Enhanced protection against malaria by a chimeric merozoite surface protein vaccine

The 42-kDa processed fragment of Plasmodium falciparum merozoite surface protein 1 (MSP-1(42)) is a prime candidate for a blood-stage malaria vaccine. Merozoite surface protein 8 contains two C-terminal epidermal growth factor (EGF)-like domains that may function similarly to those of MSP-1(42). Immunization with either MSP-1 or MSP-8 induces protection that is mediated primarily by antibodies against conformation-dependent epitopes. In a series of comparative immunogenicity and efficacy studies using the Plasmodium yoelii rodent model, we tested the ability of recombinant P. yoelii MSP-8 (rPyMSP-8) to complement rPyMSP-1-based vaccines. Unlike MSP-1, PyMSP-8-dependent protection required immunization with the full-length protein and was not induced with recombinant antigens that contained only the C-terminal EGF-like domains. Unlike PyMSP-8, the immunogenicity of the PyMSP-1 EGF-like domains was low when present as part of the rPyMSP-1(42) antigen. Immunization with a mixture of rPyMSP-1(42) and rPyMSP-8 further inhibited the antibody response to protective epitopes of rPyMSP-1(42) and did not improve vaccine efficacy. To improve PyMSP-1 immunogenicity, we produced a chimeric antigen containing the EGF-like domains of PyMSP-1 fused to the N terminus of PyMSP-8. Immunization with the chimeric rPyMSP-1/8 antigen induced high and comparable antibody responses against the EGF-like domains of both PyMSP-1 and PyMSP-8. This enhanced MSP-1-specific antibody response and the concurrent targeting of MSP-1 and MSP-8 resulted in improved, nearly complete protection against lethal P. yoelii 17XL malaria. Unexpectedly, immunization with rPyMSP-1/8 failed to protect against challenge infection with reticulocyte-restricted P. yoelii 17X parasites. Overall, these data establish an effective strategy to improve the efficacy of P. falciparum MSP-based vaccines.     

Complete protection against Plasmodium yoelii by adoptive transfer of a CD8+ cytotoxic T-cell clone recognizing sporozoite surface protein 2.

BALB/c mice immunized with irradiated Plasmodium yoelii sporozoites produce antibodies and cytotoxic T lymphocytes against the circumsporozoite protein and against a 140-kDa protein, sporozoite surface protein 2 (PySSP2). Approximately 50% of mice immunized with P815 cells transfected with the gene encoding PySSP2 are protected against malaria, and this protection is reversed by in vivo depletion of CD8+ T cells. To determine if CD8+ T cells against PySSP2 are adequate to protect against malaria in the absence of other malaria-specific immune responses, we produced three CD8+ T-cell clones by stimulating spleen cells from mice immunized with irradiated P. yoelii sporozoites with a mitomycin-treated P815 cell clone transfected with the PySSP2 gene. Adoptive transfer of clone TSLB7 protected 100% of mice against P. yoelii. The second clone protected 58% of mice, and the third clone provided no protection. Clone TSLB7 protected even when administered 3 h after sporozoite inoculation at a time when sporozoites had entered hepatocytes, suggesting that it is recognizing and eliminating infected hepatocytes. These studies demonstrate that cytotoxic T lymphocytes against PySSP2 can protect against P. yoelii sporozoite challenge in the absence of other parasite-specific immune responses.     

A recombinant 15-kilodalton carboxyl-terminal fragment of Plasmodium yoelii yoelii 17XL merozoite surface protein 1 induces a protective immune response in mice.

Since the developmental stages of malarial parasites which replicate within erythrocytes are responsible for the morbidity and mortality associated with this disease, antigens produced by these stages have been proposed as candidates for a vaccine. One surface protein of merozoites (MSP-1) has been shown to immunize both rodents and primates against virulent challenge infection in experimental systems. However, little is known of relevant epitopes on the molecule, and attempts to obtain recombinant MSP-1 polypeptides in a native configuration have proven difficult. We have found that the cysteine-rich, carboxyl-terminal region of the MSP-1 protein from the rodent malarial parasite Plasmodium yoelii yoelii can be expressed in a native configuration as a fusion protein in Escherichia coli. This recombinant polypeptide containing 15 kDa of the predicted 197-kDa protein elicits antibodies in mice which recognize the native parasite MSP-1. Most significantly, both inbred and outbred mice immunized with the fusion protein in Ribi adjuvant are partially and in some cases completely protected against challenge infection with an otherwise lethal parasite strain. This is the first observation of such significant protection obtained with a small portion of the MSP-1 produced in recombinant systems.     

