Immunization with the Gene Expressing Woodchuck He
http://www.100md.com
病菌学杂志 2005年第10期
Institut für Virologie
Zentrales Tierlaboratorium, Universit?tsklinkum Essen, Essen, Germany
ABSTRACT
A number of options are available to modify and improve DNA vaccines. An interesting approach to improve DNA vaccines is to fuse bioactive domains, like cytotoxic-T-lymphocyte-associated protein 4 (CTLA-4), to an antigen. Such fusion antigens are expressed in vivo and directed to immune cells by the specific bioactive domain and therefore possess great potential to induce and modulate antigen-specific immune responses. In the present study, we tested this new approach for immunomodulation against hepadnavirus infection in the woodchuck model. Plasmids expressing the nucleocapsid protein (WHcAg) and e antigen (WHeAg) of woodchuck hepatitis virus (WHV) alone or in fusion to the extracellular domain of woodchuck CTLA-4 and CD28 were constructed. Immunizations of mice with plasmids expressing WHcAg or WHeAg led to a specific immunoglobulin G2a (IgG2a)-dominant antibody response. In contrast, fusions of WHcAg to CTLA-4 and CD28 induced a specific antibody response with comparable levels of IgG1 and IgG2a. Furthermore, the specific IgG1 response to WHcAg/WHeAg developed immediately after a single immunization with the CTLA-4-WHcAg fusion. Woodchucks were immunized with plasmids expressing WHeAg or the CTLA-4-WHcAg fusion and subsequently challenged with WHV. CTLA-4-WHcAg showed an improved efficacy in induction of protective immune responses to WHV. In particular, the anti-WHsAg antibody response developed earlier after challenge in woodchucks that received immunizations with CTLA-4-WHcAg, consistent with the hypothesis that anti-WHs response is dependent on a Th cell response to WHcAg. In conclusion, the use of fusion genes represents a generally applicable strategy to improve DNA vaccination.
INTRODUCTION
DNA vaccination is a useful technique to induce potent antigen-specific immune responses and possesses great potential for modification and improvement (2, 8, 39, 42). In general, the application of plasmid DNA by intramuscular injections induces dominantly cell-mediated immune responses, termed Th1-type responses. A number of options are available to modify DNA vaccines, e.g., improvement of antigen expression, coadministration of cytokines, or choice of application routes. An interesting approach is to fuse a bioactive domain, like cytotoxic-T-lymphocyte-associated protein 4 (CTLA-4), to an antigen (1). Such fusion antigens have been shown to bind to antigen-presenting cells expressing B7 molecules on their surfaces (1, 7, 32, 44). Immunization with plasmids expressing fusion proteins with CTLA-4 leads to antigen-specific immune responses, preferentially with an enhanced humoral branch (termed Th2-type responses), though DNA vaccination via the intramuscular route usually induces Th1-type immune responses (1). Thus, the fusion of a specific antigen to CTLA-4 provides a simple but effective modification of DNA vaccines if balanced Th1 and Th2 immune responses are desirable. Such fusion genes showed the ability to induce and modulate antigen-specific immune responses and improved protection against challenges with the respective pathogens (3, 7, 9). For example, the use of a fusion of CTLA-4 and influenza hemagglutinin for DNA vaccination enhanced the antigen-specific immune response and conferred better protection against influenza virus challenge in mice (7).
Hepatitis B virus (HBV) infection is still one of the major infectious diseases worldwide and results in severe liver diseases, cirrhosis, and hepatocellular carcinoma (17). HBV infection in humans can be effectively controlled by vaccination with recombinant HBsAg (43). Immunizations with plasmids expressing the HBV surface antigens (HBsAg) and nucleocapsid protein (HBcAg) effectively induced specific antibody and cytotoxic-T-cell responses to the respective antigens in the mouse model (21, 29, 40; reviewed in reference 6). However, DNA immunizations in large animals like chimpanzees were not efficient, as plasmids expressing HBsAg had to be applied in a scale of milligrams to induce a measurable anti-HBs antibody response (6, 35). Thus, DNA vaccines against HBV need significant improvements.
The woodchuck (Marmota monax) model is an informative animal model used to perform vaccination trials against HBV infection (4, 11, 12, 15, 16, 23-28, 36, 37). Woodchuck hepatitis virus (WHV) causes acute self-limiting and chronic infection, like HBV in humans (reviewed in references 25, 36, and 37). The humoral and cellular immune responses to woodchuck hepatitis surface antigen (WHsAg) and core antigen (WHcAg) in acute and chronic WHV infection are similar to HBV infection. DNA vaccinations of woodchucks with plasmids expressing WHsAg and WHcAg were able to protect against subsequent challenge with WHV (23, 41). However, the induction of a measurable immune response to WHV proteins by current protocols of DNA vaccination is rather inefficient (23). In particular, antibody responses to viral antigens could only be stimulated at low levels, presumably due to the fact that DNA vaccination by the intramuscular route preferentially primes immune responses of Th1 type (12, 41).
In the present study, a new approach using fusion proteins to CTLA-4 and CD28 was tested to overcome this shortcoming. The woodchuck CD28 (wCD28) and CTLA-4 (wCTLA-4) proteins are highly similar to their mammalian counterparts and share conserved common structural features (45). A hexapeptide motif, MYPPPY, that has been shown to be important for binding to the CD80 and CD86 ligands is conserved in both wCD28 and wCTLA-4. The binding motifs for the p85 subunit of phosphoinositide 3-kinase, YMNMTPR of CD28 (amino acids [aa] 192 to 197) and YVKMPP (aa 201 to 206), are also conserved in the deduced amino acid sequences of the corresponding woodchuck proteins (14, 18). CTLA-4 has been shown to be able to interact with heterologous CD80 and CD86 proteins of other species (33). This circumstance enables us to test the abilities of our fusion proteins with wCTLA-4 and wCD28 directly in both mouse and woodchuck models. Plasmids expressing WHV proteins alone or in fusion to the extracellular domain of woodchuck CTLA-4 and CD28 were constructed. The abilities of these plasmids to induce a specific immune response to WHcAg or WHeAg were tested by immunization of mice. While immunizations with plasmids expressing WHeAg or WHcAg alone led to a specific antibody response with immunoglobulin G2a (IgG2a) as the dominant IgG subtype, fusions of WHcAg to wCTLA-4 induced rapid and enhanced antibody responses with both IgG1 and IgG2a at comparable levels. Furthermore, woodchucks were immunized with plasmids expressing WHeAg or a CTLA-4-WHcAg fusion and subsequently challenged with WHV. Immunizations with the fusion gene induced a stronger antibody response to WHcAg that was further stimulated upon challenge with WHV and resulted in protection of immunized woodchucks against WHV infection, in contrast to immunizations with plasmids expressing WHeAg. The results of these experiments indicate that the new approach significantly improves DNA vaccination in the woodchuck model.
MATERIALS AND METHODS
Woodchucks. Adult WHV-negative woodchucks trapped in the state of New York were purchased from North Eastern Wildlife (Ithaca, N.Y.). Previous exposure of these woodchucks to WHV was excluded by testing for anti-WHcAg and anti-WHsAg antibodies, and for WHsAg as well.
Construction of plasmids for DNA vaccination. A plasmid, pWHcIm, had been constructed in a previous work and used for DNA vaccination (23, 41). The preC/C region of WHV8 was amplified by PCR using primers WHpreC EV1 and WHc EV2 (Table 1). The PCR products were cloned into pcDNA3.1 vectors (Invitrogen, Karlsruhe, Germany) according to the manufacturer's instructions. A clone containing the PCR product was selected after verification of the correct nucleotide sequence of the cloned fragment by sequencing. This plasmid, designated pWHeIm, contains the WHV preC/C region under the control of the cytomegalovirus (CMV) immediate-early promoter (Fig. 1).
The cDNA fragments comprising the coding regions of wCD28 and wCTLA-4 were amplified by reverse transcription-PCR, cloned into the pCRII vector (Invitrogen), and characterized in a previous work (45). The cloned cDNA fragments of wCTLA-4 were isolated by digestion with HindIII and XhoI and inserted into the corresponding site of the pcDNA3 vector predigested with HindIII and XhoI (Invitrogen) to generate peCTLA-4. Transfection with peCTLA-4 led to the expression of wCTLA-4 in mammalian cells, as shown in a previous work (45). The coding regions for the extracellular domains of wCD28 and wCTLA-4 were amplified with the primer pairs CD28H-CD28EV and CTLA-4H-CTLA-4EV (Table 1), respectively. The PCR fragments were cloned into pCRII vectors, sequenced to exclude nucleotide misincorporation by PCR, and recloned into pcDNA3 by restriction with HindIII and XhoI. This procedure generated peWCD28E and peWCTLA-4E. These plasmids were digested with EcoRV and XhoI, and a fragment comprising the coding region of WHcAg amplified with primers WHc EV1 and WHc EV2 was inserted, resulting in plasmids pCD28-C and pCTLA-4-C (Table 1 and Fig. 1). pCTLA-4-C was modified by digestion with StuI and MluNI and religation, resulting in a small deletion of 93 base pairs between nt 24 and 117 within the coding region of wCTLA-4. This plasmid, pCTLA-4-C, encodes a CTLA-4-WHcAg fusion protein with a deletion of 31 amino acid residues of wCTLA-4. Since the deletion eliminated the signal peptide of wCTLA-4, the fusion protein was expected to be deficient for secretion. All open reading frames listed here were placed under the control of the CMV immediate-early promoter of pcDNA3.
For DNA immunizations, plasmids were prepared with the Giga plasmid purification kit (QIAGEN, Hilden, Germany). Plasmids were dissolved in phosphate-buffered saline in a concentration of 1 mg per ml.
In vitro translation. The proteins encoded by the expression vectors constructed in this study were visualized by in vitro translation. The plasmids were cut by XhoI, precipitated with ethanol, and subjected to in vitro translation with a TNT-coupled translation kit (Promega, Mannheim, Germany). The proteins were labeled by [35S]methionine and subjected to sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis. The bands of labeled proteins were detected by the exposure of the dried SDS-polyacrylamide gel to an X-ray autograph.
Transient expression and detection of WHcAg or recombinant fusion proteins in transiently transfected woodchuck cells. A woodchuck cell line, WH12/6, was used for transfection experiments. Transient transfection was performed by using Lipofectamine (Gibco BRL, Neu-Isenburg, Germany) as described previously (23). Four micrograms of plasmids was incubated with 10 μg of Lipofectamine in 100 μl of Opti-Mem for 45 min and was given to cells in 1 ml of Opti-Mem (Gibco BRL) for 6 h at 37°C, 5% CO2. Transfected cells were maintained for 48 h at 37°C, 5% CO2, and fixed with 50% methanol. The expressed WHcAg and recombinant fusion proteins were detected by indirect immunofluorescence (IF) staining using rabbit antisera to WHcAg. The expressed wCTLA-4 was detected by indirect IF staining using goat antisera to a conserved region, aa 205 to 223, of mouse CTLA-4 (Santa Cruz Biotechnology, The Netherlands).
