Molecular and morphological characterisation of Pharyngostrongylus kappa Mawson, 1965 (Nematoda: Strongylida) from Australian macropodid marsupials with the description of a new species, P. patriciae n. sp.

Background Pharyngostrongylus kappa Mawson, 1965 is a nematode (Strongyloidea: Cloacininae), endemic to the sacculated forestomachs of Australian macropodid marsupials (kangaroos and wallaroos). A recent study revealed genetic variation within the internal transcribed spacer region of the nuclear ribosomal DNA among P. kappa specimens collected from Macropus giganteus Shaw and Osphranter robustus (Gould). This study aimed to characterise the genetic and morphological diversity within P. kappa from four macropodid host species, including M. giganteus, O. robustus, O. antilopinus (Gould) and O. bernardus (Rothschild). Methods Specimens of P. kappa from M. giganteus and Osphranter spp. from various localities across Australia were examined. The first and second internal transcribed spacers (ITS1 and ITS2, respectively) were amplified using polymerase chain reaction and sequenced. Phylogenetic methods were used to determine the interspecific diversification within P. kappa and its evolutionary relationship with other congeners. Results Morphological examination revealed that P. kappa from M. giganteus, the type-host, can be distinguished from those in Osphranter spp. by the greater length and number of striations on the buccal capsules. DNA sequences showed that P. kappa from M. giganteus was genetically distinct from that in Osphranter spp., thereby supporting the morphological findings. Based on these finding, a new species from Osphranter spp., Pharyngostrongylus patriciae n. sp., is described. Conclusion Pharyngostrongylus patriciae n. sp. from Osphranter spp. is distinguished from P. kappa based on molecular and morphological evidence. The study highlights the importance of combining molecular and morphological techniques for advancing the nematode taxonomy. Although ITS genetic markers have proven to be effective for molecular prospecting as claimed in previous studies, future utilisation of mitochondrial DNA to validate ITS data could further elucidate the extent of speciation among macropodid nematodes. Electronic supplementary material The online version of this article (10.1186/s13071-018-2816-6) contains supplementary material, which is available to authorized users.


Background
Kangaroos and wallabies (family Macropodidae) are an important component of the Australian endemic fauna [1], and owing to their grazing habits, they harbour an abundant and extensive range of strongyloid helminths in their gastrointestinal tracts [2,3]. Macropodid helminths are highly host specific [4], with at least 372 species described so far [3,[5][6][7]. The primary focus of the research into macropodid helminths has been quantifying biodiversity and assessing the prevalence and distribution of these parasites [3]. Unlike helminths of livestock, pathogenic effects, host-parasite interactions and life-cycle stages of the helminth fauna of macropodid marsupials are not well understood [3,8]. Nonetheless, current knowledge suggest that the most outstanding feature observed in helminths from macropodid hosts is their remarkable diversity, particularly among the nematodes [2,3,6,9].
The subfamily Cloacininae Stossich, 1899 (Nematoda: Strongylida) comprising 36 genera and 256 species, is one of the most morphologically diverse groups of mammalian parasites [3]. All members of the Cloacininae occur exclusively in the oesophagus and forestomach of macropodid and the potoroid marsupials. York & Maplestone [10] established Pharyngostrongylus Yorke & Malplestone, 1926 for a single nematode species characterised by transverse striations on the buccal capsule, Pharyngostrongylus macropodis Yorke & Malplestone, 1926 found in the stomach of the agile wallaby Notamacropus agilis (Gould). Subsequently, additional species with these characteristics were discovered in the stomachs of other macropodid hosts [11]. Presently, 14 species of Pharyngostrongylus are known, with P. thylogale Chilton, Huby-Chilton, Gasser, Koehler & Beveridge, 2016 the most recent addition to the genus [12].
