Refined purification strategy for reliable proteomic profiling of HDL(2/3): Impact on proteomic complexity

Proteomics have extended the list of high-density lipoprotein (HDL) associated proteins to about 90. One of the major issues of global protein characterization is establishing specificity of association as opposed to contamination, a fact which has never been addressed for isolated HDL. We have developed a refined purification strategy to isolate HDL by density, followed by purification by size to generate “highly purified” fractions of HDL(2/3), which allow the reliable quantification of the HDL proteome for biomarker discovery. Mass spectrometry analysis revealed that the proteome of HDL(2/3) is composed of 10–16 different proteins, which is in striking contrast to previous reports. Importantly, proteomic analysis revealed that many proteins which have recently been described to be associated with HDL, including α-1-antitrypsin, α-2-HS-glycoprotein, serotransferrin, apolipoprotein A-IV and others, are not associated with HDL(2/3) and are exclusively found in a different molecular weight fraction containing human serum albumin, lipid-poor apolipoprotein A-I and other proteins. Interestingly, proteins found in this lower molecular weight fraction commonly share lipid-binding properties and enrichment of serum with free fatty acids/lysophophatidylcholine led to a significant increase in co-isolation of lipid-binding proteins such as albumin and α-1-antitrypsin. We propose that this refined method might become a standard in proteomic assessment of HDL(2/3) making data from clinical cohorts more comparable and reproducible.