Summary
Protein arginine phosphorylation is a recently discovered modification that affects multiple cellular pathways in Gram-positive bacteria. In particular, the phosphorylation of arginine residues by McsB is critical for regulating the cellular stress response. Given that the highly efficient protein arginine phosphatase YwlE prevents arginine phosphorylation under non-stress conditions, we hypothesized that this enzyme negatively regulates arginine phosphorylation and acts as a sensor of cell stress. To evaluate this hypothesis, we developed the first suite of highly potent and specific SO3-amidine based YwlE inhibitors. With these protein arginine phosphatase specific probes, we demonstrated that YwlE activity is suppressed by oxidative stress, which consequently increases arginine phosphorylation, thereby inducing the expression of stress response genes, which is critical for bacterial virulence. Overall, we predict that these novel chemical tools will be widely used to study the regulation of protein arginine phosphorylation in multiple organisms.
eTOC blurb
Protein arginine phosphorylation is an elusive protein modification that is involved in cell signaling. Based on the development of selective chemical probes, Fuhrmann et al. expose the molecular basis for how cells regulate protein arginine phosphorylation under oxidative stress conditions.

Introduction
Post-translational modifications (PTMs) constitute a hallmark of prokaryotic and eukaryotic cell signaling. The most widespread and best studied PTM is protein phosphorylation. This dynamic protein modification is reversibly regulated by the action of protein kinases, which transfer a phosphoryl group from a phospho-donor (usually ATP) onto the respective phospho-acceptor site, and protein phosphatases, which counteract kinase activity by catalyzing phosphoryl group hydrolysis. The most common and best characterized type of protein phosphorylation is O-phosphorylation, where the phosphoryl group is attached to the side chain hydroxyl group of serine, threonine and tyrosine residues to generate a phosphate monoester. In addition to these residues, the side chain nitrogens of histidine, arginine and lysine are also phosphorylated (i.e., N-phosphorylation) (Attwood et al., 2007; Besant et al., 2009; Klumpp and Krieglstein, 2002; Matthews, 1995). Although the existence of protein N-phosphorylation has been known for decades, the scope and importance of these PTMs, especially in eukaryotic organisms, is only slowly emerging. The lack of studies in this area likely relates to the fact that these PTMs are more difficult to study because, in contrast to the O–P bond, the N–P bond is highly unstable under the strongly acidic conditions that are usually employed during phosphopeptide analysis (Engholm-Keller and Larsen, 2013; Fuhrmann et al., 2009; Fuhrmann et al., 2015b; Kee and Muir, 2012; Kowalewska et al., 2010; Schmidt et al., 2013a).
As such, little is known about the phosphorylation of arginine residues and even less about the enzymes mediating this modification, with only a handful of reports describing the occurrence of protein arginine phosphorylation (Matthews, 1995). More recently, however, McsB was unequivocally identified as the first protein arginine kinase (PRK) in gram-positive bacteria (Fuhrmann et al., 2009; Jung and Jung, 2009). Subsequent reports identified YwlE as its cognate protein arginine phosphatase (PAP) (Elsholz et al., 2012; Fuhrmann et al., 2013a). McsB is present in more than 150 bacterial species (Suzuki et al., 2013) and is thought to have evolved from the guanidinium kinase family, whereas the arginine phosphatase YwlE belongs to the low molecular weight-protein tyrosine phosphatase (LMW-PTP) family (Ramponi and Stefani, 1997). Based on detailed structural analyses, YwlE dephosphorylates phosphoarginine (pArg) residues in a concerted, two-step process involving the initial nucleophilic attack of a highly reactive cysteine (Cys7) residue onto the phosphorus atom and the formation of a pentavalent intermediate that collapses, resulting in the cleavage of the scissile N–P bond (Figure S1) (Fuhrmann et al., 2015a; Fuhrmann et al., 2013a). YwlE residue Asp118 likely promotes this reaction by stabilizing the positive charge at the scissile amine via electrostatic interactions. Subsequently, the covalent thiophosphate reaction intermediate is hydrolyzed by an activated water molecule, which is deprotonated by Asp118. Intriguingly, a Drosophila homolog of YwlE has recently been identified, although its physiological roles have yet to be deciphered (Fuhrmann et al., 2013a).
Using a ywlE mutant strain it was possible to map more than 100 pArg sites in Bacillus subtilis (Elsholz et al., 2012; Schmidt et al., 2013b). Notably, the physiological impact of this modification appears to be widespread as the modified proteins play important roles in a variety of cellular pathways including the stress response, protein degradation, cell motility, biofilm formation and competence (Elsholz et al., 2012; Schmidt et al., 2013b). McsB is also required for stress tolerance in Staphylococcus aureus (Wozniak et al., 2012) and a ywlE mutant strain exhibits reduced stress resistance in B. subtilis (Musumeci et al., 2005). Although the impact of arginine phosphorylation is only beginning to come into focus, much remains to be learned and new chemical tools to selectively enrich and/or detect the enzymes responsible for this modification are needed to accelerate research in this area. This is especially important because McsB activity is required for virulence in S. aureus (Wozniak et al., 2012), suggesting that it represents a novel therapeutic target.
Given that arginine residues are frequently involved in protein-protein, protein-small molecule, and protein-nucleic acid interactions (Fuhrmann et al., 2015a; Thompson and Fast, 2006), it is expected that arginine phosphorylation will dramatically alter these interactions by inversing the net charge from the +1 for arginine to −1 for pArg. Such an effect was recently highlighted. Specifically, arginine phosphorylation of the transcriptional repressor CtsR, which is tightly bound to specific DNA repressor elements, reduces its affinity for DNA (Fuhrmann et al., 2009).
Since cellular protein arginine phosphorylation was only detected using either a YwlE phosphatase mutant strain or the general phosphatase inhibitor pervanadate (Schmidt et al., 2013b), we hypothesized that this type of modification, and its physiological effects might be particularly sensitive to YwlE regulation. Therefore, we generated highly specific PAP/YwlE inhibitors and chemical probes that specifically recognize the active, reduced, but not the inactive, oxidized, form of YwlE. Using these probes we found that recombinant YwlE is particularly susceptible to oxidation and propose a model of YwlE regulation during cellular oxidative stress in which YwlE is inactivated by oxidation, thereby leading to enhanced protein arginine phosphorylation and induction of the stress response system. In addition to these findings, we predict that our YwlE-specific chemical probes will assist additional efforts to characterize the physiological roles of this enzyme, and help to elucidate the precise regulatory impact of protein arginine phosphorylation in diverse organisms.
Results
Generation of pArg mimetics as potent YwlE inhibitors
YwlE is a highly efficient protein arginine phosphatase that hydrolyzes the phosphoramidate (N—P) bond in pArg (Figure 1A). Given the unique chemical structure of pArg, comprising a negatively charged phosphoryl moiety attached to a positively charged guanidinium group, we set out to develop specific inhibitors accounting for both of these characteristic features. Since the phosphoramidate bond is highly acid sensitive we envisaged the generation of acid stable pArg mimetics. Initially we derivatized the basic phosphoguanidinium unit by insertion of a methylene group linking the phosphoryl to the guanidinium moiety via an acid stable C—P bond (Figure 1B). As predicted, this simple guanidinophosphonate, 1, is stable under acidic conditions and thus acts as a suitable nonhydrolyzable pArg mimetic. To test the effect of this compound on YwlE activity, we determined IC50 values. The IC50 for compound 1 is 800 ± 115 μM (Table 1), indicating that the guanidinophosphonate is indeed able to inhibit the phosphatase activity of YwlE. To test the effect of different negatively charged functionalities, we replaced the phosphonate with a sulfonate and carboxylate group and examined the effects of the respective inhibitors on YwlE activity. Interestingly, replacement of the phosphonate moiety in 1 with a sulfonate (as in 2) substantially improved the IC50 value by 25-fold (Table 1). Replacement of the phosphonate with the carboxylate, as in 3, resulted in no inhibition below 1 mM of inhibitor. The same trend was observed for the other PAP (B. subtilis YwlE) examined in this study, which only showed appreciable inhibition with the sulfonate compound 2 (Table 1). Overall, amongst the tested compounds, the guanidinosulfonate is a faithful mimetic of a pArg.
Figure 1. Inhibitors targeting the protein arginine phosphatase YwlE.
(A) YwlE catalyzed dephosphorylation of pArg. The negatively charged phosphoryl group is highlighted in red, whereas the positively charged guanidinium group is shown in blue (see also Figure S1). (B) Structure of synthesized compounds. (C) Lineweaver-Burk plot depicting the mode of inhibition for SO3-amidine on YwlE activity using different inhibitor concentrations. Data are presented as mean ± SEM. (D) Comparing the specificity of compound 9 for the inhibition of the arginine phosphatase YwlE with the closely related tyrosine phosphatase YfkJ. Data are presented as mean ± SEM (see also Figure S2 and Figure S3).
Table 1.
IC50 values of synthesized compounds and general oxidative stress inducing agents
| Name | Compound # | IC50 (μM) YwlE G. st | IC50 (μM) YwlE B. su |
|---|---|---|---|
| guanidinophosphonate | 1 | 800 ± 115 | > 1000 |
| guanidinosulfonate | 2 | 28.7 ± 1.8 | 350 ± 50 |
| guanidinoacetate | 3 | NIa | NIa |
| SO3-guanidine | 4 | 7.3 ± 0.6 | 150 ± 2.5 |
| Taurine-guanidine | 5 | 77.3 ± 1.0 | 1600 ± 220 |
| SO3-amidine | 6 | 0.79 ± 0.05 | 10.2 ± 0.38 |
| SO3-amide | 7 | 42.5 ± 2 | 990 ± 60 |
| SO3-amidine-BpA | 8 | 4.1 ± 0.4 | 53.0 ± 13.0 |
| SO3-amidine-Ahx-BpA | 9 | 0.50 ± 0.03 | 6.6 ± 0.3 |
| Rho-SO3-amidine | 10 | 0.81 ± 0.05 | 8.6 ± 0.3 |
| Bio-SO3-amidine | 11 | 0.67 ± 0.03 | 4.6 ± 0.3 |
| H2O2 | 58.8 ± 2.9 | 27.4 ± 3.8 | |
| CuSO4 | 2.8 ± 0.15 | 5.8 ± 0.1 | |
| Diamide | 4200 ± 300 | 55.6 ± 8.8 |
NI, no inhibition at concentrations less than 1 mM.
To further improve the potency of 2 we incorporated the backbone scaffold from SeCN-amidine, a recently developed YwlE inactivator (Fuhrmann et al., 2013b). Interestingly, the presence of the aliphatic spacer and backbone moieties increases the inhibitory potential of the guanidinosulfonate 4 by ~4 fold compared to 2 (IC50 of 7.3 ± 0.6 μM). To test the effect of the linker connecting the negatively charged sulfonate and the positively charged guanidinium groups, we synthesized a taurine derivative (5) that attaches the sulfonate group via a two carbon ethylene linker to the guanidinium moiety. The IC50 value obtained for compound 5 (IC50 = 77.3 ± 1.0 μM) is ~10-fold higher than the one carbon methylene linked sulfonate derivative 4. Since extension of the spacer linking the sulfonate and the guanidinium moieties did not improve inhibitor activity, we considered whether shortening the distance might yield inhibitors with enhanced potency. Thus, we substituted the guanidinium group with an amidine functionality that was linked via a one carbon methylene group to the sulfonate moiety (6). The results for compound 6 show an almost 10-fold increase in potency relative to 4 (IC50 = 0.79 ± 0.05 μM, Table 1). To test the requirement for positive charge, we substituted the positively charged amidinium with a neutral amide group (7). As expected, the amide containing compound 7 showed a substantially higher IC50 value (42.5 ± 2 μM) comprising a more than 50-fold reduction in potency compared to the amidine analog 6. We conclude that by linking the negatively charged sulfonate group to the positively charged amidinium moiety, compound 6 represents a highly suitable bioisostere of peptidyl-pArg, that faithfully mimics the size, and, in particular, the charge distribution of a phosphoguanidinium moiety. To further define the mode of inhibition of SO3-amidine, the YwlE inhibition pattern, using different inhibitor concentrations, was determined. Based on these results and after global fits of the respective enzyme inhibition curves, we obtained a Ki value of 0.39 ± 0.04 μM for SO3-amidine (Figure 1C). Furthermore, these data indicate that SO3-amidine acts as a classic competitive inhibitor.
Design and selectivity of SO3-amidine based chemical probes
To convert the SO3-amidine-containing inhibitor into a chemical probe that can covalently trap and visualize YwlE, we initially incorporated a UV-activatable benzophenone into SO3-amidine to generate compound 8. However, in contrast to the results obtained for SO3-amidine, the BpA containing analog (SO3-amidine-BpA) is a significantly less potent YwlE inhibitor (IC50 = 4.1 ± 0.4 μM). We assumed that the presence of this bulky aromatic moiety impedes proper enzyme binding. Therefore, we synthesized a derivative in which the SO3-amidine group is connected to the BpA moiety by a hexanoic acid linker. This compound (9) inhibited YwlE with comparable efficiency to SO3-amidine; the IC50 is 0.5 ± 0.03 μM, versus 0.79 ± 0.05 μM for SO3-amidine (6). Based on MS analysis we confirmed that SO3-amidine-Ahx-BpA (9) forms a covalent enzyme-inhibitor adduct after UV-treatment; the 502 Da mass shift is consistent with the attachment of a single inhibitor molecule (Figure S2).
Having demonstrated that SO3-amidine-Ahx-BpA is a highly potent YwlE inhibitor, we next wanted to examine the selectivity of this compound, and thereby assess its ability to discriminate between highly related enzymes. Such a demonstration is especially critical for the PTP family, which shares a highly conserved active site, and for which achieving inhibitor selectivity is a major challenge (Barr, 2010; He et al., 2013; Krishnan et al., 2014; Zhang, 2002). Based on IC50 values, compound 9 is >100-fold selective for YwlE when tested against its closest related paralog, the tyrosine phosphatase YfkJ (Figure 1D); YfkJ has 43% sequence similarity to YwlE (Figure S3). In addition, SO3-amidine compounds showed no inhibition of the protein arginine deiminases (PAD) 1, 2, 3 or 4, mechanistically unrelated guanidinium modifying enzymes (data not shown). These data indicate that SO3-amidine based compounds are not only potent but also highly specific protein arginine phosphatase inhibitors.
Structural analysis of inhibitor binding
To map the specific binding site of compound 9 to YwlE, we performed protein NMR spectroscopy experiments. Comparison of the 2D [1H,15N]-HSQC spectra of apo-YwlE and SO3-amidine bound YwlE reveals that inhibitor binding affects multiple residues (Figure 2A). To unambiguously assign the individual chemical shift perturbations (CSP) upon ligand binding, we obtained the complete backbone assignment using [15N,13C]-labeled YwlE bound to compound 9; assignments for apo-YwlE were reported previously (Xu et al., 2006). The residues most affected by ligand binding cluster around the active-site region indicating that SO3-amidine specifically interacts with YwlE in the substrate binding pocket (Figure 2A). Interestingly, 2D [1H,15N]-HSQC cross-peaks of most of the active site P-loop residues, that were otherwise missing in the absence of inhibitor, become stabilized and, thus, NMR peaks appear in the SO3-amidine bound YwlE structure (Figure 2B). Consistent with the NMR results, molecular docking revealed the SO3-amidine moiety of 9 invades the active site and occupies the substrate binding pocket. The sulfonate group interacts extensively with the backbone amides of the P-loop and to the side chain of Thr11 and Arg13, whereas the amidine group is pinned between Thr11, Asp118, and Phe 120 (Figure 2C). The structural positioning of these groups is in accordance with the location of the phosphoryl and guanidinium group of the pArg substrate (Fuhrmann et al., 2013a).
Figure 2. Structure of SO3-amidine-Ahx-BpA (9) bound YwlE.
(A) Left, overlay of 2D [1H,15N]-HSQC NMR spectra of [15N,13C]-labeled apo-YwlE (orange) and [15N,13C]-YwlE bound to SO3-amidine-Ahx-BpA (9) (black). Right, inhibitor induced changes in the NMR data mapped on the structure of YwlE; the substrate binding pocket is marked by a circle or line. Spheres mark residues substantially affected by inhibitor binding, including residues with large chemical shift perturbations (blue), or residues with NMR peaks that are only present in the inhibitor bound (red) or apo-structure (orange). (B) Quantitation of chemical shift perturbation (CSP) reveals residues affected by inhibitor binding. Color coding as described in Figure 2A (C) Molecular docking, using AutoDock v.4.2.5, of 9 (cyan) into YwlE (PDB code: 4KK3, stripped of bound phosphate), which is illustrated as electrostatic surface representation. The right panel highlights a close up view of the YwlE active site with the bound inhibitor (cyan) (see also Figure S4).
The solution NMR data also reveal critical mechanistic insights into enzyme function. Notably, the NMR signal corresponding to Leu117 exhibits a large chemical shift perturbation upon inhibitor binding (inset in Figure S4A). Based on the ligand free, apo-YwlE structure, the backbone nitrogen amide of Leu117 forms a hydrogen bond to the side chain carboxyl group of Glu133 (Figure S4A). This interaction is not present in the ligand bound form of YwlE due to movement of the phosphatase D-loop. Upon ligand binding, the phosphate coordinating residue Arg13 rearranges leading to closure of the catalytic D-loop; this move is accompanied by breaking the NH-O hydrogen bond between the amide of Leu117 and the side chain of Glu133 (Figure S4A). This striking D-loop movement can be monitored by the characteristic chemical shift perturbation of Leu117. Thus, observing a strong chemical shift change of Leu117 represents a characteristic signature, which can be harnessed to screen for active site binding compounds that lock the enzyme in the D-loop closed state. Furthermore, this observation implies that substrate binding directly reorients the catalytic D-loop to position the essential aspartate (Asp118) for proper substrate cleavage. The structural flexibility of the D-loop is further corroborated by comparing the individual D-loop conformations from five distinct YwlE crystal-structures and the apo-YwlE NMR structure (Figure S4B). Specifically, structural alignments confirm that the D-loop adopts different open or closed conformations depending upon proper ligand binding. These observations highlight the highly optimized and fine-tuned nature of YwlE’s catalytic mechanism.
Rho-SO3-amidine (10) is an active site directed photo-probe
Using the SO3-amidine-Ahx-BpA scaffold, we synthesized rhodamine (Figure 3A) and biotin-tagged (Figure S5A) derivatives for use as YwlE-directed chemical photo-probes. We anticipated that these probes would be useful for visualizing and isolating YwlE. The IC50 values of the rhodamine and biotin containing compounds (10 and 11) are 0.81 ± 0.05 and 0.67 ± 0.03 μM, respectively, and thus remarkably consistent with those obtained for 9 (0.50 ± 0.03 μM). These data indicate that the reporter tag does not substantially influence the interaction between these compounds and YwlE. Having shown that 10 and 11 inhibit YwlE with comparable potency to the parent compound, we then evaluated their ability to visualize YwlE by incubating them with purified recombinant YwlE (Figure 3B and Figure S5B,C). The ideal labeling time, probe concentration, and probe sensitivity/LOD are 20 min, 5 μM and ~5 pmol, respectively (Figure S5).
Figure 3. Photo-labeling of YwlE using Rho-SO3-amidine (10).
(A) Structure of Rho-SO3-amidine (10). (B) Photo-labeling of YwlE using photo-probe 10 in the presence and absence of H2O2, the competitive inhibitor SO3-amidine and the specific oxidizing agent SeCN-amidine, which induces a disulfide bridge within the active site of YwlE. The fluorescence image is on top and the coomassie stained gel (which is below the fluorescent image) shows equal protein loading levels (see also Figure S5).
To verify that 10, the rhodamine-tagged compound, is indeed an active site directed probe, we tested whether compound 6 is a competitor of the labeling reaction. There, we observed that 6 substantially suppresses labeling of YwlE (Figure 3B). Moreover, SeCN-amidine, a recently identified YwlE inhibitor that inactivates YwlE by specifically inducing disulfide bond formation (Fuhrmann et al., 2013b), also efficiently blocks YwlE labeling by 10. Based on these data, we next evaluated the impact of a common oxidant, hydrogen peroxide, on the recognition of oxidized YwlE by this photo-probe. Oxidation of YwlE by H2O2 also efficiently blocked YwlE labeling by 10 (Figure 3B).
To characterize the molecular basis of YwlE oxidation, we performed electrospray ionization-mass spectrometry (ESI-MS) analyses of YwlE treated with H2O2 (Figure 4A–C). The spectra revealed that H202 treatment induces the formation of a disulfide bridge; indicated by a loss of 2 Dalton in YwlE (Figure 4B). The intact mass analyses further highlight the complete conversion of reduced YwlE into disulfide bridged, oxidized, YwlE. Interestingly, there are no higher oxidations such as sulfenic (+16 Da), sulfinic (+32 Da) or sulfonic acid (+48 Da) detectable, indicating that disulfide bridge formation is a highly efficient process that might have evolved as a safeguard to prevent irreversible hyperoxidation of the enzyme. We further investigated the reversibility of YwlE oxidation by ESI-MS and activity assays. Indeed, as shown in Figure 4C, the intramolecular disulfide-bridge in YwlE is readily reduced by DTT yielding the same molecular mass as non-oxidized YwlE. In addition, inactivation of YwlE by various oxidative stress inducing agents can be reversed upon addition of DTT (Figure 4D). Notably, YwlE only contains two cysteine residues, the active site nucleophile, Cys7, (Figure S1) and the ‘backdoor’ cysteine, Cys12, which are both located in the active site pocket. Moreover, recent mutational analyses in YwlE from G. stearothermophilus confirmed the critical role of these cysteine residues in the catalytic mechanism and their sensitivity to oxidation (Fuhrmann et al., 2013b). Based on a structural model of YwlE containing an active site disulfide bridge, we predict that the oxidized form has a closed P-loop that prevents access of substrates, as well as the SO3-amidine probe, to the active site pocket (Figure 4E). These observations confirmed our hypothesis that the sulfonate-amidine based compounds are selective active site directed probes that only bind and react with the active, reduced form of YwlE.
Figure 4. Functional and structural effects of YwlE oxidation.
(A–C) Mass spectrometric analyses of YwlE (A), YwlE treated with H2O2 (B) and preoxidized YwlE incubated with DTT (C). The figure panel illustrates the ESI-MS spectra including the mass and the charge status of the distinct YwlE species as well as the deconvoluted intact mass of the respective protein samples (theoretical mass of YwlE B.su: 17,850.3 Da). (D) Residual activity of YwlE after treatment with the indicated oxidative stress inducing agents. The activity was also tested after dialysis followed by treatment with the reducing agent DTT. Data are presented as mean ± SEM (n = 3). (E) The image depicts the active site architecture of YwlE bound to PO4 (left side; PDB code: 4KK3), and a model structure of YwlE containing a disulfide (S-S) bond (right side). The PDB code of YwlE bound to PO4 is 4KK3. The disulfide linked YwlE model was generated based on the apo-YwlE structure (PDB code: 1ZGG), where the two cysteine thiol groups are only 4.1 A apart. After manual cysteine side chain rotation, the model was refined by applying geometry restraints including bond distances, angles, planes and non-bonded contacts using COOT v.0.8.2 (Emsley et al., 2010).
Evaluation of Bio-SO3-amidine (11)
Although, 10 efficiently and selectively detects recombinant YwlE, we were not able to visualize endogenous YwlE in bacterial cell lysates, likely due to the low levels of cellular YwlE. Consistent with this hypothesis is the fact that YwlE is barely detectable by Western blot analyses (see Figure 5 below). As such, we next determined whether the biotinylated version, 11 (Figure S5A), could be used to selectively enrich YwlE. To first validate its utility for this purpose, we set out to further characterize its affinity for YwlE. These studies are important because such pulldown assays require strong interactions between the target enzyme and the chemical probe. Therefore, we investigated the thermodynamic signature of binding between YwlE and 11 by isothermal titration calorimetry (ITC) (Figure S6A). Consistent with the labeling studies, compound 11 only binds to active YwlE; we observed no detectable binding to the oxidized form of the enzyme (Figure S6A). The dissociation constant, KD, was 0.54 μM, which is consistent with its IC50 value (0.67 ± 0.03 μM, Table 1). The binding of 11 to YwlE is mainly driven by a favorable enthalpic (ΔH = −9.96 kcal/mol) contribution, resulting in an overall Gibbs free energy of binding (ΔG) of −8.55 kcal/mol. Having shown that 11 binds tightly to active YwlE, we next evaluated its ability to enrich recombinant YwlE in a pull down assay. For these experiments, avidin bound Bio-SO3-amidine (11) was incubated with recombinant YwlE, the beads washed, and bound proteins eluted by boiling in SDS-PAGE loading buffer. The results of these experiments indicate that 11 efficiently traps recombinant YwlE from both B. subtilis and G. stearothermophilus (Figure S6B). To demonstrate its suitability for isolating endogenous YwlE, 11 was incubated with G. stearothermophilus cell extracts and samples processed as above. As can be seen in Figure S6C, 11 successfully enriches for cellular YwlE. The identity of the detected protein band was verified by MS analysis (data not shown). In total, these data demonstrate that Bio-SO3-amidine can be used to not only enrich recombinant YwlE but also endogenous YwlE from bacterial cell extracts.
Figure 5. Regulation of cellular YwlE in B. subtilis cells.
(A) Pulldown of the active form of YwlE using 11 (upper panel) from differentially stressed B. subtilis cells and immunoprecipitation of endogenous YwlE (lower panel) from the respective cell samples (see also Figure S6). (B) Influence of various inducers of oxidative stress on the arginine phosphorylation activity of McsB in the presence of YwlE. Arginine phosphorylated McsB was monitored using anti-pArg antibody by dot blot assay (upper panel). As a control the same blot was stripped and reprobed with anti-McsB antibody (lower panel). (C) Transcriptional analysis of CtsR repressed clp gene expression in wt versus ΔmcsB B. subtilis cells analyzed by qRT-PCR. The control samples were isolated from cells grown under non-stress conditions. Data from three independent experiments are presented as mean ± SEM (*P < 0.05, **P < 0.01). (D) Model of YwlE mediated regulation of arginine phosphorylation. YwlE acts as an arginine phosphatase to reverse the kinase activity of the protein arginine kinase McsB. Under oxidative stress, YwlE is inactivated, thereby leading to an increased level of arginine phosphorylation that is important for the cellular stress response.
YwlE is inactivated during cellular oxidative stress in B. subtilis
YwlE efficiently catalyzes the dephosphorylation of substrates phosphorylated by McsB, as exemplified by the fact that protein arginine phosphorylation is only robustly detected in a ΔywlE B. subtilis strain but not in wildtype strains (Elsholz et al., 2012). Recently, we showed that arginine phosphorylation occurs in B. subtilis wildtype cells that were stressed with diamide (Fuhrmann et al., 2015b). However, the underlying molecular mechanism of the induction of arginine phosphorylation is not known. Given that the activity of YwlE is the dominant mechanism for controlling arginine phosphorylation levels in vivo, mechanisms to regulate its activity must exist. Based on our observation that YwlE is inhibited by H2O2 and other oxidants, we next tested whether endogenous YwlE activity is controlled by oxidative stress. Specifically, we used compound 11 as an active site directed affinity probe to enrich endogenous active YwlE and thereby track its response to oxidative stress. To this end, we treated B. subtilis cells with various oxidative stress inducing agents (i.e., diamide, H2O2, CuSO4) and analyzed their impact on our ability to isolate cellular YwlE. Based on our previous data (see above), we expected that oxidation of YwlE would block its isolation by 11. Consistent with this hypothesis, YwlE was not detected in the samples incubated with oxidants, whereas the level of total YwlE did not change significantly as shown by immunoprecipitation and Western blotting of YwlE (Figure 5A). These data indicate that endogenous YwlE is inactivated under cellular oxidative stress.
To evaluate the impact of oxidative stress on arginine phosphorylation, we reconstituted the McsB/YwlE couple in vitro and evaluated the impact of different oxidative stress conditions. Using a pArg specific antibody (Fuhrmann et al., 2015b), we observed that YwlE efficiently dephosphorylates McsB (Figure 5B). However, arginine phosphorylation of McsB can be detected under oxidative stress, indicating that YwlE is inactivated while McsB is active during these conditions. Overall, we conclude that YwlE is inactivated by oxidative stress, increasing arginine phosphorylation, which drives the oxidative stress response by activating downstream signaling events.
To extend this hypothesis into cells, we examined the expression of CtsR regulated genes by qRT-PCR. Since McsB is known to phosphorylate and inhibit CtsR (Fuhrmann et al., 2009), we hypothesized that oxidative stress induces gene expression by inactivating YwlE and thereby blocking dephosphorylation of McsB mediated arginine phosphorylation. To test the involvement of protein arginine phosphorylation on CtsR regulation during oxidative stress, we analyzed the transcripts of CtsR regulated genes by comparing the relative expression levels in wildtype versus ΔmcsB B. subtilis strains. The CtsR mediated repression of clpP and clpE gene expression is upregulated under oxidative stress condition in the wildtype strain, whereas the expression levels of these genes are virtually unaffected in a ΔmcsB strain (Figure 5C). In control experiments using non-stressed cells, we did not observe any significant difference in expression between wildtype and mutant ΔmcsB strains. Based on these observations, we propose the following model of YwlE regulation and hence arginine phosphorylation under oxidative stress conditions (Figure 5D). YwlE is inactivated during oxidative stress thereby releasing the phosphatase mediated suppression of protein arginine phosphorylation. Consequently McsB induces multiple downstream pathways implicated in the cellular stress response (Elsholz et al., 2012), including the inhibition of CtsR (Fuhrmann et al., 2009), the major transcriptional inhibitor of bacterial stress response genes.
Discussion
The present study identifies novel and highly potent arginine phosphatase inhibitors/probes. Based on rational, substrate inspired, inhibitor design we generated a panel of diverse nonhydrolyzable pArg analogs and identified SO3-amidine containing compounds as efficient competitive inhibitors of the arginine phosphatase YwlE. These SO3-amidine-containing compounds represent excellent substrate analogs for protein arginine phosphatases with IC50 values in the nanomolar range and high selectivity. This high specificity is particularly remarkable since a similar, recently developed YwlE inhibitor, SeCN-amidine, also targets the closely related tyrosine phosphatase YfkJ (Fuhrmann et al., 2013b). Structure-activity relationships yield a detailed model of an enzyme-inhibitor binding event that is stabilized by extensive polar interactions. The successful conversion of the inhibitor into an active site directed chemical probe further highlights the versatility of these compounds and sheds new insights onto the molecular mechanism of YwlE mediated substrate hydrolysis. In this respect, NMR measurements and structural analyses of distinct YwlE crystal structures reveal a striking mechanism of D-loop stabilization in the open state via the formation of a hydrogen bond between the nitrogen amide of Leu117 and the carboxyl side chain group of Glu133 (Figure S4). Interestingly, the human LMW-PTP HCPTPB contains a conserved glutamine (Gln144) at the corresponding position of YwlE’s Glu133, suggesting that a similar mechanism of D-loop regulation might take place in the human enzyme and other LMW-PTPs.
Since the SO3-amidine based chemical probe only recognizes active YwlE, we utilized this compound in cellular studies as an affinity based profiling reagent and showed that oxidative stress inactivates YwlE in vivo. These data were further validated by detailed in vitro analyses using reconstituted recombinant enzymes. There, we observed that YwlE can be readily oxidized and forms a disulfide bridge between both active site cysteine residues. Notably, the oxidation of YwlE is freely reversible upon treatment with a reducing agent. Therefore, we predict that YwlE is a redox sensor whose activity is adjusted according to the cellular redox levels. Since the disulfide linked, inactive, YwlE is efficiently converted into its reduced, active, form there is no need to degrade and resynthesize the enzyme. This would allow for a more efficient and rapid response once the stress diminishes. Taken together, we propose a model wherein the control of YwlE activity and consequently arginine phosphorylation is a crucial step in the regulation of the cellular stress response during oxidative stress conditions, where YwlE acts as a redox sensor that tunes the level of protein arginine phosphorylation. In this respect, we predict that a plethora of McsB targets, shown to be arginine phosphorylated in vivo (Elsholz et al., 2012; Schmidt et al., 2013b; Trentini et al., 2014), are likely to be modulated upon YwlE inactivation. Even though YwlE efficiently reverses/blocks McsB mediated arginine phosphorylation, preliminary data indicate that YwlE does not form a stable complex with McsB and has minimal effect on the McsB ATPase activity (data not shown). Therefore, we predict that YwlE counteracts the McsB activity by catalyzing the reverse reaction and does not directly inhibit McsB by blocking its activity. Potential alternative mechanisms for how YwlE inactivation might lead to increased arginine phosphorylation levels include the notion that autophosphorylation of McsB leads to its activation. This potential kinase activation mechanism would be blocked by YwlE dephosphorylation. Alternatively, a combination of both of these effects might occur where YwlE counteracts/reverses McsB mediated arginine phosphorylation and also, additionally, diminishes its activation and/or alters its substrate specificity.
Protein arginine phosphorylation was previously shown to be responsible for the coordination and the onset of distinct stress response systems in B. subtilis by the phosphorylation of the central stress response regulators CtsR and HrcA (Schmidt et al., 2013b). Moreover, it should be noted that the protein arginine kinase McsB was shown to be a crucial component of stress and pathogenicity mechanisms including biofilm formation in S. aureus (Wozniak et al., 2012), indicating that modulation of protein arginine phosphorylation might represent an attractive target to intervene in bacterial infections.
In future studies, it will be interesting to evaluate the relationship between YwlE activity and protein arginine phosphorylation during different stress conditions. In this respect, it was shown that the expression level of the McsB kinase is increased under heat stress via inactivation of the transcriptional respressor CtsR (Kirstein et al., 2005; Kruger et al., 2001). Moreover, it was demonstrated that the McsB kinase activity is also regulated by McsA and the chaperone protein ClpC (Kirstein et al., 2005). McsA is supposed to activate the kinase activity of McsB, whereas ClpC acts as an inhibitor of McsB mediated phosphorylation. Besides its kinase inhibiting activity, ClpC is also involved in the degradation of autophosphorylated McsB via the ClpCP protease complex (Kirstein et al., 2007). In order to integrate and sense diverse stress conditions, it has been proposed that McsB activity can be regulated via multiple ways involving YwlE, ClpC and McsA (Elsholz et al., 2011). This multi-level regulatory system highlights the central role and the physiological relevance of arginine phosphorylation to coordinate a proper stress response in gram-positive bacteria.
Beyond the intriguing biological implications, our data show that SO3-amidine represents a stable bioisostere for the acid labile phosphoguanidinium moiety in pArg. This nonhydrolyzable analog of pArg provides a versatile tool, which will facilitate functional studies to investigate the role of this elusive PTM in other organisms. SO3-amidine also constitutes a useful compound to screen for protein arginine phosphatase or pArg binding proteins in diverse biological samples, including humans, and represents a suitable starting point to generate highly specific arginine phosphatase inhibitors.
Significance
The protein arginine phosphatase YwlE belongs to the LMW-PTP family and was shown to be important for cellular stress responses. In the present study we present the synthesis and use of the first highly specific inhibitors/probes targeting this enzyme class. These findings are particularly important because the design of selective inhibitors targeting individual LMW-PTPs family members is largely hampered by the absence of structural differences within this protein family. However, for YwlE, we could exploit characteristic structural features of the enzyme and its preference for pArg to develop highly specific and potent inhibitors and probes. These chemical tools were used to analyze the physiological regulation of arginine phosphorylation and revealed that the oxidative inactivation of YwlE induces an increase in protein arginine phosphorylation and concomitant induction of stress response genes. Together, this work sheds new light on the biological regulation of protein arginine phosphorylation and provides novel tools to facilitate the study of this elusive protein modification in other organisms.
Experimental Procedures
A detailed description of experimental procedures is available online in the supplemental information.
Chemical synthesis
The detailed chemical syntheses and characterizations of individual compounds are described in the accompanying supplemental information.
The phosphoguanidinium mimicking derivatives 1, 2 and 3 were synthesized using a solution phase procedure starting from N, N′-di-Boc-protected pyrazole-1-carboxamidine. This compound was reacted with either aminomethanephosphonate, aminomethanelsulfonate or glycine to yield distinct guanidino acids.
The guanidine derivatives were synthesized using a solid phase procedure. Ornithine was coupled on a MBHA solid support resin. After addition of a benzoyl group to the α-amine, the ornithine side chain was reacted with Fmoc-NCS to form a thiourea derivative that was ultimately coupled with different aminosulfonates forming distinct arginine sulfonates, i.e. compounds 4 and 5.
Compounds 6 to 11 were generated in a similar manner by solid phase syntheses procedures. However, the SO3-amidine head group of compounds 6 and 8–11 were installed by reaction of the ornithine δ-amine with chloroacetimidate yielding a chloroacetamidine derivative. This compound was further reacted with sulfite to generate the SO3-amidine group.
The SO3-amide functionality of compound 7 was synthesized by reaction of the ornithine δ-amine with chloroacetylchloride to generate a chloroacetamide derivative. The sulfonate group was installed by treatment of the chloroacetamide derivative with sulfite to yield the final SO3-amide group.
Protein expression and purification
The expression and purification of His-YwlE derived from B. subtilis (YwlE B.su), His-YwlE derived from G. stearothermophilus (YwlE G.st) and His-YfkJ from B. subtilis is described in detail in the supplemental information. In brief, E. coli BL21(DE3) cells were transformed with the individual plasmid and grown in LB medium supplemented with ampicillin. Protein expression was induced when cells reached an OD600 of ~0.8 with 0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) and overexpression was performed for 2 h at 37 °C. After cell lysis by sonication, the cleared lysate was loaded on a Ni-NTA column and the bound His-tagged protein was eluted by applying a step-wise imidazole gradient. Thereafter, the eluted protein was further purified using size exclusion chromatography as described in the supplemental information.
Phosphatase activity and inhibition assay
The phosphatase assay was carried out at 25 °C in 96-well microplates using 20 mM Tris/HCl pH 7.6 buffer, containing para-nitrophenylphosphate (pNPP) as the substrate. The phosphatase reaction was monitored in 20 second intervals at 405 nm using a SpectraMax M5 (Molecular Devices) microplate reader. Reaction times, amounts of enzyme, and concentrations of substrates were optimized to have linear kinetics. The inhibitory properties of the generated compounds were tested by serial dilutions of the inhibitors. The resulting progress curves were further analyzed by plotting the percentage of enzyme inhibition versus inhibitor concentrations to obtain the compounds IC50 values. A more detailed description of the assay and IC50 calculations is provided in the supplemental information.
Isothermal titration calorimetry (ITC)
Thermodynamic values for the interaction between YwlE and 11 were determined by ITC. ITC experiments were recorded on an ITC200 (MicroCal) instrument at 25 °C. Protein and inhibitor sample were resuspended in 10 mM Tris/HCl pH 7.6, 100 mM NaCl. Sequential injections of 2 μL of compound 11 (600 μM) were dispensed into the sample cell containing 350 μL YwlE (30 μM) solution. The resulting thermodynamic data were analyzed using the program Origin.
YwlE NMR experiments
[15N,13C]-labeled YwlE was expressed in E. coli BL21(DE3) cells grown in minimal medium containing 13C-glucose and 15NH4Cl as described in the supplemental information. The protein purification procedure was identical to the unlabeled protein (see above). Prior to the NMR experiment, the protein sample buffer was exchanged to 50 mM Tris/HCl pH 7.5 containing 50 mM NaCl and 3 mM DTT. NMR data were recorded at 298 K on a Bruker 700 MHz spectrometer equipped with a QCI cryoprobe. Two-dimensional [1H,15N]-HSQC data were collected on the apo and ligand-bound [15N,13C]-labeled YwlE. To allow unambiguous chemical shift assignment for the inhibitor (9) bound YwlE structure, we collected three-dimensional NMR data including HNCO, HNCA, HNCACB, and CBCA(CO)NH to obtain the backbone resonance assignments. The inhibitor binding site was determined using the chemical shift changes from two-dimensional [1H,15N]-HSQC data. For further details, please refer to the supplemental information.
In-gel fluorescence assay using photo-probe 10
Typically, YwlE (2 μM) was labeled with photo-probe 10 (5 μM) for 10 min and then incubated for 20 min on ice under constant UV irradiation at 365 nm. After UV irradiation, the reactions were quenched by boiling with 6x SDS-PAGE loading buffer and electrophoresed on a 15% SDSPAGE gel. The gel was then imaged on a Typhoon 9410 (GE Healthcare) set at ex. 532 nm, em. 580 nm.
Enrichment of endogenous YwlE with compound 11
Bio-SO3-amidine (11) beads were prepared by incubating 5 μM Bio-SO3-amidine (11) with 50 μL neutravidin agarose beads (Pierce) for 30 min followed by extensive washing as described in the supplemental information. B. subtilis cells were then treated with 500 μM CuSO4, 500 μM H2O2 or 1 mM diamide for 20 min to induce oxidative stress. Cells were harvested by centrifugation at 8600 g at 4 °C, washed and resuspended in TBS buffer and lysed by sonication. The cleared cell lysate was split in two. One half was incubated with the Bio-SO3-amidine beads, the other one used for the YwlE immunoprecipitation experiment (see below). After extensive washing, the bound YwlE protein was eluted with 6x SDS-PAGE loading dye followed by 15% SDS-PAGE separation. YwlE was detected by Western blot analysis as detailed in the supplemental information.
Immunoprecipitation of endogenous YwlE
Immunoprecipitation experiments were performed using anti-YwlE antibody bound to proteinA/G Plus-agarose beads (Santa Cruz Biotechnology). The generated B. subtilis cell extracts (see above) were incubated with the anti-YwlE antibody protein A/G bead slurry. Following extensive washing, the bound YwlE protein was eluted with SDS-PAGE loading buffer and detected by Western blot analysis as described in the supplemental information.
McsB arginine phosphorylation assay
The McsB (2 μM) arginine autophosphorylation reaction was performed in 20 mM Tris/HCl pH 8.0, 50 mM NaCl, 20 mM MgCl2 and 2 mM ATP. To test the influence of YwlE and YwlE inhibition on McsB arginine phosphorylation, 0.5 μM YwlE was added in the absence or presence of CuSO4, H2O2 or diamide. Arginine phosphorylation was monitored by dot blot analysis using the recently generated anti-pArg antibody (Fuhrmann et al., 2015b). Protein samples were directly spotted onto the nitrocellulose membrane and tested for the presence of pArg as described in the supplemental information.
RNA isolation and quantitative RT-PCR
To test the effect of oxidative stress on clp gene expression, RNA from stressed vs. non-stressed B. subtilis strains was isolated using RNeasy-Mini RNA isolation kit (Qiagen). RNA was transcribed into cDNA by omniscript and sensiscript reverse transcriptase, the cDNA was amplified by HotStarTaq DNA polymerase and the resulting gene specific amplification products were analyzed in real time using SYBR Green. The amplified products were normalized to the cellular rpoB and 16SrRNA. The oligonucleotide sequences used for the qRT-PCR experiments are reported in the supplemental information.
Supplementary Material
Highlights.
Generation of SO3-amidine based protein arginine phosphatase inhibitors
Development of an activity based affinity probe (ABAP) for arginine phosphatases
The ABAP probes reveal that YwlE activity is negatively regulated by oxidative stress
Inhibition of YwlE activity increases McsB mediated arginine phosphorylation
Acknowledgments
We thank Dr. Kong Xiangming for help with the NMR data collection. This work was supported in part by NIH grant GM079357 (to P.R.T.), and by the Austrian Science Fund (FWF): J 3548-B21 (to J.F.).
Abbreviations
- LMW-PTP
low molecular weight-protein tyrosine phosphatase
- BpA
p-benzoylphenylalanine
- NMR
nuclear magnetic resonance
- CSP
chemical shift perturbations
- ITC
isothermal titration calorimetry
- ESI-MS
electrospray ionization-mass spectrometry
- qRT-PCR
quantitative reverse transcription-PCR
- PAP
protein arginine phosphatase
- pArg
phosphoarginine
Footnotes
Author Contributions
J.F. designed experiments, performed the experiments, analyzed the data, and wrote the manuscript, V.S. helped with research design, D.K. analyzed the protein NMR data, P.R.T. designed experiments, analyzed the data, and wrote the manuscript
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- Attwood PV, Piggott MJ, Zu XL, Besant PG. Focus on phosphohistidine. Amino Acids. 2007;32:145–156. doi: 10.1007/s00726-006-0443-6. [DOI] [PubMed] [Google Scholar]
- Barr AJ. Protein tyrosine phosphatases as drug targets: strategies and challenges of inhibitor development. Future Med Chem. 2010;2:1563–1576. doi: 10.4155/fmc.10.241. [DOI] [PubMed] [Google Scholar]
- Besant PG, Attwood PV, Piggott MJ. Focus on phosphoarginine and phospholysine. Curr Protein Pept Sci. 2009;10:536–550. doi: 10.2174/138920309789630598. [DOI] [PubMed] [Google Scholar]
- Elsholz AK, Hempel K, Michalik S, Gronau K, Becher D, Hecker M, Gerth U. Activity control of the ClpC adaptor McsB in Bacillus subtilis. J Bacteriol. 2011;193:3887–3893. doi: 10.1128/JB.00079-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Elsholz AK, Turgay K, Michalik S, Hessling B, Gronau K, Oertel D, Mader U, Bernhardt J, Becher D, Hecker M, Gerth U. Global impact of protein arginine phosphorylation on the physiology of Bacillus subtilis. Proc Natl Acad Sci U S A. 2012;109:7451–7456. doi: 10.1073/pnas.1117483109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. Acta Crystallogr, Sect D: Biol Crystallogr. 2010;66:486–501. doi: 10.1107/S0907444910007493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Engholm-Keller K, Larsen MR. Technologies and challenges in large-scale phosphoproteomics. Proteomics. 2013;13:910–931. doi: 10.1002/pmic.201200484. [DOI] [PubMed] [Google Scholar]
- Fuhrmann J, Clancy KW, Thompson PR. Chemical biology of protein arginine modifications in epigenetic regulation. Chem Rev (Washington, DC, U S) 2015a;115:5413–5461. doi: 10.1021/acs.chemrev.5b00003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fuhrmann J, Mierzwa B, Trentini DB, Spiess S, Lehner A, Charpentier E, Clausen T. Structural basis for recognizing phosphoarginine and evolving residue-specific protein phosphatases in gram-positive bacteria. Cell Rep. 2013a;3:1832–1839. doi: 10.1016/j.celrep.2013.05.023. [DOI] [PubMed] [Google Scholar]
- Fuhrmann J, Schmidt A, Spiess S, Lehner A, Turgay K, Mechtler K, Charpentier E, Clausen T. McsB is a protein arginine kinase that phosphorylates and inhibits the heat-shock regulator CtsR. Science. 2009;324:1323–1327. doi: 10.1126/science.1170088. [DOI] [PubMed] [Google Scholar]
- Fuhrmann J, Subramanian V, Thompson PR. Targeting the Arginine Phosphatase YwlE with a Catalytic Redox-Based Inhibitor. ACS Chem Biol. 2013b;8:2024–2032. doi: 10.1021/cb4001469. [DOI] [PubMed] [Google Scholar]
- Fuhrmann J, Subramanian V, Thompson PR. Synthesis and Use of a Phosphonate Amidine to Generate an Anti-Phosphoarginine-Specific Antibody. Angew Chem, Int Ed Engl. 2015b;54:14715–14718. doi: 10.1002/anie.201506737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- He R, Zeng LF, He Y, Zhang S, Zhang ZY. Small molecule tools for functional interrogation of protein tyrosine phosphatases. FEBS J. 2013;280:731–750. doi: 10.1111/j.1742-4658.2012.08718.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jung K, Jung H. A new mechanism of phosphoregulation in signal transduction pathways. Sci Signaling. 2009;2:pe71. doi: 10.1126/scisignal.296pe71. [DOI] [PubMed] [Google Scholar]
- Kee JM, Muir TW. Chasing phosphohistidine, an elusive sibling in the phosphoamino acid family. ACS Chem Biol. 2012;7:44–51. doi: 10.1021/cb200445w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kirstein J, Dougan DA, Gerth U, Hecker M, Turgay K. The tyrosine kinase McsB is a regulated adaptor protein for ClpCP. EMBO J. 2007;26:2061–2070. doi: 10.1038/sj.emboj.7601655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kirstein J, Zuhlke D, Gerth U, Turgay K, Hecker M. A tyrosine kinase and its activator control the activity of the CtsR heat shock repressor in B. subtilis. EMBO J. 2005;24:3435–3445. doi: 10.1038/sj.emboj.7600780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Klumpp S, Krieglstein J. Phosphorylation and dephosphorylation of histidine residues in proteins. Eur J Biochem. 2002;269:1067–1071. doi: 10.1046/j.1432-1033.2002.02755.x. [DOI] [PubMed] [Google Scholar]
- Kowalewska K, Stefanowicz P, Ruman T, Fraczyk T, Rode W, Szewczuk Z. Electron capture dissociation mass spectrometric analysis of lysine-phosphorylated peptides. Biosci Rep. 2010;30:433–443. doi: 10.1042/BSR20090167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Krishnan N, Koveal D, Miller DH, Xue B, Akshinthala SD, Kragelj J, Jensen MR, Gauss CM, Page R, Blackledge M, Muthuswamy SK, Peti W, Tonks NK. Targeting the disordered C terminus of PTP1B with an allosteric inhibitor. Nat Chem Biol. 2014;10:558–566. doi: 10.1038/nchembio.1528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kruger E, Zuhlke D, Witt E, Ludwig H, Hecker M. Clp-mediated proteolysis in Gram-positive bacteria is autoregulated by the stability of a repressor. EMBO J. 2001;20:852–863. doi: 10.1093/emboj/20.4.852. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Matthews HR. Protein kinases and phosphatases that act on histidine, lysine, or arginine residues in eukaryotic proteins: a possible regulator of the mitogen-activated protein kinase cascade. Pharmacol Ther. 1995;67:323–350. doi: 10.1016/0163-7258(95)00020-8. [DOI] [PubMed] [Google Scholar]
- Musumeci L, Bongiorni C, Tautz L, Edwards RA, Osterman A, Perego M, Mustelin T, Bottini N. Low-molecular-weight protein tyrosine phosphatases of Bacillus subtilis. J Bacteriol. 2005;187:4945–4956. doi: 10.1128/JB.187.14.4945-4956.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ramponi G, Stefani M. Structural, catalytic, and functional properties of low M(r), phosphotyrosine protein phosphatases. Evidence of a long evolutionary history. Int J Biochem Cell Biol. 1997;29:279–292. doi: 10.1016/s1357-2725(96)00109-4. [DOI] [PubMed] [Google Scholar]
- Schmidt A, Ammerer G, Mechtler K. Studying the fragmentation behavior of peptides with arginine phosphorylation and its influence on phospho-site localization. Proteomics. 2013a;13:945–954. doi: 10.1002/pmic.201200240. [DOI] [PubMed] [Google Scholar]
- Schmidt A, Trentini DB, Spiess S, Fuhrmann J, Ammerer G, Mechtler K, Clausen T. Quantitative phosphoproteomics reveals the role of protein arginine phosphorylation in the bacterial stress response. Mol Cell Proteomics. 2013b doi: 10.1074/mcp.M113.032292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Suzuki T, Soga S, Inoue M, Uda K. Characterization and origin of bacterial arginine kinases. Int J Biol Macromol. 2013;57:273–277. doi: 10.1016/j.ijbiomac.2013.02.023. [DOI] [PubMed] [Google Scholar]
- Thompson PR, Fast W. Histone citrullination by protein arginine deiminase: is arginine methylation a green light or a roadblock? ACS Chem Biol. 2006;1:433–441. doi: 10.1021/cb6002306. [DOI] [PubMed] [Google Scholar]
- Trentini DB, Fuhrmann J, Mechtler K, Clausen T. Chasing Phosphoarginine Proteins: Development of a Selective Enrichment Method Using a Phosphatase Trap. Mol Cell Proteomics. 2014;13:1953–1964. doi: 10.1074/mcp.O113.035790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wozniak DJ, Tiwari KB, Soufan R, Jayaswal RK. The mcsB gene of the clpC operon is required for stress tolerance and virulence in Staphylococcus aureus. Microbiology. 2012;158:2568–2576. doi: 10.1099/mic.0.060749-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Xu H, Xia B, Jin C. Solution structure of a low-molecular-weight protein tyrosine phosphatase from Bacillus subtilis. J Bacteriol. 2006;188:1509–1517. doi: 10.1128/JB.188.4.1509-1517.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhang ZY. Protein tyrosine phosphatases: structure and function, substrate specificity, and inhibitor development. Annu Rev Pharmacol Toxicol. 2002;42:209–234. doi: 10.1146/annurev.pharmtox.42.083001.144616. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.





