Highly enantioselective [2,3]-Wittig rearrangement of functionalized allyl benzyl ethers: a novel approach to lignan synthesis

Article abstract of DOI:10.1016/j.tetasy.2008.04.025

Highly enantioselective [2,3]-Wittig rearrangement of functionalized allyl benzyl ethers was accomplished using a chiral di-tert-butyl bis(oxazoline) ligand. The reaction proceeded with excellent diastereo- and enantioselectivity when no methoxy substituent was present at the ortho-position on the benzyl group. On the other hand, the enantioselectivity was drastically decreased in the presence of an ortho-methoxy group.

Full text of DOI:10.1016/j.tetasy.2008.04.025

Tetrahedron: Asymmetry 19 (2008) 1167–1170
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: http://www.elsevier.com/locate/tetasy
Highly enantioselective [2,3]-Wittig rearrangement of functionalized allyl
benzyl ethers: a novel approach to lignan synthesis
*
Yoshimi Hirokawa, Maria Kitamura, Naoyoshi Maezaki
Faculty of Pharmacy, Osaka Ohtani University, 3-11-1 Nishikiori-Kita, Tondabayashi, Osaka 584-8540, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Highly enantioselective [2,3]-Wittig rearrangement of functionalized allyl benzyl ethers was accom-
plished using a chiral di-tert-butyl bis(oxazoline) ligand. The reaction proceeded with excellent diaste-
reo- and enantioselectivity when no methoxy substituent was present at the ortho-position on the
benzyl group. On the other hand, the enantioselectivity was drastically decreased in the presence of an
ortho-methoxy group.
Received 3 March 2008
Accepted 9 April 2008
Available online 26 May 2008
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
employing a modified Box-i-Pr bearing gem-diphenyl group at
the C5 position of two oxazoline rings.6 To obtain an enantioselec-
The [2,3]-Wittig rearrangement is a strategy used to provide
building blocks for the synthesis of natural products, since two ste-
reocenters are constructed simultaneously with high diastereo-
selectivity.1 Recently, an asymmetric [2,3]-Wittig rearrangement
with an external chiral ligand has been investigated (Fig. 1).
tivity more than 85% ee, conversion of the aryl group into tricar-
bonylchromium complex was required as described in Gibson’s
report.7
Over the course of our synthetic studies on lignans,8 we inves-
tigated the asymmetric [2,3]-Wittig rearrangement of benzylic
ethers 1 as a key reaction (Scheme 1).
Lignans are phenylpropanoid dimers and possess a variety of
frameworks and substitution patterns on aryl groups. In addition,
they have interesting biological activities such as cytotoxic, antivi-
ral, antifungal, immunosuppressive, antiasthmatic, and antioxidant
activities.9 We envisioned a chiral benzylic alcohol derivative 2 as a
key intermediate for various lignans, which can be synthesized by
asymmetric [2,3]-Wittig rearrangement of oxygenated benzyl
ether 1, as shown in Scheme 1. We were interested in Honda’s
work regarding the synthesis of kallolide A and pinnatin A, in
which an asymmetric [2,3]-Wittig reaction of cyclic furfuryl ethers
R' R'
O
O
H
H
N
N
N
N
R
R
(S,S)-Box-i-Pr (R = i-Pr)
(S,S)-Box-t-Bu (R = t-Bu)
(R' = Et or Me)
(−)-Sparteine
Figure 1. External chiral ligands for [2,3]-Wittig rearrangement.
Kang pioneered the asymmetric [2,3]-Wittig reaction using an
external chiral ligand for alkenyl propargyl ethers2 and the enantio-
selectivity was improved by Nakai up to 89% ee by applying a chi-
ral diisopropyl bis(oxazoline) (Box-i-Pr).3 On the other hand, the
enantioselectivity remained moderate to low for benzyl ether-type
substrates. Manabe4 first reported an asymmetric [2,3]-Wittig
rearrangement of alkenyl benzyl ethers using a chiral amino ether
ligand, wherein (Z)-crotyl benzyl ether predominantly afforded the
syn-adduct with 64% ee. For (Z)-crotyl benzyl ether, Nakai’s ligand
was not effective (40% ee).3 In Kimachi’s study using (ꢀ)-sparteine
for ortho-functionalized benzyl ethers, the enantioselectivity did
not exceed 60% ee.5 Breeden improved the selectivity for benzyl
prenyl ether up to 66% ee with inversion of enantioselectivity by
Asymmetric
[2,3]-Wittig
Me
*
TIPSO
rearrangement
*
Me
TIPSO
HO
OR
OR
O
OR
OR
2
1
Me
Me
Me
*
*
*
OR
OR
O
*
O
OR
OR
RO
RO
O
Ar
THF lignans
Eupomatilones
* Corresponding author. Tel./fax: +81 721 24 9541.
