POTENTIATION OF CAFFEINE-INDUCED
CONTRACTURE BY RAISING EXTRACELLULAR
POTASSIUM IN FROG SKELETAL MUSCLE^?*

CHEN　KE-YING，ZHU　PEI-HONG^**
(Unit of Cell Signal Transduction, Shanghai Institute of Physiology,
Chinese Academy of Sciences, Shanghai 200031)

ABSTRACT　　The effect of raising extracellular potassium ([K⁺]_O) on caffeine contracture was inves～tigated, using small bundles dissected from frog anterior tibialis muscle. Elevating [K⁺]_O from the control of 2 mmol/L to 10 or 25 mmol/L significantly potentiated the contracture induced by 3 mmol/L caffeine. The potentiation represented by P_KC/P_C, where P_KC and P_C are the peak tension of the caffeine contracture evoked in high and normal [K⁺]_O respectively, was dependent on [K⁺]_O and the duration of conditioning high K⁺ exposure. With 10 mmol/L [K⁺]_O, the potentiation was gradually increased by prolonging conditioning exposure up to 10 min. On the contrary, with 25 mmol/L [K⁺]_O the potentiation reached a maximum within only 1 min, and then subsided to the control. These different time courses of P_KC/P_C could not be accounted for by high K⁺ induced depolarization, but were in general consistence with the time courses of the change in myoplasmic free calcium induced by corresponding high [K⁺]_O^[10]. It is suggested that, at least in frog skeletal muscle, the high [K⁺]_O induced potentiation of caffeine contracture is mainly due to an increase of myoplasmic free calcium.
Key words: skeletal muscle; ryanodine receptor; caffeine; high potassium; contracture

提高胞外钾引起的蛙骨骼肌咖啡因挛缩增强^*

陈克樱  朱培闳**

摘　　要

用蛙胫前肌小束为材料, 研究了提高胞外钾[K⁺]_O对咖啡因挛缩的作用。 [K⁺]_O从2 mmol/L提高到10或25 mmol/L, 由3 mmol/L咖啡因引起的挛缩明显增强。 以P_KC/P_C (P_KC和P_C分别为在高钾和正常钾条件下的咖啡因挛缩)表示的咖啡因挛缩增强, 依赖[K⁺]_O和高钾作用时间。 随着10 mmol/L [K⁺]_O作用时间延长, 直至10 min, 增强逐渐增加。 但是, 25 mmol/L [K⁺]_O作用1 min时增强达到最大, 然后下降到对照。 P_KC/P_C变化时程不能用高钾引起的去极化解释, 而与由相似[K⁺]_O引起的胞浆自由钙变化时程相符。 提示, 至少在蛙骨骼肌, 高钾引起的咖啡因挛缩增强主要是由胞浆自由钙升高引起的。
关键词: 骨骼肌； 钙释放通道； 咖啡因； 高钾； 挛缩
学科分类号: Q455

　　Caffeine can directly act on ryanodine receptors/calcium release channels (RyRs) to release calcium ions from the sarcoplasmic reticulum of skeletal muscle fibres and to produce contracture^[1]. Recently, it has been shown in mammalian skeletal muscle that the caffeine contracture can be potentiated by raising extracellular potassium concentration ([K⁺]_O)^[2,3].
　　It is known that the potentiation of caffeine contracture results from increased release of Ca²⁺ from the intracellular calcium stores, rather than change of calcium sensitivity of contractile proteins^[3]. But, the mechanism underlying the increased release of Ca²⁺ is still unclear. It is well established that the gating of RyRs in skeletal muscle fibres is regulated by the potentials across the transverse tubule membrane as well as by calcium ions^[4]. Several studies suggested that calcium release through caffeine sensitive pathways is controlled by the membrane potential^[3,5]. On the other hand, the elevated myoplasmic 　free calcium 　([Ca²⁺]_i) 　produced by high [K⁺]_O is 　thought 　to be 　responsible 　for 　thispotentiation^[2].
　　The predominant isoform of RyRs in mammalian skeletal muscle is RyR1 (skeletal muscle isoform), while other (1～3)% belongs to RyR3 (brain isoform). However, the skeletal muscle in nonmammalian vertebrate, including amphibian, contains other two isoforms: RyRα and RyRβ. They have been recognized being homologous to RyR1 and RyR3, respectively. However, RyRα and RyRβ in non～mammalian vertebrate skeletal muscle are present in approximately equal amounts^[6]. At present, it is unclear whether or not the diversity of the RyR isoforms serves different functions. Considering that the composition of RyR isoforms in mammalian and nonmammalian vertebrate skeletal muscle is so different, it would be interesting to investigate if caffeine contracture can be potentiated by raising [K⁺]_O in frog skeletal muscle. After finding its presence, it was attempted to see whether or not the change of membrane potential or [Ca²⁺]_i is responsible for the potentiation of caffeine contracture.

