A new entry to asymmetric synthesis of optically active α,γ-substituted γ-butyrolactones, using a carbohydrate derived amide as both a chiral auxiliary and a proton source

Article abstract of DOI:10.1021/jo0485637

(Chemical Equation Presented) A new entry for the asymmetric synthesis of optically active α,γ-substituted γ-butyrolactones was developed by using a carbohydrate-derived amide as both a chiral auxiliary and a proton source. Unlike the previously reported examples, the chiral auxiliary employed in this reaction also successfully functioned as a protonating agent. Excellent asymmetric induction could be achieved by this dual stereoselective control in the reaction process.

Full text of DOI:10.1021/jo0485637

A New Entry to Asymmetric Synthesis of Optically Active
r,γ-Substituted γ-Butyrolactones, Using a Carbohydrate Derived
Amide as Both a Chiral Auxiliary and a Proton Source
Ling-Lin Huang, Ming-Hua Xu,* and Guo-Qiang Lin
Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 354 Fenglin Lu,
Shanghai 200032, China, and Graduate School of the Chinese Academy of Sciences,
Shanghai 200032, China
xumh@mail.sioc.ac.cn
Received August 17, 2004
A new entry for the asymmetric synthesis of optically active R,γ-substituted γ-butyrolactones was
developed by using a carbohydrate-derived amide as both a chiral auxiliary and a proton source.
Unlike the previously reported examples, the chiral auxiliary employed in this reaction also
successfully functioned as a protonating agent. Excellent asymmetric induction could be achieved
by this dual stereoselective control in the reaction process.
Introduction
The SmI -mediated reductive coupling reaction of
SCHEME 1
2
aldehydes or ketones with chiral R,â-unsaturated esters
is very efficient for the synthesis of chiral γ-butyrolac-
tones that are important intermediates and building
1,2
blocks for fine chemicals and pharmaceuticals. This
3
approach, pioneered by Fukuzawa et al., has a wide
applicability. We recently reported a useful reaction
system for the asymmetric synthesis of optically active
R,γ-substituted γ-butyrolactones by the reaction of ke-
tones with 2-alkyl acrylates derived from various chiral
4
auxiliaries in the presence of protonating agents. Among
the chiral auxiliaries developed, we found that inexpen-
sive and easily accessible carbohydrate derivatives were
4
especially useful (Scheme 1). In this strategy, however,
both chiral auxiliary and bulky protonating agents such
as camphorsultam are essential for the observed excellent
enantioselectivites. To simplify the reaction system fur-
ther, we envisioned a more direct access between ketone
and chiral R,â-unsaturated ester if the employed chiral
auxiliary itself could also serve as a proton source in the
reaction. Furthermore, the chiral auxiliary that also
2
(1) For review articles on the application of SmI in organic
synthesis, see: (a) Molander, G. A. Chem. Rev. 1992, 92, 29. (b)
Molander, G. A.; Harris, C. R. Chem. Rev. 1996, 96, 307. (c) Molander,
G. A.; Harris, C. R. Tetrahedron 1998, 54, 3321. (d) Skrydstrup, T.
Angew. Chem., Int. Ed. Engl. 1997, 36, 345. (e) Krief, A.; Laval, A. M.
Chem. Rev. 1999, 99, 745. (f) Steel, P. G. J. Chem. Soc., Perkin Trans.1
2
001, 2727. (g) Kagan, H. B. Tetrahedron 2003, 59, 10351.
(
2) For γ-butyrolactones, see: (a) Depre, D.; Chen, L.-Y.; Ghosez,
L. Tetrahedron 2003, 59, 6797. (b) McMahon, L. R.; Coop, A.; France,
C. P.; Winger, G.; Woolverton, W. L. Eur. J. Pharmacol. 2003, 466,
(4) (a) Xu, M.-H.; Lin, G.-Q.; Xia, L.-J. Chin. J. Chem. 1998, 16,
561. (b) Xu, M.-H.; Wang, W.; Lin, G.-Q. Org. Lett. 2000, 2, 2229. (c)
Wang, W.; Xu, M.-H.; Lei, X.-S.; Lin, G.-Q. Org. Lett. 2000, 2, 3773.
