https://orcid.org/0000-0001-7650-5388
Figures
Abstract
Species delimitation has always been a challenge for taxonomists and for Leishmania studies there is no exception. Herein we attempt to display the usefulness of the mitochondrial gene Cytochrome Oxidase I–coI in classical and barcode-based approaches for Leishmania characterization. A total of 228 samples were analyzed, comprising 28 Leishmania related taxa, mainly from cultures of the Oswaldo Cruz Foundation`s Leishmania Collection. Primers were designed for amplification of coI; sequences were analyzed by distance-based indicators and both the Neighbor Joining and NeighborNet as species grouping techniques. Automatic Barcode Gap Discovery was applied to define species delimitation while for the character-based analysis a software for Barcoding with Logic formulas was employed. Final sequences of 486 bp with 238 parsimonious sites were aligned and edited. Robust groups were formed for most of the genus species, distinctive nucleotide positions in the barcode sequence were observed for 11 of them. A good agreement between the techniques applied and the original characterization was observed. Few species were not distinguished by coI: (i) L. (V.) peruviana, L. (V.) lindenbergi, and L. (V.) utingensis; (ii) L. (L.) venezuelensis and (iii) L. colombiensis and L. equatorensis with identical sequences. Some of these taxa have been, at one time or another, classified as controversial and, for most of them, a higher number of isolates should be studied to properly infer their taxonomic status. CoI represents a mitochondrial target that stands out as a taxonomically important asset with multiple advantages over other genes. This paper corresponds to the first report of coI analysis in Leishmania, a potentially advantageous target for the characterization of this parasite.
Citation: Mata-Somarribas C, Cardoso das Graças G, de Oliveira R. Pereira L, Côrtes Boité M, Motta Cantanhêde L, Braga Filgueira CP, et al. (2024) Applying a cytochrome c oxidase I barcode for Leishmania species typing. PLoS ONE 19(12): e0309277. https://doi.org/10.1371/journal.pone.0309277
Editor: Felipe Dutra-Rêgo, Oswaldo Cruz Foundation Rene Rachou Institute: Fundacao Oswaldo Cruz Instituto Rene Rachou, BRAZIL
Received: August 8, 2024; Accepted: September 30, 2024; Published: December 2, 2024
Copyright: © 2024 Mata-Somarribas et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All DNA sequences are available from the GenBank, NCBI database. Accession numbers are as seen in S1 Table.
Funding: CMS received a postgraduate scholarship from IOC as part of his masters’ degree. This research was funded by Projeto de Ações Estratégicas e de Fortalecimento - Instituto Oswaldo Cruz PAEF-IOC/Fiocruz, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq; 302622/2017-9) and Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro [FAPERJ; E-26/202.569/2019 (245678)]. PAEF-IOC/FIOCRUZ: https://servicos.fiotec.org.br/PortalAcessoInformacao/DetalheProjeto.aspx?k=1654 CNPq: https://www.gov.br/cnpq/pt-br FAPERJ: https://www.faperj.br/ The authors declare that the funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Defining species delimitation has been a challenge for taxonomists and for Leishmania there is no exception. The name Leishmania was first proposed for this parasite in 1903 by Ross, referring to the Leishman Donovan bodies previously described in patients’ smears suffering from kala-azar [1]. To this day, this protozoan genus has five recognized subgenera that includes around 40 species, more than 20 pathogenic to humans [2]. Many of these species were first described based on ecological, epidemiological and biological characteristics, regarded as extrinsic characters. Former reliance on these extrinsic characters changed after global application of polyphasic taxonomy to diverse groups of organisms, especially bacteria. This method takes into account all available phenotypic and genotypic data and integrates them in a consensus type of classification [3]. Application of these techniques to Leishmania was first undertaken with data obtained from isoenzyme electrophoresis, a biochemical technique introduced in the 1970s and used for decades [4]. Multilocus analyses based on enzyme electrophoresis (MLEE) came to be the gold standard for Leishmania characterization, and, until now, continues to be the reference for new species identification methods applied for the genus [5, 6].
Despite its contribution, MLEE presents drawbacks, such as the lack of resolution to differentiate gene sequences that share enzymatic phenotypes; or to interpret heterozygous phenotypes, substitutions, post-translational modifications [7, 8], intraspecies variability or convergent evolution [9]. Other important disadvantages include its requirement of bulk cultures, its inability to be applied on clinical samples or uncultured parasites, its labor intensive and time-consuming process, and its limited distribution as a readily available technique, as it can only be performed in a few reference centers [6, 10]. Even in these specialized environments, the request for reference strains to be used in every run as standards, demands additional and continuous flow of reagents, culture mediums, cryopreservation protocols and dedicated technicians, hindering practical application of this technique. Ultimately, the MLEE has never been standardized among the different specialized laboratories as a cross-sectional and homogeneous procedure. Therefore, comparison of results obtained in different laboratories was never possible.
More recently, studies have developed and employed molecular methods for Leishmania typing to overcome the above-mentioned MLEE limitations [6]. However, considering the many Leishmania species described and/or validated by MLEE, it is essential to properly compare results and thus perform a data-based revision. One common strategy of molecular studies was to use as targets the coding gene sequences of the most relevant enzymes for species identification applied in MLEE. The typing method, based on short sequences from a number of different loci, was named “multilocus sequence typing” or MLST [11]. The clearest advantage of this methodology is the higher resolution, allowing the detection of discrete variation and the identification of relationships between strains. Additionally, the approach offers data portability for sharing results between laboratories in a digital way. For these reasons, MLST has become the method of choice for typing many organisms [12].
Different studies have confirmed the agreement between the MLEE and MLST results [7, 8, 13], also reporting better resolution and congruence [14–16]. However, the lack of databases for Leishmania, added to the low number of characterized strains and the lack of consensus on markers have made it difficult to implement MLST as a reference method [17]. For these reasons, the nomenclature “multilocus sequence analysis–MLSA” is most often used in the literature on Leishmania, where the word “typing” is replaced by “analysis”. The MLSA methodology has proven to be a good tool for the characterization of strains and epidemiological surveillance, as well as for the analysis of population structure and evolutionary studies [18]. The amount of information provided by this technique opened the possibility of making inferences about the genetic diversity of the evaluated isolates, and the sequences obtained could be used to determine population structures and to point out evidence of recombination events.
Regarding the array of approaches to MLSA, the method has no immediate applicability for routine use but rather for establishing the genus’s phylogeny [10]. Its application on clinical samples is restricted due to its limited sensitivity in such samples. Routinely amplifying several genes in parallel and then sequencing them is too costly and time-consuming [6].
