Systemic and intrathecal effects of a novel series of phospholipase A 2 inhibitors on hyperalgesia and spinal prostaglandin E2 release

Article abstract of DOI:10.1124/jpet.105.091686

Phospholipase A₂ (PLA₂) forms are expressed in spinal cord, and inhibiting spinal PLA₂ induces a potent antihyperalgesia. Here, we examined the antihyperalgesic effects after systemic and i.t. delivery of four compounds constructed with a common motif consisting of a 2-oxoamide with a hydrocarbon tail and a four-carbon tether. These molecules were characterized for their ability to block group IVA calcium-dependent PLA ₂ (cPLA₂) and group VIA calcium-independent PLA ₂ (iPLA₂) in inhibition assays using human recombinant enzyme. The rank ordering of potency in blocking group IVA cPLA₂ was AX048 (ethyl 4-[(2-oxohexadecanoyl)amino]butanoate), AX006 (4-[(2- oxohexadecanoyl)amino]butanoic acid), and AX057 (tert-butyl-4-[(2- oxohexadecanoyl)amino]butanoate) > AX010 (methyl 4-[(2-oxohexadecanoyl)amino] butanoate) and for inhibiting group VIA iPLA₂ was AX048, AX057 > AX006, and AX010. No agent altered recombinant cyclooxygenase activity. In vivo, i.t. (30 μg) and systemic (0.2-3 mg/kg i.p.) AX048 blocked carrageenan hyperalgesia and after systemic delivery in a model of spinally mediated hyperalgesia induced by i.t. substance P (SP). The other agents were without activity. In rats prepared with lumbar i.t. loop dialysis catheters, SP evoked spinal prostaglandin E₂ (PGE₂) release. AX048 alone inhibited PGE₂ release. Intrathecal SR141617, a cannabinoid CB1 inhibitor at doses that blocked the effects of i.t. anandamide had no effect upon i.t. AX048. These results suggest that AX048 is the first systemically bioavailable compound with a significant affinity for group IVA cPLA ₂, which produces a potent antihyperalgesia. The other agents, although demonstrating enzymatic activity in cell-free assays, appear unable to gain access to the intracellular PLA₂ toward which their action is targeted. Copyright

Full text of DOI:10.1124/jpet.105.091686

0022-3565/06/3161-466–475$20.00
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 2006 by The American Society for Pharmacology and Experimental Therapeutics
JPET 316:466–475, 2006
Vol. 316, No. 1
91686/3069213
Printed in U.S.A.
Systemic and Intrathecal Effects of a Novel Series of
Phospholipase A2 Inhibitors on Hyperalgesia and Spinal
Prostaglandin E2 Release
Tony L. Yaksh, George Kokotos, Camilla I. Svensson, Daren Stephens,
Christoforos G. Kokotos, Bethany Fitzsimmons, Dimitra Hadjipavlou-Litina, Xiao-Ying Hua,
and Edward A. Dennis
Departments of Anesthesiology (T.L.Y., C.I.S., B.F., X.-Y.H.) and Chemistry and Biochemistry (D.S., E.A.D.), School of
Medicine, University of California, San Diego, La Jolla, California; Department of Chemistry, University of Athens, Athens,
Greece (G.K., C.G.K.); and Department of Pharmaceutical Chemistry, University of Thessaloniki, Thessaloniki, Greece (D.H.-L.)
Received June 28, 2005; accepted September 30, 2005
ABSTRACT
Phospholipase A2 (PLA2) forms are expressed in spinal cord, agent altered recombinant cyclooxygenase activity. In vivo, i.t.
and inhibiting spinal PLA2 induces a potent antihyperalgesia. (30 ␮g) and systemic (0.2–3 mg/kg i.p.) AX048 blocked carra-
Here, we examined the antihyperalgesic effects after systemic geenan hyperalgesia and after systemic delivery in a model of
and i.t. delivery of four compounds constructed with a common spinally mediated hyperalgesia induced by i.t. substance P
motif consisting of a 2-oxoamide with a hydrocarbon tail and a (SP). The other agents were without activity. In rats prepared
four-carbon tether. These molecules were characterized for with lumbar i.t. loop dialysis catheters, SP evoked spinal pros-
their ability to block group IVA calcium-dependent PLA2 taglandin E2 (PGE2) release. AX048 alone inhibited PGE2 re-
(cPLA2) and group VIA calcium-independent PLA2 (iPLA2) in lease. Intrathecal SR141617, a cannabinoid CB1 inhibitor at
inhibition assays using human recombinant enzyme. The rank doses that blocked the effects of i.t. anandamide had no effect
ordering of potency in blocking group IVA cPLA2 was AX048 upon i.t. AX048. These results suggest that AX048 is the first
(ethyl 4-[(2-oxohexadecanoyl)amino]butanoate), AX006 (4-[(2- systemically bioavailable compound with a significant affinity
oxohexadecanoyl)amino]butanoic acid), and AX057 (tert-butyl for group IVA cPLA2, which produces a potent antihyperalge-
4-[(2-oxohexadecanoyl)amino]butanoate)
Ͼ AX010 (methyl sia. The other agents, although demonstrating enzymatic ac-
4-[(2-oxohexadecanoyl)amino]butanoate) and for inhibiting tivity in cell-free assays, appear unable to gain access to the
group VIA iPLA2 was AX048, AX057 Ͼ AX006, and AX010. No intracellular PLA2 toward which their action is targeted.
Tissue injury and inflammation lead to the development of bition of spinal COX also led to reversal of the facilitated
an evident facilitation in the sensitivity to moderately aver- state (Yaksh, 1982; Taiwo and Levine, 1988). These initial
sive stimuli, e.g., hyperalgesia. It has been long appreciated findings have been widely confirmed (Yamamoto and Nozaki-
that this phenomenon is diminished by agents that block Taguchi, 1996; Turnbach and Randich, 2001) Consistent
cyclooxygenase (COX) activity (Vane, 1971). Although early
work suggested that this action resulted from a peripheral
effect (Ferreira, 1972), it was subsequently found that inhi-
with this action, persistent small afferent input, as arises
from tissue injury, was shown to evoke a significant spinal
release of prostanoids in vivo in a manner that was blocked
by spinally delivered COX inhibitors (Yaksh, 1982; Malm-
berg and Yaksh, 1992, 1995; Southall et al., 1998; Ebers-
berger et al., 1999; Samad et al., 2001; Yaksh et al., 2001). An
important element of prostaglandin (PG) synthesis is phos-
pholipase A2 (PLA2) because it is required to generate ara-
chidonic acid, which is the substrate for COX-mediated
This work was supported by AnalgesiX/University of California Biotechnol-
ogy Grant B1002-10303. The Regents of the University of California (G.K.,
E.A.D., and T.L.Y.) hold equity in AnalgesiX.
Article, publication date, and citation information can be found at
http://jpet.aspetjournals.org.
doi:10.1124/jpet.105.091686.
ABBREVIATIONS: COX, cyclooxygenase; PG, prostaglandin; PLA2, phospholipase A2; cPLA2, calcium-dependent PLA2; iPLA2, calcium-inde-
pendent PLA2; AX006, 4-[(2-oxohexadecanoyl)amino]butanoic acid; AX010, methyl 4-[(2-oxohexadecanoyl)amino]butanoate; AX048, ethyl 4-[(2-
oxohexadecanoyl)amino]butanoate; AX057, tert-butyl 4-[(2-oxohexadecanoyl)amino]butanoate; SP, substance P; NMDA, N-methyl-D-aspartate;
DMSO, dimethyl sulfoxide; SR141716A, N-(piperidin-1-yl)-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboximide hydro-
chloride; ANOVA, analysis of variance; AACOCF3, arachidonyl trifluoromethylketone; BMS-229724, 4-[4-[2-[2-[bis(4-chlorophenyl)methoxy]ethyl-
sulfonyl]ethoxy]phenyl]-1,1,1-trifluoro-2-butanone.
