Introduction

Top Abstract Introduction Methods Results Discussion References

The clinical term of myotonia is used to identify a series of dominant and recessive forms of genetic diseases of skeletal muscle characterized by abnormal membrane excitability and delayed muscle relaxation after voluntary contraction. Sodium channel myotonias, paramyotonia congenita and hyperkalemic periodic paralysis are due to mutations of the gene coding for the skeletal muscle type of voltage-gated sodium channels (SCN4). The mutated channels show different degrees of impairment of inactivation, which in turn causes membrane depolarization (Cannon, 1996; Lehmann-Horn and Rüdel, 1996). Mutations producing mild depolarizations are responsible for the clinical phenotype of hyperexcitability, whereas those exerting sustained depolarizations lead to paralytic attacks (Cannon, 1996). In the dominant myotonia congenita and recessive generalized myotonia, the characteristic hyperexcitability and SpDs of action potentials are related to an abnormally low resting G_Cl (Adrian and Bryant, 1974; Bryant and Morales-Aguilera, 1971). The protein ClC-1 has been claimed to be the putative channel responsible for the macroscopic G_Cl; this hypothesis was supported by the finding of several mutations of the ClC-1 genes in patients and myotonic animals (Gronemeier et al., 1994; Lehmann-Horn and Rüdel, 1996; Steinmeyer et al., 1991). The attempts to restore the macroscopic G_Cl of genetically myotonic goats with various pharmaceuticals known to modulate this parameter in normal subjects have been unsuccessful (Bryant and Conte Camerino, 1991).

This result is in line with the recent findings that in the goat phenotype, the genetic mutation seriously impairs the function of the chloride channel in its physiological range (Beck et al., 1996). In fact, severe malfunctions up to a total loss of function of muscular chloride channel gene product (ClC-1) can result from different mutations (Fahlke et al., 1995; Gronemeier et al., 1994). Thus, the therapy for all the myotonic syndromes, whether related to sodium or chloride channels, is mainly symptomatic and addressed at relief of membrane hyperexcitability. Actually, the antimyotonic drugs that are used clinically are orally effective lidocaine derivatives such as MEX and tocainide (Jackson et al., 1994; Lehmann-Horn and Rüdel, 1996; Rüdel et al., 1980), which are able to block the generation and propagation of the action potential in skeletal muscle by blocking the voltage-gated sodium channels. Nevertheless, their use is restricted by their possible side effects, especially those produced at the hematopoietic and central nervous system levels (Roden and Woosley, 1986). In addition, effective antimyotonic doses are as large as those used to obtain antiarrhythmic action, with a possible effect on cardiac function as well.

In the attempt to search for selective antimyotonic agents, two characteristics should be taken into account. First, the drugs should block the I_Nas in a use-dependent manner. The use-dependent blockers stabilize the bound channels in the inactivated state from which the recovery is slowed down (Catterall, 1987; De Luca et al., 1991; Grant and Wendt, 1992). This mechanism ensures a stronger potency of the compound on tissues with excessive firing of action potentials and/or permanent depolarization, such as the muscles affected by sodium or chloride channel phenotypes of myotonia, than on tissues with a physiological excitability. Second, sodium channels of the various tissues are genetically distinct and show different kinetic and pharmacological properties. It was recently reported that in addition to having different sensitivities toward tetrodotoxin, cardiac and skeletal muscle types of sodium channels are affected differently by local anesthetic-like drugs; the pharmacological profile is further exacerbated by the different inactivation properties of the two channel types (Wang et al., 1996). These observations support the possibility of designing use-dependent blockers of I_Nas to be more selective on skeletal muscle than on the heart and to be able to relieve sarcolemmal hyperexcitability in myotonic subjects with fewer side effects.

The aim of the present study was to screen the effects of newly synthetized analogs of MEX as potential antimyotonic agents by evaluating in vitro their ability to suppress the pathological hyperexcitability of skeletal muscle isolated from myotonic ADR mouse, a phenotype with a severe recessive form of low G_Cl myotonia (Gronemeier et al., 1994; Mehrke et al., 1988; Steinmeyer et al., 1991). The newly synthetized compounds were used to investigate the influence of some structural properties of the molecule on antimyotonic activity and, in particular, (1) the role of the distance between the aromatic and amino-terminal groups (compound III, fig. 1) and (2) the increase in lipophilicity produced by the insertion of a chlorine atom in the para position of the aromatic ring and of a phenyl group on the carbon atom linked to the amino group (compound II, fig. 1). These two compounds were chosen on the basis of the effects they produced on skeletal muscle I_Nas during preliminary voltage-clamp experiments and further verified in the present study. In fact, compound III showed noticeable use-dependent behavior, whereas compound II was a very potent I_Na blocker, two characteristics that may be of importance in obtaining selective antimyotonic agents.

