The kinetoplast DNA of the Australian trypanosome, Trypanosoma copemani , shares features with Trypanosoma cruzi and Trypanosoma International Journal for Parasitology

Kinetoplast DNA (kDNA) is the mitochondrial genome of trypanosomatids. It consists of a few dozen maxicircles and several thousand minicircles, all catenated topologically to form a two-dimensional DNA network. Minicircles are heterogeneous in size and sequence among species. They present one or several conserved regions that contain three highly conserved sequence blocks. CSB-1 (10 bp sequence) and CSB-2 (8 bp sequence) present lower interspecies homology, while CSB-3 (12 bp sequence) or the Universal Minicircle Sequence is conserved within most trypanosomatids. The Universal Minicircle Sequence is located at the replication origin of the minicircles, and is the binding site for the UMS binding protein, a protein involved in trypanosomatid survival and virulence. Here, we describe the structure and organisation of the kDNA of Trypanosoma copemani , a parasite that has been shown to infect mammalian cells and has been associated with the drastic decline of the endangered Australian marsupial, the woylie ( ). Deep genomic sequencing showed that T. copemani presents two classes of mini- circles that share sequence identity and organisation in the conserved sequence blocks with those of Trypanosoma cruzi and Trypanosoma lewisi . A 19,257 bp partial region of the maxicircle of T. copemani that contained the entire coding region was obtained. Comparative analysis of the T. copemani entire maxicircle coding region with the coding regions of T. cruzi and T. lewisi showed they share 71.05% and 71.28% identity, respectively. The shared features in the maxicircle/minicircle organisation and sequence between T. copemani and T. cruzi/T. lewisi suggest similarities in their process of kDNA replication, and are of signiﬁcance in understanding the evolution of Australian trypanosomes.


Introduction
The kinetoplast is a network of circular DNA (kDNA) inside a large mitochondrion. It contains the mitochondrial genome, which consists of thousands of interlocked DNA circles of two types, maxicircles and minicircles that form a complex network (Lukeš et al., 2002;Jensen and Euglund, 2012). Maxicircles comprise only a small portion of the kDNA network, with only a few dozen identical copies. Their size ranges from approximately 20-40 Kb, depending upon the species, and they encode mitochondrial gene products (Shlomai, 2004). Maxicircles are composed of two regions; the first region is a coding region that contains homologs of mitochondrial genes characteristic of other eukaryotes (Simpson et al., 1987) and a variable non-coding region, also known as the divergent region (DR). The coding region contains two rRNA genes (12S rRNA and 9S rRNA), 14 protein coding genes (ND8, ND9, ND7, COIII, Cyb, ATPase6, MURF1 now known to be ND2, ND1, COII, COI, ND4, ND3, RSP12 and ND5), four genes of unknown function (MURF2, MURF5, CR3 and CR4), and a few guide RNAs (gRNAs) (Blom et al., 1990;Kannan and Burger, 2008;Ruvalcaba-Trejo and Sturm, 2011). Minicircles comprise the major portion of the kDNA network and are present in several thousand https://doi.org/10.1016/j.ijpara.2018.02.006 0020-7519/Ó 2018 The Authors. Published by Elsevier Ltd on behalf of Australian Society for Parasitology. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). q Note: Nucleotide sequence data reported in this paper are available in the GenBank database under the accession numbers MG948554, MG948555,MG948556 and MG948557. copies that differ in size and sequence between species (Ray, 1987). They encode gRNAs that contain the genetic information for editing of mitochondrial RNA transcripts (Aphasizhev and Aphasizheva, 2014).
A considerable diversity in kDNA structure and conformation has been demonstrated within different kinetoplastids, including parasitic trypanosomatids and free-living bodonids. A classical disk-shaped kDNA network has been reported in several species, including Trypanosoma cruzi, Trypanosoma brucei, Leishmania tarentolae, and Crithidia fasciculata. Minicircles within the network are catenated, and are released from the network through decatenation by a type II DNA topoisomerase (yet to be discovered) prior to their replication (Shapiro and Englund, 1995). In contrast, the minicircles in species of more primitive kinetoplastids such as Bodo caudatus and others from the family Bodonidae are not organised in a network. Instead, they are distributed in diverse forms across the mitochondrial matrix known as poly-kDNA, pan-kDNA, mega-kDNA, and pro-kDNA (Vickerman, 1990;Lukeš et al., 2002). Correlations between these diverse patterns of kDNA organisation and genetic analysis based on nuclear rRNA genes have contributed to a better understanding of the evolution of kDNA and have facilitated the establishment of phylogenetic relationships between kinetoplastids (Lukeš et al., 2002).
