Received 29 February 2012
Two androsterone derivatives as inhibitors of androgen biosynthesis
The title compounds, (3R,5S,5'R,8R,9S,10S,13S,14S)-10,13-dimethyl-5'-(2-methylpropyl)tetradecahydro-6'H-spiro[cyclopenta[a]phenanthrene-3,2'-[1,4]oxazinane]-6',17(2H)-dione, C26H41NO3, (I), and methyl (2R)-2-[(3R,5S,8R,9S,10S,13S,14S)-10,13-dimethyl-2',17-dioxohexadecahydro-3'H-spiro[cyclopenta[a]phenanthrene-3,5'-[1,3]oxazolidin-3'-yl]]-4-methylpentanoate, C28H43NO5, (II), possess the typical steroid shape (A-D rings), but they differ in their extra E ring. The azalactone E ring in (I) shows a half-chair conformation, while the carbamate E ring of (II) is planar. The orientation of the E-ring substituent is clearly established and allows a rationalization of the biological results obtained with such androsterone derivatives.
17-Hydroxysteroid dehydrogenase type 3 (17-HSD3) is an enzyme that catalyzes the reduction of 4-androstene-3,17-dione into the androgen testosterone (Poirier, 2010; Mohler et al., 2007; Maltais et al., 2011). This latter steroid hormone, as well as its natural metabolite dehydrotestosterone, is known to stimulate the proliferation of prostate cancer cells; the Scheme shows the biosynthesis of the androgenic hormone testosterone and a representation of azalactones (I) and (III) and carbamates (II) and (IV) tested as inhibitors of 17-HSD3 (Inh = inhibition) (Poirier, 2008). Since 17-HSD3 contributes to the production of androgens in men, inhibiting this enzyme is an interesting strategy to block androgen biosynthesis and to treat prostate cancer. Androsterone (ADT) derivatives substituted at position C3 were previously reported as inhibitors of 17-HSD3 and it was established that a hydrophobic group is required to obtain good inhibition of 17-HSD3 (Maltais et al., 2002), a -oriented group producing a better inhibition than an -oriented group (Tchedam-Ngatcha et al., 2005).
In our synthetic search for inhibitors of 17-HSD3, we recently focused on the preparation of new ADT derivatives having an extra azalactone or carbamate E ring. For a better understanding of the structure-activity relationship (SAR) ofand (II) (see Scheme and Fig. 1).
The two compounds show fixed conformations of their extended side chains; no disorder is apparent in the structure analyses. Packing diagrams with van der Waals radii give the impression that these structures are not very tightly packed, in agreement with the low densities [1.162 Mg m-3 in (I) and 1.200 Mg m-3 in (II)]. The packing appears to be efficient enough, however, to be linked to the appearance of just one conformation of the side chain in each case.
The correct enantiomers, already known from the starting materials, were checked against the absolute structure parameters (Flack, 1983). For compound (I), the absolute structure parameter was found to be 0.12 (18) for 1747 Friedel pairs (83% Friedel coverage). For compound (II), it was 0.06 (18) for 2087 (100%) Friedel pairs. These two values are better than might have been expected considering that the highest anomalous contribution comes from O atoms (Flack & Bernardinelli, 2008). A Bayesian statistics analysis on Bijvoet differences was also performed using the program PLATON (Spek, 2009). The Hooft parameters obtained (Hooft et al., 2008), viz. -0.04 (7) and -0.05 (6) for compounds (I) and (II), respectively, are also in very good agreement with the known molecular absolute configurations, and were perhaps influenced positively by the use of copper radiation. Probability levels of having a false attribution, P3(false), are essentially zero for both compounds.
The steroid six-membered A, B and C rings appear as the well known fused-ring system, all in chair conformations, while the five-membered D ring is in a slightly twisted envelope conformation in both azalactone (I) and carbamate (II) (Yan et al., 2009) (Fig. 2). The heterocyclic ring of (I) is a half-chair perpendicular to the steroid backbone, while the heterocyclic ring of carbamate (II) is almost planar [maximum deviation = 0.147 (1) Å] and is perpendicular to the steroidal plane (Fig. 3). These extra heterocyclic E rings provide a restricted orientation to the isobutyl group in the azalactones (I) and (III) (see Scheme). In carbamates (II) and (IV), the methyl pentanoate group has a higher degree of liberty owing to the free rotation around the exocyclic bond to the N atom. In azalactone (I), the hydrophobic side chain has a distal orientation with respect to the longitudinal axis of the four-ring steroid system (to the right in Fig. 4a). Consequently, the isobutyl group of the other isomer, the azalactone (III), would be oriented to the left in a similar drawing, or in other words in a proximal disposition to the centerline of the four-ring system. Since the present compound (I) is a less potent inhibitor than its stereoisomer (III), it appears that the side chain must be proximal (on the left side in Fig. 4a) for better enzyme inhibition. The isobutyl group would thus point toward a hypothetical hydrophobic pocket of 17-HSD3, which was previously deduced from SAR studies (Tchedam-Ngatcha et al., 2005; Maltais et al., 2011) but not confirmed because this membrane enzyme was never crystallized. In the case of carbamates (II) (Fig. 4b) and (IV), the inhibitory activity is the same (55 and 58% of inhibition at 0.1 µM) for both compounds due to the free rotation of the methyl pentanoate group, which promotes interaction with a hydrophobic pocket of the enzyme.