Immunization of mice and woodchucks by intramuscular injection of plasmid DNA. Immunizations of mice and woodchucks were performed by the procedures described by Lu et al. (23). Briefly, BALB/cJ (H-2d) mice were kept under standard-pathogen-free conditions in the Central Animal Laboratory of the University of Essen. Mice 6 to 8 weeks of age were pretreated by intramuscular injection of 50 μl of cardiotoxin (10 μM) into the tibialis anterior muscle. After 1 week, 50 μg of plasmids at a concentration of 1 mg per ml was injected into the same muscle of both legs. Five mice per group were used for each plasmid. The plasmid injection was repeated twice at intervals of 3 weeks. The mice were sacrificed 3 weeks after the last immunization. A modified protocol was used for woodchucks. A week prior to the injection of plasmids, 500 μl of cardiotoxin (10 μM in phosphate-buffered saline) was injected into the tibialis cranialis muscle of the woodchucks. The woodchucks were vaccinated three times by intramuscular injection of 500 μl of plasmids at a concentration of 1 mg per ml into each tibialis cranialis muscle on both sides at 3-week intervals. Two woodchucks were used for each plasmid. Seven weeks after the last vaccination, the woodchucks were challenged by intravenous injection of an inoculum containing 106 WHV genome equivalents.
Serology and detection of WHV DNA. Antibodies to WHcAg or to WHeAg (anti-WHcAg/WHeAg) were detected by a new version of an enzyme-linked immunosorbent assay (ELISA). Recombinant WHcAg particles were produced in Escherichia coli and purified by a combined protocol with precipitation and 30% saturation of ammonium sulfate and chromatographic separation though a Superose 6 column. The microtiter plate was coated with 10 μg per ml of purified WHcAg. After being blocked with 5% fetal calf serum, 100 μl of mouse serum at an appropriate dilution (1:10 to 1:1,000) was added and incubated for 1 h at 37°C. The bound mouse total IgG, IgG1, or IgG2a was detected with appropriate secondary antibodies, anti-mouse IgG, anti-mouse IgG1, or anti-mouse IgG2a, labeled with horseradish peroxidase (DB Biosciences, CA) at a dilution of 1:1,000. The development of color occurred at room temperature and was read at 490 nm. The cutoff value was set as three times over negative controls. The titers of antibodies to WHcAg or WHeAg in serum samples were calculated by extrapolation of ELISA values of serially diluted samples and corresponded to the reciprocal values of the highest dilutions that were regarded as positive. This ELISA format did not strictly differentiate antibodies to WHcAg and WHeAg, since the preparation of recombinant WHcAg obviously contained a fraction of polypeptide exposing different epitopes. However, it allowed the differentiation of subtypes of WHcAg/WHeAg-specific antibodies.
The detection of anti-WHcAg/WHeAg in woodchuck sera was done as follows: the microtiter plate was coated with 10 μg per ml of purified WHcAg; after being blocked with 5% fetal calf serum, 100 μl of woodchuck serum (dilution, 1:10) was added and incubated for 1 h at 37°C, and woodchuck IgGs were detected with protein A labeled with horseradish peroxidase. Anti-WHsAg antibody in woodchuck sera was detected by ELISA as described previously (23).
The dot blot technique was routinely performed to detect WHV DNA in woodchuck sera. For PCR detection of WHV DNA in woodchuck sera, nucleic acids were extracted with a serum blood kit (QIAGEN). PCR for amplification of the WHV core gene was run with primers wc1 and wc2 as described previously (23). WHV DNAs in woodchuck sera were quantified by real-time PCR with a light cycler DNA Master SYBR Green kit (Roche). The primers used for the PCR were designed according to the method of Girones et al. (13): wc1 (5' TGG GGC CAT GGA CAT AGA TCC TTA 3' [sequences containing the restriction site are given in italic letters]; nt 2015 to 2038) and wc-149s (5' AAG ATC TCT AAA TGA CTG TAT GTT CCG 3'; nt 2467 to 2451). The reactions were run in a light cycler (Roche) at 95°C for 0 s, 53°C for 10 s, and 72°C for 12 s. A plasmid containing a full-length WHV genome was diluted and served as the standard. The detection limit of this assay was 103 WHV genome equivalents per reaction.
Measurement of WHV antigen-specific proliferation of mouse splenocytes and woodchuck peripheral blood mononuclear cells (PBMCs). Mouse splenocytes were prepared in RPMI 1680 medium, seeded in triplicates of 2 x 104 per well, and cultured in flat-bottom 96-well microtiter plates (Falcon; Becton Dickinson, N.J.) at 37°C in a humidified atmosphere containing 5% CO2. The proliferation of mouse splenocytes in response to WHcAg- or WHcAg-derived peptides was measured at a concentration of 1 μg of antigen or peptide per ml. After a 5-day incubation, the cells were labeled with 1 μCi of [2-3H]thymidine (Amersham, Braunschweig, Germany) for 20 h and collected by a cell harvester (Packard Instrument Company, Calif.). The results for triplicate cultures are presented as the mean stimulation index (SI; the mean total thymidine incorporation in stimulated mouse splenocytes divided by the mean total thymidine incorporation in the control). The standard deviations of the means were less than 30% of the mean.
Antigen-specific proliferation of woodchuck PBMCs was determined by a [2-3H]adenine assay described previously (26, 27). SIs greater than 2.1 were considered positive for lymphoproliferative responses.
RESULTS
Construction of DNA vaccines. Six plasmids used for vaccination, pWHcIm, pWHeIm, pCD28-C, pCTLA-4-C, pCTLA-4-C, and peCTLA4, are depicted in Fig. 1A. pWHcIm and pWHeIm express WHcAg and WHeAg, respectively. In pCD28-C and pCTLA-4-C, the coding region of WHcAg was fused to the 3' ends of the coding sequences for the extracellular domains of wCD28 (the N-terminal 153 aa) and wCTLA-4 (the N-terminal 161 aa), respectively. Both CD28 and CTLA-4 have a signal peptide and are cell surface proteins on the cytoplasmic membrane. In the fusion proteins, the transmembrane regions of CD28 and CTLA-4 were removed, resulting in proteins without an anchoring sequence. For the control plasmid pCTLA-4-C, a small deletion of 93 base pairs was introduced between nt 24 and 117 of the wCTLA-4 coding region, leading to the destruction of the signal peptide of the wCTLA-4-WHcAg fusion protein.
The proteins encoded by these constructs were examined by in vitro translation. The in vitro-translated products from pWHcIm and pWHeIm had molecular masses of about 20 and 22 kDa, corresponding to WHcAg and WHeAg, respectively (Fig. 1B). The fusion proteins consisting of WHcAg with the extracellular domains of wCD28 and wCTLA4 were detected as proteins at a molecular mass of about 38 kDa. The in vitro-translated product of pCTLA-4-C was slightly smaller than the fusion protein wCTLA-4-WHcAg due to the deletion of 31 aa residues within wCTLA-4.
The expression of the WHV proteins and fusion proteins was further examined by transient transfection in woodchuck 12/6 cells and by IF staining. Cells transfected with pWHcIm, pWHeIm, pCD28-C, pCTLA-4-C, or pCTLA-4-C were positively stained with anti-WHcAg antibodies (23). Figure 1C shows the IF staining of cells transiently transfected with pWHeIm and pCTLA-4-C. The expression of wCTLA-4 in cells transfected with peCTLA4 was shown by IF staining with anti-CTLA-4 antibodies (45).
Immune responses to WHcAg and WHeAg induced in mice by immunization with plasmids expressing WHV proteins and fusion proteins. The abilities of the constructed plasmids to induce a WHcAg/WHeAg-specific antibody response was tested by intramuscular DNA injection in mice as described previously (23). The antibody response induced by plasmid immunizations was monitored by measuring the serum titers of specific antibodies to WHcAg and WHeAg. Immunization with pWHcIm, pWHeIm, pCTLA-4-C, and pCD28-C led to the production of antibodies to WHcAg and WHeAg in mice, while pCTLA-4-C and peCTLA4 were not able to induce any anti-WHcAg/WHeAg antibody (Fig. 2A). For pCTLA-4-C, the deletion within the leader peptide of wCTLA-4 appeared to prevent the wCTLA-4-WHcAg fusion protein from folding into a proper conformation. Immunizations with pCTLA-4-C resulted in the highest titer of the total anti-WHcAg/WHeAg IgG, which was about twice as high as the anti-WHcAg/WHeAg IgG titers induced by pWHeIm and pCD28-C.
The specific lymphoproliferative response to WHcAg in mice immunized with pWHeIm and pCTLA-4-C was monitored. Splenocytes were prepared from mice that had received three immunizations and were cultured with 24 overlapping peptides covering the whole WHcAg at a concentration of 1 μg per ml. The cells were then labeled with [2-3H]thymidine and harvested. Splenocytes from immunized mice showed proliferative responses to various peptides, particularly peptides comprising aa 57 to 93 of WHcAg (Fig. 2B). The SI was up to 17. Splenocytes from mice without immunizations did not respond to stimulation with peptides (data not shown).
Subtypes of anti-WHcAg/WHeAg IgG and their kinetics induced in mice by plasmid immunizations. The subtypes of anti-WHcAg/WHeAg IgG in mice immunized with pWHcIm, pWHeIm, pCTLA-4-C, and pCD28-C were determined by ELISA using mouse IgG1- and IgG2a-specific secondary antibodies (Fig. 3A). The immunizations with pWHcIm led exclusively to the production of IgG2a to WHcAg/WHeAg. After three immunizations with pWHeIm, anti-WHcAg/WHeAg antibodies were detected in mice at titers of 1:151 and 1:686 for IgG1 and IgG2a, respectively. The relative dominance of the antibody subtype IgG2a over IgG1 against WHcAg/WHeAg indicates that pWHcIm and pWHeIm induced a Th1-dominant immune response. Both pCTLA-4-C and pCD28-C induced IgG2a and IgG1 antibodies to WHcAg/WHeAg at comparable levels. The ratios of IgG1 to IgG2a were 0, 0.22, 0.98, and 0.84 for pWHcIm, pWHeIm, pCTLA-4-C, and pCD28-C, respectively (Fig. 3B). Thus, immunizations with the fusion constructs with CTLA-4 and CD28 led to an enhancement of Th2-type responses, while Th1-type responses were not affected.
Figure 4 shows the kinetics of the induction of anti-WHcAg/WHeAg antibodies in mice immunized with pWHcIm and pCTLA-4-C. Mice were immunized at weeks 0, 3, and 6. Sera were taken from the immunized mice at weeks 3, 6, and 9 and tested for WHcAg/WHeAg-specific IgG2a and IgG1. The immunizations with pWHcIm led to the production of WHcAg/WHeAg-specific IgG2a. The titers of WHcAg/WHeAg-specific IgG2a increased with each immunization and reached over 1:1,000 on average after three immunizations. In contrast, no WHcAg/WHeAg-specific IgG1 was detected in these sera at a dilution of 1:20. Mice receiving only one immunization with pCTLA-4-C developed both WHcAg/WHeAg-specific IgG2a and IgG1 at titers of 1:382 and 1:962, respectively. The titers of WHAg/WHeAg-specific IgG2a in pCTLA-4-C-immunized mice reached 1:612 and 1:850 at weeks 6 and 9, while the titers of WHcAg/WHeAg-specific IgG1 were 1:1,356 and 1:985 at the same times. It appears that the induction of the specific IgG1 occurred after a single immunization at a high level and was slightly enhanced by further boosts. However, the production of the specific IgG2a was strongly dependent on boosting and was improved markedly after three immunizations.
