A reinvestigation of 4-hydroxyindole-6-carboxylate synthesis from pyrrole-2-carboxaldehyde: A facile synthesis of indoles and indolizines

Article abstract of DOI:10.1021/jo040191e

Cyclization of the Stobbe product 3 under the literature conditions (acetic anhydride/sodium acetate) affords both the indole 6 and the indolizine 8. The presence of base promotes the formation of indolizine products, and using acetic anhydride/triethylamine leads to indolizine products in good yield. Improved conditions for conversion to indoles 2, 5, and 6 are reported, and inconsistencies in some of the literature structure assignments and characterization data are noted.

Full text of DOI:10.1021/jo040191e

A Rein vestiga tion of
-Hyd r oxyin d ole-6-ca r boxyla te Syn th esis
fr om P yr r ole-2-ca r boxa ld eh yd e: A F a cile
Syn th esis of In d oles a n d In d olizin es
procedure that appears to involve a vinyl ketene inter-
5
,6
mediate 4 (Scheme 1). Several other groups have cited
this chemistry,6 but we were surprised that it has not
been used more often in view of the apparent simplicity
and the longer alternatives in the literature.2 Another
concern was that only partial NMR data had been
4
,7
,3
Musong Kim and Edwin Vedejs*
4
included in the original work, not sufficient to confirm
Department of Chemistry, University of Michigan,
30 North University Avenue, Ann Arbor, Michigan 48109
the structures. Here, we report our attempts to reproduce
the annulation from 3, along with revised procedures that
afford the desired indole 2. Spectroscopic characterization
of products is also included along with cautionary com-
ments regarding data reported in the original paper and
in subsequent work.4
9
edved@umich.edu
Received May 6, 2004
,6,7
Abstr a ct: Cyclization of the Stobbe product 3 under the
literature conditions (acetic anhydride/sodium acetate) af-
fords both the indole 6 and the indolizine 8. The presence
of base promotes the formation of indolizine products, and
using acetic anhydride/triethylamine leads to indolizine
products in good yield. Improved conditions for conversion
to indoles 2, 5, and 6 are reported, and inconsistencies in
some of the literature structure assignments and charac-
terization data are noted.
Half-ester 3 was prepared by Stobbe condensation of
pyrrole-2-carboxaldehyde and dimethyl succinate as
described,4 and the E-geometry was confirmed by NOE
methods. However, treatment of 3 with sodium acetate
a
4a
and acetic anhydride following the published procedure
resulted in a mixture of two isomeric acetates that proved
to be 6 and 8, corresponding to cyclization of the vinyl
ketene 4 via the phenolic intermediates 5 and 7. The
identity of 6 and 8 was not immediately clear, so
conditions were varied in attempts to favor one of the
In the course of a synthesis project, we were interested
in rapid access to 4-hydroxyindole-6-carboxylic acid
derivatives 1 and 2. Given the frequent occurrence of
pathways (Table 1). The combination of Et
forded a single product according to NMR assay (entry
3), and a similar reaction using Et N/methyl chlorofor-
3 2
N/Ac O af-
3
1
similar indoles in biologically active compounds and
mate was especially clean and gave a major product in
80% yield after crystallization. However, the chlorofor-
mate product proved to be the indolizine 9 according to
their relationship to FR900482 and related structures,2
there was no shortage of potential approaches. However,
we could find only three demonstrated strategies in the
literature for preparation of specific indoles having the
substitution pattern of 1 or 2.2a,3,4
,3
1
13
H and C NMR data, including a decisive DEPT
experiment that revealed the presence of five vinylic CH
carbons. Based on similarities in the NMR spectra, the
3 2
Et N/Ac O product was therefore assigned as the acetoxy
indolizine 8.
The assignment of structure 8 supported the conclusion
that indole 6 is indeed the major product under the El-
Rayyes conditions (Table 1, entry 1) and stimulated
attempts to improve the product ratio in favor of 6.
