Biological and Genetic Heterogeneity in Trypanosoma dionisii Isolates from Hematophagous and Insectivorous Bats

This study describes the morphological, biochemical, and molecular differences among Trypanosoma dionisii isolates from hemocultures of hematophagous (Desmodus rotundus; n = 2) and insectivorous (Lonchorhina aurita; n = 1) bats from the Atlantic Rainforest of Rio de Janeiro, Brazil. Fusiform epimastigotes from the hematophagous isolates were elongated, whereas those of the insectivorous isolate were stumpy, reflected in statistically evident differences in the cell body and flagellum lengths. In the hemocultures, a higher percentage of trypomastigote forms (60%) was observed in the hematophagous bat isolates than that in the isolate from the insectivorous bat (4%), which demonstrated globular morphology. Three molecular DNA regions were analyzed: V7V8 (18S rDNA), glycosomal glyceraldehyde 3-phosphate dehydrogenase gene, and mitochondrial cytochrome b gene. The samples were also subjected to multilocus enzyme electrophoresis and random amplified polymorphic DNA analysis. All isolates were identified as T. dionisii by phylogenetic analysis. These sequences were clustered into two separate subgroups with high bootstrap values according to the feeding habits of the bats from which the parasites were isolated. However, other T. dionisii samples from bats with different feeding habits were found in the same branch. These results support the separation of the three isolates into two subgroups, demonstrating that different subpopulations of T. dionisii circulate among bats.


Introduction
Trypanosomatids are widely distributed in nature and according to the number of hosts participating in their biological cycle, are classified as monoxenous (one host) or dixenous (two or more hosts) [1]. Due to its wide morphological diversity, its taxonomic classification has always been a challenge, since traditional morphological and biological studies, to the current integration with genomics analysis and improvements in microscopy approaches [2]. Among the different genera of the Trypanosomatidae family, the Trypanosoma genus includes species that are found in all classes of vertebrate hosts. The protozoan Trypanosoma dionisii is a parasite found in the Americas (Brazil, Bolivia, Pathogens 2020, 9,736 2 of 12 and the United States), Europe (Belgium, England, and Czech Republic), and Asia (China and Japan) [3][4][5][6][7][8][9]. In Brazil, T. dionisii is found in almost all biomes, including the Amazon, Atlantic Forest, Cerrado, Pantanal, and transition areas of the Pantanal and Cerrado. The protozoan infects different bat families, including Phyllostomidae, Molossidae, Noctilionidae, and Vespertilionidae in various regions from northern to southern Brazil [4,[10][11][12][13].
For many decades, T. dionisii was considered restricted to bats and believed to be transmitted only by Cimicidae, which acts as a vector [14][15][16]. Recent studies have shown that T. dionisii is not restricted to only bats; it also infects different mammalian hosts and vectors, including human cardiac tissue [17], Carnivora clot blood [18], the Didelphimorphia [18] species, and the digestive tract of the triatomine, Triatoma vitticeps [17]. These recent descriptions demonstrate that very little information is available about the dynamics of the transmission cycle of this parasite in nature, and the hosts and vectors involved.
T. dionisii is categorized into the Trypanosoma cruzi clade with other species of the subgenus Schizotrypanum, such as Trypanosoma cruzi cruzi, Trypanosoma cruzi marinkellei, and Trypanosoma erneyi. In fact, T. dionisii was referred to as T. cruzi-like for many decades due to its morphological similarity in terms of blood and cultured forms to T. c. cruzi [14]. Both T. dionisii and T. c. cruzi can invade mammalian cells [19] and alternate developmental forms between hosts, with epimastigotes and metacyclic trypomastigotes appearing in the invertebrate host and bloodstream trypomastigotes and amastigotes found in the mammalian host during the life cycle [16,[19][20][21].
Studies evaluating the biochemical and biological characteristics of T. dionisii have been undertaken for many decades [22][23][24], and current research focuses on identification and phylogenetic analysis. A number of genetic markers, such as ribosomal RNA gene (rDNA) regions (18S and internal transcribed spacer (ITS)), glycosomal glyceraldehyde 3-phosphate dehydrogenase gene (gGAPDH), and mitochondrial cytochrome b gene (Cytb), can be used for genetic analyses of T. dionisii [4,5,[25][26][27]. Previous studies have focused on recognizing T. dionisii genotypes solely on the basis of molecular data [28]. In this study, integrated analysis of morphological and biochemical (as complementary elements) characteristics, along with molecular data, played an important role in the characterization of Trypanosoma species, especially to identify intraspecific differences. These approaches were employed to compare T. dionisii isolates obtained from hemocultures: M1014 and M1015 isolates from hematophagous (Desmodus rotundus) and M1011 isolate from insectivorous (Lonchorhina aurita) bats captured in Rio de Janeiro, Brazil.

