Differential behavioral effects induced by intrathecal microinjection of opiates: Comparison of...

Brain Research, 299 (1984) 31--42 31 Elsevier

Differential Behavioral Effects Induced by Intrathecal Microinjection of Opiates: Comparison of Convulsive and Cataleptic Effects Produced by Morphine, Methadone,

and D-Ala2-Methionine-Enkephalinamide

HANAN FRENK 1, LINDA R. WATKINS 2 and DAVID J. MAYER 3

1Department of Psychology, Tel-Aviv University, Ramat A viv 69978 (Israel), 2Department of Animal Physiology, Agriculture Exp. Station, University of California, Davis, CA 95616, and 3Department of Physiology and Biophysics, Medical College of Virginia,

Virginia Commonwealth University, Richmond, VA 23298 (U.S.A.)

(Accepted September 13th, 1983)

Key words: intrathecal opiates - - convulsions - - morphine - - naltrexone - - D-ala2-methionine-enkephalinamide - - penthylenetetrazol

Intrathecai (i.t.) microinjection of 400/zg of morphine in rats induces convulsive activity restricted to the hindlimbs. This activity is potentiated rather than antagonized by naitrexone, is potentiated in animals pretreated over 6 days with increasingly higher systemic doses of morphine, and is potentiated and prolonged by high thoracic spinalization. Similar to morphine, convulsive spinal activity could also be elicited with i.t. injection of the non-opiate convulsant penthylenetetrazol (PTZ). However, methadone, D-ala2-methio- nine-enkephalinamide (DALA), or naltrexone, injected i.t. at equimolar doses with morphine did not produce similar convulsive be- havior. DALA and methadone rather produced pronounced spinal catalepsy. It was concluded: (1) that the spinal convulsive action of morphine is not mediated by specific opiate receptors; and (2) that the spinal cord may be one of the sites where high doses of systemic- ally administered morphine may produce convulsions.

INTRODUCTION

The ability of morphine and other opiates to pro- duce epileptiform phenomena both in E E G and be- havior has been known for a long time3,17. More re- cently, it has been shown that when morphine or en- dogenous opioidst4,30 are administered into the lat- eral ventricle of rats, electrographic seizure activity may evolve. This epileptiform activity is generally ac- companied by 'wet-dog shakes ' and twitches of the head and forelimbs 6,11. Although never observed for the endogenous opioids, intracerebroventr icular in- jection (i.c.v.) of morphine may, in some cases, be followed by generalized convulsions 14,23. The epilep-

togenic action of i.c.v, morphine, leu- and met-en- kephalin, and fl-endorphin can be prevented11,28, and in the case of f l-endorphin even be reversed 14 by sys- temic administration of opiate antagonists. Consis- tent with this observation, it has been shown that the epileptogenic action of met-enkephalin29 or D-ala 2-

methionine-enkephal inamide (DALA)27 is abolished in opiate tolerant rats. Thus, based upon naloxone- reversibility and opiate-tolerance studies, the i.c.v. administration of the endogenous opioids and mor- phine results in epileptiform activity which is appar- ently mediated by specific opiate receptors.

The epileptogenic activity observed following i.c.v, morphine and endogenous opioids apparent ly originates from structures in the vicinity of the ven- tricular system. Microinjections of the endogenous opioids and morphine elicit naloxone-reversible epi- leptiform E E G activity in the h ippocampus 7, as well as in the thalamuslO,34, amygdalaS,25 and caudate nu-

cleusS,8,18. Fur thermore , the opioid peptides have been found to inhibit neuronal firing in most areas of the brain, including cortex, but to induce naloxone reversible excitation of hippocampal pyramidal cells15,19, 35, an observat ion which points to the hippo- campus as a possible site of origin for opioid-me- diated, naloxone-reversible excitationl4.

Correspondence: H. Frenk, Department of Psychology, Tel-Aviv University, Ramat Aviv 69978 Israel.

0006-8993/84/$03.00 © 1984 Elsevier Science Publishers B.V.

32

In addition to the above observations which point to the presence of an opiate specific epileptogenic system in periventricular areas, there are data which

suggest the existence of a non-specific epileptogenic system as well. In rats, high doses of systemic mor- phine have been reported to cause behavioral con- vulsions 9A6 and electrographic spiking in the EEG 20

which are potentiated rather than attenuated by the administration of the agonist-antagonist nalorphinO 6

or naltrexone 9. Finally, this epileptiform activity of systemic morphine is potentiated rather than dimin- ished in rats made tolerant to high doses of mor- phineg.20.

From these data it seems apparent that, in addition

to an opiate-specific epileptogenic system, a nonspe- cific, morphine-sensitive convulsant system must be present in other sites within the central nervous sys- tem (CNS) of rats. One possible locus of this nonspe- cific system is the spinal cord since intraspinal injec- tions of morphine have been reported to cause con- vulsions in dogs 17. A more recent study has demon- strated that intrathecal morphine may produce myo- clonic twitches in the rat 22. The present paper ad- dresses whether convulsions may be elicited by mor- phine from the rat spinal cord and whether such con- vulsions are mediated by opiate specific receptors.

Additionally, specificity of the spinal cord system was tested by examining a non-opiate convulsant (penthylenetetrazol, PTZ) to determine whether it had similar effects. Lastly, an attempt was made to de- termine whether the spinal cord is part of the epilep- togenic mechanism activated by high systemic doses of morphine.

MATERIALS AND METHODS

Subjects Adult albino male Sprague-Dawley rats (Harlan)

were used in all experiments. The animals weighed 350-500 g at the time of surgery. They were individu- ally housed in metal cages with a grid floor with water and lab chow pellets available ad libitum. The ani- mals were maintained on a 12 h light cycle (light on from 07.00 to 19.00 h).

