Role of Fatty Acid Kinase in Cellular Lipid Homeostasis and SaeRS-Dependent Virulence Factor Expression in Staphylococcus aureus

ABSTRACT The SaeRS two-component system is a master activator of virulence factor transcription in Staphylococcus aureus, but the cellular factors that control its activity are unknown. Fatty acid (FA) kinase is a two-component enzyme system required for extracellular FA uptake and SaeRS activity. Here, we demonstrate the existence of an intracellular nonesterified FA pool in S. aureus that is elevated in strains lacking FA kinase activity. SaeRS-mediated transcription is restored in FA kinase-negative strains when the intracellular FA pool is reduced either by growth with FA-depleted bovine serum albumin to extract the FA into the medium or by the heterologous expression of Neisseria gonorrhoeae acyl-acyl carrier protein synthetase to activate FA for phospholipid synthesis. These data show that FAs act as negative regulators of SaeRS signaling, and FA kinase activates SaeRS-dependent virulence factor production by lowering inhibitory FA levels. Thus, FA kinase plays a role in cellular lipid homeostasis by activating FA for incorporation into phospholipid, and it indirectly regulates SaeRS signaling by maintaining a low intracellular FA pool.

FA kinase is a two-component enzyme consisting of a kinase domain protein (FakA) that phosphorylates an FA bound to a FA binding protein (FakB) (29,30). FakB transfers the acyl-phosphate to PlsX to be converted to acyl-acyl carrier protein (ACP) and enter the fatty acid biosynthetic cycle or to PlsY for the acylation of glycerol phosphate (31,32). FA kinase is the only pathway for FA activation and incorporation into S. aureus membrane phospholipids (29,33). A FA kinase-null S. aureus strain was resistant to dermcidin (34) and exhibits increased biofilm formation (35). However, the most striking phenotype of FA kinase knockout strains is the lack of ␣-hemolysin production, indicating a novel role for FA kinase in the control of virulence factor production (36). A genome-wide analysis showed that FA kinase-null strains were specifically deficient in the expression of all virulence factors controlled by the SaeRS system (29). The fact that acetyl-phosphates are known to phosphorylate response regulators suggests that FA kinase may participate in the regulatory phosphorylation cascade in the SaeRS system (29,37). Although it is clear that transcription of the SaeRS virulence regulon is supported by a functional FA kinase, the connection between FA kinase and SaeRS has not been established.
The goal of this study was to establish a biochemical connection between FA kinase and the activity of the SaeRS system. We found that FA kinase activation of virulence factor transcription requires the SaeRS two-component system, but FA kinase does not directly phosphorylate either SaeS or SaeR. Instead, FA are inhibitors of SaeS phosphorylation of SaeR and accumulate in FA kinase-null bacteria. FA removal by growth with bovine serum albumin (BSA) or by the ectopic expression of an acyl-ACP synthetase restored transcription of the saePQRS operon and downstream SaeRS-regulated genes, showing that it is FA rather than FA kinase that regulates SaeRS signaling. Thus, the 10-fold downregulation of SaeRS signaling in FA kinase-null cells is due to the accumulation of cellular FA, which in turn negatively regulates SaeRS signaling.

RESULTS
FA kinase impact on virulence factor transcription requires SaeRS. We first confirmed that FA kinase is dependent upon SaeRS to influence virulence factor transcription. A reporter construct to monitor the activity of the SaeR-controlled saePQRS promoter was created by fusing the saePQRS promoter to the chloramphenicol acetyltransferase (CAT) coding sequence to identify the DNA sequences required for FA kinase activation of the promoter (Fig. 1A). Accordingly, robust transcription of the reporter was observed in the wild-type strain USA300 but was absent in strains lacking SaeRS (strain PDJ50) or SaeS (strain PDJ51) (Fig. 1B). FA kinase inactivation significantly depressed transcription from the saePQRS promoter, showing that the putative FA kinase regulatory elements are located within the promoter construct (Fig. 1B). Activation of the saePQRS promoter requires both SaeR binding sites (11); therefore, we created promoter mutants that inactivated one or the other of the SaeR sites (Fig. 1A).
