Multiple Domains of Staphylococcal Superantigen-like Protein 11 (SSL11) Contribute to Neutrophil Inhibition

Abstract

Staphylococcus aureus is an opportunistic pathogen producing many immune evasion molecules targeting various components of the host immune defense, including the Staphylococcal superantigen-like protein (SSL 1–14) family. Despite sharing similar structures with the powerful superantigens (SAgs), which cause massive T cell activation, SSLs interfere with a wide range of innate immune defenses. SSLs are divided into two subgroups, SSLs that contain a conserved carbohydrate Sialyl Lewis X [Neu5Acα2–3Galβ1–4(Fucα1–3) GlcNAcβ, SLeX] binding site and SSLs that lack the SLeX binding site. SSL2–6 and SSL11 possess the SLeX binding site. Our previous studies showed that SSL11 arrests cell motility by inducing cell adhesion in differentiated HL60 (dHL60) cells, while SSL7 did not bind dHL60 cells. SSL7-based chimeras were engineered by exchanging the SSL7 sequence with the corresponding SSL11 sequence and assaying for a gain of SSL11 function, namely, the induction of cell spreading and motility arrest. In addition to the SLeX-binding site, we observed that three beta-strands β6, β7, and β9 and the N-terminal residues, Y16 and Y17, transitioned SSL7 to gain SSL11 activities. These studies define the structure–function properties of SSL11 that may allow SSL11 to inhibit S. aureus clearance by the host innate immune system, allowing S. aureus to maintain a carrier state in humans, an understudied aspect of S. aureus pathogenesis.

Graphical Abstract

INTRODUCTION

Staphylococcus aureus is a major opportunistic human pathogen that causes a wide variety of community- and hospital-acquired infections. The prevalence of methicillin-resistant S. aureus (MRSA) imposes a significant burden on healthcare resources.^1,2 In addition to having increased antibiotic resistance, S. aureus is a master at adapting to its host by evading the immune system at almost every level imaginable. Host innate immunity, including neutrophil killing and complement activation, is the primary host defense mechanism against S. aureus. Since host innate immunity attacks pathogens via a plethora of strategies, the evasion molecules from S. aureus display redundant functions as well.^3,4

An excellent example of redundancy of S. aureus immune evasion is the Staphylococcal superantigen-like protein (SSL) family. SSLs were identified by bioinformatic screening of the staphylococcal genome using the two conserved motifs of Staphylococcal superantigens (SAgs).⁵ The crystal structures of the SAgs/SSLs consist of two domains: an N-terminal oligosaccharide/oligonucleotide binding (OB) domain with a mixed β-barrel with Greek-key topology and a C-terminal β-grasp domain with a five-stranded, twisted, anti-parallel β-sheet capped by a central α-helix.⁶ Despite structural similarities, SSLs are not mitogenic to T cells and do not bind the MHC class II molecule due to the lack of the critical residues for SAg activities.^5,6 SSLs are divided into two subgroups depending on the presence of a Sialyl Lewis X [Neu5Acα2—3Galβ1–4(Fucα1–3) GlcNAcβ, SLeX] binding site within the C-terminal domain (Figure 1). Not all SSL functions are known, but most SSL activities identified so far involve immune evasion: SSL3 inhibits TLR2 signaling, SSL5 and SSL11 inhibit neutrophil activation and rolling, and SSL7 and SSL10 bind IgA and IgG and inhibits complement activation.^7–19 In contrast, SSL13 activates neutrophils via formyl peptide receptor 2.²⁰ Previously, we showed that SSL11 arrested neutrophil-like cell motility by inducing cell adhesion in aglycan-dependent manner, while SSL7, lacking the SLeX-binding site, did not affect neutrophil motility.¹⁸

Figure 1.

Two subfamilies of SSL proteins. (A) Crystal structures of SSL7 (PDB: 2QEJ) and SSL11 (PDB: 2RDG). SleX is shown as yellow sticks. (B) SSL1–14 sequences were aligned using Clustal Omega. The N-terminal conserved Y16 and Y17 residues are highlighted in yellow. The SLeX binding site is highlighted in yellow. Residues that are essential for SSL11 SLeX binding are highlighted in magenta.

