Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Voltage-gated K⁺ channels (Kv) are multimeric membrane proteins composed of four  subunits that are often associated with accessory  subunits. Not only are  subunit mutations responsible for human disease but also  subunits (minK and MiRP1) are involved in severe pathologies, such as the congenital long QT syndrome (Splawski et al., 1997; Abbott et al., 1999). The hKv1.5 channel encodes the ultrarapid delayed rectifier K⁺ current present in the human atrium but not in the ventricle (Fedida et al., 1993; Snyders et al., 1993; Wang et al., 1993; Feng et al., 1997). It is also widely expressed in the vascular system (Coppock and Tamkun, 2001). Two  subunits that show overlapping expression patterns with Kv1.5, Kv1.3, and Kv2.1 shift the midpoint of the activation curve toward more negative voltages and slow deactivation (England et al., 1995; Uebele et al., 1996). The Kv2.1 subunit increases the degree of slow inactivation of the current, whereas Kv1.3 converts hKv1.5 from a delayed rectifier with a modest degree of slow inactivation to a channel with both fast and slow components of inactivation (Uebele et al., 1996; Uebele et al., 1998). The regional distribution of Kv1.3 subunits in the myocardium is not homogeneous, with higher expression in the ventricle than in atria (Wang et al., 1996). If  subunits alter the pharmacology of the Kv  subunits, the molecular targets of local anesthetics and/or antiarrhythmic drugs will vary greatly depending on  subunit expression. Therefore, a further knowledge of the effects of regulatory  subunits present in the human myocardium is required. Quinidine-induced block of Kv1.5 is not modified by Kv2.1 subunit (Yeola et al., 1996). However, the possible effects of Kv1.3 on drug-hKv1.5 channel interactions have not been yet analyzed.

This study was undertaken to determine the pharmacological consequences of the interaction between Kv1.5 and Kv1.3 subunits. Bupivacaine, its quaternary ammonium derivative [R(+)-N-methyl-bupivacaine (RB⁺1C)], and quinidine sensitivity were examined. Bupivacaine is a very cardiotoxic local anesthetic that can prolong the QT interval and induce torsades de pointes (Kasten and Martin, 1985; Kasten, 1986; Covino, 1987). At the same concentrations needed to block Na⁺ channels, bupivacaine binds to two different receptors on hKv1.5 channels, one external and one internal (Valenzuela et al., 1995; Longobardo et al., 2000, 2001). Quinidine is a classic antiarrhythmic agent that has been studied extensively (Snyders et al., 1992; Yeola et al., 1996). The internal bupivacaine binding site is the same as that described for quinidine (Yeola et al., 1996; Franqueza et al., 1997). In the present study, we have analyzed the effects of bupivacaine, RB⁺1C, and quinidine on hKv1.5 channels expressed alone or with Kv1.3 in  subunit-free HEK293 cells (Uebele et al., 1996) and those produced by quinidine on Kv1.5+Kv1.3. These results are in contrast to our previous studies with these three drugs using hKv1.5 channels expressed in Ltk cells that endogenously express the Kv2.1 subunit (Uebele et al., 1996; Longobardo et al., 2000; González et al., 2001). A preliminary report of the present study has been published in abstract form (Navarro-Polanco et al., 2001).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Expression Systems and Transfection. Human Kv1.5 (22-1894 nt) and Kv1.3 (53-1500 nt) were inserted in tandem into the same pBK vector with the Kv1.5 subunit placed 3' to the Kv1.3 subunit and behind an internal ribosome entry sequence, thus generating a dual cistronic mRNA as described previously (Kwak et al., 1999). The pBK construct used for Kv1.5 alone has been described previously (Uebele et al., 1998). HEK293 cells were cultured in minimal essential medium supplemented with 10% bovine fetal serum, penicillin-streptomycin (Sigma, St. Louis, MO), and nonessential amino acids 1%. Transfection of hKv1.5 or Kv1.5+Kv1.3 channel (0.3 µg) and reporter plasmids CD8 (1.6 µg) or green fluorescent protein pCI was performed by use of LipofectAMINE (10 µl). Before experimental use, cells were incubated with polystyrene microbeads precoated with anti-CD8 antibody (Dynabeads M450; Dynal Biotech, Oslo, Norway), as described previously (González et al., 2001) or viewed under fluorescence optics to identify the transfected cells.

