Identification of phenylsulfone-substituted quinoxaline (WYE-672) as a tissue selective liver X-receptor (LXR) agonist

Article abstract of DOI:10.1021/jm100034x

A series of phenyl sulfone substituted quinoxaline were prepared and the lead compound 13 (WYE-672) was shown to be a tissue selective LXR Agonist. Compound 13 demonstrated partial agonism for LXRβ in kidney HEK-293 cells but did not activate Gal4 LXRβ fusion proteins in huh-7 liver cells. Although 13 showed potent binding affinity to LXRβ (IC₅₀ = 53 nM), it had little binding affinity for LXRα (IC₅₀ > 1.0 μM) and did not recruit any coactivator/corepressor peptides in the LXRα multiplex assay. However, compound 13 showed good agonism in THP-1 cells with respect to increasing ABCA1 gene expression and good potency on cholesterol efflux in THP-1 foam cells. In an eight-week lesion study in LDLR -/- mice, compound 13 showed reduction of aortic arch lesion progression and no plasma or hepatic triglyceride increase. These results suggest quinoxaline 13 may have an improved biological profile for potential use as a therapeutic agent.

Full text of DOI:10.1021/jm100034x

3296 J. Med. Chem. 2010, 53, 3296–3304
DOI: 10.1021/jm100034x
Identification of Phenylsulfone-Substituted Quinoxaline (WYE-672) as a Tissue Selective
Liver X-receptor (LXR) Agonist
Baihua Hu,*,† Rayomand J. Unwalla, Igor Goljer,† James W. Jetter,† Elaine M. Quinet,‡ Thomas J. Berrodin,‡
Michael D. Basso,‡ Irene B. Feingold,§ Annika Goos Nilsson, Anna Wilhelmsson, Mark J. Evans,‡ and Jay E. Wrobel†
†Chemical Sciences, Wyeth Pharmaceuticals, 500 Arcola Road, Collegeville, Pennsylvania 19426, ‡Cardiovascular and Metabolic Diseases,
Wyeth Pharmaceuticals, Collegeville, Pennsylvania 19426, §Biotransformation and Disposition, Wyeth Pharmaceuticals, Collegeville,
Pennsylvania 19426, and Karo Bio AB, Huddinge, Sweden
Received January 11, 2010
A series of phenyl sulfone substituted quinoxaline were prepared and the lead compound 13 (WYE-672) was
shown to be a tissue selective LXR Agonist. Compound 13 demonstrated partial agonism for LXRβin kidney
HEK-293 cells but did not activate Gal4 LXRβfusion proteins in huh-7 liver cells. Although 13 showed potent
binding affinity to LXRβ(IC50 = 53 nM), it had little binding affinity for LXRR (IC50 > 1.0μM) and did not
recruit any coactivator/corepressor peptides in the LXRR multiplex assay. However, compound 13 showed
good agonism in THP-1 cells with respect to increasing ABCA1 gene expression and good potency on
cholesterol efflux in THP-1 foam cells. In an eight-week lesion study in LDLR -/- mice, compound 13
showed reduction of aortic arch lesion progression and no plasma or hepatic triglyceride increase. These results
suggest quinoxaline 13 may have an improved biological profile for potential use as a therapeutic agent.
Introduction
LXR agonists also induce hypertriglyceridemia in the liver,
mainly through the upregulation of genes in fatty acid bio-
synthesis, including sterol regulatory binding element protein 1c
(SREBP-1c) and fatty acid synthase (FAS). Several strategies2,3
have been proposed for improving the therapeutic index of
LXR agonists including partial agonists, LXRβ subtype selec-
tive agonists, and gene or tissue specific agonists. The first
hypothesis is based on the premise that the partial agonists bind
to and activate the LXR receptors but elicit a smaller triglycer-
ide (TG) response than a full LXR agonist. WAY-252623 (3),
the first LXR agonist clinical candidate, possessed potent LXR
binding affinity but was a weak partial agonist in Gal4 trans-
activation assays, particularly on LXRR relative to pan agonist
1a.10 Compound 3demonstrated efficacy for reduction of aortic
arch lesion progression in a LDLR -/- atherosclerosis mice
model. In addition, this compound upregulated LXR target
gene expression (ABCA1 and ABCG1) in the duodenum of
hamsters10 and in whole blood of cynomolgus monkeys and
humans.11 Compound 3 did not have significant effect on
plasma or hepatic lipids in LDLR -/- mice11b or hamsters.10
In cynomolgus monkeys, the compound had no effect on
plasma lipids but did increase hepatic lipids dramatically at
the highest dose tested.11b Although this hepatic TG increase
may not be due directly to LXR induced lipogenic gene regu-
lation,11b we felt it was necessary to further improve our
compound profile with respect to these lipid effects and there-
fore strove to achieve greater selectivity over LXRR. Efforts to
minimize side effects through selective targeting of the LXR
β-isoform are based on the premise that LXRR is the predomi-
nant subtype expressed in the liver and that activation of LXRR
may be responsible for the observed increase in hepatic lipogen-
esis. Recent studies provide confirmation that selective LXRβ
activation reversed atherosclerosis and cellular cholesterol over-
load in mice lacking LXRR and ApoE.12 Unfortunately, there
Liver X receptors, LXRRa (expressed in adipose, intestine,
liver, kidney, and macrophages) and LXRβ (expressed ubi-
quitously), are members of the nuclear hormone receptor super
family and are involved in the regulation of cholesterol and lipid
metabolism.1-4 They are ligand-activated transcription factors
and bind to DNA as obligate heterodimers with retinoid X
receptors (RXR). Activation of LXRs (1) promotes transacti-
vation of ATP-binding cassettes (ABCA1, ABCAG1) and
apoE in macrophages, (2) prevents cholesterol accumulation
as intracellular lipid droplets, (3) promotes cholesterol efflux
from macrophages-derived foam cells of established lesions, (4)
enhances reverse cholesterol transport (RCT) process, (5)
reduces inflammation in lesion sites, and (6) reduces cholesterol
absorption in the gut. For instance, treatment with known LXR
modulators such as 1a (TO901317),5 1b (GW3965),6,7 and 2
(WAY-254011)8,9 (Figure 1) retard the advancement of athero-
sclerotic processes in mouse models of atherosclerosis. Thus, an
LXR agonist may offer potential benefits on lipid metabolism,
glucose metabolism, and vascular inflammation, thereby redu-
cing the risk of cardiovascular disease. However, these synthetic
*To whom correspondence should be addressed. Phone: 609-619-
3517. E-mail: bhu519@aol.com.
a Abbreviations: LXR, liver X receptor; ABCA1/G1/G5/G8, adeno-
sine triphosphate binding cassette, A1/G1/G5/G8; HepG2, human
hepatocellular liver carcinoma cell line; THP-1, Human acute monocytic
leukemia cell line; LDLR-/-, low density lipoprotein receptor knock-
out; TG, triglycerides; RXR, retinoic acid X receptor; RCT, reverse
cholesterol transport; SREBP-1c, sterol regulatory binding element 1c;
FAS, fatty acid synthase; LBD, ligand binding domain; PPARs (R, γ, or
δ), Peroxisome proliferator-activated receptor R, γ, or δ; FXR, farne-
soid X receptor; PXR, pregnane X receptor; TR, thyroid receptor;
NADPH, the reduced form of nicotinamide adenine dinucleotide phos-
phate; GSH, glutathione; UDPGA, uridine 50-diphospho-glucuronic
acid; HEK 293, human embryonic kidney 293 cells.
r
pubs.acs.org/jmc
Published on Web 03/30/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3297
Figure 1. Known LXR agonists.
Scheme 1a
a Reaction conditions: (a) EtOH or MeOH, RT; (b) Pd(OAc)2, 2-dicyclohexylphosphino-20,60-dimethoxybiphenyl, n-BuOH; (c) HBr or BBr3; (d)
N-phenylbis(trifluoromethanesulfonamide), KOtBu, THF, 0ꢀC to RT; (e) ArB(OH)2, Pd(PPh3)4, K3PO4, dioxane, reflux.
are only minor structural differences in the ligand binding
domains (LBD) of LXRR and LXRβ that can be exploited to
obtain LXRβ-selective ligands. The two LXR isoforms R/β
share a high sequence identity (78%) and residue differences
are located far away from the ligand binding pocket.13 This
high similarity in the binding pocket of the two LXR isoforms
constitutes a serious obstacle to the development of highly β-
selective ligands. Nevertheless, a modest level of LXRβ
selectivity in a small molecule has been achieved. N-Acylthia-
diazoline 4 with preferential affinity for LXRβ in scintillation
proximity assays (SPA) has been reported.14 However, 4 was
found to be highly protein-bound and no in vivo efficacy
(ABCA1 or SREBP1C induction) was observed in mice. N,N-
Dimethyl-3β-hydroxycholenamide (5), a synthetic oxysterol,
is a gene-selective LXR modulator that mediates potent
transcriptional activation of ABCA1 gene expression while
exhibiting minimal effects on SREBP-1c both in vitro and in
vivo in mice (dosed by IP injection only).15 Compound 5
stimulated cholesterol transport through the upregulation of
LXR target genes in liver, small intestine, and peritoneal
macrophages. However, 5 exhibited only limited activity for
increasing hepatic SREBP-1c mRNA without altering the
plasma TG. Unfortunately, 5 had poor pharmacokinetic
properties, which may limit 5 for potential use as a therapeutic
agent. Inthisreport, wedescribethe identification andSAR of
a novel series of phenyl sulfone substituted quinoxalines
leading up to the discovery of 13 (WYE-672) as an orally
active and tissue selective LXR agonist which reduced athero-
sclerotic lesion size in LDLR -/- mice without increasing
plasma or hepatic TG levels.
