Abstract

Citation:D'Avanzo N, Pekhletski R, Backx PH (2009) P-Loop Residues Critical for Selectivity in K⁺ Channels Fail to Confer Selectivity to Rabbit HCN4 Channels. PLoS ONE 4(11): e7712. https://doi.org/10.1371/journal.pone.0007712

Editor:Arnold Schwartz, University of Cincinnati, United States of America

Received:August 7, 2009;Accepted:October 9, 2009;Published: November 5, 2009

Copyright: © 2009 D'Avanzo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding:This study was supported by Heart and Stroke Foundation of Ontario (Grant No. T-4179) to PHB. PHB is a Career Investigator of the Heart and Stroke Foundation of Ontario. ND was the recipient of a Heart & Stroke Richard Lewar Center of Excellence Studentship at the University of Toronto (2004–2005), an Ontario Graduate Student Award (2005–2006), and a CIHR CGS Doctoral Award (2006–2007) during the progress of this work. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Hyperpolarization-activated cyclic-nucleotide gated (HCN) channels are expressed distributed in several excitable tissues including neurons and cardiomyocytes[1] where they contribute to pacemaker electrical activity[2],[3]. Four mammalian isoforms have been cloned (HCN1 to 4), each with different activation and deactivation kinetics as well as sensitivity of activation properties to cAMP[4]–[6]. Although, HCN channels share many sequence and structural similarities with voltage-gated K⁺ (K_v) channels, they possess unique selectivity, permeation, and gating properties. Specifically, in contrast to other voltage-gated channels, HCN channels activate slowly in response to membrane hyperpolarization despite similar voltage-sensor movement[7],[8]. HCN channels also have relatively high permeability to Na⁺ ions compared to K⁺ channels[9]–[11]. While the unique gating properties of HCN channels has been the focus of several recent studies[7],[8],[12], the molecular basis of differences in selectivity between HCN and K⁺ currently remains unknown.

One potential source of differences in selectivity between K⁺ channels and HCN channels is the selectivity sequence of the P-loop, as previously suggested[1],[13],[14]. Heteromeric K⁺ channels have a pore-lining sequence of TT(V/I)GYG (Figure 1A) which forms 4 equally spaced ion-binding sites[15],[16] that confer K⁺ selectivity to these channels[17]. By contrast, HCN channels have a pore-lining sequence of LCIGYG (Figure 1A) and have a 20–fold lower selectivity for K⁺ over Na⁺ ions compared to K⁺ channels[11],[18],[19]. Consistent with P-loop differences underlying the distinct ion selectivity between HCN and K⁺ channels, T75C mutantKcsA channels show marked reductions in K⁺ conductance and altered ion binding[20], while T441S and T442S/G mutantShaker channels have increased conductance of Rb⁺ and NH₄⁺ ions[21],[22].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Ion selectivity in WT and C479T HCN4 channels.

(A) Sequence alignment of mammalian HCN channels, and various K⁺ selective channels shows all mammalian HCN channels have a leucine and cysteine residue in place of the two threonines typically (though not always) present in the underlined K⁺ channel ‘selectivity sequence’. These residues (L478 and C479 in HCN4) were mutated to threonine to create a pore-lining sequence resembling that of a K⁺ selective channel.(B) Sample traces of WT, L478/C479T, and C479T HCN4 currents recorded in a 5 mM K⁺/135 mM Na⁺ bath solution, elicited by a 3 s prepulse to −130 mV from a holding potential of 0 mV, followed by a 1 s test pulse between +20 to −70 mV (ΔV = 10 mV).(C) Western blots performed from whole-cell lysates of untransfected (UT) CHO-K1 cells, or cells transfected with WT, L478/C479T or C479T constructs, with GAPDH used as a loading control. WT and mutant channels are highly expressed in bothN-glycosylated and core bands (←), indicating the absence of L478/C479T currents in expressed cells is not due to mis-trafficking or low protein production.

https://doi.org/10.1371/journal.pone.0007712.g001

Other regions of the P-loop sequence could also be responsible for selectivity differences between HCN and K⁺ channels. For example, two pore-helix Trp residues along with the Tyr side-chain of the GYG region stack and form a hydrophobic cuff around the selectivity filter inKcsA channels that confers pore rigidity and holds the pore open at its diameter[15]. Disruption of this aromatic sheet inShaker channels, via a W434F mutation, permits substantial Na⁺ current in the absence of K⁺ ions[23],[24]. Unlike K⁺ channels, HCN channels have Phe (F471) and Lys (K472) residues in place of Trp residues at the equivalent positions (Figure 1A). In addition, stabilization of the tertiary structure of the selectivity filter in inward rectifier K⁺ channels is also linked to salt bridge formed between E138 and R148 using Kir2.1 nomenclature[25]. Moreover, disruption of the equivalent “bow string” salt bridges of Kir3.1/Kir3.4 increases the pore diameter, increases pore flexibility, and reduces channel selectivity[26]. Interestingly, HCN channels also have an Arg residue (R484) on the extracellular side of the selectivity filter near the one found in Kir2.1 channels, however, have a Ser (S475) at the equivalent location for the negative counter ion (E138 of Kir2.1 channels) required to form a salt bridge. Thus, HCN channels may lack the molecular components required to stabilize pore structure and rigidity. Consistent with this conjecture, molecular models of spHCN1 and HCN2 channels show that HCN channels may have fewer stabilizing interactions between the selectivity filter and the P-loop α-helix than observed in K⁺ channels[27] suggesting greater flexibility within the HCN pore.

