HELLS controls mitochondrial dynamics and genome stability in liver cancer by collusion with MIEF1

Abstract

Dysregulated chromatin remodelers have emerged as critical disease targets. However, owing to the pleiotropic functions of chromatin remodelers, the underlying mechanisms of their effects on cancer have been difficult to elucidate. Here, we investigated the helicase lymphoid-specific (HELLS) oncogenic mechanism by identifying a new direct transcriptional target. Using loss or gain experiments, we identified Mitochondrial elongation factor 1 (MIEF1) as a critical target of the HELLS molecular network in liver cancer. Liver cancer patients with a poor prognosis exhibited upregulated expression of MIEF1, and MIEF1 knockdown led to the loss of tumor capabilities, indicating MIEF1 as an oncogene in liver cancer. Suppressing the HELLS-MIEF1 axis caused mitochondrial hyperfusion, energy deprivation, and further resulting senescence. HELLS knockdown globally increased histone 3 lysine 9 trimethylation (H3K9me3), especially in genomic hotspots with upregulation of SUV39H1 and further augmented DNA methylation. This stabilized genome and hyperfused mitochondria led to reduced levels of reactive oxygen species (ROS) and DNA damage. Finally, tumor cells became famished and calm. We further validated the functions of the HELLS-MIEF1 axis by MIEF1 overexpression and mitochondrial fusion drug. Our study has important implications for medical science by highlighting the crosstalk between epigenetics and metabolism through nuclear chromatin remodeler HELLS and mitochondrial protein MIEF1.

Fig. 1. HELLS targets the mitochondrial protein MIEF1.

A HELLS-dependent transcriptome analysis. Red, yellow, and green dots indicate significant changes in differentially expressed genes (DEGs) with a 2-fold change and 0.05 p-value cutoffs. Red dots represent DEGs (n = 143 and 375). Yellow dots represent genes positively correlated with HELLS expression in UALCAN (n = 32, 39). The green dot highlights MIEF1 (included in both red and yellow categories). B HELLS Hallmark analysis of HELLS positive genes (downregulated by HELLS knockdown and upregulated by HELLS overexpression, #143) and HELLS negative genes (upregulated by HELLS knockdown and downregulated by HELLS overexpression, # 375). C The expression changes of mitochondria-related gene sets upon the loss and gain of HELLS. D, E Protein levels of HELLS, MIEF1, and Beta actin upon HELLS depletion/overexpression in HepG2, Huh7, and SNU398 cells were measured using western blots. F Schematic of the MIEF1 promoter PCR region (Top), the enrichments of HELLS and H3K4me3 at the MIEF1 promoter in Huh7 cells were measured using ChIP-qPCR (mean ± S.E.M., **p < 0.01, n = 3). G The graphic model summarizes the key findings illustrated in Fig. 1.

Fig. 2. MIEF1 has an oncogenic function in liver cancer.

A MIEF1 expression in non-tumor (NT) and tumor (T) tissues presented as mean ± S.D. (***p < 0.001). Horizontal lines are the median. B Kaplan–Meier survival analysis on MIEF1 levels. The p-value was calculated using the log-rank test. C MIEF1 expression in liver cancer tissues. MIEF1 protein levels were analyzed in total liver cancer tissue (left panel) and further stratified based on HELLS expression levels (right panel) using bar charts. Representative Immunohistochemistry (IHC) staining images from each HELLS expression intensity level were included: negative (−), positive (+), and strong (++). The scale bar represents 100 μm. D MIEF1 expression levels upon MIEF1 depletion. The relative mRNA levels of MIEF1 in HepG2 and SNU398 cells were presented as mean ± S.E.M. (**p < 0.01, n = 3). The protein levels of MIEF1 and Beta actin in HepG2 and SNU398 cells were measured upon MIEF1 depletion using western blots. E Growth rates of upon MIEF1 loss. Cell numbers were presented as mean ± S.E.M. in HepG2 and SNU398 cells (***p < 0.001, n = 3). F Tumor growth in xenograft models. Tumor volumes were measured using nude mice injected with Huh7 shControl and shMIEF1 cells (n = 4 per group) and presented as mean ± S.E.M. (**p < 0.01). G The schematic model summarizes the key findings illustrated in Fig. 2.

