Presented by

Dr. Sandra Goetze

The main function of high density lipoprotein (HDL) particles in humans is to transport cholesterol from tissues to the liver, so that excessive cholesterol can be eliminated in the bile. High plasma levels of HDL cholesterol are clinically associated with a lowerrisk of coronary heart disease (CHD) and diabetes mellitus type 2 (T2DM). However, HDL could not be successfully exploited for the prevention or treatment of CHD or T2DM. This is mainly due to a lacking structure-function relationship of the HDL-particle composition. Here, we set out to characterize HDL-components (proteins/lipids) disturbed in patients with CHD and/or T2DM and to link particle composition with phenotypic functionality by implementing biological assays as readout.

Using Data-Independent Acquisition (DIA), we generated a quantitative data matrix of 182 human proteins over 166 samples including processing controls. Compared to HDL of healthy individuals, HDL of CHD-patients was characterized by the relative enrichment of specific proteins whereas HDL of T2DM patients was rather characterized by the relative loss of specific proteins. Upon elastic net analysis, several proteins emerged as independent determinants of at least one of the observed biological functions of HDL. Glycosylphosphatidyl-inositol specific phospholipase D1 (GPLD1) for instance was one of the strongest determinants for the ability of HDL to inhibit starvation induced apoptosis of HAECs. To experimentally validate our findings, we used reconstituted HDL (rHDL), consisting of purified apoA-I and dioleyl-phosphatidylcholine (DOPC) with or without the protein or lipid to be tested for biological activity. As predicted by the model, complementation of rHDL with GPLD1 improved the ability of rHDL to inhibit the apoptosis of HAECs.

In conclusion, our integrative approach unraveled several novel HDL-determinants with potentially anti-diabetic or vasoprotective functions.