Solution- and solid-phase synthesis of peptide-substituted thiazolidinediones as potential PPAR ligands

Article abstract of DOI:10.1016/S0960-894X(00)00440-6

Solution- and solid-phase methods for the preparation of peptide-substituted thiazolidinediones have been developed as an approach towards the preparation of a library of these compounds as potential ligands for the peroxisome proliferator-activated receptors (PPARs). (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00440-6

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2239±2242
Solution- and Solid-Phase Synthesis of Peptide-Substituted
Thiazolidinediones as Potential PPAR Ligands
Debra L. Mohler,a,b,* Gang Shena,b and Anthony K. Dotseb
aDepartment of Chemistry, Emory University, 1515 Pierce Drive, Atlanta, GA 30322, USA
bDepartment of Chemistry, West Virginia University, Morgantown, WV 26506, USA
Received 17 December 1999; accepted 29 June 2000
AbstractÐSolution- and solid-phase methods for the preparation of peptide-substituted thiazolidinediones have been developed as
an approach towards the preparation of a library of these compounds as potential ligands for the peroxisome proliferator-activated
receptors (PPARs). # 2000 Elsevier Science Ltd. All rights reserved.
Perturbation of the balance between energy intake and
expenditure in humans can result in a number of
unhealthy conditions including hypertension, type II
diabetes, and obesity. At the molecular level, the per-
oxisome proliferator activated receptors (PPARs) are
thought to play a role in each of these metabolic dis-
orders and thus have been the focus of much intense
investigation.1 As nuclear hormone receptors, the
PPAR isoforms (a, g, and b/d) di€erentially respond to
peroxisome proliferators (such as clo®brate),2,3 thiazol-
idinediones (such as BRL49653, 1, Fig. 1),4 and long-
chain fatty acids.5 7 The ®nding that the latter com-
pounds and their derivatives are agonists for PPARs
represents the novel concept of fatty acids behaving as
hormones controlling their own regulation,8 since the
genes regulated by PPARs include ones encoding
enzymes involved in fatty acid synthesis, storage, and
metabolism (including b- and o-oxidation, transport,
and intracellular binding).9
Interestingly, 1 exhibits a PPARg binding anity 10
times that of the thiazolidinedione MCC-555 (3), yet the
antidiabetic potency of 3 in reducing plasma glucose is
greater.13 This apparent anomaly might be explained by
di€ering modes of binding,14 although this has not been
determined. Thus, despite much work, the biological
activity of the thiazolidinediones still remains poorly
understood. Additionally, unlike PPARs a and g, the b
(or d) isoform has received little attention. Few of its
biological functions have been elucidated, most prob-
ably because no highly selective b agonist or antagonist
has been discovered.15,16
To examine the functional and structural features that
facilitate the binding of small molecules to PPARs and
to ®nd new activators and/or inhibitors, we have begun
work towards synthesizing libraries of thiazolidine-
dione±peptide hybrids. These targets allow utilization of
the `business end' of the thiazolidinedione antidiabetic
agents along with the structural diversity of commer-
cially available amino acids. Methods for the synthesis
of peptides are well-documented,17 and the solid-phase
approach is particularly e€ective for library production.
The use of peptide libraries to ®nd new biologically
active leads has proven successful for serotonin re-
uptake inhibitors18 and HIV protease inhibitors.19 To
date, the few libraries of thiazolidinediones substituted
only at the 5-position have been highly directed,20 based
on the para-alkoxybenzyl moiety in BRL49653.21
Despite the signi®cant sequence homology seen for all
PPAR isoforms,3,7,10 these enzymes do not bind or
respond equally to small molecule ligands. For example,
PPARa agonism is induced by medium- and long-chain
fatty acids, leukotriene B4, and, unexpectedly, KRP-297
(2),11,12 the only thiazolidinedione reported to activate
PPARa. The primary target for such molecules, including
the insulin sensitizer BRL49653 (1), appears to be the g
form.4
The feasibility of the synthetic approach was demon-
strated by initial studies focused on the solution-phase
synthesis of two thiazolidinedione±peptide molecules, 7
and 8 (Scheme 1). Thus, employing standard peptide
*Corresponding author. Tel.: +1-404-727-6738; fax: +1-404-727-6586;
e-mail: dmohler@emory.edu
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00440-6

2240
D. L. Mohler et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2239±2242
Figure 1. Thiazolidinedione PPAR ligands.
bond formation conditions, thiazolidinedione acid 4
(available in two steps from maleic anhydride and
thiourea)22 was treated with dipeptide 5 or 6 to furnish
each of the target compounds as a mixture of diaster-
eomers;23 however, the yields were unacceptably low.
