INTRODUCTION
Trypanosomosis is a parasitic disease caused by African trypanosomes. These unicellular protozoan parasites are mainly transmitted through the bite of a tsetse fly (Glossina species) and form a threat to human and animal health on the African continent.
Trypanosoma brucei (T. b.) is the only trypanosome species able to infect humans, and can be further subdivided into three subspecies according to host infectivity, pathogenicity and geographical occurrence. Trypanosoma brucei gambiense and Trypanosoma brucei rhodesiense are the causative agents of Human African Trypanosomosis (HAT) or Sleeping Sickness, while T. b. brucei is only infective for livestock and game, (Jackson et al. Reference Jackson, Sanders, Berry, McQuillan, Aslett, Quail, Chukualim, Capewell, MacLeod, Melville, Gibson, Barry, Berriman and Hertz-Fowler2010). Other animal infective species are Trypanosoma congolense and Trypanosoma vivax, causing a wasting disease called Nagana, and Trypanosoma evansi, Trypanosoma equiperdum and Trypanosoma suis affecting various species of economically important livestock. Together, these infections cause massive economic damage to the sub-Saharan African continent, impacting on milk and meat production as well as agriculture labour potential under the form of tracking and transport power. Important to mention is that trypanosomes such as T. vivax, T. evansi and T. equiperdum are also classified as non-tsetse transmitted trypanosomosis, and now occur beyond the borders of the African continent, i.e. in South America, Asia and even occasionally in Europe (Silva et al. Reference Silva, Arosemena, Herrera, Sahib and Ferreira1995; Reid and Copeman, Reference Reid and Copeman2000; Oliveira et al. Reference Oliveira, Hernández-Gamboa, Jiménez-Alfaro, Zeledón, Blandón and Urbina2009; Da Silva et al. Reference Da Silva, Garcia Perez, Costa, França, De Gasperi, Zanette, Amado, Lopes, Teixeira and Monteiro 2011 ; Desquesnes et al. Reference Desquesnes, Dargantes, Lai, Lun, Holzmuller and Jittapalapong2013).
Tsetse fly transmitted infection begins with the injection of non-dividing metacyclic trypomastigotes into the host bloodstream during the blood meal of a tsetse fly (Fig. 1). In the mammalian blood, the metacyclic trypomastigotes resume cell division and differentiate into long slender bloodstream trypomastigotes, which multiply by longitudinal binary fission. Eventually, the long slender forms differentiate into short-living short stumpy forms by a mechanism involving cell density and the release of the stumpy induction factor (SIF). Differentiation to this short stumpy form limits parasite growth in the mammalian host and causes the parasitaemia levels to plateau (Vassella et al. Reference Vassella, Reuner, Yutzy and Boshart1997; Reuner et al. Reference Reuner, Vassella, Yutzy and Boshart1997; Tyler et al. Reference Tyler, Higgs, Matthews and Gull2001; Rico et al. Reference Rico, Rojas, Mony, Szoor, Macgregor and Matthews2013; Szöőr et al. Reference Szöőr, Dyer, Ruberto, Acosta-Serrano and Matthews2013; Mony et al. Reference Mony, Macgregor, Ivens, Rojas, Cowton, Young, Horn and Matthews2014). The short stumpy trypomastigotes are then ingested when the tsetse fly takes a blood meal. In the fly, the parasites differentiate into procyclic forms that colonize the midgut, where they further multiply and differentiate into mesocyclic trypomastigotes that migrate to the salivary glands of the tsetse fly. In the salivary glands the parasite transforms into a proliferating epimastigote form. Finally these multiply and differentiate into non-proliferating metacyclic trypomastigotes that acquire a variant surface glycoprotein (VSG) coat and detach from the epithelium to be injected into the mammalian host (Matthews et al. Reference Matthews, Ellis and Paterou2004; Roditi and Lehane, Reference Roditi and Lehane2008; Field and Carrington, Reference Field and Carrington2009; Lacomble et al. Reference Lacomble, Vaughan, Gadelha, Morphew, Shaw, McIntosh and Gull2010; Langousis and Hill, Reference Langousis and Hill2014).
Trypanosoma brucei lifecycle: The life cycle of the extracellular protozoan parasite T. brucei begins when trypanosomes are injected into the blood of a mammal by a tsetse fly. When injected into the mammalian host, metacyclic parasite's first transform in a long slender trypomastigote form that multiplies by binary fission. Next, it differentiates into a non-proliferating short stumpy trypomastigote form. When ingested by the insect vector, short-stumpies differentiate into the procyclic form, and colonize the midgut of the tsetse fly. After migration to the salivary glands, they assume epimastigote forms. Finally they differentiate into the metacyclic trypomastigote form, which is able to infect mammals.
