Discovery of potent LPA2 (EDG4) antagonists as potential anticancer agents

Article abstract of DOI:10.1016/j.bmcl.2007.12.024

The LPA₂ protein is overexpressed in many tumor cells. We report the optimization of a series of LPA₂ antagonists using calcium mobilization assay (aequorin assay) that led to the discovery of the first reported inhibitors selective for LPA₂. Key compounds were evaluated in vitro for inhibition of LPA₂ mediated Erk activation and proliferation of HCT-116 cells. These compounds could be used to evaluate the benefits of LPA₂ inhibition both in vitro and in vivo.

Full text of DOI:10.1016/j.bmcl.2007.12.024

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1037–1041
Discovery of potent LPA2 (EDG4) antagonists
as potential anticancer agents
Hilary P. Beck,a,* Todd Kohn,a Steven Rubenstein,a, Christine Hedberg,a
Ralf Schwandner,b Kerstin Hasslinger,b Kang Dai,c Cong Li,c Lingming Liang,c
Holger Wesche,c Brendon Frank,c Songhzu An,c Dineli Wickramasinghe,c Juan Jaen,a,à
Julio Medina,a Randall Hungatea and Wang Shena
aChemistry Research & Discovery, Amgen Inc., 1120 Veterans Boulevard, South San Francisco, CA 94080, USA
bChemistry Research & Discovery, Amgen Research GmbH, Regensburg, Germany
cOncology, Amgen Inc., 1120 Veterans Boulevard, South San Francisco, CA 94080, USA
Received 18 October 2007; revised 8 December 2007; accepted 11 December 2007
Available online 4 January 2008
Abstract—The LPA2 protein is overexpressed in many tumor cells. We report the optimization of a series of LPA2 antagonists using
calcium mobilization assay (aequorin assay) that led to the discovery of the first reported inhibitors selective for LPA2. Key com-
pounds were evaluated in vitro for inhibition of LPA2 mediated Erk activation and proliferation of HCT-116 cells. These com-
pounds could be used to evaluate the benefits of LPA2 inhibition both in vitro and in vivo.
Ó 2007 Elsevier Ltd. All rights reserved.
The lysophosphatidic acid receptor 2 (LPA2, also known
as endothelial differentiation gene 4, EDG4) is a G-pro-
tein coupled receptor (GPCR) activated by lysophospha-
N O
tidic acid (LPA, 1- or 2-acyl-sn-glycerol-3-phosphate,
general structure represented by 1 in Fig. 1).1,2 LPA is
known to mediate cell adhesion and migration,3 promote
cancer cell proliferation,4 and elevated levels of LPA are
often found in some cancer patients.1b,5 LPA2 has been
shown to be overexpressed in cells from many tumor types
(breast,6 colorectal,7 gastric,8 ovarian,1b,9 and kidney).10
Recently, siRNA knock-down of LPA2 in HCT-116 co-
lon cancer cells led to growth arrest and apoptosis of cells
in vitro and in xenograph models.11 Computational mod-
eling has been used to develop LPA2 agonists12 and sev-
eral LPA2 agonists are reported,13 including some that
induce cell migration and proliferation.13c In an effort to
explore novel cancer treatment pathways, we sought to
develop a small molecule LPA2 antagonist as a potential
anticancer therapeutic.
HN
O
OH
O
OH
OH
R
O
O
O
P
3
O
O
Cl
1
F
HO
O
N
N
HN
Me
N
4
O
N
N
O
2
Keywords: LPA2; EDG4; LPA; Anticancer agents.
*
O
Corresponding author. Tel.: +1 650 244 2190; fax: +1 650 837
9396; e-mail: hbeck@amgen.com
Figure 1. Literature compounds and HTS hit.
Present address: AMRI, 7001 Performance Drive, North Syracuse,
NY 13212, USA.
à
Here we report the first potent and selective small mol-
ecule antagonists for LPA2. Using a cell-based Ca2+ flux
Present address: ChemoCentryx, Inc., 850 Maude Avenue, Mountain
View, CA 94043, USA.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.12.024

1038
H. P. Beck et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1037–1041
assay in a high throughput screening of our compound
library,14 we identified compound 2 as an inhibitor of
LPA-induced Ca2+ flux (IC50 = 1.7 lM). Some com-
pounds have been reported as antagonists for the LPA
receptors LPA1 (EDG2) and LPA3 (EDG7) (317 and
4,18 respectively). Due to the sequence similarity among
LPA1, LPA2, and LPA3,1c–e these compounds were
examined for their LPA2 potency but were found to be
inactive.
