Environment and vascular bed origin influence differences in endothelial transcriptional profiles of coronary and iliac arteries

Abstract

Atherosclerotic plaques tend to form in the major arteries at certain predictable locations. As these arteries vary in atherosusceptibility, interarterial differences in endothelial cell biology are of considerable interest. To explore the origin of differences observed between typical atheroprone and atheroresistant arteries, we used DNA microarrays to compare gene expression profiles of harvested porcine coronary (CECs) and iliac artery endothelial cells (IECs) grown in static culture out to passage 4. Fewer differences were observed between the transcriptional profiles of CECs and IECs in culture compared with in vivo, suggesting that most differences observed in vivo were due to distinct environmental cues in the two arteries. One-class significance of microarrays revealed that most in vivo interarterial differences disappeared in culture, as fold differences after passaging were not significant for 85% of genes identified as differentially expressed in vivo at 5% false discovery rate. However, the three homeobox genes, HOXA9, HOXA10, and HOXD3, remained underexpressed in coronary endothelium for all passages by at least nine-, eight-, and twofold, respectively. Continued differential expression, despite removal from the in vivo environment, suggests that primarily heritable or epigenetic mechanism(s) influences transcription of these three genes. Quantitative real-time polymerase chain reaction confirmed expression ratios for seven genes associated with atherogenesis and over- or underexpressed by threefold in CECs relative to IECs. The present study provides evidence that both local environment and vascular bed origin modulate gene expression in arterial endothelium. The transcriptional differences observed here may provide new insights into pathways responsible for coronary artery susceptibility.

Fig. 1.

Number of differentially expressed genes between coronary endothelial cells (CECs) and iliac artery endothelial cells (IECs) at 5–25% false discovery rate (FDRs) using one-class paired significance analysis of microarrays (SAM). For passage 0 (p0) and p2–p4, n = 4; for p1, n = 2.

Fig. 2.

Genes identified as differentially expressed in p0 at 5% FDR using one-class paired SAM that also exhibit significant fold differences between the two arteries in cell culture. Values are means ± SE; for p0 and p2–p4, n = 4; for p1, n = 2. CECs different from IECs: *P < 0.05, **P < 0.005. Different compared with in vivo: ⁺P < 0.05, ⁺⁺P < 0.005. See text for definition of genes.

Fig. 3.

One-class paired microarray results for genes that are putatively atheroprotective (A), associated with atherosclerosis with uncertain function (B), and atherogenic (C). Values are means ± SE; for p0 and p2–p4, n = 4; for p1, n = 2. CECs different from IECs: *P < 0.05, **P < 0.005. Different compared with in vivo: ⁺P < 0.05, ⁺⁺P < 0.005. Two-way ANOVA of microarray expression values for p2–p4, P ≤ 0.05: ^a artery and ^p passage number effects. See text for definition of genes.

Fig. 4.

PCR results for selected genes. For p0–p4, expression fold differences between CECs and IECs (A), and average ΔC_T for CECs and IECs (B) are shown. C: expression ratios for HOXA9 and HOXA10. Lines represent linear least squares fits with correlation coefficients R² as indicated. Values are means ± SE; for p1, n = 3; for p0 and p2–p4, n = 4, except that HOXA9 p0 value for CECs is based on 1 detection out of 12 wells. CEC HOXA10 not detected after 50 PCR cycles for p0. CECs different than IECs: *P < 0.02, **P < 0.002. Different compared with in vivo: ⁺P < 0.05, ⁺⁺P < 0.005. See text for definition of genes.