Lysine demethylase 5C inhibits transcription of prefoldin subunit 5 to activate c-Myc signal transduction and colorectal cancer progression

Abstract

Background: Lysine demethylase 5C (KDM5C) has been implicated in the development of several human cancers. This study aims to investigate the role of KDM5C in the progression of colorectal cancer (CRC) and explore the associated molecular mechanism.

Methods: Bioinformatics tools were employed to predict the target genes of KDM5C in CRC. The expression levels of KDM5C and prefoldin subunit 5 (PFDN5) in CRC cells were determined by RT-qPCR and western blot assays. The interaction between KDM5C, H3K4me3, and PFDN5 was validated by chromatin immunoprecipitation. Expression and prognostic values of KDM5C and PFDN5 in CRC were analyzed in a cohort of 72 patients. The function of KDM5C/PFDN5 in c-Myc signal transduction was analyzed by luciferase assay. Silencing of KDM5C and PFDN5 was induced in CRC cell lines to analyze the cell malignant phenotype in vitro and tumorigenic activity in nude mice.

Results: KDM5C exhibited high expression, while PFDN5 displayed low expression in CRC cells and clinical CRC samples. High KDM5C levels correlated with poor survival and unfavorable clinical presentation, whereas elevated PFDN5 correlated with improved patient outcomes. KDM5C mediated demethylation of H3K4me3 on the PFDN5 promoter, suppressing its transcription and thereby enhancing the transcriptional activity of c-Myc. KDM5C knockdown in CRC cells suppressed cell proliferation, migration and invasion, epithelial-mesenchymal transition, and tumorigenic activity while increasing autophagy and apoptosis rates. However, the malignant behavior of cells was restored by the further silencing of PFDN5.

Conclusion: This study demonstrates that KDM5C inhibits PFDN5 transcription, thereby activating c-Myc signal transduction and promoting CRC progression.

Keywords: Apoptosis; Autophagy; Colorectal cancer; Lysine demethylase 5C; Prefoldin subunit 5.

Fig. 1

KDM5C regulates the transcription of PFDN5 through histone demethylation. A KDM5C expression in the CRC transcriptome database in the UALCAN system; B genes exhibiting negative correlation with KDM5C in CRC in the UALCAN system (ranked by correlation coefficient, with the top 25 genes displayed); C a negative correlation between KDM5C and PFDN5 in CRC in the GEPIA system; D-E, RT-qPCR D and WB analysis E to detect the mRNA and protein levels of KDM5C and PFDN5 in the CRC cells (HCT116 and SW480) and normal NCM460 cells (two-way ANOVA); F binding relationship of KDM5C and H3K4me3 with the PFDN5 promoter in the CRC cells and normal NCM460 cells validated by the ChIP-qPCR assay (two-way ANOVA). Three biological replicates were performed. *p < 0.05 vs. NCM460/IgG

Fig. 2

Expression levels and prognostic values of KDM5C and PFDN5 in CRC patients. A, B IHC assay to detect the protein levels of KDM5C A and PFDN5 B in CRC tissues and the adjacent normal tissues; C, D RT-qPCR to detect the mRNA levels of KDM5C C and PFDN5 D in CRC tissues and the adjacent normal tissues (n = 72) (paired t test); E, a negative correlation between KDM5C and PFDN5 mRNA expression in CRC tissues (n = 72) (Pearson's correlation analysis); F–G, correlations of KDM5C F and PFDN5 G expression with the RFS of patients (Log-rank (Mantel-Cox) test). *p < 0.05

Fig. 3

KDM5C/PFDN5 regulates viability of CRC cells and affects c-Myc signal transduction. A RT-qPCR to analyze the expression of KDM5C and PFDN5 in HCT116 and SW480 cells after shRNA transfection examined by (two-way ANOVA); B CCK-8 assay to analyze the proliferation of the cells examined by the (two-way ANOVA); C, D Transwell assays to analyze migration C and invasion abilities D of HCT116 and SW480 cells (two-way ANOVA); E dual luciferase assay to examine the transcriptional activity of c-Myc in the HCT116 and SW480 cells (two-way ANOVA). Three biological replicates were performed. *p < 0.05 vs. the sh-NC group; #p < 0.05 vs. the sh-KDM5C group

Fig. 4

KDM5C/PFDN5 regulates autophagic flux in CRC cells. A immunofluorescence staining to examine the LC-3B expression in the CRC cells (two-way ANOVA); B WB analysis to analyze the protein levels of ATG7, Beclin1, P62, and LC3-II/LC3-I (two-way ANOVA); C flow cytometry to analyze the apoptosis of CRC cells (two-way ANOVA). Three biological replicates were performed. *p < 0.05 vs. the sh-NC group; #p < 0.05 vs. the sh-KDM5C group

Fig. 5

KDM5C/PFDN5 affects tumorigenesis of CRC cells in vivo. A representative images of xenograft tumors formed by HCT116 and SW480 cells on day 28 and the volume change of tumors during this period (two-way ANOVA); B weight of the xenograft tumors on day 28 (two-way ANOVA); C, activation of caspase-3 (cleaved-caspase-3) in tumor tissues examined by IHC. In each group, n = 5. *p < 0.05 vs. the sh-NC group; #p < 0.05 vs. the sh-KDM5C group

Fig. 6

KDM5C/PFDN5 affects EMT in CRC. A protein levels of Vimentin and N-cadherin in HCT116 and SW480 cells determined by WB analysis (two-way ANOVA); B protein levels of Vimentin and N-cadherin in xenograft tumors determined by western blot analysis (two-way ANOVA). For cellular experiments, three biological replicates were performed. For animal studies, n = 5 in each group. *p < 0.05 vs. the sh-NC group; #p < 0.05 vs. the sh-KDM5C group

Fig. 7

Graphical abstract. KDM5C is highly expressed in CRC. It catalyzes demethylation of H3K4me3 on the PFDN5 promoter, therefore suppressing PFDN5 transcription and leading to transcriptional activation of protooncogene c-Myc. This results in cell autophagy repression and CRC progression