Design and synthesis of fluorinated RXR modulators

Article abstract of DOI:10.1016/S0960-894X(03)00703-0

Fluorinated trienoic acid analogues of the RXR selective modulator 1 (LG101506) were synthesized, and tested for their ability to bind RXRα and activate RXR homo and heterodimers. Potency and efficacy were observed to be dependent upon the position of fluorination, and improvement in pharmacological profile was demonstrated in some cases.

Full text of DOI:10.1016/S0960-894X(03)00703-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3191–3195
Design and Synthesis of Fluorinated RXR Modulators
D. L. Gernert,b,* R. Ajamie,d R. A. Ardecky,a M. G. Bell,b M. D. Leibowitz,e
D. A. Mais,f C. M. Mapes,a P. Y. Michellys,a D. Rungta,f A. Reifel-Miller,c
J. S. Tyhonas,a N. Yumibed and T. A. Greseb
aDepartment of Medicinal Chemistry, Ligand Pharmaceuticals, Incorporated, San Diego, CA, USA
bDiscovery Chemistry Research, Lilly Research Laboratories, Indianapolis, IN, USA
cDivision of Endocrine Research, Lilly Research Laboratories, Indianapolis, IN, USA
dADME, Lilly Research Laboratories, Indianapolis, IN, USA
eDepartment of Pharmacology, Ligand Pharmaceuticals, Inc., San Diego, CA, USA
fDepartment of New Leads Discovery, Ligand Pharmaceuticals, Inc., San Diego, CA, USA
Received 14 May 2003; revised 14 May 2003; accepted 27 June 2003
Abstract—Fluorinated trienoic acid analogues of the RXR selective modulator 1 (LG101506) were synthesized, and tested for their
ability to bind RXRa and activate RXR homo and heterodimers. Potency and efficacy were observed to be dependent upon the
position of fluorination, and improvement in pharmacological profile was demonstrated in some cases.
# 2003 Elsevier Ltd. All rights reserved.
Introduction
The 5-fluorotrienoic acid derivative (2) was synthesized
as shown in Scheme 1. commercially available 2,4-di-t-
butylphenol was alkylated with 1-bromo-2,2-difluoro-
ethane and cesium carbonate in DMF and then treated
with NIS in refluxing DCM to afford the iodo com-
pound 10 in good yields. This compound then under-
went standard Sonagashira coupling with trimethylsilyl
acetylene. Treatment of 11 with dimethyl zinc and
Ni(acac)2 provided the corresponding vinyl silane 12 in
low yield. Treatment of 12 with iodine monochloride
provided the vinyl iodide 13 as a separable mixture of
isomers in a 2:1 ratio (Z/E). After separation, the Z
isomer, was coupled with the tin reagent (14),5,7 and
hydrolyzed to provide 5-fluorotrienoic acid, 2.
The retinoid X receptor (RXR) is known to regulate
gene transcription through the formation of hetero-
dimers with other nuclear receptors.1,2 In particular the
heterodimer formed with peroxisome proliferative-acti-
vated receptor gamma (PPARg) plays a major role in
the regulation of both glucose and lipid metabolism.3
Recently, selective RXR modulators such as
1
(LG101506), 6 (LG101392), and 8 have been described
as hypoglycemic agents4 (Fig. 1). As part of ongoing
SAR efforts, we endeavored to evaluate the pharmaco-
logical effects of placing a fluorine atom at different
positions of the triene system.
The 4-fluorotrienoic acid derivative (4) was prepared
from the aldehyde 15,8 as shown in Scheme 2. Con-
densation with fluoromethyl phenyl sulfone provided
the dienylsulfone 16 which underwent hydrostannyla-
tion to provide dienylstannane 17 in 38% overall yield,
after separation from isomeric material.7 Compound 17
then underwent smooth coupling with the 3-iodo-but-2-
enoic acid, to provide the desired 4-fluoro analogue 4 in
81% yield.6,9
Chemistry
Synthesis of the 6-fluoro analogue 3 has been recently
described.5 Syntheses of the 5-, 4-, and 2-fluoro posi-
tional isomers (2, 4, and 5) required the development of
alternative methods as described below.
The 2-fluorotrienoic acid derivative (5) was prepared
from the ester 18,10 as shown in Scheme 3, by four-step
*Corresponding author. Tel.+1-317-433-0430; fax:+1-317-276-2441;
e-mail: dgernert@lilly.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00703-0

