Fig. 6
circCGNL1 facilitates HDAC4-induced RUNX2 deacetylation and destabilization. A Co-IP assays were performed to detect the interaction between HDAC4 and RUNX2 in 293 T cells. An anti-lgG antibody served as the control. B The GEPIA database was used to predict the co-expression relationship between HDAC4 and RUNX2 in PC and normal tissues. A non-log scale was used for calculations, a log-scale axis was used for visualization, and Spearman’s correlation coefficient was determined. C, D Acetylation and protein expression levels of RUNX2 in HDAC4-overexpressing 293 T cells were estimated using IP-WB assays. E Co-IP assays were performed to analyze the influence of circCGNL1 overexpression on the interplay between RUNX2 and HDAC4. F The GEPIA database was used to assess RUNX2-expression levels in PC tissues and normal tissues. RUNX2 expression level was defined by log2(TPM + 1), number of pancreatic tumor tissues (T) = 179, number of normal pancreatic tissues (N) = 171. G The mRNA and protein expression levels of RUNX2 in PANC-1 and BxPC-3 cells were detected using qRT-PCR and WB analyses. H WB assays were performed to detect the protein expression levels of RUNX2 in PANC-1 and BxPC-3 cells transfected with pcDNA3.1-circCGNL1. I Co-IP-WB assays were performed to study the influence of circCGNL1 overexpression on RUNX2 acetylation. J PANC-1 cells were transfected with a circCGNL1-overexpression plasmid or the empty vector and treated with Cycloheximide (CHX) at 100 μg/ml. WB analysis was performed to detect RUNX2 stability at 3, 6, and 9 h (h) after CHX treatment (HY-12320, MCE, USA). Relative RUNX2 protein level was normalized with β-actin. R2: Goodness of fit, *p < 0.05, **p < 0.01