Surgical techniques EEG electrodes. EEG recording electrodes consis-

ted of 4 stainless-steel jeweller's screws soldered to

stainless-steel wire which was insulated with Teflon except at the tips. The screws were attached by means of the wire to connector pins (Amphenol).

The screws were threaded into the skull bilaterally overlying the prefrontal and occipital cortex. The connector pins were then snapped into a nylon con-

nector plug (Amphenol) and the whole assembly ce- mented to the skull with dental acrylic.

Intrathecal (i.t.) catheters. Following induction of

anesthesia (50 mg/kg sodium pentobarbital, i.p., supplemented with Metofane), the cisterna magna was exposed and incised. Saline-filled sterile polyeth- ylene tubing (PE 10) was then inserted into the suba- rachnoid space and gently threaded 8.5 cm caudally such that the cut end of the catheter lay at the lumbo- sacral enlargement. The PE tubing was plugged at the rostral end with a sterile piece of 30-gauge wire and secured by flowing dental acrylic around the i.t. catheter and a skull screw. All animals were treated post-operatively with gentamycin, as required.

Spinalization: After the rats were anesthetized with Metofane, a laminectomy was performed at the second thoracic vertebral level (T2) and the dura was reflected. The spinal cords of 18 rats were transected

with a heated cauterizing electrode; another 18 rats served as sham-operated controls. The exposed spi- nal cords were then covered with Gel-Foam powder

(Upjohn) and the wound closed. All animals were treated post-operatively with gentamycin, as re- quired.

After behavioral testing was completed in these animals, the rats were overdosed with sodium pento- barbital. The vertebral columns were then excised and soaked in 10% formalin for 1-2 weeks. At this time, the thoracic spinal cords of sham-operated and spinalized rats were exposed, visually examined, and manually probed for any evidence of spinal damage or incomplete transection, respectively.

Microinjection technique. The unanesthetized ani- mals were restrained in Plexiglas tubes, each of which had an opening in the top through which the i.t. catheter and electrode assembly protruded, al- lowing for both continuous EEG recording and drug delivery without handling the animals. Following baseline EEG recording (see below) animals were in- jected with one of the treatment drugs, which in all cases was dissolved in 20 ~1 of saline. The catheters were flushed with an additional 10/~1 of saline (void

TABLE I

Doses (mg/kg) injected into the animals on a 6-day tolerance schedule

33

Day l Day2 Day3 Day4 Day5 Day6

Morning 15 30 60 90 120 150 Evening 25 45 75 105 135 150

volume of the catheters). Injections were delivered over 35--45 s using Hamilton 50 pl microsyringes. Separate microinjection assemblies were used for each drug tested (for detailed methodology see ref. 31). Following microinjection animals were tak- en out of the restrainer and returned to the home- cage.

Drugs. The following drugs were administered i.t.: morphine sulfate (400pg), D-ala2-methionine-en - kephalinamide (DALA, Vega Biochemicals, 550 pg), naltrexone hydrochloride (Endo Laborato- ries, 400/~g), methadone hydrochloride (400 gg), penthylenetetrazol (PTZ, Summit Hill Laboratories, 2 mg), and physiological saline. Except for PTZ, all drugs were administered at equimolar doses.

Naltrexone hydrochloride (10 or 25 mg/kg), nal- oxone hydrochloride (10 mg/kg, Endo Laborato- ries), and morphine (300 mg/kg) were injected intra- peritoneaUy (i.p.). Naltrexone and naloxone were in- jected in 1 ml/kg physiological saline, whereas mor- phine was administered in 6 ml/kg of the same vehicle.

Induction of tolerance. Animals subjected to the tolerance regimen were injected twice daily (i.p.) fol- lowing implantation of i.t. catheters with increasingly higher doses of morphine over the course of 6 days. The doses these animals received are summarized in Table I. All injections, except for the last two, were administered in 1 ml/kg saline. The last two were ad- ministered in 2 ml/kg saline. All animals were weighed daily and the doses adjusted accordingly. Another group of animals received, following the same schedule, a matched volume of saline.

General procedure and experimental groups Behavioral assessments. All animals were ob-

served throughout the test period for convulsive ac- tivity. Generalized convulsion were scored following the 5 stages of Racine21. In addition, the animals were tested for occurrence of analgesia and catalep- sy. Analgesia was judged present if no vocalization, orientation, or struggling was observed in response

to severe manual tail pinch. Complete hindbody cat- alepsy was judged present if: (1) the hindlimbs of the animals could be hooked on a 18 cm raised ledge, and passively remained in that position for 1 min; (2) the hindpaws could be flexed and extended com- pletely without resistance from the animal; and (3) the hindquarters of the animal could be turned so that both hindpaws lost contact with the grid floor and the animal passively maintained this posture for 1 min. Full body catalepsy was judged present if: (1) the ani- mal could be hooked on a 18 cm raised ledge first with the front paws and subsequently with the hind- paws and the animal would remain in each of these positions for 1 min; and (2) if the animal could be turned on its back without eliciting a righting reflex for at least 1 min. Catatonia was judged present in a part of the body if that part showed no spontaneous movement and remained rigid when attempts at flex- ion or extension were made 13.

i.t. D A L A and morphine with EEG recording. Twenty animals were used, each of which had been implanted one week previously with cortical EEG electrodes and i.t. catheters. On the test day, the Amphenol conector plug of each animal was con- nected by way of a recording cable to a Grass Model 7B polygraph. Baseline EEG was recorded for 5 rain prior to drug administration. The animals were then randomly divided into 3 groups and microinjected through their i.t. catheters with one of the following drugs, while still connected to the polygraph. The first group (n = 10) received 400 pg of i.t. morphine, the second group (n = 5) received 550 pg i.t. DALA, whereas the last group (n = 5) received i.t. saline. Upon completion of drug delivery, the animals were returned to their homecage for a I h observation and recording session. At the end of the hour all animals received 2 injections of 10 mg/kg naloxone, 6 rain apart. EEG recordings and behavioral observations were terminated 10 min after the second injection.