The kinetics of induction of anti-WHcAg/WHeAg in mice by immunization with pWHeIm were similar to those with pWHcIm. However, pCD28-C did not induce a significant IgG1 antibody response after a single immunization (data not shown).
Immunization of woodchucks with pWHeIm and pCTLA-4-C. We showed that both pWHeIm and pCTLA-4-C were able to induce anti-WHcAg/WHeAg responses, but with different qualities in mice, though both WHeAg and the fusion protein CTLA-4-WHcAg encoded by vectors pWHeIm and pCTLA-4-C possess signal peptides. While pWHeIm induced a Th1-dominant immune response, immunization with pCTLA-4-C primed a balanced Th1-Th2 response to WHcAg/WHeAg in mice. Thus, immunizations and WHV challenge in the woodchuck model were carried out to clarify whether the induction of a balanced Th1-Th2 response to WHcAg/WHeAg is beneficial for protection against viral challenge.
Three groups of two woodchucks each received immunizations with the control plasmid peCTLA-4, pWHeIm, or pCTLA-4-C. Three immunizations were carried out at intervals of 3 weeks. Seven weeks after the last immunization, the immunized woodchucks were challenged with a stock containing 106 WHV genome equivalents and then monitored for up to 12 weeks. The anti-WHcAg/WHeAg antibody response was determined for the entire period. The anti-WHsAg antibodies and serum WHV DNA concentrations in the woodchucks were measured after challenge.
Two control woodchucks, WH12033 and WH12304, were immunized three times with peCTLA-4 and did not show any response to WHV proteins (Fig. 5). After the challenge, both woodchucks developed typical signs of acute WHV infection. WH12033 was positive for WHV DNA after challenge and had a peak level of WHV DNA at 3.65 x 1010 genome equivalents per ml at week 4 postinfection (p.i.). The WHV DNA concentration decreased rapidly in the following weeks but remained detectable until week 8 p.i. Antibodies to WHcAg/WHeAg and to WHsAg became detectable at weeks 3 and 5 p.i., respectively. The WHV DNA was at a high level, up to 6.9 x 1010 genome equivalents per ml, in WH12304 from weeks 4 to 8 p.i. Anti-WHcAg/WHeAg antibodies developed at week 6 p.i. Anti-WHsAg antibodies did not become detectable until week 8 p.i.
Two woodchucks, WH12027 and WH12034, were immunized three times with 1 mg of pWHeIm per injection. A weak anti-WHcAg/WHeAg antibody response was measured after three immunizations (Fig. 5). One woodchuck, WH12027, was completely protected against challenge with WHV, as no WHV DNA was detected by PCR. The anti-WHsAg antibody response developed at week 6 p.i. Woodchuck WH12034 was infected and showed a level of viremia of about 1.6 x 109 copies per ml for a short period during weeks 3 to 4 p.i. The WHV DNA remained detectable at a low level of 106 copies per ml until week 6. The anti-WHcAg/WHeAg antibody titer increased significantly at week 3 p.i. WH12034 became positive for anti-WHsAg at week 6 p.i.
Three immunizations with pCTLA-4-C led to the development of anti-WHcAg/WHeAg antibodies in two woodchucks, WH12035 and WH12293. In particular, anti-WHcAg/WHeAg antibodies became detectable in WH12035 after only a single immunization with pCTLA-4-C. The antibody titers increased after boosts in both woodchucks. The challenge with WHV resulted in a rapid increase of the WHcAg/WHeAg-specific antibody response that was not observed in woodchucks immunized with peCTLA-4 or pWHeIm. Both woodchucks did not develop detectable viremia. Woodchucks WH12035 and WH12293 developed anti-WHsAg antibodies in weeks 4 and 3 p.i. This unique serological profile in WH12035 and WH12293 indicated that the immunization with pCTLA-4-C did effectively change the quality of the immune response to WHV proteins in both woodchucks. Thus, pCTLA-4-C was able to induce an enhanced antibody response to WHV proteins, probably due to its ability to prime an additional antigen-specific Th2 response.
Lymphoproliferative responses to WHV proteins in woodchucks. The lymphoproliferative responses to WHcAg and WHcAg-derived peptides were measured during the period of immunization and WHV challenge. Immunizations with pWHeIm and pCTLA-4-C induced low lymphoproliferative responses to WHcAg and peptides in woodchucks (Fig. 6A). For example, the woodchuck WH12027 showed significant lymphoproliferative responses to WHcAg and four different WHcAg-derived peptides (Fig. 6B). These results were consistent with previous findings that DNA vaccination is able to prime specific lymphoproliferative responses to WHV proteins. No WHcAg-specific lymphoproliferative response was detected in two control woodchucks, WH12033 and WH12304, before WHV challenge. After challenge, PBMCs of WH12033 responded either to WHcAg and WHsAg at week 5 p.i., concomitantly with the peak viremia (Fig. 5 and Fig. 6A). For WH12304, no specific lymphoproliferative response to WHV was detected during the whole experiment. These results were consistent with previous findings.
DISCUSSION
DNA immunizations with plasmids expressing WHcAg and WHeAg by intramuscular injection primed a Th1-type-dominant immune response to the respective viral antigens in mice, as is typical for this immunization procedure. We showed here that the fusion gene of WHcAg to the extracellular domain of wCTLA-4 or wCD28 induced WHcAg/WHeAg-specific immune responses that were significantly different in quality and kinetics in the mouse model. Strikingly, a single injection of plasmids expressing wCTLA-4-WHcAg led to the production of WHcAg/WHeAg-specific antibodies of subtype IgG1. These results indicate that the fusion of WHcAg to wCTLA-4 or wCD28 dramatically changes the manner of presentation of viral antigens. It has been shown that fusion proteins of influenza hemagglutinin or human immunodeficiency virus gp120 with CTLA-4 were able to interact with antigen-presenting cells expressing B7 molecules (1, 7, 32, 44). It appears that the exogenic antigen presentation pathway was enhanced and therefore Th2-type responses were primed, in addition to the Th1-type response. It should be clarified by further investigation whether the fusion protein wCTLA-4-WHcAg is indeed taken up by antigen-presenting cells with accelerated kinetics, leading to enhanced presentation for Th cells.
Both pCD28-C and pCTLA-4-C induced a mixed IgG1-IgG2a response to WHcAg/WHeAg in mice. However, the titers of anti-WHcAg/WHeAg in pCD28-C-immunized mice were lower than those in pCTLA-4-C-immunized mice. In addition, only pCTLA-4-C showed the ability to prime the IgG1 response after a single immunization. The CTLA-4 molecule has a 1,000-fold-higher affinity for the B7 molecule than CD28 (22). This could be interpreted as the fusion protein of WHcAg to the extracellular domains of wCTLA-4 being more efficiently directed to antigen-presenting cells. It will be interesting to investigate whether the affinity of the ligand to its receptor determines the antigen-specific immune response primed by fusion proteins.
One interesting issue is whether DNA vaccinations with the fusion genes induce specific immune responses to the domain of the cellular protein. We attempted to detect wCTLA-4-specific antibodies in sera from immunized mice and woodchucks by IF staining of transiently transfected cells; however, no specific staining of wCTLA-4 was observed. Further examination with more sensitive methods will be needed to clarify the possibility of induction of a specific immune response against wCTLA-4 in immunized animals.
The immunizations with pWHeIm led to the production of anti-WHcAg/WHeAg antibodies at only a very low level in woodchucks. After viral challenge, woodchucks immunized with pWHeIm showed development of strong anti-WHcAg/WHeAg antibody response within 3 weeks. Consistently, previous experiments demonstrated that antibody responses to WHcAg or WHsAg were primed by DNA immunization and greatly stimulated by the viral challenge, even though viral replication in immunized individuals was very limited (23). The immunizations with pCTLA-4-C induced a stronger anti-WHcAg/WHeAg antibody response in woodchucks than that with pWHeIm. An immediate increase of anti-WHcAg/WHeAg was observed in two woodchucks, WH12035 and WH12239, after WHV challenge. These results demonstrated the superiority of pCTLA-4-C over pWHeIm in terms of induction of specific antibody responses.
It is highly interesting that anti-WHsAg antibodies became detectable as early as 3 weeks p.i. in woodchucks immunized with pCTLA-4-C. The early appearance of anti-WHsAg in pCTLA-4-C-immunized woodchucks may have different reasons. Since the development of anti-WHsAg during an acute WHV infection is associated with viral clearance, an efficient protection against WHV challenge by primed immune responses may result in the rapid appearance of anti-WHsAg. In addition, a strong Th response to WHcAg, particularly the enhanced Th2 response, may provide help for the development of anti-WHsAg. Milich et al. pointed out that the HBcAg-specific Th response is critical for anti-HBsAg antibody production (30). Thus, further work should be done to elucidate the relationship between WHcAg-specific Th responses and anti-WHsAg responses in woodchucks.
Inoculation of na?ve woodchucks with 106 WHV genome equivalents leads to acute infection with transiently high serum titers of WHV DNA at a level of 1010 copies per ml. Immunizations with pWHeIm led to protection against WHV challenge in one woodchuck, WH12027. Another woodchuck, WH12034, became viremic. Compared with the control woodchucks, the immunizations with pWHeIm appeared to partially limit viral replication. The peak level of WHV in WH12034 reached 108 WHV genome equivalents per ml, a viremic level that was significantly lower than that in the controls. WHV DNA was not detected in two woodchucks immunized with pCTLA-4-C by dot blot hybridization and by PCR. These results are consistent with results from our previous studies and indicate that immunizations with WHcAg may at least partially control WHV infection (41).
It is a generally applicable concept to modify a given viral antigen by fusion to a cellular protein (10, 19, 20, 31, 38). By fusion to a cellular protein, viral antigens will be engaged in protein-protein interactions that do not exist naturally. Thus, such artificial interactions lead to processing and presentation of viral antigens in an unusual context and may be beneficial in some instances. For example, the specific T-cell responses to HBV proteins are impaired in chronically HBV-infected patients. Although the HBV proteins are expressed at high levels in the patients, the host immune system is not able to mount an efficient response. It was proposed that therapeutic immunizations may stimulate the specific immune response in these patients. However, immunizations with purified viral surface antigens failed to induce an immune response that suppressed HBV replication in the majority of patients (5, 34). Therefore, modifications of viral antigens will be needed to enable efficient antigen presentation and induction of a sustained HBV-specific immune response. Thus, a regular antigen presentation pathway may be unable to stimulate the specific immune responses to viral antigens. Modifications of viral antigens, as was done in the present study, may circumvent the block by viral mechanisms and lead to efficient T- and B-cell responses.
ACKNOWLEDGMENTS
We are thankful to Michael Roggendorf for continuous support and helpful discussion. We are grateful to Thekla Kemper and Barbara Bleekmann for excellent technical assistance and to Ulf Dittmer for critical reading and helpful suggestions.
This work is partially supported by BMBF grant 01GE9909.
REFERENCES
Boyle, J. S., J. L. Brady, and A. M. Lew. 1998. Enhanced responses to a DNA vaccine encoding a fusion antigen that is directed to sites of immune induction. Nature 392:408-411.
Boyle, J. S., I. G. Barr, and A. M. Lew. 1999. Strategies for improving responses to DNA vaccines. Mol. Med. 5:1-8.
