Parasitology International 61 (2012) 461–465

Contents lists available at SciVerse ScienceDirect

Parasitology International

journal homepage: www.elsevier.com/locate/parint

Pinning down a polymorphic parasite: New genetic and morphological descriptions of macropodis from the Tammar wallaby (Macropus eugenii)

Nichola J. Hill, Carolin Richter, Michelle L. Power ⁎

Department of Biological Sciences, Macquarie University, North Ryde, NSW 2109, Sydney, Australia article info abstract

Article history: Identification of the protozoan parasite, Eimeria has traditionally relied on oocyst morphology, host range and Received 25 January 2012 life-cycle attributes. However, it is increasingly recognized that Eimeria species can vary in size and shape Received in revised form 7 March 2012 across their host range, an attribute known as ‘polymorphism’ that presents a unique challenge for identifi- Accepted 16 March 2012 cation. Advances in molecular tools hold promise for characterising Eimeria that may otherwise be misclassi- Available online 23 March 2012 fied based on morphology. Our study used morphologic and molecular traits of the oocyst life stage to identify a polymorphic parasite, Eimeria macropodis in a captive Tammar wallaby (Macropus eugenii) popula- Keywords: tion in Australia. Molecular characterization highlighted the need to use multiple genetic markers (18S SSU and cytochrome c oxidase subunit I) to accurately identify E. macropodis owing to heterozygous alleles at Species identification the 18S SSU locus. This study provided an opportunity to assess the utility and shortcomings of morphologic Cytochrome c oxidase subunit I gene and molecular techniques for ‘pinning down’ a polymorphic species. Moreover, our study was able to place 18S small subunit ribosomal RNA E. macropodis in an evolutionary context and enhance resolution of the under-studied marsupial clade. Marsupial © 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction and molecular characteristics of the parasite. Eimeria species with comparable size and shape and broad host range have proven to be Parasites belonging to the Eimeria genus are widespread among distinct species under molecular scrutiny [6]. Another challenge to hosts, with over 1700 species described (http://biology. accurate identification is polymorphism, whereby Eimeria species unm.edu/biology/coccidia/marsup.html). Some Eimeria species pose display an array of shapes and sizes depending on the host [10]. Mo- a significant threat to livestock animals, particularly chickens in lecular approaches are increasingly used to characterise the identity which outbreaks involving high morbidity and mortality occur global- of Eimeria and compliment earlier morphological descriptions [11]. ly [1,2]. Host specificity is thought to be a hallmark of the Eimeria Genetic information not only contributes to a more stable genus. A single species typically infects one genus and the host range of Eimeria but also provides insights into the evolutionary relation- is usually restricted to evolutionarily related hosts. Exceptions to this ships of this ubiquitous parasite. rule occur, as shown by field surveys that document a single Eimeria Studies of Eimeria have focused on domestic animals while Eimeria parasite infecting hosts of different species [3,4] or even generas from wildlife are under-represented, despite their value in clarifying [5,6]. As with parasites that have a resilient oocyst stage, Eimeria has evolutionary origins of the parasite. Marsupials represent an intrigu- evolved to persist in the external environment. The faecal–oral trans- ing taxon of hosts that underwent radiation in South America and mission mode of parasites is conducive to host-switching events [7,8]. Australia after their separation from eutherian mammals more than This may be common among hosts that share habitat and come into 100 million years ago [12]. For this reason, marsupials play an impor- contact with oocysts shed by sympatric species [6]. There is a clear tant role as an evolutionary outgroup, critical for resolving the early need to better understand the host–parasite associations of Eimeria evolutionary trajectory of coccidian parasites [13]. A total of 54 Eimeria using molecular tools to ensure accurate species identification. species have been described from marsupials (45 from Australia and 9 Identification of most Eimeria species has traditionally relied on from the Americas), however only one has been genetically character- oocyst morphology, host range and life-cycle attributes [9]. These ized to date, E. trichosuri [11]. In this study, we sought to improve earlier studies laid the foundation for a classification of Eimeria. How- the taxonomic resolution of the parasite by contributing molecular ever, revisions to the taxonomy of Eimeria have occurred with ad- and morphological data on Eimeria shed by the Tammar wallaby vances in diagnostic tools that shed light on novel morphometric (Macropus eugenii). Comparison of molecular and morphological data provided an opportunity to assess the utility and shortcomings ⁎ Corresponding author. Tel.: +61 2 9850 6974; fax: +61 2 9850 8245. of both approaches for identifying a polymorphic species with a wide E-mail address: [email protected] (M.L. Power). host range, Eimeria macropodis.

