Novel S1P1 receptor agonists - Part 5: From amino-to alkoxy-pyridines

Article abstract of DOI:10.1016/j.ejmech.2016.03.020

In a previous communication we reported on the discovery of aminopyridine 1 as a potent, selective and orally active S1P₁ receptor agonist. More detailed studies revealed that this compound is phototoxic in vitro. As a result of efforts aiming at eliminating this undesired property, a series of alkoxy substituted pyridine derivatives was discovered. The photo irritancy factor (PIF) of these alkoxy pyridines was significantly lower than the one of aminopyridine 1 and most compounds were not phototoxic. Focused SAR studies showed, that 2-, 3-, and 4-pyridine derivatives delivered highly potent S1P₁ receptor agonists. While the 2-pyridines were clearly more selective against S1PR₃, the corresponding 3- or 4-pyridine analogues showed significantly longer oral half-lives and as a consequence longer pharmacological duration of action after oral administration. One of the best compounds, cyclopentoxy-pyridine 45b lacked phototoxicity, showed EC₅₀ values of 0.7 and 140 nM on S1PR₁ and S1PR₃, respectively, and maximally reduced the blood lymphocyte count for at least 24 h after oral administration of 10 mg/kg to Wistar rats.

Full text of DOI:10.1016/j.ejmech.2016.03.020

European Journal of Medicinal Chemistry 115 (2016) 326e341
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Novel S1P1 receptor agonists e Part 5: From amino-to alkoxy-
pyridines
Martin H. Bolli*, Cyrille Lescop, Magdalena Birker, Ruben de Kanter, Patrick Hess,
€
Christopher Kohl, Oliver Nayler, Markus Rey, Patrick Sieber, Jorg Velker, Thomas Weller,
Beat Steiner
Actelion Pharmaceuticals Ltd., Gewerbestrasse 16, CH-4123 Allschwil, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
In a previous communication we reported on the discovery of aminopyridine 1 as a potent, selective and
orally active S1P1 receptor agonist. More detailed studies revealed that this compound is phototoxic
in vitro. As a result of efforts aiming at eliminating this undesired property, a series of alkoxy substituted
pyridine derivatives was discovered. The photo irritancy factor (PIF) of these alkoxy pyridines was
significantly lower than the one of aminopyridine 1 and most compounds were not phototoxic. Focused
SAR studies showed, that 2-, 3-, and 4-pyridine derivatives delivered highly potent S1P1 receptor ago-
nists. While the 2-pyridines were clearly more selective against S1PR3, the corresponding 3- or 4-
pyridine analogues showed significantly longer oral half-lives and as a consequence longer pharmaco-
logical duration of action after oral administration. One of the best compounds, cyclopentoxy-pyridine
45b lacked phototoxicity, showed EC50 values of 0.7 and 140 nM on S1PR1 and S1PR3, respectively,
and maximally reduced the blood lymphocyte count for at least 24 h after oral administration of 10 mg/
kg to Wistar rats.
Received 21 August 2015
Received in revised form
7 March 2016
Accepted 9 March 2016
Available online 12 March 2016
Keywords:
S1P1 receptor agonist
Pyridine
Lymphocyte count
Immunomodulation
In vitro phototoxicity
Light extinction
© 2016 Elsevier Masson SAS. All rights reserved.
1. Introduction
1 mM) and in lymph but low in tissue [1,5,6]. As maintenance of this
concentration gradient is fundamental to the biological function of
S1P, its synthesis, degradation, transport and storage are tightly
regulated [7e11].
Sphingosine-1-phosphate (S1P) is a lysophospholipid affecting
fundamental cellular processes including cell survival, differentia-
tion, proliferation or migration [1e4]. In the cell, S1P is produced
from the lipid component ceramide. Ceramidase cleaves ceramides
to give sphingosine which is phosphorylated by one of two
sphingosine kinases to form S1P. Lipid phosphate phosphatase and
other phosphatases can hydrolyze S1P back to sphingosine and
thus inactivate it. S1P may also be cleaved irreversibly by S1P lyase
to hexadecanal and aminoethanol phosphate making it unavailable
for reactivation by sphingosine kinases. S1P is transported out of
the cell to convey its biological activity in an autocrine or paracrine
fashion. Extracellular S1P exerts its activity by signalling through
five G-protein coupled receptors (S1PR1e5). S1P concentrations
greatly vary depending on the body compartment. For instance,
while S1P is stored in erythrocytes, endothelial cells, and plasma
lipoproteins, and its concentrations are high in plasma (up to nearly
While the physiological effects of S1PeS1P receptor signalling
have been studied in many organs and tissues [1] including the
nervous system [12,13], the lung [14,15], the cardiovascular system
[7,16e18] bone tissue [19], and haematopoietic and immune cells
[20e23], the observation that lymphocytes exposed to a synthetic
S1P1 receptor agonist are sequestered to lymphoid organs, removed
from circulation and are thus unable to aberrantly cause tissue
damage [24e26] is of particular interest. This novel mechanism of
immunomodulation [26e28] holds great promise in the fields of
transplantation and autoimmune diseases [29]. Today, fingolimod
(Gilenya®), a prodrug of a non-selective S1P receptor agonist, is a
successfully used approved therapy to treat relapsing multiple
sclerosis [30e34]. In addition, clinical use of S1P1 receptor agonists
in other autoimmune diseases is being evaluated [35e40].
Detailed mechanistic studies suggest that synthetic S1P1 re-
ceptor agonists in fact act as functional antagonists [41e45]. Rapid
and sustained internalization of the S1P1 receptor upon stimulation
abolishes the lymphocyte's ability to sense and migrate along the
* Corresponding author.
E-mail address: martin.bolli@actelion.com (M.H. Bolli).
http://dx.doi.org/10.1016/j.ejmech.2016.03.020
0223-5234/© 2016 Elsevier Masson SAS. All rights reserved.

M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
327
S1P gradient it needs to follow in order to exit the lymph node. In
view of this mode of action, it was postulated that also S1P1 re-
ceptor antagonists are able to sequester lymphocytes to lymph
nodes. However, this hypothesis was proven only recently [46e50].
On the other hand, it was recently demonstrated that disturbing the
S1P gradient by blocking the activity of S1P lyase with a synthetic
inhibitor can lead to lymphocyte sequestration too [51]. How S1PR1
antagonists and S1P lyase inhibitors compare to S1PR1 agonists
with respect to clinical efficacy and safety is of great interest and
results of on-going and future studies are eagerly awaited [5,52,53].
Our interest to identify novel S1P receptor agonists is based on
the following observations: Experiments with S1PR1 receptor
knock-out mice [54,55] and S1PR1 selective agonists [44,56,57],
revealed that targeting S1PR1 is sufficient to cause lymphocyte
sequestration to lymphoid organs. Originally, S1P1 receptor
agonism-triggered lymphocyte sequestration [43,54,58] was pro-
posed as the driving mechanism for the efficacy of S1P receptor
agonists in experimental autoimmune encephalomyelitis (EAE)
models in rodents [59e61]. More recent studies suggested that
S1P1 receptor modulation on neurons and astrocytes also contrib-
utes to the efficacy in these multiple sclerosis models [31,62e64].
On the other hand, activation of S1P5 receptors on astrocytes may
enhance remyelination [65]. In contrast to these beneficial effects of
S1P1 and S1P5 receptor agonism, S1P3 receptor activation is asso-
ciated with heart rate reduction [58,66e68], vaso- and broncho-
constriction [66,69], blood pressure increase [70], and fibrosis [71]
in rodents and is thus deemed undesired. As a consequence, the
majority of the corresponding drug discovery programs [72e74],
including our own, focused on identifying potent S1P1 receptor
agonists with high selectivity against S1PR3 and variable degrees of
activity on S1PR5.
groups in the dialkyl pyridine series by an alkoxy group would
again produce compounds with no phototoxicity liability. A first
alkoxy pyridine, the 2-methoxy-isonicotinic acid derivative 7a,
showed a PIF value of 0.9 in the 3T3 NRU phototoxicity test,
demonstrating that it is not phototoxic in vitro. Interestingly, during
the synthesis of 7a we also noted that aminopyridines such as 1, 2,
and 3 were strongly fluorescent under UV light (254 nm and
366 nm) when spotted on thin layer chromatography plates while
the dialkyl pyridine 4, 5, and 6 as well as the methoxy pyridine 7a
were devoid of this property. On this basis, we set out to explore the
alkoxy pyridines in more detail.
2. Results and discussion
2.1. Chemistry
The target compounds 7aek, 24aeg, 35a,b, and 45aed were
prepared in analogy to a previously reported pathway [75]. Thus, as
summarized in Scheme 1 the isonicotinic acids 12aek, the picolinic
acids 23aeg and 34a,b, and the nicotinic acids 44aed were coupled
with N-hydroxy-benzamidine 8 or 9 and the resulting amidine
ester intermediates were cyclized to the desired oxadiazoles. In
case hydroxyamidine 8 was used, the desired polar side chain (R1)
was established in steps that followed the cyclization step (for
details see Supporting information).
Schemes 2e5 summarize the synthetic pathways that were
followed to obtain the various pyridine carboxylic acid derivatives
required for this study. In general, yields have not been optimized.
For details see Supporting information.
The synthesis of the 2,6-disubstituted isonicotinic acids 12aek
required to produce target compounds 7aek is illustrated in
In a previous communication [75] we reported on the discovery
of amino-pyridine 1 as a potent and selective S1P1 receptor agonist
with high efficacy in reducing blood lymphocyte counts in the rat.
In more detailed studies involving a standardized Neutral Red
Uptake Phototoxicity Test using Balb/c 3T3 mouse fibroblasts (3T3
NRU) [76,77], the photo irritancy factor (PIF) of aminopyridine 1
was found to be >200, clearly flagging this compound as in vitro
phototoxic [78]. Testing of additional examples of our S1PR1 ago-
nists incorporating a pyridine head suggested that the phototox-
icity is an inherent property of the aminopyridine series. For
instance, while aminopyridines such as 1, 2 and 3 were clearly
phototoxic in vitro, dialkyl pyridine analogues such as compounds
4, 5 [75], and 6 [79] were devoid of this undesired property (Fig. 1).
Our discovery efforts therefore aimed at identifying novel S1P1
receptor agonists that combined the high affinity and selectivity for
S1PR1 and the favourable in vivo efficacy of the aminopyridine se-
ries [75] with the lack of the phototoxicity liability observed for the
dialkyl pyridine derivatives.
Health authority guidelines e.g. Refs. [80,81], recommend
in vitro photosafety assessment for compounds with a molar
extinction (ε) of >1000 L molꢀ1 cmꢀ1 at wavelengths >290 nm. As
the phototoxic aminopyridines 1, 2, and 3 as well as the non-
phototoxic analogues 4, 5, and 6 fulfilled this criterion, measuring
the molar extinction at >290 nm offered no guidance for our me-
dicinal chemistry program. In a recent computational approach
described by Haranosono et al. [82], the energy difference between
the HOMO and LUMO (HOMOeLUMO gap, HLG) and the
Scheme
2.
Reacting
methyl
or
ethyl
6-chloro-2-
methoxyisonicotinate 10 with an alkylmagnesium halide under
Fürstner conditions [83,84] furnished the corresponding 6-alkyl-2-
methoxy-isonicotinates 11a, 11b, and 11d. Alternatively, 6-alkyl-2-
methoxy-isonicotinic acid esters 11c and 11e were prepared by
reacting methyl or ethyl 6-chloro-2-methoxyisonicotinate with the
appropriate cycloalkylzinc bromide under Negishi conditions
[85e87]. Methyl 2-methoxy-6-phenylisonicotinate 11f was ob-
tained by Pd(PPh3)4 catalyzed Suzuki coupling [87,88] of methyl 6-
chloro-2-methoxyisonicotinate with phenylboronic acid. The
desired isonicotinic acids 12aef were then obtained by cleavage of
the corresponding esters using concentrated aqueous HCl or
aqueous NaOH or LiOH.
On the other hand, reacting 2-chloro-6-methylisonicotinic acid
13 with isopropanol under basic conditions gave access to 2-
isopropoxy-6-methylisonicotinic acid 12g. As illustrated with iso-
nicotinic acids 12he12k, 2-alkoxy-6-alkylisonicotinic acids could
also be prepared by an alternative pathway giving access to iso-
nicotinic acids bearing alkyl substituents larger than a methyl
group. Thus, reacting 2,6-dichloroisonicotinic acid 14 with cyclo-
butanolate
cyclobutoxyisonicotinic
or
cyclopentanolate
acid
furnished
or
2-chloro-6-
2-chloro-6-
15
cyclopentoxyisonicotinic acid 16, respectively. Esterification using
TMSCl in isopropanol to give 17 and 18 made the material ready for
the following Negishi coupling. Using either dimethylzinc or
diethylzinc afforded the corresponding 6-methyl or 6-ethyl-2-
cycloalkoxyisonicotinates 19, 20 and 21, 22, respectively. Saponifi-
cation furnished the desired isonicotinic acids 12hek.
maximum-conjugated
p electron number (PENMC) were calcu-
lated to predict the compound's phototoxicity potential. However,
applying this in silico method to our pyridine derivatives, we were
not able to explain the clear difference between the measured PIF
values of the aminopyridine and the dialkyl pyridine series as both
series were clearly predicted to be phototoxic. Hence, while staying
pragmatic, we wondered whether replacing one of the two alkyl
The picolinic acids 23aeg required for the study of the corre-
sponding target compounds 24aeg were prepared in a similar
fashion as outlined in Scheme 3. Thus, Negishi coupling of methyl
6-chloro-4-methoxypicolinate 25 using pentan-3-yl zinc bromide
furnished 4-methoxy-6-(pentan-3-yl)picolinic acid as its 3-pentyl
ester (26) in moderate yield. As observed earlier [75], the

