Background: FcγRI plays important roles in antibody functions.
Results: We report the first crystal structure of the extracellular human FcγRI.
Conclusion: The receptor D3 domain is positioned away from the IgG binding site, and its shorter D2 domain FG-loop is important for its high affinity.
Significance: This work provides insights to the mechanism of FcγRI function and helps to design therapeutic reagents.
Keywords: Antibodies, Cell Surface Receptor, Crystal Structure, Immunology, Protein Structure, CD64, Fc Receptor, Immunoglobulin G, Ligand Binding, Antibody
Abstract
Fcγ receptors (FcγRs) play critical roles in humoral and cellular immune responses through interactions with the Fc region of immunoglobulin G (IgG). Among them, FcγRI is the only high affinity receptor for IgG and thus is a potential target for immunotherapy. Here we report the first crystal structure of an FcγRI with all three extracellular Ig-like domains (designated as D1, D2, and D3). The structure shows that, first, FcγRI has an acute D1-D2 hinge angle similar to that of FcϵRI but much smaller than those observed in the low affinity Fcγ receptors. Second, the D3 domain of FcγRI is positioned away from the putative IgG binding site on the receptor and is thus unlikely to make direct contacts with Fc. Third, the replacement of FcγRIII FG-loop (171LVGSKNV177) with that of FcγRI (171MGKHRY176) resulted in a 15-fold increase in IgG1 binding affinity, whereas a valine insertion in the FcγRI FG-loop (171MVGKHRY177) abolished the affinity enhancement. Thus, the FcγRI FG-loop with its conserved one-residue deletion is critical to the high affinity IgG binding. The structural results support FcγRI binding to IgG in a similar mode as its low affinity counterparts. Taken together, our study suggests a molecular mechanism for the high affinity IgG recognition by FcγRI and provides a structural basis for understanding its physiological function and its therapeutic implication in treating autoimmune diseases.
Introduction
Fc receptors (FcR)2 play important roles in antibody-mediated cellular activations against microbial infections and in pathogenesis of autoimmune diseases (1–9). Most Fc receptors belong to the Ig superfamily except for FcRn and FcϵRII (CD23), which display class I major histocompatibility antigens and C-type lectins folds, respectively. Immunoglobulins G are the most abundant antibodies in humans and their receptors, which include FcγRI(CD64), FcγRIIA, B, C (CD32), and FcγRIII (CD16) (10), have been extensively characterized. Among them, FcγRI, FcγRIIA, and FcγRIII are activating receptors that either require the association with an immunoreceptor tyrosine-based activation motif (ITAM) containing FcR common γ chain (FcγRI and FcγRIII) or bear an ITAM in their cytoplasmic tail (FcγRIIA and IIC) (8). FcγRIIB is an inhibitory receptor that contains an immunoreceptor tyrosine-based inhibitory motif within its cytoplasmic domain. According to the binding affinities to their cognate antibodies, FcγRI and FcϵRI are high affinity receptors with dissociation constants ranging from 10−8 to 10−10 m (7, 8, 10), whereas other Fc receptors, such as FcγRII and FcγRIII, are low affinity receptors with dissociation constants approximately ∼10−5–10−7 m. Structurally, FcγRI is the only Fc receptor with three extracellular Ig-like domains (designated herein D1, D2, and D3), whereas all other Ig superfamily FcRs contain two Ig-like domains. To date, several structures of low affinity FcγR and their complexes with IgG-Fc have been published (11, 12). However, no atomic structural information is available for FcγRI, and consequently, the mechanism of high affinity receptor-antibody recognition remains to be elucidated.
The interaction between FcγRs and IgGs also holds profound implications for the development of therapeutic treatments for antibody-mediated autoimmune diseases. For example, the infusion of high doses of serum IgG (intravenous immunoglobulins) has been used to treat immune thrombocytopenia and Kawasaki disease in human and arthritis as well as nephrotoxic nephritis in mouse models (13–17). Because of its unique high affinity antibody binding, FcγRI has been developed as a potential therapeutic reagent to treat immune complex-mediated disease (18, 19). To further understand the high affinity antibody recognition by FcγRI and to facilitate the development of FcγRI-mediated immunotherapy, we determined the structure of the full-length extracellular domain of human FcγRI, investigated the molecular basis of its high affinity IgG binding using site-directed mutagenesis, and assessed the effects of IgG sialylation on FcγRI binding. Our study provides structural insights for understanding the physiological function of the high affinity receptor for IgG.