Neither of these mutant promoter constructs was active in the wild-type USA300 or the FA kinase knockout (JLB2) strains (Fig. 1C). These data showed that FA kinase regulation of saePQRS expression requires both SaeR binding sites, suggesting that FA kinase modifies SaeRS signaling rather than functioning independently on the promoter.
FA kinase does not phosphorylate SaeS or SaeR. The SaeRS two-component system is activated via phosphorylation (6). Response regulators in Gram-negative two-component systems can accept phosphate from acetyl-phosphate (38), leading to the hypothesis that acyl-phosphate produced by FA kinase phosphorylates either SaeS or SaeR (38). FakA, FakB2, SaeS, and SaeR proteins were expressed, purified, and used in reactions with [ 32 P]ATP to examine phosphotransfer from FA kinase to the SaeRS system. We first tested phosphorylation of SaeR. SaeS phosphotransfer to SaeR was included as a positive control (Fig. 1D). FakA was unable to phosphorylate SaeR when added at concentrations from 1 M to 10 M (Fig. 1D). Introducing FakB2 into the Fatty Acid Kinase and SaeRS Regulation ® reaction mix did not result in SaeR phosphorylation (Fig. 1D). Oleic acid (18:1) was included to drive the production of acyl-phosphate, but its presence did not lead to phosphorylated SaeR (Fig. 1D). We next examined phosphotransfer from FA kinase to SaeS. Two residues in the ATP binding region of SaeS (K252A R298A) (39) were mutated to ensure that the 32 P signal for SaeS was transferred from FA kinase and not the results of autophosphorylation. Wild-type SaeS was used as a positive control (Fig. 1E). The SaeS mutant displayed no autophosphorylation activity (Fig. 1E). Addition of FakA with the SaeS mutant did not result in SaeS phosphorylation (Fig. 1E). The same result occurred with the addition of FakB2 and oleic acid (18:1) (Fig. 1E). Acyl-phosphate produced by FakA and FakB was not detectable on SDS-PAGE gels, but it was quantified using 14 C-labeled FA as the substrate to produce [ 14 C]acyl-phosphate (29). A FA kinase assay performed in parallel showed the active formation of acyl-phosphate (specific activity, 6 mol/min/mg).
Addition of exogenous FA inhibits SaeRS signaling. An analysis of the global transcriptional response to oleic acid (18:1) addition to S. aureus cultures showed that FA strongly inhibit SaeRS signaling (25). To test if FA inhibition of SaeRS in USA300 bacteria was indeed limited to the 18:1 FA species, we added 500 M of either 16:0 (palmitate) or 18:1 in the presence of 10 mg/ml FA-free BSA to USA300 bacteria. No growth differences were noted between bacteria grown with versus without exogenous FA (40). Cells were harvested, and RNA was isolated for quantitative real-time PCR (qRT-PCR). Three representative genes controlled by SaeRS were downregulated by the addition of either of the FA species ( Fig. 2A). To further confirm the inhibitory effect of exogenous FA, we added 500 M myristate (14:0), 16:0, or 18:1 in the presence of 10 mg/ml BSA to strain USA300 harboring the CAT reporter plasmid. Transcriptional activity of the saePQRS promoter was downregulated with the addition of each FA (Fig. 2B). BSA is necessary to present the FA to the cells and prevent the toxic effects of the 14:0 FA on the bacteria (40). Krute et al. (37) reported that 14:0 did not alter SaeRS signaling in strain USA300, but that study did not employ BSA to present the FA and avoid toxicity.
