Antibodies in haystacks: how selection strategy influences the outcome of selection from molecular diversity libraries

Combining random mutagenesis, structure-guided design and next-generation sequencing to mitigate polyreactivity of an anti-IL-21R antibody

Despite substantial technological advances in antibody library and display platform development, the number of approved biotherapeutics from displayed libraries remains limited. In vivo, 20–50% of peripheral B cells undergo a process of receptor editing, which modifies the variable and junctional regions of light chains to delete auto-reactive clones. However, in vitro antibody evolution relies primarily on interaction with antigen, with no in-built checkpoints to ensure that the selected antibodies have not acquired additional specificities or biophysical liabilities during the optimization process. We had previously observed an enrichment of positive charge in the complementarity-determining regions of an anti-IL-21 R antibody during affinity optimization, which correlated with more potent IL-21 neutralization, but poor in vivo pharmacokinetics (PK). There is an emerging body of data that has correlated antibody nonspecificity with poor PK in vivo, and established a series of screening assays that are predictive of this behavior. In this study we revisit the challenge of developing an anti-IL-21 R antibody that can effectively compete with IL-21 for its highly negatively charged paratope while maintaining favorable biophysical properties. In vitro deselection methods that included an excess of negatively charged membrane preparations, or deoxyribonucleic acid, during phage selection of optimization libraries were unsuccessful in avoiding enrichment of highly charged, nonspecific antibody variants. However, a combination of structure-guided rational library design, next-generation sequencing of library outputs and application of linear regression models resulted in the identification of an antibody that maintained high affinity for IL-21 R and exhibited a desirable stability and biophysical profile.

Rationalizing Random Walks: Replicating Protective Antibody Trajectories

'Reverse vaccinology 2.0' aims to rationally reproduce template antibody responses, such as broadly neutralizing antibodies against human immunodeficiency virus-1. While observations of antibody convergence across individuals support the assumption that responses may be replicated, the diversity of humoral immunity and the process of antibody selection are rooted in stochasticity. Drawing from experience with in vitro antibody engineering by directed evolution, we consider how antibody selection may be driven, as in germline-targeting vaccine approaches to elicit broadly neutralizing antibodies and illustrate the potential consequences of over-defining a template antibody response. We posit that the prospective definition of template antibody responses and the odds of replicating them must be considered within the randomness of humoral immunity.

A comparison of three strategies for biopanning of phage-scFv library against diphtheria toxin

The biopanning process is a critical step in phage display for isolating peptides or proteins with specific binding properties. Conventional panning methods are sometimes not so effective and may result in nonspecific or low-yield positive results. In this study, three different strategies including soluble antibody-capturing, pH-stepwise elution, and conventional panning were used for enrichment of specific clones against diphtheria toxoid. The reactivity of the selected clones was evaluated using an indirect enzyme-linked immunosorbent assay. The positive clones were screened using Vero cell viability assay. The neutralizing clones were expressed in HB2151 strain of Escherichia coli and soluble single-chain fragment variable (scFv) fragments were purified by nickel-nitrilotriacetic acid affinity chromatography. Finally, the ability of scFv fragments for neutralizing diphtheria toxin (DT) were evaluated again using Vero cell viability assay. After four rounds of panning, the soluble antibody-capturing method yielded 15 positive phage-scFv clones against diphtheria toxoid. Conventional panning and pH-stepwise elution model resulted from nine and five positive phage-scFv clones, respectively. Among all positive clones, three clones were able to neutralize DT in Vero cell viability assay. Two of these clones belonged to a soluble antibody-capturing method and one of them came from conventional panning. Three neutralizing clones were used for soluble expression and purification of scFvs fragments. It was found that these soluble scFv fragments possessed neutralizing activity ranging from 0.15 to 0.6 µg against two-fold cytotoxic dose 99% of DT. In conclusion, the results of our study indicate that soluble antibody-capturing method is an efficient method for isolation of specific scFv fragments.

Many Routes to an Antibody Heavy-Chain CDR3: Necessary, Yet Insufficient, for Specific Binding

Because of its great potential for diversity, the immunoglobulin heavy-chain complementarity-determining region 3 (HCDR3) is taken as an antibody molecule’s most important component in conferring binding activity and specificity. For this reason, HCDR3s have been used as unique identifiers to investigate adaptive immune responses in vivo and to characterize in vitro selection outputs where display systems were employed. Here, we show that many different HCDR3s can be identified within a target-specific antibody population after in vitro selection. For each identified HCDR3, a number of different antibodies bearing differences elsewhere can be found. In such selected populations, all antibodies with the same HCDR3 recognize the target, albeit at different affinities. In contrast, within unselected populations, the majority of antibodies with the same HCDR3 sequence do not bind the target. In one HCDR3 examined in depth, all target-specific antibodies were derived from the same VDJ rearrangement, while non-binding antibodies with the same HCDR3 were derived from many different V and D gene rearrangements. Careful examination of previously published in vivo datasets reveals that HCDR3s shared between, and within, different individuals can also originate from rearrangements of different V and D genes, with up to 26 different rearrangements yielding the same identical HCDR3 sequence. On the basis of these observations, we conclude that the same HCDR3 can be generated by many different rearrangements, but that specific target binding is an outcome of unique rearrangements and VL pairing: the HCDR3 is necessary, albeit insufficient, for specific antibody binding.

Screening Phage-Display Antibody Libraries Using Protein Arrays

Phage-display technology constitutes a powerful tool for the generation of specific antibodies against a predefined antigen. The main advantages of phage-display technology in comparison to conventional hybridoma-based techniques are: (1) rapid generation time and (2) antibody selection against an unlimited number of molecules (biological or not). However, the main bottleneck with phage-display technology is the validation strategies employed to confirm the greatest number of antibody fragments. The development of new high-throughput (HT) techniques has helped overcome this great limitation. Here, we describe a new method based on an array technology that allows the deposition of hundreds to thousands of phages by micro-contact on a unique nitrocellulose surface. This setup comes in combination with bioinformatic approaches that enables simultaneous affinity screening in a HT format of antibody-displaying phages.

