Highly diastereoselective 1,4-addition of an organocuprate to methyl α-D-gluco-, α-D-manno-, or α-D-galactopyranosides tethering an α,β-unsaturated ester. Novel asymmetric access to β-C-substituted butanoic acids

Article abstract of DOI:10.1021/jo0101860

The 1,4-addition of magnesium divinylcuprate prepared from vinylmagnesium bromide and cuprous bromide to some 4-O-crotonyl derivatives of methyl α-D-glucopyranoside proceeds with a high level of diastereochemical induction, providing the adduct in good-to

Full text of DOI:10.1021/jo0101860

J . Org. Chem. 2001, 66, 5965-5975
5965
High ly Dia ster eoselective 1,4-Ad d ition of a n Or ga n ocu p r a te to
Meth yl r-D-Glu co-, r-D-Ma n n o-, or r-D-Ga la ctop yr a n osid es
Teth er in g a n r,â-Un sa tu r a ted Ester . Novel Asym m etr ic Access to
â-C-Su bstitu ted Bu ta n oic Acid s
Kiichiro Totani,† Takayuki Nagatsuka,† Shuhei Yamaguchi,† Ken-ichi Takao,†
Shigeru Ohba,‡ and Kin-ichi Tadano*,†
Departments of Applied Chemistry and Chemistry, Keio University,
Hiyoshi, Kohoku-ku, Yokohama 223-8522, J apan
tadano@applc.keio.ac.jp
Received February 20, 2001
The 1,4-addition of magnesium divinylcuprate prepared from vinylmagnesium bromide and cuprous
bromide to some 4-O-crotonyl derivatives of methyl R-D-glucopyranoside proceeds with a high level
of diastereochemical induction, providing the adduct in good-to-excellent yields. Other organo-
cuprates also serve as effective carbon nucleophiles for the 1,4-addition. Removal of the carbohydrate
moiety from each adduct afforded a variety of â-C-substituted butanoic esters in remarkable
enantiomeric excess. The 1,4-addition of the same cuprate to some methyl R-D-manno- or R-D-
galactopyranosidic substrates in which a crotonyl group was incorporated, each at 3-OH, was also
investigated. The reverse π-facial attack of the cuprate was observed when some D-manno-type
substrates were subjected to 1,4-addition conditions similar to those used for the D-gluco-type
substrates. Furthermore, some D-galacto-type substrates provided 1,4-adducts with higher dia-
stereoselectivities.
In tr od u ction
(alkoxyalkyl)nitrones derived from D-ribose.2 Heathcock
et al. reported in 1981 the use of enolates attached to
carbohydrate templates for stereoselective carbon-
carbon bond-forming reactions.3 Another early example
of the carbohydrate-based auxiliary concept was the
introduction of a carbohydrate titanium complex pre-
pared from diacetoneglucose and cyclopentadienyltita-
nium trichloride, originated by Duthaler et al.4 For
asymmetric synthesis of R-amino acids and â-amino
acids, Kunz et al. investigated the use of glycosylamines
as chiral auxiliaries for Strecker- and Mannich-type
reactions.5 For asymmetric cycloadditions such as cyclo-
propanations,6 Diels-Alder reactions,7 and 1,3-dipolar
additions,8 the use of carbohydrates as chiral auxiliaries
has been explored during the past two decades. A number
of carbohydrate-derived templates have been used as
chiral auxiliaries for a variety of stereoselective organic
The introduction of effective stereocontrolling factors
in the molecule is one of the prominent subjects in
stereoselective organic synthesis. In this context, the
development of novel chiral auxiliaries prepared from
readily available natural products is an ongoing subject
in the field of asymmetric synthesis; the effort is similar
to the continuous effort devoted to the search for asym-
metric catalysts. Although many existing chiral auxilia-
ries can achieve a useful level of diastereoselection,
making possible the preparation of chiral compounds
with high optical purity,1 it is still desirable to find
effective chiral auxiliaries for achieving practical asym-
metric carbon-carbon bond-forming reactions. It has
been widely recognized that a number of carbohydrate-
based templates, which in many cases exist in acyclic,
pentofuranose, or hexopyranose forms, can serve as
effective auxiliaries as a consequence of forming a ste-
reochemically biased spatial environment. In one of the
earliest studies on the use of carbohydrate-derived chiral
auxiliaries, Vasella reported in 1977 the stereoselectivity
and reactivity in the 1,3-dipolar cycloaddition of N-
(2) Vasella, A. Helv. Chim. Acta 1977, 60, 1273-1295.
(3) Heathcock, C. H.; White, C. T.; Morrison, J . J .; VanDerveer, D.
J . Org. Chem. 1981, 46, 1296-1309.
(4) (a) Riediker, M.; Duthaler, R. O. Angew. Chem., Int. Ed. Engl.
1989, 28, 494-495. (b) Duthaler, R. O.; Herold, P.; Lottenbach, W.;
Oertle, K.; Reidiker, M. Angew. Chem., Int. Ed. Engl. 1989, 28, 495-
497. (c) Bold, G.; Duthaler, R. O.; Reidiker, M. Angew. Chem., Int. Ed.
Engl. 1989, 28, 497-498. (d) Duthaler, R. O.; Hafner, A.; Riediker, M.
Pure Appl. Chem. 1990, 62, 631-642. (e) Duthaler, R. O.; Hafner, A.
Chem. Rev. 1992, 92, 807-832.
† Department of Applied Chemistry.
‡ Department of Chemistry.
(1) (a) Evans, D. A. Aldrichimica Acta 1982, 15, 23-32. (b) Evans,
D. A. In Asymmetric Synthesis; Morrison, J . D., Ed.; Academic Press:
Orlando, FL, 1984; Vol. 3, pp 1-110. (c) Oppolzer, W. Tetrahedron
1987, 43, 1969-2004. (d) Oppolzer, W. Pure Appl. Chem. 1990, 62,
1241-1250. (e) Whitesell, J . K. Chem. Rev. 1992, 92, 953-964. (f)
Seyden-Penne, J . Chiral Auxiliaries and Ligands in Asymmetric
Synthesis; Wiley: New York, 1995. (g) Waldmann, H. Synlett 1995,
133-141. (h) Enders, D.; Klatt, M. Synthesis 1996, 1403-1418. (i)
Studer, A. Synthesis 1996, 793-815. (j) Ager, D. J .; Prakash, I.; Schaad,
D. R. Aldrichimica Acta 1997, 30, 3-12. (k) Regan, A. C. J . Chem.
Soc., Perkin Trans. 1 1999, 357-373. (l) Groaning, M. D.; Meyers, A.
I. Tetrahedron 2000, 56, 9843-9873.
(5) (a) Kunz, H.; Sager, W. Angew. Chem., Int. Ed. Engl. 1987, 26,
557-559. (b) Kunz, H.; Pfrengle, W. J . Am. Chem. Soc. 1988, 110, 651-
652. (c) Kunz, H.; Pfrengle, W. Tetrahedron 1988, 44, 5487-5494. (d)
Kunz, H.; Sager, W.; Pfrengle, W.; Schanzenbach, D. Tetrahedron Lett.
1988, 29, 4397-4400. (e) Kunz, H.; Pfrengle, W. Angew. Chem., Int.
Ed. Engl. 1989, 28, 1067-1068. (f) Kunz, H.; Schanzenbach, D. Angew.
Chem., Int. Ed. Engl. 1989, 28, 1068-1069. (g) Kunz, H.; Pfrengle,
W.; Sager, W. Tetrahedron Lett. 1989, 30, 4109-4110.
(6) (a) Charette, A. B.; Coˆte´, B.; Marcoux, J .-F. J . Am. Chem. Soc.
1991, 113, 8166-8167. (b) Charette, A. B.; Coˆte´, B. J . Org. Chem. 1993,
58, 933-936. (c) Kang, J .; Lim, G. J .; Yoon, S. K.; Kim, M. Y. J . Org.
Chem. 1995, 60, 564-577.
