Calcium Oscillations and the 5th UK Calcium Signalling Conference
Ca2+-induced Ca2+ release in cardiac and smooth muscle cells T. Kamishima1 and J.M. Quayle Department of Human Anatomy and Cell Biology, University of Liverpool, Sherrington Buildings, Ashton Street, Liverpool, L69 3GE, U.K.
Abstract Ca2+ influx across plasma membranes may trigger Ca2+ release by activating ryanodine-sensitive receptors in the sarcoplasmic reticulum. This process is called Ca2+ -induced Ca2+ release, and may be important in regulating cytosolic Ca2+ concentration ([Ca2+ ]i ). In cardiac cells, the initial [Ca2+ ]i increase, caused by L-type Ca2+ current, is profoundly amplified with Ca2+ release. The synchronized opening of several ryanodinesensitive Ca2+ -releasing channels was detected as discreet and highly localized Ca2+ elevation, and termed as a ‘Ca2+ spark’. A Ca2+ spark is under local control of an L-type Ca2+ channel, and therefore a Ca2+ spark does not normally trigger subsequent Ca2+ sparks in the neighbouring area. In smooth muscle cells, the importance of Ca2+ -induced Ca2+ release in elevating [Ca2+ ]i appears to differ among preparations and species. Significant elevation in [Ca2+ ]i during depolarization was attributed to Ca2+ release in some smooth muscle cells, but not in others. Ca2+ sparks are also identified in smooth muscle cells, and may play a role as functional elementary events for Ca2+ -induced Ca2+ release. At rest, Ca2+ sparks may be also important in regulating smooth muscle membrane potential. Ca2+ sparks occurring at rest do not raise global [Ca2+ ]i , but trigger spontaneous transient outward currents (STOCs) or spontaneous transient inward currents (STICs), the former producing hyperpolarization; the latter, depolarization. Thus there may be multiple facets for Ca2+ -induced Ca2+ release in regulating the contractile status of smooth muscle cells.
Ca2+ -induced Ca2+ release in cardiac myocytes Physiological importance of Ca2+ -induced Ca2+ release was first suggested in cardiac myocytes [1,2]. Though the term is nowadays used somewhat more broadly, it originally described the observation that an increase in intracellular Ca2+ concentration ([Ca2+ ]i ), caused by an L-type Ca2+ current, was amplified by Ca2+ release from ryanodine-sensitive receptors in the sarcoplasmic reticulum. Despite some differences between species, a predominant part of the increase in [Ca2+ ]i during cardiac excitation-contraction coupling comes from Ca2+ release. More recently, two additional Ca2+ -influx pathways that may induce Ca2+ release were reported. First, T-type Ca2+ current was shown to trigger Ca2+ release [3,4]. Secondly, Ca2+ influx through Na+ /Ca2+ exchange seems to induce Ca2+ release when the membrane potential becomes positive to exchanger equilibrium potential, either by an increase in subsarcolemmal Na+ concentration ([5,6]; but see [7]) or by membrane depolarization [8,9]. However, Ca2+ influx through T-type Ca2+ channels and Na+ /Ca2+ exchangers seems substantially less effective in triggering Ca2+ release when compared with Ca2+ influx through L-type Ca2+ channels (reviewed in [10]). Thus the physiological role of non-L-type Ca2+ influx in evoking Ca2+ release remains unclear. It has been suggested Key words: Ca2+ channel, Ca2+ spark, Fura-2, ion channel, ryanodine receptor, sarcoplasmic reticulum. Abbreviations used: [Ca2+ ]i , intracellular calcium concentration; STOC, spontaneous transient outward current; STIC, spontaneous transient inward current. 1 To whom correspondence should be addressed (e-mail
[email protected]).
