Benzoyl 2-methyl indoles as selective PPARγ modulators

Article abstract of DOI:10.1016/j.bmcl.2004.10.068

A series of selective PPARγ modulators (SPPARγMs) and their development from hPPARγ active screening leads 1 and 2 is described. SPPARγM 24 displayed robust anti-diabetic activity with an improved therapeutic window in comparison to a PPARγ full agonist in a rodent efficacy model. Routine screening for human PPAR ligands yielded compounds 1 and 2, both of which were sub-micromolar hPPARγ agonists. Synthetic modifications of these leads led to a series of potent substituted 3-benzyl-2-methyl indoles, a subset of which were noted to be selective PPARγ modulators (SPPARγMs). SPPARγM 24 displayed robust anti-diabetic activity with an improved therapeutic window in comparison to a PPARγ full agonist in a rodent efficacy model.

Full text of DOI:10.1016/j.bmcl.2004.10.068

Bioorganic & Medicinal Chemistry Letters 15 (2005) 357–362
Benzoyl 2-methyl indoles as selective PPARc modulators
John J. Acton, III,a,* Regina M. Black,a A. Brian Jones,a, David E. Moller,b
Lawrence Colwell,a Thomas W. Doebber,b Karen L. MacNaul,b
Joel Bergerb and Harold B. Wooda
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Metabolic Disorders, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
Received 27 September 2004; revised 21 October 2004; accepted 21 October 2004
Available online 19 November 2004
Abstract—Routine screening for human PPAR ligands yielded compounds 1 and 2, both of which were sub-micromolar hPPARc
agonists. Synthetic modifications of these leads led to a series of potent substituted 3-benzyl-2-methyl indoles, a subset of which were
noted to be selective PPARc modulators (SPPARcMs). SPPARcM 24 displayed robust anti-diabetic activity with an improved thera-
peutic window in comparison to a PPARc full agonist in a rodent efficacy model.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Non-insulin dependent or Type 2 diabetes mellitus
(T2DM) affects a significant number of people in the
industrialized world, and is expected to be a growing
medical concern as the post-World War II generation
ages. Onset of T2DM most often occurs in middle age,
but younger people are more frequently being diag-
nosed. The sedentary lifestyles in developed countries,
combined with high caloric intake, are contributing fac-
tors in T2DM. Type 2 diabetic patients suffer a variety
of symptoms, including hyperglycemia, hyperlipidemia,
atherosclerosis, and obesity. Insulin resistance combined
with pancreatic b-cell dysfunction play a central role in
the etiology of type 2 diabetes; with insulin resistance
usually occurring first, followed by insulin insufficiency.1
Peroxisome proliferator-activated receptor gamma
(PPARc) is a transcription factor known to mediate adi-
pocyte differentiation.2 Synthetic agonists of PPARc
including pioglitazone3 and rosiglitazone4 have been
shown clinically to provide beneficial decreases in ele-
vated plasma glucose levels in T2DM patients.
preclinical rodent models. These adverse effects preclude
pioglitazone and rosiglitazone frombeing used as front-
line therapy in T2DM, suggesting that pursuit of a safer
second generation human PPARc (hPPARc) agonist is
desirable.
Selective PPARc modulators (SPPARcMs), which are
also known as hPPARc partial agonists, have recently
attracted much interest. Selective modulation of the
hPPARc nuclear receptor theoretically could provide
significant anti-diabetic activity while concurrently
reducing or eliminating PPARc-mediated side-effects
such as increased adipogenesis.5,6 Berger recently dis-
closed the SPPARcM nTZDpa,7 which in comparison
with PPARc full agonists demonstrates attenuation of
body weight gain, adipose depot size, cardiac weight
gain, and adipocyte gene expression in a rodent model
of T2DM. Equally important was the marked reduction
of hyperglycemia and hyperinsulinemia seen in these
same models. Others have reported compounds
(Fmoc-Leu8 and YM4409) that modulate PPARc with
improved in vivo profiles relative to full agonists; while
GW007210 and L-76440611 have been shown to be
hPPARc partial agonists in vitro. The antagonist
LG100641 reportedly blocks TZD-induced PPARc activ-
ation and adipocyte differentiation in vitro.12
Glitazones, while efficacious, exhibit significant liabili-
ties; they are associated with edema and weight gain in
man. In addition, cardiac hypertrophy is observed in
Keywords: PPAR; Indole; Selective PPAR modulator.