Pharyngostrongylus kappa Mawson, 1965 occurs in the eastern grey kangaroo, Macropus giganteus Shaw, throughout its entire distribution along the eastern coast of Australia. However, it also occurs in the common wallaroo, Osphranter robustus (Gould), the antilopine wallaroo, O. antilopinus (Gould), and the black wallaroo, O. bernardus (Rothschild), in the northern parts of Australia. Pharyngostrongylus kappa is morphologically distinguished from other congeners by the presence of an elongated buccal capsule with transverse striations and eight petaloid labial crown elements surrounding the apical opening [11]. A recent study of the phylogeny of the ITS region within Pharyngostrongylus found that three specimens of P. kappa obtained from M. giganteus in Victoria and O. robustus from Queensland and the Northern Territory were genetically distinct from one another [12]. These data suggest the existence of genetic variation within P. kappa. Given that occurrences of distinct genetic species in different hosts have been observed in many species within the subfamily Cloacininae [13][14][15][16][17], it was hypothesised that P. kappa from different macropodid hosts and geographical localities might exhibit genetic and possibly morphological variation. Therefore, this study aimed to characterise the genetic and morphological variation among P. kappa from the different macropodid hosts and localities within Australia. The study included the remaining species of Pharyngostrongylus known from wallaroos, namely P. papillatus Beveridge, 1982 and P. sharmani Beveridge, 1982.

Specimen collection
Adult specimens of Pharyngostrongylus spp. (n = 173) used in this study were made available from the frozen parasite collection at the Veterinary School of The University of Melbourne. These worms had been collected from the stomachs of culled or road-killed macropodid marsupials, including eastern grey kangaroos (M. giganteus), common wallaroos (O. robustus), antilopine wallaroos (O. antilopinus) and black wallaroos (O. bernardus) ( Fig. 1; Table 1).

Morphological characterisation
Prior to examination, the frozen specimens (at -80°C) were thawed and individual worms were cut into three pieces. The anterior and posterior ends were cleared in lactophenol, mounted on slides and three morphologically distinct species of Pharyngostrongylus (P. kappa, P. papillatus and P. sharmani) were identified ( Table 1). The mid-body portion of each worm was placed in an Eppendorf tube with 0. 5 ml of H 2 O, labelled and frozen at -80°C until required for molecular analysis.
Additionally, specimens of P. kappa in the collections of the South Australian Museum (SAM), Adelaide were also examined (see Additional file 1: Table S1). Ten males and ten female worms from different hosts and localities were selected for morphological measurements where available. The selected nematodes were cleared in lactophenol, mounted on slides and measurements were made using an ocular micrometer. Measurements are presented as the range and the mean in millimetres. Morphological drawings were made with the aid of a drawing tube attached to the Olympus BH-2/BHS Systems microscope. Photographs of the buccal capsules were taken with an attached microscope digital camera, Olympus DP21. The terminology used in the morphological description follows Beveridge [11] while nomenclature of hosts and parasites species follows Spratt & Beveridge [18].
Discriminant function analyses (DFA) were performed separately on male (n = 44) and female (n = 50) measurements using Microsoft Excel with XLSTAT [19]. Predictor variables included length, oesophageal length, buccal capsule length, buccal capsule width and spicule length for males. The same predictor variables were used in analysis for the females but included tail length and distance from the vulva to the posterior extremity instead of spicule length.

Molecular characterisation
To detect intraspecific genetic variation among specimens of P. kappa from different hosts and geographical localities, the first and second internal transcribed spacers (ITS1 and ITS2, respectively) were amplified using polymerase chain reaction (PCR). The ITS regions of P. papillatus and P. sharmani were also included in the molecular analyses to determine the phylogenetic relationship of P. kappa to other congeners.
Prior to DNA extraction, frozen mid-body segments of individual nematodes were thawed and rinsed three times in H 2 O. Genomic DNA was extracted using minicolumns (Wizard® SV Genomic DNA purification Kit, Promega, Madison, WI, USA) according to the manufacturer's recommendations. The concentration and purity of each extracted DNA sample were determined using a Nano-Drop spectrophotometer, and then stored in Eppendorf tubes at 4°C until required.