Immunogenicity and protective efficacy of a Plasmodium yoelii Hsp60 DNA vaccine in BALB/c mice

The gene encoding the 60-kDa heat shock protein of Plasmodium yoelii (PyHsp60) was cloned into the VR1012 and VR1020 mammalian expression vectors. Groups of 10 BALB/c mice were immunized intramuscularly at 0, 3, and 9 weeks with 100 microg of PyHsp60 DNA vaccine alone or in combination with 30 microg of pmurGMCSF. Sera from immunized mice but not from vector control groups recognized P. yoelii sporozoites, liver stages, and infected erythrocytes in an indirect fluorescent antibody test. Two weeks after the last immunization, mice were challenged with 50 P. yoelii sporozoites. In one experiment the vaccine pPyHsp60-VR1012 used in combination with pmurGMCSF gave 40% protection (Fisher's exact test; P = 0.03, vaccinated versus control groups). In a second experiment this vaccine did not protect any of the immunized mice but induced a delay in the onset of parasitemia. In neither experiment was there any evidence of a protective effect against the asexual erythrocytic stage of the life cycle. In a third experiment mice were primed with PyHsp60 DNA, were boosted 2 weeks later with 2 x 10(3) irradiated P. yoelii sporozoites, and were challenged several weeks later. The presence of PyHsp60 in the immunization regimen did not lead to reduced blood-stage infection or development of parasites in hepatocytes. PyHsp60 DNA vaccines were immunogenic in BALB/c mice but did not consistently, completely protect against sporozoite challenge. The observation that in some of the PyHsp60 DNA vaccine-immunized mice there was protection against infection or a delay in the onset of parasitemia after sporozoite challenge deserves further evaluation.     

Antibody recognition and isoelectrofocusing of antigens of the malaria parasite Plasmodium yoelii.

Inbred BALB/c mice were either immunized with Triton X-100-extracted antigens of blood-stage Plasmodium yoelii or infected with P. yoelii and cured in three successive schedules. Whereas the immunized BALB/c became only partially protected from subsequent challenge infection with blood-stage P. yoelii, the convalescent mice acquired total immunity. When total P. yoelii antigen extract was resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and immunoblotted with anti-P. yoelii serum, five major protein bands of 150, 84, 40, 19, and 16 kDa were recognized by the sera of fully protected convalescent mice but not by the sera of partially protected mice. The utility of comparing reactivities of sera from fully protected and partially protected malaria hosts and the possibility that antigens uniquely recognized by the convalescent mouse sera may contribute to immunity against P. yoelii infection are discussed. Although previously reported to be an effective adjuvant for immunization against P. yoelii infection in (BALB/c x C57BL)F1 hybrid mice, saponin did not promote protection any better than did Freund adjuvant in BALB/c mice immunized with detergent-extracted P. yoelii antigen. Most of the P. yoelii proteins (14 to 250 kDa) found in Triton X-100 extracts of P. yoelii-parasitized erythrocytes isoelectrofocused as a single peak in the pH region 4.4 to 4.6, suggesting a rationale for previous findings that the most anti-P. yoelii protective and T-helper activities are induced by antigens isoelectrically focused in a fraction of similar pH.     

Members of the merozoite surface protein 7 family with similar expression patterns differ in ability to protect against Plasmodium yoelii malaria

Previously, we described the isolation of the Plasmodium yoelii sequence-related molecules P. yoelii MSP-7 (merozoite surface protein 7) and P. yoelii MSRP-2 (MSP-7-related protein 2) by their ability to interact with the amino-terminal end of P. yoelii MSP-1 in a yeast two-hybrid system. One of these molecules was the homologue of Plasmodium falciparum MSP-7, which was biochemically isolated as part of the shed MSP-1 complex. In the present study, with antibodies directed against recombinant proteins, immunoprecipitation analyses of the rodent system demonstrated that both P. yoelii MSP-7 and P. yoelii MSRP-2 could be isolated from parasite lysates and from parasite culture supernatants. Immunofluorescence studies colocalized P. yoelii MSP-7 and P. yoelii MSRP-2 with the amino-terminal portion of MSP-1 and with each other on the surface of schizonts. Immunization with P. yoelii MSRP-2 but not P. yoelii MSP-7 protected mice against a lethal infection with P. yoelii strain 17XL. These results establish that both P. yoelii MSP-7 and P. yoelii MSRP-2 are expressed on the surface of merozoites and released from the parasite and that P. yoelii MSRP-2 may be the target of a protective immune response.     