3192
D. L. Gernert et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3191–3195
Figure 1. Fluorinated trienoic acids.
Scheme 1. 5-Fluorinated trienoic acid synthesis. Reagents and conditions: (a) (i) Cs2CO3, CF2CH2Br, DMF, 50 ꢀC (sealed tube); (ii) NIS, p-TSA,
DCM, reflux; 90%; (b) Pd(PPh3)4, CuI, TEA, TMS-acetylene, dioxane, 80 ꢀC, 57%; (c) Ni(acac)2, Me2Zn, THF/NMP(3:1); 0 ꢀC to rt, 7%; (d) ICl,
CCl4, 0 ꢀC, 39%; (e) (i) A, Pd(PPh3)2Cl2, DMF, 80 ꢀC; (ii) 1 N NaOH (aq), MeOH; 60 ꢀC, 35%; (f) ethyl 3-methyl-4-oxocrotonate, fluoromethyl
phenyl sulfone, diethyl chlorophosphate, LiHMDS, THF, À78 ꢀC; (g) SnBu3H, AIBN, benzene; reflux, 58% (desired isomer).

D. L. Gernert et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3191–3195
3193
Scheme 2. 4-Fluorinated trienoic acid synthesis. Reagents and conditions: (a) fluoromethyl phenyl sulfone, diethyl chlorophosphate, LiHMDS,
THF, À78 ꢀC; (b) SnBu3H, AIBN, benzene, reflux, 38% (desired isomer); (c) 3-iodo-but-2-enoic acid, Pd(PPh3)2Cl2, DMF, 80 ꢀC, 81%.
Scheme 3. 2-Fluorinated trienoic acid synthesis. Reagents and conditions: (a) (i) LAH, ether, 0 ꢀC; (ii) TPAP/NMO, DCM; (iii) MeMgBr, ether;
(iv) TPAP/NMO, DCM, 56%; (b) EtO2CCHFP(O)(OEt)2, NaH, DMF, 0 ꢀC to rt, 50%; (c) (i) LiOH, THF, MeOH; (ii) HPLC separation, 45%
(desired isomer).
Scheme 4. Trifluoromethyl (7) synthesis. Reagents and conditions: (a) 1,1,1-trifluoroacetone, piperidine, acetic acid, THF, 31%; (b) (i)
CH3C(O)CH2P(O)(OMe)2, NaH, ether, 84%; (ii) separation of isomers, 20% (desired isomer); (iii) LiOH, MeOH, 77%.
conversion to methyl ketone 19, followed by Horner–
Emmons condensation to provide the fluorinated ester,
20. Hydrolysis and HPLC separation of the 1:1 mixture
of isomers, provided the desired final product in 45%
yield.
provided the desired 7-trifluoromethyl product in low
yield after HPLC purification.
Biological Evaluation
In addition to our explorations of the fluorinated triene
analogues of 1, we examined the effects of replacing the
triene methyl substituents with trifluoromethyls. The 3-
trifluoromethyl analogue, 7, was synthesized from the
aldehyde 216 as shown in Scheme 4. Condensation with
1,1,1-trifluoro-propan-2-one followed by Horner–
Emmons condensation provided a 4:1 mixture of E/Z
isomers. After separation and hydrolysis the desired
Z,E,E isomer was isolated.
The binding of each compound to RXRa and the reti-
noic acid receptor (RARg), was characterized using
3[H]-9-cis-retinoic acid and 3[H]-all trans-retinoic acid
respectively and the data expressed as a Ki (Table 1).11
All of the fluorine replacements on the trienoic acid
portion showed good binding and were comparable to
the parent compound 1. Compounds 7 and 9 possessing
a 3- or 7-trifluoromethyl group, however, showed a
decrease in binding affinity relative to comparators 6
and 8. All of the compounds showed good binding
selectivity over RAR.
The 7-trifluoromethyl analogue, 9, was synthesized as
outlined in Scheme 5. The dianion of o-bromophenol
was quenched with ethyl trifluoroacetate, followed by
alkylation to provide 23. Sequential reduction, oxida-
tion, Horner–Emmons condensation, and hydrolysis
The RXR homodimer transcriptional activation profile
of each compound was determined in CV-1 cells efficacy
is reported relative to all-trans-retinoic acid.11 The

