Use of non-human primate hepatocytes for in vitro study of the pre-erythrocytic stages of malaria parasites.

Use of non-human primate hepatocytes for in vitro study of the pre-erythrocytic stages of malaria parasites P. Millet, W.E. Collins,' M. Aikawa,2 A.H. Cochrane,3 & Phuc Nguyen-Dinh' Methods were developed that allow invasion of sporozoites from simian malaria parasite species (Plasmo- dium cynomolgi, P. knowlesi, P. coatneyi, P. inui, P. gonderi, P. fragile) and development to schizont stages in rhesus and Saimiri monkey hepatocytes. The P. cynomolgi-rhesus monkey model was used to study inhibition of schizont development using monoclonal antibodies (MAbs) produced against the circum- sporozoite (CS) protein of various strains and species of malaria parasites. Immunoelectron microscopy, using gold-labelled MAbs and cultured parasites, demonstrated that the CS protein persists in 7-day old liver stages of P. cynomolgi, but is not expressed at the surface of infected hepatocytes. A rhesus monkey was immunized with autologous hepatocytes (collected by biopsy) infected in vitro with liver stages of P. cynomolgi. This immunization elicited antibodies reacting with sporozoite, liver stage, and blood-stage parasites. In addition, human malaria parasites (P. falciparum, P. vivax, P. malariae) have been cultured in Saimiri or rhesus monkey hepatocytes. The P. vivax-Saimiri monkey model was used to study inhibition activity of sera from Saimiri monkeys experimentally immunized with recombinant P. vivax CS proteins. Post-immunization sera inhibited the parasite development, thus demonstrating the induction of antibodies effective against sporozoites. No relationship, however, was detected between in vitro inhibition and in vivo protection or antibody titres determined by ELISA or IFA. Introduction In the life-cycle of malaria parasites, exoerythrocytic (EE) stages constitute an essential link between the sporozoites inoculated by a mosquito bite and the blood stages responsible for symptoms of the disease. Investigations of EE stages are hampered by their short duration, and by the fact that in mammalian malaria these stages occur in a deep organ, the liver. The study of EE parasites has been markedly enhanced by the development of techniques permit- ting their culture in vitro. Culture systems have been developed for rodent parasites (1) and for three spe- cies of human malaria parasites (2-5). However, ro- dent malaria species are not closely related to human malaria parasites, and host cells for use in in vitro studies on human malaria parasites are either difficult to obtain (human hepatocytes), or do not reflect the natural host-parasite relationship (hepatoma cells). 1 Malaria Branch, Division of Parasitic Diseases, Center for Infec- tious Diseases, Centers for Disease Control, Atlanta, GA 30333, USA. Correspondence should be sent to Dr P. Millet at this address. 2 Institute of Pathology, Case Western Reserve University, Cleve- land, OH, USA. 3 Department of Medical and Molecular Parasitology, New York University, New York, NY, USA. As an alternative, simian malaria parasites pro- vide several advantages. Rhesus monkeys (Old World monkeys) are readily available experimental animals, thus ensuring regular access to liver specimens and permitting comparisons between in vivo and in vitro results. Among the parasites infecting rhesus monkeys are some well-established models for human malarias. Plasmodium cynomolgi is a model for P. vivax; P. inui for P. malariae; P. coatneyi and P. fragile for P. falciparum. Additionally, the quotidian parasite P. knowlesi is frequently used for immunological studies on malaria. Among human malaria parasites, P. vivax and P.falciparum infect Saimiri and Aotus monkeys (New World monkeys), and vaccine trials have been con- ducted using these animals. In vitro cultured EE stages of human malaria species can be used to evalu- ate sporozoite vaccine efficacy or to isolate liver stage antigens. In vitro development of simian malaria parasites in primary cultures of simian hepatocytes will be presented here, completed by some research applica- tion of these cultures. In addition, development of human malaria species in simian hepatocytes will be presented, together with their application to in vitro evaluation of a P. vivax sporozoite immunization trial in Saimiri monkeys. 