Molecular Ecology (2005) 14, 3167–3175

doi: 10.1111/j.1365-294X.2005.02602.x

The inadvertent introduction into Australia of Trypanosoma nabiasi, the trypanosome of the European rabbit (Oryctolagus cuniculus), and its potential for biocontrol

Blackwell Publishing, Ltd.

P . B . H A M I L T O N ,* J . R . S T E V E N S ,† P . H O L Z ,‡ B . B O A G ,§ B . C O O K E ¶ and W . C . G I B S O N * *School of Biological Sciences, University of Bristol, Bristol BS8 1UG, UK, †School of Biological Sciences, University of Exeter, Exeter EX4 4PS, UK, ‡Healesville Sanctuary, Badger Creek Road, Healesville, Victoria 2777, Australia, §Scottish Crop Research Institute, Invergowrie, Dundee, DD2 5DA, UK, ¶Fundación Charles Darwin, Casilla 17-01-3891, Quito, Equador

Abstract Wild rabbits (Oryctolagus cuniculus) in Australia are the descendents of 24 animals from England released in 1859. We surveyed rabbits and rabbit fleas (Spilopsyllus cuniculi) in Australia for the presence of trypanosomes using parasitological and PCR-based methods. Trypanosomes were detected in blood from the European rabbits by microscopy, and PCR using trypanosome-specific small subunit ribosomal RNA (SSU rRNA) gene primers and those in rabbit fleas by PCR. This is the first record of trypanosomes from rabbits in Australia. We identified these Australian rabbit trypanosomes as Trypanosoma nabiasi, the trypanosome of the European rabbit, by comparison of morphology and SSU rRNA gene sequences of Australian and European rabbit trypanosomes. Phylogenetic analysis places T. nabiasi in a clade with rodent trypanosomes in the subgenus Herpetosoma and their common link appears to be transmission by fleas. Despite the strict host specificity of trypanosomes in this clade, phylogenies presented here suggest that they have not strictly cospeciated with their vertebrate hosts. We suggest that T. nabiasi was inadvertently introduced into Australia in the 1960s in its flea vector Spilopsyllus cuniculi, which was deliberately introduced as a potential vector of the myxoma virus. In view of the environmental and economic damage caused by rabbits in Australia and other islands, the development of a virulent or genetically modified T. nabiasi should be considered to control rabbits. Keywords: 18S rRNA, Australia, biological introduction, evolution of parasitism, immunocontraception, phylogeny, rabbit, Trypanosoma Received 3 February 2005; revision accepted 31 March 2005

Introduction Trypanosomes (genus Trypanosoma) are widespread parasites found in the blood of all classes of vertebrates, and several trypanosome species are agents of disease in humans and/or livestock, particularly in the tropics. Trypanosomes have been described from several indigenous Australian vertebrates (Mackerras 1959, 1961a, b; Mackerras & Mackerras 1959; Bettiol et al. 1998; Noyes et al. 1999; O’Donoghue & Adlard 2000; Jakes et al. 2001; Hamilton et al. 2005) and three species have been recorded in

Correspondence: Professor Wendy Gibson, Fax: 0117 925 7374; E-mail: [email protected] © 2005 Blackwell Publishing Ltd

introduced vertebrates in Australia, namely Trypanosoma theileri from cattle (Bos taurus), Trypanosoma melophagium from sheep (Ovis aries) and Trypanosoma lewisi from rats (Rattus rattus) (Mackerras 1959). These three trypanosome species have a cosmopolitan distribution and were probably introduced into Australia with their host species (Mackerras 1959). Wild rabbits Oryctolagus cuniculus in Australia are the descendents of 24 rabbits from England released in 1859 (Hinds et al. 1996), which multiplied and spread across most of the continent by 1910 (Stodart & Parer 1988). A number of endo- and ectoparasites have been reported in Australian rabbits, including four species of flea [Echidnophaga perilis, Echidnophaga myrmecobii, Spilopsyllus cuniculi, Xenopsylla cunicularis (Myers et al. 1989)], the latter

3168 P . B . H A M I L T O N E T A L . two having been deliberately introduced as vectors of myxomatosis, but trypanosomes have not been previously reported (Mackerras 1959; Myers et al. 1989; O’Donoghue & Adlard 2000). Trypanosoma nabiasi is a trypanosome of rabbits, which has been reported in O. cuniculus from various European countries [UK, France, Italy, Portugal (Grewal 1957)] and in domestic O. cuniculus outside Europe (Hoare 1972). Morphologically indistinguishable trypanosomes have been found in North American cottontail rabbits (Sylvilagus nuttalli, Sylvilagus auduboni and Sylvilagus floridanus), but it is not known if these are T. nabiasi (Hoare 1972). The vector of T. nabiasi is the rabbit flea, S. cuniculi (Hoare 1972), and the route of infection is by contact with faecal metatrypanosomes rather than by flea bite; rabbits ingest fleas and flea faeces when grooming, as they use the mouth and tongue to clean their toes after scratching. Channon & Wright (1927) demonstrated that rabbits became infected when suspensions of the intestinal contents of infected fleas taken from rabbit burrows were placed in their mouths. After infection of O. cuniculus with T. nabiasi, trypanosomes appear in the bloodstream 5 to 12 days later; parasitaemia increases rapidly for the next 4 to 9 days, reaching as many as 1.6 × 107 organisms per mL of blood, before gradually decreasing (Channon & Wright 1927; Kroó 1939; Grewal 1957). The duration of infection varies from 4 to 8 months, after which rabbits are immune to reinfection (Channon & Wright 1927; Kroó 1939; Grewal 1957). Experimental evidence suggests that T. nabiasi is host specific, or at least restricted to lagomorphs, as laboratory rodents (guinea pigs, rats, mice) are not susceptible to infection (Ashworth et al. 1909; Petrie 1905; Watson & Hadwen 1912; Channon & Wright 1927; Grewal 1957; Holliman 1966). Although T. nabiasi may well have been introduced into Australia in the rabbits originally imported, it is unlikely to have survived, as its only known vector, the European rabbit flea S. cuniculi, was introduced into the wild much later, in 1968, to aid the transmission of myxomatosis (Sobey & Conolly 1971), and vertical transmission between generations of rabbits is not believed to occur (Hoare 1972). Our finding of trypanosomes in rabbits and their fleas in Australia therefore poses the question of their identity and is the subject of this investigation.

