Background

Continuous in vitro Plasmodium falciparum culture (standardized in the 1970s) [1,2,3,4] has been an indispensable tool for understanding the parasite’s life cycle and identifying most proteins involved in erythrocyte invasion, some of which have been tested as vaccine components at clinical level [1,2,3,4,5,6]. Developing a methodology enabling the continuous growth and propagation of Plasmodium vivax (P. vivax being the second most important species causing malaria in humans) has thus become a challenge for several research groups studying this parasite [7,8,9,10,11,12,13].

Unfortunately, maintaining a continuous culture of P. vivax in vitro is still difficult, despite different aspects having been studied and modified, i.e. different culture media [13, 14], parasite [9, 15] and reticulocyte [7, 10] sources, added to the different methods for obtaining and enriching invasion target cells [16]. It has only been possible to maintain a culture in vitro for up to 26 months to date, having < 0.1% parasitaemia [14], which might be due to merozoites (Mrz) losing their ability to re-invade new host cells [11, 13, 14, 17, 18].

The forgoing has discouraged research orientated towards knowing in detail the mechanism used by P. vivax for specifically invading reticulocytes; consequently, there has been a delay in identifying new molecules, the function they fulfil and their antigenic and immunogenic capability; such information is essential for selecting specific proteins to be included when developing parasite control methods.

This work has been aimed at reviewing aspects which have been taken into account for standardizing an in vitro P. vivax culture and proposes some alternatives which could be considered.

The current state of Plasmodium vivax biology

Plasmodium vivax is a parasite causing malaria in humans; it has been included on the international health agenda regarding its early eradication, mainly due to the high morbidity rates it causes and its wide geographical distribution [6]. This parasite species displays particular biological characteristics, such as hypnozoite development in the liver and rapid gametocyte formation. Interestingly, the parasite exclusively infects immature erythrocytes (reticulocytes), representing just 1–2% of total red blood cells (RBC) from adult human peripheral blood. These cells are fragile, have rapid maturation and complex procedures are required for obtaining enriched samples, hence maintaining a P. vivax continuous culture in vitro is extremely difficult [16].

The absence of an in vitro culture in P. vivax could be considered as “the spoke in the wheel” which has caused a considerable delay (between 5 and 10 years) in executing certain types of studies, such as omic sciences, invasion inhibition and determining adhesin-type ligands, epitopes and antigens [19,20,21], i.e. compared to those for P. falciparum [22,23,24]. In fact, more than 50 proteins involved in P. falciparum binding to and invading target cells have now been described as well as some receptors for them [25,26,27,28]. By contrast, only 23 proteins associated with P. vivax invasion of reticulocytes have been characterized (using parasites from patients [29,30,31] and infected animals’ samples [32,33,34,35]) and few receptors have been studied (Fig. 1). The proteins characterized to date have been tryptophan rich antigens (PvTRAg26.3, PvTRAg33.5, PvTRAg34, PvTRAg35.2, PvTRAg36 (band 3 as receptor) [36], PvTRAg36.6, PvTRAg38 (basigin [37] and band 3 [38] as receptors), PvTRAg40, PvTRAg69.4, PvTRAg74 (band 3 as receptor) [36], rhoptry neck protein 5 (RON5) [39], reticulocyte-binding proteins RBP-1a, RBP-1b [40], RBP-2b (CD71 as receptor) [41, 42], erythrocyte binding protein 2 (EBP-2) [43], GPI-anchored micronemal antigen (GAMA) [44], reticulocyte binding surface antigen (RBSA) [45], the Duffy binding protein (DBP) (DARC as receptor) [46, 47], reticulocyte binding protein 1 (RBP-1) [48], merozoite surface protein 1 (MSP-1) (possible receptor, band 3) [49], apical membrane antigen 1 (AMA-1) (chymotrypsin- and neuraminidase-sensitive receptor, GPB?) [50] and rhoptry neck proteins 2 and 4 (RON2 and RON4) [51].

