Field Of the IPVIPTI_.-nn The present invention relates to compounds which inhibit P-450 _c17 or 17,20-lyase enzyme and thus inhibit the formation of testosterone, and estrodiol, compositions containing such compounds, and methods of use thereof.

BackcrronTir- f the Invention

Androgrens and estrogens play an important role in stimulating growth and development of sexual characteristics as well as in promoting anabolic effects such as retention of nitrogen, phosphorus, calcium and potassium. In addition, these steroid hormones play an important role in a variety of sex hormone dependent pathologies.

Androgens are synthesized from cholesterol and originate primarily from three glands: the testes, the adrenal cortex and the ovary. The primary androgen from the testes is testosterone, while the weakly hormonal androgens, dehydroepiandrosterone (DHEA) , its sulfate ester and androstenedione, are the primary C ₁₉ steroids secreted by the adrenal and ovary. Each gland appears to secrete its products with a circadian variation corresponding to some degree with the secretions of stimulatory proteins from the pituitary. The release of these proteins, primarily luteinizing hormone (LH) for the testis-ovary and adrenocorticotrophic hormone (ACTH) for the adrenal, in turn, is controlled primarily by hypothalamic factors, such as luteinizing hormone releasing hormone and corticotrophic releasing factor, respectively. Testosterone and its metabolite, estradiol and Sα-dihydrotestosterone are able to influence LH release at the neural and pituitary levels through a feed¬ back mechanism. The release of ACTH is sensitive to the levels of circulating cortisol. Thus, the androgen level

in an individual is dependent on a balance of several interrelated stimulatory and inhibitory processes. The end organ response to androgens, including that of feedback control, is mediated through the androgen receptor which acts in the cell at the nuclear level to modulate RNA synthesis. The nuclear androgen receptor has been found to be associated with only two natural androgens, testosterone or 5α?-dihydrotestosterone (DHT) . DHT binds more avidly to the androgen receptor than testosterone and, in certain tissues, such as the rat prostate, DHT is the exclusive androgen present in the nuclear isolates.

Genetic deficiencies in androgen-synthesizing enzymes or in the androgen receptors result in development of male fetuses with female characteristics. Likewise, female fetuses are virilized by exposure to androgen. Excess androgenic activity is implicated in precocious puberty, male hypersexuality, seborrhea, acne, male pattern baldness and hirsutism. Prostate size and maintenance are androgen dependent. Male fertility control could be achieved by blocking testosterone- stimulated spermatogenesis.

A number of androge -dependent pathologies can be treated directly by inhibiting the production of testosterone in the body. Among these pathologies are prostate cancer, benign prostatic hyperplasia, precocious masculization and testicular Leydig cell tumors. Likewise, a number of estradiol-dependent pathologies can be treated directly by inhibiting the production.of estradiol in the body. These pathologies include estradiol-dependent breast cancer, ovarian cancer and uterine cancer.

Prostate cancer is now second only to lung cancer in frequency, accounting for 18% of all male cancers (approximately 100,000 new cases per year) and generally affects men over age 50. Benign prostatic hyperplasia, although usually not fatal, is the second

leading cause of surgery in the U.S., with over 400,000 prostatectomies performed each year. However, this only represents 20-25% of the men exhibiting symptoms.

Prostate cancer and benign prostatic hyperplasia do not appear to be related, since these diseases most frequently occur independently and develop in anatomically distinct regions of the prostate gland (Hodges et al. in "Benign Prostatic Hypertrophy", F. Hinman, Jr., Ed., Springer-Verlag, New York, N.Y., 1983, p. 167). Although their etiology remains unknown, an endocrine basis is supported, since both diseases occur almost exclusively in men with functional testes and regress following castration (Coffey et al. in "Current Concepts and Approaches to the Study of Prostate Cancer," D.S. Coffey et al., Eds., Alan R. Liss, Inc., New York, N.Y. p.l;

Huggins et al., Cancer Res. 1 , 293 (1941); Huggins et al., J. Urol. 43. 705 (1940); P.C. Walsh in "New Approaches to the Treatment of Benign Prostatic Hyperplasia, " F.A. Kimball et al., Eds., Alan R. Liss, New York, N.Y. , 1984 p.l.).

The most common form of prostate cancer is a malignant transformation of epithelial cells in the peripheral region of the prostate gland (adenocarcinoma) (R.O. Petersen, in "Pathology", E. Rubin and J.L. Farber, Eds., J.P. Lippincott Co., Philadelphia, PA 1988, p.927) . Hormonal treatment is presently limited to metastatic disease and is considered palliative, since evidence supports that prostate cancer progresses from an androgen-dependent to an androgen-independent phase (Bruchovsky et al., in "Current Concepts and Approaches to the Study of Prostate Cancer", D.S. Coffey et al., Eds. Alan R. Liss, Inc., New York, N.Y. 1987 p.347.). Castration, estrogens (diethylstilbestrol, ethynyl estradiol, estramustine) , progestins (megestrol acetate, medoxyprogesterone acetate, progesterone) , androgen receptor antagonists (cyproterone acetate, flutamide) , steroidogenic inhibitors (aminoglutethimide;

spironolactone) , and LHRH agonists have all been used for the treatment of metastatic prostate cancer and have been previously reviewed (Grayhack et al., Cancer __Q.i 98 (1987); Melamed, Drug Intelligence and Clin. Pharm. 21. 247 (1987); J.A. Smith, Jr., Urol. 137. 1 (1987)).

