p53-Suppressed Oncogene TET1 Prevents Cellular Aging in Lung Cancer

Abstract

The role of transcriptional regulator ten-eleven translocation methylcytosine dioxygenease 1 (TET1) has not been well characterized in lung cancer. Here we show that TET1 is overexpressed in adenocarcinoma and squamous cell carcinomas. TET1 knockdown reduced cell growth in vitro and in vivo and induced transcriptome reprogramming independent of its demethylating activity to affect key cancer signaling pathways. Wild-type p53 bound the TET1 promoter to suppress transcription, while p53 transversion mutations were most strongly associated with high TET1 expression. Knockdown of TET1 in p53-mutant cell lines induced senescence through a program involving generalized genomic instability manifested by DNA single- and double-strand breaks and induction of p21 that was synergistic with cisplatin and doxorubicin. These data identify TET1 as an oncogene in lung cancer whose gain of function via loss of p53 may be exploited through targeted therapy-induced senescence. SIGNIFICANCE: These studies identify TET1 as an oncogene in lung cancer whose gain of function following loss of p53 may be exploited by targeted therapy-induced senescence.See related commentary by Kondo, p. 1751.

Fig. 1.. TET1 gene expression is elevated in human lung tumor samples and NSCLC derived cell lines and contributes to cell growth.

TET1 mRNA levels were assessed in AdC and SCC tumor samples (n = 25 each) compared to paired normal lung tissue (A). TET1 mRNA levels in thirteen AdC-, two large cell- (LC) and six SCC-derived cell lines were compared to mean value for HBEC cell lines (B). Horizontal line indicates 2-fold increase in TET1 levels. TET1 protein levels in NSCLC cell lines and quantification (C). Expression of TET1 in lung cancer tumors in TCGA from AdC and SCC tumors compared to normal lung tissues, (p < 0.0001, D). H1299, H1975, H226, H441, A549 and Calu6 cells were transfected with control vs. TET1 siRNA and analyzed for proliferation (E) and colony formation (F) 120 h after transfection. H1299 and H226 cells transfected with control vs. TET1 siRNA were used for xenograft tumor formation in nude mice with assessment of tumor volume (G) and mass (H). Results are presented as an average of three (in vitro) or two (in vivo) experiments ±SEM, p *<0.05, **<0.01, or ***<0.001.

Fig. 2.. TET1 expression in NSCLC cell lines is regulated by p53 binding in its proximal promoter region.

Seven deletion variants of TET1 promoter ranging from −1541bp to +29bp from transcription start site were cloned (A). Transcriptional activity of the construct carrying full TET1 promoter (−1541bp to +29bp) was analyzed in NSCLCs and HBECs with various TET1 mRNA levels (B). Deletion variants were tested in three NSCLCs lines to identify minimal active region of TET1 promoter (C). Predicted transcription factor (TF) binding sites located in −192bp/+29bp TET1 promoter fragment were identified in silico using PROMO software (D). Cells characterized by p53-null (H1299), p53-mut (H1975, Calu6, H1993 and SW900) and WT p53 (A549 and H2228) status were evaluated using ChIP for p53 enrichment on the TET1 endogenous promoter (E). H1299 cells with temperature induced WT p53 expression (cultured at 32°C) were analyzed for −192bp/+29bp TET1 promoter activity (F) and endogenous TET1 mRNA levels (G) and normalized to WT p53-negative control (cultured at 37°C). Cells lines characterized by p53-null (H1299) or p53-mut (Calu6, H1975 and SW900) status were transfected with WT p53-coding vs. GFP-coding adenoviral vectors and were analyzed for TET1 mRNA levels 48 h later with normalization to PCNA (H). Cells characterized by WT p53 status (A549, H2228 and H2023) were transfected with p53-targeting vs. control siRNA and analyzed 48 h later for TET1 mRNA (I). Experiments were performed in triplicate and presented as mean ±SEM, p *<0.05, **<0.01, or ***<0.001.

Fig. 3.. TET1 overexpression prevents NSCLC cells from cellular senescence and genomic instability.

H1299, H1975 and H226 cells were transfected with control vs. TET1 siRNA and assessed for morphology (A) and activity of β-galactosidase (B) using microscopy and β-galactosidase activity assay (scale bar = 100μm, representative images for H1299 shown). Percentage of cells positive for β-galactosidase was calculated (C). Levels of p21 mRNA were quantified 72 h and 96 h after transfection of H1299 and H1975 cells with control vs. TET1 siRNA (D), and TET1 and p21 protein was detected 96 h after transfection (E). The level of DNA double-strand breaks was assessed in H1299 and H1975 cell lines 72 h after transfection with control vs. TET1 siRNA using the micronuclei assay (F), H1299 cells were also analyzed for DNA strand breaks using the comet assay (G), and for DNA damage response activation associated with γH2AX translocation using immunofluorescence (H). The percentage of cells positive for γH2AX foci was calculated (I). Ingenuity Pathway Analysis was used to identify pathways differentially expressed in AdC and SCC tumors from TCGA characterized by TET1^high vs. TET1^low gene expression (J, K). Results are presented as an average of three experiments ±SEM, p *<0.05, **<0.01, or ***<0.001.

Fig. 4.. Depletion of WT p53 expression or function sensitizes cells to TET1 knock down-induced cellular senescence.

A549 cells were transfected with control, TET1, p53 or TET1+p53 siRNAs. Analysis of cell morphology using microscopy was performed 120 h after transfection (A). Assessment of DNA double-strand breaks was performed using micronuclei assay 72h after transfection (B). Detection of β-galactosidase activity (C) and analysis of p21 mRNA levels (D) were performed 120 h and 96 h after transfection, respectively. Number of cells was assessed 120 h after transfection (E). In chemical p53 inhibition experiments, A549 cells were incubated with 30 μM of pifithrine-α (PTF-α) for 48 h and analyzed for TET1 mRNA (F). Cells cultured with PTF-α and transfected with control vs. TET1 siRNAs were analyzed for β-galactosidase activity (G) and for cell number (H) 120 h after transfection. Experiments were performed in triplicate and presented as mean ±SEM, p *<0.05.

Fig. 5.. TET1 reprograms the lung cancer cell transcriptome via CpG oxidation-independent mechanism that in part involves chromatin remodeling through H3K9 di-methylation.

The Illumina Whole-Genome Gene Expression BeadChip followed by Ingenuity pathway analysis were used to identify pathways changed in control vs. TET1 knockdown H1299 and A549 cells 96 h after transfection with siRNA (A, B). Genomic DNA isolated from HBEC4, A549, H1975 and H1299 cells was analyzed for 5-hmC levels using Dot blot (C). Levels of di-methylated H3K9 and total H3 with normalization to β-actin were measured using Western blot in control vs. TET1 knockdown H1299 and H1975 cells 72 h after transfection with siRNA (D).

Fig. 6.. TET1 targeting sensitizes lung cancer cells to therapy-induced senescence and growth reduction.

Control vs. TET1 knockdown H1299 cells were incubated with cisplatin (CDDP) or doxorubicin (DOX) 24 h after transfection for 48 h and analyzed for p21 mRNA (A, B). Cells incubated with CDDP or DOX for 96 h (120 h after transfection with control vs. TET1 siRNAs) were analyzed for senescence using the β-galactosidase assay (C) and for total number of cells (D, E). Results are presented as an average of three experiments ±SEM, p *<0.05, **<0.01, or ***<0.001. Statistical comparison was performed for CDDP- vs. TET1 KD+CDDP-treated cells and DOX- vs. TET1 KD+DOX-treated cells. The mechanism of TET1 gene overexpression and function in lung cancer is depicted (F).