The evasion of host neutrophil recruitment to the site of infection is essential to the success of S. aureus as a pathogen.³ Precise regulation of neutrophil adhesion and de-adhesion is critical for migration toward an inflammation site.²¹ Differentiated HL60 cells (dHL60) are a widely used model of human neutrophils for migration and chemotaxis.²² In the current study, we investigated the functional domains of SSL11 activities for their ability to modulate the neutrophil function. Using targeted mutagenesis and a gain of function approach by converting SSL7 to a functional SSL11 phenotype, we determined that the C-terminal β-grasp domain of SSL11, although containing an intact SLeX-binding site was not sufficient to induce cell adhesion and spreading as the full length SSL11. Here, we show that in addition to the conserved SLeX-binding site, three beta-strands β6, β7, and β9 and the N-terminal residues, Y16 and Y17, are necessary for the expression of SSL11 activities, explaining how SSL11 inhibits neutrophil motility and chemotaxis.

RESULTS

The C-Terminal β-Grasp Domain of SSL11 Is Not Sufficient for SSL11 Activities.

SSL11 belongs to the subfamily of SSLs with a conserved carbohydrate SLeX-binding site in the C-terminal β-grasp domain. Our previous study showed that SSL11 induced cell spreading and adhesion via this SLeX-binding site, as the pre-incubation of SSL11 with SLeX blocked the SSL11 function and de-glycosylation of dHL60 cell lysate by PNGase F abolished SSL11 binding.¹⁸ We hypothesized that the C-terminal β-grasp domain (SSL11-C, SSL11 residues 91–195) was sufficient for SSL11 activities. To test this hypothesis, SSL11-C alone was expressed and incubated with dHL60 cells (Figure 2A,B). As suspension cells, quiescent dHL60 cells display low adherence. After 30 min incubation with SSL11, dHL60 cells transitioned from a non-adhesion to an adhesion phenotype, while untreated cells remained as a non-adhesion one. Cell spreading was visualized by Phalloidin staining (Figure 2C). Cell spreading was quantified by the average cell diameter (Figure 2D). Cell adhesion was quantified by crystal violet staining (Figure 2E). We found that SSL11-C did not bind to cells and failed to induce cell spreading and adhesion as SSL11 (Figure 2C–E). Therefore, the C-terminal β-grasp domain is not sufficient for SSL11 activities.

Figure 2.

The C-terminal β-grasp domain of SSL11 is not sufficient for SSL11 activities. (A) Crystal structure of SSL11 in complex with SLeX (PDB: 2RDG). The N-terminal OB domain is shown in magenta, and the C-terminal β-grasp domain (SSL11-C, SSL91–195) is shown in blue. SLeX is shown as yellow sticks. (B) A total of 1 μg of purified SSL11 and SSL11-C were separated by SDS-PAGE and stained with Coomassie Blue. (C) dHL60 cells were incubated with 80 nM of Alexa⁵⁶⁸ labeled SSL11 or SSL11-C at 37 °C for 30 min. Cells were washed, fixed, and incubated with Phalloidin⁶⁴⁷ at room temperature for 1 h, followed by DAPI staining. Cell staining is shown: SSL (green), Phalloidin (magenta). (D) The diameters (μm) of random cells were measured by ImageJ and plotted by GraphPad Prism (n = 210). (E) dHL60 cells were incubated with 80 nM of SSL11 or SSL11-C in FN-coated 96-well plates at 37 °C for 30 min followed by two PBS washes. Adherent cells were quantified by crystal violet staining and shown in adhesion arbitrary units (AU) (n = 6). (D) and (E) are from three independent experiments.

The N-Terminals Y16 and Y17 of SSL11 Contribute to SSL11 Activities.

SSL11-C failed to display SSL11 functions of inducing cell spreading and adhesion (Figure 2), suggesting that the N-terminal OB domain is required as well, in addition to the conserved SLeX-binding site in the C-terminal β-grasp domain. When SSL1–14 sequences were aligned using Clustal Omega,^23,24 Y16 and Y17 are conserved among all the SSL proteins, suggesting that Y16 and Y17 are essential for the SSL family (Figure 1B). When we examined the crystal structure of SSL11 (PDB: 2RDG), Y16 and Y17 were located on the N-terminal α-helix and in close proximity to K174, which resides in the SLeX-binding site (Figure 3A). To test whether Y16 and Y17 contribute to SSL11 activities, SSL11^Y16AY17A was constructed and expressed in E. coli. R179 binds directly to SLeX via a hydrogen bond and interacts with the carboxyl group of sialic acid via electrostatic recognition.¹⁹ Mutation of R179 in SSL11 and the corresponding R residues in SSL4 and SSL5 results in complete loss of carbohydrate binding,^6,25,26 and SSL11^R179A was constructed as an SLeX-binding null control. Without a functional SLeX-binding site, SSL11^R179A did not induce cell spreading and cell adhesion when incubated with dHL60 cells. However, with an intact SLeX-binding site, SSL11^Y16AY17A failed to stimulate cell spreading and adhesion, suggesting that the N-terminal OB domain contributes to SSL11 activities (Figure 3C,D). For the SSL subgroups with a carbohydrate-binding site, such as SSL4, SSL5, and SSL11, protein activities are dependent on the SLeX-binding site. Our study on SSL11 showed, for the first time, that the N-terminal OB domain contributes to protein activities for this subgroup of SSLs.