A common feature within minicircles is the presence of a conserved region that contains three highly conserved sequence blocks (CSBs). CSB-1 (10 bp sequence) and CSB-2 (8 bp sequence) present lower interspecies homology, while CSB-3 (12 bp sequence) or the Universal Minicircle Sequence (UMS) is conserved within most trypanosomatids and is part of the minicircle replication origin (Ray, 1989). However, the number of conserved regions and their location in each minicircle differ among species (Ponzi et al., 1984;Sugisaki and Ray, 1987;Degrave et al., 1988). The regions of the minicircles flanked by the conserved regions are heterogeneous in sequence and have been used to investigate intraspecific variations within species (Telleria et al., 2006). Due to the minicircle abundance and heterogeneity in sequence, they have been frequently used in the development of PCR-based sensitive and specific diagnostic molecular tools (Noyes et al., 1998;Botero et al., 2010;Ceccarelli et al., 2014). CSB-3 or UMS is the specific binding site for the UMS binding protein (UMSBP) (Tzfati et al., 1992(Tzfati et al., , 1995Abeliovich et al., 1993), a protein involved in kDNA replication and segregation (Milman et al., 2007). The UMSBP of C. fasciculata has been extensively studied. It is a single-stranded sequence-specific DNA binding protein that binds the UMS (12 mer) and a hexameric sequence (in the context of a 14 mer) (Abu-Elneel et al., 1999) that are conserved at the replication origins of C. fasciculata kDNA minicircles, as well as in the minicircles of other trypanosomatid species (Ray, 1989). Recent studies using antibodies raised against C. fasciculata UMSBP revealed the presence of C. fasciculata UMSBP orthologues in other trypanosomatids such as T. cruzi and T. brucei (Milman et al., 2007), as well as Leishmania donovani (Singh et al., 2016). The importance of UMSBP in the survival of trypanosomatids was demonstrated using RNA interference experiments. It was shown that simultaneous knockdown of the two UMSBP orthologous genes in T. brucei not only affects the initiation of minicircle replication, but also inhibits segregation of the daughter networks and blocks nuclear division (Milman et al., 2007). Moreover, a recent study showed that the deletion of L. donovani UMSBP induced kDNA loss, apoptosis, and regulated the virulence of the parasite in macrophages and mice (Singh et al., 2016).
Since the 1950s several species of Australian trypanosomes have been described from wildlife. These include Trypanosoma thylacis, Trypanosoma irwini, Trypanosoma vegrandis, Trypanosoma gilletti, Trypanosoma binneyi, Trypanosoma copemani, Trypanosoma noyesi, and Trypanosoma teixeirae (Mackerras, 1959;Noyes et al., 1998;Smith et al., 2008;Austen et al., 2009;Averis et al., 2009;McInnes et al., 2009McInnes et al., , 2011Paparini et al., 2011;Botero et al., 2013Botero et al., , 2016bThompson et al., 2013;Austen et al., 2011;Barbosa et al., 2016). Interestingly, T. copemani has been shown to invade cells, and has been associated with pathological changes in wildlife hosts that are similar to those seen with the human pathogen, T. cruzi (Botero et al., , 2016a. Moreover, T. copemani has been found to infect numerous endangered and threatened species of wildlife, and has been associated with the drastic decline of the critically endangered woylie (Bettongia penicillata) McInnes et al., 2011;Botero et al., 2013Botero et al., , 2016a. Despite the importance of the kDNA in trypanosomatid survival and regulation of virulence (Milman et al., 2007;Singh et al., 2016), there is a complete lack of knowledge about the kinetoplast and the organisation of the kDNA within this organelle in Australian trypanosomes.
The aim of this study was to investigate the ultrastructure and organisation of kDNA in the Australian wildlife trypanosome, T. copemani. In particular, the organisation of the maxicircles and minicircles, and the presence of the UMS element and the UMSBP were investigated.

