Trypanosoma cruzi is a unicellular eukaryotic organism that causes the Chagas disease or American Trypanosomiasis, a chronic endemic illness of Latin America, and a neglected tropical disease (Pérez-Molina and Molina, 2018). Nowadays, it has been estimated that there are around 6–7 million of chronically infected people (World Health Organization). This parasite has a very complex life cycle that includes an invertebrate hematophagous triatomine vector and a broad range of mammalian hosts (Clayton, 2010). Also, T. cruzi has different biological stages (de Souza et al., 2010; Rodrigues et al., 2014). The non-infective epimastigotes are present in the midgut of triatomines where they differentiate into infective metacyclic trypomastigotes, that after the infection of host cells are differentiated into the replicative amastigote stages and subsequently into infective trypomastigotes that reach the bloodstream (Echeverria and Morillo, 2019). Chagas disease presents an acute phase with low mortality and symptomatology. Then, the patients can remain in an asymptomatic chronic phase for life or in the 30–40% of them will produce after 10–30 years chronic myocarditis, megavisceras, or both (Rassi et al., 2012).

T. cruzi is characterized for showing a great genomic heterogeneity and plasticity across strains (Reis-Cunha et al., 2015; Callejas-Hernández et al., 2018; Herreros-Cabello et al., 2020). This diversity at the genomic level, has promoted the creation of different methods for the classification of hundreds of strains described to date (Zingales et al., 2009; Zingales et al., 2012; Barnabé et al., 2016). These classifications have been established based on some conserved genetic sequences (genomic, mitochondrial and microsatellite DNA). Besides, some researchers have suggested that both the parasite and host genetic variability could be the causes of differential clinical manifestations of Chagas disease (Manoel-Caetano Fda and Silva, 2007). Also, transcriptomic analysis revealed differences between virulent and non-virulent strains (Belew et al., 2017; Oliveira et al., 2020). Indeed, some T. cruzi strains highly differ in pathogenicity. Some are acute lethal strains as Y, whereas others, such as VFRA can produce a chronic infection in BALB/c mice (Rodriguez et al., 2014). Furthermore, proteomic analysis, comparing strains with different pathogenicity, indicates that strains inducing chronic infection have enriched antioxidant defenses, while those inducing acute infections produce nucleotides and proteins involved in parasite replication and lethality (Herreros-Cabello et al., 2019).

All the Trypanosomatids have a single large mitochondrion per cell (Maslov et al., 2019). Its mitochondrial DNA is a network of concatenated circular molecules of maxicircles and minicircles that is called the kinetoplast (kDNA). This structure contains dozens of maxicircles (20-40 Kb) and thousands of minicircles (0.5-10 Kb) with varying sizes depending on species (Shapiro, 1993; Lukes et al., 2002; Thomas et al., 2007).

Kinetoplasts may play a role in the pathogenicity of T. cruzi, and some researchers suggest the minicircles can integrate into the host genome generating autoimmune events (Simoes-Barbosa et al., 2006; Hecht et al., 2010). Also, maxicircle gene deletions have been associated with asymptomatic patients of Chagas disease (Baptista et al., 2006). Both maxicircles and minicircles have been proposed as targets for molecular detection of T. cruzi DNA (Morel et al., 1980; Schijman et al., 2011; Rusman et al., 2019).

Maxicircles contain the characteristic mitochondrial genes of other eukaryotes (Simpson, 1987) and it has been shown that their sequence is characterized by two main regions: the coding region, highly conserved across strains and the divergent/variable region, very difficult to sequence due to its repetitive nature and length variability. Recent findings of maxicircles using third-generation sequencing technologies have revealed that the sequence length may differ across strains, but more importantly, for some strains, its complete sequence is longer than previous estimations (Gerasimov et al., 2020).

Minicircles are exclusive to Trypanosomatids and they are directly involved in U-insertion/deletion editing system as they encode guide RNAs (gRNAs) (Simpson et al., 2003; Aphasizhev and Aphasizheva, 2014; Smith et al., 2020). Moreover, it is suggested that both molecule populations are heterogeneous in the cell, showing strain-specific variations (Westenberger et al., 2006; Messenger et al., 2012).

Several studies have elucidated that the set of minicircles in T. cruzi presents a conserved structure among strains of approximately 1.4 Kb, being organized into four highly conserved regions (mHCRs) of 120 bp located 90 degrees apart from each other, and an equal number of hypervariable regions (mHVRs) of 330 bp interspersed between these conserved regions (Leon et al., 1980; Gonzalez, 1986; Macina et al., 1986; Degrave et al., 1988; Avila et al., 1990; Shapiro and Englund, 1995; Guhl et al., 2002; Junqueira et al., 2005). Also, mHVRs diversity have gained attention since they are involved in specific functions that are unique for Trypanosomatids. mHVRs code for gRNAs that direct the edition of several mitochondrial mRNAs converting these primary transcripts into functional messages (Simpson et al., 2003).

