AJP - Heart Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 291: H29-H37, 2006. First published February 24, 2006; doi:10.1152/ajpheart.01317.2005
0363-6135/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
291/1/H29    most recent
01317.2005v1
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 ISI Web of Science
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 Zhu, Y.
Right arrow Articles by Lee, P. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhu, Y.
Right arrow Articles by Lee, P. J.

CALL FOR PAPERS
Computational Analysis in Ion Channelopathies

Flecainide sensitivity of a Na channel long QT mutation shows an open-channel blocking mechanism for use-dependent block

Yujie Zhu,1 John W. Kyle,3 and Peter J. Lee1,2

1Department of Medicine and Center for Cardiovascular Research and 2Department of Physiology and Biophysics, University of Illinois at Chicago; and 3Department of Medicine, University of Chicago, Chicago, Illinois

Submitted 14 December 2005 ; accepted in final form 20 February 2006


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
A long QT mutation in the cardiac sodium channel, D1790G (DG), shows enhanced flecainide use-dependent block (UDB). The relative importance of open and inactivated states of the channel in flecainide UDB has been controversial. We used a modifiable, inactivation-deficient mutant channel that contains the F1486C mutation in the IFM motif to investigate the UDB difference between the wild-type (WT-ICM) and DG (DG-ICM) channels. UDB at 5 Hz was greater in DG-ICM than WT-ICM, and IC50 values for steady-state UDB were 7.19 and 18.06 µM, respectively. When [2-(trimethyammonium) ethyl]methanethiosulfonate bromide (MTSET) was included in the pipette and fast inactivation was disabled, IC50 was 5.04 µM for DG-ICM and 12.63 µM for WT-ICM. We measured open-channel block by flecainide directly in MTSET-treated, noninactivating ICM channels. Steady-state block was higher for DG-ICM than WT-ICM (IC50 was 2.34 µM for DG-ICM and 5.87 µM for WT-ICM), suggesting that open-channel block is an important determinant of flecainide UDB. We obtained association (kon) and dissociation (koff) rates for open-channel block by the Langmuir-isotherm model. They were koff = 31.37 s–1, kon = 5.83 s–1·µM–1, and calculated Kd = 5.38 µM for WT-ICM (where Kd = koff/kon); and koff = 24.88 s–1, kon = 9.54 s–1·µM–1, and calculated Kd = 2.61 µM for DG-ICM. These Kd values were similar to IC50 measured from steady-state open-channel block. Furthermore, we modeled UDB mathematically by using these kinetic rates and found that the model predicted experimental UDB accurately. The recovery from UDB had a minor contribution to UDB. Flecainide UDB is predominantly determined by an open-channel blocking mechanism, and DG-ICM channels appeared to have an altered open-channel state with higher flecainide affinity than WT-ICM.

sodium channel; long QT syndrome; electrophysiology

IN THE MAJORITY OF CARDIAC TISSUE, an action potential is initiated by the opening of voltage-gated sodium channel (Nav1.5, the predominant isoform in heart). The inward sodium current (INa) inactivates within milliseconds, but inward Ca2+ current sustains the plateau phase. Subsequently, outward K+ current repolarizes the membrane for the next action potential. Prototypic Nav1.5 mutations that cause long QT syndrome (LQT3) disrupt channel inactivation and produce noninactivating, sustained inward Na+ current (late INa), prolonging the action potential plateau and QT intervals, and subsequently predisposes for early after depolarization (EAD) and torsades des pointes.

Flecainide is a class I antiarrhythmic drug that blocks Na+ channels. Its activity against late INa, which is observed in the majority of familial LQT3 mutations, has raised a potential therapeutic strategy for LQT3 syndrome and rekindled interest in the mechanism of flecainide block of voltage-gated Na+ channels. Flecainide blocks the sodium channel on depolarization and channel opening (14). While open-channel access to the binding site is required for channel block, determining the exact channel state to which the drug interacts during use-dependent block (UDB) has been controversial (for examples of the open-state block, see Refs. 3, 19, 20; for examples of the inactivated-state block, see Refs. 13, 14, 31).

The first LQT3 mutation identified was {Delta}KPQ within the intracellular D3–D4 linker (32), which is critical for fast inactivation of the channel and contains the IFM motif (I1485/F1486/M1487 in Nav 1.5) (34). Several recently identified LQT3 mutations that cause late INa (E1789K, Y1795C, 1795insD), however, are located in the intracellular COOH-terminal segment of the channel (reviewed in Ref. 8). This region of the channel has been suggested to play a role in channel inactivation in brain and cardiac channels (16, 17). Whether late INa is present in a recently identified D1790G (DG) LQT3 mutation (5) from the COOH-terminal region of the channel has been somewhat controversial. An et al. (2) and Wehrens et al. (33) initially reported lack of late INa in DG channels. However, Baroudi and Chahine (4) observed late INa in DG channels. Subsequently, Tateyama et al. (29) showed that late INa is enhanced by protein kinase A (PKA)-dependent phosphorylation or inhibition of phosphatases and is likely responsible for QT prolongation in ECG. Flecainide, but not lidocaine, when administered intravenously, corrected QT prolongation in patients with the DG mutation (6). The exact mechanism by which flecainide corrected QT prolongation in ECG probably involves, at least in part, blockade of the phosphorylation-dependent late INa and the increased open-channel affinity in DG channels.