Linear and multiple antigen peptides containing defined T and B epitopes of the Plasmodium yoelii circumsporozoite protein: antibody- mediated protection and boosting by sporozoite infection

We previously reported the identification of a T cell epitope in the N- terminal part of the circumsporozoite protein (CSP) of Plasmodium yoelii yoelii (Pyy). CD4+ T cell clones derived from mice immunized with a 21-mer peptide (amino acids 59-79, referred to as Py1) containing this epitope confer complete protection after passive transfer in mice. These clones proliferate in vitro in the presence of a 13-mer peptide (amino acids 59-71, referred to as Py1T). This shorter peptide was found to behave as a Th epitope in vivo, allowing overcoming of the genetic restriction for production of anti-repeat antibodies in BALB/c mice, when cross-linked to three (QGPGAP) repeats of the Pyy CSP. In this study, we report protection in BALB/c mice, against a challenge with Pyy sporozoites after immunization with linear and multiple antigen peptides containing Py1T as T epitope and three repeats QGPGAP (Py3) as B epitope. Multiple antigen peptide (MAP4-Py1T- Py3)-induced immunity was shown to be more effective than immunity induced by the linear form of the conjugate (Py1T-Py3), protecting against challenges with higher numbers of sporozoites. In both cases, levels of anti-repeat antibodies were strongly correlated with anti- parasite antibodies and protection. When tested in vitro, sera from mice immunized with the protective constructs strongly inhibited Pyy liver stages, while lymph node T cells displayed no cytotoxicity. In vivo, depletion of CD4+ or CD8+ T cells did not affect protection. Furthermore, MAP4-Py1T-Py3-immunized mice were not protected against a challenge with P. yoelii nigeriensis sporozoites, a parasite which has the same Py1T sequence but differs from Pyy in its repeated sequence. These results demonstrate that anti-repeat antibodies raised by immunization with the linear or the MAP form are exclusively responsible for the protection. Furthermore, this antibody response is boosted by a sporozoite challenge, allowing protection against a second challenge.      

Role of host cellular response in differential susceptibility of nonimmunized BALB/c mice to Plasmodium berghei and Plasmodium yoelii sporozoites.

We found BALB/c mice to be on the order of 2,000 times more susceptible to Plasmodium yoelii than Plasmodium berghei sporozoites, as measured by the ability of these sporozoites to differentiate into microscopically detectable hepatic schizonts in the livers of immunologically naive mice. One of the factors that determine the relative insusceptibility of mice to P. berghei sporozoites is the innate cellular inflammatory response that the mice mount after injection with sporozoites. The cellular inflammatory response against P. berghei is initiated soon after sporozoite injection; by 24 h, substantial histopathological changes have developed within the liver. There is considerably less of a cellular inflammatory response against P. yoelii; significant histopathological changes within the liver are not observed until well after hepatic schizonts have begun to rupture at around 44 h postinjection of sporozoites. These differences in the cellular inflammatory response against two different, closely related species of sporozoites are of considerable interest. The data strongly suggest that the BALB/c-P. berghei sporozoite system is a relatively poor biological model for sporozoite immunization studies.     

Antigenic and sequence diversity at the C-terminus of the merozoite surface protein-1 from rodent malaria isolates, and the binding of protective monoclonal antibodies

Merozoite surface protein-1 (MSP-1) is a major candidate in the development of a vaccine against malaria. Immunisation with a recombinant fusion protein containing the two Plasmodium yoelii MSP-1 C-terminal epidermal growth factor-like domains (MSP-1(19)) can protect mice against homologous but not heterologous challenge, and therefore, antigenic differences resulting from sequence diversity in MSP-1(19) may be crucial in determining the potential of this protein as a vaccine. Representative sequence variants from a number of distinct P. yoelii isolates were expressed in Escherichia coli and the resulting recombinant proteins were screened for binding to a panel of monoclonal antibodies (Mabs) capable of suppressing a P. yoelii YM challenge infection in passive immunisation experiments. The sequence polymorphisms affected the binding of the antibodies to the recombinant proteins. None of the Mabs recognised MSP-1(19) of P. yoelii yoelii 2CL or 33X or P. yoelii nigeriensis N67. The epitopes recognised by the Mabs were further distinguished by their reactivity with the other fusion proteins. The extent of sequence variation in MSP-1(19) among the isolates was extensive, with differences detected at 35 out of the 96 positions compared. Using the 3-dimensional structure of the Plasmodium falciparum MSP-1(19) as a model, the locations of the amino acid substitutions that may affect Mab binding were identified. The DNA sequence of MSP-1(19) from two Plasmodium vinckei isolates was also cloned and the deduced amino acid sequence compared with that in other species.     

