Synthesis and biological evaluation of phosphonic and thiophosphoric acid derivatives of lysophosphatidic acid

Article abstract of DOI:10.1016/j.bmcl.2004.04.061

Using an N-oleoyl ethanolamide scaffold, a series of phosphate polar head group analogues of LPA comprised of various α-substituted phosphonates and thiophosphates was prepared. In a broken cell GTP[γ³⁵S] binding assay, agonist activity was evaluated at the three LPA receptors of the endothelial differentiation gene (Edg) family. This study has resulted in the discovery of a nonhydrolyzable LPA₁-selective agonist (11). Additionally, thiophosphate 19 bears an isosteric phosphate mimetic that confers agonism at the LPA₁ receptor but not LPA₂.

Full text of DOI:10.1016/j.bmcl.2004.04.061

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3473–3476
Synthesis and biological evaluation of phosphonic and
thiophosphoric acid derivatives of lysophosphatidic acid
b
a
Webster L. Santos,c, Brian H. Heasley,a, , Renata Jarosz, Karen M. Carter,
*
Kevin R. Lynchb and Timothy L. Macdonalda
aDepartment of Chemistry, University of Virginia, PO Box 400319, McCormick Road, Charlottesville, VA 22904, USA
bDepartment of Pharmacology, University of Virginia, PO Box 400319, McCormick Road, Charlottesville, VA 22904, USA
cDepartment of Chemistry and Chemical Biology, 12 Oxford Street, Cambridge, MA 02138, USA
Received 27 March 2004; revised 14 April 2004; accepted 19 April 2004
Abstract—Using an N-oleoyl ethanolamide scaffold, a series of phosphate polar head group analogues of LPA comprised of various
a-substituted phosphonates and thiophosphates was prepared. In a broken cell GTP[c35S] binding assay, agonist activity was
evaluated at the three LPA receptors of the endothelial differentiation gene (Edg) family. This study has resulted in the discovery of a
nonhydrolyzable LPA1-selective agonist (11). Additionally, thiophosphate 19 bears an isosteric phosphate mimetic that confers
agonism at the LPA1 receptor but not LPA2.
Ó 2004 Elsevier Ltd. All rights reserved.
X
P
O
Lysophosphatidic acid (LPA, 1- or 2-O-acyl-sn-glycero-
3-phosphate) is an intercellular signaling phospholipid
that induces an array of receptor-mediated biological
effects.1 However, lipid phosphate phosphatase (LPP)-
catalyzed hydrolysis of the phosphomonoester polar
head group of LPA results in functional inactivation at
LPA receptors.2 In recent years, progress has been made
toward the development of phosphatase-resistant phar-
macologic agents that modulate LPA signaling. The
metabolically stabilized LPA analogue, 1-oleoyl-2S-O-
methyl-glycerophosphothioate (2S-OMPT) is a potent
and selective agonist for the LPA3 receptor.3 Further
development of nonhydrolyzable molecular entities such
as this may provide the necessary tools to define the
receptor specific pathophysiological responses to LPA
stimulation in intact animals (Fig. 1).
HO
HO
O
O
R1 R2
1-oleoyl-LPA: R1 = OH; R2 = H; X = O
(2S)-OMPT: R1 = H; R2 = OMe; X = S
O
H
N
HO
HO
P
O
O
1 (NAEPA)
Figure 1. Structures of LPA receptor agonists.
spatial distribution of functionality as a phosphate but it
also allows similar conformations to be accessed.