Effects of i.t. morphine and i.t. pentylenetetrazol in spinalized animals. Twenty-six animals were ira-

34

planted with i.t. catheters following either spinaliza- tion (n = 13) or sham-spinalization (n = 13). The day following surgery, 7 of the animals in each group were injected i.t. with 400 pg of morphine. The re- maining animals (n = 12) received 2 mg PTZ i.t. Each animal was observed continuously for 30 min post-injection and for an additional 5 min beginning

55 min post-injection. Effects of systemic morphine in spinalized animals.

Five animals were spinalized and 5 others were sham- spinalized. The day following surgery, all animals were first observed for 10 min and then injected with morphine (300 mg/kg i.p.). Ninety minutes following

the morphine injection animals were observed for 10 min. They were then injected with naltrexone (25 mg/kg ip). Upon completion of the naltrexone in- jection each animal was continuously observed for 10

min. Naltrexone antagonism of i.t. morphine. Sixteen

animals were prepared with i.t. catheters. One week later, 11 animals were pretreated with naltrexone i.p. (25 mg/kg) whereas the remaining animals received equivolume saline. Five minutes later, 5 naltrexone- pretreated and 5 saline-pretreated rats each received 400 pg of morphine i.t. The 6 remaining naloxone- pretreated animals received a matched volume of sa- line i.t. Animals were continuously observed for 40 min post-injection and for an additional 5 min begin- ning 55 min post-injection.

Effects of i.t. morphine in morphine tolerant ani- mals. Six morphine tolerant animals and 4 control an- imals were injected on the morning following com- pletion of the drug tolerance regimen (see above) with 400 pg i.t. morphine. Animals were continu- ously observed for 30 min following the i.t. morphine injection and for an additional 5 min starting 55 min following the injection.

Effects of i.t. naltrexone and methadone. Eleven animals were prepared with i.t. catheters. One week after surgery, 6 of these animals were injected with 400 pg of i.t. naltrexone, whereas the remaining ani- mals were injected with 400/~g of i.t. methadone. Animals were continuously observed for 30 min fol- lowing the injection. Naltrexone-injected animals were then observed for an additional 5 min beginning 55 min post-injection. Methadone-injected animals were injected i.p. with naltrexone (10 mg/kg) 30 min following the methadone injection and were ob-

served for an additional 5 min immediately following the naltrexone injection.

RESULTS

Effects of i. t. morphine on behavior and cortical EEG in intact animals

Behavior. Intrathecal administration of 400/~g of morphine, but not of D A L A (see below) or a match-

ed volume of saline vehicle, produced dramatic be- havioral changes. Within 1.9 + 0.9 min of injection, animals started to display alternating clonic contrac- tions of the hindpaws, often preceded by a tonic ex- tension of these limbs. These clonic contractions somewhat resembled normal scratching, but were much more vigorous as demonstrated by the tufts of fur the animals tore from their skin, baring and wounding it. This behavioral pattern was accompa- nied by slow rotational movements of the hindquart- ers along the longitudinal axis of the body. These ro- tational movemens often culminated with the animal turning on its back. The head and the forepaws often did not follow this movement, and in fact the animals sometimes held on to the cage grid or wall, vocalizing

loudly. In other cases the animal would start to mouth the limb that was tonically extended, stopping only when clonic contractions started. When the ani- mal's body started to rotate, the hindlimb that did not touch the grid would contract rhythmically, stopping as soon as it touched the grid. Often the other hind-

100 10

8 0 ._c

E

~ . 6 0

z

8 . . c

E Lo

6 - -

- r U

l u o,b

2

0 . . . . . . . . . . . . . O 0 10 2 0 3 0 4 0 5 0 6 0

T I M E E m i n ' l

Fig. 1. Mean number of scratching bouts (dots) and myoclonic twitches (triangles) following i.t. administration of 400 ~g of morphine (n = 10). In this and subsequent figures, data points consist of the average number of scratching bouts or twitches summed over 5 min periods. Vertical bars represent standard error of the mean.

35

limb would then start to contract. In the cases where both legs were in the air, they would contract simulta- neously.

This pattern, with varying intensity and duration (10-90 s) was repeated many times throughout the first 30 min of the session (Fig. 1). After about 10 min, these scratching bouts would become less fre- quent, and myoclonic twitches of one or both hind- limbs could be observed. By this time, the animals were nearly always in a standing position. With the occurrence of each twitch, the animal jerked violent- ly forwards, often vocalizing loudly. In some cases,

the jerk would not consist of a single paw twitch, but of a volley of twitches, resembling a short (1 s) scratching pattern. The development of the twitch frequency is shown in Fig. 1.

Cortical EEG. Neither scratching nor twitches were accompanied by any changes in EEG, which re- mained desynchronized throughout the first 15 min. Movement artifacts sometime accompanied scratch- ing and twitching (Fig. 2) but could be prevented or diminished by changing the position of the recording cable. In 6 of the 10 animals tested, spikes and some- times electrographic seizures appeared in the cortical

A

I I S

![ . , L I~ , ~ . Lli l J, II II lid JIJ. Ili.I..l~l l i I t , l l . . [ , IJ~. l LLi!.lld, i. l~L . ] 11 ,J.,J.~ l l tL .~l. IiJ.l,. Li ] ~LtLb]Jl .JJJl . i / . i. I ,[1~ 1~, .I i . J ~ . i . . I b l k I l l , l .i d , l~ l ~ J,d~ J~l. i i . l ~,J. t ~liLtJ_!~ i~Jlt J J,~, ,, ,, i~, ~111 il l i ~.,~ • ii l l i i " .....