To identify a molecular basis for differences between ion selectivity in HCN channels versus K⁺ channels, we examined the role of P-loop residues known to be critical for selectivity in K⁺ channels but naturally substituted in HCN channels. To this end, we replaced the selectivity filter leucine and cysteine residues of HCN4 channels at positions 478 and 479 respectively with threonines. Additionally, strategic site-directed mutant HCN4 channels were generated (F471W/K472W, S475D, and S475E channels in a C479T background) in an attempt to introduce stabilizing interactions that may confer K⁺ selectivity to HCN channels. Our data shows C479T channels were less, not more, selective for K⁺ ions over other monovalent cations than WT HCN4, and S475D/C479T HCN4 channels did not improve ion selectivity. L478T/C479T, F471W/K472W/C479T and S475E/C479T HCN4 channels did not produce any measurable currents despite high protein expression. The permeability sequence pattern of HCN4 and C479T HCN4 channels to alkali metals and organic cations indicates that HCN4 channels have larger pores than typical K⁺ channels, which may contribute to their reduced ionic selectivity. The C479T mutation further increases the minimum pore diameter in HCN4 channels, consequently further reducing ionic selectivity. Thus, the presence of the K⁺ channel selectivity filter sequence, TIGYG, does not confer selectivity to HCN4 channels suggesting that HCN channels may be more structurally divergent from K⁺ than anticipated. The implications of these larger pores on HCN channel conductance and gating are also discussed below.

Materials and Methods

Results

Variations in the pore-lining sequences have been suggested to contribute to differences in ion selectivity between of K⁺ channels (TT(I/V)GYG) and HCN4 channels (LCIGYG) (Figure 1A)[1],[13],[14]. Therefore, we replaced L478 and C479 residues in the HCN4 channel pore with threonines. Unfortunately, currents were not observed when L478T/C479T HCN4 channels were expressed in CHO-K1 cells whether the external solution contained 5 mM K⁺ plus 135 mM Na⁺ (Figure 1B) or 140 mM K⁺ (data not shown). To determine whether L478T/C479T channel proteins were in fact expressed, Western blots were performed on whole-cell lysates. We observed high expression levels for both WT and L478T/C479T HCN4 channels (Figure 1C) withN-glycosylated and unglycosylated bands being clearly evident. These results suggest that the absence of current in L478T/C479T channels results from non-functioning protein in the plasma membrane, since glycosylation is required for HCN protein trafficking[29]. Alternatively, the absence of current in these channels may originate from large negative shifts in their voltage-dependence for activation. (i.e. to voltages below −130 mV).

Since the threonine in the selectivity sequence at position 2 is highly conserved (Figure 1A), and forms the fourth ion binding site in K⁺ channels[15],[16], we went on to examine the biophysical properties of C479T mutant HCN4 channels. To assess the selectivity properties of C479T channels, E_rev was determined from linear fits of leak subtracted instantaneous tail currents (Figure 2). Using the Goldman-Hodgkin-Katz equation (Eq. 1), we found that selectivity of ion “X” in WT HCN4 channels, quantified as P_X/P_K, was K⁺ (1)>Rb⁺ (0.48)>Cs⁺ (0.31)≥Na⁺ (0.29)>Li⁺ (0.03) (Table 1). This is consistent with a type IV or V Eisenman sequence[30],[31] indicating a mild anionic field strength in the pore of HCN4 channels. In C479T channels, the selectivity sequence was altered to K⁺ (1)>Rb⁺ (0.85)>Cs⁺ (0.59)>Li⁺ (0.50)≥Na⁺ (0.49) with P_X/P_K increasing (P<0.05) for all test ions (Table 1) particularly for Li⁺ which has the largest hydrated radius. This sequence of permeabilities is in line with a “Li anomaly” Eisenman sequence[31] still suggestive of type IV mild anionic field strength in the mutant pore. These data establish that the TIGYG pore sequence found in most selective K⁺ channels is insufficient to confer K⁺ selectivity to HCN4 channels.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. E_rev's determined from the instantaneous I–V relationship.

(A) Sample tail currents elicited following an activating “pre-pulse” to −130 mV as shown inFigure 1B for experiments performed in 5 mM K⁺ and 135 mM X⁺ (Li⁺, Rb⁺, Cs⁺) extracellular bath solution.(B) Representative instantaneous I–V relationships for recordings performed in 5 mM K⁺ and 135 mM X⁺ (Na⁺, Li⁺, Rb⁺, Cs⁺) bath solution WT (▪) and C479T (□) channels. E_rev's were determined from the linear fits of the I–V relationship for each cell, and were then used to calculate the P_X/P_K using the GHK equation and averaged as presented inTable 1.

https://doi.org/10.1371/journal.pone.0007712.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. E_rev and P_X/P_K estimates in WT and C479T HCN4 channels.

