WO1992021972A1

WO1992021972A1 - Rapid screening assay for the identification of ligand-receptor interactions

Info

Publication number: WO1992021972A1
Application number: PCT/US1992/004640
Authority: WO
Inventors: Roland Buelow; Julian K. Hickling
Original assignee: Immulogic Pharmaceutical Corp
Current assignee: Immulogic Pharmaceutical Corp
Priority date: 1991-06-06
Filing date: 1992-06-05
Publication date: 1992-12-10
Anticipated expiration: 1993-12-06

Abstract

A rapid screening assay is provided by lysing clones having a plurality of different sequences potentially including a sequence of interest for stimulating T-cells. The poly (amino acid) sequences are enriched by cloning the sequences joined to a sequence encoding protein A to provide for a fusion protein which may then be readily purified by binding to immune complexes and removing non-specifically bound contaminants. The fused protein may be used directly or the sequence of interest cleaved from the protein A sequence and then used.

Description

RAPID SCREENING ASSAY FOR THE IDENTIFICATION OF LIGAND-RECEPTOR INTERACTIONS

INTRODUCTION

Technical Field

The field of this invention is the screening of diverse protein sequences.

Background

In many physiological processes, there are involved a large number of polymorphic proteins. Illustrative of the diversity of biologically related compounds are components of the immune system, such as immunoglobulins, major histocompatibility complex (MHC) antigens, and T-cell receptors. This diversity is predicated on the need to be able to detect an extraordinarily large number of different compounds and proteins or other sequences, while still being able to discriminate between self and non-self.

For many purposes, one wishes to be able to regulate the immune system, activating it, enhancing the response or inhibiting the response. For infection, one would like to greatly decrease the response time necessary to respond to an infection, while for autoimmune diseases or transplantation, one would wish to inhibit the immune response.

In attempting to regulate the immune response, one must understand how the immune system operates and to what sequences the immune system responds. Therefore, as to the individual aspects of the immune system, one is interested in being able to determine how cells will respond to a particular antigen or antigen fragment. In order to be able to achieve this, it is necessary to screen large numbers of compounds in an economic, efficient, and reproducible manner. To do this, one must be assured that one has introduced a minimum number of potentially interfering contaminants along with the compound of interest. It is therefore of interest to develop systems which allow for rapid screening with reasonable assurance that the results are reproducible and comparable.

Relevant Literature

References of interest include Lamb e_t .a_l. , EMBO J. 6(5): 1245-1249 (1987); Thole e± al . , Inf. Immun. 56(6) 1633-1640 (1988); Mustafa gt a_l. , J. Immunol. 141(8): 2729-2733 (1988); Oftung ___ a_l. , J. Immunol. 138(3): 927-931 (1987) and Mustafa _=£ _al.. Nature 319: 63-66 (1986).

Summary of the Invention

Methods are provided for screening large numbers of diverse poly(amino acid) sequences, where sequences undergo an initial enrichment, followed by screening of the diverse sequences with an appropriate receptor other than an immunoglobulin. The enrichment is achieved by cloning the sequence as a fusion protein to a fusion partner. The resulting clones are lysed and the fusion protein purified by complexation with a specific binding partner for the fusion partner. The enriched protein fusion product may then be used directly for screening. Brief Description of the Drawing

Figure 1 provides three graphs of the results obtained when mixing the lysates from pCA and pCA.HA and then subjecting them to various modes of purification. The top graph (Fig. 1A) purifies the lysate mixture using a conventional antibody - Sepharose column; the middle graph (Fig. IB) uses Ig coated plate purification; and the bottom graph (Fig. 1C) uses BSA-anti BSA complex purification.

Description of the Specific Embodiments

Methods are provided for screening large numbers of diverse protein sequences for binding to receptors, so as to evaluate ligand receptor interactions. Open reading frames encoding diverse poly(amino acid) sequences are joined to a fusion partner to provide a sequence encoding a fusion protein. The sequences are then cloned, so as to provide clones of a single fusion protein. The cells from each clone may then be lysed and the lysate combined with a binding partner for the fusion partner, where the binding partner is polyvalent. The resulting polyvalent complexes are separated from the lysate, washed and then used for screening of receptor binding, usually without further processing. The sequences of interest may take many forms, being from diverse sources and may find use for diverse purposes. Sources of interest may include proteins or fragments thereof encoded by a cDNA library or genomic library, where the library may have been partially separated as to one or more properties. The sequences may come from any cell

SUBSTITUTE type or organism, where there is an interest in determining binding affinity for a receptor. Rather than naturally occurring sequences, peptides and proteins may be prepared from oligodegenerate DNA sequences where the sequences are synthesized, and at one or more sites, two or more different nucleotides are added. Thus, any mixture of nucleic acid sequences, naturally occurring, synthetic or combinations thereof, may be employed in the subject invention.

