Abstract
FcγRIIb (CD32B, Online Mendelian Inheritance in Man 604590), an IgG FcR with a tyrosine-based inhibitory motif, plays a critical role in the balance of tolerance and autoimmunity in murine models. However, the high degree of homology between FcγRIIb and FcγRIIa in humans and the lack of specific Abs to differentiate them have hampered study of the normal expression profile of FcγRIIb and its potential dysregulation in autoimmune diseases such as systemic lupus erythematosus (SLE). Using our newly developed anti-FcγRIIb mAb 4F5 which does not react with FcγRIIa, we found that FcγRIIb is expressed on the cell surface of circulating B lymphocytes, monocytes, neutrophils, myeloid dendritic cells (DCs), and at very low levels on plasmacytoid DCs from some donors. Normal donors with the less frequent 2B.4 promoter haplotype have higher FcγRIIb expression on monocytes, neutrophils, and myeloid DCs similar to that reported for B lymphocytes, indicating that FcγRIIb expression on both myeloid and lymphoid cells is regulated by the naturally occurring regulatory single nucleotide polymorphisms in the FCGR2B promoter. FcγRIIb expression in normal controls is up-regulated on memory B lymphocytes compared with naive B lymphocytes. In contrast, in active SLE, FcγRIIb is significantly down-regulated on both memory and plasma B lymphocytes compared with naive and memory/plasma B lymphocytes from normals. Similar down-regulation of FcγRIIb on myeloid-lineage cells in SLE was not seen. Our studies demonstrate the constitutive regulation of FcγRIIb by natural gene polymorphisms and the acquired dysregulation in SLE autoimmunity, which may identify opportunities for using this receptor as a therapeutic target.
The only inhibitory IgG FcR in the classical FcR family (1, 2), FcγRIIb (CD32B) has an ITIM in its cytoplasmic domain and counterbalances tyrosine-based activation signals in a variety of cells (3). For example, upon coligation of FcγRIIb with the BCR by IgG immune complexes, FcγRIIb recruits and activates the SHIP and negatively regulates B cell activation and proliferation, thereby providing a critical mechanism for the feedback inhibition of IgG production (4). On monocytes and macrophages, FcγRIIb down-regulates FcγRIIa (CD32A)- and FcγRIIIa (CD16A)-mediated phagocytosis when coligated with those activating receptors (2, 5, 6). FcγRIIb also down-modulates mast cell activation by the high-affinity IgE FcR (7, 8).
On dendritic cells (DCs),3 however, a dual role of FcγRIIb has been suggested. Although interfering with the signals mediated by FcγRIIb may lead to enhanced myeloid DC (mDC) maturation and immunogenicity (9, 10, 11), several studies suggest that FcγRIIb on DCs promotes T-independent humoral responses by presenting native Ag to B cells, and that FcγRIIb on follicular DCs facilitates B cell recall responses (12, 13). Thus, FcγRIIb may have multiple effects on immune function, each of which points to an important role in immune regulation.
Indeed, the strength of FcγRIIb signaling is critical for the induction of immune tolerance or autoimmunity. Mice deficient in FCGR2B with no FcγRIIb expression have elevated serum Ig levels and enhanced anaphylactic responses (14). On a susceptible genetic background, mice deficient in FCGR2B develop anti-nuclear autoantibodies, glomerulonephritis, and other lupus-like symptoms (15). Consistent with these studies, natural polymorphisms in the mouse FCGR2B gene promoter, which lead to reduced receptor expression, have been identified in many autoimmune-prone mouse strains (16, 17, 18). Interestingly, 40% more FcγRIIb expression on B cells by retroviral delivery is sufficient to restore tolerance and prevent autoimmunity in those mouse strains (19). Thus, in mice, FCGR2B has been proposed as a distal peripheral checkpoint gene for lupus-like autoimmunity.
In humans, the FCGR2B gene is located on Chr.1q23, a systemic lupus erythematosus (SLE) susceptibility locus confirmed by several independent genome-wide scans (20). Polymorphisms in the human FCGR2B gene are associated with the SLE phenotype (21, 22, 23, 24, 25, 26). We have recently demonstrated that polymorphisms in the regulatory region of FCGR2B form two major haplotypes and that the less frequent 2B.4 haplotype is significantly overrepresented in SLE patients (22). A nonsynonymous polymorphism in the transmembrane domain of FcγRIIb is also enriched in SLE patients from Japanese and other Asian populations (23, 26, 27). However, due to the lack of specific Abs to distinguish human FcγRIIb and its highly homologous family member FcγRIIa (which is absent in mice), there is no study to date to examine the expression and regulation of FcγRIIb in normal donors and in SLE autoimmunity.
