Unique Ligand-Binding Property of the Human IgM Fc Receptor

Antibodies have dual binding activity: to Ag via their N-terminal variable regions, and to effector molecules such as FcRs via their C-terminal constant regions. FcRs are expressed by many different cell types, and their interaction with Abs can initiate a broad spectrum of effector functions essential in host defense. These functions include phagocytosis of Ab-coated microbes, lysosomal degradation of endocytosed immune complexes, Ab-dependent cell-mediated cytotoxicity, secretion of cytokines and chemokines, release of potent inflammatory mediators, regulation of Ab production by B cells, survival of plasma cells, and presentation of degradable as well as nondegradable Ags (1–7). These diverse functions depend on the Ab isotype and the cell type expressing the FcR. Structurally and functionally diverse FcRs, namely FcR for IgG (FcγRI/CD64, FcγRII/CD32, FcγRIII/CD16, FcγRIV, FcRn), IgE (FcεRI, FcεRII/CD23), IgA (FcαR/CD89), and both polymeric IgA and IgM (Fcα/μR/CD351), have been characterized extensively at both the protein and genetic levels (1–5, 8–10).

It has long been a puzzle why an FcR for IgM (FcμR), the first Ig isotype to appear during phylogeny, ontogeny, and the immune response, has defied identification, despite extensive biochemical evidence of IgM Fc–binding proteins accumulated over decades (11–13). We previously successfully identified a cDNA encoding an authentic FcμR from cDNA libraries of human B-lineage cells using a functional cloning strategy (14). FCMR is a single-copy gene located on chromosome 1q32.2, adjacent to two other IgM-binding receptor genes, Fcα/μR and polymeric Ig receptor. The predicted FcμR is a transmembrane protein that consists of a single V-set Ig-like domain responsible for Fcμ binding, an additional extracellular region with no known domain structure, a transmembrane segment containing a charged His residue, and a relatively long cytoplasmic tail (118 aa) containing three conserved Tyr and five conserved Ser residues. FcμR binds pentameric IgM with a surprisingly high avidity of ∼10 nM as determined by Scatchard plot analysis, with the assumption of a 1:1 stoichiometry of FcμR to IgM ligand. Upon ligation of FcμR with IgM ligands, both Tyr and Ser residues in the cytoplasmic tail are phosphorylated (14), and receptors are rapidly internalized into lysosomal compartments (15). Unlike other FcRs, the expression of FcμR is restricted to lymphocytes: B, T, and NK cells (14, 16). This suggests potentially distinct functions of FcμR as compared with other FcRs that are mainly expressed by myeloid cells.

Alternatively, the FcμR was initially designated as Fas apoptotic inhibitory molecule 3 (FAIM3), because coligation of Fas and FcμR/FAIM3 with an agonistic IgM anti-Fas mAb prevented Fas-mediated apoptosis (17). Unlike the effect of IgM anti-Fas mAb, however, ligation of Fas with an agonistic IgG mAb induced apoptosis irrespective of the expression of FcμR/FAIM3 (14, 16, 18). Notably, coligation of Fas and FcμR/FAIM3 with the corresponding mouse IgG mAbs plus a secondary reagent [e.g., F(ab′)₂ fragments of anti-mouse γ Ab] had no demonstrable effects on the IgG anti-Fas mAb-induced apoptosis (14). This suggests that the antiapoptotic activity of FcμR/FAIM3 depends on usage of the IgM anti-Fas mAb, and not on physical proximity of two receptors induced by artificial coligation. However, because the FcμR-mediated inhibition of the IgM anti-Fas mAb-induced apoptosis is a reproducible and well-defined function, we employed this antiapoptotic assay in the present study as one of the readouts to explore the mode of action of FcμR. In this study, we have defined the unique ligand-binding properties of FcμR and the key residues involved in its receptor function.

The human leukemic T cell (Jurkat) and mouse thymoma (BW5147) lines, which stably expressed either only GFP as a control or both GFP and human FcμR, were prepared by retrovirus-mediated transduction as previously described (14) and were maintained in media containing puromycin at 1.5 and 0.75 μg/ml, respectively. BW5147 cells stably expressing human FcμR (without GFP) and wild-type (WT) Jurkat or BW5147 cells were maintained in media only as described (14). For Ca²⁺ mobilization, mononuclear cells isolated from healthy individuals by Ficoll-Hypaque density gradient centrifugation were enriched for B cells using a B cell isolation kit II (Miltenyi Biotec) with MACS. The frequency of CD19⁺ B cells in the resultant enriched fractions was >95% as determined by flow cytometry. This study involving human subjects was approved by the institutional review board of the University of Alabama at Birmingham.

