The Pathophysiology of Myotonia Produced by Aromatic Carboxylic Acids R. E. Furman, M D , and R. L. Barchi, MD, PhD
A series of nine related aromatic monocaEboxylic acids (ACAs) previously shown to inhibit muscle membrane chloride conductance (Gcl) selectively in the rat were studied for their ability to produce myotonia. All nine
induced characteristic repetitive electrical activity and delayed relaxation in isolated muscle, although the concentrations required for this action varied widely. In each case, myotonia was observed at concentrations that correlated closely with previously determined half-maximal concentrations for inhibition of G a . Intracellular recordings from muscle made myotonic with ACA revealed prolonged latencies at rheobase, multiple driven spikes, and self-sustaining repetitive activity similar to that previously reported in hereditary goat myotonia. Phase-plane diagrams of membrane action potentials recorded after exposure to the most effective of these compounds suggested little effect on the voltage-dependent sodium system. The changes seen could be duplicated by simple removal of chloride ion. The expression of repetitive electrical activity in the presence of low membrane Gcl depends on ambient temperature and on the concentration of calcium ion. Increasing temperature and decreasing Ca++ predispose toward myotonic activity; converse conditions inhibit myotonia. Myotonia induced by ACA is inhibited by concentrations of diphenylhydantoin that are clinically effective in controlling hereditary myotonia in humans. Furman RE, Barchi R L The pathophysiology of myotonia produced by aromatic carboxylic acids. Ann Neurol 4:357-365, 1978
Myotonia, a form of pathologically delayed relaxation of striated muscle, is seen in several human and animal diseases. This delayed relaxation is the result of self-sustaining repetitive action potentials generated in the surface membrane in response to a brief stimulus [20]. Evidence suggests that the basic pathophysiological defect in myotonia may be an abnormality of chloride conductance (Ga) in the muscle fiber surface membrane [8, 21, 311. Studies on hereditary myotonia in goats [7] and on myotonia congenita in humans [2 11 have demonstrated in each case an abnormally low membrane Ga. The potential relationship between such a reduced Gcl and repetitive electrical activity has been shown in several studies using mathematical models of the muscle surface membrane 12, 3 , 51. Myotonia resembling that seen in natural disease can be produced in otherwise normal animals by a number of chemical agents, the best known of which is 2,4-dichlorophenoxyacetic acid (2,4-D) [4, 10, 121. In vitro muscle studies have documented a reduction in Gcl after exposure to 2,4-D [29] and several related carboxylic acids [9]. In a recent study of twenty-five aromatic carboxylic acids (ACAs) similar
to 2,4-D, we demonstrated that the sarcolemmal Gc1
From the Departments of Neurology and of Biochemistry and Biophysics, University of Pennsylvania School of Medicine, Philadelphia, PA.
Address reprint requests to Dr Barchi, Department of Neurology, Hospital of the University of Pennsylvania, Philadelphia, PA 19104.
was reversibly inhibited by nineteen of these acids [ 2 4 ] . Inhibition appeared to involve partitioning of the compounds into the surface membrane with subsequent specific interaction, presumably at the site of chloride translocation. Our earlier work provided us with a series of compounds having varying physical characteristics that produce inhibition of Gcl at known concentrations in vitro. In the present work we used these compounds to study the general mechanism by which repetitive activity is produced in mammalian muscle by ACAs. We demonstrated that the production of myotonia by these compounds is in each case directly related to the ability of that compound to reduce membrane Gcl and that the intracellularly recorded abnormal electrical activity closely resembles that previously documented in the inherited myotonia of the goat [ 11. Myotonia induced by ACAs exhibits many characteristics similar to those seen in hereditary myotonia congenita in man, suggesting a basic similarity between these states. A preliminary report of portions of this work has been presented [ 141.
Accepted for publication Apr 13, 1978.
0364-5134/78/0004-0410$01.25 @ 1978 by R. E. Furman
357
Methods Aromatic carboxylic acids were obtained from Aldrich Chemical Corporation, Milwaukee, WI; K and K Laboratories, Plainview, NY; and Eastman Kodak Company, Rochester, NY. Several were the generous gift of Dr R. B. Moffett of the Up john Pharmaceutical Company. Diphenylhydantoin was obtained from Elkins-Sinn, Inc, Cherry Hill, NJ. Male Wistar rats of 250 to 350 gm were used for all experiments. For twitch tension studies, extensor digitorum longus muscle removed from the tendon of origin to tendon of insertion was used. For intracellular recordings the hemidiaphragm was removed and prepared as previously described [23]. In each case the muscle was transferred to a temperature-controlled recording chamber and perfused continuously as previously described [2 31. Composition of the normal and Clk-free methylsulfate Ringer's solution was as given earlier [23, 241. Measurements of twitch tension were made with a Disa Model 5 7 tension transducer. Platinum surface electrodes spaced 1 mm apart were used for stimulation in these initial studies. Muscles were adjusted in the chamber to about 1.2 times their resting length, and a short series of constantcurrent pulses (6 pulses, 0.5 msec, 200 Hz), found to be optimal for inducing a myotonic response, was used for stimulation. Pulse amplitude was adjusted for maximal response. At least 30 seconds was allowed between each stimulus set to minimize the '' warm-up" phenomenon observed. Tension records were analyzed from photographic records. Tension responses after exposure of the muscle to an ACA were graded on the basis of ratio of peak tension generated during stimulation to plateau tension maintained more than 20 msec after termination of stimuli. Grade 1 indicates sustained tension visible but less than 10% of peak tension during stimulation; grade 2, sustained tension of 10 to 33% of peak tension; grade 3, 33 to 66%; and grade 4,66 to 100%. Average grading for agiven muscle at a particular concentration of aromatic acid was based on recordings from fifteen or more separate stimulations at multiple sites. Microelectrode techniques for intracellular recordings were as previously described [23, 241. Phase-plane recordings were obtained by electronic differentiation of the intracellularly recorded voltage response using an active network whose transfer characteristics were carefully controlled to minimize phase shift between input and output signal. The rate of change of membrane potential (dV,/dt) was displayed as a function of membrane potential (V,) on a dual-beam oscilloscope along with the usual display of V, versus time, and analyzed after enlargement from photographic records. Methods used for calibration and analysis of phase-plane diagrams were essentially those of Jenerick [17, 181. Results
Prodnction of Myotonic Activity In order to test for correlation between the inhibition of chloride conductance and induction of myotonic activity, nine ACAs previously shown to inhibit
358
Annals of Neurology Vol 4 No 4 October 1978
Table I . Benzoic Acid De7ivative.r Examined for Myotonic Effect in the Present Report"
Compound
K,(M)
1. 2.
1.1 x 8.0 x 3.0 x 3.1 x 6.0 x 9.5 x 1.5 x 2.0 x 3.0 x
Anthracene-9-carboxylicacid Pentachlorobentoic acid 3. 2,5-Dimethylbenzoic acid 4. 2,5-Dichlorobenzoic acid 5 . 2,3,5,6-Tetramethylbenzoicacid 6. 2,3,6-Trichlorobenzoic acid 7. 3,5-Dimethylbenzoic acid 8. 2,6-Dichlorobenzoic acid 9. 3,5-DimethyI-4-nitrobenzoic acid
lo-" 10-5 10-4 10-4
10-4 10-4
lo-"
aKi represents the concentration previously determined to reduce G,, by 50% (Palade and Barchi [241). Identitication numbers correspond to points in Figure 2.
