| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
-Hydroxylase/17,20-Lyase Cytochrome P450 Expression
Department of Population Health and Reproduction, School of Veterinary Medicine (F.M.M., A.J.C.), California Regional Primate Research Center (C.A.V.V.), and Center for Health and the Environment (J.W.O., B.L.L.), University of California, Davis, California 95616
Address all correspondence and requests for reprints to: Dr. Alan Conley, Veterinary Medicine, Population Health and Reproduction, University of California, One Shields Avenue, Davis, California 95616-8615. E-mail: ajconley{at}ucdavis.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/17,20-lyase cytochrome P450 (P450c17) antisera and a direct radiometric assay of 17,20-lyase activity were used to test the hypothesis that TCDD targets P450c17, thereby decreasing substrate availability for E2 synthesis by hLGC. P450c17 expression and 17,20-lyase activity were detected in hLGC with high levels of E2 secretion. Western immunoblot analysis demonstrated that TCDD treatment of hLGC decreased the expression of P450c17 by as much 50% (P < 0.05). TCDD exposure induced a 65% decrease in 17,20-lyase activity (P < 0.05), but no changes were seen in P450arom or in nicotinamide adenine dinucleotide phosphate (reduced)-cytochrome P450 oxidoreductase (reductase). Furthermore, the decreases in P450c17 and 17,20-lyase were proportional to the inhibition of E2 secretion. We conclude that the molecular target for endocrine disruption of hLGC by TCDD is P450c17, specifically decreasing the supply of androgens for E2 synthesis, and that it does not involve either P450arom or the redox partner protein reductase. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Studies investigating the molecular events that drive TCDD-induced cellular toxicity have been hampered by two major factors. Firstly, like many toxicants, the effects of TCDD differ among species, and the choice of animal models is crucial. Second, the ability to decipher mechanisms at the molecular level requires a relevant cellular or in vitro model system and the development of the proper tools. The use of human luteinizing granulosa cells (hLGC) in vitro has provided us with a useful model with which to study endocrine disruption in human tissues exposed to TCDD (9). Several studies in this laboratory as well others demonstrated that TCDD decreased E2 production by hLGC (10, 11, 12, 13, 14, 15). Logically, the effects of TCDD and other endocrine disruptors are more likely to be acutely evident if they alter a rate-limiting reaction rather than a reaction that has unlimited capability of metabolizing available substrate.
The results of our prior experiments suggested that the target for TCDD toxicity in hLGC is 17,20-lyase activity (14, 15), but we lacked the means to specifically test this hypothesis at that time. We recently raised antisera against recombinant human cytochrome P450 17-hydroxylase/17,20-lyase (16) and adopted the use of a radiometric assay previously developed (17) to directly estimate 17,20-lyase activity (18). Therefore, with these critical tools in hand, the present experiments were designed to test the hypothesis that TCDD inhibits estrogen synthesis by hLGC through effects on P450c17 by examining both protein expression and 17,20-lyase activity. In addition, as both P450c17 and P450arom activities are totally dependent on redox support by the flavoprotein nicotinamide adenine dinucleotide phosphate (reduced) (NADPH)-cytochrome P450 oxidoreductase (reductase), the level of this essential redox partner protein was measured in control and treated cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-diol-20-one [17- hydroxypregnenolone (17P5)] was purchased from Steraloids (Wilton, NH). [1,214C]Acetic acid (54 mCi/mmol) was purchased from NEN Life Science Products (Boston, MA). [21-3H]17P5 (25.9 µCi/µmol) was donated by Dr. Angela Brodie (University of Maryland Medical School, Baltimore, MD). All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO).
hLGC culture
hLGC were obtained from patients undergoing assisted reproduction treatments at the Northern California Fertility Medical Center (Roseville, CA), and all experiments using them were reviewed and approved by the human subjects advisory committee, University of California-Davis. The cells were processed as previously described (9, 15). hLGC were plated at 5 x 105 cells in 3 ml/60-mm plate or at 106 cells/100-mm plate with 5 ml conditioned medium (DMEM supplemented with antibiotic-antimycotic, 10% noncharcoal-stripped calf serum, and 2 USP/ml hCG).
