Human Molecular Genetics Advance Access originally published online on August 22, 2005
Human Molecular Genetics 2005 14(19):2881-2892; doi:10.1093/hmg/ddi320
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Fc
RIIB Ile232Thr transmembrane polymorphism associated with human systemic lupus erythematosus decreases affinity to lipid rafts and attenuates inhibitory effects on B cell receptor signaling
1Department of Allergy and Rheumatology and 2Department of Human Genetics, Faculty of Medicine, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan and 3Department of Developmental Biology, Research Institute for Radiation Biology and Medicine, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8553, Japan
* To whom correspondence should be addressed. Tel: +81 338155411 ext. 33175; Fax: +81 338155954; Email: honda-phy{at}h.u-tokyo.ac.jp
Received June 16, 2005; Accepted August 16, 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The B cell inhibitory receptor Fc
RIIB plays crucial roles in the maintenance of self-tolerance. We have identified a polymorphism FCGR2B c.695T>C that results in the non-conservative replacement of 232Ile at the transmembrane helix to Thr and demonstrated the association of the polymorphism with susceptibility to systemic lupus erythematosus (SLE) in Asians. In this study, we examined the impact of FCGR2B c.695T>C on the functional properties of FcRIIB by expressing each allele product in a human B cell line ST486 lacking endogenous FcRIIB. FcRIIB 232Thr was found to be significantly less potent than wild-type 232Ile in inhibiting B cell receptor (BCR)-mediated phosphatidylinositol-3,4,5-trisphosphate accumulation, Akt and PLC2 activation and calcium mobilization, and to display decreased levels of tyrosine phosphorylation and SH2-containing 5'-inositolphosphate phosphatase recruitment compared with 232Ile after IgG Fc-mediated coligation with BCR. Notably, a quantitative analysis of the subcellular distribution of FcRIIB using 125I-labeled anti-FcRIIB revealed that FcRIIB 232Thr is less effectively distributed to detergent-insoluble lipid rafts than 232Ile, findings in accordance with the importance of the transmembrane amino acid residues, in particular large hydrophobic amino acids including Ile, in the association of membrane proteins with lipid rafts. Given the crucial roles of lipid rafts in integrating BCR signaling, decreased association of FcRIIB 232Thr could contribute to its impaired inhibitory potential. Collectively, the present findings indicate that the Ile232Thr substitution affects the localization and function of FcRIIB and that the molecular mechanism may link the polymorphism and susceptibility to SLE. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Systemic lupus erythematosus (SLE) is a prototypical autoimmune disease characterized by autoantibody production, immune complex deposition and multiple organ damage. Although the etiology of SLE is still unknown, both genetic and environmental factors are involved in the pathogenesis of SLE. In humans, genetic association is known at various loci including those coding for human leukocytes antigens (HLAs), complements and low-affinity Fc
receptors (FcRs) (1
–3). The loci for the highly homologous human FcRs are clustered on chromosomes 1q21–24, and genome-wide linkage studies indicated that this area is one of the strongest candidate chromosomal regions for SLE (4,5). FcRs are expressed in various hematopoietic cells and play crucial roles in the regulation of the immune system through the IgG immune complex-mediated signal transduction and immune complex clearance (6–8).
Among the classical FcRs, FcRIIB is unique in that it possesses an immunoreceptor tyrosine-based inhibitory motif (ITIM), a consensus amino acid sequence present in the cytoplasmic domain of the immune inhibitory receptor family (7). Studies in mice demonstrated that FcRIIB plays crucial roles in preventing excessive immune reactions and autoimmunity in vivo. For example, FcRIIB knockout mice display an enhanced humoral immunity (9) and spontaneously develop immune complex-mediated glomerulonephritis in a strain-specific manner (10). The molecular mechanisms underlying FcRIIB-mediated immunosuppression have been precisely investigated. In humans, two FcRIIB isoforms, referred to as FcRIIB1 and FcRIIB2, are generated by alternative splicing. The long FcRIIB1 isoform is preferentially expressed in B cells and the short FcRIIB2 isoform lacking the sequence encoded by the first cytoplasmic exon in cells of myeloid lineage (6). When the B cell receptor (BCR) is coligated with FcRIIB with IgG immune complexes containing relevant antigens, the ITIM tyrosine is phosphorylated by an Src family kinase, Lyn, and recruits SH2-containing 5'-inositolphosphate phosphatase (SHIP) (11). SHIP hydrolyzes phosphatidylinositol-3,4,5-trisphosphate (PIP3) and finally attenuates the activation and survival signals in B cells (12,13).
Besides the biochemical features of the positive and negative regulations of BCR signaling, there emerge the spatio-temporal aspects of the signaling that involve a class of membrane domains, referred to as lipid rafts (14,15). Lipid rafts are enriched with glycosphingolipid and cholesterol, and several classes of upstream signal transducers including Src family kinases concentrated in lipid rafts (16–18). Upon B cell stimulation, lipid rafts promptly build up spatially compartmentalized signaling clusters, to which various signaling molecules including BCR and FcRIIB (19,20), adaptor proteins and downstream kinases and phosphatases are recruited (18). Others and we showed that mutated BCR or Src family kinases segregated from lipid rafts can no longer transduce appropriate signaling (17,19), suggesting that molecular assembly at lipid rafts is not merely coincidental, but important for signal transduction.