B m ~ . ~ ' ~ . ~ < ~ ] ~ r ~ r ~ ~ . ~ r ~ r ' 1 ~ v ~ 3 ~ ` . ~ r ~ ' ~ ] ~ , ~ ' ~ 1 ~ 1 ~ ' ~ r ~ ~ r ~ ' ~ ` ~ '

T T T T T T T T T I 3 0 0 pV I

6 s e c

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. ............ L ........ J, ........... i ......... L_ I . .L i...J . . . . . , . , . . . . . . . . . . . . . . . L,..L L....J ..... t ..... ._l. .............. J . . . . .

l I T t T l T T T l l T I T l l

,, ~ / , ~ ~ / i l \ i~!h ' ' ~/!~ ~ . . . . ~ , \ ~ I

STAGE 3 C O N V U L S I O N " ', '

1 2 0 1 ' V I

6se '¢

T l

/ ~ /,, ! /NALOXONE, , 10 mglkg / M O V E M E N T . . . . ' , t

I , , T : . ~ t ~ I : [ , I I . . . . . ",t . ~,. ,j :Lu.,,jiil,,.~lt~, ....,,a ~ ,L, ~X, ~ , ,~d . ' , i ,~ ; ,~ l i t ,~J . j , l t,, ¢, Ji~a.,l~'J.,,~t~ttiii.~!t~,J~lJ,L~,llL,~ I I I I ~' I l l~ ' i "11 ' I '

/\, ' \ \ ~'~ , , , t, ' 3 0 0 p V l 1 6 sec

Fig. 2. Cortical EEG in animals treated with 400 pg of i.t. morphine. No paroxysmal activity, but some movement artifacts accompany the scratching bouts (S) in trace A, or myoclonic twitches (T) in trace B. Once cortical spikes occur (trace C) they are not associated with the myoclonic twitches. In some animals behavioral and electrographic seizures developed (trace D). Spikes in 2 of the 6 animals showing this activity could be reversed by naloxone (trace E).

36

E E G (Fig. 2). The mean latency until the onset of spiking was 30.7 + 15.8 min for these animals. It

should be emphasized that spiking never started be- fore 15 min post-injection. In all cases, cortical spikes were dissociated from the myoclonic twitches of the hindlimbs (Fig. 2), although in some animals myoclo- nic twitches of the head and forelimbs occurred in as- sociation with these E E G spikes. In 2 animals, elec- trographic seizures were observed (Fig. 2) starting after 25 and 30 min, respectively, and repeated elec- trographic seizures occurred in one of them. All of these seizures observed were motor seizures, but none exceeded stage 4. Hindl imb involvement was

not seen. All animals were analgesic within the first 2 min of

injection, and showed evidence of full-body catato- nia at the end of the 60 min session. Catatonia and analgesia were reversed in all cases by 10 mg/kg of naloxone. E E G spikes were reversed by this dose of naloxone in 2 of the 6 animals displaying this activity

(Fig. 2) whereas naloxone failed to affect spikes and electrographic seizures in the other 4. An additional injection of naloxone (10 mg/kg) 6 min following the first had no further effect.

Behavioral and EEG changes following i.t. DALA. Animals injected with D A L A never showed scratches or twitches as displayed by those animals receiving equimolar morphine. However , within 2 min post-injection pronounced tail-pinch analgesia occurred. In addition, within 5 min all but one animal displayed catalepsy in hindpaw and lower back. Ani- mals did not react to touch in those regions, their hindpaws could be hooked on a raised ledge, their hindquarters turned so that their hindlimbs lost con- tact with the floor of the cage, and both legs could be flexed and extended without resistance by the ani- mal. By contrast, one animal showed pronounced catatonia in those same body regions; the hindlimbs were rigidly extended and could not be flexed. Apar t f rom this rigidity, this animal showed no difference

DALA SPIKES, ONSET

i

DALA NONCONVULSIVE SEIZURE .... , ! ~' ~ ~, i' " ' I ~ ' , i i II II]

, I , t~ t i r I I I , - "'~ I I' '~il'Ji I, • i ! i : ~ ;! !~ , t ' '1

300vV I DALA SPIKES, REVERSAL

i 6 sK

NAL OXONE, 10 mg/kg MOVEMENT

Fig. 3. Cortical EEG in animals treated with 550 pg i.t. DALA. In 3 of 5 animals spikes occurred after hindbody catalepsy was fully de- veloped (upper trace) and electrographic, nonconvulsive seizures occurred in 2 of these animals (middle trace). Spike reversal by nal- oxone is shown in the bottom trace.

from the other animals. Both catalepsy and catatonia lasted for the full duration (60 rain) of the session.

In 3 of the 5 animals (including the catatonic ani- mal), EEG spikes appeared at 23 + 17.8 min (Fig. 3). These spikes never appeared before 9 min post-injection, after both analgesia and pathological behavior were fully developed. In addition, 2 of these animals had repeated EEG seizures, starting at 14 and 48 min post-injection, respectively (Fig. 3), which were not accompanied by behavioral corre- lates. Simultaneous with the appearance of EEG spikes, normal morphine-like catatonia appeared in the forelimbs and head of those animals, although no change was apparent in the pathology of the hind- quarters. Those animals which had no spikes in the cortical EEG did not display behavioral changes in their upper body.

