Diphenylmethane skeleton as a multi-template for nuclear receptor ligands: Preparation of FXR and PPAR ligands

Article abstract of DOI:10.1016/j.bmcl.2006.03.075

Novel, potent farnesoid X receptor (FXR) and peroxisome proliferator-activated receptor α (PPARα) agonists were obtained by using a diphenylmethane skeleton as a substitute for a steroid skeleton.

Full text of DOI:10.1016/j.bmcl.2006.03.075

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3213–3218
Diphenylmethane skeleton as a multi-template for nuclear
receptor ligands: Preparation of FXR and PPAR ligands
Masahiko Kainuma,a, Jun-ichi Kasuga,a, Shinnosuke Hosoda,a,
Ken-ichi Wakabayashi,a Aya Tanatani,a Kazuo Nagasawa,b Hiroyuki Miyachi,a
Makoto Makishimac and Yuichi Hashimotoa,*
aInstitute of Molecular and Cellular Biosciences, University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-0032, Japan
bDepartment of Biotechnology and Life Science, Faculty of Technology, Tokyo University of Agriculture and Technology,
2-24-16 Naka-cho, Koganei, Tokyo 184-8588, Japan
cDepartment of Biochemistry, Nihon University School of Medicine, 30-1 Oyaguchi-kamicho, Itabashi-ku, Tokyo 173-8610, Japan
Received 24 February 2006; revised 14 March 2006; accepted 15 March 2006
Available online 17 April 2006
Abstract—Novel, potent farnesoid X receptor (FXR) and peroxisome proliferator-activated receptor a (PPARa) agonists were
obtained by using a diphenylmethane skeleton as a substitute for a steroid skeleton.
Ó 2006 Elsevier Ltd. All rights reserved.
Nuclear receptors (NRs) are a family of ligand-depen-
dent transcription factors which modulate gene expres-
sion and thereby influence diverse biological processes,
including cell growth, differentiation, and metabolism.1,2
Analysis of the human genome sequence indicates that
there are 48 NRs in humans.1 So far, the ligands of only
about half of them have been identified.1,3 From the
standpoint of chemical structure, the identified ligands
for NRs most commonly contain a steroid skeleton, as
typically found in the ligands for all of the classical ste-
roid hormone receptors [estrogen receptors, progester-
one receptor, androgen receptor (AR), glucocorticoid
receptor, and mineral corticoid receptor]. In addition,
oxysterols and cholic acid derivatives are considered to
act as physiological ligands for liver X (oxysterol) recep-
tors (LXRs) and farnesoid (bile acid) receptor (FXR).
The ligand for vitamin D receptor (VDR) is 1a,25-
dihydroxyvitamin D3, which has a secosteroidal skele-
ton that is metabolically derived from cholesterol,
although lithocholic acid, a steroid analog produced
metabolically from cholesterol, is also reported to be a
physiological ligand for VDR.4 Because all members
of NRs are thought to have evolved from a single gene,5
it might be expected that a ligand superfamily also
exists.6,7
There have been many structural development studies of
non-steroidal/non-secosteroidal ligands for the above
NRs from the standpoint of medicinal chemistry, includ-
ing our studies on ligands for VDR and AR.7,8 Recently,
we developed diphenylmethane-type VDR ligands,8 using
LG190155 (1) (Fig. 1) reported by Boehm et al.9 as a lead
compound. In that work, we found that a nitrogen-con-
taining diphenylpentane (DPP) derivative 2 (DPP-0113)
(Fig. 1) acts as a dual ligand for both VDR and AR.8 This
result, as well as previous structural development studies
on nuclear retinoic acid receptor (RAR) ligands,7,10,11 im-
plies that a diphenylmethane skeleton can substitute for a
steroid skeleton as a multi-template for NR ligands.
To confirm the usefulness of the diphenylmethane skele-
ton as a multi-template for NR ligands, we designed novel
ligands for FXR and peroxisome proliferator-activated
receptor-a (PPARa). These NRs were chosen as targets
because of the strict structural requirements for their li-
gands, and because steroidal ligands of FXR and PPARa,
including chenodeoxycholicacid (CDCA: 3)12 and a diph-
enylmethane-related ligand, fenofibrate (4), respective-
ly,13 are already known (Fig. 1). In this paper, we
describe the design, synthesis, and biological activity
evaluation of novel diphenylmethane-based ligands for
FXR and PPARa.
Keywords: Nuclear receptor ligands; FXR; PPAR; Drug design;
Diphenylmethane.
*
Corresponding author. E-mail: hashimot@iam.u-tokyo.ac.jp
These authors contributed equally to this work.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.03.075

