2-Imino-thiazolidin-4-one derivatives as potent, orally active S1P 1 receptor agonists

Article abstract of DOI:10.1021/jm100181s

Sphingosine-1-phosphate (S1P) is a widespread lysophospholipid which displays a wealth of biological effects. Extracellular S1P conveys its activity through five specific G-protein coupled receptors numbered S1P₁ through S1P₅. Agonists of the S1P₁ receptor block the egress of T-lymphocytes from thymus and lymphoid organs and hold promise for the oral treatment of autoimmune disorders. Here, we report on the discovery and detailed structure-activity relationships of a novel class of S1P₁ receptor agonists based on the 2-imino-thiazolidin-4-one scaffold. Compound 8bo (ACT-128800) emerged from this series and is a potent, selective, and orally active S1P₁ receptor agonist selected for clinical development. In the rat, maximal reduction of circulating lymphocytes was reached at a dose of 3 mg/kg. The duration of lymphocyte sequestration was dose dependent. At a dose of 100 mg/kg, the effect on lymphocyte counts was fully reversible within less than 36 h. Pharmacokinetic investigation of 8bo in beagle dogs suggests that the compound is suitable for once daily dosing in humans.

Full text of DOI:10.1021/jm100181s

4198 J. Med. Chem. 2010, 53, 4198–4211
DOI: 10.1021/jm100181s
2-Imino-thiazolidin-4-one Derivatives as Potent, Orally Active S1P1 Receptor Agonists†
Martin H. Bolli,* Stefan Abele, Christoph Binkert, Roberto Bravo, Stephan Buchmann, Daniel Bur, John Gatfield,
ꢀ
Patrick Hess, Christopher Kohl, Celine Mangold, Boris Mathys, Katalin Menyhart, Claus Muller, Oliver Nayler,
Michael Scherz, Gunther Schmidt, Virginie Sippel, Beat Steiner, Daniel Strasser, Alexander Treiber, and Thomas Weller
€
Drug Discovery Chemistry, Actelion Pharmaceuticals Ltd., Gewerbestrasse 16, CH-4123 Allschwil, Switzerland
Received February 11, 2010
Sphingosine-1-phosphate (S1P) is a widespread lysophospholipid which displays a wealth of biologi-
cal effects. Extracellular S1P conveys its activity through five specific G-protein coupled receptors
numbered S1P1 through S1P5. Agonists of the S1P1 receptor block the egress of T-lymphocytes from
thymus and lymphoid organs and hold promise for the oral treatment of autoimmune disorders. Here,
we report on the discovery and detailed structure-activity relationships of a novel class of S1P1
receptor agonists based on the 2-imino-thiazolidin-4-one scaffold. Compound 8bo (ACT-128800)
emerged from this series and is a potent, selective, and orally active S1P1 receptor agonist selected for
clinical development. In the rat, maximal reduction of circulating lymphocytes was reached at a dose
of 3 mg/kg. The duration of lymphocyte sequestration was dose dependent. At a dose of 100 mg/kg,
the effect on lymphocyte counts was fully reversible within less than 36 h. Pharmacokinetic
investigation of 8bo in beagle dogs suggests that the compound is suitable for once daily dosing in
humans.
Introduction
Selective agonists of the sphingosine-1-phosphate-1 (S1P1)
receptor are of current therapeutic interest for their ability to
halt the exit of lymphocytes from lymph nodes and other
secondary lymphoid organs. This interruption of lymphocyte
migration, without affecting cell types of the innate immune
system (e.g., neutrophils or macrophages), and without affect-
ing cellular reactivity of lymphocytes to antigen challenge,
promises a new immunomodulatory therapeutic principle for
a variety of autoimmune diseases.
Sphingosine-1-phosphate (S1P, Figure 1) is a lysophospho-
lipid that mediates a wealth of biological effects via extra-
cellular signaling through five distinct G-protein coupled
receptors, numbered S1P1 through S1P5.1-6 While the S1P1,
S1P2, and S1P3 receptors are expressed ubiquitously, the S1P5
receptor, for instance, is mainly found in brain tissue, in
particular in oligodendrocytes. The S1P1 receptor couples
to GRi/o only, and downstream signaling has been reported
to be mediated by the rat sarcoma GTPase (Ras), mitogen-
activated protein kinases (MAPK), extracellular signal-regu-
lated kinases (ERK), phosphoinositide 3-kinases (PI3K), and
other pathways. The other S1P receptors are promiscuous
with respect to G-protein coupling and, in addition to
GRi/o, activate G12/13 and, in the case of S1P2 and S1P3,
Figure 1. Structure of sphingosine-1-phosphate (S1P), FTY720
(1), and p-FTY720 (p-1).
S1P regulates many cellular functions by activating cell
surface S1P receptors: effects on cell survival, proliferation,
differentiation, apoptosis, adhesion, migration, and chemo-
taxis are all well-described.3,7-10 The actions of nonselective
S1P receptor agonists have been described in detail for several
tissues and organs such as the cardiovascular system,11-17 the
lung,18,19 the cells of the immune system,20-22 the nervous
system,23,24 and, most recently, bone tissue.25 The observed
effects include angiogenesis, endothelial barrier enhancement,
airway and blood vessel constriction, alveolar epithelial bar-
rier disruption, heart rate modulation, neurite extension, and
bone homeostasis. It has been proven difficult to unambiguo-
usly assign these responses to one or more of the possible S1P
receptor subtypes.
also GRq
.
†This manuscript includes four X-ray crystallographic structures.
The corresponding data sets have been deposited at the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ,
United Kingdom, http://www.ccdc.cam.ac.uk/, under the following ID
numbers: compound 2, CCDC 7711347; compound 5b, CCDC 771348;
compound 8g, CCDC 771349; compound 8e, CCDC 771350.
Attention has focused on selective S1P1 receptor agonists.
Selective S1P1 receptor agonists have therapeutic potential to
treat a variety of immune-mediated diseases such as multiple
sclerosis, psoriasis, rheumatoid arthritis, or graft versus host
*To whom correspondence should be addressed. Phone: þ 41 61 565
65 70. Fax: þ 41 61 565 65 00. E-mail: martin.bolli@actelion.com.
r
pubs.acs.org/jmc
Published on Web 05/06/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10 4199
disease. Many cells, in particular cells of the immune system,
are able to migrate along a gradient of S1P by means of the
S1P-S1P1 receptor signaling axis.3,26,27 S1P1 agonists inter-
rupt lymphocyte migration and recruitment to sites of inflam-
mation by sequestering the lymphocytes from blood circu-
lation to lymphoid organs. This became evident when it was
discovered that FTY720 (1, Figure 1), a compound which has
long been known to sequester lymphocytes from circulating
blood to lymphoid tissue,28-31 is phosphorylated in vivo to
become a potent agonist (p-FTY720, p-1, Figure 1) at all S1P
receptors except for S1P2.32-35 Experiments with S1P1 recep-
tor knockout mice36,37 as well as S1P1 selective agonists such
as compound 1analogues38 or 5-(4-phenyl-5-(trifluoromethyl)-
thiophen-2-yl)-3-(3-(trifluoromethyl)phenyl)-1,2,4-oxadiazole
(SEW2871)39,40 have demonstrated that targeting S1P1 is suffi-
cient to cause lymphocyte sequestration to lymphoid organs.
To explain the lymphocyte sequestration observed with
compound 1 and other S1P agonists,39-41 two hypotheses
have been proposed. In the “functional antagonist” hypo-
thesis, the S1P concentration gradient42 that exists between
blood plasma and lymph was proposed as a driving force for
T-lymphocytes (in contrast to B-cells43) to exit the lymph
nodes and re-enter systemic circulation.44-46 The activation of
the S1P1 receptor is thought to lead to receptor internalization
and thus desensitization of the cell toward an external S1P
gradient. The second hypothesis claims the tightening of the
lymphatic endothelial cell junctions as the main reason for the
lymphocytes’ inability to leave the lymphoid tissue.13,34,40,41,47,48
An intriguing characteristic of the immunomodulation
achieved with S1P1 agonists is the fact that lymphocytes,
although hindered in their ability to travel through the body,
survive and are still able to react to an intruding microbial
insult.30,49,50 Recent studies have shown that a low dose of
compound 1 may even enhance clearance of a chronic viral
infection.51
In clinical trials, compound 1 has shown great promise in
the treatment of multiple sclerosis patients.52-54 Results
obtained in preclinical experiments suggest that compound 1
slows down disease progression not only by blocking T cell
migration to the central nervous system but also by directly
affecting brain cells such as astrocytes and oligodendrocytes.
S1P mediated activation of astrocytes and glial cells may
trigger endogenous repair mechanisms and support survival
of neural cells.55-60
Activation of the S1P3 pathway has been reported to cause
heart rate reduction,61-64 vaso- and bronchoconstriction, and
pulmonary epithelial leakage.65 In the context of an immu-
nomodulating drug, agonistic activity on S1P3 is therefore
deemed undesirable.
An overwhelming number of patents claiming novel
selective S1P1 receptor agonists have been published in the
past few years, and in addition to compound 1, several
molecules, e.g. KRP203 (Kyorin, Novartis), BAF312
(Novartis), CS-0777 (Daiichi Sankyo), and Ono-4641 (Ono
Pharmaceutical Company), entered clinical trials.66 Here we
report on the discovery, synthesis, and detailed structure-
activity relationsphips of a novel class of S1P1 agonists which
led to the selection of a compound suitable for clinical
development.
Figure 2. Structure of the high throughput screening hit.
the scaffold of compound 2 served as a starting point for
further optimization of this class.
Chemistry
Several synthetic routes give access to substituted 2-imino-
thiazolidin-4-ones.67-73 The cyclization of substituted thio-
ureas with a halo acetic acid derivative to form the correspond-
ing 2-imino-thiazolidin-4-one has been described almost one
and a half centuries ago,74-76 albeit the structure of the
isolated products was originally assigned, incorrectly, as
2-thiohydantoin. The correct constitution of the 2-imino-
thiazolidin-4-one, also known as pseudothiohydantoin, was
first proposed by Liebermann et al. in 1879.77,78 The discus-
sion on the tautomerism of this molecule continued for more
than a century.79 Today, commercial availability of a great
variety of aryl and alkyl isothiocyanates gives easy access to
thioureas and therefore makes the pathway in Scheme 1 very
attractive for a rapid assembly of a large array of substituted
2-imino-thiazolidin-4-one scaffolds. Thus, reacting an aryl or
alkyl isothiocyanate with a primary amine furnished the
corresponding thioureaderivative, which was directly cyclized
to the iminothiazolidinone by treating this intermediate with a
bromoacetic acid ester in the presence of pyridine. It is well-
known that this synthetic route usually gives a mixture of
the two isomeric 2-imino-thiazolidin-4-ones of the general
structures A and B.76,67 The regioselectivity of the cycliza-
tion is not only influenced by the reaction conditions (e.g.,
solvent, reaction temperature, presence or absence of a
base)67,80 and the choice of the halo acetic acid derivative81-83
but also by the nature of the two substituents R1 and R2
of the thiourea intermediate. For example, it has been
reported that under thermodynamic control the more
electron withdrawing substituent of R1 and R2 usually ends
up at N2 of the 2-imino-4-thiazolidinone.84,85 The regio-
selectivity issue is further complicated by the fact that
2-alkylimino-3-(hetero)aryl-4-thiazolidinones, which form
under kinetic reaction control, isomerize to the correspond-
ing thermodynamically more stable 2-(hetero)aryl-3-alkyl-
4-thiazolidinones, for instance, when heated in alcoholic
solutions, in particular in the presence of acids.67,80 In
addition to these reports, we observed that steric effects
Scheme 1. Synthesis of the 2-Imino-thiazolidin-4-one Scaffolda
High-throughput screening (HTS) of our in-house com-
pound collection identified several iminothiazolidinone deri-
vatives as S1P1 receptor agonists (Figure 2). Compound 2 was
the most potent representative in this hit cluster and showed
an EC50 of 200 nM on S1P1 and of about 1 μM on S1P3. Thus,
a (a) R2-NH2, MeOH, rt, 1-4 h; (b) ClCH2COOMe, pyridine, rt,
16-24 h; or (a) R2-NH2, DCM, rt, 1-3 h; (b) BrCH2COBr, pyridine, rt,
1-18 h.

4200 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10
Bolli et al.
Table 1. Regioselectivity of the 2-Imino-thiazolidin-4-one Scaffold
Synthesisa
similarly to the corresponding phenyl derivatives. Mechan-
istic considerations explaining these observations have been
discussed in detail by Pihlaja et al.81-83
The results obtained with the cyclohexyl (series i, Table 1)
can be rationalized following analogous reasoning. In the
series of 1-cyclohexyl-3-isopropyl thiourea, the two nitrogens
bearing the cyclohexyl group and the isopropyl group are
electronically and sterically similar, which results in an almost
1:1 distribution of the two cyclization products 3i and 4i.
In the case of the 1-cyclohexyl-3-n-propyl thiourea, the
n-propyl group is sterically less demanding and the cycliza-
tion therefore occurs in favor of the product 6i when methyl
bromoacetate is used as the cyclization reagent. However,
when bromoacetyl bromide is employed, the comparable
electronic nature82 rather than the steric environment of
the thiourea nitrogens appears to dominate the outcome of
the cyclization reaction and an almost 1:1 mixture of the
two products 5i and 6i is obtained. In the benzyl series
(3-6j), similar arguments may explain the regioselectivity
of the cyclization reaction. The nitrogen bearing the benzyl
group is more reactive than the one bearing the isopropyl
group and reacts with the ester function of methyl bro-
moacetate to form isomer 3j. In the case of 1-benzyl-3-
n-propyl-urea, the two thiourea nitrogens display similar
reactivities and an almost 1:1 mixture of 5j and 6j is
obtained for both cyclization reagents.
