Functional characterization of LFA-1 antigens in the interaction of human NK clones and target cells

In the present studies we analyzed the role of LFA-1 antigens in the interaction between NK clones and target cells. The use of various cloned NK cell lines allowed us to analyze homogeneous populations of NK cells which ordinarily comprise only a small fraction of peripheral blood lymphocytes and are extremely heterogeneous with respect to phenotype and specificity. Indirect immunofluorescence with monoclonal antibodies against the alpha (MHM24) and beta (MHM23) chains of the LFA-1 antigen revealed similar patterns of positive reactivity with all NK clones. Both monoclonal antibodies exerted a significant blocking effect on NK cytotoxicity against target cells such as Molt-4 and CEM, whereas the inhibition was very weak against other targets such as K562 and HSB cells. Additive blocking effects were seen when both monoclonal antibodies MHM23 and MHM24 were added to the cytotoxicity assays. When we compared the inhibitory effect of MHM23 and MHM24 on uncultured peripheral blood NK cells and IL 2-activated NK cells, inhibition of cytotoxicity also was found to be primarily dependent on the individual target cells. Thus, the inhibitory activity of anti-LFA-1 antibody was shown to be independent of the phenotypic and functional heterogeneity of the NK clones, activated NK cells, and unstimulated NK cells utilized in these studies. These blocking effects were found to be independent of the LFA-1 antigen expression on the target cell membrane and inhibition occurred only when antibody was bound to the effector cells. Comparison of the effects of anti-LFA-1, anti-T3, and anti-clonotypic antibodies against a Ti-like structure of different NK clones with a mature T cell phenotype demonstrated that each of these antibodies acts on the effector cells in an independent and additive fashion. However, unlike T3 and NKTa antigen, LFA-1 antigen expression is not modulated by cell surface interaction with antibodies specific for this molecule.     

Five epitopes of a differentiation antigen on human inducer T cells distinguished by monoclonal antibodies

A series of mouse monoclonal antibodies reacting with human T cells of the helper/inducer subclass, OKT4, 4A, 4B, 4C, and 4D, have been reported. Using double-fluorescent staining and complement-mediated depletion, it was shown that the antigen(s) recognized by OKT4, 4A, 4B, 4C, and 4D antibodies are present on the same cell. Using FITC-labeled OKT4, 4A, 4B, 4C and 4D, a lack of competition between the antibodies for their epitopes was shown. Immunoprecipitation of the antigen recognized by each antibody yielded a molecule of approximately 60,000-62,000 Da. Sequential precipitation with several antibodies resulted in a minimum of additional precipitated antigen following removal of the first antigen. Capping of cell surface antigen with OKT4, followed by staining with OKT4, 4A, 4B, 4C, or 4D, indicated that the epitopes for all five antibodies co-cap. A sandwich ELISA assay using OKT4 and the other antibodies showed that molecules binding to OKT4A, 4B, 4C, and 4D also bound OKT4. It can therefore be concluded that monoclonal antibodies OKT4, 4A, 4B, 4C, and 4D recognize distinct epitopes present on a molecule of approximately 60-62,000 Da on human helper/inducer T cells.     

The murine IL 2 receptor. I. Monoclonal antibodies that define distinct functional epitopes on activated T cells and react with activated B cells

