Generation of “LYmph Node Derived Antibody Libraries” (LYNDAL) for selecting fully human antibody fragments with therapeutic potential

The development of efficient strategies for generating fully human monoclonal antibodies with unique functional properties that are exploitable for tailored therapeutic interventions remains a major challenge in the antibody technology field. Here, we present a methodology for recovering such antibodies from antigen-encountered human B cell repertoires. As the source for variable antibody genes, we cloned immunoglobulin G (IgG)-derived B cell repertoires from lymph nodes of 20 individuals undergoing surgery for head and neck cancer. Sequence analysis of unselected “LYmph Node Derived Antibody Libraries” (LYNDAL) revealed a naturally occurring distribution pattern of rearranged antibody sequences, representing all known variable gene families and most functional germline sequences. To demonstrate the feasibility for selecting antibodies with therapeutic potential from these repertoires, seven LYNDAL from donors with high serum titers against herpes simplex virus (HSV) were panned on recombinant glycoprotein B of HSV-1. Screening for specific binders delivered 34 single-chain variable fragments (scFvs) with unique sequences. Sequence analysis revealed extensive somatic hypermutation of enriched clones as a result of affinity maturation. Binding of scFvs to common glycoprotein B variants from HSV-1 and HSV-2 strains was highly specific, and the majority of analyzed antibody fragments bound to the target antigen with nanomolar affinity. From eight scFvs with HSV-neutralizing capacity in vitro, the most potent antibody neutralized 50% HSV-2 at 4.5 nM as a dimeric (scFv)₂. We anticipate our approach to be useful for recovering fully human antibodies with therapeutic potential.     

By-passing immunization. Human antibodies from V-gene libraries displayed on phage.

We have mimicked features of immune selection to make human antibodies in bacteria. Diverse libraries of immunoglobulin heavy (VH) and light (V kappa and V lambda) chain variable (V) genes were prepared from peripheral blood lymphocytes (PBLs) of unimmunized donors by polymerase chain reaction (PCR) amplification. Genes encoding single chain Fv fragments were made by randomly combining heavy and light chain V-genes using PCR, and the combinatorial library (greater than 10(7) members) cloned for display on the surface of a phage. Rare phage with "antigen-binding" activities were selected by four rounds of growth and panning with "antigen" (turkey egg-white lysozyme (TEL) or bovine serum albumin) or "hapten" (2-phenyloxazol-5-one (phOx], and the encoding heavy and light chain genes were sequenced. The V-genes were human with some nearly identical to known germ-line V-genes, while others were more heavily mutated. Soluble antibody fragments were prepared and shown to bind specifically to antigen or hapten and with good affinities, Ka (TEL) = 10(7) M-1; Ka (phOx) = 2 x 10(6) M-1. Isolation of higher-affinity fragments may require the use of larger primary libraries or the construction of secondary libraries from the binders. Nevertheless, our results suggest that a single large phage display library can be used to isolate human antibodies against any antigen, by-passing both hybridoma technology and immunization.     

Generation of “LYmph Node Derived Antibody Libraries” (LYNDAL) for selecting fully human antibody fragments with therapeutic potential

The development of efficient strategies for generating fully human monoclonal antibodies with unique functional properties that are exploitable for tailored therapeutic interventions remains a major challenge in the antibody technology field. Here, we present a methodology for recovering such antibodies from antigen-encountered human B cell repertoires. As the source for variable antibody genes, we cloned immunoglobulin G (IgG)-derived B cell repertoires from lymph nodes of 20 individuals undergoing surgery for head and neck cancer. Sequence analysis of unselected “LYmph Node Derived Antibody Libraries” (LYNDAL) revealed a naturally occurring distribution pattern of rearranged antibody sequences, representing all known variable gene families and most functional germline sequences. To demonstrate the feasibility for selecting antibodies with therapeutic potential from these repertoires, seven LYNDAL from donors with high serum titers against herpes simplex virus (HSV) were panned on recombinant glycoprotein B of HSV-1. Screening for specific binders delivered 34 single-chain variable fragments (scFvs) with unique sequences. Sequence analysis revealed extensive somatic hypermutation of enriched clones as a result of affinity maturation. Binding of scFvs to common glycoprotein B variants from HSV-1 and HSV-2 strains was highly specific, and the majority of analyzed antibody fragments bound to the target antigen with nanomolar affinity. From eight scFvs with HSV-neutralizing capacity in vitro, the most potent antibody neutralized 50% HSV-2 at 4.5 nM as a dimeric (scFv)₂. We anticipate our approach to be useful for recovering fully human antibodies with therapeutic potential.     

