J Physiol Visit Proceedings
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Physiol Volume 562, Number 3, 687-696, February 1, 2005 DOI: 10.1113/jphysiol.2004.079020
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
562/3/687    most recent
jphysiol.2004.079020v1
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by O'Connell, A. D
Right arrow Articles by Hunter, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by O'Connell, A. D
Right arrow Articles by Hunter, M.

Selectivity and interactions of Ba2+ and Cs+ with wild-type and mutant TASK1 K+ channels expressed in Xenopus oocytes

Anthony D O'Connell1, Michael J Morton1, Asipu Sivaprasadarao1 and Malcolm Hunter1

1 School of Biomedical Sciences, Worsley Building, University of Leeds, Leeds LS2 9JT, UK


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 References
 
The acid-sensitive K+ channel, TASK1 is a member of the K+-selective tandem-pore domain (K2P) channel family. Like many of the K2P channels, TASK1 is relatively insensitive to conventional channel blockers such as Ba2+. In this paper we report the impact of mutating the pore-neighbouring histidine residues, which are involved in pH sensing, on the sensitivity to blockade by Ba2+ and Cs+; additionally we compare the selectivity of these channels to extracellular K+, Na+ and Rb+. H98D and H98N mutants showed reduced selectivity for K+ over both Na+ and Rb+, and significant permeation of Rb+. This enhanced permeability must reflect changes in the structure or flexibility of the selectivity filter. Blockade by Ba2+ and Cs+ was voltage-dependent, indicating that both ions block within the pore. In 100 mM K+, the KD at 0 mV for Ba2+ was 36 ± 10 mM (n = 6), whilst for Cs+ it was 20 ± 6.0 mM (n = 5). H98D was more sensitive to Ba2+ than the wild-type (WT); in addition, the site at which Ba2+ appears to bind was altered (WT: {delta}, 0.64 ± 0.16, n = 6; H98D: {delta}, 0.16 ± 0.03, n = 5, statistically different from WT; H98N: {delta}, 0.58 ± 0.09, not statistically different from WT). Thus, the pore-neighbouring residue H98 contributes not only to the pH sensitivity of TASK1, but also to the structure of the conduction pathway.

(Received 11 November 2004; accepted after revision 7 December 2004; first published online 20 December 2004)
Corresponding author M. Hunter: School of Biomedical Sciences, Worsley Building, University of Leeds, Leeds LS2 9JT, UK. Email: m.hunter{at}leeds.ac.uk


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 References
 
TASK1 (KCNK3) is a member of the mammalian tandem-pore domain K+ channel (K2P) family (Duprat et al. 1997). Members of this group of channels have only moderate sequence homology, but are characterized by having four membrane spanning domains and two pore domains per channel subunit – functional channels are thought to comprise two subunits (for a review of K2P channels see O'Connell et al. 2002). Messenger RNA transcripts for TASK1 have been identified in numerous tissues including kidney, heart and brain in mouse, rat and human (Duprat et al. 1997; Leonoudakis et al. 1998; Kim et al. 1998; Medhurst et al. 2001). Current flow through TASK channels obeys the Goldman-Hodgkin-Katz current equation, indicating that the open probability is independent of voltage (Leonoudakis et al. 1998; Goldstein et al. 2001). Channels are open at resting membrane potentials and as such may be involved in bringing the resting membrane potential close to the equilibrium potential for K+. An important feature of TASK1 is its sensitivity to changes in extracellular pH; an increase in K+ current occurs as the external pH increases within the physiological range, with little or no current seen at pH 6 or below (Duprat et al. 1997; Leonoudakis et al. 1998; Lopes et al. 2000; Morton et al. 2003).

