Sarcolemmal K+ channel activity in developing rat skeletal muscle membranes

Journal of the Neurological Sciences, 1990, 96:321-331

321

Elsevier JNS 03326

Sarcolemmal K ÷ channel activity in developing rat skeletal muscle membranes A.C. Wareham, I.C.M. Rowe and M.A. Whittle Department of Physiological Sciences, School of Biological Sciences, Medical School, University of Manchester, Manchester (U.K.)

(Received 27 November, 1989) (Revised, received 15 January, 1990) (Accepted 15 January, 1990)

SUMMARY A method has been adapted to produce membrane vesicles suitable for routine membrane patch clamping from neonate rat skeletal muscle. Single K ÷ channel activity was recorded from cell-free inside-out patches. Most Ca 2 ÷ -activated voltage sensitive channels had large conductances of up to 300 pS, as determined from their current/voltage relationship, and an open probability (Po) approaching unity at positive membrane potentials. A lower conductance K ÷ channel, probably responsible for inward rectification, had a lower conductance of about 100 pS. Outward rectifying K ÷ channels were also observed with the lowest conductance, about 40 pS. 0.1 mM ATP when applied to the inner membrane surface reduced or blocked activity, drastically reducing Po without altering single channel conductance. Such an effect has been reported in other preparations but was different in the neonate preparation in that it blocked channels with conductances as high as 300 pS. The simple preparation described, which we have also used successfully on mature rat and mouse skeletal muscle, has potential in the analysis of channel activities in various conditions and pathologies without the need for tissue culture to produce suitable membrane preparations.

Key words: K ÷ channels; Sarcolemma; ATP modulation; Neonate

Correspondence to: Dr. A. C. Wareham, Department of PhysiologicalSciences,Schoolof Biological Sciences, Medical School, Universityof Manchester, Oxford Road, Manchester M 13 9PT, U.K.

0022-510X/90/$03.50 © 1990Elsevier Science Publishers B.V. (BiomedicalDivision)

322 INTRODUCTION

Single-channel recording techniques (Neher and Sakmann 1976; Hamill et al. 1981) have been used to examine the gating and conductance properties of very small patches of cell membrane. The method has been used most extensively with tissue-cultured cells but also with adult cells treated enzymatically to remove connective tissue and ground substance (Hamill et al. 1981) to provide a clean surface capable of forming "gigaseals" with a microelectrode tip. In this way Na+,CaZ+-activated K + and K + -delayed rectifier channels have been identified and studied in excitable membranes (Conti and Neher 1980; Barrett et al. 1982). The sarcolemma of mature skeletal muscle is difficult to clean sufficiently to obtain "gigaseals". Instead, it has been necessary to study embryonic muscle (Clapham and De Felice, 1984) or cultured muscle (Barrett et al. 1982). More recently (Standen et al. 1984) a method was described to obtain a sarcolemma preparation from frog muscle capable of producing the required "gigaseals" for single channel studies. In essence, large ( > 50 #m) vesicles were formed from whole muscle by a combination of enzyme treatment and KC1 loading. Such vesicles produced good seals and, importantly, contained Na + and K+-channels which functioned normally (Standen et al. 1985; Burton et al. 1988). We have used a simplified method, not requiring treatment of the muscle with protease (Standen et al. 1984), to produce membrane vesicles with which to study K + -channel activities in developing mammalian skeletal muscles.

METHODS

Preparation of vesicles Extensor digitorum longus muscles from neonate rats (4-30 days) were dissected and placed in a high K + solution containing 0.5 mM CaCI z, 120 mM KC1, 50 mM NaCI, 5 mM Tris, pH 7.2 at room temperature (19-22 °C). The muscle did not undergo a violent contraction due to the high K ÷ solution and could be laid out in a glassbottomed bath and gently stretched between two bars. The high K + solution was exchanged for one containing collagenase (Sigma type !A, 125 units/ml). The bath was placed in an incubator at 37 °C for 40-60 min. Subsequently the solution was replaced with Ca 2 ÷-free, high K + solution at room temperature. The muscle was well irrigated and then left for 15 min. At the end of this time the surface of the solution was gently withdrawn to remove cell debris. Examination under a microscope showed that large ( > 50 #m) vesicles were loosely attached or emerging from between the muscle fibres. Often vesicles appeared from just one or two sites along a muscle. Care had to be taken at this stage since any vibration tended to detach the vesicles which fell to the bottom of the bath. Whilst these could be patched, our experience, and others (Standen et al. 1984) is that the vesicles lying on the fibres formed the best seals, presumably because they were cleaner. The majority of vesicles emerged, or were trapped, below the muscle and could be observed using an inverted microscope. After approximately half an hour no more vesicles were