Benign prostatic hyperplasia, or nodular hyperplasia, represents a non-malignant hyperplasia and hypertrophy of both stromal and epithelial elements in the periurethral and central regions of the prostate gland. This is an extremely common disorder in men over 50. The histopathologic state can vary from a "pure" stromal to a "pure" epithelial hyperplasia/hypertrophy. Although the cause of nodular hyperplasia is still uncertain, available evidence suggest that both androgens and estrogens are involved in its genesis. Surgery via transurethral resection currently represents the leading treatment for benign prostatic hyperplasia, and has a high degree of efficacy and safety. Although benign prostatic hyperplasia responds to either surgical or chemical castration, these are not acceptable forms of treatment to a majority of the patient population. Consequently, it is important that any long-term pharmacological treatment for non-life threatening cases of benign prostatic hyperplasia have minimal side effects. Testicular tumors occur with an incidence of about 2 to 2.5 per 100,000 men per year in the United States, and account for less than 1 percent of cancer deaths in men. There are four types of germinal-cell tumors: seminomas, embryonal carcinomas, choriocarcinomas, and teratomas. Forty percent of germinal-cell tumors contain two or more cell types.

The most common stromal-cell malignancies are Leydig cell and Sertoli cell tumors. These tumors, which are rare in children, are usually benign. Leydig (interstitial) cell tumors may result in masculinization or feminization or both, or they may have no hormonal effect. Sertoli cell tumors (androblastomas) contain

cells that are arranged in tubule-like fashion and may also result in feminization or (rarely) virilization. Testicular Leydig cell tumors can be treated by inhibiting formation of testosterone, since Leydig cells are especially rich in 17,20-lyase, and neoplasms arising from these cells are particularly sensitive to compounds that interfere with the action of this enzyme.

Elevated testosterone production in patients with testicular tumors can arise by two mechanisms. In trophoblastic tumors and in tumors of Leydig and Sertoli cells, production of both hormones occurs autonomously in the tumor tissue itself; in these instances, plasma gonadotropin levels and hormone production by the uninvolved portions of the testes are depressed, and azoospermia is common. However, when gonadotropins are secreted by the tumor, the gonadotropin acts to increase estradiol and testosterone production in the unaffected areas of the testes, and azoospermia is uncommon.

About 40% of testicular tumors are composed of more than one pattern of cell involvement. The most common of these mixtures is that of teratoma and embryonal carcinoma, which comprises 25% of all testicular neoplasms. Less often, the teratoma has varying components of embryonal carcinoma, seminoma, or choriocarcinoma, as well as sarcomatous elements, and several of these features together. The designation teratocarcinoma has been applied to mixtures of teratoma with these forms of cancer.

_*

Inhibition of androgen biosynthesis has been used for the clinical treatment of hormone-dependent prostatic cancer and other androgen-dependent disorders (Van Wauwe et al., (1989) J. Med. Chem. __2 (10)2231; Gower, D.B. (1974) J. Steroid Biochem. ϋ, 501; Worgul et al., (1983) J. Urology 129. 51; Santen et al., (1983) ___. Clin. Endocrinol. Metab. __η_, 732; Feldman, D. (1986) Endocrine Rev. 2, 409; Trachtenberg et al., (1984) The Lancet. 433) .

A review of treatments for proliferative prostatic diseases appears in Annual Reports in Medicinal Chemistry 24. Chapter 21, New Horizons in the Treatment of Proliferative Prostatic Disease. Academic Press, Inc. p.197 (1989) . Thus, it would be particularly useful to provide compounds that inhibit the formation of androgens.

A number of androgen receptor antagonists have been studied, and cyproterone acetate has been approved for use in prostate cancer in several countries. However, the use of cyproterone acetate or androgen deprivation in benign prostatic hyperplasia has been questioned. Flutamide, a non-steroidal pure androgen receptor antagonist, is under investigation in the U.S. Use of flutamide in benign prostatic hyperplasia shows some improvement in subjective parameters, as well as a reduction in prostate size. Anandron is another non- steroidal androgen receptor antagonist primarily targeted for use in hormone-responsive prostate cancer. However, this compound acts as an androgen receptor antagonist in peripheral tissues, as well as centrally, and there have been concerns over the ocular toxicity of this compound. Oxendolone has been investigated for use in conjunction with other androgen receptor antagonists. Another type of compound currently in use or being investigated for use in treatment of prostate cancer or benign prostatic hyperplasia are compounds which modulate steroidogenesis. For example, ketoconazole lowers serum testosterone levels to near castrate levels at high doses, although there are pronounced side effects that limit its use as first line therapy of prostate cancer. Limited studies have been conducted on aromatase inhibitors, as there have been suggestions of a role for estrogens in spontaneous benign prostatic hyperplasia. Reductase inhibitors block the formation of DHT at the cellular levels without altering plasma testosterone levels.

