The GBS strain BJCH_SA4 used here is a high virulent type III strain which is stored in our lab. The bacteria were grown overnight in Todd–Hewitt broth (THB) at 37°C containing 5% CO₂, washed twice in nonpyrogenic PBS after centrifugation (PBS; 0.01 M phosphate, 0.15 M NaCl, pH 7.2), and resuspended to the desired concentration. Heat-killed bacteria (HK-GBS) was prepared according to the methods previously reported (10). In each experiment, the actual number of injected bacteria was determined by colony counts. In βH/C-blocking experiments, the inhibitor DPPC(dipalmitoyl phosphatidylcholine, sigma-adlrich, USA) was suspended in PBS by sonication and added to GBS for 10 min before being injected into the orbita vein.

Fpr2^-/-(Fpr2 knockout [KO]) mice were constructed from Cyagen Biosciences (Guangzhou, China) according to methods previously reported (11). The control, wild-type (WT) C57BL/6 mice (6-8 weeks old), were purchased from Charles River (Beijing, China). The mice were maintained and bred under Specific pathogen-free(SPF) conditions in the animal facilities of the Academy of Military Medical Sciences.

Six-week-old mice were injected intraperitoneally (i.p.) or intravenously (i.v.) with the BJCH_SA4 GBS strain. For the two GBS-induced infected models, mice were injected i.p. with 10⁸CFU/ml mid-logarithmic-phase GBS or mice were firstly anesthetized and then injected in the orbital vein with 10⁸CFU/ml mid-logarithmic-phase GBS and monitored daily for survival. For receptor inhibition experiment, WT mice were treated with Boc-2 (600ng/kg) or vehicle for 1 h before GBS infection. Boc-2 peptide together with the lipidated peptidomimetic Lau-(Lys-βNSpe)6-NH2 ligand, were recently found to be selective inhibitors for Fpr2. For neutrophils depletion experiment, WT mice were pretreated i.p. with 100μg of rat monoclonal anti-mouse Ly6G antibodies or rat IgG2a isotype control (BD Pharmingen, USA) 24 h before i.v. inoculation with 2×10⁸ CFU/ml of GBS. For Chemokine depletion experiment, the mice were injected i.p. 3h before the GBS challenge with 100 μg rat monoclonal antibodies (mAbs) specific for CXCL1, CXCL2, or IgG2a and IgG2b isotype controls (R&D, Minneapolis, MN, USA). For monocyte/macrophage depletion experiment, mice were pretreated i.p. with 200μg of clodronate liposome or control PBS liposomes (Yeasen Biotech Co. Ltd, China) 24 h before i.v. inoculation with 2×10⁸ CFU/ml of GBS. After the infection, the peripheral blood was taken from the orbital vein into an EDTA anticoagulation tube for follow-up experiments. The organs were placed in preweighed sterile tubes containing PBS and homogenized in a gentle tissue disruptor.

Meanwhile, the HK-GBS was injected i.p. at a dose of 0.5 mg/mouse in 0.2 ml of PBS, while the peritoneal lavage fluid was collected at predetermined times to measure cell numbers via flow cytometry and cytokine concentrations using ELISA.

Concentrations of cytokine in the plasma and supernatants were detected by multiplexed Luminex xMAP assay (Cytokine & Chemokine 36-Plex Mouse ProcartaPlex™ Panel 1A, Thermo Scientific, USA) or ELISA according to the manufacturer’s instructions.

After the mice were infected for 3 hours and 6 hours, the peripheral blood was taken into an EDTA anticoagulation tube. The red blood cell was lysed on ice for three minutes and centrifuged at 2300 rpm for 3 minutes, while the supernatant was removed to obtain the cells. The same method was used to obtain cells from the peritoneal lavage fluid. The absolute count of blood leukocytes in the blood and peritoneal lavage fluid samples were determined using a BD TruCount system (BD Biosciences, USA). The cells were stained with PE anti-mouse Ly6G, PerCP/Cyanine5.5 anti-mouse CD45, FITC-anti-mouse CD11b, Apc anti-mouse Ly6C, PE/cy7 anti-mouse CD19, and Apc anti-mouse CD3. Antibodies were purchased from Biolegend (San Diego, CA, USA). All of the above-stained cells were assayed using a BD FACSVerse flow cytometer and data were analyzed using FlowJo software.