Naloxone (10 mg/kg) administered at the end of the 60 rain session reversed EEG seizures, spikes (Fig. 3), and upper body catatonia in all but one ani- mal, where spike frequency was reduced, but not eliminated. However, complete reversal of the lower body pathology was only observed in 2 animals (in- cluding the catatonic one). Despite the injection of an additional 10 mg/kg of naloxone 6 rain following the first injection, only partial reversal (restoration of the tail pinch reflex, appearance of some sponta- neous movements in hindlimbs) was obtained in two additional animals. In one animal, no reversal of flac- cid catalepsy was observed.

180

20

.E E 140

u Z 100 t , -

I - -

60

1() 15 2'0 2'5 3'0 TIME [ m l n ]

6

220 11

" 9 .-¢

E 7 Ln

~n

5 u

1

8'0

Fig. 4. Scratching bouts (dots) and twitches (triangles) in spi- nalized (open symbols, n = 7) and sham-spinalized (filled sym- bols, n = 7) animals following 400/ag of i.t. morphine.

37

Effects of i.t. morphine in spinalized animals The spinalized animals displayed a massive poten-

tiation of the twitches, starting 20 rain post-injection, and lasting until the end of test period in response to morphine (400 pg, i.t.) (Fig. 4). At 25 and 30 min, scratching bouts were more frequent in spinalized an- imals, although no such potentiation was noticed prior to 25 min, or at 60 min. It should be noted that the scratching in spinalized animals was much less vigorous than that observed in intact animals and consisted of a rapid waving of the hindlimb, without touching the fur of the body. No noticeable decrease in twitch frequency could be detected in the spi- nalized animals following 60 min, and in most cases twitches continued for 2 or 3 h before a decrease in frequency occurred. It should be emphasized that the marked increase in the duration of twitches in spi- nalized animals cannot be accounted for by disrup- tion of CSF flow since the dura had also been re- flected in sham operated animals.

Effects of PTZ in spinalized animals The results obtained following i.t. injections of

PTZ (2 mg) in spinalized and sham-spinalized ani- mals are summarized in Fig. 5. Whereas a few twitches were observed immediately following injec- tion in the sham-spinalized animals, no repeated clonic contractions of the hindlimbs, nor rotating movements along the longitudinal axis of the body were seen. However, in spinalized animals these phenomena all occurred, qualitatively identical to those seen in the spinalized animals injected i.t. with 400 pg of morphine.

¢

E

.l-

u.-

9 9

¢

r r "~

Z 5 5 v

<

3 3 ~

1 1

0 s ~b ~s 2o 2s 3o TIME [ min]

Fig. 5. Scratching bouts (dots) and twitches (triangles) induced by i.t. PTZ in spinalized (filled symbols, n = 6) and sham-spi- nalized (open symbols, n = 6) animals. No scratching occurred in sham-spinalized animals.

38

Effects of systemic morphine (300 mg/kg) in spi- nalized animals

The injection of 300 mg/kg of morphine i.p. re- sulted in the appearance of myoclonic twitches both in spinalized and in sham-spinalized animals. Where- as, in spinalized animals, twitches originating caudal- ly to the lesion could be easily distinguished from those occurring rostrally to the lesion, no such differ- entiation was possible in sham-spinalized animals. Because of this confounding, no statistical compari- sons of twitches were made. However, both spi- nalized and sham-spinalized animals showed scratch- ing in the hindlimbs, indistinguishable from that in- duced by the i.t. injection of morphine. No difference between the groups existed in frequency of this be- havior prior to naltrexone. The injection of 25 mg/kg of naltrexone following 90 min after the morphine in- jection result_ed in a significant increase in scratching bouts both in the spinalized (from 1.8 + 0.8 to 6.4 + 4.6, P < 0.05) and sham-spinalized (from 1.4 + 2.6 to 7.6 + 6.8, P < 0.05) animals.

In one of the sham-spinalized animals, several scratching bouts were followed by two full body con- vulsions (stage 5). The contractions of the hindlimbs and the rotational movement of the body causing the animal to turn on its back during the behavioral con- vulsion were identical to the scratching observed prior to these stage 5 convulsions and the scratching observed following the i.t. administration of 400/~g of morphine in other animals.

Naltrexone antagonism Pretreatment with 25 mg/kg of naltrexone i.p. was

totally ineffective in blocking both scratching and myoclonic twitches elicited by 400 pg of morphine i.t. (Fig. 6), although it prevented both analgesia and catalepsy following this injection. In contrast, the myoclonic twitches in naltrexone-pretreated animals were potentiated when compared to saline-pre- treated controls at 20, 25, and 30 min (P < 0.05) and during the last 5 min of the test session (P < 0.05). However, like saline-pretreated animals, naltrex- one-pretreated animals displayed significantly fewer (P < 0.05) twitches at the end of the session when compared to 25 rain. Animals pretreated with nal- trexone (25 mg/kg) but injected i.t. with the saline vehicle did not display any change in behavior.

Morphine (400 I~g) i.t. in morphine tolerant animals Animals injected twice daily with increasingly

higher doses of morphine displayed tolerance to the analgesic effects of 400/~g of morphine i.t., as was evident from their vigorous response to tail pinch. However, no tolerance could be detected to scratch- ing elicited by the i.t. injection of 400/~g of morphine (Fig. 7). In fact, a significant potentiation in twitches was seen in the tolerant animals (P < 0.05), which was observed at 20, 25 and 30 min but not at the end of the session (55-60 min post-injection). It should be noted that, unlike the behaviors displayed by control animals, at the end of the 60-min session the twitch frequency for tolerant animals had not significantly decreased when compared to their own 30 min fre- quency (P < 0.30).

Naltrexone and methadone i.t. Neither naltrexone nor methadone elicited

scratching or twitches when injected i.t. at doses which were equimolar to 400ktg of morphine. Whereas naltrexone did not produce any behavioral changes, methadone elicited tail pinch analgesia within 1 min in 4 of the 5 animals tested. Analgesia lasted until naltrexone (10 mg/kg) was injected 30 min later. In all methadone-treated animals, substan- tial hindpaw catalepsy occurred. In no case was the catalepsy as pronounced as in those animals injected with DALA. The animals were able to walk when touched. However, the hindquarters of the animals could be pulled to full extension, or pushed to full

165

135

105

75

45

15

6 ~b ~s 2'o 2's 3'0 TIME [ 'min ' l

9

.S E 7 to

" r i d

%1

5

3

6'0

Fig. 6. Scratching bouts (dots) and twitches (triangles) induced by i.t. morphine (400/~g) in naltrexone (open symbols, n = 5) or saline (filled symbols, n = 5)-pretreated animals. Note the potentiation of twitches in naltrexone-pretreated animals.

flexion while the animals remained flat on their abdo- mens. Animals could be turned so that their hind- limbs did not touch the floor of the cage while their head and forepaws remained in a normal position. Hindlimbs could be placed on an elevated ledge with- out the animals attempting to move them. This situa- tion remained unchanged for 17.3 + 3.6 rain. Cat- alepsy then gradually became less pronounced, but in all animals hindlimbs could still be turned so that they did not touch the floor during the final minute of the 30-min test session. No behavioral manifestation of methadone injection could be observed 2 rain after naltrexone administration.

DISCUSSION

These experiments have demonstrated that mor- phine restricted to the spinal cord produced tonic and clonic contractions of the hindlimbs combined with rotational movements along the longitudinal body axis. These behaviors were followed by myoclonic twitches, the occurrence of which has been observed previously following i.t. injections of morphine 22.

Pharmacological evidence derived from the pre- sent study and a review of the literature provide 5 major lines of evidence that these effects are convui- sant in nature. First, whereas part of the behavior ob- served resembles hindlimb scratching, other compo- nents (the rotational movements, the pronounced tonic extensions, the subsequent myoclonic twitches) do not fit any normal pattern of grooming behavior. Second, in the present study the systemic administra- tion of morphine (300 mg/kg) to spinalized and intact animals elicited hindlimb contractions and myoclonic twitches identical to those induced by i.t. administra- tion of this drug. This behavior was potentiated by the i.p. administration of naltrexone and culminated in the observance in one animal of repeated stage 5 seizures. During these stage 5 seizures, which have been previously reported to result from this systemic drug regimen 9,16, the clonic hindlimb contractions and rotational body movements, typical of the falling phase of the stage 5 seizures, were indistinguishable from the behavior pattern induced by 400/~g of i.t. morphine. Third, the administration of 2 mg of the convulsant pentylenetetrazol (PTZ) i.t. to spinalized animals caused qualitatively identical behavior as did the 400/~g of morphine administered via the same

39

165 11

135 9

"i ~ 105 7

75 5

45 3 ~n

15 1

6 s lb is 2'o 2's 30 60 TIME [mini

Fig. 7. Scratching bouts (dots) and twitches (triangles) induced by i.t. morphine (400/~g) in animals pretreated for 6 days with morphine (open symbols, n = 6) or saline (filled symbols, n = 4). Note the potentiation of twitches in morphine-pretreated animals.

route. Fourth, the dose of morphine employed has been observed to induce epileptiform electrographic activity in other parts of the CNS. The lower body contractions and myoclonic twitches induced by i.t. morphine were not mediated by supraspinal struc- tures since those behaviors occurred in spinalized an- imals and were fully developed before EEG changes occurred in intact animals. However, when EEG changes did occur following approximately 30 min, they were invariably epileptiform.

Finally, pharmacological similarities observed in the present study between the electrographic and be- havioral seizures following systemic morphine 9 and the lower body activity induced by i.t. morphine sug- gest that these different manipulations activate, in part, the same system. Although analgesia was nal- trexone-reversible, the convulsant action of i.t. mor- phine on the hindbody was not antagonized, but rath- er potentiated, by pretreatment with this opiate an- tagonist. Naltrexone had the same effect on convul- sions induced by systemic morphine 9. Similarly, ani- mals made tolerant to the analgesic effects of mor- phine did not display decreased but rather poten- tiated convulsant activity of the hindlimbs following i.t. morphine. The same holds for convulsions in- duced with high systemic doses of morphine 9. These similarities in effects of pharmacological manipula- tions suggest that the convulsant action of high sys- temic doses of morphine is mediated, in part, at the

40

level of the spinal cord. Taken together with the ob- servations that other opiate agonists such as DALA and methadone failed to elicit convulsant activity when injected i.t., these findings also demonstrate that this convulsant system is not mediated by specif- ic opiate receptors.

