S-Glutathionylation: A Redox-Sensitive Switch Participating in Nitroso-Redox Balance

Abstract

S-Glutathionylation Uncouples eNOS and Regulates Its Cellular and Vascular Function

Cysteine side chains of proteins are being increasingly appreciated as the site of major posttranslational modifications that exert profound degrees of protein regulation.^1–3 One of the consequences of tissue nitroso-redox imbalance is a process by which regulatory thiols switch from a state of physiologic regulation by S-nitrosylation to a state of dysregulated function because of oxidation.^4,5 A key example of this is well described for regulation of the ryanodine receptor/Ca²⁺- release channels 1 and 2.^6–13

The interaction between reactive oxygen species (ROS) and reactive nitrogen species can lead to the inactivation of nitric oxide (NO), formation of further highly reactive nitrogen species, such as peroxynitrite, or the dysregulation of biological signaling processes via irreversible oxidative modification of protein thiol moieties.⁵ It is increasingly recognized that an important posttranslational modification of thiols within this spectrum of biochemical reactions is that of S-glutathionylation.^14–16 In a recent report in Nature, Chen and colleagues¹⁷ describe a fascinating feed forward system, in which a redox shift enhances S-glutathionylation of endothelial NO synthase (eNOS or NOS3) (Figure). This modification, in turn, shifts eNOS from an NO-generating enzyme to a producer of ROS.

Figure.

Schematic representation of oxidative stress-mediated S-glutathionylation (GS) of eNOS. With normal nitroso-redox balance (left), eNOS activity is finely regulated by dynamic S-nitrosylation of susceptible cysteine residues. Under these conditions, eNOS is activated by agonist-mediated denitrosylation. A shift toward oxidative stress and redox imbalance (left) dysregulates the GSH/GSSG ratio with consequent eNOS uncoupling mediated by S-glutathionylation. This causes eNOS to shift from NO to ROS production, further disrupting the nitroso-redox balance, thereby creating a feed-forward potentiation of oxidative stress.

One of the key principles of posttranslational modification of cysteine moieties is specificity; that is to say, there are specific cysteine sites that undergo posttranslational modification by S-nitrosylation, oxidation, or S-glutathionylation. In this regard, Chen et al use site-directed mutagenesis to identify Cys 689 and 908 as the key sites of regulation and show that oxidized glutathione (GSSG) stimulates protein S-glutathionylation, which in turn produces a mixed disulfide bond between the reactive Cys-thiol and reduced glutathione (GSH), a tripeptide consisting of glycine, cysteine, and glutamate.¹⁷ They further argue that addition of this bulky negatively charged group alters protein structure and function in a similar manner to the addition of a phosphate. Accordingly, this posttranslational modification must be considered in the spectrum of allosteric modifications that include S-nitrosylation and oxidation. Of note, endogenous eNOS activity or NO exposure promotes physiologic S-nitrosylation of the cysteine residues 99 and 94 of the zinc tetrathiolate cluster at the dimeric interface of human eNOS, which in turn, inhibits enzyme activity.^18,19 In resting endothelial cells, eNOS is constitutively S-nitrosylated. In response to the agonist vascular endothelial growth factor, the enzyme undergoes rapid transient denitrosylation, as a mechanism of enzyme activation and, then, is progressively renitrosylated to its original level, representing receptor-modulated reversible S-nitrosylation.¹⁹ Interestingly, agonist-induced denitrosylation is closely linked with eNOS Ser¹¹⁷⁹ phosphorylation by the phosphoinositide 3-kinase/Akt pathway.¹⁹ Together, these studies reveal dynamic regulation of eNOS by various redox-sensitive thiol modifications.

Inhibition of glutathione reductase by 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) decreases the cellular GSH/GSSG ratio, leading to protein S-glutathionylation. Chen et al. use BCNU to dose-dependently increase superoxide production and S-glutathionylation of eNOS in endothelial cells, and this is abolished by the reducing agent dithiothreitol.¹⁷ In agreement with this observation, endothelium-dependent relaxation of aortic rings is reversibly impaired by BCNU. Thus, S-glutathionylation could be considered to fall within a spectrum of oxidative modifications of thiol moieties that also include formation of sulfenic, sulfinic, and sulfonic acids, as well as disulfides.^1,20 These oxidative modifications are progressively less reversible reactions and, therefore, maladaptive to the extent that they block the more reversible and physiologic regulation mediated by S-nitrosylation.¹ Although oxidation of thiols is nonspecific and irreversible, particularly when the cysteine is oxidized to sulfinic or sulfonic acid, recent work demonstrated that oxidation of thiols (to sulfenic acid) could be specific, reversible, and controlled.²¹ Thus, as has been described by Irani et al,²² ROS can under certain circumstances serve as signaling molecules. There is also some evidence that S-glutathionylation may be a reversible modification, following restoration of a reducing GSH/GSSG ratio, and a protective mechanism against irreversible oxidation of regulatory thiols. However, it remains to be determined whether this modification is protective or detrimental in pathologic conditions associated with whole-body oxidative stress. For example, there is growing evidence that S-glutathionylated hemoglobin may be a useful biomarker of blood oxidative stress in humans,²³ shown to be increased in patients with diabetes, hyperlipidemia, and renal failure.^24,25 Whether this modification alters protein function in a way that impacts on tissue injury and disease progression or is simply a biomarker for the presence of oxidative stress is, at present, not known.²⁰

Pathophysiologic Relevance

The work by Chen et al includes a highly important description in an in vivo model of hypertension, the spontaneously hypertensive rat.¹⁷ In these studies they show increased S-glutathionylation in immunoprecipitated eNOS, which was consistent with the immunohistology of aortae demonstrating that eNOS highly colocalizes with glutathionylated protein. Importantly, the authors show that dithiothreitol reduces S-glutathionylation in this animal model concomitantly with restoration of acetylcholine-induced aortic relaxation.

The findings in this study can be considered to represent a feed forward checkpoint, whereby oxidative stress promotes S-glutathionylation of eNOS, which in turn further enhances cellular nitroso-redox imbalance by shifting eNOS production of NO toward further ROS. This would be expected to continue to promote nitroso-redox imbalance and continue to stimulate a milieu in which susceptible thiols undergo oxidative or S-glutathionylated modifications. Findings such as these lead to exciting future questions and add further motivation toward the quest for developing new therapies targeting nitroso-redox balance.

In summary, the findings in this study add to the substantial amount of evidence pointing to a critical regulatory role of posttranslational thiol modification in pathophysiologic states, such as hypertension, atherosclerosis, and heart failure. In addition, they further contribute to the development of new therapeutic drugs with thiol-reducing properties, which may improve endothelial dysfunction, restore vascular tone, enhance excitation-contraction coupling, and improve left ventricular function in patients suffering from cardiovascular diseases.