methylpyrrole-3-carboxamide could be lithiated at C-2 with
sec-BuLi-TMEDA.3 On the other hand, Meijer and co-
workers demonstrated that the lithiation of N-Boc-3-hexyl-
pyrrole proceeded at C-5 selectively on treatment with
lithium 2,2,6,6-tetramethylpiperidide (LTMP).4 Demont and
co-workers indicated that the regioselectivity of the lithiation
of 3-benzenesulfonylpyrroles depended on the nature of the
base (LTMP, n- or sec-BuLi) and the N-protecting group
(Boc, SO2Ph, SEM).5 Robertson et al. showed that 3-chloro-
N-(p-toluenesulfonyl)pyrrole was lithiated at C-2 with n-
BuLi.6 Recently, we have reported that N-benzenesulfonyl-
3-bromopyrrole (1) could also be lithiated at C-2 with lithium
diisopropylamide (LDA), and this key reaction has been
employed in the synthesis of rationally designed analogues
of the antitumor marine alkaloid lamellarin D.7 In this letter,
we describe detailed investigations on the directed lithiation
of 1.
preferential deuteration at C-5 rather than C-2 (entry 3). More
bulky LTMP might deprotonate at the less congested C-5
preferentially. Lithium isopropylcyclohexylamide (LICA)
displayed a selectivity similar to that of LDA (entry 4).
Lithium hexamethyldisilazide (LHMDS) was totally inef-
fective for the lithiation of 1 apparently due to its lower
basicity9 (entry 5). Utilization of diethyl ether or toluene
instead of THF increased C-5 selectivity (entries 6-9). These
results might be accounted for by higher aggregation of the
amide bases in the less polar solvents.10 The bulky aggregates
might deprotonate at the less congested C-5 preferentially.
Although deuterium incorporation was quite modest,11 selec-
tive C-5 lithiation was achieved using LTMP as a base in
toluene (entry 9). Unfortunately, however, use of more base
(2.4 equiv) or a longer reaction time (12 h) resulted in the
loss of high C-5 selectivity (entries 10 and 11). Further
attempts to effect selective C-5 lithiation were not performed
because we discovered a practical Br-Li exchange route to
generate the C-5 lithio species (vide infra).
First, we investigated regioselectivity via deuteration
experiments (Table 1). Treatment of 1 with 1.2 equiv of LDA
Having established the conditions for efficient and selective
C-2 lithiation, we next examined the functionalization of the
lithiated 1. Thus, 1 was treated with 1.2 equiv of LDA in
THF at -78 °C for 1 h, and the resulting C-2 lithio species
3 was reacted with 1.8 equiv of an appropriate electrophile
under the conditions A (-78 °C, 1 h) or B (-78 °C, 1 h
f 0 °C, 3 h). After the usual workup, products were
isolated by column chromatography. As shown in Table
2, in the reactions with methyl chloroformate, chlorotri-
methylsilane, diphenyl disulfide, 1,2-dibromo-1,1,2,2-
tetrafluoroethane, iodine, p-methoxybenzaldehyde, and
tert-butyl isocyanate, the corresponding 2-functionalized
pyrroles 4a-g were obtained exclusively in excellent
yields (entries 1-8). In the reactions with p-methoxy-
benzoyl chloride, however, the unexpected 5-acylated
Table 1. Survey of Reaction Conditions for Regioselective
Lithiation of 1
deuterium incorporationc
2 yield C-2 C-5 total
ratio
(C-2:C-5)
entry
base
LDA
LDA
LTMP
LICA
solvent
(%)b
(%) (%)
(%)
1
THF
THF
THF
THF
91
92
95
95
94
97
99
96
98
98
94
70
70
27
65
0
17
32
15
∼0
3
2
3
58
4
72
73
85
69
0
48
57
54
11
21
51
97:3
96:4
32:68
94:6
2d
3
4
1
pyrrole 5h was detected as a minor product by H NMR
5
6
7
8
LHMDS THF
0
-
LDA
LDA
LTMP
LTMP
LTMP
LTMP
Et2O
toluene
Et2O
31
25
39
11
18
36
35:65
56:44
28:72
1:>99
14:86
29:71
analysis (entries 9 and 10). More surprisingly, when dim-
ethylcarbamoyl chloride was used as an electrophile, 5-func-
tionalized pyrrole 5i was generated as the major product
(entry 11). Similar C-5-preferred formylation was also
observed in the reaction with N,N-dimethylformamide (entry
12). Especially when N,N-diethylformamide was reacted
under the conditions B, 5j was isolated as the sole product
(entry 14). In contrast, the reaction with ethyl formate, a more
reactive formylating agent for organolithiums, gave the
normal 2-formylated isomer 4j exclusively (entry 15). This
striking change of regioselectivity dependent on the reactivity
of electrophiles was also observed in the silylation reactions
with sterically demanding triisopropylsilyl chloride and
triisopropylsilyl triflate (entries 16-19). The less reactive
9
toluene
toluene
toluene
10e
11f
15
a MeOD (3.0 equiv) was added as a THF solution. b Isolated yield.
c
Determined by H NMR analysis (400 MHz). d Lithiation for 3 h. e 2.4
1
equiv of LTMP. f Lithiation for 12 h.
in THF at -78 °C for 1 h followed by quenching with MeOD
gave the deuterated pyrrole 2 in 91% yield after chromato-
graphic purification. The 400 MHz 1H NMR analysis of the
product indicated 72% total deuterium incorporation and
excellent C-2 regioselectivity8 (entry 1). A longer lithiation
time (3 h) did not improve the total deuterium incorporation
(entry 2). Utilization of LTMP instead of LDA resulted in
(8) Ortho-directing effect of bromine has been reported, see: (a) Arroyo,
Y.; Rodr´ıguez, J. F.; Sanz-Tejedor, M. A.; Santos, M. Tetrahedron Lett.
2002, 43, 9129–9132. (b) Mongin, F.; Marzi, E.; Schlosser, M. Eur. J. Org.
Chem. 2001, 2771–2777. (c) Lulin´ski, S.; Serwatowski, J. J. Org. Chem.
2003, 68, 5384–5387.
(3) Iwao, M.; Kuraishi, T. Tetrahedron Lett. 1985, 26, 6213–6216.
(4) Groenendaal, L.; Van Loo, M. E.; Vekemans, J. A. J. M.; Meijer,
E. W. Synth. Commun. 1995, 25, 1589–1600.
(5) Bailey, N.; Demont, E.; Garton, N.; Seow, H.-X. Synlett 2008, 185–
188.
(9) The pKa values of LTMP, LDA, and LHMDS are 37.3, 35.7, and
29.5, respectively. See: ref 1c.
(6) Robertson, J.; Kuhnert, N.; Zhao, Y. Heterocycles 2000, 53, 2415–
2420.
(10) Rutherford, J. L.; Collum, D. B. J. Am. Chem. Soc. 2001, 123,
199–202, and references cited therein.
(7) Ohta, T.; Fukuda, T.; Ishibashi, F.; Iwao, M. J. Org. Chem. 2009,
74, 8143–8153.
(11) LTMP was precipitated as white crystalline solid at -78 °C. This
might be the reason for inefficient lithiation of 1 with LTMP in toluene.
Org. Lett., Vol. 12, No. 12, 2010
2735