Sodium current and membrane potential in EDL muscle fibers from normal and dystrophic C. MATHES,
F. BEZANILLA,
AND
Ahmanson Neurobiology Laboratory Los Angeles, California 90024
R. E. WEISS
and Department
MATHES, C., F. BEZANILLA, AND R. E. WEISS. Sodium current and membrane potential in EDL muscle fibers from normal and dystrophic (mdx) mice. Am. J. Physiol. 261 (Cell Physiol. 30): C718-C725, 1991.-The macroscopic and singlechannel properties of sodium currents and membrane potential were studied in intact extensor digitorum longus (EDL) muscle fibers from mdx (C57BL/lOScSn-mdx) and normal (C57BL/ 1OSnJ) mice. The voltage dependence of activation and inactivation were determined and the associated gating charges were calculated to determine if the lack of dystrophin associated with the mdx condition has any influence on sodium channels either directly or by effects on the membrane environment of the channel. Sodium currents were recorded from cell-attached patches on EDL muscle fibers isolated by collagenase treatment and manual dissection. Both macroscopic and single-channel currents were studied. We found no apparent difference in the sodium channel properties from the two types of muscle. In addition, microelectrode measurements in both mdx and normal muscle fibers indicated similar resting membrane potentials ( Vm around -95 mV), which suggests that the normal behavior of sodium channels in the muscle sarcolemma is unaffected by the X-linked gene defect. mdx muscle; sodium
channel
gating
DUCHENNE MUSCULAR DYSTROPHY (DMD) in humans and mdx (C57BL/lOScSn-mdx) dystrophy in mice are charaterized by the absence of dystrophin in muscle and brain tissue (6, 13, 14) resulting from a single, X-linked gene defect (21). Muscle degeneration occurs in both DMD and mdx conditions with the difference that regenerative processes in mdx mice are sufficient to maintain muscle function and the viability of the animal. It has been proposed that the muscle degeneration results from increased protein degradation as a consequence of elevated intracellular calcium concentrations (25). In muscle, dystrophin exists in large, oligomeric complexes that include integral membrane proteins (7), which suggests that the sarcolemma may be involved in the dystrophic condition. Recent studies of muscle and cultured myotubes from mdx mice suggest that the electrical behavior of the sarcolemma may be an important factor. Based on patch-clamp studies of myotubes from normal and mdx mice, the source of the elevated intracellular calcium concentration has been proposed to be calcium leak channels (8) or stretch-inactivated calcium channels (9) in the cell surface membrane. In myotubes in culture derived from human normal and DMD muscle fibers, C718
0363-6143/91
$1.50
Copyright
(mdx) mice
of Physiology,
University
of California,
there are conflicting reports of differences in sodium current inactivation and the existence of calcium channels (5, 24). However, because myotubes and adult muscle fibers express different forms of sodium and calcium channels, the electrophysiological properties of intact muscle fibers cannot be extrapolated easily from the cultured cell systems. Based on findings reported from intact muscle preparations, a second hypothesis to account for the increased intracellular calcium concentrations is that spontaneous depolarizations admit calcium via conventional calcium channels. This hypothesis is supported by reports that diaphragm muscle from mdx mice is chronically depolarized by 10 mV or greater relative to normal mice (15) and that the myotonic bursts that occur in mdx muscle can be abolished by agents that block sodium channels (e.g., tetrodotoxin; Ref. 16). We have investigated the macroscopic and singlechannel properties of sodium currents and membrane potential in intact normal and mdx mouse muscle. The voltage sensitivity of activation and inactivation were measured and the associated gating charges were calculated to determine if the lack of dystrophin can be interpreted as having any influence on sodium channels either directly or by effects on the membrane environment of the channel. Our findings suggest that both membrane potential and functional sodium channel properties are unaffected by the X-linked gene defect. A preliminary report of these findings has been previously published (17). MATERIALS
AND
METHODS
Muscle preparation. All experiments were performed on extensor digitorum longus (EDL) muscles and singlemuscle fibers from normal (C57BL/lOSnJ; Jackson Labs, ME) and mdx (Jackson Labs) mice. Adult mice were killed by pentobarbitol overdose; miee younger than 4 wk were killed by cervical dislocation. For patch-clamp studies, muscles were dissected from the leg with their tendon attachments intact and pinned to the floor of a shallow dish containing low-calcium dissection saline of the following composition (in mM): 142 NaCl, 2 KCl, 0.5 CaC12, 2 MgClz, 10 N-2-hydroxyethylpiperazine-N’-2 ethanesulfonic acid (HEPES), and 5 glucose at pH 7.3. After cleaning the muscles of surface connective tissue, the bathing solution was exchanged for low-calcium dissection solution with added 1.0 mg/ml collagenase
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
NA
CURRENTS
IN
EDL
FIBERS
(Sigma type IA) and 2.0 mg/ml bovine serum albumin, which was agitated for 1 h with a gentle stream of oxygen at 30°C. After the collagenase treatment, single fibers could be separated from the muscle. Single fibers or undamaged segments of single fibers were then transferred to the experimental chamber that contained a relaxing solution composed of (in mM) 142 KCl, 10 NaCl, 0.2 CaCIZ, 0.5 ethylene glycol-bis(P-aminoethyl ether)N&N’&‘-tetraacetic acid (EGTA), 2 MgC12, 10 HEPES, and 5 glucose at pH 7.3. In the majority of experiments, potassium currents were blocked either by adding barium (2-5 mM) to the relaxing solution or by substituting CsCl for KCl. Fibers adhered to the chamber floor after several minutes, and the bathing solution was exchanged several times with fresh solution before patch-clamp experiments were begun. Fibers remained relaxed throughout the experimental period. Resting membrane potential measurement. Intracellular recordings of membrane potentials were obtained by impaling fibers of whole, freshly dissected EDL muscles bathed in a physiological saline solution containing (in mM) 110 NaCl, 30 Na sulfate, 4 KCl, 2 CaC12, 1 MgCl,, 10 HEPES, and 5 glucose at pH 7.4. Microelectrodes were pulled from lB150F-4 glass capillary tubing (World Precision Instruments) and filled with 3 M KCl. Microelectrode tip resistances were -30 MQ. Patch-clamp procedures. The patch-clamp technique followed the methods published by Hamill et al. (10). Electrodes were made from 7052 glass pipettes (Garner Glass) pulled to 3-10 MQ and filled with a normal saline solution containing (in mM) 142 NaCl, 2 KCl, 2 CaC12, 1 MgCl,, 10 HEPES, and 5 glucose at pH 7.3. Seals were reliably obtained under these conditions. Typical seal resistances were X0 GQ. Unless otherwise stated in the text, recordings were obtained from cell-attached patches on the intact sarcolemma of muscle fibers. The experimental chamber was maintained at 5 or 15°C with a Peltier device under feedback control. Data acquisition. Voltage stimuli were applied and data acquired using AT-type computer-controlled interface described previously (23). Currents were low-pass filtered with a 6-pole Bessel filter. Macroscopic currents were acquired with a bandwidth of 7-10 kHz at a sampling rate of 10 &point. Macroscopic currents were typically recorded as the average of 4-10 individual traces. Leak and capacity currents subtraction was performed on-line using the P/4 or P/-4 procedure (4). Recordings used for mean and variance analysis were acquired without on-line subtraction, and the leak and capacity currents were obtained from a series of small (20 mV) subtracting pulses in the negative voltage range. Single-channel events were recorded with a bandwidth of 5-7 kHz at 20 ps/point, also without leak and capacity subtraction. Data were stored in magnetic media for later analysis. Current amplitude analysis. Macroscopic and singlechannel amplitudes were obtained with a manually controlled computer program. Cursor bars were positioned over the baseline and peak current of each trace individually and the difference was read in picoamperes. Macroscopic current analysis. Exponential fits of the declining phase of sodium currents were obtained with
FROM
lMDX
AND
NORMAL
c719
MICE
the simplex algorithm. The normalized conductance was determined using methods described previously (23). Briefly, peak current was determined as a function of potential, and the current-voltage (I- V) relation was constructed. In the same patch, the instantaneous I-V relation was also measured. The peak conductance vs. voltage curve was obtained by dividing the current for each membrane potential of the I-V curve by the corresponding current from the instantaneous I-V curve and normalizing the resulting values to a maximum of one. Currents were not corrected for junction potentials. Estimation of the gating charge. The gating charge correlated with channel activation was estimated under the assumption that the open state is preceded by one or more closed states. In this case, the equation relating the normalized conductance to the voltage contains a number of exponential terms in the denominator, each expressing the valence of the individual transitions between the states. In the limit when the voltage is very negative, the conductance follows the equation
(0
g=
where g is the normalized conductance, a is a constant, z is the total gating charge, V is the potential, and F, R, and T are the Faraday constant, the gas constant, and the absolute temperature, respectively (2). The value of z, the effective gating charge that would move through the entire range of the membrane electric field, was estimated by plotting the log of the normalized conductance as a function of voltage and fitting a straight line with slope zF/RT to the values at the most negative potentials. Inactivation parameters. Sodium current steady-state inactivation was determined by preceding the test pulse (usually to -20 or -30 mV) with a variable 50-ms prepulse to potential V. The peak current during the test pulse (1& was normalized by the maximum peak current of the group (I Namax), plotted against V, and fitted to a Boltzmann equation (11, 12) 1
h oo- -
-
I
-
I Na,max
F(V--
vh)
1
(2)
-I
where h, is the steady-state inactivation, zh is the apparent charge of inactivation, and vh is the half-point of the inactivation. Mean and variance analysis. The mean and variance analysis followed the procedures described by Sigworth (22).The mean and variance were computed from a large number of individual trials that were recorded without on-line leak and capacity subtraction. The means of the subtracting pulses obtained during the same run were appropriately scaled to subtract leak and capacity currents from the mean of the test pulses. Means were plotted against variance for individual values of test voltage and fitted to the equation 2 g2 = i.1 - -I N
(3)
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
C720
NA CURRENTS
IN
EDL
FIBERS
where c2 is the measured variance and I is the measured mean current. The parameters determined by the fitting routine were the single-channel current (i) and the number of channels in the patch (N). Once these values were fitted, the peak open probability (PO) was calculated as P
0=
I Na ibN
FROM
lMDX
AND
NORMAL
A
(4)
where INa is the mean peak current. Our analysis of mean and variance assumed a single homogeneous population of sodium channels. This assumption was based primarily on single-channel recordings (presented below) in which only one distribution of amplitudes of singlechannel currents was observed at each membrane potential. Our analysis of macroscopic current properties in normal and mdx fibers also indicated a single class of (TTX-sensitive) sodium channels because the voltage dependence of gating was not altered in m& fibers. If a TTX-resistant current had comprised a fraction of the total sodium current greater than a few percent in rndx fibers, a detectable shift in the voltage dependence of activation and inactivation would be expected (19, 30).
