Antigenic structure and related aspects of the biology of plasmodia: the present situation*

Bulletin of the World Health Organization, 59 (3): 371-381 (1981) Antigenic structure and related aspects of the biology of plasmodia: the present situation* SCIENTIFIC WORKING GROUP ON THE IMMUNOLOGY OF MALARIA Major progress has been made in the production of monoclonal antibodies against various Plasmodium species and developmental stages. Some ofthe monoclonal antibodies, especially those directed against sporozoites, have proved to be protective in rodent and simian systems. A ntigenic analysis has been advancedby the application ofthese monoclonal antibodies, and recent studies have concentrated on membrane-bound antigens. Whilepromising leads have emergedfrom the in vitroproduction ofviable gametocytes of Plasmodium falciparum and from the in vitro cultivation of exoerythrocytic forms of rodent plasmodia, little progress has been made in the scaling up of in vitro cultures of asexualforms ofP. falciparum. Techniquesfor the separation, isolation, andpurification of parasite material have been improved. Areas that require urgent attention are identified, withparticularemphasis on antigenic analysis and the production of antigensfor vaccine development and immunodiagnosis. The elaboration of a practical and effective vaccine against human malaria has been the main objective of the Scientific Working Group on the Immunology of Malaria since its inception five years ago. The belief that such a project is feasible was originally based on the observation that malarial infections do induce specific immune responses which are protective. The best characterized of these are antibody-mediated, directed against stage-specific and species-specific antigens of sporozoites and merozoites, and act by preventing entry of these extracellular forms into target cells of the host. However, the clinical effective- ness of such specific immunity is partially offset by the antigenic complexity of the developing parasite and its ability to compromise the host immune response, for example, through the induction of immunosup- pression and polyclonal lymphocyte activation and possibly by the production of soluble antigens. Exper- imental vaccination procedures, for example with irradiated sporozoites or non-invasive merozoites, have been shown to induce far more effective immun- ity than the corresponding natural infection, indi- cating that such non-viable forms are able to stimulate protective immunity without fully activating the processes that favour parasite survival. These obser- * Requests for reprints should be addressed to Chief, Research and Technical Intelligence, Malaria Action Programme, World Health Organization, 1211 Geneva 27, Switzerland. I This article is based on the report of the fourth meeting of the Scientific Working Group on the Immunology of Malaria, which was held in Geneva in October 1980, under the auspices of the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases. A full list of participants is given on pages 380-381. vations suggest that isolated plasmodial antigens of appropriate specificity may also induce immunity and constitute effective vaccines. These considerations have stimulated interest in the detailed antigenic structure of malaria parasites. Anti- genic analysis using conventional polyclonal immune sera and various methods of immunoelectrophoresis, have revealed the great complexity of plasmodial anti- gens and the difficulty of isolating those with possible functional importance. More recently, the application of somatic cell hybridization to produce cell lines secreting monoclonal antibodies of defined specificity has greatly expedited the task of antigen isolation and characterization. Work underway with sporozoites, gametes, and blood-stage parasites promises to yield useful new serodiagnostic tests and may well culmi- nate in the development of a malaria vaccine for man. IN VITRO PRODUCTION OF PARASITES Asexual erythrocytic stages of P. falciparum Methods for the continuous cultivation of erythro- cytic stages of Plasmodium fakciparum in vitro have not been significantly improved or altered in recent years. Different laboratories have used vessels ranging from microtitration plates to large dishes according to their experimental needs. Recently, however, evidence has been obtained that the continuous flow method is particularly suitable for long-term culture as it pro- vides parasites in the best condition. This technique has supported growth of a simian malaria, P. inui, 4066 - 371- 372 SCIENTIFIC WORKING GROUP when other methods have failed. The presence of leukocytes in the blood of a single donor used as a source of red cells does not influence parasite multiplication. However, for some purposes, it may be necessary to remove leukocytes. Pooled serum from ten or more donors permits a reduction in the serum concentration in culture medium from 10% to 57o. Freshly collected pooled animal sera are unsuitable except for pig and goat sera which can support growth for two cycles. Although pooled adult bovine serum is unacceptable by itself, supplementation with neopeptone and protease pep- tone no. 3 provides continuous cultivation without adaptation, but at a significantly lower rate than pooled human sera. Several technological innovations, including auto- matic systems for changing the medium and suspen- sion cultures, have been employed for the large-scale production of parasites in culture, though the feasi- bility of the latter has not yet been fully assessed. Gametocytes of P. falciparuma Fresh isolates of P.falciparum introduced into culture produce gametocytes. Gametocyte maturation has been achieved by maintaining gametocyte-con- taining cultures at 37 °C, and susceptible mosquitos have been infected with gametes originating from this material. The procedures for production and matur- ation of gametocytes in culture are still under development in several laboratories. Exoerythrocytic forms (EEF) ofmammalian plasmodiab Only two laboratories have published information regarding the in vitro cultivation of EEF of mam- malian plasmodia. So far, purified suspensions of P. yoelii sporozoites introduced into embryonic rat brain cell monolayers in combination with MEM Alpha medium have given the best results, but have not yet produced mature segmenting schizonts. Two species of rodent malaria, namely P. berghei and P. yoelii, have been cultivated. PURIFICATION OF ASEXUAL ERYTHROCYTIC PARASITES Synchronous parasite material is essential for the study of antigenic changes associated with parasite a Recently, J. Vanderburg reported success in the in vitro produc- tion of P. falciparum gametocytes, which