Abstract
Repetitive sequences comprise two-thirds of the human genome, out of which 50% are transposable elements (TE). Cellular senescence is accompanied by induction of several cytokines, popularly referred as senescence associated secretory phenotype, 'SASP'; cause of which remains largely unknown. We reported that TE were upregulated in human senescent hematopoietic stem and progenitor cells (HSPCs). TE upregulation in senescent cells mediated interferon activation and SASP in senescence, facilitating natural killer cell mediated clearance. We recently showed that the expression of TE is suppressed, along with pathways of inflammation, in leukemic stem cells; a possible mechanism of immune escape. To elucidate the molecular mechanisms for TE upregulation in senescence, we performed whole genome bisulfite sequencing (WGBS) in senescent and active human HSPCs. we identified 61 differentially methylated regions (DMRs) in senescent vs. active HSPCs, of which 51 were hypo-methylated (hypoDMRs) and 10 hyper-methylated (hyperDMRs). By mapping the hypoDMRs to chromatin states using ChIP-seq data of primary human CD34+ cells, we found the majority of DMRs to overlap with transcriptional enhancers. CCAAT/enhancer binding proteins (CEBPA/B/G) were the dominant motif (8/10 top hits) in hypoDMRs. Interestingly, all of the hypoDMRs (51/51) but only 4/10 of the hyperDMRs overlapped with repeat elements. Based on the observation that ~50% of the genome is repetitive and that an overlap may equally likely affect hyper- and hypo-DMRs, Fisher's Exact test showed a significantly increased occurrence of repeat elements in the hypoDMRs (P<10-6). In summary, senescent human HSPCs showed co-occurrence of hypoDMRs, TE and CEBP binding motifs in the genome.
We then assessed whether CEBP plays a role in the transcriptional regulation of TE. Previous studies have shown methylation-specific increases in DNA binding affinity for CEBPβ. Hence, we speculated CEBPβ suppresses TE in a methylation-specific manner and that loss of methylation by senescence induction leads to increased TE expression. To test the hypothesis, we first checked whether CEBPβ-binding areas were enriched for repetitive elements using ChIP-seq data generated from HCT116 cells using an anti-CEBPβ antibody (ENCODE consortium). We observed a striking enrichment of repeat sequences in the CEBPβ ChIP-seq data as compared to Input, with enrichment higher than even known CEBPβ target promoters. These results strongly indicated that CEBPβ binds to TE-rich regions. qRT-PCR on spleen, T-helper cells and white adipose tissue from Cebp β-/- mice showed upregulated TE expression. This suggested that CEBPβ is a transcriptional repressor of TE and its depletion enhances their expression in vivo.
HCT116 cells, following senescence, have been shown to exhibit patterns of gene expression resembling senescence of normal cells. HCT116 cells following senescence induction showed significant upregulation of TE; similar observation was found in HCT116 after treatment with Azacytidine and in HCT DKO cells (deficient for Dnmt1 and Dnmt3a). This confirmed that the hypomethylation as well as senescence induction leads to TE upregulation.
We then wanted to test if CEBPβ binding to TE regions in the genome is disrupted during senescence induction and if this disruption is methylation-dependent. We performed chromatin immunoprecipitation using CEBPβ antibody in HCT116 cells in different conditions: senescent, demethylated and genetic loss of methylation (DKO). In all the three conditions, we found significant loss of CEBPβ binding to TE, confirming disruption of CEBPβ binding to TE regions in the DNA. CEBPβ plays a significant role in hematological malignancies and many other disease states. In this first study on senescent human HSPCs using WGBS and using in silico, in vitro and in vivo models, we show that CEBPβ binds to and regulates TE in steady state, demonstrating a novel role for CEBPβ. The role of dysregulation of CEBPβ in mediating disease pathogenesis via TE needs to be explored.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.