Conformationally constrained farnesoid X receptor (FXR) agonists: Heteroaryl replacements of the naphthalene

Article abstract of DOI:10.1016/j.bmcl.2010.12.089

To improve on the drug properties of GSK8062 1b, a series of heteroaryl bicyclic naphthalene replacements were prepared. The quinoline 1c was an equipotent FXR agonist with improved drug developability parameters relative to 1b. In addition, analog 1c lowered body weight gain and serum glucose in a DIO mouse model of diabetes.

Full text of DOI:10.1016/j.bmcl.2010.12.089

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1206–1213
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Conformationally constrained farnesoid X receptor (FXR) agonists:
Heteroaryl replacements of the naphthalene
Jonathan Y. Bass a, , Justin A. Caravella b,à, Lihong Chen c, Katrina L. Creech d, David N. Deaton a,
,
⇑
,
Kevin P. Madauss b, Harry B. Marr e, Robert B. McFadyen a, Aaron B. Miller b, Wendy Y. Mills a,§
Frank Navas III a,–, Derek J. Parks f, Terrence L. Smalley Jr. a, Paul K. Spearing a, Dan Todd g,
Shawn P. Williams b, G. Bruce Wisely f
a Department of Medicinal Chemistry, GlaxoSmithKline, Research Triangle Park, NC 27709, USA
b Department of Molecular Discovery Research, Computational and Structural Chemistry Research, GlaxoSmithKline, Research Triangle Park, NC 27709, USA
c Department of Molecular Pharmacology, GlaxoSmithKline, Research Triangle Park, NC 27709, USA
d Department of Molecular Discovery Research, Screening and Compound Profiling, GlaxoSmithKline, Research Triangle Park, NC 27709, USA
e Department of Drug Metabolism and Pharmacokinetics, GlaxoSmithKline, Research Triangle Park, NC 27709, USA
f Department of Molecular Discovery Research, Biological Reagents and Assay Development, GlaxoSmithKline, Research Triangle Park, NC, 27709, USA
g Department of Pharmaceutical Development, Physical Properties and Developability, GlaxoSmithKline, Research Triangle Park, NC 27709, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
To improve on the drug properties of GSK8062 1b, a series of heteroaryl bicyclic naphthalene replace-
ments were prepared. The quinoline 1c was an equipotent FXR agonist with improved drug developabil-
ity parameters relative to 1b. In addition, analog 1c lowered body weight gain and serum glucose in a DIO
mouse model of diabetes.
Received 8 October 2010
Revised 15 December 2010
Accepted 16 December 2010
Available online 23 December 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Farnesoid X receptor
FXR
Nuclear receptor modulator
FXR X-ray co-crystal structure
GW 4064
GSK8062
Bile acid receptor
NR1H4 GSK2324
Bile acids are synthesized from cholesterol and serve as the pri-
mary route of elimination of cholesterol from mammals. Bile acids
and their conjugates are amphipathic molecules that can form
mixed micelles, facilitating the dissolution and absorption of die-
tary lipids and fat-soluble vitamins. Since bile acids can disrupt cell
bilayers, their concentrations must be tightly controlled and nature
has developed several regulatory mechanisms to monitor and ad-
just bile acid levels.1,2 These signalling pathways include activation
of the c-Jun N-terminal kinase (JNK) pathway3, signal transduction
of the 7-transmembrane receptor family member bile acid receptor
TGR5 (GPBAR1, M-BAR)4, and transcription mediated through the
nuclear receptor family member farnesoid
X receptor (FXR,
NR1H4, bile acid receptor (BAR)).5
Besides regulating bile acid homeostasis, FXR also has additional
functions in metabolism. FXR is highly expressed in tissues impli-
cated in bile acid recirculation and elimination, including liver, gall
bladder, intestine,andkidney, aswellashaving significantexpression
inadrenalglands.6,7 The bile acid chenodeoxycholic acid(CDCA)isthe
natural ligand with the highest affinity for FXR, with cholic acid (CA),
deoxycholic acid (DCA), and lithocholic acid (LCA) also having affinity
for the receptor. Coupled to bile acids’ roles in the dietary pathway,
the signalling mechanisms of these molecules have also evolved to
regulate lipid8–10 and glucose metabolism,11–13 attenuate intestinal
infection,14–16 assist in xenobiotic elimination,17 and influence
liver regeneration18 through FXR. With FXR’s multiple physiological
roles, FXR modulators could be utilized for the treatment of
cholestasis,19–21 liver fibrosis,22–25 liver cancer,26,27 steatohepatitis,28
⇑
Corresponding author. Tel.: +1 919 483 6270.
E-mail address: david.n.deaton@gsk.com (D.N. Deaton).
Present address: Department of Chemistry, University of California, Irvine, CA
92697, USA.
à
Present address: Biogen Idec, 14 Cambridge Center, Cambridge, MA 02142, USA.
Present address: United States Army Research Laboratory, Research Triangle Park,
§
NC 27709, USA.
–
Present Address: Department of Organic and Medicinal Chemistry, Research
Triangle Institute, RTP, NC 27709, USA.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.12.089

J. Y. Bass et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1206–1213
1207
atherosclerosis,29–32 cholesterol gallstone disease,33 diabetes,13 obes-
R2
R3
X2
X3
ity,34 metabolic syndrome,35 and inflammatory bowel disease.14,15
Employing an iterative combinatorial library synthesis and
screening approach versus orphan nuclear receptors, Glaxo
Wellcome scientists discovered the potent and selective FXR
agonist GW 4064A 1a.9 This molecule, along with other FXR
ligands36–40 including 6-ethyl-CDCA41 and fexaramine,42 was
instrumental in probing the physiological roles of FXR. Recently,
Exelixis and Wyeth researchers have reported the discovery of
another chemical class of FXR modulators, exemplified by
WAY362450 (XL335).43 Although a potent FXR tool compound,
GW 4064 1a had several limitations as a drug candidate, including
UV light instability, a questionable stilbene pharmacophore, and
limited oral bioavailability. Thus, GlaxoSmithKline medicinal chem-
ists sought to improve on this ligand. Structure–activity relationship
(SAR) studies of the isoxazole substituents39 and conformational
constraints of the benzoic acid37 and internal aryl rings38 led to the
identification of GSK8062 1b (TT EC50 = 68 nM, %Max = 104), a po-
tent FXR agonist with improved light stability, good oral bioavail-
ability in dog and monkey, and an acceptable in vitro safety
profile.37 With the inherent risk associated with drug discovery, a
replacement molecule was sought to mitigate development risks
in the event that GSK8062 1b could not be advanced. In addition to
desiring a back-up molecule with the desirable properties of
GSK8062 1b, GlaxoSmithKline desired a drug candidate that had
improvements relative to GSK8062 1b. First, the limited aqueous
solubility of GSK8062 1b at neutral pH values necessitated the use
of non-standard formulations to obtain adequate exposure during
safety assessment studies. Second, the limited oral exposure of
GSK8062 1b in rats, resulting from its high clearance, required the
use of mice for rodent species safety assessment studies. Mice, be-
cause oftheir sizeand physiology, are more difficultto correctlydose
during safety assessment studies. Third, in an exploratory salt
screening study employing frequently used counterions such as so-
dium or potassium, difficulty was encountered in discovering a suit-
able crystalline salt version of GSK8062 1b, necessitating the use of a
less conventional salt version. Thus a follow-on drug candidate was
desired with similar/better potency and selectivity for FXR with im-
proved water solubility, better rat pharmacokinetics, and a more
conventional solid formulation. Believing that improved aqueous
solubility might increase rat exposure, a strategy was devised to
replace the naphthyl ring of GSK8062 1b with heteroaryl bicyclic
systems to improve hydrophilicity.