Flecainide showed increased UDB of DG channels without change in tonic block (1). What is the underlying mechanism for the increased UDB in DG channels in the experimental setting? Could this difference be used to clarify the mechanism of flecainide UDB of the channel? In our study, we determined the factors that influenced flecainide UDB in WT and DG channels empirically. We used wild-type (WT) and DG channels with and without fast inactivation to assess the contribution of open and inactivated channels toward UDB by using the F1486C mutation within the IFM motif (above). We then determined the relative importance of those factors in flecainide UDB by mathematical modeling. We found that the open-channel affinity for flecainide was increased in DG channels, and using mostly the open-channel blocking parameters, we could mathematically model UDB of WT and DG channels accurately, indicating that the increased open-channel affinity was mainly responsible for the increased UDB in DG channels. This suggests that flecainide UDB involved an open-channel blocking mechanism of the sodium channel.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Channel constructs. The hH1a {alpha}-subunit was a gift from Dr. H. A. Hartmann (11). Compared with the hH1 clone (9) used by Abriel et al. (1), hH1a is different from hH1 at four amino acid positions, including the deletion of Q1077. These differences do not result in significant differences in channel properties (15) (our data compared with published data, not shown). For historical reasons, we used the amino acid numbering of hH1 in our proposal. For example, D1790 of hH1 occurs at the 1789th position in hH1a, but we referred the DG mutation as D1790G. Human beta1-cDNA was cloned and sequenced to confirm the amino acid sequence. In all experiments, beta1 was coexpressed with the {alpha}-subunit. Typically, to express both in transfection, two separate plasmids containing the {alpha}- and beta-subunits are mixed and cotransfected. However, assuring adequate expression of the beta1-subunit for each {alpha}-subunit being expressed is difficult because there is no independent method of quantifying its expression. Thus we constructed an expression vector, which utilizes the internal ribosome entry sequence (IRES) to produce bicistronic messages to ensure a 1:1 ({alpha}1:beta1) mRNA ratio. Briefly, the vector contains a CMV promoter that drives a bicistronic mRNA containing the beta1-cDNA in the 5'-side of the IRES and the {alpha}-cDNA in the 3'-side of IRES. The vector also has a separate GFP expression cassette. In transient transfection, cells that take up the plasmid can be identified by fluorescence within 18 h. Cells with green fluorescence were patched between 18 and 48 h. Almost all cells (>95%) patched contained inward sodium current between 500 pA and 5.0 nA (an improvement from transfection with a mixture of multiple plasmids). The presence of sodium current indicates production of intact bicistronic mRNAs because a truncated version lacks the poly(A) tail and is held in the nucleus. We believe that this is an improved approach for coexpressing a consistent amount of beta1. The DG mutation was introduced into the WT channel in pRC/CMV (Invitrogen, Carlsbad, CA) and subcloned into the IRES vector above. The F1486C (IFM->ICM) mutation was introduced into the WT channels by oligonucleotide-directed mutagenesis with the Quick Change Mutagenesis Kit (Stratagene, La Jolla, CA) and subcloned into the IRES constructs containing the WT and DG channels.

Channel expression. HEK-293 cells were maintained in DMEM supplemented with 10% fetal bovine serum in humidified CO2 incubators at 37°C. For transfection experiments, the Superfect Transfection System (Qiagen, Germantown, MD), available as a kit commercially, (28) was used. All experiments were done with transiently transfected cells.

Electrophysiology. Patch-clamp experiments were conducted by using either the Multiclamp 700A or Axopatch 200B amplifier with a Digidata 1322A data-acquisition system (Axon Instruments, Union City, CA). HEK-293 cells were treated with trypsin and placed in a recording chamber containing ~0.1 ml of appropriate external buffer. All recordings were carried out at room temperature (22–25°C). Tip resistance of electrodes was between 0.6 and 1.2 M{Omega}. The standard bath solution contained (in mM) 140 NaCl, 5 KCl, 1 MgCl2, 2 CaCl2, 10 HEPES, pH 7.4 with NaOH. Pipettes contained (in mM) 35 NaCl, 105 CsF, 10 EGTA, 10 HEPES, 1 MgCl2, pH 7.3 with CsOH. This yielded a reversal potential (Vrev) of +35 mV at room temperature. Leak current and series resistance corrections were done electronically. The quality of voltage-clamp was judged by examination of data, including the Vrev from current-voltage relationship experiments. Cells expressing more than 5 nA of current at room temperature were not used in order to minimize voltage-clamp errors.

Before all experiments, the holding potential was –140 mV. Unless specified, UDB experiments involved repetitive 10-ms pulses to –10 mV from –140 mV at 5 Hz. A UDB protocol in drug-free conditions preceded the drug block experiments and showed <5% current loss in all instances. In all experiments, the whole cell configuration was held at –140 mV for at least 8 min, and when flecainide was applied, cells were held at –140 mV and superfused for at least 2 min for complete dialysis. For each cell, only one dose of flecainide was used to avoid accumulation of drug effects. Whenever possible, the flecainide was washed out, and current levels were checked. The majority of experiments involved long periods of voltage clamp because of slow flecainide recovery, and washing out the flecainide was frequently not possible. In one of four experiment sets (consisting of drug-free channel gating and UDB protocols followed by flecainide block and recovery protocols), washout data were possible. In all cases, the full recovery of current levels was observed.

Fast inactivation was disabled by including 1 mM [2-(trimethyammonium)ethyl]methanethiosulfonate bromide (MTSET) (Anatrace, Maumee, OH) in the pipette. MTSET was prepared just before the experiment. We selected cells that showed <15% of current decay at the end of a 100-ms pulse at –10 mV for analyses. Details of other experiments are described in RESULTS and in corresponding figure legends.