Identification of novel proteins binding the AU-rich element of α-prothymosin mRNA through the selection of open reading frames (RIDome)

We describe here a platform for high-throughput protein expression and interaction analysis aimed at identifying the RNA-interacting domainome. This approach combines the selection of a phage library displaying "filtered" open reading frames with next-generation DNA sequencing. The method was validated using an RNA bait corresponding to the AU-rich element of α-prothymosin, an RNA motif that promotes mRNA stability and translation through its interaction with the RNA-binding protein ELAVL1. With this strategy, we not only confirmed known RNA-binding proteins that specifically interact with the target RNA (such as ELAVL1/HuR and RBM38) but also identified proteins not previously known to be ARE-binding (R3HDM2 and RALY). We propose this technology as a novel approach for studying the RNA-binding proteome.

Rescue and In Situ Selection and Evaluation (RISE): A Method for High-Throughput Panning of Phage Display Libraries

Phage display has proven to be an invaluable instrument in the search for proteins and peptides with optimized or novel functions. The amplification and selection of phage libraries typically involve several operations and handling large bacterial cultures during each round. Purification of the assembled phage particles after rescue adds to the labor and time demand. The authors therefore devised a method, termed rescue and in situ selection and evaluation (RISE), which combines all steps from rescue to binding in a single microwell. To test this concept, wells were precoated with different antibodies, which allowed newly formed phage particles to be captured directly in situ during overnight rescue. Following 6 washing steps, the retained phages could be easily detected in an enzyme-linked immunosorbent assay (ELISA), thus eliminating the need for purification or concentration of the viral particles. As a consequence, RISE enables a rapid characterization of phage-displayed proteins. In addition, this method allowed for the selective enrichment of phages displaying a hemagglutinin (HA) epitope tag, spiked in a 104-fold excess of wild-type background. Because the combination of phage rescue, selection, or evaluation in a single microwell is amenable to automation, RISE may boost the high-throughput screening of smaller sized phage display libraries.

Artificial Avidin-Based Receptors for a Panel of Small Molecules

Proteins with high specificity, affinity, and stability are needed for biomolecular recognition in a plethora of applications. Antibodies are powerful affinity tools, but they may also suffer from limitations such as low stability and high production costs. Avidin and streptavidin provide a promising scaffold for protein engineering, and due to their ultratight binding to D-biotin they are widely used in various biotechnological and biomedical applications. In this study, we demonstrate that the avidin scaffold is suitable for use as a novel receptor for several biologically active small molecules: Artificial, chicken avidin-based proteins, antidins, were generated using a directed evolution method for progesterone, hydrocortisone, testosterone, cholic acid, ketoprofen, and folic acid, all with micromolar to nanomolar affinity and significantly reduced biotin-binding affinity. We also describe the crystal structure of an antidin, sbAvd-2(I117Y), a steroid-binding avidin, which proves that the avidin scaffold can tolerate significant modifications without losing its characteristic tetrameric beta-barrel structure, helping us to further design avidin-based small molecule receptors.

Large scale production of phage antibody libraries using a bioreactor

One of the limitations of the use of phage antibody libraries in high throughput selections is the production of sufficient phage antibody library at the appropriate quality. Here, we successfully adapt a bioreactor-based protocol for the production of phage peptide libraries to the production of phage antibody libraries. The titers obtained in the stirred-tank bioreactor are 4 to 5 times higher than in a standard shake flask procedure, and the quality of the phage antibody library produced is indistinguishable to that produced using standard procedures as assessed by Western blotting and functional selections. Availability of this protocol will facilitate the use of phage antibody libraries in high-throughput scale selections.

Diversity of the antibody response to tetanus toxoid: comparison of hybridoma library to phage display library

Monoclonal antibodies are important tools in research and since the 1990s have been an important therapeutic class targeting a wide variety of diseases. Earlier methods of mAb production relied exclusively on the lengthy process of making hybridomas. The advent of phage display technology introduced an alternative approach for mAb production. A potential concern with this approach is its complete dependence on an in vitro selection process, which may result in selection of V(H)-V(L) pairs normally eliminated during the in vivo selection process. The diversity of V(H)-V(L) pairs selected from phage display libraries relative to an endogenous response is unknown. To address these questions, we constructed a panel of hybridomas and a phage display library using the spleen of a single tetanus toxoid-immunized mouse and compared the diversity of the immune response generated using each technique. Surprisingly, the tetanus toxoid-specific antibodies produced by the hybridoma library exhibited a higher degree of V(H)-V(L) genetic diversity than their phage display-derived counterparts. Furthermore, the overlap among the V-genes from each library was very limited. Consistent with the notion that accumulation of many small DNA changes lead to increased antigen specificity and affinity, the phage clones displayed substantial micro-heterogeneity. Contrary to previous reports, we found that antigen specificity against tetanus toxoid is encoded by both V(κ) and V(H) genes. Finally, the phage-derived tetanus-specific clones had a lower binding affinity than the hybridomas, a phenomenon thought to be the result of random pairing of the V-genes.

Using phage display selected antibodies to dissect microbiomes for complete de novo genome sequencing of low abundance microbes

BACKGROUND

Single cell genomics has revolutionized microbial sequencing, but complete coverage of genomes in complex microbiomes is imperfect due to enormous variation in organismal abundance and amplification bias. Empirical methods that complement rapidly improving bioinformatic tools will improve characterization of microbiomes and facilitate better genome coverage for low abundance microbes.

METHODS

We describe a new approach to sequencing individual species from microbiomes that combines antibody phage display against intact bacteria with fluorescence activated cell sorting (FACS). Single chain (scFv) antibodies are selected using phage display against a bacteria or microbial community, resulting in species-specific antibodies that can be used in FACS for relative quantification of an organism in a community, as well as enrichment or depletion prior to genome sequencing.

RESULTS

We selected antibodies against Lactobacillus acidophilus and demonstrate a FACS-based approach for identification and enrichment of the organism from both laboratory-cultured and commercially derived bacterial mixtures. The ability to selectively enrich for L. acidophilus when it is present at a very low abundance (<0.2%) leads to complete (>99.8%) de novo genome coverage whereas the standard single-cell sequencing approach is incomplete (<68%). We show that specific antibodies can be selected against L. acidophilus when the monoculture is used as antigen as well as when a community of 10 closely related species is used demonstrating that in principal antibodies can be generated against individual organisms within microbial communities.

