Meta-analysis of peripheral blood gene expression modules for COPD phenotypes

Abstract

Chronic obstructive pulmonary disease (COPD) occurs typically in current or former smokers, but only a minority of people with smoking history develops the disease. Besides environmental factors, genetics is an important risk factor for COPD. However, the relationship between genetics, environment and phenotypes is not well understood. Sample sizes for genome-wide expression studies based on lung tissue have been small due to the invasive nature of sample collection. Increasing evidence for the systemic nature of the disease makes blood a good alternative source to study the disease, but there have also been few large-scale blood genomic studies in COPD. Due to the complexity and heterogeneity of COPD, examining groups of interacting genes may have more relevance than identifying individual genes. Therefore, we used Weighted Gene Co-expression Network Analysis to find groups of genes (modules) that are highly connected. However, module definitions may vary between individual data sets. To alleviate this problem, we used a consensus module definition based on two cohorts, COPDGene and ECLIPSE. We studied the relationship between the consensus modules and COPD phenotypes airflow obstruction and emphysema. We also used these consensus module definitions on an independent cohort (TESRA) and performed a meta analysis involving all data sets. We found several modules that are associated with COPD phenotypes, are enriched in functional categories and are overrepresented for cell-type specific genes. Of the 14 consensus modules, three were strongly associated with airflow obstruction (meta p ≤ 0.0002), and two had some association with emphysema (meta p ≤ 0.06); some associations were stronger in the case-control cohorts, and others in the cases-only subcohorts. Gene Ontology terms that were overrepresented included “immune response” and “defense response.” The cell types whose type-specific genes were overrepresented in modules (p < 0.05) included natural killer cells, dendritic cells, and neutrophils. Together, this is the largest investigation of gene blood expression in COPD with 469 cases in COPDGene, ECLIPSE and TESRA combined, with 6267 genes common to all data sets. Additional, we have 42 and 83 controls in COPDGene and ECLIPSE, respectively.

Meta-analysis of Gene Expression Microarray Datasets in Chronic Obstructive Pulmonary Disease

“…Additionally, dysregulation of the lysosomal pathway has also been previously 375 described in COPD patients [59]. 376 We observed some down-regulated genes in the adherens junction pathway for modules for each cohort [17]. Out of our 304 genes, 97 of them overlapped with their 398 findings while 207 of our genes were unique.…”

“…Training with all samples achieved an accuracy of 87.0±3.0%, (Fig 9A). The working to identify strategies that can help to clearly understand COPD, its pathology, 311 and to find biomarkers in easily accessible body fluids to promote earlier detection of 312 COPD and improve accuracy of diagnosis [15][16][17]. Our research objective was to 313 identify age, sex and smoking status effects on gene expression between COPD and 314 controls in blood.…”

“…Gene ontology results from BINGO using our 207 unique 564 statistically significant disease genes filtered for biological effect. Our 304 565 biologically significant genes were compared to Reinhold et al,[17] We ran GO with 566 BINGO on 207 of our unique genes. The node size relates to number of genes, and the 567 yellow nodes are statistically significant with a BH-adjusted p-value <0.05 and FDR568 <0.05.…”

Lung Dendritic Cells Drive Natural Killer Cytotoxicity in Chronic Obstructive Pulmonary Disease via IL-15Rα

“…Suzuki and colleagues demonstrated enrichment of genes expressed by NKs and DCs (among other cell types) in helper T-cell type 1-associated expression profiles associated with emphysematous destruction (35). Using the ECLIPSE (Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-points) and COPDGene (Genetic Epidemiology of COPD) cohorts, Reinhold and colleagues associated gene modules expressed in peripheral blood with COPD phenotypes, and also found overrepresentation of genes specific for NKs, DCs (and neutrophils) (36). In addition to providing a potential mechanistic explanation for these associations, our data imply that pathological NK-DC interactions are present when COPD is mild (by measures of airflow and gas exchange), although not necessarily while subjects are young.…”

Analysis of the transcriptional regulatory mechanisms mediated by microRNAs in Metabolic Syndrome

“…These studies find multiple pathways implicated in the gene co-expression network profiles. Although it is possible to combine multiple co-expression analyses HORVATH, 2007) , this is still a novel approach in the field (REINHOLD et al, 2017) (SUN et al, 2017 .…”