Commentary: Chronic SSRI stimulation of astrocytic 5-HT2B receptors change multiple gene expressions/editings and metabolism of glutamate, glucose and glycogen: a potential paradigm shift.

International audienceWorking on the serotonin (5-hydroxytryptamine, 5-HT) 5-HT2B receptor since several years, we have read with high interest the review by Hertz et al., (Hertz et al., 2015). Previous studies from our group demonstrated that a direct injection in mouse raphe nucleus of the 5-HT2B agonist BW723C86 has the ability to increase extracellular levels of serotonin, which can be blocked by the selective 5-HT2B receptor antagonist RS127445 (Doly et al., 2008;Doly et al., 2009). We also reported that an acute injection of paroxetine 2 mg/kg in mice knocked out for the 5-HT2B receptor gene or in wild type mice injected with RS127445 (0.5 mg/kg) triggers a strong reduction in extracellular accumulation of 5-HT in hippocampus (Diaz et al., 2012). Following these observations, we showed that acute and chronic BW723C86 injection (3 mg/kg) can mimic the fluoxetine (3 mg/kg) and paroxetine (1 mg/kg) behavioral and biochemical antidepressant effects in mice (Diaz and Maroteaux, 2011;Diaz et al., 2012). Hertz and coworkers used some of our published data on mice to support that fluoxetine and other SSRIs are acting as direct 5-HT2B receptor agonists independently of the serotonin transporter (SERT), based on their work on astrocytes. These authors forgot other important informations provided in Diaz's paper (Diaz et al., 2012), ie the absence of antidepressant effects of either fluoxetine or BW723C86 in mice lacking either the serotonin transporter (knockout for SERT) or differentiated serotonin neurons (knockout for Pet1). These data (i) rule out that the antidepressant effects of the 5-HT2B agonist BW723C86 or of fluoxetine could be independent of SERT; (ii) they indicate that serotonin neurons expressing SERT (and 5-HT2B receptors) are necessary for the 5-HT2B receptor effects independently of other cell types; (iii) they rule out the possibility that SSRIs mediate antidepressant effects only by stimulating directly putative astrocytic 5-HT2B receptors, which should be intact in these two mutant mice (SERT KO and Pet1 KO). A previous work (Launay et al., 2006) demonstrated a 5-HT2B receptor-mediated control of SERT activity in primary neurons from raphe nuclei, in which the BW723C86/5-HT2B receptor coupling promotes phosphorylations of SERT that can easily explain our findings in mice by a direct regulation of SERT by 5-HT2B receptors in a cell autonomous manner