The corpus callosum is the largest commissure in the brain of placental mammals.
Callosal axons exert a major contribution to cortical function by mediating interhemispheric communication, mainly between homotopic contralateral regions. Here, we have studied the influence of superficial callosal projecting neurons (CPNs), the most abundant subtype of CPN, in the entire columnar extension of the retrosplenial cortex, a region involved in spatial cognition and context recognition.
Our data indicate that this form of cortical input specifically targets superficial pyramidal neurons and the large bursting extratelencephalic-projecting pyramidal
neurons in upper layer 5B. In contrast, the medium-size pyramidal neurons in layers 5A,
5B and 6 and the large pyramidal neurons from lower layer 5B remained largely unresponsive, both to the direct callosal excitatory input and to PV-FS dependent inhibition.
The effect of callosal input on their two main contralateral targets was opposed, suppressing the activity of layer 2/3 pyramidal neurons and potently recruiting the large pyramids of layer 5B. A larger convergence of callosal axons on the latter and the large responsiveness of superficial parvalbumin-expressing fast-spiking interneurons (PV-FS), which potently inhibit pyramidal neurons in layers 2/3, explain the differential recruitment of both pyramidal subtypes. In addition, we demonstrate a laminar dependence in the dynamics of PV-FS recruitment by callosal input. While PV-FS cells
in superficial layers are directly recruited by callosal excitation, acting on superficial
pyramidal neurons in a feed-forward manner and markedly reducing their opportunity to
reach the firing threshold, those PV-FS neurons from deeper layers only weakly respond
to callosal input, but require the previous activation of the surrounding large bursting
pyramidal neurons to fire an action potential.
Overall, our data suggests that superficial and bursting pyramidal neurons with subcortical projections from upper layer 5B form an integrated subnetwork across contralateral retrosplenial areas, and that this is, at least partially, isolated from microcircuits including other pyramidal subtypes. In addition, we also provide new insights in the organization of cortical inhibitory networks across layers with relevant implications in the coding strategies of superficial and deep pyramidal neurons. A discussion of these results in the context of our current understanding of the general principles of cortical function is also provided.