To address this gap in knowledge, we have recently developed a novel mAb that differentiates the cell surface FcγRIIb from FcγRIIa. Using this unique reagent, we have discovered that, in addition to B cells, FcγRIIb is also expressed on monocytes, neutrophils, mDCs, and at very low levels on plasmacytoid DCs (pDCs) from a small percentage of donors. Normal donors with the 2B.4 promoter haplotype have higher FcγRIIb expression on myeloid-lineage monocytes, neutrophils, and mDCs compared with donors with the homozygous 2B.1 promoter haplotype, indicating that these promoter variants affect receptor expression in both the lymphoid and myeloid series. Interestingly, while FcγRIIb is decreased on memory and plasma B cells from SLE patients, expression on mDCs is not changed, suggesting that there may be differential regulation of expression by humoral factors. SLE is a prototype autoimmune disease characterized by overproduction of autoantibodies and tissue deposition of autoimmune complexes (28, 29). Altered regulation of FcγRIIb expression, by both genetic and disease activity related factors, could contribute significantly to the B cell hyperactivity in SLE.
Materials and Methods
Donors
SLE patients and normal controls were recruited from the University of Alabama at Birmingham-based DISCOVERY and CASSLE cohorts. All SLE patients fulfilled the revised American College of Rheumatology criteria for SLE (30) and were assessed for disease activity using the SLE Disease Activity Index (SLEDAI) (31). The human studies were reviewed and approved by the Institution Review Board, and all donors provided written informed consent.
Cell culture
PBMCs were purified from whole blood by density gradient centrifugation using Ficoll-Hypaque. PBMCs were cultured in 12-well plates at 2 × 106 cells/ml in RPMI 1640 medium supplemented with 10% human AB serum (Krackeler Scientific), glutamine, HEPES, and penicillin/streptomycin for 20–24 h in the absence or presence of indicated cytokines before flow cytometry analysis. IL-3, GM-CSF, IL-4, and IL-10 were purchased from PeproTech and used at 100 ng/ml.
Antibodies
The IV.3 and 32.2 hybridomas were purchased from American Type Culture Collection. Their fragments were prepared by Rockland Immunochemicals. The AT10 hybridoma was a gift from Dr. P. Guyre (Dartmouth Medical School, Hanover, NH) (32). mAb 4F5, a murine IgG1, was generated in BALB/c mice using the Escherichia coli-produced recombinant extracellular domain (EC) of human FcγRIIb as an immunogen. Approximately 600 mAb clones were screened by ELISA for positive reactivity with the EC of FcγRIIb and negative reactivity with the EC of FcγRIIa-R (FcγRIIa is polymorphic with arg or his at the extracellular amino acid 131 and FcγRIIb is monomorphic with arg). The clones were further screened by flow cytometry on FcγRIIb, FcγRIIa-R, or FcγRIIa-H stable transfectants for anti-FcγRIIb Abs. mAb 4F5 were purified from ascites fluid using Immunopure Protein A mouse IgG1 Purification kit (Pierce Biotechnology). All mAbs were conjugated with Alexa 488 fluorescence dye using an Alexa488 Protein Labeling kit (Invitrogen Life Technologies-Molecular Probes). The labeling efficiency of 4F5, IV.3, 3G8, AT10, and 32.2, determined following the manufacturer’s instruction, was ∼3:1 (three molecules of dyes per protein molecule). The F(ab′)2 of mAb 4F5 were made using the ImmunoPure Fab Preparation kit (Pierce Biotechnology).
CD19-allophycocyanin and CD14-Tri-Color mAbs were purchased from Caltag Laboratories. CD27-allophycocyanin mAb was purchased from eBioscience. BDCA1-allophycocyanin and BDCA2-allophycocyanin mAbs were purchased from Miltenyi Biotec. The isotype control mIgG and mIgG F(ab′)2 and F(ab′)2 goat anti-mouse IgG F(ab′)2 were purchased from Jackson ImmunoResearch Laboratories. Goat polyclonal Abs specific for the cytoplasmic domain of FcγRIIa/c were purchased from Santa Cruz Biotechnology.
Flow cytometry analysis
PBMCs or whole blood were incubated with the indicated mAbs for 45 min on ice. The cells were washed with 3 ml of ice-cold PBS plus 0.5% BSA and 0.02% NaN3. The RBC were lysed by incubation with 1.5 ml of 1× FACS Lysing Solution (BD Biosciences-BD Pharmingen) at room temperature for 15 min. The cells were washed with PBS and resuspended in PBS plus 1% paraformaldehyde for flow cytometry analysis. Flow cytometry analysis of stable transfectants was performed similarly without the RBC-lysing step.
