Pain, 57 (1994) 383-391 0 1994 Elsevier Science B.V. All rights reserved 0304-3959/94/$07.00

383

PAIN 2518

Antinociception and behavioral manifestations induced by intracerebroventricular or intra-amygdaloid administration

of cholinergic agonists in the rat

Marina A. Oliveira and Wiliam A. Prado * ~e~a~trn~nt of Pha~~~o~o~, Fact&y of Midline of RibeirGo Preto, ~n~ve~i~ of Go Pauio, Ribeinio Preto SP (Brazil)

(Received 24 March 1993, revision received 8 October 1993, accepted 19 November 1993)

Summa~ Changes in the tail-flick latency (TFL) to noxious heat st~ulation and behaviors changes produced by intracerebroventricular (i.c.v.1 or intra-amygdala administration of cholinergic agonists were studied in the rat. A significant increase in the TFL and behavioral changes were produced by carbachol (CCh, 2.2-8.8 nmol) injected into the dorsomedial portion (LVm) and inferior horn of the lateral ventricle (LVi), the effects being more prominent following injection into the LVi. Atropine (0.7 nmol), but not mecamylamine (5 nmol), fully inhibited the effects of CCh injected into the LVi. Bethanechol (4.4 nmoi) and oxotremorine (1.1-5.5 nmol), but not dimeth- ylphenyl-piperazinium (DMPP, 4.4 nmol), also increased the TFL following administration into the LVi. These cholinergic agonists were generally all less effective than CCh in eliciting behavioral changes. These results are indicative that muscarinic mechanisms of structures in the immediate vicinity of the LVi may be involved in cholinergic antinociception. When microinjected into the medial, central, basolateral, and posterior lateral nuclei of the amygdala complex (AC), both CCh and oxotremorine produced a significant increase in the TFL, but in no case was the effect stronger than that produced by stimulation of the medial nucleus. When microinjected into the same nuclei of the AC, CCh, but not oxotremorine, produced behavioral changes which were Iess frequent after stimulation of the medial nucleus. The behavioral changes, but not the antinociception, produced by CCh microinjected into the medial nucleus were inhibited by diazepam (1 mg/kg, i.p.1. These results are indicative that antinociception and behavioral changes evoked by CCh injected into the AC depend on drug action on different amygdala structures. They also suggests that antinociception from the medial nucleus is not secondary to the aversive reactions evoked by CCh and depends on the activation of mechanisms different from those required for the production of behavioral changes.

Key words: Antinociception; Cholinergic antinociception; Carbachol; Muscarinic agonist; Diazepam

Introduction

At many sites in the brain, electrical stimulation causes antinociception by activating centrifugal path- ways that act to inhibit sensory neurones in the spinal cord (see Willis 1982). This pain in~bito~ system seems to play a role in modulating affective/defensive

* Cor~s~~~~g ~u~~~: W.A. Prado, Department of Pha~acoIo~, Faculty of Medicine of Ribeirgo Preto, Ribeirgo Preto, SP, Brazil. Tel.: 16-633-3035, ext. 246; FAX: 16-633-1586.

behaviors (Watkins and Mayer 1986) since it can be activated also by a number of environmental stimuli (Mayer and Liebeskind 1974; Terman et al. 1984).

Stimulation of some parts of the pain inhibitory system, such as the periaqueductal gray (PAG) matter and nucleus raphe magmrs (NRM), evokes escape and defensive reactions (Oleson et al. 1980; Fardin et al. 1984a,b), thus leading to the notion that antinocicep- tion from these structures may be secondary to the stressfulness of brain stimulation (Prado and Roberts 1985). At Ieast for PAG, however, aversive reactions and antinociception produced by electrical stimulation can be dissociated by diazepam (Morgan et al. 1987).

SSDI 0304-3959(93)E0234-Q

384

The potential involvement of endogenous opioids and monoamines in the modulation of the pain in- hibitory system has been extensively demonstrated (see Besson and Chaouch 1987). However, there is increas- ing evidence for the critical role of cholinergic mecha- nisms in pain control and in the nociceptive suppres- sion caused by some environmental stimuli (see Green and Kitchen 1986). Strong antinociception accompa- nied by escape and defensive reactions has been shown to follow the administration of carbachol (CCh) into the dorsomedial portion of the lateral ventricle (LVm) of rats (Metys et al. 19691. The rapid onset of these effects of CCh is thought to indicate that periventricu- lar structures are involved in cholinergic antinocicep- tion (Metys et al. 1969). On the other hand, it is possible that CCh-induced antinociception is secondary to the activation of the pain inhibitory system involved in a general defensive reaction (Duggan 1984).

Based on such findings, the first objective of this study was to compare the antinociceptive and aversive effects of CCh and other cholinergic agonists when injected into the LVm and inferior horn of the lateral ventricle (LVi) of rats. The effects were stronger fol- lowing administration into the LVi, and probably de- pend on the activation of muscarinic mechanisms.

Among the structures in the immediate vicinity of the LVi, the amygdala complex (AC) was the only one shown to be sensitive to the antinociceptive effect of CCh (Klamt and Prado 1991). The AC is a limbic structure known to participate in affective/ defensive reactions (see Goddard 1964). The second objective of this study was to determine which nucleus within the AC is most sensitive to the effects of CCh. The medial and basolateral nuclei were found to be the most sensitive structures in the AC for antinociceptive and behavioral effects of CCh, respectively. In addition, it is shown that these effects of CCh can be dissociated by treating animals with diazepam.

