Am J Physiol Cell Physiol Information on EB 2010
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Cell Physiol 295: C557-C565, 2008. First published June 25, 2008; doi:10.1152/ajpcell.00154.2008
0363-6143/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
295/2/C557    most recent
00154.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 Google Scholar
Google Scholar
Right arrow Articles by Vemana, S.
Right arrow Articles by Larsson, H. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vemana, S.
Right arrow Articles by Larsson, H. P.

MEMBRANE TRANSPORTERS, ION CHANNELS, AND PUMPS

Intracellular Mg2+ is a voltage-dependent pore blocker of HCN channels

Sriharsha Vemana, Shilpi Pandey, and H. Peter Larsson

Neurological Sciences Institute, Oregon Health and Science University, Beaverton, Oregon

Submitted 15 March 2008 ; accepted in final form 22 June 2008


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels are activated by membrane hyperpolarization that creates time-dependent, inward rectifying currents, gated by the movement of the intrinsic voltage sensor S4. However, inward rectification of the HCN currents is not only observed in the time-dependent HCN currents, but also in the instantaneous HCN tail currents. Inward rectification can also be seen in mutant HCN channels that have mainly time-independent currents (5). In the present study, we show that intracellular Mg2+ functions as a voltage-dependent blocker of HCN channels, acting to reduce the outward currents. The affinity of HCN channels for Mg2+ is in the physiological range, with Mg2+ binding with an IC50 of 0.53 mM in HCN2 channels and 0.82 mM in HCN1 channels at +50 mV. The effective electrical distance for the Mg2+ binding site was found to be 0.19 for HCN1 channels, suggesting that the binding site is in the pore. Removing a cysteine in the selectivity filter of HCN1 channels reduced the affinity for Mg2+, suggesting that this residue forms part of the binding site deep within the pore. Our results suggest that Mg2+ acts as a voltage-dependent pore blocker and, therefore, reduces outward currents through HCN channels. The pore-blocking action of Mg2+ may play an important physiological role, especially for the slowly gating HCN2 and HCN4 channels. Mg2+ could potentially block outward hyperpolarizing HCN currents at the plateau of action potentials, thus preventing a premature termination of the action potential.

inward rectifying; hyperpolarization-activated current; instantaneous currents; divalent block; cysteine; hyperpolarization-activated cyclic nucleotide

RHYTHMIC ACTIVITY OF PACEMAKER cells in the heart and thalamic neurons in the brain are dependent on the inward Ih current through hyperpolarization-activated cyclic nucleotide-gated (HCN) channels (29, 36). Following an action potential in pacemaker cells, Ih contributes to the currents that slowly depolarize the membrane potential to threshold, thereby initiating another action potential (7). Four mammalian HCN channels have been cloned: HCN1–HCN4 (11, 21, 35). The deletion of HCN channels, or the presence of naturally occurring mutations in HCN channels, has been shown to have significant physiological consequences (20, 25, 37, 40). HCN2 knockout mice, for example, showed spontaneous absence seizures and cardiac sinus dysrhythmia (20).

HCN channels are members of the superfamily of voltage-gated ion channels, possessing features such as a tetrameric structure, with each subunit containing six transmembrane domains (S1–S6) (29, 36). In addition, HCN channels have a pore domain that shows conservation with voltage-gated potassium (Kv) channels, including a GYG signature motif in the selectivity filter, even though HCN channels are only modestly more selective (3:1) for K+ over Na+ (11, 21). Similar to Kv channels, HCN channels have an intracellular gate at the base of S6 that prevents access of ions to the pore (32, 33). Both Kv and HCN channels have a positively charged S4 domain that acts as the voltage sensor, controlling the opening and closing of the gate in response to membrane voltage (3, 23, 38, 41). Despite the conservation of S4 movement between HCN and Kv channels, the outward S4 movement in response to depolarization has opposite effects in the two types of channels. Depolarization opens the gate of Kv channels but closes the gate of HCN channels (17, 23). This inverse S4-to-the-gate coupling, when compared with Kv channels, and the relative nonselective cation permeability of HCN channels allow HCN channels to mediate a hyperpolarization-activated, depolarizing inward current that contributes to a variety of physiological functions in the body (2, 29).

The hyperpolarization-activation of HCN channels conferred by the S4 movement generates time-dependent, inward rectifying currents through HCN channels. However, during two-electrode recordings of HCN channels (4), we noticed that the instantaneous currents in these channels were also inwardly rectifying. For example, in response to voltage protocols designed to measure the instantaneous tail currents at different voltages following an activation prepulse to –120 mV, the inward instantaneous currents were at least twice as large as the outward instantaneous currents for similarly sized, but opposite, driving forces. In addition, mutations in the S4-S5 loop and S6 that removed the time dependence of the HCN currents, presumably by locking HCN channels in the open state, still displayed time-independent, inward rectifying currents (5). Using excised patches containing well-expressing mammalian HCN1 and HCN2 channels (the other two mammalian HCN channels, HCN3 and HCN4, did not express well enough for experiments of excised patches), we tested whether an intracellular blocking agent is responsible for this rectification or whether this rectification is due to an intrinsic voltage-gating property of HCN channels. On the basis of the following data, we suggest that intracellular magnesium binds in the pore of HCN channels in a voltage-dependent manner, thereby generating inwardly rectifying instantaneous HCN currents.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Molecular biology. The mouse HCN1 and HCN2 channels were used in the present study. The HCN1 channels had the COOH terminus deleted and a small region in the NH2 terminus deleted that maximize the expression in oocytes (42). Most experiments were done on this well-expressing HCN1 channel. However, we also conducted some parallel experiments on HCN2 channels to show that the effects of Mg2+ are also present in other mammalian HCN subtypes. The HCN2 channel had guanine-cytosine-rich domains in the NH2 terminus deleted to facilitate the use of molecular biology techniques on the HCN2 DNA (5). Site-directed mutagenesis was performed on HCN1 channels to create mutations C347S, C347G, and C347A using the QuikChange Kit (Stratagene). The mutation C347T in HCN1 was a generous gift from Dr. Steven Siegelbaum (Columbia University). The mutation R318Q/Y331S in HCN2 was a generous gift from Dr. Michael Sanguinetti (University of Utah). The HCN1 DNA was linearized with Nhe1, and the HCN2 DNA was linearized with EcoR1. RNA was synthesized in vitro using the T7 mMessage mMachine kit (Biocrest). RNA was injected (50 nl of 0.1–1 ng/nl) into Xenopus oocytes, and experiments were performed 2 to 7 days after injection. Extraction of oocytes from Xenopus laevis were conducted in accordance with the American Physiological Society Guiding Principles in the Care and Use of Animals, and protocols were approved by the Institutional Animal Care and Use Committee of Oregon Health and Science University.