The multiplicity of protocols currently available for researchers makes the choice of a particular typing methodology a difficult task. Moreover, the lack of an unequivocal classification scheme is essentially the result of using different markers to study phylogeny and of the lack of a universally applicable species definition [6]. Herein, we aim to contribute to this issue by presenting a simple, cross-laboratorial method to type and define Leishmania species through the barcode rational. Barcode is a system designed to ideally provide fast, accurate and automated identifications using species internal markers, providing resolution at the species level. Using gene sequencing, barcoding was intended so that a reliable sequence read can be obtained with a single pass on conventional sequencing platforms [19]. We bring to attention the advantages and usefulness of the mitochondrial gene Cytochrome Oxidase I–coI, both, as a single locus marker—referred as barcoding, and as part of a MLSA methodology for Leishmania characterization. Recombination disrupts signals of differentiation in the nuclear genome, and mitochondrial DNA is non-recombining, notwithstanding that more studies are needed in this particular subject. Our purpose is to evaluate the potential of coI as a barcode for Leishmania species identification. For this, we applied a combination of distance-based and character-based methods to assess the performance of coI sequences for distinguishing among different Leishmania spp.
Materials and methods
Leishmania strains
A total of 228 strains were analyzed (Table 1), comprising 28 Leishmania species or related taxa:
Of these strains, 217 were obtained from cultures of the Oswaldo Cruz Institute’s Leishmania Collection (CLIOC), previously typed by MLEE (S1 Table). The other 12 samples correspond to publicly available sequences of the coI target of interest, readily accessible in the GenBank® genomic database of the National Center for Biotechnology Information (NCBI), part of the National Institutes of Health (NIH), located after application of the BLAST tool.
Cell culture and DNA extraction
Culture strains preserved in liquid nitrogen were thawed and cultivated in Schneider’s medium at 25° and supplemented with 20% (v/v) heat-inactivated fetal bovine serum to a density of 1 x 107 promastigotes cells / mL (late log phase), as estimated by counting in a Neubauer chamber. The parasites were washed by centrifugation with TE following Internal Standards Protocol (ISP). DNA extraction was performed using the Wizard DNA purification Kit (Promega, Madison, USA) following the manufacturer’s instructions.
PCR protocol
Primers were designed for the conserved region of coI of different Trypanosomatidae family members. For this, sequences available at GenBank® for Trypanosoma brucei (M14820), T. cruzi (FJ203996), Leishmania (L.) donovani (FJ416603), L. (S.) tarentolae (NC000894) and Crithidia oncopelti (X56015) were employed (Fig 1). These primers are not genus-specific, as shown in Fig 1, where Leishmania related trypanosomatids were included in the design. The proposed amplicon has a calculated size of 572 bp for all related trypanosomatids analyzed. Primer information is shown on Table 2. Amplification reactions had, for 50 μL total volume, 0.2 μM of each primer, 1.5 mM MgCl2, 0.2 mM deoxyribonucleotide triphosphate (dNTPs), 1U Platinum Taq DNA Polymerase, 1X commercial reaction buffer and approximately 50 ng DNA. Amplification conditions were: 95° for 2 minutes, followed by 35 cycles at 95° for 30 seconds, 52° for 50 seconds and 72° for 1 minute, with a final extension at 72° for 10 minutes.
Schematic map of the A.) Coding portion of the maxicircle genome of Leishmania tarentolae and B.) The multiple sequence alignment of the coI gene of related trypanosomatids. A.) Protein coding genes abbreviations are shown inside boxes. Diagram size is proportional to gene size in Leishmania tarentolae (Accession Number NC_000894.1). Solely larger genes are shown and its corresponding abbreviations are: 12S and 9S, 12S rRNA and 9S rRNA; ND7, NADH dehydrogenase subunit 7; COIII, cytochrome c oxidase III; CYb, apocytochrome b; A6, ATPase subunit 6; URF1, maxicircle unidentified reading frame 1; ND1, NADH dehydrogenase subunit 1; COII, cytochrome c oxidase II; URF2, maxicircle unidentified reading frame 2; COI, cytochrome c oxidase I; ND4, NADH dehydrogenase subunit 4; ND5, NADH dehydrogenase subunit 5. Green arrows above the boxes represent genes that are transcribed left to right, red arrows below the boxes indicate genes transcribed in the opposite direction. Overall coI size is shown under the yellow box, spanning from 11181 to 12830 bp in the L. tarentolae published maxicircle genome, the diagram is based on Stuart and Feagin 1992 and Feagin 2000 [20, 21]. B.) The multiple sequence alignment of the coI gene between related trypanosomatids and the proposed primers was conducted using the Bioedit package (BioEdit Sequence Alignment Editor 7.2.5, Tom Hall, 1999). The annealing region can be seen marked with a yellow box, for both the forward and reverse primers, with a total length spanning from 11957 to 12575 bp. The final amplification product was calculated around 572 bp. Nucleotides that are highly conserved in these species are shaded with different colors depending on the nucleotide.
DNA sequencing
The PCR products were purified by “MinElute 96 UF PCR Purification Kit” (Qiagen, Hilden, Germany) following the protocol recommended by the manufacturer. After purification the DNA was quantified in a NanoDrop®. The DNA mass used for sequencing averaged around 7.5 ng, the plate was prepared with 0.2 μM per primer and H2O up to 7.5 μL. Samples were delivered to the DNA Sequencing Platform by Capillary Electrophoresis (Sanger)—RPT01A, part of the Fiocruz’s Network of Technological Platforms, for processing in the automatic sequencer “ABI 3730 DNA Analyzer” (Applied Biosystems, CA, USA).
Forward and reverse sequences of coI were edited, assembled and merged into consensus sequences using MEGA X (Molecular Evolutionary Genetics Analysis 10.2.2, Pennsylvania State University, PE, USA), and BioEdit (BioEdit Sequence Alignment Editor 7.2.5, Tom Hall, 1999) packages. Sequences were aligned using MEGA X and BioEdit, applying ClustalW (Conway Institute, University College Dublin, Ireland) parameters.
