INTRODUCTION
The significant morbidity and mortality attributable to Plasmodium infection (World Health Organisation, 2014) underscores the importance of a continued focus on malaria vaccine development. The quest for a vaccine has been fraught with many challenges that are due in part to the complexity of the malaria parasite. Identification and selection of a vaccine candidate is a complicated process with 6 species known to infect humans, multiple life-cycle stages and highly diverse antigen expression (Barry et al. Reference Barry, Schultz, Buckee and Reeder2009). The parasite employs many different mechanisms in its effort to evade the human immune response including antigenic variation, allelic polymorphism and immunomodulation (reviewed in (Stanisic et al. Reference Stanisic, Barry and Good2013)). These evasion strategies may explain in part why vaccine candidates that have performed successfully in animal models and the early-phase human studies have not been able to induce long-lived, protective immunity when evaluated in malaria-endemic areas (Snounou et al. Reference Snounou, Gruner, Muller-Graf, Mazier and Renia2005; Ogutu et al. Reference Ogutu, Apollo, McKinney, Okoth, Siangla, Dubovsky, Tucker, Waitumbi, Diggs, Wittes, Malkin, Leach, Soisson, Milman, Otieno, Holland, Polhemus, Remich, Ockenhouse, Cohen, Ballou, Martin, Angov, Stewart, Lyon, Heppner and Withers2009; Sagara et al. Reference Sagara, Dicko, Ellis, Fay, Diawara, Assadou, Sissoko, Kone, Diallo, Saye, Guindo, Kante, Niambele, Miura, Mullen, Pierce, Martin, Dolo, Diallo, Doumbo, Miller and Saul2009). They may also contribute to the slow and sub-optimal induction of naturally acquired immunity. There is also a poor understanding of how the parasite and the host immune system interact and which immune responses are critical for the induction and maintenance of long-term, protective immunity.
Many vaccines are designed to mimic natural exposure, and thus induce a similar immune response to an infection. However, the immune response of individuals who have developed natural immunity to malaria varies greatly from one person to another, making it very challenging to define a core set of immune correlates of protection. Furthermore, acquired sterile immunity to Plasmodium falciparum infection is not observed (Tran et al. Reference Tran, Li, Doumbo, Doumtabe, Huang, Dia, Bathily, Sangala, Kone, Traore, Niangaly, Dara, Kayentao, Ongoiba, Doumbo, Traore and Crompton2013). Immunity that does develop is acquired slowly and manifests as protection against high-density parasitemia and disease, but not from infection per se. Thus, it is clear that a vaccine acceptable to the public and to health authorities will need to induce a qualitatively and/or quantitatively different immune response to that induced during natural infection, but as previously mentioned the core parameters of natural immunity are largely un-defined. Vaccine research has essentially become a ‘best guess’ exercise based on knowledge of the efficacy of vaccine candidate homologues in rodent models, which is a testing system that can differ considerably from the experimental conditions present in human malaria infections (Wykes and Good, Reference Wykes and Good2009). It is hoped that studies involving deliberate infection of human volunteers (‘controlled human malaria infections’ [CHMI]) with P. falciparum will be more informative to vaccine development. If parasite persistence is required for the maintenance of immune responses, a vaccine that protects against the immunopathology associated with infection (without completely eradicating blood-stage parasites) may be preferable. Understanding mechanisms and targets of naturally acquired immunity will contribute to the rational design of such a malaria vaccine, but other approaches are likely to be required as well.
Immunoglobulin (Ig) transfer studies in humans demonstrated that IgG was a critical component of blood-stage immunity in humans (Cohen et al. Reference Cohen, McGregor and Carrington1961; Sabchareon et al. Reference Sabchareon, Burnouf, Ouattara, Attanath, Bouharoun-Tayoun, Chantavanich, Foucault, Chongsuphajaisiddhi and Druilhe1991). This may work alone or in combination with phagocytic cells and/or complement (Boyle et al. Reference Boyle, Reiling, Feng, Langer, Osier, Aspeling-Jones, Cheng, Stubbs, Tetteh, Conway, McCarthy, Muller, Marsh, Anders and Beeson2015b). There is also evidence from human and rodent studies that antibody-independent mechanisms (Brake et al. Reference Brake, Long and Weidanz1988; Pombo et al. Reference Pombo, Lawrence, Hirunpetcharat, Rzepczyk, Bryden, Cloonan, Anderson, Mahakunkijcharoen, Martin, Wilson, Elliott, Eisen, Weinberg, Saul and Good2002; Elliott et al. Reference Elliott, Kuns and Good2005) can control parasite growth. In blood-stage malaria vaccine development, strategies based on cellular immune responses have been largely ignored. A better understanding of the nature of these responses in malaria infection is required and exciting tools now exist to facilitate the identification of key determinants of protective cellular immunity (Cardoso et al. Reference Cardoso, Roddick, Groves and Doolan2011) as well as the functional and phenotypic characterization of cellular networks induced during malaria infection (Seder et al. Reference Seder, Darrah and Roederer2008; Tran et al. Reference Tran, Samal, Kirkness and Crompton2012). In this review, we examine key aspects of the human cellular immune response to the parasite in experimental human systems and naturally exposed individuals and discuss how this may be used to inform the development of a blood-stage vaccine.
CELL-MEDIATED IMMUNITY TO PLASMODIUM
It is thought that the acquisition of naturally acquired immunity to Plasmodium reflects a complex interplay and balance between effector and regulatory immune responses. An understanding of the nature of these responses in human experimental and natural infection is required to assist vaccine design. For Plasmodium, we know that inflammatory cellular responses are essential for the control of blood-stage parasites. Activation of innate immune cells by Plasmodium, including dendritic cells (DCs), monocytes/macrophages, natural killer (NK) cells and γδ T cells, results in the production of pro-inflammatory cytokines which limit parasite growth and also promote the induction of adaptive CD4+ and CD8+ responses (reviewed in (Korbel et al. Reference Korbel, Finney and Riley2004; Stevenson and Riley, Reference Stevenson and Riley2004; Stevenson and Urban, Reference Stevenson and Urban2006; Inoue et al. Reference Inoue, Niikura, Mineo and Kobayashi2013)) (Fig. 1). If unregulated, this inflammatory response can also contribute to the pathology that is associated with malaria infection (Artavanis-Tsakonas et al. Reference Artavanis-Tsakonas, Tongren and Riley2003; Lamb et al. Reference Lamb, Brown, Potocnik and Langhorne2006; Stevenson et al. Reference Stevenson, Ing, Berretta and Miu2011; Butler et al. Reference Butler, Harris and Blader2013; Perez-Mazliah and Langhorne, Reference Perez-Mazliah and Langhorne2014).
Fig. 1. Cellular immune responses to the blood-stage malaria parasite. Abbreviations: DC, dendritic cell; IL-2, interleukin-2; IL-10, interleukin-10; IL-12, interleukin-12; IFN-γ, interferon-γ; NK cell, natural killer cell; NO, nitric oxide; O2−, oxygen radical; pRBC, parasitized red blood cell; PRR, pattern recognition receptor; TGF-β, transforming growth factor-β; TCR, T cell receptor; TNF, tumour necrosis factor-α. Adapted from Trends In Parasitology, 29(12), DI Stanisic, AE Barry, MF Good, Escaping the Immune System: How the Malaria Parasite Makes Vaccine Development a Challenge, 612–622, 2013, with permission from Elsevier.
Cells such as CD4+, CD8+ and γδ T cells can mediate their functions directly via cytotoxicity (CD8+ and γδ T cells) or indirectly via production of cytokines such as Interferon gamma (IFN-γ) and tumour necrosis factor (TNF), which are essential mediators of the adaptive immune response and the key inflammatory cytokines for controlling the blood-stage of the parasite. Regulation of this inflammatory response requires the production of anti-inflammatory cytokines such as interleukin-10 (IL-10) and transforming growth factor-β (TGF-β), the balance and timing of which are critical for determining disease outcome.