10.1021/jo0101860 CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/11/2001

5966 J . Org. Chem., Vol. 66, No. 18, 2001
Totani et al.
reactions, such as polar, radical-mediated, and pericyclic
carbon-carbon bond-forming reactions.9 In the field of
current organic synthesis, one of the significant develop-
ments is the use of stereoselective 1,4-additions10 in the
preparation of optically active chiral compounds, which
are achieved by using a variety of chiral auxiliaries. In
the early 1980s, Oppolzer et al. reported the 1,4-addition
of an organocopper reagent to chiral enoates prepared
from (-)-8-phenylmenthol and later to chiral enoates
prepared from their camphor-derived sultam.11 Tomioka
and Koga demonstrated the 1,4-addition of organocopper
reagents to an unsaturated amide attaching a chiral
auxiliary prepared from L-glutamic acid.12 In the early
1990s, Evans et al. reported the effectiveness of their
chiral acyloxazolidinone-derived enolates as Michael
donors. The donors reacted with representative electro-
philic olefins.13 Kunz et al. reported the 1,4-addition of
organoaluminum reagents to Evans oxazolidinone-type
auxiliary-bonded unsaturated carboxylates.14 The Lewis
acid-promoted 1,4-addition of allyltrimethylsilane to
analogous unsaturated N-acylamides was also investi-
gated.15 Enders et al. demonstrated the utility of their
chiral auxiliaries SAMP/RAMP for asymmetric 1,4-ad-
dition reactions.16 The development of highly stereo-
selective 1,4-additions using chiral auxiliaries is still
being extensively explored.17 Asymmetric 1,4-additions
based on the carbohydrate-derived auxiliaries have also
been investigated in this decade.18 In this paper, we
describe the results of the diastereoselective 1,4-addition
of a variety of organocuprates to 3-O-, 4-O-, or 6-O-
crotonyl derivatives of three methyl R-D-hexopyrano-
sides.19-22
Resu lts a n d Discu ssion
Dia ster eoselective 1,4-Ad d ition of a Vin ylcu p r a te
to r,â-Un sa tu r a ted Ester s In cor p or a ted in to Meth yl
r-D-Glu cop yr a n osid e. To explore feasibility and dia-
stereoselectivity in the 1,4-addition of a vinyl group to
hexopyranosidic templates, we designed a variety of 4-O-
crotonyl derivatives of methyl R-D-glucopyranoside as the
substrates.23 First, we prepared three 6-iodo-4-O-crotonyl
esters 9-11 from known methyl 2,3-di-O-protected R-D-
glucopyranosides 1,24 2,25 and 3,26 as shown in Scheme
1. Thus, preferential tosylation of 6-OH in 1-3 provided
the respective primary tosylates 4-6. The introduction
of a crotonyl ester at C-4 was carried out for 4 or 5 by
acylation with crotonic anhydride. The resulting 4-O-
(7) (a) Felber, H.; Kresze, G.; Prewo, R.; Vasella, A. Helv. Chim. Acta
1986, 69, 1137-1146. (b) Gupta, R. C.; Larsen, D. S.; Stoodley, R. J .;
Slawin, A. M. Z.; Williams, D. J . J . Chem. Soc., Perkin Trans. 1 1989,
739-749. (c) Kunz, H.; Mu¨ller, B.; Schanzenbach, D. Angew. Chem.,
Int. Ed. Engl. 1987, 26, 267-269. (d) Shing, T. K. M.; Lloyd-Williams,
P. J . Chem. Soc., Chem. Commun. 1987, 423-424. (e) Lubineau, A.;
Queneau, Y. J . Org. Chem. 1987, 52, 1001-1007. (f) Lubineau, A.;
Queneau, Y. Tetrahedron 1989, 45, 6697-6712. (g) Arnold, T.; Orschel,
B.; Reissig, H.-U. Angew. Chem., Int. Ed. Engl. 1992, 31, 1033-1035.
(8) (a) Vasella, A.; Voeffray, R. Helv. Chim. Acta 1982, 65, 1953-
1964. (b) Huber, R.; Knierzinger, A.; Obrecht, J .-P.; Vasella, A. Helv.
Chim. Acta 1985, 68, 1730-1747. (c) Mzengeza, S.; Whitney, R. A. J .
Org. Chem. 1988, 53, 4074-4081.
(9) (a) Scho¨llkopf, U.; To¨lle, R.; Egert, E.; Nieger, M. Liebigs Ann.
Chem. 1987, 399-405. (b) Kunz, H.; Mohr, J . J . Chem. Soc., Chem.
Commun. 1988, 1315-1317. (c) Duhamel, L.; Angibaud, P.; Desmurs,
J . R.; Valnot, J . Y. Synlett 1991, 807-808. (d) Laschat, S.; Kunz, H. J .
Org. Chem. 1991, 56, 5883-5889. (e) Cintas, P. Tetrahedron 1991, 47,
6079-6111. (f) Reissig, H.-U. Angew. Chem., Int. Ed. Engl. 1992, 31,
288-290. (g) Kunz, H.; Ru¨ck, K. Angew. Chem., Int. Ed. Engl. 1993,
32, 336-358. (h) Ganz, I.; Kunz, H. Synthesis 1994, 1353-1358. (i)
Kunz, H. Pure Appl. Chem. 1995, 67, 1627-1635. (j) Bols, M.
Carbohydrate Building Blocks; J ohn Wiley and Sons: New York, 1996.
(k) Kishida, M.; Eguchi, T.; Kakinuma, K. Tetrahedron Lett. 1996, 37,
2061-2062. (l) Schumacher, R.; Reissig, H.-U. Synlett 1996, 1121-
1122. (m) Hultin, P. G.; Earle, M. A.; Sudharshan, M. Tetrahedron
1997, 53, 14823-14870. (n) Chika, J .; Takei, H. Tetrahedron Lett. 1998,
39, 605-608. (o) Ferreira, M. L. G.; Pinheiro, S.; Perrone, C. C.; Costa,
P. R. R.; Ferreira, V. R. Tetrahedron: Asymmetry 1998, 9, 2671-2680.
(p) Enholm, E. J .; J iang, S. J . Org. Chem. 2000, 65, 4756-4758.
(10) For representative reviews on this subject, see: (a) Oare, D.
A.; Heathcock, C. H. Top. Stereochem. 1989, 19, 227-407. (b) Oare,
D. A.; Heathcock, C. H. Top. Stereochem. 1991, 20, 87-170. (c) Rossiter,
B. E.; Swingle, N. M. Chem. Rev. 1992, 92, 771-806. (d) Perlmutter,
P. Conjugate Addition Reactions in Organic Synthesis; Pergamon
Press: Oxford, 1992. (e) Noyori, R. Asymmetric Catalysts in Organic
Synthesis; J ohn Wiley and Sons: New York, 1994; pp 207-212. (f)
Alexakis, A. In Organocopper Reagents, A Practical Approach; Taylor,
R. J . K., Ed.; Oxford University Press: Oxford, U.K., 1994; pp 159-
183. (g) Feringa, B. L.; de Vries, A. H. M. In Advances in Catalytic
Processes; Doyle, M. P., Ed.; J AI Press Inc.: Greenwich, CT, 1995; Vol.
1, pp 151-192. (h) Tomioka, K.; Nagaoka, Y. In Comprehensive
Asymmetric Catalysis; J acobsen, E., Pfaltz, A., Yamamoto, H., Eds.;
Springer: Berlin, 1999; Vol. 3, Chapter 31.
(16) (a) Enders, D.; Papadopoulos, K.; Herdtweck, E. Tetrahedron
1993, 49, 1821-1830. (b) Enders, D.; Heider, K.-J .; Raabe, G. Angew.
Chem., Int. Ed. Engl. 1993, 32, 598-601.
(17) (a) Ambroise, L.; Desmae¨le, D.; Mahuteau, J .; d'Angelo, J .
Tetrahedron Lett. 1994, 35, 9705-9708. (b) Krause, N.; Gerold, A.
Angew. Chem., Int. Ed. Engl. 1997, 36, 186-204. (c) Williams, D. R.;
Kissel, W. S.; Li, J . J . Tetrahedron Lett. 1998, 39, 8593-8596. (d)
Ezquerra, J .; Prieto, L.; Avendano, C.; Martos, J . L.; de la Cuesta, E.
Tetrahedron Lett. 1999, 40, 1575-1578. (e) Hanessian, S.; Gomtsyan,
A.; Malek, N. J . Org. Chem. 2000, 65, 5623-5631.
(18) (a) Hon, Y.-S.; Chen, F.-L.; Huang, Y.-P.; Lu, T.-J . Tetrahe-
dron: Asymmetry 1991, 2, 879-882. (b) Ru¨ck, K.; Kunz, H. Synlett
1992, 343-344. (c) Ru¨ck, K.; Kunz, H. Synthesis 1993, 1018-1028.
(d) Ru¨ck-Braun, K.; Stamm, A.; Engel, S.; Kunz, H. J . Org. Chem. 1997,
62, 967-975. (e) Chiappe, C.; Moro, G. L.; Munforte, P. Tetrahedron:
Asymmetry 1997, 8, 2311-2317. (f) Pinheiro, S.; Pedraza, S. F.; Peralta,
M. A.; Carvalho, E. M.; Farias, F. M. C.; Ferreira, V. F. J . Carbohydr.
Chem. 1998, 17, 901-913. (g) Becker, B.; Thiem, J . Carbohydr. Res.
1998, 308, 77-84. (h) Follmann, M.; Kunz, H. Synlett 1998, 989-990.
(i) Desai, V. N.; Saha, N. N.; Dhavale, D. D. Chem. Commun. 1999,
1719-1720.
(19) Some of the results described herein have been presented at
the 19th International Carbohydrate Symposium (University of Cali-
fornia, San Diego, August 9-14, 1998), at the 36th National Organic
Chemistry Symposium (University of WisconsinsMadison, J une 13-
17, 1999), and at the 20th International Carbohydrate Symposium
(University of Hamburg, August 29-September 1, 2000).
(20) Some parts of the present work have been preliminarily
communicated. See: Totani, K.; Nagatsuka, T.; Takao, K.; Ohba, S.;
Tadano, K. Org. Lett. 1999, 1, 1447-1450.