https://doi.org/10.1371/journal.pone.0007712.t001

In addition to selectivity filter sequence, HCN channels show residue differences in P-loop regions associated with pore rigidity and K⁺ ion selectivity. Specifically based on sequence alignments (Figure 1A), we attempted to restore stabilizing hydrophobic π−bond structures characterized inKcsA[15] andShaker channels[23],[24] by replacing F471 and K472 with tryptophans. These mutations were created on a C479T background, since Thr at this position in K⁺ channels plays a role in selectivity[20],[32] by forming the 4^th ion binding site[15],[16]. Unfortunately, F471W/K472W/C479T did not yield current in CHO-K1 cells when either 5 mM K⁺/135 mM Na⁺ (Figure 3A) or 140 mM K⁺ was used in the external solution. We also generated S475E/C479T, and S475D/C479T HCN4 channels, in an attempt to introduce a stabilizing salt bridge that has been previously characterized for inward rectifiers[25],[26]. Like F471W/K472W/C479T channels, S475E/C479T HCN4 channels did not generate currents under equivalent conditions (Figure 3A). However, both F471W/K472W/C479T and S475E/C479T HCN4 channels showed high protein expression levels, with both core andN-glycosylated bands. Thus, these mutant HCN4 channels traffic properly, despite producing no current, possibly as a result of large voltage shifts in the channel activation. By contrast, S475D/C479T HCN4 channels did generate currents whose permeability ratio for Na⁺ ions (P_Na/P_K) in S475D/C479T channels (0.51±0.04), did not differ from C479T channels (P_Na/P_K = 0.49±0.04; P = 0.80) (Figure 3B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. Na⁺ permeability in mutant HCN4 channels.

(A) Sample traces of C479T, S475D/C479T, S475E/C479T and F471W/K472W/C479T HCN4 currents (labeled S475D, S475E and WW for simplicity) recorded in a 5 mM K⁺/135 mM Na⁺ bath solution, elicited by a 3 s prepulse to −130 mV from a holding potential of 0 mV, followed by a 1 s test pulse between +20 to −70 mV (ΔV = 10 mV). S475E and WW channels did not express currents.(B) Representative instantaneous I–V relationships from tail currents of C479T (•) and S475D (◊) channels from recordings performed in a 5 mM K⁺/135 mM Na⁺ bath solution. P_X/P_K's were determined to be −17.0±1.9 (n = 4) and −16.2±2.0 (n = 5) for C479T and S475D channels respectively which were not statistically significant.(C) Western blots performed from whole-cell lysates of untransfected (UT) CHO-K1 cells, or cells transfected with C479T, S475D, S475E or WW constructs, with GAPDH used as a loading control (bottom of panel). All mutant channels are highly expressed in bothN-glycosylated and core bands (←), thus, the lack of current observed from S475E and WW channels is not due to mistrafficking, or low protein production, but rather non-functional protein at the plasma membrane.

https://doi.org/10.1371/journal.pone.0007712.g003

We measured the permeability ratios of organic cations of varying size (Figure 4A) which are poorly hydrated and therefore have been used to assess channel pore size. In WT channels, the sequence of P_X/P_K for organic cations was NH₄⁺ (0.27)>MA⁺ (0.13)>DMA⁺ (0.05)≥TMA⁺ (0.03)≥TEA⁺ (0.02). Although the sequence of organic cation selectivity was unchanged (NH₄⁺ (0.52)>MA⁺ (0.38)>DMA⁺ (0.29)>TMA⁺ (0.10)>TEA⁺ (0.01)), C479T channels showed large increases in permeability for the cations, MA⁺, DMA⁺ and TMA⁺, compared to WT channels (Table 1). Previous studies have used the “Excluded Field Theory (EFT)” to estimate the pore diameter of various multi-ion channels[28],[33]–[42] from permeation results using the crystal radii of organic cations[43]. The application of EFT to our results (Figure 4B) predicted that the minimum pore radius was larger for WT HCN4 channels (3.5±0.2 Å) than previously estimated for K⁺ channels (∼1.7 Å)[15],[16],[41],[42]. Interestingly, EFT predicted pore radii of 4.4±0.2 Å in C479T channels suggest that replacement of the relatively large cysteine residue of WT HCN4 with the smaller hydroxyl group of threonine enlarges the pore, thereby allowing increased permeability to Li⁺ and Na⁺ ions, as well as large organic cations. Moreover, the selectivity sequence of organic cations (NH₄⁺ (0.77)>MA⁺ (0.37)>DMA⁺ (0.25)>TMA⁺ (0.06)>TEA⁺ (0.04)) and the estimated minimum pore radius of S475D/C479T channels (4.1±0.3 Å) were unchanged compared to C479T channels (F = 0.23; P = 0.80).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 4. Pore size estimates for WT and C479T HCN4 channels.

