Neuroscience Vol. 42, No. 2, pp. 541-553, 1991 Printed in Great Britain

0306-4522/91 $3.00 + 0.00 Pergamon Press plc © 1991 IBRO

OPIOID SYSTEMS IN THE RESPONSE TO INFLAMMATORY PAIN: SUSTAINED BLOCKADE SUGGESTS ROLE OF x- BUT NOT #-OPIOID RECEPTORS IN THE MODULATION OF NOCICEPTION, BEHAVIOUR A N D PATHOLOGY M. J. MILLAN* and F. C. COLPAERT FONDAX-Groupe de Recherche SERVIER, 7, rue Amp6re, 92800 Puteaux, France Abstract---One day after intraplantar inoculation of Mycobacterium butyricum into the right hind-paw, unilaterally inflamed and control rats were implanted subcutaneously with osmotic minipumps delivering naloxone at 0.16 or 3.0 mg/kg/h or vehicle. As determined three days after implantation, 0.16 mg/kg/h of naloxone completely antagonized the antinociceptive action of the/~-agonist, morphine, but did not affect antinociception evoked by the x-agonist, U69,593. In contrast, at 3.0 mg/kg/h, naloxone blocked both morphine- and U69,593-induced antinociception. Thus, 0.16mg/kg ("low dose") and 3.0 mg/kg ("high dose") of naloxone block/t, or #- plus x-opioid receptors, respectively. Pumps were removed one week following their implantation. Inoculation was associated with a sustained hyperalgesia of the inflamed paw to noxious pressure, and elevation in resting core temperature, a loss of body weight, hypophagia, hypodipsia and a reduction in mobility. These parameters were differentially modified by the high as compared to the low dose of naloxone. Two days following implantation of pumps delivering the high dose of naloxone, the hyperalgesia of the inflamed paw was potentiated: by six days, this effect was lost. Further, one day after removal of pumps yielding the high dose, the inflamed paw showed a normalization of thresholds, that is a "rebound antinociception". One day later, this effect had subsided. In distinction, at no time did the low dose of naloxone modify nociceptive thresholds. The high dose of naloxone enhanced the reduction in body weight and food intake shown by unilaterally inflamed rats whereas the low dose was ineffective. Neither dose affected the reduction in water intake or hypothermia of unilaterally inflamed animals. The high dose of naloxone reduced the mobility of unilaterally inflamed rats whereas the low dose was ineffective. Finally, by 10 days following pump removal, pathology had transferred to the contralateral paw. In rats which had received the high but not the low dose, this transfer was blocked, It is concluded that blockade of x-opioid receptors with a high dose of naloxone experts pronounced functional effects in unilaterally inflamed rats. In distinction, selective blockade of/~-receptors with a low dose is ineffective. The changes seen include not only an enhancement of the hyperalgesia of the inflamed tissue, but also an exacerbation of variables (body weight, food intake and motility) which reflect pain states. These data support the argument [Millan M. J. et aL (1990) Trends Pharmac. 11, 70-76; (1988) Pain 35, 299-312; (1987) J. Neurosci. 7, 77-87; (1986) .L Neurosci. 6, 899-916] of a role of x-opioid receptors in the response to and modulation of inflammatory pain. Further, the data suggest that opioid mechanism plays a role in processes underlying the evolution of arthritic disease.

In recent years, r - o p i o i d agonists have attracted great attention concerning their therapeutic potential as novel analgesics) 9"4~ Indeed, either upon systemic administration to rodents and primates, or upon direct application into the brain or spinal cord of rodents, x-agonists reliably induce antinociception. 5,~9,4~ The opioid peptide, dynorphin, is considered to be an endogenous ligand of r-receptors and the anatomical organization of both r-receptors and dynorphin is consistent with a physiological role in the control of nociception. Thus, r-receptors and dynorphin are concentrated in those superficial (I, II) *To whom correspondence should be addressed. Abbreviations: MR 2266, (-)5,9-diethyl-2-(3-furyl-methyl)-

2'-hydroxy-6,7-benzomorphan; U69,593, (5-at,8-~t,8-fl)(+)-N-methyl-N-[7-(-pyrrolidinyl)- 1- oxaspirol (4,5) (dec-8-yl)] benzenacetamide. 541

and deep (IV, V) laminae of the spinal cord which receive primary afferent noxious information from the periphery. 3'5'51'53 Recent work in the rat has demonstrated a pronounced response of these segmental pools of dynorphin and x-receptors in two experimental models of inflammatory pain; that is, either localized inflammation of a single hindlimb or chronic polyarthritis of all four limbs. 2°'21''u'45'47'6°'71 In both models, the spinal cord reveals a pronounced rise in levels o f immunoreactive dynorphin and a striking enhancement in the staining of those dynorphin-containing neurons which receive noxious input from the inflamed tissue. 2°,~,45,47,55,60,71 The parallel elevation in levels of m R N A encoding dynorphin indicates that the functional activity of dynorphinergic neurons is enhanced. 18,2°a~'~°'TjFurther, acute pharmacological

542

M.J. MILLANand F. C. COLPAERT

blockade of opioid receptors, probably x- rather than p- and 6-receptors, potentiates the hyperalgesia of the inflamed paws to application of noxious stimuli. 26'27'38'44'4-" These data suggest that an action of dynorphin, via x-receptors, may be involved in the control of nociception under conditions of inflammatory pain. However, there remain questions concerning this interpretation which cannot be adequately addressed by the acute administration of opioid antagonists. First, to date, the only parameter which has been evaluated concerning the influence of opioid antagonists is the response of the inflamed paw to noxious pressure. There are several other parameters considered to reflect the pain experienced by the animal and which are profoundly modified in human patients suffering from chronic pain: these include body weight gain, appetite and motility. 12,14One may hypothesize that an exacerbation of pain should be accompanied by an influence upon these variables. Second, it cannot be excluded that the effects previously reported with acute application of opioid antagonists are due to a rapid influence upon the pathology of the inoculated tissue. Indeed, opioids are known to affect both inflammatory states 2~,7° and immune ~3'58,65 processes underlying arthritic disease. Interestingly, central pools of opioids have been suggested to mediate the immunosuppression associated with acute exposure to stress. 62 However, very few other data are available concerning the effects of in vivo blockade of the activity of endogenous opioid systems upon inflammatory processes and immune function. Third, one might speculate that a long-term rather than transient dysfunction or disruption of opioid mechanisms may aggravate, or even precipitate, chronic pain conditions. Alternatively, mechanisms may exist which can compensate for the inactivation of opioid systems. One approach via which such issues may be examined is that the sustained inactivation of opioid receptors by administration of opioid antagonists via osmotic mini-pumps. 74 We have recently developed a methodology for the inactivation of either p-receptors only, or both p- and x-opioid receptors, by perfusion of a low or high dose of naloxone, respectively. 48'49'54 In this study, we employed this procedure for an evaluation o f the functional consequences of differential blockade of p- and x-receptors in rats with unilateral inflammation induced by inoculation of a single hindpaw with Mycobacterium

butyricum.2O21,43,44

EXPERIMENTAL PROCEDURES

Animals Male Wistar rats (Iffa Credo, Illskirchen, France) weighing 160 g were housed singly in sawdust-lined cages with unrestricted access to standard laboratory chow and tap water. The laboratory was equipped with a 12 h/12 h light/dark cycle with lights on from 7.30a.m. to 7.30 p.m. Humidity was maintained at 60+ 5% and temperature at 21 _+ I°C. All experiments were performed in the light phase.