Figure 3.

The N-terminals Y16 and Y17 of SSL11 contribute to SSL11 activities. (A) Crystal structure of SSL11 in complex with SLeX (PDB: 2RDG). The N-terminal OB domain is shown in magenta, and the C-terminal β-grasp domain (SSL11-C, SSL91–195) is shown in blue. The SLeX-binding site is shown in green. SLeX is shown as yellow sticks. Y16A and Y17 are shown as orange sticks, and R179 is shown as green sticks. (B) A total of 1 μg of purified SSL11, SSL11^Y16AY17A, and SSL11^R179A were separated by SDS-PAGE and stained with Coomassie Blue. (C) dHL60 cells were incubated with 80 nM of SSL11, SSL11^Y16AY17A, and SSL11^R179A at 37 °C for 30 min. Cells were washed, fixed, and incubated with Phalloidin⁶⁴⁷ at room temperature for 1 h, followed by DAPI staining. The diameters (μm) of random cells were measured by ImageJ and plotted by GraphPad Prism (n = 210). (D) dHL60 cells were incubated with 80 nM of SSL11, SSL11^Y16AY17A, and SSL11^R179A in FN-coated 96-well plates at 37 °C for 30 min followed by two PBS washes. Adherent cells were quantified by crystal violet staining and shown in adhesion arbitrary units (AU) (n = 6). (C) and (D) are from three independent experiments.

An SSL7-Based SSL7/11 Chimera Induced Cell Spreading and Adhesion.

SSL7 binds to both IgA Fc and complement C5, inhibiting complement activation and serum bactericidal activity.^9,14 SSL7, unable to bind to cells because it lacks the SLeX-binding site, does not induce cell adhesion in neutrophil-like cells.¹⁸ As SSL7 has both Y16 and Y17 that are required for SSL11 activities, we asked if SSL7 could be engineered to gain SSL11 activities by acquiring the conserved SLeX-binding site (Figure 3B). The SLeX binding site from SSL11 (residues 166–182, FYTFELNKKLQTHRMGD) was engineered to replace the corresponding sequence of SSL7 to construct SSL7/11 Chimera 1 (Figure 4A,B). Although containing both the conserved N-terminal Y16 and Y17 residues and the C-terminal SLeX-binding site, SSL7/11 Chimera 1 showed no detectable cell association and no SSL11-like activities of inducing cell spreading and adhesion (Figure 4C,D). Another chimera construct SSL7/11 Chimera 2, which contains the SLeX-binding site and the adjacent β9 strand, was engineered and expressed (Figure 4A,B). However, SSL7/11 Chimera 2 did not gain SSL11 activities, suggesting that the SLeX-binding site is not sufficient for SSL11 activities.

Figure 4.

SSL7/11 Chimera 3 induces cell spreading and adhesion. (A) Crystal structures of SSL7 (PDB: 2QEJ) and SSL11 (PDB: 2RDG). The OB domain and the β-grasp domain of SSL11 are shown in magenta and blue, respectively, with the SLeX-binding site highlighted in green. The OB domain and the β-grasp domain of SSL7 are shown in wheat and cyan, respectively. β6, β7, and β9 strands of SSL11 are shown in orange when SSL7 and SSL11 crystal structures are superimposed using PyMOL. SSL7/11 Chimera 1–3 constructs are illustrated as diagrams. (B) A total of 1 μg of purified SSL7, SSL11, and SSL7/11 Chimera 1–3 were separated by SDS-PAGE and stained with Coomassie Blue. (C) dHL60 cells were incubated with 80 nM of Alex⁵⁶⁸ labeled SSL7, SSL11, and SSL7/11 Chimera 1–3 at 37 °C for 30 min. Cells were washed, fixed, and incubated with Phalloidin⁶⁴⁷ at room temperature for 1 h followed by DAPI staining. Cell staining is shown: SSL (green), Phalloidin (magenta). (D) The diameters (μm) of random cells were measured by ImageJ and plotted by GraphPad Prism (n = 210). (E) dHL60 cells were incubated with 80 nM SSL7, SSL11, and SSL7/11 Chimera 1–3 in FN-coated 96-well plates at 37 °C for 30 min followed by two PBS washes. Adherent cells were quantified by crystal violet staining and shown in adhesion arbitrary units (AU) (n = 6). (D) and (E) are from three independent experiments.