Distance-based and clustering analysis
A set of indicators were used to evaluate the performance and discriminating power of the target region (Table 3). The Overall Mean Distance (OMD) is the number of base differences per site obtained from averaging over all sequence pairs [22]. In this case, the OMD was calculated for 228 strains. The Within Group Mean Distance (WGMD) calculates the average number of base differences per site across all sequence pairs within each group, defining as a group a species for which more than one strain has been analyzed. The Between Groups Mean Distance (BGMD) represents the average number of base differences between all pairs of sequences in the proposed groups (S2 Table). The Mean Diversity Within the Population (DWP) is calculated by the number of base differences, per sequence, of the average of the diversity calculations, for the entire population and for any given subpopulations (DWS) [23]. These diversity calculations were generated automatically by MEGA X, applying equation number 12.73. The standard error of these calculations was obtained by the “bootstrap” method (1000 replicates). The codon positions included were 1st + 2nd + 3rd + Noncoding, and all ambiguous positions were removed for each pair of sequences (pairwise deletion).
Species groupings, or “clusters”, were obtained by building trees using the Neighbor Joining (NJ) method [24], which is not concerned with grouping the most closely related Operational Taxonomic Units, but rather with minimizing the size of all internal nodes and thus the size of the entire tree. The p-distance (proportion of nucleotide sites at which two sequences are different) was computed to infer the differences in nucleotide composition between sequences. Node support was evaluated with 1000 bootstrap replicates. Bootstrap equal or above 90% were considered for groups definition.
Strain clusters were also obtained by the NeighborNet method, which proposes networks based on a distance matrix given by the Uncorrected-P or Hamming distance, in which, for any two sequences of equal length, the Hamming distance is the number of positions at which the corresponding symbols are different [25]. For this goal the Splitstree program (SplitsTree4 4.17.1, Eberhard Karls University of Tübingen, Tübingen, Germany) was used.
Automatic Barcode Gap Discovery (ABGD)
To evaluate the barcode species delimitation, the ABGD method (Automatic Barcode Gap Discovery, web version 08/26/21) was applied to the DNA sequences obtained from the range of Leishmania species. This method statistically infers the “Barcode Gap” of the sequence data and partitions the species accordingly. When differences between sequence pairs are compared in a dataset, it is possible to observe a “gap” that separates intraspecific and interspecific diversity. CoI sequence data was processed in ABGD using simple distance, the value of intraspecific divergence was set between 0.001 and 0.1, with 10 recursive steps and a gap width set of 1.5.
Character-based analysis
To evaluate the barcode sequence proposed herein, the BLOG software (Barcoding with LOGic formulas v2, Istituto di Analisi dei Sistemi e Informatica, Rome, Italy) was applied. This method performs a character-based analysis that identifies, for each species in a reference library, the distinctive nucleotide positions of the DNA barcode sequence, and assigns to each species classification formulas that can tightly characterize a species [26].
Results
Herein, we gathered and analyzed nucleotide sequences covering the mitochondrial coI gene, representing 228 Leishmania related strains. These strains had been previously identified at the species level by MLEE and/or other molecular method (S1 Table). The strains included, accounted for 1 Leishmania related parasite (Genus Endotrypanum—"nomen dubium"), 3 Leishmania subgenera (Viannia, Leishmania, Sauroleishmania) and 27 Leishmania species.
Characteristics of coI sequences
PCR products of around 570bp were analyzed after coI amplification. Final sequences of 486 nucleotides were obtained after alignment and editing for all Leishmania species, showing 213 conserved sites, 273 variable, 35 singleton and 238 parsimonious sites. CoI sequences from the sampled strains showed an A + T bias (average = 68,5) relative to the C + G content (average = 31,5). The average content of each nucleotide was as follows: T = 22.9, C = 19.5, A = 45.6, and G = 12.0.
Distance-based analysis
The overall mean p-distance between sequences, OMD, was 0.126, ranging from 0.0 to 0.049 within groups, WGMD (Table 3). The lowest average between groups, BGMD, was 0.000, between L. colombiensis and L. equatorensis, and, in non-Viannia species, was 0.015 between L. (L.) infantum and L. (L.) donovani, while the highest was 0.200 between L. colombiensis/L. equatorensis and L. (L.) mexicana. Considering only L. (Viannia) species, the lowest p-distance average was 0.004, between L. (V.) braziliensis and L. (V.) peruviana, and 0.018, between L. (V.) panamensis and L. (V.) shawi, while the highest were observed between L. (V.) lainsoni and L. (V.) shawi (0.054), and L. (V.) lainsoni and L. (V.) guyanensis (0.052) (S2 Table).
After performing the NJ analysis and running 1000 replications with p-distance calculation, the generated tree clustered a large group with a high bootstrap value (100) which corresponds to the subgenus L. (Viannia) (Fig 2). In this cluster, we observed the formation of sub-groups that correlate, for the most part, with the analyzed species, with the exception of L. (V.) peruviana, L. (V.) utigensis and L. (V.) lindenbergi, which grouped together with sequences of L. (V.) braziliensis. Sequences corresponding to subgenus L. (Sauroleishmania) form a cluster with high support (100). For the genus Endotrypanum and those Leishmania species not assigned to any subgenus, a particular cluster was observed, with two branches, one including L. deanei and L. hertigi and the other including L. colombiensis and L. equatorensis.
On the left, the original NJ tree can be observed with all the analyzed species. On the right, the cutout shows L. (Viannia) species clustering. Color shading was applied to simplify tracking from the original tree.
The NJ tree also indicated that sequences from the Old-World species, classified in the subgenus L. (Leishmania), formed a robust group with subgroups clearly related to the analyzed species (Fig 3). It is important to highlight that coI allows the identification of species of the L. (L.) donovani complex, although only one sequence of L. (L.) archibaldi was analyzed and few of L. (L.) donovani. Sequences referring to the L. (L.) mexicana complex formed two groups. One group with high support is composed of L. (L.) amazonensis from La Paz, Bolivia, two L. (L.) venezuelensis from Lara, Venezuela, and all of L. (L.) mexicana. The other group was formed by the two other L. (L.) venezuelensis from Lara, Venezuela, L. (L.) garnhami and the remaining L. (L.) amazonensis from South America, mainly from Brazil.
On the left, the original NJ tree can be observed with all the analyzed species. On the right, the cutout shows L. (Leishmania) species clustering. Color shading was applied to simplify tracking from the original tree.
The NeighborNet analysis (Fig 4) provided clear identification of species clusters agreeing with the original species characterization. As in the NJ tree, NeighborNet splits show a clear definition of species clusters for the L. (Viannia) and L. (Sauroleishmania) subgenus and for L. deanei and L. hertigi. For the L. (Leishmania) subgenus, defined clusters for L. (L.) major, L. (L.) donovani, L. (L.) infantum, L. (L.) tropica, L. (L.) aethiopica, L. (L.) turanica and L. (L.) archibaldi can be observed, but interpretation remains a challenge for the L. (L.) mexicana complex, particularly when it comes to L. (L.) amazonensis.
