Research Paper Mediators of Inflammation, 9, 25–30 (2000)

NG-nitro-L-arginine methyl ester (L-NAME) has been used extensively as a paradigmatic inhibitor of NO synthase and has been shown to cause antinociception in several experimental models. We describe here how L-NAME produced a dose-dependent antinociceptive effect when injected intraperitoneally in the mouse after acetic acid induced writhings, or intraplantarly in the rat paw pressure hyperalgesia induced by carrageenin or prostaglandin E2. In contrast another NO synthase inhibitor, NG-monomethyl-L-arginine (L-NMMA), had no significant effect per se but inhibited L-NAME systemic induced antinociception in mice and local induced antinociception in the rat paw hyperalgesia test. D-NAME had no antinociceptive effect upon carrageenin-induced hyperalgesia. Pretreatment of the paws with two inhibitors of guanylate cyclase, methylene blue (MB) and 1H-:[1,2,4]-oxadiazolo-:[4,3-a] quinoxalin-1-one (ODQ) abolished the antinociceptive effect of L-NAME. L-Arginine and the cGMP phosphodiesterase inhibitor, MY 5445 significantly enhanced the L-NAME antinociceptive effect. The central antinociceptive effect of L-NAME was blocked by co-administration of L-NMMA, ODQ and MB. The present series of experiments shows that L-NAME, but not L-NMMA, has an antinociceptive effect. It can be suggested that L-NAME causes the antinociceptive effect by stimulation of the arginine/ NO/ cGMP pathway, since the antinociceptive effect of L-NAME can be antagonized by L-NMMA and abolished by the guanylate cyclase inhibitors (MB and ODQ). In addition, the NO synthase substrate, L-arginine and the cGMP phosphodiesterase inhibitor, MY5445 were seen to potentiate the effects of L-NAME. Thus, L-NAME used alone, has limitations as a specific inhibitor of the arginine-NO-cGMP pathway and may therefore be a poor pharmacological tool for use in characterising participation in pathophysiological processes.


Introduction
L-NAME (N G -nitro-L-arginine me thyl e ste r ), has be en show n to cause antinocic eption by spinal, supraspinal, loc al (intraplantar ) or systemic administration. 1 -8 As L-NAME is c onside red an spec ific nitric ox ide (NO) synthase inhibitor, 2 ,24 the se ex periments w ere taken to supp ort the hypothe sis that stimulation of the arginine /NO/c GMP pathw ay e nhance s nocice ption at various leve ls of the sensory syste m. On the othe r hand, there are several reports indicating that choline rgic or opioidergic stimulation of the arginine /NO/c GMP pathw ay c ause s ce ntral, spinal or peripheral analge sia, 9 -1 2 and some pe riphe ral analgesic s cause antinoc ice ption by stimulation of this pathw ay. 13 -1 6 Furthermore, the central analge sic, arginine , see ms to be associated w ith NO-c GMP stimulation. 17 -1 9 Thus there appe ars to be an appare nt contradic tion amongst the various ex pe riments made to asce rtain the role of the arginine /NO/c GMP pathw ay in nocice ption. A gre at de al of information has be en derive d w ith the use of L-NAME as a me thodological tool. How ever L-NAME can be se en to be either analge sic or hype ralgesic in the same te st 2 0 w hilst another NO synthase inhibitor N G -monomethyl-L-arginine (L-NMMA) doe s not show antinocic eption in the same test as L-NAME, i.e. the formalin te st in mice. 2 Re cently, L-NAME has be en reported to stimulate inducible NO synthase (iNOS) gene ex p re ssion. 2 1 Be c ause of these apparent contradic tory results, w e e valuate d the possibility that the analge sic effect of L-NAME w as due to stimulation of the arginine /NO/c GMP pathw ay. In the present study L-NAME w as initially assayed in tw o te sts in w hich nocic eption involves an inflammatory stimulus: the acetic acid induc ed w rithings in mic e and in the rat paw pre ssure hyperalgesia te st induc ed by intraplantar administration of c arragee nin. In these te sts L-NAME show ed an antinoc ice ptive effe ct w hich w as significantly inhibited by pre treatme nt of the animals w ith L-NMMA.
To furthe r inve stigate the antinoc ice ptive e ffec t of L-NAME and in order to avoid the oedema formation and fac ilitate succ essive injections of drugs, PGE 2 instead of c arragee nin w as used to induc e hyp eralge sia in the rat paw pre ssure test. In this te st the peripheral antinocic eptive synergism betw e en L-NAME and the NO synthase substrate, arginine, and the cGMP phosphodie ste rase inhibitor, as w ell the effect of tw o inhibitors of guanylate c GMP activation, me thyle ne blue (MB) and 1H-[1,2,4]-ox adiazolo-[4,3-a] quinox alin-1-one (ODQ) w as evaluate d. Finally, since L-NAME has be en show n to cause analgesia by intrac erebrove ntric ular administration, 2 w e te ste d L-NAME c o-administration w ith inhibitors of the arginine /NO/c GMP pathw ay, L-NMMA, MB and ODQ.

