Transcription factor p63 controls the reserve status but not the stemness of horizontal basal cells in the olfactory epithelium.

Adult tissue stem cells can serve two broad functions: to participate actively in the maintenance and regeneration of a tissue or to wait in reserve and participate only when activated from a dormant state. The adult olfactory epithelium, a site for ongoing, life-long, robust neurogenesis, contains both of these functional stem cell types. Globose basal cells (GBCs) act as the active stem cell population and can give rise to all the differentiated cells found in the normal tissue. Horizontal basal cells (HBCs) act as reserve stem cells and remain dormant unless activated by tissue injury. Here we show that HBC activation following injury by the olfactotoxic gas methyl bromide is coincident with the down-regulation of protein 63 (p63) but anticipates HBC proliferation. Gain- and loss-of-function studies show that this down-regulation of p63 is necessary and sufficient for HBC activation. Moreover, activated HBCs give rise to GBCs that persist for months and continue to act as bona fide stem cells by participating in tissue maintenance and regeneration over the long term. Our analysis provides mechanistic insight into the dynamics between tissue stem cell subtypes and demonstrates that p63 regulates the reserve state but not the stem cell status of HBCs.

Sox2 and Pax6 Play Counteracting Roles in Regulating Neurogenesis within the Murine Olfactory Epithelium.

In the adult olfactory epithelium, the transcription factors Pax6 and Sox2 are co-expressed in sustentacular cells, horizontal basal cells (HBCs), and less-differentiated globose basal cells (GBCs)-both multipotent and transit amplifying categories-but are absent from immediate neuronal precursor GBCs and olfactory sensory neurons (OSNs). We used retroviral-vector transduction to over-express Pax6 and Sox2 individually and together during post-lesion recovery to determine how they regulate neuronal differentiation. Both Pax6 and Sox2, separately and together, can suppress the production of OSNs, as fewer clones contain neurons than with empty vector (EV), although this effect is not absolute. In this regard, Pax6 has the strongest effect when acting alone. In clones where neurons form, Pax6 reduces neuron numbers by comparison with EV, while Sox2 expands their numbers. Co-transduction with Pax6 and Sox2 produces an intermediate result. The increased production of OSNs driven by Sox2 is due to the expansion of neuronal progenitors, since proliferation and the numbers of Ascl1, Neurog1, and NeuroD1-expressing GBCs are increased. Conversely, Pax6 seems to accelerate neuronal differentiation, since Ascl1 labeling is reduced, while Neurog1- and NeuroD1-labeled GBCs are enriched. As a complement to the over-expression experiments, elimination of Sox2 in spared cells of floxed Sox2 mice, by retroviral Cre or by K5-driven CreERT2, reduces the production of OSNs and non-neuronal cells during OE regeneration. These data suggest that Pax6 and Sox2 have counteracting roles in regulating neurogenesis, in which Pax6 accelerates neuronal production, while Sox2 retards it and expands the pool of neuronal progenitors.