An approach to open-chain 1,5-stereocontrol using a silyl group

Article abstract of DOI:10.1039/b006515n

Nucleophilic attack by an ethylcuprate on the α,β-unsaturated ketone 3 takes place anti to the silyl group to give largely (96:4) a single product 4, fragmentation of which removes the silyl group, and reveals a pair of stereogenic centres having an open-chain 1,5 relationship 9.

Full text of DOI:10.1039/b006515n

An approach to open-chain 1,5-stereocontrol using a silyl group
Ian Fleming* and Chandrashekar Ramarao
Department of Chemistry, Lensfield Road, Cambridge, UK CB2 1EW. E-mail: if10000@cam.ac.uk
Received (in Liverpool, UK) 7th August 2000, Accepted 26th September 2000
First published as an Advance Article on the web 23rd October 2000
Nucleophilic attack by an ethylcuprate on the a,b-un-
saturated ketone 3 takes place anti to the silyl group to give
largely (96+4) a single product 4, fragmentation of which
removes the silyl group, and reveals a pair of stereogenic
centres having an open-chain 1,5 relationship 9.
This time we could clearly see the signals (1H-NMR) of the
minor isomer, which were identical to those we had already seen
for the isomer 4. Clearly, the reaction had been stereo-
chemically highly controlled, and either stereoisomer, 4 or 6,
could be obtained with nearly equal ease.
We hydrolysed the silyl enol ether 4, and obtained a single
diastereoisomer of the ketone 7, which was unaffected by
treatment with sodium methoxide, suggesting that it was the
thermodynamically favourable isomer (Scheme 3). We then
applied a fragmentation reaction developed by Nishiyama and
Itoh,10 using the oxime acetate 8, and obtained the alkene 9 with
the 1,5-relationship between C-3 and C-7 revealed in an open
chain. That the alkene had a trans double bond showed that the
relative stereochemistry assigned (COSY and NOESY) to C-2
in the ketone 7 was correct, since Itoh has shown that this type
of fragmentation is stereospecifically anti.
We have developed a substantial body of chemistry in which a
stereogenic centre carrying a silyl group controls the stereo-
chemistry of electrophilic attack on a CNC double bond in the
sense 1,1 and we have also adapted these reactions to control
We proved the relative stereochemistry between the 1,5-
related centres, by carrying out the same sequence using
enantiomerically pure (5R)-methylcyclohex-2-enone 1,11 and
obtained the nitrile 9 enantiomerically pure at C-3. Ozonolysis
and borohydride reduction gave 2-methylbutanol, derivatisation
of which with Mosher’s acid gave us the known esters (Scheme
4).12 The major component 10 was the R,R-diastereoisomer
(1H-NMR), and a minor component (4%) was now detectable
and identifiable as the R,S-diastereoisomer, allowing us to
measure, more accurately than it had been possible to from the
earlier NMR spectra, the degree to which we had been
successful in the conjugate addition step 3 ? 4.
more distant open-chain relationships, including 1,4 in the
course of syntheses of tetrahydrolipstatin2 and of nonactin,3 and
1,6 in an SE2B reaction.4 The use of a silyl group to control
1,4-relationships has also been exploited by Procter5 and by
Panek.6 We have now developed a method for the control of a
1,5-relationship, which we describe here. The essence of our
approach is successively to set up two 1,3-relationships, moving
the stereochemistry three atoms along the chain each time. By
using a silyl group in the middle, we can remove the intervening
functionality and stereochemistry to reveal the open-chain
1,5-relationship.
We suggest that, if the silyl group in the a,b-unsaturated
ketone 3 were equatorial 11, it would suffer from steric
compression with the methyl group on the exocyclic double
We began with the well established 1,3-control seen in
conjugate additions to 5-substituted cyclohexenones,7 which we
already knew was well behaved when the nucleophile was a
silyl group.8 On this occasion we used the phenyldimethyl-
silylzincate reagent9 with 5-methylcyclohex-2-enone
2
(Scheme 1), and the intermediate zinc enolate readily under-
went an aldol reaction with acetaldehyde to give a b-hydroxy
ketone as a mixture of diastereoisomers. Dehydration gave, as
far as we could tell, a single a,b-unsaturated ketone 3 with the
C-5 methyl and C-3 silyl groups trans to each other and the
exocyclic double bond with a Z configuration (COSY and
NOESY). This set up the first 1,3-relationship, and we set up the
second by conjugate addition of an ethylcuprate reagent in the
presence of tert-butyldimethylsilyl chloride. This gave a silyl
enol ether 4 as a single diastereoisomer (1H-NMR, > 95+5),
which proved (see below) to have the relative configuration
illustrated.
Scheme 2 Reagents and conditions: i, PhMe2SiZnMe2 Li, THF, 278 °C, 1
h; ii, EtCHO, 278 °C, 2 h; iii, MsCl, Et3N, CH2Cl2, 0 °C, 2 h; iv, DBU,
toluene, reflux, 4 h, or NaH, THF, 0 °C ? rt, 12 h; v, Me2CuLi LiCN, THF,
278 °C, 40 min; vi, TBDMSCl, HMPA, Et3N, 278 °C, 1 h.
We carried out the complementary sequence, trapping the
zinc enolate with propionaldehyde instead of acetaldehyde, and
adding a methylcuprate to the a,b-unsaturated ketone 5, which
gave largely (86+14) the alternative stereoisomer 6 (Scheme 2).
Scheme 1 Reagents and conditions: i, PhMe2SiZnMe2 Li, THF, 278 °C, 1
h; ii, MeCHO, 278 °C, 2 h; iii, MsCl, Py, CHCl3, reflux, 16 h; iv,
TBDMSCl, Et2CuLi LiCN, THF, 278 °C, 1 h.
Scheme 3 Reagents and conditions: i, HCl, H2O, THF, rt, 14 h; ii, NH2OH
HCl, Py, EtOH, reflux, 12 h; iii, Ac2O, Py, CH2Cl2, 0 °C ? rt, 4 h; iv,
Me3SiOTf, CH2Cl2, 0 °C, 4 h.
DOI: 10.1039/b006515n
Chem. Commun., 2000, 2185–2186
This journal is © The Royal Society of Chemistry 2000
2185

