Neutralizing monoclonal antibodies elicited by mosaic RBD nanoparticles bind conserved sarbecovirus epitopes

Increased immune evasion by SARS-CoV-2 variants of concern highlights the need for new therapeutic neutralizing antibodies. Immunization with nanoparticles co-displaying spike receptor-binding domains (RBDs) from eight sarbecoviruses (mosaic-8 RBD-nanoparticles) efficiently elicits cross-reactive polyclonal antibodies against conserved sarbecovirus RBD epitopes. Here, we identified monoclonal antibodies (mAbs) capable of cross-reactive binding and neutralization of animal sarbecoviruses and SARS-CoV-2 variants by screening single mouse B cells secreting IgGs that bind two or more sarbecovirus RBDs. Single-particle cryo-EM structures of antibody-spike complexes, including a Fab-Omicron complex, mapped neutralizing mAbs to conserved class 1/4 RBD epitopes. Structural analyses revealed neutralization mechanisms, potentials for intra-spike trimer cross-linking by IgGs, and induced changes in trimer upon Fab binding. In addition, we identified a mAb-resembling Bebtelovimab, an EUA-approved human class 3 anti-RBD mAb. These results support using mosaic RBD-nanoparticle vaccination to generate and identify therapeutic pan-sarbecovirus and pan-variant mAbs.

Antibody-based redirection of universal Fabrack-CAR T cells selectively kill antigen bearing tumor cells

Background

Chimeric antigen receptor (CAR) T cells engineered to recognize and target tumor associated antigens have made a profound impact on the quality of life for many patients with cancer. However, tumor heterogeneity and intratumoral immune suppression reduce the efficacy of this approach, allowing for tumor cells devoid of the target antigen to seed disease recurrence. Here, we address the complexity of tumor heterogeneity by developing a universal CAR.

Method

We constructed a universal Fabrack-CAR with an extracellular domain composed of the non-tumor targeted, cyclic, twelve residue meditope peptide that binds specifically to an engineered binding pocket within the Fab arm of monoclonal antibodies (mAbs). As this site is readily grafted onto therapeutic mAbs, the antigen specificity of these universal Fabrack-CAR T cells is simply conferred by administering mAbs with specificity to the heterogeneous tumor.

Results

Using in vitro and in vivo studies with multiple meditope-engineered mAbs, we show the feasibility, specificity, and robustness of this approach. These studies demonstrate antigen- and antibody-specific T cell activation, proliferation, and IFNγ production, selective killing of target cells in a mixed population, and tumor regression in animal models.

Conclusion

Collectively, these findings support the feasibility of this universal Fabrack-CAR T cell approach and provide the rationale for future clinical use in cancer immunotherapy.

Manipulating mIgD-expressing B cells with anti-migis-δ monoclonal antibodies

Surface IgD and IgM doubly positive cells comprise the major population of B cells in the human immune system. The heavy chain of membrane-bound IgD (mδ) differs from that of IgD (δ) in that mδ contains a C-terminal membrane-anchor peptide. Our group previously proposed that the N-terminal extracellular segment of 27 aa residues of the membrane-anchor peptide of mδ, referred to as the mIg isotype-specific-δ (migis-δ) segment, may provide a unique antigenic site for isotype-specific targeting of mIgD(+) B cells. Here we report the preparation of mouse mAbs specific for human migis-δ. The mAbs bound to human migis-δ-containing recombinant proteins in an ELISA and to mIgD-expressing transfectants of a CHO cell line as analyzed by flow cytometry. MAb 20E6, which binds to an epitope toward the N-terminal of human migis-δ, could stain human B cell line MC116, which expressed mIgD and mIgM. MC116 cells could be induced to undergo apoptosis by treatment with 20E6 in the presence of a second crosslinking antibody. Chimeric 20E6 caused antibody-dependent cellular cytotoxicity of MC116 cells in the presence of human PBMCs as the source of effector cells. In cultures of PBMCs, 20E6 down-regulated the population of mIgD(+) B cells. The production of human IgM by transplanted MC116 cells in NOD-SCID (NOD.CB17-Prkdc(scid)/IcrCrlBltw) mice could be suppressed by 20E6. These results encourage further investigation of the potential of anti-migis-δ mAbs to control mIgD(+) B cells, when such a manipulation may alleviate a disease state.