A structure-based database of antibody variable domain diversity

The diversity of natural antibodies is limited by the genetic mechanisms that engender diversity and the functional requirements of antigen binding. Using an in vitro-evolved autonomous heavy chain variable domain (V(H)H-RIG), we have investigated the limits of structurally-tolerated diversity in the three complementarity-determining regions and a fourth loop within the third framework region. We determined the X-ray crystal structure of the V(H)H-RIG domain at 1.9A resolution and used it to guide the design of phage-displayed libraries encompassing the four loops. The libraries were subjected to selections for structural stability, and a database of structurally-tolerated sequences was compiled from the sequences of approximately 1000 unique clones. The results reveal that all four loops accommodate significantly greater diversity than is observed in nature. Thus, it appears that most sequence biases in the natural immune repertoire arise from factors other than structural constraints and, consequently, it should be possible to enhance the functions of antibodies significantly through in vitro evolution.     

Guided selection of a pan carcinoma specific antibody reveals similar binding characteristics yet structural divergence between the original murine antibody and its human equivalent

Antibody engineering provides an excellent tool for the generation of human immunotherapeutics for the targeted treatment of solid tumours. We have engineered and selected a completely human antibody to epithelial glycoprotein-2 (EGP-2), a transmembrane glycoprotein present on virtually all human simple epithelia and abundantly expressed on a variety of human carcinomas. We chose to use the procedure of "guided selection" to rebuild a high-affinity murine antibody into a human antibody, using two consecutive rounds of variable domain shuffling and phage library selection. As a starting antibody, the murine antibody MOC-31 was used. After the first round of guided selection, where the V(H) of MOC-31 was combined in Fab format with a human V(L)C(L) library, a small panel of human light chains was identified, originating from a segment of the VkappaIII family, whereas the MOC-31 V(L) is more homologous to the VkappaII family. Nevertheless, one of the chimaeric Fabs, C3, displayed an off-rate similar to MOC-31 scFv. Combining the V(L) of C3 with a human V(H) library, while retaining the V(H) CDR3 of MOC-31, clones were selected using human V(H) genes originating from the rarely used V(H)7 family. The best clone, 9E, shows over 13 amino acid mutations from the germline sequence, has an off-rate comparable to the original antibody and specifically binds to the "MOC-31"-epitope on EGP-2 in specificity and competition ELISA, FACS analysis and immunohistochemistry. In both V(L) and V(H) of antibody 9E, three germline mutations were found creating the MOC-31 homologue residue. Structural modelling of both murine and human antibodies reveals that one of the germline mutations, 53Y in V(H) CDR2, is likely to be involved in antigen binding. We conclude that, although they may bind the same epitope and have similar binding affinity to the antigen as the original murine antibody, human antibodies derived by guided selection unlike CDR-grafted antibodies, may retain only some of the original key elements of the binding site chemistry. The selected human anti-EGP-2 antibody will be a suitable reagent for tumour targeting.     

A recombinant, fully human monoclonal antibody with antitumor activity constructed from phage-displayed antibody fragments

A single-chain Fv antibody fragment specific for the tumor-associated Ep-CAM molecule was isolated from a semisynthetic phage display library and converted into an intact, fully human IgG1 monoclonal antibody (huMab). The purified huMab had an affinity of 5 nM and effectively mediated tumor cell killing in in vitro and in vivo assays. These experiments show that nonimmunized phage antibody display libraries can be used to obtain high-affinity, functional, and clinically applicable huMabs directed against a tumor-associated antigen.     

