Regioselectivity in palladium-indium iodide-mediated allylation reaction of glyoxylic oxime ether and N-sulfonylimine

Article abstract of DOI:10.1021/jo034451o

Allylation reaction of electron-deficient imines with allylic alcohol derivatives in the presence of a catalytic amount of palladium(0) complex and indium(I) iodide was studied. The reversibility of allylation was observed in the reaction of glyoxylic oxime ether having camphorsultam. As the important effect of water on regioselectivity, the γ-adducts were kinetically formed from mono-substituted allylic reagents in the presence of water. The selective formation of thermodynamically stable α-adducts was observed in anhydrous THF. In contrast, the allylation of N-sulfonylimine gave the α-adducts with high regioselectivities even under anhydrous reaction conditions.

Full text of DOI:10.1021/jo034451o

Regioselectivity in P a lla d iu m -In d iu m Iod id e-Med ia ted Allyla tion
Rea ction of Glyoxylic Oxim e Eth er a n d N-Su lfon ylim in e
Hideto Miyabe,† Yousuke Yamaoka,† Takeaki Naito,‡ and Yoshiji Takemoto*,†
Graduate School of Pharmaceutical Sciences, Kyoto University, Yoshida, Sakyo-ku,
Kyoto 606-8501, J apan, and Kobe Pharmaceutical University, Motoyamakita, Higashinada,
Kobe 658-8558, J apan
takemoto@pharm.kyoto-u.ac.jp
Received April 9, 2003
Allylation reaction of electron-deficient imines with allylic alcohol derivatives in the presence of a
catalytic amount of palladium(0) complex and indium(I) iodide was studied. The reversibility of
allylation was observed in the reaction of glyoxylic oxime ether having camphorsultam. As the
important effect of water on regioselectivity, the γ-adducts were kinetically formed from mono-
substituted allylic reagents in the presence of water. The selective formation of thermodynamically
stable R-adducts was observed in anhydrous THF. In contrast, the allylation of N-sulfonylimine
gave the γ-adducts with high regioselectivities even under anhydrous reaction conditions.
In tr od u ction
The reaction of allylmetals with electrophiles has been
developed as a fundamentally important carbon-carbon
bond-forming method.1 In general, the γ-adducts (branched
products) were predominantly obtained in the allylation
of aldehydes and imines using allylmetals. Thus, the
selective synthesis of the R-adducts (linear products) has
been a subject of current interest (Figure 1). Recently,
some highly regioselective allylations giving R-adducts
have been acheived.2
The indium-mediated allylation reactions have been
of great importance from both economic and environ-
mental points of view.3 In general, allylindium reagents
have been prepared from allylic bromides or iodides and
indium.3 Recently, the alternative method for preparation
of allylindium reagents via transient organopalladium
intermediates has been studied by Araki et al.,4 and then
several successful examples of allylation of aldehydes
under new reaction conditions were reported by Grigg,
F IGURE 1. Regioselectivity in allylation.
Kang, and our groups.5 In these reactions, allylindium
reagents were generated via transmetalation of transient
π-allylpalladium intermediates with indium(I) halides.
However, the corresponding reaction of imine derivatives
has not been widely studied;6 therefore, the allylation of
imine derivatives through the umpolung process of
electrophilic palladium intermediates into nucleophilic
indium reagents has been a subject of current interest.
As a part of our program directed toward the develop-
ment of indium-mediated reaction of imine derivatives,7,8
(5) (a) Anwar, U.; Grigg, R.; Rasparini, M.; Savic, V.; Sridharan, V.
Chem. Commun. 2000, 645. (b) Anwar, U.; Grigg, R.; Sridharan, V.
Chem. Commun. 2000, 933. (c) Kang, S.-K.; Lee, S.-W.; J ung, J .; Lim,
Y. J . Org. Chem. 2002, 67, 4376. We reported InI-Pd(0)-promoted
metalation and addition of chiral 2-vinylaziridines. See: (d) Takemoto,
Y.; Anzai, M.; Yanada, R.; Fujii, N.; Ohno, H.; Ibuka, T. Tetrahedron
Lett. 2001, 42, 1725. (e) Anzai, M.; Yanada, R.; Fujii, N.; Ohno, H.;
Ibuka, T.; Takemoto, Y. Tetrahedron 2002, 58, 5231.
(6) Recently, the palladium-indium-mediated cascade allylation of
N-sulfonylimines was reported. See: Cooper, I. R.; Grigg, R.; Mac-
Lachlan, W. S.; Thornton-Pett, M.; Sridharan, V. Chem. Commun.
2002, 1372.
(7) (a) Miyabe, H.; Ueda, M.; Nishimura, A.; Naito, T. Org. Lett.
2002, 4, 131. (b) Miyabe, H.; Nishimura, A.; Ueda, M.; Naito, T. Chem.
Commun. 2002, 1454. (c) Yanada, R.; Kaieda, A.; Takemoto, Y. J . Org.
Chem. 2001, 66, 7516.
(8) For examples of metallic indium-mediated allylation of imines,
see: (a) Lu, W.; Chan, T. H. J . Org. Chem. 2001, 66, 3467. (b) Lu, W.;
Chan, T. H. J . Org. Chem. 2000, 65, 8589. (c) Chan, T. H.; Lu, W.
Tetrahedron Lett. 1998, 39, 8605. (d) Basile, T.; Bocoum, A.; Savoia,
D.; Umani-Ronichi, A. J . Org. Chem. 1994, 59, 7766. (e) Beuchet, P.;
Marrec, N. L.; Mosset, P. Tetrahedron Lett. 1992, 33, 5959. (f) Lee, J .
