Discovery and characterization of potent and selective 4-oxo-4-(5-(5- phenyl-1,2,4-oxadiazol-3-yl)indolin-1-yl)butanoic acids as S1P1 agonists

Article abstract of DOI:10.1016/j.bmcl.2011.05.110

S1P₁ receptor driven lymphopenia has proven utility in the treatment of an array of autoimmune disease states. As a part of our efforts to develop potent and selective S1P₁ receptor agonists, we have identified a novel chemical series of 4-oxo-4-(5-(5-phenyl-1,2,4-oxadiazol-3-yl) indolin-1-yl)butanoic acid S1P₁ receptor agonists.

Full text of DOI:10.1016/j.bmcl.2011.05.110

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6013–6018
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery and characterization of potent and selective 4-oxo-4-(5-(5-phenyl-
1,2,4-oxadiazol-3-yl)indolin-1-yl)butanoic acids as S1P1 agonists
Daniel Buzard , Sangdon Han, Lars Thoresen, Jeanne Moody , Luis Lopez, Andrew Kawasaki à,
⇑
Thomas Schrader, Carleton Sage, Yinghong Gao §, Jeff Edwards –, Jeremy Barden, Jayant Thatte ||, Lixia Fu ,
Michelle Solomon, Ling Liu àà, Hussien Al-Shamma, Joel Gatlin, Minh Le, Charles Xing, Sheryll Espinola §§
Robert M. Jones
,
Arena Pharmaceuticals, 6166 Nancy Ridge Drive, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
S1P1 receptor driven lymphopenia has proven utility in the treatment of an array of autoimmune disease
states. As a part of our efforts to develop potent and selective S1P1 receptor agonists, we have identified a
novel chemical series of 4-oxo-4-(5-(5-phenyl-1,2,4-oxadiazol-3-yl)indolin-1-yl)butanoic acid S1P1
receptor agonists.
Received 23 March 2011
Revised 25 May 2011
Accepted 26 May 2011
Available online 6 June 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Sphingosine-1 phosphate receptor
S1P1 agonist
Immunomodulators
FTY720
Introduction
results from S1P1 dependent sequestration of lymphocytes in the
lymphoid tissues following phosphorylation of FTY720 to its active
S1P1 receptor mediated immunomodulation has emerged as an
intriguing mechanism for the treatment of autoimmune disease.1
FTY720 (GilenyaÒ, Fig. 1) recently became the first S1P1 receptor
agonist to gain approval in the US for relapsing forms of multiple
sclerosis (MS). In phase III clinical trials FTY720 demonstrated effi-
metabolite (FTY720P). Animal studies suggest this mechanism of
action may also be useful for the treatment of other autoimmune
diseases.3 The success of FTY720 in MS trials, and the potential
for use in other disease areas, has encouraged the development
of second generation S1P1 agonists.4 Achieving increased selectiv-
ity against the cardiac expressed S1P3 subtype has been a consis-
tent theme of this effort, as agonism of S1P3 causes bradycardia
in rodents. Transient bradycardia has been observed in humans
treated with FTY720. Efforts have also been directed at identifying
small molecules with improved pharmacokinetic properties.
FTY720 has a relatively long half-life in humans (6–9 days), and
normalization of peripheral lymphocyte levels upon cessation of
drug treatment requires 1–2 months.2 Other adverse reactions
noted in human trials of FTY720 include elevated liver enzymes,
macular edema, and respiratory effects. As a part of our efforts to
develop potent and selective S1P1 receptor agonists with desirable
pharmacokinetic and safety profiles, we have identified a novel
chemical series of 4-oxo-4-(5-(5-phenyl-1,2,4-oxadiazol-3-
yl)indolin-1-yl)butanoic acid S1P1 receptor agonists (e.g. 1). The
discovery and optimization studies pertaining to this new chemical
series are described herein.
cacy relative to both placebo and Interferon-c in reducing disease
relapse and delaying the progression of disability.2 This efficacy
⇑
Corresponding author.
E-mail address: dbuzard@arenapharm.com (D. Buzard).
Present address: Regulus Therapeutics, 3545 John Hopkins Court, San Diego, CA
92121, USA.
à
Present address: Focus Synthesis, 11475 Sorrento Valley Road, Suite 212, San
Diego, CA 92121, USA.
