Initial structure-activity relationships of lysophosphatidic acid receptor antagonists: Discovery of a high-affinity LPA1/LPA3 receptor antagonist

Article abstract of DOI:10.1016/j.bmcl.2004.03.076

A recently reported dual LPA₁/LPA₃ receptor antagonist (VPC12249, 1) has been modified herein so as to optimize potency and selectivity at LPA receptors. Compounds containing variation in the acyl lipid chain and linker region have been synthesized and screened for activity at individual LPA receptors. LPA₁-selective (14b) and LPA ₃-selective (10g,m) compounds of modest potency have been discovered. Additionally, 2-pyridyl derivative 10t exhibits a K_i value of 18nM at the LPA₁ receptor and is significantly more potent than 1 at the LPA₃ receptor. This paper describes the synthetic methods, biological evaluation, and structure-activity relationships (SARs) of LPA receptor antagonists.

Full text of DOI:10.1016/j.bmcl.2004.03.076

Bioorganic & Medicinal Chemistry Letters 14 (2004) 2735–2740
Initial structure–activity relationships of lysophosphatidic acid
receptor antagonists: discovery of a high-affinity LPA1/LPA3
receptor antagonist
Brian H. Heasley,a,* Renata Jarosz,b Kevin R. Lynchb and Timothy L. Macdonalda
aDepartment of Chemistry, University of Virginia, PO Box 400319, McCormickRoad, Charlottesville, VA 22904, USA
bDepartment of Pharmacology, University of Virginia, PO Box 400319, McCormickRoad, Charlottesville, VA 22904, USA
Received 17 February 2004; revised 21 March 2004; accepted 25 March 2004
Abstract—A recently reported dual LPA1/LPA3 receptor antagonist (VPC12249, 1) has been modified herein so as to optimize
potency and selectivity at LPA receptors. Compounds containing variation in the acyl lipid chain and linker region have been
synthesized and screened for activity at individual LPA receptors. LPA1-selective (14b) and LPA3-selective (10g,m) compounds of
modest potency have been discovered. Additionally, 2-pyridyl derivative 10t exhibits a Ki value of 18 nM at the LPA1 receptor and is
significantly more potent than 1 at the LPA3 receptor. This paper describes the synthetic methods, biological evaluation, and
structure–activity relationships (SARs) of LPA receptor antagonists.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Lysophosphatidic acid (LPA, 1- or 2-O-acyl-sn-glycero-
3-phosphate) is an endogenous glycerophospholipid that
can be generated by many cell types to elicit varied bio-
logical responses including cellular proliferation, platelet
aggregation, smooth muscle contraction, and changes in
cell morphology and mitogenesis.1–4 G-protein coupled
receptors of the endothelial differentiation gene (Edg)
family transduce many of the biological effects of LPA.5
This family consists of eight members, which are further
divided into two subfamilies based upon their specificity
for LPA or a related lysophospholipid sphingosine-1-
phosphate (S1P). Three of the eight Edg receptors,
LPA1, LPA2, and LPA3 (formerly Edg2, Edg4, and
Edg7, respectively), are high affinity LPA receptors.
Recent evidence has emerged suggesting at least two
non-Edg receptors also bind LPA with high affinity.6;7
cancer. LPA, a potent fibroblast growth factor, is now
known to be an ‘ovarian cancer activating factor’ in
ascitic fluid from ovarian cancer patients.8 Elevated
levels of LPA are present both at early and late stages in
ovarian cancer and may play a role in tumor cell pro-
liferation and invasion. LPA has also been identified as
pathogenic in the development and progression of ath-
erosclerosis.9 Finally, it was demonstrated recently that
LPA is protective at low doses against renal ischemia-
reperfusion (IR) injury.10 Furthermore, a dual LPA1/
LPA3 antagonist (VPC12249, 1) produced a profound
decrease in IR injury, a protective effect that was re-
versed by a selective LPA3 agonist.
Despite potential applications of LPA receptor antago-
nists as therapeutic agents, the detailed physiological
implications of blockade of individual LPA receptors
are largely unknown because subtype-selective antago-
nists are unavailable currently. This paper describes the
synthetic methods, biological evaluation, and initial
structure–activity relationships (SARs) of LPA receptor
antagonists.
