HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hui, C. S. Search for Related Content PubMed PubMed Citation Articles by Hui, C. S.

A slow calcium-dependent component of charge movement in Rana temporaria cut twitch fibres

Chiu Shuen Hui

Received 12 December 1997; accepted after revision 6 March 1998.

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

1. Charge movement was studied in highly stretched frog cut twitch fibres in a double Vaseline-gap voltage-clamp chamber, with the internal solution containing either 0·1 mM EGTA or 20 mM EGTA plus 1·8 mM total Ca²⁺.

2. Fibres were stimulated with TEST pulses lasting 100-400 ms. Replacement of the external Cl^- with an 'impermeant' anion, such as SO₄^2-, CH₃SO₃^-, gluconate or glutamate, greatly reduced the calcium-dependent Cl^- current in the ON segment and generated a slowly decaying inward OFF current in charge movement traces.

3. Application of 20 mM EGTA to the internal solution abolished the slow inward OFF current, implying that the activation of the current depended on the presence of Ca²⁺ in the myoplasm. The possibility that the slow inward OFF current was carried by cations flowing inwards or anions flowing outwards was studied and determined to be unlikely.

4. During a long (2000 ms) TEST pulse, a slowly decaying ON current was also observed. When the slow ON and OFF currents were included as parts of the total charge movement, ON-OFF charge equality was preserved. This slow capacitive current is named I.

5. When Cl^- was the major anion in the external solution, the OFF I was mostly cancelled by a slow outward current carried by the inflow of Cl^-.

6. The OFF I component showed a rising phase. The average values of the rising time constants in CH₃SO₃^- and SO₄^2- were similar and about half of that in gluconate.

7. The OFF I component in CH₃SO₃^- had a larger magnitude and longer time course than that in SO₄^2-. The maximum amount of Q in CH₃SO₃^- was about three times as much as that in SO₄^2-, whereas the voltage dependence of Q was similar in the two solutions.

8. Since the existence of Q depends on the presence of Ca²⁺ in the myoplasm, it is speculated that Q could be a function of intracellular calcium release.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

In measuring charge movement in cut fibres, we routinely used 20 mM EGTA in the end-pool solution to suppress fibre contraction (beginning with Hui & Chandler, 1990). The presence of EGTA has the additional advantages of stabilizing the fibre for several hours, suppressing the calcium-dependent Cl^- current (Hui & Chen, 1994) and enhancing the appearance of the I hump (Hui & Chen, 1997). However, this high [EGTA]_i reduces the resting free [Ca²⁺] in the myoplasm to far below the physiological level, thereby reducing the Ca²⁺ reloading of the sarcoplasmic reticulum (SR) after activities. Secondly, it chelates the Ca²⁺ released from the SR, thereby inhibiting any feedback of the Ca²⁺ signal on the calcium release channels or charge movement. Thirdly, it distorts the kinetics and attenuates the magnitude of the calcium release signal monitored with the help of a calcium indicator. Thus, the optimal condition for measuring the Ca²⁺ transient is to keep the [Ca²⁺]_i close to the physiological level either by reducing the [EGTA]_i or by adding Ca²⁺ to the EGTA.

We found that when [EGTA]_i was reduced from 20 to 0·1 mM, it was necessary to replace the Cl^- in the external solution with an 'impermeant' anion, such as SO₄^2-, CH₃SO₃^-, gluconate or glutamate, to abolish the calcium-dependent Cl^- current (Hui & Chen, 1994). Under this condition, a slow inward current appeared in the OFF segments of charge movement traces. Evidence will be presented to show that this slow current is not likely to be carried by an outflow of anions or by an inflow of cations. With the application of long TEST pulses, a slow outward current also appeared in the ON segments. ON-OFF charge equality will be demonstrated, suggesting that the new current is a component of charge movement, which will be referred to as I. The time course of I in the ON and OFF segments lasts hundreds of milliseconds. This explains why it was not observed in previous experiments even when a high [EGTA]_i was not used.

The remaining parts of this paper will cover additional properties of I, including calcium dependence and waveform. An explanation for the absence of I in a Cl^- external solution and a semi-quantitative comparison of the magnitude of I in different external 'impermeant' anions will be given. Some of the findings in this paper were reported to the Biophysical Society in 1995.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Solutions

Solutions are given in Table 1 and abbreviations of the chemicals are defined in the legend. Cs⁺ in solutions B-D, and TEA⁺ and Rb⁺/Cs⁺ in solutions E-H were used to suppress K⁺ currents. Tetrodotoxin in solutions E-H was used to block Na⁺ current. TEA-Cl and (TEA)₂SO₄ were bought from R. S. A. Corp. (Ardsley, NY, USA). TEA-CH₃SO₃ and TEA-gluconate were prepared by titrating, respectively, methanesulphonic acid (Aldrich) and gluconic acid lactone (Sigma) with TEA-OH (R. S. A.). Frusemide was bought from Sigma.

Table 1. Solutions

	End-pool or internal solutions
	Caesium glutamate	Cs2-CP	MgSO4	Cs2-EGTA	Cs2-ATP	Cs2-PEP	Mg-ATP	Cs-Mops	Total Ca2+
B	72·5	20	0	0·1	0	0	5·5	20	0
C	42·5	20	0	20	0	0	5·5	20	0
D	50	20	6·8	20	5·5	5	0	5	1·8

Caesium creatine phosphate was prepared from sodium creatine phosphate (Calbiochem) by ion exchange. Early batches of the chemical so prepared had a small amount of contaminating Ca²⁺, which contributed 60 µM of total Ca²⁺ to solution C, but was chelated readily by the 20 mM EGTA in the solution. For solution B, in which only 0·1 mM EGTA was used, the same amount of contaminating Ca²⁺ was undesirable. We therefore used DTPA (diethylene-triamine-pentaacetic acid anhydride) coupled to aminoethyl Biogel P-2, nicknamed 'Ca²⁺ sponge', to remove most of the Ca²⁺ (see Hui & Chen, 1994). After this purification process, solution B contained 4-5 µM of total Ca²⁺, which was chelated adequately by the 0·1 mM EGTA. With 20 mM EGTA in solution C, the amount of free Ca²⁺ was estimated to be < 10^-12 M. Solution D was prepared by adding 1·8 mM CaCl₂ to solution C. The amount of free Ca²⁺ in solution D was estimated to be 5 × 10^-8 M.

The external solutions F and H were modified to enable the application of long TEST pulses. To avoid the elicitation of the slow inward Ca²⁺ current, particularly at large depolarizations, all the Ca²⁺ was replaced by Mg²⁺.

According to Godt & Maughan (1988), the osmolarity of the cytosol in frog skeletal muscle is about 235 mosmol l^-1. The osmolarities of all the internal and external solutions were therefore adjusted accordingly by adding either water or sucrose.

Muscle and fibre preparation

All experiments were performed on cut twitch fibres from English frogs, Rana temporaria, cold acclimatized in a refrigerator at around 4°C. Animals were killed by decapitation and destruction of the brain and spinal cord.

