New neurons are generated in the granule cell layer of the dentate gyrus (DG) of the hippocampus in adult humans (Eriksson et al, 1998). Our knowledge of adult neurogenesis in humans is quite limited and it could differ from adult neurogenesis in lower mammals. In rodents, neurogenesis is necessary for learning, and some antidepressant effects are lost in the absence of adult neurogenesis, which increases with environmental enrichment and exercise (Olson et al, 2006), as well as with antidepressant treatment (Couillard-Despres et al, 2009).

We reported (Boldrini et al, 2009) that selective serotonin reuptake inhibitors and tricyclic antidepressants increase dividing and neural progenitor cells (NPCs) in the DG of depressed subjects (MDD), compared with untreated MDD patients or controls. In humans, antidepressants increase the number of mitotic cells of all phenotypes in the DG, regardless of age. On the other hand, replication of NPCs, as in lower mammals, decreases with age (Couillard-Despres et al, 2009). This might explain why there is a poor antidepressant response in the elderly.

The functional relevance of the enhancement of neurogenesis by antidepressants needs to be ascertained by determining whether increased cell proliferation is associated with improvement of symptoms in MDD. In our study (Boldrini et al, 2009), a significant proportion of subjects died by suicide, which would argue against the benefits of antidepressant-induced cell proliferation, as opposed to the potential benefits of cell maturation, survival, and integration into functional neural networks, which should have a greater role in the potential beneficial impact of adult neurogenesis. Exposure to enriched environments, learning, and neurotrophins improve the survival and differentiation of newborn cells. Factors regulating cell survival and integration should be considered when examining the role of adult neurogenesis for antidepressant efficacy.

Another open question is the role of neurogenesis in the pathogenesis of MDD. Adult neurogenesis decreases with stress in rodents and is enhanced by environmental enrichment, exercise, and antidepressants. However, blunted cell replication alone does not induce depression-like behavior in mice. Growth factors, which affect neurogenesis, are decreased in MDD. Therefore, impaired hippocampal plasticity may be involved in the pathogenesis of MDD, not merely because of impaired cell replication but also because of impaired cell connectivity and functional integration into brain circuits that regulate emotional responses to the environment.

In our study, the antidepressant-induced increase in NPCs and dividing cells was associated with a larger volume of DG. Antidepressant treatment is known to increase hippocampal volume in posttraumatic stress disorder (Bossini et al, 2007), but no similar data are available in depression, although patients with MDD have a smaller hippocampus. The volume increase could be related to a restoration of cell number or neuropil, as antidepressants reverse dendritic shrinkage and improve cell survival, activating the antiapoptotic protein Bcl-2 and brain-derived neurotrophic factor expression in mammals.

Future studies must determine whether antidepressant response is linked to increased neurogenesis, but assessing adult neurogenesis in vivo is challenging. In a recent study, magnetic resonance spectroscopy was proposed as a possible method, but the specificity of the molecule used to identify NPCs was questioned and the results have not been replicated. Positron emission tomography has the limitation of low resolution and the unknown consequences of radiolabeling newborn cells. Cerebral blood volume, which correlates with angiogenesis, may prove to be a viable method for detecting neurogenesis in vivo (Pereira et al, 2007). Linking neurogenesis to improvement in depression symptomatology would justify seeking new treatments that increase not only neurogenesis but also plasticity, cell survival, and integration into functional networks.