NeighborNet splits can be shown graphically separating groups of species. Over the NeighborNet splits the distinct ABGD groups were labelled, showing agreement between different analysis. Segments of the NeighborNet splits were cutout and zoomed in, for visualization ease. Color lettering was applied to simplify species identification.
Automatic Barcode Gap Discovery (ABGD)
Applying ABGD, nine values of “prior intraspecific divergence” (P) were evaluated to identify potential barcode gaps. Considering P values less than 0.005, since values above this threshold tend to underestimate the number of species, 28 partitions were observed for this data set (Table 4). The ABGD algorithm agreed with the previous characterization of each sample, separating species into 28 groups. The subgenus L. (Viannia) was partitioned from the rest of samples and L. (V.) panamensis, L. (V.) lainsoni, L. (V.) naiffi and L. (V.) shawi were each identified as distinct groups. However, L. (V.) braziliensis continued to be grouped with L. (V.) linderbergi, L. (V.) peruviana and L. (V.) utingensis. As previously indicated for other analysis the barcode gap failed to identify some species from the L. (L.) mexicana complex, although it allowed the separation of L. (L.) mexicana from L. (L.) venezuelensis and L. (L.) amazonensis. A high intraspecific variability was observed again in L. (L.) amazonensis, which was classified into five distinct groups. For the species not classified in any subgenera, distinct ABGD groups were observed: L. colombiensis and L. equatorensis were separated from other species as one group, while L. hertigi and L. deanei were partitioned individually.
Character-based analysis
The BLOG software was used to obtain logic formulas, also known as character-based patterns, for each species. These formulas revealed unique and unequivocal nucleotide positions in the DNA barcode sequence for 11 of the 17 taxa that had more than one sequence per species. The other 11 species, from the 28 taxa analyzed, only reported one sequence per species, and so, invalidated the BLOG software logic formulas. As shown in Table 5, the 11 species with a distinctive nucleotide positioning, of the 17 analyzed, had a 0.00 false positive rate for those diagnostic positions. Nevertheless, while analyzing these results it came to attention that two taxa showed a lower Laplacian Score. This score, which allows to rank features based on its locality preserving property [27], was notably low for L. (L.) tropica and L. hertigi, at just 0.069. After analyzing all related sequences, it was found that the distinctive nucleotide positions for L. (L.) tropica were not unique to that species. These positions were also present in L. (L.) aethiopica and L. (L.) archibaldi. The same occurred with L. hertigi, which shared its character pattern with L. deanei.
For another three species, from the group of 17 that had more than one sequence analyzed, L. (V.) braziliensis, L. (L.) amazonensis and L. colombiensis, the BLOG software was able to identify diagnostic characters but with a false positive rate > 0.00. The remaining three species of the selected group, L. (V.) peruviana, L. (L.) venezuelensis and L. equatorensis, did not show any distinctive nucleotide positioning.
Discussion
CoI nucleotide sequences of a representative set of Leishmania species were subjected to classical and barcode-based analysis. Results suggest the target represents a good species marker, considering the high agreement between the ABGD groups, distance-based analysis, monophyletic clades, species clusters and the original characterization by MLEE or molecular identification.
However, a few limitations were found, such as within the L. (V.) braziliensis complex. L. (V.) peruviana, L. (V.) lindenbergi and L. (V.) utingensis clustered with L. (V.) braziliensis strains. These strains were grouped in the same phylogenetic clade, as seen in the NJ analysis (Fig 2). No nucleotide differences were detected between L. (V.) peruviana IOCL-3659 and eight L. (V.) braziliensis strains. Interestingly, these strains were all but one from Peru and from Acre, a nearby region of the Brazilian Amazon. However, the other two L. (V.) peruviana showed a few differences when compared to L. (V.) peruviana IOCL-3659. These two strains displayed no nucleotide differences when compared to L. (V.) lindenbergi and four L. (V.) braziliensis strains not related to the Amazon region. The L. (V.) utingensis sequence was slightly different from these other sequences with an equal or inferior intraspecific variability than that within L. (V.) braziliensis strains.
Regarding L. (V.) peruviana, its status as a species has been questioned many times before [28]. Nevertheless, different analysis based on nuclear targets have seemed to show that L. (V.) braziliensis and L. (V.) peruviana are two closely-related species but distinct, monophyletic lines [29]. The close relationship between these two species involves very similar or indistinguishable genetic targets. Analyses of the mitochondrial Cytochrome b gene (cytB) conducted by independent laboratories were unable to differentiate between both species [30–32], corroborating the coI findings. Few studies have been conducted on L. (V.) lindenbergi and L. (V.) utingensis [33, 34], but they form separate and independent branches and share almost no alleles with other species, thus corroborating MLEE data [18]. More isolates from both species should be studied to properly infer their taxonomic status. Within the L. (L.) mexicana complex, a clear barcode gap was found for L. (L.) mexicana but, curiously, L. (L.) amazonensis strains were split in five clusters, with three of them showing identical nucleotide sequences between different species: i) two L. (L.) venezuelensis and seven L. (L.) amazonensis; ii) two other L. (L.) venezuelensis and three L. (L.) amazonensis); iii) one L. (L.) garnhami and four L. (L.) amazonensis. L. (L.) amazonensis showed the largest WGMD (Table 3), which was also demonstrated in the NJ and NeighborNet clustering, and in the barcode partition which resulted in five different groups of strains.
In this species complex there is still a controversial debate on the status of several taxa. L. (L.) venezuelensis has been described as a valid species on the basis of distinguishable MLEE patterns and specific monoclonal antibodies reactions [35]. Moreover, the occurrence of a heterogeneous population has been reported for this species [36]. However, molecular data is limited, and some authors strongly suggested L. (L.) venezuelensis is actually a variant of L. (L.) mexicana [37]. The current coI findings indeed reveal a great heterogenicity among these samples, but they appear more related to the L. (L.) amazonensis strains, which also show high variability. Therefore, L. (L.) venezuelensis taxonomic status remains inconclusive.
L. (L.) pifanoi and L. (L.) garnhami status has been debated practically since its determination as species [38], but more recently it has been widely accepted that they correspond to synonyms of L. (L.) mexicana and L. (L.) amazonensis, respectively [39, 40].
