IDENTIFICATION OF LIGANDS AND THEIR USE

Abstract

The invention provides carriers and kits for use in the identification of ligands. Carriers of the invention are provided to which is attached a DNA encoding a peptide and a β2 microglobulin. Carriers of the invention are capable of identifying an epitope recognized by a T-cell receptor and/or an NK-cell receptor and/or an NKT-cell receptor. The invention also provides kits for screening for a T-cell epitope and/or an NK-cell epitope and/or an NKT-cell epitope, wherein the kits comprise a carrier of the invention.

Claims

1. A carrier to which is attached a peptide, DNA encoding the peptide, and β2 microglobulin.

2. The carrier of claim 1 to which an MHC or MHC-like molecule is attached.

3. The carrier of claim 1 wherein the β2 microglobulin is attached to the carrier as part of a larger construct which includes the peptide.

4. The carrier of claim 1 wherein the peptide and β2 microglobulin are provided connected by a flexible linker.

5. The carrier of claim 1 wherein the peptide is between about 4 and about 20 amino acids.

6. The carrier of claim 1 wherein the carrier is a solid support.

7. The carrier of claim 1 wherein the carrier is a bead.

8. The carrier of claim 1 wherein the carrier is multivalent.

9. A carrier to which is attached a peptide, DNA encoding the peptide, and β2 microglobulin and/or an MHC or MHC-like molecule.

10. The use of a carrier according to claim 1 to identify an epitope recognised by a T-cell receptor and/or an NK-cell receptor and/or an NKT-cell receptor.

11. A kit for screening for a T-cell epitope and/or an NK-cell epitope and/or an NKT-cell epitope, wherein the kit comprises a carrier to which is attached a peptide, DNA encoding the peptide and β2 microglobulin.

12. The kit of claim 11 further including instructions to fold the peptide and β2 microglobulin.

Description

[0084] Preferred embodiments of the present invention will now be described, merely by way of example, with reference to the following drawings and examples.

[0085] FIG. 1—is a schematic illustration of emulsion PCR and emulsion IVTT using beads, the beads and bound DNA and peptide are then shown being passed over cell surface molecules in a screen for epitopes are recognised the cell surface molecules;

[0086] FIG. 2—shows XbaI restriction digestion of post IVTT bead-DNA-protein complexes. Lane 1-2-log DNA ladder; Lane 2-5′-bio-forward primer attached to the beads; Lane 3—supernatant of (2) after magnet separation; Lane 4-5′-bio-reverse primer attached to the beads; Lane 5—supernatant of (4); Lane 6-5′-bio-for and 5′-bio-rev; Lane 7—supernatant of (6); Lane 8—negative control: 5′-bio-forward primer attached to the beads, no DNA template; Lane 9—supernatant of (8);

[0087] FIG. 3—shows the Western blot result of human beta-2-microglobulin staining on emulsion-bead IVTT. Lane 1—shows emulsion IVTT with 5′-bio-forward primer-bead-protein, DNA template: BZLF1 construct. Lane 2—shows emulsion IVTT with 5′-bio-reverse primer-bead-protein DNA template: BZLF1 construct; Lane 3—shows emulsion IVTT with 5′-bio-forward and reverse primer-bead-protein DNA template: BZLF1 construct; Lane 4 shows emulsion IVTT negative control, no DNA template added.

[0088] FIG. 4—shows Western blot result of hemagglutinin tag staining on emulsion-bead IVTT. Lane 1—shows Emulsion IVTT with 5′-bio-forward primer-bead-protein DNA template: BZLF1 construct; Lane 2—shows Emulsion IVTT with 5′-bio-reverse primer-bead-protein DNA template: BZLF1 construct; Lane 3—shows Emulsion IVTT with 5′-bio-forward primer-bead-protein DNA template: randomised epitope construct; Lane 4—shows Emulsion IVTT with 5′-bio-reverse primer-bead-protein DNA template: randomised epitope construct.

[0089] FIG. 5—illustrates that a peptide, MHC and β2 microglobulin can be correctly folded on a carrier bead. The solid lines show the ELISA optical density (OD) for a random peptide, MHC and β2 microglobulin refolded on a bead, and the dotted lines show the OD for the a fixed BZLF1 peptide, MHC and β2 microglobulin refolded on a bead.