Flow cytometric analysis of cell surface FcμR on transductants was performed by using biotin-labeled, receptor-specific mAb (HM14, γ1κ isotype) and isotype-matched control mAb, along with PE-labeled streptavidin as described (14). For IgM binding, cells were incubated with biotin-labeled mouse myeloma IgM (TEPC183) at 1 μg/ml or PBS, washed, and then incubated with PE-labeled streptavidin. Stained cells were examined by an Accuri C6 flow cytometer (BD Biosciences), and flow cytometric data were analyzed with FlowJo software (Tree Star).

Apoptosis assays were performed essentially the same as described before (14). In brief, FcμR⁺ or control Jurkat cells were cultured for 15 h with or without agonistic IgM anti-Fas mAb (CH11, 10 ng/ml; Millipore), along with various concentrations of inhibitors: 1) IgM mAbs specific for nitrophenyl (NP) hapten (B1-8) (19) or α1-3 dextran (clone 1-21) (20), 2) their soluble immune complexes with NP₂₅-coupled BSA (molecular mass of ∼70 kDa; Biosearch Technologies) or α1-3 dextran (molecular mass of ∼1000 kDa), 3) FcμR-blocking (HM7) or nonblocking (HM3) mAbs (both γ2bκ), and 4) IgM mAb specific for the human CD2 (C373) or Jurkat TCR (C305) (21). The latter two mAbs were provided by Dr. Arthur Weiss (Howard Hughes Medical Institute, University of California, San Francisco) and were purified from ascites by Sephacryl S300 gel filtration, and the remainders were purified by Ag-coupled affinity columns. For soluble immune complexes, IgM anti-NP and IgM anti-dextran mAbs and the corresponding Ags were premixed to reach a final concentration in the cultures of 10⁴ or 10⁵ ng/ml for IgM mAbs with different molar ratios of Ab/Ag (1:10 to 1:10⁻²). Apoptotic cells were detected by staining with 7-aminoactinomycin D (7-AAD) and allophycocyanin-labeled annexin V (BD Biosciences) as described (14). For mixed cell cultures, FcμR⁺ or control Jurkat cells were cocultured with 10-fold excess number of FcμR⁺ BW5147 cells in the presence or absence of CH11 anti-Fas mAb, followed by cell surface staining with PE–anti-mouse CD3 mAb (145-2C11; BD Biosciences). Apoptosis was assessed for Jurkat cell populations by excluding CD3⁺ live BW5147 cells.

A mixture containing equal numbers of FcμR⁺ and WT Jurkat cells preloaded with 5 μM Indo-1/AM (Molecular Probes), a Ca²⁺-sensitive dye, was stimulated by various concentrations of IgM mAbs specific for CD2 or TCR or by 1 μM ionomycin as a positive control. The intracellular Ca²⁺ concentrations ([Ca²⁺]_i) in both cell populations were simultaneously assessed by the ratio of emitted fluorescence of bound Ca²⁺ at 405 nm and free Ca²⁺ at 485 nm using an LSR II flow cytometer (BD Biosciences). In some experiments, an equal mixture of control GFP⁺ and WT Jurkat cells was also assessed similarly as controls. For blood B cells, MACS-enriched B cells (>95%) were loaded with Fluo-4/AM (0.25 μM; Molecular Probes), rested, and then treated with a mixture of mouse IgM anti-human κ mAb (END-5C1, 10 μg/ml) and mouse IgG2b anti-FcμR mAb (HM7 or HM3, 100 μg/ml). Changes in Fluo-4 fluorescence in the labeled cells following treatment were measured over real time with an Accuri C6 flow cytometer and analyzed by FlowJo software as described (22).