Animals
The ex perime nts w ere performed on male Wistar rats (150-180 g ) and albino Sw is s mice (22-30 g ). The animals w ere house d unde r natural light, w ith free acce ss to food and w ater. Intrace re broventric ular (i.c.v.) inje ctions in rats w e re made follow ing the me thod described by Côrre a and Grae ff. 22 When single doses of the various drugs w ere used, the y w ere base d on dose resp onse pilot ex p eriments. All ex pe rimental procedures c onformed to the IASP guide line s on the use of animals in pain re search. Rats w ere use d once only.

(a ) Writh ing te s t in m ice .
This test w as based on the frequenc y of abdominal contortions e voked by an intraperitone al inje ction of 10 ml/kg of 0.6% ac etic acid. 8 L-NAME, L-arginine , or vehic le w as injec te d 15 min before ac etic acid administration, and the number of w rithing e vents w ere counte d for 20 min afte r the noc iceptive challe nge. For antagonism studie s, mice w e re treate d as above, ex c ept that L-NMMA w as administe red 15 min be fore L-NAME.

(b ) Hin d pa w hy p e ra lge s ia te s t in ra ts .
Our modification of the Randall-Selitto rat paw pressure test w as use d to me asure hyperalge sia. 23 In the te st, a p re ssure of 20 mmHg is c ontinuously applie d to the hind paw of the rat until the animal presents a typic al free zing re ac tion (reaction time ). After me asurement of the basal reaction time (c ontrol), hyperalge sia w as induce d eithe r by an intraplantar injection of carrage enin (Cg, 100 m g ) or PGE 2 . The inte nsity of hyperalge sia w as quantified as the diffe rence in reaction time (de lta reaction time ) me asure d 3 h after administration of Cg, from the c ontrol reaction time assessed be fore injection of the hyp eralge sic stimulus. The term nocic eption is use d in this paper to de scribe the presenc e of an overt standard behaviour induc ed by the applic ation of a nox ious stimulus in a normal tissue or a non-nox ious stimulus in previously sensitised tissue. The te rm hype ralgesia is used w he n a non-nox ious stimulus causes nocic eption w hen applie d to a se nsitized tissue e ithe r by an inflammatory stimulus like c arragee nin or an hype ralge sic mediator like prostaglandin E2.

Statistics
All results are presente d as means (SEM of five rats or six to 12 mic e pe r group. Re sults are prese nted as me ans and standard e rrors of the means of groups of at least five animals in e ach group. Diffe re nce s betw e en re sponses w e re evaluate d by ANOVA, follow ed by the Bonferroni t-te st. Results w ith P< 0.05 w ere c onside re d significant.

Results
Blockade of L-NAME effect by L-NMMA in two tests of nociception induced by inflammatory stimuli Antinociceptive effect of intraplantar administration of L-NAME but not of D-NAME on carrageenin-induced hyperalgesia and the blockade of antinociception by local L-NMMA and methylene blue (MB) co-administration. L-NMMA (100 m g/paw), MB (500 m g/paw) or saline (S) were injected 30 min before carrageenin (Cg, 100 m g/paw). L-NAME or D-NAME (50-300 m g/paw) were also injected 30 min before Cg, and the intensity of hyperalgesia was measured 3 h after the hyperalgesic stimulus (see injections diagram). The symbols are the mean + SEM of five rats/ group. *indicates significant differences (P < 0.01) in comparison with the control (saline, O) or treatment with L-NAME, MB, D-NAME. by approx imate ly 40% and 73% at dose s of 30 and 90 mg /kg, respec tively. L-NMMA (10 mg /kg i.p.), significantly blocked the antinoc ic eptive action of L-NAME, but had no effect on its ow n (Fig. 1 ).

