- Email: [email protected]
Effects of midbrain raphe lesions or systemic p-chlorophenylalanine on the development of kindled seizures in rats
Brain Research Bulletin. Vol. 4, pp. 1-7. Printed in the U.S.A.
Effects of Midbrain Raphe Lesions or Systemic p=Chlorophenylalanitie on the Development of Kindled Seizures in Rats’ RONALD
Department
of Psychology,
RACINE2
McMaster
University, Hamilton,
Ontario
AND DONALD
Section of Biopsychology,
V. COSCINA
Clarke Institute of Psychiatry, (Received
University of Toronto, Toronto, Ontario
7 August 1978)
RACINE, R. AND D. V. COSCINA. Effects of midbrain raphe lesions or systemic gchlorophenylalanine on the developof kindled seizures in rats. BRAIN RES. BULL. 4(l) l-7, 1979.-Previous research has suggested that brain serotonin (5hydroxytryptamine or 5-HT) neurons inhibit epileptiform seizure activity. To test further this possibility, experiments were performed to determine if brain S-HT depletion would enhance the occurrence and/or magnitude of seizures “kindled” from the amygdala or neocortex of rats. Two modes of 5-HT depletion were used: (1) radiofrequency heat lesions of the midbrain dorsal and median raphe nuclei, and (2) systemic injection of the 5-HT synthesis inhibitor, p-chlorophenylalanine (pCPA). Both modes of 5-HT depletion reliably enhanced the strength of motor convulsions kindled from the cortex. Systemic pCPA also reduced the duration of after-discharges (ADS) in cortically-stimulated rats. However, pCPA reduced rather than enhanced convulsions kindled from the amygdala. In contrast to this, raphe lesions appeared to sensitize rats to the effects of amygdaloid kindling, i.e., lesions lowered AD thresholds, AD durations and number of ADS to elicit motor convulsions. Viewed together, these data support the hypothesis that 5-HT neurons can serve to inhibit seizures. However, the lack of robustness across parameters of epileptogenesis as well as discrepant findings related to 5-HT depletion mode additionally suggest that kindled seizures affect other neuronal populations in addition to those under serotonergic intluence.
ment
Serotonin
Kindling
Amygdala
Neocortex
p-Chlorophenylalanine
Rat
CONSIDERABLE research has been done in attempts to determine the neurophysiological and neurochemical bases of epileptogenesis. One experimental model being studied for this purpose is the “kindled” epilepsy induced by repeated low-intensity electrical stimulations spaced, applied through electrodes implanted into brains of rats. The magnitude of both electrographic and behavioral seizures triggered by such stimulations gradually increases in a highly reliable fashion [21, 32, 33, 411. Epileptiform discharges, initially localized to the stimulated site, develop sufficient strength to recruit discharge activity at other brain sites [7, 32,35,37]. At the same time, test responses evoked between affected sites by single square-wave pulses show permanently increased amplitudes [19, 34, 36, 371. Available evidence suggests that kindled epilepsy results from either increased synaptic conduction in excitatory neural systems or
Brain
Raphe lesions
Seizures
decreased synaptic efficiency in inhibitory neural systems [4, 18, 20, 371. One research strategy currently used to characterize neural mechanisms controlling kindling is the assessment of putative neurotransmitter systems seemingly participating in this phenomenon. Thus far, brain cholinergic and catecholaminergic systems have received the most experimental attention. Systemic treatment with atropine, a cholinergic blocking agent, can retard the development of seizures elicited by both amygdaloid [4] and cortical (Racine et al., in preparation) stimulation. While these effects of atropine have been replicated several times in Sprague-Dawley rats (Arnold, 1977, personal communication), Corcoran and co-workers [ll] found no such effects on amygdaloid kindling in Long Evans rats. Reserpine, a drug which depletes catecholamines by blocking their neuronal storage,
‘Supported by funds from the National Research Council of Canada (N.R.C. Grant No. A8629 to R.R.) and the Clarke Institute of Psychiatry (D.V.C.). ‘Address reprint requests to the first author at: The Department of Psychology, McMaster University, 1280 Main Street West, Hamilton, Ontario, Canada L8S 4Kl.
Copyright 0 1979 ANKHO
International Inc.-0361-9230/79/010001-08$01.30/O
3
RACINE AND COSCINA
facilitates both amygdaloid [4] and cortical (Racine et cl/. , in preparation) kindling. Consistent with these findings, central injections of the neurotoxin, 6_hydroxydopamine, sufficient to destroy brain catecholamine nerve terminals, facilitate amygdaloid kindling [4,10]. The potential importance of brain serotonin (5hydroxytryptamine or 5-HT) systems in kindling has not been tested extensively. However, Kovaks and Zoll [27] reported that electrical stimulation of the median raphe nucleus, which is one site of origin for 5-HT fibers ascending to the forebrain, blocked kindled amygdaloid seizures induced by threshold stimulation. These data suggest an inhibitory role for brain S-HT in kindling. If this interpretation is correct, impairment of 5-HT neurotransmission should remove this source of inhibition and facilitate the development of kindled seizures. In two experiments reported here, we have tested this possibility in rats depleted of brain 5-HT: (1) chronically, by lesioning the midbrain dorsal and median raphe nuclei, and (2) acutely, by systemic injection of the tryptophan hydroxylase inhibitor, p-chlorophenylalanine (PCPA). METHOD
Animtr/.\. Forty-seven male hooded rats obtained from Canadian Breeding Farms and Laboratory (St. Eustach, Quebec) were used. Animals were housed singly, with ad lib access to food and water, on a 12-hr light, 12-hr dark cycle (lights on 0800 hr, temperature 74” F ? 3”) throughout the experiment. Surgery. All rats weighed 100-125 g at the time of surgery. Following a lo-min pre-treatment with atropine methyl nitrate (2 mg, IP), all animals were anesthetized with sodium pentobarbital (Nembutal; 50 mgikg, IP) and placed in a Kopf stereotaxic instrument. Following scalp incision and trephenation of the skull, 31 rats received radio-frequency heat lesions of both the dorsal and median raphe nuclei as previously described [13, 14, 421. Briefly, this entailed generating 55°C for 1 min each in both midbrain cell groupings at the following co-ordinates: with incisor bar 2.5 mm below interaural line, 0.35 mm posterior to interaural line, 0 mm lateral to midsaggital sinus, 5 and 7 mm below the dura mater. The remaining 16 rats received all of these surgical manipulations except that electrodes were not lowered into brains. Wounds were closed with 9 mm stainless steel wound clips. Approximately 4 weeks later, when rats weighed about 250 g, one lesioned rat was sacrificed by decapitation for fluorometric determinations of endogenous levels of NE, dopamine (DA), 5-HT, and 5-HT’s major metabolite, 5-hydroxyindoleacetic acid (5-HIAA) (see [ 161 for details of assay procedures). All remaining rats were again anesthetized with Nembutal, placed in a stereotaxic instrument, and implanted bilaterally with bipolar twisted nichrome wire electrodes into (a) the basolateral region of the amygdala (15 raphe-lesioned and 8 sham-operated rats), or (b) area 2 of the anterior neocortex (15 raphe-lesioned and 8 sham-operated rats) as previously described [32]. For the next 5 days all animals were handled 5-10 min per day. Kindling. Six days after implantation, after-discharge (AD) thresholds were measured for all rats. This was accomplished by applying a 1 set train of 1 msec biphasic square-wave pulses (60 Hz) of 20 PA intensity to the amyg-
daIa or cortex via electrodes implanted into right hemispheres. If ADS could not be triggered, current levels were raised by 20 PA and stimulation was reapplied 2 min later. This procedure was continued until an AD was triggered or until 100 PA intensity was attained. At that time, current levels were increased by 50 PA increments before subsequent stimulations. The first current intensity sufficient to trigger one AD was taken as the absolute threshold. Kindling treatment proper was then begun by stimulating all rats with a 1 set train of biphasic square-wave pulses (60 Hz) applied to right hemisphere electrodes. Peak-to-peak intensity was set at 800 PA which was sufficient to evoke epileptiform discharge in all rats. This stimulation was applied once daily until generalized convulsions appeared. The development of motor seizures by repeated amygdaloid stimulation was rated according to the following 5-point scale (see [32] for detailed description): (1) mouth or face movements, (2) clonic head movements, (3) forelimb clonus, (4) rearing (seizure-driven), (5) rearing followed by loss of postural control. The development of neocorticallydriven convulsions is more variable [33]. Nevertheless, the following 5-point scale, which describes such seizure development reasonably well, was employed: (1) head clonus, (2) unilateral forelimb clonus, (3) bilateral forelimb clonus, (4) turning or leaning of anterior trunk, or (5) loss of postural control and lying on side or back, typically in mild extension. Tcwnincltim. Upon completion of all kindling treatments, rats were sacrificed with an overdose of Nembutal. Brains were removed, fixed in formalin, sectioned and stained with thionin. The location of electrode tips as well as the extent of midbrain lesions were determined by light microscopy.
