INTRODUCTION
The genus Trypanosoma comprises unicellular flagellates that are parasites of all vertebrate classes. The vectors can be haematophagous arthropods for mammalian, avian, reptilian as well as amphibian trypanosomes, whereas fish, certain amphibian and reptilian trypanosomes are transmitted by leeches (Haag et al. Reference Haag, O'Huigin and Overath1998). Salivarian trypanosomes are known to be transmitted by biting flies of the genus Glossina commonly called tsetse flies (Donelson, Reference Donelson2003; Ravel et al. Reference Ravel, Grebaut, Cuisance and Guny2003; Stevens and Brisse, Reference Stevens, Brisse, Maudlin, Holmes and Miles2004) with the exception of Trypanosoma evansi and T. equiperdum which are transmitted mechanically by the Stomoxys and Tabanid flies and by copulation, respectively (Brun et al. Reference Brun, Hecker and Lun1998; Gibson, Reference Gibson2003). Furthermore, the stercorarian trypanosomes such as T. cruzi, T. lewisi and T. rangeli are transmitted by bugs and fleas through contaminated faeces (Stevens and Brisse, Reference Stevens, Brisse, Maudlin, Holmes and Miles2004).
The current study reports on the unknown trypanosome species that has been isolated from naturally infected Haemaphysalis hystricis ticks isolated in Kagoshima Prefecture, Japan. The H. hystricis tick is a vector of Ehrlichia and Rickettsia sp. that cause spotted fever group (SFG) ehrlichiosis and rickettsiosis, respectively (Mahara, Reference Mahara1997; Parola et al. Reference Parola, Cornet, Sanogo, Miller, Van Thien, Gonzalez, Raoult, Telford and Wongsrichanalai2003). We refer to this unknown trypanosome as Trypanosoma KG1 isolate. With the aim of identifying the taxonomic status of this unknown trypanosome species, a series of experiments was conducted including the establishment of an in vitro culture system for Trypanosoma KG1 isolate (KG1), an attempt to determine the laboratory mammalian host, experimental infection of other tick species and the development of PCR and loop-mediated isothermal amplification (LAMP) for specific detection of this trypanosome. Furthermore, in this study we cloned and sequenced the internal transcribed spacer 1 (ITS1), 18S rRNA, large subunit 28S rRNA and kinetoplast DNA (kDNA) genes of Trypanosoma KG1 isolate.
MATERIALS AND METHODS
Isolation of parasites from ticks
Three male and 3 female adult Haemaphysalis hystricis ticks were collected from vegetation at Uchinoura-cho on the Osumi Peninsula in Kagoshima prefecture, situated on the Southernmost part of Kyusyu Island, Japan. The average temperature, relative humidity and rainfall in Kagoshima prefecture are 15–23°C, 60–75 RH, and 2200–2900 mm, respectively. Vegetation is a typical subtropical forest, in which Cinnamomum camphora and Cycas revoluta plantations are dominant. Midguts of the ticks were removed aseptically and suspended as pooled samples in the sucrose-potassium-glutamate medium. A suspension of the pooled midgut contents was then inoculated into the monolayer of L929 cells cultivated with Eagle's MEM (Nissui Pharmaceutical Co. Ltd, Tokyo) supplemented with 2% fetal bovine serum (FBS) and incubated at 33°C (Honda et al. Reference Honda, Fujita, Kuramoto, Watanabe and Takada2006). However, in the current study the parasites were propagated at 37°C with the HEK 293T cell line as feeder cells. The KG1 cultures were maintained in Medium 199 (Sigma® Aldrich, U.K.) supplemented with 2% FBS and Penicillin-Streptomycin (100 U-100 μg/ml) by replacing the culture medium every second day.
Tick infection
A volume of 12 μl of KG1 culture supernatant (containing approximately 103 parasites) was experimentally injected into the haemocoel of 12 Ornithodoros moubata ticks. Another set of ticks was made to feed on the culture supernatant containing 1×106 parasites/ml through an artificial membrane as described previously (Waladde et al. Reference Waladde, Young and Morzaria1996). However, in this study we used laboratory film (Parafilm® M, Pechiney Plastic Packaging, Chigago, IL) as an artificial feeding membrane. The presence of the parasites was examined by Giemsa-stained smears made from the haemolymph, salivary glands, and midgut every 5 days for a period of 30 days. All ticks were kept at 25°C and 50–60% relative humidity in continuous darkness throughout the experiment.
