SoxR, a [2Fe-2S] transcription factor, is active only in its oxidized form.

SoxR protein is known to function both as a sensor and as a transcriptional activator for a superoxide response regulon in Escherichia coli. The activity of SoxR was tested by its ability to enable the transcription of its target gene, soxS, in vitro. The activity of the oxidized form was lost when its [2Fe-2S] clusters were reduced by dithionite under anaerobic conditions, and it was rapidly restored by autooxidation. This result is consistent with the hypothesis that induction of the regulon is effected by the univalent oxidation of the Fe-S centers of SoxR. In vivo, this oxidation may be caused by an alteration of the redox balance of electron chain intermediates that normally maintains soxR in an inactive, reduced state. Oxidized SoxR was about twice as effective as reduced SoxR in protecting the soxS operator from endonucleolytic cleavage. However, this difference could not account for a greater than 50-fold difference in their activities and therefore could not support a model in which oxidation activates SoxR by enabling it to bind to DNA. NADPH, ferredoxin, flavodoxin, or ferredoxin (flavodoxin):NADP+ reductase could not reduce SoxR directly in vitro at a measurable rate. The midpoint potential for SoxR was measured at -283 mV.

Glutathione-mediated destabilization in vitro of [2Fe-2S] centers in the SoxR regulatory protein.

SoxR is a transcription factor that governs a global defense against the oxidative stress caused by nitric oxide or excess superoxide in Escherichia coli. SoxR is a homodimer containing a pair of [2Fe-2S] clusters essential for its transcriptional activity, and changes in the stability of these metal centers could contribute to the activation or inactivation of SoxR in vivo. Herein we show that reduced glutathione (GSH) in aerobic solution disrupts the SoxR [2Fe-2S] clusters, releasing Fe from the protein and eliminating SoxR transcriptional activity. This disassembly process evidently involves oxygen-derived free radicals. The loss of [2Fe-2S] clusters does not occur in anaerobic solution and is blocked in aerobic solution by the addition of superoxide dismutase and catalase. Although H2O2 or xanthine oxidase and hypoxanthine (to generate superoxide) were insufficient on their own to cause [2Fe-2S] cluster loss, they did accelerate the rate of disassembly after GSH addition. Oxidized GSH alone was ineffective in disrupting the clusters, but the rate of [2Fe-2S] cluster disassembly was maximal when reduced and oxidized GSH were present at a ratio of approximately 1:3, which suggests the critical involvement of a GSH-based free radical in the disassembly process. Such a reaction might occur in vivo: we found that the induction by paraquat of SoxR-dependent soxS transcription was much higher in a GSH-deficient E. coli strain than in its GSH-containing parent. The results imply that GSH may play a significant role during the deactivation process of SoxR in vivo. Ironically, superoxide production seems both to activate SoxR and, in the GSH-dependent disassembly process, to switch off this transcription factor.