Hepatitis B Surface Antigen Vector Delivers Protec
http://www.100md.com
病菌学杂志 2006年第8期
Sir Albert Sakzewski Virus Research Centre, Royal Children's Hospital and Clinical Medical Virology Centre, University of Queensland, Brisbane
Department of Microbiology, Monash University, Melbourne, Australia
ABSTRACT
Although hepatitis B surface antigen (HBsAg) per se is highly immunogenic, its use as a vector for the delivery of foreign cytotoxic T-lymphocyte (CTL) epitopes has met with little success because of constraints on HBsAg stability and secretion imposed by the insertion of foreign sequence into critical hydrophobic/amphipathic regions. Using a strategy entailing deletion of DNA encoding HBsAg-specific CTL epitopes and replacement with DNA encoding foreign CTL epitopes, we have derived chimeric HBsAg DNA immunogens which elicited effector and memory CTL responses in vitro, and pathogen- and tumor-protective responses in vivo, when the chimeric HBsAg DNAs were used to immunize mice. We further show that HBsAg DNA recombinant for both respiratory syncytial virus and human papillomavirus CTL epitopes elicited simultaneous responses to both pathogens. These data demonstrate the efficacy of HBsAg DNA as a vector for the delivery of disease-relevant protective CTL responses. They also suggest the applicability of the approach of deriving chimeric HBsAg DNA immunogens simultaneously encoding protective CTL epitopes for multiple diseases. The DNAs we tested formed chimeric HBsAg virus-like particles (VLPs). Thus, our results have implications for the development of vaccination strategies using either chimeric HBsAg DNA or VLP vaccines. HBsAg is the globally administered vaccine for hepatitis B virus infection, inviting its usage as a vector for the delivery of immunogens from other diseases.
INTRODUCTION
Vaccination strategies capable of safely and effectively inducing cytotoxic T-lymphocyte (CTL) responses which are acceptable in humans have not proven easy to develop (18). The hepatitis B virus (HBV) surface antigen (HBsAg) small envelope protein (HBsAg-S) self-assembles into highly organized virus-like particles (VLPs) (11). HBsAg VLPs are extremely efficient at inducing long-lived HBsAg-specific CTL responses when administered as a vaccine, comparing favorably with traditional strong CTL inducers in this regard (40). This likely relates to the presentation of the CTL epitopes as multiple copies in repetitive arrays, a format known to enhance induction of CTL responses (21). HBsAg may also be delivered as a DNA vaccine. By use of this strategy, CTL induction by intracellular translated HBsAg protein occurs through the classical endogenous pathway as well as through the alternative exogenous pathway via secreted HBsAg particles (28, 38, 39).
HBsAg-S polypeptide is composed of 226 residues, is synthesized at the endoplasmic reticulum anchored at the lipid membrane and is secreted as empty 22-nm VLPs, with each VLP consisting of 100 to 150 polypeptides. The HBsAg molecule has a complex configuration and contains at least two transmembrane regions, an internal luminal domain and a hydrophilic external domain (the a determinant). The C-terminal domain is hydrophobic and it probably contains two additional transmembrane regions. HBsAg CTL epitopes are located primarily in the transmembrane and luminal domains (2, 42).
Although HBsAg per se is highly immunogenic, the use of chimeric HBsAg to deliver foreign epitopes has met with limited success (particularly in the delivery of CTL epitopes), probably because the strategy hitherto used, that of introducing foreign sequences into the HBV genome at sites where restriction endonuclease sites fortuitously exist or of extending the termini of HBsAg protein with epitope sequences, does not necessarily account for spatial requirements for VLP formation (2, 11). In the present study, we report a novel strategy to exploit the superior immunogenicity of HBsAg by deleting HBsAg-specific CTL epitopes and replacing them with foreign CTL epitopes of similar physical properties (i.e., size and hydrophobicity). We reasoned this approach might predispose to the maintenance of spatial requirements, the conformation of the HBsAg molecule, and, in addition, the maintenance of appropriate antigen processing.
We demonstrate that immunization with DNA encoding chimeric HBsAg in which an HBsAg-specific CTL epitope(s) was deleted and replaced with CTL epitopes of respiratory syncytial virus (RSV) or human papillomavirus (HPV) elicited strong CTL responses to the inserted epitopes. Immunization with chimeric HBsAg DNA afforded protection against RSV infection or the growth of HPV type 16 (HPV16)-associated tumors. We further show that immunization with DNA encoding HBsAg recombinant for both RSV and HPV CTL epitopes elicited CTL responses to both foreign epitopes and afforded protection against disease.
We demonstrate that chimeric HBsAg DNAs constructed using the deletion-replacement strategy form VLPs which are secreted from cells. HBsAg VLPs are globally licensed as a vaccine for HBV infections in humans (including children) with a ca. 12-year history of use in millions of recipients, inviting their usage as a vector for the delivery of immunogens from other diseases. Our results demonstrate that the powerful immunogenicity of HBsAg may be exploited to deliver foreign CTL epitopes relevant for other diseases and have generic implications for the development of HBsAg as a multivalent vaccine vector. There are also specific implications for vaccines against RSV infection and HPV-associated carcinoma.
MATERIALS AND METHODS
Cloning procedures. The plasmid pcD3-HBsAgS, which contains a AgeI restriction site at a location corresponding to amino acids (aa) 127 and 128 of the a determinant region (33), was engineered to delete the HBsAg-specific CTL epitopes IPQ and GLS (Table 1) and to introduce the restriction enzyme sites NheI and BlpI, respectively, by PCR-driven site-directed mutagenesis. Plasmid pM215IPQ (Table 2) was derived by inserting DNA encoding the RSV-specific M215 epitope at the NheI site. Further chimeric HBsAg plasmids were derived as summarized in Table 2. Plasmid pM215IPQ.RAHGLS.HWI was derived by inserting DNA sequences encoding the RSV B-cell-specific mimotope (4) into the AgeI restriction site and then sequentially inserting DNA encoding the M215 epitope and the RAH epitope at the IPQ and GLS sites, respectively. Standard molecular cloning and plasmid purification procedures were used. All plasmids were verified by DNA sequencing (PCR Sprint temperature cyclin system [Hybaid]).
Transfection. HuH-7 cells (31) were transfected with bacterial plasmid DNA by use of calcium phosphate. The supernatant was harvested 5 days later, and HBsAg was measured by the Abbott Prism HBsAg assay (Abbott Diagnostics). The level of HBsAg in the cell culture fluid was quantified by comparison with a commercially available vaccine (Engerix-B, 20 μg/ml; Smith-Kline Beecham). The transfection efficiencies of different plasmids were normalized to the activity of secreted alkaline phosphatase as previously described (1). The variation in the range of secreted alkaline phosphatase was less than twofold. The presence of hepatitis delta virus (HDV) large antigen (HDAg-L) was identified by immunoblotting using a human anti-HDV antibody and assayed by the ECL-plus detection system (Amersham) (33).
Mice. A2.1Kb mice (44) make CTL responses restricted through HLA A0201 and H-2b class 1 molecules. (A2.1Kb x BALB/c)F1 hybrid mice additionally make CTL responses restricted through H-2d. Mice were housed under specific-pathogen-free conditions and used at 7 to 15 weeks of age.
Peptides. Peptides were synthesized using 9-fluorenylmethoxy carbonyl chemistry and analyzed by high-performance liquid chromatography and by amino acid analysis by Chiron Mimotopes (Melbourne, Australia). The RSV F protein mimotope was synthesized as a tetrameric multiple antigen peptide construct as described previously (4).
Immunization and restimulation of splenocytes. Mice were immunized twice at 2-week intervals intradermally (i.d.) in the ear with 100 μg of purified plasmid DNA. Two weeks later, spleens were removed and splenocytes were restimulated in vitro for 6 days as described previously (12) with 1 μg/ml cognate peptide. For peptide immunizations, mice were immunized subcutaneously (s.c.) at the tail base with 50 μg peptide plus 0.25 μg tetanus toxoid as a source of T-helper epitopes plus 10 μg Quil A adjuvant (24). Ten days later, spleens were harvested and splenocytes were restimulated as described above.
Cells. EL4.A2 cells (12) are susceptible to specific CTL lysis through both H-2b and HLA A0201 restriction pathways. P815 is susceptible to specific CTL lysis through the H-2d restriction pathway. Cells were maintained as described previously (12).
CTL assays. CTL assays were conducted as previously described (12). In summary, target cells (104 per well), sensitized at 37°C for 1 h with 1 μg/ml cognate or irrelevant peptide or with medium alone and labeled with 100 μCi 51chromium (Cr), were incubated with effector cells at various effector-to-target cell ratios in triplicate in 96-well microtiter plates. Negative controls included wells containing target cells but no effector cells (background). Supernatants were harvested from CTL assays at 4 h, and 51Cr release was quantified by gamma counting. Results are expressed as percent cytotoxicity ± standard deviation [(51Cr release in experimental wells – background/detergent-mediated total release – background) x 100%]. Experimental and control values were compared for significant difference using Student's t test.
Murine IFN- ELISPOT assay. Epitope-specific gamma interferon (IFN-)-secreting spleen cells were enumerated ex vivo by an enzyme-linked immunospot (ELISPOT) assay with minimal CD8+ T-cell epitope peptides essentially as described previously (24). IFN- spots were counted using an AID EliSpot reader system. Results were calculated as IFN--positive cells/106 spleen cells. Experimental and control values were compared for significant difference using analysis of variance.
RSV propagation. HEp-2 cells (American Type Culture Collection, Manassas, VA) were maintained in RPMI 1640 medium (Invitrogen, Mt. Waverley, VIC, Australia) supplemented with 292 μg/ml GlutaMax-1 (Gibco BRL), 1 mM sodium pyruvate, 20 mM HEPES, 50 μM -mercaptoethanol, 100 IU/ml penicillin, 100 μg/ml streptomycin (Gibco BRL), and 10% fetal bovine serum (CSL Ltd, Parkville, Australia). RSV A2 (VR-1302) strain (American Type Culture Collection) was adsorbed onto the HEp-2 cell monolayer at a multiplicity of infection of 0.04 PFU/cell for 2 h at 37°C prior to culturing in supplemented RPMI 1640 medium for 3 to 6 days. RSV was recovered from the supernatant following freeze-thawing of monolayers displaying a cytopathic effect. RSV was quantified by immunofocus assay (16), using goat anti-RSV primary antibody (Chemicon, Australia), anti-goat immunoglobulin (Ig) horseradish peroxidase-conjugated second antibody, and DAB (tetra-aminobiphenyl hydrochloride) substrate (Sigma, Australia) according to the manufacturer's instructions.
Evaluation of RSV infection. Mice were inoculated intranasally with 8 x 105 PFU of RSV and processed for histochemical staining or immunofluorescence using goat anti-RSV antibody and fluorescein isothiocyanate anti-goat Ig-detecting antibody (Chemicon). Lung samples were also used for RSV quantitation by plaque assay (16). Plaques were confirmed by immunostaining with goat anti-RSV antibody (Chemicon) followed by treatment with anti-goat Ig horseradish peroxidase-conjugated antibody (Chemicon, Australia) and development with DAB substrate (Sigma, Australia) and urea-H2O2-NaCl according to the manufacturer's instructions (Sigma).
Tumor protection assays. Groups of H-2b mice (five per group) were immunized with 10 μg plasmid DNA i.d. or with 100 μg of E7 peptide plus tetanus toxoid plus Quil A s.c. TC-1 cells, which express the E7 tumor-associated antigen of human papillomavirus type 16 (27), were subsequently injected (2 x 105 in 0.1 ml Hanks' buffered salt solution) s.c. on the flank. (The tumor dose was predetermined by titration experiments to discern a minimal dose giving rise to tumor in 100% of unimmunized mice.) Tumor growth was monitored every 2 days, and mice were euthanized when tumor volumes exceeded 1,000 mm3. Unimmunized mice received the same number of cells and served as a control. Data are represented as Kaplan-Meier curves of the percent tumor-free mice at given time points after tumor injection.
