CZ20031044A3 - A method for diagnosing infectious agents using antigen mimetics - Google Patents

Abstract

The method consists in identifying the binding specificity of antigen and antibody molecules in serum by the method of antibody detection by antigen mimetics (ADAM), consisting in screening a phage library using serum from patients infected with the antigen and uninfected individuals by identifying antibodies binding peptides (ligands) specifically associated with the said antigen. The method is improved by in vitro maturation of ligands and binding of ligands to a common structure (core), which are, for example, MAP. A method for diagnosing hepatitis C (HCV) consisting in screening a phage library using serum from HCV patients and uninfected individuals by identifying antibodies that bind ligands specific for HCV infection.

Description

Technical area

The present invention relates to a diagnostic test for the detection of infectious agents, in particular viral agents, and in particular human hepatitis C virus (hereinafter referred to as HCV), peptides binding antibodies to infectious agents suitable for the test, and a process for preparing said peptides and kits for performing said test. The present invention relates in particular to peptides binding antibodies to anti-HCV suitable for the diagnostic test, a process for preparing them and kits for performing said test.

State of the art

Hepatitis C virus (HCV) is the main etiological agent of parenterally transmitted non-A, non-B hepatitis. This virus very often causes persistent infection and often leads to the development of chronic hepatitis and cirrhosis of the liver, thus creating a major widespread cause of chronic liver disease (Boyer, N., Marcellin, P.: Pathogenesis, diagnosis and management of hepatitis C, J. Hepatol., 32,98-112, 2000).

Viral disease is diagnosed by detecting antibodies to HCV in serum or by detecting viral RNA by nucleic acid amplification methods (Robbins, D. J., Pasupuleti, V,, Cuan, J., Chiang, C.S.: Reverse transcriptase PCR quantification of hepatitis C virus, Clin. Lab. Sci., 13, 23 - 30, Winter 2000). These latter methods are very sensitive, but expensive and

- 2 prone to technical or laboratory error. In addition, the reliability and specificity of the polymerase chain reaction (PCR) technique are not standardized (Gretch, D. R.: Diagnostic tests for hepatitis C, Hepatology, 26, (3, Suppk 1), 43S-47S, September 1997).

The present invention relates to the diagnosis of infectious diseases such as viral infections, in particular HCV infections, by detecting antibodies produced against the infectious agent, in particular antibodies against HCV.

Very recently, a system for sensitive detection of HCV core protein has been reported (Komatsu, F., Takasaki, K.: Liver, 19, (5), 375-380, October 1999).

Therefore, a significant amount of patent and non-patent literature is devoted to the topic of diagnostic methods and kits for HCV detection.

U.S. Patent 5,985,542 assigned to Sumitomo Chemical Company provides a diagnostic kit for liver diseases such as hepatitis C and alcoholic cirrhosis that contains antibodies capable of recognizing cytochrome P450.

US Patent 5,972,347 assigned to Baxter Aktiengesellschaft provides a composition for use against HCV infection comprising inactivated HCV bound to neutralizing human antibodies against HCV, wherein the antibodies are active against at least one • · • «··

- 3 protein selected from the group consisting of the HCV core protein and the NS3 protein.

US Patents 5,939,262; 5,919,625 and 5,677,124 assigned to Ambion and Cenetron provide a very broad method for determining the presence of a nucleic acid sequence of interest in a sample, characterized by obtaining a (ribo)nuclease-resistant segment of the (ribo)nucleic acid.

US Patent 5,866,139 assigned to Institut Pasteur includes a kit for determining the presence of specific antibodies to HCV E-1. See also US Patent 5,854,001 assigned to Abbott.

U.S. Patent 5,800,982 assigned to Tonen describes antigenic peptides capable of specifically reacting with antibodies directed against HCV group II. JP 10019897 to Tonen provides a kit comprising a peptide having an amino acid sequence of at least five contiguous amino acids constituting the nonstructural region 4A (NS4A) protein of HCV and a peptide having an amino acid sequence of at least five contiguous amino acids constituting the nonstructural region 4B (NS4B) protein of HCV.

For an overview of the state of the art, see also US patents US 5,871,904; US 5,869,253; US 5,750,331; US 5,747,251; US 5,667,992; US 5,645,983; US 5,625,034; US 5,610,009 and patents WO 9707400, WO 9637783; EP 593291; EP 593290; EP 586065 and JP 4349885.

- 4 Detection of HCV antibodies in serum therefore remains the most widely used test for the evaluation of HCV infection. Since 1987, when the HCV genome was cloned, various improved versions of HCV diagnostic tests have been developed. Currently available diagnostic kits detect antibodies in serum against a mixture of four recombinant antigens corresponding to non-structural region 3, non-structural region 4 and non-structural region 5 of the HCV core polypeptide (Gretch, D. R.: Diagnostic tests for hepatitis C, Hepatology, 26, (3, Suppl. 1), 43S -47S, September 1997). Samples that are evaluated as positive by this ELISA procedure must be confirmed by an immunoblot test, where the presence of antibodies against the same four antigens is separately detected. A positive diagnosis requires the detection of antibodies against at least two recombinant antigens.

A significant proportion of the HCV antibody population has been found to be reactive to only one antigen, making it impossible to formulate a definitive diagnosis. Another problem arising from the use of recombinant antigens is that they may be detected by antibodies that do not correspond to HCV, creating a false positive diagnosis. This then requires additional testing and long-term follow-up of the patient to achieve a definitive diagnosis.

The contact of a foreign antigen with the organism activates a specific immune response. The antigenic pattern detected by the humoral response is defined by the epitopes recognized by the host antibodies. Provided that the antibody binding site is a negative pattern of the epitope - the substance that specifically binds the paratope - it can • 4 • 444

- 5 represent a positive image of the epitope. This line of reasoning suggests that, with respect to the humoral response, an antigen can be explicitly described by specific ligands that bind antigen-specific antibodies. Identification of these ligands offers a way to detect a specific humoral response to an antigen, regardless of whether the same antigen is known or available or not.

This concept has been used to develop diagnostics for the detection of antibodies associated with infection, including HCV infection in humans. Screening phage libraries with HCV-positive sera identified peptide ligands that specifically bind anti-HCV antibodies. This was achieved by adjusting the selection strategy because sera from infected patients contained virus-specific antibodies that were prevalent in the population, along with a large number of other antibodies with different binding specificities.

