Pyrimidine sufficiency is required for Sae two-component system signaling in Staphylococcus aureus

ABSTRACT

Nucleotide metabolism in pathogens is essential for their virulence, supporting their growth, survival, and immune evasion during infection. Virulence in Staphylococcus aureus is driven by the production of virulence factors that facilitate nutrient acquisition and promote immune evasion and subversion. One key virulence regulatory system is the Sae two-component system, which upregulates the production of various virulence factors. The sensor histidine kinase SaeS, a member of the intramembrane family of histidine kinases (HKs), lacks a signal-binding domain, leaving the mechanisms by which these HKs sense signals and regulate gene expression unclear. We report that pyrimidine sufficiency is essential for maintaining Sae activity. Disruption of genes involved in pyrimidine biosynthesis reduces Sae-dependent promoter activity under pyrimidine-limited conditions. Phos-tag electrophoresis confirmed that pyrimidine limitation impacts SaeS kinase activity directly. The effect of pyrimidine limitation on SaeS was abrogated in a strain producing only the catalytic domain, suggesting that pyrimidines regulate SaeS activity at the membrane. Additionally, pyrimidine limitation results in cell envelope defects, specifically increased lipoteichoic acids, and incorporation of free fatty acids into the membrane. While both cell envelope aberrations are detrimental to Sae activity, we found that removal of the accumulated free fatty acids restored Sae activity. Our study highlights the interplay between nucleotide metabolism and membrane integrity in regulating virulence factor expression through signal transduction systems in pathogens.

IMPORTANCE

Virulence is often correlated with nutrient depletion, but our understanding of this coordination is incomplete. In Staphylococcus aureus, the Sae two-component system is a major regulator of virulence factor production and secretion, but as the sensor histidine kinase SaeS lacks an obvious domain to perceive its inducing signal, basic questions surrounding how the kinase is triggered persist. Here, we aimed to investigate the mechanism by which pyrimidines act to promote the activity of the SaeS kinase in S. aureus and further expand on the importance of the roles of pyrimidines in regulating envelope biogenesis. Understanding this intersection between nucleotide metabolism and virulence regulation opens up the possibility for the development of targeted anti-virulence strategies against S. aureus infections.

INTRODUCTION

Staphylococcus aureus is a Gram-positive, opportunistic human pathogen that transiently colonizes the skin and anterior nares of approximately 30% of the human population, with 20% being persistently colonized (1–3). Furthermore, those who are stably colonized are at a higher risk for developing an infection (4). S. aureus clinical manifestations can initially present as mild to severe skin and soft tissue infections (5, 6). If left untreated, these infections can lead to more devastating invasive infections such as infective endocarditis, osteomyelitis, and sepsis (7, 8). S. aureus has emerged as one of the most devastating pathogens in healthcare. Although nosocomial infections of S. aureus are a constant threat in healthcare settings, just as concerning are infections caused by isolates circulating through the broader community (9). The relatively recent increase in community-acquired infections has been attributed in part to metabolic adaptation and antimicrobial resistance. This has led to an uptick in the number of S. aureus-related deaths over the past decade, and has led to a tremendous financial burden on the US healthcare system, resulting in up to nearly $3 billion in annual hospitalization costs (9, 10).

The survival and success of S. aureus as a pathogen fundamentally depends on its ability to align its physiology to the host environment (11–13). It does so by coordinating central metabolic processes with the production of virulence factors that serve to enhance colonization while facilitating evasion from the host immune response. The combined action of central metabolism and virulence factor production contributes to the fitness of the bacterium, which is defined by the capacity of the bacterium to adapt, replicate, and persist within the host environment (14). One of the central metabolic pathways that has emerged as a critical determinant of both bacterial fitness and virulence regulation is the de novo pyrimidine biosynthetic pathway (9, 15), which provides essential precursors for the synthesis of DNA and RNA, cell wall components, and various cellular cofactors (16). De novo pyrimidine biosynthesis is a highly conserved metabolic process from microbes to animals alike. As in other organisms, S. aureus pyrimidine synthesis begins with the synthesis of N-carbamoyl-L-aspartate. This first committed step is catalyzed by aspartate transcarbamoylase (PyrB) and utilizes carbamoyl phosphate, a common metabolite in pyrimidine and nitrogen metabolism. Successive transformations by dihydroorotase (PyrC), dihydroorotate dehydrogenase (PyrD), orotate phosphoribosyltransferase (PyrE), and uridine monophosphate synthase (PyrF) result in the formation of uridine monophosphate (UMP) (Fig. 1A) (17, 18). UMP can be further processed and converted into the different pyrimidine nucleosides (uridine, cytidine, and thymidine) and nucleobases (uracil, cytosine, and thymine) required for basic cellular processes (Fig. 1A) (17–20). The process of synthesizing pyrimidines de novo is regulated by the pyrimidine biosynthesis regulator, PyrR (21). PyrR is predicted to be a bifunctional protein that acts predominantly as a uridine-dependent transcriptional regulator of expression of the pyrimidine biosynthetic gene cluster, but also as a uracil phosphoribosyltransferase (UPRTase) (21–24). PyrR regulation occurs in a phosphoribosyl pyrophosphate (PRPP)- and UMP-dependent fashion. When intracellular levels of UMP are sufficient, PyrR binds to UMP, which enhances PyrR-mRNA-binding and promotes transcriptional attenuation of the pyrimidine biosynthetic gene cluster (9, 23). Furthermore, when UMP is low and PyrR binds PRPP, there is a drop in the binding affinity of PyrR for its mRNA targets, and PyrR acts as a UPRTase to increase intracellular UMP levels. This pathway must be properly regulated in S. aureus as pyrimidine metabolism is correlated with significant impacts on the overall fitness and virulence potential of the bacterium (9, 15, 25). Although others have shown these correlations both in vitro and in vivo, the mechanisms behind pyrimidine-mediated regulation of virulence remain unclear.

Fig 1.

Loss of pyrE results in reduced basal saeP1 promoter activity and pyrimidine auxotrophy. (A) (Top) Pyrimidine biosynthetic gene cluster of S. aureus annotated by homology to that of Bacillus subtilis. The transposon insertion identified in pyrE is indicated with an inverted red triangle. (Bottom) Current understanding of de novo and salvage pathways for pyrimidine biosynthesis in S. aureus, by analogy to the model organism B. subtilis. PyrR bifunctional pyrimidine regulatory protein, uracil phosphoribosyltransferase; CarAB, carbamoyl phosphate synthetase small and large subunits; PyrB, aspartate carbamoyltransferase; PyrC, dihydroorotase; PyrD, dihydroorotate dehydrogenase; PyrE, orotate phosphoribosyltransferase; PyrF, orotidine 5′-phosphate decarboxylase; NupC, nucleotide permease; PyrP, uracil permease; Cdd, cytidine deaminase; Pdp, pyrimidine nucleoside phosphorylase; Udk, uridine kinase; Upp, uracil phosphoribosyltransferase (16, 18, 26). (B) The indicated strains were grown in tryptic soy broth to stationary phase (16 h), at which time saeP1 promoter activity was measured. Data are plotted as relative fluorescent units (RFU [GFP/OD₆₀₀]). (C) Wild-type (WT) and ΔpyrE mutant strains were grown in chemically defined medium (CDM) ± uracil supplementation (Ura; 200 µM). Growth was monitored as an increase in absorbance at 600 nm (OD₆₀₀) and plotted over time. For panels B and C, data are representative of at least three independent experiments and plotted as mean ± SEM. For panel C, error bars are plotted for all time points; in many cases, they are too small to be seen. ****P < 0.0001 relative to WT, one-way analysis of variance with Tukey’s post-test. ns, not significant.

Production of virulence factors may incur a fitness cost. This may explain in part the fact that the expression and production of virulence factors is tightly controlled by a network of regulatory systems, including transcriptional regulators, two-component systems (TCSs), and small regulatory RNAs (27–29). These systems enable S. aureus to occupy different niches by sensing and responding to the prevailing environmental conditions in a coordinated manner, producing precisely the factors required for survival. Understanding the mechanisms behind virulence factor regulation may provide novel therapeutic targets for treating staphylococcal infections. Indeed, a growing body of research suggests so-called anti-virulence strategies offer an opportunity to mitigate the damage S. aureus can cause by lowering or blocking the ability of the cells to produce virulence factors (30–32).