When SSL7 (PDB: 2QEJ) and SSL11 (PDB: 2RDG) crystal structures were superimposed, we found that two anti-parallel β-strands, β6 and β7, are longer in SSL11 than SSL7 (Figure 4A). Other SSLs with the conserved carbohydrate-binding site, such as SSL5 and SSL4, have extended β6 and β7 strands as well (Figure S1). Therefore, SSL7/11 Chimera 3, containing the SLeX-binding site, β9, β6, and β7 from SSL11, was constructed (Figure 4 A,B). When incubated with dHL60 cells, SSL7/11 Chimera 3 bound to cells and induced cell spreading and cell adhesion (Figure 4C,D). SSL7/11 Chimera 4, containing the SLeX-binding site and the β6 and β7 strands from SSL11, did not induce cell spreading (Figure S2), suggesting that in addition to the SLeX-binding site, β9, β6, and β7 strands contribute to SSL11 functions.

SSL7/11 Chimera 3 Inhibits fMLP-Mediated dHL60 Cell Motility.

Neutrophil migration requires a well-regulated balance between adhesion and de-adhesion, and interruption of this balance affects neutrophil motility.²¹ SSL7/11 Chimera 3 gained SSL11 activities of inducing cell spreading and adhesion (Figure 4). Next, chemotactic peptide fMLP-induced cell motility was tested. fMLP was added to the edge of the fibronectin (FN)-coated well, and cell motility was recorded for 30 min. SSL7/11 Chimera 3 inhibited dHL60 cell motility as well as SSL11, while SSL7 did not affect dHL60 cell motility (Figure 5A, Movies S1–S4). Upon the addition of fMLP, dHL60 cells migrated an average length of 214 (95% CI 190–238) μm and dHL60 cells pre-treated with SSL7 migrated an average length of 203 (95% CI 184–222) μm. In comparison, cells pre-incubated with SSL11 or SSL7/11 Chimera 3 migrated an average range of 71 (95% CI 60–81) μm or 92 (95% CI 79–106) μm, respectively (Figure 5B). The recorded movies revealed that SSL7/11 Chimera 3 pre-incubated dHL60 cells that responded to fMLP stimulation did not migrate due to increased cell adhesion (Movies S1–S4). Thus, SSL7/11 Chimera 3 did not inhibit cell motility by directly interfering with fMLP sensing but rather inhibited cell motility by increasing tail-localized cell adhesion, like SSL11. In conclusion, SSL7/11 Chimera 3 gained SSL11 activities of cell motility arrest.

Figure 5.

SSL7/11 Chimera 3 blocks fMLP-mediated dHL60 cell motility. (A) dHL60 cells were incubated with 80 nM of SSL11, SSL7, or SSL7/11 Chimera 3 at 37 °C for 30 min in FN-coated plates when fMLP was added, and cell motility was recorded for 30 min. Representative motility traces of 20 cells are shown as individual tracks using Ibidi Chemotaxis and Migration Tool. (B) The total length of motility of the 20 cells (μm) shown in (A) was plotted by GraphPad Prism.

DISCUSSION

S. aureus is a major opportunistic human pathogen and causes an array of diseases, from superficial complications to lethal invasive infections. The prevalence of MRSA imposes a high burden on healthcare resources. S. aureus possesses an arsenal of virulence factors: 22 secreted enzymes such as lysins and proteases, 40 known or putative toxins, including SAgs and SSLs, 20 potential adhesions, and 57 potential virulence factors, such as clumping factor, iron regulators, and lipoproteins.^3,5,27