View larger version (K): [in this window] [in a new window]   Fig. 1.   Chemical structure of MEX and its newly synthetized analogs.

	Methods

Top Abstract Introduction Methods Results Discussion References

Current-clamp recordings of macroscopic membrane excitability of ADR myotonic and healthy mice. Myotonic ADR (genotype adr/adr) mice and wild-type healthy littermates (genotype +/+ or adr/+) that were 3 months old were used for all the experiments. Myotonia in ADR mice is a recessive disease due to the insertion of a transposon into the ClC-1 gene (Gronemeier et al., 1994; Steinmeyer et al., 1991). The diseased homozygous animals develop a severe and clearly distinguishable myotonic state, whereas the genotypic adr/+ animals do not show recognizable signs of myotonia and do not have an alteration of the macroscopic G_Cl with respect to wild-type (+/+) despite the defect in one allele (Gronemeier et al., 1994; Mehrke et al., 1988).

Recordings of I_Na using the three vaseline gap voltage-clamp technique. Voltage-clamp recordings of I_Na were performed on 1-2-cm-long segment of single muscle fibers obtained by microsurgery from the ventral branch of semitendinosus muscle of rana esculenta. The cut end fiber was superfused with an internal solution and transferred into the recording chamber, filled with the same solution. The muscle chamber used was based on that of Hille and Campbell (1976). The fiber was mounted across three chamber partitions with edges that had been previously covered with strings of commercial grease (Glisseal, Borer Chemie AG, Zuchwil, Switzerland), which delineated four pools. With the fiber in position, three additional strips of grease were applied over the fiber and carefully sealed to the fiber to reduce leakage. The length of the cut ends of the fiber was ~300 µm. Four KCl/agar bridges electrodes (one for each pool) connected the recording chamber to the voltage-clamp amplifier. When the solution was lowered below the glisseal grease strips, the four pools were physically and electrically independent of each other, as established by the absence of leak current on increase of the gain amplifier. Then, the solution in one of the central pools, called the A pool (70-100 µm wide), was replaced by an external solution, and recordings were performed at 10°C. The circuit of the voltage-clamp amplifier was based on that described by Hille and Campbell (1976). Briefly, a first operational amplifier was used to compare the interior of the fiber with the ground and applied a negative feedback in pool A to force the membrane potential to 0 mV. A second operational amplifier clamped the membrane potential to the command potential via a negative feedback. The measurement of the membrane current in response to a voltage-clamp step was made by measuring the voltage drop across a resistor of a known value. The holding potential was 100 mV. The voltage-clamp amplifier was connected via a 12-bit AD/DA interface (Digidata 1200, Axon instruments, Forster City, CA) to a 80486 DX2/66 PC. The stimulation protocols and data acquisition were driven by Clampex (pClamp 6 software package, Axon Instruments). The I_Nas flowing in response to depolarizing command voltages were low-pass filtered at 10 kHz (Frequency Devices), visualized on an oscilloscope, sampled at 20 kHz and stored on the hard disk. When necessary, leak and capacities currents were subtracted using the P/4 method. The acquired traces were later analyzed with Clampfit (pClamp 6 software package, Axon Instruments). Maximal I_Namaxs were elicited with test pulses from the holding potential to 20 mV for 10 msec. The tonic block exerted by the test compounds was evaluated as percent reduction of the peak I_Na elicited by single test pulses. The evaluation of the use-dependent block by the drugs was made by using a 10-Hz train of test pulses for a period of 30 sec and normalizing the residual current at the end of this stimulation protocol with respect to that in the absence of the drug (De Luca et al., 1995).