Electrophysiology. The macropatch currents were measured with the patch-clamp technique using an Axopatch 200B amplifier (Axon Instruments). The pipette solution consisted of a 100 K solution (in mM): 100 KCl, 1.5 MgCl2, and 10 HEPES (pH = 7.1). In addition, 1 mM BaCl2, 100 µM LaCl3, and 100 µM gadolinium were added to the pipette solution to block endogenous currents including voltage-dependent potassium currents, hyperpolarization-activated chloride currents, and calcium-activated chloride currents. The patch pipettes had a resistance of around 1 M{Omega}. Capacitance compensation was done off-line using subtraction (p/n) protocols with capacitive currents from voltage steps that did not significantly activate HCN channels. The bath solution consisted of a 100 K internal solution (in mM): 100 KCl, 5 NaCl 1 EGTA, 10 HEPES, 1 EDTA, and 100 µM cAMP (pH = 7.1). A separate solution without EDTA was used for the application of MgCl2 to the cytosolic face of the patches. This solution consisted of (in mM): 100 KCl, 5 NaCl, 1 EGTA, and 10 HEPES (pH = 7.1). To this solution, we added ascending concentrations of MgCl2, including 0.25 mM, 0.5 mM, 1 mM, 5 mM, 10 mM, and 20 mM. cAMP (100 µM) was also added to saturate the cAMP-binding site in the HCN channels. An EVH-9 perfusion system (BioLogic Science Instruments) was used to rapidly alternate between the different MgCl2 concentrations applied to the cytosolic side of excised patches. The rapid perfusion system allowed for a complete dose-response experiment to be conducted in less than 40 s, thereby minimizing effects of the rundown of ionic currents seen in excised HCN channels. In addition, the patch was always returned to the original solution at the end of the experiments to measure the amount of rundown during the experiment. Patches with any significant amount of rundown during the experiments were discarded from the analysis. To chelate Mg2+ in whole oocytes, we pressure injected 50 nl of 100 mM EDTA in oocytes using a nanoinjector (Nanoject, Drummond). Based on an estimate of the oocyte volume of 500 nl, the final (after internal diffusion) concentration of EDTA in the cytosol of the oocyte would be close to 10 mM. All experiments were conducted at room temperature.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Instantaneous rectification is not due to an intrinsic voltage-dependent process. Figure 1 shows the currents from HCN1 channels recorded in the cell-attached mode from Xenopus oocytes with a 100 mM K solution in the pipette. The HCN currents exhibit the characteristic hyperpolarization-activated, time-dependent currents caused by HCN channel opening in response to negative voltage steps and the time-dependent closing in response to a depolarizing voltage step (Fig. 1A). A closer inspection of the currents, however, revealed that the amplitude of the outward currents was smaller than expected. The outward tail currents at +50 mV were not proportional to the inward currents, when normalized to the driving force at the different voltage steps (Fig. 1, A and B). For example, a voltage step to –130 mV caused a –135 pA of inward currents, but the subsequent tail currents at +50 mV was only 20 pA [instead of the expected 51 pA (= 50 mV/130 mV·135 pA) of tail currents in a 100 K external solution; Fig. 1B]. The smaller than expected tail currents suggest that there is some inward rectification of the instantaneous currents in HCN channels. To further visualize the inward rectification, we stepped to a negative potential (–130 mV) to open all HCN channels, followed by voltage steps to various potentials to record both outward and inward instantaneous currents (Fig. 1, C and D). Inward rectification is clearly evident in the plots of the instantaneous HCN1 current amplitude against voltage (Fig. 1E). HCN2 channels exhibited a similar inward rectification in the instantaneous currents (Fig. 1F).


Figure 1
View larger version (30K):
[in this window]
[in a new window]

  		Fig. 1. Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels show rectification in cell-attached patches. A: HCN1 currents elicited by voltage steps from –30 mV to –130 mV (in –10-mV increments), from a holding potential of 0 mV, followed by a step to +50 for tail currents. B: close-up of tail currents from A. C: HCN1 tail currents elicited by voltage steps from +110 mV to –115 mV (–15-mV increments), following a prepulse to –130 mV. Holding potential = 0 mV. D: close-up of tail currents from C. E: current versus voltage (I/V) plot measured at the arrow in D ({blacksquare}). The dashed line is the best linear fit to the currents for voltages between –115 and –40 mV. Inward rectification is clearly seen in the instantaneous HCN1 currents. F: I/V plot for HCN2 channels measured as in D ({blacksquare}). Inward rectification is clearly seen in the instantaneous HCN2 currents.