Chemistry
It has been demonstrated by Wyeth that heterocycle scaf-
folds such as quinoline,8 indazole,10 cinnoline,16 benzimida-
zole,17 and quinazoline18 can be utilized to afford potent LXR
agonists. In an attempt to enlarge the chemical diversity of
LXR ligands, we prepared a series of quinoxalines derivatives
as shown in Scheme 1. Condensation of 1,2-diaminobenzenes
6 with 1,2-dioxoalkane 7 gave a mixture of position isomers
which were separated into 8 and 9. The regiochemistry of 8
and 9 was assigned by 1H-15N-gHMBC NMR experiments.
Reaction of 8 (X = Cl) with phenylboronic acids under
Suzuki conditions provided 12. Alternatively, deprotection
of the methoxy group of 8 (X=OMe) by treatment with HBr
or BBr3 led to phenol 10. Phenol 10 was converted to triflate
11 by using N-phenylbis(trifluoromethanesulfonamide) in the
presence of triethylamine. The resulting triflate 11 was
coupled to aryl boronic acids to give the biaryl derivatives 12.

3298 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Table 1. LXR Binding Affinities of Quinoxalinesa
Hu et al.
compd
R1
R2
position
R3
LXRβ LXRR
1a
13
14
15
16
17
18
19
20
21
22
23
24
25
26
0.009
0.053
>1
0.013
>1
>1
8-CF3
8-CF3
8-CF3
8-CF3
8-CF3
8-CF3
8-CF3
8-CF3
8-CF3
8-CF3
8-Cl
Me
Me
Me
Me
Me
Me
Me
Me
Et
p, m0
p, p0
SO2Me
SO2Me
SO2Me
SO2Me
SO2Me
SO2Et
p, o0
>1
>1
>1
>1
m, m0
m, p0
p, m0
p, m0
p, m0
p, m0
p, m0
p, m0
p, m0
p, m0
>1
>1
>1
0.132
0.081
>1
Figure 2. Docked structure of compound 13 (magenta) overlaid
with the hLXRβ/2 (cyan) X-ray complex structure Only important
residues and helices are shown from the binding site.
SO2NHMe
CONHMe
SO2Me
SO2Me
SO2Me
SO2Me
SO2Me
>1
>1
0.018
>1
0.165
>1
H
residue, whileatthe distalendthe sulfoneoxygenmadecritical
interaction withthe backboneNHgroupof Leu330. There are
interesting differences between the two structures with regard
to the position of the core quinoline/quinoxaline rings and in
the ability of these ligands to orient the critical acid/methyl
sulfone groups toward the solvent exposed pocket. The N-1
atom of 13 is shifted further away from the His435 residue as
the ligand tries to orient the 2-methyl group in the hydro-
phobic pocket occupied by the N-3 benzyl group of 2 and the
C3 phenyl group of 13 has a different projection than that
observed for the C4 phenyl group in the X-ray bound
orientation of compound 2. In spite of the differences in the
trajectory of these phenyl groups, the terminal acid group of 2
and the methyl sulfone group of 13 are in similar position to
make the critical interaction with the Leu330 NH backbone.
This is a direct result of the phenyl group being able to twist
out of the plane and thereby allowing the longer but flexible
benzyloxy linker to orient the phenyl acetic acid group of 2
and place it in a similar position as the phenyl methyl sulfone.
The small, i.e., 3-fold increase in LXRβ binding potency for
compound 21 when compared to 13 is a result of the 2-ethyl
group being oriented toward a hydrophobic pocket sur-
rounded by the Phe340 and Phe349 residues and in a region
where the larger 3-benzyl group of compound 2 was observed
in the X-ray structure. Although the immediate residues
surrounding this pocket are not different between LXRR
and LXRβ, the small drop in LXRβ selectivity observed for
compound 21, i.e., ∼9-fold may due to the result of residue
differences within the second shell of this region.21
The potent LXRβ binders from Table 1 (Compound 13, 19,
21, and 23) were further profiled in the Gal4 transactivation
assays in huh-7 liver cells and cellular assays (ABCA1 expres-
sion in THP-1 cells, lipid accumulation in HepG2 liver cells,
Table 2).8 Literature compound 1a is a potent full agonist for
LXRR and LXRβ in the transactivation assays and showed
little differentiation between ABCA1 gene activation versus
cellular TG synthesis. In contrast, 13 showed a very favorable
profile as an LXR modulator: it increased ABCA1 gene
expression in THP-1 cells with a nearly full efficacy (79%)
and it had very low efficacy (7%) for lipid accumulation.
Compound 13 was tested in the Gal4 transactivation assays
and was inactive (efficacy 1%) against LXRR. Surprisingly, the
compound also showed basically no transactivation activity
Me
Me
0.034
0.299
0.395
0.061
0.734
>1
>1
8-CN
8-OMe Me
1.77
a IC50 in μM. The highest concentration tested was 10 μM. Results are
given as the mean of at least two independent experiments. The standard
deviations for these assays were typically (30% of the mean or less.
Results and Discussion
The LXR binding affinities of newly synthesized compounds
were evaluated (Table 1) using recombinant human LXRR or
LXRβ ligand binding domains (LBDs) with [3H]1a as a tracer.
Our first attempt with the quinoxaline scaffold (compound 13)
showed good LXRβ binding affinity (IC50 value of 53 nM) but
poor potency (IC50 value of >1 μM) for LXRR. SAR studies on
this scaffold revealed that the meta orientation (position m0) for
the methyl sulfone group (13) is preferred. The corresponding
para (14, position p0) or ortho (15, position o0) analogues were
considerably less potent. The projection of the phenyl 1 group is
also critical for the activity as is evident by 16 and 17, the meta-
substituted (position m) isomers, which had LXRβbinding IC50
values of greater than 1 μM. Compared to 13, the ethyl sulfone
18 (LXRβ IC50=132 nM) and methyl sulfonamide 19 (LXRβ
IC50=81 nM) were slightly weaker binders. Amides were pre-
viously demonstrated by Wyeth as good replacements for the
methyl sulfone group.19 However, a few amides were prepared
in the series and none of them showed appreciable LXR binding
affinity, as exemplified by 20. The group on quinoxaline C-2
also had big impact on the LXR binding affinity: the 2-ethyl
analogue 21 is a potent LXRβ binder (IC50=18 nM), while the
2-H analogue 22 had poor binding affinity (LXRβ binding
IC50 >1μM). Several small replacements9 (compounds 23-25)
for the C-8 trifluoromethyl substituent were evaluated, and the
chlorine once again proved to be a good replacement for the C-8
trifluoromethyl group because the 8-chloro quinoxaline 23 was
a slightly more potent (IC50=34 nM) LXRβ binder than 13.
Tounderstand the observed SAR for the quinoxaline series,
we docked20 13 into a previously solved in-house X-ray
structure of hLXRβ complexed with compound 2. Figure 2
shows the docked structure of 13 overlaid with the X-ray
bound pose of compound 2. Ligand recognition is achieved by
8-CF3 group making hydrogen bond interaction with His435

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3299
Table 2. Cellular Activity of Selected Compoundsa
and release of corepressors, the character of which may not be
expected simply by the in vitro binding to the LBD. To further
confirm the selectivity of 13 against the LXRR isoform, a
ligand induced coactivator recruitment assay11b for probing
conformational subtleties was conducted (Figure 3). In the
multiplex recruitment assay, compound 13 showed very little
if any interaction with the 43cofactor peptides (Figure 3A) for
the LXRR subtype, while 26 and 3 had a comparable recruit-
ment profile to that of 1a, albeit reduced in magnitude.
However, 13’s LXRβ peptide recruitment profile with the
same set of coactivator/repressor peptides was basically the
same as 1a, 3, and 26 (Figure 3B). These results may indicate a
unique conformation associated with 13 which are consistent
with the distinct pharmacology of 13 versus 26, 3, and 1a.
Compound 13 when tested for its ability to stimulate [H3]
cholesterol efflux in THP-1 foam cells induced cholesterol
efflux in a concentration-dependent manner, consistent with
its ability to induce ABCA1 gene expression. Notably, in this
assay, compound 13 had good potency (EC50 value of 72 nM
for 13 versus 4 nM for 1a) and efficacy (60% relative to 1a).
