70
Brain Re.vea;~/;. 512 {1990) 70 74 Elscvicr
BRES 15290
Stimulation-induced status epilepticus: role of the hippocampal mossy fibers in the seizures and associated neuropathology John P. Vicedomini* and J. Victor Nadler Departments of Pharmacology and Neurobiology, Duke University Medical Center, Durham, NC27710 (U.S.A.)
(Accepted 15 August 1989) Key words: Status epilepticus; Mossy fiber; Neuropathotogy; Hippocampus; Colchicine
A study of seizure activity and neuronal cell death produced by intracerebroventricular kainic acid had suggested that seizures conveyed by the hippocampal mossy fibers are more damaging to CA3 pyramidal cells than seizures conveyed by other pathways. To test this idea, the effects of a unilateral mossy fiber lesion were determined on seizure activity and neuronal degeneration provoked by repetitive electrical stimulation of the hippocampal fimbria in unanesthetized rats. Fimbrial stimulation resulted in self-sustained status epilepticus accompanied by neuronal degeneration in several brain regions, including area CA3 of the hippocampal formation. A unilateral mossy fiber lesion more readily attenuated the electrographic and behavioral seizures provoked by fimbrial stimulation than those provoked by kainic acid. If status epilepticus developed in the presence of a mossy fiber lesion, denervated CA3 pyramidal cells were still destroyed, although similar lesions protect these neurons from kainic acid-induced status epilepticus. Thus the two models of status epilepticus employ somewhat different seizure circuitries and neurodegenerative mechanisms. Seizures which involve the mossy fiber projection are not necessarily more damaging to CA3 pyramidal cells than seizures which do not.
INTRODUCTION I n t r a c e r e b r o v e n t r i c u l a r (i.c.v.) administration of kainic acid ( K A ) to rats provides an animal model of complex partial seizures, reproducing their characteristic electrographic, behavioral and neuropathologic features 1°. The neurotoxic action of i.c.v. K A results primarily from seizure activity, rather than from a direct interaction of the convulsant with the vulnerable neurons 1'5a3. H o w e v e r , the destruction of CA3 hippocampal p y r a m i d a l cells also d e p e n d s upon the integrity of mossy fiber ( M F ) afferents which originate from granule cells of the fascia dentata. I n t r a h i p p o c a m p a l injection of colchicine selectively destroys dentate granule cells, leading to d e g e n e r a t i o n of the mossy fiber pathway 4. This t r e a t m e n t protects C A 3 pyramidal cells from the subsequent i.c.v, administration of K A 7'11. A mossy fiber transection is similarly protective. The transection is effective immediately, before M F axons begin to degenerate, suggesting that the destruction of C A 3 pyramidal cells d e p e n d s on impulse flow within the M F pathway 9. I m p o r t a n t l y , an M F lesion p r o t e c t e d C A 3 pyramidal cells from i.c.v. K A even when the convulsant continued to
p r o v o k e a p r o l o n g e d status epilepticus n. Evidently, h i p p o c a m p a l seizure activity can be sustained through the activation of excitatory circuitry that bypasses the M F pathway. Pyramidal cell d e g e n e r a t i o n , however, d e p e n d s crucially on some influence of or interaction with the M F projection that enhances the n e u r o d e g e n e r a t i v e effect of the seizures. Thus it would a p p e a r that seizures which involve the M F p a t h w a y are m o r e d a m a g i n g than those which do not. It is i m p o r t a n t to know whether this conclusion applies to all e x p e r i m e n t a l complex partial seizures or only to those p r o v o k e d by i.c.v. K A . To address this issue, we utilized a m o d e l of status epilepticus based upon the repetitive stimulation of hippocampal afferent pathways 16. O u r initial studies of stimulation-induced status epilepticus (SISE) d e m o n s t r a t e d that stimulation of either the commissural or M F (activated indirectly via the perforant path) projections to a r e a C A 3 was equieffective in provoking self-sustained status epilepticus. Furthermore, the extent of neuronal d e g e n e r a t i o n , in area CA3 and elsewhere in the brain, d e p e n d e d on the duration of seizure activity, but not on which pathway was activated. The latter finding could be interpreted to
* Present address: Neurosurgery Research Laboratories, Miami Project, 1600 NW Tenth Avenue/R-48, Miami, FL 33136, U.S.A. Correspondence." J.V. Nadler, Department of Pharmacology, Box 3813, Duke University Medical Center, Durham, NC 27710, U.S.A. 0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
71 m e a n that seizures conveyed by the M F pathway are not inherently m o r e damaging to C A 3 p y r a m i d a l cells than seizures which p r o p a g a t e along o t h e r routes. It is possible, however, that commissural stimulation activ a t e d the M F p a t h w a y through a polysynaptic circuit and
that the c o m p o n e n t of the seizure that was driven by M F activity was responsible for much of the cell death in area CA3. The present study sought to resolve this issue by stimulating the commissural p r o j e c t i o n in animals whose M F pathway had previously b e e n destroyed. A brief
Fig. 1. Neuronal degeneration in hippocampal area CA3 after SISE, visualized by silver impregnation. A: sham mossy fiber transection (G indicates the granule cell body layer of the fascia dentata); B: mossy fiber transection (indicated by open arrows); C: area CA3 contralateral to the mossy fiber transection; D: vehicle injection; E: colchicine injection (note loss of dentate granule cells); F: area CA3 contralateral to the colchicine injection. Solid arrows (A-D) point to areas of somatic degeneration in area CA3. Solid triangles (A,D) denote position of the EEG recording electrode. Scale bar = 0.5 mm.
72 a c c o u n t of this w o r k has b e e n p r e s e n t e d 17. MATERIALS AND METHODS Adult male Sprague-Dawley rats were subjected to a unilateral MF lesion under pentobarbital anesthesia. Either of two procedures was used: unilateral transection of the MF pathway over a 4-ram length of the dorsal hippocampus with a Scouten wire knife (n = 6) H or unilateral destruction of dentate granule cells with intrahippocampal injections of colchieine (n = 6) 7. Histological examination of brain sections confirmed that the transection severed the MF pathway for 4 ram, with variable damage to the overlying CA1 area (Fig. 1B). Pathways in stratum oriens and stratum radiatum of area CA3 were also cut. Colchicine destroyed the great majority of dentate granule cells in 5 of the 6 rats and did not damage area CA3 (Fig. 1E). Three types of controls were used. In one group, the blade assembly of the Scouten knife was lowered into tile brain and the blade was extended and retracted without cutting any hippocampal pathways (Fig. 1A, n = 3). The second group received injections of artificial CSF instead of colchicine (Fig. ID, n = 4). Finally, 6 rats received no treatment before electrodes were implanted. Because these groups responded similarly to commissural stimulation, the data from all 13 control animals were combined. Immediately after the MF or sham transection, subjects were chronically implanted with a recording electrode in the pyramidal cell body layer of area CA3 at a site lateral to the transection (Fig. 1A) and a stimulating electrode in the contralateral fimbria (for coordinates see ref. 16). The stimulating electrode was fashioned from twisted 250-/am diameter stainless steel wire and a concentric bipolar electrode (SNEX-100; Rhodes Medical Inst., Tujunga, CA) was used for recording. Electrode positions were optimized in the dorsoventral plane to obtain the maximal electrically evoked commissural response in area CA3. The electrode leads were then mounted within an Amphenol connector and fixed to the skull with dental acrylic. Colchicine-treated and vehicle-injected controls were subjected to the same implantation protocol 3 weeks after injection. The 3-week delay was necessary to allow for changes in the shape of the hippocampal formation that followed the death of granule cells. Electrode positions were histologically verified in every animal. After a 7-day postoperative recovery, animals were subjected to SISE. Individual rats were placed in an electrically screened chamber and each electrode was connected by a commutator (Tech-Serv, Beltsville, MD) to either an $48 Grass stimulator for stimulation or to the amplifiers of a Grass EEG polygraph for recording. A 20-Hz stimulus train of 10-s duration was delivered to the fimbria every 40 s with the stimulus current set to evoke synaptic responses of maximal amplitude. The EEG was continuously monitored until 10 consecutive trains each elicited 30 s of hippocampal afterdischarge. As soon as this criterion was satisfied, stimulation was terminated. If at least 15 rain of uninterrupted self-sustained seizure activity occurred, no further stimulation was applied. If, however, seizure activity ceased within 15 min, stimulation resumed until 10 consecutive trains again each elicited 30 s of afterdischarge. This procedure was repeated until a series of 10 trains finally resulted in self-sustained seizure activity in excess of 15 min or until a maximum of 585 trains had been delivered. For quantitative purposes, the duration of SISE included only the period of spikes or spike-wave complexes that exceeded twice the preictal EEG amplitude and occurred at a frequency greater than 1 Hz. On the day after SISE, subjects received an overdose of pentobarbital and were transcardially perfused with buffered 0.9% (w/v) saline followed by buffered 4% (w/v) paraformaldehyde. Serial frozen brain sections, 40-/~m thick, were cut in either the coronal (MF transection) or horizontal (colchicine injection) plane. Degenerating neurons were visualized by silver impregnation s . Alternate sections were stained with Cresyl violet to verify electrode placements and to determine the extent of colchicine-induced degranu[ation.
Silver-impregnated sections were used for semi-quantitative analysis of neuronal degeneration. Brain regions examined were: hippocampal subfields CA1, CA2, CA3, and CA4, the fascia dentata, entorhinal cortex, anterior thalamus, and neocortex. The degeneration in each region was scored as follows: 0, no somatodendritic degeneration visible; 1, fewer than 10 silver-impregnated cell bodies per section; 2, extensive but subtotal degeneration of the neuronal population; 3, near-total degeneration. A total degencration score was computed for each animal by summing the rating of each of these 8 regions on both sides of the brain. In addition, we compared degeneration scores for area CA3 alone.
RESULTS Initially, all animals r e s p o n d e d similarly to the stimulus trains (Fig. 2). F i m b r i a l s t i m u l a t i o n e v o k e d brief afterdischarges a c c o m p a n i e d e i t h e r by b e h a v i o r a l arrest o r by limbic m o t o r seizures of v a r y i n g severity. O v e r the course of s t i m u l a t i o n , c o n t r o l rats and rats with an M F t r a n s e c t i o n r e s p o n d e d with increasingly p r o l o n g e d afterdischarges that e v e n t u a l l y e x c e e d e d the i n t e r t r a i n interval. All of t h e s e animals d e v e l o p e d S I S E , which lasted an a v e r a g e of 6 - 8 h (Table I). T h e b e h a v i o r a l e x p r e s s i o n of e l e c t r o g r a p h i c seizures in t h e s e t w o g r o u p s m a i n l y consisted of r e p e t i t i v e h e a d n o d d i n g (class 2 limbic m o t o r seizure). C o l c h i c i n e - t r e a t e d rats g e n e r a t e d h i p p o c a m p a l a f t e r d i s c h a r g e in r e s p o n s e to m o s t of the stimulus trains, but in 5 of the 6 a n i m a l s , no a f t e r d i s c h a r g e lasted for as long as 300 s. C o n s e q u e n t l y ,
t h e s e animals failed to
d e v e l o p S I S E . T h e o n e e x c e p t i o n a l rat e x p e r i e n c e d 105 min of S I S E , m u c h less t h a n transected
animals.
any of t h e c o n t r o l or
Colchicine-induced
degranulation
also r e d u c e d the severity of the b e h a v i o r a l seizures. Ictal activity was typically r e c o r d e d d u r i n g b e h a v i o r a l arrest and t h e r e w e r e few e p i s o d e s of h e a d n o d d i n g . A mossy fiber t r a n s e c t i o n did not significantly c h a n g e the d u r a t i o n of S I S E (Table I). It did, h o w e v e r , r e d u c e the n u m b e r of stimulus trains r e q u i r e d to p r o v o k e S I S E . T h e o n e c o l c h i c i n e - t r e a t e d a n i m a l that d e v e l o p e d S I S E also r e q u i r e d m a n y f e w e r stimulus trains t h a n control animals. In control rats, p r o l o n g e d S I S E d e s t r o y e d n e u r o n s within the h i p p o c a m p a l f o r m a t i o n , a m y g d a l a , a n t e r i o r and midline t h a l a m u s , lateral s e p t u m , e n t o r h i n a l c o r t e x , p y r i f o r m c o r t e x and n e o c o r t e x . D a m a g e to C A 3 pyramidal cells was essentially bilaterally s y m m e t r i c a l , but subtotal (Fig. 1 A , D ; T a b l e I). A m o s s y fiber t r a n s e c t i o n did not p r o t e c t the d e n e r v a t e d p y r a m i d a l cells f r o m d e g e n e r a t i o n (Fig. 1B,C: Table I) and did not significantly r e d u c e t h e total d e g e n e r a t i o n score. C o l c h i c i n e t r e a t e d rats that failed to d e v e l o p S I S E w e r e virtually free of s o m a t o d e n d r i t i c d e g e n e r a t i o n in t h e brain (Fig. 1E,F). D e g e n e r a t i o n was limited to a few h i p p o c a m p a l C A 4 and e n t o r h i n a l cortical n e u r o n s in o n e case. T h e o n e c o l c h i c i n e - t r e a t e d rat that d e v e l o p e d S I S E also s h o w e d
73
CONTROL
COLCHICINE
TRANSECTED '
I
I
0
I
I
'
""
• I
I II
~:"~ ~ "
I
III
t
O
II
III
[
I II III IV
b
I
I
I
" '
I
II
b
III
L
- - I
O
L
II
O
I
!"
7'
'
,
I
III
IV
- - [
V
I 0
I
[
- - ~
II
- - I
I
f
I
'
11 fill llill li~ It illI #"
"~,,i~'~,~! I' ' ~ ' ~ "
IV
;
- - 1
II III
IV
V
L 0
1
II
I
O
,r'-
5s
Fig. 2. Representative EEG recordings made at various times during the stimulation protocol. Upper panels, responses to the first stimulus train; lower panels, responses to the last stimulus train before self-sustained status epilepticus or to stimulus train 585 (coichicine); middle panels, responses to a stimulus train midway between the first and last. Solid bar beneath the traces denotes the period of stimulation. Roman numerals indicate the behavior of the animal: 0, behavioral arrest; I-V, stages of limbic motor seizure as defined by Racine 12.
evidence of s o m e C A 3 p y r a m i d a l cell d e g e n e r a t i o n on the d e g r a n u l a t e d side of the brain. The extent of the d a m a g e a p p e a r e d a p p r o p r i a t e for the relatively short d u r a t i o n of S I S E (Table I). Histological examination also r e v e a l e d that, in this animal, a substantial p r o p o r t i o n of the granule cells r e m a i n e d intact within the crest of the d e n t a t e gyrus.
TABLE I Effects of M F lesion on SISE and the associated neuronal degeneration
Status epilepticus was induced by repetitive stimulation of the fimbria. Somatodendritic degeneration was scored as described in the text. Ipsilateral and contralateral sides of the brain were defined with respect to the MF lesion. Values are means + S.E.M. or results from the one colchicine-treated rat that developed SISE.
SISE/total Stimulus trains to SISE SISE duration (min) Degeneration score Ipsilateral CA3 Contralateral CA3 All regions
Control
M F transeclion
Colchicine
13/13 304 + 26 356 + 26
6/6 193 + 10" 483 + 116
1/6 88 105
1.9 +_O.1
2.0 +_0.0
1.0
1.7 + 0.1 28.8 _+3.5
1.8 + 0.3 22.8 + 1.4
0.0 7.0
* Rats with MF transection differed from controls at P < 0.01 (Student's t-test).
DISCUSSION The most i m p o r t a n t finding of this study was the strict relationship b e t w e e n seizure duration and the degeneration of C A 3 h i p p o c a m p a l p y r a m i d a l cells, regardless of whether or not the M F fibers were intact. The M F projection p r o v e d to be crucial for the d e v e l o p m e n t of SISE. Even unilateral destruction of this p a t h w a y was sufficient to prevent SISE and to a t t e n u a t e the behavioral expression of the electrically e v o k e d seizures, provided that very few fibers r e m a i n e d intact. These findings suggest that a substantial percentage of the M F pathway on both sides of the brain must be intact for fimbrial stimulation to evoke SE and for the electrically e v o k e d seizures to activate m o t o r centers. Only if SISE develo p e d , did fimbrial stimulation d e s t r o y C A 3 p y r a m i d a l cells. M F lesions could not p r e v e n t this degeneration; CA3 p y r a m i d a l cells deprived of their mossy fiber innervation were as vulnerable to seizure activity as normally innervated p y r a m i d a l cells. Thus seizure activity conveyed by the mossy fiber p r o j e c t i o n is not necessarily more damaging to these neurons than seizure activity which p r o p a g a t e s along o t h e r routes. The differing responses to stimulation of rats subjected to an M F transection, on the one hand, and colchicinetreated rats, on the other, can p r o b a b l y be explained by the different percentage o f the M F p a t h w a y destroyed. A n M F transection only i n t e r r u p t e d these fibers in the
74 dorsal h i p p o c a m p a l formation. The more seizure-prone ventral CA3 circuitry 2'3 remained intact. Propagation of electrically e v o k e d and self-sustained seizures from the ventral to the dorsal CA3 area by way of ipsilateral associational connections TM probably accounts for the failure of an M F transection to prevent SISE. This conclusion is s u p p o r t e d by evidence that recurrent excitatory pathways within area CA3 p r o m o t e the spread of epileptiform activity6,15. In contrast, coichicine extensively d e g r a n u l a t e d both the dorsal and ventral hippoc a m p a l formation. Presumably, it was the near-total loss of M F innervation that blocked the d e v e l o p m e n t of SISE and its related neuropathology. These results differ substantially from those obtained when the neuroprotective effects of M F lesions were assessed against seizure activity induced by i.c.v. K A ~. M F lesions always p r o t e c t e d the denervated CA3 pyramidal cells against K A seizures, even when the animal e x p e r i e n c e d m a n y hours of status epilepticus. In contrast, M F lesions did not protect these cells against SISE. This result supports the view that a specific interaction between K A and the M F pathway augments the toxic effect of seizures on C A 3 pyramidal cells. H o w e v e r , the M F pathway appears to be more important for the
d e v e l o p m e n t of SISE than for the d e v e l o p m e n t ~,l K A - i n d u c e d status epilepticus. Unilateral degranulation with colchicine could prevent the d e v e l o p m e n t of SISE, whereas c o m p a r a b l e lesions failed to prevent KA-induced status epilepticus. Only bilateral degranulation was effective in the latter case. Similarly, unilateral degranulation attenuated the behavioral expression of seizures evoked by commissural stimulation, whereas bilateral degranulation was necessary to achieve the same result after i.c.v. K A . Thus the M F pathway plays a s o m e w h a t different role in electrically induced and K A - i n d u c e d complex partial seizures. It is more i m p o r t a n t for the generation of electrographic and behavioral seizures e v o k e d by fimbrial stimulation. Conversely, it strongly influences the relationship between K A - i n d u c e d seizures and C A 3 damage, but not between SISE and C A 3 damage. These differences point to the involvement of somewhat different n e u r o d e g e n e r a t i v e mechanisms and seizure circuitry in the two animal models of complex partial seizures, despite similarities in their electrographic, behavioral and neuropathologic characteristics. Acknowledgements. We thank Dr. M. Okazaki for assistance with the MF lesions. This study was supported by NIH Grant NS 17771.
REFERENCES 1 Ault, B., Gruenthal, M., Armstrong, D.R. and Nadler, J.V., Efficacy of baclofen and phenobarbital against the kainic acid limbic seizure-brain damage syndrome, J. Pharmacol. Exp. Ther., 239 (1986) 612-617. 2 Bragdon, A.C., Taylor, D.M. and Wilson, W.A., Potassiuminduced epileptiform activity in area CA3 varies markedly along the septotemporal axis of the rat hippocampus, Brain Research, 378 (1986) 169-173. 3 Gilbert, M., Racine, R.J. and Smith, G.K., Epileptiform burst responses in ventral vs dorsal hippocampal slices, Brain Research, 361 (1985) 389-391. 4 Goldschmidt, R.B. and Steward, O., Preferential neurotoxicity of colchicine for granule cells of the dentate gyrus of the adult rat, Proc. Natl. Acad. Sci. U.S.A., 77 (1980) 3047-3051. 5 Gruenthal, M., Armstrong, D.R., Ault, B. and Nadler, J.V., Comparison of seizures and brain lesions produced by intracerebroventricular kainic acid and bicuculline methiodide, Exp. Neurol., 93 (1986) 621-630. 6 Knowles, W.D., Traub, R.D. and Strowbridge, B.W., The initiation and spread of epileptiform bursts in the in vitro hippocampal slice, Neuroscience, 21 (1987) 441-455. 7 Nadler, J.V. and Cuthbertson, G.J., Kainic acid neurotoxicity toward hippocampal formation: dependence on specific excitatory pathways, Brain Research, 195 (1980) 47-56. 8 Nadler, J.V. and Evenson, D.A., Use of excitatory amino acids to make axon-sparing lesions of the hypothalamus. In P.M. Conn (Ed.), Neuroendocrine Peptide Methodology, Academic, San Diego, 1989, pp. 881-890. 9 Nadler, J.V., Evenson, D.A. and Smith, E.M., Evidence from
10
11 12 13 14 15
16 17
lesion studies for epileptogenic and non-epileptogenic neurotoxic interactions between kainic acid and excitatory innervation, Brain Research, 201 (1981) 405-410. Nadler, J.V., Okazaki, M.M., Gruenthal, M., Ault, B. and Armstrong, D.R., Kainic acid seizures and neuronal cell death: insights from studies of selective lesions and drugs. In R. Schwarcz and Y. Ben-Ari (Eds.), Excitatory Amino Acids and Epilepsy, Adv. Exp. Med. Biol., Vol. 203, Plenum, New York, 1986, pp. 673-686. Okazaki, M.M. and Nadler, J.V., Protective effects of mossy fiber lesions against kainic acid-induced seizures and neuronal degeneration, Nearoscience, 26 (1988) 763-781. Racine, R.J., Modification of seizure activity by electrical stimulation. II. Motor seizure, Electroenceph. Clin. Neurophysiol., 32 (1972) 281-294. Sater, R.A. and Nadler, J.V., On the relation between seizures and brain lesions after intracerebroventricutar kainic acid, Neurosci. Lett., 84 (1988) 73-78. Swanson, L.W., Wyss, J.M. and Cowan, W.M., An autoradiographic study of the organization of intrahippocampal association pathways in the rat, J. Comp. Neurol., 181 (1978) 681-716. Traub, R.D., Knowles, W.D., Miles, R. and Wong, R.K.S., Models of the cellular mechanism underlying propagation of epileptiform activity in CA2-CA3 region of the hippocampal slice, Neuroscience, 21 (1987) 429-440. Vicedomini, J.P. and Nadler, J.V., A model of status epilepticus based on electrical stimulation of hippocampal afferent pathways, Exp. Neurol., 96 (1987) 681-691. Vicedomini, J.P. and Nadler, J.V., Stimulation-induced status epilepticus: role of the hippocampal mossy fiber projection, Soc. Neurosci. Abstr., 13 (1987) 1155.
Brain Re.vea;~/;. 512 {1990) 70 74 Elscvicr
BRES 15290
Stimulation-induced status epilepticus: role of the hippocampal mossy fibers in the seizures and associated neuropathology John P. Vicedomini* and J. Victor Nadler Departments of Pharmacology and Neurobiology, Duke University Medical Center, Durham, NC27710 (U.S.A.)
(Accepted 15 August 1989) Key words: Status epilepticus; Mossy fiber; Neuropathotogy; Hippocampus; Colchicine
A study of seizure activity and neuronal cell death produced by intracerebroventricular kainic acid had suggested that seizures conveyed by the hippocampal mossy fibers are more damaging to CA3 pyramidal cells than seizures conveyed by other pathways. To test this idea, the effects of a unilateral mossy fiber lesion were determined on seizure activity and neuronal degeneration provoked by repetitive electrical stimulation of the hippocampal fimbria in unanesthetized rats. Fimbrial stimulation resulted in self-sustained status epilepticus accompanied by neuronal degeneration in several brain regions, including area CA3 of the hippocampal formation. A unilateral mossy fiber lesion more readily attenuated the electrographic and behavioral seizures provoked by fimbrial stimulation than those provoked by kainic acid. If status epilepticus developed in the presence of a mossy fiber lesion, denervated CA3 pyramidal cells were still destroyed, although similar lesions protect these neurons from kainic acid-induced status epilepticus. Thus the two models of status epilepticus employ somewhat different seizure circuitries and neurodegenerative mechanisms. Seizures which involve the mossy fiber projection are not necessarily more damaging to CA3 pyramidal cells than seizures which do not.
INTRODUCTION I n t r a c e r e b r o v e n t r i c u l a r (i.c.v.) administration of kainic acid ( K A ) to rats provides an animal model of complex partial seizures, reproducing their characteristic electrographic, behavioral and neuropathologic features 1°. The neurotoxic action of i.c.v. K A results primarily from seizure activity, rather than from a direct interaction of the convulsant with the vulnerable neurons 1'5a3. H o w e v e r , the destruction of CA3 hippocampal p y r a m i d a l cells also d e p e n d s upon the integrity of mossy fiber ( M F ) afferents which originate from granule cells of the fascia dentata. I n t r a h i p p o c a m p a l injection of colchicine selectively destroys dentate granule cells, leading to d e g e n e r a t i o n of the mossy fiber pathway 4. This t r e a t m e n t protects C A 3 pyramidal cells from the subsequent i.c.v, administration of K A 7'11. A mossy fiber transection is similarly protective. The transection is effective immediately, before M F axons begin to degenerate, suggesting that the destruction of C A 3 pyramidal cells d e p e n d s on impulse flow within the M F pathway 9. I m p o r t a n t l y , an M F lesion p r o t e c t e d C A 3 pyramidal cells from i.c.v. K A even when the convulsant continued to
p r o v o k e a p r o l o n g e d status epilepticus n. Evidently, h i p p o c a m p a l seizure activity can be sustained through the activation of excitatory circuitry that bypasses the M F pathway. Pyramidal cell d e g e n e r a t i o n , however, d e p e n d s crucially on some influence of or interaction with the M F projection that enhances the n e u r o d e g e n e r a t i v e effect of the seizures. Thus it would a p p e a r that seizures which involve the M F p a t h w a y are m o r e d a m a g i n g than those which do not. It is i m p o r t a n t to know whether this conclusion applies to all e x p e r i m e n t a l complex partial seizures or only to those p r o v o k e d by i.c.v. K A . To address this issue, we utilized a m o d e l of status epilepticus based upon the repetitive stimulation of hippocampal afferent pathways 16. O u r initial studies of stimulation-induced status epilepticus (SISE) d e m o n s t r a t e d that stimulation of either the commissural or M F (activated indirectly via the perforant path) projections to a r e a C A 3 was equieffective in provoking self-sustained status epilepticus. Furthermore, the extent of neuronal d e g e n e r a t i o n , in area CA3 and elsewhere in the brain, d e p e n d e d on the duration of seizure activity, but not on which pathway was activated. The latter finding could be interpreted to
* Present address: Neurosurgery Research Laboratories, Miami Project, 1600 NW Tenth Avenue/R-48, Miami, FL 33136, U.S.A. Correspondence." J.V. Nadler, Department of Pharmacology, Box 3813, Duke University Medical Center, Durham, NC 27710, U.S.A. 0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
71 m e a n that seizures conveyed by the M F pathway are not inherently m o r e damaging to C A 3 p y r a m i d a l cells than seizures which p r o p a g a t e along o t h e r routes. It is possible, however, that commissural stimulation activ a t e d the M F p a t h w a y through a polysynaptic circuit and
that the c o m p o n e n t of the seizure that was driven by M F activity was responsible for much of the cell death in area CA3. The present study sought to resolve this issue by stimulating the commissural p r o j e c t i o n in animals whose M F pathway had previously b e e n destroyed. A brief
Fig. 1. Neuronal degeneration in hippocampal area CA3 after SISE, visualized by silver impregnation. A: sham mossy fiber transection (G indicates the granule cell body layer of the fascia dentata); B: mossy fiber transection (indicated by open arrows); C: area CA3 contralateral to the mossy fiber transection; D: vehicle injection; E: colchicine injection (note loss of dentate granule cells); F: area CA3 contralateral to the colchicine injection. Solid arrows (A-D) point to areas of somatic degeneration in area CA3. Solid triangles (A,D) denote position of the EEG recording electrode. Scale bar = 0.5 mm.
72 a c c o u n t of this w o r k has b e e n p r e s e n t e d 17. MATERIALS AND METHODS Adult male Sprague-Dawley rats were subjected to a unilateral MF lesion under pentobarbital anesthesia. Either of two procedures was used: unilateral transection of the MF pathway over a 4-ram length of the dorsal hippocampus with a Scouten wire knife (n = 6) H or unilateral destruction of dentate granule cells with intrahippocampal injections of colchieine (n = 6) 7. Histological examination of brain sections confirmed that the transection severed the MF pathway for 4 ram, with variable damage to the overlying CA1 area (Fig. 1B). Pathways in stratum oriens and stratum radiatum of area CA3 were also cut. Colchicine destroyed the great majority of dentate granule cells in 5 of the 6 rats and did not damage area CA3 (Fig. 1E). Three types of controls were used. In one group, the blade assembly of the Scouten knife was lowered into tile brain and the blade was extended and retracted without cutting any hippocampal pathways (Fig. 1A, n = 3). The second group received injections of artificial CSF instead of colchicine (Fig. ID, n = 4). Finally, 6 rats received no treatment before electrodes were implanted. Because these groups responded similarly to commissural stimulation, the data from all 13 control animals were combined. Immediately after the MF or sham transection, subjects were chronically implanted with a recording electrode in the pyramidal cell body layer of area CA3 at a site lateral to the transection (Fig. 1A) and a stimulating electrode in the contralateral fimbria (for coordinates see ref. 16). The stimulating electrode was fashioned from twisted 250-/am diameter stainless steel wire and a concentric bipolar electrode (SNEX-100; Rhodes Medical Inst., Tujunga, CA) was used for recording. Electrode positions were optimized in the dorsoventral plane to obtain the maximal electrically evoked commissural response in area CA3. The electrode leads were then mounted within an Amphenol connector and fixed to the skull with dental acrylic. Colchicine-treated and vehicle-injected controls were subjected to the same implantation protocol 3 weeks after injection. The 3-week delay was necessary to allow for changes in the shape of the hippocampal formation that followed the death of granule cells. Electrode positions were histologically verified in every animal. After a 7-day postoperative recovery, animals were subjected to SISE. Individual rats were placed in an electrically screened chamber and each electrode was connected by a commutator (Tech-Serv, Beltsville, MD) to either an $48 Grass stimulator for stimulation or to the amplifiers of a Grass EEG polygraph for recording. A 20-Hz stimulus train of 10-s duration was delivered to the fimbria every 40 s with the stimulus current set to evoke synaptic responses of maximal amplitude. The EEG was continuously monitored until 10 consecutive trains each elicited 30 s of hippocampal afterdischarge. As soon as this criterion was satisfied, stimulation was terminated. If at least 15 rain of uninterrupted self-sustained seizure activity occurred, no further stimulation was applied. If, however, seizure activity ceased within 15 min, stimulation resumed until 10 consecutive trains again each elicited 30 s of afterdischarge. This procedure was repeated until a series of 10 trains finally resulted in self-sustained seizure activity in excess of 15 min or until a maximum of 585 trains had been delivered. For quantitative purposes, the duration of SISE included only the period of spikes or spike-wave complexes that exceeded twice the preictal EEG amplitude and occurred at a frequency greater than 1 Hz. On the day after SISE, subjects received an overdose of pentobarbital and were transcardially perfused with buffered 0.9% (w/v) saline followed by buffered 4% (w/v) paraformaldehyde. Serial frozen brain sections, 40-/~m thick, were cut in either the coronal (MF transection) or horizontal (colchicine injection) plane. Degenerating neurons were visualized by silver impregnation s . Alternate sections were stained with Cresyl violet to verify electrode placements and to determine the extent of colchicine-induced degranu[ation.
Silver-impregnated sections were used for semi-quantitative analysis of neuronal degeneration. Brain regions examined were: hippocampal subfields CA1, CA2, CA3, and CA4, the fascia dentata, entorhinal cortex, anterior thalamus, and neocortex. The degeneration in each region was scored as follows: 0, no somatodendritic degeneration visible; 1, fewer than 10 silver-impregnated cell bodies per section; 2, extensive but subtotal degeneration of the neuronal population; 3, near-total degeneration. A total degencration score was computed for each animal by summing the rating of each of these 8 regions on both sides of the brain. In addition, we compared degeneration scores for area CA3 alone.
RESULTS Initially, all animals r e s p o n d e d similarly to the stimulus trains (Fig. 2). F i m b r i a l s t i m u l a t i o n e v o k e d brief afterdischarges a c c o m p a n i e d e i t h e r by b e h a v i o r a l arrest o r by limbic m o t o r seizures of v a r y i n g severity. O v e r the course of s t i m u l a t i o n , c o n t r o l rats and rats with an M F t r a n s e c t i o n r e s p o n d e d with increasingly p r o l o n g e d afterdischarges that e v e n t u a l l y e x c e e d e d the i n t e r t r a i n interval. All of t h e s e animals d e v e l o p e d S I S E , which lasted an a v e r a g e of 6 - 8 h (Table I). T h e b e h a v i o r a l e x p r e s s i o n of e l e c t r o g r a p h i c seizures in t h e s e t w o g r o u p s m a i n l y consisted of r e p e t i t i v e h e a d n o d d i n g (class 2 limbic m o t o r seizure). C o l c h i c i n e - t r e a t e d rats g e n e r a t e d h i p p o c a m p a l a f t e r d i s c h a r g e in r e s p o n s e to m o s t of the stimulus trains, but in 5 of the 6 a n i m a l s , no a f t e r d i s c h a r g e lasted for as long as 300 s. C o n s e q u e n t l y ,
t h e s e animals failed to
d e v e l o p S I S E . T h e o n e e x c e p t i o n a l rat e x p e r i e n c e d 105 min of S I S E , m u c h less t h a n transected
animals.
any of t h e c o n t r o l or
Colchicine-induced
degranulation
also r e d u c e d the severity of the b e h a v i o r a l seizures. Ictal activity was typically r e c o r d e d d u r i n g b e h a v i o r a l arrest and t h e r e w e r e few e p i s o d e s of h e a d n o d d i n g . A mossy fiber t r a n s e c t i o n did not significantly c h a n g e the d u r a t i o n of S I S E (Table I). It did, h o w e v e r , r e d u c e the n u m b e r of stimulus trains r e q u i r e d to p r o v o k e S I S E . T h e o n e c o l c h i c i n e - t r e a t e d a n i m a l that d e v e l o p e d S I S E also r e q u i r e d m a n y f e w e r stimulus trains t h a n control animals. In control rats, p r o l o n g e d S I S E d e s t r o y e d n e u r o n s within the h i p p o c a m p a l f o r m a t i o n , a m y g d a l a , a n t e r i o r and midline t h a l a m u s , lateral s e p t u m , e n t o r h i n a l c o r t e x , p y r i f o r m c o r t e x and n e o c o r t e x . D a m a g e to C A 3 pyramidal cells was essentially bilaterally s y m m e t r i c a l , but subtotal (Fig. 1 A , D ; T a b l e I). A m o s s y fiber t r a n s e c t i o n did not p r o t e c t the d e n e r v a t e d p y r a m i d a l cells f r o m d e g e n e r a t i o n (Fig. 1B,C: Table I) and did not significantly r e d u c e t h e total d e g e n e r a t i o n score. C o l c h i c i n e t r e a t e d rats that failed to d e v e l o p S I S E w e r e virtually free of s o m a t o d e n d r i t i c d e g e n e r a t i o n in t h e brain (Fig. 1E,F). D e g e n e r a t i o n was limited to a few h i p p o c a m p a l C A 4 and e n t o r h i n a l cortical n e u r o n s in o n e case. T h e o n e c o l c h i c i n e - t r e a t e d rat that d e v e l o p e d S I S E also s h o w e d
73
CONTROL
COLCHICINE
TRANSECTED '
I
I
0
I
I
'
""
• I
I II
~:"~ ~ "
I
III
t
O
II
III
[
I II III IV
b
I
I
I
" '
I
II
b
III
L
- - I
O
L
II
O
I
!"
7'
'
,
I
III
IV
- - [
V
I 0
I
[
- - ~
II
- - I
I
f
I
'
11 fill llill li~ It illI #"
"~,,i~'~,~! I' ' ~ ' ~ "
IV
;
- - 1
II III
IV
V
L 0
1
II
I
O
,r'-
5s
Fig. 2. Representative EEG recordings made at various times during the stimulation protocol. Upper panels, responses to the first stimulus train; lower panels, responses to the last stimulus train before self-sustained status epilepticus or to stimulus train 585 (coichicine); middle panels, responses to a stimulus train midway between the first and last. Solid bar beneath the traces denotes the period of stimulation. Roman numerals indicate the behavior of the animal: 0, behavioral arrest; I-V, stages of limbic motor seizure as defined by Racine 12.
evidence of s o m e C A 3 p y r a m i d a l cell d e g e n e r a t i o n on the d e g r a n u l a t e d side of the brain. The extent of the d a m a g e a p p e a r e d a p p r o p r i a t e for the relatively short d u r a t i o n of S I S E (Table I). Histological examination also r e v e a l e d that, in this animal, a substantial p r o p o r t i o n of the granule cells r e m a i n e d intact within the crest of the d e n t a t e gyrus.
TABLE I Effects of M F lesion on SISE and the associated neuronal degeneration
Status epilepticus was induced by repetitive stimulation of the fimbria. Somatodendritic degeneration was scored as described in the text. Ipsilateral and contralateral sides of the brain were defined with respect to the MF lesion. Values are means + S.E.M. or results from the one colchicine-treated rat that developed SISE.
SISE/total Stimulus trains to SISE SISE duration (min) Degeneration score Ipsilateral CA3 Contralateral CA3 All regions
Control
M F transeclion
Colchicine
13/13 304 + 26 356 + 26
6/6 193 + 10" 483 + 116
1/6 88 105
1.9 +_O.1
2.0 +_0.0
1.0
1.7 + 0.1 28.8 _+3.5
1.8 + 0.3 22.8 + 1.4
0.0 7.0
* Rats with MF transection differed from controls at P < 0.01 (Student's t-test).
DISCUSSION The most i m p o r t a n t finding of this study was the strict relationship b e t w e e n seizure duration and the degeneration of C A 3 h i p p o c a m p a l p y r a m i d a l cells, regardless of whether or not the M F fibers were intact. The M F projection p r o v e d to be crucial for the d e v e l o p m e n t of SISE. Even unilateral destruction of this p a t h w a y was sufficient to prevent SISE and to a t t e n u a t e the behavioral expression of the electrically e v o k e d seizures, provided that very few fibers r e m a i n e d intact. These findings suggest that a substantial percentage of the M F pathway on both sides of the brain must be intact for fimbrial stimulation to evoke SE and for the electrically e v o k e d seizures to activate m o t o r centers. Only if SISE develo p e d , did fimbrial stimulation d e s t r o y C A 3 p y r a m i d a l cells. M F lesions could not p r e v e n t this degeneration; CA3 p y r a m i d a l cells deprived of their mossy fiber innervation were as vulnerable to seizure activity as normally innervated p y r a m i d a l cells. Thus seizure activity conveyed by the mossy fiber p r o j e c t i o n is not necessarily more damaging to these neurons than seizure activity which p r o p a g a t e s along o t h e r routes. The differing responses to stimulation of rats subjected to an M F transection, on the one hand, and colchicinetreated rats, on the other, can p r o b a b l y be explained by the different percentage o f the M F p a t h w a y destroyed. A n M F transection only i n t e r r u p t e d these fibers in the
74 dorsal h i p p o c a m p a l formation. The more seizure-prone ventral CA3 circuitry 2'3 remained intact. Propagation of electrically e v o k e d and self-sustained seizures from the ventral to the dorsal CA3 area by way of ipsilateral associational connections TM probably accounts for the failure of an M F transection to prevent SISE. This conclusion is s u p p o r t e d by evidence that recurrent excitatory pathways within area CA3 p r o m o t e the spread of epileptiform activity6,15. In contrast, coichicine extensively d e g r a n u l a t e d both the dorsal and ventral hippoc a m p a l formation. Presumably, it was the near-total loss of M F innervation that blocked the d e v e l o p m e n t of SISE and its related neuropathology. These results differ substantially from those obtained when the neuroprotective effects of M F lesions were assessed against seizure activity induced by i.c.v. K A ~. M F lesions always p r o t e c t e d the denervated CA3 pyramidal cells against K A seizures, even when the animal e x p e r i e n c e d m a n y hours of status epilepticus. In contrast, M F lesions did not protect these cells against SISE. This result supports the view that a specific interaction between K A and the M F pathway augments the toxic effect of seizures on C A 3 pyramidal cells. H o w e v e r , the M F pathway appears to be more important for the
d e v e l o p m e n t of SISE than for the d e v e l o p m e n t ~,l K A - i n d u c e d status epilepticus. Unilateral degranulation with colchicine could prevent the d e v e l o p m e n t of SISE, whereas c o m p a r a b l e lesions failed to prevent KA-induced status epilepticus. Only bilateral degranulation was effective in the latter case. Similarly, unilateral degranulation attenuated the behavioral expression of seizures evoked by commissural stimulation, whereas bilateral degranulation was necessary to achieve the same result after i.c.v. K A . Thus the M F pathway plays a s o m e w h a t different role in electrically induced and K A - i n d u c e d complex partial seizures. It is more i m p o r t a n t for the generation of electrographic and behavioral seizures e v o k e d by fimbrial stimulation. Conversely, it strongly influences the relationship between K A - i n d u c e d seizures and C A 3 damage, but not between SISE and C A 3 damage. These differences point to the involvement of somewhat different n e u r o d e g e n e r a t i v e mechanisms and seizure circuitry in the two animal models of complex partial seizures, despite similarities in their electrographic, behavioral and neuropathologic characteristics. Acknowledgements. We thank Dr. M. Okazaki for assistance with the MF lesions. This study was supported by NIH Grant NS 17771.
REFERENCES 1 Ault, B., Gruenthal, M., Armstrong, D.R. and Nadler, J.V., Efficacy of baclofen and phenobarbital against the kainic acid limbic seizure-brain damage syndrome, J. Pharmacol. Exp. Ther., 239 (1986) 612-617. 2 Bragdon, A.C., Taylor, D.M. and Wilson, W.A., Potassiuminduced epileptiform activity in area CA3 varies markedly along the septotemporal axis of the rat hippocampus, Brain Research, 378 (1986) 169-173. 3 Gilbert, M., Racine, R.J. and Smith, G.K., Epileptiform burst responses in ventral vs dorsal hippocampal slices, Brain Research, 361 (1985) 389-391. 4 Goldschmidt, R.B. and Steward, O., Preferential neurotoxicity of colchicine for granule cells of the dentate gyrus of the adult rat, Proc. Natl. Acad. Sci. U.S.A., 77 (1980) 3047-3051. 5 Gruenthal, M., Armstrong, D.R., Ault, B. and Nadler, J.V., Comparison of seizures and brain lesions produced by intracerebroventricular kainic acid and bicuculline methiodide, Exp. Neurol., 93 (1986) 621-630. 6 Knowles, W.D., Traub, R.D. and Strowbridge, B.W., The initiation and spread of epileptiform bursts in the in vitro hippocampal slice, Neuroscience, 21 (1987) 441-455. 7 Nadler, J.V. and Cuthbertson, G.J., Kainic acid neurotoxicity toward hippocampal formation: dependence on specific excitatory pathways, Brain Research, 195 (1980) 47-56. 8 Nadler, J.V. and Evenson, D.A., Use of excitatory amino acids to make axon-sparing lesions of the hypothalamus. In P.M. Conn (Ed.), Neuroendocrine Peptide Methodology, Academic, San Diego, 1989, pp. 881-890. 9 Nadler, J.V., Evenson, D.A. and Smith, E.M., Evidence from
10
11 12 13 14 15
16 17
lesion studies for epileptogenic and non-epileptogenic neurotoxic interactions between kainic acid and excitatory innervation, Brain Research, 201 (1981) 405-410. Nadler, J.V., Okazaki, M.M., Gruenthal, M., Ault, B. and Armstrong, D.R., Kainic acid seizures and neuronal cell death: insights from studies of selective lesions and drugs. In R. Schwarcz and Y. Ben-Ari (Eds.), Excitatory Amino Acids and Epilepsy, Adv. Exp. Med. Biol., Vol. 203, Plenum, New York, 1986, pp. 673-686. Okazaki, M.M. and Nadler, J.V., Protective effects of mossy fiber lesions against kainic acid-induced seizures and neuronal degeneration, Nearoscience, 26 (1988) 763-781. Racine, R.J., Modification of seizure activity by electrical stimulation. II. Motor seizure, Electroenceph. Clin. Neurophysiol., 32 (1972) 281-294. Sater, R.A. and Nadler, J.V., On the relation between seizures and brain lesions after intracerebroventricutar kainic acid, Neurosci. Lett., 84 (1988) 73-78. Swanson, L.W., Wyss, J.M. and Cowan, W.M., An autoradiographic study of the organization of intrahippocampal association pathways in the rat, J. Comp. Neurol., 181 (1978) 681-716. Traub, R.D., Knowles, W.D., Miles, R. and Wong, R.K.S., Models of the cellular mechanism underlying propagation of epileptiform activity in CA2-CA3 region of the hippocampal slice, Neuroscience, 21 (1987) 429-440. Vicedomini, J.P. and Nadler, J.V., A model of status epilepticus based on electrical stimulation of hippocampal afferent pathways, Exp. Neurol., 96 (1987) 681-691. Vicedomini, J.P. and Nadler, J.V., Stimulation-induced status epilepticus: role of the hippocampal mossy fiber projection, Soc. Neurosci. Abstr., 13 (1987) 1155.