Heating with acetic anhydride in the absence of base
(
entry 5) was promising in the sense that 8 was not
detected, but this experiment gave the diacylated indole
0 as well as 6. Better results were obtained in toluene
One of the published approaches is rather lengthy,
eight steps from 3-acetyl-1-phenylsulfonylpyrrole to an
1
acetal corresponding to a protected analogue of 1 (PhSO
2
N
A
C).2
a
using a limited amount of the anhydride. This prevented
the formation of 10 and gave better ratios of 6:8. Finally,
the optimal conditions for indole formation were found
by refluxing 3 in toluene with acetic anhydride and an
equivalent of acetic acid (entry 7). These results suggest
that the undesired cyclization to the indolizine 7 is
in place of NH; 1,3-dioxan-2-yl in place of R′O
2
more concise route was developed by Ziegler et al. using
the Batcho-Leimgruber indole synthesis to prepare a
4
-benzyloxy indole (2 with R) Bn in place of Me), but
3
six steps were required. The shortest sequence to related
indoles has been reported by El-Rayyes from the Stobbe
4
2
condensation product 3 and NaOAc/Ac O, a cyclization
(5) (a) Perri, S. T.; Moore, H. W. J . Am. Chem. Soc. 1990, 112, 1897.
(
b) Turnbull, P.; Moore, H. W. J . Org. Chem. 1995, 60, 644. (c) Karady,
(
1) General reviews: Gribble, G. W. In Rodd’s Chemistry of Carbon
S.; Amato, J . S.; Reamer, R. A.; Weinstock, L. M. Tetrahedron Lett.
1996, 37, 8277.
Compounds, 2nd ed.; 1997; Vol. 4, p 69. Lounasmaa, M. Curr. Org.
Chem. 1998, 2, 63. Leonard, J . Nat. Prod. Rep. 1999, 16, 319. Gribble,
G. W. J . Chem. Soc., Perkin Trans 1 2000, 7, 1045. Scholz, U.;
Winterfeldt, E. Nat. Prod. Rep. 2000, 17, 349.
(6) (a) Brenna, E.; Fuganti, C.; Perozzo, V.; Serra, S. Tetrahedron
1997, 53, 15029. (b) Fuganti, C.; Serra, S. J . Chem. Res., Synop. 1998,
638. (c) Fuganti, C.; Serra, S. J . Chem. Res., Miniprint 1998, 2726. (d)
Serra, S.; Fugante, C.; Moro, A. J . Org. Chem. 2001, 66, 7883.
(7) (a) Kampe, W.; Stach, K.; Thiel, M.; Bartsch, W.; Schaumann,
W. Ger. Offen. DE 2508251, 1976; Chem. Abstr. 1977, 86, 55280. (b)
Kampe, W.; Stach, K.; Thiel, M.; Bartsch, W.; Dietmann, K.; Roesch,
E.; Schaumann, W. US 4,152,446, 1980; Chem. Abstr. 1981, 94, 156743.
(c) Berthold, R.; Vogel, A. Eur. Pat. Appl. EP 191724, 1986; Chem.
Abstr. 1987, 106, 102290. (d) Berthold, R.; Ott, H. Ger. Offen. DE
3524955, 1986; Chem. Abstr. 1987, 106, 84635. (e) Mattei, P.; Pflieger,
P. PCT Int. Appl. WO 0183474, 2001; Chem. Abstr. 2001, 135, 357939.
(2) (a) Utsunomiya, I.; Muratake, H.; Natsume, M. Chem. Pharm.
Bull. 1995, 43, 37. (b) Fukuyama, T.; Goto, S. Tetrahedron Lett. 1989,
0, 6491. (c) Williams, R. M.; Rollins, S. B.; J udd, T. C. Tetrahedron
000, 56, 521.
3
2
(
3) Ziegler, F. E.; Belema, M. J . Org. Chem. 1997, 62, 1083.