Recently, we have found a non-synonymous polymorphism FCGR2B c.695T>C which alters 232Ile of FcRIIB in the middle of the transmembrane helix to Thr. We detected a significant association of the polymorphism with SLE in Asian populations (21–23) and showed that this association was not caused by linkage disequilibrium with known FCGR2A or FCGR3B polymorphisms. Therefore, the FcRIIB Ile232Thr polymorphism likely constitutes a risk factor for SLE in Asians. In this study, we examined the functional consequence of the Ile232Thr substitution and found that the amino acid substitution impairs FcRIIB association with lipid rafts and attenuates the inhibitory effects of FcRIIB on BCR signaling.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Fc
RIIB Ile232Thr substitution does not alter ligand-binding affinityAs the primary structure of Fc
RIIB shows only a modest homologue among species [for instance, amino acid identity between human FcRIIB1 and the mouse homologue (24) is 
60%], we considered that the functional consequence resulting from the minimal structural alteration in FcRIIB Ile232Thr should be examined in the context of human cells. In this study, we chose to compare the properties of FcRIIB1 variants by expressing each of them in a human B-cell line, ST486, that is inherently lacking endogenous FcRIIB (25). We generated multiple ST486 clones expressing human FcRIIB1 232Ile (hereafter referred to as h-232Ile) or FcRIIB1 232Thr (h-232Thr) (Fig. 1A) and selected four pairs of the clones expressing equivalent levels of FcRIIB1 variants (Fig. 1B).
|
We first conducted ligand-binding study using 125I-labeled human IgG, but the attempt was unsuccessful due to a high non-specific binding activity (data not shown). We thus developed whole cell binding assay using FITC-labeled human IgG (fluorescein/IgG molar ratio=3.1, Sigma) as a ligand, and the specific binding activity was successfully measured by flow cytometer as described in Materials and Methods. The mean fluorescent intensity (MFI) of bound FITC-labeled human IgG was measured and the absolute ligand number was determined from the regression line obtained by the simultaneous analysis of MFIs of quantum fluorescence microbeads (Sigma), which were labeled with known numbers of FITC molecules. Data obtained from the competitive fluorescent ligand-binding experiments were analyzed using Scatchard's plot and were fitted by linear regression analysis (Fig. 1C). Dissociation constants (Kd) for h-232Ile and h-232Thr were 6.5±1.5x10–7 and 5.7±1.0x10–7 M, respectively, which were not significantly different, indicating that Ile232Thr substitution does not significantly affect ligand-binding affinity.
Ile232Thr substitution decreases association of chimeric FcRIIB with lipid rafts
Previous studies have shown the importance of transmembrane domain of membrane proteins for their association with lipid rafts (19,26,27), suggesting the possibility that Ile232Thr substitution results in altered subcellular distribution of FcRIIB. We attempted to examine this hypothesis by using density gradient centrifugation assay (27,28). For the quantitative assessment of FcRIIB redistribution, we first tried to prepare 125I-labeled anti-human FcRIIB Fab fragment, because the use of whole IgG as a probe potentially leads to protein polymerization and misinterpretation of its localization (29). However, it was unsuccessful to prepare Fab fragment with a sufficiently high affinity for the assay from available anti-human CD32 monoclonal Abs (mAbs), FLI8.26 and AT-10 (data not shown). As an alternative approach, we prepared chimeric human FcRIIB1 232Ile and FcRIIB1 232Thr whose extracellular domains were replaced with that of mouse FcRIIB (referred to as c-232Ile and c-232Thr, respectively; Fig. 1A), because we have successfully employed Fab prepared from 2.4G2 anti-mouse FcRII/III mAb for the density gradient centrifugation assay (27). We established multiple ST486 clones expressing c-232Ile or c-232Thr, selected four clone pairs with equivalent expression levels (Fig. 1D), and the clones were subjected to the fluorescent ligand-binding assay as described above. The Kd values for c-232Ile and c-232Thr were 3.0±0.24x10–7 and 2.8±0.17x10–7 M, respectively, and there was again no significant difference between the Kd values.
To analyze the association of FcRIIB variants with lipid rafts, c-232Ile and c-232Thr were probed with 125I-2.4G2 Fab and BCR with DA4-4 anti-human IgM mAbs, and these Abs were left untreated or treated with a secondary Ab to induce BCR-FcRIIB coligation. Cells were solubilized with Brji58 and subjected to density gradient centrifugation assay as described previously (27,28). Figure 2A shows a representative profile of 125I-2.4G2 Fab distribution. Fractions 8–10, which reproducibly contain the doubly palmitoylated Csk-binding protein (Cbp) (30), were identified as lipid raft fractions. Limited amounts of c-232Ile and c-232Thr were recovered from lipid rafts under resting conditions and BCR-FcRIIB coligation noticeably induced their redistribution to lipid rafts. The amount of c-232Thr recovered from lipid rafts was consistently smaller than that of c-232Ile both under resting conditions and after BCR-FcRIIB coligation in all the three clone pairs tested and the differences were statistically significant (Fig. 2B). These findings indicate that Ile232Thr substitution decreased the association of FcRIIB with lipid rafts under resting conditions and after the coligation with BCR.
|
BCR-mediated calcium mobilization was less inhibited by c-232Thr than by c-232Ile
Next we compared Fc
RIIB variants for their impact on key events in the BCR-mediated signal transduction pathway. BCR-mediated calcium mobilization was first examined in the ST486 clones expressing c-232Ile or c-232Thr. BCR self-ligation and BCR-FcRIIB coligation were performed by the stimulation of cells with anti-human IgM F(ab')2 and anti-human IgM whole IgG (Jackson Immunoresearch), respectively. To exclude the contamination of F(ab')2 with whole IgG, purity of the Abs was routinely checked before use by gel electrophoresis. Calcium mobilization in the presence of extracellular calcium (1.8 mM) was measured as described previously (27). We used suboptimal concentrations of whole IgG and F(ab')2 [8 µg/ml whole IgG and 13 µg/ml F(ab')2] that induce comparable calcium mobilization in parent ST486 cells.
