INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Tuberculosis (TB) still causes more deaths per year worldwide than any other bacterial pathogen (33). Our ability to control this devastating epidemic could be greatly enhanced by the development of a more effective anti-TB vaccine than the currently available bacillus Calmette-Guérin (BCG). Although BCG has been widely used in many Third World countries, its efficacy in a number of clinical trials has been highly variable, with an overall effectiveness of only about 50% (4).

Among the novel TB vaccines being developed are DNA constructs which express putative protective anti-TB antigens (11, 27). These DNA vaccines induced protective immune responses in animal models of a number of parasitic, viral, and bacterial infections (7, 30). Besides their high degree of immunogenicity, these DNA vaccines offer several other potential advantages, including ease of preparation, stability, and relatively low cost. For these reasons, DNA anti-TB vaccines represent good candidates for use in developing countries, where most of the TB cases are known to occur (33). TB DNA vaccines may also offer a prophylactic alternative for immunocompromised patients, where safety issues prevent the use of live BCG.

Another attractive feature of the DNA vaccine is its ability to enhance or modulate the host's immune response by the use of inducible promoters, immunoregulatory genes, or gene fusions (22, 32). We have recently shown that the immunogenicity of Mycobacterium tuberculosis antigens expressed from DNA vaccines as tissue plasminogen activator (tPA) fusion proteins is substantially elevated compared to constructs expressing the corresponding native TB antigen (16). Recently, DNA constructs have been developed that express proteins conjugated to ubiquitin. Ubiquitin is a 76-amino-acid peptide involved in controlling the normal protein intracellular turnover in the cytoplasm of eukaryotic cells. Proteins that are to be degraded are tagged with ubiquitin molecules and are targeted to the proteasome system (29). The ubiquitin conjugation enhances proteasome-dependent degradation of the endogenously synthesized antigens and results in an increase of the cell-mediated response induced in vivo against the conjugated antigen (23, 28, 34). In the present study, DNA vaccines expressing three different forms of ubiquitin fused to the mycobacterial antigen, MPT64, were cloned and evaluated for their immunological activity and ability to generate a protective immunity in vaccinated mice. We showed that a strong Th1-oriented immune response, in the absence of any detectable humoral response, was generated by DNA vaccine expressing a specific ubiquitin-conjugated protein, resulting in a protective immune response in this mouse model of pulmonary TB.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Specific-pathogen-free C57BL/6 female mice were obtained from the National Cancer Institute, National Institutes of Health, Bethesda, Md. The mice were 8 weeks old at the time of vaccination. They were maintained under barrier conditions and fed commercial mouse chow and water ad libitum.

Microorganisms. M. tuberculosis Erdman (TMC107) and M. bovis BCG Pasteur (TMC1011) were obtained from the Trudeau Mycobacterial Culture Collection, Saranac Lake, N.Y. The Escherichia coli JM109 and Top 10 strains (Invitrogen, San Diego, Calif.) were used for cloning. For expression of histidine-tagged antigens, E. coli BL21(DE3)/pLysS strain (Invitrogen) was transformed with the pET15b expression vector.

Cloning. The genes encoding MPT64 and ESAT-6 were amplified from M. tuberculosis H37Rv and cloned into pCRBlunt (Invitrogen) as indicated previously (16). Three different chimeric proteins were generated in which the ubiquitin molecule was fused to the antigen. The ubiquitin gene was amplified from a mouse cDNA library (Clontech, Palo Alto, Calif.) and cloned in pCR2.1 (TA Cloning Kit; Invitrogen). The same 5' primer (5'-ACAAGCTTACCATGCAGATCTTCGTGAAGACC-3') with the ATG starting codon and the HindIII site was used in the three PCR reactions. However, a different 3' primer was used to generate the three different ubiquitin clones. For the UbG64 vaccine, the normal ubiquitin protein of 76 amino acids with G₇₆ was cloned in pCR2.1 using the reverse primer (5'-ACGCTAGCGCCACCGCGCAGACGCAGCAC-3') and then inserted in frame with the MPT64 antigen in the expression vector pJW4303. In the construction of the UbA64 plasmid, the 3' primer (5'-ACGCTAGCGGCACCGCGCAGACGCAC-3') was changed so that alanine in position 76 was expressed instead of a glycine. To generate the UbGR64 construct, another reverse primer (5'-ACGCTAGCACGGCCACCGCGCAGACCAC-3') was used so that an extra arginine was added after the Gly₇₆ (23). For each construct, the ubiquitin gene was inserted in the HindIII-NheI sites of the DNA vaccine vector pJW4303, upstream from the already inserted MPT64 and ESAT-6 TB gene. The preparation of DNA vaccines expressing the tPA-fused and the native forms of these antigens have been described previously (16).

