The voltage-dependent inhibition of N-type calcium current by neurotransmitters is the best-understood example of neuronal calcium channel inhibition. One of the mechanisms by which this pathway is thought to inhibit the calcium current is by reducing the permeation of divalent cations through the channel. In this study one prediction of this hypothesis was examined, that high concentrations of divalent cations reduce the maximum neurotransmitter-induced inhibition. Norepinephrine (NE)-induced inhibition was compared in external solutions containing either 2 or 100 mM Ba2+. Initially, NE dose-response curves were generated by averaging data from many neurons, and it was found that the relationship was right shifted in the high-Ba2+ external solution without an effect on maximum inhibition. The IC50 was 0.6 and 3 μM in 2 and 100 mM Ba2+, respectively. This shift was verified by comparing the effect of NE on single neurons exposed to both 2 and 100 mM Ba2+. The inhibition induced by 1 μM NE was reduced in 100 mM Ba2+ compared with that in 2 mM Ba2+. However, the response to 100 μM NE was identical between high and low Ba2+. Thus, divalent cations appear to act as a competitive inhibitor of NE binding, which likely results from these ions' interacting with negatively charged amino acids that are important for catecholamine binding to adrenergic receptors. Because the maximum inhibition induced by NE was similar in low and high Ba2+, the effect of inhibition on single N-type calcium channels was not altered by the divalent cation concentration.
Many neurotransmitters have been shown to inhibit N-type calcium channels via a voltage-dependent and membrane delimited pathway (Hille 1994; Jones and Elmslie 1997). The inhibition appears to be mediated intracellularly by G protein βγ subunits (Herlitze et al. 1996;Ikeda 1996), which are thought to bind directly to the N-type calcium channel. The voltage dependence appears to arise from the transient dissociation of the βγ subunit from the N-channel (Ehrlich and Elmslie 1995; Elmslie and Jones 1994; Golard and Siegelbaum 1993). This inhibition has been hypothesized to place the N-channel into a “reluctant” gating mode (Bean 1989; Elmslie et al. 1990). Based on whole-cell recordings, this reluctant gating mode is thought to possess a lower open probability (Po), shorter open times (Elmslie et al. 1990), and a smaller single-channel current amplitude (Kuo and Bean 1993) than the normal “willing” mode. If this is true, N-channel inhibition results from changes in both permeation and gating. Initial reports of single N-channel gating in the presence of neurotransmitter failed to observe reluctant gating (Carabelli et al. 1996; Patil et al. 1996). However, support for reluctant gating comes from whole-cell recordings showing that tail currents during inhibition are faster than control (Boland and Bean 1993;Elmslie et al. 1990). In addition, two laboratories have presented preliminary data showing that putative reluctant events can be observed at voltages depolarized to those used previously (Colecraft et al. 1999; Lee and Elmslie 1997;). As predicted by the willing-reluctant model, these putative reluctant events have low Po and brief openings compared with willing events.
It is clear that recordings from single N-channels are key to determining the effect of inhibition on the channel. However,Kuo and Bean (1993) presented data suggesting that the isotonic Ba2+ typically used to record single calcium channels may obscure the effect of inhibition on N-channel permeation. Using bullfrog sympathetic neurons they showed that N-channel inhibition by GTPγS (a direct G protein activator) was reduced in high Ba2+ (Kuo and Bean 1993). They postulated that this reduction resulted from a loss of the permeation component of inhibition. Thus inhibition in high Ba2+ may result primarily from changes in gating, whereas inhibition in low Ba2+ results from changes in both gating and permeation.
