related enzymes.
rat liver: a higher molecular mass (60 kDa) protein contains European Journal of Biochemistry 200 537–543.
the primary sequence of nonspecific lipid transfer protein as Pfeifer SM, Sakuragi N, Ryan A, Johnson AL, Deeley RG,
its C-terminal part. Proceedings of the National Academy of Billheimer JT, Baker ME & Strauss JF 1993 Chicken sterol
Sciences of the USA 88 4338–4342. carrier protein 2/sterol carrier protein x: cDNA cloning
Mustonen MVJ, Poutanen MH, Isomaa VV, Vihko PT & reveals evolutionary conservation of structure and regulated
Vihko RK 1997 Cloning of mouse 17â-hydroxysteroid expression. Archives of Biochemistry and Biophysics 304
dehydrogenase type 2, and analysing expression of the 287–293.
mRNAs for types 1, 2, 3, 4 and 5 in mouse embryos and Pollow K, Lübbert H & Pollow B 1976 Partial purification and
adult tissues. Biochemical Journal 325 199–205. evidence of heterogeneity of the cytoplasmic 17â-
Nanjo H, Adachi H, Morihana S, Mizoguchi T, Nishihara T & hydroxysteroid dehydrogenase (17â-HSD) from normal
Terada T 1995 Enzymatic characterization of a novel bovine human endometrium and endometrial carcinoma. Journal of
liver dihydrodiol dehydrogenase – reaction mechanism and Steroid Biochemistry 7 315–320.
bile acid dehydrogenase activity. Biochimica et Biophysica Poutanen M, Isomaa V, Kainulainen K & Vihko R 1990
Acta 1244 53–61. Progestin induction of 17â-hydroxysteroid dehydrogenase
Nokelainen P, Peltoketo H, Vihko R & Vihko P 1998 enzyme protein in the T-47D human breast-cancer cell line.
Expression cloning of a novel estrogenic mouse 17â- International Journal of Cancer 46 897–901.
hydroxysteroid dehydrogenase/17-ketosteroid reductase Poutanen M, Isomaa V, Peltoketo H & Vihko R 1995
(m17HSD7), previously described as a prolactin receptor- Regulation of oestrogen action: role of 17â-hydroxysteroid
associated protein (PRAP) in rat. Molecular Endocrinology 12 dehydrogenases. Annals of Medicine 27 675–682.
1048–1059. Prough RA, Webb SJ, Wu HQ, Lapenson DP & Waxman DJ
Normand T, Husen B, Leenders F, Pelczar H, Baert JL, Begue 1994 Induction of microsomal and peroxisomal enzymes by
A, Flourens AC, Adamski J & de Launoit Y 1995 Molecular dehydroepiandrosterone and its reduced metabolite in rats.
characterization of mouse 17â-hydroxysteroid dehydrogenase Cancer Research 54 2878–2886.
IV. Journal of Steroid Biochemistry and Molecular Biology 55 Qin YM, Haapalainen AM, Conry D, Cuebas DA, Hiltunen
541–548. JK & Novikov DK 1997a Recombinant 2-enoyl-CoA
Novikov DK, Vanhove G, Carchon H, Asselberghs S, Eyssen hydratase derived from rat peroxisomal multifunctional
HJ, Van Veldhoven PP & Mannaerts GP 1994 Peroxisomal enzyme 2: role of the hydratase reaction in bile acid
â-oxidation. Purification of four novel 3-hydroxyacyl-CoA synthesis. Biochemical Journal 328 377–382.
dehydrogenases from rat liver peroxisomes. Journal of Qin YM, Poutanen MH, Helander HM, Kvist AP, Siivari
Biological Chemistry 269 27125–27135. KM, Schmitz W, Conzelmann E, Hellman U & Hiltunen JK
Novikov D, Dieuaide-Noubhani M, Vermeesch JR, Fournier 1997b Peroxisomal multifunctional enzyme of beta-oxidation
B, Mannaerts GP & Van Veldhoven PP 1997 The human metabolizing -3-hydroxyacyl-CoA esters in rat liver:
peroxisomal multifunctional protein involved in bile acid molecular cloning, expression and characterization.
synthesis: activity measurement, deficiency in Zellweger Biochemical Journal 321 21–28.
syndrome and chromosome mapping. Biochimica et Reed MJ 1991 Oestradiol-17â hydroxysteroid dehydrogenase:
Biophysica Acta 1360 229–240. its family and function. Journal of Endocrinology 129
Ohba T, Rennert H, Pfeifer SM, He Z, Yamamoto R, Holt JA, 163–165.
Bilheimer JT & Strauss III JF 1994 The structure of the Scublinsky A, Marin C & Gurpide E 1976 Localization of
human sterol carrier protein x/ sterol carrier protein 2 estradiol 17â dehydrogenase in human endometrium. Journal
(SCP2) gene. Genomics 24 370–374. of Steroid Biochemistry 7 745–747.
Osumi T, Ishii N, Hijikata M, Kamijo K, Ozasa H, Furuta S, Seedorf U, Raabe M & Assmann G 1993 Cloning, expression
Miyazawa S, Kondo K, Inoue K, Kagamiyama H et al. and sequences of mouse sterol carrier protein-X encoding
1985a Molecular cloning and nucleotide sequence of the cDNAs and related pseudogene. Gene 123 165–172.
cDNA for rat peroxisomal enoyl-CoA: hydratase-3- Seedorf U, Brysch P, Engel T, Schrage K & Assmann G
hydroxyacyl-CoA dehydrogenase bifunctional enzyme. 1994 Sterol carrier protein-X is a peroxisomal 3-oxoacyl
Journal of Biological Chemistry 260 8905–8910. coenzyme A thiolase with intrinsic sterol carrier and lipid
Osumi T, Ishii N, Hijikata M, Kamijo K, Ozasa H, Furuta S, transfer activity. Journal of Biological Chemistry 269
Miyazawa S, Kondo K, Inoue K, Kagamiyama H et al. 21277–21283.
1985b Molecular cloning and nucleotide sequence of the Seedorf U, Raabe M, Ellinghaus P, Kannenberg F, Fobker M,
cDNA for rat peroxisomal enoyl-CoA: hydratase-3- Engel T, Denis S, Wouters F, Wirtz KW, Wanders RJ,
hydroxyacyl-CoA dehydrogenase bifunctional enzyme. Maeda N & Assmann G 1998 Defective peroxisomal
Journal of Biological Chemistry 260 8905–8910. catabolism of branched fatty acyl coenzyme A in mice
Palosaari PM & Hiltunen JK 1990 Peroxisomal bifunctional lacking the sterol carrier protein-2/sterol carrier protein-X
protein from rat liver is a trifunctional enzyme possessing gene function. Genes and Development 12 1189–1201.

