Fax: (61) 3 93471863
A self-adjuvanting multiepitope immunogen that induces a broadly cross-reactive antibody to hepatitis C virus†
Article first published online: 28 MAR 2007
Copyright © 2007 American Association for the Study of Liver Diseases
Volume 45, Issue 4, pages 911–920, April 2007
How to Cite
Torresi, J., Stock, O. M., Fischer, A. E., Grollo, L., Drummer, H., Boo, I., Zeng, W., Earnest-Silveira, L. and Jackson, D. C. (2007), A self-adjuvanting multiepitope immunogen that induces a broadly cross-reactive antibody to hepatitis C virus. Hepatology, 45: 911–920. doi: 10.1002/hep.21538
Potential conflict of interest: Nothing to report.
- Issue published online: 28 MAR 2007
- Article first published online: 28 MAR 2007
- Manuscript Accepted: 17 NOV 2006
- Manuscript Received: 3 JUL 2006
- National Health and Medical Research Council of Australia
- The Australian Centre for HIV and Hepatitis C Virology (ACH2)
- Cooperative Research Centre for Vaccine Technology
We describe a peptide-based strategy for HCV vaccine design that addresses the problem of variability in hypervariable region 1 (HVR1). Peptides representing antibody epitopes of HVR1 from genotype 1a were synthesized and incorporated into multideterminant immunogens that also included lipid moieties and helper T (Th) cell epitopes. Mice inoculated with these polyepitopes generated strong antibody responses. Antibody titers were highest in mice inoculated with polyepitope immunogens which contained the lipid moiety dipalmitoyl-S-glyceryl cysteine (Pam2Cys). Antisera were tested for their potential to neutralize HCV by 3 currently available assays. Antibodies elicited in mice by the polyepitope-based vaccine candidates were able to (1) bind to E2 expressed on the surface of E1/E2-transfected human embryonic kidney (HEK) 293T cells, (2) capture HCV of different genotypes (1, 2, and 3) from the serum of chronically infected humans in an immune capture RT-PCR assay and (3) inhibit HCVpp entry into Huh7 cells. Antibody present in the sera of patients chronically infected with HCV genotypes 1, 2, 3, and 4 also bound to the HVR1-based polyepitope. Conclusion: These results demonstrate the potential of self-adjuvanting epitope-based constructs in the development and delivery of cross-reactive immunogens that incorporate potential neutralizing epitopes present within the viral envelope of HCV. (HEPATOLOGY 2007;45:911–920.)
Hepatitis C virus (HCV) is a human pathogen that has major global significance. It is estimated that 170 million people worldwide and more than 10% of the population in some countries are infected with HCV.1 A range of illnesses have been associated with HCV including acute and chronic hepatitis, cirrhosis of the liver, and HCC. Hepatitis C infection is now the leading indication for liver transplantation in countries such as the United States. In the absence of a vaccine, treatment is confined to the use of the combination of interferon-α and pegylated interferon-α together with the antiviral drug ribavirin.2–5
The development of a vaccine against HCV could prevent HCV-associated diseases. However, there are 2 major impediments to the development of a HCV vaccine: the vast diversity of viral genotypes and quasispecies and the complex nature of the immune response required to clear infection. Antibodies directed to epitopes in the viral glycoprotein E2 (including HVR1) are neutralizing.6–14 Further evidence supporting the protective role of neutralizing antibody was recently highlighted by the demonstration of the presence of neutralizing antibodies in immune globulin prepared from anti-HCV serum.13 This immune globulin protected chimpanzees from infection with HCV genotypes 1a and 2a,13 strengthening the argument that neutralizing antibodies can be cross-protective. In addition, it has been shown that it is possible to prevent entry of retroviral particles pseudotyped with the E1 and E2 glycoproteins (HCVpp) using monoclonal antibodies directed to the HCV glycoproteins or the serum of chronically infected humans and chimpanzees.11, 12, 14, 15 Chronically infected patients have relatively high titers of neutralizing antibody which may demonstrate cross-reactive inhibition of HCVpp entry.12 Indeed, the importance of neutralizing antibody has been further highlighted by the recent report showing that serum from infected humans16 and a monoclonal antibody directed against epitopes in the E2 envelope protein are able to neutralize infectious cell culture-derived HCV.14