R1 X1
R6O
R3
OR6
B
R7
Cl
Cl
O
N
R7
R2
R3
X2
X3
R2
R5
R8
R1 X1
X2
5a
R8
OH
R1 X1
X3
Cl
3a-3d
O
N
Cl
b
c, d
R5
R5
R4
O
2a-2e, 2j-2l
OH
Cl
1c-1f, 1k-1n
a
2d
2e
4a-4h
Scheme 1. Reagents and conditions: (a) 2d, NEt3, Tf2O,CH2Cl2, (b) 2a–2c, 2e, or
2j–2l, 3a–3d, Na2CO3 or K3PO4, Pd(PPh3)4, or Pd(OAc)2 and PPh3, DME or dioxane,
H2O, 60–80 °C; MeSO3H, MeOH, ;", 40–98%; (c) 5a, Cs2CO3 or K2CO3, DMF, 50–70 °C,
22–70%; (d) NaOH, THF, MeOH or EtOH, H2O, 25–60 °C or microwave at 90–100 °C,
79–99%.
R2
R1
X3 R3
X2 X4
O
B
X1
O
O
R2
X2 X3
X4 R3
6a
Cl
O
R1 X1
O
N
O
N
Cl
Cl
a, b
Cl
Br
2f-2i, 2m-2s
1g-1j, 1o-1v
c, a-b
1o
1q
Scheme 2. Reagents and conditions: (a) 2f–2i, or 2m–2s, 6a, K3PO4, Pd(OAc)2 and
PPh3, dioxane, H2O, 60–80 °C, 22–91%; (b) NaOH, THF, MeOH or EtOH, H2O,
25–60 °C or microwave at 90–100 °C, 8–99%; (c) 1o, 2,6-lutidine, DMAP, Tf2O,
ClCH2CH2Cl, 60 °C, 32%.
to provide the desired carboxylic acids 1g–1j, 1o–1p, and 1r–1v
as illustrated in Scheme 2. The carboxylic acid 1q (R2 = CO2H,
R3 = Et) was prepared from the phenol 1o (R2 = CO2Me, R3 = OH).
First, conversion of the phenol to the triflate with trifluorometh-
anesulfonic anhydride, then Suzuki coupling with ethyl boronic
acid, followed by ester hydrolysis yielded the carboxylic acid 1q
(R2 = CO2H, R3 = Et).
Some of the heteroaryl esters, such as the quinolines 2a, 2c, and
2m were commercially available, while others such as quinoline
2b,45 and isoquinoline 2d46 were known in the literature. The
remaining heteroaryl esters were prepared as depicted in Schemes
3–8. The isoquinoline 2f was prepared from the phenethyl amine
9a as shown in Scheme 3. Sulfonation of the amine 9a with benzene-
sulfonyl chloride, followed by alkylation with the chloride 8a,
prepared from the thioether 7a, and tin(IV) chloride catalyzed
Friedel–Crafts cyclization gave the regioisomeric tetrahydroiso-
quinolines 10a and 10b. Subsequent base catalyzed elimination/
aromatization of the desired major product 10b afforded the iso-
quinoline 2f.
O
O
N
N
O
O
Cl
Cl
Cl
Cl
Cl
1a
1b
HO
O
HO
O
The heteroaryl carboxylic acid analogs 1c–1v were preparedby one of
two routes as depicted in Schemes 1 and 2 with the full experimental
details provided in the Supplementary data. The quinolines 2a–2c,
and 2j–2l and the isoquinoline 2e were coupled to the appropriate
boronates 3a–3d, employing the Suzuki protocol to provide the phe-
nols 4a–4h as shown in Scheme 1. Then, O-alkylation of the phenols
4a–4h with the alkyl chloride 5a37 produced aryl ethers, which upon
base-catalyzed hydrolysis of the esters afforded the corresponding
carboxylic acids 1c–1f and 1k–1n.
Alternatively, palladium-mediated Suzuki coupling of the het-
eroaryl bromides of the isoquinoline 2f, quinazoline 2g, quinoxa-
line 2h, 1,2,4-benzotriazene 2i, or quinolines 2m–2s with the
aryl boronate 6a44 gave the corresponding biaryl derivatives,
whose esters were hydrolyzed in a similar manner to Scheme 1,
The quinazoline 2g was prepared as shown in Scheme 4. The
aniline 11a was acylated with ethyl chlorooxoacetate to give the
amide, which was heated at reflux with ammonium acetate to pro-
vide the quinazoline 2g.

1208
J. Y. Bass et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1206–1213
Ph
N
Ph
N
The 1,2,4-benzotriazene 2i was synthesized as depicted in
O
O
O
O
S
S
Scheme 6. The aniline 13a was diazotized, followed by trapping with
ethyl 2-chloroacetoacetate to give the hydrazonoyl chloride 14a
(R = Cl). Then, displacement of the chloride with ammonia afforded
the imidohydrazide 14a (R = NH2). Reduction of the nitro moietyand
cyclization then provided the desired 1,2,4-benzotriazene 2i.