COUPLED PCR AND IN VITRO TRANSCRIPTION/TRANSLATION REACTION IN A BEAD EMULSION

[0090] A DNA template was designed containing the start and stop sequences for in vitro transcription translation (IVTT) surrounding the sequence of human beta-2-microglobulin, a linker and a fixed or random stretch of amino acids. Different approaches to couple the generated protein back to the beads including using a streptavidin binding peptide and using an HA tag were considered. The sequences are described below.

Materials

[0091] The following materials were used in the experiments described herein: Dynabeads® M270 streptavidin (Invitrogen); mineral oil (Sigma); Span-80 (Sigma); Tween-80 (Sigma); Triton X-100 (Sigma); RTS 100 E. coli cell-free transcription/translation high yield system (Roche); expand high fidelity PCR system; dNTP pack (Roche); tris-buffered saline (Invitrogen); H.sub.2O saturated diethyl ether (Sigma); Restriction endonuclease XbaI (New England Biolabs); Antibodies: mouse monoclonal anti-human beta-2-microglobulin (BBM1) (Abcam), Goat polyclonal anti-mouse conjugated Horseradish Peroxidase (Dako), Rat monoclonal anti-hemagglutinin conjugated Horseradish Peroxidase (Roche); Hybond®-C extra nitrocellulose membrane (GE healthcare); NuPAGE® protein system (Invitrogen); NuPAGE® 12% Bis-Tris Midi gel; NuPAGE® MES SDS running buffer; NuPAGE® MES transfer buffer PBST (PBS Tween 20 1:1000); ECL developing reagent (GE healthcare); X-ray film 130 mm×180 mm (Fujifilm).

DNA Sequences

[0092] EBV lytic cycle protein BZLF1 was used as a positive control in a DNA template for coupled emulsion PCT and IVTT. The DNA template was as follows:

TABLE-US-00001 Streptavidin T7 flexible Human binding T7 Promoter BZLF1 linker β2m linker protein linker terminator gatctcgatcccgcgaaattaatacgactcactatagggagaccacaacggtttccctctag    S  T  S  T  E  I  N  T  T  H  Y  R  E  T  T  T  V  S  L  - aaataattttgtttaactttaagaaggagatataccatgcgagcaaaatttaaacaactg  K  —  F  C  L  T  L  R  R  R  Y  T  M  R  A  K  F  K  Q  L ttaggcggtggtggcggagggtctggtgggtctggcggtagtggcggcattcagcgcact  L  G  G  G  G  G  G  S  G  G  S  G  G  S  G  G  I  Q  R  T ccgaaaatccaggtctatagccgtcatcctgcggaaaatgggaagagcaacttcctgaac  P  K  I  Q  V  Y  S  R  H  P  A  E  N  G  K  S  N  F  L  N tgctatgtttcggggtttcatccgtcggacattgaggtagacctgctgaagaacggtgaa  C  Y  V  S  G  F  H  P  S  D  I  E  V  D  L  L  K  N  G  E cgcattgagaaagtggagcacagcgatctcagcttcagtaaagattggtccttttacctg  R  I  E  K  V  E  H  S  D  L  S  F  S  K  D  W  S  F  Y  L ttgtactacacggaatttacgcccacagagaaagacgaatatgcgtgtcgcgtgaatcac  L  Y  Y  T  E  F  T  P  T  E  K  D  E  Y  A  C  R  V  N  H gtaaccctttcccagccgaaaatcgtcaaatgggatcgcgatatgggcggtggaggatca  V  T  L  S  Q  P  K  I  V  K  W  D  R  D  M  G  G  G  G  S ggtggcatggatgaaaagaccaccggttggcgtggcggtcatgtggttgaaggcttagct  G  G  M  D  E  K  T  T  G  W  R  G  G  H  V  V  E  G  L  A ggcgaactggaacagttgcgtgcacgtctggaacatcacccacaaggccaacgcgaaccg  G  E  L  E  Q  L  R  A  R  L  E  H  H  P  Q  G  Q  R  E  P taactaactaacctgcaggcgatccggtaagatccggctgctaacaaagcccgaaaggaa  —  L  T  N  L  Q  A  I  R  —  D  P  A  A  K  N  A  R  K  E gctgagttggctgctgccaccgctgagcaataactagcataaccccttggggcctctaa  A  E  L  A  A  A  T  A  E  Q  —  L  A  —  P  L  G  A  S  K cgggtcttgaggggttttttgctgaaaggaggaactatatccgga  R  V  L  R  G  F  L  L  K  G  G  T  I  S  G Total size = 767 bps
DNA sequence is disclosed as SEQ ID NO: 1 and the corresponding amino acid sequences are disclosed as SEQ ID NOS 2-6, respectively, in order of appearance.

