The murine 3β-hydroxysteroid dehydrogenase (3β-HSD) gene family

Molecular and Cellular Endocrinology 187 (2002) 213– 221 www.elsevier.com/locate/mce

The murine 3b-hydroxysteroid dehydrogenase (3b-HSD) gene family A postulated role for 3b-HSD VI during early pregnancy Lihong Peng a, Jonathan Arensburg b, Joseph Orly b, Anita H. Payne a,* a

Di6ision of Reproducti6e Biology, Department of Gynecology and Obstetrics, Stanford Uni6ersity School of Medicine, 300 Pasteur Dr., Stanford, CA 94305 -5317, USA b The Alexander Silberman Institute of Life Sciences, The Hebrew Uni6ersity of Jerusalem, Jerusalem 91904, Israel

Abstract The enzyme 3b-hydroxysteroid dehydrogenase/isomerase (3b-HSD) is essential for the biosynthesis of all active steroid hormones. The 3b-HSD enzyme consists in multiple isoforms, each the product of a distinct gene. In the mouse, six tissue-specific isoforms have been identified. These isoforms are expressed in a tissue- and temporal specific manner. Mouse 3b-HSD VI is the only isoform expressed in decidua and giant trophoblast cells during the first half of mouse pregnancy. The tissue- and temporal-specific expression of 3b-HSD VI during mouse pregnancy, as determined by in situ hybridization and immunohistochemistry, shows that 3b-HSD is expressed exclusively in the antimesometrial decidua on E6.5 and E7.5. By E9.5, expression of 3b-HSD is observed in giant trophoblast cells with a marked increase in expression by E10.5. No expression of 3b-HSD is seen in decidua after E7.5 and no expression of 3b-HSD is seen in the embryo at any of the times investigated. Giant trophoblast cells in culture from E9.5 and E10.5 synthesize progesterone with cells from E10.5 producing about 3.5-fold more progesterone during the first 24 h in culture. Western blot analysis of 3b-HSD VI protein demonstrates that the amount of 3b-HSD VI protein correlates with the amount of progesterone biosynthesis in giant trophoblast cells from E9.5 and E10.5. We propose that progesterone produced during the first half of mouse pregnancy in decidua and giant trophoblast cells acts as an immunosuppressant at the fetal maternal interface to prevent rejection of the fetus. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: 3b-Hydroxysteroid dehydrogenase; 3b-Hydroxysteroid dehydrogenase gene family; Mouse 3b-hydroxysteroid dehydrogenase VI; Pregnancy; Progesterone

1. Introduction The enzyme 3b-hydroxysteroid dehydrogenase/isomerase (3b-HSD) is essential for the biosynthesis of all active steroid hormones: the adrenal steroid hormones, cortisol, corticosterone and aldosterone; and the gonadal steroid hormones, progesterone, testosterone and estradiol (Fig. 1). The 3b-HSD enzymes exist in multiple isoforms in humans and rodents, each a product of a distinct gene. The different isoforms are indicated by roman numerals in the chronological order in which they have been isolated. The same numeral for isoforms from different species does not imply that these iso* Corresponding author. Tel.: + 1-650-725-6805; fax: +1-650-7257102. E-mail address: [email protected] (A.H. Payne).

forms are orthologous. Table 1 lists the different isoforms in humans and rodents classified according to function and tissue-specific expression. Our laboratory has isolated and characterized six distinct cDNAs in the mouse (Abbaszade et al., 1997, 1995; Bain et al., 1991; Clarke et al., 1993a,b). These six cDNAs exhibit a high degree of sequence identity. All of the cDNAs contain a 1122-bp open reading frame encoding a protein of 373 amino acids. The percent identity of the predicted amino acid sequence ranges from 72 to 93% (Abbaszade et al., 1997). The six isoforms fall into two functionally distinct groups. The first group is comprised of 3b-HSD I, III, VI, and most likely II, which function as NAD+-dependent dehydrogenase/isomerases and are therefore essential for the biosynthesis of active steroid hormones. The second group is comprised of 3b-HSD IV and V which function as

0303-7207/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 3 - 7 2 0 7 ( 0 1 ) 0 0 6 8 9 - X

214

L. Peng et al. / Molecular and Cellular Endocrinology 187 (2002) 213–221

NADPH-dependent 3-ketosteroid reductases and are most likely involved in the inactivation of active steroid hormones, such as DHT (Abbaszade et al., 1995, 1997).

