Skip to main content
PMC home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 1998 Sep 1;95(18):10938–10943. doi: 10.1073/pnas.95.18.10938

Modulation of a calcium-sensitive nonspecific cation channel by closely associated protein kinase and phosphatase activities

Gisela F Wilson 1,*, Neil S Magoski 1, Leonard K Kaczmarek 1,†
PMCID: PMC27999  PMID: 9724808

Abstract

Regulation of nonspecific cation channels often underlies neuronal bursting and other prolonged changes in neuronal activity. In bag cell neurons of Aplysia, it recently has been suggested that an intracellular messenger-induced increase in the activity of a nonspecific cation channel may underlie the onset of a 30-min period of spontaneous action potentials referred to as the “afterdischarge.” In patch clamp studies of the channel, we show that the open probability of the channel can be increased by an average of 10.7-fold by application of ATP to the cytoplasmic side of patches. Duration histograms indicate that the increase is primarily a result of a reduction in the duration and percentage of channel closures described by the slowest time constant. The increase in open probability was not observed using 5′-adenylylimidodiphosphate, a nonhydrolyzable ATP analog, and was blocked in the presence of H7 or the more specific calcium/phospholipid-dependent protein kinase C (PKC) inhibitor peptide(19–36). Because the increase in activity observed in response to ATP occurred without application of protein kinase, our results indicate that a kinase endogenous to excised patches mediates the effect. The effect of ATP could be reversed by exogenously applied protein phosphatase 1 or by a microcystin-sensitive phosphatase also endogenous to excised patches. These results, together with work demonstrating the presence of a protein tyrosine phosphatase in these patches, suggest that the cation channel is part of a regulatory complex including at least three enzymes. This complex may act as a molecular switch to activate the cation channel and, thereby, trigger the afterdischarge.

Modulation of ion channel activity by intracellular messengers underlies many prolonged changes in animal behavior (1). One dramatic example is found in Aplysia in which an increase in the excitability of bag cell neurons initiates behaviors that culminate in egg-laying (2). Specifically, brief stimulation of the pleuroabdominal connective nerve or application of reproductive tract peptides triggers a ≈30-min period of spontaneous repetitive action potentials known as the afterdischarge (3, 4). This is followed by a refractory period lasting 18 hr or more during which additional afterdischarges cannot be elicited (3, 5). Recent evidence suggests that activation of bag cell neuron nonspecific cation channels may provide the depolarizing drive underlying the afterdischarge (6, 7). As might be expected from the complex regulation of bag cell neuron activity, the regulation of cation channel activity also is complex. Cation channels are voltage- and Ca2+-sensitive and, upon patch excision, can be found in several distinguishable modes of gating. The most dramatic distinction occurs between the bursting and the continuously active gating modes. Bursting, a term that often is applied to the grouping of action potentials, here refers to groupings of channels openings. These groupings are separated by an added population of long-lived channel closures in the 1–100 s range (6). Cation channels in the bursting mode transit to a continuous high activity mode after treatment with T-cell protein tyrosine phosphatase (PTPase), which eliminates the long-lived closures. Transition from the bursting to the continuous activity mode can also be produced by treatment with cAMP-dependent protein kinase (PKA), which increases the activity of a PTPase that is endogenous to excised patches (6).

We now report that the activity of the bag cell neuron cation channel also is regulated by intracellular ATP. Because the increase in channel activity requires hydrolyzable nucleotides and because the effect of ATP is blocked by the kinase inhibitors H7 and calcium/phospholipid-dependent protein kinase C (PKC) peptide(19–36), the effect appears to be mediated by a PKC-like protein kinase. Consistent with this hypothesis, the effect of ATP is reversed by exogenously applied protein phosphatase 1 (PP1) or by a microcystin-sensitive phosphatase also endogenous to excised patches. The prevalence of the PKC-like kinase and two types of phosphatases in excised patches containing cation channels suggests that these enzymes and cation channels are colocalized. The enzyme complex may serve as part of a molecular switch that contributes to activation of the cation channel, the onset of the bag cell neuron afterdischarge, and the ensuing refractory period.

