Fig 2: NUMB p71/p72 isoforms promoted ubiquitylation of N1ICD and its degradation. (A) RKO cells were transiently transfected with si-p71/72-#1 for 48 h, then incubated in the presence of cycloheximide (CHX, 50 μg/mL) for the indicated time course, followed by Western blotting analysis. Quantification of Notch1 proteins is shown at the bottom. (B) HeLa cells were transiently transfected with si-p65/p66-#1 for 48 h, then incubated in the presence of cycloheximide (CHX, 50 μg/mL) for the indicated time course, followed by Western blotting analysis. Quantification of Notch1 proteins is shown on the bottom. (C) 293T cells were co-transfected with Flag-N1ICD, Ubiquitin (Ub) and NUMB isoform expression plasmids. Whole-cell lysates were extracted and immunoprecipitated with anti-Flag antibodies followed by Western blotting analysis. Overall, 10% of whole cell lysates were used as input. * p < 0.05, ** p < 0.01.

Fig 3: Numb and Numblike are required for sarcomere assembly.(A) Electron microscopy of the myocardial sarcomere of CTL and MLC-2v-cre DKO mice on E14.5 (original magnification, ×12,0000). (B) Electrocardiogram of CTL and MLC-2v-cre DKO mice on P60.5. (C) Electrophysiology of gastrocnemius of CTL and CKMM-cre DKO mice on P60.5. (D) Statistic data of heart rate of CTL and MLC-2v-cre DKO mouse on P60.5 (n = 5). (E) Statistic data of heart rate of CTL and α-MHC-cre DKO mice on P90.5 (n = 4). (F) Stimulus voltage (threshold and plateau) on gastrocnemius of CTL and CKMM-cre DKO mice on P60.5 (n = 6). (G) Maximum, minimum, and average tension of gastrocnemius of CTL and CKMM-cre DKO mice on P60.5 (n = 6). Scale bars: 1 μm. Two-tailed Student’s t test with error bars showing mean ± SD.

Fig 4: Gene and pathway correlation analyses of CD44 and NUMB isoforms in patient-derived colon cancers.(A) Gene correlation analysis showing the correlation of gene expression with CD44s and CD44v6 isoform expression in the The Cancer Genome Atlas (TCGA) patient cohort. Differentially expressed genes from CD44s- (red) and CD44v6-OE (blue) RNAseq data are highlighted. (B) Pathway correlation analysis showing the correlation of pathway activity CD44 and NUMB isoform expression in the TCGA patient cohort.

Fig 5: NUMB interacts with different proteins including CK2. (A) A scheme depicting the experimental approach used for mass spectrometry-based analysis of NUMB–eGFP interactor partners. (B) A volcano plot showing NUMB-interacting proteins detected by mass spectrometry. (C) Coupled immunoprecipitation (GFP antibody) and western blot analysis (GFP, cortactin, CK2α, and CK2β antibodies) of stable NUMB 1–4-overexpressing cells (n = 3; upper panel). The lower scheme illustrates the structure of the four NUMB isoforms (PTB, phospho-tyrosine-binding domain; PRR, proline-rich domain; yellow box, amino acid insert in the PTB domain) that might be responsible for interaction with specific targets. See text for details. (D) Coupled immunoprecipitation (NUMB antibody) and western blot analysis (NUMB, CK2α, and CK2β antibodies) of scramble and Numb KD cells. Non-immune immunoglobulin (IgG) was used as a negative control (n = 3). (E) Coupled immunoprecipitation (CK2α antibody) and western blot analysis (NUMB and CK2β antibodies) of scramble and Numb KD cells.