Materials and methods Collection and preparation of samples Australian samples were collected between April 2000 and October 2001. Blood samples were taken from a total of 16 rabbits from two sites in Victoria [Healesville, sample AAM (collected 8 April 2000); Bacchus Marsh, samples 1–10 (collected 22–23 September 2001)], and one site in New South Wales [Orange, samples 11–15 (collected 5 October 2001)]. The

rabbits from Bacchus Marsh were young (about 40 – 60 days old) and generally 400–600 g in body weight, whereas those from Orange were adults (> 1 year old). Rabbit fleas (Spilopsyllus cuniculi) were also sampled from wild rabbits in Healesville and Orange. Eggs laid by fleas from Orange were used to initiate a laboratory colony of fleas and provide laboratory-bred unfed fleas as experimental controls. Fleas from each location were killed by immersion in 70% ethanol and stored at ambient temperature or at 4 °C in plastic containers for up to 6 months. In the UK, rabbit fleas (S. cuniculi) were collected in the same way from 16 wild rabbits from Pitroddie, Scotland, between April and May 2004. For attempted isolation of trypanosomes into culture, 0.1–0.5 mL aliquots of blood were added to culture tubes containing 3–5 mL of modified ‘Sloppy Evans’ medium (Noyes et al. 1999), made with either rabbit or horse blood. Cultures were incubated in the dark at ambient temperature and a small drop was examined weekly for the presence of trypanosomes for up to 6 weeks. DNA was prepared from blood samples and pooled fleas by standard procedures as described by Hamilton et al. (2005). Fresh rabbit blood samples were examined for trypanosomes by phase contrast microscopy at 100× and 400× magnification. Thin films were prepared, stained with Giemsa and examined at 400× and 1000× using oil immersion. Giemsa-stained blood films from a UK rabbit were kindly supplied by Dr J. Webster (Oxford University, UK). Measurements of stained trypanosomes were made as described by Hoare (1972).

PCR analysis Fragments of the trypanosome SSU rRNA gene were amplified by polymerase chain reaction (PCR) essentially as described previously (Maslov et al. 1996; Noyes et al. 1999; Stevens et al. 1999a). Primers used were S-823 (B), S-713 (C), S-825 (D), S-827 (E), S-829 (F), S-762 (G), S- 714 (H) S-662 (I), S-757 (M) (Maslov et al. 1996) and S (Hamilton et al. 2005). Regions amplified included the most variable region of the SSU rRNA gene, as described by Noyes et al. (1999). Thermoprime polymerase (ABgene) was used for all PCRs following the manufacturer’s recommendations. The following PCR program was used: an initial denaturing step of 95 °C for 180 s, followed by 30 cycles of 60 s at 95 °C, 30 s at 55 °C and 60 s at 72 °C and a final extension period of 72 °C for 400 s. Nested PCR was used as it enabled amplification of greater lengths than standard PCR (McPherson & Moller 2000). For nested PCR, 0.5 µL of the PCR mixture from the first reaction was used as the template for the second reaction. Primers used in nested PCR are abbreviated as follows: AF-BG means that primers A and F were used in the first round, followed by B and G in the second. © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 3167–3175

T H E A U S T R A L I A N R A B B I T T R Y P A N O S O M E 3169

Sequence and phylogenetic analysis Amplified fragments were purified and cloned into a plasmid vector (pCR2.1-TOPO, Invitrogen) or sequenced directly. Purified plasmid DNA was sequenced using M13 forward and reverse, and internal primers using an automated sequencer. Consensus sequences were assembled using AutoAssembler 2.0 (ABI). Sequences were subjected to blast analysis to find the most similar sequences in the GenBank database. For phylogenetic analysis, sequences from this study and others obtained from GenBank were aligned using clustal_x (Thompson et al. 1997), with subsequent manual adjustment. For all alignments, regions judged to be poorly aligned by eye and characters with a gap in any sequence were excluded from subsequent analyses. Restricting the number of taxa in alignments 2 and 3 allowed more characters in variable regions to be included which were potentially useful for determining relationships of the shorter sequences in these alignments. Three alignments were constructed for SSU rRNA gene sequences: (i) An alignment of the 1504 bp rabbit trypanosome sequence together with 43 trypanosome and 29 other kinetoplastid sequences based on the alignments of Hamilton et al. (2004, 2005). Characters at positions 1–40, 250 – 308, 870 – 930, 960 – 976, 1240 –1480, 1690–1823, 1840–1910 and 2800–2871 were judged to be poorly aligned, so were excluded; 1128 characters of which 228 were parsimony-informative were included in subsequent analyses. (ii) An alignment of the 1504 bp