1 ms B
T
20 + -80
-40
l
a
80
-
0
To assess the physiological behavior of ion channels, determination of the resting membrane potential of the cells is essential. Muscle fibers that are chronically depolarized are likely to exhibit spontaneous electrical activity and altered intracellular ion concentrations, both of which have been observed in mdx mouse muscle (15, 16). In hemidiaphragm preparations from mdx and normal mice, previous measurements of the resting membrane potential suggest that mdx muscle is depolarized by 9-17 mV from a control value of -87 mV (15). Our measurements of membrane potential were made in EDL muscle fibers, and our results differ qualitatively from those found in the diaphragm. We found no significant difference between m& and normal fibers with mean values of -95 t 6 (SD) mV (n = 7) and -98 t 5 mV (n = 14) for mdx and normal muscle, respectively. Other than muscle type (hemidiaphragm vs. EDL), the only difference between our measurements and those of Kishi et al. (15) was the substitution of 30 mM sulfate for chloride and HEPES for phosphate and bicarbonate buffer in our bathing solution. The reduction of total chloride to physiological levels in our experiments may explain at least partly the more negative membrane potential recordings. Other considerations are given in DISCUSSION. Macroscopic sodium current. Macroscopic sodium currents were measured in cell-attached patches from a holding potential of -100 mV. A family of currents recorded from a mdx muscle fiber in response to test pulses ranging from -80 to 20 mV is shown in Fig. IA. Each test pulse was preceded by a 50-ms prepulse to -140 mV to remove fast inactivation. Figure 1B shows an I-V relation from an excised, inside-out patch from a mdx fiber. Inward current was first observed at a membrane potential ( Vm) of -60 mV and reached maximal amplitudes at -20 mV. The current reversed very near
40
I
-
RESULTS
Membrane
MICE
potential.
a 60
000 I Na
(PA)
C 1
0.01 -80
-40
-60 vm
-20
(mv>
1. A: family
of macroscopic currents recorded from a cellattached patch on a mdx muscle fiber in cesium-relaxing solution at 15°C. Test pulses (8 ms each) from -80 mV to 20 mV were preceded potential was -100 mV. by a 50-ms prepulse to -140 mV. Holding Each trace is the average of nine sweeps. B: current-voltage (I-V) relation from an inside-out, excised patch from a mdx muscle fiber recorded at 15°C in potassium relaxing solution (bath solution). Peak current amplitudes were measured from records obtained with pulse protocols as described in A. C: semilogarithmic plot of normalized conductance vs. membrane potential. Normalized conductance was determined according to method of Stimers et al. (Ref. 23; see METHODS). Straight line, linear regression fit to 4 points at most negative membrane potentials. Activation gating charge (z) is slope of line multiplied by factor RT/F. FIG.
to the calculated reversal potential for sodium, ENa = 48 mV, for which a potassium-sodium permeability ratio for the sodium channel of 0.075 was assumed. When cesium was substituted for potassium in the bathing medium, the reversal potential of the current shifted by -15 mV
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
NA
CURRENTS
IN
EDL
FIBERS
(data not shown), which agrees well with the predicted shift due to the difference between cesium and potassium permeabilities for the sodium channel. Measurement of reversal potential in cell-attached patches suggests that the internal sodium concentration was -3.5 mM in the presence of 10 mM external sodium and assuming intracellular potassium was 145 mM. Except in the vicinity of ENa, the I-V relations for cell-attached and excised patches were essentially identical. Peak sodium currents measured in 20 normal and 16 rn& muscle patches varied from 49 to 331 PA/patch and 27 to 204 PA/patch, respectively. The mean peak sodium currents per patch were 118 t 71 (SD) pA for normal and 134 t 45 pA for mdx muscle, which are not significantly different. Patches were obtained far from end plate regions of the muscle fibers to minimize sodium current density variation (29) and to avoid synaptic hot spots of sodium current (3, 20). Our results suggest that the mdx condition does not result in a dramatic reduction in sodium channel density, although our method of sampling is not optimal and our number of samples is too few for channel density studies. Sodium current actiuation. We were interested to know whether the effective valence responsible for the voltage dependence or the actual voltage dependence was affected by the dystrophic condition. We selected the limiting slope method to estimate the charge involved in the channel activation process. This method requires only that the open state be preceded by one or more closed states and is independent of any particular kinetic model that might be selected to test the data. The estimations of the parameters were obtained with the normalized conductance procedure or by direct estimation of the peak open probability with mean and variance analysis (see MATERIALS
AND METHODS).
Figure 1C shows a typical normalized conductance vs. voltage curve in semilogarithmic scale with a fit to the values at the most negative potentials. From this type of plot, the effective valence z and the midpoint of the conductance vs. voltage curve were obtained. Table 1 shows a summary of the estimated parameters for several patches from normal and rn& fibers. There were no significant differences in the activation parameters between dystrophic and normal fiber patches. Figure 2A shows an example of mean vs. variance plots for four different depolarizations of a patch from an m& muscle fiber. The fits to these types of data gave an estimate of the number of channels and the singlechannel current. Figure 2B shows the voltage dependence of the peak probability of being open (PO) and the fit to the values at most negative potentials to estimate the 1. Mean estimated gating charges and voltage mid points
TABLE
Activation z
Control Mdx Values
3.821.0 4.5t0.5
are means
VI,*, mV -44*4 -45t4
Inactivation n
z
18 11
4.0t0.7 3.9t0.6
t SD; n, no. of samples.