Auirrur/s. Fifty-three rats of the same sex and strain as described in Experiment 1 were used. Electrodes were implanted bilaterally into the amygdala (n=32) or anterior neocortex (n=21) and animals were housed as described before. After 5 days of handling, pCPA was mixed in several drops of Tween 80, suspended in saline and administered IP at doses of 150 or 300 mg/kg to 12 amygdaloid and 7 cortical animals per dose in the same volume of vehicle. The remaining 8 amygdaloid and 7 cortical rats received equal volumes of the vehicle alone. Kindling. Forty-eight hr after pCPA injection, kindling stimulation commenced. Amygdala stimulation as described in Experiment 1 was applied once every 2 hr 4 times daily for 3 days. This massed stimulation was employed so that all kindling treatment would fall within the reported period of maximal 5-HT depletion [25]. Cortically-stimulated animals received the same treatment as described in Experiment 1 except that stimulations took place only once daily for 3 days. This longer interstimulus interval is necessary since stimulations applied at shorter intervals are known to weaken the cortical response and raise the AD threshold ([33]; Burnham and Racine, in preparation). In addition, we expected that 5-HT depletion induced by pCPA would strengthen cortical responses, hence effects would show more clearly during weaker initial discharges. Stcltistictrl ~I~IN/~~sc~.s. Given the wide variety of measurement scales employed, non-parametric Mann-Whitney U analyses were used for all comparisons. As there were many statistical comparisons made, we considered pcO.02 (twotailed) confidence intervals as minimally significant.
BRAIN SEROTONIN
AND KINDLED
3
SEIZURES TABLE
1
COMPARISON OF AFTER-DISCHARGE (AD) VARIABLES AND CONVULSIVE STRENGTH INDUCED BY CORTICAL (A) OR AMYGDALOID (B) KINDLING IN RAPHE-LESIONED VS SHAM-OPERATED CONTROL RATS A. Cortex
AD Thresholds (PA) Day 20
AD Durations (set) Day 1 Day 10 Day 20
Convulsive Strength (rating scale)§ Day 1 Day 10 Day 20
Control (n=6)
mean rt
148.6 (40-600)
10.3 (7-14)
9.5 (615)
15.5 (11-20)
M$ r
2.0 (l-4)
3.5 (l-5)
3.5 (3-5)
Raphe Lesions (n=ll)
mean r
256.2 (40-800)
9.0 V-17)
9.5 (5-16)
14.7 (l&22)
M r
2.0 (l-4)
4.0 (l-5)
5.0 (4-5)
NS
NS
NS
NS
NS
NS
p
P
B. Amygdala AD Thresholds (PA)
AD Durations (see)
ADS to Generalized Convulsion (number)
Day 1
Final Day
Control (n=7)
mean r
100.0 (40-200)
14.67 (11-19)
102.1 (64-125)
12.83 (9-21)
Raphe Lesions (n=lO)
mean r
45.8 (2&80)
16.9 (7-28)
72.0 (42-129)
8.27 (4-12)
NS
~~0.05 (NS)
p
pco.01
P
tRange. *Median. § l-5 point rating scale described in text.
RESULTS
In both experiments, the distribution of electrode tips through which kindled seizures were evoked was essentially the same as that reported previously (see [4] for amygdala placements; [33] for neocortical area 2 placements). Furthermore, there were no placement differences between any of the groups tested. Several animals lost electrodes and several others died before completing the experiments. Final numbers of animals available for statistical analyses are given in Table 1. Experiment
1
The extent of midbrain raphe lesions is shown in Fig. 1. The dorsal nucleus was nearly completely destroyed in all animals. All animals sustained some additional damage to portions of the surrounding central gray substance. In the one lesioned rat randomly selected for analysis of endogenous brain levels, 5-HT and S-HIAA were reduced to 3% and 58%, respectively, of normal values while NE and DA were within normal limits (90% and 92%, respectively, of control). These data fall within the range of values obtained by one of us in previous studies of rats receiving apparently specific 5-HT-depleting raphe lesions by the methods reported here [14, 22, 421. This same lesioning method has recently been reported to be without effect on brain regional acetylcholine and y-aminobutytic acid synthetic enzymes 1151.
Prior raphe lesions appeared to alter the development of seizures kindled from both the amygdala and the neocortex. More specifically, convulsive strength generally appeared greater in cortically-kindled lesioned rats (Table 1A). This trend was statistically reliable (pcO.01) on Day 20 of kindling. For amygdaloid kindled rats, AD thresholds were significantly @
2
No assays were performed on brains from rats receiving pCPA in this study. However, additional work (Coscina, unpublished observations) has confirmed the magnitude and time course of 5-HT depletion by the dosing regimens used here which agree with those originally reported by Koe and Weissman [25]. As in Experiment 1, cortically-stimulated animals that were depleted of 5-HT showed stronger convulsions than control animals (see Table 2A). The pCPA-treated animals showed stronger convulsions on all 3 test days, although differences by our statistical criterion were significant only on Days 1 and 3. AD durations tended to be shorter in cortically-stimulated pCPA-treated animals (p
RACINE AND COSCINA
-56 l
-6.0
-64 n
8 -68
-72 D FIG. 1. Reproductions of coronal midbrain sections depicting the full range of lesion damage in Experiment 1. Reading from left to right, the numbers (-5.6 to -7.2) preceding each row represent these sections’ distance in mm behind the bregma skull suture. Column A shows the smallest lesion in terms of raphe damage. Column B shows the lesion which produced the greatest raphe damage. Column C shows the largest lesion in terms of total tissue damage.
BRAIN SEROTONIN
AND KINDLED
5
SEIZURES TABLE 2
COMPARISON OF AFTER-DISCHARGE (AD) AND CONVULSIVE VARIABLES INDUCED BY CORTICAL KINDLING IN RATS RECEIVING INTRAPERITONEAL INJECTIONS OF p-CHLOROPHENYLALANINE
(A) OR AMYGDALOID OR ITS VEHICLE
(B)
A. Cortex Day 1
AD Durations (se@ Day 2 Day 3
Convulsive Strength (rating scale)5 Stimulaton 1 Stimulation 2
Stimulation 3
Saline
mean
10.71
9.43
9.57
MS
2.0
3.0
3.0
(n=7)
rt
(7-14)
(6-43)
(5-12)
r
(l-4
(2-4)
(2-4)
pCPA I50 mglkg (n=7)
mean
7.0
7.57
7.29
M
4.0
4.0
4.0
U-10)
(4-12)
(3-12)
r
(l-5)
(l-5)
(3-5)
pCPA 300 mgkg (n=7)
mean
7.43
9.86
6.14
M
4.0
4.0
4.0
(6-12)
(6-15)
(5-8)
r
(3-5)
(3-5)
(3-5)
~~0.01
NS
pcO.05 (NS)
r
r
(drugged vspsaline)
B
pco.02
pcO.05 (NS)
pco.02
Amygdala AD Durations (se@ Initial Final
Convulsion Strength 12th stimulation (rating scale)$
Saline
mean
11.17
43.83
M
3.5
(n=6)
r
(7-16)
(33-68)
r
(2-5)
pCPA 150 mg/kg (n=ll)
mean
13.36
56.45
M
3.0
r
(7-28)
(18-90)
r
(l-5)
pCPA 300 mg/kg (n= 10)
mean
11.40
39.40
M
2.0
r
U-18)
(18-90)
r
(l-5)
NS
NS
(drugged vs’saline)
p
Percentage of Generalized convulsions (frequency) 33% (216)
27% (3/l 1)
10% (i/10)
NS
tRange. *Median. $1-5 point rating scale described in text.