3214
M. Kainuma et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3213–3218
Figure 1. Structures of LG190155 (1), DPP-0113 (VDR/AR dual ligand: 2), CDCA (FXR ligand: 3) and fenofibrate (PPARa ligand: 4).
Ligands for FXR. FXR is a well-characterized member
of the so-called ‘metabolic’ subfamily of NRs and is a
transcriptional sensor for bile acids.14 Its ligands, includ-
ing CDCA (3) (Fig. 1), act as signaling molecules and
participate in an intricate network of interactions that
ultimately govern lipid, steroid, and cholesterol homeo-
stasis, and are involved in processes such as glucose uti-
lization, inflammation, and carcinogenesis.14 Maloney
et al. reported GW4064 (5) (Fig. 2) as a potent synthetic
ligand for FXR.15 Considering the structures of CDCA
(3) and GW4064 (5), and the potential of diphenylpen-
tane (DPP) as a steroid skeleton substitute, we designed
DPP derivatives as FXR ligand candidates, that is,
DPPF-01 (6) and DPPF-13 (7) (Fig. 2). The R group
of DPPF-01 (6), containing a carboxylic acid group,
was introduced to mimic the carboxylic acid group
found in CDCA (3) and GW4064 (5). The R group of
DPPF-13 (7), containing a diol moiety, was introduced
to mimic side chains found in several oxycholesterols.
DPPF-01 (6) and DPPF-13 (7) were prepared by usual
organic synthetic methods, as summarized in Scheme
1. Briefly, bisphenol (8) and its mono-anilino derivative
(9) were prepared as already reported.8 Coupling of 8 or
9 with 3-(2,6-dichlorophenyl)-4-bromomethyl-5-isopro-
pyl-isozaxole (10) in DMF in the presence of NaH gave
11 or 12, respectively. The phenolic hydroxyl group of
11 was alkylated with methyl 2-bromoethylate, followed
by hydrolysis of the ester to give DPPF-01 (6).16 Reac-
tion of 12 with glycidol gave DPPF-13 (7).17 CDCA
(3) was purchased from Sigma Co. Ltd (Japan).
The FXR agonistic activity of DPPF-01 (6) and DPPF-
13 (7) was evaluated by a reporter gene assay method
Figure 2. Structures of GW4064 (5) and compounds designed as diphenylmethane-type ligands for FXR (6 and 7).
Scheme 1. Synthesis of compounds 6 and 7.

M. Kainuma et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3213–3218
3215
Figure 3. Transcriptional activation of FXR by DPPF-01 (6), DPPF-13 (7) and CDCA (3). Horizontal scale: Concentration of added test
compounds. Vertical scale: Relative luminescence intensity arising from luciferase reporter gene expression.
using CMX-GAL4N-hFXR as the recombinant recep-
tor gene, TK-MH100x4-LUC as the reporter gene, and
the CMX b-galactosidase gene for normalization, as
previously reported.4,12,18 Briefly, human embryonic
kidney 293 cells were cultured in Dulbecco’s modified
Eagle’s medium containing 5% fetal bovine serum at
37 °C in a humidified atmosphere of 5% CO2 in air.
Transfections were performed by the calcium phos-
phate coprecipitation method. Test compounds were
added 8 h after the transfection, and luciferase and
b-galactosidase activities were assayed using a lumino-
meter and microplate reader. The experiment was
repeated three times, and the normalized average val-
ues are presented in Figure 3. Transcriptional activa-
tion activities (and antagonistic activities) of the test
compounds toward other NRs (Figs. 4 and 7) were
similarly evaluated.4,12,18
The transcriptional activation activity of DPPF-01 (6)
seems to be specific for FXR, because DPPF-01 did
not activate VDR, PPARa, PPARc, PPARd, LXRa,
RARa or retinoid X receptor a (RXRa) at 10 lM under
these experimental conditions (Fig. 4). This specificity of
DPPF-01 (6) toward FXR would be attributed to its side
chain’s structure which had been optimized by Maloney
et al. for GW4064 (5).15 The results suggest that the
strategy adopted in this study of utilizing the diph-
enylmethane skeleton as a steroid substitute is useful
to create superior/specific ligands. The computer-assist-
ed overlaying studies of DPPF-01 (6) and CDCA (3)
are under way.
Ligands for PPARa. PPARa is a NR whose physiolog-
ical ligands are considered to be endogenous fatty acids,
and it is well known as the target molecule of fenofibrate
(4) (Fig. 1),13 a drug used to treat dyslipidemia and type
2 diabetes, and whose active form is considered to be its
hydrolyzed analog, fenofibric acid (FA: 13) (Fig. 5).
Various synthetic PPARa ligands have been reported,
including our phenylpropionic acid derivatives derived
from KCL (14) (Fig. 5).18–20 PPARa has a large Y-
shaped ligand-binding pocket of approximately 1300–
As shown in Figure 3, the agonistic activity of DPPF-01
(6) for transcriptional activation of FXR was far more
potent than that of the physiological ligand, CDCA
(3). The EC50 values are 3.4 lM for DPPF-01 (6) and
11.7 lM for CDCA (3) under the experimental condi-
tions employed. DPPF-13 (7) also showed agonistic
activity toward FXR, though it was less potent than
CDCA (3). No apparent antagonistic activity of
DPPF-13 (7) was observed.
3
21
˚
1400 A . Based on the structures of FA (13) and
KCL (14), as well as the shape of the ligand-binding
pocket of PPARa and the potential usefulness of the
Figure 4. Transcriptional activation of various NRs by DPPF-01 (6). Vertical scale: Relative luminescence intensity arising from luciferase reporter
gene expression.

