A Lipoylated Metabolic Protein Released by Staphylococcus aureus Suppresses Macrophage Activation

Summary

The virulence factors of pathogenic microbes often have single functions that permit immune suppression. However, a proportion possess multiple activities, and are considered moonlighting proteins. By examining secreted virulence factors of Staphylococcus aureus, we determine that the bacterial lipoic acid synthetase, LipA, suppresses macrophage activation. LipA is known to modify the E2 subunit of the metabolic enzyme complex pyruvate dehydrogenase (E2-PDH) with a fatty acid derivative, lipoic acid, yielding the metabolic protein lipoyl-E2-PDH. We demonstrate that lipoyl-E2-PDH is also released by S. aureus and moonlights as a macrophage immunosuppressant by reducing Toll-like receptor 1/2 (TLR1/2) activation by bacterial cell wall lipopeptides. A LipA-deficient strain induces heightened pro-inflammatory cytokine production, which is diminished in the absence of TLR2. During murine systemic infection, LipA suppresses pro-inflammatory macrophage activation, rendering these cells inefficient at controlling infection. These observations suggest that bacterial metabolism and immune evasion are linked by virtue of this moonlighting protein.

Graphical Abstract

Macrophages are crucial to the control of bacterial infection. Grayczyk et al. show that the lipoylated E2 subunit of pyruvate dehydrogenase from Staphylococcus aureus moonlights as an immunosuppressant and blocks macrophage activation by triacylated lipopeptides. The synthesis of lipoyl-proteins by S. aureus suppresses macrophage responses during infection and promotes virulence.

Introduction

Upon breaching physical barriers to infection, innate immunity initiates the rapid clearance of invading pathogens. Microbes thwart this response through the production of myriad secreted virulence factors (Finlay and Falkow, 1997). Many of these virulence factors perform distinct functions to promote microbial survival. However, a growing number of proteins with functions not traditionally linked to virulence have been found to promote pathogenesis through moonlighting activities (Henderson and Martin, 2011; Henderson and Martin, 2013; Jeffery, 1999). Many bacterial moonlighting proteins have primary functions as metabolic enzymes or molecular chaperones and typically function in the cytosol. However, these proteins are also found in the extracellular environment where they contribute to pathogenesis (Henderson and Martin, 2013). Identification of protein moonlighting in bacterial pathogens highlights a need to understand the role of these proteins in host-microbe interactions.

Staphylococcus aureus can cause disease in virtually any tissue site, most commonly manifesting as bacteremia, osteomyelitis, pneumonia, and skin infections (Tong et al., 2015). The bacterium evades phagocytic leukocytes by producing virulence factors that inhibit their antimicrobial activity (Thammavongsa et al., 2015). Macrophages are professional phagocytic leukocytes that are central to innate defenses. When activated by pathogen associated molecular patterns (PAMPs), macrophages kill microbial invaders through phagocytosis and induction of oxidative burst (Flannagan et al., 2009; Shepherd, 1986). Macrophages secrete cytokines and chemokines to regulate the innate and adaptive immune system (Mosser and Edwards, 2008). Thus, macrophages represent a significant mediator of immunity that S. aureus must overcome in order to cause disease. A growing body of work has begun to uncover mechanisms used by S. aureus to evade macrophages (Alonzo III and Torres, 2014; Flannagan et al., 2015; Peschel et al., 1999; Peschel et al., 2001). Nevertheless, the repertoire of S. aureus virulence factors that disrupt macrophage functions remains understudied.

We conducted a forward genetic screen to identify S. aureus secreted factors that perturb macrophage inflammatory responses. A mutation in the gene encoding the lipoic acid synthetase (LipA), which is required for the synthesis of lipoic acid, resulted in enhanced TLR2-dependent activation of macrophages. The hyper-inflammatory response elicited by a lipA mutant correlated with the absence of lipoylation on the E2 subunit of the pyruvate dehydrogenase complex (E2-PDH). Purified S. aureus lipoyl-E2-PDH prevented TLR1/2 activation by triacylated lipopeptides. Moreover, S. aureus lipoyl-protein production in vivo resulted in impaired activation of inflammatory macrophages and reduced host control of bacterial growth and dissemination. Overall, we report that S. aureus lipoyl-E2-PDH moonlights as an immune evasion protein by suppressing TLR-mediated macrophage activation.

Results

Identification of S. aureus Mutants That Enhance or Reduce Macrophage Activation

To determine if S. aureus releases extracellular factors that perturb macrophage function, we devised a forward genetic screen using cell free supernatants derived from 1920 annotated transposon mutants of the epidemic S. aureus clone, USA300 (JE2) (Fey et al., 2013). Supernatants were added to murine bone marrow derived macrophages (BMM) followed by measuring pro-inflammatory cytokine production using multiplexed cytokine bead arrays (Figure 1A). We identified 21 mutants that enhanced or reduced macrophage cytokine secretion (Figure S1A and Table S1). An insertion in the gene that encodes the lipoprotein signal peptidase (lspA), NE1757, led to reduced macrophage activation consistent with its role in processing lipoproteins – the bacterial PAMP recognized by TLR1/2 and TLR2/6 heterodimers – and served as an internal validation of the screen (Figure 1B) (Takeuchi et al., 2001; Takeuchi et al., 2002; Wardenburg et al., 2006). Of the other mutants identified, we prioritized NE264 for further study given its marked induction of IL-6, TNF, CCL3, and CCL4 (Figure 1B). NE264 contains an insertion in the gene that encodes the lipoic acid synthetase, LipA, an enzyme required for synthesis of lipoic acid (Cronan, 2016). Lipoic acid is an organosulfur molecule that is covalently attached to conserved lysines within E2 subunits of metabolic complexes including: pyruvate dehydrogenase (PDH), 2-oxoglutarate dehydrogenase (OGDH), and branched-chain 2-oxoacid dehydrogenase (BCODH) complexes, as well as to the H protein of the glycine cleavage system (Gcs) (Cronan, 2016; Spalding and Prigge, 2010). Prior studies indicate that lipoic acid synthesis and acquisition pathways are crucial for S. aureus infection (Zorzoli et al., 2016). Therefore, we sought to determine the mechanism behind the hyper-inflammatory response elicited by NE264.

Figure 1. LipA Suppresses TLR2-dependent Activation of Macrophages.

(A) Setup of screen used to identify macrophage immunomodulatory factors. Transposon mutants were grown to an OD600 of ~1.2, followed by addition of cell free supernatants to BMM for 24 hours. BMM supernatants were collected and assessed for cytokine and chemokine secretion using cytometric bead array (CBA).

(B) Relative abundance of IL-6, TNF, CCL3, and CCL4 produced by macrophages after addition of cell free supernatants from JE2 (WT), NE1757 (lspA::erm), NE264 (lipA::erm).

(C–E) Growth curves of WT, ΔlipA, or ΔlipA+lipA in TSB or RPMI medium with and without 25 nM lipoic acid. Coomassie stained SDS-PAGE gels of TCA precipitated exoproteins after growth in TSB or RPMI+BCFA.

(F) IL-6, TNF, CCL3, and CCL4 production (pg/mL) by BMM after addition of supernatant from WT, ΔlipA, or ΔlipA+lipA grown in RPMI+BCFA. Data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3). *, p<0.05; **, p<0.01 by 1-way ANOVA with Bonferonni-Sidak post-test.

(G) BMM production of IL-6 and TNF (pg/mL) after addition of cell-free supernatant from WT, ΔlipA, or ΔlipA+lipA to WT, TLR2^−/−, TLR4^−/−, or MyD88^−/− BMMs. Data shown are from one of at least three experiments conducted in triplicate. (−), media alone. Means ± SD are shown (n = 3). **, p<0.01; ***, p<0.001, by 1-way ANOVA with Bonferonni-Sidak post-test.

(H) NFκB activation after treatment of RAW 264.7 cells, containing an NFκB inducible secreted embryonic alkaline phosphatase reporter, with cell-free supernatant from WT, ΔlipA, or ΔlipA+lipA. Relative reporter activity (Absorbance 450 nm) from one of two experiments conducted in triplicate is shown. Means ± SD are shown. **, p<0.01; ****, p<0.0001, by 1-way ANOVA with Bonferonni-Sidak post-test.

(I) Whole cell lysates or TCA precipitated exoproteins from the indicated S. aureus strains collected after growth in RPMI+BCFA followed by immunoblotting for lipoic acid-containing proteins. See also Figure S1 and Table S1.

S. aureus Attenuation of Macrophage Activation Requires the Lipoic Acid Synthetase

To confirm that disruption of lipA led to heightened macrophage activation, we used an in-frame deletion mutant (ΔlipA) and complementation strain (ΔlipA+lipA). A ΔlipA mutant grew identically to WT in tryptic soy broth (TSB), but not Roswell Park Memorial Institute (RPMI) medium lacking free lipoic acid (Figure 1C and 1D). When free lipoic acid was supplemented into RPMI, a ΔlipA mutant grew identically to the WT strain (Figure 1D) (Zorzoli et al., 2016). The growth defect of a ΔlipA mutant was rescued in RPMI bypass medium containing branched chain carboxylic acids and sodium acetate (BCFA), leading to equivalent optical density after 7 hours and identical exoprotein profiles (Figure 1E and S1B). The addition of ΔlipA cell free supernatant, derived from cultures grown in BCFA medium, to BMMs enhanced secretion of the pro-inflammatory cytokines and chemokines IL-6, TNF, CCL3, and CCL4 (Figure 1F). Together, these data imply that LipA is required for synthesis of lipoic acid and suppresses BMM activation.

