Parasites of Selected Freshwater Snails in the Eastern Murray Darling Basin, Australia

Aquatic snails serve an important role in the ecosystem. They also play an essential role in the life cycle of many parasites as hosts and may pose risks to animal and human health. In Australia, the role of snails in the transmission of parasites of livestock is well studied. However, despite the country’s unique biodiversity and wildlife, little is known about the role of snails in the transmission and survival of parasites in other ecosystems, including aquatic and aquaculture systems. This study aimed to determine the occurrence of parasites in freshwater snails in the eastern Murray Darling Basin. A total of 275 snails were collected from various localities, including aquaculture fishery ponds and natural creeks during the summer and autumn months in the southern hemisphere. Three different species of freshwater snails, all common to the area, were found, including Bullastra lessoni (n = 11), Isidorella hainesii (n = 157), and Haitia acuta (n = 107), of which 9.1%, 1.3%, and 4.7%, respectively, were found to be harboring various developmental stages of Trematoda. No other parasite was found in the examined snails. Parasites were identified as Choanocotyle hobbsi, Plagiorchis sp. and Petasiger sp. based on the sequences of their ITS2, 18S, and 28S ribosomal DNA region. Herein, we report a native parasite Choanocotyle hobbsi in an introduced snail, Haitia acuta, from both natural and aquaculture ponds. As there are no genetic sequences for adult specimens of Petasiger spp. and Plagiorchis spp. collected in Australia for comparison, whether the specimens collected in this study are the larval stage of one of the previously described species or are a new, undescribed species cannot yet be determined. Our results also suggest snails collected from aquaculture ponds may be infected with considerably more parasites.


Introduction
Aquatic snails form a significant part of any ecosystem and are important in maintaining the balance of nature in this environment [1][2][3]. For example, because they are on the lower trophic levels of the food web, they are an important food source for many aquatic and aquatic-associated animals (from insects to lizards and snakes, fish, birds, and mammals) [1,2]. Additionally, due to their sensitivity to certain chemicals, aquatic snails can be used as environmental and water quality indicators. Unfortunately, several native freshwater snails in Australia are threatened [4], which is worrisome, considering the important role freshwater snails play in aquatic food webs.

Sample Collection
A total of 275 snails were collected from various localities, as shown in Figure 1. The collection localities were a combination of aquaculture fishery ponds (locations 1 and 2) and natural creeks (locations 3 and 4). The collection took place during summer and autumn months in the southern hemisphere (February-April 2019). The snails were collected in large specimen jars, approximately half-full of water, and were transported to the Parasitology Laboratory of Charles Sturt University. The snails were identified using Ponder (2020), and all of them were examined by autopsy as described previously [13]. Some parasite specimens were preserved in 70% ethanol for molecular work, and some were mounted permanently in glycerin jelly. Scale bar represents 20 km. Localities 1 and 2 were a golden perch aquaculture pond with soil bottom, frequented by cormorants, ducks, and egrets. Other life found at the bottom of ponds included yabbies and shrimp. Small bivalves, dipteran insects, and water scorpions were also found in Locality 2. A combination of bore water and river water (Murrumbidgee River) was used for ponds. Locality 3 was a creek flowing through a pine plantation with feral deer, feral pigs, and many native herbivores (kangaroos, wallabies, wombats) but no livestock in the collection area. Cattle and goats were present on properties upstream. Snails were among floating pondweed Potamogeton tricarinatus. Other life found among snails included leeches and dragonfly larvae. Locality 4 was a creek flowing through a cattle and sheep property. Dry ewes were in the paddock 2 weeks before collection. Snails were among water ribbons Vallisneria gigantea and water couch Paspalum paspalodes. The pond was frequented by cormorants, ducks, egrets, and pelicans. Other life found at the bottom of ponds included yabbies, shrimp, small bivalves, water scorpions, and dipteran insects.

