JOURNAL OF VIROLOGY, Nov. 2003, p. 11859–11866 0022-538X/03/$08.00⫹0 DOI: 10.1128/JVI.77.21.11859–11866.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Vol. 77, No. 21
Induction of Protective Immunity against Malaria by Priming-Boosting Immunization with Recombinant Cold-Adapted Influenza and Modified Vaccinia Ankara Viruses Expressing a CD8⫹-T-Cell Epitope Derived from the Circumsporozoite Protein of Plasmodium yoelii Gloria Gonza´lez-Aseguinolaza,1 Yurie Nakaya,2 Alberto Molano,1 Edward Dy,1 Mariano Esteban,3 Dolores Rodríguez,3 Juan Ramo ´n Rodríguez,3† Peter Palese,2 Adolfo García-Sastre,2* and Ruth S. Nussenzweig1 Department of Medical & Molecular Parasitology, NYU School of Medicine,1 and Department of Microbiology, Mount Sinai School of Medicine,2 New York, New York, and Department of Molecular and Cellular Biology, Centro Nacional de Biotecnología, CSIC, Madrid, Spain3 Received 13 June 2003/Accepted 28 July 2003
We immunized mice with an attenuated (cold-adapted) influenza virus followed by an attenuated vaccinia virus (modified vaccinia virus Ankara), both expressing a CD8ⴙ-T-cell epitope derived from malaria sporozoites. This vaccination regimen elicited high levels of protection against malaria. This is the first time that the vaccine efficacy of a recombinant cold-adapted influenza virus vector expressing a foreign antigen has been evaluated.
Malaria remains a major cause of morbidity in tropical and subtropical areas of the five continents, with an estimated 350 to 500 million individuals displaying clinical manifestations each year. In addition, the increasing drug resistance of the two most prevalent severe malaria species infecting humans, Plasmodium falciparum and Plasmodium vivax, makes it important to develop effective and long-lasting antimalaria vaccines. The feasibility of a prophylactic malaria vaccine for humans has been demonstrated with volunteers immunized with irradiated sporozoites of P. falciparum and P. vivax. This immunization protocol resulted in complete protection from disease after challenge of vaccinees by the bite of laboratory-bred mosquitoes infected with viable sporozoites (10). This proof of principle, obtained in the late 1970s, was corroborated by various investigators (9, 23). In 1976 we described the circumsporozoite (CS) protein, which covers the entire sporozoite surface with a thick coat (11). Injection, into mice, of minimal amounts of an anti-CS monoclonal antibody directed against the CS repeats of Plasmodium berghei (its major B cell epitope) abolished sporozoite infectivity (32). The CS protein is expressed not only by sporozoites but also by parasite-infected hepatocytes. Immunization of mice with irradiated sporozoites elicits CS-specific CD8⫹ T cells, which mediate the inhibition of development of the parasite’s liver stages (29). These antigen-specific CD8⫹ T cells can also be induced by immunizations with several recombinant viruses, which, by entering the cytoplasms of antigen-
* Corresponding author. Mailing address: Department of Microbiology, Box 1124, Mount Sinai School of Medicine, 1 Gustave L. Levy Pl., New York, NY 10029. Phone: (212) 241-7769. Fax: (212) 534-1684. E-mail: [email protected]
† Present address: Bionostra, Tres Cantos, 28760 Madrid, Spain.
presenting cells, induce “foreign” antigen processing and presentation by the class I pathway (18). Priming-boosting immunization approaches using two different vectors expressing the same antigen have demonstrated great potential as vaccination strategies, resulting in the induction of potent immune responses against malaria (5, 15, 25–27, 30, 33) and other infectious diseases, including AIDS (1, 16, 20, 28). A number of earlier experiments pioneered the use of different vectors in priming-boosting regimens of immunization. Specifically, mice were primed with a recombinant influenza virus and given boosters with a recombinant vaccinia virus; both viruses expressed sequences of the CS protein of Plasmodium yoelii (18). This regimen of immunization elicited a high degree of protection against sporozoite challenge. It was also shown that immunizing mice with both recombinant influenza and vaccinia viruses, which express the B-cell and cytotoxic CD8⫹-T-cell epitopes of P. falciparum, resulted in CSspecific activated CD8⫹ T cells and antibodies (21). However, these viral vectors, based on moderately attenuated recombinant viruses, could not be safely used in a human malaria vaccine, and certainly not in immunosuppressed individuals. In the present paper, we demonstrate that recombinant replication-attenuated cold-adapted (ca) influenza and modified vaccinia Ankara (MVA) virus vectors, attenuated to levels known to be safe in humans, remain effective in inducing protective immunity against P. yoelii infection in BALB/c mice. Although it is possible that this mouse model does not completely mimic P. falciparum infection in humans, our results indicate that virus vectors based on replication-attenuated influenza and vaccinia viruses are good candidates for priming-boosting vaccination approaches against malaria. Generation of a recombinant cold-adapted influenza virus