CϵmX peptide-carrying HBcAg virus-like particles induced antibodies that down-regulate mIgE-B cells (53.7)

Type-I hypersensitivity reactions play a critical role in the pathogenesis of various allergic diseases. The successful development of the anti-IgE antibody, omalizumab, has validated IgE as an effective therapeutic target for the treatment of various IgE-mediated allergic diseases. Two research groups have reported that mAbs specific for certain parts of CϵmX, a domain of 52 aa residues in human membrane-bound IgE (mIgE), can cause the lysis of mIgE-B cells by apoptosis and antibody-dependent cellular cytotoxicity (ADCC). Herein, we explore the virus-like particles (VLPs) formed by hepatitis B virus core antigen (HBcAg) that harbors the entire CϵmX peptide or its fragments as immunogens for inducing anti-CϵmX antibodies. Mice immunized subcutaneously with some of these immunogens produce antibodies that bind to recombinant CϵmX-containing human IgE.Fc in ELISA and Western blot analysis and to genetically engineered human mIgE-expressing Ramos B cell line in fluorescence flow cytometric assays. The IgG antibodies purified from the sera of immunized mice could cause the apoptosis of mIgE-expressing Ramos cells through a BCR-dependent caspase pathway. Furthermore, the IgG mediated ADCC on human mIgE-expressing murine B cell lymphoma A20. These results suggest that HBcAg-CϵmX peptide immunogens warrant further investigation as a therapeutic modality for modulating IgE in patients with IgE-mediated allergic diseases.

Lipid rafts hinder binding of antibodies to the extracellular segment of the membrane-anchor peptide of mIgA

Membrane-bound IgA (mIgA) is associated with Igα/Igβ as the B cell receptor (BCR) complex on mIgA-expressing B cells. The α chain of mIgA (mα) contains a C-terminal membrane-anchor peptide, which encompasses extracellular, transmembrane and intracellular segments. The extracellular segment, referred to as the mIg isotype-specific (migis-α) segment or the extracellular membrane proximal domain of mα, has been proposed to be a specific antigenic site suitable for isotype-specific targeting of mIgA-expressing B cells by antibodies. In this study, we developed several anti-migis-α monoclonal antibodies (mAbs), such as mAb 29C11, specific to a segment towards the N-terminus of the 26 amino acid long migis-α. The mAbs bound strongly to synthetic peptides of migis-α and to various recombinant proteins containing migis-α as revealed by ELISA. On B cells, however, flow cytometric analysis suggested that these mAbs did not bind strongly to mIgA. After lipid rafts of B cells were disrupted by cholesterol extraction, the mAbs were able to bind strongly to the treated B cells. Moreover, immunoprecipitation analysis of these mAbs indicated that mIgA could only be pulled down by the mAbs when mIgA-expressing B cells were solubilized by strong detergents, such as sodium dodecyl sulfate (SDS), or when lipid rafts were disrupted. Together, these results suggest that the migis-α region of mIgA in the BCR is associated with lipid rafts, which hinder binding of migis-α-specific antibodies to mIgA on the cell surface. Further studies are in progress to evaluate the suitability of 29C11 or its affinity-improved variants for targeting mIgA-expressing B cells.