The procedure for dissecting and mounting cut fibres from semitendinosus muscle was similar to that used by Kovacs, Rios & Schneider (1983) and Irving, Maylie, Sizto & Chandler (1987). Briefly, a stretched muscle was exposed to a Ca²⁺-free, high-K⁺ relaxing solution (solution A in Table 1) which caused a transient contraction. After the muscle had relaxed, an 6-8 mm length of a single fibre was isolated. The fibre segment was mounted in a double Vaseline-gap chamber. It was stretched to a sarcomere length of 4 µm (except the fibre in Fig. 5, which had a sarcomere length of 4·2 µm) to abolish contraction. The outer membranes of the fibre in the end-pools were permeabilized by a 2 min exposure to 0·01 % saponin. The beginning of the treatment marked time zero of an experiment. Both end-pools were rinsed with solution A, and then filled with an internal solution (solution B or D). Finally, the solution in the centre-pool was changed to an external solution (one of solutions E-H). At the 20th to 23rd minute, the voltage clamp was turned on and the fibre was repolarized. A 30 min equilibration period was allowed for the fibre to recover and for various ions to diffuse into the myoplasm in the centre-pool region. During the course of an experiment, solution change(s) was carried out with the fibre under voltage clamp. The effectiveness of the procedure in changing the internal or external solution has been evaluated in Hui & Chen (1994).

Charge movement measurement

The instrumentation for data acquisition was enhanced to facilitate the inclusion of more input channels and the digitization of a larger array of points in each channel than that in the previous set-up (Chandler & Hui, 1990). The new data acquisition module was designed and fabricated by the Biomedical Instrumentation Laboratory of Yale Department of Cellular and Molecular Physiology. Ten analog signals, including six optical signals (not shown in this paper), three electrical signals (namely the potential in one end-pool (V₁), the potential in the other end-pool (V₂) and the total current injected into the latter end-pool (I₂)), and the temperature, were connected to the input channels of the module. The cut-off frequency of the eight-pole Bessel filter in each channel was set at 0·6 kHz. Data were digitized at a rate of 10 µs per point and sent to a PDP 11/73 computer for processing. The points in each channel were compressed before storage. As a result, each point in a stored trace corresponds to 1 ms.

Holding potential was set at -90 mV. CONTROL pulses were applied from -110 mV to the holding potential, and TEST pulses from the holding potential to the potentials desired. Throughout an experiment, the condition of the fibre was tracked by monitoring the holding current. Subsequent data analysis included linear cable analysis of the CONTROL records, which yielded information about myoplasmic resistance (r_i), membrane resistance (r_m), membrane capacitance (c_m), and gap factor of the Vaseline seals defined by r_e/(r_e + r_i), where r_e is external resistance underneath the Vaseline seals (Chandler & Hui, 1990). Each CONTROL current trace was an average of four sweeps and all TEST current traces were single sweep. A TEST - CONTROL current trace was obtained by subtracting a scaled CONTROL current trace from a paired TEST current trace. All experiments were performed at 13-14°C.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Notation

Following the notation introduced by Adrian & Peres (1979), Q and Q will be used to represent the early and hump components of charge. The new ultra-slow component of charge will be named Q. I, I and I will be used to represent the currents associated with the movements of Q, Q and Q. Since it is difficult to separate I and I when [Ca²⁺]_i is close to physiological level, they will be combined and named the fast components of charge movement. Likewise, the combination of Q and Q will be called the fast charge. The words 'CONTROL' and 'TEST' in upper case will be used to refer to the electrical signals elicited by CONTROL and TEST pulses.

Charge movement in cut fibres containing 0·1 mM EGTA

Figure 1A shows TEST - CONTROL current traces from a cut fibre bathed in a TEA-Cl external solution. The internal solution contained 0·1 mM EGTA. The I humps in the ON transients had very fast kinetics, consistent with the finding of Hui & Chen (1997). This made it difficult to separate visually the I hump from the I component. Following the charge movement transient in the ON segment, a delayed outward ionic current began to appear at -50 mV. The magnitude of the current was small at the beginning of depolarization but continued to increase during the pulse. At -40 mV, the current was increased substantially and was accompanied by a small outward OFF tail current, making the OFF charge movement transient somewhat smaller. Further depolarizations enhanced the delayed outward ON current and the outward OFF tail current progressively. The outward OFF tail current even obscured the inward OFF charge movement in the bottom three traces. The delayed outward ON current has been characterized to be calcium dependent (Hui & Chen, 1994). It is absent in cut fibres containing 20 mM EGTA, because the EGTA^2- ions chelate the Ca²⁺ released from the SR, thereby prohibiting the opening of the calcium-dependent channels.

After the traces in Fig. 1A were recorded, the TEA-Cl external solution in the centre-pool was replaced with a (TEA)₂SO₄ external solution. The traces in Fig. 1B show that the delayed outward ionic current was almost completely abolished, definitely supporting our previous finding that the current was carried by Cl^- (Hui & Chen, 1994). In the absence of the outward ionic current in the ON segment, the ON charge movement transient could be visualized more clearly. Also, since the accompanying outward tail current was abolished, normal OFF charge movement transients reappeared in the traces. Surprisingly, another slow inward OFF current appeared at around -50 to -40 mV and became progressively larger as the depolarization was increased. The slow current appeared to decay exponentially and lasted hundreds of milliseconds.

The traces in Fig. 1C were recorded after the (TEA)₂SO₄ external solution was washed out with the original TEA-Cl external solution. These traces were almost identical to those in Fig. 1A, implying that the effect of replacing Cl^- with SO₄^- was completely reversible.

The slow inward OFF current observed in Fig. 1B was not exclusively generated by SO₄^2-. It was observed whenever the Cl^- in the Ca²⁺-containing external solution was replaced with other 'impermeant' anions, such as CH₃SO₃^-, gluconate and glutamate. Experiments were performed on a total of seventy-four fibres, of which forty-six were in SO₄^2-, nineteen in CH₃SO₃^-, six in gluconate and three in glutamate. Some fibres were pre-exposed to the TEA-Cl external solution, as that in Fig. 1, while others were not. The slow inward OFF current was invariably observed in all the fibres.

		View larger version [in this window] [in a new window]
Figure 1. TEST - CONTROL currents from a cut fibre containing 0·1 mM EGTA The end-pools contained solution B. Initially, the centre-pool contained a TEA-Cl external solution (solution E). From the beginning to the end of the experiment, the holding current changed from -5 to -8 nA and re/(re + ri) remained stable at 0·994. A, representative traces taken from the 59th to the 79th minute. At the 87th minute, the centre-pool solution was changed to a (TEA)2SO4 external solution (solution G). B, representative traces taken from the 102nd to the 122nd minute. At the 129th minute, the centre-pool solution was changed back to solution E. C, representative traces taken from the 139th to the 159th minute. In all panels, the numbers to the right of the traces show the potentials during the TEST pulses. The thin straight lines in the OFF segments mark the zero-current axes. Fibre diameter, 96 µm (fibre 1).

Calcium dependence of slow inward OFF current

The slow inward OFF current just discovered was never observed in any charge movement traces we published in the past, even when the external Cl^- was replaced by SO₄^2-, CH₃SO₃^- or gluconate. One main difference in the experimental protocol was that the experiment described in Fig. 1 was carried out with 0·1 mM EGTA in the internal solution, whereas those in the past were carried out with 20 mM EGTA. In this section, results from an experiment will be presented to clarify whether the difference was indeed due to the presence of high [EGTA]_i.

The traces in Fig. 2A were recorded from a fibre initially equilibrated with an internal solution containing 0·1 mM EGTA. They resembled those in Fig. 1B. After the traces were recorded, the internal solution was replaced by one containing 20 mM EGTA. The fibre was allowed to equilibrate for 30 min to let the EGTA diffuse into the myoplasm, and the traces in Fig. 2B were recorded. They resembled those we published in the past and showed that the high [EGTA]_i was very effective in suppressing both the residual outward ionic current in the ON segments and the slow inward current in the OFF segments by chelating the myoplasmic free Ca²⁺ at rest and during activity. This establishes the calcium dependence of both currents.