For each spacer region, amplicons were treated with shrimp alkaline phosphatase and exonuclease I [21] and then subjected to automated DNA sequencing using the 96-capillary 3730xl DNA Analyser, (Applied Biosystems, Foster City, CA, USA) at Macrogen Inc., South Korea. Sequencing of ITS1 and ITS2 was conducted using the specific PCR primers in separate reactions. The quality of nucleotide sequences was appraised using the program Geneious R10 (Biomatters Ltd., Auckland, New Zealand) [22], and polymorphic sites were designated using International Union of Pure and Applied Chemistry (IUPAC) codes. Voucher morphological specimens representing each of the genotypes found in the analysis were deposited in SAM (Table 1).

Phylogenetic analysis
The ITS1 and ITS2 sequences of each individual nematode generated in this study were compared with those of P. kappa previously published from M. giganteus (Genoa, Vic; GenBank: LT576294) and O. robustus (Charters Towers, Qld and Edith River, NT; GenBank: LT576296 and LT576295) [12]. Prior to phylogenetic analysis, the sequences were aligned using the program MEGA 7.0.26 [23] by employing multiple sequence comparison by log-expectation (MUSCLE) [24]. The alignments were then adjusted manually using the program BioEdit [25]. Pairwise comparisons to determine the relatedness among sequences were conducted using MEGA.
Phylogenetic analyses were performed on individual (ITS1 or ITS2) or concatenated ITS1 and ITS2 (designated as ITS+) sequence datasets using the Neighbour-Joining (NJ) and Bayesian Inference (BI) methods. Based on the sequence data, the most suitable likelihood parameters and evolutionary models were determined using the Akaike information criteria test in the program jModeltest version 3.7 [26]. jModeltest revealed that the most suitable model for the ITS1, ITS2 and ITS+ were TPM1uf+I, TPM3uf and GTR+I+G, respectively. The BI analysis was conducted, employing the Markov Chain Monte Carlo (MCMC) method in MrBayes 3.1.2 [27,28]. Posterior probabilities were calculated using 2,000,000 generations, employing four simultaneous tree-building chains, with every 100th tree being saved. A consensus tree (50% majority rule) was constructed based upon the remaining trees generated by BI. The NJ analyses were performed employing MEGA and the nodes were tested for robustness with 10,000 bootstrap replicates. The ITS sequences of Cloacina ernabella Johnston & Mawson, 1938 were used as the outgroup. The topology of each tree was examined for concordance between the NJ and BI analysis in the program FigTree. Male and female nematodes from the different macropodid hosts were similar in length and width ( Table 2). The average lengths of the specimens from M. giganteus were slightly greater than those of the specimens from Osphranter spp. (male: 8.25 vs 5.45-7.46 mm; female: 9.52 vs 6.90-9.30 mm). Other morphological features such as the distances of the nerve-ring, deirids and excretory pore to the anterior extremity fell within the same range in specimens examined from all hosts. Lengths of the male spicules also overlapped in specimens from each host. However, the length of the buccal capsule of P. kappa from M. giganteus was greater (0.12-0.17 mm) than from Osphranter spp. (0.08-0.13 mm) ( Table 2). Counting of the number of transverse striations on buccal capsules revealed that the specimens from M. giganteus had 60-64 striations, whereas those from Osphranter spp. had 38-50 striations ( Table 2).
The discriminant analyses showed that the first two functions (i.e. F1 and F2) combined were able to correctly assign 98.51% of the male specimens ( Fig. 2a) and 98.84% of female specimens (Fig. 2b). In both sexes,

Remarks
The new species described here from O. robustus is attributed to Pharyngostrongylus based on its elongated and straight-walled buccal capsule with prominent transverse striations and a labial crown composed of petaloid elements. Pharyngostrongylus patriciae n. sp. can be differentiated from all congeners apart from P. kappa by the straight-sided, elongate buccal capsule with prominent transverse striations, eight labial crown elements and the lack of flap-like valves at the junction of the buccal capsule with the oesophagus. It is differentiated from P. kappa by the shorter buccal capsule, with 38-50 transverse striations instead of 60-64 striations in P. kappa.