Organization of experiments Unless otherwise specified, each experiment employed separate groups of rats. In Table 1, the organization and experimental protocol is summarized; the general programme was as follows. One day after inoculation, control or unilaterally inflamed rats were implanted with pumps delivering vehicle, or 0.16 or 3.0 mg/kg naloxone. In the week in which the pumps were present, inflammation, nociceptive thresholds, ingestive behaviour, core temperature and locomotor activity were monitored. After one week, pumps were removed. Over the following two days, behavioural testing was pursued. Finally, 10 days following removal of pumps, transfer of pathology to the contralateral, non-inoculated paw was examined.

Inoculation and implantations of pumps As in previous studies, 43.44rats were lightly anaesthetized with ether and inoculated in the plantar surface of the right hindpaw with 0.1 ml of Mycobacterium butyricum suspended in paraffin oil (Calbiochem, La Jolla, CA). Control rats were not injected. The following day, as previously described, 48,49'54 and under ether anaesthesia, control and unilaterally inflamed rats were implanted subcutaneously on the flank with a single osmotic mini-pump (Model 2ML1, Alzet, Palo Alto, CA). The incision was closed with sterile wound clips. The pump contained either sterile water or a solution of naloxone in sterile water at a concentration equivalent to a dose of 0.16, 0.50 or 3.0 mg/kg/h. These doses were based on the weight of the rats on the day of implantation. Pumps were implanted between 2.30 and 3.30 pm: this allowed for a 4 hr wind-up time for pumps to attain their maximal rate of delivery prior to onset of the dark phase. Immediately prior to the dark phase, rats were again provided with food and water.

Plethysmography The inflammation of the paw was evaluated employing a plethysmometer (Ugo Basile, Verona, Italy). The paw was submerged to the tibio-tarsal joint in the water-filled cell and the volume of displacement, which corresponds to the paw volume, read from a digital display. Ten days following implantation, it was no longer possible to immerse the inoculated paw in the cell owing to the extreme extent of the swelling. Thus, rats were killed, the hindpaws cut just above the tibio-tarsal joint and weighed.

Determination of nociceptive thresholds For evaluation of the withdrawal response to noxious pressure, a Randall and Sellito apparatus was used (Ugo Basile, Verona, Italy): this applies an incremental pressure by means of a blunt piston of surface area 1.75 mm 2.

Table I. Experimental design Day 0

1

Inoculation Pumps in

2

3

4

5

6

Behavioural testing and plethysmography

7 Pumps out

8

9

Behavioural testing

17 Plethysmography

543

Opioids and inflammatory pain Pressure was applied to two points of the tail, respectively 4 and 5 cm from the tip, and the mean pressure in g required to elicit withdrawal was determined. In the case of the paw, pressure was applied to three separate points on the dorsal surface and the mean value determined. There was a cut-off of 500 g for the tail and paw, respectively. The experimenter was blind as to the drug treatments. Tail pressure thresholds were employed for evaluation of the antinociceptive actions of the selective/t-opioid agonist, morphine, and the selective x-opioid agonist, U69,593. Three days following pump implantation, control and unilaterally inflamed rats were injected s.c. with either 2.5 mg/kg morphine or 5.0mg/kg U69,593. The response to pressure was evaluated immediately before and 30 min following injection. For each rat, the antinociceptive action of morphine and U69,593 was expressed as a percentage of preinjection values (defined as 100%). The response of both the left and right paws to pressure was evaluated for control and unilaterally inflamed rats two and six days following implantation of pumps, in addition to one and two days subsequent to their removal. In half the rats, the left paw was evaluated prior to the right paw to avoid any possible order effects.

Behavioural observations Body weight and food and water intake were monitored in all animals. All measurements were made just prior to the onset of the dark period. On days two and six postimplantation, core temperature was determined by insertion of a vaseline-lubricated digital thermistoprobe (Testoterm, Switzerland) into the rectum to a depth of 5 cm for 45 s. Measurements were performed on rats gently restrained under paper wadding between 10.00 and 11.00 a.m. on days two and six. On days two and six postimplantation, between 9.30 and 12.00 a.m., open field locomotor activity was evaluated over 15 rain by use of an automated Digiscan apparatus (Single Axis Digiscan Analyser, Omnitech Electronics, Maywood, N J) coupled to an IBM PC. The observation chamber was constructed of plexiglass and rectangular in shape: length 36 cm, width 22 cm and height 18 cm. Activity was monitored via interruption of intersecting light beams. A total of 12 parallel light beams were orientated across the width of the chamber at a height of 3 cm above the floor such that horizontal movement in a single horizontal axis could be evaluated. The "movement time" parameter was employed. This recorded the time (s) during which the animals were ambulating in a horizontal direction, stereotyped movements involving repetitive breaking of the same beam being excluded. When the animal was stationary for > I s, the parameter ceased to be incremented. The time devoted to grooming was determined by use of a one-zero sampling method. Every 15 s, the rats were inspected and presence of grooming scored as "one". The scores for each rat, equivalent to the number of grooming bouts, were summed over 15rain (i.e. 60 sampling periods). The number of fecal pellets was counted following observation. Core temperature was evaluated both immediately prior to and following observation.

Materials Drug doses are in terms of the base. All drugs were dissolved in sterile distilled water. In the case of U69,593, a few drops of lactic acid were added and the pH adjusted to 6.0 with NaOH. The drug salts and sources were as follows. Morphine sulphate (Cooperation Franqaise, Paris, France), U69,593 benzenacetamide (Upjohn Company, Kalamazoo, MI) and naloxone hydrochloride (Sigma, Chesnes, France).

Statistics Data were subjected to Multiple Analysis of Variance (MANOVA) for evaluation of influence of treatment

(control or inoculated), drug (vehicle or naloxone) and the interaction between these variables. In addition, one-way ANOVA followed by Newman-Keuls test was used as indicated below. Student's two-tailed t-test was used where appropriate and as indicated in the text. RESULTS

Antinociceptive action o f morphine and U69,593 Naloxone exerted no influence upon basal latencies of the tail to respond to noxious pressure in either control or unilaterally inflamed rats, between which there was no difference (not shown). The selective #-agonist, morphine, 4° elicited a pronounced and comparable antinociception in vehicle-treated control and unilaterally inflamed rats (Fig. 1): the pressure eliciting tail-withdrawal was 438.1 ___25.3 g and 4 6 4 . 2 + 12.2g, respectively (P > 0 . 0 5 in Student's two-tailed t-test). In both control and unilaterally inflamed rats, all doses of naloxone blocked the antinociceptive action of morphine (Fig. 1). The selective x-agonist, U69,593, 32 likewise evoked an antinociception in control and unilaterally inflamed rats: the pressure-evoking withdrawal was 434.2 + 22.8 and 4 8 2 . 0 _ 8.0 g, respectively (P > 0.05, in Student's two-tailed t-test). The action of U69,593 was dose-dependently antagonized b y naloxone in both control and unilaterally inflamed rats with 3.0 mg/kg/h abolishing its effect. Thus, in both control and unilaterally inflamed rats, 0.16 mg/kg/h selectively blocked the action of morphine, while 3.0 mg/kg/h blocked the action of both morphine and U69,593. The doses of 0.16 and 3.0 mg/kg, which are referred to as " l o w " and " h i g h " respectively, were selected for all further experiments.

