Minor structural modifications convert a selective PPARα agonist into a potent, highly selective PPARδ agonist

Article abstract of DOI:10.1016/j.bmcl.2005.06.023

We report the solid-phase synthesis and pharmacological evaluation of a new series of small-molecule agonists of the human peroxisome proliferator- activated receptor δ (PPARδ) based on a lead structure from our PPARα program. Compound 33 showed good pharmacokinetics.

Full text of DOI:10.1016/j.bmcl.2005.06.023

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4619–4623
Minor structural modifications convert a selective PPARa
agonist into a potent, highly selective PPARd agonist
Stefan Weigand,* Hilmar Bischoff, Elke Dittrich-Wengenroth, Heike Heckroth,
Dieter Lang, Andrea Vaupel and Michael Woltering
BAYER Health Care AG, Pharma Research, D-42096 Wuppertal, Germany
Received 22 April 2005; revised 6 June 2005; accepted 7 June 2005
Available online 22 August 2005
Abstract—We report the solid-phase synthesis and pharmacological evaluation of a new series of small-molecule agonists of the
human peroxisome proliferator-activated receptor d (PPARd) based on a lead structure from our PPARa program. Compound
33 showed good pharmacokinetics.
Ó 2005 Elsevier Ltd. All rights reserved.
The peroxisome proliferator-activated receptors
(PPARs) belong to the nuclear hormone receptor super-
family and consist of the three members: PPARa,
PPARc, and PPARd. Many of the PPAR-regulated
genes are involved in fatty-acid metabolism and play a
central role in regulating the storage and catabolism of
dietary fats. Consequently, subtype selective PPAR ago-
nists are regarded as promising leads for novel drugs
used for the treatment of metabolic syndrome, diabetes,
obesity, dyslipidemia, and atherosclerosis.1,2 Unlike for
PPARa (fibrates) and PPARc (glitazones), there are
no known drugs that have been identified as working
through PPARd.3 However, oral application of 3 mg/
kg GW501516 (1, Fig. 1),4 the first PPARd-selective
agonist in clinical trials, is reported to increase HDLc
by 80%, while reducing LDL-cholesterol by 29% in a
primate model of dyslipidemia.5
Figure 1. PPARd-selective agonist GW501516.
Another series of potent PPAR agonists with up to 25-
fold selectivity for PPARd over PPARa and PPARc,
respectively, has been identified recently, starting from
a phenylacetic acid lead.6
Figure 2. Lead structures from in-house PPARa agonist program.
As part of our ongoing research in the identification of
novel nuclear receptor ligands, we report the identifica-
tion of highly selective PPARd agonists7 through solid-
phase parallel synthesis. The general lead structure 2,
specified by example 3, from our in-house PPARa
program served as the starting point for optimization
towards d-selectivity (Fig. 2).8
Keyword: PPAR delta agonists.
*
The general synthesis leading to compounds 17–35 is
depicted in Scheme 1. First, the corresponding a-bromo-
acetic acid derivatives 4 were attached to Wang resin
(p-benzyloxybenzylalcoholresin). Resinbound a-bromo-
Corresponding author. Tel.: +491749878566; fax: +49 8856 60 79
4749; e-mail: stefan.weigand@roche.com
Present address: Novartis Pharma AG, Oncology Research CH-4002
Basel, Switzerland.
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.06.023

