The Effect of Zinc on Calcium and Hydrogen Ion Currents in Intact Snail Neurones

Voltage-dependent hydrogen ion movements have been detected in a number of tissues (Thomas & Meech, 1982; Byerly et al. 1984; Barish & Baud; 1984; Mahaut-Smith, 1987, 1989; Henderson et al. 1987; Thomas, 1988a,b). In molluscan neurones, continuous activation of the H⁺ pathway can produce substantial changes in pHi (Thomas & Meech, 1982), but the physiological role of these H⁺ movements has not been investigated. A number of agents block H⁺ currents, but the most effective are heavy metal cations such as Cd²⁺, La³⁺ and Zn²⁺, which also block Ca²⁺ currents in these cells.

This study shows that H⁺ currents in intact Helix neurones are blocked by micromolar concentrations of Zn²⁺. Ca²⁺ currents are also blocked by Zn²⁺, but at higher concentrations. Zn²⁺ can therefore be used to detect the presence of H⁺ currents in studies of Ca²⁺ currents. This blocker may also be useful in studying the role of the H⁺ pathway in normal cell function. A preliminary report of the results has appeared in abstract form (Mahaut-Smith, 1987).

The preparation and experimental set-up were identical to those described in the preceding paper (Mahaut-Smith, 1989). ZnC1₂ was obtained from BDH Chemicals.

The previous paper describes conditions which allow the separation of H⁺ currents in intact Helix neurones. Briefly, cells were held under voltage-clamp using CsCl-filled microelectrodes and exposed to a 0Na⁺, TEA⁺ saline. The membrane potential was held at -30 mV for 10 min and finally at −10 or −15 mV. The low holding potentials, and the saline used, induce a cytoplasmic acidification and abolish Ca²⁺ and remaining K⁺ currents. As a result, depolarizing pulses from these low holding potentials activate fast outward currents, which are carried by H⁺, at least for pulses up to +55 mV (Mahaut-Smith, 1989).

In preliminary experiments, a number of agents known to affect voltagedependent outward currents or H⁺ movements were tested on outward H⁺ currents in intact neurones. The following compounds had no significant effect when added to the bath; V-ethylmaleimide (2 mmol l⁻¹), diethylpyrocarbonate (1 mmol l⁻¹), amiloride (100 μmol l⁻¹), apamin (Sigma) (0·3 μmol l⁻¹). 1·2 mmol l⁻¹ external Cd²⁺ blocked a large proportion of these H⁺ currents, but by far the most effective blocker was Zn²⁺.

Fig. 1A shows H⁺ currents activated by a series of depolarizing voltage steps from a holding potential of −15 mV up to 55 mV in 10 mV increments. 0·1 mmol l⁻¹ ZnCl₂ added to the bath caused a large reduction in the H⁺ current, an effect which was slow to reverse and only partially recovered after a 30 min wash in 0Na⁺, TEA⁺ saline. Subsequent application of 1 mmol l⁻¹ external ZnCl₂ virtually abolished the outward current. Hyperpolarizing voltage pulses elicited currents which behaved in an Ohmic fashion. Therefore, leak currents for depolarizing pulses were obtained by extrapolating from the Ohmic relationship, measured regularly throughout the experiment. Fig. 1B is a plot of the peak currents during a 250 ms pulse, corrected for the leak current, as a function of the membrane potential. The reduction in current caused by Zn²⁺ appears to be constant within the range of potentials used, which suggests that the block is not voltage-dependent. In the presence of 1 mmol l⁻¹ Zn²⁺, a small voltage-dependent current was observed during steps to +55 mV or more depolarized, which may be residual K⁺ current (Mahaut-Smith, 1989). At these concentrations, divalent cations such as Zn²⁺ and Cd²⁺ have little effect on the voltage-gated K⁺ current in Helix neurones that are voltage-clamped with KCl-filled electrodes in normal saline (Meech & Standen, 1975; M. P. Mahaut-Smith, unpublished observations).

The rate of activation of the time-dependent current was estimated from the time to half the peak current (see Fig. 1C). Zn²⁺ considerably slowed the activation of the outward current. Note that t_1/2 could not be measured in 1 mmol l⁻¹ Zn²⁺ and at the lower potentials used in 0·1 mmol l⁻¹ Zn²⁺ because no significant current was activated.