that the role of Ca2+ entry through Na+ /Ca2+ exchange is to refill the sarcoplasmic reticulum [8]. There may be little surprise that inhibition of Ca2+ release alone is sufficient in abolishing cell contraction [4], as the amount of Ca2+ supplied by Ca2+ release generally far exceeds that by Ca2+ current (e.g. see [11]). A rather unexpected observation was that Ca2+ transients and contractions were seriously compromised in cardiac cells obtained from rats with cardiac hypertrophy or heart failure, even though both Ca2+ influx and Ca2+ release mechanisms appear unaltered in these cells [12]. The reason why apparently normal Ca2+ current failed to activate seemingly normal ryanodinesensitive receptors appears that the sustained hypertension increased the distance between these channels [12]. The hypothesis that subtle changes in cellular microarchitecture are sufficient in uncoupling Ca2+ -induced Ca2+ release would predict that L-type Ca2+ channels are very closely located to the ryanodine-sensitive Ca2+ -releasing channels in healthy cells. This notion is supported by two lines of evidence. First, high-resolution immunocytochemical studies where Ltype Ca2+ channels and ryanodine receptors are visualized by fluorescent probes showed that these channels are indeed colocalized [13]. Secondly, the ryanodine receptors appear to detect the size of Ca2+ current passing through each L-type Ca2+ channel, suggesting a close proximity of these molecules [14]. Ca2+ release triggered by a given total Ca2+ influx, the product of unitary current and number of opening channels, was larger when the depolarization was smaller. The single Ca2+ current size is larger when the membrane potential is more negative due to increased electrochemical driving force C 2003
Biochemical Society
943
944
Biochemical Society Transactions (2003) Volume 31, part 5
on Ca2+ ions. A larger single Ca2+ channel current at more negative potential may be sensed by the ryanodine-sensitive receptors that would in turn release more Ca2+ [14]. Interestingly, rather inefficient coupling between T-type Ca2+ channels or Na+ /Ca2+ exchanger with ryanodine-sensitive receptors may be explained by an opposite hypothesis. These Ca2+ entry molecules may not be closely positioned to the ryanodine-sensitive Ca2+ -releasing channels, resulting in poor efficacy in triggering Ca2+ release (reviewed in [10]). The very nature of Ca2+ -induced Ca2+ release raises an important question. How do cardiac myocytes produce graded responses in [Ca2+ ]i increase and contraction? Since the initial elevation in [Ca2+ ]i caused by triggering Ca2+ current is generally amplified many times with the subsequent Ca2+ release, it is conceivable that, once triggered, Ca2+ induced Ca2+ release may become self-regenerative through a positive feedback loop that would be not only ‘all-ornone’-like, but also inherently rather unstable [11,15]. On the other hand, the local control theories of Ca2+ -induced Ca2+ release predict that localized Ca2+ release does not normally propagate, and therefore Ca2+ influx is of crucial importance in triggering Ca2+ release [11,14,15]. The vital evidence to support the local control theories came from the results showing Ca2+ transients occurring in discreet spots of a cell visualized by confocal microscopy [11,16,17]. Ca2+ sparks are highly localized Ca2+ transients within a cell, first observed as spontaneous Ca2+ release from ryanodine-sensitive channels [16]. Ca2+ sparks are functional units for Ca2+ release resulting from virtually synchronous activation of a cluster of several ryanodine-sensitive Ca2+ -releasing channels [11,17]. At rest, the occurrence of Ca2+ sparks is rare and stochastic, and therefore its activation does not result in propagating Ca2+ release. When depolarization initiates the excitation– contraction coupling, however, spatially uniform Ca2+ transients are seen throughout the cell, because numerous Ca2+ sparks occur simultaneously. A body of evidence has overwhelmingly suggested that a Ca2+ spark is triggered by an opening of single L-type Ca2+ channel [18]. The final proof to link single L-type Ca2+ current with a Ca2+ spark was provided by an elegant study which detected electrical and optical events within cell-attached patches [19]. Ca2+ sparks are often preceded by a Ca2+ ‘sparklet’, a local Ca2+ increase caused by a single L-type Ca2+ -channel opening [19]. The comparison of Ca2+ spark magnitude and Ca2+ sparklet size permitted the estimation that a single opening of L-type Ca2+ channel results in the simultaneous activation of about four to six ryanodine-sensitive Ca2+ releasing channels [19]. Thus cardiac Ca2+ -induced Ca2+ release and its local control mechanism, studied at the skinnedfibre level 18 years ago [1,2], can now be explained at the molecular level.
Ca2+ -induced Ca2+ release in smooth muscle cells As in the case of cardiac myocytes [1,2], pioneering studies of Ca2+ -induced Ca2+ release in smooth muscle came C 2003
Biochemical Society
from skinned-fibre experiments. Iino suggested that, since activation of Ca2+ -induced Ca2+ release in guinea pig taenia caeci required more than 1 µM Ca2+ , it is unlikely to play a primary role in physiological Ca2+ mobilization [20]. Nonetheless, the presence of Ca2+ -induced Ca2+ release in smooth muscles was often implied from the observations where an increase in [Ca2+ ]i was noted with caffeine application. Caffeine is an agent that increases the sensitivity of the ryanodine-sensitive receptors to Ca2+ [20], so that Ca2+ -induced Ca2+ release occurs at the resting [Ca2+ ]i . A caffeine-induced Ca2+ transient, however, merely indicates the presence of ryanodine-sensitive Ca2+ releasing channels. It does not prove that Ca2+ -induced Ca2+ release operates during depolarization. Furthermore, great care has to be taken in studying Ca2+ homoeostasis, since an intervention meant to modify one Ca2+ regulatory pathway often unwittingly results in modifying additional Ca2+ -handling mechanisms. For example, thapsigargin is a commonly used blocker for sarcoplasmic reticulum Ca2+ uptake and therefore a functional blocker of Ca2+ -induced Ca2+ release. However, thapsigargin also blocks L-type Ca2+ channels [21], making data interpretation complicated. To provide less ambiguous evidence, therefore, experiments need to be carried out in such a way that conditions are rigorously controlled and possible variables are measured. To unequivocally support the hypothesis that Ca2+ induced Ca2+ release plays a physiological role in smooth muscle, an increase in [Ca2+ ]i has to be significantly reduced in the presence of ryanodine with a given Ca2+ entry. Direct comparison of Ca2+ entering into the cell and Ca2+ appearing in the cell is possible by simultaneously detecting membrane current and Ca2+ transient. Since the current through L-type Ca2+ channels is almost exclusively carried by Ca2+ ions, the time-integrated Ca2+ current reflects the charge entry that can be converted into the expected increase in [Ca2+ ]i if the cell volume is known (for details, see [22]). Generally, only a few percent of Ca2+ ions entering the cell appears as free Ca2+ , as the vast majority of Ca2+ ions are quickly buffered (e.g. [22]). Therefore the Ca2+ -buffering power can be calculated as the ratio of expected increase in [Ca2+ ]i divided by the actually measured increase in [Ca2+ ]i reported by Ca2+ indicators such as Fura-2 ([22]; also see [23]). Provided that the Ca2+ buffering powers are calculated in the absence of substantial changes in Ca2+ removal rate [22], they can offer one of the most rigorous tests for the presence or absence of Ca2+ -induced Ca2+ release. If Ca2+ -induced Ca2+ release is important, the Ca2+ -buffering power of ryanodine-treated cells should be significantly larger than that of the control cells [24]. Relatively small numbers of such quantitative examination of Ca2+ -induced Ca2+ release have been carried out so far, and the results are mixed. While ryanodine treatment significantly increased Ca2+ buffering power in rat superior cerebral artery smooth muscle cells [24], it failed to do so in toad stomach [25], guinea-pig coronary artery [26], rat portal vein [22], equine airway [27] and rat femoral artery [23] smooth muscle cells. Interestingly, in rat superior cerebral artery smooth muscle cells where a
Calcium Oscillations and the 5th UK Calcium Signalling Conference
significant Ca2+ -induced Ca2+ release was detected, a smaller depolarization was more efficient in triggering Ca2+ release [24], showing voltage-dependent variable gain first described in cardiac cells [14]. Like cardiac myocytes, ryanodinesensitive Ca2+ releasing channels appear to be co-localized with L-type Ca2+ channels in some smooth muscle cells [28], and therefore they may be able to detect unitary Ca2+ current size. It is conceivable that in the cells where no significant Ca2+ -induced Ca2+ release was detected [22,23,25–27] such co-localization may not exist. Other studies where voltage-clamp technique was used, but Ca2+ -buffering power was not calculated, also indicated that Ca2+ -induced Ca2+ release may be important in smooth muscle cells. Using guinea-pig urinary bladder myocytes, Ganitkevich and Isenberg [29] suggested that about 70% of the phasic component of [Ca2+ ]i increase evoked by depolarization is due to Ca2+ release. Also, Gr´egoire et al. [30] reported that voltage-dependent Ca2+ current in rat portal vein smooth muscle cells evoked Ca2+ release occurring after membrane repolarization. It was argued that, once triggered, Ca2+ -induced Ca2+ release in those cells became regenerative, so that Ca2+ release continues in the absence of triggering Ca2+ current [30]. Moreover, Ca2+ sparks were observed in guinea-pig vas deferens and urinary bladder [31] and rabbit urinary bladder [32] smooth muscle cells following depolarization. These results suggested that, like cardiac cells, Ca2+ sparks may consist of the functional unit for smooth muscle Ca2+ -induced Ca2+ release [31,32], although the coupling of L-type Ca2+ channels and ryanodine-sensitive receptors in these cells appears to be more loose than that seen in cardiac myocytes [31,32]. It should be noted, however, that Ca2+ sparks in cerebral artery smooth muscle cells observed at rest may be important in hyperpolarization and muscle relaxation [33]. Spontaneous Ca2+ sparks at rest do not contribute in raising [Ca2+ ]i , but produce concerted opening of Ca2+ -activated K+ channels (spontaneous transient outward currents, or STOCs), causing membrane hyperpolarization [33]. More recently, it has been reported that Ca2+ sparks also evoke spontaneous transient inward currents (STICs) in guinea-pig tracheal myocytes [34]. STICs are caused by the opening of Cl− channels, and will produce membrane depolarization [34]. In these cells, Ca2+ sparks may cause biphasic current in which the spontaneous transient outward current precedes the inward current (STOICs) [34]. Taken together [33,34], Ca2+ sparks in smooth muscle cells may be important in regulating membrane potential. Some of the discrepancies regarding the importance of Ca2+ -induced Ca2+ release certainly arise from the differences among preparations and species. However, there are two possibilities that Ca2+ -induced Ca2+ release could be inadvertently identified in smooth muscle cells. First, unlike cardiac myocytes, smooth muscle cells express Ca2+ permeable cation channels that could contribute in raising [Ca2+ ]i . For example, rat portal vein smooth muscle cells were reported to produce self-regenerative Ca2+ -induced Ca2+ release that occurs in the absence of triggering L-
type Ca2+ influx [30]. However, these cells are known to express Ca2+ -permeable cation channels that are activated by elevation in [Ca2+ ]i [35]. Increase in [Ca2+ ]i persisting during repolarization did not occur when the membrane potential was clamped at the equilibrium potential for Ca2+ [22]. Therefore this delayed [Ca2+ ]i elevation seems due to Ca2+ influx through cation channels, rather than propagating Ca2+ -induced Ca2+ release [22]. Secondly, it has been shown that the open probability of ryanodine-sensitive Ca2+ releasing channels is regulated by Ca2+ concentration in the sarcoplasmic reticulum [36,37]. Therefore Ca2+ release from ryanodine-sensitive channels may occur more easily when the sarcoplasmic reticulum is Ca2+ overloaded. Such events, however, may be more related to pathological conditions [36]. The effect of Ca2+ overload on contraction is also described in cardiac muscle [38]. Since smooth muscle cells are embedded in a matrix of connective tissues, dissociation of single smooth muscle cells normally requires enzyme treatments and some mechanical agitation (e.g., see [22]). It is not uncommon that the dissociated cells are somewhat more depolarized and contracted than those found in the tissue. The sarcoplasmic reticulum of these cells could easily be Ca2+ overloaded. There are still substantial gaps in our knowledge regarding smooth muscle Ca2+ -induced Ca2+ release. Though results from the work carried out using cardiac myocytes may be used in explaining some results [24], the properties of Ca2+ induced Ca2+ release in some smooth muscle cells may differ from that in the cardiac cells [32]. Furthermore, smooth muscle cells may express various cation channels that are permeable to Ca2+ [35]. One such cation channels is the P2X receptor that is activated by ATP [39]. It has been suggested that Ca2+ influx through P2X receptors trigger further Ca2+ release in rat portal vein myocytes [40]. However, P2X receptor-mediated rat mesenteric artery contraction was unaffected by a blocker of Ca2+ -induced Ca2+ release, suggesting that Ca2+ influx through P2X receptors does not trigger substantial Ca2+ release [41]. Similarly, the field stimulation that triggered P2X receptor-mediated Ca2+ transients neither evoked Ca2+ waves in rat mesenteric arteries [42] nor released Ca2+ in mouse vas deferens [43]. It is conceivable that, like L-type Ca2+ current, P2X current may or may not trigger further Ca2+ release depending on the preparation. In summary, Ca2+ -induced Ca2+ release is important in raising [Ca2+ ]i during cardiac excitation–contraction coupling. Defects in Ca2+ -induced Ca2+ release may underlie impaired cardiac function. Ca2+ -induced Ca2+ release also plays a crucial role in smooth muscle Ca2+ homoeostasis. It may directly contribute in raising [Ca2+ ]i , or indirectly modulate [Ca2+ ]i by regulating membrane potential. Therefore Ca2+ induced Ca2+ release is an integral part of Ca2+ homoeostasis in both cardiac and smooth muscle cells.
We thank the British Heart Foundation for support. T.K. is a British Heart Foundation Intermediate Fellow (FS/2000001). C 2003
Biochemical Society
945
946
Biochemical Society Transactions (2003) Volume 31, part 5
References 1 Fabiato, A. (1985) J. Gen. Physiol. 85, 247–289 2 Fabiato, A. (1985) J. Gen. Physiol. 85, 291–320 3 Sipido, K.R., Carmeliet, E. and Van de Werf, F. (1998) J. Physiol. (Cambridge, U.K.) 508, 439–451 4 Zhou, Z. and January, C.T. (1998) Biophys. J. 74, 1830–1839 5 Leblanc, N. and Hume, J.R. (1990) Science 248, 372–376 6 Lipp, P. and Niggle, E. (1994) J. Physiol. (Cambridge, U.K.) 474, 439–446 7 Sham, J.S.K., Cleemann, L. and Morad, M. (1992) Science 255, 850–853 8 Nuss, H.B. and Houser, S.R. (1992) Am. J. Physiol. 263, H1161–H1169 9 Vornanen, M., Shepherd, N. and Isenberg, G. (1994) Am. J. Physiol. 267, C623–C632 10 Bers, D.M. (2002) Nature (London) 415, 198–205 11 Cannell, M.B., Cheng, H. and Lederer, W.J. (1994) Biophys. J. 67, 1942–1956 12 Gomez, ´ A.M., Valdivia, H.H., Cheng, H., Lederer, M.R., Santana, L.F., Cannell, M.B., McCune, S.A., Altschuld, R.A. and Lederer, W.J. (1997) Science 276, 800–806 13 Carl, S.L., Felix, K., Caswell, A.H., Brandt, N.R., Ball, Jr, W.J., Vaghy, P.L., Meissner, G. and Ferguson, D.G. (1995) J. Cell Biol. 129, 673–682 14 Wier, W.G., Egan, T.M., Lopez-L ´ opez, ´ J.R. and Balke, C.W. (1994) J. Physiol. (Cambridge, U.K.) 474, 463–471 15 Stern, M.D. (1992) Biophys. J. 63, 497–517 16 Cheng, H., Lederer, W.J. and Cannell, M.B. (1993) Science 262, 740–744 17 Cannell, M.B., Cheng, H. and Lederer, W.J. (1995) Science 268, 1045–1049 18 Wier, W.G. and Balke, C.W. (1999) Circ. Res. 85, 770–776 19 Wang, S.-Q., Song, L.-S., Lakatta, E.G. and Cheng, H. (2001) Nature (London) 410, 592–596 20 Iino, M. (1989) J. Gen. Physiol. 94, 363–383 21 Rossier, M.F., Python, C.P., Burnay, M.M., Schlegel, W., Vallotton, M.B. and Capponi, A.M. (1993) Biochem. J. 296, 309–312 22 Kamishima, T. and McCarron, J.G. (1996) J. Physiol. 492, 61–74 23 Kamishima, T., Davies, N.W. and Standen, N.B. (2000) J. Physiol. (Cambridge, U.K.) 522, 285–295 24 Kamishima, T. and McCarron, J.G. (1997) J. Physiol. (Cambridge, U.K.) 501, 497–508
C 2003
Biochemical Society
25 Guerrero, A., Singer, J.J. and Fay, F.S. (1994) J. Gen. Physiol. (Cambridge, U.K.) 104, 395–422 26 Ganitkevich, V.Y. and Isenberg, G. (1995) J. Physiol. (Cambridge, U.K.) 484, 287–306 27 Fleischmann, B.K., Wang, Y.-X., Pring, M. and Kotlikoff, M.I. (1996) J. Physiol. (Cambridge, U.K.) 492, 347–358 28 Carrington, W.A., Lynch, R.M., Moore, E.D.W., Isenberg, G., Fogarty, K.E. and Fay, F.S. (1995) Science 268, 1483–1487 29 Ganitkevich, V.Y. and Isenberg, G. (1992) J. Physiol. (Cambridge, U.K.) 458, 119–137 30 Gregoire, ´ G., Loirand, G. and Pacaud, P. (1993) J. Physiol. (Cambridge, U.K.) 474, 483–500 31 Imaizumi, Y., Torii, Y., Ohi, Y., Nagano, N., Atsuki, K., Yamamura, H., Muraki, K., Watanabe, M. and Bolton, T.B. (1998) J. Physiol. (Cambridge, U.K.) 510, 705–719 32 Collier, M.L., Ji, G., Wang, Y.-X. and Kotlikoff, M.I. (2000) J. Gen. Physiol. 115, 653–662 33 Nelson, M.T., Cheng, H., Rubart, M., Santana, L.F., Bonev, A.D., Knot, H.J. and Lederer, W.J. (1995) Science 270, 633–637 34 Zhuge, R., Sims, S.M., Tuft, R.A., Fogarty, K.E. and Walsh, Jr, J.V. (1998) J. Physiol. (Cambridge, U.K.) 513, 711–718 35 Loirand, G., Pacaud, P., Baron, A., Mironneau, C. and Mironneau, J. (1991) J. Physiol. (Cambridge, U.K.) 437, 461–475 36 Gyorke, ¨ I. and Gyorke, ¨ S. (1998) Biophys. J. 75, 2801–2810 37 Ching, L.L., Williams, A.J. and Sitsapesan, R. (2000) Circ. Res. 87, 201–206 38 Allen, D.G., Eisner, D.A., Pirolo, J.S. and Smith, G.L. (1985) J. Physiol. (Cambridge, U.K.) 364, 169–182 39 North, R.A. (2002) Physiol. Rev. 82, 1013–1067 40 Mironneau, J., Coussin, F., Morel, J.L., Barbot, C., Jeyakumar, L.H., Fleischer, S. and Mironneau, C. (2001) J. Physiol. (Cambridge, U.K.) 536, 339–350 41 Gitterman, D.P. and Evans, R.J. (2001) Br. J. Pharmacol. 132, 1201–1208 42 Lamont, C. and Wier, W.G. (2002) Circ. Res. 91, 454–456 43 Brain, K.L., Williams, D.J. and Cunnane, T.C. (2002) J. Physiol. (Cambridge, U.K.) 543P, 79P Received 1 May 2003
