Plasticity in reflex pathways controlling stepping in the cat

Previous studies have shown that stimulation of group 'I' afferents from ankle extensor muscles can prolong the cycle period in decerebrate walking cats and that the magnitude of these effects can be altered after chronic axotomy of the lateral-gastrocnemius/soleus (LGS) nerve. The effectiveness of LGS group I afferents in prolonging the cycle period decreases after axotomy, whereas the effectiveness of the uncut medial-gastrocnemius (MG) group I afferents is increased. The objectives of this investigation were to establish the time course of these changes in effectiveness and to determine whether these changes persist after transection of the spinal cord. The effects of stimulating the LGS and/or MG group I afferents on the cycle period were examined in 22 walking decerebrate animals in which one LGS nerve had been cut for 2 to 31 days. The effectiveness of LGS group I afferents declined progressively in the postaxotomy period, beginning with significant decreases at 3 days and ending close to zero effectiveness at 31 days. Large increases in the effectiveness of MG group I afferents occurred 5 days after axotomy, but there was no progressive change from 5 to 31 days. To test whether these changes in effectiveness were localized to sites within the spinal cord, the cord was transected in some decerebrate animals and stepping induced by the administration of L-DOPA L-3-4 dihydroxyphenylalanine (L-DOPA) and Nialamide. The effects of stimulating the MG and/or the LGS group I afferents on the cycle period were reexamined. In all four animals tested, stimulating the axotomized LGS group I afferents had a reduced effectiveness during locomotor activity in both the decerebrate and spinal states, whereas the increased effectiveness of the MG group I afferents was retained after transection of the spinal cord in two of five animals. Different mechanisms may be responsible for the changes in strength of the LGS and MG group I afferent pathways that project onto the rhythm generating sites in the spinal cord. This possibility follows from our observations of a linear relationship between the time after axotomy and decreased effectiveness of LGS group I afferents but no significant relationship between time postaxotomy and increased effectiveness of MG group I afferents; no significant relationship between the decreased effectiveness of LGS group I afferents and the increased effectiveness of MG group I afferents; and, after spinalization, consistent (4/4 cases) preservation of decreased LGS effectiveness but frequent (3/5 cases) loss of increased MG effectiveness.