*
Sample collection screening directed at identifying new
and structurally diverse leads active on hPPARc pro-
vided the 2-methyl indoles 1 and 2 as selective, sub-
micromolar receptor agonists. The indole core is present
Corresponding author. Tel.: +1 732 594 4722; fax: +1 732 594
9556; e-mail: john_acton@merck.com
Present address: Merck Research Laboratories, Terlings Park, United
Kingdom.
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.10.068

358
J. J. Acton, III et al. / Bioorg. Med. Chem. Lett. 15 (2005) 357–362
in a large number of pharmacologically active com-
pounds. In fact, indomethacin,13 nTZDpa,7 and 2,3-
disubstituted indole 5-acetic acids14 are all indoles that
have been reported to be PPARc agonists. Modification
of the 2-methyl indole leads and their effects on in vitro
measures of hPPARc agonismand in vivo efficacy, toxi-
city, and pharmacokinetics in rodent models are dis-
closed below.
R1
R1
X
CO2H
CO2H
O
a
.HCl
+
X
CO2H
NH2
N
H
N
H
R1
R1
X
X
CO2allyl
d
b,c
N
N
R2
O
R2
O
Modifications were made at four individual sites of the
indole core of compound 3 (Fig. 1). Scrutiny of the
screening actives and a limited synthetic survey estab-
lished that the methyl group is optimal at C-2. Modifica-
tion of the side chain at C-3 as well as variations at R1
(C-4 to C-7) and COR2 (N-1) were undertaken using
known synthetic procedures. The indoles were prepared
by treating a phenylhydrazine and a methyl ketone with
phosphoric acid in refluxing toluene (the Fischer indole
synthesis). Initially, the choices for the C-3 side chain
were limited to various aliphatic acids. The methyl ke-
tones were either commercially available or prepared
in short sequences fromavailable precursors. Following
acylation of the indole, catalytic cleavage of the allyl
ester by the method of McCombie provided the final
products.15,16 A general preparative sequence is shown
below (Scheme 1).
Scheme 1. Reagents and conditions: (a) H3PO4, toluene, reflux; (b)
Cs2CO3, allyl iodide, DMF; (c) NaHMDS, R2COCl, THF; (d)
Pd(PPh3)4, dimedone, DIEA, DMF.
can be shorter. A limited number of benzoyl group sub-
stituents (COR2) retain activity; only the 2-napthoyl and
4-methoxybenzoyl groups were notable improvements
over the 4-chlorobenzoyl screening leads. Substitution
of 4-chlorobenzyl for 4-chlorobenzoyl at N-1 led to
>50-fold reduction in hPPARc binding; transactivation
activity was not measured. Combining the most favor-
able X, R1, and COR2 substituents leads to very potent
hPPARc full agonists (IC50Õs of <40nM in the binding
assay). Unsurprisingly, these long chain alkyl com-
pounds suffer frompoor pharmacokinetic profiles and
in general fared poorly in rodent efficacy models.
All new compounds were evaluated as ligands for the
three human PPAR isoforms (c, d, and a) in competi-
tion binding assays.7,17 With rare exception, active com-
pounds were selective hPPARc ligands. Those of
adequate binding potency were then assessed for agonist
activity in a COS-1 cell-based transactivation assay (TA)
using transiently co-transfected pcDNA3-PPAR/GAL4
chimer and pUAS(5X)-tk-luc reporter vectors.7
A series of substituted 3-benzylic indoles18 were de-
signed that preserved the lipophilic nature of the acid-
bearing side chain while maintaining the carboxylate
in a similar position. Introduction of a benzylic moiety
could possibly improve pharmacokinetics relative to
the metabolically soft long-chain aliphatics. Preparation
of the first of these 3-benzylic indoles involved the reac-
tion of the benzoic acid intermediate 4 with the indo-
methacin synthetic intermediate 5 to give indole 6.