expressing a plasmodial CTL epitope. We previously generated recombinant influenza A/WSN/33 virus MNA expressing a cytotoxic-T-lymphocyte (CTL) epitope (SYVPSAEQI) derived from the CS protein of P. yoelii (24). This epitope was inserted into the neuraminidase (NA) gene of the recombinant virus (24). In order to isolate a ca influenza virus expressing the same epitope, we coinfected MDCK cells with the MNA virus and the ca influenza A/Ann Arbor/6/60 virus (provided by H. Maassab), at a multiplicity of infection of one virus particle per cell. This procedure results in the generation of viral progenies consisting of a mixture of reassortant viruses containing different combinations of the eight viral genes from each parent virus. After 3 days of incubation at 33°C, supernatants were harvested and further passaged onto fresh MDCK cell monolayers at 25°C for 3 days to select for reassortant viruses containing the polymerase genes of the ca virus parent. The amplified viruses were plaque purified in MDCK cells at 33°C. Isolated plaques were grown in MDCK cells at 25°C, and viruses were further plaque purified in MDCK cells and amplified in the allantoic cavities of 10-day-old embryonated eggs (SPAFAS). The origin of each viral RNA segment of reassortant viruses was determined by reverse transcription-PCR, followed by restriction enzyme analysis. Among the different virus isolates, we were able to identify a virus clone (no. 154), containing the hemagglutinin (HA) and NA genes derived from the MNA virus and the six remaining genes from the ca virus. This virus clone, CA-CS, is reminiscent of the ca influenza virus vaccines containing the HA and NA genes derived from circulating influenza virus strains and the remaining genes from the ca virus. The ts phenotype of the ca recombinant virus clone 154 was corroborated by infection of MDCK cells at 33 or 39°C. CA-CS grew to titers of approximately 107 PFU/ml at 33°C. However, no infectious viruses were detected in the supernatant of MDCK cells infected at 39°C (data not shown). Immunization of mice with CA-CS induces CS-specific CTLs. We immunized 8-week-old female BALB/c (H-2d) mice with 105 PFU of CA-CS using the intranasal, subcutaneous (s.c.), intramuscular (i.m.), intraperitoneal (i.p.), and intravenous (i.v.) routes of administration. Two weeks after immunization, animals were sacrificed, and the number of gamma interferon (IFN-␥)-secreting, CS peptide-specific T cells was determined by an enzyme-linked immunospot (ELISPOT) assay, as previously described (7). Briefly, splenocytes were incubated with target antigen-presenting cells for 24 h, and then the number of spots corresponding to IFN-␥-secreting cells was counted. In these assays, P815 (H-2d) cells coated with SYVPSAEQI peptide, the H-2Kd CD8 CTL epitope of the P. yoelii CS, were used as target cells. Every ELISPOT assay was performed with the corresponding controls; i.e., splenocytes obtained from the immunized mice were also cultured with P815 cells without peptide. In these control assays, no response or only a weak response was obtained (data not shown). By contrast, all routes of immunization with CA-CS resulted in the induction of IFN-␥-secreting cells specific for the CS epitope. Similarly to our previous studies with the non-ca recombinant influenza virus expressing the CS epitope (Flu-ME) (21a), no major differences were found in the magnitude of the CS-specific immune response induced through the different immunization routes, which resulted in 150 to 400 CS-specific IFN-␥ secreting cells per 106 splenocytes (Fig. 1A). In subse-
FIG. 1. CS-specific CD8⫹-T-cell immune response elicited by CACS. Numbers of CS-specific CD8⫹ T cells were measured with the IFN-␥ ELISPOT assay on fresh splenocytes 2 weeks after immunization with a single dose of 105 PFU of CA-CS following different routes (A), 2 weeks after immunization with different doses of CA-CS administered subcutaneously (B), and 4 weeks (4w) or 2 weeks (2w) after subcutaneous immunization with 106 PFU of CA-CS or 4 weeks after subcutaneous immunization with 106 PFU of CA-CS followed 2 weeks later by a homologous subcutaneous booster with the same dose of CA-CS (C). The results are those obtained in one of two identical experiments and are means plus standard deviations for four mice.