2. Tissue- and temporal-specific expression of mouse 3b-HSD isoforms The different isoforms are expressed in a tissue- and temporal-specific manner. Mouse 3b-HSD I, the ortholog to human 3b-HSD II (Simard et al., 1996), is the major or only isoform expressed in the gonads and adrenal glands and thus is responsible for the biosynthesis of gonadal and adrenal steroid hormones. 3bHSD II and III are expressed in the liver and kidney (Bain et al., 1991), with III being the major isoform expressed in the adult liver. Expression of 3b-HSD IV has been detected only in the kidney and is the major isoform expressed in both the male and female kidney (Clarke et al., 1993b). In situ hybridization analysis indicates that expression in the kidney is found only in the cortex, with the highest expression detected in the convoluted tubules (Clarke et al., 1993b). 3b-HSD V is exclusively expressed in the male liver with expression first detected during pubertal development between 30 and 40 days postnatally (Abbaszade et al., 1995; Park et al., 1996). An orthologous isoform to mouse 3b-HSD V, the rat 3b-HSD III, shows identical expression pattern to mouse V and is the only rat isoform identified to date which functions as a 3-ketosteroid reductase (de Launoit et al., 1992). Mouse 3b-HSD VI, the ortholog to human 3b-HSD I (Simard et al., 1996), is expressed in adult skin (Abbaszade et al., 1997), adult testis (Abbaszade et al., 1997; Baker et al., 1999), and during the first half of mouse pregnancy in decidua following implantation between embryonic day (E)5.5-E7.5, followed by expression in giant trophoblast cells between E8.5 and E10.5 (Abbaszade et al., 1997; Arensburg et al., 1999). The tissue- and temporal-specific expression

of mouse 3b-HSD VI will be discussed in more detail below. Studies on the ontogeny of expression of 3bHSD isoforms in fetal gonads and liver demonstrated that fetal testes expressed 3b-HSD I mRNA at E13 (the earliest time examined) (Baker et al., 1999; Greco and Payne, 1994). The expression of 3b-HSD I in fetal testes continued throughout pregnancy. Low or inconsistent expression of 3b-HSD I was observed in fetal ovaries. To our great surprise, 3b-HSD I was the major isoform expressed in both male and female livers during fetal development until the day of birth, after which time, 3b-HSD III becomes the major 3b-HSD isoform (Park et al., 1996). The functional role of the high expression of 3b-HSD I in the fetal liver of both sexes is not obvious at this time.

3. Enzymatic characteristics of the 3b-HSD recombinant proteins The enzymatic characteristics of the different mouse 3b-HSD isoforms were examined using cell-free homogenates of COS-1 or COS-7 cells which had been transiently transfected with pCMV5 expression vector containing the coding region of the different cDNAs. The cell free homogenates were incubated with 3H-labeled D53b-hydroxysteroids, pregnenolone, or dehydroepiandrosterone (DHEA), or the 5a-reduced steroids, androstanediol (Adiol) or dihydrotestosterone (DHT). NAD+ or NADP+ was used as the cofactor with pregnenolone, DHEA, and Adiol; NADH or NADPH was used as a cofactor with DHT. NAD+ was found to be the preferred cofactor for 3b-HSD I, III (Clarke et al., 1993a), and VI (Abbaszade et al., 1997). We have not studied the enzymatic characteristics of mouse 3b-HSD II. From the sequence of the coding region of 3b-HSD II, it can be concluded that it functions as an NAD+-dependent dehydrogenase (Abbaszade et al., 1995), The Km values with the different

Fig. 1. Steroid biosynthetic pathway in gonads, adrenal glands and placental trophoblast cells. P450scc, cytochrome P450 cholesterol side chain cleavage; P450c17, cytochrome P45017a-hydroxylase, C17-20 lyase; 3bHSD, 5-ene-3b-hydroxysteroid dehydrogenase/isomerase; 17KSR, 17-ketosteroid reductase; P450arom, cytochrome P450 aromatase.

L. Peng et al. / Molecular and Cellular Endocrinology 187 (2002) 213–221

215

Table 1 Classification of human, mouse and rat 3bHSD isoforms Class

Isoforms

Tissues where expressed

References

I. NAD+-dependent dehydrogenase/isomerase isoforms A Human II Adrenal glands, gonads Rat Iaa Adrenal glands, gonads, kidneys Rat Iba Adrenal glands, gonads, fat Mouse I Adrenal glands, gonads, fetal and neonatal liver B Human I Placenta, skin, mammary gland Rat IV Placenta, skin Mouse VI Maternal decidua, giant trophoblasts, skin, testis (Leydig cells) C Mouse II Kidneys, liver Mouse III Liver, kidneys

Bain et al., 1991; Park et al., 1996 Lorence et al., 1990; Rheaume et al., 1991 Simard et al., 1993 Abbaszade et al., 1997 Bain et al., 1991 Abbaszade et al., 1995; Bain et al., 1991

