Introduction

Top Summary Introduction Materials & Methods Results Discussion References

TEA and its alkyl derivatives (QA) compounds are potent blockers of voltage-gated Na⁺ and K⁺ channels that represent useful probes of ion channel function and structure. Most voltage-gated K⁺ channels have an external and an internal receptor site for QA (Hille, 1991; Yellen et al., 1991), which can be distinguished by their affinity for TEA. In squid axon and in Shaker K⁺ channels, an increase of the length of one or more alkyl chains of QA favors the interaction with the TEA internal receptor, suggesting that this binding site contains a hydrophobic region (Armstrong, 1971; Armstrong and Hille, 1972; French and Shoukimas, 1981; Swenson, 1981; Villarroel et al., 1988; Choi et al., 1993). Local anesthetics are well known ion channel blockers, and most of them are tertiary amines, which are predominantly present in their charged (cationic) form at the physiological pH (Courtney and Strichartz, 1987; Strichartz and Ritchie, 1987; Hille, 1991). Therefore, these local anesthetics can be considered as highly hydrophobic QA analogs.

Bupivacaine, ropivacaine, and mepivacaine are local anesthetics that exhibit a common chemical structure; they differ only in the length of the alkyl substituent of the tertiary nitrogen (position 1), which is a butyl, propyl, or methyl group, respectively (Fig. 1A). These three chemically related drugs possess an asymmetrical carbon and thus exist as separate (S)-()- and (R)-(+)-enantiomers (Fig. 1A). We previously reported that bupivacaine block of cardiac Na⁺ and hKv1.5 channels is stereoselective, with (R)-(+)-bupivacaine being 1.6- and 7-fold more potent than the (S)-()-enantiomer, respectively, which could explain the higher cardiotoxicity of (R)-(+)-bupivacaine versus (S)-()-bupivacaine (Åberg, 1972; Aps and Reynolds, 1978; Valenzuela et al., 1995a, 1995b). Ropivacaine is a pure (S)-()-enantiomer, developed as a less cardiotoxic alternative to the racemic bupivacaine used in the clinical practice (McClure, 1996), which blocks Na⁺ and hKv1.5 channels with a lower potency than (S)-()-bupivacaine (Akerman et al., 1988; Valenzuela et al., 1995a, 1997; Moller and Covino, 1997).

View larger version (24K): [in this window] [in a new window]   Fig. 1.   A, Chemical structures of bupivacaine, ropivacaine, and mepivacaine. *, Asymmetrical carbon of the molecule. B, Current traces obtained in the absence and presence of (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, and (S)-()-mepivacaine. Cells were held at 80 mV, and 250-msec depolarizing steps between 40 and +60 mV in increments of 20 mV were applied. Tail currents were recorded at 40 mV. Data were filtered at 2 kHz (four-pole Bessel) and digitized at 10 kHz; additional digital filtering occurred at 1 kHz.

The stereoselectivity of bupivacaine block of hKv1.5 channels seems to require both polar and hydrophobic interactions with presumably pore-lining residues in the S6-helix (Franqueza et al., 1997). Bupivacaine, ropivacaine, and mepivacaine contain two hydrophobic groups: the aromatic ring and the alkyl-substituted saturated ring (Fig. 1A). Several studies have demonstrated that similar to the binding of QA to Shaker K⁺ channels (Choi et al., 1993), the length of this side chain is a structural determinant of the potency of these local anesthetics to inhibit cardiac I_Na and I_TO (Courtney, 1980b, 1980a; Castle, 1990): the longer the side chain, the higher potency of the drug. However, there is no experimental evidence regarding to the possible role of this substituent in the degree of stereoselective K⁺ channels block. Thus, to determine whether, specifically, the length of the side chain represents one of the structural requirements implicated in the potency and in the stereoselective block of hKv1.5 channels, we studied the effects of (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, and (S)-()-mepivacaine on hKv1.5 channels stably expressed in Ltk cells. The comparison of the current results with those previously reported with bupivacaine enantiomers (Valenzuela et al., 1995a) and (S)-()-ropivacaine (Valenzuela et al., 1997) may provide valuable clues about the nature of the local anesthetic/channel interactions. Preliminary results of the current study have been published in abstract form (Longobardo et al., 1997, 1998).