Vh, mV -92t4 -88t4
n 19 15
FROM
lMDX
AND
NORMAL
MICE
C721
gating charge. The mean estimated gating charge, z, from three mdx fiber patches was 3.9 t 0.8 (SD) and the mean half-activation potential ( V& was -46 t 3 mV (SD). These values are similar to the activation parameters estimated with the normalized conductance method shown in Table 1. In four mean and variance experiments, the number of channels in the patch ranged between 20 and 507 and the peak probability at -30 mV was -0.24. A few experiments were conducted on muscles from mice younger than 21 days. The activation parameters from one such patch from an mdx muscle were z = 3.4 and Vliz = -40 mV, while for a normal muscle patch they were z = 3.0 and Vl,z = -41 mV. These values are within the range measured for 4-wk and older mice but have not been included in the group means. Sodium current inactivation. Inactivation parameters were also examined in normal and mdx muscles. Figure 3A shows the h, relation for both types of fibers, and the results for many patches are summarized in Table 1. Similar to the activation properties, there were no significant differences in sodium channel inactivation between patches from normal and dystrophic muscle fibers. When considered in conjunction with the measurements of resting membrane potential, the midpoints of both h, relations suggest that slightly more than 50% of the sodium channels may be activated under physiological conditions. The time constants of inactivation during depolarizing pulses to several potentials were also examined and are shown in Fig. 3B. Inactivation kinetics appear nearly identical in patches from both normal and mdx muscles. The similarity of sodium current activation and inactivation properties in normal and mdx fibers suggests that the sodium channels in mdx fibers are unaltered by the X-linked gene defect characteristic of the mdx dystrophic condition. These results also suggest that TTX-resistant channels, which are expressed in denervated muscle, are not measurably present in the macroscopic currents of mdx mouse muscle. In preparations where TTX-sensitive and TTX-resistant sodium channels coexist in functional states, the macroscopic voltage dependence of both activation and inactivation is altered compared with when just one type of sodium channel is functionally present (19, 30). Single-channel recording. Our objectives in recording single-channel currents were to verify two results obtained from the mean and variance analysis of macroscopic sodium currents. We wanted first to compare the single-channel current estimates with actual singlechannel current recordings. Selected single-channel currents recorded at negative membrane potentials from a cell-attached patch on a 4-wk-old mdx muscle fiber are shown in Fig. 4A. Because of the high density of sodium channels per patch, test pulses were usually preceded by a 50-ms depolarizing prepulse to inactivate the majority of the channels. Figure 4B shows single-channel I- Vplots constructed from the mean current amplitudes measured from a muscle fiber from a 5-wk-old normal mouse. With 2 mM calcium in the patch pipette, the maximum singlechannel slope conductance (filled circles) was ys = 15.5 pS. Single-channel current amplitudes from mdx and
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
C722
NA CURRENTS
IN
EDL
FIBERS
FROM
lMDX
V =-55
i=
-
50-
42
100 Current
NORMAL
MICE
mV
-1.58
V =-35 i=
Mean No
AND
pA
mV -1.18
pA
200 (-PA)
FIG. 2. A: current means plotted against variance for four different membrane potentials (V). Solid lines, fits of Eq. 3 to data. Estimated single-channel currents (i) are shown in each panel. B: maximum open probabilities (PO) obtained from mean and variance analysis are plotted against membrane potential ( Vm) in semi-log form. Straight line, linear regression fit to 3 points at most negative membrane potentials. z is slope of line multiplied by factor RT/ F. Midpoints of activation were determined by interpolation. In example shown, z = 4.6 and IQ2 = -43 mV. Peak open probability was P, = 0.27 at -35 mV.
normal muscle patches were not detectably different. The mean slope conductances estimated from all singlechannel amplitude data were 14.7 t 2.3 pS (n = 9) at 15°C and 11.8 t 2.9 pS (n = 4) at 5OC. Also shown in Fig. 4B are the estimates of single-channel current obtained from mean and variance analysis of macroscopic currents (open triangles). The single-channel current amplitudes obtained from the two methods were very similar. The two sets of single-channel amplitudes were B
also compared with the Wilcoxon matched-pairs signedranks test, which supported the null hypothesis that the two sets were not different at the 0.05level. Our second objective in recording single-channel events from mdx fibers was to look for evidence of more than one type of sodium channel. There was no indication in the single-channel recordings of the smaller conductance TTX-resistant sodium channel seen in developing and denervated mammalian skeletal muscle (27).
4 T
3 _
f P
-
i 0
1
4Z. A*
0
I -60
I
I
-40
Ada t
I
-20
J
FIG. 3. Inactivation properties of sodium channels in mdx (circles) and norma1 (triangles) muscle fibers. A: steadystate inactivation of sodium current determined by variation of prepulse potential as described in MATERIALS AND METHODS. Each daba point represents mean of 15 (mdx) and 19 (normal) experiments. Values of gating charge (zh) and midpoint of inactivation (Vh) are given in Table 1. B: inactivation time constants (7h) are plotted against membrane potential. Values of 7h were obtained from single-exponential fits of decay phase of sodium current during test pulses. Each data point represents mean of 8 (mdx) and 7 (normal) experiments.
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
NA
CURRENTS
IN EDL FIBERS
FROM lMDX AND NORMAL
MICE
C723
This result is similar to that previously shown for TTXsensitive sodium channels and in contrast to the very slight effect of calcium on TTX-resistant sodium channels in cultured rat myotubes (28). DISCUSSION
-40
mV
-20 mV
I2 PA B -80
-60
-40
-20
2
QP P
12 mM Ca
- -1.5
1 TL 7
-0.5
2 mM Ca
1 (PAI
FIG. 4. A: single-channel records (selected) at 3 test potentials from a cell-attached patch on a 4-wk-old mdx muscle fiber. Each test pulse was 18 ms in duration. Mean single-channel currents were -1.65, -1.26, and -0.94 pA, at -70, -40, and -20 mV, respectively. Single-channel conductance for this patch was ys = 16 pS. These records were obtained at a bandwidth of 3 kHz. B: single-channel I-V relations from cellattached patches from a normal EDL fiber with 2 mM (filled circles) and 12 mM (open circles) calcium in pipette solution. Data points represent mean & SD values. Single-channel current estimates (open triangles) obtained from mean and variance analysis of macroscopic currents recorded with 2 mM calcium in pipette solution are also plotted for comparison. Data points represent mean t SD values.
Although the single-channel measurements in our study are not conclusive in eliminating the possibility that TTX-resistant sodium channels exist in mdx mouse muscle, they are consistent with the results of the mean and variance experiments and the observed similarity of macroscopic current activation and inactivation in normal and mdx fibers. The effect of extracellular calcium on the single-channel recordings was also examined. As shown in Fig. 4B, with 12 mM calcium in the pipette solution, singlechannel currents (open circles) were markedly depressed.
The results presented in this study establish that the lack of dystrophin associated with the mdx condition in mice has no discernible effects on the normal resting potential of EDL muscle fibers in vivo or on the voltagesensitive characteristics of sodium channels. Our resting potential measurements are qualitatively different from results reported previously for intact hemidiaphragm muscle (15, 16) and we can identify several possible explanations. In the work of Kishi et al. (E), normal muscle fibers had a mean resting potential of -87 mV compared with -98 mV for EDL fibers in our study. The difference may be related simply to the muscle preparations used. The difference becomes more pronounced when the measurements from mdx fibers are considered because EDL fibers were unchanged, Vm = -95 mV, while hemidiaphragm fibers were depolarized significantly, Vm = -78 to -70 mV. If the difference is related to muscle type, the effect of the lack of dystrophin may be on chloride transport (channels, exchangers, etc.). Kishi et al. (15) used Tyrode solution with a chloride concentration -30 mM greater than is normal in serum. This would tend to depolarize the fibers slightly and may explain the difference in the normal resting potential values. However, the difference between the resting potentials in mdx EDL and hemidiaphragm fibers should also be -10 mV, unless the mdx hemidiaphragm fibers are even less able to cope with a slight chloride overload. This possibility could be tested by measuring the resting potentials of normal and mdx hemidiaphragm fibers in salines with different chloride concentrations. Alternatively, fiber damage during isolation may contribute to low resting potential values, as was most likely the case in a study in which the mean resting potential of the normal fibers was reported as -64 mV (18). Our findings suggest that muscle depolarization is not a chronic symptom of the mdx condition, although it must certainly occur in the final stages of muscle cell death in DMD in humans. The lack of functional effects of the mdx condition on the voltage-sensitive properties of sodium channels suggests that, at the resting potentials recorded, sodium channels are unlikely to open spontaneously. We also conclude, although with less certainty, that TTX-resistant sodium channels are not expressed in EDL muscles of mdx mice. This contention is based on our results from mean and variance analysis of macroscopic currents, comparison of the activation and inactivation of macroscopic currents from normal and mdx fibers, and a limited number of single-channel recordings. TTXresistant sodium channels have a voltage dependence that is shifted toward more negative voltages (19,27,30), and their presence has been postulated to account for spontaneous depolarizations in denervated muscle and developing myotubes (26, 27). The implications of these
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
C724
NA
CURRENTS
IN
EDL
FIBERS
results are that sodium channel-triggered myotonic bursts observed in mdx hemidiaphragm muscles in vitro (15, 16) may be conditional on the experimental conditions. In addition, the combination of an apparent lack of TTX-resistant sodium channels and very negative resting membrane potentials in mdx muscle suggests that calcium entry through conventional voltage-gated calcium channels is unlikely to occur. Our results are also relevant toward understanding inactivation of sodium channels in normal mammalian muscle under physiological conditions. The midpoint of inactivation of sodium current varies greatly in the abundant literature. In this study, the mean normal fiber Vh (-95 mV) was more negative than previously reported values. At least part of the discrepancy may be explained by technical differences in the voltage-clamp methods used. The voltage control of membrane patches with lowcurrent densities by the patch-clamp technique is much improved over older methods that attempted to control entire sections of fibers containing complex transverse tubular systems. Thus the lower value of Vh (-70 mV) obtained with the three microelectrode voltage clamps at 14°C in one of the earliest studies (1) may be related to voltage control problems. Voltage pulse protocols do not explain the differences in the reported values. Measurements of rat EDL fiber sections with the Vaseline gap voltage-clamp technique, during which the membrane potential was verified with a microelectrode, obtained a value of Vh = -90 mV (19), which is similar to our findings. Taken with our measurements of the mean resting membrane potential, we conclude that slightly less than half of the sodium channels are inactivated in a mammalian skeletal muscle fiber at rest. It is interesting that in many nerve preparations, where the resting membrane potentials are much less negative than in muscle, Vh is shifted to more positive potentials to give similar ratios of inactivated-to-total sodium channels. The earliest demonstrations in nerve may be found in the original voltage-clamp studies of squid giant axons by Hodgkin and Huxley (11, 12). Perhaps this large pool of inactivated sodium channels provides a safety factor for recovery from periods of intense muscle activity. This work was supported by grants from the Muscular Dystrophy Association and the American Heart Association to R. E. Weiss and from the Muscular Dystrophy Association and the National Institute of General Medical Sciences (GM-30376) to F. Bezanilla. Address for reprint requests: R. E. Weiss, UCLA School of Medicine, Jerry Lewis Center, 700 Westwood Plaza, Los Angeles, CA 900241770. Received
23 May
1991;
accepted
in final
form
25 July
1991.