MATERIALS AND METHODS

Abdominal ganglia were excised from adult Aplysia californica (Alacrity Marine Biological Services, Redondo Beach, CA) that had been anesthetized by injecting isotonic MgCl2. Bag cell neurons were dissociated from ganglia and maintained in artificial seawater (containing, in mM: 460 NaCl/10.4 KCl/11 CaCl2/55 MgCl2/10 Hepes, pH 7.8) supplemented with glucose (1 mg/ml), penicillin (100 units/ml), and streptomyocin (0.1 mg/ml) as described (7, 8). Channel currents were recorded by using a List EPC-7 (Adams–List, Wesbury, NY) amplifier and then low-pass filtered at 3 kHz and stored on videocassettes. Unless otherwise noted, pipettes contained artificial seawater and the bath contained, in mM: 500 K-aspartate, 70 KCl, 0.77 CaCl2, 1.2 MgCl2, 10 Hepes, 11 glucose, 0.77 EGTA, and 10 glutathione (a reducing agent), pH 7.3 (KOH). Pipettes were coated with Sylgard (Dow Corning) and had resistances ranging from 3–10 MΩ. Junction potentials were nulled immediately before seal formation. In most cases, test solutions were applied by using a gravity-driven multibarrel perfusion system (9); in a few instances, reagents were bath-applied. Experiments were performed at room temperature (≈22°C).

Steady-state recordings were performed by using a holding potential of −60 mV. Open probabilities (Po) were determined by using a minimum of 3 min of channel recording. Although single channel patches were preferred, the Po for patches containing 2 or 3 channels was determined by adding the Po at each current level and dividing by the number of current levels observed. The numbers given in the text refer to the number of patches, not channels. Kinetic analyses were performed on patches judged to contain only one channel on the basis of the single current level observed. Currents were sampled at 11 kHz and digitally filtered at 1.5 kHz (Gaussian; −3 dB). Durations were determined by using a half-amplitude threshold criterion and corrected for the filter response (ref. 10, Eq. 17). Time constants (τ) were obtained by using the binned maximum likelihood method (11) and a simplex search that was provided with the number of exponentials and an approximate starting value for each exponential as described (6). Means and standard errors are given where applicable.

PP1, isolated from rabbit skeletal muscle (12), was a gift from the laboratory of D. L. Brautigan, University of Virginia, Dept. of Microbiology. PKA inhibitor peptide(6–22), ATP (grade 2 disodium salt), GTP (type 3 disodium salt), 5′-adenylylimidodiphospate (AMPPNP), and most other nucleotides were obtained from Sigma. ATPγS, microcsytin-LR, and PKC inhibitor peptide(19–36) were obtained from Calbiochem. H7 was obtained from Alexis/LC Laboratories (San Diego, CA).

RESULTS AND DISCUSSION

To examine the regulation of cation channels, we used inside-out patches excised from the cell bodies of single cultured bag cell neurons in their resting unstimulated state. Cation channels were observed in ≈20% of these patches and, when observed, most often occurred in clusters. A nucleotide effect on channel activity was first suggested by differences in the level of activity when patches were excised into nucleotide free solutions or into intracellular solutions containing ATP and GTP at 5 and 0.1 mM concentrations, respectively. For cation channels that were in the continuously active mode after patch excision, the fraction of time channels spent in the open state, or the Po, averaged 0.028 ± 0.007 (n = 24) in the absence of ATP and GTP. In contrast, the average Po for continuously active cation channels in the presence of ATP and GTP was 0.75 ± 0.06 (n = 8). Cation channels in the bursting mode, identified by the presence of a population of exceptionally long closed times (6), were not observed in the absence of the two nucleotides.

To determine which nucleotide was responsible for the difference in Po in the two solutions, the effect of application of either ATP or GTP to the intracellular face of cation channels was examined. No increase in Po was observed after application of GTP (0.1 mM; n = 3; data not shown). In fact GTP appeared to produce a slight decrease in channel activity. In contrast, as shown in Fig. 1A (Top), addition of ATP (2 mM) in the presence of 1.2 mM Mg2+ increased the cation channel Po from a control value of 0.163 to a final value of 0.616. ATP, at concentrations ranging from 0.25 to 5 mM, increased channel activity in all but two of 39 patches examined. Application of ATP never resulted in transitions to a bursting mode. The increases in activity observed in response to ATP were initiated within the first minute after application, although in some cases a steady plateau was not reached for ≈5 min (e.g., Fig. 4B). In the continued presence of ATP, the increases in activity persisted without rundown throughout the duration of the recordings (typically 15–30 min).

