WO2010035292A1

WO2010035292A1 - The use of a human monoclonal anti-hcv antibody as a medicament for the therapeutic treatment and prevention of hcv infections

Info

Publication number: WO2010035292A1
Application number: PCT/IT2008/000609
Authority: WO
Inventors: Roberto Burioni; Massimo Clementi
Original assignee: Natimab Therapeutics S.R.L.
Priority date: 2008-09-24
Filing date: 2008-09-24
Publication date: 2010-04-01
Also published as: WO2010035292A8

Abstract

The present invention relates to human monoclonal antibodies capable of recognizing the conformational epitope of the HCV E2 glycoprotein recognized by the e8 Fab and cross-reacting with a plurality of different HCV genotypes, as a medicament for the treatment or prevention of HCV infections. A pharmaceutical composition for the treatment or prevention of HCV infections, comprising an antibody as defined above, preferably the e8 Fab, as well as pharmaceutically acceptable excipients, vehicles or diluents, is also disclosed.

Description

The use of a human monoclonal anti-HCV antibody as a medicament for the therapeutic treatment and prevention of HCV infections

The present invention relates to the use of human monoclonal antibodies directed against the E2 glycoprotein of the Hepatitis C virus (HCV) as a medicament for the therapeutic treatment and prevention of HCV infections.

HCV is a virus with a precapsid and a single stranded RNA, belonging to the Flavivirus family. Based on the genetic differences observed between different HCV isolates, this virus species is classified into 6 genotypes, each designated with a number. Each genotype in turn comprises a number of subtypes, each of which is designated with a letter. The preponderance and distribution of the different HCV genotypes varies throughout the world. In Europe, . genotype Ib is predominant, whilst in North America genotype Ia is predominant. Determining the genotype is clinically important, in that these features contribute to determining the potential response to the therapy based on the association between interferon and ribavirin, which is the most widely used treatment today. Genotypes 1 and 4, in fact, are less responsive to interferon-based treatment than genotypes 2, 3, 5 and 6.

As of today, there is no vaccine or immunotherapy available against the hepatitis C virus which has proven to be truly effective. Up to now, the high variability of the HCV antigenic structure has hindered the development of antibodies which are at the same time capable of neutralizing the virus and of cross-reacting with different virus genotypes. Thus, there is a need for anti-HCV antibodies which possess these characteristics and, as a consequence, which are truly effective in the therapy and prevention of HCV infections.

Antibodies directed against HCV antigens are disclosed in the prior art. For example, Burioni et al., Hepatology Vol. 28, n. 3, 1998, disclose cloning and characterization of sequences encoding five human recombinant antibody fragments (Fabs) specific for the HCV glycoprotein E2 (HCVE2), capable of binding to glycoproteins from different virus genotypes (cross-reactivity). Amongst the Fabs disclosed in this paper, the antibody fragment designated e8 recognises the glycoprotein E2 from the HCV genotypes Ia, Ib, 2 a, 2b and 3. The amino acid sequence of the heavy chain variable region of the antibody fragment e8 is SEQ ID NO:1. The amino acid sequence of the light chain variable region of the antibody fragment e8 is SEQ ID NO.2. Burioni et al., 1998, cit. disclosed that e8 has only a minimal activity in neutralizing the binding of HCV E2 (NOB activity). The same is provided in the International patent application WO 03/064473, filed on 29 January 2003 in the name of Roberto Burioni, in which e8 is described as incapable of inhibiting binding of HCV E2 to the target cells and incapable of neutralizing virus infection, even at high concentrations (80 μg/ml).

The present inventors have studied further e8 in the form of a Fab fragment and, with the use of more advanced neutralization activity test systems, they have surprisingly found that, contrary to what has been stated in the prior art, the e8 antibody fragment is capable of neutralising HCV infection in vitro, having a virus neutralization activity expressed as 50% inhibiting concentration as low as 10 μg/ml. Moreover, e8 stains the surface of HCV infected cells either fresh or treated with permeabilization reagent. This makes e8 particularly suitable for neutralization of HCV and elimination of HCV infected cells and for use as a medicament for immunotherapy and immunoprevention of HCV infections.

Even more importantly, the inventors have found that the e8 Fab binds to the HCV E2 glycoprotein on an epitope which has never been disclosed before.

The identification of the region recognised by the e8 Fab required, complex and time- consuming experimental work, since the region recognized and bound by e8 is a conformational epitope which is not reproduced by Pepscan or other approaches utilising linear peptides. In order to overcome such limitations, three different approaches were used.

Firstly, competition assays were carried out with a large panel of anti-E2 monoclonal mouse and rat antibodies directed against known epitopes spanning the three regions of E2 involved in binding with the CD81 molecule (412-424, 436-443, 474-495, and 520-550) (5,21). The competition assays confirmed that the e8 Fab recognition site is outside the regions on E2 engaged in CD81 binding. The CD81 molecule, which is expressed on the surface of the target cells, is known to be a necessary but not sufficient co-receptor for HCV entry into target cells. Moreover, a number of studies demonstrated that epitopes targeted by neutralizing antibodies involve amino acids that are key contact residues for E2 binding to CD81 (17,18,19,20).

Furthermore, a panel of H77-derived E1E2 proteins (HCV genotype Ia) was used, each containing single alanine substitutions at residues that are conserved across different HCV genotypes and described to be critical for CD81 binding and for the infectivity in the HCVpp assay (5). The binding of the e8 Fab to all of these mutants is not significantly inhibited compared to the wild-type E2, highlighting that the epitope recognized by e8 is outside regions crucial for CD81 binding. Collectively, these results are consistent with the experimental fact that the e8 Fab is not capable of inhibiting HCV E2 binding to CD81.