Intermediate 4 is easily accessible; the route shown in
Scheme 2 below can be used to make 4 or any number
A wide array of substituents are tolerated at the indole
5-position (R1); among those leading to potent analogs
are methoxy, trifluoromethoxy, isopropyl, fluoro, pro-
pyl, phenoxy, and N,N-dimethylcarbamate. Substitution
at the 4, 6, or 7 positions of the indole was poorly toler-
ated. Variation of the chain length at C-3 indicates that
five to six methylenes are optimal for a straight chain,
whereas the bulkier gem-dimethyl substituted chains
HO2C
O
O
O
H3PO4
NH2
O
N
O
+
N
toluene
reflux
CO2H
CO2H
S
O
H3CO
H3CO
CO2H
Cl
Cl
4
5
6
N
N
R1
O
H
O
O
a,b,c
d
X
Cl
Cl
N
1
2
X
O
X
R2
various alkyl
replacements
X=COOH,
OH,
OCH2COOH
X=COOH,
X=COOH,
OCR1R2COOH
OCH2COOH
explored sub-
X
CO2H
R1
for thioalkyl group
OCR1R2COOR
OCH2COOH
stitution at 4,5,6
and 7 group.
Explored H, Et, Ph
5-position optimal
with a variety of
substituents.
N
Scheme 2. Reagents and conditions: (a) Ph3P@CHCOCH3/THF/
reflux; (b) H2/10% Pd–C/EtOAc; (c) (if needed) BrCR1R2COOR/
K2CO3/DMF; (d) Fischer indole reaction using acylhydrazines like 5
or a phenylhydrazine (ArNHNH2) followed by acylation and ester
hydrolysis.
COR2
explored alkyl,
acyl, benzoyl,
carbamoyl.
3
Figure 1. Screening leads and initial synthetic survey.

J. J. Acton, III et al. / Bioorg. Med. Chem. Lett. 15 (2005) 357–362
Table 1. In vitro activity of 3-benzyl indoles 6–17
359
H3CO
N
X
O
Cl
Position/X substituent
hPPARc binding (SPA) IC50 (lM)a
PPARc TA EC50 (lM)c
PPARc TA % max @ 3lM
6
2; CO2H
3; CO2H
4; CO2H
6.28
0.208
4.69
NDd
0.28
ND
33%
ND
50%
25%
ND
97%
42%
ND
86%
59%
ND
7
8
ND
9
2; OCH2CO2H
3; OCH2CO2H
4; OCH2CO2H
2; OCH(Me)CO2H
3; OCH(Me)CO2H
4; OCH(Me)CO2H
2; OC(Me)2CO2H
3; OC(Me)2CO2H
4; OC(Me)2CO2H
0.333
0.191
0.077
ND
10
11
12
13
14
15
16
17
0.099 0.024b
>10
0.084
0.014
0.019 0.005b
ND
0.049
2.14
0.079
0.045
0.914
0.012
0.019
ND
a Compounds 6–17 do not bind hPPARa. Compound 10 and 16 bind hPPARd modestly (1.25 and 4.22lM IC50Õs, respectively); all others not active
on hPPARd. All data SD 15% (n = 3).
b Average of two experiments. Data presented as mean SEM.
c The EC50 is reported; this is the compound concentration at which 50% of a given compoundsÕ intrinsic maximal response has been reached. The
standard reference full agonist is rosiglitazone (EC50 = 0.023lM; 100% maximum activation at 3.0lM). Activation levels can be compared between
compounds; those that reach a nearly equivalent maximal activation to rosiglitazone are considered full agonists, while those reaching 20–60% of
rosiglitazonesÕ maximal activation are deemed partial agonists. A graphical representation of this can be seen in Figure 2.
d Not done = ND.
of positional isomers or analogs. The ketones were di-
rectly converted to the desired N-acylindoles via Fischer
indole synthesis with 5. Alternatively, Fischer indole
synthesis with the ketone and a phenylhydrazine
(ArNHNH2) followed by N-acylation and ester hydro-
lysis provided the desired indole as outlined in Scheme 1.
The ortho-lactates and isobutyrates are potent full agon-
ists of hPPARc, with all analogs except the (R)-lactate
exhibiting
binding
IC50Õs and
transactivation
EC50Õs 66nM. Data is shown in Table 2 below.
