Enantioselective synthesis of (-)-methoxyestrone

Article abstract of DOI:10.1002/ejoc.201100317

Enantioselective synthesis of unnatural (-)-methoxyestrone in 12 steps from the commercially available tetralone based on the formation of a chiral bicyclic intermediate having the A-B steroid ring framework was accomplished. The crucial synthetic step co

Full text of DOI:10.1002/ejoc.201100317

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201100317
Enantioselective Synthesis of (–)-Methoxyestrone
[a]
[a]
Robert Betík and Martin Kotora*
Keywords: Enantioselectivity / Asymmetric synthesis / Steroids / Cycloaddition / Rearrangement
Enantioselective synthesis of unnatural (–)-methoxyestrone
in 12 steps from the commercially available tetralone based
on the formation of a chiral bicyclic intermediate having the
A–B steroid ring framework was accomplished. The crucial
synthetic step comprised the enantioselective conjugate ad-
dition of vinylmagnesium bromide to a chiral imine formed
from a trisubstituted cyclic α,β-unsaturated aldehyde with
Ͼ98%ee, giving rise to the stereoselectively substituted
building block possessing the A–B steroid rings. Further
steps included the construction of the side chain containing
the triple bond, and the obtained enyne was subjected to a
Pauson–Khand reaction that furnished stereoselectively an
intermediate tetracyclic ketone. Further functional group
transformations yielded the target compound.
Introduction
intermediate with the correct relative stereochemistry under
racemic conditions by using Cp ZrBu , the use of a chiral
2
2
The total synthesis of compounds possessing the steroid
framework is attractive, as the target may possess biological
activity, and the process serves as a probing stone for syn-
thetic methodology. In this regard, estrone, thanks to its
rather complex structure, has constituted an ideal and fa-
vorable target molecule. Development of new synthetic
zirconocene derivative under catalytic or stoichiometric
conditions did not lead to the expected intermediates;
[6]
moreover, loss of stereoselectivity was observed. Obvi-
ously this led to the conclusion that a different methodol-
ogy should be applied to fulfil the desired goal. In this re-
port we would like to outline an approach for the enantio-
selective construction of the estrone framework that is
based on the initial introduction of chirality on the
B steroid ring (positions 8 and 9).
[
1]
[2]
pathways for its enantioselective synthesis has also been
fuelled by a recent interest in ent-steroids (not produced by
Nature), which are presumed to have biological activities
different from those of the natural ones.[
3]
Recently, we reported two new procedures for the dia-
stereoselective synthesis of an advanced steroid intermedi- Results and Discussion
ate bearing the required tetracyclic skeleton including the
Because we wanted to apply some parts of the previously
correct relative stereochemistry. The first approach was
based on threefold consecutive use of Cp ZrBu -mediated
used methodology in a new approach, we decided to modify
early steps of the synthesis. The overall retrosynthetic analy-
sis is outlined in Scheme 1. It was presumed that the crucial
step along the synthetic pathway – the preparation of a chi-
ral intermediate with the tetrahydronaphthalene moiety –
could be accomplished by enantioselective conjugate ad-
dition of a vinyl nucleophile to α,β-unsaturated carbonyl
compound 2a, which would introduce functional groups
2
2
reactions (Negishi reagent) to construct the steroid A–C
rings, and finally the D ring was assembled by a Ru com-
plex catalyzed ring-closing metathesis.[4] The second one
utilized two previous Cp ZrBu -mediated reactions for the
2
2
construction of the A and B rings followed by a Co-medi-
ated Pauson–Khand reaction that allowed the C and D
rings to be assembled in one step.[5] These procedures led
(
directly or after additional transformation) suitable for a
metal-mediated cyclization reaction.
