SLAMF7 regulates the inflammatory response in macrophages during polymicrobial sepsis

Abstract

Uncontrolled inflammation occurred in sepsis results in multiple organ injuries and shock, which contributes to the death of patients with sepsis. However, the regulatory mechanisms that restrict excessive inflammation are still elusive. Here, we identified an Ig-like receptor called signaling lymphocyte activation molecular family 7 (SLAMF7) as a key suppressor of inflammation during sepsis. We found that the expression of SLAMF7 on monocytes/macrophages was significantly elevated in patients with sepsis and in septic mice. SLAMF7 attenuated TLR-dependent MAPK and NF-κB signaling activation in macrophages by cooperating with Src homology 2-containing inositol-5'‑phosphatase 1 (SHIP1). Furthermore, SLAMF7 interacted with SHIP1 and TNF receptor-associated factor 6 (TRAF6) to inhibit K63 ubiquitination of TRAF6. In addition, we found that tyrosine phosphorylation sites within the intracellular domain of SLAMF7 and the phosphatase domain of SHIP1 were indispensable for the interaction between SLAMF7, SHIP1, and TRAF6 and SLAMF7-mediated modulation of cytokine production. Finally, we demonstrated that SLAMF7 protected against lethal sepsis and endotoxemia by downregulating macrophage proinflammatory cytokines and suppressing inflammation-induced organ damage. Taken together, our findings reveal a negative regulatory role of SLAMF7 in polymicrobial sepsis, thus providing sights into the treatment of sepsis.

Keywords: Bacterial infections; Infectious disease; Inflammation; Macrophages.

Figure 1. SLAMF7 expression is associated with sepsis.

(A) Heatmap depicting mRNA expression levels of inflammation-related molecules in PBMCs from patients with sepsis (S1–S5) were determined by quantitative real-time PCR and compared with levels in healthy donors (H1–H5). The fold change for each gene was normalized to β-actin expression. (B) Expression of SLAMF members in human monocyte–derived macrophages after LPS stimulation for 12 hours at different doses. (C and D) Percentage of SLAMF7⁺ cells among CD11b⁺CD14⁺ or CD3⁺ cell subsets from human healthy donors (n = 81) and patients with sepsis (n = 83). Mo, monocytes. (E) Correlation between CRP and SLAMF7⁺ monocyte frequencies in patients with sepsis. (F) Protein levels of CRP in the serum of patients with sepsis were analyzed by ELISA before clinical treatment or 1, 3, 5, and 7 days after treatment. (G) The percentage of SLAMF7⁺ cells among CD14⁺ subsets was analyzed by flow cytometry before treatment and 1, 3, 5, and 7 days after treatment. Data represent the mean ± SEM from at least 3 independent experiments. **P < 0.01 and *** P < 0.001, by 2-tailed, unpaired Student’s t test (D), Spearman’s correlation (E), and 1-way ANOVA (F and G).

Figure 2. SLAMF7 negatively regulates the production of proinflammatory cytokines.

(A) mRNA levels of Slamf7 in RAW264.7 cells were examined by real-time PCR after stimulation with Pam3Csk4 (TLR1/-2 ligand), LPS (TLR4 ligand), R848 (TLR7/8 ligand), or poly (I:C) (TLR3 ligand) for 24 hours. (B and C) mRNA levels of Slamf7 were examined in RAW264.7 cells after LPS treatment at the indicated time point (B) and concentration (Con) (C). (D) RAW264.7 cells stably expressing SLAMF7 (RAW-SLAMF7) and control RAW-vector cells were constructed. mRNA levels of Tnf, Il1b, and Il6 in RAW-SLAMF7 cells versus RAW vector cells were analyzed 0, 6, and 12 hours after LPS stimulation. (E) Gene expression of Tnf, Il1b, and Il6 in BMDMs before treatment with rmSLAMF7 protein (1 μg/mL) versus control (0.9% NaCl), 6 hours after LPS stimulation. (F) Gene expression of Tnf, Il1b, and Il6 in WT and SLAMF7-KO BMDMs after LPS stimulation at 0, 6, and 12 hours. (G) Protein levels of TNF-α, IL-1β, and IL-6 in the supernatant of BMDMs transfected with SLAMF7 siRNA (Si-SLAMF7) or negative control siRNA (Si-Control), followed by LPS stimulation for 24 and 48 hours. Data represent the mean ± SEM from at least 3 independent experiments. **P < 0.01 and ***P < 0.001, by 2-tailed, unpaired Student’s t test (D, F, and G) and 1-way ANOVA (A–C and E).

