#PAGE_PARAMS# #ADS_HEAD_SCRIPTS# #MICRODATA#

The Ps and Qs of alarmone synthesis in Staphylococcus aureus


Authors: Ning Yang aff001;  Shujie Xie aff001;  Nga-Yeung Tang aff001;  Mei Yee Choi aff001;  Ying Wang aff002;  Rory M. Watt aff001
Authors place of work: Faculty of Dentistry, The University of Hong Kong, Pokfulam, Hong Kong SAR, China aff001;  Department of Virology, Wuhan Centers for Disease Prevention and Control, Wuhan, Hubei, China aff002
Published in the journal: PLoS ONE 14(10)
Category: Research Article
doi: https://doi.org/10.1371/journal.pone.0213630

Summary

During the stringent response, bacteria synthesize guanosine-3’,5’-bis(diphosphate) (ppGpp) and guanosine-5’-triphosphate 3’-diphosphate (pppGpp), which act as secondary messengers to promote cellular survival and adaptation. (p)ppGpp ‘alarmones’ are synthesized and/or hydrolyzed by proteins belonging to the RelA/SpoT Homologue (RSH) family. Many bacteria also encode ‘small alarmone synthetase’ (SAS) proteins (e.g. RelP, RelQ) which may also be capable of synthesizing a third alarmone: guanosine-5’-phosphate 3’-diphosphate (pGpp). Here, we report the biochemical properties of the Rel (RSH), RelP and RelQ proteins from Staphylococcus aureus (Sa-Rel, Sa-RelP, Sa-RelQ, respectively). Sa-Rel synthesized pppGpp more efficiently than ppGpp, but lacked the ability to produce pGpp. Sa-Rel efficiently hydrolyzed all three alarmones in a Mn(II) ion-dependent manner. The removal of the C-terminal regulatory domain of Sa-Rel increased its rate of (p)ppGpp synthesis ca. 10-fold, but had negligible effects on its rate of (pp)pGpp hydrolysis. Sa-RelP and Sa-RelQ efficiently synthesized pGpp in addition to pppGpp and ppGpp. The alarmone-synthesizing abilities of Sa-RelQ, but not Sa-RelP, were allosterically-stimulated by the addition of pppGpp, ppGpp or pGpp. The respective (pp)pGpp-synthesizing activities of Sa-RelP/Sa-RelQ were compared and contrasted with SAS homologues from Enterococcus faecalis (Ef-RelQ) and Streptococcus mutans (Sm-RelQ, Sm-RelP). Results indicated that EF-RelQ, Sm-RelQ and Sa-RelQ were functionally equivalent; but exhibited considerable variations in their respective biochemical properties, and the degrees to which alarmones and single-stranded RNA molecules allosterically modulated their respective alarmone-synthesizing activities. The respective (pp)pGpp-synthesizing capabilities of Sa-RelP and Sm-RelP proteins were inhibited by pGpp, ppGpp and pppGpp. Our results support the premise that RelP and RelQ proteins may synthesize pGpp in addition to (p)ppGpp within S. aureus and other Gram-positive bacterial species.

Keywords:

Staphylococcus aureus – Oligomers – Recombinant proteins – Streptococcus mutans – Enterococcus faecalis – Hydrolysis – RNA synthesis – Enzyme kinetics

Introduction

The stringent response is a coordinated, multifaceted physiological response that bacterial cells initiate in response to encountering certain adverse extracellular conditions, such as nutrient deprivation. During the stringent response, bacteria produce high levels of two phosphorylated guanosine nucleotides: guanosine-3’,5’-bis(diphosphate) (ppGpp) and guanosine-5’-triphosphate 3’-diphosphate (pppGpp); collectively referred to as (p)ppGpp, ‘magic spots I and II’ or ‘alarmones’ [13]. These two (p)ppGpp alarmones act as intracellular messengers that directly or indirectly up-regulate cellular processes that promote the conservation, recycling and biosynthesis of important molecular building-blocks; initiate physiological mechanisms that preserve cellular integrity and tolerance in response to certain cytotoxic exogenous agents or harsh extracellular conditions; and down-regulate processes that are superfluous or non-essential for short-term viability (reviewed in [412]).

Alarmones are synthesized and/or hydrolyzed by proteins belonging to the RelA/SpoT-Homologue (RSH) family [13, 14]. RelA proteins, such as the Escherichia coli RelA protein, are termed monofunctional or ‘synthase-only’, in that they contain an active alarmone synthesis (SYNTH) domain, but have a catalytically-inactive form of the Histidine-Aspartate (HD) domain responsible for hydrolytic activities, and thus only catalyze (p)ppGpp production [15, 16]. This synthase activity involves the transfer of a diphosphate unit from ATP to the 3’-hydroxyl group of GTP or GDP in a Mg2+-dependent manner to produce pppGpp or ppGpp, respectively, with the concomitant release of AMP [17]. Rel (‘long-Rel’) and SpoT protein homologues are termed bifunctional or ‘synthase-hydrolase’, as they also possess the ability to hydrolyze (p)ppGpp molecules to GTP/GDP + diphosphate, via hydrolytic cleavage of the phosphoester bond at the ribose-3’ position [14, 18, 19, 20]. Rel proteins contain effective, yet opposing, (p)ppGpp synthesizing and hydrolyzing activities [16, 2123]. SpoT proteins (in species such as E. coli that possess an additional RelA homologue) have potent (p)ppGpp hydrolyzing activities, but weak (p)ppGpp synthesizing activities in the absence of activating factors [4, 24, 25].

RelA, SpoT and long-Rel proteins are ca. 700 amino acids (aa) in length, and comprise two functional domains. The ca. 350–400 aa N-terminal domain comprises adjacent HD and SYNTH domains and contains all the residues essential for catalysis, whilst the ca. 350 aa C-terminal domain (CTD) is regulatory in function [4, 14, 2023, 2536]. To avoid futile catalytic events, the SYNTH and HD domains reciprocally-modulate each other’s activities via conformational antagonism [16, 2123]. The CTD modulates the (p)ppGpp synthesizing activities of the N-terminal domain via physical interactions with various molecular effectors such as ribosomal proteins, RNA components, acyl carrier protein subunits, and/or branched chain amino acids [4, 2128]. RelA, SpoT and long-Rel homologues exhibit considerable differences in their modes of intermolecular activation and inhibition. In particular, there are notable differences and uncertainties in the putative modulatory mechanisms underpinning the (p)ppGpp synthesis versus hydrolysis activities of single copy long-Rel proteins present within diverse bacterial systems [2138].

Recent studies have shown that certain Gram positive bacterial species such as Bacillus subtilis [39], Streptococcus mutans [40], Enterococcus faecalis [41], Mycobacterium smegmatis [42], Corynebacterium glutamicum [43], and Staphylococcus aureus [44], as well as the Gram negative pathogen Vibrio cholerae [45], encode additional proteins capable of synthesizing, but not hydrolyzing, (p)ppGpp alarmones. These have been termed ‘Small Alarmone Synthetase’ (SAS) proteins [39], as they contain a ‘SYNTH’ domain, lack a ‘HD’ domain, and (generally) lack additional sensory/modulatory domains. Mutational studies have suggested that the SAS proteins are non-essential for viability, as long as there is a Rel/SpoT protein (or functionally-equivalent hydrolase enzyme) present that can degrade (p)ppGpp molecules, which are toxic if allowed to accumulate within the cell [40, 44].

In addition to a single bifunctional (long) Rel protein, many Firmicutes species encode one or two SAS homologues, which are commonly referred to as RelP or RelQ proteins; with reference to the RelP and RelQ proteins from S. mutans [40]. Firmicutes RelQ and RelP proteins are respectively ca. 230 aa and ca. 210 aa in length, and share high levels of sequence similarity and structural homology [14, 4649]. It was recently shown that the RelQ protein from E. faecalis could efficiently synthesize a third alarmone molecule: guanosine-5’-monophosphate 3’-diphosphate (pGpp) from GMP + ATP, in addition to its previously established pppGpp and ppGpp-synthesizing activities [49]. This led the authors to propose that pGpp could also play a role in the stringent response. However, the pGpp-synthesizing abilities of RelQ/RelP homologues from other Firmicutes species remains poorly explored.

In the bacterial pathogen S. aureus, the rel gene is essential [50] but the relP and relQ genes are dispensable [44]. The S. aureus Rel (Sa-Rel), RelP (Sa-RelP) and RelQ (SA-RelQ) proteins (Fig 1) have the ability to synthesize ppGpp and pppGpp [26, 44, 48]. A recent report has shed light onto the in vivo regulation of the opposing (p)ppGpp synthetase and hydrolase activities of the Sa-Rel protein [26]. A detailed structural and mechanistic analysis into the (p)ppGpp synthesis activities of the Sa-RelP protein has also recently been published [48]. However, the detailed biochemical activities of the Sa-Rel and Sa-RelQ proteins remain to be fully established. In particular, their putative roles in pGpp synthesis and/or hydrolysis remain unexplored.

Domain architecture of Sa-Rel, Sa-Rel<sub>trunc</sub>, Sa-RelQ and Sa-RelP.
Fig. 1. Domain architecture of Sa-Rel, Sa-Reltrunc, Sa-RelQ and Sa-RelP.
Schematic representation of domain structure of the Rel, RelQ and RelP proteins encoded by S. aureus. Protein lengths in amino acids (aa) are shown. The N-terminal domain of Sa-Rel comprises the HD and SYNTH domains, respectively responsible for alarmone hydrolysis and synthesis. The regulatory C-terminal domain (CTD) of Sa-Rel contains four conserved sub-domains/motifs: a ThrRS, GTPase, and SpoT (TGS) domain; an alpha helical domain (Helical); a conserved cysteine (CC) domain, also known as a (zinc-finger domain (ZFD), and an aspartokinase, chorismate mutase and TyrA (ACT) domain, which is also known as a ribosome recognition motif (RRM). The Sa-Reltrunc protein lacks the entire regulatory CTD. The Sa-RelQ and Sa-RelP proteins only contain the SYNTH domain.

Here, we have performed a detailed comparative analysis of the respective biochemical activities of the Sa-Rel, Sa-RelQ and Sa-RelP proteins. We show that Sa-RelQ and Sa-RelP have the ability to synthesize pppGpp, ppGpp and pGpp alarmones, whilst Sa-Rel can only synthesize pppGpp and ppGpp. We further reveal that the (pp)pGpp-synthesizing activities of Sa-RelQ, but not Sa-RelP are stimulated by (pp)pGpp alarmones. We further confirm that Sa-Rel has the ability to hydrolyze all three alarmones with equivalent efficiencies.

Materials and methods

Bacterial strains and DNA manipulation procedures

S. aureus subsp. aureus str. Newman was cultured aerobically in 5% Horse blood agar, 1% Hemin and Vitamin K, 37°C. S. mutans ATCC 35668 was cultured aerobically in brain heart infusion (BHI) medium. E. coli DH10B was used for plasmid maintenance and genetic manipulations, and E. coli BL21 (DE3) was for recombinant protein expression; and both were grown aerobically in Luria-Bertani (LB) broth (USB Corp.) or on LB agar plates at 37°C. Ampicillin (100 μg/mL) or kanamycin (50 μg/mL) were added as required for plasmid maintenance. All restriction enzymes were purchased from NEB. PCR products were routinely cloned into pCR2.1 TOPO vectors using a TOPO-TA cloning kit (Invitrogen, Life Technologies), and the correct gene sequences were confirmed by sequencing in both directions using M13 Forward and M13 Reverse primers (S1 Table), before subcloning into the appropriate bacterial expression vectors. All expression constructs were sequenced to confirm accurate gene insertion. DNA was introduced into freshly-prepared (electro)-competent cells by electroporation (Gene Pulser, BioRad).

The S. aureus rel (NWMN_1536), relQ (NWMN_0876) and relP (NWMN_2405) genes were PCR amplified from S. aureus Newman genomic DNA using the following primers: rel: Sa-rel-Fb2 and Sa-rel-Re2; N-terminal (catalytic) domain of rel: Sa-rel-tFnco and Sa-rel-tRsal; relP: SARelPfor and SAVRelPrev; relQ: SARelQfor and SARelQrev. The S. aureus ppaC gene (SAV1919) was PCR amplified from S. aureus ATCC 25923 using primers SA1919for1 and SA1919rev1. The S. mutans relQ and relP genes were respectively PCR amplified from S. mutans ATCC 35668 genomic DNA using the following primers: relP: SMRelPfor and SMRelPrev, relQ: SMRelQfor and SMRelQrev. Primer sequences are shown in S1 Table.

After cloning into pCR2.1 TOPO vectors, amplified genes were excised by restriction digestion (BamHI and XhoI), and ligated into digested pET28a(+) plasmid (#69864–3, Novagen, Merck Biosciences). QIAquick PCR Purification Kits (QIAGEN, Germany) were used to purify PCR products, and QIAquick Gel Extraction Kits (QIAGEN, Germany), were used to purify DNA bands excised from agarose gels, according to the manufacturer’s protocol. DNA digestion products and PCR products were routinely analyzed on 1% (w/v) agarose gels (G-10 agarose, BIOWEST, ES) in Tris-Borate-EDTA buffer (TBE: 100mM Tris, 100mM Boric acid, 2mM EDTA). The construction of the pET28a-based plasmid expressing a His6-tagged EF-RelQ (EF2671) protein has previously been reported [51].