RESULTS
Immunization with chimeric HBsAg DNA encoding the RSV M2 CTL epitope elicits M2-directed CTL. A number of HBsAg CTL epitopes in hydrophobic regions of the HBsAg molecule, restricted through murine (39, 42) and human (28) major histocompatibility complex class I (MHC-I) molecules, have been described previously. In a first set of experiments, we elected to delete two of the stronger of these HBsAg-specific CTL epitopes (IPQ and GLS [Table 1]) and replace them with a protective epitope from the M2 protein of RSV. The minimal M2 CTL epitope of RSV is a 9-mer represented by M2 residues 82 to 90 (designated M29); a larger 15-mer peptide sequence represented by residues 81 to 95 (designated M215) also functions as a CTL epitope (Table 1 and reference 20). We asked whether chimeric HBsAg DNA vaccines encoding the RSV CTL epitope would elicit M2-directed effector and memory CTL responses.
Effector responses. We quantified epitope-specific IFN- secretion of ex vivo splenocytes by ELISPOT assay. Ex vivo splenocytes from mice immunized with pM29IPQ, pM215IPQ, or pM215GLS DNA (Table 2) secreted significant amounts of IFN- when cultured in vitro with, but not without, cognate M2 peptide (Fig. 1a to c), indicating the induction of an M2-directed effector CTL response. Ex vivo splenocytes from mice immunized with wild-type HBsAg DNA (pHBsAgWT) and cultured with M2 peptide did not secrete IFN- above the level observed when culturing was done without peptide (data not shown).
Memory response. We also asked whether immunization with pM29IPQ, pM215IPQ, or pM215GLS DNA would induce memory CTL capable of being restimulated in vitro to kill target cells displaying the M2 CTL epitope. Splenocytes from pM29IPQ, pM215IPQ, or pM215GLS DNA-immunized mice were restimulated in vitro for 6 days with M215 peptide and reacted with H-2d target cells pulsed with M215 epitope peptide. Restimulated splenocytes from all these immunized groups of mice killed M215 peptide-pulsed targets (Fig. 1f to h). Notably, the efficiency of killing was comparable to that seen in identically restimulated splenocytes from M215 peptide-plus-adjuvant-immunized mice (Fig. 1i). (Immunization with high-dose CTL epitope peptide plus adjuvant is a powerful inducer of CTL responses [12] and is a standard against which other CTL-inducing modalities may be compared.) It is noteworthy that the percent killing of M2-pulsed target cells by pM29IPQ or pM215IPQ DNA-immunized mice exceeded the percent killing of target cells pulsed with the HBsAg-specific peptide IPQ by restimulated splenocytes from pHBsAgWT immunized mice (Fig. 1j). Taken together, these data indicate that replacement of DNA encoding HBsAg-specific CTL epitopes (IPQ and GLS) with DNA encoding a CTL epitope from the RSV M2 protein to generate chimeric HBsAg DNA immunogens engenders M2-directed effector and memory CTL responses of intensities comparable to those engendered by immunization with peptide plus adjuvant and to those of the HBsAg-specific epitopes engendered by immunization with pHBsAg wild-type DNA.
Chimeric HBsAg immunogen containing two copies of foreign epitope. We also asked whether two copies of the M2 epitope, inserted at the IPQ and GLS sites in HBsAg, respectively, would elicit an M2-directed CTL response that was enhanced in comparison with that elicited by chimeric HBsAg containing a single copy at the IPQ site or at the GLS site alone. Mice were immunized with a construct containing two pM215 epitopes at the different sites (pM215IPQ.GLS.HWI [Table 2]) or with constructs containing one M215 epitope but at different sites (pM215IPQ.HWI and pM215GLS.HWI). (Note that the actual insertion at the two sites differed because of the introduction of different endonuclease restriction half-sites flanking the M215 epitope [Table 2 and "Cloning procedures" in Materials and Methods].) No significant augmentation of the number of IFN--secreting ex vivo splenocytes (Fig. 1k) or of the cytotoxic efficacy of restimulated splenocytes (Fig. 1, panel l) compared to that seen for mice immunized with chimeric HBsAg DNA encoding one copy of the epitope was recorded for mice immunized with chimeric HBsAg DNA encoding two copies.
Immunization with plasmid encoding chimeric HBsAg containing a CTL epitope of RSV M2 protein partially protects against RSV infection. Groups of mice immunized with DNA encoding M2-containing HBsAg or wild-type HBsAg were inoculated with RSV, and 4 days later, lungs were removed for RSV quantitation, histology, and RSV-directed immunofluorescence. Four of five mice immunized with wild-type HBsAg DNA displayed detectable RSV, whereas only one of five mice immunized with pM215GLS or pM215IPQ DNA displayed detectable RSV in lungs (Table 3). Among those mice where RSV was detected, the viral loads per lung were significantly less in mice immunized with pM215GLS or pM215IPQ DNA than in mice immunized with DNA encoding wild-type HBsAg (Table 3).
The lungs of wild-type HBsAg-immunized mice challenged with RSV showed a pathology consistent with severe interstitial pneumonia (bronchiolar epithelial cell damage, extensive infiltrations of inflammatory cells, and hemorrhaging into the inflamed bronchiole) (Fig. 2a, panel i). However, lungs from mice immunized with plasmid pM215GLS (Fig. 2a, panel ii) or pM215IPQ (not shown) displayed less bronchiolar epithelial cell damage, little mononuclear cell infiltration of the alveoli, and less hemorrhaging into bronchioles. Peri- and extrabronchiolar RSV infection was detected in immunofluorescence-stained lung cryosections of pHBsAgWT-immunized mice (Fig. 2b, panel i) but not in those of mice immunized with pM215GLS DNA (Fig. 2b, panel ii).
Together, the data in Table 3 and Fig. 2 indicate that immunization with chimeric HBsAg DNA expressing a CTL epitope of the RSV M protein confers protection against RSV infection and associated pathology.
Immunization with plasmid encoding chimeric HBsAg containing an HPV16 E7 CTL epitope elicits E7-specific CTL responses and protects against E7-expressing tumor. We extended the concept of HBsAg-vectored delivery of protective CTL epitopes to the RAH CTL epitope of the E7 tumor-associated antigen of human papillomavirus type 16 (Table 1). We inserted DNA encoding the RAH epitope into the IPQ site of the HBsAg gene to create the recombinant construct pRAHIPQ (Table 2). Ex vivo splenocytes from mice immunized with pRAHIPQ DNA, but not from those immunized with pHBsAg wild-type DNA, and cultured with RAH peptide contained a significantly higher number of IFN--secreting cells than splenocytes cultured without RAH peptide (Fig. 3a and b). Additionally, restimulated splenocytes from pRAHIPQ DNA-immunized mice, but not from pHBsAg wild-type DNA-immunized mice, specifically lysed target cells pulsed with RAH peptide (Fig. 3d and e). These data indicate specific in vivo priming of RAH-directed effector and memory T cells by immunization with pRAHIPQ DNA. To determine whether immunization would protect against E7-expressing tumor, groups of mice were immunized with plasmids pRAHIPQ and pHBsAg and challenged with TC-1 tumor cells. Eighty percent of mice immunized with pHBsAg developed tumor within 15 days. In contrast, none of the mice immunized with pRAHIPQ DNA developed tumor (Fig. 3g). The pRAHIPQ-induced in vitro and in vivo responses compared favorably with those induced by immunization with RAH peptide plus adjuvant (Fig. 3c, f, and h).
Together, these data indicate that immunization with DNA encoding rHBsAg containing a CTL epitope (RAH) of HPV16 E7 oncoprotein protects mice against challenge with an E7-expressing tumor, and this protection is associated with RAH-directed effector and memory CTL responses.
Immunization with a plasmid encoding chimeric HBsAg containing two protective CTL epitopes from distinct pathogens elicits CTL responses to both epitopes. We extended the concept of HBsAg-vectored protective CTL epitopes to the concomitant delivery of RSV and HPV epitopes delivered by the same chimeric HBsAg DNA. Mice were immunized with pM215IPQ.RAHGLS.HW1 DNA (Table 2), and CTL responses were examined by IFN- ELISPOT assay on ex vivo splenocytes and by 51Cr release cytotoxicity assay mediated by specifically restimulated splenocytes. Splenocytes from pM215IPQ.RAHGLS.HW1-immunized mice secreted IFN- when cultured with M215 peptide (Fig. 4a). The M215-directed response was similar (P > 0.05) to that in splenocytes from mice immunized with a single recombinant encoding M215 alone (pM215IPQ) (Fig. 4a). The RAH-directed IFN- response was lost (P < 0.001) in mice immunized with the double recombinant compared with that in mice immunized with the single recombinant encoding the RAH epitope alone (pRAHGLS) (Fig. 4b).
Restimulated splenocytes from mice immunized with pM215IPQ.RAHGLS.HW1 killed target cells pulsed with either M215 peptide or RAH peptide (Fig. 4c and d). While the restimulated splenocytes killed RAH-expressing target cells significantly more than target cells not expressing RAH (50:1 effector:target cell ratio; P < 0.001) (Fig. 4d), the killing was less than that observed with restimulated splenocytes from mice immunized with pRAHGLS encoding only one inserted RAH CTL epitope (P < 0.001). (Note that RAH-restimulated splenocytes from pHBsAgWT-immunized mice fail to kill RAH-expressing target cells [data not shown].)
In view of the down-regulated CTL response to the RAH epitope in mice immunized with pM215IPQ.RAHGLS.HW1, we asked whether these mice were nonetheless protected against RAH-expressing tumor. Mice immunized with the double recombinant pM215IPQ.RAHGLS.HW1 or with the single recombinant pRAHGLS were fully protected against E7-expressing tumor. In contrast, 40 to 60% of unvaccinated mice or mice immunized with plasmids encoding wild-type HBsAg or chimeric HBsAg not containing the RAH epitope developed tumor within 8 days (Fig. 4c).
Together, these data indicate that immunization with rHBsAg DNA encoding two foreign CTL epitopes from different pathogens elicits effector and memory CTL responses to both epitopes. Although the RAH-directed CTL response measured in vitro was down-regulated in mice immunized with the chimeric HBsAg DNA encoding two epitopes, these mice were nonetheless protected against tumor challenge as effectively as mice immunized with the chimeric HBsAg DNA encoding the RAH epitope alone.
Epitope replacement allows for efficient formation and secretion of VLPs. DNA vaccines induce CTL responses through the classical endogenous pathway via intracellular translation and antigen processing of the proteins they encode. In addition, HBsAg DNA vaccines may induce CTL through the alternative exogenous pathway via secreted HBsAg VLPs (9, 28, 38, 39). It was thus relevant to ask if our strategy for deriving chimeric HBsAg DNA vaccines encoding foreign CTL epitopes also allowed for the formation of secretable HBsAg VLPs. Previous studies to derive HBsAg VLP immunogens containing foreign CTL epitopes have explored the insertion of foreign domains into the pre-S2 region (15). Indiscriminate insertions into HBsAg may lead to unpredictable stability and secretion, probably due to conformational changes in the recombinant HBsAg molecule (11). Thus, we reasoned that deletion of HBsAg-specific CTL epitopes and their replacement with foreign CTL epitopes with similar physical properties (i.e., size and hydrophobicity) would result in less structural change and thereby help to conserve HBsAg conformation and subsequent stability/secretability. To test this supposition, we therefore evaluated whether pM29IPQ, pM215IPQ, and pM215GLS encoded structurally authentic VLPs. We did this by evaluating the capacity of pM29IPQ, pM215IPQ, and pM215GLS to package HDAg-L for secretion. HDV is an RNA agent which exists as a subviral satellite of HBV. Coexpression of HDAg-L and HBsAg leads to incorporation (packaging) of HDAg-L into the empty HBsAg-S VLPs and to secretion (3, 34, 46). We cotransfected DNAs encoding HDAg-L and the chimeric HBsAg polypeptide into HuH-7 cells and subsequently determined the presence of secreted HDAg-L in the cell culture supernatant by immunoblotting. HDAg-L was confirmed in the supernatant of cells which had been cotransfected with any one of the chimeric HBsAgs (Fig. 5a, lanes 1 to 3), although to a lesser extent than in cells cotransfected with wild-type HBsAg (Fig. 5a, lane 4). No HDAg-L was detected in the supernatant of cells transfected only with the plasmid encoding HDAg-L, although HDAg-L was clearly present in the cell pellet, confirming that in the absence of HBsAg, HDAg-L was not secreted (Fig. 5b, lane 6). This result indicates that chimeric HBsAg polypeptides in which the HBsAg-specific IPQ or GLS CTL epitopes are replaced by a foreign CTL epitope (M2) retain the capacity for assembly and secretion of structurally sound HBsAg particles.