quent experiments, we used the s.c. route of administration, resulting in slightly higher numbers of IFN-␥-secreting CSspecific CD8⫹ T cells (Fig. 1A). In a dose-response immunization experiment, we found that maximal numbers of CSspecific CD8⫹ T cells (approximately 600 cells per million) were obtained with 106 PFU of CA-CS. This number of cells failed to increase upon immunization with a higher dose of CA-CS (Fig. 1B). We then compared the immune response induced by immunization with one versus two s.c. doses of CA-CS, administered at a 2-week interval. The second dose of CA-CS failed to increase greatly (boost) the initial immune response but had simply an additive effect on the number of CS-specific CD8⫹ T cells (Fig. 1C). The presence of neutralizing antibodies against
VOL. 77, 2003
the CA-CS vector induced after the first immunization is most likely responsible for the lack of a robust boosting effect. Similar additive effects after boosting with homologous viruses were seen when replication-impaired viral vectors were used, most likely indicative of low levels of infection achieved by the vector in animals previously immunized with the same vector (14). We also found that the response to a single s.c. dose of CA-CS decreased about 10 times between 2 and 4 weeks after injection (Fig. 1C). It is noteworthy that the level of the antiSYVPSAEQI CD8⫹-T-cell response to CA-CS was very similar to that obtained by immunization with the non-ca Flu-ME virus (data not shown). Cellular and antibody responses of mice primed with CA-CS or Flu-ME and boosted with either VAC-CS or MVA-CS. Previously, we found that cellular immune responses induced in mice by a recombinant influenza virus vector expressing a CTL epitope are efficiently boosted with a vaccinia virus vector expressing the same epitope. Since vaccinia virus is known to cause adverse effects in a small proportion of immunized individuals, especially if they are immunocompromised, we generated a recombinant attenuated vaccinia virus expressing the CS protein of P. yoelii based on the replication-impaired MVA strain. For this purpose, the gene encoding the entire CS protein of P. yoelii was cloned into the vaccinia virus insertion vector pJR96-01, downstream of a strong synthetic early/late virus promoter (8). The resulting plasmid, pJR-PYCS, was used to transfect chicken embryo fibroblast (CEF) cells infected with the MVA strain of vaccinia virus (provided by G. Sutter). After six rounds of plaque purification, recombinant viruses were grown on cultures of CEF cells. The purified viruses were titrated by immunostaining (13). CA-CS immunized animals were boosted 2 weeks later by using different routes of administration of the attenuated MVA virus vector expressing CS (MVA-CS) (Fig. 2A). When MVA-CS was given i.v., i.m. or i.p. to CA-CS-primed mice, the number of CS-specific CD8⫹ T cells was efficiently boosted. By contrast, poor responses were induced by the s.c. and i.d. routes of administration. In subsequent experiments, we used the i.v. route of administration for MVA-CS, resulting in slightly higher numbers of IFN-␥-secreting CS-specific CD8⫹ T cells. Similar observations were previously obtained by Hanke et al. using a recombinant MVA virus expressing CTL epitopes from HIV-1 (17). Animals immunized with only MVA-CS (not primed with the CA-CS influenza virus vector) exhibited only a modest response, similar to the one induced by CA-CS alone (Fig. 2A, last column). The number of CS-specific CD8⫹ T cells correlated closely with the viral dose used for booster immunization (Fig. 2B). Animals boosted i.v. with 108 PFU of MVA-CS displayed the largest number of IFN-␥secreting CS-specific CD8⫹ T cells. Efficient expression of CS by the MVA vector was required for efficient boosting, since a different version of MVA-CS that expressed CS under the control of a less potent promoter (the p7.5 promoter) induced 5-times-lower numbers of CS-specific CD8⫹ T cells than the version of MVA-CS used in our experiments (data not shown). We then compared the efficacy of a priming-boosting strategy using the replication-impaired CA-CS and MVA-CS vectors to that obtained with regimens based on the use of the corresponding wild-type viral strains of these vectors also expressing the same plasmodial epitope, Flu-ME and VAC-CS.
The recombinant vaccinia virus VAC-CS, derived from the Western Reverse strain, was described previously (18). Four groups of mice were primed s.c. with either Flu-ME or CA-CS and boosted i.v. with either MVA-CS or VAC-CS. When an optimal priming dose (106 PFU) of Flu-ME or CA-CS was used, the viral vectors were comparable in their ability to prime a CS-specific immune response, as measured by the numbers of splenic CS-specific IFN-␥-secreting CD8⫹ T cells at day 14 or 21 after boosting (Fig. 2C, compare columns 1 and 3 and columns 2 and 4). Only when low doses (103 or 104 PFU) of Flu-ME or CA-CS were used in priming did Flu-ME-primed mice have higher CD8 responses than those primed with CA-CS (data not shown). By contrast, when the boosting ability of VAC-CS and MVA-CS was analyzed, we found that 2 weeks after boosting the number of splenic CS-specific CD8⫹ T cells was considerably greater in the VAC-CS-boosted group of mice. By week 3, the number of CD8⫹ T cells decreased to levels similar to those found in MVA-CS-boosted animals (Fig. 2C). In contrast, mice boosted with MVA-CS had similar numbers of CS-specific CD8⫹ T cells after 2 and 3 weeks (Fig. 2C). Antisporozoite