INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The World Health Organization estimates that each year there are approximately 300 to 500 million new cases of malaria and 1.5 to 2.7 million deaths due to this disease. Most of the deaths occur in African children, but malaria remains a major health and economic problem in the tropics and subtropics of almost all underdeveloped countries. In view of the spread of drug resistance and the resistance of the mosquito vectors to insecticides, many countries of the world in which malaria is endemic have been experiencing a deterioration of the malaria situation in the last two decades (31). The availability of a cost-effective vaccine would be a valuable asset to any malaria control initiative (36). To develop a vaccine against malaria, several problems have to be taken into consideration. The identification of protective immune mechanisms capable of eliminating the parasite, identification of antigens and epitopes able to stimulate those immune mechanisms, and the selection of a vaccine delivery system able to present the antigen to the immune system in order to generate these protective immune responses are some of the key issues (14).

Immunization of mice with radiation-attenuated sporozoites induces complete protection against sporozoite challenge, and the protective role of CD8⁺ T cells and cytokines in this model has been clearly demonstrated (10, 20). DNA vaccines have provided an antigen delivery system able to generate the mechanisms of immunity that confer protection against the challenge of mice with sporozoites. They also provide a method for developing multivalent, multi-immune response vaccines (8). A DNA vaccine expressing the Plasmodium yoelii circumsporozoite protein (PyCSP) protects BALB/c mice against sporozoite challenge (25). The genetic restriction of protection elicited by this vaccine can be overcome by combining it with other antigens (9). The protective efficacy of the PyCSP DNA vaccine was further improved by coadministering it with a plasmid encoding the murine granulocyte-macrophage colony-stimulating factor (pmurGM-CSF) (35) or by boosting the PyCSP DNA vaccine-generated immune responses with a recombinant vaccinia virus expressing PyCSP (26). Combining GM-CSF and a recombinant poxvirus boost increases immunogenicity and protection more than either intervention alone (27). Incorporation of more target antigens/epitopes to protect a genetically heterogeneous population in areas where malaria is endemic requires the identification and evaluation of the protective capacity of new antigens.

Heat shock proteins (Hsp) are a highly immunogenic and conserved family of proteins that have been identified as prominent antigens in the immune response to a wide variety of infections (38). They are able to transfer exogenous peptides to major histocompatibility complex class I molecules and prime cytotoxic T lymphocytes (29), induce the expression of cytokines and adhesion molecules, induce antigen-specific CD8⁺ and/or CD4⁺ T cells, and mediate antigen-specific protection against several microorganisms (21). Plasmodium spp. Hsp60s are expressed in all the parasite development stages that occur in the vertebrate host (6, 24, 30). T cells obtained from irradiated sporozoite-vaccinated mice that proliferate in the presence of Mycobacterium tuberculosis Hsp60 have been shown to protect naive mice against sporozoite challenge (33), and T cells elicited during a P. yoelii infection respond by in vitro proliferation to Plasmodium falciparum Hsp60 (PfHsp60) and Hsp70 (15).

We recently cloned and sequenced the gene encoding P. yoelii heat shock protein 60 (PyHsp60) (24). Here we describe the evaluation of the immunogenicity and protective efficacy of a DNA vaccine that contains the PyHsp60 alone or in combination with a plasmid expressing murine GM-CSF.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Experiments reported here were conducted according to the principles set forth in the Guide for the Care and Use of Laboratory Animals, Institute of Laboratory Animal Resources, National Research Council (Department of Health and Human Services, National Institutes of Health, publication no. 86-23).

Mice and parasites. The mice used in this study were 5- to 6-week-old female BALB/cByJ mice (Jackson Laboratory, Bar Harbor, Maine). In all the experiments P. yoelii (17X NL nonlethal strain, clone 1.1) parasites were used (34). For immunization with irradiated sporozoites, sporozoites were obtained by the discontinuous gradient technique from infected mosquitoes that had been irradiated at 10 kilorads (¹³⁷Ce) 14 days after an infectious blood meal (23). For challenges, sporozoites were obtained by hand dissection of infected mosquito glands in M199 medium containing 5% normal mouse serum.