II. NADPH-dependent 3 -ketosteroid reductase isoforms D Rat III Male liver Mouse V Male liver E Mouse IV Kidneys

de Launoit et al., 1992 Abbaszade et al., 1995 Clarke et al., 1993b

a

Lachance et al., 1991 Simard et al., 1993; Zhao et al., 1991

Rat Ia and Ib refers to previously designated rat I and II, respectively (Simard et al., 1993; Zhao et al., 1991).

substrates are shown in Table 2. Both 3b-HSD I and VI exhibited very low Km values for pregnenolone (B 0.1 mM) as compared to 1 mM for 3b-HSD III, the dehydrogenase/isomerase isoform expressed in the liver. The Km values for DHEA were higher than the ones observed for pregnenolone with the three dehydrogenase/ isomerase isoforms studied. In spite of the high amino acid homology between 3b-HSD IV and V, and 3bHSD I, III, and VI, 3b-HSD IV and V do not have the capacity to convert D5-3b-hydroxysteroids to D4-3-ketosteroids. They only have the capacity to convert DHT to Adiol in the presence of NADPH. They cannot use NADH as a cofactor (Abbaszade et al., 1995). The Km value for DHT with 3b-HSD V is considerably lower than the Km value with 3b-HSD IV, suggesting that 3b-HSD V may be involved in the inactivation of DHT by converting DHT to androstanediol in the male mouse liver. The relatively high Km value found for DHT with 3b-HSD IV suggests that DHT may not be the preferred 5a-reduced 3-ketosteroid substrate (Payne et al., 1997).

4. Mouse 3b-HSD (Hsd3b) 1-6 genes The structure of mouse Hsd3b 1, 2, 4 (Clarke et al., 1996), and 6 (Peng and Payne, unpublished data) has been determined. The genes consist of four exons: the first exon is comprised entirely of the 5% untranslated sequence, the second exon contains the start site of translation, exon 3 is a short exon and exon 4 contains the majority of the coding region plus the 3% untranslated sequence. This is similar to the two human genes which have been characterized (Simard et al., 1996). The mouse genes differ in size, between 6 and 11 kb, due to the differences in the size of the introns. Linkage analysis determined that the structural Hsd3b genes are

closely linked within a 3.5-cM segment of mouse chromosome 3 between Tsh and Gba (Bain et al., 1993). This segment of mouse chromosome 3 shows conservation of gene order and physical distance with the centromeric region of human chromosome 1. The human 3b-HSD genes, HSD3B1 and HSD3B2, have been mapped to chromosome 1p13.1 at 1–2 cM of the centromeric marker D1Z5 (Simard et al., 1996). Additional analysis of the mouse Hsd3b locus using yeast artificial chromosomes (YAC) containing mouse genomic DNA established that all of the mouse Hsd3b genes are found in a region no larger than 1400 kb and that there are a total of seven Hsd3b genes (Clarke et al., 1996). We have shown that six of these genes are expressed (Abbaszade et al., 1997, 1995; Bain et al., 1993; Clarke et al., 1993a,b). We believe that the additional gene identified within the 1400-kb fragment is a Table 2 Enzyme characteristics of mouse 3b-HSD isoforms Isoform

Substrate

Preferred cofactor

Km (mM)

I

Pregnenolone DHEA Androstanediol

NAD+ NAD+ NAD+

0.076 0.14 0.16

III

Pregnenolone DHEA Androstanediol

NAD+ NAD+ NAD+

1.03 0.49 0.46

VI

Pregnenolone DHEA

NAD+ NAD+

0.035 0.12

IV

DHT

NADPH

2.20

V

DHT

NADPH

0.48

Data for 3b-HSD I, III, and IV are from Clarke et al., 1993a; for V, from Abbaszade et al., 1995; for VI, from Abbaszade et al., 1997. DHEA, dehydroepiandrosterone; androstanediol, 5a-androstane-3b, 17b-diol; DHT, dihydrotestosterone.

216

L. Peng et al. / Molecular and Cellular Endocrinology 187 (2002) 213–221

pseudogene. Further analysis of mouse genomic DNA by pulsed-field gel electrophoresis indicates that all members of the Hsd3b gene family are found within a 400-kb fragment (Clarke et al., 1996). The close linkage of the Hsd3b genes suggests that this gene family exists as a tandem cluster of related genes which have arisen through duplication and divergence of a single ancestral gene. A similar finding was reported for the human HSD3b gene family which was found to be located within a 290-kb fragment on chromosome 1 (Morissette et al., 1995). 5. Tissue- and temporal-specific expression of 3b-HSD VI

