Electroshock- and pentylenetetrazol-induced seizures in genetically epilepsy-prone rats (GEPRs): differences in threshold and pattern

Epilepsy Res., 6 (1990) 1-11 Elsevier EPIRES 00312

Research Reports Electroshock- and pentylenetetrazol-induced seizures in genetically epilepsy-prone rats (GEPRs): differences in threshold and pattern R.A. Browning a, C. Wanga, M.L.

L a n k e r a and P.C. J o b e b

aDepartment of Physiology, ¢~:tthernII!im~isUniversitySchool of Medicine, Carbondale, IL 62901.6512 (U.S.A.), and bDepartment of Basic Sciences, Universityof Illinois Collegeof Medicine, Peoria, IL 61656 (U.S.A.) (Received 9 June 1989; revision received and accepted 18 September 1989)

Key words: Electroshock seizure; Pentylenetetrazol seizure; Genetically epilepsy-prone rat; Brain-stem; Forebrain; Epilepsy

Using facial and forelimb (F&F) clonus (a proposed forebrain marker) and running-bouncing (R/B) clonus and tonus (proposed brain-stem markers), the responsiveness of forebrain and brain-stem to electroshock or pentylenetetrazol seizures was assessed in GEPRs. The most striking finding was the failure of GEPR-9s to display F&F clonus in response to transcorneal electroshock at any stimulus intensity. Indeed, GEPR-9s displayed only R/B clonus or tonus indicative of brain-stem seizure discharge. GEPR-3s and nor. mal rats, on the other hand, displayed F&F clonus in response to the least effective electroshock stimulus, and R/B clonus and tonus at higher stimulus intensities. After treatment with phenytoin (50 mg/kg) to inhibit the tonic seizure, the least effective electroshock stimulus also produced F&F clonus in GEPR-9s. These findings suggest that the threshold for triggering brain-stem seizure discharge by electroshock is lower than that for triggering forebrain seizure discharge in GEPR-9s, whereas the reverse relationship is true in normal rats and GEPR-3s. The rank ordering of the electroshock thresholds was: normals > GEPR-3s > GEPR-9s. Both GEPR-3s and GEPR-9s were found to be hyper-responsive to pentylenetetrazol as evidenced by shorter latency for the tonic seizure and a greater seizure severity than normal rats. The rank ordering of seizure severity in response to pentylenetetrazol was: GEPR-9 > GEPR-3 > normal rats.

INTRODUCTION The genetically epilepsy-prone rat (GEPR) is of increasing interest as an experimental model of epilepsy 12'2°'24. Three types of seizure predisposition characterize these animals. First, they exhibit seizures in response to stimuli (i.e., sound and byCorrespondence to: Dr. R.A. Browning, Department of Physiology, Southern Illinois University School of Medicine, Carbondale, IL 62901, U.S.A.

perthermia) which fail to cause seizures in normal, non-epileptic rats24. Second, they manifest a low incidence of spontaneous seizures which rarely culminate in status epilepticus and death 9. Third, they are abnormally sensitive or overly responsive to seizures produced by stimuli which in higher doses or magnitudes also induce seizures ~n normal animals. For example, fewer applicatiom of intralimbic currents are required to produce class 5 kindled seizures in GEPRs than in non-epileptic rats26. Moreover, responses of GEPRs to supra-

0920-1211/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

maximal electroshock are more severe than those of neurologically normal controls 17. Finally, in one type of GEPR, the threshold to convulsions induced by intracerebroventricular morphine is abnormally low~. A few reports have previously suggested abnormally low electroshock and pentylenetetrazol seizure thresholds in the GEPR. Accordingly, (3EPRs from a colony developed at the University of Arizona (i.e., UAZ:AGS(SD) minimals) which are now extinct were reported to have reduced thresholds for both minimal (EST) and maximal (MES) electroshock seizures (see reviews by Reigel et al. 24and Laird and Jobe2°). Moreover, pentylenetetrazol seizure thresholds in these extinct GEPRs were observed to be abnormally low. More recent studies have revealed that in the extant GEPR colonies, the threshold for first appearance of EEG spiking caused by intravenous pentylenetetrazol is equal to that for control subjects (see Reigel et al.24), suggesting that the extant colonies are not unusually responsive to pentylenetetrazol. The present study was designed to provide the missing electroshock data and to elucidate the possible point of divergence between the earlier and more recent data on pentylenetetrazol. Another major feature of our present work was to examine further the role of both forebrain and brain-stem seizure circuitry as determinants of seizure predisposition in the (3EPR. Previous studies indicate that neuroanatomical factors partially determine whether convulsions will be restricted to facial and forelimb clonus or whether more generalized clonus or tonus will occur. Accordingly, convulsions characterized by facial and forelimb (F&F) clonus appear to emanate from neural substrates localized within the forebrain, while clonus of all 4 limbs with bouts of running and hopping (referred to as running/bouncing (R/B) clonus) as well as tonic convulsions are believed to emanate from the brain-stem L3'4. The striking similarity of the behavioral components of the seizures produced by minimal electroshock administered through corneal electrodes and the convulsions produced by intracerebral stimulation of limbic nuclei~ provides presumptive support for the concept that convulsions arising from either stimulus modality are driven by forebrain struc-