Fig. 3. HELLS regulates mitochondrial dynamics and cellular energy homeostasis via MIEF1.

A Representative confocal images of mitochondria upon HELLS and MIEF1 depletion. HepG2 cells were transfected with mt-ro2GFP. The scale bar represents 20 μm. Quantification of mitochondrial area per cell (μM²) and mitochondrial number per total area were measured upon HELLS and MIEF1 depletion. (mean ± S.E.M., ns non-significant, **P < 0.01, n = 3). B Representative Transmission electron microscopy (TEM) images of mitochondria upon HELLS and MIEF1 depletion in HepG2. Arrows indicate mitochondria. Scale bars = 10 μm. C Representative confocal images of mitochondrial fusion upon HELLS and MIEF1 depletion. HepG2 cells were transfected with mtGreen and mtRed and fused by polyethylene glycol (PEG). The scale bar represents 10 μm. D Representative confocal images of mitochondria upon HELLS overexpression. HepG2 cells were transfected with mt-ro2GFP. The scale bar represents 20 μm. Quantification of mitochondrial area per cell (μM²) and mitochondrial number per total area were measured upon HELLS overexpression. (mean ± S.E.M., *P < 0.05, n = 3). E Representative TEM images of mitochondria upon HELLS overexpression in HepG2. Mitochondria are indicated by arrows. Scale bars = 10 μm. F Seahorse experiment data upon HELLS and MIEF1 depletion. Glycolysis, proton leak, maximal respiration, and ATP production were measured in Huh7 cells (mean ± S.E.M., *P < 0.05, **P < 0.01, n = 3). G Quantification of TMRM fluorescence intensity upon HELLS and MIEF1 depletion in HepG2 (mean ± S.E.M., *P < 0.05, ***P < 0.001, n = 3). H Protein levels of HELLS, pho-AMPK, AMPK, and Beta actin upon HELLS depletion in HepG2, Huh7, and SNU398 cells were measured using western blots. I Representative images of senescence-associated β-galactosidase (SA-β-gal) staining upon HELLS depletion in HepG2. J The schematic model summarizes the key findings illustrated in Fig. 3.

Fig. 4. Decreased HELLS induces global heterochromatin with the increased global level of H3K9me3.

A Protein levels of H3K9me3, H3K4me3, and Histone H3 upon HELLS depletion in HepG2, Huh7, SNU398, and Hep3B cells were measured using western blots. B ChIP-seq profiles showed the alterations in distribution and intensity of H3K9me3 peaks upon HELLS depletion in HepG2 cells. C Genomic distribution of differential H3K9me3 peaks upon HELLS depletion with log2 fold change (log2FC) ≥ 1 in HepG2 cells. D H3K9me3 enrichment signal across genomic regions upon HELLS depletion in HepG2 cells. E Gene expression levels from microarray analysis of genes associated with hypermethylated and hypomethylated H3K9me3 peaks at promoters and enhancers upon HELLS depletion in HepG2 cells. F Distribution of H3K9me3 peaks across various genomic hotspots upon HELLS depletion in HepG2 cells. The hotspots include microsatellites (MST), small indels (SID), single nucleotide polymorphisms (SNP), and long copy number variations (LCNV). G Protein levels of HELLS, Beta actin, H4K20me3, Histone H4, and Histone H1 upon HELLS depletion in HepG2, Huh7, SNU398, and Hep3B cells were measured using western blots. H The schematic model summarizes the key findings illustrated in Fig. 4.

Fig. 5. Loss of HELLS induces an increase in SUV39H1 expression.

A The relative expression levels of the H3K9me3-related gene set upon HELLS depletion and overexpression in HepG2. B Protein levels of HELLS, SUV39H1, Beta actin, H3K9me3, and Histone H3 upon HELLS depletion in HepG2, Huh7, SNU398, and Hep3B cells were measured using western blots. C Protein levels of HELLS, SUV39H1, Beta actin, H3K9me3, and Histone H3 upon HELLS overexpression in HepG2, Huh7, and SNU398 cells were measured using western blots. D HELLS binding at the SUV39H1 promoter. Schematic diagram of the SUV39H1 proximal promoter region (top). The enrichment of HELLS at the SUV39H1 promoter in Huh7 cells was presented using ChIP-qPCR (mean ± S.E.M., ***P < 0.001, n = 3). E The schematic model summarizes the key findings illustrated in Fig. 5.

