Molecular epidemiology of Chagas disease

The T. cruzi parasite comprises a heterogeneous population that displays clonal and/or sexual propagation due to the different cycles of transmission, and the possibility of genetic exchange that can be found in nature and has been previously reported in vitro (Gaunt et al., Reference Gaunt, Yeo, Frame, Stothard, Carrasco, Taylor, Mena, Solis, Miles, Acosta, de Arias and Miles2003; Sturm et al., Reference Sturm, Vargas, Westenberger, Zingales and Campbell2003; Westenberger et al., Reference Westenberger, Campbell, Sturm and Barnabé2005; Llewellyn et al., Reference Llewellyn, Miles, Carrasco, Lewis, Yeo, Vargas, Torrico, Diosque, Valente, Valente and Gaunt2009; Ramírez et al., Reference Ramírez, Guhl, Messenger, Lewis, Montilla, Cucunuba and Llewellyn2012; Ramírez and Llewellyn, Reference Ramírez and Llewellyn2014; Brenière et al., Reference Brenière, Barnabé and Waleckx2016). T. cruzi is genetically diverse and is classified into at least six DTUs known as TcI–TcVI based on different molecular markers and biological features. TcI and TcII are considered as two pure lines with a point of evolutionary separation of approximately 1–3 million years ago. Some authors consider TcIII and TcIV to be the result of hybridization between TcI and TcII, whereas TcV and TcVI are generally accepted to be the result of hybridization between TcII and TcIII (Gaunt et al., Reference Gaunt, Yeo, Frame, Stothard, Carrasco, Taylor, Mena, Solis, Miles, Acosta, de Arias and Miles2003; Sturm et al., Reference Sturm, Vargas, Westenberger, Zingales and Campbell2003; Westenberger et al., Reference Westenberger, Campbell, Sturm and Barnabé2005; Zingales et al., Reference Zingales, Miles, Campbell, Tibayrenc, Macedo, Teixeira, Schijman, Llewellyn, Lages-Silva, Machado, Andrade and Sturm2012; Ramírez et al., Reference Ramírez, Hernández, Cucunubá, Montilla, Flórez, Zambrano and Parra2014; Brenière et al., Reference Brenière, Barnabé and Waleckx2016; Zingales, Reference Zingales2017). Recently, other authors have proposed divergent evolutionary pathways suggesting three different ancestors (Brenière et al., Reference Brenière, Barnabé and Waleckx2016). Future studies are needed to fully understand the emergence of T. cruzi DTUs. Over the last decade, two emergent genotypes have been reported; one associated with anthropogenic bats in Brazil, Colombia, Ecuador and Panama and designated TcBat (Ramírez et al., Reference Ramírez, Hernández, Cucunubá, Montilla, Flórez, Zambrano and Parra2014; Pinto et al., Reference Pinto, Ocaña-Mayorga, Tapia, Lobos, Zurita, Aguirre-Villacís, MacDonals, Villacis, Lima, Teixeira, Grijalva and Perkins2015), and a clonal genotype linked to human infections within TcI designated TcI_Dom (Llewellyn et al., Reference Llewellyn, Miles, Carrasco, Lewis, Yeo, Vargas, Torrico, Diosque, Valente, Valente and Gaunt2009; Ramírez et al., Reference Ramírez, Guhl, Messenger, Lewis, Montilla, Cucunuba and Llewellyn2012; Zumaya-Estrada et al., Reference Zumaya-Estrada, Messenger, Lopez-Ordonez, Lewis, Flores-Lopez, Martínez-Ibarra, Pennington, Cordon-Rosales, Carrasco, Segovia, Miles and Llewellyn2012; Sánchez and Ramírez, Reference Sánchez and Ramírez2013; Ramírez and Hernández, Reference Ramírez and Hernández2017). However, these emergent genotypes are not as yet considered true DTUs by international consensus.

The DTUs of T. cruzi are broadly distributed across the American continent in diverse ecotopes (domestic, peridomestic and sylvatic transmission cycles). Discrimination of the six DTUs has become an important issue in the molecular epidemiology of T. cruzi. There are many reports showing algorithms for the molecular characterization of these DTUs based on RAPDs, polymerase chain reaction (PCR)-RFLPs, qPCR, MLST, MLMT and DNA sequencing techniques, but to date there is no gold standard protocol that is accepted internationally for strain typing (Rozas et al., Reference Rozas, Doncker, Adaui, Coronado, Barnabé, Tibyarenc and Dujardin2007; Duffy et al., Reference Duffy, Bisio, Altcheh, Burgos, Diez, Levin, Favaloro, Freilij and Schijman2009; Lewis et al., Reference Lewis, Yeo, Llewellyn, Miles, Ma and Carrasco2009; Llewellyn et al., Reference Llewellyn, Miles, Carrasco, Lewis, Yeo, Vargas, Torrico, Diosque, Valente, Valente and Gaunt2009; Burgos et al., Reference Burgos, Diez, Vigliano, Bisio, Risso, Duffy, Cura, Brusses, Favaloro, Leguizamon, Lucero, Laguens, Levin, Favaloro and Schijman2010; Ramírez et al., Reference Ramírez, Guhl, Rendón, Rosas, Marin-Neto and Morillo2010; Yeo et al., Reference Yeo, Mauricio, Messenger, Lewis, Llewellyn, Acosta, Bhattacharyya, Diosque, Carrasco and Miles2011; Higuera et al., Reference Higuera, Guhl and Ramírez2013; Cura et al., Reference Cura, Duffy, Lucero, Bisio, Péneau, Jimenez-Coello, Calabuig, Gimenez, Valencia, Kjos, Santalla, Mahaney, Cayo, Nagel, Barcán, Málaga Machaca, Acosta, Brutus, Ocampo and Aznar2015). One of the most frequently used and reliable algorithms for T. cruzi typing has been reported by the research of Burgos and colleagues, Higuera and colleagues and Ramírez and colleagues, and has also been applied to the testing of clinical samples (mainly blood samples) (Burgos et al., Reference Burgos, Diez, Vigliano, Bisio, Risso, Duffy, Cura, Brusses, Favaloro, Leguizamon, Lucero, Laguens, Levin, Favaloro and Schijman2010; Ramírez et al., Reference Ramírez, Guhl, Rendón, Rosas, Marin-Neto and Morillo2010; Higuera et al., Reference Higuera, Guhl and Ramírez2013) (Fig. 1).

Fig. 1. Algorithm for DTU discrimination of T. cruzi.

Regarding the geographical distribution of T. cruzi DTUs, it is clear that TcI predominates in the terrestrial area from the south of North America to the north of Argentina and Chile, usually due to a sylvatic transmission cycle, but TcI is also a major cause of the disease in countries endemic for trypanosomiasis due to a domestic transmission cycle. By contrast, TcII, TcV and TcVI mainly occur in the southern cone, but this distribution may not be completely accurate since the area includes countries such as Bolivia for which data may be incomplete. TcIII and TcIV are generally characterized by their sylvatic transmission cycles in areas with tropical forest ecosystems (Carrasco et al., Reference Carrasco, Segovia, Llewellyn, Morocoima, Urdaneta-Morales, Martínez, Garcia, Rodríguez, Espinosa, de Noya, Díaz-Bello, Herrera, Fitzpatrick, Yeo, Miles and Feliciangeli2012; Guhl and Ramírez, Reference Guhl and Ramírez2013; Brenière et al., Reference Brenière, Barnabé and Waleckx2016). This is particularly illuminating given that there are distinct clinical differences between patients presenting with Chagas disease in these geographical regions. Strains appear to differ in terms of pathogenicity and the response to treatment. Both TcI and TcII–VI are associated with cardiac lesions in human infections, but it seems that only TcII, TcV and TcIV are associated with digestive tract lesions (Prata, Reference Prata2001), despite a report of digestive tract lesions in Colombia caused by TcI (Prata, Reference Prata2001). In general, however, TcI is considered to be less pathogenic with lower parasitaemias (Burgos et al., Reference Burgos, Bisio, Duffy, Levin, Schijman, Altcheh, Burgos Freilij, Valadares, Freitas, Macedo, Seidenstein, Macchi and Piccinali2007) and more chronic cases being asymptomatic compared with Chagas cases caused by TcII, TcV and TcVI in Argentina, Brazil, Chile, Paraguay and Uruguay (Luquetti et al., Reference Luquetti, Tavares, Siriano, Oliveira, Campos, Morais and Oliveira2015).