1). A similar though weaker trend was seen in the cortically-kindled Ss with raphe lesions (Table 1A). Unlike midbrain raphe lesions, systemic pCPA appeared to significantly retard the development of motor seizures in amygdaloid-kindled rats as measured by the strength of convulsions on the final stimulation (Table 2B). All other comparisons yielded no significant differences.
DISCUSSION
Kovaks and Zoll [27] reported that electrical stimulation of the median raphe nucleus could block kindled amygdaloid seizures when stimulation parameters were near threshold levels. In Experiment 1 we found that dorsal plus median raphe lesions not only facilitated the development of kindled amygdaloid convulsions but also reduced thresholds for amygdaloid ADS. Therefore it is possible that the blocking
effect of raphe stimulation on seizure development observed by Kovaks and Zoll was due to an increase in AD thresholds and/or a suppression of the discharge or discharge propagation. Both effects are consistent with the view that midbrain raphe nuclei serve as the origin for ascending, presumably serotonergic, inhibitory fibers. The effects of raphe lesions on cortical kindling is also consistent with this interpretation. Again, convulsion strength developed more rapidly in lesioned rats as compared to non-lesioned ones. The effects of the 5-HT synthesis inhibitor, pCPA, on cortical kindling were similar to the effects of raphe lesions. The development of significantly stronger convulsions after pCPA treatment supports an inhibitory role for S-HT neurons in kindling. Although the data collected from the use of other seizure models has been somewhat ambiguous, much of it also supports the notion that 5-HT systems inhibit seizure activity. For example, Schlesinger et ul. [39] reported that pCPA in-
RACINE AND COSCINA
6
creased the intensity of audiogenic seizures in mice. Jobe and co-workers [23] reported this same effect in rats. De La Torre ef ~1. [ 171as well as Alexander and Kopeloff [ 11 reported that pCPA exacerbated seizures triggered by pentylenetetrazol. Chen and co-workers [9] along with Koe and Weissman [26] reported that pCPA enhanced extensor convulsions triggered by electroconvulsive shock. The effects of pCPA on amygdaloid kindling, however, present a somewhat different picture. In Experiment 2 we found that pCPA retarded the development of convulsions by amygdala stimulation. Interestin~y enough, this “anticonvulsant” effect of pCPA has also been reported to occur on the development of audiogenic seizures [3]. In addition, Alexander and Kopeloff [2] have reported an additive effect of !I-hydroxytryptophan, a S-HT precursor, and pCPA, a S-NT depletor, in the prevention of audiogenic seizures. These results suggest that pCPA may have some anticonvulsant effects independent of its 5-HT depleting properties. The time course for these two events appears to differ [3]. There are other ambiguities in the literature, however, which suggest that the explanation for these discrepant results might not be this simpIe. For example, Browning et (11. [6] have reported different effects for 5-HT depletions produced by drugs as compared to lesions. More specifically, while pharmacological depletions of brain S-HT increased to pentylenetetrazol and maximal sensitivity the electroshock-induced seizures, similar 5-HT depletions produced by raphe lesions had no effect. On the other hand, Jobe and colleagues [24] have found that pharmacological manipulation of 5-HT has little or no effect on whole-brain electroshock seizures. In view of these conflicting data, other explanations for the different effects of pCPA and raphe lesions on amygdaloid kindling in our experiments should be mentioned. An obvious possibility is that the treatments (lesion vs drug) have different effects on 5-HT and related systems (see [ 12, 13, 281 for additional discussion). The lesion effects, for example, are permanent and confined to the brain, while pCPA effects are transient, affect peripheral organs as well as spinal portions of the CNS, and could conceivably be modified by repeated brain stimulation to alter the duration of enzyme inhibition. It is also possible that the different kindling procedures used in our two experiments may have interacted with the treatment conditions to produce different results. The kindling stimulations were, of necessity, massed during the pCPA experiments and may have altered the effect of pCPA administration.
Irritative effects of the lesion as compared to the drug treatment can probably be ruled out because radio-frequency heat lesions were used. On the other hand, most of the rats in Experiment 1 did sustain some damage to areas outside the raphe nuclei, particularly the central gray. It is possible that such additional damage may have interacted with the effects of damage inflicted upon the 5-HT neural axis. We have given considerable additional thought to the possible role of denervation supersensitivity effects in contributing to our findings. While pCPA apparently does not produce such supersensitivity [40], we know of no comparable assessment after midbrain raphe lesions. Therefore, it is not at all clear how lesion vs drug treatments here would compare with respect to potential post-synaptic activational effects. Another possible factor contributing to the differential effects of 5-HT depletion mode is that 5-HT neural systems appear to contain a mixture of excitatory and inhibitory subsystems. The existence of such a mix is supported by tindings from studies employing iontophoretic application of 5-HT to single cells (e.g., IS]). Conceivably then, the functional consequences of disrupting brain 5-HT neurons may depend upon the balance of activity remaining in these subsystems. This explanation would also account for some of the apparently contradictor data reported in the literature about the role of 5-HT in seizure phenomena. Notwithstanding these considerations, the majority of experiments reported in the literature as well as much of the data which we have reported here indicate that at least some inhibitory control over seizure development is exerted by brain 5-HT neurons. The widespread neural consequences of kindling make it unlikely that any one brain neurotransmitter system is uniquely responsible for seizure development l32,33]. Whatever the exact role of brain 5-HT neurons may be in the etiology of cortical or subcortical kindling, our results suggest that they are not robust. The same appears to be true of brain cholinergic systems [4,11]. On the other hand, catecholamine neurons may provide a relatively impo~ant source of inhibition over kindled seizures 14,101. In this respect, one neural change which may contribute to the development of kindling is progressive degeneration [37], a suggestion consistent with evidence of neuronal catecholamine depletions as a result of kindling [8,35]. Of course, there are still a large number of known and unknown brain transmitter systems which additionally may be involved in kindling. Undoubtedly, those systems will receive a great deal of experimental attention in future work on the kindling phenomenon.
REFERENCES 1. Alexander, G. .I. and L. M. Kopeloff. Metrazol seizures in rats: effects of p-chiorophenylalanine. Ercrk Res. 22: 231-235, 1970.
Alexander, G. J. and L. M. Kopeloff. Additive effect of 5-hydroxytryptophan and p-chlorophenylalanine in preventing audiogenic seizures in inbred mice. Psychophormrrc. Common. 2: 379.390, 1976. 1 Alexander, G. J.. L. M. Kopeloff and R. B. Alexander. Anticonvulsive effect of p-~hlorophenylaianine in audiosensitive mice. Life Sri. 10~877-882, 1971. 4. Arnold, P., R. Racine and R. Wise. Effects of atropine, reserpine, 6-hydroxydopamine, and handling on seizure development in the rat. Expl Neural. 40: 457-470, 1973. 5. Bradley, P. and A. Dray. Modification of the response of brain stem neurones to transmitter substances by anaesthetic agents. Br. J. Pharmrrc. 48: 212-224, 1973. 7 ‘_.
_
.