Material and methods

Subjects and surgery The experiments were carried out on male albino Wistar rats

(140-160 g). Each animal was anesthetized with sodium pentobarbi-

tone (40 mg/kg, i.p.), and a 12-mm length of 23-ga stainless steel

guide cannula was stereotaxically implanted. For intracerebroventric-

ular (i.c.v.) administration of drugs, the guide cannula was implanted

to lie 4.0 mm above the LVm CAP, - 1.5 mm from bregma; L, 1.8

mm from midline) or LVi (AP, 4.0 mm from ear bars; L, 4.4 from

midline). For intra-amygdala injections, the guide cannula was im- planted to lie 4.0 mm above a target nucleus, using the coordinates

AP 4.9-4.0 mm from ear bars, and L 2.4-4.6 mm from midline

(Konig and Klippel 1963). The incisor bar was 2.4 mm below ear bars

in all cases. The guide cannula was then fixed to the skull with 2 steel

screws and dental cement. After receiving penicillin (50 mg/kg, i.m.)

the animal was allowed to recover for at least 1 week before the

experiments.

Tail-flick test Antinociception was assessed using the tail-flick (TF) test (Azami

et al. 1982). On the day of the experiments each rat was placed in a

ventilated glass tube for periods of up to 20 set, with the tail laid

across a nichrome wire coil maintained at room temperature (23_+

2°C). The coil temperature was then raised at the rate of YC/sec by

the passage of electric current, which was adjusted to ensure a tail

withdrawal reflex within 2.5-3.5 sec. A cut-off time of 6 set was

established to minimize the probability of skin damage.

Animals were tested every 10 min until a stable baseline tail-flick

latency (BTFL) was obtained over 3 consecutive trials, and then for

up to 60 min after i.c.v. or intra-amygdaloid injection, at lo-min

intervals. Each tail-flick latency (TFL) was normalized by an index of

antinociception (IA) using the formula IA = (TFL - average

BTFL)/6 - (average BTFL).

Microinjection procedures Drugs were diluted in sterile saline and injected into the lateral

ventricle or AC using a glass needle (70-90 pm, outer diameter)

protected by a system of telescoping steel tubes, as described else-

where (Azami et al. 1980). The assembly was inserted into the guide

cannula immediately before microinjection and the needle advanced

to protrude 4.0 mm beyound the guide cannula tip. The volume

injection was 1.0 ~1 (i.c.v.1 or 0.5 ~1 (intra-amygdaloid) delivered at a

constant rate over a period of 3 min, and the needle was removed 20

set after completion of this procedure.

Behavioral testing procedures The behavioral effects of intra-amygdala administration of CCh

were quantified using the open-field test. The open-field consisted of

an acrylic-walled cylinder measuring 76 cm in diameter and 45 cm in

height, the plywood floor of which was painted white and divided

into 12 fields with painted black lines.

Each rat was initially placed inside the open-field for a 30-min

period of adaptation, and then taken for BTFL recording at lo-min

intervals. The animal was replaced in the center of the open-field

arena and the behavioral activity recorded for 20 min. Diazepam (1

mg/kg) or sterile saline (1 ml/kg) was then injected intraperi-

toneally, followed 20 min later by intra-amygdaloid administration of

CCh (2.2 nmol/0.5 ~1) or sterile saline (0.5 1.~11, in a randomized

manner. Soon after these procedures, the animal was replaced in the

center of the arena for another 20 min period of observation. The TF

test was carried out at lo-min intervals throughout the experiments

and up to 60 min after intra-amygdaloid administration.

The total duration of episode of freezing (sudden and complete

cessation of movements), Straub tail reflex (tail erection), grooming

(licking the fur, washing face or scratching), rearing (standing on

hind legs), and chewing movements were measured in seconds. The

number of lines in the open-field crossed during the time of observa-

tion was taken as a measure of locomotor activity. Episodes of

recurrent shaking, tremors of the head, neck and trunk (wet dog

shakes) were also recorded. An event recorder for continuous behav-

ioral recording (Schmidek et al. 1983) was used for the quantification

of behavioral manifestations.

Histology At the end of the experiment Fast Green (0.5 ~1) dye was

microinjected to label the site of injection. The animal was then

killed by an overdose of sodium pentobarbitone and perfused through the heart with formol-saline. Dye spots were localized from 50-pm serial coronal sections stained with Neutral Red, and identified on

diagrams from the atlas of Kijnig and Klippel (1963).

Statistics Statistical analysis of the effects of drug on TFL was done using

multivariate analysis of variance (MANOVA) with repeated mea-

385

sures to compare the group over all times. The factors analysed were treatments, time and the interactions treatment x time. In the case of significant treatment X time interactions a l-way ANOVA followed by Duncan’s test was perfomed at each time. The analysis was perfomed with the statistical software package SPSS/PC + , version 3.0, and the level of significance was set at P < 0.05. Behavioral changes were submitted to x2 statistic with Yates’ correction or Fisher’s exact test when appropriate. Significance was accepted at the P < 0.05 level.

Drl&S The drugs used were: carbamylcholine chloride (carbachol), car-

bamyl-fi-methylcholine chloride (bethanechol), 1,1-dimethyl-4-phen- ylpiperazinium iodide (DMPP), atropine sulfate, and mecamylamine hydrochloride, all purchased from Sigma. Diazepam was from Roche, and oxotremorine sesquifumarate was from Aldrich. All doses were given as the salt.