466

Phospholipase A2 Inhibition and Pain
467
prostanoid formation. In recent work, we have shown the Behavioral Analysis
presence of constitutive mRNA and protein in the spinal cord
for group IVA calcium-dependent PLA2 (cPLA2) and group
VIA calcium-independent PLA2 (iPLA2) and groups II and V
secretory PLA2 forms (Lucas et al., 2005; Svensson et al.,
2005b). Inhibition of group IV cPLA2 but not group VI iPLA2
isoforms using i.t.-delivered agents suggested a role for group
IV cPLA2, but not group VI iPLA2 (Lucas et al., 2005), in
inflammation-evoked hyperalgesia.
We have reported recently the discovery of a novel struc-
tural series of 2-oxoamides that inhibit group IVA cPLA2 in
vitro and in vivo (Kokotos et al., 2002, 2004). In initial work,
2-oxoamides were observed to inhibit inflammation in the rat
paw carrageenan-induced edema assay (Kokotos et al., 2004).
In the present work, we have focused on the in vivo activity
of four related analogs of this series, AX006, AX010, AX048,
Thermal Hyperalgesia. Two approaches were employed to ini-
tiate a hyperalgesic state. An inflammation-evoked thermal hyper-
algesia was induced by subcutaneous injection of 2 mg of carra-
geenan [Sigma-Aldrich, St. Louis, MO; 100 ␮l of 20% solution (w/v) in
physiological saline] into the plantar surface of the left hind paw.
The thermally evoked paw withdrawal response was assessed (Dirig
et al., 1997). In brief, the device consists of a glass surface (main-
tained at 25°C) on which the rats are placed individually in Plexiglas
cubicles (9 ϫ 22 ϫ 25 cm). The thermal nociceptive stimulus origi-
nates from a focused projection bulb positioned below the glass
surface. The stimulus is delivered separately to either hind paw of
each test subject with the aid of an angled mirror mounted on the
stimulus source. A timer is actuated with the light source, and
latency is defined as the time required for the paw to show a brisk
withdrawal as detected by photodiode motion sensors that stop the
timer and terminate the stimulus. Paw withdrawal latencies are
and AX057. These molecules were examined for their inhib- assessed prior to any treatment (control) and at intervals after
treatment. Left (injured) and right (uninjured) paw withdrawal la-
tencies are assessed and plotted versus time. In addition, difference
latency scores (uninjured Ϫ injured) are calculated, and the average
withdrawal latencies over the postinjection observation intervals are
calculated for comparison between treatment groups. In addition to
the use of a peripheral inflammation, a thermal hyperalgesia is also
initiated by the i.t. injection of SP (20 nmol/10 ␮l). The mean paw
withdrawal latency of the left and right paws is assessed at each time
point. The mean difference between the pre- and post-i.t. SP re-
sponse latency scores is calculated for analysis.
itory effects on group IV cPLA2 and group VI iPLA2 as well as
on COX activity in in vitro assays. Their actions were then
characterized after systemic and i.t. delivery on thermal hy-
peralgesia induced by peripheral inflammation (intraplantar
carrageenan). In addition, we have shown previously that
spinal sensitization can be directly initiated in the absence of
peripheral inflammation by spinal delivery of substance P
(SP). Substance P, acting through the spinal neurokinin 1
receptor, will evoke the spinal release of PGE2 and subse-
quent thermal hyperalgesia. Both of these events are antag-
onized by spinal cyclooxygenase inhibition (Malmberg and
Yaksh, 1992; Yaksh et al., 2001). Based on these observa-
tions, we examined the effects of the PLA2 inhibitors on the
hyperalgesia and PGE2 release evoked by spinally delivered
SP. We report here that one of these agents, after systemic
delivery, displays significant antihyperalgesic effects in mod-
els of both centrally and peripherally initiated hyperalgesia
and in an effective systemic dose blocks the spinally evoked
release of spinal PGE.
Intrathecal Dialysis and PGE2 Assay. Spinal dialysis experi-
ments to define the spinal release of PGE2 were conducted in un-
anesthetized rats 3 days after dialysis catheter implantation. A
syringe pump (Harvard Apparatus Inc., Holliston, MA) was con-
nected, and dialysis tubing was perfused with artificial cerebrospinal
fluid at a rate of 10 ␮l/min. The artificial cerebrospinal fluid con-
tained 151.1 mM Naϩ, 2.6 mM Kϩ, 0.9 mM Mg2ϩ, 1.3 mM Ca2ϩ
,
122.7 mM ClϪ, 21.0 mM HCO3, 2.5 mM HPO4, and 3.5 mM dextrose
and was bubbled with 95% O2/5% CO2 before each experiment to
adjust the final pH to 7.2. The efflux (20 min/fraction) was collected
in an automatic fraction collector (Eicom, Kyoto, Japan) at 4°C. Two
baseline samples were collected following a 30-min washout and an
additional three fractions after i.t. injection of NMDA (0.6 ␮g). The
concentration of PGE2 in spinal dialysate was measured by enzyme-
linked immunosorbent assay using a commercially available kit (As-
say Designs 90001; Assay Designs, Ann Arbor, MI). The antibody is
Materials and Methods
All experiments were carried out according to protocols approved
by the Institutional Animal Care Committee of University of Cali-
fornia, San Diego.
selective for PGE2 with less than 2.0% cross-reactivity to PGF1␣
,
PGF2␣, 6-ketoPGF1␣, PGA2, or PGB2 but cross-reacts with PGE1 and
PGE3.
In Vivo Studies
Drug Delivery. Drugs were delivered systemically (i.p.) or spi-
nally (i.t.). Intraperitoneal drugs were delivered uniformly in doses
prepared in volumes of 0.5 ml/kg. Drugs injected i.t. were adminis-
tered in a total volume of 10 ␮l followed by a 10-␮l flush using
vehicle.
Animals. Male Holtzman Sprague-Dawley rats (300–350 g; Har-
lan, Indianapolis, IN) were individually housed and maintained on a
12-h light/dark cycle with free access to food and water.
Intrathecal Catheter Implantation. For spinal drug injections,
lumbar catheters were implanted in rats under isoflurane anesthesia
according to a modification of the procedure described by Yaksh and
Rudy (1976). A polyethylene catheter (PE- 5, 0.014 in outside diam-
eter; Spectranetics, Colorado Springs, CO) was inserted into the i.t.
space and advanced to the rostral edge of the lumbar enlargement
through an incision in the atlanto-occipital membrane. Five days
after implantation, rats were entered into the study. In separate
experiments to assess spinal prostaglandins release, rats were pre-
pared with lumbar loop dialysis catheters with three lumens, as
previously described (see Yaksh et al., 2001). In brief, the outer two
lumens were connected to a length of dialysis tubing (10-kDa cut-off).
The catheter was then implanted i.t. using the same technique as
described above for the i.t. catheter. A 3-day interval was allowed to
elapse prior to including the animal in a study. In all cases, the
Enzyme Assays. In vitro group IV cPLA2 and group VI iPLA2
assays were done as previously described (Kokotos et al., 2002).
Briefly, 100 ␮M lipid substrate and 100,000 cpm radiolabeled analog
were dried down under N2 and dissolved in assay buffer containing
400 ␮M Triton X-100 to yield a mixed micelle substrate solution.
