This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hattori, N. Articles by Ueda, M. Search for Related Content PubMed PubMed Citation Articles by Hattori, N. Articles by Ueda, M. Pubmed/NCBI databases Gene GEO Profiles HomoloGene UniGene

Epoxide Hydrolase Affects Estrogen Production in the Human Ovary*

Department of Gynecology and Obstetrics, Faculty of Medicine (N.H., H.F., S.F.), Institute for Frontier Medical Science (M.M.), and Institute for Virus Research (M.U.), Kyoto University, Sakyo-ku, Kyoto, 606-8507, Japan

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
To investigate the mechanisms of ovarian cell differentiation, we raised a new monoclonal antibody, HCL-3, which reacted with human luteal cells. It also reacted with human and porcine hepatocytes. The immunoaffinity-purified HCL-3 antigen from human corpora lutea (CL) was shown to be a 46-kDa protein. The N-terminal 22 amino acids of the 46-kDa protein from porcine liver exhibited high homology (82%) to human microsomal epoxide hydrolase (mEH). The purified HCL-3 antigen from human CL or porcine liver showed EH enzyme activity, confirming that HCL-3 antigen is identical to mEH, which is reported to detoxify the toxic substrates in the liver. In human follicles, mEH was immunohistochemically detected on granulosa and theca interna cells. In the menstrual and pregnant CL, mEH was also expressed on large and small luteal cells. A competitive inhibitor of EH, 1,2-epoxy-3,3,3-trichloropropane, inhibited the conversion of estradiol from testosterone by granulosa cells cultured in vitro, indicating the involvement of mEH in ovarian estrogen production. Because anticonvulsant sodium valproate and its analogues were reported to inhibit EH enzyme activity, these findings provide a new insight into the etiology of endocrine disorders that are frequently observed among epileptic patients taking anticonvulsant drugs.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
TO IDENTIFY THE differentiation-related molecules in ovarian cells, we raised several monoclonal antibodies (mAbs) against human granulosa cells (1, 2, 3). By analyzing the detected antigens, several new mechanisms in the regulation of ovarian cell function have been proposed (4, 5, 6).

Here, we show a new mAb, named HCL-3, raised against the luteal cells. An analysis of the partial amino acid sequence of purified antigen demonstrated that HCL-3 detected microsomal epoxide hydrolase (mEH, EC3.3.2.3). EH is an enzyme that converts epoxide to diol by adding H₂O to epoxide (7). mEH is expressed in hepatocytes and is considered to play an important role in detoxification of several substances (7). It was reported that this enzyme was also expressed on human adrenal gland, but its physiological role remains unclear (8). In 1977, Oesch et al. (9) reported that rat gonads contained relatively high activity of mEH, but the precise localization or function of this activity has not yet been investigated. In the present study, we confirmed the enzyme activity of mEH in human corpora lutea (CL) in the ovary and examined the precise localization of mEH in human follicles and CL of various developmental stages.

Recently, anticonvulsant sodium valproate and its analogues were revealed to inhibit EH enzyme activity (10, 11). It also has been reported that endocrine disorders, including polycystic ovary syndrome, are associated with the sodium valproate administration (12, 13). In contrast to the so-called polycystic ovary syndrome, these drug-induced disorders were not associated with a high level of serum LH, although the morphological changes of polycystic ovary and hyperandrogenism were exhibited (12, 13). At present, no definite mechanism has been proposed for the induction of polycystic ovary syndrome by sodium valproate. In this paper, to examine the physiological role of mEH in ovarian function in relation to the etiology of anticonvulsant drug-induced endocrine disorders, we also investigated the effect of an inhibitor for EH on steroid hormone production by human luteinizing granulosa cells in vitro.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Tissues
Ovarian tissues, follicles, and CL of the menstrual cycle (for microscopic observation, immunization, and enzyme assay) were obtained from 57 women, 14–46 yr old, who had undergone unilateral ovarian cystectomy or oophorectomy and contralateral wedge resection for the treatment of benign ovarian cysts. All women had normal menstrual cycles (28–30 days) and a normal luteal phase length on ovulatory normal basal body temperature charts, except a 14-yr-old girl with ovarian cysts. Six CL of pregnancy were obtained from patients, 25–33 yr old, who had ectopic pregnancies. Three CL of pregnancy were obtained from patients, 27–46 yr old, who had undergone hysterectomy because of uterine myoma and carcinoma in situ of the uterine cervix; in all three patients, fetal growth was normal on ultrasonographical examination. Macroscopically and microscopically normal regions of these tissues were used for this study. Human liver sample was obtained from patients who received lobectomy of liver because of hepatoma. Porcine ovaries and livers were obtained from a local slaughterhouse. All human materials described above were obtained with informed consent.

Production and selection of mAb
Luteal cells were isolated as described, with minor modification (14). Briefly, the CL were separated from connective tissue, then minced with scissors. They were incubated in RPMI 1640 medium (Flow Labs., Irvine, Scotland, UK) containing 0.2% collagenase (Wako Pure Chemical Industries Ltd., Osaka, Japan) and 0.01% deoxyribonuclease I (Sigma, St. Louis, MO), at 37 C for 20 min, in a test tube. Tissue fragments were then incubated in RPMI 1640 containing 10% FCS (Flow Laboratories, McLean, VA), 1000 U/ml dispase (Godo Shusei, Tokyo, Japan), and 0.01% deoxyribonuclease I, at 37 C for 30 min. After the undissociated materials were discarded, the cell suspension was collected and centrifuged for 5 min at 160 x g. Some of the cells were washed twice and suspended in PBS and immediately used for the immunization, and the rest were stocked in liquid nitrogen for further immunization. Eight-week-old BALB/c mice were injected ip with 2 x 10⁶ luteal cells every 3 weeks, for 9 weeks. The spleen cells of an immunized mouse were fused with X63Ag8.653 mouse myeloma cells, using 50% polyethylene glycol 1500 (BDH Chemicals, England, UK) on the third day after the last immunization (15). The fused cells were cultured in RPMI medium supplemented with 15% FCS and 10% BM-condimed H1 (Boehringer, Germany), including HAT (Flow Co., Scotland, UK), in 96-well plates. An indirect immunofluorescence method, using frozen sections of human CL as described below, was used for screening the supernatants of growing hybridomas. The positive hybridomas were cloned twice and were injected ip into mice previously treated with pristane (2,6,10,14-tetramethylpentadecane, Tokyo Kasei Co., Tokyo, Japan). IgG was purified from ascitic fluids with Affi-Gel protein A (Bio-Rad Laboratories, Inc., Richmond, CA). Ig isotype was determined using an isotype kit for mouse mAbs (Serotec Ltd., Oxford, UK).

