Action Potentials, Afterpotentials, and Excitation-Contraction ..., Slides of Physiology

Download Action Potentials, Afterpotentials, and Excitation-Contraction ... and more Slides Physiology in PDF only on Docsity! Action Potentials, Afterpotentials, and Excitation-Contraction Coupling in Frog Sartorius Fibers without Transverse Tubules PETER W. GAGE and ROBERT S. EISENBERG From the Department of Physiology, Duke University, Durham, North Carolina 27706. Dr. Gage's present address is Department of Physiology, University of New South Wales, Kensington, Australia. Dr. Eisenberg's present address is Department of Physiology, University of California at Los Angeles, Los Angeles, Calgornia 90024 ABSTRACT In frog sartorius muscle fibers in which the transverse tubular sys- tem has been disrupted by treatment with glycerol, action potentials which are unaccompanied by twitches can be recorded. These action potentials appear to be the same as those recorded in normal fibers except that the early afterpo- tential usually consists of a small hyperpolarization of short duration. After a train of action potentials no late afterpotential is seen even when the mem- brane potential is changed from the resting level. In fibers without transverse tubules hyperpolarizing currents do not produce a creep in potential. The interruption of excitation-contraction coupling, the changes in the afterpoten- rials, and the disappearance of creep are all attributed to the lack of a transverse tubular system. Many phenomena which have no counterpart in nerve fibers have been observed in skeletal muscle. Until the transverse tubular system was dis- covered, these phenomena remained without explanation. The discovery in muscle fibers of an extracellular space consisting of a network of narrow tubules lined with membrane has provided the structural basis for many hypotheses which at tempt to explain these phenomena. The transverse tubular system is so inaccessible, however, that direct investigations of its influence on the properties of muscle have been impossible. We have used glycerol-treated fibers in which the transverse tubular system is virtually absent (Howell and Jenden, 1967; Eisenberg and Eisenberg, 1968; Gage and Eisenberg, 1969; Eisenberg and Gage, 1969) to determine whether several phenomena of muscle do in fact reflect properties of the transverse tubular system. I t is generally believed that the transverse tubular system forms an essential part of the system by which an action potential on the surface membrane ~9a The Journal of General Physiology P. W. G^QE AND R. S. F_,ISENBERO Action Potentials and Exaltation-Contraction Coupling 299 causes contraction deep in the muscle fiber. Because localized contraction can be produced by depolarizing current applied extracellularly at localized sites along the Z line (Huxley and Taylor, 1958), it has seemed likely that the tubular system acts as a pathway into the fiber for the currents associated with the action potential. By this hypothesis action potentials should not be accompanied by contraction in fibers with a disrupted tubular system. The shape of an action potential in skeletal muscle is characterized by an afterdepolarization which is called here the early afterpotential. The time course of the later part of this early afterpotential corresponds to the time constant of the membrane (Falk, 1961 ; Persson, 1963). Because the transverse tubular system is responsible for a large part of the capacitance of muscle fibers (Fatt, 1964; Falk and Fatt, 1964; Eisenberg, 1967; Gage and Eisenberg, 1969), its removal should reduce the decay time of the early afterpotential. Other electrical properties of muscle fibers have been attributed to the transverse tubular system. The slow potential changes (late afterpotential and creep) which are present in skeletal muscle fibers have been attributed to prolonged changes in the potassium concentration of the solution filling the tubular lumen. Because of the very high ratio of the surface area of the membranes of the tubular system to the volume of the lumen of the tubules, relatively small current flows would be expected to produce significant changes in the concentration of ions. Thus, the prolonged slowly declining potential change which follows a train of action potentials (here called the late afterpotential) has been attributed (Freygang, Goldstein, and Hellam, 1964; Freygang, Goldstein, Hellam, and Peachey, 1964) to an accumulation in the lumen of the transverse tubules of the potassium which presumably crosses the tubular membrane during the train. Similarly, a slow potential change occurs when a sufficiently large inward (hyperpolarizing) current is passed across the membrane of muscle fibers: the potential does not approach a steady state with the normal time constant but continues to increase as if the membrane resistance were increasing (Adrian and Freygang, 1962). This slow increase in potential (called "creep") was attributed to a progressive decline in the potassium concentration in the tubular lumen. In the present investigation prolonged hyperpolarization and trains of action potentials have been recorded in glycerol-treated muscle fibers in order to determine whether the transverse