The Trypanosomatidae family is a group composed of flagellate protozoa that infect invertebrates, vertebrates, and plant hosts (Hoare, 1966; Vickerman, 1976). Although it is a well-studied group, new findings are continuously reported given that new parasite-host interactions and species/genotype diversity inside this family have been increasingly observed, especially after the advancement of molecular biology tools (d’Avila-Levy et al., 2016; Dario et al., 2017a; Maslov et al., 2019; Rangel et al., 2019; Rodrigues et al., 2019; Dario et al., 2021a).

The parasite species of this family are traditionally classified according to their developmental forms in their hosts during their life cycle (Hoare, 1966). Monoxenous trypanosomatids are those that evolved in only one host, namely, invertebrates. An example within this group is the parasite species from the genus Crithidia, which were first described to infect the Anopheles maculipennis mosquito (Léger, 1902) and are now recognized to infect bees, reduviids, and hoverflies (Langride and McGhee, 1967;  Lipa and Triggiani, 1988; Yurchenko et al., 2014; Ishemgulova et al., 2017). Monoxenic trypanosomatids were classically termed “inferior trypanosomatids”, demonstrating how much a reductionist and anthropocentric view can bypass important biological phenomena even from the sanitary point of view.

Trypanosomatid species that target an invertebrate, a vertebrate, or a plant host are named heteroxenous. The focus of this article is the Trypanosoma genus, which comprises species that infect vertebrates. However, the transmission cycle is only known for a minority of species, such as Trypanosoma brucei (sleeping sickness in humans), Trypanosoma evansi (surra disease in animals), and Trypanosoma cruzi (Chagas disease in humans) (Podlipaev, 2001). This finding reflects that the main studies are focused on human or animal parasites of economic interest, forgetting the profound interdependence between the health of all living beings and their environment.

Caatinga is a unique biome in Brazil that is found mainly in the northeast region and a small area in the southeast, namely, north of Minas Gerais state. The dominant climate is semiarid, and the term Caatinga means white forest in the indigenous language (Silva et al., 2004; Souza et al., 2015). Caatinga is the third most degraded biome in Brazil (Myers et al., 2000), and approximately 80% of the vegetation is completely modified due to extractivism and agriculture. Most of these areas are currently in the early or intermediate stages of ecological succession (Araújo Filho, 1996). As a result of these profound changes, the Caatinga presents large extensions where desertification is already noted (Souza et al., 2015). This biome, which represented an important endemic area for Chagas disease (Sousa et al., 2020) exhibits a diverse and unique fauna, comprising approximately 1307 animal species (Instituto Chico Mendes de Conservação da Biodiversidade, 2018). Caatinga harbors most of the triatomine species diversity in Brazil together with the Cerrado (Gurgel-Gonçalves et al., 2012), and T. cruzi infection is observed in mammals from five different orders in this biome: Artiodactyla, Carnivora, Cingulata, Didelphimorphia, and Rodentia (Jansen et al., 2018).

The order Didelphimorphia is one of the oldest and most important hosts of Trypanosoma species, including T. cruzi (Jansen et al., 2018; Rodrigues et al., 2019; Jansen et al., 2020), having as its main representative marsupials of the genus Didelphis (Jansen et al., 2015). The genus Didelphis presents a broad range distribution in nature due to its ability to adapt to different ecological niches, including anthropized environments (Olifiers et al., 2015). Therefore, it is considered a biomarker of degraded environments. These animals are nomadic, solitary, take refuge in holes and tree foliage, and are excellent climbers (Jansen et al., 2018). They are able to use all forest strata and are therefore in contact with the different transmission cycles of T. cruzi (Jansen et al., 2020).

The Rodentia order is the most diverse order within the small wild mammal taxa, and these animals can be found in different environments, ranging from tropical forests to deserts, plateaus to plains, and wild to urban environments. In nature, they are able to circulate through different forest strata and are found in terrestrial, arboreal, and semiaquatic areas (Wilson and Reeder, 2005). In relation to T. cruzi transmission, some studies have reported T. cruzi infection in rodents in Latin America (Herrera et al., 2005; Yeo et al., 2005; Herrera et al., 2007b; Vaz et al., 2007; Roque et al., 2008; Ramsey et al., 2012; Charles et al., 2013; Orozco et al., 2014; Herrera et al., 2015; Jansen et al., 2015; Morales et al., 2017). Several species of this order are highly susceptible to experimental infection by T. cruzi, demonstrating high and long-lasting parasitemia (Roque et al., 2005). However, it was suggested that this taxon plays a secondary role as a reservoir in the wild environment (Jansen et al., 2018) because the infection rate of wild rodents by T. cruzi is low, potentially due to the restricted use of the environment (restricted home range), which leads to less contact with other infected mammals and vectors (Jansen et al., 2015).