CONCLUSIONS

The approach presented here demonstrates that phage-selected antibodies against bacteria enable identification, enrichment of rare species, and depletion of abundant organisms making it tractable to virtually any microbe or microbial community. Combining antibody specificity with FACS provides a new approach for characterizing and manipulating microbial communities prior to genome sequencing.

Comprehensive interrogation of a minimalist synthetic CDR-H3 library and its ability to generate antibodies with therapeutic potential

We have generated large libraries of single-chain Fv antibody fragments (>10(10) transformants) containing unbiased amino acid diversity that is restricted to the central combining site of the stable, well-expressed DP47 and DPK22 germline V-genes. Library WySH2A was constructed to examine the potential for synthetic complementarity-determining region (CDR)-H3 diversity to act as the lone source of binding specificity. Library WySH2B was constructed to assess the necessity for diversification in both the H3 and L3. Both libraries provided diverse, specific antibodies, yielding a total of 243 unique hits against 7 different targets, but WySH2B produced fewer hits than WySH2A when selected in parallel. WySH2A also consistently produced hits of similar quality to WySH2B, demonstrating that the diversification of the CDR-L3 reduces library fitness. Despite the absence of deliberate bias in the library design, CDR length was strongly associated with the number of hits produced, leading to a functional loop length distribution profile that mimics the biases observed in the natural repertoire. A similar trend was also observed for the CDR-L3. After target selections, several key amino acids were enriched in the CDR-H3 (e.g., small and aromatic residues) while others were reduced (e.g., strongly charged residues) in a manner that was specific to position, preferentially occurred in CDR-H3 stem positions, and tended towards residues associated with loop stabilization. As proof of principle for the WySH2 libraries to produce viable lead candidate antibodies, 114 unique hits were produced against Delta-like ligand 4 (DLL4). Leads exhibited nanomolar binding affinities, highly specific staining of DLL4+ cells, and biochemical neutralization of DLL4-NOTCH1 interaction.

Synthetic antibodies: concepts, potential and practical considerations

The last 100 years of enquiry into the fundamental basis of humoral immunity has resulted in the identification of antibodies as key molecular sentinels responsible for the in vivo surveillance, neutralization and clearance of foreign substances. Intense efforts aimed at understanding and exploiting their exquisite molecular specificity have positioned antibodies as a cornerstone supporting basic research, diagnostics and therapeutic applications [1]. More recently, efforts have aimed to circumvent the limitations of developing antibodies in animals by developing wholly in vitro techniques for designing antibodies of tailored specificity. This has been realized with the advent of synthetic antibody libraries that possess diversity outside the scope of natural immune repertoires and are thus capable of yielding specificities not otherwise attainable. This review examines the convergence of technologies that have contributed to the development of combinatorial phage-displayed antibody libraries. It further explores the practical concepts that underlie phage display, antibody diversity and the methods used in the generation of and selection from phage-displayed synthetic antibody libraries, highlighting specific applications in which design approaches gave rise to specificities that could not easily be obtained with libraries based upon natural immune repertories.

Recombinant antibodies and in vitro selection technologies

Over the past decade, the accumulation of detailed knowledge of antibody structure and function has enabled antibody phage display to emerge as a powerful in vitro alternative to hybridoma methods for creating antibodies. Many antibodies produced using phage display technology have unique properties that are not obtainable using traditional hybridoma technologies. In phage display, selections are performed under controlled, in vitro conditions that are tailored to suit demands of the antigen and the sequence encoding the antibody is immediately available. These features obviate many of the limitations of hybridoma methodology, and because the entire process relies on scalable molecular biology techniques, phage display is also suitable for high-throughput applications. Thus, antibody phage display technology is well suited for genome-scale biotechnology and therapeutic applications. This review describes the antibody phage display technology and highlights examples of antibodies with unique properties that cannot easily be obtained by other technologies.

Development of phage-based single chain Fv antibody reagents for detection of Yersinia pestis

BACKGROUND

Most Yersinia pestis strains are known to express a capsule-like antigen, fraction 1 (F1)(.) F1 is encoded by the caf1 gene located on the large 100-kb pFra plasmid, which is found in Y. pestis but not in closely related species such as Yersinia enterocolytica and Yersinia pseudotuberculosis. In order to find antibodies specifically binding to Y. pestis we screened a large single chain Fv antibody fragment (scFv) phage display library using purified F1 antigen as a selection target. Different forms of the selected antibodies were used to establish assays for recombinant F1 antigen and Y. pestis detection.

METHODS

Phage antibody panning was performed against F1 in an automated fashion using the Kingfisher magnetic bead system. Selected scFvs were screened for F1-binding specificity by one-step alkaline phosphatase enzyme linked immunosorbant assay (ELISA), and assayed for binding to recombinant antigen and/or Y. pestis by flow cytometry and whole-cell ELISA.

RESULTS

Seven of the eight selected scFvs were shown to specifically bind both recombinant F1 and a panel of F1-positive Yersinia cells. The majority of the soluble scFvs were found to be difficult to purify, unstable and prone to cross-reactivity with F1-negative Yersinia strains, whereas phage displayed scFvs were found to be easy to purify/label and remarkably stable. Furthermore direct fluorescent labeling of phage displaying scFv allowed for an easy one-step flow cytometry assay. Slight cross-reactivity was observed when fixed cells were used in ELISA.

CONCLUSIONS

Our high throughput methods of selection and screening allowed for time and cost effective discovery of seven scFvs specifically binding Y. pestis F1 antigen. We describe implementation of different methods for phage-based immunoassay. Based on the success of these methods and the proven stability of phage, we indicate that the use of phage-displayed, rather than phage-free proteins, might generally overcome the shortcomings of scFv antibodies.

Beyond natural antibodies: the power of in vitro display technologies

In vitro display technologies, best exemplified by phage and yeast display, were first described for the selection of antibodies some twenty years ago. Since that time a large number of antibodies, some with remarkable properties, have been selected and improved upon using these methods. The first antibodies derived using in vitro display methods are now in the clinic, with many more waiting in the wings. Here we discuss the scope of the technology, some of the powerful antibodies selected, and the future potential in a post-genomic world. Unique advantages offered by in vitro display technologies include the ability to carefully define selection conditions, allowing the derivation of antibodies recognizing predefined epitopes or conformations, the further improvement of selected antibodies, the potential for high throughput applications and the immediate availability of genes encoding the selected antibody. We anticipate that the high throughput potential of these technologies will soon lead to their use to select antibodies against all human proteins.