(21) For our report on the stereoselective 1,4-addition of an alkyl
radical species to the D-glucose-based templates, see: Munakata, R.;
Totani, K.; Takao, K.; Tadano, K. Synlett 2000, 979-982.
(22) For our report on the stereoselective Diels-Alder reaction
achieved on some hexopyranosidic templates, see: Nagatsuka, T.;
Yamaguchi, S.; Totani, K.; Takao, K.; Tadano, K. Synlett 2001, 481-
484.
(23) We also explored the reactivity and stereoselectivity in the 1,4-
addition of two 3-O-crotonyl derivatives: methyl 2-O-benzyl-4,6-O-
benzylidene-3-O-crotonyl-R-D-glucopyranoside and methyl 2,4,6-tri-O-
benzyl-3-O-crotonyl-R-D-glucopyranoside. Under reaction conditions
similar to those used for the 4-O-crotonyl derivatives, the former 3-O-
crotonyl derivative gave the 1,4-vinyl adduct as a 62:38 diastereomeric
mixture (1H NMR analysis) in 75% yield. The latter gave the 1,4-adduct
as an approximately 1:1 diastereomeric mixture (88%). In both cases,
a useful level of stereoselectivity could not be obtained. Therefore, we
focused our interest on the substrates in which the reaction site was
installed on 4-OH or 6-OH. To date, we have not conducted 1,4-
additions using any 2-O-crotonyl derivatives.
(11) (a) Oppolzer, W.; Moretti, R.; Godel, T.; Meunier, A.; Lo¨her, H.
Tetrahedron Lett. 1983, 24, 4971-4974. (b) Oppolzer, W.; Kingma, A.
J .; Poli, G. Tetrahedron 1989, 45, 479-488.
(12) Tomioka, K.; Suenaga, T.; Koga, K. Tetrahedron Lett. 1986, 27,
369-372.
(13) Evans, D. A.; Bilodeau, M. T.; Somers, T. C.; Clardy, J .; Cherry,
D.; Kato, Y. J . Org. Chem. 1991, 56, 5750-5752.
(24) Tomic-Kulenovic, S.; Keglevic, D. Carbohydr. Res. 1980, 85,
302-306.
(14) Ru¨ck, K.; Kunz, H. Angew. Chem., Int. Ed. Engl. 1991, 30, 694-
(25) Yoshimoto, K.; Itatani, Y.; Shibata, K.; Tsuda, Y. Chem. Pharm.
Bull. 1980, 28, 208-219.
696.
(15) Wu, M.-J .; Wu, C.-C.; Lee, P.-C. Tetrahedron Lett. 1992, 33,
2547-2548.
(26) Hull, D. M. C.; Orchard, P. F.; Owen, L. N. Carbohydr. Res.
1977, 57, 51-63.

Highly Diastereoselective 1,4-Addition
J . Org. Chem., Vol. 66, No. 18, 2001 5967
Sch em e 1
Sch em e 2
mixed solution of THF and dimethyl sulfide (2:1, v/v). In
many cases, the 1,4-addition took place smoothly at that
temperature, providing each 1,4-adduct 9a -22a 30 as a
diastereomeric mixture in good-to-high yields. In some
cases, the diastereomeric ratio of the mixture could be
determined on the basis of 1H NMR analysis of each
reaction mixture after chromatographic purification on
silica gel. The 6-iodo derivative 9 gave the 1,4-adduct
9a Re with high diastereoselectivity (entry 1). Recrystal-
lization of the crude adduct 9a Re gave crystals suitable
for single-crystal analysis.31 Consequently, the R-config-
uration for the newly introduced stereogenic carbon in
9a Re was assigned unambiguously. To evaluate the
diastereoselectivity accurately, we performed a chiral
HPLC analysis to determine the enantiomeric ratio of
3-methyl-4-pentenoic acid benzyl esters (R)-a and (S)-a ,
which were prepared from a 9a Re/Si mixture by saponi-
fication followed by benzyl ester formation. The R-
configuration in the predominant diastereomer 9a Re was
further supported by the fact that the levorotatory
property of enantioenriched 3-methyl-4-pentenoic acid
prepared from 9a Re/Si was in accord with that reported
for the known (R)-(-) enantiopure sample.32 It was
established that substrate 9 afforded 9a Re with a
crotonyl derivatives 7 and 8 were treated with sodium
iodide for conversion into the respective iodides 9 and
10. On the other hand, replacement of the tosyloxy group
in 6 by an iodo group followed by acylation provided 11.
We also prepared two 6-O-protected 4-O-crotonyl deriva-
tives 14 and 15. Regioselective pivaloylation of 2 accord-
ing to a reported procedure27 provided 6-O-pivaloyl
derivative 12, which was acylated to afford the 4-O-
crotonyl derivative 14. Acylation of known 2,3,6-tri-O-
benzyl derivative 1328 gave the 4-O-crotonyl derivative
15. Moreover, we prepared 6-deoxy derivatives 19, 20,
and 22 (Scheme 2). Replacement of the tosyloxy group
in 4 by an iodo atom, followed by hydrogenolysis of the
resulting iodide 16, gave 17, which was acylated to 19.
Reductive removal of the sulfonyloxy group in the 6-O-
tosyl derivative 5 gave 18, which was acylated, affording
4-O-crotonyl derivative 20. Known 6-deoxy-2,3-di-O-
methyl derivative 2129 was acylated to give 4-O-crotonyl
ester 22. We also prepared a 4-O-cinnamoyl ester 23 as
another substrate for 1,4-addition by acylation of 18 with
cinnamoyl chloride.
We investigated the 1,4-addition of a vinylcuprate to
9-11, 14, 15, 19, 20, and 22. The results are summarized
in Table 1. All reactions were conducted at -78 °C in a
(30) The notation a in the number for each adduct (Tables 1, 4, 5,
and 7) means that the introduced carbon functionality is a vinyl group;
the notation b or c (Tables 2, 3, 6, and 8) means an ethyl or an isopropyl
group, respectively. The notation Re or Si in the number for the adduct
means that the adduct was formed as a result of the attack of the
organocopper reagent from the re-face or the si-face on the â-carbon
of the R,â-unsaturated ester, respectively.
(27) Eby, R.; Webster, K. T.; Schuerch, C. Carbohydr. Res. 1984,
129, 111-120. The regioselective pivaloylation of 2 proceeds through
a mercuric chelate formation. The 6-O-selective acetylation of 2 had
been conducted by the Schuerch group using CuCl2 (rt in THF). We
concluded that the yields (68% for 6-O-Ac and 13% for 4-O-Ac) of the
cupric chelate procedure were less effective for our purpose.
(28) Garegg, P. J .; Hultberg, H. Carbohydr. Res. 1981, 93, C10-
C11.
(31) X-ray crystallographic data on 9a Re are described in ref 20
and have been deposited at the Cambridge Crystallographic Data
Center, 12 Union Road, Cambridge CB2 1EZ, U.K.
(32) Enantioenriched (R)-3-methyl-4-pentenoic acid prepared from
the mixture of 9a Re/Si: [R]26D -17.2 (c 0.58, CHCl3). Enantiopure (R)-
(29) Shono, T.; Matsumura, Y.; Hamaguchi, H.; Naitoh, S. J . Org.
Chem. 1983, 48, 5126-5128.
3-methylpentenoic acid: [R]24 -17.42 (c 2.06, CHCl3). Uematsu, T.;
D
Umemura, T.; Mori, K. Agric. Biol. Chem. 1983, 47, 597-601.

5968 J . Org. Chem., Vol. 66, No. 18, 2001
Ta ble 1. 1,4-Ad d ition s of a Vin ylcu p r a te to 9-11, 14, 15, 19, 20, a n d 22
Totani et al.
CH2dCHMgBr
CuBr‚Me2S
(equiv)
major
adduct
d.r. of
e.r.,
entry
substrate
R
R1
(equiv)
yield (%)
adductsd
(R)-a :(S)-a f
1
2
3
4
5
6
7
8
9
9
10
11
14
15
19
19
20
20
20
22
Piv
Bn
Ac
I
I
I
10.0
10.0
10.0
10.0
10.0
10.0
5.0
10.0
4.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
0
5.0
0.1
0
9a Rea
10a Re
11a
14a Re
15a Re
19a Re
19b
20a Re
20a Re
20b
56 (84)c
75
51
e
e
94:6
87:13
g
79:21
83:17
97:3
55:45
82:18
86:14
e
Bn
Bn
Piv
Piv
Bn
Bn
Bn
Me
OPiv
OBn
H
H
H
H
H
H
81
84
61 (95)c
79
90
e
e
87:13
83:17
72 (89)c
84
10
11
10.0
5.0
22a
72 (86)c
56:44
g
a
b
Determined by single-crystal X-ray analysis. Complete recovery of the substrate. c Yield based on the recovered starting material.