	Materials and Methods

Top Summary Introduction Materials & Methods Results Discussion References

Cell culture and solutions. We used a cell line stably expressing hKv1.5 as described previously (Snyders et al., 1992, 1993). Transfected cells were cultured in Dulbecco's modified Eagle medium supplemented with 10% horse serum and 0.25 mg/ml G418, under a 5% CO₂ atmosphere. The cultures were passed every 3-5 days, using a brief trypsin treatment. Before experimental use, subconfluent cultures were incubated with 2 µM dexamethasone for 24 hr to induce the expression of hKv1.5 channels (which is driven by a dexamethasone inducible promoter). The cells were removed from the dish with a rubber policeman, a procedure that left the vast majority of the cells intact. The cell suspension was stored at room temperature (21-23°) and used within 12 hr for all the experiments reported here.

Electrical recording. Experiments were performed in a small volume (0.5-ml) bath mounted on the stage of an inverted microscope (model TMS; Nikon, Garden City, NY) continuously perfused at a flow rate of 0.5-1.0 ml/min. The hKv1.5 currents were recorded at room temperature (21-23°) using the whole-cell voltage-clamp configuration of the patch-clamp technique with an Axopatch-1C patch-clamp amplifier (Axon Instruments, Foster City, CA). Currents were filtered at 2 kHz (four-pole Bessel filter) and sampled at 4 kHz. Data acquisition and command potentials were controlled by the pClamp 5.5.1. software (Axon Instruments).

Pulse protocol and analysis. After control data were obtained, bath perfusion was switched to drug-containing solution. Drug infusion or removal was monitored with test pulses from 80 mV to +30 mV, applied every 30 sec until steady state was obtained (within 10-15 min). The holding potential was maintained at 80 mV. The cycle time within each protocol was 10 sec to avoid accumulation of block or incomplete deactivation of the current.

(1)

(2)

(3)

(4)

(5)

Statistical methods. Results are expressed as mean ± standard error. Paired Student's t test was used to compare the effects of (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, and (S)-()-mepivacaine with the control values. Statistical significance was taken as p < 0.05.

	Results

Top Summary Introduction Materials & Methods Results Discussion References

Effects of (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, and (S)-()-mepivacaine on hKv1.5 channels. Fig. 1B (top) shows the original hKv1.5 current records obtained under control conditions. The holding potential was maintained at 80 mV, and pulses of 250 msec in duration to membrane potentials between 60 and +60 mV (in steps of 20 mV) were applied. Tail currents were recorded on return to 40 mV. The current activates with a fast time course, reaches a maximum peak, and slowly inactivates during the application of depolarizing pulses as described previously (Snyders et al., 1993). On repolarization to 40 mV, current deactivates with a dominant time constant of 31 ± 3 msec (17 experiments).

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Concentration dependence of (R)-(+)-ropivacaine (Ropi), (S)-()-ropivacaine, (R)-(+)-mepivacaine (Mepi), and (S)-()-mepivacaine block of hKv1.5. Reduction of current (relative to control) at the end of depolarizing steps from 80 mV to +60 mV was used as index of block. Data are mean ± standard error of a total of 60 experiments. Data for (S)-()-ropivacaine are from Valenzuela et al. (1997).