EXPERIMENTAL PROCEDURES
Protein Expression, Purification, and Crystallization
The ectodomain of FcγRI (residues 1–282) with its native signal sequence was expressed in Chinese hamster ovary DXB-11 cells and purified via IgG affinity chromatography followed by cation exchange and gel filtration chromatography as described previously (18, 19). Before crystallization, FcγRI was dialyzed against 10 mm Hepes (pH 7.4) and 0.1 m NaCl and concentrated to a final A280 nm of 29. The crystals used for data collection were grown in hanging drops at 22 °C using 1 μl of protein and 1 μl of reservoir solution (10% PEG8000 and 10% PEG1000).
Diffraction Data Collection and Structure Determination
The crystals were immersed in cryoprotectant (10% PEG8000, 10% PEG1000, and 20% glycerol) before flash-cooling in liquid nitrogen. The x-ray data were collected to 2.65 Å resolution at SER-CAT beamlines and processed with HKL2000 (20) (see Table 1). FcγRI crystals belong to space group P3221 with the cell constants of a = b = 92.83 Å and c = 90.76 Å. The structure of the FcγRI ectodomain was solved by a molecular replacement method with the program Bables (21) and Phaser (22) in CCP4 packages (23). More specifically, the Ig domains 1 and 2 (D1 and D2) of FcγRI were solved by Bables, which picked the structure of human high affinity IgE receptor (PDB ID 1F2Q) as the best search model. Then with the D1 and D2 domains fixed in position, the D3 domain of FcγRI was solved by Phaser using the D2 domain of FcγRIIA (PDB ID 3D5O) as the search model. Model building and refinement were carried out using Coot (24) and autoBuster (25). The overall electron density of the receptor is of good quality (supplemental Fig. S1). Carbohydrate molecules were added manually using (2Fo − Fc) electron density maps contoured at 1.0σ (standard deviation of the map) and refined. The residues are numbered consistent with FCGR1_HUMAN in the Swiss-Prot entry. The final model includes residues 21–282 of the FcγRI ectodomain, with the exception of one loop region, residues 218–223. The Ramachandran statistics were generated with PROCHECK (26). Hinge angles were calculated using program HINGE (27), and the buried surface area was calculated using AreaIMol (28) in CCP4 packages. All structure figures were generated with PyMOL (29). The coordinates of FcγRI have been deposited in the Protein Data Bank under the ID code 3RJD.
TABLE 1.
Data collection and refinement statistics
| Data collection | |
| Space group | P3221 |
| Unit cell dimension (Å) | a = b = 92.8, c = 90.8 |
| Resolution range (Å) | 40.0-2.65 |
| Unique reflections | 13,522 (847)a |
| Average redundancy | 10.7 (8.2)a |
| Rsym (%)b | 10.2 (59.2)a |
| I/σ(I) | 23.7 (2.9)a |
| Completeness (%) | 99.9 (98.1)a |
| Refinement statistics | |
| Refinement resolution (Å) | 36.8-2.65 |
| Rcryst (%)c | 24.8 (28.5)d |
| Rfree (%) | 27.6 (30.4)d |
| Protein atoms | 2,036 |
| Waters and ligands | 49×H2O, 1×PEG, 9×NAG, 3×MAN, 1×FUCe |
| Root mean square deviation from ideal values | |
| Bond length (Å) | 0.007 |
| Bond angle (°) | 1.12 |
| Mean B-factor (Å2) | 84.3 |
| Wilson plot B-factor (Å2) | 87.2 |
| Ramachandran statistics | |
| Most favored region (%) | 95.3 |
| Additionally allowed (%) | 4.7 |
a Values are for the highest resolution shell in data collection (2.71-2.65 Å).
b Rsym = ΣhΣi|Ii(hkl) − 〈I(hkl)〉|/ΣhΣiIi(hkl).
c Rcryst = Σ‖Fo| − |Fc‖/Σ|Fo|, calculated from working data set. Rfree was calculated from 7.08% of data randomly chosen not to be included in refinement.
d Values are for the highest resolution shell in structure refinement (2.86-2.65 Å).
e NAG (N-acetyl glucosamine), MAN (mannose), FUC (fucose).