Stable cell lines
The parental FcR-deficient A20-IIA1.6 murine B cell line was provided by Dr. T. Wade (Dartmouth Medical Center). The FcγRIIb-IIA1.6 stable transfectants were established previously (24). The FcγRIIa cDNAs (with the “R” or “H” allele at amino acid 131) were subcloned into the pcDNA3 (Invitrogen Life Technologies) expression vector through EcoRI and NotI sites. The plasmids were transfected into A20-IIA1.6 cells using Fugene 6 transfection reagents (Roche Applied Science), and the FcγRIIa-R and FcγRIIa-H stable transfectants were selected in the presence of 1 mg/ml G418. FcγRIIc stable transfectants were established in a similar way. FcγRIIc cDNA was subcloned into pcDNA3 vector and the resulting plasmids were transfected into A20IIA1.6 cells.
RT-PCR
The desired leukocyte subsets were purified by FACS sorting after staining with their cell surface markers. pDCs were first enriched by BDCA4 microbeads using the Plasmacytoid Dendritic Cell Isolation kit (Miltenyi Biotec) and then subjected to cell sorting for BDCA2-allophycocyanin-positive cells. The purity of sorted cells was over 96% by subsequent flow cytometry analysis. The total RNA was prepared from the same number of sorted cells using TRIzol Reagents (Invitrogen Life Technologies). The RT-PCR for FCGR2B, FCGR2A, and GAPDH genes was performed using the SuperScript III One-Step RT-PCR kit (Invitrogen Life Technologies) following the manufacturer’s protocol. The FcγRIIb-specific sense primer is: 5′-TGTCCAAGCTCCCAACTCTTCACC-3′; the antisense primer is: 5′-GTGTTCTCAGCCCCAACTTTG-3′. The FcγRIIa-specific sense primer is: 5′-CACTGTCCAAGTGCCCAGCAT-3′; the antisense primer is: 5′-TTTATCATCGTCAGTAGGTGCCC-3′. The RT-PCR conditions were as follows: 56°C for 30 min, 95°C for 2 min, and 30 cycles of denaturing at 95°C for 15 s, annealing at 56°C for 30 s, and extension at 68°C for 40 s with a final extension at 68°C for 7 min.
Preparation of whole cell lysate and immunoprecipitation assay
Cells were lysed with whole cell lysis buffer at 60 μl/1 × 106 cells as previously described (24). The samples were vortexed for 10 s and incubated on ice for 30 min with a brief vortexing every 10 min. The samples were then centrifuged at 15,000 rpm at 4°C for 15 min and the supernatant was collected. For immunoprecipitation, mAbs 4F5, IV.3, or AT-10 were added to the whole cell lysate and incubated at 4°C for 2 h with mixing. Protein G Sepharose beads were added to each sample and the samples were further incubated at 4°C for 1 h with mixing. The beads were then washed four times with whole cell lysis buffer and the immunoprecipitates were subjected to Western blot analysis.
GST pull-down experiments
The EC FcγRIIb or FcγRIIa (R or H allele) was subcloned into the mammalian expression vector pEBG in frame for GST through EcoRI and NotI sites. The plasmids were transiently transfected into 293 T cells using Fugene 6 transfection reagents (Roche). The GST fusion proteins were purified from 293 T cells at 30–48 h after transfection using glutathione Sepharose beads. The purified fusion proteins were incubated with mAbs 4F5 or AT10 on ice for 1 h. After extensive washing, the bound mAbs were subjected to reduced SDS-PAGE electrophoresis and Western blot analysis using HRP-linked F(ab′)2 goat anti-mouse IgG (H+L).
Measurement of change in [Ca2+]i
Changes in intracellular [Ca2+]i induced by cross-linking of surface Igs on IIA1.6 transfectants and U937 cells were determined using an SLM 8000 spectrofluorometer monitoring the simultaneous 405/490 nm fluorescence emission ratio of the calcium-binding indo-1 fluorophore, as previously described (33). Cells (10 × 106/ml) were loaded with 5 μM indo-1-AM at 37°C for 40 min and preincubated with 5 μg/ml indicated F(ab′)2. The cells were washed and cross-linked with 20 μg/ml F(ab′)2 goat anti-mouse IgG at the 60 s time point and the data were collected for 300 s.
Statistical analysis
Mean fluorescence intensity (MFI) of mAb 4F5 staining (for the expression levels of FcγRIIb) on naive B lymphocytes was compared with that on memory or plasma B lymphocytes from the same participants by the paired Student’s t test, two-tailed. The unpaired Student’s t test, two tailed, was used to compare the MFI of mAb 4F5 staining on naive, memory, or plasma B lymphocytes between normal controls and SLE patients. The range of MFI values for 4F5 staining on myeloid-lineage cells was skewed toward low values (Lilliefors test for normality), and therefore, the Mann-Whitney U test was used to compare the expression levels of FcγRIIb on monocytes, neutrophils, or mDCs between 2B.1 and 2B.4 donors and between normal controls and SLE patients. The Spearman rank correlation test was used to assess the degree of correlation of the expression levels of FcγRIIb on monocytes, neutrophils, and mDCs.