membrane Gcl in rat diaphragm were examined for their ability to induce myotonia in the extensor digitorum longus muscle of the rat in vitro (Table 1). These compounds had inhibition constants (Ki, that concentration producing 50% inhibition) for Gcl ranging over three orders of magnitude. Muscles were equilibrated at a given ACA concentration for 15 minutes or longer and then stimulated supramaximally with surface electrodes at a number of sites away from the motor point while the resultant tension generated was recorded. Myotonia, if present, was graded o n a scale of 1 to 4 (see Methods). In initial studies using anthracene-9-carboxylic acid (g-AC), a wide range of concentrations was examined. In the course of these experiments it was noted that the myotonic response varied considerably from ohe stimulation site to another. Lowering the external calcium ion concentration reduced this variability and enhanced myotonia (see later discussion) but never caused repetitive activity by itself; Ringer's solution without Ca++ was routinely used for subsequent extracellular tension experiments. Calciumfree Ringer's solution has been used in previously reported studies o n myotonia of a similar nature [28, 29, 311. Normal (2.0 mM) Ca++ was present in all intracellular studies unless otherwise specified. In the tension studies, multiple sites in each muscle were sampled and recorded photographically, and the muscle response at a particular analog concentration was scored as the percentage of stimulations producing grade 2 or higher myotonia. A dose/ response curve using this semiquantitative approach is shown in Figure 1 for 7-AC. A sigmoidal curve can be fit to these data which is similar to that previously reported relating membrane GClto concentration of this analog [24]. A myotonic response is seen at 50% of the sites stimulated with approximately 2 x 10P M 9-AC. We had previously indicated, based on in-
10-6
I
10-5
10-4
Anthrocene-9-COOH (M)
Fig 1. Production of myotonic activity by 9-AC i n isolated extensor digitorum longus muscle. Each point represents thepercentage of sites in a given muscle that demonstrate grade 2 or higher myotonia in their twitch tension recordsfollowing equilibration with the indicated concentration of 9-AC. All experiments were done at 30°C in calcium-free buffer.
F i g 2. (A) Twitch tension recordsfrom isolated extensor digitorum longus muscle showing representative recordings of control and myotonic responses graded as indicated i n Methods. (B) Correlation between the concentration of A C A required toproduce 50% inhibition ofG,., in rat diaphragm fibers (I(,[24]) and that required to produce grade 2 myotonia in the extensor digitorum longus at 10 % or more of sites stimulated. Numbers correspond to identifirdon numbers given i n Table 1.
m Im
tracellular conductance measurements in rat diaphragm [24], that the inhibition constant for G a by M. this compound is 1 x The average concentrations of eight other analogs that resulted in a grade 2 or higher myotonic response in 10% or more of the sites stimulated were subsequently determined as a measure of the threshold concentration of each required to produce myotonia. Threshold concentrations were chosen instead of concentrations required for maximal activity since several of these analogs are cytotoxic at high concentrations, and reliable data at maximally effective doses cannot be obtained. When the threshold values are plotted against concentrations previously shown to reduce Gcl by 50% (Ki, Fig 2), a close correlation is found for all compounds tested.
Acids and the Action Potential The described correlation indicates a connection between muscle activity and reduction in membrane Gel, but a concomitant alteration in membrane sodium channel characteristics contributing to the myotonia must be considered. We examined the sodium system more directly using 9-AC as a representative example of the ACAs. This compound was chosen because it is the most effective of those studied in inhibiting Gc, (Ki = 1 x M) and is not toxic to the muscle in concentrations at least 25 times higher than this inhibitory concentration [241. We screened for effects of 9-AC on the sodium system by examining phase-plane diagrams of intracellularly recorded action potentials from a large number of fibers in muscles exposed to various concentrations of this aromatic acid. In a sample of 62 control fibers at 37"C, the maximum rate of rise was 672 v per second and maximum rate of fall, - 166 v per second. These val-
Aromatic
Control
Grade2 rnyotonia
10.3.
-8 -0
10-4
9 0
t
Grade3 rnyotonia
10.5
Grade 4 myotonia c------l
Stimulus
250 msec
6 pulses Q 200 hz
A
10.6
t 10-5
I 0-4
10-3
10-2
K , (for G c i ) M
B
Furman and Barchi: Aromatic Acids and Myotonia
359
ues are comparable to those reported by Bryant [8] for goat intercostal fibers at this temperature. Ward and Thesleff [ 321, using preconditioning pulses to - 100 mv, obtained higher values in rat extensor digitorum longus, as expected; our values are comparable when the effects of residual sodium inactivation at normal resting potentials are considered. The effects on action potential variables of 7-AC concentrations between 2 and 250 p M were evaluated in diaphragm fibers at 22°C (Table 2). Average resting potential was not significantly altered at any concentration of 9-AC. With increasing concentrations, a moderate decrease in the rate of rise of the action potential was noted though the rate of fall was not affected. The point of depolarization at which deviation of the membrane response from linearity was first detected, related to the membrane threshold potential, was unchanged. It should be noted that the concentration span of 9-AC examined ranged from 0.5 to 25 times the Ki for reduction of G,,. It has been reported previously that frog action potentials recorded in CI--free solution demonstrate decreased rates of rise [ 171. We examined 76 fibers following equilibration in Cl--free methylsulfate Ringer’s solution. These fibers demonstrated a decreased rate of rise comparable to that seen at the highest concentrations of 9-AC, with normal threshold and normal rate of fall. The effects o n the membrane sodium system after exposure to 9-AC therefore can be duplicated by lowering external C1concentration, thereby reducing membrane chloride currents. Thus it appears unlikely that the myotonic action of the ACAs involves a direct effect on the sodium conductance system.
and rat muscle treated with 7-AC were examined. In normal rat muscle the latency between onset of the current pulse and initiation of an action potential at current levels just exceeding threshold averaged 13.5 k 0.3 msec (average ? SEM for 34 fibers in two preparations). With intracellular square-current pulses of increasing amplitude, the time to onset of the action potential was progressively reduced. Additional action potentials during the current pulse were rarely seen. In the very few fibers in which more than one action potential could be induced, the second potential was usually reduced in amplitude. More than two driven action potentials were never seen. I n muscle fibers exposed to 50 p M 9-AC the membrane responses were different (Fig 3). Much smaller current pulses were required to reach threshold, as would be anticipated from the markedly increased membrane resistance accompanying exposure to 9-AC. For current pulses just exceeding threshold, the delay between onset of current pulse and initiation of action potential increased to 78 k 12 msec. As the amplitude of the current pulse was increased, this latency declined progressively, and additional driven action potentials were seen during the current pulse. For intermediate current stimuli, multiple action potentials were demonstrable during the current pulse, but usually no spontaneously regenerative activity occurred afterward. A progressive increase in the postresponse depolarization was variably observed, related to the number of action potentials driven. With higher intracellular current pulses, long trains of self-sustaining repetitive action potentials were seen after cessation of current injection, representing the membrane correlate of delayed relaxation in the intact muscle. These observations correspond closely to those previously reported by Adrian and Bryant [ l ] in the natural myotonia of the goat.
Intracellular Recording of Myotonic Activity The characteristics of intracellularly recorded responses to long current pulses in normal rat muscle
Table 2. Muscle Fiber Artion Potential Variables as Determined from Phase-Plane Diagrams”
Condition Control Control Y-AC, 2 x lo-” M 9-AC, 3.5 x lo-” M 0-AC, 5 x lo-” M 9-AC, 2.5 x M Cl--free Ringer’s CI--free Ringer’s plus 9-AC, 5 x M
Temp (“C)
37 22 22 22 22 22 22 22
No. of Preps
No. of Fibers
3
62
7 5
89
6 5 5 3 4
101 93 102 79
60 52
Resting Pote n cia1 (mv)
-82 -77 -75 -79 -76 -76 -75 -75
?
f f 2 f f f f
1 1 1 1 1 1 1 1
Recovery Potential (mv)
Amplitude of Action Potential (mv)
Maximum Rate of Rise (vlsec)
Maximum Rate of Fall
-84 2 2 -76 2 1 -742 1 -76k 1 -74 2 1 -74 2 1 -72 2 1 -72 2 1
992 1 104% 1 1092 2 112 2 1 105 2 1 105 2 2 1002 1 9622
672+9 2892 6 291 f 5 271 f 6 242 f 6 238f 5 225 f 5 216f 5
-16629
(vlsec)
-68
_C
6
-81 f 8 -76 f 6 -65 f 7 -77 -C 6 -66 f 7 -66 f 8
M reduces G(., by -60%; at 5 x lo-’ M, by >go%. I n CI--free “In all cases, p H IS 7.4. Anthracene-9-carboxylic acid (9-AC) at 2 X Ringer’s solution, methylsulfate is substituted for CI-. Recovery potential represents V, 10 rnsec after onset of action potential.
