Field of the Invention.

The present invention relates to libraries which encode and express antibody variable domains containing modified complementarity determining regions (CDRs) and the use of such libraries for providing novel binding molecules.

Background to the Invention.

A wide range of ligands for G-protein linked receptors contain two cysteine residues separated by between four and ten amino acids. In the case of human endothelin and human chorionic gonadotropin the structure has been solved and the cysteines have been found to define the boundary of a surface oriented hydrophilic loop defining what is designated a cysteine-noose.

These nooses have been implicated by mutagenesis in defining the specificity of the binding of the polypeptides to their respective receptors (Nature Structural Biology 1995,2 266-268). Other potential cysteine-nooses have been identified in gp41 of HIV, bovine serum RNase and endochitinase. Many other receptor ligands have similar loops, even if not necessarily bound by cysteines, for example a neutralising epitope of the cytokine transforming growth factor beta (TGF1) is thought to consist of an eight amino acid loop.

Examples of ligands containing cysteine-nooses include G-protein linked receptor ligands such as endothelins ET1, ET2, ET3 and SRTX-a, human chorionic gonadotropin (hCG), C5a anaphylatoxin, calcitonin, calcitonin-gene related peptides a and b, amylin, vasopressin, oxytocin and somatostatin. The cysteine nooses of these ligands comprise loops of from 4 to 10 amino acids between (and not including) two cysteine residues (i. e. providing nooses of from 6 to 12 amino acids in size).

Tyrosine kinase receptor ligands also have disulphide bridge linked motifs, for example the epidermal growth factor family, all have the structural motif:

CX. CX3sCXl0l2CXCXsGXRC in which Cl pairs with C3, C2 pairs with C4 and C5 pairs with C6.

Cysteine bridged or cyclic peptides have been isolated against a variety of target molecules from constrained peptide libraries displayed on the gene III or gene VIII protein of filamentous bacteriophage (Wright et al, Biotechnology 1995,13,165-169; O'Neil et al, Proteins 1992,14,509-515; McLafferty et al, Gene, 1993,128,29-36; Koivunene et al, J. Biol. Chem., 1993,268 (27) 20205-20210 Livnah et al, Science, 1996,273,464-471; Wrighton et al, Science, 1996,273,458-464). For example, a cysteine bridged peptide with a loop of 8 amino acids was identified as a mimetic of erythropoietin (Wrighton et al, Livnah et al, ibid).

The presence of potential cysteine-nooses in the CDR3 of antibody heavy chains is permitted, as demonstrated by a number of antibodies which have been isolated from phage display antibody libraries. Such antibodies include specificities against carcinoembryonic antigen (CEA) and a human mab (Fog-1), as reported in Griffiths et al. (1993), Human antibodies with high specificity from phage display libraries, EMBO 12,725-734.). A total of 52 examples of two cysteine residues within VH CDR3s have also been identified in naturally occurring antibodies (V-BASE sequence directory, Tomlinson, Williams, Corbett, Cox and Winter (1995) MRC Centre for Protein Engineering, Cambridge UK).

CDR loops from antibodies have been used to design mimetics which bind antigen (Saragovi et al, Biotechnology, 1992,10,773-778; Kieber-Emmons, Current Opinion in Biotechnology, 1997,8,435- 441; Monfardini et al, J. Biol. Chem., 1996,271,2966-2971; Martin et al, EMBO J., 1994,13,5305-5309). Peptide mimetics based on CDR loops have been designed in which disulphide brigdes have been introduced to the peptide after selection to provide a constrained conformation (Saragovi et al, Biotechnology, 1992, 10,773-778).

Disclosure of the Invention.

The present invention provides a repertoire of specific binding members, which binding members comprise a antibody variable domain, wherein said repertoire comprises at least 10'members which each carry a cysteine noose in at least one of their complementarity determining regions (CDR) present in said variable domains. The repertoire may comprise antibody Fab or scFv fragments.

Preferably, the antibody variable domain is a heavy chain variable domain. In a preferred aspect, the cysteine noose comprises from 2 to 8 amino acids between its cysteine residues.

Generally, the cysteine noose is present in the third CDR of the variable domain and where this is the case it is preferred that at least about 30% of said members have a tyrosine residue immediately following the C-terminal cysteine of the noose. A particularly preferred VH third CDR sequence comprises the consensus sequence GGXXC (X) nCXyXGG wherein G is glycine, X is any amino acid, C is cysteine, Xy is tyrosine in at least about 30% of the members, and n is from 3 to 7.

In another aspect, the invention provides a nucleic acid library encoding the repertoire of any one of the preceding claims. The invention also provides a method of selecting a specific binding member capable of binding a target antigen, which method comprises: screening said antigen against a repertoire of specific binding members which binding members comprise an antibody variable domain and wherein said repertoire comprises at least 103 members which each carry a cysteine noose in at least one of their complementarity determining regions (CDR) present in said variable domains; and selecting a specific binding member which is able to bind to the target antigen.

Such nucleic acid may be expressed, optionally after being

manipulated to provide, for example, co-expression of a complementary variable domain.

Methods of then invention may be used to select specific binding members for a range of antigens, including a cancer antigen, a cytokine or cytokine receptor, a growth factor or growth factor receptor, a hormone or hormone receptor, or a viral antigen or viral receptor.

The invention also provides methods of preparing a cysteine noose repertoire of the invention which methods include: a) providing a source of nucleic acid encoding an antibody variable domain; b) mutating said variable domain so as to introduce two codons which encode cysteine within a CDR of said domain, said codons being separated by one or more codons selected at random; and c) recovering the mutated nucleic acid encoding said repertoire.

The invention also provides certain specific binding members per se, which are novel by virtue of their noose size, location or target antigen, including binding members which are specific for MIP-1a receptors.

In a further aspect, the invention also provides a novel method of selecting peptide ligand mimetics capable of binding a target antigen, including the various antigens mentioned herein. In particular, this aspect of the invention provides a method for obtaining a peptide ligand mimetic capable of binding a target antigen which comprises: (a) screening said antigen against a repertoire of specific binding members which binding members comprise an antibody variable domain and wherein said repertoire comprises at least 103 members which each carry a cysteine noose in at least one of their complementarity determining regions (CDR) present in

said variable domains; (b) selecting at least one specific binding member which is able to bind to the target antigen; (c) determining the sequence of the cysteine noose of said selected specific binding member (s); and (d) making at least one peptide mimetic comprising a sequence determined in step (c).

Optionally, the process further includes the step of: (e) screening said peptide mimetic (s) against said antigen.

The process may optionally further include one or more of the following steps, in any order: -preparing further quantities of a peptide mimetic which has been selected by step (d) above; -preparing a pharmaceutical composition comprising said peptide mimetic; -preparing variants of said mimetic which include one or more amino acid substitutions, insertions or deletions in the cysteine loop; -preparing multimers of said mimetic, for example multimers comprising from 2 to 10 copies of said mimetic, for example arranged in tandem repeats; -preparing mimetics wherein the constraint is provided by a link other than a disulphide, for example by a thioether linkage; -preparing variant mimetics which have an altered loop length.

Description of the Drawings.

Figure 1 is a bar chart showing the length of amino acid sequences between two cysteine residues of VH CDR3s identified in naturally occurring antibodies.

Figure 2 shows the locations within the antibodies analysed in Figure 1 in relation to the start and end of CDR3.

Figure 3 shows the frequency of amino acid usage within the VH CDR3s analysed N-terminal and C-terminal to the cysteine noose.

Figure 4 shows the strategy for the construction of a cysteine noose library of the invention.

Figure 5 shows inhibition of MIP-la binding to CD4 cells by specific binding members of the invention.

Figure 6 shows inhibition of binding of MIP-la by a cysteine noose peptide mimetic (ML2CA5) obtained in accordance with the invention (panel a), together with a control (panel b).

Figure 7 shows inhibition of binding of MIP-la by a cysteine noose peptide mimetic (ML4CA11) obtained in accordance with the invention (panel a), together with a control (panel b).

Detailed Description of the Invention.

A. TERMINOLOGY Specific binding member This describes a member of a pair of molecules which have binding specificity for one another. The members of a specific binding pair may be naturally derived or wholly or partially synthetically produced. One member of the pair of molecules has an area on its surface, or a cavity, which specifically binds to and is therefore complementary to a particular spatial and polar organisation of the other member of the pair of molecules. Thus the members of the pair have the property of binding specifically to each other. Examples of types of specific binding pairs are antigen-antibody, biotin-avidin, hormone-hormone receptor, receptor-ligand, enzyme-substrate. This application is concerned with antigen-antibody type reactions.

An tibody This describes an immunoglobulin whether natural or partly or wholly synthetically produced. The term also covers any polypeptide or protein having a binding domain which is, or is homologous to, an antibody binding domain. These can be derived from natural sources, or they may be partly or wholly synthetically produced. Examples of antibodies are the immunoglobulin isotypes and their isotypic subclasses; fragments which comprise an antigen binding domain such as Fab, scFv, Fv, dAb, Fd; and diabodies.

It is possible to take monoclonal and other antibodies and use techniques of recombinant DNA technology to produce other antibodies or chimeric molecules which retain the specificity of the original antibody. Such techniques may involve introducing DNA encoding the immunoglobulin variable region, or the complementarity determining regions (CDRs), of an antibody to the constant regions, or constant regions plus framework regions, of a different immunoglobulin. See, for instance, EP-A-184187, GB 2188638A or EP-A-239400. A hybridoma or other cell producing an antibody may be subject to genetic mutation or other changes, which may or may not alter the binding specificity of antibodies produced.