Electrophysiological Technique and Data Acquisition. The intracellular pipette filling solution contained 80 mM K-aspartate, 50 mM KCl, 3 mM phosphocreatine, 10 mM KH₂PO₄, 3 mM MgATP, 10 mM HEPES-K, and 5 mM EGTA, and was adjusted to pH 7.25 with KOH. The bath solution contained 130 mM NaCl, 4 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, 10 mM HEPES-Na, and 10 mM glucose, and was adjusted to pH 7.40 with NaOH. Permanently charged R(+)-bupivacaine [RB⁺1C ((R)-(+)-1-butyl-1-methyl-2',6'-pipecoloxylidide)] (Astra, Södertälje, Sweden) was dissolved in ethanol (50%). Bupivacaine and quinidine (Sigma) were dissolved in distilled deionized water to yield stock solutions of 10 mM.

Statistical Analysis. Data are presented as mean values ± S.E.M. Comparisons between mean values in control conditions and mean values in the presence of drug for a single variable were performed by use of paired Student's t test. One-way analysis of variance was used to compare more than two groups. Statistical significance was set at p < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Figure 1 shows original records for hKv1.5 channels expressed alone (Fig. 1A) and with Kv1.3 (Fig. 1B) obtained after applying depolarizing pulses from a holding potential of 80 to +60 mV in 10-mV steps. Tail currents are shown after repolarization to 40 mV. In the absence of  subunits, the hKv1.5 channels inactivate by 11.6 ± 0.7% (n = 40) at the end of 200-ms depolarizing pulses to +60 mV. In the presence of Kv1.3, this current exhibited a fast initial but incomplete inactivation that averaged 77.6 ± 4.0% (n = 12) with a time constant of 3.7 ± 0.2 ms (n = 12). In addition, the Kv1.3 subunit slows the time course of the deactivation process that became monoexponential (Fig. 1, bottom).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Original records obtained upon depolarization from a holding potential of 80 mV to +60 mV in 10 mV steps and upon repolarization to 40 mV. A, current records obtained from the activation of Kv1.5 subunits. B, current records of hKv1.5 channels in the presence of the Kv1.3 subunit. Note that the deactivation process (bottom) is slower in the presence of Kv1.3.

Effects of RB⁺1C on hKv1.5 Subunits Alone and in the Presence of Kv1.3. RB⁺1C, externally applied, blocks hKv1.5 channels expressed in Ltk cells that endogenously express Kv2.1 subunits (Uebele et al., 1996) in a time- and voltage-independent manner (Longobardo et al., 2000). However, when it is applied internally, this drug mimics bupivacaine effects (Valenzuela et al., 1995; Franqueza et al., 1997; Longobardo et al., 2000). To analyze whether Kv1.3 and/or Kv2.1 modify the interaction between RB⁺1C and hKv1.5 subunits, we studied its effects on hKv1.5 subunits expressed alone or coexpressed with Kv1.3 in HEK293 cells. Figure 2A shows current traces through hKv1.5 channels in the absence and in the presence of RB⁺1C (50 µM). RB⁺1C inhibited hKv1.5 current by 22 ± 9% (n = 4) in a time- and voltage-independent manner. Figure 2B shows the effects of RB⁺1C on hKv1.5 coexpressed with Kv1.3 subunits. Under these conditions, RB⁺1C blocked K⁺ current by 16 ± 2% (n = 4). Thus, neither the Kv2.1 nor the Kv1.3 subunits altered RB⁺1C block of Kv1.5 via its external receptor site (p > 0.05).

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Effects of RB+1C (50 µM) on hKv1.5 alone (A) or in the presence of the Kv1.3 subunit (B). RB+1C induced the same degree of block under both experimental situations.