Protein expression and purification

E. coli BL21(DE3) cells freshly-transformed with the appropriate expression plasmid, were cultured at 37°C until OD600 reached ca. 0.6. Protein expression was induced by adding (isopropyl β-D-1-thiogalactopyranoside, IPTG; GE Healthcare) to a final concentration of 0.3 mM, then cells were cultured at room temperature (ca. 25°C) for 16 hours. Cells were collected by centrifugation (5,000 rpm, 10 min, 4°C), and the washed cell pellet was lysed by sonication in ‘Ni-binding buffer’ (25 mM Tris-HCl pH 7.4, 500 mM NaCl, 20 mM imidazole), containing protease inhibitors (Complete EDTA-free, Roche). Lysates were centrifuged (15,000×g, 60 min, 4°C), then supernatants were filtered (0.45 μm syringe filter, Iwaki) prior to purification on 5 ml HiTrap Chelating HP columns (GE Healthcare). Recombinant proteins were routinely purified by fast protein liquid chromatography (FPLC) using an AKTA purifier system (GE Healthcare), eluting with a linear gradient of 25 mM Tris-HCl pH 7.4, 500 mM NaCl, 200 mM imidazole versus the initial Ni-binding buffer. Proteins were routinely desalted using 5 mL HiTrap desalting columns (GE Healthcare, USA) using desalting buffer (50 mM Tris- HCl pH7.4, 150 mM NaCl, 5% glycerol). Further purification on 1 ml Resource Q columns (GE Healthcare) eluting with a linear gradient of 0.1–1 M NaCl gradient in 50 mM Tris- HCl pH7.4 was performed if required, followed by an additional desalting step. Protein concentrations were determined using the BioRad Protein assay (Bradford Reagent, BSA standard), and protein purity was determined by densitometry after 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS−PAGE). The A260/A280 levels of purified desalted proteins were determined using a Nanodrop 2000c spectrometer (ThermoFisher Scientific).

Enzymatic preparation of pGpp, ppGpp and pppGpp

The pGpp, ppGpp and pppGpp nucleotides were synthesized enzymatically using the recombinant EF-RelQ protein as previously described [44]. Briefly, reaction mixtures (100 μl) contained 5 μg of RelQ in 50 mM Tris-HCl pH 8.6, 100 mM NaCl, 10 mM MgCl2, 1 mM DTT, 1 mM ATP and 1 mM of GMP (to make pGpp) or GDP (to make ppGpp) or GTP (to make pppGpp), and were incubated for 2 hours at 30 °C. Product mixtures were loaded into a 1ml Resource Q anion exchange column (GE Healthcare), and nucleotides were separated using a gradient of 25 mM to 1 M NaCl in 25 mM Tris-HCl (pH 8.0), at a flow rate of 1.5 ml/min, using an AKTA purifier system. The eluent was monitored at 254 nm to detect and quantify nucleotide-containing fractions. Identical runs using known concentrations (0-1mM) of ATP, ADP, AMP, GTP, GDP, GMP, ppGpp and pppGpp were performed to enable unambiguous nucleotide identification and quantification. Fractions containing pure pppGpp or ppGpp were desalted by chromatography on Sephadex G-10 Sepharose chromatography with an AKTA purifier system, analogous to the method of Krohn and Wagner [52] and were characterized as described by Hardiman et al. [53].

The extinction coefficients of pGpp, ppGpp and pppGpp are highly similar to GMP, GDP and GTP, respectively (ca. 13,700 Lmol-1cm-1 at 253 nm). As a result, serial dilutions of a standard GMP, GDP and GTP solutions (2 mM, 1.5 mM, 1.0 mM, 0.5 mM and 0 mM) were used to generate a standard curve for calculating the pGpp, ppGpp and pppGpp concentrations, using a Nanodrop 2000c spectrometer at 254 nm.

Quantification of alarmone synthesis by Sa-Rel and Sa-Reltrunc proteins

The rate of alarmone synthesis was determined by quantifying AMP levels at set time-points using anion exchange chromatography (on a 1 ml Mono Q 5/50 GL column (GE Healthcare), as described above). For the Sa-Rel and Sa-Reltrunc proteins, reactions generally contained 50 mM Tris-HCl pH 7.8, 150 mM NaCl, 1 mM DTT, 6 mM MgCl2, 0.5 μM or 1 μM protein, ATP (various concentrations from 0–10 mM, kept constant at 8 mM if determining Km(GMP), Km(GDP), Km(GTP) values), and GMP/GDP/GTP (various concentrations from 0–4 mM, kept constant at 3 mM if determining Km(ATP)). Reactions were incubated at 25°C, with aliquots (20 μl) removed at set time points, which were immediately quenched by adding 20 mM EDTA, snap-frozen in liquid nitrogen, and stored at -70°C for future analysis. AMP product formation was quantified by comparing the AMP peak area on the respective chromatograms with corresponding reference values from a standard curve generated from serially-diluted concentrations of AMP (100–1000 μM).

Quantification of alarmone synthesis by Sa-RelP and Sa-RelQ

Assays were performed as described above, with minor modifications. Reactions contained 50 mM Bis-tris-propane (pH 9.0), 150 mM NaCl, 1 mM DTT, 5 mM MgCl2, 0.1 μM or 0.25 μM protein, 2.5 mM ATP and GMP/GDP/GTP (various concentrations from 0–3,000 μM). Reaction products were analyzed by ion exchange chromatography as described above, quantifying the respective nucleotide products by comparison of peak areas with values from standard curves respectively constructed using known concentrations of each product.

Quantification of alarmone hydrolysis by Sa-Rel and Sa-Reltrunc proteins

Rates of alarmone hydrolysis were determined by quantifying pyrophosphate generation using a continuous enzyme-coupled spectrophotometric assays incorporating the manganese-dependent (inorganic, type II) pyrophosphatase of S. aureus (Sa-PpaC, SAV1919) and the EnzChek phosphate Assay Kit (Thermo Fisher Scientific, USA). Briefly, reaction mixtures (200 μl) contained 50 mM Tris (pH 7.8), 150 mM NaCl, 1 mM DTT, 0.1 μM Sa-Rel/Sa-Reltrunc protein, 0.2 μM Sa-PpaC, 0.2 U purine nucleoside phosphorylase (PNP), 0.2 mM MESG, 10 mM MgCl2, and 1 mM MnCl2. Reaction mixtures were pre-mixed and pre-warmed at 25°C in 96-well plates (Corning Incorporated, USA), then the reactions were started immediately by adding serial concentrations (0~60 μM) of the respective alarmone substrate (pGpp, ppGpp and pppGpp). Assays lacking Sa-Rel/Sa-Reltrunc protein were performed in parallel, as controls. Reactions were monitored in real-time by measuring the absorbance at 360 nm using a SpectraMax M2e microplate reader (Molecular Devices, USA). A phosphate standard curve used to normalize absorbance signals was generated according to the manufacturer’s protocol.

For enzymatic assay data following Michaelis-Menten reaction kinetics, Vmax, kcat and Km values were determined by fitting data to the Michaelis-Menten equation using Prism 6.0 (GraphPad Software, USA). For enzymatic assay data following positive cooperative-binding kinetics, the Vmax, kcat and Km values and the Hill coefficient were determined by fitting data to the sigmoidal equation V = Vmax [Sn] / (Kmn + [Sn]) using Prism 6.0. Experiments were performed in 2–4 replicates, and the mean values ± standard deviation (SD) were reported.

Determination of enzymatic specific molar activities for Sa-RelP, Sa-RelQ, Sm-RelP, Sm-RelQ and Ef-RelQ

Assays mixtures contained 50 mM Bis-tris-propane (pH 9.0 or pH 7.6, as indicated in the text), 150 mM NaCl, 1 mM DTT, 2 mM MgCl2, 1 mM ATP, 1 mM GTP/GDP/GMP, with/without 100 μM pGpp/ppGpp/pppGpp, and were incubated at 37 °C for 0–60 mins. Reactions were initiated by the addition of protein (to 0.25 μM or 1 μM). As there were large variations in reaction rates for the different proteins and sets of substrates, reaction times and amounts of protein added were adjusted to ensure that substrate conversion never exceeded 15% of initial amounts present, to minimize potential product-mediated stimulatory or inhibitory effects, especially for the ‘no alarmone added’ reactions. Thus, depending upon the respective rates of the RelP/RelQ enzymes, incubation times ranged from 0–4 to 0–60 mins. Aliquots (20 μl) were collected for subsequent analysis at three different time points (as indicated in the respective text/figures), which were immediately quenched by adding 20 mM EDTA and snap frozen in liquid nitrogen, and stored at -80 degrees for subsequent analysis. Immediately prior to analysis, samples were thawed, 180 μl of buffer containing 25 mM Tris-HCl and 25 mM NaCl, was added, mixed by brief vortexing, then centrifuged (13,200 rpm for 10 min at 4°C). Nucleotide composition was analyzed as described above on Mono Q 5/50 GL columns (GE Healthcare) using an FPLC system (monitoring 254 nm absorbance), eluting with a linear gradient of 0.25–1 M NaCl in 25 mM Tris-HCl (pH 8.0). The respective rates of alarmone synthesis were determined by quantifying the rate of AMP production (which is equimolar to that of (pp)pGpp synthesis), as described above. The enzymatic specific molar activities were calculated in units of micromoles of (pp)pGpp product synthesized per minute per micromole of protein (μmol.min-1.μmol-1) according to momomeric RelP/RelQ molar concentrations. 2–4 replicates were performed for each condition, with the mean enzymatic specific molar activities reported ± standard deviation. The ‘fold change’ was calculated as the respective mean (pp)pGpp-stimulated enzymatic specific molar activity / mean non-stimulated enzymatic specific molar activity, for each of the 45 different enzymatic reactions.

Determining modulatory effects of RNA oligomer binding on alarmone synthesis rates by Sa-RelP, Sa-RelQ and EF-RelQ

Reaction mixtures contained 50 mM Bis-tris-propane (pH 9.0), 150 mM NaCl, 1 mM DTT, 1 mM GDP, 2 mM MgCl2, 1 μM of mRNA oligomer (MF, aMF, or random 24-mer; Sigma Aldrich Corporation), and 250 nM of Sa-RelQ/Sa-RelP/EF-RelQ protein (calculated as a tetramer). Reaction mixtures were pre-incubated at 37°C for 5 minutes, then ATP (1 mM) was added to initiate the reaction. Assay mixtures were incubated at 37°C for 0–30 mins, 20 μl aliquots were removed at three different time points, immediately quenched by adding 20 mM EDTA and snap frozen in liquid nitrogen, then stored at -80°C for downstream analysis by anion exchange chromatography as described above. Aliquots were removed at the following time-points, EF-RelQ: 10s, 20s, 40s; Sa-RelP: 30s, 2 min, 4 min; Sa-RelQ: 2 min, 15 min, 30 min. MF: 5’-GGCAAGGAGGUAAAAAUGUUCAAA-3’, aMF: 5’-UUUGAACAUUUUUACCUCCUUGCC-3’ [47].

Additional descriptions and experimental details are included in S1 Appendix.

Results

Multimeric arrangements of the Sa-Rel, Sa-Reltrunc Sa-RelQ and Sa-RelP proteins

The Sa-RelQ (SAS1), Sa-RelP (SAS2), Sa-Rel protein, and N-terminal domain of Sa-Rel (Sa-Reltrunc, residues 8–392), were expressed and purified in a soluble form. It should be noted that the recombinant 763 aa Sa-Rel protein studied here corresponds to residues 8–736 of the predicted coding DNA sequence for the S. aureus Rel protein [54], and includes a 34 aa N-terminal affinity tag encoded by the pET28a vector. A multiple sequence alignment containing the amino acid sequences for the recombinant Sa-Rel and Sa-Reltrunc, proteins studied here, along with several previously studied ‘long-Rel’ proteins is shown in S1 Fig; and is further discussed below). The Sa-Reltrunc protein encodes the SYNTH and HD domains that are respectively responsible for the synthesis and hydrolysis of alarmones; but lacks the putative regulatory domains found in the C-terminal region (Fig 1) [16, 26, 27, 21, 5557]. Its composition is analogous to the previously-characterized truncated form of the Rel protein from Streptococcus dysgalactiae subsp. equisimilis (‘S. equisimilis’ RelSeq NH 1–385) [16, 23]. The recombinant 396 aa Sa-Reltrunc protein contains an additional glycine residue at the penultimate N-terminal position (introduced due to the cloning strategy employed) as well as a 10 aa C-terminal hexahistidine affinity tag. The SAS homologues encoded by S. mutans: Smu_1046c (Sm-RelQ; SAS1) and Smu_926 (Sm-RelP; SAS2) [40], as well as the RelQ protein from E. faecalis (EF-RelQ; RelQEF; EF2671; SAS1) [41, 49, 58] were analogously expressed and purified, so that their respective biochemical activities could be investigated in parallel, for comparative purposes. A260/A280 ratios were determined for all proteins to determine if they were complexed to nucleic acids (DNA/RNA). Results were as follows: Sa-Rel (1.32 ± 0.07); Sa-Reltrunc (0.63 ± 0.11), Sa-RelQ (0.65 ± 0.04), Sa-Rel-P (0.62 ± 0.04), EF-RelQ (0.57 ± 0.03). By comparison with previously published reference values for protein:nucleic acid (RNA/DNA) mixtures [59], these results suggested that the purified Sa-Reltrunc, Sa-RelQ, Sa-Rel-P and EF-RelQ proteins were essentially free of (complexed) nucleic acid, whilst the purified Sa-Rel protein comprised ca. 90% protein and ca. 10% nucleic acid (originating from the E. coli expression host). The composition of the co-purifying nucleic acid component was not determined.