Site-directed Mutagenesis of FcγRIII
The recombinant FcγRIII with mutated FG loop in the D2 domain were generated by site-directed mutagenesis. Briefly, the FG loop (171LVGSKNV177) of FcγRIII that was previously cloned in pET28b vector was replaced with that of FcγRI (171MGKHRY176) or containing an additional valine insertion at residue 172 (171MVGKHRY177) using a QuikChange II site-directed mutagenesis kit (Agilent) according to the manufacture's instruction. The mutations were confirmed by DNA sequencing (ACGT Inc.). The wild type and the mutated FcγRIII proteins were expressed and purified as described previously (30).
Surface Plasmon Resonance Solution Binding Experiments
Surface plasmon resonance measurements were performed using a BIAcore 3000 instrument and analyzed with BIAevaluation 4.1 software (Biacore AB). Different human IgG subclasses were obtained from Athens Research and Technology (Athens, GA). The sialylated and asialylated Fc samples were gifts from Dr. Jeffrey Ravetch (Rockefeller University). To measure the affinity to different subclasses of human IgGs, FcγRI was immobilized on carboxylated dextran CM5 chips (Biacore AB) to 200–500 response units using a primary amine-coupling in 10 mm sodium acetate (pH 4.0). The analytes consisted of serial dilutions of IgG subclasses or sialylated Fc between 3.72 μm and 29 nm in a buffer containing 10 mm Hepes (pH 7.4) and 0.15 m NaCl. To measure the binding to FcγRIII wild type and mutant proteins, human IgG1 was immobilized on CM5 chip to 20–150 response units as described above. The analytes consisted of serial dilutions of FcγRI and FcγRIII wild type and mutant proteins between 7.89 and 0.03 μm in a buffer containing 10 mm Hepes (pH 7.4) and 0.15 m NaCl plus 0.01% P20 (Biacore AB). The dissociation constants were obtained by kinetic curve-fitting for the binding of FcγRI to IgGs and steady-state fitting for the binding of FcγRIII and its mutants to IgG1, respectively, using BIAevaluation 4.1 (BIAcore Inc.).
RESULTS AND DISCUSSION
The Overall Structure of the FcγRI Ectodomain
The crystal structure of the extracellular domain of FcγRI was determined to 2.65 Å by molecular replacement (Table 1). Carbohydrates were observed and modeled at six glycosylation sites, including asparagine residues 59, 78, 152, 159, 163, and 195 (Fig. 1A). FcγRI is the only Fc receptor that contains three Ig domains, and the overall shape of FcγRI resembles the shape of a sea horse, with the interdomain hinge angles calculated to be 35° between the D1 and D2 and 120° between the D2 and D3 domains. All three domains display a C2-set Ig fold consisting of two antiparallel sheets formed by strands ABE and CFG. The pairwise structural superimposition between the three Ig domains resulted in root mean square deviations of ∼1.7 Å.
FIGURE 1.
A, the overall structure of FcγRI is shown. The D1, D2, and D3 Ig domains are colored in orange, blue, and magenta, respectively. The carbohydrate moieties on each glycosylation sites are shown as sticks. B, D1-D2 interface is shown. C, D2-D3 interface is shown. The hydrogen bonds are represented by red dotted lines.