Since Kuo and Bean (1993) published their finding, it has been demonstrated in bullfrog sympathetic neurons that whole-cell recordings using isotonic Ba2+ are contaminated by a previously unrecognized calcium current, which we have called novel current or, more recently, Ef-current (Elmslie et al. 1994; Elmslie 1997). This current is insensitive to dihydropyridines, ω-conotoxin GVIA, and NE (Elmslie et al. 1994), and we have estimated that it comprises 30–50% of the total whole-cell calcium current in external solutions containing 100 mM Ba2+. However, Ef-current has not been observed in recordings using a 2-mM Ba2+ external solution (Elmslie et al. 1992; Elmslie et al. 1994). The presence of the neurotransmitter-insensitive Ef-current could be responsible for the apparent reduction of calcium current inhibition in high Ba2+ observed by Kuo and Bean (1993). Therefore, we compared the NE-induced inhibition of N-type calcium current in low and high Ba2+ to determine whether high Ba2+ alters inhibition. We found that, with Ef-current removed, the maximum inhibition of N-current by NE was the same in both low and high Ba2+. If gating and permeation are altered by inhibition in low Ba2+, then both are altered in high Ba2+. Thus, single N-channel recordings can be used to test the prediction that inhibition alters N-channel permeation. Part of this work has been published in abstract form.
Neurons were dissociated from paravertebral sympathetic ganglia of adult bullfrogs (Rana catesbeiana), as described previously (Elmslie 1992). Isolated neurons were maintained at 4°C until use. Whole-cell currents were recorded at room temperature.
The internal pipette solution consisted of (in mM) 82.6 NMG (N-methyl-d-glucamine) · Cl, 2.5 NMG · HEPES, 10 NMG2 · EGTA, 5 Tris2 · ATP, 6 MgCl2, and 0.3 Li2 · GTP, titrated to pH 7.2 with NMG base. The low-Ba2+ external solution contained (in mM) 2 BaCl2, 117.5 NMG · Cl, and 2.5 NMG · HEPES (pH 7.2). The high-Ba2+external contained (in mM) 100 BaCl2, 10 NMG · HEPES, 5 MnCl2, and 10 tetraethylammonium · Cl, pH 7.2.
The electrodes were fabricated from Corning 7052 glass (1.5 mm OD, 0.86 mm ID; A-M Systems, Everett, WA), and the series resistance ranged from 1 to 2 MΩ. Series resistance compensation was set to 95% using the circuitry of the Axopatch 200A amplifier (Axon Instruments, Foster City, CA). The experiment was controlled by a Macintosh II computer running S3 data acquisition software written by Dr. Stephen Ikeda (Guthrie Research Institute, Sayre, PA). Records were leak subtracted using averaged and scaled hyperpolarizing steps of one-quarter amplitude. As previously described, currents were measured as the average between 2.5 and 5 ms into the voltage step (Elmslie 1992). The facilitation ratio was calculated by dividing the postpulse current (after strong depolarization) by the prepulse current (preceding strong depolarization). In control and recovery, this value was ∼1. During NE, this value was positively correlated with the magnitude of inhibition.
In our first set of experiments, we exposed single neurons to NE concentrations of 1, 3, 10, 30, and 100 μM to obtain a dose-response relationship (Fig. 1). Only neurons exposed to at least three NE concentrations were analyzed. For these experiments each cell was exposed to a single external solution (i.e., either 2 or 100 mM Ba2+). The different NE concentrations were applied in a random order across cells to minimize the effects of desensitization on the averaged data.
However, NE responses can be highly variable when compared across cells. In our second set of experiments (Fig.2), we tested a single concentration of NE on the same cell in both external solutions. This allowed us to make within-cell comparisons, but the multiple NE applications induced desensitization. We used the following procedure to compensate for the desensitization. For each cell, one Ba2+ solution was selected as the “control” external and the other as the “test” external. The Ba2+ external solution selected as control was varied across cells. NE responses were measured in each external solution with responses in the control solution bracketing the NE response elicited in the test solution. The two NE responses in the control Ba2+ solution were averaged to compensate for desensitization during the experiment and compared with the NE response in the test external solution.