Clearance of HCV infection is associated with the development of an early broad and persistent class-1–restricted CD8+ CTL17, 18 and CD4+ T cell19, 20 response to a spectrum of HCV structural and nonstructural proteins.17, 18 However, chronic hepatitis C infection may be associated with impaired function of CD8+ T cells21 and of dendritic cell maturation22 although this has not been demonstrated in vivo.23 Although neutralizing antibody responses are not associated with viral clearance in acute hepatitis C infection, they do have a role in controlling viral replication in patients with chronic hepatitis C.24
We adopted a strategy incorporating a combinatorial epitope library into a single immunogen that uses novel chemistry to develop vaccine candidates with broad coverage against HCV.25–29 In this study, we report the design, assembly and immunological properties of self-adjuvanting multiepitope immunogens that produce broad cross-reactive antibodies capable of binding to HCV E1/E2 proteins expressed at the surface of transfected cells that are also capable of capturing different HCV quasispecies circulating in the serum of infected carriers. This approach provides a strategy for the development of a preventative vaccine against HCV.
Materials and Methods
HVR1 Peptide Library.
A peptide library of 10-mer sequences was synthesized manually using Fmoc chemistry and a strategy that introduced mixtures of amino acid at points of sequence variability within HVR1 such that a library of epitopes containing variant sequences was assembled. Four sequences TTTTGGQVGH, TTTTGGVAAH, THVVGGSQSR and THTTGGQAGH representing amino acids 2 to 12 of HVRI10, 30 were used to generate the peptide library. At each position of amino acid variability the solid phase support was divided into equal amounts corresponding to the number of different amino acids at that position (Table 1). To the individual aliquots of solid support was then coupled one of the relevant Fmoc amino acids. Following acylation the divided solid support portions were recombined, washed, the Fmoc group removed and peptide synthesis continued, until a position of variation was again reached at which point the solid support was again divided and individual and different amino acids incorporated. This strategy results in the assembly of a peptide library containing 288 different peptide sequences, which included the 4 parent sequences. Together these sequences represent 80% of the reported (GenBank) variability within HVR1.
|HCV H77 1a||T T T T G G Q V G H|
|Sequence 1||– – – – – – – – S –|
|Sequence 2||– – V V – – S Q S –|
|Sequence 3||– H – V – – S – A R|
|Sequence 4||– H – – – – – A – –|
Peptides were synthesized in-house using previously described methodologies.31 A peptide representing the HVR1 sequence of HCV H77 (ETHVTGGSAGRTTAGLVGLLTPGAKQN) reported by Farci and co-workers9 was synthesized in tandem with the helper T cell epitope (Th) KLIPNASLIENCTKAEL by chemoselective ligation using polyoxime chemistry.32 The helper T cell epitope is from the F protein of morbillivirus33 and has been found to be an efficient source of T cell help in BALB/c, CBA/H and C57Bl/6 strains of mice (data not shown).
Peptide Analysis and Purification.
Analytical HPLC was performed using a Vydac C4 column (4.6 mm × 250 mm) installed in a Waters HPLC system. All peptides, except the peptide library, were purified using a Vydac Protein C4 column (10 mm × 250 mm) installed in a Pharmacia Fast Performance Liquid Chromatography (FPLC) system. Quality assurance of peptides was determined by analytical HPLC and mass spectrometry using an Agilent Technologies 1100 series ion trap mass spectrometer configured in the positive mode and electrospray as the ion source. All peptides were isolated as single symmetrical peaks on HPLC and displayed the expected mass.
In the case of the peptide library, HPLC analysis showed a profile consistent with the presence of a large number of individual peptide species. Mass spectrometric analysis demonstrated that the peptide species within the library clustered around a mass of approximately 985 Da for the single charged species and 490 Da for the double charged species (results not shown).