REFERENCES 1. ADRIAN, R. H., AND M. W. MARSHALL. Sodium currents in mammalian muscle. J. Physiol. Lond. 268: 223-250, 1977. 2. ALMERS, W. Gating currents and charge movements in excitable membranes. Rev. Physiol. Biochem. Pharmacol. 82: 96-190,1978. 3. BEAM, K. G., J. H. CALDWELL, AND D. T. CAMPBELL. Na channels in skeletal muscle concentrated near the neuromuscular junction. Nature Lond. 313: 588~590,1985. 4. BEZANILLA, F., AND C. ARMSTRONG. Inactivation of the sodium channel. I. Sodium current experiments. J. Gen. Physiol. 70: 549566.1977.
FROM
lMDX
AND
NORMAL
MICE
5. BKAILY, G., G. JASMIN, C. TAUTU, L. PROCHEK, T. YAMAMOTO, A. SCULPTOREANU, M. PEYROW, AND D. JACQUES. A tetrodotoxinand Mn2’-insensitive Na’ current in Duchenne muscular dystrophy. Muscle Nerve 13: 939-948, 1990. 6. CHAMBERLAIN, J. S., J. A. PEARLMAN, D. M. MUZNY, R. A. GIBBS, J. E. RANIER, A. A. REEVES, AND C. T. CASKEY. Expression of murine Duchenne muscular dystrophy gene in muscle and brain. Science Wash. DC 239: 1416-1418,1988. 7. ERVASTI, J. M., K. OHLENDIECK, S. D. KAHL, M. G. GAVER, AND K. P. CAMPBELL. Deficiency of a glycoprotein component of the dystrophin complex in dystrophic muscle. Nature Lond. 345: 315319,199o. 8. FONG, P., P. R. TURNER, W. F. DENETCLAW, AND R. A. STEINHARDT. Increased activity of calcium leak channels in myotubes of Duchenne human and mdx mouse origin. Science Wash. DC 250: 673-676,199O. 9. FRANCO, A., AND J. B. LANSMAN. Calcium entry through stretchinactivated ion channels in mdx myotubes. Nuture Lond. 344: 670673,199O. 10. HAMILL, 0. P., A. MARTY, E. NEHER, B. SAKMANN, AND F. G. SIGWORTH. Improved patch-clamp techniques for high-resolution current recordings from cells and cell-free membrane patches. Pfluegers Arch. 391: 85-100, 1981. 11. HODGKIN, A. L., AND A. F. HUXLEY. The dual effect of membrane potential on sodium conductance in the giant axon of Loligo. J. Physiol. Lond. 116: 497-506, 1952. 12. HODGKIN, A. L., AND A. F. HUXLEY. A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. Lond. 117: 500-544, 1952. 13. HOFFMAN, E. P., R. H. BROWN, JR., AND L. M. KUNKEL. Dystrophin: the protein product of the Duchenne muscular dystrophy locus. Cell 51: 919-928, 1987. 14. HOFFMAN, E. P., C. M. KUDSON, K. P. CAMPBELL, AND L. M. KUNKEL. Subcellular fractionation of dystrophin to the triads of skeletal muscle. Nature Lond. 330: 754-758, 1987. 15. KISHI, M., T. KURIHARA, T. HIDAKA, AND M. KINOSHITA. The stabilizing effect of bestatin on the resting membrane potentials of cross-linked muscular dystrophy mice. Jpn. J. Psychiatry Neurol. 44: 595-600,199O. 16. KISHI, M., T. KURIHARA, H. UCHIDA, AND M. KINOSHITA. Intracellular recording of myotonia in mdx mouse and the effect of Ca antagonist in myotonia. Clin. Neurol. 29: 563-567, 1989. 17. MATHES, C., R. E. WEISS, AND F. BEZANILLA. Sodium currents in skeletal muscle fibers of mdx and control mice (Abstract). Biophys. J. 59: 201a, 1991. 18. NAGEL, A., F. LEHMANN-HORN, AND A. G. ENGEL. Neuromuscular transmission in the mdx mouse. Muscle Nerve 13: 742-749, 1990. 19. PAPPONE, P. A. Voltage-clamp experiments in normal and denervated mammalian skeletal muscle fibres. J. Physiol. Lond. 306: 377-410,198O. 20. ROBERTS, W. M. Sodium channels near end-plates and nuclei of snake skeletal muscle. J. Physiol. Lond. 388: 213-232, 1987. 21. SICINSKI, P., Y. GENG, A. S. RYDER-COOK, E. A. BARNARD, M. G. DARLISON, AND P. J. BARNARD. The molecular basis of muscular dystrophy in the mdx mouse: a point mutation. Science Wash. DC 244: 1578-1580,1989. 22. SIGWORTH, F. J. The variance of sodium current fluctuations at the node of Ranvier. J. Physiol. Lond. 307: 97-129, 1980. 23. STIMERS, J. R., F. BEZANILLA, AND R. E. TAYLOR. Sodium channel activation in the squid giant axon. J. Gen. Physiol. 85: 65-82, 1985. 24. TRAUTMANN, A., C. DELAPORTE, AND A. MARTY. Voltage-dependent channels of human muscle cultures. Pfluegbs Arch. 406: 163172,1986. 25. TURNER, P. R., T. WESTWOOD, C. R. REGEN, AND R. A. STEINHARDT. Increased protein degradation results from elevated free calcium levels found in muscle from mdx mice. Nature Lond. 335: 735-738,1988. 26. WEISS, R. E. Different types of sodium channels coexist in mammalian striated muscle. In: Transduction in Biological Systems. New York: Plenum, 1990, p. 261-274. 27. WEISS, R. E., AND R. HORN. Functional differences between two classes of sodium channels in developing rat skeletal muscle. Science Wash. DC 233: 361-364,1986. 38 WEISS, R. E., AND R. HORN. Single-channel studies of TTX-
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
NA CURRENTS
IN EDL FIBERS
sensitive and TTX-resistant sodium channels in developing rat muscle reveal different open channel properties. Ann. NY Acad. Sci. 479: 152-161,1986. 29. WEISS, R. E., W. M. ROBERTS,
Mobility
of voltage-dependent
W. ST~HMER, AND W. ALMERS. ion channels and lectin receptors in
FROM lMDX AND NORMAL
MICE
the sarcolemma of frog skeletal muscle. J. Gen. Physiol.
C725 87: 955-
983,1986. 30. WEISS,
R. E., AND N. SIDELL. Sodium currents during differentiation in a human neuroblastoma cell line. J. Gen. Physiol. 97: 521539,199l.
