INTRODUCTION
Plasmodium vivax is being increasingly recognized as a significant source of morbidity throughout an extremely wide geographic region, with more than 80 million people infected each year and more than 40% of the world's population at risk (Price et al. Reference Price, Tjitra, Guerra, Yeung, White and Anstey2007) (estimated recently at 2·5 billion people at risk (Gething et al. Reference Gething, Elyazar, Moyes, Smith, Battle, Guerra, Patil, Tatem, Howes, Myers, George, Horby, Wertheim, Price, Mueller, Baird and Hay2012)). Epidemiological studies have indicated that individuals living in endemic areas can acquire natural immunity to clinical infections, and intriguingly, at a faster rate than acquisition of protective immunity to Plasmodium falciparum. In this article we will review what is known about the underlying immunology providing clinical protection (both cellular, humoral and memory) and what still remains to be discovered.
EPIDEMIOLOGICAL OBSERVATIONS OF NATURALLY ACQUIRED IMMUNITY TO P. VIVAX
Naturally acquired immunity to P. vivax was first proposed due to epidemiological observations in which children were more likely to be susceptible to clinical disease compared with adults living in the same area (Koch, Reference Koch1900; Taliaferro, Reference Taliaferro and Boyd1949). Three detailed epidemiological studies conducted in the late 1900s demonstrated this effect quite clearly: in Vanuatu, morbidity due to P. vivax peaked in children aged 0–2 years whilst little morbidity was detected beyond the age of 6 years (Maitland et al. Reference Maitland, Williams, Bennett, Newbold, Peto, Viji, Timothy, Clegg, Weatherall and Bowden1996). In the Karen population in western Thailand, the peak incidence of P. vivax was in children aged 0–4 years, decreasing to a plateau after 25 years of age (Luxemburger et al. Reference Luxemburger, Thwai, White, Webster, Kyle, Maelankirri, Chongsuphajaisiddhi and Nosten1996). In Sri Lanka, rather than an association with age, the authors noted a decrease in the measures of malaria morbidity following successive P. vivax infections (Gunewardena et al. Reference Gunewardena, Carter and Mendis1994). Since then, similar observations have again been documented in Sri Lanka (Karunaweera et al. Reference Karunaweera, Carter, Grau and Mendis1998), Vanautu (Kaneko et al. Reference Kaneko, Chaves, Taleo, Kalkoa, Isozumi, Wickremasinghe, Perlmann, Takeo, Tsuboi, Tachibana, Kimura, Bjorkman, Troye-Blomberg, Tanabe and Drakeley2014), Papau New Guinea (PNG) (Michon et al. Reference Michon, Cole-Tobian, Dabod, Schoepflin, Igu, Susapu, Tarongka, Zimmerman, Reeder, Beeson, Schofield, King and Mueller2007), the Solomon Islands (Harris et al. Reference Harris, Sharrock, Bain, Gray, Bobogare, Boaz, Lilley, Krause, Vallely, Johnson, Gatton, Shanks and Cheng2010), Vietnam (Nguyen et al. Reference Nguyen, van den Eede, van Overmeir, Thang, Hung le, D'Alessandro and Erhart2012), Peru (Branch et al. Reference Branch, Casapia, Gamboa, Hernandez, Alava, Roncal, Alvarez, Perez and Gotuzzo2005) and Brazil (Camargo et al. Reference Camargo, Alves and Pereira da Silva1999; Alves et al. Reference Alves, Durlacher, Menezes, Krieger, Silva and Camargo2002; da Silva-Nunes et al. Reference da Silva-Nunes, Codeco, Malafronte, da Silva, Juncansen, Muniz and Ferreira2008; Ladeia-Andrade et al. Reference Ladeia-Andrade, Ferreira, de Carvalho, Curado and Coura2009). Hence, it is also possible to generate naturally acquired immunity in low-transmission regions.
Epidemiological studies have also demonstrated that this acquisition of natural immunity to clinical P. vivax infection actually occurs more rapidly than for P. falciparum. This was demonstrated in a number of the epidemiological studies mentioned above (Gunewardena et al. Reference Gunewardena, Carter and Mendis1994; Luxemburger et al. Reference Luxemburger, Thwai, White, Webster, Kyle, Maelankirri, Chongsuphajaisiddhi and Nosten1996; Maitland et al. Reference Maitland, Williams, Bennett, Newbold, Peto, Viji, Timothy, Clegg, Weatherall and Bowden1996; Michon et al. Reference Michon, Cole-Tobian, Dabod, Schoepflin, Igu, Susapu, Tarongka, Zimmerman, Reeder, Beeson, Schofield, King and Mueller2007), amongst others (Phimpraphi et al. Reference Phimpraphi, Paul, Yimsamran, Puangsa-art, Thanyavanich, Maneeboonyang, Prommongkol, Sornklom, Chaimungkun, Chavez, Blanc, Looareesuwan, Sakuntabhai and Singhasivanon2008; Lin et al. Reference Lin, Kiniboro, Gray, Dobbie, Robinson, Laumaea, Schopflin, Stanisic, Betuela, Blood-Zikursh, Siba, Felger, Schofield, Zimmerman and Mueller2010; Koepfli et al. Reference Koepfli, Colborn, Kiniboro, Lin, Speed, Siba, Felger and Mueller2013), as well as in historical trials wherein syphilis patients were infected with Plasmodium parasites as a curative measure (Boyd, Reference Boyd1947). The difference in speed of acquisition of naturally acquired immunity suggests that the functional immunology is likely quite different between the two species. It has been suggested that this may be due to different biological characteristics (Mueller et al. Reference Mueller, Galinski, Tsuboi, Arevalo-Herrera, Collins and King2013), such as the presence of hypnozoites or the large reliance on the duffy binding protein (DBP) for invasion, or due to the increased force of genetically distinct blood-stage infection seen with P. vivax (Koepfli et al. Reference Koepfli, Colborn, Kiniboro, Lin, Speed, Siba, Felger and Mueller2013).
NATURALLY ACQUIRED CELLULAR IMMUNITY
The role of naturally acquired cellular immunity to P. vivax remains poorly understood. Clearly, cellular responses are instrumental in a functional and effective response to a pathogen, however these have been difficult to define for P. vivax. Whilst induction of CD4+ T cells are critical in providing ‘help’ for B cells to produce antigen-specific antibodies, there is also evidence of cellular immunity to P. vivax being an important part of the immunoregulatory response. Herein, we first review evidence of naturally acquired P. vivax-specific cells and their relationship with protection from clinical infection, discuss the inflammatory response to P. vivax and the role of such cytokines in immunoregulation, and provide evidence of dendritic cell dysfunction during P. vivax infection. Finally, we compare the P. vivax response with the known cellular response induced by P. falciparum.
Induction of P. vivax-specific T and B cells
Whilst there is increasing recognition of the role T cells play during blood-stage infection (Riley et al. Reference Riley, Allen, Wheeler, Blackman, Bennett, Takacs, Schonfeld, Holder and Greenwood1992; Carvalho et al. Reference Carvalho, Fontes and Krettli1999; Pombo et al. Reference Pombo, Lawrence, Hirunpetcharat, Rzepczyk, Bryden, Cloonan, Anderson, Mahakunkijcharoen, Martin, Wilson, Elliott, Elliott, Eisen, Weinberg, Saul and Good2002), we know that T cells are essential to eliminate liver-stages given the parasite hides within the cell (Hoffman et al. Reference Hoffman, Isenbarger, Long, Sedegah, Szarfman, Waters, Hollingdale, van der Meide, Finbloom and Ballou1989; Weiss et al. Reference Weiss, Mellouk, Houghten, Sedegah, Kumar, Good, Berzofsky, Miller and Hoffman1990; Renia et al. Reference Renia, Marussig, Grillot, Pied, Corradin, Miltgen, Del Giudice and Mazier1991, Reference Renia, Grillot, Marussig, Corradin, Miltgen, Lambert, Mazier and Del Giudice1993). It is therefore surprising that assessment of P. vivax–specific T and B cells has largely been conducted for blood-stage proteins. There have been multiple reports of the detection of specific cellular responses to the P. vivax (Pv) antigens tryptophan rich antigens (TRAgs) (Alam et al. Reference Alam, Bora, Mittra, Singh and Sharma2008a, Reference Alam, Bora, Singh and Sharmab; Garg et al. Reference Garg, Chauhan, Singh and Sharma2008; Siddiqui et al. Reference Siddiqui, Bora, Singh, Dash and Sharma2008; Mittra et al. Reference Mittra, Singh and Sharma2010; Zeeshan et al. Reference Zeeshan, Bora and Sharma2013, Reference Zeeshan, Tyagi and Sharma2015) and merozoite surface protein 1 (MSP1) (Soares et al. Reference Soares, Levitus, Souza, Del Portillo and Rodrigues1997; Seth et al. Reference Seth, Bhat, Rao and Biswas2010; Riccio et al. Reference Riccio, Totino, Pratt-Riccio, Ennes-Vidal, Soares, Rodrigues, de Souza, Daniel-Ribeiro and Ferreira-da-Cruz Mde2013), and limited reports for the antigens apical membrane antigen 1 (AMA1) (Bueno et al. Reference Bueno, Morais, Soares, Bouillet, Bruna-Romero, Fontes, Fujiwara and Braga2009; Seth et al. Reference Seth, Bhat, Rao and Biswas2010), MSP9 (Lima-Junior et al. Reference Lima-Junior, Tran, Meyer, Singh, De-Simone, Santos, Daniel-Ribeiro, Moreno, Barnwell, Galinski and Oliveira-Ferreira2008) and DBP region II (DBPII) (Xainli et al. Reference Xainli, Baisor, Kastens, Bockarie, Adams and King2002). These antigens are all expressed during the blood-stage of infection: the large number of PvTRAgs can be expressed on both merozoites and schizonts (Zeeshan et al. Reference Zeeshan, Tyagi and Sharma2015), PvMSP1 is located on the surface of the merozoite (Mueller et al. Reference Mueller, Galinski, Tsuboi, Arevalo-Herrera, Collins and King2013), PvAMA1 is translocated to the micronemes of the merozoite near the end of asexual replication and eventually to the surface prior to invasion of erythrocytes (Bueno et al. Reference Bueno, Morais, Soares, Bouillet, Bruna-Romero, Fontes, Fujiwara and Braga2009), and PvDBP is secreted by the micronemes as the merozoite invades erythrocytes (Adams et al. Reference Adams, Sim, Dolan, Fang, Kaslow and Miller1992). The frequency of positive responders varied in all studies, likely reflecting the protein or peptide used for stimulation, the different assays employed, the transmission intensity in the region and the level of past exposure. The most extensively studied non-blood-stage protein is the circumsporozoite protein (CSP) (Rodrigues et al. Reference Rodrigues, Dutra and Yoshida1991; Herrera et al. Reference Herrera, Escobar, de Plata, Avila, Corradin and Herrera1992; Bilsborough et al. Reference Bilsborough, Carlisle and Good1993; Migot et al. Reference Migot, Millet, Chougnet, Lepers and Deloron1993; Carvalho et al. Reference Carvalho, Fontes, Fernandes, Marinuzzi and Krettli1997; Suphavilai et al. Reference Suphavilai, Looareesuwan and Good2004; Seth et al. Reference Seth, Bhat, Rao and Biswas2010), however, the majority of these studies were conducted more than 10 years ago and hence cellular responses were assessed using immune assays that are now considered outdated. Nevertheless, the frequency of recognition of various P. vivax CSP epitopes varied from less than 20% in Thai individuals (Suphavilai et al. Reference Suphavilai, Looareesuwan and Good2004), to almost 60% in a region of the Colombian Pacific Coast (Herrera et al. Reference Herrera, Escobar, de Plata, Avila, Corradin and Herrera1992) and in Caucasian volunteers who had previously lived for more than 7 years in a malaria endemic region (either PNG, the Solomon Islands or south-east Asia) (Bilsborough et al. Reference Bilsborough, Carlisle and Good1993). These findings potentially reflect differences in transmission intensity.
