Chemical composition and the potential for proteomic transformation in cancer, hypoxia, and hyperosmotic stress

Abstract

Models of chemical composition and thermodynamic potential emphasize the microenvironmental context for proteomic transformations. Here, data from 71 comparative proteomics studies were analyzed using elemental ratios and chemical components as measures of oxidation and hydration state. Experimental lowering of oxygen availability (hypoxia) and water activity (hyperosmotic stress) are reflected in decreased oxidation and hydration states of proteomes, but up-expressed proteins in colorectal and pancreatic cancer most often have higher average oxidation or hydration state. Calculations of chemical affinity were used to quantify the thermodynamic potentials for transformation of proteomes as a function of fugacity of O 2 and activity of H 2 O, which serve as scales of oxidation and hydration potential. Potential diagrams for the overall reactions between down-and up-expressed proteins in various datasets have predicted equipotential lines that cluster near unit activity of H 2 O, suggesting that protein expression is sensitive to local variations in water activity. A redox balance calculation indicates that an increase in the lipid to protein ratio in cancer cells by 20% over hypoxic cells would generate an electron sink of sufficient size for oxidation of the cancer proteomes. These findings demonstrate consistent, condition-dependent chemical differences between proteomes, identify primary microenvironmental constraints on proteomic transformations, and help to quantify the redox balance among cellular macromolecules. The datasets and computer code used here are supplied in a new R package, canprot.

Water as a reactant in the differential expression of proteins in cancer

“…An earlier compositional analysis of proteomic data provided preliminary evidence for a higher hydration state of proteomes in colorectal and pancreatic cancer 21. The compilation of differential expression data was expanded in this study to include breast, liver, lung, and prostate cancer as well as pan‐cancer transcriptomic and proteomic data from The Cancer Genome Atlas (TCGA) and Human Protein Atlas (HPA).…”