The disorders of the glutamatergic neurotransmission have been associated with pathogenesis of autism. In this study we evaluated the impact of the in vivo and ex vivo test methodology on measurements of levels of neurotransmitter amino acids in hippocampus of rats for valproic acid- (VPA) and thalidomide- (THAL) induced models of autism. The main goal was to compare the changes in concentrations of glutamate (Glu), glutamine (Gln) and GABA between both autistic groups and the control, measured in vivo and ex vivo in homogenates. The rat pups underwent three in vivo tests: ultrasonic vocalization (USV), magnetic resonance spectroscopy (MRS) and unilateral microdialysis of the hippocampus. Analyses of homogenates of rat hippocampus were performed using high-performance liquid chromatography (HPLC) and nuclear magnetic resonance (NMR) spectroscopy. For the statistical analysis, we performed univariate and multivariate tests. USV test, which is considered in rodents as an indicator of pathology similar to autism, showed decreased USV in VPA and THAL groups. In vivo MRS studies demonstrated increases of Glu content in male rat’s hippocampus in VPA and THAL groups, while the microdialysis, which allows examination of the contents in the extracellular space, detected decreases in the basal level of Gln concentrations in VPA and THAL groups. Ex vivo HPLC studies showed that levels of Glu, Gln and GABA significantly increased in male rat’s hippocampus in the VPA and THAL groups, while NMR studies showed increased levels of Gln and GABA in the VPA group. Collectively, these results are consistent with the hypothesis suggesting the role of the glutamatergic disturbances on the pathogenesis of autism. For all methods used, the values of measured changes were in the same direction. The orthogonal partial least square discriminant analysis confirmed that both animal models of autism tested here can be used to trace neurochemical changes in the brain.

Elevated homocysteine is a risk factor in cardiovascular diseases and neurodegeneration. Among the putative mechanisms of homocysteineevoked neurotoxicity, disturbances in methylation processes and NMDA receptor-mediated excitotoxicity have been suggested. Our previous studies demonstrated that group I metabotropic glutamate receptors along with NMDA receptors participate in acute homocysteine-induced neuronal damage. In this study, using propidium iodide staining, we tested whether the same mechanism may mediate chronic homocysteine neurotoxicity. Our results confirmed that the application of D,L-homocysteine in micromolar concentrations for 3 days induces neurodegeneration in primary cultures of cerebellar granule neurons. Uncompetitive NMDA receptor antagonist MK801, and mGlu1 or mGlu5 receptor antagonists (LY367385 and MPEP, respectively), given alone provided very limited neuroprotection. However, simultaneous application of the NMDA receptor antagonists MK-801, memantine or amantadine and MPEP almost completely prevented chronic homocysteine neurotoxicity. These findings suggest a novel therapeutic strategy to combat neurodegeneration induced by hyperhomocysteinemia comprising a combination of antagonists of group I metabotropic glutamate receptors and NMDA receptors. Correspondence should be addressed to J.W. Lazarewicz, Email: jerzyl@cmdik.pan.pl