Immunohistochemical examination of the antigen expression on various follicles, CL, and other tissues
The following follicles and CL were used for examination (Tables 1 and 2): primordial and primary follicles (n = 10), preantral follicles (n = 2), small follicle (less than 1 mm in diameter, n = 1), growing follicles (4–16 mm in diameter, n = 7), preovulatory follicles (18 and 20 mm in diameter, n = 2), atretic follicles (2–16 mm in diameter, n = 6), CL on CL day 2 (the day after ovulation, n = 2), CL on day 3 (n = 2), CL on day 4 (n = 4), CL on day 5 (n = 2), CL on day 6 (n = 3), CL on day 7 (n = 4), CL on day 8 (n = 2), CL on day 9 (n = 4), CL on day 10 (n = 1), CL on day 11 (n = 2), CL on day 12 (n = 3), CL on day 13 (n = 1), CL on day 14 (n = 2), and CL of pregnancy at 6, 7, 8, 9, 10, 13, 14, and 15 weeks of gestation (n = 9). Follicles obtained in the follicular phase with granulosa cells having mitotic figures and regularly shaped nuclei, cytoplasm, and stratified layers were classified as growing follicles. Follicles that were irregularly shaped with blood cell invasion and lacked mitotic figures were classified as atretic (16). If the above judgment was difficult with cryosections of follicles, the hematoxylin and eosin stained sections from the identical samples that were fixed with 10% formalin and embedded with paraffin were used. The postovulatory date of CL was evaluated according to the histological dating described by Corner GW, using hematoxylin and eosin-stained tissue sections of 10% formalin-fixed and paraffin-embedded samples (17). In this work, the term "corpus luteum (CL) day" was used according to his definition. For example, CL day 2 is the day after ovulation.

Purification of HCL-3 antigen from human CL and porcine liver
The HCL-3 antigen was purified from human CL and porcine liver as described previously, with slight modification (2). One gram of tissue was homogenized with Polytron (Kinematica AG, Luzern, Switzerland) in 10 ml of ice-cold 40 mM phosphate buffer (pH 7.3) containing 5 mM EDTA, 150 mM NaCl, 1% CHAPS (cholamidopropyl-dimethylammonio-propanesulfonic acid, Wako Pure Chemical Industries Ltd.) and the protease inhibitors, 1 mM (p-amidinophenyl) methanesulfonyl fluoride hydrochloride (Wako Pure Chemical Industries Ltd.), 10 µg/ml leupeptin, and pepstatin (Peptide Institute, Osaka, Japan). After centrifugation (10,000 x g, 20 min, 4 C), the concentration of CHAPS in the lysate was reduced to 0.3%, by dilution. The lysate was passed through the column of Affi-gel 10 that was conjugated with anti-TNP mouse IgG1 mAb (2 mg IgG/ml gel). The lysate was then incubated with 0.2 ml Affi-gel 10 that was conjugated with HCL-3 or anti-TNP mAb (2 mg IgG/ml gel) at 4 C for 3 h. After sufficient washing with the buffer, the antigen was eluted with 0.5 M NH₄OH containing 0.1% nonidet P-40 (Iwai Chemicals, Tokyo, Japan). The eluate was dried in vacuo at room temperature. The samples were dissolved in lysis buffer with 0.1 M dithiothreitol (Wako Pure Chemical Industries Ltd.) and were separated by SDS-PAGE (12% gel). Proteins in the gel were visualized by the silver stain kit (Wako Pure Chemical Industries Ltd.).

Partial amino acid sequencing of the 46-kDa protein purified from porcine liver
Porcine liver (45 g) was homogenized with Polytron in 500 ml of 40-mM phosphate buffer containing 5 mM EDTA, 150 mM NaCl, 1% CHAPS, and 1 mM (p-amidinophenyl) methanesulfonyl fluoride hydrochloride, as described above. After absorbing the nonspecific binding, the lysate was incubated with 1 ml affi-gel 10 conjugated with HCL-3 mAb at 4 C for 4 h. The antigen was eluted with 0.5 M NH₄OH containing 0.1% nonidet P-40. This procedure was repeated three times. The purified antigen was dissolved in lysis buffer for SDS-PAGE. After the electrophoresis, the proteins in the polyacrylamide gel were transblotted onto polyvinylidene difluoride (PVDF) membrane (Millipore Corp., Bedford, MA) in Tris/Boric acid buffer. The protein on PVDF membrane was stained with Coomassie blue, and the 46-kDa protein band was cut and analyzed with an amino acid sequencer PSQ-1 (Shimazu Co., Kyoto, Japan). The SWISS-PROT and GenBank data bases were used in the analysis of amino acid sequence homology.

Preparation of microsomal fraction of CL and porcine liver
Tissue (1 g) was homogenized in 5 ml PBS with Dounce homogenizer. The homogenate was centrifuged at 10,000 x g for 20 min, and the supernatant was centrifuged at 100,000 x g for 60 min. The precipitate was suspended in 1 ml PBS and was used as microsomal fraction.

