Back

Cell-type specific astrocyte activation is driven by cortical top-down modulation

Beiersdorfer, A.; Losse, K.; Bostel, J.; Popp, J. S.; Rotermund, N.; Schulz, K.; Droste, D.; Gee, C. E.; Hirnet, D.; Lohr, C.

2026-03-09 neuroscience
10.64898/2026.03.08.710364 bioRxiv
Show abstract

Cortical projections to cortical and subcortical targets provide top-down modulation that shapes neuronal performance, including gain control and excitation-inhibition balance. However, the contribution of astrocytes to this process remains poorly understood. In the olfactory bulb, the first relay station of odor information processing, bottom-up input is transmitted from olfactory sensory neurons to mitral/tufted (M/T) cells, which project to the olfactory cortex. Context- and state-dependent top-down modulation arises from feedback projections originating in the anterior piriform cortex (aPC) that target granule cells (GCs). We examined how astrocytes respond to bottom-up and top-down neuronal activity using confocal Ca{superscript 2} imaging, cell-type-specific optogenetics, electrical stimulation, and single-cell electrophysiology. We found that Ca{superscript 2} signals in astrocytes are selectively triggered by action potential-dependent ATP release from GCs while M/T cells failed to elicit significant astrocytic responses. Although synaptic input from M/T cells depolarized GCs, it was insufficient to induce action potential firing and subsequent astrocyte activation. By contrast, glutamatergic top-down input from the aPC evoked sustained GC firing, leading to ATP-dependent Ca{superscript 2} signaling in astrocytes. Our results reveal an unappreciated level of complexity in neuron-astrocyte communication, highlighting its cell-type specificity as well as its context- and state-dependence.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.