antibody titers in mice subjected to primingboosting were determined by immunofluorescence assay (IFA) using P. yoelii sporozoites, which were air-dried onto multiwell glass slides. After these antigen slides had been incubated for 1 h with the sera diluted in phosphate-buffered saline plus 1% bovine serum albumin, the slides were washed and incubated for another hour with a fluorescein isothiocyanate (FITC)labeled goat anti-mouse immunoglobulin antibody (Kirkegaard & Perry Laboratories). The slides were mounted (after repeated washes), and the antisporozoite antibody titers were determined as the highest serum dilution displaying fluorescent sporozoites (22). The antisporozoite antibody titers were very high in all four groups of mice but varied greatly in individual animals within the same group, namely, from 1,600 to 12,800 (Fig. 2C). It should be noted that since the Flu-ME and CA-CS vectors express a single CS-derived epitope in the context of the stalk of its NA protein, CS-specific antibodies were induced solely by the MVA-CS or VAC-CS vector. In order to confirm the differences in magnitude of the cellular CD8⫹ immune response induced when VAC-CS and MVA-CS were used as boosters, we also examined by tetramer staining the number of CS-specific CD8⫹ T cells present in peripheral blood of mice primed with CA-CS and boosted with MVA-CS or VAC-CS, at 5, 14, and 21 days after the booster. In these experiments, CS-specific CD8⫹ T cells were identified by using an H-2Kd major histocompatibility complex class I tetrameric complex containing the peptide SYVPSAEQI within the peptide-binding cleft. Soluble tetrameric H-2Kd SYVPSAEQI complexes were produced, as previously described (5). Freshly isolated peripheral blood lymphocytes were first stained directly, ex vivo, with the tetrameric complex and labeled with phycoerythrin, followed by staining with allophycocyanin-labeled anti-CD8 and anti-CD3–FITC-labeled monoclonal antibodies. Samples were analyzed on a FACSCalibur apparatus using CELLQUEST software (Becton Dickinson). Tetramer-positive-T-cell numbers reached a maximum on day 14, when we used VAC-CS as the booster, and decreased after this time. However, when the booster was given with MVA virus, the percentage of positive CD8⫹ T cells increased gradually from day 5 to at least day 21 after the
FIG. 2. Efficient priming and boosting of CS-specific CD8⫹-T-cell immune responses elicited by CA-CS followed by MVA-CS. (A) BALB/c mice were immunized with 106 PFU of CA-CS and boosted 2 weeks later with 5 ⫻ 107 PFU of MVA-CS by different routes. The CS-specific CD8⫹ immune response was tested 2 weeks after boosting. (B) Optimal dose of booster immunization with MVA-CS. BALB/c mice were immunized with 106 PFU of CA-CS and boosted 2 weeks later with different doses of MVA-CS. (C) Cellular and humoral immune responses of BALB/c mice immunized by different priming-boosting protocols. IFN-␥-secreting CD8⫹ cells and IFA antibody titers of mice primed with 106 PFU of Flu-ME or CA-CS and boosted 2 weeks later with 5 ⫻ 107 PFU of MVA-CS or VAC-CS were measured. At 14 and 21 days after boosting, the numbers of CS-specific CD8⫹ T cells in the spleen were determined by ELISPOT assay. Sera from all immunized mice were also collected and assayed by IFA for the presence of CS-specific antibodies. The results are those obtained in one of four identical experiments and are means plus standard deviations for three mice.
booster (end of the observation period), to approximately 30% of all CD8⫹ T cells (Fig. 3A). The same tetramer stained approximately 0.08% of CD8⫹ T cells from mice immunized with MVA-CS or VAC-CS alone and 0.03% from naive mice (data not shown). The percentage of tetramer-positive CD8⫹
T cells measured in animals immunized with CA-CS alone was similar to that found in animals immunized only with MVA-CS (between 0.5 and 0.9%). To determine the duration of the immune response, we analyzed the numbers of CS-specific CD8⫹ T cells in spleens of
VOL. 77, 2003
FIG. 3. CS-specific CD8⫹ cellular immune responses induced in mice primed with CA-CS and boosted 2 weeks later with VAC-CS or MVA-CS viruses. (A) On days 5, 14, and 21 after boosting, peripheral blood mononuclear cells from mice primed with CA-CS and boosted with either VAC-CS or MVA-CS were analyzed by flow cytometry by staining with H-2d tetramer-phycoerythrin and anti-CD8-FITC. The percentages of double-positive cells are presented. The results represent those obtained in one of two identical experiments and are means ⫾ standard deviations for 10 mice. (B) Long persistence of CS-specific IFN-␥-secreting CD8⫹ cells after priming-boosting immunization. CSspecific CD8⫹-T-cell responses in the spleen were measured by ELISPOT assay at 2 weeks and at 1, 2, and 4 months after priming with CA-CS and boosting with MVA-CS or VAC-CS. The results are those obtained in one of two identical experiments and are means ⫾ standard deviations for three mice.
immunized mice by ELISPOT. Mice were primed with CA-CS and boosted with MVA-CS or VAC-CS, and CS-specific CD8⫹-T-cell splenocytes were monitored at different times postimmunization between 14 and 120 days. There was an initial sharp increase of CD8⫹ cells, at 14 days after the VAC-CS booster, followed by a rapid decrease. This response later declined slightly, remaining detectable for up to 120 days. Similar numbers of CS-specific IFN-␥-producing CD8⫹ T cells were elicited by the MVA-CS booster, except for a smaller peak at 14 days (Fig. 3B).