(A) Sample tail currents elicited following an activating “pre-pulse” to −130 mV as shown inFigure 1B for experiments performed in 5 mM K⁺ and 135 mM X⁺ (NH₄⁺, MA⁺, DMA⁺ and TMA⁺) extracellular bath solution. Using “Exclusive Field Theory” we estimated the minimum pore diameter by plotting P_X/P_K of organic cations NH₄⁺, MA⁺, DMA⁺, TMA⁺, and TEA⁺ for WT (▪) and C479T (○) channels versus the ionic crystal radius according to Equation (2.15). Fits of these data gave minimum pore diameters of 6.9±0.4 Å (R = 0.93) and 8.7±0.3 Å (R = 0.99) for WT and C479T channels respectively (P<0.05).

https://doi.org/10.1371/journal.pone.0007712.g004

An increase in pore size for C479T channels compared to WT should be associated with an increased conductance, particularly if permeation and selectivity involves ion sieving[44], as suggested for HCN channels by the monotonic decrease in permeability with organic ion radius (Fig. 4B). To test this conjecture, unitary currents were measured using cell-attached single-channel recordings (Figure 5A) with 140 mM K⁺ in the pipette. Linear regression of the relationship between the unitary current, estimated from frequency histograms of current amplitudes as illustrated inFigure 5B at −150 mV, and the applied voltage allowed estimation of the unitary conductance (γ). For WT HCN4 channels, γ was 1.0±0.1 pS which is similar to previous estimates for native I_f channels in SA node myocytes and pyramidal neuron dendrites (∼1 and ∼0.7 pS)[45]–[47]. By contrast, γ for C479T channels was increased to 1.7±0.1 pS (P<0.05), as expected for channels consistent with an increase in pore diameter compared to WT.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 5. Unitary current (i) measurements in WT and C479T HCN4 Channels.

(A) Single-channel WT and C479T currents recorded at −120, −130, −140 and −150 mV in 140 mM extracellular K⁺. The solid line denotes zero current in the closed channel.(B) Sample histograms used to estimate unitary currents (i) of WT and C479T channels at −150 mV. Histograms were fit as a dual Gaussian distribution to obtain values fori.(C) Unitary-current-Voltage (i-V) relationship of WT (▪) and C479T () channels between −120 and −150 mV. The unitary conductance (γ), estimated from the slope of a line through the origin, was larger (P<0.05) in C479T channels (1.7±0.1 pS) compared to WT HCN4 (1.0±0.1 pS) (n≥3 for each voltage; *P<0.05).

https://doi.org/10.1371/journal.pone.0007712.g005

In addition to changes in permeation, we also routinely observed small, but reproducible differences in gating properties between WT and C479T channels. For example, the time constants of activation (τ_act) and deactivation (τ_deact) were smaller (P<0.05) in C479T channels than WT HCN4 channels with 140 mM external K⁺ (Figure 6B). These kinetic changes were also associated with altered steady-state activation properties in C479T channels compared to WT channels. Specifically, the slopes of the steady-state activation curves (k) were steeper (P<0.05) in WT channels compared to C479T channels in 140 mM K⁺ (k = 9.0±0.8 versus 12.9±0.6 respectively), with no differences (P = 0.69) in the voltage for 50% channel activation (V_1/2, = −108±4 mV for C479T versus −110±3 mV for WT) (Figure 6D). Furthermore, we also found that replacement of 140 mM external K⁺ with 5 mM K⁺/135 mM Na⁺ with (Figure 6B) accelerated (P<0.05) HCN4 channel activation, without affecting deactivation kinetics (Figure 6C). This replacement also shifted (P<0.05) to more depolarized potentials (V_1/2 = −96±2 mV) without altering the slope (k = 9.2±1.4) (Figure 6D). By contrast, and gating kinetics (Figure 6B and 6C) and steady-state activation properties (Figure 6D) of C479T channels were insensitive to replacement of extracellular K⁺ with Na⁺. This is consistent with previous reports indicating HCN channel activity is dependent on the permeant ion[19],[48],[49]. These effects of C479T replacement on channel gating may originate from a number of mechanisms including changes in pore structure and thereby ion binding, possibly leading to altered pore-gating in selectivity filter as reported previously in K⁺ channels[32],[50],[51].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 6. Gating properties of WT and C479T HCN4 channels.

(A) Representative WT and C479T activation currents elicited by the protocol in the inset. Traces shown were recorded with 5 mM K⁺/135 mM Na⁺ in the extracellular and 140 mM K⁺ intracellular solutions. Similar recordings were performed with 140 mM K⁺ in the extracellular solution.(B) Activation time constants (τ_act) measured using a mono-exponential fit of currents following the initial delay. C479T channels (circles) activated faster* than WT (squares) in 140 mM extracellular K⁺ (filled symbols) (n = 5 and 8 respectively). WT activation kinetics were faster* when most of the extracellular K⁺ was replaced with Na⁺ (5 mM K⁺/135 mM Na⁺; open symbols) (n = 9). C479T activated with the same kinetics following substitution of K⁺ with Na⁺ (n = 5).(C) Deactivation time constants (τ_deact) measured using a mono-exponential fit of currents similar to those inFigure 1 following the initial delay. C479T channels deactivated faster* than WT in both 140 mM K⁺ (n = 5 and n = 7 respectively) and 5 mM K⁺/135 mM Na⁺ (n = 7 and n = 6 respectively) extracelluar conditions.(D) Steady-state activation curves from peaks of tail currents elicited at +30 mV following pulses from −150 to −60 mV using a varying pulse duration protocol. WT (▪) channels activated with the same V_1/2 as C479T (•) channels in 140 mM K⁺ (V_1/2 = −110±3 mV and −108±4 mV respectively; P = 0.69), however, had steeper* slope factors (k = 9.0±0.8 and 12.9±0.6 respectively). Upon ion substitution to 5 mM K⁺/135 mM Na⁺, WT channels (□) activated at a +14 mV* more depolarized potential (V_1/2 = −96±2 mV) compared to 140 mM K⁺. C479T channels showed no shift in V_1/2 (−106±3 mV; P = 0.74) in 5 K⁺/135 Na⁺ () compared to those in 140 K⁺. The slope factor of activation,k, of WT channels is unchanged by ion replacement (k = 9.2±1.4 in 5 mM K⁺/135 mM Na⁺ respectively; P = 0.57 vs. 140 K⁺), however, remained steeper* than the slope factors in C479T channels (k = 11.8±0.8 in 5 K⁺/135 Na⁺; P = 0.34 vs. 140 mM K⁺). P<0.05 where indicated by *.

