Neuronal plasticity is generally active in young age and some of them disappear after a specific time window critical period. Once this period is closed, loss of plasticity limits modifiability of function in adulthood. If it is possible to manipulate critical period, it would lead to reactivation of neuronal plasticity in later stage of development.
In our in vitro slice co-culture model of corticospinal (CS) projection, CS axons innervated the entire spinal cord at an early developmental stage and then withdrew mostly from the ventral side of the spinal cord, which is NMDA receptor (NMDAR)- and its subunit GluN2B-dependent. There is critical period (6-11 DIV) for this activity-dependent synapse elimination with axonal regression.
In this study, we employed live imaging of CS axons labeled with EYFP-tagged channelrhodopsin2. If axonal withdrawal is blocked by application of NMDAR antagonist, APV during the critical period and thereafter it is removed, the number of axons on the ventral side no longer changes in wild type mice (WT). However, that was reduced robustly in GluN2A knockout mice (2a-/-) where GluN2B could not be replaced by GluN2A. Indeed, at the end of the critical period, synaptic GluN2B in 2a-/- spinal cords remained high level comparable to the peak of the critical period in WT.
Furthermore, we applied GluN2B inhibitor, ifenprodil for 2 days following APV administration in 2a-/-. After 3 days from the removal of ifenprodil, the number of axons on the ventral side decreased even after closure of the critical period, suggesting that heightened level of synaptic GluN2B re-opened the critical period.
Finally, we used pharmacological agents, proBDNF or spermine which has been known to up-regulate synaptic GluN2B. By application of these agents after the end of the critical period in WT, the axons also retracted from the ventral side and synaptic GluN2B was found to increase comparable to the critical period.
In conclusion, developmental decline of synaptic GluN2B determined the completion of the critical period for CS axonal regression and manipulation to recover synaptic GluN2B could elongate or re-open the critical period.