Non-classical binding of a polyreactive α-type anti-idiotypic antibody to B cells

Molecular Immunology 48 (2010) 98–108

Contents lists available at ScienceDirect

Molecular Immunology journal homepage: www.elsevier.com/locate/molimm

Non-classical binding of a polyreactive ␣-type anti-idiotypic antibody to B cells Tays Hernández a , Cristina Mateo de Acosta a,∗ , Alejandro López-Requena a , Ernesto Moreno b , Ruby Alonso c , Yuniel Fernández-Marrero a , Rolando Pérez d a

Immunobiology Division, Center of Molecular Immunology, P.O. Box 16040, Havana 11600, Cuba Tumor Biology Division, Center of Molecular Immunology, P.O. Box 16040, Havana 11600, Cuba c System Biology Division, Center of Molecular Immunology, P.O. Box 16040, Havana 11600, Cuba d Research and Development Direction, Center of Molecular Immunology, P.O. Box 16040, Havana 11600, Cuba b

a r t i c l e

i n f o

Article history: Received 29 June 2010 Received in revised form 10 September 2010 Accepted 14 September 2010 Available online 16 October 2010 Keywords: Alpha-type anti-idiotypic antibodies Fc␥RIIb Antibody binding site B-CLL

a b s t r a c t Detailed information on the immunological relevance of ␣-type anti-idiotypic antibodies is lacking after more than 30 years since Jerne postulated his Idiotypic Network Theory. The B7Y33 mutant is a mousehuman chimeric version of the B7 MAb, a polyreactive ␣-type anti-idiotypic antibody, generated against an anti-GM2 ganglioside IgM Ab1 antibody. It retained the unusual self-binding activity and multispecificity of the parental murine antibody, being able to recognize several anti-ganglioside IgM antibodies as well as non-immunoglobulin antigens. Previous work with the murine B7 MAb suggested that this antibody might have immunoregulatory properties, and therefore we investigated the possible interaction of B7Y33 with immune cells. We found that B7Y33 binds to human and murine B lymphocytes. Inhibition assays using flow cytometry indicated that this antibody is capable of binding the Fc ␥ receptor II (Fc␥RII). The recognition of Fc␥RII-expressing K562, Raji and Daudi human cell lines, together with the capability of inhibiting the binding of an anti-human Fc␥RII antibody to these cells, suggest that B7Y33 interacts with both the Fc␥RIIa and Fc␥RIIb isoforms. We evaluated the contribution to the binding of different surface-exposed residues at the top of the heavy chain variable region (VH) CDR loops through the construction of mutants with substitutions in the three conventional VH CDRs (HCDRs) and the “HCDR4”, located in the framework 3 (HFR3). In addition, we assessed the involvement of the Fc region by performing key mutations in the CH2 domain. Furthermore, chimeric hybrid molecules were obtained by combining the B7Y33 heavy chain with unrelated light chains. Our results indicate that the multispecificity and self-binding properties of B7Y33 are not linked to its recognition of B lineage cells, and that this phenomenon occurs in a non-classical way with the participation of both the variable and constant regions of the antibody. Two possible models for this interaction are proposed, with B7Y33 binding to two Fc␥RIIb molecules through the Fc and Fv regions, or simultaneously to Fc␥RIIb and another unknown antigen on B cells. The Fc␥RIIb has recently received great attention as an attractive target for therapies directed to B lymphocytes. The recognition of peripheral B lymphocytes from B cell chronic lymphocytic leukemia (B-CLL) patients by B7Y33 suggests its potential application for the treatment of B cell malignancies. © 2010 Elsevier Ltd. All rights reserved.

1. Introduction With the postulation of the Idiotypic Network Theory by Jerne in 1974 (Jerne, 1974), the knowledge about the function and

Abbreviations: Ab, antibody; B-CLL, B-cell chronic lymphocytic leukemia; BCR, B cell receptor; CH, heavy chain constant domain; ELISA, enzyme linked immunosorbent assay; Fc, crystallizable fragment; FITC, fluorescein isothiocyanate; Fv, variable fragment; HCDR, heavy chain complementarity determining region; HFR, heavy chain framework; IVIg, intravenous immunoglobulin; MAb, monoclonal antibody; PBMC, peripheral blood mononuclear cells; PE, phycoerithryn; PBS, phosphate buffer saline; VH, heavy chain variable region; V␬, light chain variable region. ∗ Corresponding author. Tel.: +53 7 2143160; fax: +53 7 2720644. E-mail address: [email protected] (C.M. de Acosta). 0161-5890/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.molimm.2010.09.006

structure of antibodies moved from the interaction with their antigens, to the complex relationships they can establish between them, of potential relevance for the immune system regulation (Winkler et al., 1979; Shoenfeld, 2004). The existence of ␤-type anti-idiotypic antibodies, having the ability of inhibiting the interaction of the Ab1 antibody with the nominal antigen and able to mimic the latter molecule, has been largely exploited for the design of vaccines (Poskitt et al., 1991; Betáková et al., 1998). However, the ␣-type anti-idiotypic antibodies, which do not impair the Ab1-antigen binding, have not been extensively studied. Gangliosides are normal components of the plasma membranes of most mammalian cell types, and have been also associated with malignant transformation (Marquina et al., 1996; Malykh et al.,

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

2001; Kannagi et al., 2008). Diverse strategies have been reported to elicit antibodies specific for these low immunogenic molecules (Helling et al., 1994, 1995; Livingston and Ragupathi, 1997). Our group has obtained several ganglioside-specific antibodies (Alfonso et al., 1995; Vázquez et al., 1995), some of which, in different formulations, have been able to induce a strong anti-idiotypic response in mice (Vázquez et al., 1998). We have thus isolated some monoclonal antibodies (MAbs) that behaved as ␤- or ␥-type (those that inhibit the interaction of the Ab1 antibody with the nominal antigen but do not mimic it) anti-idiotypes depending on the species in which they were used as immunogens for the obtention of Ab3 antibodies (Vázquez et al., 1998; Hernández et al., 2005). Moreover, we have also generated, immunizing mice with the anti-ganglioside antibodies mentioned above, non-paratopic specific or ␣-type anti-idiotypic MAbs. One of them, the polyreactive B7 MAb, was obtained by immunizing BALB/c mice with the anti-NeuAc-GM2 ganglioside E1 MAb (Macías et al., 1999). The B7 MAb has shown an anti-tumor effect in a murine model of melanoma. Furthermore, it also exhibits in vitro some properties that resemble those of the intravenous immunoglobulin preparation (IVIg), such as the inhibition of the proliferation of human B and T cell lines and of human normal lymphocytes activated with different mitogens (Macías et al., 1999). These findings suggested that B7 MAb might play a similar immunoregulatory role as proposed for the IVIg pool, which has been successfully used in the treatment of autoimmune and inflammatory diseases and lymphoproliferative disorders (Kazatchkine and Kaveri, 2001; Krause and Shoenfeld, 2005). The mechanisms of action of this preparation rely on both the variable and constant regions of the IgG. The Fv region is responsible for the recognition of soluble and membrane-associated self-molecules, as well as idiotypes of soluble immunoglobulins and B cell receptors (BCR). On the other hand, the Fc portion contributes to IVIg effects through interactions with Fc ␥ receptors (Fc␥R), modulation of Fc␥RIIb expression and saturation of FcRn (Negi et al., 2007). Though the reactivity of B7 MAb with antibodies and nonimmunoglobulin antigens has been well documented (Macías et al., 1999), the nature of the interaction of this antibody with different cells of the immune system has not been yet elucidated. The present work demonstrates the existence of a high avidity interaction of B7Y33, a mutated chimeric version of B7 MAb, with B lineage cells. The interaction with the low affinity receptor Fc␥RIIb proved to be critical for the recognition of this cell type. Our results point out a non-classical interaction of B7Y33 with B cells, which involves both the variable and constant regions of the antibody. The recognition of peripheral B lymphocytes from B cell chronic lymphocytic leukemia (B-CLL) patients by B7Y33 supports its potential application for the treatment of B cell malignancies.