E-mail address: maezan@osaka-ohtani.ac.jp (N. Maezaki).
Scheme 1.
0957-4166/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2008.04.025

1168
Y. Hirokawa et al. / Tetrahedron: Asymmetry 19 (2008) 1167–1170
using chiral di-tert-butyl bis(oxazoline) (Box-t-Bu) ligand pro-
ceeded with a very high selectivity (93% ee), albeit in a low yield
(19%).10 In spite of the prominent selectivity, application of the
Box-t-Bu ligand to asymmetric [2,3]-Wittig rearrangement was
limited to this cyclic substrate. We started a systematic study of
asymmetric [2,3]-Wittig rearrangements of various oxygenated
benzyl ethers using the Box-t-Bu ligand.
Herein, we report a highly diastereo- and enantioselective
asymmetric [2,3]-Wittig rearrangement of oxygenated benzylic
ethers using an external chiral ligand. We also found that an
ortho-methoxy substituent affected the regiochemistry of deproto-
nation and enantioselectivity.
The regioselectivity of the deprotonation was improved when
the 3-methoxybenzyl derivative 5c and the 2-methoxybenzyl
derivative 5d were used (entries 3 and 4). No [1,2]-Wittig rear-
rangement product was observed, and the yields of 6c and 6d
increased to 64% and 68%, respectively. We also examined the
reaction of dimethoxybenzyl derivatives 5e–h with diverse substi-
tute patterns (entries 5–8) and trimethoxybenzyl ether 5i (entry 9),
and some of these aryl groups were found in natural lignans. In
these compounds, only 3,4-dimethoxybenzyl derivative 5e affor-
ded [1,2]-Wittig rearrangement product 7e, but the ratio of the
by-product was smaller than that with the 4-methoxybenzyl
derivative 5b.
The decrease in the regioselectivity of the deprotonation in the
4-methoxybenzyl ether 5b probably originated from the weak
acidity at the benzylic position by the electron-donating resonance
effect of a para-methoxy group. Therefore, the acidity was recov-
ered in 3-methoxybenzyl ether 5c. In contrast, it was found that
the 2-methoxy substituent in 5d, with the positive resonance
effect, promotes the deprotonation at the benzylic position, there-
by improving the regioselectivity and the yield. The high regio-
selectivity in 5d would be due to the lateral metallation effect by
coordinating to the countercation of the base.
The relative stereochemistry of the [2,3]-Wittig rearrangement
product 6d was confirmed by conversion into known c-lactones
syn- and anti-8, as shown in Scheme 3.12 The 1H and 13C NMR spec-
tral data were identical to the published data. The relative stereo-
chemistry of other [2,3]-Wittig rearrangement products was
confirmed by comparison of the 1H NMR data with the data for
6d and 1-epi-6d (anti-isomer of 6d), which was prepared by oxida-
tion of 6d followed by diastereoselective reduction with L-Select-
rideÒ (Scheme 3).
Coupling constants between the C1- and C2-methine protons
were J = 4.0 Hz in 6d (syn-isomer) and J = 8.3 Hz in 1-epi-6d (anti-
isomer), respectively. Those of products 6a–i ranged from 3.4 to
4.4 Hz, thereby indicating their syn-stereochemistry. A similar
empirical rule was also reported by Nakai.13
The asymmetric version of [2,3]-Wittig rearrangement was con-
ducted with t-BuLi in hexane in the presence of (S,S)-2,20-(dimeth-
ylmethylene)bis(4-tert-butyl-2-oxazoline) 914 as an external chiral
ligand. The results are summarized in Scheme 4 and Table 2.
Unsubstituted benzyl ether 5a afforded the syn-isomer 6a exclu-
sively in 65% yield with unprecedented high diastereo- and
enantioselectivity (>98% de and 98% ee) (entry 1). High selectivity
was also observed in 4-methoxybenzyl ether 5b and 3-metho-
xybenzyl ether 5c (entries 2 and 3).