Structure Comparison with Other Fc Receptors
The structures of the low affinity FcγRII and FcγRIII can be well superimposed onto each other with root mean square deviations of 0.9–1.1 Å (30–33). The hinge angles between the D1 and D2 domains of low affinity FcγRs are remarkably similar, ranging from 55° to 60°. FcγRI, in contrast, displays a smaller D1-D2 domain hinge angle of 35° than those of FcγRII and FcγRIII but similar to the 39° hinge angle of FcϵRI (Fig. 2A) (34). This hinge angle is stabilized by extensive interactions between the D1 and D2 domains, including two interdomain salt bridges, three hydrogen bonds, and numerous hydrophobic interactions (Fig. 1B and Table 2). Some of the FcγRI hinge core residues, such as Phe-34, Glu-37, and Leu-106 in the D1-D2 hinge region are conserved with FcϵRI but not with FcγRII or RIII (Fig. 2B), further supporting their structural similarity. The hinge core in FcγRII and III, in contrast, contains one salt bridge and one hydrogen bond. Except for three conservative amino acid changes, Val-33 to Ile, Glu-99 to Gln, and Val-109 to Ala, the rest of the 15 residues in the D1-D2 hinge core are invariant between human and murine FcγRI (Fig. 2B). The D2-D3 hinge is unique to the three domains of FcγRI. It consists of one salt bridge between Glu-116 from the D2 domain and Lys-271 of the D3 domain and four hydrogen bonds involving Gly-117, Lys-157, Glu-187, and Leu-188 from the D2 domain and Lys-271 and Asn-268 from the D3 domain (Fig. 1C). Except for a conservative Asn-268 to Ser replacement, the other six residues involved in the D2-D3 hinge core are invariant between human and murine FcγRI. Despite fewer interdomain contacts and the appendage nature of the D3 domain, the entire domain has well defined electron densities and does not appear to exist in multiple conformations in the crystal. A plot of the crystallographic B-values for all Cα atoms of FcγRI shows similar thermomobility indices for the D2 and D3 domains (supplemental Fig. S1), further supporting a non-flexible conformation for the D3 domain of the receptor. The overall conservation in both the D1-D2 and the D2-D3 hinge cores between human and murine FcγRI suggests that the murine FcγRI adopts a similar structure to that of the human FcγRI.
FIGURE 2.
Structural and sequence comparisons among Fc receptors. A, shown is a comparison of the hinge angle between FcγRI and other FcRs, including FcγRIIa, FcγRIIb, FcγRIII, and FcϵRI. All structural superpositions of the receptor were based on the D2 domain alignment. The hinge angles were calculated using Hinge program. The FcϵRI and FcγRIII residues involved in Fc contact interface are displayed on their receptor structures together with the putative ligand binding residues of FcγRI. B, shown is sequence alignment of the ectodomain of FcRs. The secondary structure elements (arrow for β–strands and cylinder for helices) are indicated above the sequence. The glycosylation sites in FcγRI are indicated by asterisks. The D1-D2 and D2-D3 interface residues in FcγRI are shaded in yellow and purple, respectively. The conserved pair-wise D1-D2 interactions among FcγRI and FcϵRI are highlighted in black boxes. The Fc-contacting residues in FcγRIII are shaded in red.
TABLE 2.
Hydrogen bonds and salt bridges at D1-D2 or D2-D3 interface
| D1-D2 interface | ||||
|---|---|---|---|---|
| D1 | D2 | Distancea |
||
| FcγRI | FcϵRI | FcγRIII | ||
| Å | ||||
| Trp- 30 NE1 | Arg-123 NH1 | 2.8 | 2.8 | |
| Val-31 N | Val-109 O | 3.0 | ||
| Val-31 O | Arg-112 NH2 | 3.1 | 3.3 | |
| Ser-32 O | Gln-108 NE2 | 2.9 | 3.0 | 3.5 |
| Glu-37 OE1 | His-125 NE2 | 2.8 | 2.8 | 2.8 |
| Glu-99 OE1 | Arg-112 NH1 | 3.4 | ||
| D2-D3 interface | ||||
|---|---|---|---|---|
| D2 | D3 | Distancea |
||
| FcγRI | FcϵRI | FcγRIII | ||
| Å | ||||
| Glu-116 OE1 | Leu-188 N | 2.8 | ||
| Glu-116 OE2 | Lys-271 NZ | 3.2 | ||
| Gly-117 O | Asn-268 ND2 | 3.4 | ||
| Lys-157 NZ | Asn-268 OD1 | 3.5 | ||
| Glu-187 OE2 | Lys-271 NZ | 3.2 | ||
a Hydrogen bond distance cutoff is <3.5 Å.
