Fig 1.
Generation and validation of NOX4 knockout (KO) HeLa cell lines.
(A) T7 endonuclease 1(T7E1) assay. HeLa cells were transiently transfected with either pX459/NOX4 gRNA #1 or pX459/NOX4 gRNA #2, and genomic PCR products were analyzed by T7E1 assay (left panel). HeLa cells transfected with pX459/NOX4 gRNA’s were selected with puromycin and single colonies were isolated. The genomic PCR product of each clone was analyzed by T7E1 assay (right panel). Experiments were repeated three times with similar observations, and representative images are shown. (B) DNA sequence analysis showed the presence of the NOX4 mutation in clones #1–2, #1–5 and #1–6. The black arrows indicate the heterogeneous genomic DNA sequences in each cell lines. (C) CRISPR-Cas9 introduced the insertion and deletion (indel) mutation in the target sites. (D) Western blot of NOX4 knockout cell lines. Equal amount of HeLa control and NOX4 knockout cell lysates were probed with anti-NOX4 antibody. Experiments were repeated three times with similar observations, and representative data is shown. (E) NOX4 knockout did not influence NOX1, NOX2, NOX3, NOX5, DUOX1 and DUOX2 levels. Protein levels of NOX1, NOX2, NOX3, NOX5, DUOX1 and DUOX2 in parental HeLa cells and three clones of NOX4 knockout cells. (F) NOX4 knockout showed lower H2O2 production. The levels of H2O2 were measured with the Amplex Red assay in triplicate. The graph shows the average and the standard deviation (SD). Control vs knockout cells. *: P <0.005, **: P <0.0001.
Fig 2.
NOX4 knockout inhibits cell proliferation.
(A) Proliferation of three NOX4 knockout cell lines was quantitated by MTT assay. HeLa control and knockout clones were seeded into 24 well plate and the MTT assay were performed in quadruplicate. The graph shows the average and the standard deviation (SD). Control vs knockout cells. *: P <0.005, **: P <0.00005. (B) NOX4 knockout HeLa cells were analyzed with flow cytometry. The percentage of cells at sub-G1 and S/G2/M was measured, and is shown in the graph. Experiment was repeated three times with similar observations, a representative data is shown. (C) NOX4 knockout may affect activation of Akt phosphorylation. NOX4 depleted cells showed significantly lower levels of phospho-Akt in comparison to parental normal cell. However, NOX4 knockout did not affect phosphorylation of S6 Ribosomal Protein. Cells were serum starved by incubation in DMEM containing 1% FBS for 10 hrs, then exposed to EGF (20ng/ml) for 20 min before collection of cells for analysis. Experiments were repeated three times with similar observations, and representative results are shown.
Fig 3.
NOX4 knockout inhibits cell migration and invasion.
(A, B) NOX4 knockout inhibits the cell migration of HeLa cells. Equal numbers of cells were plated onto a 6 well plate, and cells were scraped. After 24 h and 48 h, the wound gaps were photographed using an inverted microscope and quantified using imageJ software. Bars: 100 μm. The graph shows the mean and the standard deviation (SD). Control vs NOX4 knockout cells. *: P <0.0001, **: P <0.00001, n = 6. (C, D) NOX4 knockout inhibits the cell invasion. Invasion of control and NOX4 knockout cells through matrigel in response to EGF (20 ng/ml for 24 h) was assessed by the invasion assay. Bars: 100 μm. The graph shows the average and the standard deviation (SD). Control vs NOX4 knockout cells. ***: P <0.005, ****: P <0.0005, n = 20. Experiments were repeated three times with similar observations, and representative images of the invasive cells are shown.
Fig 4.
NOX4 knockout reduces focal adhesion and invadopodium formation in HeLa cells.
(A) Cells were seeded in fibronectin coated glass-bottom dishes, and stained with anti-paxillin antibody (green) and anti-zyxin antibody (red). The images were analyzed with TIRF microscope. Bars: 20 μm. (B, C) The area distribution of focal adhesions in 20 cells were analyzed and shown in graph. (D) NOX4 is required for the efficient invadopodium formation. HeLa control and NOX4 knockout cells were cultured on coverslip coated with Cy3-gelatin (red) and stained with Cortactin (green). Gelatin degradation is visualized as darker areas. Arrowheads denote the gelatin degraded areas. Bars: 20 μm. (E) The degradation of gelatin was quantified using image analysis software. The graph shows the average and the standard error of the mean (SEM). HeLa control vs NOX4 knockout cells, *: P<0.05, n = 20.