These data prompted the inventors to further investigate the epitope bound by the e8 Fab. In particular, the e8 Fab was used to screen a phage-displayed random peptide library. Panning of phage-displayed random peptide libraries represents a very useful tool for identifying peptides able to mimic both linear and conformational epitopes of monoclonal antibodies (mimotopes), as peptides selected from a random library for their ability to bind a specific antibody may imitate the true epitope showing homology with the sequence of the native antigen (16,22,23). In particular, these mimotopes can include some amino acids of crucial importance for antibody-antigen recognition that are contiguous on the antigen primary sequence or are brought together in spatial proximity in the folded protein. This approach allowed the inventors to identify several distinct, but similar, peptide sequences with a strong specific reactivity with the e8 Fab, leading to the identification of the consensus sequence LTXPXXLXXXPR (SEQ ID NO:3), wherein L is Leu, T is Thr, P is Pro, R is Arg and X is any naturally occurring amino acid. The observation that the same amino acids exist in multiple peptide sequences that specifically bind the e8 Fab indicates that the epitope of the e8 Fab contains these specific residues, rather than amino acids with similar chemical-physical properties. There is, as yet, no three dimensional (3D) structure of HCV E2. Thus, epitope mapping was performed on the basis of the envelope glycoprotein of Tick Borne Encephalitis Virus (TBEV). Since the E2 sequence of TBEV is related to HCV E2 (12,13), structural similarity may be expected. This theoretical structure of HCV E2 of genotype Ia was scanned using two different algorithms to search for residues of consensus sequence (LTPLPR) brought into contiguity on the surface of E2. This analysis revealed that residues L₆-_U, T₆₄₈, P₅12, L₅₈₀, P₅₉i, R₅₈₈, while distant in primary sequence, are brought together in the folded E2, defining an epitope in spatial proximity, but outside the CD81 binding region located between the two E2 hypervariable domains.

Collectively, these data allowed for the first time to map precisely a conserved and neutralizing human B cell epitope on a region of HCV E2 that is not engaged in CD81 binding.

The region of the HCV E2 glycoprotein bound by the e8 Fab, designated as the conformational epitope, is composed of residues that, although distant in the primary sequence, are brought together in the folded form of the E2 molecule and in the 3D structure model of the protein obtained as illustrated below are located inside a sphere having a radius of approximately 11 A. The conformational epitope comprises residues L_64I, T₆₄₈, P₅I₂, L_58O, P₅₉i and R₅₈₈ and does not comprise the residues at positions 384-410 and 474-482 which are located in the hypervariable regions 1 and 2 (HVRl and HVR2) of HCV E2. Further, the epitope does not comprise the residues at positions 412-424, 436- 443, 474-495 and 520-550 which are located in the four regions involved in CD81 binding. The above-mentioned amino acid positions are identified based on the numbering of the primary amino acid sequence of the HCV E2 glycoprotein from genotype Ia (H77c strain), which is comprised between positions 384 and 746 of the HCV polyprotein precursor. The said amino acid sequence between residues 384 and 746 is designated in the sequence listing as SEQ ID NO:4. The inventors' further studies, herein, demonstrate that the epitope on HCV E2 that is recognized by the e8 Fab is highly conserved across the most common HCV genotypes, indicating its importance for the virus life cycle and suggesting that these residues may be involved in a crucial pre- or post-CD81 binding step of viral entry and that mutation in this key region could strongly impact the ability of the HCV virus to infect the target cells and replicate efficiently.

In the light of such observations, it can be reasonably assumed that human monoclonal antibodies, or other binding moieties, capable of recognising the same E2 conformational epitope recognised by the e8 Fab, will be provided with the same biological properties as the e8 Fab and shall be effective as a medicament for the therapeutic treatment and prevention of HCV infections.

Thus, a first subject-matter of the invention is a human monoclonal antibody, or other binding moiety, capable of binding the conformational epitope of the HCV E2 glycoprotein from a plurality of different HCV genotypes, as a medicament for the therapeutic treatment or prevention of HCV infections, wherein the conformational epitope of the HCV E2 glycoprotein is defined as follows:

(i) it comprises the amino acid residues L_64J, T₆₄₈, P₅₁₂, L_58O, P₅₉i and R₅₈₈ of the HCV E2 glycoprotein primary sequence,

(ii) it does not comprise the amino acid residues at positions 384-410 and 474-482 which are located in the hypervariable regions 1 and 2 (HVRl and HVR2) of the HCV E2 glycoprotein,

(iii) it does not comprise the amino acid residues at positions 412-424, 436-443, 474-495 and 520-550 which are located in the four regions of the HCV E2 glycoprotein involved in

CD81 binding, and

(iv) it is capable of binding a peptide comprising the consensus sequence SEQ ID NO:3, wherein the above-mentioned amino acid positions are identified based on the numbering of the primary amino acid sequence of the HCV E2 glycoprotein from genotype Ia (H77c strain) which is comprised between positions 384 and 746 of the HCV polyprotein precursor sequence (SEQ ID NO:4). The peptide comprising the consensus sequence SEQ ID NO:3, which is capable of mimicking the conformational epitope of the HCV E2 glycoprotein as defined above, is useful in therapeutic, diagnostic, prognostic and prophylactic applications relating to HCV infections, particularly as an immunogen for therapeutic or prophylactic vaccination against HCV, as a diagnostic or prognostic agent, and for the identification of diagnostic molecules and compounds that bind to the conformational epitope and exert antiviral activity.

Additionally, any kind of chemical entity comprising or expressing the conformational epitope of the HCV E2 glycoprotein would be useful in the above-mentioned medical applications.