Priming with CA-CS and boosting with MVA-CS provide complete protection of mice against P. yoelii sporozoite challenge. Mice that had been immunized by using different priming-boosting combinations were challenged with 50,000 P. yoelii sporozoites at day 21 postimmunization, and the levels of protection against parasite development were evaluated. We obtained sporozoites of P. yoelii (strain 17X NL) from dissected salivary glands of Anopheles stephensi mosquitoes, 2 weeks after their blood meal on infected (gametocyte-carrying) mice. Forty hours after the challenge, the level of inhibition of liver stage development was determined by quantifying plasmodial 18S rRNA in the livers of mice. For this purpose, liver-extracted RNA was reverse transcribed, and an aliquot of the resulting cDNA was used for real-time PCR amplification of the plasmodial 18S rRNA sequences, using a GeneAmp 5700 sequence detection system (PE Applied Biosystems, Foster City, Calif.), as described previously (6). We found that inhibition of liver stage development of the parasites, as determined by reverse transcription-PCR, was greatest in FluME- or CA-CS-primed, MVA-CS-boosted mice (98.8 and 98.2%, respectively) (Fig. 4A). The percentage of inhibition in mice boosted with VAC-CS was also high but consistently failed to reach the levels seen after MVA-CS boosting. Depletion experiments demonstrated that CD8⫹ T cells were required for the protection afforded by the CA-CS–MVA-CS priming-boosting immunization regimen (Fig. 4B). Treatment with anti-CD8 almost completely abolished the inhibition of liver stage development, whereas treatment with anti-CD4 only slightly diminished the inhibition induced by immunization with CA-CS followed by MVA-CS. In parallel experiments, protection was also assessed by examining blood smears of immunized mice, which were challenged with 100 viable P. yoelii sporozoites, for the occurrence of erythrocytic stages. Peripheral blood smears were obtained daily from day 3 to 14 postchallenge, stained with Giemsa, and examined by microscopy, to determine whether these immunized mice became parasitemic, i.e., failed to develop protection. The inhibition (⬎98%) of plasmodial RNA levels in CACS-primed, MVA-CS-boosted mice coincided with the 100% resistance of the animals to develop blood stages upon sporozoite challenge. At the same time, complete protection occurred in only 70% of mice primed with CA-CS but boosted with VAC-CS (Table 1). Moreover, in mice challenged 2 months after immunization, the percentage of completely protected mice decreased to 63% in MVA-CS-boosted mice, versus 40% in VAC-CS-boosted mice. These percentages remained unaltered at 4 months postimmunization (Table 1). Potential use of influenza and vaccinia virus vectors as malaria vaccines. We have shown that attenuated influenza and vaccinia virus vectors induce protective cellular immune responses when used in a priming-boosting immunization strategy against murine malaria. The two attenuated viruses we used for this immunization, namely, a recombinant ca influenza virus and a recombinant MVA virus, were based on viruses which had been safely used for vaccination of very large numbers of adults and children (2, 19). ca influenza viruses were obtained by passaging influenza viruses at low temperatures in tissue culture. As a result, influenza virus became adapted to efficiently replicate at 25°C while losing its ability to replicate at higher (host) temperatures. ca influenza viruses
FIG. 4. Percent inhibition of P. yoelii liver stage development in BALB/c mice primed and boosted with replication-impaired recombinant viruses. (A) Groups of five mice were immunized with recombinant constructs indicated on the x axis at the doses indicated in the legend to Fig. 2C, while one group was left untreated. Three weeks after boosting, all mice were challenged i.v. with 50,000 P. yoelii sporozoites, and the level of parasite rRNA present in their livers was determined as described in the text. The results are those obtained in one of five identical experiments and are means plus standard deviations for five mice. (B) Mice were immunized with 106 PFU of CA-CS and boosted 2 weeks later with 5 ⫻ 107 PFU of MVA-CS. Three weeks after boosting, mice were challenged i.v. with 50,000 P. yoelii sporozoites. Mice were depleted of CD4⫹ and CD8⫹ T cells by daily injections of anti-CD4 and anti-CD8 over the 3 days before challenge. The efficiency of depletion was tested by fluorescenceactivated cell sorting.
are presumed to be attenuated because they fail to replicate in the lower respiratory tract of mammalian hosts. MVA viruses were obtained by continuous passaging of vaccinia virus in tissue culture (embryonic chicken fibroblasts), which resulted in the loss of a significant part of the viral genome, leading to attenuation in mammalian hosts. These two attenuated viruses, when engineered to express either the immunodominant CTL epitope of P. yoelii CS, or the entire CS protein, were as protective as, or even more protective than, the corresponding
more virulent viral vectors used previously against malaria (18). In fact, the protective effect of the MVA-CS booster was greater than that of the VAC-CS booster. The greater protection with the MVA-CS booster was demonstrated by the nearly total inhibition of liver stage development in immunized and challenged mice (Fig. 4A). It was corroborated by subsequent experimental results, namely, that 100% of the MVA-CSboosted mice were protected in a challenge experiment 3 weeks later failing to develop parasitemia. In contrast, only
TABLE 1. MVA-CS booster confers greater resistance against P. yoelii infection than VAC-CS Immunization
Result of challengeb at no. of days after boosting: 21
No. infected/ no. challenged
No. infected/ no. challenged
No. infected/ no. challenged
CA-CS CA-CS None
VAC-CS MVA-CS None
6/20 0/20 20/20
70 100 0
12/20 8/21 20/20
40 63 0
6/10 4/10 10/10
40 60 0
Mice were primed with 106 PFU of CA-CS and boosted with 5 ⫻ 107 PFU of VAC-CS or MVA-CS. Mice received a single challenge dose consisting of 100 viable P. yoelii sporozoites at 21, 60, or 120 days after boosting.
