Stringent Response-Mediated Control of GTP Homeostasis Is Required for Long-Term Viability of Staphylococcus aureus

Abstract

Staphylococcus aureus is an opportunistic bacterial pathogen that often results in difficult-to-treat infections. One mechanism used by S. aureus to enhance survival during infection is the stringent response. This is a stress survival pathway that utilizes the nucleotides (p)ppGpp to reallocate bacterial resources, shutting down growth until conditions improve. Small colony variants (SCVs) of S. aureus are frequently associated with chronic infections, and this phenotype has previously been linked to a hyperactive stringent response. Here, we examine the role of (p)ppGpp in the long-term survival of S. aureus under nutrient-restricted conditions. When starved, a (p)ppGpp-null S. aureus mutant strain ((p)ppGpp⁰) initially had decreased viability. However, after 3 days we observed the presence and dominance of a population of small colonies. Similar to SCVs, these small colony isolates (p⁰-SCIs) had reduced growth but remained hemolytic and sensitive to gentamicin, phenotypes that have been tied to SCVs previously. Genomic analysis of the p⁰-SCIs revealed mutations arising within gmk, encoding an enzyme in the GTP synthesis pathway. We show that a (p)ppGpp⁰ strain has elevated levels of GTP, and that the mutations in the p⁰-SCIs all lower Gmk enzyme activity and consequently cellular GTP levels. We further show that in the absence of (p)ppGpp, cell viability can be rescued using the GuaA inhibitor decoyinine, which artificially lowers the intracellular GTP concentration. Our study highlights the role of (p)ppGpp in GTP homeostasis and underscores the importance of nucleotide signaling for long-term survival of S. aureus in nutrient-limiting conditions, such as those encountered during infections. IMPORTANCE Staphylococcus aureus is a human pathogen that upon invasion of a host encounters stresses, such as nutritional restriction. The bacteria respond by switching on a signaling cascade controlled by the nucleotides (p)ppGpp. These nucleotides function to shut down bacterial growth until conditions improve. Therefore, (p)ppGpp are important for bacterial survival and have been implicated in promoting chronic infections. Here, we investigate the importance of (p)ppGpp for long-term survival of bacteria in nutrient-limiting conditions similar to those in a human host. We discovered that in the absence of (p)ppGpp, bacterial viability decreases due to dysregulation of GTP homeostasis. However, the (p)ppGpp-null bacteria were able to compensate by introducing mutations in the GTP synthesis pathway that led to a reduction in GTP build-up and a rescue of viability. This study therefore highlights the importance of (p)ppGpp for the regulation of GTP levels and for long-term survival of S. aureus in restricted environments.

Keywords: (p)ppGpp; GTP; Staphylococcus aureus; stringent response.

FIG 1

Survival of wild-type and (p)ppGpp⁰ strains in nutrient-rich and nutrient-restricted media. Changes in viable count (CFU/mL) were measured over 14 days for JE2 (black) and (p)ppGpp⁰ (red) in (A) TSB, (B) human serum, (C) RPMI, and (D) DMEM with l-glutamine and FBS. Survival curves were carried out in triplicate, with error bars representing standard deviation. Data were analyzed by two-way ANOVA with Šidák’s multiple-comparison test, *, P ≤ 0.05; **, P ≤ 0.01; ****, P ≤ 0.0001.

FIG 2

(p)ppGpp⁰ S. aureus form smaller colonies when starved. (A) Representative agar plates displaying JE2 and (p)ppGpp⁰ colony morphology after 14 days culture in TSB or DMEM. (B) Histograms showing the % frequency of colonies for a given size (in cm²) for JE2 and (p)ppGpp⁰ populations after 14 days in TSB or DMEM. Experiments were performed in triplicate.

FIG 3

Phenotypic analysis of S. aureus (p)ppGpp⁰ small colony isolates. (A) Growth curves of JE2, (p)ppGpp⁰, p⁰-SCI-1, p⁰-SCI-2, and p⁰-SCI-3 in TSB and DMEM. Overnight cultures were diluted to an OD₆₀₀ of 0.05 and grown for 15 h. Graphs show the mean OD₆₀₀ of three replicates, with the standard deviation shown. (B) Serial dilutions of JE2, (p)ppGpp⁰, p⁰-SCI-1, p⁰-SCI-2, and p⁰-SCI-3 were spotted onto TSA plates containing 5% sheep’s blood and incubated at 37°C for 48 h. (C) MIC values of JE2, (p)ppGpp⁰, p⁰-SCI-1, p⁰-SCI-2, and p⁰-SCI-3 for gentamicin, vancomycin, and mupirocin were determined after overnight growth of each strain in the presence of doubling dilutions of each antibiotic. OD₆₀₀ readings were determined after 24 h growth and plotted as % growth compared with the growth in the absence of antibiotic. Plots show the mean OD₆₀₀ of three replicates and their corresponding standard deviation.

