Evans Blue Is a Specific Antagonist of the Human Epithelial Na⁺ Channel δ-Subunit

The degenerin/epithelial Na⁺ channel superfamily has striking functional diversity, including Na⁺ absorption, acid-sensing, peptide-gating, acidosis-evoked nociception, and mechanotransduction (Ugawa et al., 1998, 2002; Alvarez de la Rosa et al., 2000; Kellenberger and Schild, 2002; Welsh et al., 2002). The epithelial Na⁺ channel (ENaC) is an essential control element for Na⁺ transport pathway in cells and across epithelia. Four homologous ENaC subunits (α, β, γ, and δ) have been cloned in mammals, and the sequence identities between these subunits are ∼37% at the amino acid level (Canessa et al., 1993, 1994; McDonald et al., 1994, 1995; Waldmann et al., 1995). The α-subunit is expressed mainly in epithelia such as the kidney, lung, and colon and binds with β- and γ-subunits to play physiological roles in the control of Na⁺ balance, blood volume, and blood pressure (Alvarez de la Rosa et al., 2000; Kellenberger and Schild, 2002). On the other hand, the δ-subunit is widely distributed throughout the brain and is expressed in the heart, kidney, and pancreas (Waldmann et al., 1995; Yamamura et al., 2004a). Recently, we demonstrated that protons activate the δ-subunit, indicating that it may contribute to pH sensation in the human brain (Yamamura et al., 2004a).

In contrast to the relatively well documented α-subunit, the pharmacological profile of the δ-subunit has been poorly investigated. Amiloride, a potassium-sparing diuretic, has been described as a common blocker of the ENaC family (Canessa et al., 1993; McDonald et al., 1994; Waldmann et al., 1995; Ji et al., 2004). Recently, we reported that capsazepine is the first known chemical activator of the δ-subunit (Yamamura et al., 2004b). In addition to the δ-subunit-selective agonist, it is necessary to identify the specific inhibitor to explore the physiological and pathological functions of the δ-subunit of the ENaC.

In this investigation, the effects of Evans blue, which has been widely used as a diagnostic aid for the measurement of blood volume and vascular permeability (Rogers et al., 1989; Patterson et al., 1992), were examined on the human ENaC (hENaC) current using an electrophysiological technique in the Xenopus oocyte expression system. Here, we show that Evans blue inhibits the activity of the δ-subunit in a concentration-dependent manner, whereas the α-subunit current is not blocked but slightly increased by Evans blue. This result indicates that Evans blue is a δ-subunit-specific antagonist of ENaC.

Materials and Methods

Molecular Biology. All experiments were approved by the Ethics Committee of the Nagoya City University Graduate School of Medical Sciences and were conducted in accordance with the Declaration of Helsinki. Full-length hENaCα (GenBank accession no. X76180), hENaCβ (X87159), and hENaCγ (U48937) were isolated from human skin cDNA, and hENaCδ (U38254) was from human brain cDNA, as described previously (Yamamura et al., 2004b).

Electrophysiology. Electrophysiological studies using a two-electrode voltage-clamp technique were performed in Xenopus oocytes, as described previously (Yamamura et al., 2004b). In brief, cRNA transcript(s) (1 ng for the homomeric channel or each 0.01 ng for coexpression) was injected into Xenopus oocytes, whereas native oocytes were injected with an equal volume of nuclease-free water. After injection, oocytes were incubated at 20°C in a recording solution supplemented with 10 to 100 μM amiloride for 24 to 48 h. The recording solution had an ionic composition of 96 mM NaCl, 2 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, and 5 mM HEPES, pH 7.5. Oocytes were clamped at a holding potential of -60 mV, and the current-voltage relationship was measured using a ramp protocol from -100 to 50 mV for 15 s. All electrophysiological experiments were carried out at room temperature (25 ± 1°C).

Chemicals. Pharmacological reagents were obtained from Sigma-Aldrich (St. Louis, MO). Evans blue (6,6′-[(3,3′-dimethyl-[1,1′-biphenyl]-4,4′-diyl)bis(azo)]bis[4-amino-5-hydroxy-1,3-naphthalenedisulfonic acid] tetrasodium salt) and amiloride [3,5-diamino-N-(aminoiminomethyl)-6-chloropyrazinecarboxamide] were dissolved in dimethyl sulfoxide at the concentration of 100 mM as stock solutions. It was confirmed that up to 1% of dimethyl sulfoxide did not affect the oocyte currents.

Statistics. Pooled data are shown as the mean ± S.E.M. Statistical significance between two groups and among groups was determined by Student's t test and Scheffé's test after one-way analysis of variance, respectively. Significant difference is expressed in the figures (** or ##, p < 0.01). The data of the relationship between drug concentrations and current responses were fitted using the following equation after normalization: relative current = 1 - (1 - C)/{1 + (K/[A])ⁿ}, where C is the component resistant to the drug, K_d is the apparent dissociation constant of the drug, [A] is the drug concentration, and n is the Hill coefficient.