(d) Xu, M.-H.; Wang, W.; Xia, L.-J.; Lin, G.-Q. J. Org. Chem. 2001, 66,
3953. (e) Wang, W.; Zhong, Y.-W.; Lin, G.-Q. Tetrahedron Lett. 2003,
44, 4613.
1
13. (c) Katoh, T.; Nishide, K.; Node, M.; Hiroo, O. Heterocycles 1999,
5
0, 833. (d) Drioli, S.; Felluga, F.; Forzato, C.; Nitti, P.; Pitacco, G.;
Valentin, E. J. Org. Chem. 1998, 63, 2385.
(3) Fukuzawa, S.; Seki, K.; Tatsuzawa, M.; Mutoh, K. J. Am. Chem.
Soc. 1997, 119, 1482.
1
0.1021/jo0485637 CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/17/2004
J. Org. Chem. 2005, 70, 529-532
529

Huang et al.
SCHEME 2
TABLE 1. Enantioselective Synthesis of
functions as a protonating agent might favor stereocon-
trol in the reaction process and thus potentially improve
the stereoselection of the reaction.
r-Methyl-γ-methyl-γ-anisyl-γ-butyrolactone by Reductive
Coupling of Novel Chiral Methacrylates to
4′-Methoxyacetophenone
entry ester trans/cisa trans ee (%)b cis ee (%)b yield (%)c
Results and Discussion
1
2
3
4
5
6
7
8
3a
3b
3c
4a
4b
4c
4d
4e
72/28
65/35
71/29
31/69
67/33
75/25
25/75
33/67
95 (+)
94 (+)
96 (+)
77 (+)
99 (+)
71 (+)
90 (+)
85 (+)
29 (-)
20 (-)
57
60
44
91
36
82
59
57
In our previous studies, the wedge-shaped isosorbide
skeleton showed good performance on the induction of
high enantioselectivites, so it might be anticipated that
the same skeleton, if bearing an appropriately active
proton, as indicated in Figure 1, would also afford the
>99 (-)
87 (+)
81 (+)
79 (+)
>99 (+)
99 (+)
a
Trans and cis isomers were separated by column chromatog-
1
1
raphy and confirmed by H- H NOESY in light of their NOE
b
effect; the ratio of trans/cis was determined by HPLC or GC. The
ee values were determined by HPLC analysis on a Chiralcel
c
column. The sign of [R]D is given in parentheses. Total isolated
yield of trans and cis products.
tivity of the reaction could be readily tuned by choosing
the NHR group oriented exo or endo in the molecule.
To check the potential of this hypothesis, we first
synthesized novel methacrylate compounds 3a and 3b
from readily available carbohydrate derivative isoman-
FIGURE 1. Proposed chiral template.
5
nide (Figure 2). Early experiments were then carried out
by using these two methacrylates (Scheme 2). As we
anticipated, when a solution of chiral methacrylate 3a
and 4′-methoxyacetophenone in THF at -78 °C was
2
treated with 2 equiv of SmI , the reaction proceeded
smoothly in the absence of any other proton source, and
the diastereomeric trans- and cis-R-methyl-γ-methyl-γ-
anisyl-γ-butyrolactone resulting from this asymmetric
coupling were isolated in the ratio of about 72/28 and 57%
overall yield (Table 1, entry 1). The trans isomer was
obtained with excellent enantioselectivity (95% ee). In
contrast, much lower ee (29%) was observed for the cis
product. Similar results were found when 3b was reacted
(entry 2). To our delight, these results are very close to
FIGURE 2. New chiral methacrylates for reductive coupling
reaction.
products with high enantiomeric purities without adding
other protonating agents. Besides, it would be very
interesting if the diastereoselectivity and enantioselec-
530 J. Org. Chem., Vol. 70, No. 2, 2005

Optically Active R,γ-Substituted γ-Butyrolactones
that obtained from our previous reaction system with use
TABLE 2. Reductive Coupling of Novel Chiral
Methacrylate 3 or 4 to Ketones for the Preparation of
r,γ-Substituted γ-Butyrolactones
entry Sa Pb trans/cisc trans ee (%)d cis ee (%)d yield (%)e
of chiral auxiliary 1 and bulky protonating agent
4
d
camphorsultam, indicating the success of our new
strategy.