Results
mAb 4F5 reacts with FcγRIIb but not FcγRIIa
The EC of human FcγRIIb and FcγRIIa are ∼95% identical due to gene duplication and recombination events in the evolution of FcR cluster (34, 35, 36). Available mAbs that react with the cell surface FcγRII (CD32) include pan-reactive Abs, anti-FcγRIIa Abs, and allele-specific Abs reacting with the arg allele of FcγRIIa and with FcγRIIb (9, 32, 37, 38, 39). Studies of FcγRIIb expression on primary cells, however, have been greatly hampered by the lack of specific mAbs. To generate mAb that differentiates FcγRIIb and FcγRIIa, the E. coli-produced recombinant EC of human FcγRIIb was used as an immunogen. Approximately 600 mAb clones were first screened by ELISA and further screened by flow cytometry of stably transfected IIA1.6 cells for positive reactivity with FcγRIIb and absence of reactivity with the FcγRIIa-R or FcγRIIa-H (FcγRIIa is polymorphic with arg or his at amino acid 131 in its EC while FcγRIIb is monomorphic with arg at that site). Clone 4F5 reacted with the cell surface FcγRIIb, but not FcγRIIa-R or FcγRIIa-H, by flow cytometry analysis (Fig. 1,A). Similar expression levels of FcγRIIb and FcγRIIa on the stable transfectants were confirmed with the well-characterized pan-FcγRII mAb AT10 (Fig. 1 A). Clone 4F5 does react with stably transfected FcγRIIc (data not shown), a highly homologous gene which has been reported on NK cells from a small percentage of donors (40).
As shown by Western blot in Fig. 1,B, mAb 4F5 immunoprecipitated FcγRIIb but not FcγRIIa-R or FcγRIIa-H from the whole cell lysate of stable transfectants (Fig. 1,B, lanes 2, 5, and 8). As a comparison, mAb IV.3 preferentially immunoprecipitated FcγRIIa-R and FcγRIIa-H (Fig. 1,B, lanes 6 and 9), with a small amount of FcγRIIb also evident (Fig. 1,B, lane 3). To further establish the specific interaction between mAb 4F5 and the EC of FcγRIIb, we used a GST-pull down paradigm. GST proteins fused with the EC of FcγRIIa or FcγRIIb were incubated with mAb 4F5 or AT-10. The Abs immunoprecipitated by the GST fusion proteins were visualized by goat anti-mouse IgG Western blot analysis. The GST-FcγRIIb EC but not FcγRIIa-R/H EC fusion proteins pulled down mAb 4F5 (Fig. 1,C, lanes 1–3). As a positive control, GST-IIb and GST-IIa R/H fusion proteins pulled down the pan-mAb AT10 (Fig. 1 C, lanes 4–6). In summary, our data demonstrated that mAb 4F5 reacts with native FcγRIIb but not with FcγRIIa (R or H) on the cell surface, in the cell lysate, and as recombinant proteins.
We also tested the ability of mAb 4F5 to recognize denatured FcγRIIb in Western blot analysis. Pan-mAb AT10 or its isotype control mIgG1 were used to immunoprecipitate FcγRII from the whole cell lysate of A20IIA1.6 stable transfectants. The immunoprecipitates were subjected to Western blot analysis using either mAb 4F5 or polyclonal anti-FcγRIIa/c as blotting Abs. mAb 4F5 only blotted FcγRIIb immunoprecipated by mAb AT10 or directly from the whole cell lysate of FcγRIIb-transfectants (Fig. 1,D, upper panel, lanes 2 and 7). Western blots using polyclonal anti-FcγRIIa/c demonstrated that there were comparable amounts of FcγRIIa-R or FcγRIIa-H proteins immunoprecipitated by mAb AT10 or directly from the whole cell lysate of FcγRIIa transfectants (Fig. 1 D, upper panel, lanes 4, 6, 8, and 9). Together, our data suggest that mAb 4F5 also reacts with denatured FcγRIIb but not FcγRIIa in Western blot analysis.
