Quantification of excitatory amino acid uptake at intact glutamatergic synapses by immunocytochemistry of exogenous D-aspartate.

To study the localization and efficiency of glutamate/aspartate membrane transport in the vicinity of intact glutamatergic synapses, the avascular lamprey spinal cord was incubated with D-aspartate, a metabolically inert transporter substrate. The exogenous D-aspartate was localized by immunocytochemistry after aldehyde fixation. Incubation at 50 or 500 microM D-aspartate for 1 hr caused a prominent D-aspartate labeling of glial processes at glutamatergic synapses, while presynaptic axons and postsynaptic dendrites remained unlabeled. The glial processes surrounding glutamatergic sensory axons with a predominantly tonical firing pattern contained significantly higher levels of D-aspartate than did processes surrounding glutamatergic reticulospinal axons, which fire rarely and in brief bursts. Preparations incubated for 10 hr with 500 microM D-aspartate showed D-aspartate immunolabeling in glia as well as in the two types of glutamatergic axon, but no evidence was obtained for uptake into synaptic vesicles. Nor was such evidence obtained after high-frequency electrical stimulation. The observations suggest that excitatory amino acids delivered diffusely to the extracellular space in the intact CNS are transported almost exclusively into glia. The avid uptake in glial processes, combined with their spatial arrangement around glutamatergic synapses, appears to limit the access of exogenous D-aspartate to the nerve terminal glutamate/aspartate transporter. In physiological conditions, the glial processes are likely to impede the exchange of glutamate between the synaptic cleft and the rest of the extracellular space. The transport was more efficient in glial processes located near tonically active synapses than in ones located near synapses releasing transmitter sporadically. D-Aspartate is not a substrate of vesicular glutamate transport sites at these intact synapses.