One of the evident findings was that the magnitudes of calcium mobilization were appreciably smaller in c-232Ile-expressing cells than in parent cells and c-232Thr-expressing cells even when BCR was ligated separately from FcRIIB (Fig. 3A and B), raising the possibility that c-232Ile and c-232Thr exert constitutive, or indirect, inhibition to different degrees even without coligation with BCR at least under the experimental conditions. The differential constitutive inhibition was not due to the chimeric constructs, because such effect was reproduced in the experiments using full-length human FcRIIB variants (Fig. 6). In view of the difference in the magnitude of the constitutive inhibitions, we decided to evaluate the inhibitory potentials of FcRIIB variants by comparing the net magnitudes of calcium mobilization both under BCR self-ligation and BCR-FcRIIB coligation conditions, but not by the decrease in the magnitude of calcium mobilization by FcRIIB coligation. Traces in Figure 3A show the average magnitude of calcium mobilization from 16 experiments using four clone pairs. Cells expressing c-232Thr reproducibly displayed a higher magnitude of calcium mobilization than cells expressing c-232Ile both after BCR self-ligation and after BCR-FcRIIB coligation, and the differences at the peak (120 s after the addition of Abs) and at the sustained phase (400 s) were statistically significant (Fig. 3B). These observations indicate that BCR-mediated calcium mobilization is less inhibited by c-232Thr than by c-232Ile.
|
|
Cells expressing c-232Thr displayed greater magnitudes of BCR-mediated biochemical signaling than those expressing c-232Ile
BCR-mediated tyrosine phosphorylation of Fc
RIIB and SHIP recruitment to FcRIIB are the hallmarks of FcRIIB-mediated signal attenuation and these events are closely related to FcRIIB redistribution to lipid rafts in the vicinity of Lyn. To compare the influence of Ile232Thr substitution on these events, cells were stimulated via BCR, solubilized with n-octyl-ß-D-glucoside, and c-232Ile and c-232Thr were subjected to immunoprecipitation with 2.4G2 and a secondary Ab, followed by immunoblotting with anti-phosphotyrosine mAb and anti-SHIP Ab. As shown in a representative immunoblot (Fig. 4A), tyrosine phosphorylation of c-232Ile and c-232Thr was marginally detectable under resting conditions, the magnitudes were noticeably increased to comparable levels by BCR self-ligation and substantially enhanced after BCR-FcRIIB coligation with clear decrease in that of c-232Thr relative to c-232Ile. The association of SHIP with c-232Ile and c-232Thr was found to be detectable even under resting and BCR self-ligation conditions as well as after BCR-FcRIIB coligation, and the amount of SHIP associated with c-232Thr was noticeably smaller than that associated with c-232Ile after BCR-FcRIIB coligation, which is in good agreement with the inefficient c-232Thr tyrosine phosphorylation compared with c-232Ile under the coligation conditions. The coligation-mediated tyrosine phosphorylation of FcRIIB and SHIP recruitment to FcRIIB were abrogated by the pretreatment of cells with 2.4G2, showing the specificity of BCR-FcRIIB coligation-mediated effects. These results were reproducible in three different pairs of c-232Ile- and c-232Thr-expressing ST486 clones. These findings strongly suggest that Ile232Thr substitution decreases FcRIIB tyrosine phosphorylation level and attenuates SHIP association with FcRIIB after BCR-FcRIIB coligation.
|
Fc
RIIB coligation with BCR attenuates BCR-mediated CD19 tyrosine phosphorylation which leads to impaired intracellular calcium mobilization (31). To compare the effects of c-232Ile and c-232Thr on this event, cells were stimulated via BCR, solubilized with 1% NP-40 and 0.5% deoxycholate, and CD19 was immunoprecipitated and subjected to immunoblotting with anti-phosphotyrosine mAbs. As shown in Figure 4B, BCR self-ligation augmented CD19 tyrosine phosphorylation, and FcRIIB-BCR coligation decreased the magnitude of the phosphorylation both in c-232Ile and c-232Thr cells. The pretreatment of cells with 2.4G2 mAbs prevented the dephosphorylation of CD19, confirming the roles of FcRIIB in the dephosphorylation of CD19. Notably, the magnitudes of CD19 tyrosine phosphorylation was reproducibly greater in c-232Thr-expressing cells than those in c-232Ile-expressing cells both after BCR self-ligation and BCR-FcRIIB. These results were reproducible in three independent experiments.
We next compared the effects of c-232Ile and c-232Thr on BCR-mediated early biochemical events by focusing on the signaling induced by BCR-FcRIIB coligation, because the attenuated inhibitory effect of c-232Thr compared with that of c-232Ile on BCR-mediated calcium mobilization was consistently observed regardless of the BCR ligation conditions (Fig. 3). The most upstream event examined was time-dependent changes in PIP3 accumulation. Cells were labeled with 32P-phosphorus, 32P-labeled phospholipids were extracted and the changes in the distribution of radioactivity among phospholipids were analyzed by thin-layer chromatography. As shown in Figure 4C, PIP3 levels in c-232Thr-expressing cells were found to be significantly higher than those in c-232Ile-expressing cells under resting conditions and at an early period after BCR-FcRIIB coligation (0.5 min). The activations of Akt and PLC2, which are in part dependent on PIP3 levels and the determinants of B cell survival and calcium signaling, respectively (32), were next examined by immunoblotting using phosphorylation site-specific Abs. As shown in Figure 4D, the magnitudes of Akt and PLC2 activations induced by BCR-FcRIIB coligation were reproducibly greater in c-232Thr-expressing cells than those in c-232Ile-expressing cells. To further examine the impact of c-232Ile and c-232Thr on the signaling at lipid rafts, we prepared combined raft and non-raft fractions before and 4 min after BCR-FcRIIB coligation as indicated in Figure 2, and PLC2 activation was assessed in the separated fractions using phosphorylation site-specific Abs. As shown in Figure 4E, the magnitude of PLC2 activation at lipid rafts was greater in c-232Thr-expressing cells than that in c-232Ile-expressing cells, thereby suggesting that the augmented signaling at lipid rafts in c-232Thr cells likely explain the net increase in PLC2 activation in the whole cell lysates (Fig. 4D) as well as enhanced calcium mobilization in c-232Thr cells (Fig. 3).