rabbit trypanosome sequence with 13 long (> 1828 bp) sequences from the Trypanosoma lewisi clade (Table 1) and seven representative taxa from other trypanosome clades. All characters were judged to be well aligned; 1460 characters, of which 64 were parsimony informative were included in subsequent analyses. (iii) An alignment of the Australian and UK rabbit trypanosome sequences with 20 other sequences of T. lewisi clade trypanosomes. All characters were judged to be well aligned; 506 characters, of which 16 were parsimony informative, were included in subsequent analyses. The alignment used for 1 and 2 can be obtained via FTP from FTP.EBI.AC.UK in directory/pub/databases/embl/align or via the emblalign database via SRS at http://srs.ebi.ac.uk under accession: ALIGN_000842. Alignments 1–3 were analysed by maximum-likelihood (ML) distance analysis, maximum parsimony (MP) and alignments 2 and 3 were also analysed by ML analysis as implemented in the program paup version 4.0b10 (Swofford 2003). The appropriate model of nucleotide substitution for ML and ML distance analysis for each alignment was chosen using the Akaike information criterion (Akaike 1974) implemented in the program modeltest, version 3.06 (Posada & Crandall 1998). For alignments 1 and 2 this model was GTR + I + G with empirical nucleotide base frequencies; a four-category gamma distribution was used. For alignment 2, the chosen model was TVMef with equal base frequencies. For ML analysis 100 bootstrap replicates were performed. For MP analyses, heuristic

Table 1 Information on the origins of Trypanosoma sequences in this study Origin Trypanosome species

Isolate code

Host

Location

Accession no.

T. evotomys T. blanchardi T. grosi T. grosi T. grosi T. grosi T. lewisi T. lewisi T. microti T. microti T. microti T. microti T. microti T. musculi T. otospermophili T. otospermophili T. rabinowitschae T. sp. T. sp. T. talpae

036580025–109

vole Clethrionomys glareolus dormouse Eliomys quercinus woodmouse Apodemus sylvaticus striped field mouse Apodemus agrarius Korean woodmouse Apodemus peninsulae Japanese woodmouse Apodemus speciosus rat Rattus sp. rat Rattus norvegicus vole Microtus agrestis vole Microtus agrestis vole Microtus sp. vole Microtus sp. vole Microtus sp. house mouse Mus musculus Richardson’s ground squirrel Spermophilus richardsonii Columbia ground squirrel Spermophilus columbianus common hamster Cricetus cricetus woodmouse Apodemus sylvaticus Siberian flying squirrel Pteromys volans European mole Talpa europea

UK France UK Russia Russia Japan ? ? UK UK Alaska Alaska Alaska ? Japan import Japan import France UK Japan import UK

AY043356 AY491764 AY043355 AB175622 AB175623 AB175624 AJ009156 AJ223566 AJ009158 AY043354 AY586623 AY586621 AY586622 AJ223568 AB175625 AB190228 AY491765 AY043353 AB175626 AJ620545

110 SESUJI HANTO AKHA Molteno B3 ATCC 30085 TRL132 148 AF 61877 AF 59915 AF 59919 LUM 343

102(WM2) Pteromys

© 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 3167–3175

3170 P . B . H A M I L T O N E T A L .

Analysis of rabbit blood and rabbit flea samples

Fig. 1 Light micrographs of trypanosomes in rabbit blood from the UK and from Australia (AU: sample AAM).

searches were performed with 100 random addition replicates and TBR branch swapping. A strict consensus was made of the shortest trees. For MP and ML distance analysis 1000 bootstrap replicates were calculated.

Results Morphological comparisons Trypanosomes were seen in fresh blood from a wild rabbit caught at Healesville, Australia (sample AAM). Comparison of Giemsa-stained blood films from this rabbit and UK rabbits showed morphologically similar trypanosomes with a prominent kinetoplast, pointed posterior end and a long free flagellum (Fig. 1). The dimensions are given in Table 2. The trypanosomes differed significantly in total length and length of free flagellum (Student’s t-test: P < 0.05). However, all measurements were in the range of descriptions of the rabbit trypanosome Trypanosoma nabiasi by Channon & Wright (1927) and Kroó (1939).

Attempts to culture trypanosomes from infected blood of the Healesville rabbit AAM using sloppy Evans medium made with either horse or rabbit blood were unsuccessful. This reiterates the experience of others who were unable to culture T. nabiasi using semisolid blood culture medium or other media (Channon & Wright 1927; Grewal 1957; Holliman 1966; Mohamed & Molyneux 1987). As all of blood sample AAM was used to initiate cultures, a DNA preparation was not made from this sample. However, of the 15 other rabbit blood DNA samples from Australia, two from Bacchus Marsh in Victoria gave a PCR product with primers D and I (300 bp). Larger fragments of the SSU rRNA gene were obtained from these samples using primers C and H (900 bp) and primers AF-BG (1500 bp). These fragments were cloned and sequenced. Due to the low amounts of DNA template, the complete sequence of the SSU rRNA gene could not be obtained. For DNA samples derived from pooled rabbit fleas (Spilopsyllus cuniculi), three of four samples from Orange in New South Wales gave an amplified product with primers D and I (300 bp), as did the sample from Healesville, Victoria, in nested PCR with primers EF-MG (300 bp). These fragments were cloned and sequenced. DNA samples from unfed fleas gave no amplified products. Of 16 flea DNA samples from the UK, eight gave amplified products with primers D and S (500 bp); sequences were obtained from seven of the cloned fragments. Alignment in autoassembler (ABI) showed that all trypanosome sequences from Australian rabbits and rabbit fleas were identical in the region of overlap. For the UK flea samples, alignment in autoassembler (ABI) of the seven c. 500 bp sequences obtained revealed two different types. Three samples (UKF3, UKF4, UKF11) gave a 523 bp sequence (Accession no. AJ843896) that was identical to the consensus sequence of the Australian rabbit trypanosome, while the other four samples (UKF5, UKF8, UKF9, UKF16) gave a 517 bp sequence (Accession no. AJ843897) that differed from the 523 bp sequence by 11%. A blast search on this sequence revealed greatest similarity (93%) to the SSU rRNA gene sequence of Herpetomonas mariadeanei (Accession no. U01013), so it is likely to originate from an