The sequences of interest are joined to a sequence encoding a fusion partner. The fusion partner will allow for binding of the amino acid sequence of interest to its receptor, expression of the fusion protein, formation of high molecular weight complexes in conjunction with a binding partner, and lack of cross-reactivity with the sequences of interest.

The fusion partners may come from diverse sources and provide for a variety of ways of forming a complex. The binding partner will be polyvalent to allow for the formation of large complexes, when the fusion protein and the binding partner are brought together. Various fusion partners may be employed, such as the functional portions or all of protein A, avidin, lectins, or any polyepitopic peptide or protein, or the like. Thus, the binding partner may include sugars, where the sugars may be by themselves or in conjunction with larger aggregates, such as cells, polybiotin, where biotin may be bound to an oligomeric or polymeric backbone, or one or more antibodies, where the fusion partner is polyepitopic and has the same or different polyepitopic sites. The open reading frame encoding the sequence of interest may be joined directly or through a bridge to the sequence encoding the fusion partner. The bridge may or may not include a selective cleavage site, depending upon whether one wishes to isolate the peptide of interest separate from the fusion partner. Various methodologies may be employed for providing for preferential, if not specific, cleavage between the peptide of interest and the fusion partner. A wide variety of amino acid sequences are known for which specific proteases are available. Enzymes of interest include Factor Xa, KEX2, HIV protease, and the like.

The gene encoding the fusion protein may be prepared in a variety of ways, employing synthesis using commercially available nucleic acid synthesizing equipment, where overlapping single stranded sequences can be prepared and then brought together and ligated. Alternatively, individual sequences may be cloned, undergo appropriate restriction and then ligated together. The polymerase chain reaction may also be used. The particular manner in which the gene encoding the fusion protein is prepared " is not critical to this invention.

A wide variety of transcriptional initiation regions may be employed, which may be constitutive or inducible. For the most part, the promoters will be of weak or medium strength, where inducible promoters include promoters of β-galactosidase, cl^^s and the like. There are numerous commercially available expression vectors, as well as numerous expression vectors described in the literature. These expression vectors normally have polylinkers downstream from a transcriptional initiation region, so that one may insert the coding region in appropriate juxtaposition to the functional sequences of the transcriptional initiation region. Also present downstream will be a transcriptional termination region, which allows for processing of the transcribed sequence to a mature messenger RNA. The vector will also usually include a marker for selection, preferably positive selection, which may include antibiotic resistance, complementarily to an auxotrophic host, or the like. The marker will normally be under the control of a constitutive promoter. Illustrative selective markers include amp, cam, neo, and the like.

Also included will be at least one origin of replication which preferably allows for high copy number in the intended host. Conveniently, either plasmid or phage replication systems may be employed, desirably plasmid replication systems.

The complete vector which may be used for cloning the various sequences will usually include a plasmid or viral origin, a transcriptional initiation region, and a sequence encoding the fusion partner under the regulation of the transcriptional initiation region. Downstream from the transcriptional initiation region may be a polylinker, either 5'- or 3'- to the fusion partner. The polylinker will allow for insertion of sequences in reading frame with the fusion partner, frequently permitting random sequences to be inserted by allowing for insertion in all three reading frames to provide an open reading frame. Where the polylinker is upstream from the fusion partner, the sequence of interest will include an initiation codon or one may be provided as part of or upstream to the polylinker. The subject vector may be introduced into the host by any convenient means, including transformation, electroporation, protoplast fusion, calcium precipitated DNA, or the like. The particular host will be prokaryotic. While many different prokaryotes may find use, the host of choice will be E. coli. , particularly strains having the following mutations: recA. m^~' etc.

Once the transformants have been prepared, the cells may be cloned in accordance with conventional procedures, e.g., plating on agar, using limiting dilution, separation on gels, etc., growing the cells to high density, induction of expression of the fusion protein, as appropriate, harvesting the cells, and lysing the cells. Various techniques for lysing may be used, although some techniques may be found less efficient than other techniques. Techniques which may be used include mechanical disruption, freezing and thawing, sonication, enzymes and/or detergent treatments, where the treatments may be used individually or in combination.