Data analysis. Data are reported as means value ± SE. Data were analyzed with Clampfit (Axon Instruments, Union City, CA) and Origin 7 (Originlab, Northampton, MA) software. Two sets of data were compared by using paired or unpaired Student's t-test, and P < 0.05 was used to define statistical significance.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Rationale for design of channel constructs. First, we confirmed the findings by Abriel et al. (1) using hH1a. At 2 Hz, steady-state UDB level (ssUDB) and {tau}UDB (UDB onset time constant in pulse numbers) were 0.301 ± 0.041 and 8.1 ± 1.1 for WT and 0.515 ± 0.076 and 31.9 ± 5.8 for DG channels with 10 µM flecainide (P < 0.05 for both ssUDB and {tau}UDB, n = 5). With 2 µM flecainide, they were 0.084 ± 0.018 and 22.7 ± 3.3 for WT and 0.220 ± 0.048 and 53.4 ± 5.9 for DG channels (P < 0.05 for both parameters, n = 4 and 6 respectively). Thus, under our experimental conditions, which used hH1a and –140 mV as a holding potential, DG channels showed increased UDB compared with WT. One facet of flecainide UDB that was not described in previously published studies was UDB onset kinetics (1, 13, 14). UDB developed more slowly in DG channels (longer {tau}onset) although ssUDB was higher, in both 2 µM and 10 µM flecainide. The onset kinetics of UDB is determined by multiple factors, including drug affinities to various channel states and drug access/egress paths. To evaluate these possibilities better, we inserted the F1486C mutation within the IFM motif of the D3–D4 linker [the fast-inactivation particle (34)] into the WT and DG channels (WT-ICM and DG-ICM). With the F1486C mutation, the flecainide UDB difference between WT-ICM and DG-ICM channels persisted (Fig. 1, A, C, and D and time constants for UDB in Table 1). DG-ICM channels were more susceptible to UDB by flecainide than WT-ICM channels. Recent data suggest that flecainide reaches the binding site through the cytoplasmic pore after crossing the membrane when superfused (13). By including MTSET in the pipette, we removed fast inactivation, continuously exposing the open state of the channels during the UDB pulses (7, 12, 23, 30). This approach allowed for the direct comparison of channel block by flecainide with or without fast inactivation (with or without MTSET) in a common channel background. Previous studies have shown that internal MTSET by itself does not affect INa (24, 35).


Figure 1
View larger version (28K):
[in this window]
[in a new window]
  		Fig. 1. Use-dependent block (UDB) ± [2-(trimethyammonium) ethyl]methanethiosulfonate bromide (MTSET) of WT-ICM and DG-ICM channels. A: UDB for wild-type (WT)-ICM and D1790G LQT3 mutation (DG)-ICM without MTSET was induced by applying a train of 100 depolarizing pulses (–10 mV/10 ms) at a frequency of 5 Hz. The currents were measured after application of flecainide (2, 5, 20, 55 µM). B: UDB for WT-ICM and DG-ICM with MTSET was induced by applying a train of 100 depolarizing pulses (–10 mV/10 ms) at a frequency of 5 Hz. The currents were measured after application of flecainide (2, 5, 20, 55 µM). C, top: UDB without MTSET (flecainide, 20 µM). First-pulse block (P0, no drug; P1, first pulse of UDB train) was minimal. Pss, a pulse of the steady-state UDB (ssUDB). C, bottom: when MTSET was added, current decay was abolished, but the difference in UDB persisted. Peak current (*) levels are used for UDB (20 µM). D: IC50 for ssUDB obtained from the dose-response curves. IC50 for flecainide without MTSET is 18.06 µM (WT-ICM) and 7.19 µM (DG-ICM); IC50 for flecainide with MTSET is 12.63 µM (WT-ICM) and 5.04 µM (DG-ICM). ICM refers to modifiable, inactivation-deficient mutant channel that contains the F1486C mutation in the IFM motif.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Parameters for first-pulse block and UDB

 
Flecainide UDB of ICM channels with fast inactivation intact. As with WT and DG channels, DG-ICM was more sensitive to flecainide UDB (higher ssUDB at 5 Hz) than WT-ICM although the UDB developed more slowly (Fig. 1, A, C, and D, and time constants for UDB in Table 1). Therefore, WT-ICM and DG-ICM channels were qualitatively similar to WT and DG channels with regard to flecainide UDB.

Increased UDB in DG-ICM (and DG) channels can result from higher flecainide affinity to the closed, open, or inactivated states. A closed state with a higher affinity is unlikely because the first-pulse block (resting block) was unchanged in DG-ICM (Table 1). To minimize the contribution from the inactivated state, we measured UDB of the two channels (WT-ICM and DG-ICM) with 0.6-ms pulses, during which development of an inactivated state (both fast or slow inactivated state) is unlikely, and found that DG-ICM channels still showed higher ssUDB (0.24 ± 0.04 at 5 µM flecainide, P < 0.05, n = 4) than WT-ICM (0.12 ± 0.04, n = 3). The result indicated that the main difference in flecainide UDB between WT-ICM and DG-ICM channels was likely to be related to the open-channel state. Without the inactivated state, DG-ICM channels were still blocked more than WT-ICM channels.

Flecainide UDB of ICM channels with fast inactivation disabled. To clarify further which channel state was responsible, we measured flecainide UDB after disabling fast inactivation by including MTSET (1 mM) in the pipette. Removal of fast inactivation resulted in a small increase in ssUDB. There was a trend toward higher ssUDB with MTSET, but P > 0.05 (except for DG-ICM at 20 µM, P = 0.05) (Fig. 1, B–D, and Table 1). After fast inactivation was disabled, flecainide UDB achieved steady state quickly. DG-ICM channels were still blocked more than WT-ICM channels even in the absence of fast inactivation. Therefore, the fast-inactivated state is not likely to be involved in flecainide UDB because the lack of it did not decrease UDB.

Because MTSET in solution is very labile at room temperature [half-life, ~10 min (27)], often, fast inactivation was not completely disabled, even after a long period of exposure (>10 min). We selected cells that showed <15% of current decay at the end of a 100-ms pulse at –10 mV for analyses; the small amount of current decay might represent slow inactivation, which persists even when the fast inactivation is removed (21). Typically, one of four cells successfully patched satisfied this criterion. This stringent criterion allowed us to fit the time-dependent block of flecainide with a one-exponential function. In the study by Ramos and O'Leary (20), a two-exponential fitting was necessary to account for the slowly decaying current in channels with the IFM->QQQ mutation (see below and Fig. 2 legends on obtaining kinetic constants for open-channel block).


