Bacterial preparation

The S. aureus strain was isolated from a patient with chronic tibial osteomyelitis the pathogen of which was identified as methicillin-sensitive S. aureus. For resuscitation of S. aureus, 10 µl of frozen sample was inoculated to 3 ml tryptic soy broth (TSB) medium and shaken at 180 rpm in an incubator shaker at 37 ℃. After 16–18 h of incubation, 1 ml of bacterial suspension was collected and centrifuged at 2500 g for 5 min. The S. aureus pellet was resuspended in 1 ml PBS and centrifuged again before washes repeated for three times. The concentration of S. aureus was adjusted to an optical density (OD) of 0.5 at 600 nm, approximately equal to 1 × 108 colony forming unit (CFU) per milliliter.

Implant-associated osteomyelitis mouse models and treatments

All animal experimental protocols were approved by the Animal Care and Use Committee of Nanfang Hospital, Southern Medical University. C57BL/6 male mice were obtained from the Animal Center of Southern Medical University and housed under a 12-h light-dark cycle, 24 ± 2 ℃ room temperature, and ad libitum access to water and food. Young (8-week-old) and middle-aged (10-month-old) mice were used to establish implant-associated S. aureus osteomyelitis models, as we described previously [21]. By day 3 post-infection, infected femurs were harvested and either subjected to quantify bacterial load in bone tissue or processed to histological staining.

For the CXCR3 blockade treatment in vivo, 8-week-old mice were injected subcutaneously with AMG487 (5 mg/kg, #HY-15,319, MedChemExpress, USA) 6 h after infection, and then injected again every 12 h for a total of 6 times, while the control group was injected with the same volume of vehicle (1% DMSO). To activate CXCR3 signaling in middle-aged mice with S. aureus osteomyelitis, 10-month-old mice were injected with 100 ng recombinant mouse CXCL9 protein (#C600269, BBI Life Science, Shanghai, China) or recombinant mouse CXCL10 protein (#HY-P722, MedChemExpress, USA) into the bone marrow cavity using a microsyringe, while the control group was injected with the same volume of PBS. Subsequently, bacterial inoculation and placement of internal implants were performed. By day 3 post-infection, infected femurs were harvested and processed to histological staining.

Quantification of bacterial load in bone tissue

To quantify the colony number of S. aureus in infected femurs, mice were sacrificed by day 3 after infection. The femurs were isolated and then crushed with a tissue homogenizer. The homogenized femurs were weighed and made into homogenate at a weight to volume ratio of 0.1 g/ml in phosphate buffer saline (PBS). The homogenate was subjected to serial dilutions and incubated on TSB agar plates at 37 ℃ for 24 h.

Histological staining

The operated femurs were collected by day 3 after operation, then fixed with 4% paraformaldehyde for 24 h, decalcified with 0.5 M Ethylenediaminetetraacetic acid (EDTA) for 7 days, and finally embedded in paraffin. Paraffin sections at 4 μm were collected for hematoxylin and eosin (H&E) or immunohistochemical and immunofluorescence staining.

For immunofluorescence staining of S. aureus, after deparaffinization and rehydration of paraffin sections, antigen retrieval was performed with Tris-EDTA solution (pH = 9) at 70 ℃ for 30 min, followed by blocking with 10% goat serum at room temperature for 1 h. Sections were incubated with the rabbit anti-S. aureus antibody (#Ab20920, Abcam, USA) at 4 ℃ overnight. iFluor 594-conjugated goat anti-rabbit IgG (#HA1122, Huabio, Hangzhou, China) was used as a secondary antibody. Nuclei were counterstained with DAPI (#E607303-0002; BBI Life Science, Shanghai, China). Sections were observed under a BX53 microscope (Olympus, Tokyo, Japan) and analyzed by Image J (v1.8.0).

For immunohistochemical analysis, after deparaffinization and rehydration of paraffin sections, antigen retrieval was performed with Tris-EATA (pH = 9) for 3 h at 70 ℃ and endogenous peroxidase was inactivated by 3% H2O2 for 15 min. After being blocked with 10% goat serum for 1 h at room temperature, sections were incubated with the rabbit anti-Phospho- mixed-lineage kinase domain-like (MLKL) antibody (ET1705-51, Huabio, Hangzhou, China) or anti-CXCL9 antibody (#ABS124268, Absin, Shanghai, China) or anti-CXCL10 antibody (#DF6417, Affinity, Liyang, China) overnight at 4 °C. Next, horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG antibody (HA1001, Huabio, Hangzhou, China) was used as a secondary antibody and incubated for 1 h at room temperature. Then, peroxidase activity was revealed by 3,3-N-Diaminobenzidine tertrahydrochloride (DAB) substrate kit (#ZLI-9017, ZSGB-BIO, Beijing, China) according to manufacturer’s protocol. Finally, the nuclei were counterstained with hematoxylin (#H-3404-100, Vector). Sections were observed under a microscope (Eclipse Ci-L plus, Nikon, Tokyo, Japan) and analyzed by Image J (v1.8.0).

