IN VITRO GROWTH AND MULTIPLICATION OF THE MALARIA PARASITE PLASMODIUM KNOWLES1 BY ERIC G. BALL CHRISTIAN B. ANFINSEN QUENTIN M. GEIMAN RALPH W. MCKEE RICHARD A. ORMSBEE THE SCIENCE PRESS Reprinted from SCIEXCE, May 25, 1945, Vol. 101, i';o. 2630, pages 542-544. Reprinted from SCI~;XCB, May 25, 1945, Vol. 101, No. 2630, pages 342-544. IN VITRO GROWTH AND MULTIPLICATION OF THE MALARIA PARASITE, PLAS- MODIUM KNOWLESI', 2 WHETHER one is considering the biochemical, bio- logical, immunological or chemotherapeutic aspects of the malaria problem, questions arise which could be more readily answered by experimentation on parasites grown in vitro than on those sheltered by the host. This well-recognized fact has led many previous investigators to attempt the ila witro cultiva- tion of the malaria parasite. Undoubtedly, the most successful and best-documented work to date on the erythrocytic form has been that of Trager,3*4 who clearly demonstrated that the bird malaria parasite, P. lophurae, would survive in vitro up to 16 days at body temperature. Never did Trager observe an in- crease in vitro in the total number of parasites com- parable to that seen in the host. The parasite popu- lation in his experiments usually remained constant for the first few days and then declined. He con- cluded that, though development must be continuing in such preparations, the death rate, particularly after the first few days, far exceeds the birth rate. We wish to report here a brief summary of the results obtained during the past year on the in vitro cultivation of the erythrocytic form of the malaria parasite, P. knowlesi, in which we have regularly observed growth and good multiplication. P. knowlesi was chosen for our work because this parasite has a 24-hour cycle, produces a heavy infection of red cells in the monkey, Xacaca ndatta, and will also produce infection in man. Moreover, since our chief interest lay in making biochemical and metabolic studies on the parasite, we preferred to deal with a host pos- sessing a non-nucleated red blood cell. Two types of techniques have been developed. One, which we have termed the rocker-dilution, consists of the dilution of 1 part of whole blood with 3 parts of a nutrient fluid whose composition is given in Table 1. This mixture, usually 6.0 ml in total volume though 50 ml have been used, is placed in a tube or flask equipped with gas inlet and outlet tubes. The con- tainer is then placed on a rocking machine which just keeps the red cells in gentle motion. A slow flow of 5 per cent. CC,: 95 per cent. air is passed into the 1 From the Department of Biological Chemistry, Bar- vard Medical School, and the Department of Comparative Pathology and Tropical Medicine, Harvard School of Pub- lic Health and Harvard Medical School, Boston. 2 The work described in this paper was done under a contract, recommended by the Committee on Medical Re- search, between the Of&e of Scientific Research and De- veloument and the President and Fellows of Harvard College. 3 W. Trager, Jour. Exp. Med., 74: 441, 1941. 4 W. Trager, Jour. Exp. Med., 77: 411, 1943. TABLE 1 Coarrmus~rs or NUTRIENT MEDIUM* gm/L microgm/L t.2:: ............... ............... KC1 ................ NaCl ............... NazHPOd ............ EiEe:+ ........... ............. Difco Proteose Pe tone Stearns Amino AC ds$ . P Glycerol ............ Sodium Acetate ...... ,095 .056 .410 ":K f $X 1:50 0.50 0.25 0.15 Adenine Sulfate . . . ,,,,nt;HCl . . . . . . . . . . . . . . . gtxd;ine . . . . . . . . . ;;;;i;bic* ;Q'&' : : : : : . . . . . . . . . . . Choline . . . . . . _ . . . Cocarboxylase . . . . . Nicotinic Acid _ . . . Nicotinamide . . . . . d-Ca Pantothenate . Pyridoxine . . . . . Bibose . . . . . . . . . . . !