Determined by 1H NMR. e Could not be determined by 1H NMR analysis. f Determined by HPLC (DAICEL Chiralcel OD + OD-H, i-PrOH/
hexane ) 1:400); t(R)-a ) 27.5 min and t(S)-a ) 28.5 min. Not determined.
d
g
Ta ble 2. 1,4-Ad d ition s of Th r ee Or ga n ocop p er Rea gen ts
remarkably high diastereomeric ratio of 94:6. In the case
of substrate 10, the diastereoselectivity of the 1,4-
addition decreased to an enantiomeric ratio of 87:13 with
a preference for the formation of 10a Re (entry 2).33 No
stereoselection was observed in the case of 3-O-acetyl
derivative 11 (entry 3). The effect of the C-6 substituent
on the diastereoselectivity of the 1,4-addition was exam-
ined using substrates 14 and 15. In the cases of bulkier
6-O-pivaloyl 14 and 6-O-benzyl derivative 15, adducts
14a Re and 15Re, both formed by the re-face attack of
the cuprate, were predominantly obtained, although their
diastereoselectivities were reduced (entries 4 and 5). The
highest diastereoselectivity was observed in the case of
6-deoxy derivative 19, which produced 19a Re almost as
a single product.34 The enantiomeric excess of (R)-a was
determined to be 94%, using HPLC analysis (entry 6).
As indicated in entries 7 and 10, the 1,4-addition did not
occur in the absence of the copper(I) salt. Moreover, the
amount of cuprous salt could be reduced to 10 mol %
(entry 8 versus 9).35 On the other hand, the amount of
the Grignard reagent was critical for maintaining the
effective progress of the reaction. The use of 1.0-2.0 mol
equiv of the Grignard reagent resulted in a dramatic
reduction in the yield of the adduct. It is apparent that
the combination of a larger 3-O-substituent and a smaller
6-O-substituent results in higher diastereoselectivity.
Thus, it is possible that the proximal steric environment
to 20
entry
reagent
temp (°C) major adduct yield (%) d.r.b
1
2
3
Et2CuMgBr -78 to -18
20bRe
20bRe
20bSi
88
84:16
EtCuBF3
Et2CuLi
-78 to 0
-78
72 (81)a 73:27
76 (84)a 66:34
a
b
Yield based on the recovered starting material. Determined
by 1H NMR.
around the reaction site, formed by both the substituents
at C-3 and C-6, predominantly affects the approaching
direction of the cuprate to the crotonyl ester. The steric
bulkiness of the 3-O-substituent seems to be the more
important stereocontrolling factor. This is supported by
the result of the 1,4-addition to 3-O-methyl derivative
22, resulting in the formation of 22a Re/Si with a
complete lack of diastereoselectivity (entry 11).
Next, we investigated the inevitability of the use of
cuprates prepared from the Grignard reagents to realize
the observed diastereoselectivity. Three types of ethyl-
copper reagents were prepared for the 1,4-addition to 20.
The results are summarized in Table 2. In the case of
magnesium diethylcuprate prepared from ethylmagne-
sium bromide, the 1,4-adduct 20bRe36 was obtained with
a similar diastereomeric ratio to that observed in the case
of the 1,4-addition to 20, using magnesium divinylcu-
prate. An ethylcopper boron trifluoride complex, prepared
by mixing ethyllithium (1.1 mol equiv) and cuprous
bromide-dimethyl sulfide (1.3 mol equiv), followed by the
addition of boron trifluoride diethyl ether complex (1.3
(33) In our preliminary report (ref 20), we inadvertently described
the diastereomeric ratios of the 1,4-adducts 9a Re/Si, 10a Re/Si, 19a Re/
Si, and 20a Re/Si as >99:1. Later, we established a precise HPLC
analysis procedure for evaluating the stereoisomeric ratios more
accurately after converting the diastereomeric mixture of the 1,4-
adducts into the corresponding â-chiral butanoic acid ester or anilide.
(34) The R-configuration for the introduced stereogenic carbon in
19a Re was confirmed by coincidence with an authentic specimen,
which was prepared from the stereochemically defined 9a Re by
deiodination.
(35) Although we did not examine all the substrates, it seems that
the amount of cuprous salt does not significantly influence the
diastereoselectivity of the 1,4-addition. However, the use of a catalytic
amount of cuprous salt in general caused a decrease in the yield.

Highly Diastereoselective 1,4-Addition
J . Org. Chem., Vol. 66, No. 18, 2001 5969
Ta ble 3. 1,4-Ad d ition s of Oth er Or ga n ocu p r a tes to 19, 20, a n d 23
e.r., (S)-b:(R)-b
or (R)-c:(S)-c
entry
substrate
R
R′
R′′
Et
i-Pr
i-Pr
temp (°C)
major adduct
yield (%)
d.r. of adductb
1
2
3
4
19
19
20
23
Piv
Piv
Bn
Me
Me
Me
Ph
-78 to -18
-78 to -18
-78
19bRe
19cRe
20cRe
23a Si
71 (80)a
77 (86)a
51 (94)a
60 (90)a
c
c
98:2d
96:4e
88:12e
f
88:12
89:11
Bn
vinyl
-78
a
d
Yield based on the recovered starting material. b Determined by 1H NMR. c Could not be determined by 1H NMR analysis. Determined
by HPLC (DAICEL Chiralcel OD-H, EtOH/hexane ) 1:30); tR ) 24.5 min and tS ) 26.7 min. e Determined by HPLC (DAICEL Chiralcel
OD, i-PrOH/hexane ) 1:10); tR ) 23.3 min and tS ) 29.0 min. f Not determined.
mol equiv), gave 20bRe as the major product, but with
decreased diastereoselectivity (entry 2). Interestingly, the
reverse π-facial selection was observed in the case of
lithium diethylcuprate, which provided 20bSi in a re-
markably reduced diastereomeric ratio (entry 3). Al-
though we have no reasonable explanation for this
stereochemical reversal, we concluded that significant
stereoselectivity could be achieved when an organo-
cuprate prepared from the Grignard reagent was used.
To examine the generality of the diastereoselective 1,4-
addition achieved using magnesium divinylcuprate, we
conducted the 1,4-additions using other organocuprates.
As shown in Table 3, the 1,4-additions of two magnesium
dialkylcuprates (R′′ ) ethyl or isopropyl) to 19 provided
the 1,4-adduct 19bRe or 19cRe, respectively, with high
diastereoselectivity (entries 1 and 2). The diastereo-
selectivity in the attack of the isopropylcuprate to 20 was
slightly diminished (entry 3). The diastereomeric ratios
of 19bRe/Si, 19cRe/Si, and 20cRe/Si were confirmed by
HPLC analysis of the enantiomeric anilide pairs (S)-b/
(R)-b or (R)-c/(S)-c, prepared from each diastereomeric
mixture by hydrolysis and followed by the anilide forma-
tion of the resulting enantioenriched (S)-3-methylpen-
tanoic acid37 or (R)-3,4-dimethylpentanoic acid.38 The 4-O-
cinnamoyl derivative 23 provided the 1,4-vinyl adduct
F igu r e 1. Plausible transition state (TS) models for π-facial
23a Si39 with a diastereoselectivity of 89:11 (1H NMR
selectivity in the attack of an organocuprate to the 4-O-crotonyl
analysis) (entry 4).
derivatives prepared from methyl R-D-glucopyranoside.
A plausible mechanistic rationale for the stereochem-
ical outcome observed using the aforementioned 4-O-R,â-
unsaturated esters is speculated (Figure 1). It is most
likely that the used magnesium organocuprate or the
formed magnesium halide, each of which exists in excess
in the reaction solution, coordinates with the crotonyl
carbonyl and other ester functionalities. This causes a
severely disfavorable interaction between the metal-
coordinating carbonyl and the carbon-carbon double
bond existing in the s-cis, syn conformation. Conse-
(36) The structure of 20bRe was unequivocally determined by
chemical correlation with an authentic specimen. Namely, the stere-
ochemically defined adduct 9a Re (d.r. ) 94:6) was hydrogenated
(Raney Ni), giving methyl 6-deoxy-4-[(S)-3-methylpentanoyl]-2,3-di-
O-pivaloyl-R-D-glucopyranoside (i). On the other hand, simultaneous
hydrogenation and hydrogenolysis of the adduct 10a Re (d.r. ) 87:13)
provided methyl 6-deoxy-4-[(S)-3-methylpentanoyl]-R-D-glucopyrano-
side (ii), from which compound i was obtained by per-O-pivaloylation.
In addition, hydrogenolysis of 20bRe in the presence of Pd(OH)2 on
charcoal gave ii. This experimental correlation led to the conclusion
that the newly introduced stereogenic carbon in 20bRe is S.
(37) For enantiopure (S)-(+)-3-methylpentanoic acid, see: Mori, K.;
Kato, M.; Kuwahara, S. Liebigs Ann. Chem. 1985, 861-865. For
enantiopure (R)-(-)-3-methylpentanoic acid, see: Mori, K.; Kuwahara,
S.; Ueda, H. Tetrahedron 1983, 39, 2439-2444.
(39) The R-configuration for the newly introduced stereogenic carbon
in 23a Si was determined as follows. After hydrolytic removal of the
carbohydrate template from the mixture of 23a Si/Re, enantioenriched
(R)-3-phenyl-4-pentenoic acid thus obtained was hydrogenated to (S)-
(38) Enantioenriched (R)-3,4-dimethylpentanoic acid prepared from
20cRe/Si (d.r. ) 88:12, 1H NMR analysis): [R]14D +12.0 (c 0.54, CHCl3).
