Pinacol rearrangement for constructing asymmetric centers adjacent to heterocycles

Article abstract of DOI:10.1016/S0040-4039(02)01626-X

Stereospecific pinacol-type 1,2-shift of electron-rich heterocycles was achieved under organoaluminum-promoted conditions to afford optically pure ketones or alcohols possessing these heterocycles. The method was applied to the asymmetric formal total synthesis of indolmycin.

Full text of DOI:10.1016/S0040-4039(02)01626-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 6937–6940
Pinacol rearrangement for constructing asymmetric centers
adjacent to heterocycles
Tomoichi Shinohara and Keisuke Suzuki*
Department of Chemistry, Tokyo Institute of Technology, and CREST, Japan Science and Technology Corporation (JST),
O-okayama, Meguro-ku, Tokyo 152-8551, Japan
Received 21 June 2002; revised 31 July 2002; accepted 2 August 2002
Abstract—Stereospecific pinacol-type 1,2-shift of electron-rich heterocycles was achieved under organoaluminum-promoted
conditions to afford optically pure ketones or alcohols possessing these heterocycles. The method was applied to the asymmetric
formal total synthesis of indolmycin. © 2002 Elsevier Science Ltd. All rights reserved.
Asymmetric center adjacent to heterocyclic structures is
a motif shared by various biologically active natural/
unnatural products1–4 (Fig. 1). In many cases, the lack
of an appropriate synthetic handle renders the con-
struction of such asymmetric centers a challenge.
pinacol-type rearrangement of heterocycles in the chiral
lactate-derived mesyloxy alcohols, which is nicely pro-
moted by organoaluminum Lewis acids.
(1)
Scheme 1 shows the preparation of starting materials
for the 1,2-shift of heterocycles. For instance, diol 5a,
having a 3-indolyl group, was prepared as a 7/3
diastereomeric mixture by successive introduction of a
2-phenylethyl group and a 3-indolyl group7a to the
chiral lactamide, (S)-2,8 followed by acidic removal of
the ethoxyethyl protecting group. Other diols 5b–5f
were also prepared by introducing the respective lithi-
ated heterocycle.7
Figure 1.
We have previously described the efficient and stereose-
lective construction of asymmetric centers by Lewis
acid-promoted pinacol-type rearrangement of mesylates
and epoxides.5 It was unclear, however, if similar condi-
tions could be developed for the stereo- and regioselec-
tive
rearrangement
of
electron-rich
aromatic
heterocycles for constructing such structures via the
1,2-shift of heterocycles (Eq. (1)), where the presence of
additional coordinating groups could potentially inter-
fere with the Lewis acid-mediated process. We now
report the successful development of conditions for the
Scheme 1.
* Corresponding author. E-mail: ksuzuki@chem.titech.ac.jp
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01626-X

6938
T. Shinohara, K. Suzuki / Tetrahedron Letters 43 (2002) 6937–6940
Scheme 2 exemplifies the pinacol-type shift of a 3-
indolyl group (5a“7a).6 Methanesulfonylation of 5a
with MsCl and Et3N (CH2Cl2, 0°C, 10 min) proceeded
regioselectively at the sec-hydroxy group to give mesyl-
oxy alcohol 6a, which, without isolation, was succes-
sively treated with Et3Al (3.0 equiv.) at −78°C for 10
min.9 The indolyl group underwent smooth 1,2-shift,
giving chiral ketone 7a in 90% yield. The stereospecific-
ity of the 1,2-shift was confirmed by the enantiomeric
purity of 7a (99% e.e.) by the chiral HPLC analysis10
(DAICEL CHIRALCEL OD-H, i-PrOH/hexane=1/9),
reflecting the e.e. of the starting lactate (99% e.e.).11
Furthermore, reduction of ketone 7a with LiBEt3H
(THF, −78°C, 10 min) proceeded in excellent selectivity
to give alcohol 8a6 as a sole product (Scheme 3). The
stereostructure of 8a was deduced from the Felkin–Anh
model12 applied to the reduction process of 7a with the
indolyl group serving as a large group (Fig. 2). This
assignment was verified by the (R) absolute configura-
tion of the alcohol center according to the Mosher–
Kusumi method.13 The (S) chirality at the adjacent
center resulted from the inversive 1,2-shift.