§
Present address: EMD Millipore, 10394 Pacific Center Court, San Diego, CA 92121,
USA.
–
Present address: Amylin Pharmaceuticals, 9360 Towne Center Drive, San Diego,
CA 92121, USA.
||
Present address: PRESCOS, LLC, 3550 General Atomics Court. M/S 02-422, San
Diego, CA 92121, USA.
Present address: Synthetic Genomics, 11149 North Torrey Pines Road, La Jolla, CA
92037, USA.
àà
Present address: Pfizer, 10646 Science Center Drive, San Diego, CA 92121-1150,
An internal HTS campaign identified sulfonamide 2 (Fig. 2) as a
USA.
§§
potent agonist of the human S1P1 receptor (EC50
, Melano-
Present address: Genomics Institute of the Novartis Research Foundation, 10675
John Jay Hopkins Drive, San Diego, CA 92121, USA.
phore5 = 6.7 nM). SAR studies were initiated on the three regions
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.05.110

6014
D. Buzard et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6013–6018
F3CO
O
N
OH
NH2
N
NC
N
CO2H
HO
O
FTY720
1
Figure 1. Structures of FTY720 and Compound 1.
of the molecule indicated in Figure 2. The sulfonamide linkage al-
lowed for facile variation in each of these areas via alkylation or
sulfonylation of the anilinic nitrogen. Replacement of the allyl
group with ethyl or n-propyl was tolerated, and small lipophilic
phenyl substitution was preferred (Me, Cl, CF3, etc.). In particular,
3,5-disubstitution on lower phenyl maintained the highest degree
of potency. Unfortunately, these early analogs were not effective at
lowering lymphocyte count in mice after oral delivery. Only when
the SAR diverged from the arylsulfonamide motif did we identify a
chemical series that ultimately led to success in vivo. Indazole 3
represents a transition point in the SAR. This compound was pre-
pared as a part of a small set of 5,6-fused bicyclic analogs and re-
tained moderate activity for the S1P1 receptor. Subsequent
replacement of the sulfonamide group with the 1,2,4-oxadiazole
motif, led to a compound of improved in vitro potency (Fig. 2, Com-
pound 4).
(Fig. 4). Given the robust efficacy of indoline 20 in the acute mPLL
experiment, a lead optimization study focused on the 4-oxo-4-(5-
(5-phenyl-1,2,4-oxadiazol-3-yl)indolin-1-yl) motif was initiated.
Preparation of Compound 20 and related analogs was per-
formed utilizing commercially available indoline-5-carbonitrile
21 as starting material (Fig. 5). Two synthetic routes were utilized,
allowing for late stage variability of either the 5-phenyl-1,2,4-oxa-
diazole or butyric acid group. Additionally, we found it beneficial to
have several methods for unmasking the carboxylic acid, as some
functional groups were incompatible under basic saponification
conditions. Attempts to prepare Compound 1 via hydrolysis of a
methyl ester with KOH were not productive due to competitive
reaction at the nitrile. It was possible to introduce the butyric acid
group directly upon treatment with succinic anhydride, or in two
step sequence utilizing 4-tert-butoxy-4-oxobutanoic acid as the
coupling reagent.
Introduction of an acid side chain onto the indazole ring was
examined (5–7, Fig. 3). It was anticipated that the carboxylic acid
group would mimic the phosphate residue found within sphingo-
sine-1-phosphate. Successful phosphate replacement has been
demonstrated in several other S1P1 agonists.4 Introduction of the
carboxylic acid was tolerated, producing analogs of comparable
potency to Compound 4. Notably, masking of the carboxylic acid
as an ester (8, 9) led to a decrease in in vitro potency, supporting
the notion of a specific interaction between the carboxylic acid res-
idue and the S1P1 receptor pocket.6 A small number of alternative
ring systems were examined, both with and without an acid side
chain (e.g. 10–20). Benzimidazole and indoline were effective
replacements for the indazole group (e.g. 14, 19, 20).
Selected compounds taken from Figure 3 were examined in an
abbreviated lymphocyte lowering experiment in BALB/C mice.