TASK3, a related mammalian K2P channel, is also sensitive to extracellular pH (Kim et al. 2000; Rajan et al. 2000). Mutating a histidine (H98) residue at the outermost edge of the selectivity filter (GYGH) of the first pore domain to asparagine (N, neutral at pH 7.4) or aspartate (D, negatively charged at pH 7.4) rendered the mutants insensitive to extracellular pH (the corresponding residue in the second pore domain is D). In TASK1, mutating H98 to N also caused a reduction in extracellular pH-sensitivity (Lopes et al. 2000; Morton et al. 2003; Yuill et al. 2004). In the bacterial K+ channel, KcsA, the corresponding residue is D, and is thought to assist in attracting cations from bulk solution prior to dehydration and entry into the pore (Zhou et al. 2001) – this should apply to both permeant and blocking cations. To cause channel block, non-conducting cations, such as Ba2+ and Cs+, enter the pore but permeate at such a slow rate that they effectively stop current flow (Hille, 2001). In the case of KcsA, the Ba2+ binding site has been shown to be in the innermost cation-binding site of the selectivity filter (Jiang & Mackinnon, 2000), in agreement with earlier functional studies in Ca2+-activated K+ channels and the voltage-gated Shaker channels (Neyton & Miller, 1988; Vergara et al. 1999). This raises the possibility that substitution of two of the D residues surrounding the mouth of the pore with H residues, as occurs naturally in wild-type (WT) TASK1 channels, might affect the sensitivity of the channels to cationic blockers, and may help to explain the relative insensitivity of TASK1 channels to Ba2+ (Duprat et al. 1997; Yuill et al. 2004). In this paper we have investigated the role of the H residues surrounding the pore mouth in the determination of cation selectivity and blockade.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 References
 
Oocyte preparation

Female Xenopus laevis were immersed for 30 min in ice-cold tricane solution (0.2% 3-aminobenzoic acid ethyl ester methanesulphonate salt titrated to pH 7.4 with NaOH). Adequacy of anaesthesia was judged by lack of a righting-reflex when animals were placed in the supine position. Prior to laparotomy, animals were humanely killed in accordance with the Home Office Guidance on the Operation of the Animals (Scientific Procedures) Act of 1986.

Oocytes were removed and placed in a calcium-free Ringer (CaFR) solution containing (mM): NaCl 82, KCl 2, MgCl2 20 and Hepes 5; titrated to pH 7.4 with NaOH. The oocytes were separated into clumps of around 10 cells with watchmakers' forceps, washed in CaFR solution and placed in a Petri dish containing type 1A collagenase solution (2 mg ml–1 in CaFR solution) and gently agitated on a shaking platform for 1 h. Afterwards, cells were rinsed in CaFR solution and re-incubated with collagenase solution as described above, for a further hour. Cells were rinsed and stored in modified Barth's solution containing (mM): NaCl 96, KCl 2, CaCl2 1.8, Hepes 5, sodium pyruvate 1.5, and 5000 U penicillin, 5 mg ml–1 streptomycin, 50 mg ml–1 neomycin and 5 mg ml–1 gentamycin; titrated to pH 7.4 with NaOH. At this stage, most of the cells were separate and defolliculated, attached cells could be separated by gentle trituration with a Pasteur pipette. Cells were subsequently placed in single wells of a 96-well culture plate containing modified Barth's solution and maintained at 4°C prior to injection, which occurred no later than 24 h after defolliculation.

Messenger RNA production and oocyte injection

Plasmids containing murine TASK1 (KCNK3) were linearized and from this, capped cRNA was transcribed in vitro using a MEGAscript kit (Ambion Ltd, Huntingdon, Cambs, UK). Point mutations were made by an in-house adaptation of the Quick-change site-directed mutagenesis protocol (Stratagene, La Jolla, CA, USA) and verified by automated fluorescence sequencing (Lark Technologies Ltd, Takeley, Essex, UK). cRNA solution (50 nl of 0.05–0.1 µg µl–1) was injected into the vegetal poles of defolliculated stage V or VI oocytes. Negative control cells were injected with 50 nl RNase-free water (Sigma, UK). Cells were incubated in modified Barth's solution at 19°C for 1–3 days prior to experimentation.