to the straightforward synthesis of the known tetracyclic
intermediate over nine steps from the commercially avail-
able starting material; nonetheless, attempts to develop an
enantioselective variant of this procedure were not met with
success. Although the key synthetic step, the closing of the
B ring, could furnish a substituted tetrahydronaphthalene
Interestingly, conjugate additions to trisubstituted enals
with a similar framework to that of 2a have not been re-
ported yet. Because the conjugate additions of various nu-
cleophiles to sterically hindered enals constitute interesting
synthetic and theoretical problems, we decided to explore
its scope. We envisioned three candidates as acceptors and
potential synthetic intermediates for conjugate addition: al-
dehyde 2a, ketone 2b, and ester 2c, which were prepared
from commercially available 6-methoxy-3,4-dihydro-2H-
naphthalen-1-one (1) by using previously described meth-
[
a] Dept. Organic and Nuclear Chemistry,
Faculty of Science, Charles University in Prague
Hlavova 8, 12843 Praha 2, Czech Republic
Fax: +420-221-951-326
E-mail: kotora@natur.cuni.cz
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201100317.
[
7–10]
odology (see the Supporting Information).
Eur. J. Org. Chem. 2011, 3279–3282
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3279

R. Betík, M. Kotora
SHORT COMMUNICATION
Having secured chiral aldehyde 3a (Scheme 3), a two step
reaction sequence was required to convert it into halides 4.
Reduction of the aldehyde to an alcohol with LiAlH fol-
4
lowed by reaction with PPh /NBS or PPh /NIS furnished
3
3
corresponding bromide 4a or iodide 4b in 73 and 86% yield,
respectively. It should be noted that although aldehyde 3a
was obtained as an 8:1 trans/cis mixture, after conversion
into the halides only the corresponding trans diastereoiso-
mers were obtained. An attempt to convert bromide 4a into
the corresponding Grignard reagent followed by its cou-
pling with 2,3-dibromopropene was not met with success.
A mixture of the starting bromide, the dehydrohalogenated
product, and a dimer was obtained without any desired
product 5. Then we turned our attention to iodide 4b, which
was treated with Rieke zinc (3 equiv.) in the presence of
Et Zn (5 mol-%) to form an organozinc reagent. The pres-
2
Scheme 1. Retrosynthetic analysis of methoxyestrone synthesis.
ence of Et Zn was indispensable for the full conversion of
2
the iodide into the corresponding organozinc compound.
In its absence, the conversions were in the 20–80% range.
After securing compounds 2a–c, conjugate additions
with various vinyl nucleophiles were tried under various
catalytic and stoichiometric conditions (Scheme 2). Initially,
it was expected that the addition of vinylboronic acid ester
under Miyaura–Hayashi conditions [a Rh catalyst in com-
bination with (S)-BINAP][ could yield desired chiral in-
termediate 3a. Although this step seemed not to pose any
problem, because several synthetic procedures for enantio-
selective conjugate additions of boronic acid derivatives to
trisubstituted olefins have been reported,[12] our subsequent
studies proved otherwise. Attempts to carry out conjugate
addition of CH =CHB(OBn) catalyzed by Rh(acac)(CO) ,
11]
2
2
2
Rh(acac)(cod), or Rh(acac)(CH =CH ) in combination
2
2
with (S)-BINAP did not give rise to expected product 3a.
The conjugate addition of vinylmagnesium bromide cata-
lyzed by CuCN (10 mol-%) was somewhat more successful.
It gave 3a in 5% yield along with 50% of the 1,2-addition
product. Although we tried to improve the 1,4-selectivity
by changing the reaction conditions (e.g., combination of
[
13]
vinylmagnesium bromide with AlCl
alone or in the pres-
3
ence of HMPA, etc.),[ only complex reaction mixtures
were usually obtained. Finally, a stoichiometric method
based on the addition of vinylmagnesium bromide to the in
14]
[
15]
situ formed imine
(from 2a and l-tert-leucine tert-butyl
[
16]
ester ) gave rise to 3a in 60% yield as an 8:1 mixture of
S,S-/S,R-diastereoisomers (trans/cis isomers) with high
asymmetric induction of 98%ee (it refers to the configura-
tion on the C2 carbon atom). Conjugate additions to 2b
and 2c resulted in inferior results only (for details, see the
Supporting Information).