The role of dogs in the transmission cycle of T. cruzi can vary according to each region. In Brazil, dogs are highly exposed to T. cruzi infection based on high rates of positive serological tests. However, the detection of the parasite in blood smears or parasite isolation by hemoculture and xenodiagnosis is rare (Roque and Jansen, 2008; Noireau et al., 2009; Xavier et al., 2012, Jansen et al., 2018). A distinct scenario was described in Argentina, especially in the Gran Chaco region, where dogs are one of the main T. cruzi reservoirs, as demonstrated by high transmissibility competence (Gürtler et al., 2007; Gürtler and Cardinal, 2015). In Brazilian regions that present triatomines and small wild mammals with high T. cruzi infection rates, dogs are usually also exposed to infection as observed by serological positive tests but primarily without positive parasitological tests to date. Therefore, monitoring these animals through parasitological and serological surveys is an indication of the presence of a parasite transmission cycle occurring among wild free-ranging mammals in the area (Roque et al., 2008; Roque and Jansen, 2008; Pineda et al., 2011; Xavier et al., 2012).

Mixed infections by trypanosomatids, mainly in wild mammals, are poorly understood and rarely studied in animals. However, these infections can provide important epidemiological data because coinfections can modify the course of infection of a parasitosis. As the one-parasite-one-disease concept generally predominates, the focus has always been on T. cruzi or Leishmania sp., and the possible coinfecting trypanosomatids have been relegated to the background. Mixed infections between different trypanosomatid species and genotypes are very common in nature (Barbosa et al., 2017; Dario et al., 2017a; Jansen et al., 2018; Patiño et al., 2021). The determination of mixed infections is essential for the understanding of parasitological events, especially the resulting interactions between the parasite and its hosts. The aim of this study was to determine the trypanosomatid infection and richness in the synantropic Rattus rattus, Didelphis albiventris, and dogs in the T. cruzi transmission cycle in this semiarid Caatinga biome region.

In this study, three findings deserve attention: 1) the richness of Trypanosoma spp. found in D. albiventris but also in R. rattus in a rural area, therefore not representative of the original Caatinga landscape; 2) the first finding of C. mellificae infection in R. rattus; and 3) the richness of uncharacterized genotypes of T. cruzi. The aim of this study was to characterize trypanosomatid richness in D. albiventris and R. rattus and T. cruzi infection in dogs from a locality of the Caatinga biome that years ago was endemic for Chagas disease. Currently, no cases of children under 10 years of age have been reported (Lima et al., 2012 and unpublished results 2007), showing the results of the successful South cone initiative along with popular awareness of the disease and a certain improvement in residences.

We used the mini-exon target as a first attempt to characterize the isolates from D. albiventris and R. rattus. This technique showed that most isolates were infected by T. cruzi DTU TcI in a single infection except for one isolate that presented mixed infection with T. rangeli. Three isolates of R. rattus were demonstrated to be TcII by RFLP-PCR. The mini-exon target is a good molecular target to characterize T. cruzi DTU TcI. However, as we deal with sylvatic environment samples, which often present mixed infections, the diagnosis of the infection is unclear. Therefore, we tried to characterize the samples using SSU rDNA through Sanger sequencing, which also failed because the samples presented a double peak in the electropherogram, confirming the probably mixed infections. This is a limitation of the methodology given that it is only possible to identify them by combining Sanger sequencing with the molecular cloning methodology for the diagnosis of mixed infection, which makes the procedure considerably more laborious (Zepeda Mendoza et al., 2015; Huggins et al., 2019).

Hence, we performed NGS, which allowed us to detect 15 trypanosomatid species/genotypes infecting R. rattus, D. albiventris, and triatomines in single and mixed infections. NGS allowed the identification of a broad range of samples simultaneously, in addition to identifying the possible parasite species circulating in a given sample, even using culture isolate samples, which have gone through parasite species selection process, which could affect the results observed. Some studies have already used this methodology to identify trypanosomatids in samples from different mammal species, and it was possible to identify even more than three species in a single sample and even species of the Kinetoplastea class infecting mammals (Barbosa et al., 2017; Dario et al., 2017a; Dario et al., 2017b; Cooper et al., 2018; Huggins et al., 2019; Patiño et al., 2021). This is a promising methodology in parasitological studies that is able to unravel a world that we did not have access to before due to diagnostic limitations. This method will offer new knowledge, such as new species and genotypes and their phylogenetical relations. Here, we observed a trypanosomatid sequence that phylogenetically grouped together with sequences from the T. cruzi clade as well as a previously undescribed host-parasite interaction, namely, R. rattus infected by C. mellificae. In addition, according to Austen and Barbosa (2021), mixed infection detection in a host and the understanding of multihost parasite dynamics are priorities from now on, and detection of this type of infection is only possible through NGS.

