Hypersusceptibility of neutrophil granulocytes towards lethal action of free fatty acids contained in enzyme-modified atherogenic low density lipoprotein

Abstract

Objective

The bulk of LDL entrapped in the arterial intima is modified by hydrolytic enzymes, leading to extensive cleavage of cholesterylesters and liberation of fatty acids. The latter induce apoptosis in endothelial cells but are far less cytotoxic towards macrophages. We have compared the cytotoxic effects of enzymatically modified LDL (E-LDL) on macrophages and polymorphonuclear granulocytes (PMN).

Methods and results

E-LDL displayed toxicity towards PMN at far lower concentrations than towards monocyte-derived macrophages. Native or oxidized LDL had no effect. Free fatty acids contained in E-LDL were the cause of the observed toxicity, which could be mimicked by linoleic acid, oleic acid and arachidonic acid. E-LDL provoked Ca²⁺ influx and activated PMN, as witnessed by the generation of superoxide anions and peroxidase secretion. Inhibition of either oxidative burst or calcium influx did not diminish the cytotoxicity of E-LDL. Similar to free linoleic acid, E-LDL lysed red blood cells and rapidly rendered cells permeable to propidium iodide.

Conclusion

Possibly through their capacity to directly perturb cell membranes, free fatty acids contained in E-LDL exert potent cytotoxic effects on PMN. This may be one reason why PMN are not abundantly present in atherosclerotic lesions, and why PMN-depletion suppresses atherogenesis.

Introduction

Atherosclerosis is triggered by accumulation of LDL in the vessel wall and subsequent overactivation of the immune system [1], [2]. It is uncontested that high blood LDL levels promote atherosclerosis and that intimal LDL must be chemically modified to become atherogenic. The nature of this modification is still a matter of debate. Oxidative modifications are generally favored [1], but we have presented evidence that modification by hydrolytic enzymes may be at least equally relevant. Enzymatically modified LDL (E-LDL) is extensively present in lesions as demonstrable with the use of specific monoclonal antibodies [3], and it constitutes the major fraction of lipid extracted from early human atherosclerotic lesions [4]. Free fatty acids (FFA) and free cholesterol are liberated in the modification, which involves the combined action of proteases and cholesterylesterase. Formation of E-LDL provides a plausible explanation for the known presence of large amounts of free cholesterol in early atherosclerotic lesions [5]. Enzymatic remodeling of LDL also exposes phosphorylcholine groups which can subsequently bind CRP leading to complement activation. This accounts for the co-localization of CRP with E-LDL and activated complement components at the early stages of atherosclerotic lesion development [6]. Finally, E-LDL is a potent inducer of macrophage foam cell formation, provokes the production of inflammatory cytokines by lesional cells, and has cytotoxic properties [7], [8], [9]. We have summarized our findings in a hypothesis that provides a new concept on atherosclerosis development [6].

Polymorphonuclear granulocytes (PMN) are generally not detected in atherosclerotic lesions [10], although their presence alongside with macrophages may be expected. IL-8 is produced by E-LDL-stimulated endothelial cells [9], and chemotactic peptides generated through complement activation would further be expected to promote PMN immigration into the lesion [11]. The adhesion molecules that mediate rolling, adhesion and transmigration of PMN are present in atherosclerotic lesions [12], [13]. Indeed, neutrophils were identified as the predominant rolling leukocytes in a mouse model of atherosclerosis [14], and granulocyte-specific enzymes, including PMN elastase and matrix metalloprotease 8 have been detected in atherosclerotic lesions [15], [16], [17].

Evidence for the presence of PMN in early atherosclerotic lesions was recently produced by Zernecke et al. Moreover, lesion progression in Apoe^−/− mice was blunted by depletion of circulating PMN, indicating a significant impact of PMN on atherosclerosis in this murine model [18]. In that study, it was shown that the PMN rapidly underwent apoptosis in the lesions, and this would provide an explanation for the difficulty of their detection.

While the study of Zernecke et al. was underway, an investigation had been launched in our laboratory into the possibility that PMN might be particularly susceptible to the toxic action of E-LDL. We report that, in contrast to macrophages, human PMN are rapidly killed upon exposure to low concentrations of E-LDL. In vitro, this is accompanied by cell activation, possibly triggered by direct membrane perturbation by free fatty acids contained in the hydrolysed lipoprotein.