Molecular characterization and classification of Trypanosoma spp. Venezuelan isolates based on microsatellite markers and kinetoplast maxicircle genes

Background Livestock trypanosomoses, caused by three species of the Trypanozoon subgenus, Trypanosoma brucei brucei, T. evansi and T. equiperdum is widely distributed throughout the world and constitutes an important limitation for the production of animal protein. T. evansi and T. equiperdum are morphologically indistinguishable parasites that evolved from a common ancestor but acquired important biological differences, including host range, mode of transmission, distribution, clinical symptoms and pathogenicity. At a molecular level, T. evansi is characterized by the complete loss of the maxicircles of the kinetoplastic DNA, while T. equiperdum has retained maxicircle fragments similar to those present in T. brucei. T. evansi causes the disease known as Surra, Derrengadera or "mal de cadeiras", while T. equiperdum is the etiological agent of dourine or "mal du coit", characterized by venereal transmission and white patches in the genitalia. Methods Nine Venezuelan Trypanosoma spp. isolates, from horse, donkey or capybara were genotyped and classified using microsatellite analyses and maxicircle genes. The variables from the microsatellite data and the Procyclin PE repeats matrices were combined using the Hill-Smith method and compared to a group of T. evansi, T. equiperdum and T. brucei reference strains from South America, Asia and Africa using Coinertia analysis. Four maxicircle genes (cytb, cox1, a6 and nd8) were amplified by PCRfrom TeAp-N/D1 and TeGu-N/D1, the two Venezuelan isolates that grouped with the T. equiperdum STIB841/OVI strain. These maxicircle sequences were analyzed by nucleotide BLAST and aligned toorthologous genes from the Trypanozoon subgenus by MUSCLE tools. Phylogenetic trees were constructed using Maximum Parsimony (MP) and Maximum Likelihood (ML) with the MEGA5.1® software. Results We characterized microsatellite markers and Procyclin PE repeats of nine Venezuelan Trypanosoma spp. isolates with various degrees of virulence in a mouse model, and compared them to a panel of T. evansi and T. equiperdum reference strains. Coinertia analysis of the combined repeats and previously reported T. brucei brucei microsatellite genotypes revealed three distinct groups. Seven of the Venezuelan isolates grouped with globally distributed T. evansi strains, while TeAp-N/D1 and TeGu-N/D1 strains clustered in a separate group with the T. equiperdum STIB841/OVI strain isolated in South Africa. A third group included T. brucei brucei, two strains previously classified as T. evansi (GX and TC) and one as T. equiperdum (BoTat-1.1). Four maxicircle genes, Cytochrome b, Cythocrome Oxidase subunit 1, ATP synthase subunit 6 and NADH dehydrogenase subunit 8, were identified in the two Venezuelan strains clustering with the T. equiperdum STIB841/OVI strain. Phylogenetic analysis of the cox1 gene sequences further separated these two Venezuelan T. equiperdum strains: TeAp-N/D1 grouped with T. equiperdum strain STIB818 and T. brucei brucei, and TeGu-N/D1 with the T. equiperdum STIB841/OVI strain. Conclusion Based on the Coinertia analysis and maxicircle gene sequence phylogeny, TeAp-N/D1 and TeGu-N/D1 constitute the first confirmed T. equiperdum strains described from Latin America. Electronic supplementary material The online version of this article (doi:10.1186/s13071-015-1129-2) contains supplementary material, which is available to authorized users.