323 produced. Those that had appeared remained, provided the K + concentration was not altered and survived in a viable state for several hours.

Electrophysiology Currents from small patches of membrane, diameter less than 1 #m, were recorded using a List L/M EPC-7 patch clamp amplifier. Glass pipettes were pulled from 1.8 mm diameter pyrex glass tubing (Clarke Electromedical Instruments) containing a glass filament to effect tilling with electrolyte solutions. Electrodes were pulled on a List L/M-3-A vertical puller and tire polished. We have investigated the benefits of different types of glass in making good, noise-free seals. Essentially hard glass was less "noisy" (Corey and Stevens 1983). Electrodes pulled from Kovar glass and polished provided good seals. They had to be back filled with solution and air bubbles in the tip removed with cat's whiskers. Clarkes filament electrode glass also produced good seals and had the distinct benefit of easy filling with solution due to the fine capillary running down the length of the glass, apparently functioning even after fire-polishing. Fire polishing was not essential to seal formation with this glass but did provide higher resistance seals. Electrode filling solution contained 120 mM KC1 and usually 0.1 mM C a 2+ and 50mM Na +. Electrode resistance was 1-3 MfL Seals of resistance 5-50 GO were made by very gently pushing the electrode, held in a Narashige hydraulic micromanipulator, against a vesicle and then applying slight suction to form the seal. Channel activity was recorded via a List EPC-7 amplifier digitised by a Sony PCM 701ES analogue-to-digital converter and stored on video tape. Subsequent analysis was performed using an IBM AT personal computer with a Tecmar Labmaster interface. Channel activity was analysed using PClamp V.4.0 (Axon Instruments Inc.). Hard copy was obtained using a Racal Store 4 tape recorder replayed slowly into a Gould chart recorder.

RESULTS

Vesicles Vesicles varied in size between 10 and 100 #m. Vesicles which remained attached to muscle fibres were normally patched. Whilst many vesicles detached themselves and fell to the bottom of the bath these tended to be difficult to patch successfully since they became dirty and then flattened out, making positioning of the electrode difficult. Under the light microscope fibres have been observed to produce vesicles as a raised area which grows and then pinches off, over a time period of about 10 min. Whilst such vesicles appear to form from the sarcolemma this cannot be confirmed using the light microscope. Attempts to confirm the source of vesicles under the EM were only partly successful, since vesicles act as osmometers and very readily collapse or rupture the procedures of fixation, etc., and destroy the majority of vesicles. However, we have resolved areas of sarcolemma in mouse muscle fibres which are raised away from the underlying sarcoplasm and which could be the sources of vesicle formation (Rowe et al. 1988).

324

Patching of vesicles Single channel activity was recorded from cell-free inside-out patches. These were obtained after manipulating a patch electrode, to which slight positive pressure was applied, against a vesicle to cause a slight depression in the membrane. At this stage the electrode positive pressure was exchanged for a slight negative pressure which usually resulted in formation of a gigaseal. Progress of the seal formation was monitored by the reduction of a square wave pulse passed through the electrode. Usually the seal formed suddenly, as indicated by the loss of the square wave as resistance increased. Occasionally seals were made which took as long as 30 sec to reach maximal resistance. The electrode was then moved away from the vesicle. Gigaseals formed in this way ranged from 5 to 40 Gf~. Quite frequently channels were either distorted or absent despite good seals with low noise. We believe that this was the result of the formation of a closed "mini-vesicle" in the pipette tip. Occasionally, brief air exposure of the tip restored an inside-out configuration and normal channel activity could be observed. More often, if such exposure had to be prolonged to disrupt the "mini-vesicle" it resulted in complete loss of the patch. The lifetime of viable patches ranged from 5 min to > 2 h. Seals were robust enough to permit several exchanges of the b a t h solution without disruption. Occasionally the muscle membrane was patched directly but good seals were more difficult to obtain, probably due to the electrode becoming contaminated by cell debris close to the fibres.