Infection of laboratory animals and parasite detection
Five female ICR mice, BALB/C mice, and SD rats aged 6 weeks were inoculated intraperitoneally with 1 ml of culture supernatant containing approximately 1×106Trypanosoma KG1 parasites, and corresponding numbers of each animal were kept as uninfected control groups. One Japanese white rabbit was also inoculated with 1×106 parasites, and another was kept as control. Furthermore, a splenectomized rat and sheep were also inoculated with 1×106 and 2×106Trypanosoma KG1, respectively. Corresponding numbers of each animal species were kept as uninfected controls. All the controls of the above-mentioned mammalian species were inoculated with 1 ml of the new culture medium (Medium 199) (Sigma® Aldrich, UK). The inoculated animals were bled every 3 days for a period of 30 days and parasitological examinations (microscopy of wet smears, Giemsa-stained thin blood smears, and buffy coat preparations), PCR and LAMP were performed in order to detect the parasite from blood samples. All animals were euthanized 30 days post-infection. In order to clarify infectivity of the KG1 parasite in the animal species tested, total DNA of spleen, heart, liver, kidneys, and lymph nodes was examined by PCR and LAMP.
DNA extraction
Total DNA of the KG1 isolate, blood, and visceral organs was extracted as previously described (Sambrook and Russell, Reference Sambrook, Russell, Sambrook and Russell2001). Briefly, the extraction buffer (10 mm Tris-HCl, pH 8·0, 10 mm EDTA, 1% sodium dodecyl sulphate) and 100 μg/ml proteinase K were added to the samples and incubated overnight at 55°C. DNA was extracted with phenol-chloroform isoamyl alcohol (25:24:1) and precipitated with isopropanol. DNA was dissolved in 250 μl of TE (10 mm Tris-HCl, 1 mm EDTA, pH 8·0) buffer.
PCR
In the initial experiments, several primers of commonly known trypanosomes, including the Trypanosoma (Trypanozoon) group, T. evansi, T. rangeli, Kinetoplastida (Desquesnes and Davila, Reference Desquesnes and Davila2002), and the T. lewisi primers (Desquesnes et al. Reference Desquesnes and Davila2002), were used in our attempts to amplify DNA of Trypanosoma KG1 isolate. We also used primers for eukaryotic 18S rRNA (Countway et al. Reference Countway, Cast, Savai and Caron2005) and primers designed from the 18S rRNA of Trypanosoma pestanai (Accession no: AJ009159) for amplification of KG1 DNA and 28S rRNA LSU (Accession no: X14553) of T. brucei group (Table 1 and Fig. 1). The PCR reaction mixture contained 10×PCR buffer (100 mm Tris-HCl, pH 8·3, 500 mm KCl, 15 mm MgCl2, 0·01% (w/v) gelatin), 2 mm of each dNTP, 5 pmol of each primer, and 0·5 U of AmpliTaq Gold DNA polymerase (Applied Biosystems, Japan) in a final volume of 50 μl. The reaction mixture was heated at 94°C for 10 min and subjected to 35 cycles at 94°C for 45 sec, 1 min at 55°C, and 1 min at 72°C, with a final extension at 72°C for 7 min.
Schematic diagram of primer positions for 18S rRNA. The entire sequence of 18S rRNA of Trypanosoma KG1 is indicated by a solid line. The sequence used for phylogenetic tree construction is shown by an open box with its length and Accession number. Each primer location is indicated by open arrow and primer name.
PCR primers used for amplification of Trypanosoma KG1 genomic DNA
* Primer position on the nucleotide sequence of Trypanosoma KG1 18S rRNA (Accession no.: AB281091).
