Involvement of the Voltage-Gated Calcium Channels L- P/Q- and N-Types in Synapse Elimination During Neuromuscular Junction Development

Abstract

During the nervous system development, synapses are initially overproduced. In the neuromuscular junction (NMJ) however, competition between several motor nerve terminals and the synapses they made ends with the maturation of only one axon. The competitive signaling between axons is mediated by the differential activity-dependent release of the neurotransmitter ACh, co-transmitters, and neurotrophic factors. A multiple metabotropic receptor-driven downstream balance between PKA and PKC isoforms modulates the phosphorylation of targets involved in transmitter release and nerve terminal stability. Previously, we observed in the weakest endings on the polyinnervated NMJ that M₁ mAChR receptors reduce ACh release through the PKC pathway coupled to an excess of Ca²⁺ inflow through P/Q- N- and L-type voltage-gated calcium channels (VGCC). This signaling would contribute to the elimination of this nerve terminal. Here, we investigate the involvement of the P/Q-, N-, and L-subtype channels in transgenic B6.Cg-Tg (Thy1-YFP)16-Jrs/J mice during synapse elimination. Then, the axon number and postsynaptic receptor cluster morphologic maturation were evaluated. The results show that both L- and P/Q-type VGCC (but not the N-type) are equally involved in synapse elimination. Their normal function favors supernumerary axonal loss by jointly enhancing intracellular calcium [Ca²⁺]_i. The block of these VGCCs or [Ca2+]i i sequestration results in the same delay of axonal loss as the cPKCβI and nPKCε isoform block or PKA activation. The specific block of the muscle cell's contraction with μ-conotoxin GIIIB also delays synapse maturation, and thus, a retrograde influence from the postsynaptic site regulating the presynaptic CaV1.3 may contribute to the synapse elimination.

Keywords: Axonal competition; Motor endplate; Postnatal synapse elimination; Protein kinases; VGCC.

Fig. 1

a Representative confocal image of a nerve terminal arborization. Singly, dually, and innervated by three or more axons NMJs from YFP muscles and also images of the morphologic maturation (S1, the most inmature, and S4, almost fully differentiated, stages) of the postsynaptic clusters from P9 mice. The bar indicates 10 μm. b Confocal immunofluorescence location of α_1D L-, N-, and P/Q-type voltage-dependent calcium channels (VDCCs) at the NMJ. Triple labeling of VDCCs (green fluorescence) with syntaxin (blue fluorescence) and nAChR-α-bungarotoxin (red fluorescence) in merge images. Figure shows the presence of α_1D L-, N-, and P/Q-type-VDCC (in green) in the nerve terminal of P9 Levator auris longus (LAL) muscle endplates. The bar indicates 10 μm

Fig. 2

Western blots and histograms of α1D-L-, N-, and P/Q VGCC proteins in the LAL muscle of mice during development (P5-P7-P30). The developmental change in the P/Q VGCC protein level is parallel with the changes observed in other presynaptic molecules (nPKCε isoform and Munc18-1). Data are mean value ± SD, *p < 0.05, **p < 0.01, ***p < 0.005 (n = 5; 3 repeats)

Fig. 3

In (a) we show the percentage of singly- and polyinnervated NMJ after 4 applications over the LAL surface (one application every day between P5–P8 (observation at P9) of one of the following VGCC inhibitor substances: nitrendipine (NT 1 μM, an L-type channel blocker), ω-conotoxin-GVIA (ω-CON 1 μM, N-type channel blocker), and ω-agatoxin-IVA (ω-AGA 100 nM, P/Q-type blocker). Also, the L activator Bay-K8644 (5 μM), the P/Q- and N-type activator GV-58 (20 μM), and the intracellular calcium chelator BAPTA-AM (5 μM). The histogram in (b) shows the percentage of S1-S4 clusters in the untreated control mice (PBS) and after the 4 applications of the aforesaid substances. Data were presented as percentages of NMJ ± SD. Fisher’s test: * p < 0.05, ** p < 0.01, *** p < 0.005. The confocal images in (c) show examples of representative NMJ areas with singly, dually, and innervated by three or more axons (the corresponding number of asterisks) from YFP muscles. At the left, the L-type channel blocker nitrendipine (NT) delays axon loss because many multi-innervated NMJs persist. By the contrary, at the right, the L activator Bay-K8644 increases the number of monoinnervated junctions. The bar indicates 10 μm

Fig. 4

Diagrams are graphic representations that collectively show the pattern of action of the different agents and combinations of agents used. a Multi-innervation values. b Postsynaptic cluster (S1) values. The line between the blue and white areas means the ratio 1 (experimental value/control value) or “no effect.” Orange filled circles mean that the experimental value is not significantly different from the control value (p < 0.05). and filled circles that there is a significant difference (p > 0.05). SEMs are eliminated for clarity

Fig. 5

Graphic representation of the results. The activity-dependent signaling between the nerve terminals that are in competition through several metabotropic receptors can result in the modulation of the downstream effector kinases, specifically cPKCβI, nPKCε, and PKA. Changes in kinases activity can allow the coordinate phosphorylation of the L-type CaV1.3 and P/Q-type VGCC. The high calcium entry through these operative channels present in immature nerve endings can result in their final loss. Also, muscle CaV1.1 and contractile activity can contribute to the synapse elimination. A component of this mechanism may be mediated by a retrograde influence from the postsynaptic site, via the BDNF-TrkB pathway, on the presynaptic calcium channels