In addition, some mutant terminals presented big membrane presynaptic compartments of unknown origin and organelles with an unusual shape (clover shape) compatible with arrested vesicle budding from endosomes (Figure 7F, panels e and f). Some of those structures kept similarities with those observed in dynamin mutants (Ferguson et al., 2007 and Raimondi et al., 2011) or at the sternocleidomastoid muscles in CSP-α KO mice (Fernández-Chacón et al., 2004). To complement our analysis, we analyzed a set of junctions after electrical nerve stimulation (180 s at 30 Hz) (Figures 7E and 7 F and S5). We found

that mutants and WT terminals had a similar number of vesicles, with a significant tendency to be of bigger size in the mutant (Figure S5A). The presynaptic area Bafilomycin A1 concentration and the vesicle density were similar to the values found in resting conditions for both genotypes. However, when we restricted our measurements to the effective area where vesicles reside (by removing the area occupied by mitochondria and axonal filaments) we

found a significant increase in that area in WT and mutant upon stimulation that translated into a lower vesicle density in the mutants (Figure S5A). In addition, omega shape structures were also more frequent in the mutant upon stimulation (Figure 7F). Altogether, those observations could reflect INCB018424 alterations in membrane trafficking downstream of compensatory endocytosis. We wondered if the defects detected in the recycling could be caused by instability and/or degradation of dynamin1 Urease or other endocytic proteins that would normally require CSP-α to keep them stable over the time. Several proteins involved in vesicle recycling (intersectins, dynamins, Hsc-70, RME-8, and actin) were examined by immunoblotting of protein extracts from LAL muscles. Surprisingly, and in contrast to the strong decrease in SNAP-25 levels (Figure 2), we could not detect any reduction in the levels of endocytic proteins in the mutant terminals compared

to controls (Figures 7G and S5B–S5D). Thus, the measurements of vesicle recycling with FM2-10 suggested that the internalized membrane, upon strong stimulation, fails to be properly processed in order to initiate immediately a new wave of exocytosis and thus compromised the integrity of the recycling pool. That could be due to impairment in fast biogenesis of functional vesicles. Consistent with that view, the terminals from CSP-α KO mice exhibited plasma membrane features and unusual organelles compatible with slowed-down or arrested vesicle recycling. However, no decreased levels of endocytic proteins could be detected. CSP-α is essential to prevent activity-dependent degeneration of nerve terminals.