Approaching the C33-C38 fragments of amphotericin B and nystatin by a retro-[1,4]-Brook rearrangement and the stereoselective manipulation of the resulting allylsilane

Article abstract of DOI:10.1055/s-1998-1599

Sulfide S-4 (98.6% ee) furnished allylsilane S,S-anti-2 through a potassium naphthalenide induced, highly stereoselective retro-Brook rearrangement. Treatment with diethylborane followed by NaOOH/H₂O₂ delivered alcohol 9 as a single diastereomer. Cylization (→10), oxidation (→16), and methylation gave ketone 17. Reduction of 17 with Et₃SiH in CF₃CO₂H was accompanied by an unprecedented rearrangement of the neighboring oxasilolane ring to the C-silylated tetrahydropyran 20. It was oxidized to alcohol 21 which - if one inverted its OH group - would become a new C₃₃-C₃₈ building block for the polyol/polyene antibiotics amphotericin B and nystatin.

Full text of DOI:10.1055/s-1998-1599

February 1998
SYNLETT
201
Approaching the C -C Fragments of Amphotericin B and Nystatin by a Retro-[1,4]-Brook
33 38
Rearrangement and the Stereoselective Manipulation of the Resulting Allylsilane
‡
‡
✝
‡
Christoph Gibson , Thomas Buck , Martina Walker , Reinhard Brückner *
‡ Institut für Organische Chemie, Georg-August-Universität, Tammannstr. 2, D-37077 Göttingen, Germany
Fax: +49-551-392944; e-mail: rbrueck@gwdg.de
✝
Institut für Anorganische Chemie, Georg-August-Universität, Tammannstr. 4, D-37077 Göttingen, Germany
Received 6 November 1997
Abstract 2
4
: Sulfide S- (98.6% ee) furnished allylsilane S,S-anti-
We thought that the C -C (amphotericin numbering) array of said
33 38
through a potassium naphthalenide induced, highly stereoselective
retro-Brook rearrangement. Treatment with diethylborane followed by
compound 3 would be accessible from allylsilane S,S-anti-2. We have
recently obtained this silane − albeit racemic − by the potassium
9
naphthalenide induced retro-[1,4]-Brook rearrangement shown in
NaOOH/H O delivered alcohol as a single diastereomer. Cylization
2
2
7
10
16
17
Scheme 2.
(→ ), oxidation (→ ), and methylation gave ketone . Reduction of
Adapting this transformation for a synthesis of
optically
8
17
rearrangement of the neighboring oxasilolane ring to the C-silylated
20 21
active 3
4
with Et SiH in CF CO H was accompanied by an unprecedented
we had to prepare substrate as a pure enantiomer (Scheme 3).
3
3
2
tetrahydropyran
. It was oxidized to alcohol
which − if one
inverted its OH group − would become a new C -C building block
33 38
for the polyol/polyene antibiotics amphotericin B and nystatin.
1
2
Amphotericin B and nystatin are clinically important antimycotics
(Scheme 1). They are aminoglycosides and belong to the polyene/polyol
3
macrolides, an important class of antibiotics. Amphotericin B was
4
totally synthesized a decade ago but there are no laboratory syntheses
of nystatin or its aglycon. We were attracted by the possibility to derive
the two aglycons of Scheme 1 from precursors corresponding to the
non-identical "northern", the identical "eastern", the non-identical
"southern", and the identical "western" moieties of these structures,
respectively. Several years ago we published a synthesis of the
Scheme 2
5
1
"western"-part precursor . In the present communication, we describe
6
an approach to a building block for the "eastern" moiety. It is based on
Tiglyl alcohol (5) was obtained by reducing the commercially available
sec-butyl ester of E-2-methylbutenoic acid by LiAlH /AlCl . An
attempted displacement of the OH group of alcohol 5 with a mixture of
3
the perception that compound or a related molecule serves well such a
9
4
3
role.