(4) (a) El-Rayyes, N. R. J . Prakt. Chem. 1973, 315, 295. (b) A related
study has appeared using N-methylpyrrole derivatives where the
alternative cyclization pathway is blocked: El-Rayyes, N. R.; Al-
Salman, N. A. J . Prakt. Chem. 1976, 318, 816.
1
0.1021/jo040191e CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/09/2004
J . Org. Chem. 2004, 69, 6945-6948
6945

SCHEME 1
TABLE 1. Cycliza tion of 3 to 6, 8, or 10
the limited data available for comparison, but we note
that the methoxy indolizine 11 (from saponification of 9
and O-methylation) has similar H NMR methyl shifts
amt of
Ac2O
(equiv) (h)
1
additive
(equiv)
time
entry
solvent/T (°C)
products
compared to 2, while the melting point is 103-104 °C,
4
a
1
2
3
4
5
6
7
8
NaOAc
NaOAc
Et3N (5)
K2CO3
None
excess
excess
excess
excess
excess
10
a
4
1
2
8
16
16
6
Ac2O (rt to 75) 6:8 (1.8:1)
Ac2O (reflux)
Ac2O (rt)
Ac2O (reflux)
Ac2O (75)
toluene (reflux) 6:8 (5:1)
toluene (reflux) 6:8 (8:1)
toluene (reflux) 6:8 (7:1)
nearly the same as the reported value for “2”. This
raises the possibility that some of the original experi-
ments may have used cyclization products containing
both 6 and 8 to explore the preparation of derivatives
and that some of the data reflect the enrichment of one
isomer or the other (indole or indolizine) by crystalliza-
tion. We were able to reproduce the mp data for “1” ) 1,
but not for “2“ or ”6”. We cannot comment further on the
El-Rayyes characterization data, although the structural
8:10 (1:2)
8 (>95%)
8:10 (1:1.3)
6:10 (1:1)
None
AcOH (1) 10
AcOH (4) 10
a
The reaction mixture was stirred 14 h at rt and was then
warmed for 6 h at 75 °C.4a
4
a
assignments appear to be correct as published.
The procedure used for the synthesis of 9 is nearly
identical to a method reported in 1998 by Serra et al. for
cyclization of the Stobbe condensation product from
pyrrole-2-carboxaldehyde.6 The latter workers did not
isolate the mixed carbonate ester, but reported saponi-
fication to a product assigned as 4-hydroxyindole-6-car-
boxylic acid “1” (mp 262-4 °C). The melting point is
nearly the same as that provided by El-Rayyes for 1, and
Serra’s NMR data are similar to our chemical shift data
for 1. However, we obtained only the indolizine 8 when
using Serra’s procedure, and our attempts to saponify 8
did not give 1.8
catalyzed by base, but we have not determined whether
cyclization occurs from the vinyl ketene 4 or at the stage
of the mixed anhydride intermediate from 3.
c
The best conditions gave 6 contaminated with a
significant amount of 8, so we explored the possibility of
purification at a later stage. Crystallization was easier
after saponification of 6 to 5, and subsequent O-methy-
lation gave the desired 2. Similarities in the NMR spectra
of 2 and 6 indicate that both structures belong to the
indole series. A DEPT experiment confirmed that 2 has
four vinylic CH carbons as required, and the presence of
1
an NH proton at 8.63 ppm (br s) in the H NMR spectrum
Several patents also describe the synthesis of indole
confirmed the connectivity. Although the overall yield of
7
derivatives using the El-Rayyes method. Relevant NMR
2
from 3 was moderate (38%), the procedure was easy to
data were not provided in these reports, so we have no
basis to suggest that the structures were not the desired
indoles. Nevertheless, it would be prudent to look closely
at the spectroscopic data for the resulting indoles in view
of the structural issues and the possibility of contamina-
tion by indolizines.9
carry out on a multigram scale and solved our prepara-
tive problem.
