Synthesis of 4(5)-phenylimidazole-based analogues of sphingosine-1- phosphate and FTY720: Discovery of potent S1P1 receptor agonists

Article abstract of DOI:10.1016/j.bmcl.2005.05.097

The novel immunosuppressant FTY720 has been demonstrated to elicit immunomodulating effects via interaction with the G-protein coupled receptor S1P₁. FTY720 induced agonism at the S1P₃ receptor, however, has been shown to result in m

Full text of DOI:10.1016/j.bmcl.2005.05.097

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3568–3572
Synthesis of 4(5)-phenylimidazole-based analogues of
sphingosine-1-phosphate and FTY720: Discovery of potent
S1P1 receptor agonists
Jeremy J. Clemens,a,* Michael D. Davis,b Kevin R. Lynchb,c and Timothy L. Macdonalda
aDepartment of Chemistry, University of Virginia, Charlottesville, VA, USA
bDepartment of Biochemistry and Molecular Biology, University of Virginia, Charlottesville, VA, USA
cDepartment of Pharmacology, University of Virginia, Charlottesville, VA, USA
Received 11 March 2005; revised 13 May 2005; accepted 16 May 2005
Abstract—The novel immunosuppressant FTY720 has been demonstrated to elicit immunomodulating effects via interaction with
the G-protein coupled receptor S1P1. FTY720 induced agonism at the S1P3 receptor, however, has been shown to result in mild
bradycardia, a minor side-effect of initial FTY720 therapy. This report describes the synthesis of several potent 4(5)-phenylimidaz-
ole-based S1P1 receptor agonists that are accompanied by poor agonist activity at S1P3. For instance, compound 20 displayed an
EC50 = 4.7 1.3 nM at the S1P1 receptor and EC50 = 780 1.3 nM at the S1P3 receptor using a [c-35S]GTP-binding assay as com-
pared to phospho-FTY720 (S1P1: EC50 = 1.3 1.3 nM, S1P3: EC50 = 2.0 2.4 nM).
Ó 2005 Elsevier Ltd. All rights reserved.
The development of FTY720 for treatment against or-
gan transplant rejection has generated a great deal of
interest in the discovery of similar immunosuppressive
agents.1 The active metabolite of FTY720, phospho-
FTY720, acts as a potent agonist at four of the five
sphingosine-1-phosphate (S1P) receptors, a family of
G-protein coupled receptors whose natural ligand is
S1P.2 The recent discovery that phospho-FTY720 elicits
immunomodulatory effects via interaction with the S1P1
receptor has made that receptor a target for the synthe-
Figure 1. Structures and regions of S1P and phospho-FTY720.
sis of selective agonists.3 Agonism at the S1P3 receptor
has been shown to induce undesired cardiovascular ef-
fects including mild bradycardia, a minor side-effect of
initial FTY720 therapy.4 Therefore, the development
of S1P/FTY720 analogues with greater selectivity for
a 4(5)-phenylimidazole functionality in the linker re-
S1P1 relative to S1P3 is desired in the investigation for
have synthesized a class of analogues incorporating
gion. This class of S1P receptor agonists essentially
constitutes the insertion of a single carbon–carbon
new immunomodulators (see Fig. 1).
bond spacer into the benzimidazole functionality of
In an effort to further our understanding of the SAR
of S1P and FTY720 with respect to the S1P receptors
our previously reported S1P4 selective agonists.5a
The 4(5)-phenylimidazole class of S1P/FTY720 ana-
and to develop possible therapeutic candidates, we
logues was found to possess potent agonism at S1P1
accompanied by a relatively low affinity for the S1P3
receptor. In this class of agonists, we explored the ef-
fects of stereochemistry at the C2 amino group, meth-
ylation at the C2 position, and the use of a variety of
head groups.
Keywords: S1P receptor agonists; FTY720; S1P1; Bradycardia; Phos-
phothioate; Sphingosine-1-phosphate.
*
Corresponding author. Tel.: +1 310 825 0549; fax: +1 310 206
4038; e-mail: clemens@chem.ucla.edu
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.05.097

J. J. Clemens et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3568–3572
3569
Preparationofthe4(5)-phenylimidazole-basedS1Precep-
tor agonists commenced with the synthesis of the (2S)
40(50)-phenylimidazole compounds 6 and 7 (Scheme 1).
