Zinc Modulation of Hemi-Gap-Junction Channel Currents in Retinal Horizontal Cells

Abstract

Hemi-gap-junction (HGJ) channels of retinal horizontal cells (HCs) function as transmembrane ion channels that are modulated by voltage and calcium. As an endogenous retinal neuromodulator, zinc, which is coreleased with glutamate at photoreceptor synapses, plays an important role in shaping visual signals by acting on postsynaptic HCs in vivo. To understand more fully the regulation and function of HC HGJ channels, we examined the effect of Zn²⁺ on HGJ channel currents in bass retinal HCs. Hemichannel currents elicited by depolarization in Ca²⁺-free medium and in 1 mM Ca²⁺ medium were significantly inhibited by extracellular Zn²⁺. The inhibition by Zn²⁺ of hemichannel currents was dose dependent with a half-maximum inhibitory concentration of 37 μM. Compared with other divalent cations, Zn²⁺ exhibited higher inhibitory potency, with the order being Zn²⁺ > Cd²⁺ ≈ Co²⁺ > Ca²⁺ > Ba²⁺ > Mg²⁺. Zn²⁺ and Ca²⁺ were found to modulate HGJ channels independently in additivity experiments. Modification of histidine residues with N-bromosuccinimide suppressed the inhibitory action of Zn²⁺, whereas modification of cysteine residues had no significant effect on Zn²⁺ inhibition. Taken together, these results suggest that zinc acts on HGJ channels in a calcium-independent way and that histidine residues on the extracellular domain of HGJ channels mediate the inhibitory action of zinc.

INTRODUCTION

Horizontal cells (HCs) are second-order neurons in the vertebrate retina that receive excitatory synaptic input from, and provide inhibitory feedback to, photoreceptors (Baylor et al. 1971; Naka 1972; Wu 1992). HCs are extensively coupled via cell-to-cell gap junction channels (Baldridge et al. 1987; McMahon et al. 1989; Witkovsky et al. 1983; Zhang and McMahon 2000) and also express unapposed hemi-gap-junction (HGJ) channels in their nonjunctional membranes (DeVries and Schwartz 1992; Goodenough and Paul 2003; Malchow et al. 1993; Saez et al. 2005; Zhang and McMahon 2001). HC HGJ channels function as transmembrane ion channels (DeVries and Schwartz 1992; Malchow et al. 1993; Zhang and McMahon 2001) and have been proposed to participate in ephaptic feedback from HCs to cones (Kamermans and Fahrenfort 2004; Kamermans et al. 2001). Previous studies have demonstrated that HGJ channels are modulated by transmembrane voltage, extracellular Ca²⁺, and light-adapting retinal neurotransmitters, such as dopamine, nitric oxide, and retinoic acid (DeVries and Schwartz 1992; Retamal et al. 2006; Ripps et al. 2002; Zhang and McMahon 2000, 2001; Zhang et al. 2008). In this study, we have examined whether zinc modulates HGJ channels of HCs.

Zinc, a trace metal highly concentrated in the retina, plays a pivotal role in maintaining normal visual function (Grahn et al. 2001; Ugarte and Osborne 2001), whereas zinc deficiency results in a variety of ocular diseases (Huber and Gershoff 1975; Morrison et al. 1978). Zinc is concentrated in photoreceptor cell bodies and synaptic terminals (Lee et al. 2008; Wu et al. 1993) and is released from photoreceptors onto HCs and bipolar cells (Akagi et al. 2001; Redenti et al. 2007; Ugarte and Osborne 2001; Zhang et al. 2002). Convergent evidence indicates that zinc functions as a neuromodulator of synaptic transmission in the retina and in the rest of the brain by regulating a variety of voltage-gated and ligand-gated channels in postsynaptic neurons (Akagi et al. 2001; Dong and Werblin 1996; Han and Wu 1999; Huang 1997; Wu et al. 1993; Zhang et al. 2002, 2006). In the retina, zinc has been shown to modulate γ-aminobutyric acid (GABA) receptors, glutamate receptors, and K⁺ channels on HCs, glycine receptors on ganglion cells, and GABA receptors on bipolar cells (Dong and Werblin 1996; Han and Wu 1999; Qian et al. 1997; Wu et al. 1993; Zhang et al. 2002, 2006).