360 Annals of Neurology VoI 4 No 4 October 1978
Control
Iin
vrn
5x1
wc
0-5~
L
-
-F L
iFig 3 . Intracellular recordings of membrane response to a long current pulse in control diaphragmjibers (left) andfibers equilibrated with 5 x M 9-AC (right),allobtainedat 30°C with 2 m M Ca++.Controlfibersshow a short latency at threshold and only a single action potential with increasing currents. Myotonicfibers have very long latencies at threshold, multiple driven spikes with increasing cuwent, and finally se&sustaining repetitive activity after termination ofthe current pulse. (Ii, = intracellular stimulus current; V, = membranepotential.)
Temperature Dependency of Repetitive Activity
Under experimental conditions, the expression of membrane repetitive activity with reduced membrane Gcl is dependent on temperature. At 25°C in 50 p M 9-AC and 2 mM Ca++,fibers often generated only repetitive driven spikes during a long intracellular current pulse, with no regenerative activity after termination of that pulse. Raising ambient temperature to 27°C (Fig 4 ) resulted in an increase in the frequency of driven action potentials and in the appearance of continued spontaneous regenerative activity after the termination of current injection. In the same preparation, reduction of ambient temperature to 22°C abolished spontaneous regenerative activity and markedly diminished the frequency of driven spikes. Over the temperature range of 15" to 35"C, increasing temperature resulted in more pronounced myotonic features while decreasing temperature tended to inhibit the expression of myotonia. The exact temperature at which self-sustaining action potentials occurred in the presence of a saturating concentration of 9-AC varied from preparation to preparation but was usually near 26" to 28°C. Myotonia a n d Cai+
The effect of variations in extracellular Ca++ concen-
tration on the myotonic response was evaluated in diaphragm fibers following exposure to saturating concentrations of 9-AC ( 5 x 10+ M). In a given experiment, temperature was adjusted in the presence of normal (2 mM) Ca++ so that multiple driven spikes were seen during a current pulse but no selfsustaining repetitive activity occurred. After these baseline conditions were established, temperature was held constant and external Ca++ concentration was increased or decreased. As can be seen in Figure 5 , reduction of external calcium concentration to 0.5 mM in such a situation markedly augmented the membrane response to intracellular current pulses. Often even a brief (0.5 msec) stimulus was sufficient to induce repetitive activity. The frequency of driven action potentials was increased and long trains of self-regenerating action potentials were routinely seen following termination of the current pulse. When the Ca++ concentration was returned to normal, postpulse activity ceased, and only driven spikes at a lower frequency remained. Conversely, increasing external Ca++ from 2 to 6 mM under the initial conditions described tended to inhibit self-sustaining repetitive activity. Driven spikes were often still observed, but their frequency was greatly reduced. When self-sustaining activity did occur, the latency between spikes was prolonged. The effects of both external Ca++and temperature on the expression of repetitive activity appeared to be additive in the presence of low membrane Gel. Spontaneous repetitive activity after an intracellular current pulse could be produced at lower temperatures in the presence of reduced external Ca++. Conversely, such spontaneous activity at elevated temperatures could be inhibited if the external Ca++concentration was raised. Diphenylhydantoin
Since diphenylhydantoin (DPH) can suppress the appearance of myotonia in humans, its action on this model system was evaluated. Myotonia was produced in diaphragm fibers using 5 x M 9-AC and 2 mM Ca++. Temperature was adjusted so that long trains of regenerative action potentials were seen. Varying concentrations of DPH were then added to the perfusate, and membrane activity in response to long intracellular current pulses was monitored. At concentrations above 10 pg per milliliter, D P H reduced the incidence of regenerative activity; at concentrations between 20 and 30 pg per milliliter (Fig 6), spontaneous activity was usually completely eliminated in spite of saturating concentrations of 9-AC. In addition, the number of driven spikes occurring during a current pulse was markedly diminished; often only a single action potential could be obtained, even with the utilization of high stimulus currents. Furman and Barchi: Aromatic Acids and Myotonia 361
H 100 ms
5 x IO-'M
9-AC
B
5x
M 9-AC 30°C
C
Fig 4. Variations in ambient temperature modifr the myotonic response; raising temperature increases the frequency of driven responses and augments the appearance of self-sustaining repetitive activity. In this example. only driven spikes are seen at 22°C (A); at 27°C ( B ) ,self-sustaining activity is seen, but only afrer multiple driven spikeJ; at 30°C lC), se#-ustaining activity can be triggered by a single driuen spike. Note the long latency at threshold currents. Determinations uiere made in diaphragm muscle equilibrated in 5 x l Oe6 M 9-AC and normal Ringer's solution with 2 m M Ca++.
22"
Ca+ rE
c
2mM
0.5 mM Cat+ 22O c
2 mM Ca* 22OC
0.5mM Ca++ 22°C
E-U-J-
F i g 5 . At a ronstant temperature, the concentration of Ca++ greatly affects the appearance of myotonic activity when Grl is reduced by 9-AC. Two separate experiments are shown in which recordings u'ere made first i n 2 m M Ca++and then in 0.5 mM Ca'+ following equilibration in 5 x 1 O-%M 9-AC. Control jbers under these conditions do not j r e repetitively.
362 Annals of Neurology Vol 4 No 4 October 1978
n
5x M 9-AC 2 0 p g / m l DPH
30pglml DPH
5x'wM9-AC
LL-L-
Fig 6. Myotonic activity induced by 5 x 1O-" M 9-AC is inhibited by concentrations of diphenylhyduntoin (10 t o 3Opglml) romparable t o those used t o control hereditary myotonia in humans. Lower concentrationsfirst suppress self-sustained spikes; higher concentrations also mod& or suppress the driven spikes seen during a current pulse. Input resistance is not affected. All recordings were made at 30°C i n normal Ringer's solution with 2mM Ca++.