The lysates will then be combined with the binding partner of the fusion partner. The particular type of complex is not critical to this invention and any convenient binding partner may be employed. The binding partner may be used by itself or in conjunction with a second agent which provides for extensive cross-linking. For example, one may use combinations of antibodies, where the primary antibody recognizes the fusion partner, while the secondary agent is an antibody which recognizes a region of the primary antibody, particularly the constant region. The ratio may be selected, so that the large immune complexes may be formed. Where binding partners other than antibodies are employed, various techniques may be used to provide for extended complexation. With protein A, appropriate polyepitopic antigens may be employed, even including cells. If desired, the immune complex may be preformed by combining the primary antibody with the secondary agent antibody to form a complex which may then be mixed with a lysate containing the fusion protein.

The large immune complexes or other large binding partner bind the fusion protein and may then be separated by any convenient means, conveniently centrifugation. The complex particles which are isolated may then be washed, if desired, and re-suspended in medium, where the fusion protein is available for cellular interaction. In some instances, it may be desirable to provide for binding of the complexes to a surface, where the surface will have means for covalently or non-covalently binding the complex to the surface. This can be achieved by having the secondary antibody bound to the surface or having the binding partner bound to the surface, depending on the nature of the binding partner.

The lysate may be produced in situ by combining the cells with a cell wall degrading enzyme, such as lysozyme, following with detergent to allow for release of the fusion protein and binding to the binding partner bound to the surface. The surface may then be washed and any non-specifically bound material removed to leave the surface free of contamination from components of the host.

In many cases, particularly where one is screening large numbers of clones, usually in excess of 5,000 clones, more usually in excess of 10,000 clones, one may wish to pool clones, so that each individual determination may have from 10 to 1000 different clones involved in the determination. One can then screen the individual clones from positive mixtures. The enriched fusion protein may now be used in a variety of ways. It may be screened directly with the receptor of interest. Alternatively, as already indicated, the peptide of interest may be cleaved from the fusion partner and complex, followed by isolation and purification, if desired. Various techniques exist for purification, such as gel electrophoresis, HPLC, and the like.

The enriched or purified fusion proteins or peptides of interest are then screened with the receptor, particularly the major histocompatibility complex Class I and II molecules and the T-cell receptor, more particularly T-cells or microsomes comprising the T-cell receptor. The T-cells may be the T-cells from any particular source, such as peripheral blood, thymus, lymph node, pancreas, or other lymphoid tissue, may be homogenous or heterogenous, may be from the same or different species from which the sequences of interest were obtained, or the like. Thus, one may screen sequences for binding to a specific T-cell population of interest. One can investigate T-cell response to a mixture of peptides by combining antigen presenting cells (APCs) and T-cells in monitoring the response of the T-cells.

Various techniques exist for monitoring T-cell response. One can measure thymidine uptake, by employing radioactive thymidine; one can measure the secretion of IL-2, as an indication of T-cell response; CD69 production; or calcium flux. Thus, by measuring a secretion product associated with T-cell activation or T-cell proliferation by measuring DNA replication, one can evaluate the presence of a sequence to which T-cells may respond. Since mixtures of clones may be employed, one can then select those samples in which a T-cell response has been observed and screen individual clones or smaller mixtures of clones, from one to ten clones, whereby one can identify the particular sequence which has provided the response.

Where a protein has been found to provide for a T-cell response, one may screen the protein for the immunodominant sequence. By providing for deletions in a nucleic acid sequence, which results in the deletion of several amino acids, one can screen the entire protein by preparing clones of the mutated antigen and determining which clone(s) no longer has the ability to provide the T-cell response. Deletions may include up to about 90% of the amino acids of a protein, with various strategies being employed to ensure identification of a sequence of interest. Once one has identified a specific sequence of interest, such as an immunodominant sequence which binds to a T-cell receptor, one may further refine the functional portion of the sequence, by synthesizing various peptides, whereby the minimum number of amino acids associated with obtaining a T-cell response may be defined. Alternatively, one may restrict the cDNA encoding the protein providing for the T-cell response and introduce the various fragments into a polylinker. In this manner, one may screen various fragments to determine the T-cell response as compared to the intact antigen. By employing partial restriction, one can insure that the peptide is not missed because cleavage has occurred at an internal site of the peptide.

By employing the subject method, one can efficiently screen 10⁵-lθ6 clones for T-cell response and identify those proteins and peptides to which T-cells respond and the fragment of the protein associated with the response. Thus, one may identify those sequences which may be associated with a wide variety of pathogenic conditions, such as autoimmune diseases, infection, allergy, transplantation rejection, etc.

Alternatively, one may use the subject method to obtain samples of a plurality of cloned peptides, which may be screened with various receptors, such as surface membrane proteins, as part of a cell membrane microsome or insoluble form, or immobilized soluble receptor, e.g., thyroxine binding globulin, or the like, to determine their biological activity.

