The molecular pathogenesis of depression is often explained by the catecholamine theory. However, there are many cases of so-called refractory depression that are resistant to existing noradrenergic and serotonin activating drugs, so the true molecular pathogenesis of depression needs to be elucidated. As demonstrated by Prof Burnstock, ATP is not a merely energy currency but is an important intercellular signaling molecule that mediates information between neurons and glial cells in the central nervous system. Animal studies have reported that "ATP" is the most universally altered intercellular signaling molecule in depressive symptoms, and that a decrease in this extracellular ATP causes depressive symptoms (Nat Med 2013). However, its molecular and regulatory mechanisms remain unclear. In the present study, we tested this ATP hypothesis and elucidated the molecular mechanism of this ATP hypothesis using fluoxetine (FLX), a typical SSRI antidepressant, which increased extracellular ATP (ATPo) in the hippocampal brain in a dose-dependent manner. This ATPo increase was dependent on astrocytes rather than neurons, and the molecular mechanism of this increase was found to be an enhancement of the ATP exocytosis. We found that released ATP acts as ATP and degraded adenosine on P2Y11 and A2b receptors, respectively, to enhance the cAMP-CREB system and that this signal enhancement leads to BDNF expression. Very interestingly, the sequence of responses from ATP/adenosine to BDNF production was induced in astrocytes, but not in neurons. We further discuss the importance of astrocytes as a prime target for depression.