Synthesis and SAR of selective benzothiophene, benzofuran, and indole-based peroxisome proliferator-activated receptor δ agonists

Article abstract of DOI:10.1016/j.bmcl.2007.04.047

Recent literature has suggested the benefit of selective PPARδ agonists for the treatment of atherosclerosis and other disease states associated with the metabolic syndrome. Herein we report the synthesis and structure-activity relationships of a series of novel and selective PPARδ agonists. Our search began with identification of a novel benzothiophene template which was modified by the addition of various thiazolyl, isoxazolyl, and benzyloxy-benzyl moieties. Further elucidation of the SAR led to the identification of benzofuran and indole based templates. During the course of our research, we discovered three new chemical templates with varying degrees of affinity and potency for PPARδ versus the PPARα and PPARγ subtypes.

Full text of DOI:10.1016/j.bmcl.2007.04.047

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3630–3635
Synthesis and SAR of selective benzothiophene, benzofuran,
and indole-based peroxisome proliferator-activated
receptor d agonists
Gary F. Filzen,* Larry Bratton, Xue-Min Cheng, Noe Erasga, Andrew Geyer, Chitase Lee,
Gina Lu, Jim Pulaski, Roderick J. Sorenson, Paul C. Unangst,
B. K. Trivedi and Xiangyang Xu
Department of Chemistry, Pfizer Global Research and Development, Michigan Laboratories, Ann Arbor, MI 48105, USA
Received 14 February 2007; revised 12 April 2007; accepted 17 April 2007
Available online 24 April 2007
Abstract—Recent literature has suggested the benefit of selective PPARd agonists for the treatment of atherosclerosis and other dis-
ease states associated with the metabolic syndrome. Herein we report the synthesis and structure–activity relationships of a series of
novel and selective PPARd agonists. Our search began with identification of a novel benzothiophene template which was modified
by the addition of various thiazolyl, isoxazolyl, and benzyloxy-benzyl moieties. Further elucidation of the SAR led to the identifi-
cation of benzofuran and indole based templates. During the course of our research, we discovered three new chemical templates
with varying degrees of affinity and potency for PPARd versus the PPARa and PPARc subtypes.
ꢀ 2007 Elsevier Ltd. All rights reserved.
The peroxisome proliferator-activated receptors (PPARs)
are members of the nuclear hormone receptor super
family. The PPARs consist of three subtypes: PPARa,
PPARc, and PPARd (also known as PPARb). The
PPARs’ putative biological role is the maintenance of li-
pid and glucose homeostasis through downstream sig-
naling to the biochemical machinery responsible for
energy storage and catabolism.1 PPARa, expressed pri-
marily in the liver, is associated with lipoprotein catab-
olism. PPARc, expressed primarily in adipose tissue, is a
regulator of glucose and insulin homeostasis and fatty
acid storage. The role of PPARd, ubiquitously ex-
pressed, is less well understood but recent research has
suggested a variety of biological functions centering
around fatty acid oxidation in tissues with high energy
demands such as heart, adipose, and skeletal mu-
scle.2a–d PPARd has begun to gain more attention as a
molecular target for the treatment of cardiovascular dis-
eases. Notable is a report which states that a selective
PPARd agonist increased HDLc by 80%, while reducing
LDL-cholesterol by 29%, in rhesus monkeys.3 The po-
tential of PPARd agonists as pharmaceutical agents
for HDL-cholesterol elevation prompted our efforts to
find novel and selective compounds.
Examination of the literature revealed few reports of
selective PPARd ligands. However, it was clear that
the prototypical selective PPARd agonists (Fig. 1), such
as GW501516, L-165041, and L-783483, had either an
acetoxy or acetic acid head group attached to an aryl
core, with a variety of lipophilic tail portions.4
This was in contrast to the acidic head groups found in
most PPARa and non-TZD PPARc agonists. The a and
cPPAR subtypes can accept and prefer bulky lipophilic
substituents adjacent to the requisite carboxylic acid
function. This is due to the relatively large hydrophobic
pocket found adjacent to the activation domain in
PPARa and PPARc. The same hydrophobic pocket in
PPARd is considerably narrower and will not accommo-
date bulky substituents.5 This indicated to us that a
large component of PPARd subtype specificity (but
not exclusively) would be determined by the amount
of steric crowding around the acidic head portion rather
than what was found at the aryl core or lipophilic tail.
Our strategy was to utilize a generic PPAR scaffold
(Fig. 2) consisting of an acidic head group, aryl body,
and lipophilic tail to construct new compounds.
Keywords: PPAR; Benzothiophene; Benzofuran; Indole.
*
Corresponding author. E-mail: Gary.filzen@pfizer.com
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.04.047

G. F. Filzen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3630–3635
3631
O
O
O
HO
N
S
OH
S
O
O
O
O
OH
F3C
L-165041
GW501516
F3C
N
O
OH
O
O
S
Cl
L-783483
Figure 1. PPARd selective agonists.
Specifically, in order to increase the binding affinity of
our new ligands, we sought novel head/body combina-
tions that would have at least one less degree of rota-
tional freedom relative to the prototypical arylacetoxy
and arylacetic acid motifs. The caveat to this approach
being that the requisite carboxyl function still maintains
its ability to fit in the relatively narrow activation do-
main of PPARd.
we derived the 6-hydroxybenzofuran (8), 5-hydroxy in-
dole acetic acid (13), and 5-mercapto-indole-acetic acid
(15) cores.
Once identified, we began to combine the new acidic
head/aryl body combinations with a previously known
tail and others discovered through screening of a combi-
natorial library.
A substructure search of our proprietary chemical li-
brary using these criteria initially led us to the 6-hydrox-
ybenzothiophene-3-acetic acid core (6) and from there
The 6-hydroxybenzothiophene-3-acetic acid core was
prepared by alkylation of 3-methoxythiophenol with
methyl 4-chloroacetoacetate followed by an acid medi-
ated cyclocondensation. The cyclization to the benzothi-
ophene core resulted in a regioisomeric mixture of
compounds in a 3:2 ratio (Scheme 1). The mixture of
regioisomers could not be separated by recrystallization
or chromatography, so it was used as is in the next step.
De-methylation was accomplished using boron tribro-
mide in dichloromethane at 0 ꢁC. After work up,
chromatography, and recrystallization, the desired
6-hydroxybenzothiophen-3-acetic acid core was isolated
in 37% yield over 3 steps. Methyl and chloro substituted
analogues of 6 were prepared in a similar manner. We
Aryl Body
Lipophilc Tail
Acidic Head
rotational restrictor
Figure 2. Prototypical PPAR ligand with rotational restrictor.
O
O
O
O
4
O
S
5
MeO
Route A
a
b
6
O
SH
O
S
3
2
1
6-MeO:4-MeO
3:2
O
O
c
O
O
O
Route B
a
b
HO
SH
HO
S
HO
S
5
6
4
Scheme 1. Reagents and conditions: (a) methyl 4-chloroacetoacetate, Cs2CO3, CH3CN/DMF 5:1; (b) methanesulfonic acid, DCM; (c) BBr3, DCM,
0 ꢁC. Route A, 37% over three steps; Route B, 34% over two steps.

3632
G. F. Filzen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3630–3635
later determined that it was not necessary to have the la-
tent hydroxy group protected to arrive at 6, and in fact it
seemed preferable judging by the regiochemical ratio of
the cyclocondensation step. In the event, selective alkyl-
ation of the thiol of 3-hydroxy thiophenol with methyl
4-chloroacetoacetate gave the intermediate acetoacete-
tate derivative 5. Cyclization with excess methanesulfon-
ic acid in dichloromethane at room temperature gave the
desired 6-hydroxybenzothiophen-3-acetic acid core and
only trace amounts of the regiochemical partner. After
chromatography and recrystallization, 6 was isolated
in 34% yield over two steps.
bromoacetate using sodium hydride as base (Scheme
4). The resulting compound was then de-benzylated
using hydrogen and palladium as catalyst. The thiol
was introduced following the two-step Newman–Kwart
procedure.8 After trying several conditions to effect the
rearrangement, we found that the best yields were ob-
tained when the thiocarbamate in warm diphenyl ether
was added to diphenyl ether at 259 ꢁC.
The final indole compounds 16a–g were then prepared
in the same manner as described for compounds 10a–t
(Table 2).
The benzofuran core was made by a two-step literature
procedure (Scheme 2).6
The synthesized compounds 10a–t and 16a–g were eval-
uated for in vitro binding and agonist activity at
hPPARd using known methods.9,10 We started our
SAR by making analogues with the benzothiopheneace-
tic acid core.11 The unsubstituted analogues (10a and
10b) had a strong affinity for PPARd versus PPARa.
In general, we found that most of the compounds re-
ported herein had very little affinity for PPARc. The
example of a benzothiopheneacetic acid core and lipo-
philic thiazole tail (10a) demonstrated potency and
selectivity for PPARd similar to that of GW501516.
Replacing the lipophilic thiazole tail with a chlorophe-
nyl isoxazole (10b) resulted in only a slight loss in po-
The lipophilic tails (Fig. 3) were either commercially
available or were prepared according to literature
procedures.7
The final compounds, 10a–t, were then made in a simple
two-step procedure starting with alkylation of the free
phenol in acetonitrile with the chloride of the lipophilic
tail using cesium carbonate as base (Scheme 3). After
purification, the resulting esters 9a–t were saponified
with lithium hydroxide to give the desired carboxylic
acids in good to excellent yields (Table 1).
tency but
trifluoromethylphenyl isoxazole analogue (10c) had sim-
a
greater degree of selectivity. The
The (5-mercapto-indol-1-yl)-acetic acid core was pre-
pared by alkylating commercially available 5-ben-
zyloxyindole in dimethylformamide with methyl
ilar potency to the chlorophenyl analogue, but surpris-
a
ingly had
greater affinity for PPARa. The
benzofuran counterparts (10d–f) were less potent, per-
haps due to the smaller size of oxygen relative to sulfur,
leaving unfilled space within the PPARd activation do-
main. By comparison the indole analogues 16a–b were
about 2- to 3-fold less potent and selective for PPARd
versus PPARa than the corresponding benzothiophene
analogues. Substitution at the 4- and 5-position of the
benzothiophene core in all cases resulted in a diminution
of PPARd potency. Compounds containing benzyloxy-
benzyl tails (10n–t and 16d–f) represented an unprece-
dented addition to our stable of lipophilic tails. The
O
Cl
O
a, b
O
HO
O
O
HO
7
8
Scheme 2. Reagents and condition: (a) NaOH, reflux; (b) LiOH,
(MeO)2SO2.
N
N
N
O
O
S
F3C
O
F3C
F3C
Cl
o, m, p - D
C
A
B
Figure 3. Lipophilic tails.
O
O
O
O
OH
O
R
R
R
a
b
HO
Cl
Y
Y
O
Y
O
+
X
X
X
10a-t
X = O, S
9a-t
Y = lipophilic tail A, B, C, or D
Scheme 3. Reagents and conditions: (a) Y-CH2Cl, Cs2CO3, CH3CN, 40–87%; (b) LiOH/water/THF, 55–96%.

G. F. Filzen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3630–3635
Table 1. Activity of benzothiophene and benzofuran based compounds
3633
OH
O
R
Y
O
X
10a-t
a,c
IC50 (nM)
b,c
EC50 hPPARd
(nM)
Compound
R
X
Y = lipophilic tail
Fold selectivity: hPPARd
vs. hPPARa
hPPARd
hPPARa
hPPARc
GW501506
—
H
—
S
—
A
2.6
3.1
7.7
11
463
482
NT
178
155
496
24
4.0
a
b
c
d
e
f
>10,000
>10,000
NT
14.7
226
H
S
B
3820
261
H
S
C
305
243
H
O
O
O
S
A
14
81
679
NT
49
29
H
B
2380
1420
683
>10,000
>10,000
6260
323
H
C
25
57
1610
NT
g
h
i
5-Me
4-Cl
4-Me
5-Me
4-Cl
4-Me
4-Cl
H
A
237
58
3
18
S
A
1040
311
>10,000
8660
NT
227
638
S
B
23
37
14
j
S
B
5810
3840
1470
1350
7860
4680
963
159
10
656
k
l
S
B
375
37
NT
>10,000
NT
6360
561
S
C
40
14
m
n
o
p
q
r
S
C
97
2630
2230
172
S
o-D
o-D
m-D
p-D
o-D
m-D
p-D
244
43
>10,000
7580
5790
32
5-Me
H
S
108
65
S
15
166
H
S
3.6
662
24.1
81
229
3730
64
1070
3120
1490
3440
H
O
O
O
>10,000
2500
2380
>10,000
5400
>10,000
>15
104
29
s
H
t
H
NT = not tested.
a Concentration that inhibits 50% of the interaction between the PPAR LBD and the radiolabeled ligand.
b Concentration of test compound which produced 50% of the maximal reporter activity.
c The results are based on at least two experiments, each dose done in triplicate (SD = 10%).
O
O
H
O
O
N
N
N
b
a
BnO
BnO
HO
11
12
13
O
O
O
O
N
N
S
c
d, e, f
HS
N
O
14
15
Scheme 4. Reagents and conditions: (a) methyl bromoacetate, NaH, CH3CN/DMF; (b) H2, 10% Pd/C; (c) N,N-dimethylthiocarbamoyl chloride,
NaH, 39%; (d) diphenyl ether, 259 ꢁC, 17 h, 82%; (e) KOH, ethanol, reflux; (f) MeOH, p-TsOH, reflux, steps (e–f) 40%.
order of increasing potency for the benzothiophene
based compounds was ortho substitution, meta, and
then para. The benzofuran based compounds closely
matched their benzothiophene counterparts though they
were generally less potent. The meta substituted benzyl-
oxy-benzyl tails most closely resembled the aryl thiazole
and isoxazole tails found in the benzofuran and benzo-
thiophene examples in terms of shape, potency and
selectivity. Interestingly, even though compounds with
the ortho substituted benzyloxy-benzyl tails were less po-
tent at PPARd they had very little, if any, affinity for
PPARa and PPARc.
It is clear that there is a divergence between binding
affinity and functional activity with many of these com-
pounds.10 This is explained by the lipophilicity of the
compounds muting passive diffusion through the cell
membrane in the cell-based functional assay. We

3634
G. F. Filzen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3630–3635
Table 2. Activity of indole based compounds
O
R
OH
N
Y
X
16a-g
a,c
IC50 (nM)
b,c
EC50 PPARd
(nM)
Compound
X
R
Y = lipophilic tail
Fold selectivity: hPPARd
vs. hPPARa
hPPARd
hPPARc
hPPARa
a
b
c
d
e
f
O
O
O
O
O
O
S
H
A
6.8
31
550
835
>10,000
>10,000
NT
81
27
3
223
H
5-MeO
C
1390
225
A
41
121
H
H
H
H
o-D
m-D
p-D
A
734
13
>10,000
179
562
>10,000
>10,000
>10,000
>10,000
>14
14
6
3080
254
91
2.8
2810
4.0
g
172
61
NT, not tested.
a Concentration that inhibits 50% of the interaction between the PPAR LBD and the radiolabeled ligand.
b Concentration of test compound which produced 50% of the maximal reporter activity.
c The results are based on at least two experiments, each dose done in triplicate (SD = 10%).
8. Newman, M. S.; Karnes, H. A. J. Org. Chem. 1966, 81,
3980.
believed that the compounds were being sequestered
within the cell membrane so that the observed functional
activity was not always truly reflective of the com-
pound’s inherent potential. Nonetheless, the functional
assay confirmed that all the compounds were agonist.
9. PPAR receptor binding assay: The human PPARd and
PPARa scintillation proximity assay (SPA) was used to
measure the affinity of ligands for the respective human
PPAR receptor. The hPPARd LBD encoding amino acids
145–441 (GenBank Accession No. NM_006238) and
hPPARa LBD encoding amino acids 196–468 (GenBank
Accession No. L02932) were used. Volumes of 99 lL of
buffer (50 mM Tris, 10 mM Na-molybdate, 1 mM EDTA,
and 10% glycerol, pH 7.6) containing 50 nM of radiola-
beled ligand (3H-2-(4-(3-(4-acetyl-3-hydroxy-2 propyl-
phenoxy)propoxy)phenoxy)acetic acid (34 Ci/mmol) for
PPARd and 3H-2-(4-(2-(3-(2,4-difluorophenyl)-1-hepty-
lureido)ethyl)phenoxy)-2-methylbutanoic acid (86 Ci/
mmol) for PPARa), 0.2 mg anti-rabbit beads (Amersham,
RPN140), 0.24 lg rabbit anti-GST (Molecular Probes
Inc., A5800), and 0.2 lg purified GST/PPARhLBD were
placed into the wells of Corning 96-well tissue culture
plates. One microliter of dimethylsulfoxide (DMSO) or
1 lL of DMSO containing a test compound at a concen-
tration sufficient to give a final assay concentration of
between 1 nM and 100 lM added into each well. After
incubation with shaking at room temperature for 30 min,
radioactivity bound to the PPAR LBD-GST fusion
protein/anti-GST/SPA antibody-binding bead complex
was assessed using a Wallac MicroBeta plate reader. The
potency of interaction of a compound with the respective
PPAR LBD was determined as the concentration that
inhibits 50% of the interaction between the respective
PPAR LBD and the radiolabeled ligand.
Starting with a prototypical PPAR ligand, applying a
priori knowledge of the factors affecting subtype speci-
ficity, and a search of our proprietary chemical library,
we discovered three new templates, that when modified
had varying degrees of affinity and potency for PPARd.
These compounds represent a new class of chemical
tools that will be used to elucidate the complete biolog-
ical function of PPARd.
References and notes
1. Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B.
R. J. Med. Chem. 2000, 43, 527.
2. (a) Bedu, E.; Wahli, W.; Desvergne, B. Expert Opin. Ther.
Targets 2005, 9, 861; (b) Wolf, G. Nut. Rev. 2003, 61, 387;
(c) Wang, Y.-X.; Lee, C.-H.; Tiep, S.; Yu, R. T.; Ham, J.;
Kang, H.; Evans, R. M. Cell 2003, 113, 159; (d)
Takahashi, S.; Tanaka, T.; Kodama, T.; Sakai, J. Pharm.
Res. 2006, 53, 501.
3. Oliver, W. R.; Shenk, J. L.; Snaith, M. R.; Russell, C. S.;
Plunket, K. D.; Bodkin, N. L.; Lewis, M. C.; Winegar, D.
A.; Sznaidman, M. L.; Lambert, M. H.; Xu, H. E.;
Sternbach, D. D.; Kliewer, S. A.; Hansen, B. C.; Willson,
T. M. Proc. Natl. Acad. Sci .U.S.A. 1999, 98, 5306.
4. Zhefeng, C.; Feng, J.; Guo, Y.; Li, P.; Shen, Z.; Chu, F.;
Guo, Z. J. Biol. Chem. 1999, 274, 6718.
5. Xu, H. E.; Lambert, M. H.; Montana, V. G.; Plunket, K.
D.; Moore, L. B.; Collins, J. L.; Oplinger, J. A.; Kliewer,
S. A.; Gampe, R. T., Jr.; McKee, D. D.; Moore, J. T.;
Willson, T. M. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,
13919.
6. Dulenko, V. I.; Golyak, V. M.; Gubar, V. I.; Alekseev, N.
N. Chem. Heterocycl. Comp. 1979, 15, 815.
7. Auerbach, B. J; Bratton, L. D.; Filzen, G. F.; Geyer, A. G.;
Trivedi, B. K.;Unangst, P. C. USPatent 6,875,780 B2, 2005.
The human PPARc scintillation proximity assay (SPA) was
used to measure the affinity of ligands for the human PPARc
receptor. The hPPARc LBD encoding amino acids 206–477
(GenBankAccessionNo. NM_138712.1)wasused. Volumes
of 168 lL of buffer (1· PBS, 12 mM b-mercapto ethanol,
0.002% Tween 20, and 9% glycerol, pH 7.6) containing with
40 nM
3H-5-(4-(3-(5-methyl-2-phenyloxazol-4-yl)propa-
noyl)benzyl)thiazolidine-2,4-dione (9.57 Ci/mmol), 0.3 mg
polylysine-coated yttrium silicate beads (Amersham,
RPNQ0010), and 10 nM purified His-tagged human
PPARc LBD were placed into the wells of a 96-well white
assay plate (Corning 3604). Two microliters of dimethyl-
sulfoxide (DMSO) or 2 lL of DMSO containing a test

G. F. Filzen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3630–3635
3635
compound at a concentration sufficient to give a final assay
concentration binding curve between 1 nM and 100 lM was
added into each well. After incubation with shaking at room
temperature for 2 h, radioactivity bound to the PPARc
LBD-HIS fusion protein/yttrium bead complex was
assessed using a Wallac MicroBeta plate reader. The
potency of interaction of a compound with the PPARc
LBD was determined as the concentration that inhibits 50%
of the interaction between the PPARc LBD and the
radiolabeled ligand.
mixture was stirred at room temperature for 2 h, then
filtered through Celiteꢂ. The filtrate was concentrated and
purified using normal phase chromatography.
(6-methoxybenzo[b]thiophen-3-yl)acetic acid methyl ester(3).
Compound 2 (2.54 g, 0.01 mol) was added dropwise to
25 ml of methanesulfonic acid at room temperature, and the
solution was stirred at the same temperature for 15 min,
then the reaction mixture was added to 250 ml of ice-water.
The aqueous mixture was extracted with ethyl acetate. The
organic phase was washed with brine, sodium bicarbonate,
dried over sodium sulfate, and concentrated.
10. PPARd chimeric receptor assay (functional assay): Tran-
sient transfection assay using the HepG2 hepatoma cell line:
The GAL4 hPPARd LBD, chimeric receptor expression
constructs containing the ligand binding domain for the
human PPARd LBD (encoding amino acids 145–441,
GenBank Accession No. NM_006238), was used to
cotransfect cells with GAL4-Luciferase reporter plasmid
(p5Eb-Luc) and b-Gal plasmid. Briefly, HepG2 cells were
seeded in a 100-mm cell culture dish containing 10 ml
DMEM plus 10% serum. Transfection mix was prepared
by combining 15 lg GAL4-Luc plasmid with 15 lg of
GAL4-hPPARd LBD. b-Gal plasmid (1.5 lg) was also
added to each the as a control. LipofectAMINE 2000
reagent was used as suggested by the manufacturer
(Invitrogen, Carlsbad, CA). For each well, 2.4 ml trans-
fection mix was added and incubated at 37 ꢁC overnight.
The next day, transfected HepG2 cells were reseeded to a
96-well cell culture plate at the density of 3000 cells per
well and compounds were subsequently added to each
well. After 16 h incubation, cells were then harvested in a
lysis buffer (Promega, Madison, Wisconsin) and luciferase
activity was determined using a luminometer. Luciferase
activity was then normalized with b-Gal activity.
(6-hydroxy-benzo[b]thiophen-3-yl)acetic acid methyl ester(6).
To a stirred solution of 3 (2.20 g, 9.32 mmol) in 50 ml of
dichloromethane at À78 ꢁC was added dropwise a solution
of boron tribromide (11.68 g, 46.6 mmol) in 50 ml of
dichloromethane. After the completion of the addition of
boron tribromide, the reaction mixture was maintained at
À78 ꢁC for 1 h, then allowed to reach room temperature and
stirred at the same temperature overnight. The mixture was
cooled to 0 ꢁC, carefully quenched with 100 ml of water,
extracted with ethyl acetate, washed with brine, dried over
sodium sulfate, concentrated, and purified using normal
phase chromatography to afford 6.
Methyl 2-(6-((4-methyl-2-(4-(trifluoromethyl) phenyl)thia-
zol-5-yl)methoxy)benzo[b]thiophen-3-yl)acetate 9a.Amixture
of 6 (0.75 g, 3.4 mmol), 5-(chloromethyl)-4-methyl-2-(4-
(trifluoromethyl) phenyl)thiazole (1.0 g, 3.5 mmol), and
cesium carbonate (1.7 g, 5.1 mmol) in 10 ml acetonitrile was
stirred at 60 ꢁC for 4 h. The reaction mixture was then
concentrated in vacuo, the residue diluted with ether and
filtered through a Celiteꢂ/SiO2 sandwich, eluting with ether.
The filtrate was collected and concentrated in vacuo to give
9a pure enough for subsequent use.
11. Typical procedure for preparing thiazolyl benzothiophene
acetic acid derivatives:4-(3-Methoxy-phenylsulfanyl)-3-
oxo-butyric acid methyl ester (2). A solution of methyl
2-chloroacetoacetate (15.0 g, 0.10 mol) in 20 ml of aceto-
nitrile was added dropwise to a mixture of 3-methoxythi-
ophenol (14.0 g, 0.10 mol) and cesium carbonate (65.2 g,
0.20 mol) in 400 ml of acetonitrile over 30 min. The
2-(6-((4-methyl-2-(4-(trifluoromethyl)phenyl)thiazol-5- yl)-
methoxy)benzo[b]thiophen-3-yl)acetic acid (10a). A mixture
of 9a, in 40 ml THF/2 ml H2O, was stirred at rt for 2 h. The
reaction mixture was concentrated in vacuo, acidified to pH 1
with concd HCl, extracted with EtOAc, dried (Na2SO4), and
concentrated in vacuo. Recrystalization from chloroform/
hexanes afforded 10a.