Genetic variation in Austrostrongylus thylogale Johnston & Mawson, 1940 (Nematoda: Trichostrongylida) from the tammar wallaby, Notamacropus eugenii (Gray), and the quokka, Setonix brachyurus (Quoy & Gaimard) (Marsupialia: Macropodidae) in Australia

Background Australian marsupials harbour a diverse array of helminth parasites. Despite current attempts to assess the extent of this diversity in macropodid hosts, it has been suggested that unique parasite fauna of Australian wildlife is difficult to document comprehensively due to the common occurrence of cryptic species. This paper assessed genetic variation within Austrostrongylus thylogale Johnston & Mawson, 1940 from the tammar wallaby, Notamacropus eugenii (Gray), and the quokka, Setonix brachyurus (Quoy & Gaimard), from different localities using the molecular characterisation of the internal transcribed spacers (ITS) within the nuclear ribosomal DNA. Methods Thirty-seven specimens of A. thylogale collected from N. eugenii (from Parndana, Kangaroo Island, South Australia, and Perup, Western Australia) and S. brachyurus (from Wellington Dam, Western Australia) were characterised using a molecular-phylogenetic approach utilising the first (ITS1) and second (ITS2) internal transcribed spacers. Results Genetic variation was detected in both ITS1 and ITS2 between specimens of A. thylogale from N. eugenii and S. brachyurus; however, no variation was detected between specimens collected from N. eugenii from Parndana, South Australia, and Perup, Western Australia. Furthermore, the phylogenetic analyses of ITS sequences showed two clades of A. thylogale originating from two hosts, N. eugenii and S. brachyurus, suggesting the presence of cryptic species. Conclusions This study provides evidence of genetic variation within A. thylogale based on collections from two different host species. Morphological studies are required to fully confirm the presence of a new species or cryptic species. Further molecular studies using a larger number of specimens are warranted to explore the genetic variation between A. thylogale from different geographical localities.


Background
The Australian continent is biologically diverse, and native animals such as kangaroos and wallabies (family Macropodidae) harbour a diverse range of unique nematodes [1]. The abundance and extensive range of parasites has been attributed to the host-specific nature of the helminths found in macropodids and their herbivorous grazing habits [2,3]. Given the large numbers of nematodes, host specificity, and variable size and morphology of worms, uncommon species may often be overlooked [3][4][5].
While macropodid nematodes are diverse, they are well known for having simple body plans and morphological uniformity, which can make it difficult to distinguish between species based on morphology alone [6]. As a result, molecular methods such as DNA-based techniques have been developed to enable the definition and identification of genetic markers for the characterisation of various species [7]. A commonly used genetic method in the identification of such nematodes is the sequencing of the first and second internal transcribed spacers (ITS1 and ITS2, respectively) of the nuclear ribosomal DNA, which are known to be reliable genetic markers due to the homology of these sequences between species [7,8]. This molecular technique has been used in many studies to distinguish between morphologically similar species as well as cryptic (i.e. genetically distinct but morphologically similar) species [7,9,10].
The trichostrongylid genus Austrostrongylus Chandler, 1924 contains 13 recognised species, which occur in the duodenum of various macropodid hosts with an aberrant species in the marsupial mole (Notoryctes typhlops (Stirling)) [11][12][13]. Austrostrongylus thylogale Johnston & Mawson, 1940, was first described from tammar wallabies (Notamacropus eugenii (Gray)) on Kangaroo Island, South Australia [14] and was subsequently reported from quokkas (Setonix brachyurus (Quoy & Gaimard)) on Rottnest Island, Western Australia by Inglis [15]. Notamacropus eugenii occurs on Kangaroo Island, South Australia and, in a highly disjunct distribution, in the south-west of Western Australia including islands such as Garden Island. The population of S. brachyurus is distributed in the south-west of Western Australia, as well as on two offshore islands, Rottnest Island and Bald Island [16]. The distribution of N. eugenii is therefore sympatric with S. brachyurus in the south-west of Western Australia. A recent study analysing the ITS sequences of mainland and island populations of the strongylid nematode Labiosimplex australis Kung, 1948 suggested that genetic variation within this species was possibly due to geographical isolation on islands or between geographically distant locations [17]. Therefore, we hypothesised that similar genetic differences might also be present in populations of A. thylogale as it has a disjunct geographical distribution and occurs in different host species [6]. This paper aimed to assess the genetic variation in A. thylogale collected from different localities and host species by characterising the ITS region of the nematode.