The quinolines 2j, 2k, and 2n were prepared as shown in Scheme
7. One pot in situ tin(II) chloride reduction of the nitroarenes
15a–15c, followed by zinc(II) chloride catalyzed imine formation
with ethyl pyruvate or ethyl 2-oxobutanoate, and cyclization
provided the quinolines 2j, 2k, and 2n.
H2N
O
O
EtO
EtO
b, c
Br
10a
10b
9a
Br
Br
d
O
N
EtO
Cl
The quinolines 2o–2s were prepared as shown in Scheme 8.
Silver carbonate catalyzed alkylation of phenol 2m with various
alkyl iodides yielded the desired quinolines 2o–2s.
O
O
a
S
S
7a
8a
O
2f
O
Br
The structure–activity relationships of the heteroaryl bicyclic
analogs are depicted in Table 1. Knowing from the FXR/GSK8062
1b X-ray co-crystal structure37 that the carboxylic acid moiety
was near the solvent front, analogs that insert a nitrogen atom in
the proximal ring of the naphthalene might be permissible with
minimal desolvation cost. Also, since the naphthyl analogs, exem-
Scheme 3. Reagents and conditions: (a) N-chlorosuccinimide, CCl4, 0–25 °C, 43%;
(b) PhSO2Cl, iPr2NEt, CH2Cl2, 0–25 °C, 95%; (c) 8a, SnCl4, ClCH2CH2Cl, ";, 79% and
19%; (d) DBU, PhMe, 59%.
Br
N
a, b
Br
plified by 1b, tolerated both a- (TT EC50 = 68 nM) and b-carboxylic
O
O
acids (TT EC50 = 45 nM),37 proximal aza-analogs were prepared
with the acid functionality at either position of the ring.
N
2g
H2N
11a
O
As exemplified by the quinoline 1c (TT EC50 = 50 nM) and
isoquinoline 1f (TT EC50 = 77 nM), the nitrogen atom was well
tolerated at the 1- and 3-positions of the aza-naphthalene
template, as these analogs have similar potency in both the fluores-
cent resonance energy transfer (FRET) assay for recruitment of the
co-activator peptide of SRC-1 and the transient transfection (TT) as-
say for induction of a luciferase reporter gene. Both quinoline 1c (TT
%Max = 102) and isoquinoline 1f (TT %Max = 86) are full FXR ago-
nists in these assays. Thus, either the cost to desolvate the additional
nitrogen atom of these analogs must be small, or the lone pair on the
nitrogen atom participates in a direct or water bridging (vide infra)
hydrogen bond to the protein, where the binding energy from this
putative hydrogen bond helps compensate for the desolvation cost.
In contrast, the nitrogen atom was less well accommodated at
the 2- and 4-positions of the aza-naphthalene template as shown
by quinolines 1d (TT EC50 = 560 nM) and 1e (TT EC50 = 260 nM)
and isoquinoline 1g (TT EC50 = 190 nM), which, although full ago-
nists, exhibit diminished potency relative to their corresponding
naphthyl analogs. In addition to the desolvation cost, inferences
from X-ray structures37 suggest that the 4-aza analogs 1d and 1e
place the nitrogen lone pair near a hydrophobic part of the protein,
which is probably detrimental to binding. The carboxylic acids of
quinoline 1c and isoquinoline 1f prefer a planar conformation that
Scheme 4. Reagents and conditions: (a) EtO2CCOCl, pyridine, CH2Cl2, 0–25 °C, 85%;
(b) NH4OAc, HOAc, ";, 49%.
H2N
H2N
Br
N
N
Br
a
O
12a
2h
O
Scheme 5. Reagents and conditions: (a) EtO2CCOCH2Br, NMP, 25 °C, 13%.
O
O
O
O
O2N
NH2
Br
R
N
N
NH
N
N
O2N
a, b
c
Br
13a
14a
2i
Br
Scheme 6. Reagents and conditions: (a) NaNO2, HCl, MeOH, H2O; EtO2CCHClC-
OCH3, 0–25 °C, 63%; (b) NH3, dioxane, 25 °C, 83%; (c) Fe, HCl, HOAc, H2O, 25 °C, 41%.
R1
R1
maximizes aryl and carbonyl
p-bond resonance energy stabiliza-
Br
Br
O
a
tion and allows an intramolecular hydrogen bond between the car-
boxylic acid and the ring nitrogen. In contrast, the 2-aza analog 1g
likely has its carboxylic acid moiety twisted out of plain, similar to
the carboxylic acid of GSK8062 1b in its X-ray co-crystal struc-
ture,37 to avoid a syn-pentane interaction with the 8-hydrogen.
The lack of this intramolecular hydrogen bond likely leads to stron-
ger solvation of the isoquinoline nitrogen of 1g. Thus, a higher
desolvation penalty may result upon binding, leading to reduced
potency for isoquinoline 1g.
Inserting more than one nitrogen into the naphthalene tem-
plate, as in the quinazoline 1h (TT EC50 = 1000 nM), the quinoxa-
line 1i (TT EC50 = 690 nM), or the 1,2,4-benzotriazine 1j (TT
EC50 = >10,000 nM), resulted in a significant decrease in transient
transfection potency. Since activity in the cell free FRET assay
was maintained, the decrease in the TT assay may reflect limited
cell penetration in this assay. Alternatively, the African green mon-
key CV-1 cells used in the TT assay likely contain additional coac-
tivators and corepressors besides SRC-1, which may alter potency/
efficacy relative to the cell free FRET assay.
O
O2N
15a-15c
R2
N
R2
2j-2k, 2n
O
Scheme 7. Reagents and conditions: (a) 15a–15c, ZnCl2, SnCl2, 4Å molecular sieves,
CH3CH2COCO2Et or CH3COCO2Et, EtOH, 70 °C, 3–20%.
R1
OH
N
O
Br
Br
a
O
O
N
2m
2o-2s
O
O
Scheme 8. Reagents and conditions: (a) Ag2CO3, R1I, THF, 33–57%.
The quinoxaline 2h was synthesized as illustrated in Scheme 5.
Alkylation of the diamine 12a with ethyl bromopyruvate, followed
by intramolecular condensation yielded the quinoxaline 2h.