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
F. BEZANILLA,
AND
Ahmanson Neurobiology Laboratory Los Angeles, California 90024
R. E. WEISS
and Department
MATHES, C., F. BEZANILLA, AND R. E. WEISS. Sodium current and membrane potential in EDL muscle fibers from normal and dystrophic (mdx) mice. Am. J. Physiol. 261 (Cell Physiol. 30): C718-C725, 1991.-The macroscopic and singlechannel properties of sodium currents and membrane potential were studied in intact extensor digitorum longus (EDL) muscle fibers from mdx (C57BL/lOScSn-mdx) and normal (C57BL/ 1OSnJ) mice. The voltage dependence of activation and inactivation were determined and the associated gating charges were calculated to determine if the lack of dystrophin associated with the mdx condition has any influence on sodium channels either directly or by effects on the membrane environment of the channel. Sodium currents were recorded from cell-attached patches on EDL muscle fibers isolated by collagenase treatment and manual dissection. Both macroscopic and single-channel currents were studied. We found no apparent difference in the sodium channel properties from the two types of muscle. In addition, microelectrode measurements in both mdx and normal muscle fibers indicated similar resting membrane potentials ( Vm around -95 mV), which suggests that the normal behavior of sodium channels in the muscle sarcolemma is unaffected by the X-linked gene defect. mdx muscle; sodium
channel
gating
DUCHENNE MUSCULAR DYSTROPHY (DMD) in humans and mdx (C57BL/lOScSn-mdx) dystrophy in mice are charaterized by the absence of dystrophin in muscle and brain tissue (6, 13, 14) resulting from a single, X-linked gene defect (21). Muscle degeneration occurs in both DMD and mdx conditions with the difference that regenerative processes in mdx mice are sufficient to maintain muscle function and the viability of the animal. It has been proposed that the muscle degeneration results from increased protein degradation as a consequence of elevated intracellular calcium concentrations (25). In muscle, dystrophin exists in large, oligomeric complexes that include integral membrane proteins (7), which suggests that the sarcolemma may be involved in the dystrophic condition. Recent studies of muscle and cultured myotubes from mdx mice suggest that the electrical behavior of the sarcolemma may be an important factor. Based on patch-clamp studies of myotubes from normal and mdx mice, the source of the elevated intracellular calcium concentration has been proposed to be calcium leak channels (8) or stretch-inactivated calcium channels (9) in the cell surface membrane. In myotubes in culture derived from human normal and DMD muscle fibers, C718
0363-6143/91
$1.50
Copyright
(mdx) mice
of Physiology,
University
of California,
there are conflicting reports of differences in sodium current inactivation and the existence of calcium channels (5, 24). However, because myotubes and adult muscle fibers express different forms of sodium and calcium channels, the electrophysiological properties of intact muscle fibers cannot be extrapolated easily from the cultured cell systems. Based on findings reported from intact muscle preparations, a second hypothesis to account for the increased intracellular calcium concentrations is that spontaneous depolarizations admit calcium via conventional calcium channels. This hypothesis is supported by reports that diaphragm muscle from mdx mice is chronically depolarized by 10 mV or greater relative to normal mice (15) and that the myotonic bursts that occur in mdx muscle can be abolished by agents that block sodium channels (e.g., tetrodotoxin; Ref. 16). We have investigated the macroscopic and singlechannel properties of sodium currents and membrane potential in intact normal and mdx mouse muscle. The voltage sensitivity of activation and inactivation were measured and the associated gating charges were calculated to determine if the lack of dystrophin can be interpreted as having any influence on sodium channels either directly or by effects on the membrane environment of the channel. Our findings suggest that both membrane potential and functional sodium channel properties are unaffected by the X-linked gene defect. A preliminary report of these findings has been previously published (17). MATERIALS
AND
METHODS
Muscle preparation. All experiments were performed on extensor digitorum longus (EDL) muscles and singlemuscle fibers from normal (C57BL/lOSnJ; Jackson Labs, ME) and mdx (Jackson Labs) mice. Adult mice were killed by pentobarbitol overdose; miee younger than 4 wk were killed by cervical dislocation. For patch-clamp studies, muscles were dissected from the leg with their tendon attachments intact and pinned to the floor of a shallow dish containing low-calcium dissection saline of the following composition (in mM): 142 NaCl, 2 KCl, 0.5 CaC12, 2 MgClz, 10 N-2-hydroxyethylpiperazine-N’-2 ethanesulfonic acid (HEPES), and 5 glucose at pH 7.3. After cleaning the muscles of surface connective tissue, the bathing solution was exchanged for low-calcium dissection solution with added 1.0 mg/ml collagenase
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
NA
CURRENTS
IN
EDL
FIBERS
(Sigma type IA) and 2.0 mg/ml bovine serum albumin, which was agitated for 1 h with a gentle stream of oxygen at 30°C. After the collagenase treatment, single fibers could be separated from the muscle. Single fibers or undamaged segments of single fibers were then transferred to the experimental chamber that contained a relaxing solution composed of (in mM) 142 KCl, 10 NaCl, 0.2 CaCIZ, 0.5 ethylene glycol-bis(P-aminoethyl ether)N&N’&‘-tetraacetic acid (EGTA), 2 MgC12, 10 HEPES, and 5 glucose at pH 7.3. In the majority of experiments, potassium currents were blocked either by adding barium (2-5 mM) to the relaxing solution or by substituting CsCl for KCl. Fibers adhered to the chamber floor after several minutes, and the bathing solution was exchanged several times with fresh solution before patch-clamp experiments were begun. Fibers remained relaxed throughout the experimental period. Resting membrane potential measurement. Intracellular recordings of membrane potentials were obtained by impaling fibers of whole, freshly dissected EDL muscles bathed in a physiological saline solution containing (in mM) 110 NaCl, 30 Na sulfate, 4 KCl, 2 CaC12, 1 MgCl,, 10 HEPES, and 5 glucose at pH 7.4. Microelectrodes were pulled from lB150F-4 glass capillary tubing (World Precision Instruments) and filled with 3 M KCl. Microelectrode tip resistances were -30 MQ. Patch-clamp procedures. The patch-clamp technique followed the methods published by Hamill et al. (10). Electrodes were made from 7052 glass pipettes (Garner Glass) pulled to 3-10 MQ and filled with a normal saline solution containing (in mM) 142 NaCl, 2 KCl, 2 CaC12, 1 MgCl,, 10 HEPES, and 5 glucose at pH 7.3. Seals were reliably obtained under these conditions. Typical seal resistances were X0 GQ. Unless otherwise stated in the text, recordings were obtained from cell-attached patches on the intact sarcolemma of muscle fibers. The experimental chamber was maintained at 5 or 15°C with a Peltier device under feedback control. Data acquisition. Voltage stimuli were applied and data acquired using AT-type computer-controlled interface described previously (23). Currents were low-pass filtered with a 6-pole Bessel filter. Macroscopic currents were acquired with a bandwidth of 7-10 kHz at a sampling rate of 10 &point. Macroscopic currents were typically recorded as the average of 4-10 individual traces. Leak and capacity currents subtraction was performed on-line using the P/4 or P/-4 procedure (4). Recordings used for mean and variance analysis were acquired without on-line subtraction, and the leak and capacity currents were obtained from a series of small (20 mV) subtracting pulses in the negative voltage range. Single-channel events were recorded with a bandwidth of 5-7 kHz at 20 ps/point, also without leak and capacity subtraction. Data were stored in magnetic media for later analysis. Current amplitude analysis. Macroscopic and singlechannel amplitudes were obtained with a manually controlled computer program. Cursor bars were positioned over the baseline and peak current of each trace individually and the difference was read in picoamperes. Macroscopic current analysis. Exponential fits of the declining phase of sodium currents were obtained with
FROM
lMDX
AND
NORMAL
c719
MICE
the simplex algorithm. The normalized conductance was determined using methods described previously (23). Briefly, peak current was determined as a function of potential, and the current-voltage (I- V) relation was constructed. In the same patch, the instantaneous I-V relation was also measured. The peak conductance vs. voltage curve was obtained by dividing the current for each membrane potential of the I-V curve by the corresponding current from the instantaneous I-V curve and normalizing the resulting values to a maximum of one. Currents were not corrected for junction potentials. Estimation of the gating charge. The gating charge correlated with channel activation was estimated under the assumption that the open state is preceded by one or more closed states. In this case, the equation relating the normalized conductance to the voltage contains a number of exponential terms in the denominator, each expressing the valence of the individual transitions between the states. In the limit when the voltage is very negative, the conductance follows the equation
(0
g=
where g is the normalized conductance, a is a constant, z is the total gating charge, V is the potential, and F, R, and T are the Faraday constant, the gas constant, and the absolute temperature, respectively (2). The value of z, the effective gating charge that would move through the entire range of the membrane electric field, was estimated by plotting the log of the normalized conductance as a function of voltage and fitting a straight line with slope zF/RT to the values at the most negative potentials. Inactivation parameters. Sodium current steady-state inactivation was determined by preceding the test pulse (usually to -20 or -30 mV) with a variable 50-ms prepulse to potential V. The peak current during the test pulse (1& was normalized by the maximum peak current of the group (I Namax), plotted against V, and fitted to a Boltzmann equation (11, 12) 1
h oo- -
-
I
-
I Na,max
F(V--
vh)
1
(2)
-I
where h, is the steady-state inactivation, zh is the apparent charge of inactivation, and vh is the half-point of the inactivation. Mean and variance analysis. The mean and variance analysis followed the procedures described by Sigworth (22).The mean and variance were computed from a large number of individual trials that were recorded without on-line leak and capacity subtraction. The means of the subtracting pulses obtained during the same run were appropriately scaled to subtract leak and capacity currents from the mean of the test pulses. Means were plotted against variance for individual values of test voltage and fitted to the equation 2 g2 = i.1 - -I N
(3)
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
C720
NA CURRENTS
IN
EDL
FIBERS
where c2 is the measured variance and I is the measured mean current. The parameters determined by the fitting routine were the single-channel current (i) and the number of channels in the patch (N). Once these values were fitted, the peak open probability (PO) was calculated as P
0=
I Na ibN
FROM
lMDX
AND
NORMAL
A
(4)
where INa is the mean peak current. Our analysis of mean and variance assumed a single homogeneous population of sodium channels. This assumption was based primarily on single-channel recordings (presented below) in which only one distribution of amplitudes of singlechannel currents was observed at each membrane potential. Our analysis of macroscopic current properties in normal and mdx fibers also indicated a single class of (TTX-sensitive) sodium channels because the voltage dependence of gating was not altered in m& fibers. If a TTX-resistant current had comprised a fraction of the total sodium current greater than a few percent in rndx fibers, a detectable shift in the voltage dependence of activation and inactivation would be expected (19, 30).