		View larger version [in this window] [in a new window]
Figure 2. Abolition of the slow inward OFF current by 20 mM internal EGTA The centre-pool contained solution G. Initially, the end-pools contained solution B. From the beginning to the end of the experiment, the holding current changed from -31 to -38 nA and re/(re + ri) changed from 0·986 to 0·985. A, representative TEST - CONTROL current traces taken from the 57th to the 77th minute. At the 84th minute, the solution in the end-pools was changed to solution C. B, representative TEST - CONTROL current traces taken from the 117th to the 137th minute. In both panels, the numbers to the right of the traces show the potentials during the TEST pulses. The thin straight lines in the OFF segments mark the zero-current axes. Fibre diameter, 118 µm (fibre 2).

Possible ionic origin of the slow inward OFF current

The appearance of the slow inward OFF current is intriguing. The first obvious question that arises is whether the current is an ionic current or another unexplored component of charge movement. If it is ionic, it can be carried by anions flowing outwards. Since the internal solution contained glutamate as the primary anion, then glutamate is the most reasonable candidate to serve as the carrier. As the current is calcium dependent (see Fig. 2), it is logical to investigate whether the current is generated by glutamate flowing through calcium-dependent Cl^- channels. Figure 3A shows the presence of the slow inward OFF current in a fibre bathed in a SO₄^2- external solution. After the application of 5 mM frusemide, a well-known blocker of the calcium-dependent Cl^- channel (Evans, Marty, Tan & Trautmann, 1986; Akasu, Nishimura & Tokimasa, 1990), the traces in Fig. 3B were recorded. The slow inward OFF currents were almost identical in the corresponding traces of Fig. 3A and B. Similar results were obtained in another experiment in which frusemide was used. This rules out the involvement of a flow of glutamate through the calcium-dependent Cl^- channels but not a flow through another class of calcium-dependent anion channels which are permeable to glutamate and are resistant to frusemide. The latter possibility is considered unlikely based on the following reasons. First, it is difficult to explain why the current existed only when the external anion was impermeant but not when it was Cl^-. It would not be fruitful to search for such peculiar channels. Second, the current still existed when the external anion was glutamate.

		View larger version [in this window] [in a new window]
Figure 3. Indifference of the slow inward OFF current to the presence of external frusemide The end-pools contained solution B. Initially, the centre-pool contained solution G. From the beginning to the end of the experiment, the holding current remained stable at -17 nA and re/(re + ri) remained unchanged at 0·987. A, representative TEST - CONTROL current traces taken from the 56th to the 76th minute. At the 82nd minute, the centre-pool solution was changed to a (TEA)2SO4 external solution containing 5 mM frusemide (solution G with 3·3 mM SO42- replaced by 5 mM frusemide). B, representative TEST - CONTROL current traces taken from the 89th to the 109th minute. In both panels, the numbers to the right of the traces show the potentials during the TEST pulses. The thin straight lines in the OFF segments mark the zero-current axes. Fibre diameter, 89 µm (fibre 3).

If the slow inward OFF current is not carried by an outflow of anions, it remains to be shown whether the current is carried by an inflow of cations. In all the external solutions used, Na⁺ was completely replaced with TEA⁺ and tetrodotoxin was present. Thus, there should be no measurable inward Na⁺ current. Some Mg²⁺ is present in the myoplasm, but the driving force is in the wrong direction. Cs⁺ is much more abundant in the internal solution than in the external solution. Thus, the driving force is also in the wrong direction. The only remaining relevant cations to be considered are, therefore, K⁺ and Ca²⁺. Although all the K⁺ was replaced with Cs⁺ or Rb⁺, there remained a slight possibility that some residual K⁺ was present and the residual current was not completely blocked by TEA⁺. If that was the case, the residual K⁺ current should be blocked effectively by the potent K⁺ channel blocker 3,4-diaminopyridine (3,4-DAP; Aldrich). In the experiment shown in Fig. 4A, 2 mM 3,4-DAP was applied to the external solution. The traces clearly showed the presence of the slow inward OFF current. The same concentration of 3,4-DAP was applied to four other fibres and the slow inward OFF current was invariably present, implying that the current was not carried by K⁺.

		View larger version [in this window] [in a new window]
Figure 4. Existence of the slow inward OFF current in the presence of external 3,4-DAP or Cd2+ In both experiments, the end-pools contained solution B. A, the centre-pool contained solution G plus 2 mM 3,4-DAP. From the beginning to the end of the experiment, the holding current changed from -29 to -33 nA and re/(re + ri) changed from 0·977 to 0·974. TEST - CONTROL current traces were taken from the 90th to the 110th minute. Only representative traces are shown. Fibre diameter, 89 µm (fibre 4). B, the centre-pool contained solution G plus 2 mM Cd2+. From the beginning to the end of the experiment, the holding current changed from -30 to -31 nA and re/(re + ri) changed from 0·976 to 0·975. TEST - CONTROL current traces were taken from the 135th to the 161st minute. Only representative traces are shown. Fibre diameter, 109 µm (fibre 5). In both panels, the numbers to the right of the traces show the potentials during the TEST pulses. The thin straight lines in the OFF segments mark the zero-current axes.

The final cation to be considered is Ca²⁺. The involvement of this divalent cation is unlikely because: first, when [EGTA]_i is increased from 0·1 to 20 mM, the [Ca²⁺]_i is decreased such that the driving force for the inward flux of Ca²⁺ is increased, but the slow inward OFF current is abolished (Fig. 2) rather than increased; second, the slow inward OFF current was present when the external solution was Ca²⁺ free (see Figs 5 and 7-9 below); third, Fig. 4B shows an experiment in which 2 mM Cd²⁺ was applied to the external solution and the slow inward OFF current was not blocked. This makes it unlikely that the current was carried by Ca²⁺.

The results presented in this section indicate that the slow inward OFF current is unlikely to be an anionic or cationic current.

Is the slow inward OFF current part of charge movement?

If the slow inward OFF current is not ionic, then it must be capacitive. The main difficulty with this inference is that, if the slow inward OFF current is included as part of charge movement, then the total amount of OFF charge appears to be much larger than that of the ON charge (Figs 1B, 2A, 3 and 4). One possible explanation for this discrepancy is that, since the OFF current is so slow in kinetics, its ON counterpart could have a long time course as well. The short durations of the TEST pulses employed so far might have truncated the slow ON counterpart and, even if there was no truncation, the ON segments might not be long enough to allow an accurate determination of the baseline. The experiment shown in Fig. 5 was performed to test this hypothesis.

		View larger version [in this window] [in a new window]
Figure 5. Effect of TEST pulse duration on the OFF transient of a TEST - CONTROL current trace in Ca2+-free TEA-CH3SO3 external solution The centre-pool contained solution F and the end-pools contained solution D. From the beginning to the end of the experiment, the holding current changed from -30 to -44 nA and re/(re + ri) changed from 0·983 to 0·980. A, traces elicited, from the 213th to the 240th minute, by TEST pulses to -35 mV. The durations of the pulses were (from traces 1-9): 100, 200, 300, 500, 750, 1000, 1250, 1500 and 2000 ms. The thin straight lines in the OFF segments mark the zero-current axes. The half-widths of the OFF transients were: 17·6, 21·6, 26·5, 37·3, 41·7, 42·0, 46·0, 44·0 and 43·3 ms. The OFF charge in each trace was estimated by fitting a sum of two exponential decays and a straight line to the OFF segment and using the straight line for baseline correction. The first 10 points in the OFF segment, representing the I transient, were excluded from the fit. The amounts of OFF charge so obtained were: 42, 54, 70, 93, 100, 105, 108, 105 and 112 nC µF-1 for traces 1-9, respectively. The ON charge estimated from trace 9 after subtracting a baseline fitted to the last 500 ms of the ON segment was 109 nC µF-1. B, comparison of the OFF transients of traces 1 and 9 in A after the negative pedestals in the OFF segments had been removed. Fibre diameter, 90 µm (fibre 6).