After treatment of N-Boc-(D)-serine(Bzl)-OH with
cesium carbonate, the cesium salt was allowed to
displace the bromine of compound 1, generated by bro-
mination of 40-iodoacetophenone, to give an intermedi-
ate ketoester that was cyclized to the imidazole 2 on
treatment with ammonium acetate and azeotropic
removal of water.6 Compound 2 was then subjected to
a Sonogashira coupling to 1-octyne generating the aryl
alkyne compound 3. Compound 3 then underwent
chemoselective, simultaneous reduction of the aryl al-
kyne, as well as removal of the benzyl ether group under
Birch reduction conditions to give the free alcohol 4
without affecting the 40(50)-phenylimidazole functional-
ity. Alcohol 4 was next phosphorylated with subsequent
oxidation by hydrogen peroxide to give the protected
phosphate 5.7 Global deprotection of compound 5 pro-
vided the phosphate 6 as the TFA salt. Alcohol 7 was
obtained as a TFA salt from 4 on removal of the
N-Boc-protecting group.
treatment with cesium carbonate, the cesium salt of
N-Boc-serine(Bzl)-OH was allowed to displace the bro-
mine of 8, a product of the Friedel–Crafts acylation of
phenyl octane with bromoacetyl bromide, resulting in
the clean formation of the intermediate ketoester that
was cyclized to the imidazole 9 on treatment with
ammonium acetate and azeotropic removal of water.
Deprotection of the benzyl ether of 9 under Birch reduc-
tion conditions yielded alcohol 10, which was then phos-
phorylated with subsequent oxidation by hydrogen
peroxide to give the protected phosphate 11. Compound
11 was next globally deprotected, yielding the phosphate
12 as the TFA salt. The phosphothioate 14 was obtained
from compound 10 after phosphorylation with subse-
quent oxidation by elemental sulfur to give compound
13. Compound 14 was then obtained as the TFA salt
on global deprotection of 13 in the presence of benze-
nethiol as a cation scavenger.8
Synthesis of the racemic C2-methylated alcohol, phos-
phate, and phosphothioate compounds 17, 20, and 22,
respectively, began with the Fisher esterification of
a-methyl-DL-serine, followed by N-Boc protection, ace-
tonide protection, and finally saponification of the ini-
tially formed methyl ester to give carboxylic acid 15
(Scheme 3). Acid 15 was then coupled to 8 (synthesis
described in Scheme 1) and the resulting ketoester was
To shorten the synthesis of the (2R)-40(50)-phenylimi-
dazole based phosphate and phosphothioate com-
pounds 12 and 14, respectively, the alkyl chain was
installed prior to imidazole formation (Scheme 2). After
Scheme 1. Reagents and conditions: (i) Br2, 1:1 Et2O/dioxane, rt, 12 h, 65%; (ii) Cs2CO3, EtOH, rt, 1 h; (iii) 1, DMF, rt, 1 h, 81% (two steps);
(iv) NH4OAc, xylenes, 110 °C, 6 h, 46%; (v) 1-octyne, Pd(dba)2, Ph3P, CuI, DIEA, THF, rt, 12 h, 91%; (vi) Na °, NH3, ꢀ78 °C, 5 min, 68%;
(vii) tetrazole, di-tert-butyl diisopropylphosphoramidite, 1:1 CH2Cl2/THF, rt, 12 h; (viii) H2O2, rt, 4 h, 73% (two steps); (ix) 1:1 TFA/CH2Cl2,
rt, 4 h, quant.
Scheme 2. Reagents and conditions: (i) AlCl3, BrC(O)CH2Br, 1,2-DCE, 0 °C ! rt, 2 h, 57%; (ii) Cs2CO3, EtOH, rt, 1 h; (iii) 8, DMF, rt, 1 h; (iv)
NH4OAc, xylenes, 110 °C, 6 h, 76% (three steps); (v) Na °, NH3, ꢀ78 °C, 5 min, 42%; (vi) tetrazole, di-tert-butyl diisopropylphosphoramidite, 1:1
CH2Cl2/THF, rt, 12 h; (vii) H2O2, rt, 4 h, 62% (two steps); (viii) 1:1 TFA/CH2Cl2, rt, 4 h, 93%; (ix) S8, rt, 3 h, 39% (two steps); (x) PhSH, TMSBr, 1:1
TFA/CH2Cl2, rt, 4 h, quant.