Assay of epoxide hydrolase activity
Enzyme activity was assessed by the production rate of 7-(2',3'-dihydroxy) propoxycoumarin (DHC) from 7-glycidoxycoumarin (GOC), as described by Inoue et al. (19). Fifty microliters of the microsomal fraction and 925 µl of 173-mM Tris HCl (pH 8.6) were incubated at 37 C for 5 min. In some experiments, HCL-3 antigen-binding gel was used instead of microsomal fraction. Twenty-five microliters of GOC (2 mM) in acetone were added to the mixture and incubated at 37 C for 10 min. The reaction was stopped by cooling in ice water. Five milliliters of benzene were immediately added, and the mixture was centrifuged at 2,000 rpm for 5 min. The upper benzene fraction was discarded, and the lower water fraction was stored. This extraction procedure was repeated once more. The fluorescence intensity of the water-soluble fraction was monitored by fluorescence spectrometer F-2000 (Hitachi Scientific Instruments, Inc., Tokyo, Japan) at excitation 325 nm and emission 391 nm. DHC in methanol was used as a standard. GOC and DHC were synthesized by the method of Watabe et al. (20) and were given to us by Dr. T. Aimoto (Faculty of Pharmaceutical Science, Setsunan University, Osaka, Japan). The protein concentration was determined by the method of Lowry, and the amount of HCL-3 antigen on the Affi-gel was determined by comparing the intensities of visualized bands, by the silver staining method, after SDS-PAGE. BSA (crystallized, Sigma) was used as a standard.

Culture of human granulosa cells
Human granulosa cells were isolated from patients who had undergone treatment for in vitro fertilization-embryo transfer, as described (5). Briefly, patients receiving a GnRH agonist (buserelin acetate, Hoechst Marion Roussel, Inc. Ltd., Tokyo, Japan), beginning on the first day of the cycle, were hyperstimulated by im injection of human menopausal gonadotropin (HMG, Organon Japan Co. Ltd., Tokyo, Japan). Follicles were aspirated 36 h after the administration of hCG (5000 IU, im, Mochida Pharmaceutical Co. Ltd., Osaka, Japan). The suspension of granulosa cells was overlayered on Ficoll-Hypaque and centrifuged at 400 x g for 30 min. The cells collected from the interface were suspended in culture medium consisting of RPMI 1640 medium supplemented with 10% FCS, and 100 mg/ml kanamycin sulfate (Meiji Seika Ltd., Tokyo, Japan). To reduce contamination by adherent peripheral mononuclear cells, the collected cells were incubated, for 60 min at 37 C, in humidified air containing 5% CO2, in 60-mm diameter plastic tissue culture dishes (Falcon, Becton Dickinson and Co., Lincoln Park, NJ) pretreated with autologous serum (21). The cells were then collected after a mild wash, and viable granulosa cells suspended in culture medium (1 x 10⁵/ml, 0.1 ml/well) were inoculated into each well of 96-well plates (Falcon; Becton Dickinson and Co.). The next day (day 1), the medium was discarded, to remove unattached cells, and replaced with fresh medium [F-12/DMEM 1:1 (Life Technologies, Tokyo, Japan) containing 2 mg/ml BSA (Nitta Gelatin, Inc., Tokyo, Japan)]. The culture media were changed on day 3, and then cells were cultured in the presence or absence of 10^-7 M testosterone and various concentrations of 1,2-epoxy-3,3,3-trichloropropane (ETCP, 0–500 µM, Sigma), which is a competitive inhibitor of EH. The cultured media were collected on day 4 for steroid hormone assay.

Expression of HCL-3 antigen on cultured granulosa cells detected by indirect immunofluorescence
On days 1 and 4, the cultured granulosa cells in 8-chamber slides (Lab Tec, Chamber Slide, Nunc Inc., Naperville, IL) were washed gently three times with PBS containing 1 mM CaCl₂ and 1 mM MgCl₂, thoroughly dried, and fixed with acetone at -20 C. The slides were indirectly stained using HCL-3 and anti-TNP mAbs, as described above. Three independent experiments were performed.

Assay of steroid production by human luteinizing granulosa cells
The concentrations of progesterone and estradiol were measured using RIA kits (Daiichi Pharmaceutical Company Ltd. Radio Isotope Research Inc., Tokyo, Japan). Inter and intraassay coefficients of variation were 6.5% and 5.3% for the progesterone assay and 7.4% and 6.3% for the estradiol assay, respectively.

Effect of ETCP on the epoxide hydrolase activity of cultured granulosa cells
The granulosa cells (1 x 10⁵/ml, 0.3 ml/well) were inoculated into each well of 48-well plates (Coster, Corning, Inc., New York, NY). Cells were cultured as above; and, on day 3, various concentrations of ETCP were added to each well. Eight microliters of GOC (2 mM) in dimethylsulfoxide were added to the culture mixture. After 10 min at 37 C, the 48-well plates were cooled on ice cold water and 1 ml of benzene were immediately added, and DHC production in the water fraction was monitored by fluorescence spectrometer, as described above.

Effect of ETCP on aromatase activity in the granulosa cell and CL homogenate
The enzyme activity of aromatase in the granulosa cell and CL homogenate was measured by the method described previously, with slight modification (22, 23). Granulosa cells (3 x 10⁶) or CL tissue (50 mg) were homogenized in 2 ml of 40-mM phosphate buffer containing 0.25 M sucrose and 1 mM EDTA with sonicator. The supernatant was collected after centrifugation at 160 x g for 5 min, and was diluted with buffer to 3 mg protein/ml. The above buffer, containing ßnicotinamide adenine dinucleotide phosphate (reduced form, ß-NADPH, 2.2 mM, Sigma), dithiothreitol (2.2 mM), and testosterone (1.1 x 10^-7 M) was used as a substrate mixture for the aromatase reaction. For enzyme assay, 10 µl of each cell homogenate was added to the prewarmed substrate mixture (90 µl) containing ETCP (0–500 µM) in the tube and was incubated at 37 C for 10 min. The tube was dipped into ice-cold water to stop the reaction. The enzyme in the sample was then inactivated by boiling for 3 min. The concentration of estradiol in the reaction mixture was assayed by RIA, as described above. The boiling procedure did not affect the estradiol concentration detected by RIA assay. The estradiol production during the incubation period increased linearly for 15 min, indicating that the initial rate of aromatase enzyme activity was measured in this condition.