Compound 13 was screened against a panel of 23 nuclear
receptors, and it did not display agonist or antagonist activity
against closedrelatedreceptors such asPPARs(R, γ, δ), FXR,
PXR, RXR, GR, MR, AR, PR, ER, and TR (data not
shown).
compd Gal4 LXRβb Gal4 LXRRb
ABCA1c
lipid accumd
1a
13
19
21
23
26
0.17 (100%)
>1 (3.1%)
(2%)
0.14 (100%) 0.044 (100%) 0.14 (100%)
>1 (1%)
(1%)
1.09 (79%)
1.47 (55%)
0.90 (65%)
0.29 (50%)
1.89 (87%)
(7%)
(6%)
2.02 (70%)
0.86 (5%)
1.70 (84%)
3.59 (13)
(1%)
5.10 (36%)
0.35 (32%)
0.38 (18%)
3.18 (74%)
a EC50 in μM (%eff). The highest concentration tested was 10 μM.
Results are given as the mean of at least two independent experiments.
The standard deviations for these assays were typically (50% of the
mean or less. b The % of efficacy is relative to 1a in huh7-cell. c The % of
efficacy is relative to 1a in differentiated THP-1 cells. d The % of efficacy
is relative to 1a in HepG2 cells.
Table 3. Functional Activity in huh-7 and HEK-293 Cellsa
huh-7
HEK-293
compd
LXRβ
LXRR
LXRβ
LXRR
1a
13
26
0.17 (100%) 0.14 (100%) 0.019 (100%) 0.053 (100%)
0.58 (45%)
0.46 (48%)
>1 (3.1%)
1.70 (84%)
>1 (1%)
5.10 (36%)
1.7 (7%)
10.9 (80%)
a EC50 in μM (%eff). The highest concentration tested was 10 μM.
Results are given as the mean of at least two independent experiments.
The standard deviations for these assays were typically (50% of the
mean or less. The % of efficacy is relative to 1a.
(efficacy 3.1%) against LXRβ (vide infra). As expected, more
potent LXRβ binders 21 and 23 showed a more potent cellular
activity (Gal4, ABCA1, and lipid accumulation) over 13 while
the weaker binder (19) showed weaker cellular activity.
After being administered at a 10 mg/kg PO dose in 2%
Tween 80/0.5% methylcellulose to fasted male C57 mice,
compound 13 showed reasonable oral PK parameters: it
had a long half-life (t1/2 >20 h) and the Cmax, AUC0-¥, and
We next focused on a preliminary investigation of a possible
explanation why 13 had good activity in the THP-1 cells but
poor activity in the huh-7 liver cells. Thus we compared 2621 to
13 because it had a similar binding profile (LXRβIC50 =61nM,
LXRR IC50 = 1.77 μM) and showed similar activity in the
THP-1 cells (ABCA1 EC50 1.89 μM, 87% efficacy). However,
26 was quite efficacious in the huh-7 transactivation assay (Gal4
LXRβ EC50 1.7 μM, 84% efficacy). We initially suspected that
the activity discrepancy between 13 and 26 might be due to the
difference between the compound’s stability, solubility, and/or
cell permeability. However, no apparent difference in solubility,
stability (mouse liver microsomes), and cell uptake in the huh-7
cells was observed22 for the two compounds. When compound
13 was incubated with mouse and human liver microsomes (2
mg/mL) fortified with cytosol (2 mg/mL), NADPH (regenerat-
ing system), GSH (2 mM), and UDPGA (4 mM) at 37 ꢀC for 60
min and 100% of parent compound remained, which suggested
that the unique profile of 13 was not due to its metabolism. Inte-
restingly, when compound 13 was tested against a 293 kidney-
derived cell line (HEK-293)23 based Gal4β transactivation
assay, it showed reasonable activity (EC50 0.58 μM, 45% effi-
cacy, Table 3). As may be anticipated based on its poor LXRR
binding potency, 13 did not activate the Gal4 LXRR fusion
protein (7% efficacy) in the HEK-293 cells. In contrast, 26 and
1a showed similar transactivation potencies and efficacies bet-
ween the huh-7 liver cell line and the HEK-293 kidney cell line.
Thus, 13 appears to be a tissue selective (kidney and macro-
phage versus liver cell) LXR ligand that dissociates the desired
ABCA1 gene activation from the unwanted lipid accumulation.
Recent evidence supports the hypothesis that the lipogenic
effects of LXR ligands are primarily LXRR mediated.2,3,12
Compound 13 had little binding affinity against LXRR recep-
tor, which may confer the observed low lipogenic effects asso-
ciated with 13. However, LXR functional activity is the result
of a complex interaction between recruitment of coactivators
Tmax were 229 ng/mL, 2439 h ng/mL, and 4 h, respectively.
3
An accelerated atherosclerotic lesion study was conducted in
high fat/high cholesterol (1.25%)-fed LDLR-/- mice (n =
12 per group) for 8 weeks. Either 0.6 mg/kg/day or 3 mg/kg/
day of 13 admixed in the feed resulted in a significant reduc-
tion in lesion burden by 20% (p < 0.05), and51% (p< 0.001),
respectively, compared to the control group. In the same
experiment, the literature standard 1b (dosed at 10 mg/kg)
also significantly reduced the lesion burden by 41% (n = 12,
p < 0.001). The average plasma concentrations of 13 were
1019 ng/mL and 10782 ng/mL for the 0.6 mg/kg and 3 mg/kg
dose, respectively (at 3 h following 8 week of daily dosing in
feed). The average plasma concentration of 1b was 175 ng/
mL. Neither 13 nor 1b had a significant effect on total plasma
cholesterol levels or on plasma or hepatic triglyceride levels. In
addition, the effects of compound 13 on LXR responsive
genes were monitored in several tissues. In the duodenum, 13
at both doses and 1b significantly induced ABCA1, ABCG5,
and ABCG8 gene expression (Figure 4). Similarly, 13 at both
doses and 1b induced ABCA1 and ABCG1 expression in
mononuclear peritoneal macrophages (Figure A in the Sup-
porting Information section). In contrast, as shown in Fig-
ure 4, in the duodenum 13 exhibited much weaker activity for
SREBP1c gene expression compared to 1b. In liver, neither
compound (13 or 1b) induced hepatic SREBP-1c gene expres-
sion and both showed little regulation of LXR target genes.
(Figure B in the Supporting Information).
Conclusion
In summary, we have described the identification of phenyl
sulfone substituted quinoxaline 13 as a tissue selective LXR
agonist. The tissue selectivity was observed in Gal4 transacti-
vation assays in liver huh-7 cells versus kidney HEK-293 cells.
It showed partial agonism for LXRβ in HEK-293 cells but did
not activate Gal4 LXRR fusion proteins in either HEK-293

3300 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Hu et al.
Figure 3. Effect of 13, 1a, 3, and 26 on recruitment of cofactor peptides on human LXRR and LXRβ LBD proteins. The LXR multiplex assay
was performed with 10 μM concentration. (a) (top) represents LXRR and (b) (bottom) for LXRβ. The mean fluorescence intensity for
treatments was plotted on the Y axis.
cell line or huh-7 cell line. Quinoxaline 13 showed poor bind-
ing affinity to LXRR in the binding assay and did not recruit
coactivator/corepressor peptides in a LXRR multiplex assay.
Compound 13 showed good agonism in THP-1 cells (79%
relative to 1a) with respect to increasing ABCA1 gene expres-
sion and good potency (EC50 = 72 nM) on cholesterol efflux
in THP-1 foam cells. In an eight-week lesion study, compound
13 showed reduction of aortic arch lesion progression and
stimulation of LXR genes in the duodenum without inducing
the plasma or hepatic TG levels. Therefore, 13 may be a
promising agent for the treatment of atheroslerosis.
directly. 1H NMR were recorded on Varian INOVA 400 MHz,
Bruker AVANCE II 400 MHz, and Bruker AVANCE II 300
MHz instrumentsinthe indicatedsolventat20ꢀC.Chemicalshifts
(δ) are expressed in ppm downfield from tetramethylsilane
(TMS). High resolution mass spectra were recorded on an Agilent
6210 TOF instrument. Positive and negative electrospray mass
spectra were recorded on Waters ZQ or ZMD instruments.
The following HPLC methods were used to determine purity of
intermediates and final products. All final products tested in biol-
ogical assays were g95% pure. Method A: column, Xterra MS
C18, 5 μ, 50 mm ꢀ 2.1 mm. Mobile phase: A, 10 mM ammonium
formate in water (pH 3.5); B, 50:50 ACN:MeOH, for 2 min, hold
1.5 min. Sample concentration: 2.000 g/L. Temperature: room tem-
perature. Flow rate: 0.8 mL/min. Detection: 210-370 nm. Method
B: Column Xterra reverse phase C18, 3.5 μ, 150 mm ꢀ 2.1 mm.
Mobile phase: A, 85/15-5/95 (ammonium formate in water, pH
3.5; B:ACN þ MeOH, for 10 min, hold 4 min. Temperature:
40 ꢀC. Flow rate: 0.8 mL/min. Detection: 210-370 nm.