In anticipation of the long durations of the TEST pulses, the experimental protocols were modified in three respects. First, it was found in a few preliminary experiments that fibres equilibrated with an internal solution containing 0·1 mM EGTA contracted at long, large depolarizations even though they were stretched to sarcomere lengths of 4 µm, probably because of the massive amount of Ca²⁺ released into the myoplasm. A decision was made to chelate some of the released Ca²⁺ with 20 mM EGTA and to restore the resting [Ca²⁺]_i to physiological level by adding extra Ca²⁺ (solution D in Table 1). Second, the external solution was made Ca²⁺ free to avoid the slow inward Ca²⁺ current. Third, a total of 4096 compressed points, i.e. 4096 ms, were stored in each trace to facilitate baseline fits in the ON and OFF segments. Because of the slow time course of the signal, fitting of baseline is expected to be difficult.

All the traces in Fig. 5A were elicited by TEST pulses to the same potential, but the duration of the pulses varied from 100 to 2000 ms. A potential of -35 mV was chosen to compromise between the largest charge movement signal possible and the least contamination from ionic current. The OFF segment of trace 1, elicited by the shortest pulse, showed a fast transient possibly containing two components (see next paragraph), and an ultra-slow current lasting hundreds of milliseconds. The ON segment showed an I component and an I hump followed by what appeared to be a baseline with a small positive slope. If this was the true ON baseline, then the amount of OFF charge appeared to be much larger than the amount of ON charge. This visualization of the ON baseline is absolutely erroneous, as will become obvious immediately. With a 200 ms pulse (trace 2), the fast transients in the ON and OFF segments remained unchanged. The maintained current in the ON segment became flat and the slowly decaying current in the OFF segment became larger in amplitude. With further increases in pulse duration (traces 3-9), it became clear that the late current in the ON segment was not maintained. Instead, it decayed slowly to a well-defined baseline (trace 9). This broad outward current lasted almost 1000 ms and resembled that observed by Pape, Jong & Chandler (1996).

The slowly decaying current in the OFF segment was increased from traces 1 to 9. Figure 5B provides a closer comparison of the early parts of the OFF segments from traces 1 and 9 after baseline correction on expanded time scales. In both segments, three distinct decaying components can be visualized clearly. The fast transient was actually made up of two components, presumably I and I, and followed by the slow component. The sharp corner at the boundary between the fast and slow currents in the OFF segments of trace 1 became much more rounded in trace 9. Also, the amplitude of the slow component was increased from trace 1 to more than double in trace 9.

To test whether the slow ON and OFF currents are part of charge movement, it is important to check ON-OFF charge equality by including the charges associated with these currents in the total charge. The total OFF charge in each trace of Fig. 5A was estimated by least-squares fitting a sum of two exponential decays and a straight line to each OFF segment, with the I transient excluded from the fit, and using the straight line for baseline correction. The values listed in the figure legend show that the OFF charge increased monotonically with pulse duration from trace 1 to trace 6. Thereafter, the OFF charge remained almost constant from trace 6 to trace 9 (with an average value of 108 nC µF^-1).

The total ON charge was estimated from trace 9 after subtracting a sloping baseline fitted to the last 500 ms of the ON segment. The ON charge value of 109 nC µF^-1 matched the average value of the OFF charge. The ON charge could not be estimated reliably from traces 1-8 because the ON segments of these traces did not last long enough to provide a reliable baseline fit. Results from thirteen other experiments, five in CH₃SO₃^- and eight in gluconate, supported ON-OFF charge equality.

The general conclusion from this section is that the slow inward OFF current was matched by a slow outward ON current. The time course of the slow outward ON current was so long that it required at least 1000 ms for the flow to be complete at the potential studied. Once the flow of the slow outward ON current was complete, the magnitude of the slow inward OFF current did not increase any further with increases in pulse duration. Because of the saturation of the charge associated with the slow inward OFF current elicited by pulses lasting 1000-2000 ms and because of the equality of charge associated with the slow ON and OFF currents, it is quite likely that the newly discovered slow current is capacitive in nature and, thus, is part of charge movement. Because the new capacitive current has a time course much slower than those of I and I, it can be regarded as a separate component of charge movement and will be called I in the rest of this paper. This does not automatically imply that the associated charge Q is necessarily a distinct charge species having an origin entirely different from those of Q and Q.

When the internal solution contains 20 mM EGTA without added Ca²⁺, I has a much delayed time course with respect to I in the ON segment of charge movement traces, and I is absent in both ON and OFF segments. When the internal solution contains 0·1 mM EGTA or 20 mM EGTA plus 1·8 mM Ca²⁺, I is accelerated such that it has a time course comparable to that of I in the ON segment of charge movement traces, and the amount of Q is reduced (Hui & Chen, 1997). Moreover, a slow I is present in both ON and OFF segments and the amount of Q is surprisingly large (see below).

OFF I can be obscured by a slow tail Cl^- current

If I is indeed a slow charge movement component, why did it disappear when the external solution contained Cl^- instead of an impermeant anion? One possible explanation is that, when a fibre repolarizes after a TEST pulse, some Cl^- may enter through open Cl^- channels before they are deactivated. This inflow of Cl^- generates a slowly decaying outward OFF current that might cancel most of the OFF I. The experiment shown in Fig. 6 was aimed at testing this possibility.

This experiment was performed on a fibre bathed in a TEA-Cl external solution. Because of the existence of the calcium-dependent Cl^- current, ON charge movement could not be analysed. Hence, there was no need to use long TEST pulses nor to use the internal solution with high [EGTA] plus Ca²⁺. The OFF segments shown in Fig. 6A were recorded in the usual 0·1 mM EGTA internal solution which was Cl^- free. They resembled those in Fig. 1A and C. The OFF I components were small but not exactly zero in these traces. After 4 mM glutamate in the internal solution was replaced by Cl^-, the traces in Fig. 4B were recorded. The OFF I component was present with a slow time course similar to those in the traces of the preceding figures. Figure 6C shows that this effect of 4 mM Cl^- was reversible. Similar effects were observed in four other fibres.

The simplest explanation for the observed effects in Fig. 6A and C is that the equilibrium potential of Cl^- across the outer membranes had a large positive value. The inward flux of Cl^- generated an outward current which, by chance, almost completely cancelled the slow inward OFF I component. When [Cl^-]_i was raised to 4 mM, as in Fig. 6B, the equilibrium potential of Cl^- was brought very close to the holding potential, -90 mV. Thus, no Cl^- entered the fibre and the OFF I became visible again. Experiments were also performed in which [Cl^-]_i was raised to 10 mM (not shown). The equilibrium potential of Cl^- became more negative than -90 mV. This drove Cl^- outwards, generating an inward ionic current, which was superimposed on the inward OFF I.