Thus, the subject-matter of the invention also comprises any chemical entity comprising or expressing the conformational epitope of the HCV E2 glycoprotein for use in the above- mentioned therapeutic, diagnostic, prognostic and prophylactic applications. A chemical entity expressing the conformational epitope is for example a nucleic acid vector coding and expressing the conformational epitope.

Additionally, any kind of chemical entity which is capable of mimicking the conformational epitope of the HCV E2 glycoprotein shall be useful for identifying diagnostic compounds or antiviral compounds active against HCV.

Thus, another subject-matter of the invention is a method of identifying a compound capable of exerting antiviral activity against HCV infections or suitable for use in the diagnosis of HCV infections, comprising the steps of:

- providing a candidate compound;

- contacting the candidate compound with a peptide, a conformational epitope or a chemical entity as defined above, and

- determining whether the candidate compound binds to the peptide, the conformational epitope or chemical entity as defined above, wherein a compound that binds to the peptide, the conformational epitope or chemical entity is capable of exerting antiviral activity against HCV infections or is suitable for use in the diagnosis of HCV infections. Preferably, the human monoclonal antibody or other binding moiety used in the present invention comprises a heavy chain variable region and a light chain variable region, wherein the amino acid sequence of the heavy chain variable region is SEQ ID NO: 1 or a sequence at least 90% identical to SEQ ID NO: 1 and wherein the amino acid sequence of the light chain variable region is SEQ ID NO:2 or an amino acid sequence at least 90% identical to SEQ ID NO:2. Further preferred percentages of sequence identity are at least 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99%, wherein the expression "at least" is referred to each of the indicated percentages.

The nucleic acid sequence coding for the heavy chain variable region of SEQ ID NO: 1 is designated as SEQ ID NO: 5 in the sequence listing. The nucleic acid sequence coding for the light chain variable region of SEQ ID NO:2 is designated as SEQ ID NO:6.

The expression "antibody" is meant to include any class of full-size immunoglobulin, and any immunoglobulin fragment thereof comprising a heavy chain variable region and a light chain variable region, such as for example a Fab, a F(ab')₂, or a CDR (Complementary Determining Region), or a single chain antibody comprising both heavy and light chain variable regions or CDRs , or scaffolds comprising one or more copies of CDR fragments derived from immunoglobulin heavy and light chain variable regions. This includes functional antibody fragments described as ScFvs, diabodies, VHHs, or isolated heavy or light chains. The expression "antibody" also encompasses antibodies that may be generated using the variable region sequences of the e8 Fab in a process of light chain variable region shuffling at the scope of determining YHfVL combinations with improved properties in terms of affinity, stability and/or recombinant production. The expression "antibody" further encompasses any type of full-size immunoglobulin or immunoglobulin fragment fused with specific full-size immunoglobulin or immunoglobulin fragment that may target the e8 Fab or immunoglobulin to specific tissues, cells or soluble protein structure. The expression "binding moiety" is meant to include a peptide or chemical entity that is capable of specifically binding to the e8 Fab epitope and neutralizing HCV activity. The antibody or binding moiety used in the invention can be in a free form or in a conjugated form. A conjugated form is an antibody or binding moiety, as defined above, conjugated with a molecule that may modulate the in vivo persistency, may promote or limit the body distribution, may decrease sensitivity to proteolytic agents, may decrease antigenicity, may increase cytotoxic ability or may facilitate detection in body fluid and tissues. Non-limiting examples of molecules suitable for conjugation include human serum albumin, maltose binding protein, glutathione-S-transferase, phage coat proteins p3 or p8, peptides, sugars, PEG or PEG like molecules, toxins of animal, vegetal or microbiologic origin, cytokines, enzymes, chemiluminescent compounds, bioluminescent compounds, metal atoms, radioisotopes, fluorescent compounds, flag groups or substrates for phosphorylation, glycosylation, ubiquitination, SUMOylation or endoproteolytic cleavage. In order to facilitate conjugation, the C-terminus or N-terminus of the antibody may be modified, for example, by insertion of additional amino acid residues, for instance one or more cysteine residues that are able to form disulfide bridges. The antibody used in the invention can be also associated to human erythrocytes or other cellular carriers, to specific formulations, or to slow release delivery systems such as, but not limited to, , lyposomes, dendrimers, microsomes, nanoparticles, microcapsules, viral vectors and the like.

Another subject-matter of the invention is a pharmaceutical composition for the therapeutic treatment or prevention of HCV infections, comprising a pharmaceutically effective amount of a human monoclonal antibody or binding moiety as defined above.

The composition of the invention may be administered to a subject who is infected with HCV or to a subject who is at risk of being infected with HCV. Any suitable administration route may be used, including parenteral, oral, ocular, topical, loco-regional, enema or aerosol administration. Parenteral administration includes intramuscular injection, intravenous injection, intralymphatic injection, subcutaneous and intradermal injection and infusion.

The composition of the invention may be prepared in any pharmaceutical dosage form which is suitable for the selected route of administration, for example under the form of an injectable solution or suspension, an infusion, a tablet, a capsule, a cream, an ointment, a lotion, and a suppository.

The composition of the invention comprises the antibody, or binding moiety, as defined above as the active ingredient as well as suitable pharmaceutical excipients, vehicles or diluents known to the skilled in the art.

The invention is further described in detail in the following experimental section which is provided by way of illustration only, with reference to the enclosed drawings, wherein:

Figure 1 shows the results of the analysis of the binding of the e8 Fab to HEK 293T cells expressing E1E2 derived from HCV isolates of different genotypes (genotypes Ia, Ib, 2a, 2b, 3, 4, 5 and 6: UKN1A20.8, UKNlbl2.16, UKN2a2.4, UKN2b2.8, UKN3al.28, UKN4.21.16. UKN5.14.4 and UKN6.5.8). The c33-3 Fab was included as a negative control. The e8 and c33-3 Fabs were used at lOug/ml. Binding was assessed by immunohistology. Means and standard errors from three replicates are reported.

