- Email: [email protected]
Local inhibition of inflammatory pain by naloxone and its N-methyl quarternary analogue
European Journal of Pharmacology, 96 (1983) 277-283
277
Elsevier
LOCAL INHIBITION OF INFLAMMATORY PAIN BY N A L O X O N E A N D I T S N - M E T H Y L QUATERNARY ANALOGUE
LUDMILLA RIOS and JOSEPH J.C. JACOB * Laboratory of Pharmacology, Pasteur Institute, F 75015 Paris, France
Received 25 April 1983, revised MS received 29 July 1983, accepted 29 September 1983
L. RIOS and J.J.C. JACOB, Local inhibition of inflammatory pain by naloxone and its N-methyl quaternary analogue, European J. Pharmacol. 96 (1983) 277-283. Inflammation was induced in rats by intraplantar administration of carrageenan (500 /~g in 0.1 ml). Nociceptive thresholds were measured on both inflamed and contralateral hindpaws with the pressure test of Randal and Sellito. Low doses (0.03-10 /tg/kg) of naioxone hydrochloride (Nx) or of methylnaloxone methylsulfonate (MeNx) were injected 4 h after carrageenan in the inflamed paw: Nx (3 #g) and MeNx (1-10 #g) diminished inflammatory pain; a slight, nonspecific hyperalgesia was observed in the contralateral paw after Nx, MeNx or NaC1. Neither drug was effective when injected s.c. at the same doses and time; however activities were shown to be influenced by the experimental schedule. Low s.c. doses of Nx antagonized the analgesia produced by s.c. morphine whereas MeNx did not. This work demonstrates the local site of the analgesic action of the doses of Nx and MeNx used here, bringing new evidence in favour of the existence of cutaneous opioid receptors. Opiate antagonists might inhibit inflammatory pain by interacting with a particular population of cutaneous receptors and (or) by being dealkylated locally into agonists. Naloxone
N-Methyl naloxone
Inhibition of inflammatory pain
1. Introduction
Naloxone (Nx), a systemic opiate antagonist, is known to facilitate various nociceptive reactions, an effect which points to the regulation of pain by endogenous opioids (review by Jacob and Ramabadran, 1981). However an opposite effect, i.e. antinociception, has also been reported in human beings (Buchsbaum et al., 1977) or mice (Jacob et al., 1978) hyperreactive to pain, in patients suffering from postoperative pain (Levine et al., 1979), and in rats after intrathecal injection (Woolf, 1980). Nx analgesia has also been described in rats subjected to experimental inflammatory pain, either under acute (local injection of prostaglandin; Ferreira and Nakamura, 1979a, b) or chronic condi* To whom all correspondence should be addressed: Laboratory of Pharmacology, Pasteur Institute, 28 rue du Docteur Roux, 75015 Paris, France. 0014-2999/84/$03.00 O 1984 Elsevier Science Publishers B.V.
Peripheral, local analgesia
tions (Freund's adjuvant; Kayser and Guilbaud, 1981). Ferreira and N a k a m u r a (1979a, b) obtained analgesia with intraplantar injections of Nx in the inflamed paw and suggested a local site of action, which was questioned by Kayser and Guilbaud (1981) who used the i.v. route. Other morphinomimetic effects have been reported by Jacob and Michaud (1976) who observed myosis and sedation in dogs after repeated administration of Nx while the initial injections had been followed by mydriasis and agitation. Several interpretations have been proposed: particular characteristics of the opioid receptors involved (Ferreira and Nakamura, 1979a, b; Levine et al., 1979; Woolf, 1980), agonist effects possibly due to dealkylated metabolites (Rios and Jacob, 1982), release of endogenous opioids (Pinsky et al., 1978; Levine et al., 1979), diminution of the diffuse noxious inhibitory control (Le Bars et al., 1981). In previous experiments (Rios and Jacob, 1982),
278 we had confirmed the analgesic effect of subcutaneous Nx on acute inflammatory pain produced by the intraplantar injection of carrageenan and shown the stereospecificity of this action by using two isomers ( - ) - and (+)-5,9 ot-diethyl-2-(3-furylmethyl)-2'-hydroxy-6,7-benzomorphan; Merz et al., 1974)). Nx analgesia was somehow linked with inflammation, since it was not observed in the contralateral non-inflamed paw. However it was not due to an overall antagonism of the mediator(s) of inflammation since the edema was not influenced. We also observed analgesia with a Nmethyl-quaternary analogue of Nx (MeNx) which probably acted peripherally since analgesia was observed at doses and at times when no central effect (antagonism of morphine analgesia) could be detected. Lorenzetti and Ferreira (1982) also reported inhibition of inflammatory hyperalgesia by nalorphine and by N-methyl nalorphine injected intraperitoneally. More direct evidence for a local site of action was, however, required. The main aim of this work was to assess this possibility by comparing the effects of intraplantar injections of Nx and MeNx in the inflamed paw with those of systemic administration.
stable, whereas in our previous work it was administered when inflammation was developing, its effects being measured on progressive and relatively moderate pain. The pressure thresholds are expressed in g. The antagonism of morphine (Mo) analgesia Was studied in several groups of rats by the hot plate test (55°C; nociceptive reaction: licking). Morphine sulfate (15 m g / k g ) and the antagonist were injected (s.c.) 30 and 10 min before the test, respectively. To avoid the effect of learning, rats were exposed only once to the hot plate (cut off time: 60 s). The results are expressed in latencies to lick(s); the inhibition percentage was calculated according to the formula: I% = [(LMo L v ) / ( L M o - L c ) ] × 100 where LMo, L c and L~v are the licking latencies of rats receiving morphine, saline, or morphine and antagonist, respectively. In both procedures, groups of 12 rats were used for each experiment. Nx was used as the hydrochloride and MeNx as the methylsulfonate. The doses are expressed in terms of Nx base for the two drugs.