The third set of experiments in this study examined the effect of holding potential on the NE response in high external Ba2+ (Fig. 3). In these experiments, applications of NE were made on each cell as the holding potential was varied between −80 mV and −40 mV. One holding potential was selected as control and the other as test. NE responses during the control holding potential bracketed that during the test holding potential, and the two control responses were averaged for comparison with the test response. The control holding potential was varied across cells.
Changing the external Ba2+ concentration from 2 to 100 mM had several effects on the calcium current. The current amplitude increased approximately fourfold, voltage-dependent activation and inactivation shifted ∼40 mV to the right, and an ω-conotoxin GVIA–insensitive calcium current was revealed (Elmslie et al. 1994). We were interested in the effect of high Ba2+ on the NE-induced inhibition of N-type calcium current and therefore made several adjustments to compensate for the other effects of high Ba2+ on the current. To compensate for the shift in voltage-dependent properties, the holding potential and step voltages were generally depolarized by 40 mV when testing in high Ba2+(see Fig. 2). The depolarized holding potential had the additional benefit of almost completely inactivating Ef-current (Elmslie et al. 1994), so that the NE effect was not altered by the presence of this NE-insensitive current. Finally, Mn2+ was added to the 100-mM Ba2+ external solution to reduce the amplitude of the current to a level that could be well voltage clamped. Previously, we have demonstrated that observations made in high-Ba2+–containing external solutions are not altered by the presence of Mn2+ (Elmslie et al. 1994).
The effect of high Ba2+ on the NE response was initially examined by comparing dose-response relationships generated by exposing single neurons to a range of NE concentrations (1–100 μM) in a single external Ba2+ solution (either 2 or 100 mM). If the effect of high Ba2+ is to block the permeation component of inhibition, we predicted that inhibition would be reduced at all NE concentrations in high Ba2+ compared with low Ba2+. Alternatively, if Ba2+ was acting as a competitive inhibitor of NE binding to its receptor, N-current inhibition would be reduced at low NE concentrations, but high concentrations of NE would overcome the Ba2+ block. The data show that the NE dose-response relation was right shifted in high-Ba2+ external solution with a calculated IC50 of 3 μM in 100 mM Ba2+ and a maximum inhibition of 30% (Fig. 1). In the low-Ba2+ external, the IC50 was 0.6 μM and the maximum inhibition was 30% (Fig. 1). Although we did not test NE concentrations below 1 μM, the IC50 compares well with values previously published using 2 mM Ba2+ (Elmslie 1992; IC50 = 0.6 μM, maximum inhibition 48%). Thus, the high external Ba2+ solution shifted the NE dose-response relationship to the right without altering the maximum inhibition, as though Ba2+ were acting as a competitive inhibitor of the adrenergic receptor.
We have found large variations in NE responses measured from different cells. Therefore, we wanted to verify these results in a completely separate set of experiments in which we compared the effect of a single NE concentration in both high and low Ba2+ within the same neuron (Fig. 2). The high-Ba2+ external solution reduced the inhibition induced by 1 μM NE to an average of 18 ± 8% (mean ± SD, n = 6 cells) when compared with the average inhibition of 52 ± 14% in 2-mM Ba2+ solution (Fig. 2 B). This difference was not a result of desensitization, because smaller responses were observed when NE was applied in high Ba2+ before testing the response in low Ba2+. However, high Ba2+had less effect on the NE-induced inhibition as the concentration of NE increased and became negligible at 100 μM NE (Fig. 2, Cand D). Note in Fig. 2 that the average inhibition in 1 μM NE is larger than that in 100 μM NE. This was caused by the variability in the NE response that can result from comparing across cells (Fig. 2). Different neuronal preparations were used to test each NE concentration. The neurons used to examine 1 μM NE responded more strongly than those used to examine 100 μM NE. This demonstrates the importance of the within-cell comparison to verify our conclusion from the dose-response relationships that were generated by averaging responses from many cells.