Enantiopure (R)-(+)-3,4-dimethylpentanoic acid in the literature,
[R]20.5 +13.7 (c 1.54, CHCl3): Mori, K.; Ueda, H. Tetrahedron 1982,
3-phenylpentanoic acid with [R]27.5 +39.0 (c 0.58, benzene). For (R)-
D
D
38, 1227-1233.
3-phenylpentanoic acid, [R]25 -47.3 (c 5.3, benzene), see ref 18c.
D

5970 J . Org. Chem., Vol. 66, No. 18, 2001
Ta ble 4. 1,4-Ad d ition s of a Vin ylcu p r a te to 25 a n d 27
Totani et al.
entry
substrate
R
temp (°C)
major adduct
yield (%)
d.r. of adductsb
e.r.,c (S)-a :(R)-a
1
2
25
27
Piv
Bn
-78
-78
25a Si
27a Re
58 (73)a
62 (73)a
86:14
60:40
85:15
40:60
a
b
Yield based on the recovered starting material. Determined by 1H NMR. c Determined by HPLC.
Sch em e 3
quently, it is probable that the conformation of the R,â-
unsaturated ester changes favorably to the s-trans, syn
conformation.40,41 When the R,â-unsaturated ester is
stabilized in the s-trans, syn conformation, a C-3 sub-
stituent such as the pivaloyloxy group effectively shields
the front side, that is, the si-face on the â-carbon of the
unsaturated ester, as depicted in TS-A or TS-B. Conse-
quently, the cuprate attacks from the less-hindered re-
face (from the rear), providing the adduct possessing the
R-configuration (R′ ) vinyl). Regarding the shielding
effect of the C-3 substituent, the pivaloyloxy group is
superior to the benzyloxy group, as verified in the 1,4-
additions to 9, 10, 19, and 20 (entry 1 versus 2 or 6 versus
8, Table 1).42 In the cases of an acetoxy and a methoxy
group at C-3, the shielding effect is rather unpromising
(entries 3 and 11, Table 1). The steric effect of the C-6
substituent seems not to be critical for the formation of
the stereochemically biased chiral environment compared
to the stereochemical outcome affected by the C-3 sub-
stituent. Although the influence of pivaloyloxy or benz-
yloxy at C-6 on diastereoselectivity is less effective, as
revealed in entries 4 or 5 (Table 1), both 6-deoxy
substrates 19 and 20 revealed a comparably high dia-
stereoselectivity observed in the cases of 6-iodo counter-
parts 9 and 10 (entry 1 versus 6 and 2 versus 8, Table
1).
To obtain further insight into the chiral environment
formed by the D-glucopyranosidic template, we executed
the 1,4-addition of 6-O-crotonoyl derivatives 25 and 27.
The preparation of these substrates is summarized in
Scheme 3. Regioselective acylation of 2 gave 6-O-crotonyl
derivative 24, which was acylated to give 4-O-acylated
substrate 25. On the other hand, crotonoylation of known
methyl 2,3,4-tri-O-benzyl-R-D-glucopyranoside (26)43 af-
forded the 4-O-benzylated substrate 27. As summarized
in Table 4, a comparably high diastereoselectivity (dia-
stereomeric ratio (d.r.) ) 86:14) was observed in the case
of the 1,4-addition of the vinylcuprate to 25 with the
preferential formation of 25a Si. Regarding the intro-
duced stereogenic carbon, the major diastereomer 25a Si
possesses the reverse configuration of that in the major
adducts obtained using most 4-O-crotonyl esters. It is
apparent that the pivaloyloxy group at C-4 is important
for realizing this diastereoselective outcome. The 4-O-
benzyl derivative 27 provided the 1,4-adduct with de-
creased diastereoselectivity (at most, 1.5 to 1). The 1H
NMR analysis of 25 did not provide any decisive informa-
tion regarding the conformation of 25 in the absence or
presence of the vinylcuprate.
(40) We could not identify the exact coordinating metallic species.
We consider that the coordinating metal is magnesium in the orga-
nocuprate or the formed magnesium halide. However, the contribution
of copper as the coordinating metal cannot be ruled out. We tried to
confirm the conformational change of the unsaturated ester by the 1H
NMR technique before and after the addition of the cuprate. Unfor-
tunately, we could not obtain any useful information from the 1H NMR
spectra in the presence of magnesium divinylcuprate or magnesium-
(II) bromide, which showed a set of broad signals below 0 °C.
(41) It is reported that the s-trans, syn conformation of the R,â-
unsaturated esters attached to chiral auxiliaries is preferable to the
s-cis, syn conformation, especially in the presence of a coordinating
metallic species or in the presence of a Lewis acid. For previous
descriptions of the analogous conformational change of the reaction
site, see: (a) Oppolzer, W.; Lo¨her, H. Helv. Chim. Acta 1981, 64, 2808-
2811. (b) Oppolzer, W. Angew. Chem., Int. Ed. Engl. 1984, 23, 876-
889. (c) Loncharich, R. J .; Schwartz, T. R.; Houk, K. N. J . Am. Chem.
Soc. 1987, 109, 14-23. (d) Mezrhab, B.; Dumas, F.; d'Angelo, J .; Riche,
C. J . Org. Chem. 1994, 59, 500-503.
(42) It cannot be ruled out that the carbonyl group of the C-3
pivaloyloxy substituent in 9 or 19 participates in the transition state
(TS-A) and thus explains the superior diastereoselectivity compared
to that of the C-3 benzyloxy group (TS-B). We consider that the
bulkiness of the pivaloyloxy group works as an effective shielding factor
to govern the diastereoselection. We also observed a similar high
diastereoselectivity in the 1,4-addition of an alkyl radical species to
the substrate possessing a tert-butyldimethylsilyloxy group at C-3 (not
shown). The silyl ether at C-3 works as a bulky substituent to control
the approach of the radical species. In this particular case, the silyl
Dia ster eoselective 1,4-Ad d ition Ach ieved Usin g
th e Su bstr a tes P r ep a r ed fr om Meth yl r-D-Ma n n o-
p yr a n osid e. Our next concern was to examine the
reactivity and the stereocontrolling ability of D-manno-
pyranosidic templates for the 1,4-addition of a magne-
sium organocuprate. The preparation of the substrates
ether cannot participate as a coordinating group with
species. For this previous observation, see ref 21.
a metallic
(43) Liptak, A.; J odal, I.; Nanasi, P. Carbohydr. Res. 1975, 44, 1-11.

Highly Diastereoselective 1,4-Addition
J . Org. Chem., Vol. 66, No. 18, 2001 5971
Sch em e 4
hand, the 1,4-addition using substrates 40-42, possess-
ing an axial acyloxy group (OBz or OPiv), proceeded with
higher levels of diastereoselectivity (entries 6-13), re-
gardless of the bulkiness of the substituent at C-4 (OMe
or OBn). In the case of 40, the major diastereomer 40a Si
was obtained in crystalline form. By a single-crystal
X-ray analysis of 40a Si,47 the S-configuration was as-
signed unambiguously for the newly introduced stereo-
genic carbon. As indicated in entries 7, 11, and 12, the
1,4-addition proceeded efficiently in the presence of a
catalytic amount of cuprous salt. The amount of the
Grignard reagent can be reduced to 4 mol equiv without
a loss of the diastereoselectivity (entry 12). A significant
reduction in yield occurred in the case of 2 mol equiv of
the Grignard reagent (entry 13), and the 1,4-addition did
not proceed without the cuprous salt (entry 14). We also
examined the 1,4-addition of other organocuprates to
substrates 34 and 41. The results are summarized in
Table 6. The diastereoselectivity observed using 2-O-
pivaloyl ester 41 was superior to that obtained using 2-O-
benzyl ether 34. Each introduced stereogenic carbon in
the 1,4-adduct 34bRe or 41bSi was established through
stereochemical correlation to an authentic specimen
derived from the stereochemically defined 40a Si.48 Two
alkyl cuprates attack preferentially from the re-face on
the â-carbon for the 2-O-benzyl derivative 34 or from the
si-face for the 2-O-pivaloyl derivative 41.
A mechanistic account for the diastereoselectivity
obtained from the 3-O-crotonyl esters of methyl R-D-
mannopyranoside can be provided by assuming the
similar transition state models used for the D-gluco-type
substrates. As discussed in the case of the D-glucopyra-
nosidic substrates, the crotonyl ester in 34-37 and 40-
42 is most likely to exist in the s-trans, syn conformation
to minimize the expected steric repulsion. In the cases
of the D-manno-type substrates, the C-2 substituent
significantly governs the π-facial selection in the cuprate
attack regardless of the bulkiness of the C-4 substituent
(Figure 2). In the cases of 34-37 (TS-C), the steric effect
of the benzyloxy or methoxy group is modest. On the
other hand, the C-2 acyloxy group effectively shields the
re-face of the â-carbon on the double bond, resulting in
the preferential attack of the cuprate from the si-face
(TS-D).