Comparisons with literature structures proved some-
what difficult because El-Rayyes reported no NMR data
beyond the methyl chemical shifts in 6 (δ 3.75, 2.3 ppm
,10
4
a
in CDCl
3
6
3
). Our values for 6 are somewhat different (δ
.92, 2.41 ppm), but the methyl shifts do not distinguish
from 8 (δ 3.90, 2.45 ppm). Because 6 is the major
In conclusion, the cyclization of 3 under the literature
4
a
conditions (acetic anhydride/sodium acetate) affords
4
a
product under the reported conditions, the original
structural assignment is confirmed. However, El-Rayyes
did not comment on the formation of the isomer 8,
resulting in questions regarding other structural assign-
ments in his paper. For example, saponification of “6” to
8) Under the conditions reported by Serra,6c saponification of 8 gave
mixture of structures, apparently including the phenol 7 and
(
a
tautomeric unsaturated lactams expected from acetate cleavage, but
with the methyl ester intact. Under more forcing conditions (15%
NaOH, reflux 1 h), both the ester and the lactam were cleaved to give
2
-(1H-pyrrol-2-ylmethylene)succinic acid, the diacid corresponding to
the Stobbe condensation product 3.
“1” and conversion of “1” to “2” were reported. The
(
9) The patent literature reports a phenol assigned structure 5, mp
products were characterized without including NMR
data, but melting points were provided, “1” (mp 265 °C);
80-81 °C, but includes no NMR data for this substance;7d we find a
melting point of 148-149 °C for 5.
(
10) (a) General reviews: Swinbourne, F. J .; Hunt, J . H.; Klinkert,
“2” (mp 102 °C); “6” (mp 151 °C). Only one of these values
G. In Advances in Heterocyclic Chemistry; Katritzky, A. R., Boulton,
A. J ., Eds.; Academic Press: New York, 1978; vol. 23, pp 104-170.
Flitsch, W. In Comprehensive Heterocyclic Chemistry; Katritzky, A. R.,
Rees, C. W., Eds.; Pergamon Press: New York, 1984; Vol. 4, pp 443-
agrees with our data: 1 (mp 264-7 °C, crystallized from
hexane/EtOAc); 2 (mp 135-6 °C, crystallized from hex-
ane/EtOAc); 6 (mp 137-8 °C, crystallized from hexane).
The discrepancies cannot be fully explained because of
4
95. (b) Padwa, A.; Austin, D. J .; Precedo, L.; Zhi, L. J . Org. Chem.
1993, 58, 1144.
6
946 J . Org. Chem., Vol. 69, No. 20, 2004

both the indole 6 and the indolizine 8. The presence of
base promotes the formation of indolizine products,
especially in the case of the soluble base triethylamine.
Although mechanistic details have not been investigated,
vinyl ketene intermediates have been invoked to explain
similar annulations starting from carbocyclic analogues
of 3.5 The best ratio in favor of indole products was
obtained using 1 equiv of acetic acid and 10 equiv of acetic
anhydride in refluxing toluene, resulting in preparatively
useful conversion to the 4,6-disubstituted indoles 2 and
2 2
dissolved in CH Cl and filtered through a plug of silica gel.
Concentration of eluent afforded 2 (2.31 g, 84%) as a light yellow
solid. The analytical sample was prepared by recrystallization
from ethyl acetate/hexanes to give light yellow crystals: mp )
1
1
(
35-136 °C; H NMR (500 MHz, CDCl
3
) δ 8.63 (1H, br s) 7.84
1H, t, J ) 1.0 Hz) 7.26 (1H, d, J ) 5.6 Hz) 7.21 (1H, d, J ) 1.0
Hz) 6.70 (1H, ddd, J ) 2.9, 2.0, 0.7 Hz) 4.00 (3H, s) 3.93 (3H, s);
,6
13
C NMR (100 MHz, CDCl ) δ 168.2, 152.8, 136.2, 125.9, 124.5,
3
122.4, 107.6, 100.3, 100.0, 55.4, 52.0; IR (neat) 3323, 1690, 1580
-
1
+
cm ; MS (EI) 205 (M , 100), 190 (50), 174 (42), 146 (32), 132
+
(
2
23), 131 (24), 103 (28); HRMS calcd for C11
05.0739, found m/z 205.0744.