Molecular characterisation
Following sequencing, the flanking regions of the ITS sequences were trimmed based on sequences from Gen-Bank [12] and the duplicate sequences were removed. A total of 28 and 29 unique sequences of ITS1 and ITS2 were identified, respectively. The trimmed sequences revealed that the ITS1 sequences of P. kappa (n = 15) and   The same superscripts for GenBank IDs indicate identical sequences Fig. 4 Phylogenetic analysis of the ITS1 rDNA sequences of Pharyngostrongylus kappa and Pharyngostrongylus patriciae n. sp. from various host species and geographical locations. The sequence data were analysed using the Neighbour-Joining (NJ) and Bayesian Inference (BI) methods. There was a concordance between the topology of the BI tree and the NJ tree (not shown). Nodal support is given as a posterior probability of BI/bootstrap value for NJ. Each unique sequence is presented with a GenBank accession no. followed by the voucher number, and its host and locality. Three sequences (LT576294-LT576296) were included as reference sequences from Chilton et al. [12]. Cloacina ernabella was used as the outgroup. Scale-bar indicates the number of inferred substitutions per nucleotide site. Abbreviations: NSW, New South Wales; NT, Northern Territory; QLD, Queensland; stn, station; VIC, Victoria; WA, Western Australia P. patriciae n. sp. (n = 14) ranged between 379-380 base pairs (bp) and 377-379 bp, respectively ( Table 3). The ITS2 sequences of P. kappa (n = 17) ranged between 219-223 bp while those of P. patriciae n. sp. (n = 12) were 222 bp long ( Table 3). The ITS1 and ITS2 sequences of both P. kappa and P. patriciae n. sp. determined herein were aligned (separately) with those previously published by Chilton et al. [12] over 384 and 227 positions, respectively. Specimen XDR10.1 (GenBank: MG972145, MG972146) from M. giganteus, Inglewood, Queensland was chosen as the reference sequence as it is from the type-host and locality of P. kappa (Additional file 1: Figures S2 and S3). The ITS1 sequences from M. giganteus had fewer variations than those from Osphranter spp. Contrarily, the ITS2 sequences from Osphranter spp. exhibited greater variation compared to those from M. giganteus. However, there were consistent differences in nucleotide bases between individuals from M. giganteus and those from Osphranter spp. in both ITS1 and ITS2 regions. The only exceptions were the ITS2 sequences MG972110 and MG972112 from M. giganteus at Clermont, Queensland which shared the same sequences as specimens from Osphranter spp.
Overall pairwise nucleotide differences in the ITS2 sequences were higher (0.5-8.5%) than those of ITS1 (0.3-4.5%) ( Table 3). The greater difference was attributed to the high similarity (97-99%) between ITS2 sequences of specimens MG972110 and MG972112 from M. giganteus, from Clermont, Queensland to those of specimens from O. robustus in Charters Towers, Queensland (see Additional file 1: Figures S2  and S3).

Phylogenetic analysis
The ITS1, ITS2 and concatenated ITS (ITS+) sequence data of P. kappa and P. patriciae n. sp. were analysed using the Neighbour-Joining (NJ) and Bayesian Inference (BI) methods. All the NJ and BI trees produced were similar in topology, hence only BI trees are presented for ITS1 (Fig. 4), ITS2 (Fig. 5) and ITS+ (Fig. 6) from P. kappa and P. patriciae datasets. Phylogenetic trees of ITS+ sequences of P. kappa and P. patriciae with P. papillatus, P. sharmani and all available congener Fig. 5 Phylogenetic analysis of the ITS2 rDNA sequences of Pharyngostrongylus kappa and Pharyngostrongylus patriciae n. sp. from various host species and geographical locations. The sequence data were analysed using the Neighbour-Joining (NJ) and Bayesian Inference (BI) methods. There was a concordance between the topology of the BI tree and the NJ tree (not shown). Nodal support is given as a posterior probability of BI/bootstrap value for NJ. Each unique sequence is presented with a GenBank accession no. followed by the voucher number, and its host and locality. Three sequences (LT576294-LT576296) were included as reference sequences from Chilton et al. [12]. Cloacina ernabella was used as the outgroup. Scale-bar indicates the number of inferred substitutions per nucleotide site. Abbreviations: NSW, New South Wales; NT, Northern Territory; QLD, Queensland; stn, station; VIC, Victoria; WA, Western Australia sequences were produced using both BI and NJ methods (Fig. 7).