LipA Restriction of Macrophage Activation Occurs through TLR2

Immune cells sense and respond to S. aureus through TLR2-based recognition of lipoproteins via signaling through the adaptor protein MyD88 (Fournier and Philpott, 2005; Takeuchi et al., 2000). To determine if lipA-mediated immune suppression occurs through a TLR2-dependent pathway, we evaluated activation of WT, TLR2^−/−, TLR4^−/− or MyD88^−/− BMM treated with supernatant from WT, ΔlipA, or ΔlipA+lipA. Use of TLR2^−/− and MyD88^−/− BMM abrogated the enhanced secretion of IL-6, TNF, CCL3, and CCL4 elicited by ΔlipA supernatant (Figure 1G and S1C–D). Signaling through TLR2 leads to NFκB activation and induction of pro-inflammatory cytokine and chemokine gene expression. Indeed the increased production of IL-6, TNF, CCL3, and CCL4 correlates with increased NFκB dependent gene expression (Figure 1H). This enhanced activation by ΔlipA mutant supernatant is not limited to macrophages as murine neutrophils purified from the bone marrow also exhibited enhanced inflammatory chemokine secretion (Figure S1E). Because cell free supernatants were used in all immune cell activation studies, we postulated that the secretome of S. aureus might be disrupted in a ΔlipA mutant leading to enhancement of TLR2-dependent macrophage activation. However, we found that a ΔlipA mutant does not exhibit major alterations in its exoproteins compared to WT (Figures 1C and 1E). Instead, we found that of the four cytosolic lipoylated proteins produced by S. aureus, one is released and contains a lipoyl moiety, but is absent in the ΔlipA mutant (Figure 1I and S1F). Immunoblot of exoproteins from a ΔE2-pdh mutant indicated that this released lipoyl-protein is E2-PDH (Figure 1I and S1F). These data suggest LipA reduces TLR2-dependent phagocyte activation and WT S. aureus releases lipoyl-E2-PDH.

Release of Lipoyl-E2-PDH by S. aureus Coincides with Macrophage Suppression

Lipoic acid acquisition by S. aureus occurs through either synthesis or salvage (Zorzoli et al., 2016 and Figure 2A). LipA, LipM, and LipL are required for synthesis, while LplA1 and LplA2 are involved in salvage. LipA and LipM contribute to the direct synthesis of the lipoyl moiety and LipL is required for its transfer to E2-PDH and E2-BCODH. We hypothesized that deletion of lipM or lipL would phenocopy the inflammatory response elicited by BMM treated with ΔlipA supernatant. We found that all mutants lacking lipoyl-E2-PDH (ΔlipA, ΔlipM, and ΔlipL), but not mutants of the salvage pathway (ΔlplA1 and ΔlplA2) elicited higher activation of BMM (Figure 2B–C and S2A). A ΔlipL mutant enhanced macrophage activation even with lipoyl-E2-OGDH and lipoyl-GcvH present in the cytosol (Figure 2B–C and S2A). These data indicate that de novo lipoic acid synthesis is required to suppress BMM activation and links the presence of lipoyl-E2-PDH in the supernatant to suppression of macrophage activation.

Figure 2. The Release of lipoyl-E2-PDH by S. aureus Requires Lipoic Acid Synthesis and likely involves Atl.

(A) Model of the S. aureus lipoic acid synthesis and salvage pathways. ACP, Acyl carrier protein; GcvH, H protein of the glycine cleavage system; E2s, E2 subunits of PDH, OGDH, and BCODH. LipM, octanoyl transferase; LipA, lipoyl synthetase; LipL, lipoyl transferase; LplA1/A2, lipoate protein ligases.

(B and E) Whole cell lysates or TCA precipitated exoproteins from the indicated S. aureus strains collected after growth in RPMI+BCFA followed by immunoblotting for lipoic acid-containing proteins.

(C) IL-6, TNF, CCL3, and CCL4 production (pg/mL) by BMM after addition of supernatant from the indicated strains. Data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3). **, p<0.01; ***, p<0.001; ****, p<0.0001 by 1-way ANOVA with Bonferonni-Sidak post-test.

(D) IL-6 and TNF production (pg/mL) by BMM after addition of T3, T5, or T9 supernatant from the indicated strains. Data shown are from one of at least three experiments conducted in triplicate. (−), media alone. Means ± SD are shown (n = 3). *, p<0.05; **, p<0.001; ***, p<0.001; ****, p<0.0001 by 1-way ANOVA with Bonferonni-Sidak post-test. NS, not significant. See also Figure S2.

The detection of cytosolic proteins in the extracellular environment, such as E2-PDH, has been observed for numerous bacteria (Gründel et al., 2015; Lenz et al., 2003; Pasztor et al., 2010; Thomas et al., 2013). In S. aureus, we detected enhancement of macrophage activation with ΔlipA supernatant as early as 3 hours after subculture into BCFA medium and detectable E2-PDH in supernatant as early as 5 hours (Figure 2D and S2B–C). Listeria monocytogenes surface expression of E2-PDH is partially dependent on the secondary secretion system, SecA2 (Lenz et al., 2003), whereas the autolysin Atl is required for the release of some cytosolic proteins in S. aureus (Pasztor et al., 2010). Using transposon mutants of atl and secA2, we found that SecA2 likely contributes to, while Atl may be necessary for the release of lipoyl-E2-PDH (Figures 2E and S2D–E). Comparisons to purified recombinant lipoyl-E2-PDH suggest that S. aureus supernatant contains ~25–50 nM released lipoyl-E2-PDH that remains stable in culture for over 16 hours (Figure S2F–G). Thus, S. aureus releases nanomolar quantities of lipoyl-E2-PDH throughout the growth cycle and likely requires Atl-dependent lysis for release.

Lipoyl-E2-PDH is Sufficient to Suppress Triacylated Lipopeptide-Dependent Macrophage Activation

Free lipoic acid suppresses the respiratory burst of innate cells when used at high concentrations (Fei et al., 2016; O’Neill et al., 2008). Therefore, we hypothesized that release of lipoyl-E2-PDH by S. aureus might confer immunosuppressive properties. We first determined if free lipoic acid suppresses BMM activation by S. aureus secreted factors. Supplementation of ΔlipA supernatant with 3 mM free lipoic acid reduced BMM secretion of IL-6, TNF, CCL3, and CCL4 (Figures 3A and S3A). 3 mM free lipoic acid also suppressed the activation of BMM by Pam2CSK4 and Pam3CSK4, synthetic diacylated or triacylated lipopeptides that induce TLR2/6 or TLR1/2 heterodimer signaling (Figures 3B and S3B–C). These data indicate free lipoic acid suppresses TLR2-dependent activation of BMM at supraphysiologic concentrations.

Figure 3. Lipoyl-E2-PDH Blunts TLR1/2-dependent Activation of Macrophages.

(A and B) IL-6 and TNF production (pg/mL) after addition of ΔlipA supernatant (A) or 3 ng/mL Pam2CKS4 (Pam2) and 30 ng/mL of Pam3CSK4 (Pam3) (B) to BMM in the presence of free lipoic acid (3 mM or 0.3 mM). (−), media alone. Data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3). *, p<0.05; ***, p<0.001; ****, p<0.0001 by 1-way ANOVA with Bonferonni-Sidak post-test.

(C and D) IL-6 and TNF production (pg/mL) after addition of 10 nM lipoyl-E2-PDH (LA-PDH) (C) or synthetic tripeptides DK^LA and DKA (D) to BMMs in the presence of 1 ng/mL of Pam2 or 3 ng/mL of Pam3. Data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3). *, p<0.05; **, p<0.01; ***, p<0.001; ****, p<0.0001 by 1-way ANOVA with Bonferonni-Sidak post-test.

(E) Coomassie-stained SDS-PAGE gel of 500 ng purified SitC. IL-6, TNF, CCL3, and CCL4 production (pg/mL) after addition of SitC (0.1 ng/ml for CCL3/CCL4 and 1.0 ng/ml for IL-6/TNF) to BMM in the presence of 10 nM synthetic tripeptides DK^LA and DKA. Data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3). *, p<0.05; **, p<0.01; ***, p<0.001; ****, p<0.0001 by 1-way ANOVA with Bonferonni-Sidak post-test. See also Figure S3.

Since free lipoic acid suppresses macrophage activation we wondered whether lipoylated proteins – the main form of lipoic acid in living systems – from S. aureus are sufficient to suppress BMM activation at physiologically meaningful concentrations. We tested if lipoyl-E2-PDH suppresses BMM activation by co-incubating purified recombinant S. aureus lipoyl-E2-PDH with Pam2CSK4 and Pam3CSK4 and monitored cytokine secretion. We found that 10 nM lipoyl-E2-PDH suppressed TLR1/2 activation, but not TLR2/6 activation (Figures 3C and S3D–E). Furthermore, a synthetic tripeptide containing a lipoyl-lysine residue, DK^LA, but not the unmodified tripeptide, DKA, was sufficient to suppress TLR1/2 dependent macrophage activation at a 10 nM concentration (Figure 3D and S3F). DK^LA also suppressed macrophage activation by SitC, a TLR1/2 activating triacylated lipoprotein produced by S. aureus (Asanuma et al., 2011; Nguyen and Götz, 2016) (Figures 3E and S3G). These data indicate that lipoyl-E2-PDH is sufficient to suppress TLR1/2 activation of macrophages and that immune suppression is directly linked to the lipoyl modification.

Activity of the Lipoic Acid Synthetase Promotes S. aureus Pathogenesis

To determine if lipoyl-protein immunosuppression occurs in vivo, we induced systemic infection in mice by intraperitoneal injection of WT, ΔlipA, or ΔlipA+lipA and determined the levels of macrophage chemokines in the serum. After 16 hours, animals infected with a ΔlipA mutant had higher CCL3 and CCL4 in the serum, whereas at 72 hours chemokine levels were similar (Figure 4A–B). At 16 hours, the proportion of pro-inflammatory macrophages (CD11b⁺, F4/80⁺, CCR5⁺, I-A/I-E^high, Ly6G⁻) in the peritoneal cavity was identical among the infection groups (Alonzo III et al., 2013; Okabe and Medzhitov, 2014) (Figure 4C). However, at 72 hours, ΔlipA-infected animals had greater I-A/I-E^high/CCR5⁺ activated macrophages and I-A/I-E^high dendritic cells (Figure 4D and S4A). The total number of recruited macrophages, neutrophils, and dendritic cells were the same in all infection groups (Figure S4B). In summary, the proportion of activated pro-inflammatory phagocytes is higher in ΔlipA-infected animals.

Figure 4. LipA-dependent Suppression of Macrophage Activation Correlates with S. aureus Virulence.

(A and B) Serum CCL3 and CCL4 levels (pg/mL) 16 hours (A) or 72 hours (B) post-infection. Means ± SD are shown (WT and ΔlipA n=10, ΔlipA+lipA n=7). *, p<0.05 by 1-way ANOVA with Bonferonni-Sidak post-test. NS, not significant.

(C and D) Pro-inflammatory macrophages in the peritoneal cavity at 16 hours (C) and 72 hours (D) post-infection. Macrophages were gated on CD11b⁺ F4/80⁺ Ly6G⁻ cells followed by assessment of CCR5⁺ I-A/I-E^hi cells. Flow cytometry plots are representative of 4–8 animals per group. Scatter plots display percent CCR5⁺/I-A/I-E^hi cells within the CD11b⁺ F4/80⁺ Ly6G⁻ gate. Bars display the median. **, p<0.01; ***, p<0.001 by 1-way ANOVA with Bonferonni-Sidak post-test. NS, not significant.