proved to be reproducibly infective to Anopheles stephensi (USAID meeting on Malaria Immunology and Vaccine Development, Washington, DC, 14-16 January 1981). b Hollingdale recently succeeded in obtaining complete tissue schizogony and subsequent production in vitro of infective tissue merozoites of P. berghei in W138 lung cells. (USAID meeting on Malaria Immunology and Vaccine Development, Washington, DC, 14-16 January 1981). maturity. In some plasmodial species, e.g., P. chabaudi and P. knowlesi, synchronism is a feature of infection, but in vitro cultures of P.falciparum, for example, are often asynchronous. Synchronization of such cultures can be achieved by the use of sorbitol and sedimentation techniques in various combi- nations. The separation of infected from uninfected erythro- cytes and, to some extent, the stage-specific separ- ation of infected erythrocytes, can be achieved by density-gradient centrifugation, sedimentation, and free-flow electrophoresis. Further, separation of the parasites from their host cells can be carried out by immune and saponin lysis and physical disruption. These methods yield para- sites contaminated with red cell membranes. How- ever, relatively uncontaminated parasites can be obtained by further processing using free-flow electro- phoresis or gradient-centrifugation techniques. Isolation of merozoites has been achieved by chemical lysis, immune lysis, physical disruption, free-flow electrophoresis, and natural release. Of these, natural release methods, including cell sieving (P. knowlesi), lectin affinity chromatography (P. chabaudi, P. knowlesi, and P.falciparum), and release into culture followed by gradient-density centrifugation (P.falciparum), yield the purest and most satisfactory material. Merozoites isolated by these methods are morpho- logically intact but yields are low. The degree of contamination with red cell membranes varies with the isolation method used. The preservation of P.falci- parum merozoites has been improved by the addition of 100 ml of red cell extract per litre of culture medium. ERYTHROCYTE MEMBRANE CHANGES IN MALARIA In some species of malaria the surface of the infected erythrocyte is morphologically modified to produce caveolae-vesicle complexes, "metabolic windows", and knob protrusions. Even when the gross morphology of the infected erythrocyte is unaltered there are functional changes in membrane properties, such as altered uptake of metabolites and drugs, enhanced osmotic fragility, and an inability to extrude intracellular sodium. Prolonged in vitro cultivation of P.falciparum has resulted in the emergence of a subpopulation of parasites which do not produce the typical erythrocyte membrane alter- ations known as knobs. Light microscopy using inter- ference optics with video-image enhancement allows direct observation of these erythrocyte membrane alterations. Knobby (K+) and knobless (K-) parasite subpopulations have been cloned by three labora- ANTIGENIC STRUCTURE OF PLASMODIA tories. Studies concerning the antigenicity and patho- genicity of these clones are being conducted. In general, these structural and functional alterations cannot be correlated with modification of particular membrane components, although a parasite-derived protein has been associated with the presence of knob protrusions in P. falciparum-infected cells. The Coomassie blue-stained proteins of membranes from infected red cells generally show degradation of the spectrin duplex and band 3 (major erythrocyte membrane proteins) and intensification of lower rela- tive molecular mass bands, plus the appearance of new proteins, some of which have been shown to be syn- thesized by the parasite. In most studies, however, the entire infected cell or membranes prepared by cell lysis have been analysed causing some doubt as to the exact cause of changes in protein composition. Alterations of Coomassie blue-staining proteins from the eryth- rocyte membrane could result from insertion of para- site-derived proteins, or proteolytic cleavage by enzymes within the infected cell or in the host's serum. Radiolabelled probes have been used to identify surface protein antigens on infected erythrocytes. In order to identify the widest spectrum of new antigens in the membrane, it is recommended that different radiolabelling methods based on different reaction specificities be used- for example, lactoperoxidase or iodogen for radioiodination of tyrosine residues, pyri- doxal phosphate/NaB3 H4 for tritiation of lysine residues, galactose oxidase with or without neuramin- idase/NaB3 H4 for tritiation ofGa/GalNAc residues, diazotized sulfanilic acids which can label a variety of protein functional groups with 35S or 1251. It has already been shown that some antigens are labelled by one method and not by others. In order to determine whether radiolabelled anti- gens are in fact on the outer membrane of infected erythrocytes, it is necessary to evaluate critically the surface specificity of radiolabelling. Results obtained with several malaria species have indicated that the extent of internal labelling of infected cell proteins cannot be predicted from examination of the morpho- logy or stage of maturity of infected cells alone. Several methods are recommended for evaluation of labelling specificity, each of which has particular disadvantages and requires different assumptions to be made. These include measurement of haemoglobin labelling, demonstrating that labelled antigens react with antibody added to intact cells, and showing that addition of exogenous enzymes to intact labelled cells results in modification of the putative external antigen. It is also necessary to evaluate whether any infected cells are lysed during radiolabelling, as label- ling of internal components of lysed cells will compli- cate the interpretation of the subcellular location of identified antigens. Cell surface antigens can also be identified by absorption of monoclonal antibody or specific antibody from polyspecific sera with intact cells. Modifications in erythrocyte surface carbohy- drates, especially sialic acid, have been identified in P. knowlesi, P. berghei, P. yoelii, and P. lophurae infections. The membrane of infected erythrocytes has also been shown to lose cholesterol and unsatur- ated fatty acids, and to gain saturated fatty acids, leading to increased lipid fluidity in the host cell membrane. The relationships between alterations in erythrocyte lipids, proteins, and carbohydrates and survival of infected and uninfected erythrocytes in the circulation of infected animals is unknown. HYBRIDOMA TECHNOLOGY As in other branches of parasite immunology, the advent of hybridoma technology has added