G.; Choi, K. I.; Pae, A. N.; Koh, H. Y.; Kang, Y.; Cho, Y. S. J . Chem.
Soc., Perkin Trans. 1 2002, 1314.
† Kyoto University.
‡ Kobe Pharmaceutical University.
(1) For reviews, see: (a) Yamamoto, Y.; Asao, N. Chem. Rev. 1993,
93, 2207. (b) Bloch, R. Chem. Rev. 1998, 98, 1407. (c) Marshall, J . A.
Chem. Rev. 2000, 100, 3163.
(2) For some examples, see: (a) Tan, K.-T.; Chng, S.-S.; Cheng, H.-
S.; Loh, T.-P. J . Am. Chem. Soc. 2003, 125, 2958. (b) Loh, T.-P.; Tan,
K.-T.; Hu, Q.-Y. Tetrahedron Lett. 2001, 42, 8705. (c) Ito, A.; Kishida,
M.; Kurusu, Y.; Masuyama, Y. J . Org. Chem. 2000, 65, 494. (d)
Yanagisawa, A.; Ogasawara, K.; Yasue, K.; Yamamoto, H. J . Chem.
Soc., Chem. Commun. 1996, 367. (e) Isaac, M. B.; Chan, T. H.
Tetrahedron Lett. 1995, 36, 8957. (f) Yamamoto, Y.; Saito, Y.; Maruya-
ma, K. J . Org. Chem. 1983, 48, 5408.
(3) For a recent review, see: Li, C. J .; Chan, T. H. Tetrahedron 1999,
55, 11149. For some examples of indium-mediated reactions, see: (a)
Yang, Y.; Chan, T. H. J . Am. Chem. Soc. 2000, 122, 402. (b) Chan, T.
H.; Yang, Y. J . Am. Chem. Soc. 1999, 121, 3228. (c) Paquette, L. A.;
Rothhaar, R. R. J . Org. Chem. 1999, 64, 217. (d) Woo, S.; Sqires, N.;
Fallis, A. G. Org. Lett. 1999, 1, 573. (e) Engstrom, G.; Morelli, M.;
Palomo, C.; Mitzel, T. Tetrahedron Lett. 1999, 40, 5967. (f) Loh, T.-P.;
Zhou, J . R. Tetrahedron Lett. 1999, 40, 9115.
(4) Araki, S.; Kamei, T.; Hirashita, T.; Yamamura, H.; Kawai, M.
Org. Lett. 2000, 2, 847.
10.1021/jo034451o CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/29/2003
J . Org. Chem. 2003, 68, 6745-6751
6745

Miyabe et al.
SCHEME 1
TABLE 1. Rea ction of 1 w ith Allyl Alcoh ol Der iva tives
we now report the reactions of electron-deficient imine
derivatives such as glyoxylic oxime ether9,10 and N-
tosylimine11 with allylic acetates and carbomates in the
presence of a catalytic amount of palladium(0) complex
and indium(I) iodide. We also report that the palladium-
indium iodide-mediated allylation of glyoxylic oxime
ether afforded either γ-adducts or R-adducts by simple
change of the reaction conditions, allowing the prepara-
tion of a variety of R-amino acids.
2a ,b a n d 3a -ga
entry reagent
solvent
time (h)
product
yieldb (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
3a
3a
3a
3a
3b
2a
2b
3b
2a
2b
3c
3d
3e
3f
THF
THF
0.5
20
3
20
3
3
3
20
20
20
3
3
3
3
3
20
20
20
20
20
syn-4A/5A (1:2) 74 (10)
5A
95
90
89
94
95
94
93
85
87
90
81
71
72
66
93
95
90
93
81
H2O-THF
H2O-THF
H2O-THF
H2O-THF
H2O-THF
THF
syn-4A
syn-4A
syn-4A
syn-4A
syn-4A
5A
THF
THF
5A
5A
Resu lts a n d Discu ssion
H2O-THF
H2O-THF
H2O-THF
H2O-THF
H2O-THF
THF
THF
THF
THF
THF
syn-4B
syn-4C
syn-4D
syn-4E
syn-4F
5B
Rea ction of Glyoxylic Oxim e Eth er . We first ex-
amined the reaction of Oppolzer’s camphorsultam deriva-
tive of glyoxylic oxime ether 1, which has shown excellent
reactivity in our previous work on allylation and radical
reactions using metallic indium.7b Thus, we expected that
the direct comparison of metallic indium-mediated reac-
tions with palladium-indium iodide-mediated reactions
would lead to informative and instructive suggestions
regarding the reactivity, regioselectivity, diastereoselec-
tivity, and absolute stereochemical course in the present
method.
3g
3c
3d
3e
3f
5C
5D
5E
5F
3g
a
Reactions were carried out with 1 and allyl alcohol derivatives
b
2a ,b and 3a -g at 20 °C. Isolated yields. Yield in parentheses is
that for the recovered starting material.