SAgs are responsible for the often-fatal acute condition known as toxic shock syndrome. SAg binds MHC class II and T-cell receptors by an antigen-independent mechanism to stimulate massive T-cell activation, leading to a “cytokine storm”.^5,28,29 However, after many years of intensive research, why S. aureus produces SAgs is still unclear. In contrast to SAgs’ “immune activating” role, SSLs play an opponent role in “immune inhibition”. SSL5 and SSL11 bind P-selectin glycoprotein ligand-1 (PSGL-1) to inhibit neutrophil rolling by interfering with the binding of neutrophils to P-selectin.^19,30 SSL11 inhibits neutrophil migration by inducing cell adhesion (13). SSL3 binds to TLR2 and blocks immune cell recognition of Staphylococcal lipoproteins and peptidoglycan via TLR1-TLR2 and TLR2-TLR6 dimerization,^11,12,16 and SSL10 blocks CXCR4-mediated neutrophil chemoattraction.³¹ SSL6 binds to CD47 (integrin-associated protein), a cell surface receptor promoting migration, anti-phagocytosis, and proliferation.³² SSL7 and SSL10 bind IgA and IgG and inhibit complement activation.^7,8,14,15 S. aureus, as an opportunistic pathogen, does not benefit from causing invasive diseases to kill the host with toxins such as SAgs, Hlgs, and Hla but rather benefits from spreading from one individual to another individual to maintain a balance with the host population. SSL modulation of host innate immunity contributes to maintaining a balance between S. aureus and humans. This is consistent with the fact that ssl genes are upregulated by Rot (the transcription factor repressor of toxins) and downregulated by Agr (the accessory gene regulator), in striking contrast to virulence toxins (Ags, Hlgs, and Hla) and Δssl1–11 in the Δagr mutant strain that only showed a log reduction in bacterial burden in the kidney and no change in the heart in a mouse infection model,³³ suggesting that SSLs are not key players for infections. This model is supported by the discovery of the bifunctional staphylococcal enterotoxin-like X (SElX). SEIX, an SSL-like SAg, binds neutrophils to disrupt IgG-mediated phagocytosis in addition to Vβ-specific T-cell activation.^34,35 Recently, SSL13 was reported to activate neutrophils via formyl peptide receptor 2.²⁰

S. aureus pathogenesis and colonization are not the results of any single SSL as each SSL displays overlapping and perhaps synergistic effects. SSLs can be divided into two subgroups depending on the presence of the SLeX carbohydrate-binding site (Figure 1). All the SSLs with the SLeX-binding site, SSL 2–6 and SSL11, display functions of binding leukocytes and inhibiting leukocytes rolling or migration.^{11,18,19,25,26,32} Therefore, the SLeX-binding site plays an essential role in the SSL activities in this S. aureus subgroup. Previously, our studies showed that SSL11 activities are glycan dependent, as the exogenous SLeX inhibited SSL11 cell association and SSL11-induced cell spreading and adhesion in a dose-dependent manner.¹⁸

In the current study, we show, for the first time, that the N-terminal OB domain is also required for SSL11 activities. We found that SSL11-C failed to bind and induce cell adhesion despite containing the SLeX-binding (Figure 2) site and the N-terminal Y16 and Y17 were required for the SSL11 activities (Figure 3). The observation that the two tyrosines are conserved among all SSLs (Figure 1) suggests that Y16 and Y17 play essential roles for SSL functions. Although K174 locates in the SLeX binding site by sequence (aa 166–182), K174 points away from the SLeX-binding site. K174 and the N-terminal residues D12, L13, S14, E15, Y16, and Y17 form another potential binding site, named the YY pocket here (Figure 6, orange colored). Thus, SSL11 is likely to contain two binding sites: one is the conserved SLeX-binding site (green colored), and the other one is the YY pocket. Dual sites provide binding specificity and enhanced activities for bacterial toxins. The SAg TSST-1 has two binding sites: one in the OB domain for the MHC class II α chain and the other in the β-grasp domain for T-cell receptors.^5,36,37 SSL7 also has dual binding sites: one in the OB domain for IgA and the other in the β-grasp domain for complement C5.^5,9,38 For some bacterial toxins with dual receptors, the carbohydrate binding sites play essential roles for their activities. Tetanus toxin binds two molecules of gangliosides, while botulinum neurotoxins bind either one ganglioside and one protein or two gangliosides. For tetanus toxin, the receptor binding domain has a ganglioside site binding the sugar backbone Gal-GalNAc and a sialic acid site binding sialic acids. Both binding sites are required for tetanus toxin’s high affinity-binding to neurons.^39–41 The alternative explanation is that Y16 and Y17 are required to maintain the structure of the SLeX-binding site; this is less likely as Chemotaxis Inhibitory Protein of Staphylococcus aureus (CHIPS), sharing a structural similarity to the SSL11 C-terminal β-grasp domain (Figure S3) and lacking the SSL N-terminal OB domain, inhibits neutrophil and monocyte chemotaxis toward C5a and formylated peptides like fMLP.^42,43 Recombinant SSL11-C is as stable as SSL11, supporting that the N-terminal OB domain is not required for C-terminal β-grasp folding. Whether the YY pocket of SSL11 binds to carbohydrates or proteins requires further investigation.

