Synthesis, structure-activity relationships, and biological evaluation of fatty alcohol phosphates as lysophosphatidic acid receptor ligands, activators of PPARγ, and inhibitors of autotaxin

Article abstract of DOI:10.1021/jm049609r

We previously reported that fatty alcohol phosphates (FAP) represent a minimal pharmacophore required to interact with lysophosphatidic acid (LPA) receptors. To improve the activity of the first-generation saturated FAP series, a structure-activity relationship (SAR) study was carried out that includes modifications to the headgroup and alkyl side chain of the FAP pharmacophore. A series of unsaturated (C₁₀-C₁₈) FAP, headgroup-modified hydrolytically stable saturated (C₁₀-C₁₈) alkyl phosphonates, and saturated and unsaturated (C₁₀-C₁₈) thiophosphate analogues were synthesized and evaluated for activity in RH7777 cells transfected with individual LPA_1-3 receptors, in PC-3 cells and in human platelets that endogenously express all three isoforms. In this series we identified several LPA₁- and LPA₃-selective antagonists with IC₅₀ values in the nanomolar range. Oleoyl-thiophosphate (15g) was shown to be a panagonist, whereas tetradecyl-phosphonate (16c) was identified as a pan-antagonist. These compounds were also tested for the ability to activate the transcription factor PPARγ, an intracellular receptor for LPA, in CV1 cells transfected with the PPRE-Acox-Rluc reporter gene. All the FAP tested, along with the previously reported LPA GPCR antagonists dioctanoyl glycerol pyrophosphate (2), Ki16425 (6), and the agonist OMPT (3), were activators of PPARγ. The pan-agonist oleoyl-thiophosphate (15g) and pan-antagonist tetradecyl-phosphonate (16c) mimicked LPA in inhibiting autotaxin, a secreted lysophospholipase D that produces LPA in biological fluids.

Full text of DOI:10.1021/jm049609r

J. Med. Chem. 2005, 48, 4919-4930
4919
Synthesis, Structure-Activity Relationships, and Biological Evaluation of Fatty
Alcohol Phosphates as Lysophosphatidic Acid Receptor Ligands, Activators of
PPARγ, and Inhibitors of Autotaxin†
,§
Gangadhar G. Durgam, Tamas Virag,‡,§ Michelle D. Walker,‡ Ryoko Tsukahara,‡ Satoshi Yasuda,‡
Karoly Liliom,‡ Laurens A. van Meeteren,# Wouter H. Moolenaar,# Nicole Wilke,X Wolfgang Siess,X
,|
Gabor Tigyi,‡,| and Duane D. Miller*,
Department of Pharmaceutical Sciences, College of Pharmacy and Department of Physiology, College of Medicine, University of
Tennessee Health Science Center, 847 Monroe Avenue, Room 227C, Memphis, Tennessee 38163, Division of Cellular
Biochemistry, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands, and Institute for
Prevention of Cardiovascular Diseases, Medical Faculty, University of Munich, Munich, Germany
Received May 27, 2004
We previously reported that fatty alcohol phosphates (FAP) represent a minimal pharmacophore
required to interact with lysophosphatidic acid (LPA) receptors. To improve the activity of the
first-generation saturated FAP series, a structure-activity relationship (SAR) study was carried
out that includes modifications to the headgroup and alkyl side chain of the FAP pharma-
cophore. A series of unsaturated (C10-C18) FAP, headgroup-modified hydrolytically stable
saturated (C10-C18) alkyl phosphonates, and saturated and unsaturated (C10-C18) thiophos-
phate analogues were synthesized and evaluated for activity in RH7777 cells transfected with
individual LPA1-3 receptors, in PC-3 cells and in human platelets that endogenously express
all three isoforms. In this series we identified several LPA1- and LPA3-selective antagonists
with IC50 values in the nanomolar range. Oleoyl-thiophosphate (15g) was shown to be a pan-
agonist, whereas tetradecyl-phosphonate (16c) was identified as a pan-antagonist. These
compounds were also tested for the ability to activate the transcription factor PPARγ, an
intracellular receptor for LPA, in CV1 cells transfected with the PPRE-Acox-Rluc reporter gene.
All the FAP tested, along with the previously reported LPA GPCR antagonists dioctanoyl
glycerol pyrophosphate (2), Ki16425 (6), and the agonist OMPT (3), were activators of PPARγ.
The pan-agonist oleoyl-thiophosphate (15g) and pan-antagonist tetradecyl-phosphonate (16c)
mimicked LPA in inhibiting autotaxin, a secreted lysophospholipase D that produces LPA in
biological fluids.
Introduction
With a growing understanding of the involvement of
LPA in both normal physiology and pathology it is
evident that LPA receptor agonists and antagonists may
have therapeutic potential in treating various diseases.
LPA mimics are of therapeutic interest in wound
healing14,15 and protection against radiation- and che-
motherapy-induced apoptosis.16 Alternatively, LPA an-
tagonists may have therapeutic potential in the treat-
ment of thrombosis elicited by plaque rupture,17,18
ovarian cancer,19 atherosclerosis,20 ischemic reperfusion
injury,21 and inflammation.22
Understanding the physiological and pathological role
of LPA is a challenge due to the complexity of LPA
signaling. Most cells express multiple LPA plasma
membrane receptors, which upon stimulation interact
with multiple heterotrimeric G proteins. The recent
discovery of PPARγ as an intracellular LPA receptor
and the likely presence of more plasma membrane
receptors for LPA add to the complexity of the system.
To decipher the physiological role of individual recep-
tors, and to take advantage of LPA analogues in treating
various LPA-associated conditions, it is necessary to
identify subtype-selective agonists and antagonists.
Lysophosphatidic acid (1-acyl-2-hydroxy-sn-glycero-
3-phosphate, LPA, 1; Figure 1) is a member of phos-
pholipid growth factor (PLGF) family. LPA is generated
by activated platelets and tumor cells and elicits a wide
range of biological effects including the stimulation of
cell proliferation, and migration, as well as the promo-
tion of cell survival, platelet aggregation, apoptosis, and
smooth muscle contraction.1,2 LPA mediates many of
these cellular responses through the activation of
members of endothelial differentiation gene (EDG)
family of G-protein-coupled receptors LPA1/EDG-2,
LPA2/EDG-4, LPA3/EDG-7.3-9 Recently two non-EDG
LPA receptors, LPA4/GPR23 and the nuclear transcrip-
tion factor/receptor PPARγ, have been reported.10,11 LPA
is a product of lysophospholipase D, also known as
autotaxin, a phosphodiesterase that cleaves lysophos-
phatidylcholine to LPA and choline.12,13
† Part of this work was presented at 227th American Chemical
Society National Meeting, Anaheim, CA, March 28-April 1, 2004.
* Corresponding author. Phone: +1-901-448-6026. Fax: +1-901-
448-3446. E-mail: dmiller@utmem.edu.
Department of Pharmaceutical Sciences.
‡ Department of Physiology.
# Division of Cellular Biochemistry.
Despite their promising therapeutic potential, only a
few agents with limited subtype selectivity have been
reported to date. A number of attempts have been made
X Institute for Prevention of Cardiovascular Diseases.
§ Contributed equally.
| Senior co-authors.
10.1021/jm049609r CCC: $30.25 © 2005 American Chemical Society
Published on Web 06/22/2005

4920 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Durgam et al.
Scheme 1a
a Reagents and conditions: (a) (i) di-tert-butyl N,N-diisopropy-
lphosphoramidite, 1H-tetrazole, CH2Cl2; (ii) 50% H2O2, 68-82%;
(b) TFA, CH2Cl2, 78-86%.
Scheme 2a
a Reagents and conditions: (a) (i) bis(2-cyanoethyl) N,N-diiso-
propylphosphoramidite, 1H-tetrazole, CH2Cl2, rt; (ii) sulfur, reflux,
72-84%; (b) (i) KOH, MeOH; (ii) dil HCl, 70-86%.
analogues to selectively activate or inhibit LPA1-3
receptors. The results include identification of several
LPA1- and LPA3-selective antagonists with IC50 values
in the nanomolar range. We also identified oleoyl-
thiophosphate (15g) as a novel pan-agonist and tetrade-
cyl-phosphonate (16c) as a pan-antagonist with respect
to LPA GPCR. We also show that all the FAP we
examined are activators of PPARγ and that the pan-
agonist (15g) and pan-antagonist (16c) compounds
mimic LPA in inhibiting autotaxin/lysophospholipase D.
Chemistry. The synthesis of alkenyl phosphates
(12a-f), as outlined in Scheme 1, was carried out
according to the method of Lindberg et al.,34 with the
exception that hydrogen peroxide was used for the
oxidation step instead of m-chloroperbenzoic acid. Com-
mercially available unsaturated fatty alcohols (10a-f)
were treated with a mixture of 1H-tetrazole and di-tert-
butyl N,N-diisopropylphosphoramidite in anhydrous
methylene chloride followed by hydrogen peroxide oxi-
dation to give di-t-Boc-protected fatty alcohol phos-
phates (11a-f). Boc- protected FAP were deprotected
with TFA to yield the corresponding unsaturated FAP
(12a-f).
The synthesis of thiophosphates (15a-g) is outlined
in Scheme 2. The protected thiophosphoric acid O,O′-
bis-(2-cyanoethyl) ester O′′-alkyl/alkenyl esters were
synthesized using a modified method of Haines et al.35
Commercially available fatty alcohols (13a-g) were
treated with a mixture of 1H-tetrazole and bis(2-
cyanoethyl) N,N-diisopropylphosphoramidite in anhy-
drous methylene chloride followed by reflux in the
presence of elemental sulfur to give bis-cyanoethyl-
protected fatty alcohol thiophosphates (14a-g). These
protected thiophosphates were treated with methanolic
KOH, followed by acidification, to yield the required
thiophosphates (15a-g). The n-alkyl phosphonate ana-
logues (16a-d) used in this study were purchased from
commercial sources.
Figure 1. Chemical structures of LPA, saturated FAP, and
other known LPA receptor ligands.
to identify LPA receptor agonists and antagonists.1
N-Acylethanolamide phosphate (NAEPA) and N-palmi-
toylserine phosphate (Ser-PA) have been identified as
LPA receptor agonists in mammalian cell lines.23,24 Ser-
PA has been documented to antagonize both LPA-
induced platelet aggregation25 and oocyte depolariza-
tion.26 Fisher et al. showed that the short-chain
phosphatidates, dioctanoyl glycerol pyrophosphate
(DGPP, 2) and dioctanoyl phosphatidic acid (PA 8:0),
as selective antagonists of the LPA1 and LPA3, have 1
order of magnitude preference for LPA3.27 Rational
modifications of the glycerol backbone and the head-
group of LPA resulted in the identification of OMPT
(3),28,29 a phosphothionate analogue, and a monofluoro
phosphonate analogue 4 as selective LPA3 receptor
agonists.30 VPC12249 (5) was characterized as a selec-
tive LPA1 and LPA3 antagonist, whereas a nonlipid
compound Ki16425 (6) was shown to be an antagonist
of LPA1/3.
31,32 Two fluorinated acyl migration-resistant
LPA analogues, XY-4 (7) and XY-8 (8), have been
identified as selective PPARγ agonists.11
We previously reported a rational drug design driven
identification of fatty alcohol phosphates (FAP), which
lack a glycerol backbone and represent the minimal
pharmacophore that interacts with LPA receptors, as
subtype-selective LPA receptor agonists and antago-
nists.33 n-Decyl- (9a, FAP-10), n-dodecyl- (9b, FAP-12),
and n-tetradecyl- (9c, FAP-14) phosphates were identi-
fied as selective LPA2 agonists and LPA1/LPA3 antago-
nists (Figure 1). In the current series of modifications
we investigated the effect of headgroup modification,
side chain modification, and double bonds at specific
positions in the side chain on the ligand properties at
LPA1-3 GPCR and PPARγ. Here we provide evidence
that modification of the phosphate and/or hydrophobic
chain has significant effects on the ability of FAP
The tetradecyl-difluorophosphonate analogue was
synthesized (Scheme 3) in two steps using diethyl
difluoromethanephosphonate (17) as the starting mate-
rial.36 Compound 17 was treated with LDA at -78 °C
followed by reacting the anion with tetradecyl bromide
to give the protected phosphonate 18. Compound 18 was
deprotected using bromotrimethylsilane to yield the
required difluorophosphonate compound (19).