** ND, Not determined.
LAMP
The following LAMP primer set targeting ITS-1 gene of the Trypanosoma KG1 isolate was designed using the Primer Explorer V2 software program (Fujitsu, Japan): FIP: 5′-GAT TCC AGC TGC AGG TTC ACC AAT AGT AGG GAA GCA AAG TC-3′, BIP: 5′-GCA TGT ATG TAT GTG TAG TAT GCG TTA GAA GCT GTT GCT TCA TAC C-3′, F3: 5′-ACC GAA AGT TCA CCG TAT T-3′, and B3: 5′-TTG TGT GCG AAG AGA ACA-3′. The reaction was carried out as previously reported (Notomi et al. Reference Notomi, Okayama, Masubuchi, Yonekawa, Watanabe, Amino and Hase2000), briefly, in a final volume of 25 μl reaction mixture containing 12·5 μl of ×2 LAMP reaction buffer (40 mm Tris-HCl, pH 8·8, 20 mm KCl, 16 mm MgSO4, 20 mm (NH4)2SO4, 0·2% Tween 20, 1·6 m Betaine, 2·8 mm of each dNTPs), 2 μl template DNA and 1 μl (8 units) Bst DNA polymerase (New England BioLabs Inc., USA), 0·9 μl of primer mix (FIP and BIP at 40 pmol each, and F3 and B3 at 5 pmol each), and 8·6 μl of distilled water. The reaction mixture was incubated at 63°C for 1 h followed by 80°C for 2 min for termination of the reaction. The F3 and B3 LAMP primers were also used for PCR with the same PCR conditions as mentioned above except for annealing temperature that was modified to 57°C. Both the PCR and LAMP products were electrophoresed in a 1·5% Tris-acetic acid-EDTA (TAE) agarose gel and stained with ethidium bromide for visualization.
Gene cloning, sequencing, and analysis
The PCR products were purified using the QIAquick gel extraction Kit (Qiagen, USA) and cloned to pT7Blue-T vector (Novagen Inc., Germany) with DNA ligation kit Ver 2.1 (Takara Bio Inc., Japan). The plasmid was cut with Bam HI and Xba I restriction enzymes, and the nucleic acid sequence was determined with the BigDye terminator cycle sequencing kit (Applied Biosystems, Japan). In order to analyse similarities between the nucleic acid sequences from KG1 and the known sequences of other trypanosomes, the sequences of KG1 were subjected to NCBI BLASTn search (www.ncbi.nlm.nih.gov/BLAST). Furthermore the KG1 18S rRNA sequence was aligned with other selected stercorarian trypanosome genes using ClustalW program, and a phylogenetic tree (see Supplementary data, Figs S1 and S2 in Online version) was constructed by the bootstrapped maximum likelihood method using the PHYLIP 3.6 package program downloaded from the University of Washington website (http://evolution.gs.washington.edu/phylip.html). The Seqboot and Consense programs were used to statistically assess the strength of the tree using bootstrap resampling. The model used for the maximum likelihood classification is one that allows for inequalities of equilibrium base composition and for inequalities of the rate of transitions and transversions. It is related to the model reported by Felsenstein (Reference Felsenstein1981) but generalizes it to allow for unequal rates of transitions and transversions (Felsenstein and Churchill, Reference Felsenstein and Churchill1996).
RESULTS
Identification of the taxonomic status of the Trypanosoma KG1 isolate
Giemsa-stained smears of KG1 from in vitro cultures clearly showed common morphological features of trypanosomes such as nucleus, kinetoplast, undulating membrane, flagella pocket and the flagellum (Fig. 2A). The trypomastigote-like form of KG1 is 15–30 μm in length and has a width of 2–5 μm. The free flagellum length varied amongst cells, with a minimum of 5 μm and a maximum of 20 μm. Out of the several primers of commonly known trypanosomes that were used in an attempt to amplify the unknown trypanosome DNA, only the T. lewisi primers targeting the ITS1-5.8S, and the universal primers for the 18S rRNA of eukarytic cells, 28S rRNA and the kDNA genes, with product size of approximately 600 bp, 2000 bp, 200 bp and 500 bp, respectively, amplified Trypanosoma KG1 isolate DNA (Fig. 3). These PCR products were successfully cloned and sequenced. However, for the 18S rRNA genes we only obtained a nucleotide sequence which was about 1000 bp in length due to the quality of the sequencing reaction. Hence, in order to obtain additional sequences for more reliable phylogenetic analysis based on the 18S rRNA gene sequence which is at least 2000 bp, we further used two sets of primers designed from the 18S rRNA of T. pestanai (Table 1 and Fig. 1) with PCR products of 513 bp and 681 bp. The locations of EUK-A and B, TPE F1 and B1, and TPE F2 and B2 are schematically indicated in Fig. 1. As a result, we were able to construct a 2154 bp sequence of 18S rRNA from Trypanosoma KG1. The partial gene sequences were submitted to the GenBank as ITS1 (AB259643); 18S rRNA (AB281091); kDNA (AB259645) and 28S rRNA (AB259646). Accordingly, the nucleic acid sequences of the above-mentioned genes mainly showed close homology to a group of non-pathogenic trypanosomes as well as some Bodo, Leishmania and the Crithidia species by BLASTn search (data not shown). Similarly, the phylogenetic tree constructed with the partial 18S rRNA gene of Trypanosoma KG1 which is a conserved region with a length of 2154 bp, Trypanosoma KG1 is clustered together with Trypanosoma sp. wombat H26 and T. pestanai in the bootstrap majority-rule consensus unrooted tree obtained from 100 maximum likelihood replicates (Fig. 4). Bodo caudatus was used as an outgroup parasite.