Figure 6.

SSL11 contains a potential second binding site. (A) Crystal structure of SSL11 (PDB: 2RDG). The SLeX-binding site is shown in green, and the potential second binding site, YY pocket, is shown in orange. SleX is shown as yellow sticks. (B) The key residues for the YY pockets, Y16, Y17, and K174, are shown in orange sticks. K174 is pointing away from the SleX-binding site and interacting with Y16 and Y17.

In the current study, we found that in addition to the SLeX-binding site, β9, β6, and β7 strands contribute to SSL11 functions (Figures 4 and 5 and Figure S2). SSL4, SSL5, and SSL11 share longer β6 and β7 strands than SSL7 (Figure S1). Biosensor analysis of SSL11 on SLeX-BSA shows that SSL11 has a fast-and a slow-dissociating species. The slower-dissociating species may reflect the formation of the homodimer at the local surface to increase SSL11 avidity to surface glycoproteins and longer β6 and β7 strands involved in dimerization. It is less likely that the slower-dissociating species are from oligomers or aggregates in solution as they are from a monomer protein peak of a size exclusion chromatography.¹⁹ SSL11 may induce cell adhesion via this slow-dissociating species formed at the cell surface, and β9, β6, and β7 strands contribute to SSL11 dimerization, which enhances SLeX binding. This is consistent with previous studies that two naturally occurring alleles of SSL11, US6610, and GL10, display a 10-fold difference in affinity toward SLeX despite the identical SLeX-binding site. SSL11-US6610 and SSL11-GL10 have sequence differences in β6 strands and the linker between β6 and β7 strands, which may explain their affinity difference to SLeX.¹⁹ Previously, we showed that SSL11-US6610 medicated cell adhesion although it only shares 69% sequence identity with SSL11-USA300, suggesting that induction of cell adhesion is an important function of SSL11 alleles.¹⁸ The current study provides insight into the observed complexity and redundancy of the SSL family members and understanding their contributions to S. aureus pathogenesis.

Why does S. aureus acquire so many virulence factors while other bacterial pathogens manage with only a small number of virulence factors? This question might be explained by the fact that S. aureus maintains life-long associations with its human hosts, by colonization and infections. Approximately 30% of the human population is colonized with S. aureus.^44,45 S. aureus colonization, mainly persistent colonization, is strongly associated with higher risks for clinical infections, especially for patients undergoing surgery, dialysis, and in the intensive care unit.^46–50 However, most colonized individuals will not experience S. aureus infections, suggesting that the SSLs allow S. aureus to balance carrier state colonization between infection and immunity, a potential role for the SSLs in S. aureus pathogenesis.

Most SSLs display species specificity for humans, which might underestimate their contributions toward S. aureus pathogenesis in animal experiments.^3,45 Microarray analysis during human neutrophil phagocytosis showed that genes known for leukocyte killing such as hlgA, hlgB, and hlgC are upregulated immediately in MRSA strains while ssl genes show delayed upregulation.⁵¹ There seems to be a paradox for S. aureus pathogenesis: Why does MRSA up-regulate immune evasion genes such as ssls after neutrophil killing genes are massively expressed during S. aureus-PMN interaction? Our hypothesis is that S. aureus escapes phagocytosis by killing neutrophils with toxins such as Hlgs, PVL, PSMS, and Hla. Escaped S. aureus will attract more host immune responses. SSLs will prevent more neutrophil migration toward the infection site or prevent complement activation and phagocytosis, which is beneficial for S. aureus infection and colonization. Understanding the molecular mechanisms of action of the SSL family of toxins might shed light on understanding S. aureus pathogenesis and the relationships among immunity, colonization, carrier state, and infection that may lead to new strategies for vaccines and therapies against S. aureus.

MATERIALS AND METHODS

Reagents.