A kit may be provided for convenience, comprising a vector comprising a prokaryotic replication system, e.g., origin of replication, a marker, an expression cassette comprising a transcriptional and translational initiation region—promoter and Shine-Dalgarno sequence—the sequence encoding the fusion partner, e.g. protein A, under the transcriptional control of said region, at least one restriction site, which allows for insertion in reading frame with the fusion partner, preferably a polylinker which allows for insertion in all three reading frames, a transcriptional termination region, and such other functional sequences as may be desirable. The kit will also include a polyepitopic protein and antibodies to the protein, individually or as a complex, where one of the protein or antibodies may be bound to a surface. The following examples are offered by way of illustration and not by way of limitation.

EXPERIMENTAL

The pCA vector was prepared by employing the protein A encoding sequence from pRIT2T (Pharmacia) into pCDNA2 (Invitrogen) joined to a polylinker. The plasmid had in the clockwise direction beginning with the protein A under the transcriptional control of the β-gal promoter region, a polylinker, the fl-origin, an ampicillin resistance gene, and a pBR322 origin. To demonstrate the utility of the subject invention, the sequence coding for the hemagglutinin peptide 307-319 was inserted at BamHI and Xhol sites of the polylinker in reading frame with protein A to provide a fusion protein (pCA.HA) .

To demonstrate the subject invention, the plasmid pCA is transformed into E. coli. Tg-1 by electroporation. The resulting transformants are grown overnight in 2xYT (tryptone-yeast extract containing 100 μg ampicillin/ml) and the resulting overnight culture used to inoculate 2xYT 1:50. The second culture is grown until ODgoo ~ !• To induce expression, isopropyl thiogalactoside (IPTG) is added (1 mM) and the culture incubated for an additional four hours at 37°C. The bacteria are lysed with lysozyme/detergent. The lysate is cleared by centrifugation and stored at -20°C.

For the BSA-antibody complex purification of the fusion protein, 3 μg of bovine serum albumin is incubated with 35 μg of affinity purified anti-BSA antibody in 55 μl phosphate buffered saline (PBS) for 2h at room temperature. The pCA lysate prepared above is thawed and spun for 30 min at 4°C. To the immune complex prepared above is added 100 μl of the lysate and the mixture incubated for 30 min on ice. After spinning for 15 min at 4°C, the pellet is washed once with 500 μl sterile PBS and the pellet re-suspended in medium containing 5% human serum (for T-cell proliferation assays).

For the coated plate assay, a 96 well flat bottom tissue culture plate is coated with 50 μl of 0.1 mg/ml human IgG in PBS overnight. The wells are then washed 5X with PBS and coated with 50 μl of rabbit anti-human IgG in PBS (0.1 mg/ml), and the mixture incubated for lh at 4°C. The wells are then washed 5X with PBS, and 150 μl of 25 mM Tris pH 8.0, 50 mM EDTA, 50 mM glucose and 5 mg/ml lysozyme added. An overnight culture of pCA transformants is inoculated into 2xYT 1:10, the mixture allowed to grow for 2h at 37°C, expression induced by adding IPTG to 1 mM, and the culture allowed to grow for 4h. 10 μl per well of the culture of peripheral blood lymphocytes is added to pre-coated wells, incubated 10 min at room temperature, followed by the addition of 20 μl of 10% Triton X-100, and the mixture incubated lh on ice. After washing 5X PBS, 100 ul antigen presenting cells (APC) and T-cells (1-2X10⁴) (irradiated 0.5-1X10⁵ PBL) is added per well. One row of the plate is coated solely with antibodies as a control of the proliferation assay, while a second row is coated with pCA lysate.

Following the above procedure, various mixtures of the lysate were prepared from E. coli. containing either pCA vector or pCA.HA vector and the mixtures subjected to different modes of enrichment and assayed for the ability to stimulate T-cell proliferation. The accompanying graphs (Figure 1) indicate the results.

The top graph in Figure 1 shows that in a mixture comprising 1000 parts of neat protein A to one part of protein A bound to amino acids 307-319 of hemagglutinin (HA307-319), one can detect the presence of the fusion protein at a concentration of 0.1 μl/well by using antibody complexes. Thus, at extraordinary low concentrations of the target fusion protein, one can still observe an effect in a T-cell response test using tritiated thymidine. By contrast, in the middle graph in Figure 1, where an immunoglobulin coated plate is used, there is substantially no difference in the response observed between neat protein A and the fusion protein at the same mole ratio; however, the immunoglobulin coated plate purification, although having a lower sensitivity, is a much less complicated and lengthy procedure than an antibody - Sepharose column purification.