tubular system is involved in these phe- nomena. Some of these results have been briefly described (Eisenberg and Gage, 1967; Gage and Eisenberg, 1967). M E T H O D S The preparation and solutions have been fully described (Gage and Eisenberg, 1969), but in most of these experiments the solutions did not contain tetrodotoxin. In order 302 T H E J O U R N A L O F G E N E R A L P H Y S I O L O G Y • V O L U M E 5 ~ " ~969 electrode. A typical experiment is illustrated in Fig. 2. Action potentials were elicited at different membrane potentials and it can be seen that reversal of the early afterpotential occurred close to the resting potential. A graph of the max imum ampli tude of t h e e a r l y afterpotential against membrane potential is shown in Fig. 3. In two fibers with resting potentials of - 6 8 my (circles) and - 8 4 my (squares) the membrane potential was set at a more depolarized or hyperpolarized level by passing current. A plot of the ampli tude of the early afterpotential against membrane potential was approximately linear 25mv 1 1sec 20 mv 5msec FmugE 2. Early afterpotentials in glycerol-treated fiber at different membrane po- tentials which were set by passing constant current from an intracellular electrode. The main figure shows action potentials and early afterpotentials at the different levels of membrane potential. Inset is the same record at a slower sweep speed and lower voltage gain to show the resting membrane potential, the polarized membrane potential, and the early afterpotential following the spike. At the level of depolarization in the uppermost trace, repetitive action potentials occurred. in the range close to the resting potential. The intercepts of the lines and the abscissa give the reversal potentials; in both cases these were at a more hyper- polarized level than the membrane potential. This reversal potential differs from the reversal potential in normal fibers in that it occurs at a slightly more negative potential than the resting membrane potential whether the latter is at - 6 8 mv or --84 my. I t is tempting to speculate that the level of the first phase of the early afterpotential has an equilibrium potential set mainly by the potassium equil ibrium potential (EK). Any accumulation of potassium ions in the P. W. GAG~- AND R. S. EISENBERO Action Potentials and Excitation-Contraction Coupling 303 transverse tubules dur ing an act ion potent ial would make the net EK more positive t han would be predicted f rom the dis tr ibut ion of potassium across the surface membrane . W h e n the transverse tubules are disrupted, however, potassium ions would no t accumula te and the net EK, now equivalent to the surface E x , would remain unchanged dur ing an act ion potential . Another explanat ion migh t be tha t following or dur ing an action potent ia l the trans- verse tubules become permeable to ano ther ion wi th an equi l ibr ium potent ia l ÷ 8 , > g- Z Ul n" ~_o I U. ~-2 I1: I11 - 4 - 6 - 8 ! - 5 0 F I O U R E 3. Vm(e) S Vm(•) f , , y -60 -70 -80 -90 -I00 MEMBRANE POTENTIAL (rnv) A plot of the maximum amplitude of the first part of the early afterpotentia against the membrane potential (immediately before and after the action potential) The results from two experiments are shown, the data in one case being shown as filled circles, in the other case as filled squares. The resting potentials (V,~) of the two fibers are shown. Note that the absolute values of the reversal potential (the "x intercept" of the line) are quite different, but that the difference between resting potential and the reversal potential is about the same in the two cases. The points above the line were difficult to measure and are shown mainly for the sake of completeness. more positive t h a n Ex dur ing the first phase of the early afterpotential . Wi thou t more evidence the quest ion remains open. Late Afterpotential After repeti t ive act ion potentials in frog sartorius fibers the m e m b r a n e re- mains depolar ized for some t ime (Freygang, Goldstein, and Hel lam, 1964; Freygang et al., 1964). T h e ampl i tude of this depolar izat ion and the t ime constant of decay of the afterpotential depend on the n u m b e r of act ion poten- tials in the train. Fol lowing a t ra in of 10 act ion potentials at a f requency of 100/sec, the peak ampl i tude of the late af terpotential is about 13 mv and the 304 T H E J O U R N A L O F G E N E R A L P H Y S I O L O G Y • V O L U M E 5 3 ° ~969 t ime constant of the decay is about 300 msec (Freygang, Goldstein, and Hellam, 1964; Freygang et al., 1964). It has been proposed that these late afterpotentials are the result of accumulation in the transverse tubules of potassium ions which leave the fiber during the action potentials (see also Hel lam et al., 1965). Because of the diffusion delays caused by the long narrow tubular system, this concentrated pool of potassium takes some time to be diluted to the normal potassium concentration; during this time the tubular membrane remains depolarized. This hypothesis is based of necessity on indirect evidence but would predict that fibers without a transverse tubular system should not have late afterpotentials. This is indeed what we found. A standard train of 10 or 11 action potentials