Fig 2: Blockade of PPP1R14C phosphorylation inhibited triple-negative breast cancer (TNBC) progression in vitro and in vivo. (A–E) MTT (A), colony formation (B), anchorage-independent growth (C), transwell (D), and wound healing (E) assays were performed in vector, PPP1R14C-wild-type (WT), or PPP1R14C (T73A) mutant in MDA-MB-231 and SUM159PT cells. Data represent the means ± S.D. of three independent experiments. A two-tailed Student's t-test was used. (F) SUM159PT cells transducing with WT or T73A plasmids were orthotopically injected into mice (1 × 106/injection, n = 6/group). The tumour volumes in each group were shown. (G) IHC of PPP1R14C and Ki-67 staining showed in the indicated xenografts. Scale bars represent 50 µm. (H) In vivo metastasis assays of vector, WT and T73A groups. The visible surface metastatic lesions were counted. (I) Kaplan–Meier survival curves of mice. (J) The visible surface metastatic lesions of the lungs were counted. Two-tailed Student's t-test and log-rank test were used.

Fig 3: PPP1R14C facilitates triple-negative breast cancer (TNBC) tumour growth and metastasis. (A) Control, PPP1R14C-overexpressing, or PPP1R14C-silenced SUM159PT cells were orthotopically injected into mice (1 × 106/injection, n = 6/group). The tumour volumes in each group were shown. (B) IHC of Ki-67 staining showed in the indicated xenografts. Scale bars represent 50 µm. (C) In vivo metastasis assays of PPP1R14C-overexpressing, PPP1R14C-silenced, and control MDA-MB-231 cells. Lung metastatic burden was monitored weekly using bioluminescent imaging (BLI). Representative BLI images of the lungs were shown. The visible surface metastatic lesions were counted. (D) Kaplan–Meier survival curves of mice. (E) Lung colonization and spontaneous metastasis models of PPP1R14C-overexpressing, PPP1R14C-silenced, and control 4T1 cells. BLI quantification of lung metastasis of the indicated cells. (F) The visible surface metastatic lesions of the lungs were counted. Two-tailed Student's t-test and log-rank test were used.

Fig 4: Upregulation of PPP1R14C is associated with poor prognosis in patients with triple-negative breast cancer (TNBC). (A) Representative images of PPP1R14C staining in normal breast tissues, non-TNBC, and TNBC tissues (left panel). The staining index (SI) distribution of each group was indicated (right panel) and the Mann–Whitney U-test was used for analysis. Scale bars represent 50 µm. (B) The distribution and correlation between PPP1R14C staining and T stage, and relapse status in patients with TNBC. The ?2 test was used for analysis. Rel-, no relapse; Rel +, with relapse. (C) Kaplan–Meier 5-year overall survival (OS) and relapse-free survival (RFS) curve for patients with TNBC stratified by low (n = 41) and high PPP1R14C expression (n = 59, log-rank test). HR, hazard ratio. (D) The Kaplan–Meier Plotter (http://kmplot.com/analysis) program was used to analyse relapse-free survival of all patients with breast cancer and the basal-like subgroup (Pietenpol subtype). All settings were left at default values except for: gene symbol (PPP1R14C), survival (RFS), and auto-select best cutoff (on). (E) Multivariate Cox regression analysis to evaluate the significance of the association between PPP1R14C signature and OS, and RFS in the presence of other clinical variables.

Fig 5: PPP1R14C promotes triple-negative breast cancer (TNBC) progression in vitro. (A) Western blotting analysis of PPP1R14C in MDA-MB-231 and SUM159PT cells stably transduced with PPP1R14C-overexpressing and PPP1R14C-silencing plasmids. GAPDH was used as a loading control. (B–G) MTT (B), colony formation (C), anchorage-independent growth (D), transwell (E), wound healing (F) assays, and flow cytometric analysis (G) were performed in the indicated cells. Scale bars represent 50 µm. A two-tailed Student's t-test was used. Data represent the means ± S.D. of three independent experiments.