One major regulator of S. aureus virulence is the SaeRS TCS. This TCS is responsible for regulating over 20 different secreted and surface-associated virulence factors that facilitate nutrient acquisition and assist in evading and subverting the immune response during infection (33). The sae (S. aureus exoprotein expression) locus was discovered in 1994 by Giraudo et al. while characterizing a transposon mutant defective in exoprotein secretion (34–37). This locus consists of an operon containing the four sae genes, saeP, saeQ, saeR, and saeS. saeR encodes the DNA-binding response regulator (RR) protein, and saeS encodes the membrane-spanning sensor histidine kinase (HK) (33). saeP and saeQ encode auxiliary proteins that have been shown to stimulate phosphatase activity of SaeS toward SaeR~P, resetting the TCS to its pre-activation state (38). The expression of these genes is driven by two promoters, P1 and P3. The P1 promoter is inducible and controls the transcription of the entire sae operon, while the P3 promoter located in saeQ drives constitutive expression of saeR and saeS (39). SaeS activity is triggered by neutrophil-produced factors, most notably human neutrophil peptides (HNP1–3) by an unknown mechanism (40). In response, SaeS autophosphorylates its conserved histidine residue. Then, SaeR is autophosphorylated on a conserved aspartate residue using the phosphohistidine of SaeS as a substrate, resulting in signal transduction (41, 42). This generates SaeR~P, which binds to site-specific sequences to promote transcription of the target genes (43, 44). One target of SaeR~P is the saeP1 promoter. This creates a positive feedback loop, resulting in signal amplification and prevents Sae activity from becoming excessive. In canonical TCSs, the HK is capable of receiving input signals through a large extracytoplasmic input domain. In contrast, SaeS lacks this domain, and the extracellular-facing portion of the protein comprises only a nine-amino-acid linker (45). As such, SaeS is a nontraditional sensor HK and is a member of the intramembrane family of sensor histidine kinases (IM-HK) (46). How SaeS responds to HNPs is an open question. Even at a more foundational level, our knowledge of how IM-HKs like SaeS function is incomplete (47–49). Herein, we report that disrupting de novo pyrimidine biosynthesis reduces Sae-dependent promoter activity under pyrimidine-limited conditions. Using Phos-tag electrophoresis, we determine that pyrimidine limitation reduces SaeS kinase activity. This effect depends on the N-terminal membrane-spanning domain of SaeS, suggesting that pyrimidines act at the membrane to promote Sae kinase activity. Consistent with known roles of pyrimidine nucleoside triphosphates in cell membrane and envelope biogenesis, we show that pyrimidine limitation results in a thicker cell wall, irregular membrane, and an accumulation of free fatty acids (FFAs). Reduction of the level of free fatty acids by providing an extracellular sink for these lipids is sufficient to restore Sae activity during pyrimidine limitation. The ripple effects generated by defects in pyrimidine metabolism highlight the critical roles that central metabolic pathways play in maintaining cellular homeostasis and virulence potential in S. aureus, making pyrimidine biosynthesis a promising target for the development of novel anti-virulence therapeutics.

RESULTS

Disruption of pyrE reduces basal Sae TCS activity and results in pyrimidine auxotrophy

As a member of the IM-HK family, little is known about how SaeS perceives its signals. To gain insight into how SaeS functions mechanistically, we conducted a transposon mutagenesis screen using a pMA15 plasmid-based system (50) to identify factors required for SaeS kinase activity. Transposon mutants of S. aureus were initially screened on tryptic soy agar (TSA) plates for changes in Sae-dependent promoter activity using a GFP transcriptional reporter fusion. Candidates were followed up using liquid growth assays. Using this screen, we isolated a transposon insertion in the gene encoding orotate phosphoribosyltransferase (pyrE), a critical enzyme for de novo pyrimidine biosynthesis (Fig. 1A). To indirectly measure Sae activity, we introduced a Sae-dependent transcriptional reporter (saeP1-gfp) into S. aureus strains and grew the cells to stationary phase (16 h) in rich, complex medium (tryptic soy broth; TSB), where the Sae system is highly active due to increased incorporation of cardiolipin into the membrane (51). At this time, resulting promoter activity was measured and corrected by optical density at 600 nm (OD₆₀₀) to account for any growth defect. As expected, we detected essentially no promoter activity in the ∆sae mutant. We measured a fourfold drop in promoter activity in the pyrE::Tn mutant relative to the parent strain (Fig. 1B). To confirm that the sae defect was associated with disruption of pyrE and to eliminate the possibility of polar effects on SAUSA300_1099 immediately downstream (annotated to encode YqzH; function unknown; depicted as 1099 in Fig. 1A), we generated a ∆pyrE in-frame deletion mutant. The ∆pyrE mutant also showed the same reduction in Sae activity (Fig. 1B). We note that the Sae system is still capable of being activated by human neutrophil peptides (Fig. S1). Consistent with what is known about the role of PyrE in the de novo pyrimidine biosynthetic pathway in other Gram-positive and Gram-negative genera, the ∆pyrE mutant is a pyrimidine auxotroph and is unable to grow in chemically defined medium (CDM) unless exogenous pyrimidines like uracil are provided (Fig. 1C) (15–19, 52). Notably, the ∆pyrE mutant exhibited a significant growth yield defect in addition to the Sae defect when grown in TSB (Fig. S2A). Although we noted a slight increase in OD₆₀₀ in the ∆pyrE mutant background at the 8 h time point during growth in CDM, we did not see any further increase when we extended the growth curve to 72 h. Furthermore, we confirmed that the cells were not continuously growing by enumerating the total number of colony-forming units (CFU) in the cultures at the same time points when OD₆₀₀ was measured (Fig. S2B and C). These data strongly suggest that pyrimidines become limiting in this medium and indicate that S. aureus relies on the de novo pathway to support growth in TSB. As expected, pyrimidine auxotrophy (or limitation) can be chemically complemented with the addition of exogenous uracil in both CDM and TSB, circumventing the need for de novo synthesis (Fig. 1C; Fig. S2A). Overall, these data suggest a link between pyrimidine sufficiency and basal Sae activity.

Pyrimidine limitation leads to a reduction in basal SaeS kinase activity

As mentioned above, pyrimidines are either synthesized by the de novo pathway as needed or taken into the cell through the pyrimidine salvage pathway to support growth. Disrupting de novo pyrimidine biosynthesis can result in a buildup of pyrimidine metabolite precursors and alter expression of the genes predicted to be involved in the de novo and salvage pathways (18, 19). We noticed that the ∆pyrE mutant accumulates white, crystalline material extracellularly when grown on TSA plates (Fig. S2D). Furthermore, ∆pyrE mutant strains of other bacterial species accumulate orotate crystals extracellularly, and in some species, this is linked to defects in virulence factor production (18, 19). Therefore, we hypothesized that a pyrimidine-specific metabolite, when it accumulates intracellularly or extracellularly, reduces Sae TCS activity. To test this hypothesis, we constructed additional mutations along the de novo pathway and measured saeP1 promoter activity as a proxy for Sae activity. When we deleted the genes annotated to encode dihydroorotate dehydrogenase (pyrD) and dihydroorotase (pyrC), we measured essentially the same low saeP1-gfp promoter activity as that measured in the pyrE mutant strains (Fig. 2A). The Sae activity defect in these strains could be complemented chemically by supplementing exogenous uracil or cytidine, or genetically in the ∆pyrD mutant by ectopically expressing a wild-type (WT) copy of pyrD (pyrD⁺ ) under the control of its native promoter (Fig. 2A; Fig. S3). Although we could have tested the effects of mutations further up the de novo pathway, we opted not to so as to avoid further interference with other biosynthetic pathways (53). Since we observed an identical reduction in Sae activity in the ∆pyrD and ∆pyrC mutant cells relative to the ∆pyrE mutant cells, we reasoned that the pyrimidine-related Sae defect is due to limitation of pyrimidine nucleotides. If so, Sae activity should be unaffected during rapid growth prior to the onset of pyrimidine limitation. To test this, we collected cells that were growing exponentially in TSB (2 h) and monitored Sae activity using saeP1-gfp. Consistent with our complementation data, both the pyrE::Tn and ∆pyrE mutants showed no Sae defect when levels of pyrimidines in the medium were not limiting (Fig. 2B). Taken together, our data suggest that the defect in basal Sae activity is not due to the buildup of metabolites, but rather due to pyrimidine limitation itself.

Fig 2.

Pyrimidine limitation reduces SaeS kinase activity. The indicated strains carrying the saeP1-gfp reporter were grown in TSB ± uracil supplementation (Ura; 200 µM) to (A) stationary phase (16 h) and (B) exponential phase (2 h), at which time saeP1 promoter activity was measured. (C–F) Measurement of intracellular SaeR species. The indicated strains were grown in TSB to exponential phase (2 h) and stationary phase (16 h), at which times cell pellets were collected, normalized by OD₆₀₀, and cell lysates were then subjected to Western blot analysis. (C and D) Representative Western blots (top and bottom) and Phos-tag Western blots (middle) of cell lysates using antibodies recognizing SaeR. (E and F) Analysis of SaeR species from Western blots using Multi Gauge software (FujiFilm) showing SaeR levels relative to WT in each condition (top), and the percentage of SaeR~P relative to the total amount of SaeR in each sample (bottom). All data are from at least three independent experiments and are plotted as mean ± SEM. ****P < 0.0001, **P < 0.01, one-way analysis of variance with Dunnett’s post-test relative to respective WT. LC, loading control; nd, below detectable limit; ns, not significant.

