Ketoconazole inhibits multiple steroidogenic enzymes involved in androgen biosynthesis in the human ovary

Ketoconazole inhibits multiple steroidogenic enzymes involved in androgen biosynthesis in the human ovary

FERTILITY AND STERILITY Copyright © 1988 The American Fertility Society Vol. 49, No.1, January 1988 Printed in U. S.A. Ketoconazole inhibits multipl...

505KB Sizes 0 Downloads 52 Views

FERTILITY AND STERILITY Copyright © 1988 The American Fertility Society

Vol. 49, No.1, January 1988 Printed in U. S.A.

Ketoconazole inhibits multiple steroidogenic enzymes involved in androgen biosynthesis in the human ovary

Michael DiMattina, M.D.*t Nicole Maronian, B.S.+ Hansel Ashby, M.S.*

D. Lynn Loriaux, M.D., Ph.D.+ Barry D. Albertson, Ph.D.*+

Georgetown University Hospital, Washington, D.C., and National Institutes of Health, Bethesda, Maryland

Ketoconazole (KZ) has been shown to inhibit testicular and adrenal steroidogenesis and is useful in the medical management of gonadotropin-independent precocious puberty, prostatic cancer, and Cushing's syndrome. To determine whether KZ similarly affects ovarian steroidogenesis, the authors examined its effect on the activity of the human ovarian 3{J-hydroxysteroid dehydrogenaselisomerase (3{J-HSD), 17-hydroxylase (17-0H), and aromatase (AR) in vitro. A dose-dependent decrease in the activities of 3{J-HSD and 17-0H was observed with increasing amounts of KZ. With 10, 50, and 100-fold excess KZ, the activity of 3{J-HSD decreased by 59% (P < 0.001), 73% (P < 0.005), and 85% (P < 0.005), respectively. At equimolar concentrations with substrate (50 JLM), KZ inhibited 17-0H by 70% (P < 0.01). No significant effect on ovarian AR activity was observed, except at the highest concentration of KZ tested. The authors conclude that low concentrations of KZ profoundly inhibit the activity of human ovarian 3{J-HSD and 17-0H in vitro. These observations suggest that KZ might be useful in the medical management of women with hyperandrogenism, but further experimentation and clinical trials will be necessary. Fertil Steril 49:62, 1988

Ketoconazole (KZ) is a synthetic imidazole dioxolone derivative used clinically as an oral antifungal agent. In humans, it has been shown to inhibit adrenap,2 and testicular 3 steroidogenesis. Clinically, a single oral dose of KZ causes a transient fall in serum androstenedione (~4A) and testosterone (T), with a rise in serum 17 -hydroxyprogesterone (17-0HP).4 It inhibits testicular T production by directly inhibiting the activity of the enzymes 17hydroxylase (17 -OH) and 17,20-desmolase (17,20-

Received March 9, 1987; revised and accepted August 19, 1987. • Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology, Georgetown University Hospital. t Reprint requests: Michael DiMattina, M.D., Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology, Georgetown University Hospital, 3800 Reservoir Road, N.W., Washington, D.C. 20007. :j: Developmental Endocrinology Branch, National Institutes of Child Health and Development, National Institutes of Health.

62

DS).5 Testicular steroidogenesis, however, is more sensitive than the adrenal to inhibition by KZ since low levels of KZ profoundly inhibit testicular C-19 steroid production at a time when adrenal C-21 steroids are norma1. 1 ,3 Thus, it is possible for KZ to selectively inhibit gonadal androgen production in men. In women, excess ovarian androgen production commonly results in ovulatory dysfunction and varying degrees of virilization. Therefore, we investigated the possibility that KZ also may affect ovarian steroidogenesis by examining its direct effect on the activity of the human ovarian enzymes 3/3-hydroxysteroid dehydrogenase/isomerase (3/3HSD), 17-0H, and aromatase (AR). MATERIALS AND METHODS Preparation of Ovarian Tissue

Ovarian tissue was obtained from a 50-year-old premenopausal woman (day 6 of the menstrual

