Development of 'Redox Arrays' for identifying novel glutathionylated proteins in the secretome

Abstract

Proteomics techniques for analysing the redox status of individual proteins in complex mixtures tend to identify the same proteins due to their high abundance. We describe here an array-based technique to identify proteins undergoing glutathionylation and apply it to the secretome and the proteome of human monocytic cells. The method is based on incorporation of biotinylated glutathione (GSH) into proteins, which can then be identified following binding to a 1000-protein antibody array. We thus identify 38 secreted and 55 intracellular glutathionylated proteins, most of which are novel candidates for glutathionylation. Two of the proteins identified in these experiments, IL-1 sRII and Lyn, were then confirmed to be susceptible to glutathionylation. Comparison of the redox array with conventional proteomic methods confirmed that the redox array is much more sensitive, and can be performed using more than 100-fold less protein than is required for methods based on mass spectrometry. The identification of novel targets of glutathionylation, particularly in the secretome where the protein concentration is much lower, shows that redox arrays can overcome some of the limitations of established redox proteomics techniques.

Figure 1. Schematic representation of the experimental procedure used for the redox arrays.

THP-1 cells were incubated with cell-permeable BioGEE to label the intracellular pool of proteins. After 24 hour incubation with LPS to allow secretion of proteins, cell conditioned medium and cell lysates were collected. After removal of free BioGEE, proteins were applied to the antibody membrane-based array. Biotin-labelled proteins were detected by the addition of streptavidin-HRP, followed by chemiluminescent detection.

Figure 2. L1000 antibody arrays (composed of L507 on the left and L493 on the right) identify a number of glutathionylated proteins secreted from LPS-treated THP-1 cells.

(a) Conditioned media from LPS-treated BioGEE-loaded THP-1 cells were applied to the array. Streptavidin-HRP was used to detect the presence of bound biotinylated proteins. Triplicate positive control spots are indicated by arrows. (b) A second aliquot of the same sample was reduced with DTT and applied to a second array.

Figure 3. Recombinant IL-1 sRII is glutathionylated in vitro and in THP-1 cells.

(a) Western blotting of recombinant IL-1 sRII treated with BioGEE in vitro probed with anti-IL-1 sRII antibody (left panel) or streptavidin-HRP (right panel). The signal from the streptavidin-labelled protein is lost upon treatment with DTT. (b) Immunoprecipitation of IL-1 sRII from cell culture media of untreated THP-1 cells preincubated with BioGEE. Biotin labelling of the IL-1 sRII was confirmed by stripping of the membrane and re-probing with streptavidin-HRP. (c) Release of IL-1 sRII from THP-1 cells in response to treatment with LPS.

Figure 4. Recombinant Lyn is glutathionylated in vitro and in THP-1 cells.

(a) Western blotting of recombinant Lyn treated with BioGEE in vitro probed with anti-Lyn antibody (left panel) or streptavidin-HRP (right panel). The signal from the streptavidin-labelled protein is lost upon treatment with DTT. (b) Immunoprecipitation of Lyn from cell lysates of untreated THP-1 cells preincubated with BioGEE. Biotin labelling of Lyn was confirmed by stripping of the membrane and re-probing with streptavidin-HRP.