2. Results and discussion
The known unsaturated ester 311 was converted into triisopro-
pylsilyl (TIPS) ether 4 in 84% yield in two steps. Substrates 5a–i
were readily synthesized by the nucleophilic substitution of 4 with
alkoxides from various benzylic alcohols (Scheme 2 and Table 1).
Using substrates 5a–i, we investigated the [2,3]-Wittig rearrange-
ment with t-BuLi in tetrahydrofuran (THF). The results are shown
in Scheme 2 and Table 1. The reaction of unsubstituted benzyl
ether 5a afforded only a product arising from [2,3]-Wittig rear-
rangement in 86% yield (entry 1). The diastereoselectivity was very
high, and the syn product was produced exclusively. In the case of
the 4-methoxybenzyl derivative 5b, the reaction afforded the
desirable [2,3]-Wittig product 6b in 44% yield accompanied by
17% of the [1,2]-Wittig product 7b (entry 2). The [2,3]-Wittig rear-
rangement product 7b resulted from the deprotonation of the
allylic proton rather than the benzylic proton. The structure of
the [1,2]-Wittig rearrangement product 7b was confirmed by a
two-dimensional 1H/13C-heteronuclear multiple bond correlation
(HMBC) experiment.
MeO2C
TIPSO
c
a, b
Br
Me
Br
Me
(Table 1)
3
4
TIPSO
TIPSO
Me
d
Me
Ar
+
Me
HO
TIPSO
Ar
(Table 1)
O
HO
6a−i
Ar
7a−i
5a−i
Unexpectedly, the enantioselectivity was greatly decreased in
2-methoxybenzyl ether 5d (entry 4). A similar tendency was
observed in dimethoxybenzyl ethers 5e–h and trimethoxybenzyl
ether 5i. In contrast, while 3,4- and 3,5-dimethoxybenzyl ethers
5e and 5f showed comparatively high enantioselectivity (entries
Scheme 2. Reagents and conditions: (a) DIBAL, CH2Cl2, ꢀ20 °C; (b) TIPSOTf, 2,6-
lutidine, CH2Cl2, 0 °C to rt (84% in two steps); (c) ArCH2OH, t-BuOK, THF, rt (Table
1); and (d) t-BuLi, THF, ꢀ78 °C (Table 1).
Table 1
Synthesis of benzylic ethers 5a–i and [2,3]-Wittig rearrangement
Entry
Ar
Benzylic ether
Yielda,b (%)
[2,3]-Rearranged product
Yieldb,c (%)
[1,2]-Rearranged product
Yieldb,c (%)
1
2
3
4
5
6
7
8
9
Phenyl
5a
5b
5c
5d
5e
5f
5g
5h
5i
75
63
80
85
85
74
85
85
77
6a
6b
6c
6d
6e
6f
6g
6h
6i
86
44
64
68
43
93
72
58
79
7a
7b
7c
7d
7e
7f
7g
7h
7i
0
17
0
0
6
0
0
0
0
4-Methoxyphenyl
3-Methoxyphenyl
2-Methoxyphenyl
3,4-Dimethoxyphenyl
3,5-Dimethoxyphenyl
2,3-Dimethoxyphenyl
2,5-Dimethoxyphenyl
3,4,5-Trimethoxyphenyl
a
Reactions were carried out using benzylic alcohol (2 equiv), t-BuOK (2.2 equiv) in THF at rt for 16 h.
Isolated yield.
Reactions were carried out using benzylic ether (1 equiv), t-BuLi (10 equiv) in THF at ꢀ78 °C for 2 h.
b
c

Y. Hirokawa et al. / Tetrahedron: Asymmetry 19 (2008) 1167–1170
1169
J = 4.0 Hz
Me
Me
Me
Me
Me
OMe
OMe
a−c
a, b
a, b
OMe
OMe
OMe
OMe
O
TIPSO
(1R,2S)-6i
TIPSO
TIPSO
O
O
O
HO
HO
O
6d (syn)
syn-8
Me
10
OMe
OMe
c, d
Me
J = 8.3 Hz
Scheme 5. Reagents and conditions: (a) TBAF, THF, rt, (b) cat. TEMPO, cat. n-Bu4NI,
NCS, CH2Cl2–H2O, pH 8.6, rt; (c) H2, cat. Rh(PPh3)3Cl, toluene, rt (40% in three steps).
