- Email: [email protected]
Effect of ketoconazole on placental aromatase, 3β-hydroxysteroid dehydrogenase-isomerase and 17β-hydroxysteroid dehydrogenase
J. steroid Biochem. Vol. 25, No. 6, pp. 981-984, 1986
OOZZ-4731/86 $3.00 + 0.00 Copyright 0 1986 Pergamon Journals Ltd
Printed in Great Britain. All rights reserved
EFFECT
OF KETOCONAZOLE
ON PLACENTAL
AROMATASE,
3/l-HYDROXYSTEROID DEHYDROGENASE-ISOMERASE 17/3-HYDROXYSTEROID DEHYDROGENASE
AND
M. AYUB and S. R. STITCH* Division of Steroid End~~noio~,
Department of Chemical Pathology, university of Leeds, Leeds LS2 9LN, England (Received 26 March 1986)
Summary-Ketoconazole, an orally-active, broad spectrum mycotic agent, was shown to inhibit in vitro human placental microsomal aromatase but was without effect on 3/I-hydroxysteroid dehydrogenaseisomerase (3/I-HSD-I) and 17/Ghydroxysteroid dehydrogenase (17/l-HSD) activities. The I& of placental aromatase for testosterone was 30 f 1.1 nmol/l (mean *SEM, n I 6). Inhibition (determined by Lineweaver-Burk plot) was non-competitive with respect to substrate with a & value of 3.0 f I .4 pmol/l (mean f SEM, n =i 6). Ketoconazole was without effect on the 3fi-HSD-f and 178-HSD activities when using [3H] pregnenoione and [%I oestradiol, respectively, as substrates. Since ketoconazole is known to inhibit cytochrome P-4%dependent enzyme reactions, the results of the present study support the contention that q&chrome P-450 is involved in the aromatisation process.
INTRODUCIlON
Ketoconazoie (Nizoral) is an orally-active, broad spectrum, mycotic agent, thought to act by blocking the synthesis of ergosterol [l J. In vitro studies [2] have shown that ketoconazole inhibits the cytochrome P-450.dependent 17o-hydroxylase and 17,20desmolase activities of the rat testes. However, the drug was found to have no effect on the 17@-hydroxysteroid dehydrogenase, which is a cytochrome P~50-inde~ndent enzyme. Spectrophotometric studies using rat liver microsomal cytochrome P-450 have demonstrated that ketoconazole yields a type II binding spectra, with little effect on NADPH-cytochrome C (P-450) reductase activity [3]. The present study was designed to test whether ketoconazole, known to inhibit cytochrome P-450dependent steroidogenic enzyme activity in both gonadal and adrenal tissues [4,5,6], inhibits the human placental aromatase system in Y&O, Evidence for the involvement of cytochrome P-450 in aromatisation is conflicting. Early studies suggested that aromatisation did not involve oxygen incorporation into the substrates by means of the cytochrome P-450 mono-oxygenase enzyme system because classical inhibitors of cytochrome P-450 mono-oxygenation, such as carbon monoxide [7J had essentially no effect on the aromatisation of androsten~ione [8]. Wowever, aminoglutethimide, a known ~tochrome P-450 inhibitor, inhibits aromatase [9, IO]. Demonstration of type I binding spectra by steroid substrates and inhibitors of placental microsomal aromatase and inhibition of aromatisation by rabbit antiserum *Professor S. R. Stitch died on 15th December 1985 whilst this paper was in preparation.