1383-5769/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.parint.2012.03.003 462 N.J. Hill et al. / Parasitology International 61 (2012) 461–465

2. Materials and methods were continuous variables and therefore the Euclidean metric was used to calculate ‘straight-line’ distances between clusters. The opti- 2.1. Sample collection mal number of clusters occurring in the population was determined by Akaike's Information Criterion (AIC). Faecal samples were collected from a captive population of Secondly, we sought to assess whether oocysts from this study Tammar wallabies at the Macquarie University Fauna Park (33°46′06″ were significantly different from original descriptions of E. macropodis S, 151°06′47″ E). A total of 45 samples were collected between May by Barker et al. [3]. A multivariate ANOVA was performed to compare and July 2007 from a yard housing juvenile females less than 12 months oocyst shapes and sizes from this study with E. macropodis from nine old. The Fauna Park was a source of isolates upon which original de- species of wallaby and kangaroo [3]. The six oocyst traits (Table 1) scriptions of some Eimeria species (E. macropodis and E. bicolor) were were included as dependent variables. All statistics were performed based [3]. Returning to this study site provided a unique opportunity using SPSS 16.0 software (SPSS Ltd, Chicago). to compare attributes of the parasite over a 30-year period. Faecal samples were stored in 2% (w/v) potassium dichromate 2.4. DNA extraction at room temperature to allow oocysts to sporulate. Samples were screened for the presence of Eimeria oocysts using flotation over Genomic DNA extraction was performed following methods de- saturated zinc sulphate and microscopy [14]. Sporulated oocysts scribed by Power et al. [11].Briefly, 10 μl of purified oocysts was crushed were purified from faeces using a modified sucrose flotation method with a mini-pestle (Astral Scientific, Gymea, Australia) and added to 1 μl [15] and stored in sterile water at 4 °C. of prepGEM enzyme (20 mg/ml: ZyGEM Corporation, Hamilton, New Zealand) and 1 μl of lysozyme (5 mg/ml: Sigma-Aldrich, Castle Hill, 2.2. Morphological description Australia). DNA extraction followed prepGEM protocols with modifi- cations for oocyst wall lysis. Oocysts were examined using an oil-immersion 100× objective with an Olympus BH2 epifluorescent microscope (Ziess) and differen- 2.5. PCR amplification, cloning and sequencing tial interference contrast system. Images were captured using a Nikon DXM digital camera and measurements recorded (in μm) using Samples were screened at the 18S SSU rRNA locus using a hemi- ImageJ software (http://rsbweb.nih.gov/ij/). Oocyst and sporocyst length, nested PCR. The first reaction was performed in 50 μl volume con- width and shape (length: width) from sporulated oocysts (n=43) taining GoTaq Green 2× Mastermix (Promega, Alexandria, Australia) were documented (Table 1). The species was identified as E. macropodis and 0.2 mM each of the forward (18sF1) and reverse primer [18sR2: based on similarities with descriptions from Barker et al., [3]. both from Ref. 16].Ampliconswereusedasatemplateforasecond Scanning electron microscopy (SEM) was used to visualize fea- reaction with the 18sF1 forward primer and reverse primer EMR2— tures of the oocyst wall. Oocysts were fixed with 3% glutaraldehyde TGAGTTTCCCCGTGTTGAGT (this study). buffered in 0.1 M phosphate buffer and post-fixed with 1% osmium Amplification at the cytochrome c oxidase subunit I (COI) locus tetroxidate in 0.1 M phosphate buffer. Gradient dehydration was per- followedmethodsdescribedbyOgedengbeetal.[27].The50μlre- formed (50% ethanol for 10 min, followed by 70%, 80% and 95% for actions contained GoTaq Green polymerase and 1 mM each of the 10 min each and two steps of 100% for 15 min). Oocysts were placed forward (Cocci_COI_For) and reverse primer (Cocci_COI_Rev). inside a critical point dryer (Emitech K850, FDI Fabsurplus, TX) to en- PCR products were purified using the Qiagen spin column PCR sure all fluids were removed. Splutter coating with a 20 nm gold layer purification kit (Qiagen, Hilden Germany). Purified products were was applied to specimens (Emitech K550, FDI Fabsurplus, TX). ligated into the pCR2.1 vector using the TA Cloning® kit (Invitrogen, Images were taken with the JSM-6480 LA Analytical Scanning Electron Victoria) and positive transformants were cultured overnight in Microscope. liquid Luria–Bertani broth. The plasmid DNA was extracted using the QIAprep Spin Miniprep Kit (Qiagen, Victoria) in preparation for 2.3. Morphometric analyses sequencing. The 18S SSU clones were sequenced using three primers, 18SF1, 18SF2 and 18SR7 [11] to generate three overlapping fragments To identify any differences in size and shape among oocysts, a approximately 1195 bp in length. The COI clones were sequenced two-step cluster analysis was performed. This explorative approach using 18SF1 and 18SR2 to generate a single fragment approximately was taken because oocyst lengths were variable and associated with 780 bp in length. Sequencing was performed with an ABI 3130xl a relatively high standard deviation (3.00: Table 1). Cluster analysis Genetic Analyzer with the BigDye terminator kit (Applied Biosystems, provided a cursory assessment of whether oocysts could be separated Foster City, California). into homogeneous groups based on six oocyst traits (Table 1). Traits 2.6. Phylogenetic analysis