Results

Effects of i.c.v. administration of cholinergic agonists A significant increase in the TFL of rats was pro-

duced by injecting CCh (2.2-8.8 nmol) into the LVi (Fig. 1A) or LVm (Fig. 1B). The effect was of rapid onset in both experiments, but was stronger and longer lasting after injection into the LVi. A dose-dependent effect was demonstrated only when CCh was injected into the LVi (F,, 21 = 7.1, P = 0.002).

Administration of atropine (0.7 nmol), but not mecamylamine (5.5 nmol), into the LVi fully inhibited the antinociceptive effect of CCh (4.4 nmol) adminis- tered 20 min later by the same route (Fig. 2). ANOVA indicated that the curves on Fig. 2 are different with regard to TFL (F,, 30 = 8.4, P = 0) and time course

Fig. 1. Time course of the changes in TFL induced by injection of carbachol (CCh) or saline (control) into the inferior horn (LVi) or dorsomedial portion of the lateral ventricle (LVm). Injections were made at the moment indicated by the arrow. Points are the mean (+ S.E.M.) of 5-7 rats per curve. * Significant difference from con-

trol values (P < 0.05, Duncan’s test).

-60 -40 -20 0 20 40 60

TIME (min)

Fig. 2. Effects of administration of atropine (AT, 0.7 nmol), mecamy- lamine (MA, 5 nmol), or saline (1 ~1) into the inferior horn of the lateral ventricle (arrow 11, on the changes in the TFL produced by carbachol (CCh, 4.4 nmol) or saline (1 ~1) injected by the same route (arrow 2). Points are the mean (kS.E.M.) of 5-11 rats per curve. * Significant difference from control (saline + saline) values (P < 0.05,

Duncan’s test).

(Fll3, 180 = 2.1, P = 0.008). Duncan’s test indicates that mecamylamine-treated animals had index of antinoci- ception significantly different from rats treated with saline + saline and atropine + CCh at the times 0 (F3, 3. = 10.27; P < 0.05) and 10 min (F3 3. = 5.5; P < 0.05). At no time during the experiment did atropine-treated rats display TFL significantly different from control rats (saline + saline).

Administration of bethanechol (4.4 nmol), but not DMPP (4.4 nmol), into the LVi caused an increase of the TFL similar to that produced by an equimolar dose of CCh (Fig. 3). ANOVA followed by the Duncan’s test indicate that CCh-treated rats had TFL signifi- cantly different from control (saline) at times 0 (F3, 4o = 3.8; P< 0.051, 10 (F3, 4. = 5.9; P < 0.05), 20 (F3, 4o = 6.9 P < 0.051, and 50 min (F3, 40 = 3.8 P < 0.05). The same statistics indicated that bethanechol pro- duced effects significantly different from control only at time 0 (F3, 4o = 3.8; P < 0.05). At no time during the

s 1.0. i= 0 ULINE

B . 0.6. 0

A CCh 6Ch A DYPP

a

-20 0 20 40 60 J

TIME (min)

Fig. 3. Time course of the changes in the TFL induced by adminis- tration of bethanechol (BCh, 4.4 nmol), dimethyl-phenyl- piperazinium (DMPP, 4.4 nmol), carbachol (CCh, 4.4 nmol) or saline (control, 1.0 bl) into the inferior horn of the lateral ventricle of rats. Injections were made at the time indicated by the arrow. Points are the mean (+ S.E.M.) of 5-15 animals per curve. * Significant differ-

ence from control values (P < 0.05, Duncan’s test).

386

-20 0 20 40 60

TIME (min)

Fig. 4. Time course of the changes in TFL induced by oxotremorine

(doses in nmol/l.O ~~11 or saline (control, 1.0 ~1) injected into the

inferior horn of the lateral ventricle of rats. Points are the mean

(f S.E.M.) of 5-11 animals per curve. * Significant difference from

control values (P < 0.05, Duncan’s test).

experiment was the effect of DMPP significantly differ- ent from control. A dose-dependent effect of ox- otremorine (1.1-5.5 nmol) was also demonstrated (Fig. 4). The curves on Fig. 4 were found significantly differ- ent regarding latencies (F,, z8 = 9.57; P = 0) and time course (Fz4, 168 = 2.01; P = 0.006). In a separated group of 5 animals (data not shown in figures), CCh (4.4 nmol) was still effective when administered into the LVi 10 min after DMPP (4.4 nmol). Therefore, tachy- phylaxis or persistent depolarizations is unlike to deter- mine the lack of effect for DMPP.

. Various stereotypic behaviours and seizures were observed following i.c.v. administration of cholinergic agonists (Table I). The most common behaviors evoked by i.c.v. CCh (4.4 nmol) were freezing, wet dog shakes, and Straub tail reflex. These behaviors were signifi- cantly more frequent after injection of CCh into the LVi (x2 = 10.66, P = 0.008; x2 = 0.83, P = 0.001; and x2 = 18.36, P = 0, respectively). Other behaviors in- cluded grooming and, to a lesser extent, chewing, rear- ing, hyperexcitability to tactile stimuli and convulsion. x2 statistic with Yates’ correction yielded the values 3.29 (P = 0.006), 0.34 (P = 0.55), 0 (P = l), 3.71 (P =

0.05391, and 0 (P = l>, respectively, thus indicating that the frequency of these behaviors were not different with regard to the route of administration.