Synthesis of Pam2Cys.
Pam2Cys was synthesized using modifications to the method of Wiesmuller et al.34 as reported by Zeng et al.31 In order to incorporate lipid moieties into polyepitope constructs Pam2Cys was assembled with 12 lysine residues (polyK-Pam2Cys) to confer solubility. The N-terminal lysine was then acryloylated to enable copolymerization of lipid and peptide to form lipidated polyepitopes.
Polymerization of N-Acryloylated Peptides.
Before assembly of peptides into polyepitope immunogens Fmoc-6-aminohexanoic acid (Ahx) was manually coupled to the N-terminus of peptides using HBTU, HOBT and DIPEA. The Ahx group acted to space the peptide from the polyepitope backbone. Individual peptides were polymerized and the products purified as described.25–28 Briefly, peptides were acylated with acryloyl chloride at their N-termini while still attached to the solid support and with side chain protecting groups intact using acryloyl chloride. Acryloylation of the N-terminus of resin-bound protected peptides was carried out manually in anhydrous, de-aerated N,N′-dimethylformamide (DMF) at 0°C. A 20-fold molar excess of diisopropylethylamine (DIPEA) in DMF and a 10-fold molar excess of acryloyl chloride (in DMF) were added and the reactants mixed for 1 hour on ice and then for a further 1 hour at room temperature. Following cleavage from the support polymerization of N-acryloylated peptides was carried out in 0.5 M Tris (pH 8.3) containing 6 M guanidine-HCl and 2 mM EDTA. Assembly of peptide polyepitopes was typically carried out with 20 μmol of N-acryloyl-peptides and 50-fold molar excess of acrylamide. Polymerization was initiated by the addition of ammonium persulfate (20 μmol) and TEMED 10%; the reaction was allowed to take place overnight at room temperature.
Four polyepitope constructs were assembled (Fig. 1). The first contains only peptides from the peptide library and is called polyepitope. The second construct contained peptides from the library that were copolymerized with the helper T cell epitope Th; this construct is called polyepitope-Th. The third construct contained peptides from the library that were copolymerized with Pam2Cys and is called lipidated polyepitope. The fourth construct contains peptides from the library that were copolymerized with Th and Pam2Cys and is referred to as lipidated-polyepitope-Th.
Polyepitope and lipidated polyepitope constructs were purified by FPLC according to previously published methods.27 The polyepitope-Th and lipidated-polyepitope-Th formed viscous solutions and could not be purified by FPLC. These polyepitopes were therefore dialyzed exhaustively against water. Nondiffusible material eluted in the void volume of a Protein Pak 300 column (0.78 cm × 30 cm) (Waters, Australia, Sydney, Australia), such that high molecular weight polyepitopes were separated from low molecular weight excipients.29 Each of the immunogens were freeze-dried and stored at room temperature until required.
Inoculation of Mice.
Female Balb/c, CBA and C57Bl/6 mice, aged 6-8 weeks were obtained from the animal facility within the Department of Microbiology and Immunology and were used to examine the immunogenic properties of the polyepitope preparations. Nonlipidated preparations, polyepitope, polyepitope-Th, and HVR1-Th were emulsified in Freund's complete adjuvant for the initial inoculation and in Freund's incomplete adjuvant for subsequent doses.
Two lipidated polyepitopes—lipidated polyepitope and lipidated-polyepitope-Th—were administered in PBS, i.e., in the absence of exogenous adjuvant. Mice were inoculated subcutaneously in the base of their tails with a dose of immunogen containing 20 nmol of peptide determinant(s). Animals received the repetition doses of vaccine candidate at twice-weekly intervals and were bled prior to the initial inoculation and 2 weeks following subsequent doses.
Purification of Anti-HCV Polyepitope Mouse IgG.