2. Materials and methods

99

2.2. Monoclonal antibodies (MAbs) The E1 (Alfonso et al., 1995), P3 (Vázquez et al., 1998), and F6 (Alfonso et al., 1995) IgM MAbs were purified from mouse ascitic fluid by gel filtration chromatography using a Sephacryl S-300 high resolution column (Pharmacia, Uppsala, Sweden) equilibrated with phosphate-buffered saline (PBS) containing 0.5 M NaCl. The following chimeric antibodies were used: P3 (LópezRequena et al., 2003), two of its mutants (with arginine-to-serine replacements at HCDR1 Kabat position 31 and HCDR3 Kabat positions 98 and 100a, respectively) (López-Requena et al., 2007a), 1E10 (López-Requena et al., 2003), C5 (Roque-Navarro et al., 2003) and 14F7 (Roque-Navarro et al., 2008) (all human IgG1, ␬). The chimeric antibodies, including those obtained during this work, were purified from transfectoma culture supernatants by Protein-A Affinity Chromatography (Pharmacia, Uppsala, Sweden) and analyzed by SDS-PAGE under reducing conditions. The murine Fc␥RII/III-specific 2.4G2 antibody (rat, IgG2a) and the murine MHC-II-specific M5/114.15.2 antibody (rat, IgG2b) were purified from 2.4G2 and M5/114.15.2 hybridoma supernatants, respectively, by Protein-G Affinity Chromatography. The specificity of the purified antibodies was confirmed by enzyme-linked immunosorbent assay (ELISA). 2.3. Vectors The pAH4604 and pAG4622 vectors, containing the human ␥1 and ␬ constant regions, respectively, have been described in detail (Coloma et al., 1992) and were kindly provided by Dr. Sherrie L. Morrison, Department of Microbiology and Molecular Genetics, UCLA, USA. The pAH4604 (Ala/Ala) vector is a mutated version of pAH4604, where the leucine residues at positions 234 and 235 of the human ␥1 CH2 region were replaced by alanines (Hinojosa et al., 2010). It was used for the expression of the B7Y33LALA mutant. 2.4. Variable region genes The genes of the B7Y33 heavy chain variable region (VH) and its mutants were chemically synthesized (Geneart GmbH, Regensburg, Germany). The B7Y33 VH gene was designed from the published B7 VH MAb coding sequence (Hernández et al., 2007), with the substitution of the threonine residue at Kabat position 33 by a tyrosine. The changes contained in each mutant are detailed in Table 1. The synthetic genes were digested HincII/NheI (NEB, New England Biolabs, Ipswich, MA) from pGA4 Geneart vector and cloned into pAH4604 or pAH4604 (Ala/Ala), previously digested EcoRV/NheI (NEB). The B7 MAb light chain variable region (V␬) gene (Hernández et al., 2007) was digested EcoRV/SalI (NEB) and inserted into the equally digested pAG4622.

2.1. Cells

2.5. Chimeric antibody expression

K562 (human erythroleukemia), Raji (Burkitt’s lymphoma), Daudi (Burkitt’s lymphoma), NS0 (murine myeloma), MB16F0 (non-metastatic C57Bl/6 murine melanoma) and F3II (murine breast cancer) cell lines as well as chimeric antibodies-expressing NS0 transfectomas, were cultured at 37 ◦ C, 5% CO2 , in Dulbecco’s modified Eagles medium (DMEM) supplemented with 10% heat inactivated fetal calf serum (FCS), antibiotic mixtures of penicillin (100 U/mL) and streptomycin (100 ␮g/mL), and 2 mM l-glutamine. For the selection of whole antibody-producing transfectomas, DMEM-F12 containing 10% FCS and histidinol at 10 mM was used as selective medium.

NS0 cells were transfected by electroporation with 10 ␮g of both pAG4622 containing chimeric B7 MAb light chain, and pAH4604 or pAH4604 (Ala/Ala) bearing chimeric B7Y33 MAb heavy chain or its mutants, all linearized through PvuI (NEB) digestion. NS0 transfectoma cells expressing chimeric P3 light chain or chimeric 1E10 light chain (López-Requena et al., 2003) were transfected as above with pAH4604 bearing chimeric B7Y33 MAb heavy chain. The transfection method and the selection of transfectomas secreting chimeric IgG have been previously described (LópezRequena et al., 2007a).

100

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

Table 1 Constructed mutants of the B7Y33 VH. Numbering is according to Kabat et al. (1991). Positions Variants

28

31

53

54

57

73

75

98

99

100

B7Y33 H1 H2 H3 H4

Thr Arg

Glu Arg

Asn

Asn

Ala

Lys

Ser

Glu

Ala

Thr

Tyr

Tyr

Thr Lys

Ser

Gin

Thr

Gin

2.6. MAb-biotin conjugation The antibodies (1 mg/mL) were extensively dialyzed with borate buffer 0,2 M, pH 8,5 and then incubated with 100 mg/mL of biotin N-hydroxysuccinimide ester (Sigma, Saint Louis, MO) for 4 h with gentle agitation at room temperature. Finally, they were dialyzed against PBS (Bayer and Wilchek, 1980). 2.7. Antibody binding assays Binding of chimeric B7Y33, its mutants and hybrid molecules (VHB7Y33/V␬E10 and VHB7Y33/V␬P3) to anti-ganglioside IgMs (E1, P3 and F6 MAbs) was determined using an ELISA assay previously described (López-Requena et al., 2007a). Briefly, microtiter plates (Microlon® 600, high binding, Greiner Bio One, Germany) were coated overnight at 4 ◦ C with 10 ␮g/mL of the anti-ganglioside antibodies and blocked for 1 h at 37 ◦ C with PBS-T-BSA (phosphatebuffered saline containing 0.05% Tween 20, pH 7.5, and 1% bovine serum albumin). After an incubation of 2 h at 37 ◦ C with the chimeric antibodies, the plates were washed and an alkaline phosphatase-conjugated goat anti-human IgG (␥-chain-specific) antibody (Sigma) was added. After 1 h at 37 ◦ C the reaction was developed with p-nitrophenylphosphate substrate solution and absorbance monitored at 405 nm (Organon Teknika Inc., reader, Salzburg, Austria). A similar experiment was used to test binding of B7Y33 to chimeric P3 antibody and two of its mutants. The latter antibodies were used for coating the plates, B7Y33 was used biotinylated and reactivity was detected using alkaline-conjugated streptavidin (Jackson Immunoresearch, West Grove, PA). Three samples of each experiment were tested and the coefficient of variation was <10% for all assays. Background values of absorbance were less than 0.1. 2.8. Flow cytometry analysis 2.8.1. Isolation of spleen cells Spleen cells from naïve BALB/c mice were obtained by mechanical dissociation, and erythrocytes were lysed by hypotonic shock in ammonium chloride solution 0.15 M. The cells were washed in cold PBS containing 2% BSA (PBS–BSA 2%). 2.8.2. Isolation of PBMC Peripheral blood mononuclear cells (PBMC) from a healthy human donor and B-CLL patients were isolated by density gradient centrifugation using Ficoll-PaqueTM PLUS (Amersham Pharmacia Biotech AB, Uppsala, Sweden). PBMC were resuspended in PBS–BSA 2%. 2.8.3. Recognition of B lymphocytes and cell lines 106 BALB/c mouse spleen cells or NS0, Raji, Daudi, MB16F0 and F3II cells were incubated with chimeric antibodies on ice for 30 min, and washed with PBS–BSA 2%. The binding of chimeric antibodies was detected by incubation with a FITC-conjugated rabbit antihuman IgG F(ab )2 (Dako, Denmark) for 20 min on ice. In case of