The properties of three distinct rat monoclonal antibodies, designated 3C7, 7D4, and 2E4, to the murine IL 2 receptor have been compared in binding, biochemical, and functional assays. 3C7 appears to define an epitope near or identical to the IL 2-binding site of the receptor, because 3C7 inhibited the binding of radiolabeled IL 2 to CTL-L cells and because unlabeled IL 2 inhibited the binding of FITC-3C7 to CTL-L cells. 7D4 and 2E4 had no effect on IL 2 binding. Competitive antibody-binding studies confirmed that the epitope seen by 3C7 was distinct from the epitope(s) seen by 7D4 and 2E4. Sequential immunoprecipitation studies demonstrated that all three antibodies were reactive with the same molecular species, and that each precipitated identical components of 20,000 to 25,000 daltons, 50,000 to 60,000 daltons, and 100,000 to 120,000 daltons from the surface of CTL-L. FACS studies demonstrated a quantitatively and qualitatively identical cell distribution for the antigen defined by each antibody. They failed to stain more than 95% of resting lymphocytes, but were strongly reactive with Con A T blasts and substantially less reactive with LPS B blasts. Unlabeled IL 2 was also able to inhibit the binding of FITC-3C7 to LPS B cell blasts, suggesting the presence of IL 2-binding sites on activated B cells. Each antibody inhibited IL 2-driven proliferation of HT2 or CTL-L cells. 3C7 and 7D4 were more potent inhibitors of proliferation than was 2E4, and the combined use of 3C7 and 7D4 resulted in greater levels of inhibition of proliferation than that shown from the use of either antibody alone. Collectively, the results support the hypothesis that these antibodies detect two distinct functional regions of the IL 2 receptor.     

Development of a Streptavidin-Conjugated Single-Chain Antibody That Binds Bacillus cereus Spores

Control of microorganisms such as Bacillus cereus spores is critical to ensure the safety and a long shelf life of foods. A bifunctional single chain antibody has been developed for detection and binding of B. cereus T spores. The genes that encode B. cereus T spore single-chain antibody and streptavidin were connected for use in immunoassays and immobilization of the recombinant antibodies. A truncated streptavidin, which is smaller than but has biotin binding ability similar to that of streptavidin, was used as the affinity domain because of its high and specific affinity with biotin. The fusion protein gene was expressed in Escherichia coli BL21 (DE3) with the T7 RNA polymerase-T7 promoter expression system. Immunoblotting revealed an antigen specificity similar to that of its parent native monoclonal antibody. The single-chain antibody-streptavidin fusion protein can be used in an immunoassay of B. cereus spores by applying a biotinylated enzyme detection system. The recombinant antibodies were immobilized on biotinylated magnetic beads by taking advantage of the strong biotin-streptavidin affinity. Various liquids were artificially contaminated with 5 × 10⁴ B. cereus spores per ml. Greater than 90% of the B. cereus spores in phosphate buffer or 37% of the spores in whole milk were tightly bound and removed from the liquid phase by the immunomagnetic beads.     

Isolation of anti-angiogenesis antibodies from a large combinatorial repertoire by colony filter screening

We describe here a method, based on iterative colony filter screening, for the rapid isolation of binding specificities from a large synthetic repertoire of human antibody fragments in single-chain Fv configuration. Escherichia coli cells, expressing the library of antibody fragments, are grown on a porous master filter, in contact with a second filter coated with the antigen, onto which antibodies secreted by the bacteria are able to diffuse. Detection of antigen binding on the second filter allows the recovery of a number of E.coli cells, including those expressing the binding specificity of interest, which can be submitted to a second round of screening for the isolation of specific monoclonal antibodies. We tested the methodology using as antigen the ED-B domain of fibronectin, a marker of angiogenesis. From an antibody library of 7 × 10⁸ clones, we recovered a number of specifically-binding antibodies of different aminoacid sequence. The antibody clone showing the strongest enzyme-linked immunosorbent assay signal (ME4C) was further characterised. Its epitope on the ED-B domain was mapped using the SPOT synthesis method, which uses a set of decapeptides spanning the antigen sequence synthesised and anchored on cellulose. ME4C binds to the ED-B domain with a dissociation constant K_d = 1 × 10^–7 M and specifically stains tumour blood vessels, as shown by immunohistochemical analysis on tumour sections of human and murine origin.     

