All animal experiments were carried out in accordance with the Guide for the Care and Use of Laboratory Animals of the Chinese Association for Laboratory Animal Sciences (CALAS) and the protocol was approved by the ethics committee of Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China. Human heparinized venous blood was obtained from healthy individuals in accordance with a protocol approved by the ethics committee of Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China. All individuals gave written informed consent in accordance with the Declaration of Helsinki prior to donating blood.

Clinical S. aureus isolates were obtained from a comprehensive teaching hospital in Shanghai, China (Renji Hospital, School of Medicine, Shanghai Jiao Tong University). Bacteria were identified as staphylococci by classic microbiological methods, including Gram staining, catalase and coagulase activity on rabbit plasma. S. aureus strains were further categorized by VITEK2 automated systems (BioMérieux, France). CA-SA was defined as previously described (Li et al., 2016). Isolates were obtained either from an outpatient or from an inpatient <24 h after hospital admission and lacked the following risk factors: contact with the hospital environment in the preceding 6 months, residence in a long-term care facility in the preceding 12 months, S. aureus infection in the preceding 12 months, and presence of a central vascular catheter at the time of infection. All bacterial strains and plasmids used in this study are listed in Table 1. S. aureus was grown in tryptic soy broth (TSB; Oxoid) with 0.25% glucose or on agar plates at 37°C, and Escherichia coli was routinely grown in Luria-Bertani medium (LB; Oxoid). When necessary, antibiotics were used at the following concentrations: ampicillin, 100 μg/ml; chloramphenicol, 10 μg/ml. All oligonucleotides used in this study are listed in Table 2.

Multi-locus sequence typing (MLST) was performed as previously described (Enright et al., 2000). PCR amplicons of seven S. aureus housekeeping genes (arcC, aroE, glpF, gmk, pta, tpi, and yqiL) were obtained from chromosomal DNA. The sequences of the PCR products were compared with the existing sequences available at the MLST website¹.

Sequence alignment of esxX gene among clinical S. aureus ST398 isolates was performed using MEGA software. The esxX gene sequence was used for BLAST in the NCBI database. The amino acid sequence alignment of EsxX both from clinical isolates and the NCBI database were analyzed by GENEDOC software.

Bacteria were grown to an OD₆₀₀ of 5.0, and filtrates of the cultures were precipitated with 10% trichloroacetic acid (TCA and washed 3 times with ice-cold acetone. The samples were then resuspended in sample buffer (50 mM Tris-HCl pH 8.0, 4% SDS) and then mixed with protein loading buffer and boiled for 10 min. A 12% SDS-PAGE was used to pre-fractionate the secreted proteins and then the protein samples were subjected to in-gel digestion. Finally, the extracted peptides were vacuum-dried prior to LC-MS/MS analysis on a nanoflow liquid chromatography instrument (EASY-nLC 1000, Thermo Scientific) coupled to an ion trap mass spectrometer (LTQ Velos Pro, Thermo Scientific) in the data-dependent mode. The detailed LC-MS/MS settings for this procedure have been described previously (Hu et al., 2014).

For gene deletion in CA-SA ST398, a representative clinical isolate (RJ-ST398) was chosen; this isolate was recovered from an abscess of a patient with a skin infection (Wang et al., 2016). The homologous recombination procedure using the plasmid pKOR1 was performed as described (Bae and Schneewind, 2006). DNA fragments for the upstream and downstream sequences of esxX were PCR amplified from the chromosomal DNA of RJ-ST398, and overlap PCR was used to obtain a fused PCR product. This product was cloned into pKOR1 using a Clonase reaction and attB sites, yielding the plasmid pKOR1ΔesxX. The plasmid pKOR1ΔesxX was transferred via electroporation into S. aureus RN4220 and then into RJ-ST398. The allelic replacement procedure was performed as described by others (Vuong et al., 2003). Proper integration was verified by PCR amplification from genomic DNA using esxX-att1 and esxX-att2 primers and sequencing of the PCR product. For genetic complementation, the esxX gene was amplified by PCR with the primers esxX-SmaI-F and esxX-BamHI-flag-R. The esxX complementation plasmid was generated by cloning the esxX gene into vector pOS1 and was then transferred into RJ-ST398. The plasmid control strain was obtained by transferring the pOS1 plasmid into RJ-ST398.