Inhibitors dissolved in DMSO were added to the reaction tubes and
allowed to incubate with substrate for 5 min at 40°C. Pure enzyme
was added to yield a final volume of 500 ␮l, and digestion was carried
out at 40°C for 30 min. Reactions were quenched and extracted using
the Dole method, and products were quantified by liquid scintillation
counting (Dole, 1956). Percent inhibition was determined at a range
of inhibitor mole fraction concentrations for XI(50) calculations.
Inhibition of cyclooxygenase-1 and cyclooxygenase-2 was tested in
exclusion criteria were the presence of any neurological sequelae, vitro using the COX Activity Assay kit (catalog no. 760151) from
20% weight loss after implantation, or catheter occlusion. Cayman Chemical (Ann Arbor, MI). Assays were performed in 96-

468
Yaksh et al.
well plates using 10 ␮l of supplied COX standard (catalog no.
C24H47NO4 (413.63): C, 69.69; H, 11.45, N, 3.39. Found: C, 69.42; H,
760152) that contained COX-1 and COX-2 proteins. Activity was 11.61; N, 3.27.
detected colorimetrically at 595 nm by the appearance of oxidized
Oxidation of 2-Hydroxy-Amides
N,N,NЈ,NЈ-tetramethylphenylenediamine, which has an absorption
maximum of 611 nm (Kulmacz and Lands, 1983). Inhibitors dis-
solved in DMSO (study compounds) or ethanol (indomethacin) were
added to 50 ␮M final concentration and allowed to incubate with the
To a solution of a 2-hydroxy-amide (1.00 mmol) in a mixture of
toluene-EtOAc (15 ml), a solution of NaBr (0.11 g, 1.05 mmol) in
water (1.3 ml) was added, followed by 4-acetamido-2,2,6,6-tetra-
methylpiperidine-1-yloxy free radical (2 mg, 0.01 mmol). To the
resulting biphasic system, which was cooled at Ϫ5°C, an aqueous
solution of 0.35 M NaOCl (3.1 ml, 1.10 mmol) containing NaHCO3
(0.25 g, 3 mmol) was added drop wise while stirring vigorously at
Ϫ5°C over a period of 1 h. After the mixture had been stirred for a
further 15 min at 0°C, EtOAc (15 ml) and H2O (5 ml) were added.
The aqueous layer was separated and washed with EtOAc (10 ml).
The combined organic layers were washed consecutively with 5%
aqueous citric acid (15 ml) containing KI (0.04 g), 10% aqueous
Na2S2O3 (6 ml), and brine and dried over Na2SO4. The solvents were
evaporated under reduced pressure, and the residue was purified
by column chromatography [EtOAc-petroleum ether 1:9 (b.p., 40–
60°C)].
assay mixture including enzyme for
5 min. After addition of
N,N,NЈ,NЈ-tetramethylphenylenediamine and arachidonic acid,
samples were mixed and allowed to incubate for 5 min at room
temperature before reading absorbance at 595 nm to determine
results. Results were calculated, and percent inhibition values were
derived.
Drugs
PLA2 inhibitors employed in these studies were synthesized (see
below). These agents were prepared for delivery in a vehicle of 5%
Tween 80. Other agents used in these studies included the cannabi-
noid agonist anandamide and the CB1 antagonist SR141716A (sup-
plied courtesy of Benjamin Cravatt, Scripps Institute, La Jolla, CA).
Anandamide was prepared in 100% DMSO and SR141716A in eth-
anol Emulphor and saline (1:1:18). Control studies were run with the
respective vehicles.
AX048. Yield 86%; white solid; mp 63–64°C; 1H NMR. ␦ 7.16 (1H,
m, NH), 4.12 (2H, q, J ϭ 7 Hz, COOCH2CH3), 3.33 (2H, m, CH2NH),
2.89 (2H, t, J ϭ 7 Hz, CH2COCO), 2.34 (2H, t, J ϭ 7 Hz, CH2COO),
1.87 (2H, m, CH2CH2COO), 1.57 (2H, m, CH2CH2COCO), 1.40–1.15
(25H, m, 11ϫCH2, COOCH2CH3), 0.85 (3H, t, J ϭ 7 Hz, CH3); 13C
NMR: ␦ 199.0, 172.7, 160.2, 60.4, 38.5, 36.5, 31.7, 31.4, 29.5, 29.4,
29.3, 29.2, 28.9, 24.2, 23.0, 22.5, 14.0, 13.9; MS (FAB) m/z (%) 384
(100) [Mϩϩ H]. Anal. calcd. for C22H41NO4 (383.57): C, 68.89; H,
10.77, N, 3.65. Found: C, 68.71; H, 10.88; N,3.54.
Drug Synthesis
AX006 and AX010 were prepared as previously described (Kokotos
et al., 2002, 2004). The synthesis and the characterization of the
novel agents AX048 and AX057 are described here in detail. Figure
1 summarizes the synthesis schema.
AX057. Yield 95%; white solid; mp 61–62°C; 1H NMR. ␦ 7.11 (1H,
m, NH), 3.33 (2H, m, CH2NH), 2.91 (2H, t, J ϭ 7 Hz, CH2CO), 2.28
(2H, t, J ϭ 7 Hz, CH2COO), 1.84 (2H, m, CH2CH2COO), 1.60 (2H, m,
CH2CH2COCO), 1.45 [9H, s, C(CH3)3], 1.38–1.23 (22H, m, 11ϫCH2),
0.89 (3H, t, J ϭ 7 Hz, CH3); 13C NMR: ␦ 198.6, 171.6, 159.7, 80.0,
38.1, 36.1, 32.2, 31.3, 29.0, 28.9, 28.8, 28.7, 28.4, 27.4, 23.8, 22.5,
22.0, 13.5; MS (FAB) m/z (%) 412 (17) [Mϩϩ H], 356 (100). Anal.
calcd. for C24H45NO4 (411.62): C, 70.03; H, 11.02, N, 3.40. Found: C,
69.89; H, 11.32; N, 3.47.
For coupling of 2-hydroxy-hexadecanoic acid with esters of 4-ami-
no-butanoate, to a stirred solution of 2-hydroxy-hexadecanoic acid
(2.0 mmol) and the ester of 4-amino-butanoate (2.0 mmol) in CH2Cl2
(20 ml), Et3N (6.2 ml, 4.4 mmol) and subsequently 2-hydroxy-hexa-
decanoic acid using 1-(3-dimethylaminopropyl)-3-ethyl carbodiimide
(0.42 g, 2.2 mmol) and 1-hydroxylbenzotriazole (0.32 g, 2.0 mmol)
were added at 0°C. The reaction mixture was stirred for 1 h at 0°C
and overnight at room temperature. The solvent was evaporated
under reduced pressure, and EtOAc (20 ml) was added. The organic
layer was washed consecutively with brine, 1 N HCl, brine, 5%
NaHCO3, and brine, dried over Na2SO4, and evaporated under re-
duced pressure. The residue was purified by column chromatography
using CHCl3/MeOH (95:5) as the eluent.
Ethyl 4-[(2-hydroxyhexadecanoyl)amino]butanoate. Yield
72%; 1H NMR. ␦ 6.68 (1H, t, J ϭ 7 Hz, NH), 4.13 (3H, m, CH,
COOCH2CH3), 3.34 (2H, m, CH2NH), 2.68 (1H, b, OH), 2.32 (2H, t,
J ϭ 7 Hz, CH2COO), 1.80–1.58 (4H, m, CH2CH2COO, CH2CH),
1.45–1.23 (27H, m, 12ϫCH2, COOCH2CH3), 0.85 (3H, t, J ϭ 7 Hz,
CH3); 13C NMR: ␦ 174.0, 173.8, 72.2, 60.6, 38.5, 34.9, 31.9, 31.7, 31.4,
29.7, 29.6, 29.5, 29.4, 29.3, 25.0, 24.6, 22.7, 14.1. Anal. calcd. for
Statistics
Escape latency data are presented as the mean Ϯ S.E.M. For
carrageenan and i.t. SP analysis of thermal escape, latencies were
carried out over time and compared with one-way ANOVA. For
carrageenan, difference scores between control and injured paws
over time were calculated for each group. Comparison of drug with
vehicle treatment was performed using an unpaired Student’s t test.