Scheme 3. Reagents and conditions: (a) tert-Butyl-tert-leucine, hex-
ane, 3 Å molecular sieves; (b) CH
8%ee); (c) 1. LiAlH , THF, 0 °C; 2. PPh
PPh /NIS, THF (86%); (d) 1. Zn, Et Zn (5 mol-%), THF, 40 °C;
. CH =CBrCH Br, CuCl (3 equiv.), 0 °C (93%); (e) TBAF, DMF
96%); (f) 1. nBuLi, THF, –78 °C; 2. MeI (92%); (g) 1. Co (CO)
toluene; 2. DMSO, 80 °C (91%); (h) LiAlH /AlCl , Et O (81%);
(i) m-CPBA, CH Cl , 0 °C (91%).
2
=CHMgBr, THF, –20 °C (60%,
9
4
3
/NBS, THF (73%) or
3
2
2
(
2
2
2
8
,
4
3
2
Scheme 2. Conjugate addition to 2a.
2
2
3280
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 3279–3282

Enantioselective Synthesis of (–)-Methoxyestrone
The formed organozinc compound was immediately reacted utilized Lewis acids for the rearrangement of epoxides to
[
19]
[20]
with 2,3-dibromopropene in the presence of CuCl to form carbonyl compounds such as Bi(OTf) ,
bromodiene 5 in 93% yield.
Cu(BF4),
or
3
[
21]
IrCl3 did not give rise to the desired product; only com-
In the next four steps, bromodiene 5 was converted into plex mixtures of compounds were obtained.
a tetracyclic estrone precursor in an analogous manner to
our previously reported racemic synthesis.[ Firstly, dehy-
5]
drobromination of 5 with TBAF yielded enyne 6 (96%), its Conclusions
terminal triple bond was metalated with nBuLi, and the
In conclusion, we have outlined a potentially versatile
formed acetylide was treated with MeI to give enyne 7 in
2% yield. Then, the Pauson–Khand reaction of 7 with
Co CO proceeded diastereoselectively to give rise to a sin-
steroid construction strategy based on the synthesis of a
chiral synthon having defined configurations at C8 and C9
of the steroid B ring. The synthetic utility was validated
with the synthesis of (–)-estrone in 12 steps starting from
commercially available tetralone. It is fair to admit that the
presented synthesis is longer than those starting from
tetralone, such as the classical Torgov’s approach (6[
steps) or the recently reported enantioselective approaches
9
2
8
gle stereoisomer of tetracyclic ketone 8 in 95% yield. Fi-
nally, chemoselective reduction of the keto group with a
mixture of LiAlH /AlCl furnished chiral tetracycle 9 in
4
3
81% yield. It should be emphasized that during all these
22]
steps no racemization was observed.
As it had been shown that tetracycle 9 can be easily con-
verted into estrone,[ we decided to follow the described
procedure. Initially, tetracycle 9 was transformed into a
[
23]
based on the use of Dane’s diene in a Diels–Alder reac-
17]
[
1g,1i]
[1e]
tion (five
or seven
steps) or hydrometalation reac-
[
1f]
tions (nine steps). The number of steps in this procedure
is comparable to that of the recently reported 11-step pro-
3
.5:1 mixture of epoxides 10a/10b by reaction with m-CPBA
[
18]
in 91% yield. For the last step, that is, a Lewis acid cata-
lyzed rearrangement of the epoxy group to the carbonyl
group, only epoxide 10a with the desired stereochemistry
was used. This step proved to be more complicated than
reported and envisioned. In our hands it usually resulted in
the formation of a mixture of methoxyestrone 11, tetracy-
clic alcohol 12, and a mixture of unknown nonpolar com-
pounds 13 (Scheme 4). Some typical examples are displayed
in Table 1. When the rearrangement was carried out accord-
ing to the previously published procedure (BF ·Et O, benz-
ene, 20 °C) only a mixture of nonpolar compounds 13 was
formed, not even trace amounts of 11 or 12 were observed.