Response o f the paws to noxious pressure Inoculation of the paw was associated with a reduction in the pressure eliciting withdrawal relative both to the contralateral, non-inoculated paw and the right paw of control animals (Fig. 2 and Table 2). Irrespective of treatment, the pressure threshold tended to progressively decrease, in line with previous studies and probably reflecting the fact that rats were tested on several occasions. 43'44 The high dose of naloxone did not significantly modify the thresholds to pressure of the non-inoculated paw of unilaterally inflamed rats, or of either paw of control animals, at any time of observation (Table 2). It did, however, potentiate the hyperalgesia of the inflamed paw to noxious pressure two days postimplantation (Fig. 2 and Table 2). By six days, this effect has disappeared. The day following pump-removal, unilaterally inflamed rats which had been perfused with this high dose showed a normalization of thresholds; that is, these were no longer significantly decreased relative to the non-inoculated contralateral paw or the right paw (Fig. 2 and Table 2). Two days after removal of pumps, this effect had disappeared. In contrast to the high dose of naloxone, at no time did the low

544

M. J. MILLAN and F. C. COLPAERT MORPHINE

1 ,.

0

250

]

200

f

U 69,593

i

CONTROL A INOCULATED @

150

~*"~7,**

IR

100

v~ o.~e o15 31o NALOXONE(mg/kg/hd Fig. 1. Influence of long-term naloxone perfusion upon the antinociceptive action of the #-agonist, morphine, and the x-agonist, U69,593, in control and unilaterally inflamed (inoculated) rats. Morphine and U69,593 were applied s.c. at 2.5 and 5.0 mg/kg, respectively, and the response of the tail to noxious pressure evaluated. Data are means+ S.E.M.; n = 5-9 per point. Asterisks represent significance of naloxone vs vehicle differences for control and inoculated rats: *P < 0.05 and **P < 0.01 (Newman-Keuls test). Results of MANOVA were as follows. Morphine: inoculation, F(1,43) = 2.8, P > 0.05; naloxone, F(3,40)= 12.5, P 0.05. U69,593: inoculation, F(1,43) = 0.2, P > 0.05; naloxone, F(3,40) = 29.4, P < 0.001 and interaction F(3,40) = 0.5, P > 0.05.

dose of naloxone modify nociceptive thresholds in either unilaterally inflamed or control rats (Fig. 2 and Table 2).

Body weight Inoculation reduced body weight gain over the following 24 h: change in body weight for control rats (n = 22) was + 3 . 1 _ 0.6 g and for inoculated rats (n = 22) was - 5 . 4 ___0.5 g (P < 0.001, Student's two-tailed t-test). Thereafter, vehicle-treated unilaterally inflamed animals showed little reduction in rate of weight gain relative to control rats across the week in which pumps were present (Figs 3 and 4). The high dose of naloxone significantly reduced weight gain only on the first day following implantation in control rats (Fig. 3): across the whole week, there was no overall significant effect (Fig. 4). In unilaterally inflamed rats, in contrast, high dose naloxone exerted a more persistent reduction in body weight gain (Fig. 3) which, across the whole week, was pronounced (Fig. 4). Further, upon removal of pumps delivering the high dose, there was a rebound increase in weight gain in unilaterally inflamed but not control animals (Fig. 3). In contrast to the high dose, the low dose of naloxone exerted only a transient reduction in weight gain which, in contrast to the high dose, did not differ between control and unilaterally inflamed rats. Following pump removal, the rebound increase in weight gain was comparable in control as compared to unilaterally inflamed rats (Fig. 3).

Ingestive behaviour and core temperature Unilaterally inflamed rats receiving vehicle showed a variable degree of hypophagia and hypodipsia

relative to control animals over time: this was most pronounced immediately following inoculation (Fig. 5). Ten days after removal of pumps, at which time inflammation had spread to the contralateral limb, the reduction in food intake and water intake was pronounced (Fig. 5). As regards core temperature, vehicle-treated unilaterally inflamed rats showed a mild and significant hyperthermia throughout the week in which pumps were present (not shown and see Fig. 7). As shown in Figs 6 and 7, the action of naloxone upon food intake and water intake was expressed dose-dependently, whereas for core temperature there was no difference between the low and high doses. The high dose of naloxone significantly reduced food intake and water intake in both control and unilaterally inflamed rats; however, its action in unilaterally inflamed animals was significantly enhanced and more sustained as compared to control rats (Fig. 6). High dose naloxone also significantly reduced water intake (Fig. 6) and resting core temperature (Fig. 7) but, in contrast to food intake, there was no difference between unilaterally inflamed and control animals as concerns its influence. In the case of core temperature, the only day upon which the action of naloxone was significant was on the first day following inoculation (Fig. 7): thereafter, no significant effect was seen in either control or unilaterally inflamed rats. The low dose of naloxone also reduced food intake in both control and unilaterally inflamed animals. However, its effect was less pronounced than that of the high dose and less sustained. Indeed, its action was only significant on the first day following implan-

Opioids and inflammatory pain tation: relative to vehicle-treated rats (100%), food intake was 85.9 + 4.6 and 72.5 + 6.0%, in control and unilaterally inflamed rats, respectively (no significant difference between these values). Indeed, on no individual day did the effect of low dose naloxone upon food intake differ between control and unilaterally inflamed rats (not shown). Over the whole week, its effect was slightly more pronounced in unilaterally inflamed animals (Fig. 6). Water intake was reduced by the low dose of naloxone throughout the week in which pumps were present. There was no difference between control and unilaterally inflamed rats as concerns its effect on individual days (not shown) or across the entire week (Fig. 6). Low dose naloxone reduced core temperature in both control and unilaterally inflamed rats, these groups not differing as regards the magnitude of its effect (Fig. 7).

545

+1 +1 +1

+1 +1 -H

m. ~. m.

•

d ~ . ~~v

"0

+1 +1 +1

+1 +l +1 - ~ + ~

Open-field behaviour Two days following implantation, vehicle-treated unilaterally inflamed rats did not differ significantly from control animals as concerns their locomotor activity and grooming behaviour (Fig. 8). Further, there was no difference between these groups as concerns the hyperthermia evoked by exposure to the open-field or the number of fecal pellets deposited (not shown). High dose naloxone significantly decreased locomotion only in unilaterally inflamed animals: on the contrary, only in control animals was grooming inhibited (Fig. 8). Neither the rise in core temperature nor the number of fecal pellets deposited was modified (not shown)• The low dose of naloxone, in distinction, exerted no significant effect upon any parameters in either control or unilaterally inflamed animals (Fig. 8 and not shown)• At six days, reflecting the greater degree o f inflammation than at day two, locomotion was significantly decreased in unilaterally inflamed as compared to control rats receiving vehicle (Table 3). There was no longer any significant effect of the high dose of naloxone. As at two days, the high dose of naloxone reduced grooming in control but not in unilaterally inflamed animals (Table 3). Hyperthermia and defecation did not differ significantly between unilaterally inflamed and control rats were unaffected by naloxone (Table 3).

o~

,q~V

+1 +1 +1

+1 +1 +1

+1 +1 +1

+1 +1 + t , , z o o ~. ~. ~. . ; . ~

+1 +1 +1

+1 +1 +1

0

0

~N o +1 +1 +1 ,d ~--i "

+1 +1 +1

+1 +t +1 o.~.