[0093] Base pairs 1-101 are the T7 promoter and associated sequences.

[0094] Base pairs 102-122 encode the fixed peptide BZLF1.

[0095] Base pairs 123-488 encode a flexible linker-B2M-linker.

[0096] Base pairs 489-602 encode a streptavidin binding protein.

[0097] Base pairs 603-767 are a T7 termination sequence and associated sequences.

[0098] A further DNA template using a randomised epitope was also designed. In this template a His-tag was used for protein coupling to the beads. The DNA template was as follows:

TABLE-US-00002 Random T7 epitope flexible Human T7 Promoter 9 a.a linker β2m linker HA tag linker terminator ccatgggatctcgatcccgcgaaattaatacgactcactatagggagaccacaacggtttcc    M  G  S  R  S  R  E  I  N  T  T  H  Y  R  E  T  T  T  V  S ctctagaaataattttgtttaactttaagaaggagatataccatgnnknnknnknnknnk  L  —  K  —  F  C  L  T  L  R  R  R  Y  T  M  X  X  X  X  X nnknnknnknnkggcggtggtggtggtggttctggcggcagtggcgggtcaggcggcatt  X  X  X  X  G  G  G  G  G  G  S  G  G  S  G  G  S  G  G  I caacgtaccccgaaaatccaggtgtatagccgtcatccagctgaaaacggcaaaagcaac  Q  R  T  P  K  I  Q  V  Y  S  R  H  P  A  E  N  G  K  S  N tttctgaactgctatgtaagcgggtttcatccttcggacattgaagtcgatctgctgaag  F  L  N  C  Y  V  S  G  F  H  P  S  D  I  E  V  D  L  L  K aatggggaacgcattgagaaagtggaacacagcgatttgtccttctcgaaagactggtcc  N  G  E  R  I  E  K  V  E  H  S  D  L  S  F  S  K  D  W  S ttctaccttctgtactacacggaatttactcccacagagaaagacgaatatgcatgtcgc  F  Y  L  L  Y  Y  T  E  F  T  P  T  E  K  D  E  Y  A  C  R gttaatcacgtcaccctcagtcagccgaaaatcgtgaaatgggatcgggatatgggtggc  V  N  H  V  T  L  S  Q  P  K  I  V  K  W  D  R  D  M  G  G ggcggatctggtggatacccgtatgatgttccggattatgcgtaactaactaacctgcag  G  G  S  G  G  Y  P  Y  D  V  P  D  Y  A  -  L  T  N  L  Q gcgatccggtaagatccggctgctaacaaagcccgaaaggaagctgagttggctgctgcc  A  I  R  —  D  P  A  A  N  K  A  R  K  E  A  E  L  A  A  A accgctgagcaataactagcataaccccttggggcctctaaacgggtcttgaggggtttt  T  A  E  Q  —  L  A  —  P  L  G  A  S  K  R  V  L  R  G  F ttgctgaaaggaggaactatatccggacccggg  L  L  K  G  G  T  I  S  G  P  G Total size = 695 bps
DNA sequence is disclosed as SEQ ID NO: 7 and the corresponding amino acid sequences are disclosed as SEQ ID NOS 8-9, 4-5, and 10, respectively, in order of appearance.

[0099] Base pairs 1-107 are a T7 promoter and associated sequences.

[0100] Base pairs 108-134 encode a random epitope.

[0101] Base pairs 135-497 encode a flexible linker-β2M-linker.

[0102] Base pairs 498-551 encode a HA tag and linker.

[0103] Base pairs 525-695 are a T7 termination sequence and associated sequences.

