Repurposing of KLF5 activates a cell cycle signature during the progression from a precursor state to oesophageal adenocarcinoma

Connor Rogerson, Samuel Ogden, Edward Britton, Yeng Ang, Andrew D. Sharrocks

TL;DR

This study reveals that the transcription factor KLF5 plays a critical role in the progression of Barrett's esophagus (BO) to esophageal adenocarcinoma (OAC). By repurposing its binding activity, KLF5 drives a cell cycle gene signature in OAC cells, which is linked to poorer patient survival. This discovery offers new insights into OAC development and potential therapeutic targets.

What Problem Does This Paper Address?

Esophageal adenocarcinoma (OAC) is a deadly cancer with a low survival rate, often diagnosed at late stages. While Barrett's esophagus (BO) is a known precancerous precursor, the molecular mechanisms driving its progression to OAC remain poorly understood. This lack of knowledge hinders the development of effective early detection and treatment strategies. The paper aims to elucidate these molecular events by investigating changes in gene expression and chromatin accessibility during this transition.

Methods

The researchers employed a multi-omics approach, integrating several key techniques:

  • RNA-sequencing (RNA-seq): Used to compare gene expression profiles between BO and OAC patient samples, identifying differentially expressed genes.
  • Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq): This technique maps regions of open chromatin, indicating active regulatory elements. It was used to identify changes in the chromatin landscape between BO and OAC and to pinpoint potential transcription factor binding sites.
  • De novo motif discovery: Analyzed ATAC-seq data to identify enriched DNA motifs, suggesting the binding of specific transcription factors in accessible chromatin regions.
  • Chromatin Immunoprecipitation sequencing (ChIP-seq): Used to map the binding sites of KLF5 across the genome in different cell states (BO vs. OAC).
  • siRNA-mediated knockdowns: Employed to experimentally reduce the expression of KLF5 and other key genes (like ERBB2) to assess their functional impact on cell growth and gene expression.
  • Gene Set Enrichment Analysis (GSEA) and Gene Ontology (GO) analysis: Used to identify biological pathways and functions associated with differentially expressed genes.
  • Crystal violet assay and Propidium iodide staining: Standard assays to measure cell proliferation and cell cycle progression, respectively.

A novel aspect of their approach was the integration of ATAC-seq and RNA-seq data to link changes in chromatin accessibility to gene expression and to identify key regulatory transcription factors.

Datasets

The study utilized several datasets:

  • Human biopsy samples: Gene expression (RNA-seq) and chromatin accessibility (ATAC-seq) data from human Barrett's esophagus (BO) and esophageal adenocarcinoma (OAC) patient samples.
  • Publicly available datasets: Data from previous studies on OAC and BO, including ATAC-seq data from The Cancer Genome Atlas (TCGA).
  • Cell line models: OE19 (OAC cell line) and CP-A (BO-derived cell line) were used for experimental manipulations and mechanistic studies.

Key Insights and Ideas

  • Cell cycle signature in OAC: A striking finding was the strong upregulation of cell cycle-related genes in OAC compared to BO, suggesting a fundamental shift in cellular behavior during cancer progression.
  • KLF5 as a key regulator: The transcription factor KLF5 emerged as a critical player. Despite its expression levels remaining relatively constant between BO and OAC, its binding pattern across the genome dramatically changes.
  • Repurposing of KLF5: KLF5 appears to be "repurposed" in OAC cells. It redistributes its binding to specific regulatory regions, thereby activating a distinct set of cell cycle genes that drive proliferation. This is a key mechanistic insight.
  • Combinatorial regulation: KLF5 doesn't act alone. Its binding at target regions often coincides with motifs for other transcription factors (like AP-1, GATA, FOXA), suggesting a complex regulatory network.
  • ERBB2 convergence: The study found that KLF5 and the ERBB2 signaling pathway, often amplified in OAC, converge on the same cell cycle regulatory regions, highlighting a synergistic mechanism driving OAC progression.
  • Prognostic significance: High expression of KLF5 target genes (not KLF5 itself) was significantly correlated with poorer patient survival, suggesting these genes could serve as potential biomarkers.

Key Results

  • 905 differentially expressed genes: Identified between BO and OAC, with a strong enrichment for cell cycle-related pathways.
  • 1495 differentially accessible chromatin regions: Found between BO and OAC, indicating significant changes in regulatory element activity.
  • KLF5 motifs enriched: Found in regions with increased chromatin accessibility in OAC, implicating KLF5 as a potential regulator.
  • KLF5 directly regulates cell cycle genes: siRNA-mediated knockdown of KLF5 in OE19 cells led to significant reductions in the expression of cell cycle genes and impaired cell proliferation.
  • Altered KLF5 binding profile: ChIP-seq analysis revealed distinct KLF5 binding patterns in OAC-derived cells compared to BO-derived cells.
  • ERBB2 depletion reduces KLF5 target gene expression: Knockdown of ERBB2 in OE19 cells led to the downregulation of many KLF5-activated cell cycle genes.
  • KLF5 target gene expression predicts survival: Patients with high expression of KLF5 target genes had significantly worse survival outcomes.

Limitations and Future Work

  • Cell line limitations: While OE19 and CP-A cells were used to model BO and OAC, they may not perfectly recapitulate the complexity of patient biopsies. The authors acknowledge that further work is needed to fully understand the role of KLF5 in BO.
  • Mechanism of KLF5 redistribution: While the study shows KLF5 redistribution, the precise upstream signals or events that trigger this rewiring in OAC remain to be fully elucidated.
  • Therapeutic implications: The authors suggest that targeting KLF5 or its downstream targets could be a therapeutic strategy, but this requires further validation in preclinical models and clinical trials.
  • Role of other transcription factors: While KLF5 and AP-1 are highlighted, the precise roles and interactions of other transcription factors identified in the study (e.g., GATA, FOXA) in OAC progression warrant further investigation.
  • BO context: The study hints at KLF5 operating through the Hippo signaling pathway in BO, but this requires more direct experimental validation.

Summary updated

Abstract

Oesophageal adenocarcinoma (OAC) is one of the most common causes of cancer deaths. Barrett's oesophagus (BO) is the only known precancerous precursor to OAC, but our understanding about the molecular events leading to OAC development is limited. Here, we have integrated gene expression and chromatin accessibility profiles of human biopsies and identified a strong cell cycle gene expression signature in OAC compared to BO. Through analysing associated chromatin accessibility changes, we have implicated the transcription factor KLF5 in the transition from BO to OAC. Importantly, we show that KLF5 expression is unchanged during this transition, but instead, KLF5 is redistributed across chromatin to directly regulate cell cycle genes specifically in OAC cells. This new KLF5 target gene programme has potential prognostic significance as high levels correlate with poorer patient survival. Thus, the repurposing of KLF5 for novel regulatory activity in OAC provides new insights into the mechanisms behind disease progression.
BibTeX
@inproceedings{rogerson2020repurposing,
  author = {Connor Rogerson and Samuel Ogden and Edward Britton and Yeng Ang and Andrew D. Sharrocks},
  title = {{Repurposing of KLF5 activates a cell cycle signature during the progression from a precursor state to oesophageal adenocarcinoma}},
  booktitle = {eLife},
  year = {2020},
  doi = {10.7554/elife.57189},
}
Figure