Dimensions: mean and standard error (µm)

Total length Distance between kinetoplast and posterior Distance between kinetoplast and nucleus Length of free flagellum

Australian sample

UK sample

25.1 ± 0.37 (n = 15) 3.0 ± 1.4 (n = 18) 7.6 ± 0.10 (n = 19) 10.6 ± 0.52 (n = 11)

26.4 ± 0.28 (n = 24) 3.1 ± 0.13 (n = 24) 7.5 ± 0.15 (n = 24) 8.5 ± 0.41 (n = 24)

Table 2 Dimensions of trypanosomes from Australian and UK rabbits

© 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 3167–3175

T H E A U S T R A L I A N R A B B I T T R Y P A N O S O M E 3171 insect-only flea parasite. Trypanosomatids of genus Leptomonas have frequently been described from flea species in the past (Wallace 1966).

Phylogenetic position of Trypanosoma nabiasi Since all trypanosome sequences obtained from rabbit blood and rabbit flea samples in Australia and UK were identical, this sequence should represent that of T. nabiasi, the rabbit trypanosome. A blast search on the longest sequence (1504 bp) obtained from a Bacchus Marsh rabbit revealed greatest similarity (99%) to the SSU rRNA gene sequence of Trypanosoma microti (Accession no. AJ009158), a trypanosome from the vole (Microtus sp.). For phylogenetic placement, the 1504 bp sequence was added to an alignment of 70 other kinetoplastid taxa based on that of Hamilton et al. (2005), and this alignment was analysed by MP and distance analysis. In the resulting phylogenetic trees (not shown), the rabbit trypanosome fell in a moderately well-supported clade (bootstrap support: MP, 70%; ML distance, 84%) with Trypanosoma lewisi (a rat Rattus sp. trypanosome) and T. microti. A close relationship between the two rodent trypanosomes, T. lewisi and T. microti, was evident from previously published SSU rRNA gene phylogenies (e.g. Stevens et al. 1999a, 2001; Martin et al. 2002), and also phylogenies constructed from comparison of glycosomal glyceraldehyde phosphate dehydrogenase (gGAPDH) gene sequences (Hamilton et al. 2005). This clade will be referred to here as the ‘T. lewisi clade’. To determine the position of the rabbit trypanosome within the T. lewisi clade more precisely, the 14 long (> 1504 bp) SSU rRNA gene sequences available, together with seven representatives from outside the clade were aligned. In the ML tree shown (Fig. 2a), as well as in MP and distance trees (not shown), T. nabiasi is in the strongly supported T. lewisi clade. All trypanosomes in this clade are from mammals and the only known vectors are fleas. Most of these trypanosome species belong to subgenus Herpetosoma, except Trypanosoma talpae of the mole (Talpa europea), which is in subgenus Megatrypanum (Hoare 1972). However, other trypanosomes that have been traditionally placed in the subgenus Herpetosoma, such as Trypanosoma rangeli, fall outside the T. lewisi clade (Fig. 2a), and are more closely related to trypanosomes in the subgenus Schizotrypanum, such as Trypanosoma cruzi (Stevens et al. 1999b). Within the T. lewisi clade, the mole trypanosome is at the base on a branch by itself. The rest fall in a robustly supported clade, comprising two subclades (Fig. 2a). Subclade 1 contains trypanosomes from two rodent families: Muridae (Trypanosoma blanchardi, Trypanosoma grosi, T. lewisi and Trypanosoma musculi) and Gloridae (Trypanosoma rabinowitschae), while subclade 2 contains T. nabiasi, T. microti and two trypanosomes from squirrels. © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 3167–3175

Including a further eight partial (minimum 532 bp) SSU rRNA gene sequences in the alignment allowed more detailed analysis of the T. lewisi clade. The T. nabiasi sequence differed from all other sequences in the alignment, with unique nucleotides at five aligned sites. Phylogenies based on this alignment, e.g. Fig. 2b, also show the mole trypanosome by itself and the rest in two subclades. Subclade 2, including the rabbit trypanosome, also contains Trypanosoma evotomys and a trypanosome from a woodmouse (Trypanosoma sp. WM2). The relationships shown in these trees are in agreement with isoenzyme analysis, which places T. evotomys and T. microti in one group and T. grosi, T. musculi and T. lewisi in another (10 enzymes compared; Mohamed et al. 1987). Trypanosoma nabiasi was distinct from these, in three of four enzymes (Mohamed et al. 1987). The subclades are also consistent with development in the vertebrate host: all trypanosomes in subclade 1 divide predominantly as epimastigotes in the bloodstream; in contrast, those that have been studied in subclade 2 divide as amastigotes in lymphoid tissues [T. microti, T. evotomys, T. nabiasi (Molyneux 1976)] or spleen capillaries [T. nabiasi (Grewal 1957)] and not in the peripheral bloodstream.