328
M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
Fig. 1. Structures of the in vitro phototoxic aminopyridines 1, 2, and 3, and the non-phototoxic alkylpyridines 4, 5, and 6, and alkoxy pyridine 7a.
aReagents and conditions: (a) TBTU, DIPEA, DCM, or THF/DMF, rt, 1 h; or HOBt,
EDC HCl, THF, rt, 18 h; (b) dioxane, 80-100 °C, 7-18 h, 19-80% (2 steps); (c) (R)-
glycidol, PPh3, DEAD, THF, rt, 18 h, 76-96%, or (R)-epichlorohydrin, isopropanol, 3 N
NaOH, rt, 24 h, 35-85%; (d) 7 N NH3 in methanol, 60 °C, 18 h, 39%-quant.(crude); (e)
glycolic acid, HOBt, DIPEA, EDC HCl, THF, rt, 2 h, 23-79%. For R' and R'' see
Tables 1 to 4.
Scheme 1. Assembling the central oxadiazole ring.a
Pd(dppf)Cl2 catalyzed Negishi cross-coupling employing pentan-3-
yl zinc bromide not only led to the formation of the picolinic acid 3-
pentyl ester but also to isomerization of the 3-pentyl group
attached to the pyridine. Several groups have reported on the un-
reaction the corresponding cyclopentyl picolinic acids 23c and 23e
were obtained in good yield. On the other hand, Pd(PPh3)4-cata-
lyzed cross-coupling of ethyl 6-chloro-4-methoxypicolinate 27
with
2,4,6-tris(2-methylprop-1-en-1-yl)-1,3,5,2,4,6-
desired
b
-hydride elimination of secondary alkylzinc reagents
trioxatriborinane pyridine complex [92] furnished 4-methoxy-6-
(2-methylprop-1-en-1-yl)picolinate 28 which was hydrogenated
(29) and hydrolyzed to the desired picolinic acid 23b. 6-(Cyclo-
pentyloxy)-4-methylpicolinic acid 23f was obtained in good yield
by thermal displacement of the chlorine in 6-chloro-4-
methylpicolinic acid 32 with cyclopentanolate. To facilitate purifi-
cation, the crude product was temporarily transformed into its
methyl ester. Treating the isomeric 4-chloro-6-methylpicolinic acid
33 with cyclopentanolate furnished 4-(cyclopentyloxy)-6-
methylpicolinic acid 23g in moderate yield.
leading to isomerized cross-coupling products and significant im-
provements have been achieved using further optimized catalysts,
e.g. Refs. [89e91]. However, as we were able to obtain sufficient
material of the desired pent-3-yl-picolinic acids 23a and 23d (from
31) for our studies despite the fact that the crude reaction products
contained substantial amounts of the corresponding pent-2-yl and
pent-1-yl picolinic acid derivatives, we did not embark on further
optimizations.
b-Hydride elimination/b-migratory insertion [89]
may also occur in the case of the cyclopentyl zinc reagent but as
it has no consequence on the regioselectivity of the cross-coupling
The preparation of the two 5-cyclopentyl-picolinic acids 34a

M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
329
aReagents and conditions: (a) alkylmagnesium bromide or chloride, Fe(acac)3, THF,
NMP, -75 °C 1 h, 0 °C-rt 1 h, 40-42% (R = isobutyl (11b)); (b) alkylzinc bromide or
dialkylzinc, Pd(dppf)Cl2, dioxane, 75-80 °C, 4-48 h, 50-90% (R = 3-pentyl (11a)), 47%
(R=cyclobutyl (11c)), 67% (R = cyclopentyl (11d), 44-69% (R = cyclohexyl (11e)); (c)
(R=Ph (11f)) PhB(OH)2, Pd(PPh3)4, DME, 2 M aq. K2CO3, 80 °C, 18 h, 38%; (d) 6-7
M aq. HCl, 60-95 °C, 3-20 h, 51% (12c), 87-91% (12b), 96% - quant. (12d); (e) 3 M
aq. NaOH, ethanol, 80°C, 2 h, 86% (12a); (f) 2 N aq. LiOH, THF, MeOH, or dioxane,
rt, 2-18 h, 78-86% (12e), quant. (12f), 84%-quant. (12h), 72-88% (12i), quant. (12j),
97% (12k); (g) isopropanol, KOtBu, 80 °C, 72 h, 86% (12g); (h) cyclobutanol or
cyclopentanol, NaH, dioxane, 80 °C, 24 h, 87%-quant. (crude) (15), quant.(crude)
(16); (i) isopropanol, TMSCl, 50 °C, 24 h, 50-88% (17), 53-70% (18); (j) Me2Zn (19,
20) or Et2Zn (21, 22), Pd(dppf)Cl2; dioxane, 75 °C, 16-24 h, 61% (19), 50% (20), 79%
(21), 95% (22).
Scheme 2. Preparation of 2,6-disubstituted isonicotinic acids.a
and 34b needed for the preparation of target compounds 35a and
35b, respectively, is summarized in Scheme 4. The synthesis of 5-
cyclopentyl-4-methoxypicolinic acid 34a started by O-alkylation
of 2,5-dibromopyridin-4-ol 36 using iodomethane in the presence
of K2CO3. Regioselective Suzuki reaction of the resulting 2,5-
dibromo-4-methoxy pyridine 37 using 2,4,6-trivinyl-1,3,5,2,4,6-
trioxatriborinane pyridine complex [92] delivered 5-bromo-4-
methoxy-2-vinylpyridine 38 in good yield. Oxidative cleavage of
the vinyl group using KMnO4 established the carboxylic acid
function which subsequently was converted into its methyl ester
(39). In a second Suzuki reaction employing cyclopenten-1-yl
boronic acid the carbon framework of the 5-substituent was
introduced (40). Hydrogenation of the cyclopentene substituent
(41) and ester cleavage afforded the desired picolinic acid 34a.
Several attempts to directly introduce the cyclopentane moiety via
a Negishi reaction failed and led to decomposition of the methyl 5-
bromo-4-methoxy picolinate. Commercial availability of 5-bromo-
6-methoxypicolinic acid allowed for rapid access to picolinic acid
34b. This time, Pd(dppf)Cl2 catalyzed Negishi coupling of the ethyl
5-bromo-6-methoxypicolinate 42 with cyclopentyl zinc bromide
afforded ethyl 5-cyclopentyl-6-methoxypicolinate 43 which upon
saponification furnished the desired picolinic acid 34b.
Finally, Scheme 5 illustrates the synthetic access to the 5,6-
disubstituted nicotinic acids 44aed used to prepare the target
compounds 45aed. Methyl 6-chloro-5-methoxynicotinate 46 was
efficiently
converted
into
methyl
6-cyclopentyl-5-
methoxynicotinate 47 using cyclopentylmagnesium bromide un-
der Fürstner conditions [83,84]. Subsequent ester hydrolysis

330
M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
aReagents and conditions: (a) pentan-3-ylzinc(II) bromide, Pd(dppf)Cl2, dioxane, 75
°C, 18 h, 15-20% (26), 21% (Me ester of 23d); (b) 2 M aq. LiOH, dioxane or
methanol, rt to 75°C, 5-15 h, 98% (23a), 79 % (23c), quant. (23d), 88% (23e); (c)
2,4,6-tris(2-methylprop-1-en-1-yl)-1,3,5,2,4,6-trioxatriborinane
pyridine
complex,
Pd(PPh3)4, DME, 100 °C, 18 h, 78% (28); (d) H2, Pd/C, THF, rt, 15 h; quant. (29); (e)
6 N aq. HCl, 65°C, 12 h, quant. (23b), 84%-quant. (23c); (f) cyclopentylzinc(II)
bromide, Pd(dppf)Cl2, dioxane, 75 °C, 18 h, 90% (Me ester of 23e), 50-80% (30); (g)
(1) cyclopentanol, NaH, dioxane, 80°C, 24 h; (2) MeOH, TMSCl, 50 °C, 24 h; (3) 2 M
aq. LiOH, MeOH, rt, 2 h, 84% (23g, 3 steps); (h) cyclopentanol, NaH, 80 °C, 24 h,
35% (23g).
Scheme 3. Preparation of 4,6-disubstituted picolinic acids.a
afforded the desired nicotinic acid 44a. The preparation of the
isomeric nicotinic acid 44c followed a similar pathway. Thus, ethyl
5-chloro-6-methoxynicotinate 51 was coupled with cyclopenten-
1-yl boronic acid to the corresponding 5-(cyclopent-1-en-1-yl)-6-
methoxynicotinate 52. Hydrogenation of the cyclopentene (53)
followed by saponification of the nicotinic acid ester gave the
desired 5-cyclopentyl-6-methoxynicotinic acid 44c.
In the case of the 6-cyclopentoxynicotinic acid 44b, the cyclo-
pentoxy group was introduced by reacting 5-bromo-6-hydroxy
nicotinate 48 with cyclopentanol under Mitsunobu conditions.
Pd(dppf)Cl2 catalyzed Negishi coupling of the resulting 5-bromo-6-
cyclopentoxynicotinate 49 with dimethylzinc and subsequent

M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
331
aReagents and conditions: (a) MeI, K2CO3, DMF, 40 °C, 4 h, 35%; (b) 2,4,6-trivinyl-
1,3,5,2,4,6-trioxatriborinane pyridine, Pd(PPh3)4, PPh3, 2 M aq. K2CO3, dioxane,
90 °C, 3 h, 70%; (c) (1) KMnO4, acetone, 0°C to rt, 16 h; (2) MeOH, H2SO4, 70°C, 20
h, 53% (2 steps); (d) cyclopenten-1-ylboronic acid, Pd2(dba)3, tBu3P, Cs2CO3, 90 °C,
24 h, 65%; (e) H2, Pd/C, MeOH, THF, 60 °C, 48 h, 97%; (f) 25% aq. HCl, 70°C, 16 h,
92%; (g) cyclopentylzinc(II) bromide, Pd(dppf)Cl2, THF, 80 °C, 48 h, 31%; (h) MeOH,
2 M aq. LiOH, rt, 2 h, 90%.
Scheme 4. Preparation of 4,5- and 5,6-disubstituted picolinic acids.a
saponification afforded the desired 6-cyclopentoxynicotinic acid
44b in good overall yield. Similarly, Mitsunobu reaction of methyl
5-hydroxy-6-iodonicotinate 54 with cyclopentanol, subsequent
Negishi reaction of 55 using dimethylzinc followed by ester hy-
drolysis furnished the isomeric 5-cyclopentyloxy-nicotinic acid
44d.
pyridine derivatives incorporating either a small alkoxy and a
large alkyl or a large alkoxy and a small alkyl substituent in position
2 and 6 of the pyridine, respectively. Both substitution patterns
gave highly potent S1PR1 agonists with comparable selectivities
against S1PR3. Specifically, the 2-methoxy-6-(3-pentyl)-pyridine
7a showed the same affinity for S1PR1 as the aminopyridine 1 and
was about three times less selective against S1PR3. Similarly, the
isobutyl and cyclobutyl derivatives 7b and 7c were highly potent
S1P1 receptor agonists. These two compounds showed a further
increased affinity for S1PR3. The cyclopentyl substituted pyridine
7d reproduced the affinity profile of the 3-pentyl analogue 7a. On
the other hand, the bulkier cyclohexyl and the phenyl substituted
pyridines 7e and 7f both significantly lost affinity for S1PR1 and
S1PR3. The three compounds 7g, 7h, and 7i incorporating a large
alkoxy substituent and a methyl group were equipotent on S1PR1
and slightly more selective against S1PR3 when compared to their
closest analogues 7b, 7c, and 7d, respectively. As illustrated with
compounds 7j and 7k increasing the size of the alkyl substituent
increased the affinity towards S1PR3.
Interestingly, when the methoxy-pyridines are compared to the
corresponding alkyl pyridine analogues [75], the methyl derivative
reproduced the affinity profile of the methoxy compound much
better than the apparently structurally closer ethyl analogue. This
general finding is illustrated with the three cyclopentyl-pyridines
7d, 7l and 7m. While the methoxy pyridine 7d and methyl pyri-
dine 7l have an almost identical affinity profile for S1PR1 and S1PR3,
the ethyl derivative 7m is clearly more potent on S1PR3.
2.2. In vitro SAR discussion
First we explored the structure activity relationship (SAR) of 2-
alkoxy substituted 4-pyridines related to compound 7a. In a next
step we extended our studies to the corresponding 2-pyridines by
keeping the original meta-relationship between the alkoxy and the
alkyl substituent. We then also included 2-pyridine derivatives
wherein the alkoxy and the alkyl substituents were ortho to each
other. Finally, the SAR studies were completed with a set of pro-
totypical 3-pyridines. A selection of alkoxy substituted 4-, 2-, and 3-
pyridines shall illustrate the SAR of the three pyridine isomer series
with respect to their affinity for the S1P1 and S1P3 receptor, their
in vivo efficacy as well as phototoxicity. We focus our SAR discus-
sion on pyridine derivatives invariably incorporating the proto-
typical
(S)-3-(3-ethyl-4-(2-hydroxy-3-(2-hydroxy-acetamido)
propoxy)-5-methylphenyl)-1,2,4-oxadiazole moiety as this pattern
usually combined high receptor affinity and in vivo efficacy
[75,93,94]. EC50-values were measured with a GTPgS binding assay
using membrane preparations of CHO cells expressing either the
human S1P1 or S1P3 receptor. Table 1 compiles a series of 4-