In general, the 2-alkylimino-3-phenyl-thiazolidin-4-ones
3a-g and 5a-g can be separated from their isomeric 3-al-
kyl-2-phenylimino-thiazolidin-4-ones 4a-g and 6a-g, re-
spectively, by a simple acid-base extraction. Under acidic
conditions, the compounds 3a-g and 5a-g are extracted into
the aqueous phase while the corresponding isomers 4a-g and
6a-g remain in the organic phase (diethyl ether). Hence,
separation of the two phases separates the two isomers. The
acidic aqueous phase is basified to enable extraction of the
desired compounds 3a-g and 5a-g into diethyl ether or
dichloromethane. This behavior is reflected in the different
pKa values of compounds of scaffold A and B. For instance,
the experimentally determined pKa value for compound 3a is
4.1 while it is clearly below 2 for its isomer 4a. In the case of the
2,3-dialkyl or the alkyl-benzyl substituted 2-imino-4-thiazoli-
dinones, the extractive separation is not possible as both
isomers have similar pKa values. The separation is more
challenging and can be effected by preparative (HPLC)
chromatography. Analytically, the two isomers of structure
A and B can be separated easily under LC-MS conditions (see
ExperimentalSection), withisomer A often beingclearly more
polar than isomer B.
R2
R
isopropyl
n-propyl
OCH3
3x:4x
Br
3x:4x
OCH3
5x:6x
Br
5x:6x
R1
x
phenyl
a
b
c
d
e
f
11:1b
4:1
26:1
42:1k
63:1
54:1
18:1
47:1
49:1
59:1
1:1
1:8
25:1
41:1
34:1
28:1
21:1d
47:1
22:1
29:1j
2:1
2-methyl-phenyl
3-methyl-phenyl
4-methyl-phenyl
2,6-dimethyl-phenyl
2-chloro-phenyl
2-methoxy-phenyl
3-pyridyl
1:10
1:8
13:1
18:1
1:13j
3:1
1:7
1:100c
1:24e
1:6
1:73j
1:100
2:1k
g
h
i
10:1
5:3f
cyclohexyl
benzyl
1:2g
j
100:1i
2:1h
1:1k
a Regio-isomer ratios A:B as assessed by LC-MS run under acidic
conditions (Zorbax SB-AQ column, see Experimental Section) at 230
nm; the following ratios have also been determined by 1H NMR analysis
of the crude reaction mixtures. b 10:1. c 1:42. d 15:1. e 1:19. f 5:3. g 1:2.
h 5:2. 100:1. j Ratio determined by H NMR only. k Ratio determined
using LC-MS run under basic conditions (Zorbax Extend C18 column,
see Experimental Section).
i
1
also play an important role in this reaction. To illustrate
our observations, the regioselectivity of some cyclization
reactions carried out in the course of this study are compiled
in Table 1.
When R1 represents an unsubstituted or monosubstituted
phenyl ring(compoundseriesa-d, f, and g in Table1), amines
R2-NH2 branched at the R-position such as an isopropyl
amine (but also a tert-butyl, a cyclopropyl, or an N,N-
dimethylamino amine, see Experimental Section) furnished
the desired isomers 3a-d,f,g in good yields and gave only
small amounts of the isomers 4a-d,f,g. This is in strong
contrast to the case wherein R2 represents a straight alkyl
chain such as an n-propyl group. In this case, the cyclization of
the corresponding thioureas with methyl chloroacetatealmost
exclusively led to the formation of the undesired isomers
6a-d,f,g. Hence, the second step had to be modified in order
to reverse the product distribution in favor of isomers 5a-d,f,g.
Work by Pihlaja et al.81-83 suggested to employ the more
reactive bromo acetyl bromide (Table 1, R =Br) instead of
methyl bromoacetate. This established the isomers 5a-d,f,g
as the major product regardless of the nature of R1 and R2. In
the case where R2 represents an isopropyl group, the regios-
electivity of the cyclization reaction using methyl bromoace-
tate was reduced for the two ortho-substituted phenyl groups
(series b and f) and even reversed in favor of isomer 4e in the
2,6-dimethylphenyl case (series e), indicating thatsteric effects
influence the regioselectivity of the cyclization step. In the
n-propyl series, the two ortho-substituted phenyl series e and f
showed a strong preference for scaffold B, i.e., compounds 6e
and 6f. Nevertheless, the use of bromoacetyl bromide in the
cyclization step for both the isopropyl as well as the n-propyl
series established the compounds 3b,e,f and 5b,e,f, respec-
tively, as the prevailing isomer. The series wherein R1 repre-
sents a 3-pyridyl residue (series h, Table 1) behaves very
The 1H NMR chemical shift of the C1-protons of the
former primary alkyl-amine component is diagnostically use-
ful todistinguishbetween the two isomers.80 Forisomer A, the
R-protons of the 2-alkylimino-substituent appear in the range
of 2.7-3.5 ppm. The corresponding protons of the 3-alkyl-
substituent of isomer B are shifted toward the lower field by as
much as 1.3 ppm. This assignment is corroborated by hetero-
nuclear multiple bond correlation (HMBC) spectra85 that
were acquired for several pairs of isomer A and B. While in
isomer B the R-protons of the 3-alkyl-substituent clearly
coupled to both the carbons C2 and C4, the corresponding
R-protons of the 2-alkylimino substituent in isomer A only
showed coupling to C2 of the thiazolidinone scaffold (atom
numbering see Scheme 1).
The structure of the compounds 5a-g is further corrobo-
rated by an alternative synthesis. The pathway illustrated for

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10 4201
Figure 3. As determined by X-ray structure analysis, there are two
atropisomers of compound 5b in the asymmetric unit of single
crystals of the compound. Anisotropic thermal ellipsoids are drawn
at a 50% level. Hydrogen atoms have been omitted for reasons of
clarity. For details, see the Experimental Section and Supporting
Information.
Figure 4. Molecular structure of compound 8g as determined by
X-ray structure analysis. Anisotropic thermal ellipsoids are drawn
at a 50% level. Hydrogen atoms have been omitted for clarity
reasons. For details, see the Experimental Section and Supporting
Information.
Scheme 2. Alternative Synthesis of Scaffold 5aa
Scheme 3a
a (a) ClCOCH2Cl, DCM, Et3N, -78 to 20°C, 3h; (b) NaH, n-propyl-
NCS, rt, 4 h.
a (a) 1-2 equiv substituted benzaldehyde, ethanol, piperidine, 80°C,
18 h; (b) 1-2 equiv substituted benzaldehyde, 2 equiv NaOAc, HOAc,
60-110°C, 3-24 h.
compound 5a in Scheme 286-88 establishes the structure of 5a
in an unambiguous fashion. In a first step, aniline is reacted
with chloroacetic acid chloride to furnish 2-chloro-N-phenyl-
acetamide 7. This compound is then treated with n-propyl-
isothiocyanate under basic conditions to presumably form
3-n-propyl-1-(2-chloro-acetyl)-1-phenyl-thiourea as an inter-
mediate which immediately cyclizes to the desired 2-propyl-
imino-3-phenyl-thiazolidin-4-one 5a in acceptable yields.
The product thus obtained is identical to the one isolated as
isomer A from the synthesis shown in Scheme 1. Finally, the
structural assignment, in particular the configuration of the
exocyclic double bond, was confirmed by X-ray analysis of
crystals of compound 5b. Interestingly, there are two
atropisomers89,90 present in the asymmetric unit of crystals
of 5b (Figure 3).
A wealth of conditions are known to effect the condensa-
tion of a 2-imino-4-thiazolidinone scaffold with an aliphatic
or aromatic aldehyde to furnish the corresponding 5-alkylidene-
or 5-arylidene 4-thiazolidinone.67,68 For this study, the con-
densation of substituted benzaldehydes was usually carried
out in acetic acid in the presence of sodium acetate at temper-
atures ranging from 60 to 110 °C. For compounds 8bf to 8bk
and 8bn to 8bp, the desired side chains at the 5-benzylidene-
moiety were introduced by either condensing the appro-
priately substituted benzaldehyde or by elaborating these
side chains in steps that followed the condensation reaction
(for details, see Experimental Section and Supporting Infor-
mation). Chromatographic behavior as well as 1H NMR
spectroscopic data of compounds 8 consistently showed the
presence of only one configurational isomer in the material
isolated from the condensation reaction (Scheme 3). The Z,Z
configuration of these products was confirmed by X-ray
analysis of crystals of compounds 2, 8e (see Supporting Infor-
mation), and 8g (Figure 4). The structure of all compounds 8
was therefore assigned to the Z,Z-isomer. Compounds bear-
ing an ortho-substituted 3-phenyl-substituent (e.g., 8t, 8z, 8aa,
etc.), showed axial chirality due to hindered rotation around
N3-C1(phenyl) bond in accordance with a recent report by
Erol et al.89 In the case of 8bo (vide infra), the two atrop-
isomers were separated by means of HPLC using a chiral
stationary phase. A chloroform or methanol solution of the
pure atropisomers re-equilibrated to the original 1:1 mixture
within about one week at room temperature.
Structure-Activity Relationships in Vitro
The following SARs of the 2- and 3-substituent at the
iminothiazolidinone scaffold are illustrated with a series of
compounds incorporating a 3-chloro-4-hydroxy-benzylidene
substituent at position 5 of the iminothiazolidine scaffold, as
this pattern generally afforded more potent compounds (e.g.,
8a) than the 4-hydroxy-3-methoxy-benzylidene substituent
present in 2. The compounds compiled in Table 2 illustrate
the SAR of the 2-substitutent of the core thiazolidinone.
Starting from compound 8a, first attempts aimed at replacing
the hydrazone-type substructure by a presumably more be-
nign (alkyl)-imine moiety. As illustrated by compound 8g, this
is indeed possible without loss of activity. In a second step, the
chain length at the exocyclic imine (N2) was investigated. At a
concentration of 10 μM, the compound lacking a substituent
at this nitrogen (8b) showed only residual activity on the S1P1
and S1P3 receptor. With respect to potency on S1P1, an
n-propyl (8e) or an isopropyl chain (8g) at the exocyclic imine
appear to be optimal. The potency optimum on the S1P3
receptor is shifted toward the more bulky sec-butyl (8h) and
tert-butyl (8i) chains. As a consequence, the S1P1/3 selectivity
observed for compounds 8e and 8g is reversed for 8h and 8i.
Cycloalkyl chains (as in 8j to 8m) did not bring any advantage
over straight or branched open chains.

4202 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10
Table 2. SAR of the 2-Imino Substituenta
Bolli et al.
Table 3. SAR of the 3-Substituenta
S1P1
S1P3
compd
R
EC50 [nM] σg EC50 [nM] σg
S1P1
S1P3
8n
8o
8p
8q
8r
iso-propyl
n-hexyl
cyclohexyl
58
282
10
1.32
1.04
1.19
1.15
162
118
1.11
1.12
1.09
1.19
compd
R
EC50 [nM]
σg
EC50 [nM]
σg
8a
8b
8c
8d
8e
8f
dimethylamino
hydrogen
methyl
62
>10000
990
1.36
140
>10000
8810
1229
189
1.35
18
125
ethoxycarbonylethyl
carboxyethyl
allyl
479
>10000
55
1.47
1.08
1.64
1.21
1.35
1.36
1.20
1.66
1.63
1.32
1.13
1.20
1.20
1.29
1.31
1.23
1.18
1.36
1.13
1.13
1.42
1.26
>10000
821
120
ethyl
n-propyl
n-butyl
186
67
8s
1.60
1.35
1.43
1.22
2.01
1.33
1.15
1.26
1.34
1.49
1.53
1.36
1.16
1.23
1.45
1.31
1.56
1.18
1.60
1.36
1.39
2.10
1.06
1.26
8g
8t
phenyl
47
34
112
47
264
120
2-methyl-phenyl
3-methyl-phenyl
4-methyl-phenyl
2,6-dimethyl-phenyl
2,3-dimethyl-phenyl
2,4-dimethyl-phenyl
2-ethyl-phenyl
2-chloro-phenyl
3-chloro-phenyl
3-chloro-2-methyl-
phenyl
139
200
8g
8h
8i
isopropyl
(rac)-sec-butyl
tert-butyl
cyclopropyl
cyclobutyl
cyclopentyl
cyclohexyl
8u
8v
110
78
102
147
59
96
88
307
8w
8x
8y
8z
154
72
8j
103
202
114
128
350
458
8k
8l
179
124
54
347
584
302
310
335
425
8m
8aa
8ab
8ac
a EC50 values for compounds 8 as determined in a GTPγS assay using
membrane preparations of CHO cells overexpressing the human S1P1
and S1P3 receptor, respectively. Data are given as geometric mean and
geometric standard deviation (σg) of at least three independent measure-
ments.
35
31
129
246
8ad
3-chloro-4-methyl-
phenyl
47
1.48
144
1.29
8ae
8af
8ag
8ah
8ai
8aj
8ak
8al
2-methoxy-phenyl
3-methoxy-phenyl
4-methoxy-phenyl
2,4-dimethoxy-phenyl
3-pyridyl
106
259
251
875
356
630
925
1815
1.17
1.45
1.59
1.73
2.05
1.30
1.30
1.08
428
514
127
525
228
674
183
212
1.64
1.08
1.13
1.26
1.03
1.12
1.42
1.13
To illustrate the SAR of the N3 substituent, a selection of
compounds, wherein the 2-isopropylimino and the 5-(3-
chloro-4-hydroxy-benzylidene) substituent of the core thiazol-
idinone are kept constant, is listed in Table 3. For the
following discussion, the iminothiazolidinone 8g featuring
an unsubstituted phenyl ring at N3 shall serve as a reference
compound. Compared to this reference, the isopropyl deri-
vative 8n shows almost identical potency on S1P1 and S1P3.