Discussion The identity of the Australian rabbit trypanosome Trypanosomes were detected by microscopy and PCR in rabbit blood and by PCR in rabbit fleas from Australia. These are the first reports of trypanosomes from the rabbit in Australia. Trypanosomes were detected in three locations in two southeastern Australian states: Orange in New South Wales, and Healesville and Bacchus Marsh in Victoria. Trypanosoma nabiasi is the only known trypanosome of rabbits and rabbit fleas in Europe (Hoare 1972; Molyneux 1976), and here, European and Australian rabbit trypanosomes were found to be indistinguishable by morphology or comparison of SSU rRNA gene sequences. We therefore conclude that the Australian rabbit trypanosome is T. nabiasi, and is not a trypanosome of indigenous Australian wildlife that adapted to rabbits.

The introduction of Trypanosoma nabiasi to Australia Trypanosoma nabiasi must have been inadvertently introduced into Australia in rabbits or rabbit fleas, because until now its presence has remained unrecognized. If it was present in the first rabbits imported from Europe in the 1800s (Hinds et al. 1996), it seems unlikely to have survived, as the only known vector — the European rabbit flea Spilopsyllus cuniculi — was introduced into the wild in 1968, over 100 years later. Vertical transmission of T. nabiasi between generations of rabbits is not believed to occur (Hoare 1972). It is not known if flea species of

3172 P . B . H A M I L T O N E T A L . Fig. 2 (a) Maximum-likelihood tree based on an alignment of SSU rRNA gene sequences. It includes 22 taxa and is calculated from an alignment of 1460 characters. Values at nodes are bootstrap values (%) in order: ML, MP and ML distance. −ln L 3382.85307. Vertebrate hosts are in parentheses. (b) Maximum-likelihood tree, based on an alignment of SSU rRNA gene sequences. It includes 22 taxa and is calculated from an alignment of 506 characters. Values at nodes are bootstrap values (%) in order: ML, MP and ML distance. −ln L 907.79964. Vertebrate hosts are in parentheses.

other vertebrates that coinhabit rabbit warrens, such as Leptopsylla segnis of the house mouse, or the rat flea Nosopsyllus fasciatus, which has been reported from rabbits in Australia (Dunnet & Mardon 1974), could have maintained transmission. Two species of indigenous Australian stickfast flea, Echidnophaga perilis and Echidnophaga myrmecobii, which have a wide host range including rabbits, could not have initially maintained T. nabiasi, as they are absent from wetter regions (> 300 mm average annual

rainfall) of Australia (Dunnet & Mardon 1974), such as the site near Geelong (Victoria) where rabbits were first introduced. Likewise the mites, Listrophorus gibbus and Cheyletiella parasitivorax known from Australian rabbits (Myers et al. 1989), are unlikely vectors, as mites are not known to transmit any Trypanosoma lewisi-like trypanosome (Hoare 1972; Molyneux 1976). One possibility is that T. nabiasi was imported with 11 adult rabbit fleas (Spilopsyllus cuniculi) introduced in © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 3167–3175

T H E A U S T R A L I A N R A B B I T T R Y P A N O S O M E 3173 May 1966 from Spain to be evaluated as potential vectors of myxomatosis under Australian conditions (Sobey & Conolly 1971). Once in Australia, the fleas might have passed the infection to a captive rabbit colony on which they were reared. These fleas bred only once and died out in August 1966, but were replaced with 300 S. cuniculi pupae from the UK in November 1966 (Sobey & Menzies 1969). The UK flea pupae could not have been the source of trypanosomes, as trypanosomes are not known to be vertically transmitted between generations of invertebrates (Hoare 1972). However, the fleas that hatched from them may have subsequently maintained transmission of T. nabiasi in the infected rabbit colony until the fleas were released into populations of wild rabbits in 1968. Alternatively, T. nabiasi may have been introduced in imported domestic rabbits after the flea vector was established in the wild. However, the persistence of T. nabiasi in domestic rabbits and its transfer to wild rabbits seem unlikely as S. cuniculi does not normally persist in colonies of caged domestic rabbits. Nevertheless Hoare (1972) noted the presence of T. nabiasi in domestic rabbits outside Europe, where S. cuniculi was probably absent. The Spanish rabbit flea, Xenopsylla cunicularis, was also brought into Australia and released in 1991 to aid transmission of myxomatosis in arid areas (Cooke 1990). Although it is not known if this species can transmit T. nabiasi, it is also an unlikely route for introduction of trypanosomes, because pupae were imported to found the colony and were eventually released into the wild (Cooke 1995). At the time, importing pupae was considered an essential precaution to avoid importing rabbit diseases and parasites of fleas. Several species of spilotylenchid nematode parasitize S. cuniculi and reduce its fecundity (Launay & Deunff 1990). It is interesting that the introduction of S. cuniculi into Australia was associated with a sharp decline in rabbit populations in some areas (Williams et al. 1995). Although this was attributed to the enhanced transmission of myxomatosis at the time, in retrospect an introduction of T. nabiasi could also have been a contributing factor. Trypanosoma lewisi-like trypanosomes are generally regarded as nonpathogenic, but there are occasional reports of them causing disease (Molyneux 1976). Nevertheless, evidence for the pathogenicity of T. nabiasi is equivocal. While Petrie (1905) found blood infected with T. nabiasi to be ‘toxic’, killing two of 10 rabbits inoculated, Grewal (1957) did not detect pathogenic effects in 24 experimentally infected laboratory rabbits and two naturally infected wild rabbits. If trypanosomes were brought into Australia inadvertently in association with the deliberate introduction of fleas, it would highlight the risks inherent in introducing such agents to facilitate biological control. Strict adherence to quarantine protocols and refinement of protocols as better information becomes available, are essential to prevent such introductions in the future. On the other hand, if © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 3167–3175