10
Ph P and Ph S failed to give sulfide 6 because of a competing E→Z
3
2 2
isomerization and the occurrence of some substitution by an S ´
N
reaction. A better result was obtained by activating alcohol 5 as a
mesylate and treating the latter without isolation with sodium
phenylthiolate. The obtained sample of sulfide 6 was 95% pure. The
derived lithiosulfide ring-opened the styrene oxide R-7 (prepared from
11
R-mandelic acid with 98.6% ee ) at the less substituted side and
provided 63% of the hydroxylated allyl phenyl sulfide 8. Silylation with
12
13
tBuPh SiCl generated the silyl ether S-4. Reductive lithiation with
2
2.3 equiv. of potassium naphthalenide initiated the desired retro-Brook
rearrangement. It led to the allylsilane S,S-anti-2 in 94.5% yield after
removal of 3.8% of the syn-epimer by careful flash-chromatography on
14
silica gel.
With the allylsilane S,S-anti-2 in our hands we faced two unreconcilable
facts: (1) A tBuPh Si group cannot (could not yet) be oxidized to an OH
2
15
group be it allylic like in S,S-anti-2 or not; nonetheless, we were
obliged to realize such an oxidation at some point of our envisaged
access to the building block 3 or a similar compound. (2) No retro-[1,4]-
Brook rearrangement related to the S-4→S,S-anti-2 conversion but
placing an easier-to-oxidize silyl group in the resulting allylsilane
exhibited a high enough anti-selectivity to become synthetically useful
16
for approaching through them the compound 3 or similar compounds.
We therefore decided not to oxidize but to hydroborate our
17
rearrangement product. Based upon literature precedents we expected
that this reaction would work and reveal a considerable degree of
stereocontrol. Yet, compound S,S-anti-2 was essentially inert towards 9-
Scheme 1
BBN (70°C, THF, 1 d) or (thexyl)BH . The same was true for the
2
tBuMe Si ether derived from S,S-anti-2 with respect to added 9-BBN
2

202
LETTERS
SYNLETT
(atmospheric pressure or 10 kbar), BH •SMe or BHBr •SMe . We
3
2
2
2
18
finally found, that a THF solution of Et BH (6 equiv.) attacks
2
rearrangement product S,S-anti-2 at 15°C within 6 days: After oxidative
work-up by treatment with 10% NaOH / 15% H O we isolated the
2
2
oxidation product 9 in 85% yield (after separating 3% of the epimer by
14
flash-chromatography on silica gel ).
The stereostructure of compound 9 became evident after the next step:
19
An oxidation of the benzylic OH group succeeded with MnO
such
2
that the second OH group was not affected at all. Via the intermediacy
of ketone 12 we obtained the derived half-acetal 15 in 83% yield. We do
not know the configuration of the anomeric center of 15 but assume that
the OH group is axial to allow for anomeric stabilization. Independent
from that, the configurations of C-2 and C-3 follow from the
configuration of C-4 by a Karplus analysis of the J values in the most
vic
therefore abandoned the methylation in favor of pushing forward the
stable chair conformer of the six-membered heterocyclic ring: J
values ≥ 10 Hz are due to couplings between two axial protons while J
vic
oxidative removal of the tBuPh Si group. Heating a solution of diol 9,
2
vic
CsF, and KOEt in EtOH/THF at 65°C for almost 2 d was a first step in
this direction (Scheme 4): We displaced one Si-bound phenyl group by
incorporating the non-benzylic OH group and obtained the heterocycle
values << 7 Hz arise from couplings which include at least one
equatorial proton. As a consequence, the J values of Scheme 4 prove
vic
21
that compound 15 possesses the shown configurations of C-2 and C-3
vs. C-4 and constitutes conformer 15´. The same relative configurations
must, of course, be present in the diol precursor 9.
10 as a single stereoisomer in 81% yield. In this isomer, the tBu group
is oriented trans relative to the Ph-CH(OH)-CH substituent thus
2
causing as little steric hindrance as possible.