The Role of D3 Domain in FcγRI Recognition of IgG
Previous mutational studies suggested the importance of the receptor D2 and D3 domains to the high affinity IgG binding (35, 36). The structures of FcγRIII and FcϵRI in complex with their Fc showed a similar ligand binding mode involving the receptor BC, C′E, FG loops, and the C′ strand on the D2 domain despite the differences in their structures and binding affinities (11, 12, 37). Several of the Fc contacting residues of FcγRIII and FcϵRI, including Trp-104 and Trp-127 that form a sandwich around Pro-329 of Fc, are conserved in the high affinity FcγRI structure (Fig. 2, supplemental Fig. S2), arguing a similar Fc recognition mode for FcγRI. When the structure of FcγRI is superimposed onto that of the FcγRIII-Fc complex, its D3 domain is positioned ∼30 Å away from the putative IgG-Fc interface, suggesting that the D3 domain of FcγRI is unlikely to contact directly with IgG-Fc (Fig. 3A). However, the D3 domain of FcγRI may indirectly affect IgG binding through either stabilizing the conformation of the receptor or making the receptor protruding more into the extracellular space than its two-domain Fc receptor counterparts and, thus, more available for ligand binding.
FIGURE 3.
A, a model of FcγRI-Fc interaction is shown. FcγRIII, FcϵRI, IgG Fc, and IgE Fc are shown as schematics colored in green, cyan, yellow, and tint, respectively. FcγRI-Fc complex was modeled based on the structural alignment on the D2 domain of Fc receptors. B, shown is a comparison of the FG loop in FcγRI (blue), FcγRIII (green), and FcϵRI (magenta). The residues are shown as sticks, and the IgG Fc is represented by a yellow surface. The residue positions are numbered accordingly to FcγRI sequence.
The Mechanism of High Affinity FcγRI Recognition of IgG
Although FcγRIII and FcϵRI recognize Fc in similar binding modes, the structures of their ligand complexes show a slightly more buried interface area in FcϵRI/IgE-Fc (1883 Å2) than in FcγRIII/IgG-Fc (1798 Å2) (11, 37). It remains to be seen if FcγRI makes more extensive IgG contacts than the low affinity FcγRs. The known structures also showed that the D2 domain FG-loop on both FcϵRI and FcγRIII formed critical interface contacts with Fc, and the exchange of the D2 domain FG-loop of FcϵRI to FcγRII confers detectable IgE binding (38). Interestingly, the D2 domain FG-loop of FcγRI (171MGKHRY176) contains a single-residue deletion compared with FcγRII/FcγRIII and FcϵRI (Fig. 2B). This deletion is not present on the D1 domain FG-loop of FcγRI and is conserved in all species from human to mouse, indicating a potential functional relevance of this unique shorter FG-loop in FcγRI binding to IgG. Simple structural modeling to replace the longer FG-loop residues of FcγRIII (LVGSKNV) with those of corresponding FcγRI in the FcγRIII/IgG1-Fc structure showed that instead of making more potential contacts with IgG-Fc, the bulkier FcγRI residues were in steric clash with IgG1-Fc. Similarly, when the structure of FcϵRI was superimposed onto that of FcγRIII-Fc complex, it revealed a steric clash between the bulkier FcϵRI FG-loop residues and IgG-Fc (Fig. 3B), suggesting a potential conformational conflict between IgG-Fc and the longer FG-loop of FcϵRI. Indeed, structural overlay showed an ∼3 Å displacement in the position of the shorter FcγRI FG-loop relative to that of FcγRIII (Fig. 3B). To further investigate the contribution of the FcγRI FG-loop and its one-residue deletion to the receptor-IgG binding affinity, we mutationally replaced the FG-loop of FcγRIII (171LVGSKNV177) with both the wild type FcγRI loop (171MGKHRY176) and a mutant FcγRI loop, which contains a FcγRIII-like valine insertion at residue 172 (171MVGKHRY177). The wild type FcγRI and FcγRIII bound to immobilized IgG1 with affinities of 0.02 and 1.5 μm, respectively (Table 3, supplemental Fig. S3A). When the recombinant mutant FcγRIII carrying a shorter, wild type FcγRI FG-loop (171MGKHRY176) was examined for IgG1 binding, it resulted in an affinity of 0.1 μm, 5-fold lower than that of wild type FcγRI but 15-fold higher than that of the wild type FcγRIII (Table 3, supplemental Fig. S3A). Thus, a simple transplant of the FcγRI FG-loop to FcγRIII dramatically increased the receptor binding affinity to IgG1, illustrating the importance of the FG-loop in high affinity antibody binding. The gain in IgG1 binding affinity from swapping the FG-loop is particularly significant as the loop contains only 4 of14 potential receptor-IgG interface residues, most of which are different between FcγRI and FcγRIII. In contrast, the mutant FcγRIII with a valine inserted in the FcγRI FG-loop (171MVGKHRY177) showed a dissociation constant of 0.9 μm for IgG1 binding, similar to that of the wild type FcγRIII despite having all FcγRI residues in the FG-loop (Table 3, supplemental Fig. S3A). The 9-fold affinity difference between the two FG-loop mutations, (171MVGKHRY177) and (171MGKHRY176), suggests that the shorter FG-loop is important for the high affinity IgG binding of FcγRI.
