Reliability of Protein Abundance and Synthesis Measurements in Human Skeletal Muscle

Abstract

The repeatability of dynamic proteome profiling (DPP), which is a novel technique for measuring the relative abundance (ABD) and fractional synthesis rate (FSR) of proteins in humans, is investigated. LC-MS analysis is performed on muscle samples taken from male participants (n = 4) that consumed 4 × 50 mL doses of deuterium oxide (² H₂ O) per day for 14 days. ABD is measured by label-free quantitation and FSR is calculated from time-dependent changes in peptide mass isotopomer abundances. One-hundred one proteins have at least one unique peptide and are used in the assessment of protein ABD. Fifty-four of these proteins meet more stringent criteria and are used in the assessment of FSR data. The median (M), lower-, (Q₁ ) and upper-quartile (Q₃ ) values for protein FSR (%/d) are M = 1.63, Q₁ = 1.07, and Q₃ = 3.24, respectively. The technical CV of ABD data has a median value of 3.6% (Q₁ 1.7% to Q₃ 6.7%), whereas the median CV of FSR data is 10.1% (Q₁ 3.5% to Q₃ 16.5%). These values compare favorably against other assessments of technical repeatability of proteomics data, which often set a CV of 20% as the upper bound of acceptability.

Keywords: biosynthetic labelling; coefficient of variation; deuterium oxide; fractional synthesis rate; heavy water; precision; protein turnover; proteome dynamics; repeatability.