Additionally, some proteomic approaches have been conducted demonstrating differential protein expression among DTUs isolates (Telleria et al., Reference Telleria, Biron, Brizard, Demettre, Séveno, Barnabé, Ayala and Tibayrenc2010; Díaz et al., Reference Díaz, Torres and González2011). Moreover, there is an observed general partitioning of TcII–TcVI between sylvatic and domestic transmission cycles; with human disease cases associated with TcII, TcV and TcVI and TcIII, TcIV being predominantly sylvatic (Yeo et al., Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Zingales et al., Reference Zingales, Miles, Campbell, Tibayrenc, Macedo, Teixeira, Schijman, Llewellyn, Lages-Silva, Machado, Andrade and Sturm2012; Messenger et al., Reference Messenger, Ramirez, Llewellyn, Guhl and Miles2016). However, according to the relationship between the recent TcBat isolates and anthropogenic deaths reported in countries such as Colombia and Brazil, the zoonotic potential of this genotype is being considered (Ramírez et al., Reference Ramírez, Hernández, Cucunubá, Montilla, Flórez, Zambrano and Parra2014). In addition, TcII has also been isolated from the same type of bats from which TcBat originated, and TcII has been shown phylogenetically to be closely related to a common ancestor of the TcI genotype (Lima et al., Reference Lima, Espinosa-Álvarez, Pinto, Cavazzana, Pavan, Carranza, Lim, Campaner, Takata, Camargo, Hamilton and Teixeira2015b). Recent advances in next generation sequencing technologies have allowed researchers to obtain three complete T. cruzi genomes. The first strain to be fully sequenced was CL Brener (TcVI), which revealed a high degree of repetitive elements along the core genome (El-Sayed et al., Reference El-Sayed, Bartholomeu, Ghedin, Delcher, Blandin, Westenberger, Caler, Cerqueira, Haas, Crabtree, Feldblyum, Hou, Koo, Lacerda, Pai, Pop, Salzberg, Shetty, Simpson, Van Aken, Wortman, White, Fraser, Myler, Aggarwal, Worthey, Anupama, Attipoe, Cadag, Fazelina, Huang, Louie, Nelson, Parsons, Pentony, Rinta, Robertson, Seyler, Sisk, Vogt, Stuart, Nilsson, Tran, Branche, Arner, Bontempi, Darban, Edwards, Ferella, Kindlund, Kluge, McKenna, Mizuno, Ochaya, Tammi, Andersson, Campbell, Machado, Teixeira, Åslund, Pettersson, Bringaud, Burton, McCulloch, Mottram, Ward, Carrington, Sharma, Da Silveira, De Jong, Osoegawa, Englund, Frasch, Sanchez, Gull, Wickstead, Horn, Klingbeil, Levin, Lorenzi, Ramirez and Tarleton2005). Similarly, the recently sequenced Esmeraldo (TcII) and Sylvio X10 (TcI) genomes revealed the association between repetitive elements and mucin-like proteins that are closely associated with parasite cell invasion and survival, (Franzén et al., Reference Franzén, Ochaya, Sherwood, Andersson, Lewis, Llewellyn and Miles2011) offering new insight into the relationship between T. cruzi DTUs and pathogenicity (Andersson, Reference Andersson2011).

Pathophysiology of CCC

Chagas disease is currently considered a multifactorial disease of infectious origin, since several factors such as the mechanisms of action by which the parasite invades, the virulence factors of the parasite and the polymorphic factors of the host come into play, which together determine the severity of the clinical manifestations of the disease (Prata, Reference Prata2001). Below, the pathophysiological factors involved in the different presentations of the disease will be discussed, as well as a brief description of the manifestations involved.

The acute phase of T. cruzi infection is mediated by both innate and acquired host immunity. Initially, the infection is mediated by phagocytosis with highly virulent metacyclic trypomastigotes being ingested by macrophages. Although most are destroyed within phagocytic vacuoles, some amastigotes escape this immunological action and initiate intracellular replication, causing cell damage and rupture, allowing the parasite to spread through the bloodstream. This leads to the activation of a cascade of cytokines, cationic proteins, complement and transferrin proteins, with T. cruzi inducing the expression of interferons and cytokines such as interleukin (IL)-10 and tumour necrosis factor (TNF)-α (Sanjabi et al., Reference Sanjabi, Zenewicz, Kamanaka and Flavell2009; Teixeira et al., Reference Teixeira, Hecht, Guimaro, Sousa and Nitz2011). Regarding acquired immunity, certain humoral factors play an important role in control of the infection, among which immunoglobulin G (IgG)2b antibodies are pivotal, which also mediate inflammatory activity during the acute phase. The activation of immune cells, such as CD4+ and CD8+ cells, also plays an important role in acquired immunity (Teixeira et al., Reference Teixeira, Hecht, Guimaro, Sousa and Nitz2011).

At the moment the parasite enters the bloodstream, it is directed to the sites with the highest predilection, which in this case are the central nervous system, the cardiac and oesophageal regions, and the colon muscle tissue, as these are sites where the parasite can most readily traverse the vascular endothelium and initiate cellular parasitism (Rassi et al., Reference Rassi, Marin Neto and Rassi2017). The initial process of transmigration through the vascular endothelium prior to transport of the parasite to its target tissues, has been reported by in vitro studies in human cells, and is facilitated by chemokines such as bradykinin and chemokine ligand 2 (CCL2). However, the molecules through which the parasite interacts with the endothelium are yet to be determined (Coates et al., Reference Coates, Sullivan, Makanji, Du, Olson, Muller, Engman and Epting2013).

On reaching and invading its target organs within the host, the parasite undergoes active multiplication during a sustained immunological reaction by the cells that involves CD4+ T lymphocytes, CD8+ and B lymphocytes. This cellular reaction leads to direct induction of anti-trypanosome cytotoxicity, secretion of cytokines and the production of antibodies against the parasite. Inflammatory myocardial lesions of both chronically infected animal and human models are predominantly composed of CD8+ rather than Th1 CD4+ lymphocytes, and increased expression of genes responsible for the production of proinflammatory cytokines and chemokines (especially interferon (IFN)-γ and transforming growth factor (TGF)-β±) has been observed. Other investigators have reported reduced production of regulatory T cells and their cytokines, such as IL-10 and IL-17 (Reis et al., Reference Reis, Jones, Tostes, Lopes, Gazzinelli, Colley and McCurley1993; Sanjabi et al., Reference Sanjabi, Zenewicz, Kamanaka and Flavell2009; Guedes et al., Reference Guedes, Gutierrez, Silva, Dellalibera-Joviliano, Rodrigues, Bendhack, Rassi, Schmidt, Maciel, Marin- Neto and Silva2012; Cupello et al., Reference Cupello, Souza, Nogueira, Laranja, Sabino, Coelho, Oliveira and Paes2014). These findings are consistent with an imbalance demonstrated by the upregulation of Th1 cells and the downregulation of Treg cell activity (Fig. 2).

Fig. 2. Cytokine profile during T. cruzi infection. The acute phase shows the immune system response and the chronic phase shows the activation of cytokines.

A study using mice susceptible (BALB/c) and non-susceptible (C57BL/6) to T. cruzi infection was performed to compare the expression of CD4+ T cells in both the acute and chronic phases of the disease. During the chronic phase, a T. cruzi-specific product was amplified by PCR at higher quantities in the surviving BALB/c mice than in the C57BL/6 mice, suggesting that a greater number of parasites were present in the susceptible mice during this phase. Immunological analysis of the chronic phase samples revealed greater expression of Th1 and inflammatory cytokines, such as IFN-γ, TNF and IL-2, which triggered the low but significant activation of CD4+ T cells in the BALB/c mice. However, this cell type was not detected in the infected C57BL/6 mice, in which low levels of proinflammatory cytokines (IFN-γ and IL-2) were detected. Similarly, in BALB/c mice, the production of IL-10 and TGF-β increased, which was not observed in C57BL/6 mice. These results indicated that mice which survive the acute phase and then enter the chronic phase with parasites present in their cardiac tissue trigger a response associated with Th1 cells, but not a Th17 response (Sanoja et al., Reference Sanoja, Carbajosa, Fresno and Gironès2013).

It is also necessary to emphasize the importance of the Toll-like receptors (TLR) since they are responsible for immune recognition of parasites via pathogen-associated molecular patterns. T. cruzi has chemical structures that stimulate specific TLRs, which subsequently induce the production of nitric oxide and proinflammatory cytokines by monocytic cells. Among them, TLR-2 recognizes trypomastigote-derived glycosylphosphatidylinositol (tGPI) anchored in the mucin-like glycoproteins, TLR-4 recognizes the epimastigote glycoinositolphospholipid (eGIPL) that induces nuclear factor (NF)-κB, TLR-7 recognizes the parasitic RNA and TLR-9 recognizes DNA with abundant oligodeoxynucleotide unmethylated CpG motifs stimulating the cytochemical responses of both macrophages and dendritic cells (de Souza et al., Reference de Souza, de Carvalho and Barrias2010; Rodrigues et al., Reference Rodrigues, Oliveira and Bellio2012).