Effects of Bupivacaine on hKv1.5 Subunits. Fig. 3A shows the effects of bupivacaine (20 µM) on hKv1.5 channels expressed without a  subunit in HEK293 cells. At the end of 200-ms depolarizing pulses to +60 mV, bupivacaine inhibited this current by 61 ± 3% (n = 8). Bupivacaine decreased the current at all membrane potentials tested (Fig. 3B), although block increased more at positive than negative membrane potentials, consistent with an open channel block mechanism. All these data are similar to those found when blocking effects of bupivacaine on hKv1.5 channels were studied in Ltk cells (González et al., 2001), indicating that the Kv2.1 subunit does not modify bupivacaine affinity of hKv1.5 channels.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Effects of bupivacaine (20 µM) on hKv1.5 channels. A, current records obtained in the absence and in the presence of bupivacaine. B, IV relationships obtained in the absence () and in the presence of bupivacaine (). Block induced by bupivacaine measured at +60 mV averaged 61 ± 3%. Each point represents the mean ± S.E.M. of seven experiments. C, original records obtained after depolarization from 80 to +60 mV in the absence and in the presence of bupivacaine (20 and 50 µM). Note the fast initial decay of the current induced by the drug. Relationship between 1/Block and bupivacaine concentration. The fast time constant of the biexponential fit of the current traces in the presence of different bupivacaine concentrations were considered a good approximation of Block values (see text). For a first-order blocking scheme, a linear relation is expected: 1/Block = k × [Bupivacaine] + l. The solid line represents the linear fit, from which the apparent binding and unbinding rate constants were obtained. D, tail currents obtained upon repolarization from +60 to 40 mV in the absence and in the presence of bupivacaine. Arrow, "crossover" characteristic of an open-channel block mechanism.

1

1

1

1

1

1

Effects of Bupivacaine on hKv1.5 Subunits in the Presence of Kv1.3. Fig. 4A shows current through hKv1.5+Kv1.3 in the absence and in the presence of bupivacaine (100 µM). Bupivacaine decreased this current at +60 mV by 62 ± 4% (n = 4); i.e., its potency was ~4-fold lower than in hKv1.5 channels expressed alone or in the presence of Kv2.1 subunits. Figure 4B shows the IV relationship of hKv1.5 when expressed with Kv1.3. Block induced by bupivacaine was voltage dependent, achieving a maximum value at 0 mV and decreasing at more positive membrane potentials (74 ± 2% versus 62 ± 4%, measured at 0 and +60 mV, respectively, n = 4, p < 0.05). Figure 5 shows the concentration dependence of bupivacaine block of hKv1.5 and hKv1.5+Kv1.3 channels when using as index of block the suppression of the current at the end of 200-ms depolarizing pulses to +60 mV. In addition, Fig. 5 shows the concentration dependence of bupivacaine block of hKv1.5+Kv2.1 channels taken from a previous study (dashed line) (González et al., 2001). IC₅₀ values for blocking hKv1.5 and hKv1.5+Kv2.1 channels were similar (13.1 ± 0.8 µM, n = 25, versus 8.9 ± 1.4 µM, n = 22, respectively, p > 0.05), whereas the IC₅₀ value for the blockade of hKv1.5+Kv1.3 channels was ~4-fold higher (IC₅₀ = 47.5 ± 5.1 µM, p < 0.01). Under both experimental conditions, the n_H values were close to unity, averaging 0.96 ± 0.05 and 0.74 ± 0.06 for Kv1.5 and Kv1.5+Kv1.3 channels, respectively. When the IC₅₀ values were calculated fixing the n_H value at 1, the obtained IC₅₀ values were 12.7 ± 1.6 µM and 47.6 ± 7.8 µM for Kv1.5 and Kv1.5+Kv1.3 channels, respectively; suggesting that binding of one drug molecule was necessary to block K⁺ channel efflux.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Effects of bupivacaine (100 µM) on hKv1.5 channels in the presence of the Kv1.3 subunit. A, original records obtained in the absence and in the presence of bupivacaine. B, IV relationship obtained in the absence () and in the presence of bupivacaine (). Block induced by bupivacaine measured at +60 mV averaged 62 ± 4%. Each point represents the mean ± S.E.M. of four experiments.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Concentration dependence of bupivacaine-induced block of hKv1.5 channels in the absence and in the presence of Kv1.3 subunits. Dashed line represents the dose-response curve obtained for bupivacaine block of hKv1.5 channels expressed in Ltk cells that endogenously express Kv2.1 subunits (taken from González et al., 2001). Reduction of current (relative to control) at the end of depolarizing steps from 80 to +60 mV was used as index of block. Each point represents the mean ± S.E.M. of three to seven experiments. The continuous line represents the fit of the experimental data to a Hill equation. **, p < 0.01.