In this study, we investigated the effects of extracellular Zn²⁺ on HGJ channels in isolated retinal HCs using whole cell patch-clamp techniques. Our results demonstrate that micromolar Zn²⁺ partially suppresses HC HGJ channel currents, that HGJ channels are modulated by zinc at physiological concentrations of extracellular calcium, that zinc modulates HGJ channels in a Ca²⁺-independent way, and that the likely binding sites of Zn²⁺ are to extracellular histidine residues.

METHODS

Cell culture

Dark-adapted adult hybrid striped bass (Roccus chrysops × Roccus saxitalis) were killed in accordance with National Institutes of Health guidelines for animal use. Retinas were removed under dim red light and then incubated in L-15 medium (GIBCO) containing 20 U/ml papain (Worthington) activated with 0.3 mg/ml cysteine. Penicillin:streptomycin (1:100, Sigma) was routinely added to the medium to reduce contamination. The retinas were incubated in L-15/papain solution for 30 min, followed by six changes of fresh L-15 medium, and then dissociated by repeated passage through a serological pipette. Isolated cells were plated onto plastic 35-mm dishes containing fresh L-15 medium. Cultures were maintained at 20°C and cells were used following 1–4 days in culture.

Solutions and chemicals

The normal Ca²⁺ extracellular solution contained (in mM): 137 NaCl, 2.5 KCl, 2.5 MgCl₂, 2.5 CaCl₂, 10 HEPES, 10 glucose, and 1 mg/ml bovine serum albumin (BSA; Sigma, Fraction VII) (pH was adjusted to 7.4 using NaOH). The Ca²⁺-free extracellular solution contained (in mM): 114.5 NaCl, 30 CsCl, 2.5 KCl, 1 MgCl₂, 10 HEPES, 1 Na-pyruvate, 10 glucose, and 1 mg/ml BSA (pH was adjusted to 7.4 with NaOH). To block potassium channels, K⁺ in the normal pipette solution was replaced by Cs⁺ and tetraethylammonium chloride (TEA) was added. The pipette solution contained (in mM): 124 CsCl, 1 CaCl₂, 11 EGTA, 10 HEPES, 1 Mg-ATP, 0.1 Na-GTP, and 10 TEA (pH to 7.4 with CsOH). Solution changes during recording were performed with a perfusion system and were completed within 30 s.

The histidine-modifying reagent N-bromosuccinimide (N-BrSuc, MP Biomedicals) was dissolved in extracellular solution in 1 mM concentration. The membrane-impermeant cysteine modifier, methanethiosulfonate-ethyltrimethylammonium (MTSET, Anatrace), was prepared as a 10 mM stock in extracellular solution, set on ice for ≤1 h, then diluted to a 1 mM working concentration in extracellular solution at room temperature immediately before incubation in cell culture. The noncharged membrane-permeant cysteine modifier N-ethylmaleimide (NEM, Sigma–Aldrich) was prepared as a 100 mM stock in water and then diluted to a 1 mM working concentration in extracellular solution.

Patch-clamp recording

Recordings from solitary HCs were performed using conventional whole cell patch-clamp configuration. Patch pipettes were pulled from Corning 7052 glass (AM Systems) and fire-polished to 4–-6 MΩ resistance. The pipette series resistance and capacitance were compensated by 80%. The offset potential between the pipette and bath solutions was zeroed prior to seal formation. Voltage commands and data analysis were performed by pCLAMP 9.0 software.

Data analysis

Dose–inhibition curves were obtained by calculating the inhibition percentage of Zn²⁺ at different concentrations and were fitted with the Hill equation, I_mem = (I_max × Aⁿ)/(IC₅₀ + Aⁿ), where I_mem is the cell membrane current elicited by a given antagonist concentration (A), IC₅₀ is the antagonist concentration that elicits a half-maximal response, I_max is the measured maximal response, and n is the Hill coefficient; the normalized percentage inhibition of zinc was calculated by the equation: [(I_control − I_Zn)/I_control] × 100%. Here, the inhibition percentage was calculated by measuring the reduction of the current following application of Zn²⁺ (or other cations) and expressing this as a fraction of the peak current amplitude in the control medium without Zn²⁺ (or other cations).

The data are presented as means ± SE. P values stated were calculated using either the t-test or the paired t-test. Statistical analyses were performed with Sigmaplot 10.0.