Rapid interactome profiling by massive sequencing

We have developed a high-throughput protein expression and interaction analysis platform that combines cDNA phage display library selection and massive gene sequencing using the 454 platform. A phage display library of open reading frame (ORF) fragments was created from mRNA derived from different tissues. This was used to study the interaction network of the enzyme transglutaminase 2 (TG2), a multifunctional enzyme involved in the regulation of cell growth, differentiation and apoptosis, associated with many different pathologies. After two rounds of panning with TG2 we assayed the frequency of ORFs within the selected phage population using 454 sequencing. Ranking and analysis of more than 120 000 sequences allowed us to identify several potential interactors, which were subsequently confirmed in functional assays. Within the identified clones, three had been previously described as interacting proteins (fibronectin, SMOC1 and GSTO2), while all the others were new. When compared with standard systems, such as microtiter enzyme-linked immunosorbant assay, the method described here is dramatically faster and yields far more information about the interaction under study, allowing better characterization of complex systems. For example, in the case of fibronectin, it was possible to identify the specific domains involved in the interaction.

Charges drive selection of specific antibodies by phage display

Phage display technology has emerged as a leading approach to select proteins with improved properties for many different types of applications. The selection typically selects not only for improved binding properties but also for other factors such as efficiency of protein production and folding in Escherichia coli, the host in which the proteins and the phage are produced. Furthermore, the selection methodology is likely to influence the character of retrieved variants. We have now defined the extent whereby the charge of the displayed proteins influence the selection process, resulting in an increased average positive charge among selected proteins in comparison to the proteins that are harbored in the library before selection. Implications of and possible routes to minimize this effect are discussed.

Human monomeric antibody fragments to TRAIL-R1 and TRAIL-R2 that display potent in vitro agonism

Apoptosis through the TRAIL receptor pathway can be induced via agonistic IgG to either TRAIL-R1 or TRAIL-R2. Here we describe the use of phage display to isolate a substantive panel of fully human anti-TRAIL receptor single chain Fv fragments (scFvs); 234 and 269 different scFvs specific for TRAIL-R1 and TRAIL-R2 respectively. In addition, 134 different scFvs that were cross-reactive for both receptors were isolated. To facilitate screening of all 637 scFvs for potential agonistic activity in vitro, a novel high-throughput surrogate apoptosis assay was developed. Ten TRAIL-R1 specific scFv and 6 TRAIL-R2 specific scFv were shown to inhibit growth of tumor cells in vitro in the absence of any cross-linking agents. These scFv were all highly specific for either TRAIL-R1 or TRAIL-R2, potently inhibited tumor cell proliferation, and were antagonists of TRAIL binding. Moreover, further characterization of TRAIL-R1 agonistic scFv demonstrated significant anti-tumor activity when expressed and purified as a monomeric Fab fragment. Thus, scFv and Fab fragments, in addition to whole IgG, can be agonistic and induce tumor cell death through specific binding to either TRAIL-R1 or TRAIL-R2. These potent agonistic scFv were all isolated directly from the starting phage antibody library and demonstrated significant tumor cell killing properties without any requirement for affinity maturation. Some of these selected scFv have been converted to IgG format and are being studied extensively in clinical trials to investigate their potential utility as human monoclonal antibody therapeutics for the treatment of human cancer.

Affinity maturation generates greatly improved xyloglucan-specific carbohydrate binding modules

BACKGROUND

Molecular evolution of carbohydrate binding modules (CBM) is a new approach for the generation of glycan-specific molecular probes. To date, the possibility of performing affinity maturation on CBM has not been investigated. In this study we show that binding characteristics such as affinity can be improved for CBM generated from the CBM4-2 scaffold by using random mutagenesis in combination with phage display technology.

RESULTS

Two modified proteins with greatly improved affinity for xyloglucan, a key polysaccharide abundant in the plant kingdom crucial for providing plant support, were generated. Both improved modules differ from other existing xyloglucan probes by binding to galactose-decorated subunits of xyloglucan. The usefulness of the evolved binders was verified by staining of plant sections, where they performed better than the xyloglucan-binding module from which they had been derived. They discriminated non-fucosylated from fucosylated xyloglucan as shown by their ability to stain only the endosperm, rich in non-fucosylated xyloglucan, but not the integument rich in fucosylated xyloglucan, on tamarind seed sections.

CONCLUSION

We conclude that affinity maturation of CBM selected from molecular libraries based on the CBM4-2 scaffold is possible and has the potential to generate new analytical tools for detection of plant carbohydrates.

Automated Panning and Screening Procedure on Microplates for Antibody Generation from Phage Display Libraries

Antibody phage display technology is well established and widely used for selecting specific antibodies against desired targets. Using conventional manual methods, it is laborious to perform multiple selections with different antigens simultaneously. Furthermore, manual screening of the positive clones requires much effort. The authors describe optimized and automated procedures of these processes using a magnetic bead processor for the selection and a robotic station for the screening step. Both steps are performed in a 96-well microplate format. In addition, adopting the antibody phage display technology to automated platform polyethylene glycol precipitation of the enriched phage pool was unnecessary. For screening, an enzyme-linked immunosorbent assay protocol suitable for a robotic station was developed. This system was set up using human γ-globulin as a model antigen to select antibodies from a VTT naive human single-chain antibody (scFv) library. In total, 161 γ-globulin-selected clones were screened, and according to fingerprinting analysis, 9 of the 13 analyzed clones were different. The system was further tested using testosterone bovine serum albumin (BSA) and β-estradiol-BSA as antigens with the same library. In total, 1536 clones were screened from 4 rounds of selection with both antigens, and 29 different testosterone-BSA and 23 β-estradiol-BSA binding clones were found and verified by sequencing. This automated antibody phage display procedure increases the throughput of generating wide panels of target-binding antibody candidates and allows the selection and screening of antibodies against several different targets in parallel with high efficiency. ( Journal of Biomolecular Screening 2009:282-293)