Experimental Section
General Methods. All experiments were conducted in well-
ventilated fume hoods. Anhydrous solvents were purchased from
Aldrich Chemical Co. (Milwaukee, WI) and used directly. Bulk
solvents and chemicals were purchased from EMD and used

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3301
Figure 4. Effects of 13 and 1b on target gene relative expression in duodenum. Male LDLR-/- mice (n = 5 or 6) were fed a Western diet (black
bars) or diet supplemented to deliver 10 mg/kg of 1b (gray bars), 0.6 mg/kg of compound 13 (orange bars), and 3 mg/kg compound 13 (red bars)
for eight weeks. ** p < 0.01; *** p < 0.001.
3-Methyl-2-[30-(methylsulfonyl)biphenyl-4-yl]-5-(trifluoromethyl)-
quinoxaline (13). Step 1: A mixture of 2-nitro-3-(trifluoro-
methyl)aniline (2.6 g, 12.6 mmol), ethanol (20 mL), and 10%
Pd/C (1.0 g) was pressurized with 25 psi H2 for 1.5 h. The cata-
lyst was removed by filtering through a short pad of celite. The
filtrate (3-trifluoro-benzene-1,2-diamine solution) was mixed
with 1-(4-methoxyphenyl)propane-1,2-dione (2.5 g, 14.0 mmol).
The mixture was stirred at room temperature for 1 h, and the
solvent was removed. The residue was purified by flash chro-
matography eluted with EtOAc/hexane to give 2-(4-methoxy-
phenyl)-3-methyl-5-(trifluoromethyl)quinoxaline (8a) as a pale-
yellow solid (1.4 g, 35% for two steps). 1H NMR (DMSO-d6): δ
8.34 (d, J = 7.5 Hz, 1H), 8.21 (d, J = 7.1 Hz, 1H), 7.91 (t, J =
7.7 Hz, 1H), 7.77 (d, J = 8.7 Hz, 2H), 7.12 (d, J = 8.7 Hz, 2H),
3.86 (s, 3H), 2.79 (s, 3H). MS (ES) m/z 319.1.
8.11 (d, J = 7.0 Hz, 1H), 7.80 (d, J = 8.2 Hz, 2H), 7.48 (d, J =
8.2 Hz, 2H), 7.26 (s, 1H), 2.85 (s, 3H). MS (ES) m/z 436.9.
Step 4: A mixture of 11a (1.2 g, 2.75 mmol), 3-methylsulfo-
nylphenyl boronic acid (2.4 g, 12 mmol), K3PO4 (5.0 g, 23.6
mmol), and Pd(PPh3)4 (0.5 g, 0.43 mmol) in 40 mL of dioxane
was heated to 80 ꢀC for 1 h. The reaction mixture was poured
into water and extracted with EtOAc. The organic was concen-
trated and purified by flash chromatography eluted with
1
EtOAc/hexane to give 13 (0.59 g, 48%) as a white solid. H
NMR (DMSO-d6): δ 8.40 (d, J = 8.9 Hz, 1H), 8.29-8.26 (m,
2H), 8.19 (d, J = 7.8 Hz, 1H), 8.00-7.93 (m, 6H), 7.81 (t, J = 7.8
Hz, 1H), 3.34 (s, 3H), 2.83 (s, 3H). MS (ES) m/z 443.0. HRMS:
calcd for C23H17F3N2O2S þ Hþ, 443.10356; found (ESI, [M þ
H]þ), 443.1040.
3-Methyl-2-[40-(methylsulfonyl)biphenyl-4-yl]-5-(trifluoromethyl)-
quinoxaline (14). Prepared according to the procedure of com-
pound 13 (step 4) from 11a and 4-methylsulfonylphenylboronic
acid in 39% yield; pale-yellow solid. 1H NMR (CDCl3): δ 8.31
(d, J = 8.3 Hz, 1H), 8.10-8.06 (m, 3H), 7.76-7.87 (m, 7H), 3.13
(s, 3H), 2.90 (s, 3H). MS (ES) m/z 443.0. HRMS: calcd for
C23H17F3N2O2S þ Hþ, 443.10356; found (ESI, [M þ H]þ),
443.1039.
The regio-isomer 3-(4-methoxyphenyl)-2-methyl-5-(trifluoro-
methyl)quinoxaline (9a) was also isolated in 50% yield (2 g)
from the reaction mixture. 1H NMR (DMSO-d6): δ 8.31 (d, J =
7.9, 1H), 8.19 (d, J = 7.3 Hz, 1H), 7.81 (t, J = 8.0 Hz, 1H), 7.80
(d, J = 6.8 Hz, 2H), 7.14 (d, J = 6.8 Hz, 2H), 3.86 (s, 3H), 2.82
(s, 3H). MS (ES) m/z 319.1.
1
The regio-chemistry for 9a was assigned based on H-15N-
gHMBC NMR experiment, and the details are included in the
Supporting Information.
3-Methyl-2-[20-(methylsulfonyl)biphenyl-4-yl]-5-(trifluoromethyl)-
quinoxaline (15). Prepared according to the procedure of com-
pound 13 (step 4) from 11a and 2-methylsulfonylphenylboronic
acid in 63% yield; white solid. 1H NMR (DMSO-d6): δ 8.41 (d,
J = 7.8 Hz, 1H), 8.26 (d, J = 7.2 Hz, 1H), 8.14 (d, J = 8.0 Hz,
1H), 7.96 (t, J = 8.0 Hz, 1H), 7.89 (d, J = 8.3 Hz, 2H), 7.82 (t,
J = 7.4 Hz, 1H), 7.74 (t, J = 7.4 Hz, 1H), 7.61 (d, J = 8.3 Hz,
2H), 7.51 (d, J = 7.8 Hz, 1H), 3.31 (s, 3H), 2.94 (s, 3H). HRMS:
calcd for C23H17F3N2O2S þ Hþ, 443.10356; found (ESI, [M þ
H]þ), 443.1043. HRMS: calcd for C23H17F3N2O2S þ Naþ,
465.08550; found (ESI, [M þ Na]þ), 465.0855.
Step 2: A mixture of 8a (1.4 g, 4.39 mmol) and HBr (48% in
water, 20 mL) in 20 mL of acetic acid was heated to 90 ꢀC
overnight. Seven mL of HBr (45%) in acetic acid was added, and
the reaction mixture was heated to reflux for 5 h. The reaction
mixture was poured into ice, extracted with EtOAc. The organic
was concentrated and purified by flash chromatography eluted
with EtOAc/hexane to give 4-[3-methyl-5-(trifluoromethyl)-
quinoxalin-2-yl]phenol (10a) (1.20 g, 90%) as a gummy solid.
1H NMR (DMSO-d6): δ 9.86 (s, 1H), 8.28 (d, J = 7.5 Hz, 1H),
8.14 (d, J = 7.3 Hz, 1H), 7.85 (t, J = 7.7 Hz, 1H), 7.62 (d, J =
8.7 Hz, 2H), 6.88 (d, J = 8.7 Hz, 2H), 2.74 (s, 3H). MS (ESI) m/z
305.1. HRMS: calcd for C16H11F3N2O þ Hþ, 305.08962; found
(ESI, [M þ H]þ), 305.0894.
3-Methyl-2-[30-(methylsulfonyl)biphenyl-3-yl]-5-(trifluorome-
thyl)quinoxaline (16). Step 1: 2-(3-methoxyphenyl)-3-methyl-
5-(trifluoromethyl)quinoxaline (8b) was prepared according to
the procedure of 13 (step 1) from 3-trifluoromethyl-benzene-1,2-
diamine and 1-(3-methoxyphenyl)propane-1,2-dione in 38%;
red powder. 1H NMR (DMSO-d6): δ 9.86 (s, 1H), 8.28 (d, J =
7.5 Hz, 1H), 8.14 (d, J = 7.3 Hz, 1H), 7.85 (t, J = 7.7 Hz, 1H),
7.62 (d, J = 8.7 Hz, 2H), 6.88 (d, J = 8.7 Hz, 2H), 2.74 (s, 3H).
MS (ES) m/z 319.1.
Step 3: A mixture of 10a (1.20 g, 3.93 mmol), anhydrous THF
(40 mL), and N-phenylbis(trifluoromethanesulfonamide) (2.11
g, 5.93 mmol) was cooled to 0 ꢀC upon which potassium tert-
butoxide (0.62, 5.53 mmol) was added. The resulting mixture
was stirred at 0 ꢀC for 1 h. Another portion of N-phenylbis-
(trifluoromethanesulfonamide) (2.11 g, 5.93 mmol) and potas-
sium tert-butoxide (0.62, 5.53 mmol) was added. After one more
hour, the reaction was quenched with water, partitioned be-
tween water and EtOAc, and the organic was dried over MgSO4.