For dose-response analyses, least-squares linear regression was per-
formed, and the drug dose required to produce a 50% reduction in the
hyperalgesia otherwise observed in the vehicle-treated control ani-
mals was estimated. For release studies, release was expressed as
percentage of baseline, and the area under the release curve follow-
ing i.t. SP was calculated. Group comparisons were carried out using
nonparametric statistics for repeated measures over time and (Fried-
man analysis) across treatment groups with post hoc analyses being
undertaken with Dunns Multiple Comparison analysis. Analyses
were performed using Prism statistical software (GraphPad Prism
version 4.02 for Macintosh; GraphPad Software Inc., San Diego, CA).
C
22H43NO4 (385.58): C, 68.53; H, 11.24, N, 3.63. Found: C, 68.12; H,
11.32; N, 3.48.
Tert-Butyl 4-[(2-hydroxyhexadecanoyl)amino]butanoate.
Yield 64%; 1H NMR. ␦ 6.49 (1H, t, J ϭ 7 Hz, NH), 4.12 (1H, m, CH),
3.34 (2H, m, CH2NH), 2.73 (1H, b, OH), 2.27 (2H, t, J ϭ 7 Hz,
CH2COO), 1.82–1.49 (4H, m, CH2CH2COO, CH2CH), 1.45 [9H, s,
C(CH3)3], 1.38–1.15 (24H, m, 12ϫCH2), 0.89 (3H, t, J ϭ 7 Hz, CH3);
13C NMR: ␦ 173.9, 173.7, 80.1, 72.3, 38.3, 35.4, 31.9, 31.8, 31.4, 29.7,
29.6, 29.5, 29.4, 29.3, 28.7, 25.1, 24.5, 22.8, 14.1. Anal. calcd. for
Results
Synthesis and Physical Properties of Test Agent
Ethyl and tert-butyl 4-amino-butonate were coupled with
2-hydroxy-hexadecanoic acid using 1-(3-dimethylaminopro-
pyl)-3-ethyl carbodiimide as a condensing agent in the pres-
ence of 1-hydroxylbenzotriazole. The 2-hydroxyamides syn-
thesized were oxidized with NaOCl in the presence of a
catalytic amount of 4-acetamido-2,2,6,6-tetramethylpiperi-
Fig. 1. Scheme indicating the synthetic sequence for these AX com-
pounds.

Phospholipase A2 Inhibition and Pain
469
TABLE 1
List of compounds including physical characteristics, in vitro XI(50) values for groups IVA and VIA PLA2
N.D., 25% inhibition or less at 0.091 mole fraction; L.D., between 25 and 50% inhibition at 0.091 mole fraction.
Compound
AX006
Structure
Mol. Wt.
355.52
CLogP
6.6
Group IVA XI(50) (mole fraction)
0.024 Ϯ 0.015
Group VIA XI(50)
N.D.
AX010
AX048
AX057
369.54
383.57
411.62
7.1
7.6
8.3
N.D.
L.D.
0.022 Ϯ 0.009
0.031 Ϯ 0.017
0.027 Ϯ 0.009
0.026 Ϯ 0.014
dine-1-yloxy free radical to produce compounds AX048 and AX048, and AX057 on pure group IVA PLA2 and group VIA
AX057. PLA2 were examined, and the results are presented in Table
Characterization of PLA2 Inhibitory Activity: Enzy- 1 as XI(50). The XI(50) is the mole fraction of inhibitor in the
matic Assay. The inhibitory effects of AX006, AX010, total substrate interface required to inhibit the enzyme by
50%. The reason that XI(50) is used instead of the more
common IC50 or KI is that PLA2 is active only on phospho-
lipid surfaces such as cell membranes, phospholipid vesicles,
or phospholipid micelles, where its substrate phospholipids
reside. Almost all inhibitors of PLA2s partition at least to
some degree into the phospholipid surface because they usu-
ally have a hydrophobic portion that complements the hydro-
phobic active site of the PLA2. When these inhibitors parti-
tion into the surface, an important physical effect called
surface dilution comes into play. In this case, the affinity of
the PLA2 for the inhibitor depends not on the three-dimen-
sional (bulk) concentration of the inhibitor in molar units but
on the two-dimensional (surface) concentration of the inhib-
itor in mole fraction units. As indicated (Figs. 2 and 3; Table
1), AX048 and AX057 were potent against group IVA PLA2
and group VIA PLA2, AX006 was potent against group IVA
PLA2 alone, and AX010 was less effective against both.
Characterization of COX Inhibitory Activity. Incuba-
tion with indomethacin produced a near-complete inhibition
Fig. 2. In vitro dose-response inhibition curves of AX006 (circles), AX010
(squares), AX048 (up triangles), and AX057 (down triangles) for group
IVA cPLA2. Curves represent a fit to a logarithmic function.
of the COX activity in the assay. In contrast, incubation with
the AX compounds at concentrations that had significant
effects upon PLA2 (50 ␮M) had no inhibitory effects upon
COX activity (Fig. 4).
Fig. 4. Effects of agents on in vitro cyclooxygenase activity expressed as
percent inhibition. Figure presents the mean Ϯ S.D. for drug-treated
samples versus control. As indicated, indomethacin (Indo, 50 ␮M) but not
AX006 (50 ␮M), AX010 (50 ␮M), AX048 (50 ␮M), or AX057 (50 ␮M) served
to inhibit cyclooxygenase activity at the doses employed.
Fig. 3. In vitro dose-response inhibition curves of AX010 (squares),
AX048 (up triangles), and AX057 (down triangles) for group iVI iPLA2.
Curves represent a fit to a logarithmic function.

470
Yaksh et al.
Fig. 5. Effects of AX006, AX010, AX048, and AX057 (3 mg/kg i.p.) on thermal hyperalgesia evoked by unilateral hind paw injection of carrageenan.
Drug or vehicle was delivered at 30 min prior to intraplantar injection of carrageenan, and thermal escape latency was measured immediately before
and at intervals afterward up to 180 min. Each set of graphs shows the mean Ϯ S.E.M. of the response latency (seconds) over time for the injured (Inj)
and uninjured (Uninj) paw for drug- and vehicle-treated animals. As indicated in the legend for each graph, in control-treated groups, the carrageenan
paw displayed a significant decline in latency from baseline (one-way ANOVA). This decline was prevented only by AX048. The histogram inset
displays the mean group cumulative difference in response latencies between uninjured and injured paws over the test interval (90–180 min). As
indicated, this measure of hyperalgesia was significantly reduced only by AX048 (unpaired Student’s t test).
In Vivo Behavioral Studies
the thermal escape latency of the uninjured paw in either the
vehicle- or drug-treated animal; e.g., the agent was behaving
functionally as an antihyperalgesic agent. Comparison of the
mean group difference between response latencies of unin-
jured and injured paws revealed a significant reduction in
the AX048-treated group as compared with the vehicle-
treated group.