VOL. 77, 2003
70% of the VAC-CS-boosted mice developed sterile immunity (Table 1). Since VAC-CS and MVA-CS were not able to confer protection against sporozoite challenge when used alone (Fig. 4A), priming with recombinant influenza virus expressing a single CS epitope recognized by CTLs was essential for protection in these experiments. Although at this moment it is unclear whether the same vaccination approach will be successful in hosts with other genetic backgrounds, the generation of recombinant influenza viruses expressing multiple epitopes recognized by different class I molecules is feasible. An important finding was also the long duration of the protective response. Although levels of protection declined from 100 to 63% from day 21 to 60 after immunization, 60% of the mice were still protected at 120 days after boosting with MVA-CS (Table 1). A consistently short duration of protection has, up until now, been an important failure not only of experimental pre-erythrocytic immunization attempts but also of the corresponding human vaccine trials. In the first successful human vaccine trial, using a hybrid hepatitis B surface antigen fused with a large part of the CS protein of P. falciparum (RTS, S-BSA2), six of seven vaccinees, using certain adjuvants, were shortly thereafter protected against sporozoite challenge (31). When RTS, S-BSA2 was used as a vaccine in an area where malaria is endemic, The Gambia, the lack of duration of this vaccine’s efficacy was striking. During the first 9 weeks of surveillance, the efficacy of this vaccine formulation was reported as 71%. However, it was 0% in the following 6 weeks (4). In malaria-naïve individuals, only two of seven vaccinees who had resisted an earlier sporozoite challenge at week 31 were protected against rechallenge after 55 weeks (U. Krzych, personal communication). Vaccination protocols resulting in long-lasting immunity against Plasmodium will be needed to efficiently prevent malaria. A critical role of the CD8⫹ T cells in mediating protection was directly demonstrated in depletion experiments. Specifically, CD8 depletion and not CD4 depletion completely abolished CA-CS/MVA-CS-induced protection. However, the molecular basis of the greater protection conferred by MVA-CS (compared to VAC-CS) has not yet been clarified. We found that it is not based on a greater number of CS-specific CD8⫹ T cells or on higher antisporozoite antibody titers in the MVACS-boosted mice. Moreover, the decline of protection observed in CA-CS/MVA-CS-immunized animals between days 21 and 60 (Table 1) cannot be attributed to a decrease in numbers of CD8⫹ T cells, which remain approximately constant during days 20 and 120 postimmunization (Fig. 3B). Similarly, CS-specific antibody titers were stable over time (data not shown). The quality of the CTL response or additional immune parameters not yet identified might be responsible for the increased levels of protection induced when MVA-CS was used as a booster, compared to VAC-CS. MVA viruses, which lost approximately 30% of the vaccinia virus genome, might have lost expression of viral immunomodulatory genes, e.g., those encoding cytokines, chemokines, or their receptors, which may modulate the quality of the host immune responses (3). Defining the molecular mechanisms of the increased protection induced by the MVA-CS booster is an important issue to be investigated. We used the mouse-P. yoelii malaria model in these experiments for two reasons: First, this system provides the oppor-
tunity to observe protection and to quantitatively determine the level of parasites in the liver. This is not feasible by immunization of mice with the P. falciparum CS-expressing viral vectors, since P. falciparum sporozoite challenge does not induce disease in mice. Second, results obtained in the experimental P. yoelii rodent model by many investigators indicate that this is a good model for human malaria, and, where testable, results obtained with pre-erythrocytic malaria vaccines could be extrapolated to P. falciparum infection in humans (12). However, although our results clearly show that immunization with recombinant ca influenza virus followed by recombinant MVA virus induces protection against P. yoelii infection in mice, future experimentation is required to proof the efficacy of this vaccine approach to protect against P. falciparum in humans. We gratefully acknowledge H. F. Maassab and Gerd Sutter for providing us with the ca influenza A/Ann Arbor/6/60 virus and MVA, respectively, and Estanislao Nistal-Villa´n and Julius Hafalla for technical assistance. A.G.