Ile232Thr substitution-mediated functional alterations of FcRIIB in the context of full-length human FcRIIB constructs
Because it is theoretically possible that the functional differences between c-232Ile and c-232Thr are dependent on the chimeric structure, we attempted to examine the effects of Ile232Thr substitution on the functions of human full-length FcRIIB1. To this end, we prepared a recombinant retrovirus encoding h-232Ile or h-232Thr to avoid possible clonal variations. ST486 cells were infected with the viruses, and the cell populations were subjected to selection with puromycin as described in Materials and Methods. As shown in Figure 5A, the levels of h-232Ile and h-232Thr expressions in the cell populations were equivalent, suggesting that the Ile232Thr substitution does not critically impair protein stability or the membrane sorting of FcRIIB.
|
We first sought to compare the distributions of h-232Ile and h-232Thr to lipid rafts. Owing to the insufficient affinity of Fab fragments prepared from FLI8.26 and AT-10 for density gradient centrifugation assay (see above), we tested the use of 125I-AT10 whole IgG as the probe of human Fc
RIIB and found that 125I-AT10 does not in itself induce such unacceptable levels of FcRIIB redistribution to lipid rafts as previously noted in the case of using 2.4G2 as the probe of mouse FcRIIB [(27) and data not shown]. We thus decided to use 125I-AT10 as the probe and analyzed h-232Ile and h-232Thr distribution to lipid rafts under various stimulation conditions, including BCR self-ligation, FcRIIB self-ligation and BCR-FcRIIB coligation. To this end, h-232Ile and h-232Thr were probed with 125I-AT10, and BCR with or without biotinylated DA4-4 Fab, and the Abs were ligated with a secondary Ab or with streptavidin, as indicated in Figure 5B. Cells were subjected to density gradient centrifugation assay, and 125I radioactivities recovered from lipid raft and non-raft fractions were measured as described above (Fig. 2). As shown in Figure 5B, a small amount of FcRIIB was associated with lipid rafts under resting conditions, and BCR self-ligation increased the association to 1.5 times the basal level. FcRIIB self-ligation and BCR-FcRIIB coligation markedly enhanced FcRIIB distribution to lipid raft fractions four to five times the basal level. The amount of h-232Thr associated with lipid rafts was reproducibly smaller than that of h-232Ile under resting conditions after BCR self-ligation, after FcRIIB self-ligation and after BCR-FcRIIB coligation. These results unequivocally showed that Ile232Thr substitution attenuates the association of human FcRIIB with lipid rafts.
We next examined BCR-induced calcium mobilization. In this study, we frequently experienced difficulty in detecting IgG Fc-mediated inhibition by human and chimeric FcRIIB (Fig. 3B), in agreement with a previous indication by Van Den Herik-Oudijk et al. (33) that BCR signaling is less efficiently inhibited by coligation with human FcRIIB than by that with mouse FcRIIB. To circumvent the difficulty, we designed additional procedures that coligate FcRIIB and BCR in a more stringent and polyvalent manner via biotin–streptavidin interaction. Cells were loaded with Fluo-3, and BCR and FcRIIB were probed with a fixed concentration (1 µg/ml) of biotinylated DA4-4 Fab and with increasing concentrations of biotinylated AT10 (0–1 µg/ml) to allow the variation of number of FcRIIB molecules coligated to BCR. These Abs were ligated with an excess amount of streptavidin (100 µM), and calcium mobilization was monitored with Flexstation fluorometric scanning plate reader (Molecular Devices). The average peak [Ca2+]i (n=6) was found to be significantly greater in h-232Thr-expressing cells than in h-232Ile-expressing cells when BCR was self-ligated (AT10=0) (Fig. 6), findings in good agreement with those in c-232Ile- and c-232Thr-expressing cells (Fig. 3). We also detected a similar tendency when a low concentration of biotinylated AT10 (0.04 µg/ml) was used, although the difference was not significant (P=0.08). A prominent inhibition of calcium mobilization was achieved at higher concentrations of biotinylated AT10, but difference was not detected between the cells expressing FcRIIB variants. These findings confirmed that Ile232Thr substitution decreased the inhibition of calcium mobilization by FcRIIB when BCR is self-ligated, and suggested that under the coligation conditions, the differential inhibitory effects are evident only when a small number of FcRIIB molecules were coligated with BCR.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Previous studies using various mouse models have shown the critical roles of Fc
RIIB in preventing autoimmunity in vivo (7,9,10). We have for the first time identified a non-synonymous polymorphism Ile232Thr in FcRIIB and found its association with SLE susceptibility in Asians (21–23). So far, statistically significant association of the polymorphism with SLE susceptibility has not been detected in Caucasians or in African Americans (34,35). These apparent discrepancies may result from low frequency of 232Thr allele in Caucasians (34,35), or from allelic heterogeneity in other genes closely interacting with FcRIIB with respect to SLE susceptibility. Such epistatic interaction has been shown in mouse models of SLE (10) and we recently described an example in humans in which CD72 polymorphisms profoundly influence the association of FCGRIIB 232Thr with SLE susceptibility (36). To evaluate the possible contribution of 232Thr allele in the pathogenesis of SLE, precise functional studies of the allele products are required. Thus, we examined the properties of human FcRIIB1 232Ile and 232Thr by taking advantage of ST486 human B cell line lacking endogenous FcRIIB.