These conditions for stereoselective 1,2-shifts were
directly applicable to a variety of heterocyclic struc-
tures, and results are summarized in Table 1. Migration
of 2-thienyl, 1-methyl-2-indolyl, 3-benzofuranyl, and
2-benzothiophenyl groups proceeded smoothly, and
optically pure ketones 7b, 7c, 7e, and 7f were obtained
in high yields.6,14 The single exception in these attempts
was the sluggishness of the 1,2-shift of 2-benzofuranyl
group, resulting in a low yield of 7d with considerable
racemization, although the reason is not clear. The
inherent migratory aptitude of 2-benzofuranyl group is
not low, judging from the successful conversion of
10a“11a (vide infra), and work is in progress for
clearing this issue.
In all cases listed, subsequent reduction with LiBEt3H
(THF, −78°C) proceeded diastereoselectively to afford
alcohols 8b–f.6,15
Scheme 2.
The experimental procedure for the rearrangement of 5a
to 7a is as follows: to a solution of diol 5a (500 mg, 1.15
mmol) in CH2Cl2 (5 mL) was added Et3N (240 mL, 1.72
mmol), and MsCl (107 mL, 1.38 mmol) at 0°C. After
stirring for 10 min, the mixture was chilled to −78°C, to
which was added a hexane solution (0.92 mmol/mL) of
Et3Al (3.7 mL, 3.4 mmol). After 10 min, the mixture
was poured into aqueous KHSO4, and the products
were extracted with Et2O. The combined organic layer
was washed with water and brine, dried (MgSO4), and
concentrated. Purification with silica gel column chro-
matography (EtOAc/hexane=1/6) afforded ketone 7a
(430 mg, 90%).6
Table 1.
Scheme 3.
Figure 2.

T. Shinohara, K. Suzuki / Tetrahedron Letters 43 (2002) 6937–6940
6939
The reductive version of such 1,2-shift5c proved also
viable in stereospecific manner, as illustrated by the
reductive shift of 2-benzofuranyl group (Scheme 4).
Thus, mesyloxy ketone 10a was prepared by the reac-
tion of 2-benzofuranyllithium 4d with the lactamide
(S)-2 followed by deprotection to give a-keto alcohol
9a, which was converted into the corresponding mesyl-
ate. After simple extractive workup and drying, 10a was
dissolved in CH2Cl2, to which was added DIBAL (3
mol equiv.) and Et2AlCl (1.5 mol equiv.) at −78°C
followed by warming to 0°C in 10 min. By this proce-
dure, the carbonyl reduction in 10a occurred to gener-
ate the aluminum alkoxide I in situ, which underwent
1,2-shift of the 2-benzofuranyl group by activation with
Et2AlCl16 to give the aldehyde II, which was reduced by
the excess DIBAL to give 11a6,17 in 99% e.e.
Scheme 5 shows the application of this process to the
formal total synthesis of indolmycin (1). Treatment of
the enantiomeric lactamide (R)-2 with 3-lithioindole 4a
(vide supra) cleanly gave the corresponding ketone, and
removal of the ethoxyethyl group afforded hydroxy
ketone 12. Mesylation of 12 smoothly gave mesyloxy
ketone 13, which was successively treated with DIBAL
(3 equiv.) and Et2AlCl (1.5 equiv.) at −78°C. The
1,2-reduction occurred immediately, and was followed
by the 1,2-shift of the indolyl group and reduction of
the resulting aldehyde to give the chiral alcohol 1418 in
87% yield. The enantiomeric excess of 14 was assessed
by HPLC analysis (DAICEL CHIRALCEL OD-H,
i-PrOH/hexane=1/9), showing the stereochemical
integrity of the whole sequence.
The 1,2-shift proved to be stereospecific with inversion
of the chiral center, as shown by the absolute configura-
tion of the known ester 18, the key intermediate of
Takeda’s synthesis of indolmycin.1a Thus, mesylation of
14 to give mesylate 15, which was then treated with
KCN to afford cyanide 16. Basic hydrolysis of 16
followed by treatment of the resulting acid 17 with
diazomethane gave methyl ester 18 {[h]2D2 +11.1 (c 1.17,
C6H6), lit.1a [h]D22 +10.9 (c 2.12, C6H6)}. The enan-
tiomeric excess of 18 (99% e.e.) was further verified by
HPLC analysis19 (DAICEL CHIRALCEL OJ-H,
EtOH/hexane=1/4).