Several experimental manipulations caused poten- tiation of the convulsant action induced by i.t. mor- phine. Spinalization enhanced paroxysmal activity both in those animals receiving i.t. morphine and PTZ. In addition, potentiation was observed, al- though to a lesser degree, both in animals made toler- ant to morphine as well as those pretreated with i.p. naltrexone. Whereas the potentiation by spinaliza- tion can be most easily explained by the existence, in intact animals, of a tonically active descending inhibi- tion of spinal cord excitabilityl,Z, 33, potentiation by morphine tolerance and naltrexone could be ex- plained in two ways. On the one hand, i.t. adminis- tration of morphine could activate simultaneously a pro- and an anticonvulsant system. The existence of an opiate anticonvulsant system able to suppress electrographic seizures induced by intracerebroven- tricular (i.c.v.) opiates has been demonstrated28, 29. Consistent with these observations, it has been sug- gested that endogenous opioids exert a tonic anticon- vulsant effect within the CNS 26. The data from the present study indicate that this anticonvulsant system must be located, at least in part, within the spinal cord. First, naltrexone potentiated convulsive activ- ity induced by 300 mg/kg of morphine i.p. in the lower body of spinalized animals, demonstrating that nal- trexone's potentiation of this activity is not dependent on supraspinal structures. Second, the potentiation of twitch frequency observed in spinalized animals was longer in duration than the' potentiation of twitches in naltrexone pretreated animals. Enhanced scratching was observed in spinalized animals and not in morphine-tolerant and naltrexone-pretreated animals. On the other hand, the potentiation naltrex- one has on the convulsant activity induced by i.t. morphine could be explained in a different way by considering the interactions that opiate agonists and antagonists have with the GABAergic system. Mor- phine, but not DALA, antagonizes the actions of GABA in murine spinal cord neurons, and this effect is not reversible by naloxone3L In addition, mor- phine has been shown to displace [3H]GABA from

GABA receptor sites in homogenates of the human cerebellum 4. In this study 4 it was found that naloxone had the same effect, suggesting that naloxone (and possibly naltrexone) could display an additive effect with the convulsant properties of morphine, account- ing for the naltrexone produced potentiation of mor- phine's convulsant action observed in our present study.

An unexpected finding in our study was the obser- vation that neither DALA nor methadone-induced epileptiform activity when injected i.t., but rather produced pronounced catalepsy. The lack of convul- sive activity following equimolar doses of DALA and methadone is especially surprising in view of the pre- vious reports which demonstrated the epileptogenic potential of i.c.v. DALA in rats 27 and of systemic methadone in primates z4. However, electrographic seizures elicited by i.c.v, injections of endogenous opioids are mediated by opiate receptors, as was evi- dent from the observations that naloxone 11 and opiate tolerance 27 could prevent this EEG activity. In addition, unlike morphine, DALA does not antag- onize GABA action in murine spinal cord neurons 3z, an observation that again suggests that GABA is in- volved in the convulsant properties morphine exhib- its in the spinal cord. Concerning the lack of convul- sant action observed with methadone in the present study, it is possible that the doses we used were too low to elicit this effect. Alternatively, it could be that methadone exerts its epileptogenic action only at ce- rebral sites.

Regarding supraspinal sites of opiate action, ob- servations made both in our present study as well as subsequent ones (in preparation) suggest that the spi- nal cord is not the only anatomical locus involved in the nonspecific convulsant actions of morphine. Dif- fusion of both DALA and morphine to the cerebrum caused spikes and repeated EEG and upper body sei- zures which, in the case of morphine, were only re- versed by naloxone in 2 of 6 animals. As electrogra- phic seizures and spikes induced by morphine injec- tion into the lateral ventricle and subcortical areas can be antagonized by naloxone11,18,25, 34, it is possible that cortical areas are involved in the nonspecific convulsant action of morphine. In fact, topical appli- cation of morphine to the cerebral cortex does elicit spikes and behavioral and electrographic seizures3,12. We presently are involved in studies investigating

41

whether specific opiate receptors mediate this action

and whether such an action is shared by other

opioids.

ACKNOWLEDGEMENTS

Naltrexone and naloxone used in these studies

were generously provided by Endo Laboratories,

Garden City, NY. Supported by PHS Grant DA-

00576 to D.J .M,

REFERENCES

1 Besson, J. M., Guilbaud, G. and LeBars, D., Descending inhibitory influences exerted by the brainstem upon the ac- tivities of dorsal horn lamina V cells induced by intraarteri- al injection of bradykinin into the limbs, J. Physiol. (Lond.), 248 (1975) 725-739.

2 Cervero, F., Iggo, A. and Molony, V., Responses of spino- cervical tract neurones to noxious stimulation of the skin, J. Physiol. (Lond.), 267 (1977) 537-558.

3 Corrado, A. P. and Longo, V. G., An electrophysiological analysis of the convulsant action of morphine, codeine, and thebaine in the brain, Arch. int. Pharmacodyn. Ther., 132 (1961) 255-269.

4 Dingledine, R., Iversen, L. L. and Breuker, E., Naloxone as a GABA antagonist: evidence from iontophoretic recep- tor binding and convulsant studies, Europ. J. Pharmacol., 47 (1978) 19-27.

5 Dzoljic, M. R. and Van Der Poel-Heisterkamp, A. L., The role of the nucleus accumbens and nigrostriatum in enkeph- alin-induced myoclonus, Pharmacol. Biochem. Behav., 13 (1980) 103-106.

6 Dzoljic, M. R. and Van Der Poel-Heisterkamp, A. L., Del- ta opiate receptors are involved in the endopioid-induced myoclonic contractions, Brain Res. Bull., 8 (1982) 1-6.

7 Elazar, Z., Motles, E., Ely, Y. and Simantov, R., Acute tolerance to the excitatory effect of enkephalin microinjec- tions into hippocampus, Life Sci., 24 (1979) 541-548.

8 Elazar, Z., Simantov, R. and Motles, E., Local electro- graphic effects of Leu-enkephalin in microinjections in the brain, Electroenceph. clin. Neurophysiol., 54 (1982) 91-95.

