An immune response network associated with blood lipid levels

Abstract

While recent scans for genetic variation associated with human disease have been immensely successful in uncovering large numbers of loci, far fewer studies have focused on the underlying pathways of disease pathogenesis. Many loci which are associated with disease and complex phenotypes map to non-coding, regulatory regions of the genome, indicating that modulation of gene transcription plays a key role. Thus, this study generated genome-wide profiles of both genetic and transcriptional variation from the total blood extracts of over 500 randomly-selected, unrelated individuals. Using measurements of blood lipids, key players in the progression of atherosclerosis, three levels of biological information are integrated in order to investigate the interactions between circulating leukocytes and proximal lipid compounds. Pair-wise correlations between gene expression and lipid concentration indicate a prominent role for basophil granulocytes and mast cells, cell types central to powerful allergic and inflammatory responses. Network analysis of gene co-expression showed that the top associations function as part of a single, previously unknown gene module, the Lipid Leukocyte (LL) module. This module replicated in T cells from an independent cohort while also displaying potential tissue specificity. Further, genetic variation driving LL module expression included the single nucleotide polymorphism (SNP) most strongly associated with serum immunoglobulin E (IgE) levels, a key antibody in allergy. Structural Equation Modeling (SEM) indicated that LL module is at least partially reactive to blood lipid levels. Taken together, this study uncovers a gene network linking blood lipids and circulating cell types and offers insight into the hypothesis that the inflammatory response plays a prominent role in metabolism and the potential control of atherogenesis.

Figure 1. Network module associations with lipid traits.

For each lipid trait and each module expression profile, a Spearman rank correlation was performed. Each row corresponds to a module (arbitrarily lettered from A to W) and each column a particular lipid trait. Each cell contains the probability that a correlation exists by chance and is color-coded with red indicating a strong positive correlation and green a strong negative correlation. The module most strongly associated with lipid traits was module K. The genes composing this module, the Lipid Leukocyte (LL) module, show that many of the top signals from a standard linear regression were part of the same sub-network.

Figure 2. Topology of the network and the LL module.

The co-expression patterns of the network and the LL module were rendered using BiolayoutExpress3D . Each node is a gene (node size is not significant) and each edge is colored according to the absolute value of the Pearson correlation between two nodes, red being strong and blue being weak. The LL module has been colored yellow and, within the topology of the network (right panel), has been enlarged relative to other nodes. The topology of the network has been edge filtered (Pearson<0.65) in order to make strong correlations clearer.

Figure 3. Connectivity and trait association within the LL module.

In the LL module, there is a strong positive correlation between the intra-modular connectivity of a node (gene) and its association with APOB, HDL, and TG levels, showing that the approximately eight genes inter-connected are also the most associated with lipid traits. These genes constitute the core of the LL module.

Figure 4. Association of genetic variation with expression of FCER1A and SLC45A3.

For each cis SNP proximal to FCER1A and SLC45A3, a simple linear model is fitted and the regression P value (−log10 transformed) plotted along the vertical axis with genomic position of the SNP along the horizontal axis (both on chromosome 1). The two adjacent SNPs passing a permutation threshold of 0.05 are denoted by red dots while the SNP (rs2251746) within FCER1A found to drive expression of the LL module is circled in red. The position of the expression probe and the direction of transcription are denoted by a green arrow.

Figure 5. The directed network of core LL module, HDL, and triglycerides.

NEO was used to generate an edge-oriented network of triglycerides, HDL, and core LL module genes. Blue edges denote significant correlations between SNPs and nodes (orange circles), black edges denote significant correlations between nodes with a corresponding LEO score greater than 0.3 (predicted as a causal edge) while grey edges denote significant correlations between nodes with a LEO score less than 0.3 (causality is inconclusive). Dotted edges signify causal model fitted P values <0.05.