Lithium ion (Li+) is used as mood stabilizer in mood disorder for more than 60 years, but its mode of action remains largely obscure. Due to similarities with Mg2+, Li+ affects many biological processes dependent on Mg2+. During the past 10 years, a body of evidence has highlighted the inhibition of glycogen synthase kinase-3 (GSK-3) as a possible mechanism of therapeutic action of Li+. GSK-3 corresponds to two kinase-type enzymes (GSK-3α and GSK-3β), able to phosphorylate many proteins in neuronal and non-neuronal cells and, thereby, to exert a regulatory role in many cellular functions. GSK-3 itself is negatively regulated by phosphorylation produced by several enzymes, including Akt. It is currently believed that direct inhibition of GSK-3 by Li+ has no therapeutic relevance since only observed with Li+ concentrations toxic in humans. In contrast, Li+ concentrations consistent with therapeutic action in human activate Akt and, thereby, strongly inhibit the activity of GSK-3 [1]. Experimental animal studies have shown that the inhibitory effects of Li+ on responses to psychostimulants are related to an action on GSK-3 [2]. These behavioral responses can be compared to manic episodes and these results suggest that the antimanic effect of Li+ is mediated through GSK-3 inhibition. Several other studies suggest that antidepressant-like responses of Li+, assessed by behavioral tests in animal, depend on the GSK-3 inhibition [3]. Altogether, these preclinical data tend to attribute to the GSK-3 inhibition both antimanic and antidepressant effects, and perhaps a mood stabilizing effect. However, in the absence of clear understanding of mood disorder etiology, evaluating the exact contribution of GSK-3 inhibition to clinical effects of Li+ remains a complex issue.