against rat liver NADPH-Cytochrome C reductase led Thompson and Siiteri to suggest the involvement of cytochrome P-450 [ll]. Recently, Osawa and Higashiyana demonstrated that aromatase activity results only when fractions containing cytochrome P-450 and NADPH-cytochrome reductase activity are combined [12,13]. In the present study, ketoconazole was found to inhibit aromatase in a dose-dependent manner but not the cytochrome P4%independent 178 -hydroxysteroid dehydrogenase and 3~-hydroxysteroid dehydrogen~e-isomerase activities of human placental microsomes. EXPERIMENTAL
Materials [4, 7-3H]Pregnenolone (sp. act. 10 Ci/mmol), [4L4Cjprogesterone (sp. act. 56 mCi/mmol), [2,4,6,7‘H]oestradiol (sp. act. 101 Ci/mmol), [4-Ci4] oestrone (sp. act. 55.8 mCi/mmol) and [1,2,6, 7-3H]testosterone (sp. act. 94 Ci/mmol) were purchased from Amersham International plc, Bucks and checked for purity by TLC (chloroform-acetone, 185:15, v/v). [L4C]Oestradiol was prepared by sodium borohydride reduction of [‘4C]oestrone and purified by TLC in ethyl acetate-iso-octane, 8:2, v/v. All non-radioactive steroids were purchased from Steraioids Ltd, Croydon. Phosphate and diethyl ether were obtained from B.D.H. Chemicals Ltd, Poole, Dorset. Sucrose, NAD, NADPH and fi-mercaptoethanol were obtained from Sigma Chemical Co. Ltd, Poole, Dorset. Acetone, chloroform, ethyl acetate, iso-octane and Triton-X scintillation fluid were obtained from Vickers Laboratories Ltd., Burley-in-Wharfedale and ethanol was purchased from James Burrow, London. TLC plates were kieselgei 60 and obtained from E. 981
982
M. Avue and S. R. STITCH
Merck, Darmstadt. Ketoconazole was supplied by Janssen Pharmaceutical Ltd, Grove, Wantage, Oxfordshire. Methoa!s Preparation of human placenta/ microsomes. A human, term placenta, obtained immediately after vaginal delivery follo~ng normal pregnancy, was transported to the laboratory on ice and processed immediately at 04°C. The placenta was dissected free of adhering membrane and blood vessels and minced with scissors. Homogenization was carried out in cold phosphate buffer (25mmol/l, pH 7.4) containing sucrose (0.25 mol/l) and p-mercaptoethanol (7 mmol/l) using 3 x 10 s bursts of a motordriven homogeniser, separated by 2 min cooling periods. The homogenate was centrifuged at 10,~~ for 30 min. The supematant obtained was centrifuged at 100,OOOg for 60 min. The microsomal pellet thus obtained was re-suspended in buffer (20 ml), treated briefly with a TetIon-glass homogeniser to ensure full dispersion and divided into portions (1 ml) for storage at -20°C. The protein concentration of the suspension was determined by the method of Lowry et al.f14]. Determination of enzyme activities
(i) Aromatase: [3H]Testosterone (50 nmol/l, 0.1 PCi) was incubated with placental microsomal suspension (63 pg protein) and NADPH (I 0 pmol/l) in the phosphate-sucrose-mercaptoethanol buffer (final volume, 1 ml). (ii) 17fl-HSD: 13H]~stradiol (5 @moljl, 0.1 PCi) was incubated with placental microsomal suspension (63pg protein) and NAD (0.1 mmol/l) in the phosphate-sucrose-mercaptoethanol buffer (final volume, 1 ml). (iii) 3/I-HSD-I: [3H]Pregnenolone (2 pmol/l, 0.15 PCi) was incubated with placental microsomal suspension (63 pg protein) and NAD (1 mmoljl) in the phosphat~su~ro~mer~pt~thanol buffer (final volume, 1 ml). Each incubation mixture was pre-incubated for 5 min at 37°C without microsomes. The reaction was started by addition of the microsomal suspension. Incubations were terminated after 10 min by addition of chloroform (1 ml) containing (i) for aromatase: [‘4C]oestradiol, [14C]oestrone (each 10,000 dpm) and non-radioactive testosterone, oestradiol and oestrone (37~mol/l each); (ii) For 17@-HSD: [14C]oestrone (10,OOOdpm) and non-radioactive oestrone and oestradiol (37 pmol/l each); (iii) For 3#?-HSD-I: [‘4C]progestrone (5,000 dpm) and non-radioactive pregnenolone and progesterone (3 18 pmol/l each). For each incubation, the steroids in the mixture were extracted with ether (3 x 4ml). The extracts were combined then evaporated and the residues dissolved in ethanol (25~1) and the appropriate steroids were isolated by TLC as follows: Aromatase and 17P-HSD: ethyl acetate-iso-octane (8:2, v/v)
separated androsten~ione (& = 0.60), testosterone (0.85) oestrone (0.77) and oestradiol (0.65). 3/I-HSD-I: chloroform-acetone (185: 15, v/v) separated pregnenolone (0.56) and progesterone (0.77). The location of the radioactive steroids after chromatography was assessed from the location of authentic steroids chromatographed on the same plates. The areas of silica gel containing the labelled steroids of interest were separately transferred to counting vials and the ‘H and 14Ccontent determined by liquid scintillation counting using T&on-X scintillation fluid (10 ml). Counting efficiency for [3H] was 54%, that for [14C] was 90%, and the cross-over of 14C counts into the ‘H channels was 22%. The activity of aromatase was calculated as the sum of rH]oestrone and [3H]~stradiol formed from [3H]testosterone and expressed in terms of unit time and protein. The yield of [3H]oestrone and [3H]oestradiol was corrected for manipulative loss using the measured recovery of [i4C]oestrone and [‘4C]oestradiol, respectively. 