Parasite Morphology and Induction of Metacyclogenesis
Parasites were maintained by weekly passages in Novy, McNeal, Nicolle (NNN)/Schneider´s medium and demonstrated extensive polymorphism among the isolates. Isolates M1014 and M1015 were characterized by a large percentage of trypomastigote forms (60%) in all three intervals of observation (three, 10 and 17 days). After three and 10 days, the presence of elongated epimastigote forms and absence of rosettes could be observed in the culture. After 17 days, many forms were observed rounded and destroyed. The isolate M1011 presented some different characteristics: after three days, it presented globular (rounded) epimastigote forms and an absence of rosettes; after 10 and 17 days, the majority of forms were rounded and dividing. M1011 was also observed in degenerate forms, with an absence of rosettes, and rare trypomastigote forms (4%). After metacyclogenesis and induction in Roswell Park Memorial Institute (RPMI) medium, the percentage of trypomastigote forms in the culture moderately increased (13%).
Epimastigote forms obtained from insectivorous bats (M1011) and cultured for 10 days varied in terms of the size of the parasite body, with an average length of 9.29 ± 1.89 µm (6.31-12.71 µm) and size of the free flagellum, with an average length of 9.39 ± 2.19 µm (6.49-12.60 µm). Epimastigote forms from the hematophagous bat (M1014) demonstrated that the parasite body measured 12.97 ± 2.50 µm

Multilocus Enzyme Electrophoresis
The banding patterns of the T. dionisii isolates differed in the glucose phosphate isomerase (GPI), malate dehydrogenase (MDH), isocitrate dehydrogenase (IDH), malic enzyme (ME) isoenzymes, with a clear division into two groups: group 1 consisted of the isolate from L. aurita (M1011) and group 2 of the isolates from D. rotundus (M1015 and M1014; Supplementary Figure S1). The other five enzymes analyzed showed no differences among the T. dionisii isolates.

Molecular Characterization and Sequencing Analysis
Sequence analysis of DNA products obtained with the three molecular targets in comparison with DNA sequences of Trypanosoma spp. deposited in Genbank confirmed the identification of the three isolates as T. dionisii (Supplementary Table S1).
Phylogenic analysis based on the three different targets demonstrated that the evaluated isolates clustered (100% bootstrap) in a monophyletic assemblage with T. dionisii species, generating very similar phylogenetic trees. In the T. dionisii clade, the three sequences included in this study were clustered into two separated branches with high bootstraps according to the food habits of the bats in which the parasites were isolated; one branch with the isolate from the insectivorous bat (M1011) and the other with the two isolates from the hematophagous bats (M1014 and M1015). Phylogenetic

Multilocus Enzyme Electrophoresis
The banding patterns of the T. dionisii isolates differed in the glucose phosphate isomerase (GPI), malate dehydrogenase (MDH), isocitrate dehydrogenase (IDH), malic enzyme (ME) isoenzymes, with a clear division into two groups: group 1 consisted of the isolate from L. aurita (M1011) and group 2 of the isolates from D. rotundus (M1015 and M1014; Supplementary Figure S1). The other five enzymes analyzed showed no differences among the T. dionisii isolates.