Unphosphorylated SaeR cannot bind to target promoters to stimulate gene transcription (44). Only activated SaeR (SaeR~P) can bind. Because there is little to no crosstalk between HKs and RRs, Sae output (i.e., target gene promoter activity) is solely dependent on the activity of SaeS (54). To test the hypothesis that pyrimidine limitation affects SaeS kinase activity directly, we used Phos-tag electrophoresis (55) and Western blotting with polyclonal antibodies raised against SaeR to measure relative levels of SaeR species in ∆pyrE mutant cells during pyrimidine-limited and pyrimidine-replete conditions. During the exponential phase, we detected similar levels of SaeR in the ∆pyrE mutant relative to WT, but we were unable to detect SaeR~P in either sample. This is consistent with previous studies reporting very low levels of Sae activity during this period (56, 57). Notably, we measured a general reduction in total SaeR protein levels and a concomitant drop in the fraction of SaeR~P in the ∆pyrE mutant cells during pyrimidine limitation (Fig. 2C through F). Given that SaeR levels are dependent in part on the activity of SaeS through SaeR~P-mediated activation of the P1 promoter, this may confound our analysis, giving us a false sense that SaeR~P levels are lower in the ∆pyrE mutant. To compare SaeS kinase activity directly without confounding autoregulation, we complemented ∆sae and ∆sae ∆pyrE mutant strains with either saeRS (+RS) ectopically expressed under the control of the constitutive P3 promoter or the empty vector (+EV) and remeasured SaeR~P as a percentage of the total SaeR in these cells. As expected, we measured no difference in total SaeR levels in ∆pyrE mutant cell lysates compared to WT cell lysates. Notably, we observed a dramatic reduction in the fraction of SaeR~P in the ∆pyrE mutant cell lysates (Fig. 3A through C). Taken together, these data indicate that pyrimidine limitation results in a reduction of SaeS kinase activity.

Fig 3.

The effect of pyrimidine limitation is independent of autoregulation. SaeR species present in SaePQRS-deficient strains (Δsae) with either saeRS complemented back under the control of the constitutive P3 promoter (+RS) or the empty vector (+EV) grown in TSB to stationary phase (16 h). Cell pellets were collected, normalized by OD₆₀₀, and lysates were subsequently subjected to Western blot analysis using polyacrylamide gels ± Phos-tag reagent. (A) Representative blots are shown. (B and C) Densitometry analysis was performed, quantifying levels of the different SaeR species present in each sample. LC, loading control; nd, below detectable limit. For all panels, data are plotted as mean ± SEM, n ≥ 3. ****P < 0.0001, ***P < 0.001, *P < 0.05, one-way analysis of variance with Tukey’s post-test. ns, not significant.

The N-terminal, membrane-spanning domain of SaeS is the site of regulation of the Sae TCS by pyrimidines

The prevailing view is that the overall conformation of the N-terminal domain is critical for controlling kinase activity. To test if the N-terminal domain hinders the kinase activity of SaeS in ∆pyrE mutant cells during pyrimidine limitation, we measured SaeS kinase activity in strains producing either full-length SaeS or just the catalytic, C-terminal kinase domain located in the cytosol (SaeSc). We found that the saeP1-gfp fusion was no longer affected by pyrimidine limitation. In fact, we noted a significant increase in promoter activity when SaeSc is produced (Fig. 4A). Consistent with this observation, overall SaeR levels were increased in addition to the ratio of SaeR~P/SaeR species in the ∆pyrE mutant during limitation (Fig. 4B and C). These data indicate that pyrimidine limitation specifically targets the N-terminal membrane-spanning portion of SaeS, reducing kinase activity.

Fig 4.

SaeS variants lacking the N-terminal domain function independent of pyrimidine limitation. (A) Measurement of saeP1-gfp promoter activity in stationary phase cultures (16 h) with either EV, full-length SaeS (+RS), or the cytosolic, catalytic domain of SaeS (+RSc). <, fluorescence was undetectable relative to blank containing phosphate-buffered saline. (B) SaeR species present in SaePQRS deletion strains (Δsae) with the integrated SaeRSc construct expressed under the constitutive P3 promoter (+RSc) were grown in TSB to stationary phase (16 h). Cell pellets were then collected, normalized by OD₆₀₀, and lysates were then subjected to Western blot analysis using polyacrylamide gels (top, bottom) ± Phos-Tag reagent (middle). (C) Densitometry showing levels of the different SaeR species present in each sample. LC, loading control; nd, below detectable limit. All data represent at least three independent experiments and are plotted as mean ± SEM. Statistics: (A) ****P < 0.0001, ns, not significant, one-way analysis of variance with Tukey’s post-test; (C) *P < 0.05, unpaired Welch’s t-test relative to +RSc.

Pyrimidine limitation significantly alters the transcriptome of ∆pyrE mutant cells

Given that the Sae defect in the ∆pyrE mutant cells during pyrimidine limitation is associated with the N-terminal membrane-spanning domain of SaeS, we wondered if pyrimidine limitation is generally linked to a defect in the envelope composition. If this is true, then the effect of pyrimidine limitation likely extends beyond Sae to many proteins in the membrane, several of which are kinases for other TCSs. To determine the full extent to which pyrimidine limitation alters gene expression and help inform a potential mechanism for Sae activation, we performed transcriptional profiling in wild-type and isogenic ∆pyrE mutant cells grown to early stationary phase (i.e., pyrimidine-limiting conditions). Overall, we noticed that the transcriptome was dramatically altered in the ∆pyrE mutant cells during limitation, with 701 genes significantly downregulated and 719 genes significantly upregulated (log₂ fold change ≥1 or ≤-1 and P < 0.05; ∆pyrE/WT) (Data Table S1). This dramatic shift in the transcriptional profile of the ∆pyrE mutant likely stems from the interconnectedness of pyrimidine metabolism with other central metabolic pathways and various regulators in the cell (13, 17). To determine whether pyrimidine limitation affects specific pathways preferentially, we performed Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis on the RNA-Seq output using KEGGREST with the USA300_FPR3757 KEGG reference genome (KEGG;saa) (58, 59). Based on KEGG enrichment, the majority of differentially expressed genes (DEGs) annotated to be involved in 80 different KEGG pathways were downregulated in the ∆pyrE mutant cells during pyrimidine limitation (Data Table S2). This includes directly regulated pathways such as purine metabolism, various pathways involved in cell envelope biogenesis (e.g., fatty acid biosynthesis, teichoic acid biosynthesis, and peptidoglycan biosynthesis), and the various sugar-related metabolic pathways. Additionally, KEGG analysis also indicated that 46 out of the 53 DEGs encoding proteins annotated to be involved in TCS signaling are downregulated in the ∆pyrE mutant relative to WT during pyrimidine limitation. This led us to hypothesize that pyrimidine limitation generally affects the activity of the other TCSs in the ∆pyrE mutant cell. To test this hypothesis, we took advantage of the fact that TCSs exhibit feedback autoregulation of the HK and RR to indirectly assess the overall impact on kinase activity of the other 15 TCSs (60, 61). We saw that, in addition to the log₂ fold drop in saeRS transcripts, there was a general reduction in the expression of the other TCS coding genes in the ∆pyrE mutant cells during pyrimidine limitation (Table 1). Interestingly, the Agr quorum-sensing TCS had the highest reduction in expression relative to the other systems with a 5- to 6-log₂ fold drop in agrCA transcript abundance. Furthermore, 10 of the 14 genes previously annotated to be involved in the de novo purine biosynthetic pathway and all of the genes annotated for the different subunits of the ATP synthase were significantly downregulated in the ∆pyrE mutant during pyrimidine limitation (Table 2; Data Table S2). This suggests that the cell is reducing purine production in response to pyrimidine limitation to maintain the balance of nucleotide pools within the cell. In addition, many of the genes annotated to be involved in nucleotide salvage were upregulated in the ∆pyrE mutant cells (Table 2). Collectively, these data show that defective pyrimidine metabolism results in a global change to the transcriptional profile of S. aureus when pyrimidines are limited, including those involved in the synthesis of envelope components. Furthermore, transcriptomic analysis suggests that pyrimidine limitation generally reduces the kinase activity of the majority of TCSs in ∆pyrE mutant cells.

TABLE 1.

