Improvement of culture conditions for long-term in vitro culture of Plasmodium vivax

Background The study of the biology, transmission and pathogenesis of Plasmodium vivax is hindered due to the lack of a robustly propagating, continuous culture of this parasite. The current culture system for P. vivax parasites still suffered from consistency and difficulties in long-term maintenance of parasites in culture and for providing sufficient biological materials for studying parasite biology. Therefore, further improvement of culture conditions for P. vivax is needed. Methods Clinical samples were collected from patients diagnosed with P. vivax in western Thailand. Leukocyte-depleted P. vivax infected blood samples were cultured in a modified McCoy’s 5A medium at 5% haematocrit under hypoxic condition (5% O2, 5% CO2, and 90% N2). Reticulocytes purified from adult peripheral blood were added daily to maintain 4% reticulocytes. Parasites were detected by microscopic examination of Giemsa-stained smears and molecular methods. Results The effects of culture variables were first analysed in order to improve the culture conditions for P. vivax. Through analysis of the sources of host reticulocytes and nutrients of culture medium, the culture conditions better supporting in vitro growth and maturation of the parasites were identified. Using this system, three of 30 isolates could be maintained in vitro for over 26 months albeit parasite density is low. Conclusions Based on the analysis of different culture variables, an improved and feasible protocol for continuous culture of P. vivax was developed. Electronic supplementary material The online version of this article (doi:10.1186/s12936-015-0815-z) contains supplementary material, which is available to authorized users.


Background
Plasmodium vivax is considered to be the most widely distributed human malaria parasite in tropical and temperate countries [1]. Study of the biology of P. vivax is hindered due to the lack of a robustly propagating, continuous culture of this parasite. Several attempts have been made to grow P. vivax in vitro; some successfully established conditions for short-term culture [2][3][4], whereas others improved culture conditions that potentially support longer-term culture of this parasite [5][6][7]. However, the current lack of a continuous culture system for P. vivax requires constant accesses to freshly collected or cryopreserved parasites. Therefore, improvement of in vitro culture methods for P. vivax is needed to enhance study of its unique biology, especially underlying mechanisms of transmission and relapse, to develop better tools for elimination [8][9][10].
In this study, the effects of culture variables were further analysed in order to improve longer term in vitro culture conditions for P. vivax. Through empiric evaluation of the sources of the host reticulocytes and nutrients Open Access *Correspondence: jetsumon.pra@mahidol.edu 1 Mahidol Vivax Research Unit, Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand Full list of author information is available at the end of the article of culture medium, the culture conditions better supporting in vitro growth and maturation of the parasites were identified. Based on the optimized conditions, a more simplified and feasible protocol for continuous culture of P. vivax was developed.

Collection of Plasmodium vivax-infected blood
Clinical isolates of P. vivax were collected from symptomatic patients attending four malaria clinics in Saiyok Distict, Kanchachaburi Province, Thailand. The protocol was approved by the Ethical Review Committee of the Faculty of Tropical Medicine, Mahidol University (MUTM 2011-040-03). After informed consent was obtained, 20 mL of blood were drawn into a 50 mL tube containing lithium-heparin. Infected blood was washed once with RPMI 1640. Leukocytes were removed by passing the infected blood at 50% haematocrit through preequilibrated Plasmodipur ® filter (Europroxima). Thick and thin blood smears were prepared from 1 µL of packed cells before and after passing through Plasmodipur filter. After drying, thin smears were fixed with methanol and stained with 10% Giemsa for 10 min. Parasite density was determined from Giemsa-stained thick smear.