As antibodies can be modified in a number of ways, the term "antibody"should be construed as covering any specific binding member or substance having a binding domain with the required specificity. Thus, this term covers antibody fragments, derivatives, functional equivalents and homologues of antibodies, including any polypeptide comprising an immunoglobulin binding domain, whether natural or wholly or partially synthetic.

Chimeric molecules comprising an immunoglobulin binding domain, or equivalent, fused to another polypeptide are therefore included. Cloning and expression of chimeric antibodies are described in EP-A-0120694 and EP-A-0125023.

It has been shown that fragments of a whole antibody can perform

the function of binding antigens. Examples of binding fragments are (i) the Fab fragment consisting of VL, VH, CL and CH1 domains; (ii) the Fd fragment consisting of the VH and CH1 domains; (iii) the Fv fragment consisting of the VL and VH domains of a single antibody; (iv) the dAb fragment (Ward, E. S. et al., Nature 341,544-546 (1989)) which consists of a VH domain; (v) isolated CDR regions; (vi) F (ab') 2 fragments, a bivalent fragment comprising two linked Fab fragments (vii) single chain Fv molecules (scFv), wherein a VH domain and a VL domain are linked by a peptide linker which allows the two domains to associate to form an antigen binding site (Bird et al, Science, 242,423-426,1988; Huston et al, PNAS USA, 85,5879- 5883,1988); (viii) bispecific single chain Fv dimers (PCT/US92/09965) and (ix)"diabodies", multivalent or multispecific fragments constructed by gene fusion (W094/13804; P. Holliger et al Proc. Natl. Acad. Sci. USA 90 6444-6448,1993).

Diabodies are multimers of polypeptides, each polypeptide comprising a first domain comprising a binding region of an immunoglobulin light chain and a second domain comprising a binding region of an immunoglobulin heavy chain, the two domains being linked (e. g. by a peptide linker) but unable to associate with each other to form an antigen binding site: antigen binding sites are formed by the association of the first domain of one polypeptide within the multimer with the second domain of another polypeptide within the multimer (W094/13804).

Where bispecific antibodies are to be used, these may be conventional bispecific antibodies, which can be manufactured in a variety of ways (Holliger, P. and Winter G. Current Opinion Biotechnol. 4,446-449 (1993)), eg prepared chemically or from hybrid hybridomas, or may be any of the bispecific antibody fragments mentioned above. It may be preferable to use scFv dimers or diabodies rather than whole antibodies. Diabodies and scFv can be constructed without an Fc region, using only variable domains, potentially reducing the effects of anti-idiotypic reaction. Other forms of bispecific antibodies include the

single chain"Janusins"described in Traunecker et al, EMBO Journal, 10,3655-3659, (1991).

Bispecific diabodies, as opposed to bispecific whole antibodies, may also be particularly useful because they can be readily constructed and expressed in E. coli. Diabodies (and many other polypeptides such as antibody fragments) of appropriate binding specificities can be readily selected using phage display (W094/13804) from libraries. If one arm of the diabody is to be kept constant, for instance, with a specificity directed against antigen X, then a library can be made where the other arm is varied and an antibody of appropriate specificity selected.

Antibody variable domain This refers to the variable domains of antibodies as defined above. Variable domains will comprise three CDRs and sufficient N-and C-terminal residues to form an an antigen binding domain, if need be in association with a complementary variable domain.

Antigen binding domain This describes the part of an antibody which comprises the area which specifically binds to and is complementary to part or all of an antigen. Where an antigen is large, an antibody may only bind to a particular part of the antigen, which part is termed an epitope. An antigen binding domain may be provided by one or more antibody variable domains. Preferably, an antigen binding domain comprises an antibody light chain variable region (VL) and an antibody heavy chain variable region (VH).

Specific This may be used to refer to the situation in which one member of a specific binding pair will not show any significant binding to molecules other than its specific binding partner (s). The term is also applicable where e. g. an antigen binding domain is specific for a particular epitope which is carried by a number of antigens, in which case the specific binding member carrying

the antigen binding domain will be able to bind to the various antigens carrying the epitope.

Neutralisation This refers to the situation in which the binding of a molecule to another molecule results in the abrogation or inhibition of the biological effector function of the another molecule.

Comprise This is generally used in the sense of include, that is to say permitting the presence of one or more features or components.

Isola tex This refers to the state in which specific binding members of the invention, or nucleic acid encoding such binding members will be, in accordance with the present invention. Members and nucleic acid will be free or substantially free of material with which they are naturally associated such as other polypeptides or nucleic acids with which they are found in their natural environment, or the environment in which they are prepared (e. g. cell culture) when such preparation is by recombinant DNA technology practised in vitro or in vivo. Members and nucleic acid may be formulated with diluents or adjuvants and still for practical purposes be isolated-for example the members will normally be mixed with gelatin or other carriers if used to coat microtitre plates for use in immunoassays, or will be mixed with pharmaceutically acceptable carriers or diluents when used in diagnosis or therapy. Specific binding members may be glycosylated, either naturally or by systems of heterologous eukaryotic cells, or they may be (for example if produced by expression in a prokaryotic cell) unglycosylated.

Repertoire A repertoire is a collection of either nucleic acid sequences encoding specific binding members or specific binding members themselves which sequences or specific binding members share

common structural features, which features serve to carry a collection of varying sequences. In the context of the present invention the common structural features will be antibody framework regions which will be broadly similar in length and sequence to each other within each member of the repertoire. The varying structural features will generally be one or more of the CDRs carried by the antibody framework regions. Generally, a repertoire will comprise at least 103, preferably at least 105,<BR> for example at least 10 6 or 10 7 members. The maximum size of a repertoire is generally governed only by the technology available to make them but it is certainly possible to obtain repertoires of up to 1014, for example up to 1012 individual members.

In the present invention a repertoire of specific binding members is provided in which at least 103 members comprise a cysteine noose in at least one of their CDRs. It is not essential for all members of the repertoire to comprise a cysteine noose. For example, the repertoire could comprise 104 members of which l00 each carried the cysteine noose. However it is preferred that a substantial portion of the members, preferably at least 90% and more preferably at least 95%, 98% or 99% of the members carry a cysteine noose.

The repertoire may comprise a single antibody chain or two antibody chains, for example a light chain and a heavy chain. In the case of two antibody chains, these chains may be linked in-frame so as to provide for expression of single chain Fv molecules.

It is known that the antibody heavy chain variable domain is particularly important in providing antigen binding properties of an antibody and therefore it is preferred that the members of the repertoire carry a cysteine noose in a heavy chain CDR. However this is not to exclude the presence of cysteine nooses in light chain CDRs.

"Cysteine Noose"and"Cysteine Loop" As used herein, a cysteine noose refers to a sequence of amino acids which include an N-terminal and C-terminal cysteine residue, whereas a cysteine loop, or"loop", refers to a sequence of amino acids bounded by, but not including, the terminal cysteines. Thus for any given loop size of X amino acids the noose size will be X+2 amino acids.

B. DETAILED DESCRIPTION The cysteine noose may be present in any CDR of an antibody heavy or light chain. The third CDR is preferred and particularly preferred is the third CDR of the heavy chain variable domain.

Cysteine nooses may be of any size subject only to the maximum size of the CDR. CDRs may be defined by reference to Kabat et al, Sequences of Proteins of Immunological Interest, Fourth Edition, US Department of Health and Human Services, 1987, and updates thereof, now available on the Internet (http://immuno. bme. nwu. edu). The maximum and minimum sizes (number of amino acids) of CDRs may therefore be as follows: VL VH CDR1 11-17 5-7 CDR2 6-8 12-19 CDR3 4-16 4-30 CDR2 of the VL domain is generally 7 amino acids in size.

Cysteine nooses may therefore be present in any of the above CDRs ranging from a loop size (ie length of sequence between and not including the two cysteine residues) of from 1 to n-2 where n is the maximum size of the CDR as set out above.

The location of the cysteine noose within a CDR may be varied. In a preferred embodiment the N-terminal cysteine of the noose

will be the second, third, fourth or fifth residue of the CDR.

The C-terminal cysteine will desirably be at least one to five, for example preferably 2,3 or 4 residues away from the end of the CDR. Thus, in designing cysteine nooses according to the present invention the skilled person may either select N-and C- terminal locations for cysteines by reference to the start and finish points of a particular CDR, or by reference to only one point and a selected loop size, or a combination of both.

Cysteine noose repertoires of the invention may comprise members with uniform cysteine loop sizes or ranges of cysteine loop sizes either within a single or a mixture of CDRs. Loops sizes may be selected subject to the maximum CDR sizes and may therefore range from 1 to 28, preferably from 3 to 15, for example from 2 to 8 such as 3,4,5,6 or 7 residues in size, subject to the total noose size being no greater than that of the CDR in which it is located.

In a particularly preferred aspect of the invention the cysteine noose is present in the third CDR of a heavy chain variable domain. In one embodiment, a repertoire of such domains may be prepared which is biased to favour the presence of a tyrosine residue immediately adjacent and following the C-terminal residue of the cysteine noose. In this embodiment of the invention it is preferred at least about 30% of the members comprise tyrosine at this position, preferably at least about 50%. Although 100% of the member may comprise tyrosine at this position, desirably at least about from 20% to 70% of the members have another residue.

Another feature which may be incorporated into VH CDR3 libraries is the presence of a pair of glycine residues at the N-and/or C-termini of the CDR3, and preferably both termini. These glycines may be present in the repertoire in conjunction with or as an alternative to the tyrosine bias mentioned above.