The following reaction was intended to be a diastereoselective
methylation of the enolate derived from hydroxyketone 12.
Unfortunately, this species did not form from the half-acetal 15 to an
For a change, two uneventful steps succeeded (Scheme 5). Firstly, we
oxidized alcohol 10 to 95% of ketone 16 by the Dess-Martin
22,23
periodinane.
Secondly, we methylated the lithium enolate of this
1
extent detectable by H-NMR spectroscopy. Accordingly, we tried to
ketone and observed the expected anti-preference with respect to the
24
25
trap the TMS or TBDMS ether of 12 by treating 15 with KH and the
respective chlorosilane. However, the only new products under these
conditions were the TMS protected acetal 14 and the TBDMS ether /
inducing C−Si bond (ds = 9:1). The major diastereomer 17 was
purified by recrystallization (56% yield). Its stereostructure was
26
established by an X-ray structure analysis.
20
TBDMS enolether 11, respectively. Accessing a silyl derivative of
27
To oxidize the C −Si bond of ketone 17 to the needed C −O bond
turned out impossible in our hands.
sec
sec
hydroxyketone 12 by a mono-methyldiphenylsilylation of diol 9
followed by an oxidation of the remaining OH group failed, too; this
28,29
We suspected that the alcohol
which we had tried to liberate would be prone to follow-up reactions
because it constitutes a β-hydroxyketone. Therefore, we decided to
was because the wrong OH group reacted (→ Ph MeSi-ether 13). We
2

February 1998
SYNLETT
203
oriented "down" to explain the overall stereochemical outcome. This
would imply that the hydride reduction 17→19 proceeds with high Cram
selectivity. Alternatively, the tetrahydropyran 20 and its equatorially
substituted stereocenter C-2 could form from precursor 18 by way of an
S 1 reaction, i. e. via a benzylic cation. If so, the stereochemistry of 20
N
would be the consequence either of product development control (if 20
stems from an irreversible reaction) or of thermodynamic control (if 20
forms under equilibrating conditions).
Gratifyingly, with compound 20 in our hands we were now able to
effect the C −Si → C −O oxidation which we had so long sought. It
sec
sec
was realized under conditions recently reported by Woerpel et al. for a
29d
related transformation
which is heating at 70°C in DMF in the
33
presence of KH, KF, and tBuOOH. We obtained alcohol 21 in the first
and so far only experiment in 76% yield.
If one inverts the OH group of compound 21 − e. g. by (1) oxidation to a
ketone, (2) hydrogenolysis of the benzylic C−O bond giving an open-
chain β-hydroxy-β´-phenylketone, and (3) chelation controlled C=O
34
bond reduction − one would obtain a correctly substituted precursor of
35
the "eastern" moieties of amphotericin B and nystatin. These reactions
would add to the 10 steps required at the present stage for transforming
the sec-butyl ester of E-2-methylbutenoic acid into the advanced
intermediate 21. Considering in addition the fair overall yield of 21
(12%) and the lengths of the already known syntheses of such
6
precursors we deem it worthwhile continuing to develop the approach
presented here.
Acknowledgment: We are very grateful for support of this work by the
Fonds der Chemischen Industrie and indebted to Wacker AG
(Burghausen) for a generous gift of tBuPh SiCl.
2
References and Notes
1.
Structure elucidation: P. Ganis, G. Avitabile, W. Mechlinski, C. P.
Schaffner, J. Am. Chem. Soc. 1971, 93, 4560-4564.
2.
Structure elucidation: J. M. Lancelin, F. Paquet, J. M. Beau,
1988, , 2827-2830; K. C. Nicolaou, K. H.
29
Tetrahedron Lett.
Ahn, Tetrahedron Lett. 1989, 30, 1217-1220; J. Prandi, J. M.
Beau, Tetrahedron Lett. 1989, 30, 4517-4520; J. M. Lancelin, J.
M. Beau, Tetrahedron Lett. 1989, 30, 4521-4524 and references to
earlier work cited in these publications.
3.