Screening of peptides with a large number of negative sera eliminated those ligands that detected antibodies in negative samples (false positives), which led to the selection of a set of highly selective peptides. The occurrence of non-specific reactivities was also limited by the use of short peptide sequences unrelated to natural antigens. This technique is covered in EP 0 698 091 B1, published 28.2.1996. See also J. Mol. Biol., 222, 301 - 310, 1991; Gene, 128, 51 - 57, 1993; Gene, 148, 7 - 13, 1994; The EMBO Journal, 13, (9), 2236 - 2243, 1994; Bacterial Protein Toxins, Zb. Bakt. Suppl., 24, 415-425, 1994; The Journal of tt · • ···

- 6 Immunology, 4504 - 4513, 1996; Methods in Molecular Biology, vol. 87, pp. 195 - 208, Humana Press lne.; Combinatorial Libraries (Cortese, R., ed.) Gruyter, Walter de, chap. 8, 1996; Methods in Enzymology, 267, 116 - 129, 1996; Biol. Chem., 378, 495 - 502, June 1997; The EMBO Journal, 17, (13), 3521 - 3533, 1998; Nature Biotechnology, 16, 1068 - 1072, November 1998; a review of mimotope and phagotope technology is given in the above-mentioned EP 0 698 091 and methods of implementation.

These selection strategies identify those peptide structures that best bind immunodominant antibodies against HCV. In addition, multivalent ligand unfolding provided a significant avidity effect on the sensitivity of detection. Despite these assumptions, testing of the ADAM-HCV (ADAM = Antibody Detection by Antigen Mimics) mixture on a panel of sera that were not used for screening resulted in a sensitivity of less than 100%, although the sensitivity was relatively high. This result may be explained by the peculiar characteristics of the humoral response to HCV antibodies, as it does not target a major immunodominant epitope, but rather is directed against a large number of different viral determinants. This is further complicated by the existence of different viral genotypes.

Commonly available diagnostic kits have serum antibodies against four recombinant antigens corresponding to broad regions of HCV polypeptides (Gretch, D. R.: Diagnostic tests for hepatitis C, Hepatology, 26, (3, Supplement 1), 43S-47S, September 1997).

* 4 ·· ·· 44 »4 44 4»

- 7 A positive diagnosis requires the detection of serum antibodies against at least two recombinant antigens: for this reason, it is impossible to formulate a definitive diagnosis for a corresponding part of the population with antibodies to HCV, since specific reactivity with only one antigen can be detected. The ADAM - HCV EIA (Enzymatic Immuno Assay) uses the mimetics of many different immunodominant epitopes from different antigens, thus ensuring a higher resolution of the analysis. The immediate result can be a drastic reduction in the frequency of indeterminate samples.

We have devoted considerable effort to the technical development of the HCV ADAM assay (using the strategy essentially used in the above-mentioned EP 0 698 091). Peptides selected from phage libraries are shown to be N-terminal fusions with the major capsid protein pVIII. The use of phage as a diagnostic agent has many advantages: it is an extremely flexible reagent, which is already predestined for many types of immunological assays (Dente, et al.: 1994; Felici, F., Galfré, G., Luzzago, A., Monaci, P„ Nicosia, A and Cortese, R.: “Phage-displayed peptides as tools for the characterization of human sera”, Methods in Enzymology, 267, 116 - 129, 1996; Bartoli, et al.: Nature Biotechnology, 16, 1068 1073, November 1998), and its small-scale production is easy and inexpensive.

Despite these advantages, the size of the phage particle limits the concentration of peptide molecules that can be achieved in the assay. Interference of serum antibodies with the phage capsid requires the addition of carrier phage to the assay mixture to separate the antibodies • « • ί ·Μ

- 8 against phage. Large-scale production, if it were to occur at all, faces problems of microbial contamination of the reagent, reproducibility, quality control, purification, and production costs.

Linear synthetic peptides derived from peptide sequences that appear on phage will in most cases not provide either the sensitivity or specificity to achieve sufficiently efficient antibody detection.

The essence of the invention

It has now been found - and this is the subject of the present invention - a method for diagnosing such infectious diseases as viral infections, in particular hepatitis C infection, consisting in identifying the binding specificity of antigen-antibody molecules in serum by the method of antibody detection by antigenic mimetics (ADAM) involving screening of phage libraries using sera from patients infected with the antigen and uninfected persons, by identifying antibodies binding peptides (ligands) specifically associated with the said antigen.

In the preferred embodiment of the present invention, this relates to HCV infection,

Detailed description of the invention

In a first solution, which is preferred in the method of the present invention, said ligands are improved by in vitro maturation.

« · « ···

- 9 In a second solution, which is preferred in the method according to the present invention, the said ligands are synthetic peptides.

In a third embodiment, said ligands are linked to a common core. In a particularly preferred embodiment, said ligands are together with said common core MAP as defined below.

According to yet another aspect of the present invention, a set of HCV specific ligands can be obtained by a process comprising:

a) first screening of the phage library with n positive sera to create a first series of n-phage pools;

b) preparing n pool mixtures containing n-1 pools;

c) affinity selection of each of the n-mixtures against the serum that created the selected phage pool to obtain a second series of n-phage pools, and optionally

d) further screening each of the n-phage pools from the second series for a mixture composed of all n-original sera, except those used for the first screening;

e) immunoscreening the resulting second series of n-phage pools using a mixture of all n-origin sera to obtain positive clones;

f) testing the individual reactivities of all positive clones with a panel of positive and negative sera using a designated set of named clones as phage-secreting colonies;

g) creating replicas of said phage-secreting colonies;

• · • ·»·

- 10 h) screening each replicate for its reactivity with positive and negative serum to identify clones specifically reacting with positive serum;

i) using each of said specifically reacting phages as a ligand for affinity purification of antibodies from positive serum;

j) testing said antibodies for their reactivity with previously identified HCV peptides;

k) unification of clones detecting serum antibodies.

Individual peptides resulting from the above-mentioned process are also a subject of the present invention.

Another particular object of the present invention applied to HCV is a solution using the following peptides:

- YSREQLNKLFGIDMT;

- YSREQLNKMFGIEIS;

- YSREQLSKLFGIEPM;

- NSRWLSKAHGIEGM;

- YSREQLNKLFGIEVM;

- YSREQLSKLFGIDTQ;

- KSREQLSKLHGVDTS;

- RSREQLSKLFGIDLT;

- MWRTWLMKTHGIESW;

- MLRTWLMKYQGIESW;

- YSRSWLMKAHGLELG;

- MMRSYLMKAHGIESL;

- MSRLWLMKAHGISSE;

· * • · ···

- 11 • 44* 4444 4 4*4

44 44 4* *4 4·

- KHSEWLNKARGIESW;

- MSRTFLMKAHGIESW;

- MSRTWLMKAHGIESW;

- AEGEKKLRRSTNWGD PAK;

- ABGEFKTRRQTNYQDPAK;

- AEGEFKTLRNANRLD PAK;

- AEGEFKTLRNSNRLDPAK;

- AEGEFKKFPGSSTPKDPAKAAFDSL;

- AEGEFPQDARIFPGGGDPAKAAFDSL;

- PQDARFPGGGDPAKAAFDSL;

- AE GEFKGAGGAQTVD WALLVD THEN;

- AEGEFMGKHFGGAQWIMGDPAK;

