Introduction

Hepatitis C virus (HCV) currently infects 170 million people worldwide [1, 2]. Hundreds of thousands of individuals die each year from infection-induced liver failure and cancer. There is no current vaccine, and typical anti-HCV drugs that are used are expensive and can induce significant side effects [3]. Progress in HCV vaccine development has been hindered by the limited availability of cell culture and small-animal models.

Recent advances have enhanced the prospect of developing a new generation HCV vaccine [4]. One method that helps boost vaccine efficacy is the construction of multi-epitope vaccines. In recent years, there have been many studies indicating that virus-specific cytotoxic T lymphocytes (CTLs) play a critical role in the host cellular immune response against HCV [5, 6]. Many significant CTL epitopes have been identified in the most conserved HCV protein, the core protein, and these are mainly located at the amino terminus [7]. Therefore, the truncated HCV core gene is believed, as a candidate vaccine eliciting HCcAg-specific CTLs, to be useful for both therapy and protection against HCV infection [8]. Although hypervariable region 1 (HVR1) of the envelope protein E2 of HCV is the most variable antigenic fragment in the whole viral genome, it contains the principal neutralization epitopes, and anti-HVR1 antibodies are the only species shown so far to possess protective reactivity. The development of a prophylactic vaccine has been hindered because of the hyper-variability in HVR1, but research has indicated that mimotopes of HVR1 can induce antibodies that cross-react with a large number of viral variants [9], which is a better solution against hepatitis C virus hypervariability. It has been confirmed that mimotopes of HVR1 react with 53.3–80% of sera from HCV patients [10]. So it is an idealized strategy to select the mimotopes of HVR1 for vaccine design. Virus-specific CTLs play a critical role in preventing transmission and clearance of HCV during infection in infected individuals [11]. There are many dominant T-cell epitopes, and especially HLA-A2-restricted CTL epitopes in the non-structural region of HCV have been discovered in recent years. These epitopes have been shown to effectively induce an HCV-specific cellular immune response [12–17]. Therefore, an HCV multi-epitope vaccine would be available if applied in a suitable vehicle. Bacillus Calmette-Guerin (BCG), as an attractive live organism vaccine vehicle, has been successfully used to deliver foreign viral, bacterial, and parasite proteins and can elicit specific immune responses to recombinant proteins from many different pathogens [18–22].

In the present study, a novel multi-epitope antigen (CtEm) was designed and expressed in BCG. Our objective was to design a chimeric antigen composed of HCV structural and non-structural immunodominant epitopes, including the major epitope domains of the truncated core, mimotopes from E2, and six HLA-A2-restricted CTL epitopes from NS3, NS4 and NS5, which are conserved in different genotypes. The chimeric antigen was introduced into BCG, and HLA-A2.1 transgenic mice were immunized with recombinant BCG (rBCG). Our results indicate that HCV-specific humoral and cell-mediated immune responses were elicited, and the immunized mice were protected from challenge with the HCV recombinant vaccinia virus.

Materials and methods

Vector construction

The gene encoding signal peptide α-ss antigen and Hsp60 promoter sequences was obtained by PCR from H37Rv MTB strains and inserted into vector pOLYG (Multicopy E. coli-Mycobacterium shuttle vector; Hyg) [23] to construct the secreting-type vector pDE22. The primers for α-ss are: 5′-GC TGG CCA (BalI)CGC GCC CGC AGT TGC-3′ (sense) and 5′-GAA GGA TCC (BamHI) CAG ACG TGA GCC GAA AG-3′ (antisense), and the primers for Hsp60 promoter sequences are: 5′-GC TCT AGA (XbaI) CGG TGA CCA C-3′ (sense) and 5′-GG TGG CCA (BalI) TTG CGA AGT GA-3′ (antisense). The cDNA sequence of the truncated HCV core (aa 1-144) gene (Ct) was amplified by PCR using the primers 5′-GCC GGA TCC (BamHI) ATG AGC ACG AAT CCT AAA CCT-3′ (sense) and 5′-GC AAG CTT (PstI)TTA GCC ATG CGC CAG GGC CCT-3′ (antisense) from pBRTM/HCV1-3011 (genotype 1a), Ct was inserted into the shuttle plasmid, pDE22, after digestion with BamHI/PstI, to yield pDE22-Ct. The epitope-encoding gene Em (510 bp), composed of mimotopes from E2 and six HLA-A2-restricted CTL epitopes from non-structural regions, was synthesised with an A-A-Y spacer and PstI and HindIII sites at the ends (Table 1). The Em gene was inserted into pDE22-Ct after digestion with PstI and HindIII, and the recombinant vector pDE22-CtEm was obtained (Fig. 1). The pDE22-CS1 (containing HBV core and preS1) was used as a control [24]. CtEm was digested with BamHI and HindIII and sub-cloned into pcDNA3.1 and pQE30, creating the recombinant vectors, pcDNA3.1-CtEm and pQE30-CtEm.