+1 +1 +1

o.

0

II

N

¢"1 ~

("I

I

Inflammation and contralateral transfer of disease The progression of paw volume in the inoculated (right) and non-inoculated (left) paws of unilaterally inflamed rats is shown in Fig. 7. Values for control rats are omitted for clarity. Within one day following inoculation, the right paw was inflamed and the inflammation progressed over the week within which pumps were present. In the presence of the high dose of naloxone, the volume of the non-inoculated paw was unaffected whereas the swelling of the inoculated paw was slightly, but significantly, reduced (Fig. 9). The low dose affected the volume of neither the inoculated nor non-inoculated paw (Fig. 9).

~0

N

~. o. ,q. +1 +1 +1 o.~.~

+1 +1 +1 o. o. ~.

I ~ i ¢-1 t~,l

¢ N (.-1 ( ~ i

g~ .,r'~

e-i VM N

546

M.J. MILLAN and F. C. COLPAERT CONTROL

100

i

v

i§ 50

m

VEH 0.111 3.0 ~H 0.18 3.0 ~ IN+2

IN+,

0.18 3.0 ~ OUT+I

1 I

ode 3.0 OUT+2

Fig. 2. Influence of long-term administration of naloxone upon the response to noxious pressure of the paws of unilaterally inflamed rats. The pressure at which the inoculated (right) paw was withdrawn is expressed relative to that for the left (non-inoculated) paw. Means ± S.E.M. are shown; n = 7-15 per value. Asterisks indicate significance of naloxone vs vehicle values: *P < 0.05, **P < 0.01 (Newman-Keuls test). One-way ANOVA. In + 2: F(2,37) = 3.7, P < 0.05; In + 6: F(2,37) = 0.6, P > 0.05; Out + 1: F(2,37) = 4.4, P < 0.05; Out + 2: F(2,37) =0.2, P > 0.05.

Ten days after the removal o f the pumps, i n f l a m m a t i o n h a d transferred to the contralateral (left) paw: this was reflected in an increase in its volume (Fig. 9). In rats which h a d been receiving the high dose of naloxone, this transfer was greatly reduced: this was a p p a r e n t from lack o f increase in its volume (Fig. 9). In contrast, in rats which h a d been receiving the low dose, there was n o modification of the transfer (Fig. 9). By 10 days, the volume of the inoculated paw h a d increased to such a n extent t h a t it n o longer fitted into the m e a s u r e m e n t cell. Thus, the rats were killed a n d the paws weighed. As s h o w n in Fig. 10, the weight o f the right paw was B0gI,~/EO

20

1o

I-

/ & *

a /S

10.

5

0

DA~'SRELATIVETO IMPLJd'~'ATIONOF PUMPS

nt

0 - - 0 VEIl & ~ & 0.16 II1~11 3.0

15-

i

i

~

,,:..

i

o~,1

INOCULKI'ED [

15 v

10

z

5.

!o t

-5

i

i

i

4-s

~

o~+1

DAYSRELATIVETO IMPLANTATIONOF PUMPS

Fig. 4. Influence of long-term administration of naloxone upon body weight gain in unilaterally inflamed (inoculated) and control rats. Values represent daily change in body weight. Means _+_S.E.M. are shown; n = 9, 8 and 5 for vehicle (VEH) and 0.16 and 3.0mg/kg/h naloxone, respectively. Asterisks indicate significance of naloxone vs vehicle values: *P 0.05. MANOVA for day 1 after removal: inoculation, F(1,43) = 0.9, P > 0.05; naloxone, F(2,43) = 11.2, P < 0.001 and interaction, F(2,43) = 0.3, P > 0.05. n o t significantly reduced in rats which h a d received the high dose o f n a l o x o n e whereas the weight of the left paw was greatly reduced. The low dose of n a l o x o n e affected the weight of neither the right n o r left paw (Fig. 10).

l

l

DISCUSSION

Influence of naloxone upon nociceptive thresholds 0.111

,3.0

VEIl

0.11

3.0

mCOXONE ( m l / k l / I , r )

Fig. 3. The influence of long-term naloxone administration upon the body weight gain of control and unilaterally inflamed (inoculated) rats. The total weight gain is for the week in which pumps were present. Means-I-S.E.M. are shown; n = 5 - 9 per column. Asterisks indicate significance of difference for naloxone vs vehicle values: *P < 0.01 (Newman-Keuls test). MANOVA: inoculation, F(1,43) = 5.3, P = 0.03; naloxone, F(2,43) = 8.0, P < 0.001 and interaction F(2,43) = 0.4, P > 0.05.

In aggreement with o u r previous observations in " n a i v e " rats, 4°'49'~ in n o n - i n o c u l a t e d animals a " l o w " ( 0 . 1 6 m g / k g / h ) dose of naloxone selectively blocked #-receptors whereas a " h i g h " (3.0 m g / k g / h ) dose simultaneously blocked b o t h # a n d x-receptors. Importantly, identical findings were o b t a i n e d in unilaterally inflamed rats (Fig. 1), a finding consistent with the lack of change in the a p p a r e n t affinity of multiple opioid receptors for opioid ligands in animals with inflammation. 2°'~'47

Opioids and inflammatory pain

547

CONTROL

0--0

g

FOOD

80.

CONTROL

30.5

& ~ & WAI"I~ IN0CULAT(D

o--orooo

+

i ..0

~

37.o

0.18

3.0

VE].I 0.18

3.0

mU.OXONE( , ~ d ~ r ) -'I

;

2

PRE

3

4':'6

7

i

DAYSPOST-IMPLN,ffATION

1"0 OUT

Fig. 5. The influence of inoculation in the fight hindpaw with Mycobacteriumbutyricum upon food and water intake. All rats were implanted with pumps containing vehicle. Means + S.E.M. are shown; n = 9 for control and for inoculated rats. Asterisks indicate significance of differences between inoculated and control animals: *P < 0.05 and **P < 0.0l (Newman-Keuls test).

Fig. 7. The influence of naloxone treatment upon the core temperature of control and unilaterally inflamed rats. Core temperature was recorded one day following implantation of pumps. Means + S.E.M. are shown; n = 5-9 per column. Cross indicates significance of inoculated (VEH) vs control (VEH) values and asterisks indicate significance of naloxone vs VEH values: + , **P < 0.01 (Newman-Keuls test). MANOVA: inoculation, F(1,42)= 5.8 P = 0.02; naloxone, F(2,43) = 13.8, P < 0.001 and interaction F(2,43)= 0.05, P > 0.05.