J. Y. Bass et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1206–1213
1209
Table 1
Activation of human FXR
R3
O
N
O
Cl
Cl
R
R2
1a-1v
#
R
R1
FXR
FXR
TT
FRET
EC50 nMa (%Maxb)
EC50 nMc (%Maxb)
R2 = H
R3 = H
59
(100)
65
(100)
1a
HO
O
O
R2 = H
R3 = H
87
(134)
68
(104)
1b
1c
1d
HO
R2 = H
R3 = H
120
(105)
50
(102)
HO
N
N
O
N
R2 = H
R3 = H
340
(99)
560
(85)
HO
HO
O
N
R2 = H
R3 = H
200
(87)
260
(88)
1e
1f
O
O
R2 = H
R3 = H
120
(83)
77
(86)
HO
R2 = H
R3 = H
380
(92)
190
(102)
N
1g
1h
HO
O
N
R2 = H
R3 = H
140
(102)
1000
(127)
HO
HO
HO
HO
HO
N
N
N
N
N
O
O
O
O
O
R2 = H
R3 = H
150
(96)
690
(109)
1i
1j
1k
1l
N
R2 = H
R3 = H
360
(100)
>10,000
(—)
R2 = H
R3 = H
140
(107)
38
(101)
N
N
R2 = H
R3 = H
240
(115)
110
(107)
F
(continued on next page)

1210
J. Y. Bass et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1206–1213
Table 1 (continued)
#
R
R1
FXR
FXR
FRET
TT
EC50 nMa (%Maxb)
EC50 nMc (%Maxb)
R2 = Me
R3 = H
250
(87)
52
(91)
HO
HO
1m
1n
N
O
O
R2 = H
110
(100)
44
(94)
R3 = Me
N
R1
R1 = OH
R2 = H
R3 = H
51
(91)
>10,000
(—)
1o
1p
1q
1r
1s
1t
HO
HO
HO
HO
HO
HO
HO
HO
N
O
O
O
O
O
O
O
O
R1
R1 = Me
R2 = H
R3 = H
81
(108)
24
(107)
N
R1
R1 = Et
R2 = H
R3 = H
75
(106)
32
(109)
N
R1
R1 = OMe
R2 = H
58
(113)
17
(107)
R3 = H
N
R1
R1 = OEt
R2 = H
52
(99)
14
(84)
R3 = H
N
R1
R1 = OiPr
R2 = H
40
(113)
14
(103)
R3 = H
N
R1
R1 = OnPr
R2 = H
59
(113)
22
(93)
1u
R3 = H
N
R1
R1 = OiBu
R2 = H
78
(105)
81
(93)
1v
R3 = H
N
a
FXR ligand seeking assay measuring ligand-mediated interaction of the SRC-1 peptide (B-CPSSHSSLTERHKILHRLLQEGSPS-CONH2) with the FXR 237–472LBD, using 5 nM
biotinylated human FXR LBD coupled to 5 nM allophycocyanin-labeled streptavidin and 10 nM biotinylated human SRC-1 coupled to 5 nM Europium-labeled streptavidin as
reagents in 10 mM DTT, 0.1 g/L BSA, 50 mM NaF, 50 mM MOPS, 1 mM EDTA, and 50 lM CHAPS, at pH 7.5. The EC50 values are the mean of at least two assays.
b
Maximum percent efficacy of the test compound relative to FXR activation via GW 4064 1a.
FXR transient transfection assay measuring the ligand-mediated luminescence resulting from FXR-induced transcription of a luciferase reporter. FXR and the luciferase
c
reporter genes are transfected into African green monkey CV-1 kidney cells, then treated with test compound. The EC50 values are the mean of at least two assays.
The effect of substitution on the potency/efficacy of the aza-
naphthalenes was also explored. The 3-methyl group of quinoline
1k should force the carboxylic acid to twist out of plane to avoid
a syn-pentane interaction, possibly enhancing its interaction with
the guanidine moiety of 331Arg, while maybe increasing the desolv-
ation cost of the quinoline nitrogen due to the loss of the internal
hydrogen bond. Quinoline 1k (TT EC50 = 38 nM) exhibited similar
potency as 1c, suggesting that any gain from a stronger interaction
with 331Arg was likely offset by a larger desolvation cost.
Although small in size, the 3-fluoro group of quinoline 1l (TT
EC50 = 110 nM) could alter the rotamer conformation between
the quinoline and aryl rings affecting receptor binding, yet the drop

J. Y. Bass et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1206–1213
1211
in potency was negligible. Similarly, the 2-methyl substituent on
the phenyl ring of quinoline 1m (TT EC50 = 52 nM) could also alter
rotamer population energies between the quinoline and aryl rings,
but had no effect on receptor potency relative to 1c. Furthermore,
substitution at the 3-position of the phenyl ring with a methyl
group, as in quinoline 1n (TT EC50 = 44 nM), also maintained recep-
tor affinity, despite potentially affecting the conformation of the
oxymethylene linker. Likely, in these three cases, the receptor
could tolerate the added substituent with minimal changes in
conformation.
Similar to the 1,2,4-benzotriazine 1j, the 4-hydroxyquinoline 1o
(TT EC50 = >10,000 nM) was inactive in the transient transfection
assay, but exhibited good potency in the cell free FRET assay (FRET
EC50 = 51 nM), suggesting poor cell permeability. In contrast,
4-alkyl- and 4-alkoxy-substituents, exemplified by quinolines
1p–1v, were active in the whole cell TT assay. The methylquinoline
1p (TT EC50 = 24 nM), the ethylquinoline 1q (TT EC50 = 32 nM), the
methoxyquinoline 1r (TT EC50 = 17 nM) and the n-propoxyquino-
line 1u (TT EC50 = 22 nM) were potent, full FXR agonists. Whereas,
the ethoxyquinoline 1s (TT EC50 = 14 nM) and the iso-propoxyquin-
oline 1t (TT EC50 = 14 nM) appeared to be optimal ligands and were
modestly more potent than starting quinoline 1c, the iso-
butoxyquinoline 1v (TT EC50 = 81 nM) was less potent and its
substituent might be approaching the limit of accommodation by
the protein. The X-ray co-crystal structure of GSK8062 1b37 shows
that this region of the receptor is hydrophobic, and contains little
room for 4-position substitution on the agonist. Clearly, the ligand
binding domain is capable of moving to allow binding by these
analogs. Although these 4-substituted analogs resulted in modest
gains in potency, they came at the cost of reduced water solubility.
