Design and synthesis of novel PPARα/γ/δ triple activators using a known PPARα/γ dual activator as structural template

Article abstract of DOI:10.1016/S0960-894X(02)00881-8

Using a known dual PPARα/γ activator (5) as a structural template, SAR evaluations led to the identification of triple PPARα/γ/δ activators (18-20) with equal potency and efficacy on all three receptors. These compounds could become useful tools for studying the combined biological effects of PPARα/γ/δ activation.

Full text of DOI:10.1016/S0960-894X(02)00881-8

Bioorganic & Medicinal Chemistry Letters 13 (2003) 257–260
Design and Synthesis of Novel PPARꢀ/ꢁ/ꢂ Triple Activators
Using a Known PPARꢀ/ꢁ Dual Activator as Structural Template
John P. Mogensen,* Lone Jeppesen, Paul S. Bury, Ingrid Pettersson, Jan Fleckner,
Jan Nehlin, Klaus S. Frederiksen, Tatjana Albrektsen, Nanni Din, Steen B. Mortensen,
L. Anders Svensson, Karsten Wassermann, Erik M. Wulff, Lars Ynddal
and Per Sauerberg
˚
Novo Nordisk A/S, Novo Nordisk Park, 2760 Maløv, Denmark
Received 18 June 2002; revised 19 August 2002; accepted 7 October 2002
Abstract—Using a known dual PPARa/g activator (5) as a structural template, SAR evaluations led to the identification of triple
PPARa/g/d activators (18–20) with equal potency and efficacy on all three receptors. These compounds could become useful tools
for studying the combined biological effects of PPARa/g/d activation.
# 2002 Elsevier Science Ltd. All rights reserved.
The use of Peroxisome Proliferator-Activated Receptor g
(PPARg) activators, for example rosiglitazone (1, Fig. 1)
in the treatment of type 2 diabetes or non-insulin depen-
dent diabetes mellitus is well established due to their ability
to lower blood glucose and insulin levels and improve
insulin sensitivity.1 Furthermore, the activation of PPARa
receptors has been suggested to be the mechanism behind
the hypolipidemic effect of the fibrate drugs, for example
fenofibrate2 (2). These findings have increased the interest
in developing a PPARa/g dual activators, for example
ragaglitazar (NNC 61-0029/DRF2725)3 (3) for the treat-
ment of dyslipidemic type 2 diabetics. In light of the
recently published data4 on the selective PPARd agonist
GW501516 (4), which showed increased HDL-cholesterol
and decreased triglycerides, along with VLDL and LDL-
cholesterol levels effects in obese monkeys, PPARd acti-
vators could also be used in the treatment of dyslipidemia.
We report here the design of triple PPARa/g/d activators
using a dual PPARa/g activator as structural template.
alterations to both the lipophilic and the acidic part of
the molecule (see Fig. 1), we demonstrate here that tri-
ple PPARa/g/d activators can be designed by modifying
only the lipophilic part.
Using the recently published dual PPARa/g activator 55
as a structural template (Fig. 2), the N,N-diphenylamine
derivative 6 without the bond between the two phenyl
rings in the carbazole ring of 5 was synthesised. This
resulted in a loss in both potency and efficacy in the in
vitro transactivation of both PPARa and PPARg but it
gave no activation of the PPARd receptor (Table 1).
However, by substituting the amino group in 6 with a
methine group a more planar lipophilic structure was
obtained 7, which had equal potency on all three PPAR
receptor subtypes.
Results and Discussion
Whereas selectivity for PPARa, PPARg and PPARd
receptors has normally been obtained by making
*Corresponding author. Tel.:+45-4443-4213; fax:+45-4466-3939;
e-mail: jpm@novonordisk.com
Figure 1. Reference compounds.