Lysophosphatidic acid (LPA) antagonists have thera-
peutic potential as inhibitors of inflammation, neo-
plasms, and ischemic reperfusion injury. LPA can be
detected in various biological fluids such as serum,
plasma, tears, malignant ascites, and saliva, and its
levels are elevated in various physiological and patho-
logical conditions such as pregnancy and ovarian
O
P
HO
HO
O
Ar
O
NHCOR
Keywords: Lysophosphatidic acid; Antagonist.
* Corresponding author. Tel.: +1-434-924-0595; fax: +1-434-982-2302;
e-mail: bhh4x@virginia.edu
1 : R = 17:1 (oleoyl); Ar = Ph (Ki137 nM LPA1, 428 nM LPA3)
Figure 1. Structure of dual LPA1/LPA3 antagonist, VPC12249 (1).
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.03.076

2736
B. H. Heasley et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2735–2740
Recently, a series of 2-substituted N-acyl ethanolamine
phosphoric acid (NAEPA) derivatives were synthesized
and evaluated at LPA receptors.11 From this series of
hydrogenolysis of 3c afforded the phenol, which was O-
alkylated to the 3-methoxybenzyl aryl ether using
Mitsunobu conditions.
NAEPA derivatives emerged
a dual LPA1/LPA3
antagonist containing a bulky benzyl-4-oxybenzyl sub-
stitution at the 2-position in the linker region
(VPC12249, 1, Fig. 1). To more thoroughly elucidate the
structural parameters that contribute to potency and
selectivity in LPA antagonism, a series of VPC12249 (1)
analogues were prepared and analyzed.
As outlined in Scheme 2, the amide 7 was obtained by
selective acylation of tyrosine methyl ester followed by
protection of the phenol as the silyl ether. Methyl ester
reduction and phosphorylation followed by desilylation
with tetrabutylammonium fluoride (TBAF) afforded the
phosphotriester 9. Compound 9 was a common inter-
mediate used to generate a series of ethers/esters by the
Mitsunobu reaction or standard PyBOPꢁ chemistry,
respectively. TFA deprotection afforded phosphates
10a–u. In the case of glucuronide 10u, catalytic hy-
drogenolysis using Pearlmann’s catalyst deprotected the
Syntheses of all the compounds listed in Tables 1–3 are
described in Schemes 1–3. As outlined in Scheme 1, N-
acyl tyrosine methyl esters 3a–c,e were prepared by
esterification and subsequent N-acylation of protected
tyrosine 2. Methyl ester reduction and phosphitylation
of the resulting alcohol followed by in situ oxidation and
subsequent trifluoroacetic acid (TFA) deprotection
afforded phosphates 4a–c,e. In the case of 4d, catalytic
tetra-O-benzyl-D-glucopyranose moiety prior to TFA
deprotection of the phosphate ester. Compound 10u is
therefore the only compound in this series, which con-
tains a saturated acyl chain.
Table 1. Optimization of the N-acyl moiety as compared with lead compound 1: biological evaluation of derivatives 4a–e
O
H
N
HO
HO
P
R
O
O
Ar
O
Compounds
R
Ar
LPA1
LPA3 IC50 (nM)
IC50 (nM)
Ki (nM)
4a
4b
4c
4d
111
4e
n-C7H15
n-C13H27
Phenyl
Phenyl
Phenyl
7970
9030
N/D
N/D
N/D
N/D
137
>10,000
>10,000
6840
n-C15H31
n-C15H31
Oleoyl, 17:1a
Linoleoyl, 17:2b
>10,000
>10,000
5210
3-OMe-phenyl
Phenyl
Phenyl
7220
6450
4250
N/D
>10,000
a 18 carbon chain; cis double bond located between C-9 and C-10 from the carbonyl.
b 18 carbon chain; two cis double bonds located between C-9/C-10 and C-12/C-13 from the carbonyl.