Qualitative analysis of alarmone synthesizing abilities of the Sa-RelQ and Sa-RelP proteins

The respective GMP/GDP/GTP ‘acceptor’ substrate preferences of the Sa-RelP and Sa-RelQ proteins were determined under standardized conditions, analyzing enzymatic product mixtures using anion exchange chromatography [51]. The pGpp, ppGpp and pppGpp alarmones [collectively referred to as (pp)pGpp] were synthesized using the EF-RelQ protein; and were purified and quantified as we have previously reported [51]. These (pp)pGpp alarmones were used as standards to confirm the identity of enzymatic products, and to enable their quantitation, via analyzing peak areas on chromatograms obtained under identical conditions. Representative chromatograms of ATP/ADP/AMP, GTP/GDP/GMP, and (pp)pGpp standards are shown in Fig 2 (Panels AC). Representative chromatograms of enzymatic product mixtures respectively obtained for the incubation of Sa-RelQ and Sa-RelP with GTP + ATP, GDP + ATP and GMP + ATP are shown in Fig 2 (Panels DF and GI, respectively).

(pp)pGpp alarmone synthesis activities of Sa-RelP and Sa-RelQ.
Fig. 2. (pp)pGpp alarmone synthesis activities of Sa-RelP and Sa-RelQ.
Panels AI show representative anion-exchange chromatograms performed under analogous conditions (as described in Materials and methods). Panels AC show mixtures of pure nucleotides (A) ATP, ADP and AMP, (B) GTP, GDP and GMP, (C) pGpp, ppGpp and pppGpp. Panels D−F show product mixtures formed by the incubation of the Sa-RelQ protein with the following nucleotide mixtures: (D) ATP + GTP, (E) ATP + GDP, (F) ATP + GMP. Panels GI show product mixtures formed by the incubation of the Sa-RelP protein with the following nucleotide mixtures: (G) ATP + GTP, (H) ATP + GDP, (I) ATP + GMP.

As may be seen in Fig 2, the Sa-RelP and Sa-RelQ proteins respectively synthesized pppGpp, ppGpp and pGpp, with no competing hydrolytic processes. Minor amounts of an additional product, tentatively identified as ppGpp, were produced by Sa-RelP (but not Sa-RelQ) in the GMP + ATP assay (Panel I). This may be attributed to low-level enzymatic phosphorylation of GMP to produce GDP (which is rapidly converted to ppGpp), as has previously been noted to occur for the RelAct SAS (RelP*Cg) protein from C. glutamicum [43]. Analogous alarmone-synthesis assays were performed for the EF-RelQ protein, for comparative purposes. Consistent with previous data [49], Ef-RelQ efficiently catalyzed the respective synthesis of pppGpp, ppGpp and ppGpp (data not shown).

Sa-RelP and Sa-RelQ could also utilize inosine 5’-triphosphate (ITP) and inosine 5’-diphosphate (IDP) as substrates, in place of GTP/GDP (S2 Fig). This led to the production of phosphorylated inosine nucleotides that most probably correspond to the respective inosine-based ‘alarmone-like’ molecules pppIpp and ppIpp. Under the conditions employed, the pppIpp and ppIpp synthesizing activities of Sa-RelP were considerably higher than those of Sa-RelQ. Furthermore, Sa-RelP, but not Sa-RelQ, could utilize inosine 5’-monophosphate (IMP) for pIpp synthesis, albeit with poor efficiency. The EF-RelQ protein shared this ‘flexibility’ in substrate specificity, and could also synthesize pppIpp, ppIpp and pIpp (S2 Fig, Panels DF). It may be noted that the Sa-RelP, Sa-RelQ and EF-RelQ proteins each displayed differing preferences for IMP/IDP/ITP utilization, e.g. Sa-RelP converted ITP to pppIpp considerably more efficiently than Sa-RelQ or EF-RelQ. There was some evidence for additional phosphohydrolase/phosphotransfer activities for the EF-RelQ protein, which appeared to produce small amounts of ppIpp in addition to pppIpp, in the ITP + ATP reaction (S2 Fig, Panels C, F and I). It should be noted that ATP was essentially required for the synthesis of pppIpp by Sa-RelP, Sa-RelQ or EF-RelQ, indicating that ITP could function as a pyrophosphate acceptor for alarmone synthesis, but not a donor (data not shown).

The pH optima for the GDP + ATP activities of Sa-RelP and Sa-RelQ were determined to be ca. 8.4–9.0 and ca. 9.0, respectively (S3 Fig). Therefore, pH 9.0 buffer was used for their subsequent kinetic analysis. It may be noted that we found no discernable alterations in substrate utilization patterns for any of the SAS proteins analyzed in this study when buffers of pH 7.6 or 9.0 were used (see below).

Qualitative analysis of alarmone synthesizing abilities of the Sa-Rel and Sa-Reltrunc proteins

Analogous sets of standardized incubations were performed for the Sa-Rel and Sa-Reltrunc proteins, with the corresponding chromatograms of enzymatic product mixture shown in S4 Fig. Alarmone production levels were evaluated based on the levels of AMP byproduct formed (corresponding peak areas on the respective chromatograms), rather than the amounts of (pp)pGpp produced, due to the competing alarmone synthesis/hydrolysis reactions that may be taking place in each incubation. Results indicated that both the Sa-Rel and Sa-Reltrunc proteins catalyzed the ATP + GTP → pppGpp + AMP reaction most effectively, but the pppGpp-synthesizing activities of Sa-Reltrunc were higher than those of Sa-Rel, under the conditions used. In comparison, both Sa-Rel and Sa-Reltrunc had considerably lower ppGpp synthesizing activities, and lacked detectable pGpp synthesis activities (from ATP + GMP) (S4 Fig). As may be seen in the respective chromatograms, under the conditions employed, the majority of the nascent pppGpp or ppGpp products were hydrolyzed back to the corresponding GTP or GDP substrates. Consistent with previous reports for the long-Rel protein from C. glutamicum (RelCg) [43] small amounts of ADP and GDP were also produced during these assays (from ATP and GTP respectively), putatively via competing (enzymatic) phosphohydrolase or phosphotransferase processes. The Sa-Reltrunc protein synthesized pppIpp from ITP + ATP (data not shown), a feature that has previously been noted for the S. equisimilis RelSeq NH 1–385 and M. tuberculosis RelMtb proteins [16, 22, 60].

For the Sa-Reltrunc protein, the respective optimal pH values for the ATP + GDP and ATP + GDP reactions were ca. 7.8 and 8.8, respectively (S3 Fig). Therefore, a consensus pH value of 7.8 was chosen to determine the kinetic parameters for the (p)ppGpp synthesis activities for both Sa-Rel and Sa-Reltrunc. This is highly consistent with the properties of the RelMtb protein, which was found to exhibit a broad pH optimum for its alarmone synthesis (transferase) and hydrolase activities, with maximal activities occurring between pH 7.9–8.3. [22].

The efficiency of (p)ppGpp synthesis by the Rel proteins from S. equisimils (RelSeq) [16, 21] and M. tuberculosis (RelMtb) have previously been shown to depend on the respective molar ratios of Mg2+ ions to ATP + GTP (GDP) substrates. Mechold et al. found that the synthetic activities of RelSeq activities were optimal when Mg2+ concentrations were equal to the sum of the (molar) concentrations of the two substrates (ATP + GTP/GDP) [16]. Therefore, we determined the respective synthetic activities of Sa-Reltrunc under fourteen different conditions: using 3–4 different molar ratios of substrate (ATP + GTP, or ATP + GDP; 1–3 mM) in the presence of 3–4 different concentrations of Mg2+ ions (0.5–10 mM; see S5 Fig). Results indicated there was a complex relationship between Mg2+ ions and nucleotide substrate utilization for the Sa-Reltrunc protein. We found that the ratio of ATP:Mg2+ ions had a greater influence on overall synthetic activities, compared to the ratio of GTP(GDP): Mg2+ ions. Sa-Reltrunc synthetic activities were generally highest when the ratio of (ATP + GDP/GTP): Mg2+ ions were approximately equal. However, consistent with previous results [22], higher concentrations (molar ratios) of Mg2+ ions led to reduced alarmone synthesis activities, especially when the concentrations of Mg2+ ions were above 6 mM.

Kinetic analysis of (p)ppGpp alarmone synthesis by the Sa-Rel and Sa-Reltrunc proteins

Based on the above results, and with reference to previous investigations [16, 22], we used 8 mM ATP and 6 mM Mg2+ ions for the determination of Km(GTP) and Km(GDP) values for the Sa-Rel and Sa-Reltrunc proteins. The curves obtained for the sets of assays (Fig 3) exhibited the typical characteristics of Michaelis-Menten kinetics. As shown in Table 1, the data suggested that both the Sa-Rel and Sa-Reltrunc proteins notably preferred GTP over GDP as the acceptor substrate.

<i>V</i><sub>0</sub>/[S] plots used to calculate enzymatic kinetic parameters for (p)ppGpp synthesis activities of Sa-Rel and Sa-Rel<sub>trunc</sub>.
Fig. 3. V0/[S] plots used to calculate enzymatic kinetic parameters for (p)ppGpp synthesis activities of Sa-Rel and Sa-Reltrunc.
Panels AE respectively show the plots of rate of (p)ppGpp synthesis (V0; Y-axis, in units of μM / minute) versus ATP/GTP/GDP substrate concentration ([S]; X-axis; in millimolar units), for sets of assays performed to calculate the enzymatic kinetic parameters for the Sa-Rel and Sa-Reltrunc (contains HD and SYNTH domains only) proteins. (A) Enzymatic kinetic parameters for ATP utilization (0–10 mM) were calculated for pppGpp synthesis by Sa-Reltrunc for a fixed GTP concentration (3 mM). In Panels BE, the determination of kinetic parameters for GTP and GDP utilization were determined using a fixed ATP concentration of 8 mM. (B) GTP utilization (0–400 μM) for pppGpp synthesis by Sa-Reltrunc. (C) GDP utilization (0–4 mM) for ppGpp synthesis by Sa-Reltrunc. (D) GTP utilization (0–3 mM) for pppGpp synthesis by Sa-Rel. (C) GDP utilization (0–3 mM) for ppGpp synthesis by Sa-Rel. 2–3 replicates were performed for each condition, with mean values plotted ± standard deviation.
Tab. 1. Kinetic parameters for (p)ppGpp synthesis by Sa-Reltrunc and Sa-Rel.
Kinetic parameters for (p)ppGpp synthesis by Sa-Rel<sub>trunc</sub> and Sa-Rel.

For both the Sa-Rel and Sa-Reltrunc proteins, the respective Km values for GTP (114.3 ± 15.7 μM, 64.8 ± 14.1 μM) were several-fold lower than the corresponding Km values for GDP (434.9 ± 61.2 μM, 654.9 ± 88.1 μM). In addition, the respective kcat values for GTP were several-fold higher than those for GDP, resulting in catalytic efficiencies (kcat/Km) for pppGpp synthesis that were ca. 50-fold, and ca. 13-fold higher than those for ppGpp synthesis, for the Sa-Rel and Sa-Reltrunc proteins, respectively. The Km(ATP) was determined for the pppGpp-synthesizing activities of Sa-Reltrunc protein (fixing the GTP concentration at 3 mM) and was evaluated to be 4.70 ± 1.89 mM. This is similar to previous estimates for the Km(ATP) value for the RelSeq NH 1–385 protein of ≥ ca. 5 mM [16]. However, these values are very different to those determined for the N-terminal domain of the bifunctional long-Rel protein from M. extorquens (RSHMex1-352), where the Km(ATP) was found to be ca. 0.4 mM, which was nearly 8-fold lower than the Km(GTP) of ca. 3 mM [38].

Alarmone hydrolyzing activities of the Sa-Rel and Sa-Reltrunc proteins

The specific hydrolytic removal of the 3’-pyrophosphate group from pppGpp and ppGpp alarmones by Rel/SpoT enzymes is well-established [18, 20, 6063]. In addition, it was recently shown that the (bifunctional) long-Rel protein from E. faecalis (EF1974; RelEF) had the ability to hydrolyze pGpp (to GMP and pyrophosphate) with an efficiency that appeared to be roughly equal to that of its pppGpp and ppGpp hydrolysis activities [49]. Therefore, we sought to quantify the degree to which the Sa-Rel and Sa-Reltrunc proteins hydrolyzed pppGpp, ppGpp and pGpp to the respective guanine nucleotides. As may be seen in S6 Fig, qualitative assays performed under standardized conditions revealed that both Sa-Rel and Sa-Reltrunc possessed potent pppGpp, ppGpp and pGpp hydrolyzing activities, with GTP, GDP and GMP, specifically formed as the respective nucleotide products.