OMe
OMe
HO
O
TIPSO
Me
anti-8
1-epi-6d (anti)
Et
Et
a
Me
Scheme 3. Reagents and conditions: (a) TBAF, THF, rt; (b) cat. TEMPO, cat. n-Bu4NI,
NCS, CH2Cl2–H2O, pH 8.6, rt (50% for syn-8 and 85% for anti-8 each in two steps); (c)
cat. TPAP, NMO, CH2Cl2–CH3CN (9:1), 4 Å MS, rt (98%); and (d) L-SelectrideÒ, THF,
ꢀ78 °C then 30% H2O2, acetone, rt (91%).
TIPSO
O
HO
(1R,2S)-6j
(>99% ee)
5j
Scheme 6. Reagents and conditions: (a) t-BuLi, (S,S)-Box-t-Bu 9, hexane, ꢀ78 °C
Me Me
(69%).
Me
S
O
O
R
a
diate of 3-epi-eupomatilone 6,16 we have achieved a formal syn-
thesis of 3-epi-eupomatilone 6.
5a−i
TIPSO
Ar
N
N
HO
(Table 2)
t-Bu
t-Bu
(1R,2S)-6a−i
9
The effect of the 2-methoxy group is unclear, but the oxygen
appears to play an important role in the effect. In order to confirm
the role of the 2-methoxy group, we synthesized the 2-ethylbenzyl
equivalent 5j in 63% yield by the general procedure shown in
Scheme 2. The asymmetric [2,3]-Wittig reaction of 5j afforded 6j
in 69% yield and with excellent enantioselectivity (>99% ee)
(Scheme 6) in contrast to that of 2-methoxybenzyl ether 5d (8%
ee).à Presumably, coordination of the methoxy group to the lithium
cation rather than the steric effect would retard the high selectivity.
Further studies including the reaction mechanism are in progress.
Scheme 4. Reagents and conditions: (a) t-BuLi, (S,S)-Box-t-Bu 9, hexane, ꢀ78 °C
(Table 2).
Table 2
Asymmetric [2,3]-Wittig rearrangement of benzylic ethers 5a–ia
Entry
Ar
Product
Yieldb (%)
eec (%)
1
2d
3
4
5
6
7
8
9
Phenyl
6a
6b
6c
6d
6e
6f
6g
6h
6i
65
34
75
73
32
72
58
58
76
98
98
95
8
87
99
17
22
85
4-Methoxyphenyl
3-Methoxyphenyl
2-Methoxyphenyl
3,4-Dimethoxyphenyl
3,5-Dimethoxyphenyl
2,3-Dimethoxyphenyl
2,5-Dimethoxyphenyl
3,4,5-Trimethoxyphenyl
3. Conclusion
In conclusion, we have investigated the [2,3]-Wittig rearrange-
ment of various benzyl ethers. As a result, we have found that (S,S)-
Box-t-Bu 9 is an effective ligand for oxygenated benzylic ethers.
The substitution of the methoxy group affected the chemical yield
and enantioselectivity, depending on its position. The products will
become versatile intermediates for the synthesis of lignans.
a
Reactions were carried out using benzylic ether (1 equiv), t-BuLi (10 equiv), and
(S,S)-Box-t-Bu 9 (5 equiv) in hexane at ꢀ78 °C for 2 h.
b
Isolated yield.
c
Determined by chiral HPLC analysis.
A trace of [1,2]-Wittig rearrangement product was produced as a by-product.
d
Acknowledgements
We acknowledge the financial support of a Grant-in-Aid for Sci-
entific Research (C) from the Japan Society for the Promotion of Sci-
ence (No. 18590097) and Osaka Ohtani University Research Fund
(Pharmaceutical Sciences).
5 and 6), 2,3- and 2,5-dimethoxybenzyl ethers 5g and 5h with an
ortho-methoxy substituent afforded products with low enantio-
meric excess (entries 7 and 8). The 3,4,5-trimethoxybenzyl ether
5i, bearing no methoxy group at the ortho-position, afforded a
product with high enantiomeric excess (entry 9).