Dose Dependence. The effects of i.p. AX048 were ob-
served to be dose-dependent over the range of 0.2 to 3 mg/kg
(slope; P Ͻ 0.0004) (Fig. 6). The ED50 was defined as the dose
that reduced the hyperalgesia observed in a vehicle-treated
animal by 50%. On this basis, the estimated i.p. ED50 value
for i.p. AX048 was 1.2 mg/kg (95% confidence interval,
Ϫ0.5572 to 0.7713).
Intraperitoneal Delivery and Carrageenan-Induced
Thermal Hyperalgesia: Control. Prior to induction of hy-
peralgesia, baseline thermal escape latencies were on the
order of 10 to 12 s in all groups. Intraplantar injection of
carrageenan induced inflammation of the injected hind paw
as well as a corresponding thermal hyperalgesia that was
detectable after 60 min lasting throughout the study. The
thermal escape latency in animals treated with i.p. or i.t.
vehicle was significantly reduced to approximately 3 to 5 s
within 90 to 120 min (Figs. 5 and 6).
Intraperitoneal Delivery. Pretreatment (30 min) with 3
mg/kg (i.p.) of the four agents prior to the carrageenan injec-
tion revealed that AX048, but not AX006, AX010, or AX057,
reduced the thermal hyperalgesia otherwise observed in the
Time Course of Action. To determine the time course of
inflamed paw (Fig. 5). Importantly, there was no change in the drug action, i.p. delivery of AX048 (3 mg/kg) was under-

Phospholipase A2 Inhibition and Pain
471
cant unilateral thermal hyperalgesia as compared with the
uninjected paw (Fig. 8).
Drug Effect. Pretreatment with 30 ␮g/10 ␮l of the four
agents 15 min prior to the delivery of carrageenan revealed
that AX048, but not AX006, AX010, or AX057, attenuated the
thermal hyperalgesia (Fig. 8). Again, after i.t. delivery, there
was no change in the thermal escape latency of the uninjured
paw in either the vehicle- or drug-treated animal. Compari-
son of the mean group difference between response latencies
of uninjured and injured paws also revealed a significant
reduction in the AX048-treated group in comparison with the
vehicle-treated group.
Fig. 6. Dose-response curve for the antihyperalgesic effects of i.p. AX048
on thermal hyperalgesia evoked by unilateral hind paw injection of car-
rageenan. Each point presents the mean and S.E.M. (n ϭ 5) of the
summed difference in response latencies between injured and uninjured
paw (ء, slope, P Ͻ 0.0004). The horizontal solid and dashed line repre-
sents the mean Ϯ S.E.M. of the vehicle-treated control animals. Studies
were carried out as described in Fig. 4. Graph presents the mean Ϯ
S.E.M. of the group cumulative difference in response latencies between
uninjured and injured paws over the test interval (90–180 min) as a
function of dose. The horizontal solid and dashed line presents the
mean Ϯ S.E.M. of the thermal hyperalgesia observed in vehicle-treated
rats after carrageenan. The ED50 dose of AX048 (50% reduction in the
thermal escape latency) was 1.2 mg/kg.
Intrathecal Substance P-Induced Thermal Hyperalgesia
Control. Baseline thermal escape latencies were on the
order of 10 to 12 s. In systemic vehicle-treated animals, the
i.t. injection of SP (20 nmol/10 ␮l) evoked a significant reduc-
tion in thermal escape latency as early as 15 min after
injection, which persisted through the 45-min test interval,
returning to baseline by 60 min (Fig. 9).
Drug Effect. Pretreatment with 3 mg/kg (i.p.) of the four
agents 30 min prior to the i.t. delivery of SP revealed that
AX048, but not AX006, AX010, or AX057, completely pre-
vented the spinally evoked thermal hyperalgesia (Fig. 9). As
in the carrageenan study, there was no evidence that AX048
increased the post-treatment latency to values greater than
baseline; e.g., the agent was behaving functionally as an
antihyperalgesic agent.
Side Effect Profile. After delivery of the highest systemic
dose (3 mg/kg) or i.t. dose (20 ␮g) of any of the compounds,
there were no changes in any assessed reflex end points
including eye blink, pinnae, placing, or stepping. The ani-
mals showed no change in righting response, symmetric am-
bulation, or spontaneous activity.
Spinal Prostaglandin Release
Control. Overall baseline dialysate concentrations after
the initial washout and prior to drug treatment were deter-
mined to be 555 Ϯ 75 pg/100 ␮l perfusate. Intrathecal injec-
tion of SP (20 ␮g) but not vehicle (saline, not shown) resulted
in a statistically significant increase in PGE2 concentrations
in spinal dialysate as compared with the vehicle-treated con-
trol (Fig. 10).
Drug Effect. Pretreatment with the four agents 15 min
prior to the delivery of i.t. SP (20 ␮g/10 ␮l) revealed that the
evoked release of PGE2 was reduced only in the AX048-
treated group. Thus, of the four agents only AX048 exerted a
significant inhibitory effect upon PGE 2 synthesis and release
(Fig. 10).
Effects of CB1 Inhibition. To determine whether the
effects of the active agent AX048 might be acting directly or
indirectly through a central cannabinoid CB-1 receptor, rats
were pretreated with i.t. vehicle or i.t. SR141716, a CB1
receptor antagonist, followed after 15 min by i.t. AX048 (30
␮g) or i.t. anandamide (100 ␮g). Intrathecal SR141716 had
no effect when delivered alone (data not shown). As shown in
Table 2, in vehicle-pretreated animals, i.t. anandamide re-
sulted in a significant increase in the thermal escape latency
of the uninjured paw and that of the injured paw. Both effects
were prevented by pretreatment with i.t. SR141716. Intra-
Fig. 7. Effects of pretreatment interval on antihyperalgesic effects of
AX048 (3 mg/kg i.p.) on carrageenan-evoked thermal hyperalgesia. Drug
was delivered at 15, 30, 180, or 360 min prior to the delivery of intraplan-
tar carrageenan, and thermal escape was measured immediately before
carrageenan and at intervals afterward up to 3 h. Data are expressed as
the cumulative latency difference between injured and uninjured paws.
Maximum effects were observed at 30 min and persisted for 3 h. One-way
ANOVA (P ϭ 0.0006) followed by post hoc Bonferroni’s multiple compar-
ison test (n ϭ 4–12 / treatment group). ءء, P Ͻ 0.05 as compared with
control.
taken at Ϫ15, Ϫ30, and Ϫ180 min (Fig. 7). As indicated, peak
effects were noted at 30 min, and minimal effects were ob-
served at 15 min. The effects persisted through for 180 min
but were no different from the control by 360 min.
Intrathecal Delivery and Carrageenan-Induced Thermal
Hyperalgesia
Control. In animals receiving i.t. injections of vehicle, the thecal AX048 significantly reversed the respective hyperal-
intraplantar injection of carrageenan resulted in a signifi- gesia but had no effect upon the thermal escape latency of the

472
Yaksh et al.
Fig. 8. Effects of AX006, AX010, AX048, and AX057 (i.t. 30 ␮g/10 ␮l) on thermal hyperalgesia evoked by unilateral hind paw injection of carrageenan.