Two-electrode voltage clamp

Oocytes were introduced to a perfusion chamber (volume, ~300 µl) and superfused (~1 ml min–1) with Ringer solution containing (mM): NaCl 97, KCl 3, MgCl2 1, CaCl2 2 and Hepes 10; titrated to pH 7.4 with NaOH. Oocytes were impaled with two standard borosilicate glass microelectrodes (resistance, 1.5–2.5 M{Omega}; Clark Electromedical Instruments, UK) back-filled with 3 M KCl. Resealing of the oocyte membrane was judged to be complete when the membrane potential stabilized and the potential recorded by the two electrodes differed by 2 mV or less. Currents were measured using a Warner Oocyte Clamp OC-725B (Hamden, CT, USA) and passed through a low-pass 8-pole Bessel filter (100 Hz) prior to sampling (1 kHz). Data were acquired via a Tl–1 DMA LabMaster Interface digital-to-analog converter (Axon Instruments, Union City, CA, USA) driven by commercial software (pClamp Clampex 5.5.1, Axon Instruments). Steady-state currents were determined from the final 2 ms of the voltage pulse and were analysed off-line with Clampex 8.0 (Axon Instruments), Excel 97 (Microsoft Corporation, USA) or Origin 6.0 (Microcal, USA). Recordings were made in 100 mM K+ solution containing (mM): K+ 100 mM, MgCl2 1, CaCl2 2 and Hepes 10; adjusted to pH 8 with KOH. BaCl2 or CsCl was added to 100 mM K+ solution in concentrations between 0.3 and 10 mM. For selectivity studies, extracellular solutions were (in mM): NaCl, KCl or RbCl 100, MgCl2 1, CaCl2 2 and Hepes 10; adjusted to pH 8 with N-methyl-D-gluconate. In selectivity studies, leak-subtraction was not employed. In the blocker studies, values at –100 mV in the presence of 10 mM Ba2+ (in 100 K+ solution) were assumed to represent leak currents and were used to determine the leak conductance. This leak conductance was then used to predict linear leak currents over the whole voltage range, with an assumed reversal potential of 0 mV. This predicted current was later subtracted from all records for that particular cell. The average leak conductance in WT channels was 2.3 ± 0.6 µS (n = 6), equating to currents of –0.23 ± 0.06 µA (n = 6) at –100 mV. As the leak conductance determined in this manner was higher than the conductance at 0 mV in water-injected oocytes, this may represent an overestimation of the leak conductance, i.e. the channels were not completely blocked by 10 mM Ba2+. Nonetheless, as there was considerable variation in endogenous currents between batches of oocytes, we have used this approach to correct for leak currents as much as we were able. To guard against contamination of the data by seasonal variations in endogenous currents, water-injected oocytes were used as controls and batches of oocytes with obvious endogenous currents were discarded.

Data presentation

Data are presented as means ± S.E.M. with n number of experiments. Statistical significance was assumed at the 5% level (P < 0.05) using ANOVA and if necessary, a post hoc test (stated in the text). All reagents were purchased from Sigma-Aldrich (Poole, UK) unless otherwise stated. Experiments were performed at room temperature (22–24°C).


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 References
 
Cation selectivity

The ion selectivity of a channel may be represented in two ways, either as a permeability ratio or a conductance ratio. Here, permeability ratios were determined from the change in reversal potential upon the substitution of one ion for another and reflect the relative ion occupancies of the pore, but do not necessarily reflect the rate of permeation (Hille, 2001). They were calculated from the following equation:


{tjp_692_m1}

(1)
where PX/PK is the permeability ratio of ion X+ with respect to K+, EX and EK are the reversal potentials in solution containing ion X+ or K+, and the universal gas constant (R), valency (z) and Faraday's constant (F)have their usual values, and T was 293 K. Reversal potentials were determined from least-squares second-order polynomial fits to the current–voltage relationships. Conductance ratios reflect the ease of ions entering, crossing and exiting the pore, and were determined as the ratio of the slope conductance between –40 mV and –100 mV in the presence of one or another ion. Permeability ratios, conductance ratios and mean reversal potentials for WT and mutant channels, and water-injected oocytes are shown in Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1.  Monovalent cation selectivity
 