Dia ster eoselective 1,4-Ad d ition Ach ieved Usin g
th e Su bstr a tes P r ep a r ed fr om Meth yl r-D-Ga la cto-
p yr a n osid e. Finally, we explored 1,4-addition using five
3-O-crotonyl esters of methyl R-D-galactopyranoside,
substrates 44 and 49-52. As shown in Scheme 5, the
preparation of these substrates was started from known
methyl 2,3,6-tri-O-benzyl-R-D-galactopyranoside 4349 (for
44), known 4,6-di-O-benzyl 4550 (for 49 and 50), or 4,6-
we designed44 is summarized in Scheme 4. Starting from
known methyl 4,6-di-O-benzyl- (28),45 2,3,6-tri-O-benzyl-
(29),45 or 4,6-di-O-methyl-R-D-mannopyronoside (31),46
seven 3-O-crotonyl derivatives, 34-37 and 40-42, were
prepared using standard procedures. 1,4-Addition of
these substrates using magnesium divinylcuprate was
conducted under reaction conditions similar to those used
for the D-glucopyranosidic substrates. The results are
summarized in Table 5. The ratio of the diastereomeric
mixture obtained from each 1,4-addition was determined
by 1H NMR analysis. The stereoselectivity in the 1,4-
addition was finally confirmed by chiral HPLC analysis
of enantioenriched 3-methyl-4-pentenoic acid benzyl es-
ters (R)-a or (S)-a . The substrates carrying an axial
alkyloxy group (OBn or OMe), such as 34-37, afforded
1,4-adducts 34a Re-37a Re with the best diastereomeric
ratio of 88:12 (entries 1-5). The amount of cuprous salt
can be reduced to a catalytic amount without a significant
loss of yield or diastereoselectivity (entry 3). On the other
(47) X-ray crystallographic data for 40a Si (colorless, platelike
crystals were grown from a methanol solution): molecular formula )
C
34H38O8, MW ) 574.67, triclinic; a ) 11.298(2), b ) 15.254(2), and c
) 9.830(1) Å; R ) 99.77(1), â ) 90.34(1), and γ ) 109.41(1)°; and V )
1571.1 Å3. A space group P1 (No. 1), Z ) 2, Dc ) 1.22 g/cm3. Data
collection was done on a Rigaku AFC7R diffractometer (50 kV, 200
mA) using Cu KR radiation (µ ) 7.04 cm-1) with a rotating anode
generator. In total, 5407 reflections were collected, of which 755
reflections with I > 2.00σ(I) were variable parameters. The final R
(44) We also conducted the 1,4-addition of magnesium divinylcu-
prate to methyl 3,4,6-tri-O-benzyl-2-O-crotonyl-R-D-mannopyranoside
(-78 °C, 1 h), in which the reaction site was installed at an axial OH
on C-2. As a result, the 1,4-adducts were obtained as a nearly 1:1
factor was 0.073 (Rw ) 0.102). The final difference peaks were Fmax
)
diastereomeric mixture in
a yield of 49% (40% recovery of the
0.50 and Fmin ) -0.40 e/Å3.
substrate). In addition, methyl 2-O-crotonyl-4,6-di-O-methyl-3-O-piv-
aloyl-R-D-mannopyranoside gave the 1,4-vinyl adducts as a nearly 1:1
diastereomeric mixture (86%). Both 1,4-additions proceeded completely
devoid of diastereoselectivity.
(45) Kong, F.; Schuerch, C. Carbohydr. Res. 1983, 112, 141-147.
(46) Khan, S. H.; Matta, K. L. Carbohydr. Res. 1993, 243, 29-42.
(48) The stereochemical assignment for the isopropyl adducts 34cRe
and 41cSi was also carried out by chemical transformation and
hydrolytic removal of the carbohydrate template.
(49) Rashid, A.; Mackie, W. Carbohydr. Res. 1992, 223, 147-155.
(50) Schneider, J .; Lee, Y. C.; Flowers, H. M. Carbohydr. Res. 1974,
36, 159-166.

5972 J . Org. Chem., Vol. 66, No. 18, 2001
Ta ble 5. 1,4-Ad d ition s of a Vin ylcu p r a te to 34-37 a n d 40-42
Totani et al.
CH2dCHMgBr
CuBr‚Me2S
(equiv)
entry
substrate
R
R′
(equiv)
major adduct
yield (%)
d.r. of adductsc
e.r.,e (S)-a :(R)-a
1
2
3
4
5
6
7
8
10
11
12
13
14
34
35
35
36
37
40
40
41
42
42
42
42
42
Bn
Bn
Bn
Me
Me
Bn
Bn
Bn
Me
Me
Me
Me
Me
Bn
10.0
10.0
5.0
10.0
10.0
10.0
5.0
10.0
10.0
8.0
5.0
5.0
0.1
5.0
5.0
5.0
0.1
5.0
5.0
0.1
0.1
0.1
0
34a Re
35a Re
35a Re
36a Re
37a Re
40a Sia
40a Si
41a Si
42a Si
42a Si
42a Si
42a Si
42b
84
85
77
75
74
84
83
78
92
91
91
18
77
88:12
87:13
87:13
d
80:20
91:9
90:10
89:11
91:9
91:9
91:9
91:9
88:12
84:16
d
64:36
83:17
9:91
d
9:91
9:91
d
Me
Me
Bn
Me
Bz
Bz
Piv
Piv
Piv
Piv
Piv
Piv
4.0
2.0
5.0
d
d
a
b
d
Determined by single-crystal X-ray analysis. Recovery of the substrate. c Determined by 1H NMR analysis. Not determined.
e Determined by HPLC.
Ta ble 6. 1,4-Ad d ition s of Oth er Or ga n ocu p r a tes to 34
a n d 41
entry substrate
R
R′
major adduct yield (%) d.r.c
1
2
34
34
41
41
Bn Et
Bn i-Pr
Piv Et
Piv i-Pr
34bRe
34cRe
41bSi
41cSi
51 (63)b 68:32
74
83
83
81:19
90:10
73:27
3
4a
a
b
The reaction was carried out at -18 °C. Yield based on the
recovered starting material. c Determined by 1H NMR analysis.
di-O-methyl derivative 4751 (for 51 and 52). The results
of the 1,4-addition using these substrates are sum-
marized in Table 7. The 4-O-benzyl derivative 44 gave
the 1,4-adduct with preferential formation of 44a Si. The
4-O-acyl derivatives 49-52 provided the respective 1,4-
adduct in good yield with the best diastereomeric ratio
of 98:2. As shown in Table 8, magnesium diethyl or
diisopropylcuprate provided the respective adduct 50bRe
or 50cRe with excellent diastereoselectivity (>94% enan-
tiomeric excess for the anilides (S)-b or (R)-c).52
The high-to-excellent diastereoselectivity obtained us-
ing the 3-O-crotonyl esters of D-galactopyranoside can
also be explained by the same transition states argument
F igu r e 2. Plausible transition state models for π-facial
selective attack of the organocuprates to the 3-O-crotonyl
derivatives of D-mannopyranoside.
employed for the D-glucosidic and D-mannosidic sub-
strates (Figure 3). The crotonyl ester favorably exists in
an s-trans, syn conformation. As anticipated, the dia-
stereoselectivity was not sufficiently high in the case of
(51) Rao, A. S.; Roy, N. Carbohydr. Res. 1977, 59, 393-401.
(52) The absolute configuration of the major adduct was established
from the optical sign of the enantioenriched 3-methylpentanoic acid
or 3,4-dimethylpentanoic acid, obtained by the removal of the carbo-
hydrate template.

Highly Diastereoselective 1,4-Addition
J . Org. Chem., Vol. 66, No. 18, 2001 5973
Ta ble 8. 1,4-Ad d ition s of Oth er Or ga n ocu p r a tes to 50
Sch em e 5
e.r., (S)-b:(R)-b
entry
R
major adduct
yield (%)
or (R)-c:(S)-cb
1
2
Et
i-Pr
50bRe
50cRe
70 (84)a
44 (98)a
98:2
97:3
a
b
Yield based on the recovered starting material. Determined
by HPLC.
Ta ble 7. 1,4-Ad d ition s of th e Vin ylcu p r a te to 44 a n d
49-52
major yield d.r. of
e.r.,
entry substrate
R
R′ adduct (%) adducta (R)-a :(S)-a c
1
2
3
4
5
44
49
50
51
52
Bn Bn 44a Si
Bn Bz 49a Re
Bn Piv 50a Re
Me Bz 51a Re
Me Piv 52a Re
82
82
94
80
80
84:16
b
b
95:5
b
14:86
97:3
98:2
94:6
98:2
F igu r e 3. Plausible transition models for π-facial selective
attack of the organocuprates to the 3-O-crotonyl derivatives
of D-galactopyranoside.