a new dimension to the analysis of plasmodial antigens and immune responses to them. The application of this technology to the immunology of malaria has been stimulated by evidence that antibodies have important antiparasite effects in human, monkey, bird, and rodent malarias. The main use of hybridoma-derived monoclonal antibodies in malaria research is in the identification and isolation of target antigens for host- protective or transmission-inhibiting immunity, as part of the search for an effective malaria vaccine. Other uses include the analysis of host cell recognition by merozoites and sporozoites, the identification of cell surface molecular changes resulting from infec- tion, the screening for protein antigen expression of cloned genes, and the development of immunodiag- nostic and parasite-typing reagents. Parasites applied so far to the production of mouse or rat-mouse hybridomas are P.falciparum, P. knowlesi, P. gal- linaceum, P. berghei, P. chabaudi, and P. yoelii, anti- bodies being selected using sporozoites, blood stages, and gametes in simple parasite-binding assays as the primary screen. Sporozoites of rodent, simian, and human malarias A monoclonal IgG, antibody directed against a surface antigen (designated Pb44) ofP. berghei sporo- zoites has been produced which, upon in vitro incub- ation, destroys the sporozoite infectivity. Monovalent Fab fragments of this antibody are equally effective, and passive transfer of small quantities (e.g., 10 j.g per recipient mouse) of the purified antibody confers complete protection against infection initiated by sporozoites. Using ferritin-tagged antibody, Pb44 has been found to be distributed evenly over the surface of salivary gland sporozoites. Evidence available so far indicates that Pb44 is a dominant, readily accessible antigen of the sporozoite. It is a stage-specific differ- 373 SCIENTIFIC WORKING GROUP entiation antigen, in that, on midgut sporozoites, it is present in smaller amounts, more sparsely distributed, or is absent. It is only detectable in the exoerythrocytic stage very early after infection of host liver cells. At least one other murine plasmodium parasite, P. yoelii nigeriensis, shares an antigenic determinant with the Pb44 sporozoite antigen, a cross-reaction which was not obtained using polyspecific mouse antisera. Whereas immunoprecipitation of solubilized surface- labelled sporozoites with the anti-sporozoite hybridoma antibody reveals the presence of one band (with relative molecular mass (RMM) 44 000) on SDS- PAGE, two other higher RMM bands, which are probably not located on the sporozoite surface, are immunoprecipitated using solubilized biosyntheti- cally-labelled sporozoites. Pb44 has a rather low isoelectric point and can be separated readily from other surface-labelled material using two-dimensional gel electrophoresis. Purified Pb44 has been obtained from sporozoite extracts by combining isoelectric focusing and affinity chromatography, using the monoclonal antibody. Recently, nine IgG, monoclonal antibodies have been produced against surface antigens ofP. knowlesi sporozoites. Gel profiles of immunoprecipitates of 35 S-methionine biosynthetically-labelled proteins consist of either one or two specifically radiolabelled bands. Two hybridoma-derived antibodies tested by in vitro preincubation have been found to abolish sporozoite infectivity in rhesus monkeys. A hybrid- oma-derived antibody that reacts with surface anti- gens of P.falciparum sporozoites of Gambian origin has been obtained and cross-reaction with a Thai isol- ate of P.falciparum noted. Merozoites ofrodent malaria Two monoclonal antibodies (of IgG, and IgG2a isotypes) have been produced which have apparent merozoite specificity and which, when injected into P. yoelii-infected mice, modify the course of infection. The two antibodies in combination result in a more rapid termination of infection than either alone. It has been suggested that the action of the hybridoma antibodies in infected mice is such as to give the host more time to develop a protective immune response against the virulent P. yoelii parasite, and thus it is predicted that infected nude mice will not be protected by the antibodies. Residual parasitaemia in recipients of hybridoma-derived anti- bodies appears to be confined to reticulocytes and a satisfactory explanation for this observation has not yet been found. Several other hybridoma-derived antibodies directed against other blood stage antigens were not protective on passive transfer. The contri- butions of titre differences as compared with speci- ficity differences in the host-protective activities of different hybridoma antibodies have yet to be defined in this system. All antibodies were selected using indirect immunofluorescence on acetone-fixed smears of schizont-enriched infected mouse blood. Recent work has demonstrated that the functional mono- clonal antibodies recognize different determinants on a single molecule of merozoites and this molecule has been purified by affinity chromatography. The puri- fied antigen has been used for immunization of mice with Freund's complete adjuvant, and anti-merozoite antibodies have been detected in the sera of these mice by indirect immunofluorescence. Information is not yet available on the species specificity of the antigen. Merozoites ofsimian malaria Eleven, cloned, hybridoma cell lines producing antibodies directed against surface antigenic deter- minants of P. knowlesi merozoites have been pro- duced. Three of the hybridoma-derived antibodies agglutinated merozoites, and the efficiency of invasion blocking using purified antibodies from two ascites fluids at high concentration (0.5 g/l) correlated with the strength of agglutination. The third mono- clonal antibody, with weak agglutinating activity, failed to block invasion of rhesus monkey red blood cells in vitro, as did the eight non-agglutinating anti- bodies. All three merozoite-agglutinating antibodies immunoprecipitated a biosynthetically-labelled pro- tein ofRMM 250 000. Information is not yet available as to whether this protein is recognized by sera from infected rhesus monkeys. In another study, spleen cells from rats immunized against merozoites of P. knowlesi have been found to yield more positive hybridomas than cells from immunized mice. Merozoites and schizonts of human malarias Mice immunized with schizonts and isolated mero- zoites of P.fakciparum (or P. berghei) have been used as donors of cells for fusion and hybridoma-derived antibodies reacting with P.falciparum merozoites and schizonts selected by an indirect immunofluorescence assay. In at least two laboratories, it has been shown that mice immunized against P. berghei