The meta-lactates and isobutyrates in the ÔhybridÕ series
were potent partial agonists of hPPARc; they activate
hPPARc in the COS1 cell line to 650% of the maximum
of the comparator full agonist rosiglitazone. Closer
The initial 3-benzyl indoles prepared were 2-, 3-, and 4-
benzoic acid and phenoxyacetic acid analogs (6–11,
Table 1). Generally, the benzoic acid analogs were weak
ligands, with only meta-analog 7 showing sub-micromo-
lar binding affinity. Improvement in potency was noted
with the phenoxyacetic acid analogs; the ortho (9) and
meta (10) analogs both exhibit sub-micromolar c-bind-
ing affinity. Compound 10 also bound hPPARd weakly.
In the TA assay analogs 7, 9, and 10 showed useful func-
tional activity. Further elaboration of the phenoxyacetic
acids led to a series of compounds substituted at the car-
bon a to the carboxylate. These lactates (12, 13) and
isobutyrates (15, 16) led to another 2- to 4-fold improve-
ment in binding affinity and transactivation potency.
The addition of the methyl groups ameliorated off-target
hPPARd activity; compound 16 improved hPPARc
specificity 10-fold versus compound 10.
Table 2. In vitro activity of o-benzyl indoles 18–22
R2
R1
CO2H
O
F3CO
N
O
H3CO
R1
R2
hPPARc
binding
(SPA)
IC50 (lM)a
PPARc TA PPARc TA %
EC50 (lM) max @ 3lM
A series of compounds was then prepared that incorpo-
rated all of the optimized substituents at R1, COR2 and
the benzylic side chain at C-3. These ÔhybridsÕ include R1
as trifluoromethoxy, COR2 as 4-methoxybenzoyl and
the benzylic side chains as ortho- and meta-lactates
and isobutyrates. The individual (R) and (S) isomers
of the lactates19 were prepared as well.
18 rac-Me ––
19
20 (S)-Me
21 rac-Et ––
22 Me Me
0.001
0.003
0.021
0.002
0.006
0.050
83%
47%
97%
67%
75%
H
(R)-Me 0.023
H
0.002
0.003
0.001
a Compounds 18–22 do not bind hPPARa or hPPARd. All data
SD 15% (n = 3). rac = racemic.

360
J. J. Acton, III et al. / Bioorg. Med. Chem. Lett. 15 (2005) 357–362
examination of the profiles of the earlier benzyl indoles
(Table 1) also shows that the meta-analogs (carboxylate,
phenoxyacetate, lactate, and isobutyrate) were hSP-
PARcMs. All of the analogs in this ÔhybridÕ series were
<5nM ligands of hPPARc; maximum activation of
these analogs ranged from24% to 42% at 3 lM com-
pound concentration with nanomolar EC50Õs in the TA
assay. The actual data is shown in Table 3 below; the
graph in Figure 2 compares compound 24 with rosiglit-
azone in the transactivation assay.
CO2H
R2
O
F3CO
X
R1
N
O
Ar
Figure 3. Structural survey of meta-indole lead class: n = 0–3, Ar = 2-
napthoyl, 4-Cl-benzoyl, 2,4-Cl-benzoyl, 2,4,6-Cl-benzoyl, X = CH2,
C(CH3)2, C(CH2CH2), O.
Expansion of the SAR was undertaken to correlate the
structural parameters to partial agonism and to identify
when SPPARcM-like behavior transitions to full agon-
ismor even antagonism. These alterations are illustrated
in Figure 3. N-(Di- and tri-chlorobenzoyl) substituents
revert to full agonismwhile N-(2-napthoyl) analogs
retain the partial agonist character of the 4-chloro-
and 4-methoxybenzoyl substituents. Variation of the
length of the benzylic methylene bridge was attempted
(n = 0–3). A partial agonist was prepared fromthe
n = 0 series (R1 = R2 = CH3), while the n = 2–3 analogs
are full agonists (R1 = CH3/R2 = H; (R)-lactates). For
n = 1, every compound prepared with the ÔhybridÕ core
structure was a hSPPARcM, indicating that there is a
fairly tight structural constraint regarding full versus
partial agonism. Also in the n = 1 class, substitution
on the benzylic methylene attenuated functional activity;
the cyclopropyl and dimethyl compounds were very
weak partial agonists. When Xn = C(CH3)2, potency
dropped off sharply relative to compound 24. However,
with Xn = O at the benzylic methylene position a hSP-
PARcM that is equipotent with 24 is obtained.