TABLE 3.
Solution binding dissociation constants KD (μm) between FcγR and IgG
SA, sialylated.
| Immobilization | |||
|---|---|---|---|
| IgG1 |
FcγRI |
||
| Analytes | Affinity | Analytes | Affinity |
| μm | μm | ||
| FcγRI | 0.02 ± 0.009 | IgG1 | 0.0 4 ± 0.02 |
| FcγRIII mutant-1 (171MGKHRY176) | 0.11 ± 0.01 | IgG2 | 2.2 ± 0.3 |
| IgG3 | 0.014 ± 0.008 | ||
| FcγRIII mutant-2 (171MVGKHRY177) | 0.9 ± 0.2 | IgG4 | 0.032 ± 0.018 |
| IgG1-Fc | 0.04 ± 0.014 | ||
| FcγRIII (171LVGSKNV177) | 1.5 ± 0.2 | SA-IgG1-Fc | 0.083 ± 0.006 |
IgG Subclass Specificity
Both the high and low affinity Fc receptors discriminate among the four subtypes of IgGs (7, 39–41). In general, FcγRI displays 15–40 nm affinity bindings to IgG1, IgG3, and IgG4 but no detectable binding to IgG2 (IgG1,3 ∼ IgG4 ≫ IgG2) (41). Although the high affinity FcγRI discriminates primarily against IgG2, the low affinity FcγRIII receptors discriminate against both IgG2 and IgG4 (IgG1,3 > IgG2,4). The lower affinity binding of FcγRIII to IgG2 has been attributed to the presence of a deletion in its lower hinge relative to IgG1 (11, 12, 42). This, however, does not explain the specificity of FcγRI, which binds IgG4 but not IgG2 in similar affinity as IgG1 (Table 3, supplemental Fig. S3B). The current structure of FcγRI and its modeled complex with Fc provide a possible explanation to this isotype-dependent IgG4 binding specificity (supplemental Fig. S2). This isotype specificity difference between FcγRI and FcγRIII may be attributed, based on the modeling of FcγRI-IgG interactions, to a hydrogen bond between Tyr-176 in the FG-loop of FcγRI but not Val in FcγRIII and the conserved lower hinge residues of IgG1,3 and IgG4 but not IgG2.
Effect of Sialylation on Fc Binding by FcγRI
On both chains of Fc, ordered carbohydrate moieties are attached to one single conserved glycosylation site, Asn-297. Unlike most glycosylations, these carbohydrates on Fc are extensively interacting with the inner face of the CH2 domain to provide conformational stability for the Fc region (11, 12). Indeed, removal of these carbohydrates by enzymatic digestion dramatically decreased the binding affinity of IgGs to their receptors (43). Structural studies have shown that deglycosylated Fc undergoes remarkable conformational changes in the CH2 domain, although CH3 domains are intact (44). Even though these carbohydrate moieties are away from the receptor contact area, they are known to affect Fc receptor binding. Recently, sialylated Fc was proposed to harbor the anti-inflammatory activity of intravenous immunoglobulins as sialic acid-enriched IgG showed a 10-fold affinity reduction in binding to the low affinity FcγRIIB and FcγRIII (45). To investigate whether sialylation also affects the binding of Fc to FcγRI, we measured the receptor binding to a sialic acid-enriched Fc fragment using surface plasmon resonance analysis (Table 3, supplemental Fig. S3C). The result showed that FcγRI bound to the sialylated Fc with a modest 2-fold reduction in affinity (83 nm) compared with native Fc (42 nm). It remains to be seen whether the reduced binding affinity of FcγRI to sialylated Fc has any physiological relevance in the immunosuppressive activity of intravenous immunoglobulins.
Supplementary Material
Acknowledgments
We thank Dr. Jeffrey Ravetch for providing the sialylated and asialylated recombinant Fc fragments and Katherine Lewis and Kenneth Lewis for providing the recombinant FcγRI.