Table 2. Biological evaluation of ether/ester derivatives 10a–u
O
H
N
HO
HO
P
*
O
O
RO
Compounds
R
S/R
LPA1
LPA3 IC50 (nM)
IC50 (nM)
Ki (nM)
10a
10b
10c
10d
10e
‹ CH3
R
S
R
S
S
>10,000
>10,000
>10,000
>10,000
>10,000
N/D
N/D
N/D
N/D
N/D
>10,000
6930
6410
‹ CH2CH3
‹ CH2CH3
‹ CH2CH ¼ CH2
>10,000
>10,000
10f
S
>10,000
N/D
5920

B. H. Heasley et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2735–2740
2737
Table 2 (continued)
Compounds
R
S/R
LPA1
LPA3 IC50 (nM)
IC50 (nM)
Ki (nM)
10g
10h
R
R
>10,000
N/D
351
>10,000
N/D
3280
10i
10j
R
S
>10,000
N/D
358
4030
642
OMe
OMe
5660
10k
10l
R
S
1070
N/D
34
35.6
OMe
O
N/D
N/D
427
OMe
OMe
OMe
O
10m
R
>10,000
N/D
OMe
OMe
10n
10o
R
S
2930
134
2310
O
>10,000
N/D
>10,000
n-C5H11
O
10p
10q
S
S
>10,000
>10,000
N/D
N/D
>10,000
NH2
O
9260
N
N
O
10r
10s
R
S
7000
604
73
2630
940
156
N
N
10t
R
R
109
18
175
HO
OH
10u17
>10,000
N/D
>10,000
OH
OH
O
Weinreb’s hydroxamate 11,12 whose condensation with
the Grignard reagent 4-benzyloxyphenyl magnesium
iodide proceeded to give the expected ketone 12 in
modest yield. TFA deprotection afforded the ammo-
nium salt 13, which underwent selective acylation fol-
lowed by phosphorylation. TFA deprotection provided
ketones 14a–b.
Table 3. Comparison of ketone derivatives 14a–b with tyrosine-based
lead compound 1
Compounds S/R
LPA1
LPA3 IC50
(nM)
IC50 (nM) Ki (nM)
111
S
S
R
5210
>10,000
2490
137
N/D
N/D
6450
>10,000
>10,000
14a
14b
All compounds were characterized by 1H and 13C NMR,
mass spectroscopy and, in certain cases, elemental
analysis.13
The synthesis of ketone derivatives 14a–b is outlined in
Scheme 3. Serine was converted in three steps into

2738
B. H. Heasley et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2735–2740
O
That is, compounds that produced rightward, parallel
H
N
O
H
N
R
shifts in the concentration–response curves as a function
of antagonist concentration were considered surmount-
able antagonists for which a reliable Ki value could be
calculated for the LPA1 receptor. LPA1 IC50 values are
also listed for comparison with LPA3 data. The com-
pounds presented here were devoid of any significant
activity at the LPA2 receptor.
O
MeO
a, b or
c (R = 17:2)
HO
O
O
Ar
O
Ar
O
2
3
O
H
N
HO
HO
P
R
O
d, e, f
Optimization of the N-acyl moiety (Table 1), the out-
ermost benzyl substituent and stereochemistry of 1
(Table 2) led to the development of an SAR as described
below. In addition, two ketone derivatives of 1 (Table 3)
were evaluated.
O
Ar
O
4
Scheme 1. Synthesis of compounds 4a–e. Reagents and conditions: (a)
SOCl2, MeOH, 18 h, 100%; (b) acyl halide, DIEA, CH2Cl2, 4 h, 74–
85%; (c) linoleic acid, PyBOP, DIEA, 18 h, 70%; (d) NaBH4, CaCl2,
EtOH–THF (2:1), 18 h, 95%; (e) di-tert-butyldiisopropyl phospho-
ramidite, tetrazole, 4 h; then 30% H2O2, 4 h, 75–80%; (f) TFA–CH2Cl2
(1:2), 1 h, 95–100%.