Figure 1.

Figure 1

Open in a new tab

Steady-state recordings of continuously active cation channels in excised inside-out patches from bag cell neurons. Holding potential = −60 mV. Channel openings are shown as downward deflections. (A Top) Single channel recordings immediately before and after application of 0.5 mM ATP. (A Middle) Single channel recordings showing the lack of an increase in activity after application of 1 mM AMPPNP. (A Bottom) Single channel recordings showing the lack of an increase in activity after application of 1 mM ADP. (B) Steady-state recordings showing block of the effect of ATP by 20 μM of the PKC inhibitor peptide(19–36).

Figure 4.

Figure 4

Open in a new tab

Reversal of the effect of ATP. (A) Steady-state recordings of cation channels showing reversal of the effect of ATP after return of the patch to a zero-ATP, control intracellular solution (Left), and block of the reversal in the presence of 200 nM microcystin-LR (Right). Holding potential, −60 mV. (B) Time course of the change in Po observed during treatment with ATP (2 mM) and, subsequently, PP1 (22 nM). Each data point represents the average Po obtained for one 30-s interval. Application of ATP and PP1 is indicated by the bars at the top of the figure. The effect of ATP was noticeable almost immediately, although ≈4 min were required for the Po to stabilize at the new higher value. In response to PP1, which was applied in the continued presence of ATP, the Po oscillated for several minutes before a complete reversal.

Fig. 2 summarizes the data obtained for 24 patches containing from 1 to 3 cation channels, before and during application of ATP. The dashed lines represent the value expected if no change in Po occurred. Both the starting Po values and the magnitude of the effect of ATP varied considerably from patch to patch. Starting Po values ranged from <0.001 to 0.163 and averaged 0.028 ± 0.008. The increase in Po observed in response to ATP ranged from 1.7 to 41.9-fold and averaged 10.7 ± 2.2-fold. The magnitude of the ATP effect did not appear to correlate strongly with the concentration of ATP. In addition, when the Po in control solution was <0.02, the average Po in ATP did not exceed 0.2; however, even in these patches a substantial increase in activity was observed as evidenced by the fold increase in Po (Fig. 2 Bottom). Although the source of the variability was not investigated further, some differences in the starting Po and the effect of ATP also may be due to differences in the modification of cation channels (or enzymes, see below) at the time of patch excision. For example, phosphorylation of the cation channel or a closely associated protein by PKA has been shown to decrease channel activity (6).

Figure 2.

Figure 2

Open in a new tab

Magnitude of the increase in Po observed after application of various concentrations of ATP. Data obtained for 24 patches containing from 1 to 3 cation channels. In both plots, the dashed line represents the value expected if no increase in Po were to occur. (Top) Comparison of the final average Po as a function of the average starting Po for each patch. Note that, in five of these patches, the final Po reaches the same level initially observed when patches were excised into intracellular solution containing 5 mM ATP and 0.1 mM GTP. (Bottom) Comparison of the fold increase in Po (Po ATP/Po control) as a function of the starting Po for each patch. Although the final Po in ATP observed for patches with a low initial Po of ≈0.02 or less appears minimal, examination of the fold increase in Po reveals a substantial increase in Po after application of ATP.

We next examined single channel open and closed times to identify the changes in kinetics that underlie the increase in Po. Fig. 3A shows the open and closed time distributions, plotted on logarithmic scales, for the single cation channel of Fig. 1 (Top) before and after application of 1 mM ATP. As reported previously (6), open time distributions were best fit by a sum of two exponentials, τo1 and τo2, and the majority of open times were described by the second, slower time constant. Closed times were well fit by a sum of three exponentials, τc1–τc3. The average values for these kinetic features are presented in Table 1 and represent the averages for five patches containing only a single cation channel. Because the number of exponentials is assumed to reflect the number of states a channel may enter, these results suggest that cation channels are complexly gated and can assume a minimum of two open and three closed states. Further examination of the event distributions reveals that ATP produced dramatic changes in the open and closed times (Fig. 3 and Table 1). The most prominent change was a reduction in the duration and percentage of channel closures described by τc3, the slowest time constant. After application of ATP, channel closures were shorter and the closures described by τc3 occurred less frequently. In addition, channel openings described by both τo1 and τo2 were longer in the presence of ATP, although the relative percentage of openings described by τo1 and τo2 did not appear to change. These changes in kinetic parameters were observed in response to ATP in all channels examined, regardless of the starting Po level or the magnitude of the effect (see Fig. 3B and Table 1).

