Geminal dicarboxylates as carbonyl surrogates for asymmetric synthesis. Part II. Scope and applications

Article abstract of DOI:10.1021/ja003775g

An enantioselective synthesis of allylic esters has been achieved by a novel asymmetric alkylation of allylic gem-dicarboxylates. The catalyst derived from palladium(0) and R,R-1,2-di(2′-diphenylphosphinobenzamido)cyclohexene efficiently induced the alkyl

Full text of DOI:10.1021/ja003775g

J. Am. Chem. Soc. 2001, 123, 3687-3696
3687
Geminal Dicarboxylates as Carbonyl Surrogates for Asymmetric
Synthesis. Part II. Scope and Applications
Barry M. Trost* and Chul Bom Lee
Contribution from the Department of Chemistry, Stanford UniVersity, Stanford, CA 94305
ReceiVed October 24, 2000
Abstract: An enantioselective synthesis of allylic esters has been achieved by a novel asymmetric alkylation
of allylic gem-dicarboxylates. The catalyst derived from palladium(0) and R,R-1,2-di(2′-diphenylphosphino-
benzamido)cyclohexene efficiently induced the alkylation process with a variety of nucleophiles to provide
allylic esters as products in good yield. High regio- and enantioselectivities were observed in the alkylation
with most nucleophiles derived from malonate, whereas a modest level of ee’s was obtained in the reactions
with less reactive nucleophiles such as bis(phenylsulfonyl)ethane. In the latter case, a slow addition procedure
proved effective, leading to significantly improved ee’s. The utility of the alkylation products was demonstrated
by several synthetically useful transformations including allylic isomerizations, allylic alkylations, and Claisen
rearrangements. Using these reactions, the chirality of the initial allylic carbon-oxygen bond could be transferred
to new carbon-oxygen, carbon-carbon, or carbon-nitrogen bonds in a predictable fashion with high
stereochemical fidelity. The conversion of gem-diesters to chiral esters by the substitution reaction is the
equivalent of an asymmetric carbonyl addition by stabilized nucleophiles. In conjunction with the subsequent
reactions that occur with high stereospecificity, allylic gem-dicarboxylates serve as synthons for a double
allylic transformation.
Introduction
transformation in asymmetric synthesis. In particular, catalytic
methods for performing such transformations would be a more
desirable solution for accessing allylic systems.
Allylic esters are exceptionally versatile synthetic building
blocks in organic chemistry. This feature originates from the
steric and electronic properties of the allylic setting which allows
for easy introduction of a new bond and provides strong
diastereofacial guidance to the addition reactions of the adjacent
alkene. Thus, allylic esters are useful in a variety of reactions
exemplified by metal-catalyzed allylic alkylations,1 allylic
transpositions,2 Claisen rearrangements,3 cycloaddition reac-
tions,4 epoxidations,5 and dihydroxylations.6 These processes
generally proceed with high stereochemical fidelity, thereby
transferring or propagating the chirality of the allylic center to
other location(s) within the molecule. Therefore, the enantio-
selective preparation of allylic esters constitutes an important
A classical approach to the construction of chiral allylic
systems involves an enzymatic7 or chemical8 resolution of
racemic allylic alcohols. Although these methods have found a
number of applications, they suffer from the limitation of a 50%
theoretical yield. A recent deracemization strategy using a Pd-
catalyzed asymmetric O-alkylation reaction is free of this
limitation at least for symmetrically substituted allylic systems.9
The more widely adopted strategies utilize carbonyl addition
reactions such as the asymmetric reduction10 or alkylation of
R,â-unsaturated systems. In particular, the asymmetric addition
of dialkylzinc provides more synthetic flexibility as the allylic
stereogenic center is created through an alkylation process.11
However, this process is critically dependent on the nature of
the nucleophile and, thus far, not feasible with stabilized
nucleophiles. A simple nucleophilic addition to R,â-unsaturated
aldehydes which would create such an allylic system is
complicated by various issues of selectivity. In addition to
(7) For reviews, see: (a) Faber, K. Biotransformations in Organic
Chemistry, 3rd ed.; Springer-Verlag: Berlin, 1997. (b) Wong, C. H.;
Whitesides, G. M. Enzymes in Synthetic Organic Chemistry; Pergamon:
Oxford, 1994; pp 41-130. (c) Schmid, R. D.; Verger, R. Angew. Chem.,
Int. Ed. 1998, 37, 1608. (d) Theil, F. Chem. ReV. 1995, 95, 2203.
(8) Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.; Ikeda, M.;
Sharpless, K. B. J. Am. Chem. Soc. 1981, 103, 6237.
(9) Trost, B. M.; Organ, M. G. J. Am. Chem. Soc. 1994, 116, 10320.
(10) Corey, E. J.; Heldal, C. Angew. Chem., Int. Ed. 1998, 37, 1987 and
references therein.
(1) (a) Yamamoto, Y. Angew. Chem., Int. Ed. Engl. 1986, 25, 508. (b)
Lipshutz, B. Synthesis 1987, 325. (c) Trost, B. M.; Verhoeven, T. R. In
ComprehensiVe Organometallic Chemistry; Wilkinson, G.; Stone, F. G. A.;
Abel, E. W., Eds.; Pergamon Press: Oxford, 1982; Vol. 8, Chapter 57, p
713. (d) Godleski, S. A. In ComprehensiVe Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon Press: New York, 1991; Vol. 4, Chapter
3.3.
(2) Overman, L. E. Angew. Chem., Int. Ed. Engl. 1984, 23, 579.
(3) For reviews of the Claisen rearrangement, see: (a) Ito, H.; Taguchi,
T. Chem. Soc. ReV. 1999, 28, 43. (b) Gajewski, J. J. Acc. Chem. Res. 1997,
30, 219. (c) Wipf, P. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Paquette, L. A., Eds.; Pergamon Press: Oxford, 1991; Vol. 5,
Chapter 7.2, pp 827-874. (d) Ziegler, F. E. Chem. ReV. 1988, 88, 1423.
(e) Rhoads, S. J. Org. React. 1975, 22, 1.
(4) (a) Roush, W. R. AdV. Cycloaddit. 1990, 2, 91. (b) Carruthers, W.
Cycloaddition Reactions in Organic Synthesis; Pergamon Press: Oxford,
1990. (c) Helquist, P. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Semmelhack, M. F., Eds.; Pergamon Press: Oxford, 1991; Vol.
4, Chapter 4.6.