3. Results 2. Materials and methods
Male Sprague Dawley rats (200-250 g) were used. Inflammation was induced by intraplantar (i.pl.) injection of carrageenan (500 ~g in 0.1 ml). The opiate antagonist was administered 4 h after carrageenan, either by the i.pl. route (0.01 m l / 1 0 0 g in the inflamed paw), or by s.c. injection under the skin of the neck (0.1 ml,/100 g). The controls received saline instead of the opiate antagonist. The nociceptive thresholds were measured according to the method of Randall and Sellito (1957), with a Basile algesimeter on both inflamed and contralateral hindpaws. The measurements were performed just before as well as 2, 3 and 4 h after the injection of carrageenan and 5, 10, 30, 60 and 120 min after the antagonist. This schedule differed from that used in our previous work (see fig. 2) in that it included the evaluation of early effects of the antagonist. Furthermore, the antagonist was injected when the pain thresholds were low and
3.1. Antinociceptive effects 3.1.1. Intraplantar route (fig. 1) Nx and MeNx were tested as doses ranging from 0.3 to 10 #g. For Nx, 0.3 and 1 /~g/kg were inactive while 3 /~g/kg produced a significant effect observed from 30 min on after the injection. Increasing the dose to 1 0 / ~ g / k g did not enhance the effect which, on the contrary, disappeared. For MeNx, significant effects were observed with 1 and 3 # g / k g : these effects appeared quite rapidly after the injection (within 5 min) and had a long duration (from 30 to 60 min). A dose of 10 ~ g / k g of MeNx was somewhat less effective than the lower doses. No increase in pain threshold was observed on the contralateral, non-inflamed, noninjected paw: on the contrary, a slight hyperalgesia seemed to occur, which was however not specific since it was also observed after injection of saline in the inflamed paw.
279
LOCAL
INJECTION 1 ~9
Nx
MeNx
g 120 I00 80
I
'~'i~%,~,. •
,-, . . . . . . . . . . . . . .
•
;l- - ' ~ ' ~%-._=....
. . . . . . . _- __-.
60
~'~.~...
4O 20 0
~t~
,
I
.
d~,~ ,~..¢,I i
/
i
i
i
j b-...l
g 120 tO0 BO
I
""~"~ ",.~
_...
1
60 40 20 0
g.¢ ¢/~¢1.,k¢, I/
i
i
i
I
g
/ ~ 10
120
~ ~,~ •
iO0
+" -+ - - _ t . . . . . . . .
80
i_~_-;
60
~,~
40
'
.
°
~---,
2O
l ....
0
,t,
,?
SUBCUTANEOUS INJECTION 3~g g 120 100 80 60 40 20
ico
Nx
~
s''~e
Co
MeNx
+
+
"
30
60
¢
~ o
0
-4h
0 5 1015
30
60
17'0 rain .4h
0 5 1015
120 rain
Fig. 1. Effects o f n a l o x o n e ( N x ) a n d m e t h y l n a l o x o n e ( M e N x ) o n the p r e s s u r e t h r e s h o l d s of the i n f l a m e d a n d o f the n o n i n f l a m e d , c o n t r a l a t e r a l p a w . Ordinates: p r e s s u r e t h r e s h o l d s in g; abscissae: t i m e in m i n . All a n i m a l s received c a r r a g e e n a n ( 1' C a ) i n t r a p l a n t a r in o n e p a w 4 h b e f o r e the i n j e c t i o n ( T ) o f saline O, N x O o r M e N x + . - inflamed paw ...... c o n t r a l a t e r a l p a w . Top: local i n j e c t i o n o f N x , M e N x o r saline i n t o the i n f l a m e d p a w . Bottom: s u b c u t a n e o u s injection. Left N x , r i g h t M e N x . D o s e s : p e r k g o f b o d y weight. T h e b a r s i n d i c a t e the s t a n d a r d e r r o r s o f the m e a n s . G r o u p s o f 12 a n i m a l s . * A n t i n o c i c e p t i v e effect s i g n i f i c a n t f o r P < 0.05.
280 Nx 120 IO0 8O
\
10
\
6O
40 20 0
~mcJlkgs.c
,ca
" f '
e---4~ _4 h 0
/P---e
i 30
15
610
150
g 140 120 500
80
These results are in contrast with those previously obtained with a different schedule, where Nx and MeNx were effective in the 3 /~g to 1 m g / k g s.c. dose range (Rios and Jacob, 1982). The influence of the schedule is illustrated in fig. 2 : 1 0 /~g/kg of s.c. Nx was ineffective when injected 4 h after carrageenan when pain was stabilized and intense as indicated by the low pressure threshold (20 g), but quite effective when administered by the same route, 30 min after carrageenan, that is during the early development of inflammation; the moderate decrease in threshold observed in the control animals (from 135 to 70 g), was delayed and significantly reduced in the treated rats.
6O 40
3.2. Antagonisms of morphine analgesia (table 1)
20 0
Ca
,t -30
,t
Nx
0
. . . . . . 30
90
150
210
Fig. 2. Influence of the experimental schedule on the antinociceptive efficacy of s.c. naloxone (Nx) on inflammatory pain. Ordinates, abscissae, bars and stars: as in fig. 1. Saline controls o, Nx (3. Top: same schedule as in fig. l, the drug studied ( 1' ) was administered 240 min after carrageenan ( 1' Ca), when the pressure thresholds had decreased below 30 g. Bottom: Nx or saline (T) was administered 30 min after carrageenan (T Ca); note that the pressure thresholds in the control paw decreased only from 135 to 120 g 60 min after Ca. Groups of 12 animals.
The s.c. administration of low doses of Nx clearly antagonized the morphine-induced analgesia when the latter was measured 10 min after the injection of the antagonist, that is at a time corresponding to an early measurement of inflammatory pain. The antagonism was dose-related (10_+ 15%, 43_+ 11% and 71 _+ 14% after 1, 3 and 10/~g/kg, respectively), Under the same conditions, no significant inhibition was observed for the same doses of MeNx (0%, 13 _+ 11% and 10 _+ 8% after 1, 3 and 10/~g/kg, respectively).