Statistical analysis
The concentrations of progesterone and estradiol are given as means ± SEM. The difference of progesterone and estradiol production was analyzed by the one-way ANOVA, followed by Scheffé’s F test. The difference was considered to be significant at P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Selection of HCL-3 mAb-producing hybridoma
By the indirect immunofluorescence staining method, clone 1.12F (HCL-3 mAb-producing clone) was found to be highly reactive to small luteal cells and weakly or moderately reactive to large luteal cells of human CL. HCL-3 mAb was not reactive to connective tissues or blood vessels around CL, or the cells in the central cavity in CL. The Ig isotype of HCL-3 mAb was IgG1.

HCL-3 antigen expression in the follicles and CL
In the primordial and primary follicles, HCL-3 antigen was not detected. In the preantral follicles that have two or three layers of granulosa cells, HCL-3 antigen was not detected in granulosa or theca interna cells. On the other hand, in a small follicle less than 1 mm in diameter that contained several layers of granulosa cells, HCL-3 antigen was well detected in the theca interna cells but not in the granulosa cells (Fig. 1 and Table 1). In the growing (4–16 mm in diameter) and preovulatory follicles, HCL-3 antigen was highly expressed in the theca interna cells and was also detected in the granulosa cells with weak intensity (Fig. 2). In the atretic follicles (2–16 mm in diameter), HCL-3 antigen was expressed weakly in granulosa cells and intensely in theca interna cells (data not shown).

View larger version (100K): [in this window] [in a new window]   Figure 1. Indirect immunofluorescence staining of human small follicles with HCL-3 mAb. A–C, A preantral follicle that has only two cell layers of granulosa cells; D–F, a primordial follicle and a small follicle containing several layers of granulosa cells; A and D, hematoxylin and eosin staining; B and E, staining with HCL-3 mAb; C and F, negative controls. In a primordial follicle (E, arrowhead) and a preantral follicle (B, arrowhead), HCL-3 antigen was not detected. In a small follicle that has several layers of granulosa cells, HCL-3 antigen was highly detected in the theca interna cells (E, arrow) but not in the granulosa cells. Bar, 100 µm.

View larger version (105K): [in this window] [in a new window]   Figure 2. Indirect immunofluorescence staining of growing and preovulatory follicles with HCL-3 mAb. A–C, An antral follicle of 4 mm in diameter; D–F, a preovulatory follicle of 18 mm; A and D, hematoxylin and eosin staining; B and E, staining with HCL-3 mAb; C and F, negative controls. In both follicles, HCL-3 antigen was expressed weakly on granulosa and intensely on theca interna cells (B and E). GC, Granulosa cells. TI, theca interna cells. Str, stromal cells. Bar, 100 µm.

View larger version (109K): [in this window] [in a new window]   Figure 3. Indirect immunofluorescence staining of CL, on days 3 and 8, with HCL-3 mAb. A–C, A CL on day 3; D–F, a CL on day 8; A and D, hematoxylin and eosin staining; B and E, staining with HCL-3 mAb; C and F, negative controls. In a CL on day 3 (B), HCL-3 antigen was expressed strongly on the peripherally located luteinizing theca interna cells (PLC) and weakly on the centrally located luteinizing granulosa cells (CLC). In a CL on day 8 (E), HCL-3 antigen was expressed intensely on small luteal cells (SL) and moderately on large luteal cells (LL). Cv, Cavity. Bar, 100 µm.

View larger version (117K): [in this window] [in a new window]   Figure 4. Indirect immunofluorescence staining of CL, on day 13 and at 7 weeks of gestation, with HCL-3 mAb. A–C, A CL on day 13; D–F, a CL at 7 weeks of gestation; A and D, hematoxylin and eosin staining; B and E, staining with HCL-3 mAb; C and F, negative controls. In the CL on day 13 (B), HCL-3 antigen was moderately expressed on large luteal cells and highly expressed on small luteal cells. In the CL, at 7 weeks of gestation (E), HCL-3 antigen was moderately expressed on large luteal cells and highly expressed on small luteal cells. BV, Blood vessel. Bar, 100 µm.

View larger version (87K): [in this window] [in a new window]   Figure 5. Immunoaffinity purification of HCL-3 antigen from human CL and porcine liver. HCL-3 antigen was purified by immunoaffinity chromatography from human CL and porcine liver using HCL-3-conjugated Affi-gel 10. The purified HLC-3 antigen was visualized by silver staining on 12% SDS-PAGE under reducing conditions. Lane 1, The proteins purified with HCL-3 mAb from human CL; lane 2, the proteins purified with anti-TNP mAb (control mAb) from human CL; lane 3, the eluate from the HCL-3-conjugated Affi-gel 10 without incubation in the tissue lysate; lane 4, the eluate from the anti-TNP-conjugated Affi-gel 10 without incubation in the tissue lysate; lane 5, the proteins purified with HCL-3 mAb from porcine liver; lane 6, the proteins purified with anti-TNP mAb from porcine liver. In lanes 1 and 5, a specific protein band was observed at the molecular mass of 46 kDa (arrowheads). Bars show gel top, molecular mass markers of 97, 66, 43, and 31 kDa from the top to bottom.

