Search

This protocol describes a method for creation of a highly diverse and functional synthetic phage-displayed repertoire of fully human domain antibodies (dAbs). The repertoire is based on two human frameworks (one VH and one Vκ) that express well in bacteria and are frequently used in human antibodies. To achieve this, we first build dAb libraries, containing full synthetic diversity at key positions in the complementarity-determining regions (CDRs). We then use an antigen-independent preselection of this primary dAb repertoire on generic ligands of the VH and the Vκ scaffolds (namely, the bacterial superantigens, protein A and L) to enrich for folded dAbs. Finally, the CDRs of these preselected dAbs are randomly recombined to further expand genetic diversity. The resulting phage repertoire is in excess of 10(10) clones and is largely populated by correctly folded (over 50%) functional dAbs.

We have used phage display to isolate a range of human domain antibodies (dAbs) that bind to mouse, rat and/or human serum albumin (SA) and can be expressed at very high levels in bacterial, yeast or mammalian cell culture. In contrast to non-SA-binding dAbs, which have terminal half-lives of less than 45 min, the half-lives of these 12 kDa 'AlbudAbs' can match the half-life of SA itself. To demonstrate the use of AlbudAbs for extending the half-lives of therapeutic drugs, we created a fusion of the interleukin-1 receptor antagonist (IL-1ra) with an AlbudAb. Soluble IL-1ra is potent inhibitor of IL-1 signalling that is approved for the treatment of rheumatoid arthritis but has a relatively short in vivo half-life. Here we show that although the AlbudAb/IL-1ra fusion has a similar in vitro potency, its in vivo efficacy can be dramatically improved due to its extended serum half-life. AlbudAbs could potentially be used to generate a range of long half-life versions of many different drugs in order to improve their dosing regimen and/or clinical effect.

Protein L (PpL) is a B-cell superantigen from Peptostreptococcus magnus known to bind to mammalian Vkappa light chains. PpL from P.magnus strain 312 comprises five homologous immunoglobulin (Ig) binding domains. We first analysed the binding of the individual domains (B1-B5) of PpL(312) to human Vkappa light chains (huVkappa) subtypes 1 (huVkappaI) and 3 (huVkappaIII). Using a combination of rational design and phage selection we isolated mutants of the N-terminal B1 domain with a 14-fold increased affinity for huVkappa1 (B1kappa1) and >tenfold increased affinity for huVkappaIII (B1kappa3). We investigated the potential of the selected domains, in particular the B1kappa1 domain, as reagents in immunochemistry and immunotherapy. B1kappa1 proved a superior reagent than the wild-type domain, allowing up to tenfold more sensitive detection of human Vkappa antibody fragments in ELISA. A fusion protein of B1kappa1 with a human Vlambda antibody scFv fragment promoted the efficient recruitment of antibody encoded effector functions including complement, mononuclear phagocyte respiratory burst and phagocytosis through retargeting of IgGkappa and IgMkappa. Our results suggest that superantigens with improved affinity and/or specificity are easily accessible through protein engineering. Such engineered superantigens should prove useful as reagents in immunochemistry and may have potential as agents in immunotherapy.

Occurring naturally in "heavy chain" immunoglobulins from camels, and now produced in fully human form, domain antibodies (dAbs) are the smallest known antigen-binding fragments of antibodies, ranging from 11 kDa to 15 kDa. dAbs are the robust variable regions of the heavy and light chains of immunoglobulins (VH and VL respectively). They are highly expressed in microbial cell culture, show favourable biophysical properties including solubility and temperature stability, and are well suited to selection and affinity maturation by in vitro selection systems such as phage display. dAbs are bioactive as monomers and, owing to their small size and inherent stability, can be formatted into larger molecules to create drugs with prolonged serum half-lives or other pharmacological activities.

Monoclonal antibodies play an increasingly important role in structural biology. In this report, we develop the use of phage display technology for the isolation of an antibody that binds to a specific subunit of a macromolecular assembly. Antibodies that bind to the intact complex are selected from a phage display library and screened with a high-density Western blot to identify a subunit-specific binder. Conventional Western blotting and competition ELISA are then used to confirm the identity of the target subunit and that the antibody binds to the native protein complex and not to an epitope that is only revealed when the antibody is immobilized for phage selection. Using this technique, monoclonal scFv and Fab fragments have been produced that bind to the 51-kDa subunit of bovine complex I, a large integral membrane protein complex from mitochondria.