Lymphocyte levels were measured at a single time point (5 h), fol-
lowing a 1.0 mg/kg oral delivery of test compound. Compounds
that failed to cause lymphocyte lowering relative to vehicle at this
dose were excluded from further consideration. Often, failure to in-
duce lymphopenia was attributable to low oral bioavailability, as
was the case for benzimidazole 19. Indoline 20 was found to pro-
duce a maximal response in the lymphocyte lowering experiment
The 3-(1,2,4-oxadiazol-5-yl)-5-(trifluoromethoxy)benzonitrile
group was identified as a suitable replacement for the 5-(3,5-
bis(trifluoromethyl)phenyl) group (Fig. 5, Compound 1). This array
conferred a modest improvement in potency (hS1P1 EC50 = 1.2 nM)
and was thought to offer an advantage in physicochemical proper-
ties (Compound 1, clogP = 3.74; Compound 20, clogP = 4.68). Nota-
bly, the 3-methoxy-5-(1,2,4-oxadiazol-5-yl)benzonitrile variant
(Compound 24, Fig. 5), on the other hand, was less potent (hS1P1
EC50 [cAMP] = 10.3 nM).
A brief SAR study focused on the butyric acid group was per-
formed (Table 1). An amino acid motif was found to be a potency
driver (Compound 28); however, this compound did not produce
a substantial reduction in circulating lymphocytes in the mPLL as-
say. We attribute the lack of in vivo efficacy to low oral bioavail-
ability, as this compound is zwitterionic and poorly soluble in
water (<0.01 mg/mL). Other analogs prepared in this exercise
(25–27) exhibited equivalent or reduced potency compared to
Compound 20, thus we retained the butyric acid group as our pre-
ferred acid fragment.
An evaluation of compounds 20 and 1 across species and in a
human S1P subtype selectivity in vitro panel was undertaken.
S1P1 functional data for mouse, rat, dog, and monkey was
N
HN
anilinic region
N
N-alkyl region
HN
N
N
O
N
O S O
O S O
N
arylsulfone
region
Cl
Cl
F3C
CF3
F3C
CF3
2
3
hS1P1 EC50 = 372 nM
4
hS1P1 EC50 = 6.7 nM
hS1P1 EC50 = 9.4 nM
Figure 2. Identification the 5-(phenyl)-3-(1H-indazol-5-yl)-1,2,4-oxadiazole motif (4). Human S1P1 EC50 values were determined in Melanophore5 and are the mean of three
or more replicates.

D. Buzard et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6013–6018
6015
MeO2C
3
N
HO2C
N
N
N
N
EtO2C
HO2C
HO2C
3
N
HN
N
N
N
2
2
3
N
N
R
R
R
R
R
R
4
5
6
7
8
9
7.2 nM
9.8 nM
4.9 nM
4.2 nM
>100 µM
297 nM
H3CO
N
HN
HN
HN
HN
HN
N
N
N
R
N
R
N
R
11
1.0 µM
R
R
R
10
406 nM
12
24 nM
13
75 nM
14
4.3 nM
15
37 nM
HO2C
HO2C
HO2C
HN
O
N
O
N
HO2C
HO2C
O
N
N
N
N
N
R
R
R
R
R
16
54 nM
17
369 nM
18
41 nM
19
1.2 nM
20
4.6 nM
Figure 3. Compound 4 analogs, R = 3-(3,5-bis(trifluoromethyl)phenyl)-1,2,4-oxadiazole. Human S1P1 HTRF cAMP EC50 values are reported as the mean of three or more
replicates.
Further characterization of Compounds 20 and 1 revealed a
favorable ADME profile (Table 3). Both compounds were stable in
both human and rat liver microsomes (t1/2 > 60 min.). Not unex-
pectedly, 20 and 1 were found to be highly protein bound in both
human and rat. Pharmacokinetic analysis performed in male
Sprague Dawley rats revealed a low systemic clearance, low vol-
ume of distribution, and high Cmax. Consistent with the low volume
of distribution and systemic clearance, these compounds had good
oral bioavailability (30–50%) and a moderate terminal half-life
(3–8 h). As mouse was our pharmacologically relevant species,
mouse PK was obtained for Compound 1 and was found to be
similar to rat, with Compound 1 having a terminal half-life of
approximately 3 h.
5
4
3
2
1
0
Vehicle
(1 mg/kg)
20
Figure 4. Blood lymphocyte count at 5 h after a 0.00 (vehicle, 0.5% methylcellulose)
or 1.00 mg/kg oral dose of Compound 20 in male BALB/c mice (mean SD; n = 6 per
dose group).