Figure 3.

Figure 3

Open in a new tab

Open and closed time histograms obtained for two cation channels before and after application of 1 mM ATP. The exponential fits used to determine the time constants are indicated by the superimposed curves (see Table 1 legend and Materials and Methods). (A) Histograms for the channel shown in Fig. 1 (Top traces). The average control Po was 0.16, and the average Po in ATP was 0.62. (B) Histograms for a channel that exhibited a smaller increase in Po in response to ATP than that observed for the channel in A. For this channel, the average control Po was 0.06 and the average Po in ATP was 0.11. Note that ATP affected the same parameters for both channels.

Table 1.

Summary of the kinetic features of cation channels observed in single channel patches before and after application of ATP

Condition n Po τo1 (%) τo2 (%) τc1 (%)* τc2 (%)† τc3 (%)
Control 5 0.07  ±  0.02 1.02  ±  0.20 (43) 9.1  ±  1.5 (57) 0.46  ±  0.13 (42) 3.8  ±  1.2 (15) 384  ±  130 (44)
ATP 5 0.40  ±  0.11 1.37  ±  0.37 (43) 22.7  ±  8.5 (57) 0.58  ±  0.22 (51) 3.3  ±  1.5 (19) 69  ±  23 (25)
Open in a new tab

All τ values are expressed in milliseconds. Means are presented ± SEM. The percentage of events described by each exponential is given in parentheses. The single channel patches used for kinetic analysis were chosen on the basis of noise levels and the duration of recording; the average Po for all single channel patches of the present study (n = 14) was 0.04 ± 0.01 (control) and 0.25 ± 0.08 (in ATP). The concentration of ATP used in the five patches was, in mM, 0.5, 0.5, 1, 2, and 3. 

*

The briefest closures are poorly resolved at the employed filter frequency leading to an overestimation of τc1, and potentially obscuring any differences between the two conditions in gating events on the microsecond timescale. 

†

A somewhat improved fit was obtained if the closed durations described by τc2 (corresponding to the trough between the two major peaks) were fit with two exponentials instead of one. Variability in the τc2 column may be due to the conservative fit chosen. 

To characterize the mechanism underlying the effect of ATP on cation channel activity, we investigated the nucleotide specificity of the effect. The average Po obtained for patches containing from 1 to 3 channels before and during application of each nucleotide is given in Table 2. Examples of responses to AMPPNP and ADP are shown in the middle and bottom traces of Fig. 1A, respectively. No appreciable change in activity was noted in response to either ADP (1–2 mM, n = 8) or AMP (1 mM, n = 4). Similarly, we observed no increase in channel activity in the presence of up to 3 mM of AMPPNP (n = 8), a nonhydrolyzable ATP analog (13). In contrast, the activity of cation channels did increase in the presence of ATPγS (1–3 mM), a poorly hydrolyzable ATP analog that can substitute for ATP in many kinase reactions (n = 8) (14). As might be expected given the slower rates at which ATPγS is used in phosphotransferase reactions (14), in three of the eight patches examined, the initiation of the response to ATPγS occurred more slowly than the response to ATP, requiring between 2 and 5 min, and a final Po >0.50 was observed in only one case.

Table 2.

Effect of necleotides on cation channel Po and block of the effect

Condition n (N) Po control Po nucleotide Po wash
Nucleotides mM
 ATP 0.25-3 24 (39) 0.028  ±  0.007 0.184  ±  0.047
 ADP 1 5 (8) 0.003  ±  0.001 0.005  ±  0.002
 AMP 1 4 (4) 0.006  ±  0.005 0.006  ±  0.002
 AMP-PNP 1 3 (8) 0.042  ±  0.017 0.035  ±  0.024
 ATPγS 1 4 (8) 0.005  ±  0.003 0.020  ±  0.010
3 1 (8) 0.033 0.511
Inhibitors μM ATP
 H7 100 4 0.040  ±  0.010 0.017  ±  0.007
 PKC (19-36) 5 3 0.023  ±  0.005 0.018  ±  0.007
 PKC (19-36) 20 5 0.032  ±  0.015 0.026  ±  0.008
 PKA (6-22) 1 5 0.009  ±  0.004 0.075  ±  0.021
 Control - 6 (9) 0.053  ±  0.029 0.168  ±  0.086 0.038  ±  0.020
 Microcystin 0.010 3 (4) 0.001  ±  0.000 0.012  ±  0.001 0.004  ±  0.001
 Microcystin 0.200 4 0.007  ±  0.002 0.019  ±  0.005 0.016  ±  0.004
Open in a new tab