17fi-HSD activity was calculated as the amount of [‘Hloestrone formed from [3H]oestradiol. The yield of [3H]oestrone was corrected for manipulative loss using the measured recovery of [‘4C]oestrone. Similarly, 3b-HSD-I activity was calculated as the amount of [3H]progesterone formed from 13H]pregnenolone. The yield of [3H]progesterone was corrected for manipulative loss using the measured recovery of [‘4C]progesterone. The radiochemical purity of the product of each enzyme assay was checked after chromatography and final identification based on constant [‘H]/[r4C] ratios obtained through the following procedures: (i) rechromatographed in the solvent system ethyl acetate-iso-octane, 8:2, v/v in which the products had identical chromatographic mobilities to those of authentic standards, (ii) eluted and rechromatographed (ethyl acetate-iso-octane, 8:2, v/v) after treatment with a mixture of pyridine-acetic anhydride (2: 1, v/v; data not shown). RF3XJLTS E$ect of ketoconazole on aromatase
The amount of oestrogen formed was related linearly to the amount of microsomal protein (up to 19Opg protein) and time (data not shown). The K,,, of testosterone for aromatase was 30 rt 1.l nmol/I (mean t_SEM, n = 6) and the I’,,,,, was 2.0 It 1.9 nMf63 pg protein/min (mean ItSEM, n =6). Figure 1 shows the percentage inhibition of aromatisation as a function of increasing concentration of ketoconazole. Fifty percent enzyme inhibition was found in the presence of 2 flmol/l ketoconazole and complete inhibition occurred at 7.4 pmol/l. The type of inhibition (Fig. 2) by ketoconazole at a concentration of 2.46pmol/l was then determined. Inhibition was non~om~titive with respect to substrate with a K, value of 3.0 + 1.I @mol/I (mean ISEM, n = 6).
983
Inhibition of aromatase by ketoconazole
1lot y -:::::(z ,:“.: Y 0
I
I
I
2
4
6
(pmol/
I)
I 6
10
0
Ketaconazole
Ketoconozole
.:’
,
30
20
concentration
100
(pmol/ll
Fig. I. Effect of ketocanozole on the aromatisation of [3H]testosterone by human placental microsomes. The error bars represent the variation around the mean of duplicate incubations of two independent experiments.
Fig. 3. Effect of ketoconazole on human placental microsomal 3B-HSD-I. The error bars represent the variation around the mean of duplicate incubation of two independent experiments.
Efsects of ketoconazole on 3/?-HSD-I
corresponding decrease in amount of testosterone metabolised. This implies that ketoconazole inhibits 19-hydroxylation of testosterone, the first of three postulated hydroxylation steps in the aromatisation of androgens [ 151.The results of the present study are in Variance to those of Mason et a1.,[16] using androstenedione (K,,,, 220 nmol/l) as the substrate for aromatase, they found an I, value of 60pmol/l for ketoconazole whereas in the present study testosterone (K,, 30 nmol/l) was used as the substrate and an I, value of 2 pmol/l was found for ketoconazole. Furthermore of the imidazole derivatives studied by Mason et al., miconazole was found to be the most potent inhibitor with an I,, value of 0.6pmol/l and was a competitive inhibitor with a Ki value of 55 nmol/l. In contrast, in the present study Ketoconazole was found to be a non-competitive inhibitor of aromatase with a K, value of 3.0~mol/l.
and 17fl-HSD
Figure 3 shows that increasing concentrations of ketoconazole have no effect on [‘Hlpregnenolone conversion to [3H]progesterone by placental microsomal 3/?-HSD-I activity. Similarly, ketoconazolc was found to be without effect on the formation of [3H]oestrone from [3H]oestradiol by 17/I-HSD (Fig. 4). DISCUSSION
Ketoconazole inhibits several cytochrome P-450mediated steroid hydroxylations in both gonads and adrenals [4,5,7, 81. The ability of this compound to inhibit aromatisation by placental microsomes, as demonstrated in the present study, supports the contention that a cytochrome P-450 is involved in aromatisation. The reduction in oestrogen formation in the presence of ketoconazole correlated with a
Lack of inhibition of placental microsomal 3/3-HSD-I and 17fl-HSD by ketoconazole is in accordance with the results of Sikka et a1.[2], who also
0.25 r
‘-
.c_ E 2 a
0.15
. Ketoconarol.?
absent
0 Ketocanozole
present
-
20 t
---- Ketoconorole Ketoconazole
absenl
present
10 I 0.04
0.02
0.00
0.02
0.04
0.06
t
0.06
0
I
I
I
10
,
20
30
Ketoconazole
Fig. 2. Lineweaver-Burk plot of ketoconazole inhibition of human placental microsomal aromatase. Each point represents the mean of duplicate incubations of six independent experiments.
concentration
II
,
100
(pmol/l)
Fig. 4. Effect of ketoconazole on human placental microsomal 17b-HSD. The error bars represent the variation around the mean of duplicate incubations of two independent experiments.