-Acetoxy-1H-in d ole-6-ca r boxylic Acid Meth yl Ester (6)
a n d 7-Acetoxy-7a H-in d en e-5-ca r boxylic Acid Meth yl Ester
H
11NO
3
(M )
5
. Under basic conditions, the indolizines 8 and 9 are
4
formed, and 9 can be isolated in good yield. Few methods
are known for the preparation of 5-oxygenated indoliz-
ines,10 and the base-modified El-Rayyes conditions may
have some potential in this context.
4a
(
8). The published procedure of El-Rayyes was followed, but
the products were isolated by flash chromatography on silica
gel using 1:1 hexane/ethyl acetate instead of direct crystalliza-
4
tion. 6: R
mp 151 °C); H NMR (400 MHz, CDCl
(
f
0.51; mp 137-8 °C, crystallized from hexane (lit.
1
3
) δ 8.72 (1H, bs), 7.97
1H, s), 7.54 (1H, d, J ) 1.1 Hz), 7.25 (1H, dd, J ) 3.3, 2.6 Hz),
Exp er im en ta l Section
6
.44 (1H, m), 3.92 (3H, s), 2.41 (3H, s). 8: R
f
0.74, mp 73-75 °C
) δ 8.13
1H, s), 7.33 (1H, d, J ) 1.1 Hz), 6.94 (1H, d, J ) 1.5 Hz), 6.91
2
-(1H-P yr r ol-2-ylm eth ylen e)su ccin ic Acid 1-Meth yl Es-
1
4
a
(crystallized from hexane); H NMR (400 MHz, CDCl
3
(
(
ter (3). Following the El-Rayyes procedure, dimethyl succinate
41.3 mL, 316 mmol) was added to a stirring solution of 1H-
(
1H, dd, J ) 2.9, 1.1 Hz), 6.8 (1H, dd, J ) 4.0, 1.1 Hz), 3.9 (3H,
pyrrole-2-carbaldehyde (20.0 g, 210 mmol) in benzene (140 mL).
The solution was cooled (ice bath), and 60% NaH suspended in
mineral oil (16.8 g, 421 mmol) was added. The resulting
suspension was allowed to warm to rt and was then stirred for
s), 2.45 (3H, s).
5-Meth oxyca r bon yloxyin d olizin e-7-ca r boxylic
Acid
Meth yl Ester (9). Methyl chloroformate (5.91 mL, 76.3 mmol)
and triethylamine (25.3 mL, 181.6 mmol) were added to a
stirring solution of 3 (7.6 g, 36.3 mmol) in THF (300 mL) at 0
°C. The resulting solution was allowed to warm to rt and then
stirred for 2 h followed by evaporation (aspirator). The resulting
paste was redissolved in hexane/ethyl acetate (1:1) and filtered
through a plug of silica gel to give a brown oil (12.57 g). Crystalli-
zation from hexane/ethyl acetate afforded 9 (7.24 g, 80%)
1
6 h. The reaction mixture was quenched with water at 0 °C,
and the aqueous layer was washed with ether and 5% KOH,
followed by addition of NaCl (5 g) to the combined aqueous
layers. After acidification with concd HCl to pH < 1 at 0 °C, the
precipitate was filtered to afford 3 (42.3 g, 96%) as a brown solid,
1
mp ) 134-136 °C, that was used without purification: H NMR
(
3
