C 2005) Journal of Clinical Immunology, Vol. 25, No. 2, March 2005 ( DOI: 10.1007/s10875-005-2820-4
Immunogenic Properties of a Chimeric Plant Virus Expressing a Hepatitis C Virus (HCV)-Derived Epitope: New Prospects for an HCV Vaccine G. PIAZZOLLA,1,4 M. NUZZACI,2 C. TORTORELLA,1 E. PANELLA,1 A. NATILLA,2 D. BOSCIA,3 A. DE STRADIS,3 P. PIAZZOLLA,2 and S. ANTONACI1
ally transmitted hepatitis worldwide, with an estimated prevalence of about 3% in the world population (1). Despite a wide array of humoral and cellular host immune responses, HCV infection is characterized by a very low spontaneous recovery rate (2, 3). Epidemiological surveys indicate that in most instances HCV-related disease slowly progresses toward liver failure and cirrhosis, and also plays a leading role in the pathogenesis of hepatocellular carcinoma (4–6). At present, the development of a protective HCV vaccine remains a high priority goal since currently available therapies (pegylated α interferons either alone or in combination with ribavirin) are expensive and yield long-term responses in only about 50% of patients (7). Theoretically, an ideal vaccine should be capable of preventing initial infection by providing “sterilizing” immunity, but in the case of a persistent pathogen such as HCV this target may be very difficult to achieve. Hence, an effective HCV vaccine should at least protect against chronic and progressive disease, presumably by stimulating those immune responses that are described to be effective in patients spontaneously recovering from infection (8). With reference to this point, it is widely recognized that a critical role is played by the cellular arm of the immune system (9–12). The creation of vaccines that induce long-lived cellular immune responses is thus essential to the goal of designing immune-based strategies of prevention against HCV as well as other emerging persistent pathogens (13, 14). Another essential requirement for an HCV vaccine is that it must protect against many viral variants being HCV characterized by a high rate of mutation during replication. This feature has led to classification of the virus into distinct genotypes and subtypes whose distribution varies both geographically and among risk groups (1, 15). HCV mutations are mainly accumulated in few restricted regions, referred to as hypervariable regions (HVR). In
Accepted: November 8, 2004
A vaccine against Hepatitis C virus (HCV) is urgently needed due to the unsatisfactory clinical response to current therapies. We evaluated the immunological properties of a chimeric Cucumber mosaic virus (CMV), a plant virus engineered to express on its surface a synthetic peptide derived from many HVR1 sequences of the HCV envelope protein E2 (R9 mimotope). Evidence was obtained that the chimeric R9-CMV elicits a specific humoral response in rabbits. Furthermore, in patients with chronic HCV infection, purified preparations of R9-CMV downmodulated the lymphocyte surface density of CD3 and CD8, and induced a significant release of interferon (IFN)-γ , interleukin (IL)-12 p70 and IL-15 by lymphomonocyte cultures. Finally, an R9 mimotope-specific CD8 T-cell response, as assessed by intracellular IFN-γ production, was achieved in the majority of the patients studied. Our results open up new prospects for the development of effective vaccines against HCV infection. Moreover, the wide edible host range of CMV makes the production of an edible vaccine conceivable. KEY WORDS: HCV vaccine; T lymphocytes; cytokines; CMV; plant viruses.
INTRODUCTION
Hepatitis C virus (HCV) infection is a global health issue. HCV is, in fact, the major etiological agent of parenter-
1 Department
of Internal Medicine, Immunology and Infectious Diseases, Section of Internal Medicine, University of Bari, Policlinico, Bari, Italy. 2 Department of Biology, Plant Protection and Agrobiotechnology, University of Basilicata, 85100 Potenza, Italy. 3 Institute of Plant Virology, CNR, Section of Bari, 70126 Bari, Italy. 4 To whom correspondence should be addressed at Department of Internal Medicine, Immunology and Infectious Diseases, Section of Internal Medicine, University of Bari, Policlinico, 70124 Bari, Italy; e-mail:
[email protected].