Primers

[0104]

TABLE-US-00003 5′-bio-for (SEQ ID NO: 11) 5′-ccatgggatctcgatcccgcgaaatt-3′ 5′-bio-rev (SEQ ID NO: 12) 5′-cccgggtccggatatagttcctcctt-3′ 5′-PCR-for (SEQ ID NO: 11) 5′-ccatgggatctcgatcccgcgaaatt-3′ 5′-PCR-rev (SEQ ID NO: 12) 5′-cccgggtccggatatagttcctcctt-3′

Experimental Methods

Creating Bead-Primers

[0105] Dynabeads® M270 streptavidin beads were washed three times by separating the beads using a magnet, removing the supernatant, re-suspending the beads in binding and washing buffer (10 mM Tris-HCl, 1 mM EDTA, 2M NaCl). The appropriate quantity of biotinylated primers (5′-bio-for only, 5′-bio-rev only, or both) was then added to the re-suspended beads and the beads and primer were incubated on a rotor at room temperature. The amount of primers added was determined by the binding capacity suggested by the bead manufacturer (the maximum binding capacity of 1 mg, or 6-7×10.sup.7 of Dynabeads® M270 streptavidin to single-stranded oligonucleotides is 200 pmol).

[0106] After binding the primers to the beads, the beads are washed three times with binding and washing buffer to remove any unbound primer oligos. The beads with bound primer were then re-suspended in water and stored at 4° C.

[0107] In an alternative embodiment dual biotin labeling is employed at the 5′ end of the primer. Using double the biotin proves significantly stronger in resisting the high temperature cycle during PCR. In addition, Polyethylene Glycol (PEG) spacer may be introduced between the dual biotins and the 5′ end of the primer. This improves the accessibility of the polymerase enzyme in synthesizing the DNA template near the 5′ end.

Emulsion PCR

[0108] The emulsion oil for each PCR reaction was prepared in a universal tube as follows: 475 ul mineral oil, 22.5 ul Span-80, 2.5 ul Tween-80, 0.25 ul Triton X-100. Alternatively, ABIL EM 90 surfactant (Bis-PEG/PPG-14/14 Dimethicone, Cyclopentasiloxane, available from Degussa) may be used, this is more durable in PCR reactions. To equilibrate the emulsion oil, a magnetic stirrer was added to the tube. The tube was then placed on a magnetic spin at ≥1000 rpm in a cold room for 10 min.

[0109] Each aqueous PCR reaction (50 ul) was prepared as follows: approximately 10.sup.6 primer-coupled beads, 200 nM complement unmodified primers (vary according to the type of primers attached to the beads), 50 nM unmodified forward primer, 200 uM PCR grade dNTPs, 1.5 mM MgCl, approximately 50 ng DNA template, and 2.5 unit Expand High Fidelity Enzyme Blend. The water-in-oil emulsion was prepared by slow addition of 50 ul aqueous PCR mixture into the spinning emulsion oil, the mix was left to spin for an additional 10 min. The emulsions were then aliquoted into 100 ul each, and subjected to the following PCR cycles: 94° C. 2 min, 40 cycles of 94° C. 30s, 61° C. 30s, 72° C. 1 min, then 72° C. 7 min, followed by 4° C. incubation.

[0110] After the PCR cycles, the same reactions were pooled together in a new 1.5 ml eppendorf tube. The emulsion was broken by adding 1 ml H.sub.2O-saturated diethyl ether and vortexing for 5s. Alternatively isobutanol may be used. The broken emulsions were then centrifuged at 13,000×g for 5 min at room temperature to recover the beads. The upper solvent phase (sometimes with white aggregates) was discarded. The washing was repeated a further two times. To remove any leftover ether, the tubes were vacuum centrifuged for 5 min at room temperature. (Alternatively, the tubes were left open in the fume hood for 10 min). The beads were re-suspended in 10 ul ddH.sub.2O.

Emulsion In Vitro Transcription/Translation

[0111] The emulsion oil was prepared as described in the emulsion PCR method. Each in vitro transcription/translation reaction (RTS 100 E. coli HY kit, Roche) was prepared as follows: 12 ul E. coli lysate, 10 ul reaction mix, 12 ul amino acids (no methionine), 1 ul methionine, 5 ul reconstitution buffer, H.sub.2O up to 40 ul. 10 ul of the emulsion PCR re-suspended beads was added to 40 ul of in vitro transcription/translation mix to form a 50 ul reaction. The 50 ul reaction was mixed thoroughly by pipetting.

[0112] The 50 ul in vitro transcription/translation mixture was then added to the oil and a spinning emulsion formed. The reaction was incubated at 30° C. for 4 hours on a heat block.

[0113] After 30 minutes the emulsion was broken as described with reference to emulsion PCR. The beads were recovered as described with reference to emulsion PCT and re-suspended in 50 ul ddH.sub.2O.

