Fig 2: The effect of UBE2K can be revered by IGF2BP3 in PDAC. (A) Reverse transcription-quantitative PCR and (B) western blotting results in the groups of NC + siNC, UBE2K + siNC and UBE2K + siIGF2BP3. The (C) CCK-8 and (D) colony formation assays results in the groups of NC + siNC, UBE2K + siNC and UBE2K + siIGF2BP3. (E) Reverse transcription-quantitative PCR and (F) western blotting results in the groups of shNC + vector, shUBE2K + vector and shUBE2K + IGF2BP3. The (G) CCK-8 and (H) colony formation assays results in the groups of shNC + vector, shUBE2K + vector and shUBE2K + IGF2BP3. *P<0.05, **P<0.01, ***P<0.001. UBE2K, ubiquitin-conjugating enzyme E2K; PDAC, pancreatic ductal adenocarcinoma; NC, negative control; si, short interfering; sh, short hairpin.

Fig 3: Knockdown of UBE2K suppresses proliferation and stemness of PDAC in vitro. (A) Reverse transcription-quantitative PCR and (B) western blotting results confirmed that knockdown stably transduced strains of UBE2K was established in BxPC-3 and PANC-1. (C) CCK-8 and (D) colony formation assays indicated knockdown of UBE2K reduced proliferation ability. (E) Western blotting results indicated knockdown of UBE2K decreased stemness genes expression. (F) Sphere formation results indicated the knockdown of UBE2K decreased the size of spheres (scale bar, 200 µm). *P<0.05, **P<0.01, ***P<0.001. UBE2K, ubiquitin-conjugating enzyme E2K; PDAC, pancreatic ductal adenocarcinoma; NC, negative control; sh, short hairpin.

Fig 4: UBE2K promotes PDAC growth in vivo. (A) The tumors of UBE2K overexpression group were larger compared with NC in nude mice. The (B) volume and (C) weight of UBE2K overexpression group were larger compared with NC in the tumors of nude mice. The RNA (D) and (E) protein level of UBE2K overexpression group were higher compare to NC group. (F) The tumors of shNC group were larger compared with shUBE2K#1 and shUBE2K#2 groups in nude mice. The (G) volume and (H) weight of shNC groups were larger compared with shUBE2K#1 and shUBE2K#2 groups in the tumors of nude mice. The (I) RNA and (J) protein level of shNC were higher compared with shUBE2K#1 and shUBE2K#2 groups. **P<0.01, ***P<0.001. UBE2K, ubiquitin-conjugating enzyme E2K; PDAC, pancreatic ductal adenocarcinoma; NC, negative control; sh, short hairpin.

Fig 5: Double immunofluorescence staining of embryonic wild-type and yotari mouse kidneys with the FGFR1 (a–d), HIP2 (e), and RIP5 markers (a–e). Arrows show the expression pattern of FGFR1, HIP2, and RIP5 in metanephric mesenchyme (mm), renal vesicles (rv), glomeruli (g), convoluted tubules (Ct), ampullae (A), and collecting ducts (Cd) indicated on 4′,6-diamidino-2-phenylindole (DAPI) image. Immunoexpression of FGFR1, RIP5, DAPI staining and merged FGFR1, RIP5, and DAPI at embryonic days E13.5 and E15.5 in wild type (a,c) and yotari (b,d). Immunoexpression of HIP2, RIP5, DAPI staining and merged HIP2, RIP5, and DAPI at E15.5 in yotari (e). Wild-type and yotari mice at E13.5 of kidney development mostly correspond regarding localization and intensity of HIP2; thus, a representative image was selected from E15.5. The most prominent protein expression area is shown in inserts, corresponding to the dashed boxes. An asterisk denotes the zone where co-expression was detected. Images were taken on magnification ×40. The scale bar is 50 μm, which refers to all images. The distribution of the percentages of FGFR1 (f), and HIP2 (g) positive cells in the metanephric mesenchyme (mm), renal vesicles (rv) or glomeruli (g), convoluted tubules (Ct), and ampulla (A) or collecting ducts (Cd) of wild type and yotari kidneys at embryonic days E13.5 and E15.5. Data are presented as the mean ± SD (vertical line) and analyzed by a two-way ANOVA test followed by Tukey’s multiple comparison test. Significant differences were indicated by * p < 0.05, ** p < 0.001, **** p < 0.00001. At each time point, ten substructures were assessed.