https://doi.org/10.1371/journal.pone.0007712.g006

Discussion

HCN channels show poor selectivity for K⁺ ion versus Na⁺ ions, despite many structural similarities with K⁺ channels. Since the exquisite selectivity of K⁺ channels for K⁺ has been shown to depend on ‘selectivity sequence’ TT(V/I)GYG as well as a number of other stabilizing interactions involving other P-loop residues it has been suggested that the selectivity properties of HCN channels arise from differences these P-loop regions[1],[13],[14]. Our results demonstrate that L478T/C479T mutant HCN4 channels, which restore the TTIGYG sequence to the pore, failed to produce measurable current despite high levels of ofN-glycosylated protein expression. Since several K⁺ channels contain leucine or alanine in the equivalent position of L478 (Figure 1A) we went on to examine C479T channels. However, C479T channels exhibited reduced selectivity for K⁺ over Rb⁺, Cs⁺, Na⁺, and especially Li⁺ ions compared to WT HCN4 channels. These results establish that the TIGYG sequence known to be critical for selectivity in K⁺ channels does not confer selectivity to HCN4 channels and indicate that HCN channels may be more structurally divergent from K⁺ than anticipated.

Ion selectivity for both WT and C479T HCN4 channels follows a Type IV Eisenman sequence[30],[31], indicating that replacement of C479 with threonine has minimal impact on binding properties of ions in the pore. On the other hand, compared to WT HCN4 channels, C479T channels showed marked elevations in permeability to large organic cations, particularly DMA⁺ and TMA⁺ ions, as well as Li⁺ ions, which have the largest hydrated radius (i.e. 3.8 Å) of the alkali series of cations and are unlikely to be dehydrated due to their very negative free energy of hydration (ΔG_H = −120.1 kcal/mol)[52]. These observations suggest that replacement of C479 with threonine enlarged the pore diameter. To further assess and quantify pore size we applied “Excluded Field Theory” (EFT)[43] to our permeation studies. This method was previously applied to accurately predict the dimensions of multi-ion K⁺ channel pores[41],[42] and has been applied to estimate pore sizes for various multi-ion channels[28],[33]–[40]. EFT assumes that ion selection/permeability is dictated primarily sieving mechanisms rather than ion binding properties[43]. The validity of this assumption in HCN4 channels is supported by the monotonic decrease in permeability ratios with increased organic cation diameter (Figure 3B) and by the alkali permeability sequence measured for WT and C479T HCN4 channels which follow an Eisenman Type IV sequences indicative of weak cation binding in the pore. The application of EFT to our organic cation studies predicted that WT HCN4 channels have a pore diameter of ∼6.9 Å in WT HCN4 channels which was increased to ∼8.7 Å in C479T channels. While the basis for this increase in pore diameter remains unclear, it could simply originate from the steric differences between the two residue side-chains. Indeed, assuming the most stable S1-rotamer for the cysteine[53] and threonine side chains in an HCN2 molecular model revealed that the pore diameter increases by ∼1.8 Å (Figure 7) when C479 is replaced by threonine. This is nearly identical to our EFT predictions (∼1.9 Å). Such large pore diameters for WT and C479T HCN4 channels are predicted to readily allow passage of partially hydrated Na⁺ ions in a manner similar to the Na⁺ permeation observed in SK_Ca, NaK and mutant Kir3.1/3.4 channels[26],[54],[55]. Larger pores can also clearly explain the ∼70% larger single-channel conductance of C479T channels compared to WT HCN4 channels, particularly if permeation in these channels involves a sieving mechanism, as discussed above.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 7. Proposed pore topology in WT and C479T HCN4 channels.

A space filled model of the pore sequence in WT and C479T HCN4 channels based on the model of mHCN2 (Giorgettiet al., 2005). The top two panels depict a side view of this sequence in these channels with the intracellular side of the channel on the left. The bottom panels depict a bottom up view of the pore in these channels. The bulk of the −SH group (yellow) in WT HCN4 would limit the pore diameter, and provide a steric barrier thereby limiting ionic permeation and selectivity based on hydrated ionic radius. The C479T mutation increases the pore diameter from 5.4 Å in WT (at the edge of the electron clouds) to 6.2 Å and reduces the steric barrier for ion permeation and ionic selectivity based on effective radius.