MATERIALS AND METHODS

Strains and reagents.

E. coli strain TGI (supE hsdΔ5 thiΔ (lac^~ proAB) F' [traD36 proAB-i- lacl<J lacZΔM15] was used throughout (Immulogic Pharmaceutical Corp., Cambridge MA). T4 ligase, and all restriction enzymes were obtained from New England Biolabs (Beverley MA.). The construction of vector pCA, derived from pCDNA2 (Invitrogen, San Diego CA) , is described previously.

Synthesis and subclonino of olioonucleotides

Oligonucleotides were synthesized using an Applied Biosystems (Foster City, CA) PCR-mate-EP 391 DNA synthesizer, and purified on OPC (oligonucleotide purification column) columns (Applied Biosystems) . The oligonucleotides contained the sequence coding for the T cell determinant HA 307-319 of the influenza virus strain X31 hemagglutinin gene, followed by three stop codons (TAA, TAG TGA) , and a Bglll recognition sequence. The 5' and 3' ends of the oligonucelotides were designed to be cohesive with BamHI and Xhol cut DNA respectively. Degenerate oligonucleotides were produced by inserting the codon NNK at the position to be mutated, where N represents an equal mixture of the oligonucleotides G,A,T and C, and K represents an equal mixture of G and C.

larnHI P K Y V K Q N T L K L GATCC CCC AAG TAT GTT AAG CAA AAC ACC CTG AAG TTG G GGG TTC ATA CAA TTC GTT TTG TGG GAC TTC AAC

A T X X X GCA ACA TAA TAG TGA CGT TGT ATT ATC ACT

This produces a mixture of 32 codons coding for all 20 amino acids, along with the amber stop codon, which in the TGI strain (supE) codes for glutamic acid. Single stranded oligonucleotides were annealed by heating an equimolar mixture of complementary strand to 95°C for 10 mins, and then cooling slowly. Double stranded oligonucleotides were subcloned into the BamHI and Xho I sites downstream of the protein A coding sequence in pCA. The efficiency of subcloning was determined by testing for the presence of the novel, unique Bglll site absent in the original pCA vector.

Production and screening of fusion peptides

Products of a ligation reaction were electroporated into E. coli TGI, plated onto agar plates containing 100 μg/ml ampicillin, and cultured overnight. Individual colonies were picked and inoculated into individual wells of 96 flat bottom well tissue culture plates (Costar, Cambridge MA), containing 100 μl 2XYT broth plus 100 μg/ml ampicillin, and grown, with shaking, overnight. 10 μl of each culture was then replica plated into fresh 96 well plate. (Glycerol was added to the original wells to a final concentration of 15%, and the plates stored at -80°C to provide a source of material for sequencing.) The bacteria were cultured for 2 hrs, then induced by adding IPTG (Sigma, St. Louis MO) to a final concentration of ImM, and culturing for a further 4 hrs. At the end of this period, 10 μl of each culture was replica plated into antibody coated flat bottom wells (see below) containing 150 μl lysis buffer (50 mM glucose, 25 mM Tris.Cl pH 8.0, 50mM EDTA, 5 mg/ml lysozyme), and incubated at room temperature for 15 mins. Triton X-100 (Sigma) was added to each well to a concentration of 1%, the plates were incubated at 4°C for 1 hour, and then washed 5 times with PBS.

Flat bottom 96 well tissue culture plates (Costar were pre-coated with human IgG (lmg/ml, 50 μl/well) (Organon Teknike Corporation, West Chester PA) at 4 °C overnight, washed 3 times with PBS, then incubated with 50 μl/well IgG fraction rabbit anti-human IgG (0.lmg/ml) for 1 hr at 4°, again followed by three washes with PBS.

The number of colonies to be screened was calculated according to: P₀=(l - f)ⁿ where P₀ is the probability of any given codon being absent; f is the frequency of any given codon (ie. 1/32); and n is the number of colonies screened. n was selected so that P₀ was less than 5%.

T cell clones and proliferation assays The T cell clone 18.41 specific for HA 307-319, isolated from a donor haplotype: HLA-DR4 Dw4, DR7, DRw53, DQw8, DQw9 was used for all the experiments in this study. The clone recognized the peptide presented by DR4 Dw4, but not by DR1 Dwl, DR4 DwlO, DR4 Dwl4 nor DR4 Dwl5. T cell clones were maintained in, and proliferation assays performed in RPMI1640 (Gibco, Gaithersburg MD) supplemented with 5% human AB serum (Biocell, Carson CA), 100 U/ml penicillin, 100 μg/ml streptomycin, and 2mM L-glutamine (JRH Biosciences, Lenexa KS).