double labeling, the samples were incubated with a PE-conjugated anti-B220 antibody (BD Biosciences Pharmingen, NJ). 106 PBMC from a healthy human donor and B-CLL patients and K562 cells were incubated with biotinylated chimeric antibodies on ice for 30 min. After washing with PBS–BSA 2%, the binding was detected with PE-conjugated streptavidin (BD Biosciences Pharmingen). For the identification of the human B cell population, the samples were labeled with a FITC-conjugated antiCD19 antibody (AbD Serotec, Germany). Chimeric C5, 1E10 and P3 antibodies were used as isotype-matched controls. 2.8.4. FcRII expression in murine cell lines For the evaluation of the expression of Fc␥RII in three murine cell lines (NS0, MB16F0 and F3II), 106 cells were incubated with purified 2.4G2 MAb. For the detection of 2.4G2 binding, a biotinylated goat anti-rat immunoglobulin antibody (BD Biosciences Pharmingen) and FITC-conjugated streptavidin (BD Biosciences Pharmingen) were used. 2.8.5. Competition assays The competition experiments were performed by incubating 106 BALB/c mouse spleen cells with increasing concentrations of a donkey anti-mouse IgM (␮ chain-specific) antibody (Jackson Immunoresearch) for 30 min on ice. After washing with PBS–BSA 2%, the cells were incubated with 2.5 ␮g/ml of biotinylated B7Y33 for 30 min at 4 ◦ C. Reactivity was measured using FITC-conjugated streptavidin (BD Biosciences Pharmingen). A PE-conjugated antiB220 antibody (BD Biosciences Pharmingen) was used to identify the B cell population. The inhibition assay with an anti-mouse Fc␥RII antibody was done by incubating BALB/c mouse spleen cells or NS0 cells with purified 2.4G2 antibody for 30 min on ice. Then, cells were incubated with biotinylated B7Y33 at 2.5 ␮g/ml. Reactivity was measured using PE-conjugated streptavidin. The inhibition percentage was calculated as: % inhibition = [(% binding without inhibitor − % binding with inhibitor)/% binding without inhibitor]*100. In the experiment with spleen cells, an anti-MHCII antibody was used as control inhibitor. For the NS0 cells, the biotinylated chimeric anti-NeuGc-GM3 ganglioside 14F7 antibody was used as negative control. Human cells K562 and Raji were first incubated with B7Y33 at 15 ␮g/mL (K562) or 5 ␮g/mL (Raji), and then with PE-conjugated anti-human Fc␥RII AT10 antibody (AbD Serotec) for 30 min on ice. Binding of AT10 antibody conjugate was also assessed in the absence of B7Y33. 2.8.6. “Molecular complementarity” binding experiments H3 or H4 mutants were mixed with B7Y33LALA mutant at different ratios (1:1, 1:10 and 10:1; 1 stands for 2.5 ␮g/ml and 10 stands for 25 ␮g/ml) and incubated at 37 ◦ C for 30 min. Mixes were added to spleen cells from BALB/c mice. Reactivity was measured adding a FITC-conjugated rabbit anti-human IgG F(ab )2 (Dako) for 20 min on ice. The samples were incubated with a PE-conjugated anti-B220 antibody (BD Biosciences Pharmingen).

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

In all cytometry assays the mean fluorescence intensity (MFI) and percentage of stained cells were determined with a FACScan instrument (Becton Dickinson, NJ). The WinMDI 2.8 program was used to analyze a total of 104 cells acquired on every FACS experiment. All the assays were performed at least twice. 3. Results 3.1. The B7Y33 mutant chimeric antibody The B7 MAb is an ␣-type anti-idiotypic antibody with multispecificity and self recognition properties (Macías et al., 1999). The study of its biological properties required a constant source of antibody that was unavailable due to the instability of the secreting hybridoma (Macías A., personal communication). For this reason, we decided to construct a chimeric version of the molecule (human ␥1, ␬ constant regions), also considering the envisaged therapeutic potential and the practical benefit of avoiding the traces of host IgG contaminants coming from the murine ascites. However, the attempts to obtain a transfectoma with appropriate antibody productivity levels were ineffective, somehow in agreement with the B7 hybridoma instability. Afterwards, we obtained several variants of the recombinant antibody by introducing single point mutations in the heavy chain variable region (VH) (our unpublished data). One of these mutants, named B7Y33, where the threonine residue at position 33 (Kabat numbering) of the VH was substituted by a tyrosine, was efficiently expressed. Hence, we continued working with this mutant, taking into account that it kept the B7 MAb immunochemical properties.

101

3.2. B7Y33 binds to unrelated immunoglobulins via a non-classical binding site B7Y33, as reported before for the B7 MAb (Macías et al., 1999), is highly polyreactive, being able to interact with several anti-ganglioside IgM antibodies as well as non-immunoglobulin antigens (Supplementary Fig. 1). In order to study the molecular features of its variable region that support this multispecificity, we constructed chimeric hybrid molecules by combining the B7Y33 heavy chain with unrelated light chains. We selected the light chains of an anti-NeuGc-ganglioside MAb, named P3 (Vázquez et al., 1995) and its anti-idiotype, named 1E10 (Vázquez et al., 1998). In both cases, the mutants exhibited higher reactivity, as compared with B7Y33, with two of the IgM antibodies that were tested, while showing almost identical reactivity towards the rest (Fig. 1A). This result suggests that although the light chain variable region (V␬) modulates the interaction of B7Y33 with these antibodies, the VH domain is the main responsible for its binding properties. Based upon the sequence reported for B7 VH (Hernández et al., 2007) we constructed and expressed mutants with substitutions in each of the three HCDRs, as well as in the so-called HCDR4, located in the framework 3 (HFR3) (Franklin et al., 2004; Bond et al., 2005). We called these mutants H1, H2, H3 and H4, respectively (Table 1). As shown in Fig. 1B, and although with some differences with respect to the wild antibody, none of the mutants lost the capability of recognizing the murine IgMs. Indeed, both H1 and H2 mutations modified similarly the recognition of E1 and F6 antibodies by B7Y33, while did not have impact on the reactivity against P3. On the other hand, H3 mutations strongly influenced the reactivity against P3, unlike the E1 and F6 IgMs. These results point out the