DISCUSSION
A limited range of antigen delivery techniques are available to prime precursor CTLs to recombinant antigens. These include infection with recombinant replicating (bacterial or viral) vectors, the inoculation of antigen-encoding plasmid DNA or RNA, the association of protein antigen with selected adjuvants (e.g., squalene, Iscom), the injection of VLPs, and the injection of naked lipoprotein particles. HBsAg is the already licensed human vaccine for hepatitis B virus infection, making it an attractive vehicle for the delivery of epitopes from other pathogens and diseases, from both a regulatory and a scientific point of view. While HBsAg has been used experimentally as a carrier in recombinant vaccine studies, this primarily has been in the delivery of a variety of foreign B-cell epitopes for antibody induction, most commonly by the insertion of these epitopes into the external hydrophilic a loop (5, 23, 29, 33, 45). There are few reports of exploiting the inherent powerful immunogenicity of HBsAg to deliver foreign CTL (2, 15, 25) or T-helper (5) epitopes, and this likely relates to difficulties arising in stability and antigen processing that are incurred by the insertion of foreign sequences into hydrophobic regions of HBsAg, of which at least some are essential for correct conformation (13).
Comparison of wild-type HBsAg delivered as a DNA immunogen and that delivered as a VLP immunogen suggests that DNA is the preferred modality both economically and practically and in terms of immunogenic efficacy, at least for CTL induction (6). That a single intramuscular injection of HBsAg plasmid DNA gives CTL detectable for many months postvaccination (7, 9) suggests that continuous exposure to small doses of antigen provided from ongoing DNA transcription may be necessary to maintain immunity. Injected DNA encoding HBsAg DNA is likely to transfect a variety of cells, though only the transfection of professional antigen-presenting cells (APCs) leads to activation of the CTL response via the intracellular (endogenous) processing pathway (38). Particle formation within the APC is not an essential requirement for this to occur (38). Secretion of HBsAg particles (e.g., by DNA-transfected muscle cells) allows uptake of secreted particles by APCs for processing via the exogenous pathway (8), thus providing a second mechanism for HBsAg CTL generation (41). After a single intramuscular injection of HBsAg DNA, the antigen can be found in circulation for at least 1 month, suggesting sustained expression in muscle fibers (26). HBsAg is likely to persist elsewhere, e.g., in follicular dendritic cells, accounting for the sustenance of a prolonged immune response (17). The powerful CTL responses induced by HBsAg DNA vaccination may be explained by prolonged and higher expression, secretion for uptake of particles by APC, superior antigen processing, and the presence of multiple T-helper epitopes. Thus, DNA-based HBsAg immunization is extremely potent and may have an immune potential the same as that of natural infection (39). HBsAg-based DNA vaccination requires a dose as much as 2,500-fold lower than that used in previous clinical trials with conventional DNA (37). The fact that long-lasting CTL and antibody are efficiently induced in neonatal mice with wild-type HBsAg-encoding plasmid DNA (but not HBsAg particles) (41) argues in favor of a DNA vaccine approach for recombinant HBsAg vaccines.
In the present study, we sought to exploit the inherent immunogenicity of HBsAg DNA as a carrier for the delivery of disease-relevant foreign CTL epitopes. We argued that the simplistic approach of inserting CTL epitope-encoding sequences into the HBsAg gene at sites where restriction endonuclease sites fortuitously exist, not accounting for spatial requirements and the tertiary configuration of the HBsAg polypeptide, is likely to yield unpredictable outcomes, and this has hitherto constrained the exploitation of HBsAg for the delivery of inserted foreign CTL epitopes. We reasoned that by deleting endogenous CTL epitopes and replacing them with foreign CTL epitopes of similar size and physical characteristics (hydrophobicity), we might avoid deleterious effects on the structural integrity of the recombinant HBsAg protein. Furthermore, we reasoned that targeting foreign CTL epitopes to sites of proven HBsAg CTL epitopes might predispose to appropriate antigen processing of insert.
A number of HBsAg CTL epitopes in hydrophobic regions of the HBsAg molecule, restricted through murine (39, 42) and human (28) MHC-I molecules, have been described previously. We elected to delete two of the stronger of these endogenous CTL epitopes (IPQ and GLS [Table 2]) and replace them with protective foreign epitopes from the M2 protein of RSV and/or the E7 protein of HPV16.
We show that replacement of DNA encoding the endogenous IPQ epitope with DNA encoding either RSV M29 or M215 epitopes or the HPV oncoprotein RAH epitopes produced chimeric HBsAg DNA immunogens which elicited effector and memory CTL responses to the foreign CTL epitope (Fig. 1, 3, and 4). Moderate effector CTL responses measured by IFN- release (Fig. 1a and b) translated into strong memory responses when specific antigen was reencountered (Fig. 1f and g), as has been seen in other systems (47). We show that M2-specific CTL responses were associated with reduced proliferation of virus in the lungs of RSV-challenged mice and with amelioration in pulmonary pathology. It has been previously demonstrated that protection against RSV afforded by an RSV M2 CTL epitope vaccine is associated with CTLs which kill RSV-expressing cells in vitro (22). Similarly, RAH-specific CTL responses were associated with protection against challenge with an E7-expressing tumor. We further show HBsAg recombinant for both RSV and HPV CTL epitopes elicited simultaneous responses to both epitopes. However, there was evidence of immunodominance of the RSV M2 epitope. (Interestingly, immunodominance of the endogenous HBsAg-specific epitope at the site of the M2 insertion has also been reported [42].) While the memory CTL response to RAH was down-regulated in mice immunized with HBsAg encoding the M2 and RAH epitopes, the mice were still protected against tumor challenge (Fig. 4e). The phenomenon of relatively small CTL responses, as measured by in vitro cytotoxicity assay, translating into substantial tumor protection in vivo has been observed for other tumor vaccination systems in our laboratory (unpublished observations). Nonetheless, we have preliminary data to indicate that immunization with chimeric HBsAg DNA encoding three foreign epitopes elicits CTL responses to all three (not shown). Together, these data suggest the applicability of HBsAg as a vector for the delivery of protective CTL responses to human disease-relevant foreign epitopes and also suggest the possibility of simultaneous protection against multiple diseases with a single polyvalent HBsAg vaccine. The approach of delivering a limited number of MHC-I-restricted epitopes to target multiple diseases in humans by use of HBsAg DNA as a vector is feasible, since (i) at least 15 HBsAg CTL epitopes are candidates for replacement (28, 32, 48); (ii) in general, CTL-mediated pathogen and tumor protection is remarkably parsimonious in terms of the number of CTL epitopes to which the responses must be directed (19, 36); and (iii) a limited number of major MHC-I supertypes account for the majority of MHC polymorphism, allowing broad population coverage with very few disease epitopes (43). Furthermore, immunization with mixtures of chimeric HBsAg DNAs encoding different disease-relevant epitopes may be employed to enhance epitope delivery capacity.
CTL responses to foreign epitopes delivered in the context of chimeric HBsAg DNA compared favorably to those induced by immunization with a vast molar excess of epitope peptide plus adjuvant. This latter approach induces maximal (though short-lived) CTL responses and is a standard against which other CTL-inducing modalities may be compared. In view of the superior immunogenicity of HBsAg-based vaccines (see above) compared to that of conventional vaccine approaches, it would be interesting to compare the numbers of cognate CD8+ CTLs in the lungs of recombinant HBsAg-vaccinated mice and peptide-plus-adjuvant-vaccinated mice following RSV challenge. Additionally, we observed that responses to epitopes inserted into HBsAg could exceed the observed response to the deleted HBsAg epitope (following wild-type HBsAg immunization) which they replaced. These observations suggest that the choice of sites for inserting replacement foreign epitopes within the HBsAg molecule need not be overly influenced by the magnitude of the response to the endogenous HBsAg epitope situated at that site. Response is likely to be a function not only of primary amino acid sequence but also of the context of the epitope within the protein, such that intramolecular competition for cellular processing and HLA binding influences the magnitude of the response (35). It is also noteworthy that an HBsAg CTL epitope can be replaced with a foreign CTL epitope restricted through a different MHC-I haplotype to elicit strong insert-directed CTL responses.
In spite of our approach of replacing HBsAg-specific CTL epitopes with the M2 CTL epitope, the assembly and/or stability and secretion of recombinant HBsAg was reduced compared with those of wild-type particles (measured as reduced capacity to package hepatitis delta virus large antigen for secretion [Fig. 5]). The inference is that the replacement insertions interfere with the assembly or stability of structurally appropriate HBsAg particles. Length of insert is the major factor determining the level of secretion (11). However, it is noteworthy in our study that impaired secretion of functionally intact HBsAg particles occurred even when the replacement of the HBsAg-specific CTL epitope with foreign epitope resulted in no net change in the number of amino acid residues in the chimeric HBsAg compared to that in the wild type (pM29IPQ [Fig. 5]). This underscores the importance of specific sequence in topogenic regions essential for particle assembly (11, 13). In preliminary experiments (not shown), we found that antibody induction to a B mimotope of RSV (Table 1) inserted into the hydrophilic a determinant was lost when CTL epitopes were inserted at the IPQ and GLS sites (pM215IPQ.RAHGLS.HW1) (Table 2).
Immunization with plasmid HBsAg DNA evokes cytokines which predispose to a Th1-type response (26). Provision of T help for primary effector responses may be provided by HBsAg-specific T-helper epitopes (2, 30) or CpG dinucleotides in bacterial DNA (10). The recall of CTL-epitope specific memory (required when the pathogen or disease against which the chimeric HBsAg vaccine is directed is encountered) is likely to be T help independent (reviewed in reference 47).
In summary, using a strategy of deletion of DNA encoding HBsAg-specific CTL epitopes and replacement with DNA encoding foreign CTL epitopes, we have derived chimeric HBsAg DNA immunogens which elicited foreign epitope-directed effector and memory CTL responses associated with pathogen- and tumor-protective responses in vivo. We show that HBsAg recombinant for more than one CTL epitope elicited simultaneous responses to both epitopes. There are generic implications for the application of HBsAg, which is already licensed as human vaccine, as a vector for the delivery of multiple disease-relevant protective CTL responses. There are also specific implications for HBsAg-vectored vaccines for RSV- and HPV-associated tumors.
ACKNOWLEDGMENTS
W.-P. Woo and T. Doan contributed equally to the experimentation and should be regarded as joint first authors.
H.-J. Netter and R. W. Tindle contributed equally to the design and implementation of the study and should be regarded as joint senior authors.
We thank P. Young for providing RSV and advice and M. Mather and H. Guo for helpful discussions. D. West and her staff provided expert animal husbandry. W.-J. Liu advised with the molecular cloning. M. Duhig aided in the interpretation of lung histology.
The work was funded by grants from the Royal Children's Hospital Foundation and Queensland Cancer Fund.
Contribution no. 222 of the Sir Albert Sakzewski Virus Research Centre.
REFERENCES
Berger, J., J. Hauber, R. Hauber, R. Geiger, and B. R. Cullen. 1988. Secreted placental alkaline phosphatase: a powerful new quantitative indicator of gene expression in eukaryotic cells. Gene 66:1-10.
Bryder, K., H. Sbai, H. V. Nielsen, S. Corbet, C. Nielsen, R. G. Whalen, and A. Fomsgaard. 1999. Improved immunogenicity of HIV-1 epitopes in HBsAg chimeric DNA vaccine plasmids by structural mutations of HBsAg. DNA Cell Biol. 18:219-225.