1 ms B
T
20 + -80
-40
l
a
80
-
0
To assess the physiological behavior of ion channels, determination of the resting membrane potential of the cells is essential. Muscle fibers that are chronically depolarized are likely to exhibit spontaneous electrical activity and altered intracellular ion concentrations, both of which have been observed in mdx mouse muscle (15, 16). In hemidiaphragm preparations from mdx and normal mice, previous measurements of the resting membrane potential suggest that mdx muscle is depolarized by 9-17 mV from a control value of -87 mV (15). Our measurements of membrane potential were made in EDL muscle fibers, and our results differ qualitatively from those found in the diaphragm. We found no significant difference between m& and normal fibers with mean values of -95 t 6 (SD) mV (n = 7) and -98 t 5 mV (n = 14) for mdx and normal muscle, respectively. Other than muscle type (hemidiaphragm vs. EDL), the only difference between our measurements and those of Kishi et al. (15) was the substitution of 30 mM sulfate for chloride and HEPES for phosphate and bicarbonate buffer in our bathing solution. The reduction of total chloride to physiological levels in our experiments may explain at least partly the more negative membrane potential recordings. Other considerations are given in DISCUSSION. Macroscopic sodium current. Macroscopic sodium currents were measured in cell-attached patches from a holding potential of -100 mV. A family of currents recorded from a mdx muscle fiber in response to test pulses ranging from -80 to 20 mV is shown in Fig. IA. Each test pulse was preceded by a 50-ms prepulse to -140 mV to remove fast inactivation. Figure 1B shows an I-V relation from an excised, inside-out patch from a mdx fiber. Inward current was first observed at a membrane potential ( Vm) of -60 mV and reached maximal amplitudes at -20 mV. The current reversed very near
40
I
-
RESULTS
Membrane
MICE
potential.
a 60
000 I Na
(PA)
C 1
0.01 -80
-40
-60 vm
-20
(mv>
1. A: family
of macroscopic currents recorded from a cellattached patch on a mdx muscle fiber in cesium-relaxing solution at 15°C. Test pulses (8 ms each) from -80 mV to 20 mV were preceded potential was -100 mV. by a 50-ms prepulse to -140 mV. Holding Each trace is the average of nine sweeps. B: current-voltage (I-V) relation from an inside-out, excised patch from a mdx muscle fiber recorded at 15°C in potassium relaxing solution (bath solution). Peak current amplitudes were measured from records obtained with pulse protocols as described in A. C: semilogarithmic plot of normalized conductance vs. membrane potential. Normalized conductance was determined according to method of Stimers et al. (Ref. 23; see METHODS). Straight line, linear regression fit to 4 points at most negative membrane potentials. Activation gating charge (z) is slope of line multiplied by factor RT/F. FIG.
to the calculated reversal potential for sodium, ENa = 48 mV, for which a potassium-sodium permeability ratio for the sodium channel of 0.075 was assumed. When cesium was substituted for potassium in the bathing medium, the reversal potential of the current shifted by -15 mV
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
NA
CURRENTS
IN
EDL
FIBERS
(data not shown), which agrees well with the predicted shift due to the difference between cesium and potassium permeabilities for the sodium channel. Measurement of reversal potential in cell-attached patches suggests that the internal sodium concentration was -3.5 mM in the presence of 10 mM external sodium and assuming intracellular potassium was 145 mM. Except in the vicinity of ENa, the I-V relations for cell-attached and excised patches were essentially identical. Peak sodium currents measured in 20 normal and 16 rn& muscle patches varied from 49 to 331 PA/patch and 27 to 204 PA/patch, respectively. The mean peak sodium currents per patch were 118 t 71 (SD) pA for normal and 134 t 45 pA for mdx muscle, which are not significantly different. Patches were obtained far from end plate regions of the muscle fibers to minimize sodium current density variation (29) and to avoid synaptic hot spots of sodium current (3, 20). Our results suggest that the mdx condition does not result in a dramatic reduction in sodium channel density, although our method of sampling is not optimal and our number of samples is too few for channel density studies. Sodium current actiuation. We were interested to know whether the effective valence responsible for the voltage dependence or the actual voltage dependence was affected by the dystrophic condition. We selected the limiting slope method to estimate the charge involved in the channel activation process. This method requires only that the open state be preceded by one or more closed states and is independent of any particular kinetic model that might be selected to test the data. The estimations of the parameters were obtained with the normalized conductance procedure or by direct estimation of the peak open probability with mean and variance analysis (see MATERIALS
AND METHODS).
Figure 1C shows a typical normalized conductance vs. voltage curve in semilogarithmic scale with a fit to the values at the most negative potentials. From this type of plot, the effective valence z and the midpoint of the conductance vs. voltage curve were obtained. Table 1 shows a summary of the estimated parameters for several patches from normal and rn& fibers. There were no significant differences in the activation parameters between dystrophic and normal fiber patches. Figure 2A shows an example of mean vs. variance plots for four different depolarizations of a patch from an m& muscle fiber. The fits to these types of data gave an estimate of the number of channels and the singlechannel current. Figure 2B shows the voltage dependence of the peak probability of being open (PO) and the fit to the values at most negative potentials to estimate the 1. Mean estimated gating charges and voltage mid points
TABLE
Activation z
Control Mdx Values
3.821.0 4.5t0.5
are means
VI,*, mV -44*4 -45t4
Inactivation n
z
18 11
4.0t0.7 3.9t0.6
t SD; n, no. of samples.