FcμR cDNA constructs with point mutations (H253F, Y315F, Y366F, or Y385F) were generated by a QuickChange site-directed mutagenesis kit (Stratagene) using nonmutated FcμR cDNA as a template DNA and a set of primers: 1) 5′-GAAGGCCAAGGATTTTTCATCCTGATCCCGACCAT-3′ and 5′-ATGGTCGGGATCAGGATGAAAAATCCTTGCCCTTC-3′ for H253F, 2) 5′-CTCCCAAAACAACATCTTCAGCGCCTGCC-3′ and 5′-GGCAGGCGCTGAAGATGTTGTTTTGGGAG-3′ for Y315F, 3) 5′-GACCAGCTGTGAATTTGTGAGCCTCTACCACCAG-3′ and 5′-CTGGTGGTAGAGGCTCACAAATTCACAGCTGGTC-3′ for Y366F, and 4) 5′-GGACAGTGATTCAGATGACTTCATCAATGTTCCTG-3′ and 5′-CAGGAACATTGATGAAGTCATCTGAATCACTGTCC-3′ for Y385F (underlining indicates mutation sites), according to the manufacturer’s recommendation. For FcμR cytoplasmic tail deletion, an FcμR cDNA construct with a deletion of most of the cytoplasmic tail (A281–A390; ΔCy) but with an inclusion of the posttransmembrane basic aa-rich region (K273–K280) was generated by PCR using PrimeStar HS DNA polymerase (TaKaRa), FcμR cDNA as a template DNA, and a set of primers: 5′-GAATTCTAGAAGGGACAATGGACT-3′ and 5′-GCGGCCGCCTATTTCCTCCTTTCAACGGCC-3′ (italics indicate EcoRI and NotI sites, and underlining indicates translation start and stop sites). The amplified PCR products of the expected size were gel purified and subcloned into the ZeroBlunt TOPO vector (Invitrogen). All plasmid DNAs encoding FcμR mutants were digested with appropriate restriction enzymes, gel purified, and ligated into the pMX-PIE retroviral expression vector as described (14). After confirming the correct sequences with the expected mutation of the resultant cDNAs at both strands of DNA, they were transfected into 293T-A packaging cell line before transducing Jurkat cells as described (14). GFP⁺ transductants were enriched by FACS before IgM binding, apoptosis, and Ca²⁺ mobilization analyses.

Viable cells were incubated for 20 min on ice with Alexa Fluor 555–labeled, affinity-purified mouse monoclonal IgM (MD4 anti-hen egg lysozyme mAb) at 10 μg/ml in PBS with or without 0.1% NaN₃, washed in PBS/1% FCS/0.1% NaN₃, and cytocentrifuged onto glass slides prior to being air-dried and fixed in 95% ethanol/5% glacial acetic acid at −20°C. Some aliquots of cells were resuspended in warmed RPMI 1640/10% FCS and incubated for an additional 10 min at 37°C before cytocentrifugation. The stained cells were examined under a Leica/Leitz DMRB microscope. Images were acquired with a Hamamatsu C4742 video camera and processed with Openlab software.

In our previous studies apoptosis-prone human Jurkat cells stably expressing FcμR were shown to be protected from Fas/CD95-mediated apoptosis when ligated with an agonistic IgM anti-Fas mAb (CH11), but not when ligated with an agonistic IgG anti-Fas mAb or Fas ligand (14, 18). Notably, coligation of FcμR and Fas with the corresponding IgG mAbs plus a common secondary reagent could not prevent IgG anti-Fas mAb-induced apoptosis (14). To determine whether this apoptosis protection mediated by FcμR is affected by IgM ligands, IgM mAbs specific for NP (clone B1-8) or α1-3 dextran (clone 1-21) and their immune complexes were initially used as inhibitors in the apoptosis assays. As expected, addition of IgM anti-Fas mAb at 10 ng/ml induced robust apoptosis of control GFP⁺ cells, but not GFP⁺/FcμR⁺ cells as determined by visualizing early (annexin V⁺/7-AAD⁻) and late (Annexin V⁺/7-AAD⁺) apoptotic cells (Fig. 1A, upper second column). The addition of both types of IgM mAbs at 10⁴ ng/ml (upper third column; not shown for anti-dextran mAb) or their soluble immune complexes (upper fourth and lower first columns) did not affect the FcμR-mediated protection, suggesting the inability of such IgM molecules to block the interaction of the Fc portion of IgM Fas mAb with FcμR. The failure to inhibit protection was observed with wide ranges of Ab/Ag molar ratios (Fig. 1B, Supplemental Fig. 1). However, addition of both types of IgM at 10⁵ ng/ml (Fig. 1B, Supplemental Fig. 1) and their immune complexes with (Fig. 1A, lower second column, 1B, Supplemental Fig. 1) partially but significantly blocked the FcμR-mediated protection. All three IgM preparations (CH11, B1-8, and 1-21) were found to bind equally to FcμR⁺ cells (not shown), ruling out the possibility that the CH11 IgM has a unique posttranslational modification to bind more efficiently the FcμR compared with others. In contrast, addition of FcμR-blocking IgG2b mAb (HM7), but not FcμR nonblocking IgG2b mAb (HM3), at 10⁴ ng/ml specifically inhibited the interaction of the Fc portion of IgM Fas mAb with FcμR, thus permitting the FcμR⁺ cells to become apoptotic (Fig. 1A, lower fourth and fifth columns). These findings indicate that soluble IgM immune complexes are not good competitors in the interaction of FcμR with the Fc portion of IgM anti-Fas mAb, although FcμR is shown by Scatchard plot analysis to bind pentameric IgM with a surprisingly high avidity of ∼10 nM (14). This also suggests the possibility that FcμR may bind more efficiently to the Fc portion of IgM, which is attached to plasma membranes via its Fab region, than to the Fc portion of free IgM or its immune complexes in solution.