Drug or vehicle was delivered at 15 min prior to intraplantar injection of carrageenan and thermal escape was measured immediately before and at
intervals afterward up to 180 min. Each set of graphs shows the mean Ϯ S.E.M. of the response latency (seconds) over time for the injured (Inj) and
uninjured (Uninj) paw for drug- and vehicle-treated animals. As indicated in the legend for each graph, in control-treated groups, the carrageenan paw
displayed a decline in latency from baseline (one-way ANOVA). This decline was prevented only by AX048. The histogram inset displays the mean
group cumulative difference in response latencies between uninjured and injured paw over the test interval (90–180 min). As indicated, this measure
of hyperalgesia was significantly reduced only by AX048 (unpaired Student’s t test).
uninjured paw. The antihyperalgesic effects of i.t. AX048 and group VIA iPLA2 preferring. The profile of activity ob-
were not altered by i.t. SR141716. These observations sug-
gest that i.t. anandamide, but not i.t. AX048, were interact-
ing with a spinal CB1 receptor to alter thermal escape la-
tency.
served here suggests the importance of both group IVA
cPLA2 and group VIA iPLA2. We showed that i.t. delivery of
methyl arachidonyl fluorophosphonate and arachidonyl tri-
fluoromethylketone (AACOCF3), mixed inhibitors of group
IVA cPLA2 and group VI iPLA2, produced a dose-dependent
inhibition of hyperalgesia induced by intraplantar carra-
geenan as well as formalin-induced flinching. Moreover, i.t.
injection of AACOCF3 at antihyperalgesic doses decreased
PGE2 release into spinal dialysate evoked by i.t. NMDA
(Lucas et al., 2005). In contrast, in those studies, an irrevers-
ible group VIA iPLA2 inhibitor (bromoenol lactone), given i.t.,
failed to show any antihyperalgesic effects at doses that did
not produce motor dysfunction and, at a higher dose, failed to
block evoked spinal PGE2 release. Yeo et al. (2004) reported
that i.c.v. injection of AACOCF3 or bromoenol lactone pro-
duced antihyperalgesia as measured using facial carra-
geenan in mice (Yeo et al., 2004). Burke et al. (2001) reported
that BMS-229724, a group IVA cPLA2 inhibitor, was orally
active in inhibiting edema and neutrophil infiltration. The
present data thus continue to leave the issue open regarding
Discussion
AX048, but not the three structurally related analogs
AX006, AX010, and AX057, exerted a significant effect upon
centrally (i.t. SP) and peripherally (intraplantar carra-
geenan) initiated hyperalgesia. Because the effective i.t. dose
was 100 times less than required after systemic delivery, we
conclude that the i.t. effect represented a central action. In
addition, systemic AX048 blocked the hyperalgesia evoked by
i.t. SP in the absence of any peripheral injury. This suggests
that the activity of the systemically delivered compound was
mediated by a central action. Parallel in vitro characteriza-
tion of the selectivity of these agents in reversibly blocking
group IVA cPLA2 and group VIA iPLA2 revealed that AX010
had at best a weak effect, AX006 was group IVA PLA2 pre-
ferring, whereas AX048 and AX057 were group IVA cPLA2

Phospholipase A2 Inhibition and Pain
473
Fig. 9. Effects of AX006, AX010,
AX048, and AX057 (3 mg/kg i.p.)
on i.t. SP-evoked thermal hyperal-
gesia. Drug or vehicle was deliv-
ered at 30 prior to the i.t. delivery
of substance P (i.t. SP, 30 nmol),
and thermal escape was measured
immediately before i.t. SP and at
intervals afterward up to 60 min.
Data are expressed as the re-
sponse latency (seconds) over
time. As indicated in the legend
for each graph, one-way ANOVA
showed significant thermal hyper-
algesia reversal from vehicle for
AX048, but not the other agents.
the relative contribution of group IVA cPLA2 and group VIA the cannabinoid-1 receptor is likely. The effect upon the
iPLA2.
centrally mediated hyperalgesia excludes a peripheral can-
Role of Spinal PLA2 Isoforms in Cascade. Western nabinoid-2 receptor action. Moreover, after i.t. delivery,
blotting and reverse transcription-polymerase chain reaction anandamide elevated the thermal escape latency of the nor-
have shown that group IVA cPLA2, group VIA iPLA2, and mal paw, an effect not mimicked by the AX048. Finally,
secretory PLA2 (groups IIA and V) are constitutively ex- SR141716A, a potent CB1 antagonist (Shire et al., 1999),
pressed in the spinal cord (Sapirstein and Bonventre, 2000; given i.t. at a dose that reversed the i.t. effect of anandamide,
Lucas et al., 2005; Svensson et al., 2005b). The role of these failed to alter the effects of AX048. These data suggest an
respective isoforms has been difficult to assess given the lack effect of spinal AX048 that is independent of an action upon
of potent and selective inhibitors. Based on our earlier work either endogenous cannabinoid release or upon the receptor
noted above, we have had a particular interest in group IVA itself. These experiments provide supportive evidence consis-
cPLA2. In the present work, AX048 displayed a dose-depen- tent with the assertion that AX048 was indeed acting though
dent suppression of both centrally and peripherally evoked a PLA2 enzyme. We recognize that these are complex sys-
thermal hyperalgesia. Importantly, the comparable antihy- tems, and other potential targets might be considered in
peralgesic action of AX048 after i.t. delivery with 20 ␮g future studies and include a variety of upstream enzymes
versus the dose of 3 mg/kg, given i.p., emphasizes an impor- (such as mitogen-activated protein kinase) (Svensson et al.,
tant spinal action. The effects of systemic delivery showed an 2005a) as well as downstream effects (such as inhibition of
onset of approximately 30 min and a duration of action that prostaglandin synthases or receptors) (Guay et al., 2004;
exceeded 180 min. Importantly, this dosing was shown to Reinold et al., 2005).
have a significant effect upon i.t. SP-evoked spinal PGE2
Factors Governing Central Bioavailability and Ac-
release, a downstream biomarker believed to be essentially tivity. These compounds are constructed based on a 2-oxo-
dependent upon PLA2 activity (Svensson and Yaksh, 2002). amide with a hydrocarbon tail and four-carbon tether. An
Although the primary target of these molecules examined important consideration in the functionality of these agents
in the present study is PLA2, we note that other possibilities is their high cLog P values, in the range of 6 to 8. It is widely
may also be relevant including a direct effect upon cyclooxy- considered that agents with log P values greater than 5 may
genase or the endocannabinoid system, both of which may not be “druggable” (Lipinski et al., 2001). It is important to
lead to a change in pain behavior in a hyperalgesic state (see note that in the present systems, the target of drug action is
Rice et al., 2002; Svensson and Yaksh, 2002). The present within the cytosol. This requires that the molecule have a
studies, however, showed no effects at the highest concentra- lipophilicity that allows it to readily cross the cell membrane
tions on either COX-1 or COX-2 activity. Recent work sug- to interact with PLA2. In the present work, we found that
gested that agents interacting with the COX cascade may three of the molecules, AX048, AX057, and AX006, possessed
exert effects though an endocannabinoid pathway (Seidel et appropriate enzyme inhibitory activity in a cell-free in vitro
al., 2003). We, however, do not think that an effect through assay. Yet, only AX048 was observed to show in vivo activity.

474
Yaksh et al.
of ethyl ester angiotensin-converting enzyme inhibitors, for
example enalapril, exhibit greater oral activity than would be
expected purely from the increased lipophilicity due to the
conversion to ethyl ester. Furthermore, there is evidence that
this ethyl ester is actively absorbed by a carrier mechanism
(Swan and Tukker, 1997). These data could explain why only
ethyl ester (AX048) of the four agents is active in vivo,
whereas the other three agents are inactive at a dose of 3
mg/kg. Nevertheless, the observation that AX048 was able to
produce an antihyperalgesic effect indicates that this mole-
cule has properties that allow penetration of cellular mem-
branes. Further work will be required to define the critical
physical chemistry that defines the ability of AX048 to gain
access to the central nervous system and inhibit intracellular
PLA2.