trypanosomes were brought in with rabbits imported in the 1800s, this raises the question whether S. cuniculi is the only species that can transmit T. nabiasi. This could be answered if more was known about the distribution of T. nabiasi. For example, it would be interesting to know whether T. nabiasi is present in arid inland areas of Australia, where E. myrmecobii, E. perilis, X. cunicularis but not S. cuniculi are present, and in New Zealand where there are no hostspecific rabbit fleas. Fortunately, because of its host specificity, T. nabiasi is unlikely to harm the indigenous Australian fauna, and may even prove beneficial by contributing to the mortality of Australian rabbits, which are agricultural pests (Hinds et al. 1996) and damage the indigenous vegetation (Cooke 1987; Drollette 1997) and fauna (Smith & Quin 1996; Priddel et al. 2000). However it is possible that T. nabiasi will adapt to new hosts over time, as the phylogenetic evidence presented here shows host jumping of this group of trypanosomes has occurred in the past (see below).

The potential to develop genetically modified Trypanosoma nabiasi to control feral rabbits Viral-vectored immunocontraception has been under development as a means of controlling rabbits in Australia (Tyndale-Biscoe 1994; Kerr & Jackson 1995; Jackson et al. 1996; Kerr et al. 1999; Barlow 2000 Angulo & Cooke 2002). The myxoma virus has been transformed with genes for rabbit reproductive proteins (Gu et al. 2004; Kerr et al. 1999) and rabbits infected with the recombinant virus develop an immune response to self-reproductive proteins, decreasing fertility (Gu et al. 2004; Kerr et al. 2004). There are potential benefits of developing a recombinant T. nabiasi in a similar way to control rabbits. Genetic transformation of T. nabiasi should be feasible as other trypanosome species are routinely transformed (Beverley & Clayton 1993). Trypanosoma nabiasi could be transformed with genes for several rabbit reproductive proteins, whereas the genome size of viruses places a greater limit on the amount of foreign DNA introduced. Additionally, T. nabiasi is already present in Australia and does not appear to harm the native fauna. Its host specificity and that of its flea vector would limit exposure of the native fauna to the genetically modified trypanosome, whereas the myxoma virus is transmitted by mosquitoes in addition to fleas, increasing exposure (Fenner & Ratcliffe 1965; Hinds et al. 1996). Indeed, the phylogenetic evidence suggests that poxviruses (the group including the myxoma virus) have jumped between diverse mammalian hosts over evolutionary time (McLysaght et al. 2003).

Evolution of the Trypanosoma lewisi clade trypanosomes The T. lewisi clade described here contains trypanosomes that are parasites of rodents (order Rodentia), rabbits

3174 P . B . H A M I L T O N E T A L . (order Lagomorpha) and moles (order Insectivora). Despite this wide range of mammalian hosts, all trypanosomes in the clade are transmitted by fleas, indicating a specific host–parasite relationship. Cross-inoculation experiments indicate that T. lewisi-like trypanosomes show host restriction (Hoare 1972; Molyneux 1976), suggesting cospeciation of parasite and vertebrate host. There is some support for this idea from the SSU rRNA gene phylogenies presented here as trypanosomes of rodents in the subfamily Murinae (Trypanosoma grosi, Trypanosoma musculi and T. lewisi) fall in subclade 1, whereas trypanosomes of the subfamily Arvicolinae (Trypanosoma evotomys and Trypanosoma microti) fall in subclade 2. However, strict cospeciation is not supported and it appears that host jumping is more important in the evolution of the clade. For example, subclade 1 contains trypanosomes from rodents in the families Gliridae and Muridae which diverged at least 45 million years ago (Ma) (Huchon et al. 2002). In addition, subclade 2 also contains trypanosomes from rabbits and squirrels and trypanosomes from the mouse genus Apodemus fall in both subclades. Indeed the phylogenetic position of T. nabiasi suggests that it evolved from a rodent trypanosome; rodents are monophyletic (Huchon et al. 2002) and lagomorphs are their closest living relatives, from which they diverged approximately 80 – 70 Ma (Springer et al. 2003).

Acknowledgements We thank Dr Joanne Webster (Oxford University), Arrow Pest Control (Bristol) and Cristina Musso (CSIRO) for provision of parasite samples. The collection of Scottish material was funded by the Leverhulme Trust. PBH was funded by a Wellcome Trust Biodiversity studentship.

References Akaike H (1974) A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19, 716 – 723. Angulo E, Cooke B (2002) First synthesize new viruses then regulate their release? The case of the wild rabbit. Molecular Ecology, 11, 2703–2709. Ashworth JH, MacGowan JP, Ritchie J (1909) Note on the occurrence of a trypanosome (Trypanosoma cuniculi) in the rabbit. Journal of Pathology and Bacteriology, 13, 437. Barlow ND (2000) The ecological challenge of immunocontraception: editor’s introduction. Journal of Applied Ecology, 37, 897–902. Bettiol SS, Jakes K, Le DD, Goldsmid JM, Hocking G (1998) First record of trypanosomes in Tasmanian bandicoots. Journal of Parasitology, 84, 538 – 541. Beverley SM, Clayton CE (1993) Transfection of Leishmania and Trypanosoma brucei by electroporation. In: Protocols in Molecular Parasitology (ed. Hyde JE), pp. 333 – 348. Humana Press, Totowa, New Jersey. Channon HA, Wright HD (1927) Observations on trypanosomiasis of rabbits and its natural mode of transmission. Journal of Pathology and Bacteriology, 30, 253 – 260.