6. Browning, R. A., W. E. Hoffman and R. L. Simonton. Changes in seizure susceptibility after intracerebral treatment with 5,7dihydroxytryptamine: The role of serotonergic neurons. Ann. N. Y. Acd. Sci. 305: 437-456, 1978. 7. Burnham, W. M. Primary and “transfer”
seizure development in the kindled rat. Ctr~. J. Neural. Sci. 2: 417-428, 1975. 8. Callaghan, D. and W. Schwark. Neurochemical changes and drug effects in a model of epilepsy in the rat. Paper read at Sixth Annual Meeting of the Society for Neuroscience, Nov. 1976, Toronto, Canada (p. 257 of Abst. Vol. II, Part I, No. 363). 9. Chen, G., C. R. Ensor and B. Bohner. Drug effects on the disposition of active biogenic amines in the CNS. Life Sri. 7: 1063-1074, 1%8.
BRAIN SEROTONIN
AND KINDLED
M. E., H. C. Fibiger, J. C. McCaughran and J. A. Wada. Potentiation of amygdaloid kindling and metrazolinduced seizures by 6-hydroxydopamine in rats. Expl Neurol.
10. Corcoran,
45: 118-133, 1974. 11. Corcoran, M. E., J. A. Wada, A. Wake and H. Urstad. Failure of atropine to retard amygdaloid kindling. Exp/ Neural. 51: 271-275, 1976. 12. Coscina, D. V. Brain amines in hypothalamic obesity. In: Anorexia Nervosa, edited by R. A. Vigersky. New York: Raven Press, 1977, pp. 97-107. 13. Coscina, D. V. Effects of central 5,7_dihydroxytryptamine on the medial hypothalamic syndrome in rats. Ann. N.Y. Acad. Sci. 305: 627-644,
1978.
14. Coscina, D. V. and H. C. Stancer. Selective blockade of hypothalamic hyperphagia and obesity in rats by serotonindepleting midbrain lesions. Science 195: 416-419, 1977. 15. Coscina, D. V. and K. Cl. Lloyd. Acetylcholine synthetic capacity in brain regions from rats made hyperdipsic and hyperactive by serotonin-depleting raphe lesions. Paper read at 48th Annual Meeting of Eastern Psychological Association, April, 1977, Boston, MA (p. 189 of Abst. Vol). 16. Coscina, D. V., C. Rosenblum-Blinick, D. D. Godse and H. C. Stancer. Consummatory behaviors of hypothalamic hyperphagic rats after central injection of 6-hydroxydopamine. Pharmat. Biochem. Behav. 1: 629-642, 1973. 17. De La Terre, J. C., H. M. Kawanga and S. Mullen. Seizure susceptibility after manipulation of brain serotonin. Archs int. -. Pharmacodyn. 188: 291304, 1970. 18. Douglas, R. M. Long lasting perforant path potentiation with very brief tetanic bursts. Paper read at Sixth Annual Meeting of the Society for Neuroscience, Nov. 1976, Toronto, Canada (p. 821 of Abst. Vol. II, Part II, No. 1179). 19. Douglas, R. M. and Cl. Goddard. Long term potentiation of the perforant path granule cell synapse in the rat hippocampus. Brain Res. 86: 205-215, 1975. 20. Drucker-Colin, R. Seizure inhibition and REM sleep: a possible neurohumoral relationship. Behav. Biol. 12: 199209, 1974.
21. Goddard, G. V., D. McIntyre and C. Leech. A permanent change in brain function resultine from dailv electrical stimula. tion.-Expl Neurol. 25: 195-330, ib69. 22. Grant, L. D., D. V. Coscina, S. P. Grossman and D. X. Freedman. Muricide after serotonin depleting lesions of the midbrain raphe nuclei. Pharmac. Biochem. Behav. 1: 77-80, 1973. 23. Jobe, P., A. Picchioni and L. Chin. Role of brain 5hydroxytryptamine in audiogenic seizures in the rat. Life Sci. 13: 1-13, 1978. 24. Jobe, P., R. Stull and P. Geiges. The relative significance of norepinephrine, dopamine and 5-hydroxytryptamine in electroshock seizure in the rat. Neuropharmacology 13: 961-968, 1974.
25. Koe, B. K. and A. Weissman. p-chlorophenylalanine: a specific depletor of brain serotonin. J. Pharmac. exp. Ther. 154: 499516, 1966. 26. Koe, B. K. and A. Weissman. The pharmacology of parachlorophenylalanine, a selective depletor of serotonin stores. Adv.
Pharmac.
6B: 29-47,
7
SEIZURES
1968.
28. Lorens, S. A. Some behavioral effects of serotonin depletion depend on method: a comparison of 5,7_dihydroxytryptamine, p-chlorophenylalanine, p-chloroamphetamine and electrolytic raphe lesions. Ann. N. Y. Acad. Sci. 305: 532-555, 1978. 29. Maickle, R. P., R. H. Cox, J. Saillant and F. P. Miller. A method for the determination of serotonin and norepinephrine in discrete areas of rat brain. Int. J. Neuropharmac. 7: 275-281, 1968.
30. McNaughton, B. L. and C. A. Barnes. Physiological identitication and analysis of dentate granule cell responses to stimulation of the medial and lateral perforant paths. Paper read at Sixth Annual Meeting of the Society for Neuroscience, Nov. 1976, Toronto, Canada (p. 392 of Abst. Vol. II, Part I, No. 558). 31. Nascello, A. and E. Marichich. Effects of some cholinergic, adrenergic and serotonergic compounds, glutamic acid and GABA on hippocampal seizures. Pharmacology 9: 233-239, 1973.
32. Racine, R. Modification of seizure activity by electrical stimulation: II. Motor seizure. Electroenceph. c/in. Neurophysiol. 32: 281-294,
1972.
33. Racine, R. Modification of seizure activity of electrical stimulation: Cortical areas. Electroenceph. clin. Neurophysiol. 38: 1-12, 1975.
34. Racine, R., J. Gartner and W. Burnham. Epileptiform activity and neural plasticity in limbic structures. Brain Res. 47: 262268, 1972.
35. Racine, R., F. Newberry and W. Bumham. Post-activation potentiation and the kindling phenomenon. Electroenceph. c/in. Neurophysiol. 39: 261-271, 1975a. 36. Racine, R., L. Tuff and L. Zaide. Kindling, unit discharge patterns and neural plasticity. Can. J. Neurol. Sci. 2: 395-405, 1975b. 37. Racine, R. and J. Zaide. A further investigation into the mechanisms underlying the kindling phenomenon. In: The Continuing Evolution of the Limhic System, edited by K. Livingston and 0. Homykiewicz. New York: Plenum Press, 1978, (in press). 38. Sato, M. and T. Nakashima. Kindling: secondary epileptogenesis, sleep and catecholamines. Can. 1. Neural. Sci. 2: 446-459, 1975. 39. Schlesinger, K., W. Boggan and D. K. Freedman. Genetics of audiogenic seizures: II. Effects of pharmacological manipulation of brain serotonin, norepinephrine and gammaaminobutyric acid. Life Sci. 7: 437-447, 1%8. 40. Trulson, M. E. and B. L. Jacobs. Behavioral evidence of denervation supersensitivity after destruction of central serotonergic nerve terminals. Ann. N. Y. Acad. Sci. 305: 497-509, 1978. 41. Wada, J. A. and M. Sato. Generalized convulsive seizure induced by daily electrical stimulation of the amygdala in cats: correlative electrographic and behavioral features. Neurology 24: 565-574,
1974.
42. Warsh, J. J., D. V. Coscina and H. C. Stancer. Effect of midbrain raphe lesions or intracistemal6-hydroxydopamine administration on D,L-5-hydroxytryptophan-‘*C metabolism after peripheral decarboxylase inhibition. Brain Res. Bull. 1: 273278, 1976.
27. Kovaks, D. A. and J. G. Zoll. Seizure inhibition by median raphe nucleus stimulation in rat. Brain Res. 70: 165-169, 1974.
NOTE ADDED IN PROOF Since this paper was submitted for publication, additional evidence has appeared for differential effects of pCPA vs midbrain raphe lesions on the development of seizures. More specifically, systemic pCPA enhanced audiogenic seizure intensity in rats while dorsal or median raphe lesions were without effect (Jerlicz, M., W. Kostowski, A. Bidzinski and M. Hauptmann. Audiogenic seizure susceptibility in rats with lesioned raphe nuclei and treated with p-chlorophenylalanine. Pal. J. Pharmacol. Pharm. 30: 63-68, 1978).