This work was supported, in whole or in part, by National Institutes of Health intramural research funding (NIAID).

The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. S1–S3.
The atomic coordinates and structure factors (code 3RJD) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).
- FcR
- Fc receptor.
REFERENCES
- 1. Anthony R. M., Ravetch J. V. (2010) J. Clin. Immunol. 30, S9–14 [DOI] [PubMed] [Google Scholar]
- 2. van de Velde N. C., Mottram P. L., Hogarth P. M. (2006) Springer Semin. Immunopathol. 28, 329–338 [DOI] [PubMed] [Google Scholar]
- 3. Tamm A., Schmidt R. E. (1997) Int. Rev. Immunol. 16, 57–85 [DOI] [PubMed] [Google Scholar]
- 4. van de Winkel J. G. (2010) Immunol. Lett. 128, 4–5 [DOI] [PubMed] [Google Scholar]
- 5. Unkeless J. C., Scigliano E., Freedman V. H. (1988) Annu. Rev. Immunol. 6, 251–281 [DOI] [PubMed] [Google Scholar]
- 6. Fridman W. H. (1991) FASEB J. 5, 2684–2690 [DOI] [PubMed] [Google Scholar]
- 7. Hulett M. D., Hogarth P. M. (1994) Adv. Immunol. 57, 1–127 [DOI] [PubMed] [Google Scholar]
- 8. Ravetch J. V., Kinet J. P. (1991) Annu. Rev. Immunol. 9, 457–492 [DOI] [PubMed] [Google Scholar]
- 9. Ravetch J. V., Bolland S. (2001) Annu. Rev. Immunol. 19, 275–290 [DOI] [PubMed] [Google Scholar]
- 10. Daëron M. (1997) Annu. Rev. Immunol. 15, 203–234 [DOI] [PubMed] [Google Scholar]
- 11. Radaev S., Motyka S., Fridman W. H., Sautes-Fridman C., Sun P. D. (2001) J. Biol. Chem. 276, 16469–16477 [DOI] [PubMed] [Google Scholar]
- 12. Sondermann P., Huber R., Oosthuizen V., Jacob U. (2000) Nature 406, 267–273 [DOI] [PubMed] [Google Scholar]
- 13. Ravetch J. V. (2002) J. Clin. Invest. 110, 1759–1761 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Yuasa T., Kubo S., Yoshino T., Ujike A., Matsumura K., Ono M., Ravetch J. V., Takai T. (1999) J. Exp. Med. 189, 187–194 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Kleinau S., Martinsson P., Heyman B. (2000) J. Exp. Med. 191, 1611–1616 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Bruhns P., Samuelsson A., Pollard J. W., Ravetch J. V. (2003) Immunity 18, 573–581 [DOI] [PubMed] [Google Scholar]
- 17. Debré M., Bonnet M. C., Fridman W. H., Carosella E., Philippe N., Reinert P., Vilmer E., Kaplan C., Teillaud J. L., Griscelli C. (1993) Lancet 342, 945–949 [DOI] [PubMed] [Google Scholar]
- 18. Ellsworth J. L., Maurer M., Harder B., Hamacher N., Lantry M., Lewis K. B., Rene S., Byrnes-Blake K., Underwood S., Waggie K. S., Visich J., Lewis K. E. (2008) J. Immunol. 180, 580–589 [DOI] [PubMed] [Google Scholar]
- 19. Ellsworth J. L., Hamacher N., Harder B., Bannink K., Bukowski T. R., Byrnes-Blake K., Underwood S., Oliver C., Waggie K. S., Noriega C., Hebb L., Rixon M. W., Lewis K. E. (2009) J. Immunol. 182, 7272–7279 [DOI] [PubMed] [Google Scholar]
- 20. Otwinowski Z. M. W. (1997) Methods Enzymol. 276, 307-326 [DOI] [PubMed] [Google Scholar]
- 21. Long F., Vagin A. A., Young P., Murshudov G. N. (2008) Acta Crystallogr. D. Biol. Crystallogr. 64, 125–132 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. McCoy A. J., Grosse-Kunstleve R. W., Adams P. D., Winn M. D., Storoni L. C., Read R. J. (2007) J. Appl. Crystallogr. 40, 658–674 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. The CCP4 suite: programs for protein crystallography (1994) Acta Crystallogr. D. Biol. Crystallogr. 50, 760–763 [DOI] [PubMed] [Google Scholar]