We have combined phage display and array screening for the rapid isolation of pairs of interacting polypeptides. Our strategy, named SAC (selection by avidity capture), is based on the avidity effect, the formation of highly stable complexes formed by multivalent interactions; in our case, between a receptor (multivalently displayed on phage) and a ligand (coexpressed as a multimeric fusion protein). Capture of the long-lived interaction complex allows the isolation of phage bearing cognate interaction pairs, as we demonstrate for a range of interactions, including Ab-antigen pairs and the rapamycin-dependent interaction of FKBP-12 and FRAP. Cognate phage are enriched by SAC up to 1000-fold and interacting pairs can be identified by array screening. Application of SAC to Ab-antigen interactions as a model system yielded over 140 specific Abs to a single antigen and 92 Abs to three different fetal human brain antigens in a single round of SAC each. Our results suggest that SAC should prove useful for the identification and study of receptor-ligand interactions in particular among extracellular proteins, as well as for the rapid generation of specific Abs to multiple antigens.

Ever since DNA microarrays were first applied to the quantitation of RNA levels, there has been considerable interest in generating a protein homolog that can be used to assay cellular protein expression. A recent paper describes the first microarray that can be used for such protein profiling.

Recombinant antibodies are becoming increasingly important in the field of proteomics. Recent advances include the development of large phage-antibody libraries that contain high-affinity binders to almost any target protein, and new methods for high-throughput selection of antibody-antigen interactions. Coupled with a range of new screening technologies that use high-density antibody arrays to identify differentially expressed proteins, these antibody libraries can be applied to whole proteome analysis.

We have developed a novel technique for high-throughput screening of recombinant antibodies, based on the creation of antibody arrays. Our method uses robotic picking and high-density gridding of bacteria containing antibody genes followed by filter-based enzyme-linked immunosorbent assay (ELISA) screening to identify clones that express binding antibody fragments. By eliminating the need for liquid handling, we can thereby screen up to 18,342 different antibody clones at a time and, because the clones are arrayed from master stocks, the same antibodies can be double spotted and screened simultaneously against 15 different antigens. We have used our technique in several different applications, including isolating antibodies against impure proteins and complex antigens, where several rounds of phage display often fail. Our results indicate that antibody arrays can be used to identify differentially expressed proteins.

We have developed a system to identify highly specific antibody-antigen interactions by protein array screening. This removes the need for selection using animal immunisation or in vitro techniques such as phage or ribosome display. We screened an array of 27 648 human foetal brain proteins with 12 well-expressed antibody fragments that had not previously been exposed to any antigen. Four highly specific antibody-antigen pairs were identified, including three antibodies that bind proteins of unknown function. The target proteins were expressed at a very low copy number on the array, emphasising the unbiased nature of the screen. The specificity and sensitivity of binding demonstrates that this 'naive' screening approach could be applied to the high throughput isolation of specific antibodies against many different targets in the human proteome.

Systemic lupus erythematosus (SLE) is an autoimmune disease that is characterized by the presence of high immunoglobulin serum titers, but the mechanism by which these arise remains unclear. It has been suggested that the disease is associated with specific antibody features, including variable gene use, the presence of charged complementarity-determining region residues and/or an aberrant process of secondary light chain rearrangement. To study this in more detail, we compared variable, diversity and joining gene segment use, somatic mutation, and heavy and light chain pairings in single peripheral IgG(+) B cells between one normal (209 B cells) and two SLE (156 B cells) donors. In contrast to others, we found no systematic differences, indicating that the memory B cell repertoires in normal and SLE donors are shaped in a similar way.

We have sequenced the heavy and light chain genes from 365 IgG(+) B cells and found that 24 (6.5 %) contain somatically introduced insertions or deletions. These insertions and deletions are clustered at "hot-spots" in the antigen-binding site and frequently result in the creation of new combinations of canonical loop structures or entirely new loops that are not present in the human germline repertoire, but are similar to those seen in other species. Somatic insertion and deletion therefore provides a further mechanism for introducing structural diversity into antibodies in addition to somatic point mutation and receptor editing, which have small (single amino acid changes) and large (chain replacement) impacts on structural diversity, respectively., (Copyright 1999 Academic Press.)