at 100/sec was used to .? C D IOmsec toOomsec °. Fmvms 4. Trains of action potentials not followed by late afterpotentials in glycerol- treated muscle fibers. Each pair of records, taken from different fibers, shows repetitive action potentials at low gain and high sweep speed (above in A, B, and D; below in (3) and at high gain and slow sweep speed (noisier traces). Note the absence of the late afterpotential. allow comparison with previous results obtained in normal fibers. The late afterpotential found in normal fibers was not present in fibers without trans- verse tubules. This is illustrated in Fig. 4. Records were taken at two different gains and sweep speeds: one trace at high gain and a slow sweep speed (noisier record, below in A, B, and D, above in C) provided accurate records of the late afterpotential, the other at a lower gain and a faster sweep speed gave a good record of the action potentials. I t can be seen (Fig. 4) that there was no late afterpotenfial in any of these four fibers (A, B, C, and D) and this was invariably found. If in normal fibers the late afterpotential has an equilibrium potential more positive than the resting potential whereas in glycerol-treated fibers the reversal potential is the same as the resting potential, the lack of a late after- potential in the latter case could be simply explained by the absence of a P. W. GAOE AND R. S. EISENBERO Action Potentials and Exdtation-Contraction Coupling 307 contracting. The absence of twitches in glycerol-treated muscle fibers thus shows that the transverse tubular system is an essential link between excitation of the surface membrane and contraction. The presence of relatively normal action potentials in glycerol-treated fibers reveals some of the properties of these fibers. First, the very existence of the action potential shows that the sodium and potassium conductance systems presumably involved must lie on the surface membrane. Our results, however, do not rule out the possibility that these systems are also present in the tubular membrane. Second, the normal overshoot of the action potential shows that the internal sodium con- centration of these fibers is not greatly altered and further suggests that the internal potassium concentration is probably not grossly changed. The only part of the action potential which is appreciably changed is the early afterpotential. The later part of this potential is greatly shortened in glycerol-treated fibers supporting the suggestion that the time course of the early afterpotential is determined by the membrane time constant. Because the early afterpotential has a reversal potential within a few milli- volts of the resting membrane potential and the time course is so shortened, the early afterpotential is virtually absent in glycerol-treated preparations (Gage and Eisenberg, 1967). The polarization experiments (Figs. 2 and 3) show, however, that the mechanism underlying the early afterpotential is present, though with different properties from normal, in glycerol-treated fibers. The change in the reversal potential upon disruption of the tubules suggests that in normal fibers the level is governed to some extent by the tubular system as well as the surface membrane. As mentioned in the Results the mechanism responsible may be the potassium equilibrium potential in the tubules but the possibility that other ions may be involved has not been excluded. In order to analyze the slow potential changes of muscle, it is convenient to use an equation (Hodgkin and Horowicz, 1959) which describes the mem- brane potential in terms of the conductance and equilibrium potential of various ions and the total current flow V= = I + gK" EK' + gxt EK t gK* + gK t "}" gel" gK' + gK ~ 3r" gel' gK ° + gK t + gOl" (1) + g 01° Eot0 gK ° .{- gK ~ 3f. gOl ° Where V,, denotes membrane potent ia l ; / , membrane current; and E, with the appropriate sub- and superscripts, the equilibrium potential for the conductance system indicated. This equation has been written in terms of the potassium conductance of the surface and tubular membranes (gx ° and g~t, respectively), which in the resting state are in the ratio of about 308 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 53 " 1969 1:2; and in terms of the chloride conductance of the surface get* alone, since there is no resting chloride conductance in the tubular system (Eisen- berg and Gage, 1969). Currents which are small during the processes of interest, such as the sodium and capacity current, have been neglected. During the late afterpotential there is presumably little if any net current flowing through the membrane and the first term of the right side of equation (1) can be neglected. The membrane potential is then set only by the equi- librium potentials and relative conductances of the various systems involved. Freygang et al. (1964) have proposed that the late afterpotential is caused by a slow change in E~ t, associated with a change in the potassium con- centration in the tubular lumen. Our results support this hypothesis in two ways. First, it is clear that if the potassium concentration in the tubules is to influence the membrane potential, there must be an appreciable potassium conductance in the tubular