984
M. Avue and S. R. STITCH
failed to observe inhibition of rat testicular 17/3-HSD. Further studies are required to determine if ketoconazole inhibits aromatase in oivo; oestrogen levels in blood have not been measured in patients receiving this drug. However, interpretation of oestrogen levels in blood of treated patients would be complicated by the fact that ketoconazole also inhibits synthesis of androgen which act as oestrogen precursors. Low oestrogen concentrations therefore may not be due solely to an effect on aromatase. Deflice et a1.[17] were the first to report gynaecomastia in patients receiving the drug. An increase in the free oestrogen-androgen ratio is thought to be a prerequisite for the development of gynaecomastia [18]. However, if ketoconazole inhibits both oestrogen and androgen biosynthesis in vitro, then development of gynaecomastia by ketoconazole could be through some other mechanism. Acknowledgement-The author is grateful for the advice of Dr R. E. Oakey in the preparation of this paper.
REFERENCES
Borgers M., Van den Bossche H. and de Brabander M.: The mechanism of action of the new antimycotic ketoconazole. Am. J. Med. (Suppl.) Proceedings of a symposium on New Developments and Therapy for the Mycoscs (1983) pp. 2-8. Sikka S. C., Swerdloff R. S. and Rajfer J.: In vitro inhibition of testosterone biosynthesis by ketoconazole. Endocrinology 116 (1985) 1920-1925. Sheets J. J. and Mason J. I.: Ketoconazole: A potent inhibitor of cytochrome P-450 dependent drug metabolism in rat liver. Drug Metab. Disp. 112 (1984) 603-606. Kowai J.: The effect of ketoconazole on steroidogenesis in cultured mouse cortex tumour cells. Endocrinology 112 (1983) 1541-1543. Loose D. S., Kan P. B., Hirst M. A., Marcus R. A. and Feldman D.: Ketoconazole blocks adrenal steroido-
genesis by inhibiting cytochrome P-450 dependent enzymes. J. clin. Invest. 71 (1983) 1495-1499. 6. Pant A., Williams P. L., Loose D. S., Feldman D., Peitz R. E., Bochra C. and Stevens D. A.: Ketoconazole blocks adrenal steroid biosynthesis. Ann. Int. Med. 97 (1982) 370-372.
I. Meighs R. A. and Ryan K. J.: Cytochrome P-450 and steroid biosynthesis in the human placenta. Biochim. bioohvs. Acta 165 (1968) 47&482.
8. Chakraborty J., Hopkins R. and Parke D. V.: Inhibition studies on the aromatisation of androstenedione by human placental microsomal preparations. Biochem. J. 130 (1977) 19P-20P. 9. Graves P. E. and Schanick H. A.: Stereoselective inhibition of aromatase by enantiomers of aminoglutethimide. Endocrinology 105 (1979) 52257. IO. Thompson E. A. and Siiteri P. K.: The involvement of human placental microsomal cytochrome P-450 in aromatisation. J. biol. Chem. 249 (1974) 5373-5378. 11. Osawa Y. and Higashiyama T.: Isolation of human placental cytochrome P-450 and its mechanism of action of androgen aromatisation. In Microsomes, Drug Oxidations and Chemical Carcinogenesis (Edited by M. J. Coon, A. H. Conney, R. W. Eastbrook, A. V. Gelboin, J. R. Gillete and P. J. O’Brien). Academic Press, New York, Vol. I (1980) pp. 225-228. 12. Zachariah P. K. and Juchan M. R.: Inhibition of human placental mixed function oxidations with carbon monoxide: Reversal with monochromatic light. J. sferoid Biochem. 8 (1978) 221-228. 13. Jachan M. R. and Zachariah P. K.: Displacement
of carbon monoxide from placental cytochrome P-450 by steroids: Antagonistic effects of androstenedione and 19-norandrostenedione. Biochem. biophys. Res. Commun. 65 (1975) 16261632.
14. Lowry 0. H., Rosebrough N. S., Farr A. L. and Randall R. S.: Protein measurement with the Folin phenol reagent. J. biol. Chem. 193 (1951) 265-275. 15. Fishman J.: Biochemical mechanism of aromatization. Cancer Res. 42 (1982) 3277s-3279s. 16. Mason J. I., Murry B. A., Olcott M. and Sheets J. J.: Imidazole antimycotics: inhibitors of steroid aromatase. Biochem. Pharmac. 34 (1985) 1087-1092.
17. Deflice R., Johnson D. G. and Galgiani J. N.: Gynaecomastin with ketoconazole. Antimicrob. Agents. Chemother. 19 (1981) 1073-1074. 18. Wilson J. D. and MacDonald P. C.: The pathogenesis of gynaecomastia. Adv. Intern. Med. 25 (1980) l-32.