Molecular Characterization and Sequencing Analysis
Sequence analysis of DNA products obtained with the three molecular targets in comparison with DNA sequences of Trypanosoma spp. deposited in Genbank confirmed the identification of the three isolates as T. dionisii (Supplementary Table S1).
Phylogenic analysis based on the three different targets demonstrated that the evaluated isolates clustered (100% bootstrap) in a monophyletic assemblage with T. dionisii species, generating very similar phylogenetic trees. In the T. dionisii clade, the three sequences included in this study were clustered into two separated branches with high bootstraps according to the food habits of the bats in which the parasites were isolated; one branch with the isolate from the insectivorous bat (M1011) and the other with the two isolates from the hematophagous bats (M1014 and M1015). Phylogenetic analysis using a combined dataset of V7V8 18S rDNA and gGAPDH ( Figure 2) generated similar phylogenetic trees compared to the tree constructed using Cytb sequences (Supplementary Figure S2). analysis using a combined dataset of V7V8 18S rDNA and gGAPDH ( Figure 2) generated similar phylogenetic trees compared to the tree constructed using Cytb sequences (Supplementary Figure  S2).  Table 1.

Random Amplified Polymorphic DNA (RAPD)
The dendrogram shows that the three isolates grouped together in a branch separated from the reference samples, but once more, this clade presented two clear branches: one containing isolates from D. rotundus (M1014 and M1015) and the other the isolate from L. aurita (M1011), supporting the existence of two subgroups (Supplementary Figure S3).

Discussion
The interest in identifying differences among isolates of the same species of trypanosome began with the discovery of T. c. cruzi and its life cycle by Carlos Chagas [29]. To date, most of these studies have concentrated on the T. c. cruzi and T. rangeli species, as high morphological, biological,  Table 1.

Random Amplified Polymorphic DNA (RAPD)
The dendrogram shows that the three isolates grouped together in a branch separated from the reference samples, but once more, this clade presented two clear branches: one containing isolates from D. rotundus (M1014 and M1015) and the other the isolate from L. aurita (M1011), supporting the existence of two subgroups (Supplementary Figure S3).