 
To test whether this inward rectification is due to an intracellular factor in the oocytes, we measured HCN1 and HCN2 currents in cell-free, excised patches. Upon excision of HCN1 and HCN2 patches into a 100 K internal solution (100 KCl, 5 NaCl 1 EGTA, 10 HEPES, 1 EDTA, and 100 µM cAMP, pH = 7.1), the size of the tail currents increased approximately twofold: 122 ± 24% (n = 5; Fig. 2, A and B) in HCN1 channels and 148 ± 63% (n = 6) in HCN2 channels (Fig. 2, C and D). The increase in tail currents was not accompanied by any measurable increase in the inward currents (insets, Fig. 2, A–D). The instantaneous currents in excised patches had a linear current versus voltage (I/V) in this intracellular solution and lacked the inward rectification found in cell-attached patches (Fig. 2, E and F), suggesting that an intracellular agent is responsible for the inward rectification in intact oocytes. To further test whether an intracellular agent within the oocyte was responsible for the observed changes in tail amplitudes upon excision, we returned the excised patch back into the intracellular environment within the oocyte by using patch cramming (i.e., forcing the patch into the oocyte cytosol thereby exposing the intracellular side of the membrane again to the oocyte cytosol) (16). Upon patch cramming, the outward tail currents returned immediately to their smaller size (Fig. 3, C and D) without significantly altering the inward currents (see insets, Fig. 3), and the instantaneous currents once again became inwardly rectifying. The linear instantaneous I/V in excised patches suggests that it is not an intrinsic voltage-dependent process in HCN channels that causes the inward rectification in the instantaneous currents, but that an intracellular agent is responsible for the reduction in amplitude of the outward currents.


Figure 2
View larger version (37K):
[in this window]
[in a new window]

  		Fig. 2. HCN1 and HCN2 tail currents increase in amplitude after excision. A and B: close-ups of HCN1 tail currents in cell-attached (A) and excised (B) patches elicited by voltage steps from –30 mV to –140 mV (in –10-mV increments), followed by a step to +50 for tail currents. Currents during the hyperpolarized voltage pulses are shown in insets. Holding potential = 0 mV. C and D: close-ups of HCN2 tail currents in cell-attached (C) and excised (D) patches elicited by the same protocol as in A. Currents during the hyperpolarized voltage pulses are shown in insets. E: HCN2 currents in an excised patch in response to a –120-mV activation prepulse followed by voltage steps from +50 mV to –115 mV (in –15-mV increments). F: plot of current versus voltage for HCN2 tail currents (measured at the arrow in E) after patch excision. The I/V plot was fit with a straight line.

 

Figure 3
View larger version (35K):
[in this window]
[in a new window]

  		Fig. 3. Patch cramming reduces the amplitude of the outward currents back to cell-attached levels. A–C: tail currents from HCN1 channels elicited by voltage steps from –30 mV to –140 mV (in –10-mV increments), followed by a step to +50 for tail currents. Currents during the hyperpolarized voltage pulses are shown in insets. Holding potential = 0 mV. A: cell-attached patch. B: excised patch. C: patch crammed into the cytoplasmic environment of the Xenopus oocyte. D: a plot of tail current amplitude versus the prepulse voltage in a cell-attached patch ({blacksquare}), an excised patch (bullet), and a patch inserted back into the Xenopus oocyte ({blacktriangleup}). The approximately –10-mV shift seen between the cell-attached patch and the patch reinserted back into the Xenopus oocyte is most likely due to degradation of some of the membrane-bound phosphatidylinositol 4,5-bisphosphate, which has been shown to shift the voltage dependence of HCN channels by >20 mV (26, 47).

 
Intracellular Mg2+ blocks HCN currents at physiological concentrations. We next tested different cytosolic agents that could potentially cause the inward rectification of the instantaneous currents in HCN channels. In inward rectifying K+ channels, it has been shown that the inward rectification is mainly due to a voltage-dependent block by intracellular Mg2+ and polyamines, such as spermidine and spermine (18, 24). HCN channels play an important role as pacemaker channels in cardiac myocytes, such as in the sinoatrial node. The cytosol of cardiac myocytes contains approximate 1 mM free Mg2+ and 10 µM free polyamines, such as spermidine and spermine (45). We therefore tested the effect of applying physiological concentrations of Mg2+, spermidine, and spermine to the cytosolic side of excised patches to determine whether Mg2+ and polyamines block the outward currents through HCN channels (Fig. 4). Both the application of 10 µM spermine and the application of 10 µM spermidine blocked the outward currents through HCN1 channels. Spermine (10 µM) blocked the currents at +50 mV by 9 ± 3.6% (n = 3), and 10 µM spermidine blocked the currents at +50 mV by 16.3 ± 4.0% (n = 3). However, the application of 1 mM Mg2+ blocked the currents at +50 mV by 38.3 ± 7.2% (n = 3) in HCN1 channels (Fig. 4, B and C), while not significantly decreasing the inward currents during the step to –130 mV (Fig. 4A). That physiological concentrations of Mg2+ had the most potent block of the outward instantaneous currents suggests that it is mainly Mg2+ that causes inward rectification of the instantaneous HCN currents in intact cells.


Figure 4
View larger version (35K):
[in this window]
[in a new window]

  		Fig. 4. Physiological concentrations of intracellular magnesium block HCN outward current more than spermine or spermidine. A: HCN currents elicited by a –130-mV step from a holding potential of 0 mV, followed by a step to +50 mV for tail currents, during perfusion of control solutions, 10 µM spermidine, 10 µM spermine, and 1 mM Mg2+. B: enlargement of the tail currents showing the four sweeps where control solution, 10 µM spermidine, 10 µM spermine, and 1 mM Mg2+ were applied. Between every trace where a potential blocker was applied, an application of 100 K control solution was applied (data not shown). C: a quantitative summary of the %fraction of tail current blocked for Mg2+, spermine, and spermidine. The block by spermine and spermidine was significantly less than the block by Mg2+ (P < 0.01, paired t-test). D: average HCN2 tail currents following an activation step to –130 mV, in cell-attached mode ({blacksquare}), excised in EDTA solution ({circ}), and patch crammed into an oocyte injected with 50 nl of 100 mM EDTA ({blacktriangleup}). Norm, normalized; i-o, inside out.