4620
S. Weigand et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4619–4623
acetic acid ester 5 was coupled with 4-hydroxy benzalde-
hydes in a Williamson ether synthesis. Reductive amina-
tion lead to the secondary amine 7, which upon
treatment with trimethylsilyl protected a-bromoacetic
acid, followed by fluoride-induced ester cleavage, affor-
ded the intermediate 8. Amide coupling of 8 with various
amines and subsequent cleavage from the resin with
trifluoro acetic acid delivered the desired library of
phenoxyacetic acids of the general formula 9.
The combinatorial library was synthesized using a split-
and-pool IRORI Kan method.9 All new compounds
gave satisfactory spectral and analytical data including
1H NMR.10
In addition to the depicted solid-phase approach, sev-
eral structural features were addressed using solution
chemistry: compound 23 (see Table 1), obtained via
solid-phase chemistry according to Scheme 1, was
esterified with isobutylene in acidic media.11 Subse-
quent reduction of the amide bond with borane
dimethyl sulfide complex and acid catalyzed hydroly-
sis of the tert-butyl ester yielded the aminoethyl
substituted phenoxy acetic acid 11 (Scheme 2). The
harsh reaction conditions required for amide reduc-
tion were incompatible with our general solid-phase
approach.
Scheme 1. Synthesis of solid-phase parallel library: Reagents and
conditions: (a) Wang resin, diisopropylcarbodiimide, DMPA, Et3N,
CH2Cl2, rt; (b) Cs2CO3, dioxane/DMF (1:1), rt; (c) R4NH2, trimeth-
ylorthoformate/DMF (1:1), rt; Bu4NBH4, AcOH, DMF, rt; (d) Et3N,
dioxane, 60 °C; Bu4NF, dioxane, rt; (e) R5NH2, HATU, pyr/DMF
(2:1), rt; (f) CF3COOH, CH2Cl2, rt.
Table 1. Activitya of compounds 17–25 against human PPAR receptors
Compound
Structuresa 16
Human PPARd
EC50 (lM)14
Human PPARa
EC50 (lM)
Human PPARc
EC50 (lM)
R4
R3
H
17
18
19
20
21
22
23
24
25
0.08
0.04
0.1
0.25
0.17
P1.5b
P2.5
0.5
1.0
Me
Me
Me
Me
Me
Me
H
0.8
P 1.5
P2.5
P5
0.15
0.02
0.003
0.003
0.06
0.007
>10
P1
>10
0.3
P4
n.d.
0.015
Me
>10
a Cell-based transactivation assay.12
b P Means no saturation of dose–response curve at concentration.

S. Weigand et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4619–4623
4621
the factor 2, while retaining the same selectivity profile.
Variation of the substituent R4 at the secondary amide,
while keeping the rest of the molecule unchanged com-
pared to 18, led to compounds 19–25. Introduction of
lipophilic substituents with an n-alkyl (not shown) or
b-branched cycloalkyl chain (19, 20) did not lead to
higher d-selectivity. However, direct attachment of steri-
cally demanding cycloalkyl groups as in compounds 21–
23 resulted in a significant increase in potency at the d
receptor while decreasing the activity at a and c recep-
tors. Exchanging the methyl group at R3 for hydrogen
(24) proves the initial hypothesis that potency increases
with the introduction of substituents at that position.
Surprisingly, exchanging the cyclohexyl group in 23 by a
phenyl moiety led to the PPARd/c nonselective agonist
25, thus paving the way to compounds with a dual mode
of action for further mechanistic studies.
Scheme 2. Amide reduction: Reagents and conditions: (a) isobutylene,
H2SO4, CH2Cl2, rt, 65%; (b) BH3*SMe2, toluene, 110°C, 55%; (c)
CF3COOH, CH2Cl2, rt, 79%.
Similarly, compound 11 (Scheme 2) derived from 23 via
reduction of the amide bond shows considerable activity
on the d and c receptor (0.01 and 0.35 lM) while keep-
ing the selectivity profile against the a receptor
(>10 lM).
Cyclohexyl substituted phenoxy acetic acid 23 served as
the prototype for further optimization, as discussed in
Table 2.
Introduction of substituents in the 2- (26) and 3-position
(27) of the cyclohexyl ring reduced the activity at the d
receptor. In contrast, derivatization at the 4-position
as in ethoxy substituted cyclohexyl derivative 28, com-
bined high potency with >1000-fold selectivity against
the other PPAR receptors. Modification of the aryl
group R5 confirmed results from previous studies:8 2,4-
disubstituted anilines are superior to other substitution
patterns. In addition, introduction of polar substituents
on the aryl ring as in 29 are not tolerated.
Variation of R1–R3 on the phenoxyacetic acid head
group while keeping R4 (cyclohexyl) and R5 (2,4-
dichlorophenyl) constant (30–35) clearly shows that in
order to achieve high d-selectivity, it is mandatory to
introduce a substituent ortho to the phenol while reduc-
ing the steric bulk adjacent to the acid functionality.
Thus, unsubstituted acetic acid derivative 33 is a full
agonist at the d receptor with an EC50 of 3 nM with
virtually no effect on the a and c receptors up to greater
than 100 lM.
Scheme 3. Synthesis of glycyl amino acetic acid 15: Reagents and
conditions: (a) Et3N, CH2Cl2, rt; (b) 2,4-dichloro aniline, EtiPr2N,
DMF, 80 °C; (c) CF3COOH, CH2Cl2, rt, 55% over all steps.
The importance of the position of the carbonyl group
was further elucidated: glycylamino acetic acid 15 was
obtained by amid formation between benzylic amine
13—an intermediate of our general solid-phase ap-
proach—and bromoacetyl bromide (12), followed by
the introduction of 2,4-dichloro aniline under basic
conditions and final release of the free acid from the
resin (Scheme 3).
Inverted introduction of the bromoacetyl bromide
building block, as exemplified by compound 15 (Scheme
3), leads to a complete loss in activity on all the three
receptor subtypes (>1 lM).
Table 1 shows the EC50 for compounds 17–25 in the
transient transactivation assay against the three human
PPAR subtypes: d, a, and c, respectively.12,13 Com-
pound 17 has already been synthesized in the course of
our PPARa program.8 Due to its reversed selectivity
profile compared to the PPARa agonist 3, we chose 17
as the starting point for our optimization program to-
wards d-selectivity. Simple introduction of an ortho-
methyl substituent, as in 18, increased the potency by
The PPAR subtypes show significant sequence variation
in the residues that line the ligand-binding domain
(LBD). However, several conserved hydrogen-bonding
interactions of PPAR agonists with key amino acids of
the LBD have been determined across all subtypes.15
Despite the availability of crystal structures of all three
PPAR LBDs with and without ligand,15,16 it can only