While this result precluded the application of this
chemistry to the solid phase, it nevertheless provided an
encouraging precedent.
Gratifyingly, amide bond formation (Scheme 3)
between the N-protected thiazolidinedione acid 15 and
Phe-Gly-OMe (5) gave a much improved yield of the
desired compound 18.
Encouraged by these preliminary indications of success,
we then turned our attention to the solid-phase synthesis
of a small library of these compounds. The Fmoc pro-
tection strategy and a 2-chlorotrityl functionalized
polystryene resin were employed because of the mild
conditions available for amino group deprotection and
removal of the target from the resin. Additionally, any
excess reagents used in cleaving the compound from the
solid phase would be easily removed by evaporation,
allowing evaluation of the ®nal library without
puri®cation. Thus, standard solid-phase peptide syn-
thetic methods were applied using Fmoc-protected
glycine, leucine, or phenylalanine to produce (in par-
allel) nine resin-bound unprotected dipeptides of the
general formula 19 (Scheme 4). These were subsequently
treated with the carboxylic acid 16 in the presence of
dicyclohexylcarbodiimide (DCC), hydroxybenzotriazole
(HOBt), and triethylamine, followed by piperidine-
induced deprotection of the thiazolidinedione nitrogen.
Tri¯uoroacetic acid (TFA) caused release of the two
diastereomers of each of the nine desired compounds
Because the diculties with these and other attempted
peptide coupling methods were most likely due to com-
peting side reactions resulting from the acidity of the
thiazolidinedione proton (pKa ꢀ 6±824), it was logical to
protect the ring nitrogen. Unfortunately, none of
a variety of protecting groups (for example, TBDMS,
Bn) and conditions resulted in the selective reaction of
the thiazolidine-dione nitrogen over the carboxylic acid
moiety in 4. Therefore, an alternate approach was
adopted (Scheme 2), in which the nitrogen was func-
tionalized prior to introduction of the carboxylic acid
moiety. Thus, commercially available thiazolidinedione
(9) was doubly deprotonated and mono-alkylated25 with
allyl bromide to give 10, in which the nitrogen was subse-
quently protected with either the trimethylsilylethyl or
the ¯uorenylmethyl group under Mitsunobu conditions,
followed by ozonolysis of the alkene to give 13 or 14.
Oxidation of the aldehyde 13 to the acid 15 was
accomplished in 75% yield by NaClO2 (t-BuOH, iso-
butene in bu€ered aqueous THF) to give the carboxylic
acid 12. With KMnO4, a competing process gives a 37%
yield of the overoxidized product 17, which could be
exploited to furnish an additional diversity element.
Attempts to employ an oxidative work up for the
ozonolysis to provide 15 acid directly from 11 led only to
the complete destruction of the material. Jones oxidation
of ¯uorenylmethyl-protected compound 14 produced a
94% yield of acid 16, giving a 55% overall yield from 9.
1
from the solid support, as determined by H NMR and
mass spectrometry. Interestingly, in contrast to the
solution-phase studies, the unprotected thiazolidine-
dione acid 4 could also be employed for reactions on the
solid support with no apparent decrease in yield or
purity.
In summary, we have developed both solution- and solid-
phase approaches to the synthesis of peptide-substituted
thiazolidinediones as the ®rst steps in the preparation of
libraries of these compounds. Through the synthesis of
Scheme 1. Initial synthesis of thiazolidinedione-substituted dipeptides.
Scheme 2. Synthesis of protected thiazolidinedione acid.

D. L. Mohler et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2239±2242
2241
Scheme 3. Coupling of N-protected thiazolidinedione acid to a dipeptide.