Phylogenetic analysis of the ITS1 sequences (Fig. 4) displayed six major clades. Clade 1 contained all sequences of P. kappa from M. giganteus with weak nodal support (posterior probabilities for BI = 0.60; bootstrap value for NJ = 78%). Within this clade, a sub-clade contained sequences from NSW, Qld and Vic, with weak nodal support (BI, 0.64). Clade 2 contained all sequences from O. robustus at Charters Towers, Qld, with moderate nodal support (BI, 0.96; NJ, 84%) and one sequence (MG972161) diverging from the main clade. Two sequences from O. bernardus from Maningrida, NT, formed Clade 3, with strong statistical support (BI, 1.0; NJ, 92%). Clade 4 contained sequences from all three species of Osphranter from NT and sequences from O. robustus and O. antilopinus in WA, with moderate support (BI, 0. 98; NJ, 74%). Sequences from O. robustus at Edith River and O. bernardus from Maningrida, NT formed a fifth clade with strong statistical support (BI, 0.97; NJ, 93%). One sequence of O. robustus from Edith River, diverged in its own clade. Overall, clade formation was correlated with host species rather than the collection localities. Based on the number of clades, specimens of P. patriciae n. sp. exhibited greater genetic diversity than those of P. kappa.
Phylogenetic analysis of the ITS2 sequences yielded a tree with similar topology to ITS1 sequences with six major clades (Fig. 5). The major difference in the topology of the ITS2 tree were two sequences from M. giganteus in Clermont, Qld (MG972110 and MG972112) that grouped outside the M. giganteus clade. Instead, these two sequences clustered with the O. robustus clade from Charters Towers, Qld, though with weak nodal support (BI, 0.54; NJ, 63%).
Phylogenetic analysis of concatenated ITS sequences displayed a tree (Fig. 6) that shared a similar topology to both ITS1 and ITS2 trees. The ITS+ sequences from M. giganteus formed an identical clade and subclade to Clade 1 in the ITS1 tree. As with the ITS2 tree, the M. giganteus sequences from Clermont, Qld diverged forming an isolated clade (Fig. 6). Among the ITS+ sequences from Osphranter spp., O. robustus sequences from Charters Towers, Qld formed a separate clade (Clade 4) to the remaining sequences from the NT and WA (Fig. 6).
Phylogenetic analysis of concatenated ITS sequences of P. papillatus (GenBank: MG519827-MG519828 Fig. 6 Phylogenetic analysis of the ITS+ rDNA sequences of Pharyngostrongylus kappa and Pharyngostrongylus patriciae n. sp. from various host species and geographical locations. The sequence data were analysed using the Neighbour-Joining (NJ) and Bayesian Inference (BI) methods. There was a concordance between the topology of the BI tree and the NJ tree (not shown). Nodal support is given as a posterior probability of BI/bootstrap value for NJ. Unique sequences are presented with a GenBank accession no. followed by the voucher number, and host and locality. Three sequences (LT576294-LT576296) were included as reference sequences from Chilton et al. [12]. Cloacina ernabella was used as the outgroup. Scale-bar indicates the number of inferred substitutions per nucleotide site. Abbreviations: NSW, New South Wales; NT, Northern Territory; QLD, Queensland; stn, station; VIC, Victoria; WA, Western Australia and MG972188-MG972189) and P. sharmani (Gen-Bank: MG972190-MG972191) determined here, and all of those available on GenBank sequences of congeners [12,30] produced a tree (Fig. 7) with a similar topology to the ITS+ analyses of P. kappa and P. patriciae n. sp. The ITS+ sequences of P. kappa formed one clade (nodal support BI, 0.99 and 0.7) while those of P. patriciae diverged into two clades (nodal support BI, 0.96 and 0.95). The ITS+ sequences of P. papillatus formed one clade with strong statistical support (BI, 1.0) in between two clades of P. patriciae. The ITS+ sequences of P. sharmani also formed a clade with strong statistical support (BI, 1.0). Each Pharyngostrongylus sp. from GenBank formed a separate clade with strong statistical support (see Fig. 7).
