`Dormant' inhibitory neurons: do they exist and what is their functional impact?
Introduction
The hypothesis of `dormant' inhibitory neurons in epilepsy was developed to explain the disruption of functional inhibition, in the presence of viable inhibitory neurons and of functional GABAA receptors. The idea was proposed by R.S. Sloviter on the basis of his experiments using prolonged stimulation of the perforant path to replicate status epilepticus (Sloviter, 1987Sloviter, 1991). In these experiments, initially performed in the dentate area, he demonstrated a loss of paired pulse inhibition, despite the presence of a normal number of GABA positive basket cells. The pathology that he found consisted of a loss of mossy cells and of somatostatin and neuropeptide-Y containing neurons in the hilus. Sloviter (1987)proposed that the loss of mossy cells would reduce the excitatory drive to the basket cells and other interneurons that would normally inhibit granule cells. In a subsequent paper, Sloviter (1991)demonstrated a similar loss of paired pulse inhibition in the CA1 region in the face of preserved pyramidal cells and interneurons. In the latter case he proposed that the loss of CA3 pyramidal cells reduced the excitatory drive to CA1 inhibitory neurons, diminishing their responsiveness, and thereby weakening paired pulse inhibition.
Bekenstein and Lothman (1993)tested this hypothesis of reduced interneuronal responsiveness in the CA1 region. They used intracellular recordings in brain slices from rats made chronically epileptic by repetitive stimulation some 2–5 months previously. Their evidence for functional disconnection of the interneurons came from comparing the responses to stimulation near to (≲200 μm) and remote from (>1 mm) the impaled CA1 pyramidal cell. In control tissue, stimulation produces EPSPs, fast IPSPs and slow IPSPs, whether the stimulus is close to or remote from the recorded cell. However, in slices from chronically epileptic rats, stimulation of the remote sites failed to elicit IPSPs but still produced a fast EPSP. In contrast, stimulation within a few hundred microns of the impaled cell produced normal looking sequences of fast and slow IPSPs in the presence of drugs that block EPSPs. Bekenstein and Lothman (1993)interpreted this result to show (a) that interneuron-mediated inhibition onto pyramidal cells could be elicited directly when the stimulation site was close to the pyramidal cell, indicating that inhibitory neurons were present and viable; but (b) that these interneurons could not be excited to fire through afferents normally reaching them from more remote parts of the CA1.
Changes resembling dormancy were also found in the kainate-lesion model. Injection of kainic acid into the ventricles leads to acute epileptic seizures and focal neuronal loss, typically including the hippocampal CA3 region. In the longer term, this acute phase can be followed by the development of chronic epileptic seizures (Ben-Ari et al., 1980Lothman et al., 1981Nadler, 1981). During this chronic epileptic phase there is evidence that inhibitory neurons are present and that their synapses onto pyramidal cells can work, but that they are activated more weakly than normal. Nakajima et al. (1991)used paired intracellular recording in the CA1 region to show that the frequency of synaptic coupling between pyramidal cells and interneurons was less than in non-epileptic tissue slices, and that there was no marked increase in excitatory connections between pyramidal cells. In their study, evoked inhibition was weak, but recovered with very strong stimulation that could activate interneurons directly. In another study of the kainate model, IPSPs were reduced in the hyperexcitable parts of slices prepared from chronically epileptic rats, but pharmacologically isolated monosynaptic IPSPs remained normal, again pointing to a functional weakness in the activation of interneurons (Williams et al., 1993).
Here we will describe evidence from a different chronic experimental epilepsy, the intrahippocampal tetanus toxin model, that supports the existence of dormant interneurons. We also describe an acute experimental model to demonstrate the concept of partial disconnection of inhibitory neurons from their excitatory inputs.