View larger version (91K): [in this window] [in a new window]   Figure 6. Indirect immunofluorescence staining of porcine liver with HCL-3 mAb, and partial amino acid sequencing of the 46-kDa protein purified from porcine liver. A and B, HCL-3 mAb reacted strongly to porcine liver; A, hematoxylin and eosin staining; B, staining with HCL-3 mAb. Bar, 100 µm. C, The N-terminal amino acid sequence of the 46-kDa protein purified from porcine liver, by immunoaffinity chromatography, using the HCL-3 mAb. The 22 amino acids from the N-terminus were highly homologous to human mEH (mEH, EC 3.3.2.3), with identities 18/22 and positives 20/22.

Assay of epoxide hydrolase activity
Antigenic molecules were affinity-purified from the microsomal fraction of human CL of the midluteal phase on HCL-3 mAb-conjugated Affi-gel 10. A part of the gel was used for the assay of EH enzyme activity; and the rest, for the determination of the amount of the 46-kDa protein. The EH enzyme activity of the protein, purified from human CL, was 84 µM DHC/10 min·µg 46-kDa protein (Table 3). The activity of the protein purified from the microsomal fraction of human liver was 65 µM DHC/10 min·µg 46-kDa protein. The enzyme activity of EH in the nonpurified microsomal fraction from human CL of midluteal phase and from human liver was 0.039 and 0.105 µM DHC/10 min·µg microsomal protein, respectively. The activity of the purified protein from porcine liver was 76 µM DHC/10 min·µg 46-kDa protein, and that of the nonpurified microsomal preparation of porcine liver was 0.053 µM DHC/10 min·µg microsomal protein. HCL-3 mAb (0.1 mg/ml) or CHAPS (0.3%) did not affect the enzyme activity of EH in the microsomal fraction from human CL. In the control incubation with the proteins which bound to anti-TNP-mAb-conjugated gels, slight production of DHC was also observed, but it did not differ from the DHC production in the control buffer, indicating that the proteins associated with anti-TNP mAb had no significant EH activity.

The progesterone and estradiol production of granulosa cells cultured with ETCP was assayed. Estradiol production was significantly suppressed by ETCP at 100 µM or more (Fig. 7A). Progesterone production was also reduced by ETCP, but only slightly (Fig. 7B).

View larger version (34K): [in this window] [in a new window]   Figure 7. The effect of ETCP on the production of estradiol and progesterone in the cultured granulosa cells. A, Estradiol production. In more than 100 µM ETCP, the production of estradiol during 24 h was significantly suppressed, as compared with the ETCP-nontreated group (**, P < 0.01). B, Progesterone production. At a high dose of ETCP, the production of progesterone seemed to be decreased, but this tendency was not significant at 200 or 500 µM (*, P < 0.05). The experiment was performed in triplicate at each concentration of ETCP, and the mean values were statistically calculated of five independent experiments.

View larger version (12K): [in this window] [in a new window]   Figure 8. Effect of ETCP on enzyme activity of aromatase and EH in the cell homogenate. A, The aromatase activity was detected in both granulosa cells and CL homogenate. The experiment was performed in triplicate at each concentration of ETCP, and the mean values were statistically calculated of three independent experiments. The estradiol production was not affected by ETCP, indicating that ETCP had no direct effect on enzyme activity. Open circle, Granulosa cells; closed circle, CL. Bars represent SD. B, EH enzyme activity in the homogenate of granulosa cells. The experiment was performed in duplicate at each concentration of ETCP, and the GHC production from GOC in each concentration was normalized by that in the absence of ETCP. Each value was statistically calculated of three independent experiments. Inhibitory effect of ETCP on the EH activity was obvious at 0.5 µM (50% inhibition), and at 1 µM or more EH activity was more intense. Bars represent SD.

View larger version (11K): [in this window] [in a new window]   Figure 9. The effect of ETCP on EH enzyme activity in the cultured granulosa cells. The experiment was performed in duplicates at each concentration of ETCP, and the GHC production in each concentration was normalized by that in the absence of ETCP of each experimental condition described. Each value was statistically calculated of three independent experiments. Open square, ETCP added simultaneously with GOC to the living cells; closed square, simultaneously to the disrupted cells by sonication; open circle, ETCP added to the living cells 5 h before the addition of GOC; closed circle, 24 h before. Bars represent SD.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

mEH is an enzyme that converts epoxide to diol by adding H₂O to it, and is widely distributed not only in mammals, but also in other animals (7). mEH in the liver has been reported to be important for inactivation of most carcinogenic or cytotoxic electrophilic epoxide (7). Substrates for this enzyme include epoxide derivatives of certain pharmaceuticals, such as metabolites of phenytoin, carbamazepine, and other antiepileptic medications (30, 31). Epoxides of environmental toxins, such as the carcinogenic polyaromatic hydrocarbon, aromatic amines, and benzene, were also reported to be its substrates (32, 33, 34).

The study described here immunohistochemically demonstrated that mEH was expressed on the steroid-hormone producing cells in the human ovary. These findings imply that mEH may be involved in steroidogenesis. mEH was reportedly expressed on cortical cells of human adrenal gland (35). Papadopoulos et al. examined the effect of mEH on steroid hormone synthesis using microsomal fractions isolated from adrenal gland with ¹⁴C-labeled progesterone. They found no differences in the steroid hormone production by the treatment of inhibitors, concluding that the physiological role of mEH in the adrenal gland remained unclear (8). We could not examine the physiological role of mEH in theca interna cells or small luteal cells in this experiment. A large amount of theca interna cells or small luteal cells from human ovary sufficient for the in vitro culture experiment cannot be obtained, and the theca interna cells from porcine ovary cultured in vitro for 3 days did not maintain the expression of mEH in the cells (not shown). The granulosa cells, on the other hand, are well known to have aromatase enzyme activity like the hepatocytes. The present study showed that mEH inhibitor, ETCP, reduced estradiol production from testosterone in cultured granulosa cells (Fig. 7A). However, in the preparation of granulosa cell homogenate, the inhibitor effectively suppressed mEH activity (Fig. 8B) but not aromatase activity (Fig. 8A). It shows that the inhibitor has no direct effect on the enzyme activity of aromatase, and that granulosa cells indirectly require mEH activity for the production of estradiol.