HAT is a disease that occurs in distinct geographic locations in Africa. Occurring mostly in West- and Central Africa, T. b. gambiense is responsible for 98% of all human infections (Fig. 2). It causes a chronic disease, which progresses gradually and only has fatal outcome after several years of infection. In contrast T. b. rhodesiense infection occurs primarily in East Africa and causes an extremely virulent disease, resulting in death within weeks or months after infection. In addition to the human reservoir, both T. b. gambiense and T. b. rhodesiense have an animal reservoir, which hampers control and eradication of the disease (Njiokou et al. Reference Njiokou, Nimpaye, Simo, Njitchouang, Asonganyi, Cuny and Herder2010; Simarro et al. Reference Simarro, Diarra, Ruiz Postigo, Franco and Jannin2011; WHO, 2012). Over the last two decades, a huge effort has been made to bring down the number of actual Sleeping Sickness cases, without however bringing new therapies to the field. Today, renewed and sustained control programmes have resulted in a dramatic drop in actual case reports, reducing the number of infections to less than 10 000 reported patients in 2010. Nevertheless the WHO estimates that 57 million people are still at risk of contracting T. b. gambiense HAT (WHO, 2012; Simarro et al. Reference Simarro, Cecchi, Franco, Paone, Diarra, Ruiz-Postigo, Fèvre, Mattioli and Jannin2012). Most important for future control strategies is the fact that no vaccination strategies against trypanosomosis exist, due to specific and non-specific parasitic defence mechanisms summarized below. Hence to date, and in the foreseeable future, HAT control must mainly rely on the intense combination vector control, diagnosis and treatment.
Distribution of HAT on the African continent: T. b. gambiense, depicted in blue, occurs primarily in West- and Central Africa and is responsible for 98% of current HAT infections. Trypanosoma b. rhodesiense, depicted in pink, occurs primarily in East Africa. Based on data from ‘Report of a WHO meeting on elimination of African trypanosomiasis’ 2012.
Sleeping Sickness encompasses two disease stages: during the early haemolymphatic stage the parasites proliferate in blood and lymphatic system, while in the late meningoencephalitic stage, the parasites penetrate the blood-brain barrier and invade the central nervous system (Sternberg, Reference Sternberg2004; Blum et al. Reference Blum, Schmid and Burri2006; MacLean et al. Reference MacLean, Odiit, Chisi, Kennedy and Sternberg2010). While accurate trypanosomosis diagnosis is hard by itself, due to generally low parasite numbers, the correct stage determination of the disease poses further difficulties, and is based mainly on CSF determination of lymphocyte counts (Ngoyi et al. Reference Ngoyi, Menten, Pyana, Philippe and Lejon2013). Hence, there is an urgent need for new disease staging methods that are more specific, easy-to-use and reliable under field conditions, as administration of late stage drugs to early stage patients can lead to drug toxicity complications. For example, the current drug used for late stage treatment during T. b. rhodesiense HAT Melarsoprol, has a high treatment failure and high patient lethality and should never be administered to first-stage HAT patents (Blum, Reference Blum, Nkunku and Burri2001).
IMMUNE EVASION MECHANISMS DEVELOPED BY BOTH THE MAMMALIAN HOST AND THE TRYPANOSOMES
As indicated above, trypanosomes are obligatory extracellular parasites that dwell in the blood and lymphatics of their mammalian host. While in general the host will be unable to eliminate the infections, various host–parasite interactions do occur that have evolved in such a way that parasite survival is ensured, and that prolonged host survival allows successful population transmission. Hence, both the mammalian host, as well as the trypanosomes have selectively acquired a number of defence strategies that allow optimizing the race for survival and transmission. These defence mechanisms include toxic serum factors, antibodies and cytokines from the host's side, and antigenic variation, immune modulation and immune destruction form the parasite's side.
Innate human trypanolytic factors and parasite defence mechanisms
While HAT is to be considered a very serious human infection, it has to be stressed that humans are safe from infection by most trypanosomes, due to the presence of trypanolytic serum factors that are capable of killing most trypanosomes with exception of HAT-causing species T. b. gambiense and T. b. rhodesiense. These factors can be considered as part of the innate immunity, as they are not specifically induced during infection but are an intrinsic component of normal human serum (NHS), although their actual mode of action is to quickly lyse trypanosome upon entry in a non-immune way (Alsford et al. Reference Alsford, Currier, Guerra-Assunção, Clark and Horn2014). The factors responsible for this innate resistance are called ‘trypanolytic factors’ TLF1 and TLF2 (Rifkin, Reference Rifkin1978). The activity of both factors is considered to be mediated by the presence of apolipoprotein apoL1. This molecule is part of a larger apolipoprotein family, and also provides trypanosome resistance to gorillas and baboons. Albeit chimpanzees seem to have lost the apoL1 encoding gene, rendering them highly susceptible to trypanosomosis in general (The Chimpanzee Sequencing and Analysis Consortium, 2005; Thomson et al. Reference Thomson, Genovese, Canon, Kovacsics, Higgins, Carrington, Winkler, Kopp, Rotimi, Adeyemo, Doumatey, Ayodo, Alper, Pollak, Friedman and Raper2014). Interestingly, compared to human apoL1, the old world monkey homologue was shown to be more potent, rendering baboons even resistant to the human pathogenic T. b. rhodesiense parasite (Lugli et al. Reference Lugli, Pouliot, Portela, Loomis and Raper2004).
