Non-plaque dystrophic dendrites in Alzheimer hippocampus: a new pathological structure revealed by glutamate receptor immunocytochemistry.

Alzheimer's disease is a progressive dementia characterized by a pronounced neurodegeneration in the entorhinal cortex, hippocampal CA1, and subiculum. Excitatory amino acid receptor-mediated excitotoxicity is postulated to play a role in the neurodegeneration in Alzheimer's disease. The present study investigated immunocytochemical localization of excitatory amino acid receptor subunits in the hippocampus of twelve Alzheimer's disease and eleven controls, matched for age, sex and post mortem interval. Immunocytochemistry with antibodies specific for glutamate receptors GluR1, GluR2(4) (alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid), GluR5/6/7 (kainate) and NR1 (N-methyl-D-aspartate) receptor subunits demonstrated that virtually all projection neurons in all subfields contained subunits from each receptor class. However, regional differences in immunoreactivity were apparent in Alzheimer's disease vs normal human brain. In the vulnerable regions (i.e. CA1) immunolabelling of GluR1, GluR2(4), GluR5/6/7 and NR1 was reduced, presumably due to cell loss. In contrast, GluR2(4) immunolabelling appeared to be increased in the inner portion of the molecular layer of the dentate gyrus. In addition to cellular labelling, GluR1, GluR2(4) and NR1 immunolabelling revealed a novel pathological structure in 12 of 12 Alzheimer's disease, but none of the control brains. The lesions were juxtacellular clusters of granular immunoreactivity in the neuropil of the pyramidal cell layer. Ultrastructural analysis revealed these to be cellular processes containing dense vesicles and flocculent material with immunolabelling localized to plasma and vesicular membranes. They were not specifically associated with amyloid fibrils and did not contain paired helical filaments and they were also distinct from granulovacuolar degeneration. Several structures contained Hirano body filaments indicating that the dystrophic processes were most likely dendritic in origin. Additional studies are needed to determine the pathogenesis of these lesions, which could provide an additional index of dendritic deterioration in Alzheimer's disease.