The X-ray co-crystal structures of quinoline 1c and isoquinoline
1f with FXR were obtained and are depicted in Figures 1 and 2.
They further reinforce the receptor recognition elements necessary
for FXR gene transcription in this chemical series. In both struc-
tures, the isoxazole resides adjacent to 454Trp and 447His on the
C-terminal end of helix 10, with 469Trp located on activation func-
tion 2 region (helix 12), making an edge to face stacking interaction
with the isoxazole. Further stabilization of the active protein con-
former for recruitment of co-activator proteins for gene transcrip-
tion is provided by the iso-propyl group, which occupies a pocket
formed by 284Phe, 287Leu, 454Trp, and 461Phe. In addition, the di-
ortho substitution on the 2,6-dichlorophenyl ring makes the aryl
ring’s lowest energy conformation the rotamer that is orthogonal
to the isoxazole ring, minimizing entropic loss upon binding, while
maximizing enthalpic interactions with the ligand binding domain
in this region of the protein. While the above interactions are likely
crucial to stabilize the hydrophobic core of the receptor, leading to
an active conformation of the receptor capable of shedding
co-repressors and recruiting co-activators to induce gene tran-
scription, the phenyl scaffold connecting the isoxazole to the
heteroaryl bicycle is probably enhancing potency via hydrophobic
interactions with the receptor, as well as orienting the two termi-
nal pharmacophores for optimal interactions with the protein.
In both structures, the carboxylic acids are co-planar with their
heteroaryl ring, similar to the GW 4064 1a X-ray co-crystal struc-
ture,37 but in contrast to the GSK8062 1b X-ray co-crystal structure
in which the carboxylic acid is orthogonal to the naphthalene ring,
as shown in Figure 1. In both the quinoline 1c and the isoquinoline
1f X-ray cocrystal structures, one NH2 of the guanidine group of
331Arg on helix 5 of FXR forms electrostatic interactions with one
of the oxygen atoms of the respective carboxylic acids. In addition
to hydrophobic contacts with the protein, the quinoline nitrogen of
agonist 1c forms a water-mediated hydrogen bond with the same
NH2 that interacts with the quinoline carboxylate as well as with
Figure 1. Ligand binding domain of the X-ray co-crystal structure of quinoline 1c
(ligand 1c carbons colored green) complexed with FXR. The FXR carbons from the
co-crystal structure with 1c are colored cyan. The semi-transparent grey surface
represents the molecular surface, while hydrogen bonds are depicted as yellow
dashed lines. GSK8062 1b (ligand 1b carbons colored magenta) has been super-
imposed into this structure based on its published X-ray co-crystal structure. The
coordinates have been deposited in the Brookhaven Protein Data Bank (1b PDB code
3DCU, 1c PDB Code 3P89). This figure was generated using PYMOL version 1.3
(www.pymol.org).
interaction of the GSK8062 1b structure, the through water
hydrogen bond between the quinoline nitrogen and the side chain
of 331Arg likely offsets any reduced binding energy between the
acid and the guanidine.
As shown in Figure 2, in the isoquinoline 1f X-ray co-crystal
structure, the isoquinoline ring is rotated 180° relative to the quin-
oline ring of agonist 1c, while the carboxylic acids are only slightly
shifted, maintaining the electrostatic interaction with 331Arg. This
places the isoquinoline nitrogen of 1f in a relatively similar position
as the quinoline nitrogen of 1c. Although no water mediated hydro-
gen bond to the guanidine of 331Arg was detected in this lower res-
olution structure (1f 2.7 Å versus 1c 2.35 Å), one might infer that
Figure 2. Ligand binding domain of the X-ray co-crystal structure of isoquinoline 1f
(ligand 1f carbons colored green) complexed with FXR. The FXR carbons from the
co-crystal structure with 1f are colored cyan. The semi-transparent grey surface
represents the molecular surface, while hydrogen bonds are depicted as yellow
dashed lines. Quinoline 1c (ligand 1c carbons colored magenta) has been super-
imposed into this structure based on its X-ray co-crystal structure above. The
coordinates have been deposited in the Brookhaven Protein Data Bank (1c PDB code
3P89, 1f PDB Code 3P88). This figure was generated using PYMOL version 1.3
(www.pymol.org).
the
of the acid moiety is probably less favored than the orthogonal
e
-NH of 331Arg’s guanidine. Although the planar interaction

1212
J. Y. Bass et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1206–1213
Finally, the 340Leu iso-butyl group is visible in the higher resolution
Table 2
Pharmacokinetics of FXR agonists
structure of quinoline 1c but not in the isoquinoline 1f structure.
As shown in Table 2, quinoline 1c and isoquinoline 1f were
profiled in four species to ascertain their pharmacokinetic param-
eters. Quinoline 1c had low clearances in the mouse, beagle dog,
and cynomolgus monkey, with a medium clearance in the Spra-
gue–Dawley rat, while isoquinoline 1f had low clearances in the
rat and dog, with medium clearances in the mouse and monkey.
Although the clearance for quinoline 1c was higher in the rat, its
corresponding higher volume of distribution resulted in a longer
terminal half-life than in the other species due to their lower vol-
umes of distribution. Quinoline 1c had a good oral bioavailability
in the mouse, which decreased in higher species progressively to
an oral bioavailability of 15% in the monkey. This class of
compounds exhibits high liver to plasma ratios, thus high first pass
clearances may be limiting oral bioavailabilities.36 Isoquinoline 1f
generally exhibited higher volumes than quinoline 1c, resulting
in longer terminal half-lives, but its decreased oral bioavailabilities
and po dose-normalized areas under the curve precluded it from
further development.
a
c
#
Species
t1/2
Clb
VSS
Fd
(%)
DNAUCe
(ng h kg/mL mg)
(min)
(mL/min/kg)
(mL/kg)
1c
Mouse
Rat
Beagle
Cyno
84
170
110
120
23
32
4.4
3.1
1600
3400
310
42
23
23
15
310
130
1260
850
150
1f
Mouse
Rat
Beagle
Cyno
250
88
230
620
29
18
8.2
11
5100
1800
2000
3100
6.3
30
15
36
300
320
40
2.5
a
t1/2 is the iv terminal half-life dosed as a solution. All in vivo pharmacokinetic
values are the mean at least two experiments.
b
Cl is the iv total clearance.
VSS is the iv steady state volume of distribution.
F is the oral bioavailability dosed as a solution.