TLF-1 forms complexes with high-density lipoprotein (HDL), apolipoprotein L1 (apoL1), Haptoglobin-related protein (Hpr), apolipoprotein A1 (apoA1) and haptoglobin (Hp) while TLF-2 forms lipid-poor complexes with polyclonal IgM antibodies, apoL1, apoAi1 and Hpr (Rifkin, Reference Rifkin1978; Tomlinson and Raper, Reference Tomlinson and Raper1996; Raper et al. Reference Raper, Fung, Ghiso, Nussenzweig and Tomlinson1999, Reference Raper, Portela, Lugli, Frevert and Tomlinson2001; Vanhamme et al. Reference Vanhamme, Paturiaux-Hanocq, Poelvoorde, Nolan, Lins, Van Den Abbeele, Pays, Tebabi, Van Xong, Jacquet, Moguilevsky, Dieu, Kane, De Baetselier, Brasseur and Pays2003; Vanhollebeke and Pays, Reference Vanhollebeke and Pays2010a ; Pays et al. Reference Pays, Vanhollebeke, Uzureau, Lecordier and Pérez-Morga2014). It is generally accepted that TLF-2 is the main lytic factor in NHS (Raper et al. Reference Raper, Nussenzweig and Tomlinson1996), although it is still not clear how TLF2 is recognized/bound by the parasite. Even a recent screen using an RNAi approach did not yield an answer as to how trypanosomes take-up TLF2 (Lecordier et al. Reference Lecordier, Uzureau, Tebabi, Pérez-Morga, Nolan, Schumann Burkard, Roditi and Pays2014). The presence of a low affinity receptor (Drain et al. Reference Drain, Bishop and Hajduk2001) as well as a potential scavenger receptor (Green et al. Reference Green, Del Pilar Molina Portela, St Jean, Lugli and Raper2003) have been reported, however without providing adequate information that would allow the exact understanding of TLF2 functioning. In addition, preliminary data from our own group have indicated that a lectin-like interaction involving the complex TLF2/IgM carbohydrate side-chains might be involved (Magez et al. unpublished data). However, taken the complexity and instability of TLF2, the exact mode of TLF2 uptake remains to be elucidated, posing a challenge in the full understanding of the NHS anti-trypanosome activity. In contrast to the TLF2 situation, the uptake of TLF1 is much better understood. Here, the haptoglobin-related protein Hpr facilitates uptake of TLF1 via the trypanosome haptoglobin–haemoglobin receptor (HpHbR) responsible for parasite haeme supply (Vanhollebeke et al. Reference Vanhollebeke, De Muylder, Nielsen, Pays, Tebabi, Dieu, Raes, Moestrup and Pays2008). This receptor is however unable to discriminate between haptoglobin and haemoglobin (Hp–Hb) complexes and TLF1–Hpr–Hb (HDL) complexes. This results in the fact that in serum of ‘healthy’ individuals, TLF1 uptake might be virtually absent, as Hb concentrations usually exceed those of Hpr by a factor 100 (Vanhollebeke and Pays, Reference Vanhollebeke and Pays2010b ). When endocytosed (trough TLF2 and/or TLF1), apoL1 exhibits pore-forming activities in the lysosome, leading to osmotic imbalance and a disruption of the lysosomal membrane. This in turn causes uncontrolled swelling and parasite death (Hager et al. Reference Hager, Pierce, Moore, Tytler, Esko and Hajduk1994; Pays et al. Reference Pays, Vanhollebeke, Vanhamme, Paturiaux-Hanocq, Nolan and Perez-Morga2006). Despite the vast body of literature available to date on the biological activity of TLF1, a critical note needs reminding: detailed analysis of genetically modified parasites that lack the receptor HpHbR shows that they are still fully susceptible to TLF2 lysis, and alteration of their TLF1 mediated lysis pattern is only observed in in vitro culture. In addition, these mutants are fully sensitive to NHS lysis in physiological concentrations and an altered phenotype was observed only in conditions in which extreme low serum conditions are used, i.e. less than 1% NHS (Vanhollebeke et al. Reference Vanhollebeke, De Muylder, Nielsen, Pays, Tebabi, Dieu, Raes, Moestrup and Pays2008). These findings confirm that under normal physiological conditions, the role of TLF1 and the HpHb receptor might be minimal.
As mentioned, T. b. rhodesiense and T. b. gambiense are able to resist lysis by NHS. Hence in contrast to other trypanosomes, these parasites must have acquired resistance mechanisms that provide a defence against both TLF1 and TLF2, or against the common active compound, i.e. apoL1. Resistance of T. b. rhodesiense is not constitutive, but is induced upon the activation of transcription of a gene encoding resistance, termed serum resistance associated or sra (De Greef et al. Reference De Greef, Imberechts, Matthyssens, Van Meirvenne and Hamers1989; Van Xong et al. Reference Van Xong, Vanhamme, Chamekh, Chimfwembe, Van Den Abbeele, Pays, Van Meirvenne, Hamers, De Baetselier, Pays and Gene1998). SRA resembles a truncated version of VSG, which is located in the endocytic pathway (Shiflett et al. Reference Shiflett, Faulkner, Cotlin, Widener, Stephens and Hajduk2007). Resistance to TLF-1 is conferred by SRA interaction with apoL1 in the lysosome (Vanhamme et al. Reference Vanhamme, Paturiaux-Hanocq, Poelvoorde, Nolan, Lins, Van Den Abbeele, Pays, Tebabi, Van Xong, Jacquet, Moguilevsky, Dieu, Kane, De Baetselier, Brasseur and Pays2003). As T. b. rhodesiense is resistant to NHS lysis containing both TLF1 and TLF2, it is presumed that SRA has a similar manner of inhibiting apoL1 entering the parasite through TLF-2 uptake. One critical note that needs to be made here is the fact that recently more evidence is emerging that also human infective T. b. rhodesiense parasites exist that do not have the SRA (Enyaru et al. Reference Enyaru, Matovu, Nerima, Akol and Sebikali2006). Hence, as usual in the biology of host–parasite interaction, the full picture of NHS resistance might be more complicated than initially proposed.