Scheme 4.
1-Methyl-2-indolyl, 2-benzothiophenyl, 5-methoxy-2-
benzofuranyl,7b and 3-benzofuranyl groups also shifted
stereospecifically to give chiral alcohols 11b, 11c, 11d
and 11e, respectively6,17 (Table 2).
Table 2.
Scheme 5.

6940
T. Shinohara, K. Suzuki / Tetrahedron Letters 43 (2002) 6937–6940
In conclusion, the stereospecific pinacol-type shift of
some p-electron-rich heterocycles could be effected
under organoaluminum-promoted conditions, and
should find various utility in the selective construction
of asymmetric centers adjacent to heterocycles.
9. This one-pot procedure is particularly suitable for the
case when the intermediary mesylate was unstable.
Details for the comparison of this protocol with the
two-pot protocol5a,b we previously employed will be dis-
closed elsewhere.
10. An authentic sample of ( )-7a was prepared from ( )-
ethyl lactate along the same lines.
References
11. The e.e. of the ethyl (S)-lactate was determined to be 99%
1
by 400 MHz H NMR analysis of the (−)-MTPA ester.12a
1. Asymmetric synthesis of indolmycin: (a) Takeda, T.;
Mukaiyama, T. Chem. Lett. 1980, 163; (b) Akita, H.;
Kawaguchi, T.; Enoki, Y.; Oishi, T. Chem. Pharm. Bull.
1990, 38, 323; (c) Bando, T.; Shishido, K. Heterocycles
1997, 46, 111; (d) Hasuoka, A.; Nakayama, Y.; Adachi,
M.; Kamiguchi, H.; Kamiyama, K. Chem. Pharm. Bull.
2001, 49, 1604. Synthesis of ( )-indolmycin: (e) Dhue,
Y.-K. Tetrahedron Lett. 1996, 37, 6447.
2. (a) Inoue, M.; Frontier, A.; Danishefsky, S. J. Angew.
Chem., Int. Ed. 2000, 39, 761; (b) Appendino, G.; Casir-
aghi, G.; Zanardi, F. Chemtracts 2000, 13, 741; (c) Inoue,
M.; Carson, M. W.; Frontier, A. J.; Danishefsky, S. J. J.
Am. Chem. Soc. 1878, 123, 2001.
3. (a) Kozikowski, A. P.; Greco, M. N. J. Am. Chem. Soc.
1980, 102, 1165; (b) Guo, X.; Zhang, Z. Yaoxue Xuebao
1987, 22, 671; (c) Matsumoto, M.; Watanabe, N. Hetero-
cycles 1987, 26, 1743; (d) Dickens, M. J.; Mowlem, T. J.;
Widdowson, D. A.; Slawin, A. M. Z.; Williams, D. J. J.
Chem. Soc., Perkin Trans. 1 1992, 323; (e) Kato, K.; Ono,
M.; Akita, H. Tetrahedron Lett. 1805, 38, 1997; (f) Kato,
K.; Ono, M.; Akita, H. Tetrahedron 2001, 57, 10055.
4. Walsh, T. F.; Toupence, R. B.; Ujjainwalla, F.; Young, J.
R.; Goulet, M. T. Tetrahedron 2001, 57, 5233.
