Manganese superoxide dismutase is a p53-regulated gene that switches cancers between early and advanced stages

Abstract

Manganese superoxide dismutase (MnSOD) plays a critical role in the survival of aerobic life, and its aberrant expression has been implicated in carcinogenesis and tumor resistance to therapy. However, despite extensive studies in MnSOD regulation and its role in cancer, when and how the alteration of MnSOD expression occurs during the process of tumor development in vivo are unknown. Here, we generated transgenic mice expressing a luciferase reporter gene under the control of human MnSOD promoter-enhancer elements and investigated the changes of MnSOD transcription using the 7,12-dimethylbenz(α)anthracene (DMBA)/12-O-tetradecanoylphorbol-l3-acetate (TPA) multistage skin carcinogenesis model. The results show that MnSOD expression was suppressed at a very early stage but increased at late stages of skin carcinogenesis. The suppression and subsequent restoration of MnSOD expression were mediated by two transcription-factors, Sp1 and p53. Exposure to DMBA and TPA activated p53 and decreased MnSOD expression via p53-mediated suppression of Sp1 binding to the MnSOD promoter in normal-appearing skin and benign papillomas. In squamous cell carcinomas, Sp1 binding increased because of the loss of functional p53. We used chromatin immunoprecipitation, electrophoretic mobility shift assay, and both knockdown and overexpression of Sp1 and p53 to verify their roles in the expression of MnSOD at each stage of cancer development. The results identify MnSOD as a p53-regulated gene that switches between early and advanced stages of cancer. These findings also provide strong support for the development of means to reactivate p53 for the prevention of tumor progression.

Figure 1. Human MnSOD transgenic mouse model and MnSOD regulation in vivo

Human MnSOD promoter-enhancer-driven luciferase gene construct used to generate transgenic mice (A). Transgenic mice showing a band for the intronic fragment (I2E) were obtained by restriction digestion of genomic DNA by kpn1 and bglII (B). Schematic diagram outlining the method of DMBA and TPA treatment and non-invasive imaging (C). Bioluminescence images of mice treated with DMBA followed by TPA were acquired by CCD camera (bottom panel). Photon counts were estimated within the defined gated area on the image (top panel) (D). After 25 weeks, the animals were humanely euthanized and skin and tumor tissues were harvested and the luciferase activity (E), MnSOD mRNA (F), MnSOD protein (G) and MSOD activity (H) were measured. Data are presented as the mean ± SD, with significant differences from control indicated by *p<0.05 and **p<0.01.

Figure 2. Transcription factor binding to the MnSOD promoter and enhancer in vivo

EMSA was performed in purified nuclear extract from skin and tumor tissues. For super-shift experiments, the EMSA reaction mixture was incubated with 1 μg antibody or IgG alone. The arrows point to the protein-DNA complex and super-shifted protein-antibody complexes (top panel). NF-κB-DNA binding complex was densitometrically scanned and expressed as a relative quantity (bottom panel) (A), Sp1 binding activity (top panel). The Sp1-DNA binding complexes were quantified (bottom panel) (B). Western analysis of p50 and p65 was performed in the purified nuclear extracts. Data were quantified and the relative levels of NF-κB proteins were estimated (C). Association of transcription factors with the enhancer region of the MnSOD gene was evaluated by ChIP assay in isolated skin and tumor tissues (D). The immunoprecipitated proteins were detected by Western analysis (E). Data shown are representative of three independent experiments. Significantly different from control group, **p<0.01.

Figure 3. Increased MnSOD transcription, activity and protein in late-stage tumorigenesis

MnSOD reporter gene activity, protein levels and activity were evaluated in normal skin tissues and tissues bearing papillomas or SCC. Luciferase activity (A), protein levels (B), and MnSOD activity (C). Ap1 binding activity was measured by EMSA. The Ap1 DNA-protein complex bands were densitometrically scanned and the relative levels were determined (bottom panel) (D). PCNA levels were detected in nuclear extracts by Western blotting. The protein bands were densitometrically scanned and the relative levels were determined after normalization to lamin C (E). Each group of tissues was collected from 10 individual animals. Data presented are the mean ± SD, and significant differences from control are indicated by *p<0.05 and **p<0.01.

Figure 4. Alteration of transcription factor binding activity and nuclear p53 levels

DNA binding activity of each transcription factor was evaluated by EMSA (top panel) and quantified (bottom panel) (A) NF-κB, (B) Sp1. (C) p53. The quantification of p53-DNA complexes is shown (bottom panel) (C). Data presented are the mean ± SD, and significant differences from control, *p<0.05 and **p<0.01.

Figure 5. Knockdown of p53 alone or overexpression of Sp1 with subsequent p53 knockdown enhances Sp1-DNA binding activity and MnSOD transcription in vitro

Mouse epithelial cells (JB6) were co-transfected with control siRNA or p53 siRNA with or without Sp1 expression vector and with MnSOD reporter vector. Luciferase activity was normalized to beta-galactosidase activity and represented as a measure of MnSOD transcription (A). Suppression of p53 protein and increase of Sp1 expression upon transfection of p53 siRNA and Sp1 expression vector, respectively, were verified by Western blotting. Endogenous MnSOD and p21 protein levels were verified by reprobing the membrane with MnSOD or p21 antibody (top panel). The gels were densitometrically scanned and the relative levels were normalized to GAPDH as an internal control (bottom panel) (B). Sp1 binding activity was evaluated by EMSA following transfection of p53 siRNA with or without Sp1 expression vector (top panel). The binding complexes were densitometrically scanned and quantified (bottom panel) (C). Data consist of three representative experiments and are expressed as the mean ± SD. Significantly different from control, *p<0.05 and **p<0.01.

Figure 6. Knockdown of p53 enhances transformation phenotype and increases matrigel invasiveness in JB6 cells

JB6 cells were plated on a soft-agar dish and cultured for 14 days for colony formation. For each treatment, 6–10 dishes were used and a representative picture is shown (A). Microscopic images of transformed colonies were taken from six randomly selected fields in each dish (6–10 dishes per group). A representative microscopic view is shown (B). The number of transformed colonies was counted and quantified (C). Cells were seeded onto three-dimensional migration chamber inserts and the matrigel invasion assays were performed as described in materials and methods (D). Matrigel invasion was quantified using six different random fields per insert (4 inserts per group). Migrated cells were counted in the matrigel and control chambers following staining with crystal violet. Data were normalized to the cells that migrated from the control chamber (E). Each bar represents the average of the mean ± SD of 6 different randomly selected fields per dish (6 dishes per group). Significantly different from control, *p<0.05 and **p<0.01.