(5) Rossiter, B. E. In Asymmetric Synthesis; Morrison, J. D., Ed.;
Academic Press: Orlando, FL, 1985; Vol. 5, p 193.
(11) (a) Oguni, N.; Omi, T.; Yamamoto, Y.; Nakamura, A. Chem. Lett.
1983, 841. Oguni, N.; Omi, T. Tetrahedron Lett. 1984, 25, 2823. Oguni,
N.; Matsuda, Y.; Kaneko, T. J. Am. Chem. Soc. 1988, 110, 7877. (b) Noyori,
R.; Suga, S.; Kawai, K.; Okada, S.; Kitamura, M.; Oguni, N.; Hayashi, M.;
Kaneko, T.; Matsuda, Y. J. Organomet. Chem. 1990, 382, 19. For reviews,
see: (c) Soai, K.; Niwa, S. Chem. ReV. 1992, 92, 833. (d) Dreisbach, C.;
Wischnewski, G.; Kragl, U.; Wandrey, C. J. Chem. Soc., Perkin Tans. 1
1995, 875.
(6) For reviews, see: (a) Cha, J. K.; Kim, N.-S. Chem. ReV. 1995, 95,
1761. (b) Kolb, H. C.; Van Nieuwenhze, M. S.; Sharpless, K. B. Chem.
ReV. 1994, 94, 2483.
10.1021/ja003775g CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/28/2001

3688 J. Am. Chem. Soc., Vol. 123, No. 16, 2001
Trost and Lee
differentiation of the two prochiral faces of the carbonyl group,
stabilized nucleophiles are prone to undergo a conjugate addition
due to the reversibility and unfavorable equilibrium of a carbonyl
addition. For these reasons, chiral induction by the carbonyl
addition of a stabilized nucleophile has not been developed, in
contrast to the well-documented examples utilizing unstabilized
nucleophiles.
An alternative strategy to the carbonyl addition evolves from
the recognition of a carbonyl group as the sum of two C-O
bonds.12 In this approach, the problem of differentiating the two
enantiotopic π-faces of a carbonyl group becomes that of
asymmetric substitution of either of the enantiotopic C-O bonds
of the acetal. While this concept has proved successful in the
chiral auxiliary-based approaches, such a transformation can also
be performed in a catalytic fashion by the asymmetric allylic
alkylation (AAA) using gem-dicarboxylates 2 (eqs 1 and 2). In
hydride as the base, and THF as the solvent. Thus, employing
these conditions, diacetate 4 was reacted at 0 °C in THF in the
presence of 1 mol % π-allylpalladium chloride (6) and 3 mol
% R,R-ligand 7. The nucleophiles were generated from a mixture
of 2.0-2.5 equiv of pronucleophile (5) and 1.5-2.0 equiv of
sodium hydride. As summarized in Table 1, the alkylations of
4 with a variety of nucleophiles furnished allylic esters 8 as a
single regioisomer in good yields with high ee’s. All of the
alkylation products were stable, easily handled, and readily
purified by simple chromatography. The ee was determined by
chiral shift studies on the corresponding desilylated alcohols 9
or chiral HPLC analysis. It is worthy of note that the alkyne,
acetal, carbamate, cyano, and phenylsulfonyl functionalities did
not interfere with the asymmetric process (entries 4-8). By
contrast, attempts to use di-tert-butyl methylmalonate as the
nucleophile failed to produce the alkylation product but rather
resulted in the recovery of unreacted 4 under various conditions.
Interestingly, strong dependence of the ee on the alkylation
procedure was noted (entry 8). The initial result of the alkylation
with sulfone derivative 5h was disappointing as only 80% ee
was obtained under the standard conditions. On the other hand,
performing the same reaction using a slow addition procedure
significantly improved the ee to 92%.
The results of the alkylation with the sulfone derivative led
us to examine the reactions of 5h with other substrates. Methyl-
and phenyl-substituted gem-diacetates 10 and 11 were subjected
to alkylation with 5h under the standard conditions (eq 4).
the Pd-catalyzed alkylation with malonate esters, the readily
available acylals serve as an “activated form” of a carbonyl
group for stabilized nucleophiles, giving rise to allylic ester 3
as the product in high ee.13,14 The chemoselectivity issue of the
carbonyl versus Michael additions now becomes a question of
regiocontrol in the nucleophilic addition to the π-allyl inter-
mediate. Thus, the Pd-catalyzed allylic substitution of gem-
dicarboxylates represents a novel method for the asymmetric
synthesis of allylic esters in which chiral induction in the
carbonyl addition of a stabilized nucleophile is achieved. Herein,
we report our investigation on the scope and limitation of the
alkylation process and the synthetic utility of the alkylation
products.
Results
Although both substrates provided 12 and 13 as single regioi-
somers in 52% (99% based upon recovered starting material,
brsm) and 90% yields, respectively, chiral shift studies on the
alkylation products revealed modest ee’s of 67% for 12 and
85% for 13. Considering the excellent level of ee’s that these
substrates afforded in alkylations with 5a and 5b, these results
seemed rather suprising. In particular, the ee of phenyl derivative
13 appeared even more puzzling since the alkylation of 11
invariably exhibited ee’s greater than 95% regardless of the
nucleophile.13,14 The factors affecting the ee in this case were
probed by using different sets of conditions. In the standard
procedure, the nucleophile was generated at 25 °C and added
to the mixture containing the substrate and catalyst at 0 °C.
However, this procedure resulted in immediate precipitation of
the nucleophile from the reaction mixture. Although increasing
the solubility of the nucleophile by using tetraalkylammonium
as the counterion or DBU as the base led to a homogeneous
reaction, these modifications resulted in even slower reactions
and large deterioration in ee. It was hoped at this point that a
homogeneous reaction still employing sodium as counterion
would give an efficient alkylation. Thus, after the nucleophile
was generated, the alkylation was immediately carried out using
a slow addition procedure, thereby taking advantage of the
kinetic solubility that keeps the reaction mixture homogeneous.
Asymmetric Alkylation. Testing the efficacy of various
nucleophiles began with the reactions of the straight-chain
derivative (4), which had also been used as a model substrate
in the previous study14 (eq 3). As established in our initial
studies, the best set of conditions employed gem-diacetate rather
than dipropionate or diisobutyrate as the substrate, sodium
(12) For use of chiral acetals as chiral auxiliaries, see: Johnson, W. S.;
Harbert, C. A.; Stipanovic, R. D. J. Am. Chem. Soc. 1968, 90, 5279. Johnson,
W. S.; Ratcliffe, B. E.; Stipanovic, R. D. J. Am. Chem. Soc. 1976, 98, 6188.
For leading references, see: Sammakia, T.; Smith, R. S. J. Am. Chem. Soc.