3216
M. Kainuma et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3213–3218
Figure 5. Structures of FA (13), KCL (14), DPPK-01 (15) and DPHK-01 (16).
diphenylmethane skeleton as a scaffold for PPARa
ligands, we designed a diphenylpentane derivative
DPPK-01 (15) and a diphenylcyclohexane derivative
DPHK-01 (16) (Fig. 5), both of which possess a butyric
acid moiety. The butyric acid moiety was introduced to
mimic the carboxylic acid side chain of KCL (14), and a
cyclohexyl moiety was adopted to provide a rigid
Y-shape of the molecule, as the carbonyl group of
FA (13) does.
fibrate (4). The EC50 values were 3.5 lM for DPHK-
01 (16) and 9.2 lM for FA (13) under the experimen-
tal conditions used. The less rigid derivative, DPPK-
01 (15), showed only very weak agonistic activity to-
ward PPARa (no apparent antagonistic activity was
observed). In these preliminary studies, we used dia-
stereomeric mixtures of DPPK-01 (15) and DPHK-
01 (16). Among PPARa-activating phenylpropionic
acid derivatives, the isomers with S-configuration at
the a-carbon of the propionic acid moiety have been
established to be the eutomers (active isomers), while
the R-isomers are inactive. Therefore, existence of a
eutomer of DPHK-01 (16) that would be far more
potent than FA (13) can be expected. Evaluation of
the biological activity of the optical isomers of our
compounds is in progress. The transcriptional activa-
tion activity of DPHK-01 (16) seems to be specific
for PPARa, because DPHK-01 did not activate
VDR, FXR, PPARc, PPARd, LXRa, RARa or
RXRa at 10 lM under these experimental conditions
(Fig. 7). These results again suggest the usefulness of
the diphenylmethane skeleton as a multi-template for
NR ligands.
DPPK-01 (15)22 was prepared by reaction of a bisphenol
derivative 8 with ethyl 2-bromobutylate in DMF in the
presence of NaH. DPHK-01 (16)23 was similarly pre-
pared, but the starting bisphenol derivative was synthe-
sized by condensation of cyclohexanone with o-cresol.
FA (13) was generously supplied by ASKA Pharm.
Co. Ltd (Japan).
The PPARa agonistic activity of DPPK-01 (15) and
DPHK-01 (16) was evaluated with a reporter gene assay
method similar to that adopted for evaluation of FXR
agonistic activity (vide supra), except that CMX-
GAL4N-hPPARa was used as the recombinant receptor
gene in place of CMX-GAL4N-hFXR. The results are
shown in Figures 6 and 7.
In conclusion, we have prepared novel ligands for
FXR, DPPF-01 (6), and for PPARa, DPHK-01
(16), based on a diphenylmethane skeleton. DPPF-
01 (6) is more potent than the physiological ligand
CDCA (3), and the potency of DPHK-01 (16) is
As shown in Figure 6, DPHK-01 (16) showed more
potent agonistic activity for transcriptional activation
of PPARa than did FA (13), the active form of feno-
Figure 6. Transcriptional activation of PPARa by DPPK-01 (15), DPHK-01 (16) and FA (13). Horizontal scale: Concentration of added test
compounds. Vertical scale: Relative luminescence intensity arising from luciferase reporter gene expression.

