Contribution of supragranular layers to sensory processing and plasticity in adult rat barrel cortex

In mature rat primary somatic sensory cortical area (SI) barrel field corte x, the thalamic-recipient granular layer IV neurons project especially dens ely to layers I, II, III, and IV. A prior study showed that cells in the su pragranular layers are the fastest to change their response properties to n ovel changes in sensory inputs. Here we examine the effect of removing supr agranular circuitry on the responsiveness and synaptic plasticity of cells in the remaining layers. To remove the layer Il + III (supragranular) neuro ns from the circuitry of barrel field cortex, N-methyl-D-aspartate (NMDA) w as applied to the exposed dura over the barrel cortex, which destroys those neurons by excitotoxicity without detectable damage to blood vessels or ax ons of passage. Fifteen days after NMDA treatment, the first responsive cel ls encountered were 400-430 mu m below the pial surface. In separate cases triphenyltetrazolium chloride (TTC), a vital dye taken up by living cells, was absent from the lesion area. Cytochrome oxidase (CO) activity was absen t in the first few tangential sections through the barrel field in all case s before arriving at the GO-dense barrel domains. These findings indicate t hat the lesions were quite consistent from animal to animal. Controls consi sted of applying vehicle without NMDA under similar conditions. Responses o f D2 barrel cells were assessed for spontaneous activity and level of respo nse to stimulation of the principal D2 whisker and four surround whiskers D 1, D3, C2, and E2. In two additional groups of animals treated in the same way, sensory plasticity was assessed by trimming all whiskers except D2 and either D1 or D3 (called Dpaired) for 7 days before recording cortical resp onses. Such whisker pairing normally potentiates D2 barrel cell responses t o stimulation of the two intact whiskers (D2 + Dpaired). After NMDA lesions , cortical cells still responded to all whiskers tested. Cells in lesioned cortex showed reduced response amplitude compared with sham-operated contro ls to all D-row whiskers. In-are surround whisker (C2 or E2) responses were normal. Spontaneous activity did not change significantly in any remaining layer at the time tested. Modal latencies to stimulation of principal D2 o r surround D1 or D3 whiskers showed no significant change after lesioning. These findings indicate that there is a reasonable preservation of the resp onse properties of layer IV, V, VI neurons after removal of layer II-III ne urons in this way. Whisker pairing plasticity in layer IV-VI D2 barrel colu mn neurons occurred in both lesioned and sham animals but was reduced signi ficantly in lesioned animals compared with controls. The response bias gene rated by whisker trimming (Dpaired/Dcut + Dpaired ratio) was less pronounce d in NMDA-lesioned than sham-lesioned animals. Proportionately fewer neuron s in layer IV (52 vs. 64%) and in the infragranular layers (55 vs. 68%) exh ibited a clear response bias to paired whiskers. We conclude that receptive -field plasticity can occur in layers IV-VI of barrel cortex in the absence of the supragranular layer circuitry. However, layer I-III circuitry does play a role in normal receptive-field generation and is required for the fu ll expression of whisker pairing plasticity in granular and infragranular l ayer cells.