Fig 2: VHL mutation in A498 and CAKI-2 RCC cells does not impact on the accumulation of free ATG12, and ATG12 forming complexes.a Immunoblot analysis of ATG12 protein in CAKI-2 characterized by VHL gene mutation but a wild-type SETD2 gene, and A498 characterized by both VHL and SETD2 (loss of function) gene mutation revealed that the accumulation in free ATG12, and the ATG12-containing complex, distinct from the ATG5–ATG12 complex, is independent of the VHL gene status. b Immunoblot analysis, and its quantification in c, of ATG7 expression does not show striking differences between those two RCC cell lines. Bars represent the mean of three independent experiments, error bars represent SEM. Ns nonsignificant.

Fig 3: Free ATG12 and ATG12-associated complexes expression is increased in SETD2-deficient cells.a Immunoblot analysis of ATG12 protein in SETD2-deficient CAKI-1 cells and SETD2-competent ACHN cells, reveal the accumulation of both free ATG12 and of an additional ATG12-containing complex, distinct from the ATG5–ATG12 complex, in the SETD2-deficient RCC cells. b Immunofluorescence analysis shows an increase expression for ATG12 in CAKI-1 cells as compared with ACHN cells.

Fig 4: Characterization of SETD2, H3K36me3, LC3-I, and LC3-II expression levels in ACHN and CAKI-1 RCC cells.Immunoblot analysis of the histone modifying enzyme SETD2 (a) and its histone target H3K36me3 (b) expression levels in ACHN and CAKI-1 renal clear carcinoma cells confirmed the SETD2 deficiency in the last named RCC cell line. c Immunoblot analysis of LC3 shows decreased expression level in SETD2-deficient CAKI-1 cells as compared with SETD2-wild-type ACHN cells. Treatment with Bafilomycin A1 (BafA1, 40 nM) for 4 h shows a significant decrease in LC3-II lipidation in CAKI-1 cells as compared with ACHN cells. d Quantification of LC3-II expression compared with the expression of the housekeeping gene, actin, with and without BafA1 treatment in both cell lines. e Quantification of LC3-II/LC3-I ratio to monitor autophagic flux in ACHN and CAKI-1 cells with and without BafA1 treatment demonstrates that cells that lacks SETD2 expression exhibit a decreased autophagic flux. f Analysis of MAP1LC3B mRNA expression by qPCR in SETD2-deficient CAKI-1 cells and SETD2-competent ACHN cells. Bars display the mean of four (d, e) or three (f) experiments, error bars represent SEM; ***p = 0.001; **p = 0.01; *p = 0.05.

Fig 5: Low H3K36me3, but high ATG12 protein expression levels are observed in ccRCC tumors, and SETD2 and ATG12 gene expression levels have prognostic values for ccRCC patients.a Human kidney tissue microarrays including 30 tissues classified as ccRCC tumors (highlighted in light gray) were proceed for immunohistochemical staining for H3K36me3 (b) and ATG12 protein (c) and revealed low H3K36me3 (d) but high ATG12 (e) protein expressions. HeLa were used as a positive control for H3K36me3 protein expression in b. f, g SETD2 and ATG12 high versus low gene expression levels (expressed in FPKM and with a cut off of 5.37 for SETD2 and 4.66 for ATG12) in tumor tissue derived from ccRCC patients at the time of diagnosis and their correlation to patient survival, i.e., follow up after diagnosis (expressed in years) was extracted from the pathology atlas of the human cancer transcriptome available at the Human Protein Atlas (www.proteinatlas.org). Survival analysis, prognosis of each group of patients examined using Kaplan–Meier survival estimators, and survival outcomes of the groups compared by log-rank tests revealed that SETD2 should be considered as a favorable prognostic gene and ATG12 as an unfavorable prognostic gene for ccRCC patients.