High-affinity human antibodies from phage-displayed synthetic Fab libraries with a single framework scaffold

Phage-displayed synthetic antibody libraries were built on a single human framework by introducing synthetic diversity at solvent-exposed positions within the heavy chain complementarity-determining regions (CDRs). The design strategy of mimicking natural diversity using tailored codons had been validated previously with scFv libraries, which produced antibodies that bound to antigen, murine vascular endothelial growth factor (mVEGF), with affinities in the 100nM range. To improve library performance, we constructed monovalent and bivalent antigen-binding fragment (Fab) libraries, and explored different CDR-H3 diversities by varying the amino acid composition and CDR length. A Fab with sub-nanomolar affinity for mVEGF was obtained from a library with CDR-H3 diversity designed to contain all 20 naturally occurring amino acids. We then expanded the library by increasing the variability of CDR-H3 length and using tailored codons that mimicked the amino acid composition of natural CDR-H3 sequences. The library was tested against a panel of 13 protein antigens and high-affinity Fabs were obtained for most antigens. Furthermore, the heavy chain of an anti-mVEGF clone was recombined with a library of light chain CDRs, and the affinity was improved from low nanomolar to low picomolar. The results demonstrated that high-affinity human antibodies can be generated from libraries with completely synthetic CDRs displayed on a single scaffold.     

Identification of chimpanzee Fab fragments by repertoire cloning and production of a full-length humanized immunoglobulin G1 antibody that is highly efficient for neutralization of dengue type 4 virus

A safe and effective dengue vaccine is still not available. Passive immunization with monoclonal antibodies from humans or nonhuman primates represents an attractive alternative for the prevention of dengue virus infection. Fab monoclonal antibodies to dengue type 4 virus (DENV-4) were recovered by repertoire cloning of bone marrow mRNAs from an immune chimpanzee and analyzed for antigen binding specificity, V(H) and V(L) sequences, and neutralizing activity against DENV-4 in vitro. Fabs 5A7, 3C1, 3E4, and 7G4 were isolated from a library constructed from a chimpanzee following intrahepatic transfection with infectious DENV-4 RNA. Fabs 5H2 and 5D9, which had nearly identical V(H) sequences but varied in their V(L) sequences, were recovered from a library constructed from the same chimpanzee after superinfection with a mixture of DENV-1, DENV-2, and DENV-3. In radioimmunoprecipitation, Fab 5A7 precipitated only DENV-4 prM, and Fabs 3E4, 7G4, 5D9, and 5H2 precipitated DENV-4 E but little or no prM. Fab 3E4 and Fab 7G4 competed with each other for binding to DENV-4 in an enzyme-linked immunosorbent assay, as did Fab 3C1 and Fab 5A7. Fab 5H2 recognized an epitope on DENV-4 that was separate from the epitope(s) recognized by other Fabs. Both Fab 5H2 and Fab 5D9 neutralized DENV-4 efficiently with a titer of 0.24 to 0.58 micro g/ml by plaque reduction neutralization test (PRNT), whereas DENV-4-neutralizing activity of other Fabs was low or not detected. Fab 5H2 was converted to full-length immunoglobulin G1 (IgG1) by combining it with human sequences. The humanized chimpanzee antibody IgG1 5H2 produced in CHO cells neutralized DENV-4 strains from different geographical origins at a similar 50% plaque reduction (PRNT(50)) titer of 0.03 to 0.05 micro g/ml. The DENV-4 binding affinities were 0.42 nM for Fab 5H2 and 0.24 nM for full-length IgG1 5H2. Monoclonal antibody IgG1 5H2 may prove valuable for passive immunoprophylaxis against dengue virus in humans.     

Direct cDNA cloning of the rearranged immunoglobulin variable region

A major problem in the study of multigene families is the effort required to clone and sequence these genes. We describe a method to rapidly clone and sequence immunoglobulin variable region gene sequences without constructing cDNA libraries. Because immunoglobulin variable-region genes are flanked by conserved sequences, we have been able to apply the polymerase chain reaction (PCR) to clone and sequence both the light- and heavy-chain rearranged immunoglobulin genes from small numbers of hybridoma cells. This method will greatly facilitate the construction of chimeric mouse/human monoclonal antibodies for immunoglobulin structural studies as well as for therapeutic use.     

Antibody humanization by structure-based computational protein design

Antibodies derived from non-human sources must be modified for therapeutic use so as to mitigate undesirable immune responses. While complementarity-determining region (CDR) grafting-based humanization techniques have been successfully applied in many cases, it remains challenging to maintain the desired stability and antigen binding affinity upon grafting. We developed an alternative humanization approach called CoDAH (“Computationally-Driven Antibody Humanization”) in which computational protein design methods directly select sets of amino acids to incorporate from human germline sequences to increase humanness while maintaining structural stability. Retrospective studies show that CoDAH is able to identify variants deemed beneficial according to both humanness and structural stability criteria, even for targets lacking crystal structures. Prospective application to TZ47, a murine anti-human B7H6 antibody, demonstrates the approach. Four diverse humanized variants were designed, and all possible unique VH/VL combinations were produced as full-length IgG1 antibodies. Soluble and cell surface expressed antigen binding assays showed that 75% (6 of 8) of the computationally designed VH/VL variants were successfully expressed and competed with the murine TZ47 for binding to B7H6 antigen. Furthermore, 4 of the 6 bound with an estimated K_D within an order of magnitude of the original TZ47 antibody. In contrast, a traditional CDR-grafted variant could not be expressed. These results suggest that the computational protein design approach described here can be used to efficiently generate functional humanized antibodies and provide humanized templates for further affinity maturation.     

Transferring the Characteristics of Naturally Occurring and Biased Antibody Repertoires to Human Antibody Libraries by Trapping CDRH3 Sequences

Antibody repertoires are characterized by diversity as they vary not only amongst individuals and post antigen exposure but also differ significantly between vertebrate species. Such plasticity can be exploited to generate human antibody libraries featuring hallmarks of these diverse repertoires. In this study, the focus was to capture CDRH3 sequences, as this region generally accounts for most of the interaction energy with antigen. Sequences from human as well as non-human sources were successfully integrated into human antibody libraries. Next generation sequencing of these libraries proved that the CDRH3 lengths and amino acid composition corresponded to the species of origin. Specific CDRH3 sequences, biased towards the recognition of a model antigen either by immunizing mice or by selecting with phage display, were then integrated into another set of libraries. From these antigen biased libraries, highly potent antibodies were more frequently isolated, indicating that the characteristics of an immune repertoire is transferrable via CDRH3 sequences into a human antibody library. Taken together, these data demonstrate that the properties of naturally or experimentally biased repertoires can be effectively harnessed for the generation of targeted human antibody libraries, substantially increasing the probability of isolating antibodies suitable for therapeutic and diagnostic applications.     

Production and Characterization of Monoclonal Antibodies Specific for a Conserved Epitope within Hepatitis C Virus Hypervariable Region 1

Frequent mutations in hypervariable region 1 (HVR1) of the main envelope protein of hepatitis C virus (HCV) is a major mechanism of persistence by escaping the host immune recognition. HVR1 contains an epitope eliciting neutralizing antibodies. This study was aimed to prepare broadly cross-reacting, high-affinity, monoclonal antibodies (MAb) to the HVR1 C terminus of HCV with potential therapeutic neutralizing capacity. A conserved amino residue group of glycine (G) at position 23 and glutamic acid (Q) at position 26 in HVR1 was confirmed as a key epitope against which two MAbs were selected and characterized. MAbs 2P24 and 15H4 were immunoglobulin G1 kappa chain [IgG1(kappa)], cross-reacted with 32 and 30 of 39 random C-terminal HVR1 peptides, respectively, and did not react with other HCV peptides. The V(H) of 2P24 and 15H4 heavy chains originated from Igh germ line v gene family 1 and 8, respectively. In contrast, the V(L) kappa sequences were highly homologous. The affinity (K(d)) of 2P24 and 15H4 (10(-9) or 10(-8) M with two immunizing peptides and 10(-8) M with two nonimmunizing HVR1 peptides) paralleled the reactivity obtained with peptide enzyme immunoassay. MAbs 2P24 and 15H4 captured 25 of 31 (81%) HCV in unselected patients' plasmas. These antibodies also blocked HCV binding to Molt-4 cells in a dose-dependent fashion. The data presented suggest that broadly cross-reactive MAbs to a conserved epitope within HCV HVR1 can be produced. Clinical application for passive immunization in HCV-related chronic liver disease and after liver transplantation is considered.     

Completion of the analysis of the primary structure of the variable domain of a homogeneous rabbit antibody to type III pneumococcal polysaccharide

The amino acid sequence between residues 70 and 116 of the V (variable) region of the H (heavy) chain derived from rabbit antibody BS-5, specific for type III pneumococcal polysaccharide, was determined. The sequence of this section of the H chain which includes the hypervariable residues 94 to about 112 was unique, although minor variant sequences present in the H chain preparation would not have been detected by the techniques used in this work. Taken together with the known sequences of the N-terminal 69 residues of H chain BS-5 (Jaton & Braun, 1972) and of the V region of the light chain (Jaton, 1974b), the data establish the complete sequence of the V domain of a rabbit immunoglobulin G. The V region of H chain BS-5 is compared with the basic sequences of the three human V region subgroups known to date, with one mouse H chain, and with guinea-pig pooled H chains. Even though chains from guinea pig and mouse clearly belong to the subgroup III of variability (V(HIII)), rabbit H chain BS-5 (allotypic variant a(1)) appears more closely related to the subgroup V(HII) than to the subgroups V(HIII) or V(HI). The homology between V(L) and V(H) regions of antibody BS-5 (28%) is not greater than that observed between the V(H) region of antibody BS-5 and the V(L) regions of different rabbit antibodies.     

Molecular cloning and chimerisation of an inhibitory anti-cathepsin B antibody and its expression in Chinese hamster ovary cells

The lysosomal cysteine protease cathepsin B is one of several proteases that have been linked to tumour progression. Its increased expression and secretion in tumour cells may facilitate the degradation of extracellular matrix proteins, leading to tumour cell invasion and metastasis. Specific inhibitory monoclonal antibodies are a possible alternative to synthetic inhibitors as a therapeutic tool for cancer treatment. An inhibitory monoclonal antibody, which binds to an epitope near the active site of cathepsin B and inhibits its proteolytic activity, was prepared and its effect on invasion of ras-transformed MCF-10A neoT cells was tested in vitro. Here we present the nucleotide sequences of the heavy and light chains of the inhibitory antibody and compare them to the murine immunoglobulin germline sequences for possible somatic hypermutations. Since no harmful mutations were found, a mouse/human chimeric antibody was constructed by fusing murine V(H) and V(L) variable regions of the inhibitory antibody with human gamma 1 and K constant regions, respectively. Chinese hamster ovary K1 cells were co-transfected with expression vectors pcD-NA3L and pcDNA3H and the reactivity of the isolated chimeric antibody was tested by ELISA and Western blotting. We could demonstrate an inhibitory effect of the chimeric antibody.     

Memory in the B-cell compartment: antibody affinity maturation

In the humoral arm of the immune system, the memory response is not only more quickly elicited and of greater magnitude than the primary response, but it is also different in quality. In the recall response to antigen, the antibodies produced are of higher affinity and of different isotype (typically immunoglobulin G rather than immunoglobulin M). This maturation rests on the antigen dependence of B-cell maturation and is effected by programmed genetic modifications of the immunoglobulin gene loci. Here we consider how the B-cell response to antigen depends on the affinity of the antigen receptor interaction. We also compare and draw parallels between the two processes, which underpin the generation of secondary-response antibodies: V gene somatic hypermutation and immunoglobulin heavy-chain class switching.     