DNAUC is the oral dose normalized area under the curve.
c
d
e
Table 3
Diet-induced obese mouse study—body weight (g)a
Quinoline 1c compared favorably to GSK8062 1b in its pharma-
cokinetic parameters,37 with a similar oral dose normalized area
under the curve in monkey (1b DNAUC = 1130 ng h kg/mL mg vs
1c DNAUC = 850 ng h kg/mL mg) and a 5-fold better oral exposure
in the rat (1b DNAUC = 23 ng h kg/mL mg vs 1c DNAUC =
130 ng h kg/mL mg). Furthermore, quinoline 1c’s Cyp 450 inhibition
profile in pooled human liver microsomal assays was reasonable
(1A2 IC50 = 21,000 nM, 2C9 IC50 = 4400 nM, 2C19 IC50 = >33,000 nM,
2D6 IC50 = >33,000 nM, 3A4 [midazolam] IC50 = >33,000 nM, 3A4
[atorvastatin] IC50 = 5100 nM, 3A4 [nifedipine] IC50 = 11,000 nM).
Similar to GSK8062 1b, quinoline 1c was at least 100-fold selective
for FXR versus a panel of closely homologous nuclear receptors, con-
Diet
Treatment Vehicle
NCb
HFDc
Vehicle
HFDc
10 mg/kg
HFDc
30 mg/kg
HFDc
100 mg/kg
Baseline
Week 1
Week 2
Week 3
Week 4
26.7 0.4e
30.3 0.6
30.0 0.8 29.6 0.9
29.1 0.8
25.6 0.4f
23.2 0.5f
26.8 0.4d 29.5 0.5 29.4 0.7 28.4 0.8
26.1 0.4f
25.7 0.4f
26.2 0.5e
29.6 0.4 29.3 0.6 27.8 0.7
30.0 0.5
29.7 0.6
29.0 0.6 27.5 0.7d 24.1 0.7f
29.0 0.7 27.3 0.6
25.6 0.6f
a
The results are presented as mean SEM and analyzed by one-way ANOVA
followed by the Tukey’s honest significance test.
b
NC = normal chow diet.
HFD = high fat diet.
p <0.05 versus the high fat diet group.
p <0.01 versus the high fat diet group.
c
d
e
sisting of LRH, LXR
a, LXRb, PPARa, PPARc, PPARd, PXR, RORa, and
f
p <0.001 versus the high fat diet group.
RXR . Also, quinoline 1c was highly permeable in the Madin–Darby
a
canine kidney cell (MDCK) absorption assay47 with an apparent per-
meability factor PAPP = 183 nm/s. In contrast to GSK8062 1b
(sol. = 1 ng/mL), quinoline 1c was 60 times more soluble in fasted
state-simulated intestinal fluid (FaS-SIF) at pH 6.5 (sol. = 60 ng/
mL). The enhanced solubility of quinoline 1c allowed a simple pH
buffered water formulation to be utilized for safety assessment dos-
ing. Furthermore, a suitable crystalline potassium salt of quinoline
1c was discovered. Also, with the lower clearance of quinoline 1c
in rats, resulting in better oral exposure, the rat could be employed
this ring flip occurred to allow a similar through water interaction
with the receptor. Alternatively, or in addition, the ring might have
flipped to keep the isoquinoline nitrogen away from a hydrophobic
area formed by the side chains of 270Thr, 273Ile, 335Ile, 343Gly, and
348Leu. In this structure, small rotamer changes in the alkyl chains
of 335Ile and 450Met in the ligand binding domain of FXR are evident,
although little difference is observed between the quinoline 1c and
isoquinoline 1f conformations in those regions of the molecules.
Table 4
Diet-induced obese mouse study—serum chemistry (week 4)a
Diet
NCb
HFDc
HFDc
HFDc
HFDc
Treatment
Vehicle
Vehicle
10 mg/kg
30 mg/kg
100 mg/kg
Glucose (mg/dL)
Insulin (ng/mL)
TCg (mg/dL)
296.9 16.8
0.49 0.04
75.2 3.3f
50.0 1.3
0.42 0.02
9.0 0.7
316.5 13.2
0.80 0.12
140.0 2.8
54.4 1.5
0.46 0.02
11.0 1.3
1.7 0.1
313.0 12.2
0.63 0.07
108.5 1.9f
55.8 2.5
0.53 0.02
10.8 0.8
1.7 0.1
282.2 10.4
0.58 0.07
99.6 3.3f
51.1 2.0
0.52 0.02
10.0 1.1
1.8 0.1
241.4 16.8e
0.34 0.06d
87.9 2.7f
47.4 4.0
0.65 0.06f
9.7 0.7
TGh (mg/dL)
NEFAi (mEq/dL)
Glycerol (mg/dL)
b-HBAj (mg/dL)
1.8 0.1
3.3 1.0d
a
The results are presented as mean SEM and analyzed by one-way ANOVA followed by the Tukey’s honest significance test. Serum samples were taken at the end of the
study without fasting.
b
NC = normal chow diet.
HFD = high fat diet.
p <0.05 versus the high fat diet group.
p <0.01 versus the high fat diet group.
p <0.001 versus the high fat diet group.
TC = total cholesterol.
TG = triglycerides.
NEFA = non-esterified fatty acids.
c
d
e
f
g
h
i
j
b-HBA = b-hydroxyybutyrate.

J. Y. Bass et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1206–1213
1213
13. Zhang, Y.; Lee, F. Y.; Barrera, G.; Lee, H.; Vales, C.; Gonzalez, F. J.; Willson, T. M.;
Edwards, P. A. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 1006.
14. Inagaki, T.; Choi, M.; Moschetta, A.; Peng, L.; Cummins, C. L.; McDonald, J. G.;
Luo, G.; Jones, S. A.; Goodwin, B.; Richardson, J. A.; Gerard, R. D.; Repa, J. J.;
Mangelsdorf, D. J.; Kliewer, S. A. Cell Metab. 2005, 2, 217.
15. Inagaki, T.; Moschetta, A.; Lee, Y.-K.; Peng, L.; Zhao, G.; Downes, M.; Yu, R. T.;
Shelton, J. M.; Richardson, J. A.; Repa, J. J.; Mangelsdorf, D. J.; Kliewer, S. A. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 3920.