Scheme 4. Solid-phase synthesis of a small library of peptide-substituted thiazolidinediones.
13. Reginato, M. J.; Bailey, S. T.; Krakow, S. L.; Minami, C.;
Ishii, S.; Tanaka, H.; Lazar, M. A. J. Biol. Chem. 1998, 273,
32679.
14. Ober®eld, J. L.; Collins, J. L.; Holmes, C. P.; Goreham,
D. M.; 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. U.S.A.
1999, 96, 6102.
novel, selective PPAR ligands, it is anticipated that new
signal pathways for regulating lipid production and
metabolism may be identi®ed to ultimately provide new
therapeutic agents for treating type II diabetes, obesity,
atherosclerosis, and/or hypertension.
Acknowledgements
15. Even L165041, the most selective PPARd binding ligand
reported (Ki=6 nM versus 730 nM for the g isoform), showed
only 10-fold greater activation of PPARd/GAL4 than the
PPARg chimera.16
We gratefully acknowledge the NSF, West Virginia Uni-
versity, and Emory University for support of this work.
16. Berger, J.; Leibowitz, 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.
17. Stewart, J.; Young, J. Solid Phase Peptide Synthesis;
Pierce Chemical: Rockford, IL, 1984.
References and Notes
1. (a) Vamecq, J.; Latru€e, N. Lancet 1999, 354, 141. (b)
Kliewer, S. A.; Lehmann, J. M.; Willson, T. M. Science 1999,
284, 757.
2. Sher, T.; Yi, H. F.; McBride, O. W.; Gonzalez, F. J. Bio-
chemistry 1993, 32, 5598.
3. Issemann, I.; Green, S. Nature 1990, 347, 645.
4. Lehmann, J. M.; Moore, L. B.; Smith-Oliver, T. A.; Wil-
kison, W. O.; Willson, T. M.; Kliewer, S. A. J. Biol. Chem.
1995, 270, 12953.
5. Kliewer, S. A.; Forman, B. M.; Blumberg, B.; Ong, E. S.;
Borgmeyer, U.; Mangelsdorf, D. J.; Umesono, K.; Evans, R.
M. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 7355.
6. Keller, H.; Dreyer, C.; Medin, J.; Mahfoudi, A.; Ozato, K.;
Wahli, W. Biochemistry 1993, 90, 2160.
7. Gottlicher, M.; Widmark, E.; Li, Q.; Gustafsson, J.-A.
Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 4653.
8. Wahli, W.; Braissant, O.; Desvergne, B. Chem. Biol. 1995,
2, 261.
18. Koppel, G.; Dodds, C.; Houchins, B. M.; Hunden, D.;
Johnson, D.; Owens, R.; Chaney, M.; Usdin, T.; Ho€man, B.;
Brownstein, M. Chem. Biol. 1995, 2, 483.
19. Owens, R. A.; Gesellchen, P. D.; Houchins, B. J.; DiMarchin,
R. D. Biochem. Biophys. Res. Commun. 1991, 181, 402.
20. Yanagisawa, H.; Takamura, M.; Yamada, E.; Fujita, S.;
Fujiwara, T.; Yachi, M.; Isobe, A.; Hagisawa, Y. Bioorg. Med.
Chem. Lett. 2000, 10, 373. Tomkinson, N. C. O.; Se¯er, A. M.;
Plunket, K. D.; Blanchard, S. G.; Parks, D. J.; Willson, T. M.
Bioorg. Med. Chem Lett. 1997, 7, 2491.
21. Brummond, K. M.; Lu, J. J. Org. Chem. 1999, 64, 1723.
22. Deghenghi, R.; Daneault, G. Can. J. Chem. 1960, 38,
1255.
23. All new compounds gave satisfactory spectral and analy-
1
tical data, including H NMR: 7: (270 MHz, CD3OD) d 7.91
9. Schoonjans, K.; Martin, G.; Staels, B.; Auwerx, J. Curr.
Opin. Lipidol. 1997, 8, 159.
10. Keller, H.; Devchand, P. R.; Perroud, M.; Wahli, W. Biol.
Chem. 1997, 378, 651.