9 Frenk, H., Liban, A., Balamuth, R. and Urca, G., Opiate and nonopiate aspects of morphine induced seizures, Brain Research, 253 (1982) 253-261,

10 Frenk, H., McCarthy, B. and Liebeskind, J. C., Different brain areas mediate the analgesic and epileptic properties of enkephalin, Science, 200 (1978) 335-337.

11 Frenk, H., Urca, G. and Liebeskind, J. C., Epileptic prop- erties of leucine- and methionine-enkephalin: comparison with morphine and reversibility by naloxone, Brain Re- search, 147 (1978) 327-337.

12 Frenk, H., Watkins, L. R., Miller, J. and Mayer, D. J., Nonspecific convulsions are induced by morphine but not n-Ala2-methionine-enkephalinamide at cortical sites, Brain Research, in press.

13 Havemann, U. and Kuschinsky, K., Neurochemical as- pects of the opioid-induced 'catatonia', Neurochem. Int., 4 (1982) 199-215.

14 Henriksen, S. J., Bloom, F. E., McCoy, F., Ling, N. and Guillemin, R., fl-Endorphin induces nonconvulsive limbic seizures, Proc. nat. Acad. Sci. U.S.A., 75 (1978) 5221-5225.

15 Hill, R. G., Mitchell, J. F. and Pepper, C. M., The excita-

tion and depression of hippocampal neurones by iontopho- retically applied enkephalins, J. Physiol. (Lond.), 272 (1976) 50-51.

16 Koppanyi, T. and Karczmar, A. G., Nature of antagonism between N-allylnormorphine (Nalline) and morphine, Fed. Proc., 12 (1953) 337.

17 McGuigan, N. and Ross, E. L., Intraspinal administration of morphine, J. Amer. med. Assoc., 64 (1915) 1494-1495.

18 Neal, H. and Keane, P. E., The effects of local microinjec- tions of opiates and enkephalins into the forebrain on the electrocorticogram of the rat, Electroenceph. din. Neuro- physiol., 45 (1978) 655-665.

19 Nicoll, R. A., Siggins, G. R., Ling, N., Bloom, F. E. and Guillemin, R., Neuronal actions of endorphins and en- kephalins among brain regions: a comparative microionto- phoretic study, Proc. nat. Acad. Sci. U.S.A., 74 (1977) 2584-2588.

20 Pinsky, C., Ova, A. K. and LaBella, F. S., Peptidase inhibi- tors reduce opiate narcotic withdrawal signs, includ!ng sei- zure activity, in the rat, Brain Research, 243 (1982) 301-307.

21 Racine, R. J., Modification of seizure activity by electrical stimulation. II. Motor seizure, Electroenceph. din. Neuro- physiol., 32 (1972) 281-294.

22 Shohami, E., Evron, S., Carmon, A. and Weinstock, M., Action myoclonus induced in rats by intrathecal morphine, lsr. J. Med., in press.

23 Snead III, O. C. and Bearden, L. J., The epileptogenic spectrum of opiate agonists, Neuropharmacol., 21 (1982) 1137-1144.

24 Snyder, E. W., Shearer, D. E., Beck, E. C. and Dustman, R. E., Naloxone-induced electrographic seizures in the pri- mate, Psychopharmacol., 67 (1980) 211-214.

25 Teitelbaum, H., Blosser, J. and Catravas, G., Bilateral electroencephalographic response and unilateral tolerance to unilateral intracerebral morphine injections, Nature (Lond.), 260 (1976) 158-159.

26 Tortella, F. C., Cowan, A. and Adler, M. W., Comparison of the anticonvulsant effects of opioid peptides and etor- phine in rats after ICV administration, Life Sci., 10 (1981) 1039-1045.

27 Tortella, F. C., Moreton, J. E. and Khazan, N., Electroen- cephalographic and behavioral tolerance to and cross-toler- ance between D-Ala2-methionine-enkephalinamide and morphine in the rat, J. Pharmacol. exp. Ther., 210 (1979) 174-179.

28 Urca, G. and Frenk, H., Systemic morphine blocks the sei- zures induced by intracerebroventricular (ICV) injections of opiates and opioid peptides, Brain Research, 246 (1982) 121-126.

29 Urca, G. and Frenk, H., Intracerebral opiates block the ep- ileptic effect of intracerebroventricular (ICV) leucine-en- kephaline, Brain Research, 259 (1983) 103-110.

42

30 Urca, G., Frenk, H., Liebesi~ind, J. C. and Taylor, A. N., Morphine and enkephalin: analgesic and epileptogenic properties, Science, 197 (1977) 83-86.

31 Watkins, L. R. and Mayer, D. J., Involvement of spinal opioid systems in footshock induced analgesia: antagonism by naloxone is possible only before induction of analgesia, Brain Research, 242 (1982) 309-316.

32 Werz, M. A. and MacDonald, R. L., Opiate alkaloids an- tagonize postsynaptic glycine and GABA responses: corre- lation with convulsant action, Brain Research, 236 (1982) 107-119.

33 Winter, C. A. and Flataker, L., The effect of cortisone, de-

34

35

soxycorticosterone, and adrenocorticotrophic hormone upon the responses of animals to analgesic drugs, J. Phar- macol, exp. Ther., 103 (1951) 93-105. Yeung, J. C., Yaksh, T. L. and Rudy, T. A., Effect on the nociceptive threshold and EEG activity in the rat of mor- phine injected into the medial thalamus and the periaque- ductal gray, Neuropharmacol., 17 (1978) 525-532. Zieglg/insberger, W., French, E. D., Siggins, G. R. and Bloom, F. E., Opioid peptides may excite hippocampal py- ramidal neurons by inhibiting adjacent inhibitory interneu- rons, Science, 205 (1979) 415-417.