Chaplin, P. J., R. De Rose, J. S. Boyle, P. McWaters, J. Kelly, J. M. Tennent, A. M. Lew, and J. P. Scheerlinck. 1999. Targeting improves the efficacy of a DNA vaccine against Corynebacterium pseudotuberculosis in sheep. Infect. Immun. 67:6434-6438.
Cote, P. J., M. Shapiro, R. E. Engle, H. Popper, R. H. Purcell, and J. L. Gerin. 1986. Protection of chimpanzees from type B hepatitis by immunization with woodchuck hepatitis virus surface antigen. J. Virol. 60:895-901.
Couillin, I., S. Pol, M. Mancini, F. Driss, C. Brechot, P. Tiollais, and M. L. Michel. 1999. Specific vaccine therapy in chronic hepatitis B: induction of T cell proliferative responses specific for envelope antigens. J. Infect. Dis. 180:15-26.
Davis, H. L. 1998. DNA-based immunization against hepatitis B: experience with animal models, p. 57-68. In H. Koprowski and D. B. Weiner (ed.), DNA vaccination/genetic vaccination. Current topics in microbiology and immunology. Springer Verlag, New York, N.Y.
Deliyannis, G., J. S. Boyle, J. L. Brady, L. E. Brown, and A. M. Lew. 2000. A fusion DNA vaccine that targets antigen-presenting cells increases protection from viral challenge. Proc. Natl. Acad. Sci. USA 97:6676-6680.
Donnelly, J. J., J. B. Ulmer, J. W. Shiver, and M. A. Liu. 1997. DNA vaccines. Annu. Rev. Immunol. 15:617-648.
Drew, D. R., J. S. Boyle, A. M. Lew, M. W. Lightowlers, P. J. Chaplin, and R. A. Strugnell. 2001. The comparative efficacy of CTLA-4 and L-selectin targeted DNA vaccines in mice and sheep. Vaccine 19:4417-4428.
Drew, D. R., J. S. Boyle, A. M. Lew, M. W. Lightowlers, and R. A. Strugnell. 2001. The human IgG3 hinge mediates the formation of antigen dimers that enhance humoral immune responses to DNA immunisation. Vaccine 19:4115-4120.
Fiedler, M., M. Lu, F. Siegel, J. Whipple, and M. Roggendorf. 2001. Immunization of woodchucks (Marmota monax) with hepatitis delta virus DNA vaccine. Vaccine 19:4618-4626.
Garcia-Navarro, R., B. Blanco-Urgoiti, P. Berraondo, R. Sanchez de la Rosa, A. Vales, S. Hervas-Stubbs, J. J. La Sarte, F. Borras, J. Ruiz, and J. Prieto. 2001. Protection against woodchuck hepatitis virus (WHV) infection by gene gun coimmunization with WHV core and interleukin-12. J. Virol. 75:9068-9076.
Girones, R., P. J. Cote, W. E. Hornbuckle, B. C. Tennant, J. L. Gerin, R. H. Purcell, and R. H. Miller. 1989. Complete nucleotide sequence of a molecular clone of woodchuck hepatitis virus that is infectious in the natural host. Proc. Natl. Acad. Sci. USA 86:1846-1849.
Harper, K., C. Balzano, E. Rouvier, M. G. Mattei, M. F. Lucani, and P. Golstein. 1991. CTLA-4 and CD28 activated lymphocyte molecules are closely related in both mouse and human as to sequence, message expression, gene structure, and chromosomal location. J. Immunol. 147:1037-1044.
Hervas-Stubbs, S., J. J. Lasarte, P. Sarobe, J. Prieto, J. Cullen, M. Roggendorf, and F. Borras-Cuesta. 1997. Therapeutic vaccination of woodchucks against chronic woodchuck hepatitis virus infection. J. Hepatol. 27:726-737.
Hervas-Stubbs, S., J. J. Lasarte, P. Sarobe, I. Vivas, L. Condreay, J. M. Cullen, J. Prieto, and F. Borras-Cuesta. 2001. T-helper cell response to woodchuck hepatitis virus antigens after therapeutic vaccination of chronically-infected animals treated with lamivudine. J. Hepatol. 35:105-111.
Hollinger, F. B., and T. J. Liang. 2001. Hepatitis B virus, p. 2971-3036. In B.N. Fields, D. M. Knipe, P. M. Howley, et al. (ed.), Fields virology, 4th ed., vol. 2. Lippincott-Williams & Wilkins, Philadelphia, Pa.
June, C. H., J. A. Bluestone, L. M. Nadler, and C. B. Thompson. 1994. The B7 and CD28 receptor families. Immunol. Today 15:321-331.
Kim, S. J., C. Lee, S. Y. Lee, I. Kim, J. S. Park, T. Sasagawa, J. J. Ko, S. E. Park, and Y. K. Oh. 2003. Enhanced immunogenicity of human papillomavirus 16 L1 genetic vaccines fused to an ER-targeting secretory signal peptide and RANTES. Gene Ther. 10:1268-1273.
Kim, S. J., D. Suh, S. E. Park, J. S. Park, H. M. Byun, C. Lee, S. Y. Lee, I. Kim, and Y. K. Oh. 2003. Enhanced immunogenicity of DNA fusion vaccine encoding secreted hepatitis B surface antigen and chemokine RANTES. Virology 15:84-91.
Kuhrober, A., J. Wild, H. P. Pudollek, F. V. Chisari, and J. Reimann. 1997. DNA vaccination with plasmids encoding the intracellular (HBcAg) or secreted (HBeAg) form of the core protein of hepatitis B virus primes T cell responses to two overlapping Kb- and Kd-restricted epitopes. Int. Immunol. 9:1203-1212.
Linsley, P. S., W. Brady, M. Urnes, L. S. Grosmaire, N. K. Damle, and J. A. Ledbetter. 1991. CTLA-4 is a second receptor for the B cell activation antigen B7. J. Exp. Med. 174:561-569.
Lu, M., G. Hilken, J. Kruppenbacher, T. Kemper, R. Schirmbeck, J. Reimann, and M. Roggendorf. 1999. Immunization of woodchucks with plasmids expressing woodchuck hepatitis virus (WHV) core antigen and surface antigen suppresses WHV infection. J. Virol. 73:281-289.
Lu, M., R. Klaers, S. Menne, J. Kruppenbacher, W. Gerlich, B. Stahl, H.-P. Dienes, U. Drebber, and M. Roggendorf. 2003. Vaccination of woodchucks with chronic woodchuck hepatitis virus (WHV) infection induces specific immune responses to WHV proteins. J. Hepatol. 39:405-413.
Lu, M., and M. Roggendorf. 2001. Evaluation of new approaches of prophylactic and therapeutic vaccinations to hepatitis B viruses in the woodchuck model. Intervirology 44:214-223.
Menne, S., J. Maschke, M. Lu, H. Grosse-Wilde, and M. Roggendorf. 1998. T-cell response to woodchuck hepatitis virus (WHV) antigens during acute self-limited WHV infection and convalescence and after viral challenge. J. Virol. 72:6083-6091.
Menne, S., J. Maschke, T. K. Tolle, M. Lu, and M. Roggendorf. 1997. Characterization of T-cell response to woodchuck hepatitis virus core protein and protection of woodchucks from infection by immunization with peptides containing a T-cell epitope. J. Virol. 71:65-74.
Menne, S., C. A. Roneker, B. E. Korba, J. L. Gerin; B. C. Tennant, and P. J. Cote. 2002. Immunization with surface antigen vaccine alone and after treatment with 1-(2-fluoro-5-methyl-beta-L-arabinofuranosyl)-uracil (L-FMAU) breaks humoral and cell-mediated immune tolerance in chronic woodchuck hepatitis virus infection. J. Virol. 76:5305-5314.
Michel, M. L., H. L. Davis, M. Schleef, M. Mancini, P. Tiollais, and R. G. Whalen. 1995. DNA-mediated immunization to the hepatitis B surface antigen in mice: aspects of the humoral response mimic hepatitis B viral infection in humans. Proc. Natl. Acad. Sci. USA 92:5307-5311.
Milich, D. R., A. McLachlan, G. B. Thornton, and J. L. Hughes. 1987. Antibody production to the nucleocapsid and envelope of the hepatitis B virus primed by a single synthetic T cell site. Nature 329:547-549.
Mitchell, J. A., T. D. Green, R. A. Bright, and T. M. Ross. 2003. Induction of heterosubtypic immunity to influenza A virus using a DNA vaccine expressing hemagglutinin-C3d fusion proteins. Vaccine 21:902-914.
Nayak, B. P., G. Sailaja, and A. M. Jabbar. 2003. Enhancement of gp120-specific immune responses by genetic vaccination with the human immunodeficiency virus type 1 envelope gene fused to the gene coding for soluble CTLA4. J. Virol. 77:10850-10861.
Parsons, K. R., J. R. Young, B. A. Collins, and C. J. Howard. 1996. Cattle CTLA-4, CD28 and chicken CD28 bind CD86: MYPPPY is not conserved in cattle CD28. Immunogenetics 43:388-391.
Pol, S., B. Nalpas, F. Driss, M. L. Michel, P. Tiollais, J. Denis, and C. Brechot. 2001. Efficacy and limitations of a specific immunotherapy in chronic hepatitis B. J. Hepatol. 34:917-921.
Prince, A. M., R. Whalen, and B. Brotman. 1997. Successful nucleic acid based immunization of newborn chimpanzees against hepatitis B virus. Vaccine 15:916-919.
Roggendorf, M. and M. Lu. Woodchuck hepatitis virus. In H. Thomas (ed.), Viral hepatitis, 3rd ed., in press. Blackwell Publishing, Oxford, United Kingdom.
Roggendorf, M., and T. K. Tolle. 1995. The woodchuck: an animal model for hepatitis B virus infection in man. Intervirology 38:100-112.
Sailaja, G., S. Husain, B. P. Nayak, and A. M. Jabbar. 2003. Long-term maintenance of gp120-specific immune responses by genetic vaccination with the HIV-1 envelope genes linked to the gene encoding Flt-3 ligand. J. Immunol. 170:2496-2507.
Sanjay, G., D. M. Klinman, and R. A. Seder. 2000. DNA vaccines: immunology, application, and optimization. Annu. Rev. Immunol. 18:927-974.
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.
Siegel, F., M. Lu, and M. Roggendorf. 2001. Coadministration of gamma interferon with DNA vaccine expressing woodchuck hepatitis virus (WHV) core antigen enhances the specific immune response and protects against WHV infection. J. Virol. 75:5036-5042.
Srivstava, I. K., and M. A. Liu. 2003. Gene vaccines. Ann. Intern. Med. 138:550-559.
Szmuness, W., C. E. Stevens, E. A. Zang, E. J. Harley, and A. Kellner. 1981. A controlled clinical trial of the efficacy of the hepatitis B vaccine (Heptavax B): a final report. Hepatology 1:377-385.
Tachedjian, M., J. S. Boyle, A. M. Lew, B. Horvatic, J. P. Scheerlinck, J. M. Tennent, and M. E. Andrew. 2003. Gene gun immunization in a preclinical model is enhanced by B7 targeting. Vaccine 21:2900-2905.