As compared to T. b. rhodesiense, the situation in T. b. gambiense is even more complex. Important to mention here is that two different types of T. b. gambiense parasites have been characterized. Tbg Type 1 is the most common of the two and is characterize by a constitutive resistance to NHS. Tbg Type 2 on the other hand is characterized by an inducible level of NHS resistance, much like what is observed in T. b. rhodesiense. As both T. b. gambiense types lack SRA, and so far no common mechanisms for the resistance phenotype has been described, it appears that throughout evolution trypanosomes have acquired multiple times independent mechanisms to resist the aopL1 activity of NHS (Capewell et al. Reference Capewell, Veitch, Turner, Raper, Berriman, Hajduk and MacLeod2011). With respect to Type 1 Tbg it was recently shown that a truncated VSG-like T. b. gambiense-specific glycoprotein (TgsGP), located in the endocytic compartment is crucial for apoL1 resistance, as depletion of the TgsGP gene rendered T. b. gambiense susceptible to TLF1, apoL1 and NHS lysis (Capewell et al. Reference Capewell, Clucas, DeJesus, Kieft, Hajduk, Veitch, Steketee, Cooper, Weir and MacLeod2013; Uzureau et al. Reference Uzureau, Uzureau, Lecordier, Fontaine, Tebabi, Homblé, Grélard, Zhendre, Nolan, Lins, Crowet, Pays, Felu, Poelvoorde, Vanhollebeke, Moestrup, Lyngsø, Pedersen, Mottram, Dufourc, Pérez-Morga and Pays2013). In contrast to SRA, this mechanism does not involve direct apoL1 neutralization despite the fact that TgsGP and apoL1 co-localized, but rather has a function is altering membrane fluidity in the endocytic compartments (Uzureau et al. Reference Uzureau, Uzureau, Lecordier, Fontaine, Tebabi, Homblé, Grélard, Zhendre, Nolan, Lins, Crowet, Pays, Felu, Poelvoorde, Vanhollebeke, Moestrup, Lyngsø, Pedersen, Mottram, Dufourc, Pérez-Morga and Pays2013). Interestingly, TgsGP is absent from Tbg Type 2, indicating that T. b. gambiense needs to rely on multiple independent mechanisms to ensure NHS resistance (Radwanska et al. Reference Radwanska, Claes, Magez, Magnus, Perez-morga, Pays and Büscher2002; Gibson et al. Reference Gibson, Nemetschke and Ndung'u2010). A second resistance feature is a reduction in apoL1 sensitivity through cysteine protease activity and lower early endosomal pH (Uzureau et al. Reference Uzureau, Uzureau, Lecordier, Fontaine, Tebabi, Homblé, Grélard, Zhendre, Nolan, Lins, Crowet, Pays, Felu, Poelvoorde, Vanhollebeke, Moestrup, Lyngsø, Pedersen, Mottram, Dufourc, Pérez-Morga and Pays2013). Finally, a third resistance mechanism proposed to be involved here is the reduced uptake of TLF1, due to a single amino acid substitution in the HpHb receptor, which ablates binding and subsequent endocytosis (DeJesus et al. Reference DeJesus, Kieft, Albright, Stephens and Hajduk2013; Higgins et al. Reference Higgins, Tkachenko, Brown, Reed, Raper and Carrington2013; Uzureau et al. Reference Uzureau, Uzureau, Lecordier, Fontaine, Tebabi, Homblé, Grélard, Zhendre, Nolan, Lins, Crowet, Pays, Felu, Poelvoorde, Vanhollebeke, Moestrup, Lyngsø, Pedersen, Mottram, Dufourc, Pérez-Morga and Pays2013). Indeed, Tbg Type 1 intrinsic NHS resistance coincides with the reduced capacity of TLF1 uptake and while this might not be relevant in ‘normal’ NHS conditions (see above), it has been speculated that in co-infection condition were malaria-induced hypohaptoglobinaemia occurs, reduced TLF1 uptake could help trypanosomes to overcome NHS lysis due to reduced intracellular TLF1 accumulation (Pays et al. Reference Pays, Vanhollebeke, Uzureau, Lecordier and Pérez-Morga2014). Recent evidence comparing NHS resistant and NHS sensitive T. b. gambiense Type 2 strains has however show that here the phenotype is independent of TLF1–HpHb binding and uptake capacity. In addition, it was shown that Tbg Type 2 NHS resistance is independent of the expression site used, differentiating this activity mechanistically from the BES-associated SRA activity observed in T. b. rhodesiense NHS resistance (Capewell et al. Reference Capewell, Veitch, Turner, Raper, Berriman, Hajduk and MacLeod2011). Taken these most recent data, one could pose the critical question is to whether the reduced uptake of TLF1 by Tbg Type 1 measured in vivo really correlates with NHS resistance, or merely coincides, while having little or no biological relevant per se in an in vivo setting.
