Mycobacteria produce millimolar concentrations of mycothiol (MSH) as their major low molecular weight thiol redox buffer. MSH-deficient mutants display increased sensitivity towards reactive oxygen, nitrogen and electrophilic species as well as alkylating agents and antibiotics. MSH is maintained in its reduced thiol state by the NADPH-dependent mycothiol disulphide reductase (Mtr). However, the redoxin that uses the MSH/Mtr/NADPH pathway for reduction of MSH-mixed protein disulphides, formed during oxidative stress, has long remained unknown. In this issue, Van Laer et al. report that MSH provides the reducing power for mycoredoxin-1 (Mrx1) in reduction of synthetic MSH-mixed disulphides. The reduced (dithiol) and oxidized (disulphide) solution structures of Mrx1 have been solved by nuclear magnetic resonance (NMR) spectroscopy. NMR time course experiments have also demonstrated the transient S-mycothiolation of the active site Cys14 of oxidized Mrx1 during reduction by the MSH/Mtr/NADPH electron pathway. The paper opens a new era of research to identify S-mycothiolated Mrx1 substrates and the function of MSH in redox regulation and virulence in Mycobacterium tuberculosis.
Functions of low molecular weight thiol redox buffers in bacteria
In prokaryotic and eukaryotic cells the reduced state of the cytoplasm is maintained by low molecular weight (LMW) thiol redox buffers (Fahey, 2001; Masip et al., 2006). The best studied LMW thiol redox buffer is the tripeptide glutathione (GSH) (Fig. 1A), predominantly found in eukaryotes and most Gram-negative bacteria (Masip et al., 2006). In Escherichia coli, GSH maintains the redox balance, protects cells against chlorine compounds, methylglyoxal, acid, oxidative and osmotic stress and regulates intracellular potassium levels (Masip et al., 2006).
Most Gram-positive bacteria do not produce GSH. The Actinomycetes produce mycothiol (MSH) as their LMW thiol redox buffer and MSH-deficient mutants are very sensitive to thiol-reactive species and antibiotics that affect the redox balance (Jothivasan and Hamilton, 2008; Newton et al., 2008) (Fig. 1A). In Bacillus megaterium, Bacillus cereus and Staphylococcus aureus coenzyme A (CoASH) serves as an abundant LMW thiol (Newton et al., 1996) (Fig. 1A). Many Firmicutes bacteria, including Bacillus and Staphylococcus species, have recently been discovered to utilize bacillithiol (BSH) as their major LMW thiol redox buffer (Newton et al., 2009). BSH confers resistance to hypochlorite and to the antibiotic fosfomycin in Bacillus subtilis (Gaballa et al., 2010; Chi et al., 2011). Thus, a major function of LMW thiol redox buffers is to protect bacteria against reactive oxygen, electrophilic and nitrogen species (ROS, RES and RNS) generated during respiration, cellular metabolism or host–pathogen interactions.
Reactive oxygen species cause reversible oxidations of Cys-containing proteins, including inter- or intramolecular protein disulphides and mixed protein disulphides with LMW thiols (S-thiolations). These thiol modifications control the activities of redox-sensing transcription factors that regulate ROS detoxification pathways (Antelmann and Helmann, 2011; Vazquez-Torres, 2012). Herein, the role of emerging S-thiolation mechanisms in bacteria is summarized, which puts into context the structural and mechanistic studies of mycoredoxin-1 (Mrx1) as elucidated in this issue by Van Laer et al. (2012).
Physiological roles of S-glutathionylation and S-bacillithiolation in bacteria
Protein S-glutathionylation is the reversible oxidation of protein thiols to mixed disulphides with GSH. The mechanisms implicated in S-thiolations require activation of Cys thiols to reactive Cys oxidation intermediates, such as sulphenic acid, S-nitrosyl or sulphenylchloride intermediates (Fig. 1B) (Hawkins et al., 2003; Gallogly and Mieyal, 2007; Allen and Mieyal, 2012). In eukaryotes, S-glutathionylation controls redox-sensing transcription factors and protects active site Cys thiols against overoxidation (Dalle-Donne et al., 2009). S-glutathionylation has been shown to control energy metabolism, protein synthesis, redox balance, calcium homeostasis, cytoskeletal assembly and protein folding and stability. S-glutathionylation plays a key role in cellular signalling pathways and regulates kinases and transcription factors, essential for cellular growth, differentiation and apoptosis. S-glutathionylation is further implicated in the pathogenesis of neurodegenerative and cardiovascular diseases (Dalle-Donne et al., 2009).