Grooming was also frequently produced by adminis- tration of bethanechol (4.4 nmol), DMPP (4.4 nmol). and oxotremorine (5.5 nmol) into the LVi. x2 statistic with Yates’ correction indicates that the frequencies of grooming induced by bethanechol (x2 = 3.52; P = 0.006) and DMPP (x2 = 0.9; P = 0.34) were not differ- ent from that produced by CCh. In contrast, ox- otremorine-induced grooming was significantly more frequent than that induced by CCh (x2 = 6.88, P = 0.008).

The most frequent behaviors evoked by CCh at the LVi were less frequent or absent following administra- tion of bethanechol, DMPP, or oxotremorine. Ox- otremorine doses of less than 5.5 nmol yielded no noticeable behavioral change (data not shown).

Administration of atropine (0.7 nmol), but not mecamylamine (5 nmol) into the LVi fully inhibited the behavioral changes produced by further administration of CCh (4.4 nmol) by the same route (Table II). Fisher’s exact test applied to the data in Table II indicates that frequency of behaviors evoked by CCh in saline- and mecamylamine-treated rats were not different.

Effects of intra-amygdala administration of cholinergic agonists

Fig. 5 shows the location of the sites in which cholinergic agonists were microinjected. Carbachol (2.2 nmol) and oxotremorine (2.75 nmol) increased the TFL of rats following administration into the medial, cen- tral, basolateral, and posterior lateral nuclei of the AC (Fig. 6). ANOVA indicated that the effects obtained from medial nucleus of the AC are significantly differ- ent from control (saline) with regard to TFL (F,, 2h = 9.93; P = 0.001) and time course (F,,, ,sh = 1.86; P = 0.043). The effects obtained from central nucleus were

TABLE I

FREQUENCY OF BEHAVIORAL CHANGES PRODUCED BY CHOLINERGIC AGONISTS IN RATS

Drugs were administered into the inferior horn of the lateral ventricle (LVi). Carbachol was administered also into the dorsomedial portion of

the lateral ventricle (LVm).

Carbachol Bethanechol Oxotremorine DMPP

(4.4 nmol) (4.4 nmol) (5.5 nmol) (4.4 nmoll

LVm (n = 6) LVi (n = 18) LVi (n = 81 LVi(n= 11) LVi (n = 131

Grooming 0.67 0.17 0.63 0.73 h 0.38

Chewing 0.17 0.00 0.00 0.00 0.08 Freezing 0.17 0.94 * 0.00 h 0.00 h 0.00 h Rearing 0.17 0.22 0.13 0.09 0.23 Wet dog shakes 0.00 0.83 a 0.13 h 0.00 h 0.08 h Straub’s reflex 0.00 0.89 a 0.13 h 0.00 h 0.15 h Hyperexcitability 0.00 0.39 0.00 0.00 0.00 Convulsion 0.17 0.17 0.00 0.00 0.00

a,’ Significant difference when compared with CCh into the LVm (a) or into the LVi (b) (x2 statistic with Yates’ correction, P < 0.05).

387

TABLE II

FREQUENCY OF BEHAVIORAL CHANGES PRODUCED BY THE ADMINISTRATION OF CARBACHOL (4.4 nmol) INTO THE INFERIOR HORN OF THE LATERAL VENTRICLE OF RATS TREATED 20 MIN EARLIER WITH DRUG-FREE SALINE (1.0 PI), ATROPINE (0.7 nmol) OR MECAMYLAMINE (5 nmoI)BY THESAMEROUTE

Saline Atropine Mecamylamine (n = 6) (n = 6) (n = 7)

Grooming 0.16 0.00 0.29 Chewing 0.16 0.00 0.14 Freezing 0.66 0.00 * 0.71 Rearing 0.16 0.00 0.14 Wet dog shakes 0.50 0.00 * 0.86 Straub’s reflex 0.83 0.00 * 0.71 Hyperexcitability 0.33 0.00 0.43 Convulsion 0.50 0.00 * 0.29

* Significantly different from control (saline) (P < 0.05; Fisher’s exact test).

also different from control with regard to latency (F,, = 4.9; P = 0.0421, but not time course (F,, ,s = 1.68;

F= 0.088). The effects obtained from posterior lateral

(F2.19 = 1.15; P = 0.33 for latencies, and F,z 114 = 1.09; P = 0.37 for time course) and basolateral nuclei of the AC (F2, 15 = 1.55; P = 0.24, and F,,,,,, = 0.85; P = 0.60) were not significantly different from control. In no case was the effect stronger or longer lasting than after administration of cholinergic agonists into the medial nucleus.

Behavioral changes were also evoked by CCh (2.2 nmol), but not oxotremorine (not shown), administered into the medial, basolateral, posterior lateral, and cen- tral nuclei of the AC (Table III). The behavioral changes were more frequent following injection of CCh into the basolateral and posterior lateral nuclei. How- ever, x2 statistic with Yates’ correction indicates that the number of rats showing behavioral changes were significantly different with regard to hyperexcitability to non-noxious stimuli (x2 = 13.82; P = 0.003) and con- vulsions (x2 = 8.1; P = 0.04).

TABLE III

A 4890 u .( A4110

Fig. 5. Mapping in the coronal planes of the rat brain showing sites where carbachol and oxotremorine were administered intracere- brally. The sections were taken from the atlas of Konig and Khppel (1963) at the AP levels indicated. The occurrence (ml or absence CO)

of antinociceptive effects are also indicated. Abbreviations for amyg- daloid nuclei: ac (central), alp (posterior lateral), ace (cortical), cl (centrolateral), abl (basolateral), abm (basomedial), ala (anterior

lateral), mi (massa intercalata), mna (medial).