Background
Trypanosomes are unicellular parasites that cause important diseases in humans and animals. They comprise a wide group of parasites of vertebrates usually transmitted by haematophagous arthropods [1]. Livestock trypanosomoses caused by Trypanosoma brucei brucei, Trypanosoma equiperdum and Trypanosoma evansi, all belonging to the Trypanozoon subgenus, has a significant socio-economic impact and limits animal protein productivity throughout the world [2][3][4]. T. evansi, the first pathogenic trypanosome discovered, was found in the blood of horses and camels suffering an endemic disease known as Surra, derrengadera, murrina or "mal de cadeiras" [5]. T. evansi was introduced to America in the XV century through the importation of horses from Africa by Spanish conquerors [5][6][7]. T. evansi is mechanically transmitted and it is now widely distributed throughout tropical and subtropical regions of Northern Africa, Southeast Asia, as well as Central and South America, where livestock and native wildlife are severely affected [2,5,[8][9][10]. In Europe, the importation of dromedary camels from the Canary Islands was associated with T. evansi outbreaks in France and Spain [11,12] and chronic T. evansi infection and death was reported in a dog that returned to Germany after international travel [13].
We previously showed that Venezuelan Trypanosoma spp. isolates separated into two groups, according to their RAPD profiles [52]. Seven of the nine Venezuelan isolates clustered together, while two highly virulent horse isolates, TeAp-N/D1 and TeGu-N/D1, appeared to be genetically distinct [17,52,53]. By analyzing microsatellite loci and PE repeats, we found that the seven less virulent isolates clustered with T. evansi reference strains, while the more virulent TeAp-N/D1 and TeGu-N/D1 isolates closely matched a T. equiperdum reference strain, leading us to look for the presence of maxicircle genes.

Ethical approval
The project was approved by the COBIANIM (IVIC-DIR-1073/12) an advisory body of IVIC with regard to the ethical use of animals in research, in accordance with national and international standards. This committee oversees all research activities at IVIC, requiring the use of animals and wildlife to meet with Venezuelan law and universal ethical values. The Commission assessed the methodological, bioethical and legal aspects of this project by resolution IVIC/N o 1444 [55].

Microsatellite and Procyclin PE-typing
The PCR amplifications for microsatellite analysis were performed with an Eppendorf Mastercycler, as described [34]. Five microsatellite markers were employed, four that were previously used to characterize trypanosome isolates [34], and a fifth new genetic marker MD2.349-CA amplified with primers MD2.349-CA-F (GCATGCGTGAG GAAGTGAG) and MD2.349-CA-R (GTCCTGTTGGC CGCATTAT) and also a sequence corresponding to the Procyclin PE repeats. The lengths of the PCR products were determined using the Genescan software (Applied Biosystems), and they were sequenced using an ABI 3130 XL (Applied Biosystems) at the C.G.F.B. (Functional Genomic Center of Bordeaux, France).

Multivariate analysis
Two matrices were generated based on genotyping with microsatellite and Procyclin PE repeats (Additional file 1) and previously published data on T. brucei brucei [34]. The categorical values matrix contains all the evaluated microsatellite data, classified from the lowest to the highest number of repeats. The binary values matrix was constructed with the Procyclin PE repeats data, where (0) was the absence and (1) the presence of a determined PE repeat. When each qualitative variable is represented by binary indicator variables for each category, specifying whether an object belongs to it (1) or not (0), multiple correspondence analysis (MCA) can be formulated as a principal component analysis (PCA) of the total set of these indicator variables with respect to some predefined metrics. Thus, just as in PCA, object coordinates can be seen as component scores determined up to a rotation only [56].
By applying this transformation it was possible to use the Hill-Smith [57] method to combine categorical and binary variables to compare all Trypanosoma spp. isolates and strains in a Coinertia analysis [58,59]. This method is a combination of a multiple correspondence analysis (MCA) for categorical data matrix and MCA rotated to principal component analysis (PCA) for binary data matrix. It was implemented with functions of the ADE-4 package from the R software® [60][61][62].

PCR amplification of maxicircle genes
Four maxicircle sequences were amplified using either previously reported primers or novel primers designed for this study, based on the complete sequences of the following T. brucei genes: Cytochrome b (cytb-GenBank Accession N°M 17998); Cytochrome oxidase Subunit 1 (cox1 -GenBank Accession N°M14820); ATP synthase subunit 6 (a6-Gen-Bank Accession N°M14820); and NADH Dehydrogenase Subunit 8 (nd8 -GenBank Accession N°M63820.1). Primer sequences and PCR conditions are shown in Additional file 2: Table S1. The cytb and cox1 gene amplifications were performed with DNA from all nine Venezuelan isolates, while the a6 and nd8 amplifications were carried out only with DNA from TeAp-N/D1 and TeGu-N/D1.
The cytb, cox1, a6 and nd8 amplicons were purified prior to sequencing using the AccuPrep® PCR Purification Kit (

Sequence analysis and alignment
Maxicircle sequences from TeAp-N/D1 and TeGu-N/D1 isolates were compared to the corresponding sequences from T. equiperdum, T. brucei sp. and T. brucei brucei retrieved from the Genbank database. Sequence analysis was performed using nucleotide BLAST and the sequences were aligned by MUSCLE tools. Phylogenetic trees were constructed using Maximum Parsimony (MP) and Maximum Likelihood (ML) with the MEGA5.1® software [63] using T. cruzi as an outgroup. For the ML method, the evolution of the aligned sequences was analyzed.