Flow cytometry

The infected mice femurs and control ones were harvested on day 3 after surgery. The bone marrow cells were flushed out and collected in RPMI 1640 medium. After being filtered through a 70 μm strainer, whole bone marrow cells were counted after lysis of red blood cells. To block intracellular cytokine secretion, cells were treated with Brefeldin A (#GC17683, GLPBIO, USA) at a concentration of 10 µg/ml and incubated at 37 ℃ for 2 h. Next, cells were incubated with anti-mouse CD16/32 (#101,319, Biolegend, USA) to block non-specific staining on ice, followed by incubation with primary antibodies, including BV421-conjugated anti-mouse CD11b antibody (#101,235, Biolegend, USA), BV510-conjugated anti-mouse Ly6C antibody (#128,033, Biolegend, USA), PerCP/Cy5.5-conjugated anti-mouse Ly6G antibody (#127,615, Biolegend, USA), APC/CY7-conjugated anti-mouse F4/80 antibody (#123,117, Biolegend, USA), FITC anti-mouse CD3 (#100,203, Biolegend, USA), APC anti-mouse CD8a (#100,711, Biolegend, USA), or PE anti-mouse CD4 (#100,407, Biolegend, USA). Finally, samples were detected on the flow cytometer (LSRFortessa X-20, BD Bioscience, USA) and analyzed by FlowJo (V10).

Bioinformatics analysis

The DESeq2 R language (1.16.1) was used to analyze our previous transcriptome data (GEO: GSE166522) of bone tissue from mice (10–12 weeks old) with S. aureus osteomyelitis and control ones by day 3 after operation. Differential expressed genes (DEGs) were determined if the gene | log2FoldChange | > 1 and adjusted p values < 0.05. ClusterProfiler R software (3.6.2) was used to perform Gene Ontology (GO) enrichment analysis of differential genes, and adjusted p value < 0.05 was considered as significant enrichment.

Isolation and identification of monocytes, neutrophils and bone marrow derived macrophages (BMDMs)

After collection of bone marrow single cell suspension, bone marrow neutrophils and monocytes were isolated by density gradient centrifugation. Briefly, 3 ml of Percoll working solution and suspended single cell solution with concentrations of 1.09, 1.077 and 1.043 g/ml were successively added into a 15 ml centrifuge tube, and centrifuged at room temperature for 35 min under slow acceleration condition. The neutrophils were extracted at the interface of 1.09 g/ml and 1.077 g/ml. Monocytes-lymphocytes were extracted at the interface of 1.077 g/ml and 1.043 g/ml and then centrifuged at 1.067 g/ml Percoll working solution to further separate the upper layer of monocytes. The collected neutrophils and monocytes were washed twice with PBS and resuspended in RPMI 1640 containing 10% FBS before the next experiment.

For BMDMs isolation, bone marrow cells from 8-week-old or 10-month-old mice were flushed out with PRMI 1640 medium, centrifuged, resuspended in macrophages growth medium (RPMI 1640, 10% FBS, 20% L929 cell-conditioned medium, and 1% penicillin/streptomycin) at 2 × 106 cells /ml, and then seeded in cell culture dishes. After 7 days, bone marrow cells were differentiated into mature BMDMs.