RESULTS

Zinc inhibits HGJ channel currents

To examine the effect of Zn²⁺ on HGJ channel currents, isolated solitary HCs were recorded using whole cell voltage clamp, and outward HGJ channel currents were elicited by superfusion with Ca²⁺-free medium and cell membrane depolarization, while blocking K⁺ channels with Cs⁺, and TEA. As shown in Fig. 1A, at a membrane potential of +40 mV, outward HGJ channel current was increased by switching the extracellular solution from 2.5 mM Ca²⁺ medium to Ca²⁺-free medium. Application of 30 μM Zn²⁺ suppressed this hemichannel current from 3,040 to 1,860 pA. The action of 30 μM Zn²⁺ on HGJ channel currents was reversible on washout and the following application of 100 μM Zn²⁺ decreased the current further to 390 pA. The dose–inhibition curve for Zn²⁺ is shown in Fig. 1B and, when fit with the Hill equation, yielded a half-maximum inhibitory concentration (IC₅₀) of 37 μM. We also tested the effect of 5 μM Zn²⁺ on HGJ channel currents; no potentiation of currents was observed, suggesting that Zn²⁺ inhibits HGJ channel currents in a monophasic manner. In contrast, Ca²⁺, another divalent cation modulator for HGJ channel currents (Zhang and McMahon 2001), exhibited an IC₅₀ value of 245 μM, which is close to sevenfold higher than that of zinc (Fig. 1C).

FIG. 1.

Zinc inhibits hemi-gap-junction (HGJ) channel currents in bass horizontal cells (HCs). A: at a holding potential of +40 mV, outward hemichannel currents were induced by switching the extracellular solution from 2.5 mM Ca²⁺ medium to Ca²⁺-free medium. External application of 30 μM Zn²⁺ or 100 μM Zn²⁺ suppressed the currents to different degrees. B: normalized dose–inhibition relationships for Zn²⁺ in Ca²⁺-free medium. The reduction of currents by Zn²⁺ was normalized to the initial response obtained in Ca²⁺-free medium. C: normalized dose–inhibition relationships for Ca²⁺. The reduction of currents by Ca²⁺ was normalized to the initial response obtained in Ca²⁺-free medium. Each point is an average derived from 4 to 5 cells and dose–inhibition curves were fit with the Hill equation.

As shown in Fig. 1C, about 10% of the hemichannel current remained at 1 mM Ca²⁺, a concentration similar to that estimated in the intact retina (Dearry and Burnside 1984). To test whether Zn²⁺ inhibits HC hemichannel currents at physiological Ca²⁺ concentrations, we examined Zn²⁺ action on hemichannel currents in the presence of 1 mM Ca²⁺. Zn²⁺ (10 μM) significantly reduced the Ca²⁺-resistant hemichannel current to 19 ± 4% of control (P < 0.05, n = 5). These data indicate that Zn²⁺ can modulate HGJ channel currents at physiological Ca²⁺ concentrations.

Zinc is a potent modulator of HGJ channel currents

HGJ channels possess a high sensitivity to extracellular divalent cations, which is rarely observed on cell–cell gap junction channels (Spray et al. 2006; Srinivas et al. 2006). To compare the inhibitory potency of Zn²⁺ with other divalent cations, we added 100 μM of different cations to the Ca²⁺-free medium and determined their inhibition of HGJ channel currents (Fig. 2A). HGJ channel currents were induced under Ca²⁺-free medium by stepping membrane potential for 10 s from 0 to +40 mV. The inhibitory potency of these cations was in the order of: Zn²⁺ > Cd²⁺ ≈ Co²⁺ > Ca²⁺ > Ba²⁺ > Mg²⁺ (Fig. 2B).

FIG. 2.

Inhibitory potencies of different divalent cations on HGJ currents. A: representative current traces of the currents induced in control medium and in medium containing 100 μM Mg²⁺, Ba²⁺, Zn²⁺ (top traces), or 100 μM Ca²⁺, Co²⁺, Cd²⁺ (bottom traces). B: average inhibition by the different divalent cations. Asterisks indicate values significantly different from control, paired t-test: *P < 0.05 and ***P < 0.001 (n = 5).

Zinc modulation is independent of calcium

Ca²⁺ and retinoic acid interact in gating HGJ channels in bass retinal HCs (Zhang and McMahon 2001) and Co²⁺ and Ni²⁺ share the same binding site with Ca²⁺ on connexin hemichannels (Srinivas et al. 2006). Next, we examined the interaction between Zn²⁺ and Ca²⁺ on HGJ channel currents.

