The possibility of specifically addressing recombinant probes to mitochondria is a novel, powerful way to study these organelles within living cells. We first showed that the Ca(2+)-sensitive photoprotein aequorin, modified by the addition of a mitochondrial targeting sequence, allows to monitor specifically the Ca2+ concentration in the mitochondrial matrix ([Ca2+]m) of living cells. With this tool, we could show that, upon physiological stimulation, mitochondria undergo a major rise in [Ca2+]m, well in the range of the Ca2+ sensitivity of the matrix dehydrogenases, in a wide variety of cell types, ranging from non excitable, e.g., HeLa and CHO, and excitable, e.g., cell lines to primary cultures of various embryological origin, such as myocytes and neurons. This phenomenon, while providing an obvious mechanism for tuning mitochondrial activity to cell needs, appeared at first in striking contrast with the low affinity of mitochondrial Ca2+ uptake mechanisms. Based on indirect evidence, we proposed that the mitochondria might be close to the source of the Ca2+ signal and thus exposed to microdomains of high [Ca2+], hence allowing the rapid accumulation of Ca2+ into the organelle. In order to verify this intriguing possibility, we followed two approaches. In the first, we constructed a novel aequorin chimera, targeted to the mitochondrial intermembrane space (MIMS), i.e., the region sensed by the low-affinity Ca2+ uptake systems of the inner mitochondrial membrane. With this probe, we observed that, upon agonist stimulation, a portion of the MIMS is exposed to saturating Ca2+ concentrations, thus confirming the occurrence of microdomains of high [Ca2+] next to mitochondria. In the second approach, we directly investigated the spatial relationship of the mitochondria and the ER, the source of agonist-releasable Ca2+ in non-excitable cells. For this purpose, we constructed GFP-based probes of organelle structure; namely, by targeting to these organelles GFP mutants with different spectral properties, we could label them simultaneously in living cells. By using an imaging system endowed with high speed and sensitivity, which allows to obtain high-resolution 3D images, we could demonstrate that close contacts (< 80 nm) occur in vivo between mitochondria and the ER.