Parasites
Parasites were grown under optimal conditions for each individual species (Botero et al., 2016a). Epimastigotes of two genotypes of T. copemani (G1 and G2) were grown in 75 ml flasks containing Grace's medium with 10% Fetal Calf Serum (FCS) and penicillin-streptomycin. Trypanosoma cruzi epimastigotes were grown in 75 ml flasks containing RPMI 1640 medium with 10% FCS and penicillin-streptomycin. All strains were grown at 28°C.

Isolation of the kDNA network
Ethidium bromide (EtBr)-cesium chloride (CsCl) stepwise gradients were used to isolate the kDNA of T. copemani (Saucier et al., 1981;Hajduk et al., 1984). Approximately 1 Â 10 10 T. copemani G1 and G2 cells from the stationary phase were harvested by centrifugation at 16,000g at 4°C, for 5 min and then washed once with PBS and once with NET100 buffer (10 mM Tris-Cl pH 8, 100 mM NaCl, 100 mM EDTA, pH 8). The pellet, containing trypanosomes, was submitted to lysis by resuspending in NET100 buffer and 1 mg/ml of proteinase K. Sodium Sarkosinate (Sarcosyl) was added slowly (3% final concentration) and the suspension was gently mixed and incubated for 30 min at 4°C. Then, 24 ml of CsCl (1.386 density) and 4 ml of CsCl (1.750 density -mixed previously with EtBr to final concentration of 20 lg/ml) were added to an ultra-clear SW28 rotor tube. Ten milliliters of sample lysate were then added to the tube and centrifuged in a SW28 rotor at 21,410g at 4°C, for 15 min. The upper phase of the suspension was discarded and the middle phase (fluorescent band under ultraviolet light) was collected. The EtBr was removed by several extractions with 0.1Â SSC-saturated isoamyl alcohol, and then dialyzed overnight against Tris EDTA buffer (TE) at pH 8.0 and 4°C. The kDNA preparation was treated again with proteinase K (as above), followed by phenol-chloroform extraction and ethanol precipitation.

Deep sequencing, assembly, and annotation of minicircles and maxicircle
Approximately 100 ll of 100 ng/ml of genomic DNA from T. copemani G1 and G2 cells grown in culture was isolated using the QIAamp blood and tissue DNA MiniKit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The quality of DNA was confirmed using Nanodrop; samples tested negative for mycoplasma via PCR, and DNA was run on a 1% agarose gel to investigate signs of fragmentation prior to sequencing. Deep sequencing was performed at the Australian Genome Research Facility (AGRF, Melbourne, Australia) using an Illumina NextSeq500 platform with 300 cycles and Mid output kit. A paired end library preparation for T. copemani G1, and a mate pair preparation (gel free) for T. copemani G2 yielded fragments of 150 bp long reads. Data were generated with the Illumina bcl2fastq pipeline version 2.19 and the integrity of the files was established using TestFiles. exe on a windows platform before further analysis. Sequences were quality filtered and merged using CLC genomics v.10. Both libraries were de novo assembled into contigs using Geneious and Spades assemblers on Geneious 10.2. Maxicircle and minicircle contigs were identified by alignment with other Trypanosoma spp. on GenBank using BLAST. All contigs were aligned using MUSCLE (Edgar. 2004) and manually annotated by comparison with other sequences available on GenBank including maxicircle sequences from T. cruzi (GenBank: DQ343645), Trypanosoma lewisi (GenBank: KR072974), Trypanosoma cruzi marinkellei (GenBank: KC427240), and Trypanosoma brucei brucei (GenBank: M94286), and minicircle sequences from T. cruzi (GenBank: TCU07845), and T. lewisi (GenBank: M17995, M17996). Two full minicircle sequences and a partial sequence of the maxicircle of T. copemani G1 were deposited in GenBank under the accession numbers MG948555, MG948556, and MG948557, respectively. A partial sequence of the minicircles of T. copemani G2 was identified and was deposited in GenBank under the accession number MG948554.