		View larger version [in this window] [in a new window]
Figure 6. Cancellation of OFF I by slow tail Cl- current in a fibre bathed in TEA-Cl external solution The centre-pool contained solution E. Initially, the end-pools contained solution B. From the beginning to the end of the experiment, the holding current changed from -13 to -15 nA and re/(re + ri) changed from 0·989 to 0·985. The TEST - CONTROL current traces were elicited by TEST pulses with durations ranging from 400 ms at -70 mV to 100 ms at 0 mV. Only OFF segments of the traces are shown in all panels. A, representative traces taken from the 59th to the 79th minute. At the 85th minute, the centre-pool solution was changed to a modified solution B with 4 mM caesium glutamate replaced by CsCl. B, representative traces taken from the 101st to the 121st minute. At the 127th minute, the centre-pool solution was changed back to solution B without CsCl. C, representative traces taken from the 141st to the 161st minute. In all panels, the numbers at the right ends of the traces show the potentials during the TEST pulses. The thin straight lines in the OFF segments mark the zero-current axes. Fibre diameter, 81 µm (fibre 7).

Waveform of the OFF I component

Since the amplitude and the time course of the fast OFF transient (including I and I) in Fig. 5A remained unchanged while the pulse duration was increased, the waveform of the I component can be extracted by subtracting the OFF segment of trace 1 from those of the other traces. This was the basis of the experiment shown in Fig. 7. The traces in Fig. 7A were elicited by TEST pulses to -40 mV. The pulse duration (50 ms) of trace 0 was carefully chosen such that the I hump had just sufficient time to complete its flow. The OFF segment of trace 0 was used as a template of the OFF I and I for subtraction from the OFF segments of the other traces. The difference traces so obtained are shown in Fig. 7B on expanded scales. Interestingly, the subtractions appeared to be perfect as there was absolutely no sign of any residue of I or I, suggesting that the difference traces might represent the OFF I accurately. These difference traces showed a slow rising phase. To the first order of approximation, the current was assumed to rise exponentially to saturation. The following expression:

A1 - exp(-t/_r)}, (1)

where _r is the rising time constant, was least-squares fitted to the difference traces from the beginning of repolarization to the peak of the OFF current, and the best-fit values of _r are listed in the figure legend. Excluding trace 1, in which I was too small, the values of _r lay between 4·5 and 5·3 ms without any obvious trend of increase or decrease. It was therefore assumed that _r is independent of pulse duration for convenience in data analysis.

		View larger version [in this window] [in a new window]
Figure 7. Waveform of OFF I in Ca2+-free TEA-CH3SO3 external solution The centre-pool contained solution F and the end-pools contained solution D. From the beginning to the end of the experiment, the holding current changed from -49 to -53 nA and re/(re + ri) changed from 0·979 to 0·978. A, TEST - CONTROL current traces elicited, from the 205th to the 223rd minute, by TEST pulses of different durations to -40 mV. B, OFF segments of difference traces. Traces 1-7 were obtained by subtracting the OFF segment of trace 0 in A from those of traces 1-7 in A. The values of the rise time r, obtained by fitting expression (1) to the rising phases of the segments, are: 2·7, 4·9, 4·7, 5·3, 4·5, 5·1 and 5·2 ms for traces 1-7, respectively. C, OFF segments of difference traces. Traces 1-4 were obtained by subtracting the OFF segments of traces 0-3 in A from that of trace 7 in A. The first trace is the same as the last trace in B. The values of the rise time r, obtained by fitting expression (1) to the rising phases of the segments, are: 5·2, 6·1, 6·2 and 6·6 ms for traces 1-4, respectively. In all panels, the thin straight lines in the OFF segments mark the zero-current axes. Fibre diameter, 104 µm (fibre 8).

Since a 50 ms pulse was seldom used at intermediate depolarizations, it was necessary to find out whether the OFF segments from other traces elicited by longer pulses could be used as a template for subtraction. Another set of difference traces was obtained (Fig. 7C) by subtracting the OFF segments of traces 0-3 in Fig. 7A individually from the OFF segment of trace 7 in the same panel. Trace 1 in Fig. 7C is a duplicate of trace 7 in Fig. 7B. The amplitude of I became smaller and smaller from the top trace to the bottom trace in Fig. 7C as expected, because more and more I current was present in the template used for subtraction. Nonetheless, the waveform appeared to be similar in the four traces, and when expression (1) was least-squares fitted to the rising phases of the four traces, the best-fit values of _r did not come out to be substantially different in the four traces. Trace 4 shows that even when the OFF segment from a trace elicited by a 500 ms pulse was used as a template for subtraction, a decent I component could still be obtained. Thus, whenever the waveform of I is extracted from a pair of traces by the method just described, the duration of the shorter pulse does not have to be < 100 ms. Any pulse duration between 100 and 500 ms can be used, if a smaller fraction of I can be tolerated. There is some advantage in using a pulse longer than 50 ms, namely to ensure that I is completely activated and included in the template.

Twelve other similar experiments were performed, four in Ca²⁺-free TEA-CH₃SO₃ external solution and eight in Ca²⁺-free TEA-gluconate external solution. No general pattern of dependence of _r on the pulse duration of the template was observed.

Another important question is whether the waveform of the OFF I component is independent of the potential during the TEST pulse. In all the experiments performed thus far, the best-fit values of _r had no apparent voltage dependence. They were generally smaller at -50 and -40 mV, but became relatively constant between -35 and -20 mV. The values are less reliable at smaller depolarizations because I is smaller. Since it is impossible to estimate the values of _r in the complete voltage range in every experiment, the value obtained at around -30 mV will be used to represent, to the first order of approximation, the average value of _r at all voltages for that fibre. This value of _r will be utilized to separate the OFF I component from the fast OFF component in the analysis below. This approximation could introduce a small error in the amounts of charge at potentials < -40 mV, but the error should be negligible and only affect the foot of the Q-V plot.

The values of _r were estimated from twenty-six fibres in Ca²⁺-free TEA-CH₃SO₃ external solution and nineteen fibres in Ca²⁺-free TEA-gluconate external solution, mostly at a single potential in each fibre. The mean values of _r were 8·2 ± 0·5 and 18·3 ± 1·4 ms (means ± S.E.M.), respectively. Thus, the OFF I component rises more slowly in Ca²⁺-free TEA-gluconate external solution than in Ca²⁺-free TEA-CH₃SO₃ external solution, and the difference is statistically significant (P < 0·001, Student's two-tailed t test). The value of _r was also estimated from three fibres in Ca²⁺-free (TEA)₂SO₄ external solution and the mean value was 9·7 ± 0·3 ms, not significantly different from that for Ca²⁺-free TEA-CH₃SO₃ external solution (P > 0·2, Student's two-tailed t test).

Separation of I from I and I

In this section, an attempt is made to develop a method to estimate the amount of charge associated with I. Since it was difficult to separate I and I in fibres containing 50 nM free Ca²⁺, they are combined and referred to as the fast components, and the I component is separated from them by the methods illustrated in Fig. 8, in which an experiment was carried out in a Ca²⁺-free (TEA)₂SO₄ external solution. Figure 8A shows TEST - CONTROL current traces elicited with 2000 ms TEST pulses to the potentials indicated. Each ON segment showed the fast components of charge movement followed by the I component which was superimposed on a small negative pedestal. At -20 mV, the negative pedestal was masked by the residual outward ionic current (Hui & Chen, 1994). The slow rising phase of the ON I component at each potential made it easy to separate out visually the fast components of charge movement. Each OFF segment also showed the fast components followed by a slow I component and a small maintained negative current.

Figure 8B provides a closer view of the ON transient in the fourth trace on expanded scales. The fast ON components were isolated by fitting a baseline to the rising phase of the I component (indicated by the thin line). With this baseline correction, the amount of charge associated with the fast ON components was estimated to be 9·5 nC µF^-1.