Figure 2
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 2. Open-channel block after one 100-ms pulse. A: WT-ICM + MTSET. B: DG-ICM + MTSET. A (WT-ICM) and B (DG-ICM) are examples of open-channel block in the presence of 20 µM flecainide. Time-dependent block is evident, and the time course of current decay was fitted with a one-exponential decay function, Ae–t/{tau}open+ C, where A, t, and C are amplitude, duration of open channel exposure to flecainide, and steady-state open channel block, respectively. ({tau}open, Table 2). Steady-state block is b/(a + b). First-pulse block, (d – c)/d, is relatively small (Table 1). C: dose-response curve with IC50 values for steady-state open-channel block. D: effective blocking rate (1/{tau}open) plotted vs. flecainide concentration. DG-ICMMTSET and WT-ICMMTSET refer to DG-ICM with MTSET and WT-ICM with MTSET, respectively. Blocking rate (1/{tau}open) increased linearly with flecainide concentration where the slope and y-intercept represent the association (kon) and dissociation (koff) rate constants, respectively. 1/{tau}open = kon x D + koff. For WT-ICM, koff = 31.37 s–1 and kon = 5.83 s–1·µM–1. For DG-ICM, koff = 24.88 s–1 and kon = 9.54 s–1·µM–1. With fast inactivation disabled, the flecainide affinity for open channels was higher for DG-ICM than for WT-ICM. Calculated Kd (koff/kon) was 5.38 µM for WT-ICM and 2.61 µM for DG-ICM, in good agreement with measured IC50 values in C.

 
Open-channel affinities for flecainide. We characterized the open-channel block of WT-ICM and DG-ICM channels. After fast inactivation was disabled, an exposure of the channel binding sites to flecainide in a single 100-ms pulse (–10 mV) showed time-dependent block (Fig. 2, A–C, and Table 2). We measured the steady-state open-channel block by taking the current levels at the end of the 100-ms pulses before and after drug application.


View this table:
[in this window]
[in a new window]
  		Table 2. Parameters for open-channel block

 
The effective blocking rate (1/{tau}open) was plotted vs. flecainide concentration. The blocking rate (1/{tau}open) increased linearly with flecainide concentration, where the slope and y-intercept represent the association (kon) and dissociation (koff) rate constants, respectively. For WT-ICM, koff = 31.37 s–1 and kon =5.83 s–1·µM–1. For DG-ICM, koff = 24.88 s–1 and kon = 9.54 s–1·µM–1. The open-channel affinity for flecainide was increased in DG-ICM channels. The increased affinity was due to a combination of both increased on rate (kon) and decreased off rate (koff), estimated in the Langmuir-isotherm model (Fig. 2D). The dissociation constants Kd for WT-ICM and DG-ICM calculated from koff/kon were in good agreement with measured IC50 values from the dose-response curve (Fig. 2C). These koff and kon values are also in reasonable agreement with results by others using the IFM/QQQ mutation, considering differences in experimental designs [Ref. 10, HEK-293 system, but no beta1 and at –20 mV (vs. our –10 mV); Ref. 20, Xenopus oocyte system and no beta1].

UDB recovery. Recovery from UDB is an important determinant of ssUDB and {tau}UDB. Abriel et al. (1) showed that UDB recovery was slowed in DG channels. Recovery was not complete after 10 s for either WT or DG channels at –100 mV, and the recovery time constants were not estimated (Fig. 4 in Ref. 1). We measured recovery from flecainide UDB after a 100-pulse train (5 Hz) (Fig. 3A). The first pulse in the recovery protocol occurs 50 ms after the UDB train, allowing the channels to recover from fast inactivation. In both WT-ICM and DG-ICM channels, near-complete recovery from UDB was achieved at –140 mV, and accurate estimation of the recovery time constants was possible. Two-exponential fitting was necessary for accurate modeling of the UDB recovery, which was not dependent on the drug concentration (not shown) and was slowed in DG-ICM channels, mainly due to prolongation of the longer time constants (A2, ~70–80%, Fig. 3A and Table 3). The faster components ({tau} ~ 1 s) were not significantly different between WT-ICM and DG-ICM. Interestingly, disabling fast inactivation alone did not influence recovery from UDB, suggesting that fast inactivation did not stabilize drug binding (or vice versa), consistent with previous data with lidocaine (30). This difference in UDB recovery may have contributed to the increased UDB in DG-ICM. However, the recovery time constants ranged in seconds, with the predominant component (A2) near 100 s. It is therefore difficult to attribute the difference in UDB between WT-ICM and DG-ICM channels solely to UDB recovery when the interpulse recovery periods were 190 ms (5 Hz with 10-ms pulses).


Figure 4
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 4. Experimental UDB and modeled UDB. Using the framework of the guarded-receptor model (text), the onset kinetics of UDB ({tau}UDB) and ssUDB of UDB can be estimated independently. The graphical representation of UDB, generated by Eq. 1 (C), with (A) or without (B) fast inactivation shows good agreement between model and experiment. Note that the x-axes (pulse no.) of the graphs are adjusted in A to depict the course of UDB better. C: equations used for calculations (see text). Eqs. 1–5: an, fractional block at nth pulse; ass, fractional block at steady state; a1, fractional block at the first pulse of UDB (= first-pulse block); a{infty}, steady-state open channel block; n, pulse number, starting with 1; {lambda}, onset-rate for UDB (1/{tau}UDB); td, time duration of open-channel exposure to flecainide (area under the current tracings normalized to the peak current); tr, time duration at resting potential; {tau}d, time constant of channel block based on kon, koff, and D (= drug concentration), where kon is the on-rate during depolarization (= kon in open-channel block) and koff is the off-rate during depolarization (from the open channel); {tau}r, time constant of channel block/recovery at resting potential. We assumed that no blocking occurs while channels are closed at hyperpolarized potentials. A1 and A2, relative amplitudes for faster {tau}1(UDBrec) and slower {tau}2(UDBrec) recovery time constants in UDB recovery (see Table 3). D: time constants for UDB onset kinetics and ass (ssUDB) were calculated from the guarded-receptor model; ass was adjusted for first-pulse block. (Because ssUDB values in Table 1 were obtained by normalizing to the first pulse of the UDB train after the drug is applied, an adjustment for the first-pulse block was necessary.)

