Increased expression of fatty acid binding protein 4 and leptin in resident macrophages characterises atherosclerotic plaque rupture

Abstract

Objective: Resident macrophages play an important role in atheromatous plaque rupture. The macrophage gene expression signature associated with plaque rupture is incompletely defined due to the complex cellular heterogeneity in the plaque. We aimed to characterise differential gene expression in resident plaque macrophages from ruptured and stable human atheromatous lesions.

Methods and results: We performed genome-wide expression analyses of isolated macrophage-rich regions of stable and ruptured human atherosclerotic plaques. Plaques present in carotid endarterectomy specimens were designated as stable or ruptured using clinical, radiological and histopathological criteria. Macrophage-rich regions were excised from 5 ruptured and 6 stable plaques by laser micro-dissection. Transcriptional profiling was performed using Affymetrix microarrays. The profiles were characteristic of activated macrophages. At a false discovery rate of 10%, 914 genes were differentially expressed between stable and ruptured plaques. The findings were confirmed in fourteen further stable and ruptured samples for a subset of eleven genes with the highest expression differences (p < 0.05). Pathway analysis revealed that components of the PPAR/Adipocytokine signaling pathway were the most significantly upregulated in ruptured compared to stable plaques (p = 5.4 × 10(-7)). Two key components of the pathway, fatty-acid binding-protein 4 (FABP4) and leptin, showed nine-fold (p = 0.0086) and five-fold (p = 0.0012) greater expression respectively in macrophages from ruptured plaques.

Conclusions: We found differences in gene expression signatures between macrophages isolated from stable and ruptured human atheromatous plaques. Our findings indicate the involvement of FABP4 and leptin in the progression of atherosclerosis and plaque rupture, and suggest that down-regulation of PPAR/adipocytokine signaling within plaques may have therapeutic potential.

Fig. 1

Condition tree dendrogram demonstrating the correlation of the genomic expression profiles of the 11 laser micro-dissected samples to other samples of various types (n = 299) using Spearman correlation of the whole array genome-wide gene expression data with bootstrap confidence levels.

Fig. 2

Unsupervised explorative clustering on the entire array datasets of the samples using condition tree (Panel A) and principal components analysis (Panel B). The ruptured group cluster closely together; the stable group, although clearly separate from the ruptured group, has a greater spread (as might be expected in a histologically more heterogeneous group). In panel C, statistically significant differentially expressed genes (1187 probesets representing 914 different genes) ranked by fold difference and represented in a heatmap is shown. Red represents over-expression, green represents under-expression. The top 10 differentially expressed genes from both ends are listed. Fold change values above 1 represents higher relative expression in the ruptured group and values below 1 represents higher relative expression in the stable group. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Panel A shows the real-time qPCR results of the 12 most significantly differentially expressed genes tested in 12 ruptured and 13 stable specimens. The fold difference with the 95% confidence intervals and the ANOVA p-values are shown. Panel B shows the relative clustering and separation of the 25 samples in gene expression space using principal components analysis on all the genes tested in RTqPCR.

Fig. 4

Immunostained paraffin sections of a representative stable atheromatous plaque with a thick fibrous cap. Corresponding negative controls stained with haematoxylin, smooth muscle cells labelled by smooth muscle actin, macrophages labelled by CD68. Leptin and FABP4 immunostaining correlates with the CD68 immunostaining for macrophages.