- AEGEPLSLKGSGGGQLRALVDPAK;

- AEGEFLSLKGSGGAQLRALVDPAK;

- AEGEFYLLKRSSPPDPAKAAFDSL;

- AEGEFPILVGPYLLPRRSREEAVDPAK;

- AEGEFPILVGPYLLPRRSRBEAVDPAKGK;

- AEGEFRLGVRAPRKALDPAK;

- AE GEFRLG VRALRKALD PAK;

- AEGEFRLGVPALRKAPDPAK;

- RLGVRALRKAPDPAK;

- AEGEWFQPRGHSYQDPAK;

- AEGEFLKERAEM SARKTLGADPAK;

- AEGEFFYQIPRRMETKYGD PAK;

« « · • · ··· • · « · » · · · · * · · ·· ·· ·« ·· ·· ··

- 12 AEGEFSREQLNKLFGIEGDPAK;

AEGEFNSREWLSKAHGIEGMDPAK;

AEGEFRSREQLSKLFGIDLTDPAK;

AEGEFYSRKQLNKLFGIDMTDPAK;

AEGEFYSREQLNKMFGIETSDPAK;

AEGEFYSREQLNKLFGIEVMDPAK.

AEGEFKSREQLRKLHGPDTSDPAK;

AIEGEFKMRNYLNKAFGIEGMDPAK;

AEGEFRSRBQLSKLFGIELTDPAK;

AEGEFSRREYSNKAFGIETQDPAK;

AEGEFRRREYLNKAFGIEGDPAK;

AEGEFSRREWLNKRFGIEYLDPAK;

AEGEFMSRTWLMKAHGIESWDPAK;

AEGEFYSPEWLNKARGIDRSDPAK;

AEGEFKSREQLSKLHGVDTSDPAK;

AEGEFYSREQLNKMFGIEISDPAK;

AEGEFYSRSWLMKAHGLELGDPAK.

AEGEFMMRSYLMKAHGIESLDPAK;

AEGEFMSRLWLMIKAHGISSEDPAK;

AEGEFPQPQEVHVYREQLGLDPAKAAFDSL;

AEGEFGEVLYRGFDEVGGDPAKAAFDSL;

AGEPYVIERGMQDPAK;

AEGEFTTASPRHFLVPLDPAKAAFDSL;

AEGEFTTASPAHFLVPLDPAKAAFDSL;

• in ·

• ·

- 13 AEGEFTTASPSHFLVPLDPAKAAFDSL;

AEGEFATAPPRHYSWDPAK;

AEGEFATAPPAHYSWDPAK;

AEGEFATAPPSHYSWDPAK;

AEGEFRFWKVPDYDPPAAGGDPAK;

AEGEFTESSVSSTLADLASKTFGSADPAK;

AEGEFTLADLATMTFGSTDPAK;

AEGEFGLADLATLTFGSPDPAK;

For the set of the present invention, the preferred phage library of step a) is the pVIll-12aa library.

For the set of the present invention, the inserts of the preferred clone singlets from step k) have the following sequences:

1. SREQLNKLFGIEG;

2. RATLSNEHGITIG;

3. DQRENWFKYHGFG;

4. EWRRYMSDIHGYG;

5. DSLRYMYVMPGFG.

In the present invention, it is preferred to create a phage library in which the reacting clones, preferably the best reacting clones, are partially mutagenized such that each amino acid of the clone sequence is independently substituted with another amino acid.

In the set according to the present invention, the named clones preferably have the following inserted sequence: SREQLNKLFGIEG.

• · · • · ··« * · · · »·»· « ·· ·· · ·· ·· ··

- 14 In the set of the present invention, a random sequence may be flanked by two cysteine residues in the phage library. This aspect can be used in general in the use of the present invention and is not limited to the case of HCV.

The object of the present invention is the use of the above-mentioned kit for the preparation of a diagnostic test for the detection of infectious agents such as viruses, in particular HCV, in subjects suspected of being affected by said infectious agents such as viruses, in particular HCV.

The subject of the present invention is a kit for diagnostic purposes comprising the above-mentioned set.

The present invention provides immunogenic peptides obtainable from the process described herein, both as a set and as individual peptides. Said peptides are suitable as immunogens and therefore suitable for the preparation of vaccines, in particular HCV vaccines. The peptides are suitable in the form of the above-mentioned kit.

Compared to the antigen-based system, the diagnostic procedure we tested has an advantage of built-in scalability: for this case, selections can be made with those sera for which no response was detected. A similar methodology can be used to identify a peptide panel that discriminates between different HCV genotypes, thus replacing expensive and laborious PCR methods with cheaper and faster EIAs.

- 15 4 4 4 · 9 * 4 « 4 · · * • 4 4 4 44·· 9 44 4

44 94 94 44 4«

Design examples

The present invention will be further described in more detail with reference to examples and figures, where:

Figure 1 presents the characterization of the clones obtained by screening. The amino acid sequences of the selected clones are given in single-letter codes. The upper and lower panels show the sequences obtained from the original or secondary library. The pVIII sequences surrounding the foreign epitope are (NH ₂ ) AEGEF[foreign epitope]DPAK. The gray boxes indicate the residues more frequently present at any position in the original library clone. The residues contributing to the sequence identity of the peptides from the secondary library are indicated in bold. The data shown are the average values of two independently performed experiments and indicate the differences in absorbance (A = A _{450 nm} - A ₆₂₀ nm) measured on the marked phage and measured on the wt phage pC89 (Felici, et al., 1991). * indicates untested sera.

Figure 2 shows the identification of phage mimicking the same antigenic determinant. The affinity of antibodies purified with positive C65 serum from clones PA1, PA3, PA8, PA12 or P18 (listed in the first column) were tested by ELISA for reactivity against the same phage (listed in the first row).

The data given are the average values of two independent tests and indicate the difference in absorbance (A = A^o _nm - A ₆₂ o • # · • t ··· • ·

- 16 • * · · · * * · · · ♦ ♦ ·· ·· ·· φ« · «· _nm ) measured on tagged phage and measured on wt phage pC89 (Felici, et al., 1991).

Figure 3 shows the pool reactivity obtained by ELISA from the selection of the pVIIIA12 library. The phage pool derived from the screening of the pVIIIA library for positive sera C76 was tested by ELISA for reactivity with positive sera C12, C13, C29, C40, C47, C65, C73, C74, C76, C83 and C85. The white, grey and black bars represent the reactivity of wt phage, the pVIIIA12 library and the p76 pool. The ELISA results are expressed as A = A405 nm - _A620 nm. The values reported are the average of two independent experiments.