Cooke BD (1987) Effects of rabbit grazing on regeneration of sheoaks, Allocasuarina verticilliata, and salt-water ti-tree, Melaleuca halmaturorum, in Coorong National Park, South Australia. Australian Journal of Ecology, 13, 11–12. Cooke BD (1990) Notes on the comparative reproductive biology and the laboratory breeding of the rabbit flea Xenopsylla cunicularis Smit (Siphonaptera: Pulicidae). Australian Journal of Zoology, 38, 527–534. Cooke BD (1995) Spanish rabbit fleas, Xenopsylla cunicularis, in arid Australia: a progress report. 10th Australian Vertebrate Pest Conference, Hobart, Tasmania, 399–401. Drollette D (1997) Wide use of rabbit virus is good news for native species. Science, 275, 154–154. Dunnet GM, Mardon DK (1974) Monograph of Australian fleas (Siphonaptera). Australian Journal of Zoology Suppl. ser., No 30, 1– 272. Fenner F, Ratcliffe FN (1965) Myxomatosis. Cambridge University Press, Cambridge. Grewal MS (1957) The life cycle of the British rabbit trypanosome, Trypanosoma nabiasi Railliet, 1895. Parasitology, 47, 100–118. Gu W, Holland M, Janssens P, Seamark R, Kerr P (2004) Immune response in rabbit ovaries following infection of a recombinant myxoma virus expressing rabbit zona pellucida protein B. Virology, 318, 516–523. Hamilton PB, Stevens JR, Gaunt MW, Gidley J, Gibson WC (2004) Trypanosomes are monophyletic: evidence from genes for glyceraldehyde phosphate dehydrogenase and small subunit ribosomal RNA. International Journal for Parasitology, 34, 1393–1404. Hamilton PB, Stevens JR, Gidley J, Holz P, Gibson WC (2005) A new lineage of trypanosomes from Australian vertebrates and terrestrial bloodsucking leeches (Haemadipsidae). International Journal for Parasitology, 35, 431–443. Hinds LA, Cooke BD, Robinson TJ, Williams CK (1996) Rabbits — prospects for long term control: mortality and fertility control. A paper prepared for the Prime Minister’s Science and Engineering Council, 13 September 1996. http://www.dest.gov.au/archive/ Science/pmsec/14 meet/rcd1.html Hoare CA (1972) The Trypanosomes of Mammals. Blackwell Scientific Publications, Oxford and Edinburgh. Holliman RB (1966) A Trypanosoma lewisi-like organism from the rabbit Sylvilagus floridanus, in Virginia. Journal of Parasitology, 52, 622. Huchon D, Madsen O, Sibbald MJ et al. (2002) Rodent phylogeny and a timescale for the evolution of Glires: evidence from an extensive taxon sampling using three nuclear genes. Molecular Biology and Evolution, 19, 1053–1065. Jackson RJ, Hall DF, Kerr PJ (1996) Construction of recombinant myxoma viruses expressing foreign genes from different intergenic sites without associated attenuation. Journal of General Virology, 77, 1569–1575. Jakes KA, O’Donoghue PJ, Adlard RD (2001) Phylogenetic relationships of Trypanosoma chelodina and Trypanosoma binneyi from Australian tortoises and platypuses inferred from small subunit rRNA analyses. Parasitology, 123, 483–487. Kerr PJ, Jackson RJ (1995) Myxoma virus as a vaccine vector for rabbits: antibody levels to influenza virus haemagglutinin presented by a recombinant myxoma virus. Vaccine, 13, 1722–1726. Kerr PJ, Jackson RJ, Robinson AJ et al. (1999) Infertility in female rabbits (Oryctolagus cuniculus) alloimmunized with the rabbit zona pellucida protein ZPB either as a purified recombinant protein or expressed by recombinant myxoma virus. Biology of Reproduction, 61, 606–613. Kerr PJ, Perkins HD, Inglis B et al. (2004) Expression of rabbit IL-4 by recombinant myxoma viruses enhances virulence and overcomes genetic resistance to myxomatosis. Virology, 324, 117–128. © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 3167–3175