1992, 114, 10998. Ishihara, K.; Hanaki, N.; Yamamoto, H. J. Am. Chem.
Soc. 1993, 115, 10695. For a review, see: Alexakis, A.; Mangeney, P.
Tetrahedron: Asymmetry 1990, 1, 477.
(13) Trost, B. M.; Lee, C. B.; Weiss, J. M. J. Am. Chem. Soc. 1995,
117, 7247.
(14) See the preceding paper: Trost, B. M.; Lee, C. B. J. Am. Chem.
Soc. 2001, 123, 3671-3686.

AAA Using gem-Dicarboxylates, Part II
J. Am. Chem. Soc., Vol. 123, No. 16, 2001 3689
Table 1. Alkylations of 4 with Various Nucleophilesa
a All reactions were carried out with 1% 6 and 3% 7 in THF (0.2-0.5 M) at 0 °C unless otherwise noted. b Isolated yields. c Reference 14, the
preceding paper. d Determined by chiral shift studies on the corresponding desilylated alcohol 9 using (+)-Eu(hfc)3 in CDCl3. e Determined by
chiral HPLC using Chiralel OD or Chiralpak AD column with a 2-propanol-heptane mixture as eluent. f Slow addition procedure was applied.
g Reaction performed with S,S-ligand, ent-7. h Based upon the recovered starting material.
Indeed, this procedure increased the ee to 95%, albeit with a
moderate isolated yield of 54% mainly due to lower conversion
(80% brsm).
A cyclic nucleophile such as Meldrum’s acid derivative 16
was also tested for the alkylation (eq 5). The alkylation of the
derivative 18 was obtained in 83% yield with 85% ee from the
reaction of 15. It is of interest to note that no alkylation occurred
when 15 was subjected to the reaction with 5a in our previous
studies. The generally lower enantioselectivities using 16 as
nucleophile are typified by the alkylation of 4 that gave 19 in
86% yield with a much lower ee of 56%. This result, in
particular, provides a striking contrast to those obtained with a
variety of acyclic nucleophiles 5, in which a high level of ee
(87-93%) was uniformly observed (see Table 1).
While the ee was easily determined by chiral shift studies
on these alkylation products, derivatization of 17 to O-
methylmandelate 21 was carried out for further analysis (Scheme
1). The two-step sequence involving LAH reduction and
esterification provided tris-O-methylmandelate 21 uneventfully.
The signals from the two vinyl protons and the quaternary
methyl group of the (R)- and (S)-mandelates were completely
1
resolved in the H NMR spectrum, and integration of those
signals confirmed the measurement of 90% ee as determined
by chiral shift studies. Furthermore, the pattern of their chemi-
cal shifts was in full agreement with the reported trend
and thus established the absolute stereochemistry as depict-
ed.15
isopropyl-substituted substrate 14, which gave 95% ee in the
reaction with dimethyl methylmalonate (5a), provided 17 in 58%
yield with a somewhat decreased 90% ee. Similarly, TMS

3690 J. Am. Chem. Soc., Vol. 123, No. 16, 2001
Trost and Lee
Scheme 1. Determination of Enantioselectivity and Absolute Configuration
Scheme 2. Palladium-Catalyzed Allylic Transposition of 22
Scheme 3. Palladium-Catalyzed Allylic Transposition and Derivatization to Mandelates
Allylic Transposition. With the alkylation products in hand,
their allylic isomerization reactions were next investigated. The
equilibration reaction of allylic acetate 22 was performed using
PdCl2(CH3CN)2 as catalyst (Scheme 2).16 Toluene proved to
be the most efficient solvent as reactions in other solvents such
as benzene and THF required longer reaction times and failed
to achieve complete equilibrium. Thus, treatment of 22 (90%
ee) with 5% catalyst in refluxing toluene for 6 h generated an
inseparable mixture of 22 and 23 in a 9:91 ratio. As monitored
by GC, the ratio of the transposed acetate 23 rapidly increased
and slowly reached a constant value, at which point the reaction
was stopped. Separation was readily accomplished by hydrolysis
of the mixture to afford allylic alcohol 24 in 86% yield, while
starting acetate 22 was degraded to crotonaldehyde and dimethyl
methylmalonate (5a) by a retro-aldol process. Converting
alcohol 24 into O-methylmandelate ester 25 provided unam-
biguous determination of the ee and the absolute configuration.
Both the ee of 90% (corresponding to 100% chirality transfer,
ct) and the (S) configuration of the allylic center indicated that
the transposition reaction proceeded in a highly stereospecific
manner with conformity to suprafacial topology.
Following the same protocol, the n-propyl substituted 26 and
TBDPS derivative 8b were subjected to the transpostion reaction
(Scheme 3). The reaction of 26 (92% ee) led to the formation
of an 11:89 isomeric mixture of allylic acetates 26 and 27.
Cleavage of the acetyl group by treatment with methanolic
potassium carbonate provided allylic alcohol 29 in 87% yield
for two steps. On the other hand, isomerization of allylic acetate
8b (90% ee) under the same palladium catalysis conditions went
to completion. No starting material remained after 12 h reflux
in toluene, and only the transposed acetate 28 was isolated in
85% yield. Hydrolysis of the acetyl group followed by removal
of the silyl group with TBAF afforded diol 31 in 83% yield.
For analysis of the stereochemical outcome, alcohol 29 and diol
31 were converted to both (R)- and (S)-O-methylmandelate
esters in good yield.17 Determination of the de’s of the esters
1
by H NMR and chiral HPLC analyses gave 91-92% ee for
(15) Trost, B. M.; Belletire, J. L.; Godleski, S.; McDougal, P. G.;
Balkovec, J. M.; Baldwin, J. J.; Christy, M. E.; Ponticello, G. S.; Varga, S.
L.; Springer, J. P. J. Org. Chem. 1986, 51, 2370.
32 and 89-90% ee for 33 corresponding to 99-100% ct.
(17) Stadler, P. A. HelV. Chim. Acta 1978, 61, 1675.
(16) Overman, L. E.; Knoll, F. M. Tetrahedron Lett. 1979, 321.