Phospholipid chain length plays a critical role in LPA
receptor antagonist binding (Table 1). N-Oleoyl lead
compound 1 displays potent LPA1/LPA3 dual antago-
nism.11 At LPA1, active compounds generally contained
relatively long chain length and unsaturation (1 and 4e)
in the N- acyl region. Palmitoyl derivatives 4c–d showed
no activity at LPA1 while the unsaturated derivative 4e
displayed potency comparable to lead compound 1. At
LPA3, palmitoyl derivatives 4c–d were comparable in
activity to 1 while shorter chain lengths and increased
unsaturation led to a marked reduction in inhibitory
activity. Thus compound 4e, which contains a linoleoyl-
amide shows selectivity for LPA1 while palmitoyl
derivatives 4c–d were active only at LPA3.
A GTP[c35S] binding assay was adapted to assess in
vitro activity.14 The compounds presented in Tables 1–3
were assayed for their ability to antagonize LPA-evoked
GTP[c35S] binding. However, the differential affinity of
LPA1 versus LPA3 for 1-oleoyl-LPA prohibited a uni-
form analysis at each receptor. Specifically, the relatively
low affinity of the LPA3 receptor for 1-oleoyl-LPA im-
peded the determination of accurate Ki values for each
antagonist. The detergent-like properties of the ligands
prohibit the use of concentrations greater than 30 lM,
which are required to achieve a statistically significant
rightward shift in the sigmoidal curve requisite to the Ki
determination. Consequently, the relative potencies, in
the form of IC50 values,15 are shown for the LPA3
receptor in Tables 1–3. The Ki values as determined by
Schild regression11 are shown for the LPA1 receptor.
As described in Table 2, the outermost benzyl substituent
of LPA antagonists can be modified with various ether/
ester linkages to improve the potency and selectivity of
lead compound 1. LPA receptors also clearly exhibit
stereochemical recognition as has been documented
elsewhere.11;16 The N-oleoyl lipid chain is held constant
among derivatives 10a–u as this feature has been shown
O
O
H
N
NH3Cl
a
MeO
MeO
O
OH
RO
O
6 (R = H)
b
5
7 (R = TBDMS)
O
P
H
N
O
O
c, d
O
8 (R = TBDMS)
9 (R = H)
RO
e
O
P
H
N
HO
HO
f (X = H2) or
g (X = O), h
O
O
X
R
O
10
Scheme 2. Synthesis of compounds 10a–u. Reagents and conditions: (a) oleoyl chloride, DIEA, CH2Cl2, 4 h, 85%; (b) TBDMSCI, DIEA, DMF, 4 h,
96%; (c) NaBH4, CaCl2, EtOH–THF (2:1), 18 h, 95%; (d) di-tert-butyldiisopropyl phosphoramidite, tetrazole, 4 h; then 30% H2O2, 4 h, 83%; (e)
TBAFÆ3H2O, THF, 1 h, 95%; (f) ROH, PPh3, DIAD, CH2Cl2, 18 h, 73–88%; (g) RCOOH, DIEA, PyBOP, 18 h, 50–70%; (h) TFA–CH2Cl2 (1:2), 1 h,
95–100%.

B. H. Heasley et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2735–2740
2739
O
NH3+CF3COO-
O
HO
MeO
N
a
b
O
O
O
N
N
Boc
BnO
Boc
12
OBn
11
13
O
HO P
H
N
O
HO
c, d, e
O
O
BnO
14
Scheme 3. Synthesis of compounds 14a–b. Reagents and conditions: (a) 4-benzyloxyphenyl iodide, Mg0, Et2O, 25 ꢂC; (b) TFA–CH2Cl2 (1:2), 1 h,
then Et2O precipitation, 26% over two steps; (c) oleoyl chloride, DIEA, CH2Cl2, 4 h, 77%; (d) di-tert-butyldiisopropyl phosphoramidite, tetrazole,
4 h; then 30% H2O2, 4 h, 75%; (e) TFA–CH2Cl2 (1:2), 1 h, 100%.
to afford dual LPA1/LPA3 antagonism and good potency
LPA1
(Table 1). When the benzyl moiety of 1 is replaced with
small alkyl substituents as in compounds 10a–e, inhibi-
tory activity is generally diminished. Relatively bulkier
alkyl ethers 10f–i, display varying degrees of antagonism
at LPA3 but are inactive at LPA1. Electron-rich aromatic
systems in place of the outermost benzyl moiety confer
activity to LPA antagonists. This is exemplified by
derivatives 10j–m, which show good to excellent potency,
particularly at LPA3. Sterically demanding electron-rich
aromatic substituents, in general, retain the potency of 1
at LPA3 but exhibit diminished activity at LPA1. This
trend is particularly evident in compounds 10m and 10r.
