An acid-sensing ion channel from shark (Squalus acanthias) mediates transient and sustained responses to protons

Abstract

Acid-sensing ion channels (ASICs) are proton-gated Na(+) channels. They are implicated in synaptic transmission, detection of painful acidosis, and possibly sour taste. The typical ASIC current is a transient, completely desensitizing current that can be blocked by the diuretic amiloride. ASICs are present in chordates but are absent in other animals. They have been cloned from urochordates, jawless vertebrates, cartilaginous shark and bony fish, from chicken and different mammals. Strikingly, all ASICs that have so far been characterized from urochordates, jawless vertebrates and shark are not gated by protons, suggesting that proton gating evolved relatively late in bony fish and that primitive ASICs had a different and unknown gating mechanism. Recently, amino acids that are crucial for the proton gating of rat ASIC1a have been identified. These residues are completely conserved in shark ASIC1b (sASIC1b), prompting us to re-evaluate the proton sensitivity of sASIC1b. Here we show that, contrary to previous findings, sASIC1b is indeed gated by protons with half-maximal activation at pH 6.0. sASIC1b desensitizes quickly but incompletely, efficiently encoding transient as well as sustained proton signals. Our results show that the conservation of the amino acids crucial for proton gating can predict proton sensitivity of an ASIC and increase our understanding of the evolution of ASICs.

Figure 1. Shark ASIC1b is H⁺ sensitive

Top, representative traces of sASIC1b currents at pH 6.4 and pH 5.0. Note the sustained current at pH 6.4 and the two current components at pH 5. The current rise phase and the initial desensitization phase are also shown on an expanded time scale. Bottom, representative current trace of an uninjected oocyte. No currents are elicited by pH 5.0.

Figure 2. Characterization of the sustained sASIC1b current

A, top, representative current traces of sASIC1b that was repeatedly activated by application of either pH 6.4 or 5 for 3 s. Channels were allowed to recover in conditioning pH 7.4 for 30 s. Bottom, current amplitudes were normalized to the first amplitude. The initial amplitude of the slow current component at pH 5 decreased progressively. Absolute values of the initial amplitudes were 4.1 ± 0.5 μA (transient current at pH 6.4; n= 7), 0.3 ± 0.05 μA (sustained current at pH 6.4; n= 7), 5.8 ± 1.8 μA (transient current at pH 5; n= 6), and 1.7 ± 0.4 μA (slow current at pH 5; n= 6), respectively. B, desensitization of the sustained current at pH 6.4 by application of pH 5.0. Channels were alternately activated by pH 6.4 and pH 5.0. The amplitude of the sustained current (magnified in the insets) successively decreased after application of pH 5.0. C, current–voltage relationship for the transient and the sustained current at pH 5.0 and 6.4, respectively. For the transient currents, channels had been repeatedly activated at different holding potentials; for the sustained and slow currents, channels had been activated with pH 6.4 or 5.0, respectively, and voltage steps from −70 to +70 mV of 1 s duration were applied. Voltage steps at pH 5.0 were applied 60 s after activation when the slow current had relaxed to a constant amplitude. Absolute values of the current amplitudes at −70 mV were 19.4 ± 4.5 μA (transient current at pH 5.0; n= 6), 0.78 ± 0.12 μA (transient current at pH 6.4; n= 12), 0.33 ± 0.07 μA (sustained current at pH 5.0; n= 9–11 for voltage jumps between −70 mV and +30 mV; n= 3–5 for voltage jumps at +50 mV and +70 mV) and 0.44 ± 0.09 μA (sustained current at pH 6.4; n= 7), respectively.

Figure 3. Apparent H⁺ affinity of shark ASIC1b

A, representative current trace of oocytes expressing sASIC1b. Channels were activated for 3 s by varying low pH, as indicated. Conditioning pH 7.4 was applied for 30 s. B, channels were activated by pH 5.0 with varying pre-conditioning pH, as indicated. Conditioning pH was applied for 60 s. C, pH–response curves for activation (open circles) and steady-state desensitization (grey circles); lines represent fits to the Hill function. Dotted lines indicate EC₅₀ values. Only the transient current was analysed. The overlapping region of the activation and inactivation curves is magnified (inset). Absolute values of the current amplitudes were 4.9 ± 1.2 μA (activation curve, pH 5.0; n= 15) and 8.4 ± 1.9 μA (steady-state desensitization curve, conditioning pH 7.4; n= 11), respectively.

