GABA-labeled terminals form proportionally more synapses with dopaminergic neurons containing low densities of tyrosine hydroxylase-immunoreactivity in rat ventral tegmental area

Abstract

The levels of the catecholamine synthesizing enzyme, tyrosine hydroxylase (TH) are known to be closely regulated by neural feedback. Moreover, we have shown that intensity of TH-immunoreactivity varies with afferent input to the A10 group of dopaminergic neurons in the rat ventral tegmental area (VTA). This region is extensively and heterogenously innervated by GABAergic afferents that mediate a number of different behavioral responses to iontophoretically applied GABA mimetics. We sought to determine: (1) whether there was an ultrastructural substrate for GABAergic innervation of TH-immunoreactive neurons; and (2) whether detectable TH-immunoreactivity varied in proportion to their GABAergic input in the two major subdivisions of the VTA, the parabrachial pigmentosus and paranigral subnuclei. Rabbit antiserum to TH and rat antiserum to GABA were visualized in single coronal sections of acrolein-fixed rat brain using a combination of peroxidase-antiperoxidase (PAP) and immunoautoradiography (ARG) or PAP and silver-intensified immunogold (SIG). Two dual-labeling electron microscopic immunocytochemical methods were employed to optimize detection of antigens and to more accurately quantify densities of TH-immunoreactivity and types of synaptic associations. Ninety-six GABA-labeled terminals (43 in the parabrachial and 53 in the paranigral subdivisions) were examined with PAP and ARG; 462 (238 in parabrachial and 224 in paranigral subdivisions) were examined with PAP and SIG. Analyses of both subnuclei yielded similar results; thus, the data were combined. With both methods, most GABA-labeled terminals (63% for SIG, 66% for PAP) formed direct synapses with TH-labeled profiles. These synaptic specializations were symmetric, the type thought to mediate inhibition. In single sections where GABA-labeled terminals were presynaptic to TH-labeled profiles, they comprised 45% (PAP) to 54% (SIG) of the total number of synaptic inputs onto TH-labeled cell bodies and 65% (SIG) to 80% (PAP) of the synaptic input onto TH-labeled dendrites. This value would be significantly less, if the analysis included all sections containing only GABA or TH irrespective of their synaptic relationships. The density of TH-immunolabeling, whether low (light) or high (intense), was determined in PAP- and SIG-labeled tissue. By both labeling methods, the numbers of GABA-immunopositive terminals forming synapses with lightly and intensely TH-immunoreactive profiles appeared equal. However, lightly TH-labeled neurons received fewer synaptic contacts from unlabeled terminals and, consequently, received proportionally more GABA-labeled terminals. GABA-labeled and unlabeled terminals were often in direct apposition to each other and were surrounded laterally, but not separated from each other, by astrocytic processes. We conclude that in the rat VTA: (1) TH-labeled (dopaminergic) dendrites are the major targets of GABA-labeled terminals; and (2) the GABA-labeled terminals form proportionally more synapses with dopaminergic neurons having comparatively low densities of TH-immunoreactivity. The astrocytic processes may contribute to the uptake and metabolism of GABA while providing a diffusion barrier that facilitates presynaptic interactions between certain groups of axon terminals.