No significant change in the TFL or behavioral changes was noticed following administration of cholin- ergic agonists into the cortical, basolateral, and ante- rior lateral nuclei of the AC (data not shown).

Dissociation by diazepam of the antinociception and behavioral effects produced by administration of CCh into the medial nucleus of the AC

In these experiments rats were treated i.p. with diazepam (1 mg/kg) or drug-free saline (1 ml/kg), followed 20 min later by the injection of CCh (2.2 nmol) or saline (0.5 ~1) into the medial nucleus. Carba- chol, but not saline, caused a significant increase in the TFL of both saline- and diazepam-treated animals (Fig. 7A). ANOVA indicates that curves in Fig. 7A are significantly different with regard to latencies (F3, 31 = 10.27; P = 0) and time course (F,,, 126 = 1.96, P = 0.014). Duncan’s test demonstrates that significant dif- ferences from control (saline + saline) were found for saline + CCh-treated rats at times 0 (F3, 31 = 7.77; P < 0.051, 10 (F3, 3, = 2.15; P < 0.05), 20 (F3, 3, = 2.95; P < O.OOS), and 30 min (F3, 3, = 6.47; P < 0.005). Di- azepam + CCh-treated rats had latencies different from control at time 0 (F3, 3, = 7.77; P < 0.05) only.

PERCENTAGE OF RATS EXHIBITING BEHAVIORAL CHANGES AFTER ADMINISTRATION OF CARBACHOL (2.2 nmol) INTO THE BASOLATERAL(BL),P~STERIOR LATERAL(PL),CENTRAL(CE),AND MEDIAL (MNA) NUCLEI OFTHEAMYGDALA COMPLEX

(“n”=7) (‘n”= 8) CE MNA X2 P (n = 5) (n = 9)

Grooming 0.00 0.25 0.40 0.44 4.35 0.22 Chewing 0.14 0.25 0.40 0.55 3.41 0.33 Freezing 0.86 1.00 0.60 0.67 4.27 0.23 Rearing 0.28 0.25 0.20 0.22 0.14 0.98 Wet dog shakes 0.57 0.25 0.60 0.33 2.55 0.46 Straub’s reflex 0.71 1.00 0.60 0.55 4.71 0.19 Hyperexcitabihty 0.57 0.87 0.60 0.00 13.82 0.003 Convulsion 0.28 0.50 0.00 0.00 8.10 0.04

1. ..L . . . . ..I. . . . . . . . . ..l -20 0 20 40 60 -20 0 20 40 60

TIME (min) Fig. 6. Time course of the changes in TFL produced by carbachol

(CCh), oxotremorine (0X0) or saline (control) injected into the

medial, central, posterior lateral, and basolateral nuclei of the amyg-

dala. Injections were made at the time indicated by the arrows.

Points are the mean (sS.E.M.) of 5-15 rats per curve. Doses are

given in nmol/OS ~1. * Significant difference from control values

(P < 0.05, Duncan’s test).

The antinociceptive effect of CCh was accompanied by episodes of freezing and wet dog shakes, and Straub tail reflex in saline - but these behaviors were not observed in di~epam-treated rats (Fig. 7B,C). ANOVA indicates that the groups did not differ regarding episodes of wet dog shakes (F3, 25 = 1.27; P = 0.27). In contrast, the episodes of freezing (F3, 25 = 4.45; P = 0.012) and the occurrence of Straub tail reflex (F3.

TIME (min)

Fig. 7. Effects of i.p. administration of diazepam (DZP, 1 mg/ky) or

saline (SAL, 1 ml/kg) on the changes in TFL (A), episodes of wet

dog shakes (WDS) (B), and duration of freezing and Straub’s reflex

(Cl induced by administration of carbachol (CCh, 4.4 nmol) or saline

10.5 ~1) into the medial nucleus of the amygdala of rats. Points or

columns are the mean (_tS.E.M) of S-10 rats per curve (A) or

behavior fB and C). * Significant difference from control (0) (P i

0.05. Duncan’s test).

) ___-__ / ---1

Fig. 8. Effects of i.p. administration of diazepam (1 mg/kg) or saline

(1.0 ml/kg) on the behavioral changes produced by injection of

carbachol(4.4 nmoQ’O.5 ~1) or saline (0.5 ~1) into the medial nucleus

of the amygdala of rats. The i.p. administrations were carried out 20

min before the intra-amygdaloid injections. Locomotor activity (L.A.)

corresponds to the number of lines in the open-field crossed during

the time of observation (20 mitt): total duration of episodes of

grooming, rearing and chewing movements (CM.) are indicated (1’)

saline + saline (control); (e) diazepam f saline; ( n f salute + carbachol;

( A ) diazepam + CCh. Columns are the mean ( f S.E.M.) of S-- 10 rats

per group. Significant difference between groups was found only

after ANOVA applied to B and C. * Significantly different from

control (Duncan’s test; P < 0.05).

15 = 4.69; P = 0.~9) were significantly higher in the safine + CCh-treated rats than in the remaining experi- mental groups.