Immunization. Endotoxin-free plasmid DNA was prepared and purified with the Qiagen EndoFree Plasmid Maxi Kit (Qiagen, Chatsworth, Calif.). Groups of C57BL/6 mice were injected intramuscularly in both hind limbs on days 1, 21, and 42 with 100 µg of plasmid DNA in a total volume of 0.1 ml. As controls, mice were vaccinated subcutaneously with 5 × 10⁶ CFU BCG on day 1.

In vitro expression. Rhabdomyosarcoma (RD) cells (ATCC CCL 136, American Type Culture Collection) were grown in high-glucose Dulbecco modified Eagle medium supplemented with 10% heat-inactivated fetal calf serum, 2 mM glutamine, 20 mM HEPES, 100 U of penicillin per ml, and 100 µg of streptomycin per ml (Gibco-BRL) up to 60% confluency in six well plates. Cells were transfected with 2 µg of plasmid DNA and 6 µl of LipofectAMIN (Gibco-BRL) in serum-free medium (Opti-Mem; Gibco-BRL). After 5 h of incubation, the cells were fed with fresh complete medium. After 48 h, the transfected cells were washed twice in phosphate-buffered saline (PBS), harvested, and incubated in lysis buffer for 30 min on ice (1% Nonidet P-40; 0.1% sodium dodecyl sulfate [SDS]; 150 mM NaCl; 50 mM Tris-HCl, pH 8.0) in the presence of protease inhibitors. The cell lysates were analyzed by Western blotting. Nitrocellulose membranes (Gibco-BRL) containing SDS-polyacrylamide gel electrophoresis-separated cell lysate preparations were probed with specific mouse polyclonal antibodies, and the blots were developed using the ECL System (Amersham). Finally, the X-ray images were scanned and analyzed using ImageJ software. The amount of signal detected was expressed as the percentage of the highly expressed and stable antigen, tPA64.

Humoral response. At 28 days after the third immunization, sera were collected and pooled from the tail veins of the vaccinated mice. Immulon-1 plates (Dynatech, Chantilly, Va.) were coated overnight at 4°C with 0.1 ml of purified recombinant antigen (5 µg/ml) in a Coating Solution (KPL, Gaithersburg, Md.) and then blocked the next day with bovine serum albumin (BSA; Sigma). The recombinant-antigen purification was carried out as described earlier (16). Serum samples were applied in 0.1 ml of serial twofold dilutions, starting from 1:25. Anti-mouse immunoglobulin G (IgG) whole molecule alkaline phosphatase conjugate (Sigma, St. Louis, Mo.) was used as secondary antibody to establish the total IgG humoral response. Isotype detection was performed using goat anti-mouse IgG1 and IgG2a alkaline phosphatase conjugates (Southern Biotechnology, Birmingham, Ala.). For color development, the p-nitrophenylphosphate phosphatase system was used according to the directions supplied by the manufacturer (KPL), and the optical density (OD) was read at 405 nm on a Microplate enzyme-linked immunosorbent assay reader (BioTech Instruments). The endpoint was defined as the highest dilution of serum that gave an OD₄₀₅ value higher than 0.050 and that was twofold greater than that of the matched dilution of normal mouse sera (19).