(E and F) Bacterial burden (E) and medians of bacterial burden (F) in the peritoneal cavity and kidneys of mice 16 hours - WT (n=17), ΔlipA (n=19), ΔlipA+lipA (n=12), and 72 hours - WT (n=11), ΔlipA (n=12), ΔlipA+lipA (n=14) post IP infection. *, p<0.05; **, p<0.01; ****, p<0.0001 by non-parametric 1-way ANOVA (Kruskal-Wallis Test) with Dunn’s post-test. Dashed lines, limit of detection. N=7 and N=6 in the 72 hour dataset are the number of animals with undetectable CFU (E). NS, not significant.

(G) Survival/outgrowth of WT S. aureus after infecting WT (n=8), ΔlipA (n=8), or PBS (n=8) elicited F4/80⁺ cells. *, p<0.05, ****, p<0.0001 by 2-way ANOVA with Tukey’s post-test. See also Figure S4.

ΔlipA-infected animals had significantly fewer bacteria at 16 and 72 hours post-infection in the lavage fluid and kidney compared to WT or ΔlipA+lipA (Figure 4E). After 72 hours, a significant number of the ΔlipA-infected animals had undetectable bacteria in both the lavage fluid (N=7) and the kidneys (N=6) (Figure 4E). By 72 hours, the median bacterial CFU in ΔlipA-infected animals was near or at the limit of detection in both sites, while the median bacterial CFU of the WT and complement infected animals remained 1–2 logs higher indicating LipA promotes bacterial survival during infection (Figure 4F).

Because a greater proportion of macrophages in ΔlipA-infected mice have enhanced pro-inflammatory characteristics, we reasoned that these macrophages might have greater bactericidal activity. To test this hypothesis we isolated F4/80⁺ peritoneal cells elicited to the peritoneal cavity 72 hours after infection with WT and ΔlipA S. aureus, or mock infected with PBS. After antibiotic treatment to kill bacteria used for elicitation, we infected the sorted population of F4/80⁺ cells with WT S. aureus (Figure S4C). The F4/80⁺ peritoneal cells elicited after infection with a ΔlipA mutant inhibited the growth of S. aureus better than F4/80⁺ peritoneal cells elicited from WT or PBS treated mice (Figure 4G). Consistent with our in vitro findings, these data indicate macrophage activation is increased in ΔlipA-infected mice and coincides with better infection control.

Discussion

In this study, we found that the lipoic acid synthetase, LipA, suppresses the activation of macrophages. Immunosuppression was due to the release of lipoyl-E2-PDH, a moonlighting metabolic protein that reduces TLR1/2-dependent activation of macrophages via its lipoyl moiety. A higher proportion of activated macrophages are induced upon systemic infection with a ΔlipA mutant, coinciding with greater control of infection. Overall, these findings expand the repertoire of activities for moonlighting metabolic enzymes, and suggest that protein lipoylation has unanticipated immunosuppressive functions during infection.

The extracellular detection of lipoyl-E2-PDH is unexpected considering it lacks a discernable secretion signal sequence and functions in the cytosol as a component of the PDH complex. However, subunits of metabolic complexes, such as PDH, are found in the extracellular environment in bacteria and eukaryotes (Henderson and Martin, 2011; Henderson and Martin, 2013). In prior studies, S. aureus release of cytoplasmic proteins, including E2-PDH, correlated with cell division and the activity of the peptidoglycan-modifying enzyme, Atl (Ebner et al., 2016; Pasztor et al., 2010). Indeed, our studies support this notion, as release of lipoyl-E2-PDH appears to depend on Atl.

Lipoyl-E2-PDH suppresses TLR2 activation in nanomolar quantities while millimolar concentrations of free lipoic are needed. Others have noted that millimolar concentrations of free lipoic acid suppress respiratory burst of neutrophils and blunt the translocation of the transcription factor, NFκB, to the nucleus (Fei et al., 2016; O’Neill et al., 2008; Zhang and Frei, 2001). We hypothesize the 1000-fold difference in concentration required for immunosuppression may be due to rapid diffusion of free lipoic acid across cellular membranes where it is then attached to host metabolic proteins, whereas the protein or peptide-bound form cannot readily do so. It is also possible that free lipoic acid and lipoyl proteins suppress macrophage activation by alternate mechanisms. High concentrations of free lipoic acid broadly suppressed TLR2/6- and TLR1/2-dependent activation of macrophages while purified lipoyl-E2-PDH or DK^LA were only capable of suppressing TLR1/2-dependent activation of macrophages. It is known that high concentrations of free lipoic acid can also activate the phosphoinositide 3-kinase/Akt signaling pathway to reduce inflammatory cytokine production (Zhang et al., 2007), and has antioxidant capabilities (Packer et al., 1995). Therefore, suppression of activation by high concentrations of free lipoic acid may be due to its composite anti-inflammatory functions, whereas the TLR1/2-dependent suppression by low concentrations of lipoyl-protein is likely due to blunted signaling.

S. aureus lipoproteins are triacylated in standard growth conditions (Asanuma et al., 2011; Nguyen and Götz, 2016). Triacylated lipoproteins and peptides bind to TLR1 and 2 via interactions of the amide linked acyl chain with TLR1 and the cysteine-linked diacylglycerol with TLR2 (Jin et al., 2007; Kang et al., 2009). Lipoyl-PDH and lipoyl peptides suppress macrophage activation by Pam3CSK4, but not Pam2CSK4 suggesting a mechanism of blockade that is specific to TLR1/2. The crystal structure of TLR1/2 with Pam3CSK4 revealed a hydrophobic binding pocket in TLR1 that interacts with the amide bound lipid of Pam3CSK4 (Jin et al., 2007). This interaction is reminiscent of lipoamide binding within the hydrophobic pocket of E1 subunits (Ævarsson et al., 1999; Fujiwara et al., 2005). Furthermore, TLR1 preferentially recognizes peptide bound medium chain fatty acids (> 6 carbons) (Farhat et al., 2008; Huang et al., 2012; Lee et al., 2004). Our studies show that a tripeptide containing a lipoyl-lysine residue, DK^LA, but not an unmodified tripeptide, DKA, is sufficient to restrict macrophage activation. Thus, the lipoyl moiety is required for immune suppression. It is intriguing to surmise that the lipid-like structure of lipoic acid facilitates lipoyl-E2-PDH and DK^LA binding to TLR1 to suppress TLR1/2 activation.

A ΔlipA mutant elicited heightened levels of CCL3 and CCL4 in the serum and greater proportions of activated macrophages at the site of infection, which correlates well with our in vitro data. However, CCL3 and CCL4 levels were indistinguishable later in infection. We suspect that the early increase in macrophage chemokines is likely a meaningful early signature that promotes the enhanced recruitment and activation of macrophages at 72 hours in ΔlipA infected animals. At later time points of infection, when many ΔlipA infected mice have undetectable infection, we suspect the affects of lipoyl-E2-PDH on cytokine secretion will be less clear.

It is possible that in addition to its effect on immune cell activation, metabolic defects of a ΔlipA mutant may augment attenuation in vivo (Zorzoli et al., 2016). Three lines of evidence argue in favor of a meaningful contribution of immune cell activation to infection resolution: (i) activated macrophages elicited by a ΔlipA mutant were better able to inhibit S. aureus growth in ex vivo models (Figure 4G); (ii) in a bloodstream kidney abscess model, where phagocytic leukocytes are excluded from the central abscess (Cheng et al., 2009; Thammavongsa et al., 2013), enhanced macrophage activation is still observed in ΔlipA-infected animals yet bacterial burden is unaffected, suggesting a growth defect in the absence of macrophage infiltration is not sufficient for attenuation (Figure S4D); and (iii) ΔlipA mutants are equally susceptible to macrophage antimicrobial activity confirming that metabolic defects do not enhance killing (Figure S4E).

Overall, our observations suggest that bacterial metabolism is closely linked with the evasion of innate immune responses by virtue of the moonlighting activity of a metabolic protein. As the moonlighting roles of proteins are often conserved (Henderson and Martin, 2013), we suspect that other bacterial pathogens may promote immune evasion by similar mechanisms.

STAR Methods

Contact for Reagent and Resource Sharing

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Francis Alonzo (falonzo@luc.edu).

Experimental Model and Subject Details

Ethics Statement

All animal experiments were performed in ABSL2 facilities with protocols that are approved by Loyola University of Chicago, Health Sciences Division Institutional Animal Care and Use Committee (IACUC# 2014049) in accordance with guidelines set forth by the USDA and PHS Policy on Humane Care and Use of Laboratory Animals under the guidance of the Office of Laboratory Animal Welfare (OLAW). Loyola University Chicago, Health Sciences Division has an Animal Assurance on file with the Public Health Service (#A3117-01 approved through 02/28/2018), is a fully AAALAC International accredited institution (#000180, certification dated 11/19/2013), and is a USDA registered/licensed institution.

Mouse Experiments

Cohorts of six to eight week old female Hsd:ND4 Swiss Webster outbred mice were purchased from Envigo and housed under specific pathogen-free conditions at Loyola University of Chicago Medical campus (Maywood, IL) in an ABSL2 facility.

Cell Culture Experiments

Murine bone marrow macrophages (BMM) were differentiated from the bone marrow of six to eight week old female or male C57BL/6J mice and maintained in culture with bone marrow macrophage medium (BMM medium - DMEM (CellGro) + 1 mM Sodium Pyruvate (CellGro) + 1 mM HEPES Buffer (CellGro) + 2 mM L-glutamine (CellGro) + 20% heat inactivated FBS (Seradigm) + 30% L929 fibroblast cell supernatant + 100 μg/ml Penicillin/Streptomycin (Pen/Strep) (CellGro) + 50 μM β-mercaptoethanol (Amresco) at 37°C, 5% CO_2. Murine bone marrow neutrophils were isolated from the bone marrow of six to eight week old female or male C57BL/6J mice and maintained in culture with DMEM (Corning) + 10% heat inactivated FBS supplemented with 100 μg/ml Penicillin/Streptomycin at 37°C, 5% CO_2. NFκ^B reporter macrophages (RAW-Blue – Invivogen) were maintained in culture with DMEM (Corning) + 10% heat inactivated FBS supplemented with 100 μg/ml Penicillin/Streptomycin at 37°C, 5% CO_2.