5. (a) Suzuki, K.; Katayama, E.; Tsuchihashi, G. Tetra-
hedron Lett. 1983, 24, 4997; (b) Suzuki, K.; Katayama,
E.; Tsuchihashi, G. Tetrahedron Lett. 1817, 25, 1984; (c)
Suzuki, K.; Katayama, E.; Matsumoto, T.; Tsuchihashi,
G. Tetrahedron Lett. 1984, 25, 3715; (d) Tsuchihashi, G.;
Tomooka, K.; Suzuki, K. Tetrahedron Lett. 1984, 25,
4253; (e) Suzuki, K.; Ohkuma, T.; Tsuchihashi, G. Tetra-
hedron Lett. 1985, 26, 861; (f) Suzuki, K.; Tomooka, K.;
Matsumoto, T.; Katayama, E.; Tsuchihashi, G. Tetra-
hedron Lett. 1985, 25, 3711; (g) Suzuki, K.; Tomooka, K.;
Shimazaki, M.; Tsuchihashi, G. Tetrahedron Lett. 1985,
26, 4781; (h) Suzuki, K.; Ohkuma, T.; Miyazawa, M.;
Tsuchihashi, G. Tetrahedron Lett. 1986, 27, 373; (i)
Suzuki, K.; Tomooka, K.; Katayama, E.; Matsumoto,
T.; Tsuchihashi, G. J. Am. Chem. Soc. 1986, 108, 5221; (j)
Suzuki, K. J. Synth. Org. Chem. Jpn. 1988, 46, 365; (k)
Nagasawa, T.; Taya, K.; Kitamura, M.; Suzuki, K. J.
Am. Chem. Soc. 1996, 118, 8949.
12. (a) Che´rest, M.; Felkin, H.; Prudent, N. Tetrahedron
Lett. 1968, 2201; (b) Anh, N. T.; Eisenstein, O. Nouv. J.
Chim. 1977, 1, 61.
13. The (R) absolute configuration of the alcohol center was
deduced by the Mosher–Kusumi method based on the
chemical shifts of the corresponding (+)- and (–)-MTPA
esters. (a) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc.
1973, 95, 512; (b) Kusumi, T.; Ohtani, I.; Inoue, Y.;
Kakisawa, H. Tetrahedron Lett. 1988, 29, 4731.
14. The enantiomeric excess was determined as follows. For
7b: ketone 7b was reduced with DIBAL to give the
diastereomeric alcohols (ca. 1/1), which were converted to
the corresponding (+)-MTPA esters. For comparison, the
alcohols derived from ( )-ethyl lactate were converted to
the (+)-MTPA esters. For 7c–f: by the chiral HPLC
analysis (DAICEL CHIRALCEL OD-H, i-PrOH/hex-
ane=1/9) of the chiral ketones and ( )-ketones.
15. The diastereomers of 8b–8f were inseparable on TLC,
1
respectively. Each of the ratio was assessed by H NMR,
and the configurations of alcoholic center are based on
the analogy with 8a.
16. In this reductive version of rearrangement, the second
organoaluminum (Et2AlCl) is crucial for the smooth 1,2-
shift to occur, as noted previously.5c In its absence, the
rearrangement was much slower, and higher temperature
was necessary (see also Refs. 5d and 5g). In the case
where Et3Al was used as the second Lewis acid instead,
the yield was somewhat lower, and some byproducts were
produced, including the one from ethylation of the alde-
hyde intermediate.
6. All new compounds were fully characterized by spectro-
scopic means and combustion analysis.
17. The enantiomeric excess was determined by the chiral
HPLC analysis (DAICEL CHIRALCEL OD-H, i-PrOH/
hexane=1/9).
7. (a) 3-Indolyllithium and 3-benzofuranyllithium were gen-
erated by halogen–lithium exchange of the corresponding
bromide (n-BuLi, THF, −78°C, 1 min); (b) 2-Thienyl-
18. Although compound 14 was previously reported by
Bando et al. (Ref. 1c) in their formal total synthesis of
indolmycin via enzymatic transformation, the specific
rotation was not reported. Thus the [h]D value of the
corresponding methyl ester 18 was compared with that of
18 reported by Takeda (Ref. 1a).
lithium,
2-benzothiophenyllithium,
2-benzofuranyl-
lithium, and 5-methoxy-2-benzofuranyllithium were
generated by hydrogen–lithium exchange (n-BuLi, THF,
−78“25°C, 30 min); (c) 1-Methyl-2-indolyllithium was
generated by hydrogen–lithium exchange in the presence
of TMEDA (n-BuLi, THF, −78“25°C, 30 min).
8. Honda, Y.; Ori, A.; Tsuchihashi, G. Bull. Chem. Soc.
Jpn. 1987, 60, 1027.
19. An authentic sample of ( )-18 was prepared from ( )-
ethyl lactate along the same lines.