Figure 4. Shark ASIC1b is amiloride-sensitive

A, top, representative traces of sASIC1b currents in the presence of increasing concentrations of amiloride, as indicated. sASIC1b was activated with pH 5.0. Bottom, concentration–response curve for amiloride; the line represents a fit to the Hill function. Dotted lines indicate the EC₅₀ value. Absolute value of the current amplitude without amiloride was 4.6 ± 0.7 μA (n= 21). B, the sustained current was almost completely blocked by 1 mm amiloride (grey bar).

Figure 5. Shark ASIC1b is slightly modulated by psalmotoxin 1

A, pH–response curves for activation (squares) and steady-state desensitization (circles) with (filled symbols) and without (open symbols) pre-application of 100 nm psalmotoxin (PcTx); PcTx was present only in the conditioning period (60 s). For activation curves, channels had been activated for 3 s by varying low pH, as indicated. For steady-state desensitization curves, channels had been activated for 3 s by pH 5.0 with varying pre-conditioning pH, as indicated. Lines represent fits to the Hill function. Absolute values of the current amplitudes were 8.4 ± 2.6 μA (activation curve, pH 5.0, without PcTx; n= 6), 8.3 ± 1.8 μA (activation curve, pH 5.0, with PcTx; n= 6), 8.9 ± 2.7 μA (steady-state desensitization curve, conditioning pH 7.4, without PcTx; n= 6) and 4.5 ± 1.4 μA (steady-state desensitization curve, conditioning pH 7.4, with PcTx; n= 6), respectively. B, bar graphs comparing normalized current amplitudes at slight acidification for the data from A. Open bars, without PcTx1; filled bars, with PcTx1. For conditioning pH 6.9, significantly more channels were desensitized when PcTx was present; similarly, for activation by pH 6.8–6.4 current amplitudes were significantly larger when PcTx was present. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 6. A pair of histidines is indispensable for H⁺ sensitivity of shark ASIC1b

A, top, schematic illustration of the topology of sASIC1b. TM1, TM2: transmembrane domains. The arrow indicates the position of the N-terminal truncation in construct M27; the two conserved histidines localize to the proximal ectodomain. Bottom, representative current traces for sASIC1b-H101/102N, -M27, and -M27-H74/75N. Note that for M27-H74/75N, application of H⁺ slightly reduced the background current. B, bars representing the peak current amplitude (mean ±s.e.m.) of oocytes expressing wild-type sASIC1b (wt), the histidine mutant (H101/102N), and the two M27-mutants (n≥ 6); channels had been activated by pH 5.0. The amounts of cRNA that had been injected into each oocyte were 0.8 ng (wt and M27) or 8 ng (H101/102N and M27-H74/75N), respectively. ***P≪ 0.01. C, bars representing surface expression of sASIC1b and -M27; untagged sASIC1b served as a control (left bar). Results are expressed as relative light units (RLUs) per oocyte per second (n= 36). ***P≪ 0.01.

Figure 7. Small pH steps evoke sustained shark ASIC1b currents

A, pH was stepped from 7.4 to 6.2 in steps of 0.2 units (first step: 0.4 units; top). A representative current trace is shown (bottom). B, current versus pH relationship of sustained currents (filled circles; n= 12) measured as in A and the predicted window current (smooth curve). The window current was calculated by multiplying values at each pH of the activation and steady-state desensitization curve fits from Fig. 3C; the fit for the activation curve was refined for low H⁺ concentrations by measuring transient currents also at pH 6.95 (I= 0 μA; n= 15).

Figure 8. Phylogenetic tree illustrating the main branches of chordates

Individual ASICs are shown on the right; proton-sensitive ASICs in green, presumably proton-insensitive ASICs in red; sASIC1b is shown on a grey background. Genera from which ASICs have been cloned are also indicated. An estimate of the time of some branching events is given (Kumar & Hedges, 1998).