16. Jung, D.; Inagaki, T.; Gerard, R. D.; Dawson, P. A.; Kliewer, S. A.; Mangelsdorf, D.
J.; Moschetta, A. J. Lipid Res. 2007, 48, 2693.
17. Sonoda, J.; Rosenfeld, J. M.; Xu, L.; Evans, R. M.; Xie, W. Curr. Drug Metab. 2003,
4, 59.
18. Huang, W.; Ma, K.; Zhang, J.; Qatanani, M.; Cuvillier, J.; Liu, J.; Dong, B.; Huang, X.;
Moore, D. D. Science 2006, 312, 233.
19. Fiorucci, S.; Clerici, C.; Antonelli, E.; Orlandi, S.; Goodwin, B.; Sadeghpour, B. M.;
Sabatino, G.; Russo, G.; Castellani, D.; Willson, T. M.; Pruzanski, M.; Pellicciari, R.;
Morelli, A. J. Pharmacol. Exp. Ther. 2005, 313, 604.
20. Liu, Y.; Binz, J.; Numerick, M. J.; Dennis, S.; Luo, G.; Desai, B.; MacKenzie, K. I.;
Mansfield, T. A.; Kliewer, S. A.; Goodwin, B.; Jones, S. A. J. Clin. Invest. 2003, 112,
1678.
as the rodent toxicology species, during development of quinoline
1c.
Since quinoline 1c exhibited improved pharmacokinetics rela-
tive to GSK8062 1b, it was evaluated in a diet-induced obese
(DIO) mouse model of diabetes. C57BL/6 J mice were fed with a
high-fat diet for four weeks to establish obesity and insulin resis-
tance, then randomized based on fasting glucose and body weight
and assigned to different treatment groups. The mice were dosed
by oral gavage for four weeks with the potassium salt of quinoline
1c (mouse TT EC50 = 120 nM, %Max = 99) or vehicle, and subse-
quently sacrificed for analysis. The high fat diet increased body
weight gain relative to a normal chow diet. As shown in Table 3,
quinoline 1c caused a dose-dependent decrease in body weight
at 10, 30, and 100 mg/kg, mainly caused by a decrease in body
fat mass (data not shown). Furthermore, the FXR agonist 1c signif-
icantly and dose-dependently decreased serum glucose levels, rel-
ative to vehicle, with the maximum effect at 100 mg/kg (Table 4).
This was accompanied by a decrease in serum insulin, which sug-
gests a potential improvement in whole body insulin sensitivity. In
addition, quinoline 1c decreased triglycerides, total cholesterol,
and glycerol, relative to the high fat diet controls.
In summary, a series of aza-naphthalene analogs of GSK8062 1b
were synthesized as potential modulators of FXR. The quinoline 1c
was an equipotent full FXR agonist to GSK8062 1b with good selec-
tivity versus related nuclear receptors. In addition, it was more
water soluble and exhibited better rat pharmacokinetics than
GSK8062 1b, possibly providing development advantages over
GSK8062 1b. Furthermore, quinoline 1c dose-dependently de-
creased serum glucose and body weight gain in the DIO mouse
model. Thus, quinoline (GSK2324) 1c may prove useful in further
defining the physiological roles of FXR, as well as aiding the design
of other FXR modulators.
21. Stedman, C.; Liddle, C.; Coulter, S.; Sonoda, J.; Alvarez, J. G.; Evans, R. M.;
Downes, M. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 11323.
22. Fiorucci, S.; Antonelli, E.; Rizzo, G.; Renga, B.; Mencarelli, A.; Riccardi, L.;
Orlandi, S.; Pellicciari, R.; Morelli, A. Gastroenterology 2004, 127, 1497.
23. Fiorucci, S.; Rizzo, G.; Antonelli, E.; Renga, B.; Mencarelli, A.; Riccardi, L.;
Morelli, A.; Pruzanski, M.; Pellicciari, R. J. Pharmacol. Exp. Ther. 2005, 315, 58.
24. Fiorucci, S.; Rizzo, G.; Antonelli, E.; Renga, B.; Mencarelli, A.; Riccardi, L.;
Orlandi, S.; Pruzanski, M.; Morelli, A.; Pellicciari, R. J. Pharmacol. Exp. Ther. 2005,
314, 584.
25. Zhang, S.; Wang, J.; Liu, Q.; Harnish, D. C. J. Hepatol. 2009, 51, 380.
26. Kim, I.; Morimura, K.; Shah, Y.; Yang, Q.; Ward, J. M.; Gonzalez, F. J.
Carcinogenesis 2007, 28, 940.
27. Yang, F.; Huang, X.; Yi, T.; Yen, Y.; Moore, D. D.; Huang, W. Cancer Res. 2007, 67,
863.
28. Cariou, B. Diabetes Metab. 2008, 34, 685.
29. Evans, M. J.; Mahaney, P. E.; Borges-Marcucci, L.; Lai, K.; Wang, S.; Krueger, J. A.;
Gardell, S. J.; Huard, C.; Martinez, R.; Vlasuk, G. P.; Harnish, D. C. Am. J. Physiol.
2009, 296, G543.
30. Hanniman, E. A.; Lambert, G.; McCarthy, T. C.; Sinal, C. J. J. Lipid Res. 2005, 46,
2595.
31. Hartman, H. B.; Gardell, S. J.; Petucci, C. J.; Wang, S.; Krueger, J. A.; Evans, M. J. J.
Lipid Res. 2009, 50, 1090.
32. Zhang, Y.; Wang, X.; Vales, C.; Lee, F. Y.; Lee, H.; Lusis, A. J.; Edwards, P. A.
Arterioscler., Thromb., Vasc. Biol. 2006, 26, 2316.
Acknowledgement
33. Moschetta, A.; Bookout, A. L.; Mangelsdorf, D. J. Nat. Med. 2004, 10, 1352.
34. Hansen, M. K.; Connolly, T. M. Curr. Opin. Invest. Drugs 2008, 9, 247.
35. Cariou, B.; Staels, B. Trends Pharmacol. Sci. 2007, 28, 236.
36. Abel, U.; Schlueter, T.; Schulz, A.; Hambruch, E.; Steeneck, C.; Hornberger, M.;
Hoffmann, T.; Perovic-Ottstadt, S.; Kinzel, O.; Burnet, M.; Deuschle, U.;
Kremoser, C. Bioorg. Med. Chem. Lett. 2010, 20, 4911.