Reticulocyte preparation
Peripheral blood (type O) from healthy donors collected in CPDA1 was obtained from the Blood Bank Unit. Cord blood samples were obtained from the Rajavithi Hospital in Bangkok. The protocol for cord blood collection was approved by the Ethical Review Committee of the Rajavithi Hospital (MUTM2011-030-01). Leukocytes were removed by passing the whole blood through PALL ® leukocyte reduction filter (RC2VAE, Haemonetics Corporation). Leukocyte-depleted blood was centrifuged at 1,000×g for 10 min to remove plasma. Packed blood cells were washed twice with RPMI 1640 incomplete medium (RIM) (RPMI 1640, HEPES 5.96 g/L, NaHCO 3 2.0 g/L, d-glucose 1.0 g/L and gentamycin 40 mg/mL) and centrifuged at 1,000×g for 10 min then resuspended to 20% haematocrit in RIM. Nycodenz was diluted to 19% with a KCl buffer (115 mM KCl, 20 mM HEPES, 1 mM MgCl 2 , 1 mM NaH 2 PO 4 , 10 mM d-glucose, 0.5 mM EGTA, and 12 mM NaCl, pH 7.4). Diluted blood was overlaid on 19% Nycodenz and centrifuged at 3,000×g for 30 min. Reticulocyte-enriched fraction was collected from the interface and washed three times with RIM by centrifugation at 1,000×g for 10 min. Enriched reticulocytes were resuspended to 50% haematocrit with RIM. Enriched reticulocytes were kept at 4°C for a maximum of 2 weeks from the date of blood collection to avoid the loss of erythrocyte surface markers. Enriched reticulocytes were stained with New Methylene Blue (Retic chex) for 15 min before making thin smears. The percentage of reticulocytes was determined by counting the number of reticulin-containing cells from 5,000 cells. Cells with two or more dots of reticulin were considered reticulocytes.

Isolation of CD34 + stem cells
Packed cells of buffy coat from adult donors were obtained from the Australian Red Cross Blood Services (HREC 13/04) and used for the isolation of CD34 + stem cells. The packed cells were diluted with an equal volume of Hank's Balanced Salt Solution/Acid Citrate Dextrose (HBSS/ACD), overlaid on 20 mL of Histopaque, and centrifuged at 400×g for 30 min. The purified stem cell fraction was collected and washed three times with RIM. After treatment with 5 mL of red blood cell lysis buffer for 12 min at room temperature, CD34 + containing white blood cell fraction was collected. CD34 + cells were isolated by using the MiniMacs Direct CD34 + Progenitor Cell Isolation kit (Miltenyi Biotech 130-046-703) according to the manufacturer's instructions. Approximately 5 × 10 8 mononuclear cells with a yield of up to 5 × 10 5 CD34 + cells could be obtained from one Buffy pack (50-60 mL).

Haematopoietic stem cell culture system
The haematopoietic stem cell culture system was composed of three phases: (1) commitment to the erythroid lineage phase, (2) proliferation phase, and (3) differentiation and enucleating phase. The purified CD34 + cells were cultured in IMDM culture medium containing 10 µg/mL insulin, 5% human AB serum, and 1% Pen/ Strep at a concentration of 0.5-2 × 10 6 cells/mL for 18-19 days with daily change of the culture medium. The first phase medium contained IL-3 (1 ng/mL), SCF (40 ng/mL), transferrin (200 µg/mL) and EPO (3 U/ mL). The second phase medium contained SCF, EPO and transferrin (1,000 µg/mL). The third phase medium contained EPO and transferrin (1,000 µg/mL). After 19 days of culture, the original cell population has amplified several thousand times to yield a relatively pure reticulocyte population. In order to obtain pure and enucleated reticulocytes, the culture was filtered using leukocyte reduction filters (WBF3, Haemonetics Corporation) to remove the remaining nucleated cells and the nucleus debris. After purification, reticulocytes were stored in saline-anenine-glucose-mannitol for up to 14 days at 4°C.

Long-term in vitro culture of Plasmodium vivax
Leukocyte-depleted P. vivax-infected blood was cultured in 24-well plates with MMM complete medium at 5% haematocrit (500 µL culture volume). Parasite culture was incubated at 37°C in a hypoxic environment (5% O 2 , 5% CO 2 and 90% N 2 ). Culture medium was changed daily. Reticulocytes were successively added to the culture in order to maintain 4% reticulocytes. Giemsa-stained thick and thin blood smears were prepared from 1 µL packed cells to monitor parasites growth.