M. Kainuma et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3213–3218
3217
Figure 7. Transcriptional activation by DPHK-01 (16) toward various NRs. Vertical scale: Relative luminescence intensity arising from luciferase
reporter gene expression.
3. Giguere, V. Endocr. Rev. 1999, 20, 68.
comparable to that of FA (13), the active form of
the antihyperlipidemic drug fenofibrate (4). These
results indicate that a diphenylmethane skeleton is
useful as a steroid skeleton substitute and a multi-
template for NR ligands. It was also demonstrated
that DPPF-01 (6) and DPHK-01 (16) are highly spe-
cific for FXR and PPARa, respectively. It is well
documented that ligands with a steroidal skeleton of-
ten show cross reactivity with several NRs. For
example, lithocholic acid acts as a ligand for both
VDR and FXR.4,12 Application of a diphenylmethane
skeleton as a scaffold of NR ligands might overcome this
problem. In addition, NR ligands with a diphenylme-
thane skeleton might elicit selectivity on cell-type/co-fac-
tor(s) which is different from that of the physiological
ligands. In this preliminary study, we examined only the
diphenylmethane skeleton. However, it is very likely that
hetero-atom analogs of a diphenylmethane skeleton,
such as diphenylamine (as found in some RAR/RXR
ligands)7,24 and diphenylether (as found in thyroxine
hormones),25 would also available as substitutes for a
steroid skeleton. Further structural development studies
of the present ligands and synthesis of candidate NR
ligands based on the above ‘diphenyl X’ skeletons are
under way.
4. Makishima, M.; Lu, T. T.; Xie, W.; Whitfield, G. K.;
Domoto, H.; Evans, R. M.; Haussler, M. R.; Mangels-
dorf, D. J. Science 2002, 296, 1313.
5. Laudet, V.; Hanni, C.; Coll, J.; Catzeflis, F.; Stehelin, D.
EMBO J. 1992, 11, 1003.
6. Eyrolles, L.; Kawachi, E.; Matsushima, Y.; Nakajima, O.;
Kagechika, H.; Hashimoto, Y.; Shudo, K. Med. Chem.
Res. 1992, 2, 361.
7. Hashimoto, Y.; Miyachi, H. Bioorg. Med. Chem. 2005, 13,
5080.
8. Hosoda, S.; Tanatani, A.; Wakabayashi, K.; Nakano, Y.;
Miyachi, H.; Nagasawa, K.; Hashimoto, Y. Bioorg. Med.
Chem. Lett. 2005, 15, 4327.
9. Boehm, M. F.; Fitzgerald, P.; Zou, A.; Elgort, M. G.;
Bischoff, E. D.; Mere, L.; Mais, D. E.; Bissonnette, R. P.;
Heyman, R. A.; Nadzan, A. M.; Reichman, M.; Allegret-
to, E. A. Chem. Biol. 1999, 6, 265.
10. Hashimoto, Y.; Shudo, K. Cell Biol. Rev. 1991, 25, 209.
11. Hashimoto, Y. Cell Struct. Funct. 1991, 16, 113.
12. Makishima, M.; Okamoto, A. Y.; Repa, J. J.; Tu, H.;
Learned, R. M.; Luk, A.; Hull, M. V.; Lustig, K. D.;
Mangelsdorf, D. J.; Shan, B. Science 1999, 284, 1362.
13. Gebel, T.; Arand, M.; Oesch, F. FEBS Lett. 1992, 309,
37.
14. Pellicciari, R.; Constantino, G.; Fiorucci, S. J. Med.
Chem. 2005, 48, 5383.
15. Maloney, P. R.; Parks, D. J.; Haffner, C. D.; Fivush, A.
M.; Chandra, G.; Plunket, K. D.; Creech, K. L.; Moore,
L. B.; Wilson, J. G.; Lewis, M. C.; Jones, S. A.; Willson,
T. M. J. Med. Chem. 2000, 43, 2971.
Acknowledgments
16. DPPF-01 (6): MS (FAB) m/z 610 (M+H)+. 1H NMR
(500 MHz, CDCl3/d): 7.38 (d, J = 8.0 Hz, 2H), 7.33–7.31
(m, 1H), 6.93 (s, 2H), 6.88–6.83 (m, 2H), 6.63–6.59 (m,
2H), 4.68 (s, 2H), 4.64 (s, 2H), 3.40–3.30 (m, 1H), 2.22 (s,
3H), 2.01–1.97 (m, 7H), 1.41 (d, J = 6.7 Hz, 6H), 0.57 (t,
J = 7.3 Hz, 6H). HRMS (FAB, M+H+) calcd. for
C34H38Cl2NO5, 610.2127, found 610.2119.
The work described in this paper was partially
supported by Grants-in-Aid for Scientific Research from
The Ministry of Education, Culture, Sports, Science and
Technology, Japan, and the Japan Society for the
Promotion of Science. The authors are grateful to
ASKA Pharm. Co. Ltd (Japan) for the generous supply
of fabric acid.
17. DPPF-13 (7): MS (FAB) m/z 625 (M+H)+. 1H NMR
(500 MHz, CDCl3/d) (racemate): 7.38 (d, J = 8.0 Hz, 2H),
7.33–7.29 (m, 1H), 6.91–6.82 (m, 4H), 6.59 (d, J = 7.9 Hz,
1H), 6.53 (d, J = 7.9 Hz, 1H), 4.68 (s, 2H), 4.00 (br, 1H),
3.83–3.80 (m, 1H), 3.69–3.66 (m, 1H), 3.34–3.30 (m, 3H),
3.24–3.20 (m, 1H), 2.10 (s, 3H), 2.00–1.96 (m, 7H), 1.41 (d,
J = 6.7 Hz, 6H), 0.57 (t, J = 7.3 Hz, 6H). HRMS (FAB,
M+H+) calcd. for C35H43Cl2N2O4, 625.2600, found
625.2597.
References and notes
1. Chawla, A.; Repa, J. J.; Evans, R. M.; Mangelsdorf, D. J.
Science 2001, 294, 1866.
2. Evans, R. M. Science 1988, 240, 889.

