Novel methods for the identification of cellular S-glutathionylated proteins and sites of glutathionedependent modification using affinity chromatography and proteomic analyses

Abstract

Reactive oxygen species (ROS) are formed during normal respiration in the mitochondria through electron leakage to oxygen. During normal metabolism, several enzymes and low-molecular weight antioxidants work towards eliminating ROS in a well-coordinated manner. The tripeptide glutathione is an integral part of this antioxidant network, and functions both as a reducing factor and as a substrate by several enzymes. When the antioxidant defence of the cell is overwhelmed, "oxidative stress" develops, and cellular redox systems become oxidised. Oxidative stress can result through metabolism of foreign compounds, ischemia-reperfusion, or exposure to UV-radiation, and may also develop naturally during aging.

Glutathione (GSH) has long been known to reversibly bind cellular proteins, particularly during oxidative stress, in a reaction catalysed by glutaredoxin, previously known as thioltransferase. These modified protein substrates include several metabolic enzymes and cytoskeletal proteins, which may be thus functionally regulated as a cellular response to stress. However, a comparison of the constitutive and the stressed situations is difficult, since the current analysis protocols for high-throughput analysis are not sensitive enough to identify low level glutathionylation during constitutive metabolism. Hence, there is interest in developing more sensitive methods to accurately portray glutathionylation patterns in the cell.

A method was developed based on the specific reduction of glutathione-protein mixed disulphides by glutaredoxin, the reaction of liberated protein thiols with N-ethylmaleimide-biotin, affinity purification of tagged proteins, and identification by twodimensional gel electrophoresis and mass spectrometry. The method unequivocally identified 43 mostly novel cellular protein substrates for S-glutathionylation. These included protein chaperones, cytoskeletal proteins, cell cycle regulators, and enzymes of intermediate metabolism. The method developed in this study is high-throughput, and more importantly, can specifically identify S-glutathionylated proteins from cells - both stressed cells and cells undergoing constitutive metabolism - with minimal disruption of cellular function. Therefore, the method can successively be used to study posttranslational redox modification of cellular proteins - a potentially significant biochemical control mechanism in coordinating cellular function.

The protocol was further developed to allow for affinity purification and analysis of trypsinised S-glutathionylated proteins under non-reducing conditions, in an attempt to avoid the cumbersome 2-D gel electrophoresis step and at the same time identify the sites of S-glutathionylation by tandem mass spectrometry. The glutathionylation sites of gamma-actin, heat-shock protein 70 and elongation factor 1-alpha-1 were identified in this way, but the method was hampered by unspecific binding of peptides to the avidin-affinity column, a problem also encountered in other studies.

In this study, for the first time, a method has been developed for high-throughput identification of S-glutathionylated proteins without interference with cellular function. This protocol has also been modified to include identification of the S-glutathionylation sites of the proteins involved, although difficulties in peptide affinity capture interfere with specificity and high-throughput analysis.