Encouraged by these results, we further prepared
methacrylate substrates 3c and 4a-c with different R
substituents and NHR orientation. When 3c (R ) Cbz)
was employed under the same reaction conditions, unlike
the cases of 3a and 3b, we found that both trans and cis
products with very high enantioselectivities were ob-
tained; remarkably, a dramatic enhancement of enantio-
selectivity of the cis product (from 29% or 20% to >99%
ee) was observed (entry 3). Interestingly, when the NHR
group endo-oriented 4a was used, we found that the
major diastereomer obtained was cis (trans/cis ) 31/69),
inconsistent with the results obtained from 3a-c. More-
over, the configuation of the cis product resulting from
1
2
3
4
5
6
7
8
9
0
1
2
3a
3c
4d
3a
3c
4d
3a
3c
4d
3a
3c
4d
3a 10
3c 10
4d 10
3a 11
3c 11
3a 12
3c 12
6
6
6
7
7
7
8
8
8
9
9
9
>99/1
60/40
15/85
77/22
84/16
23/77
75/25
73/27
23/77
60/40
73/27
16/84
78/22
67/33
20/80
87/13
77/23
97 (+)
83 (+)
97 (+)
93 (-)
>99 (+)
96 (-)
96 (+)
96 (+)
90 (+)
93 (+)
96 (+)
94 (+)
78 (+)
80 (-)
70 (-)
78 (-)
92 (-)
75 (+)
86 (+)
15 (-)
21 (-)
23 (+)
23 (-)
54 (-)
22 (+)
21 (-)
32 (-)
31 (-)
6 (-)
49 (-)
24 (-)
90 (+)
84 (+)
73 (+)
54 (-)
9 (-)
74
89
78
77
75
76
53
59
60
53
60
58
61
42
56
90
78
41
40
1
1
1
13
14
1
1
1
1
5
6
7
8
4
a was also different from that resulting from 3a-c
(
entry 4). These results led to the suggestion that the
diastereoselectivity and enantioselectivity of the reaction
may be highly dependent on the orientation of the NHR
group. To confirm this consideration, the use of 4b and
19
a
Chiral methacrylate substrate. b Product. c Trans and cis iso-
mers were separated by column chromatography and confirmed
1
1
4
c was examined. Surprisingly, when tert-butyloxycar-
by H- H NOESY in light of their NOE effect; the ratio of trans/
d
bonyl-substituted 4b was used, the reaction provided the
cis was determined by HPLC or GC. The ee values were
determined by HPLC analysis on a Chiralcel column. The sign of
trans isomer as the major product with 99% ee and cis
product with 81% ee. The diastereomeric ratio (trans/
e
[
R]D is given in parentheses. Total isolated yield of trans and cis
6
products.
cis ) 67/33) was not identical with those resulting from
4a, but similar to those with 3b (entry 5). In comparison
with 4b, using benzyloxycarbonyl-substituted methacryl-
ate 4c also led to the major trans product (trans/cis )
2,4,6-triisopropylbenzenesulfonyl (Tris) substituent, rather
than a tosyl (Ts) substituent, under the same conditions
afforded much better results (entries 7 and 8). A sub-
stantial improvement in both the diastereoselectivity
(trans/cis ) 25:75) and enantioselectivity (90% ee for
trans, >99% ee for cis) were realized when 4d was
employed (entry 7). The increased selectivity when an
Mts or a Tris group was used instead of a Ts group is
undoubtedly due to the steric effect. These results also
clearly indicate that the structure of the R group in
molecule 4 has a major influence on the product distribu-
tion: cis diastereomers are favored when R is a sulfonyl
group. Thus, either enantiomer of the cis product with
extremely high ee (>99%) can be obtained by choosing
the appropriate chiral substrate (entry 3 and 7).
On the basis of the above results, methacrylates 3a,
3c, and 4d appear to be the most favorable substrates
and give excellent stereoselectivities. Reactions of these
substrates with a series of ketones were also successfully
tested to afford the corresponding γ-butyrolactones (6-
12) (Scheme 2). Table 2 summarizes the results obtained
from this procedure. As indicated in Table 2, the trans
products were attained with generally high ee values in
all cases; extremely high enantioselectivity (>99%) was
achieved with 2-acetonaphthone (entry 5). In contrast,
for the cis product, lower enantioselectivities were ob-
tained, the ee values were found to be largely dependent
on the structure of the ketones, and the highest (90% ee)
was obtained when 1-tetralone was used (entry 13). The
reaction of symmetric benzophenone was also examined,
and 86% ee was found when 3c was used (entry 19).
