Two syntheses of FF-MAS

Article abstract of DOI:10.1021/ol0343156

(Matrix presented) Follicular fluid-meiosis activating sterol (FF-MAS) has been shown to be an efficient inducer of meiotic maturation. It can potentially be used for improvements of in vitro fertilization techniques. Two short synthesis of FF-MAS are pre

Full text of DOI:10.1021/ol0343156

ORGANIC
LETTERS
2003
Vol. 5, No. 11
1837-1839
Two Syntheses of FF-MAS
Thorsten Blume,* Marc Guttzeit, Joachim Kuhnke, and Ludwig Zorn
Schering AG, Corporate Research, Research Center Europe, Medicinal Chemistry,
D-13342 Berlin, Germany
thorsten.blume@schering.de
Received February 21, 2003
ABSTRACT
Follicular fluid-meiosis activating sterol (FF-MAS) has been shown to be an efficient inducer of meiotic maturation. It can potentially be used
for improvements of in vitro fertilization techniques. Two short synthesis of FF-MAS are presented in this article. Both syntheses are based
on microbiological degradations of sterol side chains. FF-MAS can be synthesized in nine steps from commercially available starting materials
by both routes.
In 1995, Byskov et al. discovered FF-MAS 1 (follicular fluid-
meiosis activating sterol: 4,4-dimethyl-5R-cholesta-8,14,-
24-trien-3â-ol) in the follicular fluid of women.1 It has been
demonstrated that FF-MAS and analogous sterols, which can
reach physiological concentrations in the gonads in the µM
range,2 can activate the nuclear maturation of oocytes in vitro.
Furthermore, it has been shown that FF-MAS markedly
improves the quality of the mature mouse oocyte, leading
to significantly higher fertilization rates in vitro.3 Because
of its potential to be used as a fertility promoter, the
compound has gained considerable attention in the recent
literature.4
oocytes show clear potential toward improvements of in vitro
fertilization results.6
FF-MAS itself can be isolated from natural sources only
in submilligram quantities. Recently, a fermentation method
with metabolically engineered strains of Saccharomyces
cereVisiae has been published.7 With this method, only
milligram amounts of FF-MAS have been produced so far.
Therefore, chemical synthesis starting from readily avail-
able sterols or steroids seems to be the most attractive way
to obtain reasonable amounts of pure FF-MAS for biological
and clinical studies.
Several chemical syntheses of FF-MAS have been de-
scribed in the organic chemical literature8 and in patent
applications.9 All of them are partial syntheses starting from
well-known steroidal precursors. The main drawbacks of the
published syntheses are elaborate protection/deprotection
Although the exact role of MAS as a native paracrine
regulator remains controversial,5 first trials with human
(1) Byskov, A. G.; Andersen, C. Y.; Nordholm, L.; Thogersen, H.;
Guoliang, X.; Wassmann, O.; Andersen, J. V.; Guddal, E.; Roed, T. Nature
1995, 374, 559.
(2) Baltsen, M.; Byskov, A. G. Biomed. Chromatogr. 1999, 13, 382.
(3) (a) Hegele-Hartung, C.; Lessl, M.; Ottesen, J. L.; Grondahl, C. Human
Reprod. [Suppl.] 1998, 11, 193. (b) Hegele-Hartung, C.; Lessl, M.; Ottesen,
J. L.; Grondahl, C. Human Reprod. [Suppl.] 1998, 13, 98.
(4) For recent reviews see: (a) Hegele-Hartung, C.; Blume, T.; Lessl,
M.; Ottesen, J. L.; Grondahl, C. Reproduktionsmedizin 2000, 16, 129. (b)
Byskov, A. G.; Andersen, C. Y.; Leonardsen, L.; Baltsen, M. J. Exp. Zool.
1999, 285, 237. (c) Byskov, A. G.; Baltsen, M.; Andersen, C. Y. J. Mol.
Med. 1998, 76, 818.
(5) (a) Downs, S. M.; Ruan, B.; Schroepfer, G. J., Jr. Biol. Reprod. 2001,
64, 80. (b) Vaknin, K. M.; Lazar, S.; Popliker, M.; Tsafriri, A. Biol. Reprod.
2001, 64, 299. (c) Tsafriri, A.; Cao, X.; Vaknin, K. M.; Popliker, M. Mol.
Cell. Endocrinol. 2002, 187 (1-2), 197.
(6) Grondahl, C.; Hansen, T. H.; Marky-Nielsen, K.; Ottesen, J. L.; Hyttel,
P. Human Reprod. 2000, 15 (Suppl. 5, InnoVations and Limitations in
Assisted Reproduction), 3.
(7) Xu, R.; Wilson, W. K.; Matsuda, S. P. T. J. Am. Chem. Soc. 2002,
124, 918.
10.1021/ol0343156 CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/08/2003