(b ) Ra t p a w pre s s u re te s t: hy pe ra lg e s ia indu ce d by c a rra g e e n in infla m m a tio n .
Intraplantar administration of L-NAME (50 and 300 m g ) produce d a signific ant inhibition of carrage enininduc ed hype ralgesia of up to 40% for the highest dose used (Fig. 2 ). L-NAME-induce d p eriphe ral hyperalgesia w as signific antly inhib ited by pre-treatment of the paw s w ith 100 m g of L-NMMA or 500 m g of MB. Neither MB, nor L-NMMA (up to 500 m g /paw ) had hyp eralgesic effects.
Rat paw pressure test: hyperalgesia induced by PGE 2

(a ) An ta go n is m L-NMMA, MB o r ODQ o f L-NAME-ind u c e d a n tin o c ice ptio n .
Pilot ex perime nts show ed that doses higher than 200 m g /paw w ere needed to give signific ant antinocic eption. Fig. 3 show s a signific ant antinocice ptive effect upon hyp eralgesia induc ed by PGE 2 of a dose of 300 m g of L-NAME. Pre treatme nt of the paw s w ith L-NMMA (LN), MB or ODQ prevented the antinocic eptive effe ct of L-NAME. Figure 4 show s that the assoc iation of L-arginine and L-NAME treatments c ause d a significant antinocic eption compared w ith single tre atments. The re w as no difference among controls groups. L-NMMA significantly inhibited the antinocic eptive effect of the association of L-arginine and L-NAME treatment. The association of the same doses of L-NAME and D-arginine did not cause antinocic eption as c ompare d w ith the single tre atment (data not show n ). Figure 5 show s that MY 5445 enhanc ed, in a doserelated manne r, the antinocic eptive e ffec t of L-NAME. A dose of 180 m g, w hich did not p roduc e antinocic eption in our ex pe riments, p roduc ed an effe ct similar to 300 m g w he n the paw s w e re pretreated w ith MY 5445 (compare w ith Fig. 3 ).

(c ) Blo c k a de o f th e intra c e re bro ve n tricu la r a n tin o c iceptive e ffe c t o f L-NAME by c o -tre a tm e n t w ith L-NMMA, MB o r ODQ.
L-NAME, w he n administe red i.c .v. at a dose of 300 m g per rat, produc ed pote nt antinocic eption in paw s rendere d hyperalgesic by PGE 2 (Fig. 6 ). This L-NAMEinduc ed anti-hype ralgesic e ffec t w as abolishe d by coi.c.v. administration of L-NMMA (300 m g ), MB (400 m g ) and ODQ (8 m g ). Neither L-NMMA, nor MB, nor ODQ had any effe cts on PGE 2 induced hyperalge sia.

Discussion
L-NAME has be en show n by se ve ral laboratorie s 1-8 to have a pe rip heral and central antinocice ptive action.