AAA Using gem-Dicarboxylates, Part II
J. Am. Chem. Soc., Vol. 123, No. 16, 2001 3691
Scheme 4. Derivatization of Bismalonate 37 to Lactone 42
Analysis of the chemical shifts in the 1H NMR spectra of these
mandelate esters verified the absolute configuration.
The R-alkyl-R-amino acid derivative 34, which was available
from our total synthesis of sphingofungin F through the
alkylation of 4 with an azlactone nucleophile in 89% ee,18 was
also subjected to the transposition reaction in THF (eq 6).
regioisomers in good yields. While the use of either dppe or
dppp ligand gave satisfactory yields of the desired products,
no reaction was observed when triphenylphosphine was em-
ployed as ligand. Interestingly, the chiral ligand (S,S)-7, which
would be the matched ligand for this second ionization, was
found to be ineffective for the alkylation.
The complete sterochemical fidelity accompanied in the
substitution reaction was verified by chiral shift studies on 36
that revealed 91% ee (100% ct). However, various efforts to
determine the ee of the methyl analogue 37 did not provide
unambiguous values. Thus, the bis-alkylated product 37 was
further derivatized according to the sequence shown in Scheme
4. Bis-malonate 37 was treated with TBAF in refluxing THF
to give a mixture of γ-lactones 39 and 40. The formation of
these γ-lactones was readily noted by two carbonyl stretches at
1775 and 1736 cm-1 in the IR spectra. The major lactone 40
arose from dealkylative decarboxylation of 39 which was
initially generated from desilylated alcohol 38 followed by
lactonization. The relative stereochemistry of lactone 40 was
assigned as trans based on the large coupling constant (J )
11.4 Hz) between the two methine protons. The mixture of
lactones 39 and 40 was further dealkylated with lithium iodide
in refluxing 2,6-lutidine to provide an inseparable mixture of
the unconjugated (41) and conjugated (42) methyl esters in 2:1
ratio.20 Upon treatment with DBU, the â,γ-unsaturated ester 41
cleanly isomerized to R,â-unsaturated ester 42 as a single
diastereomer. Chiral HPLC analysis of 42 gave 87% ee,
corroborating the good degree of stereospecificity observed in
the second alkylation (at least 96% ct). It is worth noting that
the γ-lactone, a ubiquitous structural motif in various natural
products, could be prepared by this three-step sequence in a
highly enantio- and diastereoselective fashion.
Although the reaction gave a somewhat diminished 81:19
isomeric ratio, the rearranged acetate 35 could still be isolated
in a fairly good yield (79%). The 89% ee determined by chiral
shift studies demonstrated again the high stereospecificity
achieved in this process.
Allylic Alkylations. While the allylic transposition reaction
successfully achieved chirality transfer between the two allylic
C-O bonds, the Pd-catalyzed allylic substitution allowed for
the introduction of other types of carbon-carbon and carbon-
heteroatom bonds. Due to the highly stereospecific nature of
this process, a simple achiral ligand was sufficient for the
stereoselctive formation of a new bond.19 Allylic acetate 8a
(91% ee) was reacted with malonate-derived nucleophiles (eq
7). The reactions were performed by treating 8a with the sodium
The allylic acetate 8b (90% ee) was similarly alkylated with
dimethyl malonate to give bis-malonate 43 as a single product
in 91% yield (eq 8). Both the chiral shift experiments and chiral
salt of malonate esters in refluxing THF in the presence of 1%
6 and 3% dppp. Alkylations with both unsubstituted or methyl-
substituted malonate dimethyl esters proceeded cleanly to give
the corresponding bis-malonate derivatives 36 and 37 as single
HPLC analysis on the alkylation product uniformly afforded
(18) Trost, B. M.; Lee, C. B. J. Am. Chem. Soc. 1998, 120, 6818.
(19) (a) Tsuji, J. Tetrahedron 1986, 42, 4361. (b) References 1c and 1d.
(20) Elsinger, F. Organic Syntheses; Wiley & Sons, New York, 1973;
Collect. Vol. V, p 76.

3692 J. Am. Chem. Soc., Vol. 123, No. 16, 2001
Trost and Lee
Scheme 5. Allylic Alkylation with Phthalimide and Derivatization
90% ee. This stereochemical outcome, again, clearly demon-
strated the efficiency of chirality transfer from a C-O bond to
a new C-C bond by these Pd-catalyzed second allylic alkyla-
tions.
As was encountered in the alkylation with dimethyl malonate
(5a), the phenyl-substituted 44 (>95% ee) showed similar
aberration of optical purity in the phthalimidation (eq 10).
In contrast to the excellent results obtained from the alkyla-
tions of the TBDPS derived 8, similar alkylations of the phenyl-
substituted 44 gave rather poor stereochemical outcomes (eq
9). While the reaction of 44 (>95% ee) with dimethyl malonate
Despite the excellent yield of the reaction, the ee of allylic
phthalimide 48 was only 77% (81% ct). Variation of ligands,
counterions, catalyst concentrations, and solvents gave little
improvement. Given that other substrates could be alkylated
with high stereospecificity, the leakage of optical purity appears
to be only endemic to the phenyl-substituted substrate 44.
The introduction of a nitrogen functionality could also be
carried out using an alternative nucleophile such as sulfonamides
(eq 11). The Pd-catalyzed allylic amination of 26 (91% ee) with
using dppe as ligand gave an 81% yield of the desired alkylation
product 45, it required much longer reaction time and suffered
from severe loss of stereochemical integrity. The ee of 45 was
only 62% (65% ct) as determined by chiral shift experiments.
Various attempts to improve the stereospecificity by employing
different ligands, solvents, and bases were not fruitful.
The allylic alkylation was extended to include the introduction
of nitrogen nucleophiles. The reaction of allylic acetate 22 (90%
ee) with the cesium salt of phthalimide using dppp as ligand
and DME as solvent led to the formation of allylic phthalimide
46 as a single product in 91% yield (Scheme 5). The use of
THF as solvent or triphenylphosphine as ligand did not induce
the substitution process. Chiral HPLC analysis and chiral shift
experiments on 46 afforded ee’s of 89-90%, thereby attesting
to the complete chirality transfer from a C-O bond to a C-N
bond. To determine the absolute stereochemistry of the newly
formed C-N bond, allylic phthalimide 46 was converted to (S)-
N-phthaloyl alanine methyl ester (47). Ozonolysis of 46 ac-
cording to the known protocol21 directly generated R-amino acid
derivative 47 in 78% yield and 90% ee. Comparison of the
reported value for 47 (lit.22 [R]D ) -21.8) and the obtained
optical rotation, [R]D ) -20.2 (c 1.00, CHCl3), corroborated
the stereochemical assignment of both the initial asymmetric
and second alkylations.
the sodium salt of p-toluenesulfonamide was performed. The
initial reaction employing dppp as ligand and DME as solvent
proceeded sluggishly to provide allylic sulfonamide 49 in 85%
yield with 88% ee. The small loss of enantiopurity was ascribed
to the long reaction time which was due to the poor solubility
of the nucleophile in the reaction medium. Thus, the reaction
conditions were modified by employing cesium carbonate as
base, dppe as ligand, and a 1:1 THF-DMSO mixture as solvent.