Solutions and drugs. The mouse intercostal muscle preparations were perfused with a physiological salt solution with or without the test compounds: 148 mM NaCl, 4.5 mM KCl, 2.0 mM CaCl₂, 1.0 mM MgCl₂, 12 mM NaHCO₃, 0.44 mM NaH₂PO₄ and 5.55 mM glucose. The solutions were continuously gassed with 95% O₂/5% CO₂ (pH 7.2, 7.4). For I_Na recordings, the semitendinosus muscle fibers were perfused with an external solution consisting of 77 mM NaCl, 38 mM coline-Cl, 1.8 mM CaCl₂, 2.15 mM Na₂HPO₄ and 0.85 mM NaH₂PO₄ and dialyzed with an internal solution consisting of 105 mM CsF, 5 mM 3-(N-morpholino)propanesulfonic acid, 2 mM MgSO₄, 5 mM ethylene glycol bis(-aminoethyl ether)-N,N,N,N-tetraacetic acid and 0.55 mM Na₂ATP (pH 7.2 with NaOH concentrated solution). The compounds tested were synthetized in our laboratories. Briefly, MEX (compound I) was synthetized according to procedures previously detailed (Franchini et al., 1994); compound II [2-(4-chloro-2-methylphenoxy)-benzenethanamine] was obtained through condensation of bromoketon with the opportune phenolic derivative and transformation of the product to oxime through reaction with hydroxylamine and successive reduction to amine II² and compound III [()-S-3-(2,6-dimethylphenoxy)-2-methylpropanamine] was obtained in the optically active form through condensation of the chiral alcohol with the opportune phenolic derivative after protection of the nitrogen atom (Duranti et al., 1995) (fig. 1). All compounds were fully characterized by routine spectroscopic analyses; analytical results for C, H and N were within ±0.4% of the theoretical values. Stock solutions of MEX and compound III were prepared in physiological or external solutions for current-clamp and voltage-clamp experiments, respectively, whereas stock solutions in dimethylsulfoxide (100 µl/mg) were used for compound II. All the stock solutions were prepared daily, and the final concentrations to be tested in vitro on isolated muscle were obtained by further dilution of the stock solution as needed. Dimethylsulfoxide at the highest concentration used (0.2%) was without effect on any of the parameters recorded. On both intercostal muscle and frog muscle fibers, no more than three concentrations of the same compound were tested, and the preparations were exposed to each concentration for 10 min before recording to allow the maximum effect of the drug to be reached.

Statistical analysis. Data are expressed as mean ± S.E.M. The statistical significances of the differences between groups of mean values were calculated by unpaired Student's t test. The molar concentrations of the drugs producing a 50% block of firing or of I_Namax (IC₅₀) were determined by using a nonlinear least-squares fit of the concentration-response curves to the following logistic equation: Effect = 100/1 + {K/[drug]}ⁿ, where Effect is percent change of the parameter, 100 is the maximal effect, K is the IC₅₀ value of the tested drug, n is the logistic slope factor and [drug] is the molar concentration of the tested drug (De Luca et al., 1992, 1995). The estimates of S.E.M. and n for normalized percent values were obtained as described by Green and Margerison (1978).

	Results

Top Abstract Introduction Methods Results Discussion References

Excitability characteristics of muscle fibers of myotonic ADR mice. The excitability characteristics of intercostal muscle of myotonic ADR mice are detailed in table 1, and representative recordings are shown in figure 2. Spontaneous myotonic activity, such as high-frequency discharges of action potentials (>10 Hz) on insertion of the recording electrode, was detected in ~60% to 80% of fibers in each ADR muscle preparation (table 1 and fig. 2). The myotonic muscle fibers needed very little depolarizing current to generate one action potential and, as expected on the basis of the low G_Cl myotonic phenotype, the action potential showed a significantly longer Lat than muscle fibers from healthy mice. However, no significant differences with respect to normal animals were observed in AP or values of Th. As is typical of myotonic state, the muscle fibers of ADR mice were able to generate a significantly higher number of action potentials than healthy ones when the depolarizing stimulus was just slightly increased over the I_th (table 1). In myotonic muscle fibers, the stimulus-evoked train of action potentials was followed by self-sustained ADs at the end of the depolarizing current pulse in ~75% to 100% of the fibers sampled (table 1 and fig. 2).

View larger version (K): [in this window] [in a new window]   Fig. 2.   Typical electrophysiological recordings of excitability characteristics of mouse intercostal muscle fibers using the intracellular microelectrode current-clamp technique as described in the text. Recordings from (A) healthy and (B) myotonic ADR mice and recordings showing effects of (C) 100 µM MEX, (D) 50 µM compound III and (E) 10 µM compound II on the excitability of myotonic ADR mice are shown. Each row shows a recording from the same fiber in which the intensity of the depolarizing pulse was slightly increased (from left to right) to reach the Ith for generating a single and a train of action potentials, respectively. B, The myotonic fibers showed self-sustained ADs of action potentials at the end of the current pulse that elicited a membrane firing and SpDs on insertion of the voltage-sensitive microelectrode.