3.1.2. Subcutaneous route (figs. 1 and 2) The doses studied ranged from 0.3 to 30 # g / k g . N o analgesic effect was observed, as illustrated for 3 /zg/kg of Nx and MeNx in fig. 1 and for 10 /~g/kg of Nx, in fig. 2.
4. Discussion
The existence of peripheral analgesic effects of opiates had first been suggested but not proven by
TABLE 1 Antagonism of morphine analgesia in rat. Morphine (HC1) was administered 30 min and the antagonist 10 rain before the test (hot plate). The injections were subcutaneous. The number of rats is indicated in parentheses. The mean-licking latency of the saline controls was 5 _+1 s (18 rats). Dose of antagonist ( t~g / k g )
Mo alone (concomitant controls)
Licking latency (in s)_+S.E. Mo + Nx
Mo + MeNx
1
42_+5 (18)
38.5_+5 (12)
43_+5 (18)
3
43_+4(24)
27
_+4(18)
39_+4(18)
10
44_+4(18)
16
_+5(12)
40_+5(18)
281
Amsler et al. (quoted by Eddy, 1941). Experimental evidence was given for inflammatory pain by Ferreira and Nakamura (1979a) using local injections of morphine and by Lorenzetti and Ferreira (1982) who showed that a low dose of intraplantar N methylnalorphine inhibited the analgesic effect of systemic morphine in the injected but not in the contralateral paw. Peripheral analgesia by morphine and other opiates has also been suggested by Bentley et al. (1981) for inhibition of writhing. These authors showed with an appropriate schedule that morphine was at least 100 times more effective when injected locally (i.p.) than when injected i.v.; this action was antagonized by systemic naloxone. N-Methyl morphine was found effective in this test by Smith et al. (1982) who postulated a peripheral effect because the brain levels of [~4C]N-methylmorphine were low and considered by them as irrelevant. Our results demonstrate the peripheral, local site of the curious inhibition of inflammatory pain produced by opiate antagonists since low doses of Nx and lower ones of MeNx were effective when injected into the inflamed paw but not when administered systemically (s.c.). It seems unlikely that this effect could result from a local anesthetic property, since such activity was not observed after intradermal injection of much more concentrated solutions (3 m g / m l instead of 0.2-1 /~g/ml) in guinea-pigs. Among the various hypotheses presented in the Introduction to interpret the antinociceptive effects of opiate antagonists, the inhibition of a central diffuse inhibitory control (Le Bars et al., 1981) and a systemic release of endogenous opioids (Pinsky et al., 1978; Levine et al., 1979) cannot account for the facts observed here. Specific cutaneous receptors are probably involved since very low doses were effective and since this type of analgesia by opiate antagonist has been shown to be stereospecific (Rios and Jacob, 1982). However cutaneous opioid binding sites have not yet been identified either biochemically or histochemically, Van der Kooij (personal communication) having obtained just borderline evidence. As suggested by Ferreira and Nakamura (1979b), Levine et al. (1979) and Woolf (1980) the opioid receptors involved in analgesic effects of
antagonists might represent a particular population. According to Van der Kooy (1982), particular cutaneous receptors may indeed exist, since the local application of low concentrations of etorphine, a potent analgesic, on non-inflamed skin produced not analgesia but on the contrary, hyperalgesia which was antagonized by naloxone. Further, Bentley et al. (1981) considered that peritoneal receptors were of various types (/~, K and 6) and that those of the # type were atypical. As we have previously suggested (Rios and Jacob, 1982) the possibility also exists that Nx and MeNx may be dealkylated, yielding potent 'agonist' analgesics (oxymorphone and noroxymorphone). Since we have shown previously (Rios and Jacob, 1982) that when injected systematically Nx and MeNx inhibited pain in the inflamed paw without causing analgesia in the contralateral paw, we may assume that inflammation sensitizes the cutaneous receptors a n d / o r accelerates dealkylation. Better knowledge of the molecular mechanisms of interactions between specific binding sites and their 'agonist' or 'antagonist' ligands is needed. The analgesic effects of Nx and MeNx did not increase, but rather decreased when the dose was increased beyond an optimal one. Similar inverted U-shaped curves have been observed in the case of the central analgesic effects of mixed agonistantagonist drugs and they are thought to be due to the greater increase of the antagonist component which Would progressively limit and thereafter exceed the agonist component. The decrease of agonist effect with increasing dose was also reported in the case of the analgesic effect of naloxone administered intrathecally in rats (Woolf, 1980). For Nx, the sites of this antagonistic action may be local as suggested by Ferreira and Nakamura (1979a, b), as well as central since as we have now shown low s.c. doses of Nx antagonized the central analgesia induced by morphine and the effect was dose-related. For MeNx, we have not observed a significant central antagonism of Mo analgesia, which suggests that a local antagonist component may be prevalent. The fact that MeNx was more effective than Nx cannot yet be readily explained and we are currently investigating the factors involved using other animal models. One of these factors - indicated in
282 o u r present a n d in previous work - m a y be that M e N x had m u c h less central antagonistic action t h a n Nx when injected s.c. a fact c o n f i r m i n g that it permeates the b l o o d b r a i n barrier m u c h less easily. Similarly, Lorenzetti a n d Ferreira (1982) f o u n d that n o hyperalgesia (resulting from central a n t a g o n i s m ) interfered with the analgesic action of high doses of N - m e t h y l n a l o r p h i n e whereas it did with the action n a l o r p h i n e . However we saw n o hyperalgesia with the low doses of Nx a n d M e N x that we used, except a slight effect which appeared in the n o n i n f l a m e d paw when N x or M e N x were injected into the i n f l a m e d paw. This effect was not specific to opiate antagonists since it was also observed with the saline injection. It m a y represent a n aspect of stress hyperalgesia, as observed after other m a n i p u l a t i o n s (Vidal a n d Jacob, 1982). In the present case, this hyperalgesia became statistically significant when the results from all injected a n i m a l s were grouped. We have also d e m o n s t r a t e d that the s.c. injection of low doses of N x (up to 10 ~tg/kg) was ineffective when the i n f l a m m a t o r y pain was severe a n d stabilized whereas these doses were quite effective when the p a i n was moderate a n d developing. I n the latter case, s.c. injections of N x a n d M e N x were, as a whole, equally effective within a wide range of doses ( 3 - 3 0 0 btg; Rios and Jacob, 1982). It has previously been reported that mixed a g o n i s t - a n t a g o n i s t s were more active when the intensity of the nociceptive stimuli was diminished, e.g. p e n t a z o c i n e in the test of D ' A m o u r a n d Smith ( G r a y et al., 1970). A too intense a n d (or) stabilized p a i n m a y a c c o u n t for the failure of C h i p k i n et al. (1982) to observe an analgesic effect of n a l o x o n e in i n f l a m m a t i o n . In conclusion, the existence a n d the n a t u r e (hyperalgesia or analgesia) of the effects of opiate a n t a g o n i s t s o n nociception generally appear to be the result of m a n y factors: n a t u r e of the drug, dose, route of a d m i n i s t r a t i o n , n a t u r e of the nociceptive reaction a n d state of the animal, all of which influence the b a l a n c e between the a n t a g o n i s t a n d the (direct or indirect) agonist properties.