The reactions of 1 with allylic acetate 3a (2 equiv) were
carried out in the presence of Pd(PPh3)4 (0.05 equiv) and
indium(I) iodide (2 equiv) at 20 °C (Scheme 1). As
expected, glyoxylic oxime ether 1 exhibits a good reactiv-
ity to give the desired allylated products syn-4A and 5A,
without formation of significant byproducts such as a
reduced product. To our surprise, the selective formation
of R-adduct 5A was observed in anhydrous THF (Table
1). A 1:2 mixture of the γ-adduct syn-4A and the R-adduct
5A was obtained in 74% combined yield after being
stirred for 0.5 h, accompanied with 10% yield of the
starting material 1 (entry 1). The formation of the
R-adduct 5A was shown to be dependent on the reaction
time; thus, the prolonged reaction led to a selective
formation of the R-adduct 5A (entry 2). In our previous
studies on metallic indium-mediated reaction of 1 in
aqueous media, the γ-adducts were obtained as a single
regioisomer.7b Therefore, we next investigated the effect
of water on the palladium-indium iodide-mediated reac-
tion. In the presence of water, the γ-adduct syn-4A was
obtained in 90% yield as a single diastereomer without
the formation of the R-adduct 5A, after being stirred at
20 °C for 3 h (entry 3). Under the aqueous reaction
(9) We recently reported the radical addition to glyoxylic oxime
ether, see: Miyabe, H.; Ushiro, C.; Ueda, M.; Yamakawa, K.; Naito,
T. J . Org. Chem. 2000, 65, 176.
(10) The zinc-mediated allylation reaction of glyoxylic oxime ether
reported by Hanessian’s group. See (a) Hanessian, S.; Yang, R.-Y.
Tetrahedron Lett. 1996, 37, 5273. (b) Hanessian, S.; Bernstein, N.;
Yang, R.-Y.; Maguire, R. Bioorg. Med. Chem. Lett. 1999, 9, 1437. (c)
Hanessian, S.; Lu, P.-P.; Sanceau, J .-Y.; Chemla, P.; Gohda, K.; Fonne-
Pfister, R.; Prade, L.; Cowan-J acob, S. W. Angew. Chem., Int. Ed. 1999,
38, 3160.
6746 J . Org. Chem., Vol. 68, No. 17, 2003

Allylation Reaction of Glyoxylic Oxime Ether and N-Sulfonylimine
SCHEME 2
SCHEME 3
spectral data.7b,10b The absolute configuration of major
1
product 5A was assigned to be S since its H NMR data
showed similarity with that of the major isomer of
allylated compound of 1 obtained by our previous studies.7b
To study the reversibility in this reaction, the γ-adduct
syn-4A was treated with Pd(PPh3)4 and indium(I) iodide
in anhydrous THF. However, the formation of R-adduct
5A was not observed. Additionally, treatment of syn-4A
with n-BuLi did not lead to the formation of R-adduct
5A.
To learn about the effect of allylic reagents, we next
examined the reactions of 1 with other allylic reagents.
The branched and linear reagents 3b and 2a ,b would give
the same transient organopalladium intermediate, pro-
viding same allylic indium reagent was used. The reac-
tions with 3b and 2a ,b proceeded smoothly in H2O-THF
(1:10, v/v) to give the γ-adduct syn-4A without formation
of the R-adduct 5A (entries 5-7). Allylic reagents 3b and
2a ,b also worked well in anhydrous THF to give the
R-adduct 5A (entries 8-10). The advantage of the reac-
tion using branched allylic acetates such as 3a is that a
wide range of branched allylic acetates can be prepared
by the reaction of corresponding aldehydes with vinyl-
magnesium bromide followed by acetylation. We next
investigated the reaction of 1 with the aromatic allylic
acetates 3c-g. As expected, the γ-adducts syn-4B-F
were obtained in H2O-THF (1:10, v/v) and the R-adducts
5B-F were obtained in anhydrous THF (entries 11-20).
The absolute configuration of 4B-F and 5B-F was
assigned since these 1H NMR data showed similarity
with those of 4A and 5A. The hydrolysis of γ-adducts syn-
4B-F into oxime ether 1 and the corresponding allyl
alcohols was frequently observed, as exemplified in the
conversion of γ-adduct syn-4B into oxime ether 1 and the
corresponding alcohol, which was confirmed by the
authentic compound,14 in CDCl3 (Scheme 3). These
results also support the reversibility in the present
reaction.
In contrast to monosubstituted allylic reagents 2a ,b
and 3a -g, the reaction with disubstituted allylic reagents
2c,d and 3h -j gave exclusively γ-adducts 4G and 4H as
a single diastereomer even in anhydrous THF (Scheme
4). The absolute configuration of 4G was determined to
be S by comparison with authentic spectral data.7b,10b
These observations indicate that the allylic rearrange-
ment of indium atom from the linear indium reagent H
to the branched indium reagent I would be suppressed
due to steric effects; thus, R-adducts were not obtained
from I via the chelated six-membered ring transition
state (Scheme 5). The chemical efficiency giving γ-adduct
4G was shown to be dependent on the structure in allyl
reagents (Table 2). In the case of linear γ,γ-dimethylallyl
acetate 2c, a similar reaction procedure did not give a
good result (entry 1). The reaction of 1 with linear allyl
carbonate 2d proceeded slowly to give the desired product
conditions, the formation of the R-adduct 5A was not
observed after being stirred even for 20 h (entry 4).