Indeed, these modifications led to a rapid alkylation and
increased the yield to 87% and the ee to 91%.
(21) (a) Marshall, J. A.; Garafalo, A. W. J. Org. Chem. 1993, 58, 3675.
(b) Jumnah, R.; Williams, J. M.; Williams, A. C. Tetrahedron Lett. 1993,
34, 6619.
(22) (a) Kashima, C.; Okada, R.; Arao, H. J. Heterocycl. Chem. 1991,
28, 1241. (b) Shoda, S.; Mukaiyama, T. Chem. Lett. 1980, 391.

AAA Using gem-Dicarboxylates, Part II
J. Am. Chem. Soc., Vol. 123, No. 16, 2001 3693
Scheme 6. Claisen Rearrangement of Sulfones and Derivatization
Scheme 7. Claisen Rearrangement of 8h and Derivatization to Lactone 59
at δ 0.96, whereas the minor amide, syn-53, had the corre-
sponding signal at δ 1.01. The difference of the chemical shifts
was quite large in magnitude (∆δ ) δsyn-δanti ) +0.05 ppm),
and the shielding pattern clearly followed the proposed working
model, indicating an (S)-configuration for the C-3 methyl group
as depicted. Analysis of 54 was not possible due to the
1
incomplete resolution of the corresponding signals in the H
NMR spectrum.
Figure 1. H NMR shielding effect in syn- and anti-(S)-R-methyl-
benzylamide 53.
When the TBDPS-substituted 8h was subjected to the
rearrangement and subsequent desilylation, γ-lactone 56 was
generated in good yield presumably via hydroxyacid 55 (Scheme
7). Chiral shift studies on 56 using (+)-Eu(hfc)3 in C6D6
determined the ee to be 92%. Removal of the two phenylsulfonyl
groups with 2% sodium amalgam gave an inseparable mixture
of four alkene isomers in a 48:26:19:7 ratio. Subsequent
hydrogenation of 57 and 58 afforded â-butyl-γ-butyrolactone
(59) in 84% yield. As compared to the reported value, lit.24 [R]D
) +5.7 (c 1.73, CHCl3), the measured optical rotation of 59,
[R]D ) +4.6. (c 0.30, CHCl3), confirmed the absolute config-
uration. Therefore, this correlation study verified the absolute
stereochemistry of both the initial asymmetric alkylation product
and the stereochemical course of the subsequent Claisen
rearrangement.
Claisen Rearrangement. Studies on the Claisen rearrange-
ments were carried out using the substrates derived from the
alkylations of bis(phenylsulfonyl)ethane (5h) (Scheme 6).
Allylic acetates 12 (67% ee) and 13 (95% ee) were treated with
LDA or KHMDS and TBDMSOTf in THF at -78 °C. Upon
warming to room temperature, the generated silylketene acetals
50 rearranged smoothly to the silylesters, which were then
hydrolyzed to γ,δ-unsaturated acids 51 and 52 in 85% and 92%,
respectively. For analysis of the stereochemical outcome, these
acids were condensed with (S)-R-methylbenzylamine to give
53 and 54 both in 89% yields. 1H NMR integration of
appropriate signals readily revealed the dr of the amides and
thus the ee of acids 51 and 52. A 67% ee for 51 and a 95% ee
for 52, which were identical with those of the corresponding
starting acetate, were indicative of complete transfer of chirality
from C-O bonds to C-C bonds via the Claisen rearrangement.
In contrast to the sulfone-derived compounds, the malonate
derived ent-22 took rather a different course of reaction (eq 12).
1
Analysis of the H NMR spectra of amide 53 also provided
the establishment of the absolute configuration. A recent NMR-
based method has suggested a working model in which the
shielding of C-3 substituents by the aromatic ring in the chiral
amine moiety are different within a diastereomeric pair due to
the conformation of R-alkylbenzylamides (Figure 1).23 As a
result of this anisotropic effect, the C-3 methyl group in the
anti-isomer experiences an enhanced shielding effect and shows
up at higher field than in the syn-isomer. The C-3 methyl
chemical shift of the major amide, anti-53 (R ) CH3), appeared
(23) Hoye, T. R.; Koltun, D. O. J. Am. Chem. Soc. 1998, 120, 4638.

3694 J. Am. Chem. Soc., Vol. 123, No. 16, 2001
Trost and Lee
Scheme 8. Consecutive Allylic Alkylation of gem-Dicarboxylate
Subjection to the same conditions did not induce the rearrange-
ment but rather led to a complex mixture of products. When
LDA was employed as base in place of KHMDS, the lithium
enolate underwent carbonyl addition to give â-keto-δ-lactone
61 in 68% yield as a 5.8:1 diastereomeric mixture.
ity of a C-O bond to a different location of the molecule. The
good regioselectivities obtained from these allylic transposition
reactions appear to be reflective of the large steric differences
between the two allylic substituents. The neo-pentyl-like sub-
stituent at one allylic terminus provides a strong steric bias by
which the equilibration between the two allylic acetates under
palladium catalysis favors the transposed allylic acetates in a
high ratio. In the case of 8b, such steric differences between
the two allylic substitutents leads to exclusive formation of only
one regioisomer 28 in good yield. All rearrangements took place
with high suprafacial stereospecificity,27 as verified by the
analysis of derivatized O-methylmandelate esters. The regio-
preference of the second allylic alkylation that introduces a
nucleophile at the allylic carbon distal to the initially introduced
nucleophile moiety is in accord with that observed in the allylic
transposition reactions. It is worthwhile to note that simple
achiral ligands are sufficient to effect such transformations due
to the mechanism of the palladium(0)-catalyzed allylic substitu-
tion which leads to overall retention of stereochemistry.28 The
successful results from both carbon and nitrogen nucleophiles
invites the possibility of performing this second alkylation with
other types of nucleophiles. Through the overall transformation
to dialkylated products, gem-dicarboxylates become synthons
for a double allylic cation (Scheme 8).