Western blot analysis
Whole cell protein extracts of T. copemani G1, T. copemani G2, and T. cruzi were obtained by centrifugation of 1 Â 10 6 epimastigotes of each parasite at 5000g for 5 min, at room temperature. The pellet was washed twice with PBS and then resuspended in 40 ll of double-distilled water. Then, 10 ll of 10% SDS were added and samples were sonicated for 5 min. After sonication, cell lysates were solubilized in cracking buffer containing final concentrations of 50 mM Tris-HCl, pH 6.8, 4% SDS, 3.5% b-mercaptoethanol, 10% glycerol and 10 mM EDTA. The solution was then centrifuged at 10,000g at 4°C, for 30 min in order to remove cell debris. The supernatant was recovered and the protein concentration was determined using the Direct Detect Assay-free cards (EMD Millipore corporation, United States). Protein extract (30 and 40 lg) was heated at 70°C for 10 min and loaded onto a NuPag10 4-12% Bis-Tris gel (Invitrogen, United States). Recombinant C. fasciculata UMSBP was used as a positive control of hybridization. Protein bands on the gel were transferred onto nitrocellulose membranes using the trans-blot turbo-transfer system (Bio-Rad, United States). Then, the membrane was blocked by incubation in 5% skim dry milk (Difco, United States) diluted in Tris buffered saline with tween (50 mM Tris-Cl, 0.1% Tween-20) (TBST) pH 7.5 for 30 min with constant shaking, and probed with a 1:4000 dilution of C. fasciculata anti-UMSBP antibodies overnight at 4°C. The membrane was then washed three times with TBST and then incubated with a 1:10,000 dilution of enhanced chemiluminescent (ECL) peroxidase labelled conjugated goat anti-rabbit secondary antibody (Jackson Immuno Research Laboratories, Inc. United Sates) for 2 h, followed by ECL detection as recommended by the manufacturer (Amersham Pharmacia Biotech, United Kingdom). The preparation of C. fasciculata recombinant UMSBP was performed as reported previously (Tzfati et al., 1995;Sela and Shlomai, 2009).

DNA topoisomerase II assay
Decatenation of T. copemani kDNA was conducted in a 10 ll reaction mixture containing four units of human topoisomerase II a (TopoGen Inc, Port Orange, Florida, USA), 50 mM Tris-Cl pH 8, 120 mM KCl, 10 mM MgCl 2 , 0.5 mM of DTT, 0.5 mM of ATP and 30 lg/ ml of BSA (Topo II reaction buffer TG4040), and 0.1 lg of kDNA. Reactions were incubated for 60 min at 37°C, and then stopped by the addition of 0.1 vol of stopping buffer (5% Sarcosyl, 0.025% bromophenol blue, 50% glycerol). Samples were loaded onto 1% agarose gel containing 1 lg/ml of ethidium bromide (Life Technologies, United States) and electrophoresed at 100 V for 30 min. Topo II decatenated C. fasciculata kDNA and C. fasciculata linear kDNA markers (TopoGen Inc, United States) were used as controls in the assay.

Transmission electron microscopy
Trypanosoma copemani G1 and G2 epimastigotes were fixed in a 1:1 mixture of 5% glutaraldehyde in 0.01 M PBS:cell culture medium (pH 7.2) resulting in a working glutaraldehyde concentration of 2.5%. All subsequent processing was performed in a PELCO Biowave microwave, where samples were post-fixed in 1% OsO 4 in PBS, followed by progressive dehydration in ethanol/acetone, before being infiltrated and embedded in Procure-Araldite epoxy resin. Sections (120 nm thick) were cut with a diamond knife and mounted on copper grids. Digital images were collected from unstained sections at 120 kV on a JEOL 2100 TEM (Japan) fitted with a Gatan ORIUS1000 camera. The thickness and diameter of the kinetoplast were measured in trypanosome sections where the basal body of the flagellum was seen and where most of the DNA fibres within the kinetoplast were continuously distributed from side to side (indicating that the kinetoplast disk was cut parallel to its long axis and through its central region, respectively).

Trypanosoma copemani minicircle sequence and organisation
After removal of low quality reads from raw data, we obtained 7,000,006 clean reads from T. copemani G1 and 1,085,177 reads for T. copemani G2. These data were finally assembled into 367,392 and 310,743 contigs, respectively, using the Geneious and Spades assemblers on Geneious 10.2. Two different fully circularised minicircle contigs from T. copemani G1 were identified by BLAST search against sequences of the minicircles of other Trypanosoma spp. deposited in GenBank. Both T. copemani G1 minicircle classes presented a size of 2048 bp. The first class (which we named G1M1) contained two conserved regions present as direct repeats located approximately 180°apart, and two hypervariable regions flanked by the two conserved regions ( Fig. 2A). The second class (which we named G1M2) contained four conserved regions present as direct repeats located approximately 90°apart, and four hypervariable regions flanked by the four conserved regions (Fig. 2B). Several partial sequences of the minicircles of T. copemani G2 were also found, and all contained one conserved region.