4.
S. Omura, H. Tanaka in Macrolide Antibiotics, S. Omura (ed.),
Academic Press, Inc., Orlando, 1984, pp. 351-404. J. M. Beau in
Recent Progress in the Chemical Synthesis of Antibiotics, G.
Lukacs, M. Ohno (Eds.), Springer Verlag, Berlin, 1990, pp. 132-
182.
K. C. Nicolaou, T. K. Chakraborty, Y. Ogawa, R. A. Daines, N. S.
Simpkins, G. T. Furst, J. Am. Chem. Soc. 1988, 110, 4660-4672;
K. C. Nicolaou, R. A. Daines, J. Uenishi, W. S. Li, D. P.
Papahatjis, T. K. Chakraborty, J. Am. Chem. Soc. 1988, 110,
4672-4685; K. C. Nicolaou, R. A. Daines, T. K. Chakraborty, Y.
Ogawa, J. Am. Chem. Soc. 1988, 110, 4685-4696; K. C. Nicolaou,
R. A. Daines, Y. Ogawa, T. K. Chakraborty, J. Am. Chem. Soc.
remove the potentially harmful C=O group from substrate 17. Doing so
30
by an ionic hydrogenation with Et SiH in CF CO H or with the same
3
3
2
31
reductant and BF •Et O we isolated instead of the expected benzylic
alcohol the silylated tetrahydropyran 20 in 88% and 53% yield,
3
2
32
respectively (Scheme 5). The scheme indicates a plausible formation
1988,
, 4696-4705. Formal total synthesis: R. M. Kennedy, A.
110
Abiko, T. Takemasa, H. Okumoto, S. Masamune, Tetrahedron
1988, , 451-454 and literature cited therein.
mechanism of 20 and compiles the J values observed within its six-
vic
membered heterocycle. From them, we deduced the shown
configuration plus the favorite conformation (= chair 20´) by the same
kind of Karplus analysis detailed above in the context of the structure
elucidation of tetrahydropyran 15. The newly established stereocenter
C-2 of compound 20 bears an equatorial phenyl substituent. If 20
Lett.
29
5.
6.
R. Brückner, Tetrahedron Lett. 1988, 29, 5747-5750.
Other approaches to the "eastern" or C -C segment of the
aglycons of Scheme 1: S. Hanessian, S. P. Sahoo, M. Botta,
Tetrahedron Lett.
Takemasa, Y. Nishitani, S. Masamune, Tetrahedron Lett. 1985,
26, 5239-5242; K. C. Nicolaou, D. P. Papahatjis, D. A. Claremon,
33 38
1987,
28, 1147-1150; D. Boschelli, T.
originates from an S 2 reaction within precursor molecule 18, the C −O
N
b
bond there and the C −O bond in the preceding intermediate 19 must be
a

204
LETTERS
SYNLETT
R. L. Magolda, R. E. Dolle, J. Org. Chem. 1985, 50, 1440-1456;
S. Takano, Y. Shimazaki, Y. Sekiguchi, K. Ogasawara, Chem.
Lett. 1988, 2041-2044; M. Kinoshita, H. Takami, M. Taniguchi,
T. Tamai, Bull. Chem. Soc. Jpn. 1987, 60, 2151-2161; G. J.
McGarvey, J. M. Williams, R. N. Hiner, Y. Matsubara, T. Oh,
J. Am. Chem. Soc. 1986, 108, 4943-4952.
24. I. Fleming, J. H. M. Hill, D. Parker, D. Waterson, J. Chem. Soc.,
Chem. Commun. 1985, 318-321; R. A. N. C. Crump, I. Fleming, J.
H. M. Hill, D. Parker, N. L. Reddy, D. Waterson, J. Chem. Soc.,
Perkin Trans.
1
1992, 3277-3294. First diastereoselctive
alkylation of an ester enolate with a β-positioned oxasilolane ring:
M. R. Hale, A. H. Hoveyda, J. Org. Chem. 1994, 59, 4370-4374.
7.