Based on phylogenetic analyses of ITS sequences of all Pharyngostrongylus spp., genetic variation was associated with differences in hosts rather than with geographical localities, with one exception, the two sequences of P. kappa from M. giganteus from Clermont that formed an isolated sub-clade within the P. kappa clade (see Fig. 7).

Discussion
The present study aimed to characterise the genetic and morphological diversity of P. kappa from its macropodid hosts, including Macropus giganteus, Osphranter robustus, O. antilopinus and O. bernardus. It was hypothesised that due to the wide host range and geographical distribution, there was potential for intraspecific genetic variation within P. kappa. Molecular and morphological findings provide evidence that specimens of P. kappa from different host genera are genetically and morphologically distinct from one another.
Pharyngostrongylus kappa from Osphranter spp. are now recognised as a new species based on the length of the buccal capsule and the number of transverse striations. Specimens from Osphranter spp. had shorter buccal capsules with fewer striations compared to specimens from M. giganteus. Measurements show that the buccal capsule length of P. kappa from M. giganteus (0.12-0.17 mm) fell within range of the type-specimens (0.13-0.19 mm) described by Mawson [31]. However, the buccal capsule length of specimens from Osphranter spp. fell below this range [O. robustus (type-specimens): 0.08-0.10 mm; O. antilopinus: 0.08-0.12 mm; and O. bernardus: 0.09-0.11 Fig. 7 Phylogenetic analysis of the concatenated ITS+ rDNA sequences of Pharyngostrongylus spp. from various host species and geographical locations. The sequence data were analysed using the Neighbour-Joining (NJ) and Bayesian Inference (BI) methods. There was a concordance between the topology of the BI tree and the NJ tree (not shown). Nodal support is given as a posterior probability of BI. Cloacina ernabella was used as the outgroup. Scale-bar indicates the number of inferred substitutions per nucleotide site. Abbreviations: NSW, New South Wales; NT, Northern Territory; QLD, Queensland; VIC, Victoria; WA, Western Australia mm]. The present study also found consistent differences in the number of striations on the buccal capsule of P. kappa from M. giganteus (60-40 striations) and P. patriciae n. sp. from Osphranter spp. (38-50 striations). Although this feature was not recorded in the type-specimens nor the redescription by Beveridge [11], it could be considered taxonomically significant for other members of this genus. However, the morphological differences observed in specimens of P. patriciae n. sp. from Osphranter spp. are supported by molecular evidence strengthening the hypothesis that specimens from Osphranter spp. represent a new species of Pharyngostrongylus.
Sequence datasets of the ITS rDNA indicate that P. kappa occurring in the primary host, M. giganteus is genetically distinct from those occurring in O. robustus, O. antilopinus and O. bernardus. The length of the ITS2 sequence of P. kappa from M. giganteus from Genoa, Victoria (219 bp) is identical to the published sequence (GenBank: LT576294; [12]) from the same host and location. The ITS1 sequence amplification was unsuccessful, hence there was no comparable data for this region. Both the ITS1 and ITS2 of P. patriciae specimens from O. robustus from Charters Towers in Queensland were identical to a published sequence (GenBank: LT576296; [12]). In contrast, the lengths of ITS1 (387 bp) and ITS2 (222 bp) of P. patriciae from O. robustus from Edith River in the Northern Territory differed by one base pair from the published sequence (GenBank: LT576295; [12]) from the same host and locality (ITS1, 386 bp; ITS2, 223 bp). These differences could be attributed to the larger sample size included in the current study.