Subsequent in vitro toxicity profiling (Table 4) demonstrated
that neither compound were inhibitors of the major CYP P450 iso-
forms, and no hERG interaction was observed in 3H-astemizole
determined in the HTRF cAMP assay, and both compounds main-
tained similar cross species activity (Table 2). Selectivity for the
human S1P1 receptor versus the other high affinity subtypes
(S1P2–5) was performed in the b-arrestin platform. Both com-
pounds appeared to be less potent S1P1 agonists in b-arrestin ver-
sus cAMP. Compounds 20 and 1 were selective against human S1P2
and S1P3,7 while each maintained some activity at the human S1P4
and S1P5 receptors. Avoiding S1P3 agonist activity had been a cri-
teria of ours given the associated bradycardia observed upon acti-
vation of this subtype.8 Compounds 20 and 1 were also examined
in a Cerep diversity panel comprised of ninety-seven binding
assays and thirteen enzymatic assays. Both compounds showed
no effect, or weak effects (0–50% inhibition or stimulation) at a
binding (IC50 > 100 lM) or patch clamp (<10% at 3.0 lM). We ob-
served no liabilities in our Essential Cell Function (ECF) panel
which measures several markers of cellular toxicity such as
changes in nuclear size (NI), membrane integrity (MI), intracellular
calcium release (IC), mitochondrial membrane potential (MMP).
Ames testing showed no frame-shift or base-pair mutations rela-
tive to vehicle in the five strains examined in the presence and
absence of S9 fraction (TA97a, TA98, TA100, TA1535, and
WP2uvrApKM101).
As Compound 1 satisfied our selectivity and safety require-
ments, this compound was advanced into a mouse lymphocyte
lowering dose response experiment. Mice received an oral dose
of Compound 1 at 0 (vehicle), 1, 3, or 10 mg/kg, after which blood
samples were taken up to 24 h post-dose for lymphocyte and drug
10 lM test concentration against this panel of receptors.

6016
D. Buzard et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6013–6018
NH2
HO
NC
a, b
N
N
N
H
Boc
21
22
f, g
c, d, e
R1
NH2
O
N
h, i
HO
N
N
N
O
R2
CO2tBu
N
CO2H
O
23
20 (R1=R2=CF3), hS1P1 EC50 [cAMP] = 4.6 nM
1 (R1=CN, R2=OCF3), hS1P1 EC50 [cAMP] = 1.2 nM
24 (R1=CN, R2=OCH3), hS1P1 EC50 [cAMP] = 10.3 nM
Figure 5. Synthetic methods. (a) Boc2O, THF (b) 50% aq. hydroxylamine, EtOH (c) ArCOCl, Et3N or ArCO2H, CDI (d) TFA (e) methyl 4-chloro-4-oxobutanoate then KOH, or
succinic anhydride (f) 4-tert-butoxy-4-oxobutanoic acid, EDC (g) 50% aq. hydroxylamine, EtOH (h) ArCOCl, Et3N or ArCO2H, CDI (i) TFA.
Table 1
Table 2
Acid side chain SAR. Human S1P1 HTRF cAMP EC50 values are reported as the mean of
In vitro S1P receptor functional data. EC50 values shown in this table are the mean of
three or more replicates
three or more replicates
F3C
O
Assay/parameter
20 (nM)
1 (nM)
N
S1P1 cross species [cAMP]
N
Human
Mouse
Rat
4.6
1.4
1.3
1.3
1.1
1.2
1.1
1.0
0.70
0.80
F3C
N
R
Dog
Compound
R
hS1P1 EC50 [nM]
Monkey
Human S1P subtypes [b-arrestin]
CO2H
S1P1
S1P2
S1P3
S1P4
S1P5
39
31
20
25
4.6
6.1
>100,000
3,100
160
>100,000
>100,000
1,330
109
O
CO2H
CO2H
88
26
27
6.6
O
O
Table 3
Compounds 20 and 1 ADME profile
CO2H
102
Assay/parameter
20
1
H2N
a
Microsomal stability T1/2
>60 min (HLM, RLM)
99.4% (h), 100% (r)
>60 min (HLM, RLM)
99.7% (h), 99.3% (r)
28
1.63
CO2H
% Protein binding (species)b
O
SD Rat PKc
Cl (L/h/kg)
Vss (L/kg)
0.038
0.564
1.74
3.0
16.02
8.5
0.045
0.275
3.82
2.0
20.2
2.9
Cmax (lg/mL)
plasma concentration measurements (Fig. 6). Compound 1 pro-
duced a dose dependent reduction in circulating lymphocytes up
to 16 h post-dose with lymphocyte levels returning to vehicle lev-
els by 24 h post-dose. An indirect pharmacokinetic/pharmacody-
namic (PK/PD) model was created to estimate the blood
concentration of the drug needed to suppress lymphocyte levels
by 50% (IC50). Compound 1 was determined to have an in vivo
IC50 value of 431 ng/mL, which is in good agreement with the
b-arrestin data after correcting for plasma protein binding.