Multiplexed flow cytometry: high-throughput screening of single-chain antibodies

The development of high-throughput screening (HTS) technologies has become essential for initial characterization of recombinant antibodies and alternative affinity reagents, selected from large combinatorial libraries. Such binding ligands are routinely selected against a single antigen and screened for desired binding specificities. Recent progress with genome sequencing projects has led to widespread efforts to study corresponding proteomes; requiring selection of ligands against large numbers of gene products in a highly parallel manner. The capabilities of many routine HTS methods such as enzyme-linked immunosorbent assay (ELISA), or array-based methods, are limited to analysis of numerous different antibody clones against a single target or, individual antibody clones against many different targets. We have developed a multiplexed flow cytometry screening method that allows analysis of individual binding ligands against numerous targets in the same analytical sample. The method produces a complex analytical profile for each antibody clone in the primary screen, by allowing simultaneous determination of relative expression levels, identification of non-specific binding, and discrimination of fine specificities. The quality and quantity of data, combined with significant reductions in analysis time and antigen consumption, provide notable advantages over other standard screening methods, such as ELISA. By combining HT screening capabilities with multiplex technology, we have redefined the parameters for the initial identification of affinity reagents recovered from combinatorial libraries and removed a significant bottleneck in the generation of affinity reagents on a proteomic scale.

Eliminating helper phage from phage display

Phage display technology involves the display of proteins or peptides, as coat protein fusions, on the surface of a phage or phagemid particles. Using standard technology, helper phage are essential for the replication and assembly of phagemid particles, during library production and biopanning. We have eliminated the need to add helper phage by using 'bacterial packaging cell lines' that provide the same functions. These cell lines contain M13-based helper plasmids that express phage packaging proteins which assemble phagemid particles as efficiently as helper phage, but without helper phage contamination. This results in genetically pure phagemid particle preparations. Furthermore, by using constructs differing in the form of gene 3 that they contain, we have shown that the display, from a single library, can be modulated between monovalent (phagemid-like) and multivalent display (phage-like) without any further engineering. These packaging cells eliminate the use of helper phage from phagemid-based selection protocols; reducing the amount of technical preparation, facilitating automation, optimizing selections by matching display levels to diversity, and effectively using the packaged phagemid particles as means to transfer genetic information at an efficiency approaching 100%.

Selecting and screening recombinant antibody libraries

During the past decade several display methods and other library screening techniques have been developed for isolating monoclonal antibodies (mAbs) from large collections of recombinant antibody fragments. These technologies are now widely exploited to build human antibodies with high affinity and specificity. Clever antibody library designs and selection concepts are now able to identify mAb leads with virtually any specificity. Innovative strategies enable directed evolution of binding sites with ultra-high affinity, high stability and increased potency, sometimes to a level that cannot be achieved by immunization. Automation of the technology is making it possible to identify hundreds of different antibody leads to a single therapeutic target. With the first antibody of this new generation, adalimumab (Humira, a human IgG1 specific for human tumor necrosis factor (TNF)), already approved for therapy and with many more in clinical trials, these recombinant antibody technologies will provide a solid basis for the discovery of antibody-based biopharmaceuticals, diagnostics and research reagents for decades to come.

Efficient construction of a highly useful phage-displayed human antibody repertoire

We have constructed a highly useful phage-displayed human antibody repertoire with limited cloning efforts. Our strategy was to maximize diversity during the first steps of library construction through the use of various lymphoid sources from several donors, inclusion of different immunoglobulin isotypes, and performance of multiple separate amplification reactions with all possible combinations within a complex primer set. The resulting variable region collections were cloned to form a moderate size library, composed by 4.25x10(8) single chain antibody fragments. This repertoire was successfully used to retrieve binders to seven model antigens: six proteins and one 12 aa peptide. Binding affinities reached nanomolar and even subnanomolar range. Sequence diversity and V-gene usage variability among binders were proven. Our approach was not focused on absolute library size, but on a high quality sampling of variable regions from the human antibody repertoire.

A human monoclonal antibody neutralizes diverse HIV-1 isolates by binding a critical gp41 epitope

HIV-1 entry into cells is mediated by the envelope glycoprotein receptor-binding (gp120) and membrane fusion-promoting (gp41) subunits. The gp41 heptad repeat 1 (HR1) domain is the molecular target of the fusion-inhibitor drug enfuvirtide (T20). The HR1 sequence is highly conserved and therefore considered an attractive target for vaccine development, but it is unknown whether antibodies can access HR1. Herein, we use gp41-based peptides to select a human antibody, 5H/I1-BMV-D5 (D5), that binds to HR1 and inhibits the assembly of fusion intermediates in vitro. D5 inhibits the replication of diverse HIV-1 clinical isolates and therefore represents a previously unknown example of a crossneutralizing IgG selected by binding to designed antigens. NMR studies and functional analyses map the D5-binding site to a previously identified hydrophobic pocket situated in the HR1 groove. This hydrophobic pocket was proposed as a drug target and subsequently identified as a common binding site for peptide and peptidomimetic fusion inhibitors. The finding that the D5 fusion-inhibitory antibody shares the same binding site suggests that the hydrophobic pocket is a "hot spot" for fusion inhibition and an ideal target on which to focus a vaccine-elicited antibody response. Our data provide a structural framework for the design of new immunogens and therapeutic antibodies with crossneutralizing potential.

Perspectives for systematic in vitro antibody generation

After the completion and refinement of the human genome, the characterization of individual gene products in respect of their functions, their modifications, their cellular localization and regulation in both space and time has generated an increased demand for antibodies for their analysis. Taking into account that the human genome contains approximately 25,000 genes, and that their products are found in different splice variants and produce proteins with post-translational modifications, it can be estimated that at least 100,000 different protein products have to be investigated to gain a complete picture of what's going on in the proteome of a cell. Antibodies are preferred tools helping with the characterization and detection of proteins as well as with elucidating their individual functions. The generation of antibodies to all available human protein products by immunization and/or the hybridoma technology is not only logistically and financially enduring, but may prove to be a difficult task, as quite a number of interesting targets may evade the immune response of experimental animals, for example, allosteric variants dependent on fragile interactions to cofactors, highly conserved antigens etc. For this reason, alternative methods for the generation of antibodies have to supplement these approaches. In vitro methods for antibody generation are seen to offer this capability. In addition, they may provide a cost effective and large scale production alternative for detection reagents for the research community in their own right. Among in vitro techniques, phage display has been evolved as the most efficient option for tackling this problem and approaches optimised for automation are emerging. Maximum benefit for proteomic research could be generated by judicious and preferably international coordination of the ongoing efforts to combine the strengths of the well established animal based approaches and the novel opportunities offered by in vitro methods.