Molecular basis of a unique tumor antigen of radiation leukemia virus-induced leukemia B6RV2: its relation to MuLV gp70 of xenotropic class

Hybridomas secreting monoclonal antibodies that reacted with the B6 radiation leukemia virus (RadLV)-induced leukemia B6RV2 were produced by fusion of BALB/c NS-1 myeloma cells with spleen cells from (BALB/c X B6)F1 mice immunized with B6RV2. By direct and absorption analyses with 28 B6 and BALB/c leukemias, the monoclonal antibodies NU7-4 and NU7-99 were shown to react only with B6RV2, indicating that they recognized an individually distinct antigen on B6RV2 that was identified previously with conventional (BALB/c X B6)F1 anti-B6RV2 serum. Another monoclonal antibody, NU1-132, showed relatively restricted reactivity with B6 RadLV leukemias. These three monoclonal antibodies all precipitated material of approximately 80,000 daltons, which is the same size as that precipitated by anti-xenotropic MuLV gp70 serum. Sequential immunoprecipitation analysis revealed that the molecules precipitated by NU7-4 were not removed by pretreatment of NU7-99 or NU1-132 and that the molecules precipitated by NU7-99 were not removed by NU7-4 or NU1-132. The molecules precipitated by NU1-132 were partially removed by pretreatment with NU7-4, but not with NU7-99. The molecules precipitated by these three monoclonal antibodies were removed by pretreatment with anti-xenotropic gp70. These results suggested heterogeneity of the xenotropic MuLV gp70-related molecules expressed on B6RV2 and a possible relation between serologically defined unique tumor antigens and gp70-related molecules.     

Quantitative expression of epitopes defined by murine monoclonal antibodies on DR molecules from different HLA haplotypes

Monoclonal antibodies 50D6 and 21r5, reactive with human class II molecules, were analyzed quantitatively by flow cytometry and cellular radioimmunoassay for their binding to cells of different HLA-DR types. Monoclonal antibody 50D6 bound equally to cells of all DR types tested except DR7, where no reactivity was observed. Monoclonal antibody 21r5 was reactive with all cells. However, the percentage of DR molecules at the cell surface expressing 21r5 epitope varied with the DR type and increased as follows: DR3 = DR7 less than DR2 less than DR5 less than DR4 less than DR1. MAb 50D6 reacted with an epitope spatially related to but distinct from the 21w4 epitope present on all DR molecules. The 50D6 epitope was shown to be present on isolated DR1 molecules.     

Engagement of Two Distinct Binding Domains on CCL17 Is Required for Signaling through CCR4 and Establishment of Localized Inflammatory Conditions in the Lung

CCL17 (TARC) function can be completely abolished by mAbs that block either one of two distinct sites required for CCR4 signaling. This chemokine is elevated in sera of asthma patients and is responsible for establishing inflammatory sites through CCR4-mediated recruitment of immune cells. CCL17 shares the GPCR CCR4, with CCL22 (MDC) but these two chemokines differentially affect the immune response. To better understand chemokine mediated effects through CCR4, we have generated chimeric anti-mouse CCL17 surrogate antibodies that inhibit function of this ligand in vitro and in vivo. The affinities of the surrogate antibodies for CCL17 range from 685 pM for B225 to 4.9 nM for B202. One antibody, B202, also exhibits weak binding to CCL22 (K_D∼2 µM) and no binding to CCL22 is detectable with the second antibody, B225. In vitro, both antibodies inhibit CCL17-mediated calcium mobilization, β-arrestin recruitment and chemotaxis; B202 can also partially inhibit CCL22-mediated β-arrestin recruitment. Both B202 and B225 antibodies neutralize CCL17 in vivo as demonstrated by reduction of methacholine-induced airway hyperreactivity in the A. fumigatus model of asthma. That both antibodies block CCL17 function but only B202 shows any inhibition of CCL22 function suggests that they bind CCL17 at different sites. Competition binding studies confirm that these two antibodies recognize unique epitopes that are non-overlapping despite the small size of CCL17. Taking into consideration the data from both the functional and binding studies, we propose that effective engagement of CCR4 by CCL17 involves two distinct binding domains and interaction with both is required for signaling.     

