Introduction

Coccidian parasites are known to infect potoroid marsupials, including the critically endangered woylie or brush-tailed bettong Bettongia penicillata Gray. Although morbidity and mortality associated with coccidial infection in free-ranging macropods is uncommon (Vogelnest & Portas, 2008), disease has been documented in macropods under stress (e.g. the eastern grey kangaroo Macropus giganteus Shaw; Barker et al., 1972). Given the reliance of many threatened species on interventional management practices such as translocation, a process which has been identified as a significant stressor (Hing et al., 2017), it is imperative that we gain a greater understanding of the parasite species infecting wildlife, particularly those with the potential to cause disease in their host (e.g. coccidian parasites).

To date, no coccidian parasites (Apicomplexa: Eimeriidae) have been formally described in woylies (Duszynski, 2016), though Eimeria sp. oöcysts have been detected in faecal samples (Northover et al., 2017). Five Eimeria spp. including Eimeria potoroi Barker, O’Callaghan & Beveridge, 1988, Eimeria mundayi Barker, O’Callaghan & Beveridge, 1988, Eimeria gaimardi Barker, O’Callaghan & Beveridge, 1988, Eimeria aepyprymni Barker, O’Callaghan & Beveridge, 1988, and Eimeria burdi Hulst, Kemp & Slapeta, 2016, have been morphologically described from potoroid marsupials (Marsupialia: Potoroidae). Eimeria gaimardi from the eastern bettong Bettongia gaimardi (Desmarest), most closely morphologically resembles Eimeria sp. oöcysts found in woylies. Unfortunately, genetic characterisation of potoroid Eimeria spp. has not been undertaken. While Eimeria spp. tend to be host-specific (Barker et al., 1988), members of the genus Macropus Shaw have been known to harbor the same Eimeria spp. (e.g. Eimeria macropodis Wenyon & Scott, 1925; see Barker et al., 1989), thus it is important that additional data (e.g. genetic characterisation) is used to support the description of new species.

During this study we aimed to (i) morphologically describe and genetically characterise E. woyliei n. sp. from woylies; and (ii) genetically characterise E. gaimardi from the eastern bettong, and E. potoroi and E. mundayi from the long-nosed potoroo Potorous tridactylus (Kerr).


Materials and methods


Sample collection

Between 2014 and 2018, woylie faecal samples (n = 1,093) were collected from various sites within south-western Australia (Table 1) as part of a collaborative project with the Department of Biodiversity, Conservation and Attractions (DBCA), Kensington, Australia. Newspaper was placed beneath each trap to collect faeces, which were stored in 70% ethanol, 10% buffered formalin and/or 2% potassium dichromate until processing.

Table 1 The number of woylies screened for coccidian parasites within each sampling site
Full size table

In 2018, faecal samples collected from the eastern bettong (n = 4) were obtained from a captive population at Bonorong Wildlife Sanctuary in Brighton, Tasmania (42.71°S, 147.27°E). Samples from the long-nosed potoroo (n = 12) were acquired from wild-caught animals within the Peter Murrell reserves, south of Hobart, Tasmania (43.00°S, 147.18°E). Faecal samples from the eastern bettong and long-nosed potoroo were collected directly from traps and samples were stored in 2% potassium dichromate prior to analysis.


Identification of coccidian oöcysts in faecal samples

For woylies, the majority of faecal samples (n = 1,073) were examined for the presence of coccidian oöcysts using simple faecal flotation with sodium nitrate (NaNO3) as described by Northover et al. (2017). These samples were formalin-fixed. To describe the morphology of E. woyliei n. sp., an additional 20 samples were sporulated (using potassium dichromate) and underwent faecal flotation using zinc sulphate (ZnS04) in distilled water (SG 1.20). Briefly, faeces were placed into a 10 ml centrifuge tube (up to the 1.5–2.0 ml mark), emulsified in distilled water (tube filled to the 10-ml mark) and centrifuged (2,000 rpm for 2 min). The supernatant was removed before filling the tube with zinc sulphate solution and re-emulsifying, before final centrifugation (2,000 rpm for 2 min). A sterile wire loop was used to transfer 2–3 loops from the surface of the tube to a glass slide, and a 22 × 22 mm coverslip was placed on top. Each sample was examined at 100× magnification using an Olympus BX50 microscope. All other potoroid samples were examined using zinc sulphate. To calculate the prevalence of infection, each faecal sample was scored as positive or negative for the presence of coccidian oöcysts.