Rattus rattus is a synanthropic species that frequents human living areas. They are considered to be of economic and sanitary importance, as they cause damage to food stocks and are associated with the transmission of several zoonoses that are very general in their diet, and they present a great climbing ability. Rodents are usually suggested to play a secondary role as reservoirs of T. cruzi in the sylvatic environment (Rademaker et al., 2009; Jansen et al., 2018). However, this notion was not observed in this Caatinga region. Animals from this order can be involved in the T. cruzi transmission cycle in different ways: a) rodents can share some of their microhabitats with triatomines of the genera Triatoma and Panstrongylus (Carcavallo et al., 1998); b) as they are one of the main predation targets, rodents are important in the study of the transmission cycle of T. cruzi; and c) some rodent species are used in the peridomestic environment and thus may act as links between the transmission cycles (Roque et al., 2008). This work demonstrates a richness of trypanosomatids that was completely unknown in R. rattus and raises the question - how are these parasites that are in such low parasitemias transmitted?

Didelphis albiventris, similar to the other species of the genus, is considered a sylvatic taxon characterized by its high competence in adapting to artificial habitats. In fact, opossums are extremely adaptable to both the environment and diet. In the forest, opossums are also skilled tree climbers, although they are somewhat clumsy on the ground. They prey on eggs and small vertebrates as well as insects. The interaction of opossums with humans is quite old. These animals use human buildings as shelters, whereas humans often use opossums as a source of protein. Marsupials from the Didelphis genus are the main reservoir of T. cruzi in different biomes, as demonstrated by high rates of positive hemocultures (Jansen et al., 2018). These animals present generalist habits, use the different forest strata and easily adapt to modified environments, especially the peridomestic environment (Jansen et al., 2018). This set of events explains why these animals always have high rates of infection by T. cruzi, and this would not be different in the Caatinga.

According to our results, the T. cruzi transmission cycle has different profiles in nature, and the animals involved in transmission in this region of the Caatinga biome include D. albiventris and R. rattus species. Both studied species, but mainly Didelphis spp., are considered important reservoirs of T. cruzi (Jansen et al., 2015; Jansen et al., 2018). Moreover, in the studied localities, we observed these mammal species display subpatent parasitemias by an impressive richness by T. cruzi genotypes in the absence of new human cases, showing how powerful the adoption of easy prophylactic measures, such as improving the residences, may be effective against Chagas disease.

The DTUs TcI, TcII, and TcIV were previously described as circulating in small wild mammals with the latter noted in triatomines in the Caatinga biome. DTU TcI is the most common genotype found in T. cruzi (Zingales et al., 2012; Jansen et al., 2015; Brenière et al., 2016). DTU TcII, which was formerly associated with human disease, was the second most frequent in wild mammals, and similar findings were noted in the studied locality (Lisboa et al., 2015; Dario et al., 2017b). This is the first report of the DTU TcIV infecting triatomine in the Caatinga biome. This finding, which was previously observed only in triatomines of the Atlantic Forest (Dario et al., 2016; Dario et al., 2017b; Dario et al., 2018), demonstrates that this DTU can infect a variety of triatomine species and landscapes. This finding shows that no triatomine species is correlated with a certain DTU.

The DTU TcI is a genotype that presents an intraspecific genetic diversity diagnosed by several molecular markers (Salazar et al., 2006; Herrera et al., 2007a; O’Connor et al., 2007; Spotorno et al., 2008; Falla et al., 2009; Ramirez et al., 2011; Roman et al., 2018). Several subdivisions of TcI have been proposed. In Colombia, this genotype is subdivided into two groups through nuclear targets and multilocus sequence typing: one associated with wild transmission (TcI_SILV) and the other with domestic transmission cycles (TcI_DOM) (Ramirez et al., 2011). Here, through NGS, we were able to detect this intraspecific diversity reported in TcI, as we have observed seven different haplotypes circulating in small wild mammals and triatomines, including five new haplotypes that seem to be observed exclusively in the Caatinga biome. Roman et al. (2018) reported a set of TcI isolates from the Caatinga biome that grouped together with no other sequence from a different biome, which could indicate a geographical association. To this end, we should use the same methodology to affirm this association.

We observed different trypanosomatid distributions in Jaguaruana municipality. The Caatinguinha location was the area with the greatest diversity of TcI haplotypes and can be considered a hotspot area for this genotype. The Saquinho, Figueiredo do Bruno, and Córrego das Melancias locations presented the greatest trypanosomatid richness, and some T. cruzi DTUs were reported only in specific locations. Even though we are dealing with a single biome, we can affirm that microhabitats may influence parasite occurrence and diversity as observed in this region.