Effects of MEX and its analogs on excitability characteristics of muscle fibers of myotonic ADR mouse. All of the tested compounds affected the excitability characteristics of myotonic muscle fibers, and we took care to evaluate the ability of each compound to relieve at low concentrations the typical myotonic manifestations without changing the parameters unaffected by the disease (table 2). The ability of each compound to reduce in a concentration-dependent manner the excessive stimulus-induced firing of action potentials of ADR muscle sarcolemma was also evaluated (fig. 3). MEX at 10 µM started to have a membrane-stabilizing effect that was noted during recordings on the basis of a more stable membrane potential during impalement of the fibers. In fact, a significant reduction of the SpDs was observed at 10 µM; at 50 µM, these were completely suppressed along with a significant reduction in the occurrence of the ADs, although the stimulus-evoked firing of the membrane was not modified. This latter was appreciably reduced by 100 µM MEX (fig. 3), a concentration at which a reduction in AP and a significant 5-mV increase in Th (P < .05) were also observed (table 2 and fig. 2). The above effects of MEX were more remarkable at 300 µM, which further produced a significant shortening of the Lat to 33.5 ± 7.1 msec (n = 9) (P < .001) and a 40-nA increase in I_th (table 2). Compound III was able to decrease dose-dependently both SpDs and ADs in the range of 5 to 10 µM, being more potent than MEX on the ADs. At the above concentrations, the firing capability of the membrane was also slightly reduced (fig. 3). Compound III at 50 µM completely suppressed spontaneous myotonic manifestations and significantly reduced the maximum number of spikes that the membrane could generate from 13 ± 0.8 (n = 6) to 8.3 ± 0.8 (n = 9) (P < .005). In fact, compound III inhibited the firing capability of the membrane more effectively than MEX, with an IC₅₀ value of 66 vs. 214 µM (fig. 3). Compound III was also more potent than MEX in reducing the AP and in increasing both I_th and Th (fig. 2 and table 2). The effects of this compound increased at 100 µM, and at 180 µM we observed a complete block of the excitability in that none of the sampled fibers were able to generate one action potential (data not shown). Compound II was very potent in depressing myotonic hyperexcitability but showed a different pattern of effects compared with MEX and compound III. In fact, at concentrations as low as 10 µM, this compound not only depressed the occurrence of SpD and ADs but also produced a reduction of the firing capability comparable to that observed with 100 µM MEX, with the half-maximal concentration on this parameter being 24 µM (fig. 3 and table 2). Furthermore, at 10 µM it decreased the latency of the action potential to 66 ± 16 msec (n = 6), increased significantly I_th from 19 ± 1.5 (n = 7) to 48.7 ± 4.5 (n = 6) nA (P < .001) and slightly shifted Th toward more positive potentials (fig. 2). All these effects were concentration dependent, being more pronounced at 50 µM. At this concentration, compound II significantly shortened the Lat to 33.7 ± 7.4 msec (n = 7) (P < .001) and almost completely blocked the ability of the fiber to generate more than two action potentials. A decrease in the amplitude of the action potential was also observed (table 2). At 100 µM, compound II completely blocked the ability of the fibers to generate even one action potential. In contrast to MEX and compound III, the effects of compound II were not reversible on washout. None of the compounds tested significantly modified the resting membrane potential.

View this table: [in this window] [in a new window]   TABLE 2 Effects of MEX and its analogs on excitability characteristics of myotonic and healthy mice Shown are the effects of MEX, compound III and compound II on the excitability parameters of intercostal muscle fibers of myotonic (ADR) and healthy (Control) mice. The drug effects are illustrated as the percent modification of each parameter (+, percent increase; , percent decrease, N.C., no change) with respect to the relative value in the absence of drug ± S.E.M. from 6 to 28 normalized number of fibers.