Overnight cultures were diluted 1:100 into 10 ml of TSB and incubated at 37°C with shaking at 200 rpm until they had grown to an OD₆₀₀ of 2.0 (mid-exponential growth phase), and 1.5 ml of culture was centrifuged at 10,000 × g for 4 min to separate the cells from the supernatant. Proteins in 1 ml of supernatant were precipitated with 10% TCA using 3 ice-cold acetone washes and then resuspended in sample buffer (50 mM Tris-HCl pH 8.0, 4% SDS). The cells were resuspended in 80 μl of TE buffer (10 mM Tris-HCl, 1mM EDTA, pH 8.0). Lysostaphin (100 μg/ml) was added for a 10 μg/ml final concentration, and the samples were incubated for 30 min at 37°C. Supernatant (S) and cell lysate (CL) samples were both mixed with protein loading buffer and boiled for 10 min. For subcellular localization, 2 ml cultures were centrifuged as described above, and the supernatants were precipitated with TCA. The cell pellet of a 2 ml culture were was with TSM buffer, and then suspended in 500 μl of TSM buffer containing 10 μl of lysostaphin (2 mg/ml stock) and incubated at 37°C for 30 min. The protoplasts were collected by centrifugation at 10,000 × g for 10 min, and the cell wall fraction (W) was precipitated with TCA. Protoplasts were suspended in 5ml of membrane buffer (0.1 M Tris-HCl, 0.1 M NaCl, pH 7.5, 0.01 M MgCl₂) and sonicated on ice at full power for 10 s. Soluble proteins (C, cytoplasmic fraction) were separated from insoluble materials and membranes (M, membrane fraction) by ultracentrifugation at 45k rpm for 30 min, at 4°C. All samples were precipitated with TCA before Western blotting.

Equal volumes (10 μl) of all proteins of interest were separated by SDS-PAGE and transferred onto nitrocellulose membranes (Invitrogen). After blocking, the membranes were incubated with specific antiserum at 4°C overnight and then incubated with a horseradish peroxidase-conjugated secondary antibody at room temperature for 1 h. Images of Western blots were acquired using a Tanon-5200 system. In total, eight specific antibodies were used for Western blot analysis in this study, including α-EsxX, α-EsxA, α-EsxB (generated by GLBiochem, China); α-SrtA, α-SeaS, and α-SeaR (kindly provided by Taeok Bae, Indiana University School of Medicine-Northwest, Gary, Indiana); α-Hla (Abcam) and α-Spa (Sigma-Aldrich).

Complementary DNA (cDNA) was synthesized from total RNA using the QuantiTect RT system (Qiagen) according to the manufacturer’s instructions. Oligonucleotide primers were designed using Primer Express (Table 2). The resulting complementary DNA and negative control samples were amplified using the QuantiTect SYBR green PCR kit (Qiagen). Reactions were performed in a MicroAmp Optical 96-well reaction plate using a 7500 Sequence Detector (Applied Biosystems). Standard curves were determined for each gene using purified chromosomal DNA at concentrations of 0.005–50 ng/ml. All quantitative reverse-transcription polymerase chain reaction (qRT-PCR) experiments were performed in duplicate with gyrB as an internal control.

Strains were grown to the mid-exponential growth phase (4 h), and equal amounts (2 μl) of cells were spotted on sheep blood agar plates. After incubation at 37°C for 24 h, the zones of hemolysis (diameter) were measured.

