Quantitative Proteome Profiling of Normal Human Circulating Microparticles

Abstract

Circulating microparticles (MPs) are produced as part of normal physiology. Their numbers, origin, and composition change in pathology. Despite this, the normal MP proteome has not yet been characterized with standardized high-resolution methods. We here quantitatively profile the normal MP proteome using nano-LC-MS/MS on an LTQ-Orbitrap with optimized sample collection, preparation, and analysis of 12 different normal samples. Analytical and procedural variation were estimated in triply processed samples analyzed in triplicate from two different donors. Label-free quantitation was validated by the correlation of cytoskeletal protein intensities with MP numbers obtained by flow cytometry. Finally, the validity of using pooled samples was evaluated using overlap protein identification numbers and multivariate data analysis. Using conservative parameters, 536 different unique proteins were quantitated. Of these, 334 (63%) were present in all samples and represent an MP core proteome. Technical triplicates showed <10% variation in intensity within a dynamic range of almost 5 decades. Differences due to variable MP numbers and losses during preparative steps could be normalized using cytoskeletal MP protein intensities. Our results establish a reproducible LC-MS/MS procedure, provide a simple and robust MP preparation method, and yield a baseline MP proteome for future studies of MPs in health and disease.

Robust Label-free, Quantitative Profiling of Circulating Plasma Microparticle (MP) Associated Proteins

“…between health and disease. Despite this, only Østergaard and colleagues focused on the establishment of a reproducible workflow for quantitative proteome profiling of circulating plasma MP, achieving an intraday assay variability of less than 10% (18). We show here interassay variability of better than 3% over a period of more than 3 months by using two PFP samples prepared from different blood draws of the same individual.…”

“…7). Our proteomes and the one of Østergaard and colleagues (18) were very much alike, with plasma membrane proteins contributing a majority of proteins and being overrepresented, nucleus proteins clearly under-represented and ribosomal proteins present with only few members. This is in stark contrast with the MP proteome published by Harel et al (29), where plasma membrane proteins were not the main class of proteins and being under-represented, whereas ribosome and nucleus proteins were over-represented and the latter formed the most populated class.…”

“…For instance, lipoproteins are considered to contaminate MP prepared by differential centrifugation because of similar densities (28), hence are not MP associated. Immunoglobulin chains and complement factors had been described to associate with MP (8,18,19). The median ratios of immunoglobulin chains and complement factors, like the ones of apolipoproteins, were smaller than zero, indicating no or only loose association of these proteins with MP (supplemental Fig.…”

“…We compared the proteomes of the twelve healthy volunteer MP preparations, referred to as MP core, and the freezer study with two published proteome studies on circulating MP (18,29), and with a multicell proteome from eleven different cell lines (30), as well as one platelet proteome (31) by means of gene ontology (GO) cellular component terms (Fig. 7).…”

“…Studies did not focus on, or had problems with, repeatability rendering statistical evidence testing a difficult task (5,8,18). There are three reasons for this: First, the MP proteome represents less than 0.01% of the plasma proteome, hence variation in high abundant plasma protein contaminations do variably suppress the detection of MP proteins (19).…”