DiMattina et al. Ketoconazole inhibits human ovarian enzymes

Fertility and Sterility

cycle) undergoing hysterectomy-bilateral salpingoophorectomy for uterine fibroids and from a 33year-old woman (day 3 of the menstrual cycle) undergoing oophorectomy for removal of a benign cyst. The clinical stage of the cycle was confirmed by histologic examination of the endometrium. The ovaries, each approximately 15 gm wet weight, were immediately placed in ice-cold phosphate buffered saline and stored at -70°C until processing. The ovaries were quickly thawed, cut, rinsed with approximately 20 ml of 0.25 M sucrose, and homogenized at 4° C using a motor-driven homogenizer and a ground glass tissue grinder. Homogenates were centrifuged for 10 minutes at 8, 700, and 16,000 X g. A portion of the resulting 16,000 g supernatant from the ovaries was used to determine 3{3-HSD and 17-0H activity. The remainder of the 16,000 X g supernatant was centrifuged at 105,000 X g for 60 minutes. The resulting microsomal pellets were rehomogenized, suspended in a Tris-KCI-EDTA buffer (pH 7.4), and stored at -70°C until assayed for AR activity. The final protein concentration of both the 16,000 X g supernatant and the microsomal pellet suspension was determined by the method of Lowry et al. 6 and ranged from 9.3 to 16.22 mg/ml.

5-p,1 propylene glycol. The ethanol was evaporated under reduced pressure and the buffer containing appropriate co-factors added (Table 1). Substrate and co-factor concentrations were chosen to ensure a maximal rate of product formation in this in vitro system. The assay volume was 0.1 ml. Reaction mixtures were incubated in duplicate aliquot samples over two time points (6 and 12 minutes for 3{3-HSD; 10 and 20 minutes for 17-0H; and 15 and 30 minutes for AR) at 37°C in a Dubnoff metabolic incubator (CMS, Beltsville, MD). Reactions were terminated by adding 0.4 ml methanol containing 50 mg Ll4A, 17-0HP, and estrone (E 1) for the 3{3HSD, 17-0H, and AR assays, respectively. The product of the reactions, Ll4A, 17-0HP, and E 1, were isolated by thin-layer chromatography (TLC) on Whatman LKDF7 plates (Whitman Co. Clifton, NJ) and verified by repeat TLC after reduction and acetylation. Chromatography systems were chloroform:acetone (8:1, v/v) 3{3-HSD, chloroform:ethanol (98:2, v/v) 17-0H, and benzene:ethylacetate:heptane (50:30:20, v/v) for AR. Steroid recovery varied from 75% to 88%. The intra-assay and interassay variation was 15% for all enzyme assays. Statistics

Measurement of Enzyme Activities

Ovarian 3{3-HSD, 17 -OH, and AR activities were measured as previously described. 7- 9 The contents of the reaction mixture are listed in Table 1. Fifty p,M [4-1 4C]-dehydroepiandrosterone (51.3 mCi/ mmol), [14C]-progesterone (57.2 mCi/mmol), and [14C]Ll4A (52.0 mCi/mmol) (New England Nuclear, Boston, MA) were used as labeled substrate hormones for determination of 3{3-HSD, 17 -OH, and AR activity, respectively. All unlabeled steroids were purchased from Sigma Chemicals. 14C-Iabeled substrates and KZ, dissolved in ethanol, were added to 12 X 75-mm glass assay tubes containing

Enzyme activity in the presence or absence of KZ was compared using the Student's t-test. Comparisons were made between enzyme activities at different concentrations of KZ versus control (0 p,M KZ). RESULTS

A dose-dependent decrease in the activity of 3{3HSD was observed with increasing amounts of KZ (Fig. 1). Mean ovarian 3{3-HSD activity fell by 15% with 5 p,M KZ (not significant [NS]). At equimolar concentrations with substrate (50 p,M), KZ signifi-

Table 1 Enzyme Reaction Mixture Contents 3~-Hydroxysteroid

dehydrogenase/isomerase

Propylene glycol, 5 "I [14C)Dehydroepiandrosterone, 50 "M NAD,2mM Pyruvate, 5 mM Lactate dehydrogenase, 1 U /ml Sodium phosphate buffer (pH 7.4), 50 "M 16,000 gm supernatant containing 480 "g of protein

Vol. 49, No.1, January 1988

17-Hydroxylase

Arom~ase

Propylene glycol, 5 "I [14C)Progesterone, 50 "M NADPH, 500 "M Glucose-6-phosphate, 10 mM Glucose-6-phosphate dehydrogenase, 1.5 Vlml MgC12 , 5 mM Sodium phosphate buffer (pH 7.4), 50 "M 16,000 gm supernatant containing 650 "g of protein

Propylene glycol, 5 "I [14C)Testosterone, 5 "M NADPH, 600 "M Glucose-6-phosphate, 10 mM Glucose-6-phosphate dehydrogenase, 1.5 U /ml Sodium phosphate buffer (pH 7.4),50 "M Microsomes containing 400 "g of protein

DiMattina et aI. Ketoconazole inhibits human ovarian enzymes

63

9000

60

8000 7000 3BHSD Activity

6000

(pmol

5000

(tmol product!