Repertoires according to the invention may comprise a diverse

range of binding members which are capable of binding a diverse range of target antigens. Such libraries may be used for screening against any desired antigen. Alternatively, a repertoire of the invention may comprise a preselected subgroup of binding members with a non-random bias towards a particular target antigen. For example, this may be brought about by providing a repertoire in which one or more of the CDRs are of a defined sequence and only one CDR is varied between members of the repertoire. Alternatively, the repertoire could be derived from specific binding members with a particular ligand noose and the members of the repertoire would be varied in the noose sequence.

Repertoires according to the invention may be used to screen and select specific binding members capable of binding in a specific manner to a target antigen. Once a specific binding member has been isolated from the repertoire the nucleic acid encoding the specific binding member may be obtained and used for expression of the nucleic acid to obtain further copies of the selected specific binding member. Nucleic acid from the specific binding member may be manipulated in any manner known per se in the art.

For example, the variable domain of the specific binding member may be isolated and linked to additional antibody sequences such as sequences encoding all or part of a light or heavy chain constant region. The sequences may be linked to a complementary variable domain so that on expression an antibody two chain variable region is obtained. Manipulation also includes methods in which the CDRs of the selected antibody are combined with other framework regions so as to produce a reshaped antibody.

This may be accomplished by CDR grafting as shown in for example EP-A-239400 or framework grafting as shown in EP-B-549581.

Specific binding members may be selected by any suitable system.

The preparation of repertoires of specific binding members and the use of the repertoires for selecting antigens is as such well known in the art and this underlying technology is not part of the present invention as such. Reference may be made to for

example, McCafferty et al, 1990, Nature 348: 552-4, Winter et al, 1994, Annu. Rev. Immunol. 12: 433-455, Marks et al, 1991, J. Mol. Biol 222: 581-597 and Vaughan et al, 1196, Nature Biotechnology 14: 309- 314.

The above references also disclose in particular repertoires prepared in phage or phagemid libraries, and such type of libraries may conveniently be used in the present invention. The above references may also be used by way of guidance to those of skill in the art in methods of preparing repertoires according to the present invention. The preparation of repertoires according to the present invention will differ primarily by virtue of the use of techniques specifically designed to introduce cysteine nooses into a CDR of a variable domain. A suitable and preferred technique is the use of oligonucleotide mutagenesis in which oligonucleotides are used to amplify nucleic acid encoding all or part of an antibody variable domain wherein the oligonucleotide comprises a CDR replacement sequence comprising two codons for cysteine separated by one or more codons selected at random. By"one or more codons selected at random"it is meant that of the codons separating the two cysteine residues at least one but not necessarily all are selected at random such that a repertoire of specific binding members according to the invention is obtained. One or more of the codons flanking the N-terminal and/or C-terminal may also be selected at random.

Random selection also includes partially random selection, for example to bias a codon in favour of a particular residue or group of residues. A cysteine noose replacement primer will additionally comprise sufficient sequence flanking either or both of the codons encoding the cysteine noose such that the oligonucleotide may be used as a primer on a source of nucleic acid sequences in which members of the source comprise target sequences substantially homologous to said flanking regions of the oligonucleotide. This allows the preparation of a repertoire according to the invention by schemes such as that illustrated

in Figure 4 or methods analogous thereto.

The source nucleic acid may comprise any suitable source from which binding members of the invention can be made through methods such as that described herein. Suitable sources include existing repertoires of specific binding members or primary sources of nucleic acid such as spleen cell mRNA.

In another aspect of the invention the source of nucleic acid may be a selected group of specific binding members which are specific for a target antigen. Such a group may for example be a group in which no cysteine nooses are present and the method of the invention is carried out in order to introduce a cysteine noose in a defined CDR. The source nucleic acid may also be one which encodes a single specific binding member including a specific binding member comprising a cysteine noose. Such a source may be used for introducing additional cysteine nooses into the nucleic acid or to introduce variations in an existing cysteine noose.

The methods of preparing repertoires of specific binding members of the invention and methods of screening such repertoires may be used to select specific binding members for a variety of antigens. Specific binding members may be selected to bind to a target to antagonise or agonise the function of that target.

Such target antigens include cytokines, growth factors, hormones and viral antigens, or any of their cognate receptors, or a cancer antigen. A cancer antigen is an antigen expressed by tumour cells which is otherwise normally absent or expressed only at a low level in or on the surface of cells normally present in the human body.

Cytokines include MIP-1a, MIP-1 (3, MIP-2, RANTES, tumour necrosis <BR> <BR> <BR> factor (eg TNFa), interferons a, ß or y, interleukins (including any of IL-1 to IL-18, such as IL-2, IL-6 and IL-12), MCP-1,2 and 3, macrophage inhibitory factor (MIF), Erythropoietin (Epo) and

Tpo. Growth factors include TGFa, TGF, CSF, GCSF, PDGF, FGF, EGR, VEGF, BDNF and SCF. Viral antigens include HIV antigens such as gpl60/120, HBV surface or core antigens, HAV antigens, HCV antigens, HPV (eg HPV16) antigens, HSV-1 or-2 antigens, Epstein Barr virus (EBV) antigens, Kapos's sarcoma virus antigens, neurotropic virus antigens, adenovirus antigens, cytomegalovirus antigens, polio myelitis virus antigens. Hormones include steriod hormones such as estradiol, hCG, endothelins such as ET1 ET2 ET3 and SRTX-a, thyroid stimulating hormone, growth hormone, adrenocorticotropin, follicle stimulating hormone, prolactin, luteinizing hormone, insulin, glucagon, amylin, calcitonin CGRP a and b, oxytocin and somatostatin.

Reference to the above cytokines, growth factors, viral antigens and hormones is also intended as reference to their cognate receptors. Indeed, the cysteine noose specific binding members of the invention which bind specifically to cell surface receptors including those in the categories mentioned above form another important and preferred aspect of the invention in view of the presence of cysteine nooses in the ligands which bind some of these receptors.

Cancer antigens include CEA, alpha fetal protein (AFP), neu/HER2, polymorphic endothelia mucin (PEM), N-CAM and Lewis Y.

Some enzymes, for example bovine serum Rnase and endochitinase, and viral coat proteins, for example gp41 of HIV, also possess cysteine-nooses. The presence of cysteine-nooses in molecules such as these suggests a potential for generating antibodies with surface loops which mimic these which could block viral infection events, or inhibit enzymic reactions.

In a particular aspect, the invention provides a cysteine noose specific binding member, in particular a VH domain comprising a cysteine noose in its third CDR, and optionally in association with a cognate VL domain, capable of binding in a specific manner to MIP-la receptor. The binding member may be in the form of an

antibody fragment, for example an scFv or Fab fragment. The invention also provides compositions of such specific binding members, particularly pharmaceutical compositions as defined below. Such compositions may be of use in therapies of the human or animal body, for treating or alleviating conditions which benefit from anti-MIP-la receptor therapy. Such conditions include HIV infection. Such compositions may also be of use in vitro for investigating mechanisms of infection of HIV on CD4+ cells to which such specific binding members bind, or for investigating potential HIV therapeutics by determining the degree to which such therapeutics modulate (e. g. agonise or antagonise) the binding of such specific binding members to CD4+ cells.

Although the prior art indicates a number of antibodies which comprise a cysteine noose in one CDR are available in the art, the present invention provides for the first time a repertoire of such antibodies and the ability to select antibodies against a whole range of antigens. Accordingly, in a further aspect of the invention individual specific binding members are provided apart from those already in the art. Thus specific binding members of the invention include those against the target antigens mentioned above. Specific binding members may be provided in isolated form.

Specific binding members obtained or obtainable from the repertoires or by the methods of the present invention may be labelled with a detectable or functional label. Detectable labels include radiolabels such as 13lI or 99Tc, which may be attached to antibodies of the invention using conventional chemistry known in the art of antibody imaging. Labels also include enzyme labels such as horseradish peroxidase. Labels further include chemical moieties such as biotin which may be detected via binding to a specific cognate detectable moiety, e. g. labelled avidin.

Functional labels include substances which are designed to be

targeted to the site of a tumour to cause destruction of tumour tissue. Such functional labels include toxins such as ricin and enzymes such as bacterial carboxypeptidase or nitroreductase, which are capable of converting prodrugs into active drugs at the site of a tumour. Binding members of the invention comprising such labels may be used in methods of diagnosis and treatment of tumours in human or animal subjects, particularly solid tumours which have a necrotic centre. These tumours may be primary or secondary solid tumours of any type including, but not limited to, cervical, ovarian, prostate, lung, liver, pancreatic, colon and stomach tumours.

Specific binding members obtained or obtainable from the repertoires or by the methods of the present invention when designed for therapeutic use may be administered to a patient in need of treatment via any suitable route, for example usually by injection into the bloodstream or directly into the a particular location such as the site of the tumour. F (abc), antibody fragments may be used for both tumour imaging and tumour treatment.

Specific binding members obtained or obtainable by the present invention will usually be prepared for administration in the form of a pharmaceutical composition, which may comprise at least one component in addition to the specific binding member.

Thus pharmaceutical compositions according to the present invention, and for use in accordance with the present invention, may comprise, in addition to active ingredient, a pharmaceutically acceptable excipient, carrier, buffer, stabiliser or other materials well known to those skilled in the art. Such materials should be non-toxic and should not interfere with the efficacy of the active ingredient. The precise nature of the carrier or other material will depend on the route of administration, which may be oral, or by injection, e. g. intravenous.

