Quantitative assessment of taurine-like immunoreactivity in different cell types and processes in rat cerebellum: an electronmicroscopic study based on a postembedding immunogold labelling procedure.

Ultrathin sections of plastic-embedded rat cerebella were incubated with an antiserum against conjugated taurine and subsequently treated with a secondary antibody coupled to colloidal gold. The density of gold particles in various cellular profiles was calculated with the assistance of a computer. In the cerebellar cortex the highest density was found in the somata, dendrites, and dendritic spines of the Purkinje cells, supporting parallel light-microscopical observations in postembedding stained semithin sections from the same tissue blocks. The remaining profiles could be divided into three groups according to their immunolabelling intensity, in descending order: 1) somata and processes of granule and Golgi cells; 2) somata and processes of stellate, basket, and glial cells, and 3) mossy fiber terminals. In a representative experiment, the structures in the first and second groups showed gold particle densities in the range of 19-25%, and 4-11%, respectively, of that in the Purkinje cell somata (values corrected for background) whereas the particle density in the mossy fiber terminals was not significantly above background level. In the cerebellar nuclei, taurine-like immunoreactivity was concentrated in terminals that typically established symmetric or intermediate type contacts with weakly labelled dendrites and cell bodies. These terminals, which shared the ultrastructural features of Purkinje cell terminals, showed an average gold particle density that was about 60% higher than that of the Purkinje somata. For specificity control, ultrathin sections containing a series of different amino acid conjugates were incubated in the same drops of sera as the tissue sections. The highly selective labelling of the taurine conjugate indicated that the distribution of gold particles in the tissue was not confounded by crossreactivity with GABA, glutamate or other common amino acids but adequately reflected the distribution of fixed taurine. For additional control of specificity, the taurine antiserum was applied to the soluble fraction of a rat brain extract separated by thin layer chromatography. In this system the taurine antiserum stained a single spot that comigrated with free taurine. The present results suggest that all cell types and processes in the rat cerebellum (with the exception of the mossy fiber terminals) contain taurine. However, the concentration of taurine appears to vary considerably among the different cell types and may also differ between different parts of the same neuron.