Other factors related to cardiac involvement are neurological or microvascular alterations, immune-mediated tissue injury and parasite-dependent damage. These play a secondary role in the development of cardiac lesions and complications. However, some authors consider these critical factors in the persistence of parasites and in the inflammatory reaction and the initiation and progression of chronic myocarditis (Tarleton and Zhang, Reference Tarleton and Zhang1999, Marin-Neto et al., Reference Marin-Neto, Rassi and Maciel2015). Immune-mediated tissue injury is caused by polymorphonuclear leucocyte infiltration and the production of deleterious cytokines, mechanisms that are probably triggered by the persistence of the parasite in the tissue. Autoimmunity generated by molecular mimicry with parasite antigens and the resulting polyclonal activation have been reported; however, validation of autoimmunity is difficult and therefore remains a controversial issue (Tarleton, Reference Tarleton1991, Reference Tarleton2003). Evidence for an additional potential autoimmune mechanism came from the detection of mitochondrial DNA from the parasite in the genome of chickens in a model in which infection was induced in the egg phase (Teixeira et al., Reference Teixeira, Hecht, Guimaro, Sousa and Nitz2011).

Metabolomic studies have revealed that the proteasome is the key protease in the generation of peptides for the presentation of antigens through MHC-I. In one study, it was shown that the biosynthesis of the immunoproteasome subunits Bl1, B2i and B5i, as well as PA28B, TAP1 and MCH-I (in macrophages via the SAPK/JNK signalling pathway) were downregulated in HeLa cells by T. cruzi, although the last three were not degraded by the parasite (Camargo et al., Reference Camargo, Faria, Kloss, Favali, Kuckelkorn, Kloetzel, de Sá and Lima2014). By contrast, the parasite does not affect the transcription or expression of proteins of the constitutive subunits of the proteasome. This provides evidence that this protozoan is able to modulate infection through post-transcriptional mechanisms that affect the translation of host proteins, as well as limiting their recognition by CD8+ cells, thus favouring parasite evasion from the immune response (Camargo et al., Reference Camargo, Faria, Kloss, Favali, Kuckelkorn, Kloetzel, de Sá and Lima2014; Tarleton, Reference Tarleton2015).

Thus, when the protozoan comes into contact with neurons and glial cells, it carries out the process of cell invasion by binding to receptor tyrosine kinase A and C (TrkA and TrkC), which are normally activated by neurotrophin (NT) nerve growth factor and NT-3, respectively, and whose activation is essential for maintenance of the nervous system (Chuenkova and Pereira, Reference Chuenkova and PereiraPerrin2009; Aridgides et al., Reference Aridgides, Salvador and PereiraPerrin2013). Recognition of these receptors occurs through parasite-derived neuropathic factor (PDNF), a GPI-linked neuraminidase/trans-sialidase (TS), whose interaction, as with other protein kinases (Akt), is independent of sialic acid (Chuenkova and Pereira, Reference Chuenkova and PereiraPerrin2011). Subsequently, these same receptors were identified in cardiomyocytes (Meloni et al., Reference Meloni, Caporali, Graiani, Lagrasta, Katare, Van Linthout, Spillmann, Campesi, Madeddu, Quaini and Emanueli2010; Aridgides et al., Reference Aridgides, Salvador and PereiraPerrin2013), where it was shown that these cells are also invaded via a mechanism that involves PDNF binding to TrkC receptor primarily, or TrkA receptor, which offers some protection against oxidative stress and parasite invasion (Aridgides et al., Reference Aridgides, Salvador and PereiraPerrin2013).

Parasite invasion induces multiple responses in the heart. Studies have shown that Cox2 mRNA and protein expression is induced in murine models during parasitic loading and myocarditis. This upregulation is also associated with the induction of cardiac dysfunction markers such as endothelin-1 (ET-1) and atrial natriuretic peptide (ANP), the first considered important in signalling pathways that allow Ca activation of ERK1/2 leading to cyclin-D1 expression and inflammation-related gene expression. In addition, Cox2 is induced via the Ca/calcineurin/NFAT pathway allowing overproduction of eicosanoids (thromboxane A2 (TXA2) and prostaglandins E2 and F2a) (Petkova et al., Reference Petkova, Tanowitz, Magazine, Factor, Chan, Pestell, Bouzahzah, Douglas, Shtutin, Morris, Tsang, Weiss, Christ, Wittner and Huang2000; Salomone et al., Reference Salomone, Caeiro, Madoery, Amuchastegui, Omelinauk, Juri and Kaski2001; Hassan et al., Reference Hassan, Mukherjee, Nagajyothi, Weiss and Petkova2006; Corral et al., Reference Corral, Guerrero, Cuervo, Girones and Fresno2013), and TXA2 is induced exacerbating cardiomyocyte apoptosis, facilitating cytokine biosynthesis by monocytes, and leading to endothelial and platelet activation, aggregation and degranulation (Corral et al., Reference Corral, Guerrero, Cuervo, Girones and Fresno2013).

Virulence factors of T. cruzi

T. cruzi possesses a wide variety of virulence factors that allow it to reproduce and effectively evade host defences and generate damage or morbidity in the affected individual (Brown et al., Reference Brown, Cornforth and Mideo2012), these virulence factors are described below.

Many studies have focused on finding and characterizing the molecules involved in the process of infection by T. cruzi, but little is known about the molecules involved in cellular invasion by extracellular amastigotes. Recently, a group of amastine proteins was reported, for which there are four subfamilies (alpha, beta, delta and gamma), of which the beta and delta proteins are found in T. cruzi (Jackson, Reference Jackson2010). These proteins are located on the amastigote cell surface and in strain G were demonstrated to increase the rate of differentiation towards metacyclic trypomastigotes. Studies with recombinant proteins both in vitro and in vivo demonstrated that these proteins also participated in the adhesion to and invasion of host cells by the amastigotes, and intracellular survival of the parasites by maintaining the pH inside parasitic vacuoles potentially via a mechanism that involves proton and ion trafficking through the membrane (Cruz et al., Reference Cruz, Souza-Melo, da Silva, DaRocha, Bahia, Araújo, Teixeira and Mortara2012).

One of the most important virulence factors in T. cruzi is the TS, which allow for sialylation of T. cruzi glycoconjugate membranes at terminal positions. There are two isoforms of TS identified in T. cruzi, one of which is enzymatically active (aTS) and the other is inactive (iTS), but still retains the ability to bind to sugar substrates. These proteins are secreted into the bloodstream, where they are systemically distributed and induce modifications in host cells by sialylation (Freire-de-Lima et al., Reference Freire-de-Lima, Fonseca, Oeltmann, Mendonça-Previato and Previato2015). After intracellular replication of the parasites and transformation into trypomastigotes, a large amount of TS is released into the cytoplasm of the parasitized cell, which is then lysed leading to an increase in the concentration of systemic TS that cannot be efficiently combated by antibodies. The amount of enzyme secreted correlates with the virulence of the strain and high levels of TS are associated with a wide variety of abnormalities during the early stages of infection, including thymocyte depletion (thymocyte apoptosis), the absence of germinal centres in secondary organs, thrombocytopenia and erythropenia (Freire-de-Lima et al., Reference Freire-de-Lima, Fonseca, Oeltmann, Mendonça-Previato and Previato2015; San Francisco et al., Reference San Francisco, Gutierrez, Neira, Munoz, Sagua, Araya, Andrade, Zailberger, Catalán, Remonsellez, Vega and González2017) (Fig. 3).

Fig. 3. Virulence factors and immunological response elicited by T. cruzi infection.

TS also inhibit lymphocyte proliferation via IL-2 signalling, and are associated with the ability of the parasite to reduce IL-2Ra expression and IL-2 production by T cells. In addition, they can manipulate the maturation of CD4+ cells inducing non-protective Th2 genotypes in naive T cells while downregulating Th1 cells by IL-10 induction/expression during antigenic presentation by host cell antigens (ACP) to favour the survival and infection of the parasite. Additionally, CD8+ T cells from infected animals have been found to be highly sialylated, which reduces their ability to infiltrate tissues. This provides a dual benefit to the parasite, since it favours its intracellular replication while preventing extensive tissue damage of the infected tissue (Tarleton, Reference Tarleton2015; San Francisco et al., Reference San Francisco, Gutierrez, Neira, Munoz, Sagua, Araya, Andrade, Zailberger, Catalán, Remonsellez, Vega and González2017).

