Methylation Patterns and Chromatin Accessibility in Neuroendocrine Lung Cancer

Abstract

Lung cancer is the worldwide leading cause of death from cancer. Epigenetic modifications such as methylation and changes in chromatin accessibility are major gene regulatory mechanisms involved in tumorigenesis and cellular lineage commitment. We aimed to characterize these processes in the context of neuroendocrine (NE) lung cancer. Illumina 450K DNA methylation data were collected for 1407 lung cancers including 27 NE tumors. NE differentially methylated regions (NE-DMRs) were identified and correlated with gene expression data for 151 lung cancers and 31 human tissue entities from the Genotype-Tissue Expression (GTEx) consortium. Assay for transposase-accessible chromatin sequencing (ATAC-seq) and RNA sequencing (RNA-seq) were performed on eight lung cancer cell lines, including three NE cell lines, to identify neuroendocrine specific gene regulatory elements. We identified DMRs with methylation patterns associated with differential gene expression and an NE tumor phenotype. DMR-associated genes could further be split into six functional modules, including one highly specific gene module for NE lung cancer showing high expression in both normal and malignant brain tissue. The regulatory potential of NE-DMRs was further validated in vitro using paired ATAC- and RNA-seq and revealed both proximal and distal regulatory elements of canonical NE-marker genes such as CHGA, NCAM1, INSM1, as well as a number of novel candidate markers of NE lung cancer. Using multilevel genomic analyses of both tumor bulk tissue and lung cancer cell lines, we identified a large catalogue of gene regulatory elements related to the NE phenotype of lung cancer.

Proteogenomics of non-small cell lung cancer reveals molecular subtypes associated with specific therapeutic targets and immune-evasion mechanisms

“…For a comprehensive phenotype-level analysis of NSCLC we used isobaric labelling and HiRIEF-LC-MS 8 , 9 with data-dependent acquisition (DDA) reaching an analytical depth of 13,975 identified proteins (gene-centric search, FDR<1%, Figure 1b , Supplementary Table 2 ). In addition to MS-data, mutation analysis for cancer-associated genes was performed by panel sequencing (n=370 genes), furthermore genome-wide methylation and mRNA-level data were available for most samples 10 – 12 ( Supplementary Table 2 ).…”

“…The early-stage cohort (also referred to as the “discovery cohort”) comprised resected lung cancer tumor samples from a total of 192 patients with operable lung cancer that were surgically treated at the Skåne University Hospital in Lund, Sweden. The samples were collected as described in previous studies 10 – 12 . The late-stage cohort comprised biopsy material from inoperable lung cancer (84 samples).…”

“…Pre-processed Illumina gene expression data for 118 cases in the early-stage NSCLC cohort was obtained from Karlsson et al 10 and DNA methylation data was available from previous studies for 113/141 lung cancer tumors in this cohort (GSE60645 and GSE149521) 11 , 12 . DNA methylation data processing and filtering were performed as previously described 11 , 12 , resulting in a final dataset interrogating 459,790 genomic positions.…”

“…Pre-processed Illumina gene expression data for 118 cases in the early-stage NSCLC cohort was obtained from Karlsson et al 10 and DNA methylation data was available from previous studies for 113/141 lung cancer tumors in this cohort (GSE60645 and GSE149521) 11 , 12 . DNA methylation data processing and filtering were performed as previously described 11 , 12 , resulting in a final dataset interrogating 459,790 genomic positions. Methylation probes were annotated using the IlluminaHumanMethylation450kprobe (v2.0.6) R package and promoter regions were defined as TSS +/- 500bp and extracted using the promoters() function in the TxDb.Hsapiens.UCSC.hg19.knownGene (v3.2.2) R package.…”

Large Cell Neuro-Endocrine Carcinoma of the Lung: Current Treatment Options and Potential Future Opportunities

“…In particular, Li et al have demonstrated that the progressive loss of histone H4 acetylation at lysine16 (H4KA16) and trimethylation at lysine 20 (H4KM20) from low to high grade lung neuroendocrine tumors reflects the degree of differentiation and proliferative activity (48). Also methylation patterns have inspired interest because of their correlation with gene expression in lung neuroendocrine cancers (49).…”