Each conserved region in both T. copemani G1 and G2 minicircles (approximately 130 bp long) contained the three CSBs previously reported in most trypanosomatids. Although all CSBs were the same in the two classes of minicircles from T. copemani G1, CSB-1 and CSB-2 were different between both T. copemani G1 and G2. Alignments of T. copemani G1 (G1M1 and G1M2) and T. copemani G2 conserved regions with the minicircle conserved regions of T. cruzi (GenBank: TCU07845), and T. lewisi (GenBank: M17996) revealed similarities with both species (Table 1). The first sequence block (CSB-1), which has been shown to differ between species, was the same in T. cruzi and T. copemani G2. The CSB-1 of both classes of minicircles of T. copemani G1 differed in one nucleotide from those of T. copemani G2/T. cruzi. The second sequence block (CSB-2), which has also been shown to differ between species, was the same in T. cruzi and T. copemani G1, and in T. copemani G2 and T. lewisi, with only one nucleotide difference between T. copemani G1/T. cruzi and T. copemani G2/T. lewisi. The third sequence block CSB-3 or UMS (minicircle origin of replication), was exactly the same as the one reported in several species of trypanosomatids (Table 1).

Trypanosoma copemani maxicircle partial sequence
A long partial sequence from the maxicircle of T. copemani G1 (19,257 bp), which included a short fragment of the divergent region and all the coding region beginning at the 12S rRNA gene, and ending after the 3 0 end of the ND5 gene, was also found. The data obtained indicated that the T. copemani maxicircle encodes 20 firmly clustered genes, with the same gene order as T. cruzi, T. lewisi, T. cruzi marinkellei, and T. brucei maxicircles (Table 2).
Comparative analysis of the full coding region of the maxicircle of T. copemani G1 with that of T. cruzi and T. lewisi showed they share 70.8% and 71.2% identity, respectively. Three T. copemani G1 genes exhibited the same length as those of T. lewisi (MURF2, ND7, and ND5), and two as those of T. cruzi (12S rRNA, and CR4). However, genes CytB, COII, COI, and ND4 exhibited the same length in all three species ( Table 2). Most of these sequences are from genes where the transcripts either undergo minor or do not undergo RNA editing in T. cruzi and T. lewisi. Genes with transcripts that are subject to extensive editing generally exhibited a lower pairwise percentage of identity (Table 3).

Phylogenetic analysis based on the maxicircle coding region
Phylogenetic analysis was performed using the entire coding region (20 genes) from T. copemani G1 and the maxicircle coding region of 10 species of trypanosomatids that are available in Gen-Bank. Both ML and Mr Bayes phylogenies showed that T. copemani G1 clusters with T. lewisi and with trypanosomes within the T. cruzi clade (including T. rangeli and T. cruzi marinkellei) with 100% support (Fig. 3). However, T. copemani G1 is more closely related to T. lewisi than to T. cruzi and allied species. The tree topology obtained in this study supports previous phylogenetic analyses using either 18S rDNA and glycosomal glyceraldehyde 3phosphate dehydrogenase (gGAPDH) genes (McInnes et al., 2011).

Trypanosoma copemani contains the UMSBP
Considering that the UMS element is present in C. fasciculata minicircles, and that it was also found in T. copemani minicircles, we investigated the presence of a UMSBP in T. copemani using antibodies raised against C. fasciculata UMSBP (Tzfati et al., 1995). These anti CfUMSBP antibodies have been previously shown to detect UMSBP in other species such as T. brucei (Milman et al., 2007), T. cruzi (Coelho et al., 2003), and L. donovani (Singh et al., 2016). Western blot analysis, using anti-CfUMSBP antibodies, detected two peptides of approximately 16.4 kDa and 25.7 kDa in    (1986) D1-2, average distance (in bp) between CSB-1 and CSB-2; D2-3: average distance (in bp) between CSB-2 and CSB-3 or UMS. a Nucleotides in larger font represent residues at the indicated position which differ from those of T. cruzi. b GenBank accession numbers. both T. copemani G1 and G2 protein extracts (Fig. 4). When T. cruzi protein extracts were used, the antibody recognised two peptides of approximately 15.6 kDa and 21.6 kDa (Fig. 4). When C. fasciculata recombinant UMSBP was used, the antibody recognised a peptide of approximately 15.1 kDa and a series of higher UMSBP oligomeric forms, which could be the product of partial and not full reduction of the C. fasciculata recombinant UMSBP.