The presence of three conserved sequence blocks (CSB) in Trypanosomatids minicircles. CSB-1 (10 bp sequence) and CSB-2 (8 bp sequence) show lower interspecies homology, while CSB-3 (12 bp sequence), known as the Universal Minicircle Sequence (UMS), is conserved within most Trypanosomatids and constitutes the minicircle replication origin (Ray, 1989). Also, the number, size, and location of these CSBs in minicircles seem to differ between species (Ponzi et al., 1984; Sugisaki and Ray, 1987; Degrave et al., 1988; Botero et al., 2018). Concretely, the CSB-3 or UMS is the specific binding site of the UMS binding protein (UMSBP), which has been related with the minicircle replication and kDNA segregation (Milman et al., 2007). UMSBP has been widely studied in Crithidia fasciculate (Tzfati et al., 1995; Abu-Elneel et al., 1999; Abu-Elneel et al., 2001; Onn et al., 2006), but its presence has been revealed in other Trypanosomatids such as T. cruzi (Milman et al., 2007; Herreros-Cabello et al., 2019), Leishmania donovani (Singh et al., 2016) or T. brucei (Milman et al., 2007). According to these studies the consensus sequence for each CSB, including T. cruzi, would be: CSB-1=AGGGGCGTTC, CSB-2=CCCCGTAC and CSB-3=GGGGTTGGTGTA.

However, a similar analysis is lacking in T. cruzi. Therefore, our study aimed to assemble, annotate, and analyze the complete repertoire of maxicircle and minicircle sequences of different T. cruzi strains.

We report relevant information regarding maxicircles and minicircles kinetoplast DNA isolated from cell-derived parasites. Using second and third generation sequencing techniques we unraveled more complexity than previously reported. Being aware of the limitations of dealing with this complex parasite we were able to obtain valuable results by comparing data from different technologies.

Contrary to the first maxicircles assembled and described in T. cruzi (Westenberger et al., 2006), our results suggest that their total length may be larger than previous suggestions, at least for some strains as one analyzed here. Interestingly, this work indicated that the mitochondrial genes are highly conserved between strains and the coding region is about 16 Kb long, independently of the maxicircle length. The main difference between strains (at maxicircle level) corresponds to the variable/divergent sequence length, which also does not have conserved motifs or repetitive patterns. Besides, clear differences in %GC content between the coding and variable regions were found, suggesting, on one hand, the importance of the maxicircle stability and structure and on the other hand, that the transcription efficiency in T. cruzi is highly correlated with a proportional sequence composition.

We detected a total of 286 different minicircles. This is the first time that the complete repertoire of different minicircles of a particular T. cruzi strain is described. The general structural organization of some T. cruzi minicircles was described (Degrave et al., 1988; Avila et al., 1990; Guhl et al., 2002; Junqueira et al., 2005). Their common scheme would be a minicircle with a size around 1.4 Kb, in which four mHCRs of 120 bp are interspersed with four 330 bp regions that would correspond to the mHVRs. However, our results suggest that this organization, at least in some strains, is more complex, suggesting a new potential aggrupation in at least four main groups of minicircles according to their size and the number of mHCRs and mHVRs: 336-376 bp (group 1), 698-769 bp (group 2), 1051-1095 bp (group 3) and 1357-1448 bp (group 4). Minicircles of group 1 have only one mHCR and mHVR. Group 2 minicircles have two mHCRs and mHVRs. Those belonging to group 3 present three mHCRs and mHVRs. And finally, minicircles of group 4 have four mHCRs and mHVRs. The most abundant was group 4, with 188 members, and the less abundant the group 2 with just 9.

The circularity of minicircles was validated by PCR amplification using specific primers of different groups that amplified the common regions detected at both 3’ and 5’ ends of the reads. Primers were designed for the Y strain, but were also amplified in the Bug2148 strain, except for one set of primers, suggesting that these minicircles may be also present in this strain.

Interestingly, the four groups we have described here in Y strain differ in size by about 300 bp, and this distance is always conserved between the mHCRs in all the groups. Also, the size of the mHCRs is around 120 bp in all of the groups. Therefore, the common patterns of distance are the same as those previously described, which may explain the number of mHCRs depending on the size of the minicircle. Previous analysis detected minicircles around 1.4 Kb (Degrave et al., 1988; Avila et al., 1990; Guhl et al., 2002; Junqueira et al., 2005), as those ascribed to our group 4 (the group with more members). Thus, the new techniques of Next Generation Sequencing used here have likely improved the detection of all minicircles regardless of their abundance.

Also, we analyzed the presence and number of the CSB-1, CSB-2, and CSB-3 in the mHCRs of the Y strain minicircles. Previous studies had determined a consensus sequence for each CSB in other Trypanosomatids, although little differences in length and nucleotide composition have been described between species for the CSB-1 and CSB-2 (Barrois et al., 1981; Jasmer and Stuart, 1986; Nasir et al., 1987; Botero et al., 2018). However, in our study of the Y strain of T. cruzi we have highlighted that this theoretical consensus sequence for T. cruzi is larger than the previous one, with more conserved nucleotides in both ends in all the CSBs of the Y strain minicircles. Interestingly, there are even small differences as changes in the nucleotide proportion in a position or nucleotide insertions between mHCRs of distinct minicircle groups and between the mHCRs of the same minicircle group.

