Recent developments in vaccination against malaria: Immunization of rhesus monkeys with blood-stage antigens of Plasmodium knowlesi*

Bidief,n of Mhe W'orld Health Orgar,':fcron. 57 (SuppI 1) 139-i5I (1979) Immunization of rhesus monkeys with blood-stage antigens of Plasmodium knowlesi* K. H. RIECKMANN,I E. J. CABRERA,I G. H. CAMPBELL,3 R. C. JOST,4 R. MIRANDA,4 & T. R. O'LEARY4 Evaluation of the efficacy of 3 nonviable blood-stage antigens-schizont Ag, merozoite Ag, and "French press" Ag-ofPlasmodium knowlesi emulsified in Freund's complete adjuvant (FCA), was carried owut in a study involving 32 rhesus monkeys. After2 iinmunizations, admiinistered 6 weeks apart, monkeys were challenged with a variant of P. knowlesi differentfrom that usedfor iunmunization. A118 control monnkeys that received eitherPBS orFCA developed a severeparasitaemia and died ofthe infection within 12 days afterparasite chalienge. In the group that received thefreeze-thawedschizontAg, S ofthe 8 monkeys died at thesame time as the controls, I died a day later, and 2 survived the infection (maximum parasitaenmia: 0. 7%; 3.2%6o). In thefreeze-thawed merozoiiteAg group, I monkey died at fhesame time as the controls, S monkeys diedbetween 14 and 17 daysafter challenge, and 2 monkeys survived the infection (maximum parasitaemia: <0.01%; 2.7%o). In the lyophilized French press Ag group, 2 monkeys died at the same time as the controls, 2 died between 12 and 15 days after challenge, and 4 survived the infection (maximum parasit- aemia: 1.2%; 0.4%; 0.9%; 0.07%). Immnunized monkeysfailed to gain weightduring the period ofimmunization and abscess formation at injection sites was comparable in monkeys ofthe3 Ag groups. Reticulocytosis and anaemia developed in at mnonkeys that survived longer than 12 days after challenge, even in mnonkeys with very low-grade levels ofparasitaemia. Various tests were performned to assess humoral or cell-mediated immune responses during the period of i,n,nunization and after chllenge. No clear-cut relationship could be established between the resutis ofany of these tests and the survival ofimmunized monkeyvs after parasite challenge. Sorne of the results, however, suggest that there might be some relationship between cell-mediated responses and the survival of immunized monkeys beyond rhe survival timne of control monkeys. Immunization of rhesus monkeys with nonviable parasites ofPlasmodiurnt knowlesi was first attempted by Freund et al. (1). Formalin-treated parasitized blood was emulsified in Freund's complete adjuvant (FCA) and injected subcutaneously into rhesus mon- keys on 2 or 3 occasions; sonle of these monkeys sur- vived subsequent challenge with viable parasites of P. knowlesi. Targett & Fulton (2) confirmed these findings and also showed that protective immunity could be induced by freshly collected parasites that Fr om the Division of I ropical and Gcographic Medlicine. LUni- versity of Ncw Memseo School of Medicine. Albuquerquc, NM 87131, USA This u ork was supported by the Agcncy for inter- national Developmenm, US Department of Statc, Contract No. AID/ta-C-1259 Division Head. ' Scientist (Norwich Pharamacal Company, New Yo:k). Research Scientist Biologist. had been freed from erythrocytes by immune lysis. Protectioin was much less effective, however, in monkeys that received lyopliilized preparations of freed parasites. During subsequent studies with blood-stage anti- gens of P. ktrowlesi, Brown and hls co-workers (3) used freeze-thawed preparations of schizont-infected cells emulsified in Freund's adjuvant. They observed that imnmunized animals challenged by a heterologous variant of P. knowlesa were not protected as well as those that were challenged by a homologous variant. Later investigations by Schenkel et al. (4) and Simpson et al. (5) were carried out with lyophilized fractions of parasites disintegrated in a French press (referred to hereinafter as "French press antigen") and emulsified in FCA. Of 17 monkeys that received these antiEenic preparations on 2 occasions 4 weeks apart, 11 survived intravenous challenge with 106 parasites. The identity of the parasite stages in the 381 - 139- K. H. RIECKMANN ET AL. antigenic preparations or of the antigenic vanrants in the challenge inoculum was not determined during the course of these investigations. More recent studies by Mitchell et al. (6, 7) with merozoite antigens were prompted by earlier investigations (8, 9) which suggested that protective antibody affected extra- cellular merozoites released into the blood stream. After immunization with freshly harvested mero- zoites emulsified in FCA, 4 out of 8 monkeys survived initial challenge with the same parasite variant and the other 4 survived initial challenge with a heterologous variant. Two of the 8 animals were immunized twice at intervals of 5 weeks, whereas the other 6 were immunized on 3 occasions at variable intervals over periods ranging from 5 to 12 weeks. In 2 animals that were immunized with merozoites emulsified in Freund's incomplete adjuvant, neither animal sur- vived subsequent challenge with the heterologous variant, despite the fact that one of them received merozoites emulsified in FCA for the first immu- nization. In view of the partial protection thatcan be induced in rhesus monkeys by nonviable blood-stage antigens of P. knowlesi, greater optmism now prevails about the eventual development of a vaccine against human malaria. However, as the earlier immunization studies had been carried out in different laboratories under different experimental conditions, it was difficult to compare the relative efficacy of the different antigenic preparations with one another. Plans were, therefore, made to initiate a sizeable study for a comparative evaluation of the schizont (3), French press (4), and merozoite (6) antigens. These investigations were started after extensive discussions with various inves- tigators and, in particular, with Professor Sydney Cohen, Guy's Hospital Medical School, London, England. The results of these studies are described in this report. In addition to assessing the antimalarial efficacy of blood stage antigens, various other deter- minations were performed before, during, and after immunization. These included erythrocyte and reticu- locyte counts, indirect fluorescent antibody titres, IgG values, radioimmunoassays, B and T lymphocyte counts, lymphocyte transformations, and intradermal skin tests. In addition, limited investigations were per- formed with the opsonization (10) and merozoite inhibition (8) tests. MATERIALS AND METHODS Experimental host Rhesus monkeys (Macaca mulatta) weighing between 3.1 and 5.7 kg were used in this study. They were considered to be in good health by the staff veterinarian and their tuberculin skin tests were negative. Negative fluorescent antibody titres to P. knowlesi indicated no prior exposure to malaria. When serum samples were tested for prior exposure to 10 viruses, 3 monkeys showed positive B virus