Desymmetrization of silyl-2,5-cyclohexadiene. Synthesis of (+)-conduritol E and (-)-2-deoxy-allo-inositol

Full text of DOI:10.1021/jo960814r

5202
J . Org. Chem. 1996, 61, 5202-5203
Sch em e 1
Desym m etr iza tion of a
Silyl-2,5-cycloh exa d ien e. Syn th esis of
(+)-Con d u r itol E a n d
(-)-2-Deoxy-a llo-in ositol
Re´my Angelaud and Yannick Landais*
Institut de Chimie Organique, Universite´ de Lausanne,
Colle`ge Prope´deutique, 1015 Lausanne-Dorigny, Switzerland
Received May 3, 1996
Desymmetrization of meso compounds1 or symmetrical
bifunctional substrates2 is a very attractive method,
allowing the transformation of easily available sym-
metrical precursors into asymmetric synthons of high
value in a limited number of steps. While this approach
has been very successful and is well documented in
acyclic series, it has never been applied to cyclic systems
such as 2. Such an approach would be particularly
significant in the context of developing the potential of
polyhydroxylated cyclohexanes such as cyclitols 1a ,b3 as
inhibitors of glycosidases. Their ability to inhibit oli-
gosaccharide-processing enzymes provides a wide range
of possible applications in chemotherapy for these com-
pounds, since glycoproteins are involved in numerous
biochemical processes.4 This has stimulated enormous
synthetic efforts recently, and several asymmetric ap-
proaches to these substrates have been proposed,5 among
the most efficient of which is the microbial oxidation of
aromatic precursors.6 We propose here a general strat-
egy directed toward the synthesis of polyhydroxylated
cyclohexanes through desymmetrization of silyl-2,5-cy-
clohexadienes using Sharpless asymmetric dihydroxyla-
tion.7 Surprisingly, the functionalization of such precur-
sors has never been addressed so far.8 The silylated
analogue of 2 (i.e., 3) possesses a silicon group that can
be regarded as a hydroxy equivalent9 and is easily
accessible via a Birch reduction10 of the corresponding
arylsilane 4 (Scheme 1). Further elaboration of the
Sch em e 2
resulting synthons was anticipated to give access to the
necessary homochiral intermediates. As a demonstration
of the utility and the versatility of our methodology, we
also undertook the asymmetric syntheses of both condu-
ritol E (1a ) and 2-deoxy-allo-inositol (1b).
We speculated that the commercially available Sharp-
less dihydroxylation reagent7 (i.e., AD-mix) would be able
to differentiate the two enantiotopic double bonds and
that the silicon moiety would control the diastereofacial
selectivity by forcing the attack of the incoming electro-
phile in an anti fashion (1,2-stereocontrol)11 (Scheme 2).
The carbon-silicon group would then be oxidized with
retention of configuration using the classical Tamao-
Kumada-Fleming procedure, finally revealing the OH
group.9
2,5-Cyclohexadienylsilanol 3a was readily prepared in
70% yield12 using the Birch reduction (Li/NH3) of com-
mercially available PhMe2SiCl13 4a (Scheme 3). This
result contrasts with the studies of Eaborn et al.10b,c on
Birch reduction of PhSiMe3, where only 20% of dienyl-
silane was formed along with recovered starting material.
It is likely that, in our case, the reaction takes place on
a more reactive aminosilane intermediate (formed by
aminolysis14 of 4a ), which would then produce the silanol
3a after hydrolysis. Dihydroxylation of 3a using various
* To whom correspondence should be addressed. Tel.: 41-(021)-692-
40-05. FAX: 41-(021)-692-40-05. E-mail: ylandais@ulys.unil.ch.
(1) (a) Schreiber, S. L. Chem. Scr. 1987, 27, 563. (b) Schreiber, S.
L.; Goulet, M. T.; Schulte, G. J . Am. Chem. Soc. 1987, 109, 4718.
(2) (a) Ho, T. L. Symmetry. A basis for synthetic design; Wiley
Interscience, Ed.; J ohn Wiley & Sons: New York, 1995. (b) Schreiber,
S. L.; Schreiber, T. S.; Smith, D. B. J . Am. Chem. Soc. 1987, 109, 1525.
(c) Tamao, K.; Tohma, T.; Inui, N.; Nakayama, O.; Ito, Y. Tetrahedron
Lett. 1990, 31, 7333. (d) Takano, S.; Sakurai, K.; Hatakeyama, S. J .
Chem. Soc., Chem. Commun. 1985, 1759. (e) J a¨ger, V.; Schro¨ter, D.;
Koppenhoefer, B. Tetrahedron 1991, 47, 2195.
(3) Balci, M.; Su¨tbeyaz, Y.; Sec¸en, H. Tetrahedron 1990, 46, 3715.