The expressed human immunoglobulin Vlambda repertoire demonstrates a strong bias in the use of individual Vlambda segments. Mechanisms that underlie such biases can be divided into two categories: intrinsic genetic processes that lead to the preferential rearrangement and/or expression of certain segments; and selection following light chain expression. Here, we have used two approaches to investigate the factors that shape the human Vlambda repertoire. Firstly, we characterised 136 Vlambda rearrangements (59 productive and 77 non-productive) amplified from the human genomic DNA of peripheral blood cells. Secondly, we analysed Vlambda segment use in a library of 2000 cDNA clones from a transgenic mouse containing a 380 kb region (including 15 functional Vlambda segments) from the human immunoglobulin lambda locus. By hybridisation and sequencing we found that the patterns of use of human Vlambda segments in the transgenic mouse were similar to those found in the expressed human peripheral blood repertoire and in productive and non-productive genomic DNA rearrangements. These data indicate the importance of intrinsic genetic factors in shaping the human Vlambda repertoire and highlight the remarkable conservation of the molecular mechanisms involved in the production of the antibody repertoire in mouse and man. Therefore, transgenic mice represent a good model for analysis of the human antibody repertoire and for the production of human antibodies., (Copyright 1999 Academic Press.)

In the bone marrow, diversity in the primary antibody repertoire is created by the combinatorial rearrangement of different gene segments and by the association of different heavy and light chains. During the secondary response in the germinal centres, antibodies are diversified by somatic mutation and possibly by further rearrangements, or "receptor editing". Here, we have analysed the pairings of heavy and light chain variable domains (VH and VL) in 365 human IgG+ B cells from peripheral blood, and established that these pairings are largely random. The repertoire is dominated by a limited number of pairings of segments and folds. Among these pairings we identified two identical mutated heavy chains in combination with two different mutated light chains (one kappa and one lambda). This shows that receptor editing occurs in the human periphery and that the same antibody lineage can be subjected to both receptor editing and somatic hypermutation. This suggests that receptor editing may be used together with somatic mutation for the affinity maturation of antibodies. We also propose that receptor editing has shaped variable gene segment use and the evolution of V gene families., (Copyright 1999 Academic Press.)

We have determined the complete nucleotide sequence of the human immunoglobulin D segment locus on chromosome 14q32.3 and identified a total of 27 D segments, of which nine are new. Comparison with a database of rearranged heavy chain sequences indicates that the human antibody repertoire is created by VDJ recombination involving 25 of these 27 D segments, extensive processing at the V-D and D-J junctions and use of multiple reading frames. We could find no evidence for the proposed use of DIR segments, inverted D segments, "minor" D segments or D-D recombination. Conventional VDJ recombination, which obeys the 12/23 rule, is therefore sufficient to explain the wealth of lengths and sequences for the third hypervariable loop of human heavy chains.

Sequence diversity in the human antibody repertoire is generated in two steps: by the combinatorial assembly of V gene segments and by somatic hypermutation. Here, we have characterised these processes for the lambda (lambda) light chain using a library of 7600 lambda cDNA clones from peripheral blood lymphocytes. By hybridisation and sequencing we found that most lambda chains are derived from the cluster of V(lambda) segments closest to the J(lambda)-C(lambda) pairs and that there is considerable variation in the use of individual V(lambda) segments (ranging from 0.02% to 27%): three of the 30 functional V(lambda) segments encode half the expressed V(lambda) repertoire. As a result of these biases, sequence diversity in the primary repertoire is focused at the centre of the antigen binding site. By contrast, somatic hypermutation spreads diversity to the periphery. Comparison with the human kappa (kappa) light chain indicates that both kappa and lambda use the same strategy for searching sequence space and have almost identical patterns of diversity in the mature antibody repertoire.

We recently completed a map of the human immunoglobulin lambda (IGL) locus on chromosome 22q11.2 and showed that the V lambda genes are arranged in three distinct clusters, each containing members of different V lambda families. We have now sequenced each of these V lambda genes and determined which are functional by comparison with the expressed repertoire. Our analysis indicates that there are approximately 30 functional V lambda genes, depending on the haplotype, that belong to ten V lambda families (five V lambda 1, five V lambda 2, eight V lambda 3, three V lambda 4, three V lambda 5, one V lambda 6, two V lambda 7, one V lambda 8, one V lambda 9 and one V lambda 10). V lambda genes related to the major human V lambda families (V lambda 1, V lambda 2 and V lambda 3) predominate in species that express mainly lambda light chains.