3194
D. L. Gernert et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3191–3195
Scheme 5. Trifluoromethyl (9) synthesis. Reagents and conditions: (a) n-BuLi, ethyl trifluoroacetate, ether, À78 ꢀC, 78%, (b) NaH, iodoethane,
DMF, >99%; (c) trimethylphosphonoacetate, NaH, DMF 0 ꢀC to rt; (d) DIBAL, ether; (e) (i) TPAP/NMO, DCM; (ii) triethyl-3-methyl-4-phos-
phonocrotonate, NaH, DMF, 0ꢀC to rt; (iii) LiOH, EtOH, reflux; (iv) HPLC separation, 0.62% (desired isomer).
Table 1. In vitro evaluation of RXR modulators in CV-1 cells
Compd
RXRa binding
Ki, nMa
RXRa
RXRa
RXRa/PPARg
RARg binding
Ki, nMd
RAR agonist
synergye
Agonist %
efficacy
Antagonist %
efficacy
Agonist %
efficacy
(EC50, nM)b
(EC50, nM)b
(EC50, nM)c
1
2
3
4
5
6
7
8
9
2.7
8.2
7.8
4.8
4
270 (259)
0
100 (8712.8)
87 (6.8)
100 (4842.1)
92 (3.4)
91 (3.3)
65 (805.8)
61 (6.4)
30 (38.0)
60 (3.1)
630 (249.1)
63 (6.4)
180 (143.4)
83 (2.7)
39 (9.2)
40 (359.4)
70 (2.0)
47 (167.4)
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
8566
1.4
8.9
1.8
2.5
1.5
1.2
1.3
2
2
107
2
2
1
10.6
3.8
5208
11.4
50.3
12
22
>10,000
7
Each data point represents the mean of two measurements.
aCalculated using [3H]-9-cis-RA.
bLGD1069 was used as reference.
cCalculated using 100 nM of BRL49653, efficacy relative to BRL49653.
dCalculated using [3H]-ATRA.
eCalculated using 3 nM of TTNPB.
replacement of fluorine at the 2- and 6- positions (com-
pounds 5 and 2, respectively) showed the same RXR
homodimer antagonist profile as 1. Surprisingly, the
compounds possessing a fluorine on the 4- and 5- posi-
tions of the trienoic acid (compounds 4 and 2, respec-
tively) are RXR homodimer agonists. However, while
their binding affinity for RXR is still low nanomolar
(Ki=8.2 and 4.8 nM, respectively), they have weak
potency for RXR homodimer activation in our co-
transfection assay (Table 1, EC50 >4800 nM). Both the
3- and 7-trifluoromethyl compounds 7 and 9 demon-
strated reduced efficacy and potency as expected in light
of their lower binding affinity.
the RXR/PPARg heterodimer but did not synergize
with TTNPB to enhance activation at the RXR/RAR
heterodimer (Table 1). The RXR heterodimer activation
profiles of the fluorinated analogues demonstrated
similar trends to those observed for the homodimers
above. The 2- and 6-F compounds (5 and 3) showed
strong activation of the RXR/PPARg heterodimer in
combination with BRL49653. The 4- and 5-F com-
pounds (4 and 2) also showed much lower activation,
and the trifluoromethyl compounds showed poor acti-
vation. The 4- and 5-fluoro analogues also showed
increased synergistic activation of the RXR/RAR het-
erodimer relative to 1, while the 2- and 6-fluoro analo-
gues maintained their selectivity for the RXR/PPARg
heterodimer.
We have previously demonstrated that RXR homo-
dimer antagonists can exhibit selective profiles of RXR
heterodimer activation in combination with sub-effica-
cious concentrations of PPARg (e.g., BRL49653) or
RAR (e.g., TTNPB) ligands.12 Compound 1 was shown
to synergize with BRL49653 to enhance activation at
In order to determine whether fluorination of the triene
would affect pharmacokinetic parameters of these RXR
modulators, we conducted an in vivo study of oral
exposure (Table 2). The 6-F analogue 3, was tested as a