Additional innate host–parasite interactions
Upon successful infection of a mammalian host by trypanosomes, a range of innate immune responses will be initiated that serve to hamper parasite growth. These mechanisms are in large connected to macrophage activation, inflammatory cytokine secretion and iNOS activation. These mechanisms have been reviewed recently in detail by Beschin et al. Reference Beschin, Van Den Abbeele, De Baetselier and Pays2014. With respect to the specific innate immune aspect of trypanosome–host interactions, recent discoveries of the biological significance of the ESAG4 gene family need to highlighted here. ESAG4, belongs to a large gene subfamily encoding approximately 80 members of T. brucei adenylate cyclases, of which most are expressed constitutively (Alexandre et al. Reference Alexandre, Paindavoine, Tebabi, Halleux, Steinert and Pays1990, Reference Alexandre, Paindavoine, Hanocq-Quertier, Paturiaux-hanocq, Tebabi and Pays1996). These transmembrane receptor-like enzymes are activated under stress and among other, are implicated in cytokinesis (Salmon et al. Reference Salmon, Bachmaier, Krumbholz, Kador, Gossmann, Uzureau, Pays and Boshart2012a ). Recently it was shown that adenylate cyclases inhibit the early host innate immune response by inhibiting TNF production of liver-associated myeloid cells. By generating a double negative mutant for ESAG4, adenylate cyclase activity was reduced by 50%, which resulted in reduced parasite growth and significantly longer host survival time (Salmon et al. Reference Salmon, Vanwalleghem, Morias, Denoeud, Krumbholz, Lhommé, Bachmaier, Kador, Gossmann, Dias, De Muylder, Uzureau, Magez, Moser, De Baetselier, Van Den Abbeele, Beschin, Boshart and Pays2012b ). To underline the crucial importance of TNF in this event, experiments were repeated in TNF deficient mice, showing an abrogation of the phenotype. The diversification and abundance of adenylate cyclases in trypanosomes could be indicative for the fact that this trait to aid in growth and manipulate the immune system is essential. As adenylate cyclase synthesis is directly implicated in the inhibition of the early innate immune response, one might consider their expression as an escape mechanism. Indeed, due to the inhibition of TNF, a higher peak parasitaemia can favour transmission to the insect vector at this stage. However, transmission to the insect vector is also ensured by prolonged survival. In this aspect adenylate cyclase production must be tightly regulated in a natural host–parasite interaction setting, as accelerated parasite growth could in turn result in early host death, limiting subsequent parasite transmission potential.
Antigenic variation: a trypanosomes defence against the host adaptive immunity
The surface of the long slender bloodstream form is densely packed with 107 copies of a single VSG attached to the membrane by a glycosylphosphatidylinositol (GPI) anchor. These VSGs are highly immunogenic and enable the host to mount an effective humoral anti-VSG response, putting the parasite under continuous immune pressure. In order to deal with this pressure, trypanosomes have evolved a system called antigenic variation. This system comprises a frequent switching of the entire VSG coat, allowing the continuous evasion of new antibody attacks. Simplified, during the first ascending phase of a parasitaemia wave the majority of the parasites express the same VSG and are consequently of the major variable antigenic type (VAT) (Fig. 3). Approximately 0 × 1–1% of trypanosome divisions then produces a new VAT, thus expressing a different VSG (Robinson et al. Reference Robinson, Burman and Melville1999; Hall et al. Reference Hall, Wang and Barry2013). These new ‘antigenically distinct’ trypanosomes multiply and overgrow the first VAT, giving rise to a subsequent parasitaemia wave. This process is repeated multiple times and results in the development of a chronic infection (Pays et al. Reference Pays, Lips, Nolan, Vanhamme and Pérez-Morga2001; Baral, Reference Baral2010; Schwede and Carrington, Reference Schwede and Carrington2010; Hall et al. Reference Hall, Wang and Barry2013). Recently, studies by Hall et al. involving T. brucei VSG cDNA sequencing have shown that an African trypanosome infection comprises much more diverse parasite populations than originally described. Indeed, obtained results showed that an individual growth peak in mice can contain at least 15 distinct variants (Hall et al. Reference Hall, Wang and Barry2013). It is estimated that this number is even greater in natural hosts.
Simulating the revised concept of antigenic variation during mammalian T. brucei infection. In contrast to the previously published models of parasitaemia waves consisting of a single VAT it is now believed that parasitaemia waves comprise much more diverse parasite populations, as each individual growth peak in mice can contain at least 15 distinct variants.