However, 4-npentylbenzoyl derivative 10o was inactive at
both receptors, a result which implies some limitation to
the steric tolerance of the LPA3 antagonist binding
pocket. Polar groups such as 4-aminobenzoyl (10p) and
glucuronide 10u17 were also devoid of activity. To our
delight, 2-pyridyl derivatives 10s–t were significantly
more potent than lead compound 1 at LPA1 and LPA3
receptors. R Enantiomer 10t improves on the antagonist
activity of 1 by approximately one log order at each
receptor (Figs. 2 and 3). Indeed, all of the R derivatives in
Table 2 are more potent than their S counterparts at
LPA receptors.
600
500
400
300
200
100
0
VPC12249(1) + 1u MLPA
10t + 1uM LPA
14b + 1uM LPA
-10
-9
-8
-7
-6
-5
-4
Log[compd]
Figure 2. Effect of LPA antagonists on LPA1 receptor GTP[c35S]
binding.15
Thus, in comparison to 1, electron-rich aromatic sys-
tems in place of the outermost benzyl moiety improve
antagonist activity at both receptors (10k). LPA3 is
more tolerant of sterically bulky substituents, which
provides a strategy to realize LPA3-selective antagonists
with modest potency such as 10g and 10m. Finally, N-
heterocycle derivative 10t is a high affinity dual anta-
gonist, with a Ki value of 18 nM at the LPA1 receptor
and IC50 of 175 nM at LPA3.
LPA3
3000
2000
Two ketone derivatives of 1, oxidized at the innermost
benzylic carbon atom, were evaluated at LPA receptors
(Table 3). In this series (14a–b), a stereochemical pref-
erence for the R enantiomer is apparent at the LPA1
receptor. Installation of the carbonyl in derivatives 14a–
b completely diminishes activity at the LPA3 receptor
while 14b retains potency comparable to 1 at LPA1.
Thus, 14b represents an LPA1-selective blocker of
modest potency.
VPC12249 (1 )+ 5uM LPA
10t + 5uM LPA
10g + 5uM LPA
1000
0
-10
-9
-8
-7
-6
-5
-4
Log[compd]
Figure 3. Effect of LPA antagonists on LPA3 receptor GTP[c35S]
binding.15
In summary, an initial SAR of benzyl-4-oxybenzyl-
substituted NAEPA analogues is described with regard

2740
B. H. Heasley et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2735–2740
10. Okusa, M. D.; Ye, H.; Huang, L.; Sigismund, L.; Heise,
C.; Santos, W.; Macdonald, T.; Lynch, K. R. Am. J.
Physiol. Renal Physiol. 2003, 285, F565–F574.
11. Heise, C. E.; Santos, W. L.; Schreihofer, A. M.; Heasley,
B. H.; Mukhin, Y. V.; Macdonald, T. L.; Lynch, K. R.
Mol. Pharm. 2001, 60, 1173–1180.
12. Ageno, G.; Banfi, L.; Cascio, G.; Guanti, G.; Manghisi,
E.; Reva, R.; Rocca, V. Tetrahedron 1995, 51, 8121–8134.
13. (S)-Phosphoric acid mono-{2-octadec-9-enoylamino-3-[4-
(pyridin-2-ylmethoxy)-phenyl]-propyl} ester (10s): 1H
to LPA receptor antagonism. This study has resulted in
the discovery of a high-affinity LPA1/LPA3 receptor
antagonist (10t), which exhibits a Ki value of 18 nM at
LPA1 in our GTP[c35S] in vitro binding assay. The
findings presented herein will be a useful platform for
further optimization of lead derivative 10t. Additionally,
it will be critical to improve on the metabolic stability of
the hydrolytically-labile phosphate head group so that
in vivo studies might be employed. These efforts will be
published in due course.