Similarly, the parasite is covered by a group of mucin glycoconjugates, which are classified depending on whether they are present in the vector (TcMUC) or in the host (TcSMUG). These mucin glycoconjugates offer protection against the vector or host defences, thereby assuring cellular invasion (Osorio et al., Reference Osorio, Rios, Gutierrez and Gonzalez2012). Surface antigens also exist that are related to these glycoconjugates, such as the mucin-like gp35/50 surface antigen that compromises cellular invasion and the trypomastigote small surface antigen (TSSA) of the trypomastigote. Genetic analysis of the TSSA revealed a strong relationship with TcMUC and variations among the sequences of the six DTUs of T. cruzi (TcI–TcVI). However, unlike TcMUC, TSSA is a hypoglycosylated molecule that participates in the infectivity of the trypomastigote by acting as an adhesion molecule between the host and parasite and playing a role in amastigogenesis. There are two variants of TSSA, one of which shows adhesive properties (TSSA-CL), and the other is non-adhesive (TSSA-Sy). TSSA is a potential target as a diagnostic and therapeutic effectiveness agent (Yoshida, Reference Yoshida2006; Balouz et al., Reference Balouz, Melli, Volcovich, Moscatelli, Moroni, González, Ballering, Bisio, Ciocchini and Buscaglia2017; Cámara et al., Reference Cámara, Cánepa, Lantos, Balouz, Yu, Chen, Campetella, Mucci and Buscaglia2017).

The T. cruzi parasite has mechanisms for the evasion of reactive species of oxygen and hydrogen (e.g. O₂⁻ and H₂O₂), produced mainly by macrophages, that generate direct oxidative damage. These defence mechanisms are mainly coordinated by five different types of peroxidase (TcGPXI, TcGPXII, TcCPX, TcMPX and TcAPX), the first two conferring protection against exogenous hydroperoxides, and TcCPX, TcMPX, and small chains of organic hydroperoxides and TcAPX conferring protection against H₂O₂. T. cruzi also possesses an iron superoxide dismutase (Fe-SOD) that prevents damage by O₂⁻ from mitochondrial, cytosolic or glycosomal sources (Osorio et al., Reference Osorio, Rios, Gutierrez and Gonzalez2012; Malvezi et al., Reference Malvezi, Da Silva, De Freitas, Lovo-Martins, Tatakihara, Zanluqui, Neto, Goldenberg, Bordignon, Yamada-Ogatta, Martin-Pinge, Cecchini and Pinge-Filho2014).

T. cruzi possesses other molecules that contribute to the evasion of host immune responses, such as those that inhibit the coupling of complement molecules. Complement regulatory proteins (CRPs) expressed only in trypomastigotes, inhibit the classical and alternative pathways by binding to C3b and C4b. For example, trypomastigote decay accelerating factor (T-DAF) interferes with the coupling of C3 in both pathways, and its metacyclic trypomastigote form is able to induce blood cells to produce microvesicles that bind to the surface of the parasite and stabilize it to C3. Similarly, the complement C2 receptor inhibitor trispanning protein (CRIT) inhibits complement activation via lectin, while calreticulin (TcCRT) captures C1 molecules and uses them as host cell recognition molecules to the benefit of the parasite (Osorio et al., Reference Osorio, Rios, Gutierrez and Gonzalez2012; Henrique et al., Reference Henrique, Marques, da Silva, de Oliveira, Rodrigues, Gomez-Hernandez, Ramirez, Ferreira, Nogueira and Norris2016).

Other molecules such as calpain, a calcium-dependent lysosomal peptide cysteine, present in the endosome–lysosome system of the epimastigotes, on the epimastigote cell surface and in the amastigote–trypomastigote transitional forms, is secreted from the flagellar pocket and intervenes in the processes of cytoskeleton remodelling, proliferation, differentiation and regulation of cellular calcium. Calpain operates through the cleavage of the high molecular weight kininogen protein, which stimulates the release of calcium through inositol triphosphate, among other molecules (de Souza et al., Reference de Souza, de Carvalho and Barrias2010; Branquinha et al., Reference Branquinha, Marinho, Sangenito, Oliveira, Goncalves, Ennes-Vidal, d'Avila-Levy and Santos2013). Oligopeptidase B, a serine endopeptidase cytosolic, secreted by trypomastigotes, has also been implicated in the induction of calcium release during invasion of the parasite (Burleigh et al., Reference Burleigh, Caler, Webster and Andrews1997; Burleigh and Woolsey, Reference Burleigh and Woolsey2002; de Souza et al., Reference de Souza, de Carvalho and Barrias2010).

Finally, alteration in the metabolism of phospholipids has recently been reported as another possible mechanism by which parasites interact with the host. For example, molecules such as phospholipase A1 (PLA1) have been shown to participate in parasite–cell interactions prior to cell invasion through the generation of lipids that act as secondary messengers and co-activate kinase C (Belaunzaran et al., Reference Belaunzaran, Lammel, Gimenez, Bott, Durante de Isola, Wilkowsky and Barbieri2013). The expression of phosphatidylinositol phospholipase C (TcPI-PLC) correlates with the decrease in phosphatidylinositol-4,5-bisphosphate (PIP2) and the increase in its product inositol-1,4,5-triphosphate in the cell host demonstrating that overexpression of TcPI-PLC can inhibit the progression of trypomastigotes to amastigotes (Okura et al., Reference Okura, Fang, Salto, Singer, Docampo and Moreno2005; Osorio et al., Reference Osorio, Rios, Gutierrez and Gonzalez2012).

The main limitations in the identification of these virulence factors is that there is no clue if these are DTU-dependent. Most of the studies include merely the Y strain (typed as TcII). Therefore, future studies should establish if these virulence factors are deviated by the massive genetic diversity among T. cruzi taxon.

Host factors

Polymorphisms among certain host factors, such as those that affect the expression of cytokines involved in the immune response, can play an important role in the course of a disease by attenuating the defensive capacity of the host against an invading pathogen. The relationship between CCC and cytokine expression (e.g. for IL-1, IL-10, IL-12, IL-17 and IL-18) has been studied, along with preliminary studies to analyse the relationship between this disease and other polymorphisms among host immune factors (Leon et al., Reference Leon Rodrigez, Carmona, Echeverría, González and Martin2016).

For IL-1, a proinflammatory cytokine that has been implicated in the mediation of both acute and chronic manifestations of CCC disease, has been reported to offer protection in terms of the haplotypes IL-1A, IL-1B and IL-1RN (Flórez et al., Reference Flórez, Zafra, Morillo, Martín and González2006). For IL-10, polymorphisms that lead to reduced expression, such as IL-1082G/A, are associated with the development of CCC. For IL-17, a proinflammatory cytokine produced by CD4+ T cells, some studies have shown that lower expression of this cytokine correlates with cardiac manifestations, whereas associations between certain polymorphisms (rs2275913, rs763780) have been related to the severity of left ventricular systolic function in patients with CCC (Costa et al., Reference Costa, Moreira, Menezes, Silva, Dutra, Rocha and Gollob2009; Magalhães et al., Reference Magalhães, Villani, Nunes, Gollob, Rocha and Dutra2013; Reis et al., Reference Reis, Sakita, de Moraes, Aquino, Macedo, Mazini, Sell, de Oliveira, Bulgarelli and Laguila2017). As for IL-18, reduced expression of this cytokine has been associated with the early stage response to CCC (rs360719*C) by permitting the activation of transcription factor OCT-1; however, certain polymorphisms (rs5744258, rs360722) showed no statistically significant relationship with susceptibility to disease, and other studies found that rs2043055 was associated with the modulation of Chagas disease severity (Esper et al., Reference Esper, Utsch, Brant, Teixeira, Vieira, Machado, Soriani, Arantes, Campos, Pinho, Souza and Tanowitz2014; Nogueira et al., Reference Nogueira, Frade, Ianni, Laugier, Pissetti, Cabantous, Baron, Peixoto-Gde, Borges-Ade, Donadi, Marin-Neto, Schmidt, Dias, Saba, Wang, Fragata, Sampaio, Hirata, Buck, Mady, Martinelli, Lensi, Siqueira, Pereira, Rodrigues, Kalil, Chevillard and Cunha-Neto2015). Further studies are therefore required to investigate the relationships between host polymorphisms and CCC disease.

Similarly, polymorphisms among chemokines have been implicated in the pathology of CCC disease. Chemokines play an important role in the recruitment of inflammatory cells to heart tissue in studies of mice with both acute and chronic myocarditis. Variations in CCL2, a cytokine that participates in the elimination of parasites through NO-dependent pathways, have been associated with the development of CCC. Similarly, polymorphisms in CCR5, which is essential in the control of acute infections and in the maintenance of inflammation that causes tissue damage, have been related to CCC disease severity (CCR5 rs1799988) and other polymorphisms are more frequently associated with asymptomatic patients (CCR5 59029A/G). Furthermore, the pleomorphism rs10336 in CXCL9 [105–110] correlated with the diminution of the intensity of myocarditis. By contrast, variations in CCL4, CCL5, CCL17, CCL19 and CXCR3 showed no association with CCC (Fernández-Mestre et al., Reference Fernández-Mestre, Montagnani and Layrisse2004; Nogueira et al., Reference Nogueira, Santos, Ianni, Fiorelli, Mairena, Benvenuti, Frade, Donadi, Dias, Saba, Wang, Fragata, Sampaio, Hirata, Buck, Mady, Bocchi, Stolf, Kalil and Cunha-Neto2012).