Yang, D., M. Roggendorf, and M. Lu. 2003. Molecular characterization of CD28 and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) of woodchuck (Marmota monax). Tissue Antigens 62:225-232.(Mengji Lu, Masanori Isoga)
Zentrales Tierlaboratorium, Universit?tsklinkum Essen, Essen, Germany
ABSTRACT
A number of options are available to modify and improve DNA vaccines. An interesting approach to improve DNA vaccines is to fuse bioactive domains, like cytotoxic-T-lymphocyte-associated protein 4 (CTLA-4), to an antigen. Such fusion antigens are expressed in vivo and directed to immune cells by the specific bioactive domain and therefore possess great potential to induce and modulate antigen-specific immune responses. In the present study, we tested this new approach for immunomodulation against hepadnavirus infection in the woodchuck model. Plasmids expressing the nucleocapsid protein (WHcAg) and e antigen (WHeAg) of woodchuck hepatitis virus (WHV) alone or in fusion to the extracellular domain of woodchuck CTLA-4 and CD28 were constructed. Immunizations of mice with plasmids expressing WHcAg or WHeAg led to a specific immunoglobulin G2a (IgG2a)-dominant antibody response. In contrast, fusions of WHcAg to CTLA-4 and CD28 induced a specific antibody response with comparable levels of IgG1 and IgG2a. Furthermore, the specific IgG1 response to WHcAg/WHeAg developed immediately after a single immunization with the CTLA-4-WHcAg fusion. Woodchucks were immunized with plasmids expressing WHeAg or the CTLA-4-WHcAg fusion and subsequently challenged with WHV. CTLA-4-WHcAg showed an improved efficacy in induction of protective immune responses to WHV. In particular, the anti-WHsAg antibody response developed earlier after challenge in woodchucks that received immunizations with CTLA-4-WHcAg, consistent with the hypothesis that anti-WHs response is dependent on a Th cell response to WHcAg. In conclusion, the use of fusion genes represents a generally applicable strategy to improve DNA vaccination.
INTRODUCTION
DNA vaccination is a useful technique to induce potent antigen-specific immune responses and possesses great potential for modification and improvement (2, 8, 39, 42). In general, the application of plasmid DNA by intramuscular injections induces dominantly cell-mediated immune responses, termed Th1-type responses. A number of options are available to modify DNA vaccines, e.g., improvement of antigen expression, coadministration of cytokines, or choice of application routes. An interesting approach is to fuse a bioactive domain, like cytotoxic-T-lymphocyte-associated protein 4 (CTLA-4), to an antigen (1). Such fusion antigens have been shown to bind to antigen-presenting cells expressing B7 molecules on their surfaces (1, 7, 32, 44). Immunization with plasmids expressing fusion proteins with CTLA-4 leads to antigen-specific immune responses, preferentially with an enhanced humoral branch (termed Th2-type responses), though DNA vaccination via the intramuscular route usually induces Th1-type immune responses (1). Thus, the fusion of a specific antigen to CTLA-4 provides a simple but effective modification of DNA vaccines if balanced Th1 and Th2 immune responses are desirable. Such fusion genes showed the ability to induce and modulate antigen-specific immune responses and improved protection against challenges with the respective pathogens (3, 7, 9). For example, the use of a fusion of CTLA-4 and influenza hemagglutinin for DNA vaccination enhanced the antigen-specific immune response and conferred better protection against influenza virus challenge in mice (7).
Hepatitis B virus (HBV) infection is still one of the major infectious diseases worldwide and results in severe liver diseases, cirrhosis, and hepatocellular carcinoma (17). HBV infection in humans can be effectively controlled by vaccination with recombinant HBsAg (43). Immunizations with plasmids expressing the HBV surface antigens (HBsAg) and nucleocapsid protein (HBcAg) effectively induced specific antibody and cytotoxic-T-cell responses to the respective antigens in the mouse model (21, 29, 40; reviewed in reference 6). However, DNA immunizations in large animals like chimpanzees were not efficient, as plasmids expressing HBsAg had to be applied in a scale of milligrams to induce a measurable anti-HBs antibody response (6, 35). Thus, DNA vaccines against HBV need significant improvements.
The woodchuck (Marmota monax) model is an informative animal model used to perform vaccination trials against HBV infection (4, 11, 12, 15, 16, 23-28, 36, 37). Woodchuck hepatitis virus (WHV) causes acute self-limiting and chronic infection, like HBV in humans (reviewed in references 25, 36, and 37). The humoral and cellular immune responses to woodchuck hepatitis surface antigen (WHsAg) and core antigen (WHcAg) in acute and chronic WHV infection are similar to HBV infection. DNA vaccinations of woodchucks with plasmids expressing WHsAg and WHcAg were able to protect against subsequent challenge with WHV (23, 41). However, the induction of a measurable immune response to WHV proteins by current protocols of DNA vaccination is rather inefficient (23). In particular, antibody responses to viral antigens could only be stimulated at low levels, presumably due to the fact that DNA vaccination by the intramuscular route preferentially primes immune responses of Th1 type (12, 41).
In the present study, a new approach using fusion proteins to CTLA-4 and CD28 was tested to overcome this shortcoming. The woodchuck CD28 (wCD28) and CTLA-4 (wCTLA-4) proteins are highly similar to their mammalian counterparts and share conserved common structural features (45). A hexapeptide motif, MYPPPY, that has been shown to be important for binding to the CD80 and CD86 ligands is conserved in both wCD28 and wCTLA-4. The binding motifs for the p85 subunit of phosphoinositide 3-kinase, YMNMTPR of CD28 (amino acids [aa] 192 to 197) and YVKMPP (aa 201 to 206), are also conserved in the deduced amino acid sequences of the corresponding woodchuck proteins (14, 18). CTLA-4 has been shown to be able to interact with heterologous CD80 and CD86 proteins of other species (33). This circumstance enables us to test the abilities of our fusion proteins with wCTLA-4 and wCD28 directly in both mouse and woodchuck models. Plasmids expressing WHV proteins alone or in fusion to the extracellular domain of woodchuck CTLA-4 and CD28 were constructed. The abilities of these plasmids to induce a specific immune response to WHcAg or WHeAg were tested by immunization of mice. While immunizations with plasmids expressing WHeAg or WHcAg alone led to a specific antibody response with immunoglobulin G2a (IgG2a) as the dominant IgG subtype, fusions of WHcAg to wCTLA-4 induced rapid and enhanced antibody responses with both IgG1 and IgG2a at comparable levels. Furthermore, woodchucks were immunized with plasmids expressing WHeAg or a CTLA-4-WHcAg fusion and subsequently challenged with WHV. Immunizations with the fusion gene induced a stronger antibody response to WHcAg that was further stimulated upon challenge with WHV and resulted in protection of immunized woodchucks against WHV infection, in contrast to immunizations with plasmids expressing WHeAg. The results of these experiments indicate that the new approach significantly improves DNA vaccination in the woodchuck model.
MATERIALS AND METHODS
Woodchucks. Adult WHV-negative woodchucks trapped in the state of New York were purchased from North Eastern Wildlife (Ithaca, N.Y.). Previous exposure of these woodchucks to WHV was excluded by testing for anti-WHcAg and anti-WHsAg antibodies, and for WHsAg as well.
Construction of plasmids for DNA vaccination. A plasmid, pWHcIm, had been constructed in a previous work and used for DNA vaccination (23, 41). The preC/C region of WHV8 was amplified by PCR using primers WHpreC EV1 and WHc EV2 (Table 1). The PCR products were cloned into pcDNA3.1 vectors (Invitrogen, Karlsruhe, Germany) according to the manufacturer's instructions. A clone containing the PCR product was selected after verification of the correct nucleotide sequence of the cloned fragment by sequencing. This plasmid, designated pWHeIm, contains the WHV preC/C region under the control of the cytomegalovirus (CMV) immediate-early promoter (Fig. 1).
The cDNA fragments comprising the coding regions of wCD28 and wCTLA-4 were amplified by reverse transcription-PCR, cloned into the pCRII vector (Invitrogen), and characterized in a previous work (45). The cloned cDNA fragments of wCTLA-4 were isolated by digestion with HindIII and XhoI and inserted into the corresponding site of the pcDNA3 vector predigested with HindIII and XhoI (Invitrogen) to generate peCTLA-4. Transfection with peCTLA-4 led to the expression of wCTLA-4 in mammalian cells, as shown in a previous work (45). The coding regions for the extracellular domains of wCD28 and wCTLA-4 were amplified with the primer pairs CD28H-CD28EV and CTLA-4H-CTLA-4EV (Table 1), respectively. The PCR fragments were cloned into pCRII vectors, sequenced to exclude nucleotide misincorporation by PCR, and recloned into pcDNA3 by restriction with HindIII and XhoI. This procedure generated peWCD28E and peWCTLA-4E. These plasmids were digested with EcoRV and XhoI, and a fragment comprising the coding region of WHcAg amplified with primers WHc EV1 and WHc EV2 was inserted, resulting in plasmids pCD28-C and pCTLA-4-C (Table 1 and Fig. 1). pCTLA-4-C was modified by digestion with StuI and MluNI and religation, resulting in a small deletion of 93 base pairs between nt 24 and 117 within the coding region of wCTLA-4. This plasmid, pCTLA-4-C, encodes a CTLA-4-WHcAg fusion protein with a deletion of 31 amino acid residues of wCTLA-4. Since the deletion eliminated the signal peptide of wCTLA-4, the fusion protein was expected to be deficient for secretion. All open reading frames listed here were placed under the control of the CMV immediate-early promoter of pcDNA3.
For DNA immunizations, plasmids were prepared with the Giga plasmid purification kit (QIAGEN, Hilden, Germany). Plasmids were dissolved in phosphate-buffered saline in a concentration of 1 mg per ml.
In vitro translation. The proteins encoded by the expression vectors constructed in this study were visualized by in vitro translation. The plasmids were cut by XhoI, precipitated with ethanol, and subjected to in vitro translation with a TNT-coupled translation kit (Promega, Mannheim, Germany). The proteins were labeled by [35S]methionine and subjected to sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis. The bands of labeled proteins were detected by the exposure of the dried SDS-polyacrylamide gel to an X-ray autograph.
Transient expression and detection of WHcAg or recombinant fusion proteins in transiently transfected woodchuck cells. A woodchuck cell line, WH12/6, was used for transfection experiments. Transient transfection was performed by using Lipofectamine (Gibco BRL, Neu-Isenburg, Germany) as described previously (23). Four micrograms of plasmids was incubated with 10 μg of Lipofectamine in 100 μl of Opti-Mem for 45 min and was given to cells in 1 ml of Opti-Mem (Gibco BRL) for 6 h at 37°C, 5% CO2. Transfected cells were maintained for 48 h at 37°C, 5% CO2, and fixed with 50% methanol. The expressed WHcAg and recombinant fusion proteins were detected by indirect immunofluorescence (IF) staining using rabbit antisera to WHcAg. The expressed wCTLA-4 was detected by indirect IF staining using goat antisera to a conserved region, aa 205 to 223, of mouse CTLA-4 (Santa Cruz Biotechnology, The Netherlands).