The permeability ratio and conductance ratio selectivity sequences for WT, H98D and H98N were, K+ > Rb+ > Na+. Figure 1A and B shows raw current traces and current–voltage relationships, respectively, from oocytes injected with WT (left), H98D (middle) or H98N (right) cRNA in K+, Na+ or Rb+ solutions. In each channel, large inward currents are observed in 100 mM K+, and smaller inward currents are observed in Rb+ or Na+ bath solutions. For both mutants, Rb+, and to a lesser extent Na+, permeate more than in WT. For H98N, the Rb+ current at –100 mV is not significantly different from K+ current at –100 mV, while for H98D and WT, the two currents remain significantly different from each other. For the expressed channels, it is not certain whether the inward currents seen in the presence of Na+ reflect true inward Na+ currents through TASK1, as these currents were of a similar magnitude to the leak currents in water-injected oocytes (which were –0.06 ± 0.03 µA, n = 6). However, there is clearly an increase in PNa in H98D, because there is a large positive shift in the reversal potential in comparison to WT. Note also that the reversal potential in water-injected oocytes with Na+ solution is different to that with either WT or mutant channels, indicating that in these cases the reversal potentials were dominated by the heterologously expressed channel currents.



View larger version (26K):
[in this window]
[in a new window]
  		Figure 1.  Cation selectivity of WT and mutant channels
A, raw current traces of WT (left), H98D (middle) and H98N (right) in 100 mM K+, Na+ or Rb+ solutions. Current records were not leak-subtracted. B, current–voltage relationships of WT (left), H98D (middle) and H98N (right) in 100 mM K+, Na+ or Rb+ solutions. Steady-state currents, not leak-subtracted, were recorded at the end of the voltage pulses (WT, n = 9; H98D, n = 7; H98N, n = 6).

 
Block by Ba2+ of TASK1-WT

Figure 2 (left) shows the sensitivity of WT channels to Ba2+. As the extracellular Ba2+ concentration increases, the inward currents decreased at all potentials. Current–voltage relationships from each channel are plotted in Fig. 3A and demonstrate that Ba2+ blocks TASK1 in a voltage- and concentration-dependent manner. Continuous lines in Fig. 3A are best-fits to the mean steady-state currents, taken immediately before the end of the voltage pulse, using eqn (2):


{tjp_692_m2}

(2)
where {delta} is the fraction of the membrane electric field felt by a permeating ion, the mean of which was 0.64 ± 0.16 (n = 6), suggesting that Ba2+ binds approximately 60% of the way across the membrane electric field. The KD at 0 mV (KD0) could not be determined experimentally because there were no currents at the equilibrium potential for K+ (0 mV in 100 mM K+ solution). Therefore, eqn (2) was again used to estimate this value for each oocyte. The mean KD0 was 36 ± 10 mM Ba2+ (n = 6). KD0 and {delta} were used to create the continuous lines in Fig. 3A.



View larger version (26K):
[in this window]
[in a new window]
  		Figure 2.  Inhibition of WT and mutant channels by extracellular Ba2+
Currents in response to voltage steps from –100 mV to +20 mV in 20 mV steps (holding potential, 0 mV) from single oocytes injected with WT (left), H98D (middle) or H98N (right) bathed in concentrations of Ba2+ between 0 and 10 mM. Original current records presented here were not leak-subtracted.

 


View larger version (27K):
[in this window]
[in a new window]
  		Figure 3.  Characteristics of block by Ba2+
A, current–voltage relationships of WT (left, n = 6), H98D (middle, n = 5) and H98N (right, n = 5) at various concentrations of extracellular Ba2+. Currents were leak subtracted (see Methods). Continuous lines represent a model created using eqn (2) in which the mean KD0 and mean {delta}-values were substituted for each model (see text for values). B, dose–response curves of each channel at voltages between –40 mV and –100 mV (WT, left, n = 6; H98D, middle, n = 5; H98N, right, n = 5). Continuous lines are least-squares fits to mean data using eqn (3).

 
Dose–response curves for inhibition by Ba2+ at voltages from –40 mV to –100 mV are shown in Fig. 3B (left). The continuous lines in Fig. 3B are least-squares fits to the data using the Hill equation (eqn (3)).


{tjp_692_m3}

(3)
Hill coefficients (h) at different voltages were ~1 (no significant difference between h at the different voltages). However the KD for Ba2+ block increased at more positive voltages (Fig. 6A). There was a linear relationship between the applied voltage and log KD (continuous lines, Fig. 6A) as described by eqn (2) (Hagiwara et al. 1978).