Cross-linking of FcγRIIb by mAb 4F5 induces an inhibitory signal for BCR and FcγRIa-mediated intracellular Ca2+ flux ([Ca2+]i)
FcγRIIb negatively regulates the activation signals mediated by BCR and by activating FcRs. We next examined the ability of mAb 4F5 to modulate the Ca2+ responses induced by BCR and the activating FcγRIa. Cross-linking of mouse IgG (mIgG)-BCR alone by F(ab′)2 goat anti-mIgG induced a brisk rise in [Ca2+]i in the IIA1.6-FcγRIIb stable transfectants (Fig. 2,A). Coengagement of human FcγRIIb and mIgG-BCR by preincubation with mAb 4F5 F(ab′)2 significantly decreased the change in [Ca2+]i (Fig. 2,A). Similarly, in human monocytic U937 cells, cross-linking of FcγRIa by preincubation with mAb 32.2 F(ab′)2 induced a rise in [Ca2+]i (Fig. 2,B). However, coengagement of FcγRIIb and FcγRIa by preincubation with mAbs 4F5 F(ab′)2 and 32.2 F(ab′)2 induced a significant decrease in the Ca2+ responses (Fig. 2,B). As a control, coengagement of the activating FcγRIIa and FcγRIa by preincubation with mAbs IV.3 Fab and 32.2 F(ab′)2 induced more rapid and stronger Ca2+ responses (Fig. 2 B). Collectively, our data suggest that co-cross-linking of FcγRIIb by mAb 4F5 induces an inhibitory signal for the Ca2+ responses mediated by BCR and the activating FcγRIa.
mAb 4F5 partially competes with heat-aggregated human IgG for FcγRIIb binding
To further investigate the binding property of mAb 4F5 on FcγRIIb, we performed cross-competition binding experiments with 4F5 and heat-aggregated human IgG (ahIgG) that mimics the ligand binding of FcRs. FcγRIIb stable transfectants were preincubated with 4F5 F(ab′)2, IV.3 Fab, or isotype control mIgG1 F(ab′)2 then stained with Alexa 488-conjugated ahIgG. As expected, flow cytometry analysis demonstrated that IV.3 Fab did not affect the binding of ahIgG to FcγRIIb transfectants, but 4F5 F(ab′)2 did partially block the binding of ahIgG to FcγRIIb (Fig. 3,A, top panel). As a control, the same experiment was performed on FcγRIIa-R transfectants. IV.3 Fab almost completely blocked the binding of ahIgG to FcγRIIa, while 4F5 F(ab′)2 did not affect the binding of ahIgG to FcγRIIa (Fig. 3,A, bottom panel). These results are in agreement with previous findings that mAb IV.3 competes with the ligand binding of CD32. In a reciprocal paradigm, preincubation of FcγRIIb-transfectants with ahIgG, but not similarly prepared heat-aggregated human IgA, partially blocked the binding of 4F5 to FcγRIIb (Fig. 3,B). Similar competition experiments between mAbs 4F5 and AT10 demonstrated that 4F5 did not affect the binding of Alexa 488-conjugated mAb AT10 to FcγRIIb (Fig. 3 C). Overall, our data suggest that mAb 4F5 partially competes with ahIgG but not with the pan-mAb AT10 for receptor binding.
The expression profile of FcγRIIb on circulating leukocytes
Using mAb 4F5 and other lineage-specific markers, we were able to examine the cell surface expression of FcγRIIb on circulating leukocyte subpopulations. We performed multicolor flow cytometry analysis on whole blood or Ficoll-purified PBMCs. B lymphocytes (CD19+) express high levels of FcγRIIb as evidenced by mAb AT10 and 4F5 reactivity (Fig. 4,A). Circulating T cells (CD3+) do not express any detectable levels of FcγRIIb as determined by negative 4F5 staining (data not shown). The CD14+ monocytes and neutrophils (polymorphonuclear neutrophils (PMNs)) express FcγRIIb on the cell surface although at a much lower level than FcγRIIa (Fig. 4, B and C). A major population of mDCs in human blood is defined as CD19−BDCA1+ (41), and those mDCs express both FcγRIIa and FcγRIIb (Fig. 4 D). Of note, the level of FcγRIIb expression on mDCs is heterogeneous both between donors and within a single donor, which may reflect the heterogeneity of mDCs at different maturation and/or differentiation stages.
pDCs in human blood can be identified by a single marker, BDCA2 (41). Previous studies have suggested that FcγRIIa is the only FcR expressed on pDCs (10, 42). Because pDCs are a major producer of type I IFN (IFN-α/IFN-β) and may be involved in the “IFN signature” in SLE (43), we examined the expression of FcγRIIb on circulating pDCs from a number of donors. We found that FcγRIIb is expressed at very low levels on circulating pDCs in ∼10% donors, but in most donors, FcγRIIb is not expressed at detectable levels (Fig. 4 E, top and bottom panels, respectively). In summary, our data demonstrate the cell surface expression of FcγRIIb on myeloid-lineage cells including monocytes, neutrophils, and mDCs. Our data also suggest that FcγRIIb is expressed on circulating pDCs from a small percentage of normal donors.