Testing Amplified Double-Stranded DNA Tethering to the Beads after Emulsion PCT and IVTT

[0114] An XbaI restriction site (TCTAGA) is found at the 57.sup.th base of the BZLF1 construct, and at the 63.sup.rd base of the randomized epitope construct, this site was used to confirm that the double-stranded DNA was tethered to the bead.

[0115] 10 ul of IVTT re-suspended beads were taken and the beads were separated from the supernatant with a magnet. The beads were then re-suspended in fresh ddH.sub.2O. The following restriction digestion was prepared: 1 ul XbaI enzyme (20 units), 0.2 ul bovine serum albumin 100 ug/ml, 2 ul NEBuffer 4 10× (50 mM potassium acetate, 20 mM Tris-acetate, 10 mM magnesium acetate, 1 mM DTT, pH 7.9), 6.8 ul ddH.sub.2O, and added to the newly re-suspended beads. The beads and restriction enzyme mix were incubated at 37° C. overnight. The digest was examined on a 1% agarose gel—FIG. 2.

[0116] In the experiment depicted in FIG. 2 the BZLF1 construct was used as the DNA template for all samples. Therefore, if dsDNA was amplified and attached to the beads, XbaI restriction digest would cleave and leave approximately 700 bp of unbound DNA fragments which can be examined by the gel. Bands at 700 bps could be observed in samples with biotinylated primer-beads and DNA templates, regardless of forward or reverse primers. However, only 5′-biotin-reverse primer-beads gave a clear result (well 4), i.e. only single band at 700 bps. This could be due to the presence of the beads near the restriction recognition site, which leads to inaccurate binding of the enzyme on DNA. The presence of the smears in the supernatants suggest that a substantial amount of bacterial DNA was present in the E. coli lysate.

Examining the Integrity of Translated Proteins—Western Blotting

[0117] 20 ul of the resuspended beads produced according to the above section entitled “Emulsion in vitro transcription/translation” was mixed with 20 ul of Lithium Dodecyl Sulfate (LDS) 4×, and 40 ul of ddH.sub.2O. The reaction was incubated at 75° C. on a heat block for 10 min. The samples were then loaded onto NuPAGE® 12% Bis-Tris Midi gel, which was submerged in a gel tank filled with NuPAGE® MES SDS running buffer. The gel was run at 140V for 1.5 hours then transferred onto Hybond® nitrocellulose membrane (pre-soaked with NuPAGE® transfer buffer) with semi-dry Western blot transfer system for 30 min at 50 mA.

[0118] After transfer, the nitrocellulose membranes were blocked with blocking buffer (3% milk, PBST) for ≥1 hr on a shaker in the cold room.

[0119] The membranes were then stained with mouse monoclonal anti-human beta-2-microglobulin (anti-BBM1), 1:500 in PBST for 1 hr on a shaker at room temperature. (For staining with hemagglutinin tag, rat monoclonal anti-hemagglutinin conjugated to HRP was used at 1:500 dilution). After the staining the membranes were washed for 5 min with PBST three times. Then the membranes were stained with secondary antibody goat anti-mouse polyclonal antibody conjugated to horseradish peroxidase, 1:1000 in PBST for 1 hr on a shaker, room temperature. The membranes were washed again and developed by ECL developing reagent, filmed on X-ray films.

[0120] In the results, shown in FIG. 3, the EBV protein (BZLF1 positive control) was translated in bead-emulsion IVTT and stained for human beta-2-microglobulin. The estimated size of this protein is 18.38 kDa, which correlates to the position of the bands in sample 1, 2 and 3. The extra bands on the top of the wells are due to aggregation and impaired migration of the bead-protein-DNA complexes.

[0121] In the results shown in FIG. 4 the EBV protein (BZLF1 positive control) and the randomised epitope protein construct were translated in bead-emulsion and stained for the hemagglutinin tag. Since the BZLF1 positive control sequence did not code the hemagglutinin sequence, the staining results were negative (sample 1 and 2). The estimated size of the randomised epitope protein was 15 kDa, which correlated to the position of the bands at sample 3 and 4.

MHC Class I Refolding of Bead-DNA-Protein Complexes

[0122] The experiments described above demonstrate that emulsion PCR, followed by emulsion disruption, followed by emulsion formation, followed by emulsion IVTT is achievable on a bead. The successful translation of human beta-2-microglobulin suggests that the 9 amino-acid random epitope can be translated successfully as well.