View larger version (K): [in this window] [in a new window]   Fig. 3.   Concentration-response curves showing the percent inhibition of the maximum number of spikes elicitable (firing) produced by MEX, compound III and compound II on the intercostal muscle of myotonic ADR mice. Each value is the mean ± S.E.M. from 8 to 12 normalized fibers. The curves fitting the experimental points were obtained using the logistic function detailed in the text. The IC50 value for each compound calculated from the fit was 214 ± 26, 66 ± 6 and 24 ± 4 µM for MEX, compound III and compound II, respectively.

Effects of MEX and its analogs on excitability parameters of muscle fibers of healthy mice. The effects of the test compound on the excitability of intercostal muscles of healthy mice were studied to better evaluate the specificity of the antimyotonic activity. As shown in table 2, MEX modified the depolarization-induced excitability parameters of healthy muscle fibers at concentrations similar to those effective on the same parameters in myotonic ones. Different effects were observed with the two analogs. Compound III at 100 µM was much less effective on the excitability parameters of healthy muscle fibers than on those of ADR muscle (table 2), with the exception of the membrane firing capability, which was similarly reduced in both cases (by 70 ± 3% in healthy and by 68.5 ± 5% in ADR muscle fibers). Compound II at 50 µM reduced the amplitude of the action potential of healthy muscle fibers in a manner similar to the effect produced in myotonic fibers. However, the decrease in membrane firing capability was much less evident on normal muscle fibers. At 50 and 100 µM, compound II reduced the firing capability by 51 ± 6% and 68.3 ± 6%, respectively, effects that were much less pronounced than those observed with 50 µM in myotonic fibers. At 50 and 100 µM, a concentration-dependent decrease in I_th and a small increase in Th were observed. Under this experimental condition, the Lat was little modified by compound III; this parameter was only slightly reduced at 100 µM (table 2). In addition, on healthy muscle fibers, the test compounds did not produce any remarkable effect on resting membrane potential.

Effects of MEX and its analogs on I_Nas of single muscle fibers. The effects of in vitro application of MEX and its analogs on I_Nas of frog semitendinosus muscle fibers are illustrated in figure 4. As detailed in the text, I_Namaxs were elicited with 10-msec test pulses from the holding potential of 100 to 20 mV. Such a test pulse was applied as a single stimulus in both the absence and presence of the test drugs to evaluate the amount of tonic block (i.e., block exerted by the drug during the resting state of the channel and the membrane). After evaluation of the tonic block, the test pulse was repetitively applied at the frequency of 10 Hz for 30 sec to evaluate the ability of the test compounds to produce a use-dependent block (i.e., a block of the I_Na in a situation of high-frequency stimulation such as that occurring in the pathological myotonic state). MEX produced a tonic block of the I_Na in a concentration-dependent fashion with a calculated IC₅₀ value of 83 µM. As expected, in the presence of MEX, the repetitive stimulation at 10-Hz frequency produced a further cumulative reduction of the I_Na due to use-dependent block. After 30 sec of such a stimulation, the use-dependent block by MEX fully attained the equilibrium, and the residual current normalized with respect to that in the absence of drug allowed the calculation of the amount of use-dependent blockade. The IC₅₀ value for use-dependent block was three times lower than that for tonic block. Compound III was less potent in producing a tonic block of the currents with respect to MEX; in figure 4A, the effects of 50 µM MEX and compound III on inward sodium transients are compared. The calculated IC₅₀ value of compound III for tonic block was 108 µM. However, the 10-Hz stimulation showed a remarkable use-dependent blockade by this compound so that 50 µM compound III or MEX produced a comparable block of I_Na. Thus, the IC₅₀ value of compound III, calculated with the high-frequency stimulation protocol, was almost six times lower than that for obtaining the tonic block. Different behavior was observed with compound II; this compound produced a remarkable and irreversible tonic block of the I_Nas, with IC₅₀ values as low as 30 µM. Nevertheless, little cumulative block was observed during the 10-Hz stimulation protocol, with the calculated IC₅₀ being twice as low as that for tonic block (fig. 4).