Construction and immunogenicity of DNA plasmids. In a previous report (24) we described the cloning and complete sequence of a PyHsp60 cDNA clone. Based on this sequence, a primer containing the 5' end of the coding region with a BamHI site and a primer containing the 3' end and a BglII site were used to amplify the PyHsp60 gene from the cDNA clone, using the Gene Amp PCR reagent kit (Perkin-Elmer, Norwalk, Conn.). The PCR product was cloned into the pCR-Script Amp SK (+) cloning vector (Stratagene, La Jolla, Calif.), was sequenced using the ABI PRISM dye terminator cycle sequencing Ready Reaction kit (Perkin-Elmer) following the manufacturer's instructions, and was subcloned into the DNA vaccine plasmids VR1012 (11) and VR1020 (17) digested with BamHI and BglII. The DNA insert was sequenced, and expression of proteins encoded by the pPyHsp60-VR1012 and pPyHsp60-VR1020 plasmids was evaluated by in vitro transfection of human melanoma cells (UM449) as previously described (12). Immunoblot analysis of transfected cell lysates was done using antisera obtained from CD-1 mice immunized with pPyHsp60-VR1012 (24) or with recombinant P. falciparum Hsp60 (6). Construction and in vitro expression of plasmids containing the genes for the P. yoelii circumsporozoite protein (pPyCSP-VR1020) (26), the P. yoelii 17-kDa hepatocyte erythrocyte protein (pPyHep17-VR1012) (7), and pmurGM-CSF (35) have been described.

Immunization regimen. Two groups of mice were used as positive controls in the present study. One group received 5 × 10⁴ irradiated P. yoelii sporozoites by intravenous (i.v.) injection in the tail vein, followed by two i.v. doses of 3 × 10⁴ irradiated sporozoites 4 and 9 weeks after the first immunization. The other group was immunized with a mixture of three plasmids, pPyCSPVR1020, pPyHep17-VR1012, and pmurGM-CSF. Negative control mice were immunized with an unmodified VR1012 or VR1020 DNA plasmid alone or in combination with the pmurGM-CSF plasmid. Experimental groups included mice immunized with pPyHsp60-VR1012 or pPyHsp60-VR1020 administered alone or in combination with the pmurGM-CSF plasmid. All groups of mice immunized with DNA vaccines received a total of three doses, at 0, 3, and 9 weeks. Each dose consisted of 100 µg of plasmid DNA dissolved in sterile saline solution in a final volume of 100 µl. When plasmids were used in combination with the pmurGM-CSF plasmid, 30 µg of this DNA was combined in the same final volume. Immunizations were done by administering 50 µl of the solution into each thigh intramuscularly (i.m.) with a 30-gauge needle. In a separate experiment, 7- to 8-week-old female BALB/c mice were immunized with pPyHsps60-VR1012 DNA vaccine followed 2 weeks later by a booster immunization with either the DNA vaccine or 2 × 10³ irradiated sporozoites.

IFAT and ILSDA. Antibody responses in sera collected prior to immunization and 2 weeks after the second and third doses were evaluated by the indirect fluorescent antibody test (IFAT) using air-dried P. yoelii sporozoites and P. yoelii-infected erythrocytes as described before (3). The capacity of sera from immunized mice to inhibit sporozoite invasion into the hepatocytes was assessed by the inhibition of liver-stage development assay (ILSDA) as previously described (19). Briefly, mouse hepatocytes were seeded onto eight-well chamber slides and cultured overnight. The next day, media were aspirated from the wells and 25 µl of the serum samples, diluted 1:2, was added to triplicate wells. P. yoelii sporozoites (75,000/well) in a volume of 25 µl were added to hepatocyte cultures and incubated for 3 h, and cultures were washed extensively to remove unattached sporozoites. The infected hepatocytes were incubated for a further 2 days, fixed with methanol, and immunostained with NYLS3 (a P. yoelii-specific monoclonal antibody) and fluorescein-conjugated goat anti-mouse immunoglobulin G (IgG). The number of schizonts was counted from the triplicate cultures, and the percent inhibition was determined as follows: ([mean control  mean test]/mean control) × 100. Sera from mice immunized with unmodified VR1012 or VR1020 plasmids were used as negative controls.