5.1. Expression in mouse testes during de6elopment Mouse 3b-HSD VI is expressed in adult testes as analyzed by RT-PCR as well as by Western analysis (Abbaszade et al., 1997). An earlier analysis from our laboratory on the expression of steroidogenic enzymes in mouse fetal gonads indicated that only 3b-HSD I was expressed in fetal mouse testes (Greco and Payne, 1994). To determine possible temporal changes in the expression of 3b-HSD I and VI during testicular development, the expression of 3b-HSD I and VI mRNA was determined in fetal, neonatal and adult mouse testes by a semiquantitative RT-PCR method using isoform-specific primers. The results show that 3b-HSD I is expressed as early as E13.5 and continues to be expressed throughout development to adulthood. In contrast, 3b-HSD VI was detected in minute amounts at E13.5 and then only after day 10 postnatally with a marked increase in expression by postnatal day 15 which continued to increase until post-natal day 60, the final day examined (Baker et al., 1999). In situ hybridization using specific probes for 3b-HSD I and for 3b-HSD VI demonstrate that only 3b-HSD I is expressed in 5-day-old neonatal testes (Fig. 2A) whereas both of the isoforms are expressed in the adult testis (Fig. 2B). The in situ studies also show that expression of both 3b-HSD I and VI is confined to the interstitial cells (Fig. 2A,B). The expression of 3b-HSD VI mRNA appears to be as great or greater than the expression of 3b-HSD I in the mature mouse testis (Fig. 2A,B and Baker et al., 1999). This is in contrast to the relative expression of the two proteins (Fig. 5B and Abbaszade et al., 1997). The amount of 3b-HSD VI protein in the adult testis is always considerably less than the amount of mRNA. This observation suggests that translation of these two isoforms may be differentially regulated. The initial time of expression of 3b-HSD VI, between 10 and 15 days postnatally, coincides with the appearance of the adult type Leydig cell and suggests that expression of 3b-HSD VI mRNA may be a useful marker to study the differentiation of the adult-type Leydig cell.

5.2. Expression during early pregnancy in mice The temporal expression of 3b-HSD VI was examined by a semiquantitative RT-PCR method with RNA obtained from whole uterine tissue or isolated implantation sites between E3.5 and E10.5 (implantation of the blastocyst occurs between E4 and E5). A minute amount of 3b-HSD VI was detected at 3.5 days postcoitum prior to implantation. A 12-fold increase was observed at E4.5 and an additional sharp increase between E7.5 and E8.5 (Arensburg et al., 1999). The sharp increase between E7.5 and E8.5 reflects a switch from expression of this enzyme in the decidua at E7.5 to expression in giant trophoblast cells between E8.5 and E10.5. The cell-specific expression of 3b-HSD mRNA and protein was determined by in situ hybridization and fluorescent immunohistochemistry at E6.5, E7.5, and E9.5 (Arensburg et al., 1999). In situ hybridization and immunohistochemical staining of cross sections of E6.5 and E7.5 implantation sites show that 3b-HSD mRNA (Fig. 3A,C) and protein (Fig. 3D,E) are exclusively expressed in decidual tissue, specifically in the decidua capsularis. No expression of 3b-HSD is observed in the embryo (Fig. 3D) or in giant trophoblast cells at E7.5 (Fig. 3E). However, by E9.5, expression of 3b-HSD mRNA is seen in giant trophoblast cells with an absence of 3b-HSD VI mRNA expression in the decidua (Fig. 4). A marked increase in 3b-HSD VI mRNA is observed in giant trophoblast cells at E10.5 (Fig. 4). This increase in 3b-HSD VI mRNA in giant trophoblast cells at E10.5 is accompanied by a parallel increase in 3b-HSD VI protein as discussed below. No expression of 3b-HSD was observed in the embryo (Fig. 4).

6. Progesterone production by giant trophoblast cells To establish whether giant trophoblast cells have the capacity for de novo biosynthesis of progesterone, giant trophoblast cells were obtained from E9.5 and E10.5 implantation sites and single cell suspensions were prepared. A portion of the cells were used for the isolation of protein and the remainder were incubated for 72 h. The amount of progesterone produced during a 24-h period was determined at 24, 48 and 72 h. As shown in Fig. 5A, giant trophoblast cells from E9.5 show a slight increase in progesterone production over the 72-h incubation. Giant trophoblast cells from E10.5 produced approximately 3.5-fold more progesterone during the first 24 h in culture than cells from E9.5 followed by a time-related decrease over the next 48 h. To establish that progesterone biosynthesis in giant trophoblast cells from E10.5 is due to new synthesis and does not reflect accumulation of progesterone that is being released into the culture medium, an equal number of cells was