Cl
N
a
b, c
N
6-10
R1
11
R4
N
O
OH
R2
N
R2
R3
HN
HN
R3
In order to explore the SAR of this initial hit, we needed
a reliable method to prepare 2-alkyl-quinazolines. We
performed several attempts to prepare the desired
substituted quinazolines (Scheme 1). Our most success-
ful efforts employed 2-amino benzamides 5. Unfortu-
nately, the amides 5 failed to reproducibly afford the
desired quinazolines 6 under basic conditions (NaOEt,
EtOH, 90 °C) when reacted with either cyclopropylethyl
ester or cyclopropyl acid chloride. However, we discov-
ered that under microwave heating conditions, we could
quickly and reliably prepare the desired quinazolines 6–
10 in moderate yields by using neat alkyl carboxylic
acids (Table 1).
d, e
N
N
R1
N
R1
N
12
13-19
Scheme 2. Preparation of linker analogues. Reagents and conditions:
(a) POCl3, 100 °C, 4 h, 80%; (b) Na2CO3, NH2C(R2, R3)CO2Me,
MeCN, 12 h; (c) LiOH, 90% over two steps; (d) R4-aryl piperazine,
HOBt, EDC, CH2Cl2, 50–100%; (e) BH3, THF, 20–100%.
are shown in Table 2. 2-Trifluoromethyl-quinazoline
14 was preferred over the 2-cyclopropyl derivative 13.
Analogues 14 and 15 illustrated that (S) is the preferred
stereochemistry within the linker region. Only the
monomethyl linkers 14 and 19 were found to be active.
Other groups such as ethyl 17, dimethyl 18, and unsub-
stituted 16 linker regions resulted in greatly reduced po-
tency. Several substituents on the phenyl group were
explored with the 2-methyl 19 being the most potent
derivative in this series.
With a convenient route to 2-alkyl-quinazolines in place,
we examined several linker and substituted phenyl deriv-
atives for antagonist activity. The analogues were pre-
pared by converting the 4-hydroxyquinazolines 6–10 to
the corresponding chlorides 11 (Scheme 2). SNAr instal-
lation of the amino acid derivatives followed by amide
coupling and then reduction of the amide provided ana-
logues 13–19. Pyrimidyl ring reduction was a major side-
product of this sequence. All biologically active
compounds were characterized by 1H NMR, LC/MS
and their purity was determined to be greater than
95% by reverse phase HPLC.19
Having established the SAR of the linker region, our ef-
fort then shifted to finding the optimal heterocyclic tail
functionality for LPA2 activity. To this end, N-Boc pro-
tected L-alanine was coupled with 2-tolylpiperazine un-
der standard conditions to afford 20 (Scheme 3). Mild
reduction of 20 with borane at room temperature affor-
ded amine 22. Under reflux heating, 20 was converted to
the undesired methyl amine 21. The amine of carbamate
22 was then unmasked and coupled to various heterocy-
clic chlorides via SNAr displacement (Cs2CO3) or Pd
catalyzed coupling (23,20 NaOtBu). Many heterocycles
were well tolerated as tail moieties for LPA2 antagonists;
however, the thienopyrimidine 25 was found to be supe-
rior with an IC50 of 0.26 lM in the aequorin assay.
Compounds 13–19 were tested for LPA2 activity using a
calcium mobilization (aequorin) assay and the results
O
OH
a or b
H2N
H2N
N
R1
N
R2
R2
5
6-10
R1 = cPr, CH3, CF3, iPr
R2 = H or OMe
Since the thienopyrimidine 25 was found to be the opti-
mal tail functionality for potency, we explored the SAR
Scheme 1. Preparation of quinazolines. Reagents and conditions: (a)
R1CO2Et or R1CO2Cl, NaOEt, EtOH, 90 °C, 24 h; (b) RCO2H, l-
wave, 300 °C, 10 min.
Table 2. LPA2 activity of compounds 13–19a
Compound
R1
R2
R3
R4
IC50
Table 1. Preparation of 2-alkyl-quinazolines 6–10 using microwavea
13
14
15
16
17
18
19
c-Pr
CF3
CF3
CF3
CF3
CF3
CF3
CH3
CH3
H
H
2-OCH3
2-OCH3
2-OCH3
2-OCH3
2-OCH3
2-OCH3
2-CH3
3.2
1.3
H
Compound
R1
R2
Yield (%)
CH3
H
>30
>30
>30
>30
0.73
H
6
7
c-Pr
Me
H
52
69
37
75
29
CH2CH3
CH3
CH3
H
H
CH3
H
8
i-Pr
CF3
CF3
H
9
10
H
OMe
a IC50 values in lM (n = 2) have been determined by Ca2+ flux in
RH7777 cells co-expressing LPA2, chim, Gi4-protein, and aequorin.
a Isolated yields are reported in table.