Pyrimidine limitation negatively affects expression of most two-component systems in the ΔpyrE mutant^a

Gene	Log2 fold change (ΔpyrE/WT)	-log10P	Adjusted P-value
saeS	−2.63	6.8E − 19	0.00
saeR	−2.07	7.8E − 07	0.00
agrC	−5.03	1.2E − 307	0.00
agrA	−6.15	1.2E − 307	0.00
airS	1.35	9.8E − 08	0.00
airR	0.24	5.6E − 01	1.00
vraS	−1.08	1.2E − 04	0.00
vraR	−1.70	2.0E − 15	0.00
phoR	−2.18	5.0E − 16	0.00
phoP	−2.06	2.7E − 16	0.00
desK	−0.36	3.4E − 01	1.00
desR	−0.85	1.4E − 03	0.00
arlS	−1.71	1.6E − 10	0.00
arlR	−0.87	4.5E − 03	0.01
braS	−1.21	7.8E − 05	0.00
braR	−0.25	4.8E − 01	0.87
graS	−1.25	1.5E − 06	1.00
graR	−0.16	6.0E − 01	1.00
hptS	−1.39	1.8E − 10	1.00
hptR	−2.32	2.1E − 45	1.00
hssS	0.44	1.0E − 01	1.00
hssR	2.06	1.4E − 14	1.00
kdpD	−3.17	1.4E − 161	1.00
kdpE	−3.22	3.9E − 187	1.00
lytS	0.78	7.5E − 03	1.00
lytR	0.00	9.7E − 01	1.00
nreB	−1.90	1.9E − 14	0.00
nreC	−2.40	4.5E − 44	0.00
srrB	0.23	3.0E − 01	0.58
srrA	−0.56	4.5E − 02	0.10
walK	−0.65	3.7E − 02	0.08
walR	0.84	4.3E − 03	0.01

^a RNA-Seq analysis results quantifying the fold change in expression of TCS HK and RR genes in the ΔpyrE mutant relative to the WT after 8 h of growth in TSB (pyrimidine-limiting condition for ΔpyrE mutant cells). Raw reads were trimmed using Trimmomatic. Trimmed reads were aligned and analyzed based on the reference genome (USA300 FPR3757; NC007793) using Rockhopper. P-values were determined using a negative binomial distribution, and adjusted P-values were determined using the Benjamini-Hochberg procedure built into the script. The full data set can be found in Data Table S1.

TABLE 2.

Pyrimidine limitation results in differential expression of genes involved in purine and pyrimidine biosynthesis in the ΔpyrE mutant^a

Gene	Log2 fold change (ΔpyrE/WT)	−log10P	Adjusted P-value
Pyrimidine biosynthesis and salvage
nupC	1.12	3.08E − 04	0.00
pyrR	4.47	1.15E − 307	0.00
pyrP	3.82	9.07E − 306	0.00
pyrB	1.91	1.50E − 16	0.00
pyrC	0.69	1.56E − 01	0.32
carA	0.16	5.38E − 01	0.95
carB	−0.65	1.48E − 01	0.30
pyrF	−1.23	8.51E − 05	0.00
pyrE	nd	1.15E − 307	0.00
pyrH	1.63	1.49E − 10	0.00
cdd	−0.80	5.79E − 03	0.01
udk	0.99	7.62E − 04	0.00
upp	−2.00	1.02E − 12	0.00
pyrG	−0.70	6.45E − 02	0.14
pdp	−1.30	1.49E − 04	0.00
pyrD	3.79	1.15E − 307	0.00
Purine biosynthesis and ATP synthesis
purA	−0.31	5.83E −01	1.00
purR	−0.61	4.38E −02	0.10
purE	−1.16	8.60E −05	0.00
purK	−2.56	4.73E − 32	0.00
purC	−3.63	2.08E − 255	0.00
purS	−4.27	1.15E − 307	0.00
purQ	−5.37	1.15E − 307	0.00
purL	−6.91	1.15E − 307	0.00
purF	−7.76	1.15E − 307	0.00
purM	−8.16	1.15E − 307	0.00
purN	−8.63	1.15E − 307	0.00
purH	−7.43	1.15E − 307	0.00
purD	−7.51	1.15E − 307	0.00
purB	0.34	4.57E − 01	0.83
atpC	−5.10	1.15E − 307	0.00
atpD	−4.83	1.15E− 307	0.00
atpG	−4.99	1.15E − 307	0.00
atpA	−4.38	1.15E − 307	0.00
atpH	−3.86	1.15E − 307	0.00
atpF	−3.48	1.27E − 185	0.00
atpA	−4.38	1.15E − 307	0.00
atpE	−2.59	1.27E − 46	0.00
atpB	−1.87	1.04E − 09	0.00

^a RNA-Seq analysis results quantifying the fold change in expression of annotated purine and pyrimidine biosynthesis genes as well as the annotated ATP synthase genes in the ΔpyrE mutant relative to the WT after 8 h of growth in TSB (pyrimidine-limiting condition for ΔpyrE mutant cells). Raw reads were trimmed using Trimmomatic, and trimmed reads were aligned and analyzed based on the reference genome (USA300 FPR3757; NC007793) using Rockhopper. P-values were determined using a negative binomial distribution, and adjusted P-values were determined using the Benjamini-Hochberg procedure built into the script. nd, not detected in the ∆pyrE mutant background. The full data set can be found in Data Table S1.

Pyrimidine limitation is associated with cell wall thickening and an unstable membrane

Up to this point, we have shown that SaeS integrates pyrimidine sufficiency to regulate basal kinase activity via the membrane-spanning domain. However, how pyrimidine sufficiency is perceived by SaeS mechanistically remains unknown. Because pyrimidine nucleotides are used to activate precursors for peptidoglycan, teichoic acid, and phospholipid synthesis, and that genes that direct the synthesis of these membrane constituents are differentially expressed, we reasoned that during pyrimidine limitation, a modified cell envelope might contribute to the defect in SaeS kinase activity. We determined the minimum inhibitory concentration (MIC) of molecules that target synthesis of major cell envelope components (62). We measured no change in susceptibility to fosfomycin (first committed step in peptidoglycan synthesis), tunicamycin (wall teichoic acid synthesis), or oxacillin (peptidoglycan crosslinking) (Table S1) (62–65). In contrast, we observed a four- to eightfold increase in the susceptibility to Congo red, with a change in MIC from 0.25% in WT cells to 0.06%–0.03% in ∆pyrE mutant cells. Congo red was previously reported to be an inhibitor of LtaS (lipoteichoic acid [LTA] synthase) that is critical for LTA assembly (66). To better understand the change in susceptibility, we extracted lipoteichoic acids from WT and ∆pyrE mutant cells. Surprisingly, we found that the ∆pyrE mutant has increased production of LTAs (Fig. S4A and B). Previously, uncontrolled LTA synthesis in the absence of the small membrane regulatory protein MspA was shown to result in virulence defects (67, 68). To determine whether pyrimidine limitation acts in the same regulatory pathway as MspA, we performed a genetic epistasis experiment. The mspA::Tn single mutant showed essentially the same SaeS kinase activity defect as the ∆pyrE mutant, as judged by Phos-tag experiments. No further reduction in kinase activity was measured in the ∆pyrE mspA::Tn double mutant. The kinase defect was complemented when a wild-type copy of mspA (mspA⁺) was provided in trans in the mspA::Tn single mutant, but not in the ∆pyrE mspA::Tn double mutant, suggesting that the effects on Sae activity are independent from one another (Fig. S4C). We also overexpressed mspA⁺ in WT and ∆pyrE mutant cells and measured the effect on Sae activity using similar Phos-tag experiments (Fig. S5). Interestingly, we saw an enhancement of SaeS kinase activity when we overexpressed mspA⁺ in the WT and mspA::Tn mutant background; however, there was no effect of overexpressing mspA⁺ in the ∆pyrE mutant background (Fig. S5). These data indicate that the effect of pyrimidine limitation on Sae activity in the ∆pyrE mutant background is independent of the role of MspA.

To get a better understanding of the impact of pyrimidine limitation on the cell envelope, we performed transmission electron microscopy (TEM) on WT and ∆pyrE cells, and ∆pyrE mutant cells fed exogenous uracil. Multiple observations are consistent with an abnormal cell envelope. First, ∆pyrE mutant cells experiencing pyrimidine limitation are nearly twice as large as WT cells; this phenotype is complemented when the missing nutrient is provided. Second, ∆pyrE mutant cells under limitation conditions have a much thicker cell envelope and a visibly crenulated membrane architecture relative to WT or the chemically complemented cells (Fig. 5). Increased envelope thickness in the ∆pyrE mutant is consistent with the idea that defects in central metabolism and/or increased LTA biosynthesis result in thickening of the cell envelope (15, 64, 67). In addition to the dramatic difference in membrane morphology between WT and pyrimidine-limited (∆pyrE mutant) cells, ∆pyrE mutant cells accumulated more of the eponymous staphyloxanthin pigment compared to wild-type cells (Fig. S6A). Because increased incorporation of staphyloxanthin in the membrane has been shown to rigidify the membrane (69, 70), we wondered if increased staphyloxanthin incorporation was restraining SaeS kinase activity. To test this, we utilized an available transposon mutant in crtM encoding the 4,4′-diapophytoene synthase that is critical for staphyloxanthin synthesis (crtM::Tn) and measured saeP1-gfp promoter activity in the crtM::Tn single mutant and the ∆pyrE crtM::Tn double mutant. We found that disrupting crtM had no effect on Sae activity in either genetic background, suggesting that the increased staphyloxanthan accumulation did not explain the Sae signaling defect (Fig. S6B). Rather, the increased staphyloxanthin production may be in response to irregularities in the cell membrane and an attempt to reinforce an unstable membrane. Taken together, these data suggest that pyrimidine sufficiency is correlated with proper cell envelope biogenesis.