Method Details

Bacterial Strains and Culture Conditions

All sources of bacterial strains and plasmids used in this study are listed in the Key Resources Table. The WT and parental strain for all S. aureus mutants used in this study was the pulse field gel electrophoresis type USA300 isolate LAC cured of its plasmids (wild type – AH1263), unless noted otherwise. S. aureus RN4220 is a NCTC8325 derivative which is restriction deficient for passage of plasmid. The T7 expressing Escherichia coli strain, lysY/I^q, was used as a host to overexpress and purify recombinant 6x-Histidine tagged S. aureus E2-PDH. All S. aureus strains were either grown in rich medium - Tryptic Soy Broth (TSB) (Criterion), or in defined medium lacking lipoic acid - Roswell Park Memorial Institute 1640 medium (RPMI) (Corning) supplemented with 1% casamino acids (Amresco). E. coli was cultivated in Lysogeny Broth, Miller formulation (LB) (BD Biosciences). Unless noted, all bacteria were grown at 37°C in a shaking incubator at 200 rpm in tubes kept at a 45° angle. When solid medium was required, agar (Amresco) was supplemented into TSB and LB at 1.5%.

KEY RESOURCES TABLE.

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Anti-Lipoic Acid Rabbit pAb antibody	EMD Millipore	Cat#437695-100UL; RRID:AB_212120
IgG from human serum	Sigma	Cat#I2511; RRID:AB_1163604
Goat anti-Rabbit IgG (H+L) Secondary antibody, AP	Thermo Scientific	Cat#65-6122; RRID:AB_2533968
Goat anti-Rabbit IgG (H+L) Secondary antibody, HRP	Thermo Scientific	Cat#32460; RRID:AB_1185567
TruStain fcX(TM) (anti-mouse CD16/32) antibody (clone 93)	BioLegend	Cat#101320; RRID:AB_1574975
Alexa Fluor® 647 anti-mouse I-A/I-E antibody (clone M5/114.15.2)	BioLegend	Cat#107617; RRID:AB_493526
PE anti-mouse LY-6G antibody (clone 1A8)	BioLegend	Cat#127607; RRID:AB_1186104
Pacific Blue™ anti-mouse/human CD11b antibody (clone M1/70)	BioLegend	Cat#101224; RRID:AB_755986
APC/Cy7 anti-mouse CD49b (pan-NK cells) antibody (clone DX5)	BioLegend	Cat#108919; RRID:AB_2561457
FITC anti-mouse CD206 (MMR) antibody (clone C068C2)	BioLegend	Cat#141703; RRID:AB_10900988
Alexa Fluor® 700 anti-mouse CD11c antibody (clone N418)	BioLegend	Cat#117319; RRID:AB_528735
PE/Cy7 anti-mouse F4/80 antibody (clone BM8)	BioLegend	Cat#123113; RRID:AB_893490
Biotin anti-mouse CD195 (CCR5) antibody (clone HM-CCR5)	BioLegend	Cat#107003; RRID:AB_313298
APC anti-mouse Ly-6G antibody (clone 1A8)	BioLegend	Cat#127613; RRID:AB_1877163
FITC anti-mouse/human CD11b antibody	BioLegend	Cat#101205; RRID:AB_312788
Biotin anti-mouse F4/80 antibody (clone BM8)	BioLegend	Cat#123106; RRID:AB_893501
PerCP/Cy5.5 Streptavidin	BioLegend	Cat#405214
Bacterial and Virus Strains
DH5α E. coli	New England Biolabs	Cat#C2989K
T7 expressing lysY/Iq E. coli	New England Biolabs	Cat#C3013I
T7 expressing lysY/Iq ΔlipA E. coli	This study	N/A
T7 expressing lysY/Iq ΔlipA with E2-PDH in pET15b	This study	N/A
DH5α E. coli with pOS1-PsarA-sodRbs-SitC-6x-His	This study	N/A
USA300 LAC (AH-1264)	Boles et al., 2010	N/A
USA300 LAC	Laboratory of Victor Torres	N/A
NCTC8325 derivative RN4220	Laboratory of Victor Torres	N/A
RN4220 with pOS1-PsarA-sodRbs-SitC-6x-His	This study	N/A
Nebraska Transposon Mutant Library of USA300 JE2	Fey et al., 2013	N/A
AH-LAC ΔlipA	Zorzoli et al., 2016	N/A
AH-LAC ΔlipA + ΔlipA	Zorzoli et al., 2016	N/A
AH-LAC ΔE2-PDH	Zorzoli et al., 2016	N/A
AH-LAC ΔlipM	Zorzoli et al., 2016	N/A
AH-LAC ΔlplA1	Zorzoli et al., 2016	N/A
AH-LAC ΔlplA2	Zorzoli et al., 2016	N/A
AH-LAC ΔlipL	Zorzoli et al., 2016	N/A
atl::erm (NE460)	Fey et al., 2013	N/A
secA2:::erm (NE66)	Fey et al., 2013	N/A
Biological Samples
Chemicals, Peptides, and Recombinant Proteins
RPMI 1640	Corning	Cat#50-020-PC
Isobutyric acid	Sigma	Cat#I1754-100mL
Isovaleric acid	Sigma	Cat#129542-100mL
2-methylbutyric acid	Alfa Aesar	Cat#A11546
(±)-α-Lipoic acid	Sigma	Cat#T5625-5G
Trichloroacetic acid, 99%	Alfa Aesar	Cat#A11156
Gelcode™ Blue Stain Reagent	Thermo Scientific	Cat#24592
0.1mm Glass cell disruption beads	Scientific Industries	Cat#SI-BG01
5-bromo-4-chloro-3-indoyl phosphate	Amresco	Cat#0885-1G
Nitro blue tetrazolium	Amresco	Cat#0329-1G
Polymyxin B sulfate	Alfa Aesar	Cat#J63074
Pam2CSK4	Invivogen	Cat#TLR-PM2S-1
Pam3CSK4	Invivogen	Cat#TLRL-PMS
Lysostaphin	Ambi Products	Cat#LSPN-50
ACK Lysing Buffer (1X)	Lonza	Cat#10-548E
BD IMag Streptavidin Particles Plus-DM	BD Biosciences	Cat#557812
Saponin from quillaja bark	Sigma	Cat#57900-256
Lipoyl-E2-PDH	This study	N/A
SitC	This study	N/A
1-Step™ PNPP	Thermo Scientific	Cat#37621
Histopaque®-1077	Sigma	Cat#10771-100ML
Histopaque®-1119	Sigma	Cat#11191-100ML
Percoll™	GE Healthcare	Cat#17-0891-01
2,2,2-tribromoethanol	Sigma	Cat#T48402-25G
DKLA	AnaSpec	N/A
DKA	AnaSpec	N/A
Critical Commercial Assays
Cytometric Bead Array Flex set (Custom)	BD Biosciences	N/A
Wizard® Genomic DNA Purification Kit	Promega	Cat#A1120
Deposited Data
Experimental Models: Cell Lines
Mouse: L-929 fibroblasts	Laboratory of Victor Torres	N/A
Murine Bone Marrow Macrophages	This study	N/A
Murine Bone Marrow Neutrophils	This study	N/A
Mouse: RAW-Blue SEAP reporter cells	Invivogen	Cat#raw-sp
Experimental Models: Organisms/Strains
C57BL/6J mice	Jackson Laboratory	Cat#000664
Hsd:ND4 Swiss Webster mice	Envigo	Cat#032
Oligonucleotides
See Table S2 for full oligonucleotide list
Recombinant DNA
Plasmid: pET15b	Laboratory of Victor Torres (Novagen)	N/A
Plasmid: pKD4	Datsenko and Wanner, 2000	N/A
Plasmid: pKD46	Datsenko and Wanner, 2000	N/A
Plasmid: pOS1	Laboratory of Victor Torres	N/A
Plasmid: pOS1-pSarA-sodRBS-sGFP	Laboratory of Victor Torres	N/A
Software and Algorithms
FlowJo	FlowJo,LLC	N/A
Prism	GraphPad Software	N/A
Other

Where necessary, media was supplemented with antibiotics at the following final concentrations for S. aureus: erythromycin (Erm), 3 μg/ml (Amresco); kanamycin (Kan), 50 μg/ml (Amresco); neomycin (Neo), 50 μg/ml (Amresco); and chloramphenicol (Cam), 10 μg/ml (Amresco). Antibiotics for E. coli were supplemented at the following final concentrations: ampicillin (Amp), 100 μg/ml (Gold Biotechnology); kanamycin (Kan), 10 or 25 μg/ml (Amresco); and neomycin (Neo), 10 or 25 μg/ml (Amresco). RPMI medium used to bypass the requirement for lipoic acid (RPMI+BCFA) was supplemented with the following short branched-chain carboxylic acids: 11.23 mM isobutyric acid (Sigma), 9.5 mM 2-methylbutyric acid (Alfa Aesar), 9.69 mM isovaleric acid (Sigma), and 10 mM sodium acetate (Amresco), pH 7.4–7.5.

Bacterial Growth Curves

Growth curves of WT, ΔlipA, or ΔlipA+lipA were carried out in TSB and RPMI, or RPMI supplemented with 25 nM α-lipoic acid (Sigma) or with BCFA (RPMI+BCFA). Overnight cultures were prepared in triplicate in either TSB or RPMI+BCFA in a 96-well round-bottom polystyrene plate (CellTreat). The next day, cultures were washed three times in TSB or RPMI and the triplicate samples were subcultured 1:100 and grown in a 96-well flat-bottom polystyrene plate at 37°C with shaking at 200 rpm. Bacterial growth w as assessed hourly for 10 hours by measuring optical density at 600 nm (OD600) using an ELx800 microplate reader (BioTek) until reaching stationary phase (~10 hours).

Isolation of Bone Marrow-Derived Macrophages

Primary murine bone marrow macrophages (BMM) were derived from bone marrow progenitor cells isolated from the femurs and tibias of C57BL/6 (WT, TLR2^−/−, TLR4^−/−, MyD88^−/−) mice. 5 × 10⁶ progenitor cells were plated in 100 × 26-mm Petridishes containing 15 mL BMM medium supplemented with 100 μg/ml Penicillin/Streptomycin (Corning). After 3 days at 37°C, 5% CO₂, 10 mL of fresh BMM medium was added and macrophages were allowed to differentiate for 6 more days. To remove differentiated BMM from petri dishes, the cells were rinsed with 10 mL 1X PBS, and then incubated with 10 mL 1X PBS at 4°C for 30 minutes. Following incubation, cells were removed by manual pipetting and 1 × 10⁷ cells were resuspended in BMM medium containing 10% DMSO (Sigma) and stored in liquid nitrogen until use. For experiments using frozen BMM, the frozen vial was thawed at 37°C and then slowly diluted dropwise into fresh BMM medium, pelleted at 1500 rpm at 4°C for 5 minutes, and then resuspended in 10 ml of fresh BMM medium. The cell suspension was subsequently split into two 100 × 26-mm Petri plates containing 10 mL BMM medium with 100 μg/ml Penicillin/Streptomycin, and was cultured for 3 days at 37°C, 5% CO₂ before use.