Results
We analyzed nine Trypanosoma spp. isolates obtained from two Venezuelan states (Apure and Guárico) and from three different hosts: horses, donkeys and capybara. We previously showed that the RAPDs profile of the TeAp-N/D1 and TeGu-N/D1 isolates (horse) were quite distinct from the other seven Venezuelan isolates [52]. To further characterize these strains, all nine Trypanosoma spp. isolates were genotyped using microsatellites and Procyclin repeats followed by multivariate analysis, and the two that clustered with T. equiperdum (STIB841/ OVI), TeAp-N/D1 and TeGu-N/D1, were analyzed for specific maxicircle gene sequences.

Genotyping
The results of the microsatellite and Procyclin PE repeats analysis of the nine Trypanosoma spp. Venezuelan isolates, along with the corresponding data from fifteen T. evansi, three T. equiperdum and eighteen T. b. brucei reference strains are presented in Additional file 1. The Trypanozoon reference strains used in this study originated from Latin America (Colombia and Brazil), Asia (China) and Africa (Chad, Kenya, Ethiopia, Gambia, Ivory Coast, Burkina Faso, Nigeria, Democratic Republic of Congo, Uganda, Tanzania and South Africa).
Five SSR microsatellites and the PE repeats of the PARP genes were used to analyze the nine Venezuelan isolates. The resulting genotypes were compared to reference T. evansi and T. equiperdum strains, as well as to eighteen previously reported T. brucei brucei strains [34], by coinertia analysis.

Coinertia analysis
The rotated binary-matrix was constrained with the categorical data matrix by the Hill-Smith method to be able to perform a coinertia analysis that explained 53.68 % and 22.161 % of the observed inertia in the microsatellites hyperspace (X matrix in x axis) and the Procyclin PE repeats hyperspace (Y matrix in y axis), respectively ( Fig. 1a  and 1b). The two markers, MORF2-CA and MEST19-AT/ GT, contributed most to the construction of the coinertia first axis with 80.31 % and 62.58 % on average per allele, respectively. In the second coinertia axis, Repeats 24 and 28 contributed most with 72.67 % and 66.81 %, respectively. Figure 1c revealed three distinct groups, one with T. evansi strains, a second group with all the T. brucei brucei strains and a third group with the T. equiperdum STIB841/OVI strain and the two Venezuelan isolates from this study, TeAp-N/D1 and TeGu-N/D1, shown with black arrows. The remaining seven Venezuelan Trypanosoma spp. isolates are close to the T. evansi reference strains from the first group. The coinertia analysis of the available genotyping data showed that four of the T. evansi strains, namely JX, TC, ET and 80 are genetically distinct. Interestingly, one of them, the TC strain, is closely related to the T. b. brucei group (Fig. 1c). The B1 and BJ T. equiperdum references strains are different from the STIB841/ OVI strain (Fig. 1c). Based on the PE repeats, the T. evansi KETRI2480 and T. equiperdum BJ strains are closely related (Fig. 1c). The wide distribution of T. evansi and T. equiperdum strains among distinct groups strongly supports multiple evolutionary origins for these dyskinetoplastic strains.