Transcription of VSG occurs from specialized subtelomeric transcription units known as the bloodstream expression sites (BES). BESs are large poly-cistronic transcription units, which additionally harbour genes termed expression site-associated genes (ESAGs) (Kooter et al. Reference Kooter, van der Spek, Wagter, d'Oliveira, van der Hoeven, Johnson and Borst1987; Pays et al. Reference Pays, Lips, Nolan, Vanhamme and Pérez-Morga2001; Vanhamme et al. Reference Vanhamme, Pays, McCulloch and Barry2001a ; Borst, Reference Borst2002; Berriman et al. Reference Berriman, Hall, Sheader, Bringaud, Tiwari, Isobe, Bowman, Corton, Clark, Cross, Hoek, Zanders, Berberof, Borst and Rudenko2002), including the transmembrane proteins adenylyl cyclase (ESAG4) mentioned above (Pays et al. Reference Pays, Lips, Nolan, Vanhamme and Pérez-Morga2001) and the heterodimeric surface receptor for host transferrin (ESAG7/6). The trypanosome has an estimated 10–40 BESs harbouring VSGs. In addition the parasite can access ~1000 silent (pseudo) VSG genes scattered in the genome (Vanhamme et al. Reference Vanhamme, Pays, McCulloch and Barry2001b ). These silent genes and pseudogenes can appear into the BES by gene conversion mechanisms. VSG expression is mono-allelic, hence the appearance of a uniform VSG coat on the parasite surface. Recombination events responsible for a switch in VSG include telomere exchange, duplicative gene conversion and segmental or partial gene conversion (Vanhamme et al. Reference Vanhamme, Pays, McCulloch and Barry2001b ; Morrison et al. Reference Morrison, Marcello and McCulloch2009; Horn and McCulloch, Reference Horn and McCulloch2010; Rudenko, Reference Rudenko2011). These three mechanisms involve the exchange of genetic material between an active and a silent BES, while remaining under the original active BES promoter and remaining in the transcription body, the unique transcription apparatus where properly processed mRNA for VSG production is generated (Navarro and Gull, Reference Navarro and Gull2001). This mechanism of VSG switching is believed to be a crucial adaptation to long-term survival in a given host, as all the ESAG gene products remain the same, while only the VSG within the given BES is altered. Using this gene rearrangement, also antigenically distinct mosaic VSGs can be fabricated from silent VSG genes and pseudogenes by means of partial gene conversion, a process termed mosaicism. VSG expression follows a loose hierarchy and mosaicism occurs increasingly as the infection proceeds, contributing immensely to antigenic variation and infection chronicity (Hall et al. Reference Hall, Wang and Barry2013). Hence, the potential for trypanosome antigenic variation is enormous, and constitutes the major immune escape mechanism.
Besides VSG switching to gene rearrangement, a switch in active VSG BES transcription can also result in a new VSG variant expression. However, in this case all the ESAGs located on the BES are altered as well, meaning that the expression of for example the transferrin receptor will be altered to a new homologue as well. The latter BES switch mechanism has most likely evolved as an adaptation that allows the infection of a wide host range of mammals (Pays et al. Reference Pays, Lips, Nolan, Vanhamme and Pérez-Morga2001; Morrison et al. Reference Morrison, Marcello and McCulloch2009). It involves multiple mechanisms including transcription silencing chromatin remodelling and regulation of pre-mRNA elongation, although the exact mode of action still remains to be elucidated (Hughes et al. Reference Hughes, Wand, Foulston, Young, Harley, Terry, Ersfeld and Rudenko2007; Figueiredo et al. Reference Figueiredo, Janzen and Cross2008; Li et al. Reference Li, Yang, Lun, Ma, Xi, Chen, Song, Kang and Yang2009; Landeira et al. Reference Landeira, Bart, Van Tyne and Navarro2009).
Additional VSG-mediated defence mechanisms
Besides antigenic variation, it seems that trypanosomes might have evolved a number of addition mechanisms that provide a certain level of protection against antibody-mediated attack. First, while complement-mediated lysis is a well-established in vitro method to determine the VSG-specificity of a given antibody response, it is not clear whether this system effectively operates in vivo. Indeed, mice that lack the C5 component of the complement cascade exhibit parasitaemia control patters that are similar to fully immune competent mice. Taken the thickness of the VSG coat (approx. 200 Å) and the vast N-linked and GPI-associated carbohydrate barrier, one could argue that the final complement complex would be unable to efficiently target the lipid membrane of the parasite. Secondly, trypanosomes have developed an endocytosis mechanism that allows antibody clearance from the VSGs. This means that VSG-antibody complexes are endocytosed in the flagellar pocket and antibodies are degraded in the lysosome after which the VSG is recycled back to the surface coat (Barry, Reference Barry1979; McLintock et al. Reference McLintock, Turner and Vickerman1993; Engstler et al. Reference Engstler, Pfohl, Herminghaus, Boshart, Wiegertjes, Heddergott and Overath2007). This mechanism would provide a way to escape antibody-mediated immune attack at low to moderate antibody concentrations and could therefore pose as an escape mechanism promoting the survival of individual cells, possibly supporting their transmission to the insect vector (Engstler et al. Reference Engstler, Pfohl, Herminghaus, Boshart, Wiegertjes, Heddergott and Overath2007).