-S. and P.P. are consultants for Medimmune, Gaithersburg, Md. This work was supported by National Institute of Health grants to A.G.-S., R.S.N., and P.P. and by grants BIO99-0803 and BIO2001-2269 to D.R.M.E. financed by the Spanish Ministry of Science and Technology. REFERENCES 1. Amara, R. R., F. Villinger, J. D. Altman, S. L. Lydy, S. P. O’Neil, S. I. Staprans, D. C. Montefiori, Y. Xu, J. G. Herndon, L. S. Wyatt, M. A. Candido, N. L. Kozyr, P. L. Earl, J. M. Smith, H. L. Ma, B. D. Grimm, M. L. Hulsey, J. Miller, H. M. McClure, J. M. McNicholl, B. Moss, and H. L. Robinson. 2001. Control of a mucosal challenge and prevention of AIDS by a multiprotein DNA/MVA vaccine. Science 292:69–74. 2. Belshe, R. B., E. M. Swierkosz, E. L. Anderson, F. K. Newman, S. L. Nugent, and H. F. Maassab. 1992. Immunization of infants and young children with live attenuated trivalent cold-recombinant influenza A H1N1, H3N2, and B vaccine. J. Infect. Dis. 165:727–732. 3. Blanchard, T. J., A. Alcami, P. Andrea, and G. L. Smith. 1998. Modified vaccinia virus Ankara undergoes limited replication in human cells and lacks several immunomodulatory proteins: implications for use as a human vaccine. J. Gen. Virol. 79:1159–1167. 4. Bojang, K. A., P. J. Milligan, M. Pinder, L. Vigneron, A. Alloueche, K. E. Kester, W. R. Ballou, D. J. Conway, W. H. Reece, P. Gothard, L. Yamuah, M. Delchambre, G. Voss, B. M. Greenwood, A. Hill, K. P. McAdam, N. Tornieporth, J. D. Cohen, and T. Doherty. 2001. Efficacy of RTS,S/AS02 malaria vaccine against Plasmodium falciparum infection in semi-immune adult men in The Gambia: a randomised trial. Lancet 358:1927–1934. 5. Bruna-Romero, O., G. Gonzalez-Aseguinolaza, J. C. Hafalla, M. Tsuji, and R. S. Nussenzweig. 2001. Complete, long-lasting protection against malaria of mice primed and boosted with two distinct viral vectors expressing the same plasmodial antigen. Proc. Natl. Acad. Sci. USA 98:11491–11496. 6. Bruna-Romero, O., J. C. Hafalla, G. Gonzalez-Aseguinolaza, G. Sano, M. Tsuji, and F. Zavala. 2001. Detection of malaria liver-stages in mice infected through the bite of a single Anopheles mosquito using a highly sensitive real-time PCR. Int. J. Parasitol. 31:1499–1502. 7. Carvalho, L. H., J. C. Hafalla, and F. Zavala. 2001. ELISPOT assay to measure antigen-specific murine CD8(⫹) T cell responses. J. Immunol. Methods 252:207–218. 8. Chakrabarti, S., J. R. Sisler, and B. Moss. 1997. Compact, synthetic, vaccinia virus early/late promoter for protein expression. BioTechniques 23:1094– 1097. 9. Clyde, D. F. 1990. Immunity to falciparum and vivax malaria induced by irradiated sporozoites: a review of the University of Maryland studies, 1971– 75. Bull. W. H. O. 68(Suppl.):9–12. 10. Clyde, D. F. 1975. Immunization of man against falciparum and vivax malaria by use of attenuated sporozoites. Am. J. Trop. Med. Hyg. 24:397–401. 11. Cochrane, A. H., M. Aikawa, M. Jeng, and R. S. Nussenzweig. 1976. Antibody-induced ultrastructural changes of malarial sporozoites. J. Immunol. 116:859–867. 12. Doolan, D. L., and S. L. Hoffman. 2000. The complexity of protective immunity against liver-stage malaria. J. Immunol. 165:1453–1462. 13. Earl, P. L., N. Cooper, L. S. Wyatt, B. Moss, and M. W. Carroll. 1998. Preparation of cell cultures and vaccinia virus stocks, p. 16.16.1–16.16.13. In F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.
Smith, and K. Struhl (ed.), Current protocols in molecular biology. John Wiley & Sons, Inc., New York, N.Y. 14. Gherardi, M. M., J. L. Najera, E. Perez-Jimenez, S. Guerra, A. GarcíaSastre, and M. Esteban. 2003. Prime-boost immunization schedules based on influenza virus and vaccinia virus vectors potentiate cellular immune responses against human immunodeficiency virus Env protein systemically and in the genitorectal draining lymph nodes. J. Virol. 77:7048–7057. 15. Gilbert, S. C., J. Schneider, C. M. Hannan, J. T. Hu, M. Plebanski, R. Sinden, and A. V. Hill. 2002. Enhanced CD8 T cell immunogenicity and protective efficacy in a mouse malaria model using a recombinant adenoviral vaccine in heterologous prime-boost immunisation regimes. Vaccine 20: 1039–1045. 16. Gonzalo, R. M., D. Rodriguez, A. García-Sastre, J. R. Rodriguez, P. Palese, and M. Esteban. 1999. Enhanced CD8⫹ T cell response to HIV-1 env by combined immunization with influenza and vaccinia virus recombinants. Vaccine 17:887–892. 17. Hanke, T., T. J. Blanchard, J. Schneider, G. S. Ogg, R. Tan, M. Becker, S. C. Gilbert, A. V. Hill, G. L. Smith, and A. McMichael. 