P1: GRA Cellular and Molecular Neurobiology [cemn]
pp690-cemn-456078
December 16, 2002
20:23
Style file version Oct 23, 2000
C 2002) Cellular and Molecular Neurobiology, Vol. 22, Nos. 5/6, December 2002 (°
Glutamate NMDA Receptor Subunit R1 and GAD mRNA Expression in Human Temporal Lobe Epilepsy ˜ P. Leite,4,5 Luciano Neder,1 Valeria Valente,2 Carlos G. Carlotti Jr.,3 Joao 3 2 ˜ A. Assirati, Maria L. Pac¸o-Larson, ´ Joao and Jorge E. Moreira2 Received August 13, 2002; accepted September 13, 2002 SUMMARY 1. Molecular mechanisms underlying increased hippocampal excitability in human temporal lobe epilepsy (TLE) are largely unknown. A disturbance of the imbalance between excitatory and inhibitory neurotransmission pathways in the epileptic hippocampus may contribute substantially to a decreased seizure threshold. 2. We have extended the investigation whether TLE is associated with changes in the expression of GAD67 and NMDAR1 by assessing the relative amounts of the mRNAs in human hippocampal samples by means of semiquantitative RT-PCR. The samples included 16 hippocampal slices obtained at surgery from intractable TLE (HS, n = 14; non-HS, n = 2) and 3 postmortem control hippocampi. 3. The ratio for the GAD/NMDAR1 transcripts was significantly higher in TLE cases when compared to the nonepileptic samples. Such findings are mainly a consequence of the increased amounts of GAD mRNA detected in the epileptic hippocampus. Compared with nonepileptic samples, and without correction for neuron losses, the amounts of NMDAR1 mRNA in HS are slightly reduced, and in the non-HS samples they are significantly increased, which is consistent with an increase of NMDAR1 in the hippocampal remaining neurons, as previously reported. 4. Our results also contribute to the indication of GAD67 mRNA upregulation in human TLE. A possible functional implication for the increased GAD mRNA levels could be a mechanism to reduce neuronal hiperexcitability, synchronization, and/or the spread of seizure. KEY WORDS: hippocampus; temporal lobe epilepsy; NMDAR1; GAD; RT-PCR.
1 Department
of Pathology, Ribeirao ˜ Preto School of Medicine, University of Sao ˜ Paulo, Brazil. of Cellular and Molecular Biology, Ribeirao ˜ Preto School of Medicine, University of Sao ˜ Paulo, Brazil. 3 Department of Surgery, Ribeirao ˜ Preto School of Medicine, University of Sao ˜ Paulo, Brazil. 4 Department of Neurology, Ribeirao ˜ Preto School of Medicine, University of Sao ˜ Paulo, Brazil. 5 To whom correspondence should be addressed at Department of Neurology, University of Sao ˜ Paulo School of Medicine at Ribeirao ˜ Preto, Campus Universitario, ´ Ribeirao ˜ Preto, CEP 14049-900, Brazil; e-mail:
[email protected]. 2 Department
689 C 2002 Plenum Publishing Corporation 0272-4340/02/1200-0689/0 °
P1: GRA Cellular and Molecular Neurobiology [cemn]
690
pp690-cemn-456078
December 16, 2002
20:23
Style file version Oct 23, 2000
´ Neder, Valente, Carlotti, Leite, Assirati, Pac¸o-Larson, and Moreira
INTRODUCTION Hippocampal sclerosis (HS) is the most common neuropathologic finding in patients with intractable temporal lobe epilepsy (TLE) and appears to be centrally involved in the generation and propagation of epileptic activity. Typical pathologic features include atrophy and neuron loss of Ammon’s horn subfields, and aberrant synaptic reorganization of excitatory and inhibitory collaterals of the fascia dentata. More discrete axonal reorganization and milder neuron loss are found in hippocampi from patients with neocortical lesions, such as temporal cortical dysplasia and/or tumor (non-HS). Glutamatergic neurotransmission is involved as part of several physiological and pathological conditions with a critical role in seizures (Blumcke ¨ et al., 1999), in ischemic neuronal death after stroke (Wahlestedt et al., 1993), and is suggested to be involved in the pathogenesis of neurodegenerative disorders (Coyle and Puttfarcken, 1993). In particular, the ionotropic N-methyl-D-aspartate (NMDA) receptor, especially the receptor Subunit 1 (NR1), plays a central role in excitoxicity and in the epileptic phenomena (Kraus et al., 1994; Musshoff et al., 2000). Based on their structural and electrophysiological properties, the NMDAR1 and four NMDAR2a–d comprise the NMDA receptor gene families (Hollmann and Heinemann, 1994; Sommer and Seeburg, 1994). In situ hybridization (ISH) studies, specifically looking at hippocampal NMDA and AMPA receptor mRNAs in TLE patients, found that their levels may be decreased, unchanged, or increased compared with control necropsies (Bayer et al., 1995; Garc´ıa-Ladona et al., 1994; Lee et al., 1994; Mathern et al., 1997). This may be partly related to the research technique used and/or due to pyramidal cell loss. For example, hybridization studies showed that NMDA receptor activity is increased in fascia dentata and decreased in Ammon’s horn of sclerotic hippocampi. However, correcting for neuronal losses, HS patients have increased AMPA and NMDA mRNAs levels per neuron, in comparison to necropsies (Mathern et al., 1997, 1998). A frequently proposed hypothesis postulates that an imbalance between excitatory and inhibitory neurotransmission pathways in the epileptic hippocampus may contribute to a decreased seizure threshold (Blumck ¨ et al., 2000; Mcdonald et al., 1991; Mcnamara, 1999). γ -aminobutyric acid (GABA) is the major inhibitory neurotransmitter in the mammalian CNS. Some authors speculate that a decrease of inhibitory action on principal hippocampal neurons could be the major factor for epileptogenesis in human TLE and animal models of chronic limbic seizures (Babb et al., 1989; Cavalheiro, 1990; Esclapez et al., 1997; Mathern et al., 1995). However, the level of functional activity of GABA neurons in the epileptic hippocampal formation continues to be debated (Esclapez and Houser, 1999). Hence, it is conceivable that the glutamate/GABA ratio or its receptor subunits/synthetizing enzymes might exert a greater influence on paroxysmal neuronal discharges generation than any component per se (Ure and Perassolo, 2000). In the present study, we evaluate in the same hippocampal slice from TLE patients (HS and non-HS), the GABAsynthesizing enzyme (GAD67 ), and NMDAR1 mRNA expression by semiquantitative reverse-transcriptase—polymerase chain reaction (RT-PCR) and compare with controls (necropsies).
P1: GRA Cellular and Molecular Neurobiology [cemn]
pp690-cemn-456078
December 16, 2002
20:23
Style file version Oct 23, 2000
NMDA Receptor and GAD mRNA Expression in Temporal Lobe Epilepsy
691
MATERIAL AND METHODS Patient Evaluation and Tissue Collection Patients were evaluated at Ribeirao ˜ Preto Epilepsy Surgery Program using standardized protocols previously published and approved by the local Ethics Committee (Leite et al., 2000). Cases for this study were selected from temporal lobectomies in patients with intractable complex partial seizures of temporal lobe origin. Evaluation included a detailed history and neurological examinations, interictal scalp EEG, interictal/ictal video-EEG monitoring, an extensive neuropsychological test battery, and intracarotid amobarbital injections (Wada test) for memory and speech representation (when appropriated). Neuroimaging workup included high-resolution MRI with special protocols to visualize the hippocampus, ictal, and interictal single photon emission tomography (SPECT) scans. Preoperative data localized the focus to the anterior temporal region and patients were referred for a standardized “en bloc” resection including 3–4 cm of the hippocampus. Surgical patients were catalogved based on preoperative clinical, EEG and neuroimaging studies into those with hippocampal sclerosis (HS, n = 14) and those with temporal neocortical lesions, such as temporal cortical dysplasia and/or tumor (non-HS, n = 2). Both non-HS patients exhibited a dysembrioplasic neuronal tumor (WHO Grade I) in adjacent temporal lobe. The patient clinical profile is summarized in Table I. Control hippocampi were collected from subjects with no history of neurological disease, within