L. Peng et al. / Molecular and Cellular Endocrinology 187 (2002) 213–221

217

Fig. 2. Expression of 3b-HSD I and VI in mouse testes during development. (A) Frozen sections of testes from 5-day-old mice were subjected to in situ hybridization with 359 bp 35S-labeled sense and antisense cRNA probes representing 45 bp from the 3% end of the coding region and 314 bp from the 3%UTR of 3b-HSD VI cDNA (Abbaszade et al., 1997), and 366 bp 35S-labelled sense and antisense cRNA probes from the 3%UTR of 3b-HSD I cDNA (Bain et al., 1991). Slides were stained with eosin and hematoxylin. a and c, light field exposure; b, dark field exposure of 3b-HSD I antisense probe; d, dark field exposure of 3b-HSD VI antisense probe. Note there is no expression of 3b-HSD VI mRNA in 5-day-old testes (data for the sense probes not shown). (B) Frozen sections of 50-day-old testes were subjected to in situ hybridization as described above under (A) a and b were hybridized with 3b-HSD I antisense probe, c and d were hybridized with 3b-HSD I sense probe, e and f were hybridized with 3b-HSD VI antisense probe, g and h with 3b-HSD VI sense probe. a, c, e, and g are light field exposure, while b, d, f, and g are dark field exposure. Expression of both 3b-HSD I and VI mRNA is observed in 50-day-old mouse testes. Note expression of both I and VI is limited to the interstitial tissue.

incubated in the presence of 4 mM trilostane, a specific inhibitor of 3b-HSD activity. Trilostane markedly inhibits progesterone biosynthesis in E10.5 giant trophoblast cells (Fig. 5A). Western blot analysis of equal amounts of protein from E9.5 and E10.5 giant trophoblast cells shows that cells from E10.5 contain higher amounts of P450scc and 3b-HSD VI protein

than cells from E9.5 (Fig. 5B). The amount of enzyme protein correlates with the capacity of these cells for progesterone biosynthesis during the first 24 h in culture. The decrease in progesterone production with time in culture observed in E10.5 cells is interpreted to reflect the degeneration of the giant cells which occurs in vivo after E10.5.

218

L. Peng et al. / Molecular and Cellular Endocrinology 187 (2002) 213–221

7. Physiologic role of local progesterone production What is the physiologic role of progesterone production at or near the implantation site as shown by our studies? There are a number of reports suggesting an immunologic role for progesterone at the implantation site. Lydon et al. showed that treatment of ovariectomized wild-type mice with progesterone and estrogen results in the expected enlargement and development of the uterus (Lydon et al., 1995). In contrast, female mice that lacked functional progesterone receptors failed to develop this characteristic response. In the absence of the anti-inflammatory action of progesterone, the response of the receptorless uterus to estrogen was accompanied by a strong reaction to the immune system

(Lydon et al., 1995). These results confirm a longknown concept advocating an immunosuppressant role for progesterone at the fetal–maternal interface (Siiteri and Stites, 1982; Siiteri et al., 1977). A highly relevant study in human patients, implicating the need for production of progesterone by decidual and/or trophoblast cells for maintenance of pregnancy during the first trimester of gestation, was reported recently by Piccinni et al. (1998). They described a decreased production of leukemia inhibitory factor (LIF), IL-4 and IL-10 by decidual T cells from women with unexplained recurrent abortions (URA) during the first 8 weeks of gestation in comparison with the production of these cytokines in women with normal gestations. The URA could not be explained on the basis of conventional

Fig. 3. Localization of 3b-HSD mRNA and protein in sections of E6.5 and E7.5 implantation sites. (A) (bar, 600 mm), E6.5 histological micrograph of a sagittal-to-embryo/transverse-to uterus orientation section (hematoxylin – eosin stain). e, embryo; dc, decidua capsularis; db, decidua basalis; s, stroma; uc residual uterine cavity; bs, blood sinus; m, myometrium; mes, mesometrium. (B) (bar, 100 mm), fluorescence in situ hybridization was performed using consecutive sections of the implantation site shown in panel A (box B). Note expression of 3b-HSD mRNA in the decidual cells (dc), but not in the embryo (e) or the stroma (s). (C) (bar 20 mm), higher magnification of the boxed area in panel B. Label is specifically restricted to the cytosol and excluded from the nuclei (n). (D) (bar 45 mm), immunofluorescence staining of 3b-HSD protein in a E7.5 section viewed from the antimesometrial side of the embryo. 3b-HSD protein is not observed in the embryo (e), the degenerating old uterine epithelium (due), or migrating giant trophoblast cells (arrowheads). (E) (bar 20 mm), higher magnification to show lack of 3b-HSD protein in migrating giant trophoblast cells (arrowhead); n, giant cell nucleus; e, embryo; due, degenerating old uterine epithelium; dc, heavily stained decidua capsularis cells. Illustration from Arensburg et al. (1999).