Pharmaceutical compositions for oral administration may be in tablet, capsule, powder or liquid form. A tablet may comprise a solid carrier such as gelatin or an adjuvant. Liquid pharmaceutical compositions generally comprise a liquid carrier such as water, petroleum, animal or vegetable oils, mineral oil or synthetic oil. Physiological saline solution, dextrose or other saccharide solution or glycols such as ethylene glycol, propylene glycol or polyethylene glycol may be included.

For intravenous, injection, or injection at the site of affliction, the active ingredient will be in the form of a parenterally acceptable aqueous solution which is pyrogen-free and has suitable pH, isotonicity and stability. Those of relevant skill in the art are well able to prepare suitable solutions using, for example, isotonic vehicles such as Sodium Chloride Injection, Ringer's Injection, Lactated Ringer's Injection. Preservatives, stabilisers, buffers, antioxidants and/or other additives may be included, as required.

A composition may be administered alone or in combination with other treatments, either simultaneously or sequentially dependent upon the condition to be treated. Other treatments may include the administration of suitable doses of pain relief drugs such as non-steroidal anti-inflamatory drugs (e. g. asprin, paracetamol, ibuprofen or ketoprofen) or opitates such as morphine, or anti-emetics.

The present invention further provides an isolated nucleic acid encoding a specific binding member of the present invention.

Nucleic acid includes DNA and RNA.

The present invention also provides constructs in the form of plasmids, vectors, transcription or expression cassettes which comprise isolated nucleic acid of the invention encoding a single specific binding member.

The present invention also provides a recombinant host cell which

comprises one or more constructs as above. A nucleic acid encoding any specific binding member as provided itself forms an aspect of the present invention, as does a method of production of the specific binding member which method comprises expression from encoding nucleic acid therefor. Expression may conveniently be achieved by culturing under appropriate conditions recombinant host cells containing the nucleic acid. Following production by expression a specific binding member may be isolated and/or purified using any suitable technique, then used as appropriate.

The isolation and purification of the cysteine noose antibodies will be carried out under conditions in which the cysteine bonds of the CDR will form disulphide bridges. These conditions are normally found when the protein is exported from cells. For example in E. coli, the periplasm is an oxidising environment which promotes disulphide bridge formation. Antibodies and their fragments, such as ScFvs contain a number of internal disulphide bridges encoded within the framework regions, and these are formed when scFv are expressed in cells such as by bacteria.

Disulphide bridges formed within the noose will be more exposed to the oxidising environment than internal disulphide bridges, since they are present on the surface of the specific binding partner.

Specific binding members and encoding nucleic acid molecules and vectors according to the present invention may be provided isolated and/or purified, e. g. from their natural environment, in substantially pure or homogeneous form, or, in the case of nucleic acid, free or substantially free of nucleic acid or genes origin other than the sequence encoding a polypeptide with the required function. Nucleic acid according to the present invention may comprise DNA or RNA and may be wholly or partially synthetic.

Systems for cloning and expression of a polypeptide in a variety of different host cells are well known. Suitable host cells

include bacteria, mammalian cells, yeast and baculovirus systems.

Mammalian cell lines available in the art for expression of a heterologous polypeptide include Chinese hamster ovary cells, HeLa cells, baby hamster kidney cells and many others. A common, preferred bacterial host is E. coli.

The expression of antibodies and antibody fragments in prokaryotic cells such as E. coli is well established in the art.

For a review, see for example Pluckthun, A. Bio/Technology 9: 545-551 (1991). Expression in eukaryotic cells in culture is also available to those skilled in the art as an option for production of a specific binding member, see for recent reviews, for example Reff, M. E. (1993) Curr. Opinion Biotech. 4: 573-576; Trill J. J. et al. (1995) Curr. Opinion Biotech 6: 553-560.

Suitable vectors can be chosen or constructed, containing appropriate regulatory sequences, including promoter sequences, terminator sequences, polyadenylation sequences, enhancer sequences, marker genes and other sequences as appropriate.

Vectors may be plasmids, viral e. g. 'phage, or phagemid, as appropriate. For further details see, for example, Molecular Cloning: a Laboratory Manual: 2nd edition, Sambrook et al., 1989, Cold Spring Harbor Laboratory Press. Many known techniques and protocols for manipulation of nucleic acid, for example in preparation of nucleic acid constructs, mutagenesis, sequencing, introduction of DNA into cells and gene expression, and analysis of proteins, are described in detail in Short Protocols in Molecular Biology, Second Edition, Ausubel et al. eds., John Wiley & Sons, 1992. The disclosures of Sambrook et al. and Ausubel et al. are incorporated herein by reference.

Thus, a further aspect of the present invention provides a host cell containing nucleic acid as disclosed herein. A still further aspect provides a method comprising introducing such nucleic acid into a host cell. The introduction may employ any

available technique. For eukaryotic cells, suitable techniques may include calcium phosphate transfection, DEAE-Dextran, electroporation, liposome-mediated transfection and transduction using retrovirus or other virus, e. g. vaccinia or, for insect cells, baculovirus. For bacterial cells, suitable techniques may include calcium chloride transformation, electroporation and transfection using bacteriophage.

The introduction may be followed by causing or allowing expression from the nucleic acid, e. g. by culturing host cells under conditions for expression of the gene.

In one embodiment, the nucleic acid of the invention is integrated into the genome (e. g. chromosome) of the host cell.

Integration may be promoted by inclusion of sequences which promote recombination with the genome, in accordance with standard techniques.

The present invention also provides a method which comprises using a construct as stated above in an expression system in order to express a specific binding member or polypeptide as above.

Where the present invention is used to provide the abovementioned novel method of obtaining peptide ligand mimetics, the first step to obtain such mimetics is the same as the first step of obtaining specific binding members of the invention, namely screening a target antigen against a library of the invention.

One or more of the selected binding members which bind the target will then be selected, and the sequence of the cysteine noose determined. Most conveniently, the sequence will be determined by sequencing the nucleic acid encoding the specific binding member, although other methods may be used if required. Once the sequence of the cysteine noose is determined, a peptide mimetic comprising said sequence may be made. Conveniently, this may be done by solid phase peptide synthesis although recombinant

production may also be used, utilising methods known as such in the art.

The peptide mimetic will consist of the cysteine noose, and optionally include at its N-and/or C-terminii, additional amino acid residues. Desirably, the total size of the mimetic will be comparable to a the size of a CDR, i. e. from 5 to 30 amino acids in size. More preferably, the total size of the mimetic will be from 8 to 20 amino acids in length. The location of the N- terminal cysteine is preferably at position 2 to 5 of the mimetic, where 1 is the N-terminal residue. The location of the C-terminal cysteine is usally at X-1 to X-4, where X is the C- terminal residue.

The peptide mimetics of Example 8 below have glycine a their N- terminal and C-terminal positions. This is because the mimetics are derived from a library where the noose CDR was designed to contain this feature. Those of skill in the art will appreciate that this is not essential and that other peptide mimetics may be made, based on the production of other cysteine noose specific binding members.

Peptide mimetics selected in accordance with the invention may be screened against the target antigen for ligand binding activity.

While not wishing to be bound by any one theory, it is believed that the present method of selecting peptide ligand mimetics, particularly when selected from CDR3 cysteine noose libraries (more particularly VH cysteine noose libraries), provide a means to select a different and potentially more effective population of peptide ligands than direct display of similar cysteine noose ligands on the surface of bacteriophage. The CDR3 of an antibody chain projects above the surface of the bulk of the antibody molecule and thus may project the cysteine noose more effectively than when such a noose is directly linked to a surface component of a bacteriophage.

Once a peptide mimetic has been identifed, futher quantities of it may be made for any desired application. Applications incude formulation of the mimetic into a pharmaceutial composition comprising the mimetic and one or more suitable carriers or diluents, such as those discussed above. The mimetic may be used in monomeric form or covalently linked to other compounds, for example compounds such as biologically compatible polymers, for example polyethylene glycol based molecules designed to enhance stability of the molecules. The peptides may also be polymerised to provide multiple copies of the mimetic in a single molecule, for example in the form of tandem repeats.

The mimetic may also be used as a basis for preparing further mimetics, for example by preparing variants which include an altered loop length, or one or more amino acid substitutions.

Deletions and insertions in the cysteine loop will alter the loop length. Such alterations may be made in any part of the loop.

With insertions, these are desirably made at either the N- terminus or C-terminus of the loops, adjacent to the cysteine residues. These may be made in an iteratitve process, e. g. by inserting a range of amino acids at a first position, selecting a mimetic variant with improved properties (e. g. binding affinity or biological stability) compared to the starting variant, and then repeating this process with said variant one or more times.

With deletions, one or more deletions may be made in the loop. Again, it may be desirable to make single deletions, select a deleted variant, and then repeat the process to obtain variants with one or more deletions.

The number of such substitutions, deletions or insertions will be constrained by the size of the loop, although generally from 1 to 4, such as 1 or 2 may be made. Such altered mimetics may then be tested for binding against the target antigen, and any with a desired binding activity may be selected, manufactured and used as described above.

Mimetic variants which utilize an alternative loop constraint may also be made. Various means of constraint are known as such in the art, for example the use of a beta-turn loop (Saragovi et al, Science, 1991,253,792-795) or other means (e. g. see Kieber- Emmons et al, ibid) such as a cyclic lactam.

The invention is illustrated by the following non-limiting examples. For ease of reference, the examples below are set out in the following manner: EXAMPLE 1: Design of a cysteine-noose scFv repertoire.

EXAMPLE 2: Construction of a cysteine-noose scFv repertoire.