Kinetics of T-cell receptor binding by bivalent HLA-DR. Peptide complexes that activate antigen-specific human T-cells

Monovalent major histocompatibility complex-peptide complexes dissociate within seconds from the T-cell receptor (TCR), indicating that dimerization/multimerization may be important during early stages of T-cell activation. Soluble bivalent HLA-DR2.myelin basic protein (MBP) peptide complexes were expressed by replacing the F(ab) arms of an IgG2a antibody with HLA-DR2.MBP peptide complexes. The binding of bivalent HLA-DR2.peptide complexes to recombinant TCR was examined by surface plasmon resonance. The bivalent nature greatly enhanced TCR binding and slowed dissociation from the TCR, with a t((1)/(2)) of 2.1 to 4.6 min. Soluble bivalent HLA-DR2.MBP peptide complexes activated antigen-specific T-cells in the absence of antigen presenting cells. In contrast, soluble antibodies to the TCR.CD3 complex were ineffective, indicating that they failed to induce an active TCR dimer. TCR/CD3 antibodies induced T-cell proliferation when bound by antigen presenting cells that expressed Fc receptors. In the presence of dendritic cells, bivalent HLA-DR2. MBP peptide complexes induced T-cell activation at >100-fold lower concentrations than TCR/CD3 antibodies and were also superior to peptide or antigen. These results demonstrate that bivalent HLA-DR. peptide complexes represent effective ligands for activation of the TCR. The data support a role for TCR dimerization in early TCR signaling and kinetic proofreading.     

The interaction of monoclonal antibodies with MHC class I antigens on mouse spleen cells. I. Analysis of the mechanism of binding

We studied the mechanism of binding of radiolabeled, monoclonal anti-H-2 antibodies to mouse spleen cells to determine the number of H-2 antigen molecules per cell. Equilibrium and kinetic data were analyzed in detail according to theoretical models developed for different modes of antibody binding. The results of binding experiments from three monoclonal IgG antibodies (36-7-5, anti-Kk; 27-11-13, anti-DbDd; and 11-4-1, anti-Kk) and their F(ab')2 and F(ab') fragments show that for the IgG and F(ab')2 from all three antibodies, the monovalently and bivalently bound states of the antibody co-exist in rapid equilibrium with one another on the cell surface, with the bivalent state predominating. We show that the relative proportions of the monovalently and bivalently bound species can be estimated from dissociation kinetics experiments, and that once the mode of antibody binding has been established, the density of H-2 determinants on the cell surface can be estimated from equilibrium-binding data. We conclude that the average numbers of H-2K and H-2D molecules on B10.A spleen cells are 5 X 10(4) and 1.1 X 10(5) molecules/cell, respectively.     

Production and characterization of a monoclonal antibody specific for the human lymphocyte low affinity receptor for IgE: CD 23 is a low affinity receptor for IgE