Results

Inhibition of δ-Subunit Current by Evans Blue. The effects of Evans blue on the currents of hENaCδ and the complexes with β- and γ-subunits were examined using a two-electrode voltage-clamp technique in the Xenopus oocyte expression system. When the hENaCδ homomer was expressed in Xenopus oocytes, the mean amplitudes of the 100 μM amiloride-sensitive inward current at a holding potential of -60 mV was 59 ± 4 nA (n = 6, p < 0.01 versus native of 2 ± 1 nA, n = 7; Fig. 1). In hENaCδ-expressing oocytes, the application of Evans blue at 100 or 300 μM blocked the inward current concentration dependently (n = 6; Fig. 1B). Alternatively, the mean amplitudes of the 100 μM amiloride-sensitive inward current at -60 mV was 592 ± 19 nA in hENaCδβγ-expressing oocytes (n = 12). A similar current decrease by Evans blue was observed in hENaCδβγ-injected oocytes (n = 12; Fig. 1C). The decrease in inward currents in response to Evans blue was gradually recovered to the resting level by the removal of Evans blue in all hENaCδ- and δβγ-expressing oocytes tested (n = 6 and 12, respectively). After washing for a few minutes, the readministration of Evans blue caused a similar current blockage to the first challenge in hENaCδβγ-injected oocytes (data not shown). On the contrary, in native oocytes, the application of 100 or 300 μM Evans blue did not induce any current (n = 7; Fig. 1A).

Inhibitory Effects of Evans Blue on δ-Subunit Current. The current-voltage relationship showed that the application of 300 μM Evans blue reduced the channel activity at all voltages examined in hENaCδβγ-expressed oocytes (n = 4; Fig. 2A). At the holding potential of -60 mV, the inward currents in hENaCδβγ-injected oocytes were significantly decreased by the addition of 300 μM Evans blue from 768 ± 12 to 337 ± 19 nA (n = 12, p < 0.01; Fig. 2B). In hENaCδ- and δβγ-expressing oocytes, these 300 μM Evans blue-sensitive currents at -60 mV were 41 ± 3 (n = 6) and 432 ± 23 (n = 12) nA, respectively (p < 0.01 versus native of 2 ± 1 nA, n = 7; Fig. 2C). It was reevaluated whether the inhibition of the inward current by Evans blue was mediated though either δ-subunit alone or the accessory β or γ subunit. These amplitudes of 300 μM Evans blue-sensitive currents were normalized by inward currents sensitized with 100 μM amiloride in homomeric hENaCδ- and heteromeric hENaCδβγ-expressed oocytes. The inhibitory effects of Evans blue on amiloride-sensitive current in hENaCδ-injected oocytes (67 ± 3%, n = 6) were indistinguishable from that in hENaCδβγ-expressing oocytes (71 ± 4%, n = 12, p > 0.05; Fig. 2D).

Dose-Dependence of Current Blockage by Evans Blue. The concentration-dependence of the Evans blue-sensitive current was analyzed in hENaCδβγ-expressed oocytes. Changing the concentration of Evans blue in a range from 1 to 1000 μM showed that the inward current was significantly decreased by Evans blue at a concentration of 100 μM and more (n = 4, p < 0.05 versus control of 774 ± 20 nA), and the current inhibition was in a concentration-dependent manner (Fig. 3). The IC₅₀ value of Evans blue on the inward currents was 143 μM, and the Hill coefficient was 1.7. To compare the binding affinity between Evans blue and amiloride, a common blocker of the ENaC family, the inhibitory effects of amiloride on the inward current through hENaCδβγ were examined. The IC₅₀ value of amiloride on the inward currents was 8 μM, and the Hill coefficient was 0.8 (Fig. 3B). There was no further detectable current decrease when 1 mM amiloride was added during the application of 1 mM Evans blue (Fig. 3A).

Specific Antagonism of δ-Subunit, Not α-Subunit, by Evans Blue. Since a classical ENaC inhibitor, amiloride, acts on both the δ-subunit and another ENaC core unit, the α-subunit, we tested whether Evans blue was effective on the α-subunit current in Xenopus oocytes. When hENaCαβγ was expressed in Xenopus oocytes, the inward currents at -60 mV were mostly blocked by a lower concentration of amiloride of 10 μM (Fig. 4A). The mean amplitude of the 10 μM amiloride-sensitive current was 646 ± 20 nA (n = 7) in hENaCαβγ-injected oocytes. Unexpectedly, in contrast to the δ-subunit, the hENaCαβγ current was slightly increased by Evans blue in a concentration-dependent fashion (by 164 ± 18 nA at 300 μM, n = 7, p < 0.01). The 300 μM Evans blue-induced current was mostly abolished by the addition of 10 μM amiloride (to 149 ± 9 nA, n = 7, p < 0.01). On the other hand, the application of 300 μM Evans blue had no effects on the amiloride-sensitive currents in hENaCα-expressing oocytes (data not shown).