8.
C. Gibson, T. Buck, M. Noltemeyer, R. Brückner, Tetrahedron
Lett. 1997, 38, 2933-2936.
25. (2S)-2-[(2S,3S,4S,5S)-(2-tert-Butyl-4,5-dimethyl-2-phenyl-[1,2]-
oxasilolan-3-yl)]-1-phenyl-1-propanone (17): Mp. 87-91°C.−
25
[α]
= -130 (c = 0.98 in CH Cl ).− IR (in CDCl ): ν = 2985
2 2 3
All new compounds gave correct combustion analyses (sole
D
-1
1
cm , 2890, 1680, 1595, 1450, 1375, 1295, 1210, 1115, 1020, 935,
exception: 21) and satisfactory H NMR (300 MHz, CDCl ), and
3
1
820, 730.− H NMR (300 MHz, CDCl ): δ = 0.86 (d, J
= 6.7
IR spectra (film or CDCl solution).
3
4'-Me,4'
3
Hz, 4'-CH ), 0.98 (s, SitBu), 1.23 (d, J
= 7.2 Hz, 2-CH ),
3
3
2-Me,2
9.
A. E. Gastaminza, N. N. Ferracutti, N. M. Rodriguez, J. Org.
Chem. 1984, 49, 3859-3860.
1.36 (d, J
= 6.4 Hz, 5'-CH ), 2.06-2.20 (m, 3'-H, 4'-H), 3.59
3
5'-Me,5'
(dq, J = 10.6 Hz, J
= 7.0 Hz, 2-H), 4.21 (q, J
= 6.4
2,3'
2,2-Me
5',5'-Me
10. T. Cohen, B.-S. Guo, Tetrahedron 1986, 42, 2803-2808.
11. C. Dupin, J. F. Dupin, Bull. Soc. Chim. France 1970, 249-251.
12. S. Hanessian, P. Lavallée, Can. J. Chem. 1975, 53, 2975-2977.
Hz, 5'-H), 7.05-7.15 (m, 2 Ar-H), 7.20-7.28 (m, 3 Ar-H), 7.47-
7.55 (m, 2 Ar-H), 7.58-7.65 (m, 1 Ar-H), 7.94-8.00 (m, 2 Ar-H).−
13
APT C NMR [75 MHz, CDCl ; methyl and methine carbons up
3
("+"), methylene and quaternary carbons down ("-")]: δ = "+"
13. This method was reviewed by T. Cohen, M. Bhupathy, Acc.
Chem. Res. 1989, 22, 152-161; M. Yus, Chem. Soc. Rev. 1996,
155-161.
16.90, "+" 19.18, "+" 23.37, and "+" 26.90 (4'-CH , 2-CH , 5'-
3
3
CH and C-3'), "-" 18.29 [SiC(CH ) ], "+" 26.40 [SiC(CH ) ],
3
3 3
3 3
"+" 38.39 and "+" 40.90 (C-4', C-2'), "+" 80.08 (C-5'), "+" 127.29,
"+" 128.52, and "+" 128.70 (o- and m-C's of 1-Ph and m-C's of Si-
Ph), "+" 128.82 and "+" 133.01 (p-C's of Si-Ph and 1-Ph), "+"
134.15 (o-C's of Si-Ph), "-" 135.77 and "-" 136.04 (ipso-C's of Si-
14. Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923-
2925.
15. Allylsilane oxidations to allyl alcohols are presently limited to the
1
2
SiR
R
groups described by I. Fleming, S. B. D. Winter,
2
Ph and 1-Ph), "-" 204.54 (C-1).− C
H O Si (380.6): calcd. C
24 32 2
Tetrahedron Lett. 1993, 34, 7287-7290; M. C. Norley, P. J.
Kocienski, A. Faller, Synlett 1994, 77-78; I. Fleming, S. B. D.
Winter, Tetrahedron Lett. 1995, 36, 1733-1734; R. Angelaud, Y.