Trypanosoma copemani kDNA network is a topological catenane
In attempts to investigate whether the minicircles of T. copemani were catenated into the kDNA network, the enzyme DNA   topoisomerase II was used. Incubation of purified kDNA with DNA topoisomerase II, under decatenation conditions, revealed that the kDNA minicircles of both T. copemani G1 and G2 are topologically interlocked to form a catenane network. Minicircles, which were decatenated by topoisomerase II, were analysed by electrophoresis in agarose gels. Untreated kDNA networks failed to enter the agarose gel, while partially decatenated networks of T. copemani yielded monomeric minicircles bands, corresponding to nicked, linear and covalently closed minicircles (Fig. 5).

Discussion
The organisation of minicircles and maxicircles within kDNA differs within species and these differences have been used to cluster them into various groups (Lukeš et al., 2002). The most complex type of kDNA organisation is the kDNA network structure present in pathogenic trypanosomatids such as T. cruzi, T. brucei and Leishmania, and in the insect trypanosomatid C. fasciculata -all late diverging kinetoplastids (Lukeš et al., 2002). The kDNA network is condensed in the mitochondrial matrix into a disk-shaped structure, where minicircles are topologically relaxed, and each is catenated on average to three neighbours (Chen et al., 1995), and aligned side by side (Rauch et al., 1993). Pro-kDNA is the second most organised kDNA structure and is present in the late diverging free-living bodonid Bodo saltans. The majority of minicircles in pro-kDNA are covalently closed, topologically relaxed and organised in a single bundle-like structure with only a few catenanes (Blom et al., 1990). In contrast, all other types of kDNA structure, poly-kDNA, pan-kDNA and mega-kDNA, are present in early diverging kinetoplastids such as Dimastigella trypaniformis, Bodo caudatus, Cryptobia helicis and Trypanoplasma borreli (Lukeš et al., 2002). The kDNA fills most of the kinetoplast as with pan-kDNA, or fills various separate foci throughout the kinetoplast as with poly-kDNA, all of which lack the highly ordered kDNA packaging seen in trypanosomatids (Lukeš et al., 2002). Interestingly, the kinetoplast of T. copemani presented a classical kDNA network structure, similar to that seen in late diverging kinetoplastids such as T. cruzi and T. brucei. Furthermore, the products of decatenation by topoisomerase II indicated that the minicircles of T. copemani are catenated, confirming that the arrangement of kDNA is in the form of a catenated network as seen in T. cruzi and T. brucei. The shared features in the kinetoplast ultrastructure between late diverging kinetoplastids and T. copemani will be of value in understanding the evolution of Australian trypanosomes, although further studies are needed to better understand these relationships.