The residue was subjected to flash silica gel chromatography
(hexane:EtOAc) to afford 4-[3-methyl-5-(trifluoromethyl)quin-
oxalin-2-yl]phenyl trifluoromethanesulfonate (11a) (1.2 g, 65%)
as a colored solid. 1H NMR (CDCl3): δ 8.29 (d, J = 8.3 Hz, 1H),
The regio-isomer 3-(3-methoxyphenyl)-2-methyl-5-(trifluo-
romethyl)quinoxaline (9b) was also isolated in 24% yield; red
powder. 1H NMR (DMSO-d6): δ 8.29 (d, J = 8.4, 1H), 8.17 (d,
J = 7.2 Hz, 1H), 7.90 (t, J = 8.0 Hz, 1H), 7.45 (t, J = 8.0 Hz,
1H), 7.31-7.28 (m, 2H), 7.11-7.08 (m, 1H), 3.79 (s, 3H), 2.73 (s,
3H). MS (ES) m/z 319.1.
The regio-chemistry for 8b and 9b was confirmed based on
10b and 27 as shown in the Supporting Information.

3302 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Hu et al.
Step 2: 3-[3-methyl-5-(trifluoromethyl)quinoxalin-2-yl]phe-
nol (10b) was prepared from 8b followed the procedure of 13
(step 2). 1H NMR (DMSO-d6): δ 9.70 (s, 1H), 8.36 (d, J = 7.7,
1H), 8.24 (d, J = 7.1 Hz, 1H), 7.92 (t, J = 7.9 Hz, 1H), 7.35 (t,
J = 7.9 Hz, 1H), 7.20-7.12 (m, 2H), 6.95-6.92 (m, 1H), 2.75 (s,
3H). MS (ES) m/z 304.7. HRMS: calcd for C16H11F3N2O þ Hþ,
305.08962; found (ESI, [M þ H]þ), 305.0899.
2.84 (s, 3H). MS (ESI) m/z 422.2. HRMS: calcd for C24H18-
F3N3O þ Hþ, 422.14747; found (ESI, [M þ H]þ), 422.1475.
HRMS: calcd for C24H18F3N3O þ Naþ, 444.12941; found (ESI,
[M þ Na] þ), 444.1294.
3-Ethyl-2-[30-(methylsulfonyl)biphenyl-4-yl]-5-(trifluoromethyl)-
quinoxaline (21). Step 1: 4-[3-ethyl-8-(trifluoromethyl)quin-
oxalin-2-yl]phenol (10d) was prepared according to the proce-
dure of 13 (step 1 and 2) from 2-nitro-3-(trifluoromethyl)-
aniline and 1-(4-methoxyphenyl)butane-1,2-dione in 7% yield
for the two steps; crystal solid. 1H NMR (CDCl3): δ 8.28 (d, J =
8.7 Hz, 1H), 8.05 (d, J = 6.8 Hz, 1H), 7.76 (d, J = 8.3 Hz, 1H),
7.68 (d, J = 8.3 Hz, 2H), 7.00 (d, J = 8.3 Hz, 2H), 3.21 (q, J =
7.3 Hz, 2H), 1.37 (t, J = 7.3 Hz, 3H). HRMS: calcd for C17H13-
F3N2O þ Hþ, 319.10527; found (ESI, [M þ H]þ), 319.1053.
The regio-isomer of 10d (compound 28) was also isolated in
14% yield; pale-yellow solid. 1H NMR (CDCl3): δ 8.31 (d, J =
8.3 Hz, 1H), 8.06 (d, J = 7.2 Hz, 1H), 7.75 (d, J = 7.7 Hz, 1H),
7.53 (d, J = 8.3 Hz, 2H), 6.93 (d, J = 8.3 Hz, 2H), 3.12 (q, J =
7.3 Hz, 2H), 1.38 (t, J = 7.3 Hz, 3H); HRMS: calcd for C17H13-
F3N2O þ Hþ, 319.10527; found (ESI, [M þ H]þ), 319.1053.
The regio-chemistry for 10d and 28 was assigned as shown in
the Supporting Information.
Step 3: 16 was prepared from 10b and 3-methylsulfonylphe-
nylboronic acid following the procedure of 13 (step 3 and step 4)
in 70% yield; white solid. 1H NMR (DMSO-d6): δ 8.33 (d, J =
8.5 Hz, 1H), 8.29-8.28 (m, 1H), 8.18 (d, J = 7.3 Hz, 1H),
8.11-7.70 (m, 8H), 3.20 (s, 3H), 2.85 (s, 3H). MS (ES) m/z 443.0.
HRMS: calcd for C23H17F3N2O2S þ Hþ, 443.10356; found
(ESI, [M þ Hþ]), 443.1037.
3-Methyl-2-[40-(methylsulfonyl)biphenyl-3-yl]-5-(trifluoro-
methyl)quinoxaline (17). Prepared according to the procedure of
compound 13 (step3 and step 4) from 10b and 4-methylsulfonyl-
phenylboronic acid in 35% yield; white solid. 1H NMR (MeOH-
d4): δ 8.27 (d, J = 7.8 Hz, 1H), 8.12 (d, J = 6.9 Hz, 1H),
8.03-7.76 (m, 8H), 7.68 (t, J = 7.7 Hz, 1H), 3.12 (s, 3H), 2.81 (s,
3H). HRMS: calcd for C23H17F3N2O2S þ Hþ, 443.10356; found
(ESI, [M þ H]þ), 443.1038.
2-[30-(Ethylsulfonyl)biphenyl-4-yl]-3-methyl-5-(trifluoromethyl)-
quinoxaline (18). Prepared according to the procedure of com-
pound 13 (step 4) from 11a and 3-ethylsulfonylphenylboronic
acid in 75% yield; white solid. 1H NMR (DMSO-d6): δ 8.40 (d,
J = 8.0 Hz, 1H), 8.26 (d, J = 7.5 Hz, 1H), 8.23 (br s, 1H),
8.02-7.93 (m, 6H), 7.82 (d, J=7.8 Hz, 1H), 4.43 (q, J = 7.3 Hz,
2H), 2.82 (s, 3H), 1.18 (t, J=7.3 Hz, 3H). MS (ESI) m/z 457.2.
HRMS: calcd for C24H19F3N2O2S þ Hþ, 457.11921; found
(ESI, [M þ H]þ), 457.1199. HRMS: calcd for C24H19F3N2O2S
þ Naþ, 479.10115; found (ESI, [M þ Na]þ), 479.1012.
N-Methyl-40-[3-methyl-5-(trifluoromethyl)quinoxalin-2-yl]bi-
phenyl-3-sulfonamide (19). Step 1, 2-(4-chlorophenyl)-3-methyl-
5-(trifluoromethyl)quinoxaline (8c) and 3-(4-chlorophenyl)-2-
methyl-5-(trifluoromethyl)quinoxaline (9c): 8c was prepared
from 1-(4-chlorophenyl)propane-1,2-dione and 3-(trifluorome-
thyl)benzene-1,2-diamine according to the procedure of 13 (step
1) in 36% yield; white solid. 1H NMR (CDCl3): δ 8.28 (d, J =
8.3 Hz, 1H), 8.10 (d, J = 7.3 Hz, 1H), 7.77 (d, J = 7.8 Hz, 1H),
7.63 (d, J = 8.3 Hz, 2H), 7.54 (d, J = 8.3 Hz, 2H), 2.84 (s, 3H).
HRMS: calcd for C16H10ClF3N2 þ Hþ, 323.05573; found (ESI,
[M þ H] þ), 323.0557.
Step 2: 21 was prepared according to the procedure of 13 (step
3 and step 4) from 10d and 3-methylsulfonylphenyl boronic acid
in 33% yield; off-white solid. 1H NMR (DMSO-d6): δ 8.40 (d,
J = 8.3 Hz, 1H), 8.28-8.26 (m, 2H), 8.17 (d, J = 8.0 Hz, 1H),
8.02-7.85 (m, 6H), 7.81 (t, J = 7.8 Hz, 1H), 3.39 (s, 3H), 3.12 (q,
J = 7.8 Hz, 2H), 1.31 (t, J = 7.8 Hz, 3H). MS (ESI) m/z 457.1.
HRMS: calcd for C24H19F3N2O2S þ Hþ, 457.11921; found
(ESI, [M þ H]þ), 457.1191.
2-[30-(Methylsulfonyl)biphenyl-4-yl]-5-(trifluoromethyl)quino-
xaline (22). Step 1: 2-(4-methoxyphenyl)-5-(trifluoromethyl)-
quinoxaline (8e) was prepared according to the procedure of
13 (step 1) from 2-nitro-3-(trifluoromethyl)aniline and 2-(4-me-
thoxyphenyl)-2-oxoacetaldehyde as a solid in 6% yield. 1H
NMR (DMSO-d6): δ 9.72 (s, 1H), 8.40-8.35 (m, 3H), 8.21 (d,
J=7.1 Hz, 1H), 7.98 (t, J=7.3 Hz, 1H), 7.18 (d, J=8.8 Hz, 2H),
3.88 (s, 3H). MS (ES) m/z 304.9. The regio-isomer 2-(4-meth-
oxyphenyl)-8-(trifluoromethyl)quinoxaline (9e) was also iso-
1
lated as a pale-yellow solid in 59% yield. H NMR (DMSO-
d6): δ 9.72 (s, 1H), 8.41-8.36 (m, 3H), 8.26 (d, J = 7.2 Hz, 1H),
7.92 (t, J = 7.9 Hz, 1H), 7.20 (d, J = 7.9 Hz, 2H), 3.88 (s, 3H).