Figure 8C shows the OFF transient of the same trace on expanded scales. A sum of two exponential decays and a straight line was fitted to the OFF segment and the slower of the two exponential functions is indicated by curve 1. If this curve was taken to represent the I component, the amount of fast OFF charge, given by the area enclosed by curve 1 and the empirical trace (dotted curve), was estimated to be 7·8 nC µF^-1, which is slightly smaller than the amount of fast ON charge. The small error lay in the failure to account for the rising phase of I. Information about the rising phase was obtained with the procedure developed in the preceding section. A difference trace was obtained by subtracting another trace elicited by a 400 ms TEST pulse to -30 mV (not shown) from the fourth trace in Fig. 8A. The OFF segment showed a marked rising phase similar to those shown in Fig. 7B. The value of _r estimated for the rising phase of OFF I was 9·6 ms. Curve 2 was obtained after correcting curve 1 by expression (1) with _r set to this value. If curve 2 is taken to represent the I component, the amount of fast OFF charge, given by the area enclosed by curve 2 and the empirical trace, was estimated to be 9·7 nC µF^-1, which is much closer to the amount of fast ON charge than the amount separated by curve 1. This implies that curve 2 might provide a more accurate method for separating the OFF I component from the fast OFF components.

		View larger version [in this window] [in a new window]
Figure 8. Separation of I from I and I The centre-pool contained a Ca2+-free (TEA)2SO4 external solution (solution H) and the end-pools contained solution D. From the beginning to the end of the experiment, the holding current changed from -28 to -34 nA and re/(re + ri) changed from 0·983 to 0·980. A, representative TEST - CONTROL current traces taken from the 145th to the 161st minute. The numbers at the right ends of the traces show the potentials during the TEST pulses. The thin straight lines in the OFF segments mark the zero-current axes. B, ON transient of the 4th trace in A. Only the early part of the ON segment is shown on expanded scales. The straight line represents the baseline obtained by fitting a sum of an exponential and a straight line to the points between the tick marks. C, OFF transient of the 4th trace in A. Only the early part of the OFF segment is shown on expanded scales. Curve 1 represents the slow exponential decay obtained by fitting a sum of two exponentials and a straight line to the points between the tick mark and the end of the complete OFF segment (not the end of the part shown). Curve 2 was obtained by multiplying curve 1 by (1 - exp(-t/9·57)). Fibre diameter, 81 µm (fibre 9).

Comparison of I components in external solutions containing different major anions

Since the choice of the major anion in the external solution alters the relative proportion of I and I (Hui, 1991b; Hui & Chen, 1995), it is of interest to compare the I components in various external solutions containing different anions. An example is shown in Fig. 9. TEST - CONTROL current traces were recorded when a fibre was bathed in Ca²⁺-free TEA-CH₃SO₃ external solution (Fig. 9A) and Ca²⁺-free (TEA)₂SO₄ external solution (Fig. 9B). In this fibre, the residual outward ionic ON current did not appear until -10 mV, about 10 mV less negative than in the fibre of Fig. 8. Several differences can be observed between the traces in Fig. 9A and B. First, in SO₄^2-, the ON I component was activated after the fast components were over. This generated a dip in current between the two components (same as in Fig. 8A). In contrast, there was no such dip in CH₃SO₃^- (same as in Fig. 7A), presumably because the ON I was activated before the fast components were over, i.e. the rising phase of ON I was buried in the decay phase of the fast components. Second, the OFF I component appeared to be larger and slower in CH₃SO₃^- than in SO₄^2-. Third, in CH₃SO^3-, the OFF current settled at a negative maintained current, which was absent in SO₄^2-. Surprisingly, the fast components in SO₄^2- did not appear to be larger than those in CH₃SO₃^-, although when the external solutions contained the normal amount of Ca²⁺ and the internal solution contained 20 mM EGTA without added Ca²⁺, the amount of Q in SO₄^2- was more than three times as much as that in CH₃SO₃^- (Hui, 1991b).

		View larger version [in this window] [in a new window]
Figure 9. Comparison of charge movement components in calcium-free TEA-CH3SO3 and Ca2+-free (TEA)2SO4 external solutions The end-pools contained solution D. Initially, the centre-pool contained solution F. From the beginning to the end of the experiment, the holding current changed from -31 to -47 nA and re/(re + ri) changed from 0·979 to 0·969. A, representative TEST - CONTROL current traces taken from the 175th to the 204th minute. At the 212th minute, the centre-pool solution was changed to solution H. B, representative TEST - CONTROL current traces taken from the 227th to the 251st minute. In A and B, the numbers at the right ends of the traces show the potentials during the TEST pulses. The thin straight lines in the OFF segments mark the zero-current axes. C and E (or D and F), steady-state voltage distributions of fast (Q and Q) and slow (Q) OFF charges, respectively, obtained from the time integrals of OFF transients in TEST - CONTROL current traces, some of which are shown in Fig. 9A (or B). In C and D, and were obtained by the separation procedure illustrated by curves 1 and 2, respectively, in Fig. 8C. The timeconstants used for computing the rising phases of curve 2 were 9·9 ms in C and 9·1 ms in D. For comparison, the fast ON charges obtained after the baseline correction illustrated in Fig. 8B are shown as in D. The pairs of and overlap each other at -60, -50, -25 and -20 mV. In E and F, represents OFF Q corrected for the rising phase of I. In each of C-F, the smooth curve represents the least-squares fit of eqn (2) to the data () and the best-fit parameters are listed in columns 4-6 and 8-10 of fibre 13 in Table 2. Fibre diameter, 85 µm.

The decay phases of the OFF segments in both solutions were fitted by a sum of two exponentials and a straight line. Two decay time constants were used in the fit: _f for the fast components, and for I. Averaged over a potential range of -40 to -20 mV, the values of _f and were 20 and 372 ms in CH₃SO₃^-, respectively, and 12 and 121 ms in SO₄^2-, respectively. These values are listed in Table 2 under fibre 13. Thus, relative to SO₄^2-, CH₃SO₃^- appeared to slow the decays of both the fast OFF components and the OFF I.

Table 2. Comparison of fast (I and I) and slow (I) components of charge movement in Ca²⁺-free (TEA)₂SO₄ and TEA-CH₃SO₃ external solutions

			Fast components				Slow component
Fibre reference	External solution	cm (µF cm-1)	Vf (mV)	kf (mV)	qf,max/cm (nC µF-1)	f (ms)	V (mV)	k (mV)	q,max/cm (nC µF-1)	(ms)
10	CH3SO3-	0·212	-46·5	6·6	19·9	15	-53·2	10·9	104·7	344
	SO42-	0·221	-49·1	11·9	13·9	10	-47·7	12·8	30·0	106
11	SO42-	0·314	-48·9	8·7	6·8	8	-46·2	12·2	49·2	170
	CH3SO3-	0·274	-45·9	6·4	12·2	17	-51·3	11·6	109·1	345
12	CH3SO3-	0·226	-46·8	5·4	16·4	15	-54·2	9·8	100·8	353
	SO42-	0·246	-51·8	8·0	10·4	10	-53·7	10·5	41·5	151
13	CH3SO3-	0·183	-41·4	8·1	20·6	20	-46·3	9·3	113·9	372
	SO42-	0·195	-45·2	10·4	12·9	12	-50·7	9·0	27·8	121
14	CH3SO3-	0·155	-45·5	6·4	16·3	13	-56·8	11·3	108·5	354
	SO42-	0·162	-50·3	9·3	9·3	8	-47·2	6·0	28·2	163
15	SO42-	0·193	-47·0	7·0	10·4	8	-46·7	12·3	50·0	184
	CH3SO3-	0·168	-43·3	7·4	16·1	10	-54·2	12·1	99·1	356
Mean	SO42-		-48·8	9·2	10·6	9	-48·7	10·5	37·8	149
S.E.M.			0·9	0·7	1·0	1	1·2	1·1	4·3	12
Mean	CH3SO3-		-44·9	6·7	16·9	15	-52·7	10·8	106·0	354
S.E.M.			0·9	0·4	1·2	1	1·5	0·4	2·3	4