Discussion A number of hypotheses have been advanced to explain the production of myotonia by ACAs. It was proposed that Ca++ reaccumulation by the sarcoplasmic reticulum is reduced by 2,4-D [ 191, but this idea has received little support [30]. Brody [6] suggested a primary action on potassium flux, mediated by inhibition of membrane p-nitrophenylphosphatase. Using four aromatic acids related to those examined here, Bryant and Morales-Aguilera [9] demonstrated that a reduction in membrane Gc, accompanied the observed myotonic activity produced by these acids, and proposed that the myotonic activity was due to steric blockade of the membrane anion channels. In a recent study of twenty-five ACAs, Palade and Barchi [24] demonstrated that a majority of these compounds are capable of selectively reducing muscle membrane Gcl in a reversible manner, but that this reduction appears to be due to an alteration in selectivity sequence of the anion channel rather than to simple steric blockade. Effective concentrations of these acids varied widely, in a manner that could be related to their relative hydrophobicity, but other evidence suggested that binding of the compounds to a specific intramembrane site was important for their action. With the nine ACAs studied here, a close correlation was seen between the concentrations required to
produce recordable myotonic activity in vitro and those previously shown to be required for inhibition of membrane Gel. Minimal effects were seen on sodium-dependent excitation variables at concentrations producing myotonia, and those observed could be duplicated by removal of external C1-. Further, the intracellularly recorded myotonic activity induced in vitro by ACAs closely resembled that seen in vivo in hereditary myotonia of the goat [l]. These observations suggest that the primary pathophysiological factor in the myotonic activity produced by these aromatic acids is a reduced membrane Gcl. Initiation of repetitive activity under these circumstances appears to be related to accumulation of K+ in the t-tubular system following normal driven action potentials which, in the absence of the usual shunting C1- conductance, leads to progressive post-excitation depolarization of the surface membrane. Evidence of prominent depolarizingafterpotentials in myotonia associated with low Gcl has been presented in the goat [ l ] and in the rat diaphragm [27] by other investigators, and K+ accumulation in the t-system during multiple action potentials is a well-recognized phenomenon [ 1, 131. It is obvious, however, that the absolute value of Gcl is not the sole determinant of whether repetitive activity will be seen in a given muscle fiber. Thus, in the presence of a concentration of 9-AC known to reduce Ga to values of 10% or less of control, the appearance of myotonic activity is reversibly controllable by altering the external Ca++ concentration or the environmental temperature. A recent clinical study has demonstrated a temperature dependency similar to that reported here for myotonic activity in patients with myotonia congenita [26]. The effects of Ca++and temperature may be due to the alterations they induce in the rate constants for activation and inactivation of the membrane sodium
Furman and Barchi: Aromatic Acids and Myotonia 363
system. Alterations in divalent cation concentration shift the curves relating these activation rates along the voltage axis without changing their configuration [ l l , 151. This is thought to be due to an alteration in membrane surface charge density induced by charge screening effects 116, 221. Since the relationship between the reverse rate constant for activation of sodium conductance (Pm)and membrane potential near the resting potential is changing rapidly while similar relationships for other activation and inactivation rates are changing more slowly, moderate alterations in calcium concentration will be predominantly expressed as alterations in Pm.This activation rate is also temperature dependent, increasing approximately threefold for each 10°C increase in temperature in most systems studied. Theoretical studies have demonstrated that repetitive activity in computer models of the muscle membrane produced by lowering Gc, are especially sensitive to variations in Pm,although changes in this variable alone with normal Gcl will not produce such activity. Barchi [3] reported that shifting the curves relating Pm to membrane potential along the voltage axis by as little as 2 or 3 mv could eliminate repetitive activity in the myotonic computer model. Adrian and Marshall [2] likewise reported that in their model of the propagated muscle action potential, the appearance of self-sustaining action potentials after termination of stimulus current was critically dependent on the value of Pm and that reductions in Pm predisposed toward myotonia. The mechanism of action of DPH in suppressing myotonic activity remains unclear. At concentrations used clinically (5 to 20 pg/ml in serum), this drug has been shown to reduce the net influx of sodium ion during action potential generation in nerve [25]. At similar concentrations we have demonstrated a significant effect on the propagated muscle action potential, characterized by a reduction in the maximal rates of rise and fall as well as an inciease in the depolarization required to attain threshold for regenerative activity. These changes are clearly different from those observed with ACAs, although rates of rise are apparently affected by both. The observed effects suggest an action of DPH, either direct or indirect, on the sodium activation functions of the muscle membrane. Voltage clamp experiments are required for precise definition of the nature of this action. Supported in part by National Institutes of Health Grant NS08075 and by a grant from the Muscular Dystrophy Association.
Dr Barchi is recipient of a Research Career Development Award from the National Institutes of Health.
364 Annals of Neurology Vol 4 No 4 October 1978
References 1. Adrian RH, Bryant SH: O n the repetitive discharges in myotonic muscle fibers. J Physiol ILond) 240:50?-515, 1974 2. Adrian RH, Marshall MW: Action potentials reconstructed in normal and myotonic muscle fibers. J Physiol (Lond) 258~125-143, 1976 3. Barchi R L Myotonia: an evaluation of the chloride hypothesis. Arch Neurol 32:175-180, 1975 4. Berwick P: 2,4 Dichlorophenoxyacetic acid poisoning in man. JAMA 214:1114-1117, 1970 5. Bretag AH: Mathematical modelling of the myotonic action potential, in Desmedt JE (ed): New Developments in Electromyography and Clinical Neurophysiology. Basel, Karger, 1973, VOI 1, pp 464-482 6. Brody I: Myotonia induced by monocarboxylic aromatic acids. Arch Neurol28:243-246, 1973 7. Bryant SH: Cable properties of external intercostal muscle fibers from myotonic and non-myotonic goats. J Physiol (Lond) 204:539-550, 1969 8. Bryant SH: The electrophysiology of myotonia, with a review of congenital myotonia in goats, in Desmedt JE (ed): New Developments in Electromyography and Clinical Neurophysiology. Basel, Karger, 1974, vol 1, pp 420-450 9. Bryant SH, Morales-Aguilera A: Chloride conductance in normal and myotonic muscle fibers and the action of monocarboxylic acids. J Physiol (Lond) 2 19:367-383, 197 1 10. Bucher N L Effects of 2,4-dichlorophenoxyaceticacid on experimental animals. Proc SOC Exp Biol Med 63:204-205, 1946 11. Campbell D, Hille B: Kinetic and pharmacological properties of the sodium channel of frog skeletal muscle. J Gen Physiol 671309-323, 1976 12. Ezyaguirre C, Folk B, Zierler K, et al: Experimental myotonia and repetitive phenomena; the veratrinic effects of 2,4dichlorophenoxyacetic acid in the rat. Am J Physiol 155:6977, 1948 13. Freygang WH, Goldstein PA, Hellam DC: The after-potential that follows trains of impulses in frog muscle. J Gen Physiol 47:929-952, 1964 14. Furman RE, Barchi RL: The mechanism of myotonia induced by aromatic carboxylic acids. Neurology (Minneap) 4:378, 1977 15. Hille B: Charges and potentials at the nerve surface. Divalent cations and pH. J Gen Physiol 51:221-236, 1968 16. Hille B, Woodhull A, Shapiro B: Negative surface charge near sodium channels of nerve: Divalent ions, monovalent ions, and pH. Philos Trans R SOCLond 270:301-318, 1975 17. Jenerick H : Phase plane trajectories of the muscle spike potential. Biophys J 3:363-376, 1963 18. Jenerick H: An analysis of the striated muscle fiber action current. Biophys J 477-91, 1964 19. Kuhn E, Stein W: Modell-Myotonie nach 2,4 Dichlorophenoxyacecat bei der Ratte. Klin Wochenschr 42:12 151216, 1964 20. Landau W M The essential mechanism in myotonia. Neurology (Minneap) 2:369-388, 1952 21. Lipicky R, Bryant S, Salmon J: Cable parameters, sodium, potassium, chloride and water content, and potassium efflux in isolated external intercostal muscles of normal volunteers and patients with myotonia congenita. J Clin Invest 50:20912103, 1971 22. McLaughlin SG, Szabo G, Eisenman G: Divalent ions and surface potential of charged phospholipid membranes. J Gen Physiol 58:667-687, 1971 23. Palade PT, Barchi RL: Characteristics of the chloride conduc-
24.
25. 26. 27.
28.
tance in muscle fibers of the rat diaphragm. J Gen Physiol 691325-342, 1777 Palade PT, Barchi R L On the inhibition of musclc membrane chloride conductances by aromatic carboxylic acids. J Gen Physiol 69:875-896, 1977 Pincus JH: Diphenylhydantoin and ion flux in lobster nerve. Arch Neurol 26:4-10, 1972 Ricker K, Hertel G, Langscheid K, et al: Myotonia not aggravated by cooling. J Neurol 2169-20, 1977 Rudel R, Keller M: Intracellular recording of myotonic muscle fibers, in Bradley WG (ed): Recent Advances in Myology. Amsterdam, Excerpta Medica, 1974, ex 89, no. 360, pp 441-445 Rudel R, Senges J: Mammalian skeletal muscle: reduced chlo-
29.
30. 31.
