Fig 2: Autophagy enhances stemness in MB. a qPCR analysis of autophagy-related genes mRNA in different MB cell lines. GADPH and B2M were used as internal control. DAOY cells were arbitrarily defined as 1.00. Data are expressed as the mean value ± SEM (n = 6). Data were analysed by one-way analysis of variance (ANOVA) followed by Tukey post hoc test. (*p < 0.05; ***p < 0.001). b Both D283-Med and DAOY cells were treated with CQ (20 µM) for 1 h. Levels of LC3, P62, and ACTIN were analysed by WB. Densitometric analysis of LC3 II levels over ACTIN is also shown. Data are expressed as the mean ± SEM (n = 3) and analysed by one-way analysis of variance (ANOVA) followed by Tukey post hoc test (**p < 0.01). c Both ATG7 and BECLIN 1 expression was downregulated in D283-Med cells using specific RNAi oligonucleotides (siATG7 or siBECLIN 1, respectively) or unrelated oligos as negative control (siCTRL). Levels of CD133, BECLIN 1 and ACTIN were analysed by WB. Densitometric analysis of CD133 levels over ACTIN is also shown. Data are expressed as the mean ± SEM (n = 3) and analysed by unpaired Student's t-test (**p < 0.01). The downregulation of ATG7 was analysed by qPCR. d D283-Med cells were treated as in (c). The percentage of plasma membrane (PM) CD133 positive cells was assessed by flow-cytometry. Representative histograms of CD133 mean of fluorescence intensity (MFI) are shown. Data are expressed as the mean value ± SEM (n = 3). Data were analysed by unpaired Student's t-test (**p < 0.01). e D283-Med cells were treated with two different concentration of CQ (5 and 10 µM, respectively) for 48 h. Levels of CD133, NESTIN, LC3 and ACTIN were analysed by WB. f D283-Med cells were treated as in (e) for different time points. NESTIN mRNA levels were analysed by qPCR. GADPH and B2M were used as internal control. Data are expressed as the mean value ± SEM (n = 3). Data were analysed by one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test. (*p < 0.05, **p < 0.01). g GSAA enrichment plot showing that loss of ATG7 in D283-Med cells (3 independent experiments) results in downregulation of the pluripotency of stem cell pathway. Data are generated from RNA-seq analyses. h DAOY cells were cultured as medullospheres for 3 and 7 days. Levels of p-ULK1 758, p-ATG14, LC3, P62 and ACTIN were analysed by WB. i AMBRA1-silenced D283-Med cells (shAMBRA1) were transfected with empty, AMBRA1WT and AMBRA1AA mutant constructs, respectively. Levels of AMBRA1, CD133 and ACTIN were analysed by WB. Densitometric analysis of CD133 levels over ACTIN is also shown (right panel). Data are expressed as the mean ± SEM (n = 3) and analysed by one-way analysis of variance (ANOVA) followed by Tukey post hoc test. (*p < 0.05)

Fig 3: FUBP1 is ubiquitinated by Smurf2 in CRC regardless of KRAS genotype. (A) Representative images of FUBP1 IHC staining of 40 KRAS wild-type CRC specimens versus 14 KRAS mutation CRC specimens collected from Sun Yat-sen University Cancer Center (top, 100× magnification; bottom, 200× magnification). Scale bar, 50 µm. (B, C) Statistical analysis of the 54 CRC specimens showing low or high FUBP1 expression relative to KRAS wild-type or KRAS mutation. ***P < 0.001. (D) Overall survival analysis of 52 CRC patients which divided into three groups: KRAS wild-type and low FUBP1 expression (n = 30); KRAS wild-type and high FUBP1 expression (n = 10); KRAS mutation and high FUBP1 expression (n = 12). (E) Representative images of Smurf2 IHC staining in three groups according to FUBP1 expression (high or low) and KRAS mutation or wild-type. Scale bar, 100 µm. (F) Statistical analysis of Smurf2 staining in three groups; *P < 0.05; **P < 0.01. (G) Western blotting analysis of FUBP1, CD133, and ALDH1 in the indicated cells. ß-Actin served as a loading control. (H, I) Cycloheximide chase assay of FUBP1. (H) LoVo cells transiently transfected with vector or Smurf2 plasmids were treated with cycloheximide (CHX, 10 µg·mL-1) for indicated time points and then collected for western blotting analysis. (I) LoVo cells transfected with Smurf2 plasmid were treated with DMSO or MG132 (10 µmol·L-1) in the presence of CHX (10 µg·mL-1) for indicated time points and then collected for western blotting analysis. (J) LoVo cells transiently transfected with Vector or Smurf2 plasmids were treated with DMSO or MG132 (10 µmol·L-1) for 12 h, and then immunoprecipitated with an FUBP1 antibody. All bars represented the mean ± SD of three independent experiments. P values were determined by one-way ANOVA.

Fig 4: Effect of GMI on A549/A400 cell xenograft tumour growth in nude mice.a A549/A400 cells (approximately 5 × 106) were injected subcutaneously into each left flank of nude mice to initiate tumour growth. At 7 days after implanting the cells, the PBS group started to receive sterilised PBS, whereas the mice in GMI group received GMI protein (160 µg/mouse, N = 5, respectively). PBS and GMI proteins were administered to mice by gavage once every day. At 27 days after cell transplantation, tumour sizes were measured every 3 or 4 days and the tumour volume was calculated. The symbol ‘*’ indicates P < 0.05 for the tumour volume of the PBS group when compared with the tumour volume of the GMI group at the same time-point. b Tumour images of the PBS and GMI groups. c Tumour weights were measured after the mice were sacrificed at day 69. The symbol ‘*’ indicates P < 0.05. d IHC staining was performed to investigate CD133 expression in tumours of the PBS and GMI groups.

Fig 5: IGF2BP2-mediated upregulation of DANCR promotes cell proliferation and stemness-like properties.a IGF2BP2 siRNA down regulates DANCR expression in BXPC-3 cells, as determined by qRT-PCR, whereas ectopic expression of IGF2BP2 increases DANCR level (b). c Like IGF2BP2 siRNA, IGF2BP2 KO also suppresses DANCR expression in BXPC-3 cells. d DANCR promotes cell proliferation in BXPC-3 and SW1990 cells, as determined by CCK8 assays. e DANCR also promotes colony formation. f Ectopic expression of DANCR increases sphere formation in number and size. g DANCR-overexpression promotes expression of stemness-related genes (OCT4, NANOG, BMI 1, CD24, and CD133) in pancreatic cancer cells, as determined by qRT-PCR and at the protein level (h). i Percentage of CD24+ and CD133+ population in DANCR overexpression and control (NC) cells, as determined by flow cytometry. The statistical results of flow cytometry were showed on the right. Values in a, b, c, d, e, g, i are mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.