Neuroinflammation is a major feature of most neurodegenerative conditions, and can leadto the exacerbation of neuronal injury. Inflammatory challenges in the central nervoussystem (CNS) have been shown to activate peripheral immune cells, which subsequentlyinfiltrate into the brain. Concurrently, resident inflammatory cells in the CNS, such asmicroglia, become activated and release inflammatory mediators, including cytokines.The pro-inflammatory cytokine interleukin-1 (IL-1) is a key mediator of neuronal injury.Although two IL-1 agonists exist, IL-1α and IL-1β, the majority of research has focussedon the contribution of IL-1β to neuronal injury. Excitotoxic cell death in the rat brain,induced by striatal injection of the glutamate agonist AMPA, is exacerbated by coadministrationof recombinant IL-1β. To identify possible mediators which facilitate theexacerbation of neuronal injury by IL-1 this study investigated the early peripheral andcentral mediators of inflammation in response to AMPA + IL-1β.Neutrophil infiltration and increased neuronal activity were found to be present at 4h post-AMPA + IL-1β injection, which lead to the induction of microglial IL-1α in the ipsilateralcortex, in the absence of any IL-1β expression. To target the peripheral neutrophil responsean intervention study was performed to inhibit peripheral TNFα, which is thought tomobilise neutrophils. No significant effect of pre-treatment with etanercept, a TNFαinhibitor, was observed on neuronal injury produced in response to AMPA + IL-1β, thougha slight trend for protection was seen. To target the central IL-1α response after AMPA +IL-1β treatment an anti-IL-1α antibody was injected directly into the cerebral cortex, butthis had no effect on AMPA + IL-1β induced cell death. Therefore, using a reductionist invitro approach in organotypic slice cultures haemin, an inducer of endogenous IL-1α, wasused to investigate IL-1α mediated cell death. Haemin induced cell death was shown to beIL-1 dependent and preliminary studies using IL-1αKO mice indicated that IL-1α maypartially mediate this effect. This suggests that in the AMPA + IL-1β paradigm IL-1α is thedominant IL-1 isoform early after AMPA + IL-1β treatment, which can trigger subsequentneuronal cell death, as a result of the additive effects of neutrophil infiltration and highneuronal activity in the cortex. This study highlights the potential therapeutic value ofinhibiting IL-1α expression early following acute neuronal injury.

GABAA receptor binding is transiently increased in rat whisker barrels during the second postnatal week, at a time when neurons in the developing rat cortex are vulnerable to excitotoxic effects. To test whether these GABAA receptors might serve to protect neurons from excessive excitatory input, polymer implants containing the GABAA receptor antagonist bicuculline were placed over barrel cortex for a 4-day period in young (postnatal days 8 - 12) and adult rats. In the cortex of young, but not adult rats, the chronic blockade of GABAA receptors resulted in substantial tissue loss and neuron loss. The greater loss of neurons in young rats supports the hypothesis that a high density of GABAA receptors protects neurons from excessive excitatory input during a sensitive period in development.

In Syrian hamsters, social behavior is mediated exclusively by chemosensory cues and circulating gonadal steroid hormones. Where these two signals are processed in the brain is unknown, but the posterior bed nucleus of the stria terminalis (pBNST) has been suggested as a candidate site. Therefore, we tested male hamsters’ preference for opposite-sex odors following excitotoxic lesions of the pBNST. Lesions of the pBNST (pBNST-X) eliminated male hamsters’ preference for opposite-sex odors. Furthermore, pBNST-X males spent significantly less time investigating female odors than clean odors and significantly less time investigating female odors than control males did. Lesions of the pBNST did not change male hamsters’ investigation of male odors. The deficits observed in pBNST-X males were not due to a failure to discriminate between odors, as pBNST-X males were able to distinguish between odors. Together, these data suggest the pBNST is critical for opposite-sex odor preference in male hamsters.