Interestingly, the allyl compound 8s displays an almost
identical affinity for S1P1 but is significantly less potent on
S1P3. The n-hexyl derivative 8o, on the other hand, is equally
potent on S1P3 but has a reduced affinity for the S1P1
receptor. Compound 8q featuring a six-atom chain incor-
porating an ester functionality has a similar affinity profile as
8o. However, if the ester function in 8q is hydrolyzed to give
the corresponding acid 8r, the affinity for both the S1P1 and
S1P3 receptor is lost. The compound with a cyclohexyl
substituent at N3 (8p) is clearly more potent on S1P1 and
S1P3 when compared to the phenyl analogue 8g. The com-
pounds 8t to 8ah illustrate the rather shallow SAR of the
substitution pattern at the N3-phenyl moiety. In general, one
or two methyl groups, a chlorine, or chlorine combined with a
methyl group or a substituent in the 2- and/or 3-position of
the phenyl ring lead to potency profiles similar to the one of
the unsubstituted analogue 8g. Ethyl and methoxy groups
attached to the phenyl ring gave less potent compounds. The
selectivity for the S1P1 receptor is most pronounced for
compounds bearing a 2-substituent (e.g., 8t, 8aa, 8ae). This
selectivity is reduced for the corresponding 3-substituted
analogues (e.g., 8u, 8ab, 8af). Compounds bearing a 4-sub-
stituent are the least selective within a given series
and represent, although less potent, balanced dual S1P1/3
(e.g., 8v) or even slightly S1P3 selective (8ag) receptor
benzyl
phenethyl
4-phenyl-butyl
a EC50 values for compounds 8 determined as described for Table 2.
agonists. Replacing the phenyl substituent in 8g by a 3-pyr-
idine ring (compound 8ai) reduced the compound’s potency,
in particular on the S1P1 receptor. Similarly, moving the
phenyl ring further away from N3 as exemplified by the
benzyl (8aj), the phenethyl (8ak), and the 4-phenyl-butyl
(8al) derivatives significantly reduced the compound’s affi-
nity for the S1P1 receptor.
We then turned our attention to optimizing the benzylidene
substituent in position 5 of the core iminothiazolidinone. The
compounds compiled in Tables 4 and 5 illustrate that the S1P1
receptor is generally more susceptible to changes in the nature
and position of substituents at the 5-benzylidene moiety.
Compounds wherein R (Table 4) represents an alkyl (e.g.,
ethyl, isopropyl) or a cycloalkyl (e.g., cyclopropyl, cyclohexyl)
group didnot show any activity onthe S1P1 receptor (datanot
shown). Compound 8am with an unsubstituted benzylidene
substituent at position 5 showed only weak activity on S1P1
and moderate activity on S1P3. For the following discussion,
the 4-hydroxy-3-chloro-benzylidene derivative 8g shall again
serve as a reference compound. Compounds 8an to 8aq
illustrate the effect of a 3-substituent next to the 4-hydroxy
group of the benzylidene moiety in 8g. A chloro or a methyl
substitutent clearly improve the compound’s potency on S1P1
and, in the case of the methyl group, also on S1P3. The

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10 4203
Table 5. SAR of the para-Substituent at the 5-Benzylidene Moietya
Table 4. SAR of 5-Benzylidene Substituenta
S1P1
S1P3
compd
R
EC50 [nM] σg EC50 [nM] σg
8am
8an
8ao
phenyl
1237 1.23
317 1.02
95 1.09
99 1.36
4-hydroxy-phenyl
4-hydroxy-3-fluoro-
phenyl
122
225
1.52
1.07
8g
4-hydroxy-3-chloro-
phenyl
47
37
1.35
1.46
1.95
120 1.23
50 1.49
335 2.12
8ap
8aq
4-hydroxy-3-methyl-
phenyl
4-hydroxy-3-methoxy-
phenyl
200
8ar
8as
8at
4-methoxy-phenyl
3,4-dimethoxy-phenyl
4-dimethylamino-
phenyl
201
224
148
1.32
1.17
2.03
106 1.27
1385 1.87
165 1.63
8au
8av
8aw
8ax
4-bromo-phenyl
3-methoxy-phenyl
3-hydroxy-phenyl
3-hydroxy-4-methoxy-
phenyl
1038 1.09
136 1.76
35 1.10
103 1.02
212 1.56
185
752
444
1.39
1.55
2.35
8ay
8az
8ba
8bb
3,4-dihydroxy-phenyl
2-methyl-phenyl
2-chloro-phenyl
2-methoxy-phenyl
>10000
>10000
6452 1.43
>10000
>10000
5791 1.49
2583 1.21
6689 1.42
a EC50 values for compounds 8 determined as described for Table 2.
a EC50 values for compounds 8 determined as described for Table 2.
Table 6. EC50 Values of Compounds 8 Combining Optimized Substi-
tutentsa
methoxy group in 8aq, however, reduces the compound’s
affinity for both S1P receptors.
As exemplified by the two compounds 8ar and 8as, repla-
cing the 4-hydroxy substituent of 8an and 8aq, respectively,
by a 4-methoxy group does not improve the compounds’
affinity for the two S1P receptors. In fact, the 3,4-dimethoxy-
phenyl derivative 8as is clearly less active on S1P3 when
compared to its 4-hydroxy-3-methoxy analogue 8aq. Repla-
cing the 4-hydroxy group in 8an by a 4-dimethylamino group
yields compound 8at, which shows an almost identical affi-
nity profile. A clear loss of affinity for the S1P1 receptor is
seen when the 4-hydroxy group in 8an is replaced by a
bromine atom (8au). Removing the 4-hydroxy substituent
in 8aq while keeping the 3-methoxy group to give compound
8av has no effect on S1P1 affinity but improves the com-
pound’s ability to activate the S1P3 receptor by about 1 order
of magnitude. In contrast to the 4-substituted series, convert-
ing the 3-methoxy group in 8av to a 3-hydroxy group (com-
pound 8aw) results in a significant loss of affinity for both
receptors. As shown with compound 8ax, the affinity can be
restored partially by adding a 4-methoxy substituent. Incor-
porating a 4-hydroxy substituent to compound 8aw, how-
ever, is detrimental for the compound’s S1P agonistic activity
(compound 8ay). As illustrated by the three compounds 8az,
8ba, and 8bb, a 2-substituent at the phenyl ring significantly
hampers the compound’s affinity for the S1P1 and S1P3
receptor.
a EC50 values for compounds 8 determined as described for Table 2.
A last group of compounds shall illustrate the SAR of a
number of polar chains that have been attached to the
5-benzylidene substituent. As compared to the phenol 8an,
the corresponding hydroxymethyl substituted derivative 8bc
is less potent on S1P1. Interestingly, the activity on S1P1 is
restored when the alcohol function is moved further away

4204 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10
Bolli et al.
from the phenyl ring by either an alkyl (as in 8bd and 8be)
oran alkoxy linker (as in8bf and 8bg). In fact, when compared
to 8an, the hydroxy-propoxy substituent in 8bg even slightly
improves the compound’s affinity for the S1P1 receptor.
On the other hand, there is a weak trend for an increasing
chain length to steadily improve the compound’s affinity
for the S1P3 receptor. While the glycerol derivative 8bh shows
an affinity for the S1P1 receptor, which is comparable to the
one of the phenol 8an, its isomer 8bi is considerably less active
on both S1P receptors. Compound 8bj highlights the impor-
tance of the correct attachment point of a hydroxy-alkyl side
chain.
With respect to potency on the S1P1 receptor and selectiv-
ity against S1P3, the one-dimensional SAR studies discussed
above can be summarized as follows: at the exocyclic imine
nitrogen (N2), an isopropyl or an n-propyl chain appears to
be optimal (Table 2). At N3, a 2- and/or a 3-substituted
phenyl ring substituted with either a methyl group or a
chlorine atom gave best compounds (Table 3). A 3-chloro
or 3-methyl substituent in combination with an hydroxy, an
hydroxyalkyl, or a mono- or dihydroxyalkoxy group in
position 4 of the benzylidene moiety gave most attractive
examples (Tables 4 and 5). With these data at hand, we set out
to prepare an array of compounds wherein the substitution
patterns found to be optimal in these individual one-dimen-
sional SAR studies were combined. As illustrated by com-
pounds 8bl to 8bn (Table 6), combining these substituents
indeed led to compounds of superior potency on the S1P1
receptor.
Evaluation of in Vivo Efficacy
The potency of compounds 8g and 8bl to 8bn justified
evaluating these compounds in vivo. Hence, the number of
lymphocytes circulating in blood was measured just before, 3,
6, and 24 h after oral administration of the compound to male
Wistar rats. Because of significant interindividual variability
and the circadian rhythm of the number of circulating lym-
phocytes, a compound showing relative changes in the range
of-20%toþ40%isconsideredinactive. Onthe other hand, a
lymphocyte count (LC) reduction in the range of -60 to
-75% represents the maximal effect to be observed under the
conditions of the experiment. Following these rules, com-
pounds 8g and 8bl are clearly inactive at an oral dose of
10 (Table 7) and 30 mg/kg (data not shown). Subsequent
pharmacokinetic (PK) experiments unravelled the reason for
the lack of oral efficacy of these compounds (Table 8). The
maximal plasma concentration (Cmax) after oral administra-
tion of 3 mg/kg of 8bl reached 8.1 ng/mL only. The clearance
ofthe compounddeterminedin acorrespondingivexperiment
was found to be close to total liver blood flow. These results
demonstrate that 8bl is not bioavailable. An almost identical
PK profile was seen for the related compound 8t. The high
clearance of 8t and 8bl found in the PK experiment was also
reflected in a high in vitro intrinsic clearance in rat hepato-
cytes. These results and the fact that phenols are well-known
substrates for conjugating enzymes (see e.g., refs 91-93) made
us speculate that the phenol function present in compound 8g
and 8bl (and 8t) is at least partially responsible for poor
bioavailability and hence the lack of in vivo efficacy. Indeed,
the compounds 8bf and 8bm (see Tables 5 and 6, respectively),
wherein the phenol function was masked by a 2-hydroxy-ethyl
group, showed significantly lower in vitro intrinsic and in vivo
clearance values (Table 8). Both compounds gave reasonable
concentrations in plasma (see e.g. Cmax) and acceptable
bioavailabilities. As a consequence, compound 8bm showed
a significant reduction in circulating lymphocytes 3 and 6 h
after oral administration of 10 mg/kg (Table 7). Because of its
almost 10 times lower potency on S1P1, compound 8bf
(Table 5) had to be dosed at 100 mg/kg in order to evoke a
comparable LC reduction (data not shown). As illustrated by
example 8bn, compounds incorporating a 2,3-dihydroxy-pro-
poxy side chain at the benzylidene moiety showed favorable
pharmacokinetic behavior too (Table 8). The racemate 8bn
maintained maximal LC reduction between 3 and 6 h after
oral administration of 10 mg/kg. In fact, from the array of
Table 7. Effect on Blood Lymphocyte Counts after Oral Administra-
tion of Selected Compounds 8a
lymphocyte counts
absolute [109 cells/L]
0 h
relative [% of t = 0 h]
compd
3 h
6 h
24 h interpretation
vehicle
8g
8bl
7.5 ( 0.2
6.9 ( 0.4
6.9 ( 0.5
6.9 ( 0.4
9.3 ( 0.4
7.0 ( 0.4
7.9 ( 0.4
3.5 ( 2
2 ( 3 15 ( 3
-16 ( 4 -15 ( 8 -9 ( 4
29 ( 7 24 ( 7 40 ( 8
-54 ( 4 -38 ( 5 28 ( 3
-67 ( 3 -71 ( 1 19 ( 4
-65 ( 4 -63 ( 3 -18 ( 8
-64 ( 2 -55 ( 3 15 ( 7
inactive
inactive
active
8bm
8bn
8bo
8bp
active
active
active
a Relative blood lymphocyte countchanges ( SEM 3, 6, and 24 h after
oral administration of vehicle (n = 30) or 10 mg/kg of a compound 8
(n = 6) to male Wistar rats, as compared to absolute lymphocyte counts
before compound administration (t = 0 h).
Table 8. DMPK Data of Selected Compounds 8a
intrinsic clearance
pharmacokinetic parameters
liver microsomes
[μL/min ꢀ mg protein]
hepatocytes
[μL/min ꢀ 106 cells]
F
[%]
Cmax (dose)
ng/mL (mg/kg)
t1/2
[h]
clearance
[mL/min/kg]
Vss
[L/kg]
compd
rat
dog
human
rat
nd
dog
8g
8tb
67
158
229
4
nd
nd
nd
nd
nd
nd
7
<4
27
7
nd
nd
nd
nd
nd
nd
1.0
nd
nd
nd
nd
105
92
nd
19.8
nd
<1%
<1%
72
11 (10)
8.1 (3)
499 (10)
129 (3)
311 (10)
409 (3)
1360 (3) c
0.7
0.2
2.0
1.4
1.6
1.3
10c
3.0
1.1
4.0
5.2
5.6
3.3
0.9c
8bl
8bfb
8bm
8bnb
8bo
4
12
nd
32
48
166
27
<4
4
23
90
9.1
3.7
41
25
1.3c
56
4
35
69c
a After oral and intravenous administration of the compounds 8 to male Wistar rats (n = 3) at 3 and 0.3 mg/kg, respectively. b Oral dose 10 mg/kg, iv
dose 1 mg/kg. c After oral and intravenous administration to fed male beagle dogs (n = 3) at 3 mg/kg and 0.3 mg/kg, respectively. Vss volume of
distribution; nd not determined.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10 4205
Table 9. Potency of Compound 8bo and S1P on All Five Human S1P Receptors as Determined in a GTPγS, and 33P-S1P Binding Assaya
GTPγS EC50 [nM]
binding IC50 [nM]
S1P3
S1P1
S1P2
S1P3
S1P4
S1P5
S1P1
S1P2
S1P4
S1P5
8bo
5.7b
1.2
>10000
105
1.3
1108c
1.2
59.1d
1.9
6.0
1.4
>10000
2068
1.2
1956
1.5
142
1.2
n=5
n=3
n=5
n=2
n=4
n=8
n=4
n=3
n=4
n=7
S1P
25.3
1.1
43.9
1.2
0.7
1.3
164
1.2
121
1.2
0.1
1.1
0.7
1.1
0.2
1.1
2.9
1.1
0.2
1.1
n=5
n=3
n=5
n=2
n=4
n=4
n=4
n=4
n=4
n=4
a Values given as geometric means (first row), corresponding geometric standard deviation (second row), and number of independent experiments
(third row). b The potency of 8bo on the S1P1 was slightly shifted to a lower value in this set of experiments, which is independent of the experiments
performed to establish the SAR (Table 6). c Compound 8bo acts as a partial agonist only, maximal activity of 8bo at 10 μM corresponds to 18% of the
effect induced by S1P. d Maximal effect at 10 μM reaches 42% of the effect of S1P.