3218
M. Kainuma et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3213–3218
18. Kasuga, J.; Makishima, M.; Hashimoto, Y.; Miyachi, H.
Bioorg. Med. Chem. Lett. 2006, 16, 554.
19. Nomura, M.; Tanase, T.; Ide, T.; Tsunoda, M.; Suzuki,
M.; Uchiki, H.; Murakami, K.; Miyachi, H. J. Med.
Chem. 2003, 46, 3581.
2H), 6.54 (m, 2H), 4.58 (m, 2H), 2.23 (m, 6H), 2.0 (m, 8H),
1.11 (m, 6H), 0.58 (m, 6H). HRMS (FAB, M+) calcd. for
C27H36O6, 456.2512, found 456.2509.
1
23. DPHK-01 (16): MS m/z 468 (M)+. H NMR (500 MHz,
CDCl3/d) (diastereomeric mixture): 7.06 (m, 2H), 6.95 (m,
2H), 6.58 (m, 2H), 4.58 (m, 2H), 2.24 (m, 6H), 2.17 (m,
4H), 2.02 (m, 4H), 1.47 (m, 6H), 1.09 (m, 6H). HRMS
(FAB, M+) calcd. for C28H36O6, 468.2512, found
468.2541.
20. Kasuga, J.; Hashimoto, Y.; Miyachi, H. Bioorg. Med.
Chem. Lett. 2006, 16, 771.
21. Nolte, R. T.; Wisely, G. B.; Westin, S.; Cobb, J. E.;
Lambert, M. H.; Kurokawa, R.; Rosenfeld, M. G.; Willson,
T. M.; Glass, C. K.; Milburn, M. V. Nature 1998, 395, 137.
22. DPPK-01 (15): MS m/z 456 (M)+. 1H NMR (500 MHz,
CDCl3/d) (diastereomeric mixture) 6.93 (m, 2H), 6.87 (m,
24. Hashimoto, Y. Mini-Rev. Med. Chem. 2002, 2, 543.
25. Sagi, T.; Kagechika, H.; Fukasawa, H.; Hashimoto, Y.;
Shudo, K. Chem. Pharm. Bull. 1996, 44, 1273.