Research note. Unilateral susceptibility to audiogenic seizure impaired by contralateral lesions in the inferior colliculus

Audiogenic kindling and secondary subcortico-cortical epileptogenesis: Behavioral correlates and electrographic features

Human epilepsy is usually considered to result from cortical pathology, but animal studies show that the cortex may be secondarily involved in epileptogenesis, and cortical seizures may be triggered by extracortical mechanisms. In the audiogenic kindling model, recurrent subcortical (brainstem-driven) seizures induce secondary epileptic activation of the cortex. The present review focuses on behavioral and electrographic features of the subcortico-cortical epileptogenesis: (1) behavioral expressions of traditional and mild paradigms of audiogenic kindling produced by full-blown (generalized) and minimal (focal) audiogenic seizures, respectively; (2) electrographic manifestations of secondary epileptic activation of the cortex - cortical epileptic discharge and cortical spreading depression; and (3) persistent individual asymmetry of minimal audiogenic seizures and secondary cortical events produced by their repetition. The characteristics of audiogenic kindling suggest that this model represents a unique experimental approach to studying cortical epileptogenesis and network aspects of epilepsy. This article is part of a Special Issue entitled "Genetic and Reflex Epilepsies, Audiogenic Seizures and Strains: From Experimental Models to the Clinic".

Lateral asymmetry of early seizure manifestations in experimental generalized epilepsy

Reorganization of seizure networks during epileptogenesis involves cortico-subcortical and interhemispheric interactions. In the audiogenic kindling (AK) model of generalized tonic-clonic seizures, upstream seizure propagation along ascending brainstem-to-forebrain pathways determines progressive intensification of repeated sound-induced convulsions. Full-blown audiogenic seizures are bilaterally symmetric and their repetition results in bisynchronous recruiting the cortex in secondary epileptogenesis. The present study describes lateral asymmetry of initial behavioral and EEG manifestations of audiogenic seizures and AK in Wistar and WAG/Rij rats with acoustic hypersensitivity. These rats exhibit consistent individual lateralization of running seizures (run directionality) induced by repeated binaural stimulation. Since this initial preconvulsive running reflects seizure onset in the auditory brainstem, the running asymmetry suggests non-symmetric early epileptic activation of brainstem substrates by sound in these rats. Repetition of the asymmetric brainstem seizures led to asynchronous recruiting the cortex into seizure network and lateralization of running seizures was predictive for asymmetry of early cortical seizure manifestations in Wistar and WAG/Rij rats. Both electrographic markers of AK, spreading depression (SD) and post-running afterdischarge, first appeared in the cortex ipsilateral to run direction, suggesting lateralized brainstem-to-forebrain seizure generalization during AK. At the population level, no bias in lateralization of running and SD was found in Wistar and WAG/Rij rats but incidence of secondary cortical seizures varied, depending on strain and run laterality. Among Wistar rats, cortical seizures developed more rarely in right-runners than in left-runners, suggesting enhanced resistance of the right hemisphere to epileptogenesis in rats of this strain. WAG/Rij rats with mixed (absence and audiogenic) epilepsy showed weak lateralization of early cortical seizures and no left-right difference in their incidence during AK. Present findings suggest (1) lateralized brainstem-to-forebrain seizure propagation and hemispheric difference in its facility in Wistar rats, (2) alterations of intra- and interhemispheric seizure propagation in WAG/Rij rats with genetic absence epilepsy.

Interhemispheric difference in susceptibility to epileptogenesis: evidence from the audiogenic kindling model in Wistar rats

Audiogenic kindling (AK) represents a model of naturally occurring epileptogenesis in which intensification of repeatedly induced audiogenic seizures results from propagation of epileptic activity from the brainstem to forebrain. Previously it has been shown that unilateral cortical spreading depression (SD) is a reliable earliest manifestation of mild AK produced by repetition of minimal audiogenic seizures (running) in Wistar rats. The unilateral triggering SD suggests the existence of asymmetry in the forebrain recruitment during the kindling and the present study examined whether epileptogenesis produced by this mild AK paradigm is a lateralized process. Twenty five running episodes were induced by brief sound stimulation in Wistar rats susceptible to audiogenic seizures. Behavioral and EEG correlates of AK development were assessed. Running behavior elicited by brief sound stimulation had an asymmetrical pattern with profound preference for one direction. Most rats expressing leftward running displayed full kindling development whereas the majority of rats with rightward running were resistant to AK. The EEG marker of AK, a cortical epileptiform discharge, was recorded only in rats with leftward running and the first discharge appeared in the left cortex. Cortical SD was recorded after repeated running seizures in all rats with reproducible audiogenic response irrespective of the running lateralization and propensity to kindling. Until the late kindling stages, SD was triggered unilaterally in the cortex ipsilateral to the running direction. These findings indicate intrinsically determined lateralization of epileptogenic process in the mild AK model and enhanced vulnerability of the left hemisphere to epileptogenesis.

Development of frequency-selective domains in inferior colliculus of normal and neonatally noise-exposed rats

Topographic patterns of pure-tone responses in inferior colliculus (IC) of Wistar rats were mapped using immunohistochemical staining for the nuclear protein Fos, the translation product of the c-fos proto-oncogene. Patterns were compared in ICs of immature and mature rats and in mature rats which experienced auditory deprivation beginning on day 14, an age near the developmental onset of hearing. Neonatal hearing losses, caused here by exposure to potentially deafening noise, are known to result in audiogenic seizure susceptibility in neonatal rats. These seizures can be triggered only by high-frequency stimuli and are believed to be initiated in IC. Thus, it seemed possible that susceptibility might depend on derangements of topographic frequency representation due to neonatal auditory deprivation. The band-like frequency-response domains, characteristic of adult IC, were found to be poorly differentiated in ICs of immature rats. On day 12, only lower-frequency stimuli induced discrete bands of Fos immunoreactivity while responses to higher frequencies remained exceptionally diffuse within ventral portions of IC. Only after day 24 did responses to the highest frequencies also appear mature. Furthermore, most significantly, adult rats which were transiently deafened on day 14, retained the more voluminous response patterns which were characteristic of immature IC. Because frequency selectivity in cochlea also develops by a low-to-high frequency sequence, results are consistent with a hypothesis that topographic organization arises in IC by an activity-dependent process. Whereas neonatal noise exposure also conferred audiogenic seizure susceptibility, it appears the arrest of tonotopic organization of IC is the probable basis of this reflex epilepsy.

Recovery of susceptibility to audiogenic seizure in mice

The applicability of the auditory fatigue and anoxia hypotheses for the refractory period observed after sound-induced seizure was examined in SJL/J mice. Duration of auditory stimulation was varied independent of attainment of clonic seizure activity by use of an acoustic interruption technique. Duration of the recovery period affected the pattern of preconvulsive running. Furthermore, the motor asymmetries exhibited during clonus remained consistent across tests. Because duration of acoustic stimulation and attainment of clonus did not affect recovery rate, we conclude that neither the auditory fatigue nor the anoxia hypothesis provides a complete account for the refractory period after audiogenic seizure. We suggest that an inhibitory process, activated before clonus occurs and perhaps linked to depletion of excitatory amino acids in the inferior colliculus, may also be involved.

Monaural audiogenic seizures: evidence for control by parallel processes

SJL/J mice were binaurally sensitized and monaurally tested for susceptibility to sound-induced seizure. Most control subjects exhibited the expected biphasic pattern of running. In mice in the experimental groups, the acoustic stimulation was interrupted for less than 20 s at the conclusion of the initial burst of running. During this quiet period, the side of the blocked ear was reversed for some mice. These mice had an independent biphasic seizure when acoustic stimulation was reintroduced, indicating that the biphasic seizure progression characteristic of monaural testing is the result of a lateralized process. We conclude that the locus of this process is probably at the level of the inferior colliculus (IC).

Audiogenic seizures in unilaterally sensitized and monaurally stimulated Wistar rats

Results of previous studies (Pierson, M. and Swann, J., Epilepsia, 32 (1991) 1-9) have demonstrated that exposure of Wistar rats to noise on day 14 results in audiogenic seizure susceptibility. Experiments reported here examined whether unilateral susceptibility could be induced in rats by monaural restriction of this noise exposure. Behavioral attributes of seizures on day 28 were compared in groups that were: binaurally noise-exposed/binaurally tested, binaurally noise-exposed/monaurally tested, monaurally noise-exposed/binaurally tested and monaurally noise-exposed/monaurally tested. Effects of left- and right-ear exposures and tests were assessed separately. Unilateral susceptibility was evident since seizures could be elicited later only by stimulation of the originally noise-exposed ear. Seizures were behaviorally different in monaurally noise-exposed and binaurally noise-exposed animals. Convulsions, directional reversals during running episodes, and relatively short latencies occur only in binaurally noise-exposed rats. These behaviors occur with either monaural or binaural stimulation. Initial, running direction was random in binaurally stimulated/binaurally noise-exposed rats, but was fixed in all other groups depending on which ear was exposed in either sensitization (day 14) or testing (day 28). Right- and left-ear sensitizations or tests resulted in left-directed and right-directed running onsets respectively. Previous studies of the effect of selective CNS lesions in instances of unilateral or bilateral susceptibility have led to the understanding that seizure initiation in unilaterally susceptible animals is mediated by the crossed ascending auditory pathway.(ABSTRACT TRUNCATED AT 250 WORDS)

Audiogenic seizures induce c-fos in a model of developmental epilepsy

In rats made susceptible to audiogenic seizures by exposing them to an intense noise at a critical time during development, subsequent noise exposure elicited seizures and induced the proto-oncogene c-fos in auditory regions of the brain. Cells showing Fos-like immunoreactivity were especially dense in dorsal and external cortices of the inferior colliculus, and were nearly absent after pretreatment with the N-methyl-D-aspartate (NMDA) antagonist MK-801. Noise exposure alone (i.e. no seizure) produced a localized zone of c-fos induction within the inferior colliculus, but only when presented during the time period when susceptibility to audiogenic seizures can be most effectively induced.

Possible interaction between the inferior colliculus and the substantia nigra in audiogenic seizures in Wistar rats

Male Wistar rats were tested for sensitivity to audiogenic seizures (AS; 110 dB), using an audiogenic severity index (SI). Sensitive (S) animals were subjected to bilateral lesion of the inferior colliculus (IC) and/or the lateral lemniscus (LL). Resistant (R) animals were subjected to bilateral lesions of the IC, unilateral sequential lesions of the substantia nigra reticulata (SN) and/or IC (contralateral to one another), and unilateral thalamic and sham lesions. Bilateral lesions of the IC and LL abolish AS in S rats. Lesion of the SN resulted in more pronounced sensitivity to AS than unilateral lesion of IC, in R rats. When the SN lesion was contralateral to a previous IC lesion, the effect was not only an increase in the SI, but also a reversal of the asymmetry generated by IC lesion. Although the behavioral effects resulting from IC lesions are due to alterations in the primary structures involved in the origin of AS, unilateral SN lesions can alter critical substrates of sensorimotor integration involved in the control and expression of AS.

Ontogenetic features of audiogenic seizure susceptibility induced in immature rats by noise

Although numerous models are currently used for systematic study of the mechanisms of epileptogenesis in mature brain, few animal models have been developed that allow similar explorations in the developing nervous system. One experimental model of epilepsy supports a premise that perinatal experience can lead to eventual seizure susceptibility, however. Audiogenic seizure (AGS) susceptibility can be induced during a critical developmental period in normal mice by auditory deprivation and therefore by cochlear trauma. We studied the developmental parameters that affect success of both induction and testing of AGS-susceptibility in the rat. Intense high-frequency noise exposure was used as the traumatizing agent. The Wistar rat strain used is inherently seizure-resistant because in greater than 400 trials, untreated rats have never exhibited susceptibility at any age. Although single prolonged exposures to high-intensity noise were administered to groups of rats at ages between postnatal days (PNDs) 12 and 36, PND 14 was the age when exposure was most likely to result in eventual seizure susceptibility. Furthermore, duration of initial exposure on PND 14 determined the rate of susceptibility when measured 2 weeks later. Accordingly, we noted that single noise exposures at an intensity of 125 dB and ranging between 6 and 10 min in duration induced susceptibility in 100% of rats tested on PND 28; nonetheless, seizures among the rats exposed for 8 min were the most severe. Typically, these seizures began as wild running attacks and were followed by tonic/clonic convulsions.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of GABA (gamma-aminobutyric acid) on neurotransmission in inferior colliculus slices from the guinea pig

Inferior colliculus (IC) slices from guinea pigs were prepared by cutting the IC parasagittally and incubating the obtained slices in the standard medium. Postsynaptic field potentials (PSFP) were recorded in the superficial layer of the pericentral nucleus (PN) by stimulating the deep layer of the PN. GABA (gamma-aminobutyric acid) and its agonists (muscimol and baclofen) and antagonist (bicuculline) were applied in the perfusion medium. GABA (1-12.5 mM), muscimol (1-200 microM) and (-)-baclofen (0.1-20 microM) reduced the amplitude of the PSFP in a dose-dependent manner. Bicuculline (1-10 microM) enhanced the amplitude of the PSFP and evoked the epileptiform burst discharge (EBD). The incubation of IC slices with methoxypyridoxin (100 microM) and 3-mercaptopropionic acid (100 microM) for 2 h reduced the GABA level of the slice to about 50% of the initial value, at which point a distinct EBD appeared. The application of GABA (5 mM) to the medium in this condition blocked the EBD completely. The removal of chloride from the medium as well as the application of furosemide (100 microM) enhanced the amplitude of the PSFP and also elicited the EBD; GABA application in this case did not block the EBD. These results indicate that GABA plays an inhibitory role in the IC, and that the reduction of GABA and the blockade of GABA function cause the appearance of EBD in the IC.

Effects of unilateral microinjections of GABAergic drugs into the inferior colliculus on auditory evoked potentials and on audiogenic seizure susceptibility

Alteration of GABAergic neurotransmission within the inferior colliculus (IC) appears to be involved in the generation of the audiogenic seizure (AGS) susceptibility. In the present study, we provide evidence indicating that this susceptibility may result from IC neuronal hyperexcitability to sound induced by a decreased GABAergic inhibition. In a first experiment, a unilateral microinjection of bicuculline, a GABAa antagonist, into the IC of normal rats increased the amplitude of the collicular auditory evoked potential, while the microinjection of THIP, a GABAa agonist, decreased this response. In a second experiment, a unilateral microinjection of bicuculline into the IC induced AGS susceptibility in normal rats.

Monaural and binaural audiogenic seizures in mice

The progression of sound-induced seizures was examined in unilaterally or bilaterally sensitized SJL/J mice tested either monaurally or binaurally. An unexpected right-side advantage for becoming susceptible to audiogenic seizure was observed. In addition, two distinct patterns of seizure progression were noted, a uniphasic sequence in which a single burst of running preceded the convulsion and a biphasic pattern with two such bursts. The biphasic progression is viewed to be the result of unilaterally initiated seizures and characteristically reached only a clonic level of severity. Uniphasic seizures are concluded to be the result of bilaterally initiated seizures and, when they occurred after more than 30 s of auditory stimulation, frequently reached a tonic level of intensity. The present results support the view that audiogenic seizures are characterized by precisely timed, sequential processes dependent upon the specific priming and test procedures employed.

The sensitive period and optimum dosage for induction of audiogenic seizure susceptibility by kanamycin in the Wistar rat

The incidence and severity of audiogenic seizures in kanamycin (KM)-treated rat pups, from a Wistar strain which is inherently seizure-resistant, was analyzed as a function of (a) postnatal age at the time of KM injection (i.p.) and (b) KM dosage. The vigor of the pinna reflex response on postnatal day (PND) 28 was correlated with (a) age at the time of injection, (b) dosage and (c) individual audiogenic seizure severity scores on PND 28 or PND 32. The data indicate that PNDs 9-12 are the developmental period when the rat has its greatest sensitivity to induction of susceptibility to audiogenic seizures by KM. The pinna reflex data suggest that cochlear vulnerability to KM intoxication is also greatest during this period. The optimum dosage for the induction of susceptibility was 100 mg/kg X 4 days. Use of higher doses resulted in a reduction of both incidence and severity of audiogenic seizures. The pinna reflex generally exhibited a supranormal vigor in animals having the most severe seizures. The behavioral attributes of induced audiogenic seizures at postnatal ages of 28 and 32 days are described and discussed.

Neuroanatomical localization of structures responsible for seizures in the GEPR: lesion studies

Identification of the neural substrates subserving audiogenic convulsions in the GEPR is an important task and while it is not yet complete, many laboratories employing various techniques have contributed importantly to our current understanding. The present review focuses on the use of lesions to identify the neural substrates of audiogenic convulsions. Lesions in brain stem nuclei appear to have a much greater ability to attenuate audiogenic convulsions than do forebrain lesions. In fact, some forebrain lesions (dorsal hippocampus, caudate, intralaminar thalamic nuclei) appear to enhance the severity of audiogenic seizures. On the other hand, bilateral lesions in the inferior colliculus (IC) have been shown to completely abolish audiogenic convulsions, while lesions in the pontine reticular formation (PRF nucleus) abolish all aspects except the running episode suggesting that these two brain stem structures are important neural substrates involved in the expression of audiogenic convulsions. Large bilateral lesions of the substantia nigra also appear to attenuate audiogenic convulsions. The effect of lesions on audiogenic convulsions is basically similar to their effect on other generalized seizure models and the data appear to support the hypothesis that there are two anatomical systems involved in the expression of all generalized convulsions: a forebrain system responsible for the expression of face and forelimb clonus; and a brain stem system responsible in the expression of running-bouncing clonus and tonus.

Modulation of alpha 1-and alpha 2-adrenoceptor binding sites in the brain of audiogenic seizure susceptible mice (DBA/2J)

The binding of [3H]dihydroalprenolol ([3H]DHA), [3H]prazosin and [3H]clonidine was assayed in whole brain and various brain regions of audiogenic seizure (AS) susceptible DBA/2J (D2) mice aged 10, 24 and 50 days (i.e. before, during and after their period of AS susceptibility, respectively) and in age-matched C57BL/6J (B6) controls. In whole brain, at 24 days, [3H]DHA binding was similar in the two strains, while the binding of [3H]prazosin and [3H]clonidine was significantly lowered in D2 mice. No difference could be detected in 10 and 50 day old mice with any of the ligands. Regional studies indicated an involvement of the cerebral cortex, the olfactory bulbs and the brain-stem. alpha- (but not beta-)adrenoceptor changes were concomitant with the AS susceptibility period. These changes were unevenly distributed in the brain of D2 mice; they suggest that alpha 1- and alpha 2-adrenoceptor subtypes might play different roles in the AS of the D2 mouse strain.

Hippocampal lesions increase the severity of unilaterally induced audiogenic seizures and decrease their latency

The effect of unilateral and bilateral lesions of the hippocampus was examined in unilaterally tested SJL/J mice. The data indicate that lesions of the hippocampus decrease the latency to and increase the severity of audiogenic seizures.

A comparison of short-latency auditory-evoked potentials in two strains of mice: possible neurophysiological correlates of susceptibility to audiogenic seizures?

The short-latency auditory-evoked response was recorded from adult (over 8 wk) and young (16-22 day) mice of the audiogenic seizure-prone DBA/2J (D2) and seizure-resistant C57BL/6J (C6) strains. An auditory complex made up of eleven peaks and troughs was found within the first 20 ms after click stimuli. The latencies of the potentials tended to be shorter in the C6 mice and in the adults of both strains. The ratio of amplitude differences of late to early peaks ('amplitude index') was much larger (P less than 0.01) in D2 mice. The data are consistent with greater neuronal recruitment to acoustic stimuli in the higher-order auditory centers of D2 mice.

Effect of cortical spreading depression on audiogenic seizure priming of C57BL/6 mice

At 19 days of age, C57BL/6Bg mice received KCl-induced cortical spreading depression during which they were acoustically primed by exposure to an initial auditory stimulus. At 28 days of age, the mice were tested for susceptibility to audiogenic seizures. Cortical spreading depression had no effect on acoustic priming of C57BL/6Bg mice and it had been previously reported to have no effect on acoustic priming of SJL/J mice. These findings are discussed in the context of pharmacogenetic differences for the effects of aminooxyacetic acid on acoustic priming of C57BL/6 and SJL/J mice.