Isolation of Bone Marrow Neutrophils

Murine bone marrow neutrophils were isolated from the femurs and tibias of C57BL/6 WT mice by flushing out the bone marrow cells with DMEM + 10% heat inactivated FBS. Cells were then centrifuged at 1500 rpm at 4°C for 5 minutes, the supernatant was decanted, and red blood cells were lysed with ACK lysing buffer (Lonza) by resuspending the pellet in 2 mL of lysis buffer and incubating for 2 minutes at room temperature. Cell lysis was stopped by adding 8 mL of 1X sterile PBS, mixing gently, and pelleting at 1500 rpm, 4°C for 5 minutes. The cells were then resuspended in 10 mL of DMEM + 10% heat inactivated FBS, counted and resuspended in 1 mL of 1X PBS after pelleting at 1500 rpm at 4°C for 5 minutes. To isolate neutrophils a density gradient centrifugation was used. 3 mL of Histopaque 1077 (density 1.077 g/mL) (Sigma) was gently overlaid on 3 mL of Histopaque 1119 (density 1.119 g/mL) (Sigma) followed by addition of the bone marrow cell suspension on top of the Histopaque 1077 layer. The samples were then centrifuged for 30 minutes at 2000 rpm 25°C without brake and neutrophils were collected at the interface between the Histopaque 1119 and 1077 layers. Purified neutrophils were then washed twice by centrifuging at 1500 rpm at 4°C for 5 minutes and resuspending with DMEM + 10% heat inactivated FBS supplemented with 100 μg/ml Pen/Strep (Corning) and 50 μg/ml gentamicin (Amresco). Neutrophils were then seeded at 65000 cells per well into 96-well flat-bottom tissue culture treated plates (Corning) in 90 μl DMEM + 10% heat inactivated FBS supplemented with 100 μg/ml Pen/Strep and 50 μg/ml gentamicin. The neutrophils were activated with 10 μl OD-normalized supernatants of 9 hour S. aureus cultures followed by incubation at 37°C, 5% CO₂ for 24 hours. Supernatants were then removed and either used immediately or stored at −80°C. To measure the levels of cytokines and chemokines in the neutrophil supernatants, a custom Cytometric Bead Array Flex set (BD Biosciences) was used according to manufacturer’s specifications. Bead and supernatant mixtures were incubated for 1.5 hours at 25°C at 600 rpm using a Thermo Mixer C (Eppendorf) in a 96-well V-bottom plate (Corning). Samples were then washed using FACS Wash Buffer (1X PBS + 2% heat inactivated FBS + 0.05% (w/v) Sodium Azide), data collected on an LSR Fortessa (BD Biosciences), and analyzed using FlowJo software (FlowJo, LLC) by gating on individual bead populations and calculating the geometric mean of fluorescence relative to protein standards.

To measure the purity of the isolated bone marrow neutrophils, 100000 cells were transferred to a 96-well V-bottom plate (Corning) for cell surface staining. Prior to staining, immune cells were incubated with 50 μL of FACS Wash Buffer containing 0.2 μg/mL of anti-CD16/CD32 (93) (BioLegend) for 30 minutes on ice to block Fc receptors followed by washing with FACS Wash Buffer and surface staining with the following antibodies from BioLegend: anti-CD11b-FITC (M1/70) and anti-Ly6G-APC (1A8) for 30 minutes on ice and washed twice with FACS Wash Buffer. Samples were then washed twice and fixed using FACS Fixing Buffer (1X PBS + 2% heat inactivated FBS + 2% Paraformaldehyde + 0.05% (w/v) sodium azide). Data was collected on an LSR Fortessa (BD Biosciences) and subsequently analyzed using FlowJo software (FlowJo, LLC). Isolated bone marrow neutrophils were 85–90% pure.

Transposon Mutant library screen on BMM

An annotated transposon mutant library generated in USA300 strain JE2 was used for screening (BEI resources repository). Overnight cultures were prepared by inoculating polystyrene plates (Corning) containing sterile TSB with transposon mutants that had been spot plated and grown on solid agar in 12×8 arrays (20 plates, each containing a single 12×8 array of mutant strains) the night prior. Following overnight growth at 37°C, the mutant library was subcultured 1:100 by inoculating 2 μl of a well-mixed overnight culture into 198 μl fresh TSB in 96-well polystyrene plates (Corning). The cultures were allowed to grow for 9 hours (OD600 = 1.2) at 37°C with shaking at 200 rpm. After 9 hours, the OD600 of the strains was measured using an Envision 2103 plate reader (Perkin-Elmer) and the bacteria were pelleted at 3700 rpm for 15 minutes at 4°C. Cell free supernatant was collected and used immediately for screening or stored at −80°C. 65000 BMM derived from bone mar row of 6–8 week old male and female C57BL/6 mice were seeded into 96-well flat-bottom tissue culture treated plates (Corning) in BMM medium with 100 μg/ml Pen/Strep (Corning) and 50 μg/ml gentamicin (Amresco) the day before use. The following day, supernatants isolated from 9 hour cultures of S. aureus transposon mutants were thawed from storage at −80°C on ice followed by addition of 10 μl OD normalized supernatants to macrophages in 90 μl BMM medium and incubation at 37°C, 5% CO₂ for 24 hours. Macrophage supernatants were then removed and either used immediately in cytometric bead array or stored at −80°C for later analysis.

To measure the levels of cytokines and chemokines in the macrophage supernatants, a custom Cytometric Bead Array Flex set (BD Biosciences) was used according to manufacturer’s specifications. Bead and supernatant mixtures were incubated for 1.5 hours at 25°C at 600 rpm using a Thermo Mixer C (Eppendorf) in a 96-well V-bottom plate (Corning). Samples were then washed using FACS Wash Buffer (1X PBS + 2% heat inactivated FBS + 0.05% (w/v) Sodium Azide), data collected on an LSR Fortessa (BD Biosciences), and analyzed using FlowJo software (FlowJo, LLC) by gating on individual bead populations and calculating the geometric mean of fluorescence relative to protein standards. From the 1920 mutants in the library, 96 were initially selected for rescreening based upon an altered abundance that fell at least 1 standard deviation outside the average response induced by all 1920 mutants. These 96 mutants were rescreened (Figure S1A). Those mutants whose cytokine abundance remained 1 standard deviation above or below that of wild type induction levels were selected and are displayed in Table S1. ++ and −− in Table S1 refers to those mutants whose cytokine levels fell one standard deviation beyond wild type in the positive or negative direction respectively. For an NE264 mutant, MCP1 and KC levels were highly variable between biological replicates, therefore these cytokine changes were designated ++/ND in Table S1. All 21 mutants were rescreened in triplicate to confirm statistically significant enhancement or diminution of cytokine secretion relative to WT controls. After confirmation, the mutations were subsequently transduced into S. aureus USA300 strain LAC to confirm phenotypes before moving forward with prioritization of mutants for further studies. All transduced mutants were screened in triplicate and assessed for statistically significant differences in cytokine abundance compared to wildtype.

S. aureus Cell Free Supernatant Preparation

S. aureus overnight cultures were prepared in triplicate by inoculating three wells with an individual bacterial colony in a 96-well round-bottom polystyrene plate containing RPMI+BCFA medium that was filtered sterilized using a 0.22 μM PES syringe filter (Corning) into a plastic conical tube. Following overnight growth at 37°C with shaking at 200 rpm, triplicate samples were subcultured by inoculating 3 μl of a well-mixed overnight culture into 147 μl RPMI+BCFA. The cultures were allowed to grow for 3, 5, or 9 hours at 37°C with shaking at 200 rpm. After 3, 5 or 9 hours, the OD600 of the strains was measured using an ELx800 microplate reader (BioTek) for OD normalization and the bacteria were pelleted at 3700 rpm for 15 minutes at 4°C. Cell free supernatant was collected and used immediately or stored at −80°C.

Purification of S. aureus Lipoyl-E2-PDH

Recombinant E2-PDH was expressed and purified from E. coli lysY/I^q containing a ΔlipA::kan mutation. The ΔlipA::kan mutant was generated using lambda red mutagenesis (Datsenko and Wanner, 2000). In brief, a primer pair (LipAP1F/LipAP2R) was designed containing 50 bp of homology upstream and downstream to the lipA gene in E. coli and homology to the priming regions of pKD4 (gift of Dr. Alan Wolfe). These primers were then used to PCR amplify the kanamycin resistance cassette encoded in pKD4. Lambda red mutagenesis was carried out by electroporating this amplicon into competent lysY/I^q expressing pKD46 (gift of Dr. Alan Wolfe) and plating on LB containing either 10 or 25 μg/ml of Kan and Neo. Antibiotic resistant colonies were patched and subsequently assessed for replacement of the lipA gene with the pKD4 Kan resistance cassette using PCR with primer pair LipAF/LipAR. To generate a 6x-His-Tagged E2-PDH expression plasmid, WT genomic DNA was isolated and amplified with the primer pair 0995hisN/CF/0995hisNR to generate an E2-PDH encoding amplicon flanked by NdeI and BamHI restriction sites. This amplicon was ligated into a pET15b expression plasmid which was subsequently transformed into the ΔlipA::kan lysY/I^q strain. To confirm expression of the N terminal 6× Histidine tagged E2-PDH, protein production was assessed after induction with 1 mM IPTG at 37°C for 3 hours (Gold Biotechnology).