Vh, mV -92t4 -88t4
n 19 15
FROM
lMDX
AND
NORMAL
MICE
C721
gating charge. The mean estimated gating charge, z, from three mdx fiber patches was 3.9 t 0.8 (SD) and the mean half-activation potential ( V& was -46 t 3 mV (SD). These values are similar to the activation parameters estimated with the normalized conductance method shown in Table 1. In four mean and variance experiments, the number of channels in the patch ranged between 20 and 507 and the peak probability at -30 mV was -0.24. A few experiments were conducted on muscles from mice younger than 21 days. The activation parameters from one such patch from an mdx muscle were z = 3.4 and Vliz = -40 mV, while for a normal muscle patch they were z = 3.0 and Vl,z = -41 mV. These values are within the range measured for 4-wk and older mice but have not been included in the group means. Sodium current inactivation. Inactivation parameters were also examined in normal and mdx muscles. Figure 3A shows the h, relation for both types of fibers, and the results for many patches are summarized in Table 1. Similar to the activation properties, there were no significant differences in sodium channel inactivation between patches from normal and dystrophic muscle fibers. When considered in conjunction with the measurements of resting membrane potential, the midpoints of both h, relations suggest that slightly more than 50% of the sodium channels may be activated under physiological conditions. The time constants of inactivation during depolarizing pulses to several potentials were also examined and are shown in Fig. 3B. Inactivation kinetics appear nearly identical in patches from both normal and mdx muscles. The similarity of sodium current activation and inactivation properties in normal and mdx fibers suggests that the sodium channels in mdx fibers are unaltered by the X-linked gene defect characteristic of the mdx dystrophic condition. These results also suggest that TTX-resistant channels, which are expressed in denervated muscle, are not measurably present in the macroscopic currents of mdx mouse muscle. In preparations where TTX-sensitive and TTX-resistant sodium channels coexist in functional states, the macroscopic voltage dependence of both activation and inactivation is altered compared with when just one type of sodium channel is functionally present (19, 30). Single-channel recording. Our objectives in recording single-channel currents were to verify two results obtained from the mean and variance analysis of macroscopic sodium currents. We wanted first to compare the single-channel current estimates with actual singlechannel current recordings. Selected single-channel currents recorded at negative membrane potentials from a cell-attached patch on a 4-wk-old mdx muscle fiber are shown in Fig. 4A. Because of the high density of sodium channels per patch, test pulses were usually preceded by a 50-ms depolarizing prepulse to inactivate the majority of the channels. Figure 4B shows single-channel I- Vplots constructed from the mean current amplitudes measured from a muscle fiber from a 5-wk-old normal mouse. With 2 mM calcium in the patch pipette, the maximum singlechannel slope conductance (filled circles) was ys = 15.5 pS. Single-channel current amplitudes from mdx and
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
C722
NA CURRENTS
IN
EDL
FIBERS
FROM
lMDX
V =-55
i=
-
50-
42
100 Current
NORMAL
MICE
mV
-1.58
V =-35 i=
Mean No
AND
pA
mV -1.18
pA
200 (-PA)
FIG. 2. A: current means plotted against variance for four different membrane potentials (V). Solid lines, fits of Eq. 3 to data. Estimated single-channel currents (i) are shown in each panel. B: maximum open probabilities (PO) obtained from mean and variance analysis are plotted against membrane potential ( Vm) in semi-log form. Straight line, linear regression fit to 3 points at most negative membrane potentials. z is slope of line multiplied by factor RT/ F. Midpoints of activation were determined by interpolation. In example shown, z = 4.6 and IQ2 = -43 mV. Peak open probability was P, = 0.27 at -35 mV.
normal muscle patches were not detectably different. The mean slope conductances estimated from all singlechannel amplitude data were 14.7 t 2.3 pS (n = 9) at 15°C and 11.8 t 2.9 pS (n = 4) at 5OC. Also shown in Fig. 4B are the estimates of single-channel current obtained from mean and variance analysis of macroscopic currents (open triangles). The single-channel current amplitudes obtained from the two methods were very similar. The two sets of single-channel amplitudes were B
also compared with the Wilcoxon matched-pairs signedranks test, which supported the null hypothesis that the two sets were not different at the 0.05level. Our second objective in recording single-channel events from mdx fibers was to look for evidence of more than one type of sodium channel. There was no indication in the single-channel recordings of the smaller conductance TTX-resistant sodium channel seen in developing and denervated mammalian skeletal muscle (27).
4 T
3 _
f P
-
i 0
1
4Z. A*
0
I -60
I
I
-40
Ada t
I
-20
J
FIG. 3. Inactivation properties of sodium channels in mdx (circles) and norma1 (triangles) muscle fibers. A: steadystate inactivation of sodium current determined by variation of prepulse potential as described in MATERIALS AND METHODS. Each daba point represents mean of 15 (mdx) and 19 (normal) experiments. Values of gating charge (zh) and midpoint of inactivation (Vh) are given in Table 1. B: inactivation time constants (7h) are plotted against membrane potential. Values of 7h were obtained from single-exponential fits of decay phase of sodium current during test pulses. Each data point represents mean of 8 (mdx) and 7 (normal) experiments.
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
NA
CURRENTS
IN EDL FIBERS
FROM lMDX AND NORMAL
MICE
C723
This result is similar to that previously shown for TTXsensitive sodium channels and in contrast to the very slight effect of calcium on TTX-resistant sodium channels in cultured rat myotubes (28). DISCUSSION
-40
mV
-20 mV
I2 PA B -80
-60
-40
-20
2
QP P
12 mM Ca
- -1.5
1 TL 7
-0.5
2 mM Ca
1 (PAI
FIG. 4. A: single-channel records (selected) at 3 test potentials from a cell-attached patch on a 4-wk-old mdx muscle fiber. Each test pulse was 18 ms in duration. Mean single-channel currents were -1.65, -1.26, and -0.94 pA, at -70, -40, and -20 mV, respectively. Single-channel conductance for this patch was ys = 16 pS. These records were obtained at a bandwidth of 3 kHz. B: single-channel I-V relations from cellattached patches from a normal EDL fiber with 2 mM (filled circles) and 12 mM (open circles) calcium in pipette solution. Data points represent mean & SD values. Single-channel current estimates (open triangles) obtained from mean and variance analysis of macroscopic currents recorded with 2 mM calcium in pipette solution are also plotted for comparison. Data points represent mean t SD values.
Although the single-channel measurements in our study are not conclusive in eliminating the possibility that TTX-resistant sodium channels exist in mdx mouse muscle, they are consistent with the results of the mean and variance experiments and the observed similarity of macroscopic current activation and inactivation in normal and mdx fibers. The effect of extracellular calcium on the single-channel recordings was also examined. As shown in Fig. 4B, with 12 mM calcium in the pipette solution, singlechannel currents (open circles) were markedly depressed.
The results presented in this study establish that the lack of dystrophin associated with the mdx condition in mice has no discernible effects on the normal resting potential of EDL muscle fibers in vivo or on the voltagesensitive characteristics of sodium channels. Our resting potential measurements are qualitatively different from results reported previously for intact hemidiaphragm muscle (15, 16) and we can identify several possible explanations. In the work of Kishi et al. (E), normal muscle fibers had a mean resting potential of -87 mV compared with -98 mV for EDL fibers in our study. The difference may be related simply to the muscle preparations used. The difference becomes more pronounced when the measurements from mdx fibers are considered because EDL fibers were unchanged, Vm = -95 mV, while hemidiaphragm fibers were depolarized significantly, Vm = -78 to -70 mV. If the difference is related to muscle type, the effect of the lack of dystrophin may be on chloride transport (channels, exchangers, etc.). Kishi et al. (15) used Tyrode solution with a chloride concentration -30 mM greater than is normal in serum. This would tend to depolarize the fibers slightly and may explain the difference in the normal resting potential values. However, the difference between the resting potentials in mdx EDL and hemidiaphragm fibers should also be -10 mV, unless the mdx hemidiaphragm fibers are even less able to cope with a slight chloride overload. This possibility could be tested by measuring the resting potentials of normal and mdx hemidiaphragm fibers in salines with different chloride concentrations. Alternatively, fiber damage during isolation may contribute to low resting potential values, as was most likely the case in a study in which the mean resting potential of the normal fibers was reported as -64 mV (18). Our findings suggest that muscle depolarization is not a chronic symptom of the mdx condition, although it must certainly occur in the final stages of muscle cell death in DMD in humans. The lack of functional effects of the mdx condition on the voltage-sensitive properties of sodium channels suggests that, at the resting potentials recorded, sodium channels are unlikely to open spontaneously. We also conclude, although with less certainty, that TTX-resistant sodium channels are not expressed in EDL muscles of mdx mice. This contention is based on our results from mean and variance analysis of macroscopic currents, comparison of the activation and inactivation of macroscopic currents from normal and mdx fibers, and a limited number of single-channel recordings. TTXresistant sodium channels have a voltage dependence that is shifted toward more negative voltages (19,27,30), and their presence has been postulated to account for spontaneous depolarizations in denervated muscle and developing myotubes (26, 27). The implications of these
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
C724
NA
CURRENTS
IN
EDL
FIBERS
results are that sodium channel-triggered myotonic bursts observed in mdx hemidiaphragm muscles in vitro (15, 16) may be conditional on the experimental conditions. In addition, the combination of an apparent lack of TTX-resistant sodium channels and very negative resting membrane potentials in mdx muscle suggests that calcium entry through conventional voltage-gated calcium channels is unlikely to occur. Our results are also relevant toward understanding inactivation of sodium channels in normal mammalian muscle under physiological conditions. The midpoint of inactivation of sodium current varies greatly in the abundant literature. In this study, the mean normal fiber Vh (-95 mV) was more negative than previously reported values. At least part of the discrepancy may be explained by technical differences in the voltage-clamp methods used. The voltage control of membrane patches with lowcurrent densities by the patch-clamp technique is much improved over older methods that attempted to control entire sections of fibers containing complex transverse tubular systems. Thus the lower value of Vh (-70 mV) obtained with the three microelectrode voltage clamps at 14°C in one of the earliest studies (1) may be related to voltage control problems. Voltage pulse protocols do not explain the differences in the reported values. Measurements of rat EDL fiber sections with the Vaseline gap voltage-clamp technique, during which the membrane potential was verified with a microelectrode, obtained a value of Vh = -90 mV (19), which is similar to our findings. Taken with our measurements of the mean resting membrane potential, we conclude that slightly less than half of the sodium channels are inactivated in a mammalian skeletal muscle fiber at rest. It is interesting that in many nerve preparations, where the resting membrane potentials are much less negative than in muscle, Vh is shifted to more positive potentials to give similar ratios of inactivated-to-total sodium channels. The earliest demonstrations in nerve may be found in the original voltage-clamp studies of squid giant axons by Hodgkin and Huxley (11, 12). Perhaps this large pool of inactivated sodium channels provides a safety factor for recovery from periods of intense muscle activity. This work was supported by grants from the Muscular Dystrophy Association and the American Heart Association to R. E. Weiss and from the Muscular Dystrophy Association and the National Institute of General Medical Sciences (GM-30376) to F. Bezanilla. Address for reprint requests: R. E. Weiss, UCLA School of Medicine, Jerry Lewis Center, 700 Westwood Plaza, Los Angeles, CA 900241770. Received
23 May
1991;
accepted
in final
form
25 July
1991.