Fig. 2 shows the effect of 0·1 mmol l⁻¹ Zn²⁺ on the outward current in response to depolarizations to +45mV in two cells. Cell A was exposed to 0·1mmol l⁻¹ Zn²⁺ for 10 min, and a depolarizing step was applied every minute, whereas cell B was exposed to 0·1 mmol l⁻¹ Zn²⁺ for only 1 min, with pulses every 15 s. (The lack of effect after 15 s may have been partly a problem of dead space and mixing of solutions.) Zinc acts rapidly and the initial block does not appear to be usedependent, an effect which is observed in the block of channels by local anaesthetics (Wit & Cranefield, 1974). In cell A, a steady-state response was reached after about 9 min. The full effect of 0·1 mmol l⁻¹ Zn²⁺ is therefore shown by the current trace after 10 min in Fig. 2.

Ca²⁺ currents, relatively free from outward current, were recorded from cells in 0Na⁺, TEA⁺ saline by depolarizing voltage pulses from a holding potential of −55 mV to potentials in the range −10 to +5 mV. One concern in these experiments is that pHi will steadily fall owing to lack of pH regulation in the Na⁺-free salines. The acidosis will directly block Ca²⁺ currents (Byerly et al. 1984; Byerly & Moody, 1986) and contaminate the inward current owing to an increase in outward H⁺ current (Byerly & Moody, 1986; Mahaut-Smith, 1989). This effect was observed, but was variable; therefore, only cells in which there was a negligible decay in inward current following exposure to 0Na⁺, TEA⁺ saline were used to study Zn²⁺ block of Ca²⁺ current. Plant & Standen (1981) may have avoided this problem in their study of Ca²⁺ currents in 0 Na⁺, TEA⁺ medium because 10 mmol I^-1 4-aminopyridine also was present. 4-aminopyridine blocks H⁺ currents directly (Byerly et al. 1984), but can also inhibit H⁺ currents indirectly (Meech & Thomas, 1987) because it is a weak base which alkalinizes the cytoplasm and increases its buffering power (Szatkowski & Thomas, 1987).

Fig. 3A shows the effect of external Zn²⁺ on the Ca²⁺ current recorded at 5 mV and in Fig. 3B the peak Ca²⁺ currents, corrected for leak current (see above for method), are plotted as a function of the membrane potential in the different levels of Zn²⁺. Zn²⁺ blocked Ca²⁺ currents in Helix, but higher concentrations were required compared to the block of H⁺ currents. 0·1 mmol l⁻¹ Zn²⁺ caused only a small reduction in Ca²⁺ current; 1 mmol l⁻¹ reduced the peak inward current and the rate and extent of its decline. This is similar to the effect of low concentrations of Cd²⁺ on calcium currents in Helix and Aplysia (Plant & Standen, 1981; Chad et al. 1984). Recovery of Ca²⁺ currents following exposure to Zn²⁺ was incomplete; for example 1 mmol l⁻¹ ZnCl₂ reduced the peak Ca²⁺ current to an average of 40% of the control; the current then recovered to approximately 60% after a 10 min wash.

To compare fully the Zn²⁺ block of H⁺ and Ca²⁺ currents, a wide range of concentrations was tested. Increasing concentrations of the blocker were added to the bath, with a short period (30–45 s) of washing between doses. Fig. 4 is a plot of the peak Ca²⁺ and H⁺ current as a function of the Zn²⁺ concentration (Zn²⁺ dose on a logarithmic scale). I_max is the peak current measured in the absence of any blocker at the start of the experiment.

As discussed above, recordings of Ca²⁺ currents in a few cells in 0 Na⁺, TEA⁺ salines slowly shifted as though an outward current were developing. This effect is shown in Fig. 5A. The lower trace was recorded 4 min after exposure to the 0 Na⁺, TEA⁺ saline and the upper trace 7 min later. The difference appears to be a reduction in Ca²⁺ current (the leakage conductance was not affected). pHi often falls in neurones exposed to Na⁺-free salines (Mahaut-Smith, 1989); therefore, the apparent effect on the inward current could be due to direct inhibition of Ca²⁺ currents by H⁺ and/or an increase in outward H⁺ currents. The lower sensitivity of Ca²⁺ currents to external Zn²⁺ and the slow recovery of H⁺ currents from Zn²⁺ block, allows the latter possibility to be assessed. Fig. 5B shows the currents at −5mV before and shortly after (3 min wash) application of 0·1 mmol l⁻¹ ZnCl₂. Clearly, although there may have been some reduction in Ca²⁺ current by acidosis, the predominant effect was caused by an increased H⁺ current, which was blocked and was slow to recover from 0·1 mmol l⁻¹ Zn²⁺.