Reagents were human fibronectin (ThermoFisher), DMSO (Sigma), Phalloidin Alex647 (ThermoFisher), N-formyl-Met-Leu-Phe (fMLP, sigma), SLeX (Sigma), biotinylated ECL (Vector Laboratories), ProLong Gold Antifade Mountant (ThermoFisher), and Q5 Site-Directed Mutagenesis Kit (New England BioLabs).

Plasmid Construction.

E. coli codon-optimized sequences of S. aureus strain USA300_FPR3757 SSL7 (NCBI reference sequence: WP_000769836.1) and SSL11 (NCBI reference sequence: WP_000769163.1) were synthesized (IDT) and subcloned into a pET28a vector for expression.

Site-directed mutagenesis and chimera constructs were engineered by replacing the sequences of SSL7 with the corresponding sequences of SSL11. All constructs contain an N-terminal His₆ tag for protein purification. DNA composition of these constructs was confirmed by DNA sequencing.

SSL11-C (comprising residues 91–195 of SSL11, primers F: GATCGATCGGATCCTCTAACTACATCGACAAAGTTAAAG, R: ATCGATCGCGGCCGCTTACAGGTTAACTTCGATTTTTTCG with wt SSL11 as a template), SSL7/11 Chimera 1 (SSL7 containing residues 165–178 of SSL11, primers F: aacaaaaaactgcagacccacCGTATGGGTGACGTTCTG, R: cagttcgaaggtgtagaaaccACCGTCTTTCAGGTTGATG with wt SSL7 as a template), SSL7/11 Chimera 2 (SSL7 containing residues 151–163 and 165–178 of SSL11, primers F: aatccgtatcaccatgAAAGACGGTGGTTTCTAC, R: ttagagtctttcggttcTTTGTACAGACCGTAGTTTTTAAC with Chimera 1 as a template), SSL7/11 Chimera 3 (SSL7 containing residues 90–121, 151–163, and 165–178 of SSL11, primers F: aaacatcgactctgttacctcacctctacctcttctacctacaccATCAACAAAGAAGAAGTTTCTC, R: ttggtgatgatcaggttaacgtctttaactttgtcgatgtagttagaGTTGTTTTTTTTGGTAACAC with Chimera 2 as a template), SSL7/11 Chimera 4 (SSL7 containing residues 90–121 and 165–178 of SSL11, primers F: aaacatcgactctgttacctctacctctacctcttctacctacaccATCAACAAAGAAGAAGTTTCTC, R: ttggtgatgatcaggttaacgtctttaactttgtcgatgtagttagaGTTGTTTTTTTTGGTAACAC with Chimera 1 as a template), and SSL11^Y16AY17A (Primers F: CCTGTCTGAAgcggcgAACCGTCCGTTCTTCGAATACACCAACC, R: TCCTGGGTCGCCTGAGAA) were constructed using the Q5 Site-Directed Mutagenesis Kit. SSL11^R179A (primers F: GCAGACCCACgcgATGGGTGACG, R: AGTTTTTTGTTCAGTTCGAAG).

Protein Expression and Purification.

Plasmids encoding SSL7, SSL11, SSL11^Y16AY17A, SSL11^R179A, and SSL7/11 Chimera constructs were transformed into E. coli BL21(DE3). Transformants were grown overnight on LB agar plates containing 50 μg of kanamycin/mL, which were the inoculums for liquid cultures (LB, 400 mL) containing the same antibiotic. Cells were cultured at 37 °C to an optical density at 600 nm of 0.6 when T7 promoter expression was induced with 1 mM IPTG. Cells were cultured overnight at 250 rpm at 16 °C. Cells were pelleted and lysed with a French press and clarified by centrifugation. His₆-tagged proteins were purified using Ni²⁺-nitrilotriacetic acid (NTA) resin (Qiagen). Purified proteins were dialyzed into 20 mM Tris buffer (pH 7.9) with 200 mM NaCl and 40% glycerol. Aliquots were stored at −20 °C.

Protein Labeling by Alexa Fluor.

SSL7, SSL11, SSL11-C, and SSL7/11 chimera constructs were labeled using the Alex Fluo568 Protein Labeling Kit (Invitrogen, A10238). Purified proteins were dialyzed in PBS at 4 °C overnight. Fifty microliters of 1 M bicarbonate was added to 0.5 mL of 2 mg/ mL SSL7 or SSL11, followed by transfer of the protein solution to a vial of reactive dye. The reaction mixture was stirred for 2 h at room temperature followed by dialysis in PBS at 4 °C overnight to eliminate the free dye. Labeled proteins were stored at 4 °C.