Discussion
The interest in identifying differences among isolates of the same species of trypanosome began with the discovery of T. c. cruzi and its life cycle by Carlos Chagas [29]. To date, most of these studies have concentrated on the T. c. cruzi and T. rangeli species, as high morphological, biological, biochemical, and molecular variability has been observed in these species [30,31]. Studies regarding differences among T. dionisii isolates are scarce, and most recent publications have focused solely on the molecular identification of this parasite in both hosts and vectors from Brazil and some other countries [7,8,[10][11][12]18]. The three isolates of T. dionisii investigated (M1011, M1014, and M1015) were obtained in a study in different municipalities of Rio de Janeiro, where 22 samples of trypanosomatids were isolated by hemoculture of 84 bats [32]. Three samples were characterized as T. dionisii and 19 as T. madeirae [33]. Interestingly, we observed that T. dionisii and T. madeirae [33] were present in the same colony and were infecting the same bat species, Desmodus rotundus (data not shown). However, although co-infection is common among bats in nature, it was not detected in this study, probably because the isolation and amplification methods used for the culture of parasites exert a selective force, owing to which mixed infections are not frequently detected.
The classic identification of trypanosomatid protozoans depends on their isolation and subsequent in vitro maintenance; although it is a critical process, it enables morphological, biological, biochemical, and molecular analysis. In this study, the isolation in culture medium allowed the characterization of the three samples as T. dionisii using a DNA barcoding approach with three different molecular markers: V7V8 18S rDNA [34], gGAPDH [35], and Cytb [36]. To the best of our knowledge, this is the first report of T. dionisii infecting Lonchorhina aurita, an insectivorous bat.
The integrative methodologies used in this study were in agreement regarding of the overall result that the isolates from D. rotundus were similar to each other and different to the isolate from L. aurita and support the clear separation of the three isolates into two subgroups. T. dionisii sequences exhibited two different profiles. According to the phylogenetic analysis, the sequences obtained from the molecular targets for the M1011 samples showed greater similarity with T. dionisii sequences obtained from insectivorous (Myotis nigricans) and frugivorous (Sturnira lilium) bats. With regard to the M1014 and M1015 isolates, the sequences obtained in this study showed greater similarity with T. dionisii sequences obtained from hematophagous (D. rotundus) and frugivorous (Carollia perspicillata and Lophostoma brasiliense) bats. It was not possible to determine any type of association on subgroups of T. dionisii and feeding habits from the information obtained from phylogenetic analysis. Hamilton et al. [28] demonstrated in a previous study which used phylogenetic analysis using 18S rRNA and gGAPDH sequences, clustered T. dionisii group A contained sequences from Europe only, and group B contained sequences from Europe and South America.
Phylogenetic analysis also confirmed that T. dionisii sequences from isolates obtained from hemathophagous and insectivorous bats were more similar to T. c. cruzi and T. cruzi marinkellei and closely related to T. erneyi, all species of the Schizotrypanum subgenus ( Figure 2 and Supplementary Figure S2). Hamilton et al. [37] evaluated the relationships among T. c. cruzi, T. cruzi marinkellei, T. dionisii, and other species that comprise a single clade (clade T. cruzi), providing an interesting link between bats and T. cruzi evolution [26,38].
By multilocus enzyme electrophoresis (MLEE), random amplified polymorphic DNA (RAPD), and molecular sequencing, we found that the T. dionisii samples showed two different profiles, with samples M1014 and M1015 exhibiting the same profile and sample M1011 presenting a completely different one. Cavazzana et al. [4] demonstrated that there is only a 0.47% divergence in V7V8 of 18S DNA sequences from Brazilian T. dionisii isolates, and 2% divergence with T. dionisii from Europe [4].
Morphological analysis using light microscopy showed differences among the M1011, M1014, and M1015 samples, which were later confirmed with morphometric analysis. Another observed difference was that the isolates M1014 and M1015, naturally contained a higher percentage of trypomastigote forms compared to that of the isolate M1011. A blood smear test was also performed for the three samples, however, it was negative for all of them. Thus, it was not possible to compare the morphology of the parasites in axenic culture with parasites in blood smear.
Cimicids, the only recognized vectors of T. dionisii, have not been described in the state of Rio de Janeiro, however, we cannot confirm its absence. It is worth mentioning that T. dionisii has already been found in species of triatomines that could play a role in the transmission of this parasite, mainly through the oral route [17]. In this study, despite the three bats being from the same locality but different colonies, they could have been exposed to different vectors. In addition, numerous other vectors are involved in the transmission of different Trypanosoma species, including fleas [39], flies [40], Pathogens 2020, 9, 736 7 of 12 sandflies [41], and triatomines [42]. It is not yet known which vectors are associated with bats and possibly act as disseminators of trypanosomatids that infect these animals. Both sandflies [43] and triatomines [44] are parasited by trypanosomatids that use bats as a host. Cave-type triatomines (Cavernicola pilosa and Cavernicola lenti) are described as possible vectors of some Schizotrypanum species, such as T. cruzi marinkellei [45]. An alternative possibility of direct transmission of T. dionisii among bats in this region should be considered due to their habit of aggregating into colonies [16]. The isolate M1011 was isolated from the insectivorous bat L. aurita, described as having cave-dwelling and colony-forming habits [46]. In a captive study, Thomas et al. [47] demonstrated that bats can be infected by different species of trypanosomes orally via the ingestion of triatomines as well as through contamination when exposed to the feces and urine of these insects or through bites.
Due to little information available regarding the host specificity of most trypanosomes, the diversity of trypanosomes may have been considerably over-or underestimated with respect to their choice of host [28]. T. dionisii, previously believed to infect only bats and cimids, has been described in other mammalian hosts and vectors [17,18]. Therefore, a restricted association with its hosts cannot be guaranteed for any species of Trypanosoma. This leads us to question whether heterogeneity among isolates within the same species has to do with the adaptation to different hosts and if it is of medical importance to humans. To understand the true variability between T. dionisii isolates, studies integrating morphology, biology and molecular biology must be performed with isolates from different bat species and different locations.