T-cell depletions. For depletion of CD8⁺ T cells, mice were treated with the anti-CD8 monoclonal antibody (MAb) 2.43. On days 6, 5, 4, 2, 1, 0, and day +2 relative to sporozoite challenge, mice were injected intraperitoneally with 0.5 mg of MAb 2.43 in 0.5 ml of phosphate-buffered saline. Undepleted mice received rat IgG using the same schedule and doses used for depletion of CD8⁺ T cells. Cells obtained from mice treated with MAb 2.43 or rat IgG were stained with anti-CD8⁺ fluorescein isothyiocyanate (PharMingen, San Diego, Calif.), and the success of depletion was determined by single-color fluorescence-activated cell sorting using the FACScan (Fax 400 Royal, Beckton Dickinson Immunocytometry Systems, San Jose, Calif.).

Sporozoite challenge and protection against blood-stage infection. Two weeks after the last immunization, mice were challenged by injection of 50 sporozoites of the nonlethal strain of P. yoelii (17XNL clone 1.1). Sporozoites were injected i.v. by tail vein in a volume of 200 µl. Blood smears were obtained every day from day 4 until day 14 after challenge and were examined for the presence of parasites (Giemsa stain). Complete protection was defined as the absence of P. yoelii parasites in the blood on all days including day 14. In order to determine whether immunization of mice with the PyHsp60 DNA vaccine had any effect on the course of blood-stage infection, percent parasitemias of mice were obtained by counting the number of infected red blood cells among 5,000 erythrocytes in Giemsa-stained blood smears.

Sporozoite challenge and protection against liver-stage infection. To assess protection against liver-stage infection, the parasite burden in the liver was measured using the Taqman (Applied Biosystems, Foster City, Calif.) automated real-time PCR system using the P. yoelii-specific 18S rRNA and the mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as target sequences (37). Briefly, 2 weeks after the last immunization mice were challenged with 5 × 10⁴ sporozoites purified by the discontinuous gradient technique from mosquitoes 14 days after an infectious blood meal. Forty-two hours after challenge, livers were recovered from each mouse, were stored in RNAlater solution (Ambion Inc., Austin, Tex.), and were kept at 4°C for a maximum of 1 week. Livers were disrupted with a polytron homogenizer (Omni International), and total RNA was purified using Trizol LS reagent (Life Technologies, Gaithersburg, Md.). Purified RNA was reverse transcribed, and target sequences of the P. yoelii 18S rRNA and mouse GAPDH were amplified simultaneously using specific primers and fluorescence-labeled probes. The threshold cycle, defined as the cycle at which the fluorescence exceeds 10 standard deviations above the starting fluorescence in the system, was converted to DNA equivalents by the amplification of the respective target sequences cloned into plasmids DNA and standard curves generated by 10-fold serial dilutions of these plasmids. The parasite burden was determined by calculating the ratio of the equivalence measures of the cloned target sequences obtained for the P. yoelii rRNA over the measures obtained for the GAPDH.

Statistical analysis. The ² test or two-tailed Fisher's exact test was used to evaluate if there was any significant protection in the immunized animals compared with control groups. (Epi Info, version 6.04b, Centers for Disease Control and Prevention, Atlanta, Ga.). Mean parasitemias in groups of mice over several days of follow-up were compared using repeated-measure analysis of variance with the Scheffe test for multiple post hoc comparisons (SPSS for Windows, version 8.0; SPSS Inc., Chicago, Ill.).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Characterization of PyHsp60 constructs. The full-length sequences of the PyHsp60 insert and the cloning junctions were verified. The PCR-amplified product in the PCR-Script vector as well as the junctions at the cloning site of the mammalian expression vector showed the expected sequences. Characterization of antisera obtained from CD-1 mice immunized with the pPyhsp60-VR1012 construct has been described (24). In Western blot analyses of lysates of untransfected as well as pPyHsp60-VR1012-transfected UM449 cells, these anti-PyHsp-60 antisera reacted with a band of approximately 60 kDa (data not shown). The minimum identity of the sequences observed between any two proteins of the Hsp60 family is about 40%. It is likely that cross-reaction of the anti-PyHsp60 polyclonal antibodies with human Hsp60 accounts for the reaction with the untransfected cells. Sera from mice vaccinated with unmodified VR1012 vector as well as sera from mice obtained previous to the immunizations were negative by IFAT and Western blot (data not shown). These results and the observation that mice immunized with the pPyHsp60 DNA construct produced antibodies that recognized Hsp60 from several sources (24) led us to conclude that the vaccine plasmids were expressing the PyHsp60 gene.