Tmax (h)
AUClast (h ⁄
T1/2 (h)
%F
l
g/mL)
50.2
32.8
a
b
c
Human and rat liver microsomes.
Human and rat.
Male Sprague Dawley rat, vehicle = PEG400, Compound 20 1/1.5 mpk (IV/PO),
Compound 1 2/3 mpk (IV/PO).
The dose response experiment provided us with benchmark for
further optimization of this series. We sought to identify a mole-
cule with improved potency in the lymphocyte lowering experi-
ment. Toward this end, we initiated an SAR campaign focused on
3,4-disubstituted phenyl analogs (Table 5). Groups in the 4-posi-
tion consisted of alkyl, alkoxy, aryl, and cycloalkyl, and were paired
with small lipophilic groups occupying the adjacent position. Gen-
erally, small lipophilic groups in the 4-position compared favorably
with the 5-position although only one direct comparison was pos-
sible from this subset of compounds (39 and 24). Larger groups in
the 4-position were tolerated, and afforded several analogs of

D. Buzard et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6013–6018
6017
Table 4
Table 5
In vitro cytotoxicity profiling
Phenyl SAR. Human S1P1 HTRF cAMP EC50 values are reported as the mean of three or
more replicates
Assay/parameter
20
1
R1
CYP P450 inhibition
(IC50, lM)
1A2 > 50, 2D6 > 50
2C9 = 34.8, 2C19 > 50
3A4 > 50
1A2 > 50, 2D6 > 50
2C9 > 50, 2C19 > 50
3A4 > 50
O
N
R2
N
3H-Astemizole binding >100
(IC50 M)
>100
N
,
l
CO2H
Patch clamp (% inhib. at <10
3.0 M)
13
O
l
Compound
R1
R2
hS1P1 EC50 [nM]
0.82
ECF panel (IC50, nM)a
MI, NS, MMP, IC > 100
MI, NS, MMP, IC > 100
29
ClÀ
Micro Ames ( S9
fraction)
Negative: TA97a, TA98,
TA100 TA1535,
Negative: TA97a, TA98,
TA100 TA1535,
WP2uvrApKM101
WP2uvrApKM101
À
30
31
32
CF3
1.3
18
a
S
Essential Cell Function (ECF): membrane integrity (MI), nuclear size (NS),
mitochondrial membrane potential (MMP), intracellular calcium (IC).
O
CNÀ
O
ClÀ
45
33
34
35
36
37
38
39
40
41
42
43
44
MeOÀ
CNÀ
CHF2OÀ
1.5
1.1 (6.8)a
1.0
4.0
2.7
5.3
6.3
11
18
10
(CH3)2CHOÀ
MeOÀ
Vehicle
CF3OÀ
8
6
4
2
0
1
1
1
(1.00 mg/kg)
(3.00 mg/kg)
(10.0 mg/kg)
BrÀ
(CH3)2CHOÀ
À
À
CF3
CH3CH2
CF3OÀ
ClÀ
CNÀ
MeOÀ
MeOÀ
MeOÀ
MeOÀ
À
CF3
À
CH3
ClÀ
11
38
68
À
À
CH3SO2
HÀ
CH3CH2
CF3OÀ
0
5
10
15
20
25
a
value in parentheses is the EC50 determined in b-arrestin.
Nominal Time (hr)
Figure 6. Blood lymphocyte time-profile after
a
0.00 (vehicle), 1.00, 3.00 or
10.0 mg/kg oral dose of Compound 1 in male BALB/c mice. Symbols represent
observed values (mean SD) and the lines represent predicted values from the
pharmacokinetic/pharmacodynamic model. Vehicle = 0.5% methylcellulose in
water.