Whilst the alarmone-synthesizing (i.e. pyrophosphotransferase) activities of Rel proteins were dependent upon Mg2+ ions, the alarmone-hydrolyzing (i.e. 3’-pyrophosphohydrolase) activities of long-Rel (and SpoT) proteins were dependent-upon (or were greatly stimulated by) Mn2+ ions [16, 18, 20, 6063]. We quantified the respective pppGpp, ppGpp and pGpp-hydrolysis activities of the Sa-Reltrunc and Sa-Rel proteins in the absence or presence of Mg2+ (10 mM) or Mn2+ ions (0–2.5 mM) using an enzyme-coupled pyrophosphate-to-phosphate spectrophotometric assay. Results are summarized in S7 and S8 Figs. These assays incorporated the highly-efficient, Mn2+-dependent S. aureus type II inorganic pyrophosphatase (PpaC; SAV1919) protein [64] to hydrolyze liberated diphosphate ion to (ortho-)phosphate, which was then quantified in a continuous manner using the EnzChek Phosphate Assay Kit (Invitrogen). In control assays that contained Sa-PpaC, Mn2+ ions (0–2.5 mM) and ppGpp, but lacked Sa-Rel/Sa-Reltrunc, there was no production of phosphate (S7 Fig). Neither Sa-Reltrunc nor Sa-Rel had alarmone-hydrolyzing activities in the absence of metal ions (S7 Fig). The (pp)pGpp-hydrolyzing activities of the Sa-Reltrunc and Sa-Rel proteins were optimal in the presence of 1.0 mM Mn2+ ions. The addition of Mg2+ ions (10 mM) imbued the Sa-Rel protein with very low (but detectable) (pp)pGpp hydrolyzing activities, but led to no observable activities for the Sa-Reltrunc protein (S7 Fig). When Mn2+ ions were present at a concentration of 1 mM, the addition of 10 mM Mg2+ ions had no detectable effects on the hydrolytic activities of the Sa-Reltrunc or Sa-Rel proteins. This suggested that Mg2+ ions had no notable inhibitory nor synergistic effects on Rel−Mn2+ ion binding or hydrolytic activities. Therefore, 10 mM Mg2+ ions and 1 mM Mn2+ ions were used for the subsequent kinetic analyses of the pppGpp, ppGpp and pGpp-hydrolyzing activities of the full-length and truncated forms of the Sa-Rel protein.

In additional (single time point) qualitative experiments (S9 Fig), we further confirmed that the Sa-Rel protein had negligible ppGpp hydrolysis activities in the absence of added Mn2+ or Mg2+ ions. Consistent with the results from the spectrophotometric assays described above, 20 picomoles of Sa-Rel protein could completely hydrolyze 10 nanomoles of ppGpp to GDP in the presence of 10 mM Mg2+ ions (or 1 mM Mn2+ ions) within 2 hours (at 25°C, pH 7.8) (S9 Fig).

Results from assays performed over a range of different alarmone concentrations (0–60 μM) generated curves with characteristics typical of Michaelis-Menten kinetics (S8 Fig). The corresponding enzyme kinetic parameters are summarized in Table 2. It may be seen that Sa-Reltrunc and Sa-Rel proteins had very similar hydrolytic activities towards all three alarmone substrates. Km values were all within the range 14.2 ± 1.3 to 34.5 ± 5.5 μM, with kcat values ranging from 0.60 ± 0.03 to 1.98 ± 0.16 s-1, under the assay conditions used. Thus, the corresponding catalytic efficiency (kcat/Km) values for the full-length and truncated Sa-Rel proteins towards pGpp, ppGpp and pppGpp substrates were all fairly similar, falling within the range 27.5 ± 2.2 to 57.3 ± 4.8 mM-1s-1.

Tab. 2. Kinetic parameters for (pp)pGpp alarmone hydrolysis by Sa-Reltrunc and Sa-Rel.
Kinetic parameters for (pp)pGpp alarmone hydrolysis by Sa-Rel<sub>trunc</sub> and Sa-Rel.

Kinetic analysis of Sa-RelP and Sa-RelQ alarmone synthesizing activities

The alarmones pppGpp and ppGpp have been shown to (allosterically) stimulate the synthetic activities of the SAS proteins from E. faecalis (RelQEf) [47, 49] and B. subtilis (SAS1, YjbM) [65], but to (allosterically) inhibit the activities of the Sa-RelP protein [48]. However, the modulatory effects of alarmones on the Sa-RelQ protein remain to be determined. More notably, to the best of our knowledge, the potential modulatory effects of the putative alarmone pGpp have not yet been investigated on any SAS. Therefore, we first determined the respective enzymatic molar activities of pppGpp, ppGpp and pGpp synthesis for the Sa-RelQ and Sa-RelP proteins (i.e. micromoles of alarmone product formed per minute per micromole of protein; μmol.min-1.μmol-1), in the presence or absence of 100 μM pppGpp, ppGpp or pGpp, in sets of assays performed under standardized conditions at pH 9.0. The degree to which the respective rates for each of the alarmone-synthesizing reactions were stimulated or reduced in the presence of (pp)pGpp alarmones were calculated as a ‘fold-change’, compared to the corresponding ‘non-stimulated’ (basal) rates. Results are summarized in Table 3.

Tab. 3. Summary of RelP and RelQ molar specific activities with/without alarmone stimulation.
Summary of RelP and RelQ molar specific activities with/without alarmone stimulation.

It was found that the respective rates by which Sa-RelQ synthesized pppGpp, ppGpp and pGpp were all stimulated by the presence of (pp)pGpp alarmones: albeit to greatly varying degrees; relative to the corresponding non-stimulated rates. The addition of pppGpp, ppGpp or pGpp elicited the most notable stimulatory effects on the Sa-RelQ-mediated GMP + ATP → pGpp reaction: increasing the basal rates ca. 36-fold, 26-fold, and 4.5-fold, respectively. In contrast, (pp)pGpp alarmones elicited more modest stimulatory effects on the ppGpp and pppGpp-synthesizing activities of Sa-RelQ, increasing them only ca. 1.5 to 1.9-fold. However, it should be noted that the basal rates for Sa-RelQ-catalyzed ppGpp and pppGpp synthesis were respectively ca. 30-fold, and 18-fold higher than that for pGpp synthesis, under the conditions used. In sharp contrast, the respective rates by which Sa-RelP synthesized each of the three alarmones were all reduced by the presence of any of the three alarmones (ca. 10–30%), compared to the equivalent reactions performed in the absence of alarmones. This is discussed in greater detail below.

The biochemical activities of Sa-RelQ and Sa-RelP were further investigated in analogous sets of assays performed at pH 7.6, which may be considered more physiologically relevant. Specifically, we determined the degree to which 100 μM ppGpp respectively affected the rate of the Sa-RelQ-mediated GMP + ATP, GDP + ATP and GTP + ATP reactions. Results are shown in S2 Table. The presence of ppGpp, increased the rates of pGpp, ppGpp and pppGpp production (34 ± 1.9)-fold, (1.6 ± 0.3)-fold, and (1.4 ± 0.2)-fold, respectively, at pH 7.6. These are highly consistent with the trends shown in Table 3. In addition, the respective (unstimulated) rates of alarmone product formation by Sa-RelQ were the same at pH 7.6 and pH 9.0: ppGpp > pppGpp >> pGpp. However, the respective rates for the stimulated and unstimulated reactions at pH 7.6 were ca. 1.2–3-fold slower than the corresponding reactions performed at pH 9.0. The biochemical activities of Sa-RelP were also analogous at pH 7.6 and pH 9.0, with the rate of alarmone formation (in the absence of allosteric alarmone) following the order ppGpp > pppGpp > pGpp. Similarly, the respective rates of (pp)pGpp formation by Sa-RelP were 1.2–3-fold slower than the corresponding reactions performed at pH 9.0.

We therefore determined the kinetic parameters for pppGpp, ppGpp and pGpp synthesis by Sa-RelQ in the presence of 100 μM pppGpp, ppGpp or pGpp; and those for Sa-RelP without (pp)pGpp. For the Sa-RelQ protein, a plot of the initial reaction velocity (V0) versus GTP/GDP/GMP concentration [S] showed a distinct sigmoidal characteristic (Fig 4). This is consistent with previous reports for the B. subtilis RelQ (YjbM) protein [48, 65], indicative of positive cooperativity between the four monomers comprising the RelQ homotetramer. Thus the Hill equation (V = {Vmax [S]n} / {Km + [S]n} was employed to calculate the enzymatic kinetic parameters (where n denotes the Hill coefficient). Analogous V0/[S] plots for the Sa-RelP protein were hyperbolic in nature, and did not show any sigmoidal characteristics (Fig 5), consistent with a previous report [48]. The corresponding kinetic parameters obtained are summarized in Table 4.

<i>V</i><sub>0</sub>/[S] plots used to calculate enzymatic kinetic parameters for (pp)pGpp-stimulated (pp)pGpp-synthesis activities of Sa-RelQ.
Fig. 4. V0/[S] plots used to calculate enzymatic kinetic parameters for (pp)pGpp-stimulated (pp)pGpp-synthesis activities of Sa-RelQ.
Panels AI show plots of rate of (pp)pGpp synthesis (V0; Y-axis, in units of μM / minute) versus GTP/GDP/GMP substrate concentration ([S]; X-axis; in micromolar units), for sets of assays performed to calculate the enzymatic kinetic parameters for the Sa-RelQ protein. All assays were performed in the presence of 100 μM concentrations of pppGpp, ppGpp or pGpp stimulatory alarmone, as indicated in the respective panels. (Panels AC) pppGpp synthesis in the presence of pppGpp (A), ppGpp (B), or pGpp (C). (Panels DF) ppGpp synthesis in the presence of pppGpp (D), ppGpp (E), or pGpp (F). (Panels G−I) pGpp synthesis in the presence of pppGpp (G), ppGpp (H), or pGpp (I). Experimental details are described in the materials and methods section.
<i>V</i><sub>0</sub>/[S] plots used to calculate enzymatic kinetic parameters for (pp)pGpp synthesis activities of Sa-RelP.
Fig. 5. V0/[S] plots used to calculate enzymatic kinetic parameters for (pp)pGpp synthesis activities of Sa-RelP.
Panels AC respectively show the plots of rate of pppGpp, ppGpp and pGpp synthesis (V0; Y-axis, in units of μM / minute) versus GTP, GDP, or GMP substrate concentrations ([S]; X-axis; in micromolar units), for sets of assays performed to calculate the enzymatic kinetic parameters for the Sa-RelP protein. Experimental details are described in the materials and methods section. 2–3 replicates were performed for each condition, with mean values plotted ± standard deviation.
Tab. 4. Kinetic parameters for (pp)pGpp synthesis by Sa-RelP and Sa-RelQ.
Kinetic parameters for (pp)pGpp synthesis by Sa-RelP and Sa-RelQ.

The Sa-RelP protein utilized GTP and GDP with similar catalytic efficiencies (4.75 ± 1.00 and 4.43 ± 0.90 mM-1·s-1, respectively), ca. 50-fold more efficiently than GMP (0.093 ± 0.015 mM-1·s-1). However, the Km for GTP (80 ± 21 μM) was ca. 2-fold lower than the Km for GDP (177 ± 52 μM), whilst the Vmax for GDP (0.20 ± 0.02 μM.s-1) was ca. 5-fold higher than the Vmax for GTP (0.095 ± 0.01 μM.s-1). The higher Vmax(GDP) value almost certainly explains why Sa-RelP synthesizes ppGpp ca. 1.5–2-fold faster than pppGpp under the assays conditions used to obtain the data described in Table 3. These results are in good agreement with those recently reported by Steinchen et al. for the Sa-RelP protein (further discussed below) [48]. The considerably higher Km(GMP) value (625 ± 165 μM) and lower Vmax(GMP) value (0.058 ± 0.007 μM.s-1) result in pGpp synthesis rates being substantially lower than those of ppGpp and pppGpp under analogous conditions (Table 3).

In the presence of all three alarmones, Sa-RelQ utilized GDP more efficiently than GTP as an acceptor substrate. The Km(GDP) values (202 ± 71 to 372 ± 21 μM) were ca. 2 to 4-fold lower than the corresponding Km(GTP) values (ca. 884 ± 94 to 1334 ± 320 μM). The kcat/Km(GDP) values for Sa-RelQ (4.52 ± 0.67 to 8.57 ± 0.37 mM-1·s-1) were ca. 2 to 4-fold higher than the corresponding kcat/Km(GTP) values (1.78 ± 0.2 to 2.70 ± 0.15 mM-1s-1). The Vmax values for the Sa-RelQ-catalyzed ppGpp and pppGpp synthesis reactions were reasonably similar (ranging from 0.24 ± 0.04 to 0.82 ± 0.06 μM·s-1). The Hill coefficient values for the Sa-RelQ-catalyzed (p)ppGpp-synthesis reactions ranged from 1.2 ± 0.3 to 2.1 ± 0.4. This supports the premise that there is a certain degree of positive cooperativity between protein monomers during alarmone synthesis in the presence of 100 μM (pp)pGpp. Intriguingly, out of the three alarmones, pGpp had the largest stimulatory effects on the maximum rate, turnover numbers and catalytic efficiency of ppGpp and pppGpp synthesis by Sa-RelQ.

Regarding the kinetic parameters for the Sa-RelQ-mediated synthesis of pGpp, in the presence of additional pppGpp or ppGpp, the Km(GMP) values (948 ± 233 and 1380 ± 277 μM, respectively) and corresponding kcat/Km(GMP) values (3.43 ± 0.47 and 1.42 ± 0.13 mM-1s-1), were fairly similar to the respective values obtained for the GTP + ATP reactions. However, the Km(GMP) value for the reaction performed in the presence of pGpp (4,953 ± 772 μM) was ca. 3 to 5-fold higher than the corresponding Km(GMP) values obtained when this reaction was performed in the presence of pppGpp or ppGpp.

Comparative analysis of (pp)pGpp synthesizing activities of RelP/RelQ proteins from S. aureus, E. faecalis and S. mutans

To establish how the biochemical activities of the S. aureus RelP and RelQ proteins compared with those of other bacterial RelP/RelQ proteins, we quantified the respective (pp)pGpp synthesizing abilities of S. mutans RelP (Sm-RelP), S. mutans RelQ (Sm-RelQ), and E. faecalis RelQ (EF-RelQ), under conditions identical to the assays described above. Results are summarized in Table 3. It may be seen that the allosteric properties of Sm-RelP are analogous to those of Sa-RelP, in that they are not stimulated by any alarmone; and are in fact inhibited (ca. 1 to 4-fold) by all three alarmones at a concentration of 100 μM. The (pp)pGpp-synthesizing activities of Sm-RelP were broadly similar to those of Sa-RelP, under the conditions employed. In the absence of alarmones, Sm-RelP synthesized pppGpp most effectively, with a specific activity of 29.1 ± 10.1 μmol.min-1.μmol-1. This was ca. 2-fold higher than its corresponding rate of ppGpp synthesis, and ca. 10-fold higher than its rate of pGpp synthesis.

The pppGpp-synthesis rates of Sm-RelQ (242 ± 19 μmol.min-1.μmol-1) were ca. 30-fold higher than those of Sa-RelQ (Table 3). Its non-stimulated rates of pGpp synthesis (0.43 ± 0.14 μmol.min-1.μmol-1) and ppGpp synthesis (7.07 ± 2.17 μmol.min-1.μmol-1) were comparable to those of Sa-RelQ. However, Sm-RelQ’s pppGpp-synthesizing activities were slightly inhibited (20–40%) by the presence of (pp)pGpp alarmones. In contrast, its ppGpp and pGpp-synthesizing activities were dramatically increased in the presence of ppGpp and pppGpp alarmones. The addition of pppGpp stimulated the Sm-RelQ-mediated synthesis of ppGpp ca. 17-fold, whilst ppGpp stimulated it ca. 5-fold. Most notably, the rate of pGpp synthesis by Sm-RelQ was more than 500-fold higher in the presence of pppGpp (310 ± 80 μmol.min-1.μmol-1), compared to the non-stimulated rate. In comparison, additional (allosteric) ppGpp and pGpp alarmones had much more modest stimulatory effects on Sm-RelQ-catalyzed pGpp synthesis (ca. 3.5-fold and ca. 1.6-fold, respectively).

In almost every regard, the rates of alarmone synthesis by EF-RelQ were considerably higher than those of the other RelP/RelQ proteins investigated here. In addition, all of its respective (pp)pGpp-synthesizing activities were stimulated by the presence of (100 μM) pGpp, ppGpp and pppGpp. Consistent with previous reports [49], we found that EF-RelQ synthesized ppGpp more rapidly than pppGpp or pGpp, in the presence of 100 μM (pp)pGpp alarmones. It may be noted that ppGpp-synthesis activities were strongly-stimulated (ca. 5.5 to 7.8-fold) by the presence of (pp)pGpp alarmones. The presence of (pp)pGpp alarmones had a very modest stimulatory effect on pppGpp synthesis by EF-RelQ (ca. 1.5 to 1.9-fold increase). Levels of pGpp synthesis by EF-RelQ were most strongly stimulated by pppGpp (ca. 4.4-fold), but were barely stimulated by pGpp (ca. 1.3-fold). It is notable that in the presence of pppGpp, the rates of pGpp synthesis by EF-RelQ (367.1 ± 58.2 μmol.min-1.μmol-1), were very similar to those of Sm-RelQ under the same conditions. However, the basal rates of pGpp synthesis by EF-RelQ were ca. 190-fold higher than those of Sm-RelQ. This underlines the notable subtleties of allosteric regulation of (pp)pGpp synthesis by the three respective alarmones.

Modulation of Sa-RelP and Sa-RelQ activities by RNA oligomers

The degree to which the addition of single stranded RNA oligomers positively or negatively affected the rates of alarmone synthesis by Sa-RelQ, Sa-RelP and EF-RelQ were determined, analogously to the method previously described by Beljantseva et al. [47]. Specifically, 3 different RNA oligomers of equal length (24-mers): mRNA(MF) and antisense mRNA(MF), as well as a 24-mer RNA oligomer of random sequence, were tested as representative RNA molecules. The respective levels of ppGpp formed were quantified, and the rates of ppGpp production in the presence of RNA were normalized to the respective rates produced in the absence of RNA. Results are summarized in Fig 6. The rate of ppGpp production by EF-RelQ was strongly inhibited by equimolar concentrations of mRNA(MF) (ca. 6-fold), and to a lesser extent by RNA(random) (ca. 2.5-fold), but was barely inhibited by the addition of the antisense mRNA(MF) (ca. 0.8-fold). In contrast, the rates of ppGpp synthesis by Sa-RelQ and Sa-RelP were little-affected by the presence of any of the three RNA oligomers under the conditions used.

Single stranded RNA oligomers do not inhibit the ppGpp-synthesis activities of Sa-RelQ or Sa-RelP.
Fig. 6. Single stranded RNA oligomers do not inhibit the ppGpp-synthesis activities of Sa-RelQ or Sa-RelP.
Plots show the respective effects of adding equimolar amounts of three different single stranded RNA oligomers of equal length [MF, aMF or a 24-mer of random sequence (Random)] on the rate of ppGpp synthesis by Sa-RelQ (Panel A), Sa-RelP (Panel B) or EF-RelQ (Panel C). The inhibitory/stimulatory effects were evaluated as a ‘fold-change’ on ppGpp-synthesis rates versus analogous control reactions performed in the absence of RNA oligomer (Control). The table summarizes the respective ‘fold-change’ in rates of ppGpp-synthesis, calculated as a mean value ± standard deviation. The sequence of the MF and aMF RNA oligomers are identical to those reported previously [47].

Discussion

S. aureus synthesizes (pp)pGpp ‘alarmones’ via the combined activities of one bifunctional long-Rel protein (Sa-Rel) and two SAS proteins: Sa-RelP and Sa-RelQ [26, 44, 48, 50, 66, 67]. Sa-Rel appears to be the major source of guanosine pentaphosphate (pppGpp) and guanosine tetraphosphate (ppGpp) in S. aureus cells where the stringent response has been induced [26, 6769]. The rel gene is essential for the growth of S. aureus [50], due to its alarmone hydrolyzing activities [44, 67]. When relP and relQ are deleted, the rel gene become dispensable, as no (pp)pGpp is synthesized in the cell. Mutant strains lacking both relP and relQ have reduced survival activities in response to antibiotic agents that induce cell-envelope ‘stresses’, such as ampicillin and vancomycin [44]. In other Firmicutes taxa that contain SAS homologues whose alarmone-synthesizing activities have been determined, such as E. faecalis, S. mutans and B. subtilis; the rel gene is non-essential for viability [41, 70, 71], suggesting that there may be alternative pathways for (pp)pGpp hydrolysis, such as small alarmone hydrolases (SAHs) [14, 72], NUDIX hydrolases (Ndx8, MutT, NudG) [73, 74], or phosphohydrolases belonging to a variety of families (e.g. Mesh1, TrmE, NadR, PhoA, UshA) [74, 75]. In addition, as S. aureus lacks a GppA/PPX homologue [76], there is no currently known pathway for the direct conversion of pppGpp to ppGpp.

Recent studies have revealed that (p)ppGpp binds to dozens of target proteins in E. coli, S. aureus, and other prokaryotic systems; notably small GTPases that are involved in translation or ribosome biogenesis, phosphatases, enzymes involved in (purine) nucleotide metabolism, as well as proteins involved in DNA replication and transcription [74, 7780]. Alarmone binding generally inhibits target protein activities; and binding affinities may be in the low to mid micromolar range; which suggests that cellular (p)ppGpp production to micromolar−low millimolar levels within cells may be sufficient to have widespread physiological effects [8183].

Here, we performed a detailed investigation into the biochemical activities of Sa-Rel, Sa-RelQ (SAS1) and Sa-RelP (SAS2) proteins, to further elucidate their potential roles in (pp)pGpp metabolism in S. aureus. We compared and contrasted the (pp)pGpp-synthesizing activities of Sa-RelQ and Sa-RelP with the corresponding SAS homologues encoded by S. mutans: Sm-RelQ (Smu_1046c, SAS1) and Sm-RelP (Smu_926, SAS2) [40], as well as the RelQ protein from E. faecalis (EF-RelQ; EF2671; SAS1) [41, 49, 58]. We also characterized the synthetic and hydrolytic activities of the catalytic N-terminal domain of the Sa-Rel protein (Sa-Reltrunc), whose composition is analogous to the previously-characterized truncated form of the Rel protein from S. dysgalactiae subsp. equisimilis (RelSeq NH 1–385) [16, 23] (Fig 1). Of particular note, we studied their respective abilities to synthesize and/or hydrolyze pGpp (guanosine-5’-phosphate 3’-diphosphate), as the in vivo synthesis and biological relevance of pGpp remains enigmatic [43, 49, 72].

To the best of our knowledge, the N-terminal sequence of the native Sa-Rel protein has not been formally identified. There are methionine residues encoded at the Met1 and Met8 positions of the originally described S. aureus rel gene sequence [54], respectively encoding 736 aa and 729 versions of the Sa-Rel protein. Both 736 aa and 729 aa versions are annotated in publically-available genetic sequence repositories, although the 736 aa version predominates. The biological/biochemical activities of both the 729 aa and 736 aa versions have previously been reported [26, 44, 67], and several publications have utilized numbering systems for the 729 aa version of Sa-Rel e.g. [68]. The N-terminal Met residue of the Sa-Rel and Sa-Reltrunc proteins studied here corresponds to the N-termini of other Firmicutes (class Bacilli) Rel proteins studied to date, including Rel from Staphylococcus epidermidis, B. subtilis [84], E. faecalis [58], Listeria monocytogenes [85], S. mutans [70] and S. dysgalactiae subsp. equisimilis [20] (S1 Fig).

Whilst pGpp (or a structural isomer) appears to be synthesized to detectable levels in B. subtilis cells [86, 87], its production in S. aureus, E. faecalis or other bacterial systems remains to be shown [49]. This may be due to its absence, or may reflect technical difficulties in reliably separating pGpp from GTP and other phosphorylated nucleotides using thin layer chromatography or other approaches [9]. It has also been shown that the E. coli RelA protein synthesizes small amounts of pGpp via the in situ hydrolysis of GDP/GTP, followed by pyrophosphorylation [30].

The C-terminal domain (CTD) of long-Rel (and SpoT) proteins is of pivotal importance for modulating the balance between alarmone synthesis and hydrolysis. The (p)ppGpp-synthesizing activities of the Sa-Reltrunc protein were ca. 10 to 20-fold faster and more efficient than those of the Sa-Rel protein. This finding is consistent with the CTD reducing the overall rate/efficiency of (p)ppGpp-synthesis of Sa-Rel, in the absence of heterologous regulatory factors (e.g. branched chain amino acids or ribosome activating complex), but not greatly affecting its overall substrate utilization/preference. The removal of the CTD from the bifunctional Rel proteins from M. tuberculosis (RelMtb), S. dysgalactiae subsp. equisimilis (RelSeq) and Leptospira interrogans (RelLin) has similarly led to several-fold increases in respective (p)ppGpp synthesis levels, in the absence of stimulatory factors [16, 21, 88].

Both the full-length and truncated forms of Sa-Rel exhibited a clear preference for GTP over GDP; thus being considerably more efficient for pppGpp synthesis over ppGpp (Table 1). As Sa-Rel contains an ‘RXKD’ motif within its substrate binding pocket, this motif may underlie the substrate preference for GTP over GDP, which has been observed for other Rel/SpoT proteins [16, 30].

Sa-Rel hydrolyzes all three (pp)pGpp alarmones in a Mn2+-dependent manner with high efficiency, even though it only appears to be capable of synthesizing ppGpp and pppGpp. Thus, even low-level expression of this protein should ensure effective hydrolysis of all (pp)pGpp alarmones within S. aureus cells (i.e. reducing their concentrations to basal levels). This is fully consistent with the activities of the E. faecalis Rel and C. glutamicum Rel (RelCg) proteins, which have either non-detectable or extremely low pGpp synthesis activities, and can digest all three alarmones with similar, high efficiencies [43, 49]. Taken together, it seems likely that other Rel (SpoT) homologues should also have the ability to hydrolyze pGpp, in addition to ppGpp and pppGpp.

Both the full-length and truncated Sa-Rel proteins had equivalent alarmone-hydrolyzing activities. This is consistent with previous reports, where it was noted that the deletion of the C-terminal domain of mycobacterial Rel proteins (RelMsm, RelMtb) had little effect on alarmone-hydrolyzing activities [21, 55]. However, it is in contrast to the situation for the streptococcal RelSeq protein, where the deletion of the CTD led to a more than 150-fold reduction in pppGpp-hydrolysis levels [16]. We speculate that this suggests that there is a very delicate balance between the synthase-ON/hydrolase-OFF and synthase-OFF/hydrolase-ON protein conformations. Conclusions previously reported for the S. dysgalactiae subsp. equisimilis Rel protein (RelSeq) support this proposition [16, 23].