Fig 5.

Unlike wild-type cells, ΔpyrE mutant cells are much larger, have a thickened cell wall, and have a crenulated membrane. Representative TEM images of the (A) WT, (B) ΔpyrE mutant, and (C) ΔpyrE mutant chemically complemented with uracil (Ura; 200 µM). Inset images emphasize the crenulated cell membrane architecture of the ΔpyrE mutant. (D) Cell diameter measurements across individual cells in the TEM images. (E) Measurements of cell envelope thickness from cell membrane to cell wall edge with one measurement per individual cell indicated below the dot plot. All measurements of the indicated strains were acquired in Fiji/ImageJ as described in Materials and Methods. Data are plotted as mean ± SEM from three biological replicates, with bars indicating the mean thickness and error bars indicating SEM. White arrowheads point to the membrane; black arrowheads point to cell wall, ****P < 0.0001, one-way analysis of variance with Tukey’s post-test. ns, not significant.

Pyrimidine sufficiency promotes SaeS kinase activity by reducing free fatty acid levels in the membrane

Cytidine nucleosides play a central role in lipid synthesis in prokaryotes and eukaryotes (71). Phosphatidate cytidylyltransferase (CdsA) condenses phosphatidic acid with cytidine triphosphate (CTP) to form activated diacylglycerol (CDP-diacylglycerol; CDP-DAG) with the elimination of pyrophosphate. Displacement of CMP by glycerol phosphate and cleavage of the phosphate monoester yield phosphatidylglycerol (72, 73). If CTP is limited and CdsA activity is reduced, then we would expect that there would be an increase in unincorporated FFAs that do not get incorporated into the membrane. Based on previous studies, FFAs inhibit the Sae TCS (74, 75). To test if FFAs may be responsible for the SaeS activity defect, we grew WT and ∆pyrE mutant cells to stationary phase (16 h) in the presence of fatty acid-free bovine serum albumin (BSA) and measured saeP1 promoter activity. BSA binds and removes FFAs from the cell membrane, providing an extracellular sink for these lipids and reviving Sae TCS activity (48, 74–76). As expected, BSA treatment resulted in a modest increase in Sae activity in WT cells. Surprisingly, BSA treatment restored Sae activity to WT levels in the ∆pyrE mutant (Fig. 6A). We note that BSA addition did not restore pyrimidine auxotrophy of the mutant when grown in CDM (data not shown). A twofold increase in BSA concentration had no further effect on Sae activity, suggesting that all of the excess FFAs had been depleted (Fig. S7A). LC-MS/MS analysis of supernatant samples revealed differences in the abundances of specific FFAs recovered using BSA treatment (Fig. 6B). Interestingly, we noticed that there was a dramatic increase in the abundance of 13:0 FFAs and a moderate increase in the abundance of 15:0, 17:0, 18:0, 20:0, and 22:0 FFAs recovered from the media from ∆pyrE mutant cultures treated with BSA, where Sae activity is restored, relative to the untreated cells (Fig. 6A and B). Furthermore, there was a moderate release of 13:0, 20:0, and 24:0 FFA species in the WT cells treated with BSA that is correlated with the increase in Sae activity relative to the untreated control (Fig. 6B). Of the FFAs identified, the 13:0 species seemed to be the highest FA in abundance in the media from both the WT and ∆pyrE mutant cells treated with BSA. However, it is unclear if 13:0 is limiting Sae activity. Even though LC-MS/MS allows us to identify FFA species, we cannot distinguish between different isoforms of the fatty acids detected. Given that anteiso branched-chain fatty acids derived from isoleucine (a15:0) promote Sae activity (48, 49), we wondered if adding this fatty acid would rescue Sae activity in the ∆pyrE mutant, as it is dramatically reduced in the supernatants of ∆pyrE mutant cultures during limitation. As expected, exogenous a15:0 restored Sae activity in a strain depleted for this membrane fatty acid (lpdA::Tn mbcS1) (48, 49, 77). To our surprise, the addition of exogenous a15:0 led to a further reduction of Sae activity in the ∆pyrE mutant cells (Fig. S7B). There was no additional effect of a15:0 addition on growth. These data support the idea that the SaeRS defect seen in ∆pyrE mutant cells stems from FFA accumulation in the membrane due to a possible pyrimidine limitation-induced bottleneck during phospholipid biosynthesis. Adding exogenous a15:0 exacerbates the phenotype.

Fig 6.

Treatment with bovine serum albumin promotes Sae activity and promotes release of free fatty acids during pyrimidine limitation. (A) The indicated strains carrying the saeP1-gfp reporter were grown in TSB ± BSA (10 mg/mL) to stationary phase, at which time saeP1 promoter activity was measured. Data are plotted as mean ± SEM. (B) Culture supernatants from the indicated strains were collected, FFAs were extracted, and extracts were subjected to LC-MS/MS analysis. Data were normalized using an internal standard (d4-palmitic acid), and mean fold change (FC) in abundances of FFAs relative to untreated cell supernatants is depicted using a heat map. All data are from at least three biological replicates, ****P < 0.0001, one-way analysis of variance with Tukey’s post-test. ns, not significant.

DISCUSSION

This study aligns with the idea that nucleotide metabolism is an important factor for regulating virulence in S. aureus and other pathogens. Previous work from our lab and others has shown that mutations in de novo purine biosynthesis are correlated with altered expression of virulence genes (78–80). Specifically, disrupting GMP synthase (GuaA) results in increased activity of the AgrCA TCS and enhanced protease activity in vitro. Furthermore, it has been shown that the purine biosynthesis regulatory protein, PurR, can moonlight as a virulence regulator and represses expression of fibronectin-binding proteins and extracellular toxins in vivo and in vitro (78, 81). On the other hand, disrupting de novo pyrimidine biosynthesis is correlated with a reduction in expression of virulence factors in vitro and in vivo. One study specifically looked at the impact of inactivating thymidylate synthase (ThyA), which is responsible for the conversion of deoxy uridine monophosphate into thymidylate (15). They report that defective thymidylate synthesis is associated with a dramatic reduction in S. aureus virulence in vitro and in vivo (15). Regardless of the effects of purine and pyrimidine metabolism on enhancing or reducing virulence in vitro, S. aureus needs nucleotides to support and maintain proper envelope homeostasis and the various cellular processes for proliferation within the host environment. This dual role of nucleotide metabolism in supporting both staphylococcal fitness and virulence makes this aspect of bacterial physiology a particularly attractive target for the development of therapeutics. For instance, using an inhibitor that targets the de novo purine or pyrimidine biosynthetic pathways would significantly hinder bacterial fitness and put the bacterium in a situation where the virulence regulatory systems could not function properly, ultimately leading to host clearance. Although purine and pyrimidine metabolism are both important for virulence potential, there is little mechanistic information about how these critical pathways contribute to virulence regulation. Here, we propose a model where pyrimidines act indirectly on the activity of the SaeRS TCS in S. aureus by maintaining membrane homeostasis (Fig. 7). Importantly, this study links membrane homeostasis to virulence regulation and offers insight into how these IM-HKs, such as SaeS, are regulated.

Fig 7.

Working model for proposed mechanism by which pyrimidines promote SaeS kinase activity. During cellular growth, pyrimidine sufficiency dictates whether S. aureus needs to turn on the de novo pyrimidine biosynthetic pathway. Pyrimidines promote SaeS kinase activity, in part, by promoting and maintaining cell envelope and membrane homeostasis. Under pyrimidine-limited conditions where S. aureus cannot synthesize pyrimidines de novo, cell envelope biosynthesis goes unchecked, resulting in increased LTA production and a thickening of the cell wall. Substrate limitation for CdsA results in a bottleneck in phospholipid biosynthesis. This leads to an unstable membrane due to the accumulation of phospholipid precursors, including free fatty acids, that act to reduce SaeS kinase activity. This figure was generated using Biorender.com.