T H E A U S T R A L I A N R A B B I T T R Y P A N O S O M E 3175 Kroó H (1939) Die spontane, apathogene Trypanosomeninfektion der Kaninchen. Zeitschrift fur Immunologische Forschung, 88, 117. Launay H, Deunff J (1990) Spilotylenchus maisonabei n. sp. (Nematoda: Allantonematidae) parasite de Spilopsyllus cuniculi (Dale, 1878) (Siphonaptera: Pulicidae), puce oïoxène du lapin de garenne. Revue Nématologie, 13, 293 – 296. Mackerras MJ (1959) The haematozoa of Australian mammals. Australian Journal of Zoology, 7, 105 –135. Mackerras MJ (1961a) The haematozoa of Australian frogs and fish. Australian Journal of Zoology, 9, 123 –140. Mackerras MJ (1961b) The haematozoa of Australian reptiles. Australian Journal of Zoology, 9, 61–122. Mackerras MJ, Mackerras IM (1959) The haematozoa of Australian birds. Australian Journal of Zoology, 8, 226 – 260. Martin DS, Wright AD, Barta JR, Desser SS (2002) Phylogenetic position of the giant anuran trypanosomes Trypanosoma chattoni, Trypanosoma fallisi, Trypanosoma mega, Trypanosoma neveulemairei, and Trypanosoma ranarum inferred from 18S rRNA gene sequences. Journal of Parasitology, 88, 566 – 571. Maslov DA, Lukes J, Jirku M, Simpson L (1996) Phylogeny of trypanosomes as inferred from the small and large subunit rRNAs: implications for the evolution of parasitism in the trypanosomatid protozoa. Molecular and Biochemical Parasitology, 75, 197–205. McLysaght A, Baldi PF, Gaut BS (2003) Extensive gene gain associated with adaptive evolution of poxviruses. Proceedings of the National Academy of Sciences, USA, 100, 15655 –15660. McPherson MJ, Moller SG (2000) PCR the Basics from Background to Bench. BIOS Scientific Publishers, Oxford. Mohamed HA, Molyneux DH (1987) In vitro cultivation of Herpetosoma trypanosomes in insect cell tissue culture media. Parasitology Research, 73, 9–14. Mohamed HA, Molyneux DH, Scott CM (1987) Isoenzyme characterization of trypanosomes of the subgenus Herpetosoma. Parasitology, 94, 39– 48. Molyneux DH (1976) Biology of the trypanosomes of the subgenus Herpetosoma. In: Biology of the Kinetoplastida (eds Lumsden WHR, Evans DA), pp. 285 – 325. Academic Press, London. Myers K, Paper I, Richardson BJ (1989) Leporidae. In: Fauna of Australia (eds Walton DW, Richardson BJ). Australian Government Publishing Service, Canberra. Noyes H, Stevens JR, Teixeira M, Phelan J, Holz P (1999) A nested PCR for the ssrRNA gene detects Trypanosoma binneyi in the platypus and Trypanosoma sp. in wombats and kangaroos in Australia. International Journal for Parasitology, 29, 331– 339. O’Donoghue PJ, Adlard RD (2000) Catalogue of protozoan parasites recorded in Australia. Memoires of the Queensland Museum, 45, 1–163. Petrie GF (1905) Observations relating to the structure and geographical distribution of certain trypanosomes. Journal of Hygiene, 5, 191–200. Posada D, Crandall KA (1998) modeltest: testing the model of DNA substitution. Bioinformatics, 14, 817 – 818. Priddel D, Carlile N, Wheeler R (2000) Eradication of European rabbits (Oryctolagus cuniculus) from Cabbage Tree Island, NSW, Australia, to protect the breeding habitat of Gould’s petrel (Pterodroma leucoptera leucoptera). Biological Conservation, 94, 115–125. Smith AP, Quin DG (1996) Patterns and causes of extinction and decline in Australian conilurine rodents. Biological Conservation, 77, 243–267. © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 3167–3175

Sobey WR, Conolly D (1971) Myxomatosis: the introduction of the European rabbit flea Spilopsyllus cuniculi (Dale) into wild rabbit populations in Australia. Journal of Hygiene (Cambridge), 69, 331– 346. Sobey WR, Menzies W (1969) The introduction of the European rabbit flea Spilopsyllus cuniculi (Dale) into Australia. Australian Journal of Science, 31, 404–406. Springer MS, Murphy WJ, Eizirik E, O’Brien SJ (2003) Placental mammal diversification and the Cretaceous–Tertiary boundary. Proceedings of the National Academy of Sciences, USA, 100, 1056–1061. Stevens J, Noyes HA, Dover GA, Gibson WC (1999a) The ancient and divergent origins of the human pathogenic trypanosomes, Trypanosoma brucei and T. cruzi. Parasitology, 118, 107–116. Stevens J, Noyes HA, Schofield CJ, Gibson WC (2001) The molecular evolution of Trypanosomatidae. Advances in Parasitology, 48, 1–54. Stevens J, Teixeira MM, Bingle LE, Gibson WC (1999b) The taxonomic position and evolutionary relationships of Trypanosoma rangeli. International Journal for Parasitology, 29, 749–757. Stodart E, Parer I (1988) Colonisation of Australia by the rabbit Oryctolagus cuniculus (L.). CSIRO Division of Wildlife and Ecology, Project Report No. 6. CSIRO library catalogue number 101753, Melbourne, Australia. Swofford DL (2003) PAUP * Phylogenetic Analysis Using Parsimony (*and Other Methods). Sinauer Associates, Sunderland, Massachusetts. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The clustal_x Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research, 25, 4876–4882. Tyndale-Biscoe CH (1994) Virus-vectored immunocontraception of feral mammals. Reproductive Fertility and Development, 6, 281– 287. Wallace FG (1966) The trypanosomatid parasites of insects and arachnids. Experimental Parasitology, 18, 124–193. Watson EA, Hadwen S (1912) Trypanosomes found in Canadian mammals. Parasitology, 5, 21. Williams K, Parer I, Coman B, Burley J, Braysher M (1995) Managing Vertebrate Pests: Rabbits. Australian Government Publishing Service, Canberra.

This work formed part of P Hamilton’s PhD research on the diversity and evolution of trypanosomes under the supervision of W Gibson and J Stevens. J Stevens is a reader in molecular systematics, with ongoing work on the evolution and epidemiology of endo- and ectoparasites, and the population genetics, stock management and biogeography of Atlantic salmon and trout. P Holz is a wildlife veterinarian and has undertaken a wide range of research projects related to wildlife medicine. B Boag is a retired soil ecologist with continuing research interests in the biology and ecology of New Zealand flatworms, earthworms and soil nematodes; his research on the epidemiology of rabbit parasites was funded under a Leverhulme Grant awarded to Prof. Peter Hudson, Penn State University, USA and administered by Dr M Boots, Sheffield University, UK. B Cooke researches in the management of wildlife populations and the impact of introduced vertebrates on natural ecosystems, particularly on islands, with special interest in the control of Australian wild rabbits through the use of biological control agents. W Gibson is a professor of protozoology, with ongoing work on the genetics, evolution and epidemiology of African trypanosomes, and tsetse-trypanosome interactions.