Voltage-dependent block of hKv1.5 by (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, and (S)-()-mepivacaine. Block of hKv1.5 channels by either (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, or (S)-()-mepivacaine was voltage dependent (Fig. 3). The Fig. 3 (top) shows that the three enantiomers induced a downward curvature of the IV relationship, indicating a higher current inhibition at more positive potentials. To quantify this voltage dependence, we plotted the relative current in the presence of 50 µM (R)-(+)-ropivacaine, 500 µM (R)-(+)-mepivacaine, or 500 µM (S)-()-mepivacaine versus membrane potential (Fig. 3, bottom). As it can be observed, block induced by all three enantiomers steeply increased in the range of activation of the channel (between 30 and 0 mV), which strongly suggest that the channels must open before drug can bind and block permeation. For depolarizations positive to 0 mV, block continued increasing with a shallower voltage dependence. Because in this range of membrane potentials, activation curve is saturated, this increase in block cannot be attributed to channel opening. Both drugs are weak bases [pK_a = 8.1 and 7.8 for (R)-(+)-ropivacaine and mepivacaine enantiomers, respectively], and therefore, we attributed the increase in block in this voltage range to the effect of the transmembrane electrical field on the drug/channel interaction. This effect can be explained by a simple Woodhull model (Woodhull, 1973). Therefore, a nonlinear curve fitting of the data to eq. 5 (see Materials and Methods) yielded apparent dissociation constant at the reference potential (0 mV) (K_D*) and fractional electrical distance () values of 41 ± 2 µM and 0.156 ± 0.003 (14 experiments) for (R)-(+)-ropivacaine, 549 ± 121 µM and 0.160 ± 0.008 (10 experiments) for (R)-(+)-mepivacaine, and 538 ± 97 µM and 0.154 ± 0.016 (six experiments) for (S)-()-mepivacaine. These K_D* values were higher than those determined at +60 mV (p < 0.01), as would be expected for an open channel blocker.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Voltage-dependent block of hKv1.5 channels by (R)-(+)-ropivacaine (Ropi), (R)-(+)-mepivacaine (Mepi), and (S)-()-mepivacaine. Top, current-voltage relationship under control conditions () and in the presence of each drug (). Bottom, relative current in the presence of the drugs at each membrane potential. Dotted lines, activation curves for each particular experiment. Block steeply increases in the range of membrane potential of activation of the channel (dotted line). For potentials positive to 0 mV, block continued to increase but with a shallower voltage dependence. Solid line, fit using a Woodhull model, which yielded an equivalent electrical distance of 0.16 for each of the drugs.

Kinetics of block of hKv1.5 channels induced by (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, and (S)-()-mepivacaine. (R)-(+)-Ropivacaine block of hKv1.5 channels was time dependent at concentrations of 100 µM (i.e., the drug induced a concentration-dependent fast decline of the current during the application of a depolarizing pulse that superimposed to the slow inactivation of the current) (Fig. 4). This fast decline was more evident and faster in the presence of higher concentrations of drug, and thus, the time constant of this process was taken as an approximation of the time constant of block (_B), from which the association (k) and the dissociation rate constants (l) were derived. After this approach (see eq. 3), the k and l values were (1.37 ± 0.14) × 10⁶ M¹ sec¹ and 44.0 ± 4.4 sec¹, respectively (11 experiments). Time-dependent block induced by 100 µM (R)-(+)-ropivacaine also was evident in the tail currents, which were slower in the presence of (R)-(+)-ropivacaine increasing the time constant of deactivation from 44.1 ± 5.9 to 221.5 ± 68.0 msec (six experiments; p < 0.05) (Fig. 4). In the presence of (R)-(+)-ropivacaine, the tail currents displayed a rising initial phase that reflects the dissociation process of (R)-(+)-ropivacaine from drug-bound (not conducting) open hKv1.5 channels (OD  O). Subsequently, the tail current displayed a slower decline because some fraction of the open channels become blocked again rather than closing irreversibly (OD  OC ). Therefore, a "crossover" phenomenon between the tail current obtained in the presence of drug and that recorded under control conditions was observed, which is indicative of open channel block (Armstrong, 1971; Snyders et al., 1992; Choi et al., 1993; Valenzuela et al., 1995a). Moreover, the experimental results obtained with (R)-(+)-ropivacaine were fairly well simulated using a kinetic scheme that assumes it only binds to the open state of hKv1.5 channels (Valenzuela et al., 1995a, 1997; Delpón et al., 1996) (Fig. 5A). This represents further evidence supporting that (R)-(+)-ropivacaine blocks hKv1.5 channels by binding to the open state of the channels.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Time dependence of block of hKv1.5 induced by (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, and (S)-()-mepivacaine. Top, time-dependent block induced by 100 µM (R)-(+)-ropivacaine. Left, current records obtained in the absence and presence of 100 µM (R)-(+)-ropivacaine during the application of a depolarizing pulse from 80 to +60 mV. (R)-(+)-Ropivacaine induced a fast initial decline of the current, which superimposes to the slow inactivation of the current. Right, tail currents obtained under control conditions and in the presence of 100 µM (R)-(+)-ropivacaine. Bottom, tail current "crossover" of tails obtained in control conditions and in the presence of either (R)-(+)-mepivacaine or (S)-()-mepivacaine. Arrow, the "crossover" between the two traces.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Mathematical modeling of the effects of (R)-(+)-mepivacaine on hKv1.5 channels assuming an open-channel block mechanism. Note that only a model in which the dissociation rate constant is increased can reproduce the experimental data.