The ability of allylic esters to undergo Claisen rearrangements
imparts more significance to the enantioselective method
providing such compounds. As evident from the comparison
of optical purity between the starting acetate and the rearranged
product, the [3,3]-sigmatropic process proceeds with complete
chirality transfer. The stereochemical course is best accounted
for by the six-membered transition state 50 in which the bulky
substituent is placed in an equatorial position. Of particular
interest is the sequence shown in Scheme 7 where the phenyl-
sulfonyl groups were easily removed after the desired transfor-
mation. The formation of 61 suggests that the lithium enolate
undergoes carbonyl addition faster than silylation. Thus, the
malonate moiety participated in a potentially useful, rare
condensation process to give a â-keto-δ-lactone structure.29
Discussion
The results of the present investigation demonstrate the
feasibility of constructing chiral allylic esters by the asymmetric
allylic alkylation (AAA) reaction. The ready availability and
considerable stability of the gem-dicarboxylates make this
strategy appealing in accessing an allylic system in enantio-
merically pure form. In particular, the various issues involved
in the addition of stabilized nucleophiles to R,â-unsaturated
aldehydes is efficiently resolved by the substitution process that
produces aldol-like adducts. In our early investigation, a range
of gem-diesters could be successfully employed as substrates,
although highly substituted systems exhibited some degree of
limitation.14 On the other hand, as shown in the present study,
the choice of nucleophile seems to be less restrictive. A variety
of nucleophiles serve well in the alkylation to provide the desired
product in good yields with high ee. Many functional groups
in the nucleophiles are compatible with the alkylation process
and may prove useful for further elaboration. Pertinent to this
result is our work on the use of azlactones and phenylsulfinate
nucleophiles that have been reported elsewhere.25,26
The lower ee that initially resulted from the alkylations of
bis(phenylsulfonyl)ethane (5h) appears to be the consequence
of its diminished reactivity as compared to that of the malonate-
derived nucleophiles. The fact that the enantiodiscriminating
event lies in the ionization step excludes the effect of the
nucleophile on the ee. However, when the ionization occurs
reversibly, this premise may not be valid. For example, when
the nucleophiles are slow to add to the π-allylpalladium
intermediate, the reversibility of ionization allows a mismatched
ionization to be competitive, thereby leading to erosion of ee.
In this case, the alkylation requires more efficient capture of
the kinetic intermediate. This demand is met by the use of a
slow addition procedure which allows the π-allylpalladium
complex to experience excess nucleophile. The significantly
higher ee’s obtained through the slow addition procedure are
consistent with the results of the previous study in which the
less reactive 5b, as compared to 5a, often required a slow
addition procedure for high enantioselectivity.14 These results
suggest the kinetic importance of the fast nucleophilic addition
as well as good chiral recognition in the ionization. Although
the ee was not optimized in the case of 10, it should be noted
that unlike 5a or 5b that gave a mixture of regioisomers, the
increased steric hindrance of 5h exhibited complete regio-
selectivity.
Conclusions
In summary, a novel approach for the synthesis of chiral
allylic esters has been developed using palladium-catalyzed
asymmetric allylic alkylation of gem-diesters. High enantio- and
regioselectivities have been obtained from the alkylation of
readily available gem-diesters with a variety of nucleophiles by
(24) Kosugi, H.; Tagami, K.; Takahashi, A.; Kanna, H.; Uda, H. J. Chem.
Soc., Perkin Trans. 1 1989, 935.
(25) (a) Trost, B. M.; Ariza, X. Angew. Chem., Int. Ed. Engl. 1997, 36,
2635. (b) ref18.
(26) Trost, B. M.; Crawley, M. L.; Lee, C. B. J. Am. Chem. Soc. 2000,
122, 6120.
The distinctive value of the alkylation of gem-dicarboxylates
is more clearly manifested by the synthetic versatility of the
resulting allylic esters which permits various subsequent
transformations with a high degree of stereospecificity. The
palladium(II)-catalyzed allylic isomerization transfers the chiral-
(27) Grieco, P. A.; Takigawa, T.; Bongers, S. L.; Tanaka, H. J. Am.
Chem. Soc. 1980, 102, 7587.
(28) Reference 19.
(29) (a) Brandaenge, S.; Flodman, L.; Norberg, A. J. Org. Chem. 1984,
49, 927. (b) Kocienski, P. J.; Narquizian, R.; Raubo, P.; Smith, C.; Boyle,
F. T. Synlett 1998, 869.

AAA Using gem-Dicarboxylates, Part II
J. Am. Chem. Soc., Vol. 123, No. 16, 2001 3695
(hexanes/ethyl acetate 15:1) afforded 24 as a colorless, sticky oil (0.217
g, 86%). The ee of 24 was determined by integration of appropriate
signals in the H NMR spectrum of the corresponding O-methylman-
a simple procedure. The unusual chiral induction process
achieves an equivalent of asymmetric carbonyl addition with
stabilized nucleophiles. The problems involved in addtion to
R,â-unsaturated aldehydes are effectively resolved through the
allylic substitution reaction that differentiates between two
geminal sp3-sp3 carbon-oxygen bonds. The present asym-
metric process in conjunction with the efficient transformations
of the obtained allylic esters provides a useful approach for
asymmetric synthesis.
1
delate 25: tR ) 6.51 min (GC); [R]D -0.11 (c 2.00, CHCl3); IR (film)
3430, 1732, 1665, 1454, 1440, 1376, 1258, 1112, 1066, 975, 947 cm-1
;
1H NMR (300 MHz, CDCl3): δ 6.05 (dd, J ) 15.9, 1.2 Hz, 1H), 5.62
(dd, J ) 15.9, 6.1 Hz, 1H), 4.37-4.28 (m, 1H), 3.70 (s, 3H), 3.69 (s,
3H), 2.28 (br s, 1H), 1.52 (s, 3H), 1.23 (d, J ) 6.4 Hz, 3H); 13C NMR
(75 MHz, CDCl3): δ 171.7, 171.6, 135.8, 127.5, 68.1, 55.1, 52.7 (2),
22.9, 20.0. HRMS Calcd for C9H13O5 (M+ - CH3): 201.0762. Found:
201.0753.