Table 1 Choice of E2 mimotopes and HLA-A2-restricted CTL immunodominant epitopes from the HCV non-structural region
Full size table
Fig. 1
figure 1

Construction of the recombinant secreting-type vector pDE22. The Hsp60 promoter and signal peptide sequence were inserted into the pOLYG vector between the XbaI and BalI and BalI and BamHI sites, respectively, to give rise to the secreting-type vector pDE22. The truncated HCV core gene (Ct) was inserted into pDE22 between the BamHI and PstI sites, and the synthetic multi-epitope gene (Em) was inserted between the PstI and HindIII sites. The recombinant vector pDE22-CtEm was thus obtained

Full size image

Peptide synthesis

Peptides containing mimotope E2 and CTL epitopes from nonstructural regions were synthesized locally using Fmoc chemistry and purified by high-performance liquid chromatography (HPLC; >90%). Peptides were solubilized in water at a concentration of 1 mg/ml. Peptide HBs183–191 (FLLTRILTL) was synthesized and used as a control in this work.

Preparation of rBCG and induction

The plasmids pDE22-CtEm and pDE22-CS1 were introduced into BCG Tokyo strain by electroporation as previously described [25], giving rise to rBCG-CtEm and rBCG-CS1, respectively. These transformants were selected on a Middlebrook 7H10-agar (Difco, USA) plate containing 50 μg/ml of hygromycin and grown in Middlebrook 7H9-albumin dextrose complex (Difco) broth for 3 weeks. The rBCG cells were harvested, washed twice with PBS or physiological saline, and used for immunization. Before harvest, part of the rBCG, cells were induced for 1 h at 42°C, and the culture supernatant was collected and concentrated with PEG4000 (1:100 concentration). After dialysis with PBS, the concentrated protein was used for Western blot assay.

Expression of the multi-epitope antigen and Western blot assay

The target protein was expressed after being transformed into E. coli M15 with pQE30-CtEm and induced with 1 mM IPTG. The protein was purified on a Ni2+-NTA column, and mice were immunized to obtain mouse anti-CtEm polyclonal antibodies [26]. Purified CtEm protein and the concentrated supernatant of the rBCG cell culture medium were subjected to sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) using a 12% gradient gel (Amresco, USA). Fractionated proteins were electroblotted onto a PVDF membrane, incubated first with mouse anti-CtEm polyclonal antibodies, and then with IRDye800-conjugated goat anti-mouse IgG (LI-COR, Odyssey, USA), and visualized using Odyssey (USA).

Mice and immunization

Six-to eight-week-old HHD-2 mice, transgenic for HLA-A2.1 major histocompatibility complex (MHC) class I and deficient for both H-2Db and murine β2 microglobulin (β2 m) [27] were used. The mice were randomly divided into six equal groups (ten per group). One group was inoculated twice 1 × 106 CFU/100 μl rBCG-CtEm via the belly subcutaneous route with a 1-month interval, one group was inoculated with rBCG-CS1 control, and two groups were immunized with 100 μg/100 μl pcDNA3.1-CtEm and pcDNA3.1 control, respectively, via multidrop, inoculated intramuscularly in the anterior tibial muscle three times at 2-week intervals, with BCG and saline as controls. Two weeks after the first inoculation, the mice were bled from the vena caudalis every two weeks, and the sera were isolated and stored at −20°C until use.

Antibody assay

Polystyrene 96-well plates (Nunclon, Denmark) were coated with purified fusion proteins CtEm or mimotope E2 0.1 μg (100 μl per well) overnight at 4°C and washed five times with a solution of 0.05% Tween 20 in PBS and then blocked with 1% BSA overnight at 4°C. After washing five times, serum samples (100 μl per well) were dispensed to wells at different dilutions with PBS-Tween 20 buffer containing 0.5 M NaCl and incubated for 1 h at 37°C. After washing five times, HRP-conjugated goat anti-mouse IgG (Sigma), diluted 1:10,000, was then added, and plates were incubated for 1 h at 37°C. After washing, the color was developed using TMB according to standard procedures, and the optical density values were measured at 490 nm (OD490) in a Synergy HT Multi-Detection Microplate Reader (Bio-Tek, USA).

Assay of HCV specific lymphocyte proliferation

Mouse spleens were separated, and the lymphocytes were isolated and suspended in RPMI-1640 containing 10% fetal cattle serum. 1 × 106 spleen lymphocytes with 1 μg purified CtEm, synthetic peptide or control protein CS1 were incubated at 37°C with 5% CO2 for 68 h. The cells were pulsed with 1 μCi of [methyl-3H] thymidine per well for the last 18 h and harvested on fiber filters. Incorporation of the radiolabelled nucleotide into DNA was measured by liquid scintillation counting. Proliferation was expressed as the proliferation index (PI), calculated as the mean counts in triplicate test wells divided by the mean background counts in triplicate control wells.

Cytokine detection

Splenic lymphocytes (1 × 106) were cultured with or without 1 μg CtEm in RPMI 1640 containing 10% fetal cattle serum (800 μl per well) at 37°C in a 5% CO2 incubator. Culture supernatants were collected at day 3, and IFN-γ was quantified using a commercially available ELISA kit (MABTech, Sweden) and biotinylated anti-mouse IFN-γ. A recombinant IFN-γ standard was used to generate a standard IFN-γ curve. Standards and samples were set up in triplicate, and the mean of each triplicate was calculated. The spleen lymphocytes from immunization with rBCG-CtEm were cultured with 1 μg CtEm or medium; the culture supernatants were collected at day 3 and the cytokines IL-2, IL-4 and IL-10 were quantified using a commercially available ELISA kit (MABTech, Sweden).

Cytotoxicity assay

C1R.AAD-CtEm cells, C1R.AAD cells expressing chimeric HLA-A2-2Dd molecules [28] and transfected with pcDNA3.1-CtEm after screening by G418, were used as the target cells. In this study, the C1R.AAD cells pulsed with different peptide pools also were used as the target cells to confirm the efficacy of different single epitopes. The target cells were washed and labeled with 51Cr. 5 × 106/ml mouse spleen lymphocytes were added to CtEm or peptides to a final concentration of 10 μmol/L as effector cells. Using different target:effector ratios, the 51Cr-labeled target cells were incubated at 37°C in a 5% CO2 atmosphere for 5 h with effectors cells. Percent lysis was calculated as [(experimental release A − spontaneous release A)/(100% release A − spontaneous release A)] × 100. All experiments were performed more than three times.

Recombinant vaccinia virus challenge of mice

The HCV core gene and non-structural genes NS3–NS5 were inserted into a Kpn I restriction site of plasmid pVI75 to construct transfer plasmid pVI75-CNS. By co-transfecting CEFs with the vaccinia virus Tiantan strain, the recombinant vaccinia virus rVV-HCV-CNS was obtained through homologous recombination and screening by color difference in virus dots. HHD-2 mice immunized with rBCG-CtEm and control were challenged with 1 × 107 PFU of rVV-HCV-CNS or control wild-type vv752-1 by intraperitoneal injection 2 weeks after the last immunization. No immunized mice were infected with rVV-HCV-CNS or vv752-1 as controls. All mice were sacrificed 5 days after challenge, and their ovaries were harvested. Following freeze-thaw and homogenization, vaccinia virus titers were determined in BSC-1 cells by plaque assay.

Statistical analysis

Data analysis was carried out using a paired Student t test, and P < 0.05 was considered statistically significant.

Results

Expression of the multi-epitope antigen CtEm

The purified fusion protein and concentrated rBCG- CtEm culture supernatant were analyzed by Western blotting. The multi-epitope protein, CtEm (35.2 kDa), was detectable in both the purified fusion protein and supernatant concentrations of rBCG-CtEm but was not detected in the pQE30 empty plasmid control, which did not express the multi-epitope protein CtEm, or in the rBCG-CS1 control, which only expressed CS1 (Fig. 2).

Fig. 2
figure 2

Western blotting results for the purified fusion protein, CtEm, and the concentrated culture supernatant of rBCG-CtEm The first antibody was mouse anti-CtEm polyclonal antibody, and IRDye800-conjugated goat anti-mouse IgG was used as the second antibody for visualization with the Odyssey system. The size of the protein was 35.2 kDa. (lane 1 pQE30-CtEm; lane 2 pQE30 empty plasmid control; lane 3 rBCG-CtEm; lane 4 rBCG-CS1 control)

Full size image

Humoral immune response to rBCG

We detected HCV-specific antibodies in serum from mice immunized with the two different vaccines. The antibody level in the BCG group was similar to that in the negative control group. There were no significant difference of antibody level between the rBCG-CS1 group or the pcDNA3.1 control group and the negative control group (P > 0.05). The antibody level in mice immunized with rBCG-CtEm was higher than that of the mice immunized with DNA vaccine (P < 0.01), and CtEm-specific antibodies in the rBCG-CtEm group increased significantly after the initial inoculation and further increased after a booster injection 1 month later (Fig. 3a). The antibody disappeared there months after DNA immunization; however, the antibody could be detected six months after rBCG-CtEm inoculation (Fig. 3b). Coated with the peptide E2, the antibody could be detected in serum, and the results were nearly same as those described above, for CtEm coated with purified fusion protein (Fig. 3c). These results indicate that CtEm expression effectively induced CtEm-specific antibodies in mice, and a higher level of antibody in mice immunized with rBCG-CtEm was maintained for at least for six months, the HCV-specific antibodies were targeted directly at the mimotope from E2.

Fig. 3
figure 3

Detection of antibodies in rBCG- and DNA-immunized mice. Groups of ten female 4-to-6-week-old HDD-2 mice were immunized with 106 CFUml−1 BCG, rBCG, and a booster immunization 1 month later. The DNA immunization was carried out by the usual procedure. Every 2 weeks, blood was collected from the vena caudalis, serum was isolated, and the absorbance value (OD490) measured by enzyme immunoassay was converted into IU/ml. a CtEm-specific antibody detection 12 weeks after the first immunization. The antibody level in rBCG-immunized mice (hatched columns) was higher than that of the DNA-immunized mice (gray columns), which reached to the highest level at 6–8 weeks and then declined. b CtEm-specific antibody detection six months after the first immunization. A high level of antibody could be detected 6 months after the first immunization in rBCG-immunized mice (hatched columns), but not in mice immunized with DNA vaccine (dark stippled columns). c Detection of peptide-E2-specific antibodies 6 months after the first immunization. A high level of E2-specific antibody (hatched columns) was detected in rBCG-immunized mice. The time of rBCG booster immunizations at week 4 post-inoculation is indicated by down arrow

Full size image

Proliferative responses in splenic lymphocytes

Spleen lymphocytes were isolated and incubated with the purified CtEm or different peptide pools, and the proliferation index (PI) was calculated. The BCG and saline control groups had the same PI value. The PI of the rBCG-CS1 control group was slightly higher than that of the negative control, but was not significantly different (P > 0.05). The rBCG-CtEm group had a PI value that was notably higher than the rBCG-CS1 control (P < 0.01) (Fig. 4a). The different PI values of the spleen lymphocytes incubated with synthetic peptides were visible. When incubated with NS31073–1081, NS41807–1816 and NS52727–2735, the PI values increased notably against the control peptide HBS183–191 (P < 0.01) but were lower than the PI incubated with fusion protein CtEm (P < 0.05) (Fig. 4b). These results indicate that rBCG-CtEm effectively elicited CtEm-specific proliferative responses, and an epitope-specific proliferative response was induced. The immune response induced by CtEm was stronger than that induced by any single epitope, and this may be due to the cumulative effect of several epitopes.

Fig. 4
figure 4

Proliferation index determined by lymphocytes proliferation assay. a PI with purified protein CtEm. b PI with synthetic peptide. The PI was calculated as the mean incorporation of 3H-thymidine by splenocytes in triplicate test wells in the presence of CtEm or peptides divided by the mean background 3H-thymidine incorporated by splenocytes in triplicate control wells. *P < 0.05, **P < 0.01

Full size image

Cytokine detection

After incubation with CtEm for 3 days, the supernatants of the spleen lymphocytes were collected and the IFN-γ concentration was determined. The IFN-γ level of the rBCG-CtEm group increased markedly, reaching 1269 ± 16 pg/ml, and was significantly higher than that of the saline control group (192 ± 12 pg/ml, P < 0.01) and the control rBCG-CS1 group (533 ± 12 pg/ml, P < 0.01). The IFN-γ level of the BCG control group (521 ± 16 pg/ml) was similar to the level observed in the rBCG-CS1 control group (P > 0.05) but higher than that observed in the saline control group (P < 0.05) (Fig. 5a). The cytokines from culture supernatants of the rBCG-CtEm group were determined; the levels of IFN-γ and IL-2 increased markedly with CtEm compared to the levels of IL-4 and IL-10, indicating that the cellular immune response to rBCG-CtEm involves the recruitment of a type 1 subset of helper T cells (Fig. 5b).

Fig. 5
figure 5

Cytokine detection in splenocyte culture supernatants of mice immunized with rBCG-CtEm after incubation with CtEm or medium for 3 days. a IFN-γ concentrations measured in culture supernatants. The IFN-γ concentration in the rBCG-CtEm group was significantly higher than in the rBCG-CS1 control and vector control groups (P < 0.01). The mean level of the rBCG-CS1 control and BCG control groups was higher than that of the saline control group (P < 0.05). b Cytokine detection in splenocytes culture, IFN-γ and IL-2 increased markedly, but IL-4 and IL-10 did not increased markedly relative to medium. Results are expressed as the mean level from ten mice, ODI ± SE. *P < 0.05, **P < 0.01

Full size image

Cytotoxicity assay

Splenic lymphocytes were obtained from mice immunized with rBCG-CtEm, rBCG-CS1, or a BCG control, and their ability to recognize the MHC-matched target cells, C1R.AAD-CtEm, at different E:T ratios was assessed. A large percentage (65.3%) of effector cells elicited in rBCG-CtEm-immunized mice lysed C1R.AAD-CtEm cells (E:T:100:1); however, no such effectors cells were generated in mice immunized with the control rBCG-CS1 (P < 0.01) or BCG control groups (P < 0.01) (Fig. 6a). These results indicate that rBCG-CtEm elicited a stronger cellular immune response in transgenic mice. The specific lysis rates were higher than DNA immunization and immunization in BALB/c mice (data not shown). The E:T ratios of the BCG control were lower than that of the rBCG-CtEm group but higher than that of the saline control (P < 0.05), indicating that BCG is a powerful adjuvant with the ability to elicit specific host immunity. C1R.AAD cells pulsed with a different peptide pool were also used as the target cells to confirm the efficacy of different single epitopes. The specific lysis rates of six single CTL epitopes were different. Considerably lower lysis rates were observed for NS31073–1081 (38.5%, E:T:100:1), NS41807–1816 (42.1%, E:T:100:1) and NS52727–2735 (42.1%, E:T:100:1), which were higher than that of the control peptide (P < 0.01) but lower than that of C1R-AAD-CtEm (P < 0.05) (Fig. 6b). These results indicated that the multi-epitope elicited a more effective cellular immune response than single epitope, and the cumulative efficacy of the multi-epitope could be of advantage for a mutli-epitope rBCG vaccine.

Fig. 6
figure 6

Splenic lymphocytes from rBCG-CtEm-immunized mice developed CTL epitope-specific lysis activity. a CTL lysis activities were determined with the C1R.AAD-CtEm as the target cells. b CTL lysis activities were determined with the C1R.AAD pulsed with different synthetic peptides as the target cells. The data represent the mean percentages of the specific lysis values obtained from ten mice

Full size image

Protection from rVV-HCV-CNS

To examine the efficacy of HCV-specific CTLs generated by rBCG-CtEm immunization for clearance of the virus and protection against virus infection, mice were challenged with 1 × 107 PFU of rVV-HCV-CNS. Five days after the challenge, the mice were sacrificed, and the ovaries were harvested for measurement of vaccinia virus titers. The titers of rVV-HCV-CNS in mice immunized with rBCG-CtEm were much lower than those in mice immunized with rBCG-CS1 or BCG or in naive mice (Fig. 7a). However, the titres of wild-type rVV-752-1 did not decrease in mice immunized with rBCG-CtEm (Fig. 7b). These results indicate that rBCG-CtEm immunization confers protection against infection with rVV-HCV-CNS in vivo.

Fig. 7
figure 7

rBCG-CtEm immunization protects HHD-2 mice against a challenge with rVV-HCV-CNS. a Mice (five per group) were vaccinated by intraperitoneal injection with 1 × 107 rVV-HCV-CNS. Ovaries were harvested from all mice on day 5 (peak viral titers), followed by homogenization and freeze-thaw. Viral titers were evaluated by plaque assay. The mean viral titers (error bars represent SDs) for each group are given. b Viral challenge (1 × 107 PFU/ml) with wild type Tiantan strain vv752-1. The experiment was performed in parallel and as described above (*P < 0.05, **P < 0.01)

Full size image

Discussion

HCV infection leads to viral persistence in 50–85% of patients and can result in liver cirrhosis and liver cancer [29]. In contrast to hepatitis A and hepatitis B, no prophylactic or therapeutic vaccine is available. Many vaccine designs have been explored in recent years [30–33], but none of these has become licensed. Thus, it remains critical to design new strategies for the development of an HCV vaccine.

The goal of an HCV vaccine is the elicitation of a sustained anti-viral responses to cope with the escape from innate and adaptive immune responses [34]. Control of acute infection is associated with vigorous, broadly directed, and sustained activation of HCV-specific T cells [35]. An effective strategy to induce a cellular immune response is to target multiple HCV epitopes. A multi-epitope vaccine could involve many target antigens and adjuvant epitopes. It has the advantage of inducing a cell-mediated immune response to deal with the hypervariation of pathogens. The processing of the epitopes is critical for vaccine design strategy, so the epitopes need to be modified with flanking sequences or spacers so that they can be processed and presented separately. It has been confirmed that rational coupling of epitopes in series can efficiently stimulate a cellular immune response [36, 37]. In the present study, each epitope was extended by two or three amino acids at both ends, and an AAY spacer was added between the epitopes so that the epitope peptide could be efficiently cleaved by proteinase in antigen-presenting cells.

The multi-epitope vaccine approach has been shown to be successful [38–40]. The identification of conserved epitopes provides has provided a possibility to construct a HCV multi-epitope vaccine. Therefore, it is critical to select a suitable vector capable of sustained expression of HCV epitopes. As a new-generation vaccine for prevention and treatment, BCG has been used in many countries and has a very low incidence of serious complications. Its advantages include having a large DNA capacity, being heat-stable, and being relatively inexpensive. In addition, BCG is by itself a potent adjuvant for induction of cell-mediated immune responses. Recombinant BCGs expressing foreign proteins from different pathogens have been used successfully [18–22].

In this study, the shuttle vector pDE22 was constructed by inserting an Hsp60 promoter and signal peptide a-ss sequence. The protein of interest could be effectively secreted, guided by the signal peptide. The use of the Hsp60 promoter not only promotes expression of the exogenous gene, but also protects the expressed protein from hydrolysis by protease, and a more intense immune response can be induced when much of the expressed protein accumulated in the phagosome [41]. As a result, the multi-epitope gene was effectively expressed in rBCG-CtEm. HCV-specific antibodies were induced after immunization with rBCG-CtEm. The specific protective antibodies directly targeting neutralizing antigen epitopes from HCV E2 were sustained for at least six months; however, the antibodies induced by DNA immunization disappeared completely in the third month.

A higher level of IFN-γ was induced by CtEm protein and synthetic peptides NS31073–1081, NS41807–1816 and NS52727–2735. These results are in accordance with the prior studies [14, 15]. Cytokine levels indicated that the cell-mediated immune response to rBCG-CtEm involves the Th1-type immune response. Uno-Furuta et al. [20] found that immunization with recombinant BCG-expressed single CTL epitope from NS5 elicited major histocompatibility complex class I-restricted CD8+ HCV-NS5a-specific CTLs in BALB/c mice. In our study, the proliferation index and cytotoxicity assay results indicated that the compound epitopes elicited an HCV-specific MHC-matched cellular immune response in HHD-2 mice, and the efficacy of multi-epitope is higher than with a single epitope. The additive effect of the multi-epitope was confirmed, and these results are the foundation for the development of a multi-MHC-restricted epitope vaccine.

Finally, we examined the efficacy of HCV-specific CTLs generated by rBCG-CtEm immunization for clearing the virus and protection against recombinant vaccinia virus rVV-HCV-CNS, expressing HCV core and nonstructural protein. The results indicate that rBCG-CtEm immunization confers protection against infection with rVV-HCV-CNS in vivo. Thus, our study provides a novel vaccination strategy for construction of a prophylactic and therapeutic vaccine against HCV.