The method developed in Fig. 8 was applied to compare the amounts of fast charge and Q in the two calcium-free external solutions. The amounts of fast OFF charge in CH₃SO₃^- and SO₄^2- were estimated first without correcting for the rising phase of I (curve 1 in Fig. 8C), and are plotted against the TEST pulse potential in Fig. 9C and D, respectively(). Because of the appearance of the residual outward ionic ON current at -10 mV, the analysis was limited to -20 mV to avoid contamination of the OFF charges by tail currents. For comparison, the amounts of fast ON charge in the SO₄^2- solution, estimated after the baseline correction procedure as shown in Fig. 8B, are plotted in Fig. 9D (). The amounts of fast ON charge in the CH₃SO₃^- solution could not be estimated because the rising phases of the ON I were hidden.

The amounts of fast OFF charge should be more accurate if they are corrected for the rising phase of I (curve 2 in Fig. 8C). From the difference traces at -30 mV (not shown), the _r values of OFF I in both solutions were estimated and used to correct for the rising phases at all potentials. The amounts of fast OFF charge in CH₃SO₃^- and SO₄^2- so obtained are plotted in Fig. 9C and D (). The correction procedure increased the amounts of fast OFF charge in both solutions and improved the ON-OFF charge equality appreciably in SO₄^2-. In fact, the corrected fast OFF charge overlapped with the fast ON charge at several potentials (see figure legend).

The amounts of OFF Q in CH₃SO₃^- and SO₄^2- after correcting for the rising phase of I are plotted against the TEST pulse potential in Fig. 9E and F, respectively. The amounts without correction are not shown because they should be very close to the amounts with correction. The reason is that, in any given trace, the amounts of OFF Q without and with the rising phase are given by Aexp(-t/)dt and ‚A{\123}1 - exp(-t/_r)}exp(-t/)dt, respectively. The two amounts are proportional to and ²/(_r + ). Thus, the rising phase effectively reduces the amount of Q by a fraction equal to _r/(_r + ). Since _r is of the order of 10 ms in CH₃SO₃^- and SO₄^2- and is of the order of 200 ms, the fraction of Q reduced by the correction is of the order of 5 %. The amounts of ON Q are also not shown in the figure because, with the slow decay of the ON I and the inevitable noise level in the current, it was difficult to determine the ON baseline independently with certainty. Attempts were made to fit baselines to the last few hundred milliseconds of the ON segments, as in the analysis associated with Fig. 5. It suffices to mention that, if the cursors of the fit were chosen carefully, the value of the total ON charge almost always matched the value of the total OFF charge. In the case of SO₄^2-, since the fast ON charge matched the corrected fast OFF charge, it implied that the ON Q should be equal to the corrected OFF Q.

The values plotted as filled diamonds () in Fig. 9C-F were least-squares fitted by a Boltzmann distribution function modified from the original function used by Chandler, Rakowski & Schneider (1976). The modification corrects for the charge moved by the CONTROL pulse between -110 and -90 mV. It has been used routinely in our previous publications and is similar to the procedure used by Melzer, Schneider, Simon & Szucs (1986). In the modified function, the amount of charge Q at any potential V is given by:

Q(V) = Q_max F *(V), (2)

in which Q_max represents the maximum amount of charge and the function F * is defined by:

F *(V) =

F(V) - F(-90) - F(-110)}(V + 110)/20 - F(-110), (3)

F(V) = 1 + exp(-(V - (overbar)V/k))}^-1, (4)

where (overbar)V and k represent the equi-distribution potential and the voltage dependence (or inverse steepness) factor, respectively.

The best fit values of Q_max, (overbar)V and k for the curves in the four panels are listed in Table 2 under fibre 13. The maximum amount of fast charge in CH₃SO₃^- was almost twice as much as that in SO₄^2-. This is unexpected because, in Ca²⁺-containing external solutions and nominally Ca²⁺-free internal solution, the total amount of Q and Q in CH₃SO₃^- is less than that in SO₄^2- (Hui, 1991b). More surprisingly, the maximum amount of Q in CH₃SO₃^- was four times as much as that in SO₄^2- in this fibre.

These diversities were observed consistently in five other fibres. The order of the solution change was reversed in two fibres to minimize systematic error. The best-fit values of Q_max, {"Overbar (l.c.)" on}V{"Overbar (l.c.)" off}, k and for the fast and Q components in both solutions are also listed in Table 2. Larger amounts of fast and Q charge were always observed in CH₃SO₃^- than in SO₄^2- irrespective of which solution was used first. Averaged over the six fibres, CH₃SO₃^- increased Q more than the fast charge. It also lengthened more than _f. All the increases were statistically significant (P < 0·01 for Q_f,max and _f, P < 0·001 for Q_,max and ; Student's two-tailed t test).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

When the myoplasmic Ca²⁺ in cut fibres is chelated by 20 mM EGTA, the ON I component is manifested as a prominent hump and Q accounts for about half of the total charge (Hui & Chandler, 1990; Hui, 1991a). When the [Ca²⁺]_i is restored to the physiological level, the amount of Q is very much reduced and the remaining I is accelerated (Hui & Chen, 1997). In this paper, results are presented to show that although I is reduced in the latter situation, slow I appears in both the ON and OFF segments of TEST - CONTROL current traces. The occurrence of this slow current depends mandatorily on the presence of Ca²⁺ in the myoplasm, as 20 mM EGTA abolishes it (Fig. 2).

Capacitive nature of I

The possibility of the slow current being ionic in nature was considered. The involvement of a monovalent anion (glutamate), a monovalent cation (Cs⁺ or K⁺), or a divalent cation (Ca²⁺ or Mg²⁺) as the current carrier was ruled out systematically, suggesting that the current is more likely to be capacitive. To validate this suggestion, the experimental protocols were modified to enable the recording of TEST - CONTROL current traces with long ON and OFF segments. With these modifications, the total amounts of ON and OFF charge were found to be conserved (Fig. 5). This demonstration of ON-OFF charge equality is vital in establishing the fact that the slow current is part of charge movement. In view of the extremely long time course of the slow current, which distinguishes it from the familiar I and I, it is perceived as a new component of charge movement and named I, without any intention to reflect that all of Q necessarily has a physical origin entirely distinct from that of Q or Q. It is quite possible that some charge entities possess the properties of Q under some conditions and the properties of Q under other conditions. We have speculated that even parts of Q and Q might share a common physical origin but behave differently under different conditions (see Discussion in Hui, 1991b).

The absence of I in previous studies on cut fibres containing a trace amount of EGTA can be explained easily by the long time course of I. In the conventional measurement of charge movement, the duration of a TEST pulse was no longer than 100-200 ms. As described in conjunction with Fig. 5, this short duration truncated the ON I. The early portion of I so obtained could have been mistaken easily for the ON baseline and removed by the baseline correction procedure. A similar truncation occurred in the OFF segment. Some investigators generally marked two cursors in the later part of the OFF segment for fitting a straight line when the current between the cursors was still decaying exponentially instead of linearly. The OFF charge obtained with such baseline correction should be larger than the fast OFF charge reported in this paper or in Hui & Chen (1997), because it included part of OFF Q not completely removed by the baseline correction procedure. To properly separate the total charge movement transient from the ionic pedestal in the OFF segment, a long segment needs to be digitized. For a decay with a time constant of the order of 200 ms, ideally the length of the segment should be at least 1000 ms.