OOZZ-4731/86 $3.00 + 0.00 Copyright 0 1986 Pergamon Journals Ltd
Printed in Great Britain. All rights reserved
EFFECT
OF KETOCONAZOLE
ON PLACENTAL
AROMATASE,
3/l-HYDROXYSTEROID DEHYDROGENASE-ISOMERASE 17/3-HYDROXYSTEROID DEHYDROGENASE
AND
M. AYUB and S. R. STITCH* Division of Steroid End~~noio~,
Department of Chemical Pathology, university of Leeds, Leeds LS2 9LN, England (Received 26 March 1986)
Summary-Ketoconazole, an orally-active, broad spectrum mycotic agent, was shown to inhibit in vitro human placental microsomal aromatase but was without effect on 3/I-hydroxysteroid dehydrogenaseisomerase (3/I-HSD-I) and 17/Ghydroxysteroid dehydrogenase (17/l-HSD) activities. The I& of placental aromatase for testosterone was 30 f 1.1 nmol/l (mean *SEM, n I 6). Inhibition (determined by Lineweaver-Burk plot) was non-competitive with respect to substrate with a & value of 3.0 f I .4 pmol/l (mean f SEM, n =i 6). Ketoconazole was without effect on the 3fi-HSD-f and 178-HSD activities when using [3H] pregnenoione and [%I oestradiol, respectively, as substrates. Since ketoconazole is known to inhibit cytochrome P-4%dependent enzyme reactions, the results of the present study support the contention that q&chrome P-450 is involved in the aromatisation process.
INTRODUCIlON
Ketoconazoie (Nizoral) is an orally-active, broad spectrum, mycotic agent, thought to act by blocking the synthesis of ergosterol [l J. In vitro studies [2] have shown that ketoconazole inhibits the cytochrome P-450.dependent 17o-hydroxylase and 17,20desmolase activities of the rat testes. However, the drug was found to have no effect on the 17@-hydroxysteroid dehydrogenase, which is a cytochrome P~50-inde~ndent enzyme. Spectrophotometric studies using rat liver microsomal cytochrome P-450 have demonstrated that ketoconazole yields a type II binding spectra, with little effect on NADPH-cytochrome C (P-450) reductase activity [3]. The present study was designed to test whether ketoconazole, known to inhibit cytochrome P-450dependent steroidogenic enzyme activity in both gonadal and adrenal tissues [4,5,6], inhibits the human placental aromatase system in Y&O, Evidence for the involvement of cytochrome P-450 in aromatisation is conflicting. Early studies suggested that aromatisation did not involve oxygen incorporation into the substrates by means of the cytochrome P-450 mono-oxygenase enzyme system because classical inhibitors of cytochrome P-450 mono-oxygenation, such as carbon monoxide [7J had essentially no effect on the aromatisation of androsten~ione [8]. Wowever, aminoglutethimide, a known ~tochrome P-450 inhibitor, inhibits aromatase [9, IO]. Demonstration of type I binding spectra by steroid substrates and inhibitors of placental microsomal aromatase and inhibition of aromatisation by rabbit antiserum *Professor S. R. Stitch died on 15th December 1985 whilst this paper was in preparation.
against rat liver NADPH-Cytochrome C reductase led Thompson and Siiteri to suggest the involvement of cytochrome P-450 [ll]. Recently, Osawa and Higashiyana demonstrated that aromatase activity results only when fractions containing cytochrome P-450 and NADPH-cytochrome reductase activity are combined [12,13]. In the present study, ketoconazole was found to inhibit aromatase in a dose-dependent manner but not the cytochrome P4%independent 178 -hydroxysteroid dehydrogenase and 3~-hydroxysteroid dehydrogen~e-isomerase activities of human placental microsomes. EXPERIMENTAL
Materials [4, 7-3H]Pregnenolone (sp. act. 10 Ci/mmol), [4L4Cjprogesterone (sp. act. 56 mCi/mmol), [2,4,6,7‘H]oestradiol (sp. act. 101 Ci/mmol), [4-Ci4] oestrone (sp. act. 55.8 mCi/mmol) and [1,2,6, 7-3H]testosterone (sp. act. 94 Ci/mmol) were purchased from Amersham International plc, Bucks and checked for purity by TLC (chloroform-acetone, 185:15, v/v). [L4C]Oestradiol was prepared by sodium borohydride reduction of [‘4C]oestrone and purified by TLC in ethyl acetate-iso-octane, 8:2, v/v. All non-radioactive steroids were purchased from Steraioids Ltd, Croydon. Phosphate and diethyl ether were obtained from B.D.H. Chemicals Ltd, Poole, Dorset. Sucrose, NAD, NADPH and fi-mercaptoethanol were obtained from Sigma Chemical Co. Ltd, Poole, Dorset. Acetone, chloroform, ethyl acetate, iso-octane and Triton-X scintillation fluid were obtained from Vickers Laboratories Ltd., Burley-in-Wharfedale and ethanol was purchased from James Burrow, London. TLC plates were kieselgei 60 and obtained from E. 981
982
M. Avue and S. R. STITCH
Merck, Darmstadt. Ketoconazole was supplied by Janssen Pharmaceutical Ltd, Grove, Wantage, Oxfordshire. Methoa!s Preparation of human placenta/ microsomes. A human, term placenta, obtained immediately after vaginal delivery follo~ng normal pregnancy, was transported to the laboratory on ice and processed immediately at 04°C. The placenta was dissected free of adhering membrane and blood vessels and minced with scissors. Homogenization was carried out in cold phosphate buffer (25mmol/l, pH 7.4) containing sucrose (0.25 mol/l) and p-mercaptoethanol (7 mmol/l) using 3 x 10 s bursts of a motordriven homogeniser, separated by 2 min cooling periods. The homogenate was centrifuged at 10,~~ for 30 min. The supematant obtained was centrifuged at 100,OOOg for 60 min. The microsomal pellet thus obtained was re-suspended in buffer (20 ml), treated briefly with a TetIon-glass homogeniser to ensure full dispersion and divided into portions (1 ml) for storage at -20°C. The protein concentration of the suspension was determined by the method of Lowry et al.f14]. Determination of enzyme activities
(i) Aromatase: [3H]Testosterone (50 nmol/l, 0.1 PCi) was incubated with placental microsomal suspension (63 pg protein) and NADPH (I 0 pmol/l) in the phosphate-sucrose-mercaptoethanol buffer (final volume, 1 ml). (ii) 17fl-HSD: 13H]~stradiol (5 @moljl, 0.1 PCi) was incubated with placental microsomal suspension (63pg protein) and NAD (0.1 mmol/l) in the phosphate-sucrose-mercaptoethanol buffer (final volume, 1 ml). (iii) 3/I-HSD-I: [3H]Pregnenolone (2 pmol/l, 0.15 PCi) was incubated with placental microsomal suspension (63 pg protein) and NAD (1 mmoljl) in the phosphat~su~ro~mer~pt~thanol buffer (final volume, 1 ml). Each incubation mixture was pre-incubated for 5 min at 37°C without microsomes. The reaction was started by addition of the microsomal suspension. Incubations were terminated after 10 min by addition of chloroform (1 ml) containing (i) for aromatase: [‘4C]oestradiol, [14C]oestrone (each 10,000 dpm) and non-radioactive testosterone, oestradiol and oestrone (37~mol/l each); (ii) For 17@-HSD: [14C]oestrone (10,OOOdpm) and non-radioactive oestrone and oestradiol (37 pmol/l each); (iii) For 3#?-HSD-I: [‘4C]progestrone (5,000 dpm) and non-radioactive pregnenolone and progesterone (3 18 pmol/l each). For each incubation, the steroids in the mixture were extracted with ether (3 x 4ml). The extracts were combined then evaporated and the residues dissolved in ethanol (25~1) and the appropriate steroids were isolated by TLC as follows: Aromatase and 17P-HSD: ethyl acetate-iso-octane (8:2, v/v)
separated androsten~ione (& = 0.60), testosterone (0.85) oestrone (0.77) and oestradiol (0.65). 3/I-HSD-I: chloroform-acetone (185: 15, v/v) separated pregnenolone (0.56) and progesterone (0.77). The location of the radioactive steroids after chromatography was assessed from the location of authentic steroids chromatographed on the same plates. The areas of silica gel containing the labelled steroids of interest were separately transferred to counting vials and the ‘H and 14Ccontent determined by liquid scintillation counting using T&on-X scintillation fluid (10 ml). Counting efficiency for [3H] was 54%, that for [14C] was 90%, and the cross-over of 14C counts into the ‘H channels was 22%. The activity of aromatase was calculated as the sum of rH]oestrone and [3H]~stradiol formed from [3H]testosterone and expressed in terms of unit time and protein. The yield of [3H]oestrone and [3H]oestradiol was corrected for manipulative loss using the measured recovery of [i4C]oestrone and [‘4C]oestradiol, respectively. 17fi-HSD activity was calculated as the amount of [‘Hloestrone formed from [3H]oestradiol. The yield of [3H]oestrone was corrected for manipulative loss using the measured recovery of [‘4C]oestrone. Similarly, 3b-HSD-I activity was calculated as the amount of [3H]progesterone formed from 13H]pregnenolone. The yield of [3H]progesterone was corrected for manipulative loss using the measured recovery of [‘4C]progesterone. The radiochemical purity of the product of each enzyme assay was checked after chromatography and final identification based on constant [‘H]/[r4C] ratios obtained through the following procedures: (i) rechromatographed in the solvent system ethyl acetate-iso-octane, 8:2, v/v in which the products had identical chromatographic mobilities to those of authentic standards, (ii) eluted and rechromatographed (ethyl acetate-iso-octane, 8:2, v/v) after treatment with a mixture of pyridine-acetic anhydride (2: 1, v/v; data not shown). RF3XJLTS E$ect of ketoconazole on aromatase