View larger version (18K):
[in this window]
[in a new window]
  		Figure 6.  KD values for all channels in Ba2+ and Cs+ solutions
A, KD values for WT (n = 6), H98D (n = 5) and H98N (n = 6) in Ba2+. B, KD values for WT (n = 6), H98D (n = 7) and H98N (n = 6) in Cs+. Lines are least-squares best fits to leak-subtracted mean data using eqn (2). *Statistically significantly different KD values from WT at the same voltage were determined using ANOVA followed by Dunn's post hoc test.

 
Effect of extracellular Ba2+ on H98D and H98N

Figure 2 (middle and right) shows K+ currents from oocytes expressing H98D and H98N channels. As with WT, extracellular Ba2+ reduces inward currents in a voltage and concentration-dependent manner. Figure 3A and B (middle and right for both) show current–voltage relationships and dose–response curves, respectively. It is immediately apparent that H98D is more sensitive to Ba2+ in comparison to WT or H98N. Mean KD values for Ba2+ block of H98D and H98N are plotted in Fig. 6A. For H98D, the KD at voltages between –100 mV and –40 mV were significantly different from WT. In addition, H98D has a less-steep slope than H98N or WT channels (Fig. 6A), indicating a reduction in voltage-dependence. This is reflected in the mean {delta}-values, a reflection of the slope (H98D, 0.16 ± 0.03; H98N, 0.58 ± 0.09, n = 5 each). Hill coefficients from the same channel at different voltages, or between channels, were not significantly different from each other, and remained close to 1. Mean KD0 values for H98D and H98N were 0.44 ± 0.09 (n = 5) and 7.6 ± 3.4 mM Ba2+ (n = 5), respectively. For H98D, both KD0 and {delta}-values were significantly different from WT. For each channel, these best-fit KD0 and {delta}-values were used to create the continuous lines in Fig. 3A.

Block by Cs+ of WT-TASK1

Figure 4 shows concentration- and voltage-dependent inhibition by Cs+. The onset of block with Cs+ is much faster than with Ba2+ (cf Fig. 2), and appears to be instantaneous. The continuous lines in Fig. 5A are fits using eqn (2). The mean KD0 was 20 ± 6.0 mM Cs+ (n = 5) and the mean {delta} was 0.67 ± 0.22 (n = 5). Neither of these values are significantly different from those for block by Ba2+, which might suggest that Cs+ and Ba2+ bind at the same site, although it should be borne in mind that electrical distances do not necessarily equate to physical distances, particularly in multi-ion pores (Hille, 2001). Dose–response curves for Cs+ in Fig. 5B (left) demonstrate concentration- and voltage-dependent block. Continuous lines are least-squares best fits to the data using the Hill equation (eqn (3)). At any given Cs+ concentration, the degree of block is more pronounced at negative potentials. Hill coefficients for Cs+ block at different voltages are not significantly different from each other and remain close to 1 regardless of voltage. Figure 6B shows KD values for block by Cs+ plotted as a function of membrane potential. The continuous line is a least-squares fit to the mean data using eqn (2). At less negative potentials the data are fairly linear, as expected for a blocking ion.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 4.  Inhibition of WT and mutant channels by extracellular Cs+
Currents in response to voltage steps from –100 mV to +20 mV in 20 mV steps (holding potential, 0 mV) from single oocytes injected with WT (left), H98D (middle) or H98N (right) bathed in concentrations of Cs+ between 0 and 10 mM. Original current records presented here were not leak-subtracted.

 


View larger version (27K):
[in this window]
[in a new window]
  		Figure 5.  Characteristics of block by Cs+
A, current–voltage relationships of WT (left, n = 6), H98D (middle, n = 7) and H98N (right, n = 6) channels at varied concentrations of extracellular Cs+. Currents were leak subtracted (see Methods). Continuous lines represent a model created using eqn (2) in which the mean KD0 and mean {delta}-values were substituted in each model (see text for values). B, dose–response curves of each channel at voltages between –40 mV and –100 mV (WT, left, n = 6; H98D, middle, n = 5; H98N, right, n = 5). Continuous lines are least-squares fits to mean data using eqn (3). For H98D and H98N, meaningful mean data and least-squares fits to mean data could not be reasonably established from currents at –40 mV, and so were omitted.