D. L. Gernert et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3191–3195
3195
Table 2. In vivo evaluation of oral exposure of selected RXR
modulators
2. Schulman, I. G.; Crombie, D.; Bissonnette, R. P.; Cesario,
R.; Roegner, K.; Shao, G.; Heyman, R. A. Handbook of
Experimental Pharmacology 1999, 139, 215, and references
therein. Springer. Berlin.
3. (a) Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke,
B. R. J. Med. Chem. 2000, 44, 527, and references therein. (b)
Lehnard, J. M. Recept. Channels 2001, 27, 249. (c) Jones, A. B.
Med. Res. Rev. 2001, 21, 540. (d) Isserman, I.; Green, S. Nature
1990, 347, 645.
Compd
Dose
(mg/kg)
Oral AUC(0À6 h)
Tmax
(h)
Cmax
(mgÂh/mL)
(mgÂh/mL)
1
3
30
30
2.09Æ0.45
1
3
1.2Æ0.28
17.76Æ4.95
3.69Æ1.89
Data collected in male ICR mouse using a dose formulation of the free
acid in CMC/SLS/Povidone (30 mg/kg). Timepoints: 1, 3, 8 h (serial
sacrifice, n=3/time point).
4. Michellys, P. Y.; Ardecky, R. J.; Leibowitz, M. D.; Tyhonas,
J. S.; Mapes, C. M.; Chen, J.; Thompson, A. W.; Faulkner,
A. L.; Miller, A.; Grese, T. A.; Etgen, G.; Liu, S.; Ogilvie,
K. M.; Karanewsky, D. S.; Boehm, M. F. Abstracts of Papers,
223rd ACS National Meeting, Orlando, FL, USA, April 7–11,
2002. MEDI-005.
5. Michellys, P. Y.; Chen, J. H.; Grese, T. A.; Karanewsky,
D. S.; Leibowitz, M. D.; Liu, S.; Mais, D. A.; Mapes, C. M.;
Reifel-Miller, A.; Ogilvie, K. M.; Rungta, D.; Thompson,
A. W.; Tyhonas, J. S.; Yumibe N.; Boehm, M. F.; Ardecky,
R. J. Design and Synthesis of New RXR-Selective Modulators
with Improved Pharmacological Profile. In press
representative fluoro replacement compound and
showed a marked improvement in AUC relative to the
parent compound 1.
In summary, we have demonstrated that fluorine sub-
stitution at specific positions in the trienoic acid moiety
can maintain the desired in vitro profile exhibited by
compound 1. We have also demonstrated that the RXR
homodimer activity can be dramatically altered by the
position of the fluorine on the trienoic acid moiety.
While the 2- and 6-fluoro analogues (compounds 5 and
3) show the same RXR homodimer antagonist profile of
our reference compound 1, the 4- and 5-fluoro analo-
gues (compounds 4 and 2) are RXR homodimer ago-
nists. In some instances, fluorine introduction may also
result in improved pharmacokinetic properties for these
modulators. Further studies to examine the potential of
heterodimer selective RXR modulators for the treat-
ment of NIDDM and other metabolic disorders are
ongoing.
6. Chen, C.; Wilcoxen, K.; Zhu, Y.; Kim, K.; McCarthy, J. R.
J. Org. Chem. 1999, 64, 3476.
7. McCarthy, J. R.; Huber, E. W.; Le, T.; Laskovics, F. M.;
Matthews, P. Tetrahedron 1996, 52, 45.
8. Bell, M. G.; Grese, T. A.; Gernert, D. L.; Yumibe, N. P.;
Boehm, M. F.; Michellys, P. Y.; Ardecky, R. J.; Mapes, C. M.;
Chen, J. H.; Tyhonas, J. S. PCT Int. WO2002072528.
9. Urdaneta, F.; Neudo, A.; Herrera, Julio C.; Rodriguez-
Arias, Eloy Nouel Ciencia (Maracaibo) 1995, 3, 41.
10. Ardecky, R. J.; Boehm, M. F.; Faulkner, A. L.; Hamann,
L. G.; Jones, T. K.; Mapes, C. M.; Michellys, P.; Tyhonas,
J. S.; Thompson, A. J.; Chen, J. PCT Int. WO0119770
11. Boehm, M. F.; Zhang, L.; Badea, B. A.; White, S. K.;
Mais, D. E.; Berger, E.; Suto, C. M.; Goldman, M. E.; Hey-
man, R. A. J. Med. Chem. 1994, 37, 2930.
12. Michellys, P. Y.; Ardecky, R. J.; Chen, J. H.; Crombie, D.
L.; Etgen, G.; Faul, M.; Faulkner, A. L.; Grese, T. A.; Heyman,
R. A.; Karanewsky, D. S.; Klausing, K.; Leibowitz, M. D.;
Liu, S.; Mais, D. A.; Mapes, C. M.; Marschke, K.; Reifel-
Miller, A.; Ogilvie, K. M.; Rungta, D.; Thompson, A. W.;
Tyhonas, J. S.; Boehm, M. F. J. Med. Chem., 2003, 46, 2683.
References and Notes
1. Mangelsdorf, D. J.; Unesono, K.; Evans, R. M. The Reti-
noids: Biology, Chemistry, and Medicine, 2nd ed.; Raven: New
York, 1994; p 319.