BIOLOGY OF REPRODUCTION 56, 272-278 (1997)

Estrogen Stimulation of P450 Side-Chain Cleavage Activity in Cultures of Human Placental Syncytiotrophoblasts'

Jeffery S. Babischkin,3 Randall W. Grimes,3 Gerald J. Pepe, 4 and Eugene D. Albrecht2,3 Departments of Obstetrics/Gynecology/Reproductive Sciences and Physiology,' Center for Studies in Reproduction, University of Maryland School of Medicine, Baltimore, Maryland 21201 Department of Physiology Eastern Virginia Medical School, Norfolk, Virginia 23501

ABSTRACT biochemical differentiation of the latter cells that is mani-

fested as an increase in expression of the LDL receptor and Downloaded from https://academic.oup.com/biolreprod/article/56/1/272/2760805 by guest on 29 September 2021 The present study determined whether has a role in regulating the P450 cholesterol side-chain cleavage enzyme P450,,, enzyme system. (P450_cc) and/or de novo/deesterification cholesterol pathways Recently, we reported that LDL uptake was increased in involved in progesterone biosynthesis within human syncytiotro- human syncytiotrophoblast cells cultured with estrogen [7] phoblasts. Human placental syncytiotrophoblasts were cultured and suggested that estrogen also regulates key steps in the for 48 h with , and P450,,,cc activity was determined by placental progesterone pathway in human pregnancy. How- the formation of progesterone from 25-hydroxycholesterol. Es- ever, it remains to be determined whether estrogen also reg- tradiol at 10 7 or 10 6 M and 25-hydroxycholesterol increased ulates the P450,s,-catalyzed utilization of cholesterol sub- mean ( SE) progesterone production by syncytiotrophoblasts strate in human syncytiotrophoblasts. Moreover, although (ng/0.5 x 106 cells) to a value (19.2 - 1.1) that was 104% (p the de novo and deesterification pathways appear to have < 0.001) higher than that of the untreated controls (9.4 + 0.8) a minor role in providing cholesterol substrate for - and 52% higher (p < 0.001) than with 25-hydroxycholesterol ogenesis within the human trophoblast [8, 9], it is not alone (12.6 0.9). The stimulation of progesterone secretion known whether these cholesterol synthetic pathways are apparently was not the result of a change in progesterone me- regulated by estrogen. tabolism to its principal metabolite, because 20a-hydroxypregn- In the present study, therefore, we determined the effect 4-en-3-one represented a minor secretory component (0.7-1.7 of estrogen on P450,.c activity and mRNA levels, on the ng/0.5 x 106 cells) under these conditions, and levels were not de novo pathway involving the rate-limiting 3-hydroxy-3- substantially altered by estrogen. In contrast to the stimulatory methylglutaryl coenzyme-A (HMG-CoA) reductase en- effect of estradiol on P450scc activity, estrogen did not alter ei- zyme, and on the deesterification/esterification pathway in- ther the P450c c mRNA levels or the activities of 3-hydroxy-3- volving the rate-limiting cholesterol ester hydrolase/acyl methylglutaryl coenzyme-A reductase and cholesterol ester hy- coenzyme A-cholesterol acyltransferase (ACAT) in cultures drolase-rate-limiting enzymes for the de novo and deesterifi- of human syncytiotrophoblasts. cation pathways, respectively, for cholesterol formation in syn- cytiotrophoblasts in culture. Collectively, these results indicate that estrogen regulates the P450,cc component of the progester- MATERIALS AND METHODS one biosynthetic pathway, which we suggest signals functional/ Placental Collection and Trophoblast Preparation biochemical differentiation of syncytiotrophoblasts during pri- mate pregnancy. Human placentas were obtained immediately after un- complicated elective cesarean section at term, and cytotro- INTRODUCTION phoblasts were isolated essentially as described by Kliman et al. [10]. Briefly, placental villous tissue was dispersed at We have recently shown that there was an estrogen-de- 37°C in calcium-magnesium-free Hanks' Balanced Salt So- pendent developmental increase in low-density lipoprotein lution (GIBCO, Grand Island, NY), containing 4.0 mM so- (LDL) receptor and P450 cholesterol side-chain cleavage dium bicarbonate, 10.0 mM HEPES, 50 lpg/ml gentamicin (P450,,c) mRNA levels [1, 2] and LDL uptake [3, 4] in pla- sulfate, 50 IU/ml penicillin G, 50 pLg/ml streptomycin sul- cental syncytiotrophoblasts during advancing stages of ba- fate, 0.125% trypsin (bovine pancreas type III; Sigma boon pregnancy when estrogen levels become elevated. Chemical Co., St. Louis, MO), and 1500 kunitz U/mg DNa- Therefore, we have proposed that estrogen has a central se I (bovine pancreas type IV; Sigma). The cell suspensions integrative role in regulating those components of the pro- obtained from digestion were then layered onto a 5-70% gesterone biosynthetic pathway critical to the cellular up- (v:v) Percoll (Pharmacia LKB Biotechnology, Piscataway, take of substrate cholesterol and its metabolism to proges- NJ) gradient, and a cytotrophoblast-enriched cell fraction terone