Cytokines are essential mediators of the cellular immune response. IFN-γ is an important component of the immune response against the blood-stage of Plasmodium (reviewed in (McCall and Sauerwein, Reference McCall and Sauerwein2010)) and can be produced by numerous cell types including CD4+, CD8+ and γδ T cells. Production of Plasmodium-specific IFN-γ in vitro is associated with protection in both human experimental and natural infections (Luty et al. Reference Luty, Lell, Schmidt-Ott, Lehman, Luckner, Greve, Matousek, Herbich, Schmid, Migot-Nabias, Deloron, Nussenzweig and Kremsner1999; Dodoo et al. Reference Dodoo, Omer, Todd, Akanmori, Koram and Riley2002; Pombo et al. Reference Pombo, Lawrence, Hirunpetcharat, Rzepczyk, Bryden, Cloonan, Anderson, Mahakunkijcharoen, Martin, Wilson, Elliott, Eisen, Weinberg, Saul and Good2002; D'Ombrain et al. Reference D'Ombrain, Robinson, Stanisic, Taraika, Bernard, Michon, Mueller and Schofield2008; Robinson et al. Reference Robinson, D'Ombrain, Stanisic, Taraika, Bernard, Richards, Beeson, Tavul, Michon, Mueller and Schofield2009; McCall et al. Reference McCall, Hopman, Daou, Maiga, Dara, Ploemen, Nganou-Makamdop, Niangaly, Tolo, Arama, Bousema, van der Meer, van der Ven, Troye-Blomberg, Dolo, Doumbo and Sauerwein2010). It has numerous roles in the innate and adaptive immune responses against Plasmodium (reviewed in (McCall and Sauerwein, Reference McCall and Sauerwein2010)) with its protective effects proposed to be mediated by the activation of phagocytic cells resulting in enhancement of parasite clearance via phagocytosis (Su et al. Reference Su, Fortin, Gros and Stevenson2002; Yoneto et al. Reference Yoneto, Waki, Takai, Tagawa, Iwakura, Mizuguchi, Nariuchi and Yoshimoto2001) and production of antiparasitic nitric oxide and oxygen radicals (Ockenhouse et al. Reference Ockenhouse, Schulman and Shear1984; Su and Stevenson, Reference Su and Stevenson2000; Wang et al. Reference Wang, Liu, Liu, Chen, Zheng, Wang and Cao2009) (Fig. 1). TNF-α is produced by CD4+ T cells and activated monocytes/macrophages and has been implicated in both protection (Kremsner et al. Reference Kremsner, Winkler, Brandts, Wildling, Jenne, Graninger, Prada, Bienzle, Juillard and Grau1995; Mordmuller et al. Reference Mordmuller, Metzger, Juillard, Brinkman, Verweij, Grau and Kremsner1997; Robinson et al. Reference Robinson, D'Ombrain, Stanisic, Taraika, Bernard, Richards, Beeson, Tavul, Michon, Mueller and Schofield2009) and pathology (e.g. Grau et al. Reference Grau, Taylor, Molyneux, Wirima, Vassalli, Hommel and Lambert1989; Kern et al. Reference Kern, Hemmer, Van Damme, Gruss and Dietrich1989; Day et al. Reference Day, Hien, Schollaardt, Loc, Chuong, Chau, Mai, Phu, Sinh, White and Ho1999). It is important in parasite killing and preventing parasite replication (Taverne et al. Reference Taverne, Tavernier, Fiers and Playfair1987) via induction of nitric oxide (Rockett et al. Reference Rockett, Awburn, Cowden and Clark1991; Reference Rockett, Awburn, Aggarwal, Cowden and Clark1992) (Fig. 1). It was noted in a malaria exposed population that production of IFN-γ and TNF increased with age, whereas this age-dependent effect was not observed with IL-10 (Ramharter et al. Reference Ramharter, Winkler, Kremsner, Adegnika, Willheim and Winkler2005). Using age as a surrogate for exposure, IFN-γ and TNF were proposed as cellular correlates of natural immunity.
IL-10 is a regulatory cytokine that is produced by numerous cell types including CD4+ T cells and T regulatory cells (Tregs) (reviewed in (Freitas do Rosario and Langhorne, Reference Freitas do Rosario and Langhorne2012)). It can suppress DCs and macrophages, dampening the inflammatory immune response and block proliferation of and cytokine production by T cells (Joss et al. Reference Joss, Akdis, Faith, Blaser and Akdis2000; Moore et al. Reference Moore, de Waal Malefyt, Coffman and O'Garra2001; Maynard and Weaver, Reference Maynard and Weaver2008). This cytokine is thought to play an important role in Plasmodium infection, due to its ability to affect both innate and adaptive immune responses (reviewed in (Freitas do Rosario and Langhorne, Reference Freitas do Rosario and Langhorne2012)) (Fig. 1). While rodent studies have demonstrated a clear role for IL-10 in controlling the inflammatory response associated with malaria, its precise role in the regulation of immunopathology in humans is unclear as data from studies are conflicting. Associations between lower levels of IL-10 and increased risk of severe malaria syndromes (e.g. Kurtzhals et al. Reference Kurtzhals, Adabayeri, Goka, Akanmori, Oliver-Commey, Nkrumah, Behr and Hviid1998; Othoro et al. Reference Othoro, Lal, Nahlen, Koech, Orago and Udhayakumar1999; Awandare et al. Reference Awandare, Goka, Boeuf, Tetteh, Kurtzhals, Behr and Akanmori2006; Thuma et al. Reference Thuma, van Dijk, Bucala, Debebe, Nekhai, Kuddo, Nouraie, Weiss and Gordeuk2011) and between higher IL-10 levels and increased disease severity (e.g. Awandare et al. Reference Awandare, Goka, Boeuf, Tetteh, Kurtzhals, Behr and Akanmori2006; Ayimba et al. Reference Ayimba, Hegewald, Segbena, Gantin, Lechner, Agosssou, Banla and Soboslay2011; Stanisic et al. Reference Stanisic, Cutts, Eriksson, Fowkes, Rosanas-Urgell, Siba, Laman, Davis, Manning, Mueller and Schofield2014) have been observed. Rodent and human studies suggest that IL-10 production can impede effective clearance of the parasite (Hugosson et al. Reference Hugosson, Montgomery, Premji, Troye-Blomberg and Bjorkman2004; Weidanz et al. Reference Weidanz, Batchelder, Flaherty, LaFleur, Wong and van der Heyde2005; Couper et al. Reference Couper, Blount, Wilson, Hafalla, Belkaid, Kamanaka, Flavell, de Souza and Riley2008).
Rodent models of malaria have been used extensively to characterize and understand the role of cellular immune responses in protection, immunopathology and immune regulation (Lamb et al. Reference Lamb, Brown, Potocnik and Langhorne2006; Wykes and Good, Reference Wykes and Good2009; Perez-Mazliah and Langhorne, Reference Perez-Mazliah and Langhorne2014). While these models have allowed insight into the role of cellular immune responses associated with Plasmodium infection, they must ultimately be investigated in humans.
Investigations into the role of the Plasmodium-specific immune responses in humans have often involved cross-sectional studies that are sub-optimal for determining protective associations; few have employed a longitudinal study design to investigate the relationship between these immune responses and prospective risk of disease/re-infection (Peyron et al. Reference Peyron, Vuillez, Barbe, Boudin, Picot and Ambroise-Thomas1990; Mshana et al. Reference Mshana, Boulandi, Mshana, Mayombo and Mendome1991; Riley et al. Reference Riley, Morris-Jones, Blackman, Greenwood and Holder1993; Dodoo et al. Reference Dodoo, Omer, Todd, Akanmori, Koram and Riley2002; Le Hesran et al. Reference Le Hesran, Fievet, Thioulouse, Personne, Maubert, M'Bidias, Etye'ale, Cot and Deloron2006; D'Ombrain et al. Reference D'Ombrain, Robinson, Stanisic, Taraika, Bernard, Michon, Mueller and Schofield2008; Robinson et al. Reference Robinson, D'Ombrain, Stanisic, Taraika, Bernard, Richards, Beeson, Tavul, Michon, Mueller and Schofield2009).
Furthermore, many studies have been restricted to evaluating a limited number of immunological parameters and this has impacted on our ability to assess effective cellular immune responses. Today, advances in immunological methods to assess immune responses (Chattopadhyay et al. Reference Chattopadhyay, Hogerkorp and Roederer2008, Reference Chattopadhyay, Gierahn, Roederer and Love2014; Seder et al. Reference Seder, Darrah and Roederer2008) and access to CHMI (McCarthy et al. Reference McCarthy, Sekuloski, Griffin, Elliott, Douglas, Peatey, Rockett, O'Rourke, Marquart, Hermsen, Duparc, Mohrle, Trenholme and Humberstone2011; Engwerda et al. Reference Engwerda, Minigo, Amante and McCarthy2012) could provide important insights for vaccine approaches.
Immune cells relevant to malaria parasite control
DCs
Upon initiation of the blood stage of infection, cells of the innate immune system, such as DCs, encounter and recognize the parasitized red blood cells (pRBC) via pattern recognition receptors. This induces DC maturation resulting in IL-12 secretion, antigen presentation to CD4+ T cells and activation of NK cells (via IL-12 and contact with pRBC). DCs can also interact directly with γδ T cells (Inoue et al. Reference Inoue, Niikura, Takeo, Mineo, Kawakami, Uchida, Kamiya and Kobayashi2012) and cross-present antigen to CD8+ T cells (Lundie et al. Reference Lundie, de Koning-Ward, Davey, Nie, Hansen, Lau, Mintern, Belz, Schofield, Carbone, Villadangos, Crabb and Heath2008). Thus, DCs play a critical role in antigen presentation and the induction of adaptive cellular immune responses.
DCs are a heterogeneous cell type. Rodent models demonstrate that DCs play a critical role in inducing immunity to malaria through parasite clearance (Borges da Silva et al. Reference Borges da Silva, Fonseca, Cassado Ados, Machado de Salles, de Menezes, Langhorne, Perez, Cuccovia, Ryffel, Barreto, Marinho, Boscardin, Alvarez, D'Imperio-Lima and Tadokoro2015) and induction of CD4+ T cell responses (deWalick et al. Reference deWalick, Amante, McSweeney, Randall, Stanley, Haque, Kuns, MacDonald, Hill and Engwerda2007; Borges da Silva et al. Reference Borges da Silva, Fonseca, Cassado Ados, Machado de Salles, de Menezes, Langhorne, Perez, Cuccovia, Ryffel, Barreto, Marinho, Boscardin, Alvarez, D'Imperio-Lima and Tadokoro2015). They also indirectly contribute to immunopathology (deWalick et al. Reference deWalick, Amante, McSweeney, Randall, Stanley, Haque, Kuns, MacDonald, Hill and Engwerda2007) via the activation of CD4+ T cells. Numerous studies investigating the effect of the malaria parasite on DCs have been conducted in rodent models, but have reported contradictory findings regarding the impact of parasites on DC maturation and function (reviewed in (Wykes and Good, Reference Wykes and Good2008; Stevenson et al. Reference Stevenson, Ing, Berretta and Miu2011)). Multiple factors may have contributed to this including differences in parasite and mouse strains used and DC populations examined (reviewed in (Stanisic et al. Reference Stanisic, Barry and Good2013) and (Wykes and Good, Reference Wykes and Good2008)). Data suggests that only infections with virulent, lethal rodent parasites result in inhibition of DC function (Wykes et al. Reference Wykes, Liu, Beattie, Stanisic, Stacey, Smyth, Thomas and Good2007a). In non-lethal models, DCs that interacted with the parasite were activated and capable of stimulating T cell proliferation, secreting IL-12 and initiating production of pro-inflammatory cytokines, whereas in lethal models DCs were unable to secrete IL-12 or prime effector T cells (Wykes et al. Reference Wykes, Liu, Beattie, Stanisic, Stacey, Smyth, Thomas and Good2007a, Reference Wykes, Liu, Jiang, Hirunpetcharat and Good2007b). This differential impact on DC function resulted in either the induction of an immune response that controlled parasite growth, mediating protection and pathology (non-lethal model) or one that suppressed both pathology and induction of protective immunity (lethal model). Rodent models also suggest that modulation of DC function may be dose-dependent with high parasitemias resulting in DCs that were refractory to IL-12 and TNF production following toll-like receptor stimulation, instead producing IL-10 (Perry et al. Reference Perry, Olver, Burnett and Avery2005). This resulted in a switch from an IFN-γ-mediated T cell response at low parasitemias to an IL-10 dominated T cell response at high parasitemias. Thus, parasite load and parasite dose (in the instance of a vaccine based on whole blood-stage parasites) may be critical for inducing an effective immune response.
Few studies have investigated interactions between human parasites and DCs. In vitro studies have demonstrated that P. falciparum negatively modulates DC function (Urban et al. Reference Urban, Ferguson, Pain, Willcox, Plebanski, Austyn and Roberts1999, Reference Urban, Willcox and Roberts2001b; Giusti et al. Reference Giusti, Urban, Frascaroli, Albrecht, Tinti, Troye-Blomberg and Varani2011) and this effect is dose-dependent (Elliott et al. Reference Elliott, Spurck, Dodin, Maier, Voss, Yosaatmadja, Payne, McFadden, Cowman, Rogerson, Schofield and Brown2007). Ex vivo studies have also demonstrated modulation of function (Goncalves et al. Reference Goncalves, Salmazi, Santos, Bastos, Rocha, Boscardin, Silber, Kallas, Ferreira and Scopel2010; Arama et al. Reference Arama, Giusti, Bostrom, Dara, Traore, Dolo, Doumbo, Varani and Troye-Blomberg2011; Pinzon-Charry et al. Reference Pinzon-Charry, Woodberry, Kienzle, McPhun, Minigo, Lampah, Kenangalem, Engwerda, Lopez, Anstey and Good2013) and differences in frequencies of DC sub-types in individuals with malaria (Urban et al. Reference Urban, Mwangi, Ross, Kinyanjui, Mosobo, Kai, Lowe, Marsh and Roberts2001a; Reference Urban, Cordery, Shafi, Bull, Newbold, Williams and Marsh2006; Pichyangkul et al. Reference Pichyangkul, Yongvanitchit, Kum-arb, Hemmi, Akira, Krieg, Heppner, Stewart, Hasegawa, Looareesuwan, Shanks and Miller2004; Jangpatarapongsa et al. Reference Jangpatarapongsa, Chootong, Sattabongkot, Chotivanich, Sirichaisinthop, Tungpradabkul, Hisaeda, Troye-Blomberg, Cui and Udomsangpetch2008; 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). It is unknown whether these changes in peripheral blood DCs reflect what is occurring in the lymphoid organs and how this impacts on the development/maintenance of protective immunity. Reduced/altered DC counts from peripheral blood could reflect increased migration to lymphoid organs and this is supported by the increased expression of chemokine receptors on human DCs in vitro following exposure to pRBC (Pichyangkul et al. Reference Pichyangkul, Yongvanitchit, Kum-arb, Hemmi, Akira, Krieg, Heppner, Stewart, Hasegawa, Looareesuwan, Shanks and Miller2004; Giusti et al. Reference Giusti, Urban, Frascaroli, Albrecht, Tinti, Troye-Blomberg and Varani2011). Apoptosis could also explain this phenomenon and this has been observed both in vitro (Elliott et al. Reference Elliott, Spurck, Dodin, Maier, Voss, Yosaatmadja, Payne, McFadden, Cowman, Rogerson, Schofield and Brown2007) and in vivo in recipients of a blood-stage experimental P. falciparum infection at sub-patent parasite levels (i.e. below the levels detected by microscopy) (Woodberry et al. Reference Woodberry, Pinzon-Charry, Piera, Panpisutchai, Engwerda, Doolan, Salwati, Kenangalem, Tjitra, Price, Good and Anstey2009). The reduced number of DCs observed in the peripheral blood of these individuals was associated with increasing parasite load. DC apoptosis was also observed in individuals from endemic areas infected with P. falciparum and Plasmodium vivax, and this was mediated by IL-10 (Pinzon-Charry et al. Reference Pinzon-Charry, Woodberry, Kienzle, McPhun, Minigo, Lampah, Kenangalem, Engwerda, Lopez, Anstey and Good2013). Functionality of the DCs was recovered following effective antimalarial drug treatment (Pinzon-Charry et al. Reference Pinzon-Charry, Woodberry, Kienzle, McPhun, Minigo, Lampah, Kenangalem, Engwerda, Lopez, Anstey and Good2013). This holds implications for timing of administration of routine vaccines.
When comparing individuals naturally infected with P. vivax or P. falciparum, absolute numbers and proportions of circulating DCs were affected similarly (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). In these studies, impaired maturation of DCs was observed for a proportion of Plasmodium mono-infections (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). Further studies might enable a clearer understanding of whether human Plasmodium species differentially impact DC function.
Given the role that DCs play in T cell activation, it is likely that malaria infection contributes to sub-optimal induction of naturally acquired immunity via this mechanism. The observation that parasites, even at sub-patent levels, can compromise DC function is of great relevance for malaria and other vaccination programs. It is likely that clearance of parasites prior to vaccine administration may be required for optimal responses.
T cells
T cells from individuals resident in malaria-endemic areas respond to the malaria parasite in vitro (e.g. Troye-Blomberg et al. Reference Troye-Blomberg, Romero, Patarroyo, Bjorkman and Perlmann1984; Riley et al. Reference Riley, Jepsen, Andersson, Otoo and Greenwood1988; Mshana et al. Reference Mshana, Boulandi, Mayombo and Mendome1993; D'Ombrain et al. Reference D'Ombrain, Robinson, Stanisic, Taraika, Bernard, Michon, Mueller and Schofield2008; Robinson et al. Reference Robinson, D'Ombrain, Stanisic, Taraika, Bernard, Richards, Beeson, Tavul, Michon, Mueller and Schofield2009; McCall et al. Reference McCall, Hopman, Daou, Maiga, Dara, Ploemen, Nganou-Makamdop, Niangaly, Tolo, Arama, Bousema, van der Meer, van der Ven, Troye-Blomberg, Dolo, Doumbo and Sauerwein2010; Stanisic et al. Reference Stanisic, Cutts, Eriksson, Fowkes, Rosanas-Urgell, Siba, Laman, Davis, Manning, Mueller and Schofield2014). However, parasite-specific T cells are also detected in non-exposed individuals (Jones et al. Reference Jones, Hickling, Targett and Playfair1990; Currier et al. Reference Currier, Sattabongkot and Good1992), apparently due to cross-reactivity with environmental organisms. These T cells are CD45RO+ (memory phenotype) (Jones et al. Reference Jones, Hickling, Targett and Playfair1990; Currier et al. Reference Currier, Sattabongkot and Good1992) and can inhibit parasite growth in vitro (Fell et al. Reference Fell, Currier and Good1994), with both CD4+ and/or CD8+ T cells mediating this inhibition.
CD4+ T cells
Early rodent studies demonstrated that CD4+ T cells can mediate protective immunity (Grun and Weidanz, Reference Grun and Weidanz1983; Brake et al. Reference Brake, Long and Weidanz1988; Langhorne et al. Reference Langhorne, Gillard, Simon, Slade and Eichmann1989; Amante and Good, Reference Amante and Good1997; Elliott et al. Reference Elliott, Kuns and Good2005). They can also contribute to pathology, including anaemia (Hirunpetcharat et al. Reference Hirunpetcharat, Finkelman, Clark and Good1999) and the development of cerebral malaria (Amante et al. Reference Amante, Haque, Stanley, Rivera Fde, Randall, Wilson, Yeo, Pieper, Crabb, de Koning-Ward, Lundie, Good, Pinzon-Charry, Pearson, Duke, McManus, Loukas, Hill and Engwerda2010). It is thought that they act either as helper cells for antibody production or via the cytokines they secrete and the cells (e.g. macrophages) that are activated via these cytokines. There is uncertainty as to how these cells are activated and subsequently target parasites, as mature red cells do not express major histocompatibility complex (MHC) molecules. It is likely that CD4+ T cells are activated by professional antigen-presenting cells (APCs, e.g. DCs) that have processed antigen and then act via the production of cytokines in the spleen which activate macrophages to destroy parasites directly by engulfment or by small inflammatory molecules such as nitric oxide and oxygen radicals (Fig. 1) (reviewed in (Engwerda et al. Reference Engwerda, Beattie and Amante2005)). There are multiple types of CD4+ T cells involved in parasite control, and they can be categorized by the cytokines that they secrete. The traditional model is of the biphasic CD4+ T cell response which consists of two T cell populations (Th1, Th2) with distinct cytokine production patterns activated sequentially during infection (Langhorne et al. Reference Langhorne, Gillard, Simon, Slade and Eichmann1989; Mosmann and Coffman, Reference Mosmann and Coffman1989). It is now appreciated that multiple cell types may be involved including Th17 cells, Th22 cells, follicular helper T cells (which are a critical CD4+ T helper subset for antibody production) and multi-functional T cells such as Th1 cells which are able to secrete multiple cytokines e.g. IFN-γ and IL-10 (reviewed in (Perez-Mazliah and Langhorne, Reference Perez-Mazliah and Langhorne2014)).
Parasite-specific CD4+ T cell responses have been detected in both experimentally infected and naturally exposed individuals (e.g. Troye-Blomberg et al. Reference Troye-Blomberg, Romero, Patarroyo, Bjorkman and Perlmann1984; Mshana et al. Reference Mshana, Boulandi, Mayombo and Mendome1993; Winkler et al. Reference Winkler, Willheim, Baier, Schmid, Aichelburg, Graninger and Kremsner1998; Pombo et al. Reference Pombo, Lawrence, Hirunpetcharat, Rzepczyk, Bryden, Cloonan, Anderson, Mahakunkijcharoen, Martin, Wilson, Elliott, Eisen, Weinberg, Saul and Good2002; D'Ombrain et al. Reference D'Ombrain, Robinson, Stanisic, Taraika, Bernard, Michon, Mueller and Schofield2008; Robinson et al. Reference Robinson, D'Ombrain, Stanisic, Taraika, Bernard, Richards, Beeson, Tavul, Michon, Mueller and Schofield2009; Todryk et al. Reference Todryk, Walther, Bejon, Hutchings, Thompson, Urban, Porter and Hill2009b; McCall et al. Reference McCall, Hopman, Daou, Maiga, Dara, Ploemen, Nganou-Makamdop, Niangaly, Tolo, Arama, Bousema, van der Meer, van der Ven, Troye-Blomberg, Dolo, Doumbo and Sauerwein2010; Stanisic et al. Reference Stanisic, Cutts, Eriksson, Fowkes, Rosanas-Urgell, Siba, Laman, Davis, Manning, Mueller and Schofield2014). Many studies in humans have focused on trying to understand the dual role that these cells may play in the immune response to the parasite – being responsible for both control of the parasite and induction of immunopathology via production of cytokines and chemokines.
Following multiple sub-patent experimental infections with P. falciparum, each curtailed by drug treatment, individuals were resistant to re-infection and this was associated with robust CD4+ and CD8+ proliferative T cell responses, production of IFN-γ and nitric oxide in the absence of antibodies (Pombo et al. Reference Pombo, Lawrence, Hirunpetcharat, Rzepczyk, Bryden, Cloonan, Anderson, Mahakunkijcharoen, Martin, Wilson, Elliott, Eisen, Weinberg, Saul and Good2002). While residual drug may have contributed to some of the protection observed (Edstein et al. Reference Edstein, Kotecka, Anderson, Pombo, Kyle, Rieckmann and Good2005), it was of interest that low doses of malaria parasites were able to induce cellular immune responses in the absence of antibodies. In malaria-endemic areas, chemoprophylaxis (intermittent preventive treatment (IPT)), insecticide treated bed bets and partially effective antisporozoite vaccines may also result in similarly attenuated blood stage infections (Sutherland et al. Reference Sutherland, Drakeley and Schellenberg2007). IPT/continuous chemoprophylaxis aims to reduce the incidence of high-density parasitemias and clinical disease, but not necessarily low grade blood-stage infection. Studies have investigated whether these interventions negatively impact on acquisition of immunity due to reduced parasite exposure. The consequence of this would be in disease rebound following termination of treatment. Fortnightly administration of Maloprim for 3 years was associated with increased cellular responses (lymphocyte proliferation to soluble parasite antigen and IFN-γ production) and lower antibody responses in children who received prophylaxis compared with placebo (Otoo et al. Reference Otoo, Riley, Menon, Byass and Greenwood1989). Furthermore, in the 12 month follow-up period, there was no increase in the frequency of clinical episodes of malaria in children who received chemoprophylaxis compared with controls. Maintenance of parasites at low levels may have allowed these children to more rapidly develop T cell immunity, compared with children who were exposed to high or chronic parasitemia. However, a more recent study, using samples from children enrolled in an IPT study to assess the impact of sulfadoxine–pyrimethamine (SP) on malaria incidence, observed overall, no difference in the levels of plasma cytokines and chemokines or in the proportion of CD3+ T cells producing IFN-γ, IL-10 or IL-4 between the different treatment groups (Quelhas et al. Reference Quelhas, Puyol, Quinto, Nhampossa, Serra-Casas, Macete, Aide, Sanz, Aponte, Doolan, Alonso, Menendez and Dobano2012). The SP treatment was administered at 3, 4 and 9 months of age and had only a moderate effect on the incidence of clinical malaria (Macete et al. Reference Macete, Aide, Aponte, Sanz, Mandomando, Espasa, Sigauque, Dobano, Mabunda, DgeDge, Alonso and Menendez2006), which may explain the differential impact on the cellular immune response between the two studies. It could also be explained by differences in the immunological assays used and the age of the study participants. Overall, these data do support the induction of a protective cellular immune response through exposure to low doses of parasite.
Multi-parametric flow cytometry for immunophenotyping of peripheral blood mononuclear cells (PBMCs) allows the simultaneous assessment of the phenotype, magnitude and functional characteristics of CD4+ T cells, for a comprehensive analysis of the response to infection and vaccination (Perfetto et al. Reference Perfetto, Chattopadhyay and Roederer2004; Seder et al. Reference Seder, Darrah and Roederer2008). Antigen-specific CD4+ T cells have functional diversity; thus assessment of cells via a single parameter may be inadequate in trying to identify immune correlates of protection. For example, sterilizing immunity following vaccination with whole P. falciparum sporozoites is associated with the induction of IFN-γ/IL-2/TNF and IFN-γ/IL-2 co-producing CD4+ T cells (Roestenberg et al. Reference Roestenberg, McCall, Hopman, Wiersma, Luty, van Gemert, van de Vegte-Bolmer, van Schaijk, Teelen, Arens, Spaarman, de Mast, Roeffen, Snounou, Renia, van der Ven, Hermsen and Sauerwein2009, Reference Roestenberg, Teirlinck, McCall, Teelen, Makamdop, Wiersma, Arens, Beckers, van Gemert, van de Vegte-Bolmer, van der Ven, Luty, Hermsen and Sauerwein2011).
Differences in the percentage of multi-functional T cells between urban Kenyan and European donors compared with rural Kenyan donors suggests that an individual's antigenic experience has the capacity to influence the function of CD4+ T cells (Roetynck et al. Reference Roetynck, Olotu, Simam, Marsh, Stockinger, Urban and Langhorne2013). Thirty-two distinct subsets of CD4+ T cells were identified based on varying chemokine receptors, activation markers and cytokine profiles, which underscores the inherent complexity of examining cellular responses. A higher frequency of IFN-γ/IL-10 producers was observed in donors from rural Kenya, suggesting that exposure to the malaria parasite may impact on the expansion of specific sub-populations (Roetynck et al. Reference Roetynck, Olotu, Simam, Marsh, Stockinger, Urban and Langhorne2013). This study illustrates that detailed phenotypic and functional profiling may assist in identifying responses that could serve as biomarkers of protection or susceptibility to Plasmodium-associated disease. Establishing immunological correlates of immunity could inform vaccine design and assist in vaccine evaluation.
In children resident in endemic areas, a number of effector phenotypes of P. falciparum-specific CD4+ cells have been identified including IFN-γ/IL-10 CD4+ T cells that also expressed the Th1 transcription factor T-bet (Jagannathan et al. Reference Jagannathan, Eccles-James, Bowen, Nankya, Auma, Wamala, Ebusu, Muhindo, Arinaitwe, Briggs, Greenhouse, Tappero, Kamya, Dorsey and Feeney2014a). This sub-population dominated the T cell response in highly exposed children, was induced following recent malaria infection, but was not associated with protection against future episodes. In less exposed children, the CD4+ T cell response was more inflammatory, dominated by TNF-producing cells and the absence of these cells was associated with asymptomatic infection.
A further study showed that the effector phenotype of CD4+ T cells is influenced by both age and transmission intensity (Boyle et al. Reference Boyle, Jagannathan, Bowen, McIntyre, Vance, Farrington, Greenhouse, Nankya, Rek, Katureebe, Arinaitwe, Dorsey, Kamya and Feeney2015a). In high transmission settings, CD4+ T cells from clinically immune adults produced inflammatory cytokines (IFN-γ, TNF and/or IL-2) while the response in susceptible children was dominated by IL-10 and IFN-γ/IL-10 producing CD4+ T cells (Boyle et al. Reference Boyle, Jagannathan, Bowen, McIntyre, Vance, Farrington, Greenhouse, Nankya, Rek, Katureebe, Arinaitwe, Dorsey, Kamya and Feeney2015a). Children and adults from low transmission settings had a similar response to adults from a high transmission setting. Together these data suggest that the IL-10 response is driven by recent/persistent exposure to high levels of parasite. The role of these IL-10 secreting cells is ambiguous. Rodent studies suggest that the IL-10 from the IFN-γ/IL-10 co-producing cells may function in an immunomodulatory capacity by limiting pathology (Anderson et al. Reference Anderson, Oukka, Kuchroo and Sacks2007; Jankovic et al. Reference Jankovic, Kullberg, Feng, Goldszmid, Collazo, Wilson, Wynn, Kamanaka, Flavell and Sher2007; Freitas do Rosario and Langhorne, Reference Freitas do Rosario and Langhorne2012), possibly at the expense of parasite clearance, although IL-10 producing CD4+ T cells were not associated prospectively with protection from symptomatic disease (Jagannathan et al. Reference Jagannathan, Eccles-James, Bowen, Nankya, Auma, Wamala, Ebusu, Muhindo, Arinaitwe, Briggs, Greenhouse, Tappero, Kamya, Dorsey and Feeney2014a).
A longitudinal study in Mali may provide insight into the role of these multi-functional CD4+ T cells in repeatedly exposed children who have not acquired a broad, protective repertoire of antibodies, yet are able to remain afebrile and control parasite replication (Portugal et al. Reference Portugal, Moebius, Skinner, Doumbo, Doumtabe, Kone, Dia, Kanakabandi, Sturdevant, Virtaneva, Porcella, Li, Doumbo, Kayentao, Ongoiba, Traore and Crompton2014). It was observed that clinical malaria impacted on the children's PBMCs such that upon P. falciparum re-exposure, there was down-regulation of pro-inflammatory cytokines such as IL-6 and up-regulation of immune responses involved in the control of inflammation (e.g. CD4+CD25+Foxp3- T cell derived IL-10) and parasite clearance (e.g. CD4+ T cell derived IFN-γ) (Portugal et al. Reference Portugal, Moebius, Skinner, Doumbo, Doumtabe, Kone, Dia, Kanakabandi, Sturdevant, Virtaneva, Porcella, Li, Doumbo, Kayentao, Ongoiba, Traore and Crompton2014). A significant proportion of the P. falciparum IL-10 producing CD4+ T cells also produced IFN-γ and/or TNF. The P. falciparum-inducible IL-10 production appeared to be exposure-dependent, remaining partially upregulated only in children with ongoing asymptomatic infection. Overall, these data suggest how P. falciparum exposure may induce immunoregulatory responses that dampen inflammation and enhance antiparasite effector mechanisms. Whether these immunoregulatory cells would interfere with induction of robust effector immune responses to a malaria vaccine is unknown.
CD8+ T cells
There is a paucity of data associated with the role of CD8+ T cells in blood stage immunity. Recent data from rodent models suggest they may play a role in protective blood-stage immunity (Lundie et al. Reference Lundie, de Koning-Ward, Davey, Nie, Hansen, Lau, Mintern, Belz, Schofield, Carbone, Villadangos, Crabb and Heath2008; Horne-Debets et al. Reference Horne-Debets, Faleiro, Karunarathne, Liu, Lineburg, Poh, Grotenbreg, Hill, MacDonald, Good, Renia, Ahmed, Sharpe and Wykes2013; Imai et al. Reference Imai, Shen, Chou, Duan, Tu, Tetsutani, Moriya, Ishida, Hamano, Shimokawa, Hisaeda and Himeno2010, Reference Imai, Ishida, Suzue, Taniguchi, Okada, Shimokawa and Hisaeda2015) as well as pathology (Safeukui et al. Reference Safeukui, Gomez, Adelani, Burte, Afolabi, Akondy, Velazquez, Holder, Tewari, Buffet, Brown, Shokunbi, Olaleye, Sodeinde, Kazura, Ahmed, Mohandas, Fernandez-Reyes and Haldar2015). They contribute to parasite clearance during acute malaria and the control of chronic disease (Horne-Debets et al. Reference Horne-Debets, Faleiro, Karunarathne, Liu, Lineburg, Poh, Grotenbreg, Hill, MacDonald, Good, Renia, Ahmed, Sharpe and Wykes2013) with protection mediated predominantly by IFN-γ (Imai et al. Reference Imai, Shen, Chou, Duan, Tu, Tetsutani, Moriya, Ishida, Hamano, Shimokawa, Hisaeda and Himeno2010; Horne-Debets et al. Reference Horne-Debets, Faleiro, Karunarathne, Liu, Lineburg, Poh, Grotenbreg, Hill, MacDonald, Good, Renia, Ahmed, Sharpe and Wykes2013) which activates phagocytic cells and facilitates uptake of pRBC (Imai et al. Reference Imai, Shen, Chou, Duan, Tu, Tetsutani, Moriya, Ishida, Hamano, Shimokawa, Hisaeda and Himeno2010) (Fig. 1). CD8+ T cells generated in response to rodent parasite infection express perforin (Imai et al. Reference Imai, Shen, Chou, Duan, Tu, Tetsutani, Moriya, Ishida, Hamano, Shimokawa, Hisaeda and Himeno2010) and granzyme B (Imai et al. Reference Imai, Shen, Chou, Duan, Tu, Tetsutani, Moriya, Ishida, Hamano, Shimokawa, Hisaeda and Himeno2010; Horne-Debets et al. Reference Horne-Debets, Faleiro, Karunarathne, Liu, Lineburg, Poh, Grotenbreg, Hill, MacDonald, Good, Renia, Ahmed, Sharpe and Wykes2013) indicating cytotoxic potential (Fig. 1). Killing of pRBC would, however, be restricted to infected reticulocytes and erythroblasts, which express the MHC Class I molecules required for recognition by CD8+ T cells. Recently, a novel antiparasite mechanism mediated by CD8+ T cells has been identified. Following interaction with infected erythroblasts via a FasL-dependent mechanism, CD8+ T cells induce the externalization of phosphatidylserine on these cells, enhancing phagocytic uptake of infected cells (Imai et al. Reference Imai, Ishida, Suzue, Taniguchi, Okada, Shimokawa and Hisaeda2015).
Induction of specific CD8+ T cell responses is observed in malaria-endemic areas (Troye-Blomberg et al. Reference Troye-Blomberg, Romero, Patarroyo, Bjorkman and Perlmann1984; Ramharter et al. Reference Ramharter, Winkler, Kremsner, Adegnika, Willheim and Winkler2005). Also, the induction of proliferative CD8+ T cell responses was observed in malaria-naïve volunteers, following experimental infection (Pombo et al. Reference Pombo, Lawrence, Hirunpetcharat, Rzepczyk, Bryden, Cloonan, Anderson, Mahakunkijcharoen, Martin, Wilson, Elliott, Eisen, Weinberg, Saul and Good2002). The assumption that CD8+ T cells do not contribute to protective blood-stage immunity has contributed to the lack of studies in humans. Functional studies with P. vivax, which infects human reticulocytes and which express MHC Class I, have not been undertaken. Ex vivo phenotypic profiling of CD8+ T cells from P. vivax infected patients demonstrated CD8+ memory T cells producing IFN-γ, TNF and IL-10; infection was associated with a quantitative reduction in CD8+ memory T cell subsets (Hojo-Souza et al. Reference Hojo-Souza, Pereira, Passos, Gazzinelli-Guimaraes, Cardoso, Tada, Zanini, Bartholomeu, Fujiwara and Bueno2015). These evaluations were not antigen-specific and comparative CD4+ T cell profiling was not undertaken. Examination of the interactions between P. vivax and CD8+ T cells could identify novel correlates of P. vivax-specific immunity which may inform development of a species-specific malaria vaccine.
γδ T cells
γδ T cells are key producers of IFN-γ during a blood-stage infection. They function early in the immune response, can directly recognize pathogens through MHC-independent mechanisms (Morita et al. Reference Morita, Beckman, Bukowski, Tanaka, Band, Bloom, Golan and Brenner1995) and play a critical role in activating the adaptive immune system via production of cytokines. It has been shown that γδ T cell clones expressing the Vγ9 and Vδ2 chains can inhibit the growth of P. falciparum in vitro (Elloso et al. Reference Elloso, van der Heyde, vande Waa, Manning and Weidanz1994; Troye-Blomberg et al. Reference Troye-Blomberg, Worku, Tangteerawatana, Jamshaid, Soderstrom, Elghazali, Moretta, Hammarstrom and Mincheva-Nilsson1999). A recent study has elucidated that this inhibition occurs via a T cell receptor (TCR)-dependent mechanism following recognition of either extracellular or intracellular parasites (Costa et al. Reference Costa, Loizon, Guenot, Mocan, Halary, de Saint-Basile, Pitard, Dechanet-Merville, Moreau, Troye-Blomberg, Mercereau-Puijalon and Behr2011). It specifically targets the merozoite stage of the parasite with granulysin, not perforin, the mediator of this inhibition (Costa et al. Reference Costa, Loizon, Guenot, Mocan, Halary, de Saint-Basile, Pitard, Dechanet-Merville, Moreau, Troye-Blomberg, Mercereau-Puijalon and Behr2011) (Fig. 1).
Rodent studies show that γδ T cells increase in frequency during malaria infection and are critical in controlling the acute and chronic phases of Plasmodium chabaudi infection (Seixas and Langhorne, Reference Seixas and Langhorne1999; Weidanz et al. Reference Weidanz, LaFleur, Brown, Burns, Gramaglia and van der Heyde2010; Inoue et al. Reference Inoue, Niikura, Takeo, Mineo, Kawakami, Uchida, Kamiya and Kobayashi2012). In humans, γδ T cell frequencies during malaria infection are influenced by age and exposure (Ho et al. Reference Ho, Webster, Tongtawe, Pattanapanyasat and Weidanz1990; Roussilhon et al. Reference Roussilhon, Agrapart, Ballet and Bensussan1990; Hviid et al. Reference Hviid, Kurtzhals, Dodoo, Rodrigues, Ronn, Commey, Nkrumah and Theander1996) and are implicated in both protection and immunopathology. In an experimental model of infection in malaria-naïve individuals, involving the inoculation of sporozoites under chloroquine cover, γδ T cells were significant producers of IFN-γ in vitro in response to pRBCs for up to 14 months post infection (Teirlinck et al. Reference Teirlinck, McCall, Roestenberg, Scholzen, Woestenenk, de Mast, van der Ven, Hermsen, Luty and Sauerwein2011). These cells expressed the CD45RO+ CD62L-effector memory phenotype. Little is known about memory γδ T cells and whether they have a role in protection against re-infection with the parasite (Inoue et al. Reference Inoue, Niikura, Mineo and Kobayashi2013). In a longitudinal study in Papua New Guinea in young children, high production of IFN-γ in vitro in response to pRBCs, was shown to be associated with reduced risk of developing clinical disease (D'Ombrain et al. Reference D'Ombrain, Robinson, Stanisic, Taraika, Bernard, Michon, Mueller and Schofield2008; Robinson et al. Reference Robinson, D'Ombrain, Stanisic, Taraika, Bernard, Richards, Beeson, Tavul, Michon, Mueller and Schofield2009). γδ T cells were shown to be a major source. A further study in young children in Uganda, observed that repeated malaria exposure was associated with reduced percentages of the Vδ2+ γδ T cells and decreased proliferation and production of cytokines such as IFN-γ and TNF (Jagannathan et al. Reference Jagannathan, Kim, Greenhouse, Nankya, Bowen, Eccles-James, Muhindo, Arinaitwe, Tappero, Kamya, Dorsey and Feeney2014b). These cells did not appear to play a role in protecting against reinfection or parasitemia but the loss/dysfunction of Vδ2+ γδ T cells was associated with a reduced likelihood of symptoms upon P. falciparum re-infection. Although a causal relationship remains to be demonstrated in larger, longitudinal prospective studies, this data suggest that clinical immunity may be mediated in part by attenuation of the Vδ2+ γδ T cell pro-inflammatory response (Jagannathan et al. Reference Jagannathan, Kim, Greenhouse, Nankya, Bowen, Eccles-James, Muhindo, Arinaitwe, Tappero, Kamya, Dorsey and Feeney2014b). Given their role in the early inflammatory response to Plasmodium, γδ T cells were also investigated as a cellular source of cytokines and chemokines associated with severe malaria (Stanisic et al. Reference Stanisic, Cutts, Eriksson, Fowkes, Rosanas-Urgell, Siba, Laman, Davis, Manning, Mueller and Schofield2014). Increased TNF, IL-10, IP-10, IL-6, macrophage inflammatory protein (MIP)-1β and MIP-1α were associated with increased risk of severe malaria. γδ T cells were identified as the cellular source of TNF, MIP-1β and MIP-1α. Given their specificity for restricted TCR ligands, it was suggested that this population would be an attractive target for a preventive vaccine to protect against severe disease (Stanisic et al. Reference Stanisic, Cutts, Eriksson, Fowkes, Rosanas-Urgell, Siba, Laman, Davis, Manning, Mueller and Schofield2014).
Tregs
An inflammatory immune response induced to eradicate infecting organisms e.g. Plasmodium, can result in immunopathology if it is not regulated. Tregs modulate proliferation, cytokine production and survival of CD4+ effector cells and impact on the maturation of APCs (Tang and Bluestone, Reference Tang and Bluestone2008). The immunoregulatory cytokines, IL-10 and TGF-β, are critical for the induction and effector function of CD4+CD25+Foxp3+ Tregs (Scholzen et al. Reference Scholzen, Mittag, Rogerson, Cooke and Plebanski2009). By suppressing Th1 effector mechanisms, they may allow parasite persistence; thus, their activation has been proposed as a possible mechanism by which Plasmodium is able to evade immunity and establish chronic infection. The timing of induction appears to influence the outcome of infection – they can limit immunopathology but may also inhibit the protective immune response, allowing rapid parasite growth, if they are induced too early in infection (reviewed in (Finney et al. Reference Finney, Riley and Walther2010; Hansen and Schofield, Reference Hansen and Schofield2010; Scholzen et al. Reference Scholzen, Minigo and Plebanski2010)). These cells were examined in detail in recent publications (reviewed in (Finney et al. Reference Finney, Riley and Walther2010; Hansen and Schofield, Reference Hansen and Schofield2010; Scholzen et al. Reference Scholzen, Minigo and Plebanski2010)). There are a few key observations.
Increases in CD4+CD25+Foxp3+ Tregs have been observed in malaria infected rodents (Wu et al. Reference Wu, Wang, Zheng, Feng, Liu, Ma and Cao2007; Cambos et al. Reference Cambos, Belanger, Jacques, Roulet and Scorza2008; Couper et al. Reference Couper, Blount, Wilson, Hafalla, Belkaid, Kamanaka, Flavell, de Souza and Riley2008; Chen et al. Reference Chen, Liu, Wang, Wu, Feng, Zheng, Guo, Li, Wang and Cao2009) and humans (Walther et al. Reference Walther, Tongren, Andrews, Korbel, King, Fletcher, Andersen, Bejon, Thompson, Dunachie, Edele, de Souza, Sinden, Gilbert, Riley and Hill2005; Jangpatarapongsa et al. Reference Jangpatarapongsa, Chootong, Sattabongkot, Chotivanich, Sirichaisinthop, Tungpradabkul, Hisaeda, Troye-Blomberg, Cui and Udomsangpetch2008; Minigo et al. Reference Minigo, Woodberry, Piera, Salwati, Tjitra, Kenangalem, Price, Engwerda, Anstey and Plebanski2009). Unlike rodent malaria, an increase in Tregs is correlated with an increased parasite density in human malaria, (Walther et al. Reference Walther, Tongren, Andrews, Korbel, King, Fletcher, Andersen, Bejon, Thompson, Dunachie, Edele, de Souza, Sinden, Gilbert, Riley and Hill2005, Reference Walther, Jeffries, Finney, Njie, Ebonyi, Deininger, Lawrence, Ngwa-Amambua, Jayasooriya, Cheeseman, Gomez-Escobar, Okebe, Conway and Riley2009; Minigo et al. Reference Minigo, Woodberry, Piera, Salwati, Tjitra, Kenangalem, Price, Engwerda, Anstey and Plebanski2009), particularly in those individuals with symptomatic infections (Minigo et al. Reference Minigo, Woodberry, Piera, Salwati, Tjitra, Kenangalem, Price, Engwerda, Anstey and Plebanski2009; Walther et al. Reference Walther, Jeffries, Finney, Njie, Ebonyi, Deininger, Lawrence, Ngwa-Amambua, Jayasooriya, Cheeseman, Gomez-Escobar, Okebe, Conway and Riley2009). Treg numbers also vary according to transmission levels with an increase in both Treg and T effector cells at the end of the malaria transmission season (Finney et al. Reference Finney, Nwakanma, Conway, Walther and Riley2009), suggesting that malaria infection can drive the expansion of Tregs. The majority of studies have investigated Tregs in relation to P. falciparum; few have examined Tregs and P. vivax (Jangpatarapongsa et al. Reference Jangpatarapongsa, Chootong, Sattabongkot, Chotivanich, Sirichaisinthop, Tungpradabkul, Hisaeda, Troye-Blomberg, Cui and Udomsangpetch2008; Bueno et al. Reference Bueno, Morais, Araujo, Gomes, Correa-Oliveira, Soares, Lacerda, Fujiwara and Braga2010). Similar to P. falciparum, CD4+CD25+Foxp3+ cells were observed to increase during P. vivax infection and were associated with parasite load.
Functional studies in mice to identify the role of this cell type in malaria infection have generated conflicting results depending on the rodent model used (reviewed in (Finney et al. Reference Finney, Riley and Walther2010; Scholzen et al. Reference Scholzen, Minigo and Plebanski2010)). Human studies suggest that the propensity to activate Tregs may influence susceptibility. In human experimental infection, an early burst of TGF-β production was associated with induction of CD4+CD25+Foxp3+ cells, decreased inflammatory cytokine production, decreased antigen-specific immune responses and faster parasite growth (Walther et al. Reference Walther, Tongren, Andrews, Korbel, King, Fletcher, Andersen, Bejon, Thompson, Dunachie, Edele, de Souza, Sinden, Gilbert, Riley and Hill2005). It was proposed by the authors that monocyte-derived TGF-β may induce Treg differentiation by inducing Foxp3 expression in CD4+ T cells (Walther et al. Reference Walther, Tongren, Andrews, Korbel, King, Fletcher, Andersen, Bejon, Thompson, Dunachie, Edele, de Souza, Sinden, Gilbert, Riley and Hill2005). A functional deficit of Tregs in the ‘resistant’ Fulani compared with sympatric ethnic groups in Burkina Faso was suggested to contribute to their reduced susceptibility to clinical malaria by enabling control of parasite levels (Torcia et al. Reference Torcia, Santarlasci, Cosmi, Clemente, Maggi, Mangano, Verra, Bancone, Nebie, Sirima, Liotta, Frosali, Angeli, Severini, Sannella, Bonini, Lucibello, Maggi, Garaci, Coluzzi, Cozzolino, Annunziato, Romagnani and Modiano2008). An alternate explanation is that they are protected by other genetic traits (Arama et al. Reference Arama, Maiga, Dolo, Kouriba, Traore, Crompton, Pierce, Troye-Blomberg, Miller and Doumbo2015) and the low Treg activity is the result of lower levels of malaria infection.
If Tregs do contribute to disease severity, by interfering with clearance of the parasites, one would expect higher numbers of Tregs in individuals with severe vs uncomplicated malaria. Indeed, in severe malaria cases with hyperparasitemia, Treg numbers were significantly increased compared with severe malaria patients without hyperparasitemia (Minigo et al. Reference Minigo, Woodberry, Piera, Salwati, Tjitra, Kenangalem, Price, Engwerda, Anstey and Plebanski2009). While there was no difference in Tregs or Th1 effector:Treg ratios (as an indicator of immune homeostasis) between severe and uncomplicated malaria cases in adults, the ratio was decreased in severe malaria cases compared with asymptomatic controls, suggestive of immune suppression (Minigo et al. Reference Minigo, Woodberry, Piera, Salwati, Tjitra, Kenangalem, Price, Engwerda, Anstey and Plebanski2009). In this study, increased frequencies of TNF Receptor II-positive (TNFRII)-Treg cells were observed in adults with severe malaria, compared with uncomplicated malaria. TNFRII positive Tregs are a highly suppressive population – thus activation of a specific subset may provide a link between Tregs and disease severity in this population. In another study, the T effector:Treg ratio was elevated in severe compared with uncomplicated disease in children (Walther et al. Reference Walther, Jeffries, Finney, Njie, Ebonyi, Deininger, Lawrence, Ngwa-Amambua, Jayasooriya, Cheeseman, Gomez-Escobar, Okebe, Conway and Riley2009), indicating a potential loss of immune regulation. While there were no differences in Treg numbers or function between children with severe and uncomplicated disease, significantly higher Th1 effector responses were observed in children with severe malaria, indicating a failure to maintain immune homeostasis and regulate the response (Walther et al. Reference Walther, Jeffries, Finney, Njie, Ebonyi, Deininger, Lawrence, Ngwa-Amambua, Jayasooriya, Cheeseman, Gomez-Escobar, Okebe, Conway and Riley2009). Treg activity during acute disease was also inversely correlated with malaria-specific immune responses 28 days later. In this study, a ‘self-regulating’ population of CD4+CD45RO+Foxp3- Th1 cells co-producing IFN-γ/IL-10 was the only source of IL-10 during acute infection and was more frequent in uncomplicated malaria compared with severe malaria cases (Walther et al. Reference Walther, Jeffries, Finney, Njie, Ebonyi, Deininger, Lawrence, Ngwa-Amambua, Jayasooriya, Cheeseman, Gomez-Escobar, Okebe, Conway and Riley2009). This would argue against the classical T regulatory cell directly contributing to the immediate outcome of infection in this population. It is possible however that Tregs may contribute to susceptibility to severe disease in children and adults in different ways.
It is not yet known what parasite factors induce Tregs. Data in humans suggest that Tregs induced by acute malaria infection can impact on the subsequent memory response (Walther et al. Reference Walther, Jeffries, Finney, Njie, Ebonyi, Deininger, Lawrence, Ngwa-Amambua, Jayasooriya, Cheeseman, Gomez-Escobar, Okebe, Conway and Riley2009). In rodent models, it was demonstrated that Treg depletion prior to administration of a malaria vaccine results in higher and longer-lasting memory responses (Moore et al. Reference Moore, Gallimore, Draper, Watkins, Gilbert and Hill2005). Together, these data suggest that antimalaria treatment at the time of vaccination to ‘normalize’ the immune response may be needed to ensure induction of appropriate vaccine-specific responses.
T cell memory
Few studies have investigated mechanisms associated with the generation and persistence of T cell memory, although a generally held perception is that Plasmodium-specific immunity is short-lived in the absence of re-exposure, which may explain the observed short-lived/unstable nature of cellular immune responses in naturally exposed individuals (Migot et al. Reference Migot, Chougnet, Raharimalala, Astagneau, Lepers and Deloron1993; Riley et al. Reference Riley, Morris-Jones, Blackman, Greenwood and Holder1993; Flanagan et al. Reference Flanagan, Mwangi, Plebanski, Odhiambo, Ross, Sheu, Kortok, Lowe, Marsh and Hill2003; Bejon et al. Reference Bejon, Mwacharo, Kai, Todryk, Keating, Lowe, Lang, Mwangi, Gilbert, Peshu, Marsh and Hill2007; Dent et al. Reference Dent, Chelimo, Sumba, Spring, Crabb, Moormann, Tisch and Kazura2009; Moormann et al. Reference Moormann, Sumba, Tisch, Embury, King, Kazura and John2009). Interpretation of many of the field studies has been complicated by the challenges in controlling for malaria exposure (reviewed in (Achtman et al. Reference Achtman, Bull, Stephens and Langhorne2005)), but there is data to suggest that immunological memory exists. Malaria was eradicated from Madagascar in the 1960s, but its reappearance in the 1980s presented a unique opportunity to examine the persistence of immunity by comparing the incidence of malaria in older, previously exposed adults with younger, malaria naïve residents (Deloron and Chougnet, Reference Deloron and Chougnet1992). Previously exposed individuals >40 years of age had better protection against clinical malaria, but not infection, than younger residents who had not been alive prior to eradication, when malaria was endemic. Although a control group >40 years of age who had not grown up in the area was not available for comparison, the data suggests that clinical immunity can persist for up to 30 years. A study comparing P. falciparum malaria attacks in European patients and African immigrants who had been resident in Europe for at least 4 years showed that the latter had lower mean parasite densities, less frequent severe disease, accelerated parasite clearance and resolution of fever (Bouchaud et al. Reference Bouchaud, Cot, Kony, Durand, Schiemann, Ralaimazava, Coulaud, Le Bras and Deloron2005). This also suggests persisting clinical immunity following several years of non-exposure.
Interestingly, a recent study that carefully controlled for malaria exposure, showed that in the absence of re-infection and boosting of immune responses, malaria-specific CD4+CD45RO+ effector memory responses (IFN-γ) are relatively short-lived (a half-life of 3·3 years), whereas CD4+CD45RO+ central memory regulatory responses (IL-10) appeared to be very long-lived (a half-life of >6 years) (Wipasa et al. Reference Wipasa, Okell, Sakkhachornphop, Suphavilai, Chawansuntati, Liewsaree, Hafalla and Riley2011). In the context of human experimental malaria, memory responses were maintained (Roestenberg et al. Reference Roestenberg, Teirlinck, McCall, Teelen, Makamdop, Wiersma, Arens, Beckers, van Gemert, van de Vegte-Bolmer, van der Ven, Luty, Hermsen and Sauerwein2011; Teirlinck et al. Reference Teirlinck, McCall, Roestenberg, Scholzen, Woestenenk, de Mast, van der Ven, Hermsen, Luty and Sauerwein2011) as evidenced by the presence of Plasmodium-specific effector T cell (CD4+CD45RO+CD62L-) responses 1 year post-infection.
It may be that under conditions of heavy parasite exposure and in the presence of high parasitemias, dysregulation/modulation of the immune response negatively impacts on the generation and persistence of T cell memory, which is reflected in apparent short-lived cellular immune responses in endemic areas (Migot et al. Reference Migot, Chougnet, Raharimalala, Astagneau, Lepers and Deloron1993; Riley et al. Reference Riley, Morris-Jones, Blackman, Greenwood and Holder1993; Flanagan et al. Reference Flanagan, Mwangi, Plebanski, Odhiambo, Ross, Sheu, Kortok, Lowe, Marsh and Hill2003; Bejon et al. Reference Bejon, Mwacharo, Kai, Todryk, Keating, Lowe, Lang, Mwangi, Gilbert, Peshu, Marsh and Hill2007; Dent et al. Reference Dent, Chelimo, Sumba, Spring, Crabb, Moormann, Tisch and Kazura2009; Moormann et al. Reference Moormann, Sumba, Tisch, Embury, King, Kazura and John2009). Induction of Tregs during malaria infection may impact on generation of T cell memory (Walther et al. Reference Walther, Jeffries, Finney, Njie, Ebonyi, Deininger, Lawrence, Ngwa-Amambua, Jayasooriya, Cheeseman, Gomez-Escobar, Okebe, Conway and Riley2009). It has been proposed that apoptosis of T cells during infection may contribute to a loss of memory responses. In rodent models, IFN-γ mediated apoptosis of parasite-specific T cells has been observed (Hirunpetcharat and Good, Reference Hirunpetcharat and Good1998; Wipasa et al. Reference Wipasa, Xu, Stowers and Good2001; Xu et al. Reference Xu, Wipasa, Yan, Zeng, Makobongo, Finkelman, Kelso and Good2002) and is directly related to parasite levels (Elliott et al. Reference Elliott, Kuns and Good2005), occurring during patent but not sub-patent infection. Although some of the apoptosis may have reflected homeostatic mechanisms, it was noted that residual immunity post-infection was decreased (Xu et al. Reference Xu, Wipasa, Yan, Zeng, Makobongo, Finkelman, Kelso and Good2002). In humans, markers/characteristics of apoptosis have been observed in T cells from patients with clinical malaria (Toure-Balde et al. Reference Toure-Balde, Sarthou, Aribot, Michel, Trape, Rogier and Roussilhon1996; Kemp et al. Reference Kemp, Akanmori, Adabayeri, Goka, Kurtzhals, Behr and Hviid2002a). Recent data suggest that P. falciparum drives the expansion of a population of T cells expressing inhibitory receptors PD-1 and LAG-3 and this may contribute to sub-optimal host immunity that is unable to control the parasite (Butler et al. Reference Butler, Moebius, Pewe, Traore, Doumbo, Tygrett, Waldschmidt, Crompton and Harty2012; Illingworth et al. Reference Illingworth, Butler, Roetynck, Mwacharo, Pierce, Bejon, Crompton, Marsh and Ndungu2013). Impaired CD4+ T cell function may result in impaired T cell memory responses (Illingworth et al. Reference Illingworth, Butler, Roetynck, Mwacharo, Pierce, Bejon, Crompton, Marsh and Ndungu2013). Malaria vaccines that are required to induce a powerful effector T cell response for protection may be negatively impacted by such immune dysregulation.
CHALLENGES
There are many technical challenges and limitations associated with examining cellular immune responses in humans which need to be considered. The interpretation of data is complicated by the use of peripheral blood as the source of cells, when it is well known that they also reside in tissues.
During acute malaria and subsequent drug treatment (Chougnet et al. Reference Chougnet, Tallet, Ringwald and Deloron1992; Hviid et al. Reference Hviid, Kurtzhals, Goka, Oliver-Commey, Nkrumah and Theander1997; Worku et al. Reference Worku, Bjorkman, Troye-Blomberg, Jemaneh, Farnert and Christensson1997; Kemp et al. Reference Kemp, Akanmori, Adabayeri, Goka, Kurtzhals, Behr and Hviid2002a, Reference Kemp, Akanmori, Kurtzhals, Adabayeri, Goka and Hviid2002b), there are changes in lymphocyte activation and subset distribution suggesting that certain T cell subsets relocate from the peripheral blood. PBMCs can also be hyporesponsive at this time (Ho et al. Reference Ho, Webster, Green, Looareesuwan, Kongchareon and White1988; Kremsner et al. Reference Kremsner, Zotter, Feldmeier, Graninger, Rocha, Jansen-Rosseck and Bienzle1990), so the timing of blood sampling for immunologic analyses may affect interpretation of data. T cell-responsiveness to malaria-specific and non-specific antigens decreases in individuals with high parasitemia and severe P. falciparum malaria (Brasseur et al. Reference Brasseur, Agrapart, Ballet, Druilhe, Warrell and Tharavanij1983; Druilhe et al. Reference Druilhe, Brasseur, Agrapart, Ballet, Chanthavanich, Looareesuwan, White, Warrell, Warrell, Tharavanij and Gentilini1983) and is restored rapidly following antimalarial treatment. Thus a recent study examined cytokines/chemokines in convalescent samples to understand whether an individual's inherent immunological responsiveness was able to predict risk of developing severe malaria in the context of a case-control study design (Stanisic et al. Reference Stanisic, Cutts, Eriksson, Fowkes, Rosanas-Urgell, Siba, Laman, Davis, Manning, Mueller and Schofield2014). In this study, increased TNF, IL-10, IP-10, IL-6, MIP-1β and MIP-1α were associated with increased risk of severe malaria and severe malaria syndromes were associated with distinct cytokine/chemokine responses compared with uncomplicated cases.
The lack of standardized assays to evaluate cellular immune responses induced during malaria infection and vaccine trials slows progress. Standardized assays would enable comparisons across different sites/in different populations and would potentially facilitate the identification of an immune correlate of protection that could be utilized for evaluation of vaccine candidates in clinical trials. A good example of this is the evaluation of cytokine responses in different studies using serum, plasma, cell culture supernatants or intracellular staining. It has been demonstrated that there is no correlation or only weak associations between intracellular and plasma/serum cytokines (Jason et al. Reference Jason, Archibald, Nwanyanwu, Byrd, Kazembe, Dobbie and Jarvis2001; Quelhas et al. Reference Quelhas, Puyol, Quinto, Nhampossa, Serra-Casas, Macete, Aide, Sanz, Aponte, Doolan, Alonso, Menendez and Dobano2012). This is not surprising as intracellular cytokines measure cytokines produced by a specific cell population after stimulation with antigen in vitro, whereas plasma/serum cytokines are not antigen specific and reflect what is generally present in peripheral blood. Technical disparities such as these may contribute to differences observed in findings across studies.
Protection induced by the majority of licensed vaccines is dependent on the induction of protective antibodies. However, cellular immune responses are more complex, and may rapidly control parasitemia yet result in pathology. A key challenge would appear to be identifying and then inducing an optimal immune response that would rapidly clear parasites in concert with regulation of any inflammatory responses that may be harmful. Additionally, without specificity, it will be hard to define the contribution of any T cell subset in vivo.
There is only one defined parasite target expressed during the blood-stage that has been shown to induce a T cell-dependent immune response, without the production of antibodies. The Plasmodium purine salvage enzyme, hypoxanthine guanine xanthine phosphoribosyl transferase (HGXPRT), was shown to partially protect mice against parasite challenge in a T cell-dependent manner (Makobongo et al. Reference Makobongo, Riding, Xu, Hirunpetcharat, Keough, de Jersey, Willadsen and Good2003). HGXPRT is a purine salvage enzyme critical to parasite survival as the parasites cannot make purines de novo. It is located in electron dense regions in merozoites and in vesicles in the erythrocyte cytoplasm and is released upon erythrocyte rupture. Individuals resident in malaria-endemic areas have T cell, but not antibody responses to this antigen ((Woodberry et al. Reference Woodberry, Pinzon-Charry, Piera, Panpisutchai, Engwerda, Doolan, Salwati, Kenangalem, Tjitra, Price, Good and Anstey2009) and DI Stanisic and MF Good, unpublished data). Due to high sequence conservation between human and Plasmodium HG(X)PRT, there is concern about the possible induction of autoimmunity following vaccination with this antigen.
High throughput technology may facilitate identification of novel target antigens recognized by T cells on a proteome-wide scale (Cardoso et al. Reference Cardoso, Roddick, Groves and Doolan2011). They may not be immunogenic in all individuals and many will be polymorphic. Historically, the use of tetramer assays in assessing P. falciparum-specific T cell responses has involved the direct ex vivo examination of CD8+ T cells specific for MHC Class I restricted pre-erythrocytic antigens (e.g. Todryk et al. Reference Todryk, Pathan, Keating, Porter, Berthoud, Thompson, Klenerman and Hill2009a; Schwenk et al. Reference Schwenk, Banania, Epstein, Kim, Peters, Belmonte, Ganeshan, Huang, Reyes, Stryhn, Ockenhouse, Buus, Richie and Sedegah2013). Recent developments in the use of MHC Class II tetramers underscore the potential of utilizing this technology to interrogate P. falciparum antigen CD4+ T cell immune responses both in the context of naturally acquired immunity and vaccine trials (Nepom, Reference Nepom2012). The use of sub-unit vaccine candidates will also require the identification of a suitable human-compatible adjuvant, one of the key issues associated with the development of sub-unit vaccines. Thus, an alternative approach is to use the whole parasite. Attenuated blood stage infections are capable of inducing protective parasite-specific T cell responses in the absence of detectable antibodies (Pombo et al. Reference Pombo, Lawrence, Hirunpetcharat, Rzepczyk, Bryden, Cloonan, Anderson, Mahakunkijcharoen, Martin, Wilson, Elliott, Eisen, Weinberg, Saul and Good2002; Elliott et al. Reference Elliott, Kuns and Good2005). This has been further investigated as a vaccine approach in the context of low doses of dead parasite in combination with Alum and CpG (Pinzon-Charry et al. Reference Pinzon-Charry, McPhun, Kienzle, Hirunpetcharat, Engwerda, McCarthy and Good2010) and chemically attenuated blood-stage parasites (Good et al. Reference Good, Reiman, Rodriguez, Ito, Yanow, El-Deeb, Batzloff, Stanisic, Engwerda, Spithill, Hoffman, Lee and McPhun2013). Both methods induce T cell responses in the absence of detectable antibodies and the target antigens are conserved, given that protection is observed against different stains and species of rodent Plasmodium. Heterogeneity has been demonstrated in the early immune response to experimental infection in malaria naïve volunteers (Walther et al. Reference Walther, Woodruff, Edele, Jeffries, Tongren, King, Andrews, Bejon, Gilbert, Souza, Sinden, Hill and Riley2006). Distinct pro-inflammatory cytokine responses, defined based on the strength of the response, had a differential impact on parasite growth and clinical outcome. It will be interesting to determine, if and how such heterogeneity could impact on the development of adaptive immune responses to a whole parasite vaccine.
FUTURE DIRECTIONS
Cellular immunological correlates of immunity have been defined in naturally and experimentally infected individuals and the data to date suggest that IFN-γ is a critical cytokine to control parasite growth, probably working by activating splenic macrophages to phagocytose pRBC. Unregulated responses, however, are deleterious, and subsequent dampening of the immune response is required to protect from immunopathology. IL-10 is an important regulatory cytokine. These cytokines are generated from various cell types and the same single cell can make both cytokines. However, co-producing cells do not appear to be associated with long-term protection from malaria episodes. Responses dominated by TNF are associated with clinical symptoms. These are interesting data; however, these are still a limited number of observations and future research needs to consolidate these in different endemic settings and different populations to generate a strong hypothesis. Further research then needs to identify candidate vaccines and delivery strategies that induce the correctly ordered cytokine response.
The correct choice of antigens to be contained within a vaccine is not known. Regulatory authorities prefer that vaccines contain only the minimal antigenic material necessary to safely induce immunity. However sifting through the thousands of antigens that might be suitable is a challenging and time consuming task. It is clear that a limited/attenuated blood stage infection is able to induce protective immunity against homologous challenge. Our view is that a whole parasite approach is the only feasible way to induce cellular immunity by vaccination. This approach is successful in animal models and avoids the dangers of selecting individual antigens that may be polymorphic or poorly immunogenic. A significant challenge is to then use a delivery strategy for whole parasite vaccines that does induce the appropriate cytokine response. Human challenge studies over the next 5 years will give an indication as to whether the current whole parasite strategies are successful and will enable researchers to accumulate further cellular correlates of immunity that will guide continued vaccine development.
ACKNOWLEDGEMENTS
We acknowledge grant support from a National Health and Medical Research Council Fellowship (M.F.G) and Program Grant 1037304 (M.F.G).