The principle of redox control and protein protection by S-thiolations was also suggested in bacteria a long time ago. About 45% of the total CoASH pool was oxidized to CoASH-mixed protein disulphides in spores of B. subtilis and B. megaterium, and these are quickly reduced during germination (Setlow and Setlow, 1977). However, the reduction mechanism(s) and identities of CoASH protein-mixed disulphides are not yet known.
S-glutathionylation in E. coli has so far only been shown for a few selected proteins. The oxidative stress regulator OxyR is activated by S-glutathionylation in vitro (Kim et al., 2002). In contrast, the activities of glyceraldehyde-3-phosphate dehydrogenase, methionine synthase and the PAPS reductase are inhibited by S-glutathionylation in E. coli (Lillig et al., 2003; Hondorp and Matthews, 2004; Brandes et al., 2009).
The physiological role of S-bacillithiolation in redox regulation and protein protection has been studied in detail in B. subtilis. The MarR family OhrR repressor of B. subtilis is inactivated by S-bacillithiolation at its lone Cys15 residue after cumene hydroperoxide (CHP) and NaOCl stress resulting in derepression of ohrA transcription (Lee et al., 2007; Chi et al., 2011). The thiol-dependent OhrA peroxiredoxin protects cells against ROS, including CHP and NaOCl (Fuangthong et al., 2001; Chi et al., 2011). S-bacillithiolation is also widespread among other Firmicutes with eight common and 29 unique S-bacillithiolated proteins identified in B. subtilis, Bacillus amyloliquefaciens, Bacillus pumilus, B. megaterium and Staphylococcus carnosus (Chi et al., 2012). The S-bacillithiolome contains mainly biosynthetic enzymes for amino acids (methionine, cysteine, branched chain and aromatic amino acids), cofactors (thiamine), nucleotides (GTP); as well as translation factors, chaperones, redox and antioxidant proteins (Chi et al., 2012).
The methionine synthase MetE is the most abundant S-bacillithiolated protein in Bacillus species after NaOCl exposure (Chi et al., 2011; 2012). S-bacillithiolation of MetE occurs at its Zn binding active site Cys730, leading to methionine starvation in NaOCl-treated cells. Similarly, methionine auxotrophy is caused by S-glutathionylation of MetE in E. coli after diamide stress (Hondorp and Matthews, 2004). Since formyl methionine is required for initiation of translation, MetE inactivation could stop translation during the time of hypochlorite detoxification.
Mass spectrometry identified both S-cysteinylated and S-bacillithiolated peptides for some proteins (e.g. for MetE, PpaC, SerA, GuaB, YwaA) after diamide (Hochgräfe et al., 2007) and NaOCl stress in Bacillus strains (Chi et al., 2011; 2012). These apparent S-cysteinylation events might originate from S-bacillithiolations by fragmentation during mass spectrometry. However, MetE is inactivated by S-cysteinylation in a bshA mutant strain in B. subtilis also leading to methionine auxotrophy. Thus, in the absence of BSH, S-cysteinylation appears to contribute some thiol protection and redox-regulatory role (Chi et al., 2011). However, bshA mutants displayed an increased sensitivity towards hypochlorite, suggesting that BSH is the major player in thiol protection and NaOCl detoxification.
The recent discovery of so many S-bacillithiolated proteins raises questions about the regulation of the de-bacillithiolation process. Phylogenomic profiling identified three Trx-like proteins YphP, YqiW and YtxJ in B. subtilis that are conserved among BSH-producing bacteria (Gaballa et al., 2010; Helmann, 2011). YphP and YqiW are paralogues of the DUF1094 family with a CGC active site. YphP displays weak thiol–disulphide isomerase activity in vitro (Derewenda et al., 2009) and YtxJ has a conserved TCIPS motif reminiscent of that found in monothiol glutaredoxins. The observed S-bacillithiolation of YphP and YtxJ during NaOCl stress could represent an intermediate in a bacilliredoxin redox pathway (Chi et al., 2012), although this remains to be kinetically proven.
The mechanism of protein de-glutathionylation by glutaredoxins
The pathways for reduction of protein disulphide bonds are manifold and can involve thiol–disulphide exchange reactions with LMW thiols, with the thioredoxin (Trx)/thioredoxin reductase system or with the glutaredoxin (Grx)/GSH/GSSG reductase (Gor) system in E. coli (Fernandes and Holmgren, 2004; Inaba, 2009) (Fig. 1C). Grx proteins were first discovered in E. coli (Holmgren, 1976) where they have important functions as electron donors for ribonucleotide reductase (RNR), adenosine-5′-phosphosulphate (APS) reductase, 3′-phosphoadenosine-5′-phosphosulphate (PAPS) reductase and arsenate reductases (Holmgren, 1981; Aslund et al., 1994; Oden et al., 1994; Messens and Silver, 2006). However, the most important function of Grx is the de-glutathionylation of protein GSH-mixed disulphides.
Grx proteins are structurally classified into the classical dithiol Grxs with a CPTC redox active site and the monothiol Grxs containing a CGPS redox active site (Lillig et al., 2008). In E. coli, three dithiol Grx proteins (Grx1, Grx2 and Grx3) and one monothiol protein (Grx4) have been characterized. Grx proteins have a basic Trx fold consisting of four β strands forming a β sheet embedded by three α helices with the active site motif in the loop between β sheet 1 and α helix 1 and a cis-proline located in a conserved TVP motif (Fig. 2). The N-terminal Cys is the reactive ‘nucleophilic’ Cys that is more solvent-exposed and has a lower pKa value (∼ 3.5), which is maintained in its more reactive thiolate form at physiological pH, whereas the C-terminal ‘resolving’ Cys residue is buried and has a higher pKa (Allen and Mieyal, 2012). GSH binding involves a conserved TVP motif, a GGxdD motif and positively charged residues at the N terminus (Lillig et al., 2008).
The de-glutathionylation by Grx can occur via mono- or dithiol mechanisms. Most dithiol Grxs use monothiol mechanisms that take place in two steps. In the first step, the nucleophilic thiolate anion attacks the S-glutathionylated substrate protein, resulting in reduction of the mixed disulphide, and generates an S-glutathionylated Grx (Grx–SSG) intermediate (Fig. 1C). This Grx–SSG intermediate is regenerated by thiol–disulphide exchange with GSH in the second step, leading to glutathione disulphide (GSSG) formation. The NADPH-dependent glutathione disulphide reductase (Gor) reduces GSSG to maintain the GSH/GSSG redox balance (Allen and Mieyal, 2012). The dithiol mechanism involves intramolecular disulphide bond formation with the more buried ‘resolving’ Cys residue and has been shown for some plant Grx enzymes (Zaffagnini et al., 2012). However, this dithiol mechanism is less efficient for protein de-glutathionylation and more likely involved in the reduction of intermolecular protein disulphides (Lillig et al., 2008).
The novel mycoredoxin Mrx1 is linked to the MSH/Mtr/NADPH-electron pathway
The enzymatic pathways that catalyse the reduction of S-thiolated proteins in Gram-positive bacteria have long remained elusive. In this issue, Van Laer et al. (2012) structurally and mechanistically characterize mycoredoxin-1 (Mrx1) as a Grx-like enzyme in Mycobacteria. Mrx1 catalyses de-mycothiolation in an hydroxyethyldisulphide (HED) assay and is exclusively driven by the reducing power of the MSH/Mtr/NADPH redox pathway (Fig. 1D). Oxidized Mrx1 could not be regenerated by GSH/Gor or Trx/TrxR redox couples. The authors further showed that Mrx1 is specific for de-mycothiolation of MSH-mixed disulphides and operates via a monothiol reaction mechanism analogous to most Grxs. Moreover, using nuclear magnetic resonance (NMR) spectroscopy the formation of the transient Mrx1–SSM intermediate is followed over time during reduction of oxidized Mrx1 by the MSH/Mtr/NADPH pathway. The transient Mrx1–SSM intermediate is validated by mass spectrometry using a Mrx1CxxA mutant.
The structure of Mrx1 has a typical Trx fold with a CGYC motif as the active site at the N terminus of helix α1 and a conserved T55V56P57 motif with cis-Pro57 in a groove similar to the GSH binding site of Grx proteins (Fig. 2). In Grx, this TVP motif interacts with the Cys backbone of GSH. In Mrx1, the TVP motif likely binds to the amide bond linking the CysNAc and glucosamine groups of MSH. In the Mrx1 structure, an adjacent TNPSA patch is present that is highly conserved among Mrx1 homologues from other Actinomycetes, but is absent in Grx (which have a conserved GGxdD motif in the same location). This TNPSA motif could plausibly facilitate the recognition of the GlcN-Ins motif of S-mycothiolated protein substrates (Fig. 2). The measured redox potential of Mrx1 (E0′ = −218 mV) is similar to that of Grx proteins. However, the nucleophilic Cys14 has a higher pKa value of 5.1–5.6 when hydrogen-bonded to Cys17 in the gauche minus (g−) conformation compared with Grx proteins.
The Mrx1 homologue of Corynebacterium glutamicum was shown previously to provide electrons for the arsenate reductases Cg_ArsC1 and Cg_ArsC2 during catalysis. Mrx1 reduces the arseno-mycothiol adduct [As(V)-SM] formed during ArsC1/C2 catalysis from As(V) and MSH (Ordonez et al., 2009). The As(V)-SM adduct is reduced to arsenite [As(III)] by Mrx1 and the Mrx1–SSM intermediate reduced by the MSH/Mtr/NADPH pathway (Fig. 1D). Moreover, C. glutamicum detoxifies arsenate not only by Cg_ArsC1/C2 using Mrx1 and the MSH/Mtr/NADPH pathway, but also with Cg_ArsC1′ that uses the Trx pathway (Villadangos et al., 2011). Thus, selective redox mechanisms are employed in Actinomycetes for arsenate reduction.
Outlook for future research – exploring the ‘thiolome’
The work of Van Laer et al. (2012) will open new opportunities to address fundamental questions concerning the physiological role of Mrx1 during oxidative stress and infection-related conditions in Mycobacterium tuberculosis. It will be interesting to explore which proteins are protected and/or redox-controlled by S-mycothiolation and if Mrx1 is involved in specific de-mycothiolation and redox regulation in vivo. The MSH-utilizing Mycobacterium smegmatis is able to take up MSH via active transport (Bzymek et al., 2007). Within 3 h of loading M. smegmatis cells with either 14C-Cys- or 3H-GlcN-radiolabelled MSH, 10% of the radioactivity is consistently found in the precipitated proteins after extraction of soluble MSH in 50% acetonitrile. This suggests the presence of MSH protein-mixed disulphides. Targets for S-mycothiolation could include the methionine synthase MetE or the redox-sensitive anti-sigma factor RsrA that is oxidized after diamide stress leading to release of SigmaR, a disulphide stress-specific sigma factor. Free SigmaR transcribes genes for thioredoxin, thioredoxin reductase and MSH biosynthesis enzymes in Streptomycetes (Park and Roe, 2008). Thus, it is possible that RsrA is S-mycothiolated by diamide stress and reduced by Mrx1 via the MSH/Mtr/NADPH pathway.
There is much to be discovered to understand the role of S-mycothiolated Mrx1 substrates in oxidative stress and virulence in Actinomycetes. Mrx1-based thiol redox proteomics methods, such as the NEM Biotin switch assay (Kehr et al., 2011) can be applied to identify Mrx1 substrates at a proteome-wide scale. Looking forward, it will be interesting to compare targets for S-thiolations and their regulatory roles across different BSH-, GSH- and MSH-utilizing bacteria from a systems-wide redox proteomics perspective. The future is bright; the future is thiolated.
The authors' laboratories are supported by grants from the Deutsche Forschungsgemeinschaft (AN746/2-1 and AN746/3-1) to H. A., and by grants from the Biotechnology and Biological Science Research Council (BB/H013504/1) and the Leverhulme Trust (RPG-2012-606) to C. J. H.