 
Effect of extracellular Cs+ on H98N and H98D

Figure 4 shows K+ currents from oocytes expressing H98D and H98N. As with WT channels, with increased extracellular Cs+ concentration the inward currents decrease. Figure 5A and B show current–voltage relationships and dose–response curves of inhibition of H98D and H98N by Cs+, respectively. H98N is similarly sensitive to Cs+ compared with WT. However, at each voltage, the KD for H98D is significantly higher than that in WT (Fig. 6B). Current–voltage relationships (Fig. 5A) demonstrate that each mutant is blocked by Cs+ in a concentration- and voltage-dependent manner. Continuous lines in Fig. 5A are best fits to eqn (2) which predicts the degree of block. Dose–response curves in Fig. 5B demonstrate concentration- and voltage-dependent block by Cs+, although fits to currents at –40 mV were poor and so omitted from this study. Hill coefficients from the same channel, or between channels, are not significantly different from each other, and remained close to 1. From eqn (2), mean {delta}-values for H98D and H98N were 0.81 ± 0.03 (n = 7) and 1.0 ± 0.15 (n = 4), respectively (not significantly different from WT). Mean KD0 values were 34 ± 3.4 (n = 7) and 44 ± 8.8 mM Cs+ (n = 4) for H98D and H98N, respectively (neither of which are significantly different from WT or each other).


   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
References
 
In the present study we have determined the characteristics of cation selectivity and block by Ba2+ and Cs+ in the K2P channel, TASK1. Both Ba2+ and Cs+ block with voltage- and concentration-dependent characteristics. Furthermore, we demonstrate that a pore-neighbouring residue, H98, involved in pH sensing, also affects the permeation pathway.

Cation selectivity

Figure 1 shows that WT channels are K+-selective, and Rb+ and Na+ are essentially impermeant. However, with a single amino acid change, from H98 to D or N, at the extracellular mouth of the pore, TASK1 becomes more permeable to Na+, and particularly to Rb+, in agreement with a previous study by Yuill et al. (2004). This indicates increased access of both ions to the pore. In KcsA, strong K+ selectivity relies upon the co-ordination of the dehydrated K+ ion by the carbonyl backbone of the selectivity filter-lining residues – despite being smaller, Na+ ions are not stabilized because the pore carbonyl groups are not able to substitute effectively for the strongly bound water molecules (Doyle et al. 1998). Thus, it would appear that in H98D, the pore has become rather more flexible, allowing Na+ entry into the channel (see Table 1). As the dehydrated Na+ ion is smaller than K+ (diameter, 1.9 and 2.66 Å, respectively), the pore must have been capable of collapsing somewhat, to effectively accommodate the ion. However, in the case of H98N, the Na+ permeability was unaltered, but the ability to conduct Rb+ (diameter, 2.96 Å) was increased, as was the case for H98D. Thus in both the H98N and H98D mutants the pore must have been able to expand so that Rb+ could co-ordinate with the carbonyl oxygen atoms.

The equivalent residue to TASK1-H98 in KcsA is an aspartate residue (D80), which forms a hydrogen bond with a glutamate residue in the pore helix (E71). Together, they help to maintain the architecture of the selectivity filter. This bond shortens during collapse of the pore when the K+ concentrations is low, which leads to pore occlusion and cessation of ion flux (Zhou et al. 2001). In TASK1, the equivalent residue to E71 is a threonine (T89), which could form a hydrogen bond with H98. N and D would also be capable of hydrogen bond formation, although the altered side chain size and charge may contribute to the observed pore flexibility. Certainly, the highest selectivity was observed in the WT channel, suggesting that this is the most stable arrangement.

It is worth bearing in mind that, unlike voltage-gated and inward rectifier channels, the pore of K2P channels is asymmetrical. The first pore domain has the pore sequence GYG(H/N), whereas the second domain contains G(F/L)GD. This contrasts with the highly conserved GYGD motif of voltage-gated channels. In K2P channels, there is an interaction between the Y of the GYG(H/N) motif and the D of the neighbouring subunit; it is supposed that this deviation from the canonical voltage-gated channel signature sequence allows for fine tuning of ion selectivity (Chapman et al. 2001). In this study, and a previously published study by Yuill et al. (2004), disruption of this interaction in H98 led to impaired K+ selectivity, as predicted by the studies of Chapman et al. (2001).