Not only does the kDNA organisation differ between species, several studies have also reported differences in the size of minicircles, and in the number, position and sequence of its conserved regions. The size of the minicircles of T. lewisi, for example, are approximately 1 kb and contain two conserved regions located 180°apart (Ponzi et al., 1984). In contrast, T. cruzi minicircles are approximately 1.4 kb in size, and contain four conserved regions located at 0°, 90°, 180°and 270°apart (Degrave et al., 1988). Furthermore, T. brucei and L. tarentolae minicircles are approximately 1 kb in size and contain only one conserved region (Kidane et al., 1984;Jasmer and Stuart, 1986;Hines and Ray, 2011). Additionally, a species of Trypanosoma from South America, T. rangeli, presents two classes of minicircles with different size and molecular organisation. The first class presents a single conserved region and a size of 1764 bp, while the second class presents two conserved regions located as direct repeats 180°apart and a size of 1587 bp (Vallejo et al., 1994). The results of this study have shown that T. copemani G1 presents two distinct classes of minicircles; the first one contains two conserved regions located 180°apart, similar to T. lewisi minicircles and T. rangeli second class minicircles; and the second one presents four conserved regions located 90°apart, similar to T. cruzi minicircles. Unfortunately, only partial sequences from T. copemani G2 minicircles were obtained but all contained the full conserved region including the three conserved sequence blocks CSB-1, CSB-2 and CSB-3 (origin of replication). Interestingly, when the full conserved regions of the minicircles of T. copemani G1 and G2 were individually compared with other trypanosomatids in GenBank, the most similar sequences were T. lewisi and T. cruzi with 81.0% and 78.5% identity, respectively, with T. copemani G1, and 76.9% and 80.8% identity, respectively, with T. copemani G2. Trypanosoma copemani G1 and G2 have previously been shown to exhibit considerable differences in their biological behaviour such as growth requirements, susceptibility to drugs, and capability to infect cells in vitro (Botero et al., 2016a(Botero et al., , 2017. Furthermore, T. copemani G2 has been found in blood and tissues of marsupials with pathological changes similar to those seen during T. cruzi infections, while T. copemani G1 has only been found in blood . Adding to the differences between T. copemani G1 and G2, this study found that the conserved sequence blocks (CBS-1 and CBS-2), which have been shown to be speciesspecific, were different in both strains. Taking into account the differences in the biology, lifestyle and sequence of the minicircle conserved region, T. copemani G1 and G2 could actually be different species. Phylogenetic studies based on the 18S rDNA and gGAPDH genes have also shown that although both T. copemani G1 and G2 are closely related, T. copemani G1 forms a subclade with other genotypes from Australian wildlife such as Charlton, AAP, AAI, H26, Q3, Q10, GP63 and GP94, while T. copemani G2 sits by itself in a separate subclade (Noyes et al., 1999;Austen et al., 2009;McInnes et al., 2011;Botero et al., 2013). Identifying the full coding region of the maxicircles of T. copemani G2 could further assist in separating T. copemani G1 and G2 as separate species.
When the aligned minicircle conserved regions of T. copemani G1, G2, T. cruzi, and T. lewisi were analysed more carefully, similarities in the sequence and spacing between the three CSBs were found. These three CSBs have been reported previously in other species of trypanosomatids including C. fasciculata, L. tarentolae, T. brucei, Trypanosoma congolense and T. equiperdum (Barrois et al., 1981;Kidane et al., 1984;Ponzi et al., 1984;Jasmer and Stuart, 1986;Nasir et al., 1987;Sugisaki and Ray, 1987). However, differences in the sequence of both first and second blocks (CSB-1 and CSB-2) between different species of trypanosomatids have been reported (Ray, 1989). Interestingly, T. copemani G1 CSB-2 was identical to the CSB-2 of T. cruzi minicircles, and the CSB-1 differed only in one nucleotide, while for T. copemani G2, the CSB-2 was identical to the CSB-2 of T. lewisi, and the CSB-1 was identical to the CSB-1 of T. cruzi. As expected, both T. copemani G1 and G2 presented the conserved CSB-3 or UMS element, which has been shown to be conserved in all trypanosomatids studied. The similarities in the number, position, and sequence of the minicircle conserved regions of T. copemani G1, T. copemani G2, T. cruzi, and T. lewisi, support the theory that the UMS and the CSB-1 are involved in the initiation of minicircle L-strand and H-strand replication (reviewed in Shapiro and Englund, 1995). This suggests that T. copemani may have a common mechanism of minicircle replication with these two Trypanosoma spp. However, the significance of these genetic similarities needs to be further investigated.