Giemsa-stained slides of different forms of Trypanosoma KG1 from in vitro culture and within different parts of an Ornithodorus moubata tick. (A) Trypomastigote-like form in the in vitro culture with nucleus (nu), kinetoplast (kin), flagellar pocket (fp), undulating membrane (um), and the flagellum (fg). (B) Trypomastigote-like form in the haemocoel. (C) Trypomastigote-like form in the midgut. (D) Epimastigote-like form in the midgut. (E) Trypomastigote-like form in the salivary glands. (F) Epimastigote-like form in the salivary glands.
Amplification of Trypanosoma KG1 genomic DNA by PCR with different primers targeting different genes. Lane MI, 100 bp Marker; Lane 1, kDNA; Lane 2, TPEF1B1 for 18S rRNA; Lane 3, TPEF2B2 for 18S rRNA; Lane 4, LSU 28S rRNA; Lane 5, ITS1-5.8S; Lane 6, EUK-AB for 18S rRNA; and Lane MII, 1 kb Marker.
Phylogenetic tree based on bootstrapped maximum likelihood analysis of the 18S rRNA gene sequences performed using PHYLIP 3.6 program. The tree includes 21 taxa, values at the nodes are bootstrap values (%: 100 replicates), and the length of the 18S rRNA gene sequence of KG1 for alignment is 2154 bp. The Accession no. of T. sp. wombat H26 sequence is AJ009169. The sequences of other trypanosomes, Leishmania parasites, bodonids, Phytomonas serpens and Crithidia oncopelti were obtained from the GenBank database according to Accession numbers published by Hughes and Piontkivska (Reference Hughes and Piontkivska2003a).
In vitro and in vivo propagation of Trypanosoma KG1 isolate
Ticks were collected in the field with the aim of determining Ehrlichia infections hence, initially tick extracts were cultured with L929. However, actively motile trypanosome-like flagellates were observed in the cultures. In the current study, Trypanosoma KG1 isolate actively propagated in the presence of HEK 293T cell as a feeder cell layer. We could not find HEK 293T cells infected with the parasite throughout the culture period. Giemsa-stained smears of the haemolymph showed the presence of the Trypanosoma KG1 isolate in the O. moubata ticks for up to 30 days post-injection (time at which the experiment was terminated) into the haemocoel (Fig. 2B). In another set of O. moubata ticks that were made to feed on a medium containing KG1 parasites through an artificial membrane, the trypanosomes were detected in the tick midgut and salivary glands for up to 30 days post-infection. Two different life-cycle forms (trypomastigote-like and epimastigote-like) of this trypanosome were observed in both midgut and salivary glands of the tick, whereas in the haemolymph only the epimastigote-like form was visible (Fig. 2C–F). No parasites were observed from haemolymph of ticks fed through an artificial membrane. Following attempts to infect mice, rats, rabbits and sheep with the Trypanosoma KG1 isolate, the parasite could not be detected by microscopical examination as well as by specific gene amplification tests (data not shown).