strategies for the diene system in the steroidal core and the
separation of isomeric diene mixtures. The approaches
described herein avoid the protection of double bonds in the
steroidal core by using microbiological degradations of the
sterol side chains in multigram scale during the synthesis.
Our first approach starts from ergosterol 2 (route A,10 see
Scheme 1). After protection of the 3-hydroxy group in
tection of the 3-hydroxy group with pyridinium p-toluene-
sulfonate (PPTS),13 Oppenauer oxidation14 to the ketone with
migration of one double bond, dimethylation in position 4
with potassium tert-butoxide as the base, and subsequent
reduction of the 3-keto group with lithium aluminum hydride,
which gives 3â-alcohol 10 as the major diastereomer. The
final step is the acid-catalyzed isomerization, which includes
the fast isomerization of the ∆25 to the ∆24 double bond15
and slower isomerization of the ∆-5,7-diene system to the
thermodynamically more stable ∆-8,14-diene16 to give FF-
MAS as the major component17 of the isomerization mixture.
FF-MAS can be purified further by crystallization, HPLC,
or chromatography on silver-coated columns.18 We favored
two consecutive recrystallizations from methanol. This
procedure allowed us to isolate FF-MAS in >90% isomeric
purity and with full biological activity19 in a very convenient
way.
Scheme 1. Synthesis of FF-MAS from Ergosterol: Route Aa
Scheme 2. Synthesis of FF-MAS from
(20S)-20-Hydroxymethyl-pregn-4-en-3-one: Route Ba
a Conditions: (a) formaldehydedimethylacetal, P2O5, 100%; (b)
Mycobaterium sp.; (c) 1.5 equiv of TsCl, pyridine, rt, 8 h, 92%;
(d) Li2CuCl3, THF, -30 °C to rt, overnight, 71%; (e) PPTs, tBuOH,
reflux, 1 h, 72%; (f) Al(OiPr)3, toluene, cyclohexanone, reflux, 30
t
min, 73%; (g) KOtBu, MeI, BuOH, rt; 30 min, 68%; (h) LiAlH4,
THF, rt, 30 min, 79%; (i) 6N H2SO4, dioxane, reflux, 70 h, 69%.
ergosterol as a methoxymethyl (MOM) ether, which is
necessary for the microbiological step, the side chain was
cleaved by mycobacteria as described in the patent literature
to give 4 in a yield of 75%.11 The alcohol in the side chain
was converted to the tosylate, which can be coupled with
Grignard reagents in the presence of lithium chlorocuprate.12
We chose the homoallylic side chain for coupling, because
different attempts at coupling of an allylic (prenylic) side
chain gave unsatisfactory results. The next steps are depro-
a Conditions: (a) TIPSCl, imidazole, CH2Cl2, rt, 4 h; (b) KOtBu,
MeI, BuOH, rt; 30 min; (c) LiAlH4, THF, rt, 30 min; (d) BzCl,
t
pyridine; 0 °C, 1 h; 52% over four steps; (e) 1,3-dibromo-5,5-
dimethyl-imidazolidine-2,4-dione, benzene, hexane, 70 °C 30 min,
then 2,4,6-trimethylpyridine, toluene, reflux, 2 h; (f) TBAF, THF,
rt, 1 h, 61% over two steps; (g) p-TsCl, pyridine, 4-DMAP, 50 °C,
2 h, 72%; (h) Li2CuCl3, THF, -30 °C to rt, overnight; 100%; (i)
6 N H2SO4, dioxane, reflux, 70 h, 69%.
(8) (a) Dolle, R. E.; Schmidt, S. J.; Erhard, K. F.; Kruse., L. I. J. Am.
Chem. Soc. 1989, 111, 1, 278. (b) Ruan, B.; Watanabe, S.; Eppig, J. J.;
Kwoh, C.; Dzidic, N.; Pang, J.; Wilson, W. K.; Schroepfer, G. J., Jr. J.
Lipid Res. 1998, 39, 2005. (c) Murray, A.; Gronvald, F. C.; Nielsen, J. K.;
Faarup, P. Bioorg. Med. Chem. Lett. 2000, 10, 1067.
(9) (a) Blume, T.; Esperling, P.; Kuhnke, J. PCT Int. Appl. WO 9952930.
(b) Blume, T.; Esperling, P.; Kuhnke, J. Ger. Pat. Appl. DE 19817521,
1999. (c) Geisler, J.; Winter, E. PCT Int. Appl. WO 2000056758.
(10) Strategy of this approach (route A) has been disclosed in a German
patent application; see ref 9b.
We pursued a second synthesis (route B, see Scheme 2)
of FF-MAS starting from commercially available (20S)-20-
(13) Monti, H.; Leandri, G.; Klos-Ringuet, M.; Corriol, C. Synth.
Commun. 1983, 13, 1021.
(14) Cohen, C. F.; Louloudes, S. J.; Thompson, M. J. Steroids 1967, 9,
591.
(15) Analogous side chain isomerization under iodine catalysis has been
described: Kircher, H. W.; Rosenstein, F. U. J. Org. Chem. 1987, 52, 2586.
(16) Diene isomerizations have been described in the literature: Dolle,
R. E.; Kruse L. I. J. Org. Chem. 1986, 51, 4047 and references therein.
(11) Weber, A.; Kenneke, M.; Neef, G. PCT Int. Appl. 92/03465.
(12) For a comparable transformation, see: Morisaki, M.; Shibata, M.;
Duque, C.; Imamura N.; Ikekawa N. Chem. Pharm. Bull. 1980, 28, 606.
1838
Org. Lett., Vol. 5, No. 11, 2003