Single channel activity We have obtained patches from neonate rat skeletal muscle preparations which have shown normal functioning of the several K + -channel types described from other tissues and other muscle preparations. Depending on the recording conditions we have observed all of the major classes of K +-channels currently recognized in skeletal muscle (Moczydlowski et al. 1988). Hence, we have recorded outward-rectifying channels, anomalous or inward-rectifying channels and Ca 2 + -dependent channels. In addition we have observed the modulating effect of ATP to close inward-rectifying channels. We have recorded channel activities with a potassium gradient (10 mM K + outside and 120 mM K + inside) and with 120 mM K + on both membrane surfaces. Whilst the latter is not physiological it has the advantage of increasing single-channel conductances. Neither channel activation kinetics (Standen et al. 1984) nor distributions of open times (Standen et al. 1985) of K ÷ -channels in frog sarcolemmal vesicles are altered by raising [K ÷ ]o. The data presented here were all obtained using high [K + ]o in order to improve the resolution of single channel currents. Fig. 1. Ca2+-activated K+-channel activity. Patch clamp records from a patch isolated from vesicles • prepared from an EDL from a 14-day-old rat. The bathing solution was the same on both sides of the membrane and contained 120 m M K ÷ and 0.1 m M Ca 2÷ . (A)Single channel activity in response to stepping the membrane potential to the values shown on the left. Time scale and amplification are indicated by the calibration bars. Direction of channel opening is indicated by the arrows. The channel opened at both positive and negative membrane potentials. (B) IV relationship (pA vs. membrane potential (Em) mV) for the channel identified in (A). There was a linear relationship between - 8 0 and + 80 mV. The slope conductance was 291 pS. (C) Probability of opening plotted against membrane potential (mV). Po increased at membrane potentials above - 60 mV and was maximal at positive potentials.

325 A

20pA I lOmsec

-30

-60

-70

3020. o 10-

pA

0-10 -20-

-30 -100

I

I

I

I

I

I

I

I

I

~80

-00

-,40

-20

0

20

40

eO

80

100

Em

C 1.000 0 0.800

0

0

0

0

0

0

0

0.600 Po 0.400

0.200

0.000

-lOO

0 I

I

I

I

I

I

I

I

-80

-80

-40

-20

o

20

40

eo

Em

80

326

+40

+30

20pA

lOmsec

-8o

B

20, 15. 0

10,

0

0

0 0

5, 0

O, z~ A

-5.

o

A

A

0

-10, -15, -20 -100

I -80

i -BO

I -40

I -20

I 0

I 20

I 40

I 80

I 80

100

Em

C

1.000

0

0

0

0.800,

O.eO0 • Po 0.400-

0.2000 0

0.000 -40

I -20

o I 0

I 20

I 40 Em

I O0

I 80

1 O0

327 Ca 2 ÷ -dependent K ÷ -channels

In Ca 2 ÷ -free solutions, buffered with 1 m M E G T A , these channels were totally rectified in an outward direction, only opening with strongly positive potentials. When [Ca 2+ ] on the inner surface of the membrane patch was increased to 0.1 m M the channels lost their rectification and became active at negative potentials. In addition, there was a greatly increased activity at positive potentials where the probability of opening approached unity. Typically, these voltage sensitive Ca2÷-dependent K ÷ -channels had large conductances of 150-250 pS. In preparations from very young rat muscle it appeared that the conductance o f these channels was higher, in the region of 300 pS (Fig. 1). Inward-rectifying K + -channels