Isolation and Light Microscopy of Samples
Samples were isolated from hemocultures from two hematophagous bats, D. rotundus (M1014 and M1015), and one insectivorous bat, L. aurita (M1011), which were caught in the municipal district of Miracema in Rio de Janeiro in 2007. All bat captures and blood collections were performed in association with the Bat Maintenance Program conducted by researchers from the Rio de Janeiro State Agricultural Research Corporation (PESAGRO-RIO). The bats were captured in wild and peri-urban areas at pre-established locations, such as near corrals, water shackles, basements, or natural bat shelters, employing mist-nets that were equipped at dusk. The captured bats were carefully removed from the nets and classified according to family, gender, and clinical appearance. For blood collection, the animals were anesthetized with ketamine at a concentration calculated according to body mass, and up to 1 mL of blood was collected via cardiac puncture. Afterward, the animals were identified with a collar and released at the same capture site [32].
The flagellated protozoa were isolated and maintained in biphasic medium NNN (Novy, McNeal, Nicolle) with Schneider and supplemented with 10% fetal bovine serum at 26-28 • C. The cultures were monitored weekly for 30 days. In all experiments, the three isolates were in the same passage in the axenic culture medium described above.

Parasite Morphology and Induction of Metacyclogenesis
Light microscopy was performed on Giemsa-stained smears using M1011, M1014, and M1015 parasites from culture with NNN and Schneider medium supplemented with 10% fetal bovine serum by three, 10, and 17 days. Additionally, epimastigote forms of the M1011 isolate were transferred at 1 × 10 6 parasites/mL into Roswell Park Memorial Institute (RPMI) medium with 5% fetal calf serum at pH 8.0 for differentiation and acquisition of trypomastigote forms, following a protocol described by Koerich et al. [48]. The percentage of trypomastigote forms was determined by counting 100 randomly selected forms on the slides. Analyses were carried out under optical microscopy at 1000× magnification by examining the characteristics of the culture forms. All the experiment was performed in triplicate.
Pathogens 2020, 9, 736 8 of 12 Measurements of the parasites (µm) were taken for both the total length of the cell body and the free flagellum of 20 randomly selected parasites. Photomicrographs were obtained using Motic Image Plus 2.0 software with the optical microscope Nikon Eclipse E200. Statistical analysis was performed to compare the values of body and flagellum measurements of M1011 and M1014 samples using a one-way ANOVA test. The null hypothesis for the test was that the two means were equal, and a comparison of means whose alpha decision level was equal to or less than 0.05 was used.

Polymerase Chain Reaction (PCR) Assays
PCR assays were performed using DNA extracted from cultured trypanosomes. Approximately 20 mL of culture were centrifuged and washed twice in sterile phosphate buffer saline (1 M; pH 7.2) to pellet the cells, and DNA was extracted using DNAzol kit (Invitrogen, USA) according to the manufacturer's instructions.
PCR amplification of V7V8 18S rDNA was performed to amplify an approximately a 900 bp fragment under conditions described by Marcili et al. [34]. The gGAPDH gene from the isolates was amplified using primers G1/G2 (900 bp), as described by Hamilton et al. [35]. The primers used to amplify Cytb gene region encoded in the maxicircle DNA of the mitochondrial genome were P18/P20 (500 bp) [36]. In all assays, water was used as negative control and DNA from T. c. cruzi DTU TcII (Y strain) as a positive control. The PCR products were purified using the QIAquick Purification Kit (Qiagen, Manchester, UK) to prepare them for sequencing.