titres and were excluded from the study. Otherwise, apart from a high prevalence of prior exposure to measles virus, no abnormal fmdings were observed. The 32 monkeys used in the study were divided into 4 categories, which were similar in regard to sex, weight, and measles andbody titres. By random selection of one monkey from each category, 8 similar groups were formed. Through random allocation of these groups, 4 monkeys received phosphate- buffered saline (PBS controls), 4 monkeys received Freund's complete adjuvant alone (FCA controls), 8 monkeys received the schizont antigen and FCA, 8 monkeys received the merozoite antigen and FCA, and 8 monkeys received the French press antigen and FCA. In designing the study, care was taken to prevent any possible bias. Precautions included numerical identification of monkey groups and immunization material, sealing the identity in envelopes, etc., so that persons involved in the immunization and chal- lenge procedures or in the examination ofanimals and blood specimens were unaware of the treatment the animals had received. The clinical condition of monkeys was monitored by the staff veterinarian during the entire study. Abscess formation following immunization and any other side-effects were noted. The quantity of food consumed by each monkey was determined daily and body weights were recorded once a week. Parasites Variants of the W strain of P. knowles, were stored as glycerolized samples of parasitized blood at -70°C. The WI variant provided the parasite material for all antigenic preparations and theW3 variant was used to challenge immunized animals. To obtain parasites for these purposes, nonimmune monkeys were inoculated with thawed-out blood stabilates and parasites were collected during the acute phase of the infection. The serological identity of these parasites was confirmed by the schizont-infected cell aggluti- nation (SICA) test (11). Parasitized blood and variant- specific antisera used in these investigations were obtained from Guy's Hospital, London. Preparation of antigens Schizont antigen was prepared according to the procedure outlined by Brown et al. (3). Heparinized blood was collected from monkey 3089 when parasit- aemia had reached 32°lo and parasites were predomi- nantly schizonts. The blood was centrifuged at 350 g 140 IMMUNIZATION WITH BLOOD-STAGE ANTIGENS 141 for 8 min and the upper "brown" layer of schizont- infected cells was separated from the other cells. The layer consisted of 88% schizont-infected ce}ls (average of 6.7 merozoites per schizont), which were suspended in PBS (pH 7.2) at a concentration of 3.6 x 10 schiz- onts per ml. The preparation was freeze-thaved 3 times and stored at -70 0C until used for the first and second immunization. Merozoite antigen (Ag) for the first immunization was prepared at Guy's Hospital, London, using a modification of the procedure described by Mitchell et al. (6, 7). The procedure included the incubation of schizont-infected cells in Medium 199, enriched with NaHCO3 and glucose, but not containing any rhesus serum. When the culture contained more than 10 merozoites per infected cell, phytohaemagglutinin was added to agglutinate the remaining unruptured cells. After marked agglutination had occurred, the suspension was centrifuged at 250-280 g until the supernatant contained less than I schizont per 200 merozoites. Merozoites for the antigen preparation were collected by centrifuging the supernaEant at 3000 g for 10 min. They were harvested within 2-3 h after release of the first merozoites and were stored at -700C. The merozoite preparation was subsequently brought to our laboratory on dry-ice by Dr G. H. Mitchell. Merozoite Ag for the second immumization was obtained by different procedures from those used for the first immunization, because only a small pro- portion of schizont-infected cells ruptured during the first few hours of incubation and mature schizonts failed to release an adequate number of merozoites into the surrounding medium. When cultures N%ith high schizont-merozoite ratios were centrifuged after incubation with Bacto-phytohaemagglurinin (PRA) P, many free merozoites appeared to be trapped in or attached to sedimented ervthrocyte clumps and very few could be harvested from the supernatant fluid. In order to increase the yrield of merozoites, PHA was not added to the cultures and merozoites were col- lected by filtration of parasite suspensions through polycarbonate mcmbranes.a Merozoite antigen for the second immunization was obtained from monkey 3171 when a parasitaemia of 17%o had been reached. The "brown" layer, containing 88% schizonts (aver- age of 10.0 merozoites per schizont) was incubated in enriched Medium 199 (without serum). Periodic examination of wet films showed that infected cells ruptured slowly, 1.5 x 1010 schizonts present at the onset of culture producing only 1.2 x 1010 merozoites after 6 bours of incubation. Filtration of merozoites, initiated 2 hours later when the total merozoice count was 3.1 x 101 0, was completed lO hours after the start a From Nuclepore Corp., Pleasanton, CA, USA. of incubacion. The filtered material consisted of merozoiLes, malaria pigment, other parasite and erythrocyte debris, and some intact erythrocytes. It was centrifuged at 3000 g for 10 mn and the sediment was collected and resuspended in 8. 5 ml of phosphate- buffered saline (pH 7.2). The suspension, containing 1.64 x 1010 merozoites (1.93 x 109 merozoites per ml), was stored at -70 °C until it was thawed out just prior to the second immunization. French press antigen was prepared according to the method described by Schenkel et al. (4). Heparinized blood was obtained from monkey 2904 when parasit- aemia had reach 200o, and 75%o of the parasites were schizonts (average of 6.9 merozoites per schizont). Disintegration of erythrocytes and parasites was carried out in a French press.b The extract from the disintegrated parasites was subjected to molecular sieve column chromatographv and the void volume eluate collected as 4-ml aliquots into plastic vials. The protein content, as determined by Lowry's method, was 0.813 g/litre of eluate (3.25 mg protein/Nial). Material in the vials was then lyophilized and stored at -70°C until required for the first and second immu- nizations. Immunization Stored antigens of P. knowlesi (WI variant) were thawed out on the day of the immunization and main- tained at 4CC until immunization of animals was completed. Schizont and merozoite preparations contained approximately 2 x 109 parasites per ml and the French press preparations contained 1 mg of protein per mIl of reconstituted suspension. Antigen emulsions were prepared by mixing equal volumes of these suspensions and FCA in a Virtis homogenizer. Mixtures were tested for adequate emulsification by noting their adherence to the homogenizer blade and by the formation of discrete drops after a minute quantity of the emulsion was added to water (12). During the immunization