These experiments show that low concentrations of external Zn²⁺ blocked H⁺ currents in intact snail neurones. Zn²⁺ also reduced Ca²⁺ currents, but higher concentrations were required for an equivalent block. Because divalent cations have little effect on I_K(V) current in Helix, Zn²⁺ provides the most selective method of inhibiting H⁺ current, and is also a useful tool for investigating the physiological role of the H⁺ pathway in molluscan neurones. Zn²⁺ markedly slowed H⁺ current activation, which makes the block of the H⁺ pathway even more effective during short depolarizations such as action potentials. Theoretically, the rapid activation of H⁺ currents can cause a proton efflux within individual action potentials, which will be enhanced during bursts of activity (Byerly et al. 1984; Mahaut-Smith, 1989). If this pathway contributes significantly to pH regulation and/or to action potential repolarization, H⁺ current block will acidify the cytoplasm and/or lengthen the duration of the action potential.

This study also demonstrates (see also Byerly & Moody, 1986; Mahaut-Smith, 1989) that H⁺ currents can significantly contaminate inward Ca²⁺ current measurements. The effect is more important when pHi regulation is impaired, since acidification of the cytoplasm will increase the H⁺ current. Zn²⁺ can be used to distinguish between H⁺ and Ca²⁺ currents because they have very different K_d values and recovery times. The experiment of Fig. 5 showed that washing with a low concentration of Zn²⁺ blocked H⁺ current and revealed underlying Ca²⁺ current.

Recently, Thomas (1988a) has measured pH changes at the extracellular surface of snail neurones apparently caused by efflux of H⁺ through the H⁺ pathway. Extracellularly applied Zn²⁺ was, not surprisingly, the most effective blocker of the surface acidification induced by membrane depolarization. 1 mmol l⁻¹ Zn²⁺ completely blocked the surface acidification and the block was slow to reverse (Thomas, 1988b), which is very similar to the effect of this concentration of Zn²⁺ on the isolated H⁺ currents in the present study.

The mathematical fit for Zn²⁺ block of H⁺ current supports the model that each zinc ion binds to one receptor site. Several agents block other ionic currents according to this monomolecular reaction, where the ‘receptor’ is hypothesized to be part of the channel gating mechanism or the channel itself (see Hille, 1984; for review). I have avoided the term ‘channel’ because, at present, there is no direct evidence that voltage-dependent H⁺ movement occurs through classic ion channels. The most direct evidence for discrete H⁺ channels would come from single-channel patch-clamp experiments (Hamill et al. 1981). Byerly et al. (1984) describe the H⁺ current in Lymnaea as a channel, presumably because H⁺ current activation has many properties common to voltage-gated ionic channels. The binding model presented in this paper can only be considered as further indirect evidence that the voltage-dependent H⁺ current/pathway is conducted through ion channels. The rapid effect of Zn²⁺ on the H⁺ current suggests that the receptor for Zn²⁺ is situated on the extracellular membrane surface. In addition, the lack of voltage-dependence of the block supports a surface action for Zn²⁺ and not one in the voltage field.

Zn²⁺ has also been shown to block the delayed rectifier K⁺ conductance of frog skeletal muscle (Stanfield, 1975). The major effect of Zn²⁺ was to slow the activation of the K⁺ current by reacting with the gating particle; at 0·1 mmol l⁻¹ Zn²⁺, the rate constants were slowed by approximately 10-fold (Stanfield, 1975). This concentration of Zn²⁺ has a similar effect on H⁺ current activation, but is more effective at reducing the steady-state H⁺ conductance than the K⁺ current.

The large difference in sensitivity to Zn²⁺ suggests that voltage-dependent Ca²⁺ and H⁺ movements across the nerve membrane are conducted through different channels or pathways. The same pathway could carry both ions if the block is voltage-dependent and less effective in the range in which Ca²⁺ currents were activated. However, this seems unlikely as there was little difference in the block of the Ca²⁺ current activated by larger depolarizations (up to 35 mV), tested in cells with small outward currents (see, for example, Fig. 3B).

In conclusion, Zn²⁺ is the most selective inhibitor of H⁺ currents known and provides a method of testing for the presence and function of the H⁺ pathway.

I am grateful to Dr R. W. Meech for his guidance during these studies. Thanks also to Dr R. W. Meech and Dr M. G. Evans for comments on the manuscript and Michael Rickard for expert technical assistance. The work was supported by the Medical Research Council.