https://doi.org/10.1371/journal.pone.0007712.g007

Our estimates of pore size in WT HCN4 channels are clearly larger than K⁺ channel pores (∼3.5 Å) estimated from atomic structures[15],[16] or from EFT[41],[42], and is in good agreement with earlier estimates from I_h channel studies[56] and molecular modeling of spHCN and HCN2 channels[53]. Clearly, the basis for larger pore sizes in HCN4 channels compared to K⁺ channels are unlikely to originate from differences in the amino acid sequences in the selectivity filter. On the other hand, several molecular structures required for K⁺ rigidity and stability in K⁺ channels appear to be absent in HCN4 channels[53]. Moreover, introducing mutions in other regions of the P-loop, linked previously to pore stability in K⁺ channels[15],[23],[25],[26],[54] resulted either in non-functional channels (S475E/C479T and F471W/K472W/C479T) or channels that remained highly permeable to Na⁺ (S475D/C479T). Thus, the underlying architecture of HCN channels appears to be significantly different from K_v channels. The molecular basis for these differences in pore size and structure between HCN channels and K⁺ channels will clearly require further studies.

In addition to changes in permeation, we routinely observed reproducible differences in gating properties between C479T and WT HCN4 channels. For example, the C479T replacement resulted in a more shallow voltage-dependence for activation compared with WT HCN4 channels, without affecting the voltage for 50% activation (i.e. V_1/2). Moreover, ionic substitution had no effect on any gating properties of C479T channels. By contrast, in WT HCN4 channels substitution of extracellular K⁺ with Na⁺ accelerated activation, but not deactivation, rates and induced positive voltage shifts in the steady-state activation. Interestingly, the effects of ion substitution in HCN4 channels differed from that reported previously for HCN1 and HCN2 channels[19],[48],[49]. Clearly, additional studies will be required determine the basis for these differences between HCN4 versus HCN1 and HCN2. Regardless, it is possible that the alterations in gating seen in C479T channels originate from changes in selectivity properties or selectivity filter structure which have previously been shown to modulate “pore-gating”[32],[50],[51].

In conclusion, our data establish that the TIGYG sequence and other pore stabilizing interactions known to be critical for selectivity in K⁺ channels do not confer selectivity to HCN4 channels. Consistent with previous reports, our data suggest that HCN4 channels have much larger pore diameters than K_v channels and the C479 residue may form a steric barrier to the permeation of large ions. Overall our data suggests that HCN channels may be less similar in structure to K_v channels than originally anticipated.

Acknowledgments

We thank Dr. H. Ohmori for providing us with the WT rabbit HCN4 clone.

Author Contributions

Conceived and designed the experiments: ND RP PHB. Performed the experiments: ND RP. Analyzed the data: ND. Contributed reagents/materials/analysis tools: PHB. Wrote the paper: ND PHB.

