Intravenous immunoglobulins modulate neutrophil activation and vascular injury through FcγRIII and SHP-1

Abstract

Rationale: Intravascular neutrophil recruitment and activation are key pathogenic factors that contribute to vascular injury. Intravenous immunoglobulin (IVIG) has been shown to have a beneficial effect in systemic inflammatory disorders; however, the mechanisms underlying IVIG's inhibitory effect on neutrophil recruitment and activation are not understood.

Objective: We studied the mechanisms by which IVIG exerts protection from neutrophil-mediated acute vascular injury.

Methods and results: We examined neutrophil behavior in response to IVIG in vivo, using real-time intravital microscopy. We found that an antibody that blocks both FcγRIII and its inhibitory receptor counterpart, FcγRIIB, abrogated the inhibitory effect of IVIG on leukocyte recruitment and heterotypic red blood cell (RBC) interactions with adherent leukocytes in wild-type mice. In the context of sickle cell disease, the blockade of both FcγRIIB and III abrogated the protective effect of IVIG on acute vaso-occlusive crisis caused by neutrophil recruitment and activation. Analysis of FcγRIIB- and FcγRIII-deficient mice revealed the predominant expression of FcγRIII on circulating neutrophils. FcγRIII mediated IVIG-triggered inhibition of leukocyte recruitment, circulating RBC capture, and enhanced Mac-1 activity, whereas FcγRIIB was dispensable. In addition, FcγRIII-induced IVIG anti-inflammatory activity in neutrophils was mediated by recruitment of Src homology 2 (SH2)-containing tyrosine phosphatase-1 (SHP-1). Indeed, the protective effect of IVIG on leukocyte recruitment and activation was abrogated in SHP-1-mutant mice.

Conclusions: FcγRIII, a classic activating receptor, has an unexpected inhibitory role on neutrophil adhesion and activation via recruitment of SHP-1 in response to IVIG. Our results identify SHP-1 as a therapeutic target in neutrophil-mediated vascular injury.

Figures

Figure 1. FcγRIIB/III mediate IVIG-induced inhibition of leukocyte recruitment and RBC interactions

(A) Experimental scheme. WT mice (n= 11–14 per group) were injected with either mAb FcγRIIB/III or isotype rat IgG2b (1 mg/kg) followed by IVIG or control albumin administration (800 mg/kg) 3 h after administration of TNF-α (0.5 ug). Leukocyte behaviors were analyzed in cremasteric venules for 1 h. (B) Blood from WT mice was collected after intravital microscopy experiment and surface expression of FcγRIIB/III on the neutrophil population gated on the basis of side and forward scatter properties was examined by flow cytometry after PE-conjugated anti-FcγRIIB/III staining. (C) Numbers of circulating leukocytes. (D) The percentages of monocytes, neutrophils and lymphocytes. (E) Adherent leukocytes in venules. (F) Number of circulating RBC-adherent leukocyte interactions per min. Bars represent mean ± SEM. **P<0.01, ***P<0.001 vs albumin.

Figure 2. FcγRIII, but not FcγRIIB, is expressed on neutrophils

(A) Circulating leukocytes from WT mice were stained for FcγRIIIB/III, CD115, CD3 expression after RBC lysis. Neutrophils (blue) were gated as CD115low, monocytes (red) as CD115hi, T cell population (black) as CD3pos. (B) Blood was collected from WT (green, n=3) and Fcgr3−/− (blue, n=6) mice and then stained for Gr-1, CD115, and FcγRIIB/III expression. FcγRIIB/III expression was analyzed by gating on the neutrophil population with Gr-1hi and CD115low and monocyte population with Gr-1low-hi and CD115hi. Representative histogram of FcγRIIB/III expression and control isotype (gray, left panel) and quantification of the geometric mean of fluorescence (GMF) (right panel). (C) Representative expression of FcγRIIIB/III (left panel) on neutrophil population (Gr-1hi/ CD115low) from TNF-α treated mice deficient in the FcγRIB, FcγRIII or control WT mice (n= 2–3 per group) 1 h after administration of IVIG or control albumin. Gray histograms represent isotype control. Quantification of FcγRIIIB/III expression levels from WT, Fcgr2−/− and Fcgr3−/− mice (right panel).