To test this hypothesis, we employed additional IgM mAbs reactive with CD2 or TCR on the surface of Jurkat cells as potential competitors for the interaction of IgM Fas mAb with FcμR. Addition of IgM anti-CD2 mAb (C373) at 10³ ng/ml efficiently inhibited FcμR-mediated protection, thereby permitting the FcμR⁺ cells to undergo apoptosis (see Fig. 1A, lower third column, 1C). The anti-CD2 mAb alone did not induce apoptosis of either of cell types (not shown). Similar results were also obtained with an IgM anti-TCR mAb (C305, not shown). Thus, these findings suggest that FcμR binds more efficiently to IgM mAbs, which attach to cell surface components via their Fab regions, than to IgM immune complexes in solution.

To determine whether the engagement of FcμR with IgM attached to plasma membrane occurs in trans or cis (see Fig. 2A), we performed mixed cell cultures of FcμR⁺ (or control) Jurkat cells plus a 10-fold excess of mouse BW5147 cells stably expressing human FcμR. When the Fc portion of IgM anti-Fas mAb binds FcμR in trans between the excess of cocultured FcμR⁺ BW5147 cells, then FcμR⁺ Jurkat cells will become apoptotic. Alternatively, when the Fc portion of IgM anti-Fas mAb must bind FcμR in cis on the same cell surface to perform this protective function, FcμR⁺ Jurkat cells will still be protected against IgM anti-Fas mAb–mediated apoptosis even in the presence of a 10-fold excess number of FcμR⁺ BW5147 cells. As shown in Fig. 2B, addition of excessive FcμR⁺ BW5147 cells did not affect the FcμR-mediated protection from apoptosis (second versus fourth column), suggesting predominance of the protective mechanism involving a cis interaction of the Fc portion of IgM anti-Fas mAb with FcμR on the same cell surface. Addition of excessive FcμR⁺ BW5147 cells did not spontaneously induce apoptosis of either of FcμR⁺ or control Jurkat cells in the absence of anti-Fas mAb (third column). Collectively, these findings show that although FcμR binds soluble IgM pentamers at a high avidity (14), FcμR binds more efficiently to the Fc portion of IgM Ab when it reacts with a membrane component on the same cell surface. This finding implies that FcμR can modulate the functional activity of lymphocyte cell surface receptors or molecules recognized by either natural or immune IgM Abs.

To explore this possibility, we compared Ca²⁺ mobilization in cells with a single ligation of CD2 or TCR versus the coligation of both FcμR and CD2 or TCR. An equal mixture of WT and FcμR⁺ Jurkat cells was preloaded with the Ca²⁺ dye Indo-1/AM and simultaneously stimulated by various concentrations of IgM mAbs specific for CD2 or TCR. Although it has been well documented that induction of CD2-mediated Ca²⁺ mobilization in T cells requires ligation of CD2 with two different IgG mAbs (23–25), the IgM anti-CD2 mAb (C373) alone clearly induced Ca²⁺ mobilization (Fig. 3A). Notably, the rise in [Ca²⁺]_i occurred significantly faster in FcμR⁺ cells than in WT cells when ligated with the anti-CD2 mAb at three different concentrations (3, 10, and 30 μg/ml, data at 3 and 30 μg/ml are not shown). A similar difference, but to a much less extent, was also observed with the cross-linkage of TCR with IgM mAb (C305) (not shown). In contrast, the ionomycin-induced [Ca²⁺]_i rise occurred at the same time in both cell types. These findings suggest that the coligation of FcμR with CD2 or TCR induces a more rapid increase in [Ca²⁺]_i than does the ligation of CD2 or TCR alone.