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00881-8

258
J. P. Mogensen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 257–260
The SAR on the PPARd activity for this class of com-
pounds is in agreement with earlier published data sug-
gesting the presence of two lipophilic pockets in the
binding domain of the receptor.6 However, the fact that
both the PPARa and the PPARg receptors can accom-
modate such large conformationally constrained and
branched lipophilic groups, as in 18–20, has not yet
been reported. Although the flexible triple activator
GW2433 also supports this finding.1
FlexX7ꢀ10 was used to dock compound 18 into PPARa
co-crystallized with AZ242,11 PPARg co-crystallized
with 5,5 and PPARd co-crystallized with GW2433,6
respectively. Figure 3 shows that compound 18 docks
into the active sites with the carboxylic acid in the same
position as the co-crystallized ligands. Furthermore,
there is good agreement between the position of the
hydrophobic portion of compound 18 and the lipophilic
regions of the co-crystallized ligands.
Figure 2. Evolution of the project.
In an attempt to improve the potency of 7 and to
understand the SAR of this class of compounds, mono-
substituted analogues of 7 were synthesized and tested.
The potency was actually improved compared to the
non-substituted compound 7, especially with the more
lipophilic substituted analogues 10–15.
The bis-biphenyl compound 19 was tested in our pri-
mary diabetes model, the male db/db mouse,5 primarily
reflecting the PPARg agonist activity. Mice were treated
with 1 mg/kg orally once daily for 7 days, after which a
non-fasted blood sample was drawn. Compared to
vehicle treated animals the blood glucose (BG) and the
plasma insulin levels were lowered by 47 and 71%,
respectively. For comparison, 3 dosed at 3 mg/kg po
Activation data showed however, only minor potency
and efficacy differences between the (E) and the (Z) iso-
mers (8 vs 9; 10 vs 11; 12 vs 13 and 14 vs 15). This
prompted us to design and synthesize the bis-substituted
derivatives 16–20. The bis-substituted analogues were,
however, equipotent to the mono-substituted analogues
(10 and 11 vs 19; 12 and 13 vs 18; 14 and 15 vs 20).
Table 1. In vitro hPPARa, hPPARg, and hPPARd activation of reference and test compounds
No.
R1 (E)
R2 (Z)
PPARaa EC50
,
PPARga EC50
,
PPARdd EC50
,
mM (%max)b
mM (%max)c
mM (%max)d
7
8
9
H
F
H
Ph
H
Br
H
H
F
H
Ph
H
Br
H
6.4 (107)
5.3 (121)
3.2 (201)
0.9 (206)
0.7 (182)
1.1 (86)
2.2 (161)
0.3 (160)
0.4 (198)
0.7 (192)
2.2 (133)
0.3 (132)
1.1 (77)
2.7 (75)
8.5 (72)
4.0 (94)
0.6 (104)
1.7 (129)
1.5 (89)
4.3 (90)
0.8 (112)
2.0 (85)
3.1 (87)
2.2 (101)
0.9 (85)
0.3 (101)
2.9 (161)
5.7 (195)
8.3 (159)
3.4 (195)
2.2 (274)
2.2 (250)
0.9 (312)
3.4 (233)
0.6 (200)
2.0 (229)
2.0 (214)
0.7 (233)
0.4 (239)
0.5 (265)
1.0 (212)
10
11
12
13
14
15
16
17
18
19
20
H
2-Furanyl
H
Me
MeO
Br
Ph
2-Furanyl
Me
MeO
Br
Ph
2-Furanyl
2-Furanyl
2.0 (103)
1
Rosiglitazone
Fenofibric acid
NNC61-0029/ DRF2725
4.1 (43)
32.1 (265)
3.21 (97)
0.36 (140)
4.5 (39)
0.16 (100)
(8)
0.57 (117)
0.17 (108)
3.3 (91)
(7)
(1)
(7)
(3)
(1)
2e
3
5
6
aValues are means of 2–4 experiments in quadruple; SEMs were within 20%. Fold activation relative to maximum activation obtained with:
bWY14643 (approximately 20-fold corresponded to 100%), with:
crosiglitazone (approximately 120-fold corresponded to 100%), and with:
dcarbacyclin (approximately 250-fold corresponded to 100%).
eFenofibric acid was used.