For decades, many researchers have used for their experiments this standard organization of 1.4 Kb and 4 mHCRs and 4 mHVRs of the T. cruzi minicircles, although sequences in the NCBI databases of some minicircles suggest a variety of sizes between less than 100 bp and more than 1000 bp. However, the vast majority of submitted sequences correspond to unpublished data, partial or only the hypervariable regions of specific minicircles sequences (Telleria et al., 2006; Velazquez et al., 2008). Therefore, our contribution with the new full complement of different 286 minicircles of the Y strain supposes a great advance in the knowledge and current data that we have of this type of circular structures.

While T. cruzi has four groups of minicircles, other species as Trypanosoma rangeli (T. rangeli) display three different classes of minicircle sequences that are known as KP1, KP2, and KP3 according to the number of conserved regions. They may present one conserved region (KP1), two conserved regions located at 180 degrees (KP2), or four conserved regions, located at 90 degrees (KP3) ( (Recinos et al., 1994; Vallejo et al., 2002). However, while in T. cruzi the size of the minicircles varies between 336 and 1448 bp, T. rangeli displays a less size variation in minicircles with a range of 1.6-1.8 Kb (Vallejo et al., 1994). In Trypanosoma copemani, minicircles have 2048 bp of size and two classes were suggested considering the number of conserved and variable regions: the G1M1 minicircles with two conserved and two variable regions, and the G1M2 minicircles with four conserved and four variable regions (Botero et al., 2018).

In Leishmania major 97 different minicircles were detected, and 49 in Leishmania infantum, although in these species the size of their minicircles is very uniform, between 660-876 bp and 775-832 bp respectively (Camacho et al., 2019), while T. cruzi presents a bigger size minicircle heterogeneity. Moreover, the researchers described the minicircles of Leishmania with only one conserved region, while for the same minicircle size T. cruzi displays two, and the CSBs present differences in sequence length and nucleotide composition between both species. Taking into account that these conserved regions contain the replication origin, it seems that Leishmania only needs one to perform the completed minicircle replication, while T. cruzi needs one each 300 bp, probably due to the different efficiency in this biological process between both Trypanosomatids.

Mitochondrial metabolism and gene expression are highly regulated to deal with all those complicated environmental changes across the complete life cycle alternating between the mammalian host and insect vector, including regulation of mRNAs that require extensive uridine insertion/deletion (U-indel) editing for their maturation, as has been described in other closely related Trypanosomatids (Smith et al., 2020). To our knowledge, it is the first time describing a large repertoire of complete T. cruzi minicircle sequences.

We sequenced DNA from cell-derived trypomastigotes while in previous reports used epimastigotes. It could be argued that kinetoplast DNA may change from one stage to another. According to Riou and Gutteridge (1978), there should not be differences in DNA content between different stages of T. cruzi. On the other hand, Schwabl et al. (2019) found genomic changes over the same stage just by culture passage. Thus, it would be interesting to further investigate whether the increase in complexity could be due to changes in DNA content between epimastigotes and trypomastigotes.

Finally, considering that there are minicircles conserved among strains and their relevant role for the maturation of the maxicircle-encoded transcripts, critical for the replication and survival of the parasite, their analysis may contribute to the understanding of the mitochondrial transcription and translation processes potentially related to the Chagas disease.

The complete maxicircle sequences for Y and Bug2148 strains are available from the Genbank database accession numbers MW732647, and MW732648, respectively.

FC-H, MF, and NG conceived the idea, MF and NG acquired funding. FC-H, AH-C, and JM-S performed the experiments. FC-H and AH-C wrote the first version of the manuscript. All authors reviewed and edited the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by: “Ministerio de Economía y competitividad” and “Fondo Europeo de Desarrollo Regional” (SAF2015-63868-R (MINECO/FEDER) to NG, and SAF2016-75988-R (MINECO/FEDER) to MF); “Ministerio de Ciencia, Innovación y Universidades-Agencia Estatal de Investigación” and “Fondo Europeo de Desarrollo Regional” (PGC2018-096132-B-I00 (MICINN/FEDER) to NG); “Red de Investigación de Centros de Enfermedades Tropicales” (RICET RD12/0018/0004 to MF); Comunidad de Madrid (S-2010/BMD-2332 to MF); FC-H was recipient of a Ph.D. studentship number 411595 (“Consejo Nacional de Ciencia y Tecnología” (CONACYT, Mexico) and the “Consejo de Ciencia, Tecnología e Innovación de Hidalgo” (CITNOVA, Mexico)); AH-C was recipient of a FPU contract of the “Ministerio de Ciencia, Innovación y Universidades”; JM-S was recipient of a “Garantía Juvenil” predoctoral fellowship (Comunidad de Madrid); and Institutional grants from “Fundación Ramón Areces” and “Banco de Santander”.

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.