Figure 4 shows the reactivity of the ADAMHCV mixture with serum in diagram A. The cut-off value (CO = 0.232) was calculated as CO = N + 5σ , where N is the mean and σ is the standard deviation of the values obtained with the negative serum. Diagram B is the ADAM-HCV EIA on a panel of sera obtained from the Italian Red Cross. The cut-off value (CO = 0.252) was calculated as CO = N + 5σ , where N is the mean and σ is the standard deviation of the values obtained from five negative control sera. The ELISA assay, as described in the Materials and Methods section, detected the binding of the peptides to antibodies present in human sera. The average values of two independent experiments were pooled. The results are expressed as the ratio between the measured signal and the cut-off value (S/CO). The number of sera tested in each group of sera is also indicated.

- 17 Figure 5. ADAM/HCV EIA on a panel of unspecified sera. The leftmost column lists the names of the HCV peptides tested, grouped by their binding specificity. The following four columns list the reactivities of the listed peptides with positive (c25 and r15) and negative (r6 and r13) control sera. Each subsequent column lists the reactivities of the listed peptides with unspecified serum. Binding of antibodies to HCV peptides present in human serum was detected by ELISA as described in the Materials and Methods section. Mean values from two independent experiments were determined. For each peptide, the cutoff value was calculated as CO = N + 5σ , where N is the mean and σ is the standard deviation of the values obtained from 31 negative control sera. Results are expressed as the ratio between the measured signal and the cutoff value (S/CO).

Figure 6. ADAM-HCV/SIA on positive, negative and undetermined sera. According to specific binding, the following ADAM-HCV peptides were grouped and immobilized on a nylon membrane to form 10 bands: m1909.22 and m1913.2 (A); m1901.31, m3322.3, m3362.3 (B); m1977.1 (C); m3551.3 (D); m3566.3 (E); m858, mF78 and mH1 (F); mA12.1, mA12.2 and mA12.12 (G); mB11.17 (H); mG21.2 (I); m1929A3.1, m1929C3.4 and m1929.21 (J). An additional added line containing purified human IgG serves as an internal positive control. Antibody binding to HCV peptides present in human serum was detected as described in the Materials and Methods section.

- 18 4«*· 4 · 4 4 4 4 · 4

4· 44 4* ·· ·· ··

The best solution according to the present invention are composite antigenic peptides (MAPs) in which the C-terminus of eight identical peptide sequences is linked to a common core (Tam, J. P.: Synthetic peptide vaccine design: Synthesis and properties of high-density multiple antigenic peptide system. Proc. Natl. Acad. Sci., 85, 5409-5413, 1988), which were synthesized using substances having a level of specificity comparable to peptides derived from phages. Removal of the phage backbone allows for a higher peptide concentration of immobilized ligand, resulting in greater sensitivity.

The HCV-specific ligand panel obtainable according to the present invention comprises peptides mimicking the HCV NS3 immunodominant HCV antigen. In the preferred embodiment, this is, to our knowledge, the first described short peptide that mimics the immunodominant NS3 determinant. Peptide scanning analysis of the NS3 protein shows longer sequences that rarely react with positive sera (Khudyakov, Y., et al.: Virology, 206, 666-672,1995).

Other approaches, on the other hand (Santini, C., Brenan, D., Mennuni, C., Hoess, R. H., Nicosia, A., Cortese, R., Luzzago, A.: Efficient display of an HCV cDNA expression library as C-terminal fusion to the capsid protein D of bacteriophage lambda., J. Mol. Biol., 11, 282, (1), 125 - 135, September 1998; Pereboeva: J. Med. Virol., 56, (2), 105-111, October 1998) isolated large protein domains that were * · · · fe » · » · · « • · * · ···« ···· *· ·· ·♦ ·· ®*

- 19 retained antigenic properties. The procedure we developed therefore has the essential properties of not relying on the use or information about HCV antigens, but can also be used in systems where the antigen is unknown, thus creating a procedure that leads to antigen discovery.

Phage libraries of random peptides represent a powerful tool for identifying ligands that are specifically recognized by serum antibodies against HCV. Phage-derived peptides have a great mimicking potential, as they are able to mimic linear, conformational and non-protein epitopes (for a review see Felici, F., Luzzago, A., Monaci, P., Nicosia, A., Sollazzo, M., Traboni, C.: Peptide and protein display on the surface of filamentous bacteriophage, Biotechnol. Annu. Rev., 1,149 - 183, 1995; Zwick, M. B., Shen, J., Scott, J. K.: Phage-displayed peptide libraries, Curr. Opin. Biotechnol., 9, (4), 427 -436, August 1998).

The present invention provides the identification of a broader set of effective HCV-specific ligands and the development of new types of diagnostic kits for the detection of HCV antibodies in serum.

The peptides used in the present invention are suitable as diagnostics or as immunogens or vaccines.

Pharmaceutical preparations containing the immunogens and vaccines of the present invention are conventionally prepared in a manner known in the art.

- 20 experts having ordinary skill in the art. They can be prepared, for example, as described in EP 0 698 091.

The following examples will further clarify the invention.

Example

Identification of HCV-specific ligands with novel specificities

The pVIII-12aa phage library was plated on 8 positive sera (C13, C14, C27, C29, C40, C47, C62 and C65) to generate 8 phage pools (designated p13', p14', p27', p29', p40', p47', p62' and p65'). Eight mixtures were prepared, each containing a different combination of seven of these eight phage pools. Each mixture was affinity selected using serum that generated a separate phage pool: e.g., mixp65 consisting of pools p13', p14', p27', p29', p40', p47' and p62' was plated on serum C65. Each of the eight resulting phage pools (designated p13, p14, etc.) was individually replated on a mixture of all the original sera except the one used for the previous selection. For example, pool ρ13 ¹¹ was plated on a mixture of sera C14, C27, C29, C40, C47, C62, and C65.

Immunoscreening of the resulting eight phage pools using a mixture of all eight original sera revealed a large number of positive clones. We tested the individual reactivities of all these clones

- 21 • · · · · · · • β · · ·«· ·* ·· ·· ·· with a panel of positive and negative sera by forming an ordered set of clones as phage-secreting colonies. Following their growth, the entire set of clones was transferred as an ordered set to a nitrocellulose filter using a multi-needle device. The same procedure was repeated to create multiple replicates, which were then individually and simultaneously screened for reactivity with positive and negative sera to produce multiple clones that specifically reacted with positive sera. Each of these phages was used as a ligand to affinity purify antibodies from positive serum. These antibodies were then tested in ELISA for reactivity with any of the 4 groups of previously identified HCV peptides (Reference: Prezzi, C., Nuzzo, M., Meola, A., Delmastro, R., Galfre, C., Cortese, R., Nicosia, A. and Monaci, P.: 1. Immunology, 156, 4504-4513, 1996); (Bartoli, et al.: Nature Biotechnology, 16, 1068-1073, November 1998). This analysis identified 12 clones that detected serum antibodies with a novel binding specificity, thus indicating their mimicking antigenic determinants, distinct from those previously identified.