The authors would like to thank Judi A. McNulty for assistance
with the DIO mouse study.
37. Akwabi-Ameyaw, A.; Bass, J. Y.; Caldwell, R. D.; Caravella, J. A.; Chen, L.; Creech, K.
L.; Deaton, D. N.; Jones, S. A.; Kaldor, I.; Liu, Y.; Madauss, K. P.; Marr, H. B.;
McFadyen, R. B.; Miller, A. B.; Navas, F., III; Parks, D. J.; Spearing, P. K.; Todd, D.;
Williams, S. P.; Wisely, G. B. Bioorg. Med. Chem. Lett. 2008, 18, 4339.
38. Akwabi-Ameyaw, A.; Bass, J. Y.; Caldwell, R. D.; Caravella, J. A.; Chen, L.; Creech, K.
L.; Deaton, D. N.;Madauss, K. P.; Marr, H.B.;McFadyen,R. B.;Miller, A. B.;Navas,F.;
Parks, D. J.; Spearing, P. K.; Todd, D.; Williams, S. P.; Wisely, G. B. Bioorg. Med. Chem.
Lett. 2009, 19, 4733.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.12.089.
References and notes
39. Bass, J. Y.; Caldwell, R. D.; Caravella, J. A.; Chen, L.; Creech, K. L.; Deaton, D. N.;
Madauss, K. P.; Marr, H. B.; McFadyen, R. B.; Miller, A. B.; Parks, D. J.; Todd, D.;
Williams, S. P.; Wisely, G. B. Bioorg. Med. Chem. Lett. 2009, 19, 2969.
40. Feng, S.; Yang, M.; Zhang, Z.; Wang, Z.; Hong, D.; Richter, H.; Benson, G. M.;
Bleicher, K.; Grether, U.; Martin, R. E.; Plancher, J.-M.; Kuhn, B.; Rudolph, M. G.;
Chen, L. Bioorg. Med. Chem. Lett. 2009, 19, 2595.
41. Pellicciari, R.; Fiorucci, S.; Camaioni, E.; Clerici, C.; Costantino, G.; Maloney, P.
R.; Morelli, A.; Parks, D. J.; Willson, T. M. J. Med. Chem. 2002, 45, 3569.
42. Downes, M.; Verdecia, M. A.; Roecker, A. J.; Hughes, R.; Hogenesch, J. B.;
Kast-Woelbern, H. R.; Bowman, M. E.; Ferrer, J.-L.; Anisfeld, A. M.; Edwards, P.
A.; Rosenfeld, J. M.; Alvarez, J. G. A.; Noel, J. P.; Nicolaou, K. C.; Evans, R. M. Mol.
Cell 2003, 11, 1079.
1. Hylemon, P. B.; Zhou, H.; Pandak, W. M.; Ren, S.; Gil, G.; Dent, P. J. Lipid Res.
2009, 50, 1509.
2. Lefebvre, P.; Cariou, B.; Lien, F.; Kuipers, F.; Staels, B. Physiol. Rev. 2009, 89, 147.
3. Gupta, S.; Stravitz, R. T.; Dent, P.; Hylemon, P. B. J. Biol. Chem. 2001, 276, 15816.
4. Kawamata, Y.; Fujii, R.; Hosoya, M.; Harada, M.; Yoshida, H.; Miwa, M.;
Fukusumi, S.; Habata, Y.; Itoh, T.; Shintani, Y.; Hinuma, S.; Fujisawa, Y.; Fujino,
M. J. Biol. Chem. 2003, 278, 9435.
5. Eloranta, J. J.; Kullak-Ublick, G. A. Physiology 2008, 23, 286.
6. Forman, B. M.; Goode, E.; Chen, J.; Oro, A. E.; Bradley, D. J.; Perlmann, T.;
Noonan, D. J.; Burka, L. T.; McMorris, T.; Lamph, W. W.; Evans, R. M.;
Weinberger, C. Cell 1995, 81, 687.
43. Flatt, B.; Martin, R.; Wang, T.-L.; Mahaney, P.; Murphy, B.; Gu, X.-H.; Foster, P.;
Li, J.; Pircher, P.; Petrowski, M.; Schulman, I.; Westin, S.; Wrobel, J.; Yan, G.;
Bischoff, E.; Daige, C.; Mohan, R. J. Med. Chem. 2009, 52, 904.
44. Bass III, J. Y.; Deaton, D. N.; Caravella, J.; McFadyen, R. B.; Navas, F.; Spearing, P.
K. PCT Int. Appl. WO 096921, 2008; Chem. Abstr. 2008, 148, 472019.
45. Chan, G.; Johns, A.; Jurewicz, A.; Porter, R. A.; Widdowson, K. PCT Int. Appl. WO
09024, 1999; Chan, G.; Johns, A.; Jurewicz, A.; Porter, R. A.; Widdowson, K.
Chem. Abstr. 1999, 130, 196581.
46. Buchstaller, H.-P.; Burgdorf, L.; Finsinger, D.; Amendt, C.; Grell, M.; Sirrenberg,
C.; Zenke, F. PCT Int. Appl. WO 082858, 2005; Chem. Abstr. 2005, 143, 286297.
47. Irvine, J. D.; Takahashi, L.; Lockhart, K.; Cheong, J.; Tolan, J. W.; Selick, H. E.;
Grove, J. R. J. Pharm. Sci. 1999, 88, 28.
7. Higashiyama, H.; Kinoshita, M.; Asano, S. Acta Histochem. 2008, 110, 86.
8. Lambert, G.; Amar, M. J. A.; Guo, G.; Brewer, H. B., Jr.; Gonzalez, F. J.; Sinal, C. J. J.
Biol. Chem. 2003, 278, 2563.
9. 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.
10. Watanabe, M.; Houten, S. M.; Wang, L.; Moschetta, A.; Mangelsdorf, D. J.;
Heyman, R. A.; Moore, D. D.; Auwerx, J. J. Clin. Invest. 2004, 113, 1408.
11. Cariou, B.; van Harmelen, K.; Duran-Sandoval, D.; van Dijk, T.; Grefhorst, A.;
Bouchaert, E.; Fruchart, J.-C.; Gonzalez, F. J.; Kuipers, F.; Staels, B. FEBS Lett.
2005, 579, 4076.
12. Ma, K.; Saha, P. K.; Chan, L.; Moore, D. D. J. Clin. Invest. 2006, 116, 1102.