60 Bulletin of the World Health Organization, 66 (Suppl.): 60-65 (1990) Malaria parasites In simian hepatocytes In vitro Methods and results Primary cultures of hepatocytes were derived from liver biopsies of various simian species. A liver frag- ment (approximately 2 x 2cm) was collected by laparotomy and perfused with a collagenase-Hepes buffer solution to dissociate the hepatocytes which were cultivated as previously described (6). One drop (25p1) of the cell suspension, containing 30000 to 40000 hepatocytes, was deposited on each of 2 sites on the bottom of a 35-mm plastic tissue culture dish. Sedimentation, adherence, and resulting growth of the cells formed two disc-shaped (5-mm diameter) mono- layers. Monolayers were infected with sporozoites obtained by dissection and grinding of salivary glands from infected mosquitos. As soon as the cells were spread out, the drop of medium covering the mono- layer discs was removed and replaced with 25 p1 of medium containing sporozoites. After 2 hours, the sporozoite suspension was removed and 1 ml of cul- ture medium was added per culture dish, and changed daily. Cultures were incubated at 37°C, in an atmo- sphere of5% CO2 and 95% air. At appropriate times, the monolayers were fixed with methanol then stained with Giemsa for observation of EE stage parasites. Plasmodium cynomolgi (Fig. 1-3), P. knowlesi, P. coatneyi, and P. inui EE stages developed in primary culture of rhesus monkey hepatocytes (Table 1). From 1 to 80 schizonts were obtained per 10000 sporozo- ites, depending on the experiments. In one experiment, a few P.fragile and P. gonderi EE schizonts were also obtained. P. cynomolgi and P. coatneyi EE schizonts were also cultured in Saimiri monkey (Saimiri sciureus boliviensis) hepatocytes. These findings are surprising since all attempts to infect these monkeys with sporo- zoites from these two species have been unsuccessful. Our in vitro findings suggest that the EE stages are able to develop in the liver, but that invasion into or development within erythrocytes may not be possible. In terms of timing of development and morpho- logy, the in vitro findings in all species correlated well with the in vivo results obtained in rhesus monkeys (7-10). Nuclear division and cytomerization were fol- lowed for P. knowlesi and P. cynomolgi schizonts. The in vitro timing of release of merozoites infective to rhesus monkey erythrocytes was in concordance with the P. cynomolgi and P. knowlesi prepatent periods occurring in rhesus monkeys. Use of an inverted phase-contrast microscope allowed observation of the evolution of P. knowlesi schizonts and of the burst of a mature parasite. Based on previous studies (11), the P. cynomolgi-rhesus monkey model was used to study the inhibitory effect of anti-sporozoite monoclonal antibodies on the in vitro development of liver stages of P. cynomolgi (NIH strain of P. cynomolgi bastianel- Fig. 1-3. Exoerythrocytic stages of Plasmodlum cyno- molgi In primary culture of rhesus monkey hepatocytes. 1. 6-day-old parasite. 2. Immature 8-day-old parasite with still dividing nuclei. 3. Mature 8-day old parasite with rounded merozoites. lii) (12). MAbs against the CS proteins of five strains of P. cynomolgi (NIH, London, Gombak, Ceylon, Berok), and of P. knowlesi (H strain) were used. Incuba- tion of P. cynomolgi (NIH strain) sporozoites with the anti-NIH strain MAbs totally prevented liver stage WHO Bulletin OMS: Supplement Vol. 68 1990 B1 P. Millet et al. development; MAbs produced against the other four strains had no apparent activity. In addition, the anti- P. knowlesi MAbs had a partially inhibitory effect (80% inhibition) on parasite development (Fig. 4). These results confirm, in an in vitro functional assay, the CS protein strain specificity for simian malaria species, and the cross-reactivity between P. knowlesi and P. cynomolgi bastianellii previously demonstrated by other immunological methods (13-15). Complete inhibition of sporozoite invasion could be obtained only by using very high concentrations of antibodies. Antibodies against the CS protein had no effect when Table 1: Exoerythrocytic stages of simian and human malaria parastes established In primary cultures of simian hepatocytes New world monkeys Old world monkeys Apes Malaria species M. mulatta S. sciureus Al. griseimembra P. troglodytes Simian: P. cynomolgi 0 NT NT P. knowlesi