Nonesterified FA accumulate in FA kinase-null strains. FA are the substrates for FA kinase; therefore, we determined whether cellular FA accumulated in strains lacking the kinase. Sa178RI and USA300 strains were metabolically labeled with [ 14 C]acetic acid for 3 h, and lipids were extracted and analyzed by thin-layer chromatography (TLC). Phospholipid and diacylglycerol were the most abundant lipids in all strains. However, a 14 C-labeled nonesterifed FA band was detected in both ΔfakA mutant (strains PDJ42 Overnight cultures of USA300 were diluted and grown to an OD 600 of 0.5 before 500 M FA and 10 mg/ml BSA were added for 2 h. Fold changes shown are in relation to results with a USA300 strain grown with only 10 mg/ml BSA. The data are from two biological replicates that were each analyzed in triplicate. (B) saePQRS promoter activity in USA300 grown in the absence or presence of 500 M myristate (14:0), palmitate (16:0), or oleate (18:1) with 10 mg/ml of FA-free BSA. Cultures were grown to an OD 600 of 2 before bacteria were harvested and CAT activity in cell lysates was measured. Data are from three biological replicates. Significance was determined by the two-tailed Student t test. *, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001. and JLB2) and ΔfakB1&B2 mutant (strains PDJ43 and JLB31) backgrounds ( Fig. 3A) but appeared to be absent in the wild-type strains. Mass spectrometry was used to quantify the amount of free FA in the strains. FA was detected in wild-type strains, illustrating that FA are a normal, albeit minor, component of the normal cellular lipidome (Fig. 3B). As shown by the metabolic labeling experiment, FA accumulated in the FA kinase-null background strains to become a more significant component of the cellular lipidome (Fig. 3B).
FA, rather than FA kinase, regulates SaeRS activity. If the accumulation of cellular FA in FA kinase knockout strains inhibits SaeS signaling, reducing the cellular FA should activate SaeRS. To test this, we added FA-deficient BSA to the growth medium (10 mg/ml) to remove FA from the cells by providing an extracellular sink for these lipids. Strain JLB2 (ΔfakA) was grown with or without the addition of BSA and metabolically labeled with [ 14 C]acetate. TLC analysis showed that BSA extracted a significant portion of the intracellular FA into the medium, while FA remained predominantly cell associated without addition of BSA (Fig. 4A). Extension of these experiments to strains USA300 and JLB31 (ΔfakB1&B2) confirmed that BSA extracts FA from all strains (see Fig. S1 in the supplemental material). Quantitative mass spectrometry showed a 50% decrease of cellular FA levels in strain JLB2 (ΔfakA) grown in BSA compared to growth without BSA (Fig. 4B, inset). qRT-PCR showed that three SaeR-regulated transcripts were elevated by the removal of FA from strain JLB2 (ΔfakA) by growth with BSA (Fig. 4C). This result was corroborated by an increase in saePQRS promoter activity from the reporter plasmid in strain JLB2 (ΔfakA) grown with BSA (Fig. 4D). Likewise, BSA increased activity of the saePQRS promoter in a USA300 background, demonstrating that levels of FA in the wild-type strain USA300 background are inhibitory to SaeRS signaling (Fig. 4D). BSA had no effect on promoter activity in SaeS knockouts (Fig. 4D).
We next reduced cellular FA levels by the heterologous expression of a protein that would activate cellular FA for incorporation into phospholipids. Neisseria gonorrhoeae acyl-ACP synthetase (NgAas) (41) was cloned into the isopropyl-␤-Dthiogalactopyranoside (IPTG)-regulated vector pG164 and expressed in strain PDJ42 (ΔfakA) derived from strain Sa178R1. Three sentinel SaeRS-regulated virulence factor transcripts were elevated in strain PDJ42 (ΔfakA) by the expression of either NgAas or FakA, illustrating that NgAas expression complemented the FA kinase defect in virulence factor transcription (Fig. 5A). [ 14 C]acetate metabolic labeling and TLC demonstrated that expression of both NgAas and FakA significantly reduced the levels of intracellular FA in strain PDJ42 (ΔfakA) (Fig. 5B). Finally, quantitative mass spectrometry confirmed that NgAas or FakA expressed in strain PDJ42 (ΔfakA) reduced cellular FA (Fig. 5C). The restoration of SaeRS signaling by NgAas expression leads to the conclu- , and JLB31 (ΔfakB1&2 mutant) grown to an OD 600 of 2 to 3 followed by lipid extraction and quantitative mass spectrometry to determine the amounts of each FA in the strain. The inset shows the effect of FA kinase inactivation on the total FA pool size. The data are from three biological replicates. Significance was determined by the two-tailed Student t test. *, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001.
Fatty Acid Kinase and SaeRS Regulation ® sion that FA, rather than an activity of FA kinase itself, are responsible for the inhibition of SaeRS-dependent virulence factor transcription in ΔfakA strains.