Morphology of Parasites
Slide-mounted specimens were examined by light microscopy. Measurements of total length (TotL), body length (BL), body width (BW), tail length (TL), tail width (TW), tail width with fins (TWF), oral sucker diameter (OS), and ventral sucker diameter (VS) were taken. The numbers of collar spines were counted. Illustrations were created using a microscope equipped with a drawing tube. All measurements are given in micrometers, unless otherwise stated. Mean measurements are specified, followed by the range in parentheses. Photos were taken using a 9 MP Microscope Digital Camera (AmScope Model MU900).

Molecular Diagnostics of Parasites
Single cercaria, redia, or sporocysts were placed in individual Eppendorf tubes and stored at −20 • C until DNA extraction. The samples did not need to be cut, as they were extremely small (<1 mm), and there were many available samples. DNA extraction was completed using the QIAGEN DNeasy Blood and Tissue Kit, following the manufacturer's instructions. The ITS2, 18S, and 28S regions were amplified using primers and reagents described in Shamsi et al. [13] with the following conditions for all primers and regions: initial denaturation at 95 • C for 2 min; 40 cycles of denaturation (95 • C), annealing (58 • C for both primer pairs), and extension (72 • C) for 30, 30, and 45 s, respectively, followed by a final extension at 72 • C for 10 min. PCR products were Sanger sequenced using the same primer at the Australian Genome Research Facility (Brisbane). Sequences obtained from this study were deposited in the GenBank with accession numbers OM305031-OM305042 (28S region), OM305043-OM305054 (18S region), and OM305095-OM305107 (ITS region).
The sequences were aligned using BioEdit [14]. Primer sequences were removed from analysis. ITS2 sequences of closely related taxa were obtained from GenBank for phylogenetic analyses (Table 1). Where possible, we used sequences obtained from adult specimens associated with morphologically well-identified specimens and peer-reviewed published works. Alignments for ITS2, 28S, and 18S for group A and morphotype B were 1275, 1269, and 1777 bp, respectively. For morphotype C, the alignments of the same regions were 1523, 1225, and 1754, respectively. Descriptions of the groups/morphotypes are provided in the Results section. Alignment gaps were excluded for analyses. Pairwise genetic distances were calculated using MEGA X [15]. The GTR + G, GTR + I + G, and HKY + I models were selected for ITS2, 28S, and 18S regions, respectively, as best fit evolutionary models as inferred by the jModelTest 2 [16]. Brachycladium goliath (KR703279) was used as an outgroup for Choanocotyle and Plagiorchis sp. phylogenetic analyses, as it belongs to the same suborder Xiphidiata but different superfamily. Philophthalmus gralli (JX121229 and JQ627832) were used as an outgroup for Petasiger sp. phylogenetic analyses, as it belongs to the same superfamily but different family. The phylogeny of selected sequences was calculated using MrBayes 3.2 [17] for 3,000,000 generations for each gene region, with other parameters set as default, until the average standard deviation was lower than 0.005. The first 50% of runs from the Markov chain Monte Carlo algorithm were discarded as burn-in. The tree was visualized using Figtree v 1.4.3 [18].