One of the unexpected findings of this study is that the expression of human FcRIIB1, particularly that of 232Ile, induced a considerable level of inhibition of calcium mobilization under the BCR self-ligation conditions. Such an indirect or constitutive inhibition of BCR signaling has frequently been shown in other inhibitory co-receptors in mouse B cells including CD22 and CD72 (7,37), but has not been clearly described in mouse FcRIIB (7,9,38,39). We also noticed moderate, but noticeable levels of FcRIIB tyrosine phosphorylation and SHIP association with FcRIIB after BCR self-ligation, which is reminiscent of the CD22- and CD72-SHP-1 systems (37). It indicates that under the present experimental conditions, BCR-alone signaling could transphosphorylate FcRIIB. In addition, we experienced difficulty in detecting such distinct levels of IgG Fc-mediated inhibition of BCR-induced calcium mobilization as encountered in the studies of mouse FcRIIB (7,9,38,39) (Fig. 3B; data not shown). The apparent inconsistency with previous studies with respect to the magnitudes of the constitutive and IgG Fc-mediated inhibition of BCR signaling may be simply due to different experimental conditions including the expression levels of FcRIIB. Alternatively, in view of the modest homology between human FcRIIB1 and the mouse homologue (24), it is also possible that mouse and human FcRIIB differently transduce constitutive and IgG Fc-mediated inhibition. Although the available information on human FcRIIB is limited, Van Den Herik-Oudijk et al. (33) also suggested that the coligation-mediated inhibition of BCR signaling by human FcRIIB is less efficient than that by mouse FcRIIB. To address the possible species differences, studies are now being undertaken in our laboratory to precisely compare the properties of human and mouse FcRIIBs.
In view of the significant difference in the levels of constitutive inhibition mediated by the human FcRIIB variants, we separately compared the BCR-alone-mediated and BCR-FcRIIB coligation-mediated calcium mobilization, and concluded that the Ile232Thr substitution attenuates the inhibitory effects of FcRIIB. A detailed examination of BCR-mediated biochemical signaling demonstrated that the attenuated inhibitory effects of 232Thr were evident at the level of PIP3 metabolism. The differential inhibitory effects were overridden by excessive FcRIIB coligation to BCR with a biotin–streptavidin system, indicating that the role of 232Ile is regulatory rather than obligatory for the inhibitory function of FcRIIB.
Considering that FcRIIB 232Thr was less preferentially distributed to lipid rafts than 232Ile both after BCR self-ligation and after BCR-FcRIIB coligation, it is tempting to hypothesize that the Ile232Thr substitution decreases the amounts of FcRIIB SHIP complex associated with lipid rafts, and enhances PIP3 accumulation. It was also observed that FcRIIB 232Thr was less tyrosine-phosphorylated than FcRIIB 232Ile after BCR-FcRIIB coligation, thereby suggesting that Ile232Thr substitution renders the FcRIIB less accessible to Lyn (16,17), at least under the coligation conditions. These findings strongly suggest a causal link between the level of FcRIIB association with lipid rafts and the extent of the inhibition of BCR signaling. However, this hypothesis should be further investigated by experiments employing other FcRIIB mutants that are excluded from lipid rafts. Our preliminary experiments utilizing alanine-scanning mutagenesis have indicated that two additional FcRIIB mutants are partially excluded from lipid rafts, and concurrently show decreased levels of inhibition on BCR-mediated calcium mobilization (40). We are now investigating further the properties of artificial FcRIIB mutants.
Recently, Li et al. (34) have reported a functional analysis of FcRIIB Ile232Thr polymorphism using the IIA1.6 mouse B cell line. They compared the ratios of the decrease in BCR-mediated calcium mobilization and CD19 tyrosine phosphorylation by IgG Fc-mediated inhibition, and concluded that FcRIIB 232Thr more effectively inhibits BCR signaling than 232Ile (34). The apparent discrepancy between their conclusions (34) and ours (this study) could be due to the different behavior of human protein with one amino acid substitution in the context of human and mouse B cells, but careful reading of their data reveals intriguing similarity between their observations with ours, suggesting that these are not indeed contradictory. Thus, in the study of Li et al., the magnitudes of calcium mobilization and CD19 tyrosine phosphorylation, particularly those induced by BCR self-ligation, appear to be significantly higher in cells expressing FcRIIB 232Thr than in those expressing 232Ile. Therefore, it is possible that human FcRIIB 232Ile exerted a greater constitutive inhibition than 232Thr in IIA1.6 mouse B cells as well, and that the entire inhibitory effect of FcRIIB 232Ile was underestimated in the study by Li et al., because the constitutive inhibition of FcRIIB was not taken into consideration in their study.