Chang, F. L., P. J. Chen, S. J. Tu, C. J. Wang, and D. S. Chen. 1991. The large form of hepatitis delta antigen is crucial for assembly of hepatitis delta virus. Proc. Natl. Acad. Sci. USA 88:8490-8494.
Chargelegue, D., O. E. Obeid, S. C. Hsu, M. D. Shaw, A. N. Denbury, G. Taylor, and M. W. Steward. 1998. A peptide mimic of a protective epitope of respiratory syncytial virus selected from a combinatorial library induces virus-neutralizing antibodies and reduces viral load in vivo. J. Virol. 72:2040-2046.
Chengalvala, M. V., R. A. Bhat, B. M. Bhat, S. K. Vernon, and M. D. Lubeck. 1999. Enhanced immunogenicity of hepatitis B surface antigen by insertion of a helper T cell epitope from tetanus toxoid. Vaccine 17:1035-1041.
Davis, H. L., C. L. Brazolot Millan, M. Mancini, M. J. McCluskie, M. Hadchouel, L. Comanita, P. Tiollais, R. G. Whalen, and M. L. Michel. 1997. DNA-based immunization against hepatitis B surface antigen (HBsAg) in normal and HBsAg-transgenic mice. Vaccine 15:849-852.
Davis, H. L., M. Mancini, M. L. Michel, and R. G. Whalen. 1996. DNA-mediated immunization to hepatitis B surface antigen: longevity of primary response and effect of boost. Vaccine 14:910-915.
Davis, H. L., M. L. Michel, and R. G. Whalen. 1993. DNA-based immunization induces continuous secretion of hepatitis B surface antigen and high levels of circulating antibody. Hum. Mol. Genet. 2:1847-1851.
Davis, H. L., R. Schirmbeck, J. Reimann, and R. G. Whalen. 1995. DNA-mediated immunization in mice induces a potent MHC class I-restricted cytotoxic T lymphocyte response to the hepatitis B envelope protein. Hum. Gene Ther. 6:1447-1456.
Davis, H. L., R. Weeratna, T. J. Waldschmidt, L. Tygrett, J. Schorr, A. M. Krieg, and R. Weeranta. 1998. CpG DNA is a potent enhancer of specific immunity in mice immunized with recombinant hepatitis B surface antigen. J. Immunol. 160:870-876.
Delpeyroux, F., N. Chenciner, A. Lim, M. Lambert, Y. Malpiece, and R. E. Streeck. 1987. Insertions in the hepatitis B surface antigen. Effect on assembly and secretion of 22-nm particles from mammalian cells. J. Mol. Biol. 195:343-350.
Doan, T., M. Chambers, M. Street, G. J. Fernando, K. Herd, P. Lambert, and R. Tindle. 1998. Mice expressing the E7 oncogene of HPV16 in epithelium show central tolerance, and evidence of peripheral anergising tolerance, to E7-encoded cytotoxic T-lymphocyte epitopes. Virology 244:352-364.
Eble, B. E., D. R. MacRae, V. R. Lingappa, and D. Ganem. 1987. Multiple topogenic sequences determine the transmembrane orientation of the hepatitis B surface antigen. Mol. Cell. Biol. 7:3591-3601.
Feltkamp, M. C., H. L. Smits, M. P. Vierboom, R. P. Minnaar, B. M. de Jongh, J. W. Drijfhout, J. Ter Schegget, C. J. Melief, and W. M. Kast. 1993. Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type 16-transformed cells. Eur. J. Immunol. 23:2242-2249.
Fomsgaard, A., H. V. Nielsen, K. Bryder, C. Nielsen, R. Machuca, L. Bruun, J. Hansen, and S. Buus. 1998. Improved humoral and cellular immune responses against the gp120 V3 loop of HIV-1 following genetic immunization with a chimeric DNA vaccine encoding the V3 inserted into the hepatitis B surface antigen. Scand. J. Immunol. 47:289-295.
Graham, B. S., M. D. Perkins, P. F. Wright, and D. T. Karzon. 1988. Primary respiratory syncytial virus infection in mice. J. Med. Virol. 26:153-162.
Gray, D., M. Kosco, and B. Stockinger. 1991. Novel pathways of antigen presentation for the maintenance of memory. Int. Immunol. 3:141-148.
Gupta, R. K., and G. R. Siber. 1995. Adjuvants for human vaccines—current status, problems and future prospects. Vaccine 13:1263-1276.
Hsu, S. C., D. Chargelegue, O. E. Obeid, and M. W. Steward. 1999. Synergistic effect of immunization with a peptide cocktail inducing antibody, helper and cytotoxic T-cell responses on protection against respiratory syncytial virus. J. Gen. Virol. 80:1401-1405.
Hsu, S. C., D. M. Shaw, and M. W. Steward. 1995. The induction of respiratory syncytial virus-specific cytotoxic T-cell responses following immunization with a synthetic peptide containing a fusion peptide linked to a cytotoxic T lymphocyte epitope. Immunology 85:347-350.
Kruger, D. H., R. Ulrich, and W. H. Gerlich. 1999. Chimeric virus-like particles as vaccines. Biol. Chem. 380:275-276.
Kulkarni, A. B., M. Connors, C. Y. Firestone, H. C. Morse III, and B. R. Murphy. 1993. The cytolytic activity of pulmonary CD8+ lymphocytes, induced by infection with a vaccinia virus recombinant expressing the M2 protein of respiratory syncytial virus (RSV), correlates with resistance to RSV infection in mice. J. Virol. 67:1044-1049.
Laub, O., L. B. Rall, M. Truett, Y. Shaul, D. N. Standring, P. Valenzuela, and W. J. Rutter. 1983. Synthesis of hepatitis B surface antigen in mammalian cells: expression of the entire gene and the coding region. J. Virol. 48:271-280.
Le, T. T., D. Drane, J. Malliaros, J. C. Cox, L. Rothel, M. Pearse, T. Woodberry, J. Gardner, and A. Suhrbier. 2001. Cytotoxic T cell polyepitope vaccines delivered by ISCOMs. Vaccine 19:4669-4675.
Le Borgne, S., M. Mancini, R. Le Grand, M. Schleef, D. Dormont, P. Tiollais, Y. Riviere, and M. L. Michel. 1998. In vivo induction of specific cytotoxic T lymphocytes in mice and rhesus macaques immunized with DNA vector encoding an HIV epitope fused with hepatitis B surface antigen. Virology 240:304-315.
Leclerc, C., E. Deriaud, M. Rojas, and R. G. Whalen. 1997. The preferential induction of a Th1 immune response by DNA-based immunization is mediated by the immunostimulatory effect of plasmid DNA. Cell. Immunol. 179:97-106.
Lin, K. Y., F. G. Guarnieri, K. F. Staveley-O'Carroll, H. I. Levitsky, J. T. August, D. M. Pardoll, and T. C. Wu. 1996. Treatment of established tumors with a novel vaccine that enhances major histocompatibility class II presentation of tumor antigen. Cancer Res. 56:21-26.
Loirat, D., F. A. Lemonnier, and M. L. Michel. 2000. Multiepitopic HLA-A0201-restricted immune response against hepatitis B surface antigen after DNA-based immunization. J. Immunol. 165:4748-4755.
Michel, M. L., P. Pontisso, E. Sobczak, Y. Malpiece, R. E. Streeck, and P. Tiollais. 1984. Synthesis in animal cells of hepatitis B surface antigen particles carrying a receptor for polymerized human serum albumin. Proc. Natl. Acad. Sci. USA 81:7708-7712.
Milich, D. R. 1988. T- and B-cell recognition of hepatitis B viral antigens. Immunol. Today 9:380-386.
Nakabayashi, H., K. Taketa, K. Miyano, T. Yamane, and J. Sato. 1982. Growth of human hepatoma cell lines with differentiated functions in chemically defined medium. Cancer Res. 42:3858-3863.
Nayersina, R., P. Fowler, S. Guilhot, G. Missale, A. Cerny, H. J. Schlicht, A. Vitiello, R. Chesnut, J. L. Person, A. G. Redeker, et al. 1993. HLA A2 restricted cytotoxic T lymphocyte responses to multiple hepatitis B surface antigen epitopes during hepatitis B virus infection. J. Immunol. 150:4659-4671.
Netter, H. J., T. B. Macnaughton, W. P. Woo, R. Tindle, and E. J. Gowans. 2001. Antigenicity and immunogenicity of novel chimeric hepatitis B surface antigen particles with exposed hepatitis C virus epitopes. J. Virol. 75:2130-2141.
O'Malley, B., and D. Lazinski. 2002. A hepatitis B surface antigen mutant that lacks the antigenic loop region can self-assemble and interact with the large hepatitis delta antigen. J. Virol. 76:10060-10063.
Perkins, D. L., G. Berriz, T. Kamradt, J. A. Smith, and M. L. Gefter. 1991. Immunodominance: intramolecular competition between T cell epitopes. J. Immunol. 146:2137-2144.
Rosenberg, S. A. 2001. Progress in human tumor immunology and immunotherapy. Nature 411:380-384.
Roy, M. J., M. S. Wu, L. J. Barr, J. T. Fuller, L. G. Tussey, S. Speller, J. Culp, J. K. Burkholder, W. F. Swain, R. M. Dixon, G. Widera, R. Vessey, A. King, G. Ogg, A. Gallimore, J. R. Haynes, and F. D. Heydenburg. 2000. Induction of antigen-specific CD8+ T cells, T helper cells, and protective levels of antibody in humans by particle-mediated administration of a hepatitis B virus DNA vaccine. Vaccine 19:764-778.
Sbai, H., J. Schneider, A. V. Hill, and R. G. Whalen. 2002. Role of transfection in the priming of cytotoxic T-cells by DNA-mediated immunization. Vaccine 20:3137-3147.
Schirmbeck, R., W. Bohm, K. Ando, F. V. Chisari, and J. Reimann. 1995. Nucleic acid vaccination primes hepatitis B virus surface antigen-specific cytotoxic T lymphocytes in nonresponder mice. J. Virol. 69:5929-5934.
Schirmbeck, R., K. Melber, A. Kuhrober, Z. A. Janowicz, and J. Reimann. 1994. Immunization with soluble hepatitis B virus surface protein elicits murine H-2 class I-restricted CD8+ cytotoxic T lymphocyte responses in vivo. J. Immunol. 152:1110-1119.
Schirmbeck, R., and J. Reimann. 2002. Alternative processing of endogenous or exogenous antigens extends the immunogenic, H-2 class I-restricted peptide repertoire. Mol. Immunol. 39:249-259.
Schirmbeck, R., D. Stober, S. El Kholy, P. Riedl, and J. Reimann. 2002. The immunodominant, Ld-restricted T cell response to hepatitis B surface antigen (HBsAg) efficiently suppresses T cell priming to multiple Dd-, Kd-, and Kb-restricted HBsAg epitopes. J. Immunol. 168:6253-6262.
Sette, A., and J. Sidney. 1999. Nine major HLA class I supertypes account for the vast preponderance of HLA-A and -B polymorphism. Immunogenetics 50:201-212.
Vitiello, A., D. Marchesini, J. Furze, L. A. Sherman, and R. W. Chesnut. 1991. Analysis of the HLA-restricted influenza-specific cytotoxic T lymphocyte response in transgenic mice carrying a chimeric human-mouse class I major histocompatibility complex. J. Exp. Med. 173:1007-1015.
von Brunn, A., K. Fruh, H. M. Muller, H. W. Zentgraf, and H. Bujard. 1991. Epitopes of the human malaria parasite P. falciparum carried on the surface of HBsAg particles elicit an immune response against the parasite. Vaccine 9:477-484.
Wang, C. J., P. J. Chen, J. C. Wu, D. Patel, and D. S. Chen. 1991. Small-form hepatitis B surface antigen is sufficient to help in the assembly of hepatitis delta virus-like particles. J. Virol. 65:6630-6636.
Wherry, E. J., and R. Ahmed. 2004. Memory CD8 T-cell differentiation during viral infection. J. Virol. 78:5535-5545.