Section snippets
Methods
Adult male rats received intrahippocampal injections of <50 ng of tetanus toxin (kind gift of Wellcome Biotech, Beckenham, UK) while under general anaesthesia with a mixture of Hypnorm and Hypnovel (Whittington and Jefferys, 1994). The injection site was 2.8 mm caudal to bregma, 3.5 mm lateral to midline and 3.5 mm below cortical surface. Several days after recovery they developed an epileptic syndrome, which has been described in detail elsewhere (Mellanby et al., 1977, Hawkins and Mellanby,
Results
Rats which had received 50 ng of tetanus toxin injected into the hippocampus developed epileptic seizures that recurred intermittently over a period of up to 10 weeks after the injection. Each of these seizures lasted less than 2 min, possibly explaining the absence of consistent neuropathology in any of the hippocampal regions, either in the pyramidal cell population (Traub et al., 1994) or in the GABAergic interneurons (Najlerahim et al., 1992). We have also found that the size distribution
Acute model: low magnesium epilepsy
It has been argued that a partial loss of excitation to inhibitory neurons will have minimal impact on epileptic discharge because of the massive excitation that the interneurons should receive when pyramidal cells fire synchronously. In our work on the low-[Mg2+]o model of epilepsy, we used selective glutamate receptor antagonists to dissect out the roles of different classes of receptor in the generation of epileptic events. Incubating brain slices in ACSF containing low-[Mg2+]o produces
Discussion
The evidence from the tetanus toxin chronic epilepsy argues for a loss of function of those inhibitory neurons that remain present in the epileptogenic region. The details differ from dormant interneurons as described in chronic models of epilepsy induced by status epilepticus or kainic acid, where gross lesions occur (Nakajima et al., 1991Bekenstein and Lothman, 1993Williams et al., 1993Sloviter, 1994), in that there is no early anatomical loss of neurons in the tetanus toxin model. Recently
Acknowledgements
RDT is a Wellcome Principal Research Fellow. This work was funded by grants from the Wellcome Trust and Human Frontiers.
References (42)
- et al.
Magnesium-free medium activates seizure-like events in the rat hippocampal slice
Brain Res.
(1986) - et al.
The role of epileptic activity in hippocampal and "remote" cerebral lesions induced by kainic acid
Brain Res.
(1980) - et al.
Magnesium ions block an N-methyl-d-aspartate receptor-mediated component of synaptic transmission in rat hippocampal slices
Neurosci. Lett.
(1985) - et al.
Relative abundance of subunit mRNAs determines gating and Ca2+ permeability of AMPA receptors in principal neurons and interneurons in rat CNS
Neuron
(1995) - et al.
Ex vivo release of GABA from tetanus toxin-induced chronic epileptic foci decreased during the active seizure phase
Neurochem. Int.
(1991) - et al.
Mechanisms shaping glutamate-mediated excitatory postsynaptic currents in the CNS
Curr. Opin. Neurobiol.
(1994) - et al.
Presynaptic inhibition in the hippocampus
Trends Neurosci.
(1993) - et al.
N-methylaspartate receptors mediate epileptiform activity evoked in some, but not all, conditions in rat neocortical slices
Neuroscience
(1986) - et al.
Epileptiform activity in combined slices of the hippocampus, subiculum and entorhinal cortex during perfusion with low magnesium medium
Neurosci. Lett.
(1986) - et al.
Monosynaptic GABA-mediated inhibitory postsynaptic potentials in CA1 pyramidal cells of hyperexcitable hippocampal slices from kainic acid-treated rats
Neuroscience
(1993)
Dormancy of inhibitory interneurons in a model of temporal lobe epilepsy
Science
NMDA and non-NMDA receptors are co-localized at individual excitatory synapses in cultured rat hippocampus
Nature
Tetanus toxin: convulsant action on mouse spinal cord neurons in culture
J. Neurosci.
Recombinant ionotropic glutamate receptors: functional distinctions imparted by different subunits
Cell. Physiol. Biochem.
Paired-pulse depression of monosynaptic GABA-mediated inhibitory postsynaptic responses in rat hippocampus
J. Physiol. (Lond.)
Operative GABAergic inhibition in hippocampal CA1 pyramidal neurons in experimental epilepsy
Proc. Natl. Acad. Sci. USA
The action of calcium on the electrical properties of squid axons
Physiol. (Lond.)
Limbic epilepsy induced by tetanus toxin: a longitudinal electroencephalographic study
Epilepsia
Rat hippocampal kindling induces changes in the glutamate receptor mRNA expression patterns in dentate granule neurons
Eur. J. Neurosci.
Kainic acid-induced seizures: electrophysiologic studies
Neurology
Dual-component miniature excitatory synaptic currents in rat hippocampal CA3 pyramidal neurons
J Neurophysiol.
Cited by (19)
A mechanistic appraisal of cognitive dysfunction in epilepsy
2012, Neuroscience and Biobehavioral ReviewsCitation Excerpt :They are associated with a more prolonged depolarization and the gradual loss of the post PDS after-hyperpolarization. After discharges are believed to be sustained by the activation of NMDA receptor subtypes (Jefferys and Traub, 1998), providing prolonged depolarization of the neuronal dendrites, resulting in rapid regenerative Na+ and Ca2+ dendritic spikes, which in turn drive bursts of action potentials at the cell soma (Traub et al., 1993). After-discharges are associated with significant rises in extracellular K+.
Mossy cells in epilepsy: Rigor mortis or vigor mortis?
2002, Trends in NeurosciencesThe α7-nicotinic acetylcholine receptor and the pathology of hippocampal interneurons in schizophrenia
2000, Journal of Chemical Neuroanatomy