produce antibodies that precipitate a large number of P.falci- parum blood stage antigens. When added to P.falci- parum cultures at a concentration of 100 ml/litre, 6 of 20 ascites fluids from mice bearing the various selected hybridomas inhibited the growth of the parasites by 70-99% over a 96-h culture period. IgG fractions (0.5 g/l) of two of the ascites fluids showed full inhibitory activity, while no inhibition was obtained with myeloma proteins of appropriate isotype at 1 g/l. Inhibition was assessed by a determination of percent- age parasitaemia and by a radioisotopic read-out using 3H-hypoxanthine incorporation. Three of the 374 ANTIGENIC STRUCTURE OF PLASMODIA ascitic fluids with inhibitory activity reacted with a P.fakciparum protein ofRMM 41 000 and one with a protein ofRMM 96 000. One of the inhibitory mono- clonal antibodies reacting with the protein of RMM 41 000 was obtained from fusion of cells from mice immunized with P. berghei schizonts and merozoites, and reacted with both merozoites and schizonts of P. berghei and P.fakciparum. Two ascites fluids, one of which immunoprecipitated a biosynthetically- labelled protein of RMM 41 000, the other a protein of RMM 96 000, inhibited the growth of four geo- graphical isolates of P.falciparum in culture. More- over, all four isolates contained both proteins. Neither the mode of action of the inhibitory hybridoma anti- bodies nor the precise location of the two target anti- gens has been determined, although there is indirect evidence that the antigens are located on the merozoite surface. Attempts at affinity purification of the anti- gens in sufficient quantities for vaccination studies in monkeys are in progress. Similar studies to those outlined in this section are in progress in many laboratories throughout the world. Gametes of avian malaria Mice immunized with gametes of the chicken para- site P. gallinaceum have produced two monoclonal antibodies (of IgM and IgG isotypes), which block transmission when mixed and fed with gametocyte- carrying blood to Aedes aegypti mosquitos. Aggluti- nation of male gametes appears to be the primary mechanism by which the mixture of these antibodies prevents fertilization. Both antibodies bind to male and female gametes and the IgG antibody recognizes a radioiodinated protein of RMM 225 000 on the surface of female gametes. Because of the difficulties of working with live female gametes, it has not been possible to determine whether they are targets for antibody-mediated transmission-blocking immunity. The degree of inhibition of transmission using the monoclonal antibodies is less than that achieved with immune sera from gamete-immunized birds. This observation raises the question of whether aggluti- nation and prevention of detachment of male gametes from the residual body during gametogenesis is insufficient for complete transmission-inhibiting immunity. Alternatively, binding avidity or titre of the hybridoma antibodies used may be much less than that of polyspecific sera. Antigenic analysis ofplasmodia A strategy being used in several laboratories to generate functional hybridomas is to select hybridoma antibodies that bind to biosynthetically-labelled P.falciparum antigens; these antigens are recognized differentially by human sera, which may or may not inhibit parasite growth in vitro. Obviously, there is an urgent need for clinically and functionally defined sera for use in these studies. Considerable progress would be achieved if various stages of plasmodia could be grown or maintained in the wells used to grow hybridoma cell lines. Even when a functional hybridoma-derived anti- body is available, it is not always easy to use that anti- body to prepare sufficient quantities of antigen for active immunization attempts. Difficulties arise because mouse hybridoma-derived antibodies of various isotypes often need to be tested in hetero- logous hosts for in vivo functional activity. The availability of modified myeloma cells of human origin for fusion, or Epstein-Barr virus-transformed human B lymphocytes, will facilitate the in vivo testing of hybridoma antibodies for functional activity in passive transfer experiments. The lack of monkey hybridoma-derived antibodies is impeding progress in the identification of functional hybridomas in monkey malarias. A theoretical advan- tage of using cells from mice injected with Plas- modium spp. that do not naturally infect them, is that novel antigens that are only very weakly immunogenic in natural hosts may be identified. The possibility that immunogenic monoclonal anti- idiotypic antibodies (directed against a functional hybridoma) may substitute for antigen in sensitization against plasmodium antigenic determinants in hosts needs to be tested in a careful, systematic manner. In addition, immunization of hosts with hybridoma- derived anti-host cell auto-antibodies which block merozoite (or sporozoite) invasion may induce a degree of immunity against the parasite structures used for host cell recognition. Many aspects of this type of work are highly speculative, idiotypic regu- lation being an active area of research in basic immunology. Obviously, these novel approaches should be used in parallel with the more conventional approaches of host-protective antigen identification, isolation, and production. One explanation for the slow development of resist- ance to malaria in endemic areas is that it takes time to generate an adequate spectrum of host-protective immune responses directed against numerous parasite molecules of varying immunogenicities. Concurrent immunosuppression is also thought to play a part. Such clinical observations have led to the concept that a multiplicity of immunological effector mechanisms and effector specificities are required to maintain a high degree of protection against highly adapted organisms such as Plasmodium spp. It is therefore rather surprising that a single hybridoma-derived anti- body, directed against a very small component of the total antigenic complex, can inhibit the growth of P.falciparum blood stages in vitro. However, it is not known whether all P.falciparum isolates are equally 375 SCIENTIFIC WORKING GROUP susceptible to inhibitory hybridoma antibodies in vitro, or if residual parasites in the inhibited cultures express the target antigen of the inhibitory antibody. There is an urgent need to determine whether malaria antigens recognized by such inhibitory monoclonal antibodies also induce protective immunity in the natural host. The possibility of using in vivo screening procedures in certain selected systems should also be examined. Predictably, false negatives will occur in the in vitro screening of functional