DNA polymerase iota promotes EMT and metastasis of esophageal squamous cell carcinoma by interacting with USP7 to stabilize HIF-1α

Abstract

Esophageal squamous cell carcinoma (ESCC) is one of the most lethal cancer types, with a low 5-year survival rate of ~20%. Our prior research has suggested that DNA Polymerase iota (Pol ι), a member of Y-family DNA polymerase, plays a crucial role in the invasion and metastasis of ESCC. However, the underlying mechanism is not well understood. In this study, we utilized ChIP-PCR and luciferase reporter assays to investigate the binding of HIF-1α to the promoter of the Pol ι gene. Transwell, wound healing, and mouse models were employed to assess the impact of Pol ι and HIF-1α on the motility of ESCC cells. Co-immunoprecipitation and Western blot were carried out to explore the interaction between Pol ι and HIF-1α, while qRT-PCR and Western blot were conducted to confirm the regulation of Pol ι and HIF-1α on their downstream targets. Our results demonstrate that HIF-1α activates the transcription of the Pol ι gene in ESCC cells under hypoxic conditions. Furthermore, the knockdown of Pol ι impeded HIF-1α-induced invasion and metastasis. Additionally, we found that Pol ι regulates the expression of genes involved in epithelial-mesenchymal transition (EMT) and initiates EMT through the stabilization of HIF-1α. Mechanistically, Pol ι maintains the protein stability of HIF-1α by recruiting USP7 to mediate the deubiquitination of HIF-1α, with the residues 446-578 of Pol being crucial for the interaction between Pol ι and USP7. Collectively, our findings unveil a novel feedforward molecular axis of HIF-1α- Pol ι -USP7 in ESCC that contributes to ESCC metastasis. Hence, our results present an attractive target for intervention in ESCC.

Fig. 1. HIF-1α activated the transcription of Pol ι in ESCC.

A Correlation analysis of POL ι and HIF-1α mRNA expression in Human ESCC tissue (48 cases). B Immunohistochemistry staining for POL ι and HIF-1α in ESCC specimens (45 cases, 100×). C Correlation analysis of POL ι and HIF-1α protein expression in ESCC tissue with Pearson’s chi-square test. D qPCR analysis of Pol ι mRNA expression in ESCC cell lines. E Western blot analysis of Pol ι protein expression in ESCC cell lines. F Chromatin immunoprecipitation was conducted to assess the binding of HIF-1α to POL ι promoter region in ESCC tissues and in ECA-109 cells. G Pol ι promoter reporter luciferase activity was measured using the dual-luciferase reporter assay. H qPCR analysis of Pol ι mRNA expression in TE-1 and ECA-109 cells under normoxic and hypoxic conditions. I qPCR analysis of HIF-1α mRNA expression after exposure to 20% O₂ or 1%O₂. J, K The mRNA stability of Pol ι and HIF-1α in TE-1 and ECA-109 cells was determined under normoxic and hypoxic conditions (n = 3, bar, SD). The data were presented as mean ± SD with statistical significance denoted as *P < 0.05; **P < 0.01; and ns, P ≥ 0.05.

Fig. 2. Pol ι contributes to HIF-1α induced ESCC invasion and metastasis in vitro and in vivo.

A The levels of POL ι and HIF-1α protein were examined by western blot in ECA-109 cells. B Wound healing assay and transwell assays were performed to evaluate the migration and invasion ability of ECA-109 cells with or without overexpression of HIF-1α or knockdown of POL ι. C, D Quantitative analysis of migration and invasion abilities of ECA-109 cells. E Fluorescence imaging of live, lung and kidney from mice injected with ECA-109 cells via tail vein. F The average number of surface nodules per lung was calculated for each group of mice. G Representative images of HE staining of the lung metastatic nodules (40× and 200×). H Immunohistochemical analysis of POL ι and HIF-1α expression in the lung tissues from the mice. I Western blot analysis of POL ι and HIF-1α protein expression in POL ι overexpressed cells and POL ι knockout cells under hypoxic or normoxic. J, K qPCR analysis of POL ι and HIF-1α expression in POL ι overexpressed cells and knockout cells under hypoxic and normoxic. The data were presented as mean ± SD with statistical significance denoted as *P < 0.05; **P < 0.01; and ns, P ≥ 0.05.

Fig. 3. Pol ι initiates EMT by elevating HIF-1α and its downstream genes under hypoxic in ESCC cells.

A Giemsa staining of ECA-109-sgNC/ECA-109-sgPOLι cells and TE-1-NC/TE-1-POLι cells that were cultured under 20% O₂ or 1% O₂ conditions for 12 h (magnification: 400×). B Immunofluorescence staining for E-Cadherin, POL ι, and HIF-1α expression in ESCC cells under normoxic and hypoxic. C Transwell assay was performed to evaluate the cell migration and invasion ability. D, E The statistical results of transwell assays. F The adhesive ability of the ESCC cells was assessed by an adhesion assay. The data were presented as mean ± SD with statistical significance denoted as *P < 0.05; **P < 0.01; and ns, P ≥ 0.05. G, H EMT-related protein markers were analyzed using Western blot in ECA-109 cells and TE-1 cells after exposure to normoxic or hypoxic for 12 h.

Fig. 4. Pol ι inhibits the degradation of HIF-1α through the USP7-mediated deubiquitylation.