Cytokine ELISPOT assay. Cytokine induction was evaluated using the ELISPOT protocol as previously described (15). Briefly, 96-well Immulon-2 plates were coated with anti-gamma interferon (anti-IFN-; clone R4-6A2, Pharmingen, San Diego, Calif.) or with anti-interleukin 4 (anti-IL-4; clone BVD4-1D11; Endogen, Woburn, Mass.) mouse antibody. The plates were blocked in PBS containing 5% of BSA (Sigma) and 0.025% Tween 20. Splenocytes were pooled from three mice per group and resuspended in RPMI 1640, 5% heat-inactivated fetal calf serum, 5% nonessential amino acids, 10 mM sodium pyruvate, 2-mercaptoethanol, and 100 U of penicillin-streptomycin (complete medium) per ml. Serial dilutions of the single-cell suspension, starting at 10⁶ cells/ml, were incubated on anticytokine antibody-coated plates in complete medium for 16 h at 37°C in a humidified 5% CO₂ incubator. The purified recombinant histidine-tagged MPT64 was added when required at a concentration of 10 µg/ml. A control BCG lysate was also used in each assay. The plates were washed with 0.025% Tween 20 in distilled water and then incubated with biotinylated anti-IFN- (clone XMG1; Pharmingen) or anti-IL-4 (clone BVD6-24G2; Pharmingen) antibodies at 1 µg/ml. Individual cytokine-secreting cells were visualized by the addition of a substrate 5-bromo-4-chloro-3-indolylphosphate (BCIP)-nitroblue tetrazolium agarose mixture (Sigma).

Low-dose aerogenic challenge with M. tuberculosis. Vaccinated and control mice were infected aerogenically with about 50 CFU of M. tuberculosis Erdman using a Middlebrook chamber (Glas Col, Terre Haute, Ind.) as described previously (16). The low-dose aerogenic challenge was done 5 weeks after the final immunization. To measure the size of the challenge dose, five mice were sacrificed 24 h after challenge and the number of CFU/lung were determined as indicated before (5). The other groups of vaccinated and control mice were sacrificed by cervical dislocation 28 days after challenge.

Statistical analysis. Unpaired, two-tailed, t test statistical analysis was performed on CFU determinations from vaccinated and naive nonimmunized animals using a Microsoft Excel program on a Compaq personal computer.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Degradation of protein antigen expressed in vitro. Three pairs of DNA vaccines encoding chimeric ubiquitinated forms of the M. tuberculosis antigens ESAT-6 and MPT64, were constructed in order to obtain different degradation rates for these mycobacterial proteins (UbGAg, UbAAg, and UbGRAg [Fig. 1]). The effect of ubiquitination on the degradation rate of the MPT64 antigen was investigated by comparing protein concentrations in cells transfected with DNA vaccines expressing the tPA64 fusion protein or ubiquitin-MPT64 chimeric proteins. The DNA vaccines were transfected into RD cells, and the relative amount of antigen expressed in vitro was evaluated by immunoblot assay using a polyclonal antibody specific for MPT64. As shown in Fig. 2B, the amount of protein detected in the immunoblots ranged between high levels for tPA64 to almost undetectable levels for UbGR64. The intracellular protein concentrations were quantitated by densitometric analysis and expressed as a percentage of the most prevalent protein tPA64 (Fig. 2A). These measurements indicate that the relative concentrations of antigens expressed in transfected RD cells were as follows: tPA64 > UbG64 > UbA64 > UbGR64. The levels of protein present in culture supernatants from transfected cells were also evaluated. Considerable amounts of protein were detected in the supernatants of cells transfected with tPA64 and UbG64, while no signal was detected in the supernatants from UbA64 and UbGR64 transfected cells (data not shown).

View larger version (16K): [in this window] [in a new window]   FIG. 1.   Schematic showing the rational behind the construction of each ubiquitin fusion protein. The protease complex (PC) should cleave the UbG64 chimeric protein and release the mature antigen. The UbA64 construct exploits the reduced ability of the protease complex to cleave at position A76, leaving the intact UbA64 molecule a target for a rapid polyubiquitination. For the UbGR64 construct, the Ub conjugation is used to add an arginine to the N terminus of the mycobacterial antigen. This alteration destabilizes the protein and greatly enhances the degradation rate.

View larger version (27K): [in this window] [in a new window]   FIG. 2.   Expression of the different constructs encoding MPT64 in RD cells. The RD cells were transfected with plasmids encoding MPT64; 48 h after transfection the RD cell lysates were analyzed by immunoblot using a polyclonal antibody specific for MPT64. (A) The amount of MPT64 detected is expressed as the percentage of tPA64. (B) Immunoblot analysis of lysates from cells expressing tPA- and ubiquitin-conjugated MPT64 proteins.

Humoral responses to ubiquitinated MPT64. The humoral response induced by the different DNA vaccines was evaluated by analyzing sera collected from immunized mice 28 days after receiving the third vaccination. The total amount of IgG and the ratio of IgG isotypes induced (IgG1 and IgG2a) provide good indicators of the type of immune response generated in vivo. A strong humoral response with a prevalent IgG1 isotype was generally associated with a Th2-type immune response, while induction of a IgG2a isotype indicated a Th1-oriented immune response (26). Both tPA64 and UbG64 DNA vaccines induced a strong humoral response characterized by a mixed isotype pattern, with the first strongly polarized toward the IgG1 isotype (Table 1). The UbA64 DNA vaccine induced only a moderate humoral response, which was polarized toward the IgG2a isotype. Interestingly, no humoral response was detected in the sera of mice immunized with the UbGR64 DNA vaccine.

Cell-mediated immune responses. Recent studies have suggested that an effective TB vaccine depends on the induction of a strong Th1-type, T-cell-mediated response (14). To assess the intensity and the pattern of cell-mediated immunity induced by the ubiquitinated DNA vaccines, the numbers of splenic antigen-specific cells secreting IFN- and IL-4 were evaluated by ELISPOT assays. For these studies, the antigen-specific cytokine responses of naive mice and mice immunized with the DNA vaccines were compared with responses from BCG-vaccinated animals. In each of these experiments, the number of IFN--producing cells was virtually undetectable unless the cells were stimulated in vitro with specific antigen. In addition, very low numbers of antigen-stimulated IFN--producing cells were observed in the nonvaccinated and vector control groups. The background level of IL-4-secreting cells was higher but was consistent among the groups.

View larger version (22K): [in this window] [in a new window]   FIG. 3.   Cytokine secretion by spleen cells from vaccinated mice. Splenocytes from three mice per group were pooled and stimulated with in vitro recombinant MPT64, His-rMPT64, or with BCG crude lysate (BCG-vaccinated group). Samples were tested in triplicate, and levels are presented as IFN- (open bars) or IL-4 (shaded bars) SFU per million cells. In each group, the number of SFU obtained in the untreated cells stimulated with medium only was subtracted from the SFU detected in antigen-treated cells.

Protective responses induced by DNA vaccination. To evaluate the protective activity evoked by the different DNA vaccines, immunized and control mice were challenged aerogenically with approximately 50 CFU of M. tuberculosis Erdman. Each vaccine was tested in at least two separate protection experiments. Figure 4 shows that the bacterial burden within the lungs of mice receiving vector alone had increased 10⁴-fold 28 days after they received the small aerogenic challenge and the number of lung CFU detected were not different from the CFU numbers in naive mice. In contrast, mice vaccinated with the ubiquitin-fused forms of MPT64 had significantly fewer viable organisms in their lungs 28 days after challenge relative to the naive controls (Fig. 4). For example, a 72% reduction in lung CFU was detected for the UbGR64 group. However, the protective responses elicited by the Ub64 constructs were not statistically different from the responses evoked by the tPA64 vaccine.

View larger version (17K): [in this window] [in a new window]   FIG. 4.   The protective efficacy of DNA vaccines expressing ubiquitin-conjugated proteins in the mouse model of pulmonary TB. The reduction in CFU (log10) for vaccinated animals relative to naive controls is indicated above each bar when statistically significant (P < 0.05).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Protective immune responses against M. tuberculosis infection involve a cell-mediated rather than humoral response on the part of the host defenses. Although the precise mechanism of this anti-TB immunity has yet to be fully defined, the establishment of a Th1 phenotype with the subsequent production of IFN- seems to be a major component of the factors controlling the growth of M. tuberculosis in vivo (3, 20). Previous experiments involving DNA vaccination indicated that the presence of an eukaryotic signal sequence (tPA) fused at the N terminus of a mycobacterial protein enhanced the expression of the plasmid-encoded protein. For the tPA fusion proteins, large amounts of recombinant antigen produced in vitro seemed to correlate with a strong humoral response and moderate IFN- production in vaccinated mice (16, 17). In the present study, we evaluated a number of DNA vaccines expressing TB proteins fused at the N terminus with ubiquitin to determine whether these vaccines are more effective at inducing specific cellular immune responses against TB antigens. Conjugation of the antigen with ubiquitin should target the endogenously synthesized antigens to the proteasome, resulting in enhanced degradation of the TB proteins (23, 28, 34). Several studies suggest that cells of the nonlymphoid system are the major source of the DNA vaccine-encoded antigen (6, 9). The intact protein or even fragments of it, released by such nonlymphoid cells, would be taken up by the antigen-presenting cells, processed, and presented through both the major histocompatibility complex class I (MHC-I) and MHC-II molecules (10, 24, 25). In this model the intracellular turnover of an expressed protein within the transfected cells may have profound effects on the type of the immune response generated. Higher rates of intracellular antigen turnover should increase the number and variety of fragments and peptides available for MHC binding that may result in an increase of the cell-mediated response to the expressed antigens. Two strategies, based on the ubiquitin paradigm, were adapted to assess the impact of altered antigen degradation rates on the type of immune response generated. One approach involved generating stable ubiquitin-conjugated proteins (UbAAg) which should then be efficiently polyubiquinated. The other strategy (UbGRAg) relied on using ubiquitination to add an arginine to the N terminus of the mycobacterial fusion protein. This modification should substantially increase the turnover of the mycobacterial protein. The UbG64 construct was used as a control to assess the expression levels of the ubiquitin-conjugated proteins and to better evaluate the degradation rate of the UbA- and UbGR-cotranslated antigens. In fact, this construct should not target the mycobacterial protein to rapid degradation.

Results from the in vitro transfection experiments indicate that the strategies described above were appropriate. The high level of expression for the UbG64 antigen demonstrated that ubiquitin conjugation can provide excellent protein expression in this system. The substitution of an A₇₆ for the G₇₆ residue resulted in a significant reduction of intact antigen detected in RD cell lysates, probably due to the inhibition of cleavage at position 76, which increased polyubiquitination, and targeting of the antigen to the proteasome. The UbGR64 antigen was quickly degraded, probably because cleavage of the UbG component left the MPT64 with a very destabilizing arginine residue at the N-terminal end. It should be emphasized that the rate of protein degradation for UbGR constructs seems to be antigen dependent. For example, Wu and Kipps showed that the -galactosidase is quickly targeted for degradation when expressed as a UbGR chimera (34). In contrast, Fu, et al. (8) reported that the in vitro stability of the influenza nuclear protein was not affected by UbGR conjugation. Moreover, we have recently found using the in vitro RD transfection assay (G. Delogu and S. L. Morris, unpublished results) that only UbA and not UbGR conjugation increased the intracellular degradation rate of MTB12, another M. tuberculosis protein (31). Although the physical factors which dictate the stability of these UbGR chimeras have not been completely clarified, it is likely that the turnover rate is affected by the presence of a lysine residue at the N terminus, which must be available to allow ubiquitin binding (29). Because of the differential stability of UbGR-fused antigens, an initial assessment of protein turnover in cell culture should be performed with any new constructs to permit a rational interpretation of subsequent immunoreactivity data.