10
8
Vehicle
34 (0.3 mg/kg)
34 (1 mg/kg)
34 (3 mg/kg)
6
comparable potency to Compound 1 (e.g. 34 EC50 [cAMP] ꢀ 1 nM).
EC50 determination performed in b-arrestin; however, revealed
that Compound 34 was 4.5 fold more potent than Compound 1
in this assay (34, EC50 = 6.8 nM vs. 1, EC50 = 31 nM). It was hypoth-
esized that b-arrestin may be a more accurate predictor of in vivo
activity than the cyclase data, due to the high levels of S1P1 recep-
tor expression in the cyclase assay. Furthermore, the in vivo IC50
value for Compound 1 was in better agreement with the b-arrestin
data than cyclase.
To determine if an increase in b-arrestin potency coincided with
an improvement in in vivo potency, Compound 34 was selected for
IC50 determination in the mouse lymphocyte lowering experiment.
Given the improved potency associated with this compound, the
three dose groups were reduced to 0.3, 1.0, and 3.0 mg/kg, respec-
tively. A plot of the lymphocyte levels over time is depicted in
Figure 7. Complete suppression of circulating lymphocytes was
observed very quickly with levels returning to normal by approxi-
mately 8 h. Consistent with the b-arrestin data, Compound 34 was
determined to be more potent in vivo than Compound 1. The IC50
value for Compound 34 was estimated to be 22.6 ng/mL, which
translates into an approximate 19-fold improvement in in vivo
potency when compared to Compound 1.
4
2
0
0
10
20
30
Nominal Time (hr)
Figure 7. Blood lymphocyte-time profile after a 0.00 (Vehicle), 0.300, 1.00 and
3.00 mg/kg oral dose of Compound 34 in Male BALB/c Mice (mean SD). Symbols
represent observed values and the lines represent predicted values from the
pharmacokinetic/pharmacodynamic model.
demonstrated to possess desirable safety and ADME profiles, and
were shown to be efficacious at reducing circulating lymphocytes
in mice after a low oral dose. Further optimization of this chemical
series resulted in the identification of a new lead molecule possess-
ing significantly improved potency in the mouse lymphocyte low-
ering experiment.
In conclusion, a hit to lead optimization study was carried out
which resulted in the discovery of a new chemical series of
4-oxo-4-(5-(5-phenyl-1,2,4-oxadiazol-3-yl)indolin-1-yl)butanoic
acid S1P1 agonists. Analogs from within this series were
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.05.110.

6018
D. Buzard et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6013–6018
4. Buzard, D.; Jayant, T.; Lerner, M.; Edwards, J.; Jones, R. Exp. Opin. Ther. Pat. 2008,
References and notes
18, 1141.
5. McClintock, T. S.; Lerner, M. R. Brain Res. Prot. 1997, 2, 59.
6. Pham, T.; Fells, J.; Osborne, D.; North, E.; Naor, M.; Parrill, A. J. Mol. Graphics
Modell. 2008, 26, 1189.
7. Some insight into the selectivity observed against S1P3 may be partially
explained computationally. Please see the Supplementary data for docking
studies of Compound 1 into a homology model of the human S1P1 receptor.
8. Sanna, M. G.; Liao, J.; Jo, E.; Alfonso, C.; Ahn, M.; Peterson, M. S.; Webb, B.;
Lefebvre, S.; Chun, J.; Gray, N.; Rosen, H. J. Biol. Chem. 2004, 14, 13839.
1. (a) Brinkmann, V. Br. J. Pharmacol. 2009, 158, 1173–1182; (b) Chi, H. Trends
Pharmacol. Sci. 2011, 32, 16; (c) Cooke, N.; Zecri, F. An. Rep. Med. Chem. 2007, 42,
245.
2. <http://www.pharma.us.novartis.com/product/pi/pdf/gilenya.pdf>.
3. (a) Matsuura, M.; Imayoshi, T.; Chiba, K.; Okumoto, T. Inflamm. Res. 2000, 49,
404; (b) Kitabayashi, H.; Isobe, M.; Watanabe, N.; Suzuki, J.; Yazaki, Y.;
Sekiguchi, M. J. Cardiovasc. Pharmacol. 2000, 35, 410; (c) Daniel, C.; Sartory, N.;
Zahn, N.; Geisslinger, G.; Radeke, H.; Stein, J. J. Immunology 2007, 178, 2458.