In the present study a gamma 1 kappa monoclonal antibody, Mab 25, specific for the receptor for the Fc fragment of IgE on lymphocytes (Fc epsilon RL) was established. This antibody was generated after fusion of spleen cells from mice immunized with the EBV-transformed lymphoblastoid cell line RPMI 8866, which is known to express Fc epsilon RL at high density. Mab 25 inhibits strongly the binding of IgE to RPMI 8866 cells and to other Fc epsilon RL-positive EBV-transformed lymphoblastoid cell lines. A 50% inhibition of IgE binding was observed at a Mab 25 concentration of 10 ng/ml. The binding of IgE was also inhibited by Fab fragments of Mab 25, suggesting that the inhibition is not simply due to steric hindrance or to an eventual binding through its Fc portion. Mab 25 only binds to cell lines expressing Fc epsilon RL. Mab 25 immunoprecipitated a single polypeptide with an apparent m.w. of 42 Kd, pI 4.9. The membrane molecule bound to and eluted from a Mab 25 immunoabsorbent had the same apparent m.w. and pI as the Fc epsilon RL purified from an IgE immunoabsorbent. Additionally, when RPMI 8866 cell lysates were cleared with Mab 25, no Fc epsilon RL could be bound to or eluted from an IgE immunoabsorbent. Mab 25 was found to weakly bind to a minor proportion of blood (1 to 4%), tonsil (2 to 9%) and spleen (4 to 5%) mononuclear cells with a low intensity. By double fluorescence analysis, most of the Fc epsilon RL-positive cells were found to be CD 20-positive B lymphocytes. The staining pattern of Mab 25 and the biochemical characteristics of the antigen detected by Mab 25 were comparable to those of the CD 23 Mab. The four CD 23 Mab MHM 6, PL 13, HD 50, and Tü 1 were found to inhibit the binding of IgE. PL 13 was found to totally inhibit the binding of Mab 25 to RPMI 8866 cells, whereas Tü 1 and MHM 6 only partially inhibited Mab 25 binding. HD 50 was unable to block the binding of Mab 25. The finding that different CD 23/Fc epsilon RL-specific monoclonal antibodies recognizing distinct epitopes have in common the capacity of inhibiting the binding of IgE suggests that upon binding they induce a conformational alteration of the Fc epsilon RL resulting in a loss of the IgE binding capacity. In conclusion, our data demonstrate that the CD 23 antigen is a low affinity receptor for IgE on lymphocytes.     

Two-dimensional gel analysis of a second family of class II molecules by polymorphic HLA-DR4,5, and w9 monoclonal antibodies

Human Ia-like, class II molecules were isolated by immunoprecipitation with monoclonal antibodies from various HLA-D/DR homozygous cell lines and were analyzed by two-dimensional gel electrophoresis. The monoclonal antibody PLM12 reacted with B cells carrying DR4, DR5, DRw6.2, and DRw9 phenotypes, and its reactivity perfectly correlated with the previously defined TB21 (MB3-like) specificity. Class II molecules detected by PLM12 were structurally distinct from those precipitated by the anti-DR monoclonal antibody NC1 on all HLA-DR4, DR5, DRw6.2, and DRw9 homozygous cell lines and showed polymorphism in heavy and light chains among these cell lines. The monoclonal antibodies PLM2 and PLM9 only reacted with B cells carrying DR5 and DRw6.2 and also detected a distinct set of class II molecules from those precipitated by NC1 but identical to those of PLM12. Thus, PLM2 and PLM9 serologically detected a new subtypic antigen of the PLM12-reactive class II molecules. Furthermore, the antibody NC1 precipitated two light chains and one heavy chain from HLA-DRw6.2 homozygous cell line EBV-Sh. The result indicated the presence of three sets of class II molecules: two in a DR family and another carrying the polymorphic determinants detected by PLM2, PLM9, and PLM12 in a second family.     

Recombinant human monoclonal antibodies

To produce human monoclonal antibodies in bacteria, a gene repertoire of IgM variable regions was isolated from human peripheral B lymphocytes by the polymerase chain reaction. Alternatively, synthetic antibody genes with random hypervariable regions are being generated that may provide libraries of even higher complexity. For the selection of specific monoclonal antibodies from these libraries, we have developed twoE. coli vector systems that facilitate the surface display of an antibody physically linked to its own gene. The phagemid pSEX encodes a fusion protein of an antigen binding domain (Fv-antibody) with the docking protein (pIII) of filamentous phages. Specific antibody genes can therefore be enriched by antigen affinity chromatography. The plasmid pAP1 encodes a fusion protein of an Fv-antibody with a bacterial cell-wall protein. Bacteria carrying this plasmid express functional Fv-antibodies tightly bound to their surface. This should enable the selection of single cells with a fluorescence-assisted cell sorter (FACS) using labeled antigen or by adsorption to immobilized antigen. These vectors permit three major principles of the antibody response to be mimicked inE. coli:

1. Generation of a highly complex antibody repertoire;
2. Clonal selection procedures for library screening; and
3. The possibility of increasing a given affinity by repeated rounds of mutation and selection.