Increased FakA expression decreases FA and increases transcription in a wildtype strain. Inclusion of FA-deficient BSA in the medium increased saePQRS transcription in strain USA300 (Fig. 4D), highlighting that FA levels in the wild-type background restrain SaeRS activity. We corroborated this conclusion by examining the impact of elevated expression of FakA in a wild-type strain USA300 on FA levels and transcription of SaeRS-regulated genes. Quantitative mass spectrometry showed that the plasmiddriven expression of FakA in strain USA300 reduced the intracellular FA pool (Fig. 6A,  inset). Furthermore, the expression levels of saeP, ehp, and efb genes were elevated in FakA-overexpressing cells (Fig. 6B). These data confirmed the role of FA kinase in recycling FA into phospholipids and showed that intracellular FA in S. aureus functionally represses SaeRS signaling in wild-type strains.
Hemolytic activity is regulated by cellular FA levels. The elimination of ␣-hemolysin production was the first recognized phenotype of ΔfakA mutant strains (36). ␣-Hemolysin is regulated by SaeRS (42), and its expression can be monitored by measuring zones of clearance produced by bacteria on a blood agar plate. Plasmiddriven expression of FakA in strain USA300 increased production of ␣-hemolysin (Fig. 7A), consistent with the lowering of cellular FA in this strain (Fig. 6A). As expected, ␣-hemolysin production was almost eliminated in strain JLB2 (ΔfakA) compared to its wild-type counterpart, strain USA300 (Fig. 7A). ␣-Hemolysin production was restored in JLB2 (ΔfakA) either by growth on blood agar plates supplemented with FA-deficient BSA or by the expression of NgAas (Fig. 7A); these findings are also consistent with the reduction of FA and the transcriptional activation of other SaeRS targets by these treatments (Fig. 4 and 5). The restoration of ␣-hemolysin production independent of FA kinase function supports the role of cellular FA in inhibiting SaeRS-dependent transcription.

DISCUSSION
Our results lead to a mechanistic model for the roles of FA kinase in cellular lipid homeostasis in S. aureus (Fig. 7B). One function is to recycle FA into the phospholipid biosynthetic pathway. FA recycling begins when FakB binds to FA that spontaneously partition into the phospholipid bilayer (Fig. 7B). FakB(FA) then binds FakA, which phosphorylates the FakB-bound FA. FakB(FA~P) then delivers the acyl-phosphate to either PlsY or PlsX for phospholipid or FA synthesis (31,32), and the empty FakB returns to the membrane to pick up another FA. FA kinase is also essential for the activation of exogenous FA for incorporation into membrane phospholipids (29). However, this pathway has a limited bandwidth, and exogenous FA can be introduced to the growth medium at levels that easily overwhelm the capacity of S. aureus phospholipid biosynthesis, leading to repression of SaeRS-dependent virulence factor transcription in wild-type bacteria (25). FA synthesis is an energy-intensive process (43), and both of these functions of FA kinase represent pathways devoted to the efficient utilization of resources. Another role of FA kinase is to prevent the accumulation of membranedisrupting, nonesterified FA, which will introduce additional negative charge and impart curvature elastic stress that alters the biophysical properties of the phospholipid bilayer (44). The origin of the FA pool in S. aureus is unknown. S. aureus contains multiple genes with predicted esterase, lipase, and lysophospholipase activities (45), and their hydrolytic activities may be responsible for phospholipid turnover. Another source may be the hydrolysis of the unstable acyl-phosphate (FA~P) intermediate, which is known to occur in cells where PlsY activity is blocked (46). Regardless of the metabolic process that generates FA in S. aureus, FA kinase recycles FA into membrane phospholipids and maintains a low level of nonesterified FA in the cell.
The biochemical mechanism that explains FA inhibition of SaeRS signaling remains to be elucidated (Fig. 7B). The experimental results linking the control of SaeS activity to the conformation of the membrane anchor domain (26,27) suggest several possibilities for FA regulation. One idea is that SaeS responds to the net overall negative charge of the bilayer imparted by the presence of a higher mole fraction of FA. FA are also known to introduce curvature elastic stress into phospholipid bilayers (44), and this Overnight cultures of strains USA300/pCS119 (empty vector) and USA300/pJLB165(FakA) were diluted and grown to an OD 600 of 2, cells were harvested, RNA was extracted, and the levels of the indicated transcripts were determined by qRT-PCR. The data are from two independent experiments performed in triplicate. Significance was determined by using the two-tailed Student t test. *, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001.