4000

mlnutel

3000

~g Protein)

50 Aromatale 40 Activity

·p40.05

product!

30

minute' I'll Protem)

20 10

2000 1000

Control

0.5

5

50

500

5000

Concentration of Ketoconazole «.,.M)

Control

50

500

2500

5000

Concentration of Ketoconazole (pM)

Figure 1 Ovarian microsomal3{:1-HSD enzyme activity is expressed as femtomoles (fmol) of !J.4A produced per minute of incubation per mg of ovarian microsomal protein. Values are expressed as means ± SEM of four points, i.e., duplicate measurements at two separate time points. Substrate concentration (14C DHA) was 50 ~M in all of the groups shown. P levels are shown for various KZ concentrations versus control.

cantly inhibited 3{3-HSD by 19% (P < 0.05). Inhibition of enzyme activity by 59% (P < 0.001),73% (P < 0.005), and 85% (P < 0.005) was observed with 10,50, and 100-fold excess KZ, respectively. Mean l7-0H activity decreased by 31% with 5 JLM KZ compared with control (NS, Fig. 2). Inhibition of enzyme activity by 70% was observed at equimolar concentration of KZ and substrate (P < 0.01). Significant inhibition of enzyme activity continued with 5, 10, and 100-fold excess KZ. KZ had no significant effect on ovarian AR activity, except at the highest concentration.tested. The addition of 5000 JLM KZ to the AR assay produced a 34% decrease in enzyme activity (P < 0.005, Fig. 3). DISCUSSION

Our findings show that low concentrations of KZ directly inhibit the activity of the human ovarian enzymes 3{3-HSD and l7-0H in vitro. These two enzymes are essential for the formation of C-19 1800 1600 1400 170H Activity ('mol

product!

minute' 119 Protein)

1200 1000 800 600 400 200 Control

50

250

500

5000

Concentration 01 Keloconazole (pM)

Figure 2 Ovarian 17-0H activity is expressed as femtomoles (fmol) of 14C-17 -OHP formed. Levels shown are means ± SEM of duplicate measurements at two separate time points. Substrate concentration (14C P) was 50 ~M. P levels are shown for various KZ concentrations versus control.

64

Figure 3 Ovarian AR activity is expressed as picomoles (pmol) of 14C-El formed. Values are means ± SEM of duplica~e measurements at two separate time points. Significance IS shown at only the 5000-~M concentration of KZ versus control.

steroids such as dehydroepiandrosterone (DHA), 114A, and T. Thus, KZ could prevent the formation of these androgens by inhibiting enzymes in both the 114 and 115 biosynthetic pathways. To our knowledge, this is the first report of a direct inhibitory effect by KZ on an enzyme in the 115 pathway (i.e., 3{3-HSD), while our findings support previous studies that show inhibition of enzymes in the 114 pathway (i.e., l7_0H 2,5,lO,l1 and the l7,20-DS enzyme).4 Furthermore, KZ has been shown to inhibit cholesterol side chain cleavage enzyme in both the adrenal and testis. 12 Thus, KZ effectively inhibits ovarian androgen steroidogenesis by inhibiting at least three enzymes that are essential for androgen biosynthesis. KZ markedly depresses serum T levels in men, but serum estradiol (E 2 ) levels remain relatively unchanged. 13 This results in an increase in the E2/T ratio and perhaps contributes to the development of gynecomastia. 14 The explanation for the increased E2/T ratio and the development of gynecomastia is unclear. The present study, however, shows that KZ inhibits human ovarian AR activity at only the highest concentration of drug, while low concentrations of KZ profoundly inhibits ovarian 3{3-HSD and l7-0H activity. Taken together, these findings suggest that the increased E2/T ratio may be due to a relative insensitivity of AR to KZ, combined with a high sensitivity of the 3{3-HSD and l7-0H enzymes to the drug. In-depth kinetic studies would be helpful in clarifying the effect of KZ on the observed changes in the E2/T ratio seen in men. Syndromes of excess ovarian androgen production, such as idiopathic hirsutism and polycystic ovarian disease, are common. 15 Since KZ effectively inhibits the ovarian enzymes that are responsible for androgen biosynthesis, it may have a therapeutic value in the medical management of such conditions in women. In our study, low concentra-