- 24. Emsley P., Lohkamp B., Scott W. G., Cowtan K. (2010) Acta Crystallogr. D. Biol. Crystallogr. 66, 486–501 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Blanc E., Roversi P., Vonrhein C., Flensburg C., Lea S. M., Bricogne G. (2004) Acta Crystallogr. D. Biol. Crystallogr. 60, 2210–2221 [DOI] [PubMed] [Google Scholar]
- 26. Laskowski R. A., MacArthur M. W., Moss D. S., Thornton J. M. (1993) J. Appl. Crystallogr. 26, 283–291 [Google Scholar]
- 27. Snyder G. A., Brooks A. G., Sun P. D. (1999) Proc. Natl. Acad. Sci. U.S.A. 96, 3864–3869 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lee B., Richards F. M. (1971) J. Mol. Biol. 55, 379–400 [DOI] [PubMed] [Google Scholar]
- 29. DeLano W. L. (2002) The PyMOL Molecular Graphics System, Version 1.4, Schrodinger, LLC, New York [Google Scholar]
- 30. Zhang Y., Boesen C. C., Radaev S., Brooks A. G., Fridman W. H., Sautes-Fridman C., Sun P. D. (2000) Immunity. 13, 387–395 [DOI] [PubMed] [Google Scholar]
- 31. Maxwell K. F., Powell M. S., Hulett M. D., Barton P. A., McKenzie I. F., Garrett T. P., Hogarth P. M. (1999) Nat. Struct. Biol. 6, 437–442 [DOI] [PubMed] [Google Scholar]
- 32. Sondermann P., Huber R., Jacob U. (1999) EMBO J. 18, 1095–1103 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Sondermann P., Kaiser J., Jacob U. (2001) J. Mol. Biol. 309, 737–749 [DOI] [PubMed] [Google Scholar]
- 34. Garman S. C., Kinet J. P., Jardetzky T. S. (1998) Cell 95, 951–961 [DOI] [PubMed] [Google Scholar]
- 35. Hulett M. D., Osman N., McKenzie I. F., Hogarth P. M. (1991) J. Immunol. 147, 1863–1868 [PubMed] [Google Scholar]
- 36. Hulett M. D., Hogarth P. M. (1998) Mol. Immunol. 35, 989–996 [DOI] [PubMed] [Google Scholar]
- 37. Garman S. C., Wurzburg B. A., Tarchevskaya S. S., Kinet J. P., Jardetzky T. S. (2000) Nature 406, 259–266 [DOI] [PubMed] [Google Scholar]
- 38. Hulett M. D., McKenzie I. F., Hogarth P. M. (1993) Eur. J. Immunol. 23, 640–645 [DOI] [PubMed] [Google Scholar]
- 39. Galon J., Robertson M. W., Galinha A., Mazières N., Spagnoli R., Fridman W. H., Sautès C. (1997) Eur. J. Immunol. 27, 1928–1932 [DOI] [PubMed] [Google Scholar]
- 40. Powell M. S., Barton P. A., Emmanouilidis D., Wines B. D., Neumann G. M., Peitersz G. A., Maxwell K. F., Garrett T. P., Hogarth P. M. (1999) Immunol. Lett. 68, 17–23 [DOI] [PubMed] [Google Scholar]
- 41. Bruhns P., Iannascoli B., England P., Mancardi D. A., Fernandez N., Jorieux S., Daëron M. (2009) Blood 113, 3716–3725 [DOI] [PubMed] [Google Scholar]
- 42. Chappel M. S., Isenman D. E., Everett M., Xu Y. Y., Dorrington K. J., Klein M. H. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 9036–9040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Radaev S., Sun P. D. (2001) J. Biol. Chem. 276, 16478–16483 [DOI] [PubMed] [Google Scholar]
- 44. Feige M. J., Nath S., Catharino S. R., Weinfurtner D., Steinbacher S., Buchner J. (2009) J. Mol. Biol. 391, 599–608 [DOI] [PubMed] [Google Scholar]
- 45. Kaneko Y., Nimmerjahn F., Ravetch J. V. (2006) Science 313, 670–673 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.



