In silico study of RNA-seq and H3 Trimethylation ChIP-seq analysis in combination with laser-microdissection on human lung cancer cells

概要

Introduction and Purpose
Gene expression abnormality is the main characteristic of cancer cells, which is regulated at multiple levels. At genomic level, many studies have been taking great efforts to understand such changes in cancer gene expressions by identifying the genetic mutations. However, it was unclear if such cancer-associated changes occurred in cancer cells or other cell types in bulk tissue samples. The use of bulk tissue, which is not precisely characterized in terms of histology, has long been the basis for molecular analysis. It is undoubtedly that this approach is insufficient for a detailed analysis of molecular alterations, which might be restricted to a specific cell population, such as tumor, normal, stromal, or epithelial cells. I hypothesized that my study could provide more precise gene expression and histone modification profiles by enhanced cell selection using laser-microdissection microscopy (LMD) for clinical non–small cell lung cancer (NSCLC) tissue samples.

Materials and Methods
Tissue specimens stored in the Pathology Department and Tsukuba Human Tissue Biobank were used after surgery at the University of Tsukuba Hospital (Ibaraki, Japan). After extracting RNA and sequencing, I went through a workflow for integrative RNA- sequencing (RNA-seq) and chromatin immunoprecipitation sequencing (ChIP-seq) data analysis of such clinical tissue samples in combination with the isolation of specific cell populations by LMD.

Results
As a pilot study, I first applied the LMD-RNA-seq integrative approach in 10 cases of NSCLC, including 6 cases of lung adenocarcinoma (LUAD) and 4 cases of lung squamous cell carcinoma (LUSC). For each case, normal lung tissue, stromal and tumor cell parts were isolated by LMD. Transcriptomic profiles of NSCLC were successfully captured using LMD-isolated samples as heatmap clusters present the separation of gene expression patterns between these two NSCLC subtypes and RNA-seq analysis shows subtype-specific DEGs and enriched pathways in NSCLC. Indeed, RNA-seq results from LMD-isolated tumor samples were concordant with the Cancer Genome Atlas (TCGA) reference dataset while excluding stroma parts suggesting that my RNA-seq results were not only well representative for NSCLC but also more precise by enhanced cell selection using LMD. Next, I examined epigenetic alteration in NSCLC by using ChIP-seq on histone H3 Lysine 4 trimethylation (H3K4me3) marks. By comparing the H3K4me3 profile of lung tumor and normal tissues, I identified hundreds of somatically altered promoters. Interestingly, I observed the positive correlation between proximal H3K4me3 regions and DEGs, while altered H3K4me3-marked promoters distant from transcription start sites (TSSs) are found to be associated with enhancer activity of cancer regulatory genes. Moreover, integration with ENCODE data reveals that proximal tumor-gained promoters are associated with EZH2 and SUZ12 occupancies, which are the core components of PRC2 (polycomb repressive complex 2). Finally, I screened for genetic variation in the somatically altered H3K4me3 regions and identified possible allele-bias germline variants and potential somatic mutations that may influence the gene regulation.

Discussion
The identification of such NSCLC subtype-specific pathways and genes emphasizes the principle that the precision of my RNA-seq data comes from the precision of the cell- selection efforts using LMD. As a pilot study with a limited number of samples, comparison with a bigger platform like TCGA gives us a perspective on the correctness of my dataset. RNA-seq results show high agreement level among top DEGs with TCGA NSCLC samples, suggesting that my NSCLC gene expression profile was well-captured by LMD with the small sample-size dataset. In ChIP-seq analysis, the results suggests that most of the genes with their transcription start sites (TSS) overlapped with tumor- enriched H3K4me3 regions are highly expressed and associated with cancer-related pathways, such as PI3K-Akt signaling pathway, extracellular matrix (ECM) and adhesion in cancer. Meanwhile, distal tumor-enriched H3K4me3 regions were found to be associated with previously reported enhancer activities. Furthermore, I observed the association between proximal tumor-gained promoters with EZH2 and SUZ12 (Polycomb Repressive Complex 2 (PRC2) components) occupancies. This observation agrees with previous studies in gastric cancer, suggesting the potential common roles of these PRC2 components in multiple cancer types.

Conclusion
In conclusion, I analyzed the integrative RNA-seq and H3K4me3 ChIP-seq data from the clinical NSCLC samples isolated by LMD and showed the potential of this workflow in the perspective of human genome research. The sample size still limited the power of my study. In the future, the study would be improved with a larger cohort and incorporated with more types of omics data as well as different histone marks.