142 C 2005 Springer Science+Business Media, Inc. 0271-9142/05/0300-0142/0
CHIMERIC
CMV AS A PROSPECT FOR HCV VACCINE
particular, the highest degree of variability in the genome is shown by the 27 amino acid-long N-terminal segment of the putative envelope protein E2, namely HVR-1, which has been reported to be a critical neutralization domain (16, 17). Plant viruses are emerging as novel systems for the expression of foreign epitopes to be used as immunogens for the development of innovative vaccination strategies. Recent studies have shown that chimeric virus particles (CVPs) and hybrid capsid proteins of plant viruses are able to activate antigen (Ag)-specific cell-mediated immune responses directed against small peptides that would not be immunogenic by themselves (18). Encouraging results have been obtained with the use of plant virus strategies in studies dealing with difficult targets such as the human immunodeficiency virus (HIV), which shares with HCV the power to cause chronic, invalidating disease (19–21). Our group recently demonstrated that a chimeric Cucumber mosaic virus (CMV), genetically engineered to express an HCV epitope on its surface, is a suitable carrier that can be adopted in the development of attractive vaccination strategies (22). The selected epitope was the so-called R9 mimotope, a synthetic peptide surrogate derived from a consensus profile of more than 200 HVR1 sequences of different viral isolates (23). On the basis of these assumptions and of the consideration that no studies have yet been performed analyzing human immune responses induced by plant viruses carrying HCV epitopes, in the current work we proposed to evaluate whether potentially useful anti-viral immune responses could be stimulated in patients with chronic HCV infection by this peculiar Ag-presenting system. Firstly, we verified the ability of R9-CMV to elicit a humoral response when parenterally administered to rabbits. Subsequently, we obtained evidence that in vitro R9-CMV stimulation of peripheral blood mononuclear cells (PBMC) from HCV positive patients resulted in the activation of cellular immune responses in a significant percentage of patients infected with different HCV genotypes. All in all, our results provide the rationale for further investigation of the employment of plant-virus strategies to produce a therapeutic vaccine against HCV. MATERIALS AND METHODS
Construction of Chimeric Virus and Plant Infection Two strains of CMV, namely CMV-D and CMV-S, were propagated in Nicotiana tabacum c.v. Xanthi and purified as described by Lot et al. (24). A pseudorecombinant CMV-D/S was made, derived from the RNA3 compo-
Journal of Clinical Immunology, Vol. 25, No. 2, 2005
143
nent of the CMV-S strain carrying the coat protein (CP) gene and the RNA1,2 component of CMV-D strain. The CMV-S CP gene (AF063610) used in this study was obtained from full-length cDNA copies of CMV-S genomic RNA3 (pCMV3S) of 2078 nt (the gift of Marie Tousignant, Agricultural Research Service, U.S.D.A., Beltsville, MD 20705, USA). The R9 mimotope nucleotide sequence was inserted in position 529 of the CP gene as described by Natilla et al. (22), and the resultant plasmid served as template for in vitro generation of the corresponding CMV-S chimeric RNA 3. One microgram of DNA template was then used in a 20 µL volume reaction for synthesizing capped transcripts using the T7 mMessage mMachineTM Kit (Ambion Europe Ltd., Cambridgeshire, UK). Before transcription, the template was linearized with SmaI. The in vitro CMVS RNA3 transcript was then supplemented with the other two CMV genomic RNAs (RNA1,2/RNA3, 1:2) deriving from CMV-D, to obtain the chimeric R9-CMV. A final RNA concentration of 0.5 µg/µL in 50 mM potassium phosphate, pH 7.0, was used to inoculate Nicotiana tabacum c.v. Xanthi plants at the four-leaf stage. To verify R9-CMV ability to spread systemically in the host as well as to demonstrate that the R9 mimotope was exhibited in the planned exposed position, tissues were analyzed by RT-PCR, Western blot, ELISA and electron microscopy 10 days after inoculation, as previously described (22). R9-CMV particles were purified and quantified by measuring the optical density of the virus suspension at 260 nm (24). Virus extraction yielded an average of 10 mg/100 g of fresh tissue. CMV-D/S purified particles were used as controls in all the experiments. Rabbit Immunization Female New Zealand rabbits were subcutaneously and intramuscularly immunized with 1 mL purified R9-CMV preparations (1 mg/mL) emulsified with an equal volume of Freund’s incomplete adjuvant, followed by three booster injections (1 mL, 0.5 mg/mL) at weekly intervals. Serum samples were collected before the primary immunization and weekly after the last boost. As controls, rabbits received an identical immunization regime with purified CMV-D/S preparations. All animal experiments were conducted according to the criteria published by the Italian Ministry of Health. Purified R9-CMV or bovine serum albumin (BSA)coupled R9 mimotope were used for the determination of the antiserum titre by double diffusion in agar plates (0.7% agar, 0.85% NaCl and 0.02% sodium azide).