Semiquantitative biofilm assays were performed as described previously (Vuong et al., 2003). Cells were fixed by Bouin’s fixative. The fixative was removed after 1 h incubation, and wells were washed with phosphate-buffered saline (PBS). Organisms in the wells were then stained with crystal violet, and the floating stain was washed off with slow-running water. After drying, the stained biofilm was scanned with a Micro ELISA autoreader (BioRad) at 570 nm.

Human neutrophils were isolated from heparinized venous blood of healthy individuals with a standard method. Bacteria were grown to the mid-exponential growth phase and incubated with neutrophils at a 10:1 ratio (MOI 10) at 37°C for 3h. PBS with 0.1% Triton-X100 (100 μl) was used to determine 100% lysis. Lysis was measured using a lactate dehydrogenase (LDH) cytotoxicity detection kit (Roche) according to the manufacturer’s protocol as described elsewhere (Li et al., 2012). To determine bacteria survival rates, 100 μl of the culture above was serially diluted and plated on TSA containing 5% sheep blood for S. aureus colony forming units (CFU) counting.

For microscopic evaluation of neutrophil killing after phagocytosis and intracellular S. aureus counting, overnight cultures were diluted 1:100 into 10 ml of TSB and incubated at 37°C with shaking at 200 rpm to the mid-logarithmic growth phase (4 h). The cells were washed and resuspended in sterile PBS at a concentration of 1 × 10¹⁰ CFU/ml. S. aureus cells were mixed with heparinized venous blood at a ratio of blood/bacterial solution of 10:1 (v/v) and incubated at 37°C. After the indicated times of incubation, blood smears were prepared and cells were stained with a modified Wright-Giemsa stain. The number of neutrophils lysed by S. aureus among the total number of neutrophils that had ingested S. aureus (n = 100 analyzed) was calculated by microscopic examination at 30 min after incubation. The number of intracellular S. aureus after 30 min of infection was counted using 20 neutrophils per group.

The skin abscess model was performed as described elsewhere (Wang et al., 2007). Outbred, immunocompetent hairless female mice between 4 and 6 weeks of age were used for the experiments. Anesthetized mice were inoculated subcutaneously on the back with 100 μl of PBS containing 10⁷ CFU of live S. aureus or an equal volume of PBS alone. The next day, abscess length (L) and width (W) values were measured to calculate the area of abscesses with the formula area (A) = π (L × W)/2. The abscess skin tissue was excised, washed with saline, and fixed in 4% formalin (Sigma-Aldrich). Paraffin embedding and hematoxylin and eosin (H&E) staining were performed.

For the bacteremia model, BALB/c female mice between 4 and 6 weeks of age were used. S. aureus strains were grown to mid-logarithmic phase, washed once with sterile PBS and then suspended in PBS. Each mouse of one group was injected with 100 μl of PBS containing 10⁸ CFU of live S. aureus into the retro-orbital vein, and another group was injected with 100 μl of PBS containing 10⁹ CFU of each for survival analysis. Control animals received sterile PBS only. After inoculation, the health of the mice and the advancement of their disease state were monitored every day. The mice were euthanized immediately if they showed signs of respiratory distress, mobility loss or inability to eat and drink. The surviving mice injected with 10⁸ CFU were euthanized 4 days after injection. Kidneys were excised, washed with saline, and one kidney was fixed in 4% formalin (Sigma-Aldrich). Paraffin embedding and hematoxylin and eosin (H&E) staining were performed. The other kidney was homogenized in 0.5 ml of PBS, and the homogenized kidney tissue was diluted and plated on TSB agar for the determination of CFU numbers. All surviving mice injected with 10⁹ CFU were euthanized at 5 days.

Statistical analysis was performed using Graph Pad Prism, version 6.01. The data were analyzed using unpaired t-tests to compare two different conditions. All error bars show the standard error of the mean (SEM).