 

Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 3. Recovery from UDB (A) and recovery from inactivation (B). Recovery kinetics were fitted to a 2-exponential equation, I(t) = 1 – A1e–t/{tau}1 – A2e–t/{tau}2 – C, where I(t) is the fractional current at time t, A1 and A2 are amplitudes of exponentials, {tau}1 and {tau}2 are the recovery time constants, and C is the fraction of channels that are not fully recovered. A: recovery from UDB was measured by a test-pulse (1-ms pulse duration/–10 mV) after a train of 100 UDB pulses at 5 Hz after varying lengths of time. (see Table 3 for fitting parameters). B: recovery from inactivation was measured by a pulse to –10 mV after a 1-s inactivating pulse to –10 mV and variable length of recovery period at –140 mV. The amplitudes and time constants (A1, {tau}1, A2, {tau}2) for inactivation recovery were: for WT-ICM, 0.325, 2.01 ms, 0.675, 89.77 ms; for DG-ICM, 0.439, 1.13 ms, 0.561, 69.27 ms.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Parameters for recovery from UDB

 
Unlike UDB recovery, recovery from inactivation was faster in DG-ICM channels (Fig. 3B). A two-exponential fit of the recovery showed significant differences in the time constants between WT-ICM and DG-ICM channels (Fig. 3 legend). Therefore, the slowed flecainide-UDB recovery in DG-ICM channels does not appear to be due to slowed recovery from inactivation. Recovery from UDB (100 10-ms pulses to –10 mV at 5 Hz) without flecainide was fast and complete for both WT-ICM and DG-ICM with time constants 1–2 ms (P = not significant; not shown).

Mathematical modeling of UDB. How much does open-channel block, relative to UDB recovery, contribute to flecainide UDB? To estimate the importance of open-channel block in UDB, we used mathematical modeling to simulate flecainide UDB of WT-ICM and DG-ICM clones. Because we found that the first-pulse block is relatively small (Table 1), and Liu et al. (14) concluded that channel opening was necessary for flecainide block, we considered the receptor to be guarded at a hyperpolarized membrane potential (–140 mV). We used the equations described by Starmer and Grant (25) and Starmer et al. (26) under the guarded-receptor model (Eqs. 1, 2, and 5 in Fig. 4C) and the equation that describes the bimolecular-binding model (Eq. 3 in Fig. 4C). The recurrence equation can be solved to give the fraction of blocked channels associated with the nth stimulus (Eq. 1 in Fig. 4C). The uptake rate {lambda} is a linear combination of the two reciprocal time constants (Eq. 2 in Fig. 4C). When recovery potentials are sufficiently negative and drug block is relatively small, the steady-state blockade is simplified and can be described by Eq. 5 in Fig. 4C. When MTSET is used, the open-channel state is continuously exposed during each pulse. The equations used by Starmer and Grant (25) are not necessarily unique to the guarded-receptor model and can be used to model any UDB when several key features of flecainide UDB validate the assumptions required for Eq. 5 in Fig. 4C.

First, we note that the uptake rate {lambda} (Eq. 2 in Fig. 4C) is influenced by both td/{tau}d and tr/{tau}r, although td/{tau}d is significantly greater than tr/{tau}r. This is consistent with the slow UDB recovery (Table 3, with time constants in s) observed in our study and with the conclusion by Ramos and O'Leary (20) that flecainide is trapped by the activation gate. As shown in Fig. 4D, the UDB-onset kinetics ({tau}UDB) and ssUDB can be accurately predicted (compare the modeled parameters to those in Table 1). More interesting is whether the model can predict UDB when fast inactivation is intact where access to and egress from the receptor site is not continuous. We examined this by assuming that flecainide is a "pure" open-channel blocking drug. We used the same {tau}d as above (kon and koff from the open-channel block; Fig. 4C, Eq. 2) and adjusted the td (duration of open-channel exposure to flecainide in each pulse, "effective exposure time") by calculating the area under the current tracing. As shown in Fig. 1, DG-ICM channels show faster current decay than WT-ICM channels. A similar observation was also true for WT and DG channels (not shown). Therefore, the effective exposure time is shorter for DG-ICM channels than that for WT-ICM when fast inactivation is intact (without MTSET, 1.93 ± 0.33 ms for WT-ICM and 1.23 ± 0.10 ms for DG-ICM, P < 0.05, n = 10; with MTSET, 8.25 ± 0.25 ms for WT-ICM and 7.77 ± 0.35 ms for DG-ICM, P = not significant, n = 13). In this manner, the model estimated {tau}UDB and ssUDB well even when fast inactivation was present and the receptor was not continuously accessible (Fig. 4B and Fig. 4D table, compared with {tau}UDB and ssUDB in Table 1). Apparently, no parameters other than the kinetic constants for open-channel block are needed to model flecainide UDB (with or without fast inactivation).

In summary, we showed that the open-channel affinity for flecainide was increased, and the recovery from UDB was slowed in DG-ICM channels. Using the data from the mathematical model of UDB, we propose that this increase in the open-channel affinity underlies the increased flecainide UDB with a minor contribution from the difference in the UDB recovery and that an open-channel mechanism is responsible for flecainide UDB.


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
The primary purpose of our study was to clarify controversies in the basic mechanism of flecainide UDB by examining the drug-channel interaction of a mutant channel that shows increased flecainide UDB, i.e., we reasoned that whatever difference existed in drug-channel interaction between the WT and DG channels would be important in the flecainide UDB. In our study, we used ICM channels (WT-ICM and DG-ICM), with which we can selectively disable the fast inactivation and compare UDB in a common channel background. WT-ICM and DG-ICM channels were qualitatively similar to WT and DG channels with regard to flecainide UDB. We found that the open-state affinity of DG channels for flecainide was higher than that of WT and that this increased affinity primarily underlies the difference in UDB. This suggests an open-channel mechanism for flecainide UDB.