S1P5 receptor. Although 8bo is a slightly more potent agonist
than S1P, it never reaches the efficacy of the natural ligand. In
conclusion, compound 8bo, in particular when compared to
S1P, is best described as a potent, selective, and fully effica-
cious S1P1 receptor agonist.
In vivo dose response experiments in male Wistar rats
clearly demonstrated that the maximal lymphocyte reduction
of -65 to -75% is reached with a dose of 3 mg/kg (Figure 5).
With increasing doses, the duration of action increased signi-
ficantly. While a dose of 3 mg/kg showed maximal LC
reduction for about 3 h, a dose of 100 mg/kg maintained a
maximal effect for about 24 h. Nevertheless, even at this high
dose, the LCs recovered completely within less than 36 h (data
not shown), demonstrating the rapid reversibility of the lym-
phocyte sequestration. More details on lymphocyte subtype
sequestrationafter multiple dosing in rats have been presented
Figure 5. Effect on lymphocyte counts in the blood 3, 6, 9, 12, 18,
and 24 h after oral administration of vehicle (0), and 3 (b), 10 (2),
30 (O) and 100 (9) mg/kg of 8bo to male Wistar rats.
elsewhere.94 The pharmacokinetic behavior of compound 8bo
was also studied in male beagle dogs (Table 8). These studies
revealed a favorable bioavailability of 69%. The maximal
2-imino-thiazolidin-4-ones prepared for this study, this com-
plasma concentration of 1360 ng/mL was reached between
pound emerged as particularly interesting due to its high
4 and 8 h after oral administration of 3 mg/kg. The half-life of
affinity for the S1P1 receptor, its selectivity against S1P3,
compound 8bo was found to be around 10 h. A single oral
dose of 5 mg/kg caused maximal reduction of LC for at least
and its in vivo efficacy. The individual enantiomers 8bo and
8bp are nearly equipotent in vitro (Table 6). However, the in
vivo efficacy, in particular the duration of action of the (R)-
enantiomer 8bo, appears to be slightly better than the one of
the (S)-enantiomer 8bp (Table 7). It was therefore decided to
study compound 8bo in more detail. First, the selectivity
profile of compound 8bo with respect to all five S1P receptors
24 h in beagle dogs. Comparison of the in vitro intrinsic
clearance in rat, dog, and human liver microsomes as well as
hepatocytes (Table 8) indicates that the dog is more suitable to
predict human PK properties. The in vivo efficacy data,
together with the pharmacokinetic profile, in particular in
was determined in a GTPγS as well as a 33P-S1P binding
the dog, suggested that 8bo is suitable for a once daily dosing
regimen in humans. This was confirmed in a single-ascending-
assayusing membranes ofrecombinant chinesehamster ovary
dose human study wherein compound 8bo showed good
exposure and a half-life of 22-33 h.95
(CHO) or human embryonic kidney (HEK293) cells (for
details see Experimental Section and Supporting Informa-
tion). The results for compound 8bo as well as those for the
Conclusion
natural ligand S1P are compiled in Table 9. The binding assay
confirms that the natural ligand S1P binds very tightly to all
five S1P receptors. This is in clear contrast to the behavior of
compound 8bo, which shows a strong preference for the S1P1
receptor not only in binding but also in the GTPγS assay.
Compound 8bo binds less tightly to S1P1 than S1P but at the
same timeappears tobethe more potent agonist inthe GTPγS
assay. In both the GTPγS and the binding assay, 8bo shows no
affinity for the S1P2 receptor. The absolute potency of 8bo on
S1P3 differs significantly in the two assays. As a consequence
the S1P1 selectivity is about 20- and almost 350-fold in the
GTPγS and the binding assay, respectively. The affinities for
the S1P4 and S1P5 receptor are in the μM and 100 nM range,
respectively. On S1P4, 8bo is not only less potent than S1P but
also significantly less efficacious. Even at 10 μM, 8bo shows
only 18% of the effect of S1P. A similar pattern is seen for the
In this account, we report on the discovery, preparation,
and characterization of a novel class of 2-imino-thiazolidin-4-
ones that serve as S1P1 receptor agonists. The factors that
influence the regioselectivity during the assembly of the
2-alkylimino-3-phenyl-thiazolidin-4-one scaffold are dis-
cussed. Detailed SAR studies led to the identification of the
potent, selective, and orally active S1P1 agonist 8bo. In
contrast to FTY720,32,34 8bo does not need to be phosphory-
lated in order to be active as an S1P1 receptor agonist.
Compound 8bo showed maximal reduction of circulating
lymphocytes at an oral dose of 3 mg/kg in Wistar rats. LCs
were reduced for about 24 h when an oral dose of 100 mg/kg
was administered. At this dose, lymphocyte sequestration was
fully reversible within 36 h, reflecting the rapid decrease in the
plasma concentration of 8bo. This indicates quick recovery of

4206 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10
Bolli et al.
lymphocyte circulation upon discontinuation of 8bo. In addi-
tion, 8bo showed a favorable pharmacokinetic behavior in the
dog. Meanwhile, compound 8bo (ACT-128800) has under-
gone extensive toxicity testing and is now evaluated in clinical
trials. First results of the entry-into-humans study demon-
strate that the compound is suitable for once-a-day dosing.95
More details shall be communicated in due course.
tion. The structure was solved by direct methods using SIR92,
and refinement was performed with CRYSTALS. Full matrix
least-squares refinement was performed with anisotropic tem-
perature factors for all atoms except hydrogen, which were
included at calculated positions with isotropic temperature
factors. Coordinates, anisotropic temperature factors, bond
lengths, and angles are deposited with the Cambridge Crystral-
lographic Data Centre. In vitro potency assessment: Data
(EC50) are given as geometric means (Xgeo) with geometric
standard deviation (σg). The upper and lower 95% confidence
Experimental Section
2
2
limits are calculated as Xgeo ꢀ σg and Xgeo/σg , respectively
(results not shown). GTPγS binding assays were performed in
96-well polypropylene microtiter plates in a final volume of
200 μL. Membrane preparations of CHO cells expressing re-
combinant rat, dog, or human S1P1, S1P2, S1P3, S1P4, or S1P5
receptors were used. Assay conditions were 20 mM Hepes,
pH 7.4, 100 mM NaCl, 5 mM MgCl2, 0.1% fatty acid free BSA,
1 or 3 μM GDP (for S1P1 or S1P3, respectively), 2.5% DMSO,
and 50 pM 35S-GTPγS. Test compounds were dissolved and
diluted and preincubated with the membranes, in the absence of
35S-GTPγS, in 150 μL assay buffer at room temperature for
30 min. After addition of 50 μL of 35S-GTPγS in assay buffer,
the reaction mixture was incubated for 1 h at room temperature.
The assay was terminated by filtration of the reaction mixture
through a Multiscreen GF/C plate (Filterplate from Millipore,
MAHFC1H60), prewetted with ice-cold 50 mM Hepes pH 7.4,
100 mM NaCl, 5 mM MgCl2, 0.4% fatty acid free BSA, using a
Cell Harvester from Packard Biosciences. The filterplates were
washed with ice-cold 10 mM Na2HPO4/NaH2PO4 (70%/30%)
and dried. The plates were sealed, 25 μL MicroScint20 was added,
and membrane-bound 35S-GTPγS was determined on the Top-
Count (Packard Biosciences). Specific 35S-GTPγS binding
was determined by subtracting nonspecific binding (the signal
obtained in the absence of agonist) from maximal binding
(the signal obtained with 10 μM S1P). The EC50 of a test com-
pound is the concentration of a compound inducing 50%
of specific binding. Each compound was measured in three
independent assays to obtain an average EC50 from triplicate
measurements.
Chemistry. All reagents and solvents were used as purchased
from commercial sources (Sigma-Aldrich, Switzerland, Lan-
caster Synthesis GmbH, Germany, Acros Organics, USA).
Moisture sensitive reactions were carried out under an argon
atmosphere. Progress of the reactions was followed either by
thin-layer chromatography (TLC) analysis (Merck, 0.2 mm
silica gel 60 F254 on glass plates) or by LC-MS. LC-MS:
Finnigan MSQ plus or MSQ surveyor (Dionex, Switzerland)
with HP 1100 binary pump and DAD (Agilent, Switzerland);
column, Zorbax SB-AQ, 5 μm, 120 A, 4.6 mm ꢀ 50 mm
(Agilent); gradient, 5-95% acetonitrile in water containing
0.04% of trifluoroacetic acid, within 1 min; flow, 4.5 mL/min;
tR is given in min. Purity of the target compounds was confirmed
on two additional columns: on a Zorbax Extend C18, 5 μm, 80
A, 4.6 mm ꢀ 50 mm (Agilent), eluting with a gradient of 5-95%
of acetonitrile in water containing 13 mM of NH3; and on a
Waters XBridge C18, OBD, 5 μm, 4.6 mm ꢀ 50 mm (Waters,
Switzerland), eluting with a gradient of 5-95% of acetonitrile in
water containing 13 mM of NH3. According to these three
LC-MS analyses, all compounds showed a purity >95% (UV
at 230 nm), about half of the compounds showed a purity >98%.
Purity and identity was further confirmed by NMR spectro-
scopy and, for some compounds, by combustion analysis (perfor-
med by Solvias AG, Basel, Switzerland). Prep HPLC: Waters
XBridge Prep C18, 5 μm, OBD, 19 mm ꢀ 50 mm, gradient
of acetonitrile in water containing 0.4% of formic acid, flow
75 mL/min. Melting points were determined by differential
scanning calorimetry (DSC) using a DSC822e/400 instrument
and the HSS7 sensor from Mettler Toledo, Switzerland. The
sample was weighed in a 40 μL aluminum crucible which was
open (pierced pan) under a flow of 15 mL/min of nitrogen
during the experiment. The scan goes from 20 to 320 °C at a rate
of 5 K/min. The peak temperature of the melting endotherm was
determined with the STARe software and was assigned to the
melting point. pKa values were determined with the GLpKa
instrument from Sirius (United Kingdom) using pH-metric
titrations with cosolvent; the cosolvent mixtures were water
with 30, 40, and 50% (w/w) of methanol, and the ionic strength
was adjusted with potassium chloride at a concentration of
0.15 M. The experimental temperature was 25 ( 1 °C. The
RefinementPro’s built-in Yasuda-Shedlovski extrapolation
technique at 0% methanol gave the pKa. NMR spectroscopy:
The in vivo efficacy of the compounds 8 was assessed by
measuring the circulating lymphocytes after oral administration
of 3-100 mg/kg of a compound 8 to normotensive male Wistar
rats. The animals were housed in climate-controlled conditions
with a 12 h light/dark cycle and had free access to normal rat
chow and drinking water. Blood was collected before and 3, 6,
and 24 h after drug administration. Full blood was subjected to
hematology using Beckman Coulter Ac T 5diff CP (Beckman
3
Coulter International SA, Nyon, Switzerland). All data are
presented as mean ( SEM. Statistical analyses were performed
by analysis of variance (ANOVA) using Statistica (StatSoft) and
the Student-Newman-Keuls procedure for multiple compar-
isons. The null hypothesis was rejected when p < 0.05. For
formulation, the compounds were dissolved in DMSO. This
solution was added to a stirred solution of gelatin (7.5%) in
water. The resulting milky suspension containing a final con-
centration of 5% of DMSO was administred to the animals by
gavage. A mixture of 95% of gelatin (7.5%) in water and 5% of
DMSO served as vehicle.
1
Bruker Avance II, 400 MHz UltraShield, H (400 MHz), 13C
(100 MHz), 19F (376 MHz) (Bruker, Switzerland); chemical
shifts are reported in parts per million (ppm) relative to tetra-
methylsilane (TMS), and multiplicities are given as s (singlet), d
(doublet), t (triplet), q (quartet), quint (quintuplet), h (hextet),
hept (heptuplet), or m (multiplet). For compounds 3a, 3b, 3h, 5a,
5b, 5i, 5j, 4a, 4b, 4h, 6a, 6b, 6i, and 6j, structural assignment was
corroborated by HMBC experiments, H-C correlations were
(Z)-2-Isopropylimino-3-phenyl-thiazolidin-4-one (3a) (Method A).
To a solution of isopropylamine (1.31 g, 1.91 mL, 22.2 mmol) in
methanol (25 mL), phenyl isothiocyanate (3.00 g, 2.66 mL,
22.2 mmol) is added portionwise. The mixture which became
slightly warm (approximately 30 °C) was stirred at room
temperature for 3.5 h before bromoacetic acid methyl ester
(3.39 g, 2.04 mL, 22.2 mmol) followed by pyridine (2.63 g,
2.68 mL, 33.3 mmol) was added. Stirring of the colorless
reaction mixture was continued for 16 h. The resulting fine
suspension was diluted with 1 N aq HCl (100 mL) and extracted
with diethyl ether (150 mL). The separated organic phase
contains crude 4a. The pH of the aqueous phase was adjusted
1
established by HSQC experiments. In the H and 13C NMR
spectra of compounds bearing an ortho substituent at the
3-phenyl ring of the 2-alkylimino-3-phenyl-thiazolidin-4-one
scaffolds, the signals of the 2-alkylimino substituent indicate
the presence of atropisomers (e.g., compounds 3b, 3q, 5b, 5f,
8t, 8y, 8bn, 8bo, 8bp). X-ray diffraction: To determine the
molecular structure of compounds 2, 5b, 8e and 8g, a crystal
of the corresponding compound was mounted on a Bruker
Nonius diffractometer equipped with a CCD detector and
reflections were measured using monochromatic Mo KR radia-

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10 4207
Figure 7. Atom numbering of compound 5a as used in the HMBC
assignment.
