In order to obtain a novel binding protein against a chosen target, DNA molecules, each encoding a protein comprising one of a family of similar potential binding domains and a structural signal calling for the display of the protein on the outer surface of a chosen bacterial cell, bacterial spore or phage (genetic package) are introduced into a genetic package. The protein is expressed and the potential binding domain is displayed on the outer surface of the package. The cells or viruses bearing the binding domains which recognize the target molecule are isolated and amplified. The successful binding domains are then characterized. One or more of these successful binding domains is used as a model for the design of a new family of potential binding domains, and the process is repeated until a novel binding domain having a desired affinity for the target molecule is obtained. In one embodiment, the first family of potential binding domains is related to bovine pancreatic trypsin inhibitor, the genetic package is M13 phage, and the protein includes the outer surface transport signal of the M13 gene III protein.

Directed evolution of novel binding proteins

A member of a specific binding pair (sbp) is identified by expressing DNA encoding a genetically diverse population of such sbp members in recombinant host cells in which the sbp members are displayed in functional form at the surface of a secreted recombinant genetic display package (rgdp) containing DNA encoding the sbp member or a polypeptide component thereof, by virtue of the sbp member or a polypeptide component thereof being expressed as a fusion with a capsid component of the rgdp. The displayed sbps may be selected by affinity with a complementary sbp member, and the DNA recovered from selected rgdps for expression of the selected sbp members. Antibody sbp members may be thus obtained, with the different chains thereof expressed, one fused to the capsid component and the other in free form for association with the fusion partner polypeptide. A phagemid may be used as an expression vector, with said capsid fusion helping to package the phagemid DNA. Using this method libraries of DNA encoding respective chains of such multimeric sbp members may be combined, thereby obtaining a much greater genetic diversity in the sbp members than could easily be obtained by conventional methods.

Methods for producing members of specific binding pairs

Antibody fragments of predetermined binding specificity have recently been constructed from repertoires of antibody V genes, bypassing hybridoma technology and even immunization. The V gene repertoires are harvested from populations of lymphocytes, or assembled in vitro, and cloned for display of associated heavy and light chain variable domains on the surface of filamentous bacteriophage. Rare phage are selected from the repertoire by binding to antigen; soluble antibody fragments are expressed from infected bacteria; and the affinity of binding of selected antibodies is improved by mutation. The process mimics immune selection, and antibodies with many different binding specificities have been isolated from the same phage repertoire. Thus human antibody fragments have been isolated with specificities against both foreign and self antigens, including haptens, carbohydrates, secreted and cell surface proteins, viral coat proteins, and intracellular antigens from the lumen of the endoplasmic reticulum and ...

Making Antibodies by Phage Display Technology

Methods are disclosed for the production of anti-self antibodies and antibody fragments, being antibodies or fragments of a particular species of mammal which bind self-antigens of that species. Methods comprise providing a library of replicable genetic display packages (rgdps), such as filamentous phage, each rgdp displaying at its surface a member of a specific binding pair which is an antibody or antibody fragment, and each rgdp containing nucleic acid sequence derived from a species of mammal. The nucleic acid sequence in each rgdp encodes a polypeptide chain which is a component part of the sbp member displayed at the surface of that rgdp. Anti-self antibody fragments are selected by binding with a self antigen from the said species of mammal. The displayed antibody fragments may be scFv, Fd, Fab or any other fragment which has the capability of binding antigen. Nucleic acid libraries used may be derived from a rearranged V-gene sequences of unimmunised mammal. Synthetic or artificial libraries are described and shown to be useful.

Production of anti-self antibodies from antibody segment repertoires and displayed on phage

Compositions of, genetic constructions coding for, and methods for producing multivalent antigen-binding proteins are described and claimed. The methods include purification of compositions containing both monomeric and multivalent forms of single polypeptide chain molecules, and production of multivalent proteins from purified monomers. Production of multivalent proteins may occur by a concentration-dependent association of monomeric proteins, or by rearrangement of regions involving dissociation followed by reassociation of different regions. Bivalent proteins, including homobivalent and heterobivalent proteins, are made in the present invention. Genetic sequences coding for bivalent single-chain antigen-binding proteins are disclosed. Uses include all those appropriate for monoclonal and polyclonal antibodies and fragments thereof, including use as a bispecific antigen-binding molecule.

Multivalent antigen-binding proteins

A surface expression vector for antibody screening.

To target recombinant antibodies to the surface of Escherichia coli, we have fused single-chain variable domains to its peptidoglycan associated lipoprotein (PAL). The fusion protein was able to bind antigen and was tightly bound to the murein layer of the cell envelope. Antibody-PAL had little effect on cell growth and viability. In contrast, the expression of single chain antibody alone eventually resulted in cell lysis. Immunofluorescence studies on unfixed cells showed that functional antibodies were accessible at the surface of intact bacteria. This could provide a means of isolating single cells producing specific antibodies from libraries in E. coli by fluorescence assisted cell sorting (FACS). Pal fusions may also be of general interest for the presentation of proteins at the surface of E. coli as, for example, in the production of live vaccines.

Targeting recombinant antibodies to the surface of Escherichia coli: fusion to a peptidoglycan associated lipoprotein.

1. Information on the structure and evolution of tubulin has been obtained by comparing the available sequence data on 31 alpha-tubulins and 31 beta-tubulins. 2. Similar numbers of conserved amino acids are found amongst both alpha- and beta-tubulins (alpha: 48%, plus conservative substitutions: 72%; beta: 48%, plus conservative substitutions: 70%). About half of them are common to both subunits (23%, plus conservative substitutions: 45%). Four cysteines in the alpha-tubulins and 2 cysteines in the beta-tubulins are conserved. Only one cysteine (position 129) is conserved in all alpha- and beta-tubulins. 3. The longest unbroken stretch of identical amino acids between all the alpha- and beta-tubulins is found in positions 180-186 (Val-Val-Glu-Pro-Tyr-Asn), a region that appears to be important for binding the ribose moiety of GTP. Two other groups of amino acids implicated in GTP binding, one near position 70 and a glycine cluster at position 144 are also quite conserved. 4. Extra length differences between tubulin subunits, presumably present as extensions on the dimer surface, have been observed at position 50 and near position 360 in alpha-tubulins and in one case at position 57 in a beta-tubulin. 5. The introns of tubulin genes, many of them clustered in the first quarter of the tubulin coding region, do not appear to correspond to any particular structural or functional regions. 6. Mutation rates of tubulins vary considerably. The lowest alpha-tubulin homology (62.3%) is between a very divergent Drosophila alpha-tubulin and an alpha-tubulin from the yeast S. cerevisiae. The lowest beta-tubulin homology (63.3%) is between a yeast (S. cerevisiae) beta-tubulin and a mouse beta-tubulin expressed in hematopoietic tissue. In contrast, some mammalian and bird tubulins are almost identical. 7. Tubulin's heterogeneous C-termini are useful for identifying corresponding tubulins of different vertebrate species, many of which are remarkably conserved. Exceptions are the divergent beta-tubulins of erythrocyte and thrombocyte marginal bands. 8. We have proposed a model for tubulin evolution in metazoan organisms in which the release of structural constraints after gene duplication is a major cause of relatively rapid change.

Thomas Dr Seehaus

Papers

A surface expression vector for antibody screening.

Targeting recombinant antibodies to the surface of Escherichia coli: fusion to a peptidoglycan associated lipoprotein.

Comparative analysis of tubulin sequences

A vector for the removal of deletion mutants from antibody libraries.