332
M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
aReagents and conditions: (a) cyclopentylmagnesium bromide, Fe(acac)3, THF, 0 °C,
90 min, 71%, (b) 25% aq. HCl, 65 °C, 18 h, quant.; (c) cyclopentanol, PPh3, DEAD,
THF, 0-20°C, 1 h, 64-90% (49), 90% (55); (d) Me2Zn, Pd(dppf)Cl2, dioxane, 70-80 °C,
3-16 h, 58% (50), 68% (56); (e) 2 M aq. LiOH, MeOH or EtOH, rt, 2-15 h, 88% (44b),
quant. (44c), 83% (44d); (f) cyclopenten-1-yl boronic acid, Pd(dba)2, Cs2CO3,
dioxane, 100 °C, 18 h, 27%; (g) 5 bar H2, Pd/C, THF, EtOH, 55 °C, 48 h, 78%.
Scheme 5. Preparation of 5,6-disubstituted nicotinic acids.a
We then turned our attention to 2-pyridine derivatives while
highly potent S1PR1 agonists and both the 4-methoxy (35a) and the
6-methoxy (35b) 5-cyclopentylpyridines showed EC50 values close
to 1 nM. The two 4,5-disubstituted pyridines 35a and 35b were
more potent on S1PR3 when compared to their 4,6-disubstituted
isomer 24c and 24e, respectively.
keeping the meta substitution pattern of the 4-pyridine series. A
few examples are listed in Table 2. As in the 4-pyridine series, a 3-
pentyl (24a), an isobutyl (24b), and a cyclopentyl (24c) substituent
attached to position 6 of the 2-pyridine led to highly potent S1PR1
agonists with EC50 values around 1 nM. However, when compared
to the 4-pyridine isomers 7a, 7b, and 7d, the corresponding 2-
pyridines 24a, 24b, and 24c, respectively, were clearly more se-
lective against S1PR3. As illustrated with compounds 24d and 24e
swapping the positions of the methoxy and the alkyl substituent
again delivered highly potent S1PR1 agonists. The 3-pentyl deriv-
ative 24d was equipotent on S1PR1 but less selective against S1PR3
when compared to its isomer 24a. Conversely, the 4-cyclopentyl
pyridine 24e was less potent on both S1PR1 and S1PR3 than its
isomer 24c. Attaching a large alkoxy substituent to position 6 and a
methyl group to position 4 (e.g. compound 24f) reproduced the
affinity profile of the 6-alkyl-4-methoxy-isomer 24c. Interestingly,
however, when the large alkoxy group was placed in position 4 and
the methyl group was attached to position 6 (e.g. compound 24g),
the compound clearly lost affinity for S1PR1. Two 2-pyridine de-
rivatives wherein the substituents are ortho to each other are listed
in Table 3. Interestingly, also this substitution pattern delivered
Triggered by the observation that not only 4,6- but also 4,5- and
5,6-disubstituted 2-pyridines gave highly potent S1PR1 agonists,
we extended our SAR studies to 5,6-disubstituted 3-pyridines. A
few prototypical 3-pyridine derivatives are compiled in Table 4. Not
unexpectedly, also the 3-pyridines delivered potent S1PR1 agonists.
The potency and selectivity profile of the 5-methoxy-6-
cyclopentyl-3-pyridine 45a was very similar to the one of the 2-
pyridine isomer 35a. Moving the oxygen from position 5 to 6 to
give the 6-cyclopentoxy-5-methyl-3-pyridine 45b had little effect
on the compound's affinity towards S1PR1 and S1PR3. Swapping the
position of the small and the large substituent in 45a and 45b to
give the 3-pyridines 45c and 45d significantly increased the com-
pounds' affinity for S1PR3. On the other hand, changing the position
of the oxygen in the two compounds incorporating the larger
substituent in position 5 (i.e. 45c and 45d) clearly affected the
compounds potency on S1PR1.
In summary, while all pyridine series delivered potent S1PR1

M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
333
Table 1
SAR of 2,6-disubstituted pyridin-4-yl derivatives.
Compd R1
R2
EC50 [nM]a
%LCb (plasma conc. [nM])
DAc [h] (oral half-life [h])
S1PR1 S1PR3 3 h
6 h
24 h
48 h 72 h
96 h
144 h
1
Me
OMe
OMe
OMe
OMe
OMe
OMe
Isopropoxy
Cyclobutoxy
Cyclopentoxy Me
Cyclobutoxy Et
Cyclopentoxy Et
Cyclopentyl
Cyclopentyl
Diethyl-amino
3-Pentyl
Isobutyl
Cyclobutyl
Cyclopentyl
Cyclohexyl
Phenyl
0.6
0.4
0.3
0.3
0.6
7.3
350
96
27
22
71
690
1600
95
110
190
55
ꢀ64
ꢀ64
ꢀ70
ꢀ62
72
6 (9)
48
7a
7b
7c
7d
7e
7f
7g
7h
7i
ꢀ59 (750)
ꢀ57
ꢀ59 (470)
ꢀ59
ꢀ36 (46)
ꢀ66
bld (<3)
ꢀ59
ꢀ56
ꢀ61
ꢀ61
bld
24
ꢀ62 (2050) ꢀ65 (1860) ꢀ70 (1070)
ꢀ50 (68) ꢀ20 (12) 24 (18)
ꢀ58
ꢀ58
ꢀ72
ꢀ33
24
28
Me
Me
0.2
0.3
1.0
1.1
1.1
0.2
0.3
ꢀ66
ꢀ66
ꢀ67
ꢀ62
ꢀ74
ꢀ72
ꢀ70
ꢀ62
ꢀ28
ꢀ21
96
96
24 (13)
168
24
ꢀ69 (2200) ꢀ66 (2040) ꢀ74 (790)
ꢀ43 (67)
7j
ꢀ72
ꢀ64
ꢀ75
ꢀ71
ꢀ79
ꢀ76
7k
7le
7me
50
47
2.4
bld
Me
Et
a
EC50 values as determined in a GTPgS assay using membranes of CHO cells expressing either S1PR1 or S1PR3 [95]; EC50 values represent the geometric mean of at least
three independent measurements.
b
%Lymphocyte count reduction at indicated time point post oral administration of 10 mg/kg of compound as compared to baseline values, plasma concentrations given in
brackets, limit of quantification ¼ 3 nM.
c
DA [h] ¼ duration of maximal LC reduction as defined by the latest time point at which maximal LC reduction was observed.
d
bl ¼ LC reached baseline value.
e
Data from Ref. [75].
Table 2
SAR of 4,6-disubstituted pyridin-2-yl derivatives.
Compd
R1
R2
EC50 [nM]a
S1PR1
%LCb (plasma conc. [nM])
DAc [h] (oral half-life [h])
S1PR3
3 h
6 h
24 h
24a
24b
24c
24d
24e
24f
3-Pentyl
Isobutyl
Cyclopentyl
OMe
OMe
Cyclopentoxy
Me
OMe
OMe
OMe
3-Pentyl
Cyclopentyl
Me
1.2
1.5
0.7
1.1
4.5
1.0
11
520
1400
800
180
1500
930
ꢀ58
ꢀ66
ꢀ23
6
6
ꢀ58
ꢀ61
ꢀ30
ꢀ57 (410)
ꢀ63 (1790)
ꢀ51 (1610)
ꢀ50 (1560)
ꢀ50 (240)
ꢀ75 (1300)
ꢀ58 (890)
ꢀ36 (510)
10 (12)
2 (100)
18 (39)
33 (<9)
<3d(4)
6 (5)
6 (4)
<3d (<3)
24g
Cyclopentoxy
1400
a,b,cSee caption Table 1.
dMaximal LC reduction not reached.
agonists, there are clear differences between the SAR of the indi-
vidual isomer series. For instance, while a cyclopentoxy substituent
gave highly potent compounds in the 4-pyridine series and when
placed position 6 in the 2-pyridine series, it clearly hampered the
affinity for S1PR1 in position 4 of the 2-pyridine. In addition, the
cyclopentyl substituted 4-pyridine 7d and the corresponding 3-
pyridine 45c both showed an EC50 value of about 1 nM on S1PR1,
while performing the same structural modification in going from
the cyclopentoxy substituted 4-pyridine 7i to the corresponding 3-
pyridine analogue 45d led to a clear loss in affinity for S1PR1. (Note
that for the first pair (7d, 45c), the nature of the pyridine remains
an ‘O-alkylated 2-pyridone’, while for the second pair (7i, 45d), the

334
M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
Table 3
SAR of 4,5- and 5,6-di-substituted pyridin-2-yl derivatives.4
Compd
X
Y
EC50 [nM]a
S1PR1
%LCb (plasma conc. [nM])
DAc [h] (oral half-life [h])
S1PR3
3 h
6 h
24 h
35a
35b
N
CH
CH
N
0.6
1.4
210
920
ꢀ79 (2520)
ꢀ70 (690)
ꢀ86 (2020)
ꢀ78 (500)
ꢀ60 (71)
ꢀ23 (61)
6 (4)
6 (6)
a,b,cSee caption Table 1; additional time point: %LC(plasma conc.): 35a 72 h: back to baseline (<3 nM).
Table 4
SAR of 5,6-disubstituted pyridin-3-yl derivatives.5
Compd
R1
R2
EC50 [nM]a
S1PR1
%LCb (plasma conc. [nM])
DAc [h] (oral half-life [h])
S1PR3
3 h
6 h
24 h
45a
45b
45c
45d
Cyclopentyl
Cyclopentoxy
OMe
OMe
Me
Cyclopentyl
Cyclopentoxy
0.3
0.7
1.5
21
71
140
17
ꢀ73 (2390)
ꢀ76 (1140)
ꢀ82 (2950)
ꢀ85 (960)
ꢀ84 (970)
ꢀ85 (510)
24 (11)
24 (16)
Me
38
a,b,cSee caption Table 1; additional time points: %LC (plasma conc.): 45a 96 h: back to baseline (<10 nM); 45b 72 h: ꢀ32% (58 nM), 144 h: back to baseline (<3 nM).
nature of the pyridine changes from an ‘O-alkylated 2-pyridone’ to
a ‘3-alkoxy-pyridine’.) Interestingly, in both cases the affinity for
S1PR3 increases by a factor of 4e5. These observations clearly
indicate that steric and electronic factors contribute to the ligand
receptor binding interaction.
on S1PR1 showed in vivo efficacy.
All 4-pyridines tested in vivo showed a rapid onset of action and
the LC was maximally reduced at 3 h post administration already
(Table 1). In addition, these compounds showed a DA of at least
24 h. The cyclopentyl pyridine 7d and the cyclopentoxy pyridine 7i
showed micromolar plasma concentrations over 24 h, suggesting
that the 4-pyridines were well absorbed and slowly cleared. Oral
half-lifes of 7d and 7i were 13 and 18 h, respectively. The 3-pentyl
substituted pyridine 7a showed a DA of less than 24 h. In agreement
with earlier observations [75], and as demonstrated by the plasma
concentrations and the oral half-life the 3-pentyl pyridine 7a was
more rapidly cleared than its cyclopentyl analogue 7d. As a
consequence the LC already partially recovered at 24 h.
Interestingly, even though the 2-pyridines listed in Table 2
showed similar potencies when compared to the corresponding
4-pyridine analogues, they clearly displayed a different behaviour
in vivo. The duration of action of the 2-pyridines was consistently
shorter (e.g. compare 24b with 7b, 24c with 7d, 24g and 24f with
7i) and at a dose of 10 mg/kg none of the tested 2-pyridines pro-
duced a maximal LC reduction for 24 h. Plasma concentrations at
3 h were comparable to those reached with the corresponding 4-
pyridines but dropped much more rapidly at 6 and, in particular,
24 h (compare e.g. 24f and 24g with 7i, and 24e with 7d) translating
into rather short oral half-lifes of 3e5 h. The 4,5- and 5,6-
disubstituted 2-pyridines in Table 3 showed a very similar in vivo
behaviour. Also here, plasma concentrations were high at 3 h post
administration but dropped significantly at 24 h. With 4 and 6 h,
respectively, the oral half-life of compounds 35a and 35b was too
short to maintain maximal LC reduction over 24 h.
2.3. In vivo efficacy assessment
For prototypical compounds (usually with an EC50 ꢁ10 nM on
S1PR1) the efficacy to reduce the peripheral blood lymphocyte
count (LC) was assessed in the rat. The compounds were adminis-
tered orally at a dose of 10 mg/kg to male Wistar rats and blood LC
was measured shortly before and 3, 6, and 24 h after compound
administration. For several compounds later sampling time points
were added to confirm complete recovery of the LC. Due to sig-
nificant inter-individual variability and the circadian rhythm of the
absolute number of circulating lymphocytes, a LC change of 15%
was considered to be within baseline. On the other hand, a LC
reduction of ꢂ60% was interpreted as the maximal effect that is
achieved under the experimental conditions chosen [95]. The
duration of action (DA) was defined as the latest time point at
which maximal LC reduction was observed. For about half of the
compounds tested in vivo, plasma concentrations reached in the LC
experiment were measured and oral half-lives were calculated
based on these exposure data. Values for the measured LC and
plasma concentrations, and estimated DA and oral half-lives are
given in Tables 1e4. In line with what is reported for S1PR1 agonists
[24,25,44,54,58], our pyridine derivatives reduced the LC in a rapid
and reversible fashion. In general, compounds with an EC50 <5 nM