However, the strong electronegative property of the
oxygen clearly affects the electronic attributes of a
phosphate group. As a result, the second phosphate pKa
value is 1-log order more acidic than the corresponding
a-methylene phosphonate (ꢀ6.4 compared to ꢀ7.6).4
This discrepancy may strongly affect receptor–ligand
binding interactions wherein charge density on the polar
head group is favorable. Electronegative substituents on
the a-carbon will inductively influence charge density
thereby increasing acidity. Indeed, the introduction of
a,a-difluoro decreases the second phosphonate pKa
by as much as 2-log orders.4 Alternatively, the substi-
tution of sulfur for one of the nonbridging oxygen
atoms of a phosphate yields the metabolically stable5
Replacement of the bridging oxygen of the phosphate
group with carbon renders phosphonates hydrolytically
stable. Not only does a phosphonate possess the same
Keywords: Lysophosphatidic acid; Phosphonic acid; Thiophosphoric
acid; Phosphatase.
* Corresponding author. Tel.: +1-434-9240595; fax: +1-434-9822302;
e-mail: bhh4x@vigninia.edu
These authors contributed equally to the completion of this
manuscript.
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.04.061

3474
W. L. Santos et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3473–3476
O
P
phosphothioate, which exhibits different steric and
geometric properties as well as charge distribution in the
respective dianions.6 Thiol phosphoric acids, in which
the bridging oxygen of the phosphate is replaced by
sulfur, have been explored to a lesser extent as biologi-
cally active thio-analogues. This isostere is similarly
charged and has the same hydrogen-bonding capabilities
as the phosphate ester, except for the bridging atom.
H
N
O
O
P
a, b
17:1
OR
OR
c
HP OEt
OEt
H2N
OEt
OEt
O
2
3
4 (R = Et)
5 (R = H)
d
Scheme 1. Synthesis of phosphonate 5. Reagents and conditions: (a)
Na, EtOH, acrylonitrile, 70%; (b) NaBH4, CoCl2, ꢁ30 to 25 °C, 53%;
(c) oleoyl chloride, DIEA, 80%; (d) (i) TMSBr, CH3CN, 4 h; (ii) 95%
MeOH, 1 h, 95%.
To explore the potential of phosphonates and thio-
phosphates as potential nonhydrolyzable analogues of
LPA, we now report the synthesis and biological eval-
uation of a series of phosphate-mimetic derivatives of
the known LPA receptor agonist N-acyl ethanolamide
phosphoric acid (NAEPA, 1).7
O
c
a, b
H2N
OH
17:1
N
H
O
6
7
10 (R = tBu)
11 (R = H)
O
O
Syntheses of all compounds listed in Table 1 except 208
are described in Schemes 1–3. As outlined in Scheme 1,
phosphite anion condensed with acrylonitrile in Micheal
fashion to provide the resulting nitrile in good yield.
Cobalt(II) chloride-catalyzed sodium borohydride
reduction afforded the amine 3, which underwent rou-
tine acylation followed by transesterification of the
phosphonodiester using bromotrimethylsilane (TMSBr)
and subsequent desilylation with aqueous methanol to
give a-methylene phosphonate 5.
OR
OR
d
17:1
17:1
17:1
N
H
P
e
O
OH
O
O
OH
OtBu
OH
OH
d
f
17:1
N
H
P
OtBu
N
H
P
9
O
O
8
13 (R = tBu)
14 (R = H)
O
F
OR
OR
d
N
H
P
O
Scheme 2. Synthesis of a-substituted phosphonates 9, 11, 14. Reagents
and conditions: (a) oleoyl chloride, DIEA, 81%; (b) PCC, 47%; (c) di-
tert-butyl phosphite, NaH, 51%; (d) TFA–CH2Cl2 (1:1), 100%; (e)
PCC, 28%; (f) DAST, 0 °C, 15%.
