Cardiac-Specific External Paths for Lidocaine, Defined by Isoform-Specific Residues, Accelerate Recovery From Use-Dependent Block
Local anesthetic antiarrhythmic drugs block voltage-gated Na+ channels from the cytoplasmic side. In addition, cardiac Na+ channels can be also blocked by the membrane-impermeant local anesthetic QX via external paths not present in skeletal muscle or brain channels. Introduction of cardiac isoform-specific residues into wild-type skeletal muscle or brain channels creates access paths for external QX block. These paths should affect the characteristics of use-dependent block by influencing drug on- and off-rates. We investigated the effects of these external paths on drug kinetics of lidocaine, a lipophilic drug of clinical relevance, by studying use-dependent block using a two-electrode voltage clamp in Xenopus oocytes. Recovery from use-dependent block was slowed when cardiac isoform-specific residues important for external QX access were mutated to skeletal muscle or brain isoform-specific residues. As the fraction of charged lidocaine was decreased by raising external pH, differences in recovery kinetics diminished, indicating that these mutations mostly influenced block by charged lidocaine molecules. Data were fit into a model in which bound drug distributes into charged and neutral forms based on its pKa and external pH with separate dissociation paths and recovery-time constants. These isoform-specific mutations altered the recovery-time constants for the charged molecules with smaller effects on those for the neutral molecules. We conclude that the external egress paths created by isoform-specific residues influence the drug kinetics of lidocaine, and these residues define cardiac-specific external paths for local anesthetic drugs.
Class I local anesthetic (LA) antiarrhythmic drugs (reviewed in Butterworth and Strichartz1) block Na+ current by binding to voltage-gated Na+ channels. LA drugs in clinical use are tertiary amines and exhibit both tonic and use-dependent block (UDB) of Na+ current. UDB by LA drugs is the hallmark of their antiarrhythmic activity; it enables these drugs to be more effective when the frequency of action potentials is high such as in ventricular tachycardia. During UDB, blocked channels accumulate because of incomplete recovery of drug-bound channels during diastole. Slowed recovery of drug-bound channels can be explained by a combination of the modulated-receptor hypothesis 2,3 and the guarded-receptor model.4,5 The modulated-receptor hypothesis proposes that higher affinity for LA drugs during activated and inactivated gating states slows unbinding of drug and, consequently, recovery of the drug-bound channels. The guarded-receptor model emphasizes that a state-dependent availability of the drug access path to and from the binding site influences apparent binding and unbinding kinetics.
LA drugs block Na+ channels by binding to a site within the pore below the selectivity filter and above the activation gate at the inner pore mouth. 6,7 Specific S6 residues from domains 1, 3, and 4 (D4S6) have been identified as parts of the binding site. 8–11 Hille2 has proposed two paths to and from the binding site, a fast hydrophobic path for neutral drug and a slower hydrophilic path for charged drug. One hydrophilic path common to all channel isoforms is the inner pore, guarded by the activation gate. Lidocaine is a tertiary amine with a pKa near the physiological level of pH, and it exists in both cationic and neutral forms, allowing it to use both pathways. Onset and recovery-rate constants for UDB by lidocaine increase as the external pH is raised, but internal pH has little effect. 12,13 Thus, external pH can influence kinetics of UDB for tertiary amine drugs by influencing the fraction ionized and the relative use of fast hydrophobic and slow hydrophilic paths.
Quaternary amine QX analogues of LA drugs are membrane-impermeant and are inactive from the outside of neuronal and skeletal muscle isoforms of Na+ channels. 2,7,14 In contrast, external QX is effective in blocking cardiac Na+ channels, 15 and isoform-specific residues determine this external access path. Qu et al 16 identified an isoform-specific residue in the upper part of D4S6 (Thr in heart, Val in brain, and Cys in μ1) that is a determinant for external QX block, and we identified an isoform-specific residue within the P-loop of domain 1 (Cys in heart, Phe in brain, and Tyr in μ1).14 It is important to determine if this external access path also influences the action of lipophilic LA antiarrhythmic drugs in clinical use. In addition, better definition of the factors controlling LA kinetics may assist both in understanding the action of currently used drugs and in development of better drugs.