thereafter within the placenta ([5, 6] for reviews). was isolated by centrifugation at 1200 x g for 20 min at Moreover, because the ontogenetic increase in receptor-me- room temperature. diated LDL uptake and the P450scc enzyme occurred within syncytiotrophoblasts, we have proposed that after morpho- logical differentiation of cytotrophoblasts into syncytiotro- Culture phoblasts, there is a further estrogen-regulated functional/ Purified cytotrophoblasts (0.5-2.0 x 106 cells/ml) were incubated in 1 ml Dulbecco's Modified Eagles Medium Accepted September 11, 1996. (DMEM) containing 10% (v:v) fetal bovine serum, 44 mM Received July 8, 1996. sodium bicarbonate, 25 mM HEPES, 50 Rpg/ml gentamicin 'This work was supported by NIH Research Grant R01 HD-13294. 'Correspondence: Eugene D. Albrecht, Department of Obstetrics, Gv- sulfate, 50 IU/ml penicillin G, and 50 ,ag/ml streptomycin necology, and Reproductive Sciences, University of Maryland School of sulfate in 24-well culture plates (CoStar, Cambridge, MA) Medicine, Bressler Research Laboratories 11-017, 655 West Baltimore for 72 h at 37°C in a humidified atmosphere of 10% CO2: Street, Baltimore, MD 21201. FAX: (410) 706 5747. 90% air to allow for cell attachment, aggregation, and for- 272 P450,CC IN HUMAN PLACENTAL SYNCYTIOTROPHOBLASTS 273 mation of syncytia. Syncytiotrophoblast cultures were (Eastman Kodak, Rochester, NY). The level of mRNA ex- washed, incubated in serum-free DMEM for 48 h, and then pression was analyzed by autoradiographic densitometry incubated for 48 h in the presence of serum-free DMEM using a model 620 Video Densitometer (Bio-Rad, Rich- and various concentrations of estradiol-173 or other ster- mond, CA). oids (Sigma), 100 pxg/ml aminoglutethimide (Sigma), or 62 RIM 25-hydroxycholesterol (Steraloids, Wilton, NH) to as- HMG-CoA Reductase Activity sess P450,cc activity. Because 25-hydroxycholesterol readi- HMG-CoA reductase activity was measured by the for- ly traverses the plasma and mitochondrial membranes to 4 4 gain access to the terminal oxidase of the P450,cc, its con- mation of ['1 C]mevalonolactone from [1 C]HMG-CoA es- version into progesterone provides an index of P450,cc ac- sentially as described by Kubo and Strott [16]. Trophoblast tivity [11, 12]. In those cultures in which P450.cc mRNA cells were sonicated and then incubated (60 min; 37°C) in expression, HMG-CoA reductase, cholesterol ester hydro- 100 mM potassium phosphate buffer, pH 7.4, containing 20 lase, and ACAT were determined, the first 48-h serum-free mM glucose-6-phosphate (Sigma), 2.5 mM NADP (Sigma), 3.5 U/ml glucose-6-phosphate dehydrogenase (Sigma), 50

DMEM interval was omitted. At the end of all treatment Downloaded from https://academic.oup.com/biolreprod/article/56/1/272/2760805 by guest on 29 September 2021 periods, culture media were stored at -20°C until assayed mM NaCl, 5 mM dithiothreitol (Boehringer Mannheim), 10 mM EDTA, 0.1% (v:v) Triton X-100 (Sigma), and 2.5-200 for progesterone and 20a-hydroxypregn-4-en-3-one. Cells 4 were washed with 0.9% NaCl, and either RNA was isolated ,uM [' C]HMG-CoA (New England Nuclear-Dupont Co., for Northern blot analysis of P450,cc or cells were assessed Wilmington, DE). Reactions were terminated by addition of 0.05 ml of 5 N HCI, and after addition of 40 000 d.p.m. for activity of cholesterol metabolism enzymes as described 3 below. A small fraction of cells was also solubilized in 0.1 [ H]mevalonic acid (New England Nuclear-Dupont) as re- covery indicator, media were applied to 1-ml Bio-Rex 5 N NaOH for protein determination [13] or suspended in 1 14 3 ml 0.01 M Tris buffer and stored at -20°C until assayed (Bio-Rad) columns, and [ C/ H]mevalonolactone was elut- fluorometrically for cellular DNA content [14]. ed with 2 ml water. More than 95% of mevalonolactone was eluted from the column while 99% of [14C]HMG-CoA RIA was retained. HMG-CoA reductase specific activity was de- termined by the amount of ['4 C]mevalonolactone formed The contents of progesterone and 20a-hydroxypregn-4- in the presence of cells minus that measured in the absence en-3-one in the media were determined by RIA after puri- of cells. fication of samples on Sephadex LH-20 columns (Isolab, Inc., Akron, OH), as described previously [15], using high- Cholesterol Ester Hydrolase Activity ly specific antisera to progesterone (generously provided by Gordon Niswender, Colorado State University, Boulder, Neutral pH cholesterol ester hydrolase activity was mea- CO) and 20a-hydroxypregn-4-en-3-one (Endocrine Sci- sured in trophoblast cells by the formation of ['4 C]oleic ences, Tarzana, CA). acid from ['4 C-oleoyl]cholesteryl oleate by methods adapt- ed from Wittmaack et al. [17] and Miller and Small [18]. Northern Analysis of P450,,. Cells were incubated in 0.05 M Tris-HCI buffer (pH 7.4 at 37°C) containing 4.2 mg/ml fatty acid-free BSA (Sigma) RNA was prepared and analyzed by Northern blot hy- and 2.5-150 M [14 C-oleoyl]cholesteryl oleate (Amer- bridization as described previously [1]. Briefly, syncytiotro- sham). At the end of the incubation period, 20 000 d.p.m. phoblast cells were homogenized in 4 M guanidine isothio- [3 H]oleic acid (New England Nuclear-Dupont) was added cyanate and extracted with chloroform:isoamyl alcohol, and as recovery indicator, the reaction was terminated by ad- total RNA was obtained via cesium chloride gradient cen- dition of 5 ml chloroform:methanol (2:1), and the organic trifugation. Polyadenylated (poly A+)-enriched RNA was layer was dried under nitrogen. Lipids were resuspended in prepared from columns of oligo (deoxythymidine) cellulose 1 ml benzene:hexanes (1:1) and applied to -ml Bio-Sil HA (Pharmacia LKB) and size-fractioned by electrophoresis in (Bio Rad) columns, and ['4 C-oleoyl]cholesteryl oleate pre- 1.0% agarose gels containing 0.66 M formaldehyde and 20 cursor and [14 C]/[3H] oleic acid were eluted with benzene: mM-3 (N-morpholino) propane sulfonic acid. RNA was hexanes (1:1) and anhydrous diethyl ether. Cholesterol ester transferred to nylon membrane (Gene Screen; Dupont-New hydrolase specific activity was determined by the amount England Nuclear Corp., Boston, MA) and prehybridized in of [' 4 C]oleic acid in the presence of cells minus that in the buffer containing 50% formamide, 0.1% polyvinylpyrroli- absence of cells. done, 0.1% BSA, 0.1% Ficoll, 2.5-strength SSPE (0.375 M NaCl, 0.025 M NaH 2P0 4-H20, 0.0025 M EDTA-Na2, pH ACAT Activity 7.4), 1.0% SDS, 10% dextran sulphate, and denatured salm- on sperm DNA (100 jLg/ml) for 16-24 h at 42°C before the ACAT activity was measured in whole cells by the for- addition of probe. mation of [14C]cholesteryl oleate from [14C]oleoyl Co-A The human cDNAs for P450Scc (generously provided by and cholesterol by methods adapted from Wittmaack et al. Walter L. Miller, University of California, San Francisco, [17] and Miller and Small [18]. Cells were incubated in CA) and 3-actin (No. 65128; American Type Culture Col- 100 mM potassium phosphate buffer containing 5 mg/ml lection, Rockville, MD) were labeled with 50 iCi fatty acid-free BSA (Sigma), 2 mM glutathione (Sigma), 3 [o32P]dCTP (3000 Ci/mmol; Amersham Corp., Arlington mg/ml tyloxapol (Triton WR-1339; Sigma), 100 [Ig/ml cho- Heights, IL) using the Random-Primed DNA Labeling Kit lesterol (Sigma), and 2.5-200 tM ['4 C]oleoyl-CoA (Amer- (Boehringer Mannheim, Indianapolis, IN). Hybridization sham). At the end of the incubation period, 20 000 d.p.m. was performed at 420C for 23 h with 0.5-1.0 x 106 cpm/ml [3 H-linoleyl]cholesteryl linoleate (New England Nuclear- labeled probe. Membranes were extensively washed in dou- Dupont) was added as recovery indicator, the reaction was ble-strength SSC (0.3 M NaCl, 0.03 M sodium citrate-deu- terminated by addition of 5 ml chloroform:methanol (2:1), terium oxide), double strength SSC, 1% SDS, and 0.1- and the organic layer was dried under nitrogen. Lipids were strength SSC and were then exposed to Kodak X-AR film resuspended in 1 ml benzene:hexanes (1:1) and applied to 274 BABISCHKIN ET AL.

TABLE 2. Cellular protein and DNA content in human term syncytiotro- C U phoblasts cultured with estradiol. Uo 20 Protein DNA Treatment (Ig/0.5 x 1'0"cells) (g/0. 5 x 10" cells) x 16 + ab Control 20.6 1.2 1.8 + 0.1 12 Estradiol 20.5 0.4 2.0 0.1 a 1'Cytotrophoblasts were cultured as detailed in the first footnote of Table 1. W zw Values are the means + SE 8 _T b of 3 different cultures. .T 0 4. _ 0 strate (12.6 0.9). Although the addition of estradiol alone had no effect, estradiol at 10 6 M in the presence of 25-hy- . .r\\\\''\ H .. n _ - . C, , -- nL r droxycholesterol increased progesterone production to a Control u2 Z.J-urnC zOL-U-rltL2 value (19.2 + 1.1) that was 104% (p < 0.001) higher than Downloaded from https://academic.oup.com/biolreprod/article/56/1/272/2760805 by guest on 29 September 2021 FIG. 1 Effect of estradiol and 25-hydroxycholesterol on P450,,, activity in that of the respective untreated controls (Fig. 1), and 52% human term placental syncytiotrophoblasts in culture. Cytotrophoblasts higher (p < 0.001) than with 25-hydroxycholesterol alone. (0.5 X 106 cells) were incubated for 72 h with 10°% fetal bovine serum to permit transformation into syncytia, and syncytiotrophoblasts were then Addition of aminoglutethimide, an inhibitor of P450-me- cultured for 48 h in serum-free DMEM, and for 48 h in serum-free DMEM diated hydroxylation reactions, to the cultures caused ex- in the absence or presence of 3.7 x 10 M estradiol and/or 62 I~M 25 tensive (approximately 95%) inhibition of the synthesis of hydroxycholesterol. P450,, activity was determined by the formation of progesterone. progesterone from 25-hydroxycholesterol. Values are the means SE of Significant increases (p < 0.05) in progesterone produc- 5 separate cultures with each sample run in quadruplicate. Data points tion also occurred in syncytiotrophoblasts incubated for 48 with different letter superscripts are significantly different (p < 0.05-p < 0.001) from each other (analysis of variance and Newman-Keuls multiple h with 10 7 M estradiol and 25-hydroxycholesterol. Lower comparison test). E,, estradiol; 25-OHC, 25 hydroxycholesterol. concentrations of estradiol, however, were without effect (Table 1). The specificity of the response to estradiol was examined by incubating syncytiotrophoblasts with other 14 l-ml Bio-Sil HA columns, and [ C-oleoyl]cholesteryl ole- . In contrast to the stimulatory effect of estradiol- 3 ate and [ H-linoleyl]cholesteryl linoleate were eluted with 171, the addition of 10-6 M estradiol-l7ot, dihydrotestos- benzene:hexanes. Over 90% of cholesteryl ester eluted terone, or cortisol in the presence or absence of 25-hydrox- 4 from the column while 99.8% of the [1 C]oleic acid was ycholesterol had no significant effect on progesterone for- retained. Specific ACAT activity was determined by the mation (data not shown). Estradiol did not change the cel- 4 amount of [ C]cholesteryl oleate formed in the presence lular content of protein or DNA in syncytiotrophoblast of cells minus that measured in the absence of cells. cultures (Table 2). The effect of estrogen on the formation of 20ct-hydrox- Statistical Analysis ypregn-4-en-3-one, the major metabolite of progesterone, by syncytiotrophoblast cultures was also examined in Data were analyzed by analysis of variance with post two different cultures. Estradiol, in the presence or absence of hoc comparisons of the means by Newman-Keuls multiple 25-hydroxycholesterol, decreased 20a-hydroxypregn-4-en- comparison tests. 3-one output by approximately 50% (Table 3), or by less than I ng/0.5 x 106 cells from the control values. However, RESULTS because the level of 20a-hydroxypregn-4-en-3-one secre- P450,,, Activity tion in the presence of estradiol and 25-hydroxycholesterol was less than 5% of that of progesterone, the stimulation As demonstrated previously [10], isolated human cyto- of progesterone production was not the result of a quanti- trophoblasts cultured for 72 h in the presence of 10% fetal tatively significant alteration in progesterone metabolism. bovine serum were transformed into syncytiotrophoblasts, as determined by extensive immunocytochemical reactivity P450,,, mRNA Expression with antisera to syncytiotrophoblast-specific human placen- tal lactogen (data not shown). Progesterone synthesis (i.e., A 2.2-kb transcript for P450,,. was exhibited within hu- P450s,, activity; ng/0.5 x 106 cells) was increased (p < man syncytiotrophoblasts (Fig. 2), as previously shown by 0.05) above control levels (9.4 ±+0.8) during the 48-h in- Miller and coworkers [19]. The mRNA levels for P450s,. cubation period by addition of 25-hydroxycholesterol sub- expressed as a ratio of 3-actin, however, were similar in

TABLE 1. Effect of estradiol on progesterone formation (ngj by human term placental syncytiotrophoblasts in culture. Estradiol (M) Control 10' 10 10 10' 19.3 + 5.2 21.9 + 5.5 19.7 + 3.8 27.7 + 4.0' 26.3 + 3.41' "Cytotrophoblast (0.5 x 10' cells) were plated for 72 h to form syncytia, and syncytiotrophoblast were then cultured an additional 48 h in DMEM only, then incubated for 48 h in DMEM (without serum) supplemented with 62 M 25-hy- droxycholesterol in the absence (control) or presence of estradiol. Values are the means _+SE of 3 different cultures. -Significantly greater (p < 0.05) than control as analyzed by ANOVA and New- man-Keuls multiple comparison test. P450c c IN HUMAN PLACENTAL SYNCYTIOTROPHOBLASTS 275

TABLE 3. Effect of estradiol and 25-hydroxycholesterol on 20a-hydroxy- pregn-4-en-3-one formation by human term syncytiotrophoblasts..

20a-Hydroxypregn- 4-en-3-one Treatment (ng/0.5 x 106 cells) la Control 1.72 + 0.28 Estradiol 0.84 + 0.06 cc ' 25-Hydroxycholesterol 1.47 + 0.10 Estradiol + 25-hydroxycholesterol 0.77 _ 0.04 I l? "See legend of Figure 1 for details of the culture conditions. Values rep- resent the means SE of 2 different cultures. two different syncytiotrophoblast cultures incubated for 48 0 25 50 100 200 Downloaded from https://academic.oup.com/biolreprod/article/56/1/272/2760805 by guest on 29 September 2021 h with DMEM alone (mean = 11.5 densitometric units) or HUG-CoA ADDED (IM) with estradiol (mean = 10.2 densitometric units).

HMG-CoA Reductase Activity HMG-CoA reductase activity, measured in human pla- cental microsomes to validate the assay, was 6.3 pmol/ min/mg protein, similar to that (6 pmol/min/mg protein) reported in term placenta [20]. HMG-CoA reductase activ- Ii ity was linear over time and proportional to trophoblast cell number (data not shown); thus an assay time of 60 min and cell number of 2 x 106 were utilized. HMG-CoA reductase Vmax (pmol/min/mg cell protein) and affinity (K,,, RIM) for I substrate, determined by saturation analysis from double reciprocal plots (Fig. 3A), were similar for human syncy- tiotrophoblast cells incubated with 10-7 M or 10-6 M es- tradiol (Vma,,,, 7.0 + 1.4; K,, 8.7 3.3) or with DMEM CHOILETERM OLEATE ADDED (M) alone (Vmax, 6.3 ± 0.9; K,, 7.4 2.5).