H. P. Beck et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1037–1041
1039
N
N
N
b
a
N
N
O
N
HN
BocHN
BocHN
21
20
22
25: Het =
N
c, d or e
t-Bu
P
N
t-Bu
OAc
S
Pd
N
N
HetHN
23
24
Scheme 3. Preparation of quinazoline replacements. Reagents and conditions: (a) BH3–THF, THF, reflux, 100%; (b) BH3–THF, rt; then
NH2CH2CH2NH2, MeOH, reflux, 43%; (c) HCl, dioxane, CH2Cl2, 72%; (d) HetCl, Cs2CO3, CH3CN, 85 °C; (e) HetCl, 23, NaOtBu, toluene, 80 °C,
20–89%.
Table 3. LPA2 activity of compounds 29–35a
of the head region with this advanced tail in place. The
compounds were prepared by coupling chloride 26 with
the amine linker 27, thus affording 28 (Scheme 4). After
amine deprotection several functional groups were then
examined (amide, urea, carbamate, and alkyl). Aryl sul-
fonamides were found to be the most active.
Compound
R
IC50
29
30
31
32
33
34
35
Butyl
Benzyl
6.61
1.86
0.73
0.31
0.17
0.035
0.017
2-Me-Ph
3-Me-Ph
4-Me-Ph
3-Cl-Ph
3,4-Cl2-Ph
Compounds 29–35 were examined for LPA2 activity in
the aequorin assay and the results are shown in Table
3. Phenyl analogues 31–35 were found to be optimal
over alkyl 29 and benzyl 30 substituted derivatives for
potency. In addition, 4-methyl 33 was more active than
2-methyl 31 and 3-methyl 32. The 3-chloro derivative 34
provided a very potent compound with 3,4-dichloro 35
being the most active analogue of this series.
a IC50 values lM (n = 2) have been determined by Ca2+ flux in RH7777
cells co-expressing LPA2, chim, Gi4-protein, and aequorin.
Table 4. Selectivity for LPA2, LPA1, and LPA3 receptorsa
Compound
LPA2
LPA1
LPA3
25
35
3
0.26
0.017
>50
>50
>50
>50
>50
The LPA2 selectivity of the most potent compounds was
tested against LPA1 and LPA3 along with known LPA1
and LPA3 antagonists (3 and 4) and the results are
shown in Table 4. Heterocycle 3 was found to be a po-
tent LPA1 antagonist while exhibiting some LPA3 activ-
0.05–0.30
>50
12.00
0.04–0.08
4
>50
a IC50 values lM (n = 2) have been determined by Ca2+ flux in RH7777
cells co-expressing LPA1, LPA2, or LPA3, chim. Gi4-protein, and
aequorin.
Boc
Cl
N
ity but no LPA2 activity. Amide 4 was found to be a
select and potent LPA3 antagonist showing no activity
in either LPA1 or LPA2 receptor assay. Tolyl analogue
25 and 3,4-dichloro derivative 35 are analogues of differ-
ent chemotypes and were found to be inactive against
the LPA1 and LPA3 receptors. Thus, these analogues
of different chemotypes are potent and selective LPA2
antagonists.21
S
N
a
N
N
H2N
26
27
Boc
SO2R
N
N
N
S
N
Encouraged by these results we sought to further exam-
ine the biological effect of our selective LPA2 antago-
nists. Since activation of LPA2 by LPA may lead to
phosphorylation of Erk in LPA responsive cells, we
investigated the effect of 35 on phosphorylation of
Erk. An in-Cell Western blot analysis of cell lysates from
HCT-116 cells treated with 35 was performed. The data
shown in Figure 2 demonstrates that 35 inhibited the
phosphorylation of Erk induced by LPA in a concentra-
tion dependent manner. As a specificity control, 35
HN
HN
b,c
N
S
N
N
N
28
29-35
Scheme 4. Preparation of sulfamide replacements. Reagents and
conditions: (a) Cs2CO3, CH3CN, 93%; (b) HCl, dioxane, 100%; (c)
RSO2Cl, py, CH2Cl2, 31–81%.