EXAMPLE 3: Selection of the cysteine-noose repertoire on cell membranes.

EXAMPLE 4: Selection of the cysteine-noose repertoire on whole cells.

EXAMPLE 5: Selection of ligand inhibitors from the cysteine-noose repertoire.

EXAMPLE 6: Demonstration of the importance of the cysteine noose in antibody binding to antigen.

EXAMPLE 7: Construction of further cysteine noose repertoires with loop sizes of 4 to 10 amino acids.

EXAMPLE 8: Generation of peptide mimetics from a cysteine noose library.

EXAMPLE 1: DESIGN OF A CYSTEINE-NOOSE SCFV REPERTOIRE The human antibody sequence database (V-BASE sequence directory, Tomlinson, Williams, Corbett, Cox and Winter (1995) MRC Centre for Protein Engineering, Cambridge UK) was surveyed for the

presence of naturally occurring antibodies which contain two cysteines within the CDRs. No such motifs were found in any light chain CDRs, or in CDR1 and CDR2 of heavy chains. 52 examples of the occurrence of two cysteines were found in heavy chain CDR3's from 1200 surveyed. The number of amino acid residues present between the two cysteines in each case was found and the majority of the antibodies contained four residues between the two cysteines. The distribution of the number of residues is shown in Figure 1.

The positioning of the C (X) nC motif within the CDR3 was also assessed. The number of residues between the end of the framework 3 region of the heavy chain gene segment and the first cysteine was most commonly 3 or 4, (Figure 2a). The number of residues between the second cysteine residue and the start of the JH region of the heavy chain gene segment ranged from 0 to 9, and was most commonly between 2 or 4 residues (Figure 2b).

Codon usage in the regions of the CDR3 flanking the cysteine-noose motif was also assessed (Figures 3a and 3b). Tyrosine was found to be the residue most commonly observed immediately after the second cysteine. Glycine was commonly found at other flanking positions.

Given this information an initial consensus cysteine-noose motif was established as : FR3 GGXX C (X) C XyXGG JH (SEQ ID NO: 1) Where X is any amino acid and Xy is any amino acid, but with a bias designed to favour the presence of a tyrosine residue. FR3 represents the end of the framework three region of the heavy chain gene segment, and JH represents the start of the JH region of the heavy chain gene segment.

EXAMPLE 2: CONSTRUCTION OF A CYSTEINE-NOOSE SCFV LIBRARY

The cysteine-noose library construction was designed using an exisiting scFv antibody repertoire as a starting template. This template was then PCR amplified to give a VL gene segment repertoire and a VH gene segment repertoire. The VH gene segment repertoire was amplified using a primer which was designed to incoporate a cysteine-noose into the VHCDR3. The two repertoires were then PCR assembled to give full length scFv gene segments which were cloned into a phagemid vector. The PCR strategy is outlined in Figure 4. a) Preparation of DNA template for library amplification.

The starting material for construction of the cysteine-noose scFv library was DNA prepared from the large scFv antibody library as described in Vaughan et al., Nature Biotechnology 1996,14, 309-314 (Human Antibodies with sub-nanomolar affinities isolated from a large non-immunised phage display library). Cells from 200 yl of a bacterial glycerol stock of this library were pelleted at 10 000 x g for 2 minutes in a microcenrifuge. Miniprep DNA was then prepared from these cells using a Promega Wizard Miniprep kit. Cells were resuspended in 200 yl resuspension buffer (50mM Tris, (pH lOmM EDTA; 100 yg/ml RNase A), then lysed with 200 il lysis buffer (0.2M NaOH with lo SDS) for 10 minutes at room temperature. 200 Al neutralisation buffer (1.32M potassium acetate) was then added and the tube inverted 4 times. The lysate was then centrifuged at 10 000 x g in a microcentrifuge for 5 minutes and the supernatant collected. 1 ml of Wizard DNA purification resin was added to the supernatant and the resin was then transferred to a Minicolumn/Syringe assembly. A vacuum was applied to pull the resin/lysate mix into the minicolumn and 2 ml of Column Wash Solution (80mM potassium acetate, 8.3mM Tris-HC1, pH EDTA in 55% ethanol) was then added and drawn through the vacuum. The resin was dried under vacuum for 30 seconds and the minicolumn was then removed from the vacuum apparatus. The minicolumn was centrifuged at 10 000 x g in a microcentrifuge for 2 minutes to remove any residual Column Wash Solution. 50 yl of water was then added to the minicolumn and the minicolumn was centrifuged at 10 000 x g to elute the DNA. DNA

was stored at-20°C. b) Generation of repertoires of VL gene segments and VH qene segments which have been mutated to incoporate the cysteine-noose motif.

The primer VH Cys loop For was designed to mutate the VH CDR3 of the minipreped DNA made from the large scFv library to the following motif (mutated region is respresented in bold face): Framework 3 VH CDR3 JH TAVYYCAR GGXXC (X) 4CXyxGG WGQGTL (SEQ ID NO: 3) FR 3: T A V Y Y C A R amino acid ACG GC (CT) GTG TAT TAC TGT GC (GA) AGA nucleotide LOOP: <BR> <BR> <BR> <BR> GGXXCX. I c Xy x G G amino acid<BR> <BR> <BR> <BR> <BR> <BR> GGC GGG NNN NNN TGC (NNN) 4 TGT NTNAN NNN GGA GGT nucleotide JH: w G Q G T L amino acid (SEQ ID NO: 3) TGG GGC CA (AG) GG (AG) ACC CTG nucleotide (SEQ ID NO: 2) Oligonucleotide VH Cys loop For (reverse complement): <BR> <BR> <BR> <BR> <BR> <BR> <BR> 5'CAG GGT (CT) CC (CT) TG GCC CCA ACC TCC NNN NNTNA ACA (NNN) 4 GCA NNN NNN CCC GCC TCT (TC) GC ACA GTA ATA CAC (AG) GC CGT 3' (SEQ ID NO: 4) This primer was used in conjuction with the primer LMB3 to amplify a repertoire of heavy chains which were mutated in the VH CDR3.

LMB3

5'CAGGAAACAGCTATGAC 3' (SEQ ID NO: 5) A repertoire of VL chains was generated using the primer JH Cys Back in conjuction with the primer Fdtseq tag. The JH Cys Back primer has been designed to prime on the VH JH region with enough homology to VH Cys loop For to allow PCR assembly: JH: W G Q G T L V (SEQ ID NO: 7) TGG GGC CA (AG) GG (AG) ACC CTG GTC (SEQ ID NO: 6) Oligonucleotide JH Cys Back (as coding squence): 5'TGGGGCCA (AG) GG (AG) ACCCTGGTC 3' (SEQ ID NO: 8) Fdtseq tag 5'ATTCGTCCTATACCGTTCTACTTTGTCGTCTTTCCAGACGTTAGT 3' (SEQ ID NO: 9) PCR's to generate the VH and VL repertoires were set up in triplicate as follows: Template (miniprep's scFv library DNA) 1 ßl TAQ polymerase (Boehringer) 0.5y1 10 X PCR buffer (Boehringer) 5 ßl dNTP's (5 mM) (Pharmacia) 4/il Primer 1 (10 mM) 2.5 fil Primer 2 (10 mM) 2.5 yl dH20 34.5 yl PCR conditions were as follows: 94°C 2 min, (94°C 1 min, 60°C 1 min, 72°C 2 min) x 25,72°C 10 min.

The VH and VL repertoires generated by PCR were electrophoresed on a 1.5% low melting point agarose gel. Bands corresponding to the VH and VL gene segments were excised from the gel and the DNA extracted from the bands using Promega Wizard DNA purification columns, exactly as described in the manufacturer's instructions.

DNA was eluted from the column in 50 Ul of dH, O. c) PCR pull through reactions.

Approximately 50 ng of amplified CDR3 mutated heavy chain repertoire and 50 ng of the amplified light chain repertoire were combined. This was used in an assembly amplification after the addition of reaction buffer to 1X, dNTP's to 200nM and 5 units Taq polymerase. Amplifications consisted of 7 cycles of 94°C for 1 min, 65°C for 4 min. 5y1 of each assembly was used as the template in a'pull-through'amplification using the primers FTAG and LMB3. Amplification conditions consisted of 25 cycles of 94°C for 1 min, 55°C for 2 min and 72°C for 1 min, followed by 10 min at 72°C.

FTAG 5'TTTGTCGTCTTTCCAGACGTTAGT 3'SEQ ID NO: 10 The pull-through amplification product was separated through 1.5% agarose-TAE. A PCR product of the expected size (1.1 Kb) was visualised and excised from the gel. The gel fragment was then melted by incubation at 70°C for 15 min and the DNA purified from the gel using a Promega Magic PCR DNA purification column. lml of resin was added to the melted gel, mixed by inversion and then passed down a DNA minicolumn. The column was washed with 2ml of 80% isopropanol and spun at top speed in a minifuge for 20 seconds to dry the column. DNA was then eluted from the column in 50 yl dH2O and the DNA recovered by centrifugation in a minifuge at top speed for 20 seconds.

The purified DNA fragment was digested with the restriction endonucleases Sfi I and Not 1 (NEB) then ligated (Amersham ligation system) into Not I/Sfi I digested phagemid vector pCANTAB6 (McCafferty, J. et al., 1994. Selection and rapid purification of murine antibody fragments that bind a transition-state analog by phage display. Applied Biochemistry and Biotechnology, 47: 157-173). The ligation product was used to transform electrocompetent TG1 cells, plated out on 2YTAG (2YT

media supplemented with 100/2gel ampicillin and 2t glucose) plates and incubated overnight at 30°C. Approximately 7.2 x 108 individual clones were generated. Plates were scraped into a total of 9ml of 2TY containing 15 glycerol and 1.5 ml aliquots were made and stored at-70°C.