In our study, we characterize the properties of these paths defined by the P-loop (C373) and D4S6 (T1752) isoform-specific residues in human heart channel, hH1a, by studying recovery from UDB by lidocaine. We show that these isoform-specific residues influence the lidocaine recovery kinetics and define cardiac-specific external paths for lidocaine. The data are fit into a mathematical model17 to separate the recovery-time constants for charged and neutral lidocaine molecules. From this result, we propose a model that suggests the impact of the external path on the drug kinetics.
Materials and Methods
Clones, Molecular Biology, and Expression of Na+ Channels in Xenopus Oocytes
The hH1a α-subunit (a gift from Dr H.A. Hartmann, 18 University of Maryland, Baltimore, Md) is one of three clones of the normal human heart channel and differs in its amino acid numbering by one from the hH1 clone. 19 The mutants h-C373Y, h-T1752V, and h-(C373Y/T1752V) were prepared by oligonucleotide-directed mutagenesis with the QuikChange Multi Site-Directed Mutagenesis Kit (Stratagene). The μ1 α-subunit (provided by Dr J.R. Moorman, University of Virginia, Charlottesville, Va) and the P-loop mutant μ1-Y401C were prepared as described previously. 7,14 Stage V and stage VI Xenopus oocytes were isolated and injected with ≈50 to 100 nanograms of cRNA, in vitro-synthesized by a T7 or SP6 transcription system (Ambion), using hH1a or rat skeletal muscle (μ1) α-subunit clones. An equimolar cRNA of the human β1 subunit was included in all injections. The oocytes were incubated at 16°C for 24 to 72 hours before the experiments.
Electrophysiology and Data Analyses
Experiments used two-electrode whole-cell voltage-clamp of Xenopus oocytes using a Dagan CA-1 system (Dagan Corporation) with pCLAMP6 or pCLAMP8 software (Axon Instruments, Inc) at 50 to 60 kHz and filtered at 2 kHz. The experimental chamber of ≈200 μL was filled with a buffer containing (in mmol/L) NaCl 90, CaCl2 1, MgCl2 1, KCl 2.5, and HEPES (OR2) 5 at 23°C to 25°C with a flow rate of 0.6 mL/min during all experiments. Buffer pH was adjusted with NaOH. Lidocaine solution was made in water as 100 mmol/L stock solutions in small aliquots and stored at −20°C until dilution into OR2 at the time of experiments. The electrodes were filled with 3 mol/L KCl, with resistances between 0.3 and 0.8 MΩ. Leak currents were not subtracted, but oocytes displaying leak currents >0.05 of the peak current estimated from current-voltage relationship experiments were not used. The capacitance transients were adjusted with series resistance compensation. Data from the current-voltage relationship (−90 to +50 mV) were normalized and fit to the Boltzmann equation,
where current I is a function of t, to obtain the following: Gmax, maximum conductance; Vrev, reversal potential; V1/2, half-maximal voltage; and s, slope factor. Likewise, steady-state availability (−130 to −30 mV) experiments were fit to the Boltzmann equation to obtain Gmax, V1/2, and s. The test pulses were to −20 mV for 10 ms. Tonic-block experiments, which included a test pulse from the holding potential to −20 mV every 20 or 40 seconds to follow developing tonic block, lasted for 15 minutes in 500 μmol/L lidocaine. UDB was induced by repetitive pulses to −20 mV for 10 ms from the holding potential at 20 cycles/second after the tonic block. The UDB-recovery experiments consisted of UDB train followed by varying interval of recovery time at the holding potential. Only oocytes expressing ≤5 μA of peak currents were used, in order to minimize errors originating from limitations of the two-electrode whole-cell voltage clamp. Recovery from slow inactivation was assessed with a 5-second prepulse to −20 mV, followed by varying length of recovery time at the holding potentials. The data were fit into a double- or triple-exponential equation to obtain the time constants for slow inactivation or intermediate inactivation, IM. 20,21 Data are reported as mean±SEM. Two sets of data are compared using unpaired Student’s t test, and P<0.05 is used to define statistical significance.
Although effects of the access path on UDB may be made by direct comparison between the cardiac skeletal muscle and brain channels, these isoforms have different gating properties that also influence LA block. 22,23 Consequently, we compared lidocaine-induced UDB in the background of the cardiac isoform between the wild-type that has known QX access and mutants containing a single residue substitution in the D4 P-loop and D4S6 that should reduce access. In addition, we studied the reverse mutations in the skeletal muscle isoform for comparison with prior QX studies. To accommodate the difference in steady-state availability curves, we set the holding potentials of −120 mV for hH1a clones and −100 mV for μ1 clones.