Development of LAMP and PCR for specific detection of Trypanosoma KG1 isolate
LAMP detection method targeting ITS1 gene of Trypanosoma KG1 isolate has proved to be specific (Fig. 5A) and sensitive, whereby this method detected the genomic DNA up to 10 fg (Fig. 5B). The LAMP primers F3 and B3 have been used for PCR and were also specific (Fig. 5C), with the same sensitivity as LAMP (Fig. 5D).
LAMP and PCR for detection of ITS1 gene of Trypanosoma KG1 species. (A) Specificity of LAMP. (B) Sensitivity of LAMP. (C) Specificity of PCR. (D) Sensitivity of PCR. M and Marker indicate 100 bp ladder DNA size marker up to 1000 bp and additional band of 1500 bp.
DISCUSSION
In this study, we report on the unknown trypanosome species that has been isolated in naturally infected H. hystricis ticks. The clear visibility of common morphological features such as the kinetoplast, nucleus, the flagellar pocket, undulating membrane and the flagellum in the Giemsa-stained smears strongly suggests that the organism belongs to the genus Trypanosoma. The nucleic acid sequences of the cloned PCR products of the ITS1, 18S rRNA, 28S rRNA and kDNA genes of this unknown trypanosome isolate indicated close homology to groups of non-pathogenic trypanosomes. According to the phylogenetic tree constructed in this study from nucleic acid sequences of the 18S rRNA, Trypanosoma KG1 is clustered together with Trypanosoma (Megatrypanum) pestanai which has been reported to infect badgers (Meles meles). However, the vector is unknown (Hoare, Reference Hoare1972) and T. sp. wombat H26 isolated from wombat (Vombatus ursinus), whereby the wombat flea (Lycopsylla nova) has been suggested as a possible vector (Noyes et al. Reference Noyes, Stevens, Teixeira, Phelan and Holtz1999), as are ticks, based on a report in which trypanosomes were found in tick nymphs (Ixodes holocyclus) in Australia (Mackerras, Reference Mackerras1959).
In the phylogenetic analysis of the 18S rRNA and gGAPDH (Hughes and Piontkviska, Reference Hughes and Piontkivska2003a, Reference Hughes and Piontkivskab; Hamilton et al. Reference Hamilton, Steven, Gaunt, Gidley and Gibson2004) T. pestanai clustered with T. (Megatrypanum) theileri and T. (Megatrypanum) cyclops. Whilst, in the phylogenetic analysis of the SSU rRNA T. pestanai clustered only with T. sp. wombat H26 (Hamilton et al. Reference Hamilton, Steven, Gaunt, Gidley and Gibson2004). However, the clustering in the phylogenetic analysis of the current study is different probably because of the differences in the bootstrap support whereby they are generally low or absent for the above-mentioned analyses. Another possible reason could be due to different lengths of sequences used in alignments which are longer or shorter than those used in the current study or the fact that the model of evolution in the maximum likelihood analysis of the current study is different to evolution models used in the above-mentioned analyses.
In particular, T. theileri has also been reported to be naturally infecting the Hyalomma anatolicum anatolicum ticks (Latif et al. Reference Latif, Bakheit, Mohamed and Zweygarth2004). However, Trypanosoma KG1 is morphologically different from T. theileri. Accordingly, the trypomastigote-like form of KG1 has a curved pointed shape 15–30 μm in length (Honda et al. Reference Honda, Fujita, Kuramoto, Watanabe and Takada2006). In contrast, T. theileri is one of the largest mammalian trypanosomes with a mean length of 60–70 μm (Stevens and Brisse, Reference Stevens, Brisse, Maudlin, Holmes and Miles2004). There is great variability within trypanosome species, and a lot of incomplete knowledge about these parasites. New trypanosome species as well as new localities of known trypanosomes are constantly being described (Karbowiak and Wita, Reference Karbowiak and Wita2004). We, therefore, refer to this unknown trypanosome as Trypanosoma KG1 isolate, with reference to the locality of isolation, Kagoshima prefecture of Southern Japan.