J. P. Mogensen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 257–260
259
Figure 3. Compound 18 (purple) docked into the PPARa receptor co-crystallized with AZ 242 (left panel), into the PPAR receptor co-crystallized
with 5 (center panel) and into the PPARd receptor co-crystallized with GW2433 (right panel), respectively.
produced a 51% BG and a 79% insulin lowering in the
same experiment. After a further 2 days dosing, an oral
glucose tolerance test (OGTT) was performed. The
reduction of the blood glucose area under the curve
(AUC) after the OGTT was considered to be a measure
of improved insulin sensitivity. Compound 19 lowered
the AUC by 52% compared to 55% after treatment
with 3. The pharmacokinetic profile of 19 in rat after
oral administration (2.54 mg/kg) showed a preferred
long half life of 5.4 h with a Cmax of 132 ng/mL. Fur-
ther studies are, however, needed to fully evaluate the
combined effect of activating all three PPAR receptors.
NMR. DIBAL-H reduction of the ester group of 21
allowed the preparation of the to the corresponding
alcohol (22) which were subjected to a Mitsunobu reac-
tion with (S) ethyl 3-(4-hydroxy-phenyl)-2-ethoxy-
propionate to give the ethyl esters of the desired
products 6–13 and 16–18. Hydrolysis of the esters pro-
duced the target compounds 6–13 and 16–18.
Compounds 14 and 15 were made using the (E) and (Z)
3-(4-bromo-phenyl)-3-phenyl-prop-2-en-1-ols as pre-
cursers for the introduction of the furanyl substituent
via a Suzuki coupling with 2-furanyl boronic acid,
(Scheme 2). The disubstituted analogues 19 and 20 were
made the same way using ethyl 3,3-bis-(4-bromo-
phenyl)-prop-2-en-1-ol as the precursor.
In conclusion, our results show that it is possible to
design triple PPAR activators by keeping the acidic part
of a dual PPAR activator unchanged and modifying
only the lipophilic part of the dual activator. The data
generated on compounds 18–20 suggests that these
compounds could be tools for studies of the combined
biological effects of PPARa/g/d activation.
Scheme 2. (i) Pd(PPh3)4, DME, Na2CO3 aq 80 ꢁC.
Chemistry
Compounds 6–13 and 16–19 were synthesized12 as out-
lined in Scheme 1. Carrying out a Horner Emmons
reaction on commercial available diphenyl ketones led
to the ethyl 3,3-diphenyl-propenonates. The (E) and
(Z) mixtures of 21 were then seperated by column
chromatography and the conformations assigned using
Modelling
FlexX version 1.10.0 was used.7ꢀ10 Assignment of for-
mal charges was made. FlexX was used with the Sybyl
6.8 interface.13
In vitro transactivation
The assay as done for PPARa and PPARg has been
described previously.5 Briefly, the ligand binding
domains of the two human PPAR receptor subtypes
were fused to the DNA binding domain of the yeast
transcription factor Gal4. HEK293 cells were tran-
siently transfected with an expression vector for the
respective PPAR chimera along with a reporter con-
struct containing five copies of the Gal4 DNA binding
site driving expression of a luciferase reporter gene. All
compounds were dissolved in DMSO and diluted 1:1000
upon addition to the cells. Compounds were tested in
five concentrations ranging from 0.01 to 30 mM. Cells
were treated with compound for 24 h followed by luci-
ferase assay. The PPARd assay was done exactly as
described above for PPARa and PPARg using the
human PPARd ligand binding domain (LBD) spanning
amino acids 128—end fused to the Gal4 DNA binding
Scheme 1. (i) NaOEt, triethylphosphonoacetat, EtOH rt; (ii) DiBAL-
H, THF, ꢀ70 ꢁC, (iii) DEAD or ADDP, PPh3 or P(t-Bu)3 THF, 0 ꢁC;
(iv) 1 M NaOH, EtOH, rt ꢀ50 ꢁC.

260
J. P. Mogensen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 257–260
domain in the vector pM1, yielding the plasmid
pM1dLBD.
with brine, dried with MgSO4, filtered and evaporated. The
residue was purified by column chromatography (eluent:
ether) to give ethyl 3,3-bis-(4-bromophenyl)-acrylate as a gum.
