Plasticity after cortical stroke involves potentiating responses of pre-existing circuits but not functional remapping to new circuits

Functional recovery after stroke is thought to be mediated by adaptive circuit plasticity, whereby surviving neurons assume the roles of those that died. This "remapping" hypothesis is based on human brain mapping studies showing apparent reorganization of cortical sensorimotor maps and animal studies documenting molecular and structural changes that could support circuit rewiring. However, definitive evidence of remapping is lacking, and other studies have suggested that maladaptive plasticity mechanisms, such as enhanced inhibition in peri-infarct cortex, might actually limit plasticity after stroke. Here we sought to directly test whether neurons can change their response selectivity after a stroke that destroys a single barrel (C1) within mouse primary somatosensory cortex. Using multimodal in vivo imaging approaches, including two-photon calcium imaging to longitudinally record sensory-evoked activity in peri-infarct cortex before and after stroke, we found no evidence to support the remapping hypothesis. In an attempt to promote plasticity via rehabilitation, we also tested the effects of forced use therapy by plucking all whiskers except the C1 whisker. Again, we failed to detect an increase in the number of C1 whisker-responsive neurons in surrounding barrels even 2 months after stroke. Instead, we found that forced use therapy potentiated sensory-evoked responses in a pool of surviving neurons that were already C1 whisker responsive by significantly increasing the reliability of their responses. Together, our results argue against the long-held theory of functional remapping after stroke, but support a plausible circuit-based mechanism for how rehabilitation may improve recovery of function.