32.
ride conductance in drug-induced myotonia and induction of myotonia by low chloride solution. Naunyn Schmiedebergs Arch Pharmacol274:337-347, 1972 Rudel R, Senges J: Experimental myotonia, in Desmedt JE (ed): New Developments in Electromyography and Clinical Neurophysiology. Basel, Karger, 1973, vol 1, pp 483-491 Samaha RJ, Schroder JM, Rebieg J, et al: Studies on myotonia. Arch Neurol 17:22-33, 1967 Senges J, Rudel R Experimental myotonia in mammalian skeletal muscle. Pfluegers Arch 331:315-323, 1972 Ward MR, Thesleff S: The temperature dependence of action potentials in rat skeletal muscle fibers. Acta Physiol Scand 91~574-576, 1974
Furman and Barchi: Aromatic Acids and Myotonia 365
A series of nine related aromatic monocaEboxylic acids (ACAs) previously shown to inhibit muscle membrane chloride conductance (Gcl) selectively in the rat were studied for their ability to produce myotonia. All nine
induced characteristic repetitive electrical activity and delayed relaxation in isolated muscle, although the concentrations required for this action varied widely. In each case, myotonia was observed at concentrations that correlated closely with previously determined half-maximal concentrations for inhibition of G a . Intracellular recordings from muscle made myotonic with ACA revealed prolonged latencies at rheobase, multiple driven spikes, and self-sustaining repetitive activity similar to that previously reported in hereditary goat myotonia. Phase-plane diagrams of membrane action potentials recorded after exposure to the most effective of these compounds suggested little effect on the voltage-dependent sodium system. The changes seen could be duplicated by simple removal of chloride ion. The expression of repetitive electrical activity in the presence of low membrane Gcl depends on ambient temperature and on the concentration of calcium ion. Increasing temperature and decreasing Ca++ predispose toward myotonic activity; converse conditions inhibit myotonia. Myotonia induced by ACA is inhibited by concentrations of diphenylhydantoin that are clinically effective in controlling hereditary myotonia in humans. Furman RE, Barchi R L The pathophysiology of myotonia produced by aromatic carboxylic acids. Ann Neurol 4:357-365, 1978
Myotonia, a form of pathologically delayed relaxation of striated muscle, is seen in several human and animal diseases. This delayed relaxation is the result of self-sustaining repetitive action potentials generated in the surface membrane in response to a brief stimulus [20]. Evidence suggests that the basic pathophysiological defect in myotonia may be an abnormality of chloride conductance (Ga) in the muscle fiber surface membrane [8, 21, 311. Studies on hereditary myotonia in goats [7] and on myotonia congenita in humans [2 11 have demonstrated in each case an abnormally low membrane Ga. The potential relationship between such a reduced Gcl and repetitive electrical activity has been shown in several studies using mathematical models of the muscle surface membrane 12, 3 , 51. Myotonia resembling that seen in natural disease can be produced in otherwise normal animals by a number of chemical agents, the best known of which is 2,4-dichlorophenoxyacetic acid (2,4-D) [4, 10, 121. In vitro muscle studies have documented a reduction in Gcl after exposure to 2,4-D [29] and several related carboxylic acids [9]. In a recent study of twenty-five aromatic carboxylic acids (ACAs) similar
to 2,4-D, we demonstrated that the sarcolemmal Gc1
From the Departments of Neurology and of Biochemistry and Biophysics, University of Pennsylvania School of Medicine, Philadelphia, PA.
Address reprint requests to Dr Barchi, Department of Neurology, Hospital of the University of Pennsylvania, Philadelphia, PA 19104.
was reversibly inhibited by nineteen of these acids [ 2 4 ] . Inhibition appeared to involve partitioning of the compounds into the surface membrane with subsequent specific interaction, presumably at the site of chloride translocation. Our earlier work provided us with a series of compounds having varying physical characteristics that produce inhibition of Gcl at known concentrations in vitro. In the present work we used these compounds to study the general mechanism by which repetitive activity is produced in mammalian muscle by ACAs. We demonstrated that the production of myotonia by these compounds is in each case directly related to the ability of that compound to reduce membrane Gcl and that the intracellularly recorded abnormal electrical activity closely resembles that previously documented in the inherited myotonia of the goat [ 11. Myotonia induced by ACAs exhibits many characteristics similar to those seen in hereditary myotonia congenita in man, suggesting a basic similarity between these states. A preliminary report of portions of this work has been presented [ 141.
Accepted for publication Apr 13, 1978.
0364-5134/78/0004-0410$01.25 @ 1978 by R. E. Furman
357
Methods Aromatic carboxylic acids were obtained from Aldrich Chemical Corporation, Milwaukee, WI; K and K Laboratories, Plainview, NY; and Eastman Kodak Company, Rochester, NY. Several were the generous gift of Dr R. B. Moffett of the Up john Pharmaceutical Company. Diphenylhydantoin was obtained from Elkins-Sinn, Inc, Cherry Hill, NJ. Male Wistar rats of 250 to 350 gm were used for all experiments. For twitch tension studies, extensor digitorum longus muscle removed from the tendon of origin to tendon of insertion was used. For intracellular recordings the hemidiaphragm was removed and prepared as previously described [23]. In each case the muscle was transferred to a temperature-controlled recording chamber and perfused continuously as previously described [2 31. Composition of the normal and Clk-free methylsulfate Ringer's solution was as given earlier [23, 241. Measurements of twitch tension were made with a Disa Model 5 7 tension transducer. Platinum surface electrodes spaced 1 mm apart were used for stimulation in these initial studies. Muscles were adjusted in the chamber to about 1.2 times their resting length, and a short series of constantcurrent pulses (6 pulses, 0.5 msec, 200 Hz), found to be optimal for inducing a myotonic response, was used for stimulation. Pulse amplitude was adjusted for maximal response. At least 30 seconds was allowed between each stimulus set to minimize the '' warm-up" phenomenon observed. Tension records were analyzed from photographic records. Tension responses after exposure of the muscle to an ACA were graded on the basis of ratio of peak tension generated during stimulation to plateau tension maintained more than 20 msec after termination of stimuli. Grade 1 indicates sustained tension visible but less than 10% of peak tension during stimulation; grade 2, sustained tension of 10 to 33% of peak tension; grade 3, 33 to 66%; and grade 4,66 to 100%. Average grading for agiven muscle at a particular concentration of aromatic acid was based on recordings from fifteen or more separate stimulations at multiple sites. Microelectrode techniques for intracellular recordings were as previously described [23, 241. Phase-plane recordings were obtained by electronic differentiation of the intracellularly recorded voltage response using an active network whose transfer characteristics were carefully controlled to minimize phase shift between input and output signal. The rate of change of membrane potential (dV,/dt) was displayed as a function of membrane potential (V,) on a dual-beam oscilloscope along with the usual display of V, versus time, and analyzed after enlargement from photographic records. Methods used for calibration and analysis of phase-plane diagrams were essentially those of Jenerick [17, 181. Results
Prodnction of Myotonic Activity In order to test for correlation between the inhibition of chloride conductance and induction of myotonic activity, nine ACAs previously shown to inhibit
358
Annals of Neurology Vol 4 No 4 October 1978
Table I . Benzoic Acid De7ivative.r Examined for Myotonic Effect in the Present Report"
Compound
K,(M)
1. 2.
1.1 x 8.0 x 3.0 x 3.1 x 6.0 x 9.5 x 1.5 x 2.0 x 3.0 x
Anthracene-9-carboxylicacid Pentachlorobentoic acid 3. 2,5-Dimethylbenzoic acid 4. 2,5-Dichlorobenzoic acid 5 . 2,3,5,6-Tetramethylbenzoicacid 6. 2,3,6-Trichlorobenzoic acid 7. 3,5-Dimethylbenzoic acid 8. 2,6-Dichlorobenzoic acid 9. 3,5-DimethyI-4-nitrobenzoic acid
lo-" 10-5 10-4 10-4
10-4 10-4
lo-"
aKi represents the concentration previously determined to reduce G,, by 50% (Palade and Barchi [241). Identitication numbers correspond to points in Figure 2.