tures. These convulsions are similar to those produced in rats by the systemic administration of cholinergic agonists which have been described as 'limbic' seizures 29. In either case, the animals exhibit F&F clonus. In contrast, convulsions produced by direct stimulation of the brain-stem result in R/B clonus6'18. If progressively higher currents are used, increasing manifestations of tonic convulsions are exhibited. More direct evidence for the anatomical separation of facial and forelimb clonic convulsions from R/B clonus and tonus was provided by studies showing that rats are unable to display F&F clonus following a precollicular transection, but retain their ability to exhibit R/B clonus and tonus in response to electroshock or pentylenetetrazol 4. Interestingly, F&F clonus (limbic type seizures) cannot be produced (irrespective of the current) when the electroshock ~timulus is delivered with ear-clip electrodes. However, corneal electroshock stimulation can result in F&F clonus, R/B clonus, or tonus depending on the magnitude of the stimulating current 2. The latter findings support the hypothesis that transcorneal minimal electroshock (low current) stimulation activates forebrain seizure substrates preferentially, while minimal transauricular stimulation preferentially activates brain=stem seizure circuitry3'4. Thus, using different types or magnitudes of seizure initiating stimuli, we should be able to examine the seizure thresholds of different seizure substrates within the brain. In order to accomplish our goals relative to brain anatomy and seizure circuitry, we elected to compare the thresholds of the forebrain seizure substrate in normal Sprague-Dawley rats with those of GEPRs (Sprague-Dawley derived) using low current electroshock delivered through corneal electrodes. This approach enabled us to determine the threshold for F&F clonus. We chose to study seizure predisposition of the GEPR brain-stem circuitry through determinations of thresholds for R/B clonus and tonic extensor convulsions which occur in normal rats at currents higher than those required to produce F&F clonus. Since pentylenetetrazol-induced seizures begin with F&F clonus and progress, in a dose- and time-dependent manner, through R/B clonus to tonus, it appears that

this convulsant drug can sequentially activate the forebrain and brain-stem seizure substrates. Thus, we employed pentylenetetrazol as an additional means of assessing seizure predisposition in these 2 anatomically distinct seizure circuitries of the GEPR. Two substrains of GEPRs were used for our investigations: (1) moderate seizure animals designated as GEPR-3s; and (2) severe seizure subjects commonly referred to as GEPR-9s. The designation GEPR-3 was originally utilized because animals of this strain exhibit class 3 audiogenic seizures according to the ordinal rating scale of Jobe et al. 16. The GEPR-9 designation was first used to denote the innate propensity of severe seizure GEPRs to exhibit a class 9 audiogenic seizure (i.e., the severest form of audiogenic seizure defined by the ordinal rating scale). Once our experiments were underway, a difficult issue arose in our estimates of forebrain thresholds in GEPR-9s. Despite the use of very low stimulus intensities, F&F clonus failed to occur in these subjects. Rather, at the lowest effective seizure provoking intensities, brain-stem seizures resulted. These observations suggested that, unlike the situation which characterizes normal animals, the threshold for brain-stem seizures in GEPR-9s might be lower than those for forebrain seizures. Under these conditions, delivery of a stimulus sufficient to activate the forebrain would also initiate seizures within the brain-stem. Moreover, with simultaneous seizures within the 2 sites, brain-stem activity would probably obscure the overt manifestations of convulsions driven by the forebrain (see Kreindler et al. TMand Discussion). As a means of circumventing the special problem inherent in determining whether intact GEPR-9s can exhibit forebrain seizures in response to a classical minimal electroshock stimulus delivered through corneal electrodes, we elected to treat some of these animals with a drug which could be expected to elevate the seizure threshold of the brain-stem without a concomitant elevation of equal magnitude in the forebrain. Phenytoin was chosen for this purpose in our current experiments because of its known selective effects on tonic convulsions, and its reported absence of effect on electroshock-induced F&F clonus27.

METHODS Animals used in these studies were normal (seizure-resistant) Sprague-Dawley rats (225-350 g) obtained from Harlan-Sprague-Dawley, Madison, WI and genetically epilepsy-prone rats (GEPRs) from the colonies maintained at the University of Illinois College of Medicine at Peoria. The latter rats were obtained from either the moderate seizure colony which display running/bouncing (R/B) clonus in response to sound stimulation (called GEPR-3s) or the severe seizure colony, which display maximal tonic seizures (with tonic hind limb extension) in response to sound (called GEPR-9s). All GEPRs were tested for susceptibility to audiogenic seizures in a cylindrical chamber in which 2 door bells are mounted inside the lid. A sound of about 100 dB was delivered until the beginning of the convulsion or for a maximum of 60 sec. All rats were subjected to 3 screening tests and only those animals receiving a score of 9 (from severe seizure colony) or a score of 3 (from moderate seizure colony) on the last 2 tests were used in these studies. Normal (Sprague-Dawley) rats were also subjected to audiogenic screening tests, and none was found to convulse. The audiogenic seizure testing was carded out to ensure that the GEPRs employed in the present study were in fact seizure prone. Three to 6 weeks after the last audiogenic seizure electroshock seizures were induced, using a Whalquist electroshock apparatus and corneal ei¢ctrodes, the thresholds for facial and forelimb (F&F) clonus and R/B clonus were determined by evaluating the response to electroshock after 0.5 mA increments or decrements in stimulus intensity as previously described5. The interval between tests was 48 h. Pentylenetetrazol-induced seizures were elicited 3-6 weeks after the last audiogenic seizure by the intraperitoneal (i.p.) adminiztration of either 35 mg/kg or 50 mg/kg pentylenetetrazol (1 ml/kg). Animals were immediately placed in a circular plexiglass chamber and the time from injection until response (onset latency for the myoclonic jerk, generalized clonus, and tonus) was recorded. Rats were observed for a maximum of 15 min following pentylenetetrazol administration. Rats which