1

1

1

1

1

1

1

1

1

	Discussion

Top Summary Introduction Materials & Methods Results Discussion References

We analyzed and compared the effects of (R)-(+)-ropivacaine, (R)-(+)-mepivacaine, and (S)-()-mepivacaine on hKv1.5 channels cloned from the human ventricle. The main findings of this study are that (1) (R)-(+)-ropivacaine and both mepivacaine enantiomers block the open state of hKv1.5 channels, (2) (R)-(+)-ropivacaine is ~10-fold more potent than (R)-(+)-mepivacaine and (S)-()-mepivacaine, and (3) mepivacaine block of hKv1.5 channels is not stereoselective.

(R)-(+)-Ropivacaine and (R)-(+)- and (S)-()-mepivacaine block open hKv1.5 channels. Block of hKv1.5 channels induced by (R)-(+)-ropivacaine was concentration, time, and voltage dependent. At concentrations of 100 µM, the most prominent effect of this enantiomer was the induction of a fast decline of the current at the beginning of depolarizing pulses positive to +50 mV, which was faster at higher drug concentrations. This effect suggests an open-channel block mechanism. Moreover, block was voltage dependent, being higher at more positive step potentials, which can be explained by the effect of the transmembrane electrical field on the interaction between the cationic drug and the receptor in the channel. This voltage dependence was consistent with a  value of 0.156 ± 0.003. Finally, tail currents exhibited a "crossover" phenomenon, which is typical of open-channel block mechanism (Armstrong, 1971; Snyders et al., 1992; Choi et al., 1993; Valenzuela et al., 1995a). Block induced by either (R)-(+)-mepivacaine or (S)-()-mepivacaine did not exhibit an identifiable time dependency, but block induced by both enantiomers was voltage dependent, which is consistent with a fractional electrical distance similar than that observed with (R)-(+)-ropivacaine. Furthermore, the tail currents also exhibited a "crossover" phenomenon. The lack of time-dependent block can be attributed to the very fast dissociation rate constant, which also explains the low affinity of these enantiomers to block hKv1.5 channels. Indeed, a kinetic scheme assuming open-channel block and a very fast dissociation rate constant reproduces fairly well the experimental results (Fig. 5). This finding represents an additional evidence supporting the open-channel block induced by mepivacaine enantiomers. In fact, a similar approach was used to explain differences in potency and time dependence between quinidine and quinine on hKv1.5 channels (Snyders and Yeola, 1995).

Relationship between the affinity and the alkyl chain length of the N-substituent. Bupivacaine, ropivacaine, and mepivacaine differ only in the length of the N-substituent, which is a butyl, propyl, or methyl group, respectively. In an attempt of establish a relation among the potency of block of each enantiomer (K_D), the dissociation rate constants (l), and the number of methyl groups of the N-substituent of these drugs, we plotted l and K_D values versus the number of ---CH₂ groups (Fig. 6A). For both (S)-() and (R)-(+) enantiomers, the K_D values decreased as the number of ---CH₂ groups increased, which suggests that the length of the side chain determines the potency of the local anesthetic. The decrease in the K_D values was parallel to the decrease of the l values, which seems to indicate that the lower potency to block hKv1.5 channels by local anesthetics with shorter side chains is derived from a more unstable drug/channel interaction.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   A, Relation between KD (left panel) and dissociation rate constants (l) (right) and number of ---CH2 groups in the molecule for the two enantiomers of bupivacaine, ropivacaine, and mepivacaine. Note that KD and l decrease when the number of ---CH2 groups increase in the molecule, indicating that the higher potency observed in the presence of bupivacaine enantiomers than in the presence of mepivacaine enantiomers is due to a more stable drug/channel interaction. B, Relation among the degree of stereoselective block (), the association rate constant (k), and the number of ---CH2 groups present in the (R)-(+)- and the (S)-()-enantiomers. The degree of stereoselective block increases when the number of ---CH2 groups is higher, as do the k values for these enantiomers, whereas the k values for the (S)-()-enantiomers did not change with the number of ---CH2 groups, suggesting that the degree of stereoselective block is determined by a more favored conformation, which is reflected by a faster k value. Data for bupivacaine enantiomers and (S)-()-ropivacaine were taken from previous reports from our laboratory (Valenzuela et al., 1995a, 1997).

