Fig 2: Mutation analysis of ATP5B. The top 20 mutated genes in the breast cancer subsets stratified by ATP5B expression. The red frame represents the mutation with a statistically significant difference (p < 0.05). At the bottom is a forest plot of significantly mutated genes. Consider only genes with at least three samples mutated in one of the cohort for analysis.

Fig 3: ATP5B expression and functional enrichment in breast cancer. (A) Comparison of ATP5B mRNA expression between tumors and normal tissues, stages I–IV, and each subtype, including luminal, HER2 positive, and triple-negative cancers, in the TCGA BRCA data. (B) Representative IHC staining of the control, low and high expression of ATP5B protein in breast cancer; scale bar, 20 µm. The relative protein level of ATP5B in breast cancer tissues compared with adjacent normal tissue, stages I–III, as well as the three molecular subtypes. The data represent the median with interquartile range. (C) Volcano plot of differentially expression gene profiles between the ATP5B-high and ATP5B-low groups. The cut-off points were set at |log2(fold change)| = 1.5 and FDR < 0.05. GO annotation analysis of the overlapping genes in molecular function. Each circle represents a term, and its size represents the counts of the involved genes. (D) The three enrichment plots from the GSEA, including DNA damage recognition, E-cadherin nascent pathway and the PLK1 pathway.

Fig 4: Survival analysis of ATP5B based on clinical samples and online databases. (A) Kaplan-Meier (KM) survival curve comparing patients with high and low expression of ATP5B in TMA cohorts. Forest plot of multivariable Cox regression analysis shows that ATP5B is an independent risk factor for OS. Independent prognostic factors, including ATP5B expression and other clinical parameters, were evaluated using both univariate and multivariate Cox proportional hazards analyses. The HRs are presented as the median with 95% confidence interval. Red indicates the significant differences (p < 0.05). (B) Overall survival of ATP5B based on the TCGA (total 1096 breast cancer samples, left) and the METABRIC (total 1904 breast cancer samples, right) databases.

Fig 5: Assembly of OXPHOS complexes BN-PAGE and subsequent immuno-blotting of mitochondrial extracts from indicated HeLa cell lines showing the assembly of F1FO ATP synthase and binding of IF1. Monomers (M), dimers (D), and oligomers (O) were detected. Left panel: Coomassie-stained gel. Right panels: Immunoblotting and detection of ATP synthase and IF1. WT: wild-type HeLa cells; IF*: cells expressing IF1-H49K-HA. Stars indicate oligomeric IF1 forms. I: complex I; III: complex III.Relative levels of ATP synthase monomers, dimers, and oligomers. Semi-quantification of immunoblot shown in (A).Assembly of F1FO ATP synthase SU e-sEcGFP (Sue#) into monomeric, dimeric, and oligomeric ATP synthase (CV). Proteins were separated by BN-PAGE and immunoblotted as indicated. Left panel: detection of subunit ß (a-ATP5B); right panel: detection of sEcGFP with a-GFP; O: oligomeric ATP synthase; D: dimeric (blue **); M: monomeric F1F0 ATP synthase (blue*).Assembly of sEcGFP-tagged subunits CoxVIIIa-sEcGFP (CIV#) into complex IV, and assembly of subunits ?-sEcGFP (SU ? #) and SU e-sEcGFP (Su e#) into ATP synthase. Proteins were separated by BN-PAGE and immunoblotted. Lanes #1-#10 were cut into three pieces for further evaluation. Lanes 1–2: complex IV assembly in WT and CoxVIIIa-sEcGFP-expressing cells, detection of CIV with anti-Cox1. Lanes 3–4: complex V assembly in WT and SU ?-sEcGFP-expressing cells, detection with anti-SU ?. Lanes 8–10: assembly of CIV# into CIV, SU ?-sEcGFP (SU?#) into CV, and SU e-sEcGFP (SUe#) into CV. At lane #6 in the Coomassie-stained gel, the positions of molecular weight markers (M) are shown. SC: respiratory supercomplexes; CIV2: complex IV dimers; CIV: complex IV monomer; s1: F1-c8 subcomplexes; s2: subcomplexes of CV containing subunits b, e, and g (He et al, 2018). Blue **: dimeric F1F0 ATP synthase; blue*: monomeric F1F0 ATP synthase. Red asterisks: respiratory supercomplexes. Source data are available online for this figure.