The ess locus of USA300 or Newman has been well characterized. There are six core genes from esxA to essC and several other genes of the system (Figure 1A), in which the EsaA, EssA, EssB, and EssC proteins are membrane-located components of the secretion machinery, EsaB is a regulator, and EsxA, EsxB, EsxC, and EsxD are exported effector proteins (Burts et al., 2005, 2008; Anderson et al., 2011; Kneuper et al., 2014); three additional components of the ESS pathway, EssE, EssD, and EssI, were recently identified (Anderson et al., 2017; Ohr et al., 2017). Compared with USA300 or Newman, the ess locus of ST398 contains the same six core genes from esxA to essC, as well as another exported effector encoded gene esxB, but the esxC, esxD, essE, essD, and essI genes are all absent; instead, there are other genes of unknown function (Figure 1A). We found one gene located downstream of the esxB gene (Figure 1A) that encodes a 517 amino acid protein we named EsxX. The ESS has two identified secreted virulence factors, EsxA and EsxB, both of which are members of the WXG100 superfamily. The WXG motif of EsxA and EsxB is referred to as a signature sequence of ESAT-6-like proteins that can be secreted by ESS (Pallen, 2002). EsxX, which also belongs to the WXG superfamily, has a unique LXG domain at the N-terminal (Figure 1B; Schneewind and Missiakas, 2012), indicating that the protein might have secretion potential. This is consistent with a study proposing that EsxX might be secreted by the ESS (Warne et al., 2016). Furthermore, we performed sequence alignment of esxX among clinical S. aureus isolates and strains in the NCBI database and found that it is highly conserved in S. aureus lineage ST398. 61 isolates of S. aureus ST398 derived from infected patients and 15 isolates derived from cattle suffering from mastitis were collected in this study. Whole genome sequencing of the 76 isolates was performed, and the present study used the sequence of the esxX gene of the 76 clinical isolates for sequence alignment. According to the alignment results, 86.8% of isolates (66/76) showed 100% sequence identity, and the other 13.2% of isolates showed 99% identity, differing only by one point mutation (Supplementary Table 1). In the NCBI BLAST database, there is no homologous protein of EsxX in USA300 or other S. aureus strains. Only one ST582 and four ST15 strains were discovered to have 96% sequence identity and they had lower scores and more gaps than the other strains belonging to ST398 which showed 100 or 99% sequence identity of esxX (Supplementary Table 2). These findings suggest that EsxX is a clone-specific protein of ST398 that might have special functions.

Our previous study confirmed that ST398 ESS is responsible for the secretion of EsxA and EsxB via LC-MS/MS analysis of ST398 wild type versus essB deletion mutant culture filtrates (Wang et al., 2016). It is worth mentioning that in the same LC-MS/MS analysis, EsxX was not detectable in the essB deletion mutant culture filtrate (Figure 2A), indicating that its secretion is controlled by the ESS of ST398. To verify the secretion property of EsxX, we constructed an isogenic gene deletion mutant of esxX by a representative ST398 isolate selected previously (Wang et al., 2016). We performed a Western blot to determine whether EsxX was secreted. EsxX was detected both in the CL and the supernatant of ST398 using specific rabbit antisera (Figure 2B). Sortase-A and Hla were used as controls for proteins in the CL and the supernatant, respectively. Furthermore, to examine the subcellular localization of EsxX, cultures of ST398 were separated into cytoplasmic (C), membrane (M), cell wall (W), and supernatant (S) fractions (Figure 2C). Proteins in all fractions were examined by Western blotting with specific antibodies. EsxX was found predominantly in the supernatant and partially in the cytoplasm. As a control, Hla was detected in the supernatant, whereas Spa resided in both the cell wall and supernatant fraction, SaeS in the membrane, and SaeR in the cytoplasm (Figure 2C). Together, these results demonstrate that EsxX is a novel secreted protein controlled by ST398 ESS. Because EsxA and EsxB are two ESS effectors that play important roles in the function of ESS, we then used Western blotting to analyze whether EsxX impacts their secretion. EsxA and EsxB were detected in the supernatant fraction of both ST398 wild type and esxX mutant bacteria at the same level (Figure 2D), which suggested that the ESS was still completely functional when the esxX gene was knockout.