Fig. 1. Reactivity of B7Y33, its H1 (HCDR1), H2 (HCDR2), H3 (HCDR3) and H4 (HFR3) mutants and the hybrid VHB7Y33/V␬P3 and VHB7Y33/V␬1E10 antibodies, with antiganglioside antibodies. (A and B) Microtiter plates were coated with 10 ␮g/ml of murine IgM E1, P3 and F6 mAbs. Purified B7Y33, its mutants and the hybrid antibodies were added unconjugated at 10 ␮g/ml. Reactivity was measured using an alkaline phosphatase-conjugated anti-human IgG (␥ chain-specific) antibody. Light grey bars represent the reactivity of B7Y33, black bars of its mutants or the hybrid antibodies, and dark grey bars of the isotype-matched chimeric C5 antibody. (C) Microtiter plates were coated with 10 ␮g/ml of chimeric P3 antibody (circles) and its mutants P3S31 (HCDR1, squares) and P3S98;100a (HCDR3, triangles). Biotinylated B7Y33 was added at different concentrations and the reactivity was measured using alkaline phosphatase-conjugated streptavidin. Black curves represent the reactivity of biotinylated B7Y33 and grey curves of isotype-matched biotinylated chimeric 14F7 antibody.

102

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

Fig. 2. Self-binding assay of B7Y33 and its variable and constant region mutants. Microtiter plates were coated with 10 ␮g/mL of B7Y33 (A), its variable region mutants H1 (HCDR1) (B), H2 (HCDR2) (C), H3 (HCDR3) (D) and H4 (HFR3) (E), and its CH2 mutant B7Y33LALA (F). Rituximab (Rx) was used as isotype-matched control (G). Different concentrations of biotinylated antibodies were added to the plates coated with the same non-biotinylated molecules. Reactivity was measured using alkaline phosphataseconjugated streptavidin.

existence of at least one non-classical binding site other than the conventional one on B7Y33 that explains its polyreactivity. Taking into account that E1 and F6 MAbs belong to the same VH family, Q52 (López-Requena et al., 2007b), being almost identical (only one conservative change at framework 1), the reactivity profile of H1 and H2 mutants might indicate the recognition of a region that does not include the HCDR3 of these IgMs. In addition, we tested the binding of B7Y33 to two previously described mutants of P3 MAb, which lost the ganglioside binding capability. One of the mutants (unable also to bind to P3-specific anti-idiotypes) has a substitution in HCDR1, while the other one has two substitutions in HCDR3 (López-Requena et al., 2007a). Again, the interaction of B7Y33 was unaffected (Fig. 1C). It is thus interesting to notice that B7Y33 does not interact with P3 through the demonstrated classical binding site of this latter antibody. P3 MAb VH gene also belongs to the Q52 family, but it displays several differences at HCDRs 1 and 2 with respect to E1 and F6 MAbs (López-Requena et al., 2007b). Collectively, these results demonstrate that B7Y33 is an anti-idiotypic antibody that recognizes non-paratopic regions on different antibodies. An unusual property of the B7 MAb is its ability for self-binding (our unpublished data). As for other immunochemical properties of B7 MAb, B7Y33 also displayed a homophilic recognition capability. Interestingly, this property was not affected in any of the evaluated mutants which were able of recognizing themselves (Fig. 2) and the wild type antibody (data not shown). This last result is an additional indication of immunoglobulin interactions not involving the classical binding site. 3.3. FcRIIb is critical for the binding of B7Y33 to B cells The previous observations prompted us to test whether B7Y33 recognizes immunoglobulins not only in soluble forms, but also

as surface BCRs. We first assessed the binding of B7Y33 to B lymphocytes of both murine and human origin. B7Y33 recognized a large percentage of spleen B cells from naïve BALB/c mice (Fig. 3A and C). In addition, it interacted extensively with peripheral B lymphocytes from a healthy human donor and B-CLL patients (Fig. 3B and D). In all cases, binding was demonstrated at concentrations at which the interaction of the isotype-matched control antibody was negative. We further tested whether B7Y33 interacts directly with the BCR, using an inhibition assay. Preincubation of cells with increasing concentrations of an anti-IgM antiserum did not affect the binding of B7Y33 to murine B lymphocytes (Fig. 4A). This result is not conclusive, considering the possibility of a non-complete blocking by the antiserum of all the potential binding sites on the BCR. For that reason, we evaluated the recognition of a nonimmunoglobulin-expressing murine myeloma (NS0) by B7Y33, which resulted to be positive (Fig. 4B). Taken together, these results indicate that B7Y33 recognizes an antigen other than the BCR on B cells. Then, we evaluated the possible interaction of B7Y33 with other pan-B molecules. We first focused on the CD20 molecule, whose murine and human versions share 73% similarity (Tedder et al., 1988). For that purpose, we compared the capability of B7Y33 to recognize human B cells with that of the commercial anti-CD20 antibody (Rituximab). We used PBMC from the same three B-CLL patients, as in the previous experiments. Rituximab recognized clearly only two out of the three tested samples, indicating a heterogeneity in the expression of CD20 by these malignant B cells (Fig. 3D). In contrast, B7Y33 was able to interact also with the lymphocytes of the sample that was negative to Rituximab binding (Fig. 3D). These results rule out the CD20 molecule as a target of B7Y33 on human B lymphocytes.

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

103

Fig. 3. Binding of B7Y33 to murine and human B lymphocytes. (A and C) Spleen cells from naïve BALB/c mice; (B) PBMC from a healthy human donor; (D) PBMC from B-CLL patients. Cells were incubated with 5 ␮g/mL of unconjugated (A) or biotinylated (B and D) B7Y33 and Rituximab, or with different concentrations of unconjugated B7Y33 (C). The binding was measured using a FITC-conjugated anti-human IgG antibody (A and C) or PE-conjugated streptavidin (B and D). Chimeric C5 antibody was used as isotype-matched control. Results are representative of two performed assays.