Characterization of Plasmodium yoelii monoclonal antibodies directed against stage-specific sporozoite antigens.

A battery of monoclonal antibodies against Plasmodium yoelii sporozoites was produced. Five of these (NYS1 through NYS5) were selected for characterization. All five were positive in the indirect immunofluorescent antibody test with P. yoelii sporozoites; however, each showed a different immunofluorescence pattern. Although NYS1 (immunoglobulin G3 [IgG3]), NYS2 (IgG3), and NYS3 (IgM) were positive in the circumsporozoite precipitation test, only NYS1 and NYS2 were able to neutralize sporozoite infectivity in mice. NYS4 (IgM) and NYS5 (IgG1) were not positive in the precipitation test and did not protect mice from sporozoite infection. All except NYS4 were species as well as stage specific. NYS4 cross-reacted with sporozoites of P. berghei. Electrophoretic immunoblotting analysis showed that these monoclonal antibodies detected sporozoite antigens of various molecular weights. Inhibition enzyme-linked immunosorbent assays indicated that each recognized a different antigenic epitope. The differences in their immunochemical and biological reactivity make them useful for screening a variety of P. yoelii antigens in recombinant DNA libraries. These antigens will be used in an animal model system for vaccine development.     

Plasmodium yoelii sporozoites with simultaneous deletion of P52 and P36 are completely attenuated and confer sterile immunity against infection

Malaria infection starts when sporozoites are transmitted to the mammalian host during a mosquito bite. Sporozoites enter the blood circulation, reach the liver, and infect hepatocytes. The formation of a parasitophorous vacuole (PV) establishes their intracellular niche. Recently, two members of the 6-Cys domain protein family, P52 and P36, were each shown to play an important albeit nonessential role in Plasmodium berghei sporozoite infectivity for the rodent host. Here, we generated p52/p36-deficient Plasmodium yoelii parasites by the simultaneous deletion of both genes using a single genetic manipulation. p52/p36-deficient parasites exhibited normal progression through the life cycle during blood-stage infection, transmission to mosquitoes, mosquito-stage development, and sporozoite infection of the salivary glands. p52/p36-deficient sporozoites also showed normal motility and cell traversal activity. However, immunofluorescence analysis and electron microscopic observations revealed that p52/p36-deficient parasites did not form a PV within hepatocytes in vitro and in vivo. The p52/p36-deficient parasites localized as free entities in the host cell cytoplasm or the host cell nucleoplasm and did not develop as liver stages. Consequently, they did not cause blood-stage infections even at high sporozoite inoculation doses. Mice immunized with p52/p36-deficient sporozoites were completely protected against infectious sporozoite challenge. Our results demonstrate for the first time the generation of two-locus gene deletion-attenuated parasites that infect the liver but do not progress to blood-stage infection. The study will critically guide the design of Plasmodium falciparum live attenuated malaria vaccines.     

Immunization with parasite-derived apical membrane antigen 1 or passive immunization with a specific monoclonal antibody protects BALB/c mice against lethal Plasmodium yoelii yoelii YM blood-stage infection

We have purified apical merozoite antigen 1 (AMA-1) from extracts of red blood cells infected with the rodent malaria parasite Plasmodium yoelii yoelii YM. When used to immunize mice, the protein induced a strong protective response against a challenge with the parasite. Monoclonal antibodies specific for P. yoelii yoelii AMA-1 were prepared, and one was very effective against the parasite on passive immunization. A second protein that appears to be located in the apical rhoptry organelles and associated with AMA-1 was identified.     

A protective glycosylphosphatidylinositol-anchored membrane protein of Plasmodium yoelii trophozoites and merozoites contains two epidermal growth factor-like domains