     

UH series of monoclonal antibodies recognizing major histocompatibility complex class II antigen(s) of Japanese monkeys (Macaca fuscata)

Six mouse monoclonal antibodies were developed by immunization with a Japanese monkey cell line. These monoclonal antibodies, designated the UH series, reacted with large populations of peripheral B cells and monocytes, but not with T cells. The distribution of reactivities and the molecular weight of the membrane antigens recognized were similar to those of the HLA-DR monoclonal antibody; one inhibited the binding of HLA-DR. Human interferon-gamma induced increased expressions of all the UH antigen epitopes.     

Crystal Structure of the Engineered Neutralizing Antibody M18 Complexed to Domain 4 of the Anthrax Protective Antigen

The virulence of Bacillus anthracis is critically dependent on the cytotoxic components of the anthrax toxin, lethal factor (LF) and edema factor (EF). LF and EF gain entry into host cells through interactions with the protective antigen (PA), which binds to host cellular receptors such as CMG2. Antibodies that neutralize PA have been shown to confer protection in animal models and are undergoing intense clinical development. A murine monoclonal antibody, 14B7, has been reported to interact with domain 4 of PA (PAD4) and block its binding to CMG2. More recently, the 14B7 antibody was used as the platform for the selection of very high affinity, single-chain antibodies that have tremendous potential as a combination anthrax prophylactic and treatment. Here, we report the high-resolution X-ray structures of three high-affinity, single-chain antibodies in the 14B7 family; 14B7 and two high-affinity variants 1H and M18. In addition, we present the first neutralizing antibody-PA structure, M18 in complex with PAD4 at 3.8 Å resolution. These structures provide insights into the mechanism of neutralization, and the effect of various mutations on antibody affinity, and enable a comparison between the binding of the M18 antibody and CMG2 with PAD4.     

Antibodies binding diverse pertactin epitopes protect mice from Bordetella pertussis infection

Infection by the bacterium Bordetella pertussis continues to cause considerable morbidity and mortality worldwide. Many current acellular pertussis vaccines include the antigen pertactin, which has presumptive adhesive and immunomodulatory activities, but is rapidly lost from clinical isolates after the introduction of these vaccines. To better understand the contributions of pertactin antibodies to protection and pertactin''s role in pathogenesis, we isolated and characterized recombinant antibodies binding four distinct epitopes on pertactin. We demonstrate that four of these antibodies bind epitopes that are conserved across all three classical Bordetella strains, and competition assays further showed that antibodies binding these epitopes are also elicited by B. pertussis infection of baboons. Surprisingly, we found that representative antibodies binding each epitope protected mice against experimental B. pertussis infection. A cocktail of antibodies from each epitope group protected mice against a subsequent lethal dose of B. pertussis and greatly reduced lung colonization levels after sublethal challenge. Each antibody reduced B. pertussis lung colonization levels up to 100-fold when administered individually, which was significantly reduced when antibody effector functions were impaired, with no antibody mediating antibody-dependent complement-induced lysis. These data suggest that antibodies binding multiple pertactin epitopes protect primarily by the same bactericidal mechanism, which overshadows contributions from blockade of other pertactin functions. These antibodies expand the available tools to further dissect pertactin''s role in infection and understand the impact of antipertactin antibodies on bacterial fitness.     

Optimizing immobilization on two-dimensional carboxyl surface: pH dependence of antibody orientation and antigen binding capacity

The performance of immunosensors is highly dependent on the amount of immobilized antibodies and their remaining antigen binding capacity. In this work, a method for immobilization of antibodies on a two-dimensional carboxyl surface has been optimized using quartz crystal microbalance biosensors. We show that successful immobilization is highly dependent on surface pK_a, antibody pI, and pH of immobilization buffer. By the use of EDC/sulfo-NHS (1-ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride/N-hydroxysulfosuccinimide) activation reagents, the effect of the intrinsic surface pK_a is avoided and immobilization at very low pH is therefore possible, and this is important for immobilization of acidic proteins. Antigen binding capacity as a function of immobilization pH was studied. In most cases, the antigen binding capacity followed the immobilization response. However, the antigen-to-antibody binding ratio differed between the antibodies investigated, and for one of the antibodies the antigen binding capacity was significantly lower than expected from immobilization in a certain pH range. Tests with anti-Fc and anti-Fab₂ antibodies on different antibody surfaces indicated that the orientation of the antibodies on the surface had a profound effect on the antigen binding capacity of the immobilized antibodies.     

