Fig 2: Upregulation of RIPK1 attenuates AQP1-driven TNBC progression and lung metastasis in vivo.A Flow chart of the in vivo orthotopic breast carcinoma model establishment and experiments. B Statistical analysis of consecutive tumor volume measurements until 32 days after 4T1 cell implantation (n = 6 for each group). C Representative images of tumors in empty-vector control, AQP1-overexpressing, AQP1–RIPK1-overexpressing 4T1 cell transplanted BALB/c mice. D Statistical analysis of the final tumor volume measurements at 32 days after 4T1 cell implantation (n = 6 for each group). E Representative images of gross lung and micrometastatic nodules (×40 magnification) from empty-vector control, AQP1-overexpressing, AQP1–RIPK1-overexpressing 4T1 cell transplanted BALB/c mice. Scale bar: 200 μm. F Statistical analysis of the metastatic lesion area measurements at 32 days after 4T1 cell implantation (n = 6 for each group). G Kaplan–Meier survival plot showing the overall survival of BALB/c mice orthotopically injected with empty-vector control, AQP1-overexpressing, AQP1–RIPK1-overexpressing 4T1 cells (n = 11 for each group). Abbreviations: ad-AQP1, AQP1 overexpression; ad-RIPK1, RIPK1 overexpression. Note: **p < 0.01 vs. empty-vector control; ##p < 0.01 vs. ad-AQP1 group.

Fig 3: Inhibition of RIPK1 is required for AQP1-driven TNBC cell proliferation, migration, and invasion in vitro.A Representative blotting of Western blot analysis of MDA-MB-231 and 4T1 cells transfected with empty vector, AQP1-3×FLAG, RIPK1-HA, and/or RIPK1-siRNA. B Relative viability of indicated MDA-MB-231 cells at 24, 48, and 72 hours after cell seeding. The cell viability is displayed by the absorbance at 450 nm wavelength in the CCK-8 assay (n = 3). C Same experiment as in B, but using 4T1 cells (n = 3). D Representative images of indicated MDA-MB-231 cells in the wound healing assay at 0 and 24 hours after micropipette scratching (×100 magnification). E Quantification of the wound healing percentage at 24 hours after micropipette scratching in indicated MDA-MB-231 cell cultures (n = 3). F, G Same experiment as in D and E, but using 4T1 cells (n = 3). H Representative images of indicated MDA-MB-231 cells in the transwell migration and invasion assays (×200 magnification). Scale bar: 100 μm. I Quantification of migrated and invaded cells for indicated MDA-MB-231 cells (n = 3). J, K Same experiment as in H and I, but using 4T1 cells (n = 3). Abbreviations: ad-AQP1, AQP1 overexpression; ad-RIPK1, RIPK1 overexpression; si-RIPK1, RIPK1 knockdown with siRNA. Note: *p < 0.05 and **p < 0.01 vs. empty-vector control; #p < 0.05 and ##p < 0.01 vs. ad-AQP1 group; &p < 0.05 and &&p < 0.01 vs. ad-AQP1 group.

Fig 4: AQP1 binds with RIPK1 in TNBC.A Linear regression graph. Strong negative correlation (R2 = 0.50, p < 0.001) was revealed in TNBC between the abundance of AQP1 and RIPK1 detected by Western blotting. B Co-immunoprecipitation of RIPK1 from TNBC and normal breast tissue showing AQP1 binds with RIPK1. Abbreviations: N, normal breast tissue; C, TNBC. C Representative images of anti-AQP1 (green) and anti-RIPK1 (red) immunofluorescence staining of TNBC and normal breast tissue (×400 magnification). Orange/yellow fluorescence in the merged images represents the colocalization of AQP1 and RIPK1. D Representative images of anti-AQP1 (green) and anti-RIPK1 (red) immunofluorescence staining of MDA-MB-231 and 4T1 cells stably overexpressing AQP1 and RIPK1. Orange/yellow fluorescence in the merged images represents the colocalization of AQP1 and RIPK1. Scale bar: 20 μm.

Fig 5: AQP1 binds to the D324 site of RIPK1 and excessively activates caspase-8-mediated RIPK1 cleavage.A Schematic diagram of RIPK1 structure and functional sites. B Co-immunoprecipitation of FLAG in HEK-293T cells expressing AQP1-3×FLAG transformed with plasmids expressing RIPK1, RIPK1-K377R, RIPK1-K45A, RIPK1-K530A, RIPK1-I539A, RIPK1-S161A, or RIPK1-D324K. RIPK1-D324K was revealed with reduced interaction with AQP1. C Representative blotting of Western blot analysis of RIPK1 and cleaved RIPK1 (cl-RIPK1) in MDA-MB-231 cells stably overexpressing AQP1 and RIPK1D324K. Statistical analysis of the cl-RIPK1/RIPK1 ratio showed significant difference among empty-vector control, AQP1-overexpressing, AQP1–RIPK1D324K-overexpressing cells (n = 3). D Relative viability of indicated MDA-MB-231 cells at 72 hours after cell seeding. The cell viability is displayed by the absorbance at 450 nm wavelength in the CCK-8 assay (n = 3). E Representative images and quantification of indicated MDA-MB-231 cells in the wound healing assay at 0 and 24 hours after micropipette scratching (×100 magnification, n = 3). F Representative images and quantification of indicated MDA-MB-231 cells in the transwell migration and invasion assays (×200 magnification, n = 3). Scale bar: 100 μm. G Schematic model of the role of cytoplasmic AQP1 in regulating TNBC cell apoptosis and necroptosis by facilitating caspase-8-mediated RIPK1 cleavage. Abbreviations: RHIM, receptor-interacting protein homotypic interacting motif; ad-AQP1, AQP1 overexpression; ad-RIPK1 RIPK1, overexpression; ad-RIPK1D324K, RIPK1D324K overexpression. Note: **p < 0.01 vs. empty-vector control and ##p < 0.01 vs. ad-AQP1 group for C; *p < 0.05, **p < 0.01 vs. ad-AQP1 group; #p < 0.05 vs. ad-AQP1 + ad-RIPK1 group for D–F.