EXAMPLE 3. SELECTION OF THE CYSTEINE-NOOSE REPERTOIRE ON CELL MEMBRANES. a) Induction of the phage antibody library.

The cysteine-noose repertoire was selected to isolate antibodies which bind cell membranes of a specific cell type. Phagemid particles were recovered from the repertoire as follows. 500 ml prewarmed (37°C) 2TYAG in a 21 concical flask was inoculated with approximately 3 x 1010 cells from a glycerol stock culture of the library. The culture was grown at 37°C with good aeration until the OD 600nm reached 0.7. M13K07 helper phage (Stratagene) was added to the culture to a multiplicity of infection (moi) of approximately 10 (assuming that an OD 600nm of 1 is equivalent to 5 x 108 cells per ml of culture). The culture was incubated stationary at 37°C for 15 minutes followed by 45 minutes with light aeration (200rpm) at the same temperature. The culture was centrifuged and the supernatant drained from the cell pellet. The cells were resuspended in 500 ml 2YTAK (2YT media supplemented with 100 yg/ml amplicillin and 50 Ug/ml kanamycin, and the culture incubated overnight at 30°C with good aeration (300 rpm).

Phage particles were purified and concentrated by three polyethylene glycol (PEG) precipitations (Sambrook, J., Fritsch, E. F., & Maniatis, T. (1990). Molecular Cloning-A Laboratory Manual. Cold Spring Harbour, New York) and resuspended in PBS to 1013 transducing units (tu)/ml. b) Selection from the phage antibody library.

Phage induced from the cysteine-noose antibody repertoire were initially selected on intact cells. 100 ill of PBS containing 106 cells prepared from the breast of a 42 year old female were added to 100 ml of phage (2 x 1012) prepared from the

cysteine-noose library which had been premixed with 800 Al of 3k marvel in PBS (MPBS). Phage and cells were incubated at 37°C for 1 hour with occasional mixing. This cell type floats to the top of the liquid allowing unbound phage to be removed from under the cells with a pipette. Cells were washed three times with 1 ml of tissue culture grade PBS by mixing the cells with the PBS and then removing the PBS from under the cell layer with a pipette. The washed cells were then added directly to 5 ml of an exponentially growing culture of E coli TG1 in 2TY and incubated stationary at 37°C for 15 minutes, followed by 45 minutes with light aeration (200rpm). Infected cells were then plated onto 2TYAG medium in 243mm x 243mm dishes (Nunc) and plates were incubated overnight at 30°C. Colonies were scraped off the plates into 10 ml of 2TY broth and 15k (v/v) glycerol added for storage at-70°C.

Glycerol stock cultures from the first round of selection were rescued using helper phage to derive phagemid particles for the second round of selection. 250 il of glycerol stock was used to inoculate 50 ml 2TYAG broth, and incubated in a 250 ml conical flask at 37°C with good aeration until the OD 600nm reached 0.7.

M13K07 helper phage (moi=10) was added to the culture which was then incubated stationary at 37°C for 15 minutes followed by 45 minutes with light aeartion (200rpm) at the same temperature. The culture was centrifuged and the supernatant drained from the cell pellet. The cells were resuspended in 50 ml prewarmed 2TYAK, and the culture incubated overnight at 30°C with good areation. Phage particles were purified and concentrated by PEG precipitation (Sambrook et al., 1990) and resuspended in PBS to 1013 tu/ml.

Phage induced from the first round of selection were selected a second time as described above, except that for the second round of selection cell membrane preparations immobilised on a Nunc immunosorb tube were used as the selection surface. Immobilisation was carried out overnight at 4°C, using 1 ml of membrane preparation at a total protein concentration of 10 yg/ml in PBS.

c) Growth of individual colonies for immunoassay.

Individual colonies from the third round of selection were used to inoculate 100/il of 2TYAG into individual wells of 96 well tissue culture plates (Corning). Plates were incubated at 30°C overnight with moderate shaking (200 rpm). Glycerol to 15% was added to each well and these master plates stored at-70°C until ready for analysis. d) Phacre ELISA to identify membrane-binding-clones.

Cells from the master plates were used to inoculate fresh 96 well tissue culture plates containing 100 fit 2TYAG per well. These plates were incubated at 37°C for 6-8 hours, or until the cells in the wells were growing logarithmically (OD 600 nm 0.2-1.0).

M13K07 was added to each well to an moi of 10 and incubated stationary for 15 min then 45 min with gentle shaking (100 rpm), both at 37°C. The plates were centrifuged at 2000 rpm for 10 min and the supernatant removed. Each cell pellet was resuspended in 1001 2TYAK and incubated at 30°C overnight.

Each plate was centrifuged at 2000 rpm and the 100 au supernatant from each well recovered and blocked in 20 Hl 18hM6PBS (18t skimmed milk powder, 6 X PBS), stationary at room temperature for 1 hour. Meanwhile, flexible microtitre plates which had been blocked overnight stationary at 37°C with either 100 au 10 yg/ml of plasma membrane preparation from either the cell type used for the selection, or an unrelated cell type, were washed in PBS and blocked for 2 hours stationary at room temperature in MPBS. These plates were then washed in PBS and 50 yl of preblocked phage added to each well. The plates were incubated stationary at room temperature, after which the phage was poured off. The plates were then washed three times in PBST, followed by three washes with PBS. To each well of the plates 50 au ouf a 1: 5000 dilution of anti-Fd-HRP conjugated antibody (Pharmacia) was added and the plates incubated at room temperature for 1 hour. Plates were washed as before and developed at room temperature with 50 yl of Sigma TMB HRP substrate. Development was stopped after 1 hour by the addition of 25 yl of 0.5M H2SO4 and absorbance was measured

at 450nm using a microtitre plate reader. 30/95 clones tested by ELISA were found to bind specifically to the selected membranes, but not to the unrelated cell type membrane preparation. e) Sequencing of membrane-specific clones The nucleotide sequences of the membrane-specific antibodies were determined by first using vector specific primers to amplify the inserted DNA from each clone. Cells from an individual colony on a 2TYAG agar plate were used as the template for a PCR amplification of the inserted DNA using the primers LMB3 and fdtetseq. Amplification conditions consisted of 30 cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 2 min, followed by 10 min at 72°C. The PCR products were purified using a PCR Clean-up Kit (Promega) into a final volume of 50/il HzO. Between 2 and 5 yl of each insert preparation was used as the template for sequencing using the Taq Dye-terminator cycle sequencing system (Applied Biosystems). The primers mycseqlO and PCR-L-LINK were used to sequence the light chain of each clone and PCR-H-Link and pUC19reverse to sequence the heavy chain.

LMB3 5'-CAGGAAAACAGCTATGAC-3' (SEQ ID NO: 11) fdtetseq 5'-GTCGTCTTTCCAGACGTTAGT-3' (SEQ ID NO: 12) PCR-H-LINK 5'-ACCGCCAGAGCCACCTCCGCC-3' (SEQ ID NO: 13) PCR-L-LINK 5'-GGCGGAGGTGGCTCTGGCGGT-3' (SEQ ID NO: 14) Sequences of the VH CDR3's of the membrane specific scFv antibodies are shown below. 20 out of the 30 positive clones were taken at random and sequenced. 15 diverse sequences were identified.

CLONE VH CRD3 CL1 GGPACRSRPCQPGG (x4) (SEQ ID NO: 15) CL2 GGPQCPPAFCDPGG (x3) (SEQ ID NO: 16) CL3 GGPTCIQAKCELGG (SEQ ID NO: 17) CL4 GGAQCSSYSCYSGG (SEQ ID NO: 18) CL5 GGPGCQSRHCLPGG (SEQ ID NO: 19) CL6 GGPACRHCQCLPGG (SEQ ID NO: 20) CL7 GGTQCSFGVCATGG (SEQ ID NO: 21) CL8 GGQSCSPSGCBGGG (SEQ ID NO: 22) CL9 GGPLCYPAPCYSGG (SEQ ID NO: 23) CL10 GGPSCFFHSCEEGG (SEQ ID NO: 24) CL11 GGPACPSNACYFGG (SEQ ID NO: 25) CL12 GGFCLNPVCYHGG (SEQ ID NO: 26) CL13 GGPSCLFHSCEEGG (SEQ ID NO: 27) CL14 GGPPCYAARCYPGG (SEQ ID NO: 28) CL15 GGLPCPCDACCSGG (SEQ ID NO: 29) As can be seen from the table of VH CDR3 sequences all the selected antibodies isolated possessed the cysteine-noose. In the case of CL15 two additional cysteine residues have been incoprorated within the mutagenised region giving the possibility that more than one disulphide bridge could be formed within the VH CDR3 of the antibody. This may premit a more complex"knot" structure to be formed. In the case of CL12 one of the flanking, randomised residues had been deleted, demonstrating that variability in flanking sequence length around the cysteine-noose is possible.

EXAMPLE 4: SELECTION OF THE CYSTEINE-NOOSE REPERTOIRE ON WHOLE CELLS. a. Preparation of human CD4+ cells from blood.