Collection of specimens
Adult specimens of A. thylogale (n = 37) were sourced from the frozen parasite collection at the School of Veterinary Science, The University of Melbourne. These specimens had been collected opportunistically from culled or road-killed hosts from various localities in Australia ( Table 1). The nematodes were frozen in liquid nitrogen and stored at − 80 °C until use. Upon thawing, the anterior and posterior ends were excised, placed in lactophenol and stored for later morphological analysis.
Amplicons were purified using shrimp alkaline phosphate and exonuclease I [18] prior to automated Sanger DNA sequencing at Macrogen Incorporation, South Korea, using the same primers used in the PCR.

Sequence and phylogenetic analyses
The quality of each forward and reverse sequence was assessed, and each consensus sequence assembled using Geneious Prime 2019.0.4 (www.genei ous.com). Polymorphic sites were designated using the International Union of Pure and Applied Chemistry codes. All nucleotide sequences obtained were deposited in the Gen-Bank (Table 1). Nucleotide sequences were aligned using MUSCLE v.3.8.31 [19] within MEGA-X using default settings [20]. Pairwise comparisons of aligned sequences were performed to calculate nucleotide differences using BioEdit [21].
Unique ITS1, ITS2 and concatenated ITS (designated as ITS+) sequences collected in this study were aligned with respective reference sequences using MUSCLE in MEGA using default settings and were trimmed to uniform lengths of 356 bp (ITS1), 240 bp (ITS2) and 596 (ITS+) bp. The evolutionary models for each DNA sequence dataset [ITS1 and ITS+: Hasegawa-Kishino-Yano (HKY); ITS2 Tamura-Nei (TrN)] were determined using the Akaike and the Bayesian information criteria (AIC and BIC) tests in jModelTest v.3.7 [22]. Neighbourjoining (NJ) trees were constructed using MEGA, and Bayesian Inference (BI) trees were built using MrBayes [23]. The NJ trees were constructed with 10,000 bootstrap replicates using the Tamura-Nei distance method. Each BI analysis was run for 20,000,000 generations to calculate posterior probabilities (pp), with two runs, with every 200th tree saved. Sequences from Austrostrongylus victoriensis Cassone, 1983 were included as the outgroup. Tree topology was checked for consensus between NJ and BI analyses using the software Figtree v1.4.4 (http:// tree.bio.ed.ac.uk/softw are/figtr ee/).