Results
Three different species of freshwater snails were found. They are all common to the area. They were found to belong to three distinct families-family Lymnaeidae (Bullastra lessoni (n = 11)), family Planorbidae (Isidorella hainesii (n = 157)), and family Physidae (Haitia acuta (n = 107)). The latter species is an introduced species, which is considered invasive in Australia. Not all snails were infected with parasites. Various developmental stages of Trematoda, including sporocysts, cercariae, and metacercariae, were found in the infected snails. The highest infection rate (9.1%) was observed among Bullastra lesson; however, only 11 specimens were available in the present study. Therefore, this infection rate should be viewed with caution. Of the other two species of snails examined herein, Haitia acuta and Isidorella hainesii, 4.7% and 1.3%, respectively, were found to be infected with Trematoda parasite. No other parasite groups apart from trematodes were found in the examined snails. No mixed infection was observed. Details of the parasites found in different localities and hosts are provided in Table 2. Table 2. Snails examined in the present study and the parasites found. Locality data refer to the location numbers identified in Figure 1.
The parasites found were all at the larval stage and could not be identified to the species level. Therefore, similar morphotypes were classified into different groups, designated as A to C ( Table 2). Cercaria classified as group A did not have any distinguishing characteristics; no morphological description could be performed, as all cercaria found were not fully developed. This is possibly due to the cercaria not emerging from the snail but being removed by dissection. They were identified to the genus Plagiorchis based on their sequence data (Figure 2A-C). Sequences from this study were grouped with sequences of Plagiorchis spp., primarily from cercarial stages, from throughout Europe for both ITS2 (Figure 2A) and 28S ( Figure 2B). For 18S sequences ( Figure 2C), however, a lack of available sequences of Plagiorchis spp. placed the sequences from this study in a group with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also Table 1). Geographical area of collection of specimen indicated by a colored bar (red, North America (USA and Mexico); blue, Europe; yellow, Australia; green, Brazil; brown, Japan and China; light brown, Pakistan; light green, Rwanda). The host groups that the parasite was recovered from are shown as icons ( with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related Table 1). with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   with specimens of related genera collected from insectivorous hosts (frog, shrew) (see also   Group B was found to morphologically and genetically match Choanocotyle hobbsi as described in Shamsi, Nuhoglu, Zhu, and Barton [12] (Figure 2A-C) and is referred to as morphotype B in this paper.
Group C featured cercaria and redia with distinguishing characteristics (Figure 3), including a collar of spines, a shouldered body shape (instead of completely oval), a relatively long tail, and a larger ventral sucker in comparison to its oral sucker. The samples that are referred to as morphotype C in this study were not in a good enough condition to identify the number of collar spines. However, it was possible to see one group of four corner/posterior spines on each side of the oral sucker posteriorly. The specimens all had obvious fins along the tail. They had a total body length and width of 773.13 (705-855) and 332.14 (255-380) µm, respectively (n = 14 cercaria). Body length (excluding tail length) was 332.14 (255-380) µm. The tail was 442.50 (385-500) long. Tail width, with and without wing, was 43.75 (40-57.5) and 27.86 , respectively. Oral and ventral suckers had diameters of 48.75  and 69.81 (37.5-85), respectively. Additionally, a small group (2-3) of large granules were obvious posterior to the oral sucker in some specimens. Due to the presence of the collar spines, the cercaria were identified as members of the superfamily Echinostomatiodea [63]. They were identified as belonging to the genus Petasiger based on their sequence data (Figure 4). Morphotype C, which was identified as Petasiger sp., belongs to the suborder Echinostomata, whereas group A and morphotype B, i.e., Plagiorchis and Choanocotyle hobbsi, taxonomically belong closer to the suborder Xiphidiata. To avoid producing very large trees, separate phylogenetic trees were created for morphotype C. Sequences from this study were consistently grouped with Petasiger radiatum, collected from cormorants in Hungary (Figure 4).
were obtained. Some of these hosts are intermediate/paratenic and some are definitive hosts. Group B was found to morphologically and genetically match Choanocotyle hobbsi as described in Shamsi, Nuhoglu, Zhu, and Barton [12] (Figure 2A-C) and is referred to as morphotype B in this paper.
Group C featured cercaria and redia with distinguishing characteristics (Figure 3), including a collar of spines, a shouldered body shape (instead of completely oval), a relatively long tail, and a larger ventral sucker in comparison to its oral sucker. The samples that are referred to as morphotype C in this study were not in a good enough condition to identify the number of collar spines. However, it was possible to see one group of four corner/posterior spines on each side of the oral sucker posteriorly. The specimens all had obvious fins along the tail. They had a total body length and width of 773.13 (705-855) and 332.14 (255-380) µm, respectively (n = 14 cercaria). Body length (excluding tail length) was 332.14 (255-380) µm. The tail was 442.50 (385-500) long. Tail width, with and without wing, was 43.75 (40-57.5) and 27.86 , respectively. Oral and ventral suckers had diameters of 48.75  and 69.81 (37.5-85), respectively. Additionally, a small group (2-3) of large granules were obvious posterior to the oral sucker in some specimens. Due to the presence of the collar spines, the cercaria were identified as members of the superfamily Echinostomatiodea [63]. They were identified as belonging to the genus Petasiger based on their sequence data (Figure 4). Morphotype C, which was identified as Petasiger sp., belongs to the suborder Echinostomata, whereas group A and morphotype B, i.e., Plagiorchis and Choanocotyle hobbsi, taxonomically belong closer to the suborder Xiphidiata. To avoid producing very large trees, separate phylogenetic trees were created for morphotype C. Sequences from this study were consistently grouped with Petasiger radiatum, collected from cormorants in Hungary (Figure 4).    , mammals other than bats; , fish-eating birds). The hosts are those listed in Table 1 and include hosts from which parasites/sequences were obtained. Some of these hosts are intermediate/paratenic and some are definitive hosts.
Despite some intraspecific variation among 18S sequences belonging to C. hobbsi, the grouping of the sequences of taxa included in all three trees suggests that ITS2, 28S, and 18S are suitable for differentiation between digenean parasites. The phylogenetic tree for members of the superfamily Plagiorchioidea, including group A and morphotype B (Figure 2), also shows Australian taxa group separately from the taxa found in other parts of the world; however, for members of the superfamily Echinostomatoidea, including morphotype C, such distinction was not observed.