failed to display tonus in this time were assigned a tonic latency of 15 min. Additionally, the seizure severity was scored according to the following scale: 0 = no seizure; 1 = myoclonic jerks only; 2 = facial and forelimb clonus without loss of posture; 3 = clonus with twisting posture; 4 = running/bouncing clonus; 5 = tonic flexion (forelimb extension); 6 - tonic hind limb extension. Electroshock thresholds for individual responses (F&F clonus, R/B clonus and tonus) were compared between groups using a 2-way analysis of variance (ANOVA) with a Scheff6 test and the alpha level at 0.05. Pentylenetetrazol seizure latchties were compared using a 2-way ANOVA with a Tukey's post hoe test and seizure severity scores were compared between groups using a KruskalWallis non-parame tric AN OVA. RESULTS

Response to electroshock.induced convulsions Striking differences were observed between GEPR-9s, GEPR-3s and non-epileptic controls. Both male and female GEPR-9s failed to display F&F clonus at low currents (10-20 mA) and had R/B clonic and tonic thresholds far below the F&F clonic thresholds observed in normal rats and GEPR-3s (P < 0.05; Table I). It was indeed surprising to find GEPR-9s displaying R/B clonic and tonic seizures at stimulus intensities that had no overt effect in non-epileptic rats.

GEPR-3s were also more seizure prone than non-epileptic controls in all 3 electroshock tests employed (Table I). Accordingly, GEPR-3 thresholds were abnormally low for F&F clonus, R/B clonus and tonic forelimb extension. In these tests, the GEPR-3 thresholds were numerically located between those of the GEPR-9s and the nonepileptic controls. Moreover, the GEPR-3 thresholds were significantly different from those of either of these other 2 types of animal. Despite this intermediate position, the GEPR-3 thresholds appeared more similar to those of the controls than to those of the GEPR-9. Whereas GEPR-9s failed to exhibit F&F clonus, both GEPR-3s and controls did so. Also, the significantly reduced thresholds for R/B clonus and tonic forelimb extension for GEPR-3s differed from the corresponding thresholds of non-epileptic controls by only 19%, whereas those for GEPR-9s differed from the GEPR-3 values by approximately 71%.

Demonstration of electroshock-induced F&F clonus in GEPR-9s following phenytoin treatment Fig. 1 shows that 83% (P < 0.01 compared to control) of GEPR-9 males treated with 50 mg/kg of phenytoin (i.p.) 1 h earlier displayed F&F clonus when stimulated with 16 mA. The one rat that failed to exhibit F&F clonus displayed 'subthreshold stunning '3° suggesting that if a slightly higher stimulus intensity (e.g., 17 mA) had been used, it would also have exhibited F&F clonus. After pre-

TABLE !

Thresholds for electroshock.induced clonic and tonic convulsions N.E. -- not examined; M = male; F = female.

Rats

Sex

Threshold (mA ) F&F cionus

RIB clonus

Tonic

Normal GEPR-3 GEPR-9

M M M

23,95 _+ 0.3 (10) 19,50 + 0.1 (6) '~ Not displayed

38.9 _+ 2.7 (8) 31,5 ____0.4 (I0) a 8.7 _ 0.4 (6) a'b

39.9 + 2.7 (8) 32.5 + 0.5 (10) a 9.7 + 0.3 (6) a'b

Normal GEPR-9

F F

1 5 . 1 + 0 . 8 (6) c Not displayed

NE 6.2 + 0.2

NE 8.2+0.3

a Significantly different from normal. b Significantly different from GEPR-3. c Significantly different from normal male.

(5) c

(5) c

lOO m

=J I= 0 rJ

80

a

60

q, coincides with other tests wherein seizures occur more readily in the GEPR-9 than in either of the other 2 types of experimental subjects.

Response of GEPRs to pentylenetetrazol-induced seizures: comparison with non-epileptic rats

E 0 L

O O

40 2O

am

gh

0

15

20

50

Dose of Phenytoln (mglkg)

Fig, 1. Incidence of facial and forelimb (F&F) clonus produced by transcorneal electroshock in GEPR-9s treated with phenytoin. All rats were treated with sodium phenytoin (15, 20, or 50 mglkg) or saline vehicle (2 ml/kg) 1 h before stimulation with 16 mA electroshock. Six GEPR-9s received saline vehicle (0 mg/kg). Numbers inside bars indicate the number of rats in group. *P < 0.01 compared to vehicle control; *P < 0.05 compared to 15 or 20 mg/kg phenytoin groups using chi-square test. Rats in the 0, 15, or 20 mg/kg groups not showing F&F clonus displayed R/B elonus or tonic seizures.

treatment with 15-20 mg/kg of phenytoin, electrical stimulation caused F&F clonus in only 20% of GEPR-9s. Examination of the threshold for electroshock-induced F&F clonus in a group of young non-epileptic male rats showed that 50 mg/kg of phenytoin had no effect (treated = 20.6 + 0.6 mA (N = 9); untreated = 20.5 + 0.5 mA (N = 9)). If phenytoin does not exert an effect on the threshold for F&F clonus in GEPR-9s, our data provide a basis for estimating the forebrain threshold to be approximately 16 mA in the severe seizure GEPR9 males which have not been treated with an antieonvulsant drug. A forebrain seizure threshold of 16 mA is below that for GEPR-3 males (19 mA) and still further below that for non-epileptic controls (24 mA; see Table I). Consequently, a forward order gradation of the minimal electroshock threshold (F&F clonus) in GEPRs is apparent. Non-epileptic controls are seizure resistant and have the highest forebrain thresholds. GEPR-3s have an intermediate threshold and also exhibit other indices of a moderate degree of seizure predisposition. GEPR-9s have the lowest threshold for F&F clonus. This marked manifestation of exquisite CNS excitabili-

The response of non-epileptic rats to pentylenetetrazol always started with a myoclonic jerk and advanced to F&F clonus accompanied in most animals by rearing. Some non-epileptic animals proceeded through RIB clonus to tonic forelimb extension• However, normal rats never exhibited tonic hind limb extension (HLE) using 35 or 50 mg/kg of pentylenetetrazol given intraperitoneally. The nature of the convulsive responses of GEPRs to pentylenetetrazol diverged from those of the non-epileptic controls• Accordingly, GEPR-9s never displayed classical F&F clonus and GEPR-3s rarely displayed this behavior. In response to 35 mg/kg of pentylenetetrazol, the convulsion in GEPR-9s began with a twisting clonus (score = 3) and progressed rapidly through RIB clonus (score = 4) to tonus (score = 6). At 50 mg/kg, GEPR-9s progressed directly from class 3 to class 6 seizures without exhibiting RIB clonus. GEPR-3s displayed some F&F clonus and usually showed a running component (RIB clonus) just before showing tonic forelimb extension, especially at the 35 mg/kg dose of pentylenetetrazol. At the 50 mg/kg dose of pentylenetetrazol, most of the

[]

normal

[]

GEPR-3

[ ] GEPR-9

£ E --

10 8

6

Y,

4

Myoclonic Jerk

Clonus

Tonus

Fig. 2. Mean latency to various seizure components following 35 mg/kg pentylenetetrazol (i.p.). Numbers inside bars indicate the number of rats in group. Vertical bars indicate S.E.M. *P < 0.05 compared to tonic latency in normal rats using 2-way ANOVA with Tukey's post hoc test.

10 8

[]

normal

8

"

[ ] GEPR-3 [ ] GEPR-9

•

~

PTZ 35mo/kg

[]

PTZ S0mglkg

6.

c

4

="=

4

"J

2

2

Myocloni¢ Jerk

Clonus

Tonus

0 NORMAL

GEPR3

GEPR9

Fig, 3. Mean latency to various seizure components following 50 mg/kg pentylenetetrazol (i,p,). Numbers inside bars indicate the number of rats in group. Vertical bars indicate S.E,M. *P < 0.05 compared to tonic latency in normal rats using 2-way ANOVA with Tukey's post hoc test,

Fig. 4. Mean severity scores for pentylenetetrazol (PTZ)-induced seizures following 35 or 50 mg/kg PTZ. Numbers inside bars indicate the number of rats in group. Vertical bars indicate S.E.M. Note the absence of an S.E.M. in GEPR-9s receiving either dose FI'Z (all rats had a score of 6) and in GEPR-3s receiving 50 mg/kg (all rats had score of 5). **P < 0.05 compared to GEPR-3s and normals. *P < 0.05 compared to normal rats.

GEPR-3s also began with twisting clonus and progressed to the class 5 tonic convulsion, but unlike GEFR-9s, they usually displayed a short run just before they exhibited tonic forelimb extension and they did not display the HLE characteristic of class 6 seizures. As can be seen in Fig. 2, the latency to the onset of the myoclonic jerk or the cionic convulsion was not significantly different from that of control rats in GEPR-3s or GEPR-9s given 35 mg/kg pentyienetetrazol. However, both GEPR-3s and GEPR9s displayed a significantly shorter latency to the tonic convulsion when compared to normal rats (Fig. 2) given 35 mg/kg pentylenetetrazol. Furthermore, the seizure severity score of GEPR-9s given 35 mg/kg pentylenetetrazol was significantly greater than that of normal or GEPR-3s, and the severity of the seizure of GEPR-3s was greater than that in normal rats at the 35 mg/kg dose (Fig. 4). All GEPRs exhibited a tonic convulsion, whereas only 40% of the normal rats had a tonic seizure at 35 mg/kg pentylenetetrazol. At 50 mg/kg pentylenetetrazol the latencies for all seizure components (in all 3 groups) were shorter than at the lower dose (35 mg/kg; Fig. 3; P < 0.05 using ANOVA). However, the latency to the onset of the myoclonic jerk or clonus failed to differ significantly between the 3 groups (Fig. 3). As was the case at 35 mg/kg, the onset of the tonic seizure occurred significantly sooner in GEPR-3s

and GEPR-9s than in normal rats (Fig. 3). Moreover, the seizure severity after 50 mg/kg pentylenetetrazoi was significantly greater in GEPR-9s than in GEPR-3s and the severity scores in GEPR3s were greater than in normals (Fig. 4). DISCUSSION Our current observations with electroshock and pentylenetetrazol further solidify the concept that both GEPR-3s and GEPR-9s are characterized by an enhanced degree of seizure predisposition compared to non-epileptic controls. When interpreted in terms of the neuroanatomical hypotheses of seizure circuitry advanced by Browning ~'2 (i.e., that F&F clonic convulsions emanate from a seizure discharge within the forebrain, while R/B clonic and tonic convulsions emanate from the brainstem), our studies with electroshock suggest that the seizure thresholds of the forebrain and the brain-stem are abnormally low in GE~R-9s. Thresholds in the GEPR-3s are also reduced compared to non-epileptic control values, but they are not as low as the corresponding thresholds of GEPR-gs. In addition to the abnormally low electroshock thresholds identified for forebrain and brain-stem circuitries of both types of GEPRs, another unique dysfunction was documented in the GEPP.9. Whereas in non-epileptic rats and GEPR-3s the

electroshock thresholds of the forebrain are less than the brain-stem, the reverse relationship is found in the GEPR-9. Indeed, unlike the other 2 types of animals, the GEPR-9 does not exhibit the F&F clonus characteristic of forebrain seizures despite the application of currents below those required to produce the R/B clonus of brain-stem origin. An alternative hypothesis could be advanced to account for the failure of GEPR-9s to display F&F clonus in response to minimal electroshock. That is that the forebrain and brain-stem circuitrics are more integrated in GEPR-9s and are not as distinct as they are in normal rats and GEPR-3s. In this case activation of the forebrain circuit would always result in simultaneous brain-stem activation. However, the available evidence does not support this hypothesis and seems to argue against it. For example, it has been shown that GEPR-9s display F&F clonus in response to kindling26 (electrical forebrain stimulation), systemic kainic acid (Jobe, unpublished observation) and bicuculline infused into the deep prepiriform cortex (Browning, unpublished observations) without showing R/B clonus or tonus. Thus it is clear that activation of limbic forebrain seizures is possible in GEPR-9s without simultaneous activation of the brain-stem circuitry. These observations with electroshock provide support for the concept that the expression of F&F clonus will be overridden by the more intense manifestations of the brain-stem seizure in situations where the threshold of the more caudal substrate equal or falls below that of the forebrain~ The idea that the brain-stem would appear to dominate the motor expression of seizures under such circumstances has been suggested by Kreindler et al TM. From a neuroanatomical point of view, the translation of a forebrain seizure into convul~ve movements requires that the excessive and abnormal activity of the rostral brain be conveyed through the brain-stem and spinal cord to the appropriate muscles. We suggest that for this transfer to occur so that F&F clonus becomes the overt behavior, the brain-stem and spinal cord must retain some semblance of normal functional integrity. If the brain-stem neurons are engaged in seizure activity originating in that locus, the abnor-

mal, excessive discharges arriving from the forebrain would have little chance of doing more than contributing to seizure impulses already arising from within the more caudal structures. With major disruption of function already taking place in the brain-stem, we would not anticipate a coherent transfer of descending action potentials from rostral structures so that the facial and forelimb muscles would be selectively activated. Rather, if the seizure activity arriving from the forebrain were to have any effect on overt behavior of the animal, we would anticipate that it would be as a further contribution to the magnitude of the seizure activity already arising from the brain-stem. Thus, under conditions where the threshold for seizure activity within the brain-stern is lower than that of the forebrain, the minimal overt expression of the seizure will take the form produced by the brain-stem seizure. Interestingly, the least manifestation of seizures arising from the brain-stem or spinal cord are characterized by running movements. As would be expected, this behavior is produced by the smallest effective stimulus7. Progressively larger stimulus intensities or frequencies are required to produce R/B clonus and ultimately tonic extensor convulsions. Conversely, a stimulus sufficient to provoke tonic extensor convulsions will progressively lose this capacity in response to increasing doses of phenytoin or other anticonvulsant drugsa. Additional support for a disruption of the normal balance between forebrain and brain-stem thresholds in the GEPR-9 is derived from our experiment using phenytoin as a means of producing a normal relationship between forebrain and brain-stem thresholds. In non-epileptic animals, phenytoin has pronounced effects on tonic extensor convulsions, whereas it reportedly exerts no effects on facial and forelimb clonus27. Moreover, phenytoin has been shown to have significant effects on the brain-stem reticular formation 14'15. These factors led us to anticipate that appropriate doses of phenytoin in the GEPR-9 would normalize the balance between thresholds for forebrain and brain-stem by selectively seizures suppressing the appearance of seizure activity in the brain-stem. Itl accordance with these suppositions, the administration of phenytoin to GEPR-9s unmasked

the expression of facial and forelimb clonus following minimal electroshock. The threshold for these ¢orebrain seizures in GEFR-9s was belew that for GEPR-3s which in turn was lower than that for non-epileptic controls. Moreover, the estimated forebrain threshold in phenytoin-treated GEPR-9s was above the threshold for brain-stem seizures in GEPR-9s. These observations are compatible with the hypothesis that animals with brain-stem thresholds below those of the forebrain will exhibit convulsions characteristic of the brainstem rather than the forebrain in response to an electrical stimulus of sufficient magnitude to activate the more caudal substrate. Moreover, our findings are consistent with the concept that elevation of this current, so that it exceeds the threshold for the forebrain circuitry, will not cause the appearance of facial and forelimb clonus because of the domination of the brain-stem in controlling the convulsive pattern. Our present experiments with pentylenetetrazol extend those with electroshock by providing additional support for the concept that GEPRs are characterized by seizure predisposition. Both GEPR-3s and GEPR-9s exhibited ultrashort latencies to tonic convulsions and a higher incidence of such seizures in response to pentylenetetrazol. Moreover, following pentylenetetrazol, the mean seizure scores for both types of epileptic animals were abnormally high. Our pentylenetetrazol data also provided additional support for the concept that seizure predisposition in GEPR-9s is greater than in GEPR-3s. When treated with this convulsant drug, the maximal mean severity scores were least for non-epileptic controls, intermediate for GEPR-3s and greatest for GEPR-9s. Despite these major areas of agreement between the electroshock and chemoconvulsant parts of our present investigation, some conceptual incongruity does surface, namely, when two of the responses in the pentylenetetrazol experiments are compared with the electroshock study. The first of these partial disparities between the electroshock and pentylenetetrazol experiments occurred when we compared the relative magnitudes of forebrain thresholds in GEPRs with nonepileptic controls. The electroshock experiment clearly indicated that the forebrain threshold of

GEPR-3s and GEPR-9s was lower than that of normal rats. As a result, we anticipated that the pentylenetetrazol experiment would also provide unequivocal support for this concept. Yet, this prediction was only partially confirmed. Experimentally, forebrain activation in GEPRs begins with a class 3 response rather than seizures of lower rank as it does in non-epileptic rats. Partial incongruity occurred because the pentylenetetrazolinduced myoclonic jerk and F&F clonus (indicators of forebrain seizures) did not occur sooner in GEPR-3s and GEPR-9s than in normal rats. The second potential disparity occurred when in GEPR-gs we compared the brain-stem threshold with the foreb~ain threshold. In the electroshock study, the brain-stem seizure threshold was lower than the threshold of the forebrain circuitry. Again, we anticipated that the pentylenetetrazol experiment would provide additional support for this idea. But, the data yielded only partial verification. Supporting the concept was the observation that GEPR-9s failed to display F&F clonus at either dose of pentylenetetrazol. Lack of confirmation derived from the observation that pentylenetetrazol-induced seizures in GEPR-gs did not begin with running or R/B clonus. In lieu of these anticipated initial behaviors, the convulsions began with twisting clonus prior to the appearance of tonic hind limb extension. If twisting clonus were clearly of brain-stem origin, the second prediction might have been judged as fully confirmed. But such anatomical localization is not supported by experimental data. Rather, twisting clonus is believed to be of forebrain origin perhaps via an interaction with the brain-stem4. How can these seeming incongruities be reconciled? A review of the existing literature reveals disparities equivalent to those we have noted in the present study relative to the idea that the chemoshock threshold of the forebrain of GEPRs is abnormally low. On the question of whether the chemoshock threshold of the GEPR-9 brain-stem is below that of the GEPR-9 forebrain, the literature is uninformative since no previous investigations have been undertaken. The idea that the chemoshock threshold of the forebrain of GEPRs is abnormally low is supported by observations with flurothyl given by in-

halation ~3. It has also been supported by some ~9 but not all24 studies with pentylenetetrazol. Also, at least 2 laboratories have failed to detect differences in the F&F clonic threshold (CD-50) for bicuculline tg,~. As a means of facilitating future investigations in the 2 areas of ambiguity detected in our present investigation, we suggest that both pharmacokinetic and mechanistic factors contribute to the apparent disparities. The flurothyl study cited above may have unambiguously detected an abnormally low forebrain threshold in GEPRs because the pulmonary route of administration used for this agent resuits in a more gradual and smoother rise in brain flurothyl concentrations. In contrast, the increases in the concentration of pentylenetetrazol and bicuculline at specific sites within the brain may be more erratic so that threshold differences between GEPRs and controls are more likely to be missed. The possible validity of the pharmacokinetic hypothesis is underscored by observations with bicuculline. Accordingly, studies with systemic administration of this drug have failed to detect reduced thresholds in the GEPR, whereas Duplisse ~ has observed that the threshold for seizures induced by bicuculline injections into the inferior colliculus is abnormally low in the GEPR. Thus, results that are missed by systemic administration may be unmasked by site-specific microinfusion. With regard to our suggestion regarding apparent disparities arising from mechanistic factors, we believe that the direct plus indirect neurochemical effects of chemoconvulsants such as bicuculline and pentylenetetrazol may vary at different sites within the brain. Indeed, separate groups of investigators have shown that focal injections of bicuculline can produce either convulsant or anticonvulsant effects depending on the intracerebral sites of injection 1°'3°. In keeping with these observations, it is not surprising that the threshold for systemic bicuculline in the GEPR was not abnormal, even though the threshold upon injection into the inferior colliculus was deficient. In summary, our present electroshock and pentylenetetrazol experiments are largely in agreement with each other. We suggest that the partial disparity between these two studies probably defives from the same factors that we have proposed

to account for differences among +~hechemoconvulsant experiments. Accordingly, pharmacokinetic factors which influence regional brain levels of pentylenetetrazol would not be expected to regulate the distribution of the electroshock stimulus. Also, we want to point out that electroshock presumably produces a massive depolarization of neurons while pentylenetetrazol interferes with inhibition GABAergic synapses. As noted above for bicuculline, such action at GABAergic synapses may be seizure facilitating or seizure inhibiting depending on where the synapse is located. It is conceivable that the threshold of the forebrain and brain-stem seizure substrates may differ for electroshock and pentylenetetrazol due to the distribution and connections of the GABAergic circuitry for expression. Although some seizure circuitries in the GEPR may be exquisitely responsive to electroshock, they may be somewhat more resistant to pentylenetetrazol. Finally, it should be noted that the GEPRs used in the present study had experienced 3 (1 week apart) audiogenic seizures 3-6 weeks prior to their electroshock and/or pentylenetetrazol testing, and it is possible that this prior experience contributed to their enhanced responsiveness to electroshock and pentylenetetrazol. It has been shown previously that the severity of audiogenic seizures increases between the first and third seizure in rats experiencing 1 seizure/week 22 and it is possible that there is a permanent facilitating effect on all seizures. However, the long time interval between the third audiogenic seizure and the electroshock or pentylenetetrazol testing together with the prolonged repeated testing required for audiogeni c kindling (15-40) 22 may argue against such effects. In order to determine if prior audiogenic seizures have facilitating effects on subsequent electroshock or pentylenetetrazol seizures, the present study would have to be carried out in seizure naive GEPRs. On one hand, interpretation of findings in seizure naive GEPRs could be limited by not knowing if the animals are truly audiogenic susceptible. On the other hand, the interpretation of data in naive GEPRs will be greatly facilitated in light of the present findings and future studies should be directed along these lines.

10 ACKNOWLEDGFMENTS The authors are grateful to D. Patrick for technical assistance in some experiments, and Marion

REFERENCES 1 Browning, R.A., Role of the brain stem reticular formation in tonic-clonic seizures: lesion and pharmacological studies, Fed. Prec., 44 (1985) 2425-2431. 2 Browning, R.A., The role of neurotransmitters in electroshock seizure models. In: P.C. Jobe and H.L. Laird, II (Eds.), Neurotransmitters and Epilepsy, Humana Press, Clifton, NJ, 1987, pp. 277-320. 3 Browning, R.A. and Nelson D.K., Variation in threshold and pattern of electroshock-induced seizures in rats depending on site of stimulation, Life Sci., 37 (1985) 2205-2211. 4 Browning, R.A. and Nelson, D.K., Modification of electroshock and pentylenetetrazol seizure patterns in rats after precollicular transections, Exp. Neurol., 93 (1986) 546-556. 5 Browning, R.A., Simonton, R.L. and Turner, F.J., Antagonism of experimentally induced tonic seizures following a lesion of the midbrain tegmentum, Epilepsia, 22 (1981) 595-601. 6 Burnham, W.M., Core mechanisms in generalized convulsions, Fed. Prec., 44 (1985) 2442-2445. 7 Burnham, W.M., Electrical stimulation studies: generalized convulsions triggered from the brain-stem. In: G.H. Fromm et al. (Eds.), Epilepsy and the Reticular Formation: the Role of the Reticular Core in Convulsive Seizures, Alan R. Liss, New York, 1987, pp. 25-38. 8 Dailey, J.W. and Jobe, P.C., Anticonvulsant drugs and the genetically epilepsy-prone rat, Fed. Prec., 44 (1985) 2640-2644. 9 Dailey, J.W,, Reigel, C.E., Mishra, P.K. and Jobe, P.C., Neurobiology of seizure predisposition in the genetically epilepsy-prone rat, Epilepsy Res., 3 (1989) 3-17. 10 Dean, P. and Gale, K., Anticonvulsant action of GABA receptor blockade in the nigrotectal target region, Brain Res., 477 (1989) 391-395. 11 Duplisse, B.R., Mechanism of Susceptibility of Rats to Audiogenic Seizure, Ph.D. Dissertation, University of Arizona, 1976, 12 Faingold, C.L., Minireview. The genetically epilepsy prone rat, Gen, Pharmacoi., 19 (1988) 331-338. 13 Franck, J.E., Ginter, K.L. and Schwartzkroin, P.A., Developing genetically epilepsy-prone rats have an abnormal seizure response to flurothyl, Epilepsia, 30 (1989) 1-6. 14 Fromm, G.H., Terrence, C.F. and Chattha, A.S., Differential effects of antiepileptic and non-antiepileptic drugs on the reticular formation, Life Sci., 35 (1985) 2665-2673. 15 Fromm, G.H. and Terrence, C.F., Effect of antiepileptic

Cathey for her diUigence in typing the manuscript. This work was supported by a~l F.O. Eagles grant.

drugs on the brain-stem. In: G.H. Fromm et al. (Eds.), Epilepsy and the Reticular Formation: the Role of the Reticular Core in Convulsive Seizures, Alan R. Liss, New York, 1987, pp. 119 136. 16 Jobe, P.C., Picchioni, A.L. and Chin, L., Role of brain norepinephrine in audiogenic seizure in the rat, J. Pharmacol. Exp. Ther., 184 (1973)1-10. 17 Jobe, P.C., Lasley, S.M., Bettendorf, A.F., Frasca, J.J. and Daily, J.W., Studies of aspartame on supramaximal electroshock seizures in epileptic and non-epileptic rats, Fed. Prec., 2 (1988) A1067. 18 Kreindler, A., Zuckermann, E., Steriade, M. and Chimien, J., Electroclinicai features of convulsions induced by stimulation of brain stem, J. Neurophysioi., 21 (1958) 430-436. 19 Laird, II, H.E. and Huxtable, R.J., Taurine and audiogenic epilepsy. In: A. Barbeau and R.J. Huxtable (Eds.), Taufine and Neurological Disorders, Raven Press, New York, 1978, pp. 339-357. 20 Laird, If, H.E. and Jobe, P.C., The genetically epilepsyprone rat. In: P.C. Jobe and H.E. Laird, II (Eds.), Neurotransmitters and Epilepsy, Humana Press, Clifton, NJ, 1987, pp. 57-94. 21 Marescaux, C., Vergnes, M., Kiesmann, M., Depaulis, A., Micheletti, (3. and Warter, J,M., Kindling of audiogenic seizures in Wistar rats: an EEG study, Exp. Neurol., 97

(1987) 160-168. 22 Mishra, P.K., Dailey, J.W., Reigel, C.E., Tomsic, M.L. and Jobe, P.C., Sex-specific distinctions in audiogenic convulsions exhibited by severe seizure genetically epilepsyprone rats (GEPR-9s), Epilepsy Res., 2 (1988) 309-316. 23 Racine, R.J., Modification of seizure activity by electrical stimulation. II. Motor seizure, Electroenceph. Clin. Neurophysiol., 32 (1972) 281-294. 24 Reigei, C.E. Dailey, J.W, and Jobe, P.C., The genetically epilepsy-prone rat: an overview of seizure-prone characteristics and responsiveness to anticonvulsant drugs, Life Sci., 39 (1986) 763-774. 25 Reigel, C.E., Dailey, J.W., Jobe, P.C. and Stewart, J.J., Responsiveness of genetically epilepsy-prone rats to intracerebroventricular morphine-induced convulsions, Life Sci., 42 (1988) 1743-1749. 26 Savage, D.D., Reigel, C.E. and Jobe, P.C., Angular bundle kindling is accelerated in rats with a genetic predisposition to acoustic stimulus-induced seizures, Brain Res., 376 (1986) 412-415. 27 Swinyard, E.A., Laboratory evaluation of antiepileptic drugs: review of laboratory methods, Epilepsia, 10 (1986) 107-119.

11 28 Tacke, U., Piiiniinen, A. and Euomisto, J., Seizure thresholds and their postictal changes in audiogenic seizure (AGS)-susceptible rats, Eur. J. Pharmacol., 104 (1984) 85-92. 29 Turski, W.A., Cavalheiro, E.A., Schwarz, M., Czuczwar, S.J., Kheinrok, Z. and Turski, L., Limbic seizures produced by pilocarpine in rats: behavioral, electroencephalographic and neuropathological study, Behav. Brain Res., 9 (1983) 315-336.

30 Turski, L., Cavalheiro, E.A., Calderazzo-Fiiho, L.S., Bortolotto, Z.A., Klockgether, T., Ikonomidou, C. and Turski, W.A., The basal ganglia, the deep prepyriform cortex, and seizure spread: bicuculline is anticonvulsant in the rat striatum, Proc. Nat. Acad. Sci. (U.S.A.), 86 (1989) 1694-1697. 31 Woodbury, L.A. and Davenpo~% V.D., Design and use of a new electroshock seizure apparatus, and analysis of factors altering seizure thresholds and pattern, Arch. Int. Pharmacodyn., 92 (1952) 97-107.