Immunization of mice and woodchucks by intramuscular injection of plasmid DNA. Immunizations of mice and woodchucks were performed by the procedures described by Lu et al. (23). Briefly, BALB/cJ (H-2d) mice were kept under standard-pathogen-free conditions in the Central Animal Laboratory of the University of Essen. Mice 6 to 8 weeks of age were pretreated by intramuscular injection of 50 μl of cardiotoxin (10 μM) into the tibialis anterior muscle. After 1 week, 50 μg of plasmids at a concentration of 1 mg per ml was injected into the same muscle of both legs. Five mice per group were used for each plasmid. The plasmid injection was repeated twice at intervals of 3 weeks. The mice were sacrificed 3 weeks after the last immunization. A modified protocol was used for woodchucks. A week prior to the injection of plasmids, 500 μl of cardiotoxin (10 μM in phosphate-buffered saline) was injected into the tibialis cranialis muscle of the woodchucks. The woodchucks were vaccinated three times by intramuscular injection of 500 μl of plasmids at a concentration of 1 mg per ml into each tibialis cranialis muscle on both sides at 3-week intervals. Two woodchucks were used for each plasmid. Seven weeks after the last vaccination, the woodchucks were challenged by intravenous injection of an inoculum containing 106 WHV genome equivalents.
Serology and detection of WHV DNA. Antibodies to WHcAg or to WHeAg (anti-WHcAg/WHeAg) were detected by a new version of an enzyme-linked immunosorbent assay (ELISA). Recombinant WHcAg particles were produced in Escherichia coli and purified by a combined protocol with precipitation and 30% saturation of ammonium sulfate and chromatographic separation though a Superose 6 column. The microtiter plate was coated with 10 μg per ml of purified WHcAg. After being blocked with 5% fetal calf serum, 100 μl of mouse serum at an appropriate dilution (1:10 to 1:1,000) was added and incubated for 1 h at 37°C. The bound mouse total IgG, IgG1, or IgG2a was detected with appropriate secondary antibodies, anti-mouse IgG, anti-mouse IgG1, or anti-mouse IgG2a, labeled with horseradish peroxidase (DB Biosciences, CA) at a dilution of 1:1,000. The development of color occurred at room temperature and was read at 490 nm. The cutoff value was set as three times over negative controls. The titers of antibodies to WHcAg or WHeAg in serum samples were calculated by extrapolation of ELISA values of serially diluted samples and corresponded to the reciprocal values of the highest dilutions that were regarded as positive. This ELISA format did not strictly differentiate antibodies to WHcAg and WHeAg, since the preparation of recombinant WHcAg obviously contained a fraction of polypeptide exposing different epitopes. However, it allowed the differentiation of subtypes of WHcAg/WHeAg-specific antibodies.
The detection of anti-WHcAg/WHeAg in woodchuck sera was done as follows: the microtiter plate was coated with 10 μg per ml of purified WHcAg; after being blocked with 5% fetal calf serum, 100 μl of woodchuck serum (dilution, 1:10) was added and incubated for 1 h at 37°C, and woodchuck IgGs were detected with protein A labeled with horseradish peroxidase. Anti-WHsAg antibody in woodchuck sera was detected by ELISA as described previously (23).
The dot blot technique was routinely performed to detect WHV DNA in woodchuck sera. For PCR detection of WHV DNA in woodchuck sera, nucleic acids were extracted with a serum blood kit (QIAGEN). PCR for amplification of the WHV core gene was run with primers wc1 and wc2 as described previously (23). WHV DNAs in woodchuck sera were quantified by real-time PCR with a light cycler DNA Master SYBR Green kit (Roche). The primers used for the PCR were designed according to the method of Girones et al. (13): wc1 (5' TGG GGC CAT GGA CAT AGA TCC TTA 3' [sequences containing the restriction site are given in italic letters]; nt 2015 to 2038) and wc-149s (5' AAG ATC TCT AAA TGA CTG TAT GTT CCG 3'; nt 2467 to 2451). The reactions were run in a light cycler (Roche) at 95°C for 0 s, 53°C for 10 s, and 72°C for 12 s. A plasmid containing a full-length WHV genome was diluted and served as the standard. The detection limit of this assay was 103 WHV genome equivalents per reaction.
Measurement of WHV antigen-specific proliferation of mouse splenocytes and woodchuck peripheral blood mononuclear cells (PBMCs). Mouse splenocytes were prepared in RPMI 1680 medium, seeded in triplicates of 2 x 104 per well, and cultured in flat-bottom 96-well microtiter plates (Falcon; Becton Dickinson, N.J.) at 37°C in a humidified atmosphere containing 5% CO2. The proliferation of mouse splenocytes in response to WHcAg- or WHcAg-derived peptides was measured at a concentration of 1 μg of antigen or peptide per ml. After a 5-day incubation, the cells were labeled with 1 μCi of [2-3H]thymidine (Amersham, Braunschweig, Germany) for 20 h and collected by a cell harvester (Packard Instrument Company, Calif.). The results for triplicate cultures are presented as the mean stimulation index (SI; the mean total thymidine incorporation in stimulated mouse splenocytes divided by the mean total thymidine incorporation in the control). The standard deviations of the means were less than 30% of the mean.
Antigen-specific proliferation of woodchuck PBMCs was determined by a [2-3H]adenine assay described previously (26, 27). SIs greater than 2.1 were considered positive for lymphoproliferative responses.
RESULTS
Construction of DNA vaccines. Six plasmids used for vaccination, pWHcIm, pWHeIm, pCD28-C, pCTLA-4-C, pCTLA-4-C, and peCTLA4, are depicted in Fig. 1A. pWHcIm and pWHeIm express WHcAg and WHeAg, respectively. In pCD28-C and pCTLA-4-C, the coding region of WHcAg was fused to the 3' ends of the coding sequences for the extracellular domains of wCD28 (the N-terminal 153 aa) and wCTLA-4 (the N-terminal 161 aa), respectively. Both CD28 and CTLA-4 have a signal peptide and are cell surface proteins on the cytoplasmic membrane. In the fusion proteins, the transmembrane regions of CD28 and CTLA-4 were removed, resulting in proteins without an anchoring sequence. For the control plasmid pCTLA-4-C, a small deletion of 93 base pairs was introduced between nt 24 and 117 of the wCTLA-4 coding region, leading to the destruction of the signal peptide of the wCTLA-4-WHcAg fusion protein.
The proteins encoded by these constructs were examined by in vitro translation. The in vitro-translated products from pWHcIm and pWHeIm had molecular masses of about 20 and 22 kDa, corresponding to WHcAg and WHeAg, respectively (Fig. 1B). The fusion proteins consisting of WHcAg with the extracellular domains of wCD28 and wCTLA4 were detected as proteins at a molecular mass of about 38 kDa. The in vitro-translated product of pCTLA-4-C was slightly smaller than the fusion protein wCTLA-4-WHcAg due to the deletion of 31 aa residues within wCTLA-4.
The expression of the WHV proteins and fusion proteins was further examined by transient transfection in woodchuck 12/6 cells and by IF staining. Cells transfected with pWHcIm, pWHeIm, pCD28-C, pCTLA-4-C, or pCTLA-4-C were positively stained with anti-WHcAg antibodies (23). Figure 1C shows the IF staining of cells transiently transfected with pWHeIm and pCTLA-4-C. The expression of wCTLA-4 in cells transfected with peCTLA4 was shown by IF staining with anti-CTLA-4 antibodies (45).
Immune responses to WHcAg and WHeAg induced in mice by immunization with plasmids expressing WHV proteins and fusion proteins. The abilities of the constructed plasmids to induce a WHcAg/WHeAg-specific antibody response was tested by intramuscular DNA injection in mice as described previously (23). The antibody response induced by plasmid immunizations was monitored by measuring the serum titers of specific antibodies to WHcAg and WHeAg. Immunization with pWHcIm, pWHeIm, pCTLA-4-C, and pCD28-C led to the production of antibodies to WHcAg and WHeAg in mice, while pCTLA-4-C and peCTLA4 were not able to induce any anti-WHcAg/WHeAg antibody (Fig. 2A). For pCTLA-4-C, the deletion within the leader peptide of wCTLA-4 appeared to prevent the wCTLA-4-WHcAg fusion protein from folding into a proper conformation. Immunizations with pCTLA-4-C resulted in the highest titer of the total anti-WHcAg/WHeAg IgG, which was about twice as high as the anti-WHcAg/WHeAg IgG titers induced by pWHeIm and pCD28-C.
The specific lymphoproliferative response to WHcAg in mice immunized with pWHeIm and pCTLA-4-C was monitored. Splenocytes were prepared from mice that had received three immunizations and were cultured with 24 overlapping peptides covering the whole WHcAg at a concentration of 1 μg per ml. The cells were then labeled with [2-3H]thymidine and harvested. Splenocytes from immunized mice showed proliferative responses to various peptides, particularly peptides comprising aa 57 to 93 of WHcAg (Fig. 2B). The SI was up to 17. Splenocytes from mice without immunizations did not respond to stimulation with peptides (data not shown).
Subtypes of anti-WHcAg/WHeAg IgG and their kinetics induced in mice by plasmid immunizations. The subtypes of anti-WHcAg/WHeAg IgG in mice immunized with pWHcIm, pWHeIm, pCTLA-4-C, and pCD28-C were determined by ELISA using mouse IgG1- and IgG2a-specific secondary antibodies (Fig. 3A). The immunizations with pWHcIm led exclusively to the production of IgG2a to WHcAg/WHeAg. After three immunizations with pWHeIm, anti-WHcAg/WHeAg antibodies were detected in mice at titers of 1:151 and 1:686 for IgG1 and IgG2a, respectively. The relative dominance of the antibody subtype IgG2a over IgG1 against WHcAg/WHeAg indicates that pWHcIm and pWHeIm induced a Th1-dominant immune response. Both pCTLA-4-C and pCD28-C induced IgG2a and IgG1 antibodies to WHcAg/WHeAg at comparable levels. The ratios of IgG1 to IgG2a were 0, 0.22, 0.98, and 0.84 for pWHcIm, pWHeIm, pCTLA-4-C, and pCD28-C, respectively (Fig. 3B). Thus, immunizations with the fusion constructs with CTLA-4 and CD28 led to an enhancement of Th2-type responses, while Th1-type responses were not affected.
Figure 4 shows the kinetics of the induction of anti-WHcAg/WHeAg antibodies in mice immunized with pWHcIm and pCTLA-4-C. Mice were immunized at weeks 0, 3, and 6. Sera were taken from the immunized mice at weeks 3, 6, and 9 and tested for WHcAg/WHeAg-specific IgG2a and IgG1. The immunizations with pWHcIm led to the production of WHcAg/WHeAg-specific IgG2a. The titers of WHcAg/WHeAg-specific IgG2a increased with each immunization and reached over 1:1,000 on average after three immunizations. In contrast, no WHcAg/WHeAg-specific IgG1 was detected in these sera at a dilution of 1:20. Mice receiving only one immunization with pCTLA-4-C developed both WHcAg/WHeAg-specific IgG2a and IgG1 at titers of 1:382 and 1:962, respectively. The titers of WHAg/WHeAg-specific IgG2a in pCTLA-4-C-immunized mice reached 1:612 and 1:850 at weeks 6 and 9, while the titers of WHcAg/WHeAg-specific IgG1 were 1:1,356 and 1:985 at the same times. It appears that the induction of the specific IgG1 occurred after a single immunization at a high level and was slightly enhanced by further boosts. However, the production of the specific IgG2a was strongly dependent on boosting and was improved markedly after three immunizations.