After three hours of GBS infection, the WT and Fpr2 knockout mice were used to isolate neutrophils with the Tianjin Haoyang Neutrophil Kit. Once the purity exceeded 95%, trizol was added, and the samples were placed in liquid nitrogen and sent to the Beijing Novogene company, which conducted sequencing and analyzed the results. The reads of each gene were counted by feature counts, and the differentially expressed genes had adjusted p-values <0.01 and were differentially expressed by at least 2-fold. Quantitative PCR (qPCR) was also performed to verify the results of RNA-seq. The SYBR Green PCR method was used according to the instructions of the manufacturer of the PowerUp™ SYBR™ Green Master Mix (Thermo Scientific, USA), while the GAPDH gene was used as an internal control.

Liver FFPE tissue of mice was used as an experimental sample. All tissues were cut into 4-μm thick sections. The slides were deparaffinized in xylene for 30 mins and rehydrated in absolute ethyl alcohol for 5 mins (twice), 95% ethyl alcohol for 5 mins, 75% ethyl alcohol for 2 mins. The slides were washed with distilled water three times. A microwave oven was used for heat-induced epitope retrieval; during epitope retrieval, the slides were immersed in boiling H₂O₂ buffer (PH9.0; Histova Biotechnology, China) for 15 mins. Primary antibodies were performed as follows: Ly6G (1:500, 87048s, Cell Signaling Technology, Danvers MA, USA), CXCL1 (1:500, ab269939, Abcam, USA). All primary antibodies were incubated for 1 h at 37°C. The slides were then incubated with Opal Polymer HRP Ms+Rb (2414515; PerkinElmer, Massachusetts, USA) for 10 min at 37°C. An Opal Seven-Color IHC Kit (NEL797B001KT; PerkinElmer, Massachusetts, USA) was used for visualization. After each cycle of staining, heat-induced epitope retrieval was performed to remove all antibodies, including primary antibodies and the Opal Polymer HRP Ms+Rb. The slides were counterstained with DAPI for 5 mins and enclosed in Antifade Mounting Medium (I0052; NobleRyder, Beijing, China). The slides were scanned by multi-spectral technology [Vectra 3.0(TM)/Polaris(TM)], and analyzed by Inform software.

U73122(Catalog Number: HY-13419), PP1(Catalog Number: HY-13804), Wortmannin (Catalog Number: HY-10197) and Gallein(Catalog Number: HY-D0254) were purchased from MCE (USA). PLC specific inhibitor U73122 at 50μM, Src-specific inhibitor PP1 at 20μM,PI3K-specific inhibitor Wortmannin at 50μM, Gβγ-specific inhibitor Gallein at 20μM.

Mice were infected and subsequently euthanized, after which the liver, lung and spleen tissues were recovered and fixed in 10% buffered formalin. These tissues were stained with hematoxylin and eosin and examined via OlympusBX53 microscopy. The histological assessment was scored as previously described (12).

Peripheral blood was taken from the orbital vein 3h after infection. Myeloperoxidase (MPO) protein level was determined using a MPO Assay Kit (Mlbio, Shanghai, China) according to the manufacturer’s instructions. We isolated neutrophils and adjusted the number of neutrophils in WT and Fpr2 group to be the same, then stimulated with GBS (MOI 10:1). Nets protein level was determined using a Nets Assay Kit (Mlbio, Shanghai, China) according to the manufacturer’s instructions.

We isolated neutrophils and adjusted the number of neutrophils in WT and Fpr2 group to be the same, then stimulated with GBS(MOI 10:1), then DCFH-DA was performed for labeling and the cells were detected using the FITC channel in upstream mode. The kit was purchased from China Biyuntian Biological Company.

Mouse peripheral blood was incubated with 2×10⁶ CFU/ml GBS for 1h at 37°C. 500ul of saponing was added, after which the cells were lysed on ice for 5min. 100ul serial dilutions were prepared and spread on THB plate and incubated overnight at 37°C for the colony count.

This research was performed in compliance with the guidelines of laboratory animal care and was approved in China. All experimental procedures were approved by the Animal Ethics Committee of the Academy of Military Medical Sciences.

All data were analyzed with GraphPad Prism (version 7.0, GraphPad Software, La Jolla, CA, USA). The differences between groups were calculated using an unpaired t-test and a log-rank test, respectively. Data are shown as means ± SD. Differences were considered statistically significant when P <0.05: *P < 0.05, **P < 0.01, and ***P < 0.001.