Morphological description of the new species

Eighty-six sporulated oöcysts from a single woylie originating from Perup Sanctuary were examined at magnifications of 400–1,000×, using an Olympus BX50 microscope with a Olympus DP71 Universal Camera with Cellsens software. Photographs of sporulated oöcysts were taken using bright field microscopy. Measurements of oöcyst and sporocyst length and width, and oöcyst wall thickness were acquired using ImageJ software (US National Institute of Health, Bethesda, Maryland). All measurements are recorded in micrometres (µm) with the range followed by the mean in parentheses. We measured a single laterally positioned sporocyst within each oöcyst; if we could not identify a sporocyst in the correct position, we did not measure sporocyst length or width.


Morphological identification of other potoroid Eimeria spp.

Three distinct morphotypes of sporulated oöcysts were identified from the faeces of the eastern bettong and long-nosed potoroo. Based on their size and unique oöcyst and sporocyst characters as described by Barker et al. (1988) we identified E. gaimardi from the eastern bettong, and E. mundayi and E. potoroi from the long-nosed potoroo.


Genetic characterisation of Eimeria spp. in potoroid marsupials

For the woylie, nine faecal samples (eight ethanol- and one potassium dichromate-preserved) were used to genetically characterise E. woyliei n. sp. at the 18S rDNA and mitochondrial cytochrome c oxidase subunit 1 (cox1) loci. Eimeria gaimardi, E. mundayi and E. potoroi were genetically characterised using a single faecal sample (potassium dichromate-preserved) for each Eimeria species; as outlined above, morphological identification of sporulated oöcysts confirmed the identity of each species prior to genetic characterisation.


DNA extraction and PCR amplification

Samples stored in 70% ethanol and/or 2% potassium dichromate were exposed to four freeze/thaw cycles as described by Yang et al. (2016a) in order to achieve oöcyst lysis. Following the freeze/thaw step, faecal samples stored in 2% potassium dichromate were subjected to a wash step prior to lysis in order to remove the fixative, by centrifuging at 3000 rpm for 10 min and resuspending in phosphate buffered solution. DNA was isolated from 0.25 g of faecal sample using the PowerFecal DNA Isolation Kit (MolBio, Carlsbad, California) as per the manufacturer’s instructions. A negative control was included to rule out contamination.

Faecal samples from the woylie were screened at the 18S rDNA locus using a nested PCR with the external Eimeria spp. primers EiF1 (5′-GCT TGT CTC AAA GAT TAA GCC-3′) and EiR3 (5′-ATG CAT ACT CAA AAG ATT ACC-3′), and the internal Eimeria spp. primers EiR4 (5′-ACT CAA AAG ATT ACC TAG AC-3′) and EiF4 (5′-CTA TGG CTA ATA CAT GCG CAA T-3′) (Yang et al., 2016a). PCR reactions were carried out in a total volume of 25 µl containing 12.5 µl of 2× KAPA HiFi Hotstart ReadyMix (Millennium Science Pty. Ltd, Victoria, Australia), 0.75 µl primer (10 µM) and 2 µl DNA template. All PCR reactions were performed as described by Yang et al. (2016a) consisting of a pre-PCR step of 94°C for 3 min, followed by 45 cycles of 94°C for 30 s, 55°C annealing temperature for 30 s and 72°C for 2 min, and a final extension step of 72°C for 5 min.

Faeces from the eastern bettong and long-nosed potoroo were screened using the 18S external primers EiGTF1 (5′-TTC ACT GGT CCC TCC GAT C-3′) and EiGTR1 (5′-AAC CAT GGT AAT TCT ATG G-3′) (Yang et al., 2016b), and the internal primers EiGTF2 (5′-TTA CGC CTACTA GGC ATT CC-3′) and EiTR2 (5′-TGA CCT ATC AGC TTT CGA CG-3′). The PCR reaction contained 10 µl 2× GoTaq PCR master mix (Promega, Alexandria NSW, Australia), 1 µl DNA (50 ng), 10 µl of each primer (10 µM stock) and 7 µl distilled water. PCR cycling conditions for the external PCR were 1 cycle of 94°C for 3 min, followed by 35 cycles of 94°C for 30 s, 58°C for 30 s and 72°C for 2 min, and a final extension step of 72°C for 5 min. The conditions for the secondary PCR were the same except for 45 cycles instead of 35. All amplicons were visualised on a 1.5% agarose gel and bands were cut and purified using an in-house filter tip method defined by Yang et al. (2013).