Dimethyl (E,S)-2,6-Bis(methoxycarbonyl)-2-methyl-5-t-butyl-
diphenylsilyloxy-3-heptenedioate, (36). To a test tube charged with
sodium hydride (95% powder, 10.0 mg, 0.417 mmol) and THF (0.5
mL) was added dimethyl malonate (65.2 mg, 0.493 mmol). After stirring
for 0.5 h, a solution of 6 (0.72 mg, 0.0020 mmol), dppp (2.44 mg,
0.0059 mmol), and 8a (0.101 g, 0.197 mmol, 91% ee) in THF (1 mL)
was added via cannula. The resulting mixture was heated under reflux
for 0.5 h, cooled to room temperature, poured into 10% NaHSO4 (30
mL), and extracted with ether (20 mL × 3). The combined organic
layers were washed with 10% NaHCO3 and brine, dried over anhydrous
MgSO4, and concentrated. Purification on a silica gel column (hexanes/
ethyl acetate 10:1) yielded 36 as a colorless, sticky oil (0.109 g, 95%).
Experimental Section
Dimethyl (2′E,1′S)-2-(1′-Acetoxy-4′-tert-butyldiphenylsilyloxy-2′-
butenyl)-2-(2′,2′,2′-trichloroethoxycarbonylamino)malonate, (ent-
8f). To a suspension of sodium hydride (95% powder, 54.5 mg, 2.16
mmol) in THF (2.5 mL) was added amidomalonate 5f (0.926 g, 2.83
mmol). The resulting mixture was stirred at ambient temperature until
hydrogen gas evolution ceased. To this solution was added a mixture
of π-allylpalladium chloride dimer (6) (5.3 mg, 0.014 mmol), ligand
S,S-7 (30.0 mg, 0.0434 mmol), and geminal diacetate 4 (0.610 g, 1.44
mmol) in THF (2.5 mL) via cannula at 0°. After stirring at 0° for 6 h,
the reaction mixture was poured into 10% aqueous NaHSO4 (20 mL)
and extracted with ether (15 mL × 3). The combined organic extracts
were washed with 10% aqueous NaHCO3 (15 mL) and brine, dried
over anhydrous MgSO4, and concentrated. Purification on a silica gel
column (hexanes/ethyl acetate 10:1) yielded ent-8f as a colorless, sticky
oil (0.908 g, 92%). The ee was 90% as determined by chiral shift studies
using (+)-Eu(hfc)3 in CDCl3: [R]D +6.4 (c 1.30, CHCl3); IR (film)
1752, 1497, 1432, 1254, 1223, 1112, 1047 cm-1; 1H NMR (300 MHz,
CDCl3): δ 7.69-7.65 (m, 4H), 7.47-7.36 (m, 6H), 6.28 (s,1H), 6.23
(d, J ) 5.2 Hz, 1H), 6.00-5.84 (m, 2H), 4.75 (d, J ) 12.0 Hz, 1H),
4.63 (d, J ) 12.0 Hz, 1H), 4.20 (d, J ) 2.0 Hz, 2H), 3.80 (s, 3H), 3.78
(s, 3H), 2.07 (s, 3H), 1.08 (s, 9H); 13C NMR (75 MHz, CDCl3): δ
168.8, 165.8, 165.4, 152.7, 135.4, 134.7, 133.3, 133.2, 129.7, 128.2,
127.6, 121.9, 95.2, 74.5, 73.8, 68.5, 63.3, 53.6, 53.5, 26.7, 20.8, 19.2.
HRMS calcd for C26H27NO9Si37Cl3 (M+ - t-C4H9): 636.0432. Found:
636.0441. Anal. Calcd for C30H36Cl3NO9Si: C, 52.29; H, 5.27; N, 2.03.
Found: C, 52.43; H, 5.18; N, 1.89.
1
The ee was 91% as determined by H NMR chiral shift study using
(+)-Eu(hfc)3 in CDCl3: [R]D +16.8 (c 1.45, CHCl3); IR (film) 1738
(br), 1434, 1263, 1112, 975 824, 743, 704 cm-1; 1H NMR (300 MHz,
CDCl3): δ 7.62-7.59 (m, 4H), 7.45-7.34 (m, 6H), 6.05 (d, J ) 15.9
Hz, 1H), 5.73 (dd, J ) 15.9, 9.4 Hz, 1H), 3.83 (d, J ) 8.9 Hz, 1H),
3.75-3.65 (m, 2H), 3.71 (s, 3H), 3.69 (s, 3H), 3.67 (s, 6H), 3.32-
3.02 (m, 1H), 1.52 (s, 3H), 1.04 (s, 9H); 13C NMR (75 MHz, CDCl3):
δ 171.5 (2), 168.8, 168.5, 135.6 (2), 135.3, 133.2, 131.6, 129.8, 129.0,
127.8, 64.5, 55.5, 52.6 (2), 52.4, 52.2, 45.1, 26.6, 20.2, 19.1. HRMS
Calcd for C30H37O8Si (M+ - CH3O): 553.2258. Found: 553.2248.
Anal. Calcd for C31H40O9Si: C, 63.68; H, 6.90. Found: C, 63.97; H,
6.81.
Methyl (E,S)-2-methyl-2-methoxycarbonyl-5-phthalimido-3-hex-
enoate, (46). To a suspension of phthalimide (16.8 mg, 0.114 mmol)
and Cs2CO3 (27.9 mg, 0.0856 mmol) in DME (0.25 mL) was added a
mixture of acetate 22 (14.7 mg, 0.0570 mmol, 90% ee), 6 (0.40 mg,
0.0011 mmol), and dppp (1.4 mg, 0.0034 mmol) in DME (0.25 mL) at
room temperature. The resulting mixture was heated under reflux for
1 h, diluted with ethyl acetate (10 mL), and washed with 2 N NaOH
(5 mL), water, and brine. The organic layer was then dried over
anhydrous MgSO4 and concentrated. Purification by flash chromatog-
raphy on a silica gel column (hexanes/ethyl acetate 3:1) afforded allylic
phthalimide 46 as a colorless, sticky liquid (18.0 mg, 91%). The ee of
46 was determined to be 90% by chiral shift studies using (+)-Eu-
(hfc)3 in CDCl3 and 89% by chiral HPLC analysis: [R]D +12.4 (c 1.20,
CHCl3); IR (film) 1737, 1711, 1613, 1459, 1441, 1385, 1256, 1112,
1023 cm-1; 1H NMR (300 MHz, CDCl3): δ 7.79-7.74 (m, 2H), 7.70-
7.65 (m, 2H), 6.14 (d, J ) 16.0 Hz, 1H), 6.02 (dd, J ) 16.0, 7.1 Hz,
1H), 4.98-4.88 (m, 1H), 3.70 (s, 3H), 3.67 (s, 3H), 1.52 (d, J ) 7.0
Hz, 3H) 1.52 (s, 3H); 13C NMR (75 MHz, CDCl3): δ 171.2, 171.1,
167.6, 133.8, 131.9, 130.3, 130.2, 123.1, 55.2, 52.8, 48.2, 20.0, 18.8.