procedure, each monkey received 2 ml of antigen emulsion, consisting of 1 ml of antizen and 1 ml of FCA. Emulsions were delivcred by Luer-lock syringes and 18-gauge needles and were given as deep intramuscular injections into both thighs. Four control animals received 2 ml of PBS atone and 4 additional control animals each received 1 ml of PBS emulsified in an equal volume ofFCA. The entire immunization procedure. consisting of the preparation of emulsion. loading of syringes, and immunization of 8 monkeys, was completed in 50-56 min for each group of animals. The interval between the first and the second immu- b Manufactured by Amencan [nstrument Co., Sigver Spnogs, MD, USA 142 K. H. RIECKMANN ET AL. nization was 43 days. The same frozen schizont and lyophilized French press preparations were used for the first and second immunizations. Merozoite preparations for the first and second immunizations were prepared on two separate occasions using different procedures (see above). Challenge inoculum Monkeys were challenged by intravenous inocu- lation of the W3 variant of P. knowlesi 4 weeks after the second immunization. The inoculum consisted of about 101 trophozoite-infected red blood cells ob- tained from a recently infected donor monkey with a parasitaemia of about 1%. Collection of blood specimens Blood specimens for parasitological examination were obtained by collection of a few drops of blood from the heel. Blood was also collected once or twice a week in preservative-free sodium heparin from the femoral vein of monkeys under ketamine hydrochlo- ride anaesthesia. In order to prevent excessive with- drawal of blood, no more than 10 ml of blood was collected from an animal during the course of a week. Identical collections from 2 additional control monkeys had no effect on thelevel oftheir reticulocyte or erythrocyte counts. About 1 ml of blood was used to determine haematocrit values, erythrocyte counts, reticulocyte counts, WBC counts, and differentials. The remainder of the blood was centrifuged and the plasma was saved for subsequent determination of plasma proteins, IgG values, indirect fluorescent anti- body test, radioimmunoassay, opsonization test, and merozoite inhibition test. The packed cells were resus- pended in minimal essential medium for determi- nation of B and T cell values and for lymphocyte transformation with antigens of WI and W3 variants of P. knowlesi. Examination for malaria parasites Thick and thin blood films were prepared from each monkey once a week during the immunization period and once a day for 3 weeks, twice a week for 2 weeks and once a week for 4 months after parasite challenge. They wvere stained with 50 g/litre (5"lo) Giemsa solution and examined under oil for malaria parasites. Films were considered negative if no para- sites were seen during the examination of 100 fields of a dehaemoglobinized thick film (equivalent to about 5 x 10' red cells). At low levels of parasitaemia, thick films were examined by counting the number of para- sites per 200 leukocytes. When the parasite density ex- ceeded 1 parasite per 20 leukocytes (equivalent to about I parasite per 104 red cells), parasites were counted against RBC in methanol-fixed thin films. Parasitaemia was recorded as the number of infected erythrocytes per 104 RBC. Determination of haematologicalparameters Routine determination of haematocrit values, erythrocyie counts, reticulocyte counts, WBC counts, and differentials were performed once or twice a week throughout the period of the study. Erythrocyte and WBC values were determined in a Coulter counter. Reticulocyte percentages were determined after exam- ination of 500 erythrocytes in thin blood films stained with methylene blue. Indirect fluorescent antibody determinations Plasma specimens were forwarded to the Center for Disease Control, Atlanta, and indirect fluorescent antibody titres to P. knowlesiantigenwere determined in the laboratory of Dr W. E. Collins according to standard procedures (13). Immunoglobulin G determitnations Serum IgG levels were determined by the radial immunodiffusion method of Fahey& McKelvey (14). Rabbit anti-rhesus igG was incorporated into agar for the immunodiffusion assay. Standards, prepared from purified rhesus IgG, were included in each assay. Dilutions (1:20) of plasma were added to wells in the agar plate and the diameter of the precipitin rings was determined 72 h after incubation at 4°C. IgG values were determined from the standard curves plotted from diameters of IgG standard precipitin rings. Radioimmunoassay (RIA) A solid-phase RIA was developed by E. J. Cabrera et al. (unpublished results, 1975) in which rhesus anti- body reacting to parasite antigens bound on the sur- face of test tubes, could be detected by reaction with 125 I-labelled anti-rhesus irmunoglobulin. Lyophi- hzed WI and W3 French press antigen was reconsti- tuted with phosphate-buffered saline (pH 7.2) to a concentration of 100 mg of protein per litre. Poly- propylene Bio-vialsc containing I ml of antigen solu- tion were incubated for I h at 4 IC. After removal of antigen for later use, 1.5 ml of 10 g/litre (1%) ovalbu- min was added to cach tube and allowed to stand at room temperature for 15 min. The tubes were then rinsed twice with PBS, allowed to dry overnight, and stored at room temperature. One ml of test plasma, diluted 1:100 in PBS, was added in triplicate to antigen-sensitized vials and " From Beckman Instruments Inc., FuUeiton, CA, USA. IMMINIZATfON WITH BLOOD-STAGE ANTIGENS 143 allowed to incubate at room temperature for 1 h. After incubation, the plasma was removed and the tubes were rinsed 3 times with PBS. One ml of the 125I-labelled immunoglobulin fraction of rabbit anti- rhesus immunoglobulin G (5000 cpm/,ig protein) was added to the tubes and incubated for 2 h at room temperature. The antibody was removed and the tubes were again rinsed 3 times with PBS and capped. Radioactivity was counted in a gamma spectrometer and the data recorded as counts per minute (cpm). B and T cell determinations Peripheral blood B and T lymphocytes were esti- mated by the rosetting techniques of Jonsson (15). After successive incubation of sheep erythrocytes (E) with antibody (A) and complement (C), the number of rosettes (3 or more sheep erythrocvres suirounding a lymphocyte) among 100 lymphocytes was counted; these EAC rosettes were considered to represent the percentage of B cells in the lymphocyte population. The percentage ofT cells was determined by E rosette formation, a procedure that does not involve the prior exposure of sheep erythrocytes to antibody or com- plement. Lymphocyte transfornmation Peripheral blood lymphocytes collected by using Lymphocyte Separation Mledium,d were diluted to a concentration of 1.25 x 106 per ml in Medium 199. Each ml of this cell suspension also contained 10%o normal rhesus serum (heat inactivated at 56°C for 30 min), 20 mmol of L-glutamine, 100 IU of pemcillin, and 50 pg of streptomycin. Triplicate 