Our understanding of the role of astrocytes in the brain has been completely overturned in the last 2-3 decades by the discovery of their form of excitability (based on intracellular Ca2+ elevations) leading to the release of molecular modulators acting on neuronal synapses and vascular function. This discovery represented a conceptual paradigm shift in neuroscience, as astrocytes were seen before as simple synaptic isolators and mere metabolic suppliers to neurons. Work in the last decade aimed at deciphering the rules astrocytic language provided contradictory results. A major obstacle to the precise understanding of astrocyte functions has been the technical and conceptual difficulty in investigating astrocyte Ca2+ dynamics, as these cells display intricate 3D morphology and complex physical and functional interactions with surrounding elements (such as synapse and blood vessels) that are challenging to be studied even with the most advanced optical technologies. Because Ca2+ activity in astrocytes is mostly local, asynchronous and scattered across the entire cell structure, drawing conclusions about their overall properties based on information extracted from a single focal plane (as conventionally done in 2D microscopy studies) can be definitively misleading. For this reason, we decided to develop a new imaging approach to study astrocytes signals in 3D, to investigate in a comprehensive way their features looking, for the first time, at the entire territory of the cell, including the interface regions with synapses (even when located in optically sub-resolved domains of the cell) and blood vessels. To achieve this, we combined the most advanced two-photon technology and astrocyte-specific genetically-encoded Ca2+ indicators with a series of unique custom-made analytical tools that allowed us to detect Ca2+ activity at a completely new resolution, never described before. In our work, we were able to provide high resolution 3D maps of astrocytic activity, focusing on both qualitative and quantitative aspects of Ca2+ signaling impossible to appreciate in a comprehensive and reliable way by looking at a single focal plane. We demonstrated, in an indisputable way, that extracting Ca2+ signals from the most
accessible regions of the cell (like the soma) does not allow to infer how astrocytes communicate
at the interfaces with synapses or blood vessels, where very local forms of activity occur. When
we electrically stimulated hippocampal neuronal axons down to elicit individual action potentials, for example, we could spot correlated astrocytic responses taking place in very small regions of the cell (less than 1% of the scanned volume), that we identified as the functional neuron-glia interaction sites. The existence of such functional connections has been highly debated over the past years and one of the main reasons is that their location across the astrocytic structure is unpredictable (and so, their identification by randomly selecting a 2D focal plane within the astrocytic territory is virtually impossible). In both brain slices and in vivo, awake mice, astrocytic Ca2+ activity appeared extremely compartmentalized also at the end feet, the functional interface between astrocytes and blood vessels. Most of the Ca2+ events that we identified in in this region were fast and extremely localized and did not propagate to the neighboring processes or to the soma. Again, this explains why many studies to date failed to temporally correlate astrocyte Ca2+ dynamics and vascular changes (vasoconstriction or dilation): those studies looked (we can say, erroneously) at the temporal correlation between hemodynamic changes in the vessels and Ca2+ dynamics in the astrocytic soma/processes, that we report to have completely different features and kinetics. Our work represents a significant methodological step forward towards the comprehension of the complex language of astrocytes and gives the first solid biological indications as to how these cells are functionally integrated in their surrounding environment.