Notably, an extremely high degree of diastereofacial
selectivity (>99/1) was observed in the reaction of 3a with
acetophenone (entry 1).
75:25) but with lower enantioselectivity (71%). The minor
cis product was obtained with 79% ee and a same
configuration (entry 6). In all cases, the configurations
of the trans products were always the same whatever
methacrylate, 3 or 4, was used; however, the configura-
tions of the cis products were found opposite when 3 or
4
was used (entries 1-8). Thus, it can be concluded that
only the configuration of the cis product was largely
determined by the stereochemistry of the proton source
part. It is clear that the transition state model of the
reaction is not significantly changed if the NHR group
of the proton source is away from the reaction site,
namely exo oriented. Whatever the R group is, products
with the same configurations are obtained. However, in
the reactions of substrate 4, the NHR group is on the
same side of the reaction site, and the formation of the
transition state was strongly affected by the nature of
the R group due to either steric or chelating effects, thus
leading to different enantiomers. In the case of 4a, we
assumed that the chelation of the samarium ion and the
oxygen of the sulfonyl group in the substrate may be an
important factor for the observed diastereoselectivity. The
exact transition-state model and precise mechanistic
explanation still remain unclear at this moment.7
To investigate the steric effect of the sulfonyl group,
two other substrates, 4d and 4e, were prepared. Gratify-
ingly, the use of both 2-mesitylenesulfonyl (Mts) and
(
5) Isomannide and isosorbide are easily obtained by dehydration
of mannitol and sorbitol, respectively. (a) Wiggins, L. F. J. Chem. Soc.
945, 4. (b) Montgomery, R.; Wiggins, L. F. J. Chem. Soc. 1946, 390.
1
(
6) The recovery of the chiral auxiliary was examined in this case,
and 86% of the employed auxiliary was recovered by column chroma-
tography.
We previously demonstrated that the reaction of chiral
auxiliary 2-derived methacrylate with ketones in the
(7) Proposed working models are included in the Supporting Infor-
mation.
J. Org. Chem, Vol. 70, No. 2, 2005 531

Huang et al.
perimental results suggest that self-protonation should
be involved in the reaction process.10
Conclusions
In summary, we have shown a new entry to the
asymmetric synthesis of optically active R,γ-substituted
γ-butyrolactones using carbohydrate-derived amide as
both a chiral auxiliary and a proton source. Unlike the
previously reported examples,4 the chiral auxiliary
employed here also successfully functioned as a proto-
nating agent to afford the reaction with excellent stereo-
selectivity. This provides a new and interesting example
of dual stereoselective control in asymmetric synthesis.
In addition, the combination of a chiral auxiliary and a
proton source within one molecule has led to the develop-
ment of a new type of bifunctional chiral template for
asymmetric synthesis. It provides useful insight to the
design of multifunctional chiral templates. Detailed
mechanistic studies of the stereocontrol process in this
new reaction system as well as the extension of this
method are currently underway.
FIGURE 3. Chiral methacrylates derived from D-glucose.
b,d
TABLE 3. Reductive Coupling of Novel Chiral
Methacrylate 11 or 12 to Ketones for the Preparation of
γ-Butyrolactones
entry Sa Pb trans/cisc trans ee (%)d cis ee (%)d yield (%)e
1
2
3
4
5
6
7
8
11
11
12
12
12
12
12
12
5
7
5
5
5
7
8
9
60/40
50/50
50/50
45/55
30/70
18/72
15/85
10/90
21 (+)
25 (-)
99 (+)
64 (+)
98 (+)
77 (-)
72 (+)
90 (+)
61 (+)
17 (-)
99 (+)
56 (+)
99 (+)
80 (+)
88 (+)
82 (+)
74
32
94
52
73
70
94
68
f
f,g
a
Chiral methacrylate substrate. b Product. c Trans and cis iso-
mers were separated by column chromatography and confirmed
1
1
by H- H NOESY in light of their NOE effect; the ratio of trans/
cis was determined by HPLC or GC. d The ee values were
determined by HPLC analysis on a Chiralcel. The sign of [R]D is
Experimental Section
e
given in parentheses. Total isolated yield of trans and cis
products. In the presence of 1.0 equiv of BuOH. The reaction
system was diluted 5 times.
