Peripheral nerve injury reallocates primary afferent input through spinal parvalbumin microcircuits

Complex neural functions rely on finely tuned circuits in which the recruitment of local interneurons gates the flow of information, determining whether an input is relayed, amplified, or suppressed. Somatosensory information, such as touch or pain, is processed through such complex circuits in the dorsal horn of the spinal cord. There, local inhibitory interneurons play a key role in the proper segregation of touch and pain inputs. After nerve injury-evoked neuropathic pain, loss of inhibition impairs the function of these circuits, resulting in mechanical allodynia, where innocuous touch is perceived as painful. Disinhibition can be attributed to pruning of inhibitory synapses, reduced intrinsic excitability of inhibitory neurons, or weakened excitatory drive from primary afferents. Yet the complexity of the excitatory drive onto inhibitory neurons, and its potential modification after nerve injury, remains largely unexplored. Here we examined the nature of the synaptic drive from low-threshold A{beta} mechanoreceptors (A{beta}-LTMRs) onto parvalbumin-expressing interneurons (PVNs), and how this recruitment is affected by nerve injury. A{beta}-LTMR stimulation evoked excitatory responses in a subset of PVNs, which is expected given the exclusively excitatory nature of primary afferent inputs. However, an unexpected subset of PVNs displayed inhibitory responses, suggesting the recruitment of a feedforward inhibitory circuit. We reconciled these observations by showing that A{beta}-LTMRs engaged PVNs through both direct excitation and feedforward inhibition, which are differentially distributed between the inhibitory (iPVN) and excitatory (ePVN) subpopulation. Indeed, our results show that under naive conditions, A{beta} input preferentially recruited iPVNs, while ePVNs were predominantly suppressed by A{beta}-driven feedforward inhibition mediated by a previously unrecognized Complexin-1 (Cplx1)-expressing inhibitory interneuron. After peripheral nerve injury, this balance becomes functionally redistributed. A{beta}-to-iPVN transmission showed increased failure rates and impaired temporal precision, whereas A{beta} drive onto ePVNs shifted from inhibition toward excitation. Notably, these functional changes occurred despite preserved afferent connectivity, synapse density, and spontaneous synaptic events, indicating that the dynamic reallocation of circuit recruitment occurs in the absence of structural changes. To test the behavioral consequences of these two populations, we used chemogenetic approaches and found that iPVNs suppress, whereas ePVNs promote, mechanical hypersensitivity. Together, these findings show that nerve injury functionally reallocates primary afferent drive away from inhibitory and toward excitatory spinal PVNs, establishing functional reallocation of afferent input as a mechanism of spinal disinhibition and a key determinant of mechanical allodynia.