Plasticity in the rat spinal cord seen in response to lesions to the motor cortex during development but not to lesions in maturity.

Motor cortical inputs and proprioreceptive muscle afferents largely target the same spinal cord region. This study explored the idea that during development the two inputs interact via an activity-dependent mechanism to produce mature patterns of innervation. In rats, the forelimb motor cortex was ablated unilaterally at either postnatal day 7 (P7), the beginning of corticospinal synaptogenesis in the cervical cord, or at P50. Comparisons were made with sham-operated animals. At P70, muscle afferents from the extensor digitorum communis muscle, contralateral to the lesion, were transganglionically labeled with cholera toxin B-subunit. Lower cervical spinal cord sections were immunostained for cholera toxin B, parvalbumin, and cJun. Our small lesions had no obvious effects upon forelimb function. However, developmental lesions, but not adult lesions, were shown to significantly increase the number of muscle afferent boutons present in the contralateral ventral horn, compared with sham-operated controls. Also, the ratio of parvalbumin-positive neurons contralateral/ipsilateral to the developmental lesion (but not adult lesions) was decreased and the ratio of cJun-positive motoneurons increased. Thus, an early motor cortex lesion resulted in retention of a proportion of muscle afferent synapses to the ventral horn that are known to be lost during normal development. Parvalbumin and cJun are markers of neuronal activity suggesting that spinal circuitry develops permanently altered activity patterns in response to an early cortical lesion, although this plasticity is lost in the mature animal.

Changing pattern of expression of parvalbumin immunoreactivity during human fetal spinal cord development.

Expression of the calcium binding protein parvalbumin (PV) by different classes of spinal neuron has been shown to be developmentally regulated in both rat and monkey. From postmortem studies of eight human cervical spinal cords ranging in age from 11 to 35 weeks postconceptional age, we report that parvalbumin immunoreactivity is similarly plastic in human lower cervical spinal cord development, with many changes occurring prenatally. At 11-14 weeks postconceptional age, there was prominent immunostaining of primary sensory afferents that could be seen coursing through the dorsal horn and extensively innervating the motoneuron pools. Motoneurons were also found to be clearly immunoreactive for choline acetyltransferase by this age. A few ventral horn neurons that were not motoneurons were also parvalbumin immunoreactive. By 24-27 weeks postconceptional age, sensory afferents were still immunoreactive, as were many other axons throughout the white matter. In addition, many ventral horn neurons were now immunoreactive as well as a few dorsal horn neurons. By 31-35 weeks postconceptional age, there was extensive immunostaining of neurons throughout the spinal cord, including a few moderately immunoreactive motoneurons. There were many immunopositive axons in all the white matter tracts except the corticospinal tracts; however, staining of sensory axons traversing the grey matter was less prominent by this age. In the rat, expression of PV by primary sensory neurons coincides with the onset of fetal limb movement. The onset of expression of PV in ventral horn neurons coincides with later developmental events after the arrival of corticospinal inputs, whereas widespread PV immunoreactivity in dorsal horn neurons marks the attainment of a mature pattern of PV expression. The extent to which expression of PV immunoreactivity can be taken to indicate landmarks in human development will be discussed.