The concentration of ETCP required for the effective inhibition of estradiol production by the cultured granulosa cells (Fig. 7A) was very much higher than that for the inhibition of EH activity in the granulosa homogenate (Fig. 8B). We then examined the change of inhibitory effect of ETCP on EH activity during the granulosa cell culture period (Fig. 9). ETCP, which was added to the cell culture simultaneously with the enzyme substrate, GOC, effectively inhibited the EH activity in the low concentrations. ETCP in the cell homogenate disrupted by sonication also inhibited EH activity in the same dose-dependent manner as in the living cells, which shows that GOC and ETCP were cell membrane-permeable enough for the detection of enzyme activities and inhibitory effects in the condition described here. ETCP, added 5 or 24 h before the assay of enzyme activities, on the other hand, inhibited the EH activity only in the high concentrations, which was similar to the inhibitory effect shown in the estradiol production during 24 h of incubation (Fig. 7A). In conclusion, ETCP was shown to inhibit the EH activity and the estradiol production of the granulosa cells in a similar dose-dependent manner.

Anticonvulsant sodium valproate is reported to induce endocrine disorders such as polycystic ovary syndrome. In the drug-induced polycystic ovary syndrome, an elevated serum androgen level, irregular menstruation, and polycystic morphological change of the ovary are observed. Sodium valproate is known to interfere with the metabolism of -aminobutyric acid (GABA) and is expected to increase the GABA concentration in the central nervous system. It was therefore suspected that sodium valproate activates LH secretion through GABA systems and that the impaired LH secretion induces the hypersecretion of androgen by theca interna cells concomitant with polycystic change of the ovary. However, subsequent clinical studies did not support this explanation. Elevation of the serum LH level was not observed in the valproate-induced endocrine disorders (12, 13). The administration of sodium valproate, in normal women, was shown to induce no alteration of the basal secretion and the pulsatility of LH in both follicular and luteal phases (36, 37). In patients with polycystic ovary syndrome, sodium valproate administration also produced no significant change in the frequency and amplitude of LH and FSH secretion (38). Furthermore, it is reported that acute administration of sodium valproate suppresses the LH secretion (39). In ovariectomized women, valproate did not affect the pulsatile secretion of LH (40). Then, the effect of sodium valproate on the induction of polycystic ovary syndrome would not be caused by the alteration in the central nervous system through the GABA metabolism. Some investigators suspected that sodium valproate may inhibit conversion of testosterone to estradiol by an unknown mechanism (38). Sodium valproate and its analogues were reported to inhibit EH enzyme activity (10, 41). The present study showed that granulosa cells in the follicles of more than 4 mm in diameter express mEH and that the inhibition of mEH activity suppresses the conversion of testosterone to estradiol by granulosa cells cultured in vitro. If sodium valproate inhibits the mEH activity and then suppresses the aromatase activity in granulosa cells, testosterone produced by theca interna cells is not converted to estradiol in the follicles. This leads to the androgen-dominant microenvironment in the ovary and then to polycystic change of the ovary. This interpretation may explain the overall status of valproate-induced polycystic ovary syndrome, where an elevated serum testosterone concentration without an increase of peripheral LH and estradiol levels are manifested. It is also possible that mEH in theca interna cells regulates the steroidogenesis via the epoxide degradation, and that it leads indirectly to the suppression of the estradiol production by the granulosa cells.

In the small follicles, the protection of oocytes from adverse agents is important to preservation of healthy follicles. Although the basal lamina surrounding granulosa cells may serve as a barrier for macromolecules, it does not block low-molecular-weight reagents. In the initial developmental follicles, we observed the expression of LH receptor (42) and low-density lipoprotein receptor (43) on the theca interna cells. In addition, apolipoprotein-B was detected in the cytoplasm of theca interna cells in the small follicles less than 1 mm in diameter, suggesting the continuous uptake of low-density lipoprotein and steroid hormone synthesis by these cells (44). mEH in the liver is reported to inactivate the carcinogenic or cytotoxic electrophilic epoxides produced by the cytochrome P-450-dependent monooxygenase system (7). Interestingly, mEH was clearly detected on the theca interna cells in a small follicle with several granulosa cell layers. It is suggested that toxic substances would be inactivated by mEH, to maintain homeostasis of the microenvironment in the immature follicles, which protects oocytes from damage by some epoxides. It is also possible that a large amount of mEH found in theca interna cells in large follicles and in small luteal cells in CL inactivates the cytotoxic epoxides that were produced during the metabolic pathway of steroidogenesis.

	Acknowledgments

 
The authors acknowledge Drs. T. Aimoto and N. Inoue for donating GOC and DHC and for helpful discussion. The authors thank Dr. S. Shibamoto for help in carrying out our experiments and are grateful to Mrs. Hisako Takahashi, Mrs. Nami Kohama, and Miss Chikako Kataoka for technical assistance in RIA. The authors are also grateful to Mr. Daniel Mrozek for reading the manuscript.

	Footnotes

 
Address all correspondence and requests for reprints to: Masamichi Ueda, Institute for Virus Research, Kyoto University, Sakyo-ku, Kyoto, 606-8507, Japan.

* This work was supported, in part, by grants-in-aid for scientific research (09671673, 09671674, and 09671676).