Effects of Midbrain Raphe Lesions or Systemic p=Chlorophenylalanitie on the Development of Kindled Seizures in Rats’ RONALD
Department
of Psychology,
RACINE2
McMaster
University, Hamilton,
Ontario
AND DONALD
Section of Biopsychology,
V. COSCINA
Clarke Institute of Psychiatry, (Received
University of Toronto, Toronto, Ontario
7 August 1978)
RACINE, R. AND D. V. COSCINA. Effects of midbrain raphe lesions or systemic gchlorophenylalanine on the developof kindled seizures in rats. BRAIN RES. BULL. 4(l) l-7, 1979.-Previous research has suggested that brain serotonin (5hydroxytryptamine or 5-HT) neurons inhibit epileptiform seizure activity. To test further this possibility, experiments were performed to determine if brain S-HT depletion would enhance the occurrence and/or magnitude of seizures “kindled” from the amygdala or neocortex of rats. Two modes of 5-HT depletion were used: (1) radiofrequency heat lesions of the midbrain dorsal and median raphe nuclei, and (2) systemic injection of the 5-HT synthesis inhibitor, p-chlorophenylalanine (pCPA). Both modes of 5-HT depletion reliably enhanced the strength of motor convulsions kindled from the cortex. Systemic pCPA also reduced the duration of after-discharges (ADS) in cortically-stimulated rats. However, pCPA reduced rather than enhanced convulsions kindled from the amygdala. In contrast to this, raphe lesions appeared to sensitize rats to the effects of amygdaloid kindling, i.e., lesions lowered AD thresholds, AD durations and number of ADS to elicit motor convulsions. Viewed together, these data support the hypothesis that 5-HT neurons can serve to inhibit seizures. However, the lack of robustness across parameters of epileptogenesis as well as discrepant findings related to 5-HT depletion mode additionally suggest that kindled seizures affect other neuronal populations in addition to those under serotonergic intluence.
ment
Serotonin
Kindling
Amygdala
Neocortex
p-Chlorophenylalanine
Rat
CONSIDERABLE research has been done in attempts to determine the neurophysiological and neurochemical bases of epileptogenesis. One experimental model being studied for this purpose is the “kindled” epilepsy induced by repeated low-intensity electrical stimulations spaced, applied through electrodes implanted into brains of rats. The magnitude of both electrographic and behavioral seizures triggered by such stimulations gradually increases in a highly reliable fashion [21, 32, 33, 411. Epileptiform discharges, initially localized to the stimulated site, develop sufficient strength to recruit discharge activity at other brain sites [7, 32,35,37]. At the same time, test responses evoked between affected sites by single square-wave pulses show permanently increased amplitudes [19, 34, 36, 371. Available evidence suggests that kindled epilepsy results from either increased synaptic conduction in excitatory neural systems or
Brain
Raphe lesions
Seizures
decreased synaptic efficiency in inhibitory neural systems [4, 18, 20, 371. One research strategy currently used to characterize neural mechanisms controlling kindling is the assessment of putative neurotransmitter systems seemingly participating in this phenomenon. Thus far, brain cholinergic and catecholaminergic systems have received the most experimental attention. Systemic treatment with atropine, a cholinergic blocking agent, can retard the development of seizures elicited by both amygdaloid [4] and cortical (Racine et al., in preparation) stimulation. While these effects of atropine have been replicated several times in Sprague-Dawley rats (Arnold, 1977, personal communication), Corcoran and co-workers [ll] found no such effects on amygdaloid kindling in Long Evans rats. Reserpine, a drug which depletes catecholamines by blocking their neuronal storage,
‘Supported by funds from the National Research Council of Canada (N.R.C. Grant No. A8629 to R.R.) and the Clarke Institute of Psychiatry (D.V.C.). ‘Address reprint requests to the first author at: The Department of Psychology, McMaster University, 1280 Main Street West, Hamilton, Ontario, Canada L8S 4Kl.
Copyright 0 1979 ANKHO
International Inc.-0361-9230/79/010001-08$01.30/O
3
RACINE AND COSCINA
facilitates both amygdaloid [4] and cortical (Racine et cl/. , in preparation) kindling. Consistent with these findings, central injections of the neurotoxin, 6_hydroxydopamine, sufficient to destroy brain catecholamine nerve terminals, facilitate amygdaloid kindling [4,10]. The potential importance of brain serotonin (5hydroxytryptamine or 5-HT) systems in kindling has not been tested extensively. However, Kovaks and Zoll [27] reported that electrical stimulation of the median raphe nucleus, which is one site of origin for 5-HT fibers ascending to the forebrain, blocked kindled amygdaloid seizures induced by threshold stimulation. These data suggest an inhibitory role for brain S-HT in kindling. If this interpretation is correct, impairment of 5-HT neurotransmission should remove this source of inhibition and facilitate the development of kindled seizures. In two experiments reported here, we have tested this possibility in rats depleted of brain 5-HT: (1) chronically, by lesioning the midbrain dorsal and median raphe nuclei, and (2) acutely, by systemic injection of the tryptophan hydroxylase inhibitor, p-chlorophenylalanine (PCPA). METHOD
Animtr/.\. Forty-seven male hooded rats obtained from Canadian Breeding Farms and Laboratory (St. Eustach, Quebec) were used. Animals were housed singly, with ad lib access to food and water, on a 12-hr light, 12-hr dark cycle (lights on 0800 hr, temperature 74” F ? 3”) throughout the experiment. Surgery. All rats weighed 100-125 g at the time of surgery. Following a lo-min pre-treatment with atropine methyl nitrate (2 mg, IP), all animals were anesthetized with sodium pentobarbital (Nembutal; 50 mgikg, IP) and placed in a Kopf stereotaxic instrument. Following scalp incision and trephenation of the skull, 31 rats received radio-frequency heat lesions of both the dorsal and median raphe nuclei as previously described [13, 14, 421. Briefly, this entailed generating 55°C for 1 min each in both midbrain cell groupings at the following co-ordinates: with incisor bar 2.5 mm below interaural line, 0.35 mm posterior to interaural line, 0 mm lateral to midsaggital sinus, 5 and 7 mm below the dura mater. The remaining 16 rats received all of these surgical manipulations except that electrodes were not lowered into brains. Wounds were closed with 9 mm stainless steel wound clips. Approximately 4 weeks later, when rats weighed about 250 g, one lesioned rat was sacrificed by decapitation for fluorometric determinations of endogenous levels of NE, dopamine (DA), 5-HT, and 5-HT’s major metabolite, 5-hydroxyindoleacetic acid (5-HIAA) (see [ 161 for details of assay procedures). All remaining rats were again anesthetized with Nembutal, placed in a stereotaxic instrument, and implanted bilaterally with bipolar twisted nichrome wire electrodes into (a) the basolateral region of the amygdala (15 raphe-lesioned and 8 sham-operated rats), or (b) area 2 of the anterior neocortex (15 raphe-lesioned and 8 sham-operated rats) as previously described [32]. For the next 5 days all animals were handled 5-10 min per day. Kindling. Six days after implantation, after-discharge (AD) thresholds were measured for all rats. This was accomplished by applying a 1 set train of 1 msec biphasic square-wave pulses (60 Hz) of 20 PA intensity to the amyg-
daIa or cortex via electrodes implanted into right hemispheres. If ADS could not be triggered, current levels were raised by 20 PA and stimulation was reapplied 2 min later. This procedure was continued until an AD was triggered or until 100 PA intensity was attained. At that time, current levels were increased by 50 PA increments before subsequent stimulations. The first current intensity sufficient to trigger one AD was taken as the absolute threshold. Kindling treatment proper was then begun by stimulating all rats with a 1 set train of biphasic square-wave pulses (60 Hz) applied to right hemisphere electrodes. Peak-to-peak intensity was set at 800 PA which was sufficient to evoke epileptiform discharge in all rats. This stimulation was applied once daily until generalized convulsions appeared. The development of motor seizures by repeated amygdaloid stimulation was rated according to the following 5-point scale (see [32] for detailed description): (1) mouth or face movements, (2) clonic head movements, (3) forelimb clonus, (4) rearing (seizure-driven), (5) rearing followed by loss of postural control. The development of neocorticallydriven convulsions is more variable [33]. Nevertheless, the following 5-point scale, which describes such seizure development reasonably well, was employed: (1) head clonus, (2) unilateral forelimb clonus, (3) bilateral forelimb clonus, (4) turning or leaning of anterior trunk, or (5) loss of postural control and lying on side or back, typically in mild extension. Tcwnincltim. Upon completion of all kindling treatments, rats were sacrificed with an overdose of Nembutal. Brains were removed, fixed in formalin, sectioned and stained with thionin. The location of electrode tips as well as the extent of midbrain lesions were determined by light microscopy.
Auirrur/s. Fifty-three rats of the same sex and strain as described in Experiment 1 were used. Electrodes were implanted bilaterally into the amygdala (n=32) or anterior neocortex (n=21) and animals were housed as described before. After 5 days of handling, pCPA was mixed in several drops of Tween 80, suspended in saline and administered IP at doses of 150 or 300 mg/kg to 12 amygdaloid and 7 cortical animals per dose in the same volume of vehicle. The remaining 8 amygdaloid and 7 cortical rats received equal volumes of the vehicle alone. Kindling. Forty-eight hr after pCPA injection, kindling stimulation commenced. Amygdala stimulation as described in Experiment 1 was applied once every 2 hr 4 times daily for 3 days. This massed stimulation was employed so that all kindling treatment would fall within the reported period of maximal 5-HT depletion [25]. Cortically-stimulated animals received the same treatment as described in Experiment 1 except that stimulations took place only once daily for 3 days. This longer interstimulus interval is necessary since stimulations applied at shorter intervals are known to weaken the cortical response and raise the AD threshold ([33]; Burnham and Racine, in preparation). In addition, we expected that 5-HT depletion induced by pCPA would strengthen cortical responses, hence effects would show more clearly during weaker initial discharges. Stcltistictrl ~I~IN/~~sc~.s. Given the wide variety of measurement scales employed, non-parametric Mann-Whitney U analyses were used for all comparisons. As there were many statistical comparisons made, we considered pcO.02 (twotailed) confidence intervals as minimally significant.
BRAIN SEROTONIN
AND KINDLED
3
SEIZURES TABLE
1
COMPARISON OF AFTER-DISCHARGE (AD) VARIABLES AND CONVULSIVE STRENGTH INDUCED BY CORTICAL (A) OR AMYGDALOID (B) KINDLING IN RAPHE-LESIONED VS SHAM-OPERATED CONTROL RATS A. Cortex
AD Thresholds (PA) Day 20
AD Durations (set) Day 1 Day 10 Day 20
Convulsive Strength (rating scale)§ Day 1 Day 10 Day 20
Control (n=6)
mean rt
148.6 (40-600)
10.3 (7-14)
9.5 (615)
15.5 (11-20)
M$ r
2.0 (l-4)
3.5 (l-5)
3.5 (3-5)
Raphe Lesions (n=ll)
mean r
256.2 (40-800)
9.0 V-17)
9.5 (5-16)
14.7 (l&22)
M r
2.0 (l-4)
4.0 (l-5)
5.0 (4-5)
NS
NS
NS
NS
NS
NS
p
P
B. Amygdala AD Thresholds (PA)
AD Durations (see)
ADS to Generalized Convulsion (number)
Day 1
Final Day
Control (n=7)
mean r
100.0 (40-200)
14.67 (11-19)
102.1 (64-125)
12.83 (9-21)
Raphe Lesions (n=lO)
mean r
45.8 (2&80)
16.9 (7-28)
72.0 (42-129)
8.27 (4-12)
NS
~~0.05 (NS)
p
pco.01
P
tRange. *Median. § l-5 point rating scale described in text.
RESULTS
In both experiments, the distribution of electrode tips through which kindled seizures were evoked was essentially the same as that reported previously (see [4] for amygdala placements; [33] for neocortical area 2 placements). Furthermore, there were no placement differences between any of the groups tested. Several animals lost electrodes and several others died before completing the experiments. Final numbers of animals available for statistical analyses are given in Table 1. Experiment
1
The extent of midbrain raphe lesions is shown in Fig. 1. The dorsal nucleus was nearly completely destroyed in all animals. All animals sustained some additional damage to portions of the surrounding central gray substance. In the one lesioned rat randomly selected for analysis of endogenous brain levels, 5-HT and S-HIAA were reduced to 3% and 58%, respectively, of normal values while NE and DA were within normal limits (90% and 92%, respectively, of control). These data fall within the range of values obtained by one of us in previous studies of rats receiving apparently specific 5-HT-depleting raphe lesions by the methods reported here [14, 22, 421. This same lesioning method has recently been reported to be without effect on brain regional acetylcholine and y-aminobutytic acid synthetic enzymes 1151.
Prior raphe lesions appeared to alter the development of seizures kindled from both the amygdala and the neocortex. More specifically, convulsive strength generally appeared greater in cortically-kindled lesioned rats (Table 1A). This trend was statistically reliable (pcO.01) on Day 20 of kindling. For amygdaloid kindled rats, AD thresholds were significantly @
2
No assays were performed on brains from rats receiving pCPA in this study. However, additional work (Coscina, unpublished observations) has confirmed the magnitude and time course of 5-HT depletion by the dosing regimens used here which agree with those originally reported by Koe and Weissman [25]. As in Experiment 1, cortically-stimulated animals that were depleted of 5-HT showed stronger convulsions than control animals (see Table 2A). The pCPA-treated animals showed stronger convulsions on all 3 test days, although differences by our statistical criterion were significant only on Days 1 and 3. AD durations tended to be shorter in cortically-stimulated pCPA-treated animals (p
RACINE AND COSCINA
-56 l
-6.0
-64 n
8 -68
-72 D FIG. 1. Reproductions of coronal midbrain sections depicting the full range of lesion damage in Experiment 1. Reading from left to right, the numbers (-5.6 to -7.2) preceding each row represent these sections’ distance in mm behind the bregma skull suture. Column A shows the smallest lesion in terms of raphe damage. Column B shows the lesion which produced the greatest raphe damage. Column C shows the largest lesion in terms of total tissue damage.
BRAIN SEROTONIN
AND KINDLED
5
SEIZURES TABLE 2
COMPARISON OF AFTER-DISCHARGE (AD) AND CONVULSIVE VARIABLES INDUCED BY CORTICAL KINDLING IN RATS RECEIVING INTRAPERITONEAL INJECTIONS OF p-CHLOROPHENYLALANINE
(A) OR AMYGDALOID OR ITS VEHICLE
(B)
A. Cortex Day 1
AD Durations (se@ Day 2 Day 3
Convulsive Strength (rating scale)5 Stimulaton 1 Stimulation 2
Stimulation 3
Saline
mean
10.71
9.43
9.57
MS
2.0
3.0
3.0
(n=7)
rt
(7-14)
(6-43)
(5-12)
r
(l-4
(2-4)
(2-4)
pCPA I50 mglkg (n=7)
mean
7.0
7.57
7.29
M
4.0
4.0
4.0
U-10)
(4-12)
(3-12)
r
(l-5)
(l-5)
(3-5)
pCPA 300 mgkg (n=7)
mean
7.43
9.86
6.14
M
4.0
4.0
4.0
(6-12)
(6-15)
(5-8)
r
(3-5)
(3-5)
(3-5)
~~0.01
NS
pcO.05 (NS)
r
r
(drugged vspsaline)
B
pco.02
pcO.05 (NS)
pco.02
Amygdala AD Durations (se@ Initial Final
Convulsion Strength 12th stimulation (rating scale)$
Saline
mean
11.17
43.83
M
3.5
(n=6)
r
(7-16)
(33-68)
r
(2-5)
pCPA 150 mg/kg (n=ll)
mean
13.36
56.45
M
3.0
r
(7-28)
(18-90)
r
(l-5)
pCPA 300 mg/kg (n= 10)
mean
11.40
39.40
M
2.0
r
U-18)
(18-90)
r
(l-5)
NS
NS
(drugged vs’saline)
p
Percentage of Generalized convulsions (frequency) 33% (216)
27% (3/l 1)
10% (i/10)
NS
tRange. *Median. $1-5 point rating scale described in text.