Figure 2 shows the neutralization activity of the e8 Fab using virus pseudoparticles bearing E1E2 from genotype Ia (UKN1A20.8). Data obtained from different concentrations of immunoaffϊnity-purified e8 and c33-3 Fabs (negative control) is shown. The experiment was performed three times and the means and standard errors are reported.

Figure 3 shows the binding activity of the e8 Fab on a panel of mutated E1E2 glycoproteins. Binding activity expressed as percentage of reactivity on wild-type E1E2 (H77 strain) is reported on y axis. The means and standard errors of three replicate assays are described. The e8 Fab was used at a concentration of 10ug/ml.

Figure 4 shows the reactivity by ELISA of 31 peptide-bearing phage clones isolated after three (Figure 4a) and four (Figure 4b) rounds of panning. Fifteen distinct clones were identified after three rounds of panning (Figure 4a) and 16 additional clones were identified after 4 rounds of panning (Figure 4b). Binding activity of each clone was tested on the e8 Fab and the pool of human IgG. Means and standard errors of three replicates are shown.

Figure 5 shows the binding of the e8 Fab (lOug/ml) to cells infected with HCV E1-E2 pp of genotype 1 a by indirect immunofluorescence and flow cytometry. The e8 Fab binds to a significant proportion of HCV E1-E2 pp Ia infected cells. The binding is specific and significantly higher than binding of e8 to uninfected cells, as demonstrated on the horizontal scale, measuring log fluorescence intensity, where a significant proportion of HCV E1-E2 pp Ia infected cells show a high level of fluorescence compared to uninfected cells.

Figure 6 shows the amino acid sequence alignment, across different HCV genotypes (genotypes Ia, Ib, 2a, 2b, 3, 4, 5 and 6: UKN1A20.8, UKNlbl2.16, UKN2a2.4, UKN2b2.8, UKN3al.28, UKN4.21.16. UKN5.14.4 and UKN6.5.8), of the region of HCV E2 that contains the residues of the epitope recognized and bound by the e8 Fab. The 6 residues that define the epitope are in bold. The percentage of identity in these residues, compared to E2 from genotype Ia, is reported in brackets.

Figure 7 shows a model of the 3D structure of the HCV/E2 protein. To overcome the lack of HCV E2 3D structure, this model was built on the basis of the envelope E2 glycoprotein of Tick Borne Encephalitis Virus (TBEV) which is related to HCV E2. The model of genotype Ia HCV E2 structure revealed that the residues L₆₄₁, T₆₄₈, P₅₁₂, L₅₈₀, P591, R₅88, although distant in primary sequence, are brought in spatial proximity in the folded E2. The e8 epitope defined with this approach is outside the CD81 binding region located between the two E2 hypervariable domains.

EXPERIMENTAL PART

Generation, purification and partial characterization of the anti-HCV E2 e8 Fab used in this study are described in the prior art (6,7). The mouse monoclonal antibodies used in this study are described in (8,9). In all of the experiments, binding of the human Fab to purified antigens was detected by ELISA. Briefly, antigen-coated ELISA plates (Costar) were washed and blocked with PBS/1%BSA for Ih at 37°C, 50 μl of appropriate dilution of Fab preparation in PBS/1 %BSA was added and incubated for 2h at 37°C. Plates were washed 10 times with phosphate buffer saline (PBS)/0.05% Tween-20 (Sigma, St Louis, US) and 50μl of a 1:700 dilution in PBS of horseradish peroxidase conjugate goat anti-human Fab (Sigma, , St Louis, US) were added. In the case of mouse control Mab an anti-mouse Fc conjugate was used. After Ih at 37°C plates were washed as above, lOOμl of substrate (Sigma, St Louis, US) were added and plates were read for OD at 450nm after 30 min at room temperature in the dark. All assays were performed at least in duplicate. A negative control antigen (BSA) was always included and the OD reading was subtracted as background.

EXAMPLE 1

Evaluation ofe8 Fab binding on HCV/E2 derived from different HCV genotypes

The binding activity of the anti-HCV E2 monoclonal e8 Fab was assayed using recombinant HCV E2 proteins derived from different genotypes.

Human epithelial kidney (HEK) 293 T cells were grown in Dulbecco's modified Eagles' medium (Invitrogen, Carlsbad, CA), supplemented with 10% foetal calf serum, 5% non essential amino acids, 200 mM glutamine, streptomycin (100μg/ml)' and penicillin (100 U/ml). When the 80% of confluence was reached, 2x10⁶ HEK cells were seeded in 10 cm plates, and after 24 hours they were transfected with 3 μg of pcDNA3.1 (Invitrogen,), an expression vector encoding E1E2 glycoproteins from different HCV genotypes (10), using a calcium phosphate transfection protocol (Sigma, St Louis, US ). The medium was replaced 16 hours after transfection, and cells were then incubated at 37°C for 24 hours. The medium was discarded and the cell monolayer was washed twice with PBS. Five ml of dissociation buffer were added and the cells were incubated at 37° C for 5 minutes. Cells were washed twice with PBS and centrifuged at 1000 rpm for 5 minutes; 1.2 ml of fixation reagent (Fix and Perm, Invitrogen) was added to the pellet from each plate. Cells were incubated for 15 minutes at room temperature. Samples were washed in 5 ml of PBS supplemented with 2% of foetal calf serum (FPBS), then centrifuged at 1000 rpm for 5 minutes and treated with permeabilization reagent Fix and Perm (Invitrogen). Binding of the e8 Fab to HCV E2 expressing cells was assessed both by immunohistology (Figure 1) and fluorescence activated cell sorting (FACS, Figure 5).