Recently, several reports suggested a close relationship between intermediate inactivation (IM) and lidocaine block of the Na+ channel. 20,21 Recovery-time constants for IM were not significantly different among our wild-type and mutant cardiac channels (Table 1). Furthermore, the recovery-time constants for IM did not shorten when pH was raised (57.5±11.6, 58.2±4.8, 68.8±8.1, and 71.4±10.8 ms for h-WT, h-C373Y, h-T1752V, and h-(C373Y/T1752V), respectively; n=5 to 6), unlike those for recovery from lidocaine block. Therefore, the differences in recovery from lidocaine UDB described next likely did not result from differences in IM among the clones.
P-Loop Isoform-Specific Mutation Attenuates Lidocaine Recovery From UDB
The isoform-specific residue Cys373 in the outer vestibule of hH1a, which permits external QX access,14 is thought to be in the hydrophilic path for ion permeation because it interacts well with guanidium toxins and with external Cd2+. 24,25 If the QX path defined by C373 in the heart channel provides an egress path for lidocaine not present in μ1, mutation to the μ1-specific residue h-C373Y would be expected to slow the recovery from UDB compared with wild-type hH1a. Lidocaine has a pKa of 7.86 in bulk solution (see next section) and exists in both charged and neutral forms in physiological pH. If this path is hydrophilic, the difference in kinetics will be more pronounced at lower pH where the charged form of lidocaine predominates and will become less significant as pH is raised.
where c(t) is current after t milliseconds of recovery after the train of pulses to induce UDB; τ1 and τ2 are time constants for fast and slow components, respectively; b∞ is the current at t=∞ (equals the tonic-block level); and a1 and a2 are amplitudes for fast and slow components, respectively. The fast time constant relates to recovery from inactivation by drug-free channels, and the slower time constant is thought to reflect unbinding of the drug.26 The faster time constants ranged from 2 to 3 ms without consistent patterns and were not altered significantly by lidocaine (not shown). The slower time constants reflecting the drug unblock are listed in Table 2. As predicted, the recovery from UDB was slowed by the mutation h-C373Y (Table 2 and Figure 1A). At higher pH, the recovery-time constants for both wild-type (h-WT) and h-C373Y decreased and the differences disappeared, suggesting that the effect of C373Y was mostly on the charged drug. The reverse mutation in μ1, μ1-Y401C, accelerated the recovery of UDB. Therefore, the external QX path defined by the isoform-specific P-loop residue significantly influenced recovery kinetics for lidocaine. Furthermore, the effect was mostly on charged lidocaine, suggesting that this external path is hydrophilic.
D4S6 Isoform-Specific Residue Attenuates Recovery From UDB
Qu et al16 showed that when an isoform-specific residue in D4S6 was switched from valine in brain to threonine in heart channel, recovery from QX block slowed. The impact of “closing” this D4S6 path by T1752V mutation in the heart channel, judged by the attenuation of recovery kinetics, was somewhat less than that of the P-loop path and it diminished at pH 8.30. The S6 mutant showed a relatively constant degree of attenuation between pH 7.20 and pH 8.00, suggesting that the S6 mutation may have affected a path used by both charged and neutral forms of lidocaine (Table 2). We propose two possible models for our findings. First, there is one external path, which is influenced more significantly by the P-loop residue than by the S6 residue. Alternatively, the P-loop and S6 residues define two independent paths. The latter fits with our recent observation that cysteine-modifying MTSEA treatment of μ1-Y401C decreased external QX block, but MTSEA treatment of μ1-(Y401C/C1572T) did not effectively lessen external QX block additionally (A.S. and H.A.F., unpublished data, 2000). These two observations suggest that C1572T mutation in μ1, analogous to T1752 in hH1a, opened up a nonpore QX path, which did not overlap with that of the Y401C mutation. Taken together, this raises the interesting possibility that the D4S6 path defined by T1752, like that defined by C1572 in μ1, may represent a separate nonpore “protein path” that influences unblock of both charged and neutral forms of lidocaine. However, there are no direct data in support of two separate external paths for lidocaine at the present time.