The amount of oestrogen formed was related linearly to the amount of microsomal protein (up to 19Opg protein) and time (data not shown). The K,,, of testosterone for aromatase was 30 rt 1.l nmol/I (mean t_SEM, n = 6) and the I’,,,,, was 2.0 It 1.9 nMf63 pg protein/min (mean ItSEM, n =6). Figure 1 shows the percentage inhibition of aromatisation as a function of increasing concentration of ketoconazole. Fifty percent enzyme inhibition was found in the presence of 2 flmol/l ketoconazole and complete inhibition occurred at 7.4 pmol/l. The type of inhibition (Fig. 2) by ketoconazole at a concentration of 2.46pmol/l was then determined. Inhibition was non~om~titive with respect to substrate with a K, value of 3.0 + 1.I @mol/I (mean ISEM, n = 6).
983
Inhibition of aromatase by ketoconazole
1lot y -:::::(z ,:“.: Y 0
I
I
I
2
4
6
(pmol/
I)
I 6
10
0
Ketaconazole
Ketoconozole
.:’
,
30
20
concentration
100
(pmol/ll
Fig. I. Effect of ketocanozole on the aromatisation of [3H]testosterone by human placental microsomes. The error bars represent the variation around the mean of duplicate incubations of two independent experiments.
Fig. 3. Effect of ketoconazole on human placental microsomal 3B-HSD-I. The error bars represent the variation around the mean of duplicate incubation of two independent experiments.
Efsects of ketoconazole on 3/?-HSD-I
corresponding decrease in amount of testosterone metabolised. This implies that ketoconazole inhibits 19-hydroxylation of testosterone, the first of three postulated hydroxylation steps in the aromatisation of androgens [ 151.The results of the present study are in Variance to those of Mason et a1.,[16] using androstenedione (K,,,, 220 nmol/l) as the substrate for aromatase, they found an I, value of 60pmol/l for ketoconazole whereas in the present study testosterone (K,, 30 nmol/l) was used as the substrate and an I, value of 2 pmol/l was found for ketoconazole. Furthermore of the imidazole derivatives studied by Mason et al., miconazole was found to be the most potent inhibitor with an I,, value of 0.6pmol/l and was a competitive inhibitor with a Ki value of 55 nmol/l. In contrast, in the present study Ketoconazole was found to be a non-competitive inhibitor of aromatase with a K, value of 3.0~mol/l.
and 17fl-HSD
Figure 3 shows that increasing concentrations of ketoconazole have no effect on [‘Hlpregnenolone conversion to [3H]progesterone by placental microsomal 3/?-HSD-I activity. Similarly, ketoconazolc was found to be without effect on the formation of [3H]oestrone from [3H]oestradiol by 17/I-HSD (Fig. 4). DISCUSSION
Ketoconazole inhibits several cytochrome P-450mediated steroid hydroxylations in both gonads and adrenals [4,5,7, 81. The ability of this compound to inhibit aromatisation by placental microsomes, as demonstrated in the present study, supports the contention that a cytochrome P-450 is involved in aromatisation. The reduction in oestrogen formation in the presence of ketoconazole correlated with a
Lack of inhibition of placental microsomal 3/3-HSD-I and 17fl-HSD by ketoconazole is in accordance with the results of Sikka et a1.[2], who also
0.25 r
‘-
.c_ E 2 a
0.15
. Ketoconarol.?
absent
0 Ketocanozole
present
-
20 t
---- Ketoconorole Ketoconazole
absenl
present
10 I 0.04
0.02
0.00
0.02
0.04
0.06
t
0.06
0
I
I
I
10
,
20
30
Ketoconazole
Fig. 2. Lineweaver-Burk plot of ketoconazole inhibition of human placental microsomal aromatase. Each point represents the mean of duplicate incubations of six independent experiments.
concentration
II
,
100
(pmol/l)
Fig. 4. Effect of ketoconazole on human placental microsomal 17b-HSD. The error bars represent the variation around the mean of duplicate incubations of two independent experiments.