T cell proliferation assays were carried out in wells containing captured fusion peptide, by adding irradiated (3000 rads) DR4 Dw4 PBMC at 5 x 10 /well, followed by T cell clone 18.41 at 1 x 10⁴ cell per well. The final well volume was 200 μl. The cells were incubated for a total of 72 hrs at 37°C, in a humidified, 95% air 5% CO2 atmosphere, and pulsed with tritiated thymdine (1 μCi/well) (Amersham, Arlington Heights IL) for ^" the last 18 hrs of the incubation. Cells were harvested on a Skatron (Sterling, VA) 96 well harvester, and ³H-thymidine incorporation determined by liquid scintillation counting with a LKB-Pharmacia (Turku, Finland) 1205 Betaplate counter.

Double Stranded DNA sequencing

DNA was isolated from 1.5ml of an overnight culture to E. coli TGI by a modification of the STET boiling method (Golumbeski et al. (1990) U.S.B. Comments 16:4), and sequencing performed using Sequenase 2.0 (United States Biochemical Corp., Cleveland OH) .

Generation of degenerate oligonucleotides

To confirm the random nature of codon usage and amino acid substitution in the system, degenerate oligonucleotides were synthesized as described in materials and methods, that allowed for substitution of every possible amino acid at position 319 within HA 307-319. 52 of the resulting colonies were selected at random, and the DNA sequenced. Nucleotide usage at position one and position three in the codon was random. At position two a deficiency in the frequency of cytidines incorporated was observed (% versus the expected 25%) . This bias was likely to have been introduced during the chemical synthesis of the oligonucleotide. Despite this, the observed amino acid usage at this position did not deviate significantly from that expected on the basis of random codon frequency (x² >> 0.05).

Effect of amino acid substitutions at position 309 An initial experiment was performed to substitute each natural amino acid at position 309, and to examine the effect on recognition by the HA307-319 specific T cell clone 18.41. Monosubstitutions at this position have been shown previously to have dramatic effects on the ability of HA 307-319 to bind to HLA molecules. Only peptides containing tyrosine (the natural residue) or substitutions with bulky hydrocarbon side chains

(phenylalanine, leucine, isoleucine, valine, and methionine) retained the ability to bind to HLA-DRl

Dwl (Rothbard et al. (1989) Int. Immunol 1:479), or DR4 Dw4 on cell surfaces. Further evidence for the importance of this residue is provided by the observation that a poly-alanine 13 amino acid peptide containing tyrosine at the equivalent position to 309 has equal binding affinity for HLA-DRl Dwl (Jardetzky et al. (1990) EMBO J.7:1797) as the natural HA

307-319 sequence.

Individual E. coli colonies were picked, cultured and lysed in 96 well plates as described, and the irradiated DR4 Dw4 PBMC and clone 18.41 were added to the wells. The results were graphed with proliferation in each well expressed as a percentage of the proliferation obtained to replicate wells on the same plate containing captured natural sequence fusion peptide. Between 10,000 cpm and 30,000 cprn were routinely obtained in response to the natural sequence peptide in this and the subsequent experiments. No proliferation (ie. less that 300 cpm) was ever observed to the control irrelevant protein A fusion protein (human heat shock protein 65) also included on each plate. Fifteen of the 168 colonies screened stimulated clone 18.41 to at least 75% of the proliferation induced by the natural sequence peptide. Following DNA sequencing, these positive colonies were found to code for either tyrosine (the natural residue), or phenylalanine, leucine, valine, and isoleucine substitutions. Thus the results obtained with the oligonucleotide mutagenesis systems were entirely consistent with all previous binding data relating to 309. Clone HA1.7 has been reported to tolerate only phenylalanine and isoleucine substitutions at 309. The set of monosubstituted analogues tested was not exhaustive however, presentation was in the context of HLA-DRl Dwl, and the differences also presumably reflect differences between the TcRs of two different clones.

Amino acid requirements for T cell recognition at position 319 in HA307-319.