As outlined in Scheme 2, condensation of propanol-
amine 6 with oleoyl chloride followed by pyridinium
chlorochromate (PCC) oxidation generated the alde-
hyde 7. Phosphite addition to 7 afforded the racemic-
hydroxy phophonodiester 8, which is converted to the
phosphonic acid by treatment with trifluoroacetic acid
(TFA). Additionally, 8 is a common intermediate used
to generate the a-keto and a-fluoro derivatives, 11 and
14, respectively. PCC oxidation of 8 followed by TFA
deprotection afforded a-keto phosphonate 11. For the
latter, reaction of 8 with (diethylamino)sulfur trifluoride
(DAST) at ice-bath temperature provided the racemic
a-fluorophosphonodiester in modest yield. Subsequent
TFA deprotection gave 14.
condensed selectively with oleoyl chloride to generate
15. Phosphitylation of the alcohol 15 followed by in situ
sulfurization with elemental sulfur provided the phos-
phothioate triester 16 in good yield. Notably, TFA
deprotection in this case required the addition of at least
20 equiv of the scavenger reagent triethylsilane so as to
prevent intra- or intermolecular migration of the inter-
mediate tert-butyl cation onto sulfur. An analogous
alkyl migration has been noted elsewhere.9 Removal of
volatiles under reduced pressure afforded the pure
phosphothioate 17. The thiol phosphoric triester 18 was
prepared by an Arbuzov-type redox reaction of tri-
methyl phosphite with the thiol 15 in the presence of
As outlined in Scheme 3, two thio-analogues were also
prepared. Either ethanolamine or 2-aminoethane thiol
Table 1. Agonist activities of LPA analogues at individual LPA receptors using GTP[c35S] binding assay
O
X
Y
N
H
Compound
X
Y
LPA1
LPA2
LPA3
EC50 (nM)
Emax
EC50 (nM)
Emax
EC50 (nM)
Emax
LPA
1
NA
–O–
NA
17
197
100
100
50
6.8
30
100
110
NA
90
263
>5000
>5000
>5000
>5000
>5000
>5000
>5000
>5000
100
NA
NA
NA
NA
NA
NA
NA
NA
–PO3H2
–PO3H2
–PO3H2
–PO3H2
–PO3H2
–(PO3H2)2
–P(S)O2H2
–PO3H2
5
–CH2–
–CH(OH)–
–C(O)–
–CH(F)–
–C(OH)–
–O–
1950
1225
221
>5000
2066
1089
9
80
80
11
14
20
17
19
90
>5000
>5000
40
NA
NA
80
1960
100
NA
46
>5000
108
>5000
–S–
318
50
NA
Emax ¼ maximal efficacy of drug/maximal efficacy of 1-oleoyl-LPA; expressed as the percentage. NA, not applicable.

W. L. Santos et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3473–3476
3475
O
O
O
OR
OR
OR
OR
a
b
S
O
XH
17:1
N
H
P
17:1
N
H
P
S
17:1
N
H
X = O
X = S
O
15
16 (R = tBu)
18 (R = Et)
19 (R = H)
c
d
17 (R = H)
Scheme 3. Synthesis of thiophosphoric acids 17, 19. Reagents and conditions: (a) (i) di-tert-butyl diisopropyl phosphoramidite, tetrazole; (ii) sulfur,
reflux, 86%; (b) trimethyl phospite, TeCl4; 2,6-lutidine, 67%; (c) TFA–CH2Cl2 (1:2), triethylsilane (20 equiv), 95%; (d) (i) TMSBr, CH2Cl2, 4 h;
(ii) MeOH, 1 h, 100%.
TeCl4 and 2,6-lutidine according to the method of
Watanbe et al.10 TMSBr deprotection afforded the thiol
phosphoric acid 19.
LPA1
1250
1000
750
LPA 18:1
(1)
All compounds were characterized by 1H and 13C NMR
and mass spectroscopy.11 A membrane-based GTP[c35S]
binding assay was adapted to assess in vitro activity at
the LPA1, LPA2, and LPA3 receptors.12 It has been
demonstrated that NAEPA (1) is an excellent LPA mi-
metic, indistinguishable from LPA in aggregating hu-
man platelets and mobilizing calcium in mammalian
cells.13;14 However, this activity is primarily elicited by
activation of the LPA1 and LPA2 receptors, but not
LPA3, where potency is markedly diminished.7 We
conserved the ethanolamide scaffold of 1 and replaced
the phosphate moiety with a series of phosphate isos-
teres. The resulting GTP[c35S] binding data for all
compounds are presented in Table 1.