M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
335
On the other hand, the two 3-pyridines 45a and 45b (Table 4)
exhibited pharmacokinetic and pharmacodynamic profiles com-
parable to those of the corresponding 4 pyridines 7d and 7i,
respectively. Both compounds sustained rather high plasma con-
centrations over 24 h and as a consequence showed maximal LC
reduction for at least 24 h.
Hence, the PK and PD data suggest that the nature of the pyri-
dine isomer rather than the substitution pattern governs the
pharmacokinetic behaviour of the compound. The oral half life for
the 2-, 3-, and 4-pyridines was estimated to be 3e6 h, 11e16 h, and
13e18 h, respectively. Clearance and compound tissue distribution
(Vss) determine the half-life of a compound in vivo. To rationalize
the above PK observations, we predicted the Vss values for the
compounds tested in vivo using a rat physiology based pharmaco-
kinetic model using GastroPlus. The predicted values for Vss were
in the range of 0.8e1.5 L/kg for all three pyridine isomer classes
suggesting that the observed differences in half-life are largely
driven by differences in the clearance of the compounds.
6.5e7.5 the pyridines 7i, 35b, and 45a showed clear sign of in vitro
phototoxicity. These three compounds showed strong light ab-
sorption at wavelengths >290 nm (ε >10 000). Despite showing a
similarly high extinction (ε >10 000) at the second lmax compounds
24f, 35a and 45b were not phototoxic. Interestingly, for 35a and 45b
the second absorption maximum occurred at wavelengths below
290 nm. This observation prompted us to revisit the technical de-
tails of the 3T3 fibroblast phototoxicity assay. As recommended by
the guidelines, and due to the UVB sensitivity of the Balb/c 3T3 cell
line, the light filter used in the mouse 3T3 fibroblast phototoxicity
assay completely blocks light transmission at wavelengths
ꢁ300 nm. While some transmission occurs at 310 nm, significant
exposure is obtained above 320 nm only (for details see Supporting
information). We therefore calculated the area under the curve
(AUC) of the UV spectra between 310 and 350 nm (highest wave-
length at which light absorption was observed) normalized to the
compound concentration in mM. As shown by the results given in
Table 5, the normalized AUC values indeed correlate with the
compound's phototoxic potential. Compound 1 showing the high-
est PIF value also reached the highest AUC value (0.54). The three
phototoxic alkoxy pyridines 7i, 35b, and 45a had smaller AUC
values than 1 but scored highest when compared to the remaining,
non-phototoxic alkoxy pyridines. The threshold for the AUC to
2.4. In vitro phototoxicity assessment
As mentioned in the introduction, the observation that com-
pound 7a was not in vitro phototoxic (PIF ¼ 0.9) triggered a more
detailed investigation whether alkoxy substituted pyridine de-
rivatives would serve as potent and efficacious S1P1 receptor ago-
nists. As a first step to assess the phototoxic potential of these
compounds, UV spectra of prototypical representatives of each
pyridine isomer series were measured. While all compounds
showed an absorption maximum at around 254 nm, some showed
a second absorption maximum at wavelengths >290 nm. However,
as the molar extinction at wavelengths ꢂ290 nm of all compounds
e including 7a e were in a range (ε >1000) where the guidelines
[80,81] recommend a more detailed phototoxicity assessment, the
in vitro photo-irritancy factor (PIF) of these compounds was
determined using the Balb/c mouse 3T3 fibroblast neutral red up-
take assay [76,77]. Table 5 compiles the obtained PIF values and the
molar extinction coefficient at the absorbance maximum at
>290 nm. In case the UV spectrum showed no absorption
maximum at wavelengths >290 nm, the molar extinction at
290 nm is given.
observe in vitro phototoxicity appears to be around 0.10 OD nm/mM.
Whether this threshold is characteristic for the alkoxy-pyridine
series of the present study or could also be applied to other com-
pound series is not known and additional studies extending the
concept of integrating UV spectra to better predict in vitro photo-
toxicity of other compound classes are certainly warranted.
The above considerations clearly confirm that the UV absorption
correlates with the phototoxic potential of a compound. However, it
was much more difficult to establish a direct structure phototox-
icity relationship. By analyzing the UV spectra in more detail, a few
structural elements that affect the compound's UV spectrum and
hence its behaviour in the phototoxicity assay could be identified
and are discussed below. While all compounds showed a first ab-
sorption maximum between 249 and 264 nm, the 4-pyridines (7a,
7d, 7i), the 3-pyridines (45a, 45b) and the 2-pyridine (35b) bearing
the alkoxy substituent ortho to the pyridine nitrogen had a second
absorption maximum above 290 nm. In contrast, the 2-pyridines
wherein the alkoxy substituent is attached in para to the nitrogen
(24a (see Supporting information), 24c, 24g, 35a) show no second
absorption maximum. While the molar extinction at 290 nm of
these four 4-alkoxy-2-pyridines ranged from 4600 to 11800, their
Compared to aminopyridine 1, the PIF values of the alkoxy
pyridines listed in Table 5 are much smaller. In fact, we consider
compounds 7a, 7d, 24c, 24e, 24f, 24g, 35a, and 45b to be non-
phototoxic in vitro. In contrast, with a PIF value in the range of
Table 5
Photo-irritancy factor (PIF) values of prototypical compounds of each pyridine isomer series.a
Compound
Molar extinction coefficient (ε) at lmax >290 nm (lmax [nm])
AUC310e350 nm b [OD nm/
m
M]
PIF value
1
7a
7d
7i
24c
24e
24f
24g
35a
35b
45a
45b
2000 (380)c
2200 (310)
4300 (311)
18,100 (296)
4600f
0.54d
0.05
0.09
0.12
0
0.09
0.10
0
0.02
0.17
0.15
0.02
>200e
0.9
2.3
7.5
2.4
2.1
2.0
1.3
2.1
6.5
7.6
1.4
3700 (313)
12,500 (301)
6300f
11,800f
18,800 (302)
14,000 (302)
23,700f
a
As measured in the mouse 3T3 fibroblast neutral red uptake assay; for details see Supporting information.
b
c
d
e
f
AUC of the UV spectra between 310 and 350 nm normalized to compound concentration in
Compound is fluorescent.
AUC310e435 nm as absorption extends to 435 nm.
Please note that compound 1 did not degrade in the course of the in vitro phototoxicity assay.
No absorption maximum at wavelengths >290 nm, extinction at 290 nm given.
mM.

336
M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
AUC310e350nm value is small (0e0.02) and none of the compounds
was phototoxic in vitro. For the compounds showing a second ab-
sorption maximum >290 nm, the size of the alkyl group attached to
the oxygen appears to influence the compound's UV-absorption.
When a cyclopentyl rather than a methyl group was attached to
the oxygen the compound absorbed more UV light at wavelengths
>290 nm (compare e.g. 24e with 24f, and 7d with 7i). While UV
absorption of the 2-cyclopentyl-6-methoxy pyridine 7a is not suf-
ficient to induce in vitro phototoxicity, the one of the isomeric 2-
methyl-6-cyclopentoxy pyridine 7i is obviously strong enough to
trigger a clear phototoxicity signal. Unfortunately, a more detailed
correlation between structure and UV spectrum remained elusive
at this point.
LCeMS. LCeMS: Finnigan MSQ™ plus or MSQ™ surveyor (Dionex,
Switzerland), with HP 1100 Binary Pump and DAD (Agilent,
Switzerland), column: Zorbax SB-AQ, 5
(Agilent), or (LCeMS*) Zorbax SB-AQ, 5
m
m
m, 120 Å, 4.6 ꢃ 50 mm
m, 120 Å, 4.6 ꢃ 20 mm
(Agilent); gradient: 5e95% acetonitrile in water containing 0.04% of
trifluoroacetic acid, within 1 min, flow: 4.5 mL/min; tR is given in
min. *denotes basic conditions: gradient: 5e95% acetonitrile in
water containing 0.04% of ammonium hydroxide, within 1 min,
flow: 4.5 mL/min. Purity of all target compounds was confirmed by
a second, independent LCeMS method (LCeMS**) on a Waters
Acquity UPLC system equipped with an ACQ-PDA detector, an ACQ-
ESL detector, and an ACQ-SQ detector; column: ACQUITY UPLC BEH
C18 1.7
mm, 2.1 ꢃ 50 mm; gradient: 2e98% acetonitrile containing
In brief, the majority of the alkoxy pyridine isomers are not
phototoxic. In addition, the PIF value of those compounds that did
show in vitro phototoxicity is much lower compared to the one of
the aminopyridine 1. Furthermore, while the maximal extinction at
wavelengths ꢂ290 nm serves as an easy and rapidly available basis
for the decision whether the phototoxic potential of a compound
needs to be assessed in detail, the somewhat more sophisticated
concentration normalized integration of a compound's UV/VIS
spectrum may be more suitable to predict and rationalize the re-
sults of the in vitro phototoxicity assay.
0.045% formic acid in water containing 0.05% formic acid over
1.8 min; flow 1.2 mL/min; 60 ꢄC. According to the LCeMS analyses,
target compounds showed a purity >95% (UV at 211e214 nm).
Purity and identity were further corroborated by LC-HRMS:
Analytical pump: Waters Acquity Binary, Solvent Manager, MS:
SYNAPT G2 MS, source temperature: 150 ꢄC, desolvation tempera-
ture: 400 ꢄC, desolvation gas flow: 400 L/h; cone gas flow: 10 L/h,
extraction cone:
4 RF; lens: 0.1 V; sampling cone: 30;
capillary:1.5 kV; high resolution mode; gain: 1.0, MS function: 0.2 s
per scan, 120e1000 amu in full scan, centroid mode. Lock spray:
Leucine enkephalin 2 ng/mL (556.2771 Da) scan time 0.2 s with
interval of 10 s and average of 5 scans; DAD: Acquity UPLC PDA
3. Conclusions
Detector. Column: Acquity UPLC BEH C18 1.7
mm 2.1 ꢃ 50 mm from
In summary, starting from the phototoxic aminopyridine 1 we
identified a series of alkoxy substituted 2-, 3-, and 4-pyridine an-
alogues with no or a significantly reduced phototoxic potential.
While all three alkoxy pyridine isomer series showed similarly high
activity on the S1P1 receptor, their selectivity against the S1P3 re-
ceptor and their in vivo efficacy differed significantly. The 2-
pyridines were clearly more selective against S1PR3 when
compared to the corresponding 3- or 4-pyridine analogues (e.g.
compare 35a vs 45a, 24c and 24e vs 7d). Interestingly, despite
having a different substitution pattern at the pyridine ring, the
selectivity profile of the 3-pyridines was almost identical to the one
of the 4-pyridines (e.g. compare 45a vs 7d, 45b vs 7i). On the other
hand, the oral half-life and as a consequence the pharmacological
duration of action after oral administration of the 2-pyridines was
significantly shorter when compared to the corresponding 3- or 4-
pyridines. This indicates that the position of the pyridine nitrogen
relative to the oxadiazole rather than the position of the alkyl and
alkoxy substituents governs the pharmacokinetic behaviour of the
compound. With respect to their potency and selectivity profile and
their in vivo efficacy, the 2-pyridine 24c, the 3-pyridine 45b, and the
4-pyridine 7d are the best representatives of each pyridine isomer
series. None of these three compounds was phototoxic in the 3T3
mouse fibroblast neutral red uptake assay. Compound 45b, in
particular, incorporates an attractive combination of a long dura-
tion of action in vivo and a high selectivity against S1PR3. At this
point more detailed profiling and careful assessment of the
importance of the different properties of these compounds is
needed to judge their potential for human use.
Waters, thermostated in the Acquity UPLC Column Manager at
60 ꢄC. Eluents: water þ0.05% formic acid; B: acetonitrile þ0.05%
formic acid. Gradient: 2e98% B over 3.0 min. Flow: 0.6 mL/min.
Detection: UV 214 nm and MS, tR is given in min; and by NMR
spectroscopy: (Bruker Avance II, 400 MHz UltraShield™, 400 MHz
(1H), 100 MHz (13C); or Bruker Avance III HD, Ascend 500 MHz (1H),
125 MHz (13C) magnet equipped with DCH cryoprobe), chemical
shifts are reported in parts per million (ppm) relative to tetrame-
thylsilane (TMS), and multiplicities are given as
s (singlet),
d (doublet), t (triplet), q (quartet), quint (quintuplet), h (hextet),
hept (heptuplet) or m (multiplet), br ¼ broad, coupling constants
are given in Hz). Compounds were purified by either flash column
chromatography (CC) on silica gel 60 (Fluka SigmaeAldrich,
Switzerland), prep. TLC glass plates coated with silica gel 60 F254
(0.5 mm), or by prep. HPLC (Waters XBridge™ Prep C18, 5
mm, OBD,
19 ꢃ 50 mm, or Waters X-terra™ RP18, 19 ꢃ 50 mm, 5 m, gradient
m
of acetonitrile in water containing 0.4% of formic acid, flow 75 mL/
min), or by MPLC (Labomatic MD-80-100 pump, Linear UVIS-201
detector, column: 350 ꢃ 18 mm, Labogel-RP-18-5s-100, gradient:
10% methanol in water to 100% methanol). UV-spectra were ac-
quired on a Varian Cary 50 Scan UV/Vis Spectrometer with Varian
UV Scan Software Version 3.0 and Cary Win UV Kinetics Application
Software Version 3.0.
4.1.1. Preparation of pyridine carboxylic acid building blocks
4.1.1.1. 2-Methoxy-6-(pentan-3-yl)isonicotinic acid (12a)
(a) Under argon, Pd(dppf) (3.04 g, 4 mmol) was added to a so-
lution of 2-chloro-6-methoxy-isonicotinic acid methyl ester
(50 g, 0.248 mol) in THF (100 mL). A 0.5 M solution of 3-
pentylzincbromide in THF (550 mL) was added via drop-
ping funnel. Upon complete addition, the mixture was
heated to 85 ꢄC for 18 h before it was cooled to rt. Water
(5 mL) was added and the mixture is concentrated. The crude
product was purified by filtration over silica gel (350 g) using
heptane:EA 7:3 to give methyl 2-methoxy-6-(pentan-3-yl)-
isonicotinate 11a (53 g, 90%) as a pale yellow oil; LCeMS:
4. Experimental section
4.1. Syntheses
All reagents and solvents were used as purchased from com-
mercial sources (SigmaeAldrich Switzerland, Lancaster Synthesis
GmbH, Germany, Acros Organics USA). Moisture sensitive reactions
were carried out under an argon atmosphere. Progress of the re-
actions was followed either by thin-layer chromatography (TLC)
analysis (Merck, 0.2 mm silica gel 60 F254 on glass plates) or by
tR ¼ 0.99 min, [M þ 1]þ ¼ 238.32; 1H NMR (CDCl3):
d 7.23 (s,

M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
337
1H), 7.12 (d, J ¼ 0.6 Hz, 1H), 3.96 (s, 3H), 3.95 (s, 3H),
2.48e2.56 (m, 1H), 1.64e1.82 (m, 4H), 0.79 (t, J ¼ 7.4 Hz, 6H).
mixture was stirred at 10 ꢄC for 30 min before it was diluted
with EA and washed with water and brine. The organic
extract was dried over MgSO4, filtered and concentrated. The
crude product was purified by CC on silica gel eluting with
heptane:EA 9:1 to give ethyl 5-bromo-6-(cyclopentyloxy)
nicotinate 49 (5.87 g, 90%) as a colourless oil; LCeMS:
tR ¼ 1.12 min; [M þ 1]þ ¼ 315.85.
(b) A solution of methyl 2-methoxy-6-(pentan-3-yl)iso-
nicotinate 11a (50 g, 211 mmol) in water (50 mL), 32% aq.
NaOH (50 mL) and ethanol (250 mL) was stirred at 80 ꢄC for
2 h. The mixture was concentrated, diluted with water and
extracted with tert-butyl methyl ether (TBME). The aqueous
phase was acidified by adding 25% aq. HCl and then extracted
twice with EA (2 ꢃ 400 mL). The organic extract was
concentrated and the residue was precipitated from water
(550 mL). The product was collected, dried under high vac-
uum to give the title compound (40.2 g, 86%) as a white
powder; LCeMS: tR ¼ 0.85 min, [M þ 1]þ ¼ 224.37; 1H NMR
(c) To a solution of ethyl 5-bromo-6-(cyclopentyloxy)nicotinate
49 (2.00 g, 6.37 mmol) in dioxane (100 mL) under argon,
Pd(dppf) (104 mg, 0.127 mmol) and dimethylzinc (6.5 mL of a
1.2 M solution in toluene) was added. The mixture was stir-
red at 80 ꢄC for 3 h. The mixture was cooled to rt and the
reaction was quenched by adding water. The mixture was
extracted with EA. The organic extract was washed with
water, dried over MgSO4, filtered and concentrated. The
crude product was purified by CC on silica gel eluting with
heptane:EA 9:1 to give ethyl 6-(cyclopentyloxy)-5-
methylnicotinate 50 (920 mg, 58%) as a colourless oil;
LCeMS: tR ¼ 1.09 min; [M þ 1]þ ¼ 250.05.
(DMSO-d6): d 7.20 (s, 1H), 7.00 (s, 1H), 3.88 (s, 3H), 2.53e2.59
(m, 1H), 1.58e1.74 (m, 4H), 0.74 (t, J ¼ 7.2 Hz, 6H).
4.1.1.2. 2-Isobutyl-6-methoxyisonicotinic acid (12b)
(a) Under argon, to
a
solution of ethyl 2-chloro-6-
(d) A solution of ethyl 6-(cyclopentyloxy)-5-methylnicotinate 50
(920 mg, 3.69 mmol) in ethanol (10 mL) and 1 M aq. NaOH
(2 mL) was stirred at rt for 2 h. The mixture was acidified by
adding 1 N aq. HCl (5 mL) and extracted twice with EA. The
combined organic extracts were dried over MgSO4, filtered,
concentrated and dried under high vacuum to give the title
methoxyisonicotinate 10 (3.53 g, 16.4 mmol) in THF
(50 mL) was added Fe(acac)3 (636 mg, 1.80 mmol) followed
by NMP (2.27 g, 22.9 mmol). The mixture was cooled
to ꢀ75 ꢄC before isobutylmagnesium chloride (13 mL of a 2 M
solution in THF) was added. The mixture was stirred
at ꢀ75 ꢄC for 1 h, then at 0 ꢄC for 1 h. The reaction was
quenched by carefully adding water (100 mL) and the
mixture was extracted with diethyl ether. The organic extract
was dried over MgSO4, filtered and concentrated. The crude
product was purified by CC on silica gel eluting with hepta-
ne:EA 9:1 to give ethyl 2-isobutyl-6-methoxyisonicotinate
compound (720 mg, 88%) as
a white solid; LCeMS:
tR ¼ 0.92 min; [M þ 1]þ ¼ 222.01; 1H NMR (CDCl3):
d 8.78 (d,
J ¼ 1.7 Hz, 1H), 8.01 (s br, 1H), 5.55e5.60 (m, 1H), 2.21 (s, 3H),
1.95e2.08 (m, 2H), 1.77e1.90 (m, 4H), 1.62e1.73 (m, 2H).
11b (1.54 g, 40%) as
a
pale yellow oil; LCeMS:
4.1.2. Method A: Preparation of target compounds starting from
hydroxy-amidine 9
tR ¼ 1.06 min, [M þ 1]þ ¼ 238.13.
4.1.2.1. N-[(S)-3-(2-Ethyl-4-{5-[2-(1-ethyl-propyl)-6-methoxy-pyr-
idin-4-yl]-1,2,4-oxadiazol-3-yl}-6-methyl-phenoxy)-2-hydroxy-pro-
(b) A solution of ethyl 2-isobutyl-6-methoxyisonicotinate 11b
(1.53 g, 6.45 mmol) in 25% aq. HCl (20 mL) was stirred at
70 ꢄC for 5 h. The solvent was removed in vacuo and the
remaining residue was dried under HV to give the still water
containing hydrochloride salt of the title compound (1.45 g,
quant.) as a pale yellow solid; LCeMS: tR ¼ 0.87 min,
pyl]-2-hydroxy-acetamide (7a). To
a solution of 2-methoxy-6-
(pentan-3-yl)isonicotinic acid hydrochloride 12a (200 mg,
0.77 mmol) in DMF (3 mL) and THF (10 mL), DIPEA (200 mg,
1.54 mmol) followed by TBTU (272 mg, 0.847 mmol) was added.
The mixture was stirred at rt for 10 min before N-((S)-3-[2-ethyl-4-
(N-hydroxycarbamimidoyl)-6-methyl-phenoxy]-2-hydroxy-pro-
pyl)-2-hydroxy-acetamide 9 [75,96] (251 mg, 0.77 mmol) was
added. Stirring was continued for 30 min. The mixture was diluted
with EA (100 mL) and washed with brine. The organic extract was
concentrated and the residue was dissolved in dioxane (50 mL). The
mixture was stirred at 100 ꢄC for 2 h before it was concentrated. The
crude product was purified on prep. TLC plates using DCM:MeOH
10:1 to give the title compound (180 mg, 46%) as a colourless foam;
[M þ 1]þ ¼ 210.09; 1H NMR (DMSO-d6):
d 7.21 (s, 1H), 7.00 (s,
1H), 4.37 (s br, 2H þ H2O), 3.87 (s, 3H), 2.58 (d, J ¼ 7.1 Hz, 2H),
1.99e2.13 (m, 1H), 0.88 (d, J ¼ 6.7 Hz, 6H).
4.1.1.3. 6-Cyclopentoxy-5-methyl-nicotinic acid (44b)
(a) A suspension of 5-bromo-6-hydroxy nicotinic acid (5.00 g,
22.9 mmol) in ethanol (200 mL) and H2SO4 (0.5 mL) was
stirred at 90 ꢄC for 18 h. The solid dissolves slowly. The
mixture was concentrated and the residue was dissolved in
EA and washed with water. The washing was extracted back
once with DCM. The combined organic extracts were dried
over MgSO4, filtered and concentrated to give ethyl 5-bromo-
6-hydroxy nicotinate 48 (5.13 g, 91%) as a pale yellow oil;
LCeMS: tR ¼ 0.72 min; [M þ 1]þ ¼ 245.87.
LCeMS**: tR ¼ 1.49 min, [M þ 1]þ ¼ 513.4; 1H NMR (CDCl3):
d 7.88
(s, 1H), 7.86 (s, 1H), 7.43 (d, J ¼ 1.1 Hz, 1H), 7.31 (d, J ¼ 1.1 Hz, 1H),
7.09 (s br, 1H), 4.18e4.25 (m, 3H), 4.01 (s, 3H), 3.75e3.93 (m, 4H),
3.50e3.59 (m, 1H), 2.74 (q, J ¼ 7.5 Hz, 2H), 2.54e2.64 (m, 1H), 2.38
(s, 3H), 1.65e1.89 (m, 4H), 1.31 (t, J ¼ 7.5 Hz, 3H), 0.84 (t, J ¼ 7.4 Hz,
6H); 13C NMR (CDCl3):
d 174.4, 173.4, 168.9, 165.1, 164.5, 157.3, 137.7,
133.4, 131.6, 128.4, 126.7, 122.6, 113.5, 106.1, 74.1, 70.2, 62.2, 53.7,
51.2, 42.3, 27.9, 22.9, 16.5, 14.8, 12.2; LC-HRMS: tR ¼ 1.58 min,
[M þ H]/z ¼ 513.2713 (C27H37N4O6), found ¼ 513.2711; UV(MeOH):
249(31800), 310(2200), ε290(1800).
(b) To a solution of ethyl 5-bromo-6-hydroxy nicotinate 48
(5.10 g, 20.7 mmol) in THF (150 mL) triphenylphosphine
(8.16 g, 31.1 mmol) and cyclopentanol (2.14 g, 24.9 mmol)
was added. The mixture was cooled to 0 ꢄC before diethyl
azodicarboxylate (14.2 mL, 40% solution in toluene) was
slowly added. Upon completion of the addition, the green
4.1.2.2. N-((S)-3-{4-[5-(6-Cyclopentyl-4-methoxy-pyridin-2-yl)-
1,2,4-oxadiazol-3-yl]-2-ethyl-6-methyl-phenoxy}-2-hydroxy-pro-
pyl)-2-hydroxy-acetamide (24c). Method A, 57%, pale yellow resin;
LCeMS**: tR ¼ 1.28 min, [M þ 1]þ ¼ 511.3; 1H NMR (CDCl3):
d 7.92

338
M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
(178 mg, 408 mmol, as 40% solution in toluene) was added.
(s, 1H), 7.91 (s, 1H), 7.69 (d, J ¼ 2.3 Hz, 1H), 7.03 (t br, J ¼ 5.6 Hz, 1H),
6.94 (d, J ¼ 2.3 Hz, 1H), 4.19e4.24 (m, 3H), 3.99 (s, 3H), 3.90 (dd,
J1 ¼ 4.7 Hz, J2 ¼ 9.6 Hz, 1H), 3.84 (dd, J1 ¼ 6.3 Hz, J2 ¼ 9.5 Hz, 1H),
3.80 (ddd, J1 ¼ 3.2 Hz, J2 ¼ 6.7 Hz, J3 ¼ 14.2 Hz, 1H), 3.49e3.56 (m,
1H), 3.32e3.40 (m, 1H), 2.74 (q, J ¼ 7.6 Hz, 2H), 2.38 (s, 3H),
2.16e2.24 (m, 2H), 1.71e1.93 (m, 6H), 1.32 (t, J ¼ 7.6 Hz, 3H); 13C
Stirring was continued at rt for 18 h. The mixture was
concentrated and the crude product was purified on prep.
TLC plates using heptane:EA 7:3 to give (S)-3-(3-ethyl-5-
methyl-4-(oxiran-2-ylmethoxy)phenyl)-5-(2-isobutyl-6-
methoxypyridin-4-yl)-1,2,4-oxadiazole (88 mg, 76%) as a
colourless oil; LCeMS: tR ¼ 1.23 min; [M þ 1]þ ¼ 424.22.
NMR (CDCl3):
d 174.6, 172.6, 169.2, 168.8, 166.9, 157.2, 144.2, 137.6,
131.5,128.5,126.9,122.7,110.2,108.1, 74.0, 70.3, 62.3, 55.7, 48.1, 42.3,
33.8, 25.8, 23.0, 16.5, 14.9; LC-HRMS: tR ¼ 1.35 min, [M þ H]/
(c) A
solution
of
(S)-3-(3-ethyl-5-methyl-4-(oxiran-2-
z
¼
511.2556 (C27H35N4O6), found
¼
511.2563; UV(MeOH):
ylmethoxy)phenyl)-5-(2-isobutyl-6-methoxypyridin-4-yl)-
255(27200), ε290(4600).
1,2,4-oxadiazole (72 mg, 170 mmol) in 7 M NH3 in methanol
(5 mL) was stirred in a sealed vessel at 60 ꢄC for 16 h. The
mixture was concentrated to give crude (S)-1-amino-3-(2-
ethyl-4-(5-(2-isobutyl-6-methoxypyridin-4-yl)-1,2,4-
4.1.2.3. N-((S)-3-{4-[5-(6-Cyclopentyloxy-5-methyl-pyridin-3-yl)-
1,2,4-oxadiazol-3-yl]-2-ethyl-6-methyl-phenoxy}-2-hydroxy-pro-
pyl)-2-hydroxy-acetamide (45b). Method A, 25%, white solid
(42 mg); LCeMS**: tR ¼ 1.45 min, [M þ 1]þ ¼ 511.3; 1H NMR
oxadiazol-3-yl)-6-methylphenoxy)propan-2-ol (95 mg,
quant.) as
a
colourless oil; LCeMS: tR
¼
0.92 min;
[M þ 1]þ ¼ 441.26.
(CDCl3):
d
8.86 (d, J ¼ 2.3 Hz, 1H), 8.15 (d, J ¼ 2.3 Hz, 1H), 7.88 (d,
J ¼ 1.8 Hz, 1H), 7.86 (d, J ¼ 1.6 Hz, 1H), 6.96 (t br, J ¼ 5.0 Hz, 1H),
5.57e5.62 (m, 1H), 4.19e4.26 (m, 3H), 3.90 (dd, J1 ¼ 4.6 Hz,
J2 ¼ 9.5 Hz, 1H), 3.85 (dd, J1 ¼ 6.3 Hz, J2 ¼ 9.6 Hz, 1H), 3.81 (ddd,
J1 ¼ 3.1 Hz, J2 ¼ 6.7 Hz, J3 ¼ 14.0 Hz, 1H), 3.50e3.57 (m, 1H), 3.28 (s
br, 1H), 2.75 (q, J ¼ 7.5 Hz, 2H), 2.40 (s, 3H), 2.38 (s br, 1H), 2.27 (s,
3H),1.98e2.08 (m, 2H),1.80e1.91 (m, 4H), 1.64e1.74 (m, 2H), 1.32 (t,
(d) To a solution of the above (S)-1-amino-3-(2-ethyl-4-(5-(2-
isobutyl-6-methoxypyridin-4-yl)-1,2,4-oxadiazol-3-yl)-6-
methylphenoxy)propan-2-ol (95 mg, 170
(32 mg, 237
(42 mg, 323
m
mol) in THF, HOBt
mol), glycolic acid (18 mg, 237 mol), DIPEA
mol) and EDC HCl (46 mg, 237 mol) was
m
m
m
m
J ¼ 7.6 Hz, 3H); 13C NMR (CDCl3):
d 174.3, 172.4, 168.5, 164.6, 157.1,
added. The mixture was stirred at rt for 2 h before it was
diluted with EA and washed with sat. aq. NaHCO3 solution.
The organic extract was concentrated and the crude product
was purified on prep. TLC plates using DCM:methanol 10:1 to
give the title compound (62 mg, 73%) as a colourless foam;
LCeMS: tR ¼ 1.42 min, [M þ 1]þ ¼ 499.3; 1H NMR (CDCl3):
145.0, 137.6, 137.2, 131.5, 128.4, 126.7, 123.0, 122.2, 113.6, 79.2, 74.0,
70.4, 62.3, 42.2, 33.0, 23.9, 23.0, 16.5, 15.9, 14.9; LC-HRMS:
tR
found
ε290(23,700).
¼
1.53 min, [M
þ
H]/z
¼
511.2556 (C27H35N4O6),
¼
511.2561; UV(MeOH): 264(29,100), 285(24,100),
d
7.88 (d, J ¼ 1.8 Hz, 1H), 7.86 (d, J ¼ 1.7 Hz, 1H), 7.46 (d,
J ¼ 1.0 Hz, 1H), 7.33 (d, J ¼ 0.9 Hz, 1H), 7.04 (t, J ¼ 5.7 Hz, 1H),
4.19e4.25 (m, 3H), 4.03 (s, 3H), 3.90 (dd, J1 ¼ 4.6 Hz,
J2 ¼ 9.6 Hz, 1H), 3.85 (dd, J1 ¼ 6.2 Hz, J2 ¼ 9.5 Hz, 1H), 3.80
(ddd, J1 ¼ 3.4 Hz, J2 ¼ 6.7 Hz, J3 ¼ 14.2 Hz, 1H), 3.50e3.57 (m,
1H), 2.75 (q, J ¼ 7.5 Hz, 2H), 2.70 (d, J ¼ 7.2 Hz, 2H), 2.39 (s,
3H), 2.23 (hept, J ¼ 6.7 Hz, 1H), 1.32 (t, J ¼ 7.5 Hz, 3H), 1.00 (d,
4.1.3. Method B: Preparation of target compounds starting from
hydroxy-amidine 8
4.1.3.1. N-((S)-3-{2-Ethyl-4-[5-(2-isobutyl-6-methoxy-pyridin-4-yl)-
1,2,4-oxadiazol-3-yl]-6-methyl-phenoxy}-2-hydroxy-propyl)-2-
hydroxy-acetamide (7b)
J ¼ 6.6 Hz, 6H); 13C NMR (CDCl3):
d 174.2, 172.7, 168.9, 164.3,
(a) A solution of 2-isobutyl-6-methoxyisonicotinic acid 12b
(1.45 g, 5.90 mmol) in DCM (50 mL) DIPEA (3.05 g,
23.6 mmol) followed by TBTU (2.08 g, 6.49 mmol) was added.
The mixture was stirred at rt for 5 min before 3-ethyl-N,4-
161.5, 157.3, 137.7, 133.7, 131.7, 128.4, 126.7, 122.6, 113.8, 106.0,
74.0, 70.3, 62.3, 54.0, 47.0, 42.3, 28.8, 23.0, 22.5, 16.5, 14.9; LC-
HRMS: tR ¼ 1.50 min, [M þ H]/z ¼ 499.2556 (C26H35N4O6),
found ¼ 499.2560.
dihydroxy-5-methylbenzimidamide
8
[93,96] (1.38 g,
7.08 mmol) dissolved in DMF (5 mL) was added. Stirring was
continued at rt for 1 h. The mixture was diluted with DCM
(100 mL) and washed with sat. aq. NaHCO3 solution (100 mL)
and water (100 mL). The organic extract was dried over
MgSO4, filtered and concentrated. The residue was dissolved
in dioxane (50 mL) and the solution was stirred at 100 ꢄC for
7 h. The mixture was concentrated and the crude product
was purified by CC on silica gel eluting with heptane:EA 4:1
to give 2-ethyl-4-(5-(2-isobutyl-6-methoxypyridin-4-yl)-
1,2,4-oxadiazol-3-yl)-6-methylphenol (1.38 g, 64%) as a pale
yellow solid; LCeMS: tR ¼ 1.18 min; [M þ 1]þ ¼ 368.17; 1H
4.2. In vitro potency assessment
GTP
gS binding assays were performed in 96 well polypropylene
microtiter plates in a final volume of 200
tions of CHO cells expressing recombinant human S1P1 or S1P3
receptors were used. Assay conditions were 20 mM Hepes, pH 7.4,
m
L. Membrane prepara-
100 mM NaCl, 5 mM MgCl2, 0.1% fatty acid free BSA, 1 or 3
(for S1PR1 or S1PR3, respectively), 2.5% DMSO and 50 pM 35S-
GTP S. Test compounds were dissolved and diluted and pre-
incubated with the membranes, in the absence of 35S-GTP
S, in
150 L assay buffer at room temperature for 30 min. After addition
of 50 S in assay buffer, the reaction mixture was
L of 35S-GTP
mM GDP
g
g
NMR (DMSO-d6):
d
7.68 (s, 2H), 7.65 (d, J ¼ 2.3 Hz,1H), 7.12 (d,
m
J ¼ 2.3 Hz, 1H), 3.94 (s, 3H), 2.66 (q, J ¼ 7.3 Hz, 2H), 2.25 (s,
3H), 2.06e2.19 (m, 1H), 1.17 (t, J ¼ 7.3 Hz, 3H), 0.91 (d,
J ¼ 6.7 Hz, 6H).
m
g
incubated for 1 h at room temperature. The assay was terminated
by filtration of the reaction mixture through a Multiscreen GF/C
plate, pre-wetted 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. The
filterplates were then washed with ice-cold 10 mM Na2HPO4/
NaH2PO4 (70%/30%, w/w) containing 0.1% fatty acid free BSA. Then
(b) To a solution of 2-ethyl-4-(5-(2-isobutyl-6-methoxypyridin-
4-yl)-1,2,4-oxadiazol-3-yl)-6-methylphenol
(100
mol), (R)-glycidol (30 mg,
mol, as 40%
mg,
272
408
mmol), PPh3 (107 mg, 408 m
mmol) in THF (5 mL), DEAD (178 mg, 408
m
the plates were dried, sealed, 25
membrane-bound 35S-GTP
S was determined on the TopCount.
Specific 35S-GTP
S binding was determined by subtracting non-
specific binding (the signal obtained in the absence of agonist)
mL MicroScint20 was added, and
solution in toluene) was added at 5 ꢄC. The mixture was
stirred at rt for 3 h before another portion of (R)-glycidol
g
g
(30 mg, 408
mmol), PPh3 (107 mg, 408 mmol) and DEAD

M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
339
from maximal binding (the signal obtained with 10
m
M S1P). The
concentration of 5% of DMSO was administered to the animals by
gavage.
EC50 of a test compound is the concentration of a compound
inducing 50% of specific binding. Data (EC50) are given as geometric
means.
4.5. Pharmacokinetics in the rat
4.3. Non-GLP 3T3 neutral red uptake phototoxicity assay [76,77]
Blood samples were obtained from the above in vivo efficacy
experiments in vials containing EDTA as anticoagulant and plasma
was prepared by centrifugation at 3000 g for 10 min at 4 ꢄC. Plasma
samples were analyzed using liquid chromatography coupled to
mass spectrometry (LCeMS/MS) after protein precipitation. Phar-
macokinetic parameters were estimated with the Phoenix Win-
Nonlin 6.3 software package (Pharsight) using non-compartmental
analysis.
This test has a good predictability and identifies compounds
which act as phototoxic or as photoirritants in vivo after systemic or
topical application respectively. The 3T3 Neutral Red Uptake (NRU)
Phototoxicity assay is a 96-well cytotoxicity-based assay that uti-
lizes normal BALB/c 3T3 mouse fibroblasts to measure the
concentration-dependent reduction in neutral red (a vital dye)
uptake by the cells one day after treatment with the chemical either
in the presence or absence of a non-cytotoxic dose of UVA radiation.
Balb/c 3T3 cells, clone 31 (ECACC #86110401; C207 in BioToolBox)
were cultured overnight in 96-well microtiter plates with a density
4.6. Physiology based pharmacokinetic modelling
Chemical structures were loaded into GastroPlus™ v. 7 (Simu-
lations Plus Inc.) and Vss was predicted using the PBPKPlus module,
using the default 250 g rat physiology, based on log D and plasma
protein binding and blood-to-plasma partitioning, predicted by the
ADMET predictor module.
of 1 ꢃ 104 cells/100
mL in Dulbecco's modified Eagle's medium
(DMEM, Invitrogen #21969) containing 10% new born calf serum
(NBCS, PAA #B15-102), 4 mM
100 U/ml penicillin and 100
L
-glutamine (Invitrogen #25030),
g/mL streptomycin (Invitrogen
m
#15140). The next day, the medium is removed and the cells were
washed with phosphate buffered saline containing CaCl2 and MgCl2
(PBSþþ) (Gibco by life technologies™, #14080). Duplicate 96-well
monolayers of 3T3 fibroblasts are exposed to eight different con-
Conflict of interest
The authors declare no competing financial interest.
Acknowledgements
centrations (0.05e100
mM in DMSO) of chemical with a pre-
incubation time of 1 h (at 37 ꢄC, 5% CO2). One of the plates is
exposed to 4 J/cm2 UVA (1.7 mW/cm2 for 40 min) almost devoid of
UVB (0% transmission at 300 nm, 2.2% at 310 nm, 17.3% at 320 nm,
44.8% at 330 nm, 67.8% at 340 nm, 80% at 350 nm) using the UV-sun
simulator UVACUBE 400 equipped with a SOL500 lamp and an H1-
The authors gratefully acknowledge their colleagues and co-
workers Lucy Baumann, Maxime Boucher, Celine Bortolamiol,
ꢀ
ꢀ
Stephane Delahaye, Alexandre Flock, Elvire Fournier, Julien Gri-
€
Filter (H1 Filter and frame: Cat. Number 4730, Dr. Honle UV Tech-
mont, Hakim Hadana, Julie Hoerner, Benedikt Hofstetter, Matthias
Merrettig, Christine Metzger, Sylvie Poirey, Michel Rauser, Manuela
nology) while the other plate is kept in the dark (-UVA). After UV
exposure, PBSþþ was replaced with culture medium and cell
viability is assessed after 24 h with plate-reader at 540 nm. The
Neutral Red Uptake (NRU) by cells exposed to the test chemical in
the presence of UVA exposure (þUVA) is compared to the NRU by
cells exposed to the test chemical in the absence of UVA exposure
(ꢀUVA). Dose-response curves were constructed for each experi-
ment and the EC50 (þUVA) and EC50 (ꢀUVA) values are calculated,
i.e. the effective concentration inhibiting cell viability by 50% of
untreated controls in the presence and absence of UVA exposure.
Chlorpromazine (SigmaeAldrich #C0982-5G) served as the posi-
tive control. A Photo Irritancy Factor (PIF) is then calculated as the
ratio of toxicity for each tested chemical with and without UVA
according to the formula PIF ¼ EC50 (ꢀUVA)/EC50 (þUVA). In our
case, a PIF >2.5 was considered to be predictive for phototoxic
potential. EC50 values were calculated geometric means.
€
Schlapfer, Jürgen Seifert, Virginie Sippel, Gaby von Aesch, Aude
Weigel, and Rolf Wüst for the excellent work done and Martine
Clozel for support.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2016.03.020.
References
[1] V.A. Blaho, T. Hla, An update on the biology of sphingosine 1-phosphate re-
ceptors, J. Lipid. Res. 55 (2014) 1596e1608.
[2] T. Hla, V. Brinkmann, Sphingosine 1-phosphate (S1P): physiology and the
effects of S1P receptor modulation, Neurology 76 (2011) S3eS8.
[3] S.E. Alvarez, S. Milstien, S. Spiegel, Autocrine and paracrine roles of sphingo-
sine-1-phosphate, Trends Endocrinol. Metab. 18 (2007) 300e307.
[4] V. Brinkmann, Sphingosine 1-phosphate receptors in health and disease:
mechanistic insights from gene deletion studies and reverse pharmacology,
Pharmacol. Ther. 115 (2007) 84e105.
4.4. In vivo efficacy
In vivo efficacy of the target compounds was assessed by
measuring the circulating lymphocytes after oral administration of
10 mg/kg of a target compound to male Wistar rats (RccHan:WIST)
obtained from Harlan Laboratories (Venray, the Netherlands). 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 under isoflurane anaesthesia
by sublingual vein puncture before and 3, 6 and 24 h after drug
administration. Full blood was subjected to haematology using
Beckman Coulter Ac$T 5diff CP (Beckman Coulter International SA,
Nyon, Switzerland). All data are presented as mean of 6e8 animals.
For formulation, the compounds were dissolved in DMSO. This
solution was added to a stirred solution of succinylated gelatine
(7.5%) in water. The resulting milky suspension containing a final
[5] M. Bigaud, D. Guerini, A. Billich, F. Bassilana, V. Brinkmann, Second generation
S1P pathway modulators: research strategies and clinical developments,
Biochim. Biophys. Acta 1841 (2014) 745e758.
[6] C. Bode, S.-C. Sensken, U. Peest, G. Beutel, F. Thol, B. Levkau, Z. Li, R. Bittman,
T. Huang, M. Toelle, M.v.d. Giet, M.H. Graeler, Erythrocytes serve as a reservoir
for cellular and extracellular sphingosine 1-phosphate, J. Cell. Biochem. 109
(2010) 1232e1243.
[7] H. Obinata, T. Hla, Sphingosine 1-phosphate in coagulation and inflammation,
Semin. Immunopathol. 34 (2012) 73e91.
[8] A.J. Morris, S. Selim, A. Salous, S.S. Smyth, Blood relatives: dynamic regulation
of bioactive lysophosphatidic acid and sphingosine-1-phosphate metabolism
in the circulation, Trends Cardiovasc. Med. 19 (2009) 135e140.
[9] T. Hla, K. Venkataraman, J. Michaud, The vascular S1P gradient e cellular
sources and biological significance, Biochim. Biophys. Acta 1781 (2008)
477e482.
ꢀ
[10] H. LeStunff, P. Giussani, M. Maceyka, S. Lepine, S. Milstien, S. Spiegel, Recycling
of sphingosine is regulated by the concerted actions of sphingosine-1-
phosphate phosphohydrolase 1 and sphingosine kinase 2, J. Biol. Chem. 282

340
M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
(2007) 34372e34380.
1-phosphate (S1P) displays sustained S1P receptor agonism and signaling
through S1P lyase-dependent receptor recycling, Cell Signal. 26 (2014)
1576e1588.
[11] T. Baumruker, A. Billich, Sphingolipids in cell signalling: their function as
receptor ligands, second messengers, and raft constituents, Curr. Immunol.
Rev. 2 (2006) 101e118.
[12] T. Okada, T. Kajimoto, S. Jahangeer, S. Nakamura, Sphingosine kinase/sphin-
gosine 1-phosphate signalling in central nervous system, Cell. Signal. 21
(2009) 7e13.
[13] S. Milstien, D. Gude, S. Spiegel, Sphingosine 1-phosphate in neural signalling
and function, Acta Paed. 96 (2007) 40e43.
[14] V. Brinkmann, T. Baumruker, Pulmonary and vascular pharmacology of
sphingosine 1-phosphate, Curr. Opin. Pharmacol. 6 (2006) 244e250.
[15] S. Uhlig, E. Gulbins, Sphingolipids in the lungs, Am. J. Respir. Crit. Care Med.
178 (2008) 1100e1114.
[16] A.E. Alewijnse, S.L.M. Peters, Sphingolipid signalling in the cardiovascular
system: good, bad or both? Eur. J. Pharmacol. 585 (2008) 292e302.
[17] H. Rosen, M.G. Sanna, S.M. Cahalan, P.J. Gonzalez-Cabrera, Tipping the gate-
keeper: S1P regulation of endothelial barrier function, Trends Immunol. 28
(2007) 102e107.
[18] B.J. McVerry, J.G.N. Garcia, In vitro and in vivo modulation of vascular barrier
integrity by sphingosine 1-phosphate: mechanistic insights, Cell. Signal. 17
(2005) 131e139.
[42] F. Mullershausen, F. Zecri, C. Cetin, A. Billich, D. Guerini, K. Seuwen, Persistent
signaling induced by FTY720-phosphate is mediated by internalized S1P1
receptors, Nat. Chem. Biol. 5 (2009) 428e434.
[43] K. Chiba, FTY720, a new class of immunomodulator, inhibits lymphocyte
egress from secondary lymphoid tissues and thymus by agonistic activity at
sphingosine 1-phosphate receptors, Pharmacol. Ther. 108 (2005) 308e319.
[44] E. Jo, M.G. Sanna, P.J. Gonzalez-Cabrera, S. Thangada, G. Tigyi, D.A. Osborne,
T. Hla, A.L. Parrill, H. Rosen, S1P1-selective in vivo-active agonists from high-
throughput screening: off-the-shelf chemical probes of receptor interactions,
signaling, and fate, Chem. Biol. 12 (2005) 703e715.
€
[45] M.H. Graler, E.J. Goetzl, The immunosuppressant FTY720 down-regulates
sphingosine 1-phosphate G-protein-coupled receptors, FASEB J. 18 (2004)
551e553.
[46] S.M. Cahalan, P.J. Gonzalez-Cabrera, N. Nguyen, M. Guerrero, E.A. Cisar,
N.B. Leaf, S.J. Brown, E. Roberts, H. Rosen, Sphingosine 1-phosphate receptor 1
(S1P(1)) upregulation and amelioration of experimental autoimmune
encephalomyelitis by an S1P(1) antagonist, Mol. Pharmacol. 83 (2013)
316e321.
[19] M. Ishii, J.G. Egen, F. Klauschen, M. Meier-Schellersheim, Y. Saeki, J. Vacher,
R.L. Proia, R.N. Germain, Sphingosine-1-phosphate mobilizes osteoclast pre-
cursors and regulates bone homeostasis, Nature 458 (2009) 524e528.
[20] M.L. Allende, G. Tuymetova, B.G. Lee, E. Bonifacino, Y.-P. Wu, R.L. Proia, S1P1
receptor directs the release of immature B cells from bone marrow into blood,
J. Exp. Med. 207 (2010) 1113e1124.
[21] M.M. Price, D. Kapitonov, J. Allegood, S. Milstien, C.A. Oskeritzian, S. Spiegel,
Sphingosine-1-phosphate induces development of functionally mature
chymase-expressing human mast cells from hematopoietic progenitors,
FASEB J. 23 (2009) 3506e3515.
[47] J. Quancard, B. Bollbuck, P. Janser, D. Angst, F. Berst, P. Buehlmayer, M. Streiff,
C. Beerli, V. Brinkmann, D. Guerini, P.A. Smith, T.J. Seabrook, M. Traebert,
K. Seuwen, R. Hersperger, C. Bruns, F. Bassilana, M. Bigaud, A potent and se-
lective S1P1 antagonist with efficacy in experimental autoimmune encepha-
lomyelitis, Chem. Biol. 19 (2012) 1142e1151.
[48] D. Angst, P. Janser, J. Quancard, P. Buehlmayer, F. Berst, L. Oberer, C. Beerli,
M. Streiff, C. Pally, R. Hersperger, C. Bruns, F. Bassilana, B. Bollbuck, An oral
sphingosine 1-phosphate receptor 1 (S1P1) antagonist prodrug with efficacy
in vivo: discovery, synthesis, and evaluation, J. Med. Chem. 55 (2012)
9722e9734.
[22] J. Rivera, R.L. Proia, A. Olivera, The alliance of sphingosine-1-phosphate and its
receptors in immunity, Nat. Rev. Immunol. 8 (2008) 753e763.
[23] C.A. Oskeritzian, S. Milstien, S. Spiegel, Sphingosine-1-phosphate in allergic
responses, asthma and anaphylaxis, Pharmacol. Ther. 115 (2007) 390e399.
[24] V. Brinkmann, M.D. Davis, C.E. Heise, R. Albert, S. Cottens, R. Hof, C. Bruns,
E. Prieschl, T. Baumruker, P. Hiestand, C.A. Foster, M. Zollinger, K.R. Lynch, The
immune modulator FTY720 targets sphingosine 1-Phosphate receptors, J. Biol.
Chem. 277 (2002) 21453e21457.
[25] S. Mandala, R. Hajdu, J. Bergstrom, E. Quackenbush, J. Xie, J. Milligan,
R. Thornton, G.-J. Shei, D. Card, C.A. Keohane, M. Rosenbach, J. Hale, C.L. Lynch,
K. Rupprecht, W. Parsons, H. Rosen, Alteration of lymphocyte trafficking by
sphingosine-1-phosphate receptor agonists, Science 296 (2002) 346e349.
[26] D.D. Pinschewer, A.F. Ochsenbein, B. Odermatt, V. Brinkmann, H. Hengartner,
R.M. Zinkernagel, FTY720 immunosupression impairs effector T cell periph-
eral homing without affecting induction, expansion, and memory, J. Immunol.
164 (2000) 5761e5770.
[49] Y. Fujii, H. Ohtake, N. Ono, T. Hara, T. Sakurai, S. Takahashi, T. Takayama,
Y. Fukasawa, F. Shiozawa, N. Tsukahara, T. Hirayama, Y. Igarashi, R. Goitsuka,
Lymphopenia induced by a novel selective S1P1 antagonist structurally un-
related to S1P, Biochim. Biophys. Acta 1821 (2012) 600e606.
ꢀ
ꢁ
[50] G. Tarrason, M. Aulí, S. Mustafa, V. Dolgachev, M.T. Domenech, N. Prats,
ꢀ
M. Domínguez, R. Lopez, N. Auilar, M. Calbet, M. Pont, G. Milligan, S.L. Kunkel,
N. Godessart, The sphingosine-1-phosphate receptor-1 antagonist, W146,
causes early and short-lasting peripheral blood lymphophenia in mice, Int.
Immunopharmacol. 11 (2011) 1773e1779.
[51] S. Weiler, N. Braendlin, C. Beerli, C. Bergsdorf, A. Schubart, H. Srinivas,
B. Oberhauser, A. Billich, Orally active 7-substituted (4-benzyl-phthalazin-1-
yl)-2-methyl-piperazin-1-yl]-nicotinonitriles as active-site inhibitors of
sphingosine-1-phosphate lyase for the treatment of multiple sclerosis, J. Med.
Chem. 57 (2014) 5074e5084.
[52] M. Mori, S. Kuwabara, Are more sphingosine 1-phosphate receptor agonists a
better therapeutic option against multiple sclerosis? J. Neurol. Neurosurg.
Psychiatry 85 (2014) 1180.
[27] V. Brinkmann, S. Chen, L. Feng, D. Pinschewer, Z. Nikolova, R. Hof, FTY720
alters lymphocyte homing and protects allografts without inducing general
immunosuppression, Transplant. Proc. 33 (2001) 530e531.
[53] H. Obinata, T. Hla, Fine-tuning S1P therapeutics, Chem. Biol. 19 (2012)
1080e1082.
[28] J.G. Cyster, S.R. Schwab, Sphingosine-1-phosphate and lymphocyte egress
from lymphoid organs, Ann. Rev. Immunol. 30 (2012) 69e94.
[29] V. Brinkmann, K.R. Lynch, FTY720: targeting G-protein-coupled receptors for
sphingosine 1-phosphate in transplantation and autoimmunity, Curr. Opin.
Immunol. 14 (2002) 569e575.
[30] D.M. Wingerchuk, J.L. Carter, Multiple sclerosis: current and emerging
disease-modifying therapies and treatment strategies, Mayo Clin. Proc. 89
(2014) 225e240.
[54] M.L. Allende, J.L. Dreier, S. Mandala, R.L. Proia, Expression of the sphingosine
1-phosphate receptor, S1P1, on T-cells controls thymic emigration, J. Biol.
Chem. 279 (2004) 15396e15401.
[55] M. Matloubian, C.G. Lo, G. Cinamon, M.J. Lesneski, Y. Xu, V. Brinkmann,
M.L. Allende, R.L. Proia, J.G. Cyster, Lymphocyte egress from thymus and pe-
ripheral lymphoid organs is dependent on S1P receptor 1, Nature 427 (2004)
355e360.
[56] S.H. Wei, H. Rosen, M.P. Matheu, M.G. Sanna, S.-K. Wang, E. Jo, C.-H. Wong,
I. Parker, M.D. Cahalan, Sphingosine 1-phosphate type 1 receptor agonism
inhibits transendothelial migration of medullary T cells to lymphatic sinuses,
Nat. Immunol. 6 (2005) 1228e1235.
[31] A. Groves, Y. Kihara, J. Chun, Fingolimod: direct CNS effects of sphingosine 1-
phosphate (S1P) receptor modulation and implications in multiple sclerosis
therapy, J. Neurol. Sci. 328 (2013) 9e18.
[32] J. Ingwersen, O. Aktas, P. Kuery, B. Kieseier, A. Boyko, H.-P. Hartung, Fingoli-
mod in multiple sclerosis: mechanisms of action and clinical efficacy, Clin.
Immunol. 142 (2012) 15e24.
[33] J.A. Cohen, J. Chun, Mechanisms of fingolimod's efficacy and adverse effects in
multiple sclerosis, Ann. Neurol. 69 (2011) 759e777.
[57] J.J. Hale, W. Neway, S.G. Mills, R. Hajdu, C.A. Keohane, M. Rosenbach,
J. Milligan, G.-J. Shei, G. Chrebet, J. Bergstrom, D. Card, G.C. Koo, S.L. Koprak,
J.J. Jackson, H. Rosen, S. Mandala, Potent S1P receptor agonists replicate the
pharmacologic actions of the novel immune modulator FTY720, Bioorg. Med.
Chem. Lett. 14 (2004) 3351e3355.
[34] L. Kappos, E.-W. Radue, P. O'Connor, C. Polman, R. Hohlfeld, P. Calabresi,
K. Selmaj, C. Agoropoulou, M. Leyk, L. Zhang-Auberson, P. Burtin, A placebo-
controlled trial of oral fingolimod in relapsing multiple sclerosis, N. Engl. J.
Med. 362 (2010) 387e401.
[58] M.G. Sanna, J. Liao, E. Jo, C. Alfonso, M.-Y. Ahn, M.S. Peterson, B. Webb,
S. Lefebvre, J. Chun, N. Gray, H. Rosen, Sphingosine 1-Phosphate (S1P) receptor
subtypes S1P1 and S1P3, respectively, regulate lymphocyte recirculation and
heart rate, J. Biol. Chem. 279 (2004) 13839e13848.
[35] A. Vaclavkova, S. Chimenti, P. Arenberger, P. Hollo, P.G. Sator, M. Burcklen,
M. Stefani, D. D'Ambrosio, Oral ponesimod in patients with chronic plaque
psoriasis: a randomised, double-blind, placebo-controlled phase 2 trial, Lancet
384 (2014) 2036e2045.
[59] N. Seki, Y. Maeda, H. Kataoka, K. Sugahara, T. Sugita, K. Chiba, Fingolimod
(FTY720) ameliorates experimental autoimmune encephalomyelitis (EAE): II.
FTY720 decreases infiltration of Th17 and Th1 cells into the central nervous
system in EAE, Inflamm. Regen. 30 (2010) 542e548.
[36] ClinicalTrials.gov, Efficacy and Tolerability of BAF312 in Patients with Poly-
myositis and Dermatomyositis, 2012.
[37] ClinicalTrials.gov, Efficacy and Safety Study of RPC1063 in Ulcerative Colitis,
2012.
[60] H. Kataoka, K. Sugahara, K. Shimano, K. Teshima, M. Koyama, A. Fukunari,
K. Chiba, FTY720, sphingosine 1-phosphate receptor modulator, ameliorates
experimental autoimmune encephalomyelitis by inhibition of T cell infiltra-
tion, Cell. Mol. Immunol. 2 (2005) 439e448.
[38] ClinicalTrials.gov, Safety and Efficacy of KRP203 in Subacute Cutaneous Lupus
Erythematosus, 2012.
[39] ClinicalTrials.gov, Efficacy & Safety in Moderately Active Refractory Ulcerative
Colitis Patients, 2012.
[61] M. Webb, C.S. Tham, F.F. Lin, K. Lariosa-Willingham, N. Yu, J. Hale, S. Mandala,
J. Chun, T.S. Rao, Sphingosine 1-phosphate receptor agonists attenuate
relapsing-remitting experimental autoimmune encephalitis in SJL mice,
J. Neuroimmunol. 153 (2004) 108e121.
[40] ClinicalTrials.gov, Efficacy, Tolerability, Pharmacokinetcs and Pharmacody-
namics Study of MT-1303 in Subjects with Inflammatory Bowel Disease, 2012.
[41] J. Gatfield, L. Monnier, R. Studer, M.H. Bolli, B. Steiner, O. Nayler, Sphingosine-
[62] P.J. Gonzalez-Cabrera, S.M. Cahalan, N. Nguyen, G. Sarkisyan, N.B. Leaf,
M.D. Cameron, T. Kago, H. Rosen, S1P1 receptor modulation with cyclical
recovery from lymphopenia amelioreates mouse model of multiple sclerosis,