Fig. 1
figure 1

Plasmodium vivax and P. falciparum merozoite host cell adhesion proteins. The figure shows the P. vivax (left-hand side) and P. falciparum (right-hand side) proteins described to date having a binding-related function regarding receptors identified on target cells: reticulocyte and mature erythrocytes, respectively. Mrz proteins’ subcellular localization is indicated. An asterisk indicates those molecules with unknown subcellular localization

Full size image

The small list of characterized ligands and receptors reveals the tremendous challenge faced by researchers considering studying P. vivax in terms of continuous propagation to understand different aspects of the parasite’s basic biology. In view of this and aimed at making significant advances in clinical and basic research regarding the species, several groups have focused on standardizing a continuous in vitro culture system for P. vivax blood stages for which some essential parameters for optimizing parasite growth and development have been determined.

Culturing Plasmodium vivax

The first reports about culturing malarial Plasmodium date from the beginning of the twentieth century, some techniques being more controversial than others (i.e. parasite culture from infected water and keeping parasites alive in milk for several days) [52]. Although the first successful P. vivax in vitro culture was reported in 1912 [53, 54], a base protocol for propagating this parasite species was only established at the end of the 1970s [4, 55, 56]. Since then, one or more of the factors involved in the culture have been modified in various attempts at finding an efficient methodology (Fig. 2). However, it has not been possible to date to maintain a culture, given two main problems: parasitaemia dynamics and the amount of days for maintaining a P. vivax in vitro culture. Although is not clear why P. vivax Mrz in culture lose their ability to re-invade new host cells, the isolate or parasite strain and target cells may have intrinsic characteristics which can influence P. vivax propagation (despite modifications to the culture media). The factors related to maintaining a P. vivax in vitro culture (i.e. culture media, parasite and reticulocyte origin) will therefore be analysed.

Fig. 2
figure 2

Conditions evaluated for culturing Plasmodium vivax in vitro. The figure shows modifications made to culture medium, parasite sources and the target cells which have been used in culturing the P. vivax parasite. Each combination evaluated is described in the culture medium section. For example, 14 indicates RPMI 1640 medium, which contains HEPES, NaHCO3, neomycin, vitamin B12, hypoxanthine, ascorbic acid, MgCl2, biotin, choline and 15% AB+ human serum (box showing typology). The reticulocyte source section lists these cells’ enrichment methodology using the letters a–e. Sections of the graphic enclosed by a red discontinuous line indicate the best target cell sources and/or enrichment methods available so far

Full size image

Culture media and supplements

Several media and supplement combinations have been tested to ensure the conditions and nutrients required for optimum P. vivax growth (Fig. 2). The first reported medias for P. vivax in vitro culture were modified Harvard, RPMI-1640, Waymouth’s and SCMI 612 supplemented media [4, 53, 57] (Fig. 2); it was seen that only SCMI 612 medium improved parasite viability [58] (Table 1). Other research showed that RPMI 1640 medium supplemented with MgCl2 [59], ascorbic acid, hypoxanthine, vitamin B12, choline and biotin [60] improved parasite maturation. However, in later studies in which RPMI 1640 medium was used [12, 56, 61,62,63,64,65], even in mixture with different compounds and salts (MgS04, KH2P04 and CaCl2) and 50% AB+ human serum, there was no improvement in parasitic density, suggesting that the RPMI 1640 media is not appropriate for P. vivax continuous growth and development [65].

Table 1 Studies related to Plasmodium vivax in vitro culture development
Full size table

McCoy’s5A medium has also been routinely used [7, 9,10,11, 13, 14, 17, 18, 66, 67] in combination with various supplements such as d-glucose and l-glutamine, or just with 20% or 25% AB+ human serum [9, 10, 17, 18]. It has been reported that a medium consisting of McCoy’s5A supplemented with HEPES, NaHCO3, d-glucose, gentamycin and 50% AB+ human serum maintains parasite density (10 parasites/µL) during the first 5 days of culture. However, such parasite density can be maintained after 5 days using just media supplemented with 25% AB+ human serum [14]. Two compounds improving parasite development in McCoy’s5A medium have been reported recently: Albumax II [67] and GlutaMAX [15, 68] (Fig. 2, Table 1). GlutaMAX (l-alanyl-l-glutamine dipeptide) did indeed improve parasite viability, growth and development compared to l-glutamine as this compound does not break down to form toxic by-products, such as ammonia, formed by traditional l-glutamine [15]. This highlighted the fact that P. vivax could be very sensitive to the accumulation of waste or toxic products in in vitro conditions.

