Fig 2: The association of DYRK1B and potential regulators with prognosis in TNBC patients. (A) The mRNA levels of the 10 genes most highly correlated with DYRK1B expression were examined in DYRK1B-silenced TNBC cells, including MDA-MB-231 and Hs578T cells. CCDC97 and ZNF581 mRNA levels are consistent with DYRK1B in TNBC cells. (B) The expression correlation of DYRK1B and CCDC97 and (D) ZNF581 was analyzed using Pearson χ2 test. (C) The relationship of co-expression of DYRK1B and CCDC97 or (E) ZNF581 with prognosis was inspected using Kaplan–Meier analysis. The associations of co-expression with overall survival and disease-free survival in TNBC patients are, respectively, shown in left panels and right panels. (F) TNBC cells were silenced with DYRK1B in the absence or presence of siRNA against CCDC97 or ZNF581 for 72 h. The cell viability of transfected cells was determined by Cell-Titer Glo. The results are indicated as the mean ± SD from three individual experiments * p < 0.05, ** p < 0.01 vs. scramble siRNA (siCtrl).

Fig 3: The effects of DYRK1B knockdown on spheroid formation of TNBC cells. (A) TNBC MDA-MB-231 cells were cultured in nanoparticle plates in the presence of scrambled siRNA or siRNA against DYRK1B for spheroid cell formation. The viability of tumorspheres was measured with 3D Cell-Titer Glo (B) The MDA-MB-231 cells expressing scrambled shRNA or shRNA against DYRK1B were cultured on a nanoparticle plate and tumorsphere viability was determined with 3D Cell-Titer Glo. (C) MDA-MB-231 and Hs578T cells were transfected with scrambled siRNA or siRNA against DYRK1B in an ultra-low attachment plate and the viable or dead cells were examined by Live (Green)/Dead (Red) cytotoxicity assay. The representative results were shown. (D) The live and dead cell population was measured according to fluorescence intensity and quantitated. The quantifiable results are labeled as the mean ± SD from three individual experiments. * p < 0.05 or ** p < 0.01 vs. control siRNA (siCtrl).

Fig 4: Dyrk1b downregulates Wbp2 protein abundance. (a) Volcano plot of differentially expressed proteins between livers of Ad-EGFP (n = 3) and Ad-Dyrk1b (n = 4) female mice. (b) Heatmap of the top 20 differentially expressed proteins between livers of Ad-EGFP (n = 3) and Ad-Dyrk1b (n = 4) female mice. (c) The top 30 enriched KEGG pathways of differentially expressed proteins between livers of Ad-EGFP (n = 3) and Ad-Dyrk1b (n = 4) female mice. (d–e) Western blot and densitometry quantification of Wbp2 in livers of Ad-EGFP (n = 4) and Ad-Dyrk1b (n = 8) female mice. Unpaired Student's t-test (two-sided) was performed for Fig. 4e. (f–g) Western blot and densitometry quantification of Wbp2 in Hepa1-6 cells infected with Ad-EGFP or Ad-Dyrk1b (n = 6). Unpaired Student's t-test (two-sided) was performed for Fig. 4g. (h–i) Western blot and densitometry quantification of Wbp2 in Hepa1-6 cells infected with adenovirus expressing wild-type or kinase-defective Dyrk1b (n = 4). One-way ANOVA with Tukey's post-hoc test was performed for Fig. 4i. (j–k) Representative Western blot and densitometry quantification of Wbp2 in Hepa1-6 cells infected with Ad-EGFP or Ad-Dyrk1b and treated with cycloheximide (20 μg/ml) for the indicated time points (n = 4). Unpaired Student's t-test (two-sided) was performed for Fig. 4k. (l–m) Western blot and densitometry quantification of Wbp2 in Hepa1-6 cells infected with Ad-EGFP or Ad-Dyrk1b in the absence or presence of MG132 (5 μM, 4 h) (n = 3). Two-way ANOVA with Tukey's post-hoc test was performed for Fig. 4m. (n) HA-tagged wild-type or kinase-defective Dyrk1b was expressed in Hepa1-6 cells together with Wbp2-Flag and Ub-Myc. At 45 h post-transfection, cells were treated with MG132 (20 μM) for 3 h and levels of Wbp2 ubiquitylation were evaluated by immunoprecipitation of Wbp2 using anti-Flag antibody followed by anti-Myc immunoblotting. *P < 0.05, **P < 0.01, ***P < 0.001, n.s., not significant.

Fig 5: Dyrk1b overexpression attenuates insulin signaling in a kinase activity-dependent manner. (a) Experimental designs following Ad-Dyrk1b or Ad-EGFP infection of Hepa1-6 cells. (b) Glucose uptake of Hepa1-6 cells infected with Ad-EGFP or Ad-Dyrk1b (MOI = 100) (n = 3). Unpaired Student's t-test (two-sided) was performed for Fig. 2b. (c–e) Western blot and densitometry quantification of Hepa1-6 cells infected with Ad-EGFP or Ad-Dyrk1b (MOI = 100) (n = 3). Two-way ANOVA with Sidak's post-hoc test was performed for Fig. 2d and e. (f–g) Western blot and densitometry quantification of Hepa1-6 cells infected with Ad-EGFP or Ad-Dyrk1b in the absence or presence of AZ191 (1 μM) (n = 3). Two-way ANOVA with Tukey's post-hoc test was performed for Fig. 2g. (h–j) Western blot and densitometry quantification of Hepa1-6 cells infected with adenovirus expressing wild-type Dyrk1b or kinase-defective Dyrk1b with Y271/273F mutation. One-way analysis of variance (ANOVA) with Tukey's post-hoc test was performed for Fig. 2i and j. (k) Experimental designs for female C57 mice injected with adenovirus expressing wild-type or kinase-defective Dyrk1b. Intraperitoneal glucose tolerance test (l) and AUC of GTT curves (m) of mice injected with adenovirus expressing wild-type or kinase-defective Dyrk1b (n = 5–7). Insulin tolerance test (n) and AUC of ITT curves (o) of mice injected with adenovirus expressing wild-type or kinase-defective Dyrk1b (n = 6–7). One-way ANOVA with Tukey's post-hoc test was performed for Fig. 2l–o. *P < 0.05 and **P < 0.01 for mice injected with Ad-EGFP versus Ad-Dyrk1b. †P < 0.05 and ††P < 0.01 for mice injected with Ad-Dyrk1b versus Ad-Dyrk1b (Y271/273F). (p–r) Western blot and densitometry quantification of livers from mice injected with adenovirus expressing wild-type or kinase-defective Dyrk1b (n = 4). One-way ANOVA with Tukey's post-hoc test was performed for Fig. 2q and r. *P < 0.05, **P < 0.01, ***P < 0.001, n.s., not significant.