In Qu et al,16 substitution of D4S6 threonine residue for valine in the rat heart clone attenuated external QX block, but not to the level of the brain channel, likely reflecting the contribution of the P-loop path in the external QX block. “Shutting down” both P-loop and S6 external QX paths with a double mutant, h-(C373Y/T1752V), resulted in a more significant effect on the recovery kinetics (Table 2) that disappeared as the external pH was raised. Interestingly, the effects of the P-loop and D4S6 paths seemed to be additive. Thus, there seems to be little overlap in the impact on drug egress between the P-loop and S6 residues, perhaps suggesting two independent paths.
Tonic and UDB
Tonic block levels were similar for h-WT and h-C373Y and for μ1-WT and μ-Y401C at a given pH (Table 3 and Figure 2). UDB at 20 Hz (10-ms pulses) resulted in further decrease of current with the steady-state block reaching near 90%. Each clone showed enhanced tonic block and less UDB at pH 8.30 compared with pH 7.20, resulting in a similar level of final steady-state block (tonic plus UDB) (Figure 2). The h-T1752V and h-(C373Y/T1752V) also showed similar degree of tonic and UDB (Table 3). V1/2 values of steady-state inactivation with and without lidocaine were not significantly altered in these mutants compared with h-WT (Table 1). The results suggest that the mutations did not cause significant alterations in intrinsic affinities for the drug. We modeled the onset kinetics of UDB by fitting the normalized data with a single exponential term,
where b(n) indicates block at the nth pulse; b∞, the steady-state level of remaining fraction of current after UDB (reflecting UDB only); and τ, the first-order time constant of decay. In Table 3, time constants and amount of UDB, (1−b∞) in Equation 3, are listed. First-order onset time constants of UDB were increased in h-C373Y, h-T1752Y, and h-(C373Y/T1752V) compared with h-WT. The change with h-C373Y (versus h-WT) was not statistically significant, perhaps because of the difficulty of measuring these short time constants with a pulse train of 20 Hz, or 50 ms per episode. As in recovery, onset kinetics of UDB also accelerated as the pH was raised, and the effects of the mutations disappeared at pH 8.30.
Modeling of the Drug Egress Paths
Extracellular pH influences ionization of bound lidocaine molecules into charged and neutral forms based on the pKa. 12,13 In a simplified model of drug and channel kinetics (Figure 3A), recovery (or unbinding of drug) of the individual drug-bound channel can be described in a first-order exponential decay function with a rate constant of 1/τ. The apparent rate constant for the scheme in Figure 3A is as follows: kapparent=1/τ apparent=(fraction of BL+)/τ charged+(fraction of BL0)/τneutral. After rearranging, it becomes τapparent=(1+R)/(R/τ neutral+1/τcharged), where R denotes the fraction of BL0/BL + and equals, by Henderson-Hasselbalch relationship, 10(pH−pKa). 17 When the data in Table 2 for hH1a clones were fit into the equation, τcharged and τneutral were calculated to be 394 and 64.9 ms, respectively, in good agreement with those obtained by Broughton et al 17 using isolated myocytes at two pH points (pH 6.9 and 7.4), despite different experimental conditions and methods; τcharged and τneutral were 476 and 43 ms, respectively. Interestingly, a similar treatment for h-C373Y yielded τcharged and τneutral of 704 and 68.6 ms, respectively (Figure 3B and Table 4). We conclude, in a more quantitative manner than comparing time constants, that the isoform-specific mutation C373Y altered the egress path for charged molecules without significantly affecting that of neutral molecules; τcharged increased 78% and τneutral increased 6%. A converse relationship resulted when μ1-WT and μY401C were compared (Figure 3C). The S6 path likely influences both charged and neutral lidocaine, given modest effects on both τcharged and τneutral; τcharged increased by 34% and τneutral increased by 10%. The data for h-(C373Y/T1752V) showed the greatest increase in τcharged with a slight increase in τneutral similar to that in h-T1752V (Table 4). These results are suggestive of a model with two separate external paths in which the isoform-specific residue Cys373 in hH1a defines the external hydrophilic egress paths, and Thr1752 in hH1a defines a less potent, external nonpore protein path.
pKa values of lidocaine in a bulk solution depend on the buffer used and temperature. 1 Apparent pKa might be different near the binding site that includes aromatic residues. 27 We found similar relationships between the time constants for charged and neutral lidocaine persisted even when pKa values were assumed to be 8.00 or 8.20 (data not shown).