L. Peng et al. / Molecular and Cellular Endocrinology 187 (2002) 213–221

219

Fig. 4. Expression of 3b-HSD VI mRNA in E9.5 and E10.5 giant trophoblast cells. Implantation sites from E9.5 and E10.5 were sectioned (8 mm) and hybridized with the 3b-HSD VI-specific 35S-labeled antisense and sense probes described above for Fig. 2. The slides were stained with hematoxylin and eosin. Panels A and B are light and dark field views of E9.5 with the antisense probe. The dark field exposure shows hybridization only in giant trophoblast cells surrounding the embryonic cavity. Panels C and D are light and dark field views of E10.5. Panel C shows intense silver grains in the light field exposure exclusively in giant trophoblast cells. The much greater expression of 3b-HSD VI in giant trophoblast cells in E10.5 giant trophoblast cells compared to E9.5 is obvious from the dark field exposure (compare panels B and D). Bar, 10 mm. Data with the sense probe is not shown.

criteria, including serum concentrations of gonadotropins and steroids. They reported that the addition of progesterone to short-term cultures of T cells resulted in a marked increase in the development of both LIF- and IL-4-producing type 2 T-helper (TH2) lymphocytes. TH2 lymphocytes secrete anti-inflammatory cytokines that trigger antibody production. In addition, TH2 cytokines inhibit the production of TH1 pro-inflammatory cytokines. The reduced production of LIF, IL-4, and IL-10 by decidual T cells in women with URA was not observed in their peripheral T cells, indicating that this is not an inherent defect of their T cells, but an alteration affected by the microenvironment (lack of local progesterone in the uterus). The report by Piccinni et al. implies that production of progesterone by the corpus luteum does not prevent abortions in these women with URA and is consistent with our hypothesis that local production of progesterone plays an important role in the maintenance of early pregnancy.

8. Conclusion In conclusion, there are two major 3b-HSD isoforms expressed in the mouse, 3b-HSD I and VI, and the two orthologous human isoforms, 3b-HSD II and I, respectively, which are involved in the biosynthesis of active steroid hormones from cholesterol. Mouse 3b-HSD I and human 3b-HSD II, the gonadal and adrenal isoforms, are essential for the biosynthesis of adrenal and gonadal steroid hormones. There have been numerous reports of mutations in human 3b-HSD II which result in the classic 3b-HSD deficiency impairing steroidogenesis in both the adrenals and the gonads (Moisan et al., 1999; Simard et al., 1996). The defect in the adrenal gland results in various degrees of salt wasting in both sexes and incomplete masculinization of the male external genitalia. The severity of the symptoms differ depending on the site of the mutation and whether the mutation results in the complete absence of expression of 3b-HSD II protein or decreased activity of the mutant protein. By contrast, to date, no mutations have

220

L. Peng et al. / Molecular and Cellular Endocrinology 187 (2002) 213–221

Fig. 5. Progesterone production and 3b-HSD VI and P450scc protein in giant trophoblast cells from E9.5 and E10.5 pregnant mice. Giant trophoblast cells were isolated from E9.5 and E10.5 pregnant C57Bl/6J mice. Cells were incubated in six-well plates for 72 h. (A) Progesterone production. The amount of progesterone produced during 24 h was determined by RIA at 24, 48, and 72 h. Protein was determined by the Bradford assay. An equal number of E10.5 giant trophoblast cells was incubated in the absence or presence of trilostane (4 mM), an inhibitor of 3b-HSD activity. Each value represents the mean 9 S.E. of three different wells. The data are representative of two experiments carried out at different times with different pregnant mice. The pattern of progesterone production seen at E9.5 and E10.5 was similar for both experiments. (B) Western blot analysis of 3b-HSD and P450scc protein in giant trophoblast cells from E9.5 and E10.5 (10 mg); 50-day-old mouse testis (T, 75 mg protein), mouse ovary (O, 1.5 mg protein), mouse adrenal (A, 1.4 mg protein); VI, pCMV5.3bHSD VI transfected COS cells; I, pCMV5.3bHSD I transfected COS cells. Proteins were subjected to SDS-PAGE and Western blot analysis. Membranes were incubated sequentially, first with an antiserum generated against the human placental 3b-HSD (Abbaszade et al., 1995, 1997), stripped, followed by incubation with an antiserum generated against bovine adrenal P450scc (Anakwe and Payne, 1987).

been identified in human 3b-HSD I, which is the only isoform expressed in the placenta. In women, the corpus luteum secretes progesterone during the first trimester, followed by a sharp increase in placental progesterone production. Only 3b-HSD I is expressed in human placenta (Simard et al., 1996). We hypothesize that progesterone produced early during the first trimester in human trophoblast or in mouse decidua between 5.5 and 7.5 and giant trophoblast cells between E8.5 and 10.5, as described in this study and by Arensburg et al. (1999), acts at the fetal –maternal interface to prevent fetal loss and maintain pregnancy during this period, despite normal serum progesterone levels contributed by the corpus luteum. Because the placenta originates from the trophectoderm of the implanting blastocyst, a mutation in human b-HDS I would be lethal and would explain why no homozygous mutations in human b-HSD I have been detected. To answer the question of whether local production of progesterone is essential for maintenance of early pregnancy in the mouse and, by inference, if a mutation in human 3b-HSD I, the orthologous form to mouse 3b-HSD VI, could be a cause of many unexplained recurrent abortions early in the first trimester, current studies are directed towards producing 3b-HSD VI null mice.