To purify E2-PDH, lysY/I^q ΔlipA::kan E. coli containing the pET15b-6x-His-E2-PDH plasmid was grown overnight in 30 ml LB (BD Biosciences) supplemented with 100 μg/ml ampicillin (Gold Biotechnology) at 37°C, 220 rpm. The following day, the bacteria were subcultured 1:100 in 4 L fresh LB medium and grown for 20 hours at 37°C until reaching an OD600 of 0.25–0.3 followed by addition of IPTG (0.5 mM) (Gold Biotechnology) and incubation for 4 hours at 37°C at 220 rpm. Bacteria were pelleted by centrifugation at 8500 rpm for 10 minutes at 4°C followed by storage at −80°C. The following day, bacterial pellets were thawed at 37°C and resuspended in lysis buffer (25 mM imidazole (Alfa Aesar), 50 mM Tris-HCL, 300 mM NaCl (Amresco), pH 8) supplemented with 1 mM dithiothreitol (DTT) (Amresco) and 1 mM phenylmethane sulfonyl fluoride (PMSF) (Acròs Organics). Bacteria were lysed on ice using a sonicator (Branson) at 20-second intervals with a rate of 0.8 seconds per pulse and an output of 340 W for a total of 15 minutes. The bacterial debris was pelleted for 45 minutes at 11000 rpm, filtered through a 0.45 μm filter followed by incubation of the clarified supernatant for 1 hour with 1 ml nickel-NTA resin (Qiagen). The resin was washed with 150 mL 50 mM imidazole, 1 mM DTT, 50 mM Tris-HCL, 300 mM NaCl, pH 8 followed by elution in the same buffer containing 500 mM imidazole. The purified protein was dialyzed using snakeskin dialysis tubing (10 kDa MWCO, Thermo Scientific) into 100 mM imidazole + 50 mM Tris-HCL, 300 mM NaCl, pH 8.0 for 3 hours; 25 mM imidazole + 50 mM Tris-HCL, 300 mM NaCl, pH 8.0 overnight; and 50 mM Tris-HCL, 300 mM NaCl, pH 8 for an additional 3 hours. The concentration of the purified protein was measured using a bicinchoninic acid (BCA) kit (Thermo Scientific) and stored at −80°C.

To lipoylate purified 6x-His-E2-PDH 4×50 μl reactions were set up in 50 mM Tris-HCl, 300 mM NaCl, pH 8.0 supplemented with 6 mM ATP (Amresco), 1 mM DTT, 1 mM MgCl₂ (Amresco), 1 μM recombinant purified lipoate protein ligase, 20 μM E2-PDH, and 2.4 mM lipoic acid. The reactions were incubated for 2 hours at 37°C shaking at 600 rpm. After incubation, the reaction mixtures were loaded onto a Superdex 100 Increase 3.2/300 Gel Filtration column and fractionated by size exclusion chromatography using an AKTA FPLC system (GE Healthcare) to purify lipoyl-E2-PDH in a final buffer containing 50 mM Tris-HCl, 300 mM NaCl, pH 8.0. Protein purity and lipoylation were confirmed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie staining with GelCode Blue stain reagent (Thermo Scientific), or via immunoblot with anti-lipoic acid antibody as described in detail below.

Purification of S. aureus Lipoprotein SitC

Recombinant SitC was expressed and purified from S. aureus strain RN4220. To generate a plasmid capable of expressing SitC harboring a C-terminal 6x-His tag we first amplified the S. aureus sarA promoter (P_sarA) linked to the ribosome binding site of the S. aureus superoxide dismutase (sod_Rbs) using primer pair SitC1/SitC2 and the plasmid pOS1-P_sarA-sod_Rbs-sGFP as template. We then amplified the sitC gene from S. aureus LAC genomic DNA using primer pair SitC3/SitC4. SitC4 contains coding sequence for a 6x-His tag embedded in the primer. The resulting amplicons from these PCRs were used in a SOEing PCR reaction to generate a single amplicon harboring P_sarA-sod_RbsSitC-6xHis flanked by restriction endonuclease cut sites Pst1 and Sal1. The P_sarA-sod_RbsSitC-6xHis was digested, ligated into the pOS1 plasmid, and transformed into DH5α E. coli. The pOS1-P_sarA-sod_RbsSitC-6xHis plasmid was then purified from E. coli and electroporated into RN4220.

To purify SitC, S. aureus RN4220 containing the pOS1-P_sarA-sod_RbsSitC-6xHis plasmid was grown overnight at 37°C, 220 rpm in TSB supplemented with 10 μg of chloramphenicol (Amresco). The following day, the bacteria were subcultured 1:100 into 2 liters of fresh TSB medium (Criterion) supplemented with 10 μg of chloramphenicol (Amresco) and grown for 8 hours at 37°C. Bacteria were pelleted by centrifugation at 8500 rpm for 10 minutes at 4°C and then stored at −80°C overnight. Bacterial pellets w ere thawed on ice and resuspended in lysis buffer (20mM Tris-HCL (Alfa Aesar), 50 mM NaCl (Amresco), pH 8.0). The resuspended cells were treated with 200 μl lysostaphin (Ambi Products, NY) (2 mg/ml in 20 mM sodium acetate pH 4.5) and incubated at 37°C for 1 hour. The cells we re supplemented with 2 mM phenylmethane sulfonyl fluoride (PMSF) and lysed on ice via sonication (Branson) at 20-second intervals with a rate of 0.8 seconds per pulse and an output of 340W for a total of 30 minutes. The bacterial debris was pelleted for 1 hour at 11000 rpm and cell-free supernatants were collected followed by isolation of membranes by ultracentrifugation at 39000 rpm for 1 hour. Membrane pellets were solubilized in extraction buffer (20 mM Tris-HCl, 50 mM NaCl, 2% TritonX-100 (Amresco), pH 8.0) for 18 hours at 4°C. 1 mL of Nickel-NTA res in (Qiagen) was equilibrated in the same extraction buffer for 18 hours at 4°C. The following day, the solubilized membrane pellets were centrifuged at 11000 rpm for 1 hour to remove any insoluble debris, filtered through a 0.45 μM syringe filter, and incubated with Nickel-NTA resin for 1 hour at room temperature. The resin was washed four times with wash buffer ( 20 mM Tris-HCl, 50 mM NaCl, 0.25% TritonX-100, 40 mM imidazole (Amresco)) prior to elution in the same buffer containing 400 mM imidazole. The purified protein was dialyzed using snakeskin dialysis tubing (10 kDa MWCO, Thermo Scientific) into 100 mM imidazole, 20 mM Tris-HCl, 50 mM NaCl, 0.25% TritonX-100 pH 8.0 for 3 hours at 4°C; 25 mM imidazole, 20 mM Tris-HCl, 50 mM NaCl, 0.25% TritonX-100 pH 8.0 for 3 hours at 4°C; and 20 mM Tris-HCl, 50 mM NaCl 0.25% TritonX-1 00 pH 8.0 overnight at 4°C. The concentration of the purified SitC was measured using a bicinchoninic acid (BCA) kit (Thermo Scientific) and determined to be 2 mg/mL. The purified protein was stored at −80°C until use where it was diluted 20,000 fold before addition to cells so that the final TritonX-100 concentration was 0.00001% with no adverse affects on mammalian cells. Protein purity was confirmed by SDS-PAGE and Coomassie staining.

Exoprotein Isolation and Immunoblotting

To isolate S. aureus exoproteins, strains were first grown overnight in 5 mL TSB or RPMI+BCFA in conical tubes. The following morning the bacteria were then subcultured 1:100 into conical tubes containing 5 mL fresh TSB or RPMI+BCFA and allowed to grow for 9 hours at 37°C with shaking at 200 rpm. After 9 hours, the OD600 was measured using a Genesys 10S UV-Vis spectrophotometer (Thermo Scientific). The cultures were pelleted by centrifugation at 4200 rpm at 4°C for 15 minutes. After pelleting of the bacteria, 1.3 mL cell free supernatant was removed followed by the addition of trichloroacetic acid (TCA) (Alfa Aesar) to 10% final volume and subsequent incubation at 4°C overnight. The following day, samples were centrifuged at 13000 rpm at 4°C for 15 minutes and 1 mL of 100% EtOH was added followed by incubation at 4°C for 30 minutes. The precipitated proteins were centrifuged at 13000 rpm at 4°C for 15 minutes and EtOH was removed, followed by an additional centrifugation at 15000 rpm for 2 minutes at room temperature and removal of any excess ethanol. The pellets were then left to air dry for 1 hour, followed by resuspension in 30 μl of TCA-SDS buffer (2X SDS buffer + β-mercaptoethanol diluted 1:1 with 0.5 M Tris-HCl buffer (pH 8.0) + 4% SDS) and boiling for 10 minutes at 100°C. Samples were normalized to the highest OD600 and separated by SDS-PAGE in 10% or 12% polyacrylamide gels followed by Coomassie staining with GelCode Blue stain reagent (Thermo Scientific).

For immunoblotting, strains were first grown overnight in RPMI+BCFA. The following morning, overnights were subcultured 1:100 in 50 mL flasks containing 5 mL of RPMI+BCFA with reduced branched chain fatty acid precursors (1.123 mM isobutyric acid, 1.0 mM 2-methylbutyric acid, 1.0 mM isovaleric acid and 1.0 mM Na acetate) to reduce background on immunoblots. After 9 hours, the OD600 was measured and bacteria from 4 mL culture volume were pelleted at 4200 rpm at 4°C for 15 minutes. 1. 3 mL cell free supernatant was removed followed by the addition of 10% TCA and isolation of secreted proteins as described above. To detect lipoyl proteins in the supernatant, samples were loaded based on OD normalization to account for minor differences in final optical density, resolved on 10% or 8% SDS-PAGE gels, followed by transfer to 0.2 μm PVDF membrane (EMD Millipore) at 200V for 1.5 hours in a Quadra Mini-Vertical PAGE/Blotting System (CBS Scientific). Membranes were then incubated overnight in PBST (PBS + 0.1% Tween-20) with 5% Bovine Serum Albumin (BSA) (Amresco) or with 5% BSA in TBST (Tris-buffered saline + 1% Tween-20) at 4°C overnight. The following day, membranes were blocked with 0.9 mg/mL human IgG (Sigma) for 1 hour to mitigate nonspecific binding to S. aureus antibody binding proteins. Membranes were washed three times in PBST or TBST for 15 minutes each followed by addition of a 1:5,000 dilution of rabbit anti-lipoic acid antibody (EMD Millipore) in 10 mL of PBST + 5% BSA or TBST + 5% BSA. After 1 hour, membranes were washed three times with PBST or TBST for 15 minutes each, incubated with a 1:400 dilution of goat anti-rabbit IgG (H+L) HRP conjugate (Thermo Scientific) in 10mL of PBST + 5% BSA for 1 hour or 1:5000 goat anti-rabbit IgG (H+L) Alkaline Phosphatase (Thermo Scientific) in 10mL of TBST + 5% BSA secondary antibodies depending on the detection method, followed by an additional three 15 minute washes in PBST or TBST. Immunoblots were visualized using a FluorChemE System (Protein Simple) or exposed to film (Dot Scientific), and developed using an Alphatek Ax390 SE autoprocessor using Pierce ECL Western Blotting Substrate (Thermo Scientific) or after addition of 35 μl of a 50 mg/mL stock of 5-bromo-4-chloro-3-indoyl phosphate (Amresco) and 66 μl of a 50mg/mL stock of nitro blue tetrazolium (Amresco) to 10mL of 100 mM Tris pH 9.5 + 100 mM NaCl + 5 mM MgCl₂ (AP Buffer) to membranes followed by incubation for 1–2 minutes.