An argument for inconclusiveness regarding its status can also be made for L. (L.) waltoni, a more recent taxon described from the Dominican Republic [41], and for L. (L.) ellisi, a species isolated from an autochthonous case in Arizona, United States [42]. In the former case, an isolated group of strains from diffuse cutaneous leishmaniases lesions [43] were determined to be more closely related to L. (L.) mexicana, although showing biologically similarities with L. (L.) venezuelensis. In the latter, a single strain from a papular cutaneous lesion was isolated and determined to be closely related to L. (L.) mexicana and L. (L.) amazonensis [44]. These examples exhibit the need for a better understanding of this species complex and the designation of better markers that allow a clear definition of species. More robust molecular information seems to be needed, as it is questionable that all the species of this complex deserved specific status, and that no subspecies can be defined [45].
For the controversial taxa L. colombiensis/L. equatorensis, both were undistinguishable from each other in their coI sequence (S2 Table). Taxonomy regarding species close to the Endotrypanum genus has been debatable for decades. L. colombiensis seems to be more closely related to that genus than its resembling counterpart, L. equatorensis, and more importantly, it has been frequently associated with human disease [46]. L. equatorensis has just recently been isolated from human lesions in Colombia [47] but overall, there are few reports of natural infection by these species in other Latin American countries, either in humans or other animal reservoirs [48]. L. equatorensis and L. colombiensis comprise a distinct group within the Leishmaniinae subfamily based on minicircle analysis [49], monoclonal antibodies, enzyme electrophoretic analysis, restriction fragment analysis, schizodeme analysis and more recent molecular assays [50–54]. The similarity between these two peculiar species has been demonstrated before, and were confirmed by the present coI analysis, in which no differences between the samples could be detected.
Based on these results and previous arguments [51], we maintain Endotrypanum as a "nomem dubium", rejecting the proposal to classify L. colombiensis and L. equatorensis in that genus [46]. Furthermore, the results obtained cannot be used to assign L. hertigi and L. deanei to a separate genus [46], as this would imply a revision of the Leishmania classification, elevating other subgenera to new genera. Without a doubt, this is a picture of how complex Leishmania taxonomy really is and reinforces the need for an urgent revision, employing various markers and shared strains to be analyzed by different research groups.
Selecting the most appropriate species typing technology in a given context can be a challenging task [6]. Several approaches have been used to identify Leishmania species, including the analysis of different regions of the genome, and numerous targets have been sequenced. Identification of Leishmania species is an essential step for epidemiological studies that must address the etiological agent, but it is also pivotal for genetic diversity research in clinical scenarios, given its recognized clinical importance. For instance, in the Americas, several species of Leishmania influence clinical manifestations, severity of the disease, accuracy of diagnosis and response to treatment [55].
The structure of Leishmania’s mitochondrial genome (maxicircle and minicircles) has been characterized, shedding light on particularly useful phylogenetic targets as for example cytB. This gene has been described as a high sensitivity, high specificity uniparentally informative mitochondrial target [47, 56–59]. CytB gene has been included in different MLST sets currently used for phylogenetic studies in Leishmania [60].
As with cytB, coI represents another mitochondrial taxonomically valuable target. Since the proposal of “DNA Barcoding”, it has been widely used in barcode studies of various organisms [19]. A number of characteristics support that mitochondrial genes are good candidates to be used as species markers: rare insertions and deletions [61], offering stability–a desirable trait for species markers. Also, their high copy number that could increase sensitivity in PCR based approaches and their different mutation rates between species [62], that admits for molecular evolutionary based species typing. This marker apparently do not undergo recombination, although in Leishmania this still deserves more studies [63]; it has an assumed haploid inheritance [64] and has low ancestral polymorphism with high conservation between strains [65]. Despite the advantages and the fact that it has already been used for characterization of related pathogens as Trypanosoma cruzi [66], this target has not yet been fully applied to the identification of the Leishmania protozoan. CoI represents a good mitochondrial sequencing target to be added to the arsenal of targets already available for the identification of Leishmania species. CoI shows high discriminatory power and robustness in all analyses, relevant polymorphisms and significant informative sites between evaluated species. Herein we showed coI barcoding being successfully applied to species typing.
The present study corresponds to the first report of coI analysis in Leishmania. The few available sequences from public databases were included herein, in addition to those originally obtained by us, comprising a total of 228 sequences. Despite the number of DNA sequences, the number of strains per species was limited, partially restricting the analysis.
An additional limitation is the quality of sequences deposited in public databases. These are not always originated from taxonomic studies with rigorous postanalytical control compromising the overall quality of the information associated with Leishmania sequences. It is not uncommon to find data with sequencing blunders, wrong nomenclature, lack of biological or molecular information, misidentified species and identification errors in deposited material. For this reason also, using a standardized and curated database of Leishmania strains should be prioritize before use of other repositories. The scientific community involved in Leishmania taxonomy should work towards a standardized and curated database of strains, with the objective of simplifying phylogeny and taxonomy studies, consensus should be the final goal.
The application of mitochondrial targets for Leishmania typing shows cases of divergence when pertaining to New World Leishmania species [30]. This disagreement indicates a mismatch between kinetoplast and nuclear genes, known as mito-nuclear discordance, already reported in isolates from Latin America, specifically observed for the L. (Viannia) subgenus [31]. Further study is needed to disclose the mechanisms involved [67], which possibly include genetic exchange and the formation of hybrids in the nuclear genome, and thus, a mito-nuclear discordance. The frequency and mechanisms are still unclear, but it may be present in unexpectedly high rate (~10%) [68]. These findings invalidate any proposal that applies just one mitochondrial target for typing purposes, or application of just one target for typing, either nuclear or mitochondrial.
Finally, we encourage the scientific community to apply coI directly to clinical samples and address species characterization with this target. Although the protocol described herein could be of limited use because of the final product size, additional primers could be proposed based on the now available coI sequences. Evaluation of coI barcoding for Leishmania species identification showed great potential. Nevertheless, considering the existing limitations of typing Leishmania species with just one single locus marker, coI should be considered as a first line target for MLSA or MLST sets, incorporating both nuclear and mitochondrial genes.
Supporting information
S1 Table. DNA sequences obtained from GenBank®, of the National Center for Biotechnology Information (NCBI), and from the Oswaldo Cruz Foundation’s Leishmania Collection database, corresponding to all available coI sequences from Leishmania spp. strains, for the evaluation of the coI target.