This study may provide the first example of a disease-associated receptor polymorphism that affects the distribution of the receptor to lipid rafts, and simultaneously impairs the receptor functions. Admittedly, the functional differences resulted from the transmembrane amino acid substitution are partial and appear too modest to account for disease susceptibility when compared with the often extreme outcomes observed in genetically manipulated mouse models. However, the extents of the functional alterations caused by human polymorphic genes associated with multifactorial diseases are quantitatively at the similar level of the present results (41–44). Therefore, the partial effect of Ile232Thr substitution does not seem in itself to preclude the possible contribution of the polymorphism to the disease susceptibility. From mechanistic points of view, it is the focus of considerable interests how the Ile232Thr substitution decreases the affinity of FcRIIB to lipid rafts. In the case of the influenza HA protein, large hydrophobic amino acids including Ile and Leu seem to be critical for the association with lipid rafts (45). The potential roles of 232Ile in protein–protein or protein–lipid interaction should be further studied.
Recently, a novel human FcRIIB promoter polymorphism has been reported to be associated with SLE in Caucasian (46,47). In contrast to the FcRIIB promoter polymorphism in autoimmunity-prone mouse strains (48,49), the promoter haplotype associated with human SLE was found to enhance FcRIIB expression (47). Therefore, it is of interest to analyze the linkage disequilibrium between the two polymorphisms in the promoter and in exon 5 encoding the transmembrane of FcRIIB gene to consider potential functional interaction between the two independent functional polymorphisms. Alternatively, it may also be possible that different functional polymorphisms in other genes contribute to the susceptibility to SLE in different populations depending on their genetic background, for instance, through epistatic interaction (36). Such possibilities should be validated by further genetic and functional studies.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Cells and expression constructs
The ST486 human B cell line was maintained in Hybridoma SFM medium (Gibco/Invitrogen) supplemented with 10% FCS and antibiotics. cDNAs encoding mouse Fc
RIIB1 and human FcRIIB1 232Ile and 232Thr (referred to as h-232Ile and h-232Thr, respectively) were amplified from the mouse cDNA library (Clonetech) and from cDNAs obtained from genotyped human donors (21), and subcloned into pBluescript (Stratagene). To prepare chimeric constructs containing the mouse FcRIIB extracellular domain and human FcRIIB 232Ile or 232Thr transmembrane-cytoplasmic domain, the mouse and the human full-length cDNAs were first subcloned into pBluescript in tandem in this order, and subjected to the reverse direction PCR (50) using primers 5'-TGGAACCTGCTTTTTCTTGA-3' and 5'-TAGTCTCCCTTGGCGAATTC-3'. The resulting chimera constructs in which amino acids 1–210 of mouse FcRIIB were fused to amino acids 219–310 of human FcRIIB 232Ile or Thr (referred to c-233Ile or c-232Thr; Fig. 1A) were confirmed by DNA sequencing. The cDNAs encoding the human and chimeric FcRIIB variants were subcloned into the pCAGGS expression vector (51) and transfected by electroporation into ST486, and stable transformants were generated as described previously (27). The expression of the chimeric and human FcRIIB variants was examined by flow cytometry (EPICS XL SYSTEM II, Coulter) using FITC-conjugated 2.4G2 anti-mouse FcRII/III (Pharmingen) and FITC-conjugated FLI8.26 anti-human FcRII (Pharmingen), respectively. We also introduced human FcRIIB with the aid of an ecotropic retrovirus system. We first transfected the pSSR-
-blasticidin vector encoding the ecotropic retrovirus receptor (EcoRVR) into ST486 cells to facilitate ecotropic retrovirus infection (52). Human FcRIIB 232Ile and 232Thr cDNAs subcloned into the pMSCVpuro retrovirus vector (Clonetech) were transfected using Lipofectamine 2000 (Invitrogen) into PLAT-E packaging cells (52), and the culture supernatants were obtained. The virus stock was used to infect ST486 cells expressing EcoRVR (52). The bulk of the infected ST486 cells were selected with 0.8 µg/ml puromycin, and stable ST486 infectants were obtained.
Competitive fluorescent ligand-binding assay
Because conventional 125I-labeled ligand binding could not be applicable due to a high non-specific binding, particularly that to dead cells (data not shown), we utilized FITC-conjugated human IgG (mixture of subclasses purified from human serum, Sigma) for whole cell binding assay. Cells were incubated with 10–7 M FITC-conjugated human IgG (fluorescein/IgG molar ratio: 3.1) in the absence or presence of various concentrations (0–3.2x10–6 M) of unlabeled human IgG in PBS for 1 h on ice, washed once with ice-cold PBS and subjected to flow cytometry. The high non-specific binding activity to dead cells could be avoided by appropriate gating. Quantum fluorescence microbeads (Sigma) conjugated with four different numbers of FITC were also analyzed as a reference, and a regression line between MFIs and absolute FITC numbers was drawn. MFI of the samples was thus converted to absolute FITC-IgG number bound to one cell, and expressed as mol of IgG/cell. The binding data are subjected to Scatchard analysis, and least-square analysis was used to form regression lines.
Antibodies used for probing or ligation of BCR and FcRIIB
Anti-human IgM rabbit whole IgG and anti-human IgM rabbit IgG F(ab')2 were obtained from Jackson Immunoresearch. Before use, these Abs were subjected to gel electrophoresis followed by Coomassie Brilliant Blue staining to exclude the contamination of F(ab')2 with whole IgG. Hybridoma cell lines producing DA4-4 (anti-human IgM mouse IgG) and that producing 2.4G2 were obtained from ATCC. DA4-4 and 2.4G2 mAbs were purified from culture supernatants. The preparation of the Fab fragments of DA4-4 and 2.4G2, and the biotinylation of the Fab fragments were described previously (27). The 125I labeling of 2.4G2 Fab and AT-10 (anti-human FcRII mouse IgG, Santa Cruz) was conducted using IODO beads (Pierce) according to the manufacturer's instructions.