Quantification and Stability of Lipoyl-E2-PDH

To quantify the amount of lipoyl E2-PDH released by S. aureus, WT S. aureus was first grown overnight in RPMI+BCFA. The following morning, overnights were subcultured 1:100 in 50 mL flasks containing 5 mL of RPMI+BCFA with reduced branched chain fatty acid precursors. After 9 hours, the bacteria were pelleted at 4200 rpm at 4°C for 15 minutes. 1.3 mL cell free supernatant was removed followed by the addition of 10% TCA and secreted proteins were isolated as described above. 2.5 μL of WT TCA precipitated supernatant was loaded onto a 10% SDS-PAGE gel and resolved with a titration of purified recombinant lipoyl-E2-PDH. After performing an immunoblot to detect lipoyl-E2-PDH as described above, the amount of lipoyl E2-PDH was determined based on the intensity of the bands and found to be present in the range of approximately 25–50 nM.

To measure the stability of released lipoyl E2-PDH, WT S. aureus was first grown overnight in RPMI+BCFA. The following morning, overnights were subcultured 1:100 in 50 mL flasks containing 5 mL of RPMI+BCFA with reduced branched chain carboxylic acids. After 9 hours, the bacteria were pelleted at 4200 rpm at 4°C for 1 5 minutes. The cell free supernatant was collected and incubated for 0, 2, 4, 8 and 16 hours at 37°C. Following incubation, 1.3 mL of the cell free supernatant was removed followed by the addition of 10% TCA and secreted proteins were isolated as described above. 10 μL of the TCA precipitated supernatant was then resolved on a 10% SDS-PAGE gel and lipoyl-E2-PDH was detected via immunoblot as outlined above.

Immunoblotting for Cytosolic Lipoyl-Proteins

Whole cell lysates were collected and blotted for lipoyl-proteins as described before (Zorzoli et al., 2016). In brief, after growing strains overnight in RPMI+BCFA, cultures were diluted 1:100 in 5 mL RPMI+BCFA and grown for 9 hours at 37°C with s haking at 200 rpm. After 9 hours, OD600 was measured and the strains were pelleted at 4200 rpm for 15 minutes at 4°C. The spent culture medium was aspirated and the bacterial pellets were resuspended in 250 μl of sterile 1X PBS. The bacterial suspensions were transferred into screw-cap microcentrifuge lysing tubes (Fisher Scientific) containing 250 μl of 0.1 mm glass cell disruption beads (Scientific Industries, Inc.) and lysed using a Fast Prep-24 5G (MP Biomedicals) bead disruption system at setting 5.0 for 20 seconds followed by a 5 minute incubation on ice and additional disruption at setting 4.5 for 20 seconds. Tubes were centrifuged at 13000 rpm for 15 minutes at 4°C to pellet debris and 100 μl of clarified lysates were boiled at 100°C for 10 minutes in 6X SDS sample buffer (30% (v/v) glycerol (Amresco) + 0.5M Tris (Amresco) + 10% (w/v) SDS (Amresco) + 5% (v/v) βME + 0.012% (w/v) bromophenol blue (Amresco)) followed by storage at -20°C or immediate use in SDS-PAGE. Samples were loaded based on OD normalization to account for minor differences in final optical density and loaded onto 12% SDS-PAGE gels followed by either by Coomassie staining with GelCode Blue stain reagent (Thermo Scientific) or transfer to 0.2 μm PVDF membrane at 200V for 1 hour. Membranes were then blocked overnight with 5% BSA in PBST or with 5% BSA in TBST (Tris-buffered saline + 1% Tween-20) at 4°C. Immunoblotting was conducted as described above using 0.9 mg/mL of human IgG for blocking, 1:3000 rabbit anti-lipoic acid primary antibody, and either 1:200 goat anti-rabbit IgG (H+L) HRP conjugate or 1:5000 goat anti-rabbit IgG (H+L) Alkaline Phosphatase (Thermo Scientific) secondary antibodies depending on the detection method. Immunoblot images were then detected after addition of Pierce ECL Western Blotting Substrate, exposure to film (Dot Scientific), and developed using an Alphatek Ax390 SE autoprocessor, or after addition of 35 μl of a 50 mg/mL stock of 5-bromo-4-chloro-3-indoyl phosphate (Amresco) and 66 μl of a 50mg/mL stock of nitro blue tetrazolium (Amresco) to 10mL of 100 mM Tris pH 9.5 + 100 mM NaCl + 5 mM MgCl₂ (AP Buffer) to membranes followed by incubation for 1–2 minutes.

In vitro Macrophage Experiments

65000 macrophages were seeded into 96-well flat-bottom tissue culture treated plates (Corning) in 90 μl BMM medium supplemented with 100 μg/ml Pen/Strep and 50 μg/ml gentamicin (Amresco) the day before use. The following day, supernatants isolated from 9 hour cultures of S. aureus were thawed from storage at −80°C on ice and 10 μl was added to BMMs following OD normalization between strains to ensure equivalent addition of exoproteins followed by incubation at 37°C, 5% CO₂ for 24 hours. Macrophage supernatants were then removed and either used immediately or stored at −80°C. In experiments using either purified recombinant E2-PDH or synthetic DKA and DK^LA (AnaSpec), proteins were pretreated with 20 μg/ml of Polymyxin B sulfate (Alfa Aesar) for 1–2 hours at 37°C to mitigate aberrant LPS activation. Treated protein samples were then added to the macrophages in triplicate along with the TLR agonists Pam2CSK4 (1 ng/mL) or Pam3CKS4 (3 ng/mL) (Invivogen) where indicated. In the experiments with SitC, the recombinant lipoprotein was treated with polymyxin B sulfate as described and added to the macrophages at a concentration 0.1 ng/mL for measurement of CCL3 and CCL4 production and 1 ng/mL for measurement of IL-6 and TNF production. For activation of TLR2^−/− BMMs, 150 ng/mL of E. coli serotype 0111:B4 lipopolysaccharide (Enzo Life Sciences) was added to cells. All BMMs were activated for 24 hours at 37°C, 5% CO_2, followed by removal of macrophage and immediate use or storage at −80°C.

To measure the levels of cytokines and chemokines in the macrophage supernatants, a custom Cytometric Bead Array Flex set (BD Biosciences) was used according to manufacturer’s specifications. Bead and supernatant mixtures were incubated for 1.5 hours at 25°C at 600 rpm using a Thermo Mixer C (Eppendorf) in a 96-well V-bottom plate (Corning). Samples were then washed using FACS Wash Buffer (1X PBS + 2% heat inactivated FBS + 0.05% (w/v) Sodium Azide), data collected on an LSRFortessa (BD Biosciences), and analyzed using FlowJo software (FlowJo, LLC) by gating on individual bead populations and calculating the geometric mean of fluorescence relative to protein standards.

To measure bacterial survival upon infection of activated macrophages, 65000 BMM were first seeded into 96-well flat-bottom plates in BMM-medium without antibiotic. The following day cells were treated with 10% S. aureus cell free supernatant and incubated for 16 hours to induce macrophage activation. The day after activation, overnight cultures of WT and ΔlipA S. aureus were normalized to an OD600 of 0.32–0.33 (1×10⁸ CFU/mL) in PBS, added to macrophages at a multiplicity of infection of 1, and centrifuged for 7 minutes at 1500 rpm to synchronize the infection. Infections were carried out at 37°C, 5% CO₂ for 30 minutes, washed 3X with 1X PBS and incubated with gentamicin (50 μg/mL) at 37°C, 5% CO₂ for an additional 30 minutes. Cells were washed 3X with 1X PBS and placed in BMM medium without antibiotic. After an additional hour of incubation, saponin (0.1%) was added to the wells and incubated on ice for 30 minutes, followed by plating onto BCFA-containing tryptic soy agar plates to enumerate bacterial CFU.

NFκB Activation Assay

To measure the activation of the transcription factor NFκB, the RAW-Blue macrophage cell line harboring an NFκB inducible chromosomally integrated secreted embryonic alkaline phosphatase (SEAP) reporter construct (Invivogen) was grown to about 50–70% confluency in DMEM (Corning) + 10% heat inactivated FBS in tissue culture treated T75 flasks (Corning). Media was removed from the flask, and the cells were washed with 2mL of 0.05% trypsin + 0.53 mM EDTA (Corning). 2 mL of fresh 0.05% trypsin + 0.53 mM EDTA was added to the RAW cells for 5–10 minutes and cells were dissociated from the surface of the flasks by gentle agitation. Dissociated cells were then resuspended in DMEM + 10% heat inactivated FBS and centrifuged for 5 minutes at 1500 rpm. Cells were then counted and seeded into 96-well flat-bottom tissue culture treated plates (Corning) at 50,000 cells per well in 90 μl of DMEM + 10% heat inactivated FBS supplemented with 100 μg/ml Pen/Strep (Corning) and 50 μg/ml gentamicin (Amresco) the day before use. The following day, supernatants isolated from 9 hour cultures of S. aureus were thawed from storage at −80°C on ice and added to the RAW cells following OD normalization between strains to ensure equivalent addition of exoproteins followed by incubation at 37°C, 5% CO₂ for 24 hours. Treated RAW-Blue cell supernatants were collected and subsequently heated at 68°C for 30 minutes to deactivate endogenous phosphatases. In a 96-well polystyrene plate 50 μl of the deactivated RAW cell supernatants were mixed with an equal volume of 1-Step PNPP (Thermo Scientific). After 2–3 hours of incubation, color change was quantified by measuring the absorbance of the sample at 450 nm using an ELx800 microplate reader (BioTek).