Fpr2 receptors can play a pro-inflammatory or anti-inflammatory role when it is activated. To examine the ability of Fpr2 to mediate host responses to GBS infection, live bacteria were injected i.v. and qPCR was detected in the peripheral blood. We found that Fpr2 expression levels were the highest ( Supplementary Figure 1A ). Then Fpr2 receptor was inhibited with Boc-2 and infected with GBS i.v. and we found that the mortality rate was higher than that of WT mice ( Supplementary Figure 1B ).

During the experiment, we have found in the preliminary experiments that when the mice are not infected with GBS, the organs of the Fpr2 knockout mice are not damaged, and the mice themselves do not have any defects or injuries. To further confirm the role of Fpr2 in GBS infections, we prepared Fpr2^-/- mice and infected GBS by i.p. Our results demonstrated that the survival rate of mice in the Fpr2^-/- group was less than 50% at 48h after infection, which was significantly lower than that in the WT group (P<0.05, Figure 1A ). Meanwhile, we obtained bacterial titers from peritoneal lavage fluid and blood samples from a separate set of Fpr2^-/- or WT mice. At 3h, 6h, and 24h, the number of bacteria in the peritoneal lavage fluid and blood in the Fpr2^-/- group was significantly higher than that in WT mice (P<0.05, Figures 1B, C ). These results demonstrated that the loss of an Fpr2 receptor decreased the host’s ability to control GBS infections.

Furthermore, we used intravenous injections to observe the clearance of GBS in blood and the invasion of GBS into the organs. Our results demonstrated that 48 h after infection, the survival rate of mice in the Fpr2^-/- group decreased significantly, to less than 50%. However, only one mouse in the WT group died at 24 h, while the other mice survived for more than seven days ( Figure 1D ). The results of the colony count indicated that GBS levels in the peripheral blood, spleen, liver, and kidney of Fpr2-/- mice were significantly higher than in WT mice after 24h ( Figure 1E, F ). However, we found no difference in the number of colonies between the two groups and no symptoms of meningitis in the brain tissue, which was repeated three times. This might be related to the short observation time.

There are several reports on liver damage caused by Fpr2 deletion (13–15). In this study, we analyzed pathological damage to the liver, spleen, and kidney. Our pathological analysis demonstrated that during GBS infections, Fpr2 loss damaged the spleen, and liver though damage to the liver was the most serious ( Figure 1G ). Briefly, the pathological analysis of liver showed pathological changes characteristic of severe tissue destruction, and large areas of piecemeal necrosis following GBS infection in Fpr2^-/- mice. No obvious pathological alterations in liver were found following GBS infection in WT group. For kidney, there was no glomerular, interstitial or vascular injury in the WT group, but these pathology characteristics became readily observable in the Fpr2^-/- group. Cellular proliferation and/or membrane thickness in glomeruli, and interstitial inflammatory cell infiltration were observed. For spleen, there was little difference between the two groups ( Figure 1G ). Fpr2^-/- liver pathological injuries were determined using a semi-quantitative score ( Table 2 ), and we found a significant difference in the pathological injury score for Fpr2^-/- (P< 0.001, Figure 1H ). These results are consistent with those of the colony counts.

Altogether, our results conclusively indicated that Fpr2^-/- mice were more susceptible to GBS infections and could not control the invasion of GBS from the blood to the organs. This confirms that Fpr2 is an important receptor for hosts and is needed to resist GBS infections.

We next sought to analyze the mechanisms underlying increased susceptibility to GBS infections in the absence of Fpr2. Because the Fpr2 receptor is chemotactic, we hypothesized that Fpr2^-/- mice reduced inflammatory cell recruitment at GBS infection sites. As seen from the liver pathological sections of GBS infections, the number of inflammatory cells in the liver tissue of Fpr2^-/- mice is significantly lower than in WT mice. To determine the type of reduced inflammatory cells, we firstly determined the number and types of cells present under basal conditions in the peripheral blood and found that they did not differ between Fpr2^-/- mice and WT mice (data not shown). The mice were then infected i.v. with GBS, after which the peripheral blood was obtained at predetermined times. Compared to monocytes/macrophages, B cells, and T cells, the recruitment of neutrophils in Fpr2^-/- mice significantly blunted in early stages 3 and 6 h after the infection ( Figure 2A ). We observed that damage to the liver is most serious after the GBS infection, indicating that neutrophils in the liver suffer similarly. Therefore, we used mIF to detect Ly6G⁺ cells in the liver and found a decrease in the number of neutrophils in the liver 3h after infection. Flow cytometry also demonstrated significant lower number of neutrophils in the liver of Fpr2^-/- mice ( Figure 2B ).