Ericson et al.
® alteration in the membrane biophysical properties may alter the conformation of the SaeS membrane anchor domain. FA may act as a classical ligand for SaeS binding to a site on the transmembrane domain to stabilize its inactive conformation. Furthermore, the Sae system has the additional complexity of involving the accessory proteins SaeP and SaeQ, which modify SaeS signaling (47), and may potentially be the FA-interacting component in the system. Deciding between these possibilities will require biophysical experiments to determine how phospholipid bilayer composition impacts the conformation of the SaeS transmembrane helices and/or its interactions with accessory proteins. It is clear that not all regulation of SaeS activity is FA dependent. Defensin activation of SaeS occurs in ΔfakA strains (37), illustrating FA-independent control of SaeS activity in this circumstance. Unlike environmental signals that impact SaeRS activity (22,23,37,48,49), FA are normal constituents of the cellular lipidome whose concentrations create a cellular set point for SaeS signaling that is controlled by the activity of FA kinase. The FA kinase two-component system is conserved in Gram-  (29), and it seems likely that its biochemical roles in exogenous FA activation and cellular lipid homeostasis are common themes in this group of bacteria. FA kinase clearly has these housekeeping functions, but whether its activity is regulated to control FA levels and hence SaeS activity is an open question. More work will be needed to determine if the role of FA as cellular signaling molecules is more widespread or whether the S. aureus SaeRS two-component system is a unique example of this type of regulation.
Growth conditions and strain construction. Bacteria were grown in Luria (LB) broth. When appropriate, 1.5 g/ml of erythromycin, 10 g/ml chloramphenicol, or 100 g/ml ampicillin was added. S. aureus strains were transformed as previously described (50). Strains and plasmids used in this study are listed in Table S1. Strains PDJ50 and PDJ51 were constructed by allelic replacement (51). In brief, to make strain PDJ50, primers were ordered to amplify regions upstream (saeS-5-ERI-F and saeS-5-Kpn-R) and downstream (saeS-3-Kpn-F and saeS-3-Sal-R) of saeS. These regions were cloned into the pJB38 vector (51). Once confirmed by sequencing, the plasmid was transformed into strain Sa178RI. A single transformant was inoculated into 5 ml of tryptic soy broth with 10 g/ml chloramphenicol and grown overnight. A portion of this overnight culture was transferred to growth at 43°C for 48 h to initiate recombination. Bacteria were then plated and single colonies selected and grown at 37°C overnight. Overnight cultures were diluted and plated on tryptic soy agar containing anhydrotetracycline to initiate plasmid loss through another round of recombination. Positive colonies were grown on medium with or without chloramphenicol, and those colonies sensitive to the drug were tested by PCR. To construct strain PDJ51, saeR-ERI-F and saeR-Kpn-R primers were ordered to amplify the region upstream of SaeR and ligated into a plasmid with the downstream sequence used to produce strain PDJ50.
Plasmid construction and primers. The CAT promoter plasmid was constructed using the pCM28 plasmid as a backbone with erythromycin resistance. The CAT gene was amplified from plasmid pSK5483 and cut with PstI and SphI before ligation into pCM28. The saePQRS promoter was cloned in front of the coding sequence for CAT by using BamHI and KpnI. Primers were designed to mutate the SaeR binding sites, and the QuikChange Lightning site-directed mutagenesis kit (Agilent, Santa Clara, CA) was used according to the manufacturer's protocol. Genes encoding NgAas and S. aureus FakA were cloned into pG164 using restriction sites BamHI and HindIII. NgAas was then transferred from pG164 to pCS119 using SalI and NheI restriction enzymes. Plasmid pCS119 was created by amplifying the sarAP1 promoter from USA300 using primers sarAP1EcoRI F and sarAP1NheI R. The PCR product was digested with EcoRI and NheI and ligated into a cut pCM28 plasmid. Minipreps were performed on transformants, and all plasmids were confirmed by sequencing. Primers used in this study are listed in Table S2.