Table 1 Morphometric measurements (μm) of Eimeria oocysts from the Tammar wallaby (n=43) Sequences were assembled into optimised contigs using Geneious compared to E. macropodis pooled from 9 species of kangaroo and wallaby. Bioinformatics software (version 5.5.3, Biomatters Ltd, New Zealand). Alignments were constructed using default setting in ClustalW [17], E. macropodis E. macropodis (pooled) implemented in Geneious. Phylogenetic trees of the 18S gene included This study Barker et al. [3] 44 partial sequences from GenBank selected to represent major lineages Oocyst length Mean (S.D.) 24.39 (3.00) 24.92 (1.74) of the parasite shed by birds, mammals and reptiles. Alignments were Range 19.00–30.00 21.10–27.60 Oocyst width Mean (S.D.) 13.73 (1.30) 13.20 (0.63) trimmed to a length of 1129 bp. The COI phylogenetic trees were Range 11.00–17.00 11.90–14.20 based on only 15 reference sequences as only Eimeria species from poul- Oocyst length:width Mean (S.D.) 1.78 (0.17) 1.89 (0.12) try and rodents were available in GenBank. Alignments were trimmed – – Range 1.27 2.08 1.73 2.08 to 682 bp. Both trees included 2 outgroup sequences from Toxoplasma Sporocyst length Mean (S.D.) 8.88 (1.33) 9.02 (0.57) Range 6.00–11.00 8.00–9.80 gondii and falciparvum. The nucleotide sequences identi- Sporocyst width Mean (S.D.) 6.26 (0.73) 6.33 (0.51) fied in this study were submitted to GenBank under the accession Range 4.00–7.00 5.60–7.50 numbers JQ392574 to JQ392580. Sporocyst length:width Mean (S.D.) 1.45 (0.18) 1.43 (0.08) A best-fit nucleotide substitution model was selected using – – Range 1.04 1.87 1.31 1.53 MrModeltest 2.3 (Nylander, J.A.A. 2004. MrModeltest v2. Program N.J. Hill et al. / Parasitology International 61 (2012) 461–465 463 distributed by author. Evolutionary Biology Centre, Uppsala University). 3.2. Phylogenetics For both genes, the model with the lowest AIC corrected score was the General Time Reversible (GTR) model accounting for estimates Phylogenetic analysis indicated two types of 18S SSU alleles. of invariable sites (I) and the gamma distribution parameter (G). E. macropodis isolates formed two lineages, EM6–EM15 (98.7% iden- Phylogenetic trees were generated with a Bayesian Markov chain tity) and EM10 and EM11 (99.6% identity). The EM10–EM11 lineage Monte Carlo (MCMC) coalescent approach using Mr Bayes 3.2 [18] differed due to an insertion of ‘ATAT’ at 703–706 bp and ‘TATA’ at implemented in Geneious. A MCMC chain length of 5 million gener- 729–731 bp (Supplementary Fig. 1). Consensus trees generated with ations sampled every 1000 iterations with 4 chains, two runs and a MrBayes supported the separation of E. macropodis into two lineages burn-in value set to 100,000 was selected. The consensus tree was vi- with high statistical support (Fig. 2A). However, analysis of the COI sualized in FigTree 1.3.1 [19]. gene indicated EM10, EM11 and EM15 shared 99.9% similarity (Table 2) and originated from the same species (EM6 was unable to be amplified). Bayesian trees placed EM10, EM11 and EM15 into a 3. Results single monophyletic group (Fig. 2B). This provided evidence of two copies of the 18S SSU gene in E. macropodis from the Tammar wallaby. 