Scheme 4 Reagents and conditions: i, O3, Et2O, 278 °C, 10 min; ii, NaBH4,
H2O, 0 °C, 1 h; iii, Mosher’s R-acid, DCC, DMAP, CH2Cl2, rt, 16 h.
Scheme 5 Reagents and conditions: i, Me2CuLi, THF, 278 °C, 1 h; ii,
TBDMSCl, HMPA, Et3N, 278 °C ? rt, 1 h; iii, EtCHO, TiCl4, CH2Cl2
278 °C, 2 h; iv, TsOH, toluene, reflux, 1 h; v, Me2CuLi, THF, 278 °C, 1 h;
vi, TMSCl, HMPA, Et3N, 278 °C ? rt, 1 h.
adjacent to the stereogenic centre carrying a silyl group has
undergone nucleophilic attack. All our work in the past has
involved electrophilic attack. Since the sense of attack, anti to
the silyl group 1, is the same, it may be that we have here some
indication that the stereocontrol is largely steric in origin. This
conclusion stems from the prejudice that nucleophilic attack and
electrophilic attack would take place in opposite senses if purely
electronic effects were operative.
Although this may be a little too simplistic, we can be sure
that the silyl group is an important component in ensuring the
high levels of diastereocontrol that we have seen, for we have
carried out a similar set of reactions with the C-5 methyl group
and the C-3 silyl group in 5 interchanged (Scheme 5). Starting
with the silicon-containing cyclohexenone 16,16 and using a
methylcuprate to set up the 1,3-related centres in the enone 17.
This time, the conjugate addition of the methylcuprate gave
both possible diastereoisomers 18 in equal amounts. Pre-
sumably a methyl group on C-3, although surely held axial,
shields the bottom surface from attack to the same extent as the
upturned methyl group on the side chain shields the top
surface.
bond. In consequence, even though a trimethylsilyl group has a
larger A-value than a methyl group (2.513 and 1.74, re-
spectively), the lower-energy conformation of the enone 3
probably has the methyl group equatorial and the silyl group
axial 12, where, in any case, it only has one 1,3-diaxial
interaction. With the silyl group held on the lower surface,
nucleophilic attack can be expected to take place on the top
surface anti to the bulky group 12 (arrow), and hence give the
silyl enol ether with the relative configuration 4.
To support this argument, we carried out molecular model-
ling calculations (Macromodel), which confirmed that the
conformation with the silyl group axial 12 had the lowest
energy,14 with the stereo drawings 13 giving a more accurate
We thank Avra Laboratories, Hyderabad, for a maintenance
grant (C. R.), and Setu P. Roday for help with the molecular
modelling.
Notes and references
1 Summarised in: I. Fleming, J.. Chem. Soc., Perkin Trans. 1, 1992,
3363.
2 I. Fleming and N. J. Lawrence, J. Chem. Soc., Perkin Trans. 1, 1998,
2679.
3 M. Ahmar, C. Duyck and I. Fleming, J. Chem. Soc., Perkin Trans. 1,
1998, 2721; I. Fleming and S. K. Ghosh, J. Chem. Soc., Perkin Trans.
1, 1998, 2733.
4 I. Fleming and C. P. Leslie, J. Chem. Soc., Perkin Trans. 1, 1996,
1197.
picture of the minimised structure. Looking at the space-filling
