Figure 2. Proteomics analysis of serum glycoproteins from treated and untreated mice. 21 days after cell injection, mice underwent cryo-thermal or no treatment as control. Afterwards, mice were kept for another 2 h, 2, 5, 8, 11, 14, 21 and 28 days. For shotgun proteomics, serum were collected and pooled under each condition. After protein denaturation, reduction and alkylation, pooled sera were then digested to peptides. Subsequently, glycopeptides were captured by the hydrazide solid phase and cleaved by PNGase F enzyme. The released N-glycosylated peptides were than labeled with 8-plex iTRAQ reagent and mixed for subsequent off-gel pre-fractionation. N-glycosylated peptides extracted from healthy mice were labeled as the global reference. Each fraction was then subjected to LC-MS/MS for protein identification and quantification. Database search and analysis were performed in TPP and significant candidates were verified using quantitative PRM method for target analysis.
Figure 3. Proteomics identification and quantification. (A) The overlap of glycoproteins identified and quantified in three groups. (B) The estimated protein concentration in serum for those identified proteins, spanning 7 orders of magnitude.
Figure 4. Quantitative time course measurement of induced proteome changes. (A) The unsupervised hierarchical clustering of significant proteins’ changes over time course. The dendrogram generated represents the clustering of proteins (a-d). Biological process enrichment is based on IPA functional annotation. (B) The schematic graph of systematic response regulated in a time resolved pattern. Arrows in red represented proteins up-regulated, while in green are proteins down-regulated or in dysregulation.
Target proteomics investigating the function of acute phase response in anti-tumor activity.
To better understand how acute phase response acts in the anti-tumor activity, PRM based targeted proteomics validation on significant proteins was carried out in the following study. 5-6 mice under treated or control condition over 8 time points from the same sample as shotgun proteomics were analyzed, ending up with a total of 94 serum samples. Peptides used for target analysis were selected from previous shotgun proteomics via the following criteria: 1) N-glycosylated with fully tryptic termini, 2) commonly detected across all samples, and 3) doubly charged preferred. As a quality control, we spiked a commercially available standard angiotensin peptide in each enriched sample to evaluate the instrument performance. Additional File 7: Figure S5 showed that most of the coefficient variances (CV) of the angiotensin intensity under each condition were below 20%, and the CV value among different conditions was 22%, suggesting a good performance of instrument operation. Relative protein abundance of treat and untreated mice were expressed as fractions of the normalized value of the healthy mice which was set to 1 (base line). This PRM-based target analysis allowed us to conduct comprehensive validation of proteins in multiple biological replicates without pooling, which is not suitable for traditional approaches as western blot and ELISA, due to limitations from time-consuming and antibody dependence.