3.1. Oocyst morphology E. macropodis showed highest identity with E. trichosuri from pos- sums at the 18S SSU locus (96.7–97.1%), consistent with the high sim- Eimeria oocysts were identified in 13 of 45 samples collected from ilarity indicated by the COI gene (91.7% identity). Bayesian trees the Tammar wallaby enclosure, based on screening using zinc floata- placed E. macropodis and E. trichosuri in a monophyletic group, hence- tion. Prevalence was not estimated as samples could not be linked to forth referred to as the ‘marsupial clade’ (Fig. 2A and B). Bayesian an individual host. Oocysts were identified by their oval shape, thick trees of the 18S SSU gene indicated the marsupial clade formed a sis- cell wall and presence of four sporocysts (Fig. 1A). The 8 samples ter clade with E. reichenowi and E. gruis from waterbirds and E. arnyi with more than 20 oocysts were concentrated and used for molecular from reptiles was ancestral to these hosts (Fig. 2A). Both the 18S analysis. SSU and COI genes supported divergence of the marsupial clade The shape of the oocysts was ellipsoidal, slightly pointed at one before the large monophyletic group that comprised Eimeria from end, with a double wall approximately 0.07 μm thick consisting of a eutherian mammals and poultry. The COI consensus tree revealed a yellow outside layer (0.03–0.05 μm) and a colourless inside layer large number of polytomies, possibly owing to the low diversity of (0.02–0.03 μm). Images from SEM revealed a smooth wall with no hosts represented. As many lineages showed unresolvable evolution- protrusions (Fig. 1B). Micropyle equivocal, polar granules present ary relationships, the topology of this tree should be viewed with and oocyst residuum absent. Four ellipsoidal sporocysts with Stieda caution. bodies, no substieda body, two sporozoites filling sporocysts and conspicuous sporocyst residuum. Based on morphological similari- 4. Discussion ties with Barker et al [3] we classified the parasite as E. macropodis (Table 1). A total of 54 species of Eimeria have been described from marsu- Oocyst size and shape were homogeneous across the 43 samples pials and the majority have been identified based on morphology of investigated in this study, as indicated by cluster analysis. The cluster sporulated oocysts and host range. This has created challenges for ac- profile indicated one cohesive group, suggesting oocysts were mor- curate identification, owing to the overlap in size and shape of oocysts phologically similar and belonged to the same Eimeria species. In that belong to different species. Alternatively, some Eimeria species addition, oocysts did not differ significantly from those described by are polymorphic and display an array of shapes and sizes depending Barker et al. [3] in terms of oocyst length (F = 0.18, df= 1, P= 0.67), on the host. In a study of E. opimi, oocysts shed by 5 species of Bolivian oocyst width (F =1.64, df = 1, P= 0.21), oocyst shape (F = 3.34, rodents showed plasticity in size and shape resulting in 55% of sam- df=1, P=0.07), sporocyst length (F=0.09, df=1, P=0.76), sporo- ples being classified as the wrong species [20]. Our study used mor- cyst width (F=0.09, df=1, P=0.76) and sporocyst shape (F=0.11, phological and molecular characteristics of the oocyst life stage to df =1, P =0.74). This supported our classification of oocysts from identify the presence of E. macropodis in a captive Tammar wallaby the Tammar wallaby as E. macropodis. population. E. macropodis oocysts occurred in small and large forms