To determine the mechanistic basis for the role of pyrimidine sufficiency in regulating the activity of the SaeRS TCS, the first task was to distinguish between metabolite accumulation and pyrimidine limitation as the root cause. Based on what is known about the role of orotate in regulating virulence in other organisms (19, 82), we initially hypothesized orotate or other pyrimidine metabolites were antagonizing Sae activity. Although we observed orotate crystal formation in the ∆pyrE mutant colonies on solid medium, we determined that the defect was not due to orotate accumulation, as the ∆pyrD and ∆pyrC mutants, which do not accumulate orotate and dihydroorotate, phenocopy the ∆pyrE mutant (Fig. 2A). Further, Sae activity is unaffected in pyrimidine-replete conditions, consistent with the hypothesis that the defect is caused by pyrimidine limitation (Fig. 2B). We note that we cannot exclude the possibility that providing exogenous pyrimidines represses the pyr genes, preventing the accumulation of an inhibitory metabolite upstream of PyrC in the pathway. Moreover, it has been shown that purine and pyrimidine metabolism directly influences the activity of their respective pathways primarily through feedback inhibition of the regulatory proteins, PurR and PyrR. Although these two proteins have distinct regulatory mechanisms, they both function by sensing final or intermediate products of the pathways to control the expression of genes involved in nucleotide biosynthesis, thereby maintaining nucleotide balance within the cell (83). Given that feedback inhibition occurs, we were not surprised to see that most genes involved in purine biosynthesis were downregulated, while the genes encoding putative proteins involved in the de novo pyrimidine biosynthetic pathway (pyrB, pyrC, and pyrD) were upregulated in the ∆pyrE mutant during pyrimidine limitation (Table 2; Data Table S2; Fig. 1A). This underscores the influence that each respective biosynthetic pathway has on one another.

Our findings suggest pyrimidines affect SaeS through its N-terminal membrane-spanning domain, which is consistent with previous work showing IM-HKs, like SaeS, are sensitive to membrane composition changes (48, 49, 51, 74, 75). This is evidenced by the lack of a Sae signaling defect during pyrimidine limitation when SaeSc is produced (Fig. 3 and 4). In addition to linking the Sae phenotype to the membrane, this experiment ruled out the possibility that reduced ATP pools were to blame for the low Sae activity. We also note that SaeR protein abundance in the ∆sae ∆pyrE mutant cells complemented with the saeP3-saeRSc construct was increased compared to the new WT control (∆sae + RSc). Several possibilities could explain this unexpected result. First, the difference in expression may be due to the expression of these genes at a different locus. Second, this may be due to increased stability of SaeSc and SaeR in the ∆pyrE mutant background, possibly due to reduced activity of negative Sae regulators SaeP and SaeQ or proteases, similar to the direct and indirect effects of FtsH and ClpP on the SaeS^L18P allele in the Newman background of S. aureus (50, 84). Third, the ∆sae single mutant expressing the saeP3-saeRSc (i.e., the new WT) construct is the anomalous result, and the heightened activity seen in the ∆pyrE mutant is the expected result based on previous work showing heightened Sae activity in the otherwise WT background (45). Regardless of the expression and stability of the proteins, there is still a dramatic enhancement in the overall kinase activity of SaeSc when expressed in the ∆pyrE mutant background (Fig. 4). Further work is needed to determine the root cause of this difference in expression profiles of the P3-saeRSc construct.

We observed dramatic changes to the cell wall architecture when pyrimidines are limited. Our observations are consistent with other reports that show defects in central metabolism result in an altered cell envelope (15, 48, 49, 64, 77, 80). Since the MICs for envelope-targeting drugs and compounds remained unchanged beyond that of Congo red in ∆pyrE mutant cells under limitation, and the SaeS kinase defect is linked to its N-terminal membrane domain, our findings suggest that membrane alterations create an environment unfavorable for full SaeS kinase activity. Previous studies showed that unchecked LTA synthesis results in a virulence defect but did not measure the activities of the major virulence regulatory systems AgrCA and SaeRS (67, 68). Both the LTA and phospholipid biosynthetic pathways share diacylglycerol as a common intermediate. Our data suggest that the accumulation of LTA in the ∆pyrE mutant may be the result of diverting metabolic flux from phospholipid synthesis to LTA synthesis (Fig. S4) (85). Accumulation of LTA in the ∆pyrE mutant background may help the cells compensate for the defect in phospholipid biosynthesis as LTAs serve important roles in processes associated with the envelope including ion homeostasis, maintaining proton motive force, as well as regulating autolysin activity (86, 87). This potential compensatory effect may explain the unexpected increased susceptibility of the ∆pyrE mutant to Congo red relative to the WT cells under pyrimidine-limiting conditions. Furthermore, we also discovered that increased LTA synthesis is associated with a defect in SaeS kinase activity. This is consistent with a prior study that reported the link between MspA- and Sae-dependent virulence factors (68). However, during pyrimidine limitation, increased LTA synthesis does not seem to be responsible for the Sae defect (Fig. S4). In addition, overexpressing the LTA biosynthesis regulator MspA results in increased SaeS kinase activity (Fig. S5). Our observation that LTAs affect Sae is nevertheless important to note. However, more work is needed to dissect the connection between LTAs and SaeS.

MATERIALS AND METHODS

Bacterial strains and growth conditions

Unless specified otherwise, all Staphylococcus aureus strains used in this study are derivatives of the community-associated, methicillin resistant S. aureus (MRSA) strain USA300 LAC cured of the native plasmid pUSA03 that confers erythromycin resistance and are listed in Table S2. Unless otherwise noted, indicated S. aureus strains were cultured in TSB containing 0.25% (wt/vol) dextrose (BD Biosciences) or CDM (pH 6.5) lacking pyrimidine nucleotides as previously described. Briefly, CDM was formulated with alanine (672 µM), arginine (287 µM), aspartic acid (684 µM), cysteine (166 µM), glutamic acid (680 µM), glycine (670 µM), histidine (129 µM), isoleucine (228 µM), leucine (684 µM), lysine (342 µM), methionine (20 µM), phenylalanine (240 µM), proline (690 µM), serine (285 µM), threonine (260 µM), tryptophan (50 µM), tyrosine (275 µM), valine (684 µM), thiamine (56 µM), nicotinic acid (10 µM), biotin (0.04 µM), pantothenic acid (2.3 µM), MgCl₂ (1,000 µM), CaCl₂ (100 µM), monopotassium phosphate (40,000 µM), dipotassium phosphate (14,700 µM), sodium citrate dehydrate (1,400 µM), magnesium sulfate (400 µM), ammonium sulfate (7,600 µM), and glucose (27,753 µM) (88). Escherichia coli strains were grown in lysogeny broth without glucose (10 g/L tryptone, 5 g/L yeast extract, and 5 g/L sodium chloride) (89). When necessary, media were solidified with agar (1.5% [wt/vol]; TSA) and supplemented with antibiotics or chemicals at the indicated concentrations: ampicillin, 100 µg/mL: chloramphenicol, 5 to 10 µg/mL; erythromycin, 5 µg/mL; tetracycline, 1.5 µg/mL; kanamycin, 100 µg/mL; BSA, 10 mg/mL. When indicated, media were supplemented with exogenous pyrimidine (uracil) to a final concentration of 200 µM. Unless otherwise noted, all strains were grown at 37°C.

TSB growth curve

Unsupplemented overnight cultures were used to inoculate 125 mL DeLong flasks containing 12.5 mL TSB ± uracil supplementation (200 µM; 10:1 flask: medium ratio) to an OD₆₀₀ of 0.05 and incubated in a water bath shaking at 250 rpm. Cell growth was monitored over either 1 h or indicated increments by measuring increase in OD₆₀₀ using an Amersham Ultraspec 2100 Pro UV-visible spectrophotometer. When necessary, colony-forming units (CFU/mL) were determined by serially diluting resulting cultures in sterile 1× phosphate-buffered saline (PBS) followed by plating on TSA plates.

CDM grown curve

Overnight cultures grown in CDM ± uracil supplementation (200 µM) were used to inoculate a 96-well plate to an OD₆₀₀ of 0.05. To prevent uracil carryover from overnight cultures, 1 mL of cells was pelleted (13,000 × g for 1 min) and supernatants were removed. Cell pellets were washed twice with 1 mL of unsupplemented CDM before resuspension in 1 mL of unsupplemented CDM. Samples were then inoculated into fresh CDM medium ± uracil supplementation, and cell density was monitored over time using a computer-controlled BioTek Synergy H1 plate reader (BioTek/Aligent) running Gen5 software v.3.14.

Genetic techniques

Oligonucleotides used for this study were synthesized by Integrated DNA Technologies (Coralville, IA) and are listed in Table S3. Restriction enzymes, DNA ligases, exonucleases, DNA polymerases, genomic DNA isolation, and PCR cleanup kits were purchased from New England Biolabs (NEB). Plasmids were extracted from E. coli using the Zyppy Plasmid Miniprep kit (Zymo). Unless indicated, plasmid constructions were performed using Gibson Assembly as described previously (90, 91). The plasmids used in this study are listed in Table S4. E. coli NEB 5α (NEB) was used as host for plasmid constructions, and all plasmids were confirmed by PCR and plasmid sequencing (PlasmidSaurus) before being introduced into S. aureus strain RN4220 by electroporation (92). Plasmid and marked mutations were moved between S. aureus strains via Φ85-mediated transduction (93).