Data above suggested that the over-sensitivity to GBS infections seen in Fpr2^-/-mice is related to its reduced ability to recruit neutrophils to the infection site. Our results for GBS infection showed that a 80-90% mortality rate in both that were treated with anti-Ly6G (assessed 96 hours after GBS infection). Under the same experimental conditions, the mortality rate of all control animals was only ~20% ( Figure 2C ). Thus, our data supporting that mice lacking neutrophils are unable to control the growth of GBS, and Fpr2^-/- decreased host resistance to GBS infections, which is related to the impaired increase in neutrophils.

While we observed fewer neutrophils in Fpr2-knockout mice than in WT mice, it is unclear whether neutrophils levels are associated with low chemotaxis or the high number of living bacteria in Fpr2^-/- mice. The cytotoxin βH/C is an important virulence factor produced by GBS and can form pores on the surface of eukaryotic cells and lyse neutrophils (16). Does neutrophils impairment increase in Fpr2^-/- mice due to the high concentration of βH/C in the blood? To exclude this possibility, we used HK-GBS to avoid βH/C interference. The mice were injected with HK-GBS i.p. and 3h later, blood and the peritoneal lavage fluid were obtained to count the neutrophils number. Our results demonstrated that there was no significant difference in the chemotaxis of neutrophils in the blood, while neutrophils levels in Fpr2^-/- mice are significantly less than that in the WT group in the peritoneal lavage fluid ( Figures 3A, B ).

Thermal inactivation can affect the virulence of bacteria, leading us to block GBS with the βH/C inhibitor DPPC (16), which could block the cytotoxicity of βH/C ( Supplementary Figures 1E, F ) and perform an intravenous injection to observe the difference in neutrophils. After confirming that 250µg/mouse DPPC can effectively block the cytotoxicity of βH/C ( Figure 3C ), 2×10⁸ CFU/ml GBS was incubated with DPPC for 10 min and injected into mice i.v. Our results demonstrated that after eliminating βH/C interference, neutrophils levels were still reduced in the absence of Fpr2 ( Figure 3D ). The results of HK-GBS and DPPC blocking βH/C demonstrated that GBS slightly impacts neutrophils levels, but is not the primary reason.

The GBS infection of Fpr2^-/- mice will reduce the ability of neutrophils to recruit to the infected site, leading us to assess the factors that affect the chemotaxis of neutrophils. RNA-seq analysis was performed on neutrophils in Fpr2^-/- and WT mice infected with GBS. Of 1033 significantly differentially expressed genes (DEGs) in Fpr2^-/-, the transcripts of 239 (1.21%) genes were upregulated and the transcripts of 794 (4.01%) genes were downregulated ( Figures 4A, B ). Down-regulated gene ontology (GO) enrichment analysis demonstrated a significant defense response to other organisms and G-protein coupled receptor activity, as well as granulocyte migration/chemotaxis ( Figures 4C, D ). These were associated with the regulation of cell chemotaxis and the cytokine-mediated signaling pathway ( Figure 4D ). Additionally, KEGG pathway analysis demonstrated that the IL-17 signaling pathway, the PI3K-Akt signaling pathway, and the Cytokine-cytokine receptor interaction were all downregulated ( Figure 4E ). The data of RNA-seq indicates that the deletion of the Fpr2 receptor down-regulates many genes related to chemokine and chemokine receptors, which is related to hosting resistance to infection. The 17 infection- and acute-inflammation-related genes initially identified in our RNA-seq experiment were subsequently verified by qPCR ( Figure 4F ). Our results demonstrated that Fpr2 deletion significantly down-regulated the Saa3 expression, which is an important inflammation-related factor during the acute infection phase. Saa can induce several kinds of cytokines/chemokine (17). The classical chemokine CXCL1/2 of neutrophils and its receptor CXCR2 also decreased significantly in Fpr2^-/- mice. Another gene closely related to the host control of GBS infection, Il1b (10), was also significantly down-regulated in Fpr2^-/- mice. Other factors related to neutrophil chemotaxis and activation of host inflammation, including S100a8 (18), TREM1 (19), and the complement receptors C5ar1 (20, 21) and ITGAM (22), were significantly down-regulated in Fpr2^-/- mice ( Figure 4F ). Table 1 provides a list of the primers used in the present study.

