CrossTalk proposal: Proton permeation through HV1 requires transient protonation of a conserved aspartate in the S1 transmembrane helix

Here I present evidence in support of the premise that H+ selective permeation through the voltage gated proton channel HV1 involves obligatory protonation/deprotonation of an aspartic acid residue (Asp112 in human HV1) in the middle of the S1 transmembrane helix. An alternative view is the ‘frozen water’ mechanism, in which the channel traps one or more water molecules that block cation flux but permit Grotthuss-style proton hopping. I concede at the outset that the issue is not resolved and is not in fact resolvable based on current evidence. The historical basis for evaluating HV1 conduction mechanisms was a series of measurements of the biophysical properties of H+ currents. For over a decade the fundamental paradigm was to compare proton fluxes through HV1 with those through gramicidin (Table 1), known to be a cylindrical pore through which ions and water permeate in single file (Rosenberg & Finkelstein, 1978; Wallace & Ravikumar, 1988). The characteristics of H+ current through gramicidin differ from proton conduction in bulk water, but H+ conduction through HV1 is radically different from both (Table 1). Hence, many papers from this era concluded that the pathway through HV1 is more complex than a water-filled pore and includes obligatory protonation/deprotonation of at least one amino acid side-chain (DeCoursey & Cherny, 1994, 1997, 1998). This interpretation has been questioned, most explicitly by Ramsey and colleagues (2010), although others have espoused the view that H+ permeates HV1 via a water wire (Wood et al. 2012; Pupo et al. 2014). What follows are six relevant properties of HV1 currents, which to my knowledge are not in dispute, and their implications for the conduction mechanism. HV1 is extremely selective for H+, excluding all other ions. This extraordinary selectivity can be explained by invoking obligatory protonation/deprotonation of a titratable group during permeation (Nagle & Morowitz, 1978). Just such a group was identified as a perfectly conserved (among at least 140 species in the National Centre for Biotechnology Information database) Asp in the middle of the S1 transmembrane helix. Proton selectivity is compromised if this critical Asp is mutated to a neutral amino acid, converting the channel to anion selectivity (Musset et al. 2011). In a reduced quantum mechanical model of the selectivity filter of HV1, interaction of H3O+ with the Asp and Arg side-chains sufficed to produce H+ selective conduction, while excluding other ions (Dudev et al. 2015). Similarly, obligatory protonation/deprotonation of His37 imparts H+ selectivity to the M2 influenza A viral proton channel (Pinto et al. 1997). Despite its selectivity for H+ (DeCoursey, 2003), at non-physiologically high pH HV1 appears to conduct guanidinium ions, Gu+ (DeCoursey, 2013). The anomalous permeation of this foreign ion was attributed to its ability to denature proteins by breaking hydrogen bonds and disrupting water structure (DeCoursey, 2013). H+ selective permeation is proposed to occur when H3O+ protonates Asp, breaking its hydrogen bonds with Arg, leaving a neutral H2O molecule bridging the two side-chains (Dudev et al. 2015). Gu+ can break salt bridges between Glu and Arg or Lys by interacting specifically with carboxylate– groups (Meuzelaar et al. 2015); thus it permeates HV1 by violating the hydrogen bonds that occlude the pore to other ions. If the HV1 conduction pathway were a simple water wire, it is difficult to fathom why Gu+ would permeate but smaller or larger cations would not. The deuterium isotope effect on current amplitude is 1.9 (H+/D+) (DeCoursey & Cherny, 1997), much greater than that for H+ conduction in bulk water or in the water-filled pore of gramicidin (Table 1). The isotope effect for bulk H+ mobility is 1.4 (Lewis & Doody, 1933). Other molecules whose H+ transport pathways involve protonation/deprotonation of amino acids exhibit higher isotope effects (Blair & Berg, 1990; DeCoursey & Cherny, 1997). The isotope effect of 2 in the M2 proton channel (Mould et al. 2000) is consistent with His protonation/deprotonation kinetics. Because D+ binds more tightly than H+, the pKa of many proteins increases in heavy water; a carboxyl group binds D+ threefold more tightly than H+ (Schowen, 1977). The isotope effect is consistent with protonation/deprotonation of Asp112 during each conduction event. Both permeation and gating of HV1 have a higher Q10 (the factor by which a rate increases for a 10°C increase in temperature) than almost any other ion channel, 2–3 and 6–9, respectively (DeCoursey & Cherny, 1998; Kuno et al. 2009). The Q10 for permeation