Actin filaments and the opposing actions of CaM kinase II and calcineurin in regulating alpha7-containing nicotinic receptors on chick ciliary ganglion neurons

Abstract

Nicotinic acetylcholine receptors containing alpha7 subunits have a high relative permeability to calcium and influence numerous calcium-dependent cellular events. On chick ciliary ganglion neurons the receptors are concentrated on somatic spines containing actin filaments. Using conventional whole-cell patch-clamp recording from dissociated ciliary ganglion neurons, we show that responses from alpha7-containing receptors undergo substantial rundown when the receptors are repeatedly challenged with nicotine. Stabilization of actin filaments with phalloidin partially prevents the rundown, whereas collapse of actin filaments with latrunculin A exacerbates it. The rundown depends on calcium influx through the receptors because it requires receptor activation and can be prevented by replacing extracellular calcium with barium or by intracellular dialysis with BAPTA. Thapsigargin and ryanodine each inhibit the rundown, demonstrating further a requirement for calcium release from internal stores. Blockade of calmodulin by calmidazolium or blockade of CaM kinase II with either KN93 or autocamtide-2-related inhibitory peptide each prevents the rundown; blockade of the phosphatase calcineurin with either cyclosporin A or deltamethrin increases the rundown. The results indicate a balance of calcium-dependent kinase and phosphatase activities in regulating the function of alpha7-containing receptors. Manifestation of the rundown depends in part on the loss of intracellular components via dialysis because little rundown is seen if perforated patch-clamp recording is used to monitor receptor responses even in latrunculin A-treated cells. A membrane-permeable calcineurin inhibitor, however, still decreases the nicotinic response in a calcium-dependent manner, confirming that calcium-dependent phosphoregulation of alpha7-containing receptors occurs in the intact cell.

Fig. 1.

Activity-dependent rundown of the nicotinic response and the effects of the actin cytoskeleton. A,Whole-cell conventional patch-clamp recordings of nicotine-induced currents in dissociated chick ciliary ganglion neurons. Nicotine (20 μm; Nic) was delivered for a duration of 1 sec from a rapid applicator at 1 min intervals over a 15 min test period. The first (1st test) and last (15th test) responses are shown. Cells contained either untreated actin filaments (Control; top), actin filaments collapsed by treatment with latrunculin A (Lat; middle), or actin filaments stabilized with phalloidin (Phall; bottom). The holding potential was −60 mV. B,Time course of rundown for the nicotinic response. Nicotine-induced peak currents were normalized to the initial amplitude for each individual cell; values represent the mean ± SEM of eight cells per condition. C, Proportion of the peak current remaining at the end of the 15 min test period for cells treated as indicated and stimulated (Stim) either at 1 min (openbars) or 5 min (hatchedbars) intervals. The rundown was stimulation dependent, augmented by latrunculin A treatment (p < 0.005), and diminished by phalloidin (p < 0.01). D, Peak amplitude of the initial response to nicotine. No significant differences were found among the groups (p> 0.5). E, Decay time constants for the fast and slowly decaying components. No significant differences were found among the groups (p > 0.5).

Fig. 2.

Subtype specificity of nicotinic receptor rundown.A, Whole-cell patch-clamp recordings showing the total and α-bungarotoxin (αBgt)-resistant portion of the nicotine-induced current in the same neuron at the end of the 15 min rundown period. Nicotine (20 μm; Nic) was delivered for 1 sec at 1 min intervals over a 15 min period before testing immediately before (Control) and after (αBgt) incubating with 100 nm αBgt for 3 min. Most of the slowly desensitizing response was resistant to αBgt and therefore attributable to α3*-nAChRs. B,Proportion of the slowly desensitizing response attributable to α3*-nAChRs before and after rundown. The amplitude of the nicotine-induced current (openbars) was measured 250 msec after initiating the response in the same neurons before (BeforeRundown) and after (AfterRundown) delivering the 15 min stimulation protocol. αBgt was then applied to the same neurons to determine the proportion of the response caused by α3*-nAChRs at the end of the rundown period (hatchedbars). The proportion of the response attributable to α3*-nAChRs before rundown was determined by measuring the 250 msec values in records obtained previously (Liu and Berg, 1999). The 15 min stimulation protocol produced a decrease in the slowly desensitizing αBgt-sensitive response measured at 250 msec (p ≤ 0.01) as it did in the rapidly desensitizing peak response attributable to α7-nAChRs (see Fig. 1), but it produced no significant rundown of the α3*-nAChR response (p > 0.3). The values represent the mean ± SEM of 6–10 cells and have been normalized for the current detected initially.

Fig. 3.

Dependence of α7-nAChR rundown on intracellular calcium accumulation. A, Changes in the time course of rundown caused by replacing extracellular calcium with barium (Ba) or internally dialyzing with 10 mmBAPTA (BAPTA) both in untreated cells and in cells treated with latrunculin A (Lat). Stimulation protocols are described in Figure 1. Values represent the mean ± SEM of six to eight cells each and have been normalized for the peak current present at the outset. B, Comparison of the peak current amplitude remaining at the end of the test period for untreated (Control) and latrunculin A-treated (Lat) neurons tested with normal recording solution (openbars; taken from Fig.1E), extracellular barium instead of calcium (hatchedbars), and intracellular BAPTA (stippledbars). The calcium replacement and the internal perfusion with BAPTA each protected against rundown in both the control and latrunculin A-treated cells (p < 0.005 in all cases).

