Distribution of glutamatergic receptors and GAD mRNA-containing neurons in the vestibular nuclei of normal and hemilabyrinthectomized rats.

Vestibular compensation is an attractive model for investigations of cellular mechanisms underlying post-lesional plasticity in the adult central nervous system. Immediately after hemilabyrinthectomy, the spontaneous activity in the deafferented second-order vestibular neurons falls to zero, resulting in a strong asymmetry between the resting discharge of the vestibular complexes on the lesioned and intact sides. This asymmetry most probably causes the static and dynamic vestibular deficits observed in the acute stage. After approximately 50 h, the deafferented vestibular neurons recover a quasi-normal resting activity which is thought to be the key of the compensation of the static vestibular syndromes. However, the molecular mechanisms underlying this recovery are unknown. In this study, we investigate possible changes in the distribution of glutamatergic N-methyl-D-aspartate (NMDA) and glutamate metabotropic receptors and of glutamate decarboxylase 67k (GAD 67k) mRNAs in the deafferented vestibular neurons induced by the labyrinthine lesion. Specific radioactive oligonucleotides were used to probe sections of rat vestibular nuclei according to in situ hybridization methods. Animals were killed at different times (5 h, 3 days and 3 weeks) following the lesion. Signal was detected by means of film or emulsion autoradiography. In the normal animals, several brainstem regions including the medial, lateral, inferior and superior vestibular nuclei were densely labelled by the antisense oligonucleotide NMDAR1 probe. However, the vestibular nuclei were not labelled by the glutamate metabotropic oligonucleotide antisense probe (mGluR 1). The GAD 67k antisense oligonucleotide probe labelled numerous small- to medium-sized central vestibular neurons but not the larger cell bodies in the lateral vestibular nucleus. This agrees with previous studies. In the hemilabyrinthectomized rats, no asymmetry could be detected, at either the autoradiographic or cellular levels, between the two medial vestibular nuclei whatever the probe used and whatever the delay following the lesion. However, for the NMDAR1 probe, the mean density of silver grains in both the deafferented and intact medial vestibular neurons was 20% lower 5 h after the lesion. Three days and 3 weeks later, the intensity of labelling over all cells was the same as in the control group. Further studies are necessary to confirm the relatively weak modification of the NMDAR1 mRNAs expression and to exclude a change of GAD 65 and of other NMDA subunit mRNAs during the vestibular compensation process.