For previous asymmetric syntheses of conduritol E: (a) Takano, S.;
Yoshimitsu, T.; Ogasawara, K. J . Org. Chem. 1994, 59, 54. (b)
Hudlicky, T.; Luna, H.; Olivo, H. F.; Andersen, C.; Nugent, T.; Price,
J . D. J . Chem. Soc., Perkin Trans. 1 1991, 2907. For a previous
asymmetric synthesis of 2-deoxy-allo-inositol: (c) McCasland, G. E.;
Furuta, S.; J ohnson, L. F.; Shoolery, J . N. J . Am. Chem. Soc. 1961,
83, 2335. (d) Angyal, S. J .; Odier, L. Carbohydr. Res. 1982, 101, 209.
(e) Angyal, S. J .; Odier, L. Ibid. 1982, 100, 43.
(9) For a review on the C-Si bond oxidation, see: J ones, G. R.;
Landais, Y. Tetrahedron Report 1996, 52, 7599. (a) Tamao, K.; Ishida,
N.; Tanaka, T.; Kumada, M. Organometallics 1983, 2, 1694. (b)
Fleming, I.; Henning, R.; Parker, D. C.; Plaut, H. E.; Sanderson, P.
E.J . J . Chem. Soc., Perkin Trans. 1 1995, 317. The silicon must carry
at least one activating group (X ) F, OR, OH, NR2, ...) for the oxidation
of the C-Si bond to occur.
(4) Look, G. C.; Fotsch, C. H.; Wong, C.-H. Acc. Chem. Res. 1993,
26, 182. (b) Winchester, B.; Fleet, G. W. J . Glycobiology 1992, 2, 199.
(c) Legler, G. Adv. Carbohydr. Chem. Biochem. 1990, 48, 319. (d)
Sinnott, M. L. Chem. Rev. 1990, 90, 1171. (e) Hughes, A. B.; Rudge,
A. J . Nat. Prod. Rep. 1994, 135. (f) Bischoff, H. Eur. J . Clin. Invest.
1994, 24, 3.
(5) (a) Vogel, P.; Fattori, D.; Gasparini, F.; Le Drian, C. Synlett 1990,
173. (b) Paulsen. H.; Ro¨ben, W.; Heiker, F. R. Chem. Ber. 1981, 114,
3242.
(6) (a) Carless, H. A. J . Tetrahedron: Asymmetry 1992, 3, 795. (b)
Hudlicky, T. Chem. Rev. 1996, 96, 1. (c) Ley, S. V. Pure Appl. Chem.
1990, 62, 2031.
(7) Kolb, H. C.; vanNieuwenhze M. S.; Sharpless, K. B. Chem. Rev.
1994, 94, 2483.
(10) (a) Rabideau, P. W.; Marcinow, Z. Org. React. 1992, 42, 1. (b)
Eaborn, C.; J ackson, R. A.; Pearce, R. J . Chem. Soc., Perkin Trans. 1
1975, 470. (c) Eaborn, C.; J ackson, R. A.; Pearce, R. Ibid. 1975, 2055.
(d) Taber, D. F.; Bhamidipati, R. S.; Yet, L. J . Org. Chem. 1995, 60,
5537.
(11) Fleming, I.; Dunogue`s, J .; Smithers, R. Org. React. 1989, 37,
57.
(12) Ten to 20% of the corresponding siloxane is sometimes produced
during the workup but is easily removed by distillation. The amount
of siloxane can be reduced by diluting the reaction mixture prior to
aqueous workup.
(13) Available from Aldrich (no. 11,337-9). PhMe2SiCl is also
conveniently prepared on a 200 g scale from bromobenzene and Me2-
SiCl2: Andrianov, K. A.; Delazari, N. V. Dok. Akad. Nauk. SSSR 1958,
122, 393; Chem. Abst. 1959, 53, 2133.
(8) Bennetau, B.; Dunogue`s, J . Synlett 1993, 171.
S0022-3263(96)00814-6 CCC: $12.00 © 1996 American Chemical Society

Communications
J . Org. Chem., Vol. 61, No. 16, 1996 5203
Sch em e 3
Sch em e 4
repeating the reaction with (+)-DET we obtained the
same diol after 4 days at -20 °C in a much lower yield,
indicating a remarkable rate difference between “matched”
and “mismatched” pairs.
The total synthesis of (+)-conduritol E was completed
using LDA-mediated opening19 of the epoxide ring of 7
followed by removal of the acetonide, leading to optically
pure 1a (after one recrystallization) in 20% overall yield
from PhMe2SiCl (Scheme 4). It is worth mentioning that
such a transformation allows the functionalization of the
methylene C-4 center and, hence, functionalization of
each carbon centre of the cyclohexadiene 3a . Alterna-
tively, the treatment of epoxide 7 with a 10% AcOH-
THF mixture afforded, with complete regioselectivity,
optically pure (-)-2-deoxy-allo-inositol (1b) (after one
recrystallization) in a 30% overall yield from PhMe2SiCl.
The ring epoxide opening follows as expected a trans-
diaxial mode obeying the Fu¨rst-Plattner rule.20 Base-
catalyzed epoxide opening (PhCO2Na (cat.), H2O) likewise
afforded the regioisomer 1b in good agreement with
Hudlicky's recent observations.3b We also noticed that
opening of the epoxide is slower than the deprotection of
the acetonide since quenching the reaction mixture after
only 24 h under reflux furnished a mixture of 1b and
the epoxy triol intermediate in a 1:1 ratio.