Calcium mobilization assay
ST486 cells were loaded with Fura-2AM (Dojindo, Tokyo) in HEPES-Tyrode's solution with 1.8 mM CaCl2, 0.5 mM MgCl2 and 0.1% fatty acid-free BSA (loading buffer) as described previously (27), and changes in intracellular calcium concentration ([Ca2+]i) were measured in the same buffer using a fluorescence spectrophotometer F-2000 (Hitachi, Japan). Cells at 1.5x106 ml–1 were stimulated at room temperature with anti-human IgM rabbit whole IgG at 8 µg/ml or with F(ab')2 at 13 µg/ml, concentrations that induce comparable responses in parent ST486 cells. In some experiments, [Ca2+]i was measured using a FlexStation 96-well fluorometric imaging plate reader (Molecular Devices). In this setting, cells were loaded with Fluo-3AM (Dojindo, Tokyo) in the loading buffer containing 1% FCS, 0.02% Pluoronic acid (Molecular probe) and 2.5 mM Probenecid (Sigma) for 30 min at room temperature, washed once and plated at 1.5x105 per well (100 µl) in 96-well clear-bottomed black microplates (Corning). Cells were set in the device and preincubated for 15 min at 37°C. Then cells were treated with 1 µg/ml biotinylated DA4-4 Fab and with various concentrations (0–1 µg/ml) of biotinylated AT10 for 5 min, stimulated with 100 µM streptavidin, and the changes in fluorescence intensity were measured from the bottom with excitation at 485 nm and emission at 525 nm. Relative fluorescence unit (RFU) data are exported from the FLEXstation and converted into estimates of [Ca2+]i using the formula: [Ca2+]i=Kd(F–Fmin)/(Fmax–F), where the Kd for Fluo-3 is 390 nM. Fmax was recorded in the presence of 10 µM ionomycin, 1 µM thapsigargin and 40 mM CaCl2, and Fmin in the presence of 10 µM ionomycin, 1 µM thapsigargin and 10 mM EGTA.
Analysis of subcellular distribution of FcRIIB
ST486 cells expressing chimeric FcRIIB variants (1.5x107 ml–1) were incubated with 2 µg/ml 125I-2.4G2 Fab and 4 µg/ml DA4-4 Fab in the loading buffer for 30 min on ice, washed once, prewarmed at 37°C for 5 min, and left untreated or treated with 10 µg/ml secondary Abs (anti-rat IgG, Cappel) reactive to both the Fabs for 4 min. Cells were solubilized with 1.0 ml of ice-cold 1% Brij58 lysis buffer (28) and subjected to sucrose density gradient separation as described previously (28). Ten fractions (1 ml) were sequentially collected from the bottom and 125I radioactivity was determined by -ray counting. In the case of full-length human FcRIIBs, cells were treated with 2 µg/ml 125I-AT10 whole IgG and with or without 4 µg/ml biotinylated DA4-4 Fab as described above and BCR self-ligation, FcRIIB self-ligation and BCR-FcRIIB coligation were conducted using 10 µg/ml secondary Abs (anti-mouse IgG, Cappel) or 100 µM streptavidin for 4 min, as detailed in Figure 5. Cells were solubilized and subjected to density gradient centrifugation assay as described above.
Cell stimulation and biochemical analysis
ST486 cells expressing the chimeric or human FcRIIB variants (1x107 ml–1) were stimulated with anti-human IgM whole IgG (8 µg/ml) or with F(ab')2 (13 µg/ml) for indicated periods at 37°C, collected by brief centrifugation (30 s) in a microcentrifuge, solubilized with 2% n-octyl-ß-D-glucoside lysis buffer (20 mM Tris–HCl pH 7.4, 2% n-octyl-ß-D-glucoside, 150 mM NaCl, 1 mM EDTA, 1 mM Na3VO4, 20 mM ß-glycerophosphate and protease inhibitors), and centrifuged at 12 000 rpm (10 000 g) for 10 min at 4°C to obtain the supernatant (total cell lysate). Chimeric FcRIIB was immunoprecipitated from the total cell lysate with 2.4G2, and subjected to immunoblotting with 4G10 antiphosphotyrosine mAbs or with anti-SHIP1 Abs (Santa Cruz), and after stripping, with anti-human FcRIIB cytoplasmic antibody (Santa Cruz). For the analysis of CD19 tyrosine phosphorylation, cells were stimulated for 1 min as stated above, and solubilized with 1% NP-40 and 0.5% deoxycholate lysis buffer (20 mM Tris–HCl pH 7.4, 1% NP-40, 0.5% deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM Na3VO4, 20 mM ß-glycerophosphate and protease inhibitors). CD19 was immunoprecipitated from the cleared supernatant (total cell lysate) with anti-CD19 polyclonal antibody (Santa Cruz). The immunoprecipitates were divided into two halves: one was subjected to immunoblotting with 4G10 anti-phosphotyrosine mAbs and the other to that with anti-CD19 Abs (Santa Cruz). The total cell lysate was also subjected to immunoblotting with anti-phospho Akt (p-Ser473) Abs and anti-phospho PLC2 (p-Tyr1217) Abs (Cell Signaling Technology), and reprobed with anti-Akt and anti-PLC2 Abs (Santa Cruz) after stripping. Signals were detected using HRP-conjugated secondary Abs and an ECL chemiluminescent system (Amersham). Cell labeling with [32P]-phosphorus (Amersham) and phospholipid analysis by thin-layer chromatography were conducted as described previously (12).