Carrying out the reaction at –20 °C gave rise only to a
minor amount of 11 (4%); the major product was found to
be tetracyclic alcohol 12 (70%) along with 13 (Ͻ20%). The
highest yield of (–)-methoxyestrone (11, 25%) was obtained
at –78 °C. Interestingly, the application of other frequently
cedure based on
a
conjugate addition/allylation se-
[
1h]
quence.
However, the crucial difference in comparison
with the above-mentioned syntheses is that they all initially
introduced chirality on the D ring (positions 13 and 14),
whereas our approach relies on the introduction of chirality
on the B ring (positions 8 and 9). In this respect, it was
shown that conjugate addition to trisubstituted α,β-unsatu-
rated aldehyde led to the desired chiral intermediate with
high enantioselectivity by using the recyclable chiral auxil-
iary. We think that this methodology could be useful for the
synthesis of a wide range of natural and unnatural steroids
as well as their analogues. Further work regarding this goal
is in progress.
3
2
Experimental Section
Synthesis of (1R,2R)-1-Formyl-6-methoxy-3,4-dihydro-2-vinyl-
naphthalene (3a) as a Representative Procedure for Stoichiometric
Conjugate Additions: l-tert-Leucine tert-butyl ester (5.67 mmol,
1.06 g) was added to a solution of 2a (5.32 mmol, 1 g) in hexane
(
(
15 mL) at 20 °C. To this solution was added 3 Å molecular sieves
1 g), and the reaction mixture was stirred overnight. Then it was
left to stand without stirring for 10 min for the molecular sieves to
sediment. The solution over the sieves was transferred by cannula
to another flask and the volatiles were removed under reduced
pressure to yield a crude aldimine that was used for the next step
without further purification. Vinylmagnesium bromide (1 m in
THF, 17 mmol, 17 mL) was added to the stirred solution of the
crude aldimine in THF (60 mL) at –40 °C over a period of 2 h. The
reaction mixture was then allowed to warm up to –20 °C and kept
at this temperature for 3 h. the solution was then diluted with HCl
(1%, 100 mL) and extracted with CH Cl (3ϫ60 mL). The com-
Scheme 4. Rearrangement of 10.
Table 1. Rearrangement of 10 catalyzed by various Lewis acids.
Lewis acid
Solvent
T
Yield [%]
12
1 [
1
a]
[b]
13[b]
[
[
[
c]
BF
BF
BF
Bi(OTf)
Cu(BF
3
3
3
·Et
·Et
·Et
2
2
2
O
O
O
benzene
benzene
toluene
20
–20
–78
–20
–20
–20
0
4
0
70
55
0
0
0
Ͼ95
Ͻ20
Ͻ20
Ͼ95
Ͼ95
Ͼ95
c]
c]
2
2
bined organic fractions were dried with anhydrous MgSO , the vol-
4
25
0
[
d]
3
CH
MeCN
CH Cl
2
Cl
2
atiles were removed under reduced pressure, and column
[
d]
4
)
2
0
0
chromatography of the residue on silica gel (CH Cl /hexane, 1:1)
2
2
[
d]
IrCl
3
2
2
yielded the title compound (0.7 g, 61%, Ͼ98%ee) as a mixture of
trans/cis isomers in a 8:1 ratio. Enantiomeric ratios were deter-
mined by GC (HP-Chiral β column, 30 mϫ0.25 mm, oven: 70 °C
[
[
a] Isolated yield. [b] A mixture of unknown nonpolar compounds.
c] 4 equiv. of the Lewis acid was used. [d] 1 equiv. of the Lewis acid
was used.
for 0 min, then 0.5 °C/min to 170 °C): t(1S2S) = 176.0 min, t(1R2R)
=
Eur. J. Org. Chem. 2011, 3279–3282
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjoc.org 3281

R. Betík, M. Kotora
SHORT COMMUNICATION
1
76.6 min. Recrystallization (MeOH) yielded the pure trans isomer
[3] a) J.-F. Biellmann, Chem. Rev. 2003, 103, 2019–2033; b) D. F.