The second pan-B molecule we chose was Fc␥RIIb, considering it is conserved in mice and humans (Brooks et al., 1989; Nimmerjahn and Ravetch, 2008). This molecule belongs to a family of receptors involved in the interaction of immunoglobulins with cells, and is

the only member of this family that is expressed on B lymphocytes (Amigorena et al., 1989; Nimmerjahn and Ravetch, 2008). Preincubation of cells with the murine Fc␥RII/III-specific 2.4G2 antibody led to a marked reduction of B7Y33 binding to murine spleen B

Fig. 4. B7Y33 recognizes an antigen other than the BCR on B cells. (A) Inhibition assay of B7Y33 binding to murine B cells using an anti-mouse IgM antiserum. Spleen cells from BALB/c mice were first incubated with different concentrations of the inhibitor antiserum, and then with biotinylated B7Y33 at a non-saturating concentration (2.5 ␮g/mL). Reactivity was measured using FITC-conjugated streptavidin. A commercial PE-conjugated anti-B220 antibody was used to identify the B cell population. (B) Binding of B7Y33 to murine myeloma NS0 cell line. Cells were incubated with 10 ␮g/mL of the antibody. Binding was detected with a FITC-conjugated anti-human IgG antibody. Light grey curve represents the background fluorescence, black curve the binding of B7Y33 and dark grey curve the binding of the isotype-matched chimeric C5 antibody. Results are representative of two performed assays.

104

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

cells and to the NS0 cell line (Fc␥RIIb-positive, supplementary Fig. 2; Gillet et al., 2007). This effect was not observed either with an anti-MHC-class II antibody or with the anti-NeuGc-GM3 ganglioside 14F7 MAb (Carr et al., 2000), which binds to these murine cells (Fig. 5). Afterwards, we extended the study to other Fc␥RII isoforms, by evaluating the recognition of B7Y33 against a panel of murine and human cell lines with different expression patterns of this receptor. Consistently with the above finding, B7Y33 bound to the human K562 (Fc␥RIIa-positive; Sibéril et al., 2006), Raji (Fc␥RIIb-positive; Mackay et al., 2006) and Daudi (Fc␥RIIb-positive; Mackay et al., 2006) cell lines (Supplementary Fig. 3). No interaction was observed (Supplementary Fig. 3) with the murine MB16F0 and F3II cell lines (Fc␥RII-negative, supplementary Fig. 2). The inhibition by B7Y33 of the binding of a commercial anti-human Fc␥RII antibody to the K562 and Raji cells further corroborated the interaction of our antibody with Fc␥RII. The fact that no binding was detected for the isotype-matched control antibody (Fig. 5) indicates that binding of B7Y33 to these cells was not simply a result of the interaction of its Fc region with the Fc␥RII. 3.4. Involvement of B7Y33 constant and variable regions in its interaction with B lymphocytes In order to demonstrate the role of the B7Y33 Fc fragment in the recognition of B cells, we constructed a heavy chain constant region mutant with impaired Fc␥R binding. We selected the double leucine 234 and 235-to alanine mutation in the CH2 domain (Kabat numbering), which is well known to abolish the

IgG-Fc␥RII interaction (Huizinga et al., 1989; Sarmay et al., 1992; Xu et al., 2000; Hinojosa et al., 2010). The resulting antibody, called B7Y33LALA, retained, as expected, the recognition of E1 MAb (Supplementary Fig. 4) as well as the self-binding activity (Fig. 2F), but showed no recognition of murine B lymphocytes (Fig. 6A) and NS0 myeloma cells (Fig. 6B). This result confirms the contribution of the Fc–Fc␥RII interaction in the binding of B7Y33 to these cells. We used two approaches to test the contribution of the B7Y33 variable region to the interaction with B lineage cells. First, we tested the previously described hybrid antibodies. As shown in Fig. 7, the VHB7Y33/V␬P3 antibody did not recognize the cells, while the VHB7Y33/V␬E10 hybrid molecule did bind, although weaker than the wild antibody. We then assessed whether the H1–4 mutants were also able to bind to B and NS0 cells. The H1 and H2 mutants behaved as the wild antibody (Fig. 6A). In contrast, the interaction was affected for the H3 and H4 variants (Fig. 6A). Taken together, these results further support the conclusion that the interaction of B7Y33 with B cells does not depend on its capability of recognizing different immunoglobulins, and that both the variable and constant regions participate in the binding. Finally, we tried to demonstrate a possible way of interaction of B7Y33 with B cells based on both Fc and Fv direct contacts with Fc␥RII and the subsequent stabilization of these interactions through the homophilic and heterophilic binding activity displayed by this antibody and its H3 and H4 mutants. However, we did not detect any B cell binding when we mixed the H3 or H4 mutants with B7Y33LALA (data not shown) (see Fig. 8I, II).

Fig. 5. B7Y33 binds to the Fc␥RII. (A) Spleen cells from BALB/c mice were preincubated with different concentrations of the anti-murine Fc␥RII/III 2.4G2 antibody (left panel) or an anti-MHCII antibody (control inhibitor) (right panel); (B) murine myeloma NS0 cells were preincubated with the 2.4G2 antibody at 2 ␮g/mL. Then, cells were incubated with biotinylated B7Y33 at 2.5 ␮g/ml. Reactivity was measured using PE-conjugated streptavidin. In (B), biotinylated chimeric anti-NeuGc-GM3 ganglioside 14F7 antibody was used as negative control inhibitor. (C) Human K562 (left panel) and Raji (right panel) cells were preincubated with B7Y33 at 15 ␮g/mL (K562) or 5 ␮g/mL (Raji). Then, cells were labeled with the PE-conjugated anti-human Fc␥RII AT10 antibody at 2 ␮g/mL. Chimeric 1E10 antibody was used as isotype-matched control. The mean fluorescence intensity (MFI) and the percentage of labeled cells are indicated. Results are representative of two performed assays.

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

105

Fig. 6. Binding of B7Y33 mutants to B lineage cells. Spleen cells from BALB/c mice (A) and NS0 cells (B) were incubated with 20 ␮g/ml of B7Y33 and its mutants: H1 (HCDR1), H2 (HCDR2), H3 (HCDR3), H4 (HFR3) and B7Y33LALA (CH2), followed by labeling with a FITC-conjugated anti-human IgG antibody. A commercial PE-conjugated anti-B220 antibody was used to identify the B cell population. Chimeric C5 antibody was used as isotype-matched control. Results are representative of two performed assays. Percentage of labeled cells is indicated.

Fig. 7. Binding of VHB7Y33/unrelated V␬ hybrid antibodies to B lineage cells. Spleen cells from BALB/c mice (A) and murine myeloma NS0 cells (B) were incubated with 20 ␮g/mL of B7Y33, VHB7Y33/V␬P3, VHB7Y33/V␬1E10 and isotype-matched chimeric P3, 1E10 and C5 antibodies. Reactivity was measured using a FITC-conjugated antihuman IgG antibody. Results are representative of two performed assays.

4. Discussion The B7 MAb is an anti-idiotypic antibody (Macías et al., 1999) generated by immunization of BALB/c mice with the NeuAc-GM2 ganglioside-specific E1 MAb (Alfonso et al., 1995). Following traditional criteria, B7 is an ␣-type anti-idiotypic antibody, because it

does not inhibit the binding of E1 MAb to the ganglioside. Additionally, it exhibits a wide-spectrum reactivity that includes antibodies of different isotypes and specificities, and non-immunoglobulin antigens (Macías et al., 1999). More importantly, it has also shown anti-tumor properties (Macías et al., 1999). Therefore, it is important to study the molecular basis of the immunochemical behavior

106

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

Fig. 8. Schematic representation of the models of binding of B7Y33 to B lineage cells. (I) B7Y33 binds to Fc␥RIIb through its Fc. B7Y33-Fc␥RIIb interaction is stabilized by inter-Fv homophilic binding. (II) Two molecules of B7Y33 bound to two molecules of Fc␥RIIb, one through its Fc and the other through its variable region, interact by inter-Fv homophilic binding. (III) B7Y33 binds to one molecule of Fc␥RIIb through its Fc and to another molecule of Fc␥RIIb through its variable region. (IV) B7Y33 binds to Fc␥RIIb through its Fc, and to another unknown surface antigen through its variable region.

of this antibody and identify its targets on immune system cells, in order to understand the mechanisms underlying its potential therapeutic effects. E1 antibody, used to generate B7 MAb, is an IgM of germ line origin (López-Requena et al., 2007b), as is frequently found among the anti-ganglioside antibodies of the natural repertoire (Vollmers and Brandlein, 2007; López-Requena et al., 2007b). As demonstrated in this work, B7 MAb interacts with E1 MAb and other unrelated antigens through a non-conventional binding site. Furthermore, it also binds to B cells in a non-classical way. It is likely that, when using as antigen an antibody with the characteristics displayed by the E1 MAb, the anti-idiotypic antibodies generated may include specificities and other features of the natural repertoire, some of them with immunoregulatory properties. In that sense, we are currently studying the possible in vivo effect of B7 MAb on the antibody response against non-immunogenic self immunoglobulins, when administered together to mice. B7Y33 recognizes unrelated soluble immunoglobulins, but this observation does not explain its binding to B cells. We demonstrated that B7Y33 also binds to a myeloma cell devoid of surface immunoglobulin, which, in spite of indicating the relevance of an antigen other than BCR in the recognition, does not exclude a possible low affinity interaction with this receptor. Our results indicate that Fc␥RIIb, which is highly conserved in mice and humans (Brooks et al., 1989), is the major target of B7Y33 on B lymphocytes from both species. Though this receptor belongs to a family of molecules whose main function is associated to the binding of antibodies through their Fc, the data presented here suggest a particular type of interaction. The binding of chimeric B7Y33 antibody to the murine Fc␥RIIb, a low affinity receptor for monomeric IgG and the only Fc␥R expressed on B cells (Amigorena et al., 1989; Nimmerjahn and Ravetch, 2008), seems to be unique, as compared with other antibodies with the same human ␥1 isotype. Moreover, the recognition of different human cell lines expressing both Fc␥RIIa and Fc␥RIIb is coherent with the high similarity of the extracellular domains of these two isoforms (Brooks et al., 1989).

The construction of VH and Fc mutants of B7Y33 allowed us to elucidate the contribution of these two regions to the binding to B cells. Our results evidenced the relevance of the idiotype, which has no participation in conventional Fc-Fc␥R interactions. We proposed four possible models of interaction of B7Y33 with B cells (Fig. 8). The four possibilities were designed and evaluated using all the evidences gathered in this work, as well as the previously known polyreactive nature of B7 MAb (Macías et al., 1999; Hernández et al., 2007). Self-binding is one of the features that characterize some of the so-called “superantibodies” (Kohler and Paul, 1998; Kohler, 2000). This property consists of interactions between two identical antibody molecules through a region different from the conventional binding site. It provides a mechanism for amplifying the binding of an antibody to a cell surface antigen (Yan et al., 1996; Zhao et al., 2002). Following this concept, we proposed a first model based on the binding of B7Y33 to Fc␥RIIb through its Fc and stabilization of the complex by Fv homophilic interactions (Fig. 8-I). This model can be rejected because of the intact self binding activity of non-B cell binding VH mutants (H3 and H4). The second model suggests that self-binding through the Fv region stabilizes both the Fc- and Fv-mediated direct interaction of B7Y33 with Fc␥RIIb (Fig. 8-II). From such a model of interaction, one could expect a “molecular complementarity” of the H3/H4 and B7Y33LALA mutants for the binding to these cells: the lack of recognition of H3/H4 through the idiotype, and the impaired interaction of B7Y33LALA through the Fc, would be compensated by the interaction with Fc␥RII of the B7Y33LALA idiotype and the H3/H4 Fc, respectively. As mentioned above, we demonstrated that this type of interaction does not occur. Considering the polyspecific nature of B7 MAb (Macías et al., 1999), one antibody molecule could in principle bind to two Fc␥RIIb molecules through two different sites: one on the Fv and the other on the Fc (Fig. 8-III). Another very similar model would consist of the binding of this antibody to a surface antigen other than Fc␥RIIb, while having its Fc engaged by this receptor. We think that this is the most probable model (Fig. 8-IV), because it would be consis-

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

tent also with at least one report in which an antibody required stabilization by interaction with Fc␥RIIb for optimal binding to its antigen (CD22) on the surface of B cells (Walker and Smith, 2008). Nevertheless, with our current data we are unable to rule out the third possibility (Fig. 8-III). More experiments are needed to determine which model of interaction is the correct one. Fc␥RIIb plays an important role in B cell homeostasis and the regulation of the immune response (Rahman et al., 2007; Xiang et al., 2007; Smith and Clatworthy, 2010). It is overexpressed in lymphomas of B cell origin such as follicular lymphoma (Callanan et al., 2000). Other B cell malignancies like mantle cell, splenic marginal zone and small lymphocytic lymphomas also express this molecule (Camilleri-Broët et al., 2004; Rankin et al., 2006). These findings make the Fc␥RIIb an interesting target for treating those kinds of B cell disorders. The recognition of B lymphocytes from B-CLL patients by B7Y33 turns this antibody into a potential therapeutic tool. The well known anti-CD20 Rituximab is currently a powerful weapon against Non-Hodgkin lymphoma with outstanding results in the clinics (Rothe et al., 2004). Nonetheless, its use is limited in patients with low or no expression of the CD20 molecule on their malignant B cells. Here, we demonstrated that such cells are still recognized by B7Y33; therefore, it would be interesting to investigate the effect of our antibody on them. In summary, this work reports the high avidity binding of B7Y33, a polyreactive ␣-type anti-idiotypic antibody, to human and mouse B cells. Our data indicate the participation of both the variable and constant regions of B7Y33 in this binding, where the interaction with the Fc␥RIIb molecule on the surface of B cells seems to play a crucial role. The recognition of peripheral B lymphocytes from B-CLL patients by B7Y33 suggests its potential application in the treatment of B cell malignancies. The interaction of B7Y33 with Fc␥RIIb merits further studies in order to evaluate its possible effects on the regulation of the immune system as well as the therapeutic potential of this antibody. Conflict of interest The authors have no financial conflict of interest. Acknowledgements This work was supported by the Center of Molecular Immunology. We thank Katya Sosa for her helpful technical assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.molimm.2010.09.006. References Alfonso, M., Vázquez, A.M., Carr, A., Haerslev, T., Fernández, L.E., Lanio, M.E., Alvarez, C., Zeuthen, J., Pérez, R., 1995. T cell-independent B cell response to selfmonosialogangliosides: primary response monoclonal antibodies. Hybridoma 14, 209–216. Amigorena, S., Bonnerot, C., Choquet, D., Fridman, W.H., Teillaud, J.L., 1989. Fc gamma RII expression in resting and activated B lymphocytes. Eur. J. Immunol. 19, 1379–1385. Bayer, E.A., Wilchek, M., 1980. The use of avidin–biotin complex as a tool in molecular biology. Methods Biochem. Anal. 26, 1–45. Betáková, T., Vareckova, E., Kostolansky, F., Mucha, V., Daniels, R.S., 1998. Monoclonal anti-idiotypic antibodies mimicking the immunodominant epitope of influenza virus haemagglutinin elicit biologically significant immune responses. J. Gen. Vir. 79, 461–470. Bond, C.J., Wiesmann, C., Marsters Jr., J.C., Sidhu, S.S., 2005. A structure-based database of antibody variable domain diversity. J. Mol. Biol. 348, 699–709. Brooks, D.G., Qiu, W.Q., Luster, A.D., Ravetch, J.V., 1989. Structure and expression of human IgG FcRII (CD32). Functional heterogeneity is encoded in alternatively spliced products of multiple genes. J. Exp. Med. 170, 1369–1385.