DiMattina et al. Ketoconazole inhibits human ovarian enzymes

Fertility and Sterility

tions of KZ profoundly inhibited the activity of the ovarian 3fj-HSD and 17-0H. Thus, it is possible that low concentrations of KZ could selectively inhibit ovarian androgen steroidogenesis by affecting 3fj-HSD and 17-0H without altering adrenal cortisol biosynthesis, as has been reported in men. l •3 The concept of using low-dose KZ in the treatment of various clinical disorders is not new. The continuous use of low-dose KZ has been reported to be effective in preventing recurrent vulvovaginal candidiasis l6 and in the management of androgen excess associated with gonadotropin-independent precocious puberty in boysP Thus, it is possible that low-dose KZ would also be valuable in treating conditions in women that are characterized by excess androgen production. We conclude that low concentrations of KZ profoundly inhibit the activity of the human ovarian 3fj-HSD and 17-0H enzymes in vitro. Further experimentation and clinical trials to assess the therapeutic value of KZ in women with hyperandrogenism are warranted. Acknowledgments. We thank Vonnetta Gordon and Shirley Shirron for their expertise in preparation of the manuscript.

REFERENCES 1. Pont A, Williams PL, Loose DS, Feldman DS, Reitz RE, Bochra C, Stephens DA: Ketoconazole inhibits adrenal steroid synthesis. Ann Intern Med 97:370, 1982 2. Loose DS, Kan P, Hirst MA, Marcus RA, Feldman D: Ketoconazole blocks adrenal steroidogenesis by inhibiting cytochrome P-450-dependent enzymes. J Clin Invest 71:1495, 1983 3. Pont A, Williams PL, Anthar S, Reitz RE, Bochra C, Smith ER, Stephens DA: Ketoconazole blocks testosterone synthesis. Arch Intern Med 142:2137, 1982 4. DeCoster R, Caers I, Halterman C, Debroye M: Effect of a single administration of ketoconazole on total and physio-

Vol. 49, No.1, January 1988

logically free plasma testosterone and 17B-oestradiollevels in healthy male volunteers. Eur J Clin Pharmacol 29:489, 1985 5. Rajfer J, Sikka SC, Rivera F, Handelsman DJ: Mechanism of inhibition of human testicular steroidogenesis by oral ketoconazole. J Clin Endocrinol Metab 63:1193, 1986 6. Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ: Protein measurements with the Folin phenol reagent. J BioI Chem 193:265, 1951 7. DiMattina M, Albertson BD, Seyler DE, Loriaux DL, Falk RJ: Effect of the antiprogestin RU486 on progesterone production by cultured human granulosa cells: inhibition of the ovarian 3-hydroxysteroid dehydrogenase. Contraception 34:199, 1986 8. Munabi AK, Mericaq V, Koppelman MCS, Gelato MC, Macher AM, Albertson BD, Loriaux DL, Cassorla FG: The effects of prolactin on rat ovarian function. Steroids 43:631, 1986 9. Schiebinger RJ, Albertson BD, Cassorla FG, Bowyer DW, Geelhoed GW, Cutler GB Jr, Loriaux DL: The developmental changes in plasma adrenal androgens during infancy and adrenarche are associated with changing activities of adrenal microsomal 17-hydroxylase and 17,20-desmolase. J Clin Invest 67:1177,1980 10. Sikka SC, Swerdloff RS, Rajfer J: In vitro inhibition of testosterone biosynthesis by ketoconazole. Endocrinology 116:1920, 1985 11. Bhasin S, Sikka S, Fielder T, Sod-Moriah H, Levine HB, Swerdloff RS, Rajfer J: Hormonal effects of ketoconazole in vivo in the male rate: mechanism of action. Endocrinology 118:1229, 1986 12. Feldman D: Ketoconazole and other imidazole derivatives as inhibitors of steroidogenesis. Endocr Rev 7:409, 1986 13. Pont A, Goldman ES, Sugar AM, Suten PK, Stevens DA: Ketoconazole-induced increase in estradiol-testosterone ratio. Arch Intern Med 145:1429, 1985 14. Carlson H: Gynecomastia. N Engl J Med 303:795, 1980 15. Kirschner MA, Zucker R, Jespersen D: Idiopathic hirsuitism: an ovarian abnormality. N Engl J Med 294:637,1976 16. Sobel JD: Recurrent vulvovaginal candidiasis: a prospective study of the efficacy of maintenance ketoconazole therapy. N Engl J Med 315:1455, 1986 17. Holland JF, Fishman L, Bailey JD, Fazekas ATA: Ketoconazole in the management of precocious puberty not responsive to LHRH: analogue therapy. N Engl J Med 312:1023, 1985

DiMattina et aI. Ketoconazole inhibits human ovarian enzymes

65