For all potoroid species, amplicons were generated at the cox1 locus using the external primers COIF1 (5′-GGT TCA GGT GTT GGT TGG AC-3′) and COIF2 (5′-TAA GTA CAT CCC TAA TGT C-3′) and the internal primers COIbR1 (5′-CCA AGA GAT AAT ACR AAR TGG AA-3′) and COIbR2 (5′-ATA GTA TGT ATC ATG TAR WGC AA-3′) (Yang et al., 2016a). The PCR reactions and conditions were the same as per the 18S PCR carried out for woylie samples.


Sequencing and phylogenetic analysis

Purified amplicons were sequenced using an ABI PrismTM Dye Terminator Cycle Sequencing Kit (Applied Biosystems, California, USA) according to the manufacturer’s instructions. Samples were sequenced in the forward and reverse direction and the denaturation step was extended to 10 min to allow efficient primer binding. The sequences were analysed in Geneious v.8.1 (Kearse et al., 2012) and aligned using reference libraries with the MUSCLE (Edgar, 2004) plugin for Geneious v.8.1. All novel sequences were deposited in GenBank.

Using Bayesian methods, phylogenetic analyses were conducted for both gene regions (18S and cox1, respectively), to determine the evolutionary lineage for E. woyliei n. sp. (MK182524; MK202806), E. gaimardi (MK182525; MK202809), E. mundayi (MK182526; MK202808) and E. potoroi (MK182527; MK202807). JModelTest (Posada, 2008) was used to determine the most appropriate nucleotide substitution method for the Bayesian analyses for each gene region, which was the GTR+I+G (general time reversible gamma proportion of invariant sites) method. The Bayesian posterior probabilities were generated using MrBayes v.3.1.2 (10,000,000 generations, sampling frequency of 1,000, ‘burn in’ 3,000).


Results


The prevalence of coccidial infection in woylies is summarised by site in Table 1. It is important to note that Dryandra contains both resident woylies (endemic to the region), and translocated woylies originating from the Upper Warren region (specifically Balban, Boyicup, Corbal, Dudijup, Dwalgan and Winnejup). Likewise, Walcott and Warrup East contain both resident and translocated (originating from Perup Sanctuary) woylies. In the eastern bettong, 2 out of 4 (50%) samples were positive for E. gaimardi. In the long-nosed potoroo, 3 out of 12 (25.0%) samples were positive for E. potoroi, while 4 out of 12 (33.3%) samples were positive for E. mundayi; mixed infections with both coccidian parasites were identified in 2 out of 12 (16.7%) samples.


Family Eimeriidae Minchin, 1903

Genus Eimeria Schneider, 1875


Eimeria woyliei n. sp.


Type-host: Bettongia penicillata Gray (Mammalia: Marsupialia: Potoroidae), woylie or brush-tailed bettong.

Type-locality: Perup Sanctuary (34.16°S, 116.56°E) in the Upper Warren region, Western Australia, Australia.

Other localities: Corbal (34.10°S, 116.48°E), Dwalgan (34.07°S, 116.46°E) and Winnejup (34.07°S, 116.35°E) in the Upper Warren region, Western Australia, Australia; *Dryandra Woodland (32.80°S, 116.89°E) in the wheatbelt region, Western Australia, Australia (*translocated host, origin Dwalgan).

Type-material: Oöcysts in 10% formalin and oöcyst photosyntypes have been deposited in the Western Australian museum [reference numbers WAM Z91161 (holotype) and WAM Z91162-64 (paratypes)].

Prevalence: 21.2% (32 out of 151 specimens of the type-host; origin Perup Sanctuary).

Sporulation: Unknown.

Site of infection: Unknown, oöcysts recovered from faeces.

Representative DNA sequences: DNA sequences have been deposited in the GenBank sequence database under accession numbers MK182524-MK182527 (18S rRNA gene) and MK202806-MK202809 (cox1 gene).

Etymology: The name woyliei reflects the host species local name ‘woylie’, which is the Aboriginal name given to this animal by the Noongar people of south-western Western Australia (Abbott, 2001).

Description (Figs. 1,2A–C)

Sporulated oöcyst

Fig. 1
figure 1

Composite line drawing of a sporulated oöcyst of Eimeria woyliei n. sp.