1998. Immunogenicities of intravenous and intramuscular administrations of modified vaccinia virus Ankara-based multi-CTL epitope vaccine for human immunodeficiency virus type 1 in mice. J. Gen. Virol. 79:83–90. 18. Li, S., M. Rodrigues, D. Rodriguez, J. R. Rodriguez, M. Esteban, P. Palese, R. S. Nussenzweig, and F. Zavala. 1993. Priming with recombinant influenza virus followed by administration of recombinant vaccinia virus induces CD8⫹ T-cell-mediated protective immunity against malaria. Proc. Natl. Acad. Sci. USA 90:5214–5218. 19. Mayr, A., H. Stickl, H. K. Muller, K. Danner, and H. Singer. 1978. The smallpox vaccination strain MVA: marker, genetic structure, experience gained with the parenteral vaccination and behavior in organisms with a debilitated defence mechanism. Zentralbl. Bakteriol. 167:375–390. 20. McShane, H. 2002. Prime-boost immunization strategies for infectious diseases. Curr. Opin. Mol. Ther. 4:23–27. 21. Miyahira, Y., A. García-Sastre, D. Rodriguez, J. R. Rodriguez, K. Murata, M. Tsuji, P. Palese, M. Esteban, F. Zavala, and R. S. Nussenzweig. 1998. Recombinant viruses expressing a human malaria antigen can elicit potentially protective immune CD8⫹ responses in mice. Proc. Natl. Acad. Sci. USA 31:3954–3959. 21a..Murata, K., A. Garcı´a-Sastre, M. Tsuji, M. Rodrigues, D. Rodriguez, J. R. Rodriguez, R. S. Nussenzweig, P. Palese, M. Esteban, and F. Zavala. 1996. Characterization of in vivo primary and secondary CD8⫹ T cell responses induced by recombinant influenza and vaccinia viruses. Cell. Immunol. 173: 96–107. 22. Nardin, E., R. W. Gwadz, and R. S. Nussenzweig. 1979. Characterization of sporozoite surface antigens by indirect immunofluorescence: detection of
27. 28. 29. 30.
stage- and species-specific antimalarial antibodies. Bull. W. H. O. 57(Suppl. 1):211–217. Rieckmann, K. H. 1990. Human immunization with attenuated sporozoites. Bull. W. H. O. 68(Suppl.):13–16. Rodrigues, M., S. Li, K. Murata, D. Rodriguez, J. R. Rodriguez, I. Bacik, J. R. Bennink, J. W. Yewdell, A. García-Sastre, R. S. Nussenzweig, M. Esteban, P. Palese, and F. Zavala. 1994. Influenza and vaccinia viruses expressing malaria CD8⫹ T and B cell epitopes. Comparison of their immunogenicity and capacity to induce protective immunity. J. Immunol. 153: 4636–4648. Rogers, W. O., W. R. Weiss, A. Kumar, J. C. Aguiar, J. A. Tine, R. Gwadz, J. G. Harre, K. Gowda, D. Rathore, S. Kumar, and S. L. Hoffman. 2002. Protection of rhesus macaques against lethal Plasmodium knowlesi malaria by a heterologous DNA priming and poxvirus boosting immunization regimen. Infect. Immun. 70:4329–4335. Schneider, J., S. C. Gilbert, T. J. Blanchard, T. Hanke, K. J. Robson, C. M. Hannan, M. Becker, R. Sinden, G. L. Smith, and A. V. Hill. 1998. Enhanced immunogenicity for CD8⫹ T cell induction and complete protective efficacy of malaria DNA vaccination by boosting with modified vaccinia virus Ankara. Nat. Med. 4:397–402. Schneider, J., S. C. Gilbert, C. M. Hannan, P. Degano, E. Prieur, E. G. Sheu, M. Plebanski, and A. V. Hill. 1999. Induction of CD8⫹ T cells using heterologous prime-boost immunisation strategies. Immunol. Rev. 170:29–38. Schnell, M. J. 2001. Viral vectors as potential HIV-1 vaccines. FEMS Microbiol. Lett. 200:123–129. Schofield, L., J. Villaquiran, A. Ferreira, H. Schellekens, R. Nussenzweig, and V. Nussenzweig. 1987. Gamma interferon, CD8⫹ T cells and antibodies required for immunity to malaria sporozoites. Nature 330:664–666. Sedegah, M., G. T. Brice, W. O. Rogers, D. L. Doolan, Y. Charoenvit, T. R. Jones, V. F. Majam, A. Belmonte, M. Lu, M. Belmonte, D. J. Carucci, and S. L. Hoffman. 2002. Persistence of protective immunity to malaria induced by DNA priming and poxvirus boosting: characterization of effector and memory CD8⫹-T-cell populations. Infect. Immun. 70:3493–3499. Stoute, J. A., M. Slaoui, D. G. Heppner, P. Momin, K. E. Kester, P. Desmons, B. T. Wellde, N. Garcon, U. Krzych, and M. Marchand, et al. 1997. A preliminary evaluation of a recombinant circumsporozoite protein vaccine against Plasmodium falciparum malaria. N. Engl. J. Med. 336:86–91. Yoshida, N., R. S. Nussenzweig, P. Potocnjak, V. Nussenzweig, and M. Aikawa. 1980. Hybridoma produces protective antibodies directed against the sporozoite stage of malaria parasite. Science 207:71–73. Zavala, F., M. Rodrigues, D. Rodriguez, J. R. Rodriguez, R. S. Nussenzweig, and M. Esteban. 2001. A striking property of recombinant poxviruses: efficient inducers of in vivo expansion of primed CD8⫹ T cells. Virology 280: 155–159.