1

1

1

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Time-dependent effects of bupivacaine block of Kv1.5+Kv1.3 channels. A, original records obtained upon depolarization from 80 to +60 mV in the absence and in the presence of bupivacaine. Time constant of fast inactivation induced by Kv1.3 was faster in the presence of bupivacaine. Inset, records obtained in the absence of drug normalized to the control value. B, relationship between 1/Block and bupivacaine concentration. Block values were obtained from the fit of the sensitive current [(IControl  IDrug)/IControl] during the first 12 ms (inset) at different bupivacaine concentrations (from 10 to 100 µM). Each point represents the mean ± S.E.M. of three to four experiments. For a first-order blocking scheme, a linear relation is expected: 1/Block = k × [Bupivacaine] + l. The solid line represents the linear fit, from which the apparent binding and unbinding rate constants were obtained. C, tail currents recorded upon repolarization from +60 to 40 mV in the absence and in the presence of drug exhibited a crossover (arrow).

Effects of Quinidine on hKv1.5+Kv1.3 Channels. Quinidine and bupivacaine share a common receptor site at hKv1.5 channels that is located at the S6 segment and that involves a polar (T505) and a hydrophobic amino acid (V512) (Yeola et al., 1996; Franqueza et al., 1997). To determine whether quinidine block of Kv1.5 channels is modified also, we studied the effects of this drug on Kv1.5+Kv1.3 channels transiently transfected in HEK293 cells. Figure 7A shows current traces through hKv1.5+Kv1.3 in the absence and in the presence of quinidine (100 µM). Quinidine decreased this current at +60 mV by 62 ± 9% (n = 5); i.e., its potency was ~8-fold lower than in hKv1.5 channels expressed alone or in the presence of Kv2.1 subunits (Snyders et al., 1992; Yeola et al., 1996). Figure 7B shows the concentration-response curve for the blocking effects of quinidine on Kv1.5+Kv1.3 measured at the end of 200-ms depolarizing pulses to +60 mV. The dashed line represents the concentration dependence of quinidine block of Kv1.5+Kv2.1 as reported previously (Snyders et al., 1992), which is similar to that observed in Kv1.5 (Yeola et al., 1996). IC₅₀ values in the absence and in the presence of Kv1.3 averaged 6.2 µM and 49.6 ± 4.2 µM (n = 12, p < 0.05), respectively.

View larger version (10K): [in this window] [in a new window]   Fig. 7.   Effects of quinidine (100 µM) on hKv1.5+Kv1.3 channels. A, original records obtained in the absence and in the presence of bupivacaine. B, concentration dependence of quinidine-induced block of hKv1.5+Kv1.3 channels. Reduction of Kv1.5+Kv1.3 current (relative to control) at the end of depolarizing steps from 80 to +60 mV was used as index of block. Each point represents the mean ± S.E.M. of three to five experiments. The continuous line represents the fit of the experimental data to a Hill equation. The dashed line represents the dose-response curve obtained for quinidine block of hKv1.5 channels expressed in Ltk cells that endogenously express Kv2.1 subunits (taken from Snyders et al., 1992). **, p < 0.01.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Whereas Kv1.3 and Kv2.1 do not modify the effects of RB⁺1C on the external binding site of bupivacaine on hKv1.5 channels, Kv1.3, but not Kv2.1, reduces bupivacaine and quinidine affinity for its internal receptor site on the hKv1.5 subunit. Quinidine and bupivacaine block hKv1.5 channels after binding to an external and to an internal receptor site (Yeola et al., 1996; Franqueza et al., 1997; Longobardo et al., 2000, 2001). Although the molecular determinants of the external binding site are unknown, quinidine, and bupivacaine share a common internal receptor site located at the S6 segment that involves a polar interaction with T505 and a hydrophobic interaction with V512 (Yeola et al., 1996; Franqueza et al., 1997). The membrane impermeant form of bupivacaine, RB⁺1C, produced a similar degree of block of hKv1.5 channels expressed in HEK293 cells in the absence or in the presence of Kv1.3. This block was also similar to that previously characterized in hKv1.5 channels expressed in Ltk cells, which endogenously express Kv2.1 subunit (Uebele et al., 1996; Longobardo et al., 2000). Moreover, block was time- and voltage-independent, as that found in Kv1.5 assembled with Kv2.1, indicating that the assembly of hKv1.5 subunits with Kv1.3 or Kv2.1 subunits does not modify the binding of the charged form of bupivacaine to its external receptor site in hKv1.5 channels.