Fig. 6. The absence of HELLS promotes DNA methylation accumulation and stabilizes chromatin structures.

A Representative ICC staining images of 5hmC and quantification of 5hmC levels upon HELLS depletion in HepG2, Huh7, SNU398, and Hep3B cells. The scale bar represents 100 μm. (mean ± S.E.M., **P < 0.01, n = 3). B Dot blot images of 5hmC and 5mC upon HELLS depletion in HepG2 cells. Quantification values are indicated above each spot, representing relative intensity. C HELLS-dependent hypermethylated and hypomethylated DNA sites with beta difference (β diff) ≥│0.2│ in HepG2 cells, measured by Illumina Infinium Methylation EPIC Analysis. D, E Distribution of hypermethylated and hypomethylated DNA sites within CpG islands/ DNase I hypersensitivity regions upon HELLS depletion in HepG2 cells. F Gene expression levels from microarray analysis of genes associated with hypermethylated and hypomethylated DNA methylation sites upon HELLS depletion in HepG2 cells. G The schematic model summarizes the key findings illustrated in Fig. 6.

Fig. 7. The HELLS-MIEF1 axis promotes ROS production.

A ROS levels upon HELLS and MIEF1 depletion in HepG2 (mean ± S.E.M., **p < 0.01, n = 3). B Representative comet images stained with a green DNA staining solution and a bar graph showing the quantification of tail moment upon HELLS and MIEF1 depletion in HepG2 (mean ± S.E.M., ***p < 0.001, n = 3). C Protein levels of γH2AX and H2AX upon HELLS in HepG2, Huh7, and SNU398 upon HELLS deletion were measured using western blots. D ROS levels upon HELLS overexpression in HepG2 (mean ± S.E.M., **p < 0.01, n = 3). E Representative comet images stained with a green DNA staining solution and a bar graph showing the quantification of tail moment upon HELLS overexpression in HepG2 cells (mean ± S.E.M., ***p < 0.001, n = 3). F The schematic model summarizes the key findings illustrated in Fig. 7.

Fig. 8. MIEF1 overexpression reverses the effects induced by HELLS-MIEF1 axis suppression.

A The relative mRNA levels of HELLS and MIEF1 in HepG2 cells treated with Control, shHELLS, oeMIEF1, and shHELLS+oeMIEF1 were presented as mean ± S.E.M. (*p < 0.05, **p < 0.01, ***p < 0.001, n = 3). B Representative confocal images of mitochondria. HepG2 cells treated with Control, shHELLS, shMIEF1, oeMIEF1, shHELLS+oeMIEF1, and shMIEF1+oeMIEF1 were transfected with mt-ro2GFP. The scale bar represents 20 μm. Quantification of mitochondrial length was measured (mean ± S.E.M., ns non-significant, ***P < 0.001, n = 3). C Protein levels of pho-AMPK and AMPK in HepG2 and SNU398 cells treated with Control, shHELLS, oeMIEF1, and shHELLS+oeMIEF1 were measured using western blots. D Protein levels of HELLS, MIEF1, SUV39H1, Beta actin, H3K9me3, and Histone H3 in HepG2 and SNU398 cells treated with Control, shHELLS, oeMIEF1, and shHELLS+oeMIEF1 were measured using western blots. E ROS levels in HepG2 cells treated with Control, shHELLS, oeMIEF1, and shHELLS+oeMIEF1. (mean ± S.E.M., *p < 0.05, **p < 0.01, ***p < 0.001, n = 3). F Representative comet images stained with a green DNA staining solution and a bar graph showing the quantification of a tail moment in HepG2 cells treated with Control, shHELLS, oeMIEF1, and shHELLS+oeMIEF1 (mean ± S.E.M., ns non-significant, ***p < 0.001, n = 3). G Protein levels of γH2AX and H2AX in HepG2 and SNU398 cells treated with Control, shHELLS, oeMIEF1, and shHELLS+oeMIEF1 were measured using western blots. H The graphical summary illustrates the proposed model based on the overall findings of this study.