Protein studies in dysferlinopathy patients using llama-derived antibody fragments selected by phage display

Mutations in dysferlin, a member of the fer1-like protein family that plays a role in membrane integrity and repair, can give rise to a spectrum of neuromuscular disorders with phenotypic variability including limb-girdle muscular dystrophy 2B, Myoshi myopathy and distal anterior compartment myopathy. To improve the tools available for understanding the pathogenesis of the dysferlinopathies, we have established a large source of highly specific antibody reagents against dysferlin by selection of heavy-chain antibody fragments originating from a nonimmune llama-derived phage-display library. By utilizing different truncated forms of recombinant dysferlin for selection and diverse selection methodologies, antibody fragments with specificity for two different dysferlin domains could be identified. The selected llama antibody fragments are functional in Western blotting, immunofluorescence microscopy and immunoprecipitation applications. Using these antibody fragments, we found that calpain 3, which shows a secondary reduction in the dysferlinopathies, interacts with dysferlin.

Subtle sequence differences in a tumour-associated peptide epitope translate into major changes in antigenicity

Antigenicity, the ability to bind to members of repertoire of diverse immune receptors, is a concept that is poorly characterised with respect to its defining parameters. To learn more about its makeup, we have investigated the ability of two peptides with highly related sequences, derived from the tumour-associated antigen mucin-1, to recruit in vitro members from a large naïve repertoire of synthetic human antibody fragments. One of the peptides represents the epitope that is immunodominant in mice. We now demonstrate that the other peptide, which differs from the first only by a very conservative aspartate-threonine to glutamate-serine change, is much less antigenic than the first peptide. This is so despite the fact that there is no observable difference in the tendency of the two peptides to adopt a structure in solution. Furthermore, the peptides differ in their immunodominant parts and the less antigenic peptide selects for antibody fragments targeting residues outside of the epitope considered to be immunodominant in mice. We conclude that subtle sequence changes greatly, affect antigenicity and immunodominance of epitopes in this important tumour-associated antigen.

Oligonucleotide-assisted cleavage and ligation: a novel directional DNA cloning technology to capture cDNAs. Application in the construction of a human immune antibody phage-display library

The use of oligonucleotide-assisted cleavage and ligation (ONCL), a novel approach to the capture of gene repertoires, in the construction of a phage-display immune antibody library is described. ONCL begins with rapid amplification of cDNA ends to amplify all members equally. A single, specific cut near 5′ and/or 3′ end of each gene fragment (in single stranded form) is facilitated by hybridization with an appropriate oligonucleotide adapter. Directional cloning of targeted DNA is accomplished by ligation of a partially duplex DNA molecule (containing suitable restriction sites) and amplification with primers in constant regions. To demonstrate utility and reliability of ONCL, a human antibody repertoire was cloned from IgG mRNA extracted from human B-lymphocytes engrafted in Trimera mice. These mice were transplanted with peripheral blood lymphocytes from Candida albicans infected individuals and subsequently immunized with C.albicans glyceraldehyde-3-phosphate dehydrogenase (GAPDH). DNA sequencing showed that ONCL resulted in efficient capture of gene repertoires. Indeed, full representation of all V_H families/segments was observed showing that ONCL did not introduce cloning biases for or against any V_H family. We validated the efficiency of ONCL by creating a functional Fab phage-display library with a size of 3.3 × 10¹⁰ and by selecting five unique Fabs against GAPDH antigen.

A strategy for the generation of specific human antibodies by directed evolution and phage display

This study describes the construction of a library of single-chain antibody fragments (scFvs) from a single human donor by individual amplification of all heavy and light variable domains (1.1 x 10(8) recombinants). The library was panned using the phage display technique, which allowed selection of specific scFvs (3F and C1) capable of recognizing Cn2, the major toxic component of Centruroides noxius scorpion venom. The scFv 3F was matured in vitro by three cycles of directed evolution. The use of stringent conditions in the third cycle allowed the selection of several improved clones. The best scFv obtained (6009F) was improved in terms of its affinity by 446-fold, from 183 nm (3F) to 410 pm. This scFv 6009F was able to neutralize 2 LD(50) of Cn2 toxin when a 1 : 10 molar ratio of toxin-to-antibody fragment was used. It was also able to neutralize 2 LD(50) of the whole venom. These results pave the way for the future generation of recombinant human antivenoms.

Construction, evaluation and refinement of a large human antibody phage library based on the IgD and IgM variable gene repertoire

The ability to isolate antibodies against any antigen of interest has become increasingly important as antibodies have proved their utility both in antigen detection, quantification and as specific in vivo targeting agents. To this end, we have constructed a large antibody phage library in the single chain Fv (scFv) phagemid format based on the naive human variable (V) gene repertoire dictated by IgD and IgM. Optimizing each step of the library construction has resulted in a highly diverse and functional library, as assessed by sequencing analysis, large-scale automated expression analysis and antigen screening. Furthermore, the versatile format of the library, which comprises 14 separate sub-libraries, adds considerably flexibility with respect to which part of the antibody repertoire that is to be probed. This versatility has been further exploited to generate a refined antibody library, which exhibits one of the highest prokaryotic expression levels reported to date for a naive repertoire. The construction of the refined library was based on the functional purification of expressed V genes in the context of the protein L interaction with correctly folded V genes of the kappa light chain family. Antigen screening of this library indicated that the functional purification improved the ability to retrieve antigen specific antibodies, but at the cost of potential loss of diversity in the isolated repertoire.