In addition, a study in northern India by Seth et al. in 2010 assessed lymphocyte responses to B and T cell epitopes of four proteins, AMA1, MSP1, CSP and gametocyte surface antigen 1 (GAM1), covering each stage of the parasite's life-cycle (Seth et al. Reference Seth, Bhat, Rao and Biswas2010). They were able to identify positive responses to these epitopes amongst their study population; however, given the study design was cross-sectional, and the cellular assay used (lymphocyte proliferation assay), no further conclusions can be drawn. In fact, all the above studies have been cross-sectional in design, and hence severely limit the ability to be able to determine the relationship of P. vivax-specific cellular responses with clinical infection, and as such the need for longitudinal studies is clearly evident. Furthermore, more attention needs to be paid to the pre-erythrocytic stages of infection, particularly given the potential for P. vivax hypnozoites to lay dormant in the liver, to determine whether any cellular responses are induced to this stage during natural infection (and whether this is similar or different to that observed following P. falciparum infection).
Induction of cytokines and role of immunoregulation
It has long been recognized that the host responds to P. falciparum infection through the release of pro-inflammatory cytokines into the blood stream, with most focus on the role of tumour necrosis factor (TNF) (Grau et al. Reference Grau, Taylor, Molyneux, Wirima, Vassalli, Hommel and Lambert1989). Whilst this response acts to limit the infection (Kremsner et al. Reference Kremsner, Winkler, Brandts, Wildling, Jenne, Graninger, Prada, Bienzle, Juillard and Grau1995; Dodoo et al. Reference Dodoo, Omer, Todd, Akanmori, Koram and Riley2002), it can also induce immunopathology (Kwiatkowski et al. Reference Kwiatkowski, Molyneux, Stephens, Curtis, Klein, Pointaire, Smit, Allan, Brewster, Grau and Greenwood1993), and has been associated with severe P. falciparum infections and fatal outcomes (Grau et al. Reference Grau, Taylor, Molyneux, Wirima, Vassalli, Hommel and Lambert1989; Kwiatkowski et al. Reference Kwiatkowski, Hill, Sambou, Twumasi, Castracane, Manogue, Cerami, Brewster and Greenwood1990; Day et al. Reference Day, Hien, Schollaardt, Loc, Chuong, Chau, Mai, Phu, Sinh, White and Ho1999). The role of cytokine release in P. vivax was not assessed until the early 1990s (Karunaweera et al. Reference Karunaweera, Grau, Gamage, Carter and Mendis1992), when it was recognized that this infection also induced a TNF response, which was in fact comparatively stronger than that seen during P. falciparum infection. The induction of a strong pro-inflammatory response (cytokines such as TNF, interferon-gamma (IFN-γ), interleukin-12 (IL-12), IL-6, IL-1β and IL-8) during P. vivax infection compared with uninfected controls has since been shown in multiple studies (Torre et al. Reference Torre, Ferrario, Matteelli, Speranza, Giola, Pugliese, Cantamessa, Carosi and Fiori1998; Praba-Egge et al. Reference Praba-Egge, Montenegro, Arevalo-Herrera, Hopper, Herrera and James2003; Hemmer et al. Reference Hemmer, Holst, Kern, Chiwakata, Dietrich and Reisinger2006; Zeyrek et al. Reference Zeyrek, Kurcer, Zeyrek and Simsek2006; Fernandes et al. Reference Fernandes, Carvalho, Zanini, Ventura, Souza, Cotias, Silva-Filho and Daniel-Ribeiro2008; Andrade et al. Reference Andrade, Reis-Filho, Souza-Neto, Clarencio, Camargo, Barral and Barral-Netto2010; Jain et al. Reference Jain, Singh, Silawat, Patel, Saxena, Bharti, Shukla, Biswas and Singh2010; Medina et al. Reference Medina, Costa, Oliveira, Ventura, Souza, Gomes, Vallinoto, Povoa, Silva and Cunha2011; Goncalves et al. Reference Goncalves, Scopel, Bastos and Ferreira2012; Leoratti et al. Reference Leoratti, Trevelin, Cunha, Rocha, Costa, Gravina, Tada, Pereira, Golenbock, Antonelli and Gazzinelli2012; Mendonca et al. Reference Mendonca, Queiroz, Lopes, Andrade and Barral-Netto2013; Raza et al. Reference Raza, Ghanchi, Sarwar Zubairi, Raheem, Nizami and Beg2013; Silva et al. Reference Silva, Lacerda, Fujiwara, Bueno and Braga2013; da Costa et al. Reference da Costa, Antonelli, Costa, Pimentel, Garcia, Tarrago, dos Santos Mdo, Nogueira, Hekcmann, Sadahiro, Teixeira-Carvalho, Martins-Filho and Malheiro2014; Rodrigues-da-Silva et al. Reference Rodrigues-da-Silva, Lima-Junior Jda, e Fonseca Bde, Antas, Baldez, Storer, Santos, Banic and de Oliveira-Ferreira2014), in some cases with comparatively higher levels than seen during P. falciparum infection (Praba-Egge et al. Reference Praba-Egge, Montenegro, Arevalo-Herrera, Hopper, Herrera and James2003; Hemmer et al. Reference Hemmer, Holst, Kern, Chiwakata, Dietrich and Reisinger2006), although other reports are in contradiction (Fernandes et al. Reference Fernandes, Carvalho, Zanini, Ventura, Souza, Cotias, Silva-Filho and Daniel-Ribeiro2008; Goncalves et al. Reference Goncalves, Scopel, Bastos and Ferreira2012; Rodrigues-da-Silva et al. Reference Rodrigues-da-Silva, Lima-Junior Jda, e Fonseca Bde, Antas, Baldez, Storer, Santos, Banic and de Oliveira-Ferreira2014). The key differences between the two studies that observed higher levels of pro-inflammatory cytokines (namely TNF) and the three studies that did not were the geographic region. The studies that identified similar levels of pro-inflammatory cytokines in plasma samples from acute P. vivax and P. falciparum patients were all from the Brazilian Amazon, a region of low or unstable transmission (Fernandes et al. Reference Fernandes, Carvalho, Zanini, Ventura, Souza, Cotias, Silva-Filho and Daniel-Ribeiro2008; Goncalves et al. Reference Goncalves, Scopel, Bastos and Ferreira2012; Rodrigues-da-Silva et al. Reference Rodrigues-da-Silva, Lima-Junior Jda, e Fonseca Bde, Antas, Baldez, Storer, Santos, Banic and de Oliveira-Ferreira2014). In comparison, the study by Hemmer et al. and that of Praba-Egge et al. were conducted in non-immune European patients and in Colombians with an average of 1–2 prior episodes of malaria, respectively.