If Zn²⁺ and Ca²⁺ act on HGJ channels independently, their inhibitory effects would be expected to be additive when both are present at low concentrations (individual inhibition percentage <50%), provided that there are sufficient HGJ channel targets for both Zn²⁺ and Ca²⁺. However, if there is overlap in the binding sites of Zn²⁺ and Ca²⁺, one would expect a significantly lower total inhibition instead of simple addition in the presence of both cations. As shown in Fig. 3A, HGJ channel currents were inhibited by 24% (from 705 to 537 pA) by application of 20 μM Zn²⁺ alone, whereas application of 100 μM Ca²⁺ alone reduced the current by 40% (from 790 to 471 pA). The coapplication of 20 μM Zn²⁺ and 100 μM Ca²⁺ (Ca²⁺ added before Zn²⁺) produced a total reduction of 62% (from 790 to 490 pA), which is close to the sum of their separate inhibitory effects (64%). Statistical analyses (Fig. 3B) show that there is no significant difference between the inhibition by coapplication (58 ± 3%) versus the sum (60 ± 4%) of separate inhibitions (t-test, P > 0.05, n = 4). Similar results were also obtained when Zn²⁺ was added before Ca²⁺ during coapplication experiments (data not shown). These data suggested that Zn²⁺ and Ca²⁺ modulate HGJ channel currents independently, at least at low concentrations.

FIG. 3.

At low concentrations, Zn²⁺ and Ca²⁺ act on HGJ channels additively. A: in Ca²⁺-free medium, HGJ channel currents were evoked by stepping the holding potential from 0 to +40 mV; 20 μM Zn²⁺ alone produced a 23.8% inhibition from 705 to 537 pA; 100 μM Ca²⁺ followed by 20 μM Zn²⁺ were perfused onto the cell. B: average inhibition percentages resulting from the sum of individual application of Zn²⁺ (white bar) or Ca²⁺ (dotted bar) and resulting from the coapplication of Zn²⁺ and Ca²⁺ (solid bar). Data were collected from 4 cells.

At high concentrations of Zn²⁺ and Ca²⁺, especially those at which the sum of separate inhibition exerted by Zn²⁺ or Ca²⁺ alone would exceed 100%, it becomes impossible to examine the action mode between Zn²⁺ and Ca²⁺ by comparing the inhibition of coapplication with the sum of separate inhibitions. However, in this case, if Zn²⁺ and Ca²⁺ modulate HGJ channels independently, the fraction of inhibited channels by Zn²⁺ in Ca²⁺-free medium should remain the same as that in the presence of background Ca²⁺. To test this hypothesis, we examined inhibitions by 60 μM Zn²⁺ (78 ± 3% inhibition in Fig. 1B) on HGJ currents in the absence and in the presence of 100 μM Ca²⁺ (34 ± 2% inhibition in Fig. 1C). As shown in Fig. 4A, application of 60 μM Zn²⁺ reduced the HGJ channel current from 1,000 to 200 pA, resulting in an 80% inhibition in Ca²⁺-free medium. After washout with Ca²⁺-free medium, 100 μM background Ca²⁺ was applied and the current decreased to 640 pA. In the presence of the 100 μM Ca²⁺, additional application of 60 μM Zn²⁺ further reduced the currents from 640 to 110 pA, producing an 83% inhibition, which is similar to that observed in Ca²⁺-free medium. Again, statistical analyses showed no significant difference (Fig. 4B) between the fraction inhibited by 60 μM Zn²⁺ in the absence (78 ± 3%) and in the presence (80 ± 3%) of 100 μM Ca²⁺ (t-test, P > 0.05, n = 4). Similar results were obtained when 550 μM Ca²⁺ was examined using 20 μM Zn²⁺ as background. As shown in Fig. 4C, the inhibition of 550 μM Ca²⁺ without Zn²⁺ (65 ± 2%) remains the same as that in the presence of 20 μM Zn²⁺ (64 ± 3%, t-test, P > 0.05, n = 4). These data indicate that at high concentrations, Zn²⁺ and Ca²⁺ modulate HGJ channel currents independently.

FIG. 4.

At high concentrations, Zn²⁺ or Ca²⁺ inhibits a constant proportion of current in coapplication. A: in Ca²⁺-free medium, outward HGJ channel currents were evoked by stepping from 0 to +40 mV; 60 μM Zn²⁺, then 100 μM Ca²⁺, and additional application of 60 μM Zn²⁺ were perfused onto the cell. B: average inhibition percentages of 60 μM Zn²⁺ in Ca²⁺-free medium (left bar) and of 60 μM Zn²⁺ in the presence of 100 μM Ca²⁺ (right bar). C: average inhibition percentages of 550 μM Ca²⁺ in Zn²⁺-free medium (left bar) and of 550 μM Ca²⁺ in the presence of 30 μM Ca²⁺ (right bar). Data were collected from 4 cells.