Acknowledgements This research was supported by NICHD/NIH through cooperative agreement (U54HD 31398) as part of the Specialized Cooperative Centers Program in Reproductive Research (AHP) and by The Israel Science Foundation (c 672/00) (J.O.).

References Abbaszade, I.G., Arensburg, J., Park, C.H., Kasa-Vubu, J.Z., Orly, J., Payne, A.H., 1997. Isolation of a new mouse 3b-hydroxysteroid dehydrogenase isoform, 3b-HSD VI, expressed during early pregnancy [published erratum appears in Endocrinology 1998 Jan;139(1):218]. Endocrinology 138, 1392 – 1399. Abbaszade, I.G., Clarke, T.R., Park, C.-H.J., Payne, A.H., 1995. The mouse 3b-hydroxysteroid dehydrogenase multigene family includes two functionally distinct groups of proteins. Mol. Endocrinol. 9, 1214 – 1222. Anakwe, O.O., Payne, A.H., 1987. Noncoordinate regulation of de no6o synthesis of cytochrome P450 cholesterol side-chain cleavage and cytochrome P450 17a-hydroxylase/C17 – 20 lyase in mouse Leydig cell cultures: relation to steroid production. Mol. Endocrinol. 1, 595 – 603. Arensburg, J., Payne, A.H., Orly, J., 1999. Expression of steroidogenic genes in maternal and extraembryonic cells during early pregnancy in mice. Endocrinology 140, 5220 – 5232. Bain, P.A., Meisler, M.H., Taylor, B.A., Payne, A.H., 1993. The

L. Peng et al. / Molecular and Cellular Endocrinology 187 (2002) 213–221 genes encoding gonadal and nongonadal forms of 3b-hydroxysteroid dehydrogenase/D5-D4 isomerase are closely linked on mouse chromosome 3. Genomics 16, 219 –223. Bain, P.A., Yoo, M., Clarke, T.R., Hammond, S.H., Payne, A.H., 1991. Multiple forms of mouse 3b-hydroxysteroid dehydrogenase/ D5-D4 isomerase and differential expression in gonads, adrenal glands, liver and kidneys of both sexes. Proc. Natl. Acad. Sci. USA 88, 8870 – 8874. Baker, P.J., Sha, J.A., McBride, M.W., Peng, L., Payne, A.H., O’Shaughnessy, P.J., 1999. Expression of 3b-hydroxysteroid dehydrogenase type I and type VI isoforms in the mouse testis during development. Eur. J. Biochem. 260, 911 –916. Clarke, T.R., Bain, P.A., Burmeister, M., Payne, A.H., 1996. Isolation and characterization of several members of the murine Hsd3b gene family. DNA Cell Biol. 15, 387 –399. Clarke, T.R., Bain, P.A., Sha, L., Payne, A.H., 1993a. Enzyme characteristics of two distinct forms of mouse 3b-hydroxysteroid dehydrogenase/D5-D4-isomerase cDNAs expressed in Cos-1 cells. Endocrinology 132, 1971 – 1976. Clarke, T.R., Bain, P.A., Greco, T.L., Payne, A.H., 1993b. A novel mouse kidney 3b-hydroxysteroid dehydrogenase complementary DNA encodes a 3-ketosteroid reductase instead of a 3b-hydroxysteroid dehydrogenase/D5-D4-isomerase. Mol. Endocrinol. 7, 1569 – 1578. de Launoit, Y., Zhao, H.-F., Be´ langer, A., Labrie, F., Simard, J., 1992. Expression of liver-specific member of the 3b-hydroxysteroid dehydrogenase family, an isoform possessing an almost exclusive 3-ketosteroid reductase activity. J. Biol. Chem. 267, 4513 – 4517. Greco, T.L., Payne, A.H., 1994. Ontogeny of expression of the genes for steroidogenic enzymes P450 side-chain cleavage, 3b-hydroxysteroid dehydrogenase, P450 17a-hydroxylase/C17-20 lyase, and P450 aromatase in fetal mouse gonads. Endocrinology 135, 262 – 268. Lachance, Y., Luu-The, V., Verreault, H., Dumont, M., Rhe´ aume, E., Leblanc, G., Labrie, F., 1991. Structure of the human type II 3b-hydroxysteroid dehydrogenase/D5-D4 isomerase (3b-HSD) gene: adrenal and gonadal specificity. DNA Cell Biol. 10, 701 – 711. Lorence, M.C., Murry, B.A., Trant, J.M., Mason, J.I., 1990. Human 3b-hydroxysteroid dehydrogenase/D5-D4 isomerase from placenta: expression in nonsteroidogenic cells of a protein that catalyzes the dehydrogenation/isomerization of C21 and C19 steroids. Endocrinology 126, 2493 –2498. Lydon, J.P., Demayo, F.J., Funk, C.R., Mani, S.K., Hughes, A., Montgomery, C.A. Jr, Shyamala, G., Conneely, O.M., O’Malley, B.W., 1995. Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev. 9, 2266 – 2278. Moisan, A.M., Ricketts, M.L., Tardy, V., Desrochers, M., Mebarki, F., Chaussain, J.L., Cabrol, S., Raux-Demay, M.C., Forest, M.G., Sippell, W.G., et al., 1999. New insight into the molecular