To confirm the expression profile of FcγRIIb in leukocyte subpopulations, we performed semiquantitative RT-PCR for FcγRIIb and FcγRIIa messages from FACS-purified leukocyte subsets. FcγRIIb has two major splice variants, IIb1 which is more abundantly expressed in B lymphocytes and IIb2 which is more abundantly expressed in myeloid lineage cells (1). FcγRIIb2 lacks the 19 aa of the first exon encoding the cytoplasmic domain and is more capable of endocytosis (44). PCR primers for FcγRIIb were designed to amplify both IIb1 and IIb2 messages with amplicons distinguishable on the basis of size. The specificity of the primers and PCR were confirmed by PCRs using FcγRIIb1, FcγRIIb2, or FcγRIIa cDNA-containing plasmids as the template (Fig. 5, lanes 1–3). In agreement with previous literature (45), CD19+ B cells express more IIb1 message than IIb2 and do not express detectable IIa message while CD14+ monocytes express more IIb2 than IIb1 and abundant IIa messages (Fig. 5, lanes 4 and 5). Purified mDCs (CD19−BDCA1+) and PMNs express more FcγRIIb2 than IIb1 messages and abundant FcγRIIa message (Fig. 5, lanes 7 and 9). Because pDCs are very rare in PBMCs (∼0.1–0.5%), we first used BDCA4 microbeads (Miltenyi Biotec) to enrich pDCs and then sorted BDCA2+ cells. pDCs from donor 1 express IIb2, IIb1, and IIa messages while pDCs from donor 2 do express IIa message but no detectable IIb1 or IIb2 messages (Fig. 5, lanes 6 and 8), findings in agreement with the cell surface staining by mAbs 4F5 and IV.3 (Fig. 4 E). Taken together, we have demonstrated the expression profile of FcγRIIb on PBL subpopulations at both message and cell surface protein levels.
The expression of FcγRIIb on pDCs is induced by cytokines
Because of the potential importance of FcγRIIb expression by pDCs, we explored the possibility that FcγRIIb on pDCs could be up-regulated by cytokines. PBMCs were cultured in RPMI 1640 with 10% human AB serum in the absence or presence of cytokines for 22 h and the expression of FcγRIIb was examined by mAb 4F5 staining. Because IL-4 plus IL-10 synergistically up-regulate FcγRIIb on monocyte-derived DCs (39), we tested the effect of IL-4 plus IL-10 and IL-3 plus GM-CSF on FcγRIIb expression by pDCs. Starting with a normal donor who did not show detectable levels of FcγRIIb on pDCs before culture (Fig. 6,A), ∼8% of the pDCs were FcγRIIb-positive after in vitro culture in pooled AB serum. Approximately 20% of the pDCs were FcγRIIb-positive with IL-3 plus GM-CSF treatment, and ∼30% of the pDCs were FcγRIIb-positive with IL-4 plus IL-10 treatment (Fig. 6 B). Based on five independent experiments, our data suggest that the expression of FcγRIIb can be up-regulated by cytokines, involving 10–30% of pDCs. The expression of the inhibitory FcγRIIb on pDCs could provide an important control for immune complex-induced IFN-α production.
The expression of FcγRIIb is regulated by its promoter haplotypes
Of the two major promoter haplotypes in the human FCGR2B gene, the less frequent 2B.4 haplotype leads to increased FcγRIIb expression on B lymphocytes (21, 22). Because myeloid lineage cells express both FcγRIIa and FcγRIIb, the unique cell surface expression of FcγRIIb cannot be determined using pan-FcγRII mAbs. Using anti-FcγRIIb mAb 4F5, we next examined the cell surface expression of FcγRIIb on circulating monocytes, neutrophils, and mDCs from normal donors with homozygous 2B.1 or heterozygous 2B.1/2B.4 haplotype. Donors with the 2B.4 haplotype have increased mAb 4F5 staining on CD14+ monocytes compared with 2B.1 homozygous donors (Fig. 7,A, p < 0.002, two-tailed, Mann-Whitney U test), which is in agreement with the previously reported expression of FcγRIIb in monocytes by Western blot analysis (21). Similarly, donors with the 2B.4 haplotype have increased 4F5 staining levels on neutrophils and CD19−BDCA1+ mDCs (Fig. 7, B and C, p < 0.013 and p < 0.001, respectively, two-tailed, Mann-Whitney U test). Analysis of mAb IV.3-staining levels on the same cells did not reveal any difference between the 2B.4 and 2B.1 donors (data not shown).
Because monocytes, neutrophils, and mDCs have the same myeloid origin, we next examined the potential correlation of FcγRIIb expression levels on these three cell types. Although the expression levels of FcγRIIb varied over a several-fold range between normal donors, there was a significant correlation among the expression levels of FcγRIIb on monocytes, mDCs, and neutrophils (Spearman rank correlation coefficients of 0.692, 0.535, and 0.674 for monocytes vs neutrophils, monocytes vs mDCs, and neutrophils vs mDCs, respectively; p values for all the correlations are <0.002). Our data suggest that FcγRIIb is coordinately expressed across primary monocytes, neutrophils, and mDCs from normals in the constitutive state.