EsxA and EsxB, two canonical secreted proteins of ESS, have been described as secreted virulence factors in other S. aureus strains (Burts et al., 2005; Zhou et al., 2013). Previous work had shown that the esxA and esxB genes were under agr control at a transcriptional level (Wang et al., 2016). Because EsxX is a novel secreted protein of ST398 ESS, we questioned whether it had similar functions to the two Esx proteins. Quantitative RT-PCR and Western blotting were used to investigate EsxA, EsxB, and EsxX expression in ST398 wild type versus agr deletion mutant (Figure 3A). We found that EsxA and EsxB were agr-regulated, while EsxX was not (Figure 3B). We also found that there is no difference in growth between ST398 wild type and an esxX mutant (Figure 3C). To study the role of EsxX in ST398, we examined the capacity of the bacteria to induce hemolysis, one of the most important virulence phenotypes of S. aureus, but found no difference between the wild type and the esxX mutant strain (Figure 3D). S. aureus has been reported to form biofilms during infection, which promotes persistence and contributes to its successful colonization, making it difficult to eliminate the infection (Otto, 2008). However, the EsxX had no impact on biofilm formation (Figure 3E).

Recently, EsxA was reported to play a role in the immune response to invasive S. aureus disease (Zhang et al., 2015) and to delay apoptosis and contribute to the release of S. aureus from epithelial cells, together with EsxB (Korea et al., 2014). These reports inspired us to investigate whether the novel secreted protein EsxX impacts the interaction between S. aureus and host cells. We selected neutrophils because neutrophils represent the first line of defense against a S. aureus infection. We analyzed the lysis of neutrophils and the degree of bacterial survival by incubating wild type and esxX mutant bacteria with human neutrophils. Compared with the wild type strain, the esxX mutant showed significantly decreased neutrophil lysis (Figure 4A). The low bacterial survival of the esxX mutant (Figure 4B) showed that the mutant was more likely to be eliminated than the wild type. Further, a microscopic analysis was performed. We examined the numbers of intracellular S. aureus after 30 min of infection, and the data showed no impact of EsxX on uptake by neutrophils (Figure 4C), while the count of lysed neutrophils after ingestion (Figure 4D) suggested that EsxX contributed to neutrophil lysis after phagocytosis. These results indicated that EsxX impacts the capacity of ST398 to interact with human neutrophils and thus plays an important role in the evasion of innate host defenses.

Staphylococcus aureus causes a variety of infections, including superficial skin and soft tissues infections (SSTIs) and severe invasion infections such as bacteremia. After evaluating the impact of EsxX on the capacity of neutrophil lysis in vitro, we then explored its virulence potential in vivo using mouse skin abscess and bacteremia models. In the skin abscess model, the ST398 caused a significantly larger abscess area than the esxX mutant (Figures 5A,B), and histological examinations showed greater infiltration of inflammatory cells and skin tissue destruction in the mice infected with the wild type than those infected with the esxX mutant strain (Figure 5C). In a low-dose (10⁸ CFU) bacteremia model, the kidneys of mice infected with the ST398 wild type showed increased abscess formation (Figure 6A). Histological examination showed increased infiltration of inflammatory cells and tissue structure destruction in the kidneys of ST398 wild type-strain-infected mice compared with those esxX mutant-strain-infected mice (Figure 6B). Furthermore, the mice infected with the wild type strain showed notably higher CFU counts in the kidneys 4 days after infection than the mice infected with the esxX mutant strain (Figure 6C). In a high-dose (10⁹ CFU) bacteremia model, the mice infected with ST398 wild type all died within 96 h, while the mice infected with the esxX mutant survived for a longer time (Figure 6D); however, there was no statistical significance (P = 0.0993). The results of skin and blood infections together suggested that EsxX makes a significant contribution to the virulence of ST398.