Induction of prominent secretion of the anti-inflammatory cytokine IL-10 has also been reported (Praba-Egge et al. Reference Praba-Egge, Montenegro, Arevalo-Herrera, Hopper, Herrera and James2003; Zeyrek et al. Reference Zeyrek, Kurcer, Zeyrek and Simsek2006; Fernandes et al. Reference Fernandes, Carvalho, Zanini, Ventura, Souza, Cotias, Silva-Filho and Daniel-Ribeiro2008; Jangpatarapongsa et al. Reference Jangpatarapongsa, Chootong, Sattabongkot, Chotivanich, Sirichaisinthop, Tungpradabkul, Hisaeda, Troye-Blomberg, Cui and Udomsangpetch2008; Andrade et al. Reference Andrade, Reis-Filho, Souza-Neto, Clarencio, Camargo, Barral and Barral-Netto2010; Bueno et al. Reference Bueno, Morais, Araujo, Gomes, Correa-Oliveira, Soares, Lacerda, Fujiwara and Braga2010; Goncalves et al. Reference Goncalves, Salmazi, Santos, Bastos, Rocha, Boscardin, Silber, Kallas, Ferreira and Scopel2010, Reference Goncalves, Scopel, Bastos and Ferreira2012; Jain et al. Reference Jain, Singh, Silawat, Patel, Saxena, Bharti, Shukla, Biswas and Singh2010; Yeo et al. Reference Yeo, Lampah, Tjitra, Piera, Gitawati, Kenangalem, Price and Anstey2010; Medina et al. Reference Medina, Costa, Oliveira, Ventura, Souza, Gomes, Vallinoto, Povoa, Silva and Cunha2011; Leoratti et al. Reference Leoratti, Trevelin, Cunha, Rocha, Costa, Gravina, Tada, Pereira, Golenbock, Antonelli and Gazzinelli2012; Borges et al. Reference Borges, Fontes and Damazo2013; Mendonca et al. Reference Mendonca, Queiroz, Lopes, Andrade and Barral-Netto2013; Raza et al. Reference Raza, Ghanchi, Sarwar Zubairi, Raheem, Nizami and Beg2013; Silva et al. Reference Silva, Lacerda, Fujiwara, Bueno and Braga2013; da Costa et al. Reference da Silva-Nunes, Codeco, Malafronte, da Silva, Juncansen, Muniz and Ferreira2014; Rodrigues-da-Silva et al. Reference Rodrigues-da-Silva, Lima-Junior Jda, e Fonseca Bde, Antas, Baldez, Storer, Santos, Banic and de Oliveira-Ferreira2014), again with multiple reports of higher levels during acute P. vivax than P. falciparum infection (Praba-Egge et al. Reference Praba-Egge, Montenegro, Arevalo-Herrera, Hopper, Herrera and James2003; Fernandes et al. Reference Fernandes, Carvalho, Zanini, Ventura, Souza, Cotias, Silva-Filho and Daniel-Ribeiro2008; Goncalves et al. Reference Goncalves, Salmazi, Santos, Bastos, Rocha, Boscardin, Silber, Kallas, Ferreira and Scopel2010, Reference Goncalves, Scopel, Bastos and Ferreira2012; Yeo et al. Reference Yeo, Lampah, Tjitra, Piera, Gitawati, Kenangalem, Price and Anstey2010) and one report of equal levels between the two species (Rodrigues-da-Silva et al. Reference Rodrigues-da-Silva, Lima-Junior Jda, e Fonseca Bde, Antas, Baldez, Storer, Santos, Banic and de Oliveira-Ferreira2014). IL-10 can have an immunoregulatory, or immunosuppressive, effect and the outcome of IL-10 secretion during P. falciparum infection is still an area of contention (reviewed in (Hansen and Schofield, Reference Hansen and Schofield2010)). Interestingly, Andrade et al. found that IL-10 levels were lower in severe cases of P. vivax malaria compared with asymptomatic cases (Andrade et al. Reference Andrade, Reis-Filho, Souza-Neto, Clarencio, Camargo, Barral and Barral-Netto2010), potentially suggesting high IL-10 levels could mediate clinical immunity by reducing the effect of harmful pro-inflammatory cytokines. However, this was not confirmed in two more recent studies similarly conducted in Brazil (Goncalves et al. Reference Goncalves, Scopel, Bastos and Ferreira2012; Mendonca et al. Reference Mendonca, Queiroz, Lopes, Andrade and Barral-Netto2013). The contradicting hypothesis is that high levels of IL-10 allow the parasite to escape the host immune response, resulting in the manifestation of symptomatic or severe malaria (Hugosson et al. Reference Hugosson, Montgomery, Premji, Troye-Blomberg and Bjorkman2004). Alternatively, another possible explanation is that high levels of IL-10 are not induced until a high threshold level of pro-inflammatory cytokines has been achieved; this is supported by studies that have found strong correlations between the level of pro-inflammatory cytokines produced (such as IFN-γ, TNF or IL-16) and the level of IL-10 (Raza et al. Reference Raza, Ghanchi, Sarwar Zubairi, Raheem, Nizami and Beg2013; Silva et al. Reference Silva, Lacerda, Fujiwara, Bueno and Braga2013; da Costa et al. Reference da Costa, Antonelli, Costa, Pimentel, Garcia, Tarrago, dos Santos Mdo, Nogueira, Hekcmann, Sadahiro, Teixeira-Carvalho, Martins-Filho and Malheiro2014). Unfortunately, as these studies are cross-sectional, causal relationships cannot be inferred, so the true mechanism remains unknown.
In most studies assessing the cytokine responses, levels have been measured in the plasma and hence the cellular source of these cytokines is unknown. Whilst pro-inflammatory cytokines are likely secreted by monocytes or macrophages and are part of the innate response to P. vivax (Leoratti et al. Reference Leoratti, Trevelin, Cunha, Rocha, Costa, Gravina, Tada, Pereira, Golenbock, Antonelli and Gazzinelli2012; Antonelli et al. Reference Antonelli, Leoratti, Costa, Rocha, Diniz, Tada, Pereira, Teixeira-Carvalho, Golenbock, Goncalves and Gazzinelli2014), they can also be secreted from P. vivax-specific T cells (Salwati et al. Reference Salwati, Minigo, Woodberry, Piera, de Silva, Kenangalem, Tjitra, Coppel, Price, Anstey and Plebanski2011), including memory cells (Wipasa et al. Reference Wipasa, Okell, Sakkhachornphop, Suphavilai, Chawansuntati, Liewsaree, Hafalla and Riley2011; Silva et al. Reference Silva, Lacerda, Fujiwara, Bueno and Braga2013), and there has also been the recent suggestion that active atypical memory B cells could secrete pro-inflammatory cytokines (Requena et al. Reference Requena, Campo, Umbers, Ome, Wangnapi, Barrios, Robinson, Samol, Rosanas-Urgell, Ubillos, Mayor, Lopez, de Lazzari, Arevalo-Herrera, Fernandez-Becerra, del Portillo, Chitnis, Siba, Bardaji, Mueller, Rogerson, Menendez and Dobano2014). Regulatory cytokines such as IL-10 or transforming growth factor-beta (TGF-β) are likely secreted by T regulatory cells (Tregs), however little is known about the Treg response during P. vivax infection. Jangpatarapongsa et al. first identified significant upregulation of total CD4+ CD25+ (and also FOX3P+) Tregs during acute P. vivax infection in the Tak province, Thailand, and demonstrated that these cells were likely the cellular source of IL-10 (Jangpatarapongsa et al. Reference Jangpatarapongsa, Chootong, Sattabongkot, Chotivanich, Sirichaisinthop, Tungpradabkul, Hisaeda, Troye-Blomberg, Cui and Udomsangpetch2008). Augmented CD4+ CD25+ FOX3P+ Tregs have since been similarly observed in two other acute P. vivax infected populations in Brazil (Bueno et al. Reference Bueno, Morais, Araujo, Gomes, Correa-Oliveira, Soares, Lacerda, Fujiwara and Braga2010; Goncalves et al. Reference Goncalves, Salmazi, Santos, Bastos, Rocha, Boscardin, Silber, Kallas, Ferreira and Scopel2010), although whether these cells are protective or harmful to the host has not been successfully elucidated. There is a clear need to solve this question in order to fully understand the development of naturally acquired immunity.
In summary, whilst a multitude of studies have assessed the production of cytokines during acute P. vivax infection, little information has been gathered on what cells are producing these cytokines and the relationship of pro- and anti-inflammatory cytokines with pathology or protection to the host. Furthermore, the majority of these studies were conducted in the Brazilian Amazon and in subjects greater than 13 years of age, limiting the applicability of the results. Only two studies assessed differences in the cytokine responses between adults and children. Zeyrek et al. found that there was a positive correlation between age and IL-8 levels in Turkish P. vivax infected patients, whilst there was a negative correlation between age and IL-12 levels (Zeyrek et al. Reference Zeyrek, Kurcer, Zeyrek and Simsek2006). Jain et al. identified significantly higher IFN-γ levels in children in India than adults, and also found higher levels of IL-10 and the IFN-γ induced protein (IP-10) in children with chills and rigors compared with those without, but interestingly the same relationship was not found in adults (Jain et al. Reference Jain, Singh, Silawat, Patel, Saxena, Bharti, Shukla, Biswas and Singh2010). Hence, further work is required to clarify the relationship of cytokines with protective immunity and to extend these findings over greater geographical areas, transmission levels and age groups.
Dendritic cell dysfunction in P. vivax infection
Dendritic cells (DCs) act to link the innate and adaptive immune responses, and hence play an extremely important role in the response to infection. Studies conducted on P. falciparum infected patients have found evidence of decreased numbers of circulating DCs (Pichyangkul et al. Reference Pichyangkul, Yongvanitchit, Kum-arb, Hemmi, Akira, Krieg, Heppner, Stewart, Hasegawa, Looareesuwan, Shanks and Miller2004) or DC modulation (Urban et al. Reference Urban, Ferguson, Pain, Willcox, Plebanski, Austyn and Roberts1999, Reference Urban, Mwangi, Ross, Kinyanjui, Mosobo, Kai, Lowe, Marsh and Roberts2001, Reference Urban, Cordery, Shafi, Bull, Newbold, Williams and Marsh2006; Skorokhod et al. Reference Skorokhod, Alessio, Mordmuller, Arese and Schwarzer2004). Similarly, during P. vivax infection the majority of studies conducted have suggested that DC maturation is inhibited (Jangpatarapongsa et al. Reference Jangpatarapongsa, Chootong, Sattabongkot, Chotivanich, Sirichaisinthop, Tungpradabkul, Hisaeda, Troye-Blomberg, Cui and Udomsangpetch2008; Bueno et al. Reference Bueno, Morais, Soares, Bouillet, Bruna-Romero, Fontes, Fujiwara and Braga2009; Goncalves et al. Reference Goncalves, Salmazi, Santos, Bastos, Rocha, Boscardin, Silber, Kallas, Ferreira and Scopel2010; Pinzon-Charry et al. Reference Pinzon-Charry, Woodberry, Kienzle, McPhun, Minigo, Lampah, Kenangalem, Engwerda, Lopez, Anstey and Good2013). Such a response was first identified within P. vivax-infected patients from Thailand, where a reduction was observed compared with uninfected controls in the total number of circulating DCs (Jangpatarapongsa et al. Reference Jangpatarapongsa, Chootong, Sattabongkot, Chotivanich, Sirichaisinthop, Tungpradabkul, Hisaeda, Troye-Blomberg, Cui and Udomsangpetch2008). Further examination indicated this reduction also changed the ratio of myeloid (CD11c+) and plasmacytoid (CD123+) DCs, with a greater reduction in myeloid DCs overall. This finding was also supported by two studies in Brazil (Bueno et al. Reference Bueno, Morais, Soares, Bouillet, Bruna-Romero, Fontes, Fujiwara and Braga2009; Goncalves et al. Reference Goncalves, Salmazi, Santos, Bastos, Rocha, Boscardin, Silber, Kallas, Ferreira and Scopel2010), where more extensive phenotyping also revealed down-modulation of antigen presenting molecules such as CD1a and HLA-DR, as well as accessory molecules such as CD80 and CD86.