Fig. 1. A) Sporulated oocysts of Eimeria macropodis containing four ellipsoidal sporocysts, and B) oocyst wall showing the smooth outer surface (with mucilaginous fungi attached). 464 N.J. Hill et al. / Parasitology International 61 (2012) 461–465

A even within this single host, causing us to initially suspect the pres- ence of multiple species. Molecular analysis of the 18S SSU and COI 1 genes, coupled with statistical analysis of oocyst traits, confirmed 1 the occurrence of a single polymorphic species. 1 Our study highlights that molecular analysis of a single gene may 1 not be adequate to uncover the full genetic diversity within an 1 0.97 Eimeria species. The use of two genetic markers (18S SSU and COI)

1 was the key to the discovery of heterozygous alleles in E. macropodis 1 1 in this study. Analysis of 18S SSU indicated the presence of at least two 0.99 alleles that may have been mistaken for distinct Eimeria species, 1 0.98 without confirmation of heterozygosity by an alternate genetic marker. Validity of Eimeria species from chickens; E. mitis and E. mivati, has recently been questioned in the context of heterozygous alleles [21]. 0.99 E. mivati has been attributed to the presence of a second 18S SSU gene that is periodically isolated from E. mitis samples [21]. These studies emphasise the importance of large sample sizes to capture sequence 1 variants, as well as cloning techniques to separate out alleles that may 1 coexist in a single species. Studies of Eimeria rarely report heterozygous alleles compared to other apicomplexan parasites including, Plasmodium 1 1 [22,23] and [24,25]. It has been hypothesized that 0.99 gene variants play a role in the various stages of the apicomplexan par- 1 asite life cycle [21,26] and this needs to be considered when selecting genes for molecular identification. By genetically characterizing E. macropodis our study was able to 1 1 1 Marsupial clade place the parasite in an evolutionary context and enhance resolu- 1 tion of the marsupial clade. Analyses of partial 18S and COI gene se-

1 quences placed E. macropodis within the marsupial clade consisting of E. trichosuri from mountain brushtail possums. Posterior probabili- 1 caninum AJ271354 ties from Bayesian analysis indicated a high level of confidence in this U12138 topology (18S: 0.98, COI: 0.99). The marsupial clade diverged earlier than Eimeria from eutherian mammals including rodents and bats, 0.1 reflecting the trajectory of mammalian evolution [11]. Our second B marker, the mitochondrial COI gene has gained popularity for classi- 1 fying apicomplexan parasites. Studies comparing the utility of the

0.98 18S and COI genes indicate the latter has higher resolving power for Eimeria, especially recent speciation events [27]. COI has become the 1 0.98 target gene for the Barcode of Life project that aims to use the marker for rapid identification of animals, including parasites [28]. One draw- back of using this gene in the context of wildlife studies is the paucity of Eimeria sequences available for hosts other than poultry. Our study contributes to the species definition of E. macropodis, paving the way for more accurate and reliable identification of the parasite in marsupials. This parasite has a broad host range including nine species of kangaroo and wallaby belonging to the Macropus