221

basis of 3beta-hydroxysteroid dehydrogenase deficiency: identification of eight mutations in the HSD3B2 gene eleven patients from seven new families and comparison of the functional properties of twenty-five mutant enzymes. J. Clin. Endocrinol. Metab. 84, 4410 – 4425. Morissette, J., Rheaume, E., Leblanc, J.-F., Luu-The, V., Labrie, F., Simard, J., 1995. Genetic linkage mapping of HSD3B1 and HSD3B2 encoding human types I and II 3b-hydroxysteroid dehydrogenase/D5-D4-isomerase close to D1S514 and the centromeric D1Z5 locus. Cytogenet. Cell Genet. 69, 59 – 62. Park, C.-H.J., Abbaszade, I.G., Payne, A.H., 1996. Expression of multiple forms of 3b-hydroxysteroid dehydrogenase in the mouse liver during fetal and postnatal development. Mol. Cell. Endocrinol. 116, 157 – 164. Payne, A.H., Abbaszade, I.G., Clarke, T.R., Bain, P.A., Park, C.H., 1997. The multiple murine 3 beta-hydroxysteroid dehydrogenase isoforms: structure, function, and tissue- and developmentally specific expression. Steroids 62, 169 – 175. Piccinni, M.P., Beloni, L., Livi, C., Maggi, E., Scarselli, G., Romagnani, S., 1998. Defective production of both leukemia inhibitory factor and type 2 T-helper cytokines by decidual T cells in unexplained recurrent abortions. Nat. Med. 4, 1020 – 1024. Rheaume, E., Lachance, Y., Zhao, H.-F., Breton, N., Dumont, M., de Launoit, Y., Trudel, C., Luu-The, V., Simard, J., Labrie, F., 1991. Structure and expression of a new complementary DNA encoding the almost exclusive 3b-hydroxysteroid dehydrogenase/ D5-D4-isomerase in human adrenals and gonads. Mol. Endocrinol. 5, 1147 – 1157. Siiteri, P.K., Stites, D.F., 1982. Immunologic and endocrine interrelationships in pregnancy. Biol. Reprod. 26, 1 – 14. Siiteri, P.K., Febres, F., Clemens, L.E., Chang, R.J., B., G., and Stites, D.F., 1977. Progesterone and the maintenance of pregnancy: Is progesterone nature’s immunosuppressant? Ann. NY Acad. Sci. 258, 384 – 397. Simard, J., Durocher, F., Mebarki, F., Turgeon, C., Sanchez, R., Labrie, Y., Couet, J., Trudel, C., Rheaume, E., Morel, Y., et al., 1996. Molecular biology and genetics of the 3b-hydroxysteroid dehydrogenase/D5-D4 isomerase gene family. J. Endocrinol. 150, S189 – S207. Simard, J., Couet, J., Durocher, F., Labrie, Y., Sanchez, R., Breton, N., Turgeon, C., Labrie, F., 1993. Structure and tissue-specific expression of a novel member of the rat 3b-hydroxysteroid dehydrogenase/D5-D4 isomerase (3b-HSD) family. J. Biol. Chem. 268, 19659 – 19668. Zhao, H.-F., Labrie, C., Simard, J., de Launoit, Y., Trudel, C., Martel, C., Rheaume, E., Dupont, E., Luu-The, V., Pelletier, G., Labrie, F., 1991. Characterization of rat 3b-hydroxysteroid dehydrogenase/D5-D4 isomerase cDNAs and differential tissue-specific expression of the corresponding mRNAs in steroidogenic and peripheral tissues. J. Biol. Chem. 266, 583 – 593.