Article pubs.acs.org/jnp
Synthesis and Bioconversions of Formestane Glenroy D. A. Martin,*,† Javier Narvaez,† and Anne Marti‡ †
Chemistry, Biochemistry and Physics Department, The University of Tampa, 401 West Kennedy Boulevard, Tampa, Florida 33606-1490, United States ‡ Department of Chemistry, The University of Texas at Dallas, Richardson, Texas 75083-0688, United States S Supporting Information *
ABSTRACT: In an effort to generate new steroidal aromatase inhibitors, formestane (4-hydroxyandrost-4-ene-3,17-dione) (1) was biotransformed by Rhizopus oryzae to yield the known 4β,5α-dihydroxyandrostane-3,17-dione as the major product (5) and bioconverted by Beauveria bassiana to afford the known reduced 4,17β-dihydroxyandrost-4-en-3-one (6) and 3α,17β-dihydroxy5β-androstan-4-one (7) and the new 4,11α,17β-trihydroxyandrost-4-en-3-one (8). All the metabolites showed more potent activities than their parent congener in the aromatase and MCF-7 breast cancer assays. The bioactivities and structural elucidation of these metabolites as well as the semisynthesis of formestane (1) from testosterone (2) are reported herein.
F
ormestane (1), like exemestane, is a steroidal aromatase inhibitor that is used in the treatment of breast and ovarian cancers in postmenopausal women. It suppresses the production of estrogen by inhibiting the aromatase enzyme that converts androgens into estrogens.1 Other nonsteroidal aromatase inhibitors such as anastrazole and letrozole compete with the substrate for binding to the enzyme active site.2,3 Despite these advances in the development of aromatase inhibitors, there are still growing concerns of major side effects and possibly cellular resistance to these drugs over time. The search for better aromatase inhibitors with fewer side effects has led to the synthesis of other compounds.4 The elucidation of an aromatase active site5 has also aided in structure-guided design6 of molecules with more favorable enzyme−substrate interaction within the enzyme’s active site. The results of these efforts have been met with limited success. Of the aforementioned aromatase inhibitors, our group is focused on improving the efficacy and low bioavailability1−3,7 of formestane (1). As such, analogues8,9 of formestane (1) were prepared by microbial means with the prolific steroid transformers Beauveria bassiana10,11 and Rhizopus oryzae,12 in the hopes of finding new aromatase inhibitors with enhanced biological activities. An improved synthetic route from testosterone (2) to formestane (1) is included.
formestane (1) was confirmed by comparison of its mass spectral ([M + H]+ = 303.42) and 1H and 13C NMR data (Figures S1 and S2) with those of an authentic standard. The above synthesis utilizes an expansion of Ortar’s method15 to provide an alternate route from what was previously described in the literature.16 With the synthesis of formestane (1) completed we turned our attention toward its biotransformation by R. oryzae and B. bassiana. Bioconversion of formestane (1) with R. oryzae yielded three potential metabolites based on thin layer chromatography (TLC) comparison with the control extracts, but only one metabolite was present in sufficient quantity for characterization. Mass spectral (MS) data ([M + H]+ = 321.42)
■
RESULTS AND DISCUSSION Formestane (1) was synthesized from testosterone (2) in a facile three-step synthesis to afford an overall 23% yield (Scheme 1). The oxidation of testosterone (2) with Jones reagent13 afforded androst-4-ene-3,17-dione (3).14 Hydroxylation of 3 with OsO4/H2O2 followed by alkaline dehydration of the resultant diols 4 gave formestane (1).15 The identity of © 2013 American Chemical Society and American Society of Pharmacognosy
Received: July 17, 2013 Published: September 27, 2013 1966
dx.doi.org/10.1021/np400585t | J. Nat. Prod. 2013, 76, 1966−1969
Journal of Natural Products
Article
Scheme 1. Synthesis of Formestane (1) from Testosterone (2)
presence of carbonyl (1668 cm−1), olefinic (1619 cm−1), and hydroxyl (3354 cm−1) moieties. Comparison of 1H and 13C NMR spectroscopic data of formestane (1) and 4-hydroxytestosterone (6) with the 1D and 2D NMR (1H−1H COSY, HMQC, and HMBC) data of compound 8 (Figures S9−S14) confirmed the reduction of the C-17 carbonyl group and the hydroxylation at C-11. New proton signals were observed for H-11 (δH 4.02, dt, J = 10.32, 4.82) and H-17 (δH 3.69, t, J = 8.6). The HMBC spectrum also showed 3J couplings from H-11 (δH 4.02) to C-9 (δC 59.5), C10 (δC 39.2), and C-12 (δC 48.8) as well as correlations from H-17 (δH 3.69) to C-12 (δC 48.8) and C-18 (δC 12.2) (Figure 2 and Table 1).