â-carbon substituted butanoic acids in both enantiomeric
forms. Although the precise mechanism of the transition
state in the 1,4-addition is still unclear, it should be
emphasized that the present results show one of the
effective uses of carbohydrate scaffolds for achieving a
highly stereoselective organic reaction. We are currently
investigating the applicability of the present concept to
other types of carbon-carbon bond-forming reactions.
a
1
b
Determined by H NMR analysis. Could not be determined.
c Determined by HPLC.
the C-4 alkylated substrate 44. In contrast, the effect of
the C-4 acyloxy groups was remarkable, resulting in high
diastereoselection. As indicated by TS-E, a preferential
attack of the organocuprate from the less-hindered re-
face on the â-carbon of the double bond occurred.
Con clu sion s
Exp er im en ta l Section
We have developed a novel and highly diastereoselec-
tive 1,4-addition using a variety of hexopyranosidic
substrates as chiral templates. Consequently, a variety
of â-alkylated butanoic acids are available in a high-to-
excellent enantioenriched form. A number of O-crotonyl
derivatives of D-gluco-, D-manno-, and D-galactopyrano-
sides are approved as good stereocontrollers for realizing
the practical asymmetric 1,4-additions. Furthermore,
these substrates complementarily provide a variety of
Melting points are uncorrected. Specific rotations were
measured in a 10 mm cell. H NMR spectra were recorded at
1
270 or 300 MHz in CDCl3 solution with tetramethylsilane as
an internal standard. 13C NMR spectra were recorded at 75
MHz in CDCl3 solution. Thin-layer chromatography (TLC) was
performed with a glass plate coated with Kieselgel 60 GF254
(Merck). The crude reaction mixtures and extractive materials
were purified by chromatography on silica gel 60 K070
(Katayama Chemicals). Combined organic extracts were dried
over anhydrous Na2SO4. Solvents were removed by concentra-

5974 J . Org. Chem., Vol. 66, No. 18, 2001
Totani et al.
tion under reduced pressure using an evaporator with a water
bath at 35-45 °C. Solvents were dried (drying reagent) and
distilled prior to use: tetrahydrofuran (THF) (KANTO CHEMI-
CALS; water, 0.005% max) and CH2Cl2 (CaH2).
residue was purified by column chromatography on silica gel
(EtOAc/hexane, 1:100) to give 20.0 mg (94%) of a mixture of
(R)-a and (S)-a as a colorless oil: TLC, Rf 0.76 (EtOAc/hexane,
1:2); IR (neat) 3060-2880, 1740, 1560 cm-1 1H NMR (300
;
Gen er a l P r oced u r e for t h e 1,4-Ad d it ion s. Met h yl
6-Deoxy-6-iod o-4-O-[(R)-3-m eth yl-4-p en ten oyl]-2,3-d i-O-
p iva loyl-r-D-glu cop yr a n osid e (9a Re) (En tr y 1 in Ta ble
1). The reaction was carried out under argon. To a cold (-78
°C) solution of CuBr‚Me2S (204 mg, 0.993 mmol) in THF-Me2S
(2:1) (3 mL) was added vinylmagnesium bromide (1.04 M
solution in THF, 1.91 mL, 1.99 mmol). The solution was stirred
at -78 °C for 1 h, and then a solution of 9 (107 mg, 0.200
mmol) in THF (1 mL) was added. The solution was stirred at
-78 °C for 25 min and then quenched with saturated aq NH4-
Cl (1 mL). After being stirred for 10 min, the solution was
diluted with EtOAc (20 mL) and washed with saturated aq
NH4Cl (10 mL × 5). The organic layer was dried and
concentrated in vacuo. The residue was purified by column
chromatography on silica gel (EtOAc/hexane, 1:30) to give 66.7
mg (56%) of 9a Re as colorless crystals (enantiomeric ratio (e.r.)
of the corresponding benzyl ester ) 94:6, HPLC analysis), and
35.6 mg (33%) of 9 was recovered. 9a Re: mp 88-90 °C; TLC,
Rf 0.61 (EtOAc/hexane, 1:5); IR (neat) 1740 cm-1; 1H NMR (300
MHz) δ 1.05 (d, J ) 6.8 Hz, 3H), 1.11, 1.16 (2 s, 9 H × 2), 2.29
(d, J ) 7.1 Hz, 2H), 2.61-2.70 (m, 1H), 3.10 (dd, J ) 8.5, 11.0
Hz, 1H), 3.29 (dd, J ) 2.3, 11.0 Hz, 1H), 3.46 (s, 3H), 3.77-
3.83 (m, 1H), 4.80 (dd, J ) 3.8, 9.9 Hz, 1H), 4.91 (t, J ) 9.9
Hz, 1H), 4.96 (d, J ) 3.8 Hz, 1H), 4.98-5.07 (m, 2H), 5.52 (t,
J ) 9.9 Hz, 1H), 5.74 (ddd, J ) 7.1, 10.3, 17.3 Hz, 1H); 13C
NMR (75 MHz) δ 3.86, 20.02, 26.91 × 6, 27.01, 34.05, 38.69 ×
2, 40.89, 55.86, 68.96, 71.06, 71.97, 96.59, 113.95, 142.08,
170.85, 177.03, 177.61. Anal. Calcd for C23H37O8I: C, 48.60;
H, 6.56. Found: C, 48.48; H, 6.81.
MHz) δ 1.05 (d, J ) 6.6 Hz, 3H), 2.32 (dd, J ) 7.3, 14.8 Hz,
1H), 2.42 (dd, J ) 7.3, 14.9 Hz, 1H), 2.66-2.73 (m, 1H), 4.93-
5.04 (m, 2H), 5.11 (s, 2H), 5.77 (ddd, J ) 7.0, 10.4, 17.2 Hz,
1H), 7.30-7.38 (m, 5H); 13C NMR (75 Hz) δ 19.71, 34.43, 41.27,
66.14, 113.42, 128.18 × 2, 128.22, 128.50 × 2, 135.97, 142.34,
172.33; HRMS calcd for C13H16O2 (M+), 204.1150; found,
204.1155. The enantiomeric ratio of the mixture was deter-
mined by HPLC analysis (DAICEL Chiralcel OD + ODH,
i-PrOH/hexane ) 1:400).
Gen er a l P r oced u r e for Rem ova l of th e Ca r boh yd r a te
Tem p la te F ollow ed by An ilid e F or m a tion . F or m a tion of
(R)-c/(S)-c fr om
a Dia ster eom er ic Mixtu r e (88:12) of
20cRe a n d 20cSi. To a solution of a mixture of 20cRe/Si (48.3
mg, 0.103 mmol) in MeOH (1 mL) was added 4 M KOH
solution (1 mL). This solution was refluxed for 23 h, diluted
with H2O (5 mL), and extracted with CHCl3. The organic layer
was dried and concentrated in vacuo to give 35.1 mg (95%) of
18. The aqueous layer was acidified with 1 M HCl (to pH 2)
and extracted with CHCl3. The combined extracts were dried
and concentrated in vacuo to give 3,4-dimethylpentanoic acid,
which was subjected to the next step without purification. The
residue was dissolved in CH2Cl2 (1 mL) and 1-ethyl-(3-
dimethylaminopropyl)carbodiimide (58 mg, 0.30 mmol), and
aniline (28 µL, 0.31 mmol) and DMAP (3.8 mg, 0.31 mmol)
were added. After being stirred for 20 h, the solution was
diluted with EtOAc (10 mL) and washed with H2O (5 mL ×
3). The organic layer was dried and concentrated in vacuo. The
residue was purified by column chromatography on silica gel
(EtOAc/hexane, 1:10) to give 20.7 mg (98%) of a mixture of
(R)-c and (S)-c as colorless crystals: TLC, Rf 0.58 (EtOAc/
hexane, 1:2); IR (neat) 2960, 1660, 1600, 1550 cm-1; 1H NMR
(270 MHz) δ 0.88 (d, J ) 7.0 Hz, 3H), 0.92, 0.93 (2 d, J ) 6.6
Hz, 3H × 2), 1.67 (m, 1H), 1.96-2.10 (m, 2H), 2.43 (dd, J )
2.9, 8.8 Hz, 1H), 7.53 (m, 5H); 13C NMR (75 Hz) δ 16.00, 18.60,
20.35, 32.47, 36.70, 43.10, 120.25 × 2, 124.56, 129.34 × 2,
138.37, 171.83; HRMS calcd for C13H19NO (M+), 205.1467;
found, 205.1450. The enantiomeric ratio of the mixture was
determined by HPLC analysis (DAICEL Chiralcel OD, i-PrOH/
hexane ) 1:10).