Recent studies also demonstrated data in favor of an open-channel mechanism for flecainide block. Observation of time-dependent open-channel block supports direct interaction between the open-state and flecainide (Refs. 10 and 20 and our current study). In the recent two studies (10, 20), the "inactivation-deficient" (IFM->QQQ) channel was used to measure the open-channel block. However, the IFM->QQQ channels may not be directly comparable to wild-type channels, and the contribution of fast inactivation toward flecainide UDB and slow recovery from the drug block cannot be directly compared. In the absence of MTSET, WT-ICM and DG-ICM channels exhibited essentially intact fast inactivation and current-decay properties similar to those of WT and DG channels.

In the study by Viswanathan et al. (31), evidence for an "inactivated-state" blocking mechanism has involved correlation to enhanced "closed-state" and intermediate inactivation. Direct interaction between flecainide and these inactivated states were not demonstrated. In addition, the mutant channels showing the enhanced closed-state inactivation also had significant late INa, which could have been blocked by flecainide. Liu et al. (14), using several Na+ channel disease mutations with distinct flecainide sensitivities, concluded that channel opening is necessary but flecainide binds tightly to "an inactivated state." They estimated mean open times (0.50 ± 0.02 ms for WT and 0.42 ± 0.03 ms for DG). However, they did not consider the possibility of a change in flecainide affinity to the open-state between WT and DG [calculated open-state Kd (koff/kon) were 5.38 µM for WT-ICM and 2.61 µM for DG-ICM], which is the main finding in our study. Slowed recovery from flecainide block in the above studies (and our current study) may also represent differential voltage- and state-dependent "trapping" of the drug (20) and not necessarily an increased affinity to an inactivated state. A neutral form of flecainide, which does not depend on the pore access path for block, has been shown to shift the steady-state availability (inactivation) curve (13). However, it is not clear whether the behavior of the neutral drug truly represents that of flecainide.

We used a holding potential of –140 mV, sufficiently hyperpolarized to deactivate sodium channels fully. Our choice of such a hyperpolarized holding potential may have lessened the contribution from an inactivated state toward flecainide UDB. At –140 mV, all channels are uniformly in the closed state; it is approximately ~35 mV more negative than half-available voltage (V1/2) of the steady-state availability (SSA) curves of DG channels (not shown), similar to the conditions in the study by Abriel et al. (1). UDB recovery was nearly complete (Fig. 3A) within 200 s and was accurately estimated with a two-exponential function. Even at –140 mV, the difference in UDB recovery between WT-ICM and DG-ICM channels persisted, similar to the results by Abriel et al. (1) at –100 mV.

In our study, we found two significant differences in flecainide UDB between the WT and DG channels, increased open-channel affinity to flecainide and slowed recovery from UDB. The fast-inactivated state appeared to have little influence on flecainide block because the lack of it did not decrease UDB or prolong UDB recovery. Direct involvement of a slow-inactivated state seems unlikely because the time constants for the open-channel block ranged from 1 to 30 ms (Table 2), while slow-inactivated states have been associated with longer time constants when fast inactivation is disabled.

We used the guarded-receptor model (25, 26) to mathematically model the flecainide UDB so that the observed differences in open-channel affinities and UDB recovery accounts for the differing flecainide UDB in the WT-ICM and DG-ICM channels. The model predicted flecainide UDB accurately by using open-channel flecainide affinities and kinetic constants for UDB recovery during the interpulse intervals without contribution from any inactivated states. We believe that the difference in open-channel affinity for flecainide is the predominant factor in the increased UDB in DG channels because the recovery from UDB during each interpulse interval is relatively small. Consistent with this is the observation that, in physiological conditions, flecainide slows conduction but does not prolong refractoriness significantly (22). In fact, when the faster component of UDB recovery (see RESULTS and Table 3) was omitted in the mathematical modeling, {lambda} was completely dominated by the open-channel parameters ({tau}d). Prediction of {tau}UDB was still very accurate, and ssUDB levels were overestimated (i.e., the model predicts higher flecainide UDB than observed) only by ~10–20% in both WT-ICM and DG-ICM channels (data not shown), suggesting that the open-channel affinity for the drug predominantly underlies flecainide UDB.

Interestingly, the modeled {tau}UDB underestimated (that is, UDB develops faster in the model) the experimental observations. There are couple possibilities for this. In UDB experiments, on repolarization at the end of each pulse, a small fraction of the drug might egress quickly from the pore-receptor site before the activation gate closes. This would result in prolonging {tau}UDB (number of pulses needed to reach the steady state) in experiments. Such loss would not be reflected in kon and koff values obtained from the single-pulse open-channel experiments. Second, the 10-ms pulse duration in UDB may not be long enough to "induce" the channel state that binds to flecainide with a high affinity. This may be consistent with the fact that {tau}open (Fig. 2 and Table 2) ranges beyond 10 ms. We propose that the high-affinity site likely occurs after 0.6 ms of membrane depolarization because UDB with 0.6-ms pulses resulted in significant decrease in ssUDB, suggesting that a preopen state is not likely to be involved. The time course of block is still too fast for direct involvement of other inactivated states in flecainide binding.

We demonstrated an open-channel blocking mechanism for flecainide UDB by mathematical modeling. However, there are several limitations to our study. We conducted all of our experiments at room temperature. At more physiological temperatures, channel gating and drug-block kinetics may be different. For example, the majority of such studies are carried out at room temperature (~22°C), which is 15°C lower than the physiological body temperature. Conductance-voltage and SSA curves shift toward more depolarized potentials as temperature rises (18). We chose the holding voltage of –140 mV in part to accommodate such shifts. The hyperpolarized holding potential may have also unfairly excluded certain inactivated or closed states of the channel in our protocols. This also directly limits addressing physiological significance of our findings but made it possible to separate channel states clearly for our mathematical model.