(17)
(19)
(11)
500
250
0
-10
-9
-8
-7
-6
-5
-4
Log[compd]
Compared to NAEPA (1), a-methylene phosphonate 5
is less active by 1-log order at the LPA1 receptor and
inactive at LPA2 and LPA3. To modulate the second
pKa value and thus more effectively mimic the elec-
tronics of the phosphate, electron-withdrawing hetero-
atom substituents were introduced on the a-carbon of
phosphono-NAEPA derivatives 9, 11, 14, and 20. The
phosphonate rank order potency at the LPA1 receptor is
a-keto > a-hydroxy > a-methylene > a-fluoro. Given the
improved agonism of compounds 9 and 11, electron
density and pKa of the polar head group does seem to
impact activity at LPA receptors. However, the relative
inactivity of a-fluoro phosphonate 14 is not consistent
with this hypothesis; this moiety has been reported to
most closely approximate the electronics, acidity, and
conformation of phosphate,4 but results in a dramatic
loss of LPA receptor activation. This unexpected result
remains unresolved and will require additional studies.
Notably, 14 is fully efficacious at the LPA2 receptor
albeit at a relatively high concentration. a-Hydroxy
bisphosphonate 20 is inactive at all three receptors,
perhaps due to steric and/or electrostatic factors. Thus,
a-keto phosphonate 11 displays potency and efficacy at
the LPA1 receptor comparable to that of lead
compound NAEPA. However, 11 is greater than 1-log
order less potent than NAEPA at LPA2 and therefore
represents a nonhydrolyzable LPA1-selective agonist.
Figure 2. Agonist-induced GTP[c35S] binding11 at the LPA1 receptor.
Points are in triplicate and are representative of at least two experi-
ments.
to LPA or NAEPA. Hence 17 is a phosphatase-resis-
tant5 dual LPA agonist, displaying EC50s at LPA1, and
LPA2 of 40 and 108 nM, respectively. Additionally, thiol
phosphoric acid 19, unlike dual-agonist NAEPA, is an
LPA1-selective agonist of relatively good potency and
efficacy. This selectivity is presumably due to a lack of
compatibility between this phosphate isostere, perhaps
due to sterics or geometry, and the LPA2 receptor
binding pocket. At least one group has reported on the
chemical properties and enzymatic transformations of
the phosphorylated thiol (Fig. 2).15
In summary, a series of polar head group analogues of
LPA have been prepared and evaluated at three indi-
vidual LPA receptors. This study has resulted in the
discovery of a nonhydrolyzable LPA1-selective agonist
bearing an a-keto phosphonate head group as the
phosphate mimic. The resistance of this new pharma-
cologic agent to phosphatase-catalyzed degradation
renders 11 a useful tool to elucidate the functions of the
LPA1 receptor in vivo models. Additionally, thiol
phosphoric acid 19 is a relatively potent and efficacious
agonist at LPA1 whose head group is not tolerated at the
LPA2 receptor. Finally, a phosphothioate analogue (17)
of NAEPA is highly potent at the LPA1 and LPA2
receptors. The lack of activity at LPA3 of the head
group isosteres presented here is not surprising given the
Two thiophosphate analogues were analyzed for agonist
activity at LPA receptors. Phosphothioate 17 displayed
agonist potency comparable to that of lead compound
NAEPA. Maximal efficacy of this derivative, however, is
diminished by twofold at the LPA2 receptor, compared

3476
W. L. Santos et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3473–3476
formed which was isolated by filtration and washed with
excess methanol. The white solid 20 was judged as pure by
1H and 13C NMR and mass spectroscopy analysis.
inactivity of lead compound NAEPA itself at this
receptor. Studies to determine binding affinity at addi-
tional LPA receptors are ongoing and will be published
in due course.