Full size image
Fig. 2
figure 2

Photomicrographs of sporulated oöcysts of Eimeria woyliei n. sp. (A–C) and a sporulated oöcyst of Eimeria gaimardi (D). Abbreviations: PG, polar granule; MP, micropyle; RB, refractile body; SB/SBB, Stieda body/sub-Stieda body; SMB, submicropyle body; SP, sporocyst; SR, sporocyst residuum. Scale-bars: 10 µm

Full size image

Oöcysts (n = 86) pyriform-shaped [length 31.6–40.8 (36.7), width 22.6–31.0 (26.3), length/width (L/W) ratio 1.3–1.5 (1.4)] with smooth to slightly mammilate, bi-layered oöcyst wall, 1.3–1.7 (1.6) thick. Oöcyst wall thins at apex with visible micropyle. An irregular dome-like structure (referred to here as a submicropyle body) visible beneath micropyle. Polar granule irregular-shaped, apparent, although not detected in all sporulated oöcysts. Oöcyst residuum absent.


Sporocyst and sporozoite

Sporocysts (n = 76) ellipsoidal [length 13.3–17.8 (15.6), width 9.0–12.3 (10.3), L/W ratio 1.1–1.7 (1.5)]. Each sporocyst contains an indistinct dome-like Stieda body/sub-Stieda body. Sporocyst residuum present between 2 broadly elongate sporozoites. Sporozoites contain 2 distinct refractile bodies, one large (5–7 wide) and one small (up to 3 wide). Para-Stieda body absent.


Genetic characterisation of Eimeria spp. in potoroids


The phylogeny of the four potoroid Eimeria spp. was investigated using Bayesian analyses at two gene loci (18S and cox1). An alignment was generated for the 18S rDNA locus (1,158 bp, Fig. 3) and two alignments for the cox1 locus (686 bp, Fig. 4; 211 bp, Fig. 5). All alignments contained the novel sequences belonging to the four potoroid species as well as reference sequences downloaded from GenBank including an outgroup (Toxoplasma gondii). Similar phylogenetic relationships were observed for both loci, although some species were not observed in the cox1 tree due to lower availability of genetic data. A second cox1 alignment was generated (211 bp) in order to incorporate relevant marsupial species of smaller fragment size.

Fig. 3
figure 3

Phylogenetic relationships of Eimeria woyliei n. sp. with other Eimeria spp. using Bayesian analysis of an 1,158 bp fragment of the 18S rRNA gene. Eimeria spp. isolates obtained in this study (GenBank: MK182524-MK182527) are highlighted in bold

Full size image
Fig. 4
figure 4

Phylogenetic relationships of Eimeria woyliei n. sp. with other Eimeria spp. using Bayesian analysis of the large 686 bp fragment of the cox1 gene. Eimeria spp. isolates obtained in this study (GenBank: MK202806-MK202809) are highlighted in bold

Full size image
Fig. 5
figure 5

Phylogenetic relationships of Eimeria woyliei n. sp. with other Eimeria spp. using Bayesian analysis of the smaller 211 bp fragment of the cox1 gene. Eimeria spp. isolates obtained in this study (GenBank: MK202806-MK202809) are highlighted in bold

Full size image

18S rDNA locus

Eimeria woyliei shared 99.2% genetic similarity with E. setonicis, isolated from the quokka Setonix brachyurus (Quoy & Gaimard) (KF225639; Austen et al., 2014), and 99.1% genetic similarity with E. trichosuri, from the mountain brushtail possum Trichosurus cunninghami Lindenmayer, Dubach & Viggers (FJ8292320; Power et al., 2009). Of the potoroid Eimeria spp., E. woyliei shared 98.8% genetic similarity with E. mundayi and E. potoroi, and 98.4% similarity with E. gaimardi.


cox1 locus

Using the longer (686 bp) cox1 locus, E. woyliei shared 97.1% similarity with E. trichosuri (JN192136; Ogedengbe et al., 2015), 95.7% similarity with E. potoroi, 95.3% similarity with E. gaimardi and 95.0% similarity with E. mundayi. Using the shorter (211 bp) cox1 locus, E. woyliei shared 94.8% similarity with Eimeria kanyana Bennett, Woolford, O’Hara, Nicholls, Warren & Hobbs, 2006, from the quenda Isoodon obesulus (Shaw) (KU845563/64; Hillman, 2016), and 97.2%, 96.7% and 95.3% similarity with E. potoroi, E. gaimardi and E. mundayi, respectively.