hydroxymethyl-pregn-4-en-3-one 11.20 For our needs, this
starting material was produced from the sterol sitosterin by
side chain degradation with mycobacteria as described in
the patent literature.21
The alcohol in the side chain of 11 was protected as a
triisopropylsilyl ether. Dimethylation in position 4 and
subsequent reduction of the 3-keto group was performed in
analogy to route A. The resulting 3-â-OH group was
protected as a benzoate. In this step, the first purification
during synthesis was done. Pure 14 was isolated after a
simple crystallization from methanol in a yield of 50% over
four steps.
tosylated under standard conditions. Coupling with the
homoallyl-Grignard reagent as in route A gave the ∆-25 side
chain and concomitant deprotection of the 3-hydroxy group.
The final isomerization step is identical to route A. The
overall yield for synthesis using route B was 10% (over nine
steps).
The FF-MAS was purified as described for route A. Both
route A and route B are attractive methods for the preparation
of FF-MAS in gram scale. Route B is in our view especially
interesting because it contains fewer purification steps and
allows late introduction of side chains to produce biologically
active analogues of FF-MAS for structure-activity relation-
ship studies.23
In summary, we have developed two syntheses of FF-MAS
that can produce the product in gram scale. Both routes are
based on microbiological oxidations of sterol side chains and
require only nine steps from commercially available starting
materials and thus constitute attractive routes for the
syntheses of FF-MAS and biologically active analogues.
A bromination/dehydrobromination reaction22 gave diene
15, which was deprotected in the side chain with tetrabutyl-
ammonium fluoride (TBAF), and the resulting alcohol was
(17) Isomeric ratio of ∆24:∆25 in the side chain is approximately 8:1;
the ratio of dienes ∆8,14:∆6,8(14) is approximately 2.5:1, which corresponds
to a content of FF-MAS (∆8,14,24-triene) of ca. 60%.
(18) See ref 8b.
(19) Oocyte assay has been adopted from the literature (see ref 1). Naked
oocytes (NO) were punctured from ovaries of mice primed with follicle
stimulation hormone. After in vitro culture in the presence of 3 mM
hypoxanthine and 10 µM of the corresponding test compounds for 15-24
h, germinal vesicle breakdown (GVB) and polarbody (PB) values are
measured. The values for GVB and PB are comparable to the data reported
in the literature.
Acknowledgment. We thank Juergen Skupsch and Danie-
la Wrobel for recording NMR spectra, Barbara Boes for mass
spectra, Oliver Schenk for analytical HPLC data, and
Bernhard Lindenthal for the oocyte tests.
(20) Commercially available from Research: 10 g costs ca. $300 $.
(21) Weber, A.; Kennecke, M.; Mueller, R. Ger. Pat. Appl. DE 2757156,
1979.
(22) Johnson, R. L.; Carey, S. C.; Norman, A. W.; Okamura, W. H. J.
Med. Chem. 1977, 20, 5.
(23) For the synthesis of analogues with enhanced biological activity in
the cumulus-enclosed oocyte assay based on route B, see: Blume, T.;
Esperling, P.; Hegele-Hartung, C. Eur. Pat. Appl. EP 1245572, 2002.
1
Supporting Information Available: H NMR and mass
spectra of all intermediates and the final product. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL0343156
Org. Lett., Vol. 5, No. 11, 2003
1839