Landais, C. Maignan, Tetrahedron Lett. 1995, 36, 3861-3864. For
an alkoxy allylsilane oxidation to an allyl alcohol cf. R.-M. Chen,
W.-W. Weng, T.-Y. Luh, J. Org. Chem. 1995, 60, 3272-3273.
75.74, H 8.47; found C 75.77, H 8.37.
26. Crystal data of 17: C O Si, M = 380.58 monoclinic, space
H
24 32
2
group P2 /n, a = 962.3(2), b = 1890.2(2), c = 1197.2(7) pm, α =
1
3
90°, β = 95.45(2)°, γ = 90°, V = 2.1678(6) nm , Z = 4, D = 1.166
Mg/m , F(000) = 824, crystal dimensions 0.8 × 0.8 × 0.8 mm.−
c
3
Further details of the crystal structure investigation may be
obtained from the Fachinformationszentrum Karlsruhe, D-76344
Eggenstein-Leopoldshafen (Germany), on quoting the depository
number CSD-408041.
16. C. Gibson, Diplomarbeit, Universität Göttingen, 1996.
17. I. Fleming, N. J. Lawrence, Tetrahedron Lett. 1988, 29, 2077-
2080; J. Chem. Soc., Perkin Trans. 1 1992, 3309-3326.
18. F. Langer, A. Devasagayaraj, P.-J. Chavant, P. Knochel, Synlett
1994, 410-412.
27. For reviews of the oxidation of silanes, see: G. R. Jones, Y.
Landais, Tetrahedron 1996, 52, 7599-7662; I. Fleming,
Chemtracts: Org. Chem. 1996, 9, 1-64.
19. E. Adler, H.-D. Becker, Acta Chem. Scand. 1961, 15, 849-852.
20. We failed to methylate compound 11 by the MeI/AgOAc method
of C. W. Jefford, A. W. Sledeski, P. Lelandais, J. Boukouvalas,
Tetrahedron Lett. 1992, 33, 1855-1858.
28. We tried procedures described by K. Tamao, T. Yamauchi, Y. Ito,
Chem. Lett. 1987, 171-174; S. H. Bergens, P. Noheda, J. Whelan,
B. Bosnich, J. Am. Chem. Soc. 1992, 114, 2121-2128; M.
Murakami, M. Suginome, K. Fujimoto, H. Nakamura, P. G.
Andersson, Y. Ito, J. Am. Chem. Soc. 1993, 115, 6487-6498.
21. 1S-2-[(2S,3S,4S,5S)-(2-tert-Butyl-4,5-dimethyl-2-phenyl-[1,2]-
25
oxasilolan-3-yl)]-1-phenyl-1-ethanol ( ): [α]
10
= -11.2 (c =
D
-1
2.08 in CH Cl ).− IR (film): ν = 3415 cm , 3065, 2930, 1960,
29. Bulky silanes or silylethers which have been oxidized to alcohols
2
2
1680, 1595, 1460, 1375, 1325, 1200, 1110, 1065, 930, 825, 760,
include: a) alkyl−SiPh (H.-J. Knölker, G. Wanzl, Synlett 1995,
3
1
705.− H NMR (300 MHz, CDCl ): δ = 0.89 (d, J
= 7.1 Hz,
378-382); b) alkyl−SiMe trityl (G. P. Brengel, A. I. Meyers, J.
3
4'-Me,4'
2
4'-CH ), 0.99 (s, SitBu), 1.25 (d, J
172 (m, 2-H *, 2-H *), 1.80 (br. s, 1-OH)**, overlaps partly with
1.75-ca. 1.93 (m, 3'-H*), 1.99 (ddq, J = J = J = 6.7
= 6.4 Hz, 5'-CH ), 1.57-
3
Org. Chem.
1996
, 61, 3230-323); c) alkyl−SiPh Me (J. H.
3
5'-Me,5'
2
1
2
Smitrovich, K. A. Woerpel, J. Org. Chem. 1996, 61, 6044-6046);
d) (sec-alkyl)−Si(OR)(tBu) (P. M. Bodnar, W. S. Palmer, J. T.
4',5'
4',3'
4',4'-Me
2
Hz, 4'-CH ), 3.93 (dq, J
= J
= 6.1 Hz, 5'-H), 4.50 (br. dd,
Shaw, J. H. Smitrovich, J. D. Sonnenberg, A. L. Presley, K. A.
3
5',4'
5',5'-Me
J
= 8.1 Hz, J
= 4.4 Hz, 1-H), 6.96-7.02 (m, 2 Ar-H),
Woerpel, J. Am. Chem. Soc. 1995, 117, 10575-10576).
1,2-H(1)
1,2-H(2)
7.18-7.46 (m, 6 Ar-H), 7.62-7.72 (m, 2 Ar-H); *assignment
interchangeable; **exchangeable with D O.− O Si
(368.6): calcd. C 75.00, H 8.75; found C 75.05, H 8.78.
30. F. A. Davis, C. Clark, A. Kumar, B.-C. Chen, J. Org. Chem. 1994,
59, 1184-1190.
C
H
23 32 2
2
31. M. P. Doyle, C. T. West, S. J. Donnelly, C. C. McOsker, J.
Organomet. Chem. 1976, 117, 129-140.
22. D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155-4156; S.
D. Meyer, S. L. Schreiber, J. Org. Chem. 1994, 59, 7549-7552.
32. (6S,5S,4R,3S,2S)-4-(tert-Butylhydroxyphenylsilyl)-3,4,5,6-
23. Attempted Baeyer-Villiger oxidations of ketone 16 to the
corresponding phenylester (cf. phenylester 3, Scheme 1) failed
25
tetrahydro-2,3,5-trimethyl-6-phenyl-2H-pyran
(
): [α]
=
20
D
-1
+37.2 (c = 1.08 in CH Cl ).− IR (CDCl ): ν = 3480 cm , 2930,
2
2
3
with MCPBA / NaHCO (method: H. Kaga, S. Kobayashi, M.
1
3
2855, 1455, 1385, 1110, 1060, 1010, 910, 820, 735.− H NMR
(300 MHz; CDCl ): δ = 0.56 (d, J = 6.4 Hz, 3-CH )*, 0.91
Ohno, Tetrahedron Lett. 1988, 29, 1057-1060) or urea / H O /
2
2
3
3-Me,3
3
Na HPO (method: M. S. Cooper, H. Heaney, A. J. Newbold, W.
2
4
(d, J
= 6.8 Hz, 5-CH )*, 0.98 (s, SitBu), 1.26 (d, J
=
5-Me,5
3
2-Me,2
R. Sanderson, Synlett 1990, 533-535).
6.1 Hz, 2-CH )*, 1.64 (dd with very strong roof effect in the
3

February 1998
SYNLETT
205
*
direction of the large coupling, J = 11.7 Hz, J = 2.7 Hz, 4-H),
1.68-1.80 [m, 3-H (superimposed at high- and low-field side by 4-
H and Si-OH, respectively)], 1.82 (br. s, Si-OH)**, 2.35 (qdd,
CH ) , 1.49 (ddq, J = J = 10.2 Hz, J
= 6.4, 3-H), low-
4,3
4,5
3
3,2
3,4
3,3-Me
**
field part superimposed by 1.56 (br. s, 4-OH) , 2.19 (qdd, J
5,5-Me
= 7.1 Hz, J = 4.9 Hz, J = 2.2 Hz, 5-H), 3.26 (dq, J = 9.8
5,4
5,6
2,3
J
J
= 6.8 Hz, J = J = 2.3 Hz, 5-H), 3.25 (dq, J = 9.1 Hz,
Hz, J
4-H), 4.58 (d, J
= 6.1 Hz, 2-H), 3.65 (dd, J = 10.5 Hz, J = 4.9 Hz,
5,5-Me
5,4
5,6
2,3
2,2-Me 4,3 4,5
= 6.0 Hz, 2-H)***, 4.66 (d, J = 1.9 Hz, 6-H)***, 7.18-
= 2.2 Hz, 6-H), 7.18-7.36 (m, 5 Ar-H);
2,2-Me