Covey, Steroids 2009, 74, 577–585.
(0.55 g) as a white crystal. Data for the trans isomer: M.p. 46–
ˇ
1
[4] a) P. Herrmann, M. Kotora, M. Bud eˇ sˇ ínský, D. Saman, I.
4
=
(
7 °C. [α]
1.63–1.75 (m, 1 H, CHH), 1.94–2.04 (m, 1 H, CHH), 2.78–2.88
m, 3 H, ArCH + CH =CHCH), 3.44 (dd, J = 7.6, 3.2 Hz, 1 H,
ArCH), 3.79 (s, 3 H, OCH ), 5.05–5.10 (m, 1 H, CH=CH ), 5.08–
.14 (m, 1 H, CH=CH ), 5.84 (ddd, J = 17.6, 10.6, 7.4 Hz, 1 H,
CH=CH ), 6.68–6.74 (m, 1 H, Ar-H), 6.75–6.80 (m, 1 H, Ar-H),
.98–7.02 (m, 1 H, Ar-H), 9.47 (d, J = 3.2 Hz, 1 H, CHO) ppm.
D 3 3
= –9 (CHCl , c = 0.5). H NMR (400 MHz, CDCl ): δ
Císa rˇ ová, Org. Lett. 2006, 8, 1315–318; b) P. Herrmann, M.
Bud eˇ sˇ ínský, M. Kotora, Chem. Lett. 2007, 36, 1268–1269; c)
P. Herrmann, M. Bud eˇ sˇ ínský, M. Kotora, J. Org. Chem. 2008,
2
2
3
2
7
3, 6202–6206.
5
2
[
5] R. Betík, P. Herrmann, M. Kotora, Eur. J. Org. Chem. 2010,
2
646–655.
6
[6] P. Herrmann, M. Kotora, Tetrahedron Lett. 2007, 48, 3209–
13
C NMR (100 MHz, CDCl
3
): δ = 26.64 (CH
2
), 27.98 (CH
2
), 37.90
3212.
[
[
[
[
7] K. J. Shea, G. J. Stoddard, W. P. England, C. D. Haffner, J. Am.
(CH), 55.21 (OCH ), 56.22 (ArCH), 112.54 (Ar), 114.41 (Ar),
3
Chem. Soc. 1992, 114, 2635–2643.
1
1
2
8
1
15.40 (C=C), 121.62 (Ar), 130.42 (Ar), 138.75 (Ar), 139.83 (C=C),
58.68 (Ar), 201.06 (C=O) ppm. IR (KBr): ν˜ = 3082, 3008, 2962,
939, 2859, 2804, 2714, 1719, 1642, 1607, 1499, 1257, 1004, 919,
39, 919 cm . MS (EI): m/z (%) = 216.1 (15) [M] , 187.1 (100),
59.1 (25), 146.1 (25), 128.1 (10), 115.1 (15). HRMS (EI+): calcd.
8] A. Sultani, H. Dietrich, F. Richter, H. H. Otto, Monatsh.
Chem. 2005, 136, 1651–1669.
9] J. Quirante, X. Vila, C. Escolano, J. Jonjoch, J. Org. Chem.
–1
+
2002, 67, 2323–2328.
10] A. A. C. Wijk, M. B. Weerd, J. Lughenburg, Eur. J. Org. Chem.
2003, 863–868.
for C14
H 7.47, found C 74.84, H 7.19, these values correspond to
+ MeOH. R (CH Cl /hexane, 1:1) = 0.4.
16 2 16 2
H O 216.1150; found 216.1152. C14H O : calcd. C 77.74,
[11] a) Y. Takaya, M. Ogasawara, T. Hayashi, J. Am. Chem. Soc.
1998, 120, 5579–5580; b) T. Hayashi, Synlett 2001, 120, 879–
2C
14
H
16
O
2
f
2
2
8
2
1
87; c) T. Hayashi, K. Yamasaki, Chem. Rev. 2003, 103, 2829–
844; d) M. Sakai, H. Hayashi, N. Miyaura, Organometallics
997, 16, 4229–4231; e) T. Itoh, T. Mase, T. Nishikata, T.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures and characterization data.