of the metabolite (5) revealed the presence of a second hydroxyl group. IR data also revealed the disappearance of the C4−C5 double bond. Comparison of the 1H and 13C NMR data of this metabolite (Figures S3 and S4) with those of formestane (1) and literature data confirmed the product as 4β,5α-dihydroxyandrostane-3,17-dione (5) (Figure 1).17
Figure 2. Key 1H−1H COSY (bonds) and HMBC (arrows) correlations for compound 8. Figure 1. Bioconversion reactions of formestane (1) by R. oryzae and B. bassiana.
Table 1. 1H (300 MHz) and 13C (75 MHz) NMR Assignments of Compound 8 in CDCl3 The biotransformation of formestane (1) by B. bassiana afforded three metabolites. Purification on silica gel with increasing ethyl acetate in hexane yielded compounds 6−8 (Figure 1). MS data ([M + H]+ = 305.42) of metabolite 6 suggested that reduction had occurred with an increase of two atomic mass units (amu) when compared to formestane (1). IR and NMR data confirmed the reduction of the C-17 carbonyl to yield 4-hydroxytestosterone (6). The spectroscopic data of 6 (Figures S5 and S6) were consistent with those in the literature.8 Metabolite 7 was the product of further reduction, as indicated from the [M + H]+ with m/z of 307.33. This was established in the FTIR spectrum with the absence of the C4− C5 double bond. In addition, 1H and 13C NMR data also indicated that isomerization of the carbonyl and hydroxyl groups at C-3 and C-4 had occurred. The spectroscopic data of 7 (Figures S7 and S8) agreed well with the literature, and the metabolite was confirmed as 3α,17β-dihydroxy-5β-androstan-4one (7).8 HRMS (ESI) data ([M − H]− = 319.1912) of the most polar compound, 8, implied a molecular formula of C19H28O4. It is the product of monohydroxylation along with the reduction of one of its carbonyl groups. The IR spectrum suggested the
position 1 2 3 4 5 6b 7 8 9 10 11 12 13 14 15b 16 17 18 19
δC, type 36.3, 31.9, 194.3, 140.2, 141.6, 23.5, 30.2, 34.8, 59.5, 39.2, 69.1, 48.8, 43.5, 49.8, 23.4, 30.5, 81.1, 12.2, 18.4,
CH2 CH2 C C C CH2 CH2 CH CH C CH CH2 C CH CH2 CH2 CH CH3 CH3
δH, mult (J in Hz) 2.61, m; 1.97, m 2.58, m; 2.44, m
HMBC (H → C)a 2, 10, 19, 1, 10
6.15, br s (OH) 3.05, d (14.7); 1.94, d (5.16) 1.24, m; 0.99 d (7.15) 1.53, m 1.07, m
7, 8
4.02, dt (10.32, 4.82) 2.11, m; 1.15, m
9, 10, 12 9, 11, 13, 14, 17, 18
1.11, m 1.63, m; 1.29, m 1.83 m; 1.79, m 3.69, t (8.6) 0.82 1.31
8, 11, 13, 17 8
15, 19 1, 7, 8, 10, 11, 19
12, 18 12, 13, 14, 17 1, 9, 10
a
HMBC correlations are from proton(s) stated to the indicated carbon(s). bSignals were not distinguishable.
