Role of innate immunity-triggered pathways in the pathogenesis of Sickle Cell Disease: a meta-analysis of gene expression studies

Abstract

Despite the detailed characterization of the inflammatory and endothelial changes observed in Sickle Cell Disease (SCD), the hierarchical relationship between elements involved in the pathogenesis of this complex disease is yet to be described. Meta-analyses of gene expression studies from public repositories represent a novel strategy, capable to identify key mediators in complex diseases. We performed several meta-analyses of gene expression studies involving SCD, including studies with patient samples, as well as in-vitro models of the disease. Meta-analyses were performed with the Inmex bioinformatics tool, based on the RankProd package, using raw gene expression data. Functional gene set analysis was performed using more than 60 gene-set libraries. Our results demonstrate that the well-characterized association between innate immunity, hemostasis, angiogenesis and heme metabolism with SCD is also consistently observed at the transcriptomic level, across independent studies. The enrichment of genes and pathways associated with innate immunity and damage repair-associated pathways supports the model of erythroid danger-associated molecular patterns (DAMPs) as key mediators of the pathogenesis of SCD. Our study also generated a novel database of candidate genes, pathways and transcription factors not previously associated with the pathogenesis of SCD that warrant further investigation in models and patients of SCD.

Figure 1. Gene expression pattern from the meta-analysis.

The upper panel shows the overlap between DE genes identified in the meta-analysis (Meta-DE) and in each individual data analysis (individual-DE). Gain genes are those identified only in the meta-analysis. Loss genes are those identified in individual studies, but not in the meta-analysis. In the lower panel, a heatmap built using the top 30 differentially-expressed genes (15 up – and 15 down-regulated) comparing the gene expression pattern of studies that enrolled patients with sickle cell disease (GSE35007 and GSE53441) is shown. Class 1 and 2 refer to control and patient samples respectively, from each individual dataset.

Figure 2. Enriched gene ontology pathways identified in the meta-analysis.

The top enriched biological processes predicted from the list of up-regulated genes generated in the meta-analysis of samples from patients with sickle cell disease were grouped with the software ClueGO as a functional cluster (using a kappa score = 0.3). Each node represents a biological process. Their associated genes are represented as dots. Node and dot colors represent the functional group to which they belong. Mixed coloring nodes and dots belong to multiple groups. One ungrouped term is shown in grey. Edges represent term-term interaction or term-genes interaction. The title of the most significant term per group is shown in the network as a group title (colored text). The size of nodes reflects the enrichment significance of the terms.