Results and discussion
Following the PCR amplification of the ITS region, 14 amplicons were selected for DNA sequencing. Six and five unique sequences of ITS1 (length: 399 bp) and ITS2 (length: 240 bp) were identified, respectively. The G+C content of ITS1 sequences of A. thylogale was higher in specimens collected from S. brachyurus (42.6%) than those obtained from N. eugenii (41.5-41.6%), with no difference based on location. However, the G+C content of ITS2 sequences was higher in specimens of A. thylogale  Table 2; see Additional file 1: Figure  S1). Based on within-host species pairwise comparisons, higher nucleotide variation was observed in specimens of A. thylogale collected from S. brachyurus (0.3-2.3%) than those collected from N. eugenii (0.3-1.8%).
Nucleotide differences of 0.5-1.7% from pairwise comparisons of the ITS2 sequences of A. thylogale from the Parndana population of N. eugenii, were mainly attributed to nucleotide polymorphisms (two nucleotide positions: 2 (Y) and 148 (R)), two transitions (T ↔ C) and two transversions (T ↔ A or A ↔ C) ( Table 3; see Additional file 1: Figure S1), whereas sequences from S. brachyurus were identical. The nucleotide variability between specimens of A. thylogale from N. eugenii and S. brachyurus was linked to one transition (T ↔ C) ( Table 3; see Additional file 1: Figure S1).
Phylogenetic analyses of separate ITS sequences using NJ and BI methods revealed that the topology of both trees generated (separately) for each dataset (ITS1, ITS2 and ITS+ sequences) were similar using BI and NJ (data not shown); hence, only the NJ tree of the ITS+ sequences is presented here (Fig. 1). The ITS sequences of A. thylogale from N. eugenii originating from Parndana (South Australia) and Perup (Western Australia) grouped together with strong statistical support (posterior probability of BI = 1.00, bootstrap value of NJ = 87%) whereas those from S. brachyurus formed the separate clade, also with strong statistical support (1.00, 86%) (Fig. 1). To our knowledge, this is the first report of the genetic characterisation of the ITS+ DNA of A. thylogale and findings of this study support our hypothesis that specimens of A. thylogale collected from two different hosts i.e. N. eugenii and S. brachyurus are genetically distinct, suggesting the existence of cryptic species. Similar findings have been reported for a number of other macropodid nematodes [17,[24][25][26][27][28].
We found little variation in the ITS sequences of A. thylogale specimens collected from N. eugenii from two distant geographical locations (Parndana, South Australia, and Perup, Western Australia); however, there were four (3 in ITS1; 1 in ITS2) fixed nucleotide differences between specimens collected from N. eugenii and S. brachyurus (see Tables 2 and 3). Furthermore, the overall pairwise nucleotide differences and G+C content were also different between the specimens of A. thylogale originating from the two different host species. The ITS loci are known to be reliable genetic markers for the characterisation of strongylid nematodes as both ITS1 and ITS2 sequences can be quite conservative and only one or more fixed differences can assist in assessing interspecific differences [7,29]. Therefore, the findings of this study suggest that A. thylogale may represent more than one species. However, this hypothesis needs to be tested using the morphological and molecular examination of a large number of specimens. Preliminary morphological examination of specimens from the two host species revealed no obvious differences.
Phylogenetic analysis of the ITS1, ITS2 and ITS+ sequences consistently revealed that A. thylogale from N. eugenii and S. brachyurus formed two separate clades (Fig. 1), and individual sequences from the same host species clustered together, suggesting this grouping is more strongly associated with the host rather than the locality.
Species of Austrostrongylus occur predominantly in the related wallaby genera Notamacropus Dawson & Flannery, Dorcopsis Schlegel & Müller and Wallabia Trouessart [1] although transmission occasionally occurs to other sympatric macropodid hosts although at a lower prevalence and intensity [30]. In this instance, it appears to be a host switch from N. eugenii to S. brachyurus. This hypothesis is supported by the sympatric distribution of N. eugenii and S. brachyurus in Western Australia [16]. Studies have shown that most macropodid nematodes undergo speciation via host-switching due to overlapping host-ranges and feeding habits [25,26,31]. In the present example, apart from infecting a new host species, the parasite appears to have undergone a degree of genetic differentiation within the new host. In this study, the genetic variation in A. thylogale may represent another example of cryptic species as the specimens genetically characterised herein had very similar morphological features [14]. The existence of cryptic species amongst nematodes in macropodid hosts has been detected previously [17,32,33]. For example, Chilton et al. [24] reported four sibling species within Cloacina petrogale Johnston & Mawson, 1938 using multilocus enzyme electrophoresis. Subsequently, they used DNA sequence data to demonstrate the existence of cryptic species within four morphospecies of the genus Cloacina that occurred within rock wallabies of the genus Petrogale Gray. More recently, Chilton et al. [26] proposed 10 cryptic species within the phascolostrongyline nematode, Hypodontus macropi Mönnig, 1929 from 12 macropodid hosts. Although the reasons for the apparent common occurrence of sibling species in cloacinine and phascolostrongyline nematodes are not clear, the relatively rapid radiation of the hosts [34], as well as the opportunities for host switching, may contribute to this phenomenon [35]. However, the confirmation of these inferences requires further studies with larger sample size.

Conclusions
A molecular-phylogenetic approach using ITS sequences revealed genetic variation in 37 specimens of A. thylogale from N. eugenii and S. brachyurus from different locations. Phylogenetic analyses of the sequence data revealed distinct clades based on hosts. Additional morphological and molecular studies are required to fully confirm the presence of a new species or cryptic species of Austrostrongylus.