The experimental groups did not differ with regard to locomotor activity (F 3. 25 = 1.93; P = 0.15) and chew- ing (F,* 25 = 0.7; P = 0.55) (Fig. 8A,D), but were signifi- cantly different with regard to grooming (F3, Ls - _ ._ - 3 55; P = 0.029) and rearing (F3, 25 = 4.98; P = 0.008) (Fig. 8B,C). Duncan’s test, however, indicates that animals treated with diazepam + CCh had locomotor activity higher than rats treated with diazcpam + saline (F,. Z5 = 2.04; P < 0.051, and periods of grooming signifi- cantly shorter than control rats (F .i, 23 = 2.17; P < 0.05).

Discussion

The present study confirms an earlier demonstration that i.c.v. administration of CCh produces antinocicep- tion and evokes various behavioral changes (Metys et al. 1969). In addition, they indicate that the effects of CCh may differ in intensity and time course depending on the portion of the brain ventricle reached by the injection. In fact, it was shown that the antinociceptivc effect of CCh was stronger, longer lasting, and dose- dependent following drug administration into the LVi, in which CCh also caused higher frequencies of behav- ioral changes. These results indicate that structures in the immediate vicinity of the LVi may be involved in the antinociceptive and behavioral effects produced by cholinergic agonists.

389

By and large, cholinergic antinociception seems to be mediated by muscarinic sites within the central nervous system (see Green and Kitchen 1986). In agre- ment with this notion, the effect of CCh injected into the LVi was mimicked by the muscarinic agonists bethanechol and oxotremorine, but not by the nicotinic agonist DMPP, and antagonized by atropine, but not by mecamylamine. These results, however, do not allow us to exclude that non-cholinergic mechanisms may also be involved in the process.

The behavioral changes evoked by administration of CCh into the LVi were all inhibited by previous injec- tion of atropine by the same route, but remained unchanged after mecamylamine. Thus, it is possible that behavioral changes evoked by CCh depend on muscarinic mechanisms as well. Behavioral changes evoked by bethanechol, oxotremorine, or DMPP, how- ever, were much less frequent than those evoked by CCh. Since muscarinic agonists were used at doses sufficient to evoke antinociception, their low activity in producing behavioral changes may reflect that different cholinergic mechanisms must be activated for the ex- pression of each effect. Alternatively, drug diffusion to different structures surrounding the LVi may explain the discrepant effectiveness of cholinergic agonists in producing antinociception and behavioral effects.

Among the structures bordering the LVi, the AC has been shown to possess sites in which CCh consis- tently causes antinociception (Klamt and Prado 1991). In this study, a significant antinociceptive effect was obtained after administration of CCh and ox- otremorine into the medial, central, basolateral, and posterior lateral nuclei of the AC. In no case, however, was the effect stronger than after injection into the medial nucleus. In addition, no significant change in the TFL was obtained following CCh injection into the remaining nuclei of the AC. These nuclei are less than 1 mm from the AC nuclei shown here to be sensitive to CCh. This suggests that effective diffusion of CCh does not exceed 1 mm from the site of injection. It also suggests that the antinociceptive effect of CCh injected into the basolateral and posterior lateral nuclei may depend on drug diffusion to the most sensitive medial nucleus.

The AC receive imputs from all sensory modalities and from the interpretative cortex, and sends efferents to various structures, including the hypothalamus and PAG, which are important mediators of emotional/ motivational responses (Takeuchi et al. 1983; Thomas et al. 1984; Veening et al. 1984; Finch et al. 1986; Thompson and Cassel 1989; Wallace et al. 1989). Eletrophysiological (Bernard and Besson 1990) and anatomic studies (Bernard et al. 1993) have demon- strated efferent projections from the pontine parabrachial area to the central nucleus of the amyg- dala, which may be implicated in the emotional-affec-

tive and autonomic reactions to noxious stimuli. The medial nucleus of the AC is a fundamental structure for the retention or expression of information required for intraspecific behavior (Luiten et al. 1985). The medial nucleus has reciprocal connections with the ventromedial and premammillary nuclei of the hy- pothalamus which send inputs to the PAG (Krettek and Price 1978). A possible explanation for the CCh- induced antinociception from the medial nucleus by activation of the pain inhibitory system, therefore, would depend on the presence of indirect efferents from this nucleus to the spinal cord. However, further experiments are needed to demonstrate such a possibil- ity.

Among other behaviors elicited by i.c.v. CCh, grooming hyperexcitability, freezing, rearing, and mainly wet dog shakes have been described as specific signs of kindling, an experimental epilepsy that may be produced by the daily application of electrical stimula- tion (Goddard et al. 1969: Racine 1972; McNamara et al. 19801, or administration of cholinergic (Goddard et al. 1969), opioids (Hardy et al. 1980), GABAergic (Ap- plegate and Burchfield 1988; Uemura and Kimura 1988), noradrenergic (McIntyre and Edson 19891, or benzodiazepine agents (LeGal La Salle and Feldblum 1983) in limbic structures.

The AC is known as one of the most sensitive structures in the brain from which kindling can be evoked (Vosu and Wise 1975). The amygdala kindling induced by CCh appears to depend on muscarinic mechanisms (Wasterlain and Jonec 1983) and is not correlated with antinociception (Abbott and Melzack 1978). In this study, behavioral signs that resemble amygdala kindling were obtained following injection of CCh into the basolateral, posterior lateral, central and medial nuclei. The basolateral nucleus is known to be the AC site most sensitive to the CCh-induced kindling (Lerner-Natoli et al. 1984). Freezing and Straub tail reflex were more frequent following injection of CCh into the lateral nucleus. Except for the higher inci- dence of grooming and chewing, behavioral changes were generally less frequent after CCh administration into the medial nucleus. Hyperexcitability to non-noxi- ous stimulation and convulsions were not noticed in any rat receiving CCh into the medial nucleus. These results are indicative that behavioral changes from the medial nucleus may depend on drug diffusion to other AC nuclei that are sensitive to the behavioral effects induced by CCh.