Amplification of maxicircle genes
Because the presence of maxicircles is a universally accepted marker to distinguish T. equiperdum from T. evansi, amplification of two maxicircle genes (cytb and cox1), was attempted in the nine Venezuelan Trypanosoma spp. isolates. In agreement with the microsatellite data, no amplification of cytb and cox1 was observed in the seven Venezuelan Trypanosoma spp. isolates that belong to the main T. evansi cluster: TeAp-Cedral05, TeAp-Cedral12, TeGu-Terecay03, TeAp-El Frío01, TeAp-Mantecal01, TeGu-Terecay323 and TeGu-Terecay01. However both cytb and cox1 could be amplified from DNA of the TeAp-N/D1 and TeGu-N/D1 isolates, as were, two additional maxicircle genes, ATP synthase subunit 6 (a6) and NADH dehydrogenase subunit 8 (nd8).
Phylogenetic trees were constructed using both maximum parsimony and maximum likelihood, with T. cruzi as outgroup. The cytb analysis showed that TeAp-N/D1 and TeGu-N/D1 are closely related to the T. equiperdum STIB841/OVI strain (Fig. 2a). The phylogenetic relationship of cox1 sequences shows three different subgroups, one including TeAp-N/D1 and the T. equiperdum STIB818 and three T. brucei strains, a second group comprising TeGu-N/D1 and the T. equiperdum STIB841/ OVI strain and a third group with the T. equiperdum STIB842 and BoTat1.1 strains (Fig. 2b). The phylogenetic construction estimated with the a6 sequences shows the TeGu-N/D1 strain as a separate group, while TeAp-N/D1 showed identity with three T. brucei strains and two T. equiperdum STIB842/ BoTat1.1 strains (Fig. 3a). The nd8 sequences of TeAp-N/D1 and TeGu-N/D1 were identical and related to T. equiperdum STIB842 and BoTat1.1 strains (Fig. 3b).
The pattern and multivariate analysis of microsatellite and PE repeats from seven of the Trypanosoma spp. isolates, TeAp-Cedral05, TeAp-Cedral12, TeAp-ElFrío01, TeAp-Mantecal01, TeGu-Terecay323, TeGu-Terecay01, TeGu-Terecay03, as well as the lack of amplification of two of the maxicircle genes confirmed that these seven isolates belong to the T. evansi group, while the genotype and coinertia analysis of TeAp-N/D1 and TeGu-N/ D1 showed that they are closely related to the T. equiperdum STIB841/ OVI strain (Table 1). Further, comparison of the sequences of maxicircle gene cox1 separated the TeAp-N/D1 and TeGu-N/D1 isolates, so that TeAp-N/D1 isolate clustered with T. equiperdum STIB818 and T. brucei brucei, while the TeGu-N/D1 grouped with the T. equiperdum STIB841/OVI.