Because we found that freshly prepared blood T cells, unlike Jurkat cells, were not activated by the C373 anti-CD2 mAb (not shown), we switched to using a mitogenic IgM anti-κ mAb (END-5C1) and examined Ca²⁺ mobilization induced by BCR cross-linkage in the presence of FcμR-blocking (HM7) or nonblocking (HM3) mAbs. As shown in Fig. 2B, IgM anti-κ mAb–induced Ca²⁺ mobilization was indistinguishable in the absence or presence of FcμR nonblocking mAb (HM3). In contrast, the presence of FcμR-blocking mAb (HM7) diminished the IgM anti-κ mAb–induced Ca²⁺ mobilization, suggesting that FcμR provides a stimulatory signal upon BCR cross-linkage with IgM mAbs.

In FcRs and in other paired Ig-like receptors, it is generally observed that when the ligand-binding α-chain has a short cytoplasmic tail with no signal-transmitting potential, then it contains a charged residue in the transmembrane segment that facilitates noncovalent association with another transmembrane protein–containing ITAMs. This association allows for the transmission of activating signals to cells as seen in FcγRI, FcγRIII, FcαR, and FcεRI. Alternatively, when the ligand-binding α-chain has a conventional hydrophobic transmembrane segment, then it usually has a long ITAM-containing cytoplasmic tail (26). This recruits the phosphatase SHIP upon phosphorylation to attenuate the signals as in FcγRIIB. However, FcμR is unusual in having both features: a charged His residue (H253) in the transmembrane region and a long cytoplasmic tail containing three conserved Tyr (Y315, Y366, Y385) and five conserved Ser residues, features that have been conserved in multiple species (see Fig. 4A) (14). These characteristics suggest a dual signaling ability of FcμR: one from a potential adaptor protein noncovalently associating with the FcμR via the H253 residue, and the other from its own Tyr and/or Ser residues in the cytoplasmic tail.

To determine which part of the FcμR molecule is responsible for these potential functions, we made FcμR cDNA constructs with point mutations (H253F, Y315F, Y366F, or Y385F) or a deletion of most of the cytoplasmic tail (A281–A390; ΔCy) and expressed them in Jurkat T cells. After enriching GFP⁺ cells by FACS, the surface expression of FcμR as determined by reactivity with receptor-specific mAbs was comparable among the nonmutant and mutant FcμR transductants (Fig. 4B), suggesting that any introduced mutation does not affect the surface expression of FcμR on Jurkat cells. Remarkably, however, the IgM ligand-binding activity was significantly increased by the ΔCy mutant (Fig. 4C, 4D), suggesting the influence of the FcμR cytoplasmic tail on its ectodomain-mediated ligand binding.

Next, the fate of IgM ligands after binding to FcμR on these transductants was examined by epifluorescence microscopy. IgM binding at 4°C in the presence of NaN₃, an inhibitor of energy-dependent contraction of actin filament, was observed in a weak patchy staining pattern at the peripheral rim of transductants regardless of their mutations. However, when incubated at 4°C without NaN₃, distinct IgM binding, marked by a more intense and localized or capping pattern, was observed with the H253F mutant (Fig. 4Eb) as compared with all others except for the ΔCy mutant, which exhibited a more broadly localized staining pattern (Fig. 4Ea for nonmutated). This suggests an important role of the His²⁵³ residue in the anchoring of FcμR in the transmembrane. After washing and incubating at 37°C for 10 min without NaN₃, the IgM/FcμR complexes were internalized and localized in intracellular locations, presumably endosomal or lysosomal compartments, in the FcμR nonmutated (Fig. 4Ec), H253F (Fig. 4Ed), and Y315F transductants (not shown), but not in other mutants (Y366F, Y385F, and ΔCy, not shown). These results are consistent with previous findings reported by others that two C-terminal–conserved Tyr residues were involved in receptor-mediated endocytosis (15).