IOC-L is the prefix use by the Oswaldo Cruz Institute’s Leishmania Collection (CLIOC) as part of the deposit code applied to Leishmania strains.
https://doi.org/10.1371/journal.pone.0309277.s001
(DOCX)
S2 Table. Between Groups Mean Distance (BGMD), a distance-based indicator, showing the average number of base differences of all sequence pairs, between the proposed species groups.
https://doi.org/10.1371/journal.pone.0309277.s002
(DOCX)
Acknowledgments
To local laboratories of the Caja Costarricense de Seguro Social and their work regarding surveillance of parasitological diseases, to Ms. Wendy Valverde, Ms. Mariel Chaves and Ms. Stephanie Quirós for their technical support, to Dr. Marlon Gregori and Dr. Franklyn Samudio for their timely answers and recommendations, to the Programa de Pós-graduação em Biologia Celular e Molecular and to the Laboratório de Pesquisa em Leishmanioses do Instituto Oswaldo Cruz (IOC)–Fundação Oswaldo Cruz (Fiocruz).
References
- 1. Ross R. Note on the bodies recently described by Leishman and Donovan. Br Med J. 1903 Nov 14;2(2237):1261–2. pmid:20761169
- 2.
Pan American Health Organization. Manual of procedures for leishmaniasis surveillance and control in the Region of the Americas. 2nd ed. Washington DC: Pan American Health Organization; 2023. 222 p.
- 3. Vandamme P, Pot B, Gillis M, de Vos P, Kersters K, Swings J. Polyphasic taxonomy, a consensus approach to bacterial systematics. Microbiol Rev. 1996 Jun;60(2):407–38. pmid:8801440
- 4.
World Health Organization. Control of the leishmaniases: WHO TRS N°949 Report of a meeting of the WHO Expert Committee on the Control of Leishmaniases, Geneva, 22–26 March 2010. Geneva: World Health Organization; 2012.
- 5. Buffi G, Ceccarelli M, Diotallevi A, Abruzzese M, Bruno F, Castelli G, et al. High-resolution melting (HRM)-based detection of polymorphisms in the malic enzyme and glucose-6-phosphate isomerase genes for Leishmania infantum genotyping. Parasites & Vectors. 2023 Aug 14;16(1):282. pmid:37580789
- 6. Van der Auwera G, Dujardin JC. Species typing in dermal leishmaniasis. Clin Microbiol Rev. 2015 Apr;28(2):265–94. pmid:25672782
- 7. Mauricio IL, Yeo M, Baghaei M, Doto D, Pratlong F, Zemanova E, et al. Towards multilocus sequence typing of the Leishmania donovani complex: resolving genotypes and haplotypes for five polymorphic metabolic enzymes (ASAT, GPI, NH1, NH2, PGD). Int J Parasitol. 2006 Jun;36(7):757–69. pmid:16725143
- 8. Zemanová E, Jirků M, Mauricio IL, Horák A, Miles MA, Lukes J. The Leishmania donovani complex: genotypes of five metabolic enzymes (ICD, ME, MPI, G6PDH, and FH), new targets for multilocus sequence typing. Int J Parasitol. 2007 Feb;37(2):149–60. pmid:17027989
- 9. Jamjoom MB, Ashford RW, Bates PA, Chance ML, Kemp SJ, Watts PC, et al. Leishmania donovani is the only cause of visceral leishmaniasis in East Africa; previous descriptions of L. infantum and “L. archibaldi” from this region are a consequence of convergent evolution in the isoenzyme data. Parasitology. 2004 Oct;129(4):399–409.
- 10. Schönian G, Kuhls K, Mauricio IL. Molecular approaches for a better understanding of the epidemiology and population genetics of Leishmania. Parasitology. 2011 Apr;138(4):405–25. pmid:21078222
- 11. Maiden MCJ, Bygraves JA, Feil E, Morelli G, Russell JE, Urwin R, et al. Multilocus sequence typing: A portable approach to the identification of clones within populations of pathogenic microorganisms. PNAS. 1998 Mar 17;95(6):3140–5. pmid:9501229
- 12. Maiden MCJ, Jansen van Rensburg MJ, Bray JE, Earle SG, Ford SA, Jolley KA, et al. MLST revisited: the gene-by-gene approach to bacterial genomics. Nat Rev Microbiol. 2013 Oct;11(10):728–36. pmid:23979428
- 13. Tsukayama P, Lucas C, Bacon DJ. Typing of four genetic loci discriminates among closely related species of New World Leishmania. Int J Parasitol. 2009 Feb;39(3):355–62. pmid:18817779
- 14. El Baidouri F, Diancourt L, Berry V, Chevenet F, Pratlong F, Marty P, et al. Genetic Structure and Evolution of the Leishmania Genus in Africa and Eurasia: What Does MLSA Tell Us. PLoS Negl Trop Dis. 2013 Jun 13;7(6):e2255. pmid:23785530
- 15. Ravel C, Cortes S, Pratlong F, Morio F, Dedet JP, Campino L. First report of genetic hybrids between two very divergent Leishmania species: Leishmania infantum and Leishmania major. Int J Parasitol. 2006 Dec 1;36:1383–8. pmid:16930606
- 16. Van der Auwera G, Ravel C, Verweij JJ, Bart A, Schönian G, Felger I. Evaluation of Four Single-Locus Markers for Leishmania Species Discrimination by Sequencing. J Clin Microbiol. 2014 Apr;52(4):1098–104. pmid:24452158
- 17. Herrera G, Hernández C, Ayala MS, Flórez C, Teherán AA, Ramírez JD. Evaluation of a Multilocus Sequence Typing (MLST) scheme for Leishmania (Viannia) braziliensis and Leishmania (Viannia) panamensis in Colombia. Parasites & Vectors. 2017 May 12;10(1):236. pmid:28499458
- 18. Boité MC, Mauricio IL, Miles MA, Cupolillo E. New insights on taxonomy, phylogeny and population genetics of Leishmania (Viannia) parasites based on multilocus sequence analysis. PLoS Negl Trop Dis. 2012;6(11):e1888. pmid:23133690
- 19. Hajibabaei M, Singer GAC, Hebert PDN, Hickey DA. DNA barcoding: how it complements taxonomy, molecular phylogenetics and population genetics. Trends Genet. 2007 Apr;23(4):167–72. pmid:17316886
- 20.
Stuart K, Feagin JE. Mitochondrial DNA of Kinetoplastids. In: Wolstenholme DR, Jeon KW, editors. International Review of Cytology. Academic Press; 1992 [cited 2024 Sep 23]. p. 65–88.
