Decreased calcium influx into the neonatal rat motor nerve terminals can recruit additional neuromuscular junctions during the synapse elimination period.

Individual skeletal muscle fibers in newborn vertebrates are innervated at a single endplate by several motor axons. During the first postnatal weeks, the polyneuronal innervation decreases in an activity-dependent process of synaptic elimination by axonal competition. Because synaptic activity depends strongly on the influx of calcium from the external media via presynaptic voltage-dependent calcium channels, we investigate the relationship between calcium channels, synaptic activity and developmental axonal elimination. We studied how several calcium channel blockers affect (after 1 h of incubation) the total number of functional axons per muscle fiber (poly-innervation index) of the Levator auris longus muscle of 6-day-old rats. We determined the poly-innervation index by gradually raising the stimulus amplitude and recorded the recruitment of one or more axons that produced a stepwise increment of the endplate potential.The L-type channel blocker nitrendipine (1 microM) increased the mean poly-innervation index (35.79% +/- 3.91; P<0.05). This effect was not washed out with normal Ringer, although the poly-innervation index returned to the control value when high-calcium Ringer (5 mM) was used. The P-type channel blocker omega-agatoxin-IVA (100 nM) also increased the number of recruitable endplate potentials (27.49% +/- 1.78; P<0.05), whereas N-type channel blocker omega-conotoxin-GVIA (1 microM) was ineffective (P>0.05). However, neither nitrendipine nor omega-agatoxin-IVA modified the poly-innervation index on high-calcium Ringer (P>0.05 in both cases). A more intense inhibition of calcium influx (by the sequential use of two calcium channel blockers) did not recruit any additional silent synapses. Moderately increasing the magnesium ions (by 500 microM) in the physiological solution produces a synaptic recruitment (36.78% +/- 2.1; P<0.05) similar to that with L- and P-type calcium channel blockers incubation. This magnesium effect was not washed with normal Ringer but a Ringer that is high in calcium can reverse it. The recruited endings were identified by selective activity-dependent loading with styryl dyes. Rhodaminated alpha-bungarotoxin-labeled acetylcholine receptors were present in the postsynaptic counterpart. Based on these findings we suggest that, before their complete retraction, functionally silent nerve terminals can be manifested or recovered if calcium influx is reduced by a calcium channel blocker or if external magnesium is increased. The normal activation of this calcium-dependent silencing mechanism during development may be related to the final loss of the supernumerary axons.