References

1. 1.Robinson RB, Siegelbaum SA (2003) Hyperpolarization-activated cation currents: from molecules to physiological function. Annu Rev Physiol 65: 453–480.
   • View Article
   • Google Scholar
2. 2.DiFrancesco D (1993) Pacemaker mechanisms in cardiac tissue. Annu Rev Physiol 55: 455–472.
   • View Article
   • Google Scholar
3. 3.Pape HC (1996) Queer current and pacemaker: the hyperpolarization-activated cation current in neurons. Annu Rev Physiol 58: 299–327.
   • View Article
   • Google Scholar
4. 4.Ludwig A, Zong X, Stieber J, Hullin R, Hofmann F, et al. (1999) Two pacemaker channels from human heart with profoundly different activation kinetics. Embo J 18(9): 2323–2329.
   • View Article
   • Google Scholar
5. 5.Stieber J, Thomer A, Much B, Schneider A, Biel M, et al. (2003) Molecular basis for the different activation kinetics of the pacemaker channels HCN2 and HCN4. J Biol Chem 278(36): 33672–33680.
   • View Article
   • Google Scholar
6. 6.Mistrik P, Mader R, Michalakis S, Weidinger M, Pfeifer A, et al. (2005) The murine HCN3 gene encodes a hyperpolarization-activated cation channel with slow kinetics and unique response to cyclic nucleotides. J Biol Chem 280(29): 27056–27061.
   • View Article
   • Google Scholar
7. 7.Mannikko R, Elinder F, Larsson HP (2002) Voltage-sensing mechanism is conserved among ion channels gated by opposite voltages. Nature 419(6909): 837–841.
   • View Article
   • Google Scholar
8. 8.Vemana S, Pandey S, Larsson HP (2004) S4 movement in a mammalian HCN channel. J Gen Physiol 123(1): 21–32.
   • View Article
   • Google Scholar
9. 9.Hestrin S (1987) The properties and function of inward rectification in rod photoreceptors of the tiger salamander. J Physiol 390: 319–333.
   • View Article
   • Google Scholar
10. 10.Ho WK, Brown HF, Noble D (1994) High selectivity of the i(f) channel to Na+ and K+ in rabbit isolated sinoatrial node cells. Pflugers Arch 426(1–2): 68–74.
    • View Article
    • Google Scholar
11. 11.Ludwig A, Zong X, Jeglitsch M, Hofmann F, Biel M (1998) A family of hyperpolarization-activated mammalian cation channels. Nature 393(6685): 587–591.
    • View Article
    • Google Scholar
12. 12.Bell DC, Yao H, Saenger RC, Riley JH, Siegelbaum SA (2004) Changes in local S4 environment provide a voltage-sensing mechanism for mammalian hyperpolarization-activated HCN channels. J Gen Physiol 123(1): 5–19.
    • View Article
    • Google Scholar
13. 13.Santoro B, Tibbs GR (1999) The HCN gene family: molecular basis of the hyperpolarization-activated pacemaker channels. Ann N Y Acad Sci 868: 741–764.
    • View Article
    • Google Scholar
14. 14.Kaupp UB, Seifert R (2001) Molecular diversity of pacemaker ion channels. Annu Rev Physiol 63: 235–257.
    • View Article
    • Google Scholar
15. 15.Doyle DA, Morais-Cabral J, Pfuetzner RA, Kuo A, Gulbis JM, et al. (1998) The Structure of the Potassium Channel: Molecular Basis of K⁺ Conduction and Selectivity. Science 280(5360): 69–77.
    • View Article
    • Google Scholar
16. 16.Zhou Y, Morais-Cabral JH, Kaufman A, MacKinnon R (2001) Chemistry of ion coordination and hydration revealed by a K+ channel-Fab complex at 2.0 A resolution. Nature 414(6859): 43–48.
    • View Article
    • Google Scholar
17. 17.Heginbotham L, Lu Z, Abramson T, MacKinnon R (1994) Mutations in the K⁺ Channel Signature Sequence. Biophsyical J 66: 1061–1067.
    • View Article
    • Google Scholar
18. 18.Ishii TM, Takano M, Xie LH, Noma A, Ohmori H (1999) Molecular characterization of the hyperpolarization-activated cation channel in rabbit heart sinoatrial node. J Biol Chem 274(18): 12835–12839.
    • View Article
    • Google Scholar
19. 19.Azene EM, Xue T, Li RA (2003) Molecular basis of the effect of potassium on heterologously expressed pacemaker (HCN) channels. J Physiol 547(Pt 2): 349–356.
    • View Article
    • Google Scholar
20. 20.Zhou M, MacKinnon R (2004) A mutant KcsA K(+) channel with altered conduction properties and selectivity filter ion distribution. J Mol Biol 338(4): 839–846.
    • View Article
    • Google Scholar
21. 21.Yool AJ, Schwarz TL (1991) Alteration of ionic selectivity of a K+ channel by mutation of the H5 region. Nature 349(6311): 700–704.
    • View Article
    • Google Scholar
22. 22.Yool AJ, Schwarz TL (1996) Anomalous mole fraction effect induced by mutation of the H5 pore region in the Shaker K+ channel. Biophys J 71(5): 2467–2472.
    • View Article
    • Google Scholar
23. 23.Loboda A, Melishchuk A, Armstrong C (2001) Dilated and defunct K channels in the absence of K+. Biophys J 80(6): 2704–2714.
    • View Article
    • Google Scholar
24. 24.Starkus JG, Kuschel L, Rayner MD, Heinemann SH (1998) Macroscopic Na+ currents in the “Nonconducting” Shaker potassium channel mutant W434F. J Gen Physiol 112(1): 85–93.
    • View Article
    • Google Scholar
25. 25.Yang J, Mei Yu, Yuh Nung Jan, Lily Yeh Jan (1997) Stabilization of ion selectivity filter by pore loop ion pairs in an inwardly rectifying potassium channel. Proc Natl Acad Sci 94: 1568–1572.
    • View Article
    • Google Scholar
26. 26.Dibb KM, Rose T, Makary SY, Claydon TW, Enkvetchakul D, et al. (2003) Molecular basis of ion selectivity, block, and rectification of the inward rectifier Kir3.1/Kir3.4 K(+) channel. J Biol Chem 278(49): 49537–49548.
    • View Article
    • Google Scholar
27. 27.Giorgetti A, Carloni P, Mistrik P, Torre V (2005) A Homology Model of the Pore Region of HCN Channels. Biophys J 89(2): 932–944.
    • View Article
    • Google Scholar
28. 28.Dwyer TM, Adams DJ, Hille B (1980) The permeability of the endplate channel to organic cations in frog muscle. J Gen Physiol 75(5): 469–492.
    • View Article