The binding activity, assayed by FACS, was expressed in terms of percentage of fluorescent positive cells obtained from the percentage of cells with a higher fluorescence level than cells stained with secondary antibody alone. A human recombinant Fab (c33-3) specific for a non- structural antigen of HCV (NS3) was included in each experiment as a negative control.

The e8 Fab was shown to be broadly cross-reacting (Figure 1), being able to bind cells expressing E1E2 glycoproteins from all the most common HCV genotypes (Ia, Ib, 2a, 2b, 3), and not to cells expressing El E2 glycoproteins from the less common HCV genotypes 4, 5 and 6. The human recombinant Fab (c33-3), used as the negative control, did not show reactivity with any of the El E2 expressing cells..

EXAMPLE 2

Evaluation ofe8 Fab neutralizing activity on HCVpp bearing E1E2 of genotype Ia

The broad E2 cross-reactivity reported above, prompted the inventors to analyze the neutralizing activity of the e8 Fab against pseudoparticles (HCVpp) derived from murine leukemia virus displaying functional full-length E1E2 proteins of HCV genotype Ia (strain UKN1A20.8) using an HCVpp based neutralization assay (HCVpp = HCV pseudo particles).

HCVpp bearing E1E2 from genotype Ia were generated as previously described (11). 60ul of medium containing HCVpp were mixed with 90ul of different concentrations of e8 Fab, and incubated for 1 hour at 37°C This mixture was added to Huh-7 target cells and the cells were incubated for 3 hour at 37°C. Finally the inoculum was removed, ImI of fresh medium was added to each well, and cells were incubated at 37°C for 4 days. Cells were washed twice with PBS and then lysated with 100 μl of lysis buffer (Promega, Madison, CA), following the instructions provided from the manufacturer. The cell lysate was transferred to 96-well plate and lOOμl of substrate/buffer (Promega) were added to each well. The infection of the cells was analysed by measuring the luminescence activity (Chameleon plate reader, Hidex), given in relative light units (RLU). The neutralization activity was determined as percentage of infection, comparing the obtained luminescence to that revealed in the wells with HCVpp in the absence of antibody (neg). A human recombinant Fab (c33) specific for a non- structural antigen of HCV (NS3) was included in each experiment as a negative control.

These data showed a strong neutralizing activity of the e8 Fab, with a 50% neutralization activity (50% inhibitory concentration [IC50]) of 10 μg/ml (Figure 2).

EXAMPLE 3 Cross-competition assay

Considering the data obtained in the preceding example, an important point is the definition of the HCV E2 epitope having the potential to elicit neutralizing cross-reactive antibodies. The e8 Fab is directed against a conformational epitope, as suggested by its inability to bind linear peptides derived from HCV E2 (data not shown).

In order to better investigate the e8 Fab epitope, a competition assay with a large panel of anti-E2 mouse and rat monoclonal antibodies directed against known HCV E2 epitopes was performed. Some of these antibodies (indicated in table 1) are known to have neutralizing activity in the HCVpp assay and to inhibit binding (NOB) of HCV E2.

J

The ability of these antibodies to inhibit the binding of the e8 Fab to HCV E2 was evaluated. Competition experiments were conducted as described (4) with plates coated using HCV E2 at 400ng/ml. The e8 Fab was titrated and used at a concentration shown to give 50% of maximal binding. All competing mouse mAbs were used at a saturating concentration. The results are shown in Table 1. In no experiment was binding of the human e8 Fab inhibited by previous binding of an excess of mouse monoclonal antibody to HCV E2. This indicates that the e8 Fab epitope is outside the region on E2 recognized by this panel of monoclonal antibodies.

Table 1. Inhibition of e8 Fab binding by competing anti-HCV E2 mouse and rat monoclonal antibodies against known regions of the HCV E2 antigen. Data is expressed as % of inhibition of e8 Fab binding. HCVpp neut: HCV pseudoparticles neutralization; NOB: neutralization of binding activity.

EXAMPLE 4 Alanine-scannins mutagenesis

In order to define the critical residues of the HCV E2 sequence involved in e8 Fab binding the inventors tested the reactivity of the e8 Fab on a panel of H77-derived E1E2 (genotype Ia) mutants containing mutations within the putative CD81 binding regions. Each of conserved sites in this region was mutated to alanine (5). All of these substitutions cause the loss of infectivity in the HCVpp assay described below. The binding activity was expressed in terms of percentage of fluorescent positive cells as described above, compared to wild-type (H77) E2 (Figure 3). The binding of the e8 Fab to all of these mutants was not significantly affected compared to the wild-type (H77) E2.

EXAMPLE 5

Antibody screening of phage-display peptide library

To better investigate the epitope bound by the e8 Fab an alternative strategy was used. In particular, a phage-displayed random peptide library of 12 mer was screened on the e8 Fab.

A phage-displayed random peptide library, expressing dodecapeptides at the N-terminus of cpIII coat protein of the filamentous phage Ml 3 (New England Biolabs,Beverly, Ma).The library was screened on the e8 Fab following the instructions provided from the manufacturer. In order to discard peptide-bearing phages specific for Fab conserved region, a negative selection was performed. To this end, during the first round of panning the library of peptides was mixed with an IgGl isotype-matched Fab (el 37) directed against a different epitope on E2 (7). From the second round a more stringent negative selection was performed by mixing the library of peptides with a pool of human standard IgG. The mixture was left 1 hour at 37°C and panned on ELISA plates coated with e8 and the same pool of IgG. Non-specifically absorbed phage was removed by washing with PBS/0.5% Tween-20; (Sigma) Specifically bound peptide-bearing phages were eluted, neutralized, amplified and used for further selection rounds.

After 4 rounds the phage-display selection procedure yielded a 10-fold enrichment of eluted peptide-bearing phage on the e8 Fab compared to eluted phage on the same human IgG pool used in negative selection Afterwards, 50 peptide-bearing phage clones were analyzed in ELISA in order to study their reactivity on the e8 Fab. Phages selected on the e8 Fab from the final round of panning were used to infect E. coli strain ER2537 provided by the library kit and randomly picked single plaques were screened in an enzyme-linked immunoassay on the e8 Fab and a pool of human standard IgG following the instructions provided by the manufacturer. Briefly, antigen-coated plates (Costar) were washed and blocked with PBS/BSA 1% for Ih at 37°C; 50ul of 10⁷ phages per microliter were added and incubated for 2h at 37°C. Plates were washed 10 times with PBS (0.1% Tween-20; Sigma); afterward, 50μl of a 1 :3000 dilution in PBS of a HRP- conjugated anti-M13 antibody were added. After 2h at 37°C plates were washed as above. Specific bound phages were detected by adding lOOμl of substrate (Sigma) and plates were read for OD at 450nm after 30 min at room temperature. Clones showing an absorbance value > 0.5 on e8 compared to value measured on the pool of IgG were scored as positives and evaluated by sequence analysis following the instructions of manufacturer.

31 clones, in total, showed specific reactivity on the e8 Fab. Fifteen of these peptide- bearing clones were evident after three rounds of panning (Figure 4a) and 16 additional peptide-bearing clones were found after 4 rounds of panning (Figure 4b). Importantly, none of these clones bound strongly to the control IgG pool (Figures 4a and b), suggesting that these peptides are not directed against conserved regions of human antibodies and are able to mimic the epitope bound by the e8 Fab.

EXAMPLE 6 Epitope mapping

The sequences analysis of peptide clones able to specifically bind the e8 Fab was performed, and amino acid sequence alignment (Figure 6) showed that several residues were shared across different peptides, leading to the identification of a consensus sequence consisting of six residues LTXPXXLXXXPR (SEQ ID NO:3).

Table 2 below shows the amino acid sequence alignment and reactivity by ELISA against the e8 Fab of peptide-bearing phage clones isolated after three (III) and four (IV) rounds of panning. Binding activity of each clone was tested on the e8 Fab and the pool of human IgG. Conserved residues among different clones are shown in bold type.

In absence of the crystal structure of the HCVE2 glycoprotein, epitope mapping was accomplished using a theoretical 3D structure model based on the E2 glycoprotein of Tick Encephalitis Virus (12,13). Two different programs (Pepitope and Mimox) available online (14,15) were used to compare the sequences of antibody-selected peptides (mimotopes). In detail, peptide sequences were dissected by Pepitope in a set of overlapping amino acids pairs in order to identify specific pairs showing a frequency three standard deviations above randomness. These most frequent amino acid pairs were mapped on the theoretical 3D structure and residues brought together at a distance smaller than 7A were regarded as residues of predicted epitope candidates. In another analysis, Mimox was used to derive a consensus sequence from the alignment of positive clones. Briefly, amino acids appearing at specific positions with a frequency higher than 25% were included in the consensus sequence. Finally the program scanned the E2 structure to search for matching residues on the accessible surface of the antigen. Only residues inside a sphere within a radius of 1 lA were included in the epitope prediction (16). The epitope candidates were ranked on the basis of the sum of residue accessibility.

Five hundred epitopes were initially predicted using these analyses. Considering that the cross-reactivity of the e8 Fab suggests that it recognizes a region of E2 that is conserved across HCV genotypes, we excluded from the analysis all epitopes located in the hypervariable region 1 of HCV E2. Moreover, given the results of the competition assay demonstrating that none of the mouse and rat Mabs utilized were able to compete for e8 Fab binding (Table 1), all epitopes overlapping with regions bound by these Mabs were excluded from the analysis. Finally, those epitopes defined by residues that were shown by alanine-scanning mutagenesis to not be crucial for e8 binding, were excluded. The result of the final analysis revealed a number of overlapping epitopes ranked on the basis of the sum of residue accessibility. The best candidate epitope containing the six residues of the consensus sequence brought to juxtapose in spatial proximity in the protein folding and showing the largest accessible surface on E2 from genotype Ia was centered on residues

L₆₄I , Tό48, P₅12, L₅₈O, P591, R-588.

Considering that e8 shows broad cross-reactivity with the E2 protein of other HCV genotypes, an important point was to investigate if these critical residues are conserved across E2 from different HCVgenotypes. To this end the alignment of E2 amino acid sequences from strains tested in the binding analysis was examined (Figure 6). All 6 residues of the putative epitope were identical across E2 from genotypes Ia, Ib, 2a and 2b. All but one amino acid residue was conserved in genotype 5, while genotypes 3, 4 and 6 had conserved residues at 4 out of the 6 amino acids. . This analysis showed that the consensus sequence (SEQ ED NO:3) shared by the E2 proteins of all HCV genotypes is a good definition of the epitope recognized and bound by the e8 Fab.

CONCLUSIONS

The ability of Hepatitis C Virus to overcome the immune defences of the host establishing in the majority of cases a persistent infection, provides a clear evidence for the beneficial effects played by neutralizing antibodies produced by the host, suggesting that the overall inefficiency of the anti HCV humoral response may be explained by the fact that in the immune response antibodies having neutralizing activity coexist with clones not necessarily useful for the host (1,2). In this context, crucial information for the design of a passive or active immunotherapy relies on the identification of epitopes that are highly shared across different genotypes and targeted by broadly neutralizing antibodies and on the molecular cloning of the neutralizing antibodies themselves.

