Source of inappropriate receptive fields in cortical somatotopic maps fromrats that sustained neonatal forelimb removal

Previously this laboratory demonstrated that forelimb removal at birth in r ats results in the invasion of the cuneate nucleus by sciatic nerve axons a nd the development of cuneothalamic cells with receptive fields that includ e both the forelimb-stump and the hindlimb. However, unit-cluster recording s from primary somatosensory cortex (SI) of these animals revealed few site s in the forelimb-stump representation where responses to hindlimb stimulat ion also could be recorded. Recently we reported that hindlimb inputs to th e SI forelimb-stump representation are suppressed functionally in neonatall y amputated rats and that GABAergic inhibition is involved in this process. The present study was undertaken to assess the role that intracortical pro jections from the SI hindlimb representation may play in the functional reo rganization of the SI forelimb-stump, field in these animals. The SI foreli mb-stump representation was mapped during gamma-aminobutyric acid (GABA)-re ceptor blockade, both before and after electrolytic destruction of the SI h indlimb representation. Analysis of eight amputated rats showed that 75.8% of 264 stump recording sites possessed hindlimb receptive fields before des truction of the SI hindlimb. After the lesions, significantly fewer sites ( 13.2% of 197) were responsive to hindlimb stimulation (P < 0.0001). Electro lytic destruction of the SI lower-jaw representation in four additional con trol rats with neonatal forelimb amputation did not significantly reduce th e percentage of hindlimb-responsive sites in the SI stump field during GABA -receptor blockade (P = 0.98). Similar results were obtained from three man ipulated rats in which the SI hindlimb representation was silenced temporar ily with a local cobalt chloride injection. Analysis of response latencies to sciatic nerve stimulation in the hindlimb and forelimb-stump representat ions suggested that the intracortical pathway(s) mediating the hindlimb res ponses in the forelimb-stump field may be polysynaptic. The mean latency to sciatic nerve stimulation at responsive sites in the GABA-receptor blocked SI stump representation of neonatally amputated rats was significantly lon ger than that for recording sites in the hindlimb representation [26.3 +/- 8.1 (SD) ms vs. 10.8 +/- 2.4 ms, respectively, P < 0.0001]. These results s uggest that hindlimb input to the SI forelimb-stump representation detected in GABA-blocked cortices of neonatally forelimb amputated rats originates primarily from the SI hindlimb representation.