membranes. In the preceding paper the results suggest that most of the resting potassium conductance system is in fact in the tubular system (Eisenberg and Gage, 1969). Second, the results reported here indicate that the late afterpotential is absent in fibers without tubules and thus imply that the late afterpotential is associated with the tubular system. Another explanation of the late afterpotential consistent with our data is possible, however. If even at rest E~" were not the same as EK', then a slow change in g~* following a train of action potentials would produce a late afterpotential. Such an afterpotential could occur in the absence of any change in the concentration of ions in the tubular lumen and would, of course, be absent in fibers without tubules. Furthermore, in view of the close proximity of the membrane of the terminal cisternae of the sarcoplasmic reticulum and much of the tubular membrane (see, for instance, Fig. 4, Gage and Eisenberg, 1969), it is not inconceivable that the potassium ion concentration on the "inside" of the tubular membrane (i.e. inside the sarco- plasmic reticulum) might be different from that of the sarcoplasm, thus making Ex" different from Ex'. I t is possible, of course, that both of the pro- posed mechanisms are involved in the late afterpotential; or that the after- potential is produced by a process involving some ion as well as, or other than, potassium. The slow creep of potential which occurs during inward current flow is somewhat more difficult to analyze since in this case the first term of equation (1) is not necessarily negligible. Adrian and Freygang (1962) have suggested that this hyperpolarization is caused primarily by a reduction in the potas- sium concentration of the solution in the lumen of the tubules. Our results support this hypothesis in that we have shown that there is no creep when there are no tubules, and that there is potassium conductance in the tubular membrane (Eisenberg and Gage, 1969). It is possible, however, to explain the phenomenon of creep in terms of a slow change in the conductance g~', P. W. C,-AOE AND R. S. EISENB~RO Action Potentials andExcitation-Contraction Coupling 309 as suggested in the discussion of the late afterpotential (see also Adrian, Chandler, and Hodgkin, 1968). In the analysis of creep, the presence of a term which involves the ratio of I (current) and g~* (tubular potassium con- ductance) means that a difference in the equilibrium potential for potassium in the tubules and surface does not have to be invoked. In normal muscle fibers when rather large inward currents were passed, the potential displacement passed through a distinct maximum and then declined (bellying) (Fig. 6 A and C). This phenomenon was reversible and was not associated with the deterioration of the fiber which occurs when very large hyperpolarizing currents are passed. This maximum in the creep potential was not reported by Adrian and Freygang (1962) and may be associated with a difference in technique. In their experiments, "conductance" was recorded under conditions when the membrane potential did not vary appreciably with distance whereas we recorded potentials under conditions when membrane potential did vary with distance. Another explanation of the maximum is possible. I t is known (Hodgkin and Horowicz, 1959; Ad- rian & Horowicz, personal communication) that in muscle fibers the potas- sium conductance is reduced at very low potassium concentrations with the result that the membrane depolarizes. A decline in gx has also been found when the potassium concentration is reduced to very low levels under differ- ent conditions and it has been suggested that the potassium concentration affects potassium permeability by determining the number of K carriers in the membrane (Horowicz, Gage, and Eisenberg, 1968)i I f strong pro- longed inward currents were so to deplete the potassium concentration in the tubular lumen that gx c was reduced in a similar way, the maxima caused by the stronger hyperpolarizing currents (Fig. 6 A and C) might then be explained. It is a pleasure to thank Dr. P. Horowicz for many helpful suggestions and discussions. Received for publication 25 October 1968. R E F E R E N C E S ADMAN, R. H., W. K. CHANDLER, and A. L. HODGKIN. 1968. Voltage clamp experiments in striated muscle fibers. J. Gen. Physiol. 51(5, Pt.2):188. ADMAN, R. H., and W. H. FREYGANO. 1962. The potassium and chloride conductance of frog muscle membrane. J. Physiol. (London). 163:61. EIS~Na~-ao, R. S. 1967. The equivalent circuit of single crab muscle fibers as determined by impedance measurements with intracellular electrodes. J. Gen. Physiol. 50:I 785. EISENBERO, B., and R. S. EISENBERC-. 1968. Selective disruption of the sarcotubular sys- tem in frog sartorius muscle. J. Cell Biol. 39:451. EIS~NBZaO, R. S., and P. W. GAo~.. 1967. Changes in the electrical properties of frog skeletal muscle fibers after disruptions of the transverse tubular system. Science. 158:1700. EIS~.NB~.RO, R. S., and P. W. GAO~.. 1969. Ionic conductances of the surface and tubular mem- branes of frog sartorius fibers. J. Gaz. Physiol. 53:279. FALK, G. 1961. Electrical activity of skeletal muscle--its relation to active state. In Biophysics