Fig. 4.

Calcium-induced calcium release from internal stores helps mediate rundown of the α7-nAChR response.A, Time course of rundown for nicotinic currents after blockade of calcium release from internal stores in cells with and without latrunculin A treatment. Normal (Control) or latrunculin A-treated (Lat) cells were internally dialyzed with 10 μm ryanodine (Ryan) or 1 μm thapsigargin (Thap) before testing. In the case of thapsigargin, the cells were also incubated with the compound 10 min before patch formation. B, Peak current remaining at the end of the rundown period, normalized for the size of the initial response in the same cell. Ryanodine and thapsigargin each significantly retarded rundown in control cells (p < 0.01 and 0.005, respectively), as did thapsigargin in latrunculin A-treated cells (p < 0.005). Values represent the mean ± SEM of six to eight cells each.

Fig. 5.

Calcium-dependent phosphoregulation of α7-nAChR responses. A, Whole-cell patch-clamp recordings from neurons at the beginning (1st test) and end (15th test) of the 15 min rundown protocol when the cells were internally dialyzed with the calmodulin (CaM) antagonist calmidazolium (Calmid; 10 μm;top), the calcineurin (CaN) inhibitor cyclosporin A (cyclosp; 200 nm; middle), or the CaM kinase (CaMKII) inhibitor KN93 (KN93; 50 μm;bottom).B, Rundown time course for the α7-nAChR response when cells were internally dialyzed with the indicated compounds from the patch pipette during the rundown protocol. Nicotine was first applied 3 min after dialysis was initiated by rupture of the membrane under the patch. Deltamethrin (DeltaM; 200 nm; a calcineurin inhibitor), KN92 (KN92; 50 μm;an inactive analog of KN93), autocamtide-2-related inhibitory peptide (AIP; 5 μm; an inhibitor of CaM kinase II), and other compounds described in A were tested.C, Peak current remaining at the end of the rundown period for each of the test conditions, normalized to the initial peak response in the same cells. In Control, the patch pipette contained normal solution. The calmodulin and CaM kinase blockers each inhibited rundown (p < 0.01 and 0.005, respectively), whereas the calcineurin blockers increased it (p < 0.01); the inactive KN analog had no effect (p > 0.5). Values represent the mean ± SEM of six to eight cells each. Nic, Nicotine.

Fig. 6.

Calcium-dependent phosphoregulation in latrunculin A-treated cells. A, Time course of rundown in latrunculin A-treated (Lat) cells internally dialyzed with the indicated compounds from the patch pipette. Drug concentrations were the same as in Figure 5, except that an ATP-generating system (ATP) was included where indicated. B, Peak current remaining at the end of the rundown period for each of the test conditions, normalized to the initial peak response in the same cells. The ATP-generating system, the calmodulin blocker [calmidazolium (Calmid)], and the CaM kinase blocker [autocamtide-2-related inhibitory peptide (AIP)] each partially inhibited the rundown (p < 0.01, 0.05, and 0.001, respectively), although not as much as that seen in control cells lacking exposure to latrunculin A (see Fig. 5 and text). Combining the ATP-generating system with either the calmodulin blocker or the CaM kinase II blocker produced more substantial inhibition (p < 0.001), rivaling that obtained in cells without latrunculin A treatment. Values represent the mean ± SEM of six to eight cells each.

Fig. 7.

Dependence of rundown on intracellular dialysis.A, Perforated patch-clamp recordings (Perf-patch records) of responses elicited by 20 μm nicotine (Nic) from the same cell at the beginning (1st test) and end (15th test) of the standard rundown protocol. Top, Control cells.Second, Latrunculin A-treated cells (Lat). Third, Deltamethrin-treated cells (DeltaM). Bottom, Deltamethrin-treated cells tested in barium instead of calcium (Ba + DeltaM). B, Time course of rundown monitored with perforated patch-clamp recording in control (Perf) and latrunculin A-treated cells (Perf + Lat) under normal conditions or after bathing with 200 nm deltamethrin (Perf + DeltaM, initiated at the arrow) or after bathing with deltamethrin after replacing extracellular calcium with barium (Perf + Ba + DeltaM). C, Peak current remaining at the end of the rundown period for each of the test conditions, normalized to the initial peak response in the same cells. Rundown in the perforated patch-clamp configuration both for normal and latrunculin A-treated cells was significantly less (p < 0.001) than that seen with conventional patch-clamp recording (Control, from Fig.1C, shown for comparison). The membrane-permeant calcineurin inhibitor deltamethrin reduced the response during the perforated patch recording (p < 0.05), and the reduction was blocked by replacing the extracellular calcium with barium (p > 0.5 vs control;p < 0.05 vs deltamethrin) or by stimulating three times before applying the deltamethrin and then not again until the end (DeltaM/OneStim; p > 0.5 vs control; p < 0.05 vs deltamethrin with regular stimulation). Values represent the mean ± SEM of six to eight cells each.