3570
J. J. Clemens et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3568–3572
Scheme 3. Reagents and conditions: (i) MeOH, SOCl2, 0 °C ! rt, 12 h; (ii) (Boc)2O, NaHCO3, 1:1 H2O/THF, rt, 12 h, 33% (two steps); (iii) 2,2-
DMP, BF3 Æ OEt2, rt, 12 h, 85%; (iv) 2 M NaOH, MeOH, rt, 12 h, 85%; (v) Cs2CO3, EtOH, rt, 1 h; (vi) 8, DMF, rt, 1 h; (vii) NH4OAc, xylenes,
110 °C, 6 h, 20% (three steps); (viii) p-TsOH, MeOH, 70 °C, 3 h, 60%; (ix) (Boc)2O, Na2CO3, 1:1 H2O/THF, rt, 12 h, 57%; (x) tetrazole, di- tert-butyl
diisopropylphosphoramidite, CH2Cl2/THF, rt, 12 h; (xi) H2O2, rt, 3 h, 46% (two steps); (xii) 1:1 TFA/CH2Cl2, rt, 4 h, 91%; (xiii) S8, rt, 3 h, 46% (two
steps); (xiv) PhSH, TMSBr, 1:1 TFA/CH2Cl2, rt, 4 h, 94%.
cyclized to the imidazole compound 16. Alcohol 17 was
obtained as the free base after global deprotection of
compound 16 under acidic conditions with a basic work-
up. To produce the phosphate and phosphothioate com-
pounds 20 and 22, respectively, 17 was regioselectively
protected as the primary N-Boc to give the protected
alcohol 18. Compound 18 was then phosphorylated
and subsequently oxidized by hydrogen peroxide to give
the protected phosphate 19. Global deprotection of 19
yielded the racemic C2-methylated phosphate 20 as the
TFA salt. Oxidation of the phosphite obtained from
the phosphorylation of 18 with elemental sulfur supplied
compound 21. Global deprotection of 21 in the presence
of a cation scavenger furnished the racemic C2-methyl-
ated phosphothioate 22 as the TFA salt.
subjected to a Sonogashira coupling to 1-octyne gener-
ating the aryl alkyne compound, which was then hydro-
genated to reduce the newly formed triple bond. Global
deprotection provided the 2-amino-1,3- propanediol 25
as the HCl salt. To provide the mono-phosphate and
mono-phosphothioate compounds 28 and 30, respec-
tively, 25 was regioselectively protected as the primary
N-Boc and the resulting diol was subjected to conditions
for mono-TBS protection to give the racemic alcohol 26.
Compound 26 was next phosphorylated and then oxi-
dized by hydrogen peroxide to provide compound 27.
Concurrent deprotection of the TBS, phosphate ester,
and N-Boc groups of 27 supplied the mono-phosphate
28 as the HCl salt. To obtain the mono-phosphothioate
30, alcohol 26 was first phosphorylated and then oxi-
dized with elemental sulfur to give compound 29. Con-
current deprotection of the TBS, phosphothioate ester,
and N-Boc groups of 29 with a cation scavenger present
supplied the mono-phosphothioate 30 as the HCl salt.
Synthesis of the 2-amino-1,3-propanediol based
compounds 25, 28, and 30 was initiated with acetonide
protection of tris(hydroxymethyl)aminomethane hydro-
chloride, followed by N-Boc protection, Swern oxida-
tion, and finally sodium chlorite oxidation to give the
carboxylic acid 23 (Scheme 4). Compound 23 was subse-
quently coupled to 1 (synthesis described in Scheme 1)
and cyclized to imidazole 24. Compound 24 was then
Analysis of the phosphate compounds 6 and 12 demon-
strated them to be very potent agonists at S1P1 and
good agonists at S1P5 with good efficacy at each recep-
tor (Table 1).9 The (2R)-compound 12 displayed slightly
Scheme 4. Reagents and conditions: (i) PTSA (cat.), 2,2-DMP, DMF, rt, 12 h; (ii) (Boc)2O, NaHCO3, 1:1 THF/H2O, rt, 12 h, 69% (two steps); (iii)
(COCl)2, DMSO, TEA, DCM, ꢀ78 °C rt, 2 h, 74%; (iv) NaClO2, NaH2PO4 Æ H2O, tBuOH/H2O, 2-methyl-2-butene, rt, 1 h, 95%; (v) Cs2CO3, EtOH,
rt, 1 h; (vi) 1, DMF, rt, 1 h, 86% (two steps); (vii) NH4OAc, xylenes, 110 °C, 6 h, 50%; (viii) 1-octyne, Pd(dba)2, Ph3P, CuI, DIEA, THF, rt, 12 h,
92%; (ix) H2, 10% Pd/C, EtOH, rt, 12 h, quant.; (x) 3 N HCl, THF, rt, 4 h, 87%; (xi) (Boc)2O, NaHCO3, 1:1 THF/H2O, 50 °C, 12 h, 68%; (xii) TBSCl,
imid., DMAP (cat.), CH2Cl2, rt, 1 h, 52%; (xiii) tetrazole, di-tert-butyl diethylphosphoramidite, CH2Cl2/THF, 50 °C, 1 h; (xiv) H2O2, rt, 3 h, 48%
(two steps); (xv) 3N HCl, THF, rt, 4 h, 38%; (xvi) S8, rt, 3 h, 24% (two steps); (xvii) PhSH, TMSBr, 3 N HCl, THF, rt, 4 h, 50%.