Statistical analyses
Statistical analyses were performed using Student's t-test.
|
ACKNOWLEDGEMENTS |
|---|
We thank Dr M. Okada (Department of Oncogene Research, Research Institute for Microbial Diseases, Osaka University) for the anti-Cbp antibody, Dr T. Kitamura (Division of Cellular Therapy, Institute of Medical Science, University of Tokyo) for PLAT-E cells and ecotropic retrovirus receptor cDNA, and H. Ichijo-Ota for technical assistance. This work was supported by grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and from the Ministry of Health, Labor and Welfare of Japan and by Takeda Science Foundation.
Conflict of Interest statement. The authors have had no involvements that might raise the question of bias in the work reported or in the conclusions, implications or opinions stated.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
- Wakeland, E.K., Liu, K., Graham, R.R. and Behrens, T.W. (2001) Delineating the genetic basis of systemic lupus erythematosus. Immunity, 15, 397–408.[CrossRef][ISI][Medline]
- Kelly, J.A., Moser, K.L. and Harley, J.B. (2002) The genetics of systemic lupus erythematosus: putting the pieces together. Genes Immun., 3, S71–S85.[CrossRef][ISI][Medline]
- Tsuchiya, N., Ohashi, J. and Tokunaga, K. (2002) Variations in immune response genes and their associations with multifactorial immune disorders. Immunol. Rev., 190, 169–181.[CrossRef][ISI][Medline]
-
Moser, K.L., Neas, B.R., Salmon, J.E., Yu, H., Gray-McGuire, C., Asundi, N., Bruner, G.R., Fox, J., Kelly, J., Henshall, S. et al. (1998) Genome scan of human systemic lupus erythematosus: evidence for linkage on chromosome 1q in African-American pedigrees. Proc. Natl Acad. Sci. USA, 95, 14869–14874.
[Abstract/Free Full Text] -
Shai, R., Quismorio, F.P., Jr, Li, L., Kwon, O.J., Morrison, J., Wallace, D.J., Neuwelt, C.M., Brautbar, C., Gauderman, W.J. and Jacob, C.O. (1999) Genome-wide screen for systemic lupus erythematosus susceptibility genes in multiplex families. Hum. Mol. Genet., 8, 639–644.
[Abstract/Free Full Text] - Daeron, M. (1997) Fc receptor biology. Annu. Rev. Immunol., 15, 203–234.[CrossRef][ISI][Medline]
-
Ravetch, J.V. and Lanier, L.L. (2000) Immune inhibitory receptors. Science, 290, 84–89.
[Abstract/Free Full Text] - Takai, T. (2002) Roles of Fc receptors in autoimmunity. Nat. Rev. Immunol., 2, 580–592.[ISI][Medline]
- Takai, T., Ono, M., Hikida, M., Ohmori, H. and Ravetch, J.V. (1996) Augmented humoral and anaphylactic responses in Fc gamma RII-deficient mice. Nature, 379, 346–349.[CrossRef][Medline]
- Bolland, S. and Ravetch, J.V. (2000) Spontaneous autoimmune disease in Fc gamma RIIB-deficient mice results from strain-specific epistasis. Immunity, 13, 277–285.[CrossRef][ISI][Medline]
- Ono, M., Bolland, S., Tempst, P. and Ravetch, J.V. (1996) Role of the inositol phosphatase SHIP in negative regulation of the immune system by the receptor Fc gamma RIIB. Nature, 383, 263–266.[CrossRef][Medline]
- Scharenberg, A.M., El-Hillal, O., Fruman, D.A., Beitz, L.O., Li, Z., Lin, S., Gout, I., Cantley, L.C., Rawlings, D.J. and Kinet, J.P. (1998) Phosphatidylinositol-3,4,5-trisphosphate (PtdIns-3,4,5-P3)/Tec kinase-dependent calcium signaling pathway: a target for SHIP-mediated inhibitory signals. EMBO J., 17, 1961–1972.[CrossRef][ISI][Medline]
- Bolland, S., Pearse, R.N., Kurosaki, T. and Ravetch, J.V. (1998) SHIP modulates immune receptor responses by regulating membrane association of Btk. Immunity, 8, 509–516.[CrossRef][ISI][Medline]
- Simons, K. and Ikonen, E. (1997) Functional rafts in cell membranes. Nature, 387, 569–572.[CrossRef][Medline]
- Pizzo, P. and Viola, A. (2003) Lymphocyte lipid rafts: structure and function. Curr. Opin. Immunol., 15, 255–260.[CrossRef][ISI][Medline]
-
Timson Gauen, L.K., Linder, M.E. and Shaw, A.S. (1996) Multiple features of the p59fyn src homology 4 domain define a motif for immune-receptor tyrosine-based activation motif (ITAM) binding and for plasma membrane localization. J. Cell Biol., 133, 1007–1015.
[Abstract/Free Full Text] -
Honda, Z.-I., Suzuki, T., Kono, H., Okada, M., Yamamoto, T., Ra, C., Morita, Y. and Yamamoto, K. (2000) Sequential requirements of the N-terminal palmitoylation site and SH2 domain of Src family kinases in the initiation and progression of FcepsilonRI signaling. Mol. Cell. Biol., 20, 1759–1771.
[Abstract/Free Full Text] - Pierce, S.K. (2002) Lipid rafts and B cell activation. Nat. Rev. Immunol., 2, 96–105.[CrossRef][ISI][Medline]
-
Cheng, P.C., Brown, B.K., Song, W. and Pierce, S.K. (2001) Translocation of the B cell antigen receptor into lipid rafts reveals a novel step in signaling. J. Immunol., 166, 3693–3701.
[Abstract/Free Full Text] - Aman, M.J., Tosello-Trampont, A.C. and Ravichandran, K. (2001) Fc gamma RIIB1/SHIP-mediated inhibitory signaling in B cells involves lipid rafts. J. Biol. Chem., 276