1 genus, based on oocyst morphometrics [3]. We speculate that grazing may predispose this genus to fecal–oral transmission accounting for 1 the broad host range of Eimeria from this cohort of hosts, compared to primarily arboreal species such as possums that host Eimeria species that are more host-specific[seeRef.:29]. E. macropodis has also been detected across a large geographic region, spanning all six states of Australia from both free-ranging and captive populations, as well as a zoo in the Ukraine [30]. The ubiquity of E. macropodis across 1 host species and geographic scales makes accurate identification of paramount importance. This challenge is complicated by the poly- morphic nature of E. macropodis that may lead to misclassification and underestimation of true host range. Techniques that consider 1 ‘ ’ Marsupial clade the parasite as a moving target with dynamic morphology and multi- 1 ple life cycles offer the best chance of identifying and studying Eimeria in host populations.

Neospora caninum HM771688 1 Toxoplasma gondii HM771690 Fig. 2. Phylogenetic trees based on Bayesian analysis of A) 18S SSU and B) cytochrome c 0.1 oxidase subunit I (COI) sequences. Posterior probabilities are indicated when higher than 0.95. Isolates of Eimeria macropodis are encompassed by the marsupial clade (highlighted in grey). N.J. Hill et al. / Parasitology International 61 (2012) 461–465 465

Table 2 Pairwise similarity (%) of 18S SSU sequences (18S: upper block) and cytochrome c oxidase subunit I sequences (COI: lower block) among selected coccidian taxa.