Although the precise reason for the effect of water was
unclear, these observations can be explained by the
reversibility of allylation reaction (Scheme 2). The re-
versibility in allylation reaction of imines using other
allylmetals has been observed.2d,12 Under the anhydrous
reaction conditions, the prolonged reaction would allow
the reversibility between the γ-adduct E and the linear
indium reagent A, giving the branched indium reagent
B via the allylic rearrangement of indium atom.13 Finally,
the kinetically formed adduct E isomerized to the ther-
modynamically stable R-adduct F . Recently, a new mech-
anism based on [3,3]-sigmatropic rearrangement was
reported.5a,b Thus, an alternative mechanistic hypothesis
involving [3,3]-sigmatropic rearrangement would not be
rigorously excluded. In contrast, water suppress the
reversibility between adduct E and the indium reagent
A, as a result of the quick trapping reaction of the
branched adduct E with H2O. As an effect of water, an
alternative mechanism involving an open transition state
in the presence of water would not be excluded. The
absolute configuration of major product 4A was deter-
mined to be S and syn by comparison with authentic
(11) We recently reported that N-sulfonylimines have shown the
excellent reactivity toward nucleophilic carbon radical. See: (c) Miyabe,
H.; Ueda, M.; Naito, T. Chem. Commun. 2000, 2059.
(12) (a) Miginiac, L.; Mauze´, B. Bull. Soc. Chim. Fr. 1968, 4674. (b)
Mauze´, B.; Miginiac, L. Bull. Soc. Chim. Fr. 1973, 1832. (c) Mauze´,
B.; Miginiac, L. Bull. Soc. Chim. Fr. 1973, 1838. (b) Mauze´, B.;
Miginiac, L. Bull. Soc. Chim. Fr. 1973, 1082.
(13) Isomerization of indium reagents via the allylic rearrangement
of indium atom was reported. See: Hirashita, T.; Hayahi, Y. Mitsui,
K.; Araki, S. J . Org. Chem. 2003, 68, 1309.
(14) Engler, T. A.; LaTessa, K. O.; Iyengar, R.; Chai, W.; Agrios, K.
Bioorg. Med. Chem. 1996, 4, 1755.
J . Org. Chem, Vol. 68, No. 17, 2003 6747

Miyabe et al.
SCHEME 4
SCHEME 6
SCHEME 5
TABLE 2. Rea ction of 1 w ith Allyl Alcoh ol Der iva tives
2c,d a n d 3h -ja
entry reagent
solvent
THF
THF
THF
time (h) product
yieldb (%)
1
2c
3
4
5
6
7
8
2c
2c
2d
3h
3h
3i
20
20
20
20
3
4G
4G
4G
4G
4G
trace
trace
27 (61)
87
85
no reaction
12 (80)
no reaction
THF
H2O-THF
H2O-THF
THF
20
10
30
3i
3j
4H
THF
a
Reactions were carried out with 1 and allyl alcohol derivatives
b
2c,d and 3h -j. Isolated yields. Yields in parentheses are that
for the recovered starting material. c Reaction was carried out with
1 and 2c at 70 °C.
4G in 27% yield, accompanied with 70% yield of starting
compound 1 (entry 3). In contrast, the reaction with
branched R,R-dimethylallyl acetate 3h gave the γ-adduct
4G in 87% yield as a single diastereomer after being
stirred at 20 °C for 20 h (entry 4). These results indicate
that the branched allyl acetate 3h shows excellent
reactivity toward Pd(0) in the formation process of
transient organopalladium intermediates G (Scheme 5).
In the presence of water, the reaction with 3h also
proceeded smoothly to give 4G (entry 5). In the case of
bulky acetate 3i, the reaction proceeded slowly in anhy-
drous THF to give the γ-adduct 4H in 12% yield,
accompanied with 80% yield of starting compound 1,
although the reaction did not occur in H2O-THF (entries
6 and 7). However, the branched R,R-diphenylallyl ac-
etate 3j did not worked (entry 8).
To survey the scope and limitations of the present
reaction, the reaction of 1 with allylic acetates 3k -m was
studied (Scheme 6). The reaction with aliphatic allylic
acetate 3k in H2O-THF gave the syn/anti mixture of
γ-adduct 4I, accompanied with a small amount of R-ad-
duct 5I. The similar trend in syn/anti selectivity was
observed in the zinc-mediated reaction of 1 with aliphatic
allylic bromide.10b In contrast, the reaction with 3k in
anhydrous THF gave selectively the R-adduct 5I. The
acetates having additional olefin moiety 3l and 3m
worked. The γ-adducts syn-4J and syn-4K were obtained
from the indium reagent J as a single isomer in H2O-
6748 J . Org. Chem., Vol. 68, No. 17, 2003

Allylation Reaction of Glyoxylic Oxime Ether and N-Sulfonylimine
SCHEME 7 SCHEME 9
SCHEME 8
TABLE 3. Rea ction of 6 w ith Allyl Alcoh ol Der iva tivesa
entry
reagent
solvent
product
yieldb (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
8
THF
THF
THF
THF
THF
THF
THF
THF
7A
86
66
77
65
68
85
trace
21
36
58
65
31
16
2a
2b
3a
3b
3c
2c
2d
3h
3a
3c
3a
3h
syn-7Bc
syn-7Bc
syn-7Bc
syn-7Bc
syn-7Cc
7D
7D
7D
THF
H2O-THF
H2O-THF
H2O-EtOH
H2O-THF
syn-7Bc
syn-7Cc
syn-7Bc
7D
a
Reactions were carried out with 6 and allyl alcohol derivatives
at 20 °C. b Isolated yields. c Only syn isomers were obtained (>95%
de).
THF (Scheme 7).15 In contrast, the reactions in anhydrous
THF gave the R-adducts 5J and 5K via the indium
reagent K.16 Thus, the γ-adducts and R-adducts could be
obtained by change of the reaction conditions.