The use of Dulbecco’s Modified Eagle Medium (DMEM) for P. vivax culture supplemented with l-glutamine, HEPES and hypoxanthine has been reported recently. Parasitaemia was maintained for 233 days and was ended because of bacterial contamination [69]. The fluctuation in parasitaemia using DMEM was similar to that observed when the parasite has been grown in McCoy’s5A medium [14], suggesting that these media (McCoy’s 5A and DMEM) are useful for culturing and maintaining parasite maturation and replication in vitro. Future trials should be conducted with McCoy’s5A or DMEM medium, supplemented with 25% human serum (with Glutamax and Albumax) to evaluate whether parasite density can be maintained and/or increased in culture.

Parasite source

The parasite has been used from two sources for standardizing P. vivax in vitro culture, i.e. isolated from humans and from primates (Fig. 2, Table 1). Regardless of the source, it has been observed that keeping the culture in static conditions improves culture parasitaemia [15, 61] as well as depleting white blood cell amount in reticulocyte samples, as leukocytes’ phagocytic activity against parasites affects their invasion [53, 61]. Likewise, it has been shown that cryopreservation [9, 18] enables maintaining parasite viability and invasive capability when preserved and stored for days [13, 69] or even years [9].

Difficulty related to variation in both longevity and parasitaemia has occurred regarding in vitro culture with parasites obtained from humans. For example, it has been reported that different isolates could be maintained in culture for several days: i.e. from 10 or 30 days [12], from 2 to 8 days or up to 85 days (more than 2 months) [10]. An in vitro culture of 3 P. vivax isolates was recently maintained for more than 1 year (26 months), having ~ 0.01% parasitaemia [14]. Other research has shown that culture parasitaemia can increase almost tenfold when using parasites from isolates which were enriched during ring stage by Percoll gradient [11, 68]. Despite this, the parasite progressively loses its invasion ability, a problem which has not yet been resolved to date. These studies suggested that each P. vivax isolate has its own characteristics related to adaptation to in vitro culture and thus their invasion capability, multiplication rate and parasitaemia are variables which must be considered when standardizing a culture for each of them. This hypothesis can be supported by a study by Russell et al., who evaluated umbilical cord blood (UCB) reticulocyte invasion inhibition using 85 P. vivax clinical isolates. They found that invasion efficiency was constant for each specific isolate but that 85.79% of the total variance depended on isolate type [11]. Heterogeneity concerning human isolates’ invasion efficacy and P. vivax parasitaemia density variation thus makes the methodologies used for culturing the parasite not suitable for studying its biology and further complicates the development of a robust and reliable culture method.

Unlike parasites obtained from humans, primate-adapted P. vivax strains can be used to start in vitro culture anytime, given their availability. This is why some research groups have worked with several P. vivax strains (Fig. 2, Table 1) [56, 57, 61, 62, 70] which were able to adapt to invade erythroid cells in vitro (from humans suffering from haemochromatosis [7, 15]), owl monkey cells [7] and reticulocytes obtained from the maturation of UCB haematopoietic stem cells (HSC) - CD34+ [17], cultures reaching > 0.5% parasitaemia. These results support the notion that monkey-adapted P. vivax strains do not lose their capability to invade, regardless of cell source, and therefore, represent a good alternative for establishing a parasite culture.

Considering invasion efficacy variability in cultures from human sources and the great adaptability of strains in monkeys, it can be suggested that the same parasite strain must be used during attempts at standardization to establish the basic and necessary conditions for maintaining a long-term in vitro culture.

Target cell source

Obtaining reticulocytes for continuous supplementation in culture has been a huge inconvenience since these cells only form 1 to 2% of human peripheral blood, mature quickly, are fragile and have low viability. UCB (containing 6.9–7.9% reticulocytes), peripheral blood from humans or splenectomized monkeys, blood from haemochromatosis patients (14–17% reticulocytes) and HSC (variable reticulocyte percentages) have been used as reticulocyte sources for standardizing an in vitro P. vivax parasite culture [7, 12, 61] (Fig. 2, Table 1). Different techniques such as density gradients (Percoll and Nycodenz), ultra-centrifugation and/or immunomagnetic separation have also been used for obtaining a greater percentage of reticulocytes in culture, Nycodenz being one of the most appropriate compounds as it has had no notable toxic effects on cells [7, 14, 15] (Fig. 2).

Although UCB are a good source of reticulocytes, it has been shown that they do not support the parasite’s full development and are easily lysed [61]. Fetal haemoglobin in such erythroid cells apparently produces an inhibitory effect for P. vivax growth, equivalent to that reported for P. falciparum, which does not grow adequately in erythrocytes containing fetal haemoglobin [71, 72]. The forgoing has been supported since it has been reported that reticulocytes from CD34+ erythroid progenitors (derived from adult peripheral blood or bone marrow) and from adults’ peripheral blood lacking fetal haemoglobin could improve P. vivax invasion [66] regarding maturation and gametocyte production [14]. Whilst these studies have shown that UCB does not seem to be very suitable for standardizing an in vitro P. vivax culture, another study has shown that fetal haemoglobin caused no alteration in parasite growth and up to 0.4% parasitaemia was reached during the first days of culture [11]. Although it is not clear how haemoglobin could alter parasite development, this effect might depend on the reticulocyte’s maturation stages (variability) and the availability of these stages in the UCB source.

Another great concern related to using reticulocytes is their rapid maturation. It has been suggested that these target cells can be frozen to provide a reserve and then used for supplementing a culture when required. Different studies have reported that both fresh reticulocytes and freshly thawed reticulocytes were susceptible to invasion by P. vivax Mrz. Interestingly, such susceptibility did not depend on reticulocyte source since they were obtained from UCB [18], haemochromatosis patients (in which the cells were enriched using Percoll gradient [7, 18] or differential centrifugation [7]) and human cord HSCs [9, 17] (enriched by Percoll density gradient [10]). Notably, cryopreserved cells which were then thawed had up to 70% viability and such percentage remained stable compared to that for fresh samples [9].

According to the literature, haemochromatosis patients have been one of the best reticulocyte (fresh or cryopreserved) sources. These reticulocytes, enriched by differential centrifugation in 20% homologous plasma [7, 15], were easily invaded, able to support both parasite growth and invasion [12] and maintain a stable schizont percentage [7]. Although Percoll gradient has been widely used for enriching reticulocytes obtained from haemochromatosis patients (Fig. 2), two studies have reported that cell viability and stability could be affected by damage to or the loss of some membrane receptors which might be essential for P. vivax invasion [7, 15]. Despite this, one of the drawbacks of this target cell source is that haemochromatosis mainly occurs in Caucasians, a type of reticulocyte not normally accessible for researchers outside Europe or North America [14]. Using HSC-derived reticulocytes could guarantee a more homogenous and standardized cell population which would enable obtaining a high reticulocyte concentration (> 20%) [9], necessary for maintaining P. vivax cultures.

Factors such as culture medium, as well as parasite and reticulocyte sources have been revised and possible modifications which could improve parasite development in vitro have been pointed out. However, target cells must be analysed in depth in relation to their intrinsic characteristics enabling the parasite to invade them.

Reticulocyte receptors: the new molecular keys?

Taking into account that P. vivax Mrz only invade reticulocytes, the next question arises: Which characteristics do reticulocytes have so that P. vivax can only invade this type of cell? Immature reticulocytes in bone marrow contain ribonucleic acid (RNA) and undergo different biochemical, biophysical and metabolic changes during their maturation to normocytes within a period of 72 h [73]. The differences between reticulocytes and normocytes have been studied at molecular level in murine and human models [74,75,76]. It has been found that the main difference between these two cells is the abundance of their receptors, since more than 60% of proteins quantified in immature erythrocytes became reduced (from 2 to 100 times) as they matured to normocytes, whilst around 5% had higher expression levels. Receptors such as transferrin receptor (CD71) on reticulocyte membrane decrease progressively until their total absence in normocytes [77, 78]; this, together with cytoplasmatic RNA (Thiazole Orange stained, TO) concentration, has enabled classifying the reticulocyte population into four groups: Heilmeyer stage I (CD71highTOhigh), Heilmeyer stages II and III (CD71lowTOmed) and Heilmeyer stage IV (CD7−TOlow) [73, 79].

Most P. vivax in vitro culture studies have been restricted to using stage III (the first to emerge from bone marrow), stage IV or mature reticulocyte (CD71−TOlow) populations. Using these two reticulocyte stages and their rapid maturation could provide an explanation for why the parasite loses its infective capability through various replication cycles. This could also explain why less than 1% parasitaemia has only been achieved in most assays performed to date [12, 14]. In line with the forgoing hypothesis, it has been shown that cryopreserved parasite isolates from patients can infect stage I reticulocytes (CD71highTO high) representing only 0.02% in total blood [79]. The infected cells mature rapidly and almost completely lose reticular matter 3 h post-invasion, thereby showing that parasite invasion promotes rapid reticulocyte maturation [79, 80]. Similar results have been found by Shaw-Saliba et al., when evaluating a culture of Sal-I strain parasites adapted in A. lemurinus monkeys with CD71high reticulocytes. As expected, parasites preferentially invaded stage I reticulocytes (CD71highTOhigh) and very few managed to invade stage IV reticulocytes (CD71−TOlow) [15]. Research by Golenda and Udomsangpetch showed that P. vivax development and invasion levels were better using haemochromatosis patients’ blood; these results also support the previous hypothesis [7, 12] as people suffering this type of anaemia produce a larger amount of stage I reticulocytes (CD71highTOhigh) to balance the decrease of erythrocytes in blood flow [81].

These observations lead to another question: Why can P. vivax Mrz only invade the most immature reticulocyte stage? Several studies using different approaches could provide an answer to this question. One such was related to DARC receptor abundance on CD71highTOhigh reticulocytes and conformational changes affecting such receptor enabling the parasite to bind to and invade this type of cell [82]. It has been found that although total DARC protein remains constant throughout reticulocyte maturation there is selective exposure of one DARC amino acid sequence (QLDFEDVWNSSY) by conformational changes before maturation which causes DBP to bind more specifically to CD71high/TOhigh reticulocytes than to other mature reticulocyte or erythrocyte subpopulations [82]. Other studies showing P. vivax proteins’ preference for binding to CD71high reticulocytes [44, 45] and evidence about RBP-2b binding to CD71 membrane receptor have been published very recently [42]. The above highlights the most immature reticulocyte stage (CD71high TOhigh) as the molecular key (receptor) which P. vivax takes advantage of to invade and replicate within cells. This could suggest that using reticulocytes from bone marrow and/or from patients suffering different types of haemolytic anaemia (i.e. haemochromatosis) could be appropriate for maintaining and developing a continuous in vitro culture system involving P. vivax blood stages [83]. However, it would be ethically complicated to work with samples from patients suffering from some type of anaemia, which is why using stage I homogenous reticulocytes (CD71high TOhigh) obtained from HSC could be a viable alternative.

Others challenges to be faced

It has been demonstrated that P. vivax DBP binds more to reticulocytes having the Fya−/Fyb+ phenotype [84], which could be an advantage regarding parasite culture.

It would be expected that supplementing cultures with a CD71highTOhigh enriched reticulocyte population and having such phenotype should maintain parasitaemia, invasion efficiency and a culture for a long time.

A new challenge today concerns the fact that enough evidence has been amassed to indicate that the parasite can also invade Duffy negative cells (Fya-/Fyb-) via an as-yet-unknown alternative invasion route [85]. This finding suggests two possibilities; first, such P. vivax property remains unknown due to this parasite’s sub-microscopic and asymptomatic parasitaemia and secondly this could be a new adaptation phenomenon where ligand-receptor interaction routes different to DBP-DARC are acting as survival strategy for propagating cells having the Fya−/Fyb− phenotype. Studies aimed at ascertaining whether P. vivax target cell invasion route is via the RBP2b-CD71 interaction using Duffy negative phenotype CD71high reticulocytes are in need [42]. Future assays should evaluate whether the aforementioned factors could help standardize a P. vivax culture.

Other important considerations include knowing whether the abundance of receptors (as has been showed for DARC and CD71) or remodelling other proteins during reticulocyte maturation (as has been shown for DARC [82]) can have an impact on the development of a P. vivax in vitro culture.

This review has described different factors affecting P. vivax in vitro culture, ranging from using several strains and isolates to different target cell sources and physicochemical variations. Using the same parasite strain and CD71highTOhigh host cells could be a starting point for removing the spoke in the wheel and advance knowledge regarding P. vivax biology.