Displacement in Skeletal Muscle

In skeletal muscle, the waveform of Ca 2 release under clamp depolarization exhibits an early peak. Its decay reflects an inactivation, which locally corresponds to the termination of Ca 2 sparks, and is crucial for rapid control. In cardiac muscle, both the frequency of spontaneous sparks (i.e., their activation) and their termination appear to be strongly dependent on the Ca 2 content in the sarcoplasmic reticulum (SR). In skeletal muscle, no such role is established. Seeking a robust measurement of Ca 2 release and a way to reliably modify the SR content, we combined in the same cells the “EGTA/phenol red” method (Pape et al., 1995) to evaluate Ca 2 release, with the “removal” method (Melzer et al., 1987) to evaluate release flux. The cytosol of voltage-clamped frog fibers was equilibrated with EGTA (36 mM), antipyrylazo III, and phenol red, and absorbance changes were monitored simultaneously at three wavelengths, affording largely independent evaluations of [H ] and [Ca 2 ] from which the amount of released Ca 2 and the release flux were independently derived. Both methods yielded mutually consistent evaluations of flux. While the removal method gave a better kinetic picture of the release waveform, EGTA/phenol red provided continuous reproducible measures of calcium in the SR (Ca SR ). Steady release permeability ( P ), reached at the end of a 120-ms pulse, increased as Ca SR was progressively reduced by a prior conditioning pulse, reaching 2.34-fold at 25% of resting Ca SR (four cells). Peak P , reached early during a pulse, increased proportionally much less with SR depletion, decreasing at very low Ca SR . The increase in steady P upon depletion was associated with a slowing of the rate of decay of P after the peak (i.e., a slower inactivation of Ca 2 release). These results are consistent with a major inhibitory effect of cytosolic (rather than intra-SR) Ca 2 on the activity of Ca 2 release channels. key words: sarcoplasmic reticulum • excitation–contraction coupling • Ca channels • channel gating • channel modulation I N T R O D U C T I O N In striated muscle, contraction is switched on by Ca 2 release from the sarcoplasmic reticulum (SR). In skeletal muscle, Ca 2 release channels are controlled primarily by transverse (T) tubule membrane voltage. In principle, it should also be controlled by Ca 2 ions bound to excitatory and inhibitory cytosol-facing sites, the characterization of which has been done largely in membrane fractions and bilayer-reconstituted ryanodine receptor channels (e.g., Laver et al., 1997). Multiple experimental observations, on skeletal (Ikemoto et al.,1989; Donoso et al.,1995) and cardiac muscle (Gyorke et al., 2002; Terentyev et al., 2002, 2003), complicate this picture. Apparently the SR Ca 2 acting directly on SR-lumenal sites, or perhaps on cytosolic sites upon channel opening, may also modulate and control the state of the channels, and affect the release current. Most interestingly, in heart muscle there is now evidence for a role of the reduction in free lumenal calcium concentration, [Ca 2 ] SR , consequent to Ca 2 release, in the termination of Ca 2 sparks (Terentyev et al., 2002, 2003). While in skeletal muscle there is no comparable evidence, the issue of spark termination is equally important there, as it underlies the sharp decay of release flux that terminates the early peak observed under voltage clamp, and is understood as crucial in the self-limitation of Ca 2 release necessary to keep the process under voltage control (Stern et al., 1997). The waveform of Ca 2 release flux elicited by a voltage clamp depolarization pulse exhibits two well-defined kinetic phases. Essentially at all suprathreshold voltages, an early peak of flux is followed by a decay (sometimes oscillatory, see Rengifo et al., 2002) to a quasi-steady level that persists while the pulse is on. 20 yr after the first description of these phases (Baylor et al., 1983; The online version of this article includes supplemental material. Address correspondence to Eduardo Ríos, Dept. of Molecular Biophysics and Physiology, Rush University School of Medicine, 1750 W. Harrison St., Suite 1279JS, Chicago, IL 60612. Fax: (312) 942-8711; email: [email protected] Abbreviations used in this paper: CICR, Ca 2 -induced Ca 2 release; SR, sarcoplasmic reticulum; T, transverse. on A ril 4, 2008 w w w .jg.org D ow nladed fom http://www.jgp.org/cgi/content/full/jgp.200409071/DC1 Supplemental Material can be found at: 240 Simultaneous Measurement of [Ca 2 ] and [H ] Transients in Muscle Melzer et al., 1984), there is still no consensus view of their mechanisms. A common belief is that the kinetics are determined by Ca 2 ions, through their dual effects of activation and inhibition/inactivation of ryanodine receptors. But there is no dominant narrative of how interactions of Ca 2 and channels result in two phases under voltage clamp. The discord starts with the results of experiments repeatedly attempted on the effects of extrinsic Ca 2 buffers, added at concentrations sufficiently high to inhibit local effects of Ca 2 . While buffers suppress the kinetic features, leaving a “flat” waveform, some see the effect as a selective loss of the peak phase (Jacquemond et al., 1991; Csernoch et al., 1993), consistent with activation of this phase by Ca 2 (Ca 2 -induced Ca 2 release or CICR) as proposed earlier by us (Ríos and Pizarro, 1988). By contrast, other researchers interpret the change to a flat waveform as consequence of suppression by the buffers of the decay that terminates the peak (Baylor and Hollingworth, 1988; Hollingworth et al., 1992). This view supports a Ca 2 -dependent inactivation mechanism for the decay (first proposed by Baylor et al., 1983, and further studied by Schneider and Simon, 1988, Simon et al., 1991, and others). At the root of this discrepancy is the divergent quantification of the waveform of Ca 2 release flux as affected by buffers. While all groups obtain a similar time course, those who support the CICR view report an inhibition of the flux by the buffers (Jacquemond et al., 1991; Csernoch et al., 1993). In contrast, the interpretation based on Ca 2 -dependent inactivation requires a promotion of release, especially of its late steady phase, which is what the supporters of this view find in the buffer-modified waveforms (e.g., Hollingworth et al., 1992). Ultimately, the contention on mechanisms simply reflects lack of a uniform scaling of otherwise similar waveforms. In our view there is no common scaling because the methods used to determine flux are different. A main goal of the present work is to resolve these differences by applying both methods simultaneously. The dissension spills over to the interpretation of the steady level of flux. While the early peak is essential physiologically, as it roughly coincides with the time span of an action potential, the properties of the steady component have been studied as indicators of the underlying mechanisms of control. If the level of flux in this phase may seem constant in the short term (say 100 ms), it is clearly decaying in longer time scales, a decay that has been demonstrated to reflect SR depletion (Schneider et al., 1987b). Two views have emerged of the behavior of channels during this phase: Schneider et al. (1987b) introduced the assumption that the permeability was constant, hence the decay was a simple consequence of depletion. Under this assumption the flux becomes a proportional indicator of [Ca 2 ] SR . It is therefore simple to deduce a permeability from the flux measurement, which these workers derive by the “removal” method (i.e., fitting parameters of a common pool model of Ca 2 fluxes in the cytosol; Melzer et al., 1984, 1987). On the other hand Pape et al. (1995) introduced the “EGTA/phenol red” method to quantify Ca 2 release, and interpreted their results to demonstrate an inverse relationship between permeability and SR Ca 2 content. At the base of the discrepancies are differences in the techniques. The removal method starts from a measure of the Ca 2 concentration transient. In the presence of a slowly equilibrating buffer like EGTA (Smith et al., 1984) at high intracellular concentration, the Ca 2 transient is kinetically very close to release flux (Eq. 17 of Ríos and Pizarro, 1991). More specifically, it equals the sum of a dominant term proportional to release flux, and another roughly proportional to its time integral (Eq. A8 of Pape et al., 1995; Eq. A8 of Song et al., 1998). In contrast, the EGTA/phenol red method derives calcium release from a measure of the H transient resulting from stoichiometric displacement by released Ca 2 as it binds to EGTA. The starting signal is in this case an indirect measure of cumulated Ca 2 release, rather than its time derivative. Perhaps as a consequence, the studies with the removal method emphasize kinetic features, especially the distinctive peak of Ca 2 release (Schneider et al., 1987b; Schneider and Simon, 1988; González and Ríos, 1993; Shirokova et al., 1996), while in general the applications of EGTA/phenol red (Pape et al., 1995, 1998, 2002; Pape and Carrier, 1998, 2002; Fénélon and Pape, 2002) have not placed the same level of attention on the separate study of kinetic phases. To explore how the calcium content of the SR controls the amplitude and time course of Ca 2 release and to understand the differences between the two methods, we combined both approaches and applied them simultaneously, with a pulse protocol that induced variable SR depletion. The determination of Ca 2 release and initial calcium content in the SR (Ca SR [0]) allowed us to derive SR content, Ca SR ( t ), at all times, and evaluate the effect of its changes on release flux and release permeability. By applying the two methods in parallel, the kinetic phases of release are resolved reliably and quantitatively. The results reveal different effects of depletion on the kinetic stages of release flux. M A T E R I A L S A N D M E T H O D S The experiments were performed on single cut muscle fibers voltage clamped in a two-Vaseline gap. This technique has been extensively described (e.g., González and Ríos, 1993). Frogs Rana pipiens were killed by decapitation under deep anesthesia. Both m. semitendinosus were separated and placed in dissection chambers. A 2-cm-long piece of fiber dissected from the muscles was on A ril 4, 2008 w w w .jg.org D ow nladed fom