Cell-type–specific sub- and suprathreshold receptive fields of layer 4 and layer 2/3 pyramids in rat primary visual cortex

Abstract

Connectivity of cortical pyramidal neurons is layer-specific in the primary visual cortex (V1) and this is thought to be reflected in different receptive field (RF) properties of layer 4 and layer 2/3 pyramidal neurons (L4Ps and L2/3Ps, respectively). However, it remains unclear how the two cell populations convert incoming visually driven synaptic inputs into action potential (AP) outputs. Here I compared postsynaptic potentials (PSPs) and AP responses of L4Ps and L2/3Ps in the binocular portion of rat V1 by intrinsic optical imaging (IOI)-targeted whole-cell recordings followed by anatomical identification and dendritic reconstructions. L2/3Ps had about 2-fold longer dendritic branches and a higher number of branch points and endings in their apical portions. Functionally, L2/3Ps had more hyperpolarized resting potentials and lower rates of spontaneous APs (medians: 0.07 vs. 0.60 AP/s). PSP responses to optimally oriented moving bars were comparable in terms of amplitude (16.0±0.9 vs. 17.3±1.1 mV for L2/3Ps and L4Ps, respectively), reliability and size of the RF. The modulated component of subthreshold responses of L4Ps to optimal sinusoidal drifting gratings was larger and their PSP onset latency in response to bars flashed in the cell's RF center were shorter (60 vs. 86 ms). In contrast to the similarities of PSP responses to moving bars, AP responses of L2/3Ps were more sparse (medians: 0.7 vs. 2.9 APs/stimulus passage), less reliable, but sharper in terms of angular size. Based on the differences of subthreshold inputs, I conclude that L4Ps may receive mostly thalamic inputs, whereas L2/3Ps may receive both thalamic and cortical inputs from layer 4. The comparable subthreshold responses to moving bars are converted by L2/3Ps into sparser but sharper AP outputs possibly by cell-type-specific AP-generating mechanisms or differences in visually driven inhibitory inputs.