Plasmid constructions

pDD02

A total of 500 bp of sequence upstream and downstream of pyrE was amplified from wild-type gDNA using the following primer pairs: oDD062, oDD063 (500 upstream), and oDD64, oDD65 (500 downstream) with Q5 DNA polymerase (NEB). pKOR1 (10,374 bp) was linearized and amplified using primers oDD60 and oDD61 (94). Each primer was designed to add 25 bp overlapping for later Gibson assembly. The linearized vector was treated with DpnI for 1 h at 37°C and cleaned up using a PCR purification kit. The purified PCR products were then subjected to Gibson Assembly and then transformed into E. coli DH5α. pDD03 and pDD10 were constructed in a similar manner for pyrD and pyrC.

pDD05

pyrD⁺, including its native promoter region, was amplified from wild-type gDNA with the forward primer containing a SacI cutting site and the reverse primer containing a BamHI cutting site. Both the vector backbone, pLL39, and the pyrD insert were cut with BamHI and SacI and treated with Quick Calf Intestinal alkaline Phosphatase (CIP; NEB) to dephosphorylate the ends to prevent self-ligation. After CIP treatment, the products were cleaned up using a PCR purification kit. The purified linearized vector and insert were ligated using T4 DNA ligase, and the resulting plasmid was then transformed into E. coli DH5a.

Construction of in-frame deletions

Plasmids pDD02 (∆pyrE), pDD03 (∆pyrD), and pDD10 (∆pyrC) were moved to S. aureus strain LAC at the permissive temperature (30°C). Allelic exchange was performed as described using secY antisense counterselection (94). In-frame deletion alleles were verified by PCR and Sanger sequencing.

GFP reporter assays

Cells carrying reporter fusions were streaked out from frozen stocks on TSA plates containing appropriate antibiotics. Single colonies were used to inoculate TSB with or without supplements: 200 µM uracil, 200 µM cytidine, 10 mg/mL fatty acid-free BSA. For growth in TSB ± uracil or cytidine, cells were grown in 3 mL medium in disposable 16 × 125 mm borosilicate glass test tubes. For BSA treatment, cells were grown with the 10:1 flask:medium ratio in DeLong flasks to stationary phase (16 h). Resulting cultures were then diluted 1:5 in 1× PBS to a total volume of 100 µL per well in a 96-well plate. Both OD₆₀₀ and GFP fluorescence (485 nm excitation, 535 nm emission) were read in a computer-controlled BioTek Synergy H1 plate reader (BioTek/Aligent) running Gen5 software v.3.14. Relative fluorescent units (GFP fluorescence/OD₆₀₀) were calculated by subtracting the background fluorescence of TSB and dividing by OD₆₀₀ to correct for cell density.

Analysis of SaeS kinase activity

Separation of SaeR and SaeR~P was performed as described previously (49) using 12% polyacrylamide gels containing 100 µM manganese and 50 µM of the acrylamide-pendant Phos-tag ligand (Wako Laboratory Chemicals). Cells were grown in TSB to mid-exponential phase, early stationary phase, and late stationary phase (2 h, 8 h, and 18 h, respectively) at 37°C at a 10:1 flask to medium ratio with shaking. Cell pellets (OD₆₀₀ of 10) were collected at 13,000 × g and stored at −80°C prior to analysis. Whole-cell extracts were obtained by resuspending cell pellets in 300 µL cell extract buffer (20 mM Tris [pH 7.0], 1× Protease Inhibitor Cocktail Set I [Sigma-Aldrich]) and transferred to sterile screw cap tubes containing approximately 100 µL of 0.1 mm silica beads. The cells were homogenized at room temperature using a Precellys 24 bead beater (Bertin Technologies) for three cycles of 6,500 rpm for 30 s each, followed by 3 min pauses on ice. The tubes were then centrifuged at 8,500 × g for 15 s to settle the beads, and the supernatant was transferred into new tubes. Whole-cell extracts were normalized by protein concentration (absorbance at 280 nm [A280]) to 100 µg and electrophoresed on Phos-tag gels with standard running buffer (0.1% [wt/vol] SDS, 25 mM Tris-HCl pH 8.3, 192 mM glycine) at 4°C under constant voltage (150V) for 2 h. The gels were washed for 15 min with transfer buffer (25 mM Tris [pH 8.3], 192 mM glycine, 20% methanol) with 1 mM EDTA followed by a second wash without EDTA to remove manganese ions. Proteins were then transferred to 0.45 µM polyvinylidene fluoride (PVDF) membranes (Cytiva). Membranes were blocked in blocking buffer (5% [wt/vol] skim milk in Tris-buffered saline with Tween 20 (TBST) (20 mM Tris-HCl pH 7.6, 150 mM NaCl with 0.1% [wt/vol] Tween 20) for 1 h. Membranes were then subjected to three brief washes in TBST and incubated with polyclonal rabbit antibodies to SaeR (1.5:1,000) for 1 h. Membranes were then washed three times with and incubated with StarBright Blue 700 goat anti-rabbit IgG (1:3,500; Bio-Rad) for 1 h. Membranes were subjected to three brief washes in TBST, and signals were visualized using an Amersham ImageQuant800. The densities of SaeR~P relative to total SaeR signal were determined by quantification with Multi Gauge software (FujiFilm). The assay is linear from 0 to 100 μg protein lysate. The data are representative of three different independent experiments, and a representative image is shown.

Analysis of SaeR protein levels

The whole-cell extracts and protein concentrations described above were used to visualize total SaeR levels. Briefly, 100 µg of whole-cell extracts were subjected to 12% SDS-PAGE, and proteins were transferred to 0.45 µM PVDF membranes (Cytiva). After transfer, membranes were blocked in blocking buffer for 1 h. Membranes were then subjected to three brief washes in TBST and incubated with polyclonal rabbit antibodies to SaeR (1.5:1,000) for 1 h. Membranes were then washed three times with and incubated with StarBright Blue 700 goat anti-rabbit IgG (1:3,500; Bio-Rad) for 1 h. Membranes were subjected to three brief washes in TBST, and signals were visualized using an Amersham ImageQuant800. The densities of the SaeR signal were determined by quantification with Multi Gauge software (FujiFilm). The data are representative of three different independent experiments, and a representative image is shown.

RNA sequencing

Overnight cultures of cells grown in TSB were used to inoculate 125 mL DeLong flasks containing TSB with a 10:1 flask:media ratio. Flasks were incubated in a shaking water bath set to 37°C at 250 rpm for 8 h. A two- milliliter sample of cells of the indicated genotype were collected at 13,000 × g, resuspended in an equal volume of quenching solution (ethanol:acetone [1:1]), pelleted, and stored at −80°C until use. Cell pellets were resuspended in TRIzol and transferred to sterile screw cap tubes containing approximately 100 µL of 0.1 mm silica beads. The cells were then homogenized at room temperature using a Precellys 24 bead beater (Bertin Technologies) for three cycles of 6,800 rpm for 30 s each, followed by 5 min pauses on ice. The tubes were then centrifuged at 4°C and 13,000 × g for 1 min to settle the beads. Clarified extracts were then mixed in an equivalent volume of 100% [vol/vol] ethanol. RNAs were purified using a Direct-Zol RNA Miniprep kit (Zymo Research) according to manufacturer’s instructions, and RNA yields were determined by measuring (A260/280). RNA quality was determined using a 5200 Fragment Analyzer System (Aligent), and RNA-Seq libraries were generated using an Illumina Stranded Total RNA kit after rRNA depletion (Ribo Zero Gold; Illumina). The resulting RNA-Seq libraries were then sequenced on a HiSeq 2500 Sequencing System (Illumina).

Differential expression analysis

The sequencing files were processed with Trimmomatic (95). Alignment to the S. aureus USA300_FPR3757 genome (NC007793; NCBI) and pairwise comparisons were made using Rockhopper (Wellesley College) (96, 97).

KEGG analysis

Outputs for RNA-Seq were then processed using the KEGGREST package in R to determine which biological pathways were enriched or disturbed based on the KEGG input genome for the USA300_FPR3757 isolate (saa; KEGG) (58, 59).

Antibiotic MIC measurements

MIC measurements were performed according to the Clinical and Laboratory Standards Institute methods for broth microdilution susceptibility testing of staphylococci (62). Strains were grown on cation-adjusted Mueller-Hinton broth plates containing 2% NaCl at 37°C for 24 h and 5–10 colonies were resuspended in 0.85% saline before adjusting to McFarland standard 0.5 (A₆₀₀ = 0.1). The cell suspensions were then diluted 1:20 in sterile 1× PBS, and 300 µL was used to inoculate 3 mL media containing serially diluted antibiotics or compounds (tunicamycin, fosfomycin, oxacillin, Congo red). The tubes were incubated at 35°C for 24 h and MIC values were recorded as the lowest compound concentration where no growth was observed.