This is the second study employing NGS involving mammals and trypanosomatid infection in Brazil. The first was performed in the Atlantic Forest and investigated trypanosomatid infection in bats (Dario et al., 2017a). Here, we observed trypanosomatid species richness circulating in R. rattus and D. albiventris, i.e., nonflying mammals. Both studies revealed an unsusceptible richness of trypanosomatids in these distinct taxa, demonstrating that this diversity is common. Our findings reinforce the marsupial as a bioaccumulator of DTU TcI and demonstrate R. rattus as an important host of this genotype in this region. An interesting point to highlight is how this DTU diversity was able to maintain itself among these animals despite the low parasitemias. One explanation is that the triatomines presented the same haplotypes observed in the mammals, so they can be responsible for the transmission by being predated by the animals.

Trypanosoma rangeli lineage A and B infections were observed in D. albiventris and R. rattus. This is the first report of T. rangeli lineage B in the Caatinga, demonstrating that this lineage, together with lineage A (Dario et al., 2021b) presents a broad distribution in Brazil. Didelphis albiventris infection by T. rangeli lineage A was previously described (Dario et al., 2021b). We are adding one more rodent species infected by T. rangeli in Brazil, R. rattus. This new finding demonstrates that rodents with different ecological habits are capable of being infected with this species. Trypanosoma dionisii gains another mammalian host (R. rattus), proving to be a generalist species. Trypanosoma lainsoni, a species described in rodents (Naiff and Barrett, 2013), was observed in D. albiventris for the first time in the Caatinga biome; the first report on this species was in the Cerrado (Nantes et al., 2021), expanding its distribution spectrum. Here, we also report for the first time R. rattus infected by C. mellificae, a so-called monoxenous trypanosomatid. This species is increasingly dispersed among mammals in nature (Rangel et al., 2019; Alves et al., 2021; Dario et al., 2021a). This finding supports the likelihood that C. mellificae has a very eclectic vector or different vectors that make this species so widespread in Brazilian biomes, even in environments with a certain degree of degradation, as the infection was reported in a synanthropic animal.

Dogs presented high serological rates for T. cruzi infection in 2005 (Lima et al., 2012) and 2007 in the present study, but no parasitological exam yielded positive results. It is interesting that in Caatinguinha and Córrego das Melancias, locations where T. cruzi infection was detected in R. rattus and D. albiventris, the infection by T. cruzi was also reflected in domestic dogs, as most were infected. This observation confirmed the role of dogs’ sentinel host in the T. cruzi transmission cycle (Xavier et al., 2012), indicating the presence of the wild transmission cycle in the surroundings of the domestic environment.

In addition, the finding of so many ZOTUs/ASVs demonstrated that the Trypanosoma genus and within it, the T. cruzi clade, still offer unanswered questions. We highlight the following important questions here: What will be the impact of these mixed infections on the host in terms of their biological condition and infective potential? Are these ZOTUs/ASVs able to maintain themselves in the memory host or just in the vectors or the opposite? The same questions apply to C. mellificae, a monoxenous trypanosomatid that is increasingly found in subpatent infections in a broad spectrum of mammals. In conclusion, we can affirm that the Caatinga biome demonstrates remarkable trypanosomatid species/genotypes and haplotype richness. Two species are involved in T. cruzi enzootic transmission, D. albiventris and R. rattus, demonstrating that each animal has a different role in the transmission cycle (varies in different places and times) and therefore the importance of its investigation. The role of the dog in signaling the presence of T. cruzi was reinforced. Once again, there is much more to be unraveled than knowledge about trypanosomatids in nature given that new interactions and dispersions are observed. Thus, we are discovering a new world based on these findings.

The datasets presented in this study can be found in online repositories. The name of the repository and accession numbers can be found below: SRA, NCBI; PRJNA382386: SSR17675332 to SSR17675366.

The animal study was reviewed and approved by Ethical Committee for Animal Use of the Oswaldo Cruz Foundation (P0179-03).

AJ, MD, SX, and CF contributed to the conception and design of the study. CL and SX organized the database. MD, CF, FS, CL, FO, and SX performed the research and analyses. MD and AJ wrote the first draft of the manuscript. All authors contributed to manuscript revision and read and approved the submitted version.

This study was funded by the Fundação Oswaldo Cruz, Instituto Nacional de Câncer, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Fundaçao de Amparo a Pesquisa do Estado do Rio de Janeiro (Faperj). MD receives a postdoctoral fellow from Faperj (E-26/202.414/2019). FO receives a master’s grant from CAPES. FMS receives a postdoctoral fellow from CAPES (88887.162877/2018-00). AR is financially supported by CNPq/Universal (425293/2018-1) and Jovem Cientistas do Nosso Estado/Faperj (E-26/202.794/2019). SX received financial support from CNPq/Universal (422489/2018-2). AJ is financially supported by CNPq (Bolsista de Produtividade, nível 1A).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.