In the human immune system, antibodies with high affinities for antigen are created in two stages. A diverse primary repertoire of antibody structures is produced by the combinatorial rearrangement of germline V gene segments and antibodies are selected from this repertoire by binding to the antigen. Their affinities are then improved by somatic hypermutation and further rounds of selection. We have dissected the sequence diversity created at each stage in response to a wide range of antigens. In the primary repertoire, diversity is focused at the centre of the binding site. With somatic hypermutation, diversity spreads to regions at the periphery of the binding site that are highly conserved in the primary repertoire. We propose that evolution has favoured this complementarity as an efficient strategy for searching sequence space and that the germline V gene families evolved to exploit the diversity created by somatic hypermutation.

In humans, the gene for the V kappa domain is produced by the recombination of one of 40 functional V kappa segments and one of five functional J kappa segments. We have analysed the sequences of these germline segments and of 736 rearranged V kappa genes to determine the repertoire of main chain conformations, or canonical structures, they encode. Over 96% of the sequences correspond to one of four canonical structures for the first antigen binding loop (L1) and one canonical structure for the second antigen binding loop (L2). Junctional diversity produces some variation in the length of the third antigen binding loop (L3) and in the identity of residues at the V kappa-J kappa join. However, this is limited and 70% of the rearranged sequences correspond to one of three known canonical structures for the L3 region. Furthermore, we show that the canonical structures selected during the primary response are conserved during affinity maturation: the key residues that determine the conformations of the antigen binding loops are unmutated or undergo conservative mutation. The implications of these results for immune recognition are discussed.

The maps of the human immunoglobulin heavy-chain and kappa light-chain loci have recently been completed. We have now completed a map of the human lambda locus (IGL) located on chromosome 22q11.2. We mapped 52 V lambda genes from 10 V lambda families and 7 J lambda and C lambda genes on a 1140 kb contig constructed from eight YACs and 129 cosmid clones. The V lambda genes are arranged within 800 kb. Genes of the different V lambda families are organized in three clusters, V lambda II and III families (cluster A); V lambda I, V, VII and IX families (cluster B); V lambda IV, VI, VIII and X families (cluster C), in contrast to the dispersed organization of the different VH and V kappa families within the human VH and V kappa loci. We note that the most frequently used V lambda families (V lambda II and III) are proximal to the J lambda and C lambda genes. The VpreB gene, encoding part of the surrogate light chain, the GGT2 gene and the BCRL4 pseudogene were also mapped within the lambda locus.

A complete map of the human immunoglobulin VH locus on chromosome 14 has recently been constructed. The locus is 1100kb in length and contains 51 functional VH segments interspersed amongst a similar number of pseudogenes. Here, Graham Cook and Ian Tomlinson review the organization of the locus, its polymorphism and the repertoire it encodes.

Antibody fragments of moderate affinity (approximately microM) can be isolated from repertoires of approximately 10(8) immunoglobulin genes by phage display and rounds of selection with antigen, and the affinities improved by further rounds of mutation and selection. Here, as an alternative strategy, we attempted to isolate high affinity human antibodies directly from large repertoires. We first created highly diverse repertoires of heavy and light chains entirely in vitro from a bank of human V gene segments and then, by recombination of the repertoires in bacteria, generated a large (close to 6.5 x 10(10)) synthetic repertoire of Fab fragments displayed on filamentous phage. From this repertoire we isolated Fab fragments which bound to a range of different antigens and haptens, and with affinities comparable with those of antibodies from a secondary immune response in mice (up to 4 nM). Although the VH-26 (DP-47) segment was the most commonly used segment in both artificial and natural repertoires, there were also major differences in the pattern of segment usage. Such comparisons may help dissect the contributions of biological mechanisms and structural features governing V gene usage in vivo.

Analysis of the telomeric region of chromosome 14q has enabled us to complete a map of the immunoglobulin VH locus which accounts for almost all VH segments known to rearrange in B-lymphocytes. The human germline VH repertoire consists of approximately 50 functional VH segments--the exact number depending on the haplotype--spanning 1,100 kilobases upstream of the JH segments. A yeast artificial chromosome used to map these segments was isolated by its ability to provide telomere activity in yeast, suggesting that the VH locus may be located within a few kilobases of the 14q telomere. The limited structural diversity encoded by the functional VH segments demonstrates the importance of combinatorial diversity produced by VDJ joining and the association of heavy and light chains in producing the human antibody repertoire.