The outside QX access paths associated with P-loop and D4S6 cardiac isoform-specific residues also influence lidocaine block, as shown here for UDB kinetics. The more obvious effect is on the recovery rate from UDB, which plays an important role in the clinical actions of the tertiary amine LA drugs. Onset kinetics were also changed, but limitations of the protocols made these changes less accurate and consistent. Limited access/egress paths would be expected to result in higher level of UDB. We did observe a trend of increased UDB in our mutants [Table 3, see (1−b∞) at pH 7.20 for the heart channels], even with limitations in the protocols.
The effects of access paths on lidocaine appear to be related to the charged form of the drug, which most closely resembles the permanently charged QX analogues. This is consistent with the expectation that the P-loop mutation C373Y alters a hydrophilic path. This interpretation fits well the theoretical framework proposed by Broughton et al. 17 The role of the hydrophilic path was clear for the P-loop-induced path, but less clear for the D4S6 path. The two paths seemed to be affected differently by pH change, and they are additive, leading us to suggest that the paths are separate, with the D4S6 path being through the protein itself. However, it remains equally possible that there is only one external access path, which is more influenced by the P-loop residue than the D4S6 residue. The much faster hydrophobic path also made it more difficult to measure the outside access path(s) with alkalinization.
How important is the external path? If we assume that the mutations used in these experiments had no direct effects on the inside access path and that the C373Y and T1752V paths are additive as suggested from our results, either from sharing little overlap or being separate, we can estimate their relative effects on recovery rates. For example, in h-WT the overall recovery process through the external hydrophilic path is described as follows:
where kcharged, kP-loop, kS6, and kcytoplasmic represent rate constants for overall, P-loop, S6, and cytoplasmic paths, respectively; and τP-loop, τS6, and τcytoplasmic represent time constants for P-loop, S6, and cytoplasmic paths, respectively. The fitted time constants are taken from Table 4 and the results are shown in Figure 4. At this time, we draw one common external path for both the P-loop and S6 residues due to lack of direct evidence for their spatial independence.
We start with h-(C373Y/T1752V), in which the S6 and P-loop paths are negligible. Thus, τcytoplasmic=τ charged=1030 ms. In h-T1752V, the P-loop path (Cys373) path is “opened,” resulting in speeding up of recovery from UDB for charged molecules. Impact of opening up the P-loop path can be estimated. kcharged=1/τ charged=(1/526)= kP-loop+ kcytoplasmic=(1/τ P-loop)+(1/τ cytoplasmic)=(1/τ P-loop)+(1/1030). Thus, τP-loop=1070. In a similar manner, from h-C373Y, τS6=2220. In h-WT, both of the external paths are open. Then, kexternal=[(1/τ P-loop)+(1/τ S6)]=(1/τ charged)−(1/τ cytoplasmic)=(1/394)−(1/1030)= 1.56×10−3 ms−1, or τexternal of 638 ms. Interestingly, adding the rate constants for the P-loop and S6 paths from h-C373Y and h-T1752, we obtain kexternal=1.38×10 −3 ms−1, or τexternal of 724 ms, a fairly good agreement with our estimate obtained directly from h-WT. We conclude that the two external paths together play a more significant role in determining drug kinetics than the cytoplasmic path (τexternal= 638 ms versus τcytoplasmic=1030 ms).
The μ1 Na+ channel isoform has different gating kinetics, so it is difficult to compare its UDP recovery rates directly with those of the cardiac isoform. Nonetheless, the direction of change in recovery for the P-loop mutation Y401C is as expected. Given the different gating kinetics and holding potentials in our protocols, it is difficult to expect that the recovery-time constants for μ1-WT would be the same as those for h-(C373Y/T1752V). However, we cannot rule out the possibility that the cardiac isoform has another as yet unidentified residue contributing to its outside access path(s).
