An active role of the DeltaN isoform of p63 in regulating basal keratin genes K5 and K14 and directing epidermal cell fate

Abstract

Background: One major defining characteristic of the basal keratinocytes of the stratified epithelium is the expression of the keratin genes K5 and K14. The temporal and spatial expression of these two genes is usually tightly and coordinately regulated at the transcriptional level. This ensures the obligate pairing of K5 and K14 proteins to generate an intermediate filament (IF) network that is essential for the structure and function of the proliferative keratinocytes. Our previous studies have shown that the basal-keratinocyte restricted transcription factor p63 is a direct regulator of K14 gene.

Methodology/principal findings: Here we provide evidence that p63, specifically the DeltaN isoform also regulates the expression of the K5 gene by binding to a conserved enhancer within the 5' upstream region. By using specific antibodies against DeltaNp63, we show a concordance in the expression between basal keratins and DeltaNp63 proteins but not the TAp63 isoforms during early embryonic skin development. We demonstrate, that contrary to a previous report, transgenic mice expressing DeltaNp63 in lung epithelium exhibit squamous metaplasia with de novo induction of K5 and K14 as well as transdifferentiation to the epidermal cell lineage. Interestingly, the in vivo epidermal inductive properties of DeltaNp63 do not require the C-terminal SAM domain. Finally, we show that DeltaNp63 alone can restore the expression of the basal keratins and reinitiate the failed epidermal differentiation program in the skin of p63 null animals.

Significance: DeltaNp63 is a critical mediator of keratinocyte stratification program and directly regulates the basal keratin genes.

Figure 1. Generation of TAp63-specific antibody.

A) Specificity of the TAp63 antibody. Western blot analysis of whole cell extracts transfected with various HA-epitope tagged isoforms of p63 show reactivity of the TAp63 antibody with cells transfected with HA-TAp63α, TAp63β, and TAp63γ only. B) Immunofluorescence of cells transfected with HA-TAp63γ demonstrate the ability of the TAp63 antibody to detect TAp63 in cell culture in vivo. DAPI nuclear staining is shown in blue, HA epitope tag is shown in red, TAp63 specific staining is shown in green and a merge is shown in the lower panel. C) TAp63 expression in oocytes in sections of mouse ovary. TAp63 expression is shown in red and DAPI is shown in blue.

Figure 2. ΔNp63 is expressed during early mouse embryonic skin development.

Isoform specific expression of p63 was analyzed during various stages of skin development using three antibodies recognizing different p63 variants. Far left panel (panel 1) are H&E staining of skin using paraffin embedded whole embryos. Panel 2 demonstrates expression of ΔNp63 (green) using the RR-14 antibody. Panel 3 shows expression of p63α (green) using the H-129 antibody. K5 and K14 expression is also shown in green (panels 3 and 4 respectively). Far right panel reveals that TAp63 expression is absent during the developmental windows examined. Scale bar: 50 µm.

Figure 3. Binding of ΔNp63α to Hs II region of the K5 gene.

A) EMSA reveals binding of ΔNp63α in nuclear extracts of HaCaT cells using a radiolabeled probe corresponding to Hs II of the human K5 gene. Competition assays show that the KSC oligonucleotide, which contains a previously identified ΔNp63α binding site within the K14 gene, competes with binding of Hs II probe (lanes 2–3). A p53 consensus (lanes 4–5) and a WT oligonucleotide (lane 6–7) also compete for binding while a MT oligonucleotide did not affect Hs II binding (lanes 8–9). The highest complex (asterix) is supershifted (arrow) with the addition of antibodies raised against various domains of ΔNp63 whereas anti-TA antibodies have no effect (lanes 10–13). The lower complexes are likely to be non-specific. B) In vivo occupancy of ΔNp63α to the mouse and human K5 gene. ChIP was performed on HaCaT or primary mouse keratinocytes using two separate antibodies recognizing ΔNp63α as well as a nonspecific IgG as indicated (left panel). Input represents PCR amplification of 1% of the genomic DNA. Primers corresponding to a region of the GAPDH gene serves as a negative control. Right panel shows results obtained from real-time PCR experiments.

Figure 4. Three major isoforms of ΔNp63 can induce expression of K5 in cell culture.

Ptk2 cells were transfected with plasmids encoding different HA-epitope tagged isoforms of ΔNp63 as indicated. Cells were stained with antibodies detecting HA (red) and K5 (green). Nuclei are stained with DAPI (blue). 40× magnification is shown.

Figure 5. ΔNp63 can induce de novo expression of K5 and K14 in single-layered lung epithelia.

A) Schematic depicting the mating scheme used to generate transgenic animals expressing the HA-ΔNp63 transgene in single-layered lung epithelium using the surfactant protein C (SPC) promoter in a tetracycline inducible fashion. B) Gross morphology of E18.5 lung tissue sections stained with H&E from control animals and bi-transgenic animals (ΔNp63αBG and ΔNp63βBG). Bi-transgenic animals exhibit squamous metaplasia. Scale bar: 50 µm. C) Lung tissue sections from E18.5 stained with antibodies detecting HA epitope tag (green) and K5 or K14 (red) reveal expression of the transgene in bi-transgenic animals. Dapi staining is shown in blue. Tissues from bi-transgenic animals demonstrate induction of K5 and K14 expression (red staining) as compared to control littermates.

Figure 6. Ectopic Expression of ΔNp63β can partially rescue the p63 null phenotype.

Top left panel shows H&E staining of p63^−/− and p63^−/−,ΔNp63β rescued animals. Scale bar: 50 µm. Remaining panels illustrate immunofluorescence staining using various antibodies as indicated (in green). White hashed line demarcates the dermal epidermal boundary. Scale bar is 25 µm for immunofluorescence images.