Variations in HLA supertypes (such as HLA-A1, A2 and A24, which are present in CD8+ T cells where most of their epitopes are restricted to the HLA-A allele) have also been found to relate to disease pathology (Lasso et al., Reference Lasso, Beltrán, Guzmán, Rosas, Thomas, López, González, Cuéllar and Puerta2016), with polymorphisms in HLA I and HLA II being reported to be associated with CCC (Deghaide et al., Reference Deghaide, Dantas and Donadi1998). Of the HLA I genes, it was shown that variations in HLAC*03 have a strong association with CCC compared with asymptomatic patients. Similarly, DPB1*0401 presents a higher frequency in affected patients compared with asymptomatic patients, with asymptomatic patients showing a higher frequency of DPB1*0101 (Colorado et al., Reference Colorado, Acquatella, Catalioti, Fernandez and Layrisse2000; Layrisse et al., Reference Layrisse, Fernandez, Montagnani, Matos, Balbas, Herrera, Colorado, Catalioti and Acquatella2000). As for the HLA II genes, a comparison was made between alleles DRB1 and DQB1 and polymorphisms DRB1*14 and DQB1*0303 were found to be protective against the chronic phase of the disease. In addition, the DRB1*01, DRB1*08 and DQB1*0501 polymorphisms showed a higher frequency in patients with arrhythmia and congestive heart failure, whereas the DRB1*1501 polymorphism presented a lower frequency in such patients. These findings suggested that the HLA II genes may be associated with the development of chronic infection and cardiac tissue damage. Similarly, polymorphisms in HLA-DQ1 are considered to confer susceptibility, whereas HLA-DQ7 confers protection against the development of CCC disease (Deghaide et al., Reference Deghaide, Dantas and Donadi1998; Cunha-Neto and Chevillard, Reference Cunha-Neto and Chevillard2014).

Finally, for TNF, the TNFA-308 polymorphism and variations in the microsatellites of TNF-α from patients with end-stage CCC correlated with significantly shorter survival times compared with individuals who did not possess these variations (Drigo et al., Reference Drigo, Cunha-Neto, Ianni, Faé, Nunes, Buck, Mady, Kalil, Goldberg, Cardoso and Braga2006). Furthermore, variants in the promoter region of the IKBL gene (IKBL-62A/T, IKBL-262A/G) were associated with susceptibility to CCC (Ramasawmy et al., Reference Ramasawmy, Faé, Cunha-Neto, Borba, Ianni, Mady, Goldberg and Kalil2008). It has been recently suggested that the parasympathetic nervous system and the endocrine system may play important roles in the clinical manifestations associated with CCC (Roggero et al., Reference Roggero, Rosa Perez, Pollachini, Raquel Villar, Wildmann, Besedovsky and del Rey2016; Savino, Reference Savino2017). Further investigations are now needed on large study populations to determine the frequency of certain polymorphisms among individuals and the roles of host factor gene variations on the susceptibility and severity of CCC disease pathology (Prata, Reference Prata2001).

Cytokine profile during T. cruzi infection

The expression of cytokines and their kinetics are key factors in the development and progression of disease. At present, the pathophysiological mechanisms of Chagas’ disease have not been completely elucidated; however, two hypotheses have been established. The first relates to the early stages of infection where an appropriate inflammatory response is beneficial but a dysregulated response allows for tissue damage. In the acute phase of infection, it has been proposed that the immune response of the host may be associated with aggressive autoimmunity. The second hypothesis is based on an autoreactive process in response to the lack of adequate immune modulation between excessive proinflammatory cytokines and the loss of anti-inflammatory cytokines that plays an important role in the progression of human Chagas disease from asymptomatic to severe forms (cardiomyopathy and megavisceral syndromes) and is associated with molecular mimicry (Nagib et al., Reference Nagib, Dutra, Chiari, Conceição and Machado2007; Pissetti et al., Reference Pissetti, Correia, de Oliveira, Llaguno, Balarin, Silva-Grecco and Rodrigues2011; Longhi et al., Reference Longhi, Tasso, Gomez, Atienza, Bonato, Chiale, Perez, Buying, Pinilla, Judkowski, Balouz, Buscaglia and Santos2014).

During the acute phase of the disease, it has been suggested that the immune response employs three mechanisms to counteract the infection. The first is the detection and destruction of the parasite via macrophages and dendritic cells. The second involves the activation of dendritic cells and macrophages that initiate the presentation and activation of specific antigens to elicit immune responses. The third mechanism is the detection of infection by non-hematopoietic cells, which play a major role in protecting against the invasion of T. cruzi (Poveda et al., Reference Poveda, Fresno, Gironès, Martins-Filho, Ramírez, Santi-Rocca, Marín-Neto, Morillo, Rosas and Guhl2014).

In the chronic phase, antigens that are produced by the autoimmune response are mediated by T cells, as an important step in the progression of the disease. Several studies in children have found T cells to be predominant proinflammatory and cellular monocyte regulators. CD8+ T cells play an important role in combatting intracellular pathogens. This is due to the fact that this cell type is able to recognize infected cells, when such a function is eliminated or inhibited, there is no parasitic control in the early phase of the disease leading to the exacerbation of infection and the onset of chronic disease (Sanmarco et al., Reference Sanmarco, Visconti, Eberhardt, Ramello, Ponce, Spitale, Vozza, Bernardi, Gea, Minguez and Aoki2016).

Immune mechanisms that operate to control the parasite prior to intracellular infection are controlled by the CD4+ and CD8 cellular responses, which inhibit the replication of T. cruzi as demonstrated in vitro. These cells and the protective functions of IL-2, IFN-γ and TNF-α produced by Th1 cells, have been shown to be associated with heart disease. IL-5, IL-10 and IL-13 regulate the inflammatory humoral response and the stimulation of IgE, eosinophils and mastocytes (Teixeira et al., Reference Teixeira, Hecht, Guimaro, Sousa and Nitz2011).

A previous study detected a switch between the anti-inflammatory cytokines IL-13, IL-5 and IL-10 and proinflammatory cytokines IL-2, IL-6, IL-9 and IL-12 in 109 seropositive patients and 21 seronegative controls, who were classified into two groups, CARD (heart disease) and NON-CARD (no heart disease), and a clear lack of immune modulation was reported. By measuring the mean fluorescence intensity of IL-12, IFN-γ, IL-1, IL-6 and IL-9 for discriminant analysis of principal components (DAPC), a cluster was observed for the CARD patients that was not present with the NON-CARD patients, potentially indicating that some NON-CARD patients may be predisposed to developing cardiomyopathy or mega-viscera syndrome explaining why the cytokine levels are not homogeneous for this group and suggesting that these cytokines could be used in combination as progression markers (Poveda et al., Reference Poveda, Fresno, Gironès, Martins-Filho, Ramírez, Santi-Rocca, Marín-Neto, Morillo, Rosas and Guhl2014).

Role of cytokines in the genetic variability of T. cruzi DTUs

Previous studies to analyse the association between T. cruzi genetic variability and the different clinical manifestations of Chagas disease found an association between TcI and cardiomyopathy, and between TcII, TcV and TcVI and mega-visceral syndrome. We used sera from CCC patients infected with different DTUs (20 TcI, 20 TcII and 15 mixed TcI + TcII) to observe whether the genetic variability of the parasite was associated with the pathogenesis of Chagas disease. A proinflammatory profile was observed for all groups as expected, and this profile was similar to that identified for the cardiac group. However, higher levels of cytokines were found in the TcII and mixed groups compared with the TcI group, where the levels did not exceed 50%. By contrast, we found higher levels of the cytokines IL-6 for TcI, IL-1 for TcII and IL-22 for the mixed TcI/TcII group (Rassi et al., Reference Rassi, Rassi and Marin-Neto2009). This is interesting since it is known that IL-6 is secreted by T cells and macrophages, IL-1 by macrophages and lymphocytes, and IL-22 by dendritic cells and T cells (Rojas. et al., Reference Rojas, Anaya, Gómez, Aristizabal, Cano and Lopera2017) (Table 1).

Table 1. Association between the T. cruzi DTUs and the related clinical form and immune response

A recent study showed that the immune profile relating to the cardiac inflammatory response involved the expression of TNF, IL-2, IL-10 and IFN-γ in human cardiac cell infiltrate samples, which may suggest that these factors play an important role in the variable susceptibility to the chronic phase of the disease. Other studies demonstrated the presence of IL-2, IL-4 and IL-6 in infected cardiac tissue. Even in patients with ventricular dysfunction, IL-10, IFN-γ, IL-6, TNF and IL-1 have been reported to increase plasma levels. This suggests that there is a relationship between the secretory response of T cells in the immunological profile of infected heart tissue (Vicco et al., Reference Vicco, Ferini, Rodeles, Cardona, Bontempi, Lioi, Beloscar, Nara, Marcipar and Bottasso2013; Rodríguez et al., Reference Rodríguez, Marigorta and Navarro2014). However, a study by Vicco and colleagues in Wistar mice showed that the administration of diluted phosphorus allowed for some modulation of inflammation in the cardiac tissue via IFN-γ and TNF-α (Ferreira et al., Reference Ferreira, Portocarrero, Brustolin, Ciupa, Massini, Aleixo and de Araújo2017).

