Mitochondrial calcium signaling is important in the control of fundamental cellular functions including energy metabolism and apoptotic cell death (8,(14)(15)(16)(17). Mitochondrial calcium signals often are established by Ca 2ϩ release from sarco-endoplasmic reticulum Ca 2ϩ stores as a result of the strategic localization of the low-affinity mitochondrial Ca 2ϩ uptake sites close to Ca 2ϩ release channels. This machinery has been shown to be effective in delivering Ca 2ϩ to the mitochondria during [Ca 2ϩ ] c spikes (6-8, 10, 11, 18). Furthermore, Duchen et al. (19) monitored changes in mitochondrial membrane potential and demonstrated localized mitochondrial depolarizations in myocytes. Based on the association of depolarizations with contraction and on the effect of Ca 2ϩ transport inhibitors, they concluded that the depolarizations were due to mitochondrial Ca 2ϩ uptake. However, no measurement of localized cytosolic or mitochondrial matrix ([Ca 2ϩ ] m ) signals was shown in the study of Duchen et al. (19). Mitochondrial Ca 2ϩ uptake may be associated with membrane depolarization, but there is no expectation of a relationship between membrane potential and the subsequent duration and decay of mitochondrial Ca 2ϩ , which are the key parameters in the regulation of mitochondrial function. Moreover, mitochondrial Ca 2ϩ uptake can be associated with an increase in membrane potential because of activation of metabolism (8, 20). Thus, it is of critical importance to measure mitochondrial Ca 2ϩ directly and to address the question of whether brief opening of a few release channels during elementary Ca 2ϩ release events, such as ryanodine receptor (RyR)-mediated Ca 2ϩ sparks and IP 3 receptor-mediated Ca 2ϩ puffs, is sufficient to trigger [Ca 2ϩ ] m increases. The extent to which [Ca 2ϩ ] c sparks and puffs propagate to the mitochondria is a critical parameter for understanding the control of mitochondrial Ca 2ϩ targets and the local feedback exerted by mitochondrial Ca 2ϩ uptake on [Ca 2ϩ ] c . Using confocal microscopy and compartmentalized rhod2 to image [Ca 2ϩ ] at the level of individual mitochondria for the first time, we were able to record the elementary units of the [Ca 2ϩ ] m signal and to study how the miniature [Ca 2ϩ ] m rise is integrated into intracellular Ca 2ϩ regulation.
Experimental ProceduresCells and Solutions. H9c2 cardiac cells were cultured, plated for imaging, and allowed to differentiate to myotubes as described previously (21,22). Before use, the cells were preincubated for 30 min in extracellular medium composed of 121 mM NaCl, 5 mM NaHCO 3 , 10 mM Na⅐Hepes, 4.7 mM KCl, 1.2 mM KH 2 PO 4 , 1.2 mM MgSO 4 , 2 mM CaCl 2 , 10 mM glucose, and 2% BSA (pH 7.4) at 37°C. Loading with the dyes was carried out in the same buffer. For measurements of [Ca 2ϩ ] m in permeabilized myotubes, the cells were loaded with 4 M rhod2͞AM in the presence of 0.003% (wt͞vol) pluronic acid at 37°C for 50 min. Dye-loaded cells were washed with Ca 2ϩ -free extracellular buffer composed of 120 mM NaCl, 20 mM Na⅐...