Phylogenetic analysis of the ITS sequences indicated that specimens of P. kappa from M. giganteus and those of P. patriciae from Osphranter spp. did not form a monophyletic assemblage, providing support for earlier work by Chilton et al. [12]. Furthermore, ITS+ sequences of P. patriciae from O. robustus from Charters Towers, Queensland, formed a separate clade from specimens in Osphranter spp. from Western Australia and the Northern Territory also consistent with the study of Chilton et al. [12].
The populations of nematodes studied were taken from two different subspecies of M. robustus: M. r. robustus from the Charters Towers region of Queensland, and M. r. woodwardi from the Northern Territory and Western Australia [1] and may represent co-divergence with the two host subspecies. In addition, the two nematode populations appear to be disjunct as P. patriciae was not found in six M. robustus examined in the Mount Isa-Cloncurry region of north-western Queensland (I. Beveridge, unpublished data).
The only exception to the general pattern of clade formation was the segregation of two ITS+ sequences of P. kappa from M. giganteus from Clermont, Queensland (MG972110 and MG972112). The sharing of nucleotide changes in the ITS1 with specimens from M. giganteus and ITS2 with specimens of P. patriciae n. sp. from O. robustus may be indicative of genetic introgression. Clermont is in an area where M. giganteus occurs in sympatry with O. robustus. Pharyngostrongylus kappa may have switched hosts, resulting in the combination of ITS1 and ITS2 nucleotide changes observed. All specimens of P. kappa from Clermont were morphologically identical and conformed to the phenotype normally found in M. giganteus. Chilton et al. [15] found evidence of hybridisation between Paramacropostrongylus iugalis occurring in M. giganteus and P. typicus in M. fuliginosus in an area of host sympatry. However, there is insufficient evidence in this study to support the hypothesis of hybridisation among genetically distinct forms of P. kappa. Evidence from previous studies suggests that speciation of nematodes in macropodid marsupial hosts has occurred primarily by host-switching [7]. However, a larger sample size of P. kappa and P. patriciae from hosts within areas of sympatry with Osphranter spp. is required to confirm the occurrence of such host-switching events.
The current study also sequenced congeners of P. kappa, P. papillatus and P. sharmani from northern macropodid species. The occurrence of P. papillatus in O. bernardus documented in this study represents a new host record as P. papillatus has previously been found in O. antilopinus and O. robustus [18]. Additionally, ITS+ sequences of P. papillatus from O. bernardus formed distinct clades from specimens from O. robustus. These results suggest a potential cryptic species occurring among P. papillatus that require further morphological and molecular analysis.

Conclusions
The evidence from the current study supports the hypothesis that P. kappa found in M. giganteus is genetically distinct from P. patriciae n. sp. found in Osphranter spp. Furthermore, morphological data revealed distinctive features of specimens from Osphranter spp. supporting the morphological data. The description of a new species was made based on morphological and molecular evidence. This study contributes to earlier efforts in documenting the diversity of Australian nematodes and suggests that molecular techniques indicate that the diversity requires further detailed exploration.

Additional file
Additional file 1: Table S1. Specimens of Pharyngostrongylus kappa from Macropus giganteus examined from the South Australian Museum (SAM), Adelaide. Figure S1. Alignment of ITS1 rDNA sequences of Pharyngostrongylus kappa and P. patriciae n. sp. from different macropodid hosts. A dot indicates an identical nucleotide with respect to the sequence of XDR10.1; a dash indicates an insertion/deletion (indel) event. IUPAC codes indicate polymorphic positions in the sequences. Figure S2. Alignment of ITS2 rDNA sequences of Pharyngostrongylus kappa and P. patriciae n. sp. from different macropodid hosts. A dot indicates an identical nucleotide with respect to the sequence of XDR10.1; a dash indicates an insertion/deletion (indel) event. IUPAC codes indicate polymorphic positions in the sequences. (DOCX 69 kb)