Immune modulation: undermining the long-term immunity of the host
As if all the above described escape mechanisms were not sufficient, trypanosomes have invested in yet another way to ensure infection chronicity and hence successful parasite transmission. Trypanosomes modulate the host immune system in various ways so that the capacity of the host to mount an efficient immune response is undermined.
Infection-induced immune suppression is long considered as a hallmark of Trypanosomosis. Early studies on African trypanosomes show that the parasite overwhelms the host immune system with a massive antigenic load. This was shown to be associated with immune depression and polyclonal lymphocyte activation, and occurs during rodent, livestock and human infections (Ormerod, Reference Ormerod, Mulligan, Potts and Kershaw1970; Goodwin et al. Reference Goodwin, Green, Guy and Voller1972; Mansfield and Wallace, Reference Mansfield and Wallace1974; Diffley, Reference Diffley1983; Oka et al. Reference Oka, Yabu, Ito and Takayanagi1988). The polyclonal lymphocyte activation depletes antigen-reactive lymphocyte populations and can exhaust and supress B and T cells in the induction of antigen specific immunity against subsequent trypanosome variants or even unrelated antigens.
In both mice and cattle, B cells seem to play an important role in host protection, despite their limitation in VSG specificity (Corsini et al. Reference Corsini, Clayton, Askonas and Ogilvie1977; Campbell et al. Reference Campbell, Esser and Weinbaum1977; de Gee et al. Reference De Gee, Mccann and Mansfield1983; Guirnalda et al. Reference Guirnalda, Murphy, Nolan and Black2007; Magez et al. Reference Magez, Schwegmann, Atkinson, Claes, Drennan, De Baetselier and Brombacher2008). Although additional host factors contribute to parasite control (see further), B cells seem to be essential for post-peak parasite removal and prolonged survival (Magez et al. Reference Magez, Schwegmann, Atkinson, Claes, Drennan, De Baetselier and Brombacher2008). Mouse models of human and animal trypanosomosis show that multiple Trypanosome species cause a sustained loss in splenic and bone marrow B cell populations (Baltz et al. Reference Baltz, Baltz, Giroud and Pautrizel1981; Radwanska et al. Reference Radwanska, Guirnalda, De Trez, Ryffel, Black and Magez2008; Bockstal et al. Reference Bockstal, Geurts and Magez2011a ; Obishakin et al. Reference Obishakin, de Trez and Magez2014; La Greca et al. Reference La Greca, Haynes, Stijlemans, De Trez and Magez2014). In the spleen, microarchitecture is disrupted and different B cell subsets are undergoing apoptosis (Radwanska et al. Reference Radwanska, Guirnalda, De Trez, Ryffel, Black and Magez2008; Bockstal et al. Reference Bockstal, Guirnalda, Caljon, Goenka, Telfer, Frenkel, Radwanska, Magez and Black2011b ). In addition, B cell lymphopoiesis in the bone marrow is affected, preventing replenishment of the splenic mature B cell pool (Bockstal et al. Reference Bockstal, Guirnalda, Caljon, Goenka, Telfer, Frenkel, Radwanska, Magez and Black2011b ). The trypanosome hereby prevents the induction of a protective memory response, an additional insurance for infection chronicity. Vaccination experiments against unrelated pathogens have also shown that trypanosomes destroy previously induced vaccine-induced memory. Indeed, vaccine efficacy was abolished after the host was infected with T. brucei (Onah and Wakelin, Reference Onah and Wakelin2000; Radwanska et al. Reference Radwanska, Guirnalda, De Trez, Ryffel, Black and Magez2008). If these findings regarding B cell destruction would also hold true for field Trypanosomosis, this would complicate not only anti-Trypanosomosis vaccination, but generally any vaccination programme in Sub-Saharan Africa, implying the need of re-vaccinating HAT patients after treatment.
In light of these results, Lejon et al. (2014) conducted a field trial in Democratic Republic of Congo on T. b. gambiense infected individuals. In this study, HAT patients had higher percentages of peripheral memory T and B cells than healthy controls. In addition they investigated the immunological memory by measuring anti-measles antibodies of vaccinated subjects before and after anti-trypanosomosis treatment. Anti-measles antibodies were significantly lower in HAT patients compared to controls, and although they remained lower after treatment, the levels were above the cut off value assumed by the manufacturer to provide protection. As the authors state themselves, antibody quantification is a sub-optimal tool for the investigation of immunological memory, as they do not reflect the presence of antibody-secreting memory B cells and could be elevated in spite of immunological suppression (Onah and Wakelin, Reference Onah and Wakelin2000). In addition, polyclonal B cell activation can replace the measles-specific antibodies by low-affinity cross-reactive antibodies and hence a functional characterization is necessary to determine if the antibodies maintain their protective capacity. Despite the previously mentioned shortcomings of this study, these results could indicate that destruction of the B cell memory compartment might not be as big an issue in humans as it is in mice, at least in the case of T. b. gambiense infection. This would be encouraging news for vaccination campaigns throughout HAT-endemic regions in Africa. Further investigation into a functional antibody assay should confirm these results. In addition, this phenomenon needs to be investigated in the more virulent T. b. rhodesiense infections, as previous results indicated a correlation between immune depression and parasite load (Obishakin et al. Reference Obishakin, de Trez and Magez2014).
