Putting Humpty Dumpty Back Together Again: What Does Protein Quantification Mean in Bottom-Up Proteomics?

Abstract

Bottom-up proteomics provides peptide measurements and has been invaluable for moving proteomics into large-scale analyses. Commonly, a single quantitative value is reported for each protein-coding gene by aggregating peptide quantities into protein groups following protein inference or parsimony. However, given the complexity of both RNA splicing and post-translational protein modification, it is overly simplistic to assume that all peptides that map to a singular protein-coding gene will demonstrate the same quantitative response. By assuming that all peptides from a protein-coding sequence are representative of the same protein, we may miss the discovery of important biological differences. To capture the contributions of existing proteoforms, we need to reconsider the practice of aggregating protein values to a single quantity per protein-coding gene.

Keywords: post-translational modifications; protein grouping; proteoforms; quantitative analysis; quantitative proteomics.

Figure 1.. Effect of proteoforms on possible peptide detection.

A single protein coding gene can be modified to give rise to dozens or many thousands of proteoforms, including those harboring multiple modifications. After proteolysis, proteoforms yield peptides that may be missed in bottom-up proteomics database searching and data processing.

Figure 2.. Technical variability is reduced when peptide measurements are combined to a protein measurement.

A human cerebrospinal fluid sample digest was analyzed by DIA-MS with 8 m/z staggered windows (4 m/z after demultiplexing). The relationship between a) peptide quantities, or b) summed protein quantities across two replicate instrument runs are plotted, with each peptide colored according to calculated percent coefficient of variation. The distribution of % coefficient of variation for c) peptides and d) summed protein quantities between replicate instrument runs, with the median % coefficient of variation for each indicated by the dashed line.

Figure 3.. The effect size on the protein level is minimized for proteins with greater numbers of peptides.

An isobaric-labeled dataset associated with the Clinical Proteomics Tumor Analysis Consortium (CPTAC),() consists of 181,389 peptides mapped to 10,495 unique protein identifiers; proteins ranged from having 1 to 563 peptides associated with them. The a) log2 fold-change and b) log10 p-value is based on a comparison of tumor residual disease. The second dataset is label free and smaller, based on a Calu-3 cell culture experiment, also publicly available (MSV000079152).() This dataset has 15,953 unique protein identifiers, with proteins represented by 1 to 311 peptides. In this dataset the a) log2 fold-change and b) log10 p-value is based on a Middle East Respiratory Syndrome (MERS) infection to a sham control. Protein sum-based quantification sums all peptide measures per protein coding gene. For b) and d) the red line indicates the significance cutoff corresponding to p=0.05, with significantly different proteins falling below the line. Figures are truncated to 50 for ease of visualization.

Figure 4.. Differential abundance profiles of tryptic peptides mapping to amyloid precursor protein.

Hippocampus tissue from four experimental groups of patients were analyzed by DIA-MS; Control/No Neuropath with normal cognitive function and no neuropathologic changes of Alzheimer’s disease including no amyloid accumulation, Control/Neuropath with normal cognitive function and intermediate or severe level of neuropathologic changes of Alzheimer’s disease, Sporadic AD with dementia and intermediate or severe level of neuropathologic changes of Alzheimer’s disease, and Autosomal dominant AD with dementia and intermediate or severe level of neuropathologic changes and an autosomal dominant mutation. For all unique peptides mapping to the amyloid precursor protein sequence, peptide measures are normalized to the mean and the mean & standard error are plotted by group. Based on known protein processing we see that the two peptides with large differences map to the amyloidogenic Aβ polypeptide.

Figure 5.. Abundance profiles of tryptic peptides mapping to a) GAPDH and b) SCG2 proteins in cerebrospinal fluid.

Three groups of human cerebrospinal fluid samples were analyzed by DIA-MS: Alzheimer’s disease, Parkinson’s disease, and healthy age and sex-matched controls. Unique peptides mapping to the proteins a) GAPDH and c) SCG2 report quantitatively on their relative expression ratios. The protein-level display integrates the mean values from all peptide-level results (box-and-whisker plot at left), with the expression ratio for each individual peptide and the group shown in the bar graphs at right. b) GAPDH has been observed as three proteoforms which form homo-tetramers from human cell lines including HEK-tsa. Intact mass spectra of the monomeric form reveal a canonical form, a persulfide-modified form, and a glutathione-modified form. Reported masses represent average masses and ppm mass error from the calculated theoretical average mass. d) SCG2 is proteolytically processed to produce several peptides, has a sulfotyrosine, and can be phosphorylated at several serine residues.