The previous two sets of experiments have demonstrated that the NE dose-response relationship is right-shifted in our high-Ba2+ external solution when compared with the low-Ba2+ external solution. Mn2+ was added to the high-Ba2+ external solution to reduce the amplitude of the N-current so that we could more easily maintain voltage control in the 100-mM Ba2+ solution. Because Mn2+ was absent from the low-Ba2+ external, we wanted to test whether 100 mM Ba2+ alone could alter the NE inhibition. Using the same procedures as in the previous experiment, we examined the effect of 1 μM NE in both 2- and 100-mM Ba2+ external solutions (without Mn2+) in the same neuron. One micromolar NE was chosen for this test, because the largest difference in inhibition between low- and high-Ba2+ external solutions was observed at this concentration. In five cells examined the average inhibition was 38.7 ± 12.5% in 2 mM Ba2+and 25.6 ± 11.8% in 100 mM Ba2+. The high-Ba2+ external solution (without Mn2+) decreased the NE-induced inhibition by 33%. However, this reduction was smaller than that observed with the 100-mM Ba2+ + 5-mM Mn2+external solution. In that solution the inhibition induced by 1 μM NE was reduced by ∼65% (for data in both Figs. 1 and 2). Thus 100 mM Ba2+ alone can alter the NE inhibition. The smaller effect of high Ba2+ without Mn2+ could have resulted from poor voltage control during the large calcium currents. On the other hand, both Mn2+ and Ba2+ may have affected the NE response.
Our results show that the maximum NE response is not affected by the concentration of external Ba2+. However,Kuo and Bean (1993) showed a reduced maximal inhibition (induced by GTPγS) in high Ba2+ compared with low Ba2+. One difference is that our high-Ba2+ recordings were obtained from a holding potential of −40 mV to inactivate Ef-current, whereas Kuo and Bean (1993) used a holding potential of −90 mV. The consequence of the hyperpolarized holding potential in 100 mM Ba2+ is that Ef-current comprises a substantial portion of the total current (30–50%). However, it does not significantly contribute to the whole-cell current when using low-Ba2+external (Elmslie et al. 1992; Elmslie et al. 1994). To test whether Ef-current reduces the magnitude of the NE response, we compared the effect of 30 μM NE in high Ba2+ from holding potentials of −80 mV and −40 mV (Fig. 3). It has been demonstrated that N-current is not inactivated by the −40-mV holding potential in isotonic Ba2+ (Elmslie et al. 1994;Elmslie 1997). At a holding potential of −80 mV, the NE-induced inhibition was significantly reduced, compared with that measured from a holding potential of −40 mV in the same cell. The average inhibition was 32 ± 3% and 39 ± 7% for holding potentials of −80 mV and −40 mV, respectively (n = 5). This difference cannot be explained by desensitization, because the NE response was reduced even when NE was tested from the −80-mV holding potential before it was tested from the −40-mV holding potential. The difference in the NE responses amounted to an 18% reduction in the inhibition at −80 mV when compared with −40 mV, which is similar to the reduction observed by Kuo and Bean (1993) in the GTPγS effect when comparing low versus high external Ba2+ solutions. Thus, the presence of the neurotransmitter-insensitive Ef-current reduces the magnitude of the calcium current inhibition.
We have shown that a high-Ba2+ external solution alters the NE response by shifting the dose-response relationship to the right, but that the maximum NE inhibition is not affected. Our results differ from those of Kuo and Bean (1993), who found that the maximum inhibition induced by GTPγS was reduced by high external Ba2+. It is unlikely that the differences between the two studies result from the methods used to inhibit the current, because GTPγS and NE induce inhibitions with identical characteristics (Elmslie 1992). The most likely explanation is that the high-Ba2+ recordings of Kuo and Bean (1993) were contaminated by Ef-current. In the majority of our experiments, we used a holding potential of −40 mV when recording in high-Ba2+ external solution to inactivate Ef-current. When we used a hyperpolarized holding potential, similar to Kuo and Bean (1993), the NE response was significantly reduced.