The purity of isolated neutrophils, monocytes, and BMDMs from mice bone marrow were analyzed by detecting expression of cell surface markers and evaluating the proportion of CD11b+Ly6G+, CD11b+Ly6C+, and CD11b+F4/80+ cell populations, respectively. For freshly isolated neutrophils and monocytes, cells were incubated with anti-CD16/CD32 antibodies (#101,319, Biolegend, USA) in 0.5% BSA/PBS buffer on ice for 10 min, and then stained with a combination of FITC anti-CD11b (#101,205, Biolegend, USA) plus BV510 anti-Ly6G (#127,633, Biolegend, USA), and a combination of FITC anti-CD11b (#101,205, Biolegend, USA) plus PerCP anti-Ly6C (#128,027, Biolegend, USA), respectively, at room temperature for 30 min. For evaluating the purity of primary BMDMs, after 7 days differentiation, BMDMs were trypsinized with 0.25% Trypsin-EDTA, centrifugated and washed twice with PBS, then labeled with FITC anti-CD11b (#101,205, Biolegend, USA) and APC/Fire 750 anti-F4/80 (#123,151, Biolegend, USA) at room temperature for 30 min. Unstained cells were used as the negative control. After being washed twice, cells were resuspended in 0.5%BSA/PBS buffer and immediately detected using a Beckman CytoFLEX flow cytometer (A00-1-1102).

Cell treatment and conditional medium (CM) experiment

BMDMs (2 × 106 cells/well) were seeded in 6-well plates and stimulated with S. aureus in macrophages growth medium without antibiotics at various multiplicity of infection (MOI) (0, 2, 10 and 50) for 1 h. After 1% penicillin/streptomycin was added to the co-culture system to inhibit bacterial overgrowth, the culture was continued for 5 h. The medium was discarded and washed twice with PBS before finally TRIZOL (#AG21102, Accurate Biology, Changsha, China) was added to the well for total RNA extraction.

2 × 106 cells /ml monocytes or neutrophils suspension was seeded per well in 12-well plates. 1 ml. S. aureus at various MOI (0, 2, 10 and 50) was used to stimulate cells for 1 h, and then 1% penicillin/streptomycin was added to the co-culture system to inhibit bacterial overgrowth. The culture was continued for 5 h. After centrifugation at 300 g for 5 min, the pellet of monocytes or neutrophils was resuspended in TRIZOL for total RNA extraction. For CM preparation, the supernatant of monocytes was collected into a new centrifuge tube, centrifuged at 2500 g for 5 min to remove most of the bacteria, and finally filtered through a 0.22 μm filter. CM of monocytes treated with S. aureus (MOI = 10, CM Mono-S. aureus) and CM of monocytes treated with PBS (MOI = 0, CM Mono-vehicle) were reserved for next experiments.

For siRNA knockdown experiment, monocytes from 8-week-old mice were isolated as mentioned above, and resuspended in Opti-MEM medium (#31985070, GIBCO). Silencing sequences were transferred into cells with transfection reagents (#40806ES02, Yeasen, Shanghai, China) according to the manufacturer’s instructions and cultured for 24 h. After cells were harvested, the viable cell concentration was readjusted to 2 × 106 cells /ml. 1 ml per well was seeded into 12-well plates. Afterwards, S. aureus stimulation procedures as well as cellular RNA collection and conditioned medium collection were performed as described above. Three fragments each for CXCL9 and CXCL10 were screened for evaluating their knock-down efficiency. The sequences of siRNA fragments used for further experiments were as follows: si-CXCL9, 5’GUCGUCGUUCAAGGAAGACUAdTdT3’ and 5’UAGUCUUCCUUGAACGACGACdTdT3’; si-CXCL10, 5’CGGAAUCUAAGACCAUCAAdTdT3’ and 5’UUGAUGGGCUUAGAUUCCGdTdT3’; si-negative control (si-NC), 5’UUCUCCGAACGUGUCACGUdTdT3’ and 5’ ACGUGACACGUUCGGAGAAdTdT3’.

Quantitative real-time PCR (qPCR)

Bone tissue or cell total RNA was reverse transcribed into cDNA using a reverse transcription kit (11141ES60, Yeasen, Shanghai, China), and qPCR was performed using SYBR Green (11202ES08, Yeasen, Shanghai, China) on QuantStudio6 (Applied Biosystems, USA) according to the manufacturer’s protocol. The primer sequences were obtained from the PrimerBank database [22] as follows: CXCL9, forward: 5’GGAGTTCGAGGAACCCTAGTG3’, reverse: 5’GGGATTTGTAGTGGATCGTGC3’; IL-1β, forward: 5’TCCTGTGTAATGAAAGACGGC3’, reverse: 5’ACTCCACTTTGCTCTTGACTTC3’; CXCL10, forward: 5’CCAAGTGCTGCCGTCATTTTC3’, reverse: 5’GGCTCGCAGGGATGATTTCAA3’; GAPDH, forward: 5’TGTCGTGGAGTCTACTGGTG3’, reverse: 5’GCATTGCTGACAATCTTGAG3’.