The values given are the average Po values (± S.E.M) obtained before and after application of each nucleotide at the concentrations indicated. Inhibitors were present for all conditions (control, ATP, and wash). The concentration of ATP used with kinase inhibitors was 1 mM; the concentration of ATP used with phosphatase inhibitors was 2 mM. n is the number of patches contributing to each determination of Po; the N given in parenthesis is the number of patches for which qualitatively similar results were obtained regardless of the number of channels present and at concentrations ranging from 1 to 3 mM, except in the case of ATP in which the concentration ranged from 0.25 to 5 mM. Po determinations were limited to patches containing from 1 to 3 channels (see Materials and Methods). 

Three possible mechanisms have been described for the modulation of ion channels by ATP. First, ATP regulates ATP-sensitive channels in heart and other tissues by a mechanism involving binding to an allosteric site (15). Secondly, in the case of the cystic fibrosis transmembrane conductance regulator (CFTR), ATPase-like activity appears to mediate the effect of ATP. Although the nucleotide specificity is broad, ATPγS cannot substitute for ATP and the effect appears to require nucleotide hydrolysis (16). Finally, ATP has been shown to modulate Ca2+-activated potassium channels by a phosphorylation-dependent mechanism involving a kinase closely associated with the channel proteins (17, 18). Our results with ATP analogs suggest that, as observed for CFTR and Ca2+-activated potassium channels, ATP hydrolysis is required for the modulation of bag cell neuron cation channels. In contrast to CFTR however, there appears to be a high specificity for ATP similar to that observed for a number of kinases (e.g., refs. 19 and 20) and ATPγS was able to substitute for ATP, albeit with a possible decreased efficiency. Together, these observations suggest that the effect of ATP is mediated by a kinase endogenous to excised patches.

To test the hypothesis that phosphorylation underlies the effect of ATP on cation channels, we examined the channel response to ATP in the presence of several kinase inhibitors. When cells were preincubated (40 min to 2 hr) in a bath solution containing the broad spectrum kinase inhibitor H7 (100 μM), ATP failed to increase the cation channel Po in each of the four patches examined (Table 2). Because, in bag cell neurons, H7 has been shown to block responses mediated by PKC, but not responses mediated by PKA (21), we next examined the effect of the highly specific, synthetic PKC inhibitor peptide(19–36). This peptide corresponds in sequence to those amino acids which comprise the pseudosubstrate autoinhibitory domain of PKC (22). ATP (1 mM) failed to increase the Po when applied in the presence of the PKC inhibitor peptide(19–36) in five of eight patches tested (5 and 20 μM; Table 2). Moreover, the overall mean Po in the presence of ATP was not noticeably different than the Po for these patches in control conditions (n = 8). In contrast, a normal response to ATP was observed in the presence of 1 μM of the PKA inhibitor peptide(6–22). These results indicate that ATP serves as a phosphate donor in a phosphorylation reaction mediated by a kinase endogenous to excised patches. Moreover, the kinase is PKC-like based on the inhibition produced by the PKC inhibitor peptide(19–36).

To further verify the involvement of ATP in a phosphorylation reaction, we next examined the reversibility of the effect of ATP. Surprisingly, when patches treated with ATP were returned to a control intracellular solution, a rapid and complete reversal was observed for all patches examined (n = 9; Fig. 4A Left and Table 2 Bottom). Given that no phosphatase had been added, we hypothesized that a phosphatase, specific for serine and threonine residues and also endogenous to excised patches, might be responsible for the reversal. Indeed, when the experiment was repeated in the presence of microcystin (200 nM), a potent inhibitor of PP1 and protein phosphatase 2A (PP2A) (23), the Po failed to return to baseline levels upon removal of ATP (n = 4; Fig. 4A Right and Table 2). Lowering the concentration of microcystin from 200 to 10 nM resulted in a 30% inhibition of the return to baseline levels (n = 4; Table 2). The reported specificity of microcystin for PP1 and PP2A (23) and previous observations indicating that PP2A, applied in the presence of ATP, increases, rather than decreases, the Po of cation channels (6), suggest that the endogenous phosphatase is PP1-like.