FIG 4

S. aureus (p)ppGpp⁰ small colony isolates contain mutations within Gmk. (A) Schematic representation of the mutations found in Gmk. Gmk contains two domains, the nucleoside monophosphate (NMP)-binding domain in green and the lid domain in blue. The sizes of each domain are indicated by numbering. Deletions found in p⁰-SCI-1 and p⁰-SCI-3 are represented by a pink line. *, an amino acid change in the lid domain in p⁰-SCI-2. (B) GTP synthesis pathway. Gmk (brown) catalyzes the conversion of GMP to GDP, a reaction that is inhibited by (p)ppGpp. (C) The crystal structure of Gmk from S. aureus as a dimer (PDB: 4QRH), with one monomer in yellow bound to ppGpp (red). The lid domain region containing the T141I mutation that is not resolved is indicated. Regions from amino acids 187 to 196 at the dimer interface that are lacking in both p⁰-SCI-1 and p⁰-SCI-3 are highlighted in pink.

FIG 5

Gmk from the (p)ppGpp⁰ small colony isolates are less active than wild-type. (A) Coomassie stained gel of purified recombinant Gmk from wild-type JE2 and (p)ppGpp⁰ variants p⁰-SCI-2 and p⁰-SCI-3. Equal concentrations of Gmk, Gmk_T141I, and Gmk_Δ187-196 were run on 10% polyacrylamide gels under denaturing (left) or nondenaturing (right) conditions. Sizes in kDa are indicated on the left. (B) Gmk activity assay. The conversion of NADH to NAD⁺, in a Gmk-dependent manner, was monitored over time at an absorbance of 340 nm. Activity assays were performed in duplicate, with the standard deviation shown. (C) Relative cellular GTP levels. Overnight cultures of each strain were lysed and normalized by protein concentration. Relative intracellular GTP concentrations were determined via luminescence. Average luminescence values and standard deviations of triplicate experiments are plotted. Data were analyzed by one-way ANOVA with Dunnett’s multiple-comparison test, *, P ≤ 0.05; ***, P ≤ 0.001; ****, P ≤ 0.0001. (D) Changes in viable count (CFU/mL) were measured over 14 days for JE2 (black) and (p)ppGpp⁰ (red) in the presence (open symbols—decoyinine in DMSO) and absence (closed symbols—DMSO only) of decoyinine. Survival curves were carried out in duplicate, with error bars representing standard deviation.

FIG 6

Small colony morphology of (p)ppGpp⁰ variants is not stable. (A) TSA plates showing the colony size of each strain after streaking out from frozen (stock plate) or after 3 days of subculture. (B) Color map displaying colony phenotype at each time point, starting from the original plate of frozen stocks, overnight culture, and subsequent subculture for 3 days. (C) Schematic representation of the mutations found in Gmk in the p⁰-SCI revertants. Deletions found in p⁰-SCI-1r and p⁰-SCI-3r are represented by a pink line. *, SNPs resulting in an amino acid change or changes in the upstream promoter regions. (D) Relative cellular GTP levels. Overnight cultures of each strain were lysed and normalized by protein concentration before relative GTP levels were determined by luminescence. The average luminescence values and standard deviations of triplicate experiments are plotted. Data analyzed by two-way ANOVA with Šidák’s multiple-comparison test, **, P ≤ 0.01; ***, P ≤ 0.001.

FIG 7

Model of the requirement of GTP for starvation survival. For wild-type S. aureus experiencing starvation, (p)ppGpp levels increase. (p)ppGpp directly binds to Gmk and a number of other enzymes in the GTP synthesis pathway, lowering intracellular levels of GTP. GTP is a cofactor required for the activity of transcriptional repressor CodY. When GTP levels are lowered, CodY is derepressed, leading to the transcription of amino acid biosynthesis genes and survival. In the absence of (p)ppGpp, GTP levels rise when cells are starved as there is no inhibition of GTP synthesis enzymes. The bacteria respond by introducing mutations in Gmk (denoted as Gmk*) that lower activity, resulting in lower levels of GTP and survival.