MS (ES) m/z 304.9;
The regio-chemistry for 8e and 9e was assigned as shown in
the Supporting Information.
The regio-isomer (9c) was also isolated in 32% yield; white
solid. 1H NMR (CDCl3): δ 8.23 (d, J = 8.3 Hz, 1H), 8.07 (d, J =
7.3 Hz, 1H), 7.77 (d, J = 7.8 Hz, 1H), 7.75 (d, J = 8.3 Hz, 2H),
7.52 (d, J = 8.3 Hz, 2H), 2.86 (s, 3H). HRMS: calcd for C16H10-
ClF3N2 þ Hþ, 323.05573; found (ESI, [M þ H]þ), 323.0557.
The regio-chemistry of 8c and 9c was assigned as shown in the
Supporting Information.
Step 2: 22 was prepared according to the procedure of 13 (step
2, step 3, and step 4) from 8e and 3-methylsulfonylphenyl
boronic acid in 9% yield; off-white solid. 1H NMR (CDCl3): δ
9.54 (s, 1H), 8.39-8.35 (m, 3H), 8.26 (s, 1H), 8.12 (d, J = 7.0 Hz,
1H), 8.00-7.96 (m, 2H), 7.89-7.85 (m, 3H), 7.72 (t, J = 7.8 Hz,
1H), 3.14 (s, 3H). HRMS: calcd for C22H15F3N2O2S þ Hþ,
429.08791; found (ESI, [M þ H]þ), 429.0883. HRMS: calcd for
C22H15F3N2O2S þ Naþ, 451.06985; found (ESI, [M þ Na]þ),
451.0698.
5-Chloro-3-methyl-2-[30-(methylsulfonyl)biphenyl-4-yl]quino-
xaline (23). A mixture of 1-(4-chlorophenyl)propane-1,2-dione
(3.0 g, 16.5 mmol), 3-methylsulfonylphenyl boronic acid (6.6 g,
33 mmol), KF (2.87 g, 49.5 mmol), dicyclohexyl(biphenyl-2-
yl)phosphine (1.16 g, 3.3 mmol), and Pd(OAc)2 (0.37 g, 1.65
mmol) in 100 mL of THF was stirred at room temperature for a
day. The reaction mixture was concentrated and purified by
flash chromatography eluted with EtOAc/hexane to give
1-(30-(methylsulfonyl)biphenyl-4-yl)propane-1,2-dione as a yel-
low solid (0.3 g, ∼90% pure). The yellow solid (0.15 g, 0.5 mmol)
was dissolved in ethanol (5 mL), and 3-chloro-benzene-1,2-
diamine (0.07 g, 0.5 mmol) was added. After being stirred at
room temperature overnight, the reaction mixture was concen-
trated and purified by flash chromatography eluted with
EtOAc/hexane to give 23 (0.025 g, 12%) as a slight-colored
solid. 1H NMR (DMSO-d6): δ 8.28 (s, 1H), 8.17 (d, J = 7.9 Hz,
Step 2: A mixture of 8c (0.06 g, 0.26 mmol), 3-(N-methylsul-
famoyl)phenylboronic acid (0.15 g, 0.70 mmol), K3PO4 (0.3 g,
1.4 mmol), dicyclohexyl(20,60-dimethoxybiphenyl-2-yl)pho-
sphine (0.04 g, 0.1 mmol), and Pd(OAc)2 (0.02 g, 0.09 mmol)
in 5 mL of 1-butanol was heated to 80 ꢀC for 1 h. The reaction
mixture was concentrated and purified by flash chromato-
graphy eluted with EtOAc/hexane to give 19 (0.036 g, 42%) as
a white solid. 1H NMR (DMSO-d6): δ 8.40 (d, J = 8.3 Hz, 1H),
8.26 (d, J=7.4 Hz, 1H), 8.14 (br s, 1H), 8.12 (d, J=7.7 Hz,
1H),8.08-7.92 (m, 5H), 7.82 (d, J = 6.5 Hz, 1H), 7.78 (t, J=7.7
Hz, 1H), 7.54 (s, 1H), 3.31 (s, 3H), 2.83 (s, 3H). MS (ESI) m/z
458.1;. HRMS: calcd for C23H18F3N3O2S þ Hþ, 458.11446;
found (ESI, [M þ H]þ), 458.1145. HRMS: calcd for C23H18F3-
N3O2S þ Naþ, 480.09640; found (ESI, [M þ Na]þ), 480.0965.
N-Methyl-40-[3-methyl-5-(trifluoromethyl)quinoxalin-2-yl]bi-
phenyl-3-carboxamide (20). Prepared according to the procedure
of compound 19 (step 2) from 8c and 3-(methylcarbamoyl)phe-
nylboronic acid in 48% yield; white solid. 1H NMR (DMSO-d6):
δ 8.62-8.58 (m, 1H), 8.40 (d, J = 7.5 Hz, 1H), 8.27-8.23 (m,
2H), 7.97-7.89 (m, 7H), 7.61 (t, J = 7.7 Hz, 1H), 3.23 (s, 3H),

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3303
1H), 8.10 (d, J = 8.3 Hz, 1H), 8.04-7.90 (m, 6H), 7.82 (d, J =
7.0 Hz, 1H), 7.80 (d, J = 7.7 Hz, 1H), 3.34 (s, 3H), 2.83 (s, 3H).
MS (ESI) m/z 409.1. HRMS: calcd for C22H17ClN2O2S þ Hþ,
409.07720; found (ESI, [M þ H]þ), 409.0772. HRMS: calcd for
C22H17ClN2O2S þ Naþ, 431.05914; found (ESI, [M þ Na]þ),
431.0591.
In Vitro Assays. LXRR and LXRβ binding assays (Table 1),
Gal4 transactivation assays in huh-7 liver cells (Table 2 and 3),
ABCA1 gene regulation in THP-1 cells (Table 2), C-efflux, and
NHR cross reactivity screens were performed as described
previously.8 Gal4 transactivation assays in kidney HEK-293
(Table 3) were performed as described in ref 23.
The regio-isomer 5-chloro-2-methyl-3-[30-(methylsulfonyl)-
biphenyl-4-yl]quinoxaline (29) was also isolated as a slight-
colored solid (30 mg, 14% yield). 1H NMR (DMSO-d6): δ
8.29-8.28 (m, 1H), 8.16 (d, J = 6.8 Hz, 1H), 8.07 (d, J = 8.4
Hz, 1H), 8.05-7.70 (m, 8H), 3.34 (s, 3H), 2.82 (s, 3H). MS (ESI)
m/z 409.1. MS (ESI) m/z 447.1. HRMS: calcd for C22H17-
ClN2O2S þ Hþ, 409.07720; found (ESI, [M þ H]þ), 409.0775.
HRMS: calcd for C22H17ClN2O2S þ Naþ, 431.05914; found
(ESI, [M þ Na]þ), 431.0587.
TG Accumulation in HepG2 Cells. Results for compounds in
Table 2 were obtained using methods previously described.24
Peptide Recruitment Assay. Results for compounds in
Figure 3 were obtained using methods previously described.11b
LDLR -/- Mouse Inhibition of Lesion Progression Model.
Eight-week old male LDLR-/- mice were fed atherogenic diet
or diet supplemented to deliver compounds 1b or 13 for 8 weeks.
Aortas were obtained and analyzed using methods previously
described.8
3-Methyl-2-[30-(methylsulfonyl)biphenyl-4-yl]quinoxaline-5-
carbonitrile (24). Step 1: 2-(4-chlorophenyl)-3-methylquinox-
aline-5-carbonitrile (8f) was prepared according to the proce-
dure of 18 (step 1) from 1-(4-chlorophenyl)propane-1,2-dione
and 3-cyano-benzene-1,2-diamine in 20% yield; white solid. 1H
NMR (CDCl3): δ 8.33 (d, J = 8.4 Hz, 1H), 8.15 (d, J = 7.2 Hz,
1H), 7.78 (d, J = 8.4 Hz, 1H), 7.65 (d, J = 8.6 Hz, 2H), 7.54 (d,
J = 8.6 Hz, 2H), 2.88 (s, 3H). HRMS: calcd for C16H10ClN3 þ
Hþ, 280.06360; found (ESI, [M þ H]þ) 280.0636.
Acknowledgment. We thank the Wyeth Discovery Analy-
tic Chemistry Department for the analytical data and Anita
Halpern and Dawn Savio for biological assay data. We
gratefully acknowledge Drs. Ron Magolda (deceased on
June 1, 2008), Magid Abou-Gharbia, Tarek Mansour, Steve
Gardell, and George Vlasuk for their support of this work.