11. Lin, Q.; Ruuska, S. E.; Shaw, N. S.; Dong, D.; Noy, N.
Biochemistry 1999, 38, 185.
12. Murakami, K.; Tobe, K.; Ide, T.; Mochizuki, T.; Ohashi,
M.; Akanuma, Y.; Yazaki, Y.; Kadowaki, T. Diabetes 1998,
47, 1841.
(s, 1H), 7.13±7.33 (m, 5H), 4.66±4.69 (m, 1H), 4.50 (ddd,
J=28.9, 10.1, 4.7, 1H), 3.93±3.99 (s, 2H), 3.71 (s, 3H), 3.19
(dd, J=15.3, 7.7, 1H), 2.81±2.99 (m, 2H), 2.72 (dd, J=16.2,
9.5, 1H). 10: (300 MHz, CDCl3) d 8.466 (br s, 1H), 5.746±5.882
(m, 1H), 5.208±5.259 (m, 2H), 4.368 (dd, J=9.0, 3.8), 2.901±
2.984 (m, 1H), 2.619±2.726 (m, 1H). 11: (270 MHz, CDCl3) d
5.69±5.84 (m, 1H), 5.21 (d, J=5.5, 1H), 5.16 (s, 1H), 4.21 (dd,
J=8.7, 4.0, 1H), 3.56±3.64 (ABm, 2H), 2.87±2.98 (m, 1H),
2.52±2.65 (m, 1H), 0.84±0.93 (ABm, 2H), 0.06 (s, 9H). 12:

2242
D. L. Mohler et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2239±2242
(300 MHz, CDCl3) d 7.755 (d, J=7.5, 2H), 7.434 (dd, J=7.4,
0.9, 2H), 7.384 (d, J=7.5, 2H), 7.305 (dd, J=7.7, 0.9, 2H),
5.725±5.807 (m, 1H), 5.253 (dd, J=6.6, 1.2, 1H), 5.206 (d,
J=1.2, 1H), 4.384 (t, J=7.5, 1H), 4.329 (dd, J=9.0, 4.2, 1H),
3.913±4.041 (m, 2H), 2.917±2.985 (m, 1H), 2.559±2.662 (m,
1H). 13: (270 MHz, CDCl3) d 9.79 (s, 1H), 4.38 (dd, J=9.9,
3.4, 1H), 3.61±3.67 (ABm, 2H), 3.53 (dd, J=19.4, 3.2, 1H),
3.05 (dd, J=19.4, 9.9, 1H), 0.90±0.96 (ABm, 2H), 0.05 (s, 9H).
14: (300 MHz, CDCl3) d 9.812 (s, 1H), 7.758 (d, J=7.8, 2H),
7.370±7.452 (m, 4H), 7.268±7.324 (m, 2H), 4.483 (dd, J=11.6,
3.2, 1H), 4.400 (t, J=6.9, 1H), 3.987±4.030 (m, 2H), 3.532 (dd,
J=19.4, 3.3, 1H), 3.038 (dd, J=19.5, 10.2, 1H). 15: (270 MHz,
CDCl3) d 4.39 (dd, J=8.9, 3.7, 1H), 3.60±3.66 (ABm, 2H),
3.30 (dd, J=18.0, 3.5, 1H), 3.00 (dd, J=18.1, 8.9, 1H), 0.88±
0.95 (ABm, 2H), 0.05 (s, 9H). 16: (300 MHz, CDCl3) d 7.755
(d, J=7.2, 2H), 7.375±7.446 (m, 4H), 7.308 (dd, J=6.9, 2.4,
2H), 4.508 (dd, J=9.3, 3.2, 1H), 4.384 (t, J=6.9, 1H), 3.950±
4.025 (m, 2H), 3.308 (dd, J=19.4, 3.2, 1H), 2.970 (dd, J=19.4,
9.3, 1H). 17: (270 MHz, CDCl3) d 3.63±3.69 (ABm, 2H), 3.41
(d, J=17.3, 1H), 3.14 (d, J=17.6, 1H), 0.89±0.96 (ABm, 2H),
0.06 (s, 9H). 18: (270 MHz, CDCl3) d 7.18±7.27 (m, 5H), 6.72±