Sequence analysis revealed 5 different sequences. Culture supernatants were prepared from each of these 5 clones (PA1, PA3, PA8, PA12 and PA18) and their reactivity by ELISA was tested with 30 different positive and 24 negative sera (Figure 1). Only clones PA8 and PA12 showed statistically significant reactivity with positive sera (p < 0.2).

PA1 phage was then used to immunopurify antibodies from C65 serum. These purified antibodies specifically reacted with clones PA3, PA8, PA12, PA18, as well as with clone PA1 itself, indicating that • «4 4

4· » 4 4 t ··· ·· ·· ··

- 22 that all these phages can be classified into a single class recognized by antibodies of the same specificity (Figure 2). No reactivity was detected for any of the above clones when wild type was used as ligand or when antibodies were affinity purified from negative serum.

When different clones from the same group were used to immunopurify the serum antibodies, or when different positive sera were used, we obtained results consistent with the reactivities reported in Figure 1. When, for example, phage PA3 was used to immunopurify the antibodies from the same serum, these antibodies reacted in addition to PA3 with clones PA8 and PA12. These results indicate that clones PA1, PA3, PA8, PA12 and PA18 detect antibodies reacting with the same B-cell epitope, as indicated by the slight similarity between the peptide sequences (Figure 1).

Affinity maturation of HCV peptides

A phage library was constructed in which the sequence of clone PA12 was partially mutagenized. In this “secondary” library, designated pVIIIA12, oligonucleotides were synthesized such that each amino acid of the sequence SREQLNKLFGIEG was independently substituted with another amino acid; theoretically, substitutions should occur at each position with a frequency of 20 percent. In addition, a random fragment was inserted on both sides of the foreign peptide sequence. The pVIIIA12 library was plated twice on 12 positive sera (C8, C10, C12, C13, C22, C58, C60, C76, C83, C85, C141 and C177).

- 23 When tested by ELISA, phage pools p76, p141 and p177 (derived from selection using sera C76, C141 and C177) showed the greatest and broadest reactivity with the panel of positive sera and were further analyzed (Figure 3). Based on this reactivity, phage pools p76, p141 and p177 were immunoscreened using sera C40, C141 and C177. For each pool, this analysis yielded several clones which were then individually tested for reactivity with a wide variety of positive and negative sera by a filter replication procedure as described in more detail below. This final screen yielded 51 clones specifically reacting with positive sera.

Sequence analysis revealed 16 different sequences (8, 5 and 3 derived from the p76, p141 and p177 pools, respectively). Culture supernatants were prepared from each of these clones and their reactivity was tested by ELISA with 33 different positive and 24 negative sera (Figure 1).

Clones P40.17 and P40.7 reacted with more positive sera tested (42% each). Their combination with strains P141.7 and P177.22 evaluated 70% of the positive sera of the panel. The clustering of selected peptides defines the consensus sequence (M/Y)SRE(W/Q)L(M/N)K(A/L)(H/F)GIES(W/M).

Identification of additional HCV-specific ligands ·«

- 24 Similar selection was performed with the aim of identifying ligands with novel binding specificities different from those previously identified. Phage libraries of various lengths were screened, in which the random sequence was either completely random or flanked by cysteine fragments that restrict the conformation of the displayed peptides.

Various combinations of serum and phage pools were used to select phage libraries. Phage pools showing interesting reaction profiles with positive serum were further analyzed. Evaluation of the reactivity of a large number of individual clones by replica screening unified phages showing specific reactivity with positive sera.

We focused our attention on clones with interesting reactivity characteristics.

Screening of these clones with antibodies affinity purified with HCV peptides eliminated those clones that mimicked antigenic determinants of previously identified peptides. Clones that survived these selection steps were tested by ELISA with a large number of positive and negative sera to statistically determine their specificity for HCV.

These selected clones were further improved by generating and screening secondary libraries in which the sequence of the original clone or population of clones was partially mutagenized and again screened for variants with improved binding properties. This step was performed with a differently adapted strategy that has already been

- 25 have been described (reference to Urbanelli, Zhu). This extensive and iterative effort identified 7 new groups of ligands that specifically bound HCV-specific serum antibodies with varying binding specificity.

Screening of the repertoire of HVR1 variants represented by recombinant phages using sera isolated peptides specifically reacting with a large number of positive sera (Puntoriero, et al.: Nicosia - unpublished): The set of phage-derived HVR1 peptides obtained from this screen was analyzed with our sera panel. Three peptides with the highest and specific frequency of reactivity with positive sera were identified (mF78, mH1 and m858).

In total, 12 groups of ligands were identified, including 4 groups previously identified (Bartoli, et al.: Nature Biotechnology, 16, 1068-1073, November 1998; Prezzi, C., Nuzzo, M., Meola, A., Delmastro, R., Galfre, C., Cortese, R. Nicosia, A. and Monaci, P.: 1. Immunology, 156, 4504-4513, 1996); see Figure 5, groups A to L.

From phage-borne peptides to synthetic peptides Twenty-two peptide sequences derived from 12 groups of HCV ligands were synthesized as eight-fold branched antigenic peptides (ADAM-HCV peptides). In this molecule, eight identical sequences are linked via a lysine fork to a common core, forming a folded arrangement similar to • 4

- 26 fused pVIII peptides on the phage capsid. Below are the sequences of the ADAM-HCV peptides:

m858 ETYTTGGAAARTTSGLTSLFSPGPSQN m1901.31 AEGEFKKFPGSSTPKDPAKAAFDSL m1901.34 AEGEFPEDTFPGSKLILSGDPAKAAFDSL m 1909.2 AEGEFKTRRNTNYQDPAK m19 13.2. AEGEFKTLRNTNRLDPAK m 1929.21 AEGEFATASPTHYTSELDPAK m1929A3.1 AEGEFTTASPTHFLVPLDPAK m1929C3.4 AEGEFATAPPSHYSWDPAK m1977.1 AEGEFPYLLPRRSREEAVDPAK m3322.3 AEGEFPQDARFPGGGDPAK m3362.3 AEGEFLSLKGSGGGQLRALVDPAK m3555 1.3 AEGEFRLGVRALRKAPDPAK m3566.3 AEGEFKTSVRSVPPRARPPINGDPAK

Ma12. 1 AEGEFNSREWLSKAHGIEGMDPAK mA12.2 AEGEFRSREQLSKIFGIDLTDPAK.

mA12.13 AEGEFMSRTWLMKAHGIESWDPAK

Nb11.17 AEGEFRELLYEAFDDMEGDPAK mF78 QTHTTGGQAGHQAHSLTGLFSPGAKQN rnG2 1.2 AGEPYVIEQGMMDPAK mH 1 QTHTTGGVVGHATSGLTSLFSPGPSQN mN 15.3 AEGEFGLADLATLTFGSTDPAK mS48.5 AEGEFRPWKVPDYDPPAAGGDPAK and data on the analysis are summarized in Table 1 and figures are also attached.