References
The absolute configuration of the secondary alcohols was deter-
mined by a modified Mosher method,15 after conversion of alcohols
6b–d into the corresponding (+)- and (ꢀ)-a-methoxy-a-(trifluoro-
methyl)phenylacetic acid (MTPA) esters. Thus, the stereochemistry
of the major products was determined as (1R,2S). The absolute con-
figuration of the other products was assumed by analogy with
them. The assignment was also confirmed by conversion of 6i
1. (a) Tomooka, K. Chemistry of Organolithium Compounds 2004, 2, 749–828; (b)
Hiersemann, M.; Abraham, L.; Pollex, A. Synlett 2003, 1088–1095; (c)
McGowan, G. Aust. J. Chem. 2002, 55, 799; (d) Nakai, T.; Tomooka, K. Pure
Appl. Chem. 1997, 69, 595–600; (e) Nakai, T.; Mikami, K. Org. React. 1994, 46,
105–209; (f) Marshall, J. A. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon: New York, 1991; Vol. 3, pp 975–1014.
2. Kang, J.; Cho, W. O.; Cho, H. G.; Oh, H. J. Bull. Korean Chem. Soc. 1994, 15, 732–
739.
3. (a) Tomooka, K.; Komine, N.; Nakai, T. Tetrahedron Lett. 1998, 39, 5513–5516;
(b) Tomooka, K.; Komine, N.; Nakai, T. Chirality 2000, 12, 505–509.
4. Manabe, S. Chem. Commun. 1997, 737–738.
5. (a) Kawasaki, T.; Kimachi, T. Tetrahedron 1999, 55, 6847–6862; (b) Kawasaki, T.;
Kimachi, T. Synlett 1998, 1429–1431.
21
(85% ee) into known c-lactone 10 {½aꢁD ¼ þ67:3 (c 0.34, CHCl3)}
23
and comparison of the specific rotation {lit.16 ½aꢁD ¼ þ57:1 (c 0.2,
CHCl3)} (Scheme 5). Since the c-lactone 10 is a synthetic interme-
6. Barrett, I. M.; Breeden, S. W. Tetrahedron: Asymmetry 2004, 15, 3015–3017.
7. Gibson, S. E.; Ham, P.; Jefferson, G. R. Chem. Commun. 1998, 123–124.
The reaction of 5b using (ꢀ)-sparteine was sluggish and the expected 6b was not
obtained. Instead, SN20 reaction of t-BuLi to 5b and elimination of triisopropylsilan-
olate occurred with 53% yield along with 41% recovery of 5b.
à
The absolute configuration of 6j was determined by the modified Mosher method.

1170
Y. Hirokawa et al. / Tetrahedron: Asymmetry 19 (2008) 1167–1170
8. Kitamura, M.; Hayashi, H.; Yano, M.; Tanaka, T.; Maezaki, N. Heterocycles 2007,
71, 2669–2680.
13. Mikami, K.; Kimura, Y.; Kishi, N.; Nakai, T. J. Org. Chem. 1983, 48, 279–281.
14. (a) Ishihara, K.; Sakakura, A. Jpn. Kokai Tokkyo Koho JP2005320304, 2005.; (b)
Evans, D. A.; Woerpel, K. A.; Hinman, M. M.; Faul, M. M. J. Am. Chem. Soc. 1991,
113, 726–728.
15. Ohtani, I.; Kusumi, K.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113,
4092–4096.
9. For reviews of lignans, see: (a) Davin, L. B.; Lweis, N. G. Phytochem. Rev. 2003, 2,
257–288; (b) Umezawa, T. Phytochem. Rev. 2003, 2, 371–390; (c) Ward, R. S.
Nat. Prod. Rep. 1999, 16, 75–96; (d) Ward, R. S. Nat. Prod. Rep. 1997, 14, 43–74.
10. Tsubuki, M.; Takahashi, K.; Honda, T. J. Org. Chem. 2003, 68, 10183–10186.
11. (a) Sá, M. M.; Ramos, M. D.; Fernandes, L. Tetrahedron 2006, 62, 11652–11656;
(b) Fernandes, L.; Bortoluzzi, A. J.; Sá, M. M. Tetrahedron 2004, 60, 9983–9989.
12. Csuk, R.; Schröder, C.; Hutter, S.; Mohr, K. Tetrahedron: Asymmetry 1997, 8,
1411–1429.
16. Johnson, J. B.; Bercot, E. A.; Williams, C. M.; Rovis, T. Angew. Chem., Int. Ed. 2007,
46, 4514–4518.