Acknowledgements The authors are indebted to Drs Ferster and Vernier (Endo Laboratories. U.S.A.) for the gift of naloxone hydrochloride and to Dr Moulineau (Clin Midy, France) for the gift of methylnaloxone methylsulfonate. They wish to thank Ms N. Prudhomme for technical assistance and Mr. C. Suaudeau for the illustrations. This research was supported by grants from D.R.E.T. (contract 81.524) and i.N.S.E.R.M. (contract CRL No. 823009).
References Bentley, G.A., S.H. Newton and J. Starr, 1981, Evidence for an action of morphine and the enkephalins on sensory nerve endings in the mouse peritoneum, Br. J. Pharmacol. 73, 325. Buchsbaum, M.S., G.C. Davis and W.E. Bunney Jr., 1977, Naloxone alters pain perception and somatosensory evoked potentials in normal subjects, Nature 270, 620. Chipkin, R.E., W. Billard, L.C. Iorio and A. Barnett, 1982, Relationship of yeast-injection induced changes in brain met-enkephalin and analgesia, Peptides 3, 31. Eddy, N.B., 1941, Action on the nervous system, in: The pharmacology of the opium alkaloids, eds. H. Krueger, N.B. Eddy and M. Sumwalt (U.S. Public Health Reports, Washington) Suppl. 165 part I, p. 1. Ferreira, S.H. and M. Nakamura, 1979a, II - Prostaglandin hyperalgesia: the peripheral analgesic activity of morphine, enkephalins and opioid antagonists, Prostaglandins 18, 191. Ferreira, S.H. and M. Nakamura, 1979b, II1 - Prostaglandin hyperalgesia: relevance of the peripheral effect for the analgesic action of opioid-antagonists, Prostaglandins 18, 201. Gray, W., A. Osterberg and T. Serito, 1970, Measurement of the analgesic efficacy and potency of pentazocine by the D'Amour and Smith method, J. Pharmacol. Exp. Ther. 172, 154. Jacob, J. and G. Michaud, 1979, Production par la naloxone d'effets inverses de ceux de la morphine chez le chien 6veill~, Arch. Int. Pharmacodyn. 222, 332. Jacob, J. and K. Ramabadran, 1981, Role of opiate receptors and endogenous ligands in nociception, Pharmacol. Ther. 14, 177. Jacob, J., K. Ramabadran, J.-M. Girault and C. Suaudeau, 1978, Endorphins, training and behavioral thermoregulation, in: Characteristics and function of opioid, eds. J.M. Van Ree and L. Terenius (Elsevier/North-Holland Biomedical Press, Amsterdam) p. 171. Kayser, V. and G. Guilbaud, 1981, Dose-dependent analgesic and hyperalgesic effects of systemic naloxone in arthritic rats, Brain Res. 226, 344. Le Bars, D., D. Chitour, E. Kraus, A.H. Dickenson and J.-M. Besson, 1981, Effect of naloxone upon diffuse noxious inhibitory controls (DNIC) in the rat, Brain Res. 204, 387.
283 Levine, J.D., N.C. Gordon and H.L. Fields, 1979, Naloxone dose dependently produces analgesia and hyperalgesia in postoperative pain, Nature 278, 740. Lorenzetti, B.B. and S.H. Ferreira, 1982, The analgesic effect of quaternary analogues of morphine and nalorphine, Brazilian J. Med. Biol. Res. 15, 285. Merz, H., A. Langbein, K. Stokhaus, G. Walter and H. Wick, Structure-activity relationship in narcotic antagonists with N-furylmethyl substituents, in: Narcotic antagonist, eds. M.C. Braude. L.S. Harris, E.L. May, J.P. Smith and J.E. Villarreal (Raven Press, New York) p. 91. Pinsky, C., F.S. Labella, V. Havlicek and A.K. Dua, 1978, Apparent central agonist actions of naloxone in the unrestrained rat, in: Characteristics and Function of Opioid, eds. J.M. Van Ree and L. Terenius (Elsevier/North-Holland Biomedical Press, Amsterdam) p. 439.
Randall, L.O. and J.J. Sellito, 1957, A method for measurement of analgesic activity on inflamed tissue, Arch. Int. Pharmacodyn, 111,409. Rios, L. and J. Jacob, 1982, Inhibition of inflammatory pain by naloxone and its N methyl quaternary analogue, Life Sci. 31, 1209. Smith, T.W., P. Puchan, D.N. Parsons and S. Wilkinson, 1982, Peripheral antinociceptive effects of N methylmorphine, Life Sci. 31, 1205. Van der Kooij, D., 1982, Hyperalgesic functions of peripheral sensory nerve opiate receptors, Neurosci. Lett. Suppl. 10, $498. Vidal, C. and J. Jacob, 1982, Hyperalgesia induced by nonnoxious stress in the rat, Neurosci. Lett. 32, 75. Woolf, C.J., 1980, Analgesia and hyperalgesia produced in the rat by intrathecal naloxone, Brain Res. 189, 593.