View larger version (K): [in this window] [in a new window]   Fig. 4.   A, INa transients recorded by the three vaseline gap voltage-clamp method from single muscle fibers of frog semitendinosus muscle and (from left to right) effects of MEX (50 µM), compound III (50 µM) and compound II (10 µM), respectively. After recording the control trace elicited by a single test pulse from 100 to 20 mV (a), the test compounds were applied. After 10 min of incubation, a single test pulse was again applied. The reduction in the current under this condition was the result of the tonic block exerted by the drug (b), and its amount was quantified by percent normalization with respect to the current in the absence of drug. A train of pulses at a frequency of 10 Hz was then applied for 30 sec, and a further cumulative block of the INas over the tonic block was due to the use-dependent behavior of the compound (c). The amount of use-dependent block was calculated from percent normalization of the residual current at the end of the 30-sec stimulation with respect to the relative current in the absence of drug. As it can be seen, 50 µM MEX produced a greater tonic block of the current with respect to the same concentration of compound III; however, this latter was more use dependent, so at the end of the stimulation protocol, the percent reduction by compound III was comparable to that produced by MEX. In contrast, compound II was much more potent than both MEX and compound III in blocking INa, although this compound showed less use-dependent behavior. This protocol was repeated with various concentrations of the test compounds to construct concentration-response relationships (B). Each point is the mean ± S.E.M. of percent block of INamax from three to five fibers. The curves fitting the experimental points led to the calculated values of IC50 for both tonic and use-dependent block that were 83 ± 14 and 30 ± 5 µM for MEX; 108 ± 13 and 17 ± 4 µM for compound III and 30 ± 3 and 15 ± 2 µM for compound II, respectively.

	Discussion

Top Abstract Introduction Methods Results Discussion References

The present study was aimed at evaluating the ability of MEX and its newly synthetized analogs to abolish the abnormal membrane excitability of myotonic muscle fibers of ADR mouse. In this phenotype, the pathological hyperexcitability is due to an abnormally low macroscopic G_Cl, the parameter that ensures the electrical stability of the membrane under resting conditions and maintenance of this latter after the voluntary excitation-contraction cycle. As a consequence of their low resting G_Cl, myotonic muscles spontaneously generate trains of action potentials, which trigger involuntary spasms; the self-sustained ADs of action potentials occurring in myotonic fibers are responsible for the delayed relaxation after voluntary movements and are caused by the low G_Cl-induced potassium accumulation in the t-tubules (Adrian and Bryant, 1974; Bryant and Morales-Aguilera, 1971; Lehmann-Horn and Rüdel, 1996).

Particular attention has therefore been devoted to evaluating the effects of MEX and its derivatives on these peculiar myotonic signs of the ADR phenotype and to correlate them with the ability to block the I_Nas in a use-dependent manner. All the three compounds tested were effective, although appreciable differences were observed between them. As far as MEX is concerned, we observed that it started to reduce the abnormal SpDs of myotonic fibers at concentrations as low as 10 µM, a concentration close to that described to be clinically effective for this compound as an antiarrhythmic (Sato et al., 1995; Wu et al., 1992) and in good agreement with those able to produce a use-dependent block of the I_Nas (Sunami et al., 1993). The increased distance between the aromatic ring and the amino-terminal group by the insertion of a methylene moiety in the alkyl chain, in compound III, doubled the potency in reducing the firing capability of myotonic muscle fibers and slightly decreased the ability to modify the excitability characteristics of healthy muscle fibers. The stronger potency of compound III than MEX as an antimyotonic drug could be explained taking into account the results on the blocking mechanism of this compound on the I_Nas. Compound III was less potent than MEX in producing a tonic block of the I_Nas (i.e., a block of the sodium channels during the resting state), but it caused a stronger use-dependent accumulation of I_Na block during high-frequency stimulation. The ability of this class of compounds to produce a tonic and use-dependent block depends in part on the molecular mechanism at the level of the receptor site on the channel protein and in part on the physicochemical properties of the molecule itself. It is known that the sodium channel blockers act on a receptor on the alpha subunit of sodium channel complex that has stringent conformational requirements (i.e., it discriminates between optically active compounds) (De Luca et al., 1995; Ragsdale et al., 1994). According to the receptor-modulated hypothesis, the use-dependent mechanism is due to a change in conformation of the receptor during the voltage-dependent channel transition, and this leads to a high-affinity state when the channel is open and/or inactivated (Catterall, 1987; Grant and Wendt, 1992). The drug-blocked channels recover more slowly from the nonconductive inactivated state (Sunami et al., 1993), and this can lead to accumulation of channel block if the membrane remains depolarized or the intervals between successive depolarizing stimuli are too short with respect to the dissociation constant of the drug from the channel (Grant and Wendt, 1992). At the same time, the physicochemical properties influence the possibility of the molecule gaining access to the binding site.