REFERENCES 1. ADRIAN, R. H., AND M. W. MARSHALL. Sodium currents in mammalian muscle. J. Physiol. Lond. 268: 223-250, 1977. 2. ALMERS, W. Gating currents and charge movements in excitable membranes. Rev. Physiol. Biochem. Pharmacol. 82: 96-190,1978. 3. BEAM, K. G., J. H. CALDWELL, AND D. T. CAMPBELL. Na channels in skeletal muscle concentrated near the neuromuscular junction. Nature Lond. 313: 588~590,1985. 4. BEZANILLA, F., AND C. ARMSTRONG. Inactivation of the sodium channel. I. Sodium current experiments. J. Gen. Physiol. 70: 549566.1977.
FROM
lMDX
AND
NORMAL
MICE
5. BKAILY, G., G. JASMIN, C. TAUTU, L. PROCHEK, T. YAMAMOTO, A. SCULPTOREANU, M. PEYROW, AND D. JACQUES. A tetrodotoxinand Mn2’-insensitive Na’ current in Duchenne muscular dystrophy. Muscle Nerve 13: 939-948, 1990. 6. CHAMBERLAIN, J. S., J. A. PEARLMAN, D. M. MUZNY, R. A. GIBBS, J. E. RANIER, A. A. REEVES, AND C. T. CASKEY. Expression of murine Duchenne muscular dystrophy gene in muscle and brain. Science Wash. DC 239: 1416-1418,1988. 7. ERVASTI, J. M., K. OHLENDIECK, S. D. KAHL, M. G. GAVER, AND K. P. CAMPBELL. Deficiency of a glycoprotein component of the dystrophin complex in dystrophic muscle. Nature Lond. 345: 315319,199o. 8. FONG, P., P. R. TURNER, W. F. DENETCLAW, AND R. A. STEINHARDT. Increased activity of calcium leak channels in myotubes of Duchenne human and mdx mouse origin. Science Wash. DC 250: 673-676,199O. 9. FRANCO, A., AND J. B. LANSMAN. Calcium entry through stretchinactivated ion channels in mdx myotubes. Nuture Lond. 344: 670673,199O. 10. HAMILL, 0. P., A. MARTY, E. NEHER, B. SAKMANN, AND F. G. SIGWORTH. Improved patch-clamp techniques for high-resolution current recordings from cells and cell-free membrane patches. Pfluegers Arch. 391: 85-100, 1981. 11. HODGKIN, A. L., AND A. F. HUXLEY. The dual effect of membrane potential on sodium conductance in the giant axon of Loligo. J. Physiol. Lond. 116: 497-506, 1952. 12. HODGKIN, A. L., AND A. F. HUXLEY. A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. Lond. 117: 500-544, 1952. 13. HOFFMAN, E. P., R. H. BROWN, JR., AND L. M. KUNKEL. Dystrophin: the protein product of the Duchenne muscular dystrophy locus. Cell 51: 919-928, 1987. 14. HOFFMAN, E. P., C. M. KUDSON, K. P. CAMPBELL, AND L. M. KUNKEL. Subcellular fractionation of dystrophin to the triads of skeletal muscle. Nature Lond. 330: 754-758, 1987. 15. KISHI, M., T. KURIHARA, T. HIDAKA, AND M. KINOSHITA. The stabilizing effect of bestatin on the resting membrane potentials of cross-linked muscular dystrophy mice. Jpn. J. Psychiatry Neurol. 44: 595-600,199O. 16. KISHI, M., T. KURIHARA, H. UCHIDA, AND M. KINOSHITA. Intracellular recording of myotonia in mdx mouse and the effect of Ca antagonist in myotonia. Clin. Neurol. 29: 563-567, 1989. 17. MATHES, C., R. E. WEISS, AND F. BEZANILLA. Sodium currents in skeletal muscle fibers of mdx and control mice (Abstract). Biophys. J. 59: 201a, 1991. 18. NAGEL, A., F. LEHMANN-HORN, AND A. G. ENGEL. Neuromuscular transmission in the mdx mouse. Muscle Nerve 13: 742-749, 1990. 19. PAPPONE, P. A. Voltage-clamp experiments in normal and denervated mammalian skeletal muscle fibres. J. Physiol. Lond. 306: 377-410,198O. 20. ROBERTS, W. M. Sodium channels near end-plates and nuclei of snake skeletal muscle. J. Physiol. Lond. 388: 213-232, 1987. 21. SICINSKI, P., Y. GENG, A. S. RYDER-COOK, E. A. BARNARD, M. G. DARLISON, AND P. J. BARNARD. The molecular basis of muscular dystrophy in the mdx mouse: a point mutation. Science Wash. DC 244: 1578-1580,1989. 22. SIGWORTH, F. J. The variance of sodium current fluctuations at the node of Ranvier. J. Physiol. Lond. 307: 97-129, 1980. 23. STIMERS, J. R., F. BEZANILLA, AND R. E. TAYLOR. Sodium channel activation in the squid giant axon. J. Gen. Physiol. 85: 65-82, 1985. 24. TRAUTMANN, A., C. DELAPORTE, AND A. MARTY. Voltage-dependent channels of human muscle cultures. Pfluegbs Arch. 406: 163172,1986. 25. TURNER, P. R., T. WESTWOOD, C. R. REGEN, AND R. A. STEINHARDT. Increased protein degradation results from elevated free calcium levels found in muscle from mdx mice. Nature Lond. 335: 735-738,1988. 26. WEISS, R. E. Different types of sodium channels coexist in mammalian striated muscle. In: Transduction in Biological Systems. New York: Plenum, 1990, p. 261-274. 27. WEISS, R. E., AND R. HORN. Functional differences between two classes of sodium channels in developing rat skeletal muscle. Science Wash. DC 233: 361-364,1986. 38 WEISS, R. E., AND R. HORN. Single-channel studies of TTX-
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
NA CURRENTS
IN EDL FIBERS
sensitive and TTX-resistant sodium channels in developing rat muscle reveal different open channel properties. Ann. NY Acad. Sci. 479: 152-161,1986. 29. WEISS, R. E., W. M. ROBERTS,
Mobility
of voltage-dependent
W. ST~HMER, AND W. ALMERS. ion channels and lectin receptors in
FROM lMDX AND NORMAL
MICE
the sarcolemma of frog skeletal muscle. J. Gen. Physiol.
C725 87: 955-
983,1986. 30. WEISS,
R. E., AND N. SIDELL. Sodium currents during differentiation in a human neuroblastoma cell line. J. Gen. Physiol. 97: 521539,199l.
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 23, 2018.