Supporting Information Available: 1H-15N-gHMBC NMR
experiments, HPLC purity for the final compounds, Figure A
(Effects of 13 and 1b on Target Gene Relative Expression in
macrophages), and Figure B (Low Density array survey of gene
activation in liver indicates little regulation by compound 1b
and 13). This material is available free of charge via the Internet
at http://pubs.acs.org.
The regio-isomer 3-(4-chlorophenyl)-2-methylquinoxaline-5-
carbonitrile (9f) was also isolated in 20% yield; white solid. 1H
NMR (CDCl3): δ 8.28 (d, J = 8.4 Hz, 1H), 8.13 (d, J = 7.5 Hz,
1H), 7.80 (d, J = 7.3 Hz, 1H), 7.77 (d, J = 8.7 Hz, 2H), 7.55 (d,
J = 8.7 Hz, 2H), 2.87 (s, 3H). HRMS: calcd for C16H10ClN3 þ
Hþ, 280.06360; found (ESI, [M þ H]þ), 280.0636.
The regio-chemistry for 8f and 9f was assigned as shown in the
Supporting Information.
Step 2: 24 was prepared according to the procedure of 18 (step
2) from 8f and 3-methylsulfonylphenyl boronic acid in 29%
yield; white solid. 1H NMR (DMSO-d6): δ 8.46-8.44 (m, 2H),
8.29 (br s, 1H), 8.17 (d, J = 7.8 Hz, 1H), 8.01-7.94 (m, 6H), 7.81
(t, J = 7.8 Hz, 1H), 3.34 (s, 3H), 2.86 (s, 3H). HRMS: calcd
for C23H17N3O2S þ Hþ, 400.11142; found (ESI, [M þ H]þ,
400.1115.
References
(1) Fievet, C.; Staels, B. Liver X receptor modulators: Effects on lipid
metabolism and potential use in the treatment of atherosclerosis.
Biochem. Pharmacol. 2009, 77, 1316–1327.
(2) Zhang, Y; Chan, J.; Cummins, C. Liver X receptors as therapeutic
targets for managing cholesterol: implications for inflammatory
conditions. Clin. Lipidol. 2009, 4, 29–40.
(3) Gabbi, C; Warner, M.; Gustafsson, J. Minireview: liver X receptor
b: emerging roles in physiology and diseases. J. Mol. Endocrinol.
2009, 23, 129–136.
(4) (a) Bennett, D. J.; Carswell, E. L.; Cooke, A. J.; Edwards, A. S.;
Nimz, O. Design, structure-activity relationships and X-ray co-
crystallography of non-steroidal LXR agonists. Curr. Med. Chem.
2008, 15, 195–209. (b) Goodwin, B. J.; Zuercher, W. J.; Collins, J. L.
Recent advances in Liver X Receptor biology and chemistry. Curr.
Top. Med. Chem. 2008, 8, 781–791.
5-Methoxy-3-methyl-2-[30-(methylsulfonyl)biphenyl-4-yl]quino-
xaline (25). Step 1: 2-(4-chlorophenyl)-5-methoxy-3-methylqui-
noxaline (8g) was prepared according to the procedure of 18
(step 1) from 1-(4-chlorophenyl)propane-1,2-dione and 3-meth-
1
oxy-benzene-1,2-diamine in 21% yield; white solid. H NMR
(DMSO-d6): δ 7.79 (d, J = 8.1 Hz, 2H), 7.71 (t, J = 8.3 Hz, 1H),
7.64-7.60 (m, 3H), 7.27 (d, J = 7.8 Hz, 1H), 4.02 (s, 3H), 2.70 (s,
3H). MS (ESI) m/z 285.1. HRMS: calcd for C16H13ClN2O þ
Hþ, 285.07892; found (ESI, [M þ H]þ), 285.0794. HRMS: calcd
for C16H13ClN2O þ Naþ, 307.06086; found (ESI, [M þ Na]þ),
307.0609.
(5) Schultz, J. R.; Tu, H.; Luk, A.; Repa, J. J.; Medina, J. C.; Li, L. P.;
Schwendner, S.; Wang, S.; Thoolen, M.; Mangelsdorf, D. J.;
Lustig, K. D.; Shan, B. Role of LXRs in control of lipogenesis.
Genes Dev. 2000, 14 (22), 2831–2838.
The regio-isomer 3-(4-chlorophenyl)-5-methoxy-2-methyl-
quinoxaline (9g) was also isolated in 19% yield; white solid.
1H NMR (DMSO-d6): δ 7.77 (d, J = 8.2 Hz, 2H), 7.75 (t, J = 8.2
Hz, 1H), 7.63 (d, J = 7.8 Hz, 2H), 7.60 (d, J = 8.5 Hz, 1H), 7.25
(d, J = 7.9 Hz, 1H), 4.00 (s, 3H), 2.80 (s, 3H). MS (ESI) m/z
285.1. HRMS: calcd for C16H13ClN2O þ Hþ, 285.07892; found
(ESI, [M þ H] þ), 285.0789. HRMS: calcd for C16H13ClN2O þ
Naþ, 307.06086; found (ESI, [M þ Na] þ), 307.0609.
The regio-chemistry for 8g and 9g was assigned as shown in
the Supporting Information.
Step 2: 25 was prepared according to the procedure of 18
(step 2) from 8g and 3-methylsulfonylphenyl boronic acid in
88% yield; white solid. 1H NMR (DMSO-d6): δ 8.28 (s,
1H), 8.16 (d, J = 7.8 Hz, 1H), 8.00-7.90 (m, 5H), 7.81 (t, J=
7.8 Hz, 1H), 7.72 (t, J = 8.0 Hz, 1H), 7.63 (d, J = 8.4 Hz, 1H),
7.28 (d, J=7.9 Hz, 1H), 4.02 (s, 3H), 3.33 (s, 3H), 2.77 (s, 3H).
MS (ESI) m/z 405.2. HRMS: calcd for C23H20N2O3S þ Hþ,
405.12674; found (ESI, [M þ H]þ), 405.1267. HRMS: calcd
for C23H20N2O3S þ Naþ, 427.10868; found (ESI, [M þ Na]þ),
427.1087.
(6) Collins, J. L.; Fivush, A. M.; Watson, M. A.; Galardi, C. M.; Lewis,
M. C.; Moore, L. B.; Parks, D. J.; Wilson, J. G.; Tippin, T. K.; Binz,
J. G.; Plunket, K. D.; Morgan, D. G.; Beaudet, E. J.; Whitney, K.
D.; Kliewer, S. A.; Willson, T. M. Identification of a Nonsteroidal
Liver X Receptor Agonist through Parallel Array Synthesis of
Tertiary Amines. J. Med. Chem. 2002, 45, 1963–1966.
(7) Joseph, S. B.; McKilligin, E.; Pei, L.; Watson, M. A.; Collins, A.;
Laffitte, B. A.; Chen, M.; Noh, G.; Goodman, J.; Hagger, G. N.;
Tran, J.; Tippin, T. K.; Wang, X.; Lusis, A. J.; Hsueh, W. A.; Law,
R. E.; Collins, J. L.; Willson, T. M.; Tontonoz, P. Synthetic LXR
ligand inhibits the development of atherosclerosis in mice. Proc.
Natl. Acad. Sci. U.S.A 2002, 99, 7604–7609.
(8) Hu, B.; Collini, M.; Unwalla, R.; Miller, C.; Singhaus, R.; Quinet,
E.; Savio, D.; Halpern, A.; Basso, M.; Keith, J.; Clerin, V.; Chen,
L.; Resmini, C.; Liu, Q.-Y.; Feingold, I.; Huselton, C.; Azam, F.;
Farnegardh, M.; Enroth, C.; Bonn, T.; Goos-Nilsson, A.;
Wilhelmsson, A.; Nambi, P.; Wrobel, J. Discovery of Phenyl Acetic
Acid Substituted Quinolines as Novel Liver X Receptor Agonists
for the Treatment of Atherosclerosis. J. Med. Chem. 2006, 49,
6151–6154.
(9) Hu, B.; Jetter, J.; Kaufman, D.; Singhaus, R.; Bernotas, R.;
Unwalla, R.; Quinet, E.; Savio, D.; Halpern, A.; Basso, M.; Keith,
J.; Clerin, V.; Chen, L.; Liu, Q.-Y.; Feingold, I.; Huselton, C.;

3304 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Hu et al.
Azam, F.; Goos-Nilsson, A.; Wilhelmsson, A.; Nambi, P.; Wrobel,
J. Further Modifications on Phenyl Acetic Acid Based Quinolines
as Liver X Receptor Modulators. Bioorg. Med. Chem. 2007, 15,
3321–3333.
a New Class of Liver X Receptor Agonists with Selectivity for LXRb.
J. Med. Chem. 2007, 50, 4255–4259.
(15) (a) Kratzer, A.; Buchebner, M.; Pfeifer, T.; Becker, T.; Uray, G.;
Miyazaki, M.; Miyazaki-Anzai, S.; Ebner, B.; Chandak, P.;
Kadam, R.; Calayir, E.; Rathke, N.; Ahammer, H.; Radovic, B.;
Trauner, M.; Hoefler, G.; Kompella, U.; Fauler, G.; Levi, M.;
Levak-Frank, S.; Kostner, G.; Kratky, D. Synthetic LXR agonist
attenuates plaque formation in apoE-/- mice without inducing
liver steatosis and hypertriglyceridemia. J. Lip. Res. 2009, 50, 312–
326. (b) Quinet, E.; Savio, D.; Halpern, A.; Chen, L.; Miller, C.; Nambi,
P. Gene-selective modulation by a synthetic oxysterol ligand of the liver
X receptor. J. Lip. Res. 2004, 45, 1929–1942.
(10) Wrobel, J.; Steffan, R.; Bowen, M.; Magolda, R.; Matelan, E.;
Unwalla, R.; Basso, M.; Clerin, V.; Gardell, S.; Nambi, P.; Quinet,
E.; Reminick, J. I.; Vlasuk, G.; Wang, S.; Feingold, I.; Huselton, C.;
Bonn, T.; Farnegardh, M.; Hansson, T.; Nilsson, A.; Wilhelmsson,
A.; Zamaratski, E.; Evans, M. Indazole-based LXR Modulators
with Maintained Atherosclerotic Lesion Reduction Activity but
Diminished Stimulation of Hepatic Triglyceride Synthesis. J. Med.
Chem. 2008, 51, 7161–7168.
(11) (a) Katz, A.; Udata, C.; Ott, E.; Hickey, L.; Burczynski, M.;
Burghart, P.; Vesterqvist, O.; Meng, X. Safety, pharmacokinetics,
and pharmacodynamics of single doses of LXR-623, a novel liver
X-receptor agonist, in healthy participants. J. Clin. Pharmacol.
2009, 49, 643–649. (b) Quinet, E. M.; Basso, M. D.; Halpern, A. R.;
Yates, D. W.; Steffan, R. J.; Clerin, V.; Resmini, C.; Keith, J. C.;
Berrodin, T. J.; Feingold, I.; Zhong, W.; Hartman, H. B.; Evans, M. J.;
Gardell, S. J.; DiBlasio-Smith, E.; Mounts, W. M.; LaVallie, E. R.;
Wrobel, J.; Nambi, P.; Vlasuk, G. P. LXR ligand lowers LDL cholesterol
in primates, is lipid neutral in hamster, and reduces atherosclerosis in
mouse. J. Lip. Res. 2009, 50, 2358–2370.
(12) (a) Bradley, M. N.;Hong, C.;Chen, M.;Joseph, S. B.;Wilpitz, D. C.;
Wang, X.; Lusis, A. J.; Collins, A.; Hseuh, W. A.; Collins, J. L.;
Tangirala, R. K.; Tontonoz, P. Ligand activation of LXRb reverses
atherosclerosis and cellular cholesterol overload in mice lacking
LXRa and apoE. J. Clin. Invest. 2007, 117, 2337–2346. (b) By using
LXRa-/- mice and LXRb-/- mice the role of LXRa and LXRb has
been extensively studied, for a review on the subtype influence on TG
synthesis see: ref 3: (c) Lund, E. G.; Peterson, L. B.; Adams, A. D.; Lam,
M. N.; Burton, C. A.; Chin, J.; Guo, Q.; Huang, S.; Latham, M.; Lopez, J.
C.; Maenke, J. G.; Milot, D. P.; Mitnaul, L. J.; Rex-Rabe, S. E.; Rosa, R.
L.; Tian, J. Y.; Wright, S. D.; Sparrow, C. P. Different roles of liver X
receptor R and β in lipid metabolism: Effects of an R-selective and a dual
agonist in mice deficient in each subtype. Biochem. Pharmacol. 2006,
71, 453–463. (d) Quinet, E.; Savio, D.; Halpern, A.; Chen, L.; Schuster,
G. U.; Gustafsson, J.; Basso, M.; Nambi, P. Liver X Receptor (LXR)-
Regulation in LXR-Deficient Mice: Implications for Therapeutic Target-
ing. Mol. Pharmacol. 2006, 70, 1340–1349.
(16) Hu, B.; Unwalla, R.; Collini, M.; Quinet, E.; Feingold, I.; Goos-
Nilsson, A.; Wihelmsson, A.; Nambi, P.; Wrobel, J. Discovery and
SAR of cinnolines/quinolines as liver X receptor (LXR) agonists
with binding selectivity for LXRb. Bioorg. Med. Chem. Lett. 2009,
19, 3519–3527.
(17) Travins, J.; Bernotas, R.; Kaufman, D.; Quinet, E.; Nambi, P.;
Feingold, I.; Huselton, C.; Wilhelmsson, A.; Goos-Nilsson, A.;
Wrobel, J. 1-(3-Aryloxyaryl)benzimidazole sulfones are liver X
receptor agonists. Bioorg. Med. Chem. Lett. 2010, 20, 526–530.
(18) Travins, J.; Bernotas, R.; Unwalla, R.; Evans, M.; Quinet, E.;
Basso, M.; Nambi, P.; Feingold, I.; Wilhelmsson, A.; Goos-Nils-
son, A.; Wrobel, J. Quinazoline biarylether sulfones as LXR
ligands with improved LXRb selectivity and lower triglyceride
potential, Abstracts of Papers, 238th ACS National Meeting,
Washington, DC, August 16-20, 2009, MEDI-390.
(19) Bernotas, R.; Singhaus, R.; Kaufman, D.; Ullrich, J.; Fletcher, H.;
Quinet, E.; Nambi, P.; Unwalla, R.; Wilhelmsson, A.; Goos-
Nilsson, A.; Farnegardh, M.; Wrobel, J. Biarylether amide quino-
lines as liver X receptor agonists. Bioorg. Med. Chem. Lett. 2009,
19, 1663–1670.
(20) GLIDE, Version 4.5; Schrodinger, LLC: Portland, OR, 2007.
(21) Hu, B.; Bernotas, R.; Unwalla, R.; Collini, M.; Quinet, E.; Fein-
gold, I.; Goos-Nilsson, A.; Wilhelmsson, A.; Nambi, P.; Evans, M.;
Wrobel, J. Quinoline-3-carboxamide Containing Sulfones as Liver
X Receptor (LXR) Agonists with Binding Selectivity for LXRb
and Low Blood Brain Penetration. Bioorg. Med. Chem. Lett. 2010,
20, 632–635.
(22) Both compound 13 and 16 had good mouse liver microsome
stability (T1/2>30 min) but low solubility in the Gal4 biological
buffer (∼0.1 mM). No significant difference was found for the
uptake of these two compounds in the huh-7 cells (0.89 mmol in
cells for 13 vs 0.63 mmol for 26 after 1 h treatment; 3.79 mmol in
cells for 13 vs 1.35 mmol for 26 after 7 h treatment).
(23) Pawar, A.; Xu, J.; Jerks, E.; Mangelsdorf, D.; Jump, J. Fatty Acid
Regulation of Liver X Receptors (LXR) and Peroxisome Prolif-
erator-activated Receptor R (PPARa) in HEK293 Cells. J. Biol.
Chem. 2002, 277, 39243–39250.
(24) Hu, B.; Quinet, E.; Unwalla, R.; Collini, M.; Jetter, J.; Dooley, R.;
Andrake, D.; Nogle, L.; Savio, D.; Halpern, A.; Goos-Nilsson, A.;
Wilhelmsson, A.; Nambi, P.; Wrobel, J. Carboxylic Acid Based
Quinolines As Liver X Receptor Modulators That Have LXRb
Receptor Binding Selectivity. Bioorg. Med. Chem. Lett. 2008, 18,
54–59.
(13) (a) Faernegardh, M; Bonn, T.; Sun, S.; Ljunggren, J.; Ahola, H.;
Wilhelmsson, A.; Gustafsson, J.-A.; Carlquist, M. The Three-
dimensional Structure of the Liver X Receptor b Reveals a Flexible
Ligand-Binding Pocket That Can Accommodate Fundamentally
Different Ligands. J. Biol. Chem. 2003, 278, 38821–38828.
(b) Williams, S.; Bledsoe, R. K.; Collins, J. L.; Boggs, S.; Lambert,
M. H.; Miller, A. B.; Moore, J.; McKee, D. D.; Moore, L.; Nichols, J.;
Parks, D.; Watson, M.; Wisely, B.; Willson, T. M. X-ray Crystal
Structure of the Liver X Receptor b Ligand Binding Domain: Regula-
tion by a Histidine-Tryptophan Switch. J. Biol. Chem. 2003, 278,
27138–27143.
(14) Molteni, V.; Li, X.; Nabakka, J.; Liang, F.; Wityak, J.; Koder, A.;
Vargas, L.; Romeo, R.; Mitro, N.; Mak, P. A.; Seidel, H. M.;
Haslam, J. A.; Chow, D.; Tuntland, T.; Spalding, T. A.; Brock, A.;
Bradley, M.; Castrillo, A.; Tontonoz, P.; Saez, E. N-Acylthiadiazolines,