Intraerythrocytic Plasmodium falciparum Growth in Serum-Free Medium with an Emphasis on Growth-Promoting Factors

Malaria remains a devastating disease, particularly in the tropics. The annual incidence of malaria worldwide is estimated to be between 294 and 500 million clinical cases, while estimates of annual mortality from malaria, caused largely by the protozoan Plasmodium falciparum, range from 0.97 to 2.7 million worldwide (World Malaria Report 20101, World Health Organization; Snow et al., 2005). The emergence of resistance to conventional antimalarial drugs and insecticides means that new chemotherapeutic approaches with alternative targets are needed (Ridley, 2002). Better understandings of antimalarial drugs and the biology of the parasites are needed to allow the development of new medications.


Introduction
Malaria remains a devastating disease, particularly in the tropics. The annual incidence of malaria worldwide is estimated to be between 294 and 500 million clinical cases, while estimates of annual mortality from malaria, caused largely by the protozoan Plasmodium falciparum, range from 0.97 to 2.7 million worldwide (World Malaria Report 2010 1 , World Health Organization; Snow et al., 2005). The emergence of resistance to conventional antimalarial drugs and insecticides means that new chemotherapeutic approaches with alternative targets are needed (Ridley, 2002). Better understandings of antimalarial drugs and the biology of the parasites are needed to allow the development of new medications.
A review of the impact of continuous cultures of P. falciparum underscores their significant contributions to malaria research (Trager & Jensen, 1997). The mechanisms responsible for the growth of the parasite, however, remain largely unknown. Culture media for P. falciparum require human serum, a growth-promoting fraction derived from adult bovine plasma (GFS), or lipid-enriched bovine albumin (Asahi & Kanazawa, 1994;Asahi et al., 1996;Cranmer et al. 1997;Jensen, 1979). Elucidation of the factors able to induce the growth of P. falciparum could be of help, not only for successful culture of the parasite, but also for providing critical clues to understanding the biology of parasite proliferation during the erythrocytic phase.
In order to identify the factors controlling parasite development, and the effects of growthpromoting factors on the parasite, we initially investigated growth-promoting substances to formulate a chemically defined culture medium (CDM) suitable for sustaining the complete development and intraerythrocytic growth of P. falciparum. We also developed a simple and sensitive flow-cytometry-based assay for following each developmental stage of the parasite's erythrocytic growth. The distinct roles of the growth-promoting factors in the growth of P. falciparum were then investigated.

Parasites and culture
Cultures of the FCR3/FMG (ATCC Catalogue No. 30932, Gambia) strain of P. falciparum were used in all experiments. The parasites were routinely maintained using in vitro culture techniques. The culture medium was devoid of whole serum, and consisted of basal medium supplemented with 10% GFS (Daigo's GF21; Wako Pure Chemical Industries, Japan), as previously reported (Asahi, 2009;Asahi & Kanazawa, 1994;Asahi et al., 1996;Asahi et al., 2005;Asahi et al., 2011). This complete medium is referred to as GFSRPMI. The basal medium consisted of RPMI-1640 containing 2 mM glutamine, 25 mM 4-(2hydroxylethyl)-piperazine ethanesulfonic acid, 24 mM NaHCO 3 (Invitrogen Ltd., USA), 25 µg/ml gentamycin (Sigma-Aldrich Corp., USA) and 150 µM hypoxanthine (Sigma-Aldrich). Briefly, erythrocytes (RBC) were preserved in Alsever's solution (Sigma-Aldrich; Asahi et al., 1996) for 3-30 days. They were then washed, dispensed into 24-well culture plates at a hematocrit (% of packed RBC in medium) of 2% (1 ml of suspension/well), and cultured in a humidified atmosphere of 5% CO 2 , 5% O 2 , and 90% N 2 at 37ºC. For subculture, 3-4 days after inoculation, infected RBC and uninfected RBC were washed. Parasitemia (% of infected RBC in total RBC) was adjusted to 0.1% (for subculture) or 0.4% (for growth tests), by adding uninfected RBC, and the hematocrit was adjusted to 2% by adding the appropriate volume of either GFSRPMI or the test medium. The cultures were synchronized at the ring stage by three successive exposures to 5% (w/v) D-sorbitol (Sigma-Aldrich) at 41-and 46-h intervals. After the third sorbitol treatment, residual schizonts and cell debris were removed by isopycnic density centrifugation on 63% Percoll PLUS (GE Healthcare Bio-Science Corp., USA). The current cultivation method remains essentially the same as initially described (Jensen, 2002), with a few refinements, particularly in terms of the culture medium. The growth experiments were performed by replacing GFSRPMI with test samples. Growth rate was estimated by dividing the parasitemia of the test sample after incubation by the initial parasitemia.

CDM for continuous intraerythrocytic growth of P. falciparum using lipids
The P. falciparum parasite develops through three distinct stages within RBC during its cycle of approximately 48 h: ring, trophozoite, and schizont (Bannister et al., 2000). However, the development of P. falciparum requires the presence of currently unknown factors present in human serum (Jensen, 1979). Although numerous studies have attempted to identify the factors and substances able to sustain parasite growth (Asahi & Kanazawa, 1994;Asahi et al., 1996Asahi et al., , 2005Cranmer et al., 1997;Divo and Jensen. 1982;Lingnau et al., 1994;Mi-Ichi et al., 2006;Nivet et al., 1983;Ofulla et al., 1993;Willet and Canfield, 1984), the establishment of a fully-defined culture medium for the parasite has represented a major challenge. We previously reported that GFS supported intraerythrocytic growth of P. falciparum (Asahi and Kanazawa, 1994;Asahi et. al., 1996;Asahi et al., 2005). GFS contains lipid-rich albumin as a major component. Similarly, Cranmer et al. (1997) described a commercially available lipidenriched bovine albumin (Albumax II; Invitrogen) that could replace human serum for the in vitro cultivation of P. falciparum. These serum substitutes are currently widely employed to maintain parasite cultures. However, there is still insufficient information on these indispensable additives to allow direct identification of the functional components required for the growth of P. falciparum. The replacement of human serum or GFS in culture medium with chemically-or functionally-defined substances could not only be advantageous for www.intechopen.com parasite culture, but could also provide critical clues about the parasite's biology and its requirements for proliferation at the erythrocyte stage. To establish a CDM for continuous, intraerythrocytic growth of P. falciparum, we initially characterized the ability of the various components of GFS to sustain parasite growth (Asahi et al., 2005). Based on these results, we subsequently determined the ability of structurally defined chemicals to sustain parasite growth, and formulated a CDM for P. falciparum growth (Asahi, 2009).

The ability of GFS components to sustain parasite growth
We investigated the components of GFS and related substances that have shown an ability to sustain parasite growth (Asahi et al., 2005). A simple total lipid fraction of GFS obtained by lipid extraction has been shown to sustain complete parasite development. However, specific proteins such as bovine and human albumin, as well as the simple total lipid fraction of GFS, have also been shown to be important ( Figure 1a). The simple total lipid fraction of GFS contained phospholipids, diacylglycerides (DAG), cholesterol, monoglycerides, nonesterified fatty acids (NEFA) and cholesteryl esters. The components of the NEFA fraction were mainly cis-9-octadecenoic acid (C18:1-cis-9, 43%), hexadecanoic acid www.intechopen.com (C16:0, 21%), octadecanoic acid (C18:0, 14%), cis,cis-9,12-octadecadienoic acid (C18:2), cis-9hexadecenoic acid (C16:1), cis-5,8,11,14-eicosatetraenoic acid (C20:4), cis-5,8,11,14,17eicosapentaenoic acid (C20:5), and cis-4,7,10,13,16,19-docosahexaenoic acid (C22:6). Each NEFA enriched with bovine albumin (fatty-acid free) was tested for its ability to promote parasite growth. Mixtures of NEFA, but not individual NEFA, sustained parasite growth to a low extent (Figure 1b), but parasite growth in the presence of various combinations of NEFA was still lower than that achieved with a simple total lipid fraction of GFS, or with GFS-or human serum-containing medium. These results implied that, although the NEFA components of the simple total lipid fraction are functional factors in promoting parasite growth, other factor (s) must also contribute to the high growth-promoting activity of GFS.

CDM for intraerythrocytic growth of P. falciparum
Initial experiments designed to determine the factor (s) responsible for the high growthpromoting activity of GFS involved culture of P. falciparum with the lipid classes found in the simple total lipid fraction of GFS, and different concentrations of a mixture of the two most abundant NEFA, C18:1-cis-9 (0-60 µg/ml [212.4 µM]) and C16:0 (0-30 µg/ml [117.0 µM]) at a ratio of 2:1. The growth rate was dependent on the concentrations of the NEFA in the mixture: the maximum effect was obtained with 30 µg/ml C18:1-cis-9 plus 15 µg/ml C16:0, with declines at lower and higher concentrations. However, the growth rates were much lower than that obtained with a simple total lipid fraction of GFS or with GFSRPMI.
A mixture of all the constituents detected in a simple total lipid fraction of GFS sustained complete parasite growth. In an attempt to identify the factor (s) responsible for this growthpromoting effect, each lipid was omitted from the medium in turn. Parasite growth in the absence of phosphatidylcholine (PC) decreased to a level similar to that seen with the NEFA mixture ( Figure 2a). Phospholipids were also tested for their possible efficacies in augmenting the ability of NEFA to promote parasite growth, by adding each phospholipid to cultures with NEFA. Phospholipids, even PC alone, markedly amplified the growthpromoting ability of the NEFA mixture ( Figure 2b). These results indicate the critical importance of PC for amplifying the parasite-growth-promoting ability of NEFA mixtures. Phospholipids other than PC, such as phosphatidylethanolamine (PE), phosphatidylserine (PS), and phosphatidic acid (PA), were also beneficial for parasite growth, while the growth rates in the absence of phosphatidylinositol (PI), cholesterol, and cholesterol ester were significantly higher (Figure 2a), indicating these phospholipids were detrimental. DAG had no effect on the growth rate of the parasite at the concentrations tested.
Specific proteins such as bovine and human albumin were shown to be required for P. falciparum growth in serum-free culture with lipids, as stated above. Recombinant human albumin could replace serum albumin for sustaining parasite growth in the presence of lipids ( Figure 6).
All stages of P. falciparum cultured in the formulated CDM containing the best combination of two NEFA, phospholipids, and human, bovine, or recombinant albumin were morphologically indistinguishable from growth in complete medium (Figure 7).

Development of a measure of intraerythrocytic growth of P. falciparum using flow cytometry and SYBR Green I
Growth-promoting and antimalarial effects on plasmodia can be assessed both quantitatively and qualitatively by directly examining RBC smears from blood or cultures under a microscope; however, this method is tedious and subjective. Numerous novel in vitro assays have been introduced that are more objective, faster, more sensitive, and designed to be easier to handle. The most common of these include isotopic, enzymatic, and enzyme-linked immunosorbent assays (ELISA) (Noedl et al., 2003). Isotopic assays rely on the incorporation of radioactive 3 H-hypoxanthine into the parasite DNA (Noedl et al., 2003;Webster et al., 1985;Yayon et al., 1983). These methods are relatively reliable and objective, but not sufficiently sensitive, and require the use of hazardous radioactive material. The assays are well suited for screening large numbers of compounds. Parasite lactate dehydrogenase levels have also been used to assess the growth of malarial parasites (Asahi, et al., 2005;Makler and Hinrichs, 1993;Noedl et al., 2003). ELISA-based assays can provide measures of parasite growth by quantifying biomolecules produced during parasite development, such as histidine-rich protein 2 or parasite lactate dehydrogenase, by double-site sandwich ELISA (Druihe, et al., 2001;Noedl et al., 2002;Noedl et al., 2003). ELISA-based tests are rapid and easy to perform, and are also well suited for the screening of large number of drugs. These methods have been widely employed to detect and analyze the growth of parasites, although they are poorly suited for discriminating the developmental stages of the parasite in parasitized RBC. Flow cytometry using nucleic acid staining offers the possibility of studying the cell cycle and developmental stages of intraerythrocytic growth of malaria parasites. Flow cytometric analysis using intercalating dyes, such as acridine orange, thiazole orange, hydroethidine, propidium iodine, YOYO-1, and SYTO-16, has already been used successfully to test human and murine samples (Barkan et al., 2000;Janse and Van Vianen, 1994;Jimenez-Diaz et al., 2009;Jouin et al., 1995;Li et al., 2007;Nyakeriga, 2004;Persson et al., 2006). However, the use of flow cytometry has been limited by its lack of specificity and the complicated preparation required. We modified the flow cytometry system and introduced SYBR Green I as an intercalating dye, allowing the growth and development of P. falciparum to be analyzed with a high degree of accuracy (Izumiyama et al., 2009).

Optimization of flow cytometric measurement of infected RBC with SYBR Green I
The cytometer was equipped with a single argon-ion laser tuned to a fluorescence excitation of 488 nm for 15 mW output (PAS flow cytometer, Partec Co. Ltd., Germany). A FACSCalibur (Becton Dickinson Immunocytometry Systems, USA) was also used with a single fluorescence measurement (530 nm). Analysis was performed using FCS express software (De Novo Software Inc., Canada).
In order to distinguish infected RBC or merozoites from platelets, leukocytes, and RBC debris by flow cytometry, it is essential for them to retain their normal morphological characteristics and membrane integrity, without disruption. We evaluated the effects of (1) fixatives, (2) dilution buffer and concentration range of SYBR Green I, and (3) staining period on the intensity of fluorescence of infected RBC. (1) Infected RBC in culture were fixed using fixatives such as 1% paraformaldehyde or 1% glutaraldehyde in Tris-saline solution (20 mM Tris (hydroxymethyl) aminomethane hydrochloride at pH 7.2 and 138 mM NaCl), phosphate-buffered saline (PBS) (10 mM phosphate buffer at pH 7.2 and 138 mM NaCl), or Alsever's solution. Paraformaldehyde combined with Alsever's solution proved the most useful fixative, with no noticeable lysis or deformity of infected/uninfected RBC, such as often occurs when Tris-saline or PBS is used to dilute fixatives. (2) Concentration ranges and dilution buffers for SYBR Green I were tested to determine the optimal solution for producing clearly resolved peaks of fluorescence. Individual peaks in histograms corresponding to the development stages of infected RBC showed most clearly with SYBR Green I diluted at concentrations ranging from 0.00625-2× in Tris-saline at pH 8.8 (SYBR Green I-basic). Both fixed and unfixed infected RBC gave the best results at 1× dilution of SYBR Green I. (3) The dependence of fluorescence with time for fixed or unfixed infected RBC in SYBR Green I-basic at 1× dilution was evaluated over 30 min by time-curve analysis of P. falciparum cultures. The frequency distribution of fluorescence was similar after staining for 5 min or longer with SYBR Green I-basic, although fluorescence could be detected within seconds after addition of the fluorescent stain to fixed or unfixed infected RBC. SYBR Green I in Tris-saline at pH 6.8 and PBS resulted in inadequate signals for infected RBC stained for less than 30 min, and in deformation and hemolysis of infected RBC. Accordingly, the optimized protocol was fixation of infected/uninfected RBC by the addition of 1% paraformaldehyde combined with Alsever's solution, and staining the fixed RBC, at measure by flow cytometry, by adding into SYBR Green I-basic solution containing SYBR Green I at x1 dilution for 5 min.

Visualization of infected RBC populations with SYBR Green I-basic
Infected RBC populations were stained as described above. All developmental stages were clearly stained with SYBR Green I with no autofluorescence of RBC (Figure 8). Infected RBC were stained with SYBR Green I-basic by adding 8 x 10 5 cells into 1 ml staining solution, followed by analysis with flow cytometry. SYBR Green I-basic provided brilliant resolution of infected versus uninfected RBC and permitted visualization of infected RBC populations with high accuracy. Infected RBC were located as three clusters in twoparameter dot-plot presentations of infected/uninfected RBC from P. falciparum cultures: (1) cluster 1 (C1) contained predominantly ring forms with low DNA content (low fluorescence intensity); (2) cluster 2 (C2) contained predominantly late trophozoites and young schizonts with moderate DNA content (moderate fluorescence intensity); and (3) cluster 3 (C3) contained late schizonts and segmenters with high DNA content (high fluorescence intensity) (Figures 9a, b).
In regular cultures, schizonts burst spontaneously to release free merozoites, which then enter new RBC and increase the number of infected RBC. While the majority of released free merozoites remain in culture for a while, the merozoites after culture indicate the completion of schizogony. Parasites were synchronized at the ring form stage and cultured over 45 h under the pressure of various concentrations of the anti-schizogony drug chloroquine, to confirm the accuracy of counting the number of merozoites by flow cytometry. Merozoites released from mature schizonts were counted clearly and sensitively by flow cytometry (Figures 10a,b).

Differing effects of NEFA and phospholipids on intraerythrocytic growth of P. falciparum in serum-free medium
The efficacies of NEFA in sustaining general growth of P. falciparum varied markedly, depending on the type, total amount, and combinations. Certain structural characteristics of NEFA, such as carbon-chain length, degree and position of unsaturation, and isomerism were important. However, the mechanisms responsible for the different abilities of the various NEFA in the presence or absence of phospholipids, and of specific proteins such as bovine and human albumin for promoting parasite growth are unknown. Subsequent experiments therefore investigated the distinct effects of various NEFA on each developmental stage of P. falciparum, including schizogony, merozoite formation, and reinvasion of RBC, to provide clues to the mechanisms underlying the growth-promoting properties of NEFA.

Four typical growth patterns
To assess the effects of NEFA on each developmental stage of the parasite (schizogony, merozoite formation, and reinvasion of RBC), synchronized P. falciparum were cultured in the presence of phospholipids and bovine albumin, further supplemented with one or two NEFA. The distribution of the parasites among the different developmental stages was determined using flow cytometry with SYBR Green I-basic at 25 and 45 h during the first cycle of growth (Izumiyama et al., 2009). Late schizonts at 25 h (schizont-25h), released merozoites at 45 h (released merozoite-45h), new ring forms at 45 h (ring form-45h), and www.intechopen.com parasitemia at 45 h (parasitemia-45h) were compared between parasites grown under test conditions and those grown in complete medium GFSRPMI. Different types and combinations of NEFA exerted markedly distinct effects on parasite growth in the presence/absence of phospholipids. Four typical growth patterns were defined: no inhibition (comparable to growth in complete medium); and three rate-determining steps in growth including suppressed schizogony (SS); suppressed formation of merozoites (SMF); and inhibited invasion of merozoites into new RBC (IMI)/formation of incomplete merozoites ( Figure 11). Fig. 11. Representative modification of growth of P. falciparum cultured synchronously in the presence of various growth promoters, indicating comparable growth, SS, SMF, and IMI. Each developmental stage was compared with complete growth in GFSRPMI (control): ring forms (-○-), late schizonts (…▲…), parasitemia (-X-), and released merozoites (closed bars). Parasites at the ring stage (adjusted to 5.0% parasitemia) were maintained in different culture media.

Growth-rate-determining steps in development of P. falciparum cultured in various growth factors
All stages of the parasite cultured in medium supplemented with NEFA (C18:1-cis-9 plus C16:0) in the presence of phospholipids were comparable to those grown in GFSRPMI. Medium containing C18:1-cis-9 and C12:0 caused parasites to accumulate in clusters of ring forms, by an SS effect. SS was also observed in the presence of C16:0 alone, C18:2 plus C16:0, C20:4 plus C16:0, or C18:1-trans-9 plus C16:0. Partial SS (less suppressed) was detected when the mixture of C18:1-cis-13 plus C16:0 was added. C18:1-cis-9 alone and C18:1-cis-9 plus C22:0 suppressed the progression of parasites to merozoites following schizont formation, by an SMF effect. SMF was also observed in parasites cultured in C18:1-cis-9 plus C16:0 in the absence of phospholipids, indicating that exogenous phospholipids were crucial for the development of complete merozoites. Adding C18:1-cis-13 plus C16:0 or C16:1 plus C16:0 to the media caused accumulation of the merozoites released from mature schizonts, but the merozoites did not invade new RBC, by the IMI effect. Partial IMI (less inhibited) was detected when C18:1-cis-9 plus C14:0, C18:1-cis-6 plus C16:0, C18:1-cis-11 plus C16:0, or C18:1-cis-9 plus C18:0 were added. Any effects on steps that governed parasite growth rate disrupted the cyclic behavior of the parasite, and reduced parasitemia at 45 h culture. These results indicate that different NEFA exert distinct roles in parasite development by arresting development at different stages ( Figure 12).
NEFA Growth-rate determining step Growth level Ring ---Trophozoite ---Schizont ---Merozoite ---New ring C18:1-cis-9 SMF  Fig. 12. Growth-rate-determining step and growth level in development of P. falciparum cultured in the presence of NEFA alone or in combination. Growth of the parasite was examined in the presence of optimal phospholipids, except for C18:1-cis-9 plus C16:0 in the absence of phospholipids (*) and GFSRPMI. Bovine albumin was added to the medium, except for GFSRPMI.

Ring forms
Normal Abnormal Fig. 13. Abnormal parasites (ring forms and schizonts) grown in non-optimal culture media.

Conclusions and future perspectives
In an attempt to elucidate the mechanisms responsible for growth of P. falciparum, growthpromoting factors were identified and a CDM suitable for complete growth of the parasite was established. The CDM consists of paired NEFA, phospholipids with specific fatty acid moieties, and specific proteins dissolved in basal medium RPMI1640 supplemented with hypoxanthine. The most effective combination of NEFA was C18:1-cis-9 and C16:0. The best phospholipid crucial for serum-free culture medium supplemented with NEFA was PC-di18:1 at concentrations of 80-320 µg/ml. A simple protocol for flow cytometry with SYBR Green I was developed and used to analyze the various developmental stages of P. falciparum. Different stages of the parasite in RBC and released merozoites were quantified using this flow cytometry protocol. These techniques were applied to investigate the distinct roles of the identified growth-promoting factors in the development of the parasite, demonstrating that different combinations of NEFA and phospholipids exerted distinct roles in the growth of P. falciparum by sustaining development at different stages.
These findings can be usefully applied in diverse aspects malaria research, including drug resistance, vaccine development, genetics, parasite biochemistry, and studies of the relationship between the parasites and the host RBC. Culture in CDM produces similar results to those using the original culture method with human serum (Trager & Jensen, 1997), with the added advantage of avoiding the adverse effects caused by human serum. In particular, the methods reported here will allow the components crucial to each developmental stage of the parasite to be established. We have already performed a largewww.intechopen.com scale analysis of the genome-wide gene expression in P. falciparum cultured in the various CDM established here, to investigate the relationship between gene expression regulation and parasite development at different stages. This comprehensive analysis identified a number of genes possibly involved in arresting parasite development. These results will be reported elsewhere. Interactions between growth-promoting factors and parasite components can provide critical clues to the understanding of the general biology of P. falciparum, and will provide the foundations for future drug and vaccine development efforts aimed at eradicating this disease.