In addition to the similarities in the kDNA of T. copemani with those of T. cruzi and T. lewisi, comparative analysis of the maxicircle coding region of the three species revealed a high level of sequence similarity. Our phylogenetic analysis also confirmed the wellestablished close relatedness of T. copemani with T. cruzi and T. lewisi, with T. lewisi being more closely related. The T. copemani G1 maxicircle coding region contained the same genes and gene order seen in other trypanosomatids such as T. cruzi, T. lewisi, T. brucei, T. rangeli and Leishmania tarentolae (El-Sayed et al., 2005). Most of these maxicircle genes encode proteins that are involved in oxidative metabolism, and which are necessary in the vectorhost transmission cycle. Trypanosoma lewisi and T. cruzi are transmitted by fleas and reduviid bugs, respectively. Both parasites belong to the stercorarian group in which metacyclic trypomastigotes are located in the epithelium of the rectum of the insect, and transmission occurs when the infective forms are released in faeces near a bite site after feeding on a mammalian host (Hoare, 1972). Interestingly, T. copemani has been found to belong to the stercorarian group with the tick Ixodes australiensis acting as a vector (Austen et al., 2011). Considering that the kinetoplast plays an important role in the life cycle of trypanosomes, it could be possible that T. copemani shares more life cycle pathways with those of T. cruzi and T. lewisi, but this need to be further investigated. Trypanosoma copemani G2, for example, has been shown to be able to infect cells in vitro such as T. cruzi (Botero et al., , 2016a. Some of the maxicircle genes also encode proteins that are involved in virulence and pathogenicity. Interestingly, T. copemani has been shown to be pathogenic as well as T. cruzi, and T. lewisi. Trypanosoma cruzi is the agent of Chagas disease in humans, while T. lewisi has been implicated in the extinction of two native rats in Australia (Wyatt et al., 2008), and has recently been shown to atypically infect humans (Truc et al., 2013;de Sousa, 2014). Interestingly, T. copemani has been implicated in the drastic decline of the marsupial Bettongia penicillata or woylie. This close genetic relationship in the maxicircle coding region between T. copemani, T. lewisi and T. cruzi warrants further investigations, especially considering the mounting evidence that the two first mentioned trypanosomes have been implicated in wildlife extinctions or declines.
The UMSBP of C. fasciculata has been extensively studied. This protein is a single-stranded sequence-specific DNA binding protein that binds the UMS elements in kDNA minicircles (Tzfati et al., 1992(Tzfati et al., , 1995. In T. brucei, the two UMSBP orthologues have been suggested to act in minicircle replication initiation and kDNA segregation (Milman et al., 2007). Crithidia fasciculata UMSBP is a protein of approximately 13.7 kDa (Tzfati et al., 1992(Tzfati et al., , 1995Onn et al., 2004) and the mass of the recombinant C. fasciculata UMSBP tagged protein used here was approximately 1 kDa larger. It has been shown that antibodies raised against the UMSBP of C. fasciculata can cross-react with UMSBPs of other trypanosomatids such as T. cruzi, L. donovani, and T. brucei (Coelho et al., 2003;Milman et al., 2007;Singh et al., 2016). Trypanosoma cruzi UMSBP was previously shown to be approximately 14 kDa and to bind the dodecamer UMS element (Coelho et al., 2003). The present study confirmed these previous results and showed that antibodies raised against C. fasciculata UMSBP also recognise T. cruzi UMSBP. However, there was a difference in the size of the protein (approximately 15.1 kDa in this study). In addition, this study found a second T. cruzi protein of 21.6 kDa that reacted with the anti UMSBP antibodies, which may be the product of oligomerization, which was seen with C. fasciculata recombinant UMSBP (used as a control). This study reports for the first known time the presence of the UMSBP in T. copemani. Antibodies raised against the UMSBP of C. fasciculata detected a UMS protein in T. copemani cell extracts of approximately 16.4 kDa, and a second protein of approximately 27.7 kDa that could be the product of oligomerization or a UMSBP orthologue. Many other trypanosomatids such as Leishmania major, Leishmania infantum, Leishmania braziliensis, T. brucei, T. vivax, and T. congolense have been shown to contain two UMSBP orthologous genes (Milman et al., 2007). Our suggestion that the 27.7 kDa protein could be a UMS orthologue in T. copemani requires further investigation. The identification of T. copemani UMSBP, a protein that has been involved in the process of kDNA replication and more recently has been implicated in the regulation of virulence in other trypanosomatids, is the first step towards understanding the mechanism of kDNA replication and virulence in this parasite.
The current study has not only demonstrated strong similarities in the organisation and structure of kDNA minicircles and maxicircles between T. copemani and late emerging trypanosomatids such as T. lewisi and T. cruzi, but has also provided preliminary information and a foundation to better understanding the role of kDNA in the evolution of diversity, transmission and pathogenicity within kinetoplastids.