Intramitochondrial free calcium concentration ([Ca²⁺]_m) in non-cardiac cell types has recently been implicated in regulating or modulating whole-cell Ca²⁺ signalling (Babcock & Hille, 1998). However, research in the heart has been hampered by the difficulty of measuring [Ca²⁺]_m in living myocytes. We previously developed a method of localising indo-1 into what we believe to be the mitochondrial compartment of rat myocytes (Griffiths et al. 1997). This was done by loading cells with indo-1 AM, which localises into both mitochondrial and cytosolic compartments, and then selectively removing the cytosolic indo-1. However, indo-1 was detected using microfluorimetry, which only measures total cell fluorescence, and so the subcellular localisation of indo-1 could not be determined. In the present study, we use confocal microscopy to directly visualise the subcellular distribution of indo-1.

Male rats were killed humanely by cervical dislocation and ventricular myocytes isolated by collagenase digestion. Cells were loaded with indo-1 AM by incubation with 10 µM indo-1 AM for 15 min at room temperature, centrifuged and resuspended in Hepes buffer containing 1 mM CaCl₂. Cells were then split into two portions, and either used for measurement of total cell fluorescence, or subjected to an extended incubation protocol for measurement of [Ca²⁺]_m. Cells were maintained at room temperature for 2.5 h, and then shaken gently at 37 °C for 1.5 h to selectively remove cytosolic indo-1 (Griffiths et al. 1997). Confocal imaging was performed using eight scans for each cell, and other parameters selected to optimise resolution. Cells were illuminated with an Ar-UV laser at 351 nm for detection of indo-1 fluorescence, and light detected at 385-425 nm. Rhodamine and rhod-2 fluorescence were detected using excitation at 568 nm and emission at 590 nm.Initially, indo-1 was distributed throughout the whole cell, but following the extended incubation protocol the localisation was very different; in particular, indo-1 was no longer present in the nuclei which then appeared as dark voids (Fig. 1). On average, 44 % of the indo-1 fluorescence signal remained following the incubation. A mitochondrial localisation of indo-1 was confirmed since the distribution was identical to that of rhodamine-123, a mitochondrial marker dye. The fluorescent indicator rhod-2 has been used by other investigators to measure [Ca²⁺]_m (Trollinger et al. 1997). However, we could not localise rhod-2 exclusively into the mitochondrial compartment in the same way as indo-1; rhod-2 remained in the nuclei, and ‘hot-spots’ of fluorescence were visible in the cell, which may correspond to a lysosomal location of the dye. This work provides firm validation for our previous studies, and highlights the importance of establishing the intracellular distribution of fluorescent indicators before conclusions can be drawn from any results obtained.This work was supported by the British Heart Foundation, Garfield Weston Trust, and Medical Research Council (Cell Imaging Facility).

Figure 1. Confocal microscopy of single rat heart myocytes loaded with indo-1 in order to measure either total cell [Ca2+] (A) or only mitochondrial [Ca2+] (B).

Babcock, D.F. & Hille, B. (1998). Curr. Opin. Neurobiol. 8, 398-404.

Griffiths, E.J., Stern, M.D. & Silverman, H.S. (1997). Am. J. Physiol. 273, C37-44.

Trollinger, D.R., Cascio, W.E. & Lemasters, J.J. (1997). Biochem. Biophys. Res. Commun. 236, 738-742.