NT NT NT P. inui 0 NT NT NT P. coatneyi 0 0 NT NT P. gonderi NT NT NT P. fragile * NT NT NT Human: P. falciparum 0 0 0 NT P. vivax 0 0 0 NT P. ma/ariase NT NT 0 0 a. EE schizont development; 0 unsuccessful experiments; NT, not tried. Fig. 4. Percentage Inhibition of P. cynomolgl bastlanelill EE stage development by monoclonal antibodies directed against the CS protein of various strains of P. cynomolgl, P. knowlesI, P. vlvax, and P. takipprum. WHO Bulletin OMS: Supplement Vol. 68 1990 % inhibition 0% 50% 100% species Mobs P. cyno. bostianellii 13E1 * Is 12D6 0 Berok 2F2C4 S so 6G11 Gombak lGl London 7A9 Ceylon 6F2 P. knowlesi H 2G3 * le 51-18 77 7- P. vivax 2F2A7 S P. falciparum 2A10 0 62 Malaria parasites In simian hepatocytes In vftro added during EE stage development. Thus it appears that antibodies against CS protein can play a pro- tective role only by preventing invasion, and not by further inhibition of the EE stage development once a successful invasion has occurred. Using the same P. cynomolgi-rhesus monkey model, immunoelectron microscopy localized CS antigen in 5- and 8-day old in vitro cultures of P. cynomolgi liver stages (16). A MAb (13E1) directed against the repeat region of P. cynomolgi CS protein densely labelled, in the presence of immunogold, the plasma membrane and surface of 5-day-old liver stage parasites as well as the surrounding parasitophorous vacuole membrane and space. Density of labelling decreased as liver stages increased in size and matur- ity. The CS antigen was not associated with internal structures within developing schizonts, or with the host cell cytoplasm or its surface. This indicates that components of the CS protein are not processed and expressed on the surface of infected hepatocytes, thus avoiding recognition by antibody or lymphocyte. In order to determine the immunogenicity of EE stage antigens, a rhesus monkey was immunized with in vitro EE stages of P. cynomolgi cultured in auto- logous hepatocytes. Three intramuscular injections containing a total of 75000 EE schizonts with mur- amyl dipeptide as adjuvant were given. A control animal was immunized with autologous cultured he- patocytes following the same procedure. Results of this study are shown in Fig. 5. Antibodies against the sporozoites, the EE stages, and the blood stages were obtained, but the monkey was not protected against sporozoite challenge. The control animal did not show any antibodies against sporozoites; antibodies against the blood stages were obtained only when parasitaemia was high. Immunoelectron microscopy using 7-day-old liver stages and the post-immun- ization serum previously adsorbed on uninfected hepatocytes showed antigenic distribution around the parasitophorous vacuole and inside the schizont, but not in the cytoplasm of the host cell. A liver biopsy was performed 7 days after sporozoite challenge, and liver sections were obtained. Many EE stage parasites were detected, none being surrounded by macro- phages or other cells involved in the cellular response. However, a boost of antibodies reacting with the three stages was obtained after challenge. Western blot analysis performed with sporozoite and blood-stage extracts, and post-immunization sera showed that antibodies were produced against the CS protein, and five blood-stage proteins (data not shown). Such cross-reacting antigens have been identified for Fig. 5. Antibody titres against sporozoltes, liver stages, and blood stages of P. cynomolgl following immunization of a rhesus monkey with liver stages of P. cynomolgl cultured in autologous hepatocytes and then challenged with sporozoltes. Immunofluorescence was done using air-dried parasites for all stages. IFA titer EE stage immunization sporozoite blood sporozoites 1 st 2nd 3rd challenge parasites v V V V V liver stages1/1024- dayo day20 day34 day46 day5 I ~ ~~blood stagqs WHO Bulletin OMS: Supplement Vol. 68 1990 0 14 28 42 55 70 84 days after immunization 63 P. Millet et al. human malaria species by the use of P. vivax and P. falciparum anti-blood-stage MAbs (17, 18). Our experiment shows that some of these proteins are in sufficient quantity in the liver stage to elicit an anti- body response. Another application of in vitro cultures in liver cells from non-human primates consisted in cultures of EE stages of human malaria parasites. Plasmodium malariae schizonts were obtained in chimpanzee hep- atocytes (19). A few P. malariae schizonts were also obtained in Aotus monkey hepatocytes (19), while three attempts to grow P. falciparum failed. Sur- prisingly, hepatocytes from rhesus monkeys sup- ported the growth of some P. falciparum schizonts (4 schizonts for 100000 sporozoites). Liver stages of P. vivax developed in primary cultures of hepatocytes from Saimiri monkeys (Saimiri sciureus boliviensis) (20), but not in Aotus monkey (Aotus lemurinus griseimembra) hepatocytes in spite of the fact that both of these New World monkey species are susceptible to in vivo infection by P. vivax sporozoites. The P. vivax-Saimiri monkey model was used to correlate in vitro inhibition of liver stage parasites with immune response and immunoprotection in 18 Saimiri monkeys used in P. vivax vaccine trials using CS recombinant proteins (21). One recombinant (NS181V20) contained the repeat region of the CS protein. The other recombinant (VIVAX-1) contained the entire repeat domain and part of the surrounding regions. Both antigens were administered with alumi- nium hydroxide. In another group, six animals were immunized with irradiated sporozoites alone. Post- immunization sera from all immunized monkeys and from monkeys immunized with irradiated sporozoites inhibited parasite development when mixed with sporozoites, thus demonstrating the induction of anti- bodies effective against sporozoites, even by the non-protected animals (Table 2). However, 100% in- hibition was rarely obtained, and no relationship was detected between in vitro inhibition and antibody titres detected by IFA or ELISA. Conclusions These studies have established that most of the simian and human malaria parasite species can develop in primary culture of simian hepatocytes. The in vitro cultured EE stages were used: (a) as functional assays to study the inhibitory activity of MAbs and sera against P. cynomolgi and P. vivax sporozoites; (b) for EE stage antigen production and characterization; and (c) for antigen localization (CS protein in the EE schizonts). Most of the studies conducted here focused on the CS protein of P. cynomolgi and of P. vivax. As in studies conducted by other authors (22), our results show a strong inhibition of liver stage development by antibodies directed against the CS protein of the parasites, but two points are not very encouraging for use of the CS protein alone as a vaccine: (i) a high level of antibodies did not often entirely inhibit the schizont development; (ii) immunoelectron micro- scopy studies did not show any expression of the CS protein at the surface of the infected hepatocytes. However, other malarial antigens may be expressed at the surface of the hepatocytes which may induce a humoral or cellular immune response against the parasite. In vitro models provide a tool for searching such proteins, and as direct functional assays for studying the humoral or cellular immune response to these antibodies. Acknowledgements This study was made possible by the collaborative effort of the following: Dr J.R. Broderson, Dr B.G. Brown, Dr L.M. Hermann, E.L. Jackson, V. George, D.S. Rice, and Table 2: Summary of serological and parasitological results of vaccine trials directed against the CS protein of P. vivax using Salmirl monkeys, compared to the percentage Inhibition of P. vivaXEE stage development In vitro by sera from the Immunized monkeys % inhibition of Serologic titres sporozoites in vitroc No. of No. Antigen Adjuvant animals WFA ELISAb protected Range Mean Irradiated species 6 28960 924 2 (33)d 84-100 96 VIVAX-1 Alum 6 5120 1055 1 (17) 87-100 95 NS81 V20 Alum 6 5120 2992 0 85-100 97 None Alum 6 <20 <25 1(7) 0-45 25 a Twofold dilutions of sera were tested in the IFA starting at 1:20 using glutaraldehyde-fixed sporozoites; IFA titres shown are for the day of challenge. bTitres were expressed in OD units (the dilution of sera giving an optical density of 1.0). c Percentage of inhibition compared to the pre-immunization sera activity for each monkey. dFigures in parentheses are percentages. WHO Bulletin OMS: Supplement Vol. 68 199064 Malaria parasites In simian hepatocytes In vitro H.R. Wetmore, Office of Scientific Services; C.E. Monken, Hepatitis Branch, Division of Viral Diseases; Dr G.H. Campbell, Dr T.K. Ruebush II, P.M. Procell, V.K. Filipski, J.C. Skinner, C. Wilson, and J.H. Hartmann, Malaria Branch, Division of Parasitic Diseases, Centers for Dis- ease Control, Atlanta, GA; Dr K.K. Kamboj, Department of Medical and Molecular Parasitology, New York University, New York, NY; and Dr C.T. Atkinson, Institute of Pathology, Case Western Reserve University, Cleveland, Ohio. Plas- modium falciparum sporozoites were kindly provided by Dr M.R. Hollingdale, Biomedical Research Institute, Rock- ville, MD. This study was supported in part by USAID PASA No STB-0453-23-P-HZ-00165-03 and was done while P. Millet held a National Research Council/Centers for Disease Control Research Associateship. References 1. Holllngdale, M.R. et al. In vitro cultivation of the exoerythrocytic stage of Plasmodium berghei from sporozoites. Science, 213: 1021-1022 (1981). 2. Mazler, D. et al. Cultivation of the liver forms of Plasmodium vivax in human hepatocytes. Nature, 307: 367-369 (1984). 3. Mazler, D. et al. Complete development of hepatic stages of Plasmodium falciparum in vitro. Science, 227: 440-442 (1985). 4. Mazher, D. ot al. Plasmodium ovale: in vitro develop- ment of hepatic stages. Exp. parasitol., 64: 393-400 (1987). 5. Holllngdale, M.R. et al. In vitro culture of two popula- tions (dividing and nondividing) of exoerythrocytic parasites of Plasmodium vivax. Am. j. trop. med. hyg., 34: 216-222 (1985). 6. MIllet, P. et al. In vitro cultivation of exoerythrocytic stages of the simian malaria parasites Plasmodium cynomolgi, P. knowlesi, P. coatneyi. and P. inui in rhesus monkey hepatocytes. Am. j. trop. med. hyg., 39: 533-53 (1988). 7. Garnham, P.C.C. et al. The tissue stages and sporo- gony of Plasmodium knowlesi. Trans. Roy. Soc. Trop. Med. Hyg., 51: 384-396 (1957). 8. Eyles, D.E. The exoerythrocytic cycle of Plasmodium cynomolgi bastianellii in the rhesus monkey. Am. j. trop. med. hyg., 9: 543-555 (1960). 9. Held, J.R. & Contacos, P.G. Studies of the exoerythro- cytic stages of simian malaria. II. Plasmodium coat- neyi. J. parasitol., 53: 910-918 (1967). 10. Hold, J.R. et al. Studies of the exoerythrocytic stages of simian malaria. Ill. Plasmodium inui. J. parasitol., 54: 249-254 (1968). 11. Holllngdale, M.R. et al. Inhibition of entry of Plasmo- dium falciparum and P. vivax sporozoites into cultured cells: an in vitro assay of protective antibodies. J. immunol., 132: 909-913 (1984). 12. Millet, P. et al. In vitro exoerythrocytic development of P. cynomolgi bastianellii inhibitory activity of mono- clonal antibodies against sporozoites of different P. cynomolgi strains and of P. knowlesi. Parasite immunol: 11: 223-230 (1989). 13. Cochrane, A.H. et al. Monoclonal antibodies identify the protective antigens of sporozoites of Plasmodium knowlesi. Proc. Natl Acad. Sci., USA, 79: 5651-5655 (1982). 14. Cochrane, A.H. et al. Antigenic diversity of the circum- sporozoite proteins in the Plasmodium cynomolgi complex. Mol. biochem. parasitol., 14:111-124 (1985). 15. Cochrane, A.H. et al. Further studies on the antigenic diversity of the circumsporozoite proteins of the Plas- modium cynomolgi complex. Am. j. trop. med. hyg., 35: 479-487 (1986). 16. AtkIneon, C.T. et al. Localization of circumsporozoite antigen in exoerythrocytic schizonts of Plasmodium cynomolgi. Am. j. trop. med. hyg., 40: 131-140 (1989). 17. Aley, S.B. et al. Plasmodium vivax: exoerythrocytic schizonts recognized by monoclonal antibodies against blood-stage schizonts. Exp. parasitol., 64: 188-194 (1987). 18. Szarfman, A. et al. Mature liver stages of cloned Plasmodium falciparum share epitopes with proteins from sporozoites and asexual blood stages. Parasite immunol., 10: 339-351 (1988). 19. Millet, P. et al. In vitro cultivation of exoerythrocytic stages of the human malaria parasite Plasmodium malariae. Am. j. trop. med. hyg., 38: 470-473 (1988). 20. Millet, P. et al. Plasmodium vivax: in vitro develop- ment of exoerythrocytic stages in squirrel monkey hepatocytes and inhibition by an anti-P. cynomolgi monoclonal antibody. Exp. parasitol: 69: 91-93 (1989). 21. Collins, W.E. et al. Immunization of Saimiri sciureus boliviensis with recombinant vaccines based on the circumsporozoite protein of Plasmodium vivax. Am j. trop. med. hyg: 40: 455-464 (1989). 22. Hoffman, S.L. et al. Naturally acquired antibodies to sporozoites do not prevent malaria: vaccine develop- ment implications. Science, 237: 639-642 (1987). WHO Bulletin OMS: Supplement Vol. 68 1990