In contrast to previous studies that showed that patients with less aggressive forms of cardiomyopathy produced higher levels of IL-17 (Guedes et al., Reference Guedes, Gutierrez, Silva, Dellalibera-Joviliano, Rodrigues, Bendhack, Rassi, Schmidt, Maciel, Marin- Neto and Silva2012), we hypothesize that DTU-specific recognition by the immune system (antibodies, B cells or T cells), could lead to the differential responses observed. Our results suggest that patients with more severe cardiomyopathy would be those with TcI, followed by those with a mixed infection, and finally those infected with TcII. This is in accordance with a descriptive analysis performed by our group where we detected that patients infected with the TcI DTU displayed more cardiac alterations than those infected with TcII (Ramírez et al., Reference Ramírez, Guhl, Rendón, Rosas, Marin-Neto and Morillo2010).

Finally using IFN-γ, IL-12, IL-22 and IL-10 for DAPC analysis, we found clusters for patients infected with TcI and mixed TcI + TcII, suggesting that there is a likely association between the genetic variability of T. cruzi (TcI, TcII and mixed TcI/TcII) and the levels of some cytokines. The sympathetic nervous system is thought to play an important role in the survival of T. cruzi infection based on studies in mice (C57B1/6) given the possible participation of mechanisms that upregulate the production of proinflammatory cytokines, as well as cellular immune responses and the restriction of parasitic proliferation (Roggero et al., Reference Roggero, Rosa Perez, Pollachini, Raquel Villar, Wildmann, Besedovsky and del Rey2016).

T. cruzi heterogeneity and CCC

The classification of T. cruzi parasites is important in defining the biological, clinical and pathological characteristics associated with specific populations of T. cruzi (Mantilla et al., Reference Mantilla, Zafra, Macedo and González2010; Dias et al., Reference Dias, Ramos, Gontijo, Luquetti, Shikanai-Yasuda, Coura, Torres, Melo, Almeida, Oliveira, Silveira, Rezende, Pinto, Ferreira, Rassi, Fragata, Sousa, Correia, Jansen, Andrade, Britto, Pinto, Rassi, Campos, Abad-Franch, Santos, Chiari, Hasslocher-Moreno, Moreira, Marques, Silva, Marin-Neto, Galvão, Xavier, Valente, Carvalho, Cardoso, Silva, Costa, Vivaldini, Oliveira, Valente, Lima and Alves2016). Despite intensive research into the molecular epidemiology of T. cruzi, few studies have investigated the association between the genetic heterogeneity of T. cruzi and the clinical outcomes of Chagas disease. The DTUs of T. cruzi involved in an infection can alter the humoral response affecting the pathogenesis of the disease (Santi-Rocca et al., Reference Santi-Rocca, Fernandez-Cortes, Chillon-Marinas, Gonzalez-Rubio, Girones, Fresno and Martin2017). For this reason, different authors have proposed the use of lineage-specific serology markers to detect the serological profile of an infection according to each DTU. These assays have clearly demonstrated that the immune response elicited by each DTU may be different and highlights the need for further research in this field (Zingales et al., Reference Zingales, Andrade, Briones, Campbell, Chiari, Fernandes, Guhl, Lages-Silva, Macedo, Machado, Miles, Romanha, Sturm, Tibayrenc and Schijman2009). TSSA antigen and B-cell epitopes have been investigated for this purpose of developing specific serology markers but experiments were hindered by the lack of specificity of the in vitro assays (Bhattacharyya et al., Reference Bhattacharyya, Brooks, Yeo, Lewis, Llewellyn, Miles and Carrasco2010, Reference Bhattacharyya, Falconar, Luquetti, Costales, Grijalva, Lewis, Messenger, Tran, Ramirez, Guhl, Carrasco, Diosque, Garcia, Litvinov and Miles2014) (Fig. 4).

Fig. 4. Overview of the geographical distribution of T. cruzi DTUs associated with CCC across the American continent.

Several studies have provided evidence of the existence of a relationship between T. cruzi DTUs and clinical manifestations. TcI has been reported to be the most abundant DTU across the American continent, being detected among a wide range of reservoirs and triatomines, in which its presence is derived predominantly from sylvatic rather than domestic transmission cycles. The infection of humans by this DTU is concentrated in the north of South America reaching the centre of America, and is mainly associated with chagasic cardiomyopathy, predominantly in the domestic cycle (Zingales et al., Reference Zingales, Andrade, Briones, Campbell, Chiari, Fernandes, Guhl, Lages-Silva, Macedo, Machado, Miles, Romanha, Sturm, Tibayrenc and Schijman2009; Leiby et al., Reference Leiby, Nguyen, Proctor, Townsend and Stramer2017). It has been reported that this particular DTU plays an important role in severe forms of chagasic heart disease (Guhl, Reference Guhl2013; Guhl and Ramírez, Reference Guhl and Ramírez2013). In a study carried out in Argentina of 239 patients with a diagnosis of severe myocarditis, TcI was detected in 4.2% of the patients from either blood samples, biopsies or the organs from those who underwent a transplant (Burgos et al., Reference Burgos, Diez, Vigliano, Bisio, Risso, Duffy, Cura, Brusses, Favaloro, Leguizamon, Lucero, Laguens, Levin, Favaloro and Schijman2010; Guhl, Reference Guhl2013). In Colombia, similar studies showed that TcI caused more cardiac alterations that TcII across a large cohort of chronic symptomatic patients (Ramírez et al., Reference Ramírez, Guhl, Rendón, Rosas, Marin-Neto and Morillo2010).

Studies on TcII have shown that this DTU of T. cruzi prevails in the central and southern regions of America, and is related to transmission of the domestic cycle. TcII has been associated with the clinical manifestations of moderate chagasic cardiomyopathy, concomitant with mega syndromes such as megacolon and megaesophagus, also associated with blood donors (Zingales et al., Reference Zingales, Andrade, Briones, Campbell, Chiari, Fernandes, Guhl, Lages-Silva, Macedo, Machado, Miles, Romanha, Sturm, Tibayrenc and Schijman2009; Guhl and Ramírez, Reference Guhl and Ramírez2013; Leiby et al., Reference Leiby, Nguyen, Proctor, Townsend and Stramer2017). In addition, Bisio and coworkers confirmed that TcII was present in two patients with cardiac manifestations. However, these estimations are not absolute and patients across the continent with CCC can be infected by TcII (Burgos et al., Reference Burgos, Bisio, Duffy, Levin, Schijman, Altcheh, Burgos Freilij, Valadares, Freitas, Macedo, Seidenstein, Macchi and Piccinali2007).

In endemic areas of the Amazon region of Brazil, and in eastern Colombia and Venezuela, TcIV is considered to be the predominant DTU responsible for most of the acute diseases caused by this DTU (Carrasco et al., Reference Carrasco, Segovia, Llewellyn, Morocoima, Urdaneta-Morales, Martínez, Garcia, Rodríguez, Espinosa, de Noya, Díaz-Bello, Herrera, Fitzpatrick, Yeo, Miles and Feliciangeli2012; Guhl and Ramírez, Reference Guhl and Ramírez2013; Segovia et al., Reference Segovia, Carrasco, Martínez, Messenger, Nessi, Londoño, Espinosa, Martínez, Mijares, Bonfante-Cabarcas, Lewis, de Noya, Miles and Llewellyn2013; Monteiro et al., Reference Monteiro, Peretolchina, Lazoski, Harris, Dotson, Abad-Franch, Tamayo, Pennington, Monroy, Cordon-Rosales, Salazar-Schettino, Gómez-Palacio, Grijalva, Beard and Marcet2013a; Reference Monteiro, Margioto-Teston, Gruendling, dos Reis, Gomes, de Araújo, Bahia, Magalhães, de Oliveira-Guerra, Silveira, Toledo and Vale-Barbosa2013b; Margioto Teston et al., Reference Margioto Teston, de Abreu, Abegg, Gomes and de Ornelas Toledo2017). TcIV has also been strongly incriminated with lethal cases relating to oral transmission in Colombia and Brazil (Monteiro et al., Reference Monteiro, Peretolchina, Lazoski, Harris, Dotson, Abad-Franch, Tamayo, Pennington, Monroy, Cordon-Rosales, Salazar-Schettino, Gómez-Palacio, Grijalva, Beard and Marcet2013a; Reference Monteiro, Margioto-Teston, Gruendling, dos Reis, Gomes, de Araújo, Bahia, Magalhães, de Oliveira-Guerra, Silveira, Toledo and Vale-Barbosa2013b; Ramírez et al., Reference Ramírez, Montilla, Cucunubá, Floréz, Zambrano and Guhl2013; Dario et al., Reference Dario, Rodrigues, Barros, Xavier, Roque, Jansen and D'Andrea2016; Hernández et al., Reference Hernández, Cucunubá, Flórez, Olivera, Valencia, Zambrano, León and Ramírez2016). As for TcV and TcVI, comparative genetic studies proposed that these DTUs are TcII and TcIII hybrids that correlate with chagasic cardiomyopathy and megavisceral syndrome in the southern cone of the American continent (Zingales et al., Reference Zingales, Andrade, Briones, Campbell, Chiari, Fernandes, Guhl, Lages-Silva, Macedo, Machado, Miles, Romanha, Sturm, Tibayrenc and Schijman2009; Guhl and Ramírez, Reference Guhl and Ramírez2013). In a study from Argentina, TcV or TcII/V/VI were found to be the most prevalent DTUs among 226 of 239 patients studied (89.9%). Whereas, TcV was present in 90.9% of the TcII/V/VI group samples studied that correlated with Chagas’ moderate chronic heart disease in the study by Burgos and colleagues (Burgos et al., Reference Burgos, Diez, Vigliano, Bisio, Risso, Duffy, Cura, Brusses, Favaloro, Leguizamon, Lucero, Laguens, Levin, Favaloro and Schijman2010).