277
Elsevier
LOCAL INHIBITION OF INFLAMMATORY PAIN BY N A L O X O N E A N D I T S N - M E T H Y L QUATERNARY ANALOGUE
LUDMILLA RIOS and JOSEPH J.C. JACOB * Laboratory of Pharmacology, Pasteur Institute, F 75015 Paris, France
Received 25 April 1983, revised MS received 29 July 1983, accepted 29 September 1983
L. RIOS and J.J.C. JACOB, Local inhibition of inflammatory pain by naloxone and its N-methyl quaternary analogue, European J. Pharmacol. 96 (1983) 277-283. Inflammation was induced in rats by intraplantar administration of carrageenan (500 /~g in 0.1 ml). Nociceptive thresholds were measured on both inflamed and contralateral hindpaws with the pressure test of Randal and Sellito. Low doses (0.03-10 /tg/kg) of naioxone hydrochloride (Nx) or of methylnaloxone methylsulfonate (MeNx) were injected 4 h after carrageenan in the inflamed paw: Nx (3 #g) and MeNx (1-10 #g) diminished inflammatory pain; a slight, nonspecific hyperalgesia was observed in the contralateral paw after Nx, MeNx or NaC1. Neither drug was effective when injected s.c. at the same doses and time; however activities were shown to be influenced by the experimental schedule. Low s.c. doses of Nx antagonized the analgesia produced by s.c. morphine whereas MeNx did not. This work demonstrates the local site of the analgesic action of the doses of Nx and MeNx used here, bringing new evidence in favour of the existence of cutaneous opioid receptors. Opiate antagonists might inhibit inflammatory pain by interacting with a particular population of cutaneous receptors and (or) by being dealkylated locally into agonists. Naloxone
N-Methyl naloxone
Inhibition of inflammatory pain
1. Introduction
Naloxone (Nx), a systemic opiate antagonist, is known to facilitate various nociceptive reactions, an effect which points to the regulation of pain by endogenous opioids (review by Jacob and Ramabadran, 1981). However an opposite effect, i.e. antinociception, has also been reported in human beings (Buchsbaum et al., 1977) or mice (Jacob et al., 1978) hyperreactive to pain, in patients suffering from postoperative pain (Levine et al., 1979), and in rats after intrathecal injection (Woolf, 1980). Nx analgesia has also been described in rats subjected to experimental inflammatory pain, either under acute (local injection of prostaglandin; Ferreira and Nakamura, 1979a, b) or chronic condi* To whom all correspondence should be addressed: Laboratory of Pharmacology, Pasteur Institute, 28 rue du Docteur Roux, 75015 Paris, France. 0014-2999/84/$03.00 O 1984 Elsevier Science Publishers B.V.
Peripheral, local analgesia
tions (Freund's adjuvant; Kayser and Guilbaud, 1981). Ferreira and N a k a m u r a (1979a, b) obtained analgesia with intraplantar injections of Nx in the inflamed paw and suggested a local site of action, which was questioned by Kayser and Guilbaud (1981) who used the i.v. route. Other morphinomimetic effects have been reported by Jacob and Michaud (1976) who observed myosis and sedation in dogs after repeated administration of Nx while the initial injections had been followed by mydriasis and agitation. Several interpretations have been proposed: particular characteristics of the opioid receptors involved (Ferreira and Nakamura, 1979a, b; Levine et al., 1979; Woolf, 1980), agonist effects possibly due to dealkylated metabolites (Rios and Jacob, 1982), release of endogenous opioids (Pinsky et al., 1978; Levine et al., 1979), diminution of the diffuse noxious inhibitory control (Le Bars et al., 1981). In previous experiments (Rios and Jacob, 1982),
278 we had confirmed the analgesic effect of subcutaneous Nx on acute inflammatory pain produced by the intraplantar injection of carrageenan and shown the stereospecificity of this action by using two isomers ( - ) - and (+)-5,9 ot-diethyl-2-(3-furylmethyl)-2'-hydroxy-6,7-benzomorphan; Merz et al., 1974)). Nx analgesia was somehow linked with inflammation, since it was not observed in the contralateral non-inflamed paw. However it was not due to an overall antagonism of the mediator(s) of inflammation since the edema was not influenced. We also observed analgesia with a Nmethyl-quaternary analogue of Nx (MeNx) which probably acted peripherally since analgesia was observed at doses and at times when no central effect (antagonism of morphine analgesia) could be detected. Lorenzetti and Ferreira (1982) also reported inhibition of inflammatory hyperalgesia by nalorphine and by N-methyl nalorphine injected intraperitoneally. More direct evidence for a local site of action was, however, required. The main aim of this work was to assess this possibility by comparing the effects of intraplantar injections of Nx and MeNx in the inflamed paw with those of systemic administration.
stable, whereas in our previous work it was administered when inflammation was developing, its effects being measured on progressive and relatively moderate pain. The pressure thresholds are expressed in g. The antagonism of morphine (Mo) analgesia Was studied in several groups of rats by the hot plate test (55°C; nociceptive reaction: licking). Morphine sulfate (15 m g / k g ) and the antagonist were injected (s.c.) 30 and 10 min before the test, respectively. To avoid the effect of learning, rats were exposed only once to the hot plate (cut off time: 60 s). The results are expressed in latencies to lick(s); the inhibition percentage was calculated according to the formula: I% = [(LMo L v ) / ( L M o - L c ) ] × 100 where LMo, L c and L~v are the licking latencies of rats receiving morphine, saline, or morphine and antagonist, respectively. In both procedures, groups of 12 rats were used for each experiment. Nx was used as the hydrochloride and MeNx as the methylsulfonate. The doses are expressed in terms of Nx base for the two drugs.