Discussion
Of the snails collected and examined in the present study, Bullastra lessoni and Isidorella hainesii are native species, whereas Haitia acuta is an introduced species. Choanocotyle hobbsi, also found in the present study, is a native parasite, which has been recently reported in Isidorella hainesii [13]. Herein, we report this native parasite in an introduced snail, Haitia acuta, from both natural and aquaculture ponds. This is a case of parasite spillback where a parasite of native hosts infects an invasive host, leading to increased opportunities to infect native species [64]. In a previous study [11], researchers showed that there were only three reports of H. acuta shedding larval trematodes (cercariae) within its invasive range in Europe and the Middle East. However, due to a lack of genetic data for parasite larvae, they could not determine the origin of infection of invasive H. acuta (i.e., spillback versus spillover). As suggested by Ebbs et al. [11], including parasite genetic data, such as in the present study, is required to better understand the invasion dynamics. Parasite spillback from introduced species could potentially affect all host species in a parasite's life cycle and cause disease emergence [65]. Choanocotyle hobbsi is a parasite of freshwater turtles, many species of which are known to have had a massive decline in their population [66]. However, despite its significance, parasite spillback has been seriously neglected in the conservation plans of the ecologically fragile Murray Darling Basin in Australia. This should be brought to the attention of decision makers and conservation , fish-eating birds). The hosts are those listed in Table 1 and include hosts from which parasites/sequences were obtained. Some of these hosts are intermediate/paratenic and some are definitive hosts.
Despite some intraspecific variation among 18S sequences belonging to C. hobbsi, the grouping of the sequences of taxa included in all three trees suggests that ITS2, 28S, and 18S are suitable for differentiation between digenean parasites. The phylogenetic tree for members of the superfamily Plagiorchioidea, including group A and morphotype B (Figure 2), also shows Australian taxa group separately from the taxa found in other parts of the world; however, for members of the superfamily Echinostomatoidea, including morphotype C, such distinction was not observed.

Discussion
Of the snails collected and examined in the present study, Bullastra lessoni and Isidorella hainesii are native species, whereas Haitia acuta is an introduced species. Choanocotyle hobbsi, also found in the present study, is a native parasite, which has been recently reported in Isidorella hainesii [13]. Herein, we report this native parasite in an introduced snail, Haitia acuta, from both natural and aquaculture ponds. This is a case of parasite spillback where a parasite of native hosts infects an