FOOD INTAKE

U

WATER INTAKE

100

e~e A~A

CONTROL INOCULATED

90' A

80, IR

A**

70,

o.'le

s:o

0.'~6

s:o

Iou.oxo.c (mel~/hr)

'DoI ao

A 1

60 K

50

IA* *

i

i

i 4-6 "~ DAYS IOST-IMPLANTATION OF"PUMPS

Fig. 6. The influence o f long-term naloxone treatment upon food and water intake of control and unilaterally inflamed (inoculated) rats.) Means + S.E.M.; n = 5-9 per point. Asterisks indicate significance o f differences between naloxone and vehicle-treated values: *P < 0.05, **P < 0.01 (Newman-Keuls test). Upper panel: Dose-response relationship of naloxone effect across the week in which the pumps were present. Intake is expressed as a percentage o f values for rats receiving vehicle. MANOVA. Food intake: inoculation, F(1,43) = 106.8, P < 0.001; naloxone, F(2,43) = 108.4, P < 0.001 and interaction, F(2,43) = 9.8, P < 0.001. Water intake: inoculation, F(1,43) = 13.1, P < 0.001; naloxone, F(2,43) = 105.1, P < 0.001 and interaction F(2,43) = 2.6, P > 0.05. Lower panel: time-course o f effect o f high dose naloxone.

548

M.J. MILLANand F. C. COLPAERT

a) LOCOMOTION INOCULMIm

50

i"

I0

VEH

0.111

3.0

VEIl

0.1Q

3.0

b) GROOMING 10.

2_ 8.

_I.

pain is consistent with antinociceptive properties of exogenously administered x-agonists both in normal and in unilateral inflamed rats. 19'4~ Further, in unilateral inflammation and polyarthritis, there is an activation of spinally-localized neurons containing dynorphin, which is thought to exert its antinociceptive actions via x-opioid receptors. 4~ Third, by six days postimplantation, there was no longer any effect of naloxone infusion. This observation is in contrast to the effects of acute administration of MR 2266 which does elicit hyperalgesia at this time-point.~ This finding suggests that the loss of efficiency of long-term naloxone treatment does not reflect the disengagement of x-opioid mechanisims controlling nociception. Rather, there may be a degree of redundancy such that a non-opioid system compensates for the sustained inactivation of r-receptors. Possibly, the rebound antinociception seen upon termination of perfusion may be attributed to an unmasking of such (a) compensatory system(s). Additionally, previous studies in naive rats suggest that it may reflect an interaction of dynorphin with x-receptors rendered supersensitive by long-term treatment with naloxone.49,54, 74

VEH

0.1 §

3.0

VI[H

O.t |

3.0

NALOXONE(mg/kg/hr)

Fig. 8. The influenceof long-term naloxone treatment upon horizontal locomotor activity and grooming behaviour of control and unilaterally inflamed (inoculated) rats. Rats were observed for 15rain at two days following pump implantation in an unfamiliar Digiscan apparatus. Mean + S.E.M. are shown; n = 5-9 per column. Asterisks indicate significance of naloxone vs vehicle difference: **P 0.05; naloxone, F(2,43)= 3.6, P = 0.04 and interaction, F(2,43) = 0.7, P > 0.05. Grooming: inoculation, F(1,43) = 1.7, P > 0.05; naloxone, F(2,43) = 7.1, P = 0.002 and interaction, F(2,43) = 1.0, P > 0.05. Inoculated rats showed a clear reduction in thresholds to noxious pressure to the inflamed tissue: the potentiation of this hyperalgesia by the high but not low dose of naloxone raises several interesting points. First, this hyperalgesic action was seen at two days postimplantation and cannot reflect an aggravation of the inflammation per se, since, at this time-point, inflammation was slightly inhibited (Fig. 9). Second, the hyperalgesic effect of naloxone was selective to inflamed tissue. This finding parallels the effect of acute injections of MR 2266, an opioid antagonist with a weak preference for x- over /~-receptors "45 or doses of naloxone (probably) high enough to antagonize x-receptors. 2°'27'3s" Importantly, it is shown here that x-receptors blockade is required for the induction of hyperalgesia, p-receptor antagonism alone being ineffective. A role of x-receptors in the modulation of nociception under inflammatory

Thus, these data support a role of x-receptors in the control of nociception under inflammatory pain. However, they suggest that this action may not be indispensible, at least as regards the response of the inflamed tissue to acute imposition of noxious stimuli. Influence o f naloxone upon. behaviour

That blockade of #-receptors alone reduced food intake, water intake and body weight is in line with the effects of the selective #-antagonist, fl-funaltrexamine, which similarly suppresses food intake and water intake in rats. 22'68 However, the more pronounced influence of x-receptors blockade seen here is consistent with, first, the powerful stimulatory influence of x-agonists on food intake52'69 and, second, inhibition of food intake and water intake by either the x-agonist, nor-binaltorphimine or antibodies raised against dynorphin.6-8 In contrast to food intake and water intake, the low dose of naloxone was as effective as the high dose in reducing resting core temperature (Fig. 7): this suggests that ~t- rather than x-receptors may be involved in its control. This interpretation is consistent with the hyperthermic actions of selective #- but not K(or 6-) opioid agonists in rats. 64 Nevertheless, this study provides no support for an essential role of #- (or other) opioid receptors in the mediation of the hyperthermic response to novelty stress (Fig. 8); see Refs 42, 48, 50 and 67 for further discussion. Previous work indicates that naloxone more consistently inhibits exploration than horizontal locomotion per se 15a.30,42and, in line with this argument, naloxone failed to modify locomotor activity of control rats in the open-field (Fig. 8). Finally, acute administration of opioid antagonists depresses grooming

Opioids and inflammatory pain

549

Table 3. The influence of long-term naloxone treatment upon the open-field activity, core temperature and defecation of control and unilaterally inflamed (inoculated) rat Control VEH Movement time Grooming AT (°C) Fecal pellets

Inoculated

0.16

34.2+__2.9 36.2+5.8 9.4 + 0.6 7.2 + 0.8 1.5+__0.1 1.4+0.3 2.7 -I- 0.9 1.0 __+0.5

3.0

VEH

0.16

3.0

24.3+3.5 4.6 + 0.7** 1.7+__0.1 1.8 ___1.2

18.1+3.4"* 7.6 ___0.9 1.4+0.1 1.6 ___0.7

26.0+3.1 9.2 + 0.6 1.2+0.2 0.4 ___0.3

16.0+4.3 6.5 + 0.5 1.3+0.1 1.2 ___0.6

Animals were observed six days following implantation of pumps. VEH signifies vehicle, and 0.16 and 3.0 the dose of naloxone in mg/kg/h. AT indicates the difference in core temperature before as compared to after exposure. Mean + S.E.M. is given: n = 5-9 per value. Asterisks indicate significance of (1) naloxone vs vehicle values and (2) inflamed-VEH vs control-VEH values: **P < 0.01 (Newman-Keuls test). MANOVA was as follows. Movement time: inoculation, F(1,43) = 11.6, P < 0.001; naloxone, F(2,43) = 3.6, P = 0.04 and interaction, F(2,43)=0.6, P >0.05. Grooming: inoculation, F(1,43)=0.4, P >0.05; naloxone, F(2,43) = 10.7, P 0.5 and interaction, F(2,43) = 0.6, P > 0.05. of a functional role of x- rather than p-receptors in the response to long-term inflammatory pain,