Analysis of live/dead viability and ROS production of isolated neutrophils

Given that neutrophils in ex vivo culture have a short life-time [23], we evaluated the viability of neutrophils after S. aureus infection. The isolated neutrophils were cultured in growth medium (RPMI 1640 containing 10% FBS) for 4 h, followed by 30 min of S. aureus challenge at MOI of 10. After centrifugation at 300 g for 5 min, the cell pellet was washed twice with PBS, then the cell viability was tested using a calcein/propidium iodide (PI) viability/cytotoxicity assay kit (#C2015M, Beyotime, Shanghai, China). Cells were incubated in buffer containing calcein and PI for 30 min at 37℃ in the dark following the manufacturer’s protocol. After being washed twice with serum-free medium, cells were resuspended in 0.5 ml flow cytometry buffer and immediately detected using a Beckman CytoFLEX flow cytometer (A00-1-1102).

The activated neutrophils destroy bacteria by producing reactive oxygen species (ROS), neutrophil extracellular traps (NETs) formation and release of antimicrobial proteinase [23]. We therefore evaluated the responsivity of neutrophils by determine the levels of ROS production in response to S. aureus challenge. After 4 h of culture and 30 min of S. aureus challenge as aforementioned, cells were incubated with a staining solution containing 10 µM dihydroethidium (DHE) (#D1004, UEland, Suzhou, China) at 37℃ in the dark for 30 min. Subsequently, cells were washed twice with serum-free medium and immediately detected using a Beckman CytoFLEX flow cytometer (A00-1-1102).

Analysis of bactericidal activity of neutrophils and macrophages

For the neutrophil bactericidal assay, 1 × 106 neutrophils were stimulated with 500 µl of the medium containing different stimuli for 4 h, washed twice with PBS and resuspended in neutrophil medium. Next, S. aureus (MOI = 10) was added and co-cultured for 30 min. After centrifugation at 300 g for 5 min, the cell pellet was washed twice with PBS, and the washing solution was transferred to a new centrifuge tube together with the cell supernatant from the first centrifugation. This solution contained the remaining extracellular bacteria that had not been killed. Cell pellet was lysed with 1 ml of 0.2% Triton X-100 (#T8200, Solarbio) for 15 min at room temperature, thereby releasing residual intracellular bacteria. The remaining intracellular bacteria and the remaining extracellular bacteria were diluted in 10×, 100× and 1000×, and 10 µl of the bacterial solution was inoculated on TSA plates and incubated at 37 ℃ for 20 h.

For macrophage bactericidal assay, 1 × 106 BMDMs were stimulated with 500 µl of the medium containing different stimuli for 4 h, and subsequently co-cultured with S. aureus (MOI = 10) for 30 min. After washing twice with PBS, the extracellular bacteria were killed with gentamicin (50 µg/ml) and lysozyme (20 µg/ml) for 30 min. Then the cells were lysed with 0.2% Triton X-100 to release the remaining intracellular bacteria. After 10×, 100× and 1000× dilution, 10 µl of the bacterial solution was inoculated on TSA plates and incubated at 37 ℃ for 20 h.

The medium containing different stimuli mentioned above included: CM of S. aureus-challenged monocytes from 8-week-old mice or 10-month-old mice (CM-8 W Mono-S. aureus, CM-10 M Mono-S. aureus), CM of vehicle-treated monocytes from 8-week-old mice or 10-month-old mice (CM-8 W Mono-Vehicle, CM-10 M Mono-Vehicle), CM-8 W Mono-S. aureus pretreated with siRNA fragments targeting CXCL9, or CXCL10, or both CXCL9 and CXCL10, or pretreated with 100 ng/ml of recombinant CXCL9, or recombinant CXCL10, or a combination of CXCL9 and CXCL10. For use of CXCR3 inhibitor, it was pre-stimulated with 500 µM AMG487 for 1 h before addition of CM-8 W Mono-S. aureus, recombinant CXCL9 or CXCL10.

Statistical analysis

GraphPad Prim 9 was used for statistical analysis of the data. If the data were in accordance with normal distribution and homogeneity of variance, the Student’s t test was used to compare the statistical differences between the two groups. One-way analysis of variance (One-way ANVOA) with Bonferroni or Dunnett’s T3 post hoc test was used for statistical comparison of more than two groups. All data are expressed as mean ± SD. p < 0.05 was considered statistically significant.