Regarding genes related to chemotaxis and neutrophils activation, we detected the CXCL12/CXCR4 axis that mediates neutrophils homing and found a significant decrease in Fpr2^-/- mice ( Figure 4F ). These RNA-seq results indicate that Fpr2 deletion could affect neutrophils chemotaxis by altering the expression of CXCL1/2 and other chemokines.

Neutrophil recruitment in the blood and at infection sites is orchestrated by chemotactic cytokines/chemokines, including GM-CSF, CXCL1/2, TNF-α, and IL-6. Therefore, we next sought to ascertain whether neutrophil chemokine levels decreased in the blood of Fpr2-deficient mice inoculated with live GBS. At early time points, Fpr2^-/- mice showed a significant decrease in CXCL1, CXCL2, IL-1β, CCL2, GM-CSF, IL-6, and CCL3 levels at 1 hour and 3 hours after infection, and an increase in IL-22 and IL-23, but there is no change in TNF-α, as compared with WT mice ( Figures 5A, B ). To further confirm the effects of CXCL1/2 on the chemotactic neutrophils of Fpr2^-/-, we injected HK-GBS i.p. and found that the amount of CXCL1/2 in the WT group was significantly higher than in the Fpr2^-/- group. Compared with blood CXCL1/2, the content of CXCL1/2 in the abdominal cavity has significant difference between 3 and 6h ( Figure 5C ), which was consistent with changes in neutrophils content observed in the abdominal cavity.

The RNA-seq and luminex results of CXCL1/2 downregulation indicates that Fpr2 is needed for their production and CXCL1/2 play an important role in controlling GBS infections. Meanwhile Fpr2 absence caused an imbalance of neutrophils exudation from the bone marrow and homing.

The lack of neutrophils is not affected by GBS cytotoxins; meanwhile, chemotaxis detection demonstrates that CXCL1/2 in GBS-infected Fpr2^-/- mice slowed the increase of neutrophils. This indicates that, in Fpr2^-/- mice, the amount of neutrophils has relationship with the chemokine CXCL1/2. To demonstrate the role of CXCL1 and CXCL2 in mice infected with GBS, 100μg of anti-CXCL1 and anti-CXCL2 antibodies were injected i.p. for three hours before the infection. Orbital intravenous injection of 2×10⁸CFU/ml GBS demonstrated that pretreatment with anti-CXCL1/2 eliminated differences in the number of neutrophils in the blood between WT and Fpr2^-/-mice ( Figures 6A, B ). Compared with CXCL1, blocking CXCL2, a high-affinity ligand of CXCR2, reduced the proportion of neutrophils in the blood ( Figure 6B ). These results demonstrate that for GBS infections, Fpr2 primarily controls neutrophils levels via CXCL1/2.

CXCL1/2 can be produced by many cells including neutrophils. To determine which cell produced CXCL1/2, Ly6G antibody and clodronate liposomes were used to block neutrophils and monocytes/macrophages respectively ( Figure 2C , Supplementary Figure 1C ). We then used 2×10⁸CFU/ml GBS via orbital vein injection. Blood was then obtained for ELISA detection. Our results demonstrated that only blocking neutrophils inhibited CXCL1/2 expression which indicated that CXCL1/2 found in the blood during the early stages of GBS infection is primarily produced by neutrophils ( Figure 6C , Supplementary Figure 1D ).

The results of inhibitors showed that GBS may activate Gβγ in Fpr2 heterotrimer and PI3K pathway to produce CXCL1/2 ( Figure 6D ). The activation of Fpr2-PI3K, the release of CXCL1/2, and the subsequent chemotaxis of neutrophils all played an important role in host defense to GBS infections.

In addition to chemotaxis, Fprs also directly mediate neutrophil phagocytosis (23). Therefore, we examined what was capable of killing GBS, including the production of MPO, ROS, and NETs from WT and Fpr2^-/- mice. The peripheral blood of mice was added to GBS in the mid-log phase to induce infection, and the results demonstrated that the killing ability of Fpr2-knockout mice was weakened ( Figure 7A ). We detected the expression levels of MPO in the peripheral blood, the results showed that neutrophils of Fpr2-/- mice produced less MPO. We next detected the expression levels of ROS, and NETs in the peripheral blood neutrophils of the mice three hours after infection. These results demonstrated that the expression levels of these two were significantly reduced in Fpr2^-/- mice ( Figures 7B–D ) and that bactericidal functions in peripheral blood decreased in Fpr2 knockout mice. Therefore, the Fpr2 receptor not only affects chemotaxis but also affects other functions, including its ability to kill GBS infections.