RNA-sequencing across three matched tissues reveals shared and tissue-specific gene expression and pathway signatures of COPD

Abstract

Background: Multiple gene expression studies have been performed separately in peripheral blood, lung, and airway tissues to study COPD. We performed RNA-sequencing gene expression profiling of large-airway epithelium, alveolar macrophage and peripheral blood samples from the same subset of COPD cases and controls from the COPDGene study who underwent bronchoscopy at a single center. Using statistical and gene set enrichment approaches, we sought to improve the understanding of COPD by studying gene sets and pathways across these tissues, beyond the individual genomic determinants.

Methods: We performed differential expression analysis using RNA-seq data obtained from 63 samples from 21 COPD cases and controls (includes four non-smokers) via the R package DESeq2. We tested associations between gene expression and variables related to lung function, smoking history, and CT scan measures of emphysema and airway disease. We examined the correlation of differential gene expression across the tissues and phenotypes, hypothesizing that this would reveal preserved and private gene expression signatures. We performed gene set enrichment analyses using curated databases and findings from prior COPD studies to provide biological and disease relevance.

Results: The known smoking-related genes CYP1B1 and AHRR were among the top differential expression results for smoking status in the large-airway epithelium data. We observed a significant overlap of genes primarily across large-airway and macrophage results for smoking and airway disease phenotypes. We did not observe specific genes differentially expressed in all three tissues for any of the phenotypes. However, we did observe hemostasis and immune signaling pathways in the overlaps across all three tissues for emphysema, and amyloid and telomere-related pathways for smoking. In peripheral blood, the emphysema results were enriched for B cell related genes previously identified in lung tissue studies.

Conclusions: Our integrative analyses across COPD-relevant tissues and prior studies revealed shared and tissue-specific disease biology. These replicated and novel findings in the airway and peripheral blood have highlighted candidate genes and pathways for COPD pathogenesis.

Keywords: COPD; Emphysema; Genomics; RNA-seq; Transcriptomics.

Fig. 1

Overview of the study design illustrating the statistical and gene enrichment framework and the tissues (bronchial epithelium, peripheral blood and alveolar macrophages) and the phenotypes investigated. Findings are integrated with prior GWAS, prior lung tissue studies and the Connectivity Map

Fig. 2

Heatmap of differential gene expression correlation across all analyses. The row and column labels indicate the phenotype variable and the tissue. The results for each analysis were sorted by log2FoldChange and the Spearman correlation was calculated for each pair of results. The absolute value of these correlations is plotted in the heatmap. Clustering by euclidean distance is shown in the dendrograms. The region of correlation between the emphysema signature in blood and the smoking signature in the bronchial epithelium is outlined in the bottom black box (rows 12 & 13; columns 5–7). The region of correlation between the smoking and emphysema signatures in blood is outlined in the top black box (row 3; columns 5–7)

Fig. 3

Venn diagrams of the combined DESeq2 results intersected across tissue for the four phenotype categories (a. emphysema, b. lung function, c. smoking status, d. airway disease); an asterisk denotes significant overlap (p < 0.01)

Fig. 4

Venn diagrams of the overlap across tissue of the combined gene set enrichment results for the four phenotype categories (a. emphysema, b. lung function, c. smoking status, d. airway disease); an asterisk denotes significant overlap (p < 0.01) and lines join non-zero counts contributing to a significant overlap

Fig. 5

Heatmap summary of p-values from gene set enrichment tests using a set of significant (q-value < 0.1) airway disease results in the bronchial epithelium, and findings from previous GWAS and lung tissue studies. The top p-value corresponds to enrichment test in the up-regulated genes and the bottom (p-value) refers to enrichment in down-regulated genes. The row labels are color-coded by phenotype category (blue = lung function, red = smoking; green = emphysema, brown = airway)