Wild, J., M. J. Grusby, R. Schirmbeck, and J. Reimann. 1999. Priming MHC-I-restricted cytotoxic T lymphocyte responses to exogenous hepatitis B surface antigen is CD4+ T cell dependent. J. Immunol. 163:1880-1887.(Wai-Ping Woo, Tracy Doan,)
Department of Microbiology, Monash University, Melbourne, Australia
ABSTRACT
Although hepatitis B surface antigen (HBsAg) per se is highly immunogenic, its use as a vector for the delivery of foreign cytotoxic T-lymphocyte (CTL) epitopes has met with little success because of constraints on HBsAg stability and secretion imposed by the insertion of foreign sequence into critical hydrophobic/amphipathic regions. Using a strategy entailing deletion of DNA encoding HBsAg-specific CTL epitopes and replacement with DNA encoding foreign CTL epitopes, we have derived chimeric HBsAg DNA immunogens which elicited effector and memory CTL responses in vitro, and pathogen- and tumor-protective responses in vivo, when the chimeric HBsAg DNAs were used to immunize mice. We further show that HBsAg DNA recombinant for both respiratory syncytial virus and human papillomavirus CTL epitopes elicited simultaneous responses to both pathogens. These data demonstrate the efficacy of HBsAg DNA as a vector for the delivery of disease-relevant protective CTL responses. They also suggest the applicability of the approach of deriving chimeric HBsAg DNA immunogens simultaneously encoding protective CTL epitopes for multiple diseases. The DNAs we tested formed chimeric HBsAg virus-like particles (VLPs). Thus, our results have implications for the development of vaccination strategies using either chimeric HBsAg DNA or VLP vaccines. HBsAg is the globally administered vaccine for hepatitis B virus infection, inviting its usage as a vector for the delivery of immunogens from other diseases.
INTRODUCTION
Vaccination strategies capable of safely and effectively inducing cytotoxic T-lymphocyte (CTL) responses which are acceptable in humans have not proven easy to develop (18). The hepatitis B virus (HBV) surface antigen (HBsAg) small envelope protein (HBsAg-S) self-assembles into highly organized virus-like particles (VLPs) (11). HBsAg VLPs are extremely efficient at inducing long-lived HBsAg-specific CTL responses when administered as a vaccine, comparing favorably with traditional strong CTL inducers in this regard (40). This likely relates to the presentation of the CTL epitopes as multiple copies in repetitive arrays, a format known to enhance induction of CTL responses (21). HBsAg may also be delivered as a DNA vaccine. By use of this strategy, CTL induction by intracellular translated HBsAg protein occurs through the classical endogenous pathway as well as through the alternative exogenous pathway via secreted HBsAg particles (28, 38, 39).
HBsAg-S polypeptide is composed of 226 residues, is synthesized at the endoplasmic reticulum anchored at the lipid membrane and is secreted as empty 22-nm VLPs, with each VLP consisting of 100 to 150 polypeptides. The HBsAg molecule has a complex configuration and contains at least two transmembrane regions, an internal luminal domain and a hydrophilic external domain (the a determinant). The C-terminal domain is hydrophobic and it probably contains two additional transmembrane regions. HBsAg CTL epitopes are located primarily in the transmembrane and luminal domains (2, 42).
Although HBsAg per se is highly immunogenic, the use of chimeric HBsAg to deliver foreign epitopes has met with limited success (particularly in the delivery of CTL epitopes), probably because the strategy hitherto used, that of introducing foreign sequences into the HBV genome at sites where restriction endonuclease sites fortuitously exist or of extending the termini of HBsAg protein with epitope sequences, does not necessarily account for spatial requirements for VLP formation (2, 11). In the present study, we report a novel strategy to exploit the superior immunogenicity of HBsAg by deleting HBsAg-specific CTL epitopes and replacing them with foreign CTL epitopes of similar physical properties (i.e., size and hydrophobicity). We reasoned this approach might predispose to the maintenance of spatial requirements, the conformation of the HBsAg molecule, and, in addition, the maintenance of appropriate antigen processing.
We demonstrate that immunization with DNA encoding chimeric HBsAg in which an HBsAg-specific CTL epitope(s) was deleted and replaced with CTL epitopes of respiratory syncytial virus (RSV) or human papillomavirus (HPV) elicited strong CTL responses to the inserted epitopes. Immunization with chimeric HBsAg DNA afforded protection against RSV infection or the growth of HPV type 16 (HPV16)-associated tumors. We further show that immunization with DNA encoding HBsAg recombinant for both RSV and HPV CTL epitopes elicited CTL responses to both foreign epitopes and afforded protection against disease.
We demonstrate that chimeric HBsAg DNAs constructed using the deletion-replacement strategy form VLPs which are secreted from cells. HBsAg VLPs are globally licensed as a vaccine for HBV infections in humans (including children) with a ca. 12-year history of use in millions of recipients, inviting their usage as a vector for the delivery of immunogens from other diseases. Our results demonstrate that the powerful immunogenicity of HBsAg may be exploited to deliver foreign CTL epitopes relevant for other diseases and have generic implications for the development of HBsAg as a multivalent vaccine vector. There are also specific implications for vaccines against RSV infection and HPV-associated carcinoma.
MATERIALS AND METHODS
Cloning procedures. The plasmid pcD3-HBsAgS, which contains a AgeI restriction site at a location corresponding to amino acids (aa) 127 and 128 of the a determinant region (33), was engineered to delete the HBsAg-specific CTL epitopes IPQ and GLS (Table 1) and to introduce the restriction enzyme sites NheI and BlpI, respectively, by PCR-driven site-directed mutagenesis. Plasmid pM215IPQ (Table 2) was derived by inserting DNA encoding the RSV-specific M215 epitope at the NheI site. Further chimeric HBsAg plasmids were derived as summarized in Table 2. Plasmid pM215IPQ.RAHGLS.HWI was derived by inserting DNA sequences encoding the RSV B-cell-specific mimotope (4) into the AgeI restriction site and then sequentially inserting DNA encoding the M215 epitope and the RAH epitope at the IPQ and GLS sites, respectively. Standard molecular cloning and plasmid purification procedures were used. All plasmids were verified by DNA sequencing (PCR Sprint temperature cyclin system [Hybaid]).
Transfection. HuH-7 cells (31) were transfected with bacterial plasmid DNA by use of calcium phosphate. The supernatant was harvested 5 days later, and HBsAg was measured by the Abbott Prism HBsAg assay (Abbott Diagnostics). The level of HBsAg in the cell culture fluid was quantified by comparison with a commercially available vaccine (Engerix-B, 20 μg/ml; Smith-Kline Beecham). The transfection efficiencies of different plasmids were normalized to the activity of secreted alkaline phosphatase as previously described (1). The variation in the range of secreted alkaline phosphatase was less than twofold. The presence of hepatitis delta virus (HDV) large antigen (HDAg-L) was identified by immunoblotting using a human anti-HDV antibody and assayed by the ECL-plus detection system (Amersham) (33).
Mice. A2.1Kb mice (44) make CTL responses restricted through HLA A0201 and H-2b class 1 molecules. (A2.1Kb x BALB/c)F1 hybrid mice additionally make CTL responses restricted through H-2d. Mice were housed under specific-pathogen-free conditions and used at 7 to 15 weeks of age.
Peptides. Peptides were synthesized using 9-fluorenylmethoxy carbonyl chemistry and analyzed by high-performance liquid chromatography and by amino acid analysis by Chiron Mimotopes (Melbourne, Australia). The RSV F protein mimotope was synthesized as a tetrameric multiple antigen peptide construct as described previously (4).
Immunization and restimulation of splenocytes. Mice were immunized twice at 2-week intervals intradermally (i.d.) in the ear with 100 μg of purified plasmid DNA. Two weeks later, spleens were removed and splenocytes were restimulated in vitro for 6 days as described previously (12) with 1 μg/ml cognate peptide. For peptide immunizations, mice were immunized subcutaneously (s.c.) at the tail base with 50 μg peptide plus 0.25 μg tetanus toxoid as a source of T-helper epitopes plus 10 μg Quil A adjuvant (24). Ten days later, spleens were harvested and splenocytes were restimulated as described above.
Cells. EL4.A2 cells (12) are susceptible to specific CTL lysis through both H-2b and HLA A0201 restriction pathways. P815 is susceptible to specific CTL lysis through the H-2d restriction pathway. Cells were maintained as described previously (12).
CTL assays. CTL assays were conducted as previously described (12). In summary, target cells (104 per well), sensitized at 37°C for 1 h with 1 μg/ml cognate or irrelevant peptide or with medium alone and labeled with 100 μCi 51chromium (Cr), were incubated with effector cells at various effector-to-target cell ratios in triplicate in 96-well microtiter plates. Negative controls included wells containing target cells but no effector cells (background). Supernatants were harvested from CTL assays at 4 h, and 51Cr release was quantified by gamma counting. Results are expressed as percent cytotoxicity ± standard deviation [(51Cr release in experimental wells – background/detergent-mediated total release – background) x 100%]. Experimental and control values were compared for significant difference using Student's t test.
Murine IFN- ELISPOT assay. Epitope-specific gamma interferon (IFN-)-secreting spleen cells were enumerated ex vivo by an enzyme-linked immunospot (ELISPOT) assay with minimal CD8+ T-cell epitope peptides essentially as described previously (24). IFN- spots were counted using an AID EliSpot reader system. Results were calculated as IFN--positive cells/106 spleen cells. Experimental and control values were compared for significant difference using analysis of variance.
RSV propagation. HEp-2 cells (American Type Culture Collection, Manassas, VA) were maintained in RPMI 1640 medium (Invitrogen, Mt. Waverley, VIC, Australia) supplemented with 292 μg/ml GlutaMax-1 (Gibco BRL), 1 mM sodium pyruvate, 20 mM HEPES, 50 μM -mercaptoethanol, 100 IU/ml penicillin, 100 μg/ml streptomycin (Gibco BRL), and 10% fetal bovine serum (CSL Ltd, Parkville, Australia). RSV A2 (VR-1302) strain (American Type Culture Collection) was adsorbed onto the HEp-2 cell monolayer at a multiplicity of infection of 0.04 PFU/cell for 2 h at 37°C prior to culturing in supplemented RPMI 1640 medium for 3 to 6 days. RSV was recovered from the supernatant following freeze-thawing of monolayers displaying a cytopathic effect. RSV was quantified by immunofocus assay (16), using goat anti-RSV primary antibody (Chemicon, Australia), anti-goat immunoglobulin (Ig) horseradish peroxidase-conjugated second antibody, and DAB (tetra-aminobiphenyl hydrochloride) substrate (Sigma, Australia) according to the manufacturer's instructions.
Evaluation of RSV infection. Mice were inoculated intranasally with 8 x 105 PFU of RSV and processed for histochemical staining or immunofluorescence using goat anti-RSV antibody and fluorescein isothiocyanate anti-goat Ig-detecting antibody (Chemicon). Lung samples were also used for RSV quantitation by plaque assay (16). Plaques were confirmed by immunostaining with goat anti-RSV antibody (Chemicon) followed by treatment with anti-goat Ig horseradish peroxidase-conjugated antibody (Chemicon, Australia) and development with DAB substrate (Sigma, Australia) and urea-H2O2-NaCl according to the manufacturer's instructions (Sigma).
Tumor protection assays. Groups of H-2b mice (five per group) were immunized with 10 μg plasmid DNA i.d. or with 100 μg of E7 peptide plus tetanus toxoid plus Quil A s.c. TC-1 cells, which express the E7 tumor-associated antigen of human papillomavirus type 16 (27), were subsequently injected (2 x 105 in 0.1 ml Hanks' buffered salt solution) s.c. on the flank. (The tumor dose was predetermined by titration experiments to discern a minimal dose giving rise to tumor in 100% of unimmunized mice.) Tumor growth was monitored every 2 days, and mice were euthanized when tumor volumes exceeded 1,000 mm3. Unimmunized mice received the same number of cells and served as a control. Data are represented as Kaplan-Meier curves of the percent tumor-free mice at given time points after tumor injection.