200-pt cultures of this cell suspension were initiated in microculcure trays. Optimal concentrations of phytohaemagglu- tinin P (PHA, 10 mg/litre), concanavalin A (con A, 100 mg/litre), and WI or W3 lyophilized French press antigens (10 mg/litre) were added to appropriate cultures. Mitogen cultures were harvested on day 3 and anti- gen cultures were hanrested on day 5 of incubation at 37 IC in an atmosphere of 50/7 CO2, All cultures were pulsed witli 3.7 x 10' Bq (1 pCi) of methyl-tritiated thymidine (specific activity 7.4 x 10 IDBq (2.0 Ci) per mole), for the final 18 h of incubation. The cells were collected by a mnultiple automated sample harvesEer on glass-fibre filters, after being washed lavishly with 0.15 mol/litre saline, l0O! trichloroacetic acid, and absolute methanol. The filters were placed in 6 ml of scintillation fluid and counted in a scintillation counter (Beckman). The blastogenic index was calcu- lated as the mean counts per nuin (cpm) of triplicate stimulated cultures divided by the mean cpm of tri- d From Bioneties, Kensington, MIA. USA. plicate unstinulated control cultures. Intradermal skin test The abdominal area of the 32 monkeys was shaved at least 24 h before testing. Before the first vacci- nation, all monkeys were injected intradermally in the abdomen with 0.1 ml of PBS containing 25 lAg of P. knowlesi (WI variant) lyophilized French press antigen and. in a separate site, with 0.1 ml of PBS as a control. The test sites were checked at 0.25, 12, 24, and 48 h. All animals were again tested 2 days before challenge and skin reactions were determined at inter- vals identical to those used in the first test. The diameter of the indurated area was measured with a sliding caliper. RESULTS Clinical condition ofrhesus monkeyse The general condition of control and immunized monkeys appeared to be satisfactory during the period of immunization. Although no marked change in their appetite was observed, immunized monkeys failed to gain weight over the 10-week period between first immunization and challenge. The average weight of a monkey increased by 10 g in the schizont Ag group, decreased by 10 g in the merozoice Ag group, and iincreased by 30 g in the French press Ag group. By contrast, the average weight of a monkey increased by 300 g in the PBS control group and by 320 g in the FCA control group. Following challenge of the immunized monkeys, loss of appetite and ac- companying weight loss wras usually observed within 7-11 days (median 9 days) after inoculation. Com- plete loss ofappetite for 2 consecutivedays was always associated with a fatal outcome; this was also observed in monkey No. 30, which died wvith a low- grade parasitaemia. Local side-effects were observed in 14 of the 28 immunized monkeys. Swelling and abscesses near the injection site were considerably more frequent after the secornd immunization than after the first one. These local reactions to FCA were observed in control aniimals aiid in animals of all 3 antigen groups. Reac- tionls were recorded in 6 of the 8 survivors and in 8 of the 16 ammals that succumbed to the infection. No Correladton could be esTablished between local reac- dons and loss of weight or the subsequent develop- ment of reticulocytosis or anaemia. e A senes of tables and figurcs summarizing the data obtained in snd.xidual mronkeys is available (at cost of reproduction) on request from ti c WVHO Library. World Health Organizadon, 1211 Geneva 27, Su itzer1and. K. H. RIECKMANN ET AL. Course of infection after challenge After challenge with a heterologous variant of P. knowlesi, monkeys in the control groups that re- ceived either PBS or FCA developed a high parasit- aemia and succumbed to the infection within 10-12 days. In the schizont Ag group, 6 of the 8 monkeys died at about the same time as the controls and the remaining 2 survived the infection after developing a maximum parasitaemia of 0.7-3.2%. In the mero- zoite Ag group, one monkey died at the same time as the controls and 5 monkeys died between 14 and 17 days after challenge. Monkey No. 30 had developed only a very low parasitaemia by the time it died 15 days after challenge. Of the 2 survivors in this group, monkey No. 17 had a parasitaemia that was always below 0.01%. In the French press Ag group, 2 monkeys died at the sane time as the controls, 2 monkeys died between 13 and 15 days after challenge, and 4 monkeys survived the infection after developing a maximum parasitaemia of 0.07 to 1.2%. Erythrocyte values The erythrocyte values obtained in various monkey groups during the course of the study did not vary appreciably before or during immunization and re- mained fairly constant up to 5-8 days after challenge. The 8 control monkeys and 9 of the immunized mon- keys had succumbed to the infection by the time of the next determination on day 12. By then, erythrocyte values had started to decrease in the 15 surviving monkeys. On day 15, a further decline in values was observed in 9 of the 10 monkeys that were still alive. This was particularly marked in the 2 monkeys that did not survive the challenge and that succumbed to the infection 1 or 2 days later. Of the remaining 8 monkeys that survived the infection, 4 showed a further decrease in values on days 19 and 22 (Fig. 1). During the third week after challenge, mean erythro- cyte counts had dropped by 1.7 x 101 cells/mm3 in the schizont Ag group, 2.1 x 106 cells/mm3 in the merozoite Ag group and 1.5 x 106 cells/mm3 in the French press Ag group. By day 20, however, RBC counts were increasing in all monkeys and they had re- turned to prechallenge levels by days 36 and 43. The percentage of reticulocytes remained fairly constant both before and during the period of immu- nization and values above 3% were observed in only 2 of the 32 animals (No. 12 and 14). Both monkeys had reticulocyte counts of4-7qo on day 21 or 26 after the first immunization and on day 8 or 15 after the second immunization. Plasma collected from them on the day before challenge yielded a positive indirect Coomb's test with a titre of 1:128; other monkeys gave negative tests or had much lower titres. Reticulo- cyte values did not exceed 3% in any of the animals that failed to survive more than 12 days after challenge. Of the 6 animals that died between days 12 and 15, only 2 showed elevated reticulocyte counts (4% and 14%). On the other hand, all 8 surviving monkeys registered reticulocyte values between 7%c hnd 37%o, the peak response being observed during the third week after challenge (Fig. 1). Elevated values were even recorded in monkeys Nos. 17 and 28, in which the levels of parasitaemia never exceeded 0.0701. By the end of the fifth or sixth week, reticulo- cyte counts had returned to prechallenge values Indirectfluorescent antibody test (IFA T) Fluorescent antibody titres obtained after exami- nation of plasma samples from 24 monkeys in the 3 immunization groups showed that, during the course of imunization, all immunized monkeys gave titres