These were recorded with 120 m M K + and 0.1 m M Ca 2 + on both sides of the patch. They were easily identified by the fact that they only opened in response to membrane hyperpolarization and had conductances less then 100 pS. Their probability 'of opening increased with membrane polarization (Fig. 2). Outward-rectifying K ÷ -channels

A class of steady-state inactivated, outward-rectifying K ÷ -channels, activated by depolarization, was identified by stepping the membrane potential from values negative to E x towards zero. With 120 m M K + on both sides o f the patch these channels were inactivated when held at 0 mV. Holding the membrane at - 120 mV reprimed the channels by removing steady state inactivation after which the channels could then be activated by 2 0 - 3 0 mV depolarizing steps, although they were very quickly inactivated again. Measurements of open probability were relatively meaningless due to the rapid inactivation. Conductance could be calculated for these channels and typically lay in the region of 40 pS. A TP modulation o f K + -channels

A T P was not normally included in the solution bathing the membrane patches. In some preparations normal channel activity had been recorded when 0.1-1 m M A T P was applied to the inner surface of the patch. On m a n y occasions this had no discernable effect on channel activity. In some cases, however, A T P blocked single channel activity.

Fig. 2. Patch with two different channels present. Patch clamp records of an isolated patch taken from vesicles prepared from an EDL from a 5-day-old rat. Both sides of the membrane were bathed in the same solution which contained 120 mM K + and 0.1 mM Ca2+. (A) Channel activity in response to the membrane potential being stepped to the values shown on the left in mV. The time scale and amplification are indicated by the calibration bars. At least two different channels were activated. The larger channel opened to both depolarising and hyperpolarising steps. The smaller channel was rectified in an inward direction, only opening to hyperpolarising steps. It was distinguishable from the larger conductance channel by the fact that it opened in bursts. Opening is indicated by the arrows. (B) I vs. V relationship (pA vs. membrane potential (Era) mV) for the two identified channels in (A). The larger, non-rectified, channel had a slope conductance of 286 pS. The smaller, rectified, channel had a slope conductance of 90 pS. (C) Probability of opening of the larger conductance channel was rapidly reduced at negative potentials, unlike the more typical Ca2 ÷-activated K +-channel shown in Fig. 1.

328

20pA

[ 1.254=ec lOmuc

A

~

B ~

[l

L

l

Fig. 3. ATP modulation of channel activity. Patch clamp records from an isolated patch taken from vesicles prepared from an E D L from a 5-day-old rat. The membrane was clamped at + 30 mV and both sides of the patch were bathed in the same solution containing 120 m M K ÷ and 0.1 m M Ca 2+ . The patch was the same one from which the records in Fig. 2 were obtained. The upper trace shows channel activity on a slow time base as indicated by the 1.25 see calibration bar. The lower traces are expanded sections as indicated by letters A - D , with a time calibration of 10 msec. Section A is a recording of normal activity. Channel openings are upwards. At the first arrow A T P was added to the solution bathing the inner membrane surface to give a final concentration of 0.1 mM. The addition was complete by the second arrow. The trace shows the electrical disturbance during addition of ATP. Traces B, C and D show a progressive reduction of channel activity after ATP. There was a clear reduction in open probability without any alteration of single channel conductance. Channel conductance calculated from the I vs. V relationship determined before the addition of A T P was 286 pS (see Fig. 2).

Figure 3 shows the effect of 0.1 mM ATP on a large conductance K + -channel. A block occurred within 1 min of application and could be reversed by removal of ATP. In some preparations this process could be repeated several times after washing ATP out. The channels which were blocked by ATP in this way typically had large conductances, in the region of 250-300 pS, with 120 mM K + on both sides of the patch.