Exogenously applied PP1, purified from rabbit, also reversed the effect of ATP. PP1 was applied in the continued presence of ATP in an attempt to shift the balance between existing kinase and phosphatase activities. Fig. 4B, which plots the average Po obtained for successive 30-s intervals, shows the results of an experiment examining the effects of PP1. Upon patch excision, the cation channel displayed an initial Po ranging between 0.01 and 0.03. In response to ATP (2 mM), the Po increased, reaching a final plateau during which the average Po was 0.742. Application of PP1 (22 nM) in the continued presence of ATP caused the Po to return to baseline levels, albeit over a protracted time course. Although complete reversal required 20 min, dramatic oscillations in Po were observed ≈7 min after the addition of PP1. The oscillations, as well as the protracted time course of the reversal, are consistent with an ongoing competition between opposing kinase and phosphatase activities. PP1 also reversed the effect of ATP in the two other patches tested. No change in the average Po was observed when PP1 was applied to patches immediately after patch excision and before exposure to ATP (Po in PP1 = 0.031 ± 0.011; n = 7) suggesting that phosphorylation by the endogenous kinase is a necessary prerequisite to the effect of PP1. The simplest interpretation of these data are that purified PP1, like the endogenous phosphatase, dephosphorylates the residues phosphorylated by the endogenous kinase. Other interpretations that we cannot rule out, however, are that PP1 may dephosphorylate the endogenous, PKC-like kinase to down-regulate its activity, or that PP1 may dephosphorylate and increase the activity of the endogenous phosphatase.

In summary, our results indicate that ATP serves as a phosphate donor in a phosphorylation reaction mediated by a kinase endogenous to excised patches. This conclusion is based on evidence indicating a narrow nucleotide specificity, a requirement for hydrolyzable nucleotides, block of the effect by kinase inhibitors, and the ability of phosphatases to reverse the effect. Moreover, the kinase is PKC-like, as indicated by the inhibition produced by H7 and the PKC inhibitor peptide(19–36). Our data also indicate that a phosphatase present in excised patches serves to reverse the effect of ATP. The results obtained by using microcystin, combined with earlier results obtained by using PP2A (6), suggest that the endogenous phosphatase is PP1-like. In addition, purified PP1 also reversed the effect of ATP.

Perhaps the most striking observation is that not one, but three, enzymes appear consistently localized with cation channels in excised patches. Dendrotoxin-binding potassium channels purified from rat brain (24), type 2 Ca2+-activated potassium channels reconstituted in lipid bilayers (17, 25), and Ca2+-activated potassium channels present in excised patches of posterior pituitary nerve terminals (18, 26) also are modulated by consistently colocalized kinase, and in some cases, phosphatase activities. In the case of the cation channel, previous work has demonstrated that a PTPase activity colocalizes with these channels in patches excised from bag cell neurons (6). In the present study, a response to ATP was observed in all but two of the patches containing cation channels and reversed without the addition of phosphatase in every instance examined, demonstrating the presence of two additional enzymes. Close association between the channel and these enzymes could be the result of a broad distribution of enzymes throughout the bag cell neuron membrane or cytoskeletal network. An exceptionally high density of enzymes would however be required to account for their consistent presence in cation channel-containing patches. Alternatively, close association between the channel and these enzymes could be the result of protein-protein interactions. These interactions could be direct, or indirect and mediated by a “scaffolding” protein. Finally, the activity of at least one of these enzymes could be a property intrinsic to the channel protein. The coexistence of enzyme and ion channel in the same protein also has been suggested for some Ca2+-activated potassium channels (27, 28), adenylyl cyclase (29, 30), CFTR (16), and for Shaker potassium channels whose beta subunits belong to the aldo–keto reductase superfamily (31, 32). Given that the cation channel appears to colocalize with three enzymes, all three of these mechanisms may be involved.