HRMS Calcd for C18H19NO6 (M+): 345.1212. Found: 345.1211. Anal.
Calcd for C18H19NO6: C, 62.60; H, 5.55; N, 4.06. Found: C, 62.49;
H, 5.45; N, 4.05.
(2′E,1′R)-2,2,5-Trimethyl-5-(1′-acetoxy-3′-trimethylsilyl-2′-prope-
nyl)-1,3-dioxane-4,6-dione, (18). Following the procedure for ent-8f,
the reaction of 15 (0.1078 g, 0.468 mmol) with a mixture of 16 (0.150
g, 0.948 mmol) and sodium hydride (60% dispersion, 30 mg, 0.75
mmol) was carried out in the presence of 6 (1.7 mg, 0.0046 mmol)
and (R,R)-7 (10.0 mg, 0.0145 mmol) in THF (0.5 mL) at room
temperature for 5 h. Purification by silica gel chromatography (hexanes/
ethyl acetate 10:1) gave of 18 as a colorless oil (0.128 g, 83%), which
became crystalline upon storage in a refrigerator. The ee of 18 was
85% as determined by chiral shift studies using (+)-Eu(hfc)3 in
CDCl3: mp 79-80 °C (recryst. from pentane); [R]D -14.9 (c 1.63,
CHCl3); IR (film) 2957, 1754, 1454, 1381, 1291, 1249, 1217, 1092,
1056, 1023 cm-1; 1H NMR (300 MHz, CDCl3): δ 6.14-5.99 (m, 2H),
5.64 (d, J ) 6.0 Hz, 1H), 2.05 (s, 3H), 1.79 (s, 3H), 1.74 (s, 3H), 1.53
(s, 3H), 0.10 (s, 9H); 13C NMR (75 MHz, CDCl3): δ 168.9, 168.5,
167.3, 139.3, 137.0, 105.3, 81.0, 52.4, 29.7, 28.6, 21.1, 20.8, -1.5.
Anal. Calcd for C15H24O6Si: C, 54.86; H, 7.37. Found: C, 55.01; H,
7.20.
Dimethyl (1′E,3′S)-2-(3′-hydroxy-but-1′-enyl)-2-methylmalonate,
(24). To a solution of Pd(CH3CN)2Cl2 (15 mg, 0.058 mmol) in toluene
(45 mL) was added a solution of 22 (90% ee, 0.301 g, 1.16 mmol) in
toluene (5 mL). The resulting yellow solution was heated under reflux
for 12 h, at which point the GC isomeric ratio (tR (22) ) 6.82 min and
tR (23) ) 7.13 min) reached a constant value (8.7:91.3 ) 22:23). The
reaction mixture was then cooled to room temperature, filtered through
a short pad of silica gel, and concentrated to give a mixture of the two
isomeric acetates (0.301 g). The isomeric mixture was dissolved in
methanol (5 mL) and treated with K2CO3 (0.242 g, 1.75 mmol) at room
temperature for 0.5 h. The reaction mixture was poured into 10%
NaHSO4 (5 mL) and extracted with ether (5 mL × 3). The combined
organic layers were washed with 10% NaHCO3 and brine, dried over
anhydrous MgSO4, and concentrated. Purification on a silica gel column
(E,S)-3-Phenyl-6,6-bis(phenylsulfonyl)-4-heptenoic acid, (52). To
a mixture of KHMDS (0.5 M in toluene, 0.53 mL, 0.27 mmol) and
THF (2.0 mL) was added dropwise a solution of acetate 13 (95% ee,
0.107 g, 0.220 mmol) in THF (2.0 mL) at -78 °C. After 20 min,
TBDMSOTf (0.066 mL, 0.29 mmol) was added to the mixture via a
microliter syringe, and the resulting mixture was allowed to warm to
room temperature over for 2 h. The mixture was then heated at 65 °C
for 10 h, poured into 3 N H2SO4 (3 mL), stirred for 2 h, and extracted
with ethyl acetate (10 mL × 3). The combined organic phases were
dried over anhydrous MgSO4, concentrated, and purified on a silica
gel column (hexanes/ethyl acetate/acetic acid 50:50:0.5) to yield acid
52 as a white solid (0.098 g, 92%): mp 163-164 °C (recryst. ether);
[R]D +14.1 (c 1.25, CHCl3); IR (film) 3380, 1711, 1651, 1584, 1512,
1
1334, 1067, 982, 911 cm-1; H NMR (300 MHz, CDCl3): δ 9.90 (br

3696 J. Am. Chem. Soc., Vol. 123, No. 16, 2001
Trost and Lee
s, 1H), 7.90-7.87 (m, 2H), 7.71-7.65 (m, 3H), 7.58-7.50 (m, 3H),
7.36-7.23 (m, 5H), 7.04-7.00 (m, 2H), 5.90 (d, J ) 15.8 Hz, 1H),
5.80 (dd, J ) 15.8, 6.8 Hz, 1H), 3.87 (q, J ) 7.3 Hz, 1H), 2.71 (d, J
) 7.9 Hz, 2H), 1.72 (s, 3H); 13C NMR (75 MHz, CDCl3): δ 175.9,
142.4, 140.6, 136.7, 136.3, 134.5, 134.3, 131.1, 130.8, 128.9, 128.6,
128.5, 127.4, 127.2, 121.3, 87.3, 44.6, 39.3, 15.2. HRMS Calcd for
C19H19O4S (M+ - C6H5O2S): 343.1004. Found 343.1017.
graduate fellowship. Mass spectra were provided by the Mass
Spectrometry Regional Center of the University of California-
San Francisco, supported by the NIH Division of Research
Resources.
Supporting Information Available: Experimental proce-
dures for the preparation of all new compounds as well as
characterization data are included (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. We thank the National Science Founda-
tion and the National Institute of Health, General Medical
Sciences, for their generous support of our programs. C.B.L.
thanks Pharmacia & Upjohn for partial support through a
JA003775G