TCDD treatment
Previous studies determined that a concentration of 10 nM TCDD effectively disrupts E2 production by hLGC (10, 14). Therefore, TCDD (10-5 M) dissolved in dimethylsulfoxide (DMSO) was added directly to each plate at a dilution of 1 µl/ml conditioned medium (10 nM final concentration). TCDD treatment started on the day after recovery. Medium was changed every other day thereafter, and TCDD treatment was added immediately after medium changes. Cells were cultured in TCDD for a total of 10 d; we have previously established that the number of cells at plating does not change with either time or TCDD treatment (10, 13, 14). On the last day of culture, medium was collected and stored frozen at -20 C for hormone analysis. Cells were then harvested by scraping in PBS, pelleted by centrifugation, and stored at -80 C for further processing as described below. Controls included cells treated with DMSO (equal volume) or not.
Placental tissue
Human term placentas were obtained from patients after normal deliveries in the Obstetrics and Gynecologic Clinic at Sutter Hospital (Davis, CA) and were used as negative control tissues recognized not to express P450c17 (19) in 17,20-lyase activity assays, also after review and approval by the human subjects advisory committee, University of California-Davis. Tissues were collected within 3 h of delivery, transported on ice to the laboratory, and stored at -80 C until used for microsome preparation.
Microsome-enriched fraction preparation
Cell pellets from 100-mm dishes were resuspended in 1 ml homogenization buffer [0.1 M KHPO4 (pH 7.4), 20% glycerol, and 5 mM ß-mercaptoethanol] and 0.5 mM of the protease inhibitor phenylmethylsulfonylfluoride; after 10 strokes with a glass-glass homogenator, the cells were briefly sonicated (twice, 3 sec each time). For placenta microsome preparation an additional step of tissue grinding in the same homogenization buffer was performed (twice, 30 sec each time; until homogeneous tissue suspension was observed). Microsomal fractions were prepared by subcellular fractionation (20). The 100,000 x g pellet containing the microsomes was resuspended in 0.1 M KHPO4, 20% glycerol, 5 mM ß-mercaptoethanol, and 1 mM 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate and stored at -80 C. Protein determinations were made of the microsome-enriched preparations using the bicinchoninic acid protein assay reagent (Pierce Chemical Co., Rockford, IL) with BSA as the standard.
RIA for E2
Concentrations of E2 were determined on conditioned medium from control and TCDD-treated cells using available commercial RIA kits (Diagnostic Products, Los Angeles, CA) as previously reported (9, 10, 14, 15). The interassay coefficient of variation was less than 5%.
Western immunoblot
Twenty micrograms of microsomal proteins from control and TCDD-treated cells were separated on an 8% SDS-PAGE and transferred by electroblot onto a polyvinylidene difluoride membrane. Membranes were blotted for P450c17 with a polyclonal antisera (1:2000 dilution) raised in chickens, recently validated for primate tissue in our laboratory (16). Further verification of the specificity of the antisera for P450c17 protein was obtained in a separate experiment that included 15 µg hLGC microsomal protein, 10 µg pig testes microsome, and 0.1 µg human recombinant P450c17 (produced in our laboratory). Immunodetectable proteins were visualized with a horseradish peroxidase-linked antichicken secondary antibody (1:5000 dilution) and developed by chemiluminescence (ECL, Amersham Pharmacia Biotech, Arlington Heights, IL). Autoradiographic films were scanned, and the relative amount of protein for each blot was determined by densitometry.
Cytochrome P450 enzyme activity
Two microsomal cytochrome P450 enzyme activities were analyzed: 17,20-lyase and P450arom. Microsomal protein from control and TCDD-treated cells were assayed for 17,20-lyase activity using a procedure adapted in our laboratory and described recently (18), with minor modifications. Briefly, the assay is based on the release of the radiolabeled acetic acid from [21-3H]17P5 after cleavage of the C17–20 bond that therefore provides a direct estimate of 17,20-lyase activity in the sample (Fig. 1
) by measurement of radioactivity in the aqueous phase. One hundred micrograms of microsomal proteins from control and TCDD-treated hLGC were incubated at 37 C in assay buffer (50 mM KHPO4, 1 mM EDTA, and 1 mM 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate), at a final volume of 1 ml in the presence of 10.5 µM 17P5 (final concentration), consisting of 7 µM radiolabeled [21-3H]17P5 and 3.5 µM unlabeled 17P5 substrate. Reactions were continued for 6 h. Samples were analyzed in two assays, the internal standards (18.5 nmol/mg·h) for which were within 0.5 nmol/mg·h of each other. The limit of detection for the 17,20-lyase activity was 20 pmol/tube·h. The P450arom assay was performed on 50 µg microsomal protein from control and TCDD-treated hLGC and 100 µg placenta microsomal protein following a protocol described previously (14).