The PKA-regulated PTPase previously shown to be endogenous to isolated patches removes exceptionally long-lived fourth and fifth closed states that are responsible for the bursting gating mode of the channel (6). The primary action of the PKC-like kinase of the present study is to reduce the favorability of the third-longest closed state. Both types of regulation can dramatically increase the activity of cation channels. Moreover, both types of regulation appear to occur when crammed patches are exposed to peptides that trigger the afterdischarge of bag cell neurons (7) and the activity of both PKA and PKC appears to increase during the afterdischarge (33, 34). These observations suggest that the cation channel may serve to integrate input from several signaling pathways. The enzyme complex associated with cation channels may serve as part of a molecular switch that contributes to activation of the cation channel and the onset of the bag cell neuron afterdischarge.

Acknowledgments

We thank F. J. Sigworth for providing software and assistance in the kinetic analysis of channel activity and D. L. Brautigan for providing PP1. This work was supported by an National Institutes of Health postdoctoral fellowship (F32 NS08986) to G.F.W., a Human Frontiers Science Program postdoctoral fellowship (LT-432–96) to N.S.M., and grants from the National Institutes of Health (NS18492) to L.K.K.

ABBREVIATIONS

PKA

cAMP-dependent protein kinase

PKC

calcium/phospholipid-dependent protein kinase C

AMPPNP

5′-adenylylimidodiphosphate

Po

open probability

PP1

protein phosphatase 1

PP2A

protein phosphatase 2A

PTPase

protein tyrosine phosphatase

CFTR

cystic fibrosis transmembrane conductance regulator

τ

time constants

Footnotes

This paper was submitted directly (Track II) to the Proceedings Office.