Bupivacaine binding to its internal site inhibited current through hKv1.5 (Fig. 3) or hKv1.5+Kv2.1 (González et al., 2001) channels to a similar extent (IC₅₀ values of 13 µM and 9 µM, respectively), and in a similar fashion. Under both experimental conditions, bupivacaine induced an initial decay of the current and block increased at positive potentials over which channel activation occurred. Moreover, the superposition of the tail currents recorded under control conditions and in the presence of bupivacaine shows a `crossover' between them, indicating fast recovery from block during deactivation. All these findings are consistent with an open channel block mechanism (Armstrong, 1971). Kinetics of block differed mostly in the dissociation rate constant that was faster in the absence of Kv2.1, suggesting that block obtained with this subunit somehow stabilizes the bupivacaine-hKv1.5 interaction. Bupivacaine blocked hKv1.5+Kv1.3 channels to a lesser extent than hKv1.5 or hKv1.5+Kv2.1 channels (IC₅₀ = 48 µM). This lower potency to block Kv1.5+Kv1.3 was accompanied by a dramatic increase in the dissociation rate constant (46 s¹ versus 252 s¹), thus indicating a less stable drug-channel interaction when Kv1.3 is present. As with bupivacaine, quinidine sensitivity of Kv1.5+Kv1.3 channels increased ~8-fold in the presence of the Kv1.3 subunit (IC₅₀ = 50 µM), suggesting a common mechanism of action for local anesthetics and antiarrhythmic drugs, probably at their common internal receptor site at the S6 segment.

The most striking difference between current through hKv1.5 or hKv1.5+Kv2.1 and hKv1.5+Kv1.3 channels is the incomplete fast inactivation induced by Kv1.3. This fast inactivation involves an open channel block of the hKv1.5 subunit produced by the N terminus of the Kv1.3 subunit (inactivation "ball") (Uebele et al., 1998). These results suggest that the inactivation ball of the Kv1.3 subunit may compete with the open channel blocking drugs at the internal receptor site. Thus, this drug receptor site might be the "natural" receptor site for the inactivation ball of the Kv1.3 subunit (Yeola et al., 1996; Franqueza et al., 1997). However, internal pore mutations involved in stereoselective bupivacaine block of hKv1.5 channels, such as V512A or T505I, do not affect Kv1.3-mediated inactivation (Uebele et al., 1998). Moreover, an external pore mutation (R485Y) that decreases the slow inactivation of hKv1.5 channels and confers sensitivity to external tetraethylammonium dramatically increased the extent of Kv1.3-induced fast inactivation, suggesting that inactivation induced by Kv1.3 subunits involves open channel block that is allosterically linked to the external pore (Uebele et al., 1998). Therefore, one explanation for the present results would be that binding of the Kv1.3 inactivation particle allosterically modifies bupivacaine binding to the channel. Supporting this idea is the finding that Kv1.3 reduced bupivacaine affinity by increasing the dissociation rate constant, which could be indicative of an allosteric change in the drug binding site. Although recent studies suggest that the hydrophobic central cavity of the Shaker channel inner pore forms the receptor site for both the inactivation gate and quaternary ammonium compounds (Zhou et al., 2001), there may be significant differences between this work involving Shaker and the Kv1.5/Kv1.3 studies reported here. For example, pore mutations that modify internal tetraethylammonium block 10-fold in Shaker have minimal effects on quinidine binding to Kv1.5 (Yeola et al., 1996). In addition, the mechanism of action of the Kv1.3 N terminus may not be the same as that used by the Shaker inactivation ball (Uebele et al., 1998).

Conclusions. The present study demonstrates that the assembly of Kv1.5 and Kv1.3 subunits decreases the block induced by bupivacaine and quinidine on hKv1.5 channels (~4- and ~8-fold, respectively). Therefore, the sensitivity to hKv1.5 channel-blocking drugs will vary depending on the regional distribution of  regulatory subunits. The expression of Kv1.3 subunits in the myocardium is not homogeneous, for this subunit is expressed to a higher degree in the ventricle than in atria (Wang et al., 1996). Within various vascular beds, there are marked differences in  subunit expression, whereas Kv1.5 levels change little (Coppock and Tamkun, 2001). Thus, the differential assembly between the Kv and Kv subunits present in the cardiovascular system is another variable to be accounted for in the development of new ion channel modifying agents.