Cell Culture.

The HL60 cell (ATCC CCL-240) is a human promyelocytic cell line that can be differentiated into neutrophil-like cells.²² HL60 cells were cultured and maintained in an RMPI medium with 10% FBS and 20 mM HEPES. HL60 cells were differentiated in a complete medium with 1.3% DMSO as described by Fleck et al.²² Differentiated HL60 (dHL60) cells show ≥90% viability (trypan blue), ≥55% CD35 expression, and ≤ 20% CD71 expression. Differentiation was confirmed by flow cytometry using anti-CD11b, anti-CD35, and anti-CD71 antibodies (data not shown). dHL60 cells were used for experiments on days 6–8 after differentiation.

SSL11 Association with dHL60 Cells.

dHL60 cells (5 × 10⁵ cells/well) were incubated with Alexa568 labeled proteins (80 nM) at 37 °C for 30 min in a 24-well plate with glass coverslips. Cells were washed twice with PBS and fixed with 4% paraformaldehyde at room temperature for 15 min. Cells were incubated in blocking solution (DPBS with 10% FBS, 2.5% cold-water fish skin gelatin, 0.1% Triton-X, and 0.05% Tween 20) for 1 h followed by incubation with Phalloidin⁶⁴⁷ in incubation solution (DPBS 5% FBS, 1% cold-water fish skin gelatin, 0.1% Triton X, and 0.05% Tween 20) at RT for 1 h. Cells were washed and incubated with DAPI for nuclei staining and then were fixed again with 4% paraformaldehyde at room temperature for 15 min. After washing, coverslips were mounted using a ProLong Gold Antifade Mountant. Images were captured with a Nikon TE2000 microscope using a Photometrics CoolSnap HQ2 camera. Images were captured by epifluorescence with a Sedat Quad cube (Chroma Technology Corp). Cell spreading was quantified by measuring cell diameters using 200 random cells with Phalloidin staining by ImageJ.

dHL60 Cell Adhesion Assay.

96-well plates were coated with 100 μL of fibronectin (10 μg/mL) in PBS at 4 °C overnight followed by two PBS washes. The adhesion assay was performed as previously described with modification.⁵² dHL60 cells (3 × 10⁵ cells/well) were incubated with 80 nM SSL protein at 37 °C for 30 min followed by two PBS washes. Adherent cells were incubated with 0.5% crystal violet at RT for 10 min followed by four PBS washes. The plate was dried and followed by the addition of ethanol to solubilize cell-bound crystal violet; absorbance at 595 nm was measured to quantify crystal violet and shown in adhesion arbitrary units (AU).

SSLs Sequence Alignment.

SSLs sequences were aligned using Clustal Omega.^23,24 NCBI sequence references: SSL1, WP_000669005.1; SSL2, WP_000782618.1; SSL3, WP_000784024.1; SSL4, WP_000705644.1; SSL5, WP_000784244.1; SSL6, WP_000769845.1; SSL7, WP_000769836.1; SSL8, WP_000673479.1; SSL9, WP_000779446.1; SSL10, WP_000673051.1; SSL11, WP_000769163.1; SSL12, WP_000041896.1; SSL13, WP_001063559.1; SSL14, WP_000739523.1.

dHL60 Cell Motility Assay.

dHL60 cells (5 × 10⁵ cells) were incubated with 80 nM SSL11, SSL7, or SSL7/11 Chimera at 37 °C for 30 min. A total of 10 μM fMLP was added to an edge of the well, and cell motility was recorded by taking brightfield DIC images every 10 s for 30 min on a 37 °C heated stage with a Nikon TE2000 microscope using a Photometrics CoolSnap HQ2 camera. Twenty cells from each group were tracked using ImageJ “manual tracking”, and data were imported to the Ibidi Chemotaxis and Migration Tool and plotted by GraphPad Prism 7.03 to show the individual cell moving tracks and total migration length (μm).

Data Analysis and Statistics.

Images were generated with equal exposure times and conditions. Image intensity analysis was performed using ImageJ (NIH). Figures were compiled using Canvas X 2017 (ACD Systems) and Adobe Illustrator 2021. Data were shown as means with standard deviation (SD). Data were analyzed by unpaired two-tailed Student’s t-test using GraphPad Prism 7.03 from three independent experiments. * indicates P values of <0.05 at a 95% confidence level; ** indicates P values of <0.005; *** indicates P values of <0.001; **** indicates P values of <0.0001.