Taken together, these experiments are indicative that antinociception and behavioral changes induced by intra-amygdala injection of CCh depend on drug interaction with muscarinic sites at distinct amygdaloid nuclei. In agreement with this hypothesis, we have also shown that, with exception of locomotor activity, chew- ing movements, rearing, and episodes of wet dog

shakes, behavioral changes, but not antinociception evoked by CCh injected into the medial nucleus are inhibited by diazepam at a dose shown to have no antinociceptive effect, This result also indicates that antinociception from media1 nucleus is not secondary to the aversive reactions evoked by CCh, and depend on the activation of mechanisms different from those required for the production of behavioral changes. The AC, in addition to its involvement in a variety of emotional and motivational aspects of behaviour, is essential for fear-related modulation of spinally medi- ated nociceptive reflexes (Helmstetter and Bellgowan 1993). Descending amygdalar systems coming into play during emergency or extreme stress to modulate en- dogenous spinal mechanisms have been postulated (Kuypers 1982). It is possible, therefore, that the AC participates in the response to potentially noxious con- ditions activating mechanisms that condition adaptative behaviors and antinociception which can be dissociated pharmacologically.

Acknowledgements

This work was supported by FINEP (Grant FINEP 43.90.0159/00~ and CNPq (Grant 3~49l/BF). M.A.O. was the recipient of a CNPq fellowship. We wish to thank Dr. W. Schmideck for the use of the event recorder, Mr. P.R. Castania for valuable technical as- sistance and Miss F.H.F. Petean for typing the manuscript.

References

Abbott, F. and Melzack, R., Analgesia produced by stimulation of iimbic structures and its relation to epileptiform after-discharges. Exp. Neurol., 62 (1978) 720-734.

Applegate, C.D. and Burchfield, J.L., Microinjection of GABA ago- nists into the amygdala complex attenuates kindling seizure ex- pression in the rat, Exp. Neurol., 102 (1988) 185-189.

Azami, J., Llewelyn, M.B. and Roberts, M.H.T., An extra-fine as- sembly for intracerebral micro-injection, J. Physiol., 305 (1980) 18-19.

Azami, J., Llewelyn, M.B. and Roberts, M.H.T., The contribution of nucleus reticularis paragigantocellularis and nucleus raphe mag- nus to the analgesia produced by systemically administered mor- phine, investigated with the microinjection technique, Pain, 12 (1982) 229-246.

Bernard, J.F. and Besson, J.M., The spino (trigeminol pontoamyg- daloid pathway: electrophysioIogi~a1 evidence for an involvement in pain processes, J. Neurophysiol., 63 (1990) 473-490.

Bernard, J.F., Alden, M. and Besson, J.M. The organization of the efferent projections from the pontine parabrachial area to the amygdaloid complex: a Phaseolus vulgaris leucoagglutinin (PHA- L) study in the rat, J. Comp. Neurol., 329 (1993) 201-229.

Besson, J.M. and Chaouch, A., Peripheral and spinal mechanisms of nociception, Pharmacol. Rev., 67 (1987) 67-186.

Duggan, A.W., lnh~bition in the spinat cord: its role in the response to injury. In: L. Kruger and J.C. Liebeskind (Eds.), Advances in

Pain Research and Therapy. Vol. 6. Raven Press, New York, 1984, pp. 123-133.

Fardin, V., Oliveras, J.L. and Besson, J.M. A reinvestigation of the analgesic effects induced by stimulation of the periaqueductal gray matter in the rat. I. The production of behavioral side effects together with analgesia, Brain Res., 306 (1984al 105-123.

Fardin, V., Oliveras, J.L. and Besson, J.M. A reinvestigation of the analgesic effects induced by stimulation of the periaqueductal gray matter in the rat. II. Differential characteristics of the analgesia induced by ventral and dorsal PAG stimulation, Brain Res., (1984b) 125-139.

Finc’h, D.M., Wong, E.E., Lderian, E.L., Chen, X.H., Nowlin-Finch, N. and Brothers, L.A.. Neurophysiology of limbic system path- ways in the rat: projections from the amygdala to enthorhinal cortex, Brain Res., 370 (1986) 273-284.

Goddard, G.V., Functions of the amygdala, Phychol. Bull., 62 (1964) 89-109.

Goddard, G.V., McIntyre, DC. and Leech, C.K., A permanent change in brain function resulting from daily electrical stimula- tion, Exp. Neural., 2.5 (1969) 295-330.

Green, P.G. and Kitchen, I., Antinociception opioids and the cholin- ergic system, Prog. Neurobiol., 26 (1986) 119-146.

Hardy, C., Panksepp, J., Rossi 111, J. and Zolovich, A.J., Naloxone facilitates amygdaloid kindling in rats, Brain Res., 194 (1980) 293-297.

Helmstetter, F.J. and Bellgowan, P.S., Lesions of the amygdala block conditional hypoalgesia on the tail flick test, Brain Res., 612 (1993) 2.53-257.

Klamt, J.G. and Prado, W.A.. Antinociception and behavioral changes induced by carbachol microinjected into identified sites of the rat brain, Brain Res., 549 (199119-18.

K&rig, J.F.R. and Klippel, R.A., The Rat Brain, a Stereotaxic Atlas, Krieger, New York, 1963.