1040
H. P. Beck et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1037–1041
References and notes
LPA 3 μM
HGF 30 ng/ml
110
100
90
80
70
60
50
40
30
20
10
0
1. (a) Xu, Y.; Qian, L.; Prestwich, G. D. J. Org. Chem. 2003,
68, 5320; (b) Mills, G. B.; Eder, A.; Fang, X.; Hasegawa,
Y.; Mao, M.; Lu, Y.; Tanyi, J.; Tabassam, H.; Wiener, J.;
Lapushin, R.; Yu, S.; Parrot, J. A.; Compton, T.; Tribley,
W.; Fishman, D.; Stack, M. S.; Gaudette, D.; Jaffe, R.;
Furui, T.; Aoki, J.; Erickson, J. R. Cancer Treat. Res.
2002, 107, 259; (c) Chun, J.; Goetzl, E. J.; Hla, T.;
Igarashi, Y.; Lynch, K. R.; Moolenaar, W.; Pyne, S.;
Tigyi, G. Pharmacol. Rev. 2002, 54, 265; (d) Bandoh, K.;
Juken, A.; Taira, A.; Tsujimoto, M.; Arai, H.; Inoue, K.
FEBS Lett. 2000, 478, 159; (e) Contos, J. J. A.; Chun Mol.
Pharmacol. 2000, 58, 1188.
-2
-1
0
1
Log [35] μM
2. Intracellularly, LPA activates PPARc. See McIntyre, T.
M.; Pontsler, A. V.; Silva, A. R.; Hilaire, A. St.; Xu, Y.;
Hinshaw, J. C.; Zimmerman, G. A.; Hama, K.; Aoki, J.;
Arai, H.; Prestwich, G. D. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 131.
Figure 2. 35 inhibits LPA induced phosphorylation of Erk.22
3. (a) Ren, J.; Xiao, Y.-J.; Singh, L. S.; Zhao, X.; Zhao, Z.;
Feng, L.; Rose, T. M.; Prestwich, G. D.; Xu, Y. Cancer
Res. 2006, 66, 3006; (b) Xu, J.; Lai, Y.-J.; Lin, W.-C.; Lin,
F-T. J. Biol. Chem. 2004, 279, 10459.
4. (a) Lee, Z.; Swaby, R. F.; Liang, Y.; Yu, S.; Liu, S.; Lu, K.
H.; Bast, R. C.; Mills, G. B.; Fang, X. Cancer Res. 2006,
66, 2740; (b) Fang, X.; Yu, S.; Bast, R. C.; Liu, S.; Xu, H.-
J.; Hu, S.-X.; LaPushin, R.; Claret, F. X.; Aggarwal, B. B.;
Lu, Y.; Mills, G. B. J. Biol. Chem. 2004, 279, 9653.
5. Chou, C.-H.; Wei, L.-H.; Kuo, M.-L.; Huang, Y.-J.; Lai,
K.-P.; Chen, C.-A.; Hsieh, C.-Y. Carcinogenesis 2005, 26,
45.
6. Kitayama, J.; Shida, D.; Sako, A.; Ishikawa, M.; Hama,
K.; Aoki, J.; Arai, H.; Nagawa, H. Breast Cancer Res.
2004, 6, R640.
Figure 3. Compound 35 inhibits LPA induced proliferation of HCT-
116 colon cancer cells.23
7. Shida, D.; Watanabe, T.; Aoki, J.; Hama, K.; Kitayama,
J.; Sonoda, H.; Kishi, Y.; Yamaguchi, H.; Sasaki, S.;
Sako, A.; Konishi, T.; Arai, H.; Nagawa, H. Lab. Invest.
2004, 84, 1352.
8. Yamashita, H.; Kitayama, J.; Shida, D.; Ishikawa, M.;
Hama, K.; Aoki, J.; Arai, H.; Nagawa, H. J. Surg. Oncol.
2006, 93, 30.
showed only a minor effect on ERK phosphorylation
stimulated by HGF. Thus, an LPA2 specific antagonist
can inhibit LPA induced phosphorylation of Erk in
the HCT-116 colon cancer cells.
9. Estrella, V. C.; Eder, A. M.; Liu, S.; Pustilnik, T. B.;
Tabassam, F. H.; Claret, F. X.; Gallick, G. E.; Mills, G.
B.; Wiener, J. R. Int. J. Oncol. 2007, 31, 441.
10. Li, J.; Powers, S.; Sin, W.C.; Jianxin, Y.W.O. Patent
04046332, 2004.
11. Yang, M.; Zhong, W. W.; Srivastava, N.; Slavin, A.;
Yang, J.; Hoey, T.; An, S. Proc. Natl. Acad. Sci. U.S.A.