The reaction with R,γ-disubstituted acetate 3n in
H2O-THF gave selectively the product syn-4L in 77%
yield as a single diastereomer, as a result of the reaction
with the stable indium reagent M, although the reaction
with 3n in anhydrous THF gave a complex mixture
(Scheme 8).
gated the allylation of N-sulfonylimine 6 under the
similar reaction conditions (Scheme 9).6,17 Although the
reaction of 6 proceeded smoothly, the formation of R-ad-
ducts was not observed. In all case, the γ-adducts were
obtained even in anhydrous THF. These observations
indicate that the allylation of N-sulfonylimine 6 was not
a reversible process due to the extra stabilization of
indium-bonding adduct P by the electron-withdrawing
N-sulfonyl group. The reaction of 6 with allyl acetate 8
proceeded smoothly in THF at 20 °C to give 86% yield of
the product 7A (Table 3, entry 1). In contrast to glyoxylic
oxime ether 1, the reactions of 6 with substituted allylic
acetates and carbonates 2a ,b and 3a -c proceeded
smoothly to give the γ-adducts syn-7B and syn-7C,
Rea ction of N-Su lfon ylim in e. In comparison with
the reaction of glyoxylic oxime ether 1, we next investi-
(15) At present, the rigorous stereochemical assignment of the
mixture of diastereomers 4K was not achieved. However, a high degree
of stereocontrol regarding the stereocenter on R-carbon was generally
observed by using camphorsultam derivative of glyoxylic oxime ether
1. See refs 7b, 9, and 10.
(16) The diastereoselectivities of products 5J and 5K were deter-
mined after hydrogenolysis of the olefin moiety, which gave the single
diastereoisomer, respectively.
(17) Chan’s group reported the metallic indium-mediated allylation
of N-sulfonylimines in aqueous media through an allylindium(I)
intermediate. See ref 8a-c.
J . Org. Chem, Vol. 68, No. 17, 2003 6749

Miyabe et al.
14.1 Hz), 2.47 (2H, br m), 2.10-1.78 (5H, m), 1.46-1.23
(2H, m), 0.94, 0.91 (each 3H, s); 13C NMR (CDCl3) δ 173.2,
159.0, 137.8, 132.6, 129.9, 128.6, 128.2, 127.6, 127.4, 121.9,
113.8, 75.9, 65.1, 62.9, 55.3, 53.1, 48.5, 47.7, 44.6, 38.3, 34.3,
32.8, 26.4, 20.7, 19.9; MS (FAB) m/z 525 (M + H+, 35), 154
(100); HRMS calcd for C29H37N2O5S (M + H+) 525.2423, found
525.2409.
accompanied by a trace amount of anti isomer, without
the formation of the R-adducts (entries 2-6). The bulky
γ,γ-dimethylallyl acetate 2c and carbonate 2d were less
effective for the allylation reaction of N-sulfonylimine 6
(entries 7 and 8). The reaction of 6 with R, R-dimethylallyl
acetate 3h gave the γ-adduct 7D in 36% yield (entry 9).
The reaction of N-sulfonylimine 6 also proceeded in
aqueous media (entries 10-13).
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-5-(2-m eth oxyph en yl)-1-oxo-2-[(ph en ylm eth oxy)am i-
n o]-4-p en ten yl]-3H-3a ,6-m eth a n o-2,1-ben zisoth ia zole 2,2-
d ioxid e (5C): colorless crystal; mp 133-136 °C (AcOEt/
Con clu sion s
hexane); [R]25 -85.0 (c 1.05, CHCl3); IR (CHCl3) 3032, 1693
D
cm-1; H NMR (CDCl3) δ 7.43-7.22 (6H, m), 7.18 (1H, t, J )
1
We have demonstrated that the palladium-indium
iodide-mediated reaction of glyoxylic oxime ether afforded
either the γ-adducts or the R-adducts by change of the
reaction conditions. In contrast, the reactions of N-
sulfonylimine proceeded regioselectively to provide the
γ-adducts. In addition to the radical reaction of glyoxylic
oxime ether,7,9 the present reaction disclosed a broader
aspect of the utility of glyoxylic oxime ether for the
asymmetric synthesis of various types of R-amino acid
derivatives.
6.4 Hz) 6.88 (1H, t, J ) 6.4 Hz) 6.82 (1H, d, J ) 8.2 Hz), 6.71
(1H, d, J ) 15.9 Hz) 6.20 (1H, d, J ) 11.0 Hz) 6.10 (1H, m),
4.74, 4.68 (2H, AB q, J ) 10.6 Hz), 4.60-4.48 (1H, m), 4.01-
3.92 (1H, m), 3.79 (3H, s), 3.53-3.40 (2H, m), 2.59-2.48
(2H, m), 2.05-2.03 (2H, br m), 1.94-1.79 (2H, m), 1.79-1.72
(1H, s), 1.43-1.26 (2H, m) 0.95, 0.90 (each 3H, s); 13C NMR
(CDCl3) δ 173.2, 156.4, 137.9, 128.6, 128.1, 127.6, 126.9, 126.1,
124.7, 120.5, 110.6, 75.8, 65.0, 62.8, 55.3, 53.0, 48.4, 47.6, 44.5,
38.1, 34.7, 32.7, 26.3, 20.5, 19.7; MS (FAB) m/z 525 (M + H+,
100); HRMS calcd for C29H37N2O5S (M+) 525.2423, found
525.2438. Anal. Calcd for C29H36N2O5S: C, 66.39; H 6.92; N
5.34; S, 6.11. Found: C, 66.24; H 6.88; N 5.06; S, 6.16. Some
peaks of 13C NMR were missing due to overlap.