Figure 6. Atom numbering of compounds 3a and 4a as used in the
HMBC assignment.
6.98 (d, J=7.3 Hz, 2 H, H2b), 7.16 (t, J=7.5 Hz, 1 H, H2d), 7.37
(t, J=7.8 Hz, 2 H, H2c). 13C NMR (CDCl3): δ 11.29 (C3c), 20.56
(C3b), 32.72 (C5), 44.75 (C3a), 120.96 (C2c), 124.57 (C2d),
129.25 (C2b), 148.17 (C2a), 154.38 (C2), 171.83 (C4). HMBC
(CDCl3): H2b f C2a, C2c, C2d; H2c f C2a, C2b, C2d; H2d f
C2b, C2c; H3a f C2, C3b, C3c, C4; H3b f C3a, H3c; H3c f
C3a, C3b; H5 f C2, C4. LC-MS (ESþ): tR 0.96 min. m/z: 235 (M
þ H) (Figure 8).
to pH 8 by adding satd aq NaHCO3 solution. The aqueous
phase was extracted with diethyl ether (4 ꢀ150 mL). The
combined organic extracts containing crude 3a were dried over
MgSO4, filtered, and concentrated. The remaining crystalline
solid was washed with heptane and dried under high vacuum to
give 3a as an off-white crystalline solid (4.60 g, 88%), mp 149 °C.
1H NMR (CDCl3): δ 1.14 (d, J=6.3 Hz, 6 H, H2b), 3.52 (hept,
J=6.0 Hz, 1 H, H2a), 3.98 (s, 2 H, H5), 7.27-7.32 (m, 2 H, H3b),
7.37-7.44 (m, 1 H, H3d), 7.45-7.52 (m, 2 H, H3c). 13C NMR
(CDCl3): δ 23.35 (C2b), 32.58 (C5), 53.86 (C2a), 128.03 (C3b),
128.41 (C3d), 128.97 (C3c), 135.30 (C3a), 149.30 (C2), 171.39
(C4). HMBC (CDCl3): H2a f C2, C2b; H2b f C2, C2a; H3b f
C3a, C3d; H3c f C3a, C3d; H3d f C3a, C3b; H5 f C2, C4.
LC-LC-MS (ESþ): tR 0.59 min. m/z: 235 (M þ H). The organic
extract containing crude (Z)-3-isopropyl-2-phenylimino-thiazolidin-
4-one (4a) was concentrated and the crude byproduct was
purified by column chromatography on silica gel eluting with
heptane:ethyl acetate 3:1 to give pure 4a as a pale-yellow oil
Figure 8. Atom numbering of compound 6a as used in the HMBC
assignment.
1
(278 mg, 5%). H NMR (CDCl3): δ 1.55 (d, J=7.0 Hz, 6 H,
H3b), 3.75 (s, 2 H, H5), 4.90 (hept, J=6.8 Hz, 1 H, H3a), 6.97 (d,
J=7.5 Hz, 2 H, H2b), 7.16 (t, J=7.5 Hz, 1 H, H2d), 7.37 (t,
J=7.5 Hz, 2 H, H2c). 13C NMR (CDCl3): δ 18.72 (C3b), 32.50
(C5), 48.04 (C3a), 120.92 (C2c), 124.48 (C2d), 129.10 (C2b),
148.36 (C2a), 154.14 (C2), 171.92 (C4). HMBC (CDCl3): H2b f
C2a, C2c, C2d; H2c f C2a, C2b, C2d; H2d f C2b, C2c; H3a f
C2, C3b, C4; H3b f C3a; H5 f C2, C4. LC-MS (ESþ): tR 0.98
min. m/z: 235 (M þ H) (Figure 6).
(Z)-2-Propylimino-3-phenyl-thiazolidin-4-one (5a) (Method
C). A solution of aniline (20.4 g, 220 mmol) and triethylamine
(33.3 g, 330 mmol) in THF (750 mL) was cooled to -70 °C
before chloroacetyl chloride (24.8 g, 220 mmol) was added
slowly over a period of about 30 min with vigorous stirring.
The resulting suspension was warmed to room temperature,
diluted with water (400 mL), and extracted with ethyl acetate
(3 ꢀ 200 mL). The combined organic extracts were dried over
MgSO4, filtered, and concentrated. The remaining fine platelets
were washed with diethyl ether:heptane 1:1 and dried under high
vacuum to give 2-chloro-N-phenyl-acetamide as fine pale-beige
platelets (30.3 g, 82%), mp 136 °C. 1H NMR (CDCl3): δ 4.19 (s,
2 H), 7.19 (t, J=7.5 Hz, 1 H), 7.37 (t, J=7.5 Hz, 2 H), 7.56 (d, J=
8.0 Hz, 2 H), 8.34 (s br, 1 H). 13C NMR (CDCl3): δ 11.86, 23.57,
32.56, 54.29, 127.98, 128.63, 129.17, 135.17, 151.67, 171.37.
LC-MS (ESþ): tR 0.75 min. m/z: 170 (M þ H). To a solution
of 2-chloro-N-phenyl-acetamide (11.9 g, 70.0 mmol) and n-propyl
isothiocyanate (7.07 g, 70.0 mmol) in DMF (200 mL), NaH
(1.83 g, 55% dispersion in mineral oil, 42 mmol) was added in
four portions every 20-30 min. Upon complete addition of
NaH, stirring of the resulting brown mixture was continued for
2.5 h at rt. The mixture was diluted with ethyl acetate (500 mL)
and extracted with 1 N aq HCl (2 ꢀ 200 mL). The combined
aqueous extracts were extracted with ethyl acetate (2 ꢀ 200 mL)
after their pH had been adjusted to pH 8 by adding satd aq
NaHCO3-solution. These second organic extracts were washed
with water (2 ꢀ 200 mL) and concentrated. The remaining solid
was suspended in diethyl ether/hexane, filtered, washed with
additional diethyl ether/hexane, and dried to give 5a (4.85 g,
30%) as a pale-yellow powder, mp 143.6 °C. Spectroscopic data
were identical to those of 5a prepared according to method B.
(Z,Z)-2-(Dimethyl-hydrazono)-5-(4-hydroxy-3-methoxy-
benzylidene)-3-phenyl-thiazolidin-4-one (2) (Method D). A solution
of (Z)-2-(dimethyl-hydrazono)-3-phenyl-thiazolidin-4-one 9a
(118 mg, 0.50 mmol) and vanilline (84 mg, 0.55 mmol) in ethanol
(2 mL) and piperidine (100 μL) was stirred at 80 °C for 18 h. The
mixture was cooled to room temperature, diluted with diethyl
ether (2 mL), and the precipitate that formed was collected,
(Z)-2-Propylimino-3-phenyl-thiazolidin-4-one (5a) (Method
B). To a solution of phenylisothiocyanate (2.00 g, 14.8 mmol)
in dichloromethane (20 mL), n-propylamine (875 mg, 1.23 mL,
14.8 mmol) was added portionwise at 20 °C. The solution was
stirred at 20 °C for 15 min. The solution was cooled to 0 °C
before bromo-acetyl bromide (2.99 g, 1.29 mL, 14.8 mmol) was
added carefully such that the temperature did not rise above
5 °C. The reaction mixture was stirred at 0 °C for 15 min before
pyridine (2.40 g, 2.45 mL, 30.3 mmol) was added at 0 °C. The
mixture was stirred for another 15 min at 0 °C, then warmed to
20 °C and washed with water (10 mL). The aqueous layer is
extracted with dichloromethane (10 mL). The solvent of com-
bined organic extracts was evaporated under reduced pressure
to afford 5a (1.84 g, 53%) as an off-white solid, mp 138 °C. This
material was sufficiently pure (>90%) for further use in the
subsequent condensation step with substituted benzaldehydes.
1H NMR (CDCl3): δ 0.92 (t, J=7.3 Hz, 2 H, H2c), 1.61 (h, J=
7.0 Hz, 2 H, H2b), 3.28 (t, J=7.0 Hz, 2 H, H2a), 4.00 (s, 2 H, H5),
7.28-7.32 (m, 2 H, H3b), 7.37-7.45 (m, 1 H, H3d), 7.47-7.53
(m, 2 H, H3c). 13C NMR (CDCl3): δ 11.86 (C2c), 23.57 (C2b),
32.56 (C5), 54.29 (C2a), 127.98 (C3b), 128.63 (C3d), 129.17
(C3c), 135.17 (C3a), 151.67 (C2), 171.37 (C4). HMBC (CDCl3):
H2a f C2, C2b, C2c; H2b f C2a, C2c; H2c f C2a, C2b; H3b f
C3a, C3d; H3c f C3a, C3b; H3d f C3a, C3b; H5 f C2, C4.
LC-MS (ESþ): tR 0.60 min. m/z: 235 (M þ H) (Figure 7).
(Z)-3-Propyl-2-phenylimino-thiazolidin-4-one (6a). Method
A. From phenylisothiocyanate (34 mg, 0.25 mmol) and n-pro-
pylamine (15 mg, 0.25 mmol), colorless oil (50 mg, 85%) after
purification on prep TLC plates with heptane:ethyl acetate 4:1.
1H NMR (CDCl3): δ 1.00 (t, J=7.5 Hz, 3 H, H3c), 1.79 (h, J=
7.5 Hz, 2 H, H3b), 3.82 (s, 2 H, H5), 3.85 (t, J=7.5 Hz, 2 H, H3a),

4208 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10
Bolli et al.
washed with additional diethyl ether (2ꢀ10 mL), and dried under
high vacuum to give the title compound (150 mg, 81%) as a yellow
powder, mp 195 °C. 1H NMR (CDCl3): δ 2.57 (s, 6 H), 4.00 (s, 3
H), 7.04 (d, J=8.3 Hz, 1 H), 7.08 (s, 1 H), 7.23 (d, J=8.3 Hz, 1 H),
7.40-7.46 (m, 3 H), 7.49-7.56 (m, 2 H), 7.72 (s, 1 H). LC-MS
(ESþ): tR 0.96 min. m/z: 370 (M þ H). In another experiment
following the above procedure, compound 2 cocrystallized with 1
equiv of piperidine. For analytical data and X-ray crystallo-
graphic structure parameters of the piperidinium salt of 2, see
the Supporting Information.
(Z,Z)-5-(3-Chloro-4-hydroxy-benzylidene)-2-(dimethyl-hydra-
zono)-3-phenyl-thiazolidin-4-one (8a) (Method E). A solution of
(Z)-2-(dimethyl-hydrazono)-3-phenyl-thiazolidin-4-one 9a (235
mg, 1.00 mmol), 3-chloro-4-hydroxy-benzaldehyde (235 mg,
1.00 mmol), and sodium acetate (164 mg, 2.00 mmol) in acetic
acid (5 mL) is stirred at 85 °C for 18 h. The solution turned
yellow and a precipitate formed. The mixture was cooled to rt,
diluted with diethyl ether (150 mL), and washed with water (2 ꢀ
100 mL), satd aq NaHCO3 solution (1 ꢀ 100 mL), and water (2
ꢀ 50 mL). The organic extract was dried over MgSO4, filtered,
and concentrated. The remaining residue was suspended and
shortly refluxed in methanol (10 mL). The solid material was
collected, washed with methanol, and dried under high vacuum
to give the title compound as a bright-yellow powder (220 mg,
59%), mp >255 °C melting with decomposition. 1H NMR
(DMSO-d6): δ 2.45 (s, 6 H), 7.16 (d, J = 8.5 Hz, 1 H),
7.40-7.48 (m, 3 H), 7.48-7.57 (m, 3 H), 7.63 (s, 1 H), 7.68 (s,
1 H), 11.02 (s br, 1 H). LC-MS (ESþ): tR 1.00 min. m/z: 374 (M þ
H). Anal. (C18H16ClN3O2SCl): C, H, N, O, S, Cl.
tography on silica gel eluting with heptane:ethyl acetate 1:1 to
afford 3-chloro-4-(2-acetoxy-ethoxy)-benzaldehyde (6.44 g,
42%) as colorless solid. 1H NMR (CDCl3): δ 2.12 (s, 3H),
4.35-4.31 (m, 2H), 4.53-4.49 (m, 2H), 7.03 (d, J = 8.8 Hz,
1H), 7.75 (dd, J=1.8, 8.2 Hz, 1H), 7.91 (d, J=1.8 Hz, 1H), 9.85
(s, 1H). LC: tR 0.88 min. The title compound (276 mg, 80%) was
obtained as a pale-yellow foam after purification on prep. TLC
plates with heptane:EA 1:1 in analogy to compound 8bc starting
from 5b (200 mg, 0.81 mmol) and the above 3-chloro-4-(2-
acetoxy-ethoxy)-benzaldehyde (391 mg, 1.61 mmol). 1H NMR
(CDCl3): δ 0.94 (t, J=7.5 Hz, 3 H), 1.60-1.70 (m, 2 H), 2.22 (s, 3
H), 3.34-3.48 (m, 2 H), 4.05 (t br, J=4.3 Hz, 2 H), 4.24 (t, J=4.3
Hz, 2 H), 7.07 (d, J=8.5 Hz, 1 H), 7.21 (d, J=7.3 Hz, 1 H),
7.32-7.39 (m, 3 H), 7.48 (dd, J=8.5, 2.3 Hz, 1 H), 7.65 (d, J=2.3
Hz, 1 H), 7.69 (s, 1 H). LC-MS (ESþ): tR 1.04 min. m/z: 431 (M þ
H). Anal. (C22H23N2O3SCl): C, H, N, O, S, Cl, ash <0.2%.
rac-(Z,Z)-5-[3-Chloro-4-(2,3-dihydroxy-propoxy)-benzylidene]-
2-propylimino-3-o-tolyl-thiazolidin-4-one (8bn). Method E. Start-
ing from 5b (450 mg, 1.81 mmol) and rac-4-(2,3-dihydroxy-
propoxy)-3-chloro-benzaldehyde (835 mg, 3.62 mmol), the title
compound (262 mg, 32%) was obtained as a white powder after
extractive workup and purification of the crude product on prep.