Protein purification. FakA and FakB proteins were expressed and purified as previously described (29,30). SaeR, SaeS, and SaeS(K252A R298A) were synthesized with an amino-terminal His tag from Invitrogen (Carlsbad, CA). First, the genes were inserted into the TOPO Zero Blunt plasmid from Invitrogen and inserts were confirmed by sequencing. Genes were excised using BamHI and NdeI and ligated into a cut pET28a plasmid. After sequence confirmation, the pET28a constructs were transformed into Escherichia coli BL21 cells. Two liters of cells was grown to an optical density at 600 nm (OD 600 ) of 1 before adding 1 mM IPTG for overnight induction at 16°C. Cells were harvested by centrifugation at 4,000 relative centrifugal force (rcf) for 10 min. Cell pellets were resuspended in 25 ml of lysis buffer (20 mM Tris-HCl [pH 7.9], 0.5 M NaCl, protease inhibitor, lysozyme) and lysed with a microfluidizer. Centrifugation at 36,000 rcf for 1 h removed debris. Supernatant was added to a 4-ml Ni-nitrilotriacetic acid column that had been washed, charged with 50 mM NiSO 4 , and equilibrated with MCAC-0 (20 mM Tris-HCl [pH 7.9], 0.5 M NaCl, 10% glycerol). After the supernatant was run over the column, it was washed with 50 ml of MCAC three times with an increasing concentration of imidazole (0 mM, 25 mM, 50 mM) before the proteins were eluted in a buffer containing 400 mM imidazole. Proteins were checked on an SDS-PAGE gel to confirm correct size, concentrated, and purified on a Superdex 200 16/600 column (GE Healthcare, Little Chalfont, United Kingdom), using 20 mM Tris-HCl (pH 7.9), 0.5 M NaCl as the collection buffer.
Phosphorylation assay. Phosphorylation assays of SaeS and SaeR with FakAB were performed in buffer with 10 mM Tris-HCl (pH 7.5), 5 mM MgCl 2 , 50 mM KCl, and 0.1 Ci of [␥-32 P]ATP in a final volume of 20 l. SaeR and SaeSK252A R298A were held constant at 5 M when being tested for phosphorylation. SaeS, FakA, and/or FakB were added at increasing concentrations of 1, 2, 5, and 10 M when appropriate. FA was added to a final concentration of 10, 20, 40, or 80 M when appropriate. Phosphorylation reaction mixtures were incubated for 30 min at room temperature before being fractionated on a 10% SDSpolyacrylamide gel. Gels were stained with Coomassie brilliant blue dye, destained, and dried for 1 h before exposure on a phosphor screen from GE.
FA kinase assays. FA kinase assays were performed as previously described (29). Briefly, reactions were run in mixtures containing 0.1 M Tris-HCl (pH 7.5), 20 mM MgCl 2 , 10 mM ATP, 1% Triton X-100, Ericson et al. Addition of FakA initiated the reactions, and the mixtures were incubated at 37°C for 20 min. Forty microliters of each reaction mixture was spotted onto DE81 disks, which were washed for 30 min, three times in ethanol with 1% acetic acid. Disks were dried and scintillation was quantified on a Beckman counter (Brea, CA).
Metabolic labeling. Overnight cultures were diluted to an OD 600 of 0.05 in 10 ml of LB broth, and 50 Ci of [ 14 C]acetate was added. Cultures were grown for 3 h before being harvested. Bligh-Dyer lipid extraction was used on the cells and when appropriate on the medium (52). In brief, cell pellets were resuspended in 1 ml of water. Next, 2.4 ml of methanol with 2% acetic acid and 1 ml of chloroform were added. Mixtures were incubated at room temperature for 30 min, 1.5 ml of chloroform and 1.2 ml of water were added, and samples were vortexed and centrifuged for 10 min. The lower phase was removed and blown down under nitrogen. Because 10 ml of medium was extracted for each experimental condition, volumes added for lipid extraction from medium were 10 times higher. Lipid extractions of media were performed in a separatory funnel with the bottom organic layer easily removed. The lipid fraction was then concentrated under nitrogen before resuspension in a 1:1 chloroform-methanol mixture. Scintillation counts in 5 l of sample were determined, and equal counts of each sample were loaded onto a Silica G plate from Analtech (purchased from Sigma), which was then developed with hexane-diethyl ether-acetic acid (80/20/1 [vol/vol/vol]). Plates were allowed to dry and then exposed overnight to a phosphor screen.