One issue that we did not address in our study is how flecainide corrected QT prolongation in patients with DG mutations. QT prolongation by DG mutation is likely caused by late INa that was observed in DG channels in experiments by Baroudi and Chahine (4) and under phosphorylation-promoting conditions in the study by Tateyama et al. (29). In the study by Baroudi and Chahine (4), fluoride ion, which can inhibit phosphatase activities, was included in the patch pipette solutions (cytoplasmic side) and may have promoted late INa. In our experiments, we also observed late INa (fluoride in the pipette solution) that was ~2.4% of the peak current in DG channels, which was not present in WT channels. However, although implicitly included, the late INa contributed little toward the "effective exposure time" that was used in our mathematical modeling. The mechanism of correcting QT prolongation by flecainide likely involves both blockade of the late INa and the increased open-channel affinity in DG channels that we showed in our current work. The main focus of our study was not to elicit the mechanism of flecainide's ability to correct QT prolongation but to investigate the fundamental drug-channel interaction between flecainide and Nav1.5, specifically DG-ICM vs. WT-ICM.

In conclusion, we propose that the open-channel blocking mechanism is responsible for flecainide UDB and that increased open-channel affinity underlies the increased UDB in DG channels with a minor contribution from slowed recovery from UDB. We believe that our results provide definitive evidence for an open-channel blocking mechanism for flecainide UDB of Nav1.5. Further investigation of identifying the structural features of DG channels responsible for the increased open-channel flecainide affinity remains to be done.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This study was supported by National Heart, Lung, and Blood Institute Grant HL-072529 to P. J. Lee.


   ACKNOWLEDGMENTS
 
We thank Dr. Harry Fozzard for critical reading of the manuscript. We also thank Dr. Harvey Lardin for helpful discussions.