Leutenizing hormone releasing hormone (LHRH) agonists/antagonists when administered to males lead to a suppression of testicular steroidogenesis, or "chemical castration. " Among the other approaches used for treating prostate cancer or benign prostatic hyperplasia are alpha blockers, natural products which lower lipids or which interfere with lipid metabolism, and growth factors which regulate neoplastic prostatic growth. Steroid hormones play an important role in the physiological regulation of gene expression (Waterman et al., in Cytochrome P-450: Structure. Mechanisms and Biochemistry, ed. P.R. Ortiz de Montellano, p. 345) . By a complex biosynthetic pathway involving several P-450 enzymes, several categories of steroid hormones are produced in mammalian adrenals, gonads and placentas (Jefcoate, jUaid. , p. 387). Cholesterol is first converted to pregnenolone by mitochondrial side-chain cleavage enzyme, P450 _βcc (Burstein et al. (1975) J. Biol. Chem. 2__Q., 9028; Burstein et al., (1972) Biochemistry 11. 2883; Hume et al. (1978) Biochem. Soc. Trans. 6. 893). Pregnenolone is subsequently metabolized by three pathways to yield progestins, corticoids (mineralocorticoids and glucocorticoids) , and sex hormones (androgens and estrogens) (Miller, W.L. (1988) Endocrine Review £ (3) 298) . The key enzyme leading to these end products is cytochrome P-450 17,20-lyase, an important hemeprotein having two different activities catalyzed by a single active site: 17α-hydroxylation and cleavage of the C ₁₇ -C ₂ o bond (Nakajin et al. (1981) J. Biol. Chem. 256. 3871; Nakajin et al., (1981) J. Biol. Chem. 256. 6134; Nakajin et al., (1981) Biochemistry 20. 4037; Makin, H.L.J., ed. Biochemistry of Steroid Hormones. Blackwell Scientific Publications, London (1975); Kominami et al., (1982) Biochem. Biophys. Res. Commun. 109. 916) .

In the two pathways leading to either cortisol or sex hormones, P-450 ₇ , otherwise known as 17,20-lyase,

first hydroxylates substrates pregnenolone and progesterone at the C ₇ position to yield the intermediates, 17o;-hydroxypregnenolone and 17α- hydroxyprogesterone. These intermediates are then cleaved using molecular oxygen and reduced nicotinaminde adenine dinucleotide phosphate (NADPH) to produce dehydroepiandrosterone (DHEA) and androstenedione, respectively. Both DHEA and androstenedione are key intermediates in the biosynthesis of testosterone and estradiol. Microsomal cytochrome P-450 reductase, a flavoprotein, is required for electron mediation (Oprian et al. (1982) J. Biol. Chem. 257. 8935; French et al. (1980) J. Biol. Chem. 255. 4112) .

Steroid hormone biosynthesis begins with cholesterol, which supplies the steroid nucleus. The first step and rate limiting in this overall synthesis of steroid hormones occurs in the mitochondria where a specific enzyme, cytochrome P-450 _scc , catalyzes the side- chain cleavage of the cholesterol side chain to form pregnenolone. This product diffuses to the endoplasmic reticulum where an isomerase, 17-hydroxylase (17-OH) and 21-hydroxylase (21-OH) enzymes complete the synthesis of various hormonal intermediates for the glucocorticoid and sex hormones. Movement back to the mitochondria, where the 11-hydroxylase (11-OH) enzyme is localized, completes the synthesis of cortisol, which is the major glucocorticoid in humans. A simplified scheme for adrenal steroidogenesis is depicted in the following scheme:

Pregnenolone 17« -OH pregnenolone Dehydroepiαndrosterone

ione

11-deoxycorticosterone 11-deoxycortisol Testosterone

Aldosterone

SUBSTITUTE SH

A class of enzyme inactivators, called suicide enzyme inactivators or mechanism-based inactivators, was first described by onrad Bloch and his collaborators at Harvard University in 1970. These suicide enzyme inactivators are relatively inert molecules which so closely resemble the natural substrate of a specific enzyme that they are structurally indistinguishable from the natural substrate by the enzyme's active site. By interaction with the enzyme at its active site, the inactivator is chemically modified, converting the inactivator into a reactive compound which then reacts with the target enzyme leading to enzyme inactivation. The initial chemical interaction between inactivator and enzyme is exactly the same as that which occurs between the natural substrate and the enzyme in the normal enzyme catalyzed process. The only difference is that by acting on the suicide substrate, the target enzyme catalyzes its own destruction.

An important and distinctive property of suicide substrates which makes them useful for control of biological processes is that they are chemically non- reactive. The reactive inactivators are only generated at the active sites of the enzymes, and therefore non- specific reactions with other than the targeted enzymes do not normally occur. Suicide enzyme substrates are therefore highly selective and bring about irreversible inactivation of the target enzyme -- properties which are beneficial for compounds exhibiting the pharmacological potential of the substrates of the present invention. Reviews of suicide substrates may be found in

Annual Reports in Medicinal Chemistry. Section VI--Topics in Chemistry and Drug Design, Chapter 26 (Academic Press, 1982) ; and Tetrahedron Report Number 124, pages 871 to 909 (1982) . Wilson et al., in U.S. Patent No. 4,560,557 disclose a number of highly selective inhibitors of the cytochrome P-450 cholesterol side chain cleavage enzyme

(P-450 _8CC ). These compounds mimic the natural steroids, but introduce a trimethylsilyl group two carbons removed from (or β to) the carbon which undergoes enzymatic oxidation, or a monosubstituted sulfur group at the carbon which undergoes stereoselective enzymatic oxidation. The trimethylsilyl rationale was also applied towards the development of potent aromatase inhibitors by J.P. Burkhart et al., as reported in Steroids , 45, 357 (1985). The same group attempted to apply the sulfur technology towards development of aromatase inhibitors, but with limited success.

As shown below, the enzyme 17,20-lyase, also known as P- 450 _c17 , is responsible for the conversion of progesterone A to androstenedione B, the immediate precursor of testosterone C.

Progesterone Androstenedione Testosterone

A B C

The development of a mechanism- based inhibitor of 17, 20-lyase, specifically P-450 17, 20-lyase, and its use in the clinical treatment of prostate cancer has been investigated. One compound found to inhibit P-450 17, 20- lyase is a 17β-cyclopropylamine derivative of dihydroisoandrosterone, disclosed in a Masters thesis (Benjamin J. Taylor, Massachusetts Institute of Technology, 1985) and Angelastro, M.R. et al . Biochem. Biophys . Res . Comm. 162 , 1571 (1989) . This compound has the following

P-450 _c17 (otherwise kown as 17,20-lyase), a key enzyme for testosterone synthesis in the prostate, has been extensively studied by Hall et al. as reported in Steroids .4.:131 (1986) and purified to homogeneity. It was found that the same enzyme catalyzes both 17- hydroxylation and C ₁₇ -C ₂ o cleavage of pregnenolone or progesterone at the same active site.

^ft hrmm-i-rγ «?f ^he Invention It is an object of the present invention to overcome the deficiencies in the prior art.

It is another object of the present invention to provide compounds which inhibit the activity of P-450 _c17 or 17,20-lyase. It is yet another object of the present invention to provide compounds which can be used to inhibit testosterone production.

It is yet another object of the present invention to provide compounds which can be used to inhibit estradiol production.

It is further object of the present invention to provide methods and treatment for androgen-dependent pathologies.

It is yet a further object of the present invention to provide methods and treatment for estradiol- dependent pathologies.

It is still a further object of the present invention to provide compounds which can be used to inhibit the production of testosterone in the clinical treatment of prostate cancer and testicular cancer.

It is another object of the present invention to provide compounds which can be used to inhibit the production of estradiol in the clinical treatment of breast cancer, ovarian cancer and uterine cancer. It is yet a further object of the present invention to provide methods and compositions for the treatment of prostate cancer and benign prostate

hyperplasia.

According to the present invention, compounds of the following formulae serve to inhibit 17,20 lyase:

wherein M is sulfur or selenium (preferably sulfur) , R is cyclic or acyclic C . ₆ (preferably C^*) alkyl, C_ . _& (preferably Cj.*) alkenyl, hydroxymethyl, or hydrogen, X is 0 or NH (preferably 0) , and n is 0 or 1, with the proviso that R is not hydrogen when n is 1, or pharmaceutically acceptable salts thereof. In further preferred compounds R is other than hydrogen.

The above-described compounds inhibit the activity of P- 450 17,20-lyase when administered in effective amounts. These compounds are useful for temporarily inhibiting testosterone biosynthesis, and thus ultimately estradiol synthesis, in a way beneficial to patients with testosterone or estradiol related pathologies. Thus, the present invention also provides pharmaceutical compositions comprising a compound as defined above as an active ingredient together with a pharmaceutically acceptable carrier or excipient.

The present invention also provides a method of producing an inhibitory effect on hormonal biosynthesis in mammals, including man, which comprises the administration of an effective inhibitory amount of a compound as described above to a mammalian host sufficient to inhibit testosterone and ultimately, estradiol formation.

Brief Description of the Drawings

Figure 1 shows % control activity versus concentration for compounds 1 and 2.

SUBSTITUTESHEET

Figure 2 shows % control activity versus concentration for compound 3.

Figure 3 shows % control activity versus concentration for compound 10. Figure 4 shows % control activity versus concentration for compounds 11 and 12.

Figure 5 shows preincubation results for compounds 4-6 and 13-16. Inhibitors (lOOμM) were preincubated in the assay system for 5 and 30 minutes in the presence of NADPH. Subsequently, progesterone (5μM, final concentration) was added and incubation continued as described in the detailed description of the invention.

Figure 6 shows 17α-0H progesterone production over time for compound 1. Figure 7 shows 17o;-OH progesterone production over time for compound 10.

Figure 8 shows 17of-OH progesterone production over time for compounds 11 and 12.

Detailed Description of the Invention

Mechanism-based enzyme inhibition is a promising approach to drug design. The inhibition of biological hydroxylations mediated by cytochrome P450 enzymes has particularly wide applications, since the metabolism of drugs and the biosynthesis of many important hormones involve these enzymes. The use of anti-hormones in disease therapy ideally would target only the hormone in question.

Suicide-inhibition based on known organosilicon chemistry of stabilized β-cations or β-radicals was first reported by Nagahisa et al. in J. Am. Chem. Soc. 106. 1166, 1984. (20S) -20- (2-Trimethylsilyl) -pregn-5-en- SB,20-diol (20S-TMS) was shown to be a potent mechanism- based inhibitor of cytochrome P-450 _scc enzyme which is found in the adrenal, ovarian, and testicular glands. This enzyme catalyzes the oxidative cleavage of the cholesterol side chain to form pregnenolone. During the

catalytic cycle, 20S- TMS is believed to form a β-TMS stabilized carbocation or free radical. Such species could then form a covalent adduct with the enzyme by silylating a proximal nucleophile. As noted above, Wilson et al. in U.S. Patent No.

4,560,557, disclose both silicon-substituted and sulfur- substituted steroids as irreversible inhibitors of biosynthesis of P-450 _ecc .

The inhibitor design rationale based upon the introduction of a beta trimethylsilyl group or a replacement by a sulfur atom on the carbon which undergoes enzymatic oxidation was applied towards the design and synthesis of compounds as potential mechanism-based inhibitors for P-450 17,20-lyase. It was expected that enzymatic abstraction of a β-hydride or a β-hydrogen would generate a carbocation or radical species which would be inductively stabilized by a strategically placed trimethylsilyl or a substituted sulfur group. Such species could then form a covalent adduct with the enzyme by silylating a proximal nucleophile. The sulfur analogs inactivate the enzyme by their enzymatic oxidation to tight-binding sulfoxides.

However, it was found that the binding energies and stereochemistry of P-450 _ecc and P-450 _c17 are unpredictable and quite different from each other. The characteristics of the reactions of the two enzymes are different, such that P-450 _scc acts on substituted sulfur steroids to form essentially only one sulfoxide product which is tightly bound to the enzyme. P-450 _c17 , on the other hand, may react with sulfur substituted steroids to form sulfoxides which could readily dissociate from the enzyme.

