Quantitative Golgi study of the rat cerebellar molecular layer interneurons using principal component analysis

Abstract

In this study, we applied for the first time a multivariate analysis to describe the anatomy of cerebellar molecular layer interneurons. Forty variables extending over a variety of morphological features (geometrical, topological, and metrical) were obtained from a three-dimensional reconstruction of 26 rat rapid Golgi-stained neurons. The subsequent principal component analysis showed that the first principal component was strongly correlated with variables related to the depth of each cell's soma in the molecular layer. The second principal component was strongly correlated with parameters describing axonal morphology. Finally, an analysis of the distribution of these anatomical features suggested that these cells cannot be classified into distinct groups but, instead, represent one continuously varying population. Thus, the classical division of molecular layer neurons into deep basket cells and superficial stellate cells is not supported by our analysis. These results have important implications for the development of the cerebellar cortex as well as for the expected patterns of Purkinje cell activity following activation of the granule cell layer. J. Comp. Neurol. 393:353–373, 1998. © 1998 Wiley-Liss, Inc.