Using sera from mice immunized and protected against Plasmodium yoelii malaria, we identified a novel blood-stage antigen gene, pypag-2. The 2.1-kb pypag-2 cDNA contains a single open reading frame that encodes a 409-amino-acid protein with a predicted molecular mass of 46.8 kDa. Unlike many characterized plasmodial antigens, blocks of tandemly repeated amino acids are lacking in the pypAg-2 protein sequence. Recombinant pypAg-2, comprising the full-length protein minus the predicted N-terminal signal and C-terminal anchor sequences, was produced and used to raise a high-titer polyclonal rabbit antiserum. This antiserum was used to identify and characterize the native protein through immunoblotting, immunoprecipitation and immunofluorescence assays. Consistent with the presence of a glycosylphosphatidylinositol anchor, pypAg-2 fractionated with the detergent phase of Triton X-114-solubilized proteins and could be metabolically labeled with [(3)H]palmitic acid. By immunofluorescence, pypAg-2 expression was localized to both the trophozoite and merozoite membranes. Similar to Plasmodium falciparum merozoite surface protein 1, pypAg-2 contains two C-terminal epidermal growth factor (EGF)-like domains. Most importantly, immunization with recombinant pypAg-2 protected mice against lethal P. yoelii malaria. Thus, pypAg-2 is a target of protective immune responses and represents a novel addition to the family of merozoite surface proteins that contain one or more EGF-like domains.     

Protective immunity against Plasmodium yoelii malaria induced by immunization with particulate blood-stage antigens.

The Plasmodium yoelii murine model was used to test several combinations of blood-stage antigens and adjuvants for the ability to induce immunity to blood-stage malaria. Upon fractionation of whole blood-stage antigen into soluble and insoluble components, only the particulate antigens (pAg) induced protective immune responses. Of a number of adjuvants tested, Quil A was the most effective. Immunization with pAg plus Quil A induced solid protection against nonlethal and lethal P. yoelii challenge infection. Analysis of cytokine production revealed mRNA for Th1-type cytokines (interleukin 2 [IL-2] and gamma interferon) as well as Th2-type cytokines (IL-4 and IL-10) in the spleens of both protected and susceptible animals. The data suggested that the protective pAg response was associated with the earlier production of cytokine mRNA with a Th2 phenotype somewhat favored. Immunization of B-cell-deficient JHD mice indicated that the protection against P. yoelii induced by pAg immunization was B cell dependent. Although immunization with pAg plus Quil A increased the levels of antigen-specific antibodies of all four immunoglobulin G (IgG) isotypes, protection correlated most closely with the presence of IgG1 and IgG2b antibodies. Sera from pAg-plus-Quil A-immunized animals recognized only a limited subset of six to eight distinct P. yoelii antigens, primarily associated with the pAg fraction. These results provide the basis for the identification and characterization of potential vaccine antigens, selected solely for their ability to immunize against blood-stage malaria.     

DNA immunization by Plasmodium falciparum liver-stage antigen 3 induces protection against Plasmodium yoelii sporozoite challenge

DNA-based immunization of mice by Plasmodium falciparum liver-stage antigen 3 (PfLSA3), a novel highly conserved P. falciparum preerythrocytic antigen, was evaluated. Animals developed a dominant Th1 immune response (high gamma interferon T-cell responses and predominance of immunoglobulin G2a) to each of three recombinant proteins spanning the molecule. We have exploited the immunological cross-reactivity of PfLSA3 with its putative homologue on sporozoites of the rodent parasite Plasmodium yoelii, and we show for the first time that responses induced by PfLSA3 in mice significantly protect against a heterologous challenge by P. yoelii sporozoites. These results support a significant effect of DNA-induced immune responses on preerythrocytic stages.     

CD4+ T-Cell- and Gamma Interferon-Dependent Protection against Murine Malaria by Immunization with Linear Synthetic Peptides from a Plasmodium yoelii 17-Kilodalton Hepatocyte Erythrocyte Protein

Most work on protective immunity against the pre-erythrocytic stages of malaria has focused on induction of antibodies that prevent sporozoite invasion of hepatocytes, and CD8(+) T-cell responses that eliminate infected hepatocytes. We recently reported that immunization of A/J mice with an 18-amino-acid synthetic linear peptide from Plasmodium yoelii sporozoite surface protein 2 (SSP2) in TiterMax adjuvant induces sterile protection that is dependent on CD4(+) T cells and gamma interferon (IFN-gamma). We now report that immunization of inbred A/J mice and outbred CD1 mice with each of two linear synthetic peptides from the 17-kDa P. yoelii hepatocyte erythrocyte protein (HEP17) in the same adjuvant also induces protection against sporozoite challenge that is dependent on CD4(+) T cells and IFN-gamma. The SSP2 peptide and the two HEP17 peptides are recognized by B cells as well as T cells, and the protection induced by these peptides appears to be directed against the infected hepatocytes. In contrast to the peptide-induced protection, immunization of eight different strains of mice with radiation-attenuated sporozoites induces protection that is absolutely dependent on CD8(+) T cells. Data represented here demonstrate that CD4(+) T-cell-dependent protection can be induced by immunization with linear synthetic peptides. These studies therefore provide the foundation for an approach to pre-erythrocytic-stage malaria vaccine development, based on the induction of protective CD4(+) T-cell responses, which will complement efforts to induce protective antibody and CD8(+) T-cell responses.     

Host-parasite interactions and immunity to irradiated sporozoites

We compare and contrast the results of immunizing mice with irradiated sporozoites of Plasmodium berghei and Plasmodium yoelii. Host genetic control of protective immunity is different in the two rodent malarias. Few mouse strains are strongly protected by P. yoelii sporozoites, while all are protected by P. berghei sporozoite immunization. The role of CD8+ T cells in the protective immune response to each of these malarias varies with the strain of mouse. Moreover, a single strain will use a CD8+ T cell-dependent mechanism against one malaria, and a CD8+ independent mechanism against the other. Thus, each host-parasite pairing in these rodent malarias engenders a unique set of immune responses. Such variety should be expected in the immune response to the human malarias, and may complicate the development of universally applicable vaccines.     

Influence of adjuvants on protection induced by a recombinant fusion protein against malarial infection.

Previously, we described a protective immune response induced by the carboxyl-terminal region of the merozoite surface protein-1 (MSP-1) from the rodent malarial parasite Plasmodium yoelii yoelii 17XL, expressed as a fusion protein and designated glutathione S-transferase (GST)-PYC2. We also demonstrated that the humoral response induced by GST-PYC2 was the primary mechanism by which immunized animals controlled their blood-stage infections. We have now examined the influence of several adjuvants on the immune response to the GST-PYC2 fusion protein. While alum, Freund's adjuvant, Ribi adjuvant system, and TiterMax were efficacious in eliciting a protective response with GST-PYC2 in BALB/c mice, saponin failed to induce protection, although significant levels of PYC2-specific antibodies were produced in all immunized animals. This protection depended on the mouse strain since immunization of Swiss Webster mice with GST-PYC2 in alum did not produce levels of PYC2-specific antibodies comparable to those in BALB/c mice nor did it induce any demonstrable level of protection against parasite challenge. Swiss Webster mice were protected, however, when immunized with GST-PYC2 in other adjuvants. Immunization with PYC2, isolated free of GST induced lower levels of antigen-specific antibody; only those animals given PYC2 in Freund's adjuvant demonstrated a significant degree of protection, suggesting the possibility (of additional cellular effector mechanisms. These findings demonstrate that adjuvant, host genotype, and the fine specificity of the response significantly influence the protection induced by the carboxyl terminus of MSP-1 in vivo and illustrate the need to consider these factors in evaluating MSP-1 as a vaccine component.     

Merozoite surface protein 4/5 provides protection against lethal challenge with a heterologous malaria parasite strain

Immunization with merozoite surface protein 4/5 (MSP4/5), the murine malaria homologue of Plasmodium falciparum MSP4 and MSP5, has been shown to protect mice against challenge by parasites expressing the homologous form of the protein. The gene encoding MSP4/5 was sequenced from a number of Plasmodium yoelii isolates in order to assess the level of polymorphism in the protein. The gene was found to be highly conserved among the 13 P. yoelii isolates sequenced, even though many of the same isolates showed pronounced variability in their MSP1(19) sequences. Nonsynonymous mutations were detected only for the isolates Plasmodium yoelii nigeriensis N67 and Plasmodium yoelii killicki 193L and 194ZZ. Immunization and challenge of BALB/c mice showed that the heterologous MSP4/5 proteins were able to confer a level of protection against lethal Plasmodium yoelii yoelii YM challenge infection similar to that induced by immunization with the homologous MSP4/5 protein. To explore the limits of heterologous protection, mice were immunized with recombinant MSP4/5 protein from Plasmodium berghei ANKA and Plasmodium chabaudi adami DS and challenged with P. y. yoelii YM. Interestingly, significant protection was afforded by P. berghei ANKA MSP4/5, which shows 81% sequence identity with P. y. yoelii YM MSP4/5, but it was abolished upon reduction and alkylation. Significant protection was not observed for mice immunized with recombinant P. c. adami DS MSP4/5, which shows 55.7% sequence identity with P. y. yoelii YM MSP4/5. This study demonstrates the robustness of MSP4/5 in conferring protection against variant forms of the protein in a murine challenge system, in contrast to the situation found for other asexual-stage proteins, such as MSP1(19) and AMA1.