The FcμR-mediated protection from apoptosis was significantly diminished in the FcμR Y315F and ΔCy mutants, as the frequencies of apoptotic cells in these mutants were indistinguishable from those in the control transductant (Fig. 4F). Other FcμR mutants (H253F, Y366F, and Y385F) still had significant protective activity compared with controls (GFP⁺), but the extent of their protective activity was significantly lower than that of the nonmutated FcμR. These findings suggest that the membrane-proximal Tyr residue (Y315) is largely responsible for, but other Tyr residues also contribute minimally to, the FcμR-mediated protection. Notably, the rapid rise in IgM anti-CD2 mAb–induced [Ca²⁺]_i observed in the FcμR nonmutant was abolished by all of the FcμR mutants, suggesting that both the cytoplasmic tail including all three Tyr residues and the transmembrane His residue are involved in signaling mediated by coligation of CD2 and FcμR.

The present study is focused on the IgM ligand-binding property of FcμR, and several remarkable features of the FcμR have been identified. First, although FcμR binds soluble pentameric IgM at a relatively high avidity of ∼10 nM (14), FcμR actually interacts more efficiently with the Fc portion of IgM Abs when bound to plasma membranes via their Ag-binding Fab regions, and this interaction occurs in cis on the same cell surface. This conclusion is based on the functional assays rather than actual ligand-binding assessments. Second, the Ca²⁺ mobilization induced by cross-linkage of CD2 or BCRs with IgM mAbs differs between the ligation of CD2 or BCR alone and the coligation of FcμR and CD2 or BCR. Third, several remarkable functional changes are observed with FcμR mutants: 1) increased IgM ligand binding by the FcμR ΔCy mutant, 2) enhanced cap formation of the FcμR H253F mutant even at 4°C upon IgM ligand binding, and 3) less protective activity of the FcμR Y315F mutant in IgM anti-Fas mAb–mediated apoptosis.

The preferential interaction of FcμR with IgM attached to plasma membrane via its Fab region was initially notified in our previous reconcilement of antiapoptotic activity of Toso, the original designation given to FcμR (14, 17). In the present study, we have defined more quantitatively the ligand-binding property of FcμR. The interaction of FcμR with the Fc portion of IgM anti-Fas mAb in Jurkat cells was not inhibited by addition of 10³-fold excess of IgM mAbs or their soluble immune complexes, but it was blocked by addition of 10⁴-fold excess of IgM mAbs or immune complexes, 10³-fold excess of FcμR-blocking mAb (HM7, γ2bκ isotype), or 10-fold excess of IgM anti-CD2 or anti-TCR mAb binding to the Jurkat cells. The interaction of FcμR with the Fc of IgM anti-Fas mAb occurred in cis on the same cell surface and not in trans between neighboring cells, as evidenced by the fact that adding 10-fold excess of FcμR⁺ BW5147 cells (of mouse origin) did not affect FcμR-mediated protection from IgM anti-Fas mAb–induced apoptosis. The preferential cis engagement of FcμR is thus distinct from the trans engagement of FcγRIIb, an inhibitory FcγR, in death receptor–mediated apoptosis. The interaction of agonistic IgG mAbs against death receptors, including Fas/CD95, with FcγRIIb is essential for the death receptor–mediated apoptosis and occurs in trans, but not in cis (27–29). The cis engagement of FcμR is significant in that FcμR has the potential to modulate the functional activity of lymphocyte surface receptors on the same cell when they are recognized by natural or immune IgM Abs. The physiological relevance of this cis engagement may be related to unique features of the IgM ligand and its receptor. In idiotype network theory, BCRs (or their secreted products) are proposed to be directed toward the variable regions of other BCRs or their Ig molecules. Two thirds of the newly generated B cells in bone marrow are now known to react with self-antigens such as dsDNA, ssDNA, insulin, and LPS (30). According to our estimates, approximately a fourth of the IgM-containing culture supernatants from 96 different EBV-transformed neonatal B cell lines react with lymphocyte surface components (H. Kubagawa, K. Honjo, and Y. Kubagawa, unpublished observations). IgM anti–lymphocyte Abs are often present in individuals with autoimmune diseases or chronic viral infections, such as HIV-1, and recognize many different surface Ags (e.g., CD45, CD175/Tn, CD3ε, CD4, chemokine receptors, sphingosine-1-phosphate receptor 1), and some of those Abs regulate T cell–mediated inflammatory responses in vitro (31–41). Unlike FcRs for switched Ig isotypes, the expression of FcμR is restricted to lymphocytes, namely, B, T, and NK cells (14). Therefore, it is highly probable that FcμR on B, T, and NK cells preferentially binds the Fc portion of IgM Ab reactive with their cell surface Ags or receptors, thereby modulating the function of their target Ags or receptors. Supporting this idea, the CD2-mediated Ca²⁺ mobilization after cross-linkage with IgM mAb was clearly faster in FcμR⁺ than control Jurkat cells. The Ca²⁺ mobilization induced by cross-linkage of BCR on blood B cells with IgM anti-κ mAb was significantly different in the presence of an FcμR blocker (HM7) versus an FcμR nonblocker (HM3). The HM7 blocker diminished BCR-mediated Ca²⁺ mobilization, whereas the HM3 nonblocker did not. Unlike the coligation of FcμR and Fas receptor, FcμR provides a stimulatory signal upon coligation with CD2 or BCR.