6.81 (2d+t, 1.5H), 6.59 (t, 0.5H), 4.72±4.82 (m, 1H), 4.38
(ddd, J=25.9, 10.3, 3.8, 1H), 3.85±4.05 (m, 2H), 3.70 (s, 3H),
3.57±3.63 (ABm, 2H), 3.14 (dd, J=16.2, 3.8, 1H), 2.98±3.06
(m, 2H), 2.63±2.78 (m, 1H), 0.87±0.93 (ABm, 2H), 0.03 (s,
9H). 21: (400 MHz, CD3OD) d 7.204±7.277 (m, 5H), 4.650±
4.690 (m, 1H), 4.565 (dd, J=9.6, 4.0, 0.5H), 4.464 (dd, J=9.2, 4.4,
0.5H), 3.901±3.936 (m, 2H), 3.109±3.258 (m, 2H), 2.678±3.033 (m,
2H). 22: (400 MHz, CD3OD) d 4.593±4.668 (m, 1H), 4.420±
4.446 (m, 1H), 3.873±3.926 (m, 2H), 3.084±3.185 (m, 1H), 2.837±
3.012 (m, 1H). 23: (400MHz, CD3OD) d 7.108±7.293 (m, 5H),
4.601±4.670 (m, 1H), 4.519 (dd, J=9.6, 4.0, 0.5H), 4.440 (dd,
J=9.6, 4.0, 0.5H), 3.170±3.265 (m, 1H), 3.077±3.133 (m, 1H),
2.937±3.066 (m, 1H), 2.647±2.842 (m, 1H). 24: (400 MHz,
CD3OD) d 7.216±7.471 (m, 10H), 4.599±4.679 (m, 2H), 4.080±
4.250 (m, 1H), 2.930±3.260 (m overlapping t at 3.138, J=7.2,
6H). 25: (400 MHz, CD3OD) d 7.170±7.290 (m, 5H), 4.568±
4.663 (m, 1.5H), 4.360±4.430 (m, 1H), 4.170±4.280 (m, 0.5H),
2.975±3.264 (m, 3H), 2.784±2.911 (m, 1H), 1.580±1.670 (m,
1H), 1.458±1.558 (m, 2H), 0.877±0.974 (m, 6H). 26: (400 MHz,
CD3OD) d 4.570±4.694 (m, 1H), 4.460±4.509 (m, 1H), 4.126±
4.273, m, 2H), 3.340±3.440 (m, 1H), 2.717±2.817 (m, 1H),
1.641±1.728 (m, 3H), 0.969 (d, J=4.0, 3H), 0.927 (d, J=4.0,
3H). 27: (400 MHz, CD3OD) d 7.197±7.270 (m, 5H), 4.673
(dd, J=9.2, 5.2, 1H), 4.449±4.565 (m, 2H), 3.201 (t, J=5.4,
0.5H), 3.166 (t, J=5.4, 0.5H), 3.110 (dd, J=16.2, 3.8, 0.5H),
3.150 (dd, J=16.2, 3.8, 0.5H), 2.816±2.897 (m, 1H), 2.737 (dd,
J=16.4, 5.2, 1H), 1.679±1.726 (m, 1H), 1.630±1.652 (m, 2H),
0.958 (d, J=6.2, 3H), 0.921 (d, J=6.2, 3H). 28: (400 MHz,
CD3OD) d 4.592±4.643 (m, 1H), 4.409±4.462 (m, 2H), 3.093±
3.163 (m, 1H), 2.851±2.944 (dd overlapping dd, J=11.8, 9.0
and 12.2, 8.8, 1H), 1.548±1.728 (m, 6H), 0.895±0.974 (m, 12H).
24. The pKa of 2,4-thiazolidinedione has been reported as 6.74
at 25 ꢁC: Kanolt, C. W. J. Am. Chem. Soc. 1907, 29, 1402.
25. Zask, A.; Nowicki, J. W.; Jirkovsky, I.; Van Engen, D.
Tetrahedron Lett. 1993, 34, 2719.