Hille28 estimated the narrowest part of the Na+ channel pore to be ≈3Å by 5Å. It is difficult to imagine how molecules such as lidocaine or QX are able to access the binding site located below the selectivity filter through the pore from outside, as implied by previous studies 7,14,16 and our present report. Moorman et al29 proposed a model in which all bound drug unblocks in a neutral form through a hydrophobic route alleviating the need for permeation through the narrow selectivity filter. In their model, deprotonation rate of the bound drug is the rate-limiting step in recovery from drug block. Our data from h-WT fit their model fairly well, but it could not accommodate the data from our P-loop, D4S6, and double mutants; the deprotonation rates no longer became rate-limiting (not shown). Indeed, Wendt et al,30 by deuterium exchange experiments, reported that proton exchange was rapid and not rate-limiting. Recently, Huang et al 31 observed that large hydrophobic molecules of diameter up to 15Å could permeate through the Na+ channel with certain selectivity filter mutants and suggested that hydrophobic interphases contiguous to the pore can facilitate permeation of large molecules. Perhaps the charged aliphatic portion of the LA drug uses the direct route through the pore through the selectivity filter and the rest of the molecule “slips out” of the pore through interphases between P-loops and S6 helices. 32
Our schematic model presented above, although semiquantitative, offers an opportunity to visualize the paths important in lidocaine UDB in terms of their impact on drug kinetics. We conclude that, in the heart channel, the external paths for lidocaine may play a more significant role in drug kinetics than the cytoplasmic path. Drugs using such external paths would be specifically influenced by their ionization states, ability/inability to cross membrane, and any factors affecting the external channel structures. Drugs that block the channel only in the open state might be more profoundly affected when hydrophilic pore paths are altered. Recently, Grant et al 33 reported data suggesting that disopyramide and flecainide use different paths to block and unblock the Na+ channel during UDB. It is likely that the relative potencies of the different egress paths described in our study are drug-specific. Understanding the natures of the access/egress paths specific to an antiarrhythmic drug is likely to be crucial in understanding its characteristics of UDB of the Na+ channel.
This study was supported by grants from the NIH, HL07381 (P.J.L.) and HL65661 (A.S. and H.A.F.). We thank Ian Glasser and Blair Davis for technical help and Dr Dorothy Hanck for helpful discussions.
Original received June 8, 2001; revision received September 28, 2001; accepted September 28, 2001.
Butterworth JF, Strichartz GR. Molecular mechanisms of local anesthesia: a review. Anesthesiology. 1990; 72: 711–734.
Hille B. Local anesthetics: hydrophilic and hydrophobic pathways for the drug-receptor reaction. J Gen Physiol. 1977; 69: 497–515.
Hondeghem LM, Katzung BG. Time- and voltage-dependent interactions of the antiarrhythmic drugs with cardiac sodium channels. Biochim Biophys Acta. 1977; 472: 373–398.
Starmer CF, Grant AO, Strauss HC. Mechanisms of use-dependent block of sodium channels in excitable membranes by local anesthetics. Biophys J. 1984; 46: 15–27.
Starmer CF, Grant AO. Phasic ion channel blockade: a kinetic model and parameter estimation procedure. Mol Pharmacol. 1985; 28: 348–356.
Hille B. Mechanisms of block.In: Ionic Channels of Excitable Membranes. 2nd ed. Sunderland, Mass: Sinauer Assoc, Inc; 1992: 390–422.
Sunami A, Dudley SC, Fozzard HA. Sodium channel selectivity filter regulates antiarrhythmic drug binding. Proc Natl Acad Sci U S A. 1997; 94: 14126–14131.
Ragsdale DS, McPhee JC, Scheuer T, Catterall WA. Molecular determinants of state-dependent block of Na+ channels by local anesthetics. Science. 1994; 265: 1724–1728.
Wang GK, Quan C, Wang S-Y. Local anesthetic block of batrachotoxin-resistant muscle Na+ channels. Mol Pharmacol. 1998; 54: 389–396.
Wang S-Y, Nau C, Wang GK. Residues in Na+ channel D3-S6 segment modulate both batrachotoxin and local anesthetic affinities. Biophys J. 2000; 79: 1379–1387.
Yarov-Yarovoy V, Brown J, Sharp EM, Clare JJ, Scheuer T, Catterall WA. Molecular determinants of voltage-dependent gating and binding of pore-blocking drugs in transmembrane segment IIIS6 of the Na+ channel α subunit. J Biol Chem. 2001; 276: 20–27.
Hille B. The pH-dependent rate of action of local anesthetics on the node of Ranvier. J Gen Physiol. 1977; 69: 475–496.
Schwarz W, Palade PT, Hille B. Local anesthetics: effect of pH on use-dependent block of sodium channels in frog muscle. Biophys J. 1977; 20: 343–368.
Sunami A, Glaaser IW, Fozzard HA. A critical residue for isoform difference in tetrodotoxin affinity is a molecular determinant of the external access path for local anesthetics in the cardiac sodium channel. Proc Natl Acad Sci U S A. 2000; 97: 2326–2331.