Directed Evolution, Phage Display and Combination of Evolved Mutants: A Strategy to Recover the Neutralization Properties of the scFv Version of BCF2 a Neutralizing Monoclonal Antibody Specific to Scorpion Toxin Cn2

BCF2, a monoclonal antibody raised against scorpion toxin Cn2, is capable of neutralizing both, the toxin and the whole venom of the Mexican scorpion Centruroides noxius Hoffmann. The single chain antibody fragment (scFv) of BCF2 was constructed and expressed in Escherichia coli. Although its affinity for the Cn2 toxin was shown to be in the nanomolar range, it was non-neutralizing in vivo due to a low stability. In order to recover the neutralizing capacity, the scFv of BCF2 was evolved by error-prone PCR and the variants were panned by phage display. Seven improved mutants were isolated from three different libraries. One of these mutants, called G5 with one mutation at CDR1 and another at CDR2 of the light chain, showed an increased affinity to Cn2, as compared to the parental scFv. A second mutant, called B7 with a single change at framework 2 of heavy chain, also had a higher affinity. Mutants G5 and B7 were also improved in their stability but they were unable to neutralize the toxin. Finally, we constructed a variant containing the changes present in G5 and B7. The purpose of this construction was to combine the increments in affinity and stability borne by these mutants. The result was a triple mutant capable of neutralizing the Cn2 toxin. This variant showed the best affinity constant (KD=7.5x10(-11) M), as determined by surface plasmon resonance (BIAcore). The k(on) and k(off) were improved threefold and fivefold, respectively, leading to 15-fold affinity improvement. Functional stability determinations by ELISA in the presence of different concentrations of guanidinium hydrochloride (Gdn-HCl) revealed that the triple mutant is significantly more stable than the parental scFv. These results suggest that not only improving the affinity but also the stability of our scFv were important for recovering its neutralization capacity. These findings pave the way for the generation of recombinant neutralizing antisera against scorpion stings based on scFvs.

Fluorobodies combine GFP fluorescence with the binding characteristics of antibodies

The difficulty of deriving binding ligands to targets identified by genomic sequencing has led to a bottleneck in genomic research. By inserting diverse antibody binding loops into four of the exposed loops at one end of green fluorescent protein (GFP), we have mimicked the natural antibody binding footprint to create robust binding ligands that combine the advantages of antibodies (high affinity and specificity) with those of GFP (intrinsic fluorescence, high stability, expression and solubility). These 'fluorobodies' have been used effectively in enzyme-linked immunosorbent assays (ELISAs), flow cytometry, immuno-fluorescence, arrays and gel shift assays, and show affinities as high as antibodies. Furthermore, the intrinsic fluorescence of fluorobodies correlates with binding activity, allowing the rapid determination of functionality, concentration and affinity. These properties render them especially suitable for the high-throughput genomic scale selections required in proteomics, as well as in diagnostics, target validation and drug development.

The Remarkable Flexibility of the Human Antibody Repertoire; Isolation of Over One Thousand Different Antibodies to a Single Protein, BLyS

It is well established that the humoral immune response can generate antibodies to many different antigens. The antibody diversity required to achieve this is believed to be substantial. However, the extent to which the immune repertoire can generate structural diversity against a single target antigen has never been addressed. Here, we have used phage display to demonstrate the extraordinary capacity of the human antibody repertoire. Over 1000 antibodies, all different in amino acid sequence, were generated to a single protein, B-lymphocyte stimulator (BLyS protein). This is a highly diverse panel of antibodies as exemplified by the extensive heavy and light chain germline usage: 42/49 functional heavy chain germlines and 19/33 V(lambda) and 13/35 V(kappa) light chain germlines were all represented in the panel of antibodies. Moreover, a high level of sequence diversity was observed in the V(H) CDR3 domains of these antibodies, with 568 different amino acid sequences identified. Thus we have demonstrated that specific recognition of a single antigen can be achieved from many different VDJ combinations, illustrating the remarkable problem-solving ability of the human immune repertoire. When studied in a biochemical assay, around 500 (40%) of these antibodies inhibited the binding of BLyS to its receptors on B-cell lines. The most potent antibodies inhibited BLyS binding with sub-nanomolar IC(50) values and with sub-nanomolar affinities. Such antibodies provide excellent choices as candidates for the treatment of BLyS-associated autoimmune diseases.

Recombinant single-chain Fv antibodies that recognize the p25 protein of the Maedi-Visna virus

Single chain Fv (scFv) antibodies (generated by phage display technology, molecules representing new and efficient tools in the research and diagnostics of infectious diseases) against the capsid protein (p25) of Maedi-Visna virus were selected. Several clones of p25 specific scFv antibodies were identified; one of them was expressed as a soluble scFv molecule, purified by immobilized metal-affinity chromatography and further characterized by sequencing and determination of the kinetic equilibrium association constant. Sequence analysis showed that the rearranged VL and VH domains of the analyzed scFv clone used sequences from the VL3 family (germline DPL16/VL3.1) and VH1 family (germline VH20), respectively. The kinetic equilibrium association constant was determined as KA = 1.12 +/- 0.52 L/mumol.

Antibodies in proteomics II: screening, high-throughput characterization and downstream applications

There are many ways in which the use of antibodies and antibody selection can be improved and developed for high-throughput characterization. Standard protocols, such as immunoprecipitation, western blotting and immunofluorescence, can be used with antibody fragments generated by display technologies. Together with novel approaches, such as antibody chips and intracellular immunization, these methods will yield useful proteomic data following adaptation of the protocols for increased reliability and robustness. To date, most work has focused on the use of standard, well-characterized commercial antibodies. Such protocols need to be adapted for broader use, for example, with antibody fragments or other binders generated by display technologies, because it is unlikely that traditional approaches will provide the required throughput.

Antibodies in proteomics I: generating antibodies

The explosion in genome sequencing, and in subsequent DNA array experiments, has provided extensive information on gene sequence, organization and expression. This has resulted in a desire to perform similarly broad experiments on all the proteins encoded by a genome. Panels of specific antibodies, or other binding ligands, will be essential tools in this endeavour. Because traditional immunization will be unlikely to generate antibodies in sufficient quantity, and of the required quality and reproducibility, in vitro selection methods will probably be used. This review--the first of two--examines the strategies available for in vitro antibody selection. The second review discusses the adaptation of these methods to high throughput and the uses to which antibodies, once derived, can be put.