2005, 102, 6027.
12. Virag, T.; Elrod, D. B.; Liliom, K.; Sardar, V. M.; Parrill,
A. L.; Yokoyama, K.; Durgam, G.; Deng, W.; Miller, D.
D.; Tigyi, G. Mol. Pharmacol. 2003, 63, 1032.
Since LPA has been known to promote proliferation of
some cancer cells,6 we examined the effect of 35 on LPA
induced proliferation of HCT-116 colon cancer cells. As
shown in Figure 3, cell proliferation caused by LPA was
inhibited by 35 in a doses dependent manner. This effect
appears to be specific since HGF induced proliferation
was only slightly inhibited at high concentrations of
35. These data suggest that LPA2 antagonism can poten-
tially inhibit proliferation of cancer cells.
13. (a) Jiang, G.; Xu, Y.; Fujiwara, Y.; Tsukahara, T.;
Tsukahara, R.; Gajewiak, J.; Tigyi, G.; Prestwich, G. D.
Chem. Med. Chem. 2007, 2, 679; (b) Zhang, H.; Tsuku-
hara, R.; Tigyi, G.; Prestwich, G. D. J. Org. Chem. 2006,
71, 6061; (c) Qian, L.; Xu, Y.; Simper, T.; Jiang, G.; Aoki,
J.; Umezu-Goto, M.; Arai, H.; Yu, S.; Mills, G. B.;
Tsukahara, R.; Makarova, N.; Fujiwara, Y.; Tigyi, G.;
Prestwich, G. D. Chem. Med. Chem. 2006, 1, 376.
14. Rat hepatoma cells (RH7777) were transiently transfected
with aequorin, chim. Gi4-protein, and LPA2 using Fugene
6 transfection reagent (Roche Biosciences). In this assay,
the reduction of LPA-induced receptor activation by
library compounds in a 384-well based assay was moni-
tored by flash luminescence detection. For experimental
details, see Refs. 15,16.
In summary, a high-throughput screen and a subsequent
medicinal chemistry effort identified the first reported
compounds that are selective and highly potent LPA2
antagonists. Cell based functional assays showed that
the potent LPA2 antagonist 35 inhibited LPA induced
Erk activation and proliferation of HCT-116 colon can-
cer cells. These compounds could be used as tool com-
pounds to evaluate the anticancer effects of blocking
LPA2 mediated signaling.
Acknowledgment
The authors thank Dr. John Eksterowicz for molecular
modeling support.
15. An, S.; Thieu, B.; Yuhua, Z.; Goetzl, E. J. Mol. Pharma-
col. 1998, 54, 881.

H. P. Beck et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1037–1041
1041
16. Bandoh, K.; Aoki, J.; Hosono, H.; Kobayashi, S.;
Kobayashi, T.; Murakami-Murofushi, K.; Tsujimoto,
M.; Arai, H.; Inoue, K. J. Biol. Chem. 1999, 274, 27776.
17. Terakado, M.; Nakade, S.; Seko, T.; Takaoka, Y.W.O
Patent 04031118, 2004.
37 °C. The cells were then treated with LPA at 3 lM or
HGF at 30 ng/ml for 12 min at 37 °C followed by fixation
and permeabilization. The plate was then blocked for 2 h
in Odyssey blocking buffer. After primary and secondary
antibodies’ incubation, phosphorylated Erk (p-Erk) and
total Erk were simultaneously detected by LI-COR system
(LI-COR Biosciences). p-Erk signal was normalized using
the total ERK signal from each well.
18. Ueno, A.; Nagao, R.; Watanabe, T.; Ohta, H.; Yagi,
M.W.O. Patent 01060819, 2001.
19. HPLC (A: 0.1% TFA in H2O, B: 0.1% TFA in CH3CN,
1
10–100% B in 15 min), H NMR (Bruker, 500 MHz).
23. HCT-116 cells were plated in 96-well plates (5000 cells/
well) in serum free DMEM with 0.11% BSA. Compound
35 was added at various concentrations. LPA or HGF was
then added to a final concentration of 3 lM or 30 ng/ml
respectively. After 24-h incubation, viable cells were
determined by using the CellTiter-Glo Assay kit
(Promega).
20. Zim, D.; Buchwald, S. L. Org. Lett. 2003, 5, 2413.
21. Compounds 19 and 35 were also inactive against the S1P
receptor S1P1 (also known as EDG1).
22. HCT-116 cells at 90% confluence in a 96-well plate were
serum-starved for 2 h. Compound 35 was added at various
concentrations and incubated with the cells for 1 h at