Separation of charge movement components

The presence of I makes the study of charge movement much more complicated. Even if Q is not a new physical entity with a completely distinct origin, its properties can be studied more conveniently if it is separated from the other components. When charge movement was first discovered (Schneider & Chandler, 1973; Chandler et al. 1976), it was visualized as a simple, exponentially decaying current in the ON and OFF segments of TEST - CONTROL current traces. When the bathing solution was made slightly less hypertonic, a hump-shaped I component appeared following the I component in the ON segment (Adrian & Peres, 1977, 1979), but the two components could not be visually separated in the OFF segment, although Hui & Chandler (1991) showed that the OFF I component had a longer time course than the OFF I component. With just two components, the separation of charge movement transients into I and I (or the associated charge into Q and Q) was not straightforward, but several methods were introduced by various investigators to separate the components (Huang, 1981, 1982, 1991, 1994; Hui, 1983a, b; Hui & Chandler, 1990, 1991; Hui & Chen, 1995). Under the conditions in which I was visible, it became even more difficult to separate I and I_, and so they were combined as the fast components.

Separation of I from the fast components can be accomplished because the kinetics of I is distinctly slow and because the rising phase of I in the OFF segment can be extracted from difference traces, as shown in Fig. 7. It should be noted, however, that the results obtained from the separation procedure shown in Fig. 8C should be treated as semi-quantitative for several reasons. First, the rising time constant, _r, was assumed to be independent of pulse duration and potential, which might not be exactly correct. Second, the decay phase of I was fitted with a single time constant, , which might not be exactly accurate, especially if I is not a single component arising from a single origin. Third, I had such a long time course that any noise or drift in the current might make the fitted deviate seriously from the true time constant, thereby making the integrated area of the I curve deviate seriously from the true amount of associated charge.

The I component so separated appeared to have some other properties quite different from those of the fast components. First, the magnitude of Q and the decay time constant of OFF I were more dependent on which impermeant anion was present in the external solution (Table 2). Second, the magnitude of Q and the decay rate constant of OFF I decreased more steeply with decreasing temperature than those of the fast components (author's unpublished results). Third, the waveform of I was closely associated with that of calcium release monitored simultaneously with antipyrylazo III (Hui, 1994, and manuscript in preparation).

Origin of Q

The I component reported in this paper is similar to the slow charge movement component observed by Pape et al. (1996). They proposed that calcium release exerts a negative feedback to slow the movement of Q. So, in their scheme, the slow charge movement is generated by the retardation of part of I. In Fig. 2 of Hui & Chen (1997), a diminution of the I hump was apparent when Ca²⁺ diffused into the myoplasm. That figure also showed a growth of I. Thus, the Q lost could very well be converted to Q. In other words, Q and Q could be two kinetic isoforms of the same physical entity. An interconversion of Q between two kinetic isoforms by ryanodine was also suggested by Huang (1996). However, under the conditions of the experiments described here, the < 20 nC µF^-1 of Q lost was far too small to account for the > 100 nC µF^-1 of Q that appeared. This large amount of total charge when [Ca²⁺]_i was close to physiological level, although surprising, was also observed by Pape et al. (1996; see fibre 412911 in Table I of their paper) and, therefore, should not be viewed as an anomaly. Even their average value of 67 nC µF^-1 was significantly larger than that from fibres containing a negligible [Ca²⁺]_i. Considering the fact that their charge movement was measured in a gluconate solution, which suppressed Q (Hui & Chen, 1995), their difference was even more substantial. Thus, in addition to the amount converted from Q, a major fraction of Q must be generated from a different unknown origin. Since no I exists when the [Ca²⁺]_i is negligible and the SR is depleted of releasable Ca²⁺, it is tempting to hypothesize that part of I could be generated by calcium release from the SR. If I is generated by calcium release, and since I precedes I, then I cannot be generated by calcium release, as proposed in the 'feedback hypothesis' (Pizarro, Csernoch, Uribe, Rodriguez & Rios, 1991; Shirokova, Pizarro & Rios, 1994).

	REFERENCES

Top Abstract Introduction Methods Results Discussion References

Adrian, R. H. & Peres, A. R. (1977). A gating signal for the potassium channel? Nature 267, 800-804.

[Medline] Adrian, R. H. & Peres, A. R. (1979). Charge movement and membrane capacity in frog muscle. The Journal of Physiology 289, 83-97.

[Abstract] Akasu, T. A., Nishimura, T. & Tokimasa, T. (1990). Calcium-dependent chloride current in neurones of the rabbit pelvic parasympathetic ganglia. The Journal of Physiology 422, 303-320.

[Abstract] Chandler, W. K. & Hui, C. S. (1990). Membrane capacitance in frog cut twitch fibers mounted in a double Vaseline-gap chamber. Journal of General Physiology 96, 225-256.

[Abstract] Chandler, W. K., Rakowski, R. F. & Schneider, M. F. (1976). A non-linear voltage-dependent charge movement in frog skeletal muscle. The Journal of Physiology 254, 245-283.

[Abstract] Evans, M. G., Marty, A., Tan, Y. P. & Trautmann, A. (1986). Blockage of Ca-activated Cl conductance by furosemide in rat lacrimal glands. Pflügers Archiv 406, 65-68.

[Medline] Godt, R. E. & Maughan, D. W. (1988). On the composition of the cytosol of relaxed skeletal muscle of the frog. American Journal of Physiology 254, C591-604.

[Medline] Huang, C. L.-H. (1981). Dielectric components of charge movements in skeletal muscle. The Journal of Physiology 313, 187-205.

[Abstract] Huang, C. L.-H. (1982). Pharmacological separation of charge movement components in frog skeletal muscle. The Journal of Physiology 324, 375-387.

[Medline] Huang, C. L.-H. (1991). Separation of intramembrane charging components in low-calcium solutions in frog skeletal muscle. Journal of General Physiology 98, 249-263.

[Abstract] Huang, C. L.-H. (1994). Kinetic separation of charge movement components in intact frog skeletal muscle. The Journal of Physiology 481, 357-369.

[Abstract] Huang, C. L.-H. (1996). Kinetic isoforms of intramembrane charge in intact amphibian striated muscle. Journal of General Physiology 107, 515-534.

[Abstract] Hui, C. S. (1983a). Pharmacological studies of charge movement in frog skeletal muscle. The Journal of Physiology 337, 509-529.

[Abstract] Hui, C. S. (1983b). Differential properties of two charge components in frog skeletal muscle. The Journal of Physiology 337, 531-552.

[Abstract] Hui, C. S. (1991a). Comparison of charge movement components in intact and cut twitch fibers of the frog: Effects of stretch and temperature. Journal of General Physiology 98, 287-314.

[Abstract] Hui, C. S. (1991b). Factors affecting the appearance of the hump charge movement component in frog cut twitch fibers. Journal of General Physiology 98, 315-347.

[Abstract] Hui, C. S. (1994). A component of charge movement temporally coupled to the calcium release signal. Biophysical Journal 66, A340.

Hui, C. S. & Chandler, W. K. (1990). Intramembranous charge movement in frog cut twitch fibers mounted in a double Vaseline-gap chamber. Journal of General Physiology 96, 257-297.