830
L. Yan et al. / Bioorg. Med. Chem. Lett. 17 (2007) 828–831
5b formed only in trace amounts using conventional
thermal heating. The mechanism for this discrepancy
is not fully understood. Jeffery-Heck coupling7 of 2-
iodopyridine 9 (from 3-hydroxy-2-methyl-pyridine 8 in
two steps)8 and tert-butyl acrylate followed by hydroge-
nation delivered phenol 10 (Scheme 1b). Triflation of the
resulting phenol and cyanation gave nitrile 11, which
was subsequently converted to amidoxime 12, con-
densed with benzoic acid 13, and deprotected to afford
pyridine analog 14 (position c, Fig. 1). Heck coupling
of bromide 2 and tert-butyl acrylate afforded trans-cin-
namate 16, which was treated with trimethylsulfoxoni-
um iodide9 to yield ( )-trans-cyclopropane 16,
tethering positions a and b (Fig. 1) through a methylene
group (Scheme 1c). Again only under microwave condi-
tions amidoxime 17 was formed in good yield. Com-
pound 17 was readily converted to ( )-trans-
cyclopropanepropionic acids 18 in three steps. Vinyla-
tion10 of bromide 2 and subsequent treatment with
hydroxylamine gave amidoxime 19, which was con-
densed with benzoic acid 13 and oxidized to provide
aldehyde 20 (Scheme 1d). Formation of cis-cinnamate
21 was achieved using the Still modified Horner-Em-
mons olefination in good yield.11 Cyclopropanation of
21 followed by saponification gave ( )-cis-cyclopro-
panepriopionic acid 22.
was substituted by a C3-cyano group instead of a C3-
CF3 group (i.e., 18a vs 18b). This trend was observed
in the 3-arylpropionic acid series as well (i.e., 1a vs
1b). These structural modifications, however, did not
appear to significantly alter the binding selectivity of
S1P1 receptor against either S1P3 receptor or S1P5
receptor. All modified 3-arylpropionic acids in this study
induced PLL in mice at low dosage (<0.7 mpk).
Among pharmacokinetic properties, these modified
3-arylpropionic acids generally showed low clearances,
low volumes of distribution, and good bioavailability
in rat, but their half-lives were profoundly different.
Compound 7a having a methyl group at C2 showed
slight half-life prolongation in comparison to 1a. In con-
trast, 7b with a methyl group at C3 benzylic position had
half-life extended three-fold compared to 1a, suggesting
that the metabolism of 3-arylpropionic acids is more
sensitive to the substitution at C3 than at C2, probably
due to a steric effect. Replacement of the adjacent phenyl
ring with a pyridine group extended half-life more than
two-fold, implicating that such replacement slows the
metabolic oxidation at the C3 benzylic position of 3-
arylpropionic acids. The replacement of the propionic
acid by a cyclopropane carboxylic acid led to substantial
half-life enhancement (about five-fold or more); the
cyclopropane ring is known to be more metabolically
stable. Together, these data indicate that metabolic oxi-
dation of the propionic acid group, especially at the ben-
zylic C3 position, is responsible for the short half-life of
this series of compounds observed in rodent.
Table 1 compares the functional binding affinities (EC50)
of S1P1,3,5 receptors,12 the pharmacodynamic ED50 val-
ues of peripheral lymphocyte lowering (PLL) in mice,13
and the rat pharmacokinetic properties of 1a and 1b to
those of modified 3-arylpropionic acids. Like 1a and 1b,
all modified analogs appeared to be potent, subnanomo-
lar S1P1 full agonists. The selectivity against S1P3 recep-
tor was more pronounced when the pendant phenyl ring
In conclusion, a series of modified 3-arylpropionic acids
has been designed and synthesized as potent and selec-
tive S1P1 agonists. All these analogs induced PLL at
Table 1. S1P functional binding affinities (EC50, nM),a murine peripheral lymphocyte lowering (PLL),b and rat pharmacokinetic datac for selected
S1P receptor agonists
O N
R2
X
N
CO2H
O
R1
Y
Compound R1, R2
X
Y
S1P1 S1P3
110 40
S1P5 PLL ED50 (mg/kg, po) Rat PK
1a
1b
H, H
H, H
–CH– CF3 0.08
–CH– CN 0.08
–CH– CF3 0.3
–CH– CF3 0.12
0.3
0.1
0.1
0.1
0.2
0.2
0.2
0.2
0.3
0.7
Clp = 8.7, Vdss = 1.1, t1/2 = 1.1, %F = 88
1100
400
6.5
9.9
3.9
Clp = 2.3, Vdss = 0.2, t1/2 = 0.7, %F = 53
Clp = 7.0, Vdss = 1.5, t1/2 = 1.7, %F = 82
Clp = 2.9, Vdss = 0.6, t1/2 = 3.1, %F = 100
Clp = 1.2, Vdss = 0.3, t1/2 = 2.4, %F = 93
Clp = 2.4, Vdss = 1.1, t1/2 = 6.3, %F = 88
Clp = 1.0, Vdss = 0.4, t1/2 = 5.9, %F = 90
Clp = 3.4, Vdss = 1.0, t1/2 = 4.9, %F = 84
n.d.
7a
7b
H, CH3
CH3, H
H, H
184
14
–N–
CN 0.44
2760 28.6
123 5.1
>10,000 13.6
2600 5.2
18a
18b
18b
18b
22
–CH2–
–CH2–
–CH– CF3 0.21
–CH– CN 0.45
0,d
–CH2– enant. 1 –CH– CN 0.31
–CH2– enant. 2 –CH– CN 0.93
00,d
2660 88.4
>1000 37.7
–CH2–
–CH– CN 0.8
Clp = 1.2, Vdss = 0.7, t1/2 = 7.4, %F = 89
a Ligand-induced uptake of [35S]-GTPcS on CHO cell membranes expressing human S1P receptors. Data are reported as means for n = 3 mea-
surements. SD were generally within 20% of the average.
b The pharmacodynamic ED50 value is the effective dose of test compound that induces 50% of maximal peripheral lymphocyte lowering (PLL) in
mice.
c Plasma compound concentrations used to calculate pharmacokinetic parameters were obtained after iv administration (1.0 mpk) and po admin-
istration (2.0 mpk) of test compounds to male Sprague–Dawley rats (n = 2), respectively. The units for Clp, Vdss, and t1/2 are mL/min/kg, L/kg, and
hour, respectively.
d Enantiomers 18b0 and 18b00 were prepared by chiral separation of ( )-tert-butyl ester of 18b on Chiralcel OD 4.6 · 25 cm column (eluted with
EtOH/heptane (v:v = 1:9) at a rate of 0.5 mL/min), followed by hydrolysis.