3. Results 2. Materials and methods
Male Sprague Dawley rats (200-250 g) were used. Inflammation was induced by intraplantar (i.pl.) injection of carrageenan (500 ~g in 0.1 ml). The opiate antagonist was administered 4 h after carrageenan, either by the i.pl. route (0.01 m l / 1 0 0 g in the inflamed paw), or by s.c. injection under the skin of the neck (0.1 ml,/100 g). The controls received saline instead of the opiate antagonist. The nociceptive thresholds were measured according to the method of Randall and Sellito (1957), with a Basile algesimeter on both inflamed and contralateral hindpaws. The measurements were performed just before as well as 2, 3 and 4 h after the injection of carrageenan and 5, 10, 30, 60 and 120 min after the antagonist. This schedule differed from that used in our previous work (see fig. 2) in that it included the evaluation of early effects of the antagonist. Furthermore, the antagonist was injected when the pain thresholds were low and
3.1. Antinociceptive effects 3.1.1. Intraplantar route (fig. 1) Nx and MeNx were tested as doses ranging from 0.3 to 10 #g. For Nx, 0.3 and 1 /~g/kg were inactive while 3 /~g/kg produced a significant effect observed from 30 min on after the injection. Increasing the dose to 1 0 / ~ g / k g did not enhance the effect which, on the contrary, disappeared. For MeNx, significant effects were observed with 1 and 3 # g / k g : these effects appeared quite rapidly after the injection (within 5 min) and had a long duration (from 30 to 60 min). A dose of 10 ~ g / k g of MeNx was somewhat less effective than the lower doses. No increase in pain threshold was observed on the contralateral, non-inflamed, noninjected paw: on the contrary, a slight hyperalgesia seemed to occur, which was however not specific since it was also observed after injection of saline in the inflamed paw.
279
LOCAL
INJECTION 1 ~9
Nx
MeNx
g 120 I00 80
I
'~'i~%,~,. •
,-, . . . . . . . . . . . . . .
•
;l- - ' ~ ' ~%-._=....
. . . . . . . _- __-.
60
~'~.~...
4O 20 0
~t~
,
I
.
d~,~ ,~..¢,I i
/
i
i
i
j b-...l
g 120 tO0 BO
I
""~"~ ",.~
_...
1
60 40 20 0
g.¢ ¢/~¢1.,k¢, I/
i
i
i
I
g
/ ~ 10
120
~ ~,~ •
iO0
+" -+ - - _ t . . . . . . . .
80
i_~_-;
60
~,~
40
'
.
°
~---,
2O
l ....
0
,t,
,?
SUBCUTANEOUS INJECTION 3~g g 120 100 80 60 40 20
ico
Nx
~
s''~e
Co
MeNx
+
+
"
30
60
¢
~ o
0
-4h
0 5 1015
30
60
17'0 rain .4h
0 5 1015
120 rain
Fig. 1. Effects o f n a l o x o n e ( N x ) a n d m e t h y l n a l o x o n e ( M e N x ) o n the p r e s s u r e t h r e s h o l d s of the i n f l a m e d a n d o f the n o n i n f l a m e d , c o n t r a l a t e r a l p a w . Ordinates: p r e s s u r e t h r e s h o l d s in g; abscissae: t i m e in m i n . All a n i m a l s received c a r r a g e e n a n ( 1' C a ) i n t r a p l a n t a r in o n e p a w 4 h b e f o r e the i n j e c t i o n ( T ) o f saline O, N x O o r M e N x + . - inflamed paw ...... c o n t r a l a t e r a l p a w . Top: local i n j e c t i o n o f N x , M e N x o r saline i n t o the i n f l a m e d p a w . Bottom: s u b c u t a n e o u s injection. Left N x , r i g h t M e N x . D o s e s : p e r k g o f b o d y weight. T h e b a r s i n d i c a t e the s t a n d a r d e r r o r s o f the m e a n s . G r o u p s o f 12 a n i m a l s . * A n t i n o c i c e p t i v e effect s i g n i f i c a n t f o r P < 0.05.
280 Nx 120 IO0 8O
\
10
\
6O
40 20 0
~mcJlkgs.c
,ca
" f '
e---4~ _4 h 0
/P---e
i 30
15
610
150
g 140 120 500
80
These results are in contrast with those previously obtained with a different schedule, where Nx and MeNx were effective in the 3 /~g to 1 m g / k g s.c. dose range (Rios and Jacob, 1982). The influence of the schedule is illustrated in fig. 2 : 1 0 /~g/kg of s.c. Nx was ineffective when injected 4 h after carrageenan when pain was stabilized and intense as indicated by the low pressure threshold (20 g), but quite effective when administered by the same route, 30 min after carrageenan, that is during the early development of inflammation; the moderate decrease in threshold observed in the control animals (from 135 to 70 g), was delayed and significantly reduced in the treated rats.
6O 40
3.2. Antagonisms of morphine analgesia (table 1)
20 0
Ca
,t -30
,t
Nx
0
. . . . . . 30
90
150
210
Fig. 2. Influence of the experimental schedule on the antinociceptive efficacy of s.c. naloxone (Nx) on inflammatory pain. Ordinates, abscissae, bars and stars: as in fig. 1. Saline controls o, Nx (3. Top: same schedule as in fig. l, the drug studied ( 1' ) was administered 240 min after carrageenan ( 1' Ca), when the pressure thresholds had decreased below 30 g. Bottom: Nx or saline (T) was administered 30 min after carrageenan (T Ca); note that the pressure thresholds in the control paw decreased only from 135 to 120 g 60 min after Ca. Groups of 12 animals.
The s.c. administration of low doses of Nx clearly antagonized the morphine-induced analgesia when the latter was measured 10 min after the injection of the antagonist, that is at a time corresponding to an early measurement of inflammatory pain. The antagonism was dose-related (10_+ 15%, 43_+ 11% and 71 _+ 14% after 1, 3 and 10/~g/kg, respectively), Under the same conditions, no significant inhibition was observed for the same doses of MeNx (0%, 13 _+ 11% and 10 _+ 8% after 1, 3 and 10/~g/kg, respectively).
3.1.2. Subcutaneous route (figs. 1 and 2) The doses studied ranged from 0.3 to 30 # g / k g . N o analgesic effect was observed, as illustrated for 3 /zg/kg of Nx and MeNx in fig. 1 and for 10 /~g/kg of Nx, in fig. 2.