The kinetics of induction of anti-WHcAg/WHeAg in mice by immunization with pWHeIm were similar to those with pWHcIm. However, pCD28-C did not induce a significant IgG1 antibody response after a single immunization (data not shown).
Immunization of woodchucks with pWHeIm and pCTLA-4-C. We showed that both pWHeIm and pCTLA-4-C were able to induce anti-WHcAg/WHeAg responses, but with different qualities in mice, though both WHeAg and the fusion protein CTLA-4-WHcAg encoded by vectors pWHeIm and pCTLA-4-C possess signal peptides. While pWHeIm induced a Th1-dominant immune response, immunization with pCTLA-4-C primed a balanced Th1-Th2 response to WHcAg/WHeAg in mice. Thus, immunizations and WHV challenge in the woodchuck model were carried out to clarify whether the induction of a balanced Th1-Th2 response to WHcAg/WHeAg is beneficial for protection against viral challenge.
Three groups of two woodchucks each received immunizations with the control plasmid peCTLA-4, pWHeIm, or pCTLA-4-C. Three immunizations were carried out at intervals of 3 weeks. Seven weeks after the last immunization, the immunized woodchucks were challenged with a stock containing 106 WHV genome equivalents and then monitored for up to 12 weeks. The anti-WHcAg/WHeAg antibody response was determined for the entire period. The anti-WHsAg antibodies and serum WHV DNA concentrations in the woodchucks were measured after challenge.
Two control woodchucks, WH12033 and WH12304, were immunized three times with peCTLA-4 and did not show any response to WHV proteins (Fig. 5). After the challenge, both woodchucks developed typical signs of acute WHV infection. WH12033 was positive for WHV DNA after challenge and had a peak level of WHV DNA at 3.65 x 1010 genome equivalents per ml at week 4 postinfection (p.i.). The WHV DNA concentration decreased rapidly in the following weeks but remained detectable until week 8 p.i. Antibodies to WHcAg/WHeAg and to WHsAg became detectable at weeks 3 and 5 p.i., respectively. The WHV DNA was at a high level, up to 6.9 x 1010 genome equivalents per ml, in WH12304 from weeks 4 to 8 p.i. Anti-WHcAg/WHeAg antibodies developed at week 6 p.i. Anti-WHsAg antibodies did not become detectable until week 8 p.i.
Two woodchucks, WH12027 and WH12034, were immunized three times with 1 mg of pWHeIm per injection. A weak anti-WHcAg/WHeAg antibody response was measured after three immunizations (Fig. 5). One woodchuck, WH12027, was completely protected against challenge with WHV, as no WHV DNA was detected by PCR. The anti-WHsAg antibody response developed at week 6 p.i. Woodchuck WH12034 was infected and showed a level of viremia of about 1.6 x 109 copies per ml for a short period during weeks 3 to 4 p.i. The WHV DNA remained detectable at a low level of 106 copies per ml until week 6. The anti-WHcAg/WHeAg antibody titer increased significantly at week 3 p.i. WH12034 became positive for anti-WHsAg at week 6 p.i.
Three immunizations with pCTLA-4-C led to the development of anti-WHcAg/WHeAg antibodies in two woodchucks, WH12035 and WH12293. In particular, anti-WHcAg/WHeAg antibodies became detectable in WH12035 after only a single immunization with pCTLA-4-C. The antibody titers increased after boosts in both woodchucks. The challenge with WHV resulted in a rapid increase of the WHcAg/WHeAg-specific antibody response that was not observed in woodchucks immunized with peCTLA-4 or pWHeIm. Both woodchucks did not develop detectable viremia. Woodchucks WH12035 and WH12293 developed anti-WHsAg antibodies in weeks 4 and 3 p.i. This unique serological profile in WH12035 and WH12293 indicated that the immunization with pCTLA-4-C did effectively change the quality of the immune response to WHV proteins in both woodchucks. Thus, pCTLA-4-C was able to induce an enhanced antibody response to WHV proteins, probably due to its ability to prime an additional antigen-specific Th2 response.
Lymphoproliferative responses to WHV proteins in woodchucks. The lymphoproliferative responses to WHcAg and WHcAg-derived peptides were measured during the period of immunization and WHV challenge. Immunizations with pWHeIm and pCTLA-4-C induced low lymphoproliferative responses to WHcAg and peptides in woodchucks (Fig. 6A). For example, the woodchuck WH12027 showed significant lymphoproliferative responses to WHcAg and four different WHcAg-derived peptides (Fig. 6B). These results were consistent with previous findings that DNA vaccination is able to prime specific lymphoproliferative responses to WHV proteins. No WHcAg-specific lymphoproliferative response was detected in two control woodchucks, WH12033 and WH12304, before WHV challenge. After challenge, PBMCs of WH12033 responded either to WHcAg and WHsAg at week 5 p.i., concomitantly with the peak viremia (Fig. 5 and Fig. 6A). For WH12304, no specific lymphoproliferative response to WHV was detected during the whole experiment. These results were consistent with previous findings.
DISCUSSION
DNA immunizations with plasmids expressing WHcAg and WHeAg by intramuscular injection primed a Th1-type-dominant immune response to the respective viral antigens in mice, as is typical for this immunization procedure. We showed here that the fusion gene of WHcAg to the extracellular domain of wCTLA-4 or wCD28 induced WHcAg/WHeAg-specific immune responses that were significantly different in quality and kinetics in the mouse model. Strikingly, a single injection of plasmids expressing wCTLA-4-WHcAg led to the production of WHcAg/WHeAg-specific antibodies of subtype IgG1. These results indicate that the fusion of WHcAg to wCTLA-4 or wCD28 dramatically changes the manner of presentation of viral antigens. It has been shown that fusion proteins of influenza hemagglutinin or human immunodeficiency virus gp120 with CTLA-4 were able to interact with antigen-presenting cells expressing B7 molecules (1, 7, 32, 44). It appears that the exogenic antigen presentation pathway was enhanced and therefore Th2-type responses were primed, in addition to the Th1-type response. It should be clarified by further investigation whether the fusion protein wCTLA-4-WHcAg is indeed taken up by antigen-presenting cells with accelerated kinetics, leading to enhanced presentation for Th cells.
Both pCD28-C and pCTLA-4-C induced a mixed IgG1-IgG2a response to WHcAg/WHeAg in mice. However, the titers of anti-WHcAg/WHeAg in pCD28-C-immunized mice were lower than those in pCTLA-4-C-immunized mice. In addition, only pCTLA-4-C showed the ability to prime the IgG1 response after a single immunization. The CTLA-4 molecule has a 1,000-fold-higher affinity for the B7 molecule than CD28 (22). This could be interpreted as the fusion protein of WHcAg to the extracellular domains of wCTLA-4 being more efficiently directed to antigen-presenting cells. It will be interesting to investigate whether the affinity of the ligand to its receptor determines the antigen-specific immune response primed by fusion proteins.
One interesting issue is whether DNA vaccinations with the fusion genes induce specific immune responses to the domain of the cellular protein. We attempted to detect wCTLA-4-specific antibodies in sera from immunized mice and woodchucks by IF staining of transiently transfected cells; however, no specific staining of wCTLA-4 was observed. Further examination with more sensitive methods will be needed to clarify the possibility of induction of a specific immune response against wCTLA-4 in immunized animals.
The immunizations with pWHeIm led to the production of anti-WHcAg/WHeAg antibodies at only a very low level in woodchucks. After viral challenge, woodchucks immunized with pWHeIm showed development of strong anti-WHcAg/WHeAg antibody response within 3 weeks. Consistently, previous experiments demonstrated that antibody responses to WHcAg or WHsAg were primed by DNA immunization and greatly stimulated by the viral challenge, even though viral replication in immunized individuals was very limited (23). The immunizations with pCTLA-4-C induced a stronger anti-WHcAg/WHeAg antibody response in woodchucks than that with pWHeIm. An immediate increase of anti-WHcAg/WHeAg was observed in two woodchucks, WH12035 and WH12239, after WHV challenge. These results demonstrated the superiority of pCTLA-4-C over pWHeIm in terms of induction of specific antibody responses.
It is highly interesting that anti-WHsAg antibodies became detectable as early as 3 weeks p.i. in woodchucks immunized with pCTLA-4-C. The early appearance of anti-WHsAg in pCTLA-4-C-immunized woodchucks may have different reasons. Since the development of anti-WHsAg during an acute WHV infection is associated with viral clearance, an efficient protection against WHV challenge by primed immune responses may result in the rapid appearance of anti-WHsAg. In addition, a strong Th response to WHcAg, particularly the enhanced Th2 response, may provide help for the development of anti-WHsAg. Milich et al. pointed out that the HBcAg-specific Th response is critical for anti-HBsAg antibody production (30). Thus, further work should be done to elucidate the relationship between WHcAg-specific Th responses and anti-WHsAg responses in woodchucks.
Inoculation of na?ve woodchucks with 106 WHV genome equivalents leads to acute infection with transiently high serum titers of WHV DNA at a level of 1010 copies per ml. Immunizations with pWHeIm led to protection against WHV challenge in one woodchuck, WH12027. Another woodchuck, WH12034, became viremic. Compared with the control woodchucks, the immunizations with pWHeIm appeared to partially limit viral replication. The peak level of WHV in WH12034 reached 108 WHV genome equivalents per ml, a viremic level that was significantly lower than that in the controls. WHV DNA was not detected in two woodchucks immunized with pCTLA-4-C by dot blot hybridization and by PCR. These results are consistent with results from our previous studies and indicate that immunizations with WHcAg may at least partially control WHV infection (41).
It is a generally applicable concept to modify a given viral antigen by fusion to a cellular protein (10, 19, 20, 31, 38). By fusion to a cellular protein, viral antigens will be engaged in protein-protein interactions that do not exist naturally. Thus, such artificial interactions lead to processing and presentation of viral antigens in an unusual context and may be beneficial in some instances. For example, the specific T-cell responses to HBV proteins are impaired in chronically HBV-infected patients. Although the HBV proteins are expressed at high levels in the patients, the host immune system is not able to mount an efficient response. It was proposed that therapeutic immunizations may stimulate the specific immune response in these patients. However, immunizations with purified viral surface antigens failed to induce an immune response that suppressed HBV replication in the majority of patients (5, 34). Therefore, modifications of viral antigens will be needed to enable efficient antigen presentation and induction of a sustained HBV-specific immune response. Thus, a regular antigen presentation pathway may be unable to stimulate the specific immune responses to viral antigens. Modifications of viral antigens, as was done in the present study, may circumvent the block by viral mechanisms and lead to efficient T- and B-cell responses.
ACKNOWLEDGMENTS
We are thankful to Michael Roggendorf for continuous support and helpful discussion. We are grateful to Thekla Kemper and Barbara Bleekmann for excellent technical assistance and to Ulf Dittmer for critical reading and helpful suggestions.
This work is partially supported by BMBF grant 01GE9909.
REFERENCES
Boyle, J. S., J. L. Brady, and A. M. Lew. 1998. Enhanced responses to a DNA vaccine encoding a fusion antigen that is directed to sites of immune induction. Nature 392:408-411.
Boyle, J. S., I. G. Barr, and A. M. Lew. 1999. Strategies for improving responses to DNA vaccines. Mol. Med. 5:1-8.