RESULTS
Immunization with chimeric HBsAg DNA encoding the RSV M2 CTL epitope elicits M2-directed CTL. A number of HBsAg CTL epitopes in hydrophobic regions of the HBsAg molecule, restricted through murine (39, 42) and human (28) major histocompatibility complex class I (MHC-I) molecules, have been described previously. In a first set of experiments, we elected to delete two of the stronger of these HBsAg-specific CTL epitopes (IPQ and GLS [Table 1]) and replace them with a protective epitope from the M2 protein of RSV. The minimal M2 CTL epitope of RSV is a 9-mer represented by M2 residues 82 to 90 (designated M29); a larger 15-mer peptide sequence represented by residues 81 to 95 (designated M215) also functions as a CTL epitope (Table 1 and reference 20). We asked whether chimeric HBsAg DNA vaccines encoding the RSV CTL epitope would elicit M2-directed effector and memory CTL responses.
Effector responses. We quantified epitope-specific IFN- secretion of ex vivo splenocytes by ELISPOT assay. Ex vivo splenocytes from mice immunized with pM29IPQ, pM215IPQ, or pM215GLS DNA (Table 2) secreted significant amounts of IFN- when cultured in vitro with, but not without, cognate M2 peptide (Fig. 1a to c), indicating the induction of an M2-directed effector CTL response. Ex vivo splenocytes from mice immunized with wild-type HBsAg DNA (pHBsAgWT) and cultured with M2 peptide did not secrete IFN- above the level observed when culturing was done without peptide (data not shown).
Memory response. We also asked whether immunization with pM29IPQ, pM215IPQ, or pM215GLS DNA would induce memory CTL capable of being restimulated in vitro to kill target cells displaying the M2 CTL epitope. Splenocytes from pM29IPQ, pM215IPQ, or pM215GLS DNA-immunized mice were restimulated in vitro for 6 days with M215 peptide and reacted with H-2d target cells pulsed with M215 epitope peptide. Restimulated splenocytes from all these immunized groups of mice killed M215 peptide-pulsed targets (Fig. 1f to h). Notably, the efficiency of killing was comparable to that seen in identically restimulated splenocytes from M215 peptide-plus-adjuvant-immunized mice (Fig. 1i). (Immunization with high-dose CTL epitope peptide plus adjuvant is a powerful inducer of CTL responses [12] and is a standard against which other CTL-inducing modalities may be compared.) It is noteworthy that the percent killing of M2-pulsed target cells by pM29IPQ or pM215IPQ DNA-immunized mice exceeded the percent killing of target cells pulsed with the HBsAg-specific peptide IPQ by restimulated splenocytes from pHBsAgWT immunized mice (Fig. 1j). Taken together, these data indicate that replacement of DNA encoding HBsAg-specific CTL epitopes (IPQ and GLS) with DNA encoding a CTL epitope from the RSV M2 protein to generate chimeric HBsAg DNA immunogens engenders M2-directed effector and memory CTL responses of intensities comparable to those engendered by immunization with peptide plus adjuvant and to those of the HBsAg-specific epitopes engendered by immunization with pHBsAg wild-type DNA.
Chimeric HBsAg immunogen containing two copies of foreign epitope. We also asked whether two copies of the M2 epitope, inserted at the IPQ and GLS sites in HBsAg, respectively, would elicit an M2-directed CTL response that was enhanced in comparison with that elicited by chimeric HBsAg containing a single copy at the IPQ site or at the GLS site alone. Mice were immunized with a construct containing two pM215 epitopes at the different sites (pM215IPQ.GLS.HWI [Table 2]) or with constructs containing one M215 epitope but at different sites (pM215IPQ.HWI and pM215GLS.HWI). (Note that the actual insertion at the two sites differed because of the introduction of different endonuclease restriction half-sites flanking the M215 epitope [Table 2 and "Cloning procedures" in Materials and Methods].) No significant augmentation of the number of IFN--secreting ex vivo splenocytes (Fig. 1k) or of the cytotoxic efficacy of restimulated splenocytes (Fig. 1, panel l) compared to that seen for mice immunized with chimeric HBsAg DNA encoding one copy of the epitope was recorded for mice immunized with chimeric HBsAg DNA encoding two copies.
Immunization with plasmid encoding chimeric HBsAg containing a CTL epitope of RSV M2 protein partially protects against RSV infection. Groups of mice immunized with DNA encoding M2-containing HBsAg or wild-type HBsAg were inoculated with RSV, and 4 days later, lungs were removed for RSV quantitation, histology, and RSV-directed immunofluorescence. Four of five mice immunized with wild-type HBsAg DNA displayed detectable RSV, whereas only one of five mice immunized with pM215GLS or pM215IPQ DNA displayed detectable RSV in lungs (Table 3). Among those mice where RSV was detected, the viral loads per lung were significantly less in mice immunized with pM215GLS or pM215IPQ DNA than in mice immunized with DNA encoding wild-type HBsAg (Table 3).
The lungs of wild-type HBsAg-immunized mice challenged with RSV showed a pathology consistent with severe interstitial pneumonia (bronchiolar epithelial cell damage, extensive infiltrations of inflammatory cells, and hemorrhaging into the inflamed bronchiole) (Fig. 2a, panel i). However, lungs from mice immunized with plasmid pM215GLS (Fig. 2a, panel ii) or pM215IPQ (not shown) displayed less bronchiolar epithelial cell damage, little mononuclear cell infiltration of the alveoli, and less hemorrhaging into bronchioles. Peri- and extrabronchiolar RSV infection was detected in immunofluorescence-stained lung cryosections of pHBsAgWT-immunized mice (Fig. 2b, panel i) but not in those of mice immunized with pM215GLS DNA (Fig. 2b, panel ii).
Together, the data in Table 3 and Fig. 2 indicate that immunization with chimeric HBsAg DNA expressing a CTL epitope of the RSV M protein confers protection against RSV infection and associated pathology.
Immunization with plasmid encoding chimeric HBsAg containing an HPV16 E7 CTL epitope elicits E7-specific CTL responses and protects against E7-expressing tumor. We extended the concept of HBsAg-vectored delivery of protective CTL epitopes to the RAH CTL epitope of the E7 tumor-associated antigen of human papillomavirus type 16 (Table 1). We inserted DNA encoding the RAH epitope into the IPQ site of the HBsAg gene to create the recombinant construct pRAHIPQ (Table 2). Ex vivo splenocytes from mice immunized with pRAHIPQ DNA, but not from those immunized with pHBsAg wild-type DNA, and cultured with RAH peptide contained a significantly higher number of IFN--secreting cells than splenocytes cultured without RAH peptide (Fig. 3a and b). Additionally, restimulated splenocytes from pRAHIPQ DNA-immunized mice, but not from pHBsAg wild-type DNA-immunized mice, specifically lysed target cells pulsed with RAH peptide (Fig. 3d and e). These data indicate specific in vivo priming of RAH-directed effector and memory T cells by immunization with pRAHIPQ DNA. To determine whether immunization would protect against E7-expressing tumor, groups of mice were immunized with plasmids pRAHIPQ and pHBsAg and challenged with TC-1 tumor cells. Eighty percent of mice immunized with pHBsAg developed tumor within 15 days. In contrast, none of the mice immunized with pRAHIPQ DNA developed tumor (Fig. 3g). The pRAHIPQ-induced in vitro and in vivo responses compared favorably with those induced by immunization with RAH peptide plus adjuvant (Fig. 3c, f, and h).
Together, these data indicate that immunization with DNA encoding rHBsAg containing a CTL epitope (RAH) of HPV16 E7 oncoprotein protects mice against challenge with an E7-expressing tumor, and this protection is associated with RAH-directed effector and memory CTL responses.
Immunization with a plasmid encoding chimeric HBsAg containing two protective CTL epitopes from distinct pathogens elicits CTL responses to both epitopes. We extended the concept of HBsAg-vectored protective CTL epitopes to the concomitant delivery of RSV and HPV epitopes delivered by the same chimeric HBsAg DNA. Mice were immunized with pM215IPQ.RAHGLS.HW1 DNA (Table 2), and CTL responses were examined by IFN- ELISPOT assay on ex vivo splenocytes and by 51Cr release cytotoxicity assay mediated by specifically restimulated splenocytes. Splenocytes from pM215IPQ.RAHGLS.HW1-immunized mice secreted IFN- when cultured with M215 peptide (Fig. 4a). The M215-directed response was similar (P > 0.05) to that in splenocytes from mice immunized with a single recombinant encoding M215 alone (pM215IPQ) (Fig. 4a). The RAH-directed IFN- response was lost (P < 0.001) in mice immunized with the double recombinant compared with that in mice immunized with the single recombinant encoding the RAH epitope alone (pRAHGLS) (Fig. 4b).
Restimulated splenocytes from mice immunized with pM215IPQ.RAHGLS.HW1 killed target cells pulsed with either M215 peptide or RAH peptide (Fig. 4c and d). While the restimulated splenocytes killed RAH-expressing target cells significantly more than target cells not expressing RAH (50:1 effector:target cell ratio; P < 0.001) (Fig. 4d), the killing was less than that observed with restimulated splenocytes from mice immunized with pRAHGLS encoding only one inserted RAH CTL epitope (P < 0.001). (Note that RAH-restimulated splenocytes from pHBsAgWT-immunized mice fail to kill RAH-expressing target cells [data not shown].)
In view of the down-regulated CTL response to the RAH epitope in mice immunized with pM215IPQ.RAHGLS.HW1, we asked whether these mice were nonetheless protected against RAH-expressing tumor. Mice immunized with the double recombinant pM215IPQ.RAHGLS.HW1 or with the single recombinant pRAHGLS were fully protected against E7-expressing tumor. In contrast, 40 to 60% of unvaccinated mice or mice immunized with plasmids encoding wild-type HBsAg or chimeric HBsAg not containing the RAH epitope developed tumor within 8 days (Fig. 4c).
Together, these data indicate that immunization with rHBsAg DNA encoding two foreign CTL epitopes from different pathogens elicits effector and memory CTL responses to both epitopes. Although the RAH-directed CTL response measured in vitro was down-regulated in mice immunized with the chimeric HBsAg DNA encoding two epitopes, these mice were nonetheless protected against tumor challenge as effectively as mice immunized with the chimeric HBsAg DNA encoding the RAH epitope alone.
Epitope replacement allows for efficient formation and secretion of VLPs. DNA vaccines induce CTL responses through the classical endogenous pathway via intracellular translation and antigen processing of the proteins they encode. In addition, HBsAg DNA vaccines may induce CTL through the alternative exogenous pathway via secreted HBsAg VLPs (9, 28, 38, 39). It was thus relevant to ask if our strategy for deriving chimeric HBsAg DNA vaccines encoding foreign CTL epitopes also allowed for the formation of secretable HBsAg VLPs. Previous studies to derive HBsAg VLP immunogens containing foreign CTL epitopes have explored the insertion of foreign domains into the pre-S2 region (15). Indiscriminate insertions into HBsAg may lead to unpredictable stability and secretion, probably due to conformational changes in the recombinant HBsAg molecule (11). Thus, we reasoned that deletion of HBsAg-specific CTL epitopes and their replacement with foreign CTL epitopes with similar physical properties (i.e., size and hydrophobicity) would result in less structural change and thereby help to conserve HBsAg conformation and subsequent stability/secretability. To test this supposition, we therefore evaluated whether pM29IPQ, pM215IPQ, and pM215GLS encoded structurally authentic VLPs. We did this by evaluating the capacity of pM29IPQ, pM215IPQ, and pM215GLS to package HDAg-L for secretion. HDV is an RNA agent which exists as a subviral satellite of HBV. Coexpression of HDAg-L and HBsAg leads to incorporation (packaging) of HDAg-L into the empty HBsAg-S VLPs and to secretion (3, 34, 46). We cotransfected DNAs encoding HDAg-L and the chimeric HBsAg polypeptide into HuH-7 cells and subsequently determined the presence of secreted HDAg-L in the cell culture supernatant by immunoblotting. HDAg-L was confirmed in the supernatant of cells which had been cotransfected with any one of the chimeric HBsAgs (Fig. 5a, lanes 1 to 3), although to a lesser extent than in cells cotransfected with wild-type HBsAg (Fig. 5a, lane 4). No HDAg-L was detected in the supernatant of cells transfected only with the plasmid encoding HDAg-L, although HDAg-L was clearly present in the cell pellet, confirming that in the absence of HBsAg, HDAg-L was not secreted (Fig. 5b, lane 6). This result indicates that chimeric HBsAg polypeptides in which the HBsAg-specific IPQ or GLS CTL epitopes are replaced by a foreign CTL epitope (M2) retain the capacity for assembly and secretion of structurally sound HBsAg particles.