membrane Gcl in rat diaphragm were examined for their ability to induce myotonia in the extensor digitorum longus muscle of the rat in vitro (Table 1). These compounds had inhibition constants (Ki, that concentration producing 50% inhibition) for Gcl ranging over three orders of magnitude. Muscles were equilibrated at a given ACA concentration for 15 minutes or longer and then stimulated supramaximally with surface electrodes at a number of sites away from the motor point while the resultant tension generated was recorded. Myotonia, if present, was graded o n a scale of 1 to 4 (see Methods). In initial studies using anthracene-9-carboxylic acid (g-AC), a wide range of concentrations was examined. In the course of these experiments it was noted that the myotonic response varied considerably from ohe stimulation site to another. Lowering the external calcium ion concentration reduced this variability and enhanced myotonia (see later discussion) but never caused repetitive activity by itself; Ringer's solution without Ca++ was routinely used for subsequent extracellular tension experiments. Calciumfree Ringer's solution has been used in previously reported studies o n myotonia of a similar nature [28, 29, 311. Normal (2.0 mM) Ca++ was present in all intracellular studies unless otherwise specified. In the tension studies, multiple sites in each muscle were sampled and recorded photographically, and the muscle response at a particular analog concentration was scored as the percentage of stimulations producing grade 2 or higher myotonia. A dose/ response curve using this semiquantitative approach is shown in Figure 1 for 7-AC. A sigmoidal curve can be fit to these data which is similar to that previously reported relating membrane GClto concentration of this analog [24]. A myotonic response is seen at 50% of the sites stimulated with approximately 2 x 10P M 9-AC. We had previously indicated, based on in-
10-6
I
10-5
10-4
Anthrocene-9-COOH (M)
Fig 1. Production of myotonic activity by 9-AC i n isolated extensor digitorum longus muscle. Each point represents thepercentage of sites in a given muscle that demonstrate grade 2 or higher myotonia in their twitch tension recordsfollowing equilibration with the indicated concentration of 9-AC. All experiments were done at 30°C in calcium-free buffer.
F i g 2. (A) Twitch tension recordsfrom isolated extensor digitorum longus muscle showing representative recordings of control and myotonic responses graded as indicated i n Methods. (B) Correlation between the concentration of A C A required toproduce 50% inhibition ofG,., in rat diaphragm fibers (I(,[24]) and that required to produce grade 2 myotonia in the extensor digitorum longus at 10 % or more of sites stimulated. Numbers correspond to identifirdon numbers given i n Table 1.
m Im
tracellular conductance measurements in rat diaphragm [24], that the inhibition constant for G a by M. this compound is 1 x The average concentrations of eight other analogs that resulted in a grade 2 or higher myotonic response in 10% or more of the sites stimulated were subsequently determined as a measure of the threshold concentration of each required to produce myotonia. Threshold concentrations were chosen instead of concentrations required for maximal activity since several of these analogs are cytotoxic at high concentrations, and reliable data at maximally effective doses cannot be obtained. When the threshold values are plotted against concentrations previously shown to reduce Gcl by 50% (Ki, Fig 2), a close correlation is found for all compounds tested.
Acids and the Action Potential The described correlation indicates a connection between muscle activity and reduction in membrane Gel, but a concomitant alteration in membrane sodium channel characteristics contributing to the myotonia must be considered. We examined the sodium system more directly using 9-AC as a representative example of the ACAs. This compound was chosen because it is the most effective of those studied in inhibiting Gc, (Ki = 1 x M) and is not toxic to the muscle in concentrations at least 25 times higher than this inhibitory concentration [241. We screened for effects of 9-AC on the sodium system by examining phase-plane diagrams of intracellularly recorded action potentials from a large number of fibers in muscles exposed to various concentrations of this aromatic acid. In a sample of 62 control fibers at 37"C, the maximum rate of rise was 672 v per second and maximum rate of fall, - 166 v per second. These val-
Aromatic
Control
Grade2 rnyotonia
10.3.
-8 -0
10-4
9 0
t
Grade3 rnyotonia
10.5
Grade 4 myotonia c------l
Stimulus
250 msec
6 pulses Q 200 hz
A
10.6
t 10-5
I 0-4
10-3
10-2
K , (for G c i ) M
B
Furman and Barchi: Aromatic Acids and Myotonia
359
ues are comparable to those reported by Bryant [8] for goat intercostal fibers at this temperature. Ward and Thesleff [ 321, using preconditioning pulses to - 100 mv, obtained higher values in rat extensor digitorum longus, as expected; our values are comparable when the effects of residual sodium inactivation at normal resting potentials are considered. The effects on action potential variables of 7-AC concentrations between 2 and 250 p M were evaluated in diaphragm fibers at 22°C (Table 2). Average resting potential was not significantly altered at any concentration of 9-AC. With increasing concentrations, a moderate decrease in the rate of rise of the action potential was noted though the rate of fall was not affected. The point of depolarization at which deviation of the membrane response from linearity was first detected, related to the membrane threshold potential, was unchanged. It should be noted that the concentration span of 9-AC examined ranged from 0.5 to 25 times the Ki for reduction of G,,. It has been reported previously that frog action potentials recorded in CI--free solution demonstrate decreased rates of rise [ 171. We examined 76 fibers following equilibration in Cl--free methylsulfate Ringer’s solution. These fibers demonstrated a decreased rate of rise comparable to that seen at the highest concentrations of 9-AC, with normal threshold and normal rate of fall. The effects o n the membrane sodium system after exposure to 9-AC therefore can be duplicated by lowering external C1concentration, thereby reducing membrane chloride currents. Thus it appears unlikely that the myotonic action of the ACAs involves a direct effect on the sodium conductance system.
and rat muscle treated with 7-AC were examined. In normal rat muscle the latency between onset of the current pulse and initiation of an action potential at current levels just exceeding threshold averaged 13.5 k 0.3 msec (average ? SEM for 34 fibers in two preparations). With intracellular square-current pulses of increasing amplitude, the time to onset of the action potential was progressively reduced. Additional action potentials during the current pulse were rarely seen. In the very few fibers in which more than one action potential could be induced, the second potential was usually reduced in amplitude. More than two driven action potentials were never seen. I n muscle fibers exposed to 50 p M 9-AC the membrane responses were different (Fig 3). Much smaller current pulses were required to reach threshold, as would be anticipated from the markedly increased membrane resistance accompanying exposure to 9-AC. For current pulses just exceeding threshold, the delay between onset of current pulse and initiation of action potential increased to 78 k 12 msec. As the amplitude of the current pulse was increased, this latency declined progressively, and additional driven action potentials were seen during the current pulse. For intermediate current stimuli, multiple action potentials were demonstrable during the current pulse, but usually no spontaneously regenerative activity occurred afterward. A progressive increase in the postresponse depolarization was variably observed, related to the number of action potentials driven. With higher intracellular current pulses, long trains of self-sustaining repetitive action potentials were seen after cessation of current injection, representing the membrane correlate of delayed relaxation in the intact muscle. These observations correspond closely to those previously reported by Adrian and Bryant [ l ] in the natural myotonia of the goat.
Intracellular Recording of Myotonic Activity The characteristics of intracellularly recorded responses to long current pulses in normal rat muscle
Table 2. Muscle Fiber Artion Potential Variables as Determined from Phase-Plane Diagrams”
Condition Control Control Y-AC, 2 x lo-” M 9-AC, 3.5 x lo-” M 0-AC, 5 x lo-” M 9-AC, 2.5 x M Cl--free Ringer’s CI--free Ringer’s plus 9-AC, 5 x M
Temp (“C)
37 22 22 22 22 22 22 22
No. of Preps
No. of Fibers
3
62
7 5
89
6 5 5 3 4
101 93 102 79
60 52
Resting Pote n cia1 (mv)
-82 -77 -75 -79 -76 -76 -75 -75
?
f f 2 f f f f
1 1 1 1 1 1 1 1
Recovery Potential (mv)
Amplitude of Action Potential (mv)
Maximum Rate of Rise (vlsec)
Maximum Rate of Fall
-84 2 2 -76 2 1 -742 1 -76k 1 -74 2 1 -74 2 1 -72 2 1 -72 2 1
992 1 104% 1 1092 2 112 2 1 105 2 1 105 2 2 1002 1 9622
672+9 2892 6 291 f 5 271 f 6 242 f 6 238f 5 225 f 5 216f 5
-16629
(vlsec)
-68
_C
6
-81 f 8 -76 f 6 -65 f 7 -77 -C 6 -66 f 7 -66 f 8
M reduces G(., by -60%; at 5 x lo-’ M, by >go%. I n CI--free “In all cases, p H IS 7.4. Anthracene-9-carboxylic acid (9-AC) at 2 X Ringer’s solution, methylsulfate is substituted for CI-. Recovery potential represents V, 10 rnsec after onset of action potential.