Ba2+ and Cs+ block

Ba2+ and Cs+ block WT channels in both voltage- and concentration-dependent manners. This indicates that the blocking site is within the membrane field, presumably within the pore of the channel. Block for Ba2+ was quite well described by eqn (2) (Fig. 3A), which suggests that the blocking ion enters the pore, but does not permeate (Hagiwara et al. 1978). As the membrane potential was made more negative, the degree of channel block increased as the positively charged blocking ions were attracted into the pore. For Cs+, channel blockade was so fast that it appeared to be instantaneous, whereas for Ba2+ there was a clear time-dependence of voltage-induced block (Figs 2 and 4).

In the bacterial K+ channel KcsA, Ba2+ binds to a site deep within the selectivity filter, where it is flanked by K+ ions in the outer binding site and cavity binding site (Jiang & Mackinnon, 2000). This finding, based upon X-ray crystallography studies, supported the conclusions from previous single channel studies of Ca2+-activated K+ channels (Neyton & Miller, 1988), suggesting close structural and functional homology between the selectivity filters of KcsA and Ca2+-activated K+ channels. As the pore sequence of TASK1 is similar to that of these channels, it seems reasonable to assume that, at least in WT channels, Ba2+ blocks TASK1 at a similar site to that in KcsA, accounting for the observed voltage-dependence. It is also possible that, at least in WT channels, Cs+ binds to the same site, because the {delta}-values (fraction of the membrane electric field felt by an ion) for Ba2+ and Cs+ are not significantly different (~0.6). H98D was more sensitive to Ba2+ than WT, which is indicative of either easier pore-entry or stronger binding within the pore. However, in H98D, {delta} for Ba2+ block is reduced to 0.16 and is significantly lower than that for H98N or WT (both ~0.6). Thus in H98D, Ba2+ appears to bind with greater efficiency to a site closer to the extracellular mouth of the pore, perhaps by stabilization of Ba2+ at the external K+ binding site within the selectivity filter. Thus the enhanced flexibility of H98D's pore, which results in reduced selectivity, allows stabilization of Ba2+ at one of the outermost cation-binding sites within the selectivity filter, although this is not seen for H98N.

It is also important to realize that in both WT and mutant channels, Ba2+ and Cs+ must be binding within the selectivity filter and not directly to residue 98, as block was voltage-dependent. pH sensing in TASK-1 and -3 occurs at H98, whereas these pH-induced changes in channel activity are essentially voltage-independent as may be expected because this residue lies exterior to the selectivity filter (Duprat et al. 1997; Leonoudakis et al. 1998; Lopes et al. 2000; Morton et al. 2003).

In summary, the pore-neighbouring residue H98 is not only a pH sensor, able to command channel closure in times of extracellular acidic challenge, but is also involved in the maintenance of pore structure.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Chapman ML, Krovetz HS & VanDongen AMJ (2001). GYGD pore motifs in neighbouring potassium channel subunits interact to determine ion selectivity. J Physiol 530, 21–33.[Abstract/Free Full Text]

Doyle DA, Cabral JM, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL et al. (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69–77.[Abstract/Free Full Text]

Duprat F, Lesage F, Fink M, Reyes R, Heurteaux C & Lazdunski M (1997). TASK, a human background K+ channel to sense external pH variations near physiological pH. EMBO J 16, 5464–5471.[CrossRef][Medline]

Goldstein SA, Bockenhauer D, O'Kelly I & Zilberberg N (2001). Potassium leak channels and the KCNK family of two-P-domain subunits. Nat Rev Neurosci 2, 175–184.[Medline]

Hagiwara S, Miyazaki S, Moody WJ & Patlak J (1978). Blocking effects of barium and hydrogen ions on the potassium current during anomalous rectification in the starfish egg. J Physiol 279, 167–185.[Abstract/Free Full Text]

Hille B (2001). Ion Channels of Excitable Membranes, 3rd edn. Sinauer Associates Inc, Sunderland, MA, USA.