400 MHz, CD OD) δ 10.94 (1H, br s) 7.72 (1H, s), 6.96 (1H,
1
ddd, J ) 2.7, 2.7, 1.4 Hz) 6.52 (1H, dm, J ) 2.5 Hz) 6.26 (1H, m)
as a light orange solid: mp ) 92-94 °C; H NMR (400 MHz,
3
1
1
.77 (3H, s) 3.62 (2H, s); 13C NMR (125 MHz, CD
70.5, 133.4, 133.3, 128.8, 123.4, 123.3, 118.4, 114.6, 114.5, 112.1,
3
OD) δ 175.2,
CDCl ) δ 8.16 (1H, d, J ) 0.7 Hz), 7.42 (1H, m), 7.03 (1H,
3
d, J ) 1.5 Hz), 6.93 (1H, dd, J ) 4.0, 2.9 Hz), 6.82 (1H, dd,
-
1
13
12.0; IR (neat) 3323, 1684, 1613 cm ; MS (DCI/NH
3
) m/z 210
J ) 4.0, 1.5 Hz), 4.00 (3H, s), 3.91 (3H, s); C NMR (100
+
(
MH , 40), 209 (15), 194 (14), 193 (19), 192 (100), 191 (46), 178
MHz, CDCl ) δ 166.0, 152.0, 139.2, 133.3, 120.5, 118.8, 115.9,
3
+
(20), 166 (30); HRMS calcd for C10
H
12NO
4
(MH ) 210.0766, found
-1
1
10.7, 105.9, 98.3, 56.4, 52.1; IR (neat) 1779, 1713 cm ; MS
) m/z 250 (MH , 100), 205 (21), 192 (24), 191 (20), 190
(44); HRMS calcd for C12H12NO5 (MH ) 250.0715, found m/z
m/z 210.0765.
-H yd r oxy-1H -in d ole-6-ca r b oxylic Acid Met h yl E st er
5). Acetic acid (2.73 mL, 47.8 mmol) and acetic anhydride (45.1
+
(DCI/NH
4
+
4
(
250.0715.
mL, 478 mmol) were added to a stirring solution of 3 (10 g, 47.8
mmol) in toluene (400 mL) and refluxed for 16 h. After being
cooled to 0 °C, the resulting suspension was diluted with ether
4
-Acetoxy-1-a cetyl-1H-in d ole-6-ca r boxylic Acid Meth yl
Ester (10). The procedure described for cyclization was per-
formed as described for the preparation of 5, except that toluene
was replaced by acetic anhydride as the solvent (4 h, reflux).
Chromatography as described for preparation of 8 gave a polar
fraction consisting of 10. Crystallization from hexane/ethyl
(
200 mL) and quenched with satd NaHCO
was extracted with ether, and the combined organics were
washed with satd NaHCO , H O, brine, and dried (Na SO ) to
3
. The aqueous layer
3
2
2
4
afford a brown solid (9.42 g). The crude product was redissolved
in MeOH (350 mL), and NaOMe (10.9 g, 201 mmol) was added.
The resulting solution was refluxed for 1 h. After being cooled
to 0 °C, the reaction mixture was diluted with ether (150 mL)
and acidified with 1 N HCl. The aqueous layer was extracted
with ether, and the combined organic layers were washed with
1
acetate gave 10 as fine filament-like crystals: mp 165-7 °C; H
3
NMR (400 MHz, CDCl ) δ 9.01 (1H, s), 7.75 (1H, s), 7.57 (1H, d,
J ) 3.7 Hz), 6.59 (1H, d, J ) 3.7 Hz), 3.95 (3H, s), 2.68 (3H, s),
2
.41 (3H, s); MS calcd for C14
m/ z 298.0679 (M + Na ).