107

Callanan, M.B., Le Baccon, P., Mossuz, P., Duley, S., Bastard, C., Hamoudi, R., Dyer, M.J., Klobeck, G., Rimokh, R., Sotto, J.J., Leroux, D., 2000. The IgG Fc receptor FcgRIIB, is a target for deregulation by chromosomal translocation in malignant lymphoma. Proc. Natl. Acad. Sci. U.S.A. 97, 309–314. Camilleri-Broët, S., Cassard, L., Broët, P., Delmer, A., Le Touneau, A., Diebold, J., Fridman, W.H., Molina, T.J., Sautès-Fridman, C., 2004. FcgammaRIIB is differentially expressed during B cell maturation and in B cell lymphomas. Br. J. Haematol. 124, 55–62. Carr, A., Mullet, A., Mazorra, Z., Vázquez, A.M., Alfonso, M., Mesa, C., Rengifo, E., Pérez, R., Fernández, L.E., 2000. A mouse IgG1 monoclonal antibody specific for Nglycolyl GM3 ganglioside recognized breast and melanoma tumors. Hybridoma 19, 241–247. Coloma, M.J., Hastings, A., Wims, L.A., Morrison, S.L., 1992. Novel vectors for the expression of antibody molecules using variable regions generated by polymerase chain reaction. J. Immunol. Methods 152, 89–104. Franklin, M.C., Carey, K.D., Vajdos, F.F., Leahy, D.J., de Vos, A.M., Sliwkowski, M.X., 2004. Insights into ErbB signaling from the structure of the ErbB2–pertuzumab complex. Cancer Cell. 5, 317–328. Gillet, L., May, J.S., Colaco, S., Stevenson, P.G., 2007. The murine gammaherpesvirus68 gp150 acts as an immunogenic decoy to limit virion neutralization. PLoS One 2, e705. Helling, F., Shang, A., Calves, M., Zhang, S., Ren, S., Yu, R.K., Oettgen, H.F., Livingston, P.O., 1994. GD3 vaccines for melanoma: superior immunogenicity of keyhole limpet hemocyanin conjugate vaccines. Cancer Res. 54, 197–203. Helling, F., Zhang, S., Shang, A., Adluri, S., Calves, M., Koganty, R., Longenecker, B.M., Yao, T.-J., Oettgen, H.F., Livingston, P.O., 1995. GM2-KLH conjugate vaccine: Increased immunogenicity in melanoma patients after administration with immunological adjuvant QS-21. Cancer Res. 55, 2783–2788. Hernández, A.M., Rodríguez, M., López-Requena, A., Beausoleil, I., Pérez, R., Vázquez, A.M., 2005. Generation of anti-Neu-glycolyl-ganglioside antibodies by immunization with an anti-idiotype monoclonal antibody: a self versus nonself-matter. Immunobiology 210, 11–21. Hernández, T., López-Requena, A., Mateo de Acosta, C., Mustelier, G., Roque-Navarro, L., Pérez, R., 2007. B7 and 34B7 monoclonal antibodies: a theoretical approach to the molecular basis of immunoglobulin cross-reactive antibodies. Hybridoma (Larchmt). 26, 66–72. Hinojosa, L.E., Hernández, T., Mateo de Acosta, C., Montero, E., Pérez, R., LópezRequena, A., 2010. Construction of a recombinant non-mitogenic anti-human CD3 antibody. Hybridoma 29, 115–124. Huizinga, T.W., Kerst, M., Nuyens, J.H., Vlug, A., von dem Borne, A.E., Roos, D., Tetteroo, P.A., 1989. Binding characteristics of dimeric IgG subclass complexes to human-neutrophils. J. Immunol. 142, 2359–2364. Jerne, N.K., 1974. Towards a network theory of the immune system. Ann. Immunol. 125, 373–389. Kabat, E.A., Wu, T.T., Perry, H.M., Gotessmann, K.S., Foeller, C., 1991. Sequences of proteins of immunological interest. Dept of Health and Human Service. US Government Printing Office. Kannagi, R., Yin, J., Miyazaki, K., Izawa, M., 2008. Current relevance of incomplete synthesis and neo-synthesis for cancer-associated alteration of carbohydrate determinants—Hakomori’s concepts revisited. Biochim. Biophys. Acta 1780, 525–531. Kazatchkine, M.D., Kaveri, S.V., 2001. Immunomodulation of autoimmune and inflammatory diseases with intravenous immunoglobulin (IVIg). N. Engl. J. Med. 345, 747–755. Kohler, H., 2000. Superantibodies: synergy of innate and acquired immunity. Appl. Biochem. Biotechnol. 83, 1–9. Kohler, H., Paul, S., 1998. Superantibody activities: new players in innate and adaptive immune responses. Immunol. Today 19, 221–227. Krause, I., Shoenfeld, Y., 2005. Intravenous immunoglobulin treatment for fibrosis, atherosclerosis, and malignant conditions. Methods Mol. Med. 109, 403–408. Livingston, P.O., Ragupathi, G., 1997. Carbohydrate vaccines that induce antibodies against cancer. Previous experience and future. Cancer Immunol. Immunother. 45, 10–19. López-Requena, A., Mateo de Acosta, C., Moreno, E., González, M., Puchades, Y., Talavera, A., Vispo, N.S., Vázquez, A.M., Pérez, R., 2007a. Gangliosides Ab1 and Ab2 antibodies I. Towards a molecular dissection of an idiotype–anti-idiotype system. Mol. Immunol. 44, 423–433. López-Requena, A., Mateo de Acosta, C., Pérez, A., Valle, A., Lombardero, J., Sosa, K., Pérez, R., Vázquez, A.M., 2003. Chimeric anti-N-glycolyl-ganglioside and its anti-idiotypic MAbs: immunodominance of their variable regions. Hybrid Hybridomics 22, 235–243. López-Requena, A., Mateo De Acosta, C., Vázquez, A.M., Pérez, R., 2007b. Immunogenicity of autologous immunoglobulins: principles and practices. Mol. Immunol. 44, 3076–3082. Macías, A., Arce, S., Leon, J., Mustelier, G., Bombino, G., Domarco, A., Perez, R., Lage, A., 1999. Novel cross-reactive anti-idiotype antibodies with properties close to the human intravenous immunoglobulin (IVIg). Hybridoma 18, 263–272. Mackay, M., Stanevsky, A., Wang, T., Aranow, C., Li, M., Koenig, S., Ravetch, J.V., Diamond, B., 2006. Selective dysregulation of the Fc gamma IIB receptor on memory B cells in SLE. J. Exp. Med. 203, 2157–2164. Malykh, Y.N., Schauer, R., Shaw, L., 2001. N-Glycolylneuraminic acid in human tumours. Biochimie 83, 623–634. Marquina, G., Waki, H., Fernandez, L.E., Kon, K., Carr, A., Valiente, O., Perez, R., Ando, S., 1996. Gangliosides expressed in human breast cancer. Cancer Res. 56, 5165–5171.