It has been suggested that the reduction of DCs during malarial infection could reflect sequestration of such cells to lymphoid tissues (Jangpatarapongsa et al. Reference Jangpatarapongsa, Chootong, Sattabongkot, Chotivanich, Sirichaisinthop, Tungpradabkul, Hisaeda, Troye-Blomberg, Cui and Udomsangpetch2008; Pinzon-Charry et al. Reference Pinzon-Charry, Woodberry, Kienzle, McPhun, Minigo, Lampah, Kenangalem, Engwerda, Lopez, Anstey and Good2013), however one study conducted in Indonesian Papua provides an alternative theory. Pinzon-Charry et al. demonstrated clearly that the reduction in circulating DCs in patients infected with either P. vivax or P. falciparum was due to induced apoptosis of DCs (Pinzon-Charry et al. Reference Pinzon-Charry, Woodberry, Kienzle, McPhun, Minigo, Lampah, Kenangalem, Engwerda, Lopez, Anstey and Good2013). They also extended the phenotypic analysis of the remaining DCs by demonstrating the functional effect: impairment of these DCs to mature, as well as their ability to capture and present antigen to T cells. Furthermore, they also described a clear role for IL-10 in this process, whereby IL-10 levels in the plasma not only correlated with DC apoptosis but also blocking of IL-10 could prevent apoptosis in vitro (Pinzon-Charry et al. Reference Pinzon-Charry, Woodberry, Kienzle, McPhun, Minigo, Lampah, Kenangalem, Engwerda, Lopez, Anstey and Good2013). Pinzon-Charry et al. described a similar effect on DCs from both P. vivax and P. falciparum; in contrast, Goncalves et al. had only found impaired maturation within P. vivax and not P. falciparum infected patients (Goncalves et al. Reference Goncalves, Salmazi, Santos, Bastos, Rocha, Boscardin, Silber, Kallas, Ferreira and Scopel2010).
There have also been two studies on monocyte/macrophage populations during P. vivax infection. DCs likely arise from monocytes (Guilliams et al. Reference Guilliams, Ginhoux, Jakubzick, Naik, Onai, Schraml, Segura, Tussiwand and Yona2014) and they are differentiated based on the expression of different CD markers; however, both have a role in phagocytosis. Antonelli et al. recently found that the absolute numbers of monocytes (CD14+ CD16−, CD14low CD16+ and CD14− CD16+) increase during P. vivax infection and that they have a greater ability for phagocytosis, compared to cells of uninfected controls (Antonelli et al. Reference Antonelli, Leoratti, Costa, Rocha, Diniz, Tada, Pereira, Teixeira-Carvalho, Golenbock, Goncalves and Gazzinelli2014). However, Fernandez-Arias et al. previously described a reduction in the number of monocytes/macrophages (CD16+ CD10−) and a decrease in expression of surface complement receptor 1 (CR1) (which they related to decreased clearance of immune complexes) (Fernandez-Arias et al. Reference Fernandez-Arias, Lopez, Hernandez-Perez, Bautista-Ojeda, Branch and Rodriguez2013). As the two studies utilized different methods of classifying monocytes further analysis needs to be performed, in particular to determine whether an increase in monocytes and subsequent phagocytosis could be complementary to the decreased function of DCs.
Comparison with the cellular response induced by P. falciparum
As P. falciparum has long been considered the most clinically important species of Plasmodium, given its greater lethality than P. vivax, more extensive research has been conducted into the immunological response to this infection. This is also due to the greater ease of working with P. falciparum parasites (i.e. well established in vitro culture) and hence being able to determine P. falciparum-specific responses. A relatively limited number of studies have directly compared responses between P. falciparum and P. vivax infected patients, likely because in most areas of the world one species is dominant (Baird, Reference Baird2013). Whilst it is difficult to directly compare antigen-specific B and T cell responses between species given the antigens do differ, it is possible to directly compare cytokine responses. It has largely been accepted that P. vivax is more pyrogenic than P. falciparum, in that it induces a greater cytokine response leading to fever at relatively lower parasite loads (Price et al. Reference Price, Tjitra, Guerra, Yeung, White and Anstey2007). However the actual evidence is somewhat contradictory, as discussed above (whilst not all studies assessed cytokine concentration per parasite, in those that did not the parasitaemia was not significantly different between P. falciparum and P. vivax infected patients (Fernandes et al. Reference Fernandes, Carvalho, Zanini, Ventura, Souza, Cotias, Silva-Filho and Daniel-Ribeiro2008; Rodrigues-da-Silva et al. Reference Rodrigues-da-Silva, Lima-Junior Jda, e Fonseca Bde, Antas, Baldez, Storer, Santos, Banic and de Oliveira-Ferreira2014)). Whilst IL-10 responses clearly seem to be greater in response to P. vivax infection (Praba-Egge et al. Reference Praba-Egge, Montenegro, Arevalo-Herrera, Hopper, Herrera and James2003; Fernandes et al. Reference Fernandes, Carvalho, Zanini, Ventura, Souza, Cotias, Silva-Filho and Daniel-Ribeiro2008; Goncalves et al. Reference Goncalves, Salmazi, Santos, Bastos, Rocha, Boscardin, Silber, Kallas, Ferreira and Scopel2010, Reference Goncalves, Scopel, Bastos and Ferreira2012; Yeo et al. Reference Yeo, Lampah, Tjitra, Piera, Gitawati, Kenangalem, Price and Anstey2010), the evidence is less certain for pro-inflammatory cytokines. A number of reasons could account for any observed differences, such as the age of the subjects, the transmission level and geographical location, the stage of infection at which the sample was taken and the method used for enumerating the relative cytokine concentration. Importantly, a greater effort needs to be made to use standardized assays for measurement of cytokine responses, as recently described (Moncunill et al. Reference Moncunill, Aponte, Nhabomba and Dobano2013).
NATURALLY ACQUIRED HUMORAL IMMUNITY
Antibodies are considered to provide what is known as ‘naturally acquired immunity’ to malaria. This is largely based upon a pivotal study conducted in 1961 that demonstrated that transferring sera from P. falciparum immune adults to children could subsequently protect those children from clinical disease (Cohen et al. Reference Cohen, Mc and Carrington1961); this finding was independently confirmed 30 years later (Sabchareon et al. Reference Sabchareon, Burnouf, Ouattara, Attanath, Bouharoun-Tayoun, Chantavanich, Foucault, Chongsuphajaisiddhi and Druilhe1991). Antibody-mediated effector functions can include the inhibition of red blood cell (RBC) invasion, neutralization, opsonization and antibody-dependent cellular inhibition (Beeson et al. Reference Beeson, Osier and Engwerda2008). In this section we will focus on what is known about the production of P. vivax-specific antibodies, focusing on DBP as well as some other proteins of vaccine interest, in addition to the potential role of P. vivax-specific antibodies in blocking transmission.
Global production of antibodies
The P. vivax parasite contains over 5000 proteins (Aurrecoechea et al. Reference Aurrecoechea, Brestelli, Brunk, Dommer, Fischer, Gajria, Gao, Gingle, Grant, Harb, Heiges, Innamorato, Iodice, Kissinger, Kraemer, Li, Miller, Nayak, Pennington, Pinney, Roos, Ross, Stoeckert, Treatman and Wang2009); whilst a number of these have been selected as vaccine candidates by various methods, we still do not know which are responsible for clinical immunity observed in endemic areas, nor do we have a suitable, highly efficacious candidate for a blood-stage malaria vaccine (for either P. vivax or P. falciparum). Therefore, whilst some protein-specific responses will be discussed in more detail below, we first wanted to review what has been learnt from studies where the focus was on the global production of antibodies. This generally involves using whole P. vivax lysate as a stimulant, or more recently, large-scale protein arrays.
An early study by Ray et al. demonstrated that a high proportion of Indian patients (94%), positive for P. vivax, responded with antibodies against P. vivax blood-stage lysate (Ray et al. Reference Ray, Ansari and Sharma1994). They also demonstrated a high proportion of responsiveness in non-infected patients living in India (50%), compared with zero of nine controls from abroad. This demonstrated that sero-conversion following P. vivax infection is highly likely, and that potentially these responses can be maintained, which would account for the high rate in non-infected Indian controls. Whilst this study was useful in indicating that P. vivax-specific antibodies are highly prevalent, it did not shed any light on how many of the 5000-odd proteins are immunogenic, or whether this response was due to one immunodominant antigen.
Since this study was undertaken the P. vivax genome and transcriptome has been completed (Bozdech et al. Reference Bozdech, Mok, Hu, Imwong, Jaidee, Russell, Ginsburg, Nosten, Day, White, Carlton and Preiser2008; Carlton et al. Reference Carlton, Adams, Silva, Bidwell, Lorenzi, Caler, Crabtree, Angiuoli, Merino, Amedeo, Cheng, Coulson, Crabb, Del Portillo, Essien, Feldblyum, Fernandez-Becerra, Gilson, Gueye, Guo, Kang'a, Kooij, Korsinczky, Meyer, Nene, Paulsen, White, Ralph, Ren and Sargeant2008; Westenberger et al. Reference Westenberger, McClean, Chattopadhyay, Dharia, Carlton, Barnwell, Collins, Hoffman, Zhou, Vinetz and Winzeler2010). Together with the development of efficient systems for protein production (Davies et al. Reference Davies, Liang, Hernandez, Randall, Hirst, Mu, Romero, Nguyen, Kalantari-Dehaghi, Crotty, Baldi, Villarreal and Felgner2005; Tsuboi et al. Reference Tsuboi, Takeo, Iriko, Jin, Tsuchimochi, Matsuda, Han, Otsuki, Kaneko, Sattabongkot, Udomsangpetch, Sawasaki, Torii and Endo2008), this has enabled the development of P. vivax-specific protein arrays. Three such studies, screening between 89 and 152 proteins with sera from 20 to 60 patients, identified the rate of immunogenic proteins to be between 11 and 27·5% (Chen et al. Reference Chen, Jung, Wang, Ha, Lu, Lim, Takeo, Tsuboi and Han2010; Molina et al. Reference Molina, Finney, Arevalo-Herrera, Herrera, Felgner, Gardner, Liang and Wang2012; Lu et al. Reference Lu, Li, Wang, Cheng, Kong, Cui, Ha, Sattabongkot, Tsuboi and Han2014). However, a much larger study was recently conducted. Finney et al. screened 1936 P. vivax proteins (approximately 40% of the predicted 5000 proteins in P. vivax) with sera from 224 PNG children (Finney et al. Reference Finney, Danziger, Molina, Vignali, Takagi, Ji, Stanisic, Siba, Liang, Aitchison, Mueller, Gardner and Wang2014). They demonstrated that over 50% of these proteins were recognized in these children, which far outweighs the number of proteins that have traditionally been assessed for antibody responses (discussed below). Interestingly, the authors also had the opportunity to assess the contribution of these responses to clinical immunity, by comparing symptomatic and asymptomatic children, and did find that symptomatic children carried fewer antibodies (Finney et al. Reference Finney, Danziger, Molina, Vignali, Takagi, Ji, Stanisic, Siba, Liang, Aitchison, Mueller, Gardner and Wang2014). Such protein arrays are likely to provide useful information in the future, not only for identifying protective antibodies, but also for identifying those that act as markers of exposure and increasing our understanding of the humoral immune response to P. vivax in general.
The role of DBP antibodies and the association with protective immunity
DBP is the ligand the P. vivax parasite uses to bind to the Duffy antigen receptor for chemokines (DARC) on RBCs in order to invade. However, more recently it has been shown that this interaction is not essential for invasion, as the parasite can also infect Duffy negative individuals (although this is extremely rare) (Ryan et al. Reference Ryan, Stoute, Amon, Dunton, Mtalib, Koros, Owour, Luckhart, Wirtz, Barnwell and Rosenberg2006; Cavasini et al. Reference Cavasini, Mattos, Couto, Bonini-Domingos, Valencia, Neiras, Alves, Rossit, Castilho and Machado2007; Menard et al. Reference Menard, Barnadas, Bouchier, Henry-Halldin, Gray, Ratsimbasoa, Thonier, Carod, Domarle, Colin, Bertrand, Picot, King, Grimberg, Mercereau-Puijalon and Zimmerman2010; Mendes et al. Reference Mendes, Dias, Figueiredo, Mora, Cano, de Sousa, do Rosario, Benito, Berzosa and Arez2011; Woldearegai et al. Reference Woldearegai, Kremsner, Kun and Mordmuller2013). DBP is a type 1 membrane protein and it is the amino cysteine-rich domain, known as region II, which binds to DARC (Chitnis and Miller, Reference Chitnis and Miller1994; Chitnis et al. Reference Chitnis, Chaudhuri, Horuk, Pogo and Miller1996). Given the importance of this protein in invasion, there has been strong interest in using DBP as a vaccine target. In 1997 it was demonstrated that naturally infected individuals do make antibodies to DBP region II (DBPII) (Fraser et al. Reference Fraser, Michon, Barnwell, Noe, Al-Yaman, Kaslow and Adams1997), and subsequently it was also shown that such antibodies could block binding of DBPII to RBCs (Michon et al. Reference Michon, Fraser and Adams2000). The level of inhibition correlated with the antibody response, suggesting that with the development of enough antibodies against DBP this could result in protective immunity. This was further supported by the association of increased levels of antibodies to DBPII with age (Michon et al. Reference Michon, Arevalo-Herrera, Fraser, Herrera and Adams1998; Xainli et al. Reference Xainli, Cole-Tobian, Baisor, Kastens, Bockarie, Yazdani, Chitnis, Adams and King2003) or with level of exposure (Ceravolo et al. Reference Ceravolo, Bruna-Romero, Braga, Fontes, Brito, Souza, Krettli, Adams and Carvalho2005). Many subsequent studies have provided further support for the development of antibodies against DBP in various geographical regions (Cole-Tobian et al. Reference Cole-Tobian, Cortes, Baisor, Kastens, Xainli, Bockarie, Adams and King2002; Suh et al. Reference Suh, Choi, Lee, Woo, Kang, Won, Cho and Lim2003; Tran et al. Reference Tran, Oliveira-Ferreira, Moreno, Santos, Yazdani, Chitnis, Altman, Meyer, Barnwell and Galinski2005; Barbedo et al. Reference Barbedo, Ricci, Jimenez, Cunha, Yazdani, Chitnis, Rodrigues and Soares2007; Maestre et al. Reference Maestre, Muskus, Duque, Agudelo, Liu, Takagi, Ntumngia, Adams, Sim, Hoffman, Corradin, Velez and Wang2010; Souza-Silva et al. Reference Souza-Silva, da Silva-Nunes, Sanchez, Ceravolo, Malafronte, Brito, Ferreira and Carvalho2010; Zakeri et al. Reference Zakeri, Babaeekhou, Mehrizi, Abbasi and Djadid2011; Kano et al. Reference Kano, Sanchez, Sousa, Tang, Saliba, Oliveira, Nogueira, Goncalves, Fontes, Soares, Brito, Rocha and Carvalho2012; Valizadeh et al. Reference Valizadeh, Zakeri, Mehrizi and Djadid2014), with a subset also assessing and demonstrating inhibitory activity (Ceravolo et al. Reference Ceravolo, Souza-Silva, Fontes, Braga, Madureira, Krettli, Souza, Brito, Adams and Carvalho2008, Reference Ceravolo, Sanchez, Sousa, Guerra, Soares, Braga, McHenry, Adams, Brito and Carvalho2009; Grimberg et al. Reference Grimberg, Udomsangpetch, Xainli, McHenry, Panichakul, Sattabongkot, Cui, Bockarie, Chitnis, Adams, Zimmerman and King2007; King et al. Reference King, Michon, Shakri, Marcotty, Stanisic, Zimmerman, Cole-Tobian, Mueller and Chitnis2008; Chootong et al. Reference Chootong, Ntumngia, VanBuskirk, Xainli, Cole-Tobian, Campbell, Fraser, King and Adams2010; Chootong et al. Reference Chootong, Panichakul, Permmongkol, Barnes, Udomsangpetch and Adams2012; Souza-Silva et al. Reference Souza-Silva, Torres, Santos-Alves, Tang, Sanchez, Sousa, Fontes, Nogueira, Rocha, Brito, Adams, Kano and Carvalho2014).
However, whilst these responses have been identified in P. vivax infected individuals, in each study conducted, a significant proportion of infected or exposed individuals did not make antibodies against DBP. Similarly, of the studies listed above that described inhibition of binding or invasion of RBCs, the proportions of individuals with antibodies capable of this functionality in each study were low. Such variability could be due to the polymorphism within DBPII (Tsuboi et al. Reference Tsuboi, Kappe, al-Yaman, Prickett, Alpers and Adams1994; Cole-Tobian and King, Reference Cole-Tobian and King2003), which may favour immune evasion (and be the result of earlier selection), or due to difference in host genotype of the DARC molecule. DARC is a glycosylated membrane protein encoded by five common genotypes, resulting in three phenotypes that are associated with differential levels of expression of DARC (Maestre et al. Reference Maestre, Muskus, Duque, Agudelo, Liu, Takagi, Ntumngia, Adams, Sim, Hoffman, Corradin, Velez and Wang2010). It has been demonstrated that antibody responses against DBPII are greater within individuals with suspected lower levels of DARC (Maestre et al. Reference Maestre, Muskus, Duque, Agudelo, Liu, Takagi, Ntumngia, Adams, Sim, Hoffman, Corradin, Velez and Wang2010). This is potentially due to the suggested ability of DARC to down regulate the host immune response, although this has yet to be fully elucidated and more recent studies have disputed this finding (King et al. Reference King, Adams, Xianli, Grimberg, McHenry, Greenberg, Siddiqui, Howes, da Silva-Nunes, Ferreira and Zimmerman2011; Souza-Silva et al. Reference Souza-Silva, Torres, Santos-Alves, Tang, Sanchez, Sousa, Fontes, Nogueira, Rocha, Brito, Adams, Kano and Carvalho2014).
The above findings make it clear that there is questionable evidence over whether DBP, or DBPII, would make a successful vaccine target. However, a study by King et al. did find an association between the presence of high-level inhibitory antibodies to DBPII and a delayed time to P. vivax re-infection, compared with subjects with low-level inhibitory antibodies (King et al. Reference King, Michon, Shakri, Marcotty, Stanisic, Zimmerman, Cole-Tobian, Mueller and Chitnis2008). Perhaps most importantly, the authors also found that such high-level inhibitory antibodies associated with protection were able to overcome strain-specific responses, with similar levels of inhibition against P. vivax with six different DBPII haplotypes. These findings suggest that a vaccine may be successful, if predominantly protective, high-level inhibitory antibodies can be induced over non-, or low-level, inhibitory antibodies for DBPII.
Production of antigen-specific antibodies
Apart from the interest in DBP as a vaccine candidate, a number of other P. vivax blood-stage proteins have received a significant amount of interest. The majority of studies have been conducted on MSP1 (Soares et al. Reference Soares, Levitus, Souza, Del Portillo and Rodrigues1997, Reference Soares, Oliveira, Souza and Rodrigues1999b; Ak et al. Reference Ak, Jones, Charoenvit, Kumar, Kaslow, Maris, Marwoto, Masbar and Hoffman1998; Egan et al. Reference Egan, Burghaus, Druilhe, Holder and Riley1999; Park et al. Reference Park, Moon, Yeom, Lim, Sohn, Jung, Cho, Jeon, Ju, Ki, Oh and Choe2001; Soares and Rodrigues, Reference Soares and Rodrigues2002; Braga et al. Reference Braga, Barros, Reis, Fontes, Morais, Martins and Krettli2002a; Rodrigues et al. Reference Rodrigues, Cunha, Machado, Ferreira, Rodrigues and Soares2003; Suh et al. Reference Suh, Choi, Lee, Woo, Kang, Won, Cho and Lim2003; Lim et al. Reference Lim, Park, Yeom, Lee, Yoo, Oh, Sohn, Bahk and Kim2004; Morais et al. Reference Morais, Soares, Carvalho, Fontes, Krettli and Braga2005; Valderrama-Aguirre et al. Reference Valderrama-Aguirre, Quintero, Gomez, Castellanos, Perez, Mendez, Arevalo-Herrera and Herrera2005; Nogueira et al. Reference Nogueira, Alves, Fernandez-Becerra, Pein, Santos, Pereira da Silva, Camargo and del Portillo2006; Barbedo et al. Reference Barbedo, Ricci, Jimenez, Cunha, Yazdani, Chitnis, Rodrigues and Soares2007; Bastos et al. Reference Bastos, da Silva-Nunes, Malafronte, Hoffmann, Wunderlich, Moraes and Ferreira2007; Ladeia-Andrade et al. Reference Ladeia-Andrade, Ferreira, Scopel, Braga, Bastos Mda, Wunderlich and Coura2007; Pitabut et al. Reference Pitabut, Panichakorn, Mahakunkijcharoen, Hirunpetcharat, Looareesuwan and Khusmith2007; Wickramarachchi et al. Reference Wickramarachchi, Illeperuma, Perera, Bandara, Holm, Longacre, Handunnetti and Udagama-Randeniya2007; Zeyrek et al. Reference Zeyrek, Babaoglu, Demirel, Erdogan, Ak, Korkmaz and Coban2008; Mehrizi et al. Reference Mehrizi, Zakeri, Salmanian, Sanati and Djadid2009; Seth et al. Reference Seth, Bhat, Rao and Biswas2010; Fernandez-Becerra et al. Reference Fernandez-Becerra, Sanz, Brucet, Stanisic, Alves, Camargo, Alonso, Mueller and del Portillo2010; Storti-Melo et al. Reference Storti-Melo, Souza-Neiras, Cassiano, Taveira, Cordeiro, Couto, Povoa, Cunha, Echeverry, Rossit, Arevalo-Herrera, Herrera and Machado2011; Lima-Junior et al. Reference Lima-Junior, Rodrigues-da-Silva, Banic, Jiang, Singh, Fabricio-Silva, Porto, Meyer, Moreno, Rodrigues, Barnwell, Galinski and de Oliveira-Ferreira2012; Mourao et al. Reference Mourao, Morais, Bueno, Jimenez, Soares, Fontes, Guimaraes Lacerda, Xavier, Barnwell, Galinski and Braga2012; Riccio et al. Reference Riccio, Totino, Pratt-Riccio, Ennes-Vidal, Soares, Rodrigues, de Souza, Daniel-Ribeiro and Ferreira-da-Cruz Mde2013; Versiani et al. Reference Versiani, Almeida, Melo, Versiani, Orlandi, Mariuba, Soares, Souza, da Silva Balieiro, Monteiro, Costa, del Portillo, Lacerda and Nogueira2013). This is largely due to the extensive interest in a P. falciparum MSP1 vaccine given the presence of naturally induced antibodies and some associations with protection (reviewed in (Holder, Reference Holder2009)), and the identification of blocks of conserved sequence between P. falciparum and P. vivax (del Portillo et al. Reference del Portillo, Longacre, Khouri and David1991). Of the studies measuring antibody responses against P. vivax MSP1, only two were able to associate MSP1-specific antibodies with protection (Nogueira et al. Reference Nogueira, Alves, Fernandez-Becerra, Pein, Santos, Pereira da Silva, Camargo and del Portillo2006; Versiani et al. Reference Versiani, Almeida, Melo, Versiani, Orlandi, Mariuba, Soares, Souza, da Silva Balieiro, Monteiro, Costa, del Portillo, Lacerda and Nogueira2013). However, all studies detected a relatively high frequency of P. vivax infected or previously infected individuals who were sero-positive for MSP1 (30–98%, largely dependent on the region of PvMSP1 assessed), suggesting such antibodies are rather more reflective of exposure than clinical protection.
Other studies assessing naturally induced antibodies have been against blood-stage proteins including MSP3 (Lima-Junior et al. Reference Lima-Junior, Jiang, Rodrigues-da-Silva, Banic, Tran, Ribeiro, Meyer, De-Simone, Santos, Moreno, Barnwell, Galinski and Oliveira-Ferreira2011, Reference Lima-Junior, Rodrigues-da-Silva, Banic, Jiang, Singh, Fabricio-Silva, Porto, Meyer, Moreno, Rodrigues, Barnwell, Galinski and de Oliveira-Ferreira2012; Mourao et al. Reference Mourao, Morais, Bueno, Jimenez, Soares, Fontes, Guimaraes Lacerda, Xavier, Barnwell, Galinski and Braga2012; Stanisic et al. Reference Stanisic, Javati, Kiniboro, Lin, Jiang, Singh, Meyer, Siba, Koepfli, Felger, Galinski and Mueller2013), MSP9 (Soares and Rodrigues, Reference Soares and Rodrigues2002; Lima-Junior et al. Reference Lima-Junior, Tran, Meyer, Singh, De-Simone, Santos, Daniel-Ribeiro, Moreno, Barnwell, Galinski and Oliveira-Ferreira2008; Stanisic et al. Reference Stanisic, Javati, Kiniboro, Lin, Jiang, Singh, Meyer, Siba, Koepfli, Felger, Galinski and Mueller2013), AMA1 (Rodrigues et al. Reference Rodrigues, Rodrigues, Oliveira, Comodo, Rodrigues, Kocken, Thomas and Soares2005; Wickramarachchi et al. Reference Wickramarachchi, Premaratne, Perera, Bandara, Kocken, Thomas, Handunnetti and Udagama-Randeniya2006; Barbedo et al. Reference Barbedo, Ricci, Jimenez, Cunha, Yazdani, Chitnis, Rodrigues and Soares2007; Mufalo et al. Reference Mufalo, Gentil, Bargieri, Costa, Rodrigues and Soares2008; Seth et al. Reference Seth, Bhat, Rao and Biswas2010; Bueno et al. Reference Bueno, Lobo, Morais, Mourao, de Avila, Soares, Fontes, Lacerda, Chavez Olortegui, Bartholomeu, Fujiwara and Braga2011; Dias et al. Reference Dias, Somarathna, Manamperi, Escalante, Gunasekera and Udagama2011) and various TRAgs (Garg et al. Reference Garg, Chauhan, Singh and Sharma2008; Siddiqui et al. Reference Siddiqui, Bora, Singh, Dash and Sharma2008; Mittra et al. Reference Mittra, Singh and Sharma2010; Zeeshan et al. Reference Zeeshan, Bora and Sharma2013). Similar studies have focused on the P. vivax variant proteins (VIRs) (Oliveira et al. Reference Oliveira, Fernandez-Becerra, Jimenez, Del Portillo and Soares2006), the reticulocyte binding proteins (RBPs) (Tran et al. Reference Tran, Oliveira-Ferreira, Moreno, Santos, Yazdani, Chitnis, Altman, Meyer, Barnwell and Galinski2005), the sporozoite antigen CSP (Migot et al. Reference Migot, Millet, Chougnet, Lepers and Deloron1993; Carvalho et al. Reference Carvalho, Fontes, Fernandes, Marinuzzi and Krettli1997; Arevalo-Herrera et al. Reference Arevalo-Herrera, Roggero, Gonzalez, Vergara, Corradin, Lopez and Herrera1998; Suh et al. Reference Suh, Choi, Lee, Woo, Kang, Won, Cho and Lim2003; Seth et al. Reference Seth, Bhat, Rao and Biswas2010) and the gametocyte antigen GAM1 (Seth et al. Reference Seth, Bhat, Rao and Biswas2010). However, it is difficult to determine the relative utility of such studies, when antibodies to only a small number of proteins are measured at only one time-point and a detailed history of malaria is unknown. A systematic review and meta-analysis of 18 studies (mostly cross-sectional) was recently conducted, in an attempt to combine all the relevant P. vivax antibody data. Overall, IgG responses to PvCSP, PvMSP1, PvMSP9 and PvAMA1 were associated with increased odds of P. vivax infection in a diverse range of populations (Cutts et al. Reference Cutts, Powell, Agius, Beeson, Simpson and Fowkes2014), hence acting as markers of exposure. Only PvMSP1 (described above), PvMSP3 and PvMSP9 had associations with protection from P. vivax malaria.
The main determinant of whether antibodies associate with increased risk of infections or of protection is likely to be the previous lifetime exposure (Koepfli et al. Reference Koepfli, Colborn, Kiniboro, Lin, Speed, Siba, Felger and Mueller2013). In a recent longitudinal study in PNG children 1–4 years of age that acquired approximately 2·5-times as many distinct P. vivax than P. falciparum infections (Mueller et al. Reference Mueller, Schoepflin, Smith, Benton, Bretscher, Lin, Kiniboro, Zimmerman, Speed, Siba and Felger2012), antibodies to PvMSP3 block II and PvMSP9 N-terminus were found to be independently associated with reduced risk of clinical malaria even after correction for individual variation in exposure (Stanisic et al. Reference Stanisic, Javati, Kiniboro, Lin, Jiang, Singh, Meyer, Siba, Koepfli, Felger, Galinski and Mueller2013). In the same children antibodies to a panel of P. falciparum merozoite antigens, however, predicted an increased risk of P. falciparum malaria, an association that disappeared when correcting for individual difference in exposure (Stanisic et al. Reference Stanisic, Fowkes, Koinari, Javati, Lin, Kiniboro, Richards, Robinson, Schofield, Kazura, King, Zimmerman, Felger, Siba, Mueller and Beeson2015).
In order to investigate antibodies as potential markers of exposure or as correlates of (clinical) protection, it is therefore essential to account both for difference in exposure as well as for the boosting of antibody responses by concurrent infections. By combining such well designed cohort studies with cutting-edge approaches (e.g. protein arrays) it will be become possible to screen a much larger part of the P. vivax proteome and rationally prioritize antigens for further evaluation as potential vaccine candidates. However, the lack of a continuous P. vivax in vitro culture system makes it significantly more difficult to develop an assay that can evaluate not only the presence but also investigate the functionality of antibodies against P. vivax.
Role of P. vivax-specific antibodies in blocking transmission
Interestingly, as the presence of malarial infections diminishes with age, so does the presence of gametocytes (Nguyen et al. Reference Nguyen, van den Eede, van Overmeir, Thang, Hung le, D'Alessandro and Erhart2012). Whilst this could be simply due to a decrease in the number of total parasites (and hence gametocytes included), there also could be an induction of antibodies specific to sexual-stage antigens and hence naturally acquired transmission-blocking immunity. Early studies conducted in Sri Lanka indicated that this could well be the case (Mendis et al. Reference Mendis, Munesinghe, de Silva, Keragalla and Carter1987; de Zoysa et al. Reference de Zoysa, Herath, Abhayawardana, Padmalal and Mendis1988; Peiris et al. Reference Peiris, Premawansa, Ranawaka, Udagama, Munasinghe, Nanayakkara, Gamage, Carter, David and Mendis1988; Ranawaka et al. Reference Ranawaka, Munesinghe, de Silva, Carter and Mendis1988; Gamage-Mendis et al. Reference Gamage-Mendis, Rajakaruna, Carter and Mendis1992). These studies demonstrated that sera from P. vivax infected patients were very effective at suppressing infection of the parasites in mosquitoes. They identified that this was likely due to the presence of antibodies against surface antigens of extracellular gametes or zygotes, as the level of these antibodies correlated closely with the level of suppression of infectivity. They also demonstrated that both the reduction in infectivity and the level of antibodies increased when the patient had frequent, recent P. vivax infections. However, beyond 4 months since the last infection this pattern was not present, suggesting that there was no true maintenance (or perhaps, generation) of memory to this form of transmission-blocking immunity.
However, following these studies in Sri Lanka a similar experiment was conducted along the southern coast of Mexico (Ramsey et al. Reference Ramsey, Salinas and Rodriguez1996), and the authors found slightly different results. They also identified transmission-blocking activity using sera from patients with acute P. vivax malaria, however only in those who had experienced at least one previous malaria episode, and the duration since this past episode was not important (i.e. even if the previous episode was more than 7 months prior, they still exhibited good transmission-blocking activity). Transmission-blocking activity has since also been identified using P. vivax infected patients from western Thailand (Sattabongkot et al. Reference Sattabongkot, Maneechai, Phunkitchar, Eikarat, Khuntirat, Sirichaisinthop, Burge and Coleman2003; Coleman et al. Reference Coleman, Kumpitak, Ponlawat, Maneechai, Phunkitchar, Rachapaew, Zollner and Sattabongkot2004) and Colombia (Arevalo-Herrera et al. Reference Arevalo-Herrera, Solarte, Zamora, Mendez, Yasnot, Rocha, Long, Miller and Herrera2005). Nearly all such studies have been conducted on symptomatic patients, clearly demonstrating that such transmission-blocking activity may be of benefit to the community, but not to the individual patient.
LONGEVITY OF IMMUNITY: EVIDENCE OF PREMUNITION, MEMORY OR EXHAUSTION
The longevity of immunity to malaria has long been a hot topic, with observations for P. falciparum pointing to a situation of ‘premunition’, whereby constant antigenic stimulation is required to maintain the immune response (Smith et al. Reference Smith, Felger, Tanner and Beck1999). There has also been more recent evidence for P. falciparum that the immune response may be dysfunctional, with studies identifying atypical B cells and exhausted T cells (Weiss et al. Reference Weiss, Crompton, Li, Walsh, Moir, Traore, Kayentao, Ongoiba, Doumbo and Pierce2009, Reference Weiss, Clark, Li, Traore, Kayentao, Ongoiba, Hernandez, Doumbo, Pierce, Branch and Crompton2011; Illingworth et al. Reference Illingworth, Butler, Roetynck, Mwacharo, Pierce, Bejon, Crompton, Marsh and Ndungu2013; Scholzen et al. Reference Scholzen, Teirlinck, Bijker, Roestenberg, Hermsen, Hoffman and Sauerwein2014), and we know that memory responses to all pathogens or vaccines are not equal (Bottiger et al. Reference Bottiger, Gustavsson and Svensson1998; Amanna and Slifka, Reference Amanna and Slifka2010). How this contributes to the overall immune response and whether this is related to the perceived lack of memory for P. falciparum has not yet been fully quantified, however some evidence suggests that these cells are not, in fact, ‘dysfunctional’ (Muellenbeck et al. Reference Muellenbeck, Ueberheide, Amulic, Epp, Fenyo, Busse, Esen, Theisen, Mordmuller and Wardemann2013; Speiser et al. Reference Speiser, Utzschneider, Oberle, Munz, Romero and Zehn2014). Conversely, the response to P. vivax appears to be different, with multiple lines of evidence pointing to sustained immunological memory in the absence of concurrent infection, an area of wide interest within the field of immunology.
There is still debate over what constitutes the basis of immunological memory in humans (Crotty and Ahmed, Reference Crotty and Ahmed2004). In regards to malarial infections, it has been suggested that constant antigenic stimulation is required to maintain memory T and B cell populations (Bilsborough et al. Reference Bilsborough, Carlisle and Good1993; Berenzon et al. Reference Berenzon, Schwenk, Letellier, Guebre-Xabier, Williams and Krzych2003), and that dormant hypnozoites may provide this stimulation for P. vivax. In fact, whether antigenic stimulation is required for maintenance of memory cells has been historically controversial (Gray and Skarvall, Reference Gray and Skarvall1988; Schittek and Rajewsky, Reference Schittek and Rajewsky1990; Maruyama et al. Reference Maruyama, Lam and Rajewsky2000; Zinkernagel, Reference Zinkernagel2002). However, a study of the successful smallpox vaccine found this was not the case, and that memory B cells could be maintained for many years without specific antigenic stimulation (Crotty et al. Reference Crotty, Felgner, Davies, Glidewell, Villarreal and Ahmed2003). Interestingly, it has also been demonstrated that memory B cells were still identifiable when plasma antibodies no longer existed (Bauer and Jilg, Reference Bauer and Jilg2006; Ndungu et al. Reference Ndungu, Olotu, Mwacharo, Nyonda, Apfeld, Mramba, Fegan, Bejon and Marsh2012), providing further support for the maintenance of memory cells without antigenic stimulation. Clearly, if there was stimulation, memory B cells would differentiate into antibody secreting cells and hence antibodies should be detectable.
Whilst memory B cells may survive without antigenic stimulation, to maintain circulating antibodies either of three scenarios must exist: memory B cells must be stimulated to differentiate into antibody secreting cells by specific antigen, either through re-infection or from that captured by follicular DCs (Tew et al. Reference Tew, Phipps and Mandel1980; Ochsenbein et al. Reference Ochsenbein, Pinschewer, Sierro, Horvath, Hengartner and Zinkernagel2000); alternatively, memory B cells may be stimulated by some form of polyclonal activation (Bernasconi et al. Reference Bernasconi, Traggiai and Lanzavecchia2002); or, finally, long-lived plasma cells must be generated. Long-lived plasma cells can survive for many years without antigenic stimulation (Manz et al. Reference Manz, Thiel and Radbruch1997, Reference Manz, Lohning, Cassese, Thiel and Radbruch1998; Slifka et al. Reference Slifka, Antia, Whitmire and Ahmed1998), however their location (in the bone marrow) makes them unsuitable for measurement in human studies. The presence of antibodies in the absence of memory B cells is generally considered evidence of long-lived plasma cells, and other indirect evidence exists (Sfikakis et al. Reference Sfikakis, Boletis and Tsokos2005). Apart from the maintenance of antibodies, the generation of memory T cells is also extremely important (Hammarlund et al. Reference Hammarlund, Lewis, Hansen, Strelow, Nelson, Sexton, Hanifin and Slifka2003; Tsai and Yu, Reference Tsai and Yu2014).
In summary, it seems appropriate that where possible, all such measures should be considered as potential surrogates of immunological memory, including the generation of memory T and B cells and the maintenance of antigen-specific antibodies. All such factors will be discussed below in relation to the generation of memory to P. vivax. Secondly, we will also consider whether there is any evidence of dysfunctional memory to P. vivax, as has been suggested for P. falciparum.
Memory responses to P. vivax
Memory (or in some cases, persisting) T cells have been identified in a number of malaria endemic populations. In 1993, Bilsborough et al. identified vivax-specific T cell responses (to epitopes of CSP) in individuals who had not experienced a malaria episode for up to 49 years (Bilsborough et al. Reference Bilsborough, Carlisle and Good1993). This study was conducted in volunteers who no longer lived in a malaria endemic area, but had done so previously for an average of 7 years. A similar result was found the following year in a study conducted in Thailand (Zevering et al. Reference Zevering, Khamboonruang, Rungruengthanakit, Tungviboonchai, Ruengpipattanapan, Bathurst, Barr and Good1994). The authors hypothesized that the maintenance of these P. vivax CSP-specific T cell responses in the absence of infection was due to the presence of hypnozoites, given the same pattern was not found for P. falciparum CSP. Further studies have also identified a greater percentage of memory T cells (largely CD4+) in acute patients, as well as uninfected patients who lived in endemic areas, compared with un-exposed controls. However, it is difficult to determine whether these were (or would be) long-lived responses or not (Jangpatarapongsa et al. Reference Jangpatarapongsa, Sirichaisinthop, Sattabongkot, Cui, Montgomery, Looareesuwan, Troye-Blomberg and Udomsangpetch2006, Reference Jangpatarapongsa, Xia, Fang, Hu, Yuan, Peng, Gao, Sattabongkot, Cui, Li and Udomsangpetch2012; Silva et al. Reference Silva, Lacerda, Fujiwara, Bueno and Braga2013; Zeeshan et al. Reference Zeeshan, Bora and Sharma2013).
Despite the obvious interest from the vaccine community in inducing memory B cells or long-lived plasma cells, this remains an area of limited research in studies of naturally acquired immunity. Wipasa et al. investigated the response to some well-known P. vivax antigens, AMA1, MSP1-19 and DBP, and found that within a group of 26 individuals from Northern Thailand with previous malaria exposure, 35% had memory B cells specific to one or more of these P. vivax antigens (Wipasa et al. Reference Wipasa, Suphavilai, Okell, Cook, Corran, Thaikla, Liewsaree, Riley and Hafalla2010). Requena et al. recently identified an increase in the percentage of atypical memory B cells (active and resting, defined as IgD− CD27− CD21− and IgD− CD27− CD21+, respectively) and active classical memory B cells (IgD CD27+ CD21−) in a malaria exposed population (both P. vivax and P. falciparum) compared with a non-exposed population (Requena et al. Reference Requena, Campo, Umbers, Ome, Wangnapi, Barrios, Robinson, Samol, Rosanas-Urgell, Ubillos, Mayor, Lopez, de Lazzari, Arevalo-Herrera, Fernandez-Becerra, del Portillo, Chitnis, Siba, Bardaji, Mueller, Rogerson, Menendez and Dobano2014). It will be of great interest to the field to follow-up these interesting studies with further longitudinal experiments, including the addition of peripheral blood mononuclear cells stimulated with whole P. vivax parasites or lysate, to determine whether the response is truly specific.
Wipasa et al. also measured P. vivax-specific antibodies, and whilst the overall level of seropositivity was low (25% of volunteers previously exposed to P. vivax, with no significant difference to those with no previous exposure), they found that of those who were positive there was no significant decline in titre over the 12 months of the study (Wipasa et al. Reference Wipasa, Suphavilai, Okell, Cook, Corran, Thaikla, Liewsaree, Riley and Hafalla2010). Other studies have also indicated that there is generally a good maintenance of positive antibody responses to P. vivax antigens (Braga et al. Reference Braga, Fontes and Krettli1998; Park et al. Reference Park, Moon, Yeom, Lim, Sohn, Jung, Cho, Jeon, Ju, Ki, Oh and Choe2001; Lim et al. Reference Lim, Park, Yeom, Lee, Yoo, Oh, Sohn, Bahk and Kim2004; Morais et al. Reference Morais, Soares, Carvalho, Fontes, Krettli and Braga2006; Barbedo et al. Reference Barbedo, Ricci, Jimenez, Cunha, Yazdani, Chitnis, Rodrigues and Soares2007), although in general the titres did decline over time. However, there was significant variation in the amount of time these studies classified as ‘maintenance’, varying from 1 year post-exposure (Park et al. Reference Park, Moon, Yeom, Lim, Sohn, Jung, Cho, Jeon, Ju, Ki, Oh and Choe2001), to up to 30 years (Lim et al. Reference Lim, Park, Yeom, Lee, Yoo, Oh, Sohn, Bahk and Kim2004). This could depend on numerous factors including the antigen studied or the age of the subjects, as well as the underlying level of transmission subjects were exposed to or the number of malarial infections they experienced prior to the study. Conversely, another study found that P. vivax MSP1-specific antibodies declined rapidly over a period of 2–4 months (Soares et al. Reference Soares, da Cunha, Silva, Souza, Del Portillo and Rodrigues1999a), and inhibitory antibodies against DBPII are also known to be short-lived (Ceravolo et al. Reference Ceravolo, Sanchez, Sousa, Guerra, Soares, Braga, McHenry, Adams, Brito and Carvalho2009). This could again be attributed to some of the factors outlined above.
Even though the underlying immunological evidence demonstrating a successful memory response against P. vivax may not yet be clear, a strong epidemiological study supports this finding. In 1991, a malaria elimination program was initiated on Aneityum Island in Vanuatu, and by 1996 P. vivax was considered to be eliminated. However, in 2002 an outbreak occurred, with 77 infections identified from 759 individuals (Kaneko et al. Reference Kaneko, Chaves, Taleo, Kalkoa, Isozumi, Wickremasinghe, Perlmann, Takeo, Tsuboi, Tachibana, Kimura, Bjorkman, Troye-Blomberg, Tanabe and Drakeley2014). The infections showed a clear age-structure, with adults remaining largely P. vivax free. This suggested that those who were born before the elimination strategy begun still retained immunity from exposure prior to 1996. Combining all evidence, it does seem that immunological memory to P. vivax is a strong possibility, however the exact mechanism by which this occurs (and to which antigens this is directed) still needs to be uncovered.
Dysfunctional immunity
Interestingly, there has clearly been an inability of some populations to maintain antibody responses to specific antigens, such as MSP1 or DBP, as mentioned earlier (Soares et al. Reference Soares, da Cunha, Silva, Souza, Del Portillo and Rodrigues1999a, Reference Soares, Oliveira, Souza and Rodriguesb; Ceravolo et al. Reference Ceravolo, Sanchez, Sousa, Guerra, Soares, Braga, McHenry, Adams, Brito and Carvalho2009). One hypothesis is that memory generation to immunodominant antigens is dysfunctional in P. vivax infection, conceivably as an immune defence mechanism or evasion strategy of the parasite. Perhaps surprisingly, this has also been evident for cellular responses. A small study of 33 people in Sri Lanka found that residents who had lived their entire life in a malaria endemic area were less able to respond to stimulation with a soluble extract of P. vivax infected erythrocytes than individuals who did not have life-long exposure but had acquired malaria on a visit to an endemic area (Goonewardene et al. Reference Goonewardene, Carter, Gamage, Del Giudice, David, Howie and Mendis1990). This suggests that life-long exposure had led to immunosuppression in response to P. vivax antigens. Again, these results could be interpreted in another manner: perhaps such ‘immunosuppression’ is required in individuals with life-long exposure in order to not overwhelm the immune response, and would hence then be referred to as immune adaptation. However, in this particular study, the volunteers were all symptomatic, even though they all reported having experienced ‘repeated malaria infections’. This finding contradicts the evidence demonstrating effective immunity against clinical P. vivax symptoms in endemic areas (including those of low-transmission intensity). A similar study more than 10 years later in Brazil identified the same phenomenon: individuals who had been resident for more than 10 years in an endemic area had a lower proliferative response to P. vivax CSP than those with less than 1 year total in an endemic area (Braga et al. Reference Braga, Carvalho, Fontes and Krettli2002b). In this case given only CSP was studied, immunosuppression cannot be as easily assigned; particularly given the antibody response to CSP was comparable between the two groups. Furthermore, as both studies focused on patients with acute malaria, they could be considered biased towards individuals with a poorer immune response to malaria. Clearly, further studies are required, but there does seem to be some evidence of immune dysfunction in particular groups of patients.
CROSS-SPECIES IMMUNITY
Another important aspect of naturally acquired immunity to P. vivax is whether this immunity is species specific, or whether exposure to P. vivax could also protect against P. falciparum (or vice versa). A number of the highly immunogenic P. vivax antigens mentioned so far, such as AMA1, CSP, MSP1 and MSP3, have homologs in P. falciparum. Whilst it is still unknown as to whether cross-species immunity could actually provide functional protection, there has long been evidence of cross-species recognition of immune determinants (Diggs and Sadun, Reference Diggs and Sadun1965).
It has been shown through the use of a competition enzyme-linked immunosorbent assay that cross-reactive antibody responses can be induced by natural infection between Pv and PfMSP5 (Woodberry et al. Reference Woodberry, Minigo, Piera, Hanley, de Silva, Salwati, Kenangalem, Tjitra, Coppel, Price, Anstey and Plebanski2008). The same group subsequently demonstrated that cellular responses specific to PfMSP5 can also be boosted by P. vivax infection (Salwati et al. Reference Salwati, Minigo, Woodberry, Piera, de Silva, Kenangalem, Tjitra, Coppel, Price, Anstey and Plebanski2011). This is supported by other studies where infection with either species seemed to be capable of boosting existing responses to either P. vivax or P. falciparum antigens or crude lysate (del Portillo et al. Reference del Portillo, Levitus, Camargo, Ferreira and Mertens1992; Carvalho et al. Reference Carvalho, Fontes, Fernandes, Marinuzzi and Krettli1997; Chuangchaiya et al. Reference Chuangchaiya, Jangpatarapongsa, Chootong, Sirichaisinthop, Sattabongkot, Pattanapanyasat, Chotivanich, Troye-Blomberg, Cui and Udomsangpetch2010; Wipasa et al. Reference Wipasa, Okell, Sakkhachornphop, Suphavilai, Chawansuntati, Liewsaree, Hafalla and Riley2011). Furthermore, in an area of only P. vivax transmission, P. falciparum antigens were also able to stimulate the immune response in P. vivax infected patients (Jangpatarapongsa et al. Reference Jangpatarapongsa, Xia, Fang, Hu, Yuan, Peng, Gao, Sattabongkot, Cui, Li and Udomsangpetch2012). This would suggest that there are conserved regions between the two species that the human immune system can respond to; however, we do not know whether these are protective determinants. However, two papers have proposed the theory that prior exposure of P. vivax may ameliorate a subsequent course of P. falciparum infection, based largely on the unusual finding that malaria specific mortality was very low in the two populations studied (Gunewardena et al. Reference Gunewardena, Carter and Mendis1994; Maitland et al. Reference Maitland, Williams, Bennett, Newbold, Peto, Viji, Timothy, Clegg, Weatherall and Bowden1996). This hypothesis remains to be experimentally proven.
CONCLUSIONS/FUTURE DIRECTIONS
Gaps in our knowledge of naturally acquired immunity to P. vivax
Whilst many immuno-epidemiological studies have now been undertaken in the hope of understanding naturally acquired immunity or identifying promising vaccine candidates, much remains unknown. We have identified several key questions that remain unanswered:
(i) Are cellular responses induced to pre-erythrocytic antigens, and is this to a greater extent than seen for P. falciparum (given the potential for an extended liver-stage)?
(ii) Do the newly identified innate lymphoid cells respond to P. vivax? If so, could this partly be responsible for the high cytokine response?
(iii) How does immunoregulation influence naturally acquired immunity? What are the roles of regulatory T cells and IL-10, and how do they change with acute compared with asymptomatic infections?
(iv) What role do DCs and phagocytes play in naturally acquired immunity, and are their roles interchangeable?
(v) What other proteins does the parasite use for invasion of RBCs in addition to DBP, and are they targets of naturally induced antibodies?
(vi) What are the targets of antibodies that provide naturally acquired transmission-blocking immunity?
(vii) What constitutes immunological memory to P. vivax and what cell types are involved in generating such a long-lived response? Alternatively, does P. vivax infection lead to atypical B cells and exhausted T cells?
(viii) Are follicular helper T cells essential for naturally acquired immunity, given their importance in generating high-affinity antibodies? Do they contribute to the longevity of immunity?
(ix) Is cross-species protection a true phenomenon, and can this be answered through studies of natural immunity?
Answering such questions might give further input to the overall aim of many immuo-epidemiological studies: defining key targets of long-lasting natural immunity and ultimately identifying correlates of protection. To answer such questions, researchers will need to use the right tools, such as protein arrays, multi-parameter flow cytometry, the generation of P. vivax lysate (from multiple strains) to determine specific responses and the use of samples from well defined, preferentially longitudinal, studies. Furthermore, collaborative efforts will be required to draw all such information together to build a working model of naturally acquired immunity to P. vivax.
Lessons learned: applications in vaccine development and sero-diagnostics
Although many questions surrounding the development and maintenance of naturally acquired immunity remain to be answered, what we know so far gives great promise for the development of a P. vivax vaccine. We know that immunity to clinical infection is acquired faster for P. vivax than for P. falciparum, and that long-lasting immune responses to defined antigens can be induced following P. vivax infection (Bilsborough et al. Reference Bilsborough, Carlisle and Good1993; Zevering et al. Reference Zevering, Khamboonruang, Rungruengthanakit, Tungviboonchai, Ruengpipattanapan, Bathurst, Barr and Good1994; Wipasa et al. Reference Wipasa, Suphavilai, Okell, Cook, Corran, Thaikla, Liewsaree, Riley and Hafalla2010). The difficulty will be choosing what targets to incorporate into a successful vaccine, and until we uncover the mechanisms of naturally acquired immunity, defining what type of immune response we want such a vaccine to induce will be difficult. Furthermore, more effort needs to be placed into understanding naturally acquired immunity to P. vivax in children, rather than adults, given children encompass the target demographic for a malaria vaccine in endemic regions.
Apart from contributing to rational vaccine design, a greater understanding of naturally acquired immunity can also be used to design new sero-diagnostic tools (Cutts et al. Reference Cutts, Powell, Agius, Beeson, Simpson and Fowkes2014). Whilst P. vivax antigens used for targets in vaccines will need to induce long-lived immune responses, those that only induce short-lived antibody responses may prove useful for population-level surveillance. Sero-surveillance has the potential to identify areas of remaining P. vivax transmission, asymptomatic individuals and potentially individuals harbouring hypnozoites, and would be a unique and useful tool for monitoring the success of malaria elimination programs. In order to develop such a tool, detailed information must be learned about antigen-specific antibody responses and their longevity in various geographical regions, transmission settings and age groups.
ACKNOWLEDGEMENTS
We acknowledge Timothy Cole (Office of Research Services, Mahidol University) for the careful review of this manuscript and the National Research Council of Thailand for their support.
FINANCIAL SUPPORT
We acknowledge the financial support from the National Institute of Allergy and Infectious Diseases, USA (NIH grant number 5R01 AI 104822) and the Foundation for Innovative New Diagnostics. IM is supported by a NHMRC Senior Research Fellowship (#1043345).