Mutational analyses of FcμR revealed several remarkable alterations in its function. Whereas nonmutated and mutant FcμR transductants expressed comparable levels of cell surface FcμR as judged by receptor-specific mAbs, the IgM ligand-binding activity was significantly increased in the ΔCy mutant. This suggests either the formation of oligomeric FcμR due to its presumably mobile nature within the plasma membrane or the inside–out regulation of ligand-binding activity of FcμR by its cytoplasmic tail. Given the precedent of inside–out signaling regulation in the integrin family of cell surface adhesion molecules (42) and the FcαR (43), both possibilities are equally conceivable and remain to be elucidated.

When the fate of IgM-bound FcμR was examined by immunofluorescence microscopy, enhanced cap formation was clearly observed with the H253F mutant even at 4°C without NaN₃ as compared with FcμR nonmutated or other mutant transductants, except the ΔCy mutant, which displayed a more broadly localized staining pattern. This finding suggests a functional role of His²⁵³ residue in the anchoring of FcμR in the transmembrane. Although we have not yet identified a membrane protein noncovalently associating with the ligand-binding chain of FcμR through its H253, there are several possible candidates, including the ITAM-containing CD3ζ, FcεRIγ, or DAP12, the PI3K-recruiting DAP10, or the unknown 40-kDa protein (p40), which was often coprecipitated with the 60-kDa FcμR in our previous analyses (14). Notably, unlike other multichain FcRs, the FcμR H253F mutant presented on the cell surface without a potentially associated membrane protein.

The findings that the Y315F and ΔCy mutants had no significant protective activity against IgM anti-Fas mAb–induced apoptosis is of interest, given the unique sequence around the Y³¹⁵ residue: ³⁰⁸P·R·S/T·Q·N·N·I/V·Y·S/T·A·C·P·R·R·A·R³²³ (bold type indicates conserved amino acids). This does not match any known Tyr-based signaling motifs. The mechanism of Toso/FcμR-mediated protection from apoptosis was suggested to result from potentiation of the cellular FLIP, a master antiapoptotic regulator (17), or alternatively by prevention of the internalization of Fas, an important step for apoptosis signaling (44, 45), owing to simultaneous cross-linkage of both Fas and FcμR with IgM Fas mAb (16). With regard to the latter, it is noteworthy that the coligation of FcμR with CD2 or TCR induced a more rapid increase in intracellular Ca²⁺ levels than in the ligation of CD2 or TCR alone.

In conclusion, FcμR, the newest member of the FcR family, interacts more efficiently with the Fc portion of the IgM Abs, which are bound to plasma membranes via their Ag-binding Fab region, as compared with the Fc portion of IgM Abs or their immune complexes in solution. The former interaction occurs in cis on the same cell surface and not in trans between neighboring cells. This strongly implies that FcμR expressed on B, T, and NK cells may have a potential to modulate the function of target Ags or receptors, which are recognized by natural or immune IgM Abs, on the same cell surface.

We thank Dr. Arthur Weiss for the gift of IgM mAbs specific for CD2 (C373) or Jurkat TCR (C305) and for valuable suggestions, Dr. Peter Burrows for critical reading of the manuscript, Enid Keyser for FACS sorting, and Marion Spell for Ca²⁺ mobilization.

The authors have no financial conflicts of interest.