Hepatitis C virus (HCV)-induced immunoglobulin hypermutation reduces the affinity and neutralizing activities of antibodies against HCV envelope protein

Hepatitis C virus (HCV) often causes persistent infection despite the presence of neutralizing antibodies against the virus in the sera of hepatitis C patients. HCV infects both hepatocytes and B cells through the binding of its envelope glycoprotein E2 to CD81, the putative viral receptor. Previously, we have shown that E2-CD81 interaction induces hypermutation of heavy-chain immunoglobulin (V(H)) in B cells. We hypothesize that if HCV infects antibody-producing B cells, the resultant hypermutation of V(H) may lower the affinity and specificity of the HCV-specific antibodies, enabling HCV to escape from immune surveillance. To test this hypothesis, we infected human hybridoma clones producing either neutralizing or non-neutralizing anti-E2 or anti-E1 antibodies with a lymphotropic HCV (SB strain). All of the hybridoma clones, except for a neutralizing antibody-producing hybridoma, could be infected with HCV and support virus replication for at least 8 weeks after infection. The V(H) sequences in the infected hybridomas had a significantly higher mutation frequency than those in the uninfected hybridomas, with mutations concentrating in complementarity-determining region 3. These mutations lowered the antibody affinity against the targeting protein and also lowered the virus-neutralizing activity of anti-E2 antibodies. Furthermore, antibody-mediated complement-dependent cytotoxicity with the antibodies secreted from the HCV-infected hybridomas was impaired. These results suggest that HCV infection could cause some anti-HCV-antibody-producing hybridoma B cells to make less-protective antibodies.     

Human antibodies from immunized donors are protective against anthrax toxin in vivo

A panel of Fabs that neutralize anthrax toxin in vitro was selected from libraries generated from human donors vaccinated against anthrax. At least two of these antibodies protect rats from anthrax intoxication in vivo. Fabs 83K7C and 63L1D bind with subnanomolar affinity to protective antigen (PA) 63, and Fab 63L1D neutralizes toxin substoichiometrically, inhibits lethal factor (LF) interaction with PA63 and binds to a conformational epitope formed by PA63.     

Selection and characterisation of binders based on homodimerisation of immunoglobulin V(H) domains

The antigen-binding surface of antibodies is formed by the heterodimerisation of the two variable domains of the light (V(L)) and heavy (V(H)) chains. We have previously described the spontaneous formation of V(H) dimers (VHD) in both bacteria and mammalian cells. The self-association of a single domain produces a homo-VHD, in which the two identical V(H) domains generate a unique symmetric surface for antigen binding that is never found in the normal V(L)/V(H) antibody binding site. We developed a phagemid vector for the construction of phage display libraries in which a cysteine residue, introduced at the C-terminus of the only V(H) cloned, allowed display of homo-VHDs. Panning of the library on different proteins yielded antigen specific binders against lysozyme, glutathione S-transferase and streptavidin. A lysozyme specific homo-VHD was further characterised with an apparent affinity determined to be 216+/-6.6 nM. Importantly, the results showed that its binding activity was fully dependent on the dimerisation of both identical V(H) domains.     

Multi‐constraint computational design suggests that native sequences of germline antibody H3 loops are nearly optimal for conformational flexibility

The limited size of the germline antibody repertoire has to recognize a far larger number of potential antigens. The ability of a single antibody to bind multiple ligands due to conformational flexibility in the antigen‐binding site can significantly enlarge the repertoire. Among the six complementarity determining regions (CDRs) that generally comprise the binding site, the CDR H3 loop is particularly variable. Computational protein design studies showed that predicted low energy sequences compatible with a given backbone structure often have considerable similarity to the corresponding native sequences of naturally occurring proteins, indicating that native protein sequences are close to optimal for their structures. Here, we take a step forward to determine whether conformational flexibility, believed to play a key functional role in germline antibodies, is also central in shaping their native sequence. In particular, we use a multi‐constraint computational design strategy, along with the Rosetta scoring function, to propose that the native sequences of CDR H3 loops from germline antibodies are nearly optimal for conformational flexibility. Moreover, we find that antibody maturation may lead to sequences with a higher degree of optimization for a single conformation, while disfavoring sequences that are intrinsically flexible. In addition, this computational strategy allows us to predict mutations in the CDR H3 loop to stabilize the antigen‐bound conformation, a computational mimic of affinity maturation, that may increase antigen binding affinity by preorganizing the antigen binding loop. In vivo affinity maturation data are consistent with our predictions. The method described here can be useful to design antibodies with higher selectivity and affinity by reducing conformational diversity. Proteins 2009. © 2008 Wiley‐Liss, Inc.