Relationship between the N-alkyl chain length and stereoselectivity. In the current study, we show that block of hKv1.5 channels induced by (R)-(+)-ropivacaine exhibited a K_D value of 32 µM [i.e., it was 2.5-fold more potent than (S)-()-ropivacaine] (Valenzuela et al., 1997). It is interesting to note that the stereoselectivity previously described for bupivacaine enantiomers was much higher, with (R)-(+)-bupivacaine being 7-fold more potent than (S)-()-bupivacaine (Valenzuela et al., 1995a; Franqueza et al., 1997). Moreover, in the current study, we also found that mepivacaine block of hKv1.5 channels is not stereoselective. All these results demonstrate that the length of the N-substituent is a key structural determinant of stereoselective block of bupivacaine-related local anesthetics, which suggests that stereoselective block of hKv1.5 channels by this type of local anesthetics involves a hydrophobic interaction between the side chain of the local anesthetic at position 1 and the receptor in the channel. Because the dissociation rate constants are similar for each pair of enantiomers studied and most of their physicochemical properties are identical, stereoselectivity requires a more favored conformation of one of the enantiomers of each drug. This would lead to a faster association rate constant for the more potent enantiomers (R) versus the less potent ones (S), which would increase as the length of the side chain becomes longer and therefore more hydrophobic. To analyze this hypothesis, we plotted the relationship among the degree of stereoselectivity [ = K_D (R)-(+)/K_D (S)-()] (Courtney and Strichartz, 1987), the association rate constants, and the number of ---CH₂ groups at the N-position (Fig. 6B). As can be observed, the  values increased with the number of ---CH₂ groups. This increase in stereoselectivity (from 1.3 to 7) was accompanied by a parallel increase in the k values obtained for (R)-(+)-enantiomers, whereas the k values for the (S)-()-enantiomers remained constant for the three drugs studied, suggesting that (S)-() enantiomers need to adopt an energetically less favored conformation than (R)-(+) enantiomers to block hKv1.5 channels.

Hydrophobic interactions determine potency and degree of stereoselectivity. Several important molecular determinants for block of hKv1.5 channels and Shaker K⁺ channels by alkyl-TEA derivatives, as well as quinidine and bupivacaine, are located in the S6 segment of the  subunit (Yellen et al., 1991; Choi et al., 1993; Yeola et al., 1996; Franqueza et al., 1997). In a previous study, we reported that threonine at position 477 (i.e., hKv1.5 internal TEA binding site) also is involved in the binding of bupivacaine (Franqueza et al., 1997), thus suggesting that bupivacaine binds to a receptor site in the S6 segment, which overlaps the internal TEA binding site, as also was proposed by Baukrowitz and Yellen (1996) for alkyl-TEA derivatives and local anesthetics in Shaker K⁺ channels. The potency of block of hKv1.5 channels induced by alkyl-TEA derivatives in Shaker channels is related to the length of the side chain (Villarroel et al., 1988; Choi et al., 1993). Our results demonstrate that the length of the N-substituent at position 1 in bupivacaine-related local anesthetics determines the potency of these drugs to block hKv1.5 channels. The observed decrease in the dissociation rate constants with the number of ---CH₂ groups in the side chain could be explained if the stability of the drug/channel complex is mainly due to a hydrophobic interaction between the alkyl chain of the local anesthetic and some amino acid residue of the channel protein.

Conclusions. The length of the N-substituent of bupivacaine-related local anesthetics ropivacaine and mepivacaine determines the potency of block, being higher with longer chains. Moreover, stereoselectivity also is determined by the length of this side chain, in such a way that only bupivacaine-related anesthetics with alkyl substitutions of at least more than one ---CH₂ group block hKv1.5 channels in a stereoselective manner.