A compound whose pK_a favors an increase in the ratio between charged/uncharged forms usually shows stronger use-dependent behavior with respect to an uncharged or a highly lipophilic compound because the former needs a channel opening to gain access to the receptor site (Hille, 1977). The minimal chemical modification introduced with compound III can increase the pK_a of the molecule (with an increased amount of charged form at physiological pH) and change the affinity of the drug/receptor interaction, since in a series of lidocaine derivatives, the presence of an alkyl chain of two carbons has been found to optimize the binding to the receptor (Sheldon et al., 1991). In turn, the high use-dependent behavior can account for the ability of compound III to act on high-frequency discharges of action potentials, either spontaneously occurring or induced by the depolarizing stimuli, at lower concentrations with respect to those of MEX. On the other hand, an increase in the lipophilicity of the molecule via the insertion of a chlorine atom in the para position on the aromatic ring and of a phenyl group on the carbon atom linked to the amino group led to compound II, which proved to be almost 10 times more potent than MEX in reducing the excitability characteristics of myotonic muscle fibers although with less graduation of the effect on the various myotonic manifestations (e.g., the SpDs). Concomitantly, compound II produced a remarkable and irreversible tonic block of I_Nas during voltage-clamp experiments, and the additional use-dependent block was slight, likely due to its high lipophilicity.

The weak use-dependent behavior, irreversible action and lipophilicity made compound II appear to be of little interest from a therapeutical point of view. One would predict a toxicological potential due to poor discrimination between pathological and physiological excitability patterns and long-lasting accumulation in various tissues. Nevertheless, compound II shows interesting features; in particular, (1) it was effective at low concentrations, (2) in contrast to MEX, it was more effective on myotonic fibers than on healthy ones and (3) it was able to shorten the Lat in a concentration-dependent manner in the myotonic but not the healthy fibers. This latter effect, which was also observed with MEX at doses as high as 300 µM, may be beneficial for therapeutic interventions on the low-G_Cl myotonic state, such as the ADR mouse. In fact the long latency, and the consequent prolongation of action potential duration, is a characteristic feature of the decrease in G_Cl (Adrian and Bryant, 1974). The Lat reflects the electrotonic response of the fibers before reaching the threshold, and it is mainly due to the resting ionic conductances to chloride and potassium ions. In the absence of G_Cl, as in ADR fibers, a shortening of the latency can be mediated by increasing potassium conductance through a modulation of potassium channels active at resting potential. It has been recently shown that MEX, at high concentrations, can open K_ATP channels in the heart and thus shorten cardiac action potential duration (Sato et al., 1995). It is tempting to speculate that a similar mechanism can account for the decrease in latency observed with MEX and compound II in ADR fibers. K_ATP channels are abundantly present on sarcolemma of striated fibers and are predisposed to opening at resting potential when the level of intracellular ATP falls or the cytoplasmic pH decreases (Longman and Hamilton, 1992; Tricarico and Conte Camerino, 1994), metabolic situations likely to occur in the myotonic muscle fibers because of the intensive muscle work.

The hypothesis that compound II can act on K_ATP channels at concentrations close to those effective on sodium channels has yet to be validated through more direct experimental evidence. However, it is a particularly attractive hypothesis because the classic openers of K_ATP channels have been proposed as alternative antimyotonic agents (Longman and Hamilton, 1992). Furthermore, the above hypothesized mechanism would be specifically active in myotonic fibers, since the healthy fibers have a large G_Cl and lack the metabolic alterations that facilitate opening of the K_ATP channels. Other different mechanisms can also account for the effectiveness of compound II on the myotonic fibers at lower concentrations than on healthy ones. It is possible that in the healthy fiber, the compound acts with additive mechanisms able to counteract the block of I_Nas or, instead, that compound II has a very high affinity for inactivated sodium channels, which would predominate in the high discharging low-G_Cl fibers. For instance, riluzole, a novel psychotropic agent with anticonvulsant properties, has been found to have a high affinity for inactivated sodium channels despite its weak use dependency (Hebert et al., 1994). Further investigation into the mechanisms underlying the effects of compound II, such as verification of its ability to open K_ATP channels, and on its structure-activity relationship is of interest to improve the therapeutical potential of this mexiletine derivative. Finally, our results show that the newly synthetized analogs of MEX have different features from the parent compound and are of interest in the development of more potent blockers of the skeletal muscle sodium channels and potential antimyotonic agents.