DISCUSSION
A limited range of antigen delivery techniques are available to prime precursor CTLs to recombinant antigens. These include infection with recombinant replicating (bacterial or viral) vectors, the inoculation of antigen-encoding plasmid DNA or RNA, the association of protein antigen with selected adjuvants (e.g., squalene, Iscom), the injection of VLPs, and the injection of naked lipoprotein particles. HBsAg is the already licensed human vaccine for hepatitis B virus infection, making it an attractive vehicle for the delivery of epitopes from other pathogens and diseases, from both a regulatory and a scientific point of view. While HBsAg has been used experimentally as a carrier in recombinant vaccine studies, this primarily has been in the delivery of a variety of foreign B-cell epitopes for antibody induction, most commonly by the insertion of these epitopes into the external hydrophilic a loop (5, 23, 29, 33, 45). There are few reports of exploiting the inherent powerful immunogenicity of HBsAg to deliver foreign CTL (2, 15, 25) or T-helper (5) epitopes, and this likely relates to difficulties arising in stability and antigen processing that are incurred by the insertion of foreign sequences into hydrophobic regions of HBsAg, of which at least some are essential for correct conformation (13).
Comparison of wild-type HBsAg delivered as a DNA immunogen and that delivered as a VLP immunogen suggests that DNA is the preferred modality both economically and practically and in terms of immunogenic efficacy, at least for CTL induction (6). That a single intramuscular injection of HBsAg plasmid DNA gives CTL detectable for many months postvaccination (7, 9) suggests that continuous exposure to small doses of antigen provided from ongoing DNA transcription may be necessary to maintain immunity. Injected DNA encoding HBsAg DNA is likely to transfect a variety of cells, though only the transfection of professional antigen-presenting cells (APCs) leads to activation of the CTL response via the intracellular (endogenous) processing pathway (38). Particle formation within the APC is not an essential requirement for this to occur (38). Secretion of HBsAg particles (e.g., by DNA-transfected muscle cells) allows uptake of secreted particles by APCs for processing via the exogenous pathway (8), thus providing a second mechanism for HBsAg CTL generation (41). After a single intramuscular injection of HBsAg DNA, the antigen can be found in circulation for at least 1 month, suggesting sustained expression in muscle fibers (26). HBsAg is likely to persist elsewhere, e.g., in follicular dendritic cells, accounting for the sustenance of a prolonged immune response (17). The powerful CTL responses induced by HBsAg DNA vaccination may be explained by prolonged and higher expression, secretion for uptake of particles by APC, superior antigen processing, and the presence of multiple T-helper epitopes. Thus, DNA-based HBsAg immunization is extremely potent and may have an immune potential the same as that of natural infection (39). HBsAg-based DNA vaccination requires a dose as much as 2,500-fold lower than that used in previous clinical trials with conventional DNA (37). The fact that long-lasting CTL and antibody are efficiently induced in neonatal mice with wild-type HBsAg-encoding plasmid DNA (but not HBsAg particles) (41) argues in favor of a DNA vaccine approach for recombinant HBsAg vaccines.
In the present study, we sought to exploit the inherent immunogenicity of HBsAg DNA as a carrier for the delivery of disease-relevant foreign CTL epitopes. We argued that the simplistic approach of inserting CTL epitope-encoding sequences into the HBsAg gene at sites where restriction endonuclease sites fortuitously exist, not accounting for spatial requirements and the tertiary configuration of the HBsAg polypeptide, is likely to yield unpredictable outcomes, and this has hitherto constrained the exploitation of HBsAg for the delivery of inserted foreign CTL epitopes. We reasoned that by deleting endogenous CTL epitopes and replacing them with foreign CTL epitopes of similar size and physical characteristics (hydrophobicity), we might avoid deleterious effects on the structural integrity of the recombinant HBsAg protein. Furthermore, we reasoned that targeting foreign CTL epitopes to sites of proven HBsAg CTL epitopes might predispose to appropriate antigen processing of insert.
A number of HBsAg CTL epitopes in hydrophobic regions of the HBsAg molecule, restricted through murine (39, 42) and human (28) MHC-I molecules, have been described previously. We elected to delete two of the stronger of these endogenous CTL epitopes (IPQ and GLS [Table 2]) and replace them with protective foreign epitopes from the M2 protein of RSV and/or the E7 protein of HPV16.
We show that replacement of DNA encoding the endogenous IPQ epitope with DNA encoding either RSV M29 or M215 epitopes or the HPV oncoprotein RAH epitopes produced chimeric HBsAg DNA immunogens which elicited effector and memory CTL responses to the foreign CTL epitope (Fig. 1, 3, and 4). Moderate effector CTL responses measured by IFN- release (Fig. 1a and b) translated into strong memory responses when specific antigen was reencountered (Fig. 1f and g), as has been seen in other systems (47). We show that M2-specific CTL responses were associated with reduced proliferation of virus in the lungs of RSV-challenged mice and with amelioration in pulmonary pathology. It has been previously demonstrated that protection against RSV afforded by an RSV M2 CTL epitope vaccine is associated with CTLs which kill RSV-expressing cells in vitro (22). Similarly, RAH-specific CTL responses were associated with protection against challenge with an E7-expressing tumor. We further show HBsAg recombinant for both RSV and HPV CTL epitopes elicited simultaneous responses to both epitopes. However, there was evidence of immunodominance of the RSV M2 epitope. (Interestingly, immunodominance of the endogenous HBsAg-specific epitope at the site of the M2 insertion has also been reported [42].) While the memory CTL response to RAH was down-regulated in mice immunized with HBsAg encoding the M2 and RAH epitopes, the mice were still protected against tumor challenge (Fig. 4e). The phenomenon of relatively small CTL responses, as measured by in vitro cytotoxicity assay, translating into substantial tumor protection in vivo has been observed for other tumor vaccination systems in our laboratory (unpublished observations). Nonetheless, we have preliminary data to indicate that immunization with chimeric HBsAg DNA encoding three foreign epitopes elicits CTL responses to all three (not shown). Together, these data suggest the applicability of HBsAg as a vector for the delivery of protective CTL responses to human disease-relevant foreign epitopes and also suggest the possibility of simultaneous protection against multiple diseases with a single polyvalent HBsAg vaccine. The approach of delivering a limited number of MHC-I-restricted epitopes to target multiple diseases in humans by use of HBsAg DNA as a vector is feasible, since (i) at least 15 HBsAg CTL epitopes are candidates for replacement (28, 32, 48); (ii) in general, CTL-mediated pathogen and tumor protection is remarkably parsimonious in terms of the number of CTL epitopes to which the responses must be directed (19, 36); and (iii) a limited number of major MHC-I supertypes account for the majority of MHC polymorphism, allowing broad population coverage with very few disease epitopes (43). Furthermore, immunization with mixtures of chimeric HBsAg DNAs encoding different disease-relevant epitopes may be employed to enhance epitope delivery capacity.
CTL responses to foreign epitopes delivered in the context of chimeric HBsAg DNA compared favorably to those induced by immunization with a vast molar excess of epitope peptide plus adjuvant. This latter approach induces maximal (though short-lived) CTL responses and is a standard against which other CTL-inducing modalities may be compared. In view of the superior immunogenicity of HBsAg-based vaccines (see above) compared to that of conventional vaccine approaches, it would be interesting to compare the numbers of cognate CD8+ CTLs in the lungs of recombinant HBsAg-vaccinated mice and peptide-plus-adjuvant-vaccinated mice following RSV challenge. Additionally, we observed that responses to epitopes inserted into HBsAg could exceed the observed response to the deleted HBsAg epitope (following wild-type HBsAg immunization) which they replaced. These observations suggest that the choice of sites for inserting replacement foreign epitopes within the HBsAg molecule need not be overly influenced by the magnitude of the response to the endogenous HBsAg epitope situated at that site. Response is likely to be a function not only of primary amino acid sequence but also of the context of the epitope within the protein, such that intramolecular competition for cellular processing and HLA binding influences the magnitude of the response (35). It is also noteworthy that an HBsAg CTL epitope can be replaced with a foreign CTL epitope restricted through a different MHC-I haplotype to elicit strong insert-directed CTL responses.
In spite of our approach of replacing HBsAg-specific CTL epitopes with the M2 CTL epitope, the assembly and/or stability and secretion of recombinant HBsAg was reduced compared with those of wild-type particles (measured as reduced capacity to package hepatitis delta virus large antigen for secretion [Fig. 5]). The inference is that the replacement insertions interfere with the assembly or stability of structurally appropriate HBsAg particles. Length of insert is the major factor determining the level of secretion (11). However, it is noteworthy in our study that impaired secretion of functionally intact HBsAg particles occurred even when the replacement of the HBsAg-specific CTL epitope with foreign epitope resulted in no net change in the number of amino acid residues in the chimeric HBsAg compared to that in the wild type (pM29IPQ [Fig. 5]). This underscores the importance of specific sequence in topogenic regions essential for particle assembly (11, 13). In preliminary experiments (not shown), we found that antibody induction to a B mimotope of RSV (Table 1) inserted into the hydrophilic a determinant was lost when CTL epitopes were inserted at the IPQ and GLS sites (pM215IPQ.RAHGLS.HW1) (Table 2).
Immunization with plasmid HBsAg DNA evokes cytokines which predispose to a Th1-type response (26). Provision of T help for primary effector responses may be provided by HBsAg-specific T-helper epitopes (2, 30) or CpG dinucleotides in bacterial DNA (10). The recall of CTL-epitope specific memory (required when the pathogen or disease against which the chimeric HBsAg vaccine is directed is encountered) is likely to be T help independent (reviewed in reference 47).
In summary, using a strategy of deletion of DNA encoding HBsAg-specific CTL epitopes and replacement with DNA encoding foreign CTL epitopes, we have derived chimeric HBsAg DNA immunogens which elicited foreign epitope-directed effector and memory CTL responses associated with pathogen- and tumor-protective responses in vivo. We show that HBsAg recombinant for more than one CTL epitope elicited simultaneous responses to both epitopes. There are generic implications for the application of HBsAg, which is already licensed as human vaccine, as a vector for the delivery of multiple disease-relevant protective CTL responses. There are also specific implications for HBsAg-vectored vaccines for RSV- and HPV-associated tumors.
ACKNOWLEDGMENTS
W.-P. Woo and T. Doan contributed equally to the experimentation and should be regarded as joint first authors.
H.-J. Netter and R. W. Tindle contributed equally to the design and implementation of the study and should be regarded as joint senior authors.
We thank P. Young for providing RSV and advice and M. Mather and H. Guo for helpful discussions. D. West and her staff provided expert animal husbandry. W.-J. Liu advised with the molecular cloning. M. Duhig aided in the interpretation of lung histology.
The work was funded by grants from the Royal Children's Hospital Foundation and Queensland Cancer Fund.
Contribution no. 222 of the Sir Albert Sakzewski Virus Research Centre.
REFERENCES
Berger, J., J. Hauber, R. Hauber, R. Geiger, and B. R. Cullen. 1988. Secreted placental alkaline phosphatase: a powerful new quantitative indicator of gene expression in eukaryotic cells. Gene 66:1-10.
Bryder, K., H. Sbai, H. V. Nielsen, S. Corbet, C. Nielsen, R. G. Whalen, and A. Fomsgaard. 1999. Improved immunogenicity of HIV-1 epitopes in HBsAg chimeric DNA vaccine plasmids by structural mutations of HBsAg. DNA Cell Biol. 18:219-225.
Chang, F. L., P. J. Chen, S. J. Tu, C. J. Wang, and D. S. Chen. 1991. The large form of hepatitis delta antigen is crucial for assembly of hepatitis delta virus. Proc. Natl. Acad. Sci. USA 88:8490-8494.