of 1:20 or higher. Meanwhile, the 8 control monkeys continued to give negative results or had titres of less than 1:10. Of the 13 immunized monkeys that attained titres of 1:80 or higher following the second immuniz- ation, 3 died at the same time as the controls after the challenge inoculum, 4 died later than the controls, and 6 survived the infection. No clear relationship could be established between the levels of antibody titre reached during the course of imnunization and subsequent protection against malaria. Within 4 days after challenge, a general decline in antibody levels was observed. Following the development of parasit- aemia, there was a rise in the antibody titre of sur- viving monkeys and levels equalled or surpassed those observed during the course of immunization. Immunoglobulin G values In general, IgG values in plasma samples from 24 immunized monkeys tended to increase in all 3 immu- nization groups following the second immunization, but the increases were more pronounced in the mero- zoite Ag group and least obvious in the schizont Ag group. In the PBS control group, values remained fairly constant throughout the period of immuniz- ation and the highest level recorded did not exceed a mean of 15.3 + 2.1 g/litre. In the FCA control group, however, the mean irnmunoglobulin G concentration had reached 22.8 ± 1.6 g/litre by the 28th day after the second immunization. Within a week after challenge, values were similar to those observed before chal- lenge. Yet, 2 or 4 weeks later, IgG levels in the sur- vivors were generally higher than those recorded earlier. Radioimmunoassay (RIA) values In the control groups, antibody levels measured by RIA in the 8 monkeys varied from 17.4 to 34.2 x 102 144 IMIsIUNIZATION WITH BLOOD-STAGE ANTIGENS O he mmUUl1 fmmf. %.* ftft U ~~ I I Mr I I 15 122 29 36 43 * a l I F ._9. ; *,1s*% SCHIZONT ANTIGEN 25 -15 1S 5 3. 2 - 1 MEROZOITE ANTIGEN -31 - -25 - -26 4 -15 3- 10 2- 5 1- UI I I I 35 22 29 31 43 3I SII a I I 11 I 1 15 22 29 36 13 ,,,o oA,,%.\._-0-" . _ma UT t I 15 22 "! 3i 43 FRENCH PRESS ANTIGEN 64. **% 15 22 23 3C 43 -U-,.-.I._U-bv 2 5 4 30 25 20 is a1 5 a -30 - 25 -20 - 15 5 2-1 1' 0 5 - J 1- a- 4f~~~4 ._-os~~~. . U.,. .,.., s_ _* _ r* * . 1.,,I_ O I 15 22 21 36 43 I I *.. I 15 22 29 3i 43 is 22 21 36 43 DAY DAY Fig. 1. Erythrocyte and reticuklcyte values in monkeys that survived challenge with 104 parasites on day 0 -a n -N 15 is .5 I S 5 4 3 2 6 I5 4 .3 2 a :1- I a Ho 3, 25 21 15 5 I. I- *31 °: I.) v -25 i u 20 It5 u 3D 25 20 15 13 S A. .F ts 0 .*so/*M ~o m.-fUw I I I I 145 K. H. RIECKMANN ET AL. counts per minute. In the immunized group, antibody levels above those in the control groups were observed 26 days after the first immunization. Following the second immunization, an increase in antibody levels was observed in most monkeys belonging to the mero- zoite Ag group and a few monlkeys belonging to the French press Ag group. In comparing antibody levels with the susceptibility of monkeys to malaria, sur- vivors in the merazoite Ag group had slightly higher levels than nonsurvivors on the day of the challenge (second immunization, day 28). Otherwise, antibody levels measured by this technique appeared to show little relationship to subsequent protection of monkeys against parasite challenge. Following the onset of parasitaemia after parasite challenge, no increase in antibody levels was observed. Opsonizatuon and merozoite inhibition tests The opsonization test (10) detects opsonizing anti- bodies during the course of a P. knowlesi infection. Although differences in phagocytosis of parasites could be detected between normal and immune sera, reproducible results could not be obtained with the same or different mouse pentoneal macrophages, normal or immune serum samples, "donor" para- sites, etc. As quantitative standardization of this test was not possible, examination of plasmna samples by this procedure was not pursued any further. The merozoite inhibition test (8) detects antimero- zoite antibodies and has been used in investigations to assess theimmune status of rhesus monkeys to malaria (16, 17). Quantitative standardization of this test for alongitudinal evaluation of the development of mero- zoite antibodies was difficult, from both a technical and a logistical point of view, and therefore, detailed studies with this tecbnique were not carried out. B and T cell values The mean percentage ofB and T lymphocytes in the blood of monkeys, as measured by their ability to form EAC and E rosettes, respectively, remained relatively constant during the period of immunization and no differences were observed between the monkey groups. Three weeks after challenge, how- ever, a marked decrease in the percentage of T cells was noted in the 8 surviving monkeys. This was ac- companied by a corresponding increase in the mean percentage of null cells lacking demonstrable surface receptors. Expression of values in terms of the absol- ute number of lymphocyte types per mm3 was not attempted because WBC counts and differentials were extremely variable from one estimation to another. Lymphocyte transformation test (LTT) The blastogenic response of peripheral blood lym- phocytes from immunized monkeys to W3 variant French press Ag showed characteristic changes. Prior to their first immunization, the 24 monkeys had blastogenic indices of less than 2. Blastogenic indices increased in the 3 immunized groups following the first and second immunization, while those in the control groups remained at baselinie levels. No apparent differences were observed in the blastogenic responses of survivors and nonsurvivors. On the other hand, monkeys that died at the same time as the control monkeys often had lower blastogenic indices than those that survived beyond 12 days after chal- lenge. By day 8 after challenge, blastogenic indices in the immunized monkeys dropped sharply; this was generally more marked in those monkeys that had al- ready developed patent parasitaemia. One week later, on day 15, values were still depressed and they had dropped further for those monkeys that had become patent since day 8. On days 22 and 36, a rise in blastogenic indices was observed in all of the surviving monkeys. Blastogenic indices after transformation of lymphocytes with the WI variant antigen were, gener- ally speaking, slightly higher than those obtained for the W3 variant antigen. Lymphocyte transformation after stimulation with the plant rnitogens concanavalin A (con A) or phyto- haemagglutinin (PHA) was similar in control and im- munized groups. In the PBS group, great variability was observed in the blastogenic indices of individual monkeys, values ranging from 27 to 364 and from 3 to 47 after stimulation with con A and PHA, respec- tively. During the period of immunization, no marked difference could be detected in the blastogenic indices of monkeys belonging to the control or immu- nized groups or between monkeys that survived and those that did not survive