- 21. Feagin JE. Mitochondrial genome diversity in parasites. Int J Parasitol. 2000 Apr 1;30(4):371–90. pmid:10731561
- 22. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Mol Biol Evol. 2018 Jun 1;35(6):1547–9. pmid:29722887
- 23.
Nei M, Kumar S. Molecular Evolution and Phylogenetics. Oxford University Press; 2000. 348 p.
- 24. Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol. 1987 Jul 1;4(4):406–25. pmid:3447015
- 25. Wang C, Kao WH, Hsiao CK. Using Hamming Distance as Information for SNP-Sets Clustering and Testing in Disease Association Studies. PLOS ONE. 2015 Aug 24;10(8):e0135918. pmid:26302001
- 26. Weitschek E, Van Velzen R, Felici G, Bertolazzi P. BLOG 2.0: a software system for character-based species classification with DNA Barcode sequences. What it does, how to use it. Mol Ecol Resour. 2013 Nov;13(6):1043–6. pmid:23350601
- 27. He X, Cai D, Niyogi P. Laplacian Score for Feature Selection. Adv Neural Inf Process Syst. 2005;18.
- 28. Bañuls AL, Dujardin JC, Guerrini F, De Doncker S, Jacquet D, Arevalo J, et al. Is Leishmania (Viannia) peruviana a distinct species? A MLEE/RAPD evolutionary genetics answer. J Eukaryot Microbiol. 2000 Jun;47(3):197–207. pmid:10847336
- 29. Bañuls AL, Hide M, Prugnolle F. Leishmania and the leishmaniases: a parasite genetic update and advances in taxonomy, epidemiology and pathogenicity in humans. Adv Parasitol. 2007;64:1–109. pmid:17499100
- 30. Davila M, Pineda V, Calzada JE, Saldaña A, Samudio F. Evaluation of cytochrome b sequence to identify Leishmania species and variants: the case of Panama. Mem Inst Oswaldo Cruz. 2021 Apr 19;116. pmid:33886871
- 31. Kato H, Cáceres AG, Seki C, García CRS, Mauricci CH, Martínez SCC, et al. Further insight into the geographic distribution of Leishmania species in Peru by cytochrome b and mannose phosphate isomerase gene analyses. PLOS Neglected Tropical Diseases. 2019 Jun 20;13(6):e0007496. pmid:31220120
- 32. Koarashi Y, Cáceres AG, Saca FMZ, Flores EEP, Trujillo AC, Alvares JLA, et al. Identification of causative Leishmania species in Giemsa-stained smears prepared from patients with cutaneous leishmaniasis in Peru using PCR-RFLP. Acta Trop. 2016 Jun;158:83–7. pmid:26943992
- 33. Cantanhêde LM, Mattos CB, de Souza Ronconi C, Filgueira CPB, da Silva Júnior CF, Limeira C, et al. First report of Leishmania (Viannia) lindenbergi causing tegumentary leishmaniasis in the Brazilian western Amazon region. Parasite. 2019;26.
- 34. Cantanhêde LM, Cupolillo E. Leishmania (Viannia) naiffi Lainson & Shaw 1989. Parasites & Vectors. 2023 Jun 8;16(1):194.
- 35. Rivas AK, Alcover M, Martínez-Orellana P, Montserrat-Sangrà S, Nachum-Biala Y, Bardagí M, et al. Clinical and diagnostic aspects of feline cutaneous leishmaniosis in Venezuela. Parasites & Vectors. 2018 Mar 20;11(1):141. pmid:29554979
- 36. Bonfante-Garrido R, Meléndez E, Barroeta S, de Alejos MA, Momen H, Cupolillo E, et al. Cutaneous leishmaniasis in western Venezuela caused by infection with Leishmania venezuelensis and L. braziliensis variants. Trans R Soc Trop Med Hyg. 1992 Apr;86(2):141–8. pmid:1440772
- 37. Kato H, Watanabe J, Mendoza Nieto I, Korenaga M, Hashiguchi Y. Leishmania species identification using FTA card sampling directly from patients’ cutaneous lesions in the state of Lara, Venezuela. Trans R Soc Trop Med Hyg. 2011 Oct;105(10):561–7. pmid:21907375
- 38. Momen H, Grimaldi G. On the identity of Leishmania mexicana pifanoi and L. mexicana garnhami. Trans R Soc Trop Med Hyg. 1984;78(5):701–2. pmid:6506164
- 39. Akhoundi M, Kuhls K, Cannet A, Votýpka J, Marty P, Delaunay P, et al. A Historical Overview of the Classification, Evolution, and Dispersion of Leishmania Parasites and Sandflies. PLoS Negl Trop Dis. 2016 Mar 3;10(3):e0004349. pmid:26937644
- 40. Schönian G, Mauricio I, Cupolillo E. Is it time to revise the nomenclature of Leishmania? Trends in Parasitology. 2010 Oct;26(10):466–9. pmid:20609626
- 41. Shaw J, Pratlong F, Floeter-Winter L, Ishikawa E, El Baidouri F, Ravel C, et al. Characterization of Leishmania (Leishmania) waltoni n.sp. (Kinetoplastida: Trypanosomatidae), the Parasite Responsible for Diffuse Cutaneous Leishmaniasis in the Dominican Republic. Am J Trop Med Hyg. 2015 Sep 2;93(3):552–8. pmid:26149864
- 42. de Almeida M, Zheng Y, Nascimento FS, Bishop H, Cama VA, Batra D, et al. Cutaneous Leishmaniasis Caused by an Unknown Leishmania Strain, Arizona, USA. Emerg Infect Dis. 2021 Jun;27(6):1714–7. pmid:34013870
- 43. Schnur LF, Walton BC, Bogaert-diaz H. On the identity of the parasite causing diffuse cutaneous leishmaniasis in the Dominican Republic. Transactions of the Royal Society of Tropical Medicine and Hygiene. 1983 Jan;77(6):756–62. pmid:6665827
- 44. Sapp SGH, Low R, Nine G, Nascimento FS, Qvarnstrom Y, Barratt JLN. Genetic characterization and description of Leishmania (Leishmania) ellisi sp. nov.: a new human-infecting species from the USA. Parasitol Res. 2024;123(1):52.