A CHX treatment and western blot assay were conducted to measure HIF-1α protein stability in ECA-109 cells. Cells were incubated with 1% O₂ for 12 h and then incubated in the presence of CHX for 0, 10, 30, 60,120, or 180 min. B Protein half-life curves were made based on the data. Data were presented as mean ± SD (n = 3). C A CHX chase assay was performed to analyze the half-life of the HIF-1α protein in TE-1 cells. D Protein half-life curves were made based on the data. Data were presented as mean ± SD (n = 3). E, F HIF-1α ubiquitination was determined by Western blot with anti-ubiquitin antibody in ECA-109 cells after treatment with 1% O₂ for 12 h. G HIF-1α ubiquitination was determined by western blot with anti-ubiquitin antibody in TE-1 cells after treatment with 1% O₂ for 12 h. H Silver staining was conducted to exhibit POL ι interacted proteins. I Mass spectrum analysis of peptides from POL ι and USP7 protein. J IP-western blot was conducted to determine the interaction between POL ι, USP7, and HIF-1α in HEK293T cells transfected with Flag- POL ι, Myc-USP7, and HA-HIF-1α. K Co-immunoprecipitation of endogenous HIF-1α, POL ι, and USP7 in TE-1 and ECA-109 cells. L, M GST pulldown assay was performed to determine the interaction between POL ι, USP7, and HIF-1α. N Co-immunoprecipitation (co-IP) of endogenous HIF-1α, POL ι, and USP7 in 20% O₂ or 1% O₂-exposeded TE-1-NC/TE-1-POL ι and ECA-109-shNC/ECA-109-shPOL ι cells.

Fig. 5. Pol ι enhances the binding of USP7 to HIF-1α for its deubiquitination via residues 446–578aa within the C-terminal of Pol ι.

A CHX and western blot assays were used to measure the HIF-1α protein stability in POL ι and USP7 knockdown ECA-109 cells. B Protein half-life curves were made based on the data in ECA-109 cells. C CHX and western blot assays were used to measure HIF-1α protein stability in POL ι and USP7 overexpressed TE-1 cells. D Protein half-life curves were made based on the data in TE-1 cells. Data were presented as mean ± SD (n = 3). E Western blot analysis of the HIF-1α-immunoprecipitated lysates with the anti-ubiquitin antibody in ECA-109 cells. F Western blot analysis of the HIF-1α-immunoprecipitated lysates with the anti-ubiquitin antibody in TE-1 cells. G, H Co-IP assay was performed to determine the binding ability of USP7, POL ι, and HIF-1α in POL ι/USP7 knockdown or overexpression ESCC cells incubated under 1 or 20% oxygen. I, J Quantitative analysis of the co-IP interaction data. K Construction of POL ι wild-type (WT) and deletion mutants with Flag-tag. L In vivo ubiquitination assay and Co-IP were conducted to detect HIF-1α ubiquitination and the binding of POLι mutants to USP7 in HEK293T cells.

Fig. 6. Blocking the specific binding of Pol ι and USP7 inhibits HIF-1α induced EMT in ESCC cells.

A Images and statistics of the migrated and invaded ECA-109 cells with Pol ι and USP7 knockdown (left panel). The numbers of normalized cell numbers were quantified (right panel). The data (mean ± SD, n = 3) are expressed as normalized cell numbers. **P < 0.01. B Transwell assay was conducted to evaluate the effects of Pol ι and USP7 co-expression on the migration and invasion of TE-1 cells. The numbers of normalized cell numbers were quantified (right panel). Representative images and statistical analysis. The data (mean ± S.D, n = 3) are expressed as normalized cell numbers. **P < 0.01. C, D Adhesion assays to examine adhesion capacity of Pol ι and USP7 knockdown/overexpression ECA-109 and TE-1 cells. E The EMT-related protein levels of Pol ι, USP7, HIF-1α, and EMT-related markers in ECA-109 and TE-1 cells were detected by western blot. F, G Promoter activities of the Snail and Slug genes in ECA-109 and TE-1 cells was measured using a Dual-Luciferase Reporter Assay. The data are presented as mean ± SD with statistical significance denoted as *P < 0.05; **P < 0.01; and ns, P ≥ 0.05. H Live animal imaging of mice at various time points after tail vein injection of ECA-109 cells expressing GFP. I In vivo imaging of mice after intraperitoneal inoculation of ECA-109 cells expressing GFP. J Mice showed liver and spleen metastases after intraperitoneal injection with ECA-109 cells expressing luciferase. K In vivo imaging of mice injected with ECA-109 cells expressing GFP via the left armpit. L Results of quantitative and statistical analyses of the bioluminescence signal intensities in the NC, shPOLι, shPOLι + HIF-1α, and shPOLι + USP7 groups. M Representative image of the transwell assay to evaluate the migration and invasion ability of TE-1 cells expressing different POLι mutants (right panel). The quantitative and statistical analyses are based on normalized cells per field (left panel). N Luciferase reporter assays were used to measure the snail and slug promoter activities in TE-1 cells transfected with empty vector, POLι-WT, and POLι-Δ1–4 plasmids. (n = 3, bar, SD). O EMT-associated protein expression in POLι mutants was analyzed through a western blot assay.

Fig. 7. Diagram of the feedforward loop of the HIF-1α-POLι-USP7 regulatory axis.

The figure was created using Figdraw.