Discussion
The classification of trypanosomes within the Trypanozoon sub-genera was originally based on morphological and morphometrical criteria, as well as clinical manifestations, host range and geographical distribution. In the last several years, molecular markers such as microsatellite loci and sequence analysis of the rRNA and gGAPDH  genes have been used to describe the evolutionary relationships among organisms. This has led to the re-evaluation of what constitutes a trypanosome species [5,34,64,65]. The lack of maxicircle genes in T. evansi has been used to differentiate it from T. equiperdum [33], both being considered as petite mutants of T. brucei [51]. Subtle genomic changes were found between the akinetoplastic strain T. evansi STIB805 and the T. b. brucei TREU 927/4 strain, supporting the view that T. evansi should be considered a subspecies of T. brucei [66]. The species or subspecies status of both T. evansi and T. equiperdum has been debated by several authors and continues to be polemical [4,16,51,66,67].
Using RAPD analysis, we previously showed that nine Venezuelan Trypanosoma spp. isolates cluster into two separate groups, one with seven isolates that share a similar genetic pattern and a second, distinct group of two horse isolates, TeAp-N/D1 and TeGu-N/D1 that appeared identical except with one of the forty random primers [52]. Since these nine Venezuelan Trypanosoma spp. isolates were presumed to belong to the Trypanozoon group, we used microsatellites markers and PARP-PE repeats to compare them to T. evansi and T. equiperdum reference strains. These loci exhibited limited polymorphism among seven of our isolates and the T. evansi reference strains from various hosts, horse, dog, coati, bovine, buffalo, mule and camels and geographical origins, South America, Asia and Africa. However, the microsatellite analysis clearly clustered two of the Venezuelan Trypanosoma spp. horse isolates with the T. equiperdum STIB841/OVI strain from South Africa.
Coinertia analysis of the microsatellite and PE repeats revealed that seven of the nine Venezuelan isolates closely match eleven T. evansi reference strains from around the world, while TeAp-N/D1 and TeGu-N/D1 are identical to the South African T. equiperdum STIB841/OVI strain. Principal Component Analysis (PCA) of microsatellite markers showed that this T. equiperdum strain is closely related to the T. brucei brucei Kiboko group [66]. In agreement with previous studies of Trypanozoon microsatellite loci [34], our coinertia analysis showed that BoTat-1.1 and BJ were highly heterogeneous T. equiperdum strains, genetically distant from the group that included the two Venezuelan T. equiperdum strains and STIB841/OVI. Four independent T. evansi/T. equiperdum genotypes have been recently described by Carnes et al [66]. They classified two of the reference T. equiperdum strains included in this study, BoTat-1.1 (Teq21) and STIB841/OVI, within groups 2 and 3, respectively.
Since T. equiperdum is distinguished from T. evansi by the presence of partially deleted maxicircles [14,51,67], we amplified and sequenced maxicircle genes from the TeAp-N/D1 and TeGu-N/D1 isolates and analyzed the resulting phylogenetic relationships. In agreement with previous reports, the two Venezuelan T. equiperdum strains analyzed in this study, TeAp-N/D1 and TeGu-N/ D1, have retained at least four maxicircle genes [51]. ND8 and A6 constitute complexes I and V of the oxidative phosphorylation system, and their expression is essential in the T. brucei bloodstream form [68,69], as is the expression of cytb [70]. A6 is important for maintaining the mitochondrial membrane potential and several mutations that affect its function have been described in diskinetoplastic trypanosomes [71][72][73]. Mitochondrial genes have been proposed as excellent molecular markers for discriminating closely related species [74][75][76]. The four maxicircle genes from the TeAp-N/D1 and TeGu-N/ D1 isolates revealed a close relationship to both T. brucei and T. equiperdum strains. The phylogenetic analysis of cox1 gene sequences is concordant with the four distinct groups of T. evansi and T. equiperdum strains that suggest four independent origins of these diskinetoplastic parasites [66]. Interestingly, while the microsatellite loci and Procyclin PE repeats and coinertia analyses showed identical genotypes for TeAp-N/D1 and TeGu-N/D1, the cox1 and a6 mitochondrial markers separated them into different clusters. These results confirm the first molecular report of T. equiperdum strains (TeAp-N/D1 and TeGu-N/ D1) isolated in Venezuela or in any part of Latin America.
Interestingly, T. equiperdum is not a typical American parasite but dourine, a sexually transmitted chronic disease in horses, mules and donkeys [5,17], has been n.a.
Positive for cytb, cox1, a6 and nd8 na: not amplified by PCR. The nomenclature, natural hosts and molecular characteristics of the nine Venezuelan strains included in this study are presented. The geographical origin of the strains has been previously reported [52] sporadically reported in the American continent from Canada (1921) to the USA (1934) and Mexico (1973) [77]. In Venezuela, clinical dourine was first recorded by the presence of dourine plaques in a domestic male horse, but no parasites were observed [78]. The actual geographical distribution of dourine is not known and trade restrictions appear to limit notification of the disease. The disease was widespread in the past but was eradicated from many countries of Europe after the 1940s [22,79]. Currently, the disease is endemic in parts of Africa, Asia and Russia, and dourine outbreaks or incidents have been occasionally reported in the Middle East and Europe [16,[80][81][82]. However, dourine may exist in areas where diagnostic tests are not routinely performed [3,17]. In 2011, seven dourine outbreaks were confirmed in various regions of Italy [18][19][20][21]83], all linked to the movement of breeding animals showing the characteristic plaques and lesions. The presence of T. equiperdum was confirmed by RT-PCR. T. equiperdum is morphologically indistinguishable from T. evansi and T. brucei, [5,17]. T. evansi was initially proposed to have evolved from an ancestral T. brucei when infected camels were introduced to Glossina-free areas [5,14] and is characterized by the absence of maxicircle structures [45,84]. Several years ago, Claes et al [16] proposed that T. equiperdum does not exist as a separate species and that extant strains are either T.b. equiperdum, or misidentified T.b. brucei or T. evansi strains. Based on genetic analysis, other authors have also proposed that T. evansi and T. equiperdum should be considered subspecies of T. brucei [16,17,51,65,66]. In contrast, Desquesnes et al [4] recommend keeping the current species status for T. evansi and T. equiperdum, in agreement with the rules of the international code for zoological nomenclature and based on their significant biological differences. Further genomic analysis of the two Venezuelan T. equiperdum strains should shed new light on the evolution, origin and pathogenic effects of these trypanosomes.
The evolution of T. evansi and T. equiperdum has been revisited in recent years. Lun et al [67] proposed two sequential steps in the spreading of T. brucei out of Africa, the first involved the homogenization of minicircles in the T. b. brucei bloodstream form and the loss of the ability to differentiate within the insect vector, resulting in T. b. equiperdum. Over generations, the gradual loss of maxicircles occurred due to the lack of selective pressure to preserve them, giving rise to T. b. equiperdum and eventually T. b. evansi, which lacks all maxicircles. An alternative model proposed that an ancestral trypanosome lost maxicircle genes in three [85] or four independent occasions [66], to generate stable diskinetoplastic forms, a loss that is compensated by distinct mutations on the ATP synthase γ-subunit [85]. A third, alternative hypothesis proposes that T. b. evansi and T. b. equiperdum underwent separate evolutionary processes from a T. b. brucei ancestor [17]. Our results are quite consistent with the existence of partially distinct evolutionary lineages, with the two new Venezuelan T. equiperdum strains, TeAp-N/D1 and TeGu-N/D1 corresponding to the recently proposed T. equiperdum groups 1 and 3, respectively [66].