144
Western Blot Analysis Purified virions (1 µg) and R9 mimotope (1 µg) were electrophoresed by 12.5% SDS-PAGE (45 min at 100 V) and then electroblotted onto nitrocellulose membrane for 1 h at 100 V. The membrane was blocked with powdered milk and incubated with R9-CMV polyclonal antiserum (1:500). It was finally treated with goat antirabbit alkaline phosphatase-conjugated antibodies (Abs) (Sigma Chemical Co., Milan, Italy) (diluted 1:2000). The reactivity was detected using the Sigma FastTM kit (Sigma).
PIAZZOLLA
et al.
In 30 patients histological features of minimal or mild portal/periportal necro-inflammation and absent or minimal fibrosis (grading 1–2/staging 1–2) (25) were found. Seven specimens showed moderate or severe periportal inflammation (grading 2–3) associated with bridging fibrosis or cirrhosis (staging 3–4). However, no differences in immunological responses were found according to the histological or clinical aspects of liver disease or to the HCV genotype among the patients enrolled. Thirty HCV negative healthy donors were also included in the study as controls.
Immune Electron Microscopy
Cell Cultures
Specimen grids were sensitized with 1:50 diluted antiR9-CMV polyclonal antiserum for 30 min at room temperature. R9-CMV or CMV-D/S virus particles from crude extracts were incubated on sensitized grids for 1 h at room temperature and then exposed for 15 min to the same antiserum diluted 1:25. Grids were washed with water, stained with 2% aqueous uranyl acetate and immediately processed for electron microscopy assays.
PBMC were isolated from heparinized venous blood by Lympholyte (Cedarlane Laboratories, Hornby, Ontario, Canada) density gradient centrifugation. The cell suspensions recovered at the interface were washed and resuspended in RPMI 1640 (Sigma) supplemented with penicillin (200 IU/mL), streptomycin (100 µg/mL), l-glutamine (2 mM) and 10% heat-inactivated fetal calf serum (PCS) (complete medium). The monocyte concentration in PBMC suspensions was approximately 15%, as evaluated by morphological (Giemsa staining) and cytochemical (non-specific esterase) criteria. PBMC (2 × 105 cells per well) were incubated for various periods of time (4, 7 and 14 days) at 37◦ C 5% CO2 in 96-well round-bottom microtiter plates (Costar, Cambridge, MA, USA) in the presence of medium alone or with increasing concentrations of R9-CMV or CMVD/S (1, 2.5, 5, 10 and 20 µg/mL). For intracellular cytokine production experiments, brefeldin A (10 µg/mL) was added to PBMC cultures during the last 4 h to block cytokine secretion. Where indicated, phorbol myristate acetate (PMA, 5 ng/mL) plus ionomycin (1 µM) were added to unstimulated cell cultures during the final 4 h. All reagents were LPS free, as assessed by the Limulus amebocyte lysate assay (FBI International, Milan, Italy).
Patients Forty-five untreated patients with chronic HCV infection (25 males and 20 females; mean age 54 years, range 19–68 years), admitted to the Department of Internal Medicine of Bari Medical School between January 2003 and December 2003, were enrolled in the study, after giving informed consent. Each patient had abnormal alanine aminotransferase (ALT) serum levels (range 1.4–3.8 times higher than the upper normal limit) for at least 6 months before inclusion in the study. Exclusion criteria included alcoholism, use of hepatotoxic drugs, clinical and/or histological evidence of liver autoimmune hepatitis, inherited metabolic disorders and co-infection with other hepatotropic viruses (i.e., HBV and HDV). HCV infection was assessed in all patients using the Ortho Third generation HCV ELISA Test System (Ortho Diagnostic System, Raritan, NJ, USA), followed by detection of serum HCVRNA by nested reverse transcription-polymerase chain reaction (RT-PCR), using primers of the 5 noncoding region of HCV. The HCV genotype was determined by Inno-Lipa HCV II (Innogenetics N.V., Ghent, Belgium), that allows the genotyping of the six major HCV types and their most common subtypes. With this approach we found genotype 1b in 22 patients, 2a/2c in 21 patients, while 2 patients had genotypes 4c/4d and 5a, respectively. Percutaneous needle liver biopsy specimens were obtained from 37 patients and used for histological examination; in 8 patients liver biopsy was not performed due either to low patient compliance or clinical contraindications.
Flow Cytometric Analysis of T Lymphocyte Surface Antigens After harvesting, PBMC were triple stained with fluorescein isothiocyanate (FITC)-labeled anti-CD3, phycoerythrin (PE)-labeled anti-CD4 and PE-Texas Red (ECD)-conjugated anti-CD8 monoclonal Abs (mAbs) (all purchased from Beckman Coulter Inc., Miami, FL, USA) on ice for 30 min, washed and then analyzed on a flow cytometer/cell sorter COULTER Epics Elite (Beckman Coulter Inc.). Lymphocytes were defined by their forwardand side-scatter properties. Data on CD3, CD4 and CD8 expression were collected in terms of both percentage and mean fluorescence intensity (MFI) of the positive cells.