• 4

4·

- 27 4· 4·

In many cases, pVIII sequences surrounding foreign epitopes (NH2AEGEF and/or DPAK-COOH) were reported, which are relevant for the binding specificity of the corresponding peptide.

A mixture containing these 22 ADAM-HCV peptides (ADAM-HCV mixture) was used to detect the presence of antibodies to HCV by EIA (ADAM-HCV EIA). The ADAM-HCV peptide mixture was immobilized by passive coating on the bottom of a multiwell ELISA plate and incubated with serum samples diluted 1:40 for 40 minutes. Human antibodies bound to the peptides were detected by incubation for 20 minutes with an anti-human conjugate and measured by a chromogenic enzymatic reaction.

As shown in Figure 4A, the ADAM-HCV EIA effectively discriminates between positive and negative sera.

ADAM-HCV/EIA

A panel of HCV positive and HCV negative sera was tested for the presence of anti-HCV antibodies using the ADAM-HCV-EIA (Figure 4B). The assay identified all positive sera given to the panel and also demonstrated 100% specificity in identifying all negative samples.

Undetermined samples

A set of sera diagnosed as indeterminate by commercially available confirmatory tests for HCV detection was obtained

- 28 from different sources. 23 ADAM-HCV peptides were individually tested for reactivity by ELISA with 31 samples (Figure 5). 6 samples did not react with any of the tested MAPs and were therefore considered negative. 8 samples detected only one antigen, thus confirming the uncertainty of the analysis. 17 samples finally showed two or more reactions against different groups of peptides, thus being identified as positive.

ADAM-HCV Strip Immunoblot Assay (ADAM-HCV/SIA) ADAM-HCV peptides were covalently immobilized on an activated nylon membrane to obtain strips with ten bands. Each row contained different peptides with the same binding specificity, as detailed in the legend to Figure 6. A control row containing purified human IgG was also included as a positive control. A selected number of samples from a panel of undetermined sera were tested for reactivity with the immobilized antigens by incubating the serum samples on the strip. Anti-HCV antibodies captured by the individual antigens were visualized by colorimetric enzymatic reaction by incubating the strips with anti-human enzyme conjugate. The reactivity of the samples with the peptide bands was determined by visually comparing the intensity of each band with the internal control positive band row.

ADAM-HCV/SIA showed reactivity in all 8 positive sera tested for various viral determinants with ADAM-HCV mimic peptides. No reactivity was detected in 8 negative sera tested (Figure 6).

• · * ·

- 29 We also analyzed a selected number of undetermined samples using ADAM-SIA. As shown in Figure 6, the analysis of 8 undetermined sera confirmed the results obtained by ADAM-HCV/EIA using individual peptides, indicating comparable sensitivity of both assays.

Materials and methods used

Phage libraries

Four different libraries of random peptides containing phages were used as a source of ligands: pVHI9a, pVlllaa_cys, pVIII12aa, pVIIHSaa and pVIIIA12. pVIII9aa (Felici, et al., 1991), pVIII12aa and pVIII15aa are three different libraries, composed of random nonamers, dodecamers and pentadecamers, which appear on filamentous phage as a condensed NH ₂ terminus of the main coat protein pVIII. pVIIl9aa_cys is a library in which the nine-fold peptides are surrounded by two cysteine fragments (Luzzago, et al., 1993). In this last library, the cysteines induce the formation of a disulfide bridge, which to some extent restricts the conformation of the displayed peptides. The pVIIIA12 library was constructed by synthesizing oligonucleotides encoding the amino acid sequence SREQLNKLFGIEG. By modifying the synthetic high-molecular-weight cleavage method (Glaser, et al., 1992), each amino acid position can be substituted with an NNS triplet at a frequency of 20 percent. In addition, random fragments are present at both ends of the foreign peptide sequence. All libraries were constructed according to the literature (Folgori, et al.,

- 30 1998). The library complexity, as measured by the number of individual clones obtained by bacterial transformation, was around 1 x 10 ⁸ in each of the five libraries.

Human sera

The human sera used in this study were collected randomly from rejected units of donated blood from transfusion centers and various hospital departments and from units obtained from healthy volunteers. Namely, many of the unspecified samples used in this study were obtained from the Laboratory of Virology, Istituto Superiore di Sanita, Rome (Italy) and from the Centro Nazionale Transfusione Sangue della Croce Rossa Italiana, Rome, (Italy). The sera were tested for the presence of antibodies to HCV with the second-generation HCV ELISA test system (Ortho Diagnostic Systems, Bersee, Belgium) and confirmed with the first-generation RIBA HCV dot blot immunoassay system (Chiron Co., Emmeryville, CA). The sera were also tested for the absence of antibodies to HBsAg and to HIV-1/HIV-2 with the AUSAB EIA test (Abbott Labs, Chicago, IL) and the third-generation HIV-1/HIV-2 EIA test (Abbott Labs, South Pasadena, CA). Samples positive for anti-HCV antibodies but negative for anti-HBsAg and anti-HIV antibodies were included in this study as HCV positive sera. Sera negative for antibodies to all three antigens were included in this study as HCV negative sera.

4

4

4

- 31 Affinity selection and immunoscreening

HCV positive sera were used for affinity selection of random peptide phage libraries (Folgori, et al., 1998; Felici, et al., 1996; Prezzi, et al., 1996). Clones derived from affinity selection were immunoscreened using sera as described in the literature (Prezzi, et al., 1996; Minenkova, et al.).

ELISA test using phage clones

The ELISA using phage supernatant and human sera was performed as follows; Phage supernatants were prepared from infected DH5α-F' cells as previously described (Felici, et al., 1991). Multiwell plates (Immunoplate Maxisorp, Nunc, Roskilde, Denmark) were coated overnight at 4°C with 200 μl of anti-pIII monoclonal antibody 57D1 (Dente, et al., 1994) at a concentration of 1/vg antibody per ml in 50 mM NaHCO ₃ pH 9.6. After removal of the coating solution, the plates were incubated at 37°C for 60 minutes with ELISA blocking buffer (0.1% casein, 1% Triton-X100 in PBS). The plates were washed several times with PBS/0.05% Tween-20 (washing buffer). A 1:1 mixture of ELISA blocking buffer and clarified phage supernatant was added to each well and allowed to react at 37°C for 1 hour. Human serum diluted 1:40 was incubated for 30 minutes at room temperature with 5.10 ¹⁰ plaque-forming units of phage f11.1 per ml (pfu.ml· ¹ ) (Dente, et al., 1996), 25 µl.ml ^·1 protein extract from DH5-F cells infected with phage f11.1 (Dente, et al.) and 25 µl.ml· ¹ supernatant from unrelated rat hybridoma cells in blocking buffer.