J. J. Clemens et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3568–3572
3571
Table 1. EC50 (nM) and Emax values for S1P and synthetic analogues at S1P receptors determined by a [c-35S]GTP binding assayab
Compound
S1P1
S1P2
S1P3
S1P4
S1P5
EC50
4.5 1.1
1.3 1.3
1.2
2700 1.2
1.3
6.4 1.3
naa
Emax
EC50
Emax
EC50
2
Emax
EC50
Emax
EC50
9.2 1.1
Emax
S1P
Phospho-FTY720
1.00
1.00
0.75
0.98
0.69
0.67
0.00
0.88
0.87
0.66
0.91
0.66
8.3 1.2
naa
naa
1.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.95
0.00
8.7 1.1
1.00
0.50
0.56
0.38
0.42
0.45
0.00
0.53
0.44
0.00
0.21
0.68
270 1.2
41 1.2
400 1.1
7400 1.2
150 1.2
190 1.2
naa
1.00
0.78
1.00
0.90
0.79
0.82
0.00
0.94
0.68
0.79
0.87
0.72
1.00
0.56
0.73
0.83
0.69
0.82
0.00
0.72
0.60
0.48
0.66
0.56
2.4
40 1.2
37 1.2
860 1.3
12 1.3
29 1.7
6
7
1700 3.2
2700 1.3
330 1.4
690 1.2
naa
7
naa
naa
12
14
17
20
22
25
28
30
4
naa
naa
naa
4.7 1.3
9.8 1.2
5100 1.2
7.9 1.1
150 1.1
naa
naa
780 1.3
1600 1.3
naa
91 1.4
170 1.4
660 2.1
160 1.2
89 1.5
26 1.1
51 1.2
naa
18 6.3
naa
6700 2.4
17 1.2
16 2.0
630 3.7
790 1.3
a Values are means of three experiments (naa = no agonist activity).
b Emax = maximal efficacy of analogue/maximal efficacy of S1P at the indicated receptor.
more potent agonism at S1P1 and S1P5 than its enantio-
mer 6. As was expected, alcohol 7 was a poor agonist at
all of the S1P receptors, approximately 2- to 3-fold less
potent than the phosphate counterpart 6. The phospho-
thioate 14 retained relative potency and efficacy at S1P1,
as compared to the phosphate counterpart 12, while dis-
playing slightly decreased potency and a mild increase in
efficacy at S1P5.
phothioate head group to obtain significant potency. We
have also demonstrated the retention of agonism on
methylation at the C2 position and that incorporation
of a mono-phosphate 2-amino-1,3-propanediol head
group results in retention of agonist activity at S1P1 as
compared to compounds 6 and 12. The mono-phospho-
thioate compound 30, however, significantly lost agonist
activity at S1P1. Our findings have helped to develop the
SAR of S1P/FTY720 analogues with regard to selectiv-
ity between S1P1 and S1P3 agonism, and will serve as the
basis for future SAR and in vivo studies.