Mononuclear cells were prepared from a 50ml buffy coat using Ficoll-Paque (Pharmacia) density gradient centrifugation (600g for 20 min at 20°C). CD4+ cells were then isolated from the 1.5 x 108 recovered cells using a Biotex CD4 column, following the

manufacturer's instructions, although PBS/2% foetal calf serum (FCS) was used throughout. Eluted cells were pelleted at 600g for 5 min and resuspended in 3001 PBS/ 2% FCS. 8.3 x 10 cells were recovered using this procedure. The recovered cells were analysed by flow cytometry and approximately 59t of the cells were found to be CD4+. b) Selection of cell specific binders from the cysteine-noose repertoire.

The cysteine-noose phage antibody library was induced as described in Example 2 part a). Approximately 1 x 1012 phage induced from the cysteine-noose antibody library were then selected on 1 x 106 CD4+ cells in a total volume of lml MPBS.

Phage and cells were incubated at room temperature for 1 hour, and cells were then pelleted at low speed (2000 rpm) in a minifuge for 2 min. The cells were washed three times in 1 ml PBS, and then resuspended in 500y1 PBST to lyse the cells.

Lysates were used to directly infect 5 ml of an exponentially growing culture of E coli in 2TY. Infection and plating out was carried out as previously described. Phage selected from the first round of cell selection were induced as described and second and third rounds of selection were performed exactly as the first round. c) Phage ELISA to identify anti-CD4+ cell scFv's.

1 x 105 CD4+ cells were spun onto 96 well culture wells which had been precoated with poly-L-lysine for 30 min at room tempertaure.

Cells were blocked in 100 yl MPBS for 2 hours at 37°C, and rinsed once in PBS. The phage supernatants were then added to the cells and incubated for 1 hr at room temperature, then washed 3 times in PBS. 100 Al of a 1 : 5000 dilution of sheep anti-fd antibody (Pharmacia) in MPBS was added and the plates incubated at room temperaure for 1 hr. Plates were washed 3 times in PBS and 100 y1 of a 1 : 5000 dilution of donkey anti-sheep alkaline phosphatase conjugate (Sigma) in MPBS was added and incubated for 1 hr at room temperature. Plates were washed 3 times in PBS and alkaline phosphatase activity was visualised using the

chromagenic substrate pNPP (Sigma). Absorbance was measured at 405mm using a microtitre plate reader. 23/95 clones screened were found to bind CD4+ cells. d) Seauencing of the CD4+ cell binding cysteine-noose scFv's.

6 of the 23 positive clones were chosen at random and sequenced as previously described. All six clones were found to be diverse, and all possessed the cysteine-noose motif. 3/6 of the clones had a tyrosine residue as the first residue after the second cysteine, which is the bias which was designed into the original mutagenesis strategy.

Sequences of the VH CDR3's of the CD4+ binding clones: Clone VH CDR3 1CD4 GGLSCVLISCYPGG (SEQ ID NO: 30) 2CD4 GGTTCPRYQCKHGG (SEQ ID NO: 31) 3CD4 GGLSCVLISCYPGG (SEQ ID NO: 32) 4CD4 GGSACLLSSCSSGG (SEQ ID NO: 33) 5CD4 GGVACPPMSCNEGG (SEQ ID NO: 34) 6CD4 GGPRCYSALCYPGG (SEQ ID NO: 35) EXAMPLE 5: SELECTION OF LIGAND INHIBITORS FROM THE CYSTEINE-NOOSE REPERTOIRE.

This example describes use of the biotin tyramine signal transfer selection procedure (as described in patent application GB 9712818.5 and in Osbourn, J. K. et al., Pathfinder selection; in vivo isolation of novel antibodies. Immunotechnology, submited), in a two step manner to isolate antibodies from the cysteine-noose antibody repertoire which inhibit binding of the initial guide molecule to cells. The cysteine-noose repertoire has been designed to potentially provide a population of scFv's which are biased towards binding cellular receptors. This selection procedure could be applied to the generation of inhibitors to any ligand, small molecule, or antibody. The process involves an initial first stage of the selection to biotinylate and capture phage antibodies which bind around the

site of ligand binding. The biotinylated phage are then used directly, without amplification, to guide a second stage of selection using cells in the absence of ligand. In this way antibodies which bind in the ligand binding site can be biotinylated by signal transfer procedure, then captured and screened for inhibition of ligand binding. In this example the chemokine MIP-la, which is a ligand for CCRK1, CCRK4 and CCRK5 chemokine receptors which are found on CD4+ cells, was used as the guide ligand to select for cysteine-noose scFv's which inhibit MIP-la binding to CD4+ cells. a) Selection procedure and capture of biotinylated phage.

1 x 105 CD4+ lymphocytes were incubated with 2 x 1012 phage prepared from the cysteine-noose phage display library in either the presence of biotinylated MIP-la (R and D Systems) at a final concentration of 375nM. The final volume for each selection was made up 40y1 with PBS containing 2.-marvel (MPBS). Selections were incubated for 14 hr at 4°C. Cells were pelleted by centrifugation at 600g for 3 min, and washed in 1 ml MPBS. A total of three washes were carried out. 100 ut of streptavidin-HRP was added at a dilution of 1: 1000 in MPBS. This was incubated for 2 hr, then washed as before. Biotin tyramine was then added in 100y1 of 150mM NaCl/50mM TrisHC1 pH 7.4 containing 3% H202 and incubated for 10 min at room temperature. Cells were washed and resuspended in 10oral TE containing 0.5% triton. Biotinylated phage were captured on 10 lil of MPBS-blocked streptavidin-coated magnetic beads (Dynal). The beads were washed three times in lml PBS/0. 1% Tween 20 (PBST). The beads were then resuspended in 100 yl 100mM triethlyamine for 10 minutes at 37°C, which elutes phage from the beads. TEA was then neutralised with 50 ßl 1M Tris-HCl pH 7.4. After elution the beads were taken out of solution and the supernatant containing the biotinylated phage was taken for use in the second phase of the selection.

The population of biotinylated phage which had been recovered from the first stage of the selection were added directly to 1

x 10"CD4+ lymphocytes in a total volume of 200 y1 in MPBS. Phage were allowed to bind to the cells for 1 hr at room temperature, and cells were then washed 3 times in 1 ml PBS. Cells were pelleted at 4000 rpm for 2 min in a minifuge between washes. A further aliquot of the scFv phage library (2 x 1012 phage) was then added to the cells in lml MPBS and allowed to bind for 1 hr at room temperature. Cells were washed 3 times in PBS as above and then resuspended in 200 ßl MPBS containing 2 yl of streptavidin-HRP complex (Amersham). This was allowed to bind for 30 min at room temperature, and cells were then washed as before.

Biotin tyramine treatment of the cells was carried out as described above. Cells were then lysed by resuspension in 100y1 PBST and 30 81 preblocked streptavidin-coated Dynal beads were added to the lysate. Beads and lysate were rotated at room tempertaure for 20 min, and the beads then taken out of solution on a magnet. Beads were washed 3 times in 1 ml PBST, followed by 3 times in 1 ml PBS. Washed beads were used to directly infect an exponentially growing culture of E coli TG1. A total of around 4 x 103 clones were recovered from this selection procedure. b) Growth of single selected clones for immunoassay and phage ELISA Growth of clones and phage ELISAs on CD4+ cells were carried out as described in the previous example. 24/48 clones were found to be positive for CD4+ cells. c) Assessment of anti-CD4 scFv's to inhibit binding of MIP-la to CD4+ cells.

6 scFv's from the 24 positives were chosen at random and were purified using nickel agarose metal affinity chromatography (Quiagen). 1 x 105 CD4 cells were preincubated with the purified CD4+ cell-binding scFv's, or with an irrelevant control scFv for 1 hr at room temperature in PBS containing 0.1% BSA in a total volume of 100 au. Approximately 5-10 Ug of scFv was used per sample. Cells were pelleted at 4000 rpm in a minifuge and washed once in 1 ml PBS. Biotinylated MIP-la (R and D Systems) was made up according to manufacture's instructions and 5 fit (equivalent

to 5ng) added to the cells in 100 Ul MPBS and incubated at room temperature for 1 hr. Cells were washed as before. 100 y1 of streptavidin-FITC (Sigma) at a dilution of 1 : 100 in MPBS was added and incubated for 30 min at room temperature, and cells were washed as before. Fluorescence was detected using a Coulter Epics-XL flow cytometer. Results are shown in Figure 5. As can be see from the figure MIP-la gives signifcant shift in the fluorescence of the cells when no scFv, or control scFv is added to the cells. In the presence of scFv from the selected clones from the cysteine-noose library MIP-la binding to the cells is significantly inhibited. Inhibition varies from clone to clone. d) Sequencing of the cysteine-noose repertoire clones which inhibit MIP-la binding to CD4+ cells.

Antibody sequences were determined as previously described.

Clone VH CDR3 MI1 GGIRCPARACHPGG (SEQ ID NO: 36) MI2 GGARCVLECYHGG (SEQ ID NO: 37) MI3 GGLRCLISGCLEGG (SEQ ID NO: 38) MI4 GGLRCSTTRCYYGG (SEQ ID NO: 39) MI5 GGNSCCPQBCYNGG (SEQ ID NO: 40) MI6 GGLACQIAPCYFGG (SEQ ID NO: 41) As can be seen from the sequences all six clones contain the cysteine-noose motif. MI2 has a truncated three residue loop, rather than four residues, demonstrating the potential for variation in loop length, whilst retaining cell binding activity.

EXAMPLE 6: DEMONSTRATION OF THE IMPORTANCE OF THE CYSTEINE NOOSE IN ANTIBODY BINDING TO ANTIGEN.