4622
S. Weigand et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4619–4623
Table 2. Further optimization of 23 by variation of R1–R5
Compound
Structures 9
Human PPARd
EC50 (lM)14
Human PPARa
EC50 (lM)
Human PPARc
EC50 (lM)
R5
R4
R3
R1
R2
26
27
Me
Me
Me
Me
0.4
0.1
>10
P4
P4
Me
Me
P4
28
Me
Me
Me
0.005
P5
P5
29
30
31
32
33
34
35
Me
OMe
OMea
Me
Me
H
Me
Me
1.5
0.2
0.5
Me
Me
0.03
0.03
0.01
0.003
0.05
0.015
>10
P1
1
>10
P2
Me
Me
Meb
H
>10
>100
>10
>10
H
H
>100
>10
>10
H
H
Cl
H
H
a meta-OMe instead of ortho-OMe.
b Racemic mixture.
be speculated about the nature of the selectivity of our
novel PPARd ligands.
Starting from a highly selective PPARa agonist, 3, we
were able to identify a series of potent and selective
PPARd agonists (e.g. 33) by using combinatorial chem-
istry on solid support. These phenoxyacetic acid deriva-
tives will be valuable tools for further elucidation of the
biological role of the PPARd receptor as well as new
therapeutic agents for the treatment of metabolic diseas-
es like dyslipidemia, obesity, and atherosclerosis.
Pharmacokinetics of 33 were evaluated in rats (n = 2)
with an oral dose of 5 mg/kg. The maximum plasma
concentration was found to be 1.24 lM, the half life
being 1.3 h. Incubation of 33 with rat or human liver
microsomes in vitro revealed that metabolism of the par-
ent compound occurs mainly by hydroxylation of the
cyclohexyl moiety. Consequently, introduction of polar
or sterically demanding groups in this position should
further improve half-life of novel congeners. Investiga-
tions in this direction as well as in-depth pharmacologi-
cal studies are currently ongoing in our group.
Acknowledgments
We thank K.-D. Bremm, H. Haning, and H. Wild for
helpful discussions.