Exp er im en ta l Section
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-5-(4-m eth ylph en yl)-1-oxo-2-[(ph en ylm eth oxy)am in o]-
4-p en t en yl]-3H -3a ,6-m et h a n o-2,1-b en zisot h ia zole 2,2-
d ioxid e (5D): colorless crystal; mp 157-161 °C (AcOEt/
Gen er a l P r oced u r e for Allyla tion of Oxim e Eth er 1 in
THF . A mixture of 1 (47.1 mg, 0.125 mmol), allylic reagents
(0.25 mmol), Pd(PPh3)4 (7.3 mg, 0.0063 mmol), and indium
iodide (60 mg, 0.25 mmol) in THF (1.0 mL) was stirred under
argon atmosphere at 20 °C for 20 h. The reaction mixture was
diluted with saturated aqueous potassium sodium (+)-tartrate
and then extracted with AcOEt. The organic phase was dried
over MgSO4 and concentrated at reduced pressure. Purification
of the residue by preparative TLC (hexane/AcOEt ) 4:1, 2-fold
development) afforded the R-adducts.
Gen er a l P r oced u r e for Allyla tion of Oxim e Eth er 1 in
H2O-THF (1:10, v/v). A mixture of 1 (47.1 mg, 0.125 mmol),
allylic reagents (0.25 mmol), Pd(PPh3)4 (7.3 mg, 0.0063 mmol),
and indium iodide (60 mg, 0.25 mmol) in H2O-THF (1:10, v/v,
1.1 mL) was stirred under argon atmosphere at 20 °C for 3 h.
The reaction mixture was diluted with saturated aqueous
potassium sodium (+)-tartrate and then extracted with AcOEt.
The organic phase was dried over MgSO4 and concentrated at
reduced pressure. Purification of the residue by preparative
TLC (hexane/AcOEt ) 4:1, 2-fold development) afforded the
γ-adducts.
hexane); [R]25 -67.0 (c 0.93, CHCl3); IR (CHCl3) 1693 cm-1
;
D
1H NMR (CDCl3) δ 7.43-7.22 (5H, m), 7.18 (2H, d, J
)
7.4 Hz), 7.08 (2H, d, J ) 7.4 Hz), 6.34 (1H, d, J ) 15.3 Hz),
6.19 (1H, br d, J ) 10.7 Hz), 6.06 (1H, m), 4.73, 4.68 (2H, AB
q, J ) 11.6 Hz), 4.52 (1H, br m), 3.96 (1H, br m), 3.48, 3.46
(2H, AB q, J ) 14.4 Hz), 2.47 (2H, br m), 2.32 (3H, s), 2.07-
1.75 (5H, m), 1.47-1.25 (2H, m), 0.95, 0.91 (each 3H, s); 13C
NMR (CDCl3) δ 173.2, 137.8, 137.0, 134.2, 133.0, 129.0, 128.7,
128.1, 127.6, 126.2, 123.1, 75.9, 65.1, 62.8, 53.1, 48.5, 47.7, 44.5,
38.3, 34.3, 32.8, 26.4, 21.2, 20.7, 19.8; MS (FAB) m/z 509
(M + H+, 100); HRMS calcd for C29H37N2O4S (M + H+)
509.2474, found 509.2482. Anal. Calcd for C29H36N2O4S: C,
68.47; H, 7.13; N, 5.51; S, 6.30. Found: C, 68.20; H, 7.07; N,
5.25; S, 6.37.
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-5-(2-m eth ylph en yl)-1-oxo-2-[(ph en ylm eth oxy)am in o]-
4-p en t en yl]-3H -3a ,6-m et h a n o-2,1-b en zisot h ia zole 2,2-
d ioxid e (5E): colorless crystal; mp 129-132 °C (AcOEt/
The diastereoselectivities of products were determined by
1H NMR analysis of diastereomeric mixture obtained after
rough purification of chromatography (hexame/AcOEt ) 2:1).
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-1-oxo-5-p h en yl-2-[(p h en ylm et h oxy)a m in o]-4-p en -
ten yl]-3H-3a ,6-m eth a n o-2,1-ben zisoth ia zole 2,2-d ioxid e
hexane); [R]25 -75.8 (c 0.87, CHCl3); IR (CHCl3) 3032, 1693
D
cm-1
;
1H NMR (CDCl3) δ 7.37-7.24 (6H, m), 7.13-7.10
(3H, m), 6.69 (1H, d, J ) 15.6 Hz), 6.20 (1H, m), 6.00 (1H, m),
4.74, 4.67 (2H, AB q, J ) 11.9 Hz), 4.54 (1H, m), 3.97 (1H, br
t), 3.48 (2H, m) 2.55 (2H, m), 2.26 (3H, s), 2.05 (2H, m), 2.19-
1.79 (3H, m), 1.42-1.30 (2H, m) 1.26, 1.15 (each 3H, s); 13C
NMR (CDCl3) δ 173.0, 137.8, 136.0, 135.0, 131.1, 131.0, 128.6,
128.1, 127.6, 127.2, 125.9, 125.7, 125.5, 75.9, 65.0, 62.9, 62.7,
53.0, 48.5, 47.7, 44.5, 38.2, 34.5, 32.7, 26.2, 20.6, 19.7; MS
(FAB) m/z 509 (M + H+, 100); HRMS calcd for C29H37N2O4S
(5A): colorless oil; [R]33 -59.4 (c 0.94, CHCl3); IR (CHCl3)
D
3270, 1692, 1601 cm-1
;
1H NMR (CDCl3) δ 7.39-7.15 (10H,
m), 6.38 (1H, d, J ) 15.6 Hz), 6.13 (1H, dt, J ) 15.6, 7.0 Hz),
4.73, 4.68 (2H, AB q, J ) 11.6 Hz), 4.54 (1H, br s), 3.96 (1H,
br t, J ) 6.1 Hz), 3.48, 3.45 (2H, AB q, J ) 13.8 Hz), 2.51-
2.45 (2H, m), 2.10-1.74 (5H, m), 1.46-1.26 (2H, m), 0.93, 0.90
(each 3H, s); 13C NMR (CDCl3) δ 173.1, 137.8, 137.0, 133.1,
128.6, 128.4, 128.1, 127.6, 127.3, 126.3, 124.3, 75.9, 65.1, 62.8,
53.0, 48.5, 47.6, 44.5, 38.2, 34.3, 32.8, 26.3, 20.6, 19.8; MS
(FAB) m/z 495 (M + H+, 90), 91 (100); HRMS calcd for
(M + H+) 509.2474, found 509.2479. Anal. Calcd for C29H36
-
N2O4S: C, 68.47; H 7.13; N 5.51; S, 6.30. Found: C, 68.25; H
7.06; N 5.27; S, 6.42.