versions 14 (from above) and 15 (from below) illustrates the
difference between the two surfaces of the enone 3—the b-
carbon (darker grey) in the top view is exposed, but in the
bottom view it is hindered by the substituents on the silyl
group.15
5 G. Procter, A. T. Russell, P. J. Murphy, T. S. Tan and A. N. Mather,
Tetrahedron, 1988, 44, 3953.
6 C. E. Masse and J. S. Panek, Chem. Rev., 1995, 95, 1293.
7 N. L. Allinger and C. K. Riew, Tetrahedron Lett., 1966, 1269. For the
most recent discussion, see S. Mori and E. Nakamura, Chem. Eur. J.,
1999, 5, 1534. For the first application in synthesis, see G. Stork, R. A.
Kretchmer and R. H. Schlessinger, J. Am. Chem. Soc., 1968, 90, 1647;
G. Stork, Pure Appl. Chem., 1968, 17, 383.
8 I. Fleming, R. Henning, D. C. Parker, H. E. Plaut and P. E. J. Sanderson,
J. Chem. Soc., Perkin Trans. 1, 1995, 317.
9 R. A. N. C. Crump, I. Fleming and C. J. Urch, J. Chem. Soc., Perkin
Trans. 1, 1994, 701.
10 H. Nishiyama, K. Sakuta, N. Osaka, H. Arai, M. Matsumoto and K. Itoh,
Tetrahedron, 1988, 44, 2413.
11 D. Caine, K. Procter and R. A. Cassell, J. Org. Chem., 1984, 49,
2647.
12 H. Oikawa, I. Matsuda, T. Kagawa, A. Ichihara and K. Kohmoto,
Tetrahedron, 1994, 50, 13347.
Furthermore, the calculation for the second a,b-unsaturated
ketone 5 revealed why it suffered a lower degree of stereo-
control in the conjugate addition step—the methyl group of the
ethyl group was oriented upwards, hindering the top surface
more than the top surface of the a,b-unsaturated ketone 3 = 12
= 13.
The stereocontrol in this system is interesting from two points
of view. In the first place, it shows that the silyl group is not so
hindering that it cannot adopt the axial position, and yet
simultaneously it does hinder the approach of the incoming
nucleophile. The long Si–C bond takes the silyl group far
enough away from the cyclohexane ring to make 1,3-diaxial
interactions less severe than they would be for a carbon-based
group, and yet the length of the Si–C bonds to the three other
substituents on the silicon atom causes them to occupy much of
the space below the double bond. In the second place, this is the
first reaction that we have studied in which the double bond
13 W. Kitching, H. A. Olszowy, G. M. Drew and W. Adcock, J. Org.
Chem., 1982, 47, 5153.
14 The chair conformation with the silyl group equatorial is 46.5 kJ mol21
higher in energy, and there are two families of boat conformations in
between, at 13 and 40 kJ mol21 above the minimum.
15 Although the calculation places the phenyl group under the carbonyl
group, a methyl group would be nearly as much hindrance at the b-
carbon; we do not think that the high degree of selectivity is dependent
upon the presence of the phenyl group on silicon.
16 M. Laguerre, J. Dunogue`s R. Calas and N. J. Duffaut, J. Organomet.
Chem., 1975, 93, C17; M. Asaoka, K. Shima and H. Takei, Tetrahedron
Lett., 1987, 28, 5669.
2186
Chem. Commun., 2000, 2185–2186