Unique epitopes on C epsilon mX in IgE-B cell receptors are potentially applicable for targeting IgE-committed B cells

Membrane-bound IgE (mIgE) is part of the IgE-BCR and is essential for generating isotype-specific IgE responses. On mIgE(+) B cells, the membrane-bound epsilon-chain (mepsilon) exists predominantly in the long isoform, mepsilon(L), containing an extra 52 aa CepsilonmX domain between CH4 and the C-terminal membrane-anchoring segment; the short isoform of mepsilon, mepsilon(S), exists in minor proportions. CepsilonmX thus provides an attractive site for immunologic targeting of mIgE(+) B cells. In this study, we show that nine newly prepared CepsilonmX-specific mAbs, as well as the previously reported a20, bound to mIgE.Fc(L)-expressing CHO cells, while only 4B12 and 26H2 bound to mIgE.Fc(L)-expressing B cell line Ramos cells. The mAb 4B12 bound to the N-terminal part, 26H2 the middle part, and all others the C-terminal part of CepsilonmX. Expression of Igalpha and Igbeta on the mIgE.Fc(L)-CHO cells reduces the binding of a20 to CepsilonmX as compared with that of 4B12 and 26H2. The chimeric mAbs c4B12 and c26H2, when cross-linked by secondary antibodies, lysed mIgE.Fc(L)-Ramos cells by apoptosis through a BCR-dependent caspase pathway. Using PBMCs as the source of effector cells, c4B12 and c26H2 demonstrated Ab-dependent cellular cytotoxicity toward mIgE.Fc(L)-Ramos cells in a dose-dependent fashion. In cultures of PBMCs from atopic dermatitis patients, c4B12 and c26H2 inhibited the synthesis of IgE driven by anti-CD40 and IL-4. These results suggest that 4B12 and 26H2 and an immunogen using the peptide segments recognized by these mAbs are potentially useful for targeting mIgE(+) B cells to control IgE production.

Anti-CmεX monoclonal antibodies potentially applicable for targeting mIgE-expressing B cells (140.20)

On B lymphocytes, membrane-bound IgE (mIgE) is part of BCR and essential for generating isotype-specific IgE response. On mIgE+ B cells in man, the membrane-bound ε chain (mε) exists predominantly in the long form (mεL), containing 52-a.a. CεmX domain between CH4 and the C-terminal membrane-anchoring segment; the conventional form of mε (mεS) exists in minor proportions. CεmX thus provides an attractive site for immunological targeting of mIgE+ cells. Here we have shown that 9 newly prepared CεmX-specific monoclonal antibodies (mAbs), as well as previously reported a20, can bind to mIgE.FcL-expressing CHO cells, while only 4B12 and 26H2 can bind to mIgE.FcL-expressing B cell line Ramos cells. In epitope mapping studies using synthetic peptides representing different parts of CεmX, 4B12 was shown to bind to the N-terminal part, 26H2 the middle part, and all others the C-terminal part of CεmX. These results suggest that the C-terminal part of CεmX is sterically hindered for mAb binding by associated components in the BCR. The chimerized mAbs c4B12 and c26H2, when cross-linked by secondary antibodies, could lyse mIgE.FcL-expressing Ramos cells by apoptosis through a BCR-dependent caspase pathway. Using PBMCs as the source of effector cells, c4B12 and c26H2 could cause ADCC against mIgE.FcL-expressing Ramos cells in a does-dependent fashion. These results indicate that anti-CεmX mAbs, 4B12 and 26H2, can bind to native mIgE on B cells and are potentially useful for targeting mIgE+ B cells for controlling IgE production.

Humanization of an anti-CεmX monoclonal antibody 4B12 for potential clinical use in targeting mIgE+B cells (140.19)