|
Data analysis
All data are expressed as the mean ± SEM of the treatment groups. Hormonal concentration data were log-transformed before statistical analysis and are expressed as nanograms of E2 per number of cells initially plated per unit of time. Results for the immunoblot analysis of P450c17 are presented as relative units of protein expression for each treatment after the densitometric analysis as explained above. The results for the 17,20-lyase and P450arom activity assays are expressed as picomoles of substrate converted per milligram of protein per hour. Comparisons for lyase activity and the Western blot data were made by one-way ANOVA, followed by a pairwise multiple comparison procedure using the Student-Newman-Keuls method. E2 production in culture was analyzed by two-way ANOVA, using the patient as the experimental block. Differences with P < 0.05 were considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
); thus, all future comparisons reported in reference to TCDD treatment effects were made using the DMSO vehicle-treated cell cultures as the control group. The E2 concentration was significantly reduced by TCDD treatment from 6.0 ± 2.2 to 2.2 ± 0.8 ng/ml (n = 12; P < 0.001). As we previously reported (14), there was a wide range in the level of E2 production among patients (0.1–23.2 ng/ml) necessitating that the data be log-transformed for statistical analysis. However, it is important to note that E2 production by cells from individual patients was inhibited regardless of the level of basal secretion for that patient, and there was no correlation between the level of E2 and the degree of TCDD-induced inhibition.
|
and 4A). Pig testis microsomal protein, used as a secondary internal reference, also demonstrated a single immunodetectable band, if at a slightly smaller apparent molecular size. Microsomal proteins from cells of patients producing sufficient E2 (>1 ng/ml) were analyzed further for immunodetectable P450c17 after TCDD treatment. Microsomal proteins (20 µg) from control and TCDD-treated cells were run in adjacent lanes, and comparisons were made among the DMSO control and TCDD treatment for hLGC from each patient. TCDD decreased P450c17 protein expression in cells from all patients (Fig. 4A). Based on densitometric analysis of immunodetectable protein levels, TCDD decreased the expression (P < 0.05) of P450c17 in hLGC by over 50% (Fig. 4B). There was no change observed in immunodetectable levels of reductase (data not shown).
|
|
; P < 0.05), and this decrease corresponded very closely to the inhibition of E2 production (Fig. 2). In contrast to 17,20-lyase, there was no effect of TCDD on P450arom or reductase activities of hLGC. P450arom and reductase activities in hLGC microsomes were comparable to those obtained with human placenta microsomes (Table 1), which exhibited no detectable 17,20-lyase activity.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The in vitro culture of hLGC permits the investigation of the specific effects of TCDD on estrogen production in a relevant human cell type. Studies in this laboratory (10, 11, 13, 14, 15) and others (12) have shown consistently that TCDD inhibits E2 production by hLGC, but the molecular target(s) of toxicity has not been identified. We reasoned that at the very least, TCDD was likely to inhibit the activity of the rate-limiting step in estrogen synthesis by these cells, and the results of the current experiments indicate that this is the 17,20-lyase activity of P450c17. This is consistent with several previously reported observations from studies conducted in our laboratory. First, TCDD treatment resulted in relatively little change in the levels of either progesterone or 17-hydroxyprogesterone, suggesting that the enzymatic step in the steroidogenic cascade most affected by TCDD must occur after cytochrome P450 side-chain cleavage and the action of 3ß-hydroxysteroid dehydrogenase on 
5-C21 steroids (14). Similarly, neither P450arom protein expression nor activity was affected by TCDD (14), and neither was correlated with E2 secretion. In the same study TCDD did not change the ability of hLGC to metabolize E2, also consistent with the effects of toxicity inhibiting synthesis rather than increasing clearance (14). Furthermore, we and others have shown that the addition of androgens can reverse the negative effect of TCDD on E2 production in vitro, indicating that the supply of substrate for aromatase is limited in TCDD-treated cells (12, 14, 26). The lack of measurable androgens in medium under basal culture conditions (15) also suggests that aromatization occurs more rapidly in hLGC than does the synthesis of androgens. Therefore, we (14, 15), as have others (12), proposed that the 17,20-lyase activity of P450c17 limits the rate of E2 production by hLGC.