FcγRIIb expression is dysregulated on B lymphocytes in active SLE patients
Recent studies in mice have highlighted the important role of FcγRIIb in the development of lupus autoimmunity (19, 46). Therefore, we compared FcγRIIb expression on circulating leukocytes from SLE patients and normal controls. We first examined FcγRIIb expression on circulating B cell subpopulations: naive (CD19+CD27−), memory (CD19+CD27+), and a subgroup of plasma (CD19lowCD27high) B cells. We observed that in patients with active SLE, there is a significant increase in the number and percentage of CD19lowCD27high plasmablast B cells: 12.9% in a representative SLE patient vs 2.1% in a normal donor (Fig. 8,A). The number of naive B cells was substantially decreased and the percentage of memory B cells was relatively increased in SLE patients (Fig. 8,A), as previously reported (47, 48, 49). In individual normal donors, FcγRIIb expression on memory B lymphocytes appeared to be up-regulated relative to naive B lymphocytes (Fig. 8,B, left panel), a finding confirmed in a group of 30 normal controls with ∼20% more expression on memory B cells than on their naive B cells (Fig. 8,C; p < 0.0001, two-tailed, paired, Student’s t test). The expression levels of FcγRIIb on CD27high plasma B cells of normal controls were not different from those on the naive B compartments (Fig. 8,C). However, in 19 active SLE patients (SLEDAI ≥2), FcγRIIb expression was decreased on memory and plasma B lymphocytes compared with their own naive B compartments (Fig. 8, B and C; p < 0.05 and p < 0.04, respectively, two-tailed, paired, Student’s t test). The FcγRIIb expression levels on active SLE memory B lymphocytes was more significantly decreased when compared with those on normal memory B lymphocytes (∼30% less; Fig. 8,C; p < 0.0001, two-tailed, unpaired, Student’s t test). Similarly, the FcγRIIb expression levels on active SLE plasma B cells was also decreased compared with normal plasma B cells (∼15% less; Fig. 8,C; p < 0.03, two-tailed, unpaired, Student’s t test). However, the expression levels of FcγRIIb on memory or plasma B lymphocytes from five SLE patients in remission (SLEDAI = 0) were not significantly different from those on their naive B compartments, nor from those on normal memory or plasma B lymphocytes (Fig. 8,C). The expression levels of FcγRIIb on naive B compartments were not different between normals, active SLE, or SLE in remission (Fig. 8 C). In summary, our data strongly suggest that the expression of FcγRIIb is regulated during normal B cell differentiation and activation and that this regulation is abnormal in patients with active SLE.
Comparison of FcγRIIb expression levels on myeloid-lineage cells in normal controls and SLE patients
In contrast, the expression of FcγRIIb on myeloid-lineage cells including monocytes, neutrophils, and mDCs was not altered in SLE. Analyses of 26 normal controls and 14 active SLE patients revealed no difference in the expression levels of FcγRIIb on CD19−BDCA1+ mDCs (Fig. 9). Similarly, there was no significant difference in the expression levels of FcγRIIb on CD14+ monocytes and neutrophils between normal controls and active SLE patients despite an increased expression of FcγRIa (CD64) on monocytes (data not shown) as previously described (50).
Discussion
The ITIM-bearing FcγRIIb is expressed on lymphoid- and myeloid-lineage cells and plays an important role in immune regulation (2). Studies in mice have suggested that the expression levels of FcγRIIb greatly influence humoral immune responses and the development of autoimmunity. Mice with reduced FcγRIIb expression exhibit enhanced IgG responses and an autoimmune-prone phenotype (16, 17, 18). Partial restoration of FcγRIIb expression on B cells in lupus-prone mouse strains is sufficient to restore tolerance and prevent autoimmunity (19). On a susceptible genetic background, FCGR2B-deficient mice develop a lupus-like autoimmune disease (15). Reciprocally, recent studies suggest that FcγRIIb on DCs promote T-independent humoral responses and B cell recall responses (12, 13). In each model, the data strongly suggest that FcγRIIb is critical for the modulation of immune responses.
However, studies of FcγRIIb in the human system have been hampered by the lack of mAbs able to differentiate between the highly homologous FcγRIIa and FcγRIIb. In this study, we have developed mAb 4F5 that reacts with human FcγRIIb but not FcγRIIa under both native and denatured conditions. mAb 4F5 partially blocks the ligand-binding of FcγRIIb and mediates an inhibitory signal by co-cross-linking FcγRIIb with activating FcRs. Thus, mAb 4F5 not only enables study of the expression, regulation, and function of this important receptor, but also may point to the therapeutic applications where the modulation of FcγRIIb is required.