Antibody titers (IFAT) and ILSDA. In order to determine the immunogenicity of the pPyHsp60-VR1012 and pPyHsp60-VR1020 DNA vaccines in BALB/c mice, sera from immunized mice were evaluated by IFAT against P. yoelii sporozoite and blood-stage parasites. Sera obtained from mice before the injection of DNA vaccines (data not shown) or from mice immunized with the DNA plasmid control (groups 1.H and 1.I, Table 1) did not show any significant reaction by IFAT against P. yoelii sporozoite or blood-stage parasites at serum dilutions of 1:20. Immunization of mice with the plasmids pPyHsp60-VR1012 and pPyHsp60-VR1020 induced the production of antibodies that reacted by IFAT with sporozoites and blood-stage parasites after the second and third immunizations (groups 1.A, 1.B, 1.C, 1.D, and 1.E, Table 1). The magnitude of this immune response against sporozoites was less than that induced by immunization with the positive control vaccine (PyCSP and PyHep17 plasmid mixture) (group 1.G, Table 1). The PyHsp60 vaccine-elicited antibodies also showed strong reactivity against P. yoelii-infected red blood cells. The addition of GM-CSF plasmid to the PyHsp60 DNA vaccines induced only a modest increase in the antibody titers as observed by IFAT (groups 1.B, 1.C, and 1.D) after the third immunization. By Western blot, the antiserum produced by immunization with pPyHsp60-VR1012 in combination with pmurGM-CSF recognized mycobacterial Hsp60, recombinant PfHsp60, and a band of around 60 kDa in sporozoites and blood-stage parasites (data not shown). To determine whether the antibodies induced in mice immunized with the PyHsp60 DNA vaccine in combination with pmurGM-CSF plasmid were able to inhibit the invasion of sporozoites into the hepatocytes, sera from mice immunized with pPyHsp60 alone or in combination with pmurGM-CSF were tested by ILSDA. In this assay the MAb NYS1 that recognizes the PyCSP on the surface of sporozoites (3, 4) and the MAb NYSL3 (5) that reacts with PyHep17, which is located on the parasitophorous vacuole membrane of the liver and blood stages of the parasite life cycle, consistently inhibit invasion or development of the parasite in hepatocytes by greater than 90%. Indeed, in the experiment conducted in the present study, NYS1 showed a 93% inhibition of sporozoite invasion. In contrast, none of the sera obtained from mice immunized with the PyHsp60-based DNA vaccines alone or in combination with pmurGM-CSF were able to inhibit invasion of sporozoites into the hepatocyte at the dilutions tested (Fig. 1).

View this table: [in this window] [in a new window]   TABLE 1.   Experiment no. 1immunization regimen, antibody titers, and protective efficacy of PyHsp60 DNA vaccines

View larger version (15K): [in this window] [in a new window]   FIG. 1.   ILSDA. Sera with the highest titer of antibody from mice immunized with pPyHsp60-VR1012 DNA vaccine alone or in combination with pmurGM-CSF were pooled and tested in the assay. The percent inhibition was calculated using the mean of the schizonts counted on triplicate cultures and determined as follows: (mean of control  mean of test)/mean control × 100.

Protection after sporozoite challenge. We first evaluated protective efficacy in mice immunized with the pPyHsp60-VR1012 and pPyHsp60-VR1020 DNA vaccines alone or in combination with pmurGM-CSF DNA. Mice were considered protected if parasites were not observed in the blood on day 14 as evaluated by Giemsa-stained blood films. In this experiment, 100% of the mice that were immunized with irradiated sporozoites (group 1.G, Table 1) and 90% of the mice immunized with the combination pPyCSPVR1020, pPyHep-17-VR1012, and pmurGM-CSF (group 1.F, Table 1) did not develop parasitemia during the 14 days of follow-up and were considered protected. Ninety percent of mice in the negative control group 1.H that were immunized with unmodified pVR1020 DNA alone (Table 1) and all the mice in the negative control group 1.I immunized with VR1020 in combination with pmurGM-CSF developed parasitemia by day 14 postchallenge. Because there did not appear to be a significant difference in the behavior of the two negative control groups, in subsequent analysis we pooled them. In contrast, 40% of mice immunized with the combination of pPyHsp60-VR1012 and pmurGM-CSF did not develop parasitemia during the 14 days postchallenge (group 1.B, Table 1). Only this group had statistically significant protection on day 14 as compared with the pooled controls (two-tailed Fisher's exact test: P = 0.031, group 1.B versus group 1.H + group 1.I [Table 1]). Only 10% of mice in groups 1.A, 1.C, and 1.D that were immunized with the other PyHsp60-based DNA vaccine alone or in combination with pmurGM-CSF plasmid did not develop parasitemia. Only 10% of mice that were depleted of CD8⁺ T cells did not develop parasitemia. However, this difference was not statistically significant when compared with immunized undepleted mice (Fisher's exact test: P > 0.05, group 1.B versus group 1.C [Table 1]).

View larger version (17K): [in this window] [in a new window]   FIG. 2.   Geometric mean parasitemias of BALB/c mice immunized with the PyHsp60 DNA vaccines (experiment no. 1). As described in Table 1, mice were challenged with 50 P. yoelii 17X (NL) sporozoites. Percent parasitemia was calculated by counting the number of infected red blood cells among 5,000 erythrocytes on Giemsa-stained blood smears from each mouse. Only mice which developed parasitemia were included in the analysis.

View this table: [in this window] [in a new window]   TABLE 2.   Experiment no. 2immunization regimen and protective efficacy of PyHsp60DNA vaccine

View larger version (20K): [in this window] [in a new window]   FIG. 3.   Geometric mean parasitemias of BALB/c mice immunized with the PyHsp60 DNA vaccines (experiment no. 2). As described in Table 2, mice were challenged with 50 P. yoelii 17X (NL) sporozoites. Percent parasitemia was calculated by counting the number of infected red blood cells among 5,000 erythrocytes on Giemsa-stained blood smears from each mouse. Only mice which developed parasitemia were included in the analysis.

Boosting of pPyHsp60-VR1012 DNA with irradiated sporozoites and protection against liver-stage infection. We observed in the first of the two experiments described above that immunization with the pPyHsp60-VR1012 DNA vaccine used in combination with the plasmid pmurGM-CSF resulted in 40% protection. On the other hand, in the second experiment there was no protection on day 14 although there was a delay in the onset of parasitemia in the groups that had received the PyHsp60 plasmid DNA. It seemed from the first two experiments that a PyHsp60 vaccine administered as a DNA plasmid may have induced immune responses with protective activity against the preerythrocytic stages of P. yoelii but that this immunity was not as protective as that elicited by immunization with the combination of PyCSP and PyHEP17 DNA vaccines with the plasmid expressing murine GM-CSF. Since mice immunized with PyHsp60 DNA vaccine produced antibodies against sporozoites, we speculated that the PyHsp60 DNA vaccine might prime for boosting with irradiated sporozoites. Furthermore, since protection against blood-stage parasitemia is an all or none phenomenon that does not allow us to detect 90% of protective activity against infected hepatocytes, we utilized a Taqman assay developed to detect liver-stage P. yoelii ribosomal RNA as our endpoint measure (37). Table 3 shows the experimental design of this experiment. In Fig. 4 we report the mean ratios of the measures of amplified P. yoelii 18S rRNA over amplified mouse GADPH obtained from the livers of all immunized mice. Parasite rRNA was not detected in mice that received a first dose of the combination of pPyCSP-VR1020, pPyHep17-VR1012, and GM-CSF and were boosted with irradiated sporozoites. The maximum measurement of rRNA was obtained in the group of mice that received two doses of unmodified VR1012 control plasmid or in those that received two doses of pPyHsp60-VR1012 plasmid. A 75% reduction of parasite burden was observed in the mice that received either PyHsp60-VR1012 or VR1012 negative control plasmid as a first dose and irradiated sporozoites as the second dose. The results show that in contrast to immunization with the combination pPyCSP-VR1020, pPyHep17-VR1012, and pmurGM-CSF, immunization with pPyHsp60-VR1012 DNA vaccine neither diminished the parasite burden in the liver on its own nor primed immunized mice for a greater protective response to a single dose of irradiated sporozoites.

View this table: [in this window] [in a new window]   TABLE 3.   Boosting of pPyHsp60-VR1012 with irradiated sporozoitesimmunization regimen

View larger version (29K): [in this window] [in a new window]   FIG. 4.   Liver-stage infection burden. Mean ratios of DNA equivalent measures for the P. yoelii 18S rRNA over the DNA equivalent for murine GAPDH are shown. Groups of mice were immunized with DNA vaccines and were boosted with irradiated sporozoites (as described in Table 3), and 2 weeks after the last immunization mice were challenged by intravenous injection of 5 × 104 sporozoites. Forty-two hours later livers were recovered and RNA was extracted. The analysis included triplicate Taqman measures from livers recovered from three mice per group.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study we evaluated the immunogenicity and protective efficacy of pPyHsp60-VR1012 and pPyHsp60-VR1020 DNA vaccines alone or in combination with pmurGMCSF plasmid and compared them with the protection against sporozoite challenge afforded by immunization with the combination pPyCSP-VR1020, pPyHep17-VR1012, and pmurGM-CSF. As a negative control in all experiments, we used mice injected with the unmodified DNA vaccine vector alone or in combination with the pmurGM-CSF plasmid. Both pPyHsp60-VR1012 and pPyHsp60-VR1020 DNA vaccines were immunogenic in BALB/c mice, inducing antibodies that recognized P. yoelii parasites by IFAT. The levels of antibodies induced against blood-stage parasites were higher than those induced against sporozoites. It has been shown that expression of PfHsp70 is not observed (16) or is observed at a very low level in P. falciparum sporozoites (32). The low antisporozoite antibody titer observed in mice immunized with the PyHsp60 DNA vaccines may reflect a lower level of expression of this protein in this stage of the parasite compared with the blood stage. An intriguing result from the present study is the suggestion that immunization with PyHsp60 may be capable of inducing a degree of protection against sporozoite challenge, although the findings are inconsistent between different challenge experiments. Immunization of mice with the plasmid pPyHsp60-VR1012 in combination with pmurGM-CSF revealed a statistically significant 40% protection of BALB/c mice in one of the experiments (two-tailed Fisher's exact test: P = 0.031, group 1.B versus groups 1.H + 1.I [Table 1]) and only a delay in the onset of blood-stage parasites in 40 to 50% of the mice immunized with the PyHsp60-VR1012 DNA vaccine in the second experiment (two-tailed ² test: P = 0.0634, groups 2.A + 2.C versus 2.F + 2.G, day 4 [Table 2]). Some of the best protection obtained against challenge with P. yoelii sporozoites in BALB/c mice is conferred by immunization with PyCSP in combination with pmurGM-CSF for the first dose and boosting 3 weeks later with recombinant vaccinia virus expressing PyCSP (27). Under these conditions approximately 80% of the mice are protected. On the other hand, protection of BALB/c mice against sporozoite challenge ranging from 25 to 75% has been consistently obtained by immunizing with the PyCSP DNA vaccine alone (9, 26, 35). An attractive explanation for the variability in the level of protection from experiment to experiment may be the variability in sporozoite viability and/or infectivity between sporozoites isolated from different batches of infected mosquitoes. As demonstrated by McKenna et al. (18), the liver parasite burden in groups of mice infected with P. yoelii sporozoites purified from different batches of infected mosquitoes varies dramatically. One approach to asking whether there has been experiment-to-experiment variation in the infectivity of sporozoites is to consider the level of protection achieved with the positive control immunization regimen of PyCSP, PyHEP17, and pmurGM-CSF. Indeed, in the first experiment this regimen provided 100% protection, while in the second experiment it provided only 40% protection, suggesting that the challenge sporozoites in the second experiment were more infective.