6,5
6,5
*
7.26 (m, 1 Ar-H), 7.29-7.39 (m, 7 Ar-H), 7.57-7.65 (m, 2 Ar-H);
*assignment interchangeable; **exchangeable with D O;
assignment based on the occurrence of the required cross-peak in
**
13
a H,H-COSY spectrum; exchangeable with D O.− APT
C
2
2
***signal unchanged after H,D exchange with D O − as required
NMR [75 MHz, CDCl ; methyl and methine carbons up ("+"),
methylene and quaternary carbons down ("-")]: δ = "+" 5.83 (5-
2
3
13
for the suggested structure.− APT C NMR [75 MHz, CDCl ;
3
*
*
*
methyl and methine carbons up ("+"), methylene and quaternary
carbons down ("-")]: δ = "+" 11.60, "+" 19.00, "+" 20.19, "+"
CH ) , "+" 13.53 (3-CH ) , "+" 19.52 (2-CH ) , "+" 39.34 (C-
3 3 3
**
**
***
***
3) , "+" 40.93 (C-5) , "+" 77.04 (C-4) , "+" 78.26 (C-2)
"+" 79.62 (C-6) , "+" 125.59 and "+"128.13 (o-C’s and m-C’s
,
***
33.01, "+" 36.22, and "+"36.68 (C-4, 3-CH , 5-CH , 6-CH , C-3,
3
3
3
C-5), "-" 19.44 [SiC(CH ) ], "+" 27.32 [SiC(CH ) ], "+" 81.62,
of Ph), "+"126.76 (p-C of Ph), "-" 141.32 (ipso-C von Ph);
3 3
3 3
* ** ***
and "+" 83.06 (C-2, C-6), "+" 125.43, "+" 127.58, and "+" 127.92
(o-C's of 2-Ph, m-C's of 2-Ph, m-C's of Si-Ph), "+" 126.47 and "+"
128.88 (p-C's of 2-Ph and Si-Ph), "+" 133.64 (o-C's of Si-Ph), "-"
,
,
assignment based on the occurrence of corresponding
cross-peaks in a C,H correlation spectrum.
34. syn-Reduction of β-hydroxyketones: K. Narasaka, F.-C. Pai,
Tetrahedron 1984, 40, 2233-2238.
138.89 and "-" 142.22 (ipso-C's of 2-Ph and Si-Ph).− C
H O Si
24 34 2
(382.6): calcd. C 75.34, H 8.96; found C 75.39, H 8.94.
35. The degradation Ph-CH - → RO C- is conceivable in the
2
2
33. (2S,3R,4R,5R,6S)-3,4,5,6-Tetrahydro-2,3,5-trimethyl-6-phenyl-
following steps: (1) ozonolysis (Ph-CH -
→ HO C-CH -)
2 2
2
25
2H-4-pyranol (21): Mp. 90-92°C.− [α]
= 9.98 (c = 1.64 in
D
followed by spontaneous formation of a γ-lactone with
hydroxylated side-chain; (2) treatment of this lactone with 1
-1
CHCl ).− IR (KBr): ν = 3355 cm , 2965, 2925, 2890, 2850, 2360,
3
1
1635, 1450, 1385, 1315, 1150, 1115, 1060, 1030, 930, 755.−
H
*
equiv. each of LDA and tBuMe SiCl; (3) ozonolysis of the
2
NMR (300 MHz; CDCl ): δ = 0.69 (d, J
= 7.1 Hz, 5-CH ) ,
3
5-Me,5
3
resulting silylketene acetal.
*
1.01 (d, J
= 6.4 Hz, 3-CH ) , 1.33 (d, J
= 6.0 Hz, 2-
3-Me,3
3
2-Me,2