The importance of the cysteine noose in antibody-antigen interactions is demonstrated by in vitro mutagenesis of one or both of the cysteines which contribute to the noose. a) Design of mutacrenesis strategy

Using the membrane specific clone CL1 (SEQ ID NO: 15) as a model antibody to be mutated, the two cysteines within the VH CDR3 are mutated to serine residues. Serine is the preferred amino acid to which cysteine is mutated, since it is structurally very similar to cysteine; the side chain group of serine is CH2OH, compared to the CTLSH of cysteine.

The following oligonucleotides (reverse complement sequences) are utilised in the mutagenesis (mutated residues are marked in bold): <BR> <BR> <BR> <BR> <BR> <BR> CL1. MUT. 1<BR> <BR> <BR> <BR> <BR> <BR> P A S R S R P C Q P amino acid (SEQ ID NO: 43) 5'CCC GCA AGC CGT AGC CGT CCT TGT CAA CCT 3'nucleotide (coding) (SEQ ID NO: 42) CL1. MUT. 2 P A C R S R P S Q P amino acid (SEQ ID NO: 45) 5'CCC GCA TGT CGT AGC CGT CCT AGC CAA CCT 3'nucleotide (coding) (SEQ ID NO: 44) CL1. MUT. 3 P A S R S R P S Q P amino acid (SEQ ID NO: 47) 5'CCC GCA AGC CGT AGC CGT CCT AGC CAA CCT 3'nucleotide (coding) (SEQ ID NO: 46) b) Generation of mutated versions of CL1 These oligonucleotides are used individually in combination with LMB3 (SEQ ID NO: 5) in a PCR, using conditions exactly as described in Example 2b, to generate VH gene segments with either one, or both cysteines mutated to serine (s). Template for this PCR is minipreped DNA from phagemid clone CL1. The CL1 VL gene

segment is PCR amplified from minipreped DNA from phagemid clone CL1 using the primers JH (SEQ ID NO: 8) and Fdtseqtag (SEQ ID NO: 9). The VH and VL gene segments are then PCR assembled using a pull through reaction exactly as described in Example 2b, using the primers FTAG (SEQ ID NO: 10) and LMB3 (SEQ ID NO: 5). The resultant PCR products are digested with the restriction endonucleases Sfi I and Not I, then ligated into Not I/Sfi I difested phagemid vector pCANTAB 6.

The clones are designated: CLIP (parental CL1) CL1M1 (first cysteine of the noose mutated to serine) CL1M2 (second cysteine of the noose mutated to serine) CL1M3 (both cysteines of the noose mutated to serine) c) Phase ELISA to assess whether mutants retain ability to bind membrane fractions Phage ELISA is carried out exactly as described in Example 2d.

The clone CLIP gives a high ELISA signal under these conditions (>0.5 after 30 minutes development), whereas the other three mutated clones give a signal greatly reduced compared to the parent clone.

EXAMPLE 7: CONSTRUCTION OF AN EXPANDED CYSTEINE-NOOSE REPERTOIRE.

Example 1 above describes the design of a cysteine-noose scFv repertoire which includes in the VH CDR3 a cysteine noose of 6 amino acids in size (i. e. a 4 amino acid loop). In order to provide a more diverse source of antibodies which could mimic these naturally occurring ligands the cysteine-noose repertoire was expanded to cover a wider range of potential noose lengths. a) Construction of the expanded library Seven variants of the oligonucleotide VH Cys loop For were generated with the number of residues present between the two

cysteines varying from 4 to 10 as follows: Oligonucleotide VH Cys loop For; <BR> <BR> <BR> <BR> <BR> <BR> 5'CAG GGT (CT) CC (CT) TG GCC CCA CAA TCC NNN NNN ACA (NNN) 4-10 GCA NNN NNN CCC GCC TCT (TC) GC ACA GTA ATA CAC (A% E ; EETNO34'8) These primers were used individually in conjunction with the primer LMB3 (SEQ ID NO: 5) to amplify a repertoire of heavy chains which were mutated in the VH CDR3 to give a variety of noose lengths. The PCR template used was DNA prepared from the large scFv antibody library exactly as described in Example 2.

The repertoire of VL chains generated using primers JH Cys Back (SEQ ID NO: 8) and Fdtseq tag (SEQ ID NO: 9) as described in Example 2 were used as partners for the new set of mutated VH chains.

All primary PCR's and PCR pull through reactions were carried out as described above. PCR products containing the VH CDR3's with different noose lengths were kept separate at all times and were cloned into pCANTAB6 in separate aliquots using the protocols detailed in Example 2. This effectively resulted in the generation of seven separate scFv repertoires each containing VH chains with cysteine nooses in the CDR3 of either 4,5,6,7,8, 9, or 10 residues not including the cysteines. From this process approximately 5 x 107 individual clones were generated for each individual noose length, giving a total combined library size of 3.5 x 108. b) Validation of the expanded library in terms of percentage recombinants and sequence.

Single colonies were picked from the output of the pCANTAB6 cloning step to assess the percentage of colonies which contained scFv inserts. The presence of inserts was determined by PCR using primers LMB3 (SEQ ID NO: 11) and fdtetseq (SEQ ID NO : 12) and the PCR products taken on to be sequenced as described in

Example 3 part e).

Percentage of clones which were recombinant (i. e. contained a scFv insert) were as shown below. The average o recombination was 92% Loop length recombinant 4 94 5 94 6 79 7 100 8 89 9 94 10 94 A random selection of PCR scFv products were sequenced to ensure the mutation process had been successful and that cysteine-nooses were present in the VH CDR3's. Cysteine noose-encoding sequences containing open reading frames included sequences coding for the following: Loop-length VH CDR3 5 GGALC NHRHR CTGGG SEQ ID NO: 49 GGAPC ATAIR CLIGG SEQ ID NO: 50 6 GGDPC RLSRTK CNNGG SEQ ID NO: 51 7 GENDC RVHPPDT CJMVG SEQ ID NO: 52 8 GGSTC RVTSTRYG CYSGG SEQ ID NO: 53 9 GGNHC YVERIAGTS CLLGG SEQ ID NO: 54 10 GGLSC GHNYPAGSTE CLIGG SEQ ID NO: 55

Hence it has been demonstrated that this method of library construction can be used to produce antibody clones with a range of noose lengths.

EXAMPLE 8: USE OF THE CYSTEINE NOOSE LIBRARY TO GENERATE PEPTIDE LIGAND MIMETICS.

Example 5 described the selection of ligand inhibitors of MIP-la from the cysteine-noose repertoire. Six different scFv antibody clones were isolated which gave varying degrees of inhibition of MIP-la binding to CD4+ cells. The VH CDR3's of five of these clones were taken and used as the basis for the design of peptides covering the cysteine-noose region. a) Design of peptides.

Peptides were synthesised which included the cysteine-noose region of the VH CDR3. Analagous peptides were also generated which contained serine residues in place of the noose cysteines.

Serine was chosen because it possesses a similar R group to cysteine except that the SH group is replaced by OH, hence serine is incapable of forming a di-sulphide cross-linked noose.

The following peptides were synthesised: ML1CA3 GIRCPARACHPG SEQ ID NO: 56 ML1SA3 GIRSPARASHPG SEQ ID NO: 57 ML2CA5 GARCVLECYHG SEQ ID NO: 58 ML2SA5 GARSVLESYHG SEQ ID NO: 59 ML3CA9 GLRCLISGCLEG SEQ ID NO: 60 ML3SA9 GLRSLISGSLEG SEQ ID NO: 61 ML4CA11 GLRCSTTRCYYG SEQ ID NO: 62 ML4SA11 GLRSSTTRSYYG SEQ ID NO: 63 ML5CC1 GLACQIAPCYFG SEQ ID NO: 64

ML5SC1 GLASQIAPSYFG SEQ ID NO: 65 All peptides were examined by mass spectrometry to ensure synthesis had been completed successfully. Peptides were resuspended in 100W DMSO at approximately 2 mg/ml and stored at -20°C. b) Assessment of ability of synthesised peptides to inhibit MIP-la binding to Jurkat cells.

1 x 105 Jurkat cells were used per sample and were incubated with 1OAl of 2 mg/ml peptide in a total volume of 100 fit PBS/ 0. 1% BSA at 4°C for 1 hour. Cells were then washed in 5ml of PBS BSA and pelleted at 12000 rpm in a Sorvall bench top centrifuge for 5 min. Cells were resuspended in 25 fit of PBS/BSA and 10 il of biotinylated MIP-la (equivalent to 5 ng) added and incubated for 1 hour at 4°C. Cells were washed as before and resuspended in 10oral of streptavidin-FITC (Sigma) made up at a dilution of 1: 100 in PBS/BSA. Cells were washed as before and fluorescence was detected using a Coulter Epics-XL flow cytometer. Control samples with no peptide pre-incubation were also included. c) Summary of results.

Two of the five cysteine peptides tested showed some inhibition of MIP-la binding to Jurkat cells in the cysteine form but not the serine form. Peptide ML4CA11 gave partial inhibition of MIP-la binding, and ML2CA5 gave almost complete inhibition of MIP-la binding. The results are shown in Figures 6a and 6b. The maximal observable shift due to MIP-la binding to the cells with no peptide present is represented by the unshaded solid line peak. The unshaded dotted line peak represents the background fluorescence due to streptavidin-FITC binding to the cells. The level of labelling after pre-incubation with the various peptides is shown by the shaded peak.

These inhibition profiles demonstrate that the VH CDR3's of antibodies selected from the cysteine-noose library can be used as starting points to generate peptide mimetics of naturally occurring ligands.