Our data suggest that the Sa-Rel protein is notably different from the long-Rel proteins from M. tuberculosis and S. dysgalactiae [16, 20, 21 60] regarding its ability to also utilize Mg2+ ions for alarmone hydrolysis (when present at a concentration of 10 mM), albeit at rates that are many-fold less than when Mn2+ ions are present (S7 and S9 Figs). The biophysical/biochemical factors underlying this major difference remain to be established.

Our results are reasonably consistent with those recently reported by Gratani et al. [26] who noted that the deletion of the CTD of Sa-Rel led to a retention of (p)ppGpp synthesis activities (albeit at levels lower than those of the full-length Sa-Rel protein), and had negligible effects on the rate of (p)ppGpp hydrolysis. However, these authors noted considerable differences in the net balance of (p)ppGpp synthesis to hydrolysis activities when different in vivo or in vitro assays systems were utilized, highlighting the complex set of mechanisms underpinning the switch between alarmone synthesis and hydrolysis. It appears that the recombinant Sa-Rel and Sa-Rel N-terminal domain proteins studied by Gratani et al. (respectively encoded by plasmids pCG511 and pCG512) [26] start from the Met8 position, and are analogous to the proteins studied here. Secondly, it appears that these investigators did not include Mn2+ ions in the buffers used for their in vitro alarmone hydrolysis assays, which we show greatly enhances the rate of hydrolysis for Sa-Rel.

Our experimental analysis of the Sa-Rel protein has several limitations. We were unable to accurately determine the multimeric status of the recombinant Sa-Rel protein, and thus we cannot ascertain whether or not this is a key factor underpinning biochemical activities, as has been shown for E. coli RelA and RelMtb [21, 32, 56]. Secondly, all experiments were performed in vitro with purified recombinant proteins without the addition of known regulatory factors, therefore our results only provide a snapshot of Sa-Rel (and Sa-Reltrunc) activities under one set of conditions. In addition, the putative effects of the small amounts of nucleic acid that co-purified with the Sa-Rel protein (originating from the E. coli expression host) remain to be established. It was previous noted by Gratani et al. [26] that Sa-Rel proteins respectively expressed within S. aureus and E. coli host cells had notable differences in biochemical activity. This may possibly arise from differences in Rel protein expression rates or folding dynamics, or may be due to its binding (complexation) with different biomolecules within the two respective bacterial systems. These issues require further detailed investigation in the future.

During the final revision of this paper, there was a publication showing that the ribosome-associated GTPase Era and the DEAD-box RNA helicase CshA form binding interations with the Sa-Rel protein [80]. The binding of Sa-Rel had a small positive effect on the GTPase activities of Era, but had a negative effect on the helicase activities of CshA. The authors suggested that these binding interactions may modulate the role Sa-Rel plays in the maturation of 30S ribosome subunits. It remains to be established how these respective binding events affect the biochemical and biophysical activities of Sa-Rel, and how this impacts the stringent response during various physiological scenarios.

In S. aureus, the (p)ppGpp alarmones produced due to the constitutive, low-level expression of RelP and RelQ make the alarmone hydrolysis activities of Rel essential [44, 67]. Geiger et al. reported that relP and relQ were not up-regulated during glucose or iron starvation [44]. However, levels of relP and relQ, but not rel, were up-regulated in cells treated with antibiotics that induced cell-wall stress (e.g. ampicillin, vancomycin). The transcription of relP is tightly controlled by the VraS/R two-component system [89], as is relQ, but putatively to a lesser extent [44]. RelP also appears to play a more prominent role in (p)ppGpp synthesis in S. aureus, as has also been proposed to occur in S. mutans [40]. However, recent reports have suggested that there may be differences in the respective levels of induction (physiological roles) of relP and relQ in methicillin resistant versus methicillin susceptible S. aureus strains in response to treatment with beta-lactam antibiotics; and that RelQ may play a more important role in (p)ppGpp production and related beta-lactam resistance mechanisms than was previously thought [90, 91]. In E. faecalis, which encodes a RelQ protein and a bifunctional long-Rel protein, but lacks a RelP homologue, the RelQ protein is similarly up-regulated during vancomycin-induced cell stress, leading to an elevation of (p)ppGpp [41].

Sa-RelQ shares the highest levels of (aligned) amino acid (aa) identity with EF-RelQ (ca. 66%), then the RelQ protein from B. subtilis (BsRelQ, YjbM) (ca. 61%), and considerably less aa identity with Sm-RelQ (ca. 38%) (S10 Fig). Whilst kinetic parameters for alarmone synthesis by EF-RelQ and BsRelQ (YjbM) have previously been reported [4749], the biochemical activities of Sm-RelP and Sm-RelQ have only been determined in a semi-quantitative manner [40, 49]. Sm-RelP and Sm-RelQ were reported to have ppGpp and pppGpp synthesis activities that were “weaker” than those of EF-RelQ, and pGpp synthesizing abilities that were “much weaker” than those of EF-RelQ [49]. Our results here show that the unstimulated rate of ppGpp synthesis of Sm-RelQ was ca. 2-fold less than that of Sa-RelQ under analogous conditions. However, in the presence of pppGpp or ppGpp it was stimulated ca. 17-fold or ca. 5-fold, respectively, whilst the corresponding rates of ppGpp synthesis by Sa-RelP were only stimulated ca. 1.5-fold.

The kinetic parameters we report for the Sa-RelQ protein here (Table 4) correlate extremely well to those previously reported for the EF-RelQ protein, which were similarly determined in the presence of 100 μM of pppGpp or ppGpp [49]. The Km values for GDP and GTP for the BsRelQ were reported to be 1.7 ± 0.1 and 1.2 ± 0.1 mM, respectively [48]. Whilst the (alarmone-stimulated) Km(GTP) values for Sa-RelQ determined here (ca. 0.9–1.3 mM) correspond closely to that of BsRelQ, the Km(GDP) values for Sa-RelQ (ca. 0.2–0.4 mM) are ca. 5 to 8-fold lower. However, these authors showed that the addition (p)ppGpp considerably reduced the respective Km(GTP) and Km(GDP) values of BsRelQ [48].

In its crystal structure, BsRelQ form an oval-shaped homotetramer, which contains a central cleft within which two allosteric pppGpp molecules bind [65]. This has served as a reference by which structure-activity relationships have been investigated for the EF-RelQ protein [47, 48]. Sa-RelQ analogously forms a homotetramer in solution (data not shown), and thus it seems likely that it adopts a 3D fold highly similar to that of the BsRelQ protein, and is allosterically-regulated in an analogous manner. Recent reports have shown that the Sa-RelP protein also crystalizes in a homotetrameric form, containing a central cleft, which is highly similar to that of BsRelQ [46, 48]. Similar to BsRelQ, Sa-RelP also shares high levels of structural homology with the N-terminal domain of Rel from S. dysgalactiae subsp. equisimilis [23]. However, unlike BsRelQ, no (allosteric) alarmones were located within the central cleft.

In the BsRelQ structure [65], residues Lys21, Lys25, Arg28, Glu41, Phe42 and Thr44 (located within helix α1 and the β1-strand) as well as Asn148, have been shown to play pivotal roles in binding the guanine base and α−ε phosphate units of allosteric pppGpp molecules. The corresponding residues in Sa-RelQ, EF-RelQ and Sm-RelQ are highlighted in S10 Fig. Whilst these residues are highly-conserved in Sa-RelQ and EF-RelQ, they are very poorly conserved in Sm-RelQ; with the exception of Glu31. However, Sm-RelQ is still highly stimulated by (p)ppGpp alarmones. In the absence of an Sm-RelQ crystal structure, the factors underlying this observation remain to be established. The Lys25 residue of Bs-RelQ has been proposed to form an important interaction with the 5’-γ-phosphate of pppGpp, which may be responsible for pppGpp having higher stimulatory effects than ppGpp [65]. This corresponds to Lys23 in Sa-RelQ, but corresponds to Arg34 in EF-RelQ. This semi-conservative polymorphism may be related to our finding that ppGpp and pppGpp stimulate (pp)pGpp synthesis by EF-RelQ to fairly similar levels, consistent with the data reported by Gaca et al. [49].

The binding of substrate GTP/GDP (and GMP) molecules is mediated by residues within the ‘G-loop’, which connects the β3 and β3 units of BsRelQ [48]. The conformation of the G-loop has been proposed to play a key role in governing the functions of RelP and RelQ [48]. The G-loop is inherently ordered in Sa-RelP, but is disordered in Bs-RelQ. The binding of pppGpp orders the structure of the G-loop, and reorients helices α4 and α5 via the formation of a salt bridge between Arg117 and Glu178, resulting in increased (p)ppGpp synthesis via improved GTP/GDP binding. Arg 117 and Glu178 are well conserved in Sa-RelQ, EF-RelQ and Sm-RelQ (and other RelQ homologues) supporting the widespread operation of this mechanism. However, the structural factors underlying the differences in the respective allosteric effects of pppGpp, ppGpp and pGpp remain to be established.

In the recent investigation by Steinchen et al. [48], the Km(GDP) and Km(GTP) values for Sa-RelP were reported to be 300 ± 200 μM and 100 ± 100 μM, respectively. These are in very good agreement with the values determined here: Km(GDP) = 177 ± 52 μM, and Km(GTP) = 80 ± 21 μM. In the recent report by Manav et al. [46] the authors reported that the ppGpp-synthesis activities of Sa-RelP were inhibited by pppGpp and ppGpp with IC50 values of ca. 45 μM and 94 μM, respectively. This trend is consistent with our data for the Sa-RelP and Sm-RelP proteins, in that pppGpp was the most potent inhibitor of (pp)pGpp synthesis (Table 3).

Beljantseva et al. have reported that EF-RelQ binds ssRNA molecules in a sequence-specific manner [47]. The binding of certain RNA molecules [as exemplified by an RNA-oligomer that models mRNA encoding MF-dipeptide; mRNA(MF)] inhibits (p)ppGpp synthesis. The mRNA and (p)ppGpp molecules reciprocally destabilize each other’s binding to EF-RelQ. Thus, (p)ppGpp and mRNA have opposing effects on EF-RelQ activities; being stimulatory and inhibitory, respectively. This does not appear to be the case for Sa-RelQ, whose pppGpp-synthesis levels were largely unaffected by the addition of three different ssRNA oligomers. However, we cannot exclude the possibility that other RNA sequences may modulate Sa-RelQ activities. Indeed, RNA may regulate RelQ-mediated alarmone synthesis in a selective manner [92]. Manav et al. recently reported that the MF ssRNA oligomer did not affect the ppGpp-synthesis activities of Sa-RelP [46]. Our findings confirm this result, and also show that two other RNA oligomers similarly have negligible effects on the ppGpp-synthesizing activities of Sa-RelP. As noted above, whilst we cannot exclude the possibility that Sa-ReP is inhibited by RNA molecules, it seems unlikely. It has been proposed that RelP alarmone-synthesis activities may be regulated by other molecular effectors within the cell [46].

The results presented here confirm that RelP and RelQ protein homologues have notably distinct alarmone synthesis activities and alarmone-regulated properties, validating their assignments as distinct protein families [14, 40]. They highlight two key features of RelQ activities: firstly, the basal rates of alarmone synthesis exhibit large variations in magnitude; and secondly, even though the proteins share high levels of sequence similarity, they exhibit large variations in the degree to which their basal activities are allosterically stimulated by added alarmones (e.g. at 100 μM concentration). However, the precise mechanisms underlying these notable differences in biochemical preferences and properties remain to be fully elucidated.

As has been previously proposed [46, 47] RelP appears to function as an active, ‘always-ON’ source of (pp)pGpp, whilst RelQ is a more regulable source of alarmones, which can be activated: having a lower basal activity than RelP, but a considerably higher (pp)pGpp-stimulated rate, which can be dampened via the binding of ssRNA molecules. Therefore, in a ‘relaxed’ state (non-stressed, nutritionally replete conditions) the synthesis of alarmones should be driven primarily via the activities of RelP, but these should be rapidly hydrolyzed by Rel.

In summary, our results indicate that the respective sets of Rel, RelQ and RelP proteins from S. aureus, S. mutans and E. faecalis (Rel and RelQ only) play equivalent roles in (pp)pGpp metabolism, whilst they may exhibit considerable variation in their precise catalytic efficiencies, substrate specificities and allosteric regulatory properties.

Supporting information

S1 Table [pdf]
Sequences of PCR primers used.

S2 Table [pdf]
Sa-RelP and Sa-RelQ molar specific activities with/without ppGpp stimulation at pH 7.6.

S1 Fig [tiff]
Multiple sequence alignment of recombinant Sa-Rel, Sa-Reltrunc and representative long-Rel proteins from selected Firmicutes taxa.

S2 Fig [pp]
Abilities of Sa-RelQ, Sa-RelP and EF-RelQ proteins to synthesize inosine-based alarmone-like molecules (pp)pIpp.

S3 Fig [blue]
Optimal pH for (p)ppGpp synthesis by Sa-RelQ, Sa-RelP and Sa-Reltrunc.

S4 Fig [pp]
Synthesis of (p)ppGpp by Sa-Rel and Sa-Reltrunc.

S5 Fig [p]
Variation of (p)ppGpp-synthesis rates for Sa-Reltrunc with differing Mg2+ and nucleotide ratios.

S6 Fig [a]
Hydrolysis of (pp)pGpp by Sa-Rel and Sa-Reltrunc.

S7 Fig [pp]
Metal ion requirements by Sa-Rel and Sa-Reltrunc for (pp)pGpp hydrolysis.

S8 Fig [pp]
0/[S] plots used to calculate kinetic parameters for (pp)pGpp hydrolytic activities of Sa-Rel and Sa-Reltrunc.

S9 Fig [a]
Utilization of Mg2+ and Mn2+ ions by Sa-Rel for ppGpp hydrolysis.

S10 Fig [tif]
Multiple sequence alignment of RelQ proteins highlighting conserved residues of function importance.

S1 Appendix [pdf]
Supplementary methods.

S1 File [xlsx]
Experimental data.


Zdroje

1. Cashel M. The control of ribonucleic acid synthesis in Escherichia coli. IV. Relevance of unusual phosphorylated compounds from amino acid-starved stringent strains. J Biol Chem. 1969;244(12):3133–41. 4893338.

2. Cashel M, Kalbacher B. The control of ribonucleic acid synthesis in Escherichia coli. V. Characterization of a nucleotide associated with the stringent response. J Biol Chem. 1970;245(9):2309–18. 4315151.

3. Cashel M, Gallant J. Two compounds implicated in the function of the RC gene of Escherichia coli. Nature. 1969;221(5183):838–41. doi: 10.1038/221838a0 4885263.

4. Potrykus K, Cashel M. (p)ppGpp: still magical? Annu Rev Microbiol. 2008;62:35–51. doi: 10.1146/annurev.micro.62.081307.162903 18454629.

5. Braeken K, Moris M, Daniels R, Vanderleyden J, Michiels J. New horizons for (p)ppGpp in bacterial and plant physiology. Trends Microbiol. 2006;14(1):45–54. doi: 10.1016/j.tim.2005.11.006 16343907.

6. Dalebroux ZD, Svensson SL, Gaynor EC, Swanson MS. ppGpp conjures bacterial virulence. Microbiol Mol Biol Rev. 2010;74(2):171–99. doi: 10.1128/MMBR.00046-09 20508246.

7. Srivatsan A, Wang JD. Control of bacterial transcription, translation and replication by (p)ppGpp. Curr Opin Microbiol. 2008;11(2):100–5. doi: 10.1016/j.mib.2008.02.001 18359660.

8. Dalebroux ZD, Swanson MS. ppGpp: magic beyond RNA polymerase. Nat Rev Microbiol. 2012;10(3):203–12. doi: 10.1038/nrmicro2720 22337166.

9. Gaca AO, Colomer-Winter C, Lemos JA. Many means to a common end: the intricacies of (p)ppGpp metabolism and its control of bacterial homeostasis. J Bacteriol. 2015;197(7):1146–56. doi: 10.1128/JB.02577-14 25605304.

10. Hauryliuk V, Atkinson GC, Murakami KS, Tenson T, Gerdes K. Recent functional insights into the role of (p)ppGpp in bacterial physiology. Nat Rev Microbiol. 2015;13(5):298–309. doi: 10.1038/nrmicro3448 25853779.

11. Irving SE, Corrigan RM. Triggering the stringent response: signals responsible for activating (p)ppGpp synthesis in bacteria. Microbiology. 2018;164(3):268–76. doi: 10.1099/mic.0.000621 29493495.

12. Cashel M G D, Hernandez VJ, Vinella D. The Stringent Response. In: Escherichia coli and Salmonella Cellular and Molecular Biology. Neidhardt F, editor. Washington, DC: ASM Press; 1996. p. 1458–96.

13. Mittenhuber G. Comparative genomics and evolution of genes encoding bacterial (p)ppGpp synthetases/hydrolases (the Rel, RelA and SpoT proteins). J Mol Microbiol Biotechnol. 2001;3(4):585–600. 11545276.

14. Atkinson GC, Tenson T, Hauryliuk V. The RelA/SpoT homolog (RSH) superfamily: distribution and functional evolution of ppGpp synthetases and hydrolases across the tree of life. PLoS One. 2011;6(8):e23479. doi: 10.1371/journal.pone.0023479 21858139.

15. Aravind L, Koonin EV. The HD domain defines a new superfamily of metal-dependent phosphohydrolases. Trends Biochem Sci. 1998;23(12):469–72. doi: 10.1016/s0968-0004(98)01293-6 9868367.

16. Mechold U, Murphy H, Brown L, Cashel M. Intramolecular regulation of the opposing (p)ppGpp catalytic activities of Rel(Seq), the Rel/Spo enzyme from Streptococcus equisimilis. J Bacteriol. 2002;184(11):2878–88. doi: 10.1128/JB.184.11.2878-2888.2002 12003927.

17. Haseltine WA, Block R, Gilbert W, Weber K. MSI and MSII made on ribosome in idling step of protein synthesis. Nature. 1972;238(5364):381–4. doi: 10.1038/238381a0 4559580.

18. An G, Justesen J, Watson RJ, Friesen JD. Cloning the spoT gene of Escherichia coli: identification of the spoT gene product. J Bacteriol. 1979;137(3):1100–10. 374338.

19. Xiao H, Kalman M, Ikehara K, Zemel S, Glaser G, Cashel M. Residual guanosine 3',5'-bispyrophosphate synthetic activity of relA null mutants can be eliminated by spoT null mutations. J Biol Chem. 1991;266(9):5980–90. 2005134.

20. Mechold U, Cashel M, Steiner K, Gentry D, Malke H. Functional analysis of a relA/spoT gene homolog from Streptococcus equisimilis. J Bacteriol. 1996;178(5):1401–11. doi: 10.1128/jb.178.5.1401-1411.1996 8631718.

21. Avarbock A, Avarbock D, Teh JS, Buckstein M, Wang ZM, Rubin H. Functional regulation of the opposing (p)ppGpp synthetase/hydrolase activities of RelMtb from Mycobacterium tuberculosis. Biochemistry. 2005;44(29):9913–23. doi: 10.1021/bi0505316 16026164.

22. Avarbock D, Avarbock A, Rubin H. Differential regulation of opposing RelMtb activities by the aminoacylation state of a tRNA.ribosome.mRNA.RelMtb complex. Biochemistry. 2000;39(38):11640–8. doi: 10.1021/bi001256k 10995231.

23. Hogg T, Mechold U, Malke H, Cashel M, Hilgenfeld R. Conformational antagonism between opposing active sites in a bifunctional RelA/SpoT homolog modulates (p)ppGpp metabolism during the stringent response. Cell. 2004;117(1):57–68. doi: 10.1016/s0092-8674(04)00260-0 15066282.

24. Murray KD, Bremer H. Control of spoT-dependent ppGpp synthesis and degradation in Escherichia coli. J Mol Biol. 1996;259(1):41–57. doi: 10.1006/jmbi.1996.0300 8648647.

25. Battesti A, Bouveret E. Acyl carrier protein/SpoT interaction, the switch linking SpoT-dependent stress response to fatty acid metabolism. Mol Microbiol. 2006;62(4):1048–63. doi: 10.1111/j.1365-2958.2006.05442.x 17078815.

26. Gratani FL, Horvatek P, Geiger T, Borisova M, Mayer C, Grin I, et al. Regulation of the opposing (p)ppGpp synthetase and hydrolase activities in a bifunctional RelA/SpoT homologue from Staphylococcus aureus. PLoS Genet. 2018;14(7):e1007514. doi: 10.1371/journal.pgen.1007514 29985927.

27. Ronneau S, Caballero-Montes J, Coppine J, Mayard A, Garcia-Pino A, Hallez R. Regulation of (p)ppGpp hydrolysis by a conserved archetypal regulatory domain. Nucleic Acids Res. 2019;47(2):843–54. doi: 10.1093/nar/gky1201 30496454.

28. Arenz S, Abdelshahid M, Sohmen D, Payoe R, Starosta AL, Berninghausen O, et al. The stringent factor RelA adopts an open conformation on the ribosome to stimulate ppGpp synthesis. Nucleic Acids Res. 2016;44(13):6471–81. doi: 10.1093/nar/gkw470 27226493.

29. Martinez-Costa OH, Fernandez-Moreno MA, Malpartida F. The relA/spoT-homologous gene in Streptomyces coelicolor encodes both ribosome-dependent (p)ppGpp-synthesizing and -degrading activities. J Bacteriol. 1998;180(16):4123–32. 9696759.

30. Sajish M, Tiwari D, Rananaware D, Nandicoori VK, Prakash B. A charge reversal differentiates (p)ppGpp synthesis by monofunctional and bifunctional Rel proteins. J Biol Chem. 2007;282(48):34977–83. doi: 10.1074/jbc.M704828200 17911108.

31. Gentry DR, Cashel M. Mutational analysis of the Escherichia coli spoT gene identifies distinct but overlapping regions involved in ppGpp synthesis and degradation. Mol Microbiol 1996:19:1373–1384. doi: 10.1111/j.1365-2958.1996.tb02480.x 8730877

32. Gropp M, Strausz Y, Gross M, Glaser G. Regulation of Escherichia coli RelA requires oligomerization of the C-terminal domain. J Bacteriol. 2001;183(2):570–9. doi: 10.1128/JB.183.2.570-579.2001 11133950.

33. Haseltine WA, Block R. Synthesis of guanosine tetra- and pentaphosphate requires the presence of a codon-specific, uncharged transfer ribonucleic acid in the acceptor site of ribosomes. Proc Natl Acad Sci USA. 1973;70(5):1564–8. doi: 10.1073/pnas.70.5.1564 4576025.

34. Loveland AB, Bah E, Madireddy R, Zhang Y, Brilot AF, Grigorieff N, et al. Ribosome*RelA structures reveal the mechanism of stringent response activation. eLife. 2016;5. doi: 10.7554/eLife.17029 27434674.

35. Fang M, Bauer CE. Regulation of stringent factor by branched-chain amino acids. Proc Natl Acad Sci USA. 2018;115(25):6446–51. doi: 10.1073/pnas.1803220115 29866825.

36. Brown A, Fernández IS, Gordiyenko Y, Ramakrishnan V. Ribosome-dependent activation of stringent control. Nature. 2016;534(7606):277–280. doi: 10.1038/nature17675 27279228.

37. Singal B, Balakrishna AM, Nartey W, Manimekalai MSS, Jeyakanthan J, Grüber G. Crystallographic and solution structure of the N-terminal domain of the Rel protein from Mycobacterium tuberculosis. FEBS Lett. 2017;591(15):2323–2337. doi: 10.1002/1873-3468.12739 28672070.

38. Sobala M, Bruhn-Olszewska B, Cashel M, Potrykus K. Methylobacterium extorquens RSH Enzyme Synthesizes (p)ppGpp and pppApp in vitro and in vivo, and Leads to Discovery of pppApp Synthesis in Escherichia coli. Front. Microbiol. 2019;10:859. doi: 10.3389/fmicb.2019.00859 31068922.

39. Nanamiya H, Kasai K, Nozawa A, Yun CS, Narisawa T, Murakami K, et al. Identification and functional analysis of novel (p)ppGpp synthetase genes in Bacillus subtilis. Mol Microbiol. 2008;67(2):291–304. doi: 10.1111/j.1365-2958.2007.06018.x 18067544.

40. Lemos JA, Lin VK, Nascimento MM, Abranches J, Burne RA. Three gene products govern (p)ppGpp production by Streptococcus mutans. Mol Microbiol. 2007;65(6):1568–81. doi: 10.1111/j.1365-2958.2007.05897.x 17714452.

41. Abranches J, Martinez AR, Kajfasz JK, Chavez V, Garsin DA, Lemos JA. The molecular alarmone (p)ppGpp mediates stress responses, vancomycin tolerance, and virulence in Enterococcus faecalis. J Bacteriol. 2009;191(7):2248–56. doi: 10.1128/JB.01726-08 19168608.

42. Murdeshwar MS, Chatterji D. MS_RHII-RSD, a dual-function RNase HII-(p)ppGpp synthetase from Mycobacterium smegmatis. J Bacteriol. 2012;194(15):4003–14. doi: 10.1128/JB.00258-12 22636779.

43. Ruwe M, Kalinowski J, Persicke M. Identification and Functional Characterization of Small Alarmone Synthetases in Corynebacterium glutamicum. Front Microbiol. 2017;8:1601. doi: 10.3389/fmicb.2017.01601 28871248.

44. Geiger T, Kastle B, Gratani FL, Goerke C, Wolz C. Two small (p)ppGpp synthases in Staphylococcus aureus mediate tolerance against cell envelope stress conditions. J Bacteriol. 2014;196(4):894–902. doi: 10.1128/JB.01201-13 24336937.

45. Das B, Pal RR, Bag S, Bhadra RK. Stringent response in Vibrio cholerae: genetic analysis of spoT gene function and identification of a novel (p)ppGpp synthetase gene. Mol Microbiol. 2009;72(2):380–98. doi: 10.1111/j.1365-2958.2009.06653.x 19298370.

46. Manav MC, Beljantseva J, Bojer MS, Tenson T, Ingmer H, Hauryliuk V, et al. Structural basis for (p)ppGpp synthesis by the Staphylococcus aureus small alarmone synthetase RelP. J Biol Chem. 2018;293(9):3254–64. doi: 10.1074/jbc.RA117.001374 29326162.

47. Beljantseva J, Kudrin P, Andresen L, Shingler V, Atkinson GC, Tenson T, et al. Negative allosteric regulation of Enterococcus faecalis small alarmone synthetase RelQ by single-stranded RNA. Proc Natl Acad Sci USA. 2017;114(14):3726–31. doi: 10.1073/pnas.1617868114 28320944.

48. Steinchen W, Vogt MS, Altegoer F, Giammarinaro PI, Horvatek P, Wolz C, et al. Structural and mechanistic divergence of the small (p)ppGpp synthetases RelP and RelQ. Sci Rep. 2018;8(1):2195. doi: 10.1038/s41598-018-20634-4 29391580.

49. Gaca AO, Kudrin P, Colomer-Winter C, Beljantseva J, Liu K, Anderson B, et al. From (p)ppGpp to (pp)pGpp: Characterization of Regulatory Effects of pGpp Synthesized by the Small Alarmone Synthetase of Enterococcus faecalis. J Bacteriol. 2015;197(18):2908–19. doi: 10.1128/JB.00324-15 26124242.

50. Gentry D, Li T, Rosenberg M, McDevitt D. The rel gene is essential for in vitro growth of Staphylococcus aureus. J Bacteriol. 2000;182(17):4995–7. doi: 10.1128/jb.182.17.4995-4997.2000 10940046.

51. Choi MY, Wang Y, Wong LL, Lu BT, Chen WY, Huang JD, et al. The two PPX-GppA homologues from Mycobacterium tuberculosis have distinct biochemical activities. PLoS One. 2012;7(8):e42561. doi: 10.1371/journal.pone.0042561 22880033.

52. Krohn M, Wagner R. A procedure for the rapid preparation of guanosine tetraphosphate (ppGpp) from Escherichia coli ribosomes. Anal Biochem. 1995;225(1):188–90. Epub 1995/02/10. doi: 10.1006/abio.1995.1138 7778781.

53. Hardiman T, Windeisen V, Ewald JC, Zibek S, Schlack P, Rebell J, et al. In vitro synthesis and characterization of guanosine 3',5'-bis(diphosphate). Anal Biochem. 2008;383(2):337–9. doi: 10.1016/j.ab.2008.07.042 18789883.

54. Fujimura T, Murakami K. Increase of methicillin resistance in Staphylococcus aureus caused by deletion of a gene whose product is homologous to lytic enzymes. J Bacteriol. 1997;179(20):6294–301. doi: 10.1128/jb.179.20.6294-6301.1997 9335275.

55. Jain V, Saleem-Batcha R, China A, Chatterji D. Molecular dissection of the mycobacterial stringent response protein Rel. Protein Sci. 2006;15(6):1449–64. doi: 10.1110/ps.062117006 16731979.

56. Yang X, Ishiguro EE. Dimerization of the RelA protein of Escherichia coli. Biochem Cell Biol. 2001;79(6):729–36. doi: 10.1139/o01-144 11800013.

57. Schreiber G, Metzger S, Aizenman E, Roza S, Cashel M, Glaser G. Overexpression of the relA gene in Escherichia coli. J Biol Chem. 1991;266(6):3760–7. 1899866.

58. Yan X, Zhao C, Budin-Verneuil A, Hartke A, Rince A, Gilmore MS, et al. The (p)ppGpp synthetase RelA contributes to stress adaptation and virulence in Enterococcus faecalis V583. Microbiology. 2009;155(Pt 10):3226–37. doi: 10.1099/mic.0.026146-0 19608607.

59. Glasel JA. Validity of nucleic acid purities monitored by 260nm/280nm absorbance ratios. Biotechniques 1995, 18(1):62–62 7702855

60. Avarbock D, Salem J, Li LS, Wang ZM, Rubin H. Cloning and characterization of a bifunctional RelA/SpoT homologue from Mycobacterium tuberculosis. Gene. 1999;233(1–2):261–9. doi: 10.1016/s0378-1119(99)00114-6 10375643.

61. Heinemeyer EA, Richter D. Mechanism of the in vitro breakdown of guanosine 5'-diphosphate 3'-diphosphate in Escherichia coli. Proc Natl Acad Sci USA. 1978;75(9):4180–3. doi: 10.1073/pnas.75.9.4180 212739.

62. Heinemeyer EA, Geis M, Richter D. Degradation of guanosine 3'-diphosphate 5'-diphosphate in vitro by the spoT gene product of Escherichia coli. Eur J Biochem. 1978;89(1):125–31. doi: 10.1111/j.1432-1033.1978.tb20904.x 359325.

63. Sy J. In vitro degradation of guanosine 5'-diphosphate, 3'-diphosphate. Proc Natl Acad Sci USA. 1977;74(12):5529–33. doi: 10.1073/pnas.74.12.5529 414222.

64. Gajadeera CS, Zhang X, Wei Y, Tsodikov OV. Structure of inorganic pyrophosphatase from Staphylococcus aureus reveals conformational flexibility of the active site. J Struct Biol. 2015;189(2):81–6. doi: 10.1016/j.jsb.2014.12.003 25576794.

65. Steinchen W, Schuhmacher JS, Altegoer F, Fage CD, Srinivasan V, Linne U, et al. Catalytic mechanism and allosteric regulation of an oligomeric (p)ppGpp synthetase by an alarmone. Proc Natl Acad Sci USA. 2015;112(43):13348–53. doi: 10.1073/pnas.1505271112 26460002.

66. Cassels R, Oliva B, Knowles D. Occurrence of the regulatory nucleotides ppGpp and pppGpp following induction of the stringent response in staphylococci. J Bacteriol. 1995;177(17):5161–5. doi: 10.1128/jb.177.17.5161-5165.1995 7665499.

67. Geiger T, Goerke C, Fritz M, Schafer T, Ohlsen K, Liebeke M, et al. Role of the (p)ppGpp synthase RSH, a RelA/SpoT homolog, in stringent response and virulence of Staphylococcus aureus. Infect Immun. 2010;78(5):1873–83. doi: 10.1128/IAI.01439-09 20212088.

68. Gao W, Chua K, Davies JK, Newton HJ, Seemann T, Harrison PF, et al. Two novel point mutations in clinical Staphylococcus aureus reduce linezolid susceptibility and switch on the stringent response to promote persistent infection. PLoS Pathog. 2010;6(6):e1000944. doi: 10.1371/journal.ppat.1000944 20548948.

69. Mwangi MM, Kim C, Chung M, Tsai J, Vijayadamodar G, Benitez M, et al. Whole-genome sequencing reveals a link between beta-lactam resistance and synthetases of the alarmone (p)ppGpp in Staphylococcus aureus. Microb Drug Resist. 2013;19(3):153–9. doi: 10.1089/mdr.2013.0053 23659600.

70. Lemos JA, Brown TA Jr., Burne RA. Effects of RelA on key virulence properties of planktonic and biofilm populations of Streptococcus mutans. Infect Immun. 2004;72(3):1431–40. doi: 10.1128/IAI.72.3.1431-1440.2004 14977948.

71. Wendrich TM, Marahiel MA. Cloning and characterization of a relA/spoT homologue from Bacillus subtilis. Mol Microbiol. 1997;26(1):65–79. doi: 10.1046/j.1365-2958.1997.5511919.x 9383190.

72. Ruwe M, Ruckert C, Kalinowski J, Persicke M. Functional Characterization of a Small Alarmone Hydrolase in Corynebacterium glutamicum. Front Microbiol. 2018;9:916. doi: 10.3389/fmicb.2018.00916 29867827.

73. Ooga T, Ohashi Y, Kuramitsu S, Koyama Y, Tomita M, Soga T, et al. Degradation of ppGpp by nudix pyrophosphatase modulates the transition of growth phase in the bacterium Thermus thermophilus. J Biol Chem. 2009;284(23):15549–56. doi: 10.1074/jbc.M900582200 19346251.

74. Zhang Y, Zbornikova E, Rejman D, Gerdes K. Novel (p)ppGpp Binding and Metabolizing Proteins of Escherichia coli. mBio. 2018;9(2). doi: 10.1128/mBio.02188-17 29511080.

75. Sun D, Lee G, Lee JH, Kim HY, Rhee HW, Park SY, et al. A metazoan ortholog of SpoT hydrolyzes ppGpp and functions in starvation responses. Nat Struct Mol Biol. 2010;17(10):1188–94. doi: 10.1038/nsmb.1906 20818390.

76. Koonin EV. Yeast protein controlling inter-organelle communication is related to bacterial phosphatases containing the Hsp 70-type ATP-binding domain. Trends Biochem Sci. 1994;19(4):156–7. doi: 10.1016/0968-0004(94)90275-5 8016863.

77. Wang B, Dai P, Ding D, Del Rosario A, Grant RA, Pentelute BL, Laub MT. Affinity-based capture and identification of protein effectors of the growth regulator ppGpp. Nat Chem Biol. 2019;15(2):141–150. doi: 10.1038/s41589-018-0183-4 30559427.

78. Steinchen W, Bange G. The magic dance of the alarmones (p)ppGpp. Mol Microbiol. 2016;101(4):531–44. doi: 10.1111/mmi.13412 27149325.

79. Corrigan RM, Bellows LE, Wood A, Gründling A. ppGpp negatively impacts ribosome assembly affecting growth and antimicrobial tolerance in Gram-positive bacteria. Proc Natl Acad Sci USA. 2016;113(12):e1710–9. doi: 10.1073/pnas.1522179113 26951678.

80. Wood A, Irving SE, Bennison DJ, Corrigan RM. The (p)ppGpp-binding GTPase Era promotes rRNA processing and cold adaptation in Staphylococcus aureus. PLoS Genet. 2019;15(8):e1008346. doi: 10.1371/journal.pgen.1008346 31465450.

81. Mechold U, Potrykus K, Murphy H, Murakami KS, Cashel M. Differential regulation by ppGpp versus pppGpp in Escherichia coli. Nucleic Acids Res. 2013;41:6175–6189. doi: 10.1093/nar/gkt302 23620295.

82. Varik V, Oliveira SRA, Hauryliuk V, Tenson T. HPLC-based quantification of bacterial housekeeping nucleotides and alarmone messengers ppGpp and pppGpp. Sci Rep 2017;7:11022. doi: 10.1038/s41598-017-10988-6 28887466.

83. Liu K, Bittner AN, Wang JD. Diversity in (p)ppGpp metabolism and effectors. Curr Opin Microbiol. 2015;24:72–79. doi: 10.1016/j.mib.2015.01.012 25636134.

84. Wendrich TM, Marahiel MA. Cloning and characterization of a relA/spoT homologue from Bacillus subtilis. Mol Microbiol. 1997;26(1):65–79. doi: 10.1046/j.1365-2958.1997.5511919.x 9383190.

85. Okada Y, Makino S, Tobe T, Okada N, Yamazaki S. Cloning of rel from Listeria monocytogenes as an osmotolerance involvement gene. Appl Environ Microbiol. 2002;68(4):1541–7. doi: 10.1128/AEM.68.4.1541-1547.2002 11916666.

86. Nishino T, Gallant J, Shalit P, Palmer L, Wehr T. Regulatory nucleotides involved in the Rel function of Bacillus subtilis. J Bacteriol. 1979;140(2):671–9. 115847.

87. Pao CC, Gallant J. A new nucleotide involved in the stringent response in Escherichia coli. Guanosine 5'-diphosphate-3'-monophosphate. J Biol Chem. 1979;254(3):688–92. Epub 1979/02/10. 368059.

88. He P, Deng C, Liu B, Zeng L, Zhao W, Zhang Y, et al. Characterization of a bifunctional enzyme with (p)ppGpp-hydrolase/synthase activity in Leptospira interrogans. FEMS Microbiol Lett. 2013;348(2):133–42. doi: 10.1111/1574-6968.12279 24111633.

89. Kuroda M, Kuroda H, Oshima T, Takeuchi F, Mori H, Hiramatsu K. Two-component system VraSR positively modulates the regulation of cell-wall biosynthesis pathway in Staphylococcus aureus. Mol Microbiol. 2003;49(3):807–21. doi: 10.1046/j.1365-2958.2003.03599.x 12864861.

90. Bhawini A, Pandey P, Dubey AP, Zehra A, Nath G, Mishra MN. RelQ Mediates the Expression of β-Lactam Resistance in Methicillin-Resistant Staphylococcus aureus. Front Microbiol. 2019;10:339. doi: 10.3389/fmicb.2019.00339 30915038.

91. Matsuo M, Yamamoto N, Hishinuma T, Hiramatsu K. Identification of a Novel Gene Associated with High-Level β-Lactam Resistance in Heterogeneous Vancomycin-Intermediate Staphylococcus aureus Strain Mu3 and Methicillin-Resistant S. aureus Strain N315. Antimicrob Agents Chemother. 2019;63(2). doi: 10.1128/AAC.00712-18 30455230.

92. Hauryliuk V, Atkinson GC. Small Alarmone Synthetases as novel bacterial RNA-binding proteins. RNA Biol. 2017;14(12):1695–9. doi: 10.1080/15476286.2017.1367889 28820325.


Článek vyšel v časopise

PLOS One


2019 Číslo 10
Nejčtenější tento týden
Nejčtenější v tomto čísle
Kurzy Podcasty Doporučená témata Časopisy
Přihlášení
Zapomenuté heslo

Zadejte e-mailovou adresu, se kterou jste vytvářel(a) účet, budou Vám na ni zaslány informace k nastavení nového hesla.

Přihlášení

Nemáte účet?  Registrujte se

#ADS_BOTTOM_SCRIPTS#