Protein Aggregation Capture on Microparticles Enables Multipurpose Proteomics Sample Preparation

Abstract

Universal proteomics sample preparation is challenging because of the high heterogeneity of biological samples. Here we describe a novel mechanism that exploits the inherent instability of denatured proteins for nonspecific immobilization on microparticles by protein aggregation capture. To demonstrate the general applicability of this mechanism, we analyzed phosphoproteomes, tissue proteomes, and interaction proteomes as well as dilute secretomes. The findings present a practical, sensitive and cost-effective proteomics sample preparation method.

Keywords: Affinity proteomics; Automation; Mass Spectrometry; Phosphoproteome; Protein Denaturation*; Secretome; magnetic beads; microparticles; protein aggregation; sample preparation.

Graphical abstract

Fig. 1.

Protein aggregation driven immobilization on microparticles. A, The hypothesis of HILIC based bead interactions was tested by inducing bead - protein interaction on 20 μg of HeLa protein lysate (in 1% SDS) with the addition of acetonitrile (70% final concentration) and carboxyl coated microparticles (20 μg of Sera-mag beads) separated by magnet. The resulting supernatant was analyzed by SDS gel electrophoresis. Beads were sequentially washed three times with the indicated buffers. All washes were analyzed by SDS-PAGE for protein elution by beads including washes by milli-Q water and the lanes indicated in red. After bead washing, LDS buffer was added to the beads (see materials section) and analyzed. B, U2OS protein lysates (in 0.1% NP-40) were treated with different conditions as indicated in green. Carboxyl coated magnetic beads were added to the lysates after each treatment and the resulting supernatant analyzed by SDS-PAGE.

Fig. 2.

Elucidating the mechanism of protein aggregation capture on microparticles. A, Acetonitrile was added to HeLa lysate (in 1% SDS) to a final concentration of 70% and equal amounts of microparticles with different surface chemistries were added to the lysates and the supernatant removed. LDS buffer was added to the microparticles and the resulting supernatant analyzed by SDS-gel after removal by magnet. B, Aggregation of equal amount of HeLa lysate (20 μg at 0.25 μg/μl after addition of acetonitrile) was induced in a similar fashion as indicated above and carboxyl coated microparticles were added to the lysate at different amounts as indicated in the figure. The supernatant was removed, and the LDS buffer was added to the different samples and analyzed by SDS-PAGE after separation of microparticles by magnet. C, The hypothesized model for protein aggregation capture (PAC) on microparticles is illustrated based on the above observations.

Fig. 3.

Effects of protease digestion and post-translation modification (PTM) analysis of proteins immobilized on microparticles. A, Average (based on duplicates) percentage of peptides containing missed cleavages at arginine or lysine after digestion with Trypsin and Lys-c proteases in different combinations and ratios are displayed by heatmap. Missed cleavage rates were investigated on lysates prepared by protein aggregation on microspheres or in-solution digestion. B, Average number of unique phosphopeptide variants were counted (after removal of contaminant or reverse hits as defined by MaxQuant analysis) for the different experiments. Phosphopeptides were tallied after enrichment from lysates prepared with in-solution or PAC digestion. C, Average number of phosphorylation sites with high localization probabilities (as defined by site localization probability ≥0.75) are presented between the two different methods. D, Overlap of localized phosphorylation sites (localization probably ≥0.75) between the two experiments. The site had to be identified in two of the four replicates in both experiments for it to be valid.

Fig. 4.

Exploring the boundaries of PAC for proteomics analysis of different sample types. A, Average peptide recovery after protease digestion of equal amounts (∼1.8 mg) of mouse skeletal tissue prepared using the FASP or PAC protocol measured by nanodrop absorbance at 280/260 nm. B, The average number of proteins with SILAC ratios were counted between the two different experiments after removal of contaminating proteins and reveres hits. C, Overlap of statistically regulated protein between two preparation methods as determined by t test with permutation-based FDR of 0.05 with 250 randomizations and s0 of 0.1 (supplemental Table 4). D, Number of proteins containing LFQ intensities from the different secretome workflows. E, Number of secreted proteins between the different experiments (see Methods). F, Median Log10 transformed LFQ intensities of the proteins identified in PAC are plotted against those identified in the base proteome of Raw264.7 cells. Classical and nonclassical secreted proteins are highlighted. *Error bars represent standard deviation in all cases.