Krettek, J.E. and Price, J.L., Amygdaloid projections to subcortical structures within the basal forebrain and brainstem in the rat and cat, J. Comp. Neurol., 178 (1978) 225-245.

Krettek, J.E. and Price, J.L., A description of the amygdaloid com- plex in the rat and cat with observations on intra-amygdaIoid axonal connections, J. Comp. Neurol., 178 (1978) 255-280.

Kuypers. H.G.J.M., A new look at the organization of the motor system. In: H.G.J.M. Kuypers and G.F., Martin (Eds.), Anatomy of Descending Pathways to the Spinal Cord. Vol. 57. Prog. in Brain Res., Elsevier, Amsterdam, 1982. pp. 381-403.

LeGal La Salle, G. and Feldbrum, S., Reversal of the anticonvulsi- vant effects of diazepam on amygdaloid kindled seizures by a specific benzodiazepine antagonists Ro15-1788, Eur. J. Pharma- col.. 86 (1983) 91-93.

Lerner-Natoli, M. Rodouin, G. and Baldy-Moulinier, M., Evaluation of wet dog shakes during kindling in rats: comparison between hippocampal and amygdala kindling, Exp. Neurol., 83 (1984)

Luiten, P.M.G., Koolhaas, J.M., Deboer, S. and Koopmans, S.J., The corticomedial amygdala in the central nervous system organiza- tion of agonistic behavior, Brain Res., 332 (1985) 283-297.

Mayer, D.J. and Liebeskind, J.C., Pain reduction by focal electrical stimulation of the brain: an anatomical and behavioural analysis, Brain Res., 68 (1974) 73-93.

McIntyre, D.C. and Edson, N., ~ndling-based status epilepticus: effect of NE depletion with &OHDA, Exp. Neurol., 104 (1989) 10-14.

McNamara, O., Byrne, M.C., Dasheiff, R.M. and Fitz, J.G., The kindling model of epilepsy: a review, Prog. Neurobiol., 15 (1980) 139-159.

Metys, J. Wagner, N., Metysova, J. and Herz, A., Studies on the central antino~i~eptive action of cholinomimetic agents, Int. J. Neuropharmacol., 8 (1969) 413-425.

Morgan, M.M., DePaulis, A. and Liebeskind, J.C., Diazepam dissoci-

391

ates the analgesia and aversive effects of periaqueductal gray stimulation in the rat, Brain Res., 423 (1987) 395-398.

Oleson, T.D., Kirkpatrich, D.B. and Goodman, S.J., Elevation of pain threshold to tooth shock by brain stimulation in primates, Brain Res., 194 (1980) 79-95.

Prado, W.A. and Roberts, M.H.T., An assessment of the antinoci- ceptive and aversive effects of stimulating identified sites in the rat brain, Brain, Res., 340 (1985) 219-228.

Racine, R.J., Modification of seizure activity by electrical stimula- tion. I. After-discharge threshold, Eletroenceph. Clin. Neuro- physiol., 32 (1972) 269-279.

Schmidek, W.R., NegrHo, N., Nishida, S.M. and Vieira, R., An inexpensive event recorder for continuous behavioral recording, Braz. J. Med. Biol. Res., 16 (1983) 161-164.

Takeushi, V., Matsushima, S., Matsushima, R. and Hopkins, D.A., Direct amygdaloid projection to the dorsal motor nucleus of the vagus: a light and electromicroscopic study in the rat, Brain Res., 280 (1983) 143-147.

Terman, G.N., Shavit, Y., Vewis, J.W., Cannon, J.T. and Liebeskind, J.C., Intrinsic mechanisms of pain inhibition: activation by stress, Science, 226 (1984) 1270-1277.

Thomas, S.R., Assaf, S.Y. and Yversen, D., Amygdaloid complex modulates neurotransmission from the entorhinal cortex to the dentate gyrus of the rat, Brain Res., 307 (1984) 363-365.

Thompson, R.L. and Cassel, M.D., Differential distribution and non-collaterization of central amygdala neurons projecting to different medullary regions, Neurosci. Lett., 97 (1989) 245-251.

Uemura, S. and Kimura, H., Amygdaloid kindling with bicuculline methiodide in rats, Exp. Neurol., 102 (1988) 346-353.

Veening, J.G., Swanson, L.W. and Sawchenko, P.E., The organiza- tion of projections from the central nucleus of the amygdala to brainstem sites involved in central autonomic regulation: a com- bined retrograde transport immunohistochemical study, Brain Res., 303 (1984) 337-357.

Vosu, H. and Wise, R., Cholinergic seizure kindling in the rat: comparison of caudate, amygdala and hippocampus, Behav. Biol., 13 (1975) 491-495.

Wallace, D.M., Magnuson, D.T. and Gray, T.S., The amygdalo- brainstem pathways: selective innervation of dopaminergic, nora- drenergic and adrenergic cells in the rat, Neurosci. Lett., 97 (1989) 252-258.

Wasterlain, C. and Jonec, V., Chemical kindling by muscarinic amyg- daloid stimulation in the rat, Brain Res., 271 (1983) 311-323.

Watkins, L.R. and Mayer, D.J., Multiple endogenous opiate and non-opiate analgesic systems: evidence of their existence and clinical implications, Ann. NY Acad. Sci., 467 (1986) 273-299.

Willis, D.J., Progress in Sensory Physiology, Vol. 3, Springer, Berlin, 1982.