(a), Set-up for live imaging of head skin epidermis. (b), Visualization of LV-Cre-mGFP transduced cells in vivo. Arrows indicate hair follicles. Suprabasal cells are identified based on their flattened morphology. Basal cells are identified based on their cuboidal morphology and proximity to the basement membrane (blue; second harmonic generation signal from Collagen). Scale bar, 50 μm. Experiment was repeated in transduced head skin of 4 animals independently with similar results. (c-e), Activation rate and bias of LV-CreER transduced epidermis treated with low dosage of Tamoxifen. At P20, CreER activity, based on conversion of mT to mG, is detected in ~1% of epidermal cells (c), majority of clones (~90%) are found as single cells (d), and most of the single cells ( > 98%) are localized to the basal/progenitor layer (e). Statistics based on n = 7 WT animals and 6 Pik3ca 2X animals. Error bar: SD, center value: mean. (f), FACS plot of GFP+/- cells sorted from Pik3ca 2X R26mT/mG epidermis transduced with LV-Cre or with LV-CreER and treated with low dose of Tamoxifen. Experiment was repeated 3 times independently with similar results. (g), Schematic of genotyping primers31 detecting substitution of wild type exon 20 with H1047R exon 20 of Pik3ca. (h), Genotyping PCR shows that low dose of Tamoxifen can efficiently activate Pik3caH1047R. Experiment was repeated 3 times independently with similar results. (i), Rate of progenitor cell renewal in WT and Pik3ca 2X epidermis observed using direct two-photon imaging is consistent with EdU-BrdU pulse-chase assay quantification. Statistics based on n = 3 animals of each condition. One-way ANOVA, P-value as indicated. Error bar: SD, center value: mean.