DISCUSSION

The preparation used to produce membrane vesicles from mammalian skeletal muscle fibres is simple and quick and does not require any cell culture techniques. There is, however, no certainty that the channels identified originated from the sarcolemma. Whilst our attempts at viewing stages in vesicle formation under the electron microscope did not resolve the origin of vesicles, areas of detaching sarcolemma were observed (Rowe et al. 1988). There is evidence that acetylcholine channels and other membrane proteins became mobile in similarly treated membranes (Tank et al. 1982). Hence, even if our vesicles do originate from sarcolemma, it is quite possible that some channels may have migrated from T-tubule regions into a forming vesicle. Nevertheless, the types of

329 K ÷-channel we have described in our preparations are consistent with expectations based on studies of whole muscle fibres and with results from patch-clamp studies on vesicles prepared from frog skeletal muscle (Standen et al. 1984, 1985) and human intercostal muscle (Burton et al. 1988). It was not our initial intention to characterise any of the channels in detail. Hence, we have not at this stage determined ion selectivity nor studied channel kinetics. The purpose of the study was to demonstrate the viability of the preparation for further, more detailed studies. Hence, we have restricted this report to a documentation of the types of K + -channel activity observed in neonate rat skeletal muscle. We have also prepared vesicles from mature rat and mouse skeletal muscle which are equally suitable for patching. Probably the most common channel in excitable cells is the Ca 2 ÷-activated K + -channel of which two principle types are recognised (Rudy 1988), The SK, with a small single channel conductance, is blocked by apamin (Blatz and Magleby 1984) and the BK or maxi channel with a large single channel conductance, is blocked by external TEA and by charybdotoxin (Blatz and Magelby 1984; Miller et al. 1985). The BK channel is activated by depolarisation in low Ca 2+ over a voltage range that becomes more positive the lower the internal free Ca 2 + (Barrett et al. 1982). In our experiments this channel could be activated between + 80 and + 100 even when calcium was buffered with 1 mM EGTA. The channel had a low open probability in the absence of Ca 2 + although Po was voltage-dependent. In the presence of excess Ca 2 + concentrations the channel became active at negative potentials and the I/V curve was linear. A large conductance K + -channel with very similar characteristics has been described in human intercostal muscle (Burton et al. 1988). The macroscopic current through BK channels is typically a large long-lasting outward current contributing to action potential repolarization and the after-hyperpolarization (Rudy 1988). Inward, or anomalous, rectification is an important property of skeletal muscle (Adrian 1964) where conductance increases with hyperpolarization, allowing K ÷ entry. In our preparations we frequently encountered channels which only opened at negative potentials and had conductances in the region of 50 pS. Interestingly, open probability of these channels was rarely greater than 0.5 and was not as voltage sensitive as the BK channels. This may indicate that the skeletal muscle channel contains an intrinsic inactivating mechanism (Hutter et al. 1988) as it does in cardiac muscle (Sakmann and Trnbe 1984). Outward rectifying channels or delayed rectifier channels, were also observed. Low conductance 'A' type channels were activated by depolarising steps from negative holding potentials and may be similar to those responsible for the 'A' current first described by Connor and Stevens (1971) in molluscan neurons. In the whole cell these channels would be briefly activated during the hyperpolarisation following an action potential and the resulting transient outward current would serve to stabilize the membrane. It is well documented that ATP blocks a class of low conductance voltage dependent K÷-channels in frog skeletal muscle (Spruce et al. 1987) although these authors were unable to demonstrate ATP-dependent K ÷ -channels in rat myotubes. In

330 our preparation, large c o n d u c t a n c e K + -channels were found in n e o n a t e preparations which were blocked by ATP. The previously reported direct action of A T P on channel activity has been of a block in skeletal muscle (Spruce et al. 1987), cardiac cells (Kakei et al. 1985) and pancreatic beta-cells (Cook and Hales 1984). One suggestion (Spruce et al. 1987) for the presence of A T P - d e p e n d e n t channels in muscle is to act to stabilise m e m b r a n e s in times of excessive use when A T P levels may be expected to fall to a low level. The n e o n a t e muscle fibre may be more vulnerable to fatigue and have developed a greater dependence u p o n this m e m b r a n e stabilisation action of low A T P levels by employing large conductance A T P dependent channels. The m e m b r a n e vesicle preparation described here will be useful in a variety of situations where cell culture techniques are unsuitable. F o r example, studies of channel activities during normal muscle development after birth and in a variety of pathogenic conditions, such as muscular dystrophy, which affect mature muscle fibres are n o w in progress in our laboratory.