Sequencing Analysis
All nucleotide sequences were determined using an automatic sequencer (3730 DNA Analyzer, Applied Biosystems) and analyzed using the Basic Local Alignment Search Tool (BLAST) program (http://blast.ncbi.nlm.nih.gov/Blast.cg). All nucleotide sequences obtained for each molecular target (V7V8 18S rDNA, gGAPDH, and Cytb) were aligned and compared using the Molecular Evolutionary Genetics Analysis (MEGA) program version 7.0.26 [51] with bat trypanosome species and other trypanosomatids available from GenBank (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC540017/).
The concatenated alignment V7V8 18S rDNA and gGAPDH results were analyzed by neighbor-joining (NJ), maximum likelihood (ML), and Bayesian (B) methods using Kimura 2-parameter model with gamma distributed rate variation. To evaluate the robustness of the nodes in the resulting phylogenetic tree for NJ and ML methods, 1000 bootstrap replications were performed using MEGA version 7.0.26 [52]. The models of sequence evolution and their parameters were calculated using the jModelTest in MEGA version 7.0.26 [52]. Bayesian inference (BI) was run in MrBayes (version 3.1.1) [53] with a Kimura 2-parameter model with gamma distributed rate variation. The runs converged after 1,000,000 generations and discarding the first 25% of the trees as burn-in. Phylogenetic tree inferred by Maximum likelihood with the Tamura-Nei model of partial sequences of Cytb genes of T dionisii from this study and reference sequences deposited in GenBank were aligned with the MEGA program version 7.0.26 [53]. Sequences of bat trypanosomes from this study, the species included in the phylogenetic trees and their respective host, geographical origin, and GenBank accession numbers are shown in Table 1.

Random Amplified Polymorphic DNA (RAPD)
The genetic variability among isolates was assessed by random DNA amplification using four arbitrary sequence primers with 10 nucleotides (Pharmacia Biothec Ready-To-Go TM RAPD Analysis Kit), according to the manufacturer's recommendations. The amplified products were analyzed on an agarose gel (2%), stained with ethidium bromide, and visualized under ultraviolet light. The RAPD profiles were analyzed using the Jaccard similarity coefficient to determine the proportion of similar pieces among all isolates. The matrix was transformed into a dendrogram by an unweighted method of grouping in pairs using the arithmetic mean (unweighted pair group method using arithmetic averages (UPGMA)) [54]. Numerical analysis was performed using the NTSYS-pc software program (Version1.70, Exeter software).
Supplementary Materials: The following are available online at http://www.mdpi.com/2076-0817/9/9/736/s1, Figure S1: Schematic representation of the isoenzyme patterns for glucose phosphate isomerase (GPI), malate dehydrogenase (MDH), isocitrate dehydrogenase (IDH), malic enzyme (ME). A-Trypanosoma rangeli (Choachi), B and C-Trypanosoma c. cruzi (Y, CL Brener), D-Trypanosoma desterrensis, E, F and G-Trypanosoma dionisii (M1011, M1014, M1015). Figure S2: Phylogenetic analyses demonstrating the clear separation of Trypanosoma dionisii isolates into two groups. Phylogenetic tree inferred by Maximum likelihood with the Tamura-Nei model of partial sequences of cytochrome b genes of T dionisii from this study and reference sequences deposited in GenBank were aligned with the Molecular Evolutionary Genetics Analysis (MEGA) program version 7.0.26. Outgroup: Trypanosoma rangeli. Bootstrap test results (1000 replicates) are shown next to the branches GenBank accession numbers are shown in Table 1. Figure S3: Unweighted pair group method using arithmetic averages (UPGMA) dendrogram built with the simple matching coefficient of similarity based on the genetic profiles obtained from Random Amplified Polymorphic (RAPD) among isolates. T. dionisii M1011 was clustered separated M1014 and M1015 samples. T. c. cruzi Y, T. cruzi ClBrenner, T. rangeli Choachi and T. desterrensis were used as reference and were clustered totally separated of the T. dionisii samples. Table S1: The sequence analysis of DNA products obtained with the three molecular targets in comparison with DNA sequences of Trypanosoma spp. deposited in GenBank confirmed the identification of the three isolates as Trypanosoma dionisii. Graphic S1: Body and flagellum measurements of parasites used in morphometric analysis. A. M1011, B. M1014.