Ca2+ influx and cAMP elevation overcame botulinum toxin A but not tetanus toxin inhibition of insulin exocytosis

Previous reports showed that cleavage of vesicle-associated membrane protei n-2 (VAMP-2) and synaptosomal-associated protein of 25 kDa (SNAP-25) by clo stridial neurotoxins in permeabilized insulin-secreting p-cells inhibited C a2+-evoked insulin secretion. In these reports, the soluble N-ethylmaleimid e-sensitive factor attachment protein target receptor proteins might have f ormed complexes, which preclude full accessibility of the putative sites fo r neurotoxin cleavage. In this work, VAMP-2 and SNAP-25 were effectively cl eaved before they formed toxin-insensitive complexes by transient transfect ion of insulinoma HIT or INS-1 cells with tetanus toxin (TeTx) or botulinum neurotoxin A (BoNT/A), as shown by immunoblotting and immunofluorescence m icroscopy. This resulted in an inhibition of Ca2+ (glucose or KCl)-evoked i nsulin release proportionate to the transfection efficiency (40-50%) and an accumulation of insulin granules. With the use of patch-clamp capacitance measurements, Ca2+-evoked exocytosis by membrane depolarization to -10 mV w as abolished by TeTx (6% of control) but only moderately inhibited by BoNT/ A (30% of control). Depolarization to 0 mV to maximize Ca2+ influx partiall y overcame BoNT/A (60% of control) but not TeTx inhibition. Of note, cAMP a ctivation potentiated Ca2+-evoked secretion by 129% in control cells but on ly 55% in BoNT/A-transfected cells and had negligible effects in TeTx-trans fected cells. These results indicate that, whereas VAMP-2 is absolutely nec essary for insulin exocytosis, the effects of SNAP-25 depletion on exocytos is, perhaps on insulin granule pool priming or mobilization steps, could be partially reversed by higher levels of Ca2+ or cAMP potentiation.