108

T. Hernández et al. / Molecular Immunology 48 (2010) 98–108

Negi, V., Sriramulu, E., Siberil, S., Graff-Dubois, S., Mouthon, L., Kazatchkine, M.D., Lacroix-Desmazes, S., Bayry, J., Kaveri, S.V., 2007. Intravenous immunoglobulin: an update on the clinical use and mechanisms of action. J. Clin. Immunol. 27, 233–245. Nimmerjahn, F., Ravetch, J.V., 2008. Fcgamma receptors as regulators of immune responses. Nat. Rev. Immunol 8, 34–47. Poskitt, D.C., Jean-Francois, M.J., Turnbull, S., Macdonald, L., Yasmeen, D., 1991. Internal image (Ab2 beta) anti-idiotype vaccines. Theoretical and practical aspects. Vaccine 9, 792–796. Rahman, Z.S., Alabyev, B., Manser, T., 2007. FcgammaRIIB regulates autoreactive primary antibody-forming cell, but not germinal center B cell activity. J. Immunol. 178, 897–907. Rankin, C.T., Veri, M.C., Gorlatov, S., Tuaillon, N., Burke, S., Huang, L., Inzunza, H.D., Li, H., Thomas, S., Johnson, S., Stavenhagen, J., Koenig, S., Bonvini, E., 2006. CD32B, the human inhibitory Fc-gamma receptor IIB, as a target for monoclonal antibody therapy of B-cell lymphoma. Blood 108, 2384–2391. Roque-Navarro, L., Chakrabandhu, K., de León, J., Rodríguez, S., Toledo, C., Carr, A., de Acosta, C.M., Hueber, A.O., Pérez, R., 2008. Anti-ganglioside antibody-induced tumor cell death by loss of membrane integrity. Mol. Cancer Ther. 7, 2033–2041. Roque-Navarro, L., Mateo, C., Lombardero, J., Mustelier, G., Fernández, A., Sosa, K., Morrison, S.L., Pérez, R., 2003. Humanization of predicted T-cell epitopes reduces the immunogenicity of chimeric antibodies: new evidence supporting a simple method. Hybrid Hybridom. 22, 245–257. Rothe, A., Schulz, H., Elter, T., Engert, A., Reiser, M., 2004. Rituximab monotherapy is effective in patients with poor risk refractory aggressive non-Hodgkin’s lymphoma. Haematologica 89, 875–876. Sarmay, G., Lund, J., Rozsnyay, Z., Gergely, J., Jefferis, R., 1992. Mapping and comparison of the interaction sites on the Fc region of the IgG responsible for the triggering antibody dependent cellular cytotoxicity (ADCC) through different types of human Fcg receptor. Mol. Immunol. 29, 633–639. Shoenfeld, Y., 2004. The idiotypic network in autoimmunity: antibodies that bind antibodies that bind antibodies. Nat. Med. 10, 17–18. Sibéril, S., Dutertre, C.A., Boix, C., Bonnin, E., Ménez, R., Stura, E., Jorieux, S., Fridman, W.H., Teillaud, J.L., 2006. Molecular aspects of human FcgammaR interactions with IgG: functional and therapeutic consequences. Immunol. Lett. 106, 111–118.

Smith, K.G., Clatworthy, M.R., 2010. FcgRIIb in autoimmunity and infection: evolutionary and therapeutic implications. Nat. Rev. Immunol. 10, 328–343. Tedder, T.F., Klejman, G., Disteche, C.M., Adler, D.A., Schlossman, S.F., Saito, H., 1988. Cloning of a complementary DNA encoding a new mouse B lymphocyte differentiation antigen, homologous to the human B1 (CD20) antigen, and localization of the gene to chromosome 19. J. Immunol. 141, 4388–4394. Vázquez, A.M., Alfonso, M., Lanne, B., Karlsson, K.A., Carr, A., Barroso, O., Fernández, L.E., Rengifo, E., Lanio, M.E., Alvarez, C., Zeuthen, J., Pérez, R., 1995. Generation of a murine monoclonal antibody specific for N-glycolylneuraminic acidcontaining gangliosides that also recognizes sulfated glycolipids. Hybridoma 14, 551–556. Vázquez, A.M., Pérez, A., Hernández, A.M., Macías, A., Alfonso, M., Bombino, G., Pérez, R., 1998. Syngeneic anti-idiotypic monoclonal antibodies to an anti-NeuGccontaining ganglioside monoclonal antibody. Hybridoma 17, 527–534. Vollmers, H.P., Brandlein, S., 2007. Natural antibodies and cancer. J. Autoimmun. 29, 295–302. Walker, J.A., Smith, K.G.C., 2008. Dependence of surface monoclonal antibody binding on dynamic changes in FcgRIIb expression. Immunology 124, 412–418. Winkler, M., Franssen, J.D., Collignon, C., Leo, O., Mariamé, B., van de Walle, P., de Groote, D., Urbain, J., 1979. Idiotypic regulation of the immune system Common idiotypic specificities between idiotypes and antibodies raised against anti-idiotypic antibodies in rabbits. J. Exp. Med. 150, 184–195. Xiang, Z., Cutler, A.J., Brownlie, R.J., Fairfax, K., Lawlor, K.E., Severinson, E., Walker, E.U., Manz, R.A., Tarlinton, D.M., Smith, K.G., 2007. FcgammaRIIb controls bone marrow plasma cell persistence and apoptosis. Nat. Immunol. 8, 419–429. Xu, D., Alegre, M.L., Varga, S.S., Rothermel, A., Collins, A.M., Pulito, V.L., Hanna, L.S., Dolan, K.P., Parren, P., Bluestone, J.A., Jolliffe, L.K., Zivin, R.A., 2000. In Vitro characterization of five humanized OKT3 effector function variant antibodies. Cell. Immunol. 200, 16–26. Yan, X., Evans, S.V., Kaminki, M.J., Gillies, S.D., Reisfeld, R.A., Houghton, A.N., Chapman, P.B., 1996. Characterization of an Ig VH idiotope that results in specific homophilic binding and increased avidity for antigen. J. Immunol. 157, 1582–1588. Zhao, Y., Lou, D., Burke, J., Kohler, H., 2002. Enhanced anti-B-cell tumor effects with anti-CD20 superantibody. J. Immunother. 25, 57–62.