Discussion


Five species of Eimeria have been previously described in potoroid marsupials (Barker et al., 1988; Hulst et al., 2016). Of these, E. woyliei oöcysts most closely resemble E. gaimardi oöcysts from the eastern bettong in shape and size (range: 31.6–40.8 × 22.6–31.0 vs 32.0–39.2 × 20.8–26.4 µm; mean: 36.7 × 26.3 vs 34.6 × 24.3 µm) (see Barker et al., 1988). Sporocyst size of E. woyliei and E. gaimardi is also similar (range: 13.3–17.8 × 9.0–12.3 vs 13.6–16.0 × 9.0–10.4 µm; mean: 15.6 × 10.3 vs 15.0 × 9.6 µm). However, oöcysts and sporocysts in E. woyliei are slightly larger on average. Eimeria woyliei oöcysts also contain a micropyle, submicropyle body and a polar granule; E. gaimardi oöcysts do not according to the original morphological description by Barker et al. (1988). Based on the sporulated E. gaimardi oöcysts examined during this study however, we propose that E. gaimardi oöcysts do possess a micropyle, submicropyle body and polar granule (Fig. 2D). Thus, the major distinguishing morphological characteristic between E. woyliei and E. gaimardi is the appearance of the oöcyst wall, which for E. gaimardi appears mammilated and more robust (Fig. 2D). The oöcyst wall of E. woyliei in comparison is smooth to slightly mammilate.

Eimeria woyliei oöcysts are noticeably larger than E. potoroi oöcysts [24.0–29.6 × 16.8–22.4 (26.2 × 18.5) µm] (Barker et al., 1988) and E. mundayi oöcysts [13.6–20.8 × 13.6–19.2 (16.9 × 16.2) µm] from the long-nosed potoroo (Barker et al., 1988). Eimeria woyliei oöcysts are comparable in size to E. aepyprymni oöcysts [32.0–42.8 × 18.4–25.2 (36.7 × 21.9) µm] from the rufous bettong Aepyprymnus rufescens (Gray) (Barker et al., 1988); however, E. aepyprymni oöcysts are ovoid rather than pyriform. While similar in shape (pyriform), E. woyliei oöcysts are considerably larger than E. burdi oöcysts [21.0–24.0 × 14.0–16.0 (22.6 × 14.9) µm] from the burrowing bettong Bettongia lesueur (Quoy & Gaimard) (Hulst et al., 2016).

Using the 18S rDNA locus, E. woyliei was grouped within the marsupial clade and was most similar to E. setonicis from the quokka. Eimeria woyliei oöcysts can be morphologically distinguished from oöcysts of E. setonicis by shape (pyriform vs ellipsoidal) (Austen et al., 2014). Despite the morphological similarity between oöcysts of E. woyliei and E. gaimardi, E. woyliei was more genetically similar to E. mundayi and E. potoroi, rather than E. gaimardi, though this difference was minimal.

Using the cox1 locus, E. woyliei was most similar to E. trichosuri from the mountain brushtail possum and grouped within the same clade as E. macropodis from the tammar wallaby Macropus eugenii (Desmarest) (Hill et al., 2012), and E. kanyana from the quenda (Hillman et al., 2016). Eimeria woyliei oöcysts can be differentiated from oöcysts of E. trichosuri by shape (pyriform vs ellipsoidal; Power et al., 2009); E. trichosuri oöcysts are also larger [34.4–49.2 × 18.4–27.8 (41.4 × 22.7) µm] (O’Callaghan & O’Donoghue, 2001). Unexpectedly, Isospora amphiboluri Canon, 1967, from the central bearded dragon Pogona vitticeps (Ahl) was also present within the same phylogenetic clade (GenBank: KR108297). Although cox1 sequences are deemed superior to nuclear 18S rDNA sequences for delimiting species (Ogedengbe et al., 2011), cox1 sequences on GenBank are limited. Nonetheless the supposed relationship between E. woyliei and I. amphiboluri remains unclear.

Within Australia various coccidial species (Eimeria and less commonly Isospora) have been recorded in captive and free-ranging macropods (Mykytowycz, 1964; Barker et al., 1989; Yang et al., 2012; Duszynski, 2016). Traditional morphological criteria have been useful for identifying coccidial species infecting wildlife. However, the implementation of genetic characterisation combining both 18S rDNA and cox1 loci helps to discriminate between morphologically similar species and provides an accurate measure of the evolutionary relationship between coccidial species (Power et al., 2009; Yang et al., 2012). As some Eimeria spp. are capable of infecting more than one marsupial host (Barker et al., 1989) and marsupials tend to harbour multiple Eimeria spp. (Power et al., 2009), this knowledge may be useful for predicting potential avenues of disease spread during the management of threatened populations (e.g. during fauna translocation). This study contributes toward our knowledge of Eimeria spp. infecting potoroid marsupials. Eimeria woyliei parasitising woylies is the sixth Eimeria spp. to be formally described from potoroid marsupials and we have genetically characterised four of the six known potoroid Eimeria species.