1. Our previous studies demonstrated that protein kinase C (PKC) activity is required for acetylcholine (ACh) sensitivity to recover fully at snake twitch fiber end plates after prolonged exposure to carbachol. In the present studies, we have investigated whether protein phosphatase(s), activated during carbachol exposure, dephosphorylated critical membrane proteins, which required rephosphorylation by PKC to maintain end-plate sensitivity. End-plate sensitivity was assessed from measurements of miniature end-plate currents (MEPCs) and carbachol-activated currents (EPCCARBS). Conductance of ACh-activated channels was determined from patch-clamp recordings of single-channel currents. 2. Pretreatment of snake muscle preparations with the protein kinase inhibitor staurosporine (0.5 microM), followed by a 10-min exposure to 540 microM carbachol, reduced mean MEPC amplitudes to values 30-40% less than those recorded before carbachol exposure. Conversely, at control end plates exposed to carbachol, the mean MEPC amplitude was reduced by only approximately 5% compared with precarbachol values. This staurosporine-induced decrease in ACh sensitivity could be prevented by pretreatment with the protein phosphatase 2B (calcineurin) inhibitor deltamethrin (0.5 microM), whereas okadaic acid (5 microM) and calyculin A (0.5 microM), inhibitors of protein phosphatases 1 and 2A, had no effect. 3. After a 10-min exposure to 540 microM carbachol, EPCCARB amplitudes (produced by local superfusion with 20 microM carbachol) were significantly smaller at staurosporine-treated end plates than at control end plates. In contrast, the EPCCARB amplitude recorded from end plates pretreated with both deltamethrin and staurosporine was not significantly different from that recorded at control end plates. 4. Substitution of 10 mM Mn2+ for external Ca2+ during the exposure to 540 microM carbachol prevented the decrease in MEPC amplitude recovery at staurosporine-treated end plates. These results suggested that the alteration in sensitivity at staurosporine-treated end plates was calcium dependent. 5. At control end plates, a single population of ACh-activated channels (45-50 pS) is observed both before and after a 10-min exposure to 540 microM carbachol. Conversely, at staurosporine-treated end plates, after exposure to carbachol, a second population of small-conductance (25-30 pS) ACh-activated channels is present in addition to the predominant 45- to 50-pS ACh-activated channels. In preparations pretreated with both deltamethrin and staurosporine, after carbachol exposure, there was a significant decrease in the frequency of small-conductance ACh-activated channels. Deltamethrin treatment alone produced no small-conductance channels before or after a 10-min exposure to 540 microM carbachol. Also, no small-conductance ACh-gated channels were recorded at PKC-inhibited end plates after carbachol exposure either with pretreatment with 10 microM cyclosporin A (another inhibitor of calcineurin) or with the substitution of 10 mM Mn2+ for Ca2+ during the 10-min agonist exposure. 6. We propose that during prolonged exposure to the nicotinic agonist carbachol, calcium influx through ACh-gated channels elevates the level of ionized calcium at the inner surface of the post-junctional membrane and that this local rise in intracellular calcium activates the calcium-dependent phosphatase calcineurin. Dephosphorylation of some key membrane protein by calcineurin leads to a decrease in the extent of recovery from desensitization. Under normal conditions, this process is effectively reversed by PKC activity and end-plate sensitivity recovers fully. However, when PKC is inhibited, the extent of recovery of end-plate sensitivity is decreased, and associated with this decrease is the presence of small-conductance ACh-activated channels not normally recorded at snake twitch fiber end plates.