Alpert LA, Fozzard HA, Hanck DA, Makielski JC. Is there a second external lidocaine binding site on mammalian cardiac cells? Am J Physiol. 1989; 257: H79–H84.
Qu Y, Rogers J, Tanada T, Scheuer T, Catterall WA. Molecular determinants of drug access to the receptor site for antiarrhythmic drugs in the cardiac Na+ channel. Proc Natl Acad Sci U S A. 1995; 92: 11839–11843.
Broughton A, Grant AO, Starmer CF, Klinger JK, Stambler BS, Strauss HC. Lipid solubility modulates pH potentiation of local anesthetic block of Vmax reactivation in guinea pig myocardium. Circ Res. 1984; 55: 513–523.
Hartmann HA, Tiedeman AA, Chen SF, Brown AM, Kirsch GE. Effects of III-IV linker mutations on human heart Na+ channel inactivation gating. Circ Res. 1994; 75: 114–122.
Gellens ME, George AL, Chen LQ, Chahine M, Horn R, Barchi RL, Kallen RG. Primary structure and functional expression of the human cardiac tetrodotoxin-insensitive voltage-dependent sodium channel. Proc Natl Acad Sci U S A. 1992; 89: 554–558.
Kambouris NG, Hastings LA, Stepanovic S, Marban E, Tomaselli GF, Balser JR. Mechanistic link between lidocaine block and inactivation probed by outer pore mutations in the rat μ1 skeletal muscle sodium channel. J Physiol. 1998; 512: 693–705.
Chen Z, Ong BH, Kambouris NG, Marban E, Tomaselli GF, Balser JR. Lidocaine induces a slow inactivated state in rat skeletal muscle sodium channels. J Physiol. 2000; 524(pt 1): 37–49.
Wright SN, Wang SY, Kallen RG, Wang GK. Differences in steady-state inactivation between Na channel isoforms affect local anesthetic binding affinity. Biophys J. 1997; 73: 779–788.
Nuss HB, Kambouris NG, Marban E, Tomaselli GF, Balser JR. Isoform-specific lidocaine block of sodium channels explained by differences in gating. Biophys J. 2000; 78: 200–210.
Satin J, Kyle JW, Chen M, Bell P, Cribbs LL, Fozzard HA, Rogart RB. A mutant of TTX-resistant cardiac sodium channels with TTX-sensitive properties. Science. 1992; 256: 1202–1205.
Backx PH, Yue DT, Lawrence JH, Marban E, Tomaselli GF. Molecular localization of an ionic binding site within the pore of mammalian sodium channels. Science. 1992; 257: 248–251.
Bean BP, Cohen CJ, Tsien RW. Lidocaine block of cardiac sodium channels. J Gen Physiol. 1983; 81: 613–642.
Nettleton J, Wang GK. pH-dependent binding of local anesthetics in single batrachotoxin-activated Na+ channels. Biophys J. 1990; 58: 95–106.
Hille B. Selective permeability: independence.In: Ionic Channels of Excitable Membranes. 2nd ed. Sunderland, Mass: Sinauer Assoc, Inc; 1992: 337–389.
Moorman JR, Yee R, Bjornsson T, Starmer CF, Grant AO, Strauss HC. pKa does not predict pH potentiation of sodium channel blockade by lidocaine and W6211 in guinea pig ventricular myocardium. J Pharmacol Exp Ther. 1986; 238: 159–166.
Wendt DJ, Starmer CF, Grant AO. pH dependence of kinetics and steady-state block of cardiac sodium channels by lidocaine. Am J Physiol. 1993; 264: H1588–H1598.
Huang CJ, Favre I, Moczydlowski E. Permeation of large tetra-alkylammonium cations through mutant and wild-type voltage-gated sodium channels as revealed by relief of block at high voltage. J Gen Physiol. 2000; 115: 435–454.
Lipkind GM, Fozzard HA. KcsA crystal structure as framework for a molecular model of the Na+ channel pore. Biochemistry. 2000; 39: 8161–8170.
Grant AO, Chandra R, Keller C, Carboni M, Starmer CF. Block of wild-type and inactivation-deficient cardiac sodium channels IFM/QQQ stably expressed in mammalian cells. Biophys J. 2000; 79: 3019–3035.