The different rates of protein turnover suggested by the in vitro assays for the MPT64 fusion proteins were reflected in the various immune responses to DNA immunization. The constructs expressing relatively stable antigens (tPA64 and UbG64) induced substantial humoral responses and only moderate levels of IFN-. Expression of the less-stable UbA64 fusion protein after DNA vaccination yielded a weaker, but Th1-polarized, humoral response and substantial cytokine production (Fig. 3). In contrast, immunization with the plasmid that encoded the UbGR64 chimera generated a very robust IFN- response but no anti-MPT64 antibodies (Table 1). The complete abrogation of the humoral response to this chimera suggests that the R64 protein is rapidly and completely degraded intracellularly, leaving insufficient intact protein to interact with the B cells. The elevated IFN- response to UbGR64 also suggests that targeting the endogenously synthesized Ub fusion protein to the proteasome for cleavage resulted in effective antigen presentation, which was able to induce a highly polarized Th1 type immune response compared to the tPA64 construct.

Using the mouse model of pulmonary TB, we demonstrated that the constructs expressing the Ub conjugates and the tPA fusion proteins elicited an enhanced level of resistance compared to vector controls. Both the UbA64 and UbGR64 DNA vaccines and the UbAES6 and UbGRES6 constructs evoked similar levels of resistance compared to that evoked by their tPA vaccine counterpart. These data suggest that the expression of other TB antigens as UbA or UbGR fusion proteins may yield protective immunity against M. tuberculosis challenge. For both UbA64 and UbGR64 constructs, a predominant Th1 type immunity was established. Surprisingly, despite the much higher levels of IFN- produced in vitro, UbGR64 did not provide a better immune response in vivo than that achieved by UbA64. This result seems to support a recent report that in vitro IFN- stimulation assays may be poorly predictive of the in vivo response developed by the host defenses (13). Alternatively, our data could indicate that IFN- production is necessary, but not sufficient in itself, for complete protection against this intracellular pathogen. We had reported earlier that the protective response observed might not correlate directly with the level of IFN- production generated after immunization (16). It is likely that IFN-, as well as other as-yet-undefined immunomodulators, must be induced by the vaccine in order to achieve improved protection. Long-term studies designed to investigate whether an increase in protective immunity is associated with one particular construct will be needed before dissecting the specific T-cell populations involved in the protective immune response. In a viral system where cytotoxic T lymphocyte (CTL) induction is a correlate of protective immunity, DNA vaccines expressing antigens cotranslated with ubiquitin have been used to increase the CD8-mediated CTL response (23, 28). The development of Th1 type immunity is considered the major factor leading to the establishment of a protective immunity against TB, while the specific role of CTL responses remains to be fully elucidated (14). For this reason, we did not use methods to specifically assess the CTL response in these studies.

Despite the promising results, immunization with constructs expressing single antigens did not elicit protective responses that exceeded the response afforded by BCG. Vaccination with plasmids expressing multiple epitopes, either as combinations of different plasmids or in the minigene form (12), will probably be needed to further improve the protective immunity evoked. We have recently demonstrated that immunization with a combination of four single DNA vaccines expressing M. tuberculosis antigens as tPA fusion proteins enhanced the anti-TB response relative to that observed with single constructs (18). Since our studies suggest that DNA vaccines encoding mycobacterial proteins cotranslated with ubiquitin modulate the immune response, other constructs expressing Ub-conjugated proteins should be evaluated as combination vaccines. Although the short-term assays have shown that the immunity elicited by the Ub fusion and the tPA fusion vaccines is similar, the constructs expressing Ub chimeras may offer long-term advantages. Orme and coworkers have shown that the continuing lung granulomatous response to the infection, and not the increasing bacterial growth, is responsible for the death of aerogenically challenged animals with M. tuberculosis (1). A dampening of the immunological burden that takes place in the lung during chronic infection may lengthen the survival times for the vaccinated animals. The UbGR64 construct induces a cell-mediated rather than a humoral response, resulting in a partial control of the M. tuberculosis growth in vivo compared to control mice. Thus, the induction of an enhanced cell-mediated immunity in the absence of a humoral response by DNA vaccine expressing UbGR-conjugated mycobacterial antigens may limit the tissue damage associated with the inflammatory responses in the lung (21) and may ultimately prolong the survival of the challenged animals. For this reason, study of constructs expressing ubiquitin conjugates should continue in the search for new improved anti-TB vaccines.