Genetic network identifies novel pathways contributing to atherosclerosis susceptibility in the innominate artery

Abstract

Background: Atherosclerosis, the underlying cause of cardiovascular disease, results from both genetic and environmental factors.

Methods: In the current study we take a systems-based approach using weighted gene co-expression analysis to identify a candidate pathway of genes related to atherosclerosis. Bioinformatic analyses are performed to identify candidate genes and interactions and several novel genes are characterized using in-vitro studies.

Results: We identify 1 coexpression module associated with innominate artery atherosclerosis that is also enriched for inflammatory and macrophage gene signatures. Using a series of bioinformatics analysis, we further prioritize the genes in this pathway and identify Cd44 as a critical mediator of the atherosclerosis. We validate our predictions generated by the network analysis using Cd44 knockout mice.

Conclusion: These results indicate that alterations in Cd44 expression mediate inflammation through a complex transcriptional network involving a number of previously uncharacterized genes.

Figure 1

Overview of analysis: This flowchart presents a brief overview of the analysis and subsequent experiments performed. 1. Construction of the weighted gene Co-expression network Analysis and relationship to innominate artery atherosclerosis. 2. Relationships between modules and atherosclerosis were confirmed using independent and publically available gene expression datasets. 3. Ontology analysis was performed using DAVID and identified macrophages as a potential cell type for validation of the network. In-vitro experiments are performed to characterize module genes. 4. Causality analysis is performed and experiments using macrophages from gene targeted mice are used to validate predictions.

Figure 2

Genetic network associated with innominate artery atherosclerosis. Weighted Gene Co-expression analysis of liver RNA identifies 10 modules of highly co-expressed genes. Networks are visualized in Cytoscape. Note size of the module denotes the overall connectivity of the hubs and strength of the topological overall is denoted by the length of the edge.

Figure 3

Co-expression Network analysis identifies the Brown module as related to Atherosclerosis. (A). Mean MS score for each of the 11 network modules. (B) Mean module gene expression in atherosclerotic aorta tissue relative to non-atherosclerotic tissue at 6 weeks. (C) Mean module gene expression in atherosclerotic aorta tissue relative to non-atherosclerotic tissue at 36 weeks. In panels (B–C) expression is presented as the mean log2 expression for each gene in a module in aorta from C57BL/6 J Apoe^-/- mice minus log2 expression from wild type C57BL/6 J mice. (D) A sub-network of co-expressed genes in the Brown Module including Cd44 and the brown module hub genes. Gene connectivity determines size of each node. Distance between nodes is determined by the topological overlap. * denotes significant differences (p < 0.05).

Figure 4

Strain specific expression of hub genes. Peritoneal macrophages were isolated from C57BL/6 J and C3H/HeJ mice and treated in triplicate with media or media with 10 ng/ml LPS for 4 hours. Expression of Apbb1ip, Cd44, Evl, Fermt3, Gpsm3, Ncf2, Nckap1l, Plcg2, Tnf, Trpv2, and Was were normalized to Rpl4 and expressed relative to non-stimulated cells (Panels A-K respectively). * indicates significant differences P < 0.05. Values represent mean ± sem.

Figure 5

Cd44 modulates brown module hub gene expression. Peritoneal macrophages were isolated from C57BL/6 J and Cd44^-/- mice and treated in triplicate with media or media with 10 ng/ml LPS for 4 hours. Expression of Apbb1ip, Cd44, Evl, Fermt3, Gpsm3, Ncf2, Nckap1l, Plcg2, Tnf, Trpv2, and Was were normalized to Rpl4 and expressed relative to non-stimulated cells from C57BL/6 J mice (Panels A-K respectively). Genotype of the cells and treatment condition are indicated below the x-axis * indicates significant differences P < 0.05. Values represent mean ± sem.

Figure 6

Confirmation of novel Cd44 target genes. Peritoneal macrophages were isolated from C57BL/6 J and Cd44^-/- mice and treated in triplicate with media or media with 10 ug/ml LPS for 4 hours. Expression of Calhm2, Dapp1, Ehd4, Fmnl2, Kcng2, Neurl2, Pltp, Tex14, and Vsig4 were normalized to Rpl4 and expressed relative to non-stimulated cells (Panels A-H respectively). * indicates significant differences P < 0.05. Values represent mean ± sem.

Figure 7

Brown Module Hub Genes are differentially expressed in human atherosclerosis. Publically available microarray data, GSE43292, was analyzed for differential expression of brown module hub genes, Apbb1ip, Cd44, Evl, Fermt3, Gpsm3, Ncf2, Nckap1l, Plcg2, Tnf, and Trpv2. Expression levels determined by robust multi-array normalization (RMA) for atherosclerotic sample (grey) and matched intact samples from the patients (white). * indicates significant differences P < 0.05. Values represent mean ± sem.