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-5-(4-flu or oph en yl)-1-oxo-2-[(ph en ylm eth oxy)am in o]-
4-p en t en yl]-3H -3a ,6-m et h a n o-2,1-b en zisot h ia zole 2,2-
d ioxid e (5F ): colorless crystal; mp 157-161 °C (AcOEt/
C
28H35N2O4S (M + H+) 495.2318, found 495.2327.
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-5-(4-m eth oxyph en yl)-1-oxo-2-[(ph en ylm eth oxy)am i-
n o]-4-p en ten yl]-3H-3a ,6-m eth a n o-2,1-ben zisoth ia zole 2,2-
hexane); [R]25 -67.0 (c 0.93, CHCl3); IR (CHCl3) 1693 cm-1
;
D
1H NMR (CDCl3) δ 7.43-7.22 (5H, m), 7.18 (2H, d, J
)
d ioxid e (5B): colorless oil; [R]34 -130 (c 2.17, CHCl3); IR
7.4 Hz), 7.08 (2H, d, J ) 7.4 Hz), 6.34 (1H, d, J ) 15.3 Hz),
6.19 (1H, br d, J ) 10.7 Hz), 6.06 (1H, m), 4.73, 4.68 (2H, AB
q, J ) 11.6 Hz), 4.52 (1H, br m), 3.96 (1H, br m), 3.48, 3.46
(2H, AB q, J ) 14.4 Hz), 2.47 (2H, br m), 2.32 (3H, s), 2.07-
1.75 (5H, m), 1.47-1.25 (2H, m), 0.95, 0.91 (each 3H, s); 13C
D
(CHCl3) 3270, 1692, 1606 cm-1; 1H NMR (CDCl3) δ 7.47-7.20
(7H, m), 6.81 (2H, d, J ) 8.5 Hz), 6.32 (1H, d, J ) 15.9 Hz),
5.97 (1H, m), 4.74, 4.69 (2H, AB q, J ) 11.9 Hz), 4.53 (1H, br
s), 3.96 (1H, br m), 3.80 (3H, s), 3.48, 3.45 (2H, AB q, J )
6750 J . Org. Chem., Vol. 68, No. 17, 2003

Allylation Reaction of Glyoxylic Oxime Ether and N-Sulfonylimine
NMR (CDCl3) δ 173.2, 137.8, 137.0, 134.2, 133.0, 129.0, 128.7,
128.1, 127.6, 126.2, 123.1, 75.9, 65.1, 62.8, 53.1, 48.5, 47.7, 44.5,
38.3, 34.3, 32.8, 26.4, 21.2, 20.7, 19.8; MS (FAB) m/z 509
(M + H+, 100); HRMS calcd for C29H37N2O4S (M + H+)
509.2474, found 509.2482. Anal. Calcd for C29H36N2O4S: C,
68.47; H, 7.13; N, 5.51; S, 6.30. Found: C, 68.20; H, 7.07; N,
5.25; S, 6.37.
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-5-cycloh exyl-1-oxo-2-[(p h en ylm et h oxy)a m in o]-4-
p en t en yl]-3H -3a ,6-m et h a n o-2,1-b en zisot h ia zole 2,2-d i-
38.2, 38.1, 34.1, 32.8, 28.6, 26.4, 26.3, 26.2, 25.8, 20.7, 19.8,
18.1 (× 2); MS (FAB) m/z 473 (M + H+, 100); HRMS calcd for
C
26H37N2O4S (M + H+) 473.2474, found 473.2478. Some peaks
of 13C NMR were missing due to overlap.