| || Figure 1 |
The molecular structures of (a) azalactone (I) and (b) carbamate (II), with non-H atoms represented by 50% probability displacement ellipsoids.
| || Figure 2 |
The structures of (a) (I) and (b) (II), showing the conformations of the steroid A, B, C and D rings.
| || Figure 3 |
The E-ring conformation in (a) (I) and (b) (II).
| || Figure 4 |
The E-ring substituent conformation in (a) azalactone (I) and (b) carbamate (II). The isobutyl group of (I) is oriented away from the longitudinal axis of the steroid ring system, while the methyl pentanoate group of (II) has free rotation.
Compounds (I) and (III) were synthesized by modification of an approach developed in our laboratory (Rouillard et al., 2008). Briefly, aminolysis of the oxirane at C3 of androsterone with L- and D-isoleucine methyl esters provided the corresponding amino alcohols, which were treated with sodium methoxide to afford azalactones (I) and (III). Compounds (II) and (IV) were prepared according to the method of Maltais et al. (2002) by treatment of each amino alcohol with triphosgene. X-ray quality crystals of (I) and (II) were obtained by slow diffusion of hexane into a methylene chloride solution of the compound (hexane-methylene chloride 1:1 v/v) followed by slow evaporation.
H atoms were treated as riding, with C-H = 1.00 (methyl), 0.99 (methylene) and 0.98 Å (remaining H atoms), and with Uiso(H) = 1.5Ueq(C) for methyl H atoms and 1.2Ueq(C) otherwise. The N-bound H atom of (I) was located by difference Fourier synthesis; its coordinates were refined and its Uiso(H) value was constrained to 1.2Ueq(N).
Data collection: APEX2 (Bruker, 2011); cell refinement: SAINT (Bruker, 2011); data reduction: XPREP (Bruker, 2008); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: SHELXTL (Sheldrick, 2008); software used to prepare material for publication: UdMX (Maris, 2004).
Supplementary data for this paper are available from the IUCr electronic archives (Reference: FA3273 ). Services for accessing these data are described at the back of the journal.
The authors thank the Canadian Institutes of Health Research (CIHR) for financial support. LCK thanks the Faculty of Medicine (Laval University) for a fellowship.
Bruker (2008). XPREP. Bruker AXS Inc., Madison, Wisconsin, USA.
Bruker (2011). APEX2 and SAINT. Bruker AXS Inc., Madison, Wisconsin, USA.
Flack, H. D. (1983). Acta Cryst. A39, 876-881.
Flack, H. D. & Bernardinelli, G. (2008). Chirality, 20, 681-690.
Hooft, R. W. W., Straver, L. H. & Spek, A. L. (2008). J. Appl. Cryst. 41, 96-103.
Maltais, R., Fournier, M. A. & Poirier, D. (2011). Bioorg. Med. Chem. 19, 4652-4668.
Maltais, R., Luu-The, V. & Poirier, D. (2002). J. Med. Chem. 45, 640-653.
Maris, T. (2004). UdMX. University of Montréal, Montréal, QC, Canada.
Mohler, M. L., Narayanan, R., He, Y., Miller, D. D. & Dalton, J. T. (2007). Recent Pat. Endocr. Metab. Immune Drug Discov. 1, 103-118.
Poirier, D. (2008). Drug Dev. Res. 69, 304-318.
Poirier, D. (2010). Expert Opin. Ther. Pat. 20, 1123-1145.
Rouillard, F., Roy, J. & Poirier, D. (2008). Eur. J. Org. Chem. 14, 2446-2453.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.
Spek, A. L. (2009). Acta Cryst. D65, 148-155.
Tchedam-Ngatcha, B., Luu-The, V., Labrie, F. & Poirier, D. (2005). J. Med. Chem. 48, 5257-5268.
Yan, X., Xu, S., Wang, J., Chen, Y. & Xia, P. (2009). Acta Cryst. E65, o587.