Multiple axon collaterals of single corticospinal axons in the cat spinal cord.

To investigate intraspinal branching patterns of single corticospinal neurons (CSNs), we recorded extracellular spike activities from cell bodies of 408 CSNs in the motor cortex in anesthetized cats and mapped the distribution of effective stimulating sites for antidromic activation of their terminal branches in the spinal gray matter. To search for all spinal axon branches belonging to single CSNs in the "forelimb area" of the motor cortex, we microstimulated the gray matter from the dorsal to the ventral border at 100-micron intervals at an intensity of 150-250 microA and systematically mapped effective stimulating penetrations at 1-mm intervals rostrocaudally from C3 to the most caudal level of their axons. From the depth-threshold curves, the comparison of the antidromic latencies of spikes evoked from the gray matter and the lateral funiculus, and the calculated conduction times of the collaterals, we could ascertain that axon collaterals were stimulated in the gray matter rather than stem axons in the corticospinal tract due to current spread. Virtually all CSNs examined in the forelimb area of the motor cortex had three to seven branches at widely separated segments of the cervical and the higher thoracic cord. In addition to terminating at the brachial segments, they had one to three collaterals to the upper cervical cord (C3-C4), where the propriospinal neurons projecting to forelimb motoneurons are located. About three quarters of these CSNs had two to four collaterals in C6-T1. This finding held true for both fast and slow CSNs. About one third of the CSNs in the forelimb area of the motor cortex projected to the thoracic cord below T3. These CSNs also sent axon collaterals to the cervical spinal cord. CSNs in the "hindlimb area" of the motor cortex had three to five axon branches in the lumbosacral cord. These branches were mainly observed at L4 and the lower lumbosacral cord. None of these CSNs had axon collaterals in the cervical cord. CSNs terminating at different segments of the cervical and the thoracic cord were distributed in a wide area of the motor cortex and were intermingled. To determine the detailed trajectory of single axon branches, microstimulation was made at a matrix of points of 100 or 200 micron at the maximum intensity of 30 microA, and their axonal trajectory was reconstructed on the basis of the location of low-threshold foci and the latency of antidromic spikes.(ABSTRACT TRUNCATED AT 400 WORDS)