 
To further test whether intracellular Mg2+ causes the rectification of HCN currents, we crammed patches containing HCN2 channels into oocytes that had been injected with high concentrations of EDTA to chelate intracellular Mg2+. As shown in Fig. 3 for HCN1 channels, excising a patch into a solution containing EDTA increased the HCN2 tail currents at depolarized voltages and removed the rectification of the HCN2 channel currents (Fig. 4D, squares to circles). Patch cramming an excised patch from a bath solution containing EDTA into control oocytes reduced the currents and returned the currents in the patch-crammed mode to the levels of the currents measured in cell-attached mode (n = 3). In contrast, patch cramming an excised patch into oocytes that had been injected with EDTA reduced the HCN2 channel currents by only 8 ± 15% (n = 3). The currents in the patch-crammed mode were 214 ± 65% larger than the currents measured in cell-attached mode (n = 3; Fig. 4D). The effect of intracellular EDTA suggests that the rectification of HCN channel currents in intact cells is mainly due to intracellular Mg2+. The small, remaining reduction in the tail currents, when cramming a patch into an oocytes injected with EDTA, is most likely due to the block by intracellular polyamines, such as spermidine and spermine (Fig. 4B).

Mg2+ blocks HCN currents in a voltage-dependent manner. We next measured the affinity of Mg2+ for HCN1 and HCN2 channels by applying different concentrations of Mg2+ and measuring the size of the outward tail currents at +50 mV, following an activation prepulse to –150 mV to open all channels. The tail currents decreased for increasing Mg2+ concentrations including (in mM) 0.25, 0.5, 1, 5, 10, and 20 (Fig. 5, A and B). The amplitudes of the inward currents (at –150 mV) were also reduced for the highest concentrations of Mg2+ (10 and 20 mM; Fig. 5A). The dose-response curve was well fit with a Michaelis-Menton curve with a Hill coefficient of 1.11 ± 0.04 (n = 3). The IC50 was 0.82 ± 0.27 mM at +50 mV (n = 3) for HCN1 channels (Fig. 5C), and the IC50 was 0.53 ± 0.12 mM (n = 3) at +50 mV for HCN2 channels (Fig. 5D).


Figure 5
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 5. Mg2+ blocks HCN outward currents at physiological concentrations. A: to test Mg2+ block of HCN1 outward currents, different concentrations of Mg2+ were applied while the currents were recorded in response to a voltage step to –150 mV from a holding potential of 0 mV, followed by a voltage step to +50 mV for tails. No Mg2+ was applied during the first sweep, followed in succession by different concentrations of Mg2+ (in mM): 0.25, 0.5, 1, 5, 10, and 20. B: a close-up of the tail currents shows the concentration-dependent block of the outward tail currents in response to increasing concentrations of Mg2+. The dashed line shows the current when returned to 0 Mg2+. The arrow marks the time point at which values were taken to create a dose-response curve for Mg2+ block of an HCN1 channel. The remaining currents in the high Mg2+ concentrations are most likely due to endogenous oocyte currents from hyperpolarization-activated chloride channels that close during the depolarizing voltage step. C: dose-response curve for Mg2+ block of the currents through HCN1 channels at +50 mV. The values were fit with the equation: I(Mg2+)/Imax = 1/(1 + Km/[Mg2+]) + C, where Imax is the current in 0 Mg2+ and C is the current in saturating Mg2+ concentrations. D: dose-response curve for Mg2+ block of the currents through HCN2 channels at +50 mV, from similar experiment as in A.

 
The rectification induced by intracellular Mg2+ (Fig. 1) suggests that Mg2+ binding is voltage dependent. The voltage dependence of Mg2+ block was determined by measuring the IC50 at different tail potentials for HCN1 channels (Fig. 6). The affinity for Mg2+ was clearly voltage dependent: the IC50 was 0.81 ± 0.23 mM (n = 3) at +80 mV and 1.8 ± 0.1 mM (n = 3) at +20 mV (P < 0.05; Fig. 6A). The Mg2+ affinity versus voltage was fit to the Woodhull equation (44), K1/2(V) = K1/2(0 mV) exp(–{delta}zV/kT). The electrical distance ({delta}) was found to be 0.19 ± 0.06 (n = 3) and the Mg2+ affinity at 0 mV, K1/2 (0 mV), was found to be 2.3 ± 0.47 mM (n = 3) (Fig. 6B). The K1/2 (0 mV) and the relatively shallow voltage dependence for Mg2+ block ({delta} = 0.19) are consistent with the amount of block of the inward currents at –150 mV seen at higher concentrations of Mg2+ (Fig. 5A). The voltage dependence of the affinity is consistent with Mg2+ blocking HCN channels in a voltage-dependent manner.


Figure 6
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 6. Voltage dependence of Mg2+ block. A: seven different concentrations of Mg2+ were applied while the currents were recorded from HCN1 channels in response to a voltage step to –150 mV, followed by a voltage step to +20 mV, +50 mV, or +80 mV. Holding potential = 0 mV. The order of application for Mg2+ was (in mM) 0, 0.25, 0.5, 1, 5, 10, and 20. Current amplitudes from time points during the tail currents were used (as in Fig. 5) to create dose-response curves for each of the three different tail voltages applied. B: the IC50 values for the Mg2+ block of the tail currents at +20 mV, +50 mV, and +80 mV plotted and fit to the Woodhull equation: K1/2(V) = K1/2(0 mV) exp(–{delta}zV/kT). The fraction of the membrane electric field traversed by the Mg2+ ion was {delta} = 0.19. The intersection of the fitted line with the y-axis gave the Km value at 0 mV: K1/2(0 mV) = 2.3 mM.

 
The effect of a pore mutation suggests that Mg2+ binds to the selectivity filter. The voltage dependence of Mg2+ block could be explained by Mg2+ binding to a site in the pore of the HCN channels. Divalent ions have been shown to block currents in other channels by binding to sites in or near the selectivity filter in the pore. For example, in the KcsA channels, it has been shown by X-ray crystallography that Ba2+ binds to a binding site just below the GYG motif in the selectivity filter (14). In HCN channels, Cd2+ has been shown to bind to a cysteine (C347 in HCN1) located two amino acids before the GYG (31). In the KcsA crystal structure, the four homologous residues to this cysteine, one from each subunit, are very close together in space and could easily form an ion-binding site (Fig. 9). In contrast, the acidic residue in S6 that has been identified to bind Mg2+ in inward rectifying K (Kir) channels (19, 43) is not conserved in HCN channels. We therefore tested whether mutations of C347 would affect Mg2+ binding to HCN channels.


Figure 9
View larger version (53K):
[in this window]
[in a new window]

  		Fig. 9. Putative location of the Mg2+ binding site in the pore. Model for the putative binding site for Mg2+ in a homology model for the HCN2 channel [modeled after the KcsA crystal structure (12)], with the four C347 residues highlighted. A: side view with only two subunits shown. C347 is shown in space fill. B: view from the cytosolic side, showing the close proximity of the four C347 side chains in the model. C: proposed Mg2+ binding site formed by the four C347 at the electrical distance, {delta}, from the cytosol. The dipoles (arrows) formed by the pore helices may also contribute to the binding affinity for divalent ions, such as Mg2+, to this site.

 
The mutations C347S and C347T in the HCN1 channel were still susceptible to block by intracellular Mg2+ (Fig. 7). However, the Mg2+ affinities of the C347S and C347T channels were reduced significantly to an IC50 = 2.8 ± 0.3 mM (n = 3, P < 0.01) and an IC50 = 3.9 ± 1.0 mM (n = 3, P < 0.01) at +50 mV, respectively. Mutating C347 to a more radically different residue, such as to small hydrophobic alanine or glycine residues, to further test whether residue 347 contributes to the Mg2+ binding site, resulted in nonfunctional channels. However, the change in affinity with the serine and threonine substitutions at C347 is consistent with our hypothesis that C347 is part of the Mg2+ binding site, deep within the pore domain.


Figure 7
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 7. The mutants C347S and C347T altered the magnesium affinity in HCN1 channels. Dose-response curve at +50 mV for a representative patch with wild-type HCN1 channels ({blacksquare}), C347S HCN1 channels ({circ}), and C347S HCN1 channels ({blacktriangleup}) measured as in Fig. 5. The data were fitted with the Michaelis-Menton equation (gray line for C347S, black lines for wild type and 347T).

 
Magnesium causes inward rectification in a permanently open HCN channel. In a previous study, the double mutation R318Q/Y331S in HCN2 channels was shown to generate ionic currents with no time-dependent component in response to hyperpolarized or depolarized voltage steps (5). Since these channels did not display any time dependence in their currents in response to hyperpolarizations or depolarizations, the mutant channel was assumed to be in a permanently open state. The authors suggested that mutation R318Q in S4 prevents S4 from moving, whereas mutation Y331S in the S4-S5 loop decouples S4 from the gate, thereby creating a permanently open channel. However, HCN2 R318Q/Y331S channels still exhibit inward rectifying currents, with larger inward currents than outward currents (Fig. 8A). The currents from HCN2 R318Q/Y331S channels look very similar to the inward rectifying currents through Kir channels. We tested whether this inward rectification was mainly due to intracellular Mg2+ block of the outward currents. Upon patch excision, the outward currents exhibit less rectification, and the current/voltage relationship becomes almost linear through all negative and positive potentials (Fig. 8, B and D), reminiscent of the increase in tail amplitudes seen for wild-type HCN1 and HCN2 channels after patch excision. We attribute the remaining rectification in R318Q/Y331S channels to a small time-dependent voltage activation gated by S4 movement. Application of 1 mM Mg2+ was able to restore inward rectification to the currents from R318Q/Y331S channels in excised patches (Fig. 8, C and D). This ability of Mg2+ to block the outward currents even in a permanently open HCN channel further suggests that Mg2+ acts as an extrinsic inward rectifier for outward currents in HCN channels.


Figure 8
View larger version (40K):
[in this window]
[in a new window]

  		Fig. 8. Magnesium causes inward rectification in a permanently open HCN channel. A–C: current from HCN2 R318Q/Y331S channels in response to voltage steps from +80 mV to –110 mV (in –10-mV increments) from a holding potential of 0 mV, followed by a step to +50 mV. A: cell-attached patch. B: excised patch with 0 Mg2+ applied. C: excised patch with 1 mM Mg2+ applied. D: I/V curves from A–C showing inward rectification for ionic currents in cell-attached patches ({blacksquare}). The rectification is decreased in an excised patch exposed to 0 Mg2+ ({circ}). The rectification is restored, however, on application of 1 mM Mg2+ ({blacktriangleup}).

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Using cell-attached and excised patches with HCN1 and HCN2 channels, we have here shown that HCN channels are susceptible to an "extrinsic" rectification in which intracellular Mg2+ blocks the outward currents of HCN channels at positive potentials. Mg2+ blocks these outward currents significantly at physiological concentrations with an IC50 of 0.53 mM for HCN2 channels and 0.81 mM for HCN1 channels at +50 mV. The concentration of free, unbound Mg2+ has been estimated to be between 0.5 and 1.5 mM in the intracellular environment of mammalian cells (1, 6, 13, 15, 30), suggesting that the Mg2+ block of HCN channels is physiological. Lyashchenko and Tibbs (22) recently (in an article that was published during the final preparation of our manuscript) concluded that intracellular Mg2+ blocks outward HCN currents in symmetrical K solutions, as well as in more physiological solutions (22). Similar to Lyashchenko and Tibbs (22), we find that polyamines, at physiological concentrations, contribute less than Mg2+ to the rectification of HCN channels. Intracellular Mg2+ concentrations in Xenopus oocytes have been estimated to be 0.3 mM (10). In our cell-attached experiments on Xenopus oocytes, we observed that HCN channels with Km in the range from 0.53 mM to 0.81 mM were more than 50% blocked, suggesting that either the Mg2+ concentrations in Xenopus oocytes have previously been underestimated or that Mg2+ is concentrated at the intracellular mouth of HCN channels by some unknown mechanism. Lyashchenko and Tibbs (22) speculated that the presence of labile intracellular anionic lipids, such as free fatty acids, raises the Mg2+ affinity of HCN channels.

The voltage dependence of the Mg2+ block suggests a binding site within the transmembrane electric field of HCN channels, presumably in the pore. In the open pore of Kv channels, it is assumed that 80% of the transmembrane electric field primarily lies over the selectivity filter (8, 46). Therefore, the {delta}-value of 0.19 obtained for Mg2+ of HCN1 channels suggests that the binding site for Mg2+ is located internally to the selectivity filter in HCN channels. This location for the Mg2+ binding site in HCN channels is consistent with {delta}-values obtained from other channels, for which the binding site for divalent cations such as Mg2+ and Ba2+ lies in, or close to, the selectivity filter. For instance, in small-conductance Ca2+-activated K+ channel SK2, divalent ions bind with a {delta}-value of 0.39 to a residue in the selectivity filter (39) and, in KcsA, Ba2+ binds to the selectivity filter with a {delta}-value of 0.3 (14). Our mutagenesis of a conserved cysteine located two residues intracellular to the GYG selectivity sequence supports this location for the Mg2+ binding site in HCN channels. We showed that mutations of this cysteine altered the affinity of the binding site, suggesting that this cysteine forms part of the binding site. The four cysteines, one from each subunit, form a ring of cysteines at the cytosolic entrance to the selectivity filter (12) and can easily be envisioned to create a binding site for divalent cations (Fig. 9). This cysteine is homologous to a residue in SK channels that has been shown to bind divalent ions in a voltage dependent manner in SK channels (39). In addition, Vincent Torre's group showed that the same cysteine binds Cd2+ in HCN1 channels and that these cysteines can spontaneously crosslink with disulfide bonds, suggesting that these four cysteines are located close together (31). Individual sulfur atoms are assumed to bind Mg2+ poorly (9). However, four symmetrical located cysteines close together at the entrance to the selectivity filter in HCN channels might form a low-affinity divalent binding site, especially if one or more of the four cysteines would be deprotonated and, thereby, negatively charged. Mutating C347 to a serine or a threonine did change the affinity for Mg2+ binding, but did not change binding affinity drastically. However, serine and threonine have hydroxyl side chains that could also create a good binding environment for the Mg2+. We were unable to mutate C347 to a more radically different residue, such as a small hydrophobic alanine or glycine residues, to further test whether C347 contributes to the Mg2+ binding site. However, the changes in affinity with the serine substitution are consistent with our hypothesis that C347 is part of the Mg2+ binding site, deep within the pore domain. The C347 residues are located at the COOH-terminal end of the pore helices, which have electrical dipoles associated with them that have been suggested to stabilize cations in the pore (34). These electrical dipoles point their negative poles towards the C347 residues and may also contribute to the binding affinity for Mg2+ at this site (Fig. 9). Other residues in the pore cavity might also participate in the coordination of the Mg2+. It is also possible that mutations of C347 may indirectly affect the binding of Mg2+ and that the Mg2+ binding site is located, for example, deeper in the selectivity filter.

The voltage-dependent Mg2+ block described in the present study for HCN channels creates an extrinsic inward rectification of the instantaneous HCN currents. This is most clearly seen in the Mg2+ block of the currents through permanently lock-open HCN channels, which gives rise to currents similar to those through inward rectifying Kir channels. Kir channel gating properties rely mainly on the voltage-dependent block by intracellular Mg2+ and polyamines (17a). However, HCN channels also possess a major "intrinsic" gating mechanism in the form of the S4 voltage sensor, which acts to open and close an intracellular activation gate of HCN channels in response to voltage (23). This voltage-dependent opening and closing of the activation gate is generally assumed to constitute the major voltage-dependent gating mechanism in HCN channels, underlying, for example, the physiological role of HCN channels in pacemaker cells. However, HCN channels exhibit a nonnegligible instantaneous current (27), suggested to be contributed by a subpopulation of HCN channels that are constitutively open and do not close in response to depolarizations (28). This "instantaneous" current, or leak current, through HCN channels, has been proposed to be physiologically important by, for example, increasing the rate of repolarization of plateau action potentials in the heart (28). The voltage-dependent Mg2+ block described here for HCN channels creates an extrinsic inward rectification of these instantaneous currents and could prevent too much outward K+ flow during plateau action potentials that could otherwise prematurely terminate the action potential by this hyperpolarizing K+ current. In addition, some of the members in the HCN channel family have very slow kinetics. For example, HCN4 channels open and close with time constants in the range of seconds (36). These channels would not have time to all close during a depolarizing action potential. For these slower HCN channels, the Mg2+ block might be a faster, secondary voltage-dependent gating mechanism to reduce outward currents through HCN channels at depolarized potentials. Therefore, the Mg2+ block in HCN channels might be a mechanism to reduce unwanted outward currents through open HCN channels at prolonged depolarized potentials, such as plateau action potentials.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by National Institutes of Health Grants NS-043259 and NS-051169.


   ACKNOWLEDGMENTS
 
We thank Drs. Andrew Bruening-Wright, John Adelman, Lane Brown, James Maylie, and Fredrik Elinder for comments and suggestions on the manuscript.


   FOOTNOTES
 

Address for reprint requests and other correspondence: H. P. Larsson, Neurological Sciences Institute, Oregon Health & Science Univ., 505 NW 185th Ave., Beaverton, OR 97006 (e-mail: larssonp{at}ohsu.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
1. Almulla HA, Bush PG, Steele MG, Ellis D, Flatman PW. Loading rat heart myocytes with Mg2+ using low-[Na+] solutions. J Physiol 575: 443–454, 2006.[Abstract/Free Full Text]

2. Baruscotti M, Bucchi A, Difrancesco D. Physiology and pharmacology of the cardiac pacemaker ("funny") current. Pharmacol Ther 107: 59–79, 2005.[CrossRef][Web of Science][Medline]

3. Bell DC, Yao H, Saenger RC, Riley JH, Siegelbaum SA. Changes in local S4 environment provide a voltage-sensing mechanism for mammalian hyperpolarization-activated HCN channels. J Gen Physiol 123: 5–19, 2004.[CrossRef][Web of Science][Medline]

4. Bruening-Wright A, Elinder F, Larsson HP. Kinetic relationship between the voltage sensor and the activation gate in spHCN channels. J Gen Physiol 130: 71–81, 2007.[Abstract/Free Full Text]

5. Chen J, Mitcheson JS, Tristani-Firouzi M, Lin M, Sanguinetti MC. The S4–S5 linker couples voltage sensing and activation of pacemaker channels. Proc Natl Acad Sci USA 98: 11277–11282, 2001.[Abstract/Free Full Text]

6. Csernoch L, Bernengo JC, Szentesi P, Jacquemond V. Measurements of intracellular Mg2+ concentration in mouse skeletal muscle fibers with the fluorescent indicator mag-indo-1. Biophys J 75: 957–967, 1998.[Web of Science][Medline]

7. DiFrancesco D. Pacemaker mechanisms in cardiac tissue. Annu Rev Physiol 55: 455–472, 1993.[CrossRef][Web of Science][Medline]

8. Doyle DA, Morais Cabral J, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT, MacKinnon R. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280: 69–77, 1998.[Abstract/Free Full Text]

9. Dudev T, Lin YL, Dudev M, Lim C. First-second shell interactions in metal binding sites in proteins: a PDB survey and DFT/CDM calculations. J Am Chem Soc 125: 3168–3180, 2003.[CrossRef][Web of Science][Medline]

10. Gabriel TE, Gunzel D. Quantification of Mg2+ extrusion and cytosolic Mg2+-buffering in Xenopus oocytes. Arch Biochem Biophys 458: 3–15, 2007.[CrossRef][Web of Science][Medline]

11. Gauss R, Seifert R, Kaupp UB. Molecular identification of a hyperpolarization-activated channel in sea urchin sperm. Nature 393: 583–587, 1998.[CrossRef][Web of Science][Medline]

12. Giorgetti A, Carloni P, Mistrik P, Torre V. A homology model of the pore region of HCN channels. Biophys J 89: 932–944, 2005.[CrossRef][Web of Science][Medline]

13. Handy RD, Gow IF, Ellis D, Flatman PW. Na-dependent regulation of intracellular free magnesium concentration in isolated rat ventricular myocytes. J Mol Cell Cardiol 28: 1641–1651, 1996.[CrossRef][Web of Science][Medline]

14. Jiang Y, MacKinnon R. The barium site in a potassium channel by x-ray crystallography. J Gen Physiol 115: 269–272, 2000.[Abstract/Free Full Text]

15. Kim SJ, Kang HS, Lee MY, Lee SJ, Seol JW, Park SY, Kim IS, Kim NS, Kim SZ, Kwak YG, Kim JS. Ketamine-induced cardiac depression is associated with increase in [Mg2+]i and activation of p38 MAP kinase and ERK 1/2 in guinea pig. Biochem Biophys Res Commun 349: 716–722, 2006.[CrossRef][Web of Science][Medline]

16. Kramer RH. Patch cramming: monitoring intracellular messengers in intact cells with membrane patches containing detector ion channels. Neuron 4: 335–341, 1990.[CrossRef][Medline]

17. Larsson HP. The search is on for the voltage sensor-to-gate coupling. J Gen Physiol 120: 475–481, 2002.[Free Full Text]

17a. Lopatin AN, Makhina EN, Nichols CG. Potassium channel block by cytoplasmic polyamines as the mechanism of intrinsic rectification. Nature 372: 366–369, 1994.[CrossRef][Web of Science][Medline]

18. Lu Z. Mechanism of rectification in inward-rectifier K+ channels. Annu Rev Physiol 66: 103–129, 2004.[CrossRef][Web of Science][Medline]

19. Lu Z, MacKinnon R. Electrostatic tuning of Mg2+ affinity in an inward-rectifier K+ channel. Nature 371: 243–246, 1994.[CrossRef][Web of Science][Medline]

20. Ludwig A, Budde T, Stieber J, Moosmang S, Wahl C, Holthoff K, Langebartels A, Wotjak C, Munsch T, Zong X, Feil S, Feil R, Lancel M, Chien KR, Konnerth A, Pape HC, Biel M, Hofmann F. Absence epilepsy and sinus dysrhythmia in mice lacking the pacemaker channel HCN2. EMBO J 22: 216–224, 2003.[CrossRef][Web of Science][Medline]

21. Ludwig A, Zong X, Jeglitsch M, Hofmann F, Biel M. A family of hyperpolarization-activated mammalian cation channels. Nature 393: 587–591, 1998.[CrossRef][Web of Science][Medline]

22. Lyashchenko AK, Tibbs GR. Ion binding in the open HCN pacemaker channel pore: fast mechanisms to shape "slow" channels. J Gen Physiol 131: 227–243, 2008.[Abstract/Free Full Text]

23. Mannikko R, Elinder F, Larsson HP. Voltage-sensing mechanism is conserved among ion channels gated by opposite voltages. Nature 419: 837–841, 2002.[CrossRef][Web of Science][Medline]

24. Nichols CG, Lopatin AN. Inward rectifier potassium channels. Annu Rev Physiol 59: 171–191, 1997.[CrossRef][Web of Science][Medline]

25. Nolan MF, Malleret G, Lee KH, Gibbs E, Dudman JT, Santoro B, Yin D, Thompson RF, Siegelbaum SA, Kandel ER, Morozov A. The hyperpolarization-activated HCN1 channel is important for motor learning and neuronal integration by cerebellar Purkinje cells. Cell 115: 551–564, 2003.[CrossRef][Web of Science][Medline]

26. Pian P, Bucchi A, Robinson RB, Siegelbaum SA. Regulation of gating and rundown of HCN hyperpolarization-activated channels by exogenous and endogenous PIP2. J Gen Physiol 128: 593–604, 2006.[Abstract/Free Full Text]

27. Proenza C, Angoli D, Agranovich E, Macri V, Accili EA. Pacemaker channels produce an instantaneous current. J Biol Chem 277: 5101–5109, 2002.[Abstract/Free Full Text]

28. Proenza C, Yellen G. Distinct populations of HCN pacemaker channels produce voltage-dependent and voltage-independent currents. J Gen Physiol 127: 183–190, 2006.[Abstract/Free Full Text]

29. Robinson RB, Siegelbaum SA. Hyperpolarization-activated cation currents: from molecules to physiological function. Annu Rev Physiol 65: 453–480, 2003.[CrossRef][Web of Science][Medline]

30. Romani A. Regulation of magnesium homeostasis and transport in mammalian cells. Arch Biochem Biophys 458: 90–102, 2007.[CrossRef][Web of Science][Medline]

31. Roncaglia P, Mistrik P, Torre V. Pore topology of the hyperpolarization-activated cyclic nucleotide-gated channel from sea urchin sperm. Biophys J 83: 1953–1964, 2002.[Web of Science][Medline]

32. Rothberg BS, Shin KS, Phale PS, Yellen G. Voltage-controlled gating at the intracellular entrance to a hyperpolarization-activated cation channel. J Gen Physiol 119: 83–91, 2002.[Abstract/Free Full Text]

33. Rothberg BS, Shin KS, Yellen G. Movements near the gate of a hyperpolarization-activated cation channel. J Gen Physiol 122: 501–510, 2003.[Abstract/Free Full Text]

34. Roux B, MacKinnon R. The cavity and pore helices in the KcsA K+ channel: electrostatic stabilization of monovalent cations. Science 285: 100–102, 1999.[Abstract/Free Full Text]

35. Santoro B, Liu DT, Yao H, Bartsch D, Kandel ER, Siegelbaum SA, Tibbs GR. Identification of a gene encoding a hyperpolarization-activated pacemaker channel of brain. Cell 93: 717–729, 1998.[CrossRef][Web of Science][Medline]

36. Santoro B, Tibbs GR. The HCN gene family: molecular basis of the hyperpolarization-activated pacemaker channels. Ann NY Acad Sci 868: 741–764, 1999.[CrossRef][Web of Science][Medline]

37. Schulze-Bahr E, Neu A, Friederich P, Kaupp UB, Breithardt G, Pongs O, Isbrandt D. Pacemaker channel dysfunction in a patient with sinus node disease. J Clin Invest 111: 1537–1545, 2003.[CrossRef][Web of Science][Medline]

38. Sesti F, Rajan S, Gonzalez-Colaso R, Nikolaeva N, Goldstein SA. Hyperpolarization moves S4 sensors inward to open MVP, a methanococcal voltage-gated potassium channel. Nat Neurosci 6: 353–361, 2003.[CrossRef][Web of Science][Medline]

39. Soh H, Park CS. Localization of divalent cation-binding site in the pore of a small conductance Ca2+-activated K+ channel and its role in determining current-voltage relationship. Biophys J 83: 2528–2538., 2002.[Web of Science][Medline]

40. Ueda K, Nakamura K, Hayashi T, Inagaki N, Takahashi M, Arimura T, Morita H, Higashiuesato Y, Hirano Y, Yasunami M, Takishita S, Yamashina A, Ohe T, Sunamori M, Hiraoka M, Kimura A. Functional characterization of a trafficking-defective HCN4 mutation, D553N, associated with cardiac arrhythmia. J Biol Chem 279: 27194–27198, 2004.[Abstract/Free Full Text]

41. Vemana S, Pandey S, Larsson HP. S4 movement in a mammalian HCN channel. J Gen Physiol 123: 21–32, 2004.[CrossRef][Web of Science][Medline]

42. Wainger BJ, DeGennaro M, Santoro B, Siegelbaum SA, Tibbs GR. Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411: 805–810, 2001.[CrossRef][Web of Science][Medline]

43. Wible BA, Taglialatela M, Ficker E, Brown AM. Gating of inwardly rectifying K+ channels localized to a single negatively charged residue. Nature 371: 246–249, 1994.[CrossRef][Web of Science][Medline]

44. Woodhull AM. Ionic blockage of sodium channels in nerve. J Gen Physiol 61: 687–708, 1973.[Abstract/Free Full Text]

45. Yan DH, Nishimura K, Yoshida K, Nakahira K, Ehara T, Igarashi K, Ishihara K. Different intracellular polyamine concentrations underlie the difference in the inward rectifier K+ currents in atria and ventricles of the guinea-pig heart. J Physiol 563: 713–724, 2005.[Abstract/Free Full Text]

46. Yellen G, Jurman ME, Abramson T, MacKinnon R. Mutations affecting internal TEA blockade identify the probable pore-forming region of a K+ channel. Science 251: 939–942, 1991.[Abstract/Free Full Text]

47. Zolles G, Klocker N, Wenzel D, Weisser-Thomas J, Fleischmann BK, Roeper J, Fakler B. Pacemaking by HCN channels requires interaction with phosphoinositides. Neuron 52: 1027–1036, 2006.[CrossRef][Web of Science][Medline]





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
295/2/C557    most recent
00154.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 Google Scholar
Google Scholar
Right arrow Articles by Vemana, S.
Right arrow Articles by Larsson, H. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vemana, S.
Right arrow Articles by Larsson, H. P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.