1967
dx.doi.org/10.1021/np400585t | J. Nat. Prod. 2013, 76, 1966−1969
Journal of Natural Products
Article
breast cancer patients.21 The aforementioned compounds 5−7 in addition to other reported phase I and phase II metabolites8,20,21 were seen in human eukaryote metabolism studies of formestane (1). In summary, a new bioactive metabolite, 4,11α,17βtrihydroxyandrost-4-en-3-one (8), was obtained from the microbial transformation of formestane (1). The range of metabolites obtained from our two fermentations were from fungi (R. oryzae and B. bassiana) possessing different cytochrome P-450 enzymes. However, their combined metabolic profiles mimic those observed in other organisms. The knowledge derived from fungal metabolism studies continues to serve as a powerful tool in research, as it provides useful information about phase I metabolites and therefore aids in the development of future drugs.
The large coupling constant of H-11 (J = 10.3 Hz) and NOESY 1D correlations (Figure S15) with the H-18 (δH 0.82) and H-19 (δH 1.31) methyl groups and H-12β (δH 2.11) both implied that H-11 (δH 4.02) was axial with β-stereochemistry (Figure 3). The stereochemistry of the hydroxyl group was
■
Figure 3. Key NOE correlations for metabolite 8.
therefore assigned as C-11αOH. The stereochemistry of the second hydroxyl group was assigned as C-17βOH based on the lack of NOE correlation between H-17 and the H-18β methyl group. This new metabolite was therefore assigned as 4,11αdihydroxytestosterone (8). Since formestane (1) is a known aromatase inhibitor and is active against breast cancer cells (MCF-7),15 compound 1 and its analogues 5−8 were evaluated for their potential as antiproliferative agents against MCF-7 cells in culture as well as aromatase inhibition.18 All the metabolites were active in both assays and were more potent than the parent congener (Table 2). Their cytotoxicities toward MCF-7 cells revealed Table 2. Results of the Cytotoxic and Aromatase Inhibition Assays of Formestane (1) and Its Metabolites (5−8) IC50, μM compound 1 5 6 7 8 vinblastine naringenin
MCF-7 20.0 15.0 16.5 16.4 12.5 10.9
± ± ± ± ± ±
1.1 3.8 1.1 1.5 2.4 2.4 nM
aromatase 58.6 29.1 47.0 3.9 10.9
± ± ± ± ±
EXPERIMENTAL SECTION
General Experimental Procedures. Melting points were determined on a Digimelt melting point apparatus and are uncorrected. Infrared spectra were recorded on a Thermo Fisher Nicolet iS10 FT-IR spectrometer. Optical rotations were performed on a Rudolph Research Analytical Autopol IV automatic polarimeter. Ultraviolet−visible measurements were conducted on an Agilent Technologies Cary 60 UV−vis instrument. 1H and 13C NMR data were obtained on a JEOL ECS 300 MHz NMR instrument. Deuterated chloroform (CDCl 3 ) was used as solvent with tetramethylsilane (TMS) as internal standard. HRMS data were determined on an Agilent 6210 TOF-MS, and LRMS data were obtained on a Thermo Fisher LTQ-XL. Column chromatography was performed on silica gel (37−63 μm dia.). Detection of compounds on thin layer chromatography was achieved by spraying the plates with phosphomolybdic acid solution followed by heating until the color developed. Rhizopus oryzae ATCC 11145 and Beauveria bassiana ATCC 7159 were obtained from the American Type Culture Collection (ATCC), Rockville, MD, USA. Culture Conditions. R. oryzae ATCC 11145 and B. bassiana ATCC 7159 were maintained on potato dextrose agar slants at 28 °C for two weeks. For each fungus, 10 slants (4 days old) were used to inoculate ten 500 mL Erlenmeyer flasks, each containing 250 mL of liquid culture medium. The liquid medium (2.5 L) for R. oryzae contained glucose (20 g/L), yeast extract (5 g/L), sodium chloride (5 g/L), and dipotassium hydrogen phosphate (5 g/L). The flasks were shaken at 180 rpm at 27 °C. The liquid medium (2.5 L) for B. bassiana consisted of potato dextrose broth (24 g/L). Formestane (1) (1 g), dissolved in acetone, was pulse fed to the growing fungus (nine flasks) in portions of 10, 20, 30, and 40% at 24, 36, 48, and 60 h, respectively, after inoculation. One flask served as the control, in which no substrate was added. The fermentation was allowed to proceed for 5 days. The fermentation beer was pooled, and the fungal cells were separated from the broth via vacuum filtration. The broth was extracted with ethyl acetate (2 × 750 mL), and the fungal cells were homogenized and extracted in warm ethyl acetate (1000 mL). The organic solutions from the broth and fungal cells were dried separately with anhydrous sodium sulfate, filtered, and concentrated in vacuo to afford crude extracts. Semisynthesis of Formestane (1). Testosterone (2) (18.1 g, 0.0627 mol) was oxidized to androst-4-ene-3,17-dione (3) (13.2 g, 73%) according to the previously reported Jones oxidation method.13 To a solution of androst-4-ene-3,17-dione (3) (1.4 g, 4.9 mmol) in tBuOH (50 mL) was added a solution of osmium tetroxide (40 mg, 0.16 mmol) in t-BuOH (2 mL) and 30% H2O2 (7.5 mL). The resulting solution was allowed to stir at room temperature for 3 days, diluted with brine (100 mL), and extracted with CH2Cl2 (2 × 100 mL). The organic phase was washed with brine (100 mL), 10% NaHSO3 solution (50 mL), 10% Na2CO3 solution (50 mL), and brine (100 mL). The organic solution was dried with anhydrous Na2SO4 and evaporated in vacuo to afford a crude residue (1.96 g). To the residue dissolved in MeOH (10 mL) was added a solution of KOH (393 mg, 7
6.3 0.9 4.9 1.3 0.6
3.3 ± 0.2
IC50 values ranging from 12.5 to 15.5 μM compared to formestane (IC50 of 20 μM). Metabolite 8 was shown to be the most potent of all, with an IC50 of 12.5 μM. The aromatase assay18 showed IC50 values of metabolite inhibition ranging from 3.9 to 47 μM. Compound 7 exhibited potency (3.9 μM) similar to that of the positive control naringenin (3.3 μM), with metabolite 8 being slightly less active (10.9 μM). The range of hydroxylase, oxido-reductase, and isomerase activities observed in the above fungal transformation reactions suggest that microbes possess similar cytochrome P-450 enzyme activities to those observed in animals and human eukaryotes. This is consistent with the early findings of Smith and Rosazza in that microbial biostransformation reactions function as models for mammalian metabolism.19 The 4β,5αdihydroxy metabolite (5) obtained in our R. oryzae fermentation, for example, was also observed in studies with rat hepatocytes and separately in breast and prostatic cancer patients.20,21 Likewise, metabolites 6 and 7 from our B. bassiana fermentation were also noted in formestane (1) metabolism studies in male rhesus monkeys, female rats,20 and human 1968
dx.doi.org/10.1021/np400585t | J. Nat. Prod. 2013, 76, 1966−1969
Journal of Natural Products
Article
mmol) in MeOH (3 mL), and the solution was stirred at 55 °C for 10 min. Acetic acid (0.3 mL) was added, and the reaction mixture was diluted with brine (100 mL) and extracted with CH2Cl2 (2 × 100 mL). The combined organic phases were washed with brine (100 mL), dried, and concentrated to yield a dark brown residue. The residue (1.25 g) was purified on silica gel with elution in 20% ethyl acetate in hexane to afford formestane (1) (0.452 g, 31%) [mp 195−200 °C (lit.15 199−202 °C)]. See Figures S1 and S2 for 1H and 13C NMR spectra, respectively, of semisynthetic formestane (1). Bioconversion with R. oryzae. Formestane (1) (1 g) was pulse fed to R. oryzae. Workup yielded broth and mycelial extracts, which were pooled and purified on silica gel. Elution in 25% ethyl acetate in hexane afforded 4β,5α-dihydroxyandrostane-3,17-dione (5) (86 mg) [mp 165−166 °C (lit.17 168−170 °C)]. See Figures S3 and S4 for 1H and 13C NMR spectra, respectively, of metabolite 5. Biotransformation with B. bassiana. The bioconversion of formestane (1) (1 g) with B. bassiana provided mycelial and broth extracts that were combined and chromatographed on silica gel. Elution in 20% ethyl acetate in hexane yielded untransformed formestane (1) (95 mg), while elution in 25% ethyl acetate in hexane afforded 4,17β-dihydroxyandrost-4-en-3-one (6) (53 mg) [mp 199− 204 °C (lit.8 205−210 °C)]. The metabolite 3α,17β-dihydroxy-5βandrostan-4-one (7) (9 mg) [mp 180−183 °C (lit.8 168−170 °C)] was obtained in 30% ethyl acetate in hexane, and elution in 50% ethyl acetate in hexane provided 4,11α,17β-trihydroxyandrost-4-en-3-one (8) (0.24 g). See Figures S5−S15 for NMR spectra of compounds 6, 7, and 8. 4,11α,17β-Trihydroxyandrost-4-en-3-one (8): amorphous solid; [α]20D +57.5 (c 0.73, CHCl3); UV (EtOH) λmax (log ε) 280 (1.20) nm; IR νmax 3350, 1648, 1614, 1386, 1159 cm−1; 1H NMR (CDCl3, 300 MHz) and 13C NMR (CDCl3, 75 MHz), see Table 1; ESIMS m/z 321 [M + H]+ (100), 239 (28), 228 (67), 220 (24); HRMS(ESI) m/z 319.1912 [M − H]− (calcd for C19H27O4, 319.1915). SRB Assay. The cytotoxic potential of formestane (1) and its metabolites (5−8) toward MCF-7 cancer cells was determined using the previously described SRB assay.18 These experiments were performed in triplicate. Test compounds (dissolved in DMSO) were transferred to 96-well plates and incubated in a CO2 incubator for 72 h at 37 °C. The incubation was terminated with trichloroacetic acid. The cells were washed, air-dried, and stained with SRB solution, and optical densities were determined at 515 nm using a microplate reader. In each case, a zero-day control was performed by adding an equivalent number of cells to several wells, incubating at 37 °C for 30 min, and processing as described above. Percent of cell survival was calculated using the formula (ODcells+tested compound − ODday0)/(ODcells+10%DMSO − ODday0) × 100. Vinblastin was used as the positive control.18 Aromatase Assay. Compounds 1 and 5−8 were evaluated for aromatase inhibition according to a previously established protocol.18
■
Prof. J. Pezzuto and T. Kondratyuk (University of Hawaii at Hilo) for conducting cytotoxic and aromastase inhibition assays. We are grateful to S. Shaw for technical assistance. Dr. E. Ballard is thanked for helpful discussions.
■
ASSOCIATED CONTENT
S Supporting Information *
The NMR spectra of compounds 1 and 5−8 are available free of charge via the Internet at http://pubs.acs.org.
■
REFERENCES
(1) Wiseman, L. R.; Goa, K. L. Drugs Aging 1996, 9, 292−306. (2) Varela, C.; Tavares da Silva, E. J.; Amaral, C.; Correia da Silva, G.; Baptista, T.; Alcaro, S.; Costa, G.; Carvalho, R. A.; Teixeira, N. A. A.; Roleira, F. M. F. J. Med. Chem. 2012, 55, 3992−4002. (3) Dutta, U.; Pant, K. Med. Oncol. 2008, 25, 113−124. (4) Pokhrel, M.; Ma, E. Molecules 2011, 16, 9868−9885. (5) Ghosh, D.; Griswold, J.; Erman, M.; Pangborn, W. Nature 2009, 457, 219−223. (6) Ghosh, D.; Lo, J.; Morton, D.; Valette, D.; Xi, J.; Griswold, J.; Hubbell, S.; Egbuta, C.; Jiang, W.; An, J.; Davies, H. M. L. J. Med. Chem. 2012, 55, 8464−8476. (7) Joseph, J. K.; Lim, A. K. Med. J. Malaysia 1998, 53, 37−41. (8) Kohler, M.; Parr, M. K.; Opfermann, G.; Thevis, M.; Schlorer, N.; Marner, F. J.; Schanzer, W. Steroids 2007, 72, 278−286. (9) Lonning, P. E.; Geisler, J.; Johannessen, D. C.; Gschwind, H. P.; Waldmeier, F.; Schneider, W.; Galli, B.; Winkler, T.; Blum, W.; Kriemler, H. P.; Miller, W. R.; Faigle, J. W. J. Steroid Biochem. Mol. Biol. 2001, 77, 39−47. (10) Huszcza, E.; Dmochowska-Gładysz, J.; Bartmańska, A. Z. Naturforsch. 2005, 60c, 103−108. (11) Zhan, J.; Gunatilaka, A. A. L. In Advances in Fungal Biotechnology; Rai, M., Ed.; I.K. International: New Delhi, 2009; Chapter 8, pp 226−249. (12) Martin, G. D. A. Curr. Org. Chem. 2010, 14, 1−14. (13) Martin, G. D. A.; Reynolds, W. F.; Reese, P. B. Phytochemistry 2004, 65, 701−710. (14) Huang, C.-L.; Chen, Y.-R.; Liu, W.-H. Enzyme Microb. Tech. 2006, 39, 296−300. (15) Ciattini, P. G.; Morera, E.; Ortar, G. Synth. Commun. 1992, 22, 1949−1952. (16) Campos Neves, A. S.; Sá e Melo, M. L.; Moreno, M. J. S. M.; Tavares da Silva, E. J.; Salvador, J. A. R.; Pereira da Costa, S. C.; Martins, R. M. L. Tetrahedron 1999, 55, 3255−3264. (17) Foster, A. B.; Jarman, M.; Mann, J.; Parr, I. B. J. Steroid Biochem. 1986, 24, 607−617. (18) Schupp, P. J.; Kohlert-Schupp, C.; Whitefield, S.; Engemann, A.; Rohde, S.; Hemscheidt, T.; Pezzuto, J. M.; Kondratyuk, T. P.; Park, E. J.; Marler, L.; Rostama, B.; Wright, A. D. Nat. Prod. Commun. 2009, 4, 1717−1728. (19) Azerad, R. Adv. Biochem. Eng./Biotechnol. 1999, 63, 169−218 and references therein. (20) Poon, G. K.; Jarman, M.; Rowlands, M. G.; Dowsett, M.; Firth, J. J Chromatogr. B 1991, 565, 75−88. (21) Poon, G. K.; Jarman, M.; McCague, R.; Davies, J. H.; Heeremans, C. E. M.; Van der Hoeven, R. A. M.; Niessen, W. M. A.; van der Greef, J. J. Chromatogr. B 1992, 576, 235−244.
AUTHOR INFORMATION
Corresponding Author
*Tel: 813 257 1805. Fax: 813 258 7496. E-mail: gdmartin@ut. edu. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was partly funded by a grant provided by the Biological Chemistry Summer Research Program (Merck/ AAAS), the American Chemical Society (ACS) Project SEED endowment, the Hillsborough County Public Schools Academic Programs, and the ACS Tampa Bay local section. We thank 1969
dx.doi.org/10.1021/np400585t | J. Nat. Prod. 2013, 76, 1966−1969
