Subcortically generated movements activate motor cortex during sleep and wake in rats through postnatal day 24

The development of motor control in primary motor cortex (M1) requires both movement and neural activity. In rats, cortical motor control first appears around postnatal day (P) 25, prior to which movements are generated by subcortical motor nuclei such as the red nucleus (RN). While these subcortical movements are thought to provide the activity that guides the development of M1, the types of movements that drive M1 activity and their somatotopic and temporal precision remain unknown. Here, we recorded activity in the forelimb region of M1 of P12-24 rats as they cycled between sleep and wake, and compared M1 and RN activity in P24 rats. At every age, M1 neurons showed somatopically precise activity to REM sleep twitches, along with strong responses to wake movements. From P12 to P24, the proportion of neurons that showed twitch-related activity decreased, twitch-related activity became more temporally refined, and a larger fraction of spikes occurred before movement onset. At P24, M1 showed less premovement activity than RN. Further, in contrast to the non-selective activity seen in M1, some RN neurons showed selective movement-related activity during wake, firing only during particular wake movements. These findings reveal that movement-related activity in M1 is somatotopically precise by P12 and temporally precise by P24. But M1 still lacks the strong premovement activity and selectivity characteristic of RN, suggesting that at P24, subcortical outputs remain the main drivers of M1s movement-related activity. Significance StatementIn adults, primary motor cortex (M1) is widely recognized for its role in initiating voluntary movements and guiding motor learning. However, during early development, our findings show that M1 processes activity related to self-generated limb movements but does not drive the production of those movements. Using simultaneous recordings from M1 and the red nucleus (RN) of developing rats, we show that although M1 shows movement-related activity during wake and sleep, RN remains the principal driver of behavior through at least postnatal day 24. These results refine our understanding of early movement-related activity and highlight the possibility that both sleep-related twitches and wake movements shape sensorimotor development by driving neural activity in M1 and RN.