Regulation of Competitive Interactions During Neuromuscular Synapse Elimination

Citation

Callaway, Edward Matthew (1988) Regulation of Competitive Interactions During Neuromuscular Synapse Elimination. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/s60x-n119. https://resolver.caltech.edu/CaltechTHESIS:01172013-084333489

Abstract

The subsequent chapters of this thesis address a number of issues related to the phenomena that occur during neuromuscular synapse elimination and to the rules and mechanisms that govern them. The results they describe are therefore based on observations of developmental processes in both normal animals and in those whose normal developmental interactions have been perturbed by alteration of activity for some of the elements involved.

Chapter 2 addresses the question of whether the rate of neuromuscular synapse elimination might normally depend on the level of postsynaptic activity. Previous studies had strongly implicated activity in regulation of synapse elimination rate; e.g., increased activity evoked by chronic stimulation increased the elimination rate; but these studies did not differentiate between pre- and postsynaptic activity. Duxson (1982) treated the rat soleus muscle with α-bungarotoxin (α-BGT), completely blocking postsynaptic activity during the normal period of synapse elimination and reported that the number of terminal profiles per endplate observed in electron micrographs did not decrease as it does normally. I did not consider this study to be conclusive because complete activity blockade might invoke influences that are not normally present in active muscles during synapse elimination, and because the assay was indirect - an increase in the number of terminal profiles per endplate might reflect an increase in terminal complexity rather than maintenance of more terminals.

For the experiments described in Chapter 2, postsynaptic activity was partially blocked by α-BGT superfusion of the neonatal rabbit soleus muscle. The toxin treatment resulted in slower synapse elimination, as assessed both physiologically and anatomically, even for muscle fibers whose activity was not completely blocked. While the interpretation of this result is dependent on the possibility that α-BGT has influences other than decreased activity, it appears quite likely that a partial block of postsynaptic activity can slow the rate of neuromuscular synapse elimination.

Chapter 3 describes a separate series of experiments in which motor unit twitch tensions were assayed for the soleus muscles of neonatal rabbits. Synapse loss could be assayed separately for fast and slow populations by separating the motor units, based on their twitch rise times. Estimates of the rate of synapse elimination for the two populations suggested that slow muscle fibers were initially more heavily polyinnervated than fast fibers and that they lost synapses at a faster rate, so that both populations of fibers became predominantly singly innervated at about the same time. The remainder of the issues in Chapter 3 are related to the question of whether there are particular attributes of motor neurons that might place the inputs from some neurons at a competitive advantage over others.

The first such issue was whether motor neurons with relatively large axonal arbors are at an advantage or disadvantage in the competition for synaptic sites. If this were the case, it would be expected that the diversity in motor unit sizes would decrease during synapse elimination if a large arbor were a disadvantage, and thediversity would increase if it were an advantage. Contrary to both hypotheses, no significant change in the diversity of motor unit sizes was observed.

The next issue was whether motor neurons from particular positions in the spinal cord were at an advantage or disadvantage compared to the others. To test this issue, mean sizes of motor units from both rostral and caudal extremes of the soleus motor pool were compared to the mean sizes for those from the middle of the pool. At the earliest age tested, when the soleus is still heavily polyinnervated, the motor units from the extremes were no smaller than those from the middle. However, just four days later, the units from each extreme were significantly smaller than those from the middle; this difference persisted in older, singly innervated muscles. There was no significant difference between the rostral and caudal motor units at any age tested. It is concluded that motor neurons from rostral and caudal extremes are at a disadvantage when in competition with those from the middle of the motor pool during synapse elimination in the rabbit soleus.

Finally, a small portion of the rabbit soleus motor neurons was inactivated by implantation of a tetrodotoxin-laden Silastic plug during synapse elimination. Since there was nearly complete overlap between the inactivated and active motor units at the time of the implant, this allowed a test of whether the level of activity of a motor neuron can influence the ability of its terminals to compete for sole occupancy of endplates. It was found that the inactive motor units ended up significantly larger than their active counterparts in normal and control implanted animals, and remained larger even after the endplates were virtually all singly innervated. It is concluded that inactivity can result in a significant competitive advantage during synapse elimination. The generality of these conclusions and their implications in terms of the ways in which neuromuscular synapse elimination might be regulated are discussed in detail.

Thesis Files