1). A similar though weaker trend was seen in the cortically-kindled Ss with raphe lesions (Table 1A). Unlike midbrain raphe lesions, systemic pCPA appeared to significantly retard the development of motor seizures in amygdaloid-kindled rats as measured by the strength of convulsions on the final stimulation (Table 2B). All other comparisons yielded no significant differences.
DISCUSSION
Kovaks and Zoll [27] reported that electrical stimulation of the median raphe nucleus could block kindled amygdaloid seizures when stimulation parameters were near threshold levels. In Experiment 1 we found that dorsal plus median raphe lesions not only facilitated the development of kindled amygdaloid convulsions but also reduced thresholds for amygdaloid ADS. Therefore it is possible that the blocking
effect of raphe stimulation on seizure development observed by Kovaks and Zoll was due to an increase in AD thresholds and/or a suppression of the discharge or discharge propagation. Both effects are consistent with the view that midbrain raphe nuclei serve as the origin for ascending, presumably serotonergic, inhibitory fibers. The effects of raphe lesions on cortical kindling is also consistent with this interpretation. Again, convulsion strength developed more rapidly in lesioned rats as compared to non-lesioned ones. The effects of the 5-HT synthesis inhibitor, pCPA, on cortical kindling were similar to the effects of raphe lesions. The development of significantly stronger convulsions after pCPA treatment supports an inhibitory role for S-HT neurons in kindling. Although the data collected from the use of other seizure models has been somewhat ambiguous, much of it also supports the notion that 5-HT systems inhibit seizure activity. For example, Schlesinger et ul. [39] reported that pCPA in-
RACINE AND COSCINA
6
creased the intensity of audiogenic seizures in mice. Jobe and co-workers [23] reported this same effect in rats. De La Torre ef ~1. [ 171as well as Alexander and Kopeloff [ 11 reported that pCPA exacerbated seizures triggered by pentylenetetrazol. Chen and co-workers [9] along with Koe and Weissman [26] reported that pCPA enhanced extensor convulsions triggered by electroconvulsive shock. The effects of pCPA on amygdaloid kindling, however, present a somewhat different picture. In Experiment 2 we found that pCPA retarded the development of convulsions by amygdala stimulation. Interestin~y enough, this “anticonvulsant” effect of pCPA has also been reported to occur on the development of audiogenic seizures [3]. In addition, Alexander and Kopeloff [2] have reported an additive effect of !I-hydroxytryptophan, a S-HT precursor, and pCPA, a S-NT depletor, in the prevention of audiogenic seizures. These results suggest that pCPA may have some anticonvulsant effects independent of its 5-HT depleting properties. The time course for these two events appears to differ [3]. There are other ambiguities in the literature, however, which suggest that the explanation for these discrepant results might not be this simpIe. For example, Browning et (11. [6] have reported different effects for 5-HT depletions produced by drugs as compared to lesions. More specifically, while pharmacological depletions of brain S-HT increased to pentylenetetrazol and maximal sensitivity the electroshock-induced seizures, similar 5-HT depletions produced by raphe lesions had no effect. On the other hand, Jobe and colleagues [24] have found that pharmacological manipulation of 5-HT has little or no effect on whole-brain electroshock seizures. In view of these conflicting data, other explanations for the different effects of pCPA and raphe lesions on amygdaloid kindling in our experiments should be mentioned. An obvious possibility is that the treatments (lesion vs drug) have different effects on 5-HT and related systems (see [ 12, 13, 281 for additional discussion). The lesion effects, for example, are permanent and confined to the brain, while pCPA effects are transient, affect peripheral organs as well as spinal portions of the CNS, and could conceivably be modified by repeated brain stimulation to alter the duration of enzyme inhibition. It is also possible that the different kindling procedures used in our two experiments may have interacted with the treatment conditions to produce different results. The kindling stimulations were, of necessity, massed during the pCPA experiments and may have altered the effect of pCPA administration.
Irritative effects of the lesion as compared to the drug treatment can probably be ruled out because radio-frequency heat lesions were used. On the other hand, most of the rats in Experiment 1 did sustain some damage to areas outside the raphe nuclei, particularly the central gray. It is possible that such additional damage may have interacted with the effects of damage inflicted upon the 5-HT neural axis. We have given considerable additional thought to the possible role of denervation supersensitivity effects in contributing to our findings. While pCPA apparently does not produce such supersensitivity [40], we know of no comparable assessment after midbrain raphe lesions. Therefore, it is not at all clear how lesion vs drug treatments here would compare with respect to potential post-synaptic activational effects. Another possible factor contributing to the differential effects of 5-HT depletion mode is that 5-HT neural systems appear to contain a mixture of excitatory and inhibitory subsystems. The existence of such a mix is supported by tindings from studies employing iontophoretic application of 5-HT to single cells (e.g., IS]). Conceivably then, the functional consequences of disrupting brain 5-HT neurons may depend upon the balance of activity remaining in these subsystems. This explanation would also account for some of the apparently contradictor data reported in the literature about the role of 5-HT in seizure phenomena. Notwithstanding these considerations, the majority of experiments reported in the literature as well as much of the data which we have reported here indicate that at least some inhibitory control over seizure development is exerted by brain 5-HT neurons. The widespread neural consequences of kindling make it unlikely that any one brain neurotransmitter system is uniquely responsible for seizure development l32,33]. Whatever the exact role of brain 5-HT neurons may be in the etiology of cortical or subcortical kindling, our results suggest that they are not robust. The same appears to be true of brain cholinergic systems [4,11]. On the other hand, catecholamine neurons may provide a relatively impo~ant source of inhibition over kindled seizures 14,101. In this respect, one neural change which may contribute to the development of kindling is progressive degeneration [37], a suggestion consistent with evidence of neuronal catecholamine depletions as a result of kindling [8,35]. Of course, there are still a large number of known and unknown brain transmitter systems which additionally may be involved in kindling. Undoubtedly, those systems will receive a great deal of experimental attention in future work on the kindling phenomenon.
REFERENCES 1. Alexander, G. .I. and L. M. Kopeloff. Metrazol seizures in rats: effects of p-chiorophenylalanine. Ercrk Res. 22: 231-235, 1970.
Alexander, G. J. and L. M. Kopeloff. Additive effect of 5-hydroxytryptophan and p-chlorophenylalanine in preventing audiogenic seizures in inbred mice. Psychophormrrc. Common. 2: 379.390, 1976. 1 Alexander, G. J.. L. M. Kopeloff and R. B. Alexander. Anticonvulsive effect of p-~hlorophenylaianine in audiosensitive mice. Life Sri. 10~877-882, 1971. 4. Arnold, P., R. Racine and R. Wise. Effects of atropine, reserpine, 6-hydroxydopamine, and handling on seizure development in the rat. Expl Neural. 40: 457-470, 1973. 5. Bradley, P. and A. Dray. Modification of the response of brain stem neurones to transmitter substances by anaesthetic agents. Br. J. Pharmrrc. 48: 212-224, 1973. 7 ‘_.
_
.
6. Browning, R. A., W. E. Hoffman and R. L. Simonton. Changes in seizure susceptibility after intracerebral treatment with 5,7dihydroxytryptamine: The role of serotonergic neurons. Ann. N. Y. Acd. Sci. 305: 437-456, 1978. 7. Burnham, W. M. Primary and “transfer”
seizure development in the kindled rat. Ctr~. J. Neural. Sci. 2: 417-428, 1975. 8. Callaghan, D. and W. Schwark. Neurochemical changes and drug effects in a model of epilepsy in the rat. Paper read at Sixth Annual Meeting of the Society for Neuroscience, Nov. 1976, Toronto, Canada (p. 257 of Abst. Vol. II, Part I, No. 363). 9. Chen, G., C. R. Ensor and B. Bohner. Drug effects on the disposition of active biogenic amines in the CNS. Life Sri. 7: 1063-1074, 1%8.
BRAIN SEROTONIN
AND KINDLED
M. E., H. C. Fibiger, J. C. McCaughran and J. A. Wada. Potentiation of amygdaloid kindling and metrazolinduced seizures by 6-hydroxydopamine in rats. Expl Neurol.
10. Corcoran,
45: 118-133, 1974. 11. Corcoran, M. E., J. A. Wada, A. Wake and H. Urstad. Failure of atropine to retard amygdaloid kindling. Exp/ Neural. 51: 271-275, 1976. 12. Coscina, D. V. Brain amines in hypothalamic obesity. In: Anorexia Nervosa, edited by R. A. Vigersky. New York: Raven Press, 1977, pp. 97-107. 13. Coscina, D. V. Effects of central 5,7_dihydroxytryptamine on the medial hypothalamic syndrome in rats. Ann. N.Y. Acad. Sci. 305: 627-644,
1978.
14. Coscina, D. V. and H. C. Stancer. Selective blockade of hypothalamic hyperphagia and obesity in rats by serotonindepleting midbrain lesions. Science 195: 416-419, 1977. 15. Coscina, D. V. and K. Cl. Lloyd. Acetylcholine synthetic capacity in brain regions from rats made hyperdipsic and hyperactive by serotonin-depleting raphe lesions. Paper read at 48th Annual Meeting of Eastern Psychological Association, April, 1977, Boston, MA (p. 189 of Abst. Vol). 16. Coscina, D. V., C. Rosenblum-Blinick, D. D. Godse and H. C. Stancer. Consummatory behaviors of hypothalamic hyperphagic rats after central injection of 6-hydroxydopamine. Pharmat. Biochem. Behav. 1: 629-642, 1973. 17. De La Terre, J. C., H. M. Kawanga and S. Mullen. Seizure susceptibility after manipulation of brain serotonin. Archs int. -. Pharmacodyn. 188: 291304, 1970. 18. Douglas, R. M. Long lasting perforant path potentiation with very brief tetanic bursts. Paper read at Sixth Annual Meeting of the Society for Neuroscience, Nov. 1976, Toronto, Canada (p. 821 of Abst. Vol. II, Part II, No. 1179). 19. Douglas, R. M. and Cl. Goddard. Long term potentiation of the perforant path granule cell synapse in the rat hippocampus. Brain Res. 86: 205-215, 1975. 20. Drucker-Colin, R. Seizure inhibition and REM sleep: a possible neurohumoral relationship. Behav. Biol. 12: 199209, 1974.
21. Goddard, G. V., D. McIntyre and C. Leech. A permanent change in brain function resultine from dailv electrical stimula. tion.-Expl Neurol. 25: 195-330, ib69. 22. Grant, L. D., D. V. Coscina, S. P. Grossman and D. X. Freedman. Muricide after serotonin depleting lesions of the midbrain raphe nuclei. Pharmac. Biochem. Behav. 1: 77-80, 1973. 23. Jobe, P., A. Picchioni and L. Chin. Role of brain 5hydroxytryptamine in audiogenic seizures in the rat. Life Sci. 13: 1-13, 1978. 24. Jobe, P., R. Stull and P. Geiges. The relative significance of norepinephrine, dopamine and 5-hydroxytryptamine in electroshock seizure in the rat. Neuropharmacology 13: 961-968, 1974.
25. Koe, B. K. and A. Weissman. p-chlorophenylalanine: a specific depletor of brain serotonin. J. Pharmac. exp. Ther. 154: 499516, 1966. 26. Koe, B. K. and A. Weissman. The pharmacology of parachlorophenylalanine, a selective depletor of serotonin stores. Adv.
Pharmac.
6B: 29-47,
7
SEIZURES
1968.
28. Lorens, S. A. Some behavioral effects of serotonin depletion depend on method: a comparison of 5,7_dihydroxytryptamine, p-chlorophenylalanine, p-chloroamphetamine and electrolytic raphe lesions. Ann. N. Y. Acad. Sci. 305: 532-555, 1978. 29. Maickle, R. P., R. H. Cox, J. Saillant and F. P. Miller. A method for the determination of serotonin and norepinephrine in discrete areas of rat brain. Int. J. Neuropharmac. 7: 275-281, 1968.
30. McNaughton, B. L. and C. A. Barnes. Physiological identitication and analysis of dentate granule cell responses to stimulation of the medial and lateral perforant paths. Paper read at Sixth Annual Meeting of the Society for Neuroscience, Nov. 1976, Toronto, Canada (p. 392 of Abst. Vol. II, Part I, No. 558). 31. Nascello, A. and E. Marichich. Effects of some cholinergic, adrenergic and serotonergic compounds, glutamic acid and GABA on hippocampal seizures. Pharmacology 9: 233-239, 1973.
32. Racine, R. Modification of seizure activity by electrical stimulation: II. Motor seizure. Electroenceph. c/in. Neurophysiol. 32: 281-294,
1972.
33. Racine, R. Modification of seizure activity of electrical stimulation: Cortical areas. Electroenceph. clin. Neurophysiol. 38: 1-12, 1975.
34. Racine, R., J. Gartner and W. Burnham. Epileptiform activity and neural plasticity in limbic structures. Brain Res. 47: 262268, 1972.
35. Racine, R., F. Newberry and W. Bumham. Post-activation potentiation and the kindling phenomenon. Electroenceph. c/in. Neurophysiol. 39: 261-271, 1975a. 36. Racine, R., L. Tuff and L. Zaide. Kindling, unit discharge patterns and neural plasticity. Can. J. Neurol. Sci. 2: 395-405, 1975b. 37. Racine, R. and J. Zaide. A further investigation into the mechanisms underlying the kindling phenomenon. In: The Continuing Evolution of the Limhic System, edited by K. Livingston and 0. Homykiewicz. New York: Plenum Press, 1978, (in press). 38. Sato, M. and T. Nakashima. Kindling: secondary epileptogenesis, sleep and catecholamines. Can. 1. Neural. Sci. 2: 446-459, 1975. 39. Schlesinger, K., W. Boggan and D. K. Freedman. Genetics of audiogenic seizures: II. Effects of pharmacological manipulation of brain serotonin, norepinephrine and gammaaminobutyric acid. Life Sci. 7: 437-447, 1%8. 40. Trulson, M. E. and B. L. Jacobs. Behavioral evidence of denervation supersensitivity after destruction of central serotonergic nerve terminals. Ann. N. Y. Acad. Sci. 305: 497-509, 1978. 41. Wada, J. A. and M. Sato. Generalized convulsive seizure induced by daily electrical stimulation of the amygdala in cats: correlative electrographic and behavioral features. Neurology 24: 565-574,
1974.
42. Warsh, J. J., D. V. Coscina and H. C. Stancer. Effect of midbrain raphe lesions or intracistemal6-hydroxydopamine administration on D,L-5-hydroxytryptophan-‘*C metabolism after peripheral decarboxylase inhibition. Brain Res. Bull. 1: 273278, 1976.
27. Kovaks, D. A. and J. G. Zoll. Seizure inhibition by median raphe nucleus stimulation in rat. Brain Res. 70: 165-169, 1974.
NOTE ADDED IN PROOF Since this paper was submitted for publication, additional evidence has appeared for differential effects of pCPA vs midbrain raphe lesions on the development of seizures. More specifically, systemic pCPA enhanced audiogenic seizure intensity in rats while dorsal or median raphe lesions were without effect (Jerlicz, M., W. Kostowski, A. Bidzinski and M. Hauptmann. Audiogenic seizure susceptibility in rats with lesioned raphe nuclei and treated with p-chlorophenylalanine. Pal. J. Pharmacol. Pharm. 30: 63-68, 1978).