in other ion channels is typically 1.2–1.5 (DeCoursey & Cherny, 1998), like that for ion conductivity in bulk solution. This suggests either that diffusive entry into the pore of other channels is rate determining, or their permeation path mimics aqueous diffusion. The temperature dependence of H+ conductivity in bulk solution is lower than that of other ions, so the high Q10 for permeation through HV1 indicates an energetically difficult pathway. For example, rotation of a protonated side-chain could advance the proton across a narrow hydrophobic region. An imidazolium ring flip likely occurs during H+ conduction in the M2 viral proton channel, which also has a high Q10 (Lin & Schroeder, 2001; Hu et al. 2010). In contrast, H+ permeates the gramicidin water wire easily, with Q10 1.35 (Table 1). In addition, the apparent mobility of H+ inside gramicidin is not much lower than in bulk solution (Cukierman, 2000). Ion carriers and pumps exhibit high Q10 in the range of HV1 (Blair & Berg, 1990; DeCoursey & Cherny, 1998). A telling example is the Shaker voltage sensing domain with the Arg→His mutation R365H, in which the introduced His acts as a proton carrier, shuttling protons with a Q10 of 2.6 (Starace et al. 1997). In contrast with the water-filled pore of gramicidin (Decker & Levitt, 1988), diffusion of protonated buffer to the HV1 channel is not rate limiting because the current decreases only slightly when buffer concentration is lowered 100-fold, from 100 to 1 mm (DeCoursey & Cherny, 1996). At even lower buffer concentration, 150 μm, H+ current through M2 channels was decreased by 89% (Mould et al. 2000). H+ current through the gramicidin water wire is roughly proportional to [H+] over many orders of magnitude (DeCoursey, 2003). In contrast, the HV1 unitary conductance increases only 3.7-fold unit−1 as pHi is lowered (Cherny et al. 2003). The HV1 permeation mechanism must be capable of conducting at least 105 H+ s−1. This gives a mean transit time of ≤10 μs at pHi 5.5 at 120 mV above EH with a unitary conductance of 140 fS (Cherny et al. 2003). The gramicidin water-filled pore can conduct up to 2 × 109 H+ s−1 (Cukierman, 2000); showing that a water wire is kinetically competent, if not excessively so. Nagle and Morowitz estimated that a generic hydrogen bonded chain (HBC) involving amino acid side-chains might conduct 4 × 105 H+ s−1 (Nagle & Morowitz, 1978). The Shaker voltage sensing domain with R362H mutation acts as a hyperpolarization-activated proton channel, with the introduced His shuttling 5.6 × 104 H+ s−1 (Starace & Bezanilla, 2004). Carbonic anhydrase II shuttles protons via His64 and is the fastest known enzyme, with a turnover rate of 106 s−1 (Rowlett & Silverman, 1982). These examples show that obligatory proton shuttling by an amino acid side-chain (including protonation, side-chain excursion, and deprotonation) can occur at appropriate rates. One could measure water flux. Rapid water flux through HV1 would argue against the frozen water model, but could be compatible with ion exclusion as occurs in aquaporins. NMR could detect the protonation state and protonation kinetics of key amino acids. High-resolution crystal structures of open and closed HV1 channels might show the presence or absence of a fully-formed water wire capable of transferring H+. Then again they might not. Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief (250 word) comment. Comments may be submitted up to 6 weeks after publication of the article, at which point the discussion will close and the CrossTalk authors will be invited to submit a ‘Last Word’. Please email your comment, including a title and a declaration of interest, to jphysiol@physoc.org. Comments will be moderated and accepted comments will be published online only as ‘supporting information’ to the original debate articles once discussion has closed. Thomas E. DeCoursey's scientific path meandered through a childhood highlighted by summers spent in King's Canyon and Wind Cave National Parks, then to Cincinnati (mentor: Shirley H Bryant), Glasgow (Otto F Hutter), Irvine, California (Michael D Cahalan), and finally Rush University in Chicago where he is Professor of Physiology & Biophysics. Potassium and other channels served as ‘gateway’ channels leading to the discovery of voltage-gated proton channels in mammals in 1991. For the past quarter century his focus has been on all aspects of proton channels, from their critical roles in phagocytes and dinoflagellates to site-directed mutagenesis aimed at elucidating structure-function relationships. Disclaimer: Supporting information has been peer-reviewed but not copyedited. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. None declared. Support: NIH 1R01GM102336, NSF MCB-1242985.