In vitro invasion assay
Leukocyte-depleted P. vivax-infected blood was incubated with McCoy's 5A medium supplemented with 25% heat inactivated human AB serum (standard McCoy's 5A complete medium) in T75 cm 2 tissue culture flask. Parasites were cultured as described above until reaching the schizont stage (~18-24 h). Parasite culture was washed once with RIM and centrifuged at 1,000×g for 10 min. Packed infected blood was diluted to 20% haematocrit with RIM and overlaid on 45% Percoll (diluted in 1 × PBS, pH 7.4). The schizont-enriched fraction was collected after centrifugation at 1,200×g for 20 min from the interface and washed three times with RIM by centrifugation at 500×g for 10 min. To optimize the reticulocyte content in the culture, enriched schizonts were incubated with 1, 4, or 8% of reticulocytes obtained from cord blood at a final parasitaemia of 0.01%. The schizont-reticulocyte mixture was cultured in standard McCoy's 5A complete medium in 24-well plates (5% haematocrit; 500 µL culture volume) and incubated at 37°C under hypoxic conditions. After 18 h, ring-stage parasites were observed microscopically after Giemsa staining. The proportion of ring-stage parasites was determined from 5,000 cells.

Mosquito membrane feeding assay
Membrane feeding assay was performed according to a published protocol [11]. Parasite culture was transferred to 1.5 mL tubes and spun at 1,000×g for 3 min. The culture supernatant was removed and the packed infected blood was washed once with warmed RIM. The packed infected blood was resuspended to 50% haematocrit with warmed human AB serum and offered to 50 female Anopheles dirus mosquitoes for 30 min. Engorged mosquitoes were collected and maintained on 10% sugar. Seven days after membrane feeding, mosquitoes were dissected to check the presence of oocysts on the midguts.

Nested-PCR
Nested-PCR was performed to confirm the presence of P. vivax parasites in cultures. Genomic DNA was extracted from 100 µL packed cells with QIAamp ® DNA Mini Kit (QIAGEN) according to the manufacture's protocol. Three primers were used in nested-PCR: P1F forward primer 5′-ACGATCAGATACCGTCGTAATCTT-3′, P2R reverse primer 5′-GAACCCAAAGACTTTGATTTCTC AT-3′, and VR reverse primer 5′-CAATCTAAGA ATAAACTCCGAGAGGAAA-3′. The 18s rDNA was first amplified by using P1F and P2R primers and P. vivax was confirmed subsequently by using vivax-specific P1F and VR primers.

Optimization of proportion of reticulocytes in culture
Because reticulocytes are required for P. vivax invasion, the affect of the proportion of reticulocytes in the in vitro culture was first tested. Using reticulocytes purified from cord blood (RC), the optimization of the proportions of reticulocytes for the invasion of P. vivax merozoites was performed with at least nine clinical isolates of P. vivax. Purified schizonts (2.5 × 10 4 ) were incubated with 1, 4, and 8% (v/v) RC to give a final of parasitaemia of 0.01%. Enumeration of ring-stage parasites 18 h later revealed that with 1% RC the culture contained ~0.02% ring-stage parasitaemia (Fig. 1). An increase of RC to 4% resulted in ~2-fold increase of ring-stage parasitaemia. However, a further increase of RC to 8% did not lead to a significant increase in ring parasitaemia (Fig. 1). Thus a 4% reticulocyte concentration was selected for further optimization of vivax culture conditions.
The influence of sources of reticulocytes to the invasion and maturation of vivax parasites was determined. For comparison, reticulocytes were obtained from two additional sources: purified from peripheral blood (RP), and cultured from haematopoietic stem cells (cRBC). All reticulocytes were seeded at 4% and cultured with the same clinical isolates of P. vivax. Invasion and maturation of the parasites in these cultures were monitored at 18 and 48 h, respectively. The published criteria for developmental assessment of P. vivax was used to differential stage of the parasite [2]. Ring-stage parasitaemias were not significantly (one way ANOVA, F = 0.192, P = 0.827) different among the three sources of reticulocytes. In addition, the morphology of the parasites residing in RP, RC and cRBC at 18 h did not show obvious differences (Fig. 2a). However, at 48 h parasites in RP showed better maturation than those in RC and cRBC (Fig. 2b). Nonetheless, parasite growth in all three sources of reticulocytes was slightly delayed and none of the parasite isolates tested could complete blood stage schizogony within 48 h. Further, the long-term effect of different sources of reticulocytes on parasite culture was studied with 11 P. vivax clinical isolates (Fig. 3). Reticulocytes were added daily to maintain 4% and the parasite density was monitored daily for 7 days. Overall, parasite densities among the three reticulocyte sources were not significantly different (one way ANOVA, F value = 0.024, P = 0.976). Yet, depending on the parasite isolates, some grew better in one source of reticulocytes than others (see Additional file 1). Although RC reticulocytes contain fetal haemoglobin, which was reported to have an inhibitory effect on parasite growth [13], RC better supported parasite maturation as well as gametocyte production in four of the tested parasite isolates (see Additional file 2). Given that RP could support parasite growth for most of the tested isolates, and are a more reliable source of reticulocytes, RP reticulocytes were chosen for subsequent studies to improve long-term culture of P. vivax.