Examination of the C2-methylated phosphate compound
20 demonstrated retention of agonism as compared to the
non-methylated counterparts 6 and 12. As with the phos-
phate counterparts, compound 20 displayed potent agon-
ism at S1P1 and good potency at S1P5. The 2-methylated
phosphothioate compound 22 retained the agonism of the
phosphate counterpart 20 with only slight decreases in
potency and efficacy at S1P1, S1P3, S1P4, and S1P5. The
2-methylated alcohol compound 17 showed a total lack
of agonism at each of the S1P receptors.
Acknowledgments
This work was supported by NIH grants (NIGMS R01
GM067958 (KRL) and NIGMS F31 GM064101
(MDD)).
Incorporation of a 2-amino-1,3-propanediol head group
gave several interesting results. As expected, the diol
compound 25 was a poor agonist at all of the S1P recep-
tors. The mono-phosphate compound 28, however,
demonstrated very high levels of potency and efficacy
as an agonist at S1P1 and S1P2, to our knowledge the
first synthetic compound to demonstrate potent agonism
at S1P2. Compound 28 showed only moderate selectivity
however, as it was also a relatively potent agonist at
S1P4, and S1P5. Compared to 28, the mono-phospho-
thioate 30 lost potency and efficacy as an agonist at
S1P1 and all agonist activity at S1P2. With regard to
S1P3, S1P4 and S1P5, compound 30 retained the proper-
ties of compound 28, acting as a poor agonist at S1P3,
while displaying rather high potency and moderate effi-
cacy at S1P4 and S1P5.
References and notes
1. (a) Hale, J. J.; Neway, W.; Mills, S. G.; Hajdu, R.;
Keohane, C. A.; Rosenbach, M.; Milligan, J.; Shei, G. J.;
Chrebet, G.; Bergstrom, J.; Card, D.; Coo, G. C.; Koprak,
S. L.; Jackson, J. J.; Rosen, H.; Mandala, S. Bioorg. Med.
Chem. Lett. 2004, 14, 3351; (b) Lim, H. S.; Park, J. J.; Ko,
K.; Lee, M. H.; Chung, S. K. Bioorg. Med. Chem. Lett.
2004, 14, 2499.
2. Brinkmann, V.; Davis, M. D.; Heise, C. E.; Albert, R.;
Cottens, S.; Hof, R.; Bruns, C.; Prieschl, E.; Baumruker,
T.; Hiestand, P.; Foster, C. A.; Zollinger, M.; Lynch, K. R.
J. Biol. Chem. 2002, 277, 21453.
3. Matloubian, M.; Lo, C. G.; Cinamon, G.; Lesneski, M. J.;
Xu, Y.; Brinkmann, V.; Allende, M. L.; Proia, R. L.;
Cyster, J. G. Nature 2004, 427, 355.
4. Sanna, M. G.; Liao, J.; Jo, E.; Alfonso, C.; Ahn, M. Y.;
Peterson, M. S.; Webb, B.; Lefebvre, S.; Chun, J.; Gray, N.;
Rosen, H. J. Biol. Chem. 2004, 279, 13839.
5. (a) Clemens, J. J.; Davis, M. D.; Lynch, K. R.; Macdonald,
T. L. Bioorg. Med. Chem. Lett. 2004, 14, 4903; For
previously reported S1P receptor agonists see: (b) Clemens,
J. J.; Davis, M. D.; Lynch, K. R.; Macdonald, T. L. Bioorg.
Med. Chem. Lett. 2003, 13, 3401; (c) Im, D. S.; Clemens, J.;
Macdonald, T. L.; Lynch, K. R. Biochemistry 2001, 40,
14053.
To summarize, we have synthesized a series of S1P/
FTY720 analogues that incorporate a 4(5)-phenylimi-
dazole ring system in the ꢀlinkerꢁ region of the pharma-
cophore. This structural modification has resulted in
the generation of highly potent S1P1 agonists. We have
determined a slight preference in potency for 2R-config-
uration, as well as a necessity for the phosphate or phos-

3572
J. J. Clemens et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3568–3572
6. Gordon, T. D.; Singh, J.; Hansen, P. E.; Morgan, B. A.
Tetrahedron Lett. 1993, 34, 1901.
7. Beaucage, S. L.; Caruthers, M. H. Tetrahedron Lett. 1981,
22, 1859.
8. Debont, D. B. A.; Moree, W. J.; Vanboom, J. H.; Liskamp,
R. M. J. J. Org. Chem. 1993, 58, 1309.
9. A brief assay protocol may be found in Ref. 5a.