Another study revealed that in the southern region of Latin America (in particular Argentina), blood samples and cultures isolated from patients with CCC were predominately of the TcII, TcV and TcVI T. cruzi DTUs. To date, in this geographic location, TcI has not been considered the dominant DTU in heart disease (Cura et al., Reference Cura, Bisio, Duffy, Schijman, Lucero, Formichelli, Brusés, Merino, Oshiro, Sosa-Estani, Burgos, Lejona, Anchart, Hernández, Severini, Velazquez, Lattes, Altcheh, Freilij, Diez, Nagel, Vigliano, Favaloro and Favaloro2012). However, in terms of clinical associations, descriptive molecular epidemiology studies linked TcI with severe forms of myocarditis in cardiac samples from CCC patients in Argentina and no specific clinical manifestations related to T. cruzi DTUs in Bolivian CCC patients showing the pleomorphism of T. cruzi (Moncayo and Yanine, Reference Moncayo and Yanine2006; Ramírez et al., Reference Ramírez, Guhl, Umezawa, Morillo, Rosas, Marin-Neto and Restrepo2009; Burgos et al., Reference Burgos, Diez, Vigliano, Bisio, Risso, Duffy, Cura, Brusses, Favaloro, Leguizamon, Lucero, Laguens, Levin, Favaloro and Schijman2010; del Puerto et al., Reference del Puerto, Nishizawa, Kikuchi, Iihoshi, Roca, Avilas, Gianella, Lora, Velarde, Gutierrez, Renjel, Miura, Higo, Komiya, Maemura and Hirayama2010). The direct detection of T. cruzi DTUs in the blood of CCC patients was established by amplification of the 24Sα rDNA divergent domain and the mitochondrial housekeeping genes (Mantilla et al., Reference Mantilla, Zafra, Macedo and González2010). In this study, molecular characterization of T. cruzi DTUs showed that most of the patients were infected with TcI and some were infected with TcII (9.9%). Recently, a new approach for T. cruzi DTU detection in CCC patients has been developed that showed that TcI was the predominant DTU and TcII was also detected, furthermore, the genetic characteristics of the TcII parasites found in Colombia were similar to those of the TcII parasites found in Bolivia and Chile (Mantilla et al., Reference Mantilla, Zafra, Macedo and González2010). Regarding the genetic variability of the parasite, prognosis markers based on mitochondrial genes are being developed, since specific mutations in these genes can trigger complications in the chronic phase of the disease in asymptomatic patients (Carranza et al., Reference Carranza, Valadares, D’Ávila, Baptista, Moreno, Galvão, Chiari, Sturm, Gontijo, Macedo and Zingales2009).

Despite the genetic variability, it is important to consider the presence of T. cruzi clones that have been detected in different tissues. Several studies have demonstrated specific histotropism of T. cruzi in mice showing differences in the pathological, immunological and clinical features that the parasite can elicit in the host (Carareto et al., Reference Carareto, Manoel-Caetano, Aparecida, Silva, Borim and Miyazaki2008; Ramírez et al., Reference Ramírez, Guhl, Rendón, Rosas, Marin-Neto and Morillo2010; Cruz et al., Reference Cruz, Santos-Miranda, Monti-Rocha, Machado, Sales, Campos and Roman-Campos2016; Leon et al., Reference Leon, Ramirez, Montilla, Vanegas, Castillo and Parra2017). Moreover, some authors have shown that the T. cruzi population in a patient's bloodstream may differ from the parasite population that causes tissue damage (Vago et al., Reference Vago, Andrade, Leite, d'Avila-Reis, Macedo, Adad, Tostes, Moreira, Filho and Pena2000; Macedo et al., Reference Macedo, Machado, Oliveira and Pena2004). Differences were found in T. cruzi populations in the bloodstreams of patients with CCC and of chagasic patients without cardiomyopathy (Venegas et al., Reference Venegas, Coñoepan, Pichuantes, Miranda, Apt, Arribada, Zulantay, Coronado, Rodriguez, Reyes, Solari and Sanchez2009). Microsatellite analyses have shown multiclonality in samples from the heart and bloodstream of infected patients demonstrating that specific populations of T. cruzi may determine the disease outcome (Burgos et al., Reference Burgos, Bisio, Duffy, Levin, Schijman, Altcheh, Burgos Freilij, Valadares, Freitas, Macedo, Seidenstein, Macchi and Piccinali2007; Valadares et al., Reference Valadares, Pimenta, de Freitas, Duffy, Bartholomeu, Oliveira, Chiari, Moreira, Filho, Schijman, Franco, Machado, Pena and Macedo2008). Finally, studies from clones obtained from hemoculture from one patient (chronic symptoms) showed divergent multilocus genotypes, reinforcing this hypothesis (Ramírez et al., Reference Ramírez, Guhl, Messenger, Lewis, Montilla, Cucunuba and Llewellyn2012, Reference Ramírez, Montilla, Cucunubá, Floréz, Zambrano and Guhl2013). One group proposed a model of clonal histiotropism to explain how a composite of clones may be related to disease outcomes (Macedo and Segatto, Reference Macedo and Segatto2010). Recent reports from Colombia corroborate this premise since they describe cardiac biopsies infected with sylvatic genotypes and TcI circulating in the bloodstream of patients (Ramírez et al., Reference Ramírez, Guhl, Rendón, Rosas, Marin-Neto and Morillo2010; Zafra et al., Reference Zafra, Mantilla, Jacome, Macedo and Gonzalez2011). The case of mixed infections appears to be the rule rather than the exception in chagasic patients. It has been proposed that such mixed infections play a major role in Chagas pathogenicity, with some paradox traits: in animal models, infection by a ‘slow’ genotype and by a ‘fast’ genotype is faster than the faster genotype alone. This has been recently corroborated by Llewellyn et al. in Reference Llewellyn, Messenger, Luquetti, Garcia, Torrico, Tavares, Cheaib, Derome, Delepine, Baulard, Deleuze, Sauer, Miles and Michael2015 where they explored T. cruzi infection multiclonality in the context of age, sex and clinical profile among a cohort of chronic patients, as well as paired congenital cases from Cochabamba, Bolivia and Goias, Brazil using amplicon deep sequencing technology (Llewellyn et al., Reference Llewellyn, Messenger, Luquetti, Garcia, Torrico, Tavares, Cheaib, Derome, Delepine, Baulard, Deleuze, Sauer, Miles and Michael2015). Their conclusions showed that no specific association was found between the number and diversity of parasite genotypes in each patient with their age, sex or disease status. Also, they were able to detect the transmission of multiple parasite genotypes between mother and foetus. This clearly opens the window about the need to explore in deep the case of mixed infections across a well-characterized cohort of chagasic patients and demonstrate the pivotal role of multiclonality in a plausible severity of disease outcome (Llewellyn et al., Reference Llewellyn, Messenger, Luquetti, Garcia, Torrico, Tavares, Cheaib, Derome, Delepine, Baulard, Deleuze, Sauer, Miles and Michael2015).