Mass spectrometry. Cells were grown to an OD 600 between 2 and 3 in all experiments to minimize any variations in FA content as a function of growth phase. Cells (10 ml) were harvested and lipids were extracted using the Bligh-Dyer materials and methods (52). At the beginning of the lipid extraction, D4-16:0 (4 nmol) was added to each sample to serve as an internal standard for quantification of FA levels in the sample. Lipid extracts were dried completely under N 2 gas and resuspended in 400 l of chloroform-methanol (1:1). Free fatty acids were analyzed using a Shimadzu Prominence ultrafast liquid chromatograph attached to a Sciex QTrap 4500 apparatus equipped with a Turbo V ion source (Sciex, Framingham, MA). Samples were injected onto an Acquity UPLC BEH HILIC column (1.7 m, 2.1 by 150 mm; Waters, Milford, MA), using a flow rate of 0.2 ml/min. Solvent A was 100% acetonitrile, and solvent B was 15 mM ammonium acetate (pH 3). The high-performance liquid chromatography program was the following: 0 to 2 min isocratic at 4% B; 2 to 20 min linear gradient to 80% B; 20 to 23 min isocratic at 80% B; 23 to 25 min linear gradient to 4% B; 25 to 30 min isocratic with 4% B. The QTrap 4500 was operated in the negative mode, and the ion source parameters were the following: ion spray voltage, Ϫ4,500 V; curtain gas, 25 lb/in 2 ; temperature, 350°C; ion source gas 1, 40 lb/in 2 ; ion source gas 2, 60 lb/in 2 ; declustering potential, Ϫ40 V. The system was controlled and analyzed by using Analyst software (Sciex). Cell numbers were determined by making serial dilutions of a culture and counting the colonies.
Chloramphenicol acetyltransferase assay. For CAT assays, overnight cultures were diluted in 10-ml volumes to an OD 600 of 0.05 in LB. When appropriate, BSA was added. After growth for 3 to 4 h, cells were harvested and resuspended in 100 l of a lysis buffer of 25 mM Tris-HCl, 10 mM EDTA, protease inhibitor, and DNase I. Six microliters of 5 mg/ml lysostaphin was added to each sample, and then the mixtures were incubated at room temperature for 20 min. Lysed cells were centrifuged at 14,000 rpm for 30 min, and 5 l of supernatant was used for the CAT assay. The CAT assay reaction buffer contained 100 mM Tris, 400 mM DTNB, and 1 M acetyl coenzyme A. Ninety-three microliters of this mixture was added to the supernatant. Lastly, 2 l of a 1.6-mg/ml chloramphenicol solution was added to each reaction mixture. The production of TNB from these reactions was quantified using a SpectraMax machine (Molecular Devices, Sunnyvale, CA) in kinetic mode, which read the change in absorbance at 412 nm every 10 s. To convert the rate of the change of absorbance to millimoles of TNB produced, the change was divided by 13.6, as described previously (53). Rates of TNB conversion were normalized by running a bicinconinic acid (BCA) assay for determination of protein concentration. Before cells were harvested for the CAT assay, 200 l of each sample was collected and cells were harvested and then resuspended in 100 l of 100 mM Tris-HCl (pH 7.5), 0.1% SDS, and 0.1% Triton X-100. Mixtures were incubated on ice for 20 min and then at 100°C for 30 min. Samples were then centrifuged at 14,000 rpm for 15 min, and 25 l was added to the working reagent made from a BCA protein kit from Thermo Fisher (Waltham, MA). BSA was used as a standard to quantify protein amounts. After the sample and working reagent were mixed, they were incubated at 37°C for 30 min. The protein amount was measured using a SpectraMax reader based on absorption at 562 nm. The protein determinations correlated with the cell numbers, and normalization of CAT activity based on the OD 600 gave the same relative results.