ACKNOWLEDGEMENTS This work was supported by a Medical Research Project grant No. G8600958N. M . A . Whittle is a S E R C supported Research Student.

REFERENCES Adrian, R.H. (1964) The potassium and rubidium permeabilties of frog muscle membrane, J. Physiol. (Lond.), 175: 134-159. Barrett, J.N., K.L. Magleby and B.S. Pallotta (1982) Properties of single calcium-activated potassium channels in cultured rat muscle, J. Physiol. (Lond.), 331: 211-230. Blatz, A. L. and K. L. Magleby(1984) Ion conductance and selectivityof singlecalcium-activatedpotassium channels in cultured rat muscle, J. Gen. Physiol., 84: 1-23. Burton, F., U. Dorstelmann and O.F. Huner (1988) Single-channelactivity in sarcolemmal vesicles from human and other mammalian muscles, Muscle Nerve, 11: 1029-1038. Clapham, D.E. and L.J. De Felice (1984) Voltage-activatedK channels in embryonic chick heart, Biophys. J., 45: 40-42. Connor, J.A. and C. F Stevens (1971) Voltage clamp studies of a transient outward membrane current in gastropod neural somata, J. Physiol. (Lond.), 231: 21-30. Cook, D.L. and C.N. Hales (1984) Intracellular ATP directly blocks K + channels in pancreatic B cells, Nature, 311 : 271-273. Conti, F. and C. Neher (1980) Singlechannel recordings of K currents in squid axons, Nature, 285: 140-143. Corey, D.P. and C.F. Stevens (1983) Science and technology of patch-recording electrodes. In: SingleChannel Recording, Sakmann, B. and E. Neher (Eds.), Plenum Press, New York, pp. 53-68. Hamill, O. P., A. Marty, E. Neher, B. Sakmann and F.J. Sigworth (1981) Improved patch-clamp techniques for high-resolution current recording from cell and cell-free membrane patches, Pflugers Arch., 391: 85-100. Kakei, M., A. Noma and T. Shibasaki (1985) Properties of adenosine-triphosphate-regulatedpotassium channels in guinea-pigventricular cells, J. Physiol. (Lond.), 363: 441-462. Miller, C., E. Moczydlowski,R. Latorre and M. Phillips(1985) Charybdotoxin,a protein inhibitor of single C a 2 + -activated K ÷ -channels from mammalian skeletal muscle, Nature, 313: 316-318. Moczydlowski, E., K. Lucchesi and A. Ravindran (1988) An emerging pharmacology of peptide toxins targeted against potassium channels, J. Membrane BioL, 105:95-111.

331 Neher, E. and B. Sakmann (1976) Single-channel currents recorded from membrane of denervated frog muscle fibres, Nature, 260: 799-802. Rowe, I.C.M., A.C. Wareham, and M.A. Whittle (1988) Sarcolemmal ultrastructure during enzymic treatment of mammalian skeletal muscle to produce vesicles for patch clamping, J. Physiol., 403: 6P. Rudy, B. (1988) Diversity and ubiquity of K channels, Neuroscience, 25: 729-749. S akmann, B. and G. Traube (1984) Conductance properties of single inward rectifying potassium channels in ventricular cells from guinea-pig heart, J. Physiol. (Lond.), 347: 641-657. Spruce, A. E., N.B. Standen and P.R. Stanfield (1987) Studies of the unitary properties of adenosine-5-triphosphate-regulated potassium channels of frog skeletal muscle, J. Physiol. (Lond.), 382: 213-236. Standen, N.B., P.R. Stanfield, T.A. Ward and S.W. Wilson (1984) A new preparation for recording single-channel currents from skeletal muscle, Proc. R. Soc. Lond. B, 221:455-464 Standen, N.B., P.R. Stanfield and T.A. Ward (1985) Properties of single potassium channels in vesicles formed from the sarcolemma of frog skeletal muscle, J. Physiol. (Lond.), 364: 339-358. Tank, D.W., E. S. Wu and W.W. Webb (1982) Enhanced molecular diffusibility in muscle membrane blebs. Release of lateral constraints, J. Cell Biol., 92: 207-212.