- 32 ELISA buffer. After removing the mixture containing the phage supernatant, the plates were washed with the buffer and 200 μl of the pre-incubated serum mixture was added to each well and incubated at 37°C for 60 minutes. The plates were then washed with the buffer and a 1:20,000 dilution of goat anti-human IgG HRP-conjugate (Jackson Immuno Research Laboratories, Inc., West Grove, PA) in secondary antibody blocking buffer (1% Triton, 1% horse serum, 50% fetal calf serum in PBS, Mab supernatant 5 μl per well) was added to each well. After a 30 minute incubation at 37°C, the plates were washed and peroxidase activity was determined by incubation with 200 μl of the TMB liquid substrate system (Sigma, St. Louis, MO). After 15 minutes of exposure, the reaction was stopped by adding 25 μl 2M H ₂ SO ₄ . The plates were read using an automated ELISA reader (Labsystems Multiskan Bichromatic, Helsinki, Finland) and the results were expressed as A = A450 nm - A ₆₂ o nm. The ELISA data reported are the average values from two independent determinations. They are considered statistically significant if they are greater than 0.3 above the cutoff and differ by more than 3σ (σ = {1 /2[cr ² p + ίΑν]} ^1/2 ) from the background observed for the wt tag.

The p-value referring to clones PA8 and PA12 is the probability that the observed frequency distribution of reactivities of positive and negative sera is statistically equal according to the % ² -test.

« *

- 33 Affinity purification of phagotope-specific antibodies from serum

A 60 mm diameter polystyrene Petri dish (Becton Dickinson Labware, NJ) was coated with a solution of 1.10 ¹¹ .ml' ¹ of CsCl-purified phage particles in 50 mM NaHCO ₃ pH 9.6 at 4°C overnight. After washing with PBS/Tween, the dish was incubated at 37°C for 60 minutes with ELISA blocking buffer. A mixture of human serum (1:100 dilution in ELISA blocking buffer), 1.10 ¹² f11.1 pfu.ml' ¹ and 25 //l.ml' ¹ of XLI-blue cell protein extract was incubated at room temperature for 60 minutes. After removing the blocking buffer, the pre-incubated mixture was placed on the plate and incubated at 4°C overnight. The diluted serum was removed and the dish was washed with washing buffer. Bound antibodies were eluted with 0.1 M glycine.HCl pH 2.7, with 10 µg.ml ^-1 BSA added and neutralized.

Characterization of phage clones

Affinity-purified antibodies were tested for reactivity against phagetopes by standard ELISA. Multiwell plates were routinely coated with 100 μΙ of a solution of 1.10 ¹¹ TU.ml' ¹ of CsCl-purified phage in 50 mM NaHCO ₃ pH 9.6 per well and left at 4 °C overnight. After washing with PBS/Tween, the plates were incubated at 37 °C for 60 minutes with blocking buffer. Then, 100 μΙ of affinity-purified antibody was added to each well and left to react at 4 °C overnight. The plates were then washed with cold PBS/Tween and 100 μΙ of goat anti-human IgG (Fc specific) antibodies were added.

- 34 conjugated to alkaline phosphatase (Sigma, St. Louis, MO), diluted 1:5000 with blocking buffer. After incubation at room temperature for 2 hours, the plates were washed and alkaline phosphatase was determined as described above.

Synthetic peptides

We used synthetic compound eight-fold branched peptides (MAP): (Tam, 199X). The synthesis was performed by the flow polyamide method. The peptides were dissolved in dimethyl sulfoxide.

ELISA using synthetic peptides

Multiwell plates (Immuno plate Maxisorp, Nunc, Roskilde, Denmark) were coated overnight at 4°C with a MAP solution at a concentration of ¹⁰ //g.ml·1 in 50 mM NaHCO ₃ at pH 9.6. After removal of the coating solution, the plates were incubated at 37°C for 60 minutes with ELISA blocking buffer (0.1% casein, 1% Triton-X100 in PBS). The plates were washed several times with PBS/0.05% Tween-20 (washing buffer). Human serum diluted 1:40 was added to each well and incubated at 37°C for 40 minutes. The plates were then washed with wash buffer and goat anti-human IgG HRP-conjugate diluted 1:20,000 (Jackson Immuno Research Laboratories, Inc., West Grove, PA) in ELISA blocking buffer was added to each well. After a 20-minute incubation at 37°C, the plates were washed and peroxidase activity was determined by incubation with TMB liquid substrate (SIGMA, St. Louis, MO). After 15

- 35 • 4 *4« · · · 4 4 · 4 • 4 4 · · « 4 · 4 4 4 4 • 4 44 4* ·· 44 4« After a minute reaction, the reaction was stopped by the addition of 2M H ₂ SO ₄ . The plates were read with an automatic ELISA reader (Labsystems Multiskan Bichromatic, Helsinki, Finland) and the results expressed as A = A4 _50nm - A ₆₂ onm· The ELISA values reported are the average of two independent determinations.

Links

Smith, C.P. and Petrenko, V.A.: Phage display. Chem, Rev., 97, 391-410, 1997

Folgori, A., Tafi, R., Meola, A., Felici, F., Galfre, G., Cortese, R., Monaci, P. and Nicosia, A.: A general strategy to identity -mimotopes of pathological antigens using only random peptide libraries and human sera. EMBO 1.13, 2236 - 2243, 1994

Prezzi, C., Nuzzo, M., Meola, A., Delmastro, R., Galfre, C., Cortese, R., Nicosia, A. and Monaci, P.: 1. Immunology. 156, 45044513, 1996.

Alter, H.j.: To C or not to C: these are the questions'. Blood. 85, 1681, 1995

Felici, F., Castagnoli, L., Musacchio, A., Jappelli, R. and Cesareni, C.: Selection of antibodies ligands from a large library of oligopeptides expressed on a multivalent exposition vector, f. Mol. Biol. 222, 301-310, 1991.

Luzzago, A., Felici F., Tramontano, A., Pessi, A. and Cortese, R.: Mimicking of discontinuous epitopes by phage displayed peptides, ·«** · · · · · · · · ·· ·· ·* ·♦ ·· ··

- 36 I. Epitope mapping of human H ferritin using a phage library of constrained peptides. Price, 128:51-57, 1993

Dente, L., Cesareni, G., Micheli, C., Felici, F., Folgori, A., Luzzago, A., Monaci, P., Nicosia, A. and Delmastro, P.: Monoclonal antibodies that recognize inflammatory phage. Useful tools for phage display technology., Gene, 148, 7, 1994.