Position 319 is situated outside the core of the peptide shown to be responsible for binding to HLA molecules, and can be deleted without any loss of binding activity, providing the C terminus of the peptide is amidated. Thus, this position might be expected to be more tolerant of amino acid substitution than position 309. Indeed, the results showed this to be the case. Of the 168 colonies screened with 18.41, 114 stimulated proliferation greater than 75% of that obtained to the natural sequence, and only 22 colonies failed to stimulate clone 18.41 above background. Forty three colonies that stimulated proliferation above background were selected at random and sequenced. Nearly all the possible amino acid substitutions were detected in this sample at least once (exceptions were: asparagine, aspartic acid and methionine) , and gave at least 75% of the proliferation obtained to the natural sequence (exceptions: alanine; 65%, serine; 51%, tyrosine; 63%). The omissions were most likely due to the size of the sample sequenced. When colonies that failed to stimulate T cell proliferation were sequenced, no amino acid was found to be consistently present, suggesting that no single amino acid was responsible for abrogating T cell recognition. Most of the negative colonies sequenced contained deleted or truncated oligonucleotide sequences.

Amino acid reouirements for T cell recognition at positions 310, 313 and 316 in HA307-319.

Residues at position 310 and 316 of HA 307-319 are thought to interact both with the T cell receptor and the MHC molecule in the ternary complex, whilst position 313 is thought to interact solely with the T cell receptor. Long chain biotin can be added to each of these positions without any loss of binding activity (Rothbard et al., supra) . and a variety of polar, non-polar, and charged substitutions can be made without affecting binding to DR4 Dw4. In contrast, the side chain requirements at each of these positions for T cell recognition are much more stringent. Degenerate oligonucleotides were constructed which coded for amino acid substitutions at positions 310, 313, and 316 and screened as before. The results were graphed as previously described.

In the case of substitutions at position 310, 10 of the 176 colonies screened stimulated proliferation greater than or equal to 75% of that elicited by the natural sequence. These colonies were found to code for valine (the natural amino acid at this position) isoleucine, leucine and threonine. That only a small number of residues were found to stimulate 18.41 supports the model that this position is interacting with the T cell receptor. Interestingly, all the positive substitutions have a methylene group at the 5 b carbon, and a methylene or methyl group at the g carbon. Both these features seem to be necessary, as neither alanine nor serine were positive. The fact that methione at 310 also was not found to stimulate 18.41 suggests that either the size or polarity of

10 the sulphur atom is deleterious for recognition. In experiments with the HLA-DRl restricted T cell clone HA1.7 specific for HA307-319, only valine and glutamine were tolerated at this position, again reflecting variation in requirements for binding to

15 the TcRs of the two T cell clones.

Position 313 is thought to interact solely with the T cell receptor. Clone HA1.7 responded weakly (40-50% of response to natural sequence) to serine and phenylalanine monosubstitution, and failed to

20 recognize lysine and aspartic acid substitutions at 313. An exhaustive set of analogues was not tested however, so the possibility still exists that other residues are tolerated at this position by the T cell receptor. In terms of T cell recognition by clone

25 18.41 however, this position was found to be the least tolerant of amino acid substitution out of all those studied. Only 3 colonies out of 168 screened stimulated proliferation equivalent to the natural sequence, and these were subsequently found to code

30 for asparagine (the natural residue) at 313. The pool of 168 colonies screened can be assumed to have contained all possible amino acid substitutions however. When 38 colonies were selected and sequenced at randon, 19 out of 20 amino acids were present, the exception being tryptophan.

Position 316 was more tolerant of amino acid substitution than 310 or 313, with more colonies stimulating proliferation above background levels. Seventeen out of 176 colonies gave 75% or more of the proliferation induced by the natural sequence, and these were found to code for amino acids; lysine (the natural residue), arginine, leucine, methionine and glutamine. Position 316 is thought to pack against position 74 in the β chain of the Dr4 Dw4 molecule, and to be oriented up and over the HLA-DR β chain helix, imposing a requirement for an unbranched side chain (at least until the γ carbon) In HLA-DR4 Dw4 position 74 is an alanine residue, explaining the requirement for a hydrophobic centre around the β and γ carbon atoms at 316. The fact that several side chain termini are tolerated suggests that the putative contact with the T cell receptor is not too rigorous, though acidic side chains clearly are not tolerated.

It is evident from the above results, that the subject method provides a powerful procedure for screening large numbers of diverse peptide sequences or DNA sequences in determining their physiological activity. Thus, one may screen peptides for binding to a wide variety of receptors, particularly surface membrane protein receptors, where one is interested in determining the agonist or antagonist activity of such peptide and the receptor provides for a detectable signal. All publications and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

The invention now being fully described, it will be apparent to one of ordinary skill in the art that many changes and modifications can be made thereto without departing from the spirit or scope of the appended claims.