360 Annals of Neurology VoI 4 No 4 October 1978
Control
Iin
vrn
5x1
wc
0-5~
L
-
-F L
iFig 3 . Intracellular recordings of membrane response to a long current pulse in control diaphragmjibers (left) andfibers equilibrated with 5 x M 9-AC (right),allobtainedat 30°C with 2 m M Ca++.Controlfibersshow a short latency at threshold and only a single action potential with increasing currents. Myotonicfibers have very long latencies at threshold, multiple driven spikes with increasing cuwent, and finally se&sustaining repetitive activity after termination ofthe current pulse. (Ii, = intracellular stimulus current; V, = membranepotential.)
Temperature Dependency of Repetitive Activity
Under experimental conditions, the expression of membrane repetitive activity with reduced membrane Gcl is dependent on temperature. At 25°C in 50 p M 9-AC and 2 mM Ca++,fibers often generated only repetitive driven spikes during a long intracellular current pulse, with no regenerative activity after termination of that pulse. Raising ambient temperature to 27°C (Fig 4 ) resulted in an increase in the frequency of driven action potentials and in the appearance of continued spontaneous regenerative activity after the termination of current injection. In the same preparation, reduction of ambient temperature to 22°C abolished spontaneous regenerative activity and markedly diminished the frequency of driven spikes. Over the temperature range of 15" to 35"C, increasing temperature resulted in more pronounced myotonic features while decreasing temperature tended to inhibit the expression of myotonia. The exact temperature at which self-sustaining action potentials occurred in the presence of a saturating concentration of 9-AC varied from preparation to preparation but was usually near 26" to 28°C. Myotonia a n d Cai+
The effect of variations in extracellular Ca++ concen-
tration on the myotonic response was evaluated in diaphragm fibers following exposure to saturating concentrations of 9-AC ( 5 x 10+ M). In a given experiment, temperature was adjusted in the presence of normal (2 mM) Ca++ so that multiple driven spikes were seen during a current pulse but no selfsustaining repetitive activity occurred. After these baseline conditions were established, temperature was held constant and external Ca++ concentration was increased or decreased. As can be seen in Figure 5 , reduction of external calcium concentration to 0.5 mM in such a situation markedly augmented the membrane response to intracellular current pulses. Often even a brief (0.5 msec) stimulus was sufficient to induce repetitive activity. The frequency of driven action potentials was increased and long trains of self-regenerating action potentials were routinely seen following termination of the current pulse. When the Ca++ concentration was returned to normal, postpulse activity ceased, and only driven spikes at a lower frequency remained. Conversely, increasing external Ca++ from 2 to 6 mM under the initial conditions described tended to inhibit self-sustaining repetitive activity. Driven spikes were often still observed, but their frequency was greatly reduced. When self-sustaining activity did occur, the latency between spikes was prolonged. The effects of both external Ca++and temperature on the expression of repetitive activity appeared to be additive in the presence of low membrane Gel. Spontaneous repetitive activity after an intracellular current pulse could be produced at lower temperatures in the presence of reduced external Ca++. Conversely, such spontaneous activity at elevated temperatures could be inhibited if the external Ca++concentration was raised. Diphenylhydantoin
Since diphenylhydantoin (DPH) can suppress the appearance of myotonia in humans, its action on this model system was evaluated. Myotonia was produced in diaphragm fibers using 5 x M 9-AC and 2 mM Ca++. Temperature was adjusted so that long trains of regenerative action potentials were seen. Varying concentrations of DPH were then added to the perfusate, and membrane activity in response to long intracellular current pulses was monitored. At concentrations above 10 pg per milliliter, D P H reduced the incidence of regenerative activity; at concentrations between 20 and 30 pg per milliliter (Fig 6), spontaneous activity was usually completely eliminated in spite of saturating concentrations of 9-AC. In addition, the number of driven spikes occurring during a current pulse was markedly diminished; often only a single action potential could be obtained, even with the utilization of high stimulus currents. Furman and Barchi: Aromatic Acids and Myotonia 361
H 100 ms
5 x IO-'M
9-AC
B
5x
M 9-AC 30°C
C
Fig 4. Variations in ambient temperature modifr the myotonic response; raising temperature increases the frequency of driven responses and augments the appearance of self-sustaining repetitive activity. In this example. only driven spikes are seen at 22°C (A); at 27°C ( B ) ,self-sustaining activity is seen, but only afrer multiple driven spikeJ; at 30°C lC), se#-ustaining activity can be triggered by a single driuen spike. Note the long latency at threshold currents. Determinations uiere made in diaphragm muscle equilibrated in 5 x l Oe6 M 9-AC and normal Ringer's solution with 2 m M Ca++.
22"
Ca+ rE
c
2mM
0.5 mM Cat+ 22O c
2 mM Ca* 22OC
0.5mM Ca++ 22°C
E-U-J-
F i g 5 . At a ronstant temperature, the concentration of Ca++ greatly affects the appearance of myotonic activity when Grl is reduced by 9-AC. Two separate experiments are shown in which recordings u'ere made first i n 2 m M Ca++and then in 0.5 mM Ca'+ following equilibration in 5 x 1 O-%M 9-AC. Control jbers under these conditions do not j r e repetitively.
362 Annals of Neurology Vol 4 No 4 October 1978
n
5x M 9-AC 2 0 p g / m l DPH
30pglml DPH
5x'wM9-AC
LL-L-
Fig 6. Myotonic activity induced by 5 x 1O-" M 9-AC is inhibited by concentrations of diphenylhyduntoin (10 t o 3Opglml) romparable t o those used t o control hereditary myotonia in humans. Lower concentrationsfirst suppress self-sustained spikes; higher concentrations also mod& or suppress the driven spikes seen during a current pulse. Input resistance is not affected. All recordings were made at 30°C i n normal Ringer's solution with 2mM Ca++.