In the third challenge study, immunizations with either the experimental vaccine or the protective DNA vaccine mix were boosted with irradiated sporozoites and the liver parasite burden was used as a read-out for the effect of the regimen on the intrahepatocytic growth of the parasites. Priming immunization with the PyHsp60 followed by boosting with irradiated sporozoites, however, did not have any effect on the liver parasite burden that was evaluated 42 h after the injection of 5 × 10⁴ sporozoites. In contrast, immunization with the protective combination in the first dose and boosting with irradiated sporozoites completely abolished the infection of parasites in the infected hepatocyte. In this case it seems likely that whatever potential protective effect of immunization with PyHsp60 was present was overwhelmed by the large challenge dose.

Since PyHsp60 is a protein expressed in all the parasite stages found in the vertebrate host, we assessed whether immune responses elicited by immunization with PyHsp60 DNA had antiparasite effects at different stages of the life cycle. Data from the ILSDA indicated that the antibodies induced did not prevent sporozoite invasion of hepatocytes or inhibit the growth of parasites developing within hepatocytes. Assessment of the course of parasitemias in immunized mice provided no evidence for an anti-infected erythrocyte effect. However, the protection in the first experiment, the delay in onset of parasitemia in the second experiment, and the slightly lower parasitemias early in infection (Fig. 3 and 4) suggested a pre-erythrocytic stage effect. Since antibodies apparently had no effect, this suggests that T-cell responses against the infected hepatocyte may have been responsible for this effect. The suggestion in experiment no. 1 that CD8⁺ T-cell depletion reduced the limited protection observed, although not statistically significant, is consistent with a role for T-cell responses in protection.

In summary, we have suggestive evidence that immunization with PyHsp60 can induce a partially protective immune response against challenge with P. yoelii sporozoites. The magnitude of the response is small and it is easily overwhelmed by large doses of sporozoites and even by variation in the infectivity of different sporozoite preparations. Nonetheless, the findings suggest that PyHsp60 may still have relevance as a vaccine target. It is certainly possible that, as in the case of the PyCSP DNA vaccine, optimal protective efficacy will be obtained only by boosting the DNA vaccine immunization with recombinant poxvirus encoding PyHsp60. In order to fully evaluate the potential of PyHsp60 as a vaccine antigen it will be important to construct and characterize recombinant poxvirus expressing this protein and to evaluate it in prime-boost experiments.

In other animal models, it has been shown that Hsp60 is a prominent immunogen able to induce a variety of immune responses or act as a carrier when administered with other antigens (2, 22, 28). In the P. yoelii-mouse model, T cells elicited by immunization with irradiated sporozoites or elicited by blood-stage infection proliferate in the presence of Hsp60 (15, 33). Furthermore, Hsp60-specific ⁺ T cells elicited by immunization with irradiated sporozoites seem to contribute to the decrease of liver parasite burden in this model (33). In the present preliminary study we have produced DNA vaccines based on PyHsp60 that were immunogenic in BALB/c mice. We also demonstrated that this vaccine did seem to have an effect on the parasites in mice challenged with 50 sporozoites. Given the preexisting literature on the potential of Hsp60 as a vaccine target and our results suggesting a modest protective effect in the P. yoelii model, we conclude that PyHsp60 is an interesting candidate vaccine antigen and that further experiments using more immunogenic vaccination regimens, such as DNA-virus prime-boost approaches, should be carried out to further evaluate the usefulness of PyHsp60 as a vaccine candidate. Additionally, it has been demonstrated that immunization of mice with a recombinant Hsp60 in fusion with an influenza virus protein efficiently primes CD8⁺ cytotoxic T lymphocytes against this viral antigen (1). Our long-term goal in the design of a DNA vaccine against malaria is to induce different mechanisms of immunity in the heterogeneous population able to recognize and eliminate the different stages of the parasite life cycle (13). The potential of PyHsp60 DNA vaccines as an adjuvant for use in combination with other malaria antigens should also be explored.