Dendritic reorganisation in the basal forebrain under degenerative conditions and its defects in Alzheimer's disease. I. Dendritic organisation of the normal human basal forebrain.

In the present study, the dendritic organisation of neurones in the normal human basal forebrain was analysed as a prerequisite for the evaluation of pathological changes occurring in Alzheimer's disease and related conditions (see other Arendt et al. papers in this issue). Neurones in the basal nucleus of Meynert (NbM), the nucleus of the vertical limb of the diagonal band, and the medial septal nucleus were examined after Golgi impregnation. Cells were classified according to the dendritic branching pattern and soma shape as either reticular neurones or multipolar giant neurones. The reticular type of neurones constitutes more than 90% of neurones in the BnM. Cholinergic neurones also belong to this cell type. Reticular neurones were further subdivided into four subtypes. Morphological features and arrangement of reticular basal forebrain neurones were identical to those described for "reticular formation cells" or "isodendritic" neurones. Dendritic trees of reticular neurones show a spatial orientation perpendicular to passing fibres as well as a high degree of overlap, both of which are hallmarks of "open nuclei." The qualitative classification of Golgi-impregnated basal forebrain neurones was substantiated by a computer-based three-dimensional analysis. Topologic and metric parameters of the dendritic tree were calculated for each type of neurone to characterise the degree of dendritic branching, the shape and orientation of the dendritic arborisation, the spatial extension of the dendritic tree, and soma size. The classification criteria were evaluated according to their power of discrimination between different cell types by means of a discriminant analysis. The quantitative approach applied in the present study not only provides an objective measure for the description and comparison of the structure of various types of neurones but also makes it possible to elucidate fine structural changes that might occur under pathologic conditions and that are not evident during qualitative studies alone.