984
M. Avue and S. R. STITCH
failed to observe inhibition of rat testicular 17/3-HSD. Further studies are required to determine if ketoconazole inhibits aromatase in oivo; oestrogen levels in blood have not been measured in patients receiving this drug. However, interpretation of oestrogen levels in blood of treated patients would be complicated by the fact that ketoconazole also inhibits synthesis of androgen which act as oestrogen precursors. Low oestrogen concentrations therefore may not be due solely to an effect on aromatase. Deflice et a1.[17] were the first to report gynaecomastia in patients receiving the drug. An increase in the free oestrogen-androgen ratio is thought to be a prerequisite for the development of gynaecomastia [18]. However, if ketoconazole inhibits both oestrogen and androgen biosynthesis in vitro, then development of gynaecomastia by ketoconazole could be through some other mechanism. Acknowledgement-The author is grateful for the advice of Dr R. E. Oakey in the preparation of this paper.
REFERENCES
Borgers M., Van den Bossche H. and de Brabander M.: The mechanism of action of the new antimycotic ketoconazole. Am. J. Med. (Suppl.) Proceedings of a symposium on New Developments and Therapy for the Mycoscs (1983) pp. 2-8. Sikka S. C., Swerdloff R. S. and Rajfer J.: In vitro inhibition of testosterone biosynthesis by ketoconazole. Endocrinology 116 (1985) 1920-1925. Sheets J. J. and Mason J. I.: Ketoconazole: A potent inhibitor of cytochrome P-450 dependent drug metabolism in rat liver. Drug Metab. Disp. 112 (1984) 603-606. Kowai J.: The effect of ketoconazole on steroidogenesis in cultured mouse cortex tumour cells. Endocrinology 112 (1983) 1541-1543. Loose D. S., Kan P. B., Hirst M. A., Marcus R. A. and Feldman D.: Ketoconazole blocks adrenal steroido-
genesis by inhibiting cytochrome P-450 dependent enzymes. J. clin. Invest. 71 (1983) 1495-1499. 6. Pant A., Williams P. L., Loose D. S., Feldman D., Peitz R. E., Bochra C. and Stevens D. A.: Ketoconazole blocks adrenal steroid biosynthesis. Ann. Int. Med. 97 (1982) 370-372.
I. Meighs R. A. and Ryan K. J.: Cytochrome P-450 and steroid biosynthesis in the human placenta. Biochim. bioohvs. Acta 165 (1968) 47&482.
8. Chakraborty J., Hopkins R. and Parke D. V.: Inhibition studies on the aromatisation of androstenedione by human placental microsomal preparations. Biochem. J. 130 (1977) 19P-20P. 9. Graves P. E. and Schanick H. A.: Stereoselective inhibition of aromatase by enantiomers of aminoglutethimide. Endocrinology 105 (1979) 52257. IO. Thompson E. A. and Siiteri P. K.: The involvement of human placental microsomal cytochrome P-450 in aromatisation. J. biol. Chem. 249 (1974) 5373-5378. 11. Osawa Y. and Higashiyama T.: Isolation of human placental cytochrome P-450 and its mechanism of action of androgen aromatisation. In Microsomes, Drug Oxidations and Chemical Carcinogenesis (Edited by M. J. Coon, A. H. Conney, R. W. Eastbrook, A. V. Gelboin, J. R. Gillete and P. J. O’Brien). Academic Press, New York, Vol. I (1980) pp. 225-228. 12. Zachariah P. K. and Juchan M. R.: Inhibition of human placental mixed function oxidations with carbon monoxide: Reversal with monochromatic light. J. sferoid Biochem. 8 (1978) 221-228. 13. Jachan M. R. and Zachariah P. K.: Displacement
of carbon monoxide from placental cytochrome P-450 by steroids: Antagonistic effects of androstenedione and 19-norandrostenedione. Biochem. biophys. Res. Commun. 65 (1975) 16261632.
14. Lowry 0. H., Rosebrough N. S., Farr A. L. and Randall R. S.: Protein measurement with the Folin phenol reagent. J. biol. Chem. 193 (1951) 265-275. 15. Fishman J.: Biochemical mechanism of aromatization. Cancer Res. 42 (1982) 3277s-3279s. 16. Mason J. I., Murry B. A., Olcott M. and Sheets J. J.: Imidazole antimycotics: inhibitors of steroid aromatase. Biochem. Pharmac. 34 (1985) 1087-1092.
17. Deflice R., Johnson D. G. and Galgiani J. N.: Gynaecomastin with ketoconazole. Antimicrob. Agents. Chemother. 19 (1981) 1073-1074. 18. Wilson J. D. and MacDonald P. C.: The pathogenesis of gynaecomastia. Adv. Intern. Med. 25 (1980) l-32.