Differential loss of neuromuscular connections according to activity level and spinal position of neonatal rabbit soleus motor neurons

Abstract

We have tested whether the ability of synapses to compete for occupancy of endplates during neuromuscular synapse elimination is affected by differences in the spinal position or in the activity level of the parent motor neuron. To test the role of spinal position, the relative sizes of motor units for motor neurons from middle and extreme (rostral/caudal) positions in the rabbit soleus motor pool were determined at 3 postnatal ages: 4-5 d ("early" ages, when the soleus is heavily polyinnervated), 8-9 d ("intermediate"), and 11-15 d ("late," when the soleus has just reached singly innervated state). Average motor unit sizes from extreme ventral roots were similar to those from middle ventral roots in early-aged soleus muscles but were significantly smaller (by 18-27%) for both intermediate and late muscles. Thus, motor neurons from extreme positions evidently compete less effectively for retention of synapses than those from middle positions. To test the role of differential activity, inactive and active synapses were pitted directly against one another by implanting Silastic plugs laden with tetrodotoxin (TTX) into one of the spinal nerves containing a minority of the soleus motor axons. Differential activity was maintained during a period of extensive synapse loss, from the time of the implant at day 4 or 5 until the intermediate age (day 8-9). Motor unit twitch tensions were subsequently measured to determine the relative number of synapses retained by individual active and inactive motor neurons. The inactivated motor units were on average significantly larger (by more than 50%) than the corresponding group from normal and control-implanted animals. The abnormally large size of inactivated motor units persisted in animals allowed to recover from the TTX block and examined after multiple innervation had disappeared. Hence, the effect of the TTX block cannot be attributed to a simple slowing of synapse elimination specifically among the inactive motor neurons. We conclude that complete presynaptic inactivity improves the chances of survival relative to that for normal activity during synapse elimination in the neonatal rabbit soleus muscle. This difference in competitive ability may contribute to the development of an important characteristic of adult muscles, the correlation between motor unit size and recruitment threshold.