Sambrook, J., Fritsch, T. and Maniatis, T.: Molecular Cloning: a laboratory manual (second edition), 1989

Takamizawa, A., Mori, C., IFuke, I., Manabe, S., Murakami, S.,

Fujita, J., Onoshi, K., Andoh, T., Yoshida, I. and Okayama, H.: Structure and organization of the Hepatitis C virus genome isolated from human carriers. I. Virol. 65,1105-1113, 1991,

Smith, D.B.: Purification of glutathione S-Transferase fusion proteins. Methods in molecular and cellular biology. 4, 220-229, 1993

Frangioni, f.V. and Neel, B.J.: Solubilization and purification of enzymatically active glutathione S-transferase (pGex) fusion proteins. Analyte. Biochem. 210, 179-187, 1993

Kornatsu F, Takasaki K: Determination of serum hepatitis C virus (HCV) core protein using a novel approach for quantitative evaluation of HCV viraemia in anti-HCV-positive patients. Liver, 19, 375-80, 1999

Table 1. Reactivity of ADAM-HCV peptides

N = number of sera tested; S = measured signal; CO = cut-off value; (CO = neg + 3σ, where neg and σ are the mean and standard deviations of the values from the negative serum value);

negative positive sera sera

group peptide N N %S/CO>1 mean value S/CO>1 A m1913.2 33 . 21 100 45.0 m1909.2 33 21 100 50.3 B ΠΊ1901.31 39 38 61 3.2 m1901.34 39 38 39 5.4 m3322.3 30 18 78 10.1 m3362.3 31 37 84 4.4 C m1977.1 32 28 96 13.6 D m3551.3 37 26 46 3.0 E m3566.3 14 12 . 50 2.1 F m858 40 41 73 8.9 mH1 60 60 68 9.8 mF78 64 51 80 5.8 G mA12.1 37 45 64 8.4 mA12.2 37 45 33 8.6 m12,13 38 45 31 4.4 H mB11.17 35 24 46 2.7 I mG21.2 34 34 26 4.0 J m1929C3.4 35 33 27 7.1 m1929A3.1 43 25 40 3 Q mt 929.21 13 18 72 97 To mS48.5 29 9 67 7.0 L mN15.3 36 21 10 15.2

- 37 • · · · ♦ ♦ * · · * · ·

Claims (25)

Patent claims 1. A method of diagnosing antigen comprising identifying the binding specificity of an antibody molecule with an antigen in a serum by an antigen mimetic (ADAM) antibody detection method comprising screening a phage library using a serum of patients infected with antigen and uninfected individuals; said antigen. 2. The method of claim 1 wherein said ligands are enhanced by in vitro maturation. Method according to claim 1 or 2, characterized in that said ligands are synthetic peptides. The method of claim 1, 2 or 3, wherein said ligands are attached to common core structures. 5. The method of claim 4, wherein said ligands together with said core structures are MAP. 6. An array of antigen-specific ligands, characterized in that it is obtained by a process comprising: a) first screening the phage library with n positive sera to form the first series of n-phage pools; b) preparing n pool mixtures containing n-1 pools; - 38 4 4 «44 4 4 4 · 4 · 4 ♦ • 4 4 4 4 4 44 ·· • 4 · 4 * 4 44 44 ·· c) selecting the affinity of each of the n-mixtures against serum, which generated a selected phage pool to obtain a second series of n-phage pools, and optionally d) further screening of each of the second series n-phage pools against a mixture of all n-original sera, except for those used for the first screening; e) immunoscreening the resulting second series of n-phage pools using a mixture of all n-parent sera to obtain positive clones; f) testing individual reactivities of all positive clones with a panel of positive and negative sera using a selected set of said clones as phage-secreting colonies; g) creating replicas of said phage-secreting colonies; h) screening each replica for its positive and negative serum reactivity to detect clones specifically reacting with the positive serum; i) using each of said specifically reacting phages as ligands for caffinite purification of antibodies from positive serum; j) assaying said antibodies for their reactivity with previously identified HCV peptides; k) unifying clones detected by serum antibodies. 7. A method for diagnosing hepatitis C by identifying the binding specificities of anti-HCV antibody molecules in serum by a method of detecting the antibody by antigen mimetics. I I «· · 44 44« «« «« «« « 43 4 · ·· ·· · · * 39 (ADAM) consisting of screening a phage library using serum from HCV patients and uninfected individuals, by identifying antibodies that bind peptides (ligands) specifically associated with HCV infection. 8. The method of claim 7, wherein said ligands are enhanced by in vitro maturation. The method of claims 7 or 8, wherein said ligands are synthetic peptides. Method according to claims 6 or 7 or 8, characterized in that said ligands are attached to common core structures. The method of claim 8, wherein said ligands together with a common core structure are MAP. A set of HCV-specific ligands obtainable by a process comprising: a) first screening the phage library with n positive sera to form the first series of n-phage pools; b) preparing n pool mixtures containing n-1 pools; c) selecting the affinity of each of the n-mixtures against the serum that formed the selected phage pool to obtain a second series of n-phage pools, and optionally (d) further screening of each of the second series n-phage pools into a mixture consisting of all n-original sera, except for those used for the first screening; «« » - 40 • · · ··· e) immunoscreening the resulting second series of n-phage pools using a mixture of all n-parent sera to obtain positive clones; f) testing individual reactivities of all positive clones with a panel of positive and negative sera using a selected set of said clones as phage secreting colonies; g) creating replicas of said phage-secreting colonies; h) screening each replica for its positive and negative serum reactivity to detect clones specifically reacting with the positive serum; i) using each of said specifically reacting phages as a ligand to caffinitively purify antibodies from positive serum; j) assaying said antibodies for their reactivity with previously identified HCV peptides; k) unifying clones detected by serum antibodies. 13. The collection of claim 12, wherein the phage library of step a) is pVII 1-12aa. The collection of claims 12 and 13, wherein the inserted clone singlets of step k) of claim 12 have the following sequences: SREQLNKLFGIEG; RATLSNEHGITIG; DQRENWFKYHGFG; EWRRYMSDIHGYG; DSLRYMYVMPGFG. A collection according to claim 12, characterized in that a phage library is generated, each amino acid of the clone sequence is independently substituted with some other amino acid. . Ϊ tet · · · · · 44 · · · a · ··· 44 ··· · ··· «« ·· - 41 16. The collection of claim 12, wherein a phage library is formed in which the best responding clones are partially mutagenized by substituting each amino acid of the clone sequence independently for some other amino acid. 17. The collection of claim 15 or 16, wherein said clones have an inserted sequence SREQLNKLFGIEG. A collection according to any one of claims 12 to 17, characterized in that in said phage library the random sequence is surrounded by two cysteine fragments. A collection according to any one of claims 12 to 18, wherein the peptide sequence is linked to a common core structure. 20. The assembly of claim 17, wherein said peptide together with a common core structure is MAP. Use of a kit according to any one of claims 12 to 20 in the preparation of a diagnostic kit for detecting HCV individuals suspected of having been affected by said HCV. A kit for diagnostic purposes comprising the kit of claim 6. An HCV diagnostic kit comprising the kit of any one of claims 12 to 20. A kit according to claims 22 or 23 comprising strips, wherein said stripe is immobilized thereon. 25. The kit of claim 24, wherein said strip further comprises an internal standard.