A three-step strategy for targeting drug carriers to human ovarian carcinoma cells in vitro.

To improve tumor-to-tissue ratios of anticancer agents in radioimmunotherapy, a three-step targeting approach was used to deliver biotinylated liposomes to human ovarian cancer cells (NIH:OVCAR-3, SK-OV-3) in vitro. Targeting was based upon the use of two antibodies specific for the CA-125 antigen that is highly expressed on NIH:OVCAR-3 cells but not expressed on SK-OV-3 cells. Briefly, the approach consists of prelabeling target cells with biotinylated anti-CA-125 antibody and FITC-labeled streptavidin (SAv) prior to administration of biotinylated liposomes containing a marker dye for visualization by confocal laser scanning microscopy (CLSM). In addition, the two anti-CA-125 antibodies (B27.1 and B43.13) were labeled with FITC and incubated with ovarian cancer cells at 37 degrees C from 30 min to 24 h to study binding and uptake kinetics. Shedding kinetics of bound antibody from tumor cells was performed using radiolabeled B27.1. Results demonstrated that both B27.1 and B43.13 specifically bound to the cell surface of OVCAR-3 cells but not to SK-OV-3 cells. Biotinylation, FITC-labeling and radiolabeling of the antibodies did not compromise immunoreactivity. Less than 6% of the bound B27.1 was shed from tumor cells by 4 h following incubation, and the antibody-antigen complex resided predominantly on the cell surface by 4 h at 37 degrees C with slow internalization by 12-24 h. Biotinylated, conventional liposomes were specifically and effectively delivered to OVCAR-3 cells prelabeled with biotinylated B27.1 and SAv. The slow internalization and shedding properties of these antibodies are useful for multistep pretargeting methods. Thus, a modified targeting strategy, utilizing a bispecific antibody and liposomes, may be feasible for radioimmunoliposomal therapy of ovarian cancer.     

Human Germline Antibody Gene Segments Encode Polyspecific Antibodies

Structural flexibility in germline gene-encoded antibodies allows promiscuous binding to diverse antigens. The binding affinity and specificity for a particular epitope typically increase as antibody genes acquire somatic mutations in antigen-stimulated B cells. In this work, we investigated whether germline gene-encoded antibodies are optimal for polyspecificity by determining the basis for recognition of diverse antigens by antibodies encoded by three V_H gene segments. Panels of somatically mutated antibodies encoded by a common V_H gene, but each binding to a different antigen, were computationally redesigned to predict antibodies that could engage multiple antigens at once. The Rosetta multi-state design process predicted antibody sequences for the entire heavy chain variable region, including framework, CDR1, and CDR2 mutations. The predicted sequences matched the germline gene sequences to a remarkable degree, revealing by computational design the residues that are predicted to enable polyspecificity, i.e., binding of many unrelated antigens with a common sequence. The process thereby reverses antibody maturation in silico. In contrast, when designing antibodies to bind a single antigen, a sequence similar to that of the mature antibody sequence was returned, mimicking natural antibody maturation in silico. We demonstrated that the Rosetta computational design algorithm captures important aspects of antibody/antigen recognition. While the hypervariable region CDR3 often mediates much of the specificity of mature antibodies, we identified key positions in the V_H gene encoding CDR1, CDR2, and the immunoglobulin framework that are critical contributors for polyspecificity in germline antibodies. Computational design of antibodies capable of binding multiple antigens may allow the rational design of antibodies that retain polyspecificity for diverse epitope binding.