Ta ble 1. In flu en ce of th e Na tu r e of th e Liga n d s in
AD-m ix on th e Asym m etr ic Dih yd r oxyla tion of 3a
(Sch em e 3)
liganda
eeb
entc
(DHQD)2PHAL
DHQD-IND
(DHQD)2PYR
(DHQ)2PYR
44
40
52
65
(-)
(-)
(-)
(+)
a
DHQ:dihydroquinine; DHQD:dihydroquinidine; PHAL:phthala-
b
zine; PYR:diphenylpyrimidine; IND:indole. Enantiomeric excess
measured using 1H NMR (Eu(hfc)3) or capillary GC. c Enantiomer
formed.
AD-mix formulations (Table 1) afforded the desired 1,2-
diol,15 which was directly protected as the acetonide 5.
Interestingly, the silanol was converted into the silyl-
methyl ether during the diol protection. Examination of
1
the H NMR of 5 revealed that only one diasteroisomer
In summary, we have demonstrated that the desym-
metrization of a readily available silyl-2,5-cyclohexadiene
using asymmetric dihydroxylation affords an efficient and
rapid entry into cyclohexane skeletons having four or five
stereogenic centers in a stereocontrolled manner. This
strategy is competitive in terms of cost, efficiency, and
versatility with other approaches to similar substrates.3,5,6
The chlorosilane precursor 4a is inexpensive, and both
enantiomers are theoretically accessible. Further ma-
nipulation of intermediates such as 6 or 7 should also
provide the means to access other cyclitols and related
analogues. Finally, it is worth mentioning that we have
only reported the desymmetrization of 3a using asym-
metric electrophilic processes. Other asymmetric non-
electrophilic functionalizations of these dienes may,
however, be envisioned and are now under study in our
laboratory. In addition, the conversion of the syn-1,2-
diol moiety into useful functions such as 1,2-amino
alcohols7,21 or 1,2-thio alcohols22 has now been reported
and could well broaden the scope of our methodology.
was formed under these conditions and that according
to our expectations (as proved by difference NOE experi-
ments), dihydroxylation anti relative to the bulky silicon
group had occurred in a highly selective manner. The
carbon-silicon bond was then oxidized16 to afford the
allylic alcohol 6, diastereoisomerically pure and with 65%
ee.17
This moderate enantioselectivity was not surprising if
one considers the observations of Sharpless and co-
workers,7 who mentioned that Z-olefins and cyclic olefins
usually give poor results using the commercially available
AD-mix ((DHQD)2PHAL). Different AD-mixes were pre-
pared by varying the chiral ligands and the spacers as
described by these authors.7 Our results summarized in
Table 1 indicate that the best enantioselectivity was
obtained using (DHQ)2PYR, which correlates with the
observations made during asymmetric dihydroxylations
on closely related substrates.
With enantiomerically enriched 6 in hand, we pro-
ceeded to examine the functionalization of the remaining
double bond. Several routes were envisaged, and Sharp-
less epoxidation was retained due to the known efficiency
of this kinetic resolution process.18 Therefore, epoxida-
tion using the standard conditions ((-)-DET) gave the
expected syn epoxide 7 in a completely diastereoselective
fashion and 90% ee (Scheme 3). At the same time, by
Ack n ow led gm en t. We thank the Office Fe´de´ral de
l'Education et de la Science for financial support through
the COST (D2) program. We also thank Dr. S. Ainge
for helpful comments.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and spectral data for compounds 3a , 5, 6, 7, 1a , and
1b (3 pages).
(14) On addition of the chlorosilane to the reaction mixture (NH3-
THF), one could observe the immediate formation of a heavy precipi-
tate, presumably NH4Cl.
J O960814R
(15) The reaction between 3a and AD-mix was complete after 12 h
at 0 °C and led after careful extraction to the desired 1,2-diol in 65-
75% crude yield. The presence of more polar, intractable tetrols in the
aqueous phase cannot, however, be excluded.
(16) DMF was used instead of the standard THF-MeOH mixture,
which led to incomplete oxidation.
(17) Enantiomeric excesses were determined either by capillary GC
(Cyclodex-B), 1H NMR with Eu(hfc)3, or 1H NMR of Mosher's esters.
(18) Pfenninger, A. Synthesis 1986, 89.
(19) (a) Marshall, J . A.; Audia, V. H. J . Org. Chem. 1987, 52, 1106.
(b) Morgans, D. J ., J r.; Sharpless, K. B.; Traynor, S. G. J . Am. Chem.
Soc. 1981, 103, 462.
(20) Fu¨rst, A.; Plattner, P. A. Helv. Chim. Acta 1949, 32, 275.
(21) Lohray, B. B.; Gao, Y.; Sharpless, K. B. Tetrahedron Lett. 1989,
30, 2623.
(22) Nishimura, Y.; Umezawa, Y.; Adachi, H.; Kondo, S.; Takeuchi,
T. J . Org. Chem. 1996, 61, 480.