Hard ticks of the genera Hyalomma and Rhipicephalus have been reported to harbour trypanosome (el Kady, Reference el Kady1998; Latif et al. Reference Latif, Bakheit, Mohamed and Zweygarth2004) and Leishmania (Coutinho et al. Reference Coutinho, Bueno, Sterzik, Fujiwara, Bolteho, De Maria, Genaro and Linardi2005) parasites, respectively. Trypanosoma KG1 isolate was also isolated in hard tick H. hystricis. Additionally we have been able to propagate the trypanosome in vivo by infecting the soft tick O. moubata and the parasite could be isolated in the midgut and the salivary glands. This observation suggests that this trypanosome has the ability to infect both the hard and soft tick species. Some of the Stercorarian trypanosomes to which Trypanosoma KG1 isolate is closely related, have a narrow host range for their vertebrate hosts where they live extracellularly, primarily in the bloodstream (Sato et al. Reference Sato, Ishita, Matsuo, Inaba, Kamiya and Ito2003). This could explain our unsuccessful attempts to propagate Trypanosoma KG1 isolate in the experimental animals (mice, rats, rabbits and sheep).
According to Yamakuti et al. (Reference Yamaguti, Tipton, Keegan and Toshioka1971), H. hystricis distribution is restricted to a subtropical and temperate belt of Eastern Asia, and the adult ticks have a fairly wide host range including humans, domestic dogs, buffalo, pigs, wild boars, tigers, hog-badgers, goat-like deer (Muntiacus reevesi), short-eared rabbit (Lepus sinensis) and the Ryuku black rabbit (Yamaguti et al. Reference Yamaguti, Tipton, Keegan and Toshioka1971; Mahara, Reference Mahara1997; Cao et al. Reference Cao, Gao, Zhang, Zhang, Dai, Dumler, Fang and Yang2000; Parola et al. Reference Parola, Cornet, Sanogo, Miller, Van Thien, Gonzalez, Raoult, Telford and Wongsrichanalai2003). This wide host range of H. hystricis suggests that one or more of the above-mentioned mammals could be harbouring Trypanosoma KG1 isolate, particularly the wild pig (Sus scrofa), as it has been reported to be a host of H. hystricis in Kagoshima (Yamaguti et al. Reference Yamaguti, Tipton, Keegan and Toshioka1971), badgers also (which are hosts for T. pestanai which is clustered together with KG1) are possible hosts. In Sudan, Morzaria et al. (Reference Morzaria, Latif, Jongejan and Walker1986) demonstrated the first biological transmission of T. theileri to cattle by the tick H. a. anatolicum. It is therefore highly possible that Trypanosoma KG1 might be transmitted by the ticks during feeding since we detected the parasite in the salivary glands of experimentally infected O. moubata tick.
LAMP reaction amplifies DNA with high specificity, efficiency and rapidity under isothermal conditions using only simple incubators (Notomi et al. Reference Notomi, Okayama, Masubuchi, Yonekawa, Watanabe, Amino and Hase2000), and has recently been developed for diagnosis of African trypanosomes (Kuboki et al. Reference Kuboki, Inoue, Sakurai, Di Cello, Grab, Suzuki, Sugimoto and Igarashi2003; Thekisoe et al. Reference Thekisoe, Inoue, Kuboki, Tuntasuvan, Bunnoy, Borisutsuwan, Igarashi and Sugimoto2005), whilst PCR is already an established and widely used diagnostic technique for trypanosomes (Zarlenga and Higgins, Reference Zarlenga and Higgins2001; Desquesnes and Davila, Reference Desquesnes and Davila2002; Cox et al. Reference Cox, Tilley, McOdimba, Fyfe, Eisler, Hide and Welburn2005). Thus, this study has also developed sensitive and specific LAMP and PCR methods that could later be used in attempts to identify vertebrate host(s) and epidemiological studies of the Trypanosoma KG1 isolate.
We are now faced with challenges of identifying the possible vertebrate host(s), which may lead to a better understanding of its transmission and parasite host-relationship. There is also a need for further studies of other trypanosome species and ticks as this might give information on non-survival of other trypanosomes in ticks, eventually identifying responsible molecules that could be used as trypanosome vaccine candidates.
This study was supported by a Grant-in-Aid for Scientific Research to N. I., K. F. and C. S. from the Japan Society for the Promotion of Science. The authors are thankful to Professor T. Itagaki, Laboratory of Veterinary Parasitology, Iwate University, Japan, for his excellent advice on phylogenetic analysis.