Chaplin, P. J., R. De Rose, J. S. Boyle, P. McWaters, J. Kelly, J. M. Tennent, A. M. Lew, and J. P. Scheerlinck. 1999. Targeting improves the efficacy of a DNA vaccine against Corynebacterium pseudotuberculosis in sheep. Infect. Immun. 67:6434-6438.
Cote, P. J., M. Shapiro, R. E. Engle, H. Popper, R. H. Purcell, and J. L. Gerin. 1986. Protection of chimpanzees from type B hepatitis by immunization with woodchuck hepatitis virus surface antigen. J. Virol. 60:895-901.
Couillin, I., S. Pol, M. Mancini, F. Driss, C. Brechot, P. Tiollais, and M. L. Michel. 1999. Specific vaccine therapy in chronic hepatitis B: induction of T cell proliferative responses specific for envelope antigens. J. Infect. Dis. 180:15-26.
Davis, H. L. 1998. DNA-based immunization against hepatitis B: experience with animal models, p. 57-68. In H. Koprowski and D. B. Weiner (ed.), DNA vaccination/genetic vaccination. Current topics in microbiology and immunology. Springer Verlag, New York, N.Y.
Deliyannis, G., J. S. Boyle, J. L. Brady, L. E. Brown, and A. M. Lew. 2000. A fusion DNA vaccine that targets antigen-presenting cells increases protection from viral challenge. Proc. Natl. Acad. Sci. USA 97:6676-6680.
Donnelly, J. J., J. B. Ulmer, J. W. Shiver, and M. A. Liu. 1997. DNA vaccines. Annu. Rev. Immunol. 15:617-648.
Drew, D. R., J. S. Boyle, A. M. Lew, M. W. Lightowlers, P. J. Chaplin, and R. A. Strugnell. 2001. The comparative efficacy of CTLA-4 and L-selectin targeted DNA vaccines in mice and sheep. Vaccine 19:4417-4428.
Drew, D. R., J. S. Boyle, A. M. Lew, M. W. Lightowlers, and R. A. Strugnell. 2001. The human IgG3 hinge mediates the formation of antigen dimers that enhance humoral immune responses to DNA immunisation. Vaccine 19:4115-4120.
Fiedler, M., M. Lu, F. Siegel, J. Whipple, and M. Roggendorf. 2001. Immunization of woodchucks (Marmota monax) with hepatitis delta virus DNA vaccine. Vaccine 19:4618-4626.
Garcia-Navarro, R., B. Blanco-Urgoiti, P. Berraondo, R. Sanchez de la Rosa, A. Vales, S. Hervas-Stubbs, J. J. La Sarte, F. Borras, J. Ruiz, and J. Prieto. 2001. Protection against woodchuck hepatitis virus (WHV) infection by gene gun coimmunization with WHV core and interleukin-12. J. Virol. 75:9068-9076.
Girones, R., P. J. Cote, W. E. Hornbuckle, B. C. Tennant, J. L. Gerin, R. H. Purcell, and R. H. Miller. 1989. Complete nucleotide sequence of a molecular clone of woodchuck hepatitis virus that is infectious in the natural host. Proc. Natl. Acad. Sci. USA 86:1846-1849.
Harper, K., C. Balzano, E. Rouvier, M. G. Mattei, M. F. Lucani, and P. Golstein. 1991. CTLA-4 and CD28 activated lymphocyte molecules are closely related in both mouse and human as to sequence, message expression, gene structure, and chromosomal location. J. Immunol. 147:1037-1044.
Hervas-Stubbs, S., J. J. Lasarte, P. Sarobe, J. Prieto, J. Cullen, M. Roggendorf, and F. Borras-Cuesta. 1997. Therapeutic vaccination of woodchucks against chronic woodchuck hepatitis virus infection. J. Hepatol. 27:726-737.
Hervas-Stubbs, S., J. J. Lasarte, P. Sarobe, I. Vivas, L. Condreay, J. M. Cullen, J. Prieto, and F. Borras-Cuesta. 2001. T-helper cell response to woodchuck hepatitis virus antigens after therapeutic vaccination of chronically-infected animals treated with lamivudine. J. Hepatol. 35:105-111.
Hollinger, F. B., and T. J. Liang. 2001. Hepatitis B virus, p. 2971-3036. In B.N. Fields, D. M. Knipe, P. M. Howley, et al. (ed.), Fields virology, 4th ed., vol. 2. Lippincott-Williams & Wilkins, Philadelphia, Pa.
June, C. H., J. A. Bluestone, L. M. Nadler, and C. B. Thompson. 1994. The B7 and CD28 receptor families. Immunol. Today 15:321-331.
Kim, S. J., C. Lee, S. Y. Lee, I. Kim, J. S. Park, T. Sasagawa, J. J. Ko, S. E. Park, and Y. K. Oh. 2003. Enhanced immunogenicity of human papillomavirus 16 L1 genetic vaccines fused to an ER-targeting secretory signal peptide and RANTES. Gene Ther. 10:1268-1273.
Kim, S. J., D. Suh, S. E. Park, J. S. Park, H. M. Byun, C. Lee, S. Y. Lee, I. Kim, and Y. K. Oh. 2003. Enhanced immunogenicity of DNA fusion vaccine encoding secreted hepatitis B surface antigen and chemokine RANTES. Virology 15:84-91.
Kuhrober, A., J. Wild, H. P. Pudollek, F. V. Chisari, and J. Reimann. 1997. DNA vaccination with plasmids encoding the intracellular (HBcAg) or secreted (HBeAg) form of the core protein of hepatitis B virus primes T cell responses to two overlapping Kb- and Kd-restricted epitopes. Int. Immunol. 9:1203-1212.
Linsley, P. S., W. Brady, M. Urnes, L. S. Grosmaire, N. K. Damle, and J. A. Ledbetter. 1991. CTLA-4 is a second receptor for the B cell activation antigen B7. J. Exp. Med. 174:561-569.
Lu, M., G. Hilken, J. Kruppenbacher, T. Kemper, R. Schirmbeck, J. Reimann, and M. Roggendorf. 1999. Immunization of woodchucks with plasmids expressing woodchuck hepatitis virus (WHV) core antigen and surface antigen suppresses WHV infection. J. Virol. 73:281-289.
Lu, M., R. Klaers, S. Menne, J. Kruppenbacher, W. Gerlich, B. Stahl, H.-P. Dienes, U. Drebber, and M. Roggendorf. 2003. Vaccination of woodchucks with chronic woodchuck hepatitis virus (WHV) infection induces specific immune responses to WHV proteins. J. Hepatol. 39:405-413.
Lu, M., and M. Roggendorf. 2001. Evaluation of new approaches of prophylactic and therapeutic vaccinations to hepatitis B viruses in the woodchuck model. Intervirology 44:214-223.
Menne, S., J. Maschke, M. Lu, H. Grosse-Wilde, and M. Roggendorf. 1998. T-cell response to woodchuck hepatitis virus (WHV) antigens during acute self-limited WHV infection and convalescence and after viral challenge. J. Virol. 72:6083-6091.
Menne, S., J. Maschke, T. K. Tolle, M. Lu, and M. Roggendorf. 1997. Characterization of T-cell response to woodchuck hepatitis virus core protein and protection of woodchucks from infection by immunization with peptides containing a T-cell epitope. J. Virol. 71:65-74.
Menne, S., C. A. Roneker, B. E. Korba, J. L. Gerin; B. C. Tennant, and P. J. Cote. 2002. Immunization with surface antigen vaccine alone and after treatment with 1-(2-fluoro-5-methyl-beta-L-arabinofuranosyl)-uracil (L-FMAU) breaks humoral and cell-mediated immune tolerance in chronic woodchuck hepatitis virus infection. J. Virol. 76:5305-5314.
Michel, M. L., H. L. Davis, M. Schleef, M. Mancini, P. Tiollais, and R. G. Whalen. 1995. DNA-mediated immunization to the hepatitis B surface antigen in mice: aspects of the humoral response mimic hepatitis B viral infection in humans. Proc. Natl. Acad. Sci. USA 92:5307-5311.
Milich, D. R., A. McLachlan, G. B. Thornton, and J. L. Hughes. 1987. Antibody production to the nucleocapsid and envelope of the hepatitis B virus primed by a single synthetic T cell site. Nature 329:547-549.
Mitchell, J. A., T. D. Green, R. A. Bright, and T. M. Ross. 2003. Induction of heterosubtypic immunity to influenza A virus using a DNA vaccine expressing hemagglutinin-C3d fusion proteins. Vaccine 21:902-914.
Nayak, B. P., G. Sailaja, and A. M. Jabbar. 2003. Enhancement of gp120-specific immune responses by genetic vaccination with the human immunodeficiency virus type 1 envelope gene fused to the gene coding for soluble CTLA4. J. Virol. 77:10850-10861.
Parsons, K. R., J. R. Young, B. A. Collins, and C. J. Howard. 1996. Cattle CTLA-4, CD28 and chicken CD28 bind CD86: MYPPPY is not conserved in cattle CD28. Immunogenetics 43:388-391.
Pol, S., B. Nalpas, F. Driss, M. L. Michel, P. Tiollais, J. Denis, and C. Brechot. 2001. Efficacy and limitations of a specific immunotherapy in chronic hepatitis B. J. Hepatol. 34:917-921.
Prince, A. M., R. Whalen, and B. Brotman. 1997. Successful nucleic acid based immunization of newborn chimpanzees against hepatitis B virus. Vaccine 15:916-919.
Roggendorf, M. and M. Lu. Woodchuck hepatitis virus. In H. Thomas (ed.), Viral hepatitis, 3rd ed., in press. Blackwell Publishing, Oxford, United Kingdom.
Roggendorf, M., and T. K. Tolle. 1995. The woodchuck: an animal model for hepatitis B virus infection in man. Intervirology 38:100-112.
Sailaja, G., S. Husain, B. P. Nayak, and A. M. Jabbar. 2003. Long-term maintenance of gp120-specific immune responses by genetic vaccination with the HIV-1 envelope genes linked to the gene encoding Flt-3 ligand. J. Immunol. 170:2496-2507.
Sanjay, G., D. M. Klinman, and R. A. Seder. 2000. DNA vaccines: immunology, application, and optimization. Annu. Rev. Immunol. 18:927-974.
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.
Siegel, F., M. Lu, and M. Roggendorf. 2001. Coadministration of gamma interferon with DNA vaccine expressing woodchuck hepatitis virus (WHV) core antigen enhances the specific immune response and protects against WHV infection. J. Virol. 75:5036-5042.
Srivstava, I. K., and M. A. Liu. 2003. Gene vaccines. Ann. Intern. Med. 138:550-559.
Szmuness, W., C. E. Stevens, E. A. Zang, E. J. Harley, and A. Kellner. 1981. A controlled clinical trial of the efficacy of the hepatitis B vaccine (Heptavax B): a final report. Hepatology 1:377-385.
Tachedjian, M., J. S. Boyle, A. M. Lew, B. Horvatic, J. P. Scheerlinck, J. M. Tennent, and M. E. Andrew. 2003. Gene gun immunization in a preclinical model is enhanced by B7 targeting. Vaccine 21:2900-2905.
Yang, D., M. Roggendorf, and M. Lu. 2003. Molecular characterization of CD28 and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) of woodchuck (Marmota monax). Tissue Antigens 62:225-232.(Mengji Lu, Masanori Isoga)