Terminal differentiation of adult hippocampal progenitor cells is a step functionally dissociable from proliferation and is controlled by Tis21, Id3 and NeuroD2

Cell proliferation and differentiation are interdependent processes. Here, we have asked to what extent the two processes of neural progenitor cell amplification and differentiation are functionally separated. Thus, we analyzed whether it is possible to rescue a defect of terminal differentiation in progenitor cells of the dentate gyrus, where new neurons are generated throughout life, by inducing their proliferation and/or their differentiation with different stimuli appropriately timed. As a model we used the Tis21 knockout mouse, whose dentate gyrus neurons, as demonstrated by us and others, have an intrinsic defect of terminal differentiation. We first tested the effect of two proliferative as well as differentiative neurogenic stimuli, one pharmacological (fluoxetine), the other cognitive (the Morris water maze (MWM) training). Both effectively enhanced the number of new dentate gyrus neurons produced, and fluoxetine also reduced the S-phase length of Tis21 knockout dentate gyrus progenitor cells and increased the rate of differentiation of control cells, but neither factor enhanced the defective rate of differentiation. In contrast, the defect of terminal differentiation was fully rescued by in vivo infection of proliferating dentate gyrus progenitor cells with retroviruses either silencing Id3, an inhibitor of neural differentiation, or expressing NeuroD2, a proneural gene expressed in terminally differentiated dentate gyrus neurons. This is the first demonstration that NeuroD2 or the silencing of Id3 can activate the differentiation of dentate gyrus neurons, complementing a defect of differentiation. It also highlights how the rate of differentiation of dentate gyrus neurons is regulated genetically at several levels and that a neurogenic stimulus for amplification of neural stem/progenitor cells may not be sufficient in itself to modify this rat