Fig 2: κB-Ras deficiency promotes acinar-to-ductal metaplasia and Sox9 expression.a Resected tumours of 10-week-old 1SKO/R and pDKO/R mice (left) and exemplary haematoxylin/eosin staining of a section of 10-week-old pDKO/R mice (right). Scale bar: 500 µm. b Fluorescent IHC staining of CK19 (magenta) and amylase1 (green) in pancreatic sections of 1SKO/R and pDKO/R littermate mice (4 weeks old). Scale bar: 200 µm. c FACS analysis of pancreatic cell populations. Acinar cells were excluded from analysis based on FSC/SSC and residual cells analysed for ductal (MIC-1C3+) and acinar cell markers (MPx1+) by surface staining. d Fluorescent IHC staining of Cpa1 (magenta, acinar cell marker), Sox9 (green) and nuclei (DAPI, blue) in paraffin sections of 1SKO/R and pDKO/R littermate mice (4 weeks old). Scale bar: 200 µm.

Fig 3: The Ras-rescue screen. a Schematic diagram of the temperature-sensitive Ras-rescue system used to identify membrane-associated effectors in yeast. b Experimental setup for construction of the Ras*-effector fusion library and mutants that alleviated yeast growth inhibition. Ras*-effector-HA fusion proteins (labeled red-green) were expressed in cdc25ts and assayed for growth inhibition. Catalytically dead mutations (if known) were made in each effector that caused yeast growth defects, allowing for those effectors to be included in the screen. The catalytic residues for EspL2Ec, LpnELp, SseFSt, IpaJSf, and HopAA1-2Ps were unknown and, therefore, these effectors could not be included in the screen. c Growth assay for the cdc25ts yeast strain harboring the indicated effector genes grown at the non-restrictive temperature (25 °C) or the restrictive temperature (37 °C). Effectors that allow for yeast growth at the restrictive temperature reconstituted Ras* localization to the membrane. Effectors known to have transmembrane domains are colored in blue as confirmation of the screen

Fig 4: KIKK motif of KRAS4A functions by providing a positive charged region. a Morphology of cultured NIH3T3 cells stably expressing KRAS4AG12D/C180S with KIKK motif mutated to QIKK, KIQK, KIKQ or QIQK, QIKQ, KIQQ was observed under light microscope (original magnification, ×200). b, d Quantification of anchorage-independent growth of NIH3T3 cells expressing the indicated proteins. Colonies were counted on day 14 after plating in triplicate. Student t test was used for the statistical analysis. ****P < 0.0001. c Morphology of cultured NIH3T3 cells stably expressing KRAS4AG12D, KRAS4AG12D/C180S, KRAS4AG12D/QIQQ, or KRAS4AG12D/C180S/QIQQ was observed under light microscope (original magnification, ×200). e Serum-starved lysates of NIH3T3 cells expressing the indicated proteins were analyzed by western blotting to detect the phosphorylation levels of c-RAF, MEK, ERK1/2, AKT, and S6RP. Expressions of RAS, GFP, and β-actin were used as loading controls

Fig 5: Schematic diagram of the molecular mechanism underlying the promoting roles of the expression of CSE on the progression of breast cancer. CSE, cystathionine-γ-lyase; VEGF, vascular endothelial growth factor; PI3K, phosphatidylinositol 3-kinase; Akt, protein kinase B; FAK, focal adhesion kinase; Ras, rat sarcoma; Raf, rapidly accelerated fibrosarcoma; ERK, extracellular signal-regulated protein kinase.