    • Google Scholar
29. 29.Much B, Wahl-Schott C, Zong X, Schneider A, Baumann L, et al. (2003) Role of subunit heteromerization and N-linked glycosylation in the formation of functional hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 278(44): 43781–43786.
    • View Article
    • Google Scholar
30. 30.Eisenman G (1962) Cation selective glass electrodes and their mode of operation. Biophys J 2(2 Pt 2): 259–323.
    • View Article
    • Google Scholar
31. 31.Eisenman G, Horn R (1983) Ionic selectivity revisited: the role of kinetic and equilibrium processes in ion permeation through channels. J Membr Biol 76(3): 197–225.
    • View Article
    • Google Scholar
32. 32.Zheng J, Sigworth FJ (1997) Selectivity changes during activation of mutant Shaker potassium channels. J Gen Physiol 110(2): 101–117.
    • View Article
    • Google Scholar
33. 33.Hille B (1971) The permeability of the sodium channel to organic cations in myelinated nerve. J Gen Physiol 58(6): 599–619.
    • View Article
    • Google Scholar
34. 34.Hille B (1972) The permeability of the sodium channel to metal cations in myelinated nerve. J Gen Physiol 59(6): 637–658.
    • View Article
    • Google Scholar
35. 35.Coronado R, Smith JS (1987) Monovalent ion current through single calcium channels of skeletal muscle transverse tubules. Biophys J 51(3): 497–502.
    • View Article
    • Google Scholar
36. 36.McCleskey EW, Almers W (1985) The Ca channel in skeletal muscle is a large pore. Proc Natl Acad Sci U S A 82(20): 7149–7153.
    • View Article
    • Google Scholar
37. 37.Cohen BN, Labarca C, Davidson N, Lester HA (1992) Mutations in M2 alter the selectivity of the mouse nicotinic acetylcholine receptor for organic and alkali metal cations. J Gen Physiol 100(3): 373–400.
    • View Article
    • Google Scholar
38. 38.Eickhorst AN, Berson A, Cockayne D, Lester HA, Khakh BS (2002) Control of P2X(2) channel permeability by the cytosolic domain. J Gen Physiol 120(2): 119–131.
    • View Article
    • Google Scholar
39. 39.Voets T, Janssens A, Droogmans G, Nilius B (2004) Outer pore architecture of a Ca2+-selective TRP channel. J Biol Chem 279: 15223–15230.
    • View Article
    • Google Scholar
40. 40.Wang HZ, Veenstra RD (1997) Monovalent ion selectivity sequences of the rat connexin43 gap junction channel. J Gen Physiol 109(4): 491–507.
    • View Article
    • Google Scholar
41. 41.Bezanilla F, Armstrong CM (1972) Negative conductance caused by entry of sodium and cesium ions into the potassium channels of squid axons. J Gen Physiol 60(5): 588–608.
    • View Article
    • Google Scholar
42. 42.Hille B (1973) Potassium channels in myelinated nerve. Selective permeability to small cations. J Gen Physiol 61(6): 669–686.
    • View Article
    • Google Scholar
43. 43.Sun YM, Favre I, Schild L, Moczydlowski E (1997) On the structural basis for size-selective permeation of organic cations through the voltage-gated sodium channel. Effect of alanine mutations at the DEKA locus on selectivity, inhibition by Ca2+ and H+, and molecular sieving. J Gen Physiol 110(6): 693–715.
    • View Article
    • Google Scholar
44. 44.Coronado R, Rosenberg RL, Miller C (1980) Ionic selectivity, saturation, and block in a K+-selective channel from sarcoplasmic reticulum. J Gen Physiol 76(4): 425–446.
    • View Article
    • Google Scholar
45. 45.Bucchi A, Baruscotti M, DiFrancesco D (2002) Current-dependent block of rabbit sino-atrial node I(f) channels by ivabradine. J Gen Physiol 120(1): 1–13.
    • View Article
    • Google Scholar
46. 46.DiFrancesco D, Ferroni A, Mazzanti M, Tromba C (1986) Properties of the hyperpolarizing-activated current (If) in cells isolated from the rabbit sino-atrial node. J Physiol 377: 61–88.
    • View Article
    • Google Scholar
47. 47.Kole MH, Hallermann S, Stuart GJ (2006) Single Ih channels in pyramidal neuron dendrites: properties, distribution, and impact on action potential output. J Neurosci 26(6): 1677–1687.
    • View Article
    • Google Scholar
48. 48.Azene EM, Sang D, Tsang SY, Li RA (2005) Pore-to-gate coupling of HCN channels revealed by a pore variant that contributes to gating but not permeation. Biochem Biophys Res Commun 327(4): 1131–1142.
    • View Article
    • Google Scholar
49. 49.Macri V, Proenza C, Agranovich E, Angoli D, Accili EA (2002) Separable gating mechanisms in a Mammalian pacemaker channel. J Biol Chem 277(39): 35939–35946.
    • View Article
    • Google Scholar
50. 50.Zheng J, Sigworth FJ (1998) Intermediate conductances during deactivation of heteromultimeric Shaker potassium channels. J Gen Physiol 112(4): 457–474.
    • View Article
    • Google Scholar
51. 51.Berneche S, Roux B (2005) A gate in the selectivity filter of potassium channels. Structure (Camb) 13(4): 591–600.
    • View Article
    • Google Scholar
52. 52.Hush NS (1948) The Free Energies of Hydration of Gaseous Ions. Australian Journal of Scientific Research 1(4): 480–494.
    • View Article
    • Google Scholar
53. 53.Giorgetti A, Carloni P, Mistrik P, Torre V (2005) A homology model of the pore region of HCN channels. Biophys J 89(2): 932–944.
    • View Article
    • Google Scholar
54. 54.Guidoni L, Torre V, Carloni P (1999) Potassium and sodium binding to the outer mouth of the K+ channel. Biochemistry 38(27): 8599–604.
    • View Article
    • Google Scholar
55. 55.Noskov SY, Roux B (2007) Importance of Hydration and Dynamics on the Selectivity of the KcsA and NaK Channels. J Gen Physiol 129(2): 135–143.
    • View Article
    • Google Scholar
56. 56.Wollmuth LP, Hille B (1992) Ionic selectivity of Ih channels of rod photoreceptors in tiger salamanders. J Gen Physiol 100(5): 749–765.
    • View Article
    • Google Scholar