Transcriptome analysis of genes regulated by cholesterol loading in two strains of mouse macrophages associates lysosome pathway and ER stress response with atherosclerosis susceptibility.

Cholesterol loaded macrophages in the arterial intima are the earliest histological evidence of atherosclerosis. Studies of mouse models of atherosclerosis have shown that the strain background can have a significant effect on lesion development. We have previously shown that DBA/2 ApoE(-/-) mice have aortic root lesions 10-fold larger than AKR ApoE(-/-) mice. The current study analyzes the response to cholesterol loading of macrophages from these two strains. Macrophages from the atherosclerosis susceptible DBA/2 strain had significantly higher levels of total and esterified cholesterol compared to atherosclerosis resistant AKR macrophages, while free cholesterol levels were higher in AKR cells. Gene expression profiles were obtained and data were analyzed for strain, cholesterol loading, and strain-cholesterol loading interaction effects by a fitted linear model. Pathway and transcriptional motif enrichment were identified by gene set enrichment analysis. In addition to observed strain differences in basal gene expression, we identified many transcripts whose expression was significantly altered in response to cholesterol loading, including P2ry13 and P2ry14, Trib3, Hyal1, Vegfa, Ccr5, Ly6a, and Ifit3. Eight pathways were significantly enriched in transcripts regulated by cholesterol loading, among which the lysosome and cytokine-cytokine receptor interaction pathways had the highest number of significantly regulated transcripts. Of the differentially regulated transcripts with a strain-cholesterol loading interaction effect, we identified three genes known to participate in the endoplasmic reticulum (ER) stress response, Ddit3, Trib3 and Atf4. These three transcripts were highly up-regulated by cholesterol in AKR and either down-regulated or unchanged in loaded DBA/2 macrophages, thus associating a robust ER stress response with atherosclerosis resistance. We identified significant transcripts with strain, loading, or strain-loading interaction effect that reside within previously described quantitative trait loci as atherosclerosis modifier candidate genes. In conclusion, we characterized several strain and cholesterol induced differences that may lead to new insights into cellular cholesterol metabolism and atherosclerosis.