S. aureus Chromosomal DNA Isolation

To verify that the atl::erm and secA2::erm transposon mutants carried the anticipated transposon within the gene of interest, primers were designed that directly flanked the annotated insertion site (see Key Resources Table). atl::erm and secA2::erm transposon mutant chromosomal DNA was isolated using the Wizard Genomic DNA purification kit (Promega) following the manufacturers protocol with minor modifications. A 1.2 mL overnight culture of atl::erm or secA2::erm was pelleted at 14000 rpm for 3 minutes at room temperature and resuspended in 200 μL of TSM buffer (50 mM Tris, 0.5 M D-Sucrose,10 mM MgCl₂ pH 7.5) followed by addition of 2.5 μl of lysostaphin (Ambi Products, NY) stock solution (2 mg/mL in 20 mM sodium acetate, pH 4.5) and incubation at 37°C f or 15 minutes to digest the cell wall. The bacteria were then pelleted at 14000 rpm for 5 minutes at room temperature and the chromosomal DNA was extracted as detailed by the manufacturer’s instructions. Purified DNA samples were used as template with the aforementioned primers to amplify the ~3.2 kb bursa aurealis transposon using a Flexid Mastercycler (Eppendorf) and Phusion High-Fidelity DNA Polymerase (New England Biolabs).

Murine Systemic Infection Models

Single colonies of freshly struck out bacteria were inoculated in TSB with shaking at 37°C and grown overnight. The overnight culture was subcultured 1:100 in 15mL fresh TSB and incubated at 200 rpm at 37°C for 3 hours. Cultures were then centrifuged for 5 minutes at 4200 rpm at 4°C, and the resulting cell pellets were washed 3 times in 1X sterile PBS. Bacterial suspensions were then normalized to an OD600 of 1.1–1.2 (~1 × 10⁹ CFU/mL) for intraperitoneal infection or to an OD600 of 0.32–0.33 (~1 × 10⁸ CFU/mL) for intravenous infection. Six to eight week old female Swiss Webster mice (Envigo) were then injected intraperitoneally with 100 μL of PBS containing 1 × 10⁸ CFU of S. aureus. For intravenous infection, six to eight week old female Swiss Webster mice were anesthetized with 2,2,2-tribromoethanol (Avertin) (250 mg/kg) (Sigma), via intraperitoneal injection followed by inoculation with 100 μL PBS containing 1 × 10⁷ CFU of S. aureus directly into the bloodstream via retro-orbital sinus. After infection, the remaining bacterial suspensions were plated onto TSA plates to ensure accuracy of infection inoculums (all mice received between 1.0 and 2.0 × 10⁸ CFU for IP infection and between 1.0 and 2.0 × 10⁷ CFU for IV infection). Mice were monitored daily and after 16 or 72 hours for IP infection and 96 hours for IV infection, serum was collected via the facial vein and mice were immediately euthanized by CO2 narcosis followed by aseptic isolation of kidneys and lavage samples. Kidneys were homogenized, and lavage fluid and kidney homogenates were serially diluted onto TSA plates followed by incubation at 37°C overnight to enumerate CFU.

Flow Cytometry of Immune Cells

Assessment of immune cell recruitment to the peritoneal cavity was determined by performing a lavage of the peritoneal cavity of euthanized mice 16 or 72 hours post-infection with ~7 mL of 1X sterile PBS in a 10 mL sterile syringe with an 18-gauge needle. Isolated cells were pelleted at 1500 rpm at 4°C for 5 minutes, supernatant was decanted, and red blood cells were lysed with ACK lysing buffer (Lonza) by resuspending the pellet in 2 mL of lysing buffer and incubating for 2 minutes at room temperature. Cell lysis was stopped by adding 8 mL of 1X sterile PBS, mixing gently, and pelleting at 1500 rpm at 4°C for 5 minutes. Cells were then suspended in 1X sterile PBS and kept on ice while they were counted. 1–2 million cells were transferred to a 96-well V-bottom plate (Corning) for cell surface staining. Prior to staining, immune cells were incubated with 50 μL of FACS Wash Buffer containing 0.2 μg/mL of anti-CD16/CD32 (93) (BioLegend) for 30 minutes on ice to block Fc receptors followed by washing with FACS Wash Buffer and surface staining with the following antibodies from BioLegend: anti-CD11b-Pacific Blue (M1/70), anti-Ly6G-PE (1A8), anti-DX5-APC-Cy7 (DX5), anti-CD206-FITC (C068C2), anti-CD11c-Alexa700 (N418), anti-I-A/I-E-Alexa647 (M5/114.15.2), anti-F4/80-PE-Cy7 (BM8), anti-CCR5-biotin (HM-CCR5) for 30 minutes on ice and washed twice with FACS Wash Buffer. To stain for CCR5, streptavidin conjugated PerCP-Cy5.5 was added on ice for 30 minutes. Samples were then washed twice, fixed using FACS Fixing Buffer (1X PBS + 2% heat inactivated FBS + 2% Paraformaldehyde + 0.05% (w/v) sodium azide). Data was collected on an LSRFortessa (BD Biosciences) and subsequently analyzed using FlowJo software (FlowJo, LLC).

To assess the immune cells recruited to the kidneys in intravenously infected mice after 96 hours, immune cell suspensions were purified using a 40/80 Percoll (GE Healthcare) density gradient centrifugation. Kidneys were isolated and broken down using glass mortar and pestle homogenizers (Kontes Glass Co) in 5 mL of RPMI (Corning). Kidney homogenate was then transferred to a 50 mL conical tube and centrifuged at 1500 rpm, 4°C for 5 minutes. Supernatant was decanted and red blood cells were lysed with ACK lysing buffer (Lonza) by resuspending the pellet in 2 mL of lysis buffer and incubating for 2 minutes at room temperature. Cell lysis was stopped by adding 8 mL of RPMI (Corning), mixing gently, and pelleting at 1500 rpm, 4°C for 5 minutes. Cells were resuspended in 10 mL RPMI (Corning) and passed through a 70 μm nylon mesh cell strainer (Corning). A 100% Percoll solution was made by mixing 9 parts Percoll with 1 part 10X PBS (Corning), and then diluted to 80% or 40% in RPMI + 10% heat inactivated FBS. Cells were then centrifuged at 1500 rpm, at 4°C for 5 minutes and resuspended in 5 mL of 40% RPMI Percoll solution and overlaid carefully on top of 5 mL of the 80% Percoll RPMI solution. Percoll gradients were rested at room temperature for 10 minutes and then centrifuged for 20 minutes 2500 rpm at room temperature with no brake. The top layer of parenchymal cells was aspirated off, and the immune cells present in at the 40%/80% Percoll interface were collected. The immune cells were then washed twice in RPMI + 10% heat inactivated FBS and stained for flow cytometry as described above.

S. aureus survival in F4/80⁺ Sorted Cells

Six to eight week old female Swiss Webster mice were infected via the peritoneal route as described previously with WT and ΔlipA S. aureus or sterile 1X PBS. 72 hours post infection mice were euthanized and peritoneal cells were isolated by lavaging the peritoneal cavity with ~7 mL of 1X sterile PBS in a 10 mL sterile syringe with an 18-gauge needle. Cells were then pelleted at 1500 rpm at 4°C for 5 minutes followed by decanting the supernatant and resuspending in complete RPMI cell culture medium (Corning) (RPMI + 10% heat inactivated FBS) supplemented with 100 μg/ml Pen/Strep and 50 μg/ml gentamicin and incubated on ice for 30 minutes to 1 hour. After incubation, cells were washed 3X in complete RPMI medium without antibiotics and incubated for 30 minutes in the antibiotic free medium. F4/80⁺ cells were sorted from total peritoneal cells using BD Imag Cell Separation (BD Biosciences) after incubation with anti-CD16/CD32 (93) (BioLegend) and biotinylated anti-F4/80 antibody (BM8) (BioLegend). In brief, streptavidin-conjugated magnetic beads were added to cells and allowed to bind to a magnet for 8 minutes followed by removal of the unbound fraction. Bound beads were then resuspended in complete RPMI medium and allowed to re-bind to the magnet for 6 minutes. This washing step was repeated once more to ensure purity of the bound fraction. Sorted cells were stored in RPMI complete medium overnight at 4°C. The following day an overnight culture of WT LAC grown in TSB was washed 3 times in sterile 1X PBS, normalized to an OD600 of ~0.32 (~1.0×10⁸ CFU/mL), and opsonized by incubation with 10% mouse serum for 30 minutes at 37°C followed by washing 3 times in sterile 1X P BS. The sorted F4/80⁺ cells were pelleted at 1500 rpm at 4°C for 5 minutes, resuspended in fresh RPMI complete medium and counted. The opsonized bacteria were then used to infect 1 million F4/80⁺ cells at a multiplicity of infection 1 for 30 minutes at 37°C in sterile 1.5 ml microcentrifuge tubes placed on a rotisserie. Following the 30 minute infection, samples were centrifuged at 1500 rpm at room temperature in a benchtop centrifuge (Eppendorf) for 5 minutes and washed 3 times in sterile 1X PBS. Samples were then resuspended in 1 mL complete RPMI medium and placed at 37°C on a rotisserie. S. aureus CFU were enumerated hourly for 6 hours by removing 100 μl aliquots, lysing with 0.1% saponin (Sigma) for 20 minutes on ice, followed by serial diluting and plating the dilutions on TSA plates supplemented with 50 μg/ml of Kan and Neo.

Quantification and Statistical Analysis

Experimental data analyzed for significance were from at least three replicates, repeated at least three independent times. Statistical analysis was completed using Prism software (GraphPad) and the specific tests used are indicated in the figure legends. Statistical significance was defined as p<0.05. N represents number of experimental replicates or animals in treatment groups, and it is specified in the figure legends. For any data without the use of animal models, we assumed a Gaussian distribution and used 1-way ANOVA with Bonferonni-Sidak post-test for pair-wise comparison. Data are reported as means with standard deviation unless otherwise mentioned. For any data with the use of animal models, we assumed a non-Gaussian disruption due to the possibilities of outliers. As such, we used non-parametric 1-way ANOVA (Kruskal-Wallis Test) with Dunn’s post-test. For S. aureus ex vivo survival/growth assays with F4/80⁺ sorted cells a 2-way ANOVA with Tukey’s multiple comparison test was used.

Highlights.

• de novo synthesis of lipoic acid by S. aureus restricts macrophage activation

• S. aureus releases lipoyl-E2-PDH into the extracellular environment

• Lipoyl-E2-PDH blunts TLR1/2 macrophage activation by triacylated lipopeptides

• Lipoic acid synthesis augments immune suppression in vivo to promote virulence