Jiang Y & Mackinnon R (2000). The barium site in a potassium channel by x-ray crystallography. J GenPhysiol 115, 269–272.[Abstract/Free Full Text]

Kim D, Fujita A, Horio Y & Kurachi Y (1998). Cloning and functional expression of a novel cardiac two-pore background K+ channel (cTBAK-1). Circ Res 82, 513–518.[Abstract/Free Full Text]

Kim Y, Bang H & Kim D (2000). TASK-3, a new member of the tandem pore K+ channel family. J Biol Chem 275, 9340–9347.[Abstract/Free Full Text]

Leonoudakis D, Gray AT, Winegar BD, Kindler CH, Harada M, Taylor DM et al. (1998). An open rectifier potassium channel with two pore domains in tandem cloned from rat cerebellum. J Neurosci 18, 868–877.[Abstract/Free Full Text]

Lopes CM, Gallagher PG, Buck ME, Butler MH & Goldstein SA (2000). Proton block and voltage gating are potassium dependent in the cardiac leak channel Kcnk3. J Biol Chem 275, 16969–16978.[Abstract/Free Full Text]

Lopes CM, Zliberberg N & Goldstein SA (2001). Block of Konk3 by protons. Evidence that 2-P-domain potassium channel subunits function as homodimers. J Biol Chem 276, 24449–24452.[Abstract/Free Full Text]

Medhurst AD, Rennie G, Chapman CG, Meadows H, Duckworth MD, Kelsell RE et al. (2001). Distribution analysis of human two pore domain potassium channels in tissues of the central nervous system and periphery. Mol Brain Res 86, 101–114.[Medline]

Morton MJ, O'Connell AD, Sivaprasadarao A & Hunter M (2003). Determinants of pH sensing in the two-pore domain K+ channels TASK1 and -2. Pflugers Arch 445, 577–583.[Medline]

Neyton J & Miller C (1988). Discrete Ba2+ block as a probe of ion occupancy and pore structure in the high-conductance Ca2+-activated K+ channel. J Gen Physiol 92, 569–586.[Abstract/Free Full Text]

O'Connell AD, Morton MJ & Hunter M (2002). Two-pore domain K+ channels-molecular sensors. Biochim Biophys Acta 1566, 152–156.[Medline]

Rajan S, Wischmeyer E, Xin LG, Preisig-Muller R, Daut J, Karschin A et al. (2000). TASK-3, a novel tandem pore domain acid-sensitive K+ channel. An extracellular histidine as pH sensor. J Biol Chem 275, 16650–16657.[Abstract/Free Full Text]

Vergara C, Alvarez O & Latorre R (1999). Localization of the K+ lock-in and the Ba2+ binding sites in a voltage-gated calcium-modulated channel. Implications for survival of K+ permeability. J Gen Physiol 114, 365–376.[Abstract/Free Full Text]

Yuill K, Ashmore I & Stanfield PR (2004). The selectivity filter of the tandem pore potassium channel TASK1 and its pH-sensitivity and ionic selectivity. Pflugers Arch 448, 63–69.[CrossRef][Medline]

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 Å resolution. Nature 414, 43–48.[CrossRef][Medline]


   Acknowledgements
 
We are grateful to Dr Steve Goldstein (University of Chicago, IL, USA) for the gift of the TASK1 (KCNK3, OAT1) clone. Thanks also to Dr Thomas W. Claydon (currently at the University of British Columbia, Canada) and Mr Timothy Munsey (University of Leeds, UK) for oocyte injection. We gratefully acknowledge the financial support of the Medical Research Council (UK) and the Wellcome Trust.




This article has been cited by other articles:


Home page
 J. Neurophysiol.Home page
H. Toyoda, M. Saito, H. Sato, Y. Dempo, A. Ohashi, T. Hirai, Y. Maeda, T. Kaneko, and Y. Kang
cGMP Activates a pH-Sensitive Leak K+ Current in the Presumed Cholinergic Neuron of Basal Forebrain
J Neurophysiol, May 1, 2008; 99(5): 2126 - 2133.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
562/3/687    most recent
jphysiol.2004.079020v1
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by O'Connell, A. D
Right arrow Articles by Hunter, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by O'Connell, A. D
Right arrow Articles by Hunter, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS