Authors

Taotao Sheng ^# 1 2, Shamaine Wei Ting Ho ^# 2 3, Wen Fong Ooi ⁴, Chang Xu ^2 3, Manjie Xing ^2 4, Nisha Padmanabhan ², Kie Kyon Huang ², Lijia Ma ⁵, Mohana Ray ⁵, Yu Amanda Guo ⁶, Ngak Leng Sim ⁶, Chukwuemeka George Anene-Nzelu ^7 8 9 10, Mei Mei Chang ⁶, Milad Razavi-Mohseni ¹¹, Michael A Beer ¹¹, Roger Sik Yin Foo ^7 8, Raghav Sundar ^2 12, Yiong Huak Chan ¹³, Angie Lay Keng Tan ², Xuewen Ong ², Anders Jacobsen Skanderup ⁶, Kevin P White ^14 15, Sudhakar Jha ^{16 17 18 19}, Patrick Tan ^{20 21 22 23 24 25}

Affiliations

¹Department of Biochemistry, National University of Singapore, Singapore, 117596, Singapore.

Abstract
Background: Enhancers are distal cis-regulatory elements required for cell-specific gene expression and cell fate determination. In cancer, enhancer variation has been proposed as a major cause of inter-patient heterogeneity-however, most predicted enhancer regions remain to be functionally tested.

Methods: We analyzed 132 epigenomic histone modification profiles of 18 primary gastric cancer (GC) samples, 18 normal gastric tissues, and 28 GC cell lines using Nano-ChIP-seq technology. We applied Capture-based Self-Transcribing Active Regulatory Region sequencing (CapSTARR-seq) to assess functional enhancer activity. An Activity-by-contact (ABC) model was employed to explore the effects of histone acetylation and CapSTARR-seq levels on enhancer-promoter interactions.

Results: We report a comprehensive catalog of 75,730 recurrent predicted enhancers, the majority of which are GC-associated in vivo (> 50,000) and associated with lower somatic mutation rates inferred by whole-genome sequencing. Applying CapSTARR-seq to the enhancer catalog, we observed significant correlations between CapSTARR-seq functional activity and H3K27ac/H3K4me1 levels. Super-enhancer regions exhibited increased CapSTARR-seq signals compared to regular enhancers, even when decoupled from native chromatin contexture. We show that combining histone modification and CapSTARR-seq functional enhancer data improves the prediction of enhancer-promoter interactions and pinpointing of germline single nucleotide polymorphisms (SNPs), somatic copy number alterations (SCNAs), and trans-acting TFs involved in GC expression. We identified cancer-relevant genes (ING1, ARL4C) whose expression between patients is influenced by enhancer differences in genomic copy number and germline SNPs, and HNF4? as a master trans-acting factor associated with GC enhancer heterogeneity.

Conclusions: Our results indicate that combining histone modification and functional assay data may provide a more accurate metric to assess enhancer activity than either platform individually, providing insights into the relative contribution of genetic (cis) and regulatory (trans) mechanisms to GC enhancer functional heterogeneity.

Keywords: CapSTARR-seq; Enhancer heterogeneity; Enhancer landscape; Enhancer-promoter interactions; Gastric cancer.

© 2021. The Author(s).

PMID: 34635154

PMCID: PMC8504099

DOI: 10.1186/s13073-021-00970-3