An experimental electron microscopy study of afferent connections to the primate motor and somatic sensory cortices

Abstract

An experimental electron microscope (e.m.) study has been made of the termination of the afferent connections to the primate sensori-motor cortex. Following large, stereotaxically placed thalamic lesions, degeneration in the motor and somatic sensory cortices was studied at survival periods of 4 and 5 days. Degenerating thalamocortical terminals had asymmetric membrane specializations. In the motor cortex 89.5% made synapses on to dendritic spines, 9% on to dendritic shafts and 1.5% on to cell somata; in the somatic sensory area 89% made synapses on to spines, 11 % on to dendritic shafts and one example contacted a cell soma and a spine. A considerable number of the spines receiving synapses from degenerating thalamo-cortical terminals were traced to their parent dendrites and these were of the pyramidal type whereas the dendritic shafts and cell somata contacted by degenerating thalamo-cortical terminals were mostly of the large stellate type. Most of the thalamo-cortical degeneration in both cortical areas occurred in a dense band in the upper two thirds of layer IV and the lower half of layer III but a number of degenerating terminals were found deep to this; in the motor cortex a second, less dense, band of degeneration was present in the lower part of layer V and top of layer VI. Degenerating thalamo-cortical terminals making synapses on to dendritic shafts and cell somata were scattered through the deep half of the cortex and not concentrated in the dense band of degeneration and so formed a greater proportion of the degeneration in the deep layers, particularly in the motor cortex. Sections cut parallel to the pial surface in layer IV of the motor cortex showed a statistically significant association between the degenerating thalamocortical axon terminals and the bundles of apical dendrites present at this level. Degeneration of commissural fibres was studied after removal of the contralateral sensori-motor cortex. Degenerating terminals had asymmetric membrane specializations. In the motor cortex 96% made synapses on to dendritic spines, 3% contacted dendritic shafts and one example made an axosomatic synapse; in area 3 97% made synapses on to dendritic spines and 3% contacted dendritic shafts. A number of the spines receiving synapses from degenerating commissural axon terminals were traced to their parent dendrites and these were of the pyramidal type. The cell soma and the majority of the dendritic shafts receiving synapses from commissural terminals were of the large stellate type although some of the dendritic shafts were probably those of small stellate cells. In the motor cortex degenerating commissural axon terminals were found in all cortical layers but were relatively more dense in layer I, the upper part of layer III, the upper part of layer V and the lowest part of layer V with layer V I; in the somatic sensory cortex most degenerating commissural terminals were found in the superficial half of the cortex. Following lesions of the primary somatic sensory cortex (SI) or of area 6 of the premotor cortex, degenerating terminals making asymmetric synapses were found in the motor cortex. Of the terminals of association fibres from SI, 82% made synapses on to dendritic spines and 18% on to dendritic shafts; of those fibres from area 6, 76% made synapses on to dendritic spines and 24% on to dendritic shafts. For both these association fibre connections, a proportion of the dendritic shafts contacted were clearly identifiable as those of large stellate cells. Terminals of both association connections occurred in all cortical layers with no obvious concentrations at any particular depth.

Glutamate in thalamic fibers terminating in layer IV of primary sensory cortex

“…difference is suggested by published photomicrographs (Jones and Powell, 1970;Garey and Powell, 197 1;LeVay and Gilbert, Discussion 1976;Peters and Feldman, 1976;Sloper and Powell, 1978; Morphological observations Hersch and White, 198 1;McGuire et al, 1984; White and KelThalamic fibers terminating in layer IV of different regions of ler, 1987;LeVay, 1988;Enhanane and White, 1990; Czeiger cortex were generally similar, and differed markedly from terand White, 1993), the present study provides morphometric . Scatterplot showing Glu immunostaining over vesicle-containing regions in thalamocortical terminals (circles) and GABAergic terminals (triangles); regression lines are least squares fits.…”

Morphological Characterization of a Cortico-cortical relay in the cat sensorimotor cortex

“…However, the paucity of axospinous synapses appears to be an actual feature of this connection because all but three contacts seen at the EM level were axodendritic. This is in spite of the fact that, similar to numerous projections that link different cortical areas of many species through axospinous projections (Jones and Powell, 1970;Lund and Lund, 1970;Sloper and Powell, 1979;Cipolloni and Peters, 1983;Innocenti, 1984;Fisken et al, 1985), the majority of the synapses formed between area 3a axons and unidentified area 2 neurons are axospinous (57%; Porter, 1991). Therefore, the particular population of neurons studied here appears to form fewer axospinous synapses with area 3a afferents than the overall population of area 2 neurons.…”

“…Conversely, they are the targets of the majority of intrinsic axon collaterals (Kisvardy et al, 1986;Gabbott et al, 1987;Elhanany and White, 1990;McGuire et al, 1991;White and Czeiger, 1991;Keller and Asanuma, 1993) and of inter-areal cortical afferents (e.g. Jones and Powell, 1970;Lund and Lund, 1970;Fisken et al, 1975;Sloper and Powell, 1979;Cipolloni and Peters, 1983;Ichikawa et al, 1985;Porter and Sakamoto, 1988) arising from pyramidal neurons in many cortical areas of different species. A handful of studies have focused on the specific patterns of synaptic interactions between the local axon collaterals of pyramidal neurons and their target cells in various regions of the cortex.…”

Excitation by geniculocortical synapses is not ‘vetoed’ at the level of dendritic spines in cat visual cortex.

“…One necessary structural prerequisite for the veto mechanism is that the spine should receive input from two different types of synapses. Electron microscopic studies have shown that 5-20 % of all spines are thought to carry two different types of synapses, one GABA (y-aminobutyric acid)-ergic and thought to be inhibitory, the other excitatory (Jones & Powell, 1969;Peters & Kaiserman-Abramof, 1970;Sloper & Powell, 1979;Koch & Poggio, 1983;Beaulieu & Colonnier, 1985;Somogyi & Soltesz, 1986). However, previous studies either did not identify the source of the synapse, or did not quantify the number of dual inputs in the case of identified geniculocortical synapses (Freund et al 1985b), which form between 5 and 30 % of synapses in layer 4 (Garey & Powell, 1971;LeVay & Gilbert, 1976).…”