Further evidence of this unpredictability is demonstrated by testing trimethylsilyl derivative compounds which were analogous to the trimethylsilyl derivatives of Wilson et al., U.S. Patent No. 4,560,557, which were found to be successful suicide substrates for

P-450 _8cc . None of the trimethylsilyl derivatives tested was an inhibitor of the 17o;-hydroxylase activity of 17,20- lyase. This is consistent with the demonstration of Vickery et al. in Proc. Natl. Acad. Sci.. USA 22:5773, 1982, which showed that (20R) -20-phenyl-5-pregnene-3-B,20- diol (20-PPD) , a potent inhibitor of P-450 _8CC , does not inhibit 17,20-lyase activity. The lack of effect was attributed to the bulky phenyl substituent at the C-20 position. It appears that the active site pocket of 17,20-lyase, unlike P-450 _8CC , binds to its substrates such that there is little room left to accommodate large substituents such as TMS or a phenyl group.

Cyclobutyl and cyclopropylamines have been demonstrated to be mechanism-based inhibitors of oxidative enzymes such as cytochrome P-450 and monoamine oxidase (MAO) . Guengrich et al., in J. Am. Chem. Soc. 106:6446 (1984) proposed that the cyclopropylamino moiety inactivated enzymes via a β-iminium radical intermediate. While cyclopropylamine derivatives of dehydroisoandrosterol have been found to inhibit 17,20- lyase, as shown below, the cyclobutyl compounds were found to be inactive.

It has been found that incorporation of sulfur into strategic positions along the side chain as well as steroid nucleus has led to potent mechanism-based inhibitors for enzymes involved in steroid biosynthesis such as P-450 _8cc , aromatase, P-450c 11, and lanosterol sterol methyltransferase. See Waterman et al., supra. However, because P-450 _scc is the first step in steroid hormone biosynthesis, compounds that interfere with P- 450 _8CC shut down the production of all steroid hormones. In order to inhibit the formation of testosterone, and ultimately estradiol, 17,20-lyase is an important target for selective inhibition. Hall et al. in Biochem. Biophys. Res. Commun.

111:512. 1984, observed that intermediate 17α-hydroxyprogesterone (and 17α-hydroxypregnenolone) is

not tightly held within the active site, unlike the tightly bound hydroxylated intermediates in the cholesterol side-chain cleavage reaction. This important feature distinguishes between the two side-chain cleavage enzymes despite the fact that they employ similar catalytic mechanisms. Therefore, one would expect that inhibition of P-450 _8CC would not necessarily result in inhibition of P-450 _c17 .

The biological effects of a number of compounds as inhibitors of 17,20-lyase were investigated in the porcine testicular microsome system. 17α-hydroxylase activity was assayed based on methods developed by Hall as reported in Nakajin et al, 1981, J. Biol. Chem.. 256. 3871. Biological evaluation of several of these compounds has evidenced inhibition of androgen and estrogen production; these compounds are effective in treatment of several diseases, including prostate cancer.

In the examples given below, the general 17oi- hydroxylase assay was conducted as follows: A 1 ml assay mixture contained in final concentrations ¹⁴ C-progesterone (14mm), MgCl ₂ (5 x 10" ⁴ M), glucose-6-phosphate (5 x 10" ³ ), glucose-6-phosphate dehydrogenase (0.15 units/ml) , NADPH (200 nmoles) in potassium phosphate buffer (50 mM) , pH, 7.25) and 0.076 nmol P-450. The components of this assay are readily commercially available from a variety of sources including Sigma Chemicals (St. Louis, MO) . The reaction was initiated by addition of NADPH after three minute preincubation at 37°C in a shaking water bath. After a fifteen minute incubation, duplicates of the mixture were quenched with 1 N HCl. Controls were run under identical conditions lacking either NADPH or protein. The assay mixture was then filtered through mini reverse phase C-18 silica columns. The columns were washed twice with water and eluted with HPLC grade methanol. Progesterone, 17o;- progesterone, and androstenedione carrier steroids, all available from Sigma Chemicals, were added in quantities

of 40 μg each, the eluant was dried, applied to 250 micron silica gel polyester plates (Aldrich Chemicals, Milwaukee, Wisconsin) , and developed twice in hexane/ ethyl acetate (3:1). The carrier steroids were located by UV, cut out, and ¹⁴ C was measured by liquid scintillation counting. The 17of-hydroxylase activity was determined by adding the ¹⁴ C contents of the 17oi-hydroxyprogesterone and androstenedione carrier zones as described by Hall.

The inhibitor titration assay was conducted essentially as above, with slight modifications. At T=0, 5 μl inhibitor stock ethanolic solution was added and the tubes were placed into a shaking 37°C water bath. The inhibitor concentrations ranged between 0.1 and 100 μM final concentration and an equivalent volume of ethanol was added to the controls, which contained no inhibitor. At T = 3 minutes, 200 nmoles of NADPH was added and, after fifteen minutes, the sample was quenched and the steroids were extracted as described above.

The time course production assay for 17α- hydroxyprogesterone was conducted by fixing the inhibitor concentrations between 1 and 5 μM and the time interval between addition of NADPH and quenching varied between 0, 5, 10, 15, 30, 45, and 60 minutes.