The CεmX domain of 52 a.a. residues, located between the CH4 domain and membrane-anchor segment of human membrane-bound ? chain on mIgE+ B cells, is a unique antigenic site for the immunological targeting of those B cells. We have demonstrated that a chimeric form of anti-CεmX monoclonal antibody (mAb), 4B12, can induce apoptosis and ADCC against a human mIgE.Fc+ B cell transfectoma via effector cells from PBMC in vitro (presented by another abstract in this meeting). To enhance the clinical applicability of 4B12, we have humanized it by first grafting its six complementary-determining regions (CDRs) into human Ig VH subgroup IV and V? subgroup II frameworks (FWs). Those residues in the FWs of VH and VL of the parental murine 4B12 that are essential for Fv dimer interaction and CDR conformation were identified by analyzing the binding affinity of a number of designed Ab variants with site-directed mutations, using competitive ELISA and surface plasmon resonance analysis. A humanized 4B12 retaining the affinity of its parent was obtained by adopting all FWs of the selected human V? template without any change and all FWs of the selected VH with only one residue change. The Val at position 71 of the human VH was replaced by Arg at the corresponding position of the murine VH. Based on molecular modeling analysis, it is likely that such a change is probably crucial for maintaining the proper structural packing of Arg71 with Ile29 in the CDRH1 and Gly55 in the CDRH2. The KD of the chimeric and humanized 4B12 are 5.37×10-9 M-1 and 5.76×10-9 M-1, respectively. Affinity enhancement of humanized 4B12 by selective CDR mutations is in progress.

Alleles and isoforms of human membrane-bound IgA1

In humans, IgA exists as two subclasses, IgA1 and IgA2, which contain distinct alpha1 and alpha2 heavy chains, respectively. Both subclasses also have membrane-bound forms (mIgA1 and mIgA2) containing the corresponding malpha1 and malpha2 heavy chains, which differ from alpha1 and alpha2 by an additional "membrane-anchor" peptide segment extending from the CH3 domain of alpha1 and alpha2. The membrane-anchor segment has three parts: an extracellular, a transmembrane, and an intracellular segment. The heavy chain malpha1 exists in short and long isoforms, referred to as malpha1S and malpha1L, with the latter containing extra 6 amino acid residues, GSCSVA, at the N-terminus of the extracellular segment (residues 453-458). By studying the genomic and mRNA sequences of malpha1 and malpha2 from 30 individuals residing in Taiwan, we have found that, in addition to the known malpha1 allele, referred to as malpha1(456S), malpha1 also has a previously unknown allele, referred to as malpha1(456C) (GenBank accession no. EU431191). This newly identified allele is present in the donor population at a similar proportion to malpha1(456S), and appears to exist only as the long isoform, i.e. malpha1L, rather than the short isoform, malpha1S. Furthermore, we confirmed that malpha2 exists only as the short isoform. Future studies will examine whether these mIgA1 variations affect the regulation of IgA synthesis and whether mIgA1 can provide an antigenic site for the immunological targeting of IgA-expressing B cells.

Alleles and isoforms of human membrane-bound IgA1 (35.50)

In humans, IgA has two subclasses, IgA1 and IgA2, which contain α1 and α2 heavy chains, respectively. For both subclasses, there are also membrane-bound forms, mIgA1, and mIgA2, containing mα1 and mα2 heavy chains. Mα1 and mα2 differ from α1 and α2 by having a “membrane-anchor” peptide segment extending from the CH3 domain of α1 and α2. The membrane-anchor segment has three parts: an extracellular, a trans-membrane, and an intracellular segments. It was found previously that mα1 exists in a short and a long isoforms, referred to as mα1(I) and mα1(II), with the latter containing extra 6 amino acid residues, GSCSVA (residues 453–458), at the N-terminal of the extracellular segment. By studying the genomic and mRNA sequences of mα1 and mα2 from 30 individuals residing in Taiwan, we have found that, in addition to mα1(456S) allele, mα1 has a previously unknown allele, referred to as mα1(456C), which is present in the donor population at about the same proportion as mα1(456S). Moreover, mα1(456C) allele exists in the long isoform only and does not yield a short isoform corresponding to mα1(I). Furthermore, we have clarified that mα2 exist as short isoforms only. Future studies will examine the variations in mIgA1 in other races and the possible association of such variations with the regulation of IgA synthesis.

This research is supported by a grant, #94-2320-B007-004, from the National Science Council, Taiwan.