Supplementary data to this article can be found online at doi:10. [13] Power ML. Biology of Cryptosporidium from marsupial hosts. Experimental Parasi- tology 2010;124:40–4. 1016/j.parint.2012.03.003. [14] Daugschies A, Imarom S, Bollwahn W. Differentiation of porcine Eimeria spp. by morphologic algorithms. Veterinary Parasitology 1999;81:201–10. Acknowledgements [15] Truong Q, Ferrari BC. Quantitative and qualitative comparisons of Cryptosporidium faecal purification procedures for the isolation of oocysts suitable for proteomic analysis. International Journal for Parasitology 2006;36:811–9. The authors wish to thank John Barta and Joseph Ogedengbe [16] Zhao X, Duszynski DW. Molecular phylogenies suggest the oocyst residuum can of the University of Guelph for assistance with analysing the COI be used to distinguish two independent lineages of Eimeria spp in rodents. – gene. Funding for this project was provided through the Macquarie Parasitology Research 2001;87:638 43. [17] Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, University Research Fellowship Scheme and an Australian Academy et al. Clustal W and Clustal X version 2.0. Bioinformatics 2007;23:2947–8. of Science Travel Grant. [18] Huelsenbeck JP, Ronquist F. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 2001;17:754–5. [19] Rambaut A. FigTree—tree figure drawing tool. Accessed 14 Nov 2011: http://tree. References bio.ed.ac.uk/software/figtree. [20] Gardner SL, Duszynski DW. Polymorphism of eimerian oocysts can be a problem [1] Sun XM, Pang W, Jia T, Yan WC, He G, Hao LL, et al. Prevalence of Eimeria species in in naturally infected hosts: an example from subterranean rodents in Bolivia. – broilers with subclinical signs from fifty farms. Avian Diseases 2009;53:301–5. Journal of Parasitology 1990;76:805 11. [2] Aarthi S, Dhinakar Raj G, Raman M, Gomathinayagam S, Kumanan K. Molecular [21] Vrba V, Poplstein M, Pakandl M. The discovery of the two types of small subunit prevalence and preponderance of Eimeria spp. among chickens in Tamil Nadu, ribosomal RNA gene in Eimeria mitis contests the existence of E. mivati as an inde- – India. Parasitology Research 2010;107:1013–7. pendent species. Veterinary Parasitology 2011;183:47 53. [3] Barker IK, O'Callaghan MG, Beveridge I. Host-parasite associations of Eimeria [22] McCutchan TF, de la Cruz VF, Lal AA, Gunderson JH, Elwood HJ, Sogin ML. Primary spp. (Apicomplexa: ) in kangaroos and wallabies of the genus Macropus sequences of two small subunit ribosomal RNA genes from Plasmodium – (Marsupialia: Macropodidae). International Journal for Parasitology 1989;19: falciparum. Molecular and Biochemical Parasitology 1988;28:63 8. 241–63. [23] Qari SH, Goldman IF, Pieniazek NJ, Collins WE, Lal AA. Blood and sporozoite fi [4] Zhao X, Duszynski DW. Phylogenetic relationships among rodent Eimeria species stage-speci c small subunit ribosomal RNA-encoding genes of the human malaria – determined by plastid ORF470 and nuclear 18S rDNA sequences. International parasite . Gene 1994;150:43 9. Journal for Parasitology 2001;31:715–9. [24] Hill NJ, Deane EM, Power ML. Prevalence and genetic characterization of [5] Heckscher SK, Wickesberg BA, Duszynski DW, Gardner SL. Three new species of Cryptosporidium isolates from common brushtail possums (Trichosurus vulpecula) Eimeria from Bolivian marsupials. International Journal for Parasitology 1999;29: adapted to urban settings. Applied and Environmental Microbiology 2008;74: – 275–84. 5549 55. [6] Zhao X, Duszynski DW, Loker ES. Phylogenetic position of Eimeria antrozoi, a bat [25] Le Blancq SM, Khramtsov NV, Zamani F, Upton SJ, Wu TW. Ribosomal RNA gene coccidium (Apicomplexa: Eimeriidae) and its relationship to morphologically organization in . Molecular and Biochemical Parasitology – similar Eimeria spp. from bats and rodents based on nuclear 18S and plastid 23S 1997;90:463 78. rDNA sequences. Journal of Parasitology 2001;87:1120–3. [26] Corredor V, Enea V. The small ribosomal subunit RNA isoforms in Plasmodium – [7] Dick CW, Patterson BD. Against all odds: explaining high host specificity in cynomolgi. Genetics 1994;136:857 65. fi dispersal-prone parasites. International Journal for Parasitology 2007;37:871–6. [27] Ogedengbe JD, Hanner RH, Barta JR. DNA barcoding identi es Eimeria species and [8] Pedersen AB, Altizer S, Poss M, Cunningham AA, Nunn CL. Patterns of host speci- contributes to the phylogenetics of coccidian parasites (Eimeriorina, Apicomplexa, – ficity and transmission among parasites of wild primates. International Journal Alveolata). International Journal for Parasitology 2011;41:843 50. for Parasitology 2005;35:647–57. [28] Ratnasingham S, Hebert PD. BOLD: the barcode of life data system. (http://www. – [9] Duszynski DW, Wilber PG. A guideline for the preparation of species descriptions barcodinglife.org). Molecular Ecology Notes 2007;7:355 64. in the Eimeriidae. Journal of Parasitology 1997;83:333–6. [29] O'Callaghan MG, O'Donoghue PJ. A new species of Eimeria (Apicomplexa: Eimeriidae) [10] Joyner LP. Host and site specificity. In: Long PL, editor. The biology of coccidia. from the brushtail possum, Trichosurus vulpecula (Diprotodontia: Phalangeridae). – Baltimore: University Park Press; 1982. p. 35–62. edn. Transactions of the Royal Society of South Australia 2001;125:129 32. [11] Power ML, Richter C, Emery S, Hufschmid J, Gillings MR. Eimeria trichosuri: [30] Yakimoff WL, Matschoulsky SN. Coccidiosis in the kangaroo. Journal of Parasitology – phylogenetic position of a marsupial coccidium, based on 18S rDNA sequences. 1936;22:514 5. Experimental Parasitology 2009;122:165–8. [12] Clemens WA. Origin and early evolution of marsupials. Evolution 1968;22:1–18.