M.H. Bolli et al. / European Journal of Medicinal Chemistry 115 (2016) 326e341
341
Mol. Pharmacol. 81 (2012) 166e174.
validation of a new concentration response analysis software and biostatis-
tical analyses related to the use of various prediction models, Altern. Lab.
Anim. 30 (2002) 415e432.
[63] J.W. Choi, S.E. Gardell, D.R. Herr, R. Rivera, C.-W. Lee, K. Noguchi, S.T. Teo,
Y.C. Yung, M. Lu, G. Kennedy, J. Chun, FTY720 (fingolimod) efficacy in an
animal model of multiple sclerosis requires astrocyte sphingosine 1-
phosphate receptor 1 (S1P1) modulation, Proc. Natl. Acad. Sci. U. S. A. 108
(2011) 751e756.
[78] A.M. Lynch, P. Wilcox, Review of the performance of the 3T3 NRU in vitro
phototoxicity assay in the pharmaceutical industry, Exp. Toxicol. Pathol. 63
(2011) 209e214.
[64] B. Balatoni, M.K. Storch, E.M. Swoboda, V. Schonborn, A. Koziel, G.N. Lambrou,
P.C. Hiestand, R. Weissert, C.A. Foster, FTY720 sustains and restores neuronal
function in the DA rat model of MOG-induced experimental autoimmune
encephalomyelitis, Brain Res. Bull. 74 (2007) 307e316.
[79] M. Bolli, D. Lehmann, B. Mathys, C. Mueller, O. Nayler, B. Steiner, J. Velker, in:
Pyridine-3-yl Derivatives as Immunomodulating Agents, Actelion Pharma-
ceuticals Ltd., 2008.
[80] Food and Drug Administration, in: S10 Photosafety Evaluation of Pharma-
ceuticals, Center for Drug Evaluation and Research (CDER), 2012.
[81] European Medicine Agency, in: ICH Guideline S10 Guidance on Photosafety
Evaluation of Pharmaceuticals, Committee for Medicinal Products for Human
Use, 2012.
[65] V.E. Miron, S.K. Ludwin, P.J. Darlington, A.A. Jarjour, B. Soliven, T.E. Kennedy,
J.P. Antel, Fingolimod (FTY720) enhances remyelination following demyelin-
ation of organotypic cerebellar slices, Am. J. Pathol. 176 (2010) 2682e2694.
[66] A. Murakami, H. Taksugi, S. Ohnuma, Y. Koide, A. Sakurai, S. Takeda,
T. Hasegawa, J. Sasamori, T. Konno, K. Hayashi, Y. Watanabe, K. Mori, Y. Sato,
A. Takahashi, N. Mochizuki, N. Takakura, Sphingosine 1-phosphate (S1P)
regulates vascular contraction via S1P3 receptor: investigation based on a new
S1P3 receptor antagonist, Mol. Pharmacol. 77 (2010) 704e713.
[67] J.J. Hale, G. Doherty, L. Toth, S.G. Mills, R. Hajdu, C.A. Keohane, M. Rosenbach,
J. Milligan, G.-J. Shei, G. Chrebet, J. Bergstrom, D. Card, M. Forrest, S.-Y. Sun,
S. West, H. Xie, N. Nomura, H. Rosen, S. Mandala, Selecting against S1P3 en-
hances the acute cardiovascular tolerability of 3-(N-benzyl)amino-
propylphosphonic acid S1P receptor agonists, Bioorg. Med. Chem. Lett. 14
(2004) 3501e3505.
[68] M. Forrest, S.-Y. Sun, R. Hajdu, J. Bergstrom, D. Card, G. Doherty, J. Hale,
C. Keohane, C. Meyers, J. Milligan, S. Mills, N. Nomura, H. Rosen, M. Rosenbach,
G.-J. Shei, I.I. Singer, M. Tian, S. West, V. White, J. Xie, R.L. Proia, S. Mandala,
Immune cell regulation and cardiovascular effects of sphingosine 1-phosphate
receptor agonists in rodents are mediated via distinct receptor subtypes,
J. Pharmacol. Exp. Ther. 309 (2004) 758e768.
[69] S. Salomone, E.M. Potts, S. Tyndall, P.C. Ip, J. Chun, V. Brinkmann, C. Waeber,
Analysis of sphingosine 1-phosphate receptors involved in constriction of
isolated cerebral arteries with receptor null mice and pharmacological tools,
Brit. J. Pharmacol. 153 (2008) 140e147.
[70] R.M. Fryer, A. Muthukumarana, P.C. Harrison, S. Nodop Mazurek, R.R. Chen,
K.E. Harrington, R.M. Dinallo, J.C. Horan, L. Patnaude, L.K. Modis, G.A. Reinhart,
The clinically-tested S1P receptor agonists, FTY720 and BAF312, demonstrate
subtype-specific bradycardia (S1P(1)) and hypertension (S1P(3)) in rat, PLoS
One 7 (2012) e52985.
[82] Y. Haranosono, M. Kurata, H. Sakaki, Establishment of an in silico phototox-
icity prediction method by combining descriptors related to photo-absorption
and photo-reaction, J. Toxicol. Sci. 39 (2014) 655e664.
[83] A. Fürstner, A. Leitner, Iron-catalyzed cross-coupling reactions of alkyl-
Grignard reagents with aryl chloride, tosylates and triflates, Angew. Chem.
Int. Ed. 41 (2002) 609e612.
ꢀ
[84] A. Fürstner, A. Leitner, M. Mendez, H. Krause, Iron-catalyzed cross coupling
reactions, J. Am. Chem. Soc. 124 (2002) 13856e13863.
[85] H. Matsushita, E. Negishi, Palladium-catalyzed reactions of allylic electrophiles
with organometallic reagents. A regioselective 1,4-elimination and a regio-
and stereoselective reduction of allylic derivatives, J. Org. Chem. 47 (1982)
4161e4165.
[86] J.A. Casares, P. Espinet, B. Fuentes, G. Salas, Insights into the mechanism of the
Negishi reaction: ZnRX versus ZnR2 reagents, J. Am. Chem. Soc. 129 (2007)
3508e3509.
[87] C.C.C. Johansson Seechurn, M.O. Kitching, T.J. Colacot, V. Snieckus, Palladium-
catalyzed cross-coupling: a historical contextual perspective to the 2010
nobel prize, Angew. Chem. Int. Ed. Engl. 51 (2012) 5062e5085.
[88] H. Doucet, SuzukieMiyaura cross-coupling reactions of alkylboronic acid de-
rivatives or alkyltrifluoroborates with aryl, alkenyl or alkyl halides and tri-
flates, Eur. J. Org. Chem. 2008 (2008) 2013e2030.
[89] J.T. Binder, C.J. Cordier, G.C. Fu, Catalytic enantioselective cross-couplings of
secondary alkyl electrophiles with secondary alkylmetal nucleophiles:
Negishi reactions of racemic benzylic bromides with achiral alkylzinc re-
agents, J. Am. Chem. Soc. 134 (2012) 17003e17006.
[71] K. Sobel, K. Menyhart, N. Killer, B. Renault, Y. Bauer, R. Studer, B. Steiner,
M.H. Bolli, O. Nayler, J. Gatfield, Sphingosine-1-phosphate (S1P) receptor ag-
onists mediate pro-fibrotic responses in normal human lung fibroblasts via
S1P2 and S1P3 receptors and Smad-independent signaling, J. Biol. Chem. 288
(2013) 14839e14851.
[72] M. Urbano, M. Guerrero, H. Rosen, E. Roberts, Modulators of the sphingosine
1-phosphate receptor 1, Bioorg. Med. Chem. Lett. 23 (2013) 6377e6389.
[73] E. Roberts, M. Guerrero, M. Urbano, H. Rosen, Sphingosine 1-phosphate re-
ceptor agonists: a patent review (2010e2012), Exp. Opin. Ther. Pat. 23 (2013)
817e841.
[74] M.H. Bolli, C. Lescop, O. Nayler, Synthetic sphingosine 1-phosphate receptor
modulators e opportunities and potential pitfalls, Curr. Top. Med. Chem. 11
(2011) 726e757.
[75] M.H. Bolli, S. Abele, M. Birker, R. Bravo, D. Bur, R. de Kanter, C. Kohl, J. Grimont,
P. Hess, C. Lescop, B. Mathys, C. Muller, O. Nayler, M. Rey, M. Scherz,
G. Schmidt, J. Seifert, B. Steiner, J. Velker, T. Weller, Novel S1P1 receptor ag-
onists e part 3: from thiophenes to pyridines, J. Med. Chem. 57 (2014)
110e130.
[90] Z. Liu, N. Dong, M. Xu, Z. Sun, T. Tu, Mild Negishi cross-coupling reactions
catalyzed by acenaphthoimidazolylidene palladium complexes at low catalyst
loadings, J. Org. Chem. 78 (2013) 7436e7444.
[91] C. Han, S.L. Buchwald, Negishi coupling of secondary alkylzinc halides with
aryl bromides and chlorides, J. Am. Chem. Soc. 131 (2009) 7532e7533.
[92] F. Kerins, D.F. O'Shea, Generation of substituted styrenes via Suzuki cross-
coupling of aryl halides with 2,4,6-trivinylcyclotriboroxane, J. Org. Chem. 67
(2002) 4968e4971.
[93] M.H. Bolli, C. Müller, B. Mathys, S. Abele, M. Birker, R. Bravo, D. Bur, P. Hess,
C. Kohl, D. Lehmann, O. Nayler, M. Rey, S. Meyer, M. Scherz, G. Schmidt,
B. Steiner, A. Treiber, J. Velker, T. Weller, Novel S1P1 receptor agonists e part
1: from pyrazoles to thiophenes, J. Med. Chem. 56 (2013) 9737e9755.
[94] M.H. Bolli, J. Velker, C. Müller, B. Mathys, M. Birker, R. Bravo, D. Bur, R. de
Kanter, P. Hess, C. Kohl, D. Lehmann, S. Meyer, O. Nayler, M. Rey, M. Scherz,
B. Steiner, Novel S1P1 receptor agonists e part 2: from bicyclo[3.1.0]hexane
fused thiophenes to isobutyl substituted thiophenes, J. Med. Chem. 57 (2014)
78e97.
[95] M.H. Bolli, S. Abele, C. Binkert, R. Bravo, S. Buchmann, D. Bur, J. Gatfield,
P. Hess, C. Kohl, C. Mangold, B. Mathys, K. Menyhart, C. Müller, O. Nayler,
M. Scherz, G. Schmidt, V. Sippel, B. Steiner, D. Strasser, A. Treiber, T. Weller, 2-
Imino-thiazolidin-4-one derivatives as potent, orally active S1P1 receptor
agonists, J. Med. Chem. 53 (2010) 4198e4211.
[76] H. Spielmann, M. Balls, J. Dupuis, W.J. Pape, G. Pechovitch, O. de Silva,
H.G. Holzhutter, R. Clothier, P. Desolle, F. Gerberick, M. Liebsch, W.W. Lovell,
T. Maurer, U. Pfannenbecker, J.M. Potthast, M. Csato, D. Sladowski, W. Steiling,
P. Brantom, The international EU/COLIPA in vitro phototoxicity validation
study: results of phase II (blind trial). Part 1: the 3T3 NRU phototoxicity test,
Toxicol. In Vitro 12 (1998) 305e327.
[96] G. Schmidt, S. Reber, M.H. Bolli, S. Abele, Practical and scalable synthesis of
S1P1 receptor agonist ACT-209905, Org. Proc. Res. Dev. 16 (2012) 595e604.
[77] B. Peters, H.G. Holzhutter, In vitro phototoxicity testing: development and