Chargelegue, D., O. E. Obeid, S. C. Hsu, M. D. Shaw, A. N. Denbury, G. Taylor, and M. W. Steward. 1998. A peptide mimic of a protective epitope of respiratory syncytial virus selected from a combinatorial library induces virus-neutralizing antibodies and reduces viral load in vivo. J. Virol. 72:2040-2046.
Chengalvala, M. V., R. A. Bhat, B. M. Bhat, S. K. Vernon, and M. D. Lubeck. 1999. Enhanced immunogenicity of hepatitis B surface antigen by insertion of a helper T cell epitope from tetanus toxoid. Vaccine 17:1035-1041.
Davis, H. L., C. L. Brazolot Millan, M. Mancini, M. J. McCluskie, M. Hadchouel, L. Comanita, P. Tiollais, R. G. Whalen, and M. L. Michel. 1997. DNA-based immunization against hepatitis B surface antigen (HBsAg) in normal and HBsAg-transgenic mice. Vaccine 15:849-852.
Davis, H. L., M. Mancini, M. L. Michel, and R. G. Whalen. 1996. DNA-mediated immunization to hepatitis B surface antigen: longevity of primary response and effect of boost. Vaccine 14:910-915.
Davis, H. L., M. L. Michel, and R. G. Whalen. 1993. DNA-based immunization induces continuous secretion of hepatitis B surface antigen and high levels of circulating antibody. Hum. Mol. Genet. 2:1847-1851.
Davis, H. L., R. Schirmbeck, J. Reimann, and R. G. Whalen. 1995. DNA-mediated immunization in mice induces a potent MHC class I-restricted cytotoxic T lymphocyte response to the hepatitis B envelope protein. Hum. Gene Ther. 6:1447-1456.
Davis, H. L., R. Weeratna, T. J. Waldschmidt, L. Tygrett, J. Schorr, A. M. Krieg, and R. Weeranta. 1998. CpG DNA is a potent enhancer of specific immunity in mice immunized with recombinant hepatitis B surface antigen. J. Immunol. 160:870-876.
Delpeyroux, F., N. Chenciner, A. Lim, M. Lambert, Y. Malpiece, and R. E. Streeck. 1987. Insertions in the hepatitis B surface antigen. Effect on assembly and secretion of 22-nm particles from mammalian cells. J. Mol. Biol. 195:343-350.
Doan, T., M. Chambers, M. Street, G. J. Fernando, K. Herd, P. Lambert, and R. Tindle. 1998. Mice expressing the E7 oncogene of HPV16 in epithelium show central tolerance, and evidence of peripheral anergising tolerance, to E7-encoded cytotoxic T-lymphocyte epitopes. Virology 244:352-364.
Eble, B. E., D. R. MacRae, V. R. Lingappa, and D. Ganem. 1987. Multiple topogenic sequences determine the transmembrane orientation of the hepatitis B surface antigen. Mol. Cell. Biol. 7:3591-3601.
Feltkamp, M. C., H. L. Smits, M. P. Vierboom, R. P. Minnaar, B. M. de Jongh, J. W. Drijfhout, J. Ter Schegget, C. J. Melief, and W. M. Kast. 1993. Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type 16-transformed cells. Eur. J. Immunol. 23:2242-2249.
Fomsgaard, A., H. V. Nielsen, K. Bryder, C. Nielsen, R. Machuca, L. Bruun, J. Hansen, and S. Buus. 1998. Improved humoral and cellular immune responses against the gp120 V3 loop of HIV-1 following genetic immunization with a chimeric DNA vaccine encoding the V3 inserted into the hepatitis B surface antigen. Scand. J. Immunol. 47:289-295.
Graham, B. S., M. D. Perkins, P. F. Wright, and D. T. Karzon. 1988. Primary respiratory syncytial virus infection in mice. J. Med. Virol. 26:153-162.
Gray, D., M. Kosco, and B. Stockinger. 1991. Novel pathways of antigen presentation for the maintenance of memory. Int. Immunol. 3:141-148.
Gupta, R. K., and G. R. Siber. 1995. Adjuvants for human vaccines—current status, problems and future prospects. Vaccine 13:1263-1276.
Hsu, S. C., D. Chargelegue, O. E. Obeid, and M. W. Steward. 1999. Synergistic effect of immunization with a peptide cocktail inducing antibody, helper and cytotoxic T-cell responses on protection against respiratory syncytial virus. J. Gen. Virol. 80:1401-1405.
Hsu, S. C., D. M. Shaw, and M. W. Steward. 1995. The induction of respiratory syncytial virus-specific cytotoxic T-cell responses following immunization with a synthetic peptide containing a fusion peptide linked to a cytotoxic T lymphocyte epitope. Immunology 85:347-350.
Kruger, D. H., R. Ulrich, and W. H. Gerlich. 1999. Chimeric virus-like particles as vaccines. Biol. Chem. 380:275-276.
Kulkarni, A. B., M. Connors, C. Y. Firestone, H. C. Morse III, and B. R. Murphy. 1993. The cytolytic activity of pulmonary CD8+ lymphocytes, induced by infection with a vaccinia virus recombinant expressing the M2 protein of respiratory syncytial virus (RSV), correlates with resistance to RSV infection in mice. J. Virol. 67:1044-1049.
Laub, O., L. B. Rall, M. Truett, Y. Shaul, D. N. Standring, P. Valenzuela, and W. J. Rutter. 1983. Synthesis of hepatitis B surface antigen in mammalian cells: expression of the entire gene and the coding region. J. Virol. 48:271-280.
Le, T. T., D. Drane, J. Malliaros, J. C. Cox, L. Rothel, M. Pearse, T. Woodberry, J. Gardner, and A. Suhrbier. 2001. Cytotoxic T cell polyepitope vaccines delivered by ISCOMs. Vaccine 19:4669-4675.
Le Borgne, S., M. Mancini, R. Le Grand, M. Schleef, D. Dormont, P. Tiollais, Y. Riviere, and M. L. Michel. 1998. In vivo induction of specific cytotoxic T lymphocytes in mice and rhesus macaques immunized with DNA vector encoding an HIV epitope fused with hepatitis B surface antigen. Virology 240:304-315.
Leclerc, C., E. Deriaud, M. Rojas, and R. G. Whalen. 1997. The preferential induction of a Th1 immune response by DNA-based immunization is mediated by the immunostimulatory effect of plasmid DNA. Cell. Immunol. 179:97-106.
Lin, K. Y., F. G. Guarnieri, K. F. Staveley-O'Carroll, H. I. Levitsky, J. T. August, D. M. Pardoll, and T. C. Wu. 1996. Treatment of established tumors with a novel vaccine that enhances major histocompatibility class II presentation of tumor antigen. Cancer Res. 56:21-26.
Loirat, D., F. A. Lemonnier, and M. L. Michel. 2000. Multiepitopic HLA-A0201-restricted immune response against hepatitis B surface antigen after DNA-based immunization. J. Immunol. 165:4748-4755.
Michel, M. L., P. Pontisso, E. Sobczak, Y. Malpiece, R. E. Streeck, and P. Tiollais. 1984. Synthesis in animal cells of hepatitis B surface antigen particles carrying a receptor for polymerized human serum albumin. Proc. Natl. Acad. Sci. USA 81:7708-7712.
Milich, D. R. 1988. T- and B-cell recognition of hepatitis B viral antigens. Immunol. Today 9:380-386.
Nakabayashi, H., K. Taketa, K. Miyano, T. Yamane, and J. Sato. 1982. Growth of human hepatoma cell lines with differentiated functions in chemically defined medium. Cancer Res. 42:3858-3863.
Nayersina, R., P. Fowler, S. Guilhot, G. Missale, A. Cerny, H. J. Schlicht, A. Vitiello, R. Chesnut, J. L. Person, A. G. Redeker, et al. 1993. HLA A2 restricted cytotoxic T lymphocyte responses to multiple hepatitis B surface antigen epitopes during hepatitis B virus infection. J. Immunol. 150:4659-4671.
Netter, H. J., T. B. Macnaughton, W. P. Woo, R. Tindle, and E. J. Gowans. 2001. Antigenicity and immunogenicity of novel chimeric hepatitis B surface antigen particles with exposed hepatitis C virus epitopes. J. Virol. 75:2130-2141.
O'Malley, B., and D. Lazinski. 2002. A hepatitis B surface antigen mutant that lacks the antigenic loop region can self-assemble and interact with the large hepatitis delta antigen. J. Virol. 76:10060-10063.
Perkins, D. L., G. Berriz, T. Kamradt, J. A. Smith, and M. L. Gefter. 1991. Immunodominance: intramolecular competition between T cell epitopes. J. Immunol. 146:2137-2144.
Rosenberg, S. A. 2001. Progress in human tumor immunology and immunotherapy. Nature 411:380-384.
Roy, M. J., M. S. Wu, L. J. Barr, J. T. Fuller, L. G. Tussey, S. Speller, J. Culp, J. K. Burkholder, W. F. Swain, R. M. Dixon, G. Widera, R. Vessey, A. King, G. Ogg, A. Gallimore, J. R. Haynes, and F. D. Heydenburg. 2000. Induction of antigen-specific CD8+ T cells, T helper cells, and protective levels of antibody in humans by particle-mediated administration of a hepatitis B virus DNA vaccine. Vaccine 19:764-778.
Sbai, H., J. Schneider, A. V. Hill, and R. G. Whalen. 2002. Role of transfection in the priming of cytotoxic T-cells by DNA-mediated immunization. Vaccine 20:3137-3147.
Schirmbeck, R., W. Bohm, K. Ando, F. V. Chisari, and J. Reimann. 1995. Nucleic acid vaccination primes hepatitis B virus surface antigen-specific cytotoxic T lymphocytes in nonresponder mice. J. Virol. 69:5929-5934.
Schirmbeck, R., K. Melber, A. Kuhrober, Z. A. Janowicz, and J. Reimann. 1994. Immunization with soluble hepatitis B virus surface protein elicits murine H-2 class I-restricted CD8+ cytotoxic T lymphocyte responses in vivo. J. Immunol. 152:1110-1119.
Schirmbeck, R., and J. Reimann. 2002. Alternative processing of endogenous or exogenous antigens extends the immunogenic, H-2 class I-restricted peptide repertoire. Mol. Immunol. 39:249-259.
Schirmbeck, R., D. Stober, S. El Kholy, P. Riedl, and J. Reimann. 2002. The immunodominant, Ld-restricted T cell response to hepatitis B surface antigen (HBsAg) efficiently suppresses T cell priming to multiple Dd-, Kd-, and Kb-restricted HBsAg epitopes. J. Immunol. 168:6253-6262.
Sette, A., and J. Sidney. 1999. Nine major HLA class I supertypes account for the vast preponderance of HLA-A and -B polymorphism. Immunogenetics 50:201-212.
Vitiello, A., D. Marchesini, J. Furze, L. A. Sherman, and R. W. Chesnut. 1991. Analysis of the HLA-restricted influenza-specific cytotoxic T lymphocyte response in transgenic mice carrying a chimeric human-mouse class I major histocompatibility complex. J. Exp. Med. 173:1007-1015.
von Brunn, A., K. Fruh, H. M. Muller, H. W. Zentgraf, and H. Bujard. 1991. Epitopes of the human malaria parasite P. falciparum carried on the surface of HBsAg particles elicit an immune response against the parasite. Vaccine 9:477-484.
Wang, C. J., P. J. Chen, J. C. Wu, D. Patel, and D. S. Chen. 1991. Small-form hepatitis B surface antigen is sufficient to help in the assembly of hepatitis delta virus-like particles. J. Virol. 65:6630-6636.
Wherry, E. J., and R. Ahmed. 2004. Memory CD8 T-cell differentiation during viral infection. J. Virol. 78:5535-5545.
Wild, J., M. J. Grusby, R. Schirmbeck, and J. Reimann. 1999. Priming MHC-I-restricted cytotoxic T lymphocyte responses to exogenous hepatitis B surface antigen is CD4+ T cell dependent. J. Immunol. 163:1880-1887.(Wai-Ping Woo, Tracy Doan,)