Figure 3. SLAMF7 attenuates MAPK/NF-κB signaling pathways by activating SHIP1.

(A–C) Phosphorylation of AKT (A), MAPKs (ERK, JNK, and p38) (B), and IKKα/β (C) in RAW-SLAMF7 versus RAW-vector cells was examined by Western blotting after LPS stimulation at the indicated time points. (D) WT and SLAMF7-KO BMDMs were challenged with LPS for the indicated durations, followed by Western blotting to determine the phosphorylation of MAPKs and AKT. (E and F) Protein levels of the NF-κB p65 subunit in nuclei and the cytosolic fraction of LPS-treated RAW-vector versus RAW-SLAMF7 cells (E) and WT versus SLAMF7-KO BMDMs (F). (G and H) mRNA expression of Tnf, Il1b, and Il6 in WT versus SLAMF7-KO BMDMs before treatment with an inhibitor targeting PI3K/Akt (Ly294002) (G) and NF-κB (JSH23) (H), followed by LPS stimulation for 6 hours. (I) mRNA expression of Eat2, Sap, and Ship1 in BMDMs after stimulation with rmSLAMF7, followed by LPS stimulation. (J and K) Immunoblot (IB) analysis of p-SHIP1 in RAW-vector versus RAW-SLAMF7 cells (J) and WT versus SLAMF7-KO BMDMs (K) stimulated with LPS for 0, 15, 30, and 60 minutes. (L) mRNA expression of Tnf, II1b, and Il6 in RAW264.7 cells after pretransfection with SHIP1 siRNA, followed by LPS stimulation for 6 hours. Data represent the mean ± SEM from at least 3 independent experiments. ns, not significant. *P < 0.05, **P < 0.01, and ***P < 0.001, by 2-tailed, unpaired Student’s t test (I) and 1-way ANOVA (G, H, and L).

Figure 4. SLAMF7 cooperates with SHIP1 to inhibit TRAF6 K63 ubiquitination.

(A and B) IP using anti-Flag or anti-HA antibodies from lysate of HEK 293T cells transfected with Flag-tagged SLAMF7 alone, or HA-tagged SHIP1. (C) Immunoassay of lysate of RAW264.7 cells stimulated with LPS, followed by IP with IgG or anti-SLAMF7 and IB analysis with anti-TRAF6 or anti-SHIP1. (D and E) IP using anti-Flag or anti-HA antibodies from lysate of HEK 293T cells transfected with HA-tagged TRAF6 and Flag-tagged SLAMF7 (D) or HA-tagged SHIP1 and Flag-tagged TRAF6 (E). (F) Confocal microscopy of HEK 293T cells cotransfected with Flag-tagged SLAMF7 and HA-tagged SHIP1 (top row) or HA-tagged TRAF6 (bottom row). White arrows indicate colocalization. Scale bars: 10 μm. (G) IB of TRAF6-Ubs precipitated with anti-HA antibodies from lysates of HEK 293T cells transfected with Flag-tagged TRAF6, HA-tagged Ubs, and Myc-tagged SHIP1. (H) IB analysis of TRAF6 ubiquitination from precipitation of lysates of 293T cells transfected with Flag-tagged TRAF6, HA-tagged ubiquitin with or without SLAMF7 and SHIP1. (I) IB of lysates of TRAF6 ubiquitination in HEK 293T cells transfected with HA-tagged ubiquitin, HA-tagged K48 ubiquitin, HA-tagged K63 ubiquitin, HA-tagged K48R ubiquitin, and HA-tagged K63 ubiquitin. (J) IB of TRAF6 ubiquitination of LPS-stimulated macrophages transfected with control siRNA or SLAMF7 siRNA, followed by LPS stimulation for 30 minutes, with or without MG132 treatment. (K) Relative mRNA expression of Tnf, Il1b, Il6, and Il10 after transfection with a constructed TRAF6 plasmid or control plasmid. Data represent the mean ± SEM from at least 3 independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001, by 1-way ANOVA (K).

Figure 5. Binding of SLAMF7 to SHIP1 by tyrosine and the EEP domain is required for negative regulation of TLR responses.

