Final Report Summary - ASTROMNESIS (The language of astrocytes: multilevel analysis to understand astrocyte communication and its role in memory-related brain operations and in cognitive behavior)
The “Astromnesis” project studies the role of astrocytes in learning and memory processes at four interconnected levels: sub-cellular, cellular, circuital, and behavioral. It is a recent acquisition that astrocytes interact with synapses and neuronal circuits underlying cognitive behavior. However, the role (if any) that neuron-astrocyte signal exchanges have in cognitive processes is unclear. Likewise, it is unclear how the two cell types communicate given their different properties. Our project aimed at bringing light on these aspects via experimental and technological innovations. Studies at the subcellular level better defined the structural-functional relations. We identified several proteins involved in astrocyte-neuron communication and defined their subcellular distribution using light and electron microscopy. Some of these proteins accumulate at the interface between synaptic and astrocytic structures. Next, we moved analysis to the ultrastructural level. Because current methods do not preserve well the astrocyte ultrastructure, we developed a new method based on cryopreservation of biological samples. Using a special electron microscope, FIB-SEM, we reconstructed in 3D the ultrastructure of astrocytes with unprecedented resolution and started study of their internal organization. This method sets a new standard in astrocyte research. For studies at cellular level, we developed 3D two-photon Ca2+ imaging, a new technology allowing for the first time to record the activity of entire astrocytes. This major scientific and technical breakthrough provides a fully new view on astrocyte biology. The findings are surprising, astrocytes display activity (monitored as Ca2+ elevations) everywhere, mostly fast, local and asynchronous, suggesting that they contain “mini-compartments” that likely communicate with synapses and blood vessels independently. Our 3D information challenges previous interpretations of the functional roles of astrocytes based on recordings of slow and large somatic Ca2+ events, as these “central” events are much less frequent than the local events. They also overturn the dominating idea that astrocytes respond only to intense neuronal activity, because we detected responses even to minimal activity, but in very tiny regions, trackable only by analyzing the whole cell volume. This study was published in the prestigious journal Science. Next, we showed that astrocyte Ca2+-dependent signaling controls synaptic function in the hippocampus in a circuit-specific manner. This thanks to activation of a special class of NMDA receptors that are located in sub-populations of excitatory nerve terminals, in direct contact with astrocytes, and display atypical molecular composition that imparts them “astrocyte-friendly” functional properties. At the circuit level, we set-up a method to monitor astrocyte Ca2+ activity in a cognitive area in freely-moving animals while they perform a cognitive task involving that area. This was achieved by using microendoscopy, a technique never utilized before in astrocyte studies. For studies at behavioral level, we generated several new mouse lines carrying genetic modifications in specific components of astrocyte signaling. These exclusive lines are precious for defining astrocyte contributions to cognitive processing. We discovered that one of our lines, in which only astrocytes can transduce the action of the pro-inflammatory cytokine TNFalpha, displays a specific cognitive defect in a murine model of a human pathology. This proves that astrocyte signaling is critically involved in the functioning of cognitive circuits, at least during pathological processes. This study was published in the prestigious journal, Cell. The project generated additional advances. For example, handling of “big data” (GBs/min) produced by our imaging studies necessitated development of innovative IT platforms. Likewise, analysis of astrocytic Ca2+ activity required design of innovative and dedicated software tools, that we provide open source to the scientific community. Finally, as an educational outcome, we were invited by some of the most prestigious Neuroscience journals (Neuron, Nature Reviews Neuroscience, Nature Neuroscience) to review research advances in the domain of our project.