Molecular epidemiology studies of T. cruzi have attempted to establish the effects of different DTUs in the clinical progression of Chagas disease. Several studies have shown the effect of genetic variability on the host immune response (Ramírez et al., Reference Ramírez, Guhl, Umezawa, Morillo, Rosas, Marin-Neto and Restrepo2009). However, a study carried out in the Chilean population, in which the DTUs TcI, TcII, TcV and TcVI were present and TcV predominated, sought to establish a relationship between the parasitic burden, the DTUs and the clinical manifestations of the disease, but found no correlation between the DTUs and the cardiac manifestations of the disease (Apt et al., Reference Apt, Zulantay, Saavedra, Araya, Arriagada, Arribada, Solari, Ortiz and Rodríguez2015). It had previously been established that the cardiopathologies in southern cone countries were caused by TcII, TcV and TcVI, but it has recently been demonstrated that TcI can also play an important role specifically in the severe cardiopathologies related to Chagas disease. Studies of cardiac biopsies from Argentinean patients revealed that patients with severe myocarditis were infected with TcI, whereas those with moderate or absent myocarditis were infected with TcII, TcV or TcVI (Burgos et al., Reference Burgos, Diez, Vigliano, Bisio, Risso, Duffy, Cura, Brusses, Favaloro, Leguizamon, Lucero, Laguens, Levin, Favaloro and Schijman2010). Furthermore, among patients with CCC, the TcIDOM genotype was most commonly found in the bloodstream, whereas sylvatic-like TcI parasites were most commonly found in cardiac biopsies. These results were consistent with reports from patients in Colombia, where the least and most prevalent TcI genotypes in adult patients with CCC were sylvatic-like TcI parasites and TcIDOM, respectively (Ramírez et al., Reference Ramírez, Guhl, Rendón, Rosas, Marin-Neto and Morillo2010; Hernández et al., Reference Hernández, Cucunubá, Flórez, Olivera, Valencia, Zambrano, León and Ramírez2016). These results suggest potential histotropism by TcI genotypes and the epidemiological importance of this DTU in the southern American countries, where cardiopathologies were previously thought to be caused primarily by TcII, TcV and TcVI.

The main problem in establishing the real picture of T. cruzi heterogeneity in Chagas disease patients is the low parasitic load in the chronic phase of the disease. Furthermore, a temporal variation pattern has been detected whereby the T. cruzi population may change at 10-day intervals (Sánchez and Ramírez, Reference Sánchez and Ramírez2013). Therefore, it is imperative to improve the current methodologies for strain typing. Recently, with the rise of next generation sequencing technologies, researchers have been able to deploy multilocus sequence typing (MLST) schemes to infer the genetic divergence of this parasite (Lauthier et al., Reference Lauthier, Tomasini, Rumi, D'Amato, Ragone, Diosque, Barnabé, Tibayrenc, Yeo, Lewis, Llewellyn, Miles and Basombrío2012; Messenger et al., Reference Messenger, Llewellyn, Bhattacharyya, Franzén, Lewis, Ramírez, Carrasco, Andersson and Miles2012). These methodologies have been applied to specific clinical phenotypes such as oral Chagas disease, which is an eminent public-health problem in those areas were vectorial transmission has been interrupted. In Colombia, six outbreaks of oral Chagas disease have been reported and strains isolated from these outbreaks were analysed by MLST schemes. The results showed a predominance of TcI in the cases with a foreseen infection of TcIV, suggesting the unlikely relatedness of sylvatic strains with the oral outbreaks (Hernández et al., Reference Hernández, Cucunubá, Flórez, Olivera, Valencia, Zambrano, León and Ramírez2016).

T. cruzi I has shown a relevant genetic heterogeneity and some authors have subdivided it into at least two near-clades (domestic and sylvatic TcI) (Ramírez and Hernández, Reference Ramírez and Hernández2017). However, other studies showed that only domestic TcI is a robust genotype across the American continent which clearly reflects that TcI isolates are highly variable (Zumaya-Estrada et al., Reference Zumaya-Estrada, Messenger, Lopez-Ordonez, Lewis, Flores-Lopez, Martínez-Ibarra, Pennington, Cordon-Rosales, Carrasco, Segovia, Miles and Llewellyn2012). Recent genomic approaches by whole genome sequencing and multi-SNP typing will most probably uncover lesser genetic subdivisions within it. This intra-DTU diversity is remarkable in the light of plausible histiotropism where has been revealed that domestic TcI circulates in the bloodstream and sylvatic TcI shows tropism for heart tissue (Burgos et al., Reference Burgos, Diez, Vigliano, Bisio, Risso, Duffy, Cura, Brusses, Favaloro, Leguizamon, Lucero, Laguens, Levin, Favaloro and Schijman2010). Such micro-molecular epidemiology tends to become routine in bacteria and has started in parasites (Wong et al., Reference Wong, Baker, Pickard, Parkhill, Page, Feasey, Kingsley, Thomson, Keane, Weill, Edwards, Hawkey, Harris, Mather, Cain, Hadfield, Hart, Thieu, Klemm, Glinos, Breiman, Watson, Kariuki, Gordon, Heyderman, Okoro, Jacobs, Lunguya, Edmunds, Msefula, Chabalgoity, Kama, Jenkins, Dutta, Marks, Campos, Thompson, Obaro, MacLennan, Dolecek, Keddy, Smith, Parry, Karkey, Mulholland, Campbell, Dongol, Basnyat, Dufour, Bandaranayake, Naseri, Singh, Hatta, Newton, Onsare, Isaia, Dance, Davong, Thwaites, Wijedoru, Crump, De Pinna, Nair, Nilles, Thanh, Turner, Soeng, Valcanis, Powling, Dimovski, Hogg, Farrar, Holt and Dougan2015). Such studies show that upper evolutionary units such as T. cruzi I are to broad units of analysis for refined epidemiological studies. Future studies must consider intra-DTU genetic variation and plausible disease outcome.

The analysis of specific haplotypes incriminated the TcI sylvatic-like strains in the oral cases of Chagas disease highlighting the need to improve current epidemiological surveillance systems in endemic areas to detect the invasion of sylvatic-like genotypes in domestic cycles of transmission (Ramírez et al., Reference Ramírez, Montilla, Cucunubá, Floréz, Zambrano and Guhl2013). This type of analysis was also conducted in the largest urban oral outbreak of Chagas disease ever reported in Caracas, Venezuela, where the authors were able to track the source of infection attributing it to TcI sylvatic-like strains (Segovia et al., Reference Segovia, Carrasco, Martínez, Messenger, Nessi, Londoño, Espinosa, Martínez, Mijares, Bonfante-Cabarcas, Lewis, de Noya, Miles and Llewellyn2013). This scenario proposes the relevance of the typing schemes to track the phylodynamics of T. cruzi and also indicates that some TcI DTUs are likely more susceptible to causing oral infections. However, some authors have proposed that TcI and TcII express different glycoproteins that facilitate their survival in the gastric mucosa leading to infection via the oral route (Yoshida et al., Reference Yoshida, Tyler and Llewellyn2011; Sánchez and Ramírez, Reference Sánchez and Ramírez2013). This area therefore requires further study.

Studies that encompass molecular epidemiology and molecular biology techniques, such as PCR, allow us to establish the relationship between different DTUs of T. cruzi and the clinical manifestations of the disease (including Chagas’ cardiomyopathy) in various populations. The findings of such studies indicate that certain genotypes of this parasite exert damage at the cellular level and therefore are not the same as those detected in the blood that cause parasitaemia and infection during the acute phase of the disease. In addition, the various serotypes of trypanosomes that cause non-heart-related pathologies and those that cause Chagas cardiomyopathy may differ, meaning that organisms belonging to a specific genotype may predominate in particular clinical manifestations. This suggests that specific genotypes may determine the clinical course of the disease (Guhl and Ramírez, Reference Guhl and Ramírez2013).

One interesting and important study was conducted by Santi-Rocca et al. in 2017 using syngeneic mice infected acutely or chronically with six DTUs, 66 parameters were analysed, including parasite tropism, organ and immune responses (local and systemic) and clinical presentations of CCC. The authors of that study found that the parasite genetic background consistently impacts most of these parameters, but they remain highly variable impeding reliable one-dimensional association with phases, strains and damage, but the use of multi-dimensional statistics overcame this extreme intra-group variability and revealed some pathophysiological patterns that accurately allow defining (i) the infection phase, (ii) the infecting parasite strains and (iii) organ damage type and intensity (Santi-Rocca et al., Reference Santi-Rocca, Fernandez-Cortes, Chillon-Marinas, Gonzalez-Rubio, Girones, Fresno and Martin2017). This study was very important towards the understanding of the association between T. cruzi genetic diversity, host genetics and disease outcome. However, the results are not merely conclusive and research must focus on this topic.

However, the existence of studies showing the non-existence of a relationship between the clinical manifestations of Chagas disease and the genetic heterogeneity of the parasite must be considered. According to one study, no lineage was found to have a significant association with any clinical manifestation in particular, nor exclusively with patients who presented with the chronic phase of the disease (del Puerto et al., Reference del Puerto, Nishizawa, Kikuchi, Iihoshi, Roca, Avilas, Gianella, Lora, Velarde, Gutierrez, Renjel, Miura, Higo, Komiya, Maemura and Hirayama2010).