Proteomic analysis of HDL isolates reveals method-driven variability: An interlaboratory approach

The high-density lipoprotein (HDL) is the most heterogeneous and protein-rich lipoprotein, and its complex proteome has been correlated with many distinct properties. However, the lack of a standardized approach for HDL isolation, combined with the fact that different methods capture overlapping, but distinct HDL subspecies make the establishment of a composition-function relationship a challenge. Key factors influencing HDL proteomic profile are the isolation methodology and the technical variability associated with it. Importantly, interlaboratory technical variability associated with HDL isolation methodologies and how it affects the HDL proteome has never been determined. Here, we used two common methods to isolate HDL particles, ultracentrifugation (UC) and immunoaffinity chromatography against APOA1 (IAC), and performed a thorough evaluation of intralaboratory repeatability as well as intra- and interlaboratory reproducibility of these isolation methods, assessing their influence on HDL-associated proteins composition and abundance. Our results demonstrate that methodological variability has a greater impact on the HDL proteome than interlaboratory differences, with almost 60% of the variance in the data explained by the method of isolation. Importantly, the top 15 HDL proteins account for > 90% of the protein mass in HDL, regardless of the isolation method, and the variability in protein quantification is inversely associated with protein abundance. A joint analysis combining interlaboratory reproducibility of top 15 HDL proteins in multiple days shows 11 and 10 proteins with CV < 25% for UC and IAC, respectively. In summary, our findings demonstrate that standardized isolation methods can achieve acceptable reproducibility within and across laboratories, but they may capture distinct HDL particles.