Claims

1. A method for screening peptides for physiological activity as a result of peptide binding to a receptor, where binding of a ligand to said receptor can provide a detectable response, said method comprising: a) cloning and expressing a group of fusion proteins, said fusion proteins comprising a fusion partner joined to diverse poly(amino acid) sequences, whereby individual clones are produced, each clone expressing a single fusion protein; b) lysing the individual or combinations of clones to produce a plurality of lysates; c) combining each of said lysates with a binding partner which binds to said fusion partner to produce a high molecular weight cross-linked complex comprising said fusion protein and said binding partner; d) separating the complex from the other components of each of said lysates; and e) screening said fusion proteins for their physiological activity with receptors.

2. A method according to Claim 1, wherein said binding partner comprises immunoglobulin and said fusion partner comprises at least the functional binding portion of protein A.

3. A method according to Claim 1, wherein said binding partner comprises a cell and said fusion partner comprises protein A.

4. A method according to Claim 1, wherein said complex is separated by centrifugation.

5. A method according to Claim 1, wherein said fusion protein is freed from said complex prior to contacting with said receptor.

6. A method of determining the physiological activity of an individual member of a group of related peptides in binding to a surface membrane protein receptor, wherein said receptor provides for a detectable response, said method comprising: a) cloning and expressing a group of fusion proteins comprising a fusion partner and said peptide sequences to produce individual clones, each clone expressing a single fusion protein; b) lysing the individual or combination of clones to produce a plurality of lysates; c) contacting each of said lysates with a binding partner for said fusion partner dispersed in a liquid medium to produce an extended complex; d) separating said complex from the other components of each of said lysates; and e) combining said fusion proteins with said receptor to determine the effect of binding of said fusion protein on said receptor.

7. A method according to Claim 6, wherein said fusion partner is at least the functional portion of protein A and said binding partner is an immune complex.

8. A method according to Claim 6, wherein said surface membrane protein receptor is the T-cell receptor and is present as part of a viable T-cell.

9. A method according to Claim 6, wherein said surface membrane protein receptor is a major histocompatibility complex antigen or combination of major histocompatibility complex antigens.

10. A method according to Claim 6, wherein said related peptides are related by being fragments of the same protein.

11. A method according to Claim 6, wherein said peptides are related by being encoded by a synthetic DNA sequence, wherein a plurality of different nucleotides were added at at least one stage in the synthesis of said DNA sequence.

12. A method according to Claim 11, wherein said surface membrane protein receptor is a major histocompatibility complex antigen, and said DNA sequence comprises sequences encoding a plurality of amino acids at a single site for use in determining the effect on binding affinity to said receptor.

13. A method for screening peptides for physiological activity as a result of peptide binding to a receptor, where binding of a ligand to said receptor can provide a detectable response, said method comprising: a) cloning and expressing a group of fusion proteins comprising at least the functional portion of protein A and the sequences of said related peptides separated by a selectable cleavage site to produce individual clones, each clone expressing a single fusion protein; b) lysing the individual or combination of clones to produce a plurality of lysates; c) contacting each of said lysates with an immune complex, whereby said fusion proteins bind to said immune complex; d) separating the immune complex-fusion protein from the other components of each of said lysates; e) freeing said fusion protein from said immune complex to provide said fusion protein in substantially pure form; f) cleaving said fusion proteins at said selectable cleavage site to provide oligopeptides; g) combining said oligopeptides with cells comprising said receptor; and h) determining the effect of binding of said oligopeptides on said cells.

14. A method according to Claim 13, wherein a plurality of clones are mixed in a step at least prior to said combining to provide a mixture of fused proteins; and prior to said combining, including the additional steps of: a) bringing together said mixture of fusion proteins with cells comprising said receptor; b) determining the effect of binding of said mixture of fusion proteins on said cells; and c) using individual clones from mixtures having an effect on said cells in said combining step.

15. A method according to Claim 13, wherein said related peptides are related by being fragments of the same protein.

16. A method according to Claim 13, wherein said peptides are related by being encoded by a synthetic DNA sequence wherein a plurality of nucleotides were added at successive stages in the synthesis of said DNA sequence.

17. A kit comprising a vector comprising a replication system functional in a prokaryotic host, a marker and in the direction of transcription, an expression cassette comprising a transcriptional and translational initiation region, an open reading frame encoding at least the functional portion of protein A, at least -31-

one restriction site providing for insertion in reading frame with said open reading frame and a transcriptional termination region, a polyepitopic protein and antibodies to said 5 protein.

18. A kit according to Claim 17, wherein one of said protein or antibodies is bound to a surface.

10 19. A kit according to Claim 18, wherein said surface is a microtiter well.

20. A kit according to Claim 16, wherein said surface is a particle.

15

21. A kit according to Claim 15, wherein said open reading frame is joined to a sequence encoding a selectable cleavage site.