Table 3. In vitro activity of m-benzyl indoles 23–28
CO2H
O
Following the identification of multiple hSPPARcMs
with good intrinsic potency in vitro, it was necessary
to determine whether these compounds would retain
useful in vivo efficacy in diabetes models while exhibit-
ing an enhanced safety profile with respect to known
PPARc mechanism-based liabilities. Easily monitored
liabilities include weight gain, cardiac hypertrophy,
and brown adipose tissue (BAT) accumulation. Armed
with numerous SPPARcMs, in vivo testing in rodent
models was used to ascertain whether this objective
could be attained. The incorporation of the more meta-
bolically robust 3-benzylic indole side chain improved
pharmacokinetics in Sprague–Dawley (SD) rats versus
the 3-aliphatic compounds described above, validating
one of the hypotheses posited earlier. Db/db mice,20 a ge-
netic model of obese, insulin resistant T2DM, were then
used to evaluate the in vivo efficacy of analogs with ade-
quate pharmacokinetic profiles. While a number of
SPPARcMs were tested in pharmacokinetics and in
the db/db mouse model, only representative compound
24 is described in detail (Table 4). SPPARcM 24 effec-
tively reduces hyperglycemia in the db/db model at 3
and 10mg/kg/day for 11days (79% and 90% correction,
respectively). Rosiglitazone is effective as well, but
slightly less so under these test conditions (10mg/kg/
day, 60% glucose correction). Body weight gain meas-
ured in the db/db mouse amounted to only 4% at 3
and 10mg/kg/day of 24, while a 10% body weight gain
was noted for rosiglitazone at 10mg/kg/day. PPARc-
mediated toxicities were evaluated by dosing lean Spra-
gue–Dawley rats21,22 with high doses of the requisite
PPARc ligand for 14days, followed by analysis of serum
chemistries, necropsy, and organ weight determination.
There were two notable differences between compound
24 and rosiglitazone in this study. Compound 24
resulted in a 3.5% heart weight increase and a 45%
BAT weight increase at a plasma exposure 100 times
greater than the exposure needed for efficacy in db/db
F3CO
R2
R1
N
O
H3CO
R1
R2
hPPARc/d/a
PPARc TA PPARc TA
binding (SPA) EC50 (lM) % max @
IC50 (lM)
3lM
23 rac-Me ––
24
0.001/>15/3.90 0.002
(S)-Me 0.001/>15/>15 0.002
24%
21%
34%
36%
42%
24%
H
25 (R)-Me
26 rac-Et
H
––
0.003/>15/1.65 0.002
0.003/>15/1.64 0.003
0.004/>50/2.75 0.002
0.003/>15/2.77 0.006
27 rac-nPr ––
28 Me Me
All data SD 15% (n = 3). rac = racemic.
120
100
80
Rosiglitazone
EC50 = 23 nM(100%)
60
40
20
0
Compound 24
= 3 nM(20%)
10-9
10-7
10-6
10-8
[Compound] (M)
Figure 2. Human PPARc transactivation of rosiglitazone and com-
pound 24.

J. J. Acton, III et al. / Bioorg. Med. Chem. Lett. 15 (2005) 357–362
Table 4. In vivo data of Compound 24 versus rosiglitazone
361
db/db Mouse20
Normal Sprague–Dawley rats21
Glucose
correctiona
(dose, mg/kg)
Body
weight gain
(dose, mg/kg)
Plasma AUC of
parent drug
(dose, mg/kg)
Heart wt.
(dose, mg/kg)
Brown adipose
tissue wt.
(BAT) (dose, mg/kg)
Plasma AUC of
parent drug
(dose, mg/kg)
Vehicle
24
None
ND
None
8.7lMh (3)
1.13 0.06g
1.17 0.09g (30)
0.31 0.03
0.45 0.05 (30)
None
90 11% (10)b
79 15% (3)c
60 17%b (10)
4% (10)
4% (3)
10% (10)
20.2lMh (2)
955lMh (30)
1607lMh (150)
Rosiglitazone
317lMh (10)
1.41 0.07g (150)
1.37 0.08g (30)
0.88 0.15 (150)
a Glucose correction is calculated as the percent correction of nonfasting hyperglycemia present in vehicle control db/db mice relative to normal
glucose levels in control db/lean mice, all on day 11 of dosing. Data is presented as mean glucose correction SEM.
b Average of two experiments.
c Average of four experiments.
5. Cock, T.-A.; Houten, S. M.; Auwerx, J. EMBO Rep. 2004,
5, 142; Also: Miller, A. R.; Etger, G. J. Expert Opin.
Investig. Drugs 2003, 12, 1489.
mice, while rosiglitazone resulted in a 23% heart weight
increase and a 183% BAT weight increase at an exposure
only five times greater than the exposure needed for effi-
cacy in db/db mice. Based upon these findings, we con-
clude that SPPARcM 24 possesses important
pharmacological advantages relative to the PPARc full
agonist rosiglitazone in these models.
6. Gurnell, M.; Savage, D. B.; Chatterjee, V. K. K.;
OÕRahilly, S. J. Clin. Endocrinol. Metab. 2003, 88, 2412.
7. Berger, J. P.; Petro, A. E.; McNaul, K. L.; Kelly, L. J.;
Zhang, B. B.; Richards, K.; Elbrecht, A.; Johnson, B. A.;
Zhou, G.; Doebber, T. W.; Biswas, C.; Parikh, M.;
Sharma, N.; Tanen, M. R.; Thompson, G. M.; Ventre, J.;
Adams, A. D.; Mosley, R.; Surwit, R. S.; Moller, D. E.
Mol. Endocrinol. 2003, 17, 662.
8. Rocchi, S.; Picard, F.; Vamecq, J.; Gelman, L.; Potier, N.;
Zeyer, D.; Dubuquoy, L.; Bac, P.; Champy, M.-F.;
Plunket, K. D.; Leesnitzer, L. M.; Blanchard, S. G.;
Desreumaux, P.; Moras, D.; Renaud, J.-P.; Auwerx, J.
Mol. Cells 2001, 8, 743.
9. Shimaya, A.; Kurosaki, E.; Nakano, R.; Hirayama, M.;
Shibasaki, M.; Shikama, H. Metabolism 2000, 49, 411.
10. Oberfield, J. L.; Collins, J. L.; Holmes, C. P.; Goreham, D.
P.; Cooper, J. P.; Cobb, J. E.; Lenhard, J. M.; Hull-Ryde,
E. A.; Mohr, C. P.; Blanchard, S. G.; Parks, D. J.; Moore,
L. B.; Lehmann, J. M.; Plunket, K.; Miller, A. B.;
Milburn, M. V.; Kliewer, S. A.; Willson, T. M. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 6102.
During the course of these investigations, structural
modifications of full hPPARc agonists 1 and 2 led to
the identification of a series of potent partial agonists
or hSPPARcMs, typified by the meta-lactate substituted
3-benzylic indole 24. The series of compounds described
here feature improved pharmacokinetics as compared to
the original screening leads, while also demonstrating
excellent efficacy in rodent diabetes models. Further
evaluation demonstrated that separation of efficacy
fromhPPAR c mechanism-based liabilities can be
achieved in rodent models. These results indicate that
hSPPARcMs are attractive candidates for future
development.
11. Elbrecht, A.; Chen, Y.; Adams, A.; Berger, J.; Griffin, P.;
Klatt, T.; Zhang, B.; Menke, J.; Zhou, G.; Smith, R. G.;
Moller, D. E. J. Biol. Chem. 1999, 27, 7913.
Acknowledgements
12. Mukherjee, R.; Hoener, P. A.; Jow, L.; Bilakovics, J.;
Klausing, K.; Mais, D. E.; Faulkner, A.; Croston, G. E.;
Paterniti, J. R., Jr. Mol. Endocrinol. 2000, 14, 1425.
13. Lehmann, J. M.; Lenhard, J. M.; Oliver, B. B.; Ringold,
G. M.; Kliewer, S. A. J. Biol. Chem. 1997, 272, 3406.
14. Henke, B. R.; Adkison, K. K.; Blanchard, S. G.; Leesn-
itzer, L. M.; Mook, R. A., Jr.; Plunket, K. D.; Ray, J. A.;
Roberson, C.; Unwalla, R.; Willson, T. M. Bioorg. Med.
Chem. Lett. 1999, 3329.
We are indebted to Dr. Peter Meinke for critical reading
of this manuscript. We also wish to acknowledge the
skills and hard work of the many scientists who pro-
vided biological support including Margaret Wu, John
Ventre, Roger Meurer, NeelamSharma, Chhabi Biswas,
Raul Alvaro, and Zhesheng Chen.
15. Jeffrey, P. D.; McCombie, S. W. J. Org. Chem. 1982, 47,
587.
References and notes
16. Arnould, J. C.; Landier, F.; Pasquet, M. J. Tetrahedron
Lett. 1992, 33, 7133.
1. Reviews on PPARc and insulin resistance: Olefsky, J. M.;
Saltiel, A. R. Trends Endocrinol. Metab. 2000, 11, 362;
Berger, J.; Wagner, J. A. Diabetes Technol. Ther. 2002, 4,
163; Berger, J.; Moller, D. E. Annu. Rev. Med. 2002, 53,
409.
17. The experimental procedure used for the PPAR binding
experiment is in the following reference: Berger, J.;
Liebowitz, M. D.; Doebber, T. W.; Elbrecht, A.; Zhang,
B.; Zhou, G.; Biswas, C.; Cullinan, C. A.; Hayes, N. S.; Li,
Y.; Tanen, M.; Ventre, J.; Wu, M. S.; Berger, G. D.;
Mosley, R.; Marquis, R.; Santini, C.; Sahoo, S. P.;
Tolman, R. L.; Smith, R. G.; Moller, D. E. J. Biol. Chem.
1999, 274, 6718.
2. Tontonoz, P.; Hu, E.; Spiegelman, B. M. Cell 1994, 79,
1147.
3. Momose, Y.; Meguro, K.; Ikeda, H.; Hatanka, C.; Oi, S.;
Sodha, T. Chem. Pharm. Bull. 1991, 39, 1440.
4. Cantello, B. C. C.; Cawthorne, M. A.; Haigh, D.; Hindley,
R. M.; Smith, S. A.; Thurlby, P. J. Med. Chem. 1994, 37,
3977.
1
18. All compounds described in this paper gave consistent H
NMR and LC/MS data. Further detail describing
the synthetic sequences used, synthetic experimental

362
J. J. Acton, III et al. / Bioorg. Med. Chem. Lett. 15 (2005) 357–362
procedures, and spectral data can be found in: Acton J. J.;
21. Male Sprague–Dawley rats (n = 8 rats/group; ca. 200g
body weight) were provided ad lib access to water and
rodent chow (Purina #5008). Body weights were measured
prior to dosing and on days 1, 4, 6, 8, 11, and 14 of dosing.
The animals were dosed 1 · daily by oral gavage with the
compound indicated or with vehicle (0.5% methylcellulose,
10mL/kg). At 24h following the final dose the animals
were euthanatized by CO2 inhalation. Tissues of interest
were removed and weighed; serum was prepared from
blood samples and assayed for serum chemistry. Organ
weights are presented as an average standard deviation.
22. Pharmacology of rosiglitazone in rodents is described
here: Pickavance, L. C.; Tadayyon, M.; Widdowson, P. S.;
Buckingham, R. E.; Wilding, J. P. H. Br. J. Pharmacol.
1999, 128, 1570.
Black, R. M.; Jones, A. B.; Wood, H. B. U.S. Patent
6,525,083, 2001.
19. Stereochemistry of lactate isomers determined by infer-
ence. These compounds are made by the Mitsunobu
reaction of the requisite phenol and a lactic acid ester;
inversion of configuration is known to occur in this
reaction. Stereochemical purity (eeÕs) were determined for
a subset of the chiral compounds by injection onto a chiral
HPLC column and comparison with the racemate.
20. db/db Mice aged 11–13weeks were used in the db/db
assay otherwise consistent with the procedure described in
this reference: Berger, J.; Bailey, P.; Biswas, C.; Cullinan,
C. A.; Doebber, T. W.; Hayes, N. S.; Saperstein, R.; Smith,
R. G.; Liebowitz, M. D. Endocrinology 1996, 137, 4189.