5
H
13NNaO
5
298.0691 (ES/Na), found
+
-Met h oxyin d olizin e-7-ca r b oxylic Acid Met h yl E st er
H
(
(
2
O and brine and dried (Na
8.84 g) as a brown oil. Crystallization from CH
4.23 g, 46% over two steps) as a shiny light brown solid: mp )
2
SO
4
) to give crude hydroxy indole
(
11). Sodium methoxide (95%, 4.22 g, 72.2 mmol) was added to
2
2
Cl afforded 5
a stirring solution of 9 (4.50 g, 18.1 mmol) in methanol (150 mL)
at 0 °C. The resulting solution was quenched with pH 3
phosphate buffer after 20 min. The aqueous layer was extracted
1
1
7
)
48-149 °C; H NMR (400 MHz, CD
.70 (1H, t, J ) 1.1 Hz) 7.31 (1H, d, J ) 3.3 Hz) 7.08 (1H, d, J
1.5 Hz) 6.61 (1H, dd, J ) 2.9, 0.7 Hz) 3.88 (1H, s); 13
OD) δ 170.2, 151.1, 138.6, 127.4, 124.8,
23.4, 107.4, 104.5, 100.0, 52.3; IR (neat) 3399, 1676, 1584
3
OD) δ 10.83 (1H, br s)
C
with CH
2 2
Cl , and the combined organic layers were dried over
NMR (100 MHz, CD
1
3
Na SO to give a brown solid (3.49 g). Attempts to carry out the
2
4
7
e
O-alkylation using iodomethane/KO-t-Bu gave products of
C-methylation. For this reason, the crude product was dissolved
-1
cm ; MS calcd for C10
9 3
H NNaO 214.1 (ES/Na), found m/ z 214.1
+
(M + Na ).
in CH
romethanesulfonate (3.10 mL, 27.4 mmol) was added (no added
base). The resulting mixture was quenched with satd NH Cl (40
mL) after stirring for 20 h at rt. The aqueous layer was extracted
with CH Cl , and the combined organic layers were dried over
Na SO and evaporated (aspirator) to give a dark green oil (4.17
2 2
Cl (180 mL) and cooled to 0 °C, and methyl trifluo-
4
-Met h oxy-1H -in d ole-6-ca r b oxylic Acid Met h yl E st er
(
2). Solid K CO (5.58 g, 40.4 mmol) and MeI (0.839 mL, 13.5
2
3
4
mmol) were added to a stirring solution of 5 (2.58 g, 13.5 mmol)
in DMF (68 mL) at 0 °C. After being stirred for 16 h at 0 °C, the
resulting suspension was quenched with 1 N HCl (30 mL) and
the aqueous layer was extracted with ether. The combined
2
2
2
4
g). Filtration through a plug of silica gel (1:1 hexanes/ethyl
acetate) afforded 11 (3.13 g, 84% over two steps) as an off-white
solid. A sample was prepared by flash chromatography (3:1,
organic phase was washed with H
2
O and brine and dried over
Na SO to give a brown solid (2.66 g). The crude product was
2
4
J . Org. Chem, Vol. 69, No. 20, 2004 6947

hexanes/ethyl acetate) and crystallization from hexane: mp )
Ack n ow led gm en t. This work was supported by the
National Institutes of Health (CA17918).
1
1
03-104 °C; H NMR (400 MHz, CDCl
3
) δ 7.92 (1H, d, J ) 0.7
Hz) 7.53 (1H, ddd, J ) 1.5, 1.5, 0.7 Hz) 6.87 (1H, dd, J ) 3.7,
2
4
1
.6 Hz) 6.71 (1H, dd, J ) 4.0, 1.5 Hz) 6.36 (1H, d, J ) 1.1 Hz)
.10 (3H, s) 3.91 (3H, s); 13C NMR (125 MHz, CDCl
) δ 167.0,
48.5, 132.6, 119.6, 115.8, 115.0, 110.7, 104.44, 83.59, 56.1, 51.98;
Su p p or tin g In for m a tion Ava ila ble: NMR spectra of key
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
3
-
1
+
IR (neat) 3167, 1703, 1625 cm ; MS (EI) m/z 205 (M , 83), 190
+
3
(100), 131 (14), 103 (11); HRMS calcd for C11H11NO (M )
2
05.0739, found m/z 205.0730.
J O040191E
6
948 J . Org. Chem., Vol. 69, No. 20, 2004