1　MATERIALS AND METHODS
1.1　Preparation and solution 　　The experiments were performed on small bundles of anterior tibialis muscle dissected out from pithed frog Rana nigromaculata at a temperature of 15℃. Due to the limited sensitivity of transducer, the bundle used in this study comprised about 10～20 fibres.
　　The composition of Ringer′s solution was (in mmol/L): NaCl 120, KCl 2, CaCl₂ 1.8, sucrose 10, and HEPES 4. The solution was titrated to pH 7.2 with NaOH. In high K⁺ medium, potassium was equivalently substituted for sodium. In order to keep the product of [K⁺] and [Cl^－] constant, Cl^－ was partially replaced by CH₃HSO^?₃. Caffeine was dissolved in the Ringer′s solution or in the high K⁺ medium.
1.2　Contractile assessment and experimental protocols　　The dissected muscle bundle was put in a perfusion chamber with one end fixed by a clamp and the other end connected to the lever of a transducer. The preparation was then perfused with Ringer′s solution at a rate of about 2 ml/min. After being adjusted to an optimal length, the preparation was stimulated for a few minutes with single pulses (pulse duration 1 ms) or repetitive pulses (0.5 s train, 50 Hz) via a pair of parallel platinum electrodes, and stable twitch (Pt) and tetanus tensions (P_O) were obtained. Afterwards, the preparation was treated with one of the following three protocols.
　　(1) As a control, the preparations were repeatedly exposed to caffeine. During the interval (20 min) between caffeine exposures, the preparation was perfused with Ringer′s solution, and no electrical stimuli were applied. To check the condition of the preparation, a few stimuli were delivered just before each caffeine exposure. The peak tension of caffeine contracture and the tetanus tension in each caffeine exposure were designated as P_CX and P_OX, respectively, where X represents the time of caffeine exposure. (2) Different from (1), the second caffeine contracture was evoked in high [K⁺]_O, 10 or 25 mmol/L. The peak tension of caffeine contracture evoked in normal and raised [K⁺]_O was respectively represented by P_C and P_KC. Because the preparation used in the present study was relatively thick, the factor of diffusion should be considered. Therefore, the preparation was perfused with high potassium medium in advance of caffeine exposure for various times. As a result, the potentiation of caffeine contracture was found to be dependent on the duration of conditioning high K⁺ exposure. (3) Different from (2), the time of conditioning high K⁺ exposure was fixed at 2 min, and the preparation was perfused with Ringer′s solution for 1～10 min before the second caffeine exposure. Moreover, the second caffeine contracture was evoked in normal [K⁺]_O. The effect of conditioning high K⁺ exposure on caffeine contracture was represented by P_C2 /P_C1, where P_C2 and P_C1 are the peak tension of caffeine contracture with and without conditioning high K⁺ exposure, respectively.
　　The sensitivity to caffeine varied considerably among the muscles isolated from different frogs and even among different bundles dissected from one muscle, perhaps due to the presence of different types of the fibres. It is known that caffeine sensitivity of different types of the fibres is different^[2]. Therefore, choosing proper caffeine concentration is essential. After trials, 3 mmol/L caffeine was adopted throughout this study.
1.3　Data analysis　　Since considerable variation of caffeine sensitivity was present among preparations, P_C and P_KC are represented relatively by P_C/P_O and P_KC/P_C, where P_O is the tetanus tension, and P_KC and P_C are the peak tension of caffeine contracture initiated in raised and normal [K⁺]_O, respectively. Student′s t test was used for comparison of different groups.

2　RESULTS
2.1　Effect of repeated caffeine exposures
　　It has been shown that the caffeine contracture was gradually reduced by repeated caffeine exposures in frog skeletal muscle, but this reduction varied under different experimental conditions^[7]. As a control, the experiments were carried out, using protocol (1).
　　To evoke caffeine contracture, the preparation was usually perfused with a medium containing 3 mmol/L caffeine until the contracture reached a peak, and then returned to Ringer′s solution. The time to peak was variable among preparations, and increased with repeated caffeine exposures. Typically, it was about 2 min in the first caffeine exposure.
　　Caffeine exposure did not significantly affect the tetanus tension. P_O2/P_O1 was 0.96±0.08 (mean±S.D., n=12). Even after two caffeine exposures, only few preparations showed a small depression of the tetanus. Consequently, P_O3/P_O1 was reduced to 0.89±0.10 (n=11). But, the effect of the first caffeine exposure on the following caffeine contractures was considerably variable. An evident depression as well as potentiation of P_C2 was seen in some bundles, but more preparations did not show significant change in P_C2. In 12 preparations, P_C2/P_C1 ranged between 0.25 and 1.80, and had a mean of 1.00±0.47. Since no obvious correlation was observed between P_C2/P_C1 and P_O2/P_O1, the depression of the second caffeine contracture in some preparations does not seem to result from some deterioration of the preparation. However, after two caffeine exposures, the caffeine contracture was consistently depressed. P_C3/P_C1 was reduced to 0.46±0.30 (n=11).
2.2　Potentiation of caffeine contracture by raising [K⁺]_O
　　Using protocol 2, the effect of raising [K⁺]_O on caffeine contracture was examined. 10 mmol/L K⁺ usually did not produce any detectable mechanical responses, while a transient contracture was induced by 25 mmol/L K⁺ exposure. It can be seen that caffeine contracture was clearly potentiated after either 10 or 25 mmol/L K⁺ conditioning exposure of 1 min (Fig.1). It is also evident that, after two caffeine exposures, caffeine contracture was significantly depressed or almost abolished.

Fig.1　Representative records of the effect of raising [K⁺]_O on caffeine contracture of small bundles dissected from frog anterior tibialis muscle
The left and right panels indicate contractures evoked by 3 mmol/L caffeine in normal [K⁺]_O, i.e. the pre- and post-control, respectively. The middle one was caffeine contracture evoked in raised [K⁺]_O, 10 mmol/L (A) and 25 mmol/L (B). Two different preparations. The duration of conditioning high K⁺ exposure: 1 min. An interval of 20 min was allowed between runs. 3C: 3 mmol/L caffeine; 10 or 25 K: 10 or 25 mmol/L [K⁺]_O; R: Ringer′s solution.

　　More interestingly, it was found that the extent of the potentiation was dependent on the duration of conditioning high K⁺ exposure, as shown in Fig.2. The potentiation in this figure is represented by P_KC /P_C, where P_KC and P_C are the peak tension of caffeine contracture evoked in raised and normal [K⁺]_O, respectively. Since the caffeine contracture was significantly depressed after two caffeine exposures (Fig.1), the post control was not taken into account, although observed. Only one P_KC /P_C was obtained from each preparation. The control value of P_KC /P_C was 1.00±0.47 (n=12), which was obtained with protocol (1), i. e. P_C2/P_C1. With 10 mmol/L K⁺, the potentiation was gradually increased with conditioning high K⁺ exposure. But, due to an unknown reason, the potentiation at 5 min was statistically insignificant. Although the potentiation induced by 25 mmol/L K⁺ conditioning exposure was also dependent on the duration, the time dependence was obviously different from that seen with 10 mmol/L K⁺. In the presence of 25 mmol/L K⁺, the potentiation reached a maximum within 1 min, and then declined gradually. It completely disappeared at 10 min.

Fig.2　Effect of the duration of 10 mmol/L (○) or 25 mmol/L (●) [K⁺]_O
conditioning exposure (protocol 2) on the potentiation (P_KC/P_C) P_KC and P_C are the peak tension of 3 mmol/L caffeine contracture evoked in raised and normal [K⁺]_O, respectively. The abscissa is the duration of conditioning
high K⁺ exposure. The data point with the bar represents ±s. The figure near each data point expresses the number of the examined preparations. The control P_KC/P_C (□) was 1.00±0.47 (n=12), which was obtained with protocol 1, i.e. P_C2/P_C1. ^*　P<0.05,^*　*　P<0.01,^*　*　*　P<0.001 vs control.

　　As described above, P_O was not changed after one and even two caffeine exposures. Therefore, the potentiation can be estimated by P_KC /P_O, where P_O is the tetanus tension obtained before the first caffeine contracture. The time course of the change of P_KC /P_O was basically similar to that of P_KC /P_C (data unshown).
　　With 25 mmol/L K⁺ conditioning exposure, caffeine contracture may be superimposed on potassium contracture. However, due to inactivation of excitation-contraction coupling^[8], the contracture induced by 25 mmol/L K⁺ usually returned to the baseline level within about 2 min in spite of raised K⁺. Consequently, the K⁺ contracture subsided at the time when the caffeine contracture reached its peak, i.e. in about 2 min. Therefore, the estimation of P_KC did not take the residual potassium contracture into account.
2.3　Effect of conditioning high K⁺ exposure on caffeine contracture
　　To examine if caffeine contracture evoked in normal [K⁺]_O is affected by conditioning high K⁺ exposure, the following experiments were performed with protocol (3). The time of conditioning high K⁺ exposure was fixed at 2 min, and the preparation was perfused with Ringer′s solution for between 1 and 10 min before the second caffeine exposure. Moreover, the second caffeine contracture was evoked in normal [K⁺]_O.
　　Representative records shown in Fig.3 indicate that, following conditioning high K⁺ exposure, caffeine contracture evoked in normal [K⁺]_O was significantly depressed. In this case, the preparations had been perfused with Ringer′s solution for 2 min before the second caffeine exposure. The time course of the recovery of caffeine contracture following conditioning high K⁺ exposure is shown in Fig.4, indicating that caffeine contracture could recover completely after 10 mmol/L K⁺ of 2 min, while the depression induced by 25 mmol/L K⁺ of 2 min was irreversible.

Fig.3　Representative records of the effect of conditioning high K⁺ exposure (protocol 3) on caffeine contracture evoked in normal [K⁺]_O
The left and right panel indicates caffeine contracture without and with conditioning high K⁺ exposure, respectively. K⁺ concentration: 10 mmol/L (A) and 25 mmol/L (B). Two different preparations. The duration of conditioning high K⁺ exposure was fixed at 2 min. In this case, the preparation was perfused with Ringer′s solution for 2 min before the second caffeine exposures. An interval of 20 min was allowed between runs. 3C: 3 mmol/L caffeine; 10 or 25 K: 10 or 25 mmol/L [K⁺]_O; R: Ringer′s solution.

Fig.4　Time course of the recovery of caffeine contracture following 10 mmol/L (○) or 25 mmol/L (●) [K⁺]_O conditioning exposure (protocol 3)
The duration of conditioning high K⁺ exposure was fixed at 2 min. The abscissa represents the perfusion duration with Ringer′s solution before the second caffeine exposure. P_C2 and P_C1 indicate the peak tension of caffeine contracture with and without conditioning high K⁺ exposure, respectively. The data point with the bar represents ±s. The figure near each data point expresses the number of the examined preparations. The control P_C2/P_C1 (□) was obtained with protocol 1. ^*　P<0.05, ^*　*　P<0.01, vs control.

3　DISCUSSION
　　The present study clearly indicates that caffeine contracture in frog skeletal muscle could be potentiated by raising extracellular potassium as in mammalian skeletal muscle^[2,3]. In comparison with the latter^[2,3,5], the extent of the potentiation was significantly less in frog skeletal muscle. It is unclear whether or not it is due to the different isoforms of RyRs expressed in mammalian and frog skeletal muscle^[6]. The present study has shown a time dependence of the potentiation in frog skeletal muscle. Whether this is also present in mammalian skeletal muscle remains to be verified.
　　Since high K⁺-induced membrane depolarization in such a small bundle is a relatively fast process and a depolarization would continue in the presence of high K⁺, obviously the alteration of the membrane potential can not account for the different time courses of P_KC/P_C with 10 and 25 mmol/L K⁺.
　　Two potassium concentrations, 10 and 25 mmol/L, were used in the present study. It has been shown in frog skeletal muscle fibres that myoplasmic free calcium concentration [Ca²⁺]_i began to increase, when external potassium was increased to higher than 5 mmol/L^[9]. At 10 mmol/L K⁺, [Ca²⁺]_i gradually rose and maintained at a high level. At a potassium concentration higher than 16 mmol/L, [Ca²⁺]_i increased to a peak with a faster rate, and then returned to a baseline level or declined to a plateau. The time course of P_KC /P_C at 10 and 25 mmol/L [K⁺]_O displayed in Fig.2 was generally consistent with that of the change of [Ca²⁺]_i induced by corresponding [K⁺]_O, suggesting the role of high K⁺ induced increase of [Ca²⁺]_i in the potentiation of caffeine contracture. This deduction is in harmony with the result obtained from mammalian skeletal muscle by pharmacological means^[2]. More recently, it has been indicated that caffeine-induced release of calcium ions was not affected by the holding potential^[10]. The present result provides further evidence to support the role of [Ca²⁺]_i in regulating caffeine-sensitive pathway.
　　In the presence of caffeine, the phasic and tonic components of calcium release were similarly potentiated^[10]. In addition, under this condition membrane potential was still operational to control the calcium release, but could not alter caffeine-induced permeability. It is suggested that the gating of ryanodine receptors/calcium release channels is independently regulated by caffeine and the voltage sensor^[10]. Therefore, voltage may not directly affect caffeine-sensitive calcium release channel, if high K⁺-induced depolarization plays some roles in the potentiation of caffeine contracture.
　　This study evidently demonstrated a depression of caffeine contracture following conditioning high K⁺ exposure (Fig.4). This depression was dependent on [K⁺]_O and probably the duration of conditioning high K⁺ exposure. Under certain conditions, e.g. 25 mmol/L K⁺ conditioning exposure for 2 min, this depression became irreversible (Fig.4), indicating that conditioning high K⁺ exposure may somehow reduce the sensitivity of ryanodine receptors/calcium release channels to caffeine. At present, we are ignorant of its mechanism. To what extent the time course of P_KC/P_C obtained with protocol 2 (Fig.2) is influenced by this effect is still obscure.

^*?This study was supported by a grant from National Natural Science Foundation of China (No.39670242).
^**? To whom correspondence should be addressed. Tel: 021-64370080, ext.147; Fax: 86-21-64332445; E-mail: phzhu@server.shcnc.ac.cn
^*国家自然科学基金资助 (No.39670242)
**联系作者. 　Tel: 021-64370080, ext.147; Fax: 86-21-64332445; E-mail: phzhu@server.shcnc.ac.cn

作者单位:陈克樱  朱培闳　中国科学院上海生理研究所细胞信号转导组，　 上海　200031

REFERENCES

　[1]　Rousseau E, Ladine J, Liu QY, et al. Activation of the Ca²⁺ release channel of skeletal muscle sarcoplasmic reticulum by caffeine and related compounds. Arch Biochem Biophys, 1988, 267: 75～86.
　[2]　Gallant EM, Lentz LR, Taylor SR. Modulation of caffeine contractures in mammalian skeletal muscles by va～ri～a～tion of extracellular potassium. J Cell Physiol, 1995, 165: 254～260.
　[3]　Hidalgo J, Niemeyer MI, Jaimovich E. Voltage control of calcium transients elicited by caffeine and tetracaine in cultured rat muscle cells. Cell Calcium, 1995, 18: 140～154.
　[4]　Coronado R, Morrissette J, Sukhareva M, et al. Structure and function of ryanodine receptors. Am J Physiol, 1994, 266: C1485～C1504.
　[5]　Suda N, Penner R. Membrane repolarization stops caffeine-induced Ca²⁺ release in skeletal muscle cells. Proc Natl Acad Sci USA, 1994, 91: 5725～5729.
　[6]　Franzini-Armstrong C, Protasi F. Ryanodine receptors of striated muscles: a complex channel capable of multiple interactions. Physiol Rev, 1997, 77: 699～729.
　[7]　Raymond G, Potreau D, Cognard C, et al. Stimulation frequency and external ionic composition control the re～priming of caffeine-induced contractures in frog skeletal muscle. Can J Physiol Pharmacol, 1985, 65: 704～710.
　[8]　　Dulhunty AF. Activation and inactivation of excitation-contraction coupling in rat soleus muscle. J Physiol (Lond), 1991, 439: 605～626.
　[9]　Snowdowne KW. Subcontracture depolarization increases sarcoplasmic ionized calcium in frog skeletal muscle. Am J Physiol, 1985, 248: C520～C526.
　[10]　Shirokova N, Rios E. Activation of Ca²⁺ release by caffeine and voltage in frog skeletal muscle. J Physiol (Lond), 1996, 493: 317～339.　　

Received 1998-06-08　　Revised 1998-09-07