References

  • 1.Kaczmarek L K, Levitan I, editors. Neurmodulation: The Biochemical Control of Neuronal Excitability. New York: Oxford Univ. Press; 1987. [Google Scholar]
  • 2.Conn P J, Kaczmarek L K. Mol Neurobiol. 1989;3:237–273. doi: 10.1007/BF02740607. [DOI] [PubMed] [Google Scholar]
  • 3.Kupfermann I, Kandel E R. J Neurophysiol. 1970;33:865–876. doi: 10.1152/jn.1970.33.6.865. [DOI] [PubMed] [Google Scholar]
  • 4.Kaczmarek L K, Jennings K R, Strumwasser F. Proc Natl Acad Sci USA. 1978;75:5200–5204. doi: 10.1073/pnas.75.10.5200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kauer J A, Kaczmarek L K. J Neurosci. 1985;5:1339–1345. doi: 10.1523/JNEUROSCI.05-05-01339.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Wilson G F, Kaczmarek L K. Nature (London) 1993;366:433–438. doi: 10.1038/366433a0. [DOI] [PubMed] [Google Scholar]
  • 7.Wilson G F, Richardson F C, Fisher T E, Olivera B M, Kaczmarek L K. J Neurosci. 1996;16:3661–3671. doi: 10.1523/JNEUROSCI.16-11-03661.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Kaczmarek L K, Finbow M, Revel J P, Strumwasser F J. J Neurobiol. 1979;10:535–550. doi: 10.1002/neu.480100604. [DOI] [PubMed] [Google Scholar]
  • 9.Yellen G. Nature (London) 1982;296:357–359. doi: 10.1038/296357a0. [DOI] [PubMed] [Google Scholar]
  • 10.Colquhoun D, Sigworth F J. In: Single-Channel Recording. Sakmann B, Neher E, editors. New York: Plenum; 1983. pp. 191–263. [Google Scholar]
  • 11.Sigworth F J. In: Single-Channel Recording. Sakmann B, Neher E, editors. New York: Plenum; 1983. pp. 301–321. [Google Scholar]
  • 12.Brautigan D L, Shriner C L. Methods Enzymol. 1988;159:339–346. doi: 10.1016/0076-6879(88)59034-1. [DOI] [PubMed] [Google Scholar]
  • 13.Yount R G, Ojala D, Babcock D. Biochemistry. 1971;10:2490–2496. doi: 10.1021/bi00789a010. [DOI] [PubMed] [Google Scholar]
  • 14.Gratecos D, Fischer E. Biochem Biophys Res Commun. 1974;58:960–967. doi: 10.1016/s0006-291x(74)80237-8. [DOI] [PubMed] [Google Scholar]
  • 15.Ashcroft F M. Annu Rev Neurosci. 1988;11:97–118. doi: 10.1146/annurev.ne.11.030188.000525. [DOI] [PubMed] [Google Scholar]
  • 16.Anderson M P, Berger H A, Rich D P, Gregory R J, Smith A E, Welsh M J. Cell. 1991;67:775–784. doi: 10.1016/0092-8674(91)90072-7. [DOI] [PubMed] [Google Scholar]
  • 17.Chung S K, Reinhart P H, Martin B L, Brautigan D, Levitan I B. Science. 1991;253:560–562. doi: 10.1126/science.1857986. [DOI] [PubMed] [Google Scholar]
  • 18.Bielefeldt K, Jackson M B. Biophys J. 1994;66:1904–1914. doi: 10.1016/S0006-3495(94)80984-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Lemaire S, Labrie F, Gauthier M. Can J Biochem. 1974;52:137–141. doi: 10.1139/o74-021. [DOI] [PubMed] [Google Scholar]
  • 20.Sasaki N, Rees-Jones R W, Zick Y, Nissley S P, Rechler M M. J Biol Chem. 1985;260:9793–9804. [PubMed] [Google Scholar]
  • 21.Conn P J, Strong J A, Azhderian E M, Nairn A C, Greengard P, Kaczmarek L K. J Neurosci. 1989;9:473–479. doi: 10.1523/JNEUROSCI.09-02-00473.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.House C, Kemp B E. Science. 1987;238:1726–1728. doi: 10.1126/science.3686012. [DOI] [PubMed] [Google Scholar]
  • 23.MacKintosh C, Beattie K A, Klumpp S, Cohen P, Codd G A. FEBS Lett. 1990;264:187–192. doi: 10.1016/0014-5793(90)80245-e. [DOI] [PubMed] [Google Scholar]
  • 24.Rehm H, Pelzer S, Cochet C, Chambaz E, Tempel B L, Trautwein W, Pelzer D, Lazdunski M. Biochemistry. 1989;28:6455–6460. doi: 10.1021/bi00441a044. [DOI] [PubMed] [Google Scholar]
  • 25.Reinhart P H, Levitan I B. J Neurosci. 1995;15:4572–4579. doi: 10.1523/JNEUROSCI.15-06-04572.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Bielefeldt K, Jackson M B. J Physiol. 1994;475:241–254. doi: 10.1113/jphysiol.1994.sp020065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Atkinson N S, Robertson G A, Ganetzky B. Science. 1991;253:551–555. doi: 10.1126/science.1857984. [DOI] [PubMed] [Google Scholar]
  • 28.Adelman J P, Shen K, Kavanaugh M P, Warren R A, Wu Y, Lagrutta A, Bond C T, North R A. Neuron. 1992;9:209–216. doi: 10.1016/0896-6273(92)90160-f. [DOI] [PubMed] [Google Scholar]
  • 29.Krupinski J, Coussen F, Bakalyar H A, Tang W-J, Feinstein P G, Orth K, Slaughter C, Reed R R, Gilman A G. Science. 1989;244:1558–1564. doi: 10.1126/science.2472670. [DOI] [PubMed] [Google Scholar]
  • 30.Schultz J E, Klumpp S, Benz R, Schurhoff-Goeters W J, Schmid A. Science. 1992;255:600–603. doi: 10.1126/science.1371017. [DOI] [PubMed] [Google Scholar]
  • 31.McCormack T, McCormack K. Cell. 1994;79:1133–1135. doi: 10.1016/0092-8674(94)90004-3. [DOI] [PubMed] [Google Scholar]
  • 32.Chouinard S W, Wilson G F, Schlimgen A K, Ganetzky B. Proc Natl Acad Sci, USA. 1995;92:6763–6767. doi: 10.1073/pnas.92.15.6763. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kaczmarek L K, Jennings K R, Strumwasser F. Proc Natl Acad Sci, USA. 1978;75:5200–5204. doi: 10.1073/pnas.75.10.5200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Conn P J, Strong J A, Kaczmarek L K. J Neurosci. 1989;9:480–487. doi: 10.1523/JNEUROSCI.09-02-00480.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES