Research Toward Vaccines Against Malaria

Louis Miller (National Institute of Allergy and Infectious Diseases) and Stephen Hoffman (Naval Medical Research Institute) review progress toward developing malaria vaccines. They argue that multiple antigens from different stages may be needed to protect the diverse populations at risk, and that an optimal vaccine would Induce Immunity against all stages. Vaccines for African children, In whom the major mortality occurs, must induce immunity against ase,cual blood stages. Research toward vaccines against malaria

Malaria is one of the major causes of the only stage of the life cycle that causes disease and death between the Tropic LOUIS H. MILLER1 disease. The stages before the asexual of Cancer and Tropic of Capricorn. & STEPHEN L. HOFFMAN2 blood stage are lumped together and Plasmodium falciparum has an especially called pre-erythrocytic. A small propor profound impact on infants and children in sub-Saharan Africa, tion of the asexual blood stages differentiate into sexual stages, where its effect on health is increasing as chloroquine resistance the gametocytes in red cells that infect mosquitoes; vaccines to spreads across the continent. We believe that vaccination the mosquito stages are called transmission-blocking vaccines. against P. falciparum is the intervention with the greatest poten The parasites' multistage life cycle and the fact that immune re tial to reduce malaria-associated severe morbidity and mortality sponses that recognize one stage often do not affect the next in areas with the most intense transmission and that it may do stage have made vaccine development for malaria more diffi so without necessarily preventing blood stage infection. Malaria cult than for antiviral and antibacterial vaccines. vaccines also would be the optimal method for preventing The pre-erythrocytic stages include sporozoites and para malaria in travelers to countries where malaria is transmitted. sites that develop in hepatic parenchymal cells. Sporozoites, Such vaccines will have to entirely prevent the development of after inoculation by the mosquito, spend only a few minutes any clinical manifestations of infection with P. falciparum and in the blood stream before they invade hepatocytes, in which of the second most common human malaria parasite, P. vivax. development takes a minimum of five and a half days in P. This will require preventing blood stage infection. falciparum. Antibodies to sporozoites block their infection of Infants and young children at risk from P. falciparum in Africa hepatocytes. The liver stage in rodent malarias, and probably and nonimmune travelers to areas endemic for P. falciparum and in human malarias, is attacked primarily by CDS· T cells, but P. vivax represent the extremes of target groups in whom malaria also by co4• T cells. These cells can directly eliminate the in- vaccines would be useful. Even fected cells or secrete inter in countries where the overall Slngea Goals Ta,vel Populallon feron (IFN)-y that induces

numbers of afflicted individuals A Prevent dlJHH A Nonrnmune nitric oxide-dependent killing by blockmg "1111C · lravelen: and are not as great as in Africa, lion belofe re-nllfrom within hepatocytes. Killing all emargence ar•u of low parasites during the sporo malaria wreaks its havoc in from IMtr t,ansmtSSIOf\ many epidemiologic settings (o.g., lndta) zoite stage or during the he and population groups, dramati B. Reduu disease B. Ch,khn and patic infection prevents all by a "8CC11lt p,ognan1 women cally undermines the productiv lhal combtnes Ill.,... of high disease. Thus, stand-alone par 11ally •H• transrn.ssion ity of workers, and drains pr1t•orythrocyuc (o.g., Afnca) vaccines against pre-erythro national budgets. From the tnd blood suoge cytic stages are being devel- components Amazon basin to the Indian sub oped for protection of

continent, Southeast Asia, and Reduce disuse by Chddlen and travelers. They may also have raduang blood pregnanl women a role in the general popula Oceania, malaria is often a major stage asuuaf ,n r&as ol high national public health problem. ~,tebu,oon transm1SSt0n tion in countries with low en- (eg., Afnca) In these settings, vaccines offer demicity such as India. Their the greatest potential for reduc application on their own in ing malaria's debilitating effects. areas with intense transmis The spread of drug-resistant P. sion is considered by some to falciparum and the recent emer PreYonUon or be inadvisable. There is a con 1rat111mls1lon cern that if the vaccines were gence and spread of chloro A Eradlcabon A. Vacanate enllr• quine-resistant P. vivax have community 1n too good, they would prevent ISantibody or cellular immunity. gated if a pre-erythrocytic

stage vaccine were com invades midgut epithelial cells. Parasites on the midgut ep bined with a blood stage ithelium develop into thousands of sporozoites that invade vaccine, a strategy some the salivary glands to transmit the infection while feeding on consider to be optimal. people. Transmission-blocking vaccines under development The blood stage begins contain gamete, zygote and ookinete surface proteins and se when parasites leave liver creted proteins from ookinetes. They induce antibodies that, cells and invade red cells. when ingested with the blood meal, block fertilization, de The parasite develops and velopment, penetration of the peritrophic membrane or in j multiplies within the red vasion of midgut epithelium. Such vaccines would block ! cell; 16 or more merozoites transmission of mutant parasites resistant to drugs or to vac j appear 48 hours after inva cines against other stages. Furthermore, they could be used " sion (in P. falciparum and P. for epidemic control in areas of unstable transmission. The f vivax). The infected red cell third use, in combination with other interventions such as .u lyses, releasing merozoites, insecticides, would be for malaria eradication in island com Human red cells infected with each capable of invading munities (such as Sri Lanka) where transmission is low. Plasmodium falciparum. other red cells. The para Several constructs that contain a surface protein on sitemia can increase ten-fold ookinetes, Pfs25, are under development and are being tested each 48 hours. No disease occurs until the parasitemia in normal volunteers. reaches a level critical for induction of fever and shaking Unlike most viral vaccines, malaria vaccines will be pro chills. The serious complications of anemia and coma ap duced as subunits that will contain only a few of the esti pear later in the infection as the parasitemia rises. The cur mated 5,000-7,000 proteins in the parasite or parts of these rent vaccine approach is to design blood stage vaccines that proteins. Below, we identify obstacles to successful subunit will prime the individual for boosting of immunity by the vaccine development for pre-erythrocytic and blood stage parasite during each infection, thereby controlling para vaccines and present our perspective on how to overcome sitemia before serious complications occur. There are data them. For those interested in the details of antigens and indicating that disease in young children in Africa is influ strategies for vaccine development, see refs. 4 & 5. enced by the intensity of transmission (number of infectious bites per year), the child's genetics, and perhaps virulence Pre-erythrocytic stage vaccines factors in P. falciparum 1• For example, in many areas, cere Immunization of mice and humans with radiation-attenu bral malaria (coma and seizures with high mortality) is un ated sporozoites of rodent and human malaria parasites, re common before two years of age2• In low-transmission areas spectively, induces complete protection against the (< 15 bites/individual by infected mosquitoes per year), se pathologic and clinical manifestations of the infection. The vere anemia occurs early in life and cerebral malaria later, protective immunity prevents the parasites from emerging suggesting a developmental susceptibility to cerebral from the liver into the bloodstream, is not strain specific, malaria in children2 • In high-transmission areas (> 100 and lasts in humans for at least nine months (ref. 6 for re bites/individual by infected mosquitoes per year), severe view). This would be an ideal vaccine for travelers but is im anemia occurs early in life and cerebral malaria occurs infre practical because it requires exposure to thousands of quently1·3, possibly because children become immune before infected, irradiated mosquitoes. the age at which cerebral malaria occurs. Despite these ob Efforts for 30 years have focused on characterizing the servations, the pathogenesis of disease in African children is protective immune responses and their antigenic targets and poorly defined. Current development efforts assume that re developing vaccine delivery systems to induce comparable ducing parasitemia will limit all complications of malaria. immunity. Antibodies induced by irradiated sporozoites This assumption may be wrong, and some complications have modest antisporozoite inhibitory activity in mice and may be worsened by vaccination. The humans, but protection in mice is al cost of vaccine trials, however, de ways dependent on CD8' T lympho mands this simplifying assumption to cytes. These T cells eliminate test the multiple antigens with differ parasite-infected hepatocytes from cul ent formulations that will be needed ture. The CD8· T cell immunity may be for blood stage vaccine development. the result of lysis of infected hepatoc The young children used to test se tyes by cytotoxic T lymphocytes (CTL), lected candidates need to be followed but the T cells also release IFN-y, which closely for increased complications induces the infected hepatocyte to pro after vaccination. duce nitric oxide that eliminates the Throughout the blood stage of the in infected cells. Depending on the indi- fection, a percentage of asexual blood ~ vidual and the antigen, both mecha i stage parasites convert to gametocytes nisms of killing are probably that are infectious to mosquitoes. important6 • Humans immunized with Factors in the mosquito midgut induce irradiated P. fa lciparum sporozoites gametocytes to transform into gametes have been shown to have CD8· CTLs and zygotes that develop into ookinetes, Human red cells infected with gametocytes against four different P. falciparum pro- the stage that penetrates the mosquito's (sexual stage) and ring forms (asexual stage) of teins expressed in infected hepatoctyes chitinous peritrophic membrane and Plasmodium falciparum. (refs. 6 & 7 for review); subunit vac-

NATURE MEDICINE VACCINE SUPPLEMENT • VOLUME 4 • NUMBER 5 • MAY 1998 521 REVIEW ········ • ©...... 1998...... Nature.. Publishing Group http://www.nature.com/naturemedicine

cines of the rodent malaria parasite orthologues of three of these induce cos· T cell-dependenta- 10 and, in one case, also

IFN-y- and nitric oxide-dependent protection in mice 10 • It must be remembered, however, that irradiated sporozoites may induce immunity against hundreds or thousands of proteins expressed in infected hepatocytes, not just the handful currently under study. The challenge, as in vaccinology in general, has been to de velop subunit vaccines that induce protective CDS· T cell re sponses in humans. Intramuscular immunization with recombinant, attenuated vaccinia expressing several liver stage P. falciparum proteins induced modest CTLs but mini P/asmodium falciparum-infected red cell (center) binding uninfected red cells. Courtesy of Kjell Hultenby & Carl Johan Treutiger. mal protection11 • Based on work in mice and rhesus mon keys, there is now great hope that ongoing clinical trials with DNA vaccines will demonstrate cos· T cell-dependent pro bospondin-related anonymous protein (TRAP) (ref. 18). Thus tection in humans12• Regardless of the results of these early far, efforts to induce highly inhibitory antibodies against this trials with DNA alone, preliminary studies in mice indicate domain, which bears some homology to human throm that DNA vaccine-induced immunity and protection against bospondin, have been unsuccessful. Additional sporozoite sur malaria can be improved by co-administration of plasmids face proteins have been identified, and some have been shown expressing cytokines13 and by prime boost approaches11•15 • It to induce inhibitory antibodies, but to date none has been in is likely that these and other16 approaches will be studied in cluded in an experimental human vaccine. humans in the next few years. Once the problem of optimal During the past decade, the major efforts on pre-erythro induction of these responses is achieved, a successful vaccine cytic malaria vaccine development have focused on induc will have to include adequate cos• T cell epitopes to over ing antibodies to the sporozoite surface and CDS· T ceUs come the problems of genetic restriction of cos· T cell re against infected hepatoctyes; however, there is now consid sponses', variation of epitopes in the field' , and perhaps even erable interest in protective co4• T cell responses against in epitope antagonism". fected hepatocytes. New data indicate that in several strains Although the protective immunity induced by irradiated of mice, protection induced by the irradiated sporozoite vac sporozoites is probably primarily mediated by T cells, the cine is eliminated by treatment of the mice with anti-CD4 best (80-100%), most consistent and least genetically re antibodies or anti-CD8 antibodies. CD4· T cell clones stricted protection achieved in rodents has been with vac against the PyCSP eliminate infected hepatocytes from cul cines that induce antibodies that prevent sporozoite ture and adoptively transfer protection23 • Immunization invasion of hepatocytes'". Monoclonal antibodies against a with linear synthetic peptides from two liver stage proteins tandem repeat domain on the circumsporozoite protein induces consistent, genetically restricted, CD4· T cell- and (CSP) inhibit sporozoite invasion of hepatocytes in vitro and IFN-y-dependent protection in mice24 • The RTS,S PfCSP vac passively transfer protection to mice and monkeys. This do cine described above is a potent activator of IFN-y-producing main is conserved among au P. falciparum isolates. Based on co4• T cells. In addition to antibodies, co4• T cells may these studies, clinical trials with PfCSP repeat region recom have played an important role in the protection of volun binant protein and synthetic peptide vaccines have been teers induced by this vaccine. conducted for more than a decade. The principle that such The irradiated sporozoite vaccine has served as an impor vaccines could protect was proved in the first trials in which tant model for developing pre-erythrocytic stage vaccines; aluminum hydroxide was used as adjuvant; two of the four however, particularly in regard to protective antibodies and volunteers with the highest levels of antibodies were com CD4· T cell responses, it seems that we can do better than the pletely protected19•20• Increasing the complexity of the im model. The challenge for the future is to develop a pre-ery munogen and including more potent adjuvants has resulted throcytic vaccine capable of inducing not only protective an in a steady increase in immunogenicity. Recently, RTS,S-a tibodies against sporozoites but also co4• and CDS• T cell hybrid containing the central repeats and most of the C-ter responses against infected hepatocytes, incorporating into minus of the PfCSP fused to hepatitis B surface antigen in a these vaccines the important parasite proteins expressed on complex adjuvant mixture (SBAS4) 31 - completely protected the sporozoite surface or early in the liver stage identified by six of seven volunteers against sporozoite challenge three the malaria genome project. weeks after the last immunization. Volunteers immunized with RTS,S in other adjuvants had little protection. Blood stage vaccines Anti-PfCSP antibodies were extremely high in most protected Two observations make us confident that a blood stage vac volunteers but some unprotected volunteers had comparable cine is possible. First, people in endemic areas become im antibody titers. The six protected volunteers were challenged mune after repeated infections. Immune older children and approximately six months after the first challenge and one adults have blood stage infections but are usually asympto was protected22• A field study of RTS,S-SBAS4 in The Gambia matic. Thus, immunity is measured by the absence of dis and additional clinical trials of RTS,S and of a PfCSP multiple ease, not by resistance to reinfection, and the goal of a blood antigenic peptide vaccine18 are now in progress. stage vaccine is to prevent disease, not infection. Second, A ligand that binds to a receptor on hepatoctyes has been immunization of animals induces protection that is more ef identified on the CSP and on a second sporozoite surface pro fective than immunity induced by chronic, repeated infec tein called sporozoite surface protein 2 (SSP2) or throm- tions. For example, monkeys immunized against P. knowlesi

522 NATURE MEDICINE VACCINE SUPPLEMENT • VOLUME 4 • NUMBER 5 • MAY 1998 © 1998 Nature Publishing Group http://www.nature.com/naturemedicine • ····· ···· ··· ···REVIEW parasites in Freund's complete adjuvant can rapidly clear is less variable than other parts of the molecule and binds to challenge infections with both homologous and heterolo this common receptor on endothelium, it may have a con

gous strains25• served structure. Thus, it may be possible to develop an im Effective approaches to antigen identification have been munogen of this domain that induces antibody to block through monoclonal antibodies to merozoite surface mole endothelial attachment. Although this is counterintuitive cules and organellar proteins and through the use of Triton (variation is designed to escape immunity), the epitopes X-114 to differentially solubilize surface proteins26. The most around this domain may not normally induce immunity be promising candidate to date has been a merozoite surface cause they are cryptic, that is, not immunogenic during in protein (MSP-1) that was identified by monoclonal antibod fection, but protection can be induced after vaccination. ies. Vaccination with immunoaffinity-purified MSP-1 from a If many antigens derived from blood stage parasites have rodent parasite27 and from P. falciparum induced protection been identified and some have been shown to be effective in mice and monkeys, respectively. Protective monoclonal after vaccination of mice and monkeys, why have they not antibodies reacted with an epidermal growth factor-like do been tested in humans? The main obstacle to vaccine devel main in the C-terminus [MSP-1(19)]; immunization with re opment has been the absence of industrial interest in blood combinant proteins from this region protected against lethal stage vaccines because they do not expect profits that would P. yoelii in mice and against P. falciparum in Aotus monkeys5. warrant the investment. Thus, a gap exists between the re In MSP-1(19)-immunized mice, immunity is primarily anti search laboratories that lack the costly facilities to produce body mediated; high titer sera in immunized mice correlates blood stage vaccines for human trials, and industry that has with protection28. Other promising antigens and immune the capability, but not the motivation, to produce them. mechanisms such as antibody-dependent cellular inhibi Transition from a research laboratory setting to vaccine pro tion29 are also under study30• duction for human trials will require a different mind set for Patarroyo and his team from Colombia identified potential the scientist, and costly facilities5. The National Institutes of candidate antigens by immunizing Aotus monkeys with pep Health is in the process of setting up a unit to attempt to ef tides linked to bovine serum albumin and challenging them fect this transition. with virulent P. falciparum'1. The three peptides that induced some protection were synthesized as a single 66 amino acid Looking ahead peptide, polymerized and bound to aluminum hydroxide. The most exciting period for the scientific community will This vaccine was tested in Latin America, Africa and Asia with be when we extensively test vaccines in endemic countries, evidence of protection in some studies32 but not in others33 • and particularly in Africa. We will be studying the effect of The lack of in vitro correlates of immunity made interpreta vaccines in the natural host and under conditions in which tion of the variable results difficult. the host is repeatedly infected. One concern for an anti Parasite ligands for invading red cells and for binding of body-based vaccine is the selection of parasites resistant to infected red cells to endothelium may be candidates for vac the vaccine. We know that every parasite protein exposed to cine development. Apicomplexa usually invade host cells by the immune system has many polymorphisms. It is possible specific ligands in the apical end. These proteins and the do that structural requirements for critical domains limit the mains within the proteins for red cell binding have been flexibility of the parasite, especially if the vaccine contains identified in Plasmodium. One domain is conserved in P. fa/ multiple epitopes. Combining a transmission-blocking vac ciparum and P. vivax, although the red cell receptors differ34·35. cine with a blood stage vaccine may prolong the life of a vac Recombinant proteins from these domains are now being cine by reducing transmission. tested for inducing neutralizing antibodies in monkey malar In addition to antigenic polymorphism, the parasite may ias. One concern is that these receptors may be exposed on have evolved other mechanisms of immune evasion. There the red cell surface for a short period between release of may be ways that the parasite drives a nonprotective im merozoites from one red cell and invasion of another red mune response or induces apoptosis of immune T cells. Th e cell. A further concern is that the receptors are found within large number of parasite proteins with repeat epitopes sug an organelle at the apical end and may only be released for gested to Anders the possibility that crossreactions may in binding after the merozoite has attached to a red cell, a step terfere with maturation of the immune response to produce that signals the merozoite to begin the invasion process. A high affinity antibodies38. The low complexity of many membrane protein on merozoites involved in signaling malarial proteins with no fixed globular structure39 may also would be a prime candidate for a vaccine, but none has been be a mechanism of immune evasion. Hopefully, the globular identified to date. structure of the vaccine candidates under study such as the Infected red cells bind to endothelium in a receptor-spe C-terminus of MSP-1 and AMA-1 may circumvent these cific manner to protect the infected red cell from the killing problems. effects of the spleen. A number of endothelial receptors have Another scenario is the possible effect of repeated infec been identified; the parasite ligands for these receptors are tions that may give rise to greater immunity in humans after encoded by a highly variant protein with up to 150 copies vaccination, more so than we have seen in animal studies. per parasite genome. Curiously, the parasite ligands for bind We know that previously infected animals have a secondary ing endothelium have been found on a domain that contains response to first immunization. Although vaccination of homology to the red cell binding domain involved in inva rhesus monkeys with AMA-1 of P. knowlesi afforded minimal sion36. Each variant protein on the red cell surface has multi protection, the immunized, previously infected animals ple copies of this domain. One of these domains binds were strongly protected from reinfection40• Thus, infection endothelium via CD36 (ref. 37), which is thought to be a re after immunization was critical to induce high levels of im ceptor to which most infected red cells bind. As this domain munity.

Conclusion protein vaccine against Plasmodium falciparum malaria. RTS, S Malaria Vaccine The malaria community has characterized immune mecha Evaluation Group. N. Engl. J. Med. 336, 86-91 (1997). 22. Stoute, J.A. et al. Long-term efficacy and immune responses following immu• nisms responsible for protection, identified numerous target nization with the RTS,S malaria vaccine. /. Infect. Dis. (1998) (in press). antigens and is on the cutting edge of research in vaccinology. 23. Renia, L et al. In vitro activity of CD4+ and CDS+ lymphocytes frommice im• In the coming years, we will use the advances in immunology munized with a synthetic malaria peptide. Proc. Natl. Acad. Sci. USA 88, 7963-7967 (1991 ). and vaccinology and discover new potential vaccine targets 24. Wang, R. et al. Protection against malaria by Plasmodium yoelii sporozoite sur• from the Plasmodium genome project. We know that our goals face protein 1 linear peptide induction of CD4+ T cell- and IFN-y-dependenl elimination of infected hepatocytes. /. lmmunol. 157, 4061-4067 (1996). are achievable because of successful vaccination of humans 25. Mitchell, G.H., Butcher, G.A. & Cohen, S. Merozoite vaccination against with irradiated sporozoites and monkeys with blood stage par Plasmodium knowlesi malaria. Immunology 29, 397-407 (1975). asites and because of the fact that people become immune. It is 26. Smythe, J.A. et al. Identification of two integral membrane proteins of Plasmodium falciparum. Proc. Natl. Acad. Sci. USA 85, 5195-5199 (1988). now time to make the commitment to GMP vaccine manufac 27. Holder, A.A. & Freeman, R.R. Immunization against blood stage rodent ture and human trials, including phase III mortality studies. malaria using purified parasite antigens. Nature 294, 361-366 (1982). 28. Hirunpetcharat, C., et al. Complete protective immunity induced in mice by The seriousness of the human problem demands that we de immunization with the 19-kilodalton carboxyl-terminal fragment of the mero•

velop this most cost-effective tool. zoite surface protein (MSP1 19) of Plasmodium yoelii expressed in Saccharomyces cerevisiae: correlation of protection with antigen-specific antibody titer, but 1. Miller, LH., Good, M.F. & Milon, G. Malaria pathogenesis. Science 264, not with effector CD4' T cells./. lmmunol. 159, 3400-3411 (1997). 1878-1883 (1994). 29. Bouharoun-Taylor, H., Oeuvray, C., Lunel, F. & Druilhe, P. Mechanisms under• 2. Marsh, K. Malaria-a neglected disease? Parasitology 104, S53-S69 (1992). lying the monocyte-mediated antibody-dependent killing of Plasmodium falci• 3. Snow, R.W. et al. Relation between severe malaria morbidity in children and parum asexual blood stages./. Exp. Med. 182, 409-418 (1995). level of Plasmodium falciparum transmission in Africa. Lancet 349, 1650-1654 30. Holder, A.A. Preventing merozoite invasion of erythrocytes. In: Malaria Vaccine (1997). Development: A Multi-Immune Response Approach, Hoffman, S.L (ed.) 4. Hoffman, S.L (ed.) Malaria Vaccine Development: A Multi-Immune Response Washington, D.C.: American Society of Microbiology, pp. 77-104 (1996). Approach. Washington, D.C.: ASM Press (1996). 31. Patarroyo, M.E. et al. Induction of protective immunity against experimental 5. Good, M.F., Kaslow, D.C. & Miller, L.H. Pathways and strategies for develop• infection with malaria using synthetic peptides. Nature 328, 629-632 (1987). ing a malaria blood stage vaccine. Ann. Rev. lmmunol. 16, 57-87 (1998). 32. Alonso, P.L et al. Randomised trial of efficacy of SPf6 vaccine against 6. Hoffman, S.L, Franke, E.D., Hollingdale, M.R. & Druilhe, P. Attacking the in• Plasmodium falciparum malaria in children in southern Tanzania. Lancet 344, fected hepatocyte. In: Malaria Vaccine Development: A Multi-Immune Response 1175-1181 (1994). Approach, Hoffman, S.L. (ed.) Washington, D.C.: American Society of 33. d'Allessandro, U. et al. Efficacy trial of malaria vaccine SPf66 in Gambian in• Microbiology, pp. 35-75 (1996). fants. Lancet 346, 462-467 (1995). 7. Doolan, D.L et al. Degenerate cytotoxic T cell epitopes from Plasmodium falci• 34. Chitnis, C.E. & Miller, L.H. Identification of the erythrocyte binding domains of parum restricted by multiple HLA-A and HLA-B supertype alleles. Immunity 7, Plasmodium vivax and Plasmodium knowlesi proteins involved in erythrocyte in• 97-112 (1997). vasion./. Exp. Med. 180, 497-506 (1994). 8. Aggarwal, A. et al. Oral Salmonella: malaria circumsporozoite recombinants in• 35. Sim, B.K.L, Chitnis, C.E., Wasniowska, K., Hadley, T.J. & Miller, L.H. Receptor duce specific cos• cytotoxic T cells./. Exp. Med. 172, 1083-1090 (1990). and ligand domains for invasion of erythrocytes by Plasmodium falciparum. 9. Khusmith, S., et al. Protection against malaria by vaccination with sporozoite Science 264, 1941-1944 (1994). surface protein 2 plus CS protein. Science 252, 715-718 (1991 ). 36. Su, X.-Z. et al. The large diverse gene family var encodes proteins involved in 10. Doolan, D.L et al. Circumventing genetic restriction of protection against the cytoadherence and antigenic variation of Plasmodium falciparum-infected malaria with multigene DNA immunization: CDS' cell-, interferon--, and ni• erythrocytes. Ce// 82, 89-100 (1995). tric oxide-dependent immunity./. Exp. Med. 183, 1739-1746 (1996). 37. Baruch, D.I. et al. Identification of a region of PIEMPl that mediates adherence 11. Ockenhouse, C.F. et al. Phase I/Ila safety, immunogenicity, and efficacy of of Plasmodium falciparum infected erythrocytes to CD36: conserved function NYVAC-Pf7, a pox-vectored, multiantigen, multistage vaccine candidate for with variant sequence. Blood 90, 3766-3775 (1997). Plasmodium falciparum malaria./. Infect. Dis. (in the press). 38. Anders, R.F. Multiple cross-reactivities amongst antigens of Plasmodium falci• 12. Hoffman, S.L. et al. Toward clinical trials of DNA vaccines against malaria. parum impair the development of protective immunity against malaria. lmmunol. Cell Biol. 75, 376-381 (1997). Parasite lmmunol. 8, 529-539 (1986). 13. Weiss, W. et al. A plasmid encoding murine GMCSF increases protection con• 39. Wooton, J.C. Non-globular domains in protein sequence: automated segmen• ferred by a malaria DNA vaccine. /. lmmunol. (1998) (in press). tation using complexity measures. Computers Chem. 18, 269-285 (1994). 14. Schneider, J. et al. Enhanced immunogenicity for CDS+ T cell induction and 40. Deans, J.A., et al. Vaccination trials in rhesus monkeys with a minor, invariant, complete protective efficacy of malaria DNA vaccination by boosting with Plasmodium knowlesi 66 kD merozoite antigen. Parasite lmmunol. 10, 535-552 modified vaccinia virus Ankara . Nature Med. 4, 397-402 (1998). (1988). 15 Sedegah, M. et al. Boosting with recombinant vaccinia increases immuno• genicity and protective efficacy of malaria DNA vaccine. Proc. Natl. Acad. Sci. USA (1998) (in press). 1 16. Rodrigues, E.G., Zavala, F., Eichinger, D., Wilson, J.M. & Tsuji, M. Single immu• Laboratory ofParasitic Diseases, National Institute ofAllergy and nizing dose of recombinant adenovirus efficiently induces CDS+ T cell-medi• Infectious Diseases, National Institutes of Health, Bethesda, ated protective immunity against malaria./. lmmunol. 158, 1268-1274 (1997). Maryland 20892 USA 17. Gilbert, S.C. et al. Association of malaria parasite population structure, HLA, and immunological antagonism. Science 279, 1173-1177 (1998). 2 Malaria Program, Naval Medical Research Institute, Bethesda, 18. Sinnis, P. & Nussenzweig, V. Preventing sporozoite invasion of hepatocytes. Maryland 20889-5607 USA In: Malaria Vaccine Development: a Multi-Immune Response Approach, Hoffman, Correspondence should be addressed to L.H.M. S.L. (ed.) Washington, D.C.: American Society of Microbiology, pp. 15-33 (1996). e-mail: louis_ [email protected] 19. Herrington, D.A. et al. Safety and immunogenicity in man of a synthetic pep• tide malaria vaccine against Plasmodium falciparum sporozoites . Nature 328, The opinions and assertions herein are those of the author (S.L.H.) and are 257-259 (1987). not to be construed as official or as reflecting the views of the U.S. Navy or 20. Ballou, W.R. et al. Safety and efficacy of a recombinant DNA Plasmodium falci• parum sporozoite vaccine. Lancet 1, 1277-1281 (1987). naval services at large. This work was supported in part by STO F 6.3a- 21 Stoute, J.A. et al. A preliminary evaluation of a recombinant circumsporozoite 63002AA0101HFX.1433.

524 NATURE MEDICINE VACCINE SUPPLEMENT • VOLUME 4 • NUMBER 5 • MAY 1998