Using mAb 4F5, we have determined the cell surface expression profile of FcγRIIb on different leukocyte subpopulations. In addition to expression on B lymphocytes, monocytes, neutrophils, and myeloid DCs, FcγRIIb is also constitutively expressed at low levels on pDCs from some donors, and this expression on pDCs can be up-regulated by incubation with certain cytokines. Because pDCs are a major producer of type I IFNs (IFN-α/IFN-β) and enhanced expression of a range of IFN-α-regulated genes (“IFN-α-signature”) has been observed in active SLE patients (43, 51, 52), the expression of the inhibitory FcγRIIb on pDCs could provide an important control for immune complex-mediated production of IFN-α. Modulation of FcγRIIb expression on pDCs might provide a novel therapeutic pathway to induce SLE remission.
We have previously identified two major promoter haplotypes in the human FCGR2B gene: 2B.1 and 2B.4 (21, 22). The less frequent 2B.4 haplotype has higher binding capacity for GATA4 and YYI transcription factors and leads to higher expression in a luciferase reporter system. Donors with the 2B.4 haplotype have increased FcγRIIb expression on B lymphocytes, and in this study, we have used the newly developed anti-FcγRIIb mAb 4F5 to demonstrate that donors with the 2B.4 haplotype also have increased FcγRIIb expression on myeloid-lineage cells, including monocytes, neutrophils, and mDCs. Our data suggest that the expression of FcγRIIb is regulated by its natural promoter polymorphisms in both lymphoid and myeloid-lineage cells. However, our data showing differential regulation of FcγRIIb between normals and SLE patients in some, but not all, cell subsets also suggest that FcγRIIb expression may have tissue-specific properties.
Our comparison of FcγRIIb expression in normal donors and SLE patients suggests a dysregulation of FcγRIIb on SLE B lymphocytes. In normal donors, the expression of FcγRIIb is up-regulated on memory B cells compared with naive B cells, perhaps as a means to negatively control humoral responses. In contrast, the expression of FcγRIIb in active SLE patients is significantly down-regulated on memory B cells compared with their own naive B cells and memory B cells from normal donors. Similarly, the expression of FcγRIIb is decreased on SLE plasma cells compared with their naive B cells and normal plasma B cells. Our observation of decreased expression of FcγRIIb on SLE memory and plasma B lymphocytes together with previous observation of defective FcγRIIb signaling in SLE B cells may provide an important mechanism for the overproduction of autoantibodies and other B cell abnormalities in SLE (53). In a transgenic mouse model with anti-DNA BCR, FcγRIIb is shown to mediate peripheral tolerance by limiting the cell number of anti-DNA plasma cells (46). Further studies to examine the function of FcγRIIb in B cell differentiation and development will be particularly interesting because they will not only help to understand the normal control of humoral immune responses but also shed light on the pathogenesis of autoimmune diseases. In addition, the mechanism for FcγRIIb dysregulation in SLE B cells may represent a more general defect in the B cell development and differentiation in patients with SLE.
The expression level of FcγRIIb on myeloid-lineage cells showed a broad distribution among normal donors and SLE patients, a pattern which was pronounced on mDCs and may reflect stages of differentiation. Although monocytes from SLE patients showed higher levels of FcγRIa (CD64), consistent with an IFN effect (50, 51, 52), comparison of FcγRIIb expression on monocytes, neutrophils, and mDCs between SLE patients and normal donors did not reveal a significant difference. This result may be due, in part, to the relative small sample size and, in part, to the multiple factors of activation and differentiation states of cells within the myeloid compartment. Indeed, this suggests that FcγRIIb may be regulated in a lymphoid and myeloid lineage-specific manner under certain circumstances and the role of FcγRIIb in autoimmunity may depend upon the specific cell types involved.
In summary, the expression profile of human FcγRIIb is regulated in both the lymphoid and myeloid compartments not only by its natural gene polymorphisms but also by active autoimmunity exemplified by SLE. Our data further suggest several pathways through which FCGR2B may be an important disease susceptibility gene for SLE and implicate FcγRIIb as a potential therapeutic target for human diseases.
Acknowledgments
We thank Drs. G. S. Alarcón, H. Bastian, W. Chatham, and B. Fessler for identification assessment of SLE patients and Susan Kimberly for assistance in patient recruitment.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported in part by Grants P01 AR49084, R01 AR42476, R01 AR33062, and N01-AI40068 from the National Institutes of Health and in part by the University of Alabama at Birmingham Rheumatic Diseases Core Center (P30 AR48311) and the General Clinical Research Center (M01 RR-00032).
Abbreviations used in this paper: DC, dendritic cell; mDC, myeloid DC; pDC, plasmacytoid DC; SLE, systemic lupus erythematosus; SLEDAI, SLE Disease Activity Index; EC, extracellular domain; MFI, mean fluorescence intensity; [Ca2+]i, intracellular Ca2+ concentration; mIgG, mouse IgG; ahIgG, heat-aggregated human IgG; PMN, polymorphonuclear neutrophil.