Inflammation and transfer of arthritic disease In a recent study, we observed that in rats implanted 24 h previously with pumps yielding 3.0 mg/kg/h of naloxone, the inflammation induced by inoculation of the paw was inhibited.43 Similarly, this dose inhibited the swelling of the inoculated paw upon implantation of pumps subsequent to inoculation (Fig. 9), indicating that naloxone can partially inhibit (reverse) inflammatory oedema. Evidently, opioids differentially modify the swelling as compared to the hyperaigesia of inflammation (see above). Inflammatory mechanisms are known to be influenced by opioids and dynorphin induces plasma extravasation in the rat. 2'~°'74Since functional r-opioid receptors exist in the inflamed paw 66 it is conceivable that naloxone interferes with a pro-

inflammatory action of dynorphin, thereby reducing the oedema. In contrast to the primary, local inflammation, the spread of pathology in unilaterally inflamed rats reflects mechanisms involved in the development of (rheumatoid) arthritis. As such, (1) autoimmune processes, (2) the hypothalamo-adrenocorticotropin axis and the sympathetic nervous system and (3) CNS mechanisms are implicated. 16'3337,39,61 Each of these mutually interactive systems are subject to modulation by opioid mechanisms (including x-receptors) and could be involved in the inhibition by high dose naloxone of transfer of disease to the contralateral limb (Figs 9 and 10). (1) The simplest interpretation that opioids contribute to the activation of immune mechanisms responsible for evolution of arthritic disease would be consistent with findings of an in vitro enhancement of immunocompetent cell activity by endorphins (including dynorphin); however, there are also data showing immunosuppressant actions of opioids, and their modulation of immune function is exceedingly complex. 9'13,2s'Ss'6~Further, it is difficult to specify with confidence potential mechanisms since the role(s) of individual opioid receptor types remain unelucidated and very few in vivo data are available. (2) Since, upon acute administration, naloxone enhances the activity of the hypothalamoadrenocortical axis, 23,46 long-term treatment might be associated with a sustained facilitation of corticosterone secretion which could secondarily influence, via immune mechanisms, arthritic disease. Further, the genesis of experimental arthritis in the rat is dependent upon an intact sympathetic transmission.35'36As catecholamine secretion into the circulation is powerfully modulated by opioids, 24'29'46'56'57 naloxone treatment may inhibit transfer of pathology via modulation of sympathetic activity. (3) Finally, as regards CNS mechanisms, repeated i.c.v, administration of morphine inhibited experimental arthritis by, it was suggested, reducing central sympathetic outflow. 36 In addition, substance P in primary afferents contributes to mechanisms of inflammation and arthritic disease33,39 and morphine presynaptically diminishes substance P release in rats. 31'72'73 These data imply that p-opioid agonists might inhibit arthritic processes. However, since blockage of p-receptors did not alter symptomology in the present study, there is no evidence that endogenous p-opioids may fulfil such a physiological role. In fact, ~c-agonists fail to modulate peripheral release of substance p,59 but such studies have not as yet been performed in inflamed tissue in which x-receptors are functionally more active.66 Further, a spinally-localized mechanism is involved in the contralateral spread of inflammation and hyperalgesia between limbs 11'15'25'34and it is the spinal cord which is the locus of the pronounced effects upon the functional activity of dynorphin and x-receptors (see above) seen in unilateral inflammation.

Opioids and inflammatory pain Irrespective of the mechanisms involved, since pathology on the contralateral paw is only apparent 10 days after discontinuation of naloxone treatment, opioid receptors are implicated in early events determining disease evolution; this is analogous to, for example, adrenoceptor-mediated mechanisms.35 Whether likewise, in analogy to sympathetic blockade, naloxone treatment attenuates the progression of disease after it has been established remains to be seen. Further, whether naloxone treatment postpones or permanently prevents transfer over longer time-periods requires clarification. CONCLUSIONS

The present data show that the functional effects of sustained opioid receptor antagonism are modified under conditions of long-term inflammatory pain, as regards their influence not only upon nociceptive thresholds but also body weight, appetite and motility; these are variables highly responsive to, and characteristic of, long-term pain states. Mu-receptors appear to be of comparatively little importance and the effects observed here may relate primarily to x-receptors. However, it is essential to introduce two caveats. First, 6-receptors are antagonized by the high dose of naloxone.49'~ Delta-sites do not appear to play an important role in the modulation of ingestive behaviour, body weight, core temperature or open-field behaviour, nor are they involved in the

551

control of nociceptive thresholds or inflammation in unilaterally inflamed rats. 4,~,45,~ However, it cannot be excluded that their blockade contributes to the effects of naloxone. Further, that combined blockade of/t-, x- and 6-receptors is required for the effects observed also cannot be discounted. Second, it is possible that the alterations in the influence of opioid receptor blockade under unilateral inflammation reflect changes in opioid mechanisms for the control of appetite, activity and core temperature per se. These may themselves be affected under inflammation. However, differing pools of (x)-opioid receptors control the various parameters affected. Thus, the most parsimonious explanation is that the primary effect of naloxone is to enhance hyperalgesia by blocking x-receptors and that this action results in a secondary modification of other variables such as body weight, etc. Finally, the data show that a sustained antagonism of (non-#), x- and/or ~-opioid receptors inhibits the contralateral transfer of pathology in this model of experimental arthritis. These data suggest a pathophysiological role of opioid systems in the modulation of immune or other processes determining the evolution of arthritic disease. Le Marouille is thanked for excellent technical work. S. Demoulin, K. Chadwick, S. Carrive and V. Green are thanked for preparation of the manuscript. Acknowledgements--S.

REFERENCES

1. Aloyo V. J., Spruijt B., Zwiers H. and Gispen W. H. (1983) Peptide-induced excessive grooming in the rat: the role of opiate receptors. Peptides 4, 833-836. 2. Bartho L. and Szolcsanyi J. (1981) Opiate agonists inhibit neurogenic plasma extravation in the rat. Eur. J. Pharmac. 73, 101-104. 3. Basbaum A. I., Cruz L. and Weber E. (1986) Immunoreactive dynorphin in sacral primary afferent fibres of the cat. J. Neurosci. 6, 127-133. 4. Benton D., Dalrymple-Alford J. C., McAllister K. H., Brain P. F. and Brain S. (1984) Comparison in the mouse of the effect of the opiate delta receptor antagonists ICI 154129 and naloxone in tests of extinction, passive avoidance and food intake. Psychopharmacology 82, 41-45. 5. Besson J.-M. and Chaouch A. (1987) Peripheral and spinal mechanisms of nociception. Physiol. Rev. 67, 67-186. 6. Calcagnetti D. J., Calcagnetti R. L. and Fanselow M. S, (1990) Centrally administered opioid antagonists, nor-binaltorphimine, 16-methylcyprenorphine and MR 2266, suppress intake of a sweet solution. Pharmac. Biochem. Behav. 35, 69-73. 7. Carr K. D. and Bak T. H. (1990) Rostral and caudal ventricular infusion of antibodies to dynorphin A (1-17) and dynorphin (1-8): effects on electrically-elicitedfeeding in the rat. Brain Res. 5117, 289-294. 8. Carr K. D., Bak T. H., Simon E. J. and Portoghese P. S. (1989) Effects of the selective kappa opioid antagonist, nor-binaltorphimine, on electrically-elicitedfeeding in the rats. Life Sci. 45, 1787-1792. 9. Carr R. J. J., DeCosta B. R., Jacobson A. E., Rice K. C. and Blalock J. E. (1990) Corticotropin-releasing hormone augments natural killer cell activity through a naloxone-sensitive pathway. J. Neuroimmun. 28, 53-61. 10. Chahl L. ~. and Chahl J. S. (1986) Plasma extravasation induced by dynorphin-(1-13) in rat skin. Eur. J. Pharmac. 124, 343-347. 11. Coderre T. J. and Melzack R. (1987) Cutaneous hyperalgesia: contributions of the peripheral and central nervous systems to the increase in pain sensitivity after injury. Brain Res. 4tl4, 95-106. 12. Colpaert F. C. (1987) Evidence that adjuvant arthritis in the rat is associated with chronic pain. Pain 211, 201-222. 13. Dantzer R. and Kelley K. W. (1989) Stress and immunity: an integrated view of relationships between the brain and the immune system. Life Sci. 44, 1995-2008. 14. Dardick S. J., Basbaum A. I. and Levine J. D. (1986) The contribution of pain to disability in experimentally induced arthritis. Arth. Rheum. 29, 1917-1922. 15. Denko C. W. and Petricevic M. (1978) Sympathetic or reflex swelling due to crystal-induced inflammation in the opposite foot. Inflammation 3, 81-86. 15a. File S. E. (1980) Naloxone reduces social and exploratory activity in the rat. Psychopharmacology 71, 41-44. 16. Fitzgerald M. (1989) Arthritis and the nervous system. Trends Neurosci. 12, 86-87.