In contrast to the lack of effect of TCDD on P450arom and consistent with the above-stated hypothesis, the present results show clearly that both P450c17 protein and 17,20-lyase activity were reduced by toxicant exposure. Just as importantly, the degree of enzyme inhibition correlated closely with the decrease in E2 production. Collectively, these data indicate that the enzymatic step inhibited by TCDD that results in reduced estrogen synthesis involves 17,20-lyase, and that it is brought about at least in part by a decrease in the expression of P450c17. No effect was seen on the microsomal redox partner protein, reductase. Moreover, as another microsomal P450 also involved in the steroidogenic pathway leading to estrogen synthesis, namely P450arom, is not affected by TCDD exposure, this effect appears to be specific for P450c17 expression. Although aromatization is commonly held to be the rate-limiting step in estrogen biosynthesis, this is clearly not consistent with prior (12, 14, 15) or our current results. The data reported herein indicate that there are substantial differences in the capacity of hLGC to synthesize androgens and estrogens; 17,20-lyase was 15-fold lower than P450arom activity in cells exhibiting the highest levels of 17,20-lyase activity. As noted above, hLGC provides a good model of human luteal function (9); thus, we believe that this has great relevance in understanding the control of estrogen production by the human corpus luteum.
In the context of physiological and toxicological relevance to human health, it is important to note that the levels of 17,20-lyase and aromatase activities reported here for hLGC are within the range of values obtained ex vivo. Sano et al. (34) investigated steroid metabolism by human ovarian tissues collected at various stages of the menstrual cycle. Aromatase activity of cells in our study (2300 pmol/mg·h) was higher than that found in late luteal tissue (around 600 pmol/mg·h). However, our estimates were also made on enriched microsomal protein, whereas those by Sano et al. (34) were made on tissue lysates and would therefore be expected to be lower. Similarly, the estimates of 17,20-lyase activity in late luteal tissue were also lower than those found in hLGC. Regardless, aromatase was also higher than 17,20-lyase in human luteal lysates (34). Changes in P450c17 and P450arom transcripts during the menstrual cycle generally support this view (35, 36), and previous studies have also found relatively low levels of P450c17 mRNA in hLGC. However, circulating androgens of luteal origin increase in parallel with E2 in women during the midluteal phase of the menstrual cycle and increase further with E2 immediately after implantation in conceptive menstrual cycles (37). This early pregnancy-associated rise in androgens and E2 is most likely due to an increase in ovarian P450c17 expression in response to hCG that drives steroid synthesis before placental steroidogenic competency is established. The observation that these midluteal and postimplantation rises in androgens are similar within women, but highly variable between women, can be interpreted to indicate that the ability to induce P450c17 varies between individuals. Individual differences in P450c17 also provide a plausible explanation of the observed ethnic differences in androgen levels during the menopausal transition (38). Variability in the level of expression of a primary target of toxicity may also explain individual differences in susceptibility to TCDD toxicity. Therefore, P450c17 may become an especially sensitive target for TCDD-induced endocrine disruption in women during the late luteal phase, early pregnancy, and even later in life when it is perhaps more limiting than P450arom in estrogen synthesis.