In addition to the major human immunoglobulin heavy-chain locus on chromosome 14q32.3, VH and D segments are known to be present on chromosomes 15 and 16. We have now amplified and sequenced 24 such VH segments from somatic cell hybrids and have assigned them to 15q11.2 and 16p11.2 using cosmid and yeast artificial chromosome clones. In addition, we have located a cluster of D segments on 15q11.2, previously thought to be located on 14q32.3. We propose that the segments on chromosome 16 arose by an interchromosomal duplication and identify the corresponding region on chromosome 14. Taken together with the completion of a map of the human VH locus on 14q32.3, the total number of VH segments now identified is 117. We can now account for most, if not all human germ-line VH segments.

From the genomic DNA of a single individual, we have amplified, cloned and sequenced 37 human germ-line V kappa segments. Four of these segments were new. We then compiled a comprehensive directory of all germ-line V kappa segments and identified 50 different sequences with open reading frames. Comparison with 236 rearranged sequences revealed that no more than 24 of these germ-line sequences could be assigned rearranged counterparts, that some of these were rarely used, and that only about 11 sequences are used frequently. This suggests that the expressed V kappa repertoire is mainly derived from a limited number of segments. Most surprisingly, the J kappa-distal region of the locus appears to be rarely used: we could unambiguously assign 162 rearranged sequences to V kappa segments of the J kappa-proximal region, but only 5 to segments of the J kappa-distal region.

The display of repertoires of antibody fragments on the surface of filamentous bacteriophage offers a new way of making antibodies with predefined binding specificities. Here we explored the use of this technology to make immunochemical reagents to a range of antigens by selection from a repertoire of > 10(8) clones made in vitro from human V gene segments. From the same 'single pot' repertoire, phage were isolated with binding activities to each of 18 antigens, including the intracellular proteins p53, elongation factor EF-1 alpha, immunoglobulin binding protein, rhombotin-2 oncogene protein and sex determining region Y protein. Both phage and scFv fragments secreted from infected bacteria were used as monoclonal and polyclonal reagents in Western blots. Furthermore the monoclonal reagents were used for epitope mapping (a new epitope of p53 was identified) and for staining of cells. This shows that antibody reagents for research can be readily derived from 'single pot' phage display libraries.

We have identified and sequenced 14 human immunoglobulin VH segments cloned in a yeast artificial chromosome, and have used a rapid PCR-based technique (HAPPY mapping, 12) to derive the order and approximate distances between them. The sequences mapped comprise thirteen germline VH segments and one rearranged VH3 gene. Comparison of our map with other data suggests the existence of at least two distinct haplotypes, differing in the presence or absence of the consecutive genes DP-78, DP-46 and DP-64, and in the duplication of segments DP-49 and DP-65. Screening of ten individuals confirms the existence of both haplotypes, and indicates that both are common amongst the population.

The VH gene segments produce the part of the VH domains of antibodies that contains the first two hypervariable regions. The sequences of 83 human VH segments with open reading frames, from several individuals, are currently known. It has been shown that these sequences are likely to form a high proportion of the total human repertoire and that an individual's gene repertoire produces about 50 VH segments with different protein sequences. In this paper we present a structural analysis of the amino acid sequences produced by the 83 segments. Particular residue patterns in the sequences of V domains imply particular main-chain conformations, canonical structures, for the hypervariable regions. We show that, in almost all cases, the residue patterns in the VH segments imply that the first hypervariable regions have one of three different canonical structures and that the second hypervariable regions have one of five different canonical structures. The different observed combinations of the canonical structures in the first and second regions means that almost all sequences have one of seven main-chain folds. We describe, in outline, structures of the antigen binding site loops produced by nearly all the VH segments. The exact specificity of the loops is produced by (1) sequence differences in their surface residues, particularly at sites near the centre of the combining site, and (2) sequence differences in the hypervariable and framework regions that modulate the relative positions of the loops.

We have used the polymerase chain reaction and VH family-based primers to clone and sequence 74 human germline VH segments from a single individual and built a directory to include all known germline sequences. The directory contains 122 VH segments with different nucleotide sequences, 83 of which have open reading frames. The directory indicates that the structural diversity of the germline repertoire for antigen binding is fixed by about 50 groups of VH segments: each group encodes identical hypervariable loops. The directory should help in mapping the VH locus, in estimating somatic mutation and VH segment usage and in designing and constructing synthetic antibody libraries.