Mediators of Inflammation · Vol 9 · 2000
FIG. 4. Antinociceptive synergism between L-NAME and L-arginine. The measurements were made 3 h after the intraplantar injections of PGE 2 (100 ng/paw). L-NMMA, L-Arginine (L-Arg) and L-NAME were injected into the paw 1, 1.5 and 2 h after the PGE 2 challenge, respectively. The bars are the mean ± SEM of five rats/group. The asterisks mean significant differences (P < 0.05): (a) *in comparison with the control PGE 2 treated either with L-arg, L-NAME, or saline (S); (b) **in comparison with the group which received L-NMMA + L-arg + L-NAME.
FIG. 5. Potentiation of intraplantar injections of L-NAME by the cGMP phosphodiesterase inhibitor, MY 5445. The measurements were made 3 h after the intraplantar injections of PGE 2 (100 ng/paw). MY 5445 and (or) L-NAME were injected into the paw 1 and 2 h after PGE 2 , respectively. The bars are the mean ± SEM of five rats/group. The asterisk means significant differences (P < 0.05) compared with respective doses of L-NAME. There was no significant differences between the controls groups (C) treated with saline or MY 5445.
FIG. 6. Antinociceptive effect of intracerebroventricular administration of L-NAME (N G -nitro-L-arginine methyl ester) on PGE 2 -induced hyperalgesia, and its blockade by i.c.v. L-NMMA (LN, 300 m g,), MB (400 m g) or ODQ (8 m g). These drugs and vehicle were co-injected with L-NAME 2 h after PGE 2 (100 ng/paw). The intensity of hyperalgesia was measured 1 h after the i.c.v. injections. The bars are the mean, SEM of five rats/group. *,**mean significant differences (P < 0.01) in comparison with the saline-(Sal) or L-NAME-treated group, respectively. He re, w e confirm the ir obse rvations using tw o te sts of noc ice ption. In the ac etic acid w rithing te st, systemic administration (i.p .) of L-NAME but not L-NMMA caused an antinocic eptive e ffec t (Fig. 1 ). In addition, w e show ed that L-NMMA inhibite d the antinocic eptive ac tivity of L-NAME in this test as w e ll as in the rat paw p re ssure te st in w hich hype ralgesia w as induce d e ithe r by an inflammatory stimuli like Cg or by an hyperalgesic mediator, PGE 2 (Figs 2 -4 ). The D-is omer of NAME show e d no antinoc icep tive ac tivity (Fig. 2 ) in c arragee nin-induc ed hype ralgesia in the rat pressure te sts (Fig. 2 ). It has already be en show n that this is ome r had no antinoc ice ptive effe ct in other te sts . 2 Co-inje ction of L-NMMA i.c .v. also inhibited the ce ntral analge sic action of L-NAME (Fig. 6 ).
The observed antinocic eptive effe ct of L-NAME appare ntly supports the ide a that the arginine /NO/ cGMP pathw ay c ontributes to nocice ption induce d by inflammatory stimuli, partic ularly bec ause L-NAME is c onside red to be a sele ctive NO synthase inhibitor. 2,2 4 L-arginine -derive d inhibitors, how e ver, have bee n found to have bizarre pharmacologic al e ffec ts. For instance , it has be en de monstrated that L-NAME is a poor inhibitor of L-arginine transport, w hereas L-NMMA and L-NIO substantially inhibit this e ve nt. 2 5 Moreove r, L-NAME, but not L-NMMA, is a muscarinic antagonist. 26 On the othe r hand, L-NMMA, has be en show n to be have as a partial agonist, sinc e it antagonize s NO synthesis in some tissue s, but stimulate s NO synthesis in is olated arte rial rings. 27 Since, in our ex periments, L-NMMA in the doses used did not show any effe ct pe r se, but blocked the antinocic eptive effe ct of L-NAME, w e assume that it may be acting as a NO synthase inhibitor. On the othe r hand, L-NAME may be ac ting either as a substrate for or as an iNOS stimulator.
It is know n that during c arragee nin (but not PGE 2 ) induc ed hype ralgesia the arginine /NO/c GMP pathw ay is ac tivated. 28 The abse nce of activity of the arginine NO cGMP pathw ay in PGE 2 -induc ed hyperalge sia is illustrated here by the fac t that neither the NO synthase inhibitors (MB or ODQ) nor the c GMP phosp hodie ste rase inhibitor, MY5445, have any effe ct upon control hype ralgesia (Figs 3 -5 ). Neverthele ss, L-NAME displaye d an antinocic eptive effe ct. Rec ently, L-NAME has bee n described as ac ting as a partial agonist, 27 c ausing rapid induction of iNOS ge ne ex pre ssion. 21 Stimulation of NO synthesis may ex plain the observed L-NAME antinocic eption, in mode ls like those used in this inve stigation. In these mode ls it has be en previously show n that e ithe r NO donors or drugs w hich stimulate the arginine NO cGMP pathw ay c ause analge sia. 9,1 0,13 -18 Thus, the simple st comprehensive ex planation for the fac t that L-NAME antinocic eption w as inhibite d by a NO synthase inhibitor, L-NMMA and w as abolishe d by guanylate cyclase inhibitors, MB or ODQ as w e ll as pote ntiated by e ithe r the NO synthase substrate, arginine or by the cGMP phosphodieste rase inhibitor MY5445 is that L-NAME is activating iNOS. The fac t that i.c .v. co-inje ctions of MB or ODQ inhibited the ce ntral antinocic eptive e ffec t of L-NAME suggests a similar me chanism of action for the pe rip heral and ce ntral action for this age nt.
Finally, it must be pointed out that, in contrast w ith the re sults prese nte d he re , there are several observations indic ating that the intraplantar or syste mic administration of L-NAME has similar e ffec ts to other NO synthase inhibitors in causing antinoc iception. 4 ,2 0 This contradic tion may ex plaine d by considering that the ac tivation of the arginine /NO/cGMP pathw ay causes hyp eralgesia or analge sia de pending on the predominant type of fibres involved in the nocice ptive response or depending on the tissue level of NO. 2 0 From the the rape utic p oint of view, how e ve r, it se ems that during inflammatory pain in man, the activation of the arginine /NO/cGMP p athw ay causes analgesia. This suggestion is in line w ith the observations that NO donors are either e ffec tive as analgesic s by themselves or in c onjunction w ith other analgesics. 2 9 -32 In conclusion, the pre sent study confirms that L-NAME c ause s analgesia and de monstrated that L-NMMA, another NO synthase inhibitor, significantly blocked both the peripheral and c entral antinocice ptive actions of L-NAME in rats and mic e. L-NAME antinocic eption w as also blocke d by inhibitors of guanilate-c yclase activation and pote ntiated by arginine and by a cGMP phosphodie ste rase inhibitor. These re sults allow us to spe culate that L-NAME causes antinocic eption by ac ting as a partial agonist, thus stimulating iNOS ac tivation in the nocice ptive te sts use d. Furthermore our results draw into que stion the use of L-NAME alone as a me thodological tool to charac te rise the nocic eptive role of the arginine-NO-cGMP pathw ay in physiopathological proc esses, in the absenc e of confirmation w ith anothe r NO synthase inhibitor.