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-1-oxo-7-p h en yl-2-[(p h en ylm eth oxy)a m in o]-4,6-h ep -
ta d ien yl]-3H-3a ,6-m eth a n o-2,1-ben zisoth ia zole 2,2-d iox-
id e (5K): amorphous substance as a 6:1 E/Z mixture; [R]26
D
-54.1 (c 1.37, CHCl3); IR (CHCl3) 3280, 1730, 1693, 1601
cm-1;1H NMR (CDCl3) δ 7.42-7.17 (10H, m), 6.97 (1/7H, dd,
J ) 14.9, 11.6 Hz), 6.67 (6/7H, dd, J ) 15.5, 10.7 Hz), 6.53
(1/7H, d, J ) 15.8 Hz), 6.43 (6/7H, d, J ) 15.8 Hz), 6.25-6.12
(2H, m), 5.72 (6/7H, m), 5.47 (1/7H, m), 4.73, 4.67 (2H, AB q,
J ) 11.6 Hz), 4.51 (1H, br s), 3.96 (1H, br m), 3.50, 3.47 (2H,
AB q, J ) 13.7 Hz), 2.72 (1/7H, m), 2.47 (6/7H, m), 2.43 (1H,
m), 2.15-1.76 (5H, m), 1.48-1.20 (2H, m), 1.11, 0.94 (each 3H,
s); 13C NMR (CDCl3) δ 173.1, 137.8, 137.3, 133.8, 131.5, 128.7,
128.6, 128.4, 128.2, 127.6, 127.4, 126.5, 126.3, 75.9, 65.0, 62.7,
53.0, 48.5, 47.7, 44.5, 38.2, 34.0, 32.7, 26.3, 20.7. 19.8; MS
(FAB) m/z 521 (M + H+, 100); HRMS calcd for C30H37N2O4S
(M + H+) 521.2474, found 521.2470. 13C NMR values are for
the major E-isomer.
oxid e (5I): colorless crystal; mp 96-98 °C (AcOEt/hexane);
1
[R]27 -74.5 (c 1.08, CHCl3); IR (CHCl3) 3280, 1694 cm-1; H
D
NMR (CDCl3) δ 7.40-7.22 (5H, m), 6.13 (1H, br d, J
)
7.9 Hz), 5.39 (1H, dd, J ) 15.3, 6.4 Hz), 5.25 (1H, m), 4.71,
4.66 (2H, AB q, J ) 11.9 Hz), 4.41 (1H, br s), 3.96 (1H, br m),
3.49, 3.46 (2H, AB q, J ) 13.7 Hz), 2.34, 2.26 (each 1H, m),
2.10-2.00 (2H, m), 1.95-1.78 (4H, m), 1.69-1.55 (5H, m),
1.48-0.90 (7H, m), 1.14, 0.97 (each 3H, s); 13C NMR (CDCl3)
δ 173.0, 140.5, 137.9, 128.6, 128.1, 127.6, 121.1, 75.8, 64.9, 62.9,
53.0, 48.5, 47.7, 44.5, 40.5, 38.3, 33.7, 32.8, 32.6, 26.3, 26.0,
25.9, 20.8, 19.8; MS (FAB) m/z 91 (100), 501 (M + H+, 98);
HRMS calcd for C28H41N2O4S (M + H+) 501.2787, found
501.2782. Anal. Calcd for C28H40N2O4S: C, 67.17; H, 8.05; N,
5.59; S, 6.40. Found: C, 67.39; H, 7.89; N, 5.57; S, 6.69.
(3a S,6R,7a R)-1,4,5,6,7,7a -H exa h yd r o-8,8-d im et h yl-1-
[(2S)-5-m et h yl-1-oxo-2-[(p h en ylm et h oxy)a m in o]-4-h ex-
en yl]-3H -3a ,6-m et h a n o-2,1-b en zisot h ia zole 2,2-d ioxid e
Ack n ow led gm en t. This work was supported in part
by The J apan Health Sciences Foundation and Grant-
in-Aid for Scientific Research (C) (Y.T.) and a Grant-
in-Aid for Scientific Research (B) (T.N.) from the Min-
istry of Education, Culture, Sports, Science and Tech-
nology of J apan and the Science Research Promotion
Fund of the J apan Private School Promotion Foundation
for research grants.
(5J ): colorless oil as a 1:1 E/Z mixture; [R]24 -94.0 (c 1.31,
D
CHCl3); IR (CHCl3) 3280, 1693 cm-1;1H NMR (CDCl3) δ 7.43-
7.23 (5H, m), 6.23 (1H, m), 6.14 (1H, m), 5.97 (1/2H, d, J )
11.6 Hz), 5.72 (1/2H, d, J ) 11.0 Hz), 5.42 (1/2H, m), 5.26 (1/2H,
m), 4.73, 4.67 (1/2 × 2H, AB q, J ) 11.9 Hz), 4.72, 4.66 (1/2 ×
2H, AB q, J ) 11.9 Hz), 4.47 (1H, br s), 3.96 (1H, br m), 3.50,
3.46 (2H, AB q, J ) 13.7 Hz), 2.58 (1/2H, m), 2.39 (1H, m), 2.35
(1/2H, m), 2.13-1.78 (5H, m), 1.77, 1.74, 1.71, 1.69 (each 1/2 ×
3H, s), 1.46-1.25 (2H, m), 1.13, 1.11 (each 1/2 × 3H, s), 0.96
(3H, s); 13C NMR (CDCl3) δ 173.3, 173.2, 137.8, 136.6, 134.4,
130.3, 128.7, 128.2, 127.7, 127.6, 124.7, 124.3, 122.0, 120.0,
75.8, 65.0 (× 2), 62.8, 62.7, 53.0, 48.5 (× 2), 47.7 (× 2), 44.6,
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal procedure for reaction of N-sulfonylimine 6 and character-
ization data for γ-adducts 4A-L and 7A-D and 1H and 13C
NMR spectra of all obtained compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O034451O
J . Org. Chem, Vol. 68, No. 17, 2003 6751