f
t
g
2
General Procedure of the SmI -Mediated Asymmetric
Reductive Coupling Reaction. To a dry Schlenk flask with
samarium metal powder (200 mg, 1.3 mmol) was added a
solution of diiodomethane (freshly distilled, 0.081 mL, 1.0
mmol) in THF (5 mL) at room temperature under nitrogen.
After being stirred for 1 h, the deep blue solution was cooled
to -78 °C, and a mixture of methacrylate (0.5 mmol) and
ketone (0.5 mmol) in THF (5 mL) was added. The resulting
mixture was stirred for 1 h at the same temperature and then
allowed to warm slowly; the reaction was subsequently
quenched at -10 °C with 5% aqueous HCl, extracted with
presence of a bulky protonating agent could give satisfac-
tory results in both diastereoselectivities (in favor of cis
isomers) and enantioselectivities (up to 99%).4e To extend
our study, we prepared two new methacrylates 11 and
1
2 that have similar skeletons to that of 2 and a free
8
amide proton (Figure 3). Studies were carried out with
different ketones and the results are summarized in
Table 3. As we can see, treatment of 11 or 12 with
ketones in THF resulted in successful conversion to the
desired γ-butyrolactones. In most cases, cis isomers were
indeed obtained as major products and relatively high
ee values were attained when 12 was used (entries 3 and
ethyl ether, washed with aqueous NaHCO
3
and brine, and
dried over anhydrous Na SO . The solution was concentrated
2
4
and the resulting residue was purified by flash column
chromatography on silica gel to afford the optically active
γ-butyrolactones.
6
-8). It is noteworthy that the reaction of 12 with 4′-
Acknowledgment. Financial support from the Na-
tional Natural Science Foundation of China (203900506,
methoxyacetophenone afforded a 50:50 diastereomeric
ratio and extremely high enantiomeric excess (99% ee)
for both isomers (entry 3). To gain some insight into the
protonation mechanism, we performed the same reaction
in the presence of tert-butyl alcohol. We wished to see if
the observed excellent stereoselection was due to the self-
protonation in the reaction process. We first run the
reaction under similar reaction conditions with additional
protonating agent tert-butyl alcohol; although the dia-
stereomeric ratio remained similar (45/55), the ee values
of both trans and cis products were found to decrease a
lot (64% for trans and 56% for cis) (entry 4). However,
when the above reaction was performed in a diluted
system, excellent ee values (98% for trans and 99% for
2
0402018), the Major State Basic Research Develop-
ment Program (G2000077506), and the Shanghai Mu-
nicipal Committee of Science and Technology is ac-
knowledged.
Supporting Information Available: Characterization
data of all compounds and supposed working models of the
reaction. This material is available free of charge via the
Internet at http://pubs.acs.org.
JO0485637
(
9) The difference between the trans/cis ratio in entry 3 and that in
cis) were observed again regardless of the presence of tert-
entry 5 might be explained by the conformation change of the reaction
transition state in the diluted system. The exact reason is not clear at
this time.
butyl alcohol (entry 5),9 indicating a intramolecular
protonation process. Thus, these two independent ex-
(10) Other experimental results that are also supportive: comparing
with the results in entry 5 of Table 1, no significant difference was
found (trans/cis ) 69/31, 99% ee for trans and 79% ee for cis) when
substrate 4b was reacted with 4′-methoxyacetophenone (5) in the
presence of 1.0 equiv of tert-butyl alcohol under diluted (5 times)
conditions.
(
8) Prepared from D-glucose, see: (a) Tian, H.; She, X.; Shi, Y. J.
Org. Chem. 2002, 67, 2435. (b) Shu, L.; Shen, Y.-M.; Burke, C.; Goeddel,
D.; Shi, Y. J. Org. Chem. 2003, 68, 4963.
532 J. Org. Chem., Vol. 70, No. 2, 2005