Journal of Clinical Immunology, Vol. 25, No. 2, 2005
CHIMERIC
CMV AS A PROSPECT FOR HCV VACCINE
To overcome inter-assay variations of the fluorescence, a flow cytometer daily setting was established using the Flow-Set Fluorospheres (Beckman Coulter Inc.). These Fluorospheres have a uniform size and fluorescence intensity and contain a dye which has a fluorescence emission range of 525 ran to 700 nm when excited at 488 nm, thus allowing fluorescence intensity standardization. Cytokine Assay To assess cytokine production, culture supernatants were harvested and kept frozen at −80◦ C until assayed. Interferon (IFN)-γ , interleukin (IL)-12 p70 and IL-15 amounts were determined by ELISA kits (for IFN-γ and IL-12 p70: Euroclone, Paignton, Devon, UK; for IL-15: BioSource International, Camarillo, CA, USA) following the manufacturer’s protocol. Assay sensitivities were 1.2, 2 and 11 pg/mL, respectively. Flow Cytometric Analysis of CD8+ T-Cell Cytokine Production Cells recovered at different intervals of stimulation [5 × 105 cells per test in phosphate buffered saline (PBS)] were stained with FITC-labeled anti-CD3 and ECDconjugated anti-CD8 mAbs on ice for 30 min. Cells were then washed with 2 mL PBS 1% BSA and fixed with 0.5 mL fix/per solution (containing 4% paraformaldehyde and 0.2% saponin in PBS) for 15 min on ice. Cells were washed once with 1 × Per solution (0.1% saponin in PBS 1% BSA), resuspended in 50 µL 2 × Per buffer solution (0.2% saponin in PBS 1% BSA), and stained with PEconjugated anti-IFN-γ mAbs (Beckman Coulter Inc.) on ice for 30 min. Then cells were washed twice, suspended in 0.5 mL 1% paraformaldehyde PBS buffer, and analyzed on flow cytometry. Histograms and dot plots were generated after forward- and side-scatter gating on lymphocytes. After further gating on CD3+ T cells, the percentage of IFN-γ producing T lymphocytes within the CD8+ population was extrapolated. In each subject,unstimulated cells were used to set negative staining. Statistical Analysis Results were analyzed using the paired Student’s t-test or linear regression. In both cases, p values of 0.05 or less were regarded as significant. RESULTS
Immunogenicity of R9-CMV in Rabbits To determine whether CVPs of R9-CMV were immunogenic in vivo, purified R9-CMV or CMV-D/S prepaJournal of Clinical Immunology, Vol. 25, No. 2, 2005
145
rations in Freund’s incomplete adjuvant were parenterally administered to rabbits. After four injections, a humoral response to R9-CMV was revealed in immunized rabbits by Immunoelectron microscopy (I.E.M.). In Figure 1, I.E.M. decoration of whole CVPs following exposition to rabbit anti-R9-CMV serum is shown. The presence of a specific anti-R9 response was assessed by double diffusion gel, showing an anti-R9 titre of 1:32. To further confirm the presence of anti-R9 antibodies, sera of immunized animals were subjected to immunoblot analysis. As shown in Figure 2, anti-R9-CMV and anti-CMV-D/S sera reacted with the chimeric proteins (lane 1) and the CMV-D/S proteins (lane 3), respectively, but only the anti-R9-CMV serum reacted also against the R9 mimotope (see lanes 2 and 4). Study of Chimeric R9-CMV Effects on T Cell Surface Molecules in Patients Chronically Infected with HCV CD3, CD8 and CD4 are key membrane structures involved in T-cell activation. Although they are known to be modulated in response to viral antigens, very little data are available on HCV-dependent regulation of these molecules. To investigate whether the R9 mimotope expressed on the chimeric CMV was able to influence the expression of T cell surface molecules, PBMC from 25 HCV patients were incubated with R9-CMV or CMVD/S, and then analyzed by multiparameter flow cytometric techniques after staining for CD3, CD8 and CD4 antigens. The MFI of each antigen on stimulated T cells was extrapolated in single subjects and compared with values obtained in unstimulated (NS) cell cultures. All values were then expressed in terms of MFI ratio of stimulated to NS cell cultures. Preliminary experiments of the dose– effect relationship and time course showed that 10 µg/mL of virus particles, and a 7-day interval of stimulation, induced maximal effects on the expression of T-cell receptors. As positive controls, 1-week unstimulated cell cultures supplemented with PMA plus ionomycin during the last 4 h were considered. The most striking effects of R9-CMV stimulation were observed on surface CD3 expression. As illustrated in Figure 3a, the CD3 MFI of cells cultured with R9-CMV was markedly lower than that observed in unstimulated cells, this finding being evident on the whole T cell population as well as on CD8+ and CD4+ T cell subsets. Although CMV-D/S cell challenge also resulted in T cell CD3 MFI down-modulation with respect to NS cells, this effect was significantly lower than that resulting from R9CMV stimulation (Figure 3a). In this regard, it should be further emphasized that the entity of the R9-CMVdependent CD3 MFI decrease was comparable with that induced by PMA/ionomycin cell treatment.
146
PIAZZOLLA
et al.
Fig. 1. (a) Particles of R9-CMV; (b) R9-CMV exposed to homologous polyclonal rabbit antiserum.
With reference to CD8 expression, we found that R9CMV-stimulated PBMC cultures underwent a CD8 MFI decrease that was significantly greater than that obtained with CMV-D/S, but lower than that induced by PMA + iononomycin (Figure 3b). Conversely, a similar reduction in CD4 MFI was observed following cell incubation with both plant viruses as compared with unstimulated cells (R9-CMV/NS CD4 MFI ratio = 0.71 ± 0.19; CMVDS/NS CD4 MFI ratio = 0.72 ± 0.18, both expressed as mean ± SD). Finally, no changes in the relative frequency of CD3+ , CD4+ or CD8+ T lymphocytes occurred in patient PBMC
cultures following either R9-CMV or CMV-D/S cell challenge (data not shown). R9-CMV-Induced Cytokine Release in PBMC Cultures of HCV Patients IFN-γ , IL-15 and IL-12 p70 are, among cytokines, those mainly involved in cell-mediated immune responses. Hence, the capacity of R9-CMV to trigger the production of these cytokines in PBMC cultures of the same 25 HCV positive patients was investigated. Twenty HCV negative healthy donors were used as controls.
Fig. 2. Western blot analysis of R9-CMV proteins (lane 1), CMV-D/S proteins (lane 3) and R9 mimotope (lanes 2 and 4) with anti-R9-CMV or anti-CMV-D/S serum. The 29 kDa protein in lane 1 is due to the 3 kDa increase of the 26 kDa CMV protein induced by the mimotope. M: Prestained Protein Marker, Broad Range (New England BioLabs). Journal of Clinical Immunology, Vol. 25, No. 2, 2005
CHIMERIC
CMV AS A PROSPECT FOR HCV VACCINE
147
Fig. 3. CD3 (a) and CD8 (b) T cell surface density following CMV-D/S stimulation (white columns), R9-CMV stimulation (black columns), or PMA + ionomycin stimulation (grey columns) of PBMC from HCV patients (n = 25). Results are expressed as ratio of stimulated to unstimulated T cell CD3 or CD8 MFI (mean ± SD).
Spontaneous and CMV-D/S-induced cytokine releases were also assessed in all subjects. A concentration of 10 µg/mL R9-CMV or CMV-D/S and a 7-day incubation time were used on the basis of pivotal experiments that demonstrated that these stimulation conditions gave rise to substantial cytokine production. The results, depicted in Figure 4, show that a significantly higher IFN-γ , IL-12 p70 and IL-15 release occurred in R9-CMV-stimulated PBMC with respect to unstimulated cells in 18/25 (72%), 18/25 (72%) and 13/25 (52%) HCV+ patients, respectively. Cell stimulation was partially ascribable to the virus itself, as assessed by CMV-D/S cell challenge. However, when compared with cytokine release induced by CMVD/S, an R9 mimotope-specific IFN-γ , IL-12 p70 and IL15 production was still found in 8 (32%), 10 (40%) and 9 (36%) of overall patients, respectively (see Figure 4). Interestingly, a positive correlation between the production of IFN-γ and IL-12 p70 (R = 0.579; p = 0.007) or IL-12 p70 and IL-15 (R = 0.471; p = 0.036) was seen in patient PBMC cultures following R9-CMV stimulation. Conversely, no significant differences in cytokine production were found between R9-CMV- and CMVD/S-stimulated PBMC cultures from healthy donors (R9-CMV-induced, CMV-D/S-induced and spontaneous cytokine productions, expressed as mean ± SD, were the following: IFN-γ = 1654 ± 1220, 1488 ± 1200, and 320 ± 190 pg/mL; IL-12 p70 = 4.1 ± 2.9, 4.2 ± 3.4 and