The effect of our high-Ba2+ external solution on the NE response is consistent with a competitive block of the adrenergic receptor. Because the concentrations of both Ba2+ and Mn2+ were altered between the low- and high-Ba2+ external solutions, both could be blockers of the receptor. When we used a high-Ba2+ external without Mn2+, the reduction in NE inhibition was smaller than when Mn2+ was present. Thus, it is possible that both Mn2+ and Ba2+participate in the apparent competitive block of the adrenergic receptor. Effects of divalent cations on the binding of agonists and antagonists to adrenergic receptors have been previously investigated by examining the binding of radioactive compounds to membrane preparations. At concentrations in the 0.1–10-mM range, divalent cations were found to facilitation agonist binding and to inhibit antagonist binding to adrenergic receptors (Asakura et al. 1984; Jarrott et al. 1982; Loftus et al. 1984; Nomura et al. 1984; Rouot et al. 1982). The effect on agonist binding primarily resulted from an increased number of binding sites, which was thought to result from interactions between divalent cations and G proteins to switch the receptor from a low-affinity to a high-affinity state (Asakura et al. 1984). Unlike these binding studies, the applied divalent cations in our study did not have access to the intracellular face of the membrane. In addition, our internal solution contained 10 mM EGTA to prevent increases in the internal concentration of divalent cations. Thus, we should not observe these effects. In the binding studies, concentrations of divalent cations higher than ∼10 mM resulted in lower agonist binding (Asakura et al. 1984;Loftus et al. 1984; Nomura et al. 1984). However, this effect of divalent cations was not examined in detail; thus, no explanation was proposed for this effect. Since these binding experiments were conducted, much information has been revealed about the composition of the catecholamine-binding site on the adrenergic receptors, which provides clues about where divalent cations may be exerting their blocking effect. It is known that one or more aspartic acid residues are crucial for high-affinity binding of catecholamines to adrenergic receptors (Strader et al. 1994). The negatively charged acidic group on the aspartate residue is thought to interact with the positively charged amine of the catecholamine. Divalent cations may be interacting with the negatively charged amino acids in the binding site. Such an interaction could interfere with the binding of NE to produce a competitive block of the adrenergic receptor that we observe.
The conclusion of Kuo and Bean (1993) that voltage-dependent inhibition alters N-channel permeation was primarily supported by reduced inhibition observed when monovalent cations were permeating the N-channel. Their observation of a reduction of inhibition by isotonic Ba2+ was interpreted within this framework as evidence that the permeation component of inhibition was reduced by the high Ba2+concentration. If this were true, the high Ba2+typically used in single-channel recordings would mask the effect of neurotransmitters on permeation. Our results show that after adjusting for the right-shift in the dose-response relationship, the NE response was not reduced by high Ba2+. Therefore, if inhibition results in changes in N-channel permeation, these changes should be observed in single-channel studies using isotonic Ba2+. Previous single N-channel studies have observed a neurotransmitter-induced increase in the latency-to-first channel opening, but these studies have failed to observe changes in open times and single-channel current amplitude (Carabelli et al. 1996; Patil et al. 1996). These studies focused on N-channel activity at voltages less than +40 mV. Data from two recent preliminary reports show that reluctant openings can be observed only at voltages of +40 mV or more (Colecroft et al. 1999; Lee and Elmslie 1997). A detailed analysis of these openings will determine whether both gating and permeation are altered during neurotransmitter inhibition of N-type calcium current. We conclude that if permeation is a mechanism of N-channel inhibition, it should be observed in single-channel recordings using high external Ba2+ as the charge carrier.
We thank Drs. Geoffrey G. Schofield and Yong Sook Goo and H. Liang for helpful comments on the manuscript.
This study was supported by National Institute of Neurological Disorders and Stroke Grant NS-33671.
Address for reprint requests: K. S. Elmslie, Dept. of Physiology, SL-39, Tulane University Medical School, 1430 Tulane Ave., New Orleans, LA 70112.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2000 The American Physiological Society