The lack of adequate methods to investigate brain energy metabolism with the required spatio-temporal resolution in the intact organism has hampered significant advances in the field. Förster resonance energy transfer (FRET) sensors specific for energy substrates, such as glucose, lactate and pyruvate have been developed and successfully used in cultured cells and in brain slices. A major advantage of these FRET sensors is that they do not interfere with the intrinsic metabolite concentrations and pathways. In addition to unsurpassed spatial resolution, FRET microscopy can also detect fast metabolic dynamics. Furthermore, these sensors have great potential for in vivo studies in combination with two-photon microscopy. Likewise, the ongoing development of novel genetically encoded calcium indicators have revealed complex spatiotemporal patterns of astrocytic calcium transients.
We report novel results on concurrent measurement of astrocyte and neuron calcium signals in the awake mouse. Astrocyte sub-cellular compartments show a delayed calcium response to sensory stimulation compared to neurons. However, fast signals that are similar in time scale to neurons occur in some astrocyte processes. This is most easily seen with membrane bound, fast GCaMP6f. Disruption of locus coeruleus signaling (via DSP4 injection) reduced the number of spontaneous astrocyte calcium signals but did not affect the astrocyte response to stimulation.
We present results using the genetically encoded FRET sensors for lactate and pyruvate in vivo. Recombinant adeno-associated virus (AAV) was used with appropriate promoters to express the sensors in astrocytes and neurons. Experiments were carried out under anesthesia and in awake, head-fixed mice. Various pharmacological interventions were developed and applied to compare the basal concentration and transients of energy substrates in single cells. We demonstrate that FRET sensors for energy substrates are powerful tools for in vivo investigations of the cellular compartmentalization of energy metabolism. We will present evidence for a significant lactate concentration gradient from astrocytes to neurons. This gradient is in support of a vectorial flux of lactate from astrocytes to neurons, as suggested by the astrocyte-neuron lactate shuttle hypothesis. Finally, we demonstrate rapid lactate release from astrocytes upon arousal-mediated cortical activation, followed by lactate production in astrocytes and update in neurons.