For kinetic determination of inhibition type, inhibitor concentrations used were 0.0, 0.5, 1.0 and 2 μM final concentrations. The substrate concentrations used were 1.0, 1.66, 2.5, and 5 μM final concentrations. The time between addition of NADPH and quenching remained at fifteen minutes. The compounds evaluated were as follows:

R'= R = methyl

SUB T

Compounds 1-16 above were initially evaluated in porcine testicular microsomes for their ability to inhibit the 17o;-hydroxylase activity of P-450 17,20-lyase. Since 17α-hydroxylase and 17,20-lyase activities are mediated by a single hemeprotein, it would be expected that both activities would be affected by an inhibitor. The concentrations studied ranged between 0 and 100 μM. These results, shown in Figures 1-8, are based on the percent of control activity. As demonstrated in Figures 1-8, compounds 1, 2, and 10-12 were found to be effective inhibitors of hydroxylase activity. Compounds 3-6 and 13-16, represented in Figure 5, showed no inhibitory abilities, with only marginal effects seen for compound 4. Based upon these preliminary concentration titration results, the most promising inhibitors were further evaluated for time dependence of inhibition. 17of- progesterone time course production results are shown in Figures 6-8. In the presence of substrate and inhibitor compounds 10-12, 17oi-progesterone production is essentially linear, with no significant conversion of the steroid inhibitors. As shown in Figure 6, compound 1 showed linear product production for approximately fifteen minutes and was followed by a lag in activity. Separate incubation of 2.0 μM of compound 1 with bovine adrenal cytochrome P-450 _8CC showed no inhibition.

Compounds 10-12 were determined to be competitive inhibitors of progesterone with apparent K _f values represented in Table 1 below.

TABLE 1

COMPOUND APPAR. K,- (μM) Is ₀ (μM) INHIBITION-TYPE

Besides being useful as pharmaceutical agents, enzyme inhibitors are important because they can define the structural requirements for active site binding and, in the case of mechanism-based activators, are useful in elucidating reaction mechanisms. The most successful inhibitors of P-450 17,20-lyase were those which contain moieties with lone electron pairs which coordinate with the heme iron but are also able to interact with the hydrophobic binding pocket in the active site. Examples of these include pyridines, imidazoles, and amines. However, these inhibitors all suffer from a lack of specificity towards the cytochrome P-450, which lack of specificity makes them less useful for specific inhibition of production of testosterone.

The inhibitor design rationale based upon the introduction of the trimethylsilyl (TMS) group β to the carbon which undergoes enzymatic oxidation has been applied towards the design and synthesis of compounds 13- 16 as potential mechanism-based inhibitors for 17,20- lyase. It was expected that enzymatic abstraction of a β-hydride or β-hydrogen would generate a carbocation or radical species which would be inductively stabilized by a strategically placed TMS group. Such species could then form a covalent adduct with the enzyme by silylating a proximal nucleophile. Compounds 4-6 were prepared as a

new class of mechanism-based inhibitors based on the design of aminoalkyl TMS mechanism-based irreversible inhibitors for bovine MAO.

It is clearly shown in Figure 5 that none of the seven TMS derivatives was an inhibitor of the 17α- hydroxylase activity of 17,20-lyase. Compound 4, the best for this series of compounds, exhibited marginal inhibition at high concentrations (100 μM) and retained 77% control activity at concentrations 20 times greater than that of the substrate. The apparent lack of inhibitory activity for compounds 4-6 and 13-16 can be attributed to the TMS group, whose steric bulk may prevent this group of compounds from binding to the enzyme active site. Vickery et al. (Proc. Natl. Acad. Sci.. USA 2£:5773, 1982) demonstrated that (20R) -20-phenyl-

5-pregnene-3-β,20-diol (20-PPD) , a potent inhibitor of P-450 _8cc , does not inhibit 17,20-lyase activities. The lack of ef ect was attributed to the bulky phenyl substituent at the C-20 position. It appears that the active site pocket of 17,20-lyase, unlike P-450 _8CC , binds to its substrates such that there is little room left to accommodate large substituents such as TMS or a phenyl group.

Another class of steroids evaluated as possible mechanism-based inhibitors of P-450 _c17 are 17β- cyclobutylamine and cyclopropylamine derivatives of dihydroisoandrosterone compounds 1-3. The design of these steroids was inspired by the demonstrated effect of cyclobutyl- and cyclopropylamines as mechanism-based inhibitors of oxidative enzymes such as cytochrome P-450 and monoamine oxidase (MAO). Guengrich (op. cit.) proposed that the cycloproplyamino moiety inactivated enzymes via a β-iminium radical intermediate.

Both compounds 1 and 2 were determined to be concentration-dependent inhibitors of 17,20-lyase. As can be seen in Figure 1, compound 1 is slightly more inhibitory than compound 2, although the two compounds

differ only by a methyl group substituent on the amine. The slight inhibitory differences can be attributed to experimental error (+.10%) . The cyclobutylamine derivative 3, on the other hand, is not an inhibitor of hydroxylase activity. It should also be noted that compound 1 is not an inhibitor of P-450 _8cc . To confirm that inhibition by the cyclopropylamine steroids is due to the cyclopropyl ring rather than the mere presence of a nitrogen atom at the nor-20-position, three aminoalkyl steroid derivatives were evaluated as potential inhibitors of 17α-hydroxylates activity, compounds 7-9. These compounds are not inhibitors for 17,20-lyase. Based upon these findings, it appears that the nitrogen atom at the nor C-20 position does not aid in binding or heme coordination, as the distance between the nitrogen lone pair and heme iron center may be too great to allow for orbital interaction. The inhibitory properties of compounds 1 and 2 may be attributable to the ring-opening mechanism proposed for cyclopropylamines. The lack of effect by compound 3 may be due to the steric bulk of a cyclobutyl ring which could either block access to the nitrogen atoms or, more likely, prevent the steroid from binding to the active site. Methylsulfide substitution, compound 10, was found to be a concentration-dependent inhibitor, as shown in Figure 3, and both time course production and double reciprocal plots indicate a competitive mode of with an apparent j of 0.70 μM. It is interesting to note that heteroatom substitution at the nor C-20 position from nitrogen to sulfur permitted compound 10 to be a potent inhibitor relative to its methylamine analog, compound 8. One possible reason is that the electron density at the sulfur relative to nitrogen may allow for some coordination between the lone electron pairs of sulfur to the empty d-orbitals of iron, and this may contribute to the binding interaction. Alternatively, the methylsulfide

could assume a desirable side-chain conformation which allows it to interact preferentially.

It should also be noted that compound 10 did not exhibit biphasic curvature of the time course production of 17o!-hydroxyprogesterone product as seen with 20(S)-22- nor-22-thiacholesterol and its respective enzyme, P-450 _scc . The reaction rate is linear in the presence of inhibitor, which may indicate that no significant conversion of inhibitor 10 occurred. Alternatively, a sulfoxide metabolite, if produced, may not be a significantly greater inhibitor of the enzyme relative to compound 10, and this could also account for the linearity of the reaction rate. Finally, a sulfoxide metabolite, if produced, could readily dissociate from the active site. To determine whether P-450 _cl7 can distinguish between the two possible sulfoxide configurations, sulfoxides 11 and 12 were synthesized and evaluated. As seen from Figure 4, both compounds 11 and 12 show concentration-dependent inhibition. Moreover, the results from time course production of 17of-progesterone (Figure

11) in the presence of 2 μM concentrations of compounds 11 and 12 indicated that both compounds behave as equi-potent competitive inhibitors, in contrast to the case with P-450 _8cc . These compounds have identical apparent K _? values of 0.38 μ values. Coπtpound 12 reduced slightly the reaction rate by 5% relative to compound 11. Based on these results, it is clear that both sulfoxide diastereomers bind equally well to 17,20-lyase.

As shown in the following reaction scheme

35

SUBSTITUTE SHEET

synthesis of potential sulfur steroid inhibitors of P- 450 _c17 is based on the preparation of a secondary mercaptan as described by Wilson and co-workers (Wilson et al, 1983, Orcr. Syn. 61. 74; Guengerich et al 1984, J__ Chem. Sci. 104. 205) . All chemical reagents, starting materials and solvents are commercially available from Aldrich Chemicals (Milwaukee, Wisconsin) . (÷)Dehydroisoandrosterone 17 was treated with ethanedithiol in the presence of BF ₃ etherate complex to give the corresponding thioketal 18. Subsequent cleavage of 33 with excess n-butyl lithium generated the desired 17-mercaptosteroid 19 which was then converted to the lithium salt and quenched with iodomethane to yield the methylsulfide steroid 10. Oxidation of compound 10 with excess 30% H ₂ 0 ₂ gave a mixture of sulfoxide isomers 11 and 12 which was subsequently separated by careful column chromatography and recrystallized. X-ray crystallography and NMR spectroscopy confirmed the stereochemical assignment of the 20-nor-sulfoxides.

17-Met__γlthi oanr -r -ti-em-rrmf-

To prepare 17-methylthioandrosterone, one gram (3.2625 mmoles) of 17-mercaptoandrosterol was dissolved in 100 ml of anhydrous ether contained in a dried 250 ml three-neck flask, then cooled to 0°C. Then, 1.610 ml of 2.5 M n-butyllithium (in hexane, 4.025 mmoles, 1.23 equiv.) was added dropwise, and the reaction was stirred for two hours at room temperature. After cooling to -78°C, 1 ml of hexamethylphosphoramide (HMPA) was added, followed by 0.385 ml (1.190 moles) of iodomethane. The reaction was stirred for 30 minutes at room temperature and was quenched with saturated ammonium chloride solution. The ether layer was separated, washed with 10% HCl and brine, and then dried. Removal of solvent under reduced pressure gave 1.18 gram of a semi-solid white material. Purification via flash column chromatography (10- 20% ethylacetate in hexane) gave 1 gram (yield =

95.6%) of an oil that crystallized upon standing.

Exact mass (C ₂₀ H ₃₂ OS) : Calculated: 320.2174; Found: 320.2184

To prepare sulfoxides of compound 10, 152 mg (0.472 mmoles) of methylsulfide androsterol (compound 10) was dissolved in 20 ml acetone in a 50 ml round bottom flask. Five ml (excess) of 30% hydrogen peroxide was added, and the reaction mixture was stirred for three days at room temperature and pressure. Water was added to the solution, and the precipitated product was extracted with ethyl ether. The ether layer was separated, washed twice with saturated sodium sulfite and brine, then dried. Removal of solvent afforded 440 mg of a crude clear oil which contained a mixture of sulfoxides 11 and 12 in a 1.3:1 ratio,respectively. This material was subsequently purified via column chromatography (50-75% ethylacetate in hexanes) to afford 34 mg (yield - 21.3%) of a faster moving sulfoxide (12) and 19 mg (yield = 11.9%) of a slower moving sulfoxide (11) . Both sulfoxides were recrystallized from ethanol. However, the faster moving sulfoxide crystals were amenable to X-ray analysis, allowing assignment of the sulfoxide stereochemistry.

In view of the data presented above, it can be concluded that the compounds of the present invention are useful as pharmaceutical preparations that temporarily inhibit testosterone biosynthesis in a way beneficial to patients with testosterone or estradiol-related pathologies. Thus, the present invention also provides pharmaceutical compositions comprising a compound according to the present invention as defined above as an active ingredient together with a pharmaceutically acceptable carrier or excipient.

Compositions according to the present invention may be administered parenterally in combination with conventional injectable liquid carriers such as sterile pyrogen-free water, sterile peroxide-free ethyl oleate, dehydrated alcohol, or propylene glycol. Conventional

pharmaceutical adjuvants for injection solution such as stabilizing agent, solubilizing agents and buffers, such as ethanol, complex forming agents such as ethylene diamine tetraacetic acid, tartrate and citrate buffers, and high-molecular weight polymers such as polyethylene oxide for viscosity regulation may be added. Such compositions may be injected intramuscularly, intraperitoneally, or intravenously.

Further non-limiting examples of carriers and diluents include albumin and/or other plasma protein components such as low density lipoproteins, high density lipoproteins and the lipids with which these serum proteins are associated. These lipids include phosphatidyl choline, phosphatidyl serine, phosphatidyl ethanolamine and neutral lipids such as triglycerides. Lipid carriers also include, without limitation, tocopherol and retinoic acid. Additional lipid and lipoprotein drug delivery systems that may be included herein are described more fully in "Biological Approaches to Controlled Delivery of Drugs," Annals of the New York Academy of Sciences. 507. 775-88, 98-103, and 252-271, which disclosure is hereby incorporated by reference.

The compositions may also be formulated into orally administrable compositions containing one or more physiologically compatible carriers or excipients, and may be solid or liquid in form. These compositions may, if desired, contain conventional ingredients such as binding agents, for example, syrups, acacia, gelatin, sorbitol, tragacanth, or polyvinylpyrrolidone; fillers, such as lactose, mannitol, starch, calcium phosphate, sorbitol, cyclodextran, or methylcellulose; lubricants such as magnesium stearate, high molecular weight polymers such as polyethylene glycols, high molecular weight fatty acids such as stearic acid or silica; disintegrants such as starch; acceptable wetting agents as, for example, sodium lauryl sulfate.

The oral compositions may assume any convenient

form, such as tablets, capsules, lozenges, aqueous or oily suspensions, emulsions, or dry products suitable for reconstitution with water or other liquid medium prior to use. The liquid oral forms may, of course, contain flavors, sweeteners, preservatives such as methyl or propyl p-hydroxybenzoates; suspending agents such as sorbitol, glucose or other sugar syrup, methyl, hydroxymethyl, or carboxymethyl celluloses or gelatin; emulsifying agents such as lecithin or sorbitan monooleate or thickening agents. Non-aqueous compositions may also be formulated which comprise edible oils as, for example, fish-liver or vegetable oils. These liquid compositions may conveniently be encapsulated in, for example, gelatin capsules in a unit dosage amount. The pharmaceutical compositions according to the present invention may also be administered, if appropriate, either topically as an aerosol or, formulated with conventional bases as a cream or ointment.

The pharmaceutical compositions of the present invention can also be administered by incorporating the active ingredient into colloidal carriers, such as liposomes. Liposome technology is well known in the art, having been described by Allison et al. in Nature 252: 252-254 (1974) and Dancy et al., J. Immunol. 120: 1109- 1113 (1978) .

Alternatively, active-targeting vesicles can be used as carriers for the active components of the present invention by placing a recognition sequence, i.e., from an antibody, onto the vesicles such that it is taken up more rapidly by certain cell types, such as cancer cells (cf. Papahadjopoulou et al., Annals of the New York Academy of Sciences. 507: 67-74 (1987)). As further embodiments of the present invention, the active components can be administered in the form of sustained release products, by incorporating the active components in a suitable polymer. Polymer systems found to be useful in delivering steroids in a time-dependent manner are known in the art and

include subdermal reservoir implants composed of nondegradable polymers that release steroids for extended periods of time (Folkman et al., J. Surσ. Res. 4.: 139 (1964)) and subdermal implants or injectable microspheres comprising degradable materials, such as lactide-glycolide copolymers and polycaprolactones (Langer, Science 249: 1527-1532 (1990)).

It will be understood by the skilled practitioner that the compositions of the present invention may be administered in conjunction with, as well as formulated with, at least one other therapeutic agent to produce a combination composition and/or therapy effective for ameliorating certain androgen-dependent pathologies, i.e., prostate cancer and benign prostatic hyperplasia; or estrogen dependent pathologies such as uterine, breast or ovarian cancer. "In conjunction" is defined herein to mean the present compositions may be administered first and other chemotherapeutic agents later, or vice versa. Therapeutic agents include, by way of non-limiting examples, LHRH agonists/antagonists such as leuprolide acetate and androgen receptor antagonists such as cyproterone acetate and flutamide.

A particular aspect of the present invention comprises a compound of the present invention in an effective unit dose form. By "effective unit dose" is meant a predetermined amount sufficient to bring about the desired inhibitory effect, which can be readily determined by one skilled in the art.

The dosage of the compounds of the present invention or their pharmaceutically acceptable salts or derivatives will depend, of course, on the degree of testosterone (or estrogen) inhibition desired. Dosages of pharmaceutically active compounds such as those disclosed in the present invention are conventionally given in amounts sufficient to bring about the desired testosterone (or estrogen) inhibition relative to the condition being treated (prostate cancer, benign prostatic hyperplasia.

precocious asculization, testicular cancer, testicular Leydig cell tumors; ovarian, breast or uterine cancer etc.) without causing undue burden upon the host.

The foregoing description of the specific embodiments will so fully reveal the general nature of the invention that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept, and therefore such adaptations and modifications are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation.