Malaria

Plasmodium is a genus of intracellular apicomplexan protists, which are the causative agent of malaria. While over 100 species of Plasmodium have been identified (Tuteja, Reference Tuteja2007), only five of these cause malaria in humans. These species are Plasmodium falciparum, Plasmodium ovale, Plasmodium malariae, Plasmodium knowlesi and Plasmodium vivax (Kantele and Jokiranta, Reference Kantele and Jokiranta2011); with P. falciparum infection strongly associated with higher disease severity and mortality.

The life cycle of Plasmodium spp. is complex, with infection acquired by humans through sporozoite transmission from infected female Anopheles mosquitoes during feeding. Sporozoites migrate to and replicate within hepatocytes for ~7–10 days, resulting in merozoite release into the bloodstream (Tuteja, Reference Tuteja2007). The importance of hepatocytes to the Plasmodium life cycle is still not completely elucidated; however, a recent review by Kaushansky and Kappe (Reference Kaushansky and Kappe2015) summarizes key molecular mechanisms thought to support the specific infection of hepatocytes, and the environment provided by hepatocytes, which permits the parasite to transform into the merozoite stage (Kaushansky and Kappe, Reference Kaushansky and Kappe2015). Once released, merozoites infect and asexually replicate within red blood cells, with some merozoites undergoing sexual differentiation (Tuteja, Reference Tuteja2007). These gametocytes are then transmitted to mosquitoes during feeding, with fertilization occurring within the mosquito's gut. The discovery that the blood stage of Plasmodium spp. can infect cells other than erythrocytes, namely plasmacytoid dendritic cells (DCs) (Wykes et al. Reference Wykes, Kay, Manderson, Liu, Brown, Richard, Wipasa, Jiang, Jones, Janse, Waters, Pierce, Miller, Stow and Good2011; Wykes and Horne-Debets, Reference Wykes and Horne-Debets2012), has led researchers to believe that this parasite can adopt the Trojan-horse style of survival and dissemination.

The host immune response to the malarial parasite is a key factor in the development and outcome of the infection. Indeed, as with many infections, the host immune response can control parasitaemia; however, excessive inflammation often leads to infection-associated pathology. Few stages of the Plasmodium spp. life cycle are extracellular, making it much easier for the parasite to evade the host immune system. Upon initial infection, immunity is induced, with subsequent exposures leading to induction of protective mechanisms in the host, such as antibodies against each stage of infection, cytokine production, induction of cytotoxic T cells (Schofield et al. Reference Schofield, Villaquiran, Ferreira, Schellekens, Nussenzweig and Nussenzweig1987) and Th1 responses (Torre et al. Reference Torre, Speranza, Giola, Matteelli, Tambini and Biondi2002; Perez-Mazliah and Langhorne, Reference Perez-Mazliah and Langhorne2014). Interferon-gamma (IFN-γ) has been identified as a key cytokine in the response against malaria infection (Artavanis-Tsakonas and Riley, Reference Artavanis-Tsakonas and Riley2002; McCall and Sauerwein, Reference McCall and Sauerwein2010). Indeed, it has been found that IFN-γ and interleukin (IL)-12 production occurs in a Toll like receptor (TLR)-9 and MyD88-dependant manner, with these cytokines then further enhancing the expression of TLRs by host immune cells (Franklin et al. Reference Franklin, Parroche, Ataide, Lauw, Ropert, de Oliveira, Pereira, Tada, Nogueira, da Silva, Bjorkbacka, Golenbock and Gazzinelli2009). This is one mechanism thought to contribute to the excessive pro-inflammatory response responsible for the symptoms of the infection.

With such a profound immune reaction to the presence of the parasite, the role of chemokines and their receptors in the promotion of inflammation during malarial infection has been explored. A particular focus of many studies has been the development and pathogenesis of cerebral malaria (CM), a complication of P. falciparum infection in humans which results in the obstruction of small blood vessel in the brain with infected erythrocytes, and is often fatal (Hora et al. Reference Hora, Kapoor, Thind and Mishra2016). Clinically, CM is characterized by seizures and loss of consciousness, due to encephalopathy (Newton et al. Reference Newton, Hien and White2000).

One receptor which has been extensively studied in the context of CM is CCR5. CCR5 is expressed by monocytes (Combadiere et al. Reference Combadiere, Ahuja, Tiffany and Murphy1996; Ubogu et al. Reference Ubogu, Callahan, Tucky and Ransohoff2006), activated T cells (Dragic et al. Reference Dragic, Litwin, Allaway, Martin, Huang, Nagashima, Cayanan, Maddon, Koup, Moore and Paxton1996; Ubogu et al. Reference Ubogu, Callahan, Tucky and Ransohoff2006; Jiang et al. Reference Jiang, Zhao, Bao, Xiao and Xiong2009), DCs (Lee et al. Reference Lee, Sharron, Montaner, Weissman and Doms1999) and natural killer (NK) (Khan et al. Reference Khan, Thomas, Moretto, Lee, Islam, Combe, Schwartzman and Luster2006) cells, with ligands including CCL3, 4, 5, 7, 8, 13, 14 and 16, and as will be discussed, has been identified as playing key roles in the pathogenesis of several protist infections. CCR5-deficient mice exhibit substantially reduced development of CM in comparison with their wild-type counterparts, with wild-type mice exhibiting higher numbers of CD8+ T cells and both serum levels and splenocyte production of the pro-inflammatory Th1 cytokine tumour necrosis factor alpha (TNF-α) also increased (Belnoue et al. Reference Belnoue, Kayibanda, Deschemin, Viguier, Mack, Kuziel and Renia2003b ).

Other chemokine receptors have also been implicated in the pathogenesis of malaria. CCR2 has been shown in murine experimental models to play little role in the development of CM (Belnoue et al. Reference Belnoue, Costa, Vigario, Voza, Gonnet, Landau, Van Rooijen, Mack, Kuziel and Renia2003a ). CCR2^-/- mice typically exhibit the inability of monocytes to traffic out of the bone marrow (Serbina and Pamer, Reference Serbina and Pamer2006; Tsou et al. Reference Tsou, Peters, Si, Slaymaker, Aslanian, Weisberg, Mack and Charo2007; Jia et al. Reference Jia, Serbina, Brandl, Zhong, Leiner, Charo and Pamer2008; Shi et al. Reference Shi, Jia, Mendez-Ferrer, Hohl, Serbina, Lipuma, Leiner, Li, Frenette and Pamer2011; Shi and Pamer, Reference Shi and Pamer2011). During experimental Plasmodium chabaudi infection, CCR2^-/- mice exhibit prolonged phases of acute parasitaemia when compared to their wild-type counterparts (Sponaas et al. Reference Sponaas, Freitas do Rosario, Voisine, Mastelic, Thompson, Koernig, Jarra, Renia, Mauduit, Potocnik and Langhorne2009). CCL2, a known CCR2 ligand, is a chemoattractant for monocytes and macrophages, and an experimental murine model of malaria utilizing the Plasmodium yoelli nigeriensis strain has shown that production (or overproduction) of this chemokine can contribute to immunopathology and host mortality (Pattaradilokrat et al. Reference Pattaradilokrat, Li, Wu, Qi, Eastman, Zilversmit, Nair, Huaman, Quinones, Jiang, Li, Zhu, Zhao, Kaneko, Long and Su2014).

Another chemokine receptor of importance during CM in humans and mice is that of CXCR3 and its ligands CXCL4, CXCL9 and CXCL10. Indeed, CXCL10 has been highlighted as a potential diagnostic and prognostic biomarker for CM in humans (Armah et al. Reference Armah, Wilson, Sarfo, Powell, Bond, Anderson, Adjei, Gyasi, Tettey, Wiredu, Tongren, Udhayakumar and Stiles2007; Jain et al. Reference Jain, Armah, Tongren, Ned, Wilson, Crawford, Joel, Singh, Nagpal, Dash, Udhayakumar, Singh and Stiles2008; Lucchi et al. Reference Lucchi, Jain, Wilson, Singh, Udhayakumar and Stiles2011; Wilson et al. Reference Wilson, Jain, Roberts, Lucchi, Joel, Singh, Nagpal, Dash, Udhayakumar, Singh and Stiles2011). In an experimental murine model, using the P. berghei strain, both CXCL9 and CXCL10 were found in the brain at significantly high levels, with infected CXCR3^-/- mice exhibiting protection against development of CM (Campanella et al. Reference Campanella, Tager, El Khoury, Thomas, Abrazinski, Manice, Colvin and Luster2008). Subsequent analysis confirmed that these mice had significantly fewer T cells in the brain compared with their wild-type counterparts. Furthermore, CXCL4, a platelet derived chemokine, is elevated in the brains of mice infected with P. berghei, with platelets becoming activated early on in infection and CXCL4 contributing to the development of CM through destruction of the blood–brain barrier (Srivastava et al. Reference Srivastava, Cockburn, Swaim, Thompson, Tripathi, Fletcher, Shirk, Sun, Kowalska, Fox-Talbot, Sullivan, Zavala and Morrell2008). Histological examination of brains from infected CXCL4^-/- and CXCR3^-/- mice show less cerebral inflammation and damage than their wild-type counterparts, driven by fewer CD4+ and CD8+ T cells in the brain (Srivastava et al. Reference Srivastava, Cockburn, Swaim, Thompson, Tripathi, Fletcher, Shirk, Sun, Kowalska, Fox-Talbot, Sullivan, Zavala and Morrell2008).

Toxoplasmosis

Toxoplasma gondii is another Apicomplexan protist, first identified by Nicolle and Manceaux in 1908. This is an intracellular pathogen, which can infect most warm-blooded animals (Dubey, Reference Dubey and S.1996). The infection has a high prevalence in humans, with previous studies identifying an overall seroprevalence of 22·5% for anti-Toxoplasma antibodies in the USA (Jones et al. Reference Jones, Kruszon-Moran, Wilson, McQuillan, Navin and McAuley2001). Despite this high prevalence, many individuals are unaware of their infection status, as a competent immune system has the ability to asymptomatically control this infection (Paspalaki et al. Reference Paspalaki, Mihailidou, Bitsori, Tsagkaraki and Mantzouranis2001). The infection often occurs as a result of ingesting T. gondii bradyzoites (slow, dormant cysts), which may be present in the undercooked meat of an infected animal (Dubey, Reference Dubey2009). In recent times, the ability of T. gondii to be transmitted in the environment and through water sources has been recognized (Jones and Dubey, Reference Jones and Dubey2010). Once ingested, the parasite has the ability to cross the intestinal epithelial barrier and replicate within cells of the lamina propria (Speer and Dubey, Reference Speer and Dubey1998). From here, tachyzoites (the fast, replicating stage) can disseminate to various tissues throughout the body (Barragan and Sibley, Reference Barragan and Sibley2002), such as brain and muscle tissue, where the parasite can exist dormant for long periods of time as cysts (Dubey et al. Reference Dubey, Lindsay and Speer1998).

Toxoplasmosis has the potential to be particularly detrimental in immunocompromised individuals; often causing encephalitis and multiple organ failure (Paspalaki et al. Reference Paspalaki, Mihailidou, Bitsori, Tsagkaraki and Mantzouranis2001). The infection may also occur through congenital transmission; with the severity of symptoms ranging from retinochoroiditis, to hydrocephalus and abortion of the developing foetus (Gilbert et al. Reference Gilbert, Tan, Cliffe, Guy and Stanford2006). The severity of the symptoms is largely dependent on the gestational age of transmission, with early transmission associated with a disease phenotype of a higher severity (Pinard et al. Reference Pinard, Leslie and Irvine2003).

The development of immunity to T. gondii is complex, with most leucocytes involved in some manner (Miller et al. Reference Miller, Boulter, Ikin and Smith2009; Munoz et al. Reference Munoz, Liesenfeld and Heimesaat2011; Coombes and Hunter, Reference Coombes and Hunter2015). The chemokine system has a critical role in the response to the initial stages of T. gondii infection, allowing establishment of an effective innate immune response to the parasite. For example, knockout studies have demonstrated that CCR5 is essential for the recruitment of NK cells and host survival during infection (Khan et al. Reference Khan, Thomas, Moretto, Lee, Islam, Combe, Schwartzman and Luster2006). The CCR5^-/- models have identified that without CCR5-mediated NK cell recruitment; there is a reduction in the CCR5 ligands, CCL3, CCL4 and CCL5, as well as reduced IFN-γ production in spleen, lung, liver and small intestine. Together, infected CCR5^-/- mice did not exhibit the typical tissue damage as a consequence of excessive inflammation; however, they did have an increased parasite burden (Khan et al. Reference Khan, Thomas, Moretto, Lee, Islam, Combe, Schwartzman and Luster2006).

Interestingly, the parasite can also influence the host immune response through secretion of immunomodulatory proteins, such as Cyclophilin 18 (TgCyp18) and the dense granule protein GRA24. TgCyp18 binds to CCR5 (Aliberti et al. Reference Aliberti, Valenzuela, Carruthers, Hieny, Andersen, Charest, Reis e Sousa, Fairlamb, Ribeiro and Sher2003), as illustrated in Fig. 2A, and acts to mimic the known CCR5-binding chemokines CCL3, CCL4 and CCL5 (Aliberti et al. Reference Aliberti, Valenzuela, Carruthers, Hieny, Andersen, Charest, Reis e Sousa, Fairlamb, Ribeiro and Sher2003; Ibrahim et al. Reference Ibrahim, Xuan and Nishikawa2010). Through stimulating IL-12, TNF-α and nitric oxide production by DCs and macrophages, TgCyp18 can promote Th1 type responses in a CCR5-dependent manner (Aliberti et al. Reference Aliberti, Valenzuela, Carruthers, Hieny, Andersen, Charest, Reis e Sousa, Fairlamb, Ribeiro and Sher2003; Ibrahim et al. Reference Ibrahim, Bannai, Xuan and Nishikawa2009). As illustrated in Fig. 2B, in addition to the production of pro-inflammatory cytokines, immune cell recruitment is further promoted through host cell chemokine production, including CCL2, 3, 4 and 5 in response to the parasite. Recruitment of inflammatory monocytes, DCs and NK cells, and induction of Th1 and CD8+ cytotoxic T cells, targets infected cells for destruction, but provides new potential host cells for the parasite.

Fig. 2. Leucocyte trafficking during T. gondii infection. (A) Known CCR5 ligands. (B) Infection of host immune cells results in the production of a number of inflammatory chemokines, which serve to recruit more immune cells. Promotion of a Th1-type response, with the involvement of CD8+ cytotoxic T cells is critical for the destruction of infected host cells.

CCL2 is produced in response to exposure of intestinal cells to the tachyzoite form of the parasite (Rachinel et al. Reference Rachinel, Buzoni-Gatel, Dutta, Mennechet, Luangsay, Minns, Grigg, Tomavo, Boothroyd and Kasper2004; Gopal et al. Reference Gopal, Birdsell and Monroy2011), in particular to the surface protein SAG1 (Brenier-Pinchart et al. Reference Brenier-Pinchart, Villena, Mercier, Durand, Simon, Cesbron-Delauw and Pelloux2006). In addition, the T. gondii dense granule protein GRA24 has been shown to induce CCL2 production by infected host cells, in an infection model utilizing the type II strain Pru (Braun et al. Reference Braun, Brenier-Pinchart, Yogavel, Curt-Varesano, Curt-Bertini, Hussain, Kieffer-Jaquinod, Coute, Pelloux, Tardieux, Sharma, Belrhali, Bougdour and Hakimi2013). CCL3, CCL4 and CCL5 are also produced by intestinal cells (Gopal et al. Reference Gopal, Birdsell and Monroy2011) and blood leucocytes (Bliss et al. Reference Bliss, Marshall, Zhang and Denkers1999) in response to T. gondii tachyzoites. It seems paradoxical that this parasite deliberately triggers the recruitment of cells designed to destroy it, however as previously mentioned, and illustrated in Fig. 2B, this parasite adopts the Trojan-horse method of dissemination to establish infection by attracting further cells to infect (Da Gama et al. Reference Da Gama, Ribeiro-Gomes, Guimaraes and Arnholdt2004; Lambert et al. Reference Lambert, Vutova, Adams, Lore and Barragan2009; Elsheikha and Khan, Reference Elsheikha and Khan2010; Sanecka and Frickel, Reference Sanecka and Frickel2012; Coombes et al. Reference Coombes, Charsar, Han, Halkias, Chan, Koshy, Striepen and Robey2013).

Other chemokine receptor knockout mice have been utilized in murine models of T. gondii infection, as summarized in Table 2, to demonstrate the role of certain cell types in the pathogenesis or control of infection. For example, CCR1^-/- mice have defects in neutrophil trafficking, and so use of these mice allowed researchers to dissect the neutrophilic response during early phases of T. gondii infection (Khan et al. Reference Khan, Murphy, Casciotti, Schwartzman, Collins, Gao and Yeaman2001). In addition, the necessity of CCR1 and its ligand CCL3 in driving the recruitment of inflammatory monocytes to the intestine during T. gondii-induced ileitis has been demonstrated through the use of CCR1^-/- mice (Schulthess et al. Reference Schulthess, Meresse, Ramiro-Puig, Montcuquet, Darche, Bègue, Ruemmele, Combadière, Di Santo, Buzoni-Gatel and Cerf-Bensussan2012). Moreover, studies using mice deficient in either CCR2 or its ligand CCL2, clearly demonstrated the importance of early monocyte recruitment to the intestine during T. gondii infection, with knockout mice unable to control parasite replication (Dunay et al. Reference Dunay, Damatta, Fux, Presti, Greco, Colonna and Sibley2008). Subsequent studies by the same group confirmed this important role of monocytes, but showed that neutrophil depletion had no impact on the ability to control infection, but indeed may actually contribute to pathology (Dunay et al. Reference Dunay, Fuchs and Sibley2010).

As with many parasites studies, knowledge has been gained through the use of animal models, but this always leaves the question as to relevance for human infection. While studies using human cells are limited, it has been found that infection of human epithelial and fibroblast with the highly virulent RH strain of T. gondii results in production of the monocyte and neutrophil chemoattractants CCL2, CXCL1 and CXCL8 (Denney et al. Reference Denney, Eckmann and Reed1999). Importantly, host cell invasion and lysis was necessary for this response, with stimulation with T. gondii lysates not sufficient to induce the response (Denney et al. Reference Denney, Eckmann and Reed1999). More recently, studies examining the impact of neuronal T. gondii infection showed similar results, with human bone marrow endothelial cells and microglial cells showed significant expression of CCL2 and CXCL1 after 8 h of infection with either the RH or PRU strains, whereas infection of a neuroblastoma cell line only stimulated production of CXCL1 after 24 h of infection (Mammari et al. Reference Mammari, Vignoles, Halabi, Darde and Courtioux2014). CXCL8 was produced by all cell types at much earlier time-points post-infection, indicating a key role in initiating immune responses (Mammari et al. Reference Mammari, Vignoles, Halabi, Darde and Courtioux2014).

Cryptosporidiosis

The causative agents of cryptosporidiosis, Cryptosporidium, such as Plasmodium and T. gondii are Apicomplexan protists. The main species of Cryptosporidium, which causes gastroenteritis in humans are Cryptosporidium parvum and Cryptosporidium hominus and are mainly transmitted via the oral–fecal route (Tzipori and Ward, Reference Tzipori and Ward2002). Cryptosporidiosis is usually a short-term infection, but it can cause severe disease in those who are immunocompromised and in the young, and it can persist in the lower gastrointestinal tract for up to 5 weeks. Cryptosporidiosis is one of the leading causes of diarrhoeal disease in developing countries. This parasite is highly resistant to chlorine (Korich et al. Reference Korich, Mead, Madore, Sinclair and Sterling1990), therefore meaning it presents a particular challenge to remove it from water systems, which is the main route of transmission.

Unlike Plasmodium and Leishmania, Cryptosporidium does not require a vector, and unlike T. gondii, it is minimally invasive for humans. Controlling Cryptosporidium infection is highly dependent on immune cell recruitment. Indeed, parasite infection is restricted to the intestinal epithelia and in addition, in neonates (Auray et al. Reference Auray, Lacroix-Lamandé, Mancassola, Dimier-Poisson and Laurent2007; Lantier et al. Reference Lantier, Lacroix-Lamandé, Potiron, Metton, Drouet, Guesdon, Gnahoui-David, Le Vern, Deriaud, Fenis, Rabot, Descamps, Werts and Laurent2013) and in the immunodeficient host, there is a low representation of immune cells in the intestinal mucosa (Steege et al., Reference Steege, Buurman and Forget1997) as well as chemokines (Lantier et al. Reference Lantier, Lacroix-Lamandé, Potiron, Metton, Drouet, Guesdon, Gnahoui-David, Le Vern, Deriaud, Fenis, Rabot, Descamps, Werts and Laurent2013). Several papers have reviewed the immune response to this parasite and like many of the other parasites discussed, IFN-γ, IL-12 and the induction of the Th1-type response is key to immune control of parasite (Borad and Ward, Reference Borad and Ward2010; Petry et al. Reference Petry, Jakobi and Tessema2010; McDonald et al. Reference McDonald, Korbel, Barakat, Choudhry and Petry2013).

Of the CC chemokines, both CCL5 and CCL20 have been shown to play crucial roles in mediating immune responses against the parasite in models of both human and mouse infection, respectively. CCL5 is chemotactic for Th1T cells (Kawai et al. Reference Kawai, Seki, Hiromatsu, Eastcott, Watts, Sugai, Smith, Porcelli and Taubman1999) and monocytes (Schall et al. Reference Schall, Bacon, Toy and Goeddel1990), and has been shown in a human intestinal cell line model to be produced, along with CXCL8 and TGF-β, in response to C. parvum infection (Laurent et al. Reference Laurent, Eckmann, Savidge, Morgan, Theodos, Naciri and Kagnoff1997; Maillot et al. Reference Maillot, Gargala, Delaunay, Ducrotte, Brasseur, Ballet and Favennec2000) to draw in cells associated with aiding recovery. The mucosal-tissue associated chemokine, CCL20, which is known to exhibit antimicrobial activity (Yang et al. Reference Yang, Chen, Hoover, Staley, Tucker, Lubkowski and Oppenheim2003) is downregulated in a neonatal mouse model of C. parvum infection, resulting in increased parasite burden (Guesdon et al. Reference Guesdon, Auray, Pezier, Bussiere, Drouet, Le Vern, Marquis, Potiron, Rabot, Bruneau, Werts, Laurent and Lacroix-Lamande2015).

Due to immunocompetent adults being fairly resistant to infection, little work has been done utilizing chemokine receptor knockout mice; indeed, in a study using adult CCR5^-/- mice, attempts to establish the infection were not successful (Campbell et al. Reference Campbell, Stewart and Mead2002). Despite this, the role of CCR5 in neonatal infection has been explored. Expression of mRNA for CCR5 and its ligands (CCL3, 4, 5) increases in the intestine of neonates from 4 to 6 days post-infection (Lacroix-Lamandé et al. Reference Lacroix-Lamandé, Mancassola, Auray, Bernardet and Laurent2008; Lantier et al. Reference Lantier, Lacroix-Lamandé, Potiron, Metton, Drouet, Guesdon, Gnahoui-David, Le Vern, Deriaud, Fenis, Rabot, Descamps, Werts and Laurent2013), suggesting a key role for CCR5 in the control of neonatal infection; however, CCR5^-/- neonates showed a higher parasite burden only at the initial stages of infection, and recovered as well as the wild-type controls (Lacroix-Lamandé et al. Reference Lacroix-Lamandé, Mancassola, Auray, Bernardet and Laurent2008). The authors hypothesized that the redundancy in the chemokine system prevented the absence of CCR5 from inhibiting cell recruitment. Indeed, the authors noted that CCR5^-/- mice has double the amount of CCR2 mRNA (Lacroix-Lamandé et al. Reference Lacroix-Lamandé, Mancassola, Auray, Bernardet and Laurent2008). CCR5 is an important chemokine receptor involved in monocyte trafficking, as is CCR2. Almost no inflammatory monocytes are found in the intestine of neonatal mice; however a recent study by (de Sablet et al. Reference de Sablet, Potiron, Marquis, Bussière, Lacroix-Lamandé and Laurent2016) has shown that upon infection with C. parvum this increases significantly. Infected neonatal CCR2^-/- mice have, as expected, very few inflammatory monocytes in their intestine; however it has also been found that these mice differ from their wild-type counterparts in the permeability of the intestinal wall (de Sablet et al. Reference de Sablet, Potiron, Marquis, Bussière, Lacroix-Lamandé and Laurent2016). CCR2^-/- mice did not lose epithelial barrier function to the same extent as wild-type mice, indicating that monocytes may contribute to this during C. parvum infection in neonates (de Sablet et al. Reference de Sablet, Potiron, Marquis, Bussière, Lacroix-Lamandé and Laurent2016).

Development and initiation of a successful immune response is highly dependent on the recruitment and functioning of professional antigen presenting cells such as DCs. As with the monocytic populations, neonatal intestines have few DCs, but upon infection with C. parvum, this number increases (Lantier et al. Reference Lantier, Lacroix-Lamandé, Potiron, Metton, Drouet, Guesdon, Gnahoui-David, Le Vern, Deriaud, Fenis, Rabot, Descamps, Werts and Laurent2013). CXCL10 is a ligand for CXCR3 (Loetscher et al. Reference Loetscher, Gerber, Loetscher, Jones, Piali, Clark-Lewis, Baggiolini and Moser1996), and attracts T cells, B cells, NK cells, DCs and macrophages (Sallusto et al. Reference Sallusto, Lenig, Mackay and Lanzavecchia1998; Liu et al. Reference Liu, Guo, Hibbert, Jain, Singh, Wilson and Stiles2011). To analyse the role of CXCL10 and its receptor CXCR3 in neonatal infection, CXCR3^-/- neonates were infected and it was found that DCs were not recruited to the intestine and that these mice did not control infection as well as their wild-type counterparts (Lantier et al. Reference Lantier, Lacroix-Lamandé, Potiron, Metton, Drouet, Guesdon, Gnahoui-David, Le Vern, Deriaud, Fenis, Rabot, Descamps, Werts and Laurent2013).

As previously mentioned, cryptosporidiosis is also a severe and life-threatening disease for the immunocompromised, for example AIDS patients. The severity of disease in these individuals highlights the necessity for a functional CD4+ T cell response to the parasite. CXCL10 has also been studied in the context of C. parvum infection as a complication of AIDS. In jejunal biopsies from AIDS patients with, and without C. parvum infection, CXCL10 was found in high levels in those patients with the dual infection, and this correlated with parasite burden (Wang et al. Reference Wang, Dann, Okhuysen, Lewis, Chappell, Adler and White2007). CXCL10 was localized to epithelial cells, and has been hypothesized to play a key role in the resolution of C. parvum infection in AIDS patients (Wang et al. Reference Wang, Dann, Okhuysen, Lewis, Chappell, Adler and White2007).

Leishmaniasis

Leishmania species are protists, which are members of the Sarcomastigophora phylum. Leishmania species exist in two forms; the intracellular amastigote form, and the promastigote form. The life cycle of the parasite involves amastigote and promastigote morphologies (Wheeler et al. Reference Wheeler, Gluenz and Gull2011), with extracellular promastigotes colonizing the sandfly midgut, while the intracellular amastigotes are associated with the mammalian host (Bates, Reference Bates2007). On initial infection of the host, promastigotes are engulfed by phagocytes and reside within the phagolysosome, where the parasite has the ability to inhibit phagolysosome biogenesis though its surface glycolipid lipophosphoglycan (Moradin and Descoteaux, Reference Moradin and Descoteaux2012). The parasite then differentiates into amastigotes within macrophages, replicating within the phagolysosome, before being released to infect cells in various tissues around the body (Moradin and Descoteaux, Reference Moradin and Descoteaux2012). The host can then transmit the parasite to uninfected sandflies, where it is believed that the sandfly ingests amastigote-infected macrophages during feeding (Dostálová and Volf, Reference Dostálová and Volf2012).

Leishmaniasises can be described as visceral, cutaneous or mucocutaneous, causing 20 000–30 000 deaths annually (World Health Organization figures, 2016). Visceral leishmaniasis is the most severe form of the disease, causing enlargement of the spleen and liver, most commonly caused by the Leishmania chagasi and Leishmania donovani strains. Cutaneous leishmaniasis is disfiguring to those infected, and is caused by the Leishmania amazonensis, Leishmania major and Leishmania mexicana strains, whereas mucocutaneous leishmaniasis is mostly associated with the Leishmania brasiliensis strain. A common feature of these infections is the extensive inflammatory reaction (Nylen and Gautam, Reference Nylen and Gautam2010; Oliveira et al. Reference Oliveira, Ribeiro, Schrieffer, Machado, Carvalho and Bacellar2014; Melo et al. Reference Melo, Silva, Grano, Souza and Machado2015; Rodrigues et al. Reference Rodrigues, Mazotto, Cardoso, Alves, Amaral, Silva, Pinheiro and Vermelho2015).

The immune response elicited against Leishmania spp. is complex, particularly as the parasite preferentially infects phagocytic cells, including DCs, macrophages and neutrophils. While the importance of granulocytes for the control of Leishmania infections is clear, studies have shown this parasite can actually evade the killing mechanisms of phagocytes, and use these cells to hide and spread infection (Laskay et al. Reference Laskay, van Zandbergen and Solbach2003; van Zandbergen et al. Reference van Zandbergen, Klinger, Mueller, Dannenberg, Gebert, Solbach and Laskay2004; Shapira and Zinoviev, Reference Shapira and Zinoviev2011). Resistance to leishmaniasis requires induction of a Th1-type response, with both IFN-γ and IL-12 crucial for this response (Kemp et al. Reference Kemp, Hey, Kurtzhals, Christensen, Gaafar, Mustafa, Kordofani, Ismail, Kharazmi and Theander1994; Kurtzhals et al. Reference Kurtzhals, Hey, Jardim, Kemp, Schaefer, Odera, Christensen, Githure, Olafson and Theander1994; Ghalib et al. Reference Ghalib, Whittle, Kubin, Hashim, el-Hassan, Grabstein, Trinchieri and Reed1995; Alexander et al. Reference Alexander, Satoskar and Russell1999). A number of studies and reviews have examined the expression of chemokines, and their role in mediating immunity against Leishmania spp. (Teixeira et al. Reference Teixeira, Teixeira, Andrade, Barral-Netto and Barral2006; Oghumu et al. Reference Oghumu, Lezama-Davila, Isaac-Marquez and Satoskar2010). Again, with examples for L. donovani and L. major included in Table 2, chemokine receptor knockout mice have been instrumental in dissecting the pathways of inflammation involved in developing immunity to infection, along with development of tissue pathology.

In a murine model of visceral leishmaniasis, utilizing L. donovani, CCL2, CCL3 and CXCL10 were found to be produced in the first week of infection by the liver, at a point when inflammation is being initiated (Cotterell et al. Reference Cotterell, Engwerda and Kaye1999). By comparing wild-type BALB/c mice with SCID mice, the authors found that early T-cell-independent chemokine expression does not mediate the inflammation associated with development of the characteristic granulomas found in the livers of infected mice (Cotterell et al. Reference Cotterell, Engwerda and Kaye1999). Leishmania donovani-infected CCL3^−/- mice have shown that this chemokine is not essential for control of the parasite; however, spleen cells from mice infected for 8 weeks produce significantly higher levels of IFN-γ than the wild-type controls, with CD4+ T cells the main source of this cytokine within the spleen cell mixture (Sato et al. Reference Sato, Kuziel, Melby, Reddick, Kostecki, Zhao, Maeda, Ahuja and Ahuja1999).

The role of CCL3 has also been examined in a murine model of cutaneous leishmaniasis, using L. major as the infective parasite (Steigerwald and Moll, Reference Steigerwald and Moll2005). DCs exposed to L. major were found to reduce expression of the receptors CCR2 and CCR5, as well as their ability to respond to CCL2 and CCL3; however, these cells upregulated expression of the receptor CCR7, and were therefore more responsive to its ligand CCL21 (Steigerwald and Moll, Reference Steigerwald and Moll2005). The authors of this particular study postulate that in response to the parasite, the immune system modulates the recruitment of DCs to the secondary lymphoid organs for antigen presentation. Indeed, a more recent study has shown that the lymph node hypertrophy that occurs during in vivo murine L. major infection, is due to bystander DCs expressing elevated levels of CCR7 in a TLR-9-dependent manner, leading to enhanced immune cell recruitment during the first weeks post-infection (Carvalho et al. Reference Carvalho, Petritus, Trochtenberg, Zaph, Hill, Artis and Scott2012). Additionally, studies looking at the hours immediately after exposure to L. major promastigotes demonstrated immediate and transient upregulation of CCL2 and CXCL1 production by murine macrophages (Racoosin and Beverley, Reference Racoosin and Beverley1997).

These studies, amongst many others, demonstrate that chemokines are induced at early and late stages of infection. However, these are in murine models of infection. What is known about chemokines in human leishmaniasis? Lesions from humans suffering from different degrees of cutaneous leishmaniasis (localized, intermediate and diffuse) showed elevated CCL2, CCL3, CCL4, CCL8 and CXCL10 in localized lesions and CCL3 in diffuse lesions (Ritter et al. Reference Ritter, Moll, Laskay, Brocker, Velazco, Becker and Gillitzer1996; Díaz et al. Reference Díaz, Zerpa and Tapia2013). Indeed, incubation of human blood mononuclear cells with L. major results in production of CCL2 and CXCL8 (Badolato et al. Reference Badolato, Sacks, Savoia and Musso1996).

Concluding remarks

The chemokine, and chemokine receptor system is complex, with slight variations between humans and mice, for example, human CCL13 has no mouse equivalent, and while mouse CCL8 binds CCR8, human CCL8 does not, suggesting that human and mouse CCL8 are not equivalent chemokines (Zlotnik and Yoshie, Reference Zlotnik and Yoshie2012). This highlights the need for careful translation of research findings from animal model to human disease. This review has focused on four protists, responsible for a significant amount of disease burden globally, and not only highlights the complexity of the chemokine system, but clearly demonstrates gaps in our understanding of the immune response to these organisms. The interesting paradox with chemokines is that these proteins draw in the cells that can facilitate resistance, but in some cases, may actually facilitate infection spread through providing more cells for the parasite to infect, allowing the parasites to adopt a ‘Trojan-horse’ style of deception from the host immune system. Therefore, there is a necessity for controlled balance between immunity for host protection, and immunity contributing to disease pathology.

The parasites discussed within this article are all members of the Protista kingdom; however, the diseases they cause in humans are quite different, mainly due to the specificities of each life cycle. Common to all is the importance of a Th1-type immune response for control of parasite replication, and with that, the inflammation-induced pathologies driven by excessive inflammation. Studies utilizing chemokine and chemokine-receptor knockout mice have been crucial in elucidating the immune pathways driven by these infections. For example, as highlighted previously, studies have been conducted with all of the protists discussed using CCR5^-/- mice (Table 2), giving insight into the role of CCR5 expressing cells (monocytes, activated T cells, DCs, NK cells) in protective immune responses and development of pathology.

In recent times, the development and investigation of the use of chemokine receptor agonists, and antagonists for disease therapy (Castellani et al. Reference Castellani, Bhattacharya, Tagen, Kempuraj, Perrella, De Lutiis, Boucher, Conti, Theoharides, Cerulli, Salini and Neri2007; Mohit and Rafati, Reference Mohit and Rafati2012) has been an area of investigation in the field of chemokine biology; however much of this work focuses on the use these new therapies as cancer treatments (Ruffini et al. Reference Ruffini, Morandi, Cabioglu, Altundag and Cristofanilli2007; Wu et al. Reference Wu, Lee, Chevalier and Hwang2009) and could be extended to infectious disease. This, along with the recognition that chemokines may also prove useful as potential biomarkers for infectious diseases, for example, as that described for CXCL10 in CM, highlights the need for research focus in this area.