NMR (300 MHz, CD3OD/CDCl3)
d 0.62 (t, 3H,
J ¼ 7:2 Hz), 0.95–1.10 (m, 11H), 1.22–1.32 (m, 2H),
1.70–1.79 (m, 4H), 1.89 (t, 2H, J ¼ 7:9 Hz), 2.51 (dd,
1H, J ¼ 7:7, 13.6 Hz), 2.61 (dd, 1H, J ¼ 7:7, 13.6 Hz),
3.68–3.73 (m, 2H), 3.92–4.03 (m, 1H), 4.91 (s, 2H), 5.07 (t,
2H, J ¼ 5:5 Hz), 6.66 (d, 2H, J ¼ 7:9 Hz), 6.91 (d, 2H,
J ¼ 7:9 Hz), 7.09 (t, 1H, J ¼ 6:4 Hz), 7.84 (d, 1H,
J ¼ 8:1 Hz), 7.69 (t, 1H, J ¼ 8:6 Hz), 8.29 (d, 1H,
J ¼ 5:0 Hz); 13C NMR (300 MHz, CD3OD/CDCl3) d
13.19, 22.01, 25.21, 26.52, 28.52, 28.68, 28.87, 29.11,
31.27, 35.15, 35.68, 50.16, 50.26, 66.21, 69.15, 114.15,
121.58, 122.75, 129.11, 129.27, 129.80, 130.10, 137.68,
147.62, 148.31, 156.19, 156.43; MS (ESI) m=z 601.74
[MÀHþ, 100%]; anal. calcd for C33H51N2O6P: C, 65.76; H,
8.53; N, 4.65; found: C, 65.06; H, 8.57; N, 4.75.
(R)-Phosphoric acid mono-{2-octadec-9-enoylamino-3-[4-
(pyridin-2-ylmethoxy)-phenyl]-propyl} ester (10t): This
product had spectroscopic properties identical to its
enantiomer.
Acknowledgements
Support by NIH (R01-GM52722).
We thank Dr. Mahendra Chordia for reviewing this
manuscript.
References and notes
1. Tigyi, G.; Parrill, A. L. Prog. Lipid Res. 2003, 42, 498–526.
2. Mills, G. B.; Moolenaar, W. H. Nat. Rev. Cancer 2003, 3,
582–591.
3. Lynch, K. R.; Macdonald, T. L. Prost. Lipid Med. 2001,
64, 33–45.
14. 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.
4. Tokumura, A. Prog. Lipid Res. 1995, 34, 151–184.
5. Contos, J. J. A.; Ishii, I.; Chun, J. Mol. Pharm. 2000, 58,
1188–1196.
6. 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. Proc. Natl. Acad. Sci.
U.S.A. 2003, 100, 131–136.
7. Noguchi, K.; Ishii, S.; Shimizu, T. J. Biol. Chem. 2003,
278, 25600–25606.
8. Xu, Y.; Gaudette, D. C.; Boynton, J. D.; Frankel, A.;
Fang, X.-J.; Sharma, A.; Hurteau, J.; Casey, G.; Good-
body, A.; Mellors, A.; Holub, B. J.; Mills, G. B. Clin.
Cancer Res. 1995, 1, 1223–1232.
15. DNAs encoding LPA1, LPA2, or LPA3 receptors were
transfected into HEK293T cells along with DNAs encod-
ing Gai2, Gb, and Gc2 proteins. After 48 h, microsomes
were prepared and used in a GTP[c35S] binding assay, as
described.14 For determination of IC50 values, a concen-
tration of 1-oleoyl-LPA equal to the EC80 value (LPA1:
1 lM; LPA3: 10 lM) was collided with antagonists at
concentrations ranging from 0.1 to 30,000 nM. Points are
in triplicate and are representative of at least two
experiments.
16. 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.
9. Hayashi, K.; Takahashi, M.; Nishida, W.; Yoshida, K.;
Ohkawa, Y.; Kitabatake, A.; Aoki, J.; Arai, H.; Sobue, K.
Circ. Res. 2001, 89, 251–258.
17. For reasons mentioned in the text above, 10u is the only
compound in Table 2 which contains a saturated 18
carbon length acyl chain.