Meth yl 2-O-Ben zoyl-4,6-d i-O-ben zyl-3-O-[(S)-3-m eth yl-
4-p en ten oyl]-r-D-m a n n op yr a n osid e (40a Si) (En tr y 6 in
Ta ble 5). As described for the preparation of 9a Re, 1.05 g
(2.00 mmol) of 40 in THF (3.25 mL) was treated with CuBr‚
Me2S (2.06 g, 10.0 mmol) in THF-Me2S (2:1) (21 mL) and
vinylmagnesium bromide (1.10 M solution in THF, 18.0 mL,
20.0 mmol) at -78 °C for 1 h. Extractive workup and
purification by column chromatography on silica gel (EtOAc/
hexane, 1:5) gave 0.93 g (84%) of a mixture of 40a Si/Re (e.r.
of the corresponding benzyl ester ) 91:9, HPLC analysis) as
colorless crystals, which was recrystallized from isopropyl
ether to give pure 40a Si: mp 80.5-81.5 °C; TLC, Rf 0.71
(EtOAc/hexane, 1:2); IR (neat) 1730, 1600, 1500 cm-1; 1H NMR
(270 MHz) δ 0.95 (d, J ) 7.0 Hz, 3H), 2.11 (dd, J ) 7.9, 15.0
Hz, 1H), 2.28 (dd, J ) 6.6, 15.0 Hz, 1H), 2.53-2.63 (m, 1H),
3.42 (s, 3H), 3.75-3.94 (m, 3H), 4.20 (t, J ) 9.2 Hz, 1H), 4.55,
4.68 (2 d, J ) 11.1 Hz, 1H × 2), 4.55, 4.76 (2 d, J ) 11.5 Hz,
1H × 2), 4.80-4.94 (m, 2H), 4.84 (d, J ) 2.9 Hz, 1H), 5.44-
5.50 (m, 2H), 5.67 (ddd, J ) 6.8, 10.4, 17.2 Hz, 1H), 7.17-8.08
(m, 15H); 13C NMR (75 MHz) δ 19.51, 33.95, 40.96, 55.15,
68.78, 70.50, 71.42, 72.00, 72.90, 73.50, 74.75, 98.52, 113.26,
127.55 × 3, 127.62 × 2, 127.70, 128.32 × 2, 128.36 × 2, 128.45
× 2, 129.67, 129.84 × 2, 133.26, 138.00, 138.32, 142.17, 165.42,
171.44; HRMS calcd for C34H38O8 (M+), 574.2566; found,
574.2564.
Met h yl 2,3-Di-O-b en zyl-6-d eoxy-4-O-[(R)-3-m et h yl-4-
p en ten oyl]-r-D-glu cop yr a n osid e (20a Re). Use of a Ca ta -
lytic Am ou n t of Cu Br ‚Me2S (En tr y 9 in Ta ble 1). 47.1 mg
(0.110 mmol) of 20 in THF (1 mL) was treated with CuBr‚
Me2S (2.3 mg, 11.2 µmol) in THF-Me2S (2:1) (4.5 mL) and
vinylmagnesium bromide (0.95 M solution in THF, 0.46 mL,
0.44 mmol) at -78 °C for 50 min. Extractive workup and
purification by column chromatography on silica gel (EtOAc/
hexane, 1:16) gave 35.9 mg (72%) of the mixture of 20a Re/Si
(e.r. of the corresponding benzyl ester ) 83:17, HPLC analysis),
and 8.9 mg (19%) of 20 was recovered. 20a Re/Si as a colorless
oil: TLC, Rf 0.69 (EtOAc/hexane, 1:2); IR (neat) 1740 cm-1
;
1H NMR for the major 20a Re (300 MHz) δ 1.01 (d, J ) 6.8
Hz, 3H), 1.12 (d, J ) 6.3 Hz, 3H), 2.13 (dd, J ) 7.3, 15.4 Hz,
1H), 2.25 (dd, J ) 7.0, 15.4 Hz, 1H), 2.56-2.68 (m, 1H), 3.38
(s, 3H), 3.57 (dd, J ) 3.7, 9.5 Hz, 1H), 3.69-3.79 (m, 1H), 3.87
(t, J ) 9.5 Hz, 1H), 4.53 (d, J ) 3.7 Hz, 1H), 4.63, 4.64 (2 d, J
) 11.9 Hz, 1H × 2), 4.77 (t, J ) 9.5 Hz, 1H), 4.77 (d, J ) 12.2
Hz, 1H), 4.86-5.02 (m, 3H), 5.73 (ddd, J ) 6.7, 10.4, 17.3 Hz,
1H), 7.23-7.31 (m, 10H); 13C NMR for the major 20a Re (75
MHz) δ 17.45, 19.64, 33.87, 47.11, 55.25, 65.32, 73.40, 74.98,
75.14, 78.97, 79.90, 98.06, 113.44, 127.45, 127.63 × 2, 127.93,
128.13 × 2, 128.26 × 2, 128.44 × 2, 137.98, 138.65, 142.27,
171.34. Anal. Calcd for C27H34O6: C, 71.34; H, 7.54. Found:
C, 71.32; H, 7.55.
Gen er a l P r oced u r e for Rem ova l of th e Ca r boh yd r a te
Tem p la te F ollow ed by Ben zyl Ester F or m a tion . F or m a -
tion of (R)-a /(S)-a fr om a Dia ster eom er ic Mixtu r e (87:
13) of 20a Re/Si. To a solution of the mixture (47.4 mg, 0.104
mmol) in MeOH (1 mL) was added 4 M KOH solution (1 mL).
This solution was refluxed for 16 h, diluted with H2O (5 mL),
and extracted with CHCl3. The organic layer was dried and
concentrated in vacuo to give 36.0 mg (97%) of 18. The aqueous
layer was acidified with 1 M HCl (to pH 2) and extracted with
CHCl3. The combined extracts were dried and concentrated
in vacuo to give 3-methyl-4-pentenoic acid, which was benzy-
lated without purification. The residue was dissolved in DMF
(1 mL), and NaHCO3 (18 mg, 0.22 mmol) and benzyl bromide
(24.5 µL, 0.20 mmol) were added. After being stirred at 50 °C
for 19 h, the solution was diluted with EtOAc (10 mL) and
washed with H2O (5 mL) and saturated aq NaHCO3 (5 mL ×
2). The organic layer was dried and concentrated in vacuo. The
Meth yl 2,6-Di-O-ben zyl-3-O-[(R)-3-m eth yl-4-pen ten oyl]-
4-O-p iva loyl-r-D-ga la ctop yr a n osid e (50a Re) (En tr y 3 in
Ta ble 7). As described for the preparation of 9a Re, 44.5 mg
(84.5 mmol) of 50 in THF (1 mL) was treated with CuBr‚Me2S
(3.2 mg, 16 µmol) in THF-Me2S (2:1) (1.2 mL) and vinylmag-
nesium bromide (0.95 M solution in THF, 0.44 mL, 0.42 mmol)
at -78 °C for 1 h. Extractive workup and purification by

Highly Diastereoselective 1,4-Addition
J . Org. Chem., Vol. 66, No. 18, 2001 5975
column chromatography on silica gel (EtOAc/hexane, 1:5) gave
44.1 mg (94%) of 50a Re (e.r. of the corresponding benzyl ester
) 98:2, HPLC analysis) as a colorless oil: TLC, Rf 0.63 (EtOAc/
of Education, Science, Sports and Culture, J apan. The
financial support through the Sasakawa Scientific
Research Grant from the J apan Science Society to K.T.
(11-120) is gratefully acknowledged.
1
hexane, 1:2); IR (neat) 1740 cm-1; H NMR (300 MHz) δ 1.04
(d, J ) 6.8 Hz, 3H), 1.15 (s, 9H), 2.10 (dd, J ) 8.3, 15.1 Hz,
1H), 2.29 (dd, J ) 6.1, 15.1 Hz, 1H), 2.58-2.72 (m, 1H), 3.37-
3.49 (m, 5H), 3.81 (dd, J ) 3.7, 10.4 Hz, 1H), 4.14-4.18 (m,
1H), 4.42, 4.52 (2 d, J ) 11.9 Hz, 1H × 2), 4.60, 4.68 (2 d, J )
12.4 Hz, 1H × 2), 4.75 (d, J ) 3.7 Hz, 1H), 4.95 (dt, J ) 1.5,
10.5 Hz, 1H), 5.01 (dt, J ) 1.5, 17.3 Hz, 1H), 5.36 (dd, J ) 3.4,
10.4 Hz, 1H), 5.44-5.45 (m, 1H), 5.78 (ddd, J ) 6.6, 10.5, 17.3
Hz, 1H), 7.24-7.35 (m, 10H); 13C NMR (75 MHz) δ 19.28, 27.09
× 3, 33.65, 38.95, 40.94, 55.45, 67.59, 68.41, 68.71, 69.75, 73.02,
73.22, 73.43, 98.37, 113.19, 127.62 × 2, 127.68, 127.94, 128.06
× 2, 128.37 × 4, 137.68, 137.77, 142.39, 171.14, 177.06; HRMS
calcd for C32H42O8 (M+), 554.2880; found, 554.2887.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails for the preparation and characterization of 4-12, 14-
20, 22-25, 27, 32, 34-42, 44, 46, 48-52, 10a Re, 19a Re,
19bRe, 19cRe, 20a Re, 20bRe, 25a Si, 34a Re, 41bSi, 42a Si,
1
and 50bRe; copies of H and 13C NMR spectra of 4-12, 14-
20, 22-25, 27, 32, 34-42, 44, 46, 48-52, 10a R e, 19a R e,
19bRe, 19cRe, 20a Re, 20bRe, 25a Si, 34a Re, 41bSi, 42a Si,
and 50bRe; and the ORTEP plot for 40a Si. This material is
available free of charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. This work was supported by a
Grant-in-Aid for Scientific Research from the Ministry
J O0101860