[0123] In order to rapidly screen the epitopes which bind to known MHC heavy chains, a refolding reaction was carried out. The correctly refolded bead-MHC-peptide complex was screened by ELISA.

Refolding of the Peptide, the β2 Microglobulin and the HLA

[0124] To prepare the refolding mixture the bead-proteins were collected from the bead-emulsion IVTT reaction, with a typical volume of 50 ul. The refolding buffer was prepared as follows: for 1 L refolding buffer, 100 ml 1M Tris pH8.4 ml 0.5M EDTA, 84.28 g L-arginine HCl, 1.54 g reduced glutathione, 0.31 g oxidised glutathione. The buffer is equilibrated with a magnetic stirrer.

[0125] 1 ml of refolding buffer was placed in a 1.5 ml eppendorf tube in the cold room, the tubes were fixed on a rotor. 50 ul of IVTT product was added to the 1 ml of refolding buffer. 10 ul of HLA heavy chain (stock 30 mg/ml) was slowly added into refolding buffer. For BZLF1 positive control, HLA-B8 is added (recombinant HLA-B*0801 synthesized in house). The refolding mixture and bead-protein and HLA was then left on the rotor overnight in the cold room. To reduce the amount of non-specific aggregates, the refolding mixture is centrifuged at 3000 rpm, for 5 min. The supernatant is then removed and concentrated by passing through a 15 ml centrifuge filter unit (Millipore). The tubes are spun at 1500 rpm for 5 min, to concentrate the volume to approximately 100 ul. The concentrate can then be stored at 4° C. The refolding mixture was then centrifuged at 4000 rpm for 20 min. The supernatant was removed, to remove aggregates. A millipore 15 ml concentrator was then used to concentrate the 1 ml refolding mixture down to approximate 200 ul. 100 ul of the sample was loaded onto an ELISA well pre-coated with W6/32—an antibody that will only recognise correctly refolded HLA and β2 microglobulin (mouse monoclonal anti-HLA-A/B/C from eBiosciences), 1:300, and blocked with 3% BSA. The plate was incubated for 1 hr at 37° C. The plate was then washed and the wells loaded with rat monoclonal anti-b2m (1:1000). The plates were then incubated for 2 hours at RT. The plates were then washed with excess PBS by gentle aspiration in the wells, before adding 100 ul goat anti-rat polyclonal antibody conjugated to horseradish peroxidase (Abcam) at 1:1000 dilution. The plates were incubated for 1 hr at room temperature and washed again. 100 ul Tetramethyl Benzidine chromogen (TMB) solution (Invitrogen) was added to each well, and 5-10 min was waited for the colour change. 100 ul ELISA stop solution (Invitrogen) was then added to each well. Absorbance readings for each well at 450 nm was then collected by ELISA plate reader.

[0126] The results in FIG. 5 demonstrate that the peptide, HLA and β2 microglobulin are correctly folded on the bead.

Identification of Epitopes for T-Cell Receptors and/or NK-Cell Receptors.

[0127] Beads produced as described above by emulsion PCR and IVTT displaying a peptide, known or random, the encoding DNA, β2 microglobulin and MHC or MHC-like molecule on their surface are applied to T-cells and/or NK-cells in suspension or attached to a support. The beads and cells are incubated for about 20 minutes at 37° C. Unbound beads are washed off by using size exclusion filtration or alternative approaches including differential magnetic migration of unbound beads compared to beads attached to cells. The DNA is then purified from the cells and beads. The recovered DNA is then amplified by PCR using primers designed to amplify the DNA encoding the peptide epitope. The amplified DNA is then sequenced. As the sequences are detected by PCR and sequencing, only a few cells are required to test the specificity. Visualisation of reactive cells using techniques such as flow cytometry is an alternative but requires larger numbers of cells.

[0128] The T-cells or NK-cells used in the method of the invention may comprise part of the cell mix in a homogenised tissue sample, purification of the T-cells or NK-cells from the sample is not necessary. The method allows a very small number of cells to be used, this is important if there is only a limited sample, such as a biopsy sample. Alternatively, the T-cells or NK-cells may be in a blood sample, for example, a blood sample from an individual infected with a pathogen such as staphylococcus, dengue or influenza. By using tissue samples from individuals with a known condition or infection the method of the invention will allow more to be learnt about disease pathogenesis and will also allow potential drug targets or biomarkers to be identified.

