SIRT6 promotes COX-2 expression and acts as an oncogene in skin cancer

Abstract

SIRT6 is a SIR2 family member that regulates multiple molecular pathways involved in metabolism, genomic stability, and aging. It has been proposed previously that SIRT6 is a tumor suppressor in cancer. Here, we challenge this concept by presenting evidence that skin-specific deletion of SIRT6 in the mouse inhibits skin tumorigenesis. SIRT6 promoted expression of COX-2 by repressing AMPK signaling, thereby increasing cell proliferation and survival in the skin epidermis. SIRT6 expression in skin keratinocytes was increased by exposure to UVB light through activation of the AKT pathway. Clinically, we found that SIRT6 was upregulated in human skin squamous cell carcinoma. Taken together, our results provide evidence that SIRT6 functions as an oncogene in the epidermis and suggest greater complexity to its role in epithelial carcinogenesis.

Figure 1. Role of SIRT6 in human skin SCC and mouse skin tumorigenesis

A, Immunoblot analysis of SIRT6 and β-actin in human normal skin and SCC (n=6). B,.Real-time PCR analysis of SIRT6 mRNA levels in normal human skin and SCC (n=6). C, Schematic for the DMBA-TPA treatment of SIRT6 WT and cKO mice. D, percentage of tumor-free mice from SIRT6 WT and cKO mice treated with DMBA-TPA (n = 15). E, average tumor number per mouse from SIRT6 WT and cKO mice treated with DMBA-TPA (n = 15).

Figure 2. SIRT6 loss suppresses proliferation and hyperplasia in mouse skin

A, histological analysis of control, treated non-tumor skin, and tumors in SIRT6 WT and cKO mice (n=3). B, immunohistochemical analysis of Ki67-positive cells in SIRT6 WT and cKO mouse skin following treatment with DMBA-TPA (n=3). C, histological analysis of SIRT6 WT and cKO mouse skin post-UVB (350 mJ/cm²) or –sham irradiation (three times in one week) (n=3). Scale bar=100 μm. D, quantification of epidermal thickness (μm) in C. *, P<0.05, significant differences between SIRT6 WT and cKO groups. E, immunohistochemical analysis of Ki67-positive cells in SIRT6 WT and cKO mouse skin post-UVB or -sham irradiation (three times in one week) (n=3). Scale bar=50 μm. F, quantification of Ki67-positive (Ki67+) cells in E. *, P<0.05, significant differences between SIRT6 WT and cKO groups.

Figure 3. SIRT6 loss suppresses COX-2 expression

A, Real-time PCR analysis of COX-2 mRNA expression in NHEK cells transfected with NC or siSIRT6. B, Immunoblot analysis of COX-2, SIRT6, and GAPDH in NHEK cells transfected with NC or siSIRT6. C, Immunoblot analysis of COX-2 and GAPDH in SIRT6 WT and cKO mouse skin tissue (n=6). Three skin samples were pooled. D, Immunoblot analysis of COX-2, SIRT6, and GAPDH in NHEK cells at 48 h post-infection with an adenoviral vector expressing empty vector or SIRT6 (Ad-SIRT6) for 6 h. E, Immunoblot analysis of COX-2 and GAPDH in two representative tumor and adjunct skin from DMBA-TPA-treated SIRT6 WT and cKO mouse (n=6). Protein levels were quantified using the ImageJ software shown in the right panel. *, P<0.05, significant differences between comparison groups. F, Immunoblot analysis of SIRT6, COX-2 and GAPDH in NHEK cells transfected with NC or siSIRT6 at 1.5 h and 6 h post-UVB (30 mJ/cm²) or –sham irradiation. G, Immunoblot analysis of SIRT6, COX-2, and GAPDH in NHEK cells transfected with NC or siSIRT6 upon treatment with DMBA (0, 10, 100 μM) for 24h. H, Immunoblot analysis of SIRT6, COX-2, and GAPDH in NHEK cells transfected with NC or siSIRT6 following treatment with TPA (0, 100, or 500 nM) for 24 h. These results were obtained from three independent experiments.

Figure 4. Role of COX-2 in SIRT6 regulation of cell survival post-UVB

A, Analysis of UVB-induced apoptosis by flow cytometric analysis of sub-G1 cells in NHEK cells transfected with negative control siRNA (NC) or siRNA targeting SIRT6 (siSIRT6). B, Quantitation of apoptotic cells (%) in A. *, P<0.05, significant differences between SIRT6 WT and cKO groups. C, Immunoblot analysis of SIRT6, cleaved-caspase 3, and GAPDH in NHEK cells transfected with siRNA targeting SIRT6 (siSIRT6) or negative control siRNA (NC) at 14 h post-UVB (50 mJ/cm²) or –sham irradiation. D, Immunoblot analysis of COX-2, p-AKT, and GAPDH in NC- or siCOX-2-transfected NHEK cells at 1.5 and 6 h post-UVB (30 mJ/cm²) or -sham irradiation. E, Immunoblot analysis of COX-2, cleaved-caspase 3, and GAPDH in NC- or siCOX-2-transfected NHEK cells at 24 h post-UVB (50 mJ/cm²) or –sham irradiation. F, Immunoblot analysis of SIRT6, COX-2 and GAPDH in NHEK cells transfected with NC, siSIRT6, or a combination of siSIRT6 and a COX-2-expressing construct. G, Immunoblot analysis of cleaved-caspase 3 and GAPDH in NHEK cells transfected with NC, siSIRT6, or a combination of siSIRT6 and a COX-2-expressing construct at 1.5, 6, and 24 h post-UVB (50 mJ/cm²) or –sham irradiation. These results were obtained from three independent experiments.

Figure 5. Role of the AMPK pathway in SIRT6 regulation of COX-2

A, Luciferase reporter assay of COX-2-luc in NHEK cells transfected with NC or siSIRT6. B, Real-time PCR analysis of COX-2 mRNA in NHEK transfected with NC or siSIRT6 upon treatment with actinomycin (1 μM) for 0.5, 1, 2, 6, or 24 h. C, Immunoblot analysis of SIRT6, p-ACC, ACC, p-AMPK, AMPK, p-AKT, AKT, and GAPDH in NHEK cells transfected with NC or siSIRT6. D, Immunoblot analysis of p-ACC, ACC, COX-2, SIRT6, and GAPDH in NHEK cells transfected with NC or siSIRT6 following treatment with DMSO or Compound C (10 μM) for 24h. These results were obtained from three independent experiments.

Figure 6. UVB up-regulates SIRT6 through activating AKT

A, Immunoblot analysis of COX-2, SIRT6, acH3K9 and GAPDH in NHEK cells at 24 h post-sham or –UVB (40 mJ/cm²). B, Immunoblot analysis of SIRT6, COX-2, p-AKT, AKT, and GAPDH in NHEK cells pretreated with vehicle (Vec) or LY294002 (LY, 10 μM) at 24 h post-sham or –UVB (40 mJ/cm²). C, Real-time PCR analysis of SIRT6 mRNA level in NHEK cells treated as in B. D, Immunoblot analysis of SIRT6, p-AKT, AKT and GAPDH in NC- or siAKT1-transfected NHEK cells at 24 h post-UVB (40 mJ/cm²). E, Immunoblot analysis of SIRT6, p-AKT, AKT and GAPDH in NHEK cells at 72h post-transfection with NC or siSIRT6. These results were obtained from three independent experiments. F, a schematic for the role of SIRT6 in skin carcinogenesis.