Multiple Effects of PLA2 Inhibition. In the face of pe-
ripheral inflammation and tissue injury, an exaggerated pro-
cessing of nociceptive stimuli ensues. This facilitation re-
flects in part an afferent-evoked cascade leading to enhanced
nociceptive processing at the spinal level. An important com-
ponent of this cascade is associated with the actions of spi-
nally released prostanoids. Support for this thesis arises
from the observation that the spinal delivery of prostaglan-
dins will induce hyperalgesia and that these lipidic acids are
released into the spinal extracellular space after tissue injury
(see references in Introduction). In addition, i.t. COX inhib-
itors reduce prostaglandins release and the facilitated state
induced by peripheral injury or by the direct activation of
these circuits with i.t. SP and/or glutamate (see Svensson
and Yaksh, 2002). This cascade suggests the relevance of
pursuing the upstream PLA2 linkages which precede those
mediated by cyclooxygenase. We note, however, that there is
substantial evidence that other products of PLA2 activity are
important in nociceptive processing. Arachidonic acid gener-
ated by PLA2 isoforms can directly augment NMDA iono-
phore function (Richards et al., 2003). The NMDA receptor is
believed to play an important role in pre- and postsynaptic
facilitation at the spinal level (L’Hirondel et al., 1999; Rich-
ards et al., 2003). Arachidonic acid formed by the action of
PLA2s also provides the essential substrate necessary for the
cyclooxygenase-independent synthesis of isoprostanes. Spi-
nal isoprostanes initiate facilitated transmitter release and
neuronal discharge, and their spinal delivery will lead to
hyperalgesia (Evans et al., 2000). Platelet-activating factor,
an alkyl-phospholipid, arises from the membrane lipid hy-
drolysis by PLA2. Platelet-activating factor produces a prom-
inent allodynia after spinal delivery (Morita et al., 2004).
This lipid mediator is present in the spinal cord and is re-
leased from stimulated microglia cells (Jaranowska et al.,
1995). PLA2 activity forms lysophosphates. These products
have been implicated in facilitated states of pain processing
(Inoue et al., 2004; Seung Lee et al., 2005). In short, we
hypothesize that a more pronounced effect on spinal nocicep-
tive processing might arise by blocking these linkages up-
stream to COX,. Finally, the present studies showing the
development of systemically bioavailable PLA2-selective
agents may be relevant to therapeutic targets other than
pain. A variety of neuron inflammatory processes may also be
mediated through their activation of neuraxial PLA2 iso-
Fig. 10. Unanesthetized rats prepared with spinal dialysis catheters
received i.p. injections of vehicle or AX006, AX010, AX048, and AX057
(3 mg/kg i.p.) followed 20 min later by i.t. injections of substance P (i.t.
SP, 20 nmol). Top, time course of PGE2 release determined in sequen-
tial 15-min samples out through 45 min following i.t. SP in animals
pretreated with i.p. vehicle or i.p. AX048 (3 mg/kg). Intrathecal SP
evoked a time-dependent increase in release following i.p. vehicle but
not following i.p. AX048 (ء, P Ͻ 0.05). Bottom, area under the time
effect curve for PGE2 release from 0 to 45 min in rats receiving vehicle,
AX006, AX010, AX048, or AXO57. As indicated, after i.p. AX006,
AX010, or AX057, i.t. SP evoked a significant increase as compared
with vehicle only [Kruskall Wallace, P Ͻ 0.008; ء, P Ͻ 0.05; ءء, P Ͻ
0.01, Dunns multiple comparison versus vehicle (VEH)]. In contrast,
following i.p. AXO48, there was no difference between release as
compared with i.p. vehicle alone (P Ͼ 0.05).
TABLE 2
Effects of i.t. SR141716A on the effects of i.t. anandamide and AX048
at 2 h postcarrageenan on thermal escape latency of the uninjured and
injured paws
n ϭ 4 to 6 rats/treatment group.
Uninjured
Paw
Treatment
Injured Paw
i.t. Vehicle ϩ i.t. vehicle
10.4 Ϯ 1.5
3.3 Ϯ 1.1*
i.t. Vehicle ϩ i.t. anandamide (100 ␮g)
18 Ϯ 3.2** 15.2 Ϯ 4.2**
i.t. SR141716A (20 ␮g) ϩ i.t. anandamide 10.8 Ϯ 2.4
(100 ␮g)
i.t. vehicle ϩ i.t. AX048 (30 ␮g)
i.t. SR141716A ϩ i.t. AX048 (30 ␮g)
3.1 Ϯ 1.3*
10.3 Ϯ 2.0
11.7 Ϯ 2.9
7.8 Ϯ 1.8**
7.1 Ϯ 1.9**
* P Ͻ 0.05 vs. uninjured paw; two-tailed paired Student’s t test.
** P Ͻ 0.05 vs. respective vehicle-treated paw; two-tailed unpaired Student’s t
test.
We suspect that the dissociation between in vitro and in vivo
activity that these agents display may well depend on the
complex issue of distribution that these molecules face. At
present, we believe that AX048 may be acting as a prodrug.
The most common prodrug moiety in marketed drugs is the
esterification of an acid group with a simple alkyl alcohol. A
variety of ester prodrugs, in particular ethyl esters, are sum-
marized in a recent review (Beaumont et al., 2003). A number forms.

Phospholipase A2 Inhibition and Pain
475
Reinold H, Ahmadi S, Depner UB, Layh B, Heindl C, Hamza M, Pahl A, Brune K,
Narumiya S, Muller U, and Zeilhofer HU (2005) Spinal inflammatory hyperalgesia
is mediated by prostaglandin E receptors of the EP2 subtype. J Clin Investig
115:673–679.
Rice AS, Farquhar-Smith WP, and Nagy I (2002) Endocannabinoids and pain: spinal
and peripheral analgesia in inflammation and neuropathy. Prostaglandins Leukot
Essent Fatty Acids 66:243–256.
Richards DA, Bliss TV, and Richards CD (2003) Differential modulation of NMDA-
induced calcium transients by arachidonic acid and nitric oxide in cultured hip-
pocampal neurons. Eur J Neurosci 17:2323–2328.
Samad TA, Moore KA, Sapirstein A, Billet S, Allchorne A, Poole S, Bonventre JV,
and Woolf CJ (2001) Interleukin-1beta-mediated induction of Cox-2 in the CNS
contributes to inflammatory pain hypersensitivity. Nature (Lond) 410:471–475.
Sapirstein A and Bonventre JV (2000) Specific physiological roles of cytosolic phos-
pholipase A(2) as defined by gene knockouts. Biochim Biophys Acta 1488:139–148.
Seidel K, Hamza M, Ates M, and Guhring H (2003) Flurbiprofen inhibits capsaicin
induced calcitonin gene related peptide release from rat spinal cord via an endo-
cannabinoid dependent mechanism. Neurosci Lett 338:99–102.
Seung Lee W, Hong MP, Hoon Kim T, Kyoo Shin Y, Soo Lee C, Park M, and Song JH
(2005) Effects of lysophosphatidic acid on sodium currents in rat dorsal root
ganglion neurons. Brain Res 1035:100–104.
Shire D, Calandra B, Bouaboula M, Barth F, Rinaldi-Carmona M, Casellas P, and
Ferrara P (1999) Cannabinoid receptor interactions with the antagonists SR
141716A and SR 144528. Life Sci 65:627–635.
Southall MD, Michael RL, and Vasko MR (1998) Intrathecal NSAIDS attenuate
inflammation-induced neuropeptide release from rat spinal cord slices. Pain 78:
39–48.
Svensson CI, Fitzsimmons B, Azizi S, Powell HC, Hua XY, and Yaksh TL (2005a)
Spinal p38beta isoform mediates tissue injury-induced hyperalgesia and spinal
sensitization. J Neurochem 92:1508–1520.
Svensson CI, Lucas KK, Hua XY, Powell HC, Dennis EA, and Yaksh TL (2005b)
Spinal phospholipase A(2) in inflammatory hyperalgesia: role of the small, secre-
tory phospholipase A(2). Neuroscience 133:543–553.
Svensson CI and Yaksh TL (2002) The spinal phospholipase-cyclooxygenase-
prostanoid cascade in nociceptive processing. Annu Rev Pharmacol Toxicol 42:
553–555.
Swan PW and Tukker JJ (1997) Molecular determinants of recognition for the
intestinal peptide carrier. J Pharm Sci 86:596–602.
Taiwo YO and Levine JD (1988) Prostaglandins inhibit endogenous pain control
mechanisms by blocking transmission at spinal noradrenergic synapses. J Neuro-
sci 8:1346–1349.
Turnbach ME and Randich A (2001) The effect of spinal and systemic administration
of indomethacin on zymosan-induced edema, mechanical hyperalgesia and ther-
mal hyperalgesia. J Pain 2:25–35.
Acknowledgments
We thank Alan Moore for carrying out the prostaglandin E2 as-
says.
References
Beaumont K, Webster R, Gardner I, and Dack K (2003) Design of ester prodrugs to
enhance oral absorption of poorly permeable compounds: challenges to the discov-
ery scientist. Curr Drug Metab 4:461–485.
Burke JR, Davern LB, Stanley PL, Gregor KR, Banville J, Remillard R, Russell JW,
Brassil PJ, Witmer MR, Johnson G, et al. (2001) BMS-229724 is a tight-binding
inhibitor of cytosolic phospholipase A2 that acts at the lipid/water interface and
possesses anti-inflammatory activity in skin inflammation models. J Pharmacol
Exp Ther 298:376–385.
Dirig DM, Salami A, Rathbun ML, Ozaki GT, and Yaksh TL (1997) Characterization
of variables defining hindpaw withdrawal latency evoked by radiant thermal
stimuli. J Neurosci Methods 76:183–191.
Dole VP (1956) A relation between non-esterified fatty acids in plasma and the
metabolism of glucose. J Clin Investig 35:150–154.
Ebersberger A, Grubb BD, Willingale HL, Gardiner NJ, Nebe J, and Schaible HG
(1999) The intraspinal release of prostaglandin E2 in a model of acute arthritis is
accompanied by an up-regulation of cyclo-oxygenase-2 in the spinal cord. Neuro-
science 93:775–781.
Evans AR, Junger H, Southall MD, Nicol GD, Sorkin LS, Broome JT, Bailey TW, and
Vasko MR (2000) Isoprostanes, novel eicosanoids that produce nociception and
sensitize rat sensory neurons. J Pharmacol Exp Ther 293:912–920.
Ferreira SH (1972) Prostaglandins, aspirin-like drugs and analgesia. Nat New Biol
240:200–203.
Guay J, Bateman K, Gordon R, Mancini J, and Riendeau D (2004) Carrageenan-
induced paw edema in rat elicits a predominant prostaglandin E2 (PGE2) response
in the central nervous system associated with the induction of microsomal PGE2
synthase-1. J Biol Chem 279:24866–24872.
Inoue M, Rashid MH, Fujita R, Contos JJ, Chun J, and Ueda H (2004) Initiation of
neuropathic pain requires lysophosphatidic acid receptor signaling. Nat Med 10:
712–718.
Jaranowska A, Bussolino F, Sogos V, Arese M, Lauro GM, and Gremo F (1995)
Platelet-activating factor production by human fetal microglia. Effect of lipopoly-
saccharides and tumor necrosis factor-alpha. Mol Chem Neuropathol 24:95–106.
Kokotos G, Kotsovolou S, Six DA, Constantinou-Kokotou V, Beltzner CC, and Dennis
EA (2002) Novel 2-oxoamide inhibitors of human group IVA phospholipase A(2).
J Med Chem 45:2891–2893.
Kokotos G, Six DA, Loukas V, Smith T, Constantinou-Kokotou V, Hadjipavlou-Litina
D, Kotsovolou S, Chiou A, Beltzner CC, and Dennis EA (2004) Inhibition of group
IVA cytosolic phospholipase A2 by novel 2-oxoamides in vitro, in cells and in vivo.
J Med Chem 47:3615–3628.
Kulmacz RJ and Lands WE (1983) Requirements for hydroperoxide by the cycloox-
ygenase and peroxidase activities of prostaglandin H synthase. Prostaglandins
25:531–540.
L’Hirondel M, Cheramy A, Artaud F, Godeheu G, and Glowinski J (1999) Contribu-
tion of endogenously formed arachidonic acid in the presynaptic facilitatory effects
of NMDA and carbachol on dopamine release in the mouse striatum. Eur J Neu-
rosci 11:1292–1300.
Lipinski CA, Lombardo F, Dominy BW, and Feeney PJ (2001) Experimental and
computational approaches to estimate solubility and permeability in drug discov-
ery and development settings. Adv Drug Deliv Rev 46:3–26.
Lucas KK, Svensson CI, Hua XY, Yaksh TL, and Dennis EA (2005) Spinal phospho-
lipase A(2) in inflammatory hyperalgesia: role of Group IVA cPLA(2). Br J Phar-
macol 144:940–952.
Malmberg AB and Yaksh TL (1992) Hyperalgesia mediated by spinal glutamate or
substance P receptor blocked by spinal cyclooxygenase inhibition. Science (Wash
DC) 257:1276–1279.
Malmberg AB and Yaksh TL (1995) Cyclooxygenase inhibition and the spinal release
of prostaglandin E2 and amino acids evoked by paw formalin injection: a micro-
dialysis study in unanesthetized rats. J Neurosci 15:2768–2776.
Morita K, Morioka N, Abdin J, Kitayama S, Nakata Y, and Dohi T (2004) Develop-
ment of tactile allodynia and thermal hyperalgesia by intrathecally administered
platelet-activating factor in mice. Pain 111:351–359.
Vane JR (1971) Inhibition of prostaglandin synthesis as a mechanism of action for
aspirin-like drugs. Nat New Biol 231:232–235.
Yaksh TL (1982) Central and peripheral mechanisms for the antianalgesic action of
acetylsalicylic acid, in Acetylsalicylic Acid: New Uses for an Old Drug (Barnet JM,
Hirsh J, and Mustard JF eds) pp 137–152, Raven Press, New York.
Yaksh TL, Dirig DM, Conway CM, Svensson C, Luo D, and Isakson PC (2001) The
acute antihyperalgesic action of NSAIDs and release of spinal PGE2 is mediated by
the inhibition of constitutive spinal COX-2 but not COX-1. J Neurosci 21:5847–
5853.
Yaksh TL and Rudy TA (1976) Chronic catheterization of the spinal sub-arachnoid
space. Physiol Behav 17:1031–1036.
Yamamoto T and Nozaki-Taguchi N (1996) Analysis of the effects of cyclooxygenase
(COX)-1 and COX-2 in spinal nociceptive transmission using indomethacin, a
non-selective COX inhibitor and NS-398, a COX-2 selective inhibitor. Brain Res
739:104–110.
Yeo JF, Ong WY, Ling SF, and Farooqui AA (2004) Intracerebroventricular injection
of phospholipases A2 inhibitors modulates allodynia after facial carrageenan in-
jection in mice. Pain 112:148–155.
Address correspondence to: Dr. Tony L. Yaksh, Department of Anesthesi-
ology University of California, San Diego, 9500 Gilman Drive, La Jolla, CA
92093-0818. E-mail: tyaksh@ucsd.edu