4. Discussion
The existence of peripheral analgesic effects of opiates had first been suggested but not proven by
TABLE 1 Antagonism of morphine analgesia in rat. Morphine (HC1) was administered 30 min and the antagonist 10 rain before the test (hot plate). The injections were subcutaneous. The number of rats is indicated in parentheses. The mean-licking latency of the saline controls was 5 _+1 s (18 rats). Dose of antagonist ( t~g / k g )
Mo alone (concomitant controls)
Licking latency (in s)_+S.E. Mo + Nx
Mo + MeNx
1
42_+5 (18)
38.5_+5 (12)
43_+5 (18)
3
43_+4(24)
27
_+4(18)
39_+4(18)
10
44_+4(18)
16
_+5(12)
40_+5(18)
281
Amsler et al. (quoted by Eddy, 1941). Experimental evidence was given for inflammatory pain by Ferreira and Nakamura (1979a) using local injections of morphine and by Lorenzetti and Ferreira (1982) who showed that a low dose of intraplantar N methylnalorphine inhibited the analgesic effect of systemic morphine in the injected but not in the contralateral paw. Peripheral analgesia by morphine and other opiates has also been suggested by Bentley et al. (1981) for inhibition of writhing. These authors showed with an appropriate schedule that morphine was at least 100 times more effective when injected locally (i.p.) than when injected i.v.; this action was antagonized by systemic naloxone. N-Methyl morphine was found effective in this test by Smith et al. (1982) who postulated a peripheral effect because the brain levels of [~4C]N-methylmorphine were low and considered by them as irrelevant. Our results demonstrate the peripheral, local site of the curious inhibition of inflammatory pain produced by opiate antagonists since low doses of Nx and lower ones of MeNx were effective when injected into the inflamed paw but not when administered systemically (s.c.). It seems unlikely that this effect could result from a local anesthetic property, since such activity was not observed after intradermal injection of much more concentrated solutions (3 m g / m l instead of 0.2-1 /~g/ml) in guinea-pigs. Among the various hypotheses presented in the Introduction to interpret the antinociceptive effects of opiate antagonists, the inhibition of a central diffuse inhibitory control (Le Bars et al., 1981) and a systemic release of endogenous opioids (Pinsky et al., 1978; Levine et al., 1979) cannot account for the facts observed here. Specific cutaneous receptors are probably involved since very low doses were effective and since this type of analgesia by opiate antagonist has been shown to be stereospecific (Rios and Jacob, 1982). However cutaneous opioid binding sites have not yet been identified either biochemically or histochemically, Van der Kooij (personal communication) having obtained just borderline evidence. As suggested by Ferreira and Nakamura (1979b), Levine et al. (1979) and Woolf (1980) the opioid receptors involved in analgesic effects of
antagonists might represent a particular population. According to Van der Kooy (1982), particular cutaneous receptors may indeed exist, since the local application of low concentrations of etorphine, a potent analgesic, on non-inflamed skin produced not analgesia but on the contrary, hyperalgesia which was antagonized by naloxone. Further, Bentley et al. (1981) considered that peritoneal receptors were of various types (/~, K and 6) and that those of the # type were atypical. As we have previously suggested (Rios and Jacob, 1982) the possibility also exists that Nx and MeNx may be dealkylated, yielding potent 'agonist' analgesics (oxymorphone and noroxymorphone). Since we have shown previously (Rios and Jacob, 1982) that when injected systematically Nx and MeNx inhibited pain in the inflamed paw without causing analgesia in the contralateral paw, we may assume that inflammation sensitizes the cutaneous receptors a n d / o r accelerates dealkylation. Better knowledge of the molecular mechanisms of interactions between specific binding sites and their 'agonist' or 'antagonist' ligands is needed. The analgesic effects of Nx and MeNx did not increase, but rather decreased when the dose was increased beyond an optimal one. Similar inverted U-shaped curves have been observed in the case of the central analgesic effects of mixed agonistantagonist drugs and they are thought to be due to the greater increase of the antagonist component which Would progressively limit and thereafter exceed the agonist component. The decrease of agonist effect with increasing dose was also reported in the case of the analgesic effect of naloxone administered intrathecally in rats (Woolf, 1980). For Nx, the sites of this antagonistic action may be local as suggested by Ferreira and Nakamura (1979a, b), as well as central since as we have now shown low s.c. doses of Nx antagonized the central analgesia induced by morphine and the effect was dose-related. For MeNx, we have not observed a significant central antagonism of Mo analgesia, which suggests that a local antagonist component may be prevalent. The fact that MeNx was more effective than Nx cannot yet be readily explained and we are currently investigating the factors involved using other animal models. One of these factors - indicated in
282 o u r present a n d in previous work - m a y be that M e N x had m u c h less central antagonistic action t h a n Nx when injected s.c. a fact c o n f i r m i n g that it permeates the b l o o d b r a i n barrier m u c h less easily. Similarly, Lorenzetti a n d Ferreira (1982) f o u n d that n o hyperalgesia (resulting from central a n t a g o n i s m ) interfered with the analgesic action of high doses of N - m e t h y l n a l o r p h i n e whereas it did with the action n a l o r p h i n e . However we saw n o hyperalgesia with the low doses of Nx a n d M e N x that we used, except a slight effect which appeared in the n o n i n f l a m e d paw when N x or M e N x were injected into the i n f l a m e d paw. This effect was not specific to opiate antagonists since it was also observed with the saline injection. It m a y represent a n aspect of stress hyperalgesia, as observed after other m a n i p u l a t i o n s (Vidal a n d Jacob, 1982). In the present case, this hyperalgesia became statistically significant when the results from all injected a n i m a l s were grouped. We have also d e m o n s t r a t e d that the s.c. injection of low doses of N x (up to 10 ~tg/kg) was ineffective when the i n f l a m m a t o r y pain was severe a n d stabilized whereas these doses were quite effective when the p a i n was moderate a n d developing. I n the latter case, s.c. injections of N x a n d M e N x were, as a whole, equally effective within a wide range of doses ( 3 - 3 0 0 btg; Rios and Jacob, 1982). It has previously been reported that mixed a g o n i s t - a n t a g o n i s t s were more active when the intensity of the nociceptive stimuli was diminished, e.g. p e n t a z o c i n e in the test of D ' A m o u r a n d Smith ( G r a y et al., 1970). A too intense a n d (or) stabilized p a i n m a y a c c o u n t for the failure of C h i p k i n et al. (1982) to observe an analgesic effect of n a l o x o n e in i n f l a m m a t i o n . In conclusion, the existence a n d the n a t u r e (hyperalgesia or analgesia) of the effects of opiate a n t a g o n i s t s o n nociception generally appear to be the result of m a n y factors: n a t u r e of the drug, dose, route of a d m i n i s t r a t i o n , n a t u r e of the nociceptive reaction a n d state of the animal, all of which influence the b a l a n c e between the a n t a g o n i s t a n d the (direct or indirect) agonist properties.