Analysis of culture medium
Next we compared three types of culture medium, McCoy's 5A, RPMI 1640 and Waymouth's. Human AB serum from only Duffy positive blood donors were used in all experiments. The percentage of serum varied between 25 and 50% for each culture medium, yielding six combinations: McCoy +25% and McCoy +50% serum, RPMI +25% and RPMI +50% serum, and Waymouth's +25% and Waymouth's +50% serum. The relative effect of each culture medium-serum combination on parasite growth of seven isolates was determined by counting parasite density in Giemsa-stained thick smears every other day for 9 days (Fig. 4). For RPMI 1640 and Waymouth's medium, an increase from 25 to 50% serum did not result in discernable differences in parasite densities. However, the effect of serum content in the culture medium was more pronounced for McCoy's 5A medium. During the first few days of the culture, P. vivax parasite cultured in McCoy +50% AB had higher parasite density than in McCoy +25% AB. Yet, during the later days of culture, this trend was reversed with McCoy +25% AB having the higher parasite density. Overall, McCoy +25% showed greater parasite densities and was, therefore, selected as the standard culture medium in this system.

Modification of the culture medium
Whereas the standard McCoy's 5A medium has been widely used for P. vivax culture, delayed maturation of parasites in the second cycle after reinvasion of reticulocytes was observed, which reflected a delayed death type of phenotype. In most cases there was a noticeable lengthening of the blood-stage asexual development cycle, since the parasites were not able to complete schizogony within 48 h. To determine whether the inclusion of additional nutrients could improve parasite growth during in vitro culture, McCoy's 5A complete medium was supplemented with ascorbic acid 0.0116 g/L (ASC), cholesterol 750 mg/L (CHO), reduced glutathione 2.2 mM (GSH), hypoxanthine 360 µM (HT), orotic acid 360 µM (ORA), inosine 5′-monophosphate 360 µM (IMP) and d-glucose 4 g/L (GLU). Each additive was evaluated individually and then in combination for the ability to improve parasite growth over a 7-day period in culture, although there was not a significant difference in parasitaemia among the media tested (Fig. 5). In contrast, a combination of all additives appeared to promote in vitro growth of the parasite, since only P. vivax parasites cultured in the Modified McCoy's 5A medium (MMM) could develop to late schizont stage within 48 h (Fig. 6). Moreover no parasites persisted for a month in cultures with McCoy's 5A complete medium containing no or single nutrient additives, but parasites were still detected in cultures with MMM (2-10 parasites/1 µL packed cells) (Fig. 7a). Therefore, MMM was selected as the medium for long-term culture of P. vivax parasite.

Long-term culture of clinical isolates of P. vivax
Thirty P. vivax isolates were further tested in MMM in 24-well plates (500 µL/well) at 5% haematocrit with daily change of medium with RP added daily to maintain 4% reticulocytes. The parasites were cultured in Fig. 2 Maturation of Plasmodium vivax cultured in different sources of reticulocytes. The invasion assay was performed. The purified schizonts were cultured with 4% of RP, RC, and cRBC to give a final concentration of 0.01% schizonts. The maturation of the parasites was monitored at 18 and 48 h. The developmental assessment of the parasite followed the published guideline [2]. a Parasites resided RP, RC, and cRBC at 18 h of cultured showed similar morphology of large ring stage and irregular shape with polymorphic cytoplasm. The parasite occupied 1/3 area of the erythrocyte. b The maturation of the parasites were continued to follow at 48 h. Parasites residing RP showed better maturation compared to RC and cRBC. The parasites reside RP developed to early trophozoite with irregular cytoplasm and light brown pigment. The parasite occupied ½ of enlarged erythrocyte. While the parasites resided RC, and cRBC showed similar morphology of large ring stage parasite and irregular shape with polymorphic cytoplasm. The parasite occupied 1/3 area of the erythrocyte.

Fig. 3
Parasite density of Plasmodium vivax cultured with different sources of reticulocytes in 7 days. a Parasite density during the 7 day cultured period; b Magnification of the graph from day 4 to day 7. Reticulocytes purified from peripheral blood (RP), cord blood (RC) and haematopoietic stem cell (cRBC) were used to cultured fresh isolates of P. vivax (10 isolates). Reticulocytes were added to the cultures daily at a final 4% and the parasite density was determined for 7 days from Giemsa-stained thick smears. Each box represents average parasite density from 7 days obtained from 10 P. vivax isolates. Fig. 4 Parasite density of Plasmodium vivax cultured with different culture media in 9 days. a Parasite density from day 3 to day 9. b Parasite density during the 9 day culture period. P. vivax parasites (7 isolates) were cultured in McCoy +25%, McCoy +50%, RPMI +25%, RPMI +50%, Waymouth's +25% and Waymouth's +50% AB serum. The parasite density was determined for 7 days from Giemsa-stained thick smears. Each box represents the parasite density from 7 days obtained from 7 P. vivax isolates.
≥2 wells at the beginning for each isolates and parasite growth was monitored by Giemsa-stained thick and thin smears (1 µL packed cells) once a week. Parasite cultures were split once the haematocrit was higher than 5%. Most isolates shared a similar growth pattern with fluctuations of parasitaemia (Fig. 7b).
When enough cultured packed cell volume was available, 100 µL of packed cells from the cultures were used for parasite detection by nested-PCR. Considering the growth pattern of the parasite cultured mentioned above in this system, the culture wells that were negative by nested-PCR for four consecutive times were discarded. The RT-QMAL was also performed for all of the cultures. Three out of 30 isolates were positive by this method (see Additional file 3). The longest culture period was over 26 months.

Screening of P. vivax isolates producing infective gametocytes
In order to determine whether gametocytes could form and mature using the MMM culture medium and protocol, aliquots of day 7 cultures were fed to An. dirus mosquitoes. Assuming a 3-day longevity of P. vivax gametocytes [14], the 7-day culture period was selected to make sure that the available parasites were not too low and gametocytes present in the culture were not the ones present in the original samples. After mosquito feeding, two out of 30 isolates were infective to mosquitoes (Fig. 8).

Discussion
During the past decades, many efforts have been undertaken to establish a continuous culture system for P. vivax. Several established systems were shown to be able to support short-term growth with some potentially supporting longer-term maintenance of the parasites [2][3][4][5][6][7]. One of the major challenges in the development of in vitro culture of P. vivax lies in its preference to invade reticulocytes [15]. Provision of sufficient reticulocytes is a critical factor for the success of continuous culture of this parasite. In addition, P. vivax also invades Duffy positive (Fy antigen) reticulocytes [16,17]. While testing Fy antigen is not performed by most of the blood banks, in-house testing could be done by using a simple agglutination test with commercially available reagents (orthoclinical diagnostic) or flow cytometry using Fy 6 antibody [18].
The level of available reticulocytes in P. vivax in vitro culture should be of particular importance since too many reticulocytes may dilute the parasite in culture, whereas too few may limit host cells for parasites to invade. In this study, the optimum level of the reticulocytes for parasite invasion was ~4%, whereas further increases did not result in any significant increase in parasitaemia. Various studies have used different sources of reticulocytes, including those from peripheral blood [5], haemochromatosis blood [6], and cord blood [4,19]. The advantage of using reticulocytes from haemochromatosis blood is that they contain adult haemoglobin and the starting proportion of reticulocytes is high. However, since haemochromatosis occurs mainly in Caucasians, this type of reticulocytes for P. vivax culture is normally not accessible to researchers outside Europe or Americas. Another source of reticulocytes is cord blood, which contains a high percentage of young reticulocytes. Although the purified reticulocytes from cord blood can be cryopreserved in Glycerolyte 57 for later uses [19], they contain high-level fetal haemoglobin that could affect parasite development. A major breakthrough is the use of haematopoietic stem cell culture technology to generate highly pure young reticulocytes for P. vivax culture [7,20,21]. Yet, major limitations lie in the low reticulocyte yield and high cost of this method. Comparison of reticulocytes obtained from three different sources did not show significant differences in their abilities to support parasite invasion and maturation. Nevertheless, parasites grown in reticulocytes from peripheral blood seemed to develop better than those grown in reticulocytes from the other two sources.
A variety of reticulocyte purification procedures have been used to enrich reticulocytes for adding to P. vivax cultures. Differential centrifugation in autonomous plasma could purify reticulocytes for up to 20% from haemochromatosis blood and supported parasite growth without requiring additional reagents [6]. However, the method is labour-intensive and needs an ultracentrifuge. Several density gradient centrifugation methods have also been developed. The Percoll/Renografin-60 method could enrich reticulocytes up to 15% from adult peripheral blood, but Renografin-60 has toxic effect on the parasite [5]. Without Renografin-60, 71% Percoll centrifugation also allowed enrichment of reticulocytes from cord blood without toxicity to the parasite [9,19], but this method seems less effective when using peripheral blood. A recently developed protocol for reticulocyte enrichment by using aqueous multiphase systems (AMPS) could obtain highly pure reticulocyte, but the yield was very low [22].
In this study, a simple and reproducible method for purifying reticulocytes from adult blood using 19% Nycodenz centrifugation was employed. Using this method, reticulocytes up to ~80% reticulocytes could be highly enriched from adult peripheral blood (with a normal range of 0.5-2% reticulocytes) to yield 0.5-2 mL packed cells from 30 mL of packed blood cells (Fig. 9). With this method, it requires only 1-2 reticulocyte purifications per week to maintain P. vivax cultures from 30 isolates with at least two wells per isolate. More importantly, reticulocytes purified by this method could support parasite invasion in vitro and in vivo [23] without any noticeable toxic effect on parasites. This method alleviates the difficulty in preparing reticulocytes during long-term P. vivax culture. Fig. 8 Oocysts on mosquito midguts stained with mercurochrome. Parasite cultures from 30 isolates were fed to female Anopheles dirus mosquitoes on day 7 of parasite culture using membrane feeding. Two out of 30 isolates were infective to mosquitoes with oocysts.
Regardless of the source of reticulocytes used for in vitro culture, delayed parasite maturation was observed. To determine whether this delayed parasite maturation was due to the lack of certain nutrients needed for parasite growth, different culture media and specific nutrient additives were evaluated to improve parasite growth. A combination of additives to McCoy's 5A medium (MMM) was determined to support improved parasite growth and efficiency of schizont maturation better than other media during long-term culture.

Conclusions
In summary, this study compared the effects of different culture variables, both of host origin and culture environments, in order to identify the optimum conditions for long-term culture of P. vivax. This is a further improvement of conditions for long-term culture of P. vivax. By using this system, three of 30 isolates could be maintained in vitro for with the longest culture period over 26 months, albeit parasite density was low. Future efforts should be directed towards increasing parasite density in culture in order to provide sufficient materials for studying parasite biology.