Journal of Molecular Endocrinology (1999) 22, 227–240

240    and   · 17â-HSD 4

Sierralta W, Truitt AJ & Jungblut PW 1978 Studies on the oxidation pathway: identification of pristanal as product of
involvement of lysosomes in estrogen action. I. Isolation and the decarboxylation of 2-hydroxyphytanoyl-CoA. Biochemical
enzymatic properties of pig endometrial lysosomes. Hoppe- and Biophysical Research Communications 237 33–36.
Seyler’s Zeitschrift für Physiologische Chemie 359 517–528. Wanders RJA, Schutgens RBH & Barth PG 1995 Peroxisomal
Srivastava S & Srivastava SP 1991 Effect of di(2- disorders: a review. Journal of Neuropathology and
ethylhexyl)phthalate on 17â-hydroxysteroid dehydrogenase Experimental Neurology 54 726–739.
in testis of rat. Toxicology Letters 57 235–239. Wanders RJ, Denis S, Wouters F, Wirtz KW & Seedorf U
Steinman HM & Hill RL 1975 Bovine liver crotonase (enoyl 1997 Sterol carrier protein-X (SCPx) is a peroxisomal
coenzyme A hydratase). Methods in Enzymology 35 136–139. branched-chain beta-ketothiolase specifically reacting with
Suzuki Y, Jiang LL, Souri M, Miyazawa S, Fukuda S, Zhang 3-oxo-pristanoyl-CoA: a new, unique role for SCPx in
Z, Une M, Shimozawa N, Kondo N, Orii T & Hashimoto T branched-chain fatty acid metabolism in peroxisomes.
1997 -3-hydroxyacyl-CoA dehydratase/-3-hydroxyacyl- Biochemical and Biophysical Research Communications 236
CoA dehydrogenase bifunctional protein deficiency: a newly 565–569.
identified peroxisomal disorder. American Journal of Human Wirtz KW 1997 Phospholipid transfer proteins revisited.
Genetics 61 1153–1162. Biochemical Journal 324 353–360.
Swinkels BW, Gould SJ, Bodnar AG, Rachubinski RA & Wu L, Einstein M, Geissler WM, Chan HK, Elliston KO &
Subramani S 1991 A novel, cleavable peroxisomal targeting Andersson S 1993 Expression cloning and characterization of
signal at the amino-terminus of the rat 3-ketoacyl-CoA human 17 beta-hydroxysteroid dehydrogenase type 2, a
thiolase. EMBO Journal 10 3255–3262. microsomal enzyme possessing 20 alpha-hydroxysteroid
Tremblay Y & Beaudoin C 1993 Regulation of 3â- dehydrogenase activity. Journal of Biological Chemistry 268
hydroxysteroid dehydrogenase and 17â-hydroxysteroid 12964–12969.
dehydrogenase messenger ribonucleic acid levels by cyclic Yamamoto R, Kallen CB, Babalola GO, Rennert H, Billheimer
adenosine 3 ,5 -monophosphate and phorbol miristate acetate JT & Strauss JF 1991 Cloning and expression of a cDNA
in human choriocarcinoma cells. Molecular Endocrinology 7 encoding human sterol carrier protein 2. Proceedings of the
355–364. National Academy of Sciences of the USA 88 463–467.
Tseng L & Gurpide E 1974 Estradiol and 20á- Zhang Y, Dufort I, Soucy P, Labrie F & Luu-The V 1995
dihydroprogesterone dehydrogenase activities in human Cloning and expression of human type V 17â-hydroxysteroid
endometrium during the menstrual cycle. Endocrinology 94 dehydrogenase. Proceedings of the 77th Annual Meeting of the
419–425. Endocrine Society 622.
Tseng L & Gurpide E 1975 Induction of human endometrial
estradiol dehydrogenase by progestins. Endocrinology 97    3 December 1998
825–833.
Tseng L & Gurpide E 1979 Stimulation of various 17â- and
20á-hydroxysteroid dehydrogenase activities by progestins in
human endometrium. Endocrinology 104 1745–1750. NOTE ADDED IN PROOF
Tseng L & Mazella J 1981 Kinetic studies of human
endometrial hydroxysteroid dehydrogenase. Journal of
Steroid Biochemistry 14 437–442. Radiation hybrid mapping by Genethon assigned
Verhoeven NM, Schor DS, ten Brink HJ, Wanders RJ & HSD17B4 gene to interval D5S471–D5S393 with a
Jakobs C 1997 Resolution of the phytanic acid alpha- physical position of 484.22 cR3000 on chromosome 5.

Journal of Molecular Endocrinology (1999) 22, 227–240