Cholesterol Ester Hydrolase Activity Cholesterol ester hydrolase activity, determined in ho- mogenates of rat adrenal to validate the assay, was 16.9, 3.8, and 6.5 pmol/min/mg cell protein when performed at pH of 4.5, 7.4, and 8.2, respectively, which corresponds to values (20.0 and 4.9 pmol/min/mg at pH 4.5 and pH 8.2, respectively) previously reported [21, 22]. Cholesterol ester hydrolase activity was linear over time and proportional to human trophoblast cell number (data not shown); therefore, I an assay time of 60 min and cell number of 6-8 x 106 were utilized in all experiments. Cholesterol ester hydrolase Vmax (pmol/min/mg cell protein, Fig. 3B) and affinity for OLEOYL-CoA ADDED (M) substrate (Kin, RIM) did not differ in cells incubated with 10 7 M or 10-6 M estradiol (Vmax, 4.0 1.0; K,,, 8.4) or FIG. 3. Activities of HMG-CoA reductase (A), cholesterol ester hydrolase with DMEM alone (Vma, 3.7 ± 0.8; K,, 7.5). (B), and acylcoenzyme-A-cholesterol acyltransferase (C) in human syn- cytiotrophoblasts cultured for 48 h without (solid circles) or with 3.7 x 10 6 M estradiol (open circles). A) Saturation analysis of specific conver- sion of [ 4C]HMG-CoA to ['4Clmevalonolactone by 2 106 human syn- cytiotrophoblast cells. B) Saturation analysis of specific liberation of [' 4C]oleic acid from [4Clcholesteryl oleate by 6-8 x 106 human syncy- tiotrophoblast cells. C) Saturation analysis of specific conversion of ['4C]oleoyl-CoA and cholesterol to ['4C]cholesteryl oleate by 4 x 106 human syncytiotrophoblast cells. Values are the mean + SE of 4 separate cultures.

ACAT Activity ACAT activity, measured in human placental micro- somes to validate the assay, was 14.1 pmol/min/mg protein, FIG. 2. Northern blot of P450.,. mRNA from human term placental syn- similar to that (15 pmol/min/mg protein) previously re- cytiotrophoblast cultures (n = 2) incubated for 48 h in serum-free DMEM +- ported [9]. ACAT activity was linear over time and pro- in the absence (C) or presence of 3.7 x 10-6 M estradiol (E2). Poly(A) enriched RNA (approximately 10 pig/lane) was processed from syncytio- portional to trophoblast cell number (data not shown); thus trophoblasts and hybridized at 42°C for 23 h with approximately 106 an assay time of 60 min and cell number of 4 X 106 was cpm/ml 32P-labeled P450-c~ cDNA and exposed to x-ray film for 12 h. utilized. ACAT Vmax (pmol/min/mg cell protein) and affin- 276 BABISCHKIN ET AL. ity for substrate (K,,,; F.M), determined by saturation anal- biosynthetic pathways [10, 29], we propose that syncytio- ysis from double reciprocal plots (Fig. 3C), did not differ trophoblasts then undergo a further functional/biochemical in cells exposed to 10 7 M or 10-6 M estradiol (V,,,,,x, 13.5 differentiation process that is regulated by estrogen and that - 1.0 K,,i, 35.6) or to DMEM alone (V,,,,,, 15.2 + 2.3; K,,,, results in an enhanced capacity for the production of pro- 42.6). gesterone. Although the increases in P450s,, activity and LDL up- take [7] observed with estradiol in human syncytiotropho- DISCUSSION blast cultures were significant, the relative effects were less marked than The present study shows that estradiol stimulated P450,,, those observed on these aspects of steroido- activity, as determined by the formation of progesterone genesis in vivo in estrogen-depleted/repleted baboons [3, from 25-hydroxycholesterol, in cultures of human syncytio- 30]. One likely explanation for this difference is that other factors of trophoblasts. In contrast, estrogen did not change the activ- maternal, fetal, and/or extratrophoblast placental ities of HMG-CoA reductase, cholesterol ester hydrolase, origin, and thus not present in the culture system, interact with estrogen in vivo to regulate syncytiotrophoblast and ACAT-rate-limiting enzymes for the de novo and func- Downloaded from https://academic.oup.com/biolreprod/article/56/1/272/2760805 by guest on 29 September 2021 deesterification/esterification pathways for cholesterol for- tion. It is also possible that trophoblasts obtained at term mation. The increase in P450,cc activity observed in human for culture had already been programmed by the high levels syncytiotrophoblasts cultured with estrogen occurred in a of estrogen to which they were exposed in utero during human dose-dependent manner and was not associated with a pregnancy or in vitro in the presence of fetal bovine change in cellular protein or DNA. Although the increase serum during the initial 72-h transformation period. For ex- in P450,CC activity elicited by estrogen was not associated ample, we previously demonstrated that significant amounts of estrogen with a corresponding increase in mRNA levels, a rapid were endogenously produced by human tropho- turnover of P450,,, mRNA may explain these apparently blasts cultured with 10% fetal bovine serum [15]. Further- disparate results. Nuclear run-on assay and immunoblot as- more, we have shown that cytotrophoblasts obtained from say of P450,, protein are needed to definitively determine estrogen-treated baboons displayed a pattern of corticoste- roid metabolism after transformation the latter possibility. The level of estradiol (10 7 to 10 6 into syncytiotropho- blasts in M) shown to be effective under cell culture conditions was vitro that mimicked the pattern observed in the comparable to that to which syncytiotrophoblasts are ex- elevated estrogen condition in vivo [31]. An additional pos- posed in vivo during human pregnancy [23, 24]. The stim- sibility is that progesterone secretion by trophoblasts would ulation of progesterone secretion in syncytiotrophoblasts have been enhanced with the addition in medium of a pu- apparently was not the result of a change in progesterone rified steroid-binding protein such as albumin, which has metabolism to its principal metabolite, because 20ct-hy- been shown to greatly enhance steroid secretion under in droxypregn-4-en-3-one represented a minor secretory com- vitro