[0129] The peptide library used could be random or could be tailored for a particular application. For example, if looking a dengue virus infection, the peptide library could be based on known dengue proteins.

Identification of Peptides from a Bead-Protein-DNA Library which Recognise HLA-A*0201-Restricted GILGFVFTL-Influenza Matrix Specific T-Cell Clones (“GILGFVFTL” Disclosed as SEQ ID NO: 13)

[0130] A2-specific T-cell clones (HLA-A*0201-restricted GILGFVFTL-influenza matrix specific T-cell clones (“GILGFVFTL” disclosed as SEQ ID NO: 13)) were washed with RPMI 1640 medium and re-suspended at 10.sup.4 cells in 400 ul RPMI 1640 medium for each reaction. 50 ul of a protein-DNA-bead resuspension was incubated with 10.sup.4 T-cell clones for 30 min at 37° C. The protein-DNA-beads used were as described earlier and comprised beads with DNA encoding a random peptide and beta-2-microglobulin attached thereto, which has been transcribed and translated in vitro on the beads to express the peptide and the beta-2-microglobulin which are anchored to the bead, HLA-A2 heavy chain, which is required to correctly present the peptide, was added exogenously. The peptide library size was in the order of 10exp7.

[0131] Each reaction was then transferred to a well of a 24-well Millipore® Millicell® filter culture plate, and the plate was spun at 1500 rpm, for 90 seconds. Unbound beads were washed through the filter, and the remaining cells were resuspended in ddH.sub.2O and the reaction was transferred onto ice. Depending on the rate of reaction, if a large number of beads were internalized by the cells due to T-cell receptor-MHC complex interaction, cell lysis treatment was required to harvest the beads. If the beads were attached on the cell surface, and not internalised, the DNA could be amplified directly.

[0132] Standard PCR amplification (94° C. 30 sec, 50° C. 30 sec, 72° C. 30 sec, 35 cycles) with primers flanking the randomized protein region was used. PCR samples were purified using the PCR purification kit (Qiagen). The purified DNA samples were then amplified by extension PCR with a 5′ primer adding an additional 4 bases (CACC) on the 5′ end of the DNA template. This was needed for directional TOPO cloning. The PCR samples were purified again with the PCR purification kit (Qiagen). A directional TOPO cloning reaction (pENTR Directional TOPO Cloning Kits, Invitrogen) was performed as described by the manufacturer's instructions. The TOPO cloned reactions were plated on kanamycin supplemented agar plates, and incubated at 37° C. overnight. The next day 20-50 colonies were picked and expanded by incubating the bacteria in 5 ml LB culture supplemented with kanamycin at 37° C. with shaking until the OD600 is 0.5 or above. Plasmids were then extracted using the Qiagen Miniprep Kit according to the manufacturer's instructions. The recovered plasmids were then sequenced to understand the exact protein sequence which had interacted with the A2-specific T-cell clones.

[0133] The results of the sequencing showed that nine sequences were recovered. In this example conventional sequencing was used, but if high throughput sequencing was used many millions of sequences could be examined.

[0134] The sequences obtained using conventional cloning and sequencing in this example were:

TABLE-US-00004 1. (SEQ ID NO: 14) Q V xxxxxx R 2. (SEQ ID NO: 15) A I xxxxxx I 3. (SEQ ID NO: 16) M A xxxxxx W 4. (SEQ ID NO: 17) L A xxxxxxx 5. (SEQ ID NO: 18) Q R xxxxxx A 6. (SEQ ID NO: 19) M A xxxxxxxx R 7. (SEQ ID NO: 20) S S xxxxxx V 8. (SEQ ID NO: 21) G I xxxxxx L 9. (SEQ ID NO: 22) C L xxxxxx D

[0135] Sequence 8 shows identical anchor residues to the influenza matrix peptide GILGFVFTL (SEQ ID NO: 13). Therefore 1/9 of the sequences potentially identified the epitope recognised by the specific T cells after just a single round of selection. Using large numbers of sequences generated with high throughput sequencing approaches, it will be possible to use bioinformatics to identify relevant sequences, for example through comparing sequences obtained from control cells. Alternative strategies may be to use varying times and temperatures of incubation of the beads with the cells and to use blocking strategies such as control beads (with known defined sequences) and/or using anti-CD8 to limit CD8/bead association. In addition, it is possible to re-derive the beads using the sequences obtained from round 1 of selection in order to undertake further rounds of selection to identify relevant sequences.