challenge. Intradermal skin test P. knowlesi antigen was administered intradermally to 8 control and 24 immunized monkeys 2 days before they were challenged with 10 parasites. None of the control monkeys showed a positive response 24 to 48 h after administration of the test. Results ob- tained after administration of the test to the immu- nized monkeys are show%n in Table 1. Fifteen of the 24 immunized monkeys had a positive skin reaction. In the 8 monkeys that survived challenge, 7 were posi- tive; in the 16 monkeys that died, 8 were positive (Table 2). The difference observed between these 2 groups is statistically not significant and the results indicate that this delayed hypersensitivity reaction does ,aot correlate with survival beyond 17 days after challenge. Survival beyond 12 days, on the other 146 IMMUNIZATION WITH BLOOD-STAGE ANTIGENS Table 1 Results of intradermal skin tests and lymphocyte transformation tests (LTT) performed on 24 immunized monkeys at the time of parasite challenge Reaction to skin testb Survival ___ ___ Survival beyond Biastogenic index d Monkey Monkey after 12 days after Mean diameter (mm) after lymphocyte group No. challenge chalengea ot induration InterpretationC transformation with 24 h 48h Wl Ag W3 Ag Schizont 23 Yes Yes 11 5 + 9 12 antigen 27 Yes Yes 17 13 + 10 13 13 No Yes 4 0 - 8 B 2 No No tO 0 + 3 1 15 No No 4 0 - 11 11 1 No No 3 0 - 8 1 8 No No 0 0 - 54 48 20 No No 0 0 - 13 11 Merozoite 9 Yes Yes 9 4 - 11 12 antigen 17 Yes Yes I0 16 - 16 14 25 No Yes 16 13 418 17 16 No Yes 10 17 + 27 20 30 No Yes 13 17 28 30 3 No Yes 9 B + 45 25 32 No Yes 10 3 I- 11 8 11 No No 2 0 - 3 4 French 12 Yes Yes 19 17 + 43 22 press 14 Yes Yes 2 0 - 9 5 antigen 26 Yes Yes 5 13 + 47 29 29 Yes Yes 8 0 I 32 23 6 No Yes 6 7 -I 176 118 21 No Yes 17 17 + 27 t8 4 No No 0 0 - 10 5 31 No No 0 0 - 10 9 8 control monkeys that received only PBS or FCA died wiThin 10-12 eays after parasitle challenge 4 All monkeys were skin-tested 2 days before challenge wiLn 10' parasitized cells. c Positive reaction was defineEt as skin induwaion exceediig 5 mm in mean dianler 24 hours after initradermal injection of 25 wg of WI vanant P know/em (lyophilized) anbgen suspended in 0 1 ml 0t PBS Negal,ve reactions were racorded in he B =on rol monkeys d Blastogenic indices were determined on the day of Darasite cnallengc. 1kdices below 4 wse'e recorded in th.e e ccntrol mronkeys Table 2. Comparison of positive intradermal skin test reactions in monkeys that survived or died after parasite challenge ,?ositr,e sKn lest " i' 6surviors ce-rond Monkey group survivors 12 davs a.ter challerge Yes Not Yes Noc Schizont antgen 2/2 1,6 2/3 1J5 Merozoite antigen 2i2 5:5 7/7 0:1 French pres antigen 3/4 2/4 5/6 012 Total 7W8d 8. 16d 14 l e 1/8" PBS control - 0/4 - 0/4 FCA control 0/4 - 014 D All monkeys were skin tesied 2 days before parasite challenge b Nonsurvivors died between 10 and 17 days after challenge c Nonsurvivors died at the same brne as control monkeys within 10-12 days after challenge d This difference is statiatzcally not signiftcant IX' = 3 2, P >0 051 " This difference is statistcally highlv si3nificant (X - 12.7; P <0 0051 hand, showed a marked correlation with positive skin reactions. Whereas positi e responses were observed in 14 of th,e 16 monkeys that survived for longer than 12 days after challenge, they were recorded in only I of the 8 monkeys that died at the same time as the con- trols 12 days or less after challenge (Table 2). In comparing the results obtained after skin testing with those obtained after lymphocyte transformation in the presence of WVl or W3 antigen (on day of chal- lenge). negative skin tests were invariably associated with lower blastogenic indices in monkeys that were immunized with merozoite or French press Ag (Table 1). Such a relationship was not observed in monkeys that received the schizont Ag. D[SCUSSION The findings described in tbis report confirm pre- vious observations that partial protection against malaria can be induced in rhesus monkeys after the 147 K. H. RIECKMANN ET AL. administration of blood-stage antigens ofP. knowksi emulsified in Freund's complete adjuvant. After 24 immunized monkeys were challenged with a variant of P. knowlesi different from that used for immu- nization, patent parasitaemia developed in all of the monkeys, 8 monkeys survived the infection, and 8 of the remaining monkeys died later than monkeys in the control groups, i.e., more than 12 days after chal- lenge. The outcome in monkeys that received the schizont or French press antigens was similar to that reported previously (3, 4). As in previous studies, antigen preparations had been stored either frozen (schizontAg) or freeze-dried (French press Ag). In the group of monkeys that received the merozoite Ag, protection was less conisistent than had been observed after immunization with freshly prepared merozoites (7). Storage of the merozoite Ag in a frozen state prior to its use may account for the lower level of pro- tection observed in these studies. Better results might also have been obtained if the 2 immunizations had been administered at an interval of less than 6 weeks. The prolonged period over which merozoites were collected to obtain sufficient antigen for the second immunization may be another reason for the lower degree of protection observed in these studies. Although approximately 60% of the merozoites were released during the last 2 h of incubation, the total incubation period of 8 h was considerably longer than that required to collect merozoites for the first antigen preparation. Protective antigens on the surface coat of the merozoites may have been eluted into the medium or destroyed by proteolytic enzymes during the period of incubation. In a subsequent study, merozoite Ag (WI variant) was prepared by Dr Mitchell in London and forwarded to our laboratory in a frozen state. Except for a short period ofexposure to about -20 IC, the preparation was stored at -196°C. Following immunization of 8 monkeys with this merozoite Ag, according to procedures outlined in this report, 4 of the monkeys survived challenge with the W3 variant of P. knowlesi. The results obtained in these studies do not clearly demonstrate the superiority of one or other of the 3 antigenic preparations in protecting rhesus monkeys against P. knowlesi. In another recent study, 3 owl monkeys (Aofus trivirgatus) were partialy protected against malaria by mature schizonts of P.falcparum (18). This suggests that mature intracellular mero- zoites may be as effective as spontaneously released extracellular merozoites of P.falciparum (19) in achieving protection against malaria. Whatever the relative antigenicities of intracellular or extracellular parasites might be, red cell contaminants will undoubtedly have to be excluded from antigenic preparations that will be used eventually for human immunization. It seems more reasonable, therefore, to pursue further immunization studies with mero- zoites thaL have already been released sponltaneously fiom red blood cells than to attemnpt the freeing and separation of intracellular parasites from their host cells. Despite the relative advantages of using extra- cellular merozoites, considerable additional work is required to define optimum experimental conditions for their release, collection, and storage. For example, serum should probably be incorporated into the cul- ture medium in order to ensure a more prompt and consistent release of merozoites. Particular attention should be given to environmental conditions that might adversely affect the stability of merozoite anti- gens. In addidon, procedures for harvesting mero- zoites will have to be improved in order to exclude the substantial amount of erythrocyte debris currently included in merozoite preparati-ons. Examination of haematological parameters during the course of these studies showed a reticulocyte res- ponse in 2 monkeys after the second immunization, but it did not reveal any change in erythrocyte values. A marked decline in erythrocyte counts was, bow- ever, observed in all surviving monkeys during the second or third week after the challenge inoculum. Anaemia even developed in monkeys 17 and 28, des- pite the fact that parasitaemia did not exceed 0.007%l at any time during the course of their infection. The development of anaemia in the presence of very low levels of parasitaemia is a well-known phenomenon in malaria infections. In drug-suppressed falciparum infections in both man and Aotus, anaemia of the magnitude encountered in these investigations is fre- quently seen a few weeks after the onset of patency. This is observed despite the fact that a parasitaemia of less than 0.01% is maintained throughout the course of infection. Although many etiological mechanisms for such anaemia have been postulated, including various autoimmune phenomena, the precise causa- tive factors have not yet been identified (20). To pre- vent its occurrence, malaria vaccines may have to be developed that completely inhibit any patent or, pos- sibly, subpatent parasitaenia. Various immunological procedures were carried out during the course of these studies, including a number of tests that were reported to have shown some correlation with clinical immunity, e.g., the opsonization test (10), the merozoite inhibition test (8), radioimmunoassay (Cabrera, unpublished re- sults, 1975), and the intradermal skin test (21). Be- cause of the delay in administering the first immuniz- ation, extensive baseline values were obtained over a period of about 3 months. With the exception of the opsonization test and the merozoite inhibition test, the various procedures were reproducible and yielded fairly consistent results over this period of time. In assessing the response of IgG and humoral antibodies 148 IMMUNIZATION WITH BLOOD-STAGE ANTIGENS to immunization and subsequent challenge, an in- crease in values was generally observed after exposure to malaria antigen. No clearcut relationship could be established, however, between antibody levels attained during immunization and subsequent protec- tion observed after challenge. The tendency for anti- body levels to decline, in some instances, immediately after challenge may possibly reflect an absorption of some of the available antibodies by recently intro- duced parasites. In assessing the nonspecific cellular immune res- ponse after immunization, measurement of any change in the total number ofcirculating lymphocytes was hampered by the wide range in leukocyte counts, among individual monkeys from one estimation to another. Although blood samples were collected under anaesthesia at about the same time of the day, the well-known "alarm reaction" probably accounted for the marked variation in total leukocyte levels (22). Despite these limitations, it is noteworthy that the percentage ofB and T cell lymphocytes was similar in the control and immunized groups of monkeys and that values remained fairly constant during the period of immunization and during the first 2 weeks after challenge. By the end of 3 weeks after challenge, a marked decrease in the percentage of T cells and a corresponding increase in the proportion of null cells was observed in surviving monkeys from each of the 3 immunized groups. The changes in circulating lym- phocyte subpopulations observed in our studies have been noted previously in patients infected with falci- parum malaria (23). It has been postulated that tlus phenomenon may be due to the sequestration of mature circulatiing T cells in the spleen and their re- placement by null cells representing either immature lymphocytes lacking demonstrable surface receptors for C'3 or B cells whose C'3 recepcors were blocked by immune complexes (23). When cell-mediated immunity was measured more specifically by stimulation of peripheral blood lym- phocytes with the Wl and W3 variant Frencb press Ag, a marked response was observed after immuniz- ation of monkeys with any of the 3 antigen prepar- ations. Although an increase in the responsiveness of lymph node and spleen cells had been observed previously in monkeys sensitized to P. knowlesi anti- gen and FCA (24), no information has been forth- coming on any correlation between the blastogenic response of lymphocytes before challenge and the length of survival of monkeys after challenge. Although immunized monkeys that died at the same time as the control monkeys tended to have lower blastogenic indices on the day of challenge than those that survived for longer than 12 days after challenge, no clearcut relationship could be established between the duration of monkey survival and the intensity of lymphocyte responsiveness to this particular malaria antigen. Within a week after parasite challenge, a marked decline in blastogenic indices was observed in monkeys of all 3 immunized groups, being least pro- nounced in surviving monkeys that showed no patent parasitaemia at this time (No. 12, 17, and 28). During the course of falciparum infections in patients or Aotus monkeys, no lymphocyte hyporesponsiveness was observed after antigen (25) or mitogen (26) stimulation at comparable or lower levels of parasit- aemia. At higher parasite counts, however, suppres- sion of lymphocyte response to mitogen stimulation was observed (26). Three weeks after challenge, blastogenic indices in 6 surviving monkeys showed a 5- to 21-fold increase. Values in the other 2 survivors (No. 12 and 14) remained low at this time, but showed a 15- to 22-fold increase within a week after the dis- appearance of parasites. It is interesting to note that the lymphocyte responsiveness of the 8 surviving monkeys coincided with the decline in the proportion of T cells observed 3 or 4 weeks after challenge. The results obtained with the skin test should en- courage further investigations using this approacb to study cell-mediated imune mechanisms in malaria. Althougb no correlation between this delayed-type hypersensitivityreaction (19) and survival of monkeys could be established, the mechanisms responsible for inducing positive skin reactions seemed to be related in some way to the extension of survival of monkeys beyond the time that control monkeys succumbed to the infection. Furthermore, negative skin reactions in monkeys of the merozoite and French press Ag groups were invariably observed in those monkeys that had the lowest lymphocyte response after Ag stimulation. This was not observed, however, in monkeys immu- nized with the schizont Ag. The findings observed in these studies emphasize the fact that, despite recent advances, considerable additional work is still -required before a blood-stage malaria vaccine becomes available for human use (27, 28). To prevent undesirable systemic or local side- effects in recipients of the vaccine, particular atten- tion should be given to developing antigenic prepar- ations that do not contain contaminating blood elements and to finding a clinically acceptable substi- tute for Freund's adjuvant. The results of our in vitro assays and skin tests illustrate the complex and dynamic nature of the immune response to malarial antigens. An irnproved understanding o'the sequence of eventS involved in the evolution of this response might provide a means for assessing the host's protec- tive immunity after inoculation. It may also be of value in selecting optimum doses of antigens and adjuvants, the frequency and route of immunization, and other factors that need to be considered in the design of an effective immunization schedule. 149 150 K. H. RIECKMANN ET AL. ACKNOWLEDGEMENTS We wish to thank Professor S. Cohen for contibuting to the design of the study, reviewing its results, and releasing a member of his staff, Dr G. H. Mitchell, to spend about 10 weeks in our laboratory. We also thank Dr Mitchell for his will- ingness to participate in the initial phases of this study and for his expertise and assistance during this period. Others from whom we received valuable advice included Dr L. H. Miller, Dr R. L. Beaudoin, and Dr C. L. Diggs. The fluorescent antibody tests were kindly performed by Dr W. E. Collins and members of his staff, and their contri- buuion to this study is very much appreciated. Finally, we would like to acknowledge the active participation and support of Dr P. Day, Dr B. Snyder, and other members of the Animal Resource Facility in these investigations. RitSUUM IMMUNISATION DE SINGES RHESUS AU MOYEN D'ANT1GtNES DES STADES SANGUINS DE PLASMODIUMKNOWLESI Les travaux d&crits visaient A 1'6valuation, A partir d'un modAle constitu6 par 32 singes rh6sus, de 1'efficacitt protec- trice de trois antig6nes parasitaires des stades sanguins de Pkasmodrum knowkesi-schizonte Ag, m6rozolte Ag et "French press" Ag-non viables et 6mulsionn6s dans de I'adjuvant dc Freund complet (AFC). Apr6s deux injections d'antig6nes opkr&s A 6 semaines d'intervalle, les singes ont t6 6prouvts avec un variantde P. knowlesidifftrent de celii employ6 pour l'immunisation. Chez les 8 singes t6moins ayant regu uniquement soit I'AFC soit une solution saline tamponn6e au phosphate (PBS), une forte parasittmie s'est d6velopp6e et la mort est survenue 12 jours aprts l'inoculation d'lpreuve . Dans le groupe de 8 immunis, avec schizonte Ag d&congelt-recongelk, 5 singes sont morts en meme temps que les singes du groupe thmoin et I un jour plus tard; les 2 autres ont surv6cu (parasit6mie maximale: 0,7; 3,2%o). Dans le groupe immunis6 avec m6rozoite Ag d6congel6-recongelt, I singe est mort en meme temps que les singes ttmoins et5 de 14A 17 jours apres avoir t infectks; Ii aussi, 2 ont surv6cu (parasit6mie maximale: < 0,01%; 2,7%). Dans le groupe irnmunis6 avec French press Ag Iyophlius, 2 singes sont morts en meme temps que les tkmoins et 2 de 12 i 15 jours apr6s l'inoculation d'6preuvc; 4 singes ont survbcu (parasit6mie maximale: 1,2%#; 0,4%; 0,9%; 0,07%). Les singes immunis6s n'ont pas pris de poids pendant la periode d'immunisation et les abc,s qui sont apparus au point d'inoculation Ctaient comparables dans les trois groupes ayant reu un antig6ne. Tous les singes qui ont survOcu plus de 12 jours apr6s l'inoculation d'6preuve ont prhsent6 de P'anmie et un taux accru de r6ticulocytes, meme lorsque la parasit&miectait de faible intensitk. Divers tests ont &t6 effectu&s pour tvaluer les reponses immunitaires humorale et A m6diation ceilulaire pendant la p6riode d'immurisation et apr6s l'preuve. Aucun des tests n'a permis d'1tablir une relation nette entre la survie des singes et les constatations faites. Certains r6sultats ont cependant donn6 A penser qu'il pouvait y avoir un rapport entre les r6ponses A m6diation celulaire et la survie des singes immunisCs au-deli de la dur6e dc survie des singes tAmoins. REFERENCES 1. FREUND, J. ET AL. American journal of tropical medi- cine, 28: 1-22 (1948). 2. TARGETT, G. A. T. & FULTON, 1. D. Experimental parasitology, 17: 180-193 (1965). 3. BROWN, K. N. ET AL Experimental parasitology, 28: 304-317 (1970). 4. SCHENKEL, R. H. ET AL Bulletin of the World Health Organization, 48: 597-604 (1973). 5. SiMPSON, 0. L. ET AL Nature (London), 247: 304-305 (1974). 6. MITCHELL, G. H. ET AL Nature (London), 252: 311- 313 (1974). 7. MITCHELL, G. H. ET AL Immunology, 29: 397-407 (1975). 8. COHEN, S. ETAL. Nature (London), 223: 368-371 (1969). 9. COHEN, S. & BUrCHFR, G. A. Immunology, 19: 369- 383 (1970). 10. BROWN, K. N. & HILLS, L. A. Transactions of the Royal Society of Tropical Medicine and Hygiene, 68: 139-142 (1974). II. BROWN, K. N. Nature (London), 242: 49-50 (1973). 12. HERBERT, W. J. Mineral-oil adjuvants and the immu- nization of laboratory animals. In: Weir, D. M., ed. Handbook of experimental immunology, 2nd ed., Oxford, Blackwell, 1973, p. A2.6. 13. COLLINS, W. E. &SKINNER, J. C. Amencanjournalof tropical medicine and hygiene, 21: 690-695 (1972). 14. FAHEY, J. L. & MCKELVEY, E. M. Journal of immu- nology, 94: 84-90 (1965). 15. JONSSON, V. Scandinavianjournal ofhaematology, 13: 361-369 (1974). 16. BUTCHER, G. A. ET AL Immunology, 34: 77-86(1978). 17. MILLER, L. H. El AL. Experimental parasitology, 41: 105-111 (1977). 18. SIDDIQUI, W. A. Science, 197: 388-389 (19177). 19. MITCHELL, G. H. ET AL. Lancer, 1: 1335-1338 (1977). 20. ZUCKERMAN, A. Military medicine, 131 (Suppl.): 1201- 1216 (1966). 21. CABRERA, E. J. eIT AL Zeitschrift fur Parasitenkunde, S0: 31-42 (1976). IMIMUNIZATION WITH BLOOD-STAGE ANTIGENS 151 22. SCHALM, 0. W. ET AL Veterinary hematology, 3rd ed., Philadelphia, Lea & Febiger, 1975, p. 261. 23. WYLER, J. Clinical experimental immunology, 23:471- 476 (1976). 24. PHILLIPS, R. S. ET AL. Experimental parsrtology. 28: 339-355 (1970). 25. WYLER, K. J. & BROWN, J. Clinical and experimental immunology, 29: 40140 (1977). 26. TAYLOR, D. W. & SIDDIQX,'[ W. A. Amencan journal of tropical medicine and hygiene, 27: 738-742 (1978). 27. MILLER, L. H. Journal ofinfectiousdiseases, 135: 855- 864 (1977). 28. COHEN, S. Journal of the Royal Society of Medicine, 71: 476-478 (1978).