Mouse IgG was purified from sera using Protein G Sepharose 4 Fast Flow (Amersham Biosciences, Piscataway, NJ). We added 25 μl of a 50% Protein G Sepharose slurry (swollen Protein G Sepharose beads, suspended in an equal volume of 20 mM sodium phosphate, pH 7.0, buffer) was added to a 1.5 ml microcentrifuge tube. We added 10 μl of pooled mouse sera to the Protein G Sepharose and held the mixture at room temperature for 30 minutes with agitation every 5 minutes. The solid support was then washed twice with 20 mM sodium phosphate buffer and the Protein G Sepharose sedimented by brief centrifugation. The supernatant was removed and 100 μl of 0.1 M acetic acid added to the Protein G Sepharose. This was then agitated for 5 minutes, centrifuged briefly, and the supernatant removed to a new microcentrifuge tube. The eluate was neutralized by the addition of 400 μl of 1M Tris-HCl, pH 8.0. The eluate was then concentrated 10-fold using a Centricon YM-100 filter (Millipore, Bedford, MA).
Recombinant E2 Glycoprotein.
Recombinant truncated E2 protein was produced using the plasmid pCDNAtpaE2-661myc.35 This plasmid encodes the HCV E2 ectodomain residues 384–661 of genotype 1a HCV-H and contains a C-terminal myc epitope tag that is secreted into culture supernatants of transfected cells. HEK 293T cells grown in OptiMEM (Gibco, Grand Island, NY) were transfected using 5 μg pCDNAtpaE2-661myc DNA and Fugene6 (Roche, Mannheim, Germany) in a ratio of 2:1. After 3 days, the culture supernatants were collected and clarified by centrifugation at 2205g for 10 minutes. Clarified supernatants were then filtered through a 0.45 μm filter (Pall Corp., Ann Arbor, MI) and concentrated by ultrafiltration using a Centricon YM-30 filter (Millipore). Recombinant E2 was quantitated by Lowry assay (Bio-Rad Laboratories, Richmond, CA).
Enzyme-Linked Immunoadsorbent Assay.
HCV-containing sera were obtained from patients with chronic or convalescent hepatitis C infection who attended the hepatitis clinic at the Royal Melbourne Hospital, Melbourne, Australia. Approval to collect blood samples was obtained from the Human Research and Ethics Committee of the Royal Melbourne Research Directorate. Serum samples were determined to be positive or negative for hepatitis C by determining the presence or absence of antibody using a commercial kit (Axsym HCV 3.0, Abbot) and the presence or absence of viral RNA by Roche Amplicor. Determination of HCV genotype was determined by the Department of Microbiology, Royal Melbourne Hospital, Australia.
FACS Analysis of E2 Expressed on the Surface of Transfected Cells.
Twenty-four hours before transfection, 6-well dishes (Nunc) were seeded with 3.0 × 105 HEK 293T cells per well and incubated at 37°C with 5% CO2. The cells were transiently transfected using 2 μg of either pcDNA4 His MaxC or pcDNA 4 His MaxC E1E2 (HCV genotype 1a) and Fugene 6 reagent in a ratio of Fugene to DNA of 3:1. At 48 hours after transfection, cells were harvested in fluorescence-activated cell sorting (FACS) buffer (1% BSA, 0.02% NaN3 in PBS). Cells (5 × 105) were stained with either anti-lipidated-polyepitope-Th serum (diluted 1:50), naïve serum (diluted 1:50), goat anti-E2 antibody (Virogen) (diluted 1:100), or monoclonal antibody H53 (provided by J. Dubuisson, Lille, France, and diluted 1:100). The total volume of cells with antibody was 200 μl and staining was performed at 4°C in the dark for 30 minutes. Cells were then washed twice in FACS buffer and counter stained for 30 minutes at 4°C in the dark with Alexa anti-mouse or anti-goat 488 which was diluted 1:500. Cells were washed twice, resuspended in 200 μl of FACS buffer, and binding of antibodies was examined by fluorescence-activated cell sorting on a FACSCalibur (Becton Dickinson) instrument.
HCV Immune Capture RT-PCR.
PCR tubes (Axygen, Union City, CA) were coated with 100 μl (5 μg/ml) of anti-HVR1 peptide IgG or anti-lipidated-polyepitope-Th diluted in sterile PBS and incubated at 37°C for 1 hour followed by 4°C overnight. Plastic unoccupied by IgG was blocked by addition of 200 μl of 10% skim milk powder in 0.5% Tween 20/PBS. Tubes were washed 3 times with buffer 1 (0.05% Tween 20, 0.02% of NaN3 in PBS) before adding 65 μl PBS and 25 μl of viremic human serum and incubating at 4°C for 1 hour. Tubes were washed 4 times with buffer 2 (25 mM Tris-HCl pH 8.0, 300 mM NaCl) before extracting viral RNA using the QIAamp Viral RNA Mini Kit (Qiagen, Valencia, CA). Viral RNA was used as template and HCV5′ UTR ER (see below) as the reverse primer to produce cDNA with Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA) according to the manufacturer instructions and followed by nested PCR. First-round PCR was performed using 12.5 μl of cDNA as template, 1.25 μl each of 2.5 mM dATP, dCTP, dGTP, dTTP, 3 μl of 25 mM MgCl2, 1 μl of (100 pmol) HCV5′ UTR EF primer (CCATGGCGTTAAGTATGAGTG), 1 μl of (100 pmol) HCV5′ UTR ER primer (TGCACGGTCTACGAGACCT), 22 μl of H2O, 5 μl of 10× PCR Buffer (Sigma, Steinheim, Germany), and 0.5 μl of (1 U/μl) RedTaq (Sigma). First-round PCR cycle conditions included 5 minutes at 95°C, 30 cycles of 15 seconds denaturation at 95°C, 20 seconds annealing at 48°C and 25 seconds extension at 72°C followed by a final extension of 10 minutes at 72°C. Second-round PCR was performed with 1 μl of the first-round PCR reaction as template together with 1.25 μl each of 2.5 mM dATP, dCTP, dGTP, dTTP; 3 μl of 25 mM MgCl2, 33.5 μl H2O, 1 μl (100 pmol) HCV5′ UTR NF primer (AGTGTCGTGCAGCCTCCAGG), 1 μl (100 pmol) HCV5′ UTR NR primer (CACTCGCAAGCACCCTATC), 5 μl of 10× PCR Buffer and 0.5 μl of RedTaq (Sigma) in a final reaction volume of 50 μl using the same cycle conditions as above to generate a 235-bp product.
E1E2-HIV-1 Pseudotype Particle Entry Assay.
HCV pseudoparticles were prepared according to Drummer et al.15 The HIV-1 luciferase reporter vector NL4-3.LUC.R-E- was obtained from Dr N. Landau through the NIH AIDS Research and Reference Reagent program.38 Briefly, 3 × 105 293T cells, seeded into wells of 6-well culture plates, were transfected with 1 μg each of NL4-3.LUC.R-E- and either pE1E2H77c (HCV genotype 1a) or pCDNA4HisMax with Fugene 6 transfection reagent (Roche, Indianapolis, IN). Three days later, tissue culture fluid containing the HCVpps was collected and filtered (0.45 μm). Dilutions of heat-inactivated (56°C, 20 min) mouse antisera in DMF10 were incubated with an equal volume of HCVpps for 1 hour before addition to Huh7 cells (30,000/well) in 48-well culture plates. After 4 hours incubation at 37°C, the inoculum was removed and cells were cultured for a further 3 days. Luciferase activity was measured in a Polarstar microplate reader fitted with luminescence optics (BMG Labtechnologies) using the Promega luciferase reagent system. Percentage viral particle entry was calculated by using the expression: (luciferase units in the presence of sera)/(luciferase units in the absence of sera) × 100.
Statistical analyses were performed using a nonparametric, 1-tailed Mann-Whitney test with 95% confidence interval.
Reactivity of Human Sera to Polyepitope.
We determined whether antibodies present in serum from patients chronically infected with HCV could recognize epitopes incorporated into polyepitope. A panel of sera collected from 40 patients with chronic hepatitis C infection was tested for reactivity against polyepitope (Fig. 2). Four broad groupings were identified. Three patients infected with genotypes 1a, 3a, and 4 exhibited high reactivity, with antibody titers > 5.0Log10. Four patients infected with genotype 1a and 2 with 3a had intermediate reactivity, their antibody titers ranging between 2.0 Log10 and 3.6 Log10. An additional 7 patients demonstrated low reactivity to polyepitope with antibody titers 2.0 Log10; three of these patients were infected with genotype 3a, 1 with genotype 1, 1 with genotype 1a/1b, 1 with genotype 2a/2c, and 1 with genotype 4. The remaining 26 patients were seronegative. There were no distinguishing clinical, virological (viral genotype), laboratory, or histological (liver) features noted among the 4 patient groups. These results highlight the cross-reactivity of sera from patients infected with different genotypes of HCV for epitopes contained within polyepitope, although it does not reflect neutralizing activity of that antibody.
Immunogenicity of Polymerized N-Acryloylated HVR1 Peptides.
The ability of the various epitope-based vaccine constructs to elicit antibody responses was examined by assessing antibody titers in BALB/c mice that had been inoculated with 20 nmol of the different polyepitopes. Antibody titers were analyzed by ELISA using antisera obtained 2 weeks following the fourth dose of vaccine. Polyepitope was used as the coating antigen. The results (Fig. 3) indicate that each of the polyepitopes were immunogenic although the highest antibody titers were achieved after inoculation with self-adjuvanting lipidated-polyepitope-Th. The hierarchy of antibody responses achieved was lipidated-polyepitope-Th > polyepitope-Th > lipidated-polyepitope = polyepitope > peptide library. Because polyepitope-Th was administered in Freund's adjuvant, these results indicate that the endogenous Pam2Cys is a superior adjuvant. It should also be noted that no site reactions were apparent with the self-adjuvanting vaccine candidates, which is consistent with our previous findings using Pam2Cys.39
The antibodies elicited by lipidated-polyepitope-Th were also tested for their ability to bind to a synthetic peptide representing the sequence of HVR1 or to a recombinant E2 protein (Table 2). Antibodies raised against this immunogen bound well to each of these antigens confirming the ability of anti-lipidated-polyepitope-Th antibodies to bind to HVR1 presented in the context of E2 protein as well as to the HVR1 peptide itself. Because the lipidated-polyepitope-Th produced the highest antibody titers in mice the anti-lipidated-polyepitope-Thantiserum was used in all subsequent analyses.
|Coating antigen||Antibody titer (log10)|
|Polyepitope||4.853 ± 0.036|
|Recombinant E2||2.893 ± 0.023|
|HVR1 synthetic peptide||2.847 ± 0.022|
Binding of Anti-Lipidated-Polyepitope-Th Antibody to E1E2 Expressed at the Cell Surface.
E1E2 glycoprotein dimers expressed at the surface of transfected HEK 293T cells are indistinguishable from the glycoforms of E1E2 incorporated into HCVpps11, 15 and represent a functional form of the viral glycoprotein complex that is able to mediate viral entry into Huh7 cells. Binding of anti-polyepitope antibodies to E2 expressed on the surface of E1E2-transfected HEK 293T cells was examined and the pattern of binding compared to a commercially available antibody and the anti-E2 monoclonal antibody H53.
By using FACS analysis, we were able to show that anti-lipidated-polyepitope-Th antibody bound to 14% of transfected cells compared to 32% obtained with monoclonal antibody H53, 22% with the Virogen anti-E2 antibody and 4% with naïve serum. These results demonstrate that the anti-lipidated-polyepitope-Th induced an antibody response that included specificities directed against epitopes present in the sequence of HVR1 of HCV genotype 1a H77 and provide good supporting evidence for the potential of the polyepitope immunogens to produce antibody responses that are able to recognize E2 expressed in mammalian cells.
Capture of HCV by Anti-Lipidated Polyepitope-Th Antibodies.
The functional importance of antibodies raised against epitopes present in E2 can also be demonstrated by the ability of antibody to bind to mature virions present in the serum of patients infected with HCV.40 In order to determine whether anti-Lipidated-Th Polyepitope antibodies are able to bind to HCV in viremic human serum we tested anti-Lipidated-Th Polyepitope IgG in an immune capture RT-PCR. Viremic serum collected from 15 HCV PCR positive patients was analysed; four of these patients were infected with genotype 1a, two were infected with 2a and nine patients with 3a. The anti-Lipidated-Polyepitope-Th IgG captured virus of each of the three genotypes present in the serum of seven of these patients. In contrast, IgG obtained from mice prior to inoculation with the vaccine candidate failed to capture HCV from the serum of these patients. The results (Fig. 4) of a representative assay show that anti-lipidated-polyepitope-Th captured virus from the sera of 5 patients that were chronically infected with either, HCV genotype 3a, genotype 1a or genotype 2a. The variation in the ability of the anti-lipidated-polyepitope-Th IgG to capture HCV was not a reflection of differences in serum HCV viral loads; the mean viral load of samples that produced a positive (2.5 × 106 copies/ml [±SD 3 × 106]) result by immune capture was not significantly different to those of negative samples (1.2 × 106 copies/ml [±SD 1.15 × 106]; P > 0.4). These results show that a broad range of epitopes can be assembled into a polyvalent vaccine capable of inducing antibody populations that are cross-reactive with different genotypes of HCV. HCV bound to IgG in circulation is not infectious41 and the result of the immune capture RT-PCR is encouraging because the ability of antibody to capture HCV is an indication of its potential to neutralize virus.42
Inhibition of HCVpp Entry into Susceptible Cells.
Antibodies elicited by lipidated polyepitope-Th vaccine candidate was also able to inhibit the entry of HCVpps into Huh7 cells (P < 0.01) (Fig. 5). No such inhibition was observed with the preimmune mouse antibodies. Although the antibody titer elicited by lipidated polyepitope-Th was significantly higher than those elicited by Th-HVR1 (Table 2), anti-Th-HVR1 produced greater inhibition of HCVpp cell entry. This demonstrates that antibody detected by ELISA does not necessarily reflect the ability of such antibody to neutralize HCV. The inhibition of HCVpp cell entry by the anti-lipidated polyepitope-Th however reinforces the biological activity of antibody elicited with this vaccine candidate and its potential to neutralize HCV.
Neutralizing antibody responses clearly develop as a consequence of HCV infection.11–13, 43 with the host responding to emergence of HCV variants by the sequential development of variant-specific antibodies that are able to neutralize homologous and closely related HCV types.42 The dilemma facing vaccine designers is how to deal with the immense variability that occurs within potential neutralizing regions such as HVR1.
By making use of published consensus sequences,30, 44 we devised a method for generating a peptide library that covers approximately 80% of the reported variability within HVR1 and have incorporated these sequences into a polyvalent vaccine candidate. Our rationale for using sequences corresponding to the N-terminus of HVR1 was based on the observation of Zibert and co-workers who have shown an association between clearance of hepatitis C virus and the development of an antibody response directed predominantly to the N-terminus of HVR1.10 Although this association does not prove a causal link, the finding that an antibody response to the C-terminus of HVR1 is associated with chronic infection reinforces the importance of anti-N-terminal antibodies. Epitopes present at the C-terminus of HVR1 may therefore represent immunodominant decoys to be avoided in the design of vaccines.45 These polyepitope constructs elicited high antibody titers to multiple peptide epitopes with the response being augmented by the copolymerization of the peptide library with a promiscuous T helper epitope and particularly by the incorporation of Pam2Cys as an endogenous adjuvant.
High-density lipoprotein (HDL) has been shown to inhibit the activity of neutralizing antibodies through its interaction with SR-B1.46 The interaction of HDL/SR-B1 with HCV only appears to inhibit antibodies that block the binding of CD81 to HCV-E2. Antibodies that do not inhibit the binding between E2 and CD81 are not affected by the presence of HDL in human serum. For example, antibodies to HVR1 and E1 had the same IC90 in the presence and absence of human serum.46 Recently, the CD81 binding motifs in the HCV E2 glycoprotein have been described and are located downstream of HVR1.47 Our Polyepitope only includes sequences from the N-terminal part of HVR1; therefore, the polyclonal antibodies produced against this vaccine are unlikely to interact with CD81 or SR-B1 and therefore the neutralizing antibodies induced by this vaccine are unlikely to be affected by HDL in human serum.
The antibody response induced by the lipidated-polyepitope-Th was strong and broad and the functional potential of the antibody was demonstrated in 3 different assays. Furthermore, the ability of antibodies elicited by the Lipidated-Polyepitope-Th to bind to E2 expressed at the cell surface provides an additional assessment of whether these antibodies are likely to bind to E2 on the surface of HCV virions.11, 12, 15
In order to determine whether antibodies raised to lipidated-polyepitope-Th can bind to native hepatitis C virus, an immune-capture RT-PCR assay was developed. Although immune-capture of HCV does not predict neutralizing activity of antibody it has been shown that neutralizing antibodies are able to bind homologous and closely related viral quasispecies in an immune capture RT-PCR while antibodies that are not neutralizing fail to do so.40, 42, 48 In addition, the dilution of antibody required to capture HCV closely parallels the neutralizing antibody titers in the serum of chronically infected chimpanzees.42 Immune capture is therefore a useful adjunct to approaches designed to examine the neutralizing activity of antibody.40
Supportive evidence for the potential of the anti-lipidated-polyepitope-Th to neutralize HCV was provided by the inhibition of HCVpp entry. The level of inhibition achieved with this antibody was lower than the level observed with the anti-HVR1-Th. This is not unexpected because the lipidated Th-polyepitope potentially contains 288 different epitopes derived from the HVR1 sequence and only some of these will be cross-reactive with the sequences represented by the HCVpps, recombinant E2 and synthetic HVR1.
We reported29 a vaccine strategy that allows different epitopes to be assembled into a polyepitope structure which then induces neutralizing antibody against multiple and different serotypes of group A streptococcus.29 In the present study we have applied this technology to different genotypes and quasispecies of HCV and in addition have shown for the first time that a polyepitope vaccine can be self-adjuvanting if the lipid moiety Pam2Cys is incorporated. In the context of single epitope-based vaccines the lipid has been shown to target dendritic cells31 and also to activate them through Toll-like receptor-2.49 It is very likely that the multivalent anti-HCV candidate vaccines that we described here act by a similar mechanism.
Epitope-based strategies in which the self-adjuvanting properties of lipid structures are exploited provide potential for the development of effective vaccines that incorporate cross-protective neutralizing epitopes. The extreme viral diversity that exists with HCV presents a significant problem for the design of a HCV vaccine. However, an epitope-based strategy has the ability to overcome this limitation and provides promise for the development of an effective vaccine against HCV. In the future, it would be important to dissect out the ability of anti-polyepitope antibodies to cross-neutralize not only different genotypes of HCV but also across quasispecies variants.
We thank Prof C Rice, Rockefeller University, New York, NY, and Dr Jean Dubuisson, Institut de Biologie de Lille & Institut Pasteur de Lille, Lille, France, for the generous provision of reagents.
- 2Randomised trial of interferon alpha2b plus ribavirin for 48 weeks or for 24 weeks versus interferon alpha2b plus placebo for 48 weeks for treatment of chronic infection with hepatitis C virus. International Hepatitis Interventional Therapy Group (IHIT). Lancet 1998; 352: 1426–1432., , , , , , et al.