- 45. Fraga J, Montalvo AM, De Doncker S, Dujardin JC, Van der Auwera G. Phylogeny of Leishmania species based on the heat-shock protein 70 gene. Infect Genet Evol. 2010 Mar;10(2):238–45. pmid:19913110
- 46. Espinosa OA, Serrano MG, Camargo EP, Teixeira MMG, Shaw JJ. An appraisal of the taxonomy and nomenclature of trypanosomatids presently classified as Leishmania and Endotrypanum. Parasitology. 2018 Apr;145(4):430–42. pmid:27976601
- 47. Ramírez JD, Hernández C, León CM, Ayala MS, Flórez C, González C. Taxonomy, diversity, temporal and geographical distribution of Cutaneous Leishmaniasis in Colombia: A retrospective study. Sci Rep. 2016 Jun 22;6(1):28266. pmid:27328969
- 48. Salgado-Almario J, Hernández CA, Ovalle CE. Geographical distribution of Leishmania species in Colombia, 1985–2017. Biomedica. 2019 15;39(2):278–90.
- 49. Fernandes O, Catanho MP, Segura I, Labrada LA, Derre R, Saravia N, et al. Minicircle Variable Region Probes for Characterization of Leishmania (Viannia) Species. The Journal of Parasitology. 1999 Jun;85(3):563. pmid:10386456
- 50. Cupolillo E, Pereira LO, Fernandes O, Catanho MP, Pereira JC, Medina-Acosta E, et al. Genetic Data Showing Evolutionary Links between Leishmania and Endotrypanum. Mem Inst Oswaldo Cruz. 1998 Sep;93(5):677–83. pmid:9830537
- 51. Cupolillo E, Medina-Acosta E, Noyes H, Momen H, Grimaldi G. A revised classification for Leishmania and Endotrypanum. Parasitol Today. 2000 Apr;16(4):142–4. pmid:10725899
- 52. Grimaldi G, Kreutzer RD, Hashiguchi Y, Gomez EA, Mimory T, Tesh RB. Description of Leishmania equatorensis sp. n. (Kinetoplastida: Trypanosomatidae), a new parasite infecting arboreal mammals in Ecuador. Mem Inst Oswaldo Cruz. 1992 Jun;87:221–8.
- 53. Kreutzer RD, Corredor A, Grimaldi G, Grogl M, Rowton ED, Young DG, et al. Characterization of Leishmania colombiensis sp. n (Kinetoplastida: Trypanosomatidae), a new parasite infecting humans, animals, and phlebotomine sand flies in Colombia and Panama. Am J Trop Med Hyg. 1991 Jun;44(6):662–75. pmid:1677544
- 54. Yang BB, Chen DL, Chen JP, Liao L, Hu XS, Xu JN. Analysis of kinetoplast cytochrome b gene of 16 Leishmania isolates from different foci of China: different species of Leishmania in China and their phylogenetic inference. Parasites Vectors. 2013 Dec;6(1):1–12. pmid:23383990
- 55.
Pan American Health Organization. Guideline for the Treatment of Leishmaniasis in the Americas. Second Edition. Second. Pan American Health Organization; 2022.
- 56. Camacho E, Rastrojo A, Sanchiz Á, González-de La Fuente S, Aguado B, Requena JM. Leishmania Mitochondrial Genomes: Maxicircle Structure and Heterogeneity of Minicircles. Genes. 2019 Sep 26;10(10):758. pmid:31561572
- 57. Asato Y, Oshiro M, Myint CK, ichi Yamamoto Y, Kato H, Marco JD, et al. Phylogenic analysis of the genus Leishmania by cytochrome b gene sequencing. Experimental Parasitology. 2009 Apr 1;121(4):352–61. pmid:19159626
- 58. Foulet F, Botterel F, Buffet P, Morizot G, Rivollet D, Deniau M, et al. Detection and identification of Leishmania species from clinical specimens by using a real-time PCR assay and sequencing of the cytochrome B gene. J Clin Microbiol. 2007 Jul;45(7):2110–5. pmid:17475750
- 59. Myint CK, Asato Y, ichi Yamamoto Y, Kato H, Bhutto AM, Soomro FR, et al. Polymorphisms of cytochrome b gene in Leishmania parasites and their relation to types of cutaneous leishmaniasis lesions in Pakistan. J Dermatol. 2008 Feb;35(2):76–85. pmid:18271802
- 60.
Kuhls K, Mauricio I. Phylogenetic Studies. In: Clos J, editor. Leishmania. New York, NY: Springer New York; 2019. p. 9–68.
- 61. Hebert PDN, Cywinska A, Ball SL, deWaard JR. Biological identifications through DNA barcodes. Proc Biol Sci. 2003 Feb 7;270(1512):313–21. pmid:12614582
- 62. Yatawara L, Le TH, Wickramasinghe S, Agatsuma T. Maxicircle (mitochondrial) genome sequence (partial) of Leishmania major: Gene content, arrangement and composition compared with Leishmania tarentolae. Gene. 2008 Nov;424(1–2):80–6. pmid:18718512
- 63. Van den Broeck F, Savill NJ, Imamura H, Sanders M, Maes I, Cooper S, et al. Ecological divergence and hybridization of Neotropical Leishmania parasites. Proc Natl Acad Sci U S A. 2020 Oct 6;117(40):25159–68. pmid:32958676
- 64. Kay C, Williams TA, Gibson W. Mitochondrial DNAs provide insight into trypanosome phylogeny and molecular evolution. BMC Evol Biol. 2020 Dec;20(1):161. pmid:33297939
- 65. Callejas-Hernández F, Herreros-Cabello A, del Moral-Salmoral J, Fresno M, Gironès N. The Complete Mitochondrial DNA of Trypanosoma cruzi: Maxicircles and Minicircles. Front Cell Infect Microbiol. 2021 Jun 29;11. pmid:34268138
- 66. Rodrigues MS, Morelli KA, Jansen AM. Cytochrome c oxidase subunit 1 gene as a DNA barcode for discriminating Trypanosoma cruzi DTUs and closely related species. Parasit Vectors. 2017 Oct 16;10(1):488.
- 67. Tabbabi A, Cáceres AG, Chauca TPB, Seki C, Choochartpong Y, Mizushima D, et al. Nuclear and kinetoplast DNA analyses reveal genetically complex Leishmania strains with hybrid and mito-nuclear discordance in Peru. PLOS Neglected Tropical Diseases. 2020 Oct 19;14(10):e0008797. pmid:33075058
- 68. Kato H, Cáceres AG, Gomez EA, Tabbabi A, Mizushima D, Yamamoto DS, et al. Prevalence of Genetically Complex Leishmania Strains With Hybrid and Mito-Nuclear Discordance. Front Cell Infect Microbiol. 2021 Feb 24;11:625001. pmid:33732663