(A) Functional domains of SLAMF and SHIP1. (B and C) SLAMF7 plasmids lacking extracellular (Δextra), transmembrane (Δtrans), or intracellular (Δintra) domains were transfected into HEK 293T cells with SHIP1 (B) or TRAF6 (C) plasmid to determine the interactions. (D and E) IP was performed in HEK 293T cells cotransfected with SLAMF7 (D) or TRAF6 (E) plasmids as well as with full-length SHIP1 or SHIP1 plasmids lacking the SH2 domain (ΔSH2), the EEP domain (ΔEEP), and the P-rich domain (ΔP rich). (F) IB assay of endogenous ubiquitination of TRAF6 using anti-HA from lysate immunoprecipitated with anti-Flag from HEK 293T cells transfected with HA-tagged ubiquitin (Ubs-HA) or Flag-tagged TRAF6 (TRAF6-Flag), plus SLAMF7 alone or SLAMF7 plus EEP-HA or SLAMF7 plus ΔEEP-HA. (G and H) Tyrosine sites at amino acids 261, 266, and 281 of the SLAMF7 amino acid sequence, ignoring signal peptide were manually mutated to phenylalanine (Y→F), respectively. IP was performed in HEK 293T cells after the transfection of SHIP1 (G) or TRAF6 (H) plasmid with full-length SLAMF7 or SLAMF7 tyrosine mutations. (I) Tnf, Il1b, and Il6 mRNA levels in RAW264.7 cells transfected with Y261F-, Y266F-, or Y281F-SLAMF7 plasmids, followed by LPS stimulation for 6 hours. Data represent the mean ± SEM from at least 3 independent experiments. *P < 0.05 and **P < 0.01, by 1-way ANOVA (I).

Figure 6. SLAMF7 protects against sepsis by inhibiting inflammation and lung injury.

A sepsis model was established in C57BL/6 (B6) mice by i.p. injection with LPS (25 mg/kg) or P. aeruginosa (PA) (2 × 10⁷ CFU/kg) or by CLP surgery. (A) Percentage of SLAMF7⁺ macrophages in the PL was analyzed by flow cytometry. Mɸ, macrophages. (B–F) Mice were i.p. injected with rmSLAMF7 or vehicle control (0.9% NaCl), followed by the establishment of LPS, P. aeruginosa, or CLP sepsis 6 hours later. (B) Schematic of rmSLAMF7 administration prior to sepsis. (C–E) Survival rates of mice challenged with LPS (C), P. aeruginosa (D), or CLP (E). (F) H&E staining of lung sections was examined in rmSLAMF7 and vehicle control–treated mice 24 hours after CLP. Scale bars: 200 μm. (G–K) Mice were i.p. injected with LPS, P. aeruginosa, or CLP to establish sepsis models. Six hours after injection, rmSLAMF7 or vehicle control was i.p. administrated. (G) Schematic of rmSLAMF7 administration after sepsis induction. (H–J) Survival rates of mice after LPS (H), P. aeruginosa (I), or CLP (J) challenge. (K) H&E staining was performed 24 hours later to evaluate lung injury. Scale bars: 200 μm. Data represent the mean ± SEM from at least 3 independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001, by 1-way ANOVA (A), log-rank test (C–E and H–J), and 2-tailed, unpaired Student’s t test (F and K).

Figure 7. SLAMF7 deficiency exacerbates sepsis by aggravating inflammation and lung damage.

WT mice and SLAMF7-KO mice were subjected to LPS- (25 mg/kg), P. aeruginosa– (2 × 10⁷ CFU/kg), or CLP-induced sepsis. (A–C) Survival curves were calculated after LPS (A), P. aeruginosa (B), or CLP (C) challenge. (D) H&E staining of lung sections was examined 24 hours after CLP. Scale bars: 200 μm. (E–H) LPS- (25 mg/kg), P. aeruginosa– (2 × 10⁷ CFU/kg), or CLP-induced sepsis models were established in control (SLAMF7^fl/fl) and SLAMF7 conditional-KO mice (SLAMF7^fl/fl Lyz2^Cre). (E–G) Survival rates of mice after LPS (E) or P. aeruginosa (F) injection or CLP surgery (G). (H) Twenty-four hours after CLP surgery, H&E staining was performed to assess injury and inflammatory infiltration into lung tissues. Scale bars: 200 μm. Data represent the mean ± SEM and represent 3 individual experiments *P < 0.05, **P < 0.01, and ***P < 0.001, by log-rank test (A–C and E–G) and 2-tailed, unpaired Student’s t test (D and H).