Evidence for a microglial reaction within the vestibular and cochlear nuclei following inner ear lesion in the rat

Following unilateral inner ear lesion, astrocytes undergo hypertrophy in th e deafferented vestibular and cochlear nuclei as shown by an increase in th e level of glial fibrillary acid. The present study extends our understandi ng of vestibular and cochlear system plasticity by examining microglial cha nges in these deafferented nuclei. The microglial reaction was studied 1, 2 , 4, 8, 14, 21, 28 and 42 days following the lesion with a monoclonal OX-42 antibody and lectins (Griffonia simplicifolia, B-4 isolectin) labelled wit h horseradish peroxidase or fluorescein. The deafferented nuclei were also examined for apoptotic cells by terminal transferase-mediated nick end labe lling of nuclear DNA fragments. In control and sham-operated rats, the dist ribution of the resting microglial cells was uniform in both the vestibular and cochlear nuclei. In the deafferented vestibular complex, the microglia l cells increased in number, became hypertrophied and were distributed in t he medial, lateral, superior and inferior vestibular nuclei. Reactive micro glial cells were also detected in the ipsilateral cochlear nuclei. Some of the immunostained cells were hypertrophic whereas others presented an amebo id morphology with few short and stout processes. The microglial reaction w as confined to the antero- and posteroventral cochlear nuclei. Finally, rea ctive microglia was also observed in the prepositus hypoglossi ipsilateral to the lesion. The microglial reactions within the prepositus hypoglossi, t he vestibular and the cochlear nuclei were detectable as early as one day a fter the lesion and persisted several weeks in both the vestibular and coch lear nuclei. Apoptotic cells were not detected in the vestibular nuclei at any stage following the lesion. In contrast, terminal deoxynucleotidyl tran sferase-mediated digoxygenin-11-dUTP nick end labelling-positive cells were first detected in the deafferented cochlear nuclei on the 3rd day followin g the lesion. They reached an apparent maximum by day 8 and then declined u ntil day 24. Double labelling experiments demonstrate that these cochlear t erminal deoxynucleotidyl transferase-mediated digoxygenin-11-dUTP nick end labelling-positive cells were also lectin-positive suggesting that reactive cochlear lectin-positive microglia cells were eliminated by a programmed c ell death. Our results establish the two experimental models as reliable tools to unde rstand the role of microglia in adult brain plasticity. The cochlear microg lial reaction was probably induced by the degeneration of the acoustic nerv e which follows the acoustic ganglion destruction. Interestingly, the same reasoning cannot apply to the vestibular microglial reaction following unil ateral labyrinthectomy: the vestibular ganglion was spared and the primary vestibular neurons did not degenerate, at least during the first week follo wing the lesion. (C) 1999 IBRO. Published by Elsevier Science Ltd.