Received December 13, 1999.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Fujiwara H, Ueda M, Imai K, Fukuoka M, Yasuda K, Takakura K, Suginami H, Kanzaki H, Inoko H, Mori T, Maeda M 1993 Human leukocyte antigen-DR is a differentiation antigen for human granulosa cells. Biol Reprod 49:705–715[Abstract]
2. Fujiwara H, Maeda M, Ueda M, Fukuoka M, Yasuda K, Imai K, Takakura K, Mori T 1993 A differentiation-related molecule on the cell surface of human granulosa cells. J Clin Endocrinol Metab 76:956–961[Abstract]
3. Honda T, Fujiwara H, Ueda M, Maeda M, Mori T 1995 Integrin 6 is a differentiation antigen of human granulosa cells. J Clin Endocrinol Metab 80:2899–2905[Abstract/Free Full Text]
4. Fujiwara H, Kataoka N, Honda T, Ueda M, Yamada S, Nakamura K, Suginami H, Mori T, Maeda M 1998 Physiological roles of integrin 6ß1 in ovarian functions. Horm Res [Suppl 2] 50:25–29
5. Hattori N, Ueda M, Fujiwara H, Fukuoka M, Maeda M, Mori T 1995 Human luteal cells express leukocyte functional antigen (LFA)-3. J Clin Endocrinol Metab 80:78–84[Abstract]
6. Hashii K, Fujiwara H, Yoshioka S, Kataoka N, Yamada S, Hirano T, Mori T, Fujii S, Maeda M 1998 Peripheral blood mononuclear cells stimulate progesterone production by luteal cells derived from pregnant and non-pregnant women: possible involvement of interleukin-4 and 10 in corpus luteum function and differentiation. Hum Reprod 13:2738–2744[Abstract/Free Full Text]
7. Seidegard J, DePierre JW 1983 Microsomal EH, properties, regulation and function. Biochim Biophys Acta 695:251–270[Medline]
8. Papadopoulos D, Jörnvall H, Rydström J, DePierre JW 1994 Purification and initial characterization of microsomal EH from the human adrenal gland. Biochim Biophys Acta 1206:253–262[CrossRef][Medline]
9. Oesch F, Glatt H, Schmassmann H 1977 The apparent ubiquity of epoxide hydrolase in rat organs. Biochem Pharmacol 26:603–607[CrossRef][Medline]
10. Kerr BM, Rettie AE, Eddy AC, Loiseau P, Guyot M, Wilensky AJ, Levy RH 1989 Inhibition of human liver microsomal epoxide hydrolase by valproate and valpromide: in vitro/in vivo correlation. Clin Pharmacol Ther 46:82–93[Medline]
11. Robbins DK, Wedlund PJ, Elsbero S, Oesch F, Thomas H 1992 Interaction of valproic acid and some analogues with microsomal epoxide hydrolase. Biochem Pharmacol 43:775–783[CrossRef][Medline]
12. Isojärvi JI, Laatikainen TJ, Pakarinen AJ, Juntunen KT, Myllyla VV 1993 Polycystic ovaries and hyperandrogenism in women taking valproate for epilepsy. N Engl J Med 329:1383–1388[Abstract/Free Full Text]
13. Sharma S, Jacobs HS 1997 Polycystic ovary syndrome associated with treatment with the anticonvulsant sodium valproate. Curr Opin Obstet Gynecol 9:391–392[Medline]
14. Fujiwara H, Ueda M, Hattori N, Mori T, Maeda M 1996 A differentiation antigen of human large luteal cells in corpora lutea of the menstrual cycle and early pregnancy. Biol Reprod 54:1173–1183[Abstract]
15. Köhler G, Milstein C 1975 Continuous cultures of fused cells secreting antibody of predetermined specificity. Nature 256:495–497[CrossRef][Medline]
16. Ryan RJ 1981 Follicular atresia: some speculations of biochemical markers and mechanisms. In: Schwartz NB, Hunzicker-Dunn M (eds) Dynamics of Ovarian Function. Raven Press, New York, pp 1–11
17. Corner GW 1956 The histological dating of the human corpus luteum of menstruation. Am J Anat 98:377–401[CrossRef][Medline]
18. Tsujimura K, Park Y, Miyama-Inaba M, Meguro T, Ohno T, Ueda M, Masuda T 1990 Comparative studies on FcR (FcRII, FcRIII and FcR) function of murine B cells. J Immunol 144:4571–4578[Abstract]
19. Inoue N, Yamada K, Imai K, Aimoto T 1993 Sex hormone-related control of hepatic EH activities in mice. Biol Pharm Bull 16:1004–1007[Medline]
20. Watabe T, Hiratsuka A, Ishikawa K, Isobe M, Ozawa N 1984 7-Glycidoxycoumarin (GOC): a fluorophotometric epoxide substrate for the assay of glutathione S-transferase activity. Biochem Pharmacol 33:1839–1844[CrossRef][Medline]
21. Emi N, Kanzaki H, Yoshida M, Takakura K, Kariya M, Okamoto N, Imai K, Mori T 1991 Lymphocytes stimulate progesterone production by cultured human granulosa luteal cells. Am J Obstet Gynecol 165:1469–1474[Medline]
22. Hutchison JB, Steimer TH, Duncan R 1981 Behavioral action of androgen in the dove: effects of long-term castration on response specificity and brain aromatization. J Endocrinol 90:167–178[Abstract]
23. Gibb W, Lavoie JC 1980 Substrate specificity of the placental microsomal aromatase. Steroids 36:507–519[CrossRef][Medline]
24. Skoda RC, Demierre A, McBride OW, Gonzalez FJ, Meyer UA 1988 Human microsomal xenobiotic epoxide hydrolase. Complementary DNA sequence, complementary DNA-directed expression in COS-1 cells, and chromosomal localization. J Biol Chem 263:1549–1554[Abstract/Free Full Text]
25. Falany CN, McQuiddy P, Kasper CB 1987 Structure and organization of the microsomal xenobiotic epoxide hydrolase gene. J Biol Chem 262:5924–5930[Abstract/Free Full Text]
26. Oesch F, Daly JW 1971 Solubilization, purification, and properties of a hepatic EH. Biochim Biophys Acta 227:692–697[Medline]
27. Oesch F 1974 Purification and specificity of a human microsomal EH. Biochem J 139:77–88[Medline]
28. Lu AY, Ryan D, Jerina DM, Daly JW, Levin W 1975 Liver microsomal epoxide hydrolase. Solubilization, purification, and characterization. J Biol Chem 250:8283–8288[Abstract/Free Full Text]
29. Halpert J, Glaumann H, Ingelmann-Sundberg M 1979 Isolation and incorporation of rabbit liver EH into phospholipid vesicles. J Biol Chem 254:7434–7441[Abstract/Free Full Text]
30. Buehler BA, Delimont D, van Waes M, Finnell RH 1990 Prenatal prediction of risk of the fetal hydantoin syndrome. N Engl J Med 322:1567–1572[Abstract]
31. Kaneko S 1991 Antiepileptic drug therapy and reproductive consequences: functional and morphological effects. Reprod Toxicol 5:179–198[CrossRef][Medline]
32. Gonzalez FJ, Samore M, McQuiddy P, Kasper CB 1982 Effects of 2-acetylaminofluorene and N-hydroxy-2-acetylaminofluorene on the cellular levels of epoxide hydrolase, cytochrome P-450b, and NADPH-cytochrome c (P-450) oxidoreductase messenger ribonucleic acids. J Biol Chem 257:11032–11036[Abstract/Free Full Text]
33. Jerina DM 1983 Metabolism of aromatic hydrocarbons by the cytochrome P-450 system and epoxide hydrolase. Drug Metab Dispos Biol Fate Chem 11:1–4[Medline]
34. Wood AW, Chang RL, Levin W, Yagi H, Thakker DR, van Bladeren PJ, Jerina DM, Conney AH 1983 Mutagenicity of the enantiomers of the diastereomeric bay-region benz(a)anthracene 3,4-diol-1,2-epoxides in bacterial and mammalian cells. Cancer Res 43:5821–5825[Medline]
35. Papadopoulos D, Gröndal S, Rydström J, DePierre JW 1992 Levels of cytochrome P-450, steroidogenesis and microsomal and cytosolic epoxide hydro- lases in normal human adrenal tissue and corresponding tumors. Cancer Biochem Biophys 12:283–291[Medline]
36. Lado-Abeal J, Cabezas-Agricola J, Paz-Carreira JM, Cabezas-Cerrato J 1991 Influence of sodium valproate on late follicular phase pulsatile LH secretion in normal women. Clin Endocrinology (Oxf) 35:477–483[Medline]
37. Lado-Abeal J, Rey-Losada C, Cabezas-Agricola JM, Cabezas-Cerrato J 1994 Influence of sodium valproate on medium-late luteal phase pulsatile LH secretion in normal women. Clin Endocrinol (Oxf) 40:127–131[Medline]
38. Popovic V, Spremovic S 1995 The effect of sodium valproate on luteinizing hormone secretion in women with polycystic ovary disease. J Endocrinol Invest 18:104–108[Medline]
39. Popovic V, Spremovic-Radjenovic S, Eric-Marinkovic J, Grossman A 1996 Effect of sodium valproate on luteinizing hormone secretion in pre- and postmenopausal women and its modulation by naloxone infusion. J Clin Endocrinol Metab 81:2520–2524[Abstract]
40. Lado-Abeal J, Liz JL, Rey C, Febrero M, Cabezas-Cerrato J 1996 Effects of valproate-induced alteration of the GABAergic system on pulsatile luteinizing hormone secretion in ovariectomized women. Eur J Endocrinol 135:293–298[Abstract]
41. Bernus I, Dickinson RG, Hooper WD, Eadie MJ 1997 The mechanism of the carbamazepine-valproate interaction in humans. Br J Clin Pharmacol 44:21–27[CrossRef][Medline]
42. Takao Y, Honda T, Ueda M, Hattori N, Yamada S, Maeda M, Fujiwara H, Mori T, Wimalasena J 1997 Immunohistochemical localization of LH/hCG receptor in human ovary: hCG enhances cell surface expression of LH/hCG receptor on luteinizing granulosa cells in vitro. Mol Hum Reprod 3:569–578[Abstract/Free Full Text]
43. Yamada S, Fujiwara H, Kataoka N, Honda T, Nakayama T, Higuchi T, Mori T, Maeda M 1998 Stage-specific uptake of apolipoprotein-B in ovarian follicles and corpora lutea of the menstrual cycle and early pregnancy. Hum Reprod 13:944–952[Abstract/Free Full Text]
44. Yamada S, Fujiwara H, Ueda M, Honda T, Nakayama T, Higuchi T, Mori T, Maeda M 1998 A monoclonal antibody, HCL-2, raised against human luteal cells reacts with apolipoprotein-B and detects the uptake of LDL by luteinizing granulosa cells. Hum Reprod 13:936–943[Abstract/Free Full Text]