In addition, we also recorded HGJ channel currents at varying concentrations of Zn²⁺ in the presence of 100 μM Ca²⁺. The dose–response curve of Zn²⁺ in the presence of 100 μM Ca²⁺ exhibited an IC₅₀ of 32 μM (data not shown), which is similar to the IC₅₀ of Zn²⁺ in Ca²⁺-free medium (37 μM), further supporting that Zn²⁺ and Ca²⁺ act on HGJ channels independently.

Zinc inhibits HGJ channel currents via histidine not cysteine residues

The action of Zn²⁺ on proteins is generally mediated by complexing zinc ions with histidine or cysteine residues (Connolly and Wafford 2004; Nevin et al. 2003; Norregaard et al. 1998; Seebungkert and Lynch 2001). To test the hypothesis that the inhibition of connexin hemichannels by Zn²⁺ is mediated by binding to histidine/cysteine residues, the inhibition exerted by 50 μM Zn²⁺ was compared in the presence or the absence of histidine or cysteine residue modifying agents. In control experiments, with no pretreatment, 50 μM Zn²⁺ resulted in a 78 ± 2% inhibition of HGJ channel currents (n = 5, Fig. 5). When the brominating agent N-BrSuc was used to modify the external histidine residues by preincubating HCs with 1 mM N-BrSuc for 30 min, the inhibitory effect exerted by 50 μM Zn²⁺ was significantly reduced to only 40 ± 2% inhibition (t-test, P < 0.001, n = 5, Fig. 5). We then tested whether cysteine residues are involved in mediating the action of Zn²⁺ on hemichannels. When HCs were preincubated with the fast-reacting, lipid-insoluble cysteine-modifier MTSET, no apparent effect was observed on the inhibition produced by 50 μM Zn²⁺ (75 ± 3% inhibition, n = 5, Fig. 5), indicating that MTSET-accessible extracellular cysteines are not involved in Zn²⁺ action. To further confirm the lack of cysteine involvement, a noncharged, membrane-permeant cysteine modifier, NEM, was tested and similar results were obtained (74 ± 3%, n = 5, Fig. 5).

FIG. 5.

Effects of histidine and cysteine modification on zinc inhibition. Histogram shows 50 μM zinc inhibition of HGJ channels with or without pretreating with histidine/cysteine modifying reagents. Asterisks indicate values significantly different from the control, t-test: ***P < 0.001 (n = 5).

DISCUSSION

We have characterized the effects of Zn²⁺ on HGJ channels of solitary retinal HCs and our principal findings are as follows: 1) Zn²⁺ inhibits HGJ channel currents at micromolar concentrations in a monophasic manner. 2) Zn²⁺ and Ca²⁺ modulate HGJ currents independently. 3) The inhibition of Zn²⁺ is mediated through histidine residues on the extracellular domain of hemichannels, whereas the contribution of cysteines is limited. Taken together, our data suggest that Zn²⁺ is a potent regulator of native HGJ channel currents.

Zinc acts monophasically and extracellularly

Previous studies have shown that extracellular Zn²⁺ modulates perch Cx35 and human Cx26 HGJ channels expressed on Xenopus oocytes in a biphasic pattern, with low micromolar concentrations of Zn²⁺ enhancing HGJ currents and high concentrations suppressing them (Chappell et al. 2003, 2004). This biphasic action of Zn²⁺ has also been reported for other Zn²⁺-binding proteins (Han and Wu 1999). In contrast, in this study of native HC hemichannels, we found that Zn²⁺ suppressed HGJ channel currents at both low and high micromolar concentrations. This monophasic inhibitory effect of Zn²⁺ indicates that the Zn²⁺ modulation of bass HGJ channels is different from the biphasic model, which has two metal ion-binding sites with different affinities. It is interesting that a similar monotonic Zn²⁺ inhibition of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)-type glutamate receptors has also been reported in bass retinal HCs (Zhang et al. 2002).

