The inputs to several different populations of spinal cord neurons (identified by morphological criteria, tract tracing or intracellular staining) have been examined with immunogold labeling for Glu and other compounds. Inherent to the variation of Glu immunogold labeling density within a terminal population (see Section 1), the criteria for defining a terminal as putatively glutamatergic vary somewhat between laboratories. Thus, comparisons of the percentages obtained in different studies should be made with this caution in mind.
Murphy et al. (1996) examined terminals that contacted the soma or dendrites of ret-rogradely labeled motoneurons in the phrenic nucleus. Fifty-five percent of such terminals contained round vesicles and were enriched in Glu. Nerve terminals containing flattened vesicles comprised 42%, and presumably contain inhibitory amino acids. Cell bodies and dendrites displayed similar percentages of round and flattened vesicle terminals. Linda and collaborators examined the inputs to large cell bodies of alpha-motoneuron size in the L7 segment (Ornung, 1997). Of the terminals containing spherical vesicles (types S and M), all M terminals and most S terminals were enriched in Glu and displayed a positive correlation between the Glu immunogold labeling density and the density of synaptic vesicles. A proportion of S terminals contained low levels of Glu and displayed no such correlation. C-type (cholinergic) terminals contained levels of Glu intermediate to those detected in terminals with flattened and spherical vesicles, respectively, and displayed no correlation between the densities of synaptic vesicles and gold particles. Between 13.9 and 17.2% of the inputs to neuronal soma and proximal dendrites were classified as glutamatergic. Ornung et al. (1998) examined the inputs to motoneuron dendrites. Thirty-five percent of the boutons were immunopositive for Glu and 59% for GAB A and/or glycine. First-order (stem) dendrites differed from more distal dendrites by having a lower proportion of Glu-enriched terminals (18%) and a higher proportion of glycine and/or GABA-immunoreactive terminals (69%). Maxwell et al. (1997) examined inputs to dendrites of group-II spinal interneurons and identified two types of boutons. The first of these bouton types contained spherical vesicles and was enriched in Glu. The second type contained pleomorphic vesicles and was immunoreactive for GABA and/or glycine.
Neurons at the origin of several ascending somatosensory pathways have been examined for the presence of glutamatergic inputs. Westlund et al. (1992) investigated inputs to three intracellularly labeled spinothalamic tract neurons in the deep dorsal horn. Of the terminals located on the soma of these neurons, 46.5% were enriched in Glu. The proportion of Glu-immunoreactive terminals on the dendrites of these neurons was slightly higher (50.5%). In the superficial dorsal-horn, Glu-immunoreactive terminals constitute 37% and GABA-immunoreactive terminals 20% of the total terminal length on retrogradely labeled spinothalamic tract cells (Lekan and Carlton, 1995). The percentages were similar on soma and dendrites of these cells. Maxwell et al. (1992, 1995) have examined neurons at the origin of the spinocervical tract and the postsynaptic dorsal column pathway. The percentages of Glu-enriched terminals on these cells were estimated to 42% and 53%, respectively.
Llewellyn-Smith et al. (1992, 1998) examined the inputs to sympathetic preganglionic neurons. They detected enrichment of Glu in about two-thirds of the terminals contacting sympathoadrenal neurons and in about half of the terminals contacting neurons which project to the superior cervical ganglion. Most of the remaining terminals were immunoreactive for GABA.
Was this article helpful?