TLC plates with toluene:ethyl acetate 1:3. 1H NMR (CDCl3): δ
0.94 (t, J=7.3Hz, 3 H), 1.60-1.70 (m, 2 H), 2.12 (t br, J=5.8 Hz, 1
H), 2.21 (s, 3 H), 2.76 (d br, J=3.8 Hz, 1 H), 3.34-3.48 (m, 2 H),
3.83-3.97 (m, 2 H), 4.17-4.27 (m, 3 H), 7.07 (d, J=8.5 Hz, 1 H),
7.21 (d, J=7.3 Hz, 1 H), 7.31-7.39 (m, 3 H), 7.49 (dd, J=8.8, 1.8
Hz, 1 H), 7.64 (d, J=1.8 Hz, 1 H), 7.69 (s, 1 H). LC-MS (ESþ): tR
0.96 min. m/z: 461 (M þ H). Anal. (C23H25N2O4SCl) H, N, S, Cl;
C: calcd 59.93, found 60.41.
(Z,Z)-5-(3-Chloro-4-hydroxy-benzylidene)-2-isopropylimino-
3-phenyl-thiazolidin-4-one (8g). Method E. The title compound
(1.13 g, 71%) was isolated as a beige to yellow powder from the
reaction mixture starting from 3a (1.00 g, 4.27 mmol) and
3-chloro-4-hydroxy-benzaldehyde (668 mg, 4.27 mmol); mp
268 °C. 1H NMR (DMSO-d6): δ 1.11 (d, J=6.0 Hz, 6 H), 3.56
(hept, J=5.8 Hz, 1 H), 7.15 (d, J=8.5 Hz, 1 H), 7.37 (d, J=7.5
Hz, 2 H), 7.44 (d, J=7.3 Hz, 1 H), 7.51 (t, J=7.8 Hz, 3 H), 7.66 (s,
1 H), 7.69 (d, J=1.0 Hz, 1 H), 10.99 (s, 1 H). LC-MS (ESþ): tR
1.01 min. m/z: 373 (M þ H). Anal. (C19H17N2O2SCl): C, H, N,
O, S, Cl. A small sample was recrystallized from ethanol/DCM
to obtain crystals suitable for X-ray analysis, pale-yellow nee-
dles, mp 268 °C. X-ray crystallographic structure parameters:
Crystals of 8g (C19H17ClN2O2S, formula weight 372.87) formed
in the triclinic space group P1. A total of 29341 reflections was
measured at 173 K. Molecules/unit cell Z=2, cell dimensions a=
9.2574(2) A, b=9.8000(2) A, c=10.6365(2) A, R=99.1910(10)°,
β=99.6080(10)°, γ=108.6490(10)°; calculated density=1.41 g
cm-3. The final R factor of 0.031 was obtained for 4805
observed reflections (I > 3σ(I)); largest difference peak and
hole were 0.42 and -0.41 eA-3, respectively (CCDC771349).
(Z,Z)-5-(3-Chloro-4-hydroxy-benzylidene)-2-propylimino-3-
o-tolyl-thiazolidin-4-one (8bl). Method E. The title compound
(1.14 mg, 69%) was isolated from the reaction mixture as a pale-
yellow crystalline solid starting from 5b (1.06 g, 4.27 mmol) and
3-chloro-4-hydroxy-benzaldehyde (668 mg, 4.27 mmol); mp 200 °C.
1H NMR (DMSO-d6): δ 0.84 (t, J=7.3 Hz, 3 H), 1.46-1.57 (m,
2 H), 2.09 (s, 3 H), 3.24-3.37 (m, 2 H), 7.15 (d, J=8.5 Hz, 1 H),
7.25-7.28 (m, 1 H), 7.30-7.36 (m, 1 H), 7.36-7.40 (m, 2 H),
7.53 (dd, J=8.5, 2.0 Hz, 1 H), 7.67 (s, 1 H), 7.71 (d, J=2.3 Hz, 1
H), 11.01 (s, 1 H). LC-MS (ESþ): tR 1.03 min. m/z: 387 (M þ H).
Anal. (C20H19N2O2SCl): C, H, N, S, Cl.
(Z,Z)-5-[3-Chloro-4-((2R)-2,3-dihydroxy-propoxy)-benzylidene]-
2-propylimino-3-o-tolyl-thiazolidin-4-one (8bo). To a solution of
3-chloro-4-hydroxybenzaldehyde (4.21 g, 26.9 mmol) in degas-
sed toluene (100 mL), ((4S)-2,2-dimethyl-[1,3]dioxolan-4-yl)-
methanol (5.35 g, 40.5 mmol) and 1,10-(azodicarbonyl)dipiperi-
dide (13.63 g, 54 mmol) followed by tributylphosphine (10.93 g,
54 mmol) was added. The mixture became slightly warm and a
precipitate formed. The reaction mixture was diluted with
degassed toluene (500 mL) and was stirred at room temperature
for 2 h, then at 60 °C for further 18 h before it was washed with
1 N aq NaOH (3 ꢀ 150 mL) and water (150 mL). The organic
phase was collected, dried over MgSO4, filtered, and concen-
trated to leave a dark-brown oil which was purified by column
chromatography on silica gel eluting with hexane:ethyl acetate
4:1 to give 3-chloro-4-((4S)-2,2-dimethyl-[1,3]dioxolan-4-yl-
methoxy)-benzaldehyde (4.30 g, 59%) as a yellow oil. LC: tR
0.93 min. 1H NMR (CDCl3): δ 1.41 (s, 3H), 1.47 (s, 3H),
4.06-4.00 (m, 1H), 4.14-4.08 (m, 1H), 4.23-4.17 (m, 2H),
4.56-4.43 (m, 1H), 7.05 (d, J=8.2 Hz, 1H), 7.74 (dd, J=1.8, 8.2
Hz, 1H), 7.89 (d, J = 1.8 Hz, 1H), 9.82 (s, 1H). A solution
of 5b (1.98 g, 8.00 mmol), the above 3-chloro-4-((4S)-2,2-
dimethyl-[1,3]dioxolan-4-ylmethoxy)-benzaldehyde (4.32 g, 16.0
mmol) and anhydrous sodium acetate (1.31 g, 16.0 mmol) in
acetic acid (50 mL) was stirred at 110 °C for 4 h. Water (0.3 mL)
was added, and stirring was continued at 100 °C for 1 h before
the mixture was concentrated. The remaining residue was dis-
solved in ethyl acetate (250 mL), washed with satd aq NaHCO3
solution (250 mL) and water (2 ꢀ 150 mL), dried over MgSO4,
filtered, and again concentrated. Sodium methoxide (500 mg,
9.25 mmol) was added to a solution of the residue in methanol
(100 mL). The mixture was stirred at 40 °C for 30 min before it
was concentrated. The remaining residue was dissolved in ethyl
acetate (250 mL) and washed twice with water (2 ꢀ 150 mL). The
washings were extracted back with ethyl acetate (150 mL), and
the combined organic extracts were dried over MgSO4, filtered,
and concentrated. The crude product was purified by column
chromatography on silica gel eluting with heptane:ethyl acetate
1:4 to give the title compound (1.34 g, 37%) as a pale-yellow
foam. 1H NMR (CDCl3): δ 0.94 (t, J=7.3 Hz, 3 H), 1.58-1.70
(m, 2 H), 2.21 (s, 3 H), 3.32-3.48 (m, 2 H), 3.82-3.95 (m, 3 H),
(Z,Z)-5-[3-Chloro-4-(2-hydroxy-ethoxy)-benzylidene]-2-pro-
pylimino-3-o-tolyl-thiazolidin-4-one (8bm). A mixture of 3-chloro-
4-hydroxybenzaldehyde (10 g, 63.9 mmol), K2CO3 (26.5 g, 191.6
mmol), and 2-bromoethyl acetate (26.7 g, 159.7 mmol) in ace-
tone (250 mL) was refluxed for 18 h before it was diluted with
diethyl ether (200 mL) and washed with water (3 ꢀ 200 mL). The
washings were extracted with diethyl ether (200 mL). The
combined organic extracts were dried over MgSO4 and concen-
trated. The remaining residue was purified by column chroma-

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10 4209
4.12-4.27 (m, 4 H), 7.07 (d, J=8.8 Hz, 1 H), 7.21 (d, J=7.0 Hz,
1 H), 7.31-7.39 (m, 3 H), 7.49 (dd, J=8.5, 2.0 Hz, 1 H), 7.64 (d,
J=2.0 Hz, 1 H), 7.69 (s, 1 H). 13C NMR (CDCl3): δ 11.83, 17.68,
23.74, 55.42, 63.46, 69.85, 70.78, 133.48, 120.75, 123.71, 127.05,
128.25, 128.60, 129.43, 130.06, 131.13, 131.50, 134.42, 136.19,
146.98, 154.75, 166.12. LC-MS (ESþ): tR 0.96 min. m/z: 461 (M
þ H). HPLC (ChiralPak AD-H, 4.6 mm ꢀ 250 mm, 0.8 mL/min,
70% hexane in ethanol): tR 11.8 min. Anal. (C23H25N2O4SCl):
C, H, N, O, S, Cl.
(7) Maceyka, M.; Payne, S. G.; Milstien, S.; Spiegel, S. Sphingosine
kinase, sphingosine-1-phosphate, and apoptosis. Biochim. Biophys.
Acta 2002, 1585, 193–201.
(8) Spiegel, S.; Milstien, S. Sphingosine-1-phosphate: An Enig-
matic Signalling Lipid. Nature Rev. Mol. Cell Biol. 2003, 4, 397–
407.
(9) Hait, N. C.; Oskeritzian, C. A.; Paugh, S. W.; Milstien, S.; Spiegel,
S. Sphingosine kinases, sphingosine 1-phosphate, apoptosis and
diseases. Biochim. Biophys. Acta 2006, 1758, 2016–2026.
(10) Kihara, A.; Mitsutake, S.; Mizutani, Y.; Igarashi, Y. Metabolism
and biological functions of two phosphorylated sphingolipids,
sphingosine 1-phosphate and ceramide 1-phosphate. Prog. Lipid
Res. 2007, 46, 126–144.
(Z,Z)-5-[3-Chloro-4-((2S)-2,3-dihydroxy-propoxy)-benzylidene]-
2-propylimino-3-o-tolyl-thiazolidin-4-one (8bp). Starting from
3-chloro-4-hydroxybenzaldehyde (2.00 g, 12.8 mmol) and ((4R)-
2,2-dimethyl-[1,3]dioxolan-4-yl)-methanol (2.53 g, 19.2 mmol),
3-chloro-4-((4R)-2,2-dimethyl-[1,3]dioxolan-4-ylmethoxy)-ben-
zaldehyde (1.95 g, 56%) was obtained as a brownish oil following
the procedure given above for its (S)-isomer. LC: tR 0.93 min.
1H NMR (CDCl3): δ 1.43 (s, 3 H), 1.49 (s, 3 H), 4.06 (dd, J=
8.5, 5.8 Hz, 1 H), 4.11-4.18 (m, 1 H), 4.20-4.26 (m, 2 H),
4.52-4.59 (m, 1 H), 7.09 (d, J=8.5 Hz, 1 H), 7.78 (dd, J=8.5, 2.0
Hz, 1 H), 7.93 (d, J=2.0 Hz, 1 H), 9.88 (s, 1 H). Starting from 5b
(248 mg, 1.00 mmol) and the above 3-chloro-4-((4R)-2,2-di-
methyl-[1,3]dioxolan-4-ylmethoxy)-benzaldehyde (405 mg, 1.50
mmol), the title compound (210 mg, 46%) was obtained as a
pale-yellow solid following the procedure given for compound
8bo. 1H NMR (CDCl3): δ 0.94 (t, J=7.4 Hz, 3 H), 1.59-1.72 (m,
2 H), 2.21 (s, 3 H), 2.31 (s br, 1 H), 2.90 (s br, 1 H), 3.34-3.48 (m,
2 H), 3.80-3.95 (m, 2 H), 4.15-4.25 (m, 3 H), 7.06 (d, J=8.6 Hz,
1 H), 7.21 (d, J=7.2 Hz, 1 H), 7.31-7.40 (m, 3 H), 7.48 (dd, J=
8.6, 2.1 Hz, 1 H), 7.63 (d, J=2.1 Hz, 1 H), 7.67 (s, 1 H). LC-MS
(ESþ): tR 0.96 min, m/z: 461 (M þ H). HPLC (ChiralPak AD-H,
4.6 mm ꢀ 250 mm, 0.8 mL/min, 70% hexane in ethanol): tR 13.6
min. Anal. (C23H25N2O4SCl, 0.1H2O): C, H, N, O, S, Cl, ash
<0.2%.
ꢀ
ꢁ
(11) Auge, N.; Negre-Salvayre, A.; Salvayre, R.; Levade, T. Sphingo-
myelin metabolites in vascular cell signaling and atherogenesis.
Prog. Lipid Res. 2000, 39, 207–229.
(12) Alewijnse, A. E.; Peters, S. L. M.; Michel, M. C. Cardiovascular
effects of sphingosine-1-phosphate and other sphingomyelin meta-
bolites. Br. J. Pharmacol. 2004, 143, 666–684.
(13) McVerry, B. J.; Garcia, J. G. N. In vitro and in vivo modulation of
vascular barrier integrity by sphingosine 1-phosphate: mechanistic
insights. Cell. Signalling 2005, 17, 131–139.
(14) Yatomi, Y. Sphingosine 1-Phosphate in Vascular Biology: Possible
Therapeutic Strategies to Control Vascular Disease. Curr. Pharm.
Des. 2006, 12, 575–587.
(15) Rosen, H.; Sanna, M. G.; Cahalan, S. M.; Gonzalez-Cabrera, P. J.
Tipping the gatekeeper: S1P regulation of endothelial barrier
function. Trends Immunol. 2007, 28, 102–107.