Figure 3. FcγRIII mediates IVIG-induced inhibition of leukocyte adhesion and Mac-1 activity

IVIG or control albumin (800 mg/kg) was administrated 3 h after TNF-α injection (0.5 µg) in WT, Fcgr2b−/−, and Fcgr3−/−mice (n= 5-7 per group). Cremasteric venules were analyzed by intravital microscopy for 1 h. (A) Adherent leukocytes in venules. (B) Number of circulating RBC-adherent leukocyte interactions per min. Bars represent mean ± SEM. *P<0.05, **P<0.01 vs albumin. (C) Representative images of fluosphere bound to leukocytes in venules from WT mice. Images were acquired in the brightfield and FITC channel. Asterisks represent RBCs that interact with adherent leukocytes and arrows indicate the direction of flow. Scale bar, 20 µm. (D) Binding of albumin-coated fluospheres to leukocytes in WT, Fcgr2b−/−, and Fcgr3−/− mice. Mice prepared for intravital microscopy were intravenously injected with 109 albumin-coated fluosphere 10 min after administration of either IVIG or control albumin. Each dot represents the average number of fluospheres bound per leukocyte within individual venules (n= 35–45 venules from four mice per group). *P<0.05, ***P<0.001, Mann-Whitney test.

Figure 4. Abrogation of IVIG-induced anti-inflammatory activity in mev/mev mice

(A) SHP-1 association with FcγRIIB/III in response to IVIG in neutrophils. Bone marrow neutrophils (BMNs) isolated from control WT mice (n= 3) were incubated with IVIG or albumin (6.7 mg/ml) at 37 °C for 15 min, and then lysates were prepared. Lysates were immunoprecipitated with anti-FcγRIIB/III or control isotype rat IgG2b followed by immunoblotting (IB) with anti-SHP-1/2 ab. (B) Adherent leukocytes in venules. (C) Number of circulating RBC-adherent leukocyte interactions per min. Bars represent mean ± SEM. *P<0.05, ***P<0.001 vs albumin. (D) Binding of albumin-coated flurosphere to leukocytes10 min after either IVIG or control albumin administration in control chimeric WT or mev/mev mice. Each dot represents the average number of fluospheres bound per leukocyte within individual venules (n= 20–28 venules from 4 mice per group). ***P<0.001 vs albumin, Mann-Whitney test.

Figure 5. Blockade of FcγRIIB and III abrogates the protective actions of IVIG against acute vaso-occlusion in SCD mice

(A) Experimental scheme. Three hours after administration of TNF-α (0.5 ug), SCD mice (n= 8 per group) were injected with either anti-FcγRIIB/III monoclonal antibody (mAb) or isotype rat IgG2b, (1 mg/kg) before administration of IVIG or an equivalent volume of human albumin (800 mg/kg). (B) Blood from SCD mice was collected after intravital experiment and surface expression of FcγRIIB/III on the neutrophils was examined with PE-conjugated anti-FcγRIIB/III mAb by flow cytometry. (C) Adherent leukocytes in venules. (D) Number of circulating sickle RBC-adherent leukocyte interactions per min. (E) Blood flow rates. Bars represent mean ± SEM. *P<0.05, **P<0.01, ***P<0.001 vs albumin. (F) Representative images of each group after IVIG or control albumin administration showing leukocyte recruitment and heterotypic interactions. Scale bars, 20 µm. (G) The Kaplan-Meier survival curves for individual SCD mice. P=0.03, log-rank test, IVIG vs albumin in IgG2b-treated group. P=0.93, log-rank test, IVIG vs albumin in anti-FcγRIIB/III mAb-treated group.