Searching for novel cardiometabolic biomarkers : method development for differential approaches and analysis of lipids involved in lipoprotein metabolism

Abstract

The current strategy for the prevention of cardiovascular diseases (CVD) is to stratify patients at risk for CVD using different algorithms. However, this has resulted in a high false-positive rate, for example, of 68% and 86% for high and intermediate CVD risks, respectively, in the Munster Heart Study (PROCAM). Novel biomarkers are hence needed to improve the reliability and the cost-benefit relationship of preventative measures. Since a biomarker should ideally be an indicator of normal and pathogenic processes as well as pharmacologic responses to a therapeutic intervention, the potential candidate new markers are those that are relevant to the pathophysiology of atherosclerosis. Several candidates have emerged, for example 27-hydroxycholesterol (27OHC), sphingosine-1-phosphate (S1P), sphingolipids and phospholipids. Research over the last decade has revealed that HDL plays a pivotal role in the pathophysiology of atherosclerosis, mainly due to its role in removing the excess of cholesterol in peripheral cells via a reverse cholesterol transport pathway. However, HDL also displays pleiotropic effects on the cardiovascular system which are independent of the reverse cholesterol transport mechanism. 27OHC and S1P are of potential interest due to their roles in HDL metabolism and function. Published LC-MS methods for the analysis of 27OHC still lacked sensitivity for the measurement in animal models (single mouse) and lipoprotein fractions. Due to this reason, an LC-MS method using atmospheric pressure photoionization (APPI) has been developed and validated. The method was more sensitive than the published LC-APCI-MS or GC-MS methods, enabling us to analyse as little as 15 L sample volume with a lower limit of quantification (LLOQ) of 40 ng/mL or 4.9 pmol on column. Subsequently, the role of 27OHC in relation to cholesterol metabolism was evaluated for the first time in patients with monogenic disorders affecting HDL metabolism. In most cases, the defects in HDL metabolism affected the 27OHC concentrations in HDL in a similar degree as the cholesterol concentrations. The most important exception was observed in samples from individuals with mutations in the lecithin:cholesteryl acyl transferase (LCAT) gene. A reduced LCAT activity appeared to impair the 27OHC esterification in the plasma compartment more pronouncedly than the cholesterol esterification. In addition, several defects in HDL metabolism led to a re-distribution of 27OHC into apoB-containing lipoproteins, which was not observed for cholesterol. An LC-MS method was developed and validated for the analysis of S1P in plasma and lipoprotein fractions. The method was subsequently used to investigate the determinants of circulating S1P in patients with inborn errors of HDL metabolism. Although only ≤5% of HDL particles carry one S1P molecule, both mildly and severely lowered HDL concentrations limited the quantity of S1P in plasma. By contrast, high concentrations of HDL-cholesterol and apoA-I did not influence S1P levels in plasma. Since S1P exerts several anti-atherogenic functions of HDL, these findings are in agreement with the concept of a threshold concentration of HDL or apoA-I which is needed for atheroprotection. Since there exist biochemical interactions among different lipids involved in cholesterol metabolism, profiling of lipids may reveal more key determinants in the pathogenesis of atherosclerosis. A sphingo- and phospholipid profiling project has been started, in which a method has been developed using a combination of normal phase HPLC separation and parent-ion scanning by a triple quadrupole mass spectrometer. Feasibility of the method was demonstrated in patients with stable coronary heart disease.