(16) Michel, M. C.; Mulders, A. C. M.; Jongsma, M.; Alewijnse, A. E.;
Peters, S. L. M. Vascular effects of sphingolipids. Acta Paed. 2007,
96, 44–48.
(17) Alewijnse, A. E.; Peters, S. L. M. Sphingolipid signalling in the
cardiovascular system: good, bad or both? Eur. J. Pharmacol. 2008,
585, 292–302.
(18) Brinkmann, V.; Baumruker, T. Pulmonary and vascular pharma-
cology of sphingosine 1-phosphate. Curr. Opin. Pharmacol. 2006, 6,
244–250.
(19) Uhlig, S.; Gulbins, E. Sphingolipids in the lungs. Am. J. Resp. Crit.
Care Med. 2008, 178, 1100–1114.
(20) Oskeritzian, C. A.; Milstien, S.; Spiegel, S. Sphingosine-1-phos-
phate in allergic responses, asthma and anaphylaxis. Pharmacol.
Ther. 2007, 115, 390–399.
(21) Rosen, H.; Goetzl, E. J. Sphingosine 1-phosphate and its receptors:
an autocrine and paracrine network. Nature Rev. Immunol. 2005, 5,
560–570.
(22) Rivera, J.; Proia, R. L.; Olivera, A. The alliance of sphingosine-
1-phosphate and its receptors in immunity. Nature Rev. Immunol.
2008, 8, 753–763.
(23) Okada, T.; Kajimoto, T.; Jahangeer, S.; Nakamura, S. Sphingosine
kinase/sphingosine 1-phosphate signalling in central nervous sys-
tem. Cell. Signalling 2009, 21, 7–13.
(24) Milstien, S.; Gude, D.; Spiegel, S. Sphingosine 1-phosphate in
neural signalling and function. Acta Paed. 2007, 96, 40–43.
(25) Ishii, M.; Egen, J. G.; Klauschen, F.; Meier-Schellersheim, M.;
Saeki, Y.; Vacher, J.; Proia, R. L.; Germain, R. N. Sphingosine-
1-phosphate mobilizes osteoclast precursors and regulates bone
homeostasis. Nature 2009, 458, 524–528.
(26) Liu, F.; Verin, A. D.; Wang, P.; Day, R.; Wersto, R. P.;
Chrest, F. J.; English, D. K.; Garcia, J. G. N. Differential
Regulation of Sphingosine-1-Phosphate- and VEGF-Induced
Endothelial Cell Chemotaxis. Am. J. Respir. Cell Mol. Biol. 2001,
24, 711–719.
Acknowledgment. We gratefully acknowledge the excellent
ꢀ
ꢀ
work done by our co-workers Celine Bortolamiol, Stephane
Delahaye, Judith Frei, HakimHadana, Julie Hoerner, Katalin
Menyhart, Christine Metzger, Rodolphe Mielke, David
€
Monnard, Markus Rey, Christine Seeger, Jurgen Seifert,
Marco Tschanz, Gaby von Aesch, Daniel Wanner, and Aude
Weigel, and Martine Clozel and Walter Fischli for support.
Note Added in Proof. Reference 51 has been retracted by
Premenko-Lanier et al., Nature 2010, 464, 942 (8 April 2010).
Supporting Information Available: Experimental details on
the synthesis and characterization, including biological assays,
of the 2-imino-thiazolidin-4-one scaffolds 3, 4, 5, 6, additional
2-imino-thiazolidin-4-one scaffolds, and target compounds 8
prepared for this study. This material is available free of charge
via the Internet at http://pubs.acs.org.
(27) Graeler, M.; Goetzl, E. J. Activation-regulated expression and
chemotactic function of sphingosine 1-phosphate receptors in
mouse splenic T cells. FASEB J. 2002, 16, 1874–1878.
(28) Adachi, K.; Kohara, T.; Nakao, N.; Arita, M.; Chiba, K.; Mishina,
T.; Sasaki, S.; Fujita, T. Design, synthesis, and structure-activity
relationships of 2-substituted-2-amino-1,3-propanediols: disco-
very of a novel immunosuppressant, FTY720. Bioorg. Med. Chem.
Lett. 1995, 5, 853–856.
(29) Chiba, K.; Yanagawa, Y.; Masubuchi, Y.; Kataoka, H.;
Kawaguchi, T.; Ohtsuki, M.; Hoshino, Y. FTY720, a Novel
Immunosuppressant, Induces Sequestration of Circulating Mature
Lymphocytes by Acceleration of Lymphocyte Homing in Rats. I.
FTY720 Selectively Decreases the Number of Circulating Mature
Lymphocytes by Acceleration of Lymphocyte Homing. J. Immu-
nol. 1998, 160, 5037–5044.
References
(1) Chun, J.; Goetzl, E. J.; Hla, T.; Igarashi, Y.; Lynch, K. R.;
Moolenaar, W.; Pyne, S.; Tigyi, G. International Union of
Pharmacology. XXXIV. Lysophospholipid Receptor Nomencla-
ture. Pharmacol. Rev. 2002, 54, 265–269.
(2) Ishii, I.; Fukushima, N.; Ye, X.; Chun, J. Lysophospholipid
Receptors: Signaling and Biology. Annu. Rev. Biochem. 2004, 73,
321–354.
(3) Brinkmann, V. Sphingosine 1-phosphate receptors in health and
disease: Mechanistic insights from gene deletion studies and reverse
pharmacology. Pharmacol. Ther. 2007, 115, 84–105.
(4) Meyer-zu-Heringdorf, D.; Jakobs, K. H. Lysophospholipid recep-
tors: Signalling, pharmacology and regulation by lysophospholipid
metabolism. Biochim. Biophys. Acta 2006, 1768, 923–940.
(5) Gardell, S. E.; Dubi, A. E.; Chun, J. Emerging medicinal roles for
lysophospholipid signaling. Trends Mol. Med. 2006, 12, 65–75.
(6) Rosen, H.; Gonzalez-Cabrera, P. J.; Sanna, M. G.; Brown, S.
Sphingosine 1-Phosphate Receptor Signaling. Annu. Rev. Biochem.
2009, 78, 743–768.
(30) Pinschewer, D. D.; Ochsenbein, A. F.; Odermatt, B.; Brinkmann,
V.; Hengartner, H.; Zinkernagel, R. M. FTY720 Immunosupres-
sion Impairs Effector T Cell Peripheral Homing Without Affecting
Induction, Expansion, and Memory. J. Immunol. 2000, 164, 5761–
5770.

4210 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10
Bolli et al.
(31) Brinkmann, V.; Chen, S.; Feng, L.; Pinschewer, D.; Nikolova, Z.;
Hof, R. FTY720 Alters Lymphocyte Homing and Protects Allo-
grafts Without Inducing General Immunosuppression. Transplant.
Proc. 2001, 33, 530–531.
(50) Graeler, M. H.; Huang, M.-C.; Watson, S.; Goetzl, E. J. Immuno-
logical Effects of Transgenic Constitutive Expression of the Type 1
Sphingosine 1-Phosphate Receptor by Mouse Lymphocytes.
J. Immunol. 2005, 174, 1997–2003.
(51) Premenko-Lanier, M.; Moseley, N. B.; Pruett, S. T.; Romagnoli,
P. A.; Altman, J. D. TransientFTY720treatmentpromotesimmune-
mediated clearance of a chronic viral infection. Nature 2008, 454,
894–898.
(52) Baumruker, T.; Billich, A.; Brinkmann, V. FTY720, an immuno-
modulatory sphingolipid mimetic: translation of a novel mechan-
ism into clinical benefit in multiple sclerosis. Expert Opin. Invest.
Drugs 2007, 16, 283–289.
(53) Mansoor, M.; Melendez, A. J. Recent Trials for FTY720
(Fingolimod): A New Generation of Immunomodulators Structu-
rally Similar to Sphingosine. Rev. Recent Clin. Trials 2008, 3, 62–
69.
(54) O’Connor, P.; Comi, G.; Montalban, X.; Antel, J.; Radue, E. W.;
deVera, A.; Pohlmann, H.; Kappos, L. Oral fingolimod (FTY720)
in multiple sclerosis. Neurology 2009, 72, 73–79.
(32) Brinkmann, V.; Davis, M. D.; Heise, C. E.; Albert, R.; Cottens, S.;
Hof, R.; Bruns, C.; Prieschl, E.; Baumruker, T.; Hiestand, P.;
Foster, C. A.; Zollinger, M.; Lynch, K. R. The Immune Modulator
FTY720 Targets Sphingosine 1-Phosphate Receptors. J. Biol.
Chem. 2002, 277, 21453–21457.
ꢀ
(33) Billich, A.; Bornancin, F.; Devay, P.; Mechtcheriakova, D.; Urtz,
N.; Baumruker, T. Phosphorylation fo the Immunomodulatory
Drug FTY720 by Sphingosine Kinases. J. Biol. Chem. 2003, 278,
47408–47415.
(34) Mandala, S.; Hajdu, R.; Bergstrom, J.; Quackenbush, E.; Xie, J.;
Milligan, J.; Thornton, R.; Shei, G.-J.; Card, D.; Keohane, C. A.;
Rosenbach, M.; Hale, J.; Lynch, C. L.; Rupprecht, K.; Parsons,
W.; Rosen, H. Alteration of Lymphocyte Trafficking by Sphingo-
sine-1-Phosphate Receptor Agonists. Science 2002, 296, 346–349.
€
(35) Albert, R.; Hinterding, K.; Brinkmann, V.; Guerini, D.; Muller-
Hartwieg, C.; Knecht, H.; Simeon, C.; Streiff, M.; Wagner, T.;
(55) Foster, C. A.; Howard, L. M.; Schweitzer, A.; Persohn, E.;
Hiestand, P. C.; Balatoni, B.; Reuschel, R.; Beerli, C.; Schwartz,
M.; Billich, A. Brain Penetration of the Oral Immunomodulatory
Drug FTY720 and Its Phosphorylation in the Central Nervous
System during Experimental Autoimmune Encephalomyelitis:
Consequences for Mode of Action in Multiple Sclerosis. J. Phar-
macol. Exp. Ther. 2007, 323, 469–476.
ꢀ
Welzenbach, K.; Zecri, F.; Zollinger, M.; Cooke, N.; Francotte, E.
Novel Immunomodulator FTY720 Is Phosphorylated in Rats and
Humans To Form a Single Stereoisomer. Identification, Chemical
Proof, and Biological Characterization of the Biologically Active
Species and Its Enantiomer. J. Med. Chem. 2005, 48, 5373–5377.
(36) Allende, M. L.; Dreier, J. L.; Mandala, S.; Proia, R. L. Expression
of the Sphingosine 1-Phosphate Receptor, S1P1, on T-cells Con-
trols Thymic Emigration. J. Biol. Chem. 2004, 279, 15396–15401.
(37) Matloubian, M.; Lo, C. G.; Cinamon, G.; Lesneski, M. J.; Xu, Y.;
Brinkmann, V.; Allende, M. L.; Proia, R. L.; Cyster, J. G.
Lymphocyte egress from thymus and peripheral lymphoid organs
is dependent on S1P receptor 1. Nature 2004, 427, 355–360.
(38) Hale, J. J.; Neway, W.; Mills, S. G.; Hajdu, R.; Keohane, C. A.;
Rosenbach, M.; Milligan, J.; Shei, G.-J.; Chrebet, G.; Bergstrom,
J.; Card, D.; Koo, G. C.; Koprak, S. L.; Jackson, J. J.; Rosen, H.;
Mandala, S. Potent S1P receptor agonists replicate the pharmaco-
logic actions of the novel immune modulator FTY720. Bioorg.
Med. Chem. Lett. 2004, 14, 3351–3355.
€
(56) Balatoni, B.; Storch, M. K.; Swoboda, E.-M.; Schonborn, V.;
Koziel, A.; Lambrou, G. N.; Hiestand, P. C.; Weissert, R.; Foster,
C. A. FTY720 sustains and restores neuronal function in the DA
rat model of MOG-induced experimental autoimmune encephalo-
myelitis. Brain Res. Bull. 2007, 74, 307–316.
(57) Miron, V. E.; Hall, J. A.; Kennedy, T. E.; Soliven, B.; Antel, J. P.
Cyclical and Dose-Dependent Responses of Adult Human Mature
Oligodendrocytes to Fingolimod. Am. J. Pathol. 2008, 173, 1143–
1152.
(58) Miron, V. E.; Schubart, A.; Antel, J. P. Central nervous system-
directed effects of FTY720 (fingolimod). J. Neurol. Sci. 2008, 274,
13–17.
(39) Jo, E.; Sanna, M. G.; Gonzalez-Cabrera, P. J.; Thangada, S.; Tigyi,
G.; Osborne, D. A.; Hla, T.; Parrill, A. L.; Rosen, H. S1P1-
Selective In Vivo-Active Agonists from High-Throughput Screen-
ing: Off-the-Shelf Chemical Probes of Receptor Interactions, Sig-
naling, and Fate. Chem. Biol. 2005, 12, 703–715.
(40) Wei, S. H.; Rosen, H.; Matheu, M. P.; Sanna, M. G.; Wang, S.-K.;
Jo, E.; Wong, C.-H.; Parker, I.; Cahalan, M. D. Sphingosine
1-phosphate type 1 receptor agonism inhibits transendothelial
migration of medullary T cells to lymphatic sinuses. Nature
Immunol. 2005, 6, 1228–1235.
(41) Rosen, H.; Alfonso, C.; Surh, C. D.; McHeyzer-Williams, M. G.
Rapid induction of medullary thymocyte phenotypic maturation
and egress inhibition by nanomolar sphingosine 1-phosphate
receptor agonist. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 10907–
10912.
(42) Hla, T.; Venkataraman, K.; Michaud, J. The vascular S1P gradient;
cellular sources and biological significance. Biochim. Biophys. Acta
2008, 1781, 477–482.