   FOOTNOTES
 

Address for reprint requests and other correspondence: P. J. Lee, Dept. of Medicine/Cardiology, Univ. of Illinois at Chicago, 840 S. Wood ST, M/C 715, Chicago, IL 60612 (e-mail: PeterL{at}uic.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. Abriel H, Wehrens XH, Benhorin J, Kerem B, and Kass RS. Molecular pharmacology of the sodium channel mutation D1790G linked to the long-QT syndrome. Circulation 102: 921–925, 2000.[Abstract/Free Full Text]
  2. An RH, Wang XL, Kerem B, Benhorin J, Medina A, Goldmit M, and Kass RS. Novel LQT-3 mutation affects Na+ channel activity through interactions between {alpha}- and beta1-subunits. Circ Res 83: 141–146, 1998.[Abstract/Free Full Text]
  3. Anno T and Hondeghem LM. Interactions of flecainide with guinea pig cardiac sodium channels. Importance of activation unblocking to the voltage dependence of recovery. Circ Res 66: 789–803, 1990.[Abstract/Free Full Text]
  4. Baroudi G and Chahine M. Biophysical phenotypes of SCN5A mutations causing long QT and Brugada syndromes. FEBS Lett 487: 224–228, 2000.[CrossRef][ISI][Medline]
  5. Benhorin J, Goldmit M, MacCluer JW, Blangero J, Goffen R, Leibovitch A, Rahat A, Wang Q, Medina A, Towbin J, and Kerem B. Identification of a new SCN5A mutation, D1840G, associated with the long QT syndrome. Mutations in brief no 153. Online Hum Mutat 12: 72, 1998.
  6. Benhorin J, Taub R, Goldmit M, Kerem B, Kass RS, Windman I, and Medina A. Effects of flecainide in patients with new SCN5A mutation: mutation-specific therapy for long-QT syndrome? Circulation 101: 1698–1706, 2000.[Abstract/Free Full Text]
  7. Chahine M, Deschenes I, Trottier E, Chen LQ, and Kallen RG. Restoration of fast inactivation in an inactivation-defective human heart sodium channel by the cysteine modifying reagent benzyl-MTS: analysis of IFM-ICM mutation. Biochem Biophys Res Commun 233: 606–610, 1997.[CrossRef][ISI][Medline]
  8. Clancy CE and Kass RS. Inherited and acquired vulnerability to ventricular arrhythmias: cardiac Na+ and K+ channels. Physiol Rev 85: 33–47, 2005.[Abstract/Free Full Text]
  9. Gellens ME, George AL Jr, Chen LQ, Chahine M, Horn R, Barchi RL, and Kallen RG. Primary structure and functional expression of the human cardiac tetrodotoxin-insensitive voltage-dependent sodium channel. Proc Natl Acad Sci USA 89: 554–558, 1992.[Abstract/Free Full Text]
  10. Grant AO, Chandra R, Keller C, Carboni M, and Starmer CF. Block of wild-type and inactivation-deficient cardiac sodium channels IFM/QQQ stably expressed in mammalian cells. Biophys J 79: 3019–3035, 2000.[Abstract/Free Full Text]
  11. Hartmann HA, Tiedeman AA, Chen SF, Brown AM, and Kirsch GE. Effects of III-IV linker mutations on human heart Na+ channel inactivation gating. Circ Res 75: 114–122, 1994.[Abstract/Free Full Text]
  12. Kellenberger S, Scheuer T, and Catterall WA. Movement of the Na+ channel inactivation gate during inactivation. J Biol Chem 271: 30971–30979, 1996.[Abstract/Free Full Text]
  13. Liu H, Atkins J, and Kass RS. Common molecular determinants of flecainide and lidocaine block of heart Na+ channels: evidence from experiments with neutral and quaternary flecainide analogues. J Gen Physiol 121: 199–214, 2003.[Abstract/Free Full Text]
  14. Liu H, Tateyama M, Clancy CE, Abriel H, and Kass RS. Channel openings are necessary but not sufficient for use-dependent block of cardiac Na+ channels by flecainide: evidence from the analysis of disease-linked mutations. J Gen Physiol 120: 39–51, 2002.[Abstract/Free Full Text]
  15. Makielski JC, Ye B, Valdivia CR, Pagel MD, Pu J, Tester DJ, and Ackerman MJ. A ubiquitous splice variant and a common polymorphism affect heterologous expression of recombinant human SCN5A heart sodium channels. Circ Res 93: 821–828, 2003.[Abstract/Free Full Text]
  16. Mantegazza M, Yu FH, Catterall WA, and Scheuer T. Role of the C-terminal domain in inactivation of brain and cardiac sodium channels. Proc Natl Acad Sci USA 98: 15348–15353, 2001.[Abstract/Free Full Text]
  17. Motoike HK, Liu H, Glaaser IW, Yang AS, Tateyama M, and Kass RS. The Na+ channel inactivation gate is a molecular complex: a novel role of the COOH-terminal domain. J Gen Physiol 123: 155–165, 2004.[Abstract/Free Full Text]
  18. Nagatomo T, Fan Z, Ye B, Tonkovich GS, January CT, Kyle JW, and Makielski JC. Temperature dependence of early and late currents in human cardiac wild-type and long Q-T {Delta}KPQ Na+ channels. Am J Physiol Heart Circ Physiol 275: H2016–H2024, 1998.[Abstract/Free Full Text]
  19. Ragsdale DS, McPhee JC, Scheuer T, and Catterall WA. Common molecular determinants of local anesthetic, antiarrhythmic, and anticonvulsant block of voltage-gated Na+ channels. Proc Natl Acad Sci USA 93: 9270–9275, 1996.[Abstract/Free Full Text]
  20. Ramos E and O'Leary ME. State-dependent trapping of flecainide in the cardiac sodium channel. J Physiol 560: 37–49, 2004.[Abstract/Free Full Text]
  21. Richmond JE, Featherstone DE, Hartmann HA, and Ruben PC. Slow inactivation in human cardiac sodium channels. Biophys J 74: 2945–2952, 1998.[Abstract/Free Full Text]
  22. Roden DM. Antiarrhythmic drugs. In: Goodman & Gilman's The Pharmacological Basis of Therapeutics. (10th ed.), edited by Hardman JG, Limbird LE, and Gilman AG. New York: McGraw-Hill, 2001, p. 933–970.
  23. Sheets MF and Hanck DA. Charge immobilization of the voltage sensor in domain IV is independent of sodium current inactivation. J Physiol 563: 83–93, 2005.[Abstract/Free Full Text]
  24. Sheets MF, Kyle JW, Kallen RG, and Hanck DA. The Na channel voltage sensor associated with inactivation is localized to the external charged residues of domain IV, S4. Biophys J 77: 747–757, 1999.[Abstract/Free Full Text]
  25. Starmer CF and Grant AO. Phasic ion channel blockade. A kinetic model and parameter estimation procedure. Mol Pharmacol 28: 348–356, 1985.[Abstract]
  26. Starmer CF, Grant AO, and Strauss HC. Mechanisms of use-dependent block of sodium channels in excitable membranes by local anesthetics. Biophys J 46: 15–27, 1984.[Abstract/Free Full Text]
  27. Stauffer DA and Karlin A. Electrostatic potential of the acetylcholine binding sites in the nicotinic receptor probed by reactions of binding-site cysteines with charged methanethiosulfonates. Biochemistry 33: 6840–6849, 1994.[CrossRef][Medline]
  28. Tang MX, Redemann CT, and Szoka FC Jr. In vitro gene delivery by degraded polyamidoamine dendrimers. Bioconjug Chem 7: 703–714, 1996.[CrossRef][ISI][Medline]
  29. Tateyama M, Rivolta I, Clancy CE, and Kass RS. Modulation of cardiac sodium channel gating by protein kinase A can be altered by disease-linked mutation. J Biol Chem 278: 46718–46726, 2003.[Abstract/Free Full Text]
  30. Vedantham V and Cannon SC. The position of the fast-inactivation gate during lidocaine block of voltage-gated Na+ channels. J Gen Physiol 113: 7–16, 1999.[Abstract/Free Full Text]
  31. Viswanathan PC, Bezzina CR, George AL Jr, Roden DM, Wilde AA, and Balser JR. Gating-dependent mechanisms for flecainide action in SCN5A-linked arrhythmia syndromes. Circulation 104: 1200–1205, 2001.[Abstract/Free Full Text]
  32. Wang Q, Shen J, Splawski I, Atkinson D, Li Z, Robinson JL, Moss AJ, Towbin JA, and Keating MT. SCN5A mutations associated with an inherited cardiac arrhythmia, long QT syndrome. Cell 80: 805–811, 1995.[CrossRef][ISI][Medline]
  33. Wehrens XH, Abriel H, Cabo C, Benhorin J, and Kass RS. Arrhythmogenic mechanism of an LQT-3 mutation of the human heart Na+ channel {alpha}-subunit: a computational analysis. Circulation 102: 584–590, 2000.[Abstract/Free Full Text]
  34. West JW, Patton DE, Scheuer T, Wang Y, Goldin AL, and Catterall WA. A cluster of hydrophobic amino acid residues required for fast Na+-channel inactivation. Proc Natl Acad Sci USA 89: 10910–10914, 1992.[Abstract/Free Full Text]
  35. Yang N, George AL Jr, and Horn R. Molecular basis of charge movement in voltage-gated sodium channels. Neuron 16: 113–122, 1996.[CrossRef][ISI][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
291/1/H29    most recent
01317.2005v1
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 ISI Web of Science
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 Zhu, Y.
Right arrow Articles by Lee, P. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhu, Y.
Right arrow Articles by Lee, P. J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2006 by the American Physiological Society.