H
N
H2O3P
HO
17:1
20
H2O3P
HO
H2O3P
NH2
H2O3P
O
a
Acknowledgements
+
b
H
O
H2O3P
O
H2O3P
Support by NIH (R01-GM52722).
N
17:1
17:1
O
8
Pamidronateâ was obtained as a generous gift from
Novartis Pharma AG, Basel, Switzerland.
Synthesis of α-OH bisphosphonate 20. (a) oleoyl chloride, 2M K2CO3 (aq);
(b) 2M LiOH (aq)/THF, 80% over two steps.
9. de Bont, D. B. A.; Moree, W. J.; van Boom, J. H.;
Liskamp, R. M. J. J. Org. Chem. 1993, 58, 1309–1317.
10. Watanbe, Y.; Inoue, S.; Yamamoto, T.; Ozaki, S.
Synthesis 1995, 1243–1244.
References and notes
1. Tigyi, G.; Parrill, A. L. Prog. Lipid Res. 2003, 42, 498–526.
2. Sciorra, V. A.; Morris, A. J. Biochim. Biophys. Acta 2002,
1582, 45–51.
3. Quian, L.; Xu, Y.; Hasegawa, Y.; Aoki, J.; Mills, G. B.;
Prestwich, G. D. J. Med. Chem. 2003, 46, 5575–5578.
4. Nieschalk, J.; Batsanov, A. S.; O’Hagan, D.; Howard, J.
A. K. Tetrahedron 1996, 52, 165–176.
5. 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. J. Biol. Chem. 2003, 278, 11962–11969, and
references cited therein.
11. Thiophosphoric acid 2-((Z)-octadec-9-enoylamino)-ethyl
ester (17): 1H NMR (300 MHz, CDCl3) d 0.87 (t, 3H,
J ¼ 6:7 Hz), 1.22–1.39 (m, 11H), 1.60–1.70 (m, 2H), 1.95–
2.02 (m, 4H), 2.33–2.46 (m, 2H), 3.54–3.64 (m, 2H), 4.08–
4.19 (m, 2H), 5.34 (t, 2H, J ¼ 6:0 Hz); 13C NMR
(300 MHz, CDCl3) 14.13, 22.67, 25.88, 27.24, 29.16,
29.32, 29.54, 29.78, 31.92, 32.66, 36.03, 40.41, 40.81,
50.80, 65.14, 129.70, 130.02, 177.34; MS (ESI) m=z 419.8
[MꢁHþ, 100%].
12. Im, D. S.; Heise, C. E.; Harding, M. A.; George, S. R.;
O’Dowd, B. F.; Theodorescu, D.; Lynch, K. R. Mol.
Pharm. 2000, 57, 753–759.
6. Swierczek, K.; Pandey, A. S.; Peters, J. W.; Hengge, A. C.
J. Med. Chem. 2003, 46, 3703–3708.
7. Lynch, K. R.; Macdonald, T. L. Biochim. Biophys. Acta
2002, 1582, 289–294.
13. Hooks, S. B.; Ragan, S. P.; Hopper, D. W.; Honemann, C.
W.; Durieux, M. E.; Macdonald, T. L.; Lynch, K. R. Mol.
Pharm. 1998, 53, 188–194.
8. In brief, Pamidronateâ (phosphoric acid (3-amino-1-
hydroxypropylidene)bisdisodium salt) was condensed with
oleoyl chloride in 2 M aqueous K2CO3. This afforded a
mixture of mono- and di-acylated bisphosphonates, which
were exposed to a solution of 1:1 2 M LiOH–THF. Upon
addition of methanol to the reaction mixture, a precipitate
14. Sugiura, T.; Tokumura, A.; Gregory, L.; Nouchi, T.;
Weintraub, S. T.; Hanahan, D. J. Arch. Biochem. Biophys.
1994, 311, 358–368.
15. Knight, W. B.; Sem, D. S.; Smith, K.; Miziorko, H. M.;
Rendina, A. R.; Cleland, W. W. Biochemistry 1991, 30,
4970–4977.