The effect of antidepressants cannot be explained by the classical monoaminergic theory. In particular, that model does not explain the delay in clinical response with antidepressants. Many hypotheses have been developed to understand the mechanism of action of antidepressants, each of them involving the regulation of different receptors. In parallel, functional brain imaging and neurobiological techniques have revealed specific neuroanatomical lesions in affective disorders. Depression in particular is associated with a neuronal loss in specific brain regions. These anatomical changes are reduced after antidepressant treatment. In the last decade, a new pathophysiological concept of affective disorders has emerged, integrating preferentially molecular and cellular antidepressant-induced changes leading to rehabilitation of synaptic activity. In the present review, we will summarize recent crucial data that establish the link between depression and neuroplasticity.