State-of-the-art housekeeping proteins for quantitative western blotting: Revisiting the first draft of the human proteome

Abstract

Western blotting (WB) analysis is the most popular and widely used methodology for protein detection and characterization over recent decades. In accordance with the advancement of the technologies for the acquisition of WB signals, a quantitative value is used to present the abundance of target proteins in a complex sample, thereby requiring the use of specific proteins as internal references that represent total proteins. Heretofore, proteins encoded by housekeeping genes such as GAPDH, β-tubulin and β-actin have been commonly used as loading controls without any hesitation because their mRNA expression levels tend to be high and constant in many different cells and tissues. Experimentally, however, some of the housekeeping reference proteins are often displayed with inconsistent expression levels in both homogeneous and heterogeneous tissues, and, in terms of mRNA levels, they have a weak correlation to the abundance of proteins. To estimate accurate, reliable, and reproducible protein quantifications, it is crucial to define appropriate reference controls. For this paper, we explored the recently released large-scale, human proteomic database ProteomicsDB including 16 857 liquid chromatography tandem-mass-spectrometry data from 27 human tissues, and suggest 20 ubiquitously- and constitutively-expressed, putative internal-reference controls for the quantification of differential protein expressions. Intriguingly, the most commonly used, known housekeeping genes were entirely excluded in our newly defined candidates. Although the applications of the candidates under many different biological conditions and in other organisms are yet to be empirically verified, we propose reliable, potential loading controls for a WB analysis in this paper.

Keywords: Bioinformatics; Housekeeping proteins; Reference control; Western blot analysis.