%2%? . . . . . . . . . . . . . . . . . * the The freezing point of this medium is - 0.60" C. +_ .02, and air pH after equilibration with 5 per cent. CO2 : 95 per cent. IS 7.45i 0.1. t ing Added as NazCO; and converted to bicarbonate by pass- COz gas through the solution. $ Fortified with glycine and histidine. vessel without being allowed to bubble through the liquid. The whole procedure is carried out under sterile conditions and the cultivation performed at a temperature of 38.5' C. Best results have been obtained when the number of total parasitized cells in the initial mixture is limited to not more than 25,000 per cumm., and the total number of red blood cells to 1.25 x 10" per cu.mm. This is usually accom- plished by diluting highly parasitized blood with several times its volume of normal blood in order to ensure the presence of an adequate amount of nor- mal blood constituents in the final mixture. The other technique, which has been termed the rocker-perfusion, employs a Cellophane membrane to separate whole parasitized blood from the nutrient medium. Several types of apparatus for this pur- pose have been developed. For volumes of blood approximating 1 ml, t,ype 1 is used in which a Cello- phane membrane is stretched over one end of a glass tube 30 mm in diameter, and the blood placed within the tube directly upon the Cellophane membrane. This tube is mounted in a vessel containing about 30 ml of nutrient fluid in such a way that the mem- brane end is just wetted by the nutrient fluid. For volumes of blood approximating 15 ml, type 2 is used in which a coil of Cellophane tubing (8/32 in. dia.) supported by a glass form is immersed in the blood and nutrient fluid allowed to flow through the Cello- phane tubing from a reservoir at a rate of approxi- mately 1,500 ml per 24 hours. In both types of . . apparatus, the container is rocked with a slow stream of 5 per cent. CO,: 95 per cent. air passing through it. All operations are performed under sterile con- ditions and cultivation carried out at 38.5' C. As in the rocker-dilution technique, we have usually em- ployed parasitized blood diluted with normal. 2 TABLE 2 SUMMARY OF IN VITRO CULTIVATION EXPERIMENTS IN wHICII TWOFOLD OB BETTER MULTIPLICATION OF PARASITES HAS OCCC~RED IN 2(r24 Hooas Number of experiments with the Total multiplication listed Avg. Method expts. fo2ld 6-8 multipli- f :d f :d f cfld cation fold p;;; Rock:r 23 2 5 8 5 3 4.0 Perfusion 22 3 9 3 3 4 4.0 22Eerl Perfusion 44 19 8 10 4 3 3.2 Type 2 Cultivation with all methods is usually carried on for a period of 24 hours. If subculture is performed, then a fresh setup is made each 24 hours, the culti- vated parasitized blood being diluted with normal whole blood and the nutrient fluid replaced by fresh. A summary of the results obtained with the various techniques is given in Table 2. In the 89 experi- ments given here, the average increase in parasite count within 24 hours has been three- to fourfold. In some sixty monkeys studied to date, the average increase ir. viwo for a similar period, and at compar- able percentages of parasitization has also been of this order of magnitude. TABLE 3 DE.?AIL~D COUNT ON CULTIVATION EXPERIMENTS RUN BY BOTH TECHNIQUES Time (hrs.) 0 5 10 24 Techniaue* R.D. R.P. R.D. R.P. R.D. R.P. R.D. R.P. Red cells/ cmm x 106 Parasite count, per cent. Dlfferentlal Dbti8t;tion per'cent. Tropho- -" -~~. per cent. Degener- cite and unreeog- nizable, per cent. 1.32 3.98 1.30 4.07 1.8 2.0 2.0 2.2 1 1 1 2 96 83 29 38 3 14 69 52 0 1 0 6 0 1 1 2 0 0 0 0 . . 4.04 1.10 3.78 . . 9.2 7.4 9.6 . . 61 3 12 . 31 95 80 . . 3 0 1 . . 2 0 0 * . 1 2 4 . . 2 0 3 * R.D. = Rocker Dilution. R.P. = Rocker Perfusion, Type 2. The blood for these two experiments came from two different monkeys 80 that the similarity in count is merely fortuitous. In Table 3, a complete differential count for one experiment run by each technique is given. The shift in distribution of the various forms with time is clearly evident, segmentation having occurred sometime between the fifth and tenth hour of ita vitro cultivation. At the end of 24 hours, the distribution of forms is about the same as at the start of the experiment though four to five times as many para- sites are then present. Differential counts for all the 59 experiments summarized in Table 2 are also on record and show a similar picture. Using the rocker-dilution technique, three succes- sive generations of P. knowlesi have been grown in vifro, while with the rocker-perfusion (type 1) technique six successive generations have been ob- tained. In the latter case, the infectivity of the blood after six days' culture in vitro was demonstrated by inoculation into a monkey. It should be pointed out, however, that in order to obtain subculture with the rocker-perfusion technique, it is necessary to replace about one fourth of the nutrient medium with blood serum. Indeed, we have recently found that better in vitro multiplication and growth result during the first 24 hours by the perfusion technique if some serum is present in the nutrient medium. This fact emphasizes one of the basic reasons for our employ- ment of two types of cultivation techniques. We find that the perfusion technique is particularly useful for the study of the nutrient requirements of the parasites, because any low molecular weight material essential for growth and not present in the nutrient fluid will be dialyzed away from the parasite and its deficiency rapidly made apparent. In the rocker- dilution technique, it is obvious that such deficiencies in the nutrient media will not be so readily observ- able. However, the rocker-dilution technique, be- cause of its simplicity and lack of a Cellophane mem- brane, has proved very useful in testing the action of antimalarial drugs, immune serum, etc., on the growth of the malaria parasite iva vitro. Chemical and metabolic studies have also been made 011 parasites grown ir, vivo and in vitro. Increases have been observed in the content of fatty acids, flavine adenine dinueleotide, total phosphorus, ELmin- ute hydrolyzable phosphorus, phospholipid phos- phorus and nucleic acid phosphorus (by difference) in the red blood cells as their parasite count increases either in vitro or in vivo. Similar studies on glucose and oxygen consumption and lactate production have also been carried out. We have encountered a strik- ing difference between in vitro and C vivo grown parasites only in their oxygen consumption. Multi- plication of parasites in &tro has not been attended by the same increase in oxygen consumption that is observed during multiplication in wizro. We have as yet no explanation for this phenomenon, but it may be a reflection of some deficiency or toxic agent in our media or an indication that, if the parasite is well supplied with nutrients, it can exist largely on energy derived from anaerobic processes. In sup- port of the latter explanation is our finding that a gas phase low in 0, (0.37 per cent. 0, : 5 per cent. CO,: 94.63 per cent. N2) permits at least as good growth and multiplication in vitro as in 95 per cent. 3 air: 5 per cent. CO,. Definitely detrimental to in vitro growth is a gas phase high in oxygen content (95 per cent. 0, : 5 per cent. CO,). The composition of the nutrient medium employed and given in Table 1 was arrived at by a priori reasoning. We can not say at present how many of the components of this medium are essential for growth of the parasite. Omission of the proteose peptone is, however, definitely detrimental to growth and multiplication. Recent experiments indicate that para-amino benzoic acid is probably the chief essen- tial growth component furnished by the proteose pep- tone, a fact which may help to explain the observa- tion of Coggeshal15 that sulfanilamide will eradicate I'. I~~o~wZesi infections in monkeys. The techniques described here are also now being . applied with the assistance of Dr. J. W. Ferrebee to the ilz vitro cultivation of human malaria para- sites. Results to date have been encouraging and . will be reported at a later time. ERIC 0. BALL CHRISTIAN B. AXFINSEN QUENTIN M. GEIMAN RALPH W. MCKEE RICHARD A. ORMSBEE 5 L. T. Coggeshall, Jour. Eep. Med., 71: 13, 1940.