Studies based on cross-competition assays and alanine-scanning mutagenesis experiments have clustered neutralizing epitopes within two E2 regions involved in binding to CD81, a crucial receptor for virus entry (3). These studies showed that epitopes targeted by neutralizing antibodies involve amino acids that are key contact residues for E2 binding to CD81 (17,18,19,20).

Although many aspects of HCV binding and entry are not completely known, it was widely accepted that CD81 is a necessary but not sufficient co-receptor for HCV entry. Moreover, several cell lines expressing CD81 are not permissive to HCV infection, suggesting that other cell surface molecules are required for HCV entry (21). Therefore epitopes targeted by neutralizing antibodies unable to inhibit the E2-CD81 interaction are of the greatest importance both for an effective immunotherapy and as potential targets for epitope-based vaccines.

However, the definition of conserved human epitopes outside the CD81 -binding region, while crucial for eliciting or administering antibodies with the ability to hit the virus in a different region, has not been described previously. f

Due to the complex and time-consuming experimental work described in the present patent application, the present inventors succeeded in identifying a conformational epitope having the above-mentioned features, which is schematically illustrated in Figure 7. The e8 Fab employed in the present patent application, in fact, proved to be incapable of inhibiting E2 binding to CD81 in a binding neutralisation assay (NOB activity). However, the e8 Fab used in the present patent application possesses a strong neutralising activity for HCVpp bearing E1E2 from genotype Ia. Collectively, these characteristics indicate that the epitope recognized by the e8 Fab includes some regions highly shared across the most common HCV genotypes, crucial for the life cycle of the virus and not directly involved in E2-CD81 binding.

REFERENCES

1. Burioni R, Matsuura Y, Mancini N, Tani H, Miyamura T, Varaldo PE, Clementi M. J Virol.2002;76:11775-9.

2. Zhang P, Wu CG, Mihalik K, Virata-Theimer ML, Yu MY, Alter HJ, Feinstone SM. Proc Natl Acad Sci U S A. 2007; 104:8449-54.

3. Keck ZY, Op De Beeck A, Hadlock KG, Xia J, Li TK, Dubuisson J, Foung SK. J Virol. 2004;78:9224-32.

4. Law M, Maruyama T, Lewis J, Giang E, Tarr AW, Stamataki Z, Gastaminza P, Chisari FV, Jones IM, Fox RI, Ball JK, McKeating JA, Kneteman NM, Burton DR. Nat Med. 2008; 14:25-7.

5. Owsianka AM, Timms JM, Tarr AW, Brown RJ, Hickling TP, Szwejk A, Bienkowska-Szewczyk K, Thomson BJ, Patel AH, Ball JK. J Virol. 2006;80:8695-704.

6. Bugli, F., N. Mancini, C. Y. Kang, C. DiCampli, A. Grieco, A. Manzin, A. Gabrielli, A. Gasbarrini, G. Fadda, P. E. Varaldo, M. Clementi, and R. Burioni. J Virol. 2001;75:9986-9990.

7. Burioni, R., P. Plaisant, A. Manzin, D. Rosa, V. Delli Cam, F. Bugli, L. Solforosi, S. Abrignani, P. E. Varaldo, G. Fadda, and M. Clementi. Hepatology 1998; 28:810-4.

8. Flint, M., C. Maidens, L. D. Loomis-Price, C. Shotton, J. Dubuisson, P. Monk, A. Higginbottom, S. Levy, and J. A. McKeating. J Virol 1999; 73:6235-44.

9. Flint, M., and J. A. McKeating. J Gen Virol 1999; 80: 1943-7.

10. Lavillette D, Tarr AW, Voisset C, Donot P, Bartosch B, Bain C, Patel AH, Dubuisson J, Ball JK, Cosset FL. Hepatology. 2005;41 :265-74.

11. Bartosch B, Dubuisson J, Cosset FL. J Exp Med. 2003;197:633-42.

12. Yagnik AT, Lahm A, Meola A, Roccasecca RM, Ercole BB, Nicosia A, Tramontano A. Proteins. 2000;40:355-66.

13. Spiga,M.Padula,M.Scarselli,A.Ciutti,A.Bernini,C.Falciani, L.Lozzi,et al. Theoretical model of e2 glycoprotein dimer. Available from: http://www.ebi.ac.uk/swissprot/

14. Huang J, Gutteridge A, Honda W, Kanehisa M. BMC Bioinformatics. 2006;7:451.

15. Tarnovitski N, Matthews LJ, Sui J, Gershoni JM, Marasco WA. J MoI Biol. 2006;359: 190-201.

16. Moreau V, Granier C, Villard S, Laune D, Molina F. Bioinformatics. 2006;22: 1088-95.

17. Keck ZY, Li TK, Xia J, Gal-Tanamy M, Olson O, Li SH, Patel AH, Ball JK, Lemon SM, Foung SK. J Virol. 2008;82:6061-6.

18. Owsianka AM, Tarr AW, Keck ZY, Li TK, Witteveldt J, Adair R, Foung SK, Ball JK, Patel AH. J Gen Virol. 2008;89:653-9.

19. Johansson DX, Voisset C, Tarr AW, Aung M, Ball JK, Dubuisson J, Persson MA. Proc Natl Acad Sci U S A. 2007; 104: 16269-74.

20. Perotti M, Mancini N, Diotti RA, Tarr AW, Ball JK, Owsianka A, Adair R, Patel AH, Clementi M, Burioni R. J Virol. 2008;82: 1047-52.

21. Drummer HE, Boo I, Maerz AL, Poumbourios P. J Virol. 2006;80:7844-53.

22. Rowley MJ, O'Connor K, Wijeyewickrema L. Biotechnol Annu Rev. 2004; 10:151- 88.

23. O'Connor KH, Kόnigs C, Rowley MJ, Irving JA, Wijeyewickrema LC, Pustowka A, Dietrich U, Mackay IR. J Immunol Methods. 2005;299:21-35.

Claims

1. A human monoclonal antibody, or binding moiety, capable of binding to the conformational epitope of the HCV E2 glycoprotein from a plurality of different HCV genotypes, as a medicament for the therapeutic treatment or prevention of HCV infections, wherein the conformational epitope of the HCV E2 glycoprotein is defined as follows:

(i) it comprises the amino acid residues L₆₄₁, T₆₄₈, P₅₁₂, L₅₈o, P₅₉₁ and R₅₈₈ of the HCV E2 glycoprotein primary sequence,

(ii) it does not comprise the amino acid residues at positions 384-410 and 474-482 which are located in the hypervariable regions 1 and 2 (HVRl and HVR2) of the HCV E2 glycoprotein, and

CD81 binding, and

(iv) it is capable of binding a peptide comprising the consensus sequence SEQ ID NO:3, wherein the above-mentioned amino acid positions are identified based on the numbering of the primary amino acid sequence of the HCVE2 glycoprotein from genotype 1 a (H77c strain) which is comprised between positions 384 and 746 of the HCV polyprotein precursor sequence (SEQ ID NO:4).

2. The human monoclonal antibody, or binding moiety, according to claim 1 , which is capable of binding to the HCV E2 glycoprotein from HCV genotypes Ia, Ib, 2a, 2b and 3.

3. The human monoclonal antibody, or binding moiety, according to claim 1 or 2, wherein the human monoclonal antibody comprises a heavy chain variable region and a light chain variable region, wherein the amino acid sequence of the heavy chain variable region is SEQ ID NO:1 or a sequence at least 90% identical to SEQ ID NO:1 and wherein the amino acid sequence of the light chain variable region is SEQ ID NO:2 or an amino acid sequence at least 90% identical to SEQ ID NO:2.

4. The human monoclonal antibody, or binding moiety, according to claim 1 or 2, wherein the human monoclonal antibody comprises a heavy chain variable region encoded by the nucleic acid sequence SEQ ID NO:5 and a light chain variable region encoded by the nucleic acid sequence SEQ ID NO:6.

5. A human monoclonal antibody, or binding moiety, according to any of claims 1 to 4, wherein the antibody is a full-size immunoglobulin or an immunoglobulin fragment comprising at least a heavy chain variable region and a light chain variable region.

6. A human monoclonal antibody, or binding moiety, according to claim 5, wherein the antibody is selected from a Fab, a F(ab')₂, a CDR (Complementary Determining Region), or a single chain antibody comprising both heavy and light chain variable regions or CDRs , or scaffolds comprising one or more copies of CDR fragments derived from immunoglobulin heavy and light chain variable regions.

7. A human monoclonal antibody, or binding moiety, according to any of claims 1 to

6, wherein the binding moiety is a peptide or chemical entity that specifically binds to a conformation epitope defined as in claim 1.

8. A human monoclonal antibody, or binding moiety, according to any of claims 1 to

7, wherein the antibody is in a free form.

9. A human monoclonal antibody, or binding moiety, according to any of claims 1 to 7, which is conjugated with a molecule able to modulate the in vivo persistency, promote or limit the body distribution, decrease sensitivity to proteolytic agents, decrease antigenicity, increase cytotoxic ability and/or facilitate detection in body fluid and tissues.

10. A human monoclonal antibody, or binding moiety, according to any of claims 1 to 9, which is fused with a specific full-size immunoglobulin or immunoglobulin fragment capable of targeting the antibody to specific tissues, cells or soluble protein structures.

11. A peptide comprising the consensus sequence SEQ ID NO:3.

12. A pharmaceutical composition for the therapeutic treatment or prevention of HCV infections, comprising a pharmaceutically effective amount of a human monoclonal antibody or binding moiety as defined in any of claims 1 to 10, or a pharmaceutically effective amount of a peptide as defined in claim 11.

13. The pharmaceutical composition according to claim 12, in a pharmaceutical dosage form suitable for parenteral, oral, ocular, topical, loco-regional, enema or aerosol administration.

14. The pharmaceutical composition according to claim 12 or 13, in the form of an injectable solution or suspension, an infusion, a tablet, a capsule, a cream, an ointment, a lotion, or a suppository.

15. The peptide according to claim 11 , as a medicament for the therapeutic treatment or prevention of HCV infections.

16. The peptide according to claim 11 , as a diagnostic or prognostic agent.

17. The use of a conformational epitope as defined in claim 1, or a chemical entity comprising or expressing the conformational epitope as defined in claim 1, for the therapeutic treatment or prevention of HCV infections.

18. The use of a conformational epitope as defined in claim 1, or a chemical entity comprising or expressing the conformational epitope as defined in claim 1, as a diagnostic or prognostic agent.

19. A method of identifying a compound capable of exerting antiviral activity against HCV infections or is suitable for use in the diagnosis of HCV infections, comprising the following steps:

- providing a candidate compound;

- contacting the candidate compound with a peptide as defined in claim 11 or a conformational epitope or chemical entity as defined in claim 17, and - determining whether the candidate compound binds to the peptide as defined in claim 11 or to the conformational epitope or chemical entity as defined in claim 17, wherein a compound that binds to the peptide as defined in claim 11 or to the conformational epitope or chemical entity as defined in claim 17 is capable of exerting antiviral activity against HCV infections or is suitable for use in the diagnosis of HCV infections.