It is also significant that the effect of TCDD is specific for P450c17 and does not affect P450arom expression, if only because TCDD induces the transcriptional expression of other microsomal P450s (39, 40). Studies are in progress to determine whether there is a decrease in transcript or perhaps an increased rate of P450c17 degradation, although this seems less likely based on the specificity of the change in levels for this microsomal P450. However, even if P450c17 transcript levels do decline, this does not necessarily imply that the effect of TCDD is directly on CYP17. Putative dioxin-response elements in the human CYP17 gene are fully 3 kb upstream from the transcriptional start site (Dr. Mike Denison, University of California-Davis, unpublished observations). Others have reported a transient inhibition of E2 secretion before stimulation in the first 48 h of culture, but only when micromolar concentrations of TCDD were used (12). In our studies E2 secretion by hLGC is highly variable in the first 2 d of culture, as is the response to TCDD, and a consistent depression begins by d 4 of culture (14). However, in contrast to others (12), in our studies hLGC are cultured in the continued presence of hCG and nanomolar concentrations of TCDD. Studies in vivo and in vitro demonstrated that TCDD decreased gonadotropic support or response (41, 42, 43), and this may still play a role in the response of reproductive tissues to toxic exposure, for which numerous disease consequences have been suspected (44, 45). Although this mechanism could contribute to the reduction of P450c17 noted here, the lack of effect on P450arom argues strongly that is not a major factor, as both P450c17 and P450arom are believed to be responsive to gonadotropins in human ovarian cells (46, 47). It is not known whether P450c17 expression is a target for TCDD in other tissues, such as the adrenal cortex or testes. The means of transcriptional regulation of CYP17 expression is not necessarily the same in the adrenal gland and gonad, even though the same promoter region is used (48). Further studies are needed to elucidate the specific mode of action of TCDD in depressing P450c17 expression in hLGC.
In summary, to our knowledge this is the first report of P450c17 protein expression and 17,20-lyase activity in hLGC, and the demonstration of its susceptibility to TCDD toxicity. TCDD treatment decreased both the protein expression and the 17,20-lyase activity of P450c17 in microsomal fractions from hLGC by more than 50% and 60%, respectively. To our knowledge, this is the first report of a clear, specific molecular target of endocrine disruption resulting from TCDD toxicity in any human cell.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: DMSO, Dimethylsulfoxide; E2, estradiol; hCG, human chorionic gonadotropin; hLGC, human luteinized granulosa cells; NADPH, nicotinamide adenine dinucleotide phosphate (reduced); P450arom, aromatase cytochrome P450; P450c17, 17-hydroxylase/17,20-lyase cytochrome P450; 17P5, 17-hydroxypregnenolone; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin.
Received August 5, 2002.
Accepted for publication October 16, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase-C17,20-lyase inhibitors. Anal Biochem 267:319–330[CrossRef][Medline]
-hydroxylase/17,20-lyase) in cultured human granulosa cells. J Clin Endocrinol Metab 63:202–207[Abstract]
-hydroxylase/lyase activities in proliferating human theca interna cells in long term monolayer culture. Endocrinology 125:1959–1966[Abstract]
-hydroxylase gene expression in the bovine ovary: mechanisms regulating expression differ from those in adrenal cells. J Steroid Biochem Mol Biol 59:21–29[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  J. T. Sanderson The Steroid Hormone Biosynthesis Pathway as a Target for Endocrine-Disrupting Chemicals Toxicol. Sci., November 1, 2006; 94(1): 3 - 21. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. N. Fru, C. A. VandeVoort, and C. L. Chaffin Mineralocorticoid Synthesis During the Periovulatory Interval in Macaques Biol Reprod, October 1, 2006; 75(4): 568 - 574. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Myllymaki, T. E. Haavisto, L. J. S. Brokken, M. Viluksela, J. Toppari, and J. Paranko In Utero and Lactational Exposure to TCDD; Steroidogenic Outcomes Differ in Male and Female Rat Pups Toxicol. Sci., December 1, 2005; 88(2): 534 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L.-A. Li and P.-W. Wang PCB126 Induces Differential Changes in Androgen, Cortisol, and Aldosterone Biosynthesis in Human Adrenocortical H295R Cells Toxicol. Sci., May 1, 2005; 85(1): 530 - 540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Payne and D. B. Hales Overview of Steroidogenic Enzymes in the Pathway from Cholesterol to Active Steroid Hormones Endocr. Rev., December 1, 2004; 25(6): 947 - 970. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B.-M. Huang, K.-Y. Hsiao, P.-C. Chuang, M.-H. Wu, H.-A. Pan, and S.-J. Tsai Upregulation of Steroidogenic Enzymes and Ovarian 17{beta}-Estradiol in Human Granulosa-Lutein Cells by Cordyceps sinensis Mycelium Biol Reprod, May 1, 2004; 70(5): 1358 - 1364. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