This article has been cited by other articles:

				K. S. Rajapaksa, I. G. Sipes, and P. B. Hoyer Involvement of Microsomal Epoxide Hydrolase Enzyme in Ovotoxicity Caused by 7,12-Dimethylbenz[a]anthracene Toxicol. Sci., April 1, 2007; 96(2): 327 - 334. [Abstract] [Full Text] [PDF]

				G. J. Tranah, E. Giovannucci, J. Ma, C. Fuchs, S. E. Hankinson, and D. J. Hunter Epoxide hydrolase polymorphisms, cigarette smoking and risk of colorectal adenoma in the Nurses' Health Study and the Health Professionals Follow-up Study Carcinogenesis, July 1, 2004; 25(7): 1211 - 1218. [Abstract] [Full Text] [PDF]

				E. A. Cannady, C. A. Dyer, P. J. Christian, I. G. Sipes, and P. B. Hoyer Expression and Activity of Microsomal Epoxide Hydrolase in Follicles Isolated from Mouse Ovaries Toxicol. Sci., July 1, 2002; 68(1): 24 - 31. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hattori, N. Articles by Ueda, M. Search for Related Content PubMed PubMed Citation Articles by Hattori, N. Articles by Ueda, M. Pubmed/NCBI databases Gene GEO Profiles HomoloGene UniGene