Crystallization from hexanes gave white crystals in 8.77 g
1
(73%) yield. H NMR (CDCl3, 300 MHz), d 1.20 (3H, t), 4.05
Acknowledgements
(2H, q), 6.35 (1H, s), 7.0–7.1 (4H, m), 7.40–7.52 (4H, m). Step
2: Ethyl 3,3-bis-(4-bromophenyl)-acrylate (8.75 g, 21.3 mmol)
was dissolved in dry THF (35 mL). DIBAL-H (1.5 M in tolu-
ene, 43 mL, 64.0 mmol) was added at ꢀ15 ꢁC and the reaction
mixture was stirred for 30 min. A solution of NH4Cl in water
was added and the mixture was extracted with EtOAc (3ꢂ50
mL). The combined organic phases were washed with brine,
dried with MgSO4, filtered and evaporated to give 3,3-bis-(4-
bromophenyl)-pro-2-en-1-ol in 6.0 g (76%) yield. 1H NMR
(CDCl3, 300 MHz), d 1.15 (1H, br s), 4.16–4.20 (2H, dd), 6.25
(1H, t), 7.0–7.1 (4H, m), 7.40-7.52 (4H, m). Step 3: To a solu-
tion of 3,3-bis-(4-bromophenyl)-pro-2-en-1-ol (1.00 g, 2.72
mmol) and Pd(PPh3)4 (0.11 g, 0.1 mmol) in DME (40 mL) was
added phenylboronic acid (1.33 g, 10.88 mmol) and the reac-
tion was stirred for 10 min at rt under N2. Na2CO3 (2 M aq,
16.3 mL, 32.6 mmol) was added and the reaction was stirred at
80 ꢁC overnight followed by 3 days at rt. Water was added to
the reaction mixture and the product extracted with CH2Cl2
(3ꢂ50 mL). The combined organic phases were washed with
brine, dried with MgSO4, filtered and evaporated. The residue
was purified on column chromatography [eluent/EtOAc/hex-
anes (3:1)] to give 3,3-bis-(biphenyl-4-yl)-pro-2-en-1-ol in 630
The technical assistance from Rikke Burgdorf, Hanne
Nielsen, Lene Priskorn, Helle Bach, Anette Heerwagen,
Anette Zeneca, Kirsten M. Klausen, Kent Pedersen,
Brian Rosenberg, Otto Larsson, Sanne von Eyben,
Bjørn Metzler, Per Klifforth, and Alice Ravn is highly
appreciated.
References and Notes
1. Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B.
R. J. Med. Chem. 2000, 43, 527.
2. Staels, B.; Dallongeville, J.; Auwerx, J.; Schoonjans, K.;
Leitersdorf, E.; Fruchart, J.-C. Circulation 1998, 98, 2088.
3. Lohray, B. B.; Lohray, V. B.; Bajji, A. C.; Kalchar, S.;
Poondra, R. R.; Padakanti, S.; Chakrabarti, R.; Vikrama-
dithyan, R. K.; Misra, P.; Juluri, S.; Mamidi, N. V. S. R.;
Rajagopalan, R. J. Med. Chem. 2001, 44, 2675.
4. 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. 2001, 98, 5306.
5. Sauerberg, P.; Pettersson, I.; Jeppesen, L.; Bury, P. S.;
Mogensen, J. P.; Wassermann, K.; Brand, C. L.; Sturis, J.;
Woldike, H. F.; Fleckner, J.; Andersen, A.-S. T.; Mortensen,
S. B.; Svensson, L. A.; Rasmussen, H. B.; Lehmann, S. V.;
Polivka, Z.; Sindelar, K.; Panajotova, V.; Ynddal, L.; Wulff,
E. M. J. Med. Chem. 2002, 45, 789.
6. Xu, H. E.; Lambert, M. H.; Montana, V. G.; Parks, D. J.;
Blanchard, S. G.; Brown, P. J.; Sternbach, D. D.; Lehmann,
J. M.; Wisely, G. B.; Willson, T. M.; Kliewer, S. A.; Milburn,
M. V. Mol. Cell 1999, 3, 397.
1
mg (64%) yield. H NMR (CDCl3, 300 MHz), d 1.4 (1H, t),
4.32 (2H, dd), 6.45 (1H, t), 7.2–7.7 (18H, m). Step 4: 3,3-bis-
(biphenyl-4-yl)-2-pro-2-en-1-ol (363 mg, 1.0 mmol) and (S)-
ethyl-2-ethoxy-3-(4-hydroxyphenyl)-propionate (262 mg, 1.1
mmol) and Bu3P (303 mg, 1.5 mmol) were dissolved in dry
benzene (20 mL) and cooled to 0 ꢁC. 1,10-(Azodicarbonyl)di-
piperidine (ADDP) (378 mg, 1.5 mmol) was added and the
reaction mixture was stirred at room temp. for 48 h. Water (50
mL) was added and the mixture was extracted with EtOAc
(3ꢂ50 mL). The combined organic phases were washed with
brine, dried with MgSO4, filtered and evaporated. The residue
was purified by column chromatography [eluent/EtOAc/
hexanes (3:1)] to give (S)-ethyl-3-(4-(3,3-bis-(biphenyl-4-yl)-
allyloxy)-phenyl)-2-ethoxy-propionate in 445 mg (76%)
yield. 1H NMR (CDCl3, 300 MHz), d 1.15 (3H, t), 1.25
(3H, t), 2.92 (1H, d), 3.28-3.41 (1H, m), 3.52-3.66 (1H, m),
4.0 (1H, dd), 4.22 (2H, q), 4.65 (2H, d), 6.40 (1H, t), 6.79
(2H, d), 7.12 (2H, d), 7.30-7.70 (18H, m). Step 5: A sus-
pension of (S)-ethyl-3-(4-(3,3-bis-(biphenyl-4-yl)-allyloxy)-
phenyl)-2-ethoxy-propionate (370 mg, 0.64 mmol) in EtOH (4
mL) and 1 M NaOH (1.27 mL, 1.27 mmol) was heated to give
a clear solution that was stirred for 5 days at rt. The reaction
mixture was washed with EtOAc and 1 N HCl was added to
pH<2. The water phase was extracted with EtOAc (3ꢂ50 mL)
and the combined organic phases were washed with brine,
dried with MgSO4, filtered and evaporated. The residue was
purified by column chromatography (eluent: 5% MeOH in
CH2Cl2) to give (S)-3-(4-(3,3-bis-(biphenyl-4-yl)-allyloxy)-
7. Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. J. Mol.
Biol. 1996, 261, 470.
8. Kramer, B.; Rarey, M.; Lengauer, L. Proteins: Struct.,
Funct. Genet. 1999, 37, 228.
9. Rarey, M.; Wefing, S.; Lengauer, T. J. Comput.-Aided Mol.
Des. 1996, 10, 41.
10. Rarey, M.; Kramer, B.; Lengauer, T. J. Comput.-Aided
Mol. Des. 1997, 11, 369.
11. Cronet, P.; Petersen, J. F. W.; Folmer, R.; Blomberg, N.;
Sjøblom, K.; Karlsson, U.; Lindstedt, E.-L.; Barnberg, K.
Structure 2001, 9, 699.
12. Target compounds were synthesized according to the fol-
lowing general procedure, using the synthesis of compound 19
as an example: Step 1: To a solution of NaH (3.53 g, 88.2
mmol) in dry toluene (300 mL) was added dropwise at 0 ꢁC a
solution of trietylphosphonoacetate (13.2 g, 58.8 mmol) in
toluene (100 mL). The reaction mixture was stirred for 30 min
after which a solution of 4,4-dibromobenzophenone (10.0 g,
29.4 mmol) in THF (100 mL) was added. The reaction mixture
was stirred for 48 h. EtOH (10 mL) and water (300 mL) were
added and the mixture was extracted with EtOAc–MeOH
(2%, 2ꢂ150 mL). The combined organic phases were washed
1
phenyl)-2-ethoxy-propionic acid in 275 mg (78%) yield. H
NMR (CDCl3, 300 MHz), d 1.13 (3H, t), 2.93 (1H, dd), 3.05
(1H, dd), 3.35-3.50 (1H, m), 3.50-3.64 (1H, m), 4.02 (1H, dd),
4.62 (2H, d), 6.39 (1H, t), 6.80 (2H, dm), 7.02 (2H, dm), 7.27–
7.72 (18H, m).
13. SYBYL1 6.8; Tripos Inc.: 1699 South Hanley Rd., St.
Louis, MO, 63144, USA.