Discussion

Ion channels in ENaC, members of the degenerin/ENaC superfamily, play pivotal control elements for Na⁺ homeostasis in cells and across epithelia. Since the ENaCαβγ complex contributes to the regulations of Na⁺ balance, blood volume, and blood pressure in epithelia such as the kidney, lung, and colon, the physiological and pharmacological properties have been well studied (Alvarez de la Rosa et al., 2000; Kellenberger and Schild, 2002). On the other hand, the physiological and pathological functions of the δ-subunit had not yet been identified. Recently, we have shown that the δ-subunit is widely distributed throughout the brain and is activated by protons, indicating that it may act as a pH sensor in the human brain (Yamamura et al., 2004a). Moreover, from a pharmacological aspect, we demonstrated that capsazepine, which was originally developed as a competitive antagonist for the vanilloid receptor (Bevan et al., 1992; Caterina et al., 1997), is a chemical agonist for the δ-subunit (Yamamura et al., 2004b). In this study, we found that the application of Evans blue inhibits the activity of the δ-subunit in a concentration-dependent manner. The most interesting finding is that Evans blue blocked the δ-subunit but not the α-subunit.

When the hENaCδβγ complex was expressed in Xenopus oocytes, the application of Evans blue at a concentration of 100 μM and more significantly reduced inward current. Because the reversible current decreased by Evans blue was not observed in native oocytes, Evans blue-sensitive currents were resulted from the ENaCδβγ expression. In addition to hENaCδβγ, the homomeric δ-subunit was also inhibited by Evans blue, indicating that it acts directly on the δ-subunit itself. Evans blue failed to decrease the current amplitude of another ENaC core unit, α-subunit with 37% amino acid identity to δ-subunit, and unexpectedly, Evans blue caused a slight increase in hENaCαβγ current. It has been reported that there are differences in Na⁺ permeability (α/δ = ∼1.6: 0.6 as P_Li/P_Na) and amiloride sensitivity (α/δ = 0.1:2.6 μM as IC₅₀ values) between the α- and δ-subunits (Kellenberger and Schild, 2002). Although the affinity of Evans blue with the δ-subunit was approximately 20-fold lower than that of amiloride in this study, Evans blue was a δ-subunit-specific inhibitor that distinguished α- and δ-subunits. The α-subunit is reported to be expressed in the human kidney and lung (McDonald et al., 1994, 1995), whereas the δ-subunit is expressed in the brain as well as non-neuronal tissues such as the heart, kidney, and pancreas in humans (Waldmann et al., 1995; Yamamura et al., 2004a). Tissue distribution implies that both the α- and δ-subunit proteins may coexpress to play physiological roles in the human kidney. It seems to be difficult for amiloride to separate the component of either the α- or δ-subunit from the amiloride-sensitive current in the human tissues because of the lower selectivity of amiloride. Our screening of the chemical agonists and antagonists for the δ-subunit revealed capsazepine and icilin as specific agonists as described previously (Yamamura et al., 2004b, 2005) and Evans blue as a specific antagonist in this study. Although the electrophysiological signals mediated by the δ-subunit are not detected in intact cells or tissues, these findings of chemical compounds strongly influencing the δ-subunit may lead to the elucidation of the physiological functions of the δ-subunit in vitro and in vivo in humans.

Evans blue, an azo dye, has been widely used as an indicator in the determination of blood volume and in studies of vascular permeability because of its strong affinity for albumin (Rogers et al., 1989; Patterson et al., 1992). Several recent studies have shown that this compound is able to modulate some receptors and ion channels at submillimolar concentrations that are necessary to examine microvascular or epithelial permeability. It has been described that Evans blue modulates the non-N-methyl-d-aspartate receptor in rat thalamic neurons (Leßmann et al., 1992), blocks the P₂X-purinergic receptor in rat vas deferens (Bultmann and Starke, 1993), and inhibits the glutamate transporter in rat brain synaptic vesicles (Roseth et al., 1995). Furthermore, recent evidence suggests that Evans blue activates large-conductance Ca²⁺-activated K⁺ channels in sheep bladder myocytes (Hollywood et al., 1998) and cultured endothelial cells of human umbilical veins (Wu et al., 1999). In addition to these actions on receptors and channels, in this investigation, we have clarified that Evans blue inhibits the δ-subunit of the ENaC.

In conclusion, we found that Evans blue acts selectively on the δ-subunit of the ENaC rather than the α-subunit and causes inhibition of the δ-subunit current, indicating that Evans blue is a specific antagonist for δ-subunit of the ENaC. In addition to the physiological function of δ-subunit as a pH sensor in the human brain, this finding in our study provides a starting point for a number of exciting follow-up investigations into the physiological and pathological roles of ENaCδ in humans.