conditions by diminishing intracellular unbound ste- ponent of the progestin pathway under these conditions. roid concentration and thus negative feedback on produc- Cyanoketone, an inhibitor of A-3-hydroxysteroid5 dehy- tion [32]. The action of estrogen on the trophoblast appears drogenase (33-HSD), was not used in this study because it to be mediated via the classical , which binds to oxidized placental cytochrome P450 to inhibit ac- we have shown is expressed in the nuclei of syncytiotro- tivity [25]; however, the estrogen-stimulated conversion of phoblasts in culture [33]. Thus, it is possible that the dif- 25-hydroxycholesterol to progesterone does not appear to ferences in relative effects of estrogen observed in vivo and reflect a change in 3-HSD activity in trophoblasts. Thus, in vitro also reflect corresponding differences in expression progesterone was the major product formed from 25-hy- of the estrogen receptor. droxycholesterol in human syncytiotrophoblast cultures The inability of estradiol alone to significantly increase [10], and 3-HSD activity determined in vitro in baboon progesterone formation probably reflected the absence of trophoblast cells or in vivo across the baboon placenta was exogenous LDL cholesterol in the serum-free culture me- not altered by estrogen [26]. We have also recently reported dium and the inability of estrogen to stimulate the de novo that estrogen stimulated LDL cholesterol uptake in human or deesterification pathways for cholesterol production in syncytiotrophoblasts [7] and that there was an estrogen- syncytiotrophoblasts. Collectively, therefore, the selective dependent developmental increase in LDL receptor mRNA effects demonstrated with estrogen on P450. and receptor- expression [1], LDL uptake [3, 4], and P450,,, mRNA ex- mediated LDL uptake in the baboon and human placenta pression and activity [2, 27] within syncytiotrophoblasts indicate that the LDL receptor and P450scc components of during baboon pregnancy, while adrenodoxin and 3-HSD the steroidogenic pathway, and not the de novo or deester- mRNA expression/activity were unaltered [2, 28]. Collec- ification pathways, are the principal sites of action of es- tively, these studies utilizing both in vitro and in vivo ex- trogen in regulating cholesterol metabolism within syncy- perimental approaches indicate that estrogen selectively tiotrophoblasts during primate pregnancy. This does not regulates key sites for the receptor-mediated uptake and uti- mean that in the absence of LDL the de novo/deesterifi- lization of cholesterol substrate for progesterone biosynthe- cation pathways for cholesterol formation are unimportant. sis within syncytiotrophoblasts of both the nonhuman pri- Indeed, activity of these enzymes was measurable in, and mate and human placenta. significant amounts of progesterone were formed by, syn- The estrogen-dependent ontogenetic increase in LDL re- cytiotrophoblasts in the absence of exogenous cholesterol ceptor mRNA levels/LDL uptake and P450s, mRNA levels substrate in serum-free medium as employed in the present previously shown in baboon pregnancy was exhibited with- study. Moise et al. [34] have reported that human chorio- in a syncytiotrophoblast-rich cellular fraction. In the current carcinoma cells cultured in the absence of LDL produced study, the stimulatory effect of estradiol on P450c, activity up to 50% of the progesterone formed in the presence of also occurred within human syncytiotrophoblasts. There- LDL. Thus, although the de novo and deesterification path- fore, although the morphological differentiation of cytotro- ways within trophoblasts cultured in the presence of LDL, phoblasts into syncytiotrophoblasts represents an important which has the capacity to down-regulate the HMG-CoA step in the onset of expression of the steroid and protein reductase and cholesterol ester hydrolase enzymes [35], P450,,, IN HUMAN PLACENTAL SYNCYTIOTROPHOBLASTS 277 seem to represent a minor source of substrate for steroid- REFERENCES ogenesis [8, 9], these intracellular sites may become more prominent sources of cholesterol substrate under various 1. Albrecht ED, Babischkin JS, Koos RD, Pepe GJ. Developmental in- experimental conditions. The regulatory effect of estrogen crease in low density lipoprotein receptor messenger ribonucleic acid on the LDL receptor and P450,cc components levels in placental syncytiotrophoblasts during baboon pregnancy. En- of the pro- docrinology 1995; 136:5540-5546. gesterone pathway observed in vivo in pregnant baboons 2. Babischkin JS, Pepe GJ, Albrecht ED. Developmental expression of in the presence of endogenous LDL further highlights the placental trophoblast P-450 cholesterol side-chain cleavage, adreno- specific effects that estrogen has upon these aspects of syn- doxin and A5-3{-hydroxysteroid dehydrogenase/isomerase messenger cytiotrophoblast function. ribonucleic acids during baboon pregnancy. Placenta 1996; (in press). Because aromatizable or estradiol were 3. Albrecht ED, Henson MC, Pepe GJ. Regulation of placental low den- sity lipoprotein uptake in baboons by estrogen. Endocrinology 1991; effective in stimulating placental P450sCC activity in vivo in 128:450-458. baboons [36], we chose to use estradiol for the study of 4. Henson MC, Pepe GJ, Albrecht ED. Developmental increase in pla- P450¢cc in human trophoblasts. Androgens and cental low density lipoprotein uptake during baboon pregnancy. En- docrinology 1992; 130:1698-1706. have been reported to have either stimulatory or inhibitory Downloaded from https://academic.oup.com/biolreprod/article/56/1/272/2760805 by guest on 29 September 2021 effects on placental progesterone secretion in vitro depend- 5. Albrecht ED, Pepe GJ. Placental steroid hormone biosynthesis in pri- mate pregnancy. Endocr Rev 1990; 11:124-150. ing upon the particular cellular conditions. For example, 6. Pepe GJ, Albrecht ED. Actions of placental and fetal adrenal steroid although estrogen [37] or sulfate hormones in primate pregnancy. Endocr Rev 1995; 16:608-648. [34] inhibited 3-HSD activity and progesterone formation 7. Grimes RW, Pepe GJ, Albrecht ED. Regulation of human placental from LDL in isolated human placental mitochondria or trophoblast low density lipoprotein uptake in vitro by estrogen. J Clin JEG-3 choriocarcinoma cells, respectively, androstenedione Endocrinol & Metab 1996; 81:2675-2679. 8. Simpson ER, Porter JC, Milewich L, Bilheimer DW, MacDonald PC. stimulated progesterone secretion in early human placental Regulation by plasma lipoproteins of progesterone biosynthesis and explants [38]. The apparently disparate effects may reflect 3-hydroxy-3-methyl glutaryl coenzyme A reductase activity in cul- the enzyme kinetic parameters associated with subcellular tured human choriocarcinoma cells. J Clin Endocrinol & Metab 1978; fractions and intact cells, because although 3P3-HSD activity 47:1099-1105. was inhibited by androgens or estrogen in placental micro- 9. Simpson ER, Burkhart ME Acyl coA:cholesterol acyl transferase ac- tivity in human placental microsomes: inhibition by progesterone. somes [23], it was unaltered or stimulated in placental frag- Arch Biochem Biophys 1980; 200:79-85. ments [23] or explants [38]. 10. Kliman HJ, Nestler JE, Sermasi E, Sanger JM, Strauss III JE Purifi- The relatively small, albeit significant, increase in pro- cation, characterization, and in vitro differentiation of cytotrophoblasts gesterone synthesis achieved with 25-hydroxycholesterol from human term placentae. Endocrinology 1986; 118:1567-1582. alone in human syncytiotrophoblasts of the present study 11. Toaff ME, Schleyer H, Strauss III JE Metabolism of 25-hydroxycho- lesterol by rat luteal mitochondria and dispersed cells. Endocrinology was surprising, considering that this soluble substrate 1982; 111:1785-1790. caused a marked increase in progesterone formation by 12. Lambeth JD, Kitchen SE, Farooqui AA, Tuckey R, Kamin H. Cyto- short-term incubates of human cytotrophoblasts [10]. The chrome P-450c,,-substrate interactions. Studies of binding and cata- differences may reflect the use of cytotrophoblasts versus lytic activity using hydroxysterols. J Biol Chem 1982; 257:1876- syncytiotrophoblasts and the in vitro conditions used in the 1884. 13. Bradford MM. A rapid and sensitive method for the quantitation of two studies. Because steroidogenic acute regulatory (StAR) microgram quantities of protein utilizing the principle of protein-dye protein, which regulates mitochondrial cholesterol translo- binding. Anal Biochem 1976; 72:248-254. cation in other tissues [39], is not expressed in the human 14. Brunk CE Jones KC, James TW. Assay for nanogram quantities of placenta [40], cholesterol transport to the P450s,, site of the DNA in cellular homogenates. Anal Biochem 1979; 92:497-500. inner mitochondrial membrane may be facilitated by other 15. Wunsch DM, Anderson LD, Pepe GJ, Albrecht ED. Regulation of progesterone formation by human placental cells in culture. Endocri- means in the primate placenta. Although this may explain nology 1986; 119:998-1003. in part the relatively minor effect of 25-hydroxycholesterol 16. Kubo M, Strott CA. Differential activity of 3-hydroxy-3-methylglu- on progesterone synthesis in syncytiotrophoblasts of the taryl coenzyme A reductase in zones of the adrenal cortex. Endocri- present study, additional investigation is required to eluci- nology 1987; 120:214-221. date this aspect of steroidogenesis. 17. Wittmaack FM, Holt JA, Schreiber JR. Cholesterol metabolism in es- trogen-sensitive progestin synthesis by rabbit corpus luteum. Am J In summary, the present study shows that, in cultures of Physiol 1986; 251(Endocrinol Metab 14):E457-E463. human syncytiotrophoblasts, estrogen enhanced P450sc ac- 18. Miller KW, Small DM. Triolein-cholesteryl oleate-cholesterol-lecithin tivity but not the activity of the de novo or deesterification emulsions: structural models of triglyceride-rich lipoproteins. Bio- pathways for cholesterol formation. Collectively, the results chemistry 1983; 22:443-451. obtained in vitro with human syncytiotrophoblasts and in 19. Chung BC, Matteson KJ, Voutilainen R, Mohandas TK, Miller WL. Human cholesterol side-chain cleavage enzyme, P-450,cc: cDNA clon- vivo in pregnant baboons indicate that there is a selective ing, assignment of the gene to chromosome 15, and expression in the estrogen-dependent developmental regulation of receptor- placenta. Proc Natl Acad Sci USA 1986; 83:8962-8966. mediated uptake of cholesterol substrate and its metabolism 20. Boguslawski W, Sokolowski W. HMG-CoA reductase activity in the to progesterone. Therefore, we propose that estrogen has a microsomal fraction from human placenta in early and term pregnan- role in regulating the P450¢cc component of the progester- cy. Int J Biochem 1984; 16:1023-1026. 21. Mikani K, Nishikawa T, Saito Y, Tamura Y, Matsuoka N, Kumagai one biosynthetic pathway that signals functional/biochem- A, Yoshida S. Regulation of cholesterol metabolism in rat adrenal ical differentiation of syncytiotrophoblasts during primate glands: effect of adrenocorticotropin, cholesterol, and corticosteroids pregnancy. on acyl-coenzyme A synthetase and cholesterol ester hydrolase. En- docrinology 1984; 114:136-140. 22. Nishikawa T, Mikami K, Saito Y, Tamura Y, Kumagai A. Studies on ACKNOWLEDGMENTS cholesterol esterase in the rat adrenal. Endocrinology 1981; 108:932- 936. 23. Genti Raimondi S, Olivier NS, Patrito LC, Flury A. Regulation of the We sincerely appreciate the excellent secretarial assistance of Mrs. 31-hydroxysteroid dehydrogenase activity in tissue fragments and mi- Wanda James with the computer preparation of the figures and manuscript, crosomes from human term placenta: kinetic analysis and inhibition and the technical assistance of Mrs. Laura Watkins and Mr. Gregory Kane. by steroids. J Steroid Biochem 1989; 32:413-420. 278 BABISCHKIN ET AL.

24. Tulchinsky D. Placental secretion of unconjugated , estradiol 32. Ewing LL, Chubb CE, Robaire B. Macromolecules, steroid binding and into the maternal and the fetal circulation. J Clin Endocri- and secretion by rabbit testes. Nature 1976: 264:84-85. nol & Metab 1973; 36:1079-1087. 33. Billiar RB, Pepe GJ. Albrecht ED. Immunocytochemical identification 25. Graves PE, Uzgiris VI, Querner M, Kashiwagi K, Mcintosh EN, Sal- of the estrogen receptor in the nuclei of cultured human term placental hanick HA. Interaction of cyanoketone and other steroid nitriles with trophoblasts. Placenta 1996; (in press). cytochrome oxidase, hemoglobin, and cytochrome P-450. Endocri- 34. Moise J, Ilekis J, Scommegna A, Benveniste R. The effect of dehy- nology 1978; 102:1077-1087. droepiandrosterone sulfate on de novo and low-density lipoprotein- 26. Henson MC, Pepe GJ, Albrecht ED. Transuterofeto placental conver- stimulated progesterone secretion by human choriocarcinoma JEG-3 sion of pregnenolone to progesterone in -treated baboons. cells. Am J Obstet Gynecol 1986; 154:1080-1085. Endocrinology 1987; 121:1265-1271. 35. Strauss III JF Menon KMJ. Lipoprotein and Cholesterol Metabolism 27. Babischkin JS, Pepe GJ, Albrecht ED. Regulation of progesterone in Steroidogenic Tissues. Philadelphia. PA: George E Stickley Co.: biosynthesis by estrogen during baboon pregnancy:placental mito- 1985. chondrial cholesterol side-chain cleavage activity in antiestrogen 36. Babischkin JS, Pepe GJ, Albrecht ED. Estrogen regulation of placental (ethamoxytriphetol, MER-25)-treated baboons. Endocrinology 1989; P-450 cholesterol side-chain cleavage enzyme messenger ribonucleic 124:1638-1645. acid levels and activity during baboon pregnancy. Endocrinology 28. Henson MC, Babischkin JS, Pepe GJ. Albrecht ED. Effect of the 1996; (in press). 37. Rabe T. Kiesel L, Runnebaum B. Regulation of human placental pro- Downloaded from https://academic.oup.com/biolreprod/article/56/1/272/2760805 by guest on 29 September 2021 antiestrogen ethamoxytriphetol (MER-25) on placental low density li- J Steroid poprotein uptake and degradation in baboons. Endocrinology 1988: gesterone synthesis in vitro by naturally occurring steroids. Biochem 1985; 22:657-664. 122:2019-2026. 38. Begum-Hasan J, Murphy BEP. In vitro stimulation of placental pro- 29. Strauss Ill JE Kido S, Sayegh R, Sakuragi N, Gafvels ME. The cAMP gesterone production by 19-nortestosterone and C19 steroids in early signalling system and human trophoblast function. Placenta 1992; 13: human pregnancy. J Clin Endocrinol & Metab 1992: 75:838-845. 389-403. 39. Stocco DM, Clark BJ. Regulation of the acute production of steroids 30. Henson MC, Pepe GJ. Albrecht ED. Regulation of placental low den- in steroidogenic cells. Endocr Rev 1996; 17:221-244. sity lipoprotein uptake in baboons by estrogen: dose-dependent effects 40. Sugawara T Holt JA, Driscoll D, Strauss III JF, Lin D, Miller WL, of the antiestrogen ethamoxytriphetol (MER-25). Biol Reprod 1991; Patterson D, Clancy KP. Hart IM, Clark BJ. Human steroidogenic 45:43-48. acute regulatory protein: functional activity on COS-I cells, tissue 31. Baggia S. Albrecht ED, Babischkin JS, Pepe GJ. Interconversion of specific expression, and mapping of the structural gene to 8p 11.2 and cortisol and cortisone in baboon trophoblast and decidua cells in cul- a pseudogene to chromosome 13. Proc Natl Acad Sci USA 1995: 92: ture. Endocrinology 1990; 127:1735-1741. 4778-4782.