Fig 2: General ROS content in 6DT1 cells.A, 6DT1 CLIC4-WT cells were either left untreated or treated with 500 nM H2O2 for 24 h and incubated with 0.5 µM Mito-tracker orange for Mito-Tracker Orange CMTMRos (orange) for 30 min before fixation with 4% paraformaldehyde. Cells were then prepared for staining with CLIC4-specific antibody and Alexafluor-488 (green) and DAPI (blue) for nuclear staining. Cells were examined with confocal microscopy for immunofluorescence at 63× magnification. B, CLIC4-WT and CLIC4 KO (CLIC4-KO clone sg# 1) 6DT1 cells were incubated with 1 µM H2O2 for 24 h and stained with 3 µM CellROX green for 30 min and examined by confocal microscopy at 63× optical magnification. Green fluorescence is indicative of ROS levels and blue is Hoechst nuclear staining. C, stacked histogram plot of a representative FACS analysis of WT or KO clone sg# 1 6DT1 cells at 80% confluence that were treated with 1 µM H2O2 or 50 µM NAC or both for 24 h followed by addition of 5 µM DCFDA in 1× PBS for 30 min. The shift in cell populations to the right under different treatments represents the increase in DCFDA fluorescence. D, the mean fluorescence intensity of three independent experiments of 6DT1 cells with indicated treatments, incubated with DCFDA, and evaluated by FACS. The results were plotted using prism pad 7 and data were presented as means ± SD of three independent experimental repeats; *p < 0.05. CLIC4, chloride intracellular channel-4; ROS, reactive oxygen species; NAC, N-acetyl cysteine.

Fig 3: CLIC4-deficient 6DT1 cells have higher mitochondrial superoxides and H2O2content.A, CLIC4-WT and KO sg# 1 6DT1 cells were grown to 80% confluence in phenol red–free Dulbecco’s modified Eagle’s medium incubated with/without 1 µM H2O2 for 24 h and exposed to 2 µM MitoSOX red in HBSS for 10 min. Blue is Hoechst dye for nuclear staining. Live cell confocal imaging was performed at 63× magnification. B, CLIC4-WT and KO clone sg# 1 6DT1 cells were untreated (NT) or treated with 1 µM H2O2 or 50 µM NAC or both for 24 h, exposed to 2 µM of MitoSOX red for 10 min, and harvested with trypsin-EDTA and prepared for FACS analysis. The representative stacked histogram plots of FACS analysis indicate the shift in cell populations toward the right suggesting higher MitoSOX fluorescence. C, averaged mean fluorescence intensity reading plotted from three independent repeats of FACS analysis using MitoSOX red on cells treated as in (B) to depict the relative superoxide content by genotype and treatment. D, CLIC4-WT, KO clone sg# 1, and KO+CLIC4 lentivirus or control vector (Empty) 6DT1 cells were untreated or incubated with 1 µM H2O2 only or 50 µM NAC only or with both (as indicated with the plus + signs under the plot) for 24 h. Amplite substrate was added to fresh cell lysates for 30 min to quantify H2O2 content measuring fluorescence at 650 nm. E, peroxidase activity was determined in 6DT1 CLIC4-WT cells, CLIC4-KO clone sg# 1, and KO+CLIC4 untreated or treated with 1 µM H2O2 with or without 50 µM NAC for 24 h. The red-fluorescent oxidation product, resorufin, indicative of peroxidase activity in 6DT1 cell lysates was measured at 530 to 560/590 nm. *p < 0.05, **p < 0.005 and ***p < 0.0005. CLIC4, chloride intracellular channel-4; NAC, N-acetyl cysteine.

Fig 4: CLIC4 is expressed in human breast cancers and varies with patient age and survival probability.(A) Representative distribution of CLIC4 expression in human breast cancer tissue arrays by IHC. Scale bar, 100 µm. Each panel illustrates a different tumor. (B) Box plots of CLIC4 expression in breast tumors separated by stage from the METABRIC database. (C) Correlation analysis of CLIC4 mRNA expression and patient age in human breast cancer patients (FPKM = fragments per kilobase of exon per million reads mapped). Left panel, TCGA-BRCA (n = 1075). Right panel, METABRIC (n = 1992). (D) Comparison of CLIC4 expression in cancers from women with breast cancer below and above 55 years of age (left panel, n = 1074). Comparison of CLIC4 expression in cancers from women with triple-negative breast cancer (TNBC) and other breast cancer types (right panel, n = 1075). Source data from TCGA-BRCA (FPKM = fragments per kilobase of exon per million reads mapped). (E) METABRIC and TCGA-BRCA breast cancer datasets, with 75th percentile as the cut-off and 10-year censored data, show the high CLIC4 group had poor survival compared to the low CLIC4 expression group compared by the Kaplan-Meier curves.

Fig 5: Clic4 host deficiency impairs neovascularization of the primary tumor.(A) Representative CD31 immunostaining in primary tumors from Clic4 wildtype (WT) and knockout (KO) tumor-bearing mice at 28 days post-implantation. Scale bar, 200 µm. (B) Quantification of tumor vascularization in primary tumors from Clic4 wildtype (WT) and knockout (KO) tumor-bearing mice at 28 days post-implantation (n = 3 per group). Values represent the mean ± s.d. from whole tissue sections analyzed by Image Scope. P values were obtained after unpaired t test. (C) Quantification of solid tumor expansive density area, vimentin immuno-labeled area, and necrosis area in primary tumors from Clic4 wildtype (WT) and knockout (KO) tumor-bearing mice at 28 days post-implantation (n = 3 per group). Values represent the mean ± s.d. from whole tissue sections analyzed by HALO imaging software. P values were obtained after unpaired t test. (D) Representative vimentin immunostaining (left) and pseudo-colored analysis mask (right) showing the segmentation of tumor (red), necrosis (green), and intervening loose areolar and adipose connective tissue areas (yellow) in primary tumors that were used for the quantification shown in (C). Scale bars, 50 µm (left) and 500 µm (right).