In Syrian hamsters, social behavior is mediated exclusively by chemosensory cues and circulating gonadal steroid hormones. Where these two signals are processed in the brain is unknown, but the posterior bed nucleus of the stria terminalis (pBNST) has been suggested as a candidate site. Therefore, we tested male hamsters’ preference for opposite-sex odors following excitotoxic lesions of the pBNST. Lesions of the pBNST (pBNST-X) eliminated male hamsters’ preference for opposite-sex odors. Furthermore, pBNST-X males spent significantly less time investigating female odors than clean odors and significantly less time investigating female odors than control males did. Lesions of the pBNST did not change male hamsters’ investigation of male odors. The deficits observed in pBNST-X males were not due to a failure to discriminate between odors, as pBNST-X males were able to distinguish between odors. Together, these data suggest the pBNST is critical for opposite-sex odor preference in male hamsters.

Kainic acid induced excitotoxicity causes pyramidal cell death in the CA3a/b region of the hippocampus. Electrical synapses, gap junctional communication, and single membrane channels in non-junctional membranes (hemichannels) composed of connexin36 (Cx36) have been implicated in both seizure propagation and the spread of excitotoxic cell death. In rats, Cx36 protein is expressed by pyramidal neurons. Localization of protein in mouse, however, is highly controversial. Expression is reported to be restricted to hippocampal interneurons yet the same excitotoxic mechanisms (electrical and metabolic coupling between pyramidal neurons) are invoked to explain the role of Cx36 in excitotoxic pyramidal loss in murine brain. To address this controversy, I show by confocal immunofluorescence and in situ hybridization that Cx36 protein expression is restricted to interneurons and microglia in murine hippocampus and is not expressed by, or is below level of detection in pyramidal neurons. Using behavioural and electrophysiological measures, seizure propagation was found to be moderately enhanced in the absence of Cx36 likely due to the loss of interneuron-mediated synchronous inhibition of the pyramidal cells. Further, CA3a/b neurons die post kainic acid injury in the presence of Cx36 but are protected in Cx36-/- mice. When delayed excitotoxic cell death is maximal, Cx36 is primarily expressed by activated microglia as demonstrated by confocal immunofluorescence, in situ hybridization, and Western blotting. These activated microglia are located in the direct vicinity of, and surrounding cells in the damaged Ca3a/b region. Finally, I show that loss of Cx36 from activated microglia in mice is sufficient to prevent excitotoxic cell death in the CA3a/b with surviving neurons functional as assessed by both electrophysiological and behavioural measures. Together, these data identify a new mechanism of excitotoxic injury, mediated by neuronal-glial interactions, and dependent on microglial Cx36 expression.

No description available.

Glutamate excitotoxicity is the predominant mechanism implicated in neuronal cell death associated with neurological disorders such as stroke, epilepsy, traumatic brain injury and ALS. Excessive stimulation of NMDA subtypes of glutamate receptors leads to protracted intracellular calcium elevations triggering calcium mediated neurotoxic mechanisms culminating in delayed neuronal cell death. In addition, glutamate excitotoxicity induces a NMDA dependent extended neuronal depolarization mediated by continuous calcium influx that correlates with delayed neuronal death. Attempts to prevent neuronal death by blocking calcium entry into the neurons using calcium channel blockers or NMDA receptor antagonists have failed to provide any beneficial effects in clinical trials. Thus, calcium continues to enter the neurons despite the presence of calcium entry blockers. This phenomenon is known as the "calcium paradox of stroke" and represents a major problem in developing effective therapies for treatment of stroke. Here employing a combination of patch clamp recordings, fluorescent calcium imaging and neuronal cell death assays in well-characterized in vivo and in vitro models of glutamate excitotoxicity, we report the discovery of a novel calcium permeable ion channel that is activated by excitotoxic glutamate injury and mediates a calcium current that is an early initiating step in causing neuronal death. Blocking this calcium permeable channel with high concentrations of Zn2+ or Gd3+ by removing extracellular calcium for a significant time period after the initial injury is effective in preventing calcium entry, apoptosis and neuronal death, thus accounting for the calcium paradox. This injury induced-calcium permeable channel provides a unique mechanism for calcium entry following stroke and offers a new target for extending the therapeutic window for preventing neuronal death after the initial excitotoxic (stroke) injury.

Many cyanobacteria are reported to produce the nonprotein amino acid β-N-methylamino-L-alanine (BMAA). Cyanobacteria are extensively distributed in terrestrial and aquatic environments and recently BMAA was detected in temperate aquatic ecosystems, e.g. the Baltic Sea. Little is known about developmental effects of the mixed glutamate receptor agonist BMAA. Brain development requires an optimal level of glutamate receptor activity as the glutamatergic system modulates many vital neurodevelopmental processes. The aim of this thesis was to investigate the developmental neurotoxicity of BMAA, and its interaction with the pigment melanin. Autoradiography was utilized to determine the tissue distribution of 3H-labelled BMAA in experimental animals. Behavioral studies and histological techniques were used to study short and long-term changes in the brain following neonatal exposure to BMAA. Long-term changes in protein expression in the brain was also investigated using matrix-assisted laser desorption ionization (MALDI) imaging mass spectrometry (IMS). A notable targeting of 3H-BMAA to discrete brain regions e.g. hippocampus and striatum in mouse fetuses and neonates was determined by autoradiography. BMAA treatment of neonatal rats on postnatal days 9–10 induced acute but transient ataxia and hyperactivity. Postnatal exposure to BMAA also gave rise to reduced spatial learning and memory abilities in adulthood. Neonatal rat pups treated with BMAA at 600 mg/kg showed early neuronal cell death in the hippocampus, retrosplenial and cingulate cortices. In adulthood the CA1 region of the hippocampus displayed neuronal loss and astrogliosis. Lower doses of BMAA (50 and 200 mg/kg) caused impairments in learning and memory function without any acute or long-term morphological changes in the brain. The MALDI IMS studies, however, revealed changes in protein expression in the hippocampus and striatum suggesting more subtle effects on neurodevelopmental processes. The studies also showed that BMAA was bound and incorporated in melanin and neuromelanin, suggesting that pigmented tissues such as in the substantia nigra and eye may be sequestering BMAA. In conclusion, the findings in this thesis show that BMAA is a developmental neurotoxin in rodents. The risks posed by BMAA as a potential human neurotoxin merits further consideration, particularly if the proposed biomagnifications in the food chain are confirmed.

Ce travail de thèse avait pour objectif d’étudier le rôle de l’hippocampe (HPC) ventral et des noyaux reuniens (Re) et rhomboïde (Rh) du thalamus dans les processus cognitifs qui sous-tendent la mémoire spatiale chez le Rat. Par l’utilisation d’approches complémentaires combinant l’imagerie cérébrale, la lésion excitotoxique, l’inactivation fonctionnelle réversible et des évaluations comportementales, nos résultats ont mis en évidence : (1) l’implication spécifique de l’HPC ventral uniquement dans le rappel d’informations spatiales ; (2) un rôle-clé des noyaux Re et Rh dans la persistance d’un souvenir spatial ; (3) l’implication des noyaux Re et Rh dans le labyrinthe du double-H, un nouveau test nécessitant d’une part, l’utilisation d’informations spatiales dépendant de l’intégrité de l’HPC dorsal, et d’autre part, une flexibilité comportementale, impliquant le cortex préfrontal médian. Ainsi, l’ensemble de ces résultats permet de proposer l’existence d’un circuit HPC-préfronto-thalamique impliqué dans divers aspects du traitement des informations spatiales. / The main objective of this thesis was to investigate the role of the ventral hippocampus (HPC) and the reuniens (Re) and rhomboid (Rh) thalamic nuclei in the cognitive processes underlying spatial memory in the Rat. If our results first confirmed, in the Morris water maze, the role of the dorsal HPC in the acquisition and retrieval of a spatial reference memory, we demonstrated the specific involvement of the ventral HPC only in the recall of spatial information. In addition, by using complementary approaches combining brain imaging, excitotoxic lesion and reversible functional inactivation, we were able to show for the first time a key role for the Re and Rh in the persistence of a spatial memory (25 days). Finally, the third set of experiments has highlighted the involvement of the Re and Rh in a mnemonic task performed in a new test, the double-H maze, which requires the use of spatial information depending on the integrity of the dorsal HPC, and a behavioral flexibility, involving the medial prefrontal cortex. Thus, taken together, these results suggest the involvement of a HPC-prefronto-thalamic network in various aspects of spatial information processing.

No description available.