The presence of 1 mM external Ca²⁺ did not affect the Zn²⁺ inhibition of HGJ currents, indicating that Zn²⁺ acts on HGJ channels via specific binding interactions, not due to nonspecific cationic surface charge actions. We also found that Zn²⁺ suppressed HGJ currents at both positive and negative potentials. This voltage independence of Zn²⁺ modulation further suggests that the binding sites for Zn²⁺ are unlikely to be located within the pore of HGJ channels. Furthermore, we observed no significant effect of extracellular Zn²⁺ on cell-to-cell coupling of bass HCs (data not shown), indicating that the binding sites for Zn²⁺ are located on the extracellular surfaces exposed on unapposed HGJ channels. Taken together, these results suggest that the modulation of HGJ channels by Zn²⁺ occurs at specific binding sites on the extracellular surface, as has been suggested for Cx46 hemichannels (Verselis and Srinivas 2008). The distinct regulatory pattern and potency of Zn²⁺ observed on Cx35 and Cx26 HGJ channels heterologously expressed in oocytes (Chappell et al. 2003, 2004) may be due either to sequence differences in those connexin proteins compared with whatever bass HC connexin protein that forms the hemichannels we recorded or to differences in zinc regulation of native HC versus heterologously expressed non-HC connexins.

Zinc acts independently of calcium

Previous studies have shown that Ca²⁺ has interactions with retinoic acid (Zhang and McMahon 2001) and shares the same binding sites with Co²⁺ and Ni²⁺ in regulating HGJ channel currents (Srinivas et al. 2006). In this study, we found that at low concentrations of both Zn²⁺ and Ca²⁺, their actions were additive, whereas at high concentrations the fraction of current inhibited by one cation remained the same in the presence of the other. These results indicate that Zn²⁺ regulates HGJ channels by coordinating with binding sites that are distinct from those for Ca²⁺. Furthermore, the IC₅₀ of Zn²⁺ does not change significantly in the presence of Ca²⁺, suggesting that the affinity of HGJ channels for Zn²⁺ remains the same under those conditions. All these data suggest that Zn²⁺ regulates HGJ channel currents in a Ca²⁺-independent manner.

Zinc acts on histidine residues

Histidines and cysteines are the most common coordinating residues at Zn²⁺-binding sites (Connolly and Wafford 2004; Horenstein and Akabas 1998; Katz and Luong 1999; Nevin et al. 2003; Norregaard et al. 1998; Seebungkert and Lynch 2001). Although the specific sequences of the bass HC connexins mediating hemichannel currents have not been identified, connexin proteins possess conserved cysteine residues on the extracellular loops that stabilize the folding and cell–cell interactions of connexin proteins (Foote et al. 1998; Richard et al. 1998; Toyofuku et al. 1998). Histidine residues have also been found conserved in many different connexins (Beahm and Hall 2002; Stergiopoulos et al. 1999). Our data show that Zn²⁺ inhibition was significantly attenuated by pretreating cells with a histidine modifier, N-BrSuc, a common brominating agent used to modify histidines in different proteins (Mancini et al. 1992; Poe et al. 1979), indicating that the Zn²⁺ inhibition involves binding to one or more exposed histidine residues. To assess the contribution of cysteine residues to the Zn²⁺ inhibition, the effects of two cysteine modifiers, a membrane-impermeant agent MTSET and a membrane-permeant agent NEM, were also examined and neither exerted significant influence on Zn²⁺ modulation of HGJ currents. Overall, these results suggest that histidines are the major residues for coordinating Zn²⁺, whereas cysteine residues contribute little to this inhibition. Interestingly, conserved histidine residues on HGJ channels have been identified on the extracellular domains of connexin proteins (Beahm and Hall 2002). However, since the modification of histidine residues did not block the Zn²⁺ inhibition completely, our data do not exclude the possibility that other residues might also contribute to the coordination of Zn²⁺.

Physiological relevance of retinal zinc

Endogenous Zn²⁺ has been shown to be highly concentrated in the synaptic terminals of photoreceptors (Lee et al. 2008; Wu et al. 1993). Recently, both intracellular redistribution of Zn²⁺ and depolarization-induced Zn²⁺ release have been reported in isolated zebrafish photoreceptors (Redenti et al. 2007). Furthermore, Zn²⁺ has also been shown to modulate GABA_C and AMPA receptors in teleost retinal HCs, indicating that Zn²⁺ can potentially regulate visual signal processing within retinal circuits (Dong and Werblin 1996; Zhang et al. 2002). The experiments with isolated HCs reported here, which allow direct measurement of native hemichannel currents not possible in HCs in intact retinas, have established novel aspects of HC HGJ currents and their modulation by zinc and suggest the potential for these mechanisms to contribute to the regulation of HC function.

GRANTS

This work was supported by National Eye Institute Grant R01 EY-09256 to D. G. McMahon.