S. Weigand et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4619–4623
4623
9. Nicolaou, K. C.; Xiao, X.-P.; Parandoosh, Z.; Senyei, A.;
Nova, M. P. Angew. Chem. 1995, 107, 2476; Angew. Chem.
Int. Ed. Eng. 1995, 34, 2289.
References and notes
1. Linsel-Nitscke, P.; Tall, A. R. Nature Rev. Drug Discov.
2005, 4, 193.
2. Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B.
R. J. Med. Chem. 2000, 43, 527.
3. Willson, T. M.; Wahli, W. Curr. Opin. Chem. Biol. 1997, 1,
235.
10. e.g. 33: 1H NMR (300 MHz, CDCl3): d = 1.10–1.40 (m,
3H), 1.55 (m, 2H), 1.70 (m, 1H), 1.93 (m, 2H), 2.14 (m,
2H), 2.16 (s, 3H), 3.41 (m, 1H), 4.09 (s, 2H), 4.21 (s, 2H),
4.62 (s, 2H), 6.57 (d, J = 8.3 Hz, 1H), 7.03 (dd, J = 1.9 and
8.2 Hz, 1H), 7.10 (d, J = 1.8 Hz, 1H), 7.17 (dd, J = 2.3 and
8.7 Hz, 1H), 7.31 (d, J = 2.3 Hz, 1H), 7.63 (d, J = 8.7 Hz,
1H), 9.57 (s, 1H).
4. Sznaidman, M. L.; Haffner, C. D.; Maloney, P. R.;
Fivush, A.; Chao, E.; Goreham, D.; Sierra, M. L.;
LeGrumelec, C.; Xu, H. E.; Montana, V. G.; Lambert,
M. H.; Willson, T. M.; Oliver, W. R., Jr.; Sternbach, D. D.
Bioorg. Med. Chem. Lett. 2003, 13, 1517.
5. Oliver, W. R., Jr.; 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. 2001, 98,
5306.
6. Adams, A. D.; Yuen, W.; Hu, Z.; Santini, C.; Jones, A. B.;
MacNaul, K. L.; Berger, J. P.; Doebber, T. W.; Moller, D.
E. Bioorg. Med. Chem. Lett. 2003, 13, 931.
7. Bischoff, H.; Dittrich-Wengenroth, E.; Heckroth, H.;
Vaupel, A.; Woltering, M.; Weigand, S. (Bayer Health
Care AG), WO 03/035603 2003; Chem. Abstr. 2003, 138,
353734.
11. Liu, K. G.; Lambert, M. H.; Leesnitzer, L. M.; Oliver, W.,
Jr.; Ott, R. J.; Plunket, K. D.; Stuart, L. W.; Brown, P. J.;
Willson, T. M.; Sternbach, D. D. Bioorg. Med. Chem.
Lett. 2001, 11, 2959.
12. The different PPAR subtypes used in the assay are fusion
proteins containing the ligand binding domain for PPARd
(aa 139–442), PPARa (aa 167–468) and PPARc (aa 203–
506), respectively, fused to the GAL4 DNA-binding
domain (aa 1–147). The assay was performed as stated
in literature.7,8
13. The results are based on two to three independent
experiments each dose done in triplicate (SD = 20%).
14. All compounds (except 29) are full agonists on PPARd
with >70% efficacy, e.g. 33: PPARd 2 nM, 75% efficacy;
PPARa /c >10 lM, 30% efficacy.
15. Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B.
R. J. Med. Chem. 2004, 43, 527.
16. Xu, H. E.; Lambert, M. H.; Montana, V. G.; Parks, D. J.;
Blanchard, S. G.; Brown, P. J.; Sternbach, D. D.; Lehman,
J. M.; Wisely, G. B.; Willson, T. M.; Kliewer, S. A.;
Milburn, M. V. Mol. Cell 1999, 3, 397.
8. Urbahns, K.; Woltering, M.; Nikolic, S.; Pernerstorfer,
J.; Hinzen, B.; Dittrich-Wengenroth, E.; Bischoff, H.;
Hirth-Dietrich, C.; Lustig, K. (Bayer Health Care
AG), WO 02/028821 2002; Chem. Abstr. 2002, 136,
294653.