hybridoma antibodies if the secondary consequences of antibody binding do not occur. Moreover, false positives will occur if the inhi- bition assay used is vulnerable to non-specific effects of ascites fluids, culture supernates, or even high concentrations of isotypes of purified mouse immu- noglobulins. The ideal screening procedure should employ as many genetic variants of the parasite popu- lation as possible and should use freshly isolated as well as cultured forms of plasmodium parasites. Moreover, the use of Fab and F(ab)+ fragments of hybridoma antibodies provides a means of discrimi- nating between combining site and immunoglobulin Fc interactions with parasite molecules and provides clues to the mode of antibody action (e.g., receptor inhibition versus agglutination). Currently, for immunodiagnostic purposes, the demonstration of circulating antigen is considered preferable to the demonstration of circulating anti- body. In this respect, a judiciously selected single specificity hybridoma-derived antibody may have great value. The strength of hybridoma technology in providing probes for molecular analysis is obvious. However, it is not the only approach to identification, isolation, and characterization of parasite molecules and anti- gens. Some important molecules may be very poor immunogens in mice and development of a satisfac- tory battery of mouse hybridoma-derived antibodies may not be possible. OTHER APPROACHES TO ANTIGENIC ANALYSIS The analysis of antigens at the molecular level is being carried out on sporozoites, merozoites, infected erythrocytes, purified schizonts, and gametes from a variety of malaria species, including P.falciparum,P. knowlesi, P. cynomolgi, P. berghei, P. yoelii, P. cha- baudi, P. gallinaceum, and P. lophurae. Particular attention has been focused on membrane-bound, potentially protective surface antigens and on charac- terization of appropriate antigens for immunodiag- nostic requirements. Antigenic determinants may involve proteins, glycoproteins, glycolipids, or other complex macromolecules. Theymaybe biosynthesized by the parasite or may represent host components modified by parasite factors (e.g., proteases). Free parasites may absorb host serum components, thereby modifying their intrinsic antigenicity. Major eukaryotic cell surface antigens are generally glycoproteins, with antigenicity residing in the poly- peptide, the carbohydrate, or both. Some malaria antigens are stage-specific, and may include key protective antigens. Similarly, some antigens will show species specificity, but degrees of cross-reaction may be encountered which do not necessarily correlate with protection data. Variant antigens have been reported to occur on the surface of P. knowlesi- infected erythrocytes. It is most important that malaria antigens used in protection experiments are pure and characterized unambiguously by several independent criteria, including relative molecular mass, charge, compo- sition, cellular origin, and functional and immuno- logical properties. It is also important to avoid spurious results arising, for example, from high levels of proteolytic activity which may appear when work- ing with parasites isolated at high parasitaemia, or from sialic acid modification. Isolation and identification of membrane-associ- ated antigens of malaria parasites pose specific problems; firstly, in erythrocytic stages, there are, in fact, three membranes-host cell, parasitophorous vacuole, and parasite plasma membrane. Secondly, cell surface labelling may give unreliable results because of leaking from schizont-stage infected host cells. The morphological appearance of schizont- infected erythrocytes and the degree of maturity of the schizont are not always reliable guides to the integrity of the erythrocyte membrane with respect to pene- tration of a probe. Also, rupture of a small proportion of infected cells during an experiment may alter the efficiency of surface radiolabelling. To identify the maximum possible number of surface components, a variety of radiolabelling probes, with different specificities, should be used. The relative importance of merozoite and infected cell-surface antigens in protection is unclear. It is therefore still important to distinguish accurately whether certain antigens are exposed on the infected cell surface. Some suggested criteria for establishing the location of an antigen on the red cell surface are: (a) demonstration of low labelling of intracellular components when surface-labelling probes are used, (b) demonstration that radiolabelled antigens are susceptible to degradation by extracellular enzymes, (c) demonstration that antibody recognizes antigen on living cells, (d) localization of antigen by electron microscopy techniques. The permeability of schizont-infected cells to Ig is 376 ANTIGENIC STRUCTURE OF PLASMODIA also relevant in this context as well as in the killing of intracellular parasites. The use of highly synchronized parasite populations could also facilitate interpret- ation of radiolabelling patterns. Sporozoites of P. berghei Classical immunochemical methods have given results complementary to those obtained using mono- clonal antibodies in identifying one protein (Pb44) as the major surface antigen of P. berghei sporozoites. This antigen has an acidic isoelectric point of about 3.9 under non-denaturing conditions, and can be highly purified by a combination of isoelectric focus- ing and affinity chromatography using monoclonal antibodies. Carbohydrate residues have not been detected on Pb44 using lectins or by treatment with neuraminidase. Pb44 appears to be a stage-specific antigen, covering homogeneously the surface of mature sporozoites and being absent from other developmental stages, including the majority of immature (oocyst) sporozoites. Metabolic labelling followed by immunoprecipitation also shows twin bands of about RMM 60 000 which share antigenic determinants with Pb44. Preliminary pulse-chase experiments suggest that these higher RMM bands, which are not surface-labelled, are cytoplasmic precursors of Pb44. Monoclonal antibodies against a surface antigen of P. y. nigeriensis sporozoites immunoprecipitated Pb44, indicating structural simi- larity between the sporozoite antigens of these two rodent malaria species. Gametes and gametocytes (P. gallinaceum) Surface labelling of microgametes is difficult because of their instability, but surface iodination of macrogametes in conjunction with precipitation by monoclonal antibodies has identified one surface anti- gen of RMM 250 000. Immunoprecipitation using polyvalent antisera suggests that while micro- and macrogametes possess certain common antigens, the macrogamete has others which are apparently unique. It would be interesting to determine whether surface changes occur on P.falciparum gametocytes during development. Merozoites Merozoites in amounts sufficient for surface labelling studies have so far been obtained only from P. knowlesi. The decreasing invasive ability of isolated merozoites indicates possible loss of a major mero- zoite component, which may prejudice its identifi- cation by labelling techniques. Intraerythrocytic stages Antigenic variation of P. knowlesi-infected red cells was first described 15 years ago, and cell surface antigens are implicated. The putative variant antigens have not yet been characterized. Alternative studies have led to the identification of parasite-synthesized antigens on the surface of P. knowlesi-infected rhesus erythrocytes. Sera from rhesus monkeys rendered immune by P. knowlesi infection/challenge were analysed by immunoelectro- phoresis and immune precipitation and identified two P. knowlesi-specific glycoproteins of RMM 65 000 and 90 000 (gp65 and gp9O) exposed on the surface of infected erythrocytes. A third component (RMM 125 000) can be metabolically labelled but not surface labelled. The three proteins appear common to the MalaysianH and Philippine strains ofP. knowlesi and are also precipated by sera of Gambian adults. In addition, the Gambian and rhesus immune sera pre- cipitated a component ofRMM 60 000-65 000 from surface-iodinated P.falciparum-infected squirrel monkey erythrocytes. The gp65 is therefore of par- ticular interest as analysis of 25 sera from rhesus monkeys revealed a positive correlation between immunity and presence of antibodies against gp65, which has an isoelectric point of about 4.6. In P. knowlesi and P. cynomolgi, tryptic peptide map- ping after iodination suggests about 70% homology in the amino acid sequences of the corresponding anti- gens. Both gp65 and gp9O appear by their antigenic properties to be present also on the surface of merozoites. Attempts are being made to correlate analyses of P. knowlesi antigens made in different laboratories using polyclonal antisera and crossed immunoelectro- phoresis. The most extensively investigated cell surface anti- gen ofP.falciparum-infected cells is the knob protein. Studies of K+ and K- lines point to a component of RMM 80 000. The knob protein cannot be surface labelled, but can be partly removed by incubation of cells with pronase. Metabolic labelling suggests that the knob protein is preferentially labelled with histi- dine and proline, and the evidence suggests that the chemical similarities with histidine-rich protein (HRP) of P. lophurae. The functional role of knob protein is uncertain. Antibodies to knobs have been detected in sera from P.falciparum immune monkeys and human subjects from endemic areas. A stage-specific glyco- protein of RMM 180 000 has been labelled on the surface of P.falciparum-infected erythrocytes. HRP of P. lophurae Under defined conditions, HRP of P. lophurae, which is located in the parasite granules, has been shown to be a protective antigen in ducks. Its function is still unclear, and HRP immunization of mice does not affect the course of P. berghei infection. 377 SCIENTIFIC WORKING GROUP Exoerythrocytic stages Early exoerythrocytic stages of P. berghei contain sporozoite-specific antigens detected by immuno- fluorescence using monoclonal antibodies to Pb44. These antigens are not detectable after 30 hours of development. Late exoerythrocytic stages react with antibodies to the blood forms of the parasite. Antigen production forprotection studies The development of vaccines using cellular antigens is hampered by the very low efficiency of the processes for obtaining purified merozoites, with yields often as low as 50/ of the theoretical level, and residual con- tamination with host and parasite debris. The early studies with P. knowlesi merozoites, P.falciparum and P. berghei sporozoites, and P. gailinaceum gametes have given clear indications as to the location and specificity of protective antigens. The relative scarcity of stage-specific natural plasmodial antigens seems likely to influence the effective strategies for obtaining protective antigens for initial testing and ultimately for large-scale trials. With the availability of monoclonal antibodies and powerful purification techniques, it will now be possible to obtain good yields of highly purified molecular antigens which are adequate for experi- mental purposes. The protective capacity of such purified antigens in animal models is the primary focus of several groups working particularly with sporozoite and asexual blood stages. However, it is hardly possible to secure adequate amounts of any natural sporozoite antigen for immunization trials and this work will probably move directly towards the application of recombinant DNA technology. Gametocytes may be obtainable in adequate amounts to enable purification of specific antigens for preliminary testing in an avian system only. Asexual blood stages may yield adequate amounts of purified antigens for initial protection tests with rodent, simian, and human malaria parasites. Recombinant DNA technology offers the most promising possibility for providing adequate supplies of pure antigens and for their characterization. The cloning of genes from animal malarias may provide useful probes to isolate genes from the human para- sites. Gene cloning will provide a valid approach only if, as seems probable, the relevant antigenic deter- minants are polypeptide in nature. However, there could be major problems if determinants of complex conformation are involved. Purification of antigens, whether from the parasites themselves or from genetically engineered sources, should not involve insurmountable problems. As several techniques converge to identify a limited range of proven or likely candidate protective anti- gens, research in the immediate future is likely to be dominated by the application of monoclonal anti- bodies and gene cloning. RECOMMENDATIONS Parasite production andpurification - Studies of the in vitro cultivation of malaria parasites should be encouraged, particularly studies of species of interest in immunological investigations, biochemical characterization of antigens, and the development of recombinant DNA technology. - For studies relating to biological, biochemical, immunological, and drug susceptibility aspects of malaria, clones of parasites should be used in definitive investigations when possible. - Improvement of techniques for the production and purification of merozoites for antigen isolation and characterization is required. - In order to obtain larger quantities of P.falci- parum gametes, and subsequently sporozoites, the conditions that maximize the rates of induction and maturation of sexual stages must be elucidated. - In view of the encouraging results of cultivation of rodent EEFs as far as the merozoite stage, further studies are needed, related to primate pathogenic species not yet cultured. - In the United States of America, the American Type Culture Collection has already shown interest in setting up aP.falciparum strain bank, and is currently preserving three strains. It is recommended that simi- lar centres be established in Europe and other areas and that the Natural Environment Research Council, in the United Kingdom, which already has a culture collection of algae and protozoa, be approached to determine their interest in participating. Such banks should contain strains previously characterized according to drug susceptibilities and other para- meters of general interest, such as gametocyte production. - In view of the danger of infection from laboratory material, all investigators should use prudence when handling infectious materials, especially in disposing of contaminated plastic and glassware. Furthermore, it should be remembered that handling of blood and blood products can result in bacterial or viral infections. Erythrocyte membrane changes - Technical advances are required in the puri- fication of the outer membrane of infected cells from parasitophorous vacuoles and all other internal com- ponents for unambiguous identification of: 378 ANTIGENIC STRUCTURE OF PLASMODIA 379 (a) alterations in erythrocyte membrane com- ponents; and (b) identification of external membrane surface antigens. Attention should be given to the possibility that parasite-derived or host proteases may degrade important parasite antigens during lysis and sub- cellular fractionation of infected cells. Inhibitors of protease enzymes of different reaction specificity are available and should be used if necessary. Critical evaluation is required of the extent of surface-specific versus internal labelling when protein radiolabelling probes are used. Systematic studies should be encouraged to investigate the permeability properties of infected cells to small and large molecules. - Use of a variety of radiolabelling methods based on different reaction specificities is recommended in order to radiolabel the maximum number of antigens. Studies on changes in erythrocyte surface carbo- hydrates, and proteins during malaria should be extended, as changes in these outer membrane com- ponents may determine the lifetime of circulating erythrocytes. Hybridoma technology Additional functional tests should be developed for the screening of monoclonal antibodies against various plasmodial developmental stages and used in conjunction with immunofluorescence as a primary screen. It is considered essential to develop a battery of monoclonal antibodies against any one given parasite stage in order to increase the probability of detecting minor or weakly immunogenic determinants which may have important functional roles. - It will be of particular value to develop mono- clonal antibodies directed against accessible mem- brane components of the various parasite stages and against membrane components characteristic of infected host cells. - Functional tests should be carried out using not only purified intact antibodies, but also Fab frag- ments. These studies will provide insight into the mode of action of hybridoma-derived antibodies and the roles of the respective target antigens. - The functional assay for monoclonal antibodies should not be limited to the testing of single-specificity monoclonal antibodies, but extended to include com- binations of such antibodies. - It is appreciated that studies on monoclonal anti- bodies directed against antigens of P. vivax, P. mal- ariae, and P. ovale are lagging behind those applied to P.falciparum, and that screens should be developed to detect shared determinants amongst these four Plasmodium spp. Such monoclonal antibodies of known specificity would be of primary importance in setting up immunodiagnostic assays. - Efforts should be made to develop centres for testing the protective effects of selected hybridoma- derived antibodies and target antigens in vivo in non- human primate systems. Such in vivo assays are essen- tial to assess the validity of in vitro inhibitory tests. Analysis and production of antigens - Further efforts should be made to correlate antigens presently defined by different criteria, and to develop a unified nomenclature. - Antigens are unlikely to be available in large amounts for general distribution in the near future, but exchange of material on a personal basis between laboratories should be encouraged, especially to resolve differences in results or to compare products from different sources. RdSUMit STRUCTURE ANTIGENIQUE ET ASPECTS CONNEXES DE LA BIOLOGIE DES PLASMODIUMS: SITUATION ACTUELLE La quatrieme reunion du Groupe de travail scientifique sur l'immunologie du paludisme (SWG/IMMAL), qui s'est tenue sous les auspices du Programme speial PNUD/ Banque mondiale/OMS de recherche et de formation concernant les maladies tropicales, a e consacree a la structure antigenique et aux proprietes biologiques connexes des plasmodiums, y compris la production de materiel para- sitaire. I1 n'y a pas eu recemment de progres importants dans la culture in vitro des stades erythrocytaires asexues de P. falci- parum. On est parvenu A obtenir A partir de cultures des gametes de P. falciparum infectants pour les moustiques, mais les conditions optimales de production des gametocytes n'ont pas encore e etablies. Les formes exoerythrocytaires de P. berghei et P. yoelii ont ete cultivees en monocouches de cellules cerebrales d'embryons de rat.c I1 existe des methodes permettant la synchronisation des cultures de P. falciparum et la separation des erythrocytes infectes de ceux qui ne le sont pas. Les parasites provenant de cellules de l'h8te lysees peuvent etre obtenus sous forme relativement pure et les meilleures methodes d'isolement des merozoites sont des c Depuis que ce rapport a et redige, on a obtenu une schizogonie tissulaire complete et la production ulterieure in vitro de mtrozoltes tissulaires infectants de P. berghei dans des cellules pulmonaires W138 (Hollingdale, USAID meeting on Malaria Immunology and Vaccine Development, Washington, DC, 14-16 janvier 1981). 380 SCIENTIFIC WORKING GROUP methodes de liberation naturelle; cependant, le rendement et la viabilite doivent etre ameliores. Des anticorps monoclonaux ont e produits A l'egard d'antigenes de plusieurs especes de parasites et stades de developpement. Un anticorps monoclonal dirige contre une proteine superficielle des sporozoites de P. berghei (Pb44) a manifeste une activite neutralisante in vitro et in vivo. Grace A cet anticorps, on a montre que Pb44 etait un antigene de differenciation specifique de stade, et il a ete recemment purifie. Des anticorps monoclonaux A l'egard d'antigenes de surface des sporozoites de P. knowlesi ont egalement protege contre l'infection. Des anticorps provenant d'hybri- domes ont et prepares A 1'egard d'antigenes de mtrozoYtes de plusieurs especes; dans le modele P. yoelii deux de ces anticorps agissant ensemble venaient A bout de l'infection plus rapidement que chacun d'eux separement. Cependant, la question de l'importance relative du titre et de la speci- ficite dans la protection de l'h6te necessite de nouvelles etudes. Un antigene de merozoite unique reagissant avec plusieurs anticorps monoclonaux fonctionnels a ete purifie et il est utilis6 dans des experiences de vaccination chez la souris. On a mis en evidence une correlation entre l'activite agglutinante des anticorps monoclonaux A l'6gard des mero- zoites de P. knowlesi et leur capacite de bloquer l'invasion des erythrocytes. Des anticorps provenant de souris immunisees contre P. berghei precipitaient un certain nombre d'antigenes de P.falciparum et certains inhibaient la croissance de ce parasite in vitro. On a mis en evidence une immunite bloquant la transmission due A des anticorps monoclonaux A l'egard de gametes de P. gallinaceum, le meanisme primordial etant une agglutination des gametes males. On procede A l'analyse des antigenes au niveau molecu- laire par des methodes immunochimiques classiques sur des parasites appartenant A plusieurs especes et plusieurs stades de developpement, en mettant l'accent sur les antigenes lies A la membrane susceptibles de jouer un r6le important dans la protection ainsi que sur les antigenes pouvant servir A l'immunodiagnostic. Les principaux antigenes de la surface cellulaire sont habituellement des glycoproteines et ils comprennent des antigenes sp6cifiques de stade et sptci- fiques d'espece. Des antigbnes variants ont e signales A la surface d'erythrocytes infect6s de P. knowlesi, mais ils n'ont pas encore e caracterises. Des etudes portant sur des erythrocytes infectes de ce dernier parasite ont permis d'identifier deux glycoproteines (gp 65 et gp 90) qui sont synthetisees par le parasite et apparaissent sur la surface erythrocytaire: l'immunite contre l'infection s'est r6v6lke en correlation avec la presence d'anticorps A l'6gard de gp 65, laquelle est probablement pr6sente egalement sur les mero- zoites. La proteine des protuberances qui apparaissent sur les erythrocytes infectes de P.falciparum ainsi que la proteine riche en histidine de P. lophurae ont egalement e etudies en detail et des similitudes chimiques entre elles ont W decouvertes. Les secteurs prioritaires du programme ont et identifies et des recommandations formulees au sujet d'aspects parti- culiers de la production et de l'analyse des antigenes necessi- tant la poursuite des etudes. LIST OF PARTICIPANTS R. L. Beaudoin, Head, Malaria Branch, Division of Immunoparasitology, Naval Medical Research Institute, Bethesda, MD 20014, USA S. Cohen, Professor of Chemical Pathology, Department of Chemical Pathology, Guy's Hospital Medical School, London, England W. E. Collins, Bureau of Tropical Diseases, Department of Health and Human Services, Centers for Disease Control, Atlanta, GA 30333, USA G. A. M. Cross, Department of Immunochemistry, Wellcome Research Laboratories, Beckenham, Kent, England R. Howard, Malaria Section, Laboratory of Parasitic Diseases, National Institute of Allergy and Infec- tious Diseases, National Institutes of Health, Bethesda, MD 20012, USA Liu Er Hsiang, Associate Professor, Department of Immunology, Institute of Basic Medical Sciences, Beijing, China T. Ch. Ifediba, Division of Parasitology, Department of Microbiology, New York University Medical Center, New York, NY 10016, USA J. B. Jensen, Michigan State University, East Lansing, MI 48824, USA A. Kilejian, Laboratory of Parasitology, The Rockefeller University, New York, NY 10021, USA F. Kourilsky, Centre d'Immunologie de Marseille- Luminy, 13288 Marseille, France I. McGregor, 210 Hyde End Road, Spencers Wood, Reading, Berkshire, England L. Miller, Head, Malaria Section, Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20014, USA G. Mitchell, Experimental Pathology Unit, Walter and Eliza Hall Institute of Medical Research, Royal Melbourne Hospital, Melbourne, Victoria 3050, Australia R. S. Nussenzweig, Head, Division of Parasitology, Department of Preventive Medicine, New York University Medical Center, New York, NY 10016, USA V. Nussenzweig, Department of Pathology, New York University Medical Center, New York, NY 10016, USA L. Perrin, Division of Haematology, Cantonal Hospital, University of Geneva, 1211 Geneva 4, Switzerland ANTIGENIC STRUCTURE OF PLASMODIA 381 T. Ponnudurai, Department of Medical Parasitology, Faculty of Medicine, University of Nijmegen, Nijmegen, The Netherlands P. K. Russell, Director, Walter Reed Army Institute of Research, Walter Reed Army Medical Center, Washington, DC 20012, USA R. Schmidt-Ullrich, Assistant Professor, Radio- biology Division, Therapeutic Radiology Depart- ment, Tuft's New England Medical Center, Boston, MA 02111, USA I. W. Sherman, Department of Biology, University of California, Riverside, CA 92521, USA C. R. Sterling, Department of Immunology and Microbiology, Wayne State Medical School, Wayne State University, Detroit, MI, USA S. Tharavanij, Faculty of Tropical Medicine, Mahidol University, Bangkok 4, Thailand W. Trager, Laboratory of Parasitology, Rockefeller University, New York, NY 10021, USA WHO Secretariat V. Houba, Immunology, Division of Noncommuni- cable Diseases, WHO, Geneva, Switzerland L. Martinez, Consultant, Secretary, Steering Committee of the Scientific Working Group on the Immunology of Malaria, UNDP/World Bank/ WHO Special Programme for Research and Train- ing in Tropical Diseases, WHO, Geneva, Switzer- land. G. Torrigiani, Chief, Immunology, Division of Non- communicable Diseases, WHO, Geneva, Switzer- land P. I. Trigg, Secretary, Steering Committee of the Scientific Working Group on the Chemotherapy of Malaria, UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases, WHO, Geneva, Switzerland W. H. Wernsdorfer, Chief, Research and Technical Intelligence, Malaria Action Programme, and Secretary, Scientific Working Group on the Immunology of Malaria, UNDP/World Bank/ WHO Special Programme for Research and Train- ing in Tropical Diseases, WHO, Geneva, Switzer- land Observers R. Dayal, Department of Medicine, University of Geneva, Geneva, Switzerland J. Erickson, USAID, Washington, USA