Discussion A number of hypotheses have been advanced to explain the production of myotonia by ACAs. It was proposed that Ca++ reaccumulation by the sarcoplasmic reticulum is reduced by 2,4-D [ 191, but this idea has received little support [30]. Brody [6] suggested a primary action on potassium flux, mediated by inhibition of membrane p-nitrophenylphosphatase. Using four aromatic acids related to those examined here, Bryant and Morales-Aguilera [9] demonstrated that a reduction in membrane Gc, accompanied the observed myotonic activity produced by these acids, and proposed that the myotonic activity was due to steric blockade of the membrane anion channels. In a recent study of twenty-five ACAs, Palade and Barchi [24] demonstrated that a majority of these compounds are capable of selectively reducing muscle membrane Gcl in a reversible manner, but that this reduction appears to be due to an alteration in selectivity sequence of the anion channel rather than to simple steric blockade. Effective concentrations of these acids varied widely, in a manner that could be related to their relative hydrophobicity, but other evidence suggested that binding of the compounds to a specific intramembrane site was important for their action. With the nine ACAs studied here, a close correlation was seen between the concentrations required to
produce recordable myotonic activity in vitro and those previously shown to be required for inhibition of membrane Gel. Minimal effects were seen on sodium-dependent excitation variables at concentrations producing myotonia, and those observed could be duplicated by removal of external C1-. Further, the intracellularly recorded myotonic activity induced in vitro by ACAs closely resembled that seen in vivo in hereditary myotonia of the goat [l]. These observations suggest that the primary pathophysiological factor in the myotonic activity produced by these aromatic acids is a reduced membrane Gcl. Initiation of repetitive activity under these circumstances appears to be related to accumulation of K+ in the t-tubular system following normal driven action potentials which, in the absence of the usual shunting C1- conductance, leads to progressive post-excitation depolarization of the surface membrane. Evidence of prominent depolarizingafterpotentials in myotonia associated with low Gcl has been presented in the goat [ l ] and in the rat diaphragm [27] by other investigators, and K+ accumulation in the t-system during multiple action potentials is a well-recognized phenomenon [ 1, 131. It is obvious, however, that the absolute value of Gcl is not the sole determinant of whether repetitive activity will be seen in a given muscle fiber. Thus, in the presence of a concentration of 9-AC known to reduce Ga to values of 10% or less of control, the appearance of myotonic activity is reversibly controllable by altering the external Ca++ concentration or the environmental temperature. A recent clinical study has demonstrated a temperature dependency similar to that reported here for myotonic activity in patients with myotonia congenita [26]. The effects of Ca++and temperature may be due to the alterations they induce in the rate constants for activation and inactivation of the membrane sodium
Furman and Barchi: Aromatic Acids and Myotonia 363
system. Alterations in divalent cation concentration shift the curves relating these activation rates along the voltage axis without changing their configuration [ l l , 151. This is thought to be due to an alteration in membrane surface charge density induced by charge screening effects 116, 221. Since the relationship between the reverse rate constant for activation of sodium conductance (Pm)and membrane potential near the resting potential is changing rapidly while similar relationships for other activation and inactivation rates are changing more slowly, moderate alterations in calcium concentration will be predominantly expressed as alterations in Pm.This activation rate is also temperature dependent, increasing approximately threefold for each 10°C increase in temperature in most systems studied. Theoretical studies have demonstrated that repetitive activity in computer models of the muscle membrane produced by lowering Gc, are especially sensitive to variations in Pm,although changes in this variable alone with normal Gcl will not produce such activity. Barchi [3] reported that shifting the curves relating Pm to membrane potential along the voltage axis by as little as 2 or 3 mv could eliminate repetitive activity in the myotonic computer model. Adrian and Marshall [2] likewise reported that in their model of the propagated muscle action potential, the appearance of self-sustaining action potentials after termination of stimulus current was critically dependent on the value of Pm and that reductions in Pm predisposed toward myotonia. The mechanism of action of DPH in suppressing myotonic activity remains unclear. At concentrations used clinically (5 to 20 pg/ml in serum), this drug has been shown to reduce the net influx of sodium ion during action potential generation in nerve [25]. At similar concentrations we have demonstrated a significant effect on the propagated muscle action potential, characterized by a reduction in the maximal rates of rise and fall as well as an inciease in the depolarization required to attain threshold for regenerative activity. These changes are clearly different from those observed with ACAs, although rates of rise are apparently affected by both. The observed effects suggest an action of DPH, either direct or indirect, on the sodium activation functions of the muscle membrane. Voltage clamp experiments are required for precise definition of the nature of this action. Supported in part by National Institutes of Health Grant NS08075 and by a grant from the Muscular Dystrophy Association.
Dr Barchi is recipient of a Research Career Development Award from the National Institutes of Health.
364 Annals of Neurology Vol 4 No 4 October 1978
References 1. Adrian RH, Bryant SH: O n the repetitive discharges in myotonic muscle fibers. J Physiol ILond) 240:50?-515, 1974 2. Adrian RH, Marshall MW: Action potentials reconstructed in normal and myotonic muscle fibers. J Physiol (Lond) 258~125-143, 1976 3. Barchi R L Myotonia: an evaluation of the chloride hypothesis. Arch Neurol 32:175-180, 1975 4. Berwick P: 2,4 Dichlorophenoxyacetic acid poisoning in man. JAMA 214:1114-1117, 1970 5. Bretag AH: Mathematical modelling of the myotonic action potential, in Desmedt JE (ed): New Developments in Electromyography and Clinical Neurophysiology. Basel, Karger, 1973, VOI 1, pp 464-482 6. Brody I: Myotonia induced by monocarboxylic aromatic acids. Arch Neurol28:243-246, 1973 7. Bryant SH: Cable properties of external intercostal muscle fibers from myotonic and non-myotonic goats. J Physiol (Lond) 204:539-550, 1969 8. Bryant SH: The electrophysiology of myotonia, with a review of congenital myotonia in goats, in Desmedt JE (ed): New Developments in Electromyography and Clinical Neurophysiology. Basel, Karger, 1974, vol 1, pp 420-450 9. Bryant SH, Morales-Aguilera A: Chloride conductance in normal and myotonic muscle fibers and the action of monocarboxylic acids. J Physiol (Lond) 2 19:367-383, 197 1 10. Bucher N L Effects of 2,4-dichlorophenoxyaceticacid on experimental animals. Proc SOC Exp Biol Med 63:204-205, 1946 11. Campbell D, Hille B: Kinetic and pharmacological properties of the sodium channel of frog skeletal muscle. J Gen Physiol 671309-323, 1976 12. Ezyaguirre C, Folk B, Zierler K, et al: Experimental myotonia and repetitive phenomena; the veratrinic effects of 2,4dichlorophenoxyacetic acid in the rat. Am J Physiol 155:6977, 1948 13. Freygang WH, Goldstein PA, Hellam DC: The after-potential that follows trains of impulses in frog muscle. J Gen Physiol 47:929-952, 1964 14. Furman RE, Barchi RL: The mechanism of myotonia induced by aromatic carboxylic acids. Neurology (Minneap) 4:378, 1977 15. Hille B: Charges and potentials at the nerve surface. Divalent cations and pH. J Gen Physiol 51:221-236, 1968 16. Hille B, Woodhull A, Shapiro B: Negative surface charge near sodium channels of nerve: Divalent ions, monovalent ions, and pH. Philos Trans R SOCLond 270:301-318, 1975 17. Jenerick H : Phase plane trajectories of the muscle spike potential. Biophys J 3:363-376, 1963 18. Jenerick H: An analysis of the striated muscle fiber action current. Biophys J 477-91, 1964 19. Kuhn E, Stein W: Modell-Myotonie nach 2,4 Dichlorophenoxyacecat bei der Ratte. Klin Wochenschr 42:12 151216, 1964 20. Landau W M The essential mechanism in myotonia. Neurology (Minneap) 2:369-388, 1952 21. Lipicky R, Bryant S, Salmon J: Cable parameters, sodium, potassium, chloride and water content, and potassium efflux in isolated external intercostal muscles of normal volunteers and patients with myotonia congenita. J Clin Invest 50:20912103, 1971 22. McLaughlin SG, Szabo G, Eisenman G: Divalent ions and surface potential of charged phospholipid membranes. J Gen Physiol 58:667-687, 1971 23. Palade PT, Barchi RL: Characteristics of the chloride conduc-
24.
25. 26. 27.
28.
tance in muscle fibers of the rat diaphragm. J Gen Physiol 691325-342, 1777 Palade PT, Barchi R L On the inhibition of musclc membrane chloride conductances by aromatic carboxylic acids. J Gen Physiol 69:875-896, 1977 Pincus JH: Diphenylhydantoin and ion flux in lobster nerve. Arch Neurol 26:4-10, 1972 Ricker K, Hertel G, Langscheid K, et al: Myotonia not aggravated by cooling. J Neurol 2169-20, 1977 Rudel R, Keller M: Intracellular recording of myotonic muscle fibers, in Bradley WG (ed): Recent Advances in Myology. Amsterdam, Excerpta Medica, 1974, ex 89, no. 360, pp 441-445 Rudel R, Senges J: Mammalian skeletal muscle: reduced chlo-
29.
30. 31.
32.
ride conductance in drug-induced myotonia and induction of myotonia by low chloride solution. Naunyn Schmiedebergs Arch Pharmacol274:337-347, 1972 Rudel R, Senges J: Experimental myotonia, in Desmedt JE (ed): New Developments in Electromyography and Clinical Neurophysiology. Basel, Karger, 1973, vol 1, pp 483-491 Samaha RJ, Schroder JM, Rebieg J, et al: Studies on myotonia. Arch Neurol 17:22-33, 1967 Senges J, Rudel R Experimental myotonia in mammalian skeletal muscle. Pfluegers Arch 331:315-323, 1972 Ward MR, Thesleff S: The temperature dependence of action potentials in rat skeletal muscle fibers. Acta Physiol Scand 91~574-576, 1974
Furman and Barchi: Aromatic Acids and Myotonia 365