Murine models have also shown that aside from B cells, VSG-specific Th1 cells and IFNγ regulate another major component of host resistance to African trypanosomes (Hertz et al. Reference Hertz, Filutowicz and Mansfield1998; Paulnock et al. Reference Paulnock, Freeman and Mansfield2010). The parasites first activate macrophages and dendritic cells through the production of pathogen-associated molecular patterns (PAMPS) which comprise the GPI anchors of shed VSG and CpG DNA (Mansfield and Paulnock, Reference Mansfield and Paulnock2005). Particularly splenic dendritic cells seem to be an important cell subset for the induction of VSG-specific T cell responses (Dagenais et al. Reference Dagenais, Freeman, Demick, Paulnock and Mansfield2009). Th1 cells are an important source of IFNγ, which is necessary to activate macrophages for the production of trypanolytic factors (Magez et al. Reference Magez, Radwanska, Beschin, Sekikawa and De Baetselier1999; Drennan et al. Reference Drennan, Stijlemans, Van Den, Quesniaux, Barkhuizen, De Baetselier, Ryffel and Magez2005). In addition the trypanosomes induce macrophages to produce suppressor effector molecules like prostaglandins, which inhibit the VSG-specific T cells to proliferate (Schleifer and Mansfield, Reference Schleifer and Mansfield1993) and constitute another way for the parasite to ensure its survival.
CONCLUSION AND DISCUSSION
African trypanosomes have evolved multiple mechanisms to ensure their survival in the host and consequently establish a chronic infection (Fig. 4). These immune evasion mechanisms have prevented the design of a prophylactic vaccine so far. As the main reservoir for T. b. rhodesiense are African domestic animals like cattle and African wildlife, the full eradication of the parasite from this reservoir is impossible. Therefore the only way to protect the human population against re-infection is through prophylactic vaccination. Over the last decades different vaccination strategies have been designed, but not a single one obtained 100% sterile immunity or made its way to a field trial (La Greca and Magez, Reference La Greca and Magez2011). Due to the parasite's antigenic variation system, vaccination against VSG is impossible. Vaccination protocols involving invariant antigens such as invariant surface glycoproteins like the transferrin receptor ESAG6/7 (Lança et al. Reference Lança, Pires de Sousa, Atouguia, Prazeres, Monteiro and Sousa Silva2011), the flagellar pocket proteins (Mkunza et al. Reference Mkunza, Aloho and Powell1995; Radwanska et al. Reference Radwanska, Magez, Dumont, Pays, Nolan and Pays2000) only rendered animals partially protected against low parasite dose challenge. In addition to the antigenic variation system, immunosuppression, and in particular depletion of immune memory, could be another way that parasites ensure infection chronicity. Trypanosomes could have invested in these additional evasion strategies to block any previous antibody reaction against their newly synthesized (mosaic) VSGs, are antigenically distinct (Hall et al. Reference Hall, Wang and Barry2013).
Schematic overview of the escape mechanisms used by African trypanosomes. (I) VSG associated immune evasion strategy (antigenic variation) encompasses the most important escape mechanism used by trypanosomes. In addition, antibody clearance by the VSG coat can aid escape on the single cell level. (II) In contrast to other trypanosome species, human-infective trypanosomes can resist lysis by human serum factors TLF-1 and TLF-2 via different strategies. (III) Adenylate cyclase, produced by the trypanosome can inhibit a part of the innate immune response. (IV) Trypanosomes modulate their hosts immune response by exhausting and supressing the host immune system.
In animal Trypanosomosis it would be useful to protect the host from disease-associated complications, as many animals can harbour infection without developing severe symptoms, suggesting the deadly outcome in human infections to be a consequence of the host reaction. An alternative to sterile immunity is therefore anti-disease vaccination, to target the infection-associated pathology. This strategy has given some positive results in an experimental setting against trypanosome cysteine proteases (Authie et al. Reference Authie, Biulange, Muteti, Lalmanach, Gauthier and Musoke2001), but has so far not resulted in a field application.
Given that to date, experimental vaccine attempt have not resulted in any promising results, it must be mentioned that the use of murine model for Trypanosomosis in initial experimental settings might not represent the ideal host–parasite context for research regarding vaccination against trypanosomes. However, this model has given us valuable insights into host parasite interactions and biology of antigenic variation. Hence, future efforts are needed to validate the use of these mouse models in trypanosome vaccine research, and, alternative models that better reflect the parasite–host interaction will need to be evaluated as well.
ACKNOWLEDGEMENTS
This work was supported by the Interuniversity Attraction Poles Programme – Belgian Science Policy, the Research Foundation Flanders (FWO) and the Vrije Universiteit Brussel (VUB).
FINANCIAL SUPPORT
This work was supported by the Fonds voor Wetenschappelijk Onderzoek (FWO) Project fwoTM597 and Krediet aan Navorsers grant #1517313N/KN251.