Lipoteichoic acid extraction and blotting

Extraction

Lipoteichoic acid extraction was performed as previously described (64, 67, 98). Single colonies grown on TSA plates were used to inoculate 3 mL cultures in disposable 16 × 125 mm borosilicate glass tubes (Fisher Scientific). Cells were grown to stationary phase overnight (16 h) at 37°C and incubated with rotation. After incubation, cells were normalized to an OD₆₀₀ of 3 and collected at 7,800 rpm. The supernatant was removed, and pellets were resuspended in 1 mL sterile 1× PBS and transferred to screw cap tubes containing approximately 100 µL of 100 µm glass beads. The cells were homogenized at room temperature using a FastPrep-24 (MP Biomedicals) for three cycles of 6 m/s for 40 s each, followed by a 4 min rest. Tubes were then centrifuged at 200 × g for 1 min to settle the beads, and 700 µL of the supernatant was transferred into a fresh 1.5 mL Eppendorf tube. The tubes containing the supernatant were then centrifuged at 14,000 rpm for 15 min to collect the cell debris. After centrifugation, the supernatant was removed, and the remaining cell debris pellet was resuspended in 80 µL 2× Laemmli sample buffer (Bio-Rad) and incubated at 95°C for 20 min. After heating, samples were centrifuged at 14,000 rpm for 5 min, and the supernatant was stored at −20°C until ready for Western analysis.

Blotting

A total of 20 µL of the LTA extracts was run on 15% polyacrylamide gels at room temperature under constant voltage (100V) until the bromophenol blue reached the bottom of the gel and transferred to PVDF. After transfer, membranes were blocked in blocking buffer for 1 h. Membranes were then subjected to three brief washes in TBST and incubated with monoclonal mouse antibodies specific for LTAs (mAB55; Hycult Biotech; 1:5,000). Membranes were then washed three times with and incubated with StarBright Blue 700 goat anti-mouse IgG (1:3,500; Bio-Rad) for 1 h. Membranes were subjected to three brief washes in TBST, and signals were visualized using an Amersham ImageQuant800. The densities of the LTA signal were determined by quantification with Multi Gauge software (FujiFilm). The data are representative of three different independent experiments, and a representative image is shown.

Staphyloxanthin analysis

The extraction of pigment from S. aureus was performed as previously described (99). Single colonies grown on TSA plates were used to inoculate 3 mL cultures in disposable 16 × 125 mm borosilicate glass tubes (Fisher Scientific). Cells were grown to stationary phase (24 h) at 37°C and incubated with rotation. After incubation, 1 mL of culture was transferred to a 1.5 mL microcentrifuge tube and pelleted at 13,000 × g for 1 min. The supernatant was removed, pellets were washed with sterile distilled water, followed by re-centrifugation and removal of supernatant. Cell pellets were then resuspended in 200 µL of methanol followed by incubation at 55°C for 3 min in an Eppendorf ThermoMixer. After incubation, the cell-methanol mixture was centrifuged at 13,000 × g to remove cell debris, and the supernatant containing the pigment was transferred to a new microfuge tube. To ensure total removal of pigment, the extraction was repeated, and the extracts were pooled together. A total of 100 µL of extracts was then transferred to quartz cuvettes containing 900 µL methanol and mixed thoroughly by pipetting before measuring at 465 nm in an Amersham Ultraspec 2100 Pro UV-visible spectrophotometer. Absorbance was normalized by CFU enumeration by plating serially diluted cells to TSA plates.

Transmission electron microscopy

Single colonies grown on TSA plates were used to inoculate 4 mL of TSB ± uracil supplementation (200 µM). Cells were grown to stationary phase (16 h) at 37°C and incubated with rotation. After incubation, cells were pelleted in 2 mL microcentrifuge tubes until the pellet was ~2 mm tall from the bottom of the tube. Cell pellets were then washed three times in sterile 1× PBS (pH 7.4). Samples were then fixed overnight at 4°C in sterile 1× PBS (pH 7.4) containing 2.5% [vol/vol] electron microscopy grade glutaraldehyde before being processed at the Laboratory of Biological Ultrastructure at the University of Maryland, College Park, MD. Subsequently, cells were subjected to a second fixing step in 1% [vol/vol] osmium tetroxide for 1 h. Samples were then washed three times in double distilled water and postfixed in 2% [vol/vol] uranyl acetate solution for 60 min. Samples were then dehydrated using five incremental ethanol washes (35%, 50%, 70%, 95%, 100% [vol/vol] for 10 min each wash); the final wash using ethanol (100% [vol/vol]) was repeated three times. Following dehydration, cells were infiltrated using propylene oxide (100% [vol/vol]) two times for 15 min each, followed by overnight incubation with agitation in a 1:1 (vol:vol) mixture of propylene oxide (100% [vol/vol]) and resin (100% [vol/vol]). The following morning, samples were subjected to incremental infiltration steps (1:2, 1:3, 100% resin; propylene oxide [100% {vol/vol}]: resin [100% {vol/vol}] for 60 min each step); the final infiltration step with 100% resin was repeated before embedding the samples in fresh resin. Samples were then solidified by incubating the resin at 70°C for a total of 12 days before thin-sectioning and staining. Stained sections were loaded onto mesh grids for imaging. Images were obtained on a Hitachi HT7700 Transmission Electron Microscope at the Laboratory of Biological Ultrastructure at the University of Maryland, College Park, MD. Cell diameter and envelope thickness were measured in FIJI (100). Measurements were determined by first setting pixel values to the length of the scale bar, followed by use of the line tool to measure cell diameter or envelope thickness (from cell membrane to the outside of the cell wall), respectively.

Medium FFA analysis using QTRAP 5500

This method is designed to measure 21 classes of lipid molecules which include FFAs using QTRAP 5500 LC-MS/MS System (Sciex). Single colonies were used to inoculate 125 mL DeLong flasks containing TSB ± BSA (10 mg/mL) with a 10:1 flask:medium ratio. Flasks were incubated in a shaking water bath set to 37°C at 250 rpm for 16 h. Resulting cultures were checked for promoter activity, then pelleted at 3,500 × g for 20 min in a sterile 50 mL falcon tube. After pelleting the cells, 9 mL of the supernatant was transferred to a fresh 50 mL falcon tube. For FFA analysis, 2 mL supernatant from resulting promoter activity cultures was freeze-dried and used for sample preparation. A total of 125 µL of pre-chilled isopropanol containing internal standard (FFA-d4-palmitic acid—1 ug per mL; Cambridge Isotope Laboratories) was added, and samples were vortexed. The sample tube was plunged into dry ice for 30 s followed by a 90 s incubation in a 37°C water bath. The freeze-thaw cycle was repeated three times, followed by sonication for 30 s. After sonication, samples were vortexed for 1 min and were incubated at −20°C overnight for complete protein precipitation. The samples were then centrifuged at 13,000 rpm for 20 min at 4°C. After centrifugation, the supernatant of each sample was transferred to an MS vial for LC-MS analysis. 5 µL of each sample was injected onto an Xbridge amide 3.5 µm, 4.6 × 100 mm column (Waters) using SIL-30 AC autosampler (Shimadzu) connected with a high-flow LC-30AD solvent delivery unit (Shimadzu) and CBM-20A communication bus module (Shimadzu), online with QTRAP 5500 (Sciex, MA, USA) operating in positive and negative ion mode. A binary solvent comprising acetonitrile/water 95/5 with 10 mM ammonium acetate as solvent A and acetonitrile/water 50/50 with 10 mM ammonium acetate as solvent B was used for the resolution. Lipids were resolved at a 0.7 mL/min flow rate. Initial gradient conditions started with 100% of solvent A at 0.2 mL per min flow rate, shifting toward 99.9% of solvent A over a time period of 3 min, 94% of solvent A over a time period of 2 min at 5 min, and 75% of solvent A over a period of 2 min. Finally, the column was re-equilibrated the to initial conditions (100% of solvent A) over a time period of 3 min using auto sampler temperature 15°C and oven temperature 35°C. Source and gas settings were used as follows: curtain gas = 30, collision gas  = medium, ion spray voltage = 5.5 kV in positive mode and −4.5 kV in negative mode, temperature = 550°C, nebulizing gas = 50 and heater gas = 60. The data were then normalized to the respective internal standard area (FFA-d4-palmitic acid; Cambridge Isotope Laboratories) and processed using MultiQuant 3.0.3 (Sciex). The quality and reproducibility of LC-MS data were ensured using several measures. The column was conditioned using the pooled quality control samples initially and was also injected periodically to monitor shifts in signal intensities and retention time as measures of reproducibility and data quality of the LC-MS data.

Statistical analysis

Data shown are the results of at least three independent experiments. Statistical significance was determined using Prism v.10 (GraphPad) with the indicated tests.

DATA AVAILABILITY

RNA-sequencing data have been deposited in the NCBI database under the BioProject accession number PRJNA1232367 and in the Sequence Read Archive (SRA) under accession number(s) (SRX27890085 through SRX27890090).

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jb.00115-25.

Supplemental materials. jb.00115-25-s0001.docx.

Supplemental figures, tables, and references.

Data Table S1. jb.00115-25-s0002.xlsx.

Transcriptomic analysis of ∆pyrE mutant cells during pyrimidine limitation.

Data table S2. jb.00115-25-s0003.xlsx.

KEGG analysis reveals shifts in many different biochemical pathways in the ∆pyrE mutant during pyrimidine limitation.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.