The main findings obtained from our study are (1) The DC-chemokine receptor CCR7 and its ligands CCL21/CCL19 are significantly downregulated in atherosclerotic plaques; (2) Circulating CCL21 levels are upregulated in serum from atherosclerotic patients; (3) oxLDL impairs CCR7 expression in DCs and CCL21 expression in microvascular ECs.
Our data support the concept that modulation of chemokine receptors (mediated, e.g., by oxLDL) in the plaque may trigger retention of DCs, thereby impeding the vascular innate and adaptive immunity [29
]. Accordingly, Angeli et al. found that oxLDL and other lipid mediators are jointly responsible for trapping of DCs in the vascular wall [29
]. In our in vivo
study, oxLDL serum concentrations showed no correlation with CCL21/CCL19 or CCR7 expression in plaques. However, CCR7 and CCL21/CCL19 were downregulated in atherosclerotic plaques compared to nonatherosclerotic aortic tissue. This discrepancy may be explained by the fact that oxLDL is unstable in serum but accumulates in the subintimal space over time. Hereby, it reaches much 70 fold higher subintimal concentrations as compared to serum levels [31
]. This fact especially appeases to monocytes rich plaques [31
We found CCR7 expression to be downregulated after 24 hours of stimulation with oxLDL. Alongside, CCL21 was downregulated on HMECs-1 after incubation with oxLDL, in a concentration which predominates in atherosclerotic plaques. Trogan et al. previously demonstrated that phases of hyperlipidemia severely reduce CCR7 mRNA and the corresponding protein levels in an ApoE−/−
-mice model. CCR7 increased in the presence of normolipidemia. In their work, CCR7 expression was found to be essential for the migration of DCs from atherosclerotic plaque. In atherosclerotic plaques of ApoE−/−
-mice, CCR7 receptor expression was reduced. When transferred to wild-type mice, this process was accompanied by an upregulation of CCR7 itself [32
Under a high-dose statin therapy, Damas et al. found a significant decrease in CCL19 and CCL21 levels in serum [2
]. Our data support this observation, showing reduced CCL21 serum protein expression in the control group, characterized by low oxLDL levels. The difference in controls was evident despite a low-dose statin therapy.
Our results suggest that DC maturation is triggered by increased oxLDL deposits in the subintimal space. At the same time DCs, and even T-cells do not seem to express enough CCR7 receptors in order to migrate. As a consequence, oxLDL, DCs and T-cells appear to accumulate, thereby enhancing plaque inflammation processes and possibly plaque rupture.
These results complement the former findings that oxLDL increased the adhesion and promotes the maturation and differentiation of DCs from monocytes [9
]. Furthermore, oxLDL supports the building of foam-cell accumulations and also directly induces chemotaxis of immune cells like T-cells via upregulation of endothelial adhesion molecules [9
]. Keeping in mind that DCs are responsible for priming naive T-cells to oxLDL-specific T-cells [9
], the interaction of oxLDL and DCs must be taken into account as a key factor for the induction and progression of atherosclerosis.
oxLDL not only seems to have an influence on the expression of CCR7, but also to induce downregulation of CCL21 on ECs in vitro
as well as decreased expression of CCL21 and CCL19 in the plaque. Comparing CCL19 with CCL21, the latter appears to be affected more directly by local plaque homing factors. In plaque versus normal aortic tissue, we found a much higher response to CCL21 on ECs as opposed to CCL19. Typically, the expression of CCL21 on ECs is higher than CCL19 [11
], since CCL21 has a C-terminus indicating a strong affinity to glycosaminoglycan. This is crucial for effective presentation of CCL21 on ECs [11
In contrast to our study, other investigators demonstrated an upregulation of CCL21/CCL19 in carotid plaques [7
]. In previous studies, 2–4 coronary or renal vessels from autopsy material were taken as controls. To overcome these limitations, we used 14 age-matched healthy aortic tissue samples taken from valve operations serving as controls, which were immediately frozen after their operative preparation, similar to the procedures of carotid tissue preparation. Autolysis and coincidence caused by the small number of controls were therefore ruled out in our experimental setting. A further aspect which points against the renal vessels as controls is the fact, that the vascular-associated lymphoid tissue (including DC and T-cells) has been characterized in human large arteries, including the aorta, carotid, and coronary arteries [35
] but not in renal vessels.