Lysophosphatidic Acid Receptor Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4921
Scheme 3a
transients or inhibited the endogenous purinergic recep-
tor response elicited by 10 µM ATP (data not shown).
The effects of unsaturation at different positions, modi-
fication of headgroup to phosphonate, difluoro phospho-
nate, and thiophosphate with/without unsaturation on
the activity of C-14 analogues at LPA1-3 receptors, are
shown in Figure 2. These modifications dramatically
changed the pharmacological properties of FAP on
LPA1-3 receptors. The monounsaturated FAP analogues
(12a-e) showed increased potency and/or efficacy when
compared to saturated analogues, except C-10 ana-
logues, without changing their properties as agonists
or antagonists at the LPA2 and LPA3 receptors (Table
1). The position of the double bond also had an impact
on the activity. Comparison of the activities between
n-decenyl regioisomers 12a and 12b suggests that the
C9dC10 double bond, as found in LPA 18:1, was pre-
ferred over C4dC5 in activating the LPA2 receptor (EC50
) 3800 nM for 12a versus >10000 nM for 12b). Though
12b (Ki ) 370 nM) was moderately more active than
12a (Ki ) 504 nM), the preference for the double bond
position was much less pronounced for inhibition of the
LPA3 receptor. Similarly, the LPA2 receptor showed
preference for C9dC10 unsaturation between the tet-
radecenyl isomers 12d (EC50 ) 397 nM) and 12e (EC50
) 4100 nM), and LPA3 showed no significant preference
for double bond position. In contrast, LPA1 preferred
C11dC12 over C9dC10 between 12d (Ki ) 1146 nM) and
12e (Ki ) 457 nM), indicating the possibility of a
differential conformational requirement in the side
chain for each of the three LPA receptors (Figure 2). In
the unsaturated series, only the tetradecenyl compounds
(12d, 12e) antagonized the LPA response at LPA1
receptor. This further supports our hypothesis that the
length of the side chain is critical for interaction with
LPA receptors.
a Reagents and conditions: (a) (i) LDA, -78 °C, THF; (ii)
14H29Br, 40%; (b) (i) TMSBr, CH2Cl2, 6 h, rt; (b) MeOH/H2O, 78%.
C
Biology. Compounds were tested for their ability to
induce or inhibit LPA-induced calcium transients in
RH7777 rat hepatoma cells stably expressing LPA1,
LPA2, and LPA3 receptors and in PC-3 cells that express
LPA1-3 endogenously using a FlexStation II automated
fluorometer (Molecular Devices, Sunnyvale, CA).27,33
The detailed protocol used is included in the Experi-
mental Section. Selected compounds were tested for
activation of human platelets17,18 and PPARγ activation
in CV1 cells, transfected with an acyl-coenzyme A
oxidase-luciferase (PPRE-Acox-Rluc) reporter gene con-
struct as previously reported.37 The two compounds with
pan-agonist and antagonist properties were assayed for
the effect on lysophospholipase D activity.38
Results and Discussion
According to our original two-point contact model,39,40
both a polar phosphate headgroup and a hydrophobic
tail are required for specific interactions with LPA
GPCR.27 On the basis of this model, we identified 2 and
PA 8:0 as selective LPA1 and LPA3 antagonists and FAP
as subtype-selective agonists/antagonists of LPA1-3
receptors.27,33 Bandoh et al. showed, using Sf-9 insect
cells transiently transduced by LPA GPCR, that LPA3
prefers unsaturated fatty acyl LPA species over satu-
rated LPAs.41 Replacement of the phosphate with a
phosphonate renders compounds metabolically stable
against degradation by lipid phosphate phosphatases.
The phosphonate modification also affects ligand-
receptor interactions by reducing charge density on the
polar headgroup. Phosphonate analogues of LPA have
been studied recently and are found to be less potent
than LPA.42,43 Alternatively, thiophosphate in place of
phosphate yielded metabolically stable compounds with
increased charge on the polar headgroup such as 3, a
selective LPA3 agonist.28,29
To explore the effects of these modifications along
with variations in the side chain of FAP structure, we
synthesized a series of FAP analogues with an unsat-
uration at different positions in the alkyl side chain
(12a-f), thiophosphates (15a-g), and phosphonates
(16a-d, 19). These new analogues were evaluated as
agonists and antagonists with respect to LPA1-3. Satu-
rated FAP analogues containing 10, 12, or 14 carbons
(9a-c) were previously shown to be the most effective
agonists and/or inhibitors at LPA1-3 in our initial
study.33 For this reason we synthesized and character-
ized modified FAP analogues with these optimum chain
lengths.
The replacement of phosphate with a thiophosphate
as the headgroup in 10-, 12-, and 14-carbon saturated
FAP analogues (15a-c) had a significant impact on
their agonist/antagonist properties at all three LPA
receptor subtypes. At LPA1, the thiophosphate modifica-
tion completely abolished the inhibitory effects of the
original FAP analogues. At LPA2 on the other hand, the
thiophosphate invariably increased the efficacy of the
original FAP to 100%. At the LPA3 receptor, the
saturated thiophosphate FAP analogues consistently
showed improved inhibition of the LPA response com-
pared to the original FAP. Dodecyl-thiophosphate (15b)
is the most potent agonist and antagonist in the
saturated thiophosphate analogues at LPA2 (EC50
)
1000 nM) and LPA3 (Ki ) 14 nM), respectively. These
results are consistent with our two-point contact model
as the increase in the charge density, influenced by the
properties of the hydrophobic tail, increased the agonist
or antagonist properties of the FAP.
Next, we investigated the effect of combining a
thiophosphate headgroup with monounsaturation (C9d
C10) in the side chain. The combination of the thiophos-
phate headgroup with C9dC10 unsaturation resulted in
analogues (15d-f), with properties that were a combi-
nation of the properties of saturated thiophosphates and
unsaturated FAP, substantially lowering the EC50 and
IC50 values. Similar to the saturated thio analogues,
compounds 15d-f were inactive at the LPA1 receptor.
Each FAP analogue was tested for the ability to
induce Ca2+ transients in RH7777 cells transfected with
LPA1-3 receptors (agonism), as well as the ability to
inhibit LPA-induced Ca2+ transients in the same cells
(antagonism) (Table 1). Every FAP analogue used in this
study was tested for activity in nontransfected RH7777
cells. None of the FAP analogues, when applied up to a
concentration of 30 µM, induced intracellular Ca2+

4922 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Durgam et al.
Table 1. Effects of FAP Analogues 12a-f, 15a-g, 16a-d, and 19 on LPA1-3 Transfected RH7777 Cells and Comparison of the
Activities with the Previously Reported Compounds
LPA1
LPA2
LPA3
PC-3
EC50
IC50
EC50
IC50
(Ki)
nM
EC50
IC50
(Ki)
nM
EC50
IC50
a
(Emax
)
(Ki)
(Emax
)
(Emax
)
(Emax
)
(Ki)
compd
X
Y
R
nM
nM
nM
nM
nM
nM
9ab
9bb
9cb
12a
12b
12c
12d
12e
12f
O
O (CH2)9CH3
O (CH2)11CH3
O (CH2)13CH3
NEc
NE
NE
NE
NE
1800 (82)
NE
NE
NE
NE
384 (121)
128 (61)
422 (211)
770 (504)
830 (370)
32 (27)
96 (58)
103 (40)
NE
122 (49)
28 (14)
162 (76)
340 (128)
27 (14)
28 (14)
NE
NDd ND
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
2800 (1354) 3100 (50)
2300 (1082) NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
546
(78)
NE
NE
ND
ND
NAe
NA
NA
NA
ND
ND
ND
1510 (574)
1300 (735)
916 (390)
241 (123)
ND
O (CH2)8CHdCH2
>10000
>10000
>10000
3000 (1146) 397 (58)
2200 (457) 4100 (75)
NE
NE
NE
NE
NE
NE
NE
NE
3800 (100) NE
>10000
717 (78)
O (CH2)3CHdCH(CH2)4CH3 NE
O (CH2)8CHdCHCH2CH3 NE
O (CH2)8CHdCH(CH2)3CH3 NE
O (CH2)10CHdCHCH2CH3 NE
NE
NE
NE
NE
NE
O (CH2)8CHdCH(CH2)7CH3 NE
NE
- (11) NA
15a
15b
15c
15d
15e
15f
S
S
S
S
S
S
S
(CH2)9CH3
NE
NE
NE
NE
NE
4570 (100) NE
1000 (100) NE
2500 (100) NE
>10000(56) NE
NA
NA
NE
NA
1220 (521)
(CH2)11CH3
2838 (1300)
NE
(CH2)13CH3
(CH2)8CHdCH2
(CH2)8CHdCHCH2CH3
(CH2)8CHdCH(CH2)3CH3 NE
(CH2)8CHdCH(CH2)7CH3 193
(80)
1000 (533)
677 (100)
480 (150)
244 (175)
NE
NE
NE
- (27) 2972 (1460)
- (40) 938 (397)
- (30) NA
15g
16a
16b
16c
16d
19
CH2 O (CH2)8CH3
CH2 O (CH2)10CH3
CH2 O (CH2)12CH3
CH2 O (CH2)16CH3
CF2 O (CH2)13CH3
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
1200 (680) NA
654 (303) NA
3100 (1120) NA
NE NE
1513 (575) ND
454 (202) ND
3122 (1500)
2638 (1270)
9674 (4620)
NE
ND
ND
∼10000
5500 (3550) NE
NE
NE
NE
NE
NE
2500 (788)f ∼10000 (40) NE
2g
5500
(4300)
WA
NE
NE
5h
6i
NE
NE
NE
NE
NE
NE
NE
NE
1186 (588) NA
301 (148) NA
WA
3384
(1740)
762
(425)
Ser-PA
1850 NE
(100)
NE
>10000
∼11000
NE
1600 (100) NE
- (42) NA
Tyr-PA
NE
NE
NE
5570 (2325) - (25) WAj
a Emax ) maximal efficacy of the drug/maximal efficacy of LPA 18:1, expressed as the percentage. b Previously reported in Virag et al.
(ref 33). c NE ) no effect was shown at the highest concentration (30 µM) tested. d ND ) not determined. e NA ) not applicable. f Partial
antagonist with 40% inhibition of the LPA response. g Reported Ki values of 2 are 106 nM and 6.6 µM at LPA3 and LPA1, respectively (ref
27). h Reported Ki values of 5 are 137 nM and 428 nM at LPA1 and LPA3, respectively (ref 31). i Reported Ki values of 6 are 250 nM, 360
nM, and 5.6 µM at LPA1, LPA3, and LPA2, respectively (ref 32). j WA ) weak antagonist.
The unsaturated C12 and C14 thiophosphates showed
increased potency at LPA2 with a minimal change in
the potency at LPA3 receptor as compared to their
saturated forms. The tetradec-9-enyl thiophosphate
(15f) compound showed a unique pharmacology. It
retained the features of the saturated thio analogues
at LPA1, as it had no effect on the LPA-induced Ca2+
mobilization. On the other hand, 15f was found to be
the best agonist at LPA2 (EC50 ) 480 nM) and most
potent antagonist at LPA3 (Ki ) 14 nM) among all C-10,
-12, and -14 thiophosphate analogues tested (Figures
2B and 2D). These differences in the ligand properties
of the thiophosphate analogues at the LPA receptor
subtypes may provide a practical advantage in improv-
ing subtype-selective agonists and antagonists, as short-
chain thiophosphates interact selectively with LPA2 and
LPA3 receptors.
Oleoyl-phosphate (12f), an unsaturated FAP analogue
of oleoyl-LPA, neither inhibited nor activated Ca2+
mobilization in cells expressing LPA1-3. However, it
potentiated LPA response at all three LPA receptors
when the two compounds were coapplied (data not
shown). This observation led us to the hypothesis that
by increasing the charge density of the 12f headgroup
by replacing the phosphate with a thiophosphate, we
may turn this analogue into an agonist. To test this
hypothesis, we synthesized and evaluated the oleoyl-
thiophosphate (15g) at LPA1-3 receptors. In agreement
with our prediction, compound 15g was a partial agonist
at LPA1, LPA3, and a potent and full agonist at LPA2
with the EC50 of 244 nM (Emax ) 175% of LPA response),
lower than that of oleoyl-LPA (EC50 ) 300 nM). The
dose-responses of 15g, comparing its effects with LPA
18:1 at LPA1-3 receptors, are shown in Figure 3.
The phosphonate analogues (16a-d) were weaker
inhibitors and agonists at the LPA receptors than their
phosphate counterparts, consistent with data reported
previously.42 However, tetradecyl-phosphonate (16c)
inhibited LPA-induced Ca2+ mobilization at all three
receptor subtypes with IC50 values in the micromolar
range, thus becoming, to our knowledge, the first pan-
antagonist of the LPA receptors (Figure 2). Compound
16c caused a rightward shift in the LPA dose-response
curve and reduced the maximal response elicited by 10
µM LPA in LPA1, LPA2, and LPA3 expressing RH7777
cells (Figure 4). Compound 16c is the only known
inhibitor of the LPA2 receptor subtype apart from 6 that
exerts only a modest and partial inhibition.32 Compound
16c, with a simpler structure and phosphonate head-
group, is presumably resistant to degradation by lipid

Lysophosphatidic Acid Receptor Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4923
Figure 2. The effects of modified C-14 analogues on RH7777 cells stably transfected with the LPA1-3 receptors. 200 nM of LPA
18:1 was coapplied with increasing concentrations of C-14 analogues to RH7777 cells stably expressing LPA1 and LPA3. Increasing
concentrations of various C-14 analogues were applied to measure their agonist properties at LPA2. Data points represent an
average of four measurements. (A) Inhibition of the LPA response by C-14 analogues at LPA1; (B) activation of LPA2 by C-14 FAP
analogues; (C) inhibition of the LPA response by C-14 phosphonate 16c at LPA2; (D) inhibition of the LPA response by C-14
analogues at LPA3.
phosphate phosphatases. These features make this
molecule an attractive lead structure for further devel-
opment of pan-antagonists for the LPA1-3 receptors. We
synthesized compound 19, which contains a difluoro-
phosphonate group as an isosteric replacement for
phosphonate in 16c, and tested at LPA1-3 receptors.
This compound retains the metabolic stability against
phosphatases and at the same time increases the acidity
of the phosphonate group, which presumably increases
receptor binding. Increasing the acidity of phosphonate
group by the addition of two fluorine atoms in compound
19 converted the compound from an antagonist to a
weak and partial agonist with an EC50 of 10 µM (Emax
) 40%) at the LPA2 receptor. Compound 19 showed
improved antagonist activity at LPA3 compared to 16c,
while it showed partial antagonism at LPA1.
All FAP antagonists have been tested for the revers-
ibility of the antagonist effect. In these experiments
RH7777 cells stably expressing LPA2 or LPA3 were
exposed to LPA followed by LPA plus the inhibitor at
30 µM, and finally to LPA after a washout period. The
LPA1 receptor that shows very rapid and substantial
desensitization could not be tested in this paradigm.
However, as shown in Figure 5, 16c exerted fully
reversible antagonism in LPA2 and LPA3. Similar
results were seen with the other FAP antagonists (data
not shown). These observations are in complete agree-
ment with our earlier report on FAP analogues.33
LPA was shown to activate mitogenic and motogenic
signaling in PC-3 cells.44 RT-PCR analysis of PC-3 cells,
an androgen-independent human prostate cancer cell
lines, showed expression of transcripts encoding all
three LPA receptors.45 We tested the FAP analogues in
PC-3 cells, which unlike the transfected RH7777 cells
endogenously express LPA1-3 receptors. Since PC-3 cells
express LPA1-3 receptors, the effects shown by the FAP
compounds (Table 1) represent the combination of the
effects of these compounds at the three LPA receptors.
These experiments confirmed the pharmacological prop-
erties of the FAP analogues obtained from RH7777 cells
expressing each LPA receptor individually. Thiophos-
phate analogues (15e and 15f) showed both independent
activation and inhibition of LPA-induced Ca2+ tran-
sients in PC-3 cells, as they have different effects at each
of the LPA1-3 receptors. Oleoyl-thiophosphate (15g)
showed a maximal response of 30% of the maximal LPA
response, with no inhibition of LPA response. These
results are in consistent with the data obtained in
transfected RH7777 cells. The consistency of the results
obtained from PC-3 cells that endogenously express LPA
receptors with those results obtained using transfected
RH7777 cells validates our assay systems.

4924 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Durgam et al.
Figure 4. The effect of increasing concentrations of compound
16c on the LPA dose-response curve in RH7777 cells stably
transfected with LPA1-3 receptors. LPA-alone (closed squares)
and LPA coapplied (closed trianges and open squares) with
different concentrations of compound 16c. Each point repre-
sents the mean ( s.d. for n ) 3. Compound 16c caused a
rightward shift in the dose-response curve increasing the
apparent EC50. It also caused a modest inhibition of the
maximal response to LPA.
Figure 3. Oleoyl-thiophosphate (15g) is an agonist at LPA1,
LPA2, and LPA3 receptors expressed in RH7777 cells. Intra-
cellular Ca2+ transients were measured in response to the
application of increasing concentrations of 15g and compared
to transients elicited by the corresponding amount of LPA 18:
1. Data points represent the average of four measurements.
Dose-response relationships for LPA 18:1 and 15g in RH7777
cells expressing LPA1 (A), LPA2 (B), and LPA3 (C).
antagonist 16c, all compounds tested were agonists and
elicited platelet shape change, which was reversible
after 5 min (Table 2). The most potent compound was
the pan-agonist 15g, which induced shape change with
an EC50 of 16 nM, only about 2-times higher than the
EC50 of LPA 18:1. Compounds 12c and 15e, selective
agonists of LPA2, activated platelets with much lower
potency, EC50 of 1.5 µM and 1.0 µM, respectively, than
the pan-agonist 15g. These results suggest the presence
of functionally active LPA2 receptors on platelets and
also indicate that LPA2 is not as important as LPA1 and
LPA3 in platelet activation.
LPA has been shown to activate platelets, and this
effect is blocked by LPA GPCR antagonists.17 LPA1,
LPA2, and LPA3 have been detected in platelets by RT-
PCR. With the exception of the LPA receptor pan-

Lysophosphatidic Acid Receptor Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4925
receptor desensitization is reversible, whereas LPA
receptor antagonism is not as long as the antagonist is
present. Therefore, platelets were incubated for 30 min
with LPA 16:0, the pan-agonist 15g or the LPA1/3
antagonist 9c (which is an agonist in platelets). Indeed,
after 30 min of incubation, platelets incubated with LPA
16:0 were again fully responsive to LPA, whereas the
LPA response was inhibited in platelets incubated with
15g or 9c. The IC50 of 15g and 9c after 30 min
incubation were 0.016 ( 0.01 µM and 0.3 ( 0.07 µM
(mean ( SD, n ) 3), respectively, which were similar
to the IC50 values obtained after 5 min incubation (Table
2). There are two alternative explanations for these
results: LPA receptors became desensitized or inhibited.
Thus, the fatty alcohol phosphates and fatty alcohol
thiophosphates might inhibit platelet LPA receptors
through either receptor desensitization or antagonist
action.
To compare the effects of these FAP analogues at LPA
receptors with other available agonists and antagonists,
we tested compounds 2, 5, 6, Ser-PA, and N-acyl
tyrosine phosphoric acid (Tyr-PA), in our RH7777 cell
system. This comparison, where a single test system is
used for all compounds, has the benefit of providing
reliable information on the relative effectiveness of these
compounds despite the inherent shortcomings the in-
dividual test systems may have. Our results were
consistent with previously published data for 2, 6, and
Ser-PA. However, we encountered differences for com-
pounds 5 and Tyr-PA (Table 1). Compound 2 was
identified by our group as a subtype-selective inhibitor
for LPA3 and LPA1, with Ki values of 106 nM and 6.6
µM, respectively.27 To test our high throughput screen-
ing system, we evaluated the effects of 2 in the same
stably transfected RH7777 cell lines. The Ki values were
202 nM for LPA3 and 4.3 µM for LPA1 (Table 1). These
results convincingly showed the reproducibility of the
results of 2, even after the modification of the original
assay method. Compound 6 was synthesized and identi-
fied as a subtype-selective antagonist for LPA1 and LPA3
with a very weak inhibitory effect on LPA2 with Ki
values 250 nM, 360 nM, and 5.6 µM, respectively, using
GTPγS loading assay in HEK293T cells transfected with
LPA receptors.32 When this compound was tested in our
intracellular Ca2+ monitoring system, we obtained
similar Ki values for LPA1 (425 nM) and LPA3 (148 nM);
however, 6 seemed to inhibit LPA3 better compared to
LPA1 (Table 1). Ser-PA and Tyr-PA were originally
identified as inhibitors of LPA-induced platelet aggrega-
tion25 and inhibitors of the LPA-induced Cl- current in
Xenopus oocytes.26 In a mammalian cell line, however,
Ser-PA was found to be an LPA-like agonist.24 It was
also shown to be an agonist at LPA1 and LPA2 when
these receptor subtypes were heterologously expressed
in TAg-Jurkat T-cells.4 In our experiments, Ser-PA was
a full agonist at LPA1 (EC50 ) 1.85 µM), but only a weak
agonist at LPA2. At LPA3, Ser-PA was also a weak but
full agonist with an EC50 value of 1.6 µM (Table 1).
Figure 5. Reversible inhibition of the LPA2 and LPA3
response by compound 16c. LPA2 (A) or LPA3 (B) cells were
exposed to repeated concentrations of LPA or LPA plus 30 µM
16c. Compound 16c completely inhibited the response, which
was fully recovered after a wash for both receptor types. The
traces are representative of at least three experiments.
Table 2. Effect of Fatty Alcohol Phosphates and Fatty Alcohol
Thiophosphates on Induction and Inhibition of Platelet Shape
Change
agonist activity, antagonist activity,
compd
Emax,a % EC50 (SD (µM)
IC50 (SD (µM)
n
9b
94 ( 3
87 ( 9
0.25 ( 0.2
0.16 ( 0.05
0.42 ( 0.28
(0.3 ( 0.07)b
0.7 ( 0.3
3
5
3
3
4
4
3
3
3
4
9c
0.43 ( 0.18
12c
15e
15g
89 ( 13
73 ( 3
112 ( 11
1.5 ( 0.8
1.0 ( 0.02
0.016 ( 0.005
0.7 ( 0.26
0.018 ( 0.012
(0.016 ( 0.01)b
33 ( 8.0
16c
0
inactive
0.010 ( 0.006
0.009 ( 0.006
LPA 16:0 120
LPA 18:1 120
0.017 ( 0.010
NDc
a Emax ) maximal shape change induced by test compound/
shape change induced by 0.025 µM LPA 16:0. Shape change
induced by 0.025 µM LPA was 82 ( 6% (mean ( SD, n ) 7) of
maximal and set to 100%. Values are mean ( SD from different
experiments with different platelet donors. Inhibitory activity was
tested 5 min after addition of the test compounds, i.e., after the
shape change induced by the compounds became reversible. b IC50
values after 30 min of incubation with the test compound. c ND )
not determined.
All the compounds tested inhibited shape change
induced by 25 nM LPA 16:0. The IC50 values of the fatty
alcohol phosphates and fatty alcohol thiophosphates
were in the range of the EC50 values. Since platelet LPA
receptors show a homologous desensitization 5 min after
activation,46 it is possible that the fatty alcohol phos-
phates and fatty alcohol thiophosphates might exert
their inhibition through desensitizing LPA receptors
rather than receptor antagonism. To distinguish be-
tween these two possibilities, we reasoned that LPA
An et al. showed that Tyr-PA did not affect LPA
signaling at LPA1 and LPA2 receptors when applied at
a concentration of 1 µM.4 Tyr-PA in our experiments
had no effect on LPA1; however, it was found to be a
weak agonist at LPA2 (EC50 ) 11 µM) and an inhibitor
at LPA3 (Ki ) 2.3 µM) as shown in Table 1. Compound

4926 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Durgam et al.
Figure 7. Concentration-dependent inhibition of autotaxin
(ATX) phosphodiesterase activity by compounds 15g and 16c,
as compared to LPA18:1. ATX activity was measured by the
hydrolysis of the fluorescent substrate CPF4 (2 µM), as
described in ref 38.
Figure 6. Results of in vitro PPARγ activation by selected
compounds in CV1 cells transfected with PPARγ and PPRE-
Acox-Rluc reporter gene, comparing the effects with rosigli-
tazone, a known PPARγ agonist. CV1 cells were treated with
1% DMSO or 10 µM of test compound dissolved in DMSO for
20 h. Luciferase and â-galactosidase activities (mean ( SEM)
were measured in the cell lysate (n ) 4). *P < 0.05 and **P <
0.01, significant differences over vehicle control.
agonists (3) and antagonists (2, 5, and 6), FAP ana-
logues, regardless of their LPA1-3 agonist/antagonist
activities, can activate PPRE-Acox-Rluc reporter. These
results are consistent with previously reported results,37
in that LPA GPCR ligands can activate PPARγ. Ηow-
ever, the results also emphasize that the SAR of PPARγ
activation is different from GPCR.
Recent evidence shows that autotaxin, the major LPA-
producing exo-phosphodiesterase/lysophospholipase D,
exhibits product inhibition by LPA.38 For this reason,
we tested the pan-agonist 15g and the pan-antagonist
16c for their effects on autotaxin phosphodiesterase
activity using a fluorescence-based biosensor.38 The
results showed that both analogues inhibited the en-
zyme, although 15g was more effective than 16c and
even slightly more potent than LPA 18:1 itself (Figure
7). Compound 16c exerted only a partial inhibition
whereas 15g showed a complete dose-dependent auto-
taxin inhibition at micromolar concentrations.
5 is a 2-substituted analogue of the NAEPA that was
identified as a subtype-selective inhibitor of the LPA1
and LPA3 receptors, using a GTPγS-loading assay with
cell membranes isolated from LPA1-, LPA2-, or LPA3-
expressing HEK293T cells. Compound 5 was a better
antagonist at LPA1 (Ki ) 137 nM) than at LPA3 (Ki )
428 nM).31 In our experiments however 5 was only a
weak inhibitor at LPA1 and a better inhibitor at LPA3
with a Ki value of 588 nM (Table 1). This value is in
reasonable agreement with the published data in ad-
dition to the observation that 5 did not affect LPA
signaling through LPA2 (Table 1). Analogous to the FAP,
these compounds also showed effects that are combina-
tion of effects at three LPA receptors on PC-3 cells,
further validating our assay system.
In addition to its plasma membrane receptors, LPA
was shown to be an agonist of the nuclear transcription
factor PPARγ.11 Many agents have been reported to
activate PPARγ, including the thiazolidinedione family
represented by rosiglitazone, oxidized phospholipids,
fatty acids, eicosanoids, and oxidized LDL. Zhang et al.
showed that unsaturated and alkyl ether analogues of
LPA, 1,1-difluorodeoxy-(2R)-palmitoyl-sn-glycero-3-phos-
phate (7), its monofluoro analogue 8, and the oxidized
phosphatidylcholine 1-O-hexadecyl-2-azeleoyl-phosphati-
dylcholine, induced neointima formation, an early step
leading to the development of atherogenic plaques,
through PPARγ activation.37 The SAR of neointima
formation by LPA analogues in vivo was identical to
PPARγ activation in vitro and different from LPA
G-protein-coupled receptors.37 We tested selected com-
pounds including FAP-12 (9b), unsaturated thiophos-
phate analogues (15d-g), tetradecyl-phosphonate 16c,
previously reported LPA1/LPA3 antagonists 2, 5, 6, and
the selective LPA3 agonist 3, for PPARγ activation in
vitro in CV1 cells using the PPRE-Acox-Rluc reporter
gene assay. Interestingly, results from this assay (Fig-
ure 6) indicate that along with previously reported
Conclusions
The present study extended our previously described
two-point contact model as the minimal requirement to
elicit specific interactions with LPA GPCR and provides
further refinement of the minimal pharmacophore FAP
by identifying modifications that allowed the synthesis
of a pan-agonist and a pan-antagonist and several
subtype-selective ligands. A systematic SAR study of the
FAP pharmacophore with phosphonate, thiophosphate,
and introduction of unsaturation in the side chain
outlined important principles for the design of subtype-
selective LPA receptor agonists and antagonists. The
results of the FAP analogues, and previously reported
LPA agonists and antagonists by other groups, obtained
in transfected RH7777 cells expressing individual LPA
receptors were consistent with results obtained in PC-3
cells that endogenously express LPA1-3 receptors. In
addition to their ligand properties on LPA GPCR, we
showed that FAP also activate nuclear transcription
factor PPARγ with an SAR distinct from LPA GPCR.
Based on the principles that emerged from the SAR of

Lysophosphatidic Acid Receptor Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4927
Phosphoric acid di-tert-butyl ester dec-4-enyl ester
9b, oleoyl-thiophosphate (15g) was synthesized and
identified as a novel pan-agonist at all three LPA
receptors. Tetradecyl-phosphonate (16c) was identified
as a metabolically stable pan-antagonist that could
serve as a lead structure for further development of
LPA1-3 receptor antagonists that are stable against
degradation by lipid phosphate phosphatases. Our
results provide the first comprehensive evaluation of
LPA GPCR ligands as activators of PPARγ. It was an
unexpected surprise that all analogues tested, with the
exception of 5, were PPARγ activators, regardless of
whether they were agonists or antagonists of LPA
GPCR. Therefore, future synthesis must focus on de-
veloping LPA GPCR ligands that lack PPARγ agonist
properties. Finally, two of our most interesting lead
compounds, regardless of their agonist or antagonist
properties on LPA GPCR and PPARγ, inhibited auto-
taxin/lysophospholipase D. Thus, the FAP analogues we
have synthesized have multiple biological targets and
serve as leads for future structural modifications to
develop receptor type-, subtype-, and autotaxin-selective
agents.
1
(11b): Isolated as clear oil (68% yield). H NMR (CDCl3): δ
5.15-5.30 (m, 2H), 3.84 (q, J ) 6.8 Hz, 2H), 2.05 (q, J ) 7.0
Hz, 2H), 1.98 (q, J ) 6.8 Hz, 2H), 1.58-1.62 (m, 2H), 1.42 (s,
18H), 1.22 (br s, 6H), 0.80 (t, J ) 7.2 Hz, 3H); 31P NMR (CD3-
OD): δ 7.90 (s); MS: [M + 23Na] at m/z 371.3.
Phosphoric acid di-tert-butyl ester dodec-9-enyl ester
1
(11c): Isolated as clear oil (70% yield). H NMR (CDCl3): δ
5.15-5.45 (m, 2H), 3.88 (q, J ) 6.6 Hz, 2H), 1.92-1.99 (m,
4H), 1.57-1.61 (m, 2H), 1.42 (s, 18H), 1.24 (br s, 10H), 0.89
(t, J ) 7.5 Hz, 3H); 31P NMR (CDCl3): δ 7.80 (s); MS: [M +
23Na] at m/z 399.5.
Phosphoric acid di-tert-butyl ester tetradec-9-enyl
ester (11d): Isolated as clear oil (68% yield). 1H NMR
(CDCl3): δ 5.34 (t, J ) 5.2 Hz, 2H), 3.94 (q, J ) 6.6 Hz, 2H),
2.01 (m, 4H), 1.60-1.69(m, 2H), 1.48 (s, 18H), 1.30 (br s, 18H),
0.90 (t, J ) 7.4 Hz, 3H); 31P NMR (CDCl3): δ 7.90 (s); MS: [M
+
23Na] at m/z 427.4.
Phosphoric acid di-tert-butyl ester tetradec-11-enyl
ester (11e): Isolated as clear oil (82% yield). 1H NMR
(CDCl3): δ 5.28-5.40 (m, 2H), 3.94 (q, J ) 6.5 Hz, 2H), 1.98-
2.06 (m, 4H), 1.62-1.70 (m, 2H), 1.48 (s, 18H), 1.23 (br s, 14H),
0.95 (t, J ) 7.4 Hz, 3H); 31P NMR (CDCl3): δ 8.10 (s); MS: [M
+
23Na] at m/z 427.4.
Phosphoric acid di-tert-butyl ester octadec-9-enyl
ester (11f): Isolated as clear oil (72% yield). 1H NMR
(CDCl3): δ 5.28-5.44 (m, 2H), 3.94 (q, J ) 6.9 Hz, 2H), 1.90-
2.10 (m, 4H), 1.60-1.70 (m, 2H), 1.48 (s, 18H), 1.28 (br s, 22H),
0.88 (t, J ) 6.6 Hz, 3H); 31P NMR (CDCl3): δ 8.10 (s); MS: [M
Experimental Section
General Methods. All reagents were purchased from
Sigma-Aldrich Chemical Co., Fisher Scientific (Pittsburgh,
PA), Bedukian Research (Danbury, CT), and Toronto Research
Chemicals (North York, ON, Canada) and were used without
further purification. Phosphonate analogues were purchased
from Lancaster (Pelham, NH; n-decyl-phosphonate (16a)),
PolyCarbon (Devens, MA; n-dodecyl-phosphonate (16b)), Alfa
Aesar (Ward Hill, MA; n-tetradecyl-phosphonate (16c), and
n-octadecyl-phosphonate (16d)). LPA 18:1, 2, Ser-PA, and Tyr-
PA were obtained from Avanti Polar Lipids (Alabaster, AL).
Melting points were determined on a Thomas-Hoover capillary
melting point apparatus and are uncorrected. Routine thin-
layer chromatography (TLC) was performed on 250 µm glass-
backed Uniplates (Analtech, Newark, DE). Flash chromatog-
raphy was performed on pre-packed silica gel columns using
a Horizon HPFC system (Biotage, Charlottesville, VA). 1H and
31P NMR spectra were obtained on a Bruker AX 300 (Billerica,
+
23Na] at m/z 483.5.
General Procedure for Synthesis of Phosphoric Acid
Monoalkenyl Esters (12a-f). To a solution of 100 mg of
11a-f in anhydrous methylene chloride (20 mL) was added
trifluroacetic acid (0.3 mL). The mixture was allowed to stir
for 4 h, and TLC showed the completion of the reaction.
Solvents were evaporated, and the residue was washed with
methylene chloride (2 × 20 mL) and concentrated under
vacuum to yield the desired phosphoric acid monoalkenyl
esters (12a-f) as colorless oils.
Phosphoric acid monodec-9-enyl ester (12a): Isolated
as an oil (85%). 1H NMR (CD3OD): δ 5.52-5.80 (m, 1H), 4.80-
4.95 (m, 2H), 3.90 (q, J ) 6.6 Hz, 2H), 2.01 (q, J ) 6.9 Hz,
2H), 1.53-1.70 (m, 2H), 1.28 (br s, 10H); 31P NMR (CD3OD):
δ 17.84 (s); MS: [M - H]- at m/z 235.2. Anal. (C10H21O4P‚
0.1H2O) C, H.
Phosphoric acid monodec-4-enyl ester (12b): Isolated
as an oil (78%). 1H NMR (CD3OD): δ 5.25.35 (m, 2H), 3.84 (q,
J ) 6.8 Hz, 2H), 2.05 (q, J ) 7.0 Hz, 2H), 1.98 (q, J ) 6.8 Hz,
2H), 1.58-1.65 (m, 2H), 1.22 (br s, 6H), 0.80 (t, J ) 7.2 Hz,
3H); 31P NMR (CD3OD): δ 17.45 (s); MS: [M - H]- at m/z
235.2. Anal. (C10H21O4P‚0.5H2O) C, H.
1
MA) spectrometer. Chemical shifts for H NMR are reported
as parts per million (ppm) relative to TMS. Chemical shifts
for 31P NMR are reported as parts per million (ppm) relative
to 0.0485 M triphenyl phosphate in CDCl3. Mass spectral data
was collected on a Bruker ESQUIRE electrospray/ion trap
instrument in the positive and negative ion modes. Elemental
analyses were performed by Atlantic Microlab Inc., Norcross,
GA.
Phosphoric acid monododec-9-enyl ester (12c): Iso-
1
lated as an oil (82%). H NMR (DMSO-d6/CD3OD): δ 5.20-
General Procedure for Synthesis Phosphoric Acid Di-
tert-butyl Ester Alkenyl Esters (11a-f). To a stirred
solution of alcohol (10a-f, 2.5 mmol) and di-tert-butyl N,N-
diisopropylphosphoramidite (1.51 g, 4 mmol) in anhydrous
methylene chloride (60 mL) was added 1H-tetrazole (578 mg,
8.25 mmol). After 30 min of stirring the mixture was cooled
to 0 °C and 0.3 mL of 50% hydrogen peroxide was added. The
mixture was stirred for 1 h, diluted with methylene chloride
(100 mL), and washed with10% sodium metabisulfite (2 × 50
mL), saturated sodium bicarbonate (2 × 50 mL), water (50
mL), and brine (50 mL). The organic layer was dried over
anhydrous sodium sulfate, filtered, and concentrated under
vacuum. The resulting crude products were purified by silica
gel chromatography using hexane/ethyl acetate (7:3) to elute
the desired products (11a-f).
5.40 (m, 2H), 3.82 (q, J ) 6.6 Hz, 2H), 1.88-2.10 (m, 4H), 1.50-
1.62 (m, 2H), 1.25 (br s, 10H), 0.88 (t, J ) 7.2 Hz, 3H); 31P
NMR (CD3OD): δ 16.22 (s); MS: [M - H]- at m/z 263.0. Anal.
(C12H25O4P‚0.6H2O) C, H.
Phosphoric acid monotetradec-9-enyl ester (12d): Iso-
1
lated as an oil (84%). H NMR (CDCl3/CD3OD): δ 5.12-5.30
(m, 2H), 3.84 (q, J ) 6.5 Hz, 2H), 1.81-2.10 (m, 4H), 1.48-
1.60 (m, 2H), 1.20 (br s, 14H), 0.78 (m, 3H); 31P NMR (CD3-
OD): δ 16.20 (s); MS: [M - H]- at m/z 291.4. Anal. (C14H29O4P‚
0.25H2O) C, H.
Phosphoric acid monotetradec-11-enyl ester (12e):
1
Isolated as an oil (78%). H NMR (CD3OD): δ 5.18-5.30 (m,
2H), 3.88 (q, J ) 6.6 Hz, 2H), 1.90-2.00 (m, 4H), 1.58-1.62
(m, 2H), 1.25 (br s, 14H), 0.86 (t, J ) 7.1 Hz, 3H); 31P NMR
(CD3OD): δ 16.20 (s); MS: [M - H]- at m/z 291.3. Anal.
(C14H29O4P) C, H.
Phosphoric acid di-tert-butyl ester dec-9-enyl ester
1
(11a): Isolated as clear oil (75% yield). H NMR (CDCl3): δ
5.73-5.85 (m, 1H), 4.90-5.02 (m, 2H), 3.95 (q, J ) 7.5 Hz,
2H), 2.03 (q, J ) 7.1 Hz, 2H), 1.60-1.69 (m, 2H), 1.48 (s, 18H),
1.30 (br s, 10H); 31P NMR (CDCl3): δ 7.90 (s); MS: [M +
23Na] at m/z 371.3.
Phosphoric acid monooctadec-9-enyl ester (12f): Iso-
lated as an oil (86%). 1H NMR (CD3OD): δ 5.25-5.40 (m, 2H),
3.91 (q, J ) 6.6 Hz, 2H), 1.90-2.05 (m, 4H), 1.55-1.70 (m,
2H), 1.26 (br s, 22H), 0.86 (t, J ) 6.0 Hz, 3H); 31P NMR (CD3-

4928 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Durgam et al.
OD): δ 16.21 (s); MS: [M - H]- at m/z 347.4. Anal. (C18H37O4P‚
Thiophosphoric acid O-dodecyl ester (15b): Isolated
as light yellow colored oil (73% yield). 1H NMR (DMSO-d6): δ
3.54-3.80 (m, 2H), 1.51-1.64 (m, 2H), 1.24 (br s, 18H), 0.83
(t, J ) 6.9 Hz, 3H); 31P NMR (DMSO-d6): δ 78.55 (s); MS: [M
- H]- at m/z 280.9. Anal. (C12H27O3PS‚0.5H2O) C, H.
Thiophosphoric acid O-tetradecyl ester (15c): Isolated
as light yellow colored oil (70% yield). 1H NMR (DMSO-d6): δ
3.78-3.92 (m, 2H), 1.50-1.62 (m, 2H), 1.24(br s, 22H), 0.85
(t, J ) 6.0 Hz, 3H); 31P NMR (DMSO-d6): δ 77.54 (s); MS: [M
- H]- at m/z 309.4. Anal. (C14H31O3PS‚0.25H2O) C, H.
Thiophosphoric acid O-dec-9-enyl ester (15d): Isolated
as light yellow colored oil (76% yield). 1H NMR (DMSO-d6): δ
5.70-5.90 (m, 1H), 4.85-5.05 (m, 2H), 3.75-3.98 (m, 2H), 2.01
(q, J ) 6.6 Hz, 4H), 1.48-1.68 (m, 2H), 1.26 (br s, 10H); 31P
NMR (DMSO-d6): δ 77.75 (s); MS: [M - H]- at m/z 251.1.
Anal. (C10H21O3PS) C, H.
0.4H2O) C, H.
General Method for Synthesis of Thiophosphoric Acid
O,O′-Bis-(2-cyanoethyl) Ester O′′-Alkyl/Alkenyl Esters
(14a-g). A solution of alcohol (13a-g, 2.0 mmol), bis-(2-
cyanoethyl) N,N-diisopropylphosphoramidite (1.085 g, 4 mmol)
and 1H-tetrazole (420 mg, 6 mmol) in anhydrous methylene
chloride (40 mL) was stirred for 30 min at room temperature,
followed by the addition of elemental sulfur (200 mg), and the
mixture was refluxed for 2 h. The reaction mixture was cooled
to room temperature, and solvents were evaporated under
vacuum. Addition of ethyl acetate (30 mL) precipitated excess
sulfur, which was filtered out, and the solvent was evaporated
to give the crude mixture. The mixture was purified by flash
chromatography using hexanes/ethyl acetate (8:1) to give the
desired products (14a-g) as colorless oils.
Thiophosphoric acid O-dodec-9-enyl ester (15e): Iso-
lated as light yellow colored oil (80% yield). 1H NMR (DMSO-
d6): δ 5.20-5.38 (m, 2H), 3.85 (q, J ) 6.6 Hz, 2H), 1.90-2.05
(m, 4H), 1.50-1.62 (m, 2H), 1.26 (br s, 10H), 0.91 (t, J ) 7.5
Hz, 3H); 31P NMR (DMSO-d6): δ 79.50 (s); MS: [M - H]- at
m/z 279.5. Anal. (C12H25O3PS‚0.35H2O) C, H.
Thiophosphoric acid O,O′-bis-(2-cyanoethyl) ester O′′-
decyl ester (14a): Isolated as colorless oil (72% yield). 1H
NMR (CDCl3): δ 4.21-4.35 (m, 4H), 4.10-4.13 (m, 2H), 2.8
(t, J ) 6.3 Hz, 4H), 1.62-1.78 (m, 2H), 1.26 (br s, 14H), 0.88
(t, J ) 6.0 Hz, 3H); 31P NMR (CDCl3): δ 84.56 (s); MS: [M +
23Na] at m/z 383.4.
Thiophosphoric acid O-tetradec-9-enyl ester (15f):
Isolated as light yellow colored oil (72% yield). 1H NMR
(DMSO-d6): δ 5.28-5.45 (m, 2H), 3.76-3.96 (m, 2H), 1.90-
2.08 (m, 4H), 1.50-1.62 (m, 2H), 1.26 (br s, 14H), 0.86 (t, J )
6.9 Hz, 3H); 31P NMR (DMSO-d6): δ 78.50 (s); MS: [M - H]-
at m/z 307.5. Anal. (C14H29O3PS‚0.3H2O) C, H.
Thiophosphoric acid O,O′-bis-(2-cyanoethyl) ester O′′-
dodecyl ester (14b): Isolated as colorless oil (84% yield). 1H
NMR (CDCl3): δ 4.26-4.33 (m, 4H), 4.07-4.19 (m, 2H), 2.8
(t, J ) 6.2 Hz, 4H), 1.65-1.80 (m, 2H), 1.26 (br s, 14H), 0.88
(t, J ) 6.6 Hz, 3H); 31P NMR (CDCl3): δ 84.55 (s); MS: [M +
23Na] at m/z 411.4.
Thiophosphoric acid O-octadec-9-enyl ester (15g):
Isolated as light yellow colored oil (82% yield). 1H NMR
(DMSO-d6): δ 5.28-5.42 (m, 2H), 3.75-3.95 (m, 2H), 1.91-
2.05 (m, 4H), 1.51-1.62 (m, 2H), 1.24 (br s, 22H), 0.85(t, J )
6.9 Hz, 3H); 31P NMR (DMSO-d6): δ 79.75 (s); MS: [M - H]-
at m/z 363.5. Anal. (C18H37O3PS‚0.3H2O) C, H.
Thiophosphoric acid O,O′-bis-(2-cyanoethyl) ester O′′-
1
tetradecyl ester (14c): Isolated as clear oil (82% yield). H
NMR (CDCl3): δ 4.25-4.33 (m, 4H), 4.08-4.17 (m, 2H), 2.8
(t, J ) 6.0 Hz, 4H), 1.65-1.80 (m, 2H), 1.26 (br s, 18H), 0.88
(t, J ) 6.6 Hz, 3H); 31P NMR (CDCl3): δ 84.56 (s); MS: [M +
23Na] at m/z 439.5.
(1,1-Difluoropentadecyl)-phosphonic Acid Diethyl Es-
ter (18). To a solution of diethyl difluoromethanephosphonate
(17, 1.0 g, 5.316 mmol) in anhydrous THF (50 mL) was added
2 M LDA (626 mg, 5.847 mmol) at -78 °C, and the mixture
was stirred for 30 min. Tetradecyl bromide (1.474 g, 5.316
mmol) in anhydrous THF (10 mL) was added to the mixture
at -78 °C, and the reaction mixture was stirred overnight at
room temperature. THF was evaporated and the residual oil
was purified by flash chromatography using hexanes/ethyl
acetate (7:3) as eluent to give 817 mg (40%) of compound 18
as colorless oil. 1H NMR (CDCl3): δ 4.21-4.31 (m, 4H), 1.80-
2.10 (m, 2H), 1.51-1.64 (m, 2H), 1.37 (t, J ) 6.9 Hz, 6H), 1.25
(br s, 22H), 0.87 (t, J ) 6.6 Hz, 3H); 31P NMR (CDCl3): δ 22.55
(t, J ) 90.96 Hz); MS: [M + 23Na] at m/z 407.2.
(1,1-Difluoro-pentadecyl)-phosphonic Acid (19). To a
solution of vacuum-dried 18 (225 mg, 0.585 mmol) in anhy-
drous methylene chloride (5 mL) was added bromotrimethyl-
silane (895 mg, 5.85 mmol), and the mixture was stirred at
room temperature for 6 h. Solvents were removed under
reduced pressure, and the residue was stirred in 95% methanol
in water (3 mL) for 1 h. The mixture was concentrated under
reduced pressure and dried under vacuum to give 150 mg
(78%) of 19 as light yellow solid. mp 66-69 °C; 1H NMR (CD3-
OD): δ 1.80-2.06 (m, 2H), 1.42-1.58 (m, 2H), 1.24 (br s, 22H),
0.90 (t, J ) 6.6 Hz, 3H); 31P NMR (CD3OD): δ 23.85 (t, J )
107.18 Hz); MS: [M - H]- at m/z 327.3. Anal. (C15H31F2O3P‚
0.2H2O) C, H.
Thiophosphoric acid O,O′-bis-(2-cyanoethyl) ester O′′-
1
dec-9-enyl ester (14d): Isolated as clear oil (76% yield). H
NMR (CDCl3): δ 5.72-5.91 (m, 1H), 4.88-5.05 (m, 2H), 4.22-
4.32 (m, 4H), 4.07-4.17 (m, 2H), 2.8 (t, J ) 6.3 Hz, 4H), 2.01
(t, J ) 6.6 Hz, 4H), 1.65-1.78 (m, 2H), 1.31 (br s, 10H); 31P
NMR (CDCl3): δ 84.58 (s); MS: [M + 23Na] at m/z 381.3.
Thiophosphoric acid O,O′-bis-(2-cyanoethyl) ester O′′-
dodec-9-enyl ester (14e): Isolated as clear oil (80% yield).
1H NMR (CDCl3): δ 5.26-5.46 (m, 2H), 4.25-4.33 (m, 4H),
4.08-4.17 (m, 2H), 2.8 (t, J ) 6.0 Hz, 4H), 2.07 (m, 2H), 1.62-
1.76 (m, 2H), 1.31 (br s, 10H), 0.96 (t, J ) 7.5 Hz, 3H); 31P
NMR (CDCl3): δ 84.55 (s); MS: [M + 23Na] at m/z 409.5.
Thiophosphoric acid O,O′-bis-(2-cyanoethyl) ester O′′-
tetradec-9-enyl ester (14f): Isolated as clear oil (75% yield).
1H NMR (CDCl3): δ 5.33-5.40 (m, 2H), 4.25-4.33 (m, 4H),
4.08-4.18 (m, 2H), 2.78 (t, J ) 6.0 Hz, 4H), 2.02 (m, 2H), 1.64-
1.80 (m, 2H), 1.31 (br s, 14H), 0.90 (t, J ) 7.2 Hz, 3H); 31P
NMR (CDCl3): δ 84.57 (s); MS: [M + 23Na] at m/z 437.5.
Thiophosphoric acid O,O′-bis-(2-cyanoethyl) ester O′′-
octadec-9-enyl ester (14 g): Isolated as clear oil (72% yield).
1H NMR (CDCl3): δ 5.31-5.42 (m, 2H), 4.27-4.31 (m, 4H),
4.09-4.21 (m, 2H), 2.78 (t, J ) 6.0 Hz, 4H), 2.02 (m, 2H), 1.65-
1.78 (m, 2H), 1.27 (br s, 22H), 0.88 (t, J ) 7.2 Hz, 3H); 31P
NMR (CDCl3): δ 84.56 (s); MS: [M + 23Na] at m/z 493.5.
General Procedure for Preparation of Thiophospho-
ric Acid O-Alkyl/Alkenyl Esters (15a-g). A solution of 100
mg of 14a-g in 1 M methanolic KOH (10 mL) was stirred for
2 h, and TLC showed the completion of the reaction. The
solvent was evaporated to give the crude product, which was
dissolved in water (20 mL) and acidified with HCl. The
aqueous mixture was extracted with ethyl acetate (2 × 50 mL),
and the organic layer was dried over sodium sulfate and
concentrated under vacuum to give the desired compound
(15a-g) as light yellow colored oil.
Thiophosphoric acid O-decyl ester (15a): Isolated as
light yellow colored oil (80% yield). 1H NMR (DMSO-d6): δ
3.75-3.95 (m, 2H), 1.50-1.65 (m, 2H), 1.24 (br s, 14H), 0.86
(t, J ) 6.0 Hz, 3H); 31P NMR (DMSO-d6): δ 77.50 (s); MS: [M
- H]- at m/z 253.2. Anal. (C10H23O3PS) C, H.
High Throughput Ca2+ Measurements. RH7777 cells
stably expressing either LPA1, LPA2, or LPA327,33 or PC-3 cells
were plated on poly-D-lysine-coated black-wall clear-bottom 96-
well plates (Becton Dickinson, San Jose, CA) with a density
of 50000 cells/well and cultured overnight. The culture medium
(DMEM containing 10% FBS) was then replaced with modified
Krebs solution (120 mM NaCl, 5 mM KCl, 0.62 mM MgSO4,
1.8 mM CaCl2, 10 mM HEPES, 6 mM glucose, pH 7.4), and
the cells were serum starved for 6-8 h (12 h for PC-3 cells).
Cells were loaded with Fura-2 AM for 35 min in modified
Krebs medium containing 2% pluronic acid. The dye loading
medium was removed and replaced with 100 µL of modified
Krebs medium/well prior to analysis in a Flex Station II

Lysophosphatidic Acid Receptor Ligands
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4929
instrument. Changes in intracellular Ca2+ concentration were
monitored by measuring the ratio of emitted light intensity
at 520 nm elicited by excitation at 340 or 380 nm wavelength
lights, respectively. Each well was monitored for 80-120 s. A
50 µL amount of the test compound (3× stock solution in
modified Krebs) was added automatically after 15 s of starting
the measurement. Time courses were recorded using the
SoftMax Pro software (Molecular Devices, Sunnyvale, CA).
Ca2+ transients were quantified automatically by calculating
the difference between maximum and baseline ratio values for
each well.
To evaluate the whether the FAP antagonists are reversible,
we applied assay conditions described in our previous publica-
tions.47 Briefly, RH7777 cells stably transfected with LPA
GPCR were plated at a density of 4 × 105 on poly-L-lysine-
coated glass coverslips in growth medium (DMEM containing
10% FBS) overnight at 37 °C. The normal growth medium was
replaced with serum-free DMEM for 6 h before Ca2+ measure-
ments. Cells were loaded with Fura-2 AM in serum-free
DMEM for 35 min. The coverslips were gently washed twice
with Krebs solution in order to remove excess Fura 2-AM. Cells
were exposed to LPA, washed for 500 s, and exposed to FAP
antagonists at a concentration of 30 µM together with LPA.
This concentration completely inhibited LPA-induced Ca-
transients. The cells were washed for an additional ∼500 s
and reexposed to the original concentration of LPA.
and CPF4 was added at a concentration of 2 µM. CPF4
fluorescence was monitored (at 37 °C) in a BMG Fluorstar 96-
well plate reader (excitation at 355 nm; emission at 460 and
520 nm).
Acknowledgment. This work was supported by
research grants CA92160 and HL 61469 from the
National Institutes of Health, the Deutsche Forschungs-
gemeinschaft (Si 274/9), and the Dutch Cancer Society.
We thank Dr. Gordon B. Mills (M. D. Anderson Cancer
Center, Houston, TX) for providing compound 3. Com-
pound 5 was a kind gift from Dr. Kevin R. Lynch
(University of Virginia, Charlottesville, VA). Compound
6 was provided from the Kirin Brewery Co., Pharma-
ceutical Research Division, Tokyo. We also thank Dr.
Yehia Daaka (Duke University, Durham, NC) for gener-
ously providing PC-3 cells. Finally, we thank Drs.
Daniel L. Baker and Abby L. Parrill for their valuable
comments and critical reading of the manuscript.
Supporting Information Available: Table S1, listing the
elemental analyses of the final compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
PPARγ Activation Assay. Determination of PPARγ acti-
vation in CV1 cells transfected with an acyl-coenzyme A
oxidase-luciferase (PPRE-Acox-Rluc) reporter gene construct
was run as previously reported.37 Briefly, CV-1 cells were
plated in 96-well plates (5 × 103 cells per well) in Dulbecco’s
modified Eagle’s medium supplemented with 10% fetal bovine
serum. The next day, the cells were transiently transfected
with 125 ng of pGL3-PPRE-Acox-Rluc, 62.5 ng of pcDNAI-
PPARγ, and 12.5 ng of pSV-â-galactosidase (Promega, Madi-
son, WI) using LipofectAMINE 2000 (Invitrogen). Twenty-four
hours after transfection, cells were treated with 1% FBS
supplemented OptiMEMI (Invitrogen) containing DMSO or 10
µM test compound dissolved in DMSO for 20 h. Luciferase and
â-galactosidase activities were measured with the Steady-Glo
Luciferase Assay System (Promega) and the Galacto-Light
Plus System (Applied Biosystems, Foster City, CA), respec-
tively. Samples were run in quadruplicate and the mean (
standard errors were calculated. Data are representative of
at least two independent transfections. Student’s t-test was
used for null hypothesis testing, and P < 0.05 was considered
significant (in the figures P < 0.05 is denoted by * and P <
0.01 is **).
References
(1) Tigyi, G.; Parrill, A. L. Molecular mechanisms of lysophospha-
tidic acid action. Prog. Lipid Res. 2003, 42, 498-526.
(2) Mills, G. B.; Moolenaar, W. H. The emerging role of lysophos-
phatidic acid in cancer. Nat. Rev. Cancer 2003, 3, 582-591.
(3) Hecht, J. H.; Weiner, J. A.; Post, S. R.; Chun, J. Ventricular zone
gene-1 (vzg-1) encodes a lysophosphatidic acid receptor expressed
in neurogenic regions of the developing cerebral cortex. J. Cell.
Biol. 1996, 135, 1071-1083.
(4) An, S.; Bleu, T.; Zheng, Y.; Goetzl, E. J. Recombinant human G
protein-coupled lysophosphatidic acid receptors mediate intra-
cellular calcium mobilization. Mol. Pharmacol. 1998, 54, 881-
888.
(5) An, S.; Bleu, T.; Hallmark, O. G.; Goetzl, E. J. Characterization
of a novel subtype of human G protein-coupled receptor for
lysophosphatidic acid. J. Biol. Chem. 1998, 273, 7906-7910.
(6) Bandoh, K.; Aoki, J.; Hosono, H.; Kobayashi, S.; Kobayashi, T.;
Murakami-Murofushi, K.; Tsujimoto, M.; Arai, H.; Inoue, K.
Molecular cloning and characterization of a novel human G-
protein-coupled receptor, EDG7, for lysophosphatidic acid. J.
Biol. Chem. 1999, 274, 27776-27785.
(7) Moolenaar, W. H. Bioactive lysophospholipids and their G
protein-coupled receptors. Exp. Cell Res. 1999, 253, 230-238.
(8) Contos, J. J.; Ishii, I.; Chun, J. Lysophosphatidic acid receptors.
Mol. Pharmacol. 2000, 58, 1188-1196.
(9) Im, D. S.; Heise, C. E.; Harding, M. A.; George, S. R.; O’Dowd,
B. F.; Theodorescu, D.; Lynch, K. R. Molecular cloning and
characterization of a lysophosphatidic acid receptor, Edg-7,
expressed in prostate. Mol. Pharmacol. 2000, 57, 753-759.
(10) Noguchi, K.; Ishii, S.; Shimizu, T. Identification of p2y9/GPR23
as a novel G protein-coupled receptor for lysophosphatidic acid,
structurally distant from the Edg family. J. Biol. Chem. 2003,
278, 25600-25606.
(11) McIntyre, T. M.; Pontsler, A. V.; Silva, A. R.; St Hilaire, A.; Xu,
Y.; Hinshaw, J. C.; Zimmerman, G. A.; Hama, K.; Aoki, J.; Arai,
H.; Prestwich, G. D. Identification of an intracellular receptor
for lysophosphatidic acid (LPA): LPA is a transcellular PPAR-
gamma agonist. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 131-
136.
(12) Umezu-Goto, M.; Kishi, Y.; Taira, A.; Hama, K.; Dohmae, N.;
Takio, K.; Yamori, T.; Mills, G. B.; Inoue, K.; Aoki, J.; Arai, H.
Autotaxin has lysophospholipase D activity leading to tumor cell
growth and motility by lysophosphatidic acid production. J. Cell
Biol. 2002, 158, 227-233.
(13) Tokumura, A.; Majima, E.; Kariya, Y.; Tominaga, K.; Kogure,
K.; Yasuda, K.; Fukuzawa, K. Identification of human plasma
lysophospholipase D, a lysophosphatidic acid-producing enzyme,
as autotaxin, a multifunctional phosphodiesterase. J. Biol.
Chem. 2002, 277, 39436-39442.
(14) Hines, O. J.; Ryder, N.; Chu, J.; McFadden, D. Lysophosphatidic
acid stimulates intestinal restitution via cytoskeletal activation
and remodeling. J. Surg. Res. 2000, 92, 23-28.
Isolation of Platelets and Measurement of Platelet
Shape Change. For measurement of shape change, human
platelets were treated with acetylsalicylic acid and isolated
in the presence of apyrase as described previously.48 Platelets
were resuspended at a concentration of 4 × 105/µL in buffer
(20 mM HEPES, 138 mM NaCl, 2.9 mM KCl, 1 mM MgCl2,
0.36 mM NaH2PO4, 5 mN glucose; 0.6 U ADPase/ml apyrase;
pH 7.4). The test compounds were dissolved in methanol
(generally at 2 mM concentration) and stored at -80 °C. Before
the experiment, methanol was evaporated under a stream of
nitrogen, and the lipids were resuspended by sonication in
buffer (20 mM HEPES, 138 mM NaCl, BSA 0.25 mM) to obtain
a 1 mM solution (with a ratio of lipid/albumin of 4:1).
Suspensions of washed platelets were transferred into aggre-
gometer cuvettes, incubated at 37 °C with stirring (1100 rpm),
and exposed to various concentrations of the compounds or
vehicle control. After a fixed incubation time, LPA 16:0 was
added. Shape change was measured by the decrease in light
transmission of the stirred platelet suspension in a LABOR
aggregometer (Fresenius, Bad Homburg, Germany).
Autotaxin Assay. Autotaxin phosphodiesterase activity
was determined using the fluorescence-resonance energy-based
sensor CPF449 as a substrate, as described by van Meeteren
et al.38 Briefly, recombinant ATX in Tris-buffered saline (140
mM NaCl, 5 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 50 mM Tris;
pH 8.0) was incubated with or without the test compounds,
(15) Sturm, A.; Sudermann, T.; Schulte, K. M.; Goebell, H.; Dignass,
A. U. Modulation of intestinal epithelial wound healing in vitro
and in vivo by lysophosphatidic acid. Gastroenterology 1999, 117,
368-377.

4930 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Durgam et al.
(16) Deng, W.; Balazs, L.; Wang, D. A.; Van Middlesworth, L.; Tigyi,
G.; Johnson, L. R. Lysophosphatidic acid protects and rescues
intestinal epithelial cells from radiation- and chemotherapy-
induced apoptosis. Gastroenterology 2002, 123, 206-216.
(17) Rother, E.; Brandl, R.; Baker, D. L.; Goyal, P.; Gebhard, H.;
Tigyi, G.; Siess, W. Subtype-selective antagonists of lysophos-
phatidic Acid receptors inhibit platelet activation triggered by
the lipid core of atherosclerotic plaques. Circulation 2003, 108,
741-747.
(33) Virag, T.; Elrod, D. B.; Liliom, K.; Sardar, V. M.; Parrill, A. L.;
Yokoyama, K.; Durgam, G.; Deng, W.; Miller, D. D.; Tigyi, G.
Fatty alcohol phosphates are subtype-selective agonists and
antagonists of lysophosphatidic acid receptors. Mol. Pharmacol.
2003, 63, 1032-1042.
(34) Lindberg, J.; Ekeroth, J.; Konradsson, P. Efficient synthesis of
phospholipids from glycidyl phosphates. J. Org. Chem. 2002, 67,
194-199.
(35) Haines, A. H.; Massy, D. J. R. Synthesis of the dipotassium salts
of methyl R-D-mannopyranoside 6-phosphorothioate and D-
mannose 6-phosphorothioate. Synthesis 1996, 12, 1422-1424.
(36) Halazy, A.; Ehrhard, A.; Danzin, C. 9-(Difluorophosphonoalkyl)-
guanines as a new class of multisubstrate analogue inhibitors
of purine nulceoside phosphorylase. J. Am. Chem. Soc. 1991,
113, 315-317.
(37) Zhang, C.; Baker, D. L.; Yasuda, S.; Makarova, N.; Balazs, L.;
Johnson, L. R.; Marathe, G. K.; McIntyre, T. M.; Xu, Y.;
Prestwich, G. D.; Byun, H. S.; Bittman, R.; Tigyi, G. Lysophos-
phatidic acid induces neointimaformation through PPARγ ac-
tivation. J. Exp. Med. 2004, 199, 763-774.
(18) Spector, A. A. Plaque rupture, lysophosphatidic acid, and
thrombosis. Circulation 2003, 108, 641-643.
(19) Xu, Y.; Gaudette, D. C.; Boynton, J. D.; Frankel, A.; Fang, X. J.;
Sharma, A.; Hurteau, J.; Casey, G.; Goodbody, A.; Mellors, A.
Characterization of an ovarian cancer activating factor in ascites
from ovarian cancer patients. Clin. Cancer Res. 1995, 1, 1223-
1232.
(20) Hayashi, K.; Takahashi, M.; Nishida, W.; Yoshida, K.; Ohkawa,
Y.; Kitabatake, A.; Aoki, J.; Arai, H.; Sobue, K. Phenotypic
modulation of vascular smooth muscle cells induced by unsatur-
ated lysophosphatidic acids. Circ. Res. 2001, 89, 251-258.
(21) Okusa, M. D.; Ye, H.; Huang, L.; Sigismund, L.; Macdonald, T.;
Lynch, K. R. Selective blockade of lysophosphatidic acid LPA3
receptors reduces murine renal ischemia-reperfusion injury. Am.
J. Physiol. Renal Physiol. 2003, 285, F565-574.
(22) Graler, M. H.; Goetzl, E. J. Lysophospholipids and their G
protein-coupled receptors in inflammation and immunity. Bio-
chim. Biophys. Acta 2002, 1582, 168-174.
(23) Lynch, K. R.; Hopper, D. W.; Carlisle, S. J.; Catalano, J. G.;
Zhang, M.; MacDonald, T. L. Structure/activity relationships in
lysophosphatidic acid: the 2-hydroxyl moiety. Mol. Pharmacol.
1997, 52, 75-81.
(24) Hooks, S. B.; Ragan, S. P.; Hopper, D. W.; Honemann, C. W.;
Durieux, M. E.; Macdonald, T. L.; Lynch, K. R. Characterization
of a receptor subtype-selective lysophosphatidic acid mimetic.
Mol. Pharmacol. 1998, 53, 188-194.
(25) Sugiura, T.; Tokumura, A.; Gregory, L.; Nouchi, T.; Weintraub,
S. T.; Hanahan, D. J. Biochemical characterization of the
interaction of lipid phosphoric acids with human platelets:
comparison with platelet activating factor. Arch. Biochem.
Biophys. 1994, 311, 358-368.
(26) Liliom, K.; Bittman, R.; Swords, B.; Tigyi, G. N-palmitoyl-serine
and N-palmitoyl-tyrosine phosphoric acids are selective competi-
tive antagonists of the lysophosphatidic acid receptors. Mol.
Pharmacol. 1996, 50, 616-623.
(27) Fischer, D. J.; Nusser, N.; Virag, T.; Yokoyama, K.; Wang, D.;
Baker, D. L.; Bautista, D.; Parrill, A. L.; Tigyi, G. Short-chain
phosphatidates are subtype-selective antagonists of lysophos-
phatidic acid receptors. Mol. Pharmacol. 2001, 60, 776-784.
(28) Hasegawa, Y.; Erickson, J. R.; Goddard, G. J.; Yu, S.; Liu, S.;
Cheng, K. W.; Eder, A.; Bandoh, K.; Aoki, J.; Jarosz, R.; Schrier,
A. D.; Lynch, K. R.; Mills, G. B.; Fang, X. Identification of a
phosphothionate analogue of lysophosphatidic acid (LPA) as a
selective agonist of the LPA3 receptor. J. Biol. Chem. 2003, 278,
11962-11969.
(29) Qian, L.; Xu, Y.; Hasegawa, Y.; Aoki, J.; Mills, G. B.; Prestwich,
G. D. Enantioselective responses to a phosphorothioate analogue
of lysophosphatidic acid with LPA3 receptor-selective agonist
activity. J. Med. Chem. 2003, 46, 5575-5578.
(30) Xu, Y.; Qian, L.; Prestwich, G. D. Synthesis of monofluorinated
analogues of lysophosphatidic acid. J. Org. Chem. 2003, 68,
5320-5330.
(38) van Meeteren, L. A.; Ruurs, P.; Christodoulou, E.; Goding, J.
W.; Takakusa, H.; Kikuchi, K.; Perrakis, A.; Nagano, T.;
Moolenaar, W. H. Inhibition of autotaxin by lysophosphatidic
acid and sphingosine-1-phosphate. J. Biol. Chem. 2005, 280,
21155-21161.
(39) Wang, D. A.; Lorincz, Z.; Bautista, D. L.; Liliom, K.; Tigyi, G.;
Parrill, A. L. A single amino acid determines lysophospholipid
specificity of the S1P1 (EDG1) and LPA1 (EDG2) phospholipid
growth factor receptors. J. Biol. Chem. 2001, 276, 49213-49220.
(40) Sardar, V. M.; Bautista, D. L.; Fischer, D. J.; Yokoyama, K.;
Nusser, N.; Virag, T.; Wang, D. A.; Baker, D. L.; Tigyi, G.; Parrill,
A. L. Molecular basis for lysophosphatidic acid receptor antago-
nist selectivity. Biochim. Biophys. Acta 2002, 1582, 309-317.
(41) Bandoh, K.; Aoki, J.; Taira, A.; Tsujimoto, M.; Arai, H.; Inoue,
K. Lysophosphatidic acid (LPA) receptors of the EDG family are
differentially activated by LPA species. Structure-activity
relationship of cloned LPA receptors. FEBS Lett. 2000, 478,
159-165.
(42) Hooks, S. B.; Santos, W. L.; Im, D. S.; Heise, C. E.; Macdonald,
T. L.; Lynch, K. R. Lysophosphatidic acid-induced mitogenesis
is regulated by lipid phosphate phosphatases and is Edg-receptor
independent. J. Biol. Chem. 2001, 276, 4611-4621.
(43) Xu, Y.; Prestwich, G. D. Synthesis of chiral (alpha,alpha-
difluoroalkyl)phosphonate analogues of (lyso)phosphatidic acid
via hydrolytic kinetic resolution. Org. Lett. 2002, 4, 4021-4024.
(44) Kue, P. F.; Taub, J. S.; Harrington, L. B.; Polakiewicz, R. D.;
Ullrich, A.; Daaka, Y. Lysophosphatidic acid-regulated mitogenic
ERK signaling in androgen-insensitive prostate cancer PC-3
cells. Int. J. Cancer. 2002, 102, 572-579.
(45) Daaka, Y. Mitogenic action of LPA in prostate. Biochim. Biophys.
Acta 2002, 1582, 265-269.
(46) Siess, W.; Zangl, K. J.; Essler, M.; Bauer, M.; Brandl, R.;
Corrinth, C.; Bittman, R.; Tigyi, G.; Aepfelbacher, M. Lysophos-
phatidic acid mediates the rapid activation of platelets and
endothelial cells by mildly oxidized low-density lipoprotein and
accumulates in human atherosclerotic lesions. Proc. Natl. Acad.
Sci. U.S.A. 1999, 96, 6931-6936.
(47) Tigyi, G.; Liliom, K.; Fischer, D. J.; Guo, Z. Phospholipid growth
factors: Identification and mechanism of action. Lipid Second
Messengers; CRC Press: Washington, DC, 1999; pp 51-81.
(48) Negrescu, E. V.; de Quintana, K. L.; Siess, W. Platelet shape
change induced by thrombin receptor activation. Rapid stimula-
tion of tyrosine phosphorylation of novel protein substrates
through an integrin- and Ca(2+)-independent mechanism. J.
Biol. Chem. 1995, 270, 1057-1061.
(31) Heise, C. E.; Santos, W. L.; Schreihofer, A. M.; Heasley, B. H.;
Mukhin, Y. V.; Macdonald, T. L.; Lynch, K. R. Activity of
2-substituted lysophosphatidic acid (LPA) analogs at LPA recep-
tors: discovery of a LPA1/LPA3 receptor antagonist. Mol.
Pharmacol. 2001, 60, 1173-1180.
(32) Ohta, H.; Sato, K.; Murata, N.; Damirin, A.; Malchinkhuu, E.;
Kon, J.; Kimura, T.; Tobo, M.; Yamazaki, Y.; Watanabe, T.; Yagi,
M.; Sato, M.; Suzuki, R.; Murooka, H.; Sakai, T.; Nishitoba, T.;
Im, D. S.; Nochi, H.; Tamoto, K.; Tomura, H.; Okajima, F.
Ki16425, a subtype-selective antagonist for EDG-family lyso-
phosphatidic acid receptors. Mol. Pharmacol. 2003, 64, 994-
1005.
(49) Takakusa, H.; Kikuchi, K.; Urano, Y.; Sakamoto, S.; Yamaguchi,
K.; Nagano, T. Design and synthesis of an enzyme-cleavable
sensor molecule for phosphodiesterase activity based on fluo-
rescence resonance energy transfer. J. Am. Chem. Soc. 2002,
124, 1653-1657.
JM049609R