(43) Sinha, R. K.; Park, C.; Hwang, I.-Y.; Davis, M. D.; Kehrl, J. H. B
Lymphocytes Exit Lymph Nodes through Cortical Lymphatic
Sinusoids by a Mechanism Independent of Sphingosine-1-Phos-
phate-Mediated Chemotaxis. Immunity 2009, 30, 434–446.
(44) Dorsam, G.; Graeler, M. H.; Seroogy, C.; Kong, Y.; Voice, J. K.;
Goetzl, E. J. Transduction of Multiple Effects of Sphingosine
1-Phosphate (S1P) on T Cell Functions by the S1P1 G Protein-
Coupled Receptor. J. Immunol. 2003, 171, 3500–3507.
(45) Schwab, S. R.; Cyster, J. G. Finding a way out: lymphocyte egress
from lymphoid organs. Nature Immunol. 2007, 8, 1295–1301.
(46) Pappu, R.; Schwab, S. R.; Cornelissen, I.; Pereira, J. P.; Regard,
J. B.; Xu, Y.; Camerer, E.; Zheng, Y.-W.; Huang, Y.; Cyster, J. G.;
Coughlin, S. R. Promotion of Lymphocyte Egress into Blood and
Lymph by Distinct Sources of Sphingosine-1-Phosphate. Science
2007, 316, 295–298.
(47) Sanna, M. G.; Wang, S.-K.; Gonzalez-Cabrera, P. J.; Don, A.;
Marsolais, D.; Matheu, M. P.; Wei, S. H.; Parker, I.; Jo, E.; Cheng,
W.-C.; Cahalan, M. D.; Wong, C.-H.; Rosen, H. Enhancement of
capillary leakage and restoration of lymphocyte egress by a chiral
S1P1 antagonist in vivo. Nature Chem. Biol. 2006, 8, 434–441.
(48) Marsolais, D.; Rosen, H. Chemical modulators of sphingosine-1-
phosphate receptors as barrier-oriented therapeutic molecules.
Nature Rev. Drug Discovery 2009, 8, 297–307.
(59) Dev, K. K.; Mullershausen, F.; Mattes, H.; Kuhn, R. R.; Bilbe, G.;
Hoyer, D.; Mir, A. Brain sphingosine-1-phosphate receptors:
implication for FTY720 in the treatment of multiple sclerosis.
Pharmacol. Ther. 2008, 117, 77–93.
(60) Choi, J. W.; Herr, D.; Lee, C.-W.; Teo, S.; Kennedy, G.; Chun, J.
S1P1 receptor signaling on cells of astrocytic lineages in experi-
mental autoimmune encephalomyelitis: a role in disease progres-
sion and the efficacy of fingolimod (FTY720). Mult. Scler. 2008, 14
(S37), P29.
(61) Hale, J. J.; Doherty, G.; Toth, L.; Mills, S. G.; Hajdu, R.; Keohane,
C. A.; Rosenbach, M.; Milligan, J.; Shei, G.-J.; Chrebet, G.;
Bergstrom, J.; Card, D.; Forrest, M.; Sun, S.-Y.; West, S.; Xie,
H.; Nomura, N.; Rosen, H.; Mandala, S. Selecting against S1P3
enhances the acute cardiovascular tolerability of 3-(N-ben-
zyl)aminopropylphosphonic acid S1P receptor agonists. Bioorg.
Med. Chem. Lett. 2004, 14, 3501–3505.
(62) Forrest, M.; Sun, S.-Y.; Hajdu, R.; Bergstrom, J.; Card, D.;
Doherty, G.; Hale, J.; Keohane, C.; Meyers, C.; Milligan, J.; Mills,
S.; Nomura, N.; Rosen, H.; Rosenbach, M.; Shei, G.-J.; Singer,
I. I.; Tian, M.; West, S.; White, V.; Xie, J.; Proia, R. L.; Mandala, S.
Immune Cell Regulation and Cardiovascular Effects of Sphingo-
sine 1-Phosphate Receptor Agonists in Rodents ARe Mediated via
Distinct Receptor Subtypes. J. Pharmacol. Exp. Ther. 2004, 309,
758–768.
(63) Sanna, M. G.; Liao, J.; Jo, E.; Alfonso, C.; Ahn, M.-Y.; Peterson,
M. S.; Webb, B.; Lefebvre, S.; Chun, J.; Gray, N.; Rosen, H.
Sphingosine 1-Phosphate (S1P) Receptor Subtypes S1P1 and S1P3,
Respectively, Regulate Lymphocyte Recirculation and Heart Rate.
J. Biol. Chem. 2004, 279, 13839–13848.
(64) Salomone, S.; Potts, E. M.; Tyndall, S.; Ip, P. C.; Chun, J.;
Brinkmann, V.; Waeber, C. Analysis of shingosine 1-phosphate
receptors involved in constriction of isolated cerebral arteries with
receptor null mice and pharmacological tools. Br. J. Pharmacol.
2008, 153, 140–147.
(65) Gon, Y.; Wood, M. R.; Kiosses, W. B.; Jo, E.; Sanna, M. G.;
Chun, J.; Rosen, H. S1P3 receptor-induced reorganization of
epithelial tight junctions compromises lung barrier integrity and
is potentiated by TNF. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
9270–9275.
(66) Buzard, D. J.; Thatte, J.; Lerner, M.; Edwards, J.; Jones, R. M.
Recent progress in the development of selective S1P1 receptor
agonists for the treatment of inflammatory and autoimmune
disorders. Expert Opin. Ther. Pat. 2008, 18, 1141–1159.
(67) Brown, F. C. 4-Thiazolidinones. Chem. Rev. 1961, 61, 463–521.
(49) Brinkmann, V.; Pinschewer, D. D.; Feng, L.; Chen, S. FTY720:
Altered lymphocyte traffic results in allograft protection. Trans-
plantation 2001, 72, 764–769.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 10 4211
(68) Singh, S. P.; Parmar, S. S.; Raman, K.; Stenberg, V. I. Chemistry
and Biological Activity of Thiazolidinones. Chem. Rev. 1981, 81,
175–203.
1,3-thiazolidin-4-ones. Rapid Commun. Mass Spectrom. 2004, 18,
87–95.
(83) Geci, I.; Valtamob, P.; Imrich., J.; Kivela, H.; Kristian, P.; Pihlaja,
K. Regiospecific Synthesis, Structure and Electron Ionization
Mass Spectra of 1,3-Thiazolidine-4-ones Containing the Acridine
Skeleton. J. Heterocycl. Chem. 2005, 42, 907–918.
(84) Sahu, M.; Garnaik, B. K.; Behera, R. K. Influence of Substituents
on the Synthesis of Thiazolidinones. Indian J. Chem., Sect. B: Org.
Chem. Incl. Med. Chem. 1987, 26, 779–781.
(85) Laurent, D. R. S.; Gao, Q.; Wu, D.; Serrano-Wu, M. H. Regiose-
lective synthesis of 3-(heteroaryl)-iminothiazolidin-4-ones. Tetra-
hedron Lett. 2004, 45, 1907–1910.
ꢀ
€
(69) Bhargave, P. N.; Prakash, S.; Lakhan, R. Synthesis of 2-(40,
50-Disubstituted-thiazol-20-ylimino)-3-(m-methylphenyl)-5-methyl
(or H) -4-thiazolidinones and Their Fungicidal Activity. Indian
J. Chem., Sect. B: Org. Chem. Incl. Med. Chem. 1981, 20, 927–
929.
(70) Gruner, M.; Rehwald, M.; Eckert, K.; Gewald, K. New Synthesis
of 2-Alkylthio-4-oxo-3,4-dihydro-quinazolines, 2-Alkylthioquina-
zolines, as well as Their Hetero Analogues. Heterocycles 2000, 53,
2363–2377.
(71) Sedlak, M.; Hejtmankova, L.; Hanusek, J.; Machacek, V. Synthe-
sis and ring transformation of substituted S-(1-phenylpyrrolidine-
2-one-3-yl)isothiuronium salts to substituted 2-imino-5-[2-(phenyl-
amine)ethyl]thiazolidin-4-ones. J. Heterocycl. Chem. 2002, 39,
1105–1107.
(72) Blanchet, J.; Zhu, J. Reeve’s synthesis of 2-imino-4-thiazolidinone
from alkyl (aryl) trichloromethylcarbinol revisited, a three-compo-
nent process from aldehyde, chloroform and thiourea. Tetrahedron
Lett. 2004, 45, 4449–4452.
(86) Johnson, T. B. On the Molecular Rearrangement of Thiocyanace-
tanilides into Labile Pseudothiohydantoins and on the Molecular
Rearrangement of the Later into Stable Isomers. J. Am. Chem. Soc.
1903, 25, 483–491.
(87) Bolli, M.; Scherz, M.; Mu€ller, C.; Mathys, M.; Lehmann, D.
5-(Benz-(Z)-ylidene)-thiazolidin-4-one derivatives as immunosup-
pressant agents. Patent WO2005054215, 2005.
(88) Moghaddam, F. M.; Hojabri, L. A Novel Synthesis of Some
2-Imino-4-thiazolidinone Derivatives. J. Heterocycl. Chem. 2007,
44, 35–38.
(89) Erol, S.; Dogan, I. Axially Chiral 2-Arylimino-3-aryl-thiazolidine-
4-one Derivatives: Enantiomeric Separation and Determination of
Racemization Barriers by Chiral HPLC. J. Org. Chem. 2007, 72,
2494–2500.
ꢀ
ꢀ
(73) Gabillet, S.; Lecercle, D.; Loreau, O.; Carboni, M.; Dezard, S.;
Gomis, J.-M.; Taran, F. Phosphine-Catalysed Construction of
Sulfur Heterocycles. Org. Lett. 2007, 9, 3925–3927.
(74) Maly, R. Ueber Sulfhydantoın (Glycolylsulfoharnstoff). Justus
¨
Liebigs Ann. Chem. 1873, 168, 133–139.
(75) Volhard, J. Ueber Glycolylsulfoharnstoff. Justus Liebigs Ann.
Chem. 1873, 166, 383–384.
(90) Clayden, J.; Moran, W. J.; Edwards, P. J.; LaPlante, S. R. The
Challenge of Atropisomerism in Drug Discovery. Angew. Chem.,
Int. Ed. 2009, 48, 6398–6401.
(76) Meyer, P. J. Ueber substitutierte Sulfohydantoıne. Berichte Dtsch.
¨
Chem. Ges. 1877, 10, 1965–1967.
(77) Liebermann, C.; Lange, A. Ueber die Formeln der Sulfhydantoine.
Berichte Dtsch. Chem. Ges. 1879, 12, 1588–1595.
(78) Liebermann, C. Zur Constitution der Sulfhydantoıne und Sulfur-
¨
(91) Taylor, J. B.; Triggle, D. J. Comprehensive Medicinal Chemistry II.
Elsevier: New York, 2007; Vol. 5, pp 133-165.
(92) Jeong, E. J.; Liu, X.; Jia, X.; Chen, J.; Hu, M. Coupling of
Conjugating Enzymes and Efflux Transporters: Impact on Bio-
availability and Drug Interactions. Curr. Drug Metab. 2005, 6, 455–
468.
(93) Crosignani, S.; Bombrun, A.; Covini, D.; Maio, M.; Marin, D.;
Quattropani, A.; Swinnen, D.; Simpson, D.; Sauer, W.; Franc-on,
B.; Martin, T.; Cambet, Y.; Nichols, A.; Martinou, I.; Burgat-
Charvillon, F.; Rivron, D.; Donini, C.; Schott, O.; Eligert, V.;
Novo-Perez, L.; Vitte, P.-A.; Arrighi, J.-F. Discovery of a novel
series of potent S1P1 agonists. Bioorg. Med. Chem. Lett. 2010, 20,
1516–1519.
(94) Piali, L.; Hess, P.; Kohl, C.; Nayler, O.; Bolli, M.; Steiner, B. S.76.
Selective S1P1 Receptor Agonist ACT-128800 Rapidly and Re-
versibly Reduces Blood Lymphocyte Count. FOCIS 2009, San
Francisco, June 11-14, 2009; p S153.
ethane. Ann. Chem. 1881, 207, 121–167.
(79) Steel, P. J.; Guard, J. A. M. Heterocyclic tautomerism. VI. A
redetermination and reassignment of the structure of 2-aminothia-
zol-4(5H)-one. Acta Crystallogr. Sect. C: Cryst. Struct. Commun.
1994, 50, 1721–1723.
ꢁ
(80) Ottana, R.; Maccari, R.; Berreca, M. L.; Bruno, G.; Rotondo, A.;
Rossi, A.; Chiricosta, G.; Paola, R. D.; Sautebin, L.; Cuzzocrea, S.;
Vigorita, M. G. 5-Arylidene-2-imino-4-thiazolidinones: design and
synthesis of novel anti-inflammatory agents. Bioorg. Med. Chem.
2005, 13, 4243–4252.
ꢀ
(81) Klika, K. D.; Janovec, L.; Imrich, J.; Suchar, G.; Pavol, K.;
Sillanpaa, R.; Kalevi, P. Regioselective synthesis of 2-imino-1,3-
thiazolidin-4-ones by treatment of N-(anthracen-9-yl)-N0-ethyl-
thiourea with bromoacetic acid derivatives. Eur. J. Org. Chem.
2002, 1248–1255.
(95) Brossard, P.; Hofmann, S.; Cavallaro, M.; Halabi, A.; Dingemanse,
J. Entry-into-humans study with ACT-128800, a selective S1P1
receptor agonist: tolerability, safety, pharmacokinetics, and pharma-
codynamics. ASCPT 2009, Annual Meeting of the American Society
for Clinical Pharmacology & Therapeutics,, Washington, DC,
March 18-21, 2009; pp S63-S64, PII-87 .
ꢀ
(82) Klika, K. D.; Valtamo, P.; Janovec, L.; Suchar, G.; Kristian, P.;
€
€
Imrich, J.; Kivela, H.; Alfoldi, J.; Pihlaja, K. Regioselective
syntheses, structural characterization, and electron ionization mass
spectrometric behavior of regioisomeric 2,3-disubstituted 2-imino-