Quantitative real-time PCR. For qRT-PCR, S. aureus strains USA300, JLB2 (ΔfakA), JLB31 (⌬fakB1&2), Sa178R1, Sa178R1/pNgAas, and Sa178R1/pFakA were grown to an OD 600 of 2. For BSA experiments, JLB2 (ΔfakA) cells were grown with or without FA-free BSA, and for FA experiments strains were grown to an OD 600 of 0.5 before FA was added to BSA to a final concentration of 500 M. RNA was isolated with an RNAqueous purification kit (Ambion, Austin, TX) according to the manufacturer's specifications. Purified RNA was then mixed with 0.5 vol of LiCl precipitation solution (Ambion) and left at Ϫ20°C for 30 min. Turbo DNase (Ambion) was added to the precipitated RNA to a final concentration of 1 U of DNase per 5 g of RNA. This mixture was incubated at 37°C for 30 min. Integrity of the RNA was assessed by agarose gel electrophoresis and by using an Agilent Technologies 2100 Bioanlyzer Lab-on-a-Chip system, before the RNA was used in the qRT-PCR. Primers for ehp, efb, and saeP have been previously reported (22). Each 20-l RT reaction mixture contained 500 ng of RNA, 12.5 ng/l of random hexamers (Invitrogen), 10 mM deoxynucleoside triphosphates (Sigma), 40 U of RNaseout (Invitrogen), and 10 U/l of SuperScript II reverse transcriptase (Invitrogen). After the RT reaction, 2-l aliquots of the cDNA product were added to an RT-PCR reaction mixture with 1ϫ TaqMan universal PCR master mix (Applied Biosystems, Foster City, CA), 660 nM of each forward and reverse primer, and 150 nM of probe. Specific products were detected on an ABI Prism 7700 sequence detection system (Applied Biosystems) using the following conditions: 50°C for 2 min, 95°C for 10 min, 95°C for 15 s for 40 cycles, and 60°C for 1 min. Samples were processed in 96-well plates. Three RT-PCRs were performed on each cDNA sample. cDNA was prepared from separate cultures for each biological condition. No-template mixtures and reaction mixtures without reverse transcriptase were run as negative controls. Real-time values were evaluated using the threshold cycle (C T ) method, with genes being normalized to a gmk calibrator (54).
FA removal by BSA. Overnight cultures of strains USA300, JLB2 (ΔfakA), and PDJ50 (ΔsaeS) were diluted to an OD 600 of 0.05 in 10 ml of LB medium in the absence or presence of 10 mg/ml FA-free BSA. For metabolic labeling, cells were grown with FA-free BSA for 3 h before cell pellets were harvested and lipid extraction was performed. For CAT assays, bacterial cultures were grown to an OD 600 of 2 and used for further analysis.
Hemolysis assay. For the hemolysis assay, overnight cultures were diluted to an OD 600 of 0.05 and allowed to grow to mid-log phase. Upon reaching mid-log growth, 1 l of the culture was spotted onto an LB agar plate supplemented with 3% rabbit blood. Plates were incubated at 37°C overnight, after which zones of clearance were examined.
Statistics. Data sets were compared using the two-tailed Student t test, and statistical significance is indicated in the figures (*, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001).
Availability of data. Strain descriptions and data that support the findings of this study are available from the corresponding author upon request.

ACKNOWLEDGMENTS
We thank Karen Miller for protein purification, Pam Jackson for construction of cell lines and plasmids, Katie Creed for performing qRT-PCR experiments, the Hartwell Center DNA Sequencing Shared Resource for DNA sequencing, Josh Dodson for pJD01, Neville Firth for plasmid pSK5483, and Nachum Kaplan for the strain Newman. We thank Taeok Bae for providing strains and constructs used in our initial experiments and for helpful discussions about the use of the reagents.
M.E.E., C.S., and C.O.R. conceived the experiments, M.E.E., C.S., and M.W.F. performed the experiments and analyzed the data, and all authors participated in writing and reviewing the paper. This work was supported by National Institutes of Health grants GM034496 (C.O.R.) and Cancer Center support grant CA21765 and the American Lebanese Syrian Associated Charities.