Iyama, H. Tachikawa, Y. Kobayashi, Y. Yamamoto, N. Mi-
yaura, Tetrahedron 2006, 62, 9610–9621.
[
[
[
[
[
12] R. Itooka, Y. Iguchi, N. Miyaura, J. Org. Chem. 2003, 68,
Acknowledgments
6000–6004.
13] M. V. Augustin, A. Alexakis, Chem. Eur. J. 2007, 13, 9467–
9662.
This work was supported by the Ministry of Education, Youth and
Sports of the Czech Republic through grant project 1M0508 and
MSM0021620857.
14] P. Perlmutter, Conjugate Addition Reactions in Organic Synthe-
sis, Pergamon Press, Oxford, U.K., 1992.
15] H. Kogen, K. Tomioka, S. Hashimoto, K. Koga, Tetrahedron
Lett. 1980, 21, 4005–4008.
16] A. B. Smith, L. Kurti, A. H. Davulen, Y. S. Cho, Org. Process
Res. Dev. 2007, 11, 19–24.
[
1] For enantioselective syntheses of estrone during the last dec-
ade, see: a) L. F. Tietze, T. Nöbel, M. Spescha, J. Am. Chem.
Soc. 1998, 120, 8971–8977; b) L. F. Tietze, M. Wiegand, C.
Vock, J. Organomet. Chem. 2003, 687, 346–352; c) K. Hanada,
N. Miyazawa, K. Ogasawara, Chem. Pharm. Bull. 2003, 51,
[17] P. A. Bartlett, W. S. Johnson, J. Am. Chem. Soc. 1973, 95,
7501–7502.
[18] M. B. Groen, B. Hindriksen, F. J. Zeelen, Recl. Trav. Chim.
Pays-Bas 1982, 101, 148–155.
[19] K. A. Bhatia, K. J. Eash, N. M. Leonard, M. C. Oswald, R. S.
Mohan, Tetrahedron Lett. 2001, 42, 8129–8132.
[20] M. W. C. Robinson, K. S. Pillinger, A. E. Graham, Tetrahedron
Lett. 2006, 47, 5919–5921.
1
04–106; d) G. Pattenden, M. A. Gonzales, S. McCulloch, A.
Walter, S. J. Woodhead, Proc. Natl. Acad. Sci. USA 2004, 101,
2024–12029; e) Q.-Y. Hu, P. D. Rege, E. J. Corey, J. Am.
Chem. Soc. 2004, 126, 5984–5986; f) D. Soorukam, P. Knochel,
1
Org. Lett. 2007, 9, 1021–1023; g) E. Canales, E. J. Corey, Org. [21] I. Karamé, M. L. Tommasino, M. Lemaire, Tetrahedron Lett.
Lett. 2008, 10, 3271–3273; h) V. Foucher, B. Guizzardi, M. B.
Groen, M. Light, B. Linclau, Org. Lett. 2010, 12, 680–683; i)
M. Weimar, G. Dürner, J. W. Bats, M. W. Gödel, J. Org. Chem.
2003, 44, 7687–7689.
[22] a) S. N. Anachenko, I. V. Torgov, Tetrahedron Lett. 1963, 4,
1553–1558; b) S. N. Anachenko, I. V. Torgov, Steroids 1964, 4,
33–40.
2010, 75, 2718–2721.
[
2] For syntheses of ent-steroids, see: a) E. J. Westover, D. F. Covey, [23] E. Dane, J. Schmitt, Justus Liebigs Ann. Chem. 1939, 537, 246–
Steroids 2003, 68, 159–166; b) E. J. Westover, D. F. Covey, Ste-
roids 2006, 71, 484–488; c) Z. Y. Cai, D. F. Covey, Steroids
249.
Received: March 5, 2011
Published Online: May 13, 2011
2007, 72, 351–359.
3282
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 3279–3282