A new helper phage and phagemid vector system improves viral display of antibody Fab fragments and avoids propagation of insert-less virions

Phage display technology (PDT) is a powerful method for isolating functional gene products such as antigen-specific monoclonal antibodies (mAbs). To improve the effectiveness of PDT, we sought to optimize display of Fab-g3p (antibody fragment fused with viral gene 3 protein) on phagemid virions and to optimize the yield of such phage. To do so, we constructed a novel helper phage, Phaberge, having a conditional deficiency in g3p production. Unlike most other published g3p-deficient helper phage, Phaberge is produced at high levels, 10(11) PFU/ml. As compared to g3p-sufficient helper phage, Phaberge caused a 5-20-fold increase in display level. Another novel feature is that Phaberge only packages insert-containing, not insert-less, phagemid into infectious virions. This should prove useful in preserving quality of phagemid libraries during propagation. In addition, other parameters were also found to affect production of phagemid virions. In particular, the choice of bacterial host cell, phagemid construct and growth temperature had a substantial impact on display levels, but generally no effect on number of phagemid virions produced. In short, we have established a set of parameters that improve production and quality of phagemid virions which we expect to facilitate the isolation of mAbs or other gene products by PDT.

Polyclonal Fab phage display libraries with a high percentage of diverse clones to Cryptosporidium parvum glycoproteins

The protozoan parasite Cryptosporidium parvum is regarded as a major public health problem world-wide, especially for immunocompromised individuals. Although no effective therapy is presently available, specific immune responses prevent or terminate cryptosporidiosis and passively administered antibodies have been found to reduce the severity of infection. Therefore, as an immunotherapeutic approach against cryptosporidiosis, we set out to develop C. parvum-specific polyclonal antibody libraries, standardised, perpetual mixtures of polyclonal antibodies, for which the genes are available. A combinatorial Fab phage display library was generated from the antibody variable region gene repertoire of mice immunised with C. parvum surface and apical complex glycoproteins which are believed to be involved in mediating C. parvum attachment and invasion. The variable region genes used to construct this starting library were shown to be diverse by nucleotide sequencing. The library was subjected to one round of antigen selection on C. parvum glycoproteins or a C. parvum oocyst/sporozoite preparation. The two selected libraries showed specific reactivity to the glycoproteins as well as to the oocyst/sporozoite preparation, with 50-73% antigen-reactive members. Fingerprint analysis of individual clones from the two antigen-selected libraries showed high diversity, confirming the polyclonality of the selected libraries. Furthermore, immunoblot analysis on the oocyst/sporozoite and glycoprotein preparations with selected library phage showed reactivity to multiple bands, indicating diversity at the antigen level. These C. parvum-specific polyclonal Fab phage display libraries will be converted to libraries of polyclonal full-length antibodies by mass transfer of the selected heavy and light chain variable region gene pairs to a mammalian expression vector. Such polyclonal antibody libraries would be expected to mediate effector functions and provide optimal passive immunity against cryptosporidiosis.

Flow-cytometric isolation of human antibodies from a nonimmune Saccharomyces cerevisiae surface display library

A nonimmune library of 10(9) human antibody scFv fragments has been cloned and expressed on the surface of yeast, and nanomolar-affinity scFvs routinely obtained by magnetic bead screening and flow-cytometric sorting. The yeast library can be amplified 10(10)-fold without measurable loss of clonal diversity, allowing its effectively indefinite expansion. The expression, stability, and antigen-binding properties of >50 isolated scFv clones were assessed directly on the yeast cell surface by immunofluorescent labeling and flow cytometry, obviating separate subcloning, expression, and purification steps and thereby expediting the isolation of novel affinity reagents. The ability to use multiplex library screening demonstrates the usefulness of this approach for high-throughput antibody isolation for proteomics applications.

Antibody discovery: phage display

Phage display has proven to be a robust and convenient technology for the selection of high-quality human antibodies from diverse libraries. Besides enabling the identification of antibodies in a fast, high-throughput mode, which allows comprehensive protein expression analyses, phage display has been used to identify a fully human therapeutic antibody presently undergoing the regulatory process for market approval.

Polyclonal anti-colorectal cancer Fab phage display library selected in one round using density gradient centrifugation to separate antigen-bound and free phage

A combinatorial Fab phage display library generated from antibody variable (V) region genes of BALB/c mice immunized with the human colorectal cancer cell lines SW480, SW948, and SW837, was used to isolate an anti-colorectal cancer library. In an attempt to preserve as many anti-colorectal cancer specificities as possible, the original Fab phage display library was selected for binding to a suspension of the human colorectal cancer cells using density gradient centrifugation, instead of washes, to separate cell-bound and free phage. The method consists of placing the cell-phage mixture on a layer of fetal bovine serum (FBS) which had been overlaid on a "cushion" of percoll density medium in a soft, polyallomer tube. After centrifugation, free phage remain on top of the serum layer, whereas the colorectal cancer cells with bound phage are recovered from the serum-percoll interface with a syringe. Analysis of the selected phage display library by enzyme linked immunosorbent assay (ELISA) and diagnostic restriction enzyme digests of individual members (fingerprinting), revealed about 90% anti-colorectal cancer diverse clones after only one round of selection. After conversion to a library of full-length antibodies, such an anti-cancer polyclonal library could be useful for therapeutic and/or diagnostic applications. The density gradient centrifugation method presented here holds great promise for the generation of polyclonal antibody libraries (PCALs) to complex antigens. It is also applicable for selection of peptide or other phage display libraries on any insoluble ligand.

A naturally occurring NAR variable domain binds the Kgp protease from Porphyromonas gingivalis

The new antigen receptor (NAR) from sharks consists of a single immunoglobulin variable domain attached to five constant domains, and is hypothesised to function as an antibody. Two closely related NARs with affinity for the Kgp (lysine-specific) gingipain protease from Porphyromonas gingivalis were selected by panning an NAR variable domain library. When produced in Escherichia coli, these recombinant NARs were stable, correctly folded, and specifically bound Kgp (K(d)=1.31+/-0.26x10(-7) M). Binding localised to the Kgp adhesin domains, however without inhibiting adhesin activity. These naturally occurring proteins indicate an immune response to pathogenic bacteria and suggest that the NAR is a true antibody-like molecule.