552

M.J. MILLAN and F. C. COLPAERT

17. Green E. J., Isaacson R. L., Dunn A. J. and Lanthorn T. H. (1979) Naloxone and haloperidol reduce grooming occurring as an after effect of novelty. Behav. Neural Biol. 27, 546-551. 18. Hollt V., Haarmann I., Millan M. J. and Herz A. (1987) Prodynorphin gene expression is enhanced in the spinal cord of chronic arthritic rats. Neurosci. Lett. 73, 90-94. 19. HorweU D. C. (1988) Kappa opioid analgesics. Drugs Future 13, 1061 1071. 20. Iadorola M. J., Brady L. S., Draisci G. and Dubner R. (1988) Enhancement of dynorphin gene expression in spinal cord following experimental inflammation: stimulus specificity, behavioural parameters and opioid receptor binding. Pain 35, 313-326. 21. Iadorola M. J., Douglass J., Civelli O. and Naranjo J. R. (1988) Differential activation of spinal cord dynorphin and enkephalin neurones during hyperalgesia: evidence using cDNA hybridization. Brain Res. 455, 205-212. 22. Islam A. K. and Bodnar R. J. (1990) Selective opioid antagonists effects upon intake of a high-fat diet in rats. Brain Res. 508, 293-296. 23. Iyengar S., Kim H. S. and Wood P. L. (1986) Kappa opiate agonists modulate the hypothalamic-pituitaryadrenocortical axis in the rat. J. Pharmac. exp. Ther. 238, 429-436. 24. Jackson A. and Cooper S. J. (1986) An observational analysis of the effect of the selective kappa opioid agonist, U-50,488H on feeding and related behaviours in the rat. Psychopharmacology 90, 217-221. 25. Kayser V., Basbaum A. and Guilbaud G. (1990) Deafferentiation in the rat increases mechanical nociceptive threshold in the innervated limbs. Brain Res. 508, 329-332. 26. Kayser V. and Guilbaud G. (1990) Differential effects of various doses of morphine and naloxone on two nociceptive test thresholds in arthritic and normal tests. Pain 41, 353-363. 27. Kayser V. and Guilbaud G. (1981) Dose-dependent analgesic and hyperalgesic effects of systemic naloxone in arthritic rats. Brain Res. 226, 344-348. 28. Khansari D. N., Murgo A. J. and Faith R. E. (1990) Effects of stress on the immune system. Immun. Today 11, 170-175. 29. Kiritsky-Roy J. A., Appel N. M., Bobbit F. G. and van Loon G. R. (1986) Effects of mu-opioid receptor stimulation in the hypothalamic paraventricular nucleus on basal and stress-induced catecholamine secretion and cardiovascular response. J. Pharmac. exp. Ther. 239, 814-822. 30. Koek W. and Stangen J. L. (1984) Acute effects of naloxone and naltrexone, but lack of delayed effects on exploratory behaviour in the rat. Psychopharmacology 84, 383-387. 31. Kuraishi Y., Hirota N., Sugimoto M., Satoh M. and Takagi H. (1983) Effects of morphine on noxious stimuli-induced release of substance P from rabbit dorsal horn in vivo. Life Sci. 33, Suppl. 1, 693-696. 32. Lahti R. A., Mickelson M. M., McCall J. M. and von Voigtlander P. (1985) 3[H]U69,593: a highly selective ligand of the opioid x-receptors. Eur. J. Pharmac. 109, 181-184. 33. Levine J. D., Clark R., Devor M., Helms C., Moskowitz M. A. and Basbaum A. (1984) Intraneuronal substance P contributes to the severity of experimental arthritis. Science 226, 547-549. 34. Levine J. D., Dardick S. J., Basbaum A. I. and Scipio E. (1985) Reflex neurogenic inflammation: I. Contribution of the peripheral nervous system to spatially remote inflammatory responses that follow injury. J. Neurosci. 5, 1380-1386. 35. Levine J. D., Dardick S. J., Roizan M. F., Helms C. and Basbaum A. T. (1988) a2-Adrenergic mechanisms in experimental arthritis. Proc. natn. Acad. Sci. U.S.A. 85, 4553-4556. 36. Levine J. D., Dardick S. J., Roizan M. F., Helms C. and Basbaum A. I. (1986) Contribution of sensory afferents and sympathetic efferents to joint injury in experimental arthritis. J. Neurosci. 6, 3423-3429. 37. Limberger N. (1988) Interaction between presynaptic opioid K-receptors, ~t2-adrenoceptors and purine Pt-receptors at noradrenergic axon terminals. In Regulatory Roles of Opioid Peptides (eds Illes P. and Farsarg C.), pp. 282-296. VCM Press, Weinheim, F.R.G. 38. Lombard M. C. and Besson J.-M. (1989) Electrophysiological evidence for a tonic activity of the spinal cord intrinsic opioid systems in a chronic pain model. Brain Res. 477, 48-56. 39. Lotz M., Carson D. A. and Vaughan J. H. (1987) Substance P activation of rheumatoid synviocytes: neural pathway in pathogenesis of arthritis. Science 235, 893-895. 40. Magnan J., Paterson S. J., Tavani A. and Kosterlitz H. W. (1982) The binding spectrum of narcotic analgesic drugs with different agonist and antagonist properties. Naunyn-Schmiedeberg's Arch. Pharmac. 319, 2197-2205. 41. Millan M. J. (1990) Kappa-opioid receptors and analgesia, Trends Pharmac. 11, 70-76. 42. Millan M. J. (1981) Stress and endogenous peptides: a review. Mod. Probl. Pharmacopsychiat. 17, 49-67. 43. Millan M. J. and Colpaert F. C. (1990) The influence of sustained opioid receptor blockade in a model of long-term, localized inflammatory pain in the rat. Neurosci. Lett. 113, 50-55. 44. Millan M. J., Czlonkowski A., Morris B., Stein C., Arendt R., Huber A., Hollt V. and Herz A. (1988) Inflammation of the hind limb as a model of unilateral, localized pain: influence on multiple opioid systems in the spinal cord of the rat. Pain 35, 299-312. 45. Millan M. J., Czlonkowski A., Pilcher C. W. T., Almeida O. F. X., MiUan M. H., Colpaert F. C. and Herz A. (1987) A model of chronic pain in the rat: functional correlates of alterations in the activity of opioid systems. J. Neurosci. 7, 77-87. 46. Millan M. J. and Herz A. (1985) The endocrinology of the opioids. Int. Rev. Neurobiol. 26, 1-83. 47. Millan M. J., Millan M. H., Czlonkowski A., Hollt V., Pilcher C. W. T., Colpaert F. C. and Herz A. (1986) A model of chronic pain in the rat: response of multiple opioid systems to adjuvant-induced arthritis. J. Neurosci. 6, 899-916. 48. Millan M. J. and Morris B. J. (1988) Long-term blockade of/l-opioid receptors suggests a role in control of ingestive behaviour, body weight and core temperature. Brain Res. 450, 247-258. 49. Millan M. J., Morris B. J. and Herz A. (1988) Antagonist-induced opioid up-regulation. I. Characterization of supersensitivity to selective/t- and x-agonists. J. Pharmac. exp. Ther. 247, 721-728. 50. Millan M. J., Przewlocki R., Jerlicz M., Gramsch Ch., Hollt V. and Herz A. (1981) Stress-induced release of brain and pituitary fl-endorphin: major role of endorphins in generation of hyperthermia, not analgesia. Brain Res. 208, 325-338. 51. Miller K. E. and Seybold V. S. (1987) Comparison of met-enkephalin-dynorphin A- and neurotensin-immunoreactive neurones in the cat and rat spinal cords. I. Lumbar cord. J. comp. Neurol. 255, 293-304. 52. Morley J. E., Bartness T. J., Gosnell B. A. and Levine A. S. (1985) Peptidergic regulation of feeding. Int. Rev. Biol. 27, 207398.