Conclusion
The microsatellite data divided the nine Venezuelan trypanosoma isolates into two groups, one closely related to T. evansi and a second, closely related to T. Microsatellites and four maxicircle genes were used to characterize nine Trypanosoma spp. Venezuelan isolates. Seven isolates were shown to be closely related to T. evansi reference strains, while two were classified as T. equiperdum equiperdum (STIB841/OVI). The classification of TeAp-N/D1 and TeGu-N/D1 as T. equiperdum is supported by the sequences of their four maxicircle genes that are nearly identical to the orthologous T. brucei and T. equiperdum genes. Coinertia analysis of microsatellites and Procyclin PE repeats also place these two Venezuelan isolates close to the STIB841/OVI T. equiperdum strain. Phylogenetic analysis of the sequences of the cox1 gene, an exceptional discriminative molecular marker, separated the two isolates: TeAp-N/D1 clustered with the T. equiperdum STIB818 strain; while TeGu-N/D1 was grouped with the T. equiperdum STIB841/OVI strain. This constitutes the first molecular report of T. equiperdum strains isolated in Latin America (Fig. 4).

Competing interests
The authors declare that they have no competing interests.
Authors' contributions ES propagated some of the Trypanosoma spp. field isolates, performed maxicircle gene amplifications, analysis and phylogenetic trees and contributed to the manuscript draft. TP initiated the molecular characterization of the Venezuelan Trypanosome spp. isolates, including the genetic studies, participated in its coordination and contributed to the manuscript draft. GPR propagated some of the Trypanosoma spp. field isolates, performed maxicircle sequence analysis and alignments. IC propagated some of the Trypanosoma spp. field isolates, performed maxicircle sequence analysis and alignments. NB participated in the design, coordination and analysis of the microsatellite and Procyclin PE-genotyping experiments and contributed some unpublished genotyping results. PMA conceived the molecular characterization of the Venezuelan Trypanosome spp. isolates, including the genetic studies and participated in its coordination. AM participated in the design and coordination of the kDNA analysis, conceived the multivariate analysis of the microsatellite data and contributed to the manuscript draft. TB conceived and coordinated the microsatellites study and participated in its design and coordination. DB contributed to the design and coordination of the kDNA analysis. LBN designed and performed the Coinertia analysis. MG conceived the microsatellites and maxicircle gene analysis, participated in its design and coordination and contributed to the manuscript draft. All authors read and approved the final manuscript.