Acknowledgements The authors are indebted to Drs Ferster and Vernier (Endo Laboratories. U.S.A.) for the gift of naloxone hydrochloride and to Dr Moulineau (Clin Midy, France) for the gift of methylnaloxone methylsulfonate. They wish to thank Ms N. Prudhomme for technical assistance and Mr. C. Suaudeau for the illustrations. This research was supported by grants from D.R.E.T. (contract 81.524) and i.N.S.E.R.M. (contract CRL No. 823009).
References Bentley, G.A., S.H. Newton and J. Starr, 1981, Evidence for an action of morphine and the enkephalins on sensory nerve endings in the mouse peritoneum, Br. J. Pharmacol. 73, 325. Buchsbaum, M.S., G.C. Davis and W.E. Bunney Jr., 1977, Naloxone alters pain perception and somatosensory evoked potentials in normal subjects, Nature 270, 620. Chipkin, R.E., W. Billard, L.C. Iorio and A. Barnett, 1982, Relationship of yeast-injection induced changes in brain met-enkephalin and analgesia, Peptides 3, 31. Eddy, N.B., 1941, Action on the nervous system, in: The pharmacology of the opium alkaloids, eds. H. Krueger, N.B. Eddy and M. Sumwalt (U.S. Public Health Reports, Washington) Suppl. 165 part I, p. 1. Ferreira, S.H. and M. Nakamura, 1979a, II - Prostaglandin hyperalgesia: the peripheral analgesic activity of morphine, enkephalins and opioid antagonists, Prostaglandins 18, 191. Ferreira, S.H. and M. Nakamura, 1979b, II1 - Prostaglandin hyperalgesia: relevance of the peripheral effect for the analgesic action of opioid-antagonists, Prostaglandins 18, 201. Gray, W., A. Osterberg and T. Serito, 1970, Measurement of the analgesic efficacy and potency of pentazocine by the D'Amour and Smith method, J. Pharmacol. Exp. Ther. 172, 154. Jacob, J. and G. Michaud, 1979, Production par la naloxone d'effets inverses de ceux de la morphine chez le chien 6veill~, Arch. Int. Pharmacodyn. 222, 332. Jacob, J. and K. Ramabadran, 1981, Role of opiate receptors and endogenous ligands in nociception, Pharmacol. Ther. 14, 177. Jacob, J., K. Ramabadran, J.-M. Girault and C. Suaudeau, 1978, Endorphins, training and behavioral thermoregulation, in: Characteristics and function of opioid, eds. J.M. Van Ree and L. Terenius (Elsevier/North-Holland Biomedical Press, Amsterdam) p. 171. Kayser, V. and G. Guilbaud, 1981, Dose-dependent analgesic and hyperalgesic effects of systemic naloxone in arthritic rats, Brain Res. 226, 344. Le Bars, D., D. Chitour, E. Kraus, A.H. Dickenson and J.-M. Besson, 1981, Effect of naloxone upon diffuse noxious inhibitory controls (DNIC) in the rat, Brain Res. 204, 387.
283 Levine, J.D., N.C. Gordon and H.L. Fields, 1979, Naloxone dose dependently produces analgesia and hyperalgesia in postoperative pain, Nature 278, 740. Lorenzetti, B.B. and S.H. Ferreira, 1982, The analgesic effect of quaternary analogues of morphine and nalorphine, Brazilian J. Med. Biol. Res. 15, 285. Merz, H., A. Langbein, K. Stokhaus, G. Walter and H. Wick, Structure-activity relationship in narcotic antagonists with N-furylmethyl substituents, in: Narcotic antagonist, eds. M.C. Braude. L.S. Harris, E.L. May, J.P. Smith and J.E. Villarreal (Raven Press, New York) p. 91. Pinsky, C., F.S. Labella, V. Havlicek and A.K. Dua, 1978, Apparent central agonist actions of naloxone in the unrestrained rat, in: Characteristics and Function of Opioid, eds. J.M. Van Ree and L. Terenius (Elsevier/North-Holland Biomedical Press, Amsterdam) p. 439.
Randall, L.O. and J.J. Sellito, 1957, A method for measurement of analgesic activity on inflamed tissue, Arch. Int. Pharmacodyn, 111,409. Rios, L. and J. Jacob, 1982, Inhibition of inflammatory pain by naloxone and its N methyl quaternary analogue, Life Sci. 31, 1209. Smith, T.W., P. Puchan, D.N. Parsons and S. Wilkinson, 1982, Peripheral antinociceptive effects of N methylmorphine, Life Sci. 31, 1205. Van der Kooij, D., 1982, Hyperalgesic functions of peripheral sensory nerve opiate receptors, Neurosci. Lett. Suppl. 10, $498. Vidal, C. and J. Jacob, 1982, Hyperalgesia induced by nonnoxious stress in the rat, Neurosci. Lett. 32, 75. Woolf, C.J., 1980, Analgesia and hyperalgesia produced in the rat by intrathecal naloxone, Brain Res. 189, 593.