Opioids and inflammatory pain

553

53. Morris B. J. and Herz A. (1987) Distinct distribution of opioid receptor types in rat lumbar spinal cord. Naunyn-Schiedeberg 's Arch. Pharmac. 336, 240-243. 54. Morris B. J., Millan M. J. and Herz A. (1988) Antagonist induced opioid up-regulation. II. Regionally specific modulation of/z-, 6- and r-binding sites revealed by quantitative autoradiography. J. Pharmac. exp. Ther. 247, 729-736. 55. Nahin R. L., Hylden J. L. K., Iadorola M. J. and Dubner R. (1989) Peripheral inflammation is associated with increased dynorphin immunoreactivity in both projection and local circuit neurons in the superficial dorsal horn of the rat lumbar spinal cord. Neurosci. Lett. 96, 247-252. 56. Pfeiffer A., Feuerstein G., Faden A. and Kopin I. J. (1982) Evidence of involvement of #, 3- and r-opiate receptors in sympathetically and parasympathetically mediated cardiovascular responses to opiates upon anterior hypothalamic injection. Life Sci. 31, 1279-1282. 57. Pfeiffer A., Feuerstein G., Zerbe R. L., Faden A. L. and Kopin L. (1983) Receptors mediate opioid cardiovascular effects at anterior hypothalamic sites through sympatho-adrenomedullary and parasympathetic pathways. Endocrinology 113, 929-938. 58. PlotnikoffN. P., Faith R. E., Murgo A. J. and Good F. A. (eds) (1987) Enkephalins andEndorphins, Stress andlmmune System. Plenum Press, New York. 59. Pohl M., Mauborgue A., Bourgoin S., Benoliel J. J., Hamon M. and Cesselin F. (1989) Neonatal capsaicin treatment abolishes substance P release from rat spinal cord slices. Neurosci. Lett. 96, 102-107. 60. Ruda M. A., Iadorola M. J., Cohen L. V. and Young W. S. (1988) In situ hybridization histochemistry and immunocytochemistry reveal an increase in spinal dynorphin biosynthesis in a rat model of peripheral inflammation and hyperalgesia. Proc. natn. Acad. Sci. U.S.A. 85, 622-626. 61. Schoenfeld Y. and Isenberg D. A. (1988) Mycobacteria and immunity. Immun. Today 9, 178-182. 62. Shavit Y., Lewis J. W., Terman G. W., Gale R. P. and Leibeskind J. C. (1984) Opioid peptides mediate the suppressive effect of stress on natural killer cell cytotoxicity. Science 223, 188-190. 63. Shippenberg T. S., Stein C., Huber A., Millan M. J. and Herz A. (1988) Motivational effects of opioids in an animal model of prolonged inflammatory pain: alteration in the effects of r- but not #-receptor agonists. Pain 35, 179-186. 64. Spencer R. L., Hruby V. J. and Burks T. F. (1990) Alteration of thermoregulatory set point with opioid agonists. J. Pharmac. exp. Ther. 252, 696-705. 65. Stefano G. B. (1989) Role of opioid neuropeptides in immunoregulation. Prog. NeurobioL 33, 149-159. 66. Stein C., Millan M. J., Shippenberg T. S., Peter K. and Herz A. (1989) Peripheral opioid receptors mediating antinociception in inflammation: evidence for involvement of #, 6 and r-receptors. J. Pharmac. exp. Ther. 248, 1269-1275. 67. Stewart J. and Eikelboom R. (1979) Stress masks the hypothermic effect of naloxone in rats. Life. Sci. 25, 1165-1172. 68. Ukai M. and Holtzman S. G. (1988) Effects of fl-funaltrexamine on ingestive behaviours in the rat. Eur. J. Pharmac. 153, 161-165. 69. Ukai M. and Holtzman S. G. (1988) Effects of intrahypothalamic administration of opioid peptides selective for /~-, K- and f-receptors on different schedules of water intake in the rat. Brain Res. 459, 275-281. 70. Wei E. T., Kiang J. G., Buchan P, and Smith T. W. (1986) Corticotropin-releasing factor inhibits neurogenic plasma extravasation in the rat paw. J. Pharmac. exp. Ther. 238, 783-784. 71. Weihe E., Millan M. J., Hollt V., Nohr D. and Herz A. (1989) A model of chronic pain in the rat: induction of the gene encoding pro-dynorphin by arthritis results in a lighting-up of dynorphin neurons in lumbosacral spinal cord. Neuroscience 31, 77-95. 72. Yaksh T. L. (1988) Substance P release from knee joint afferent terminals: modulation by opioids. Brain Res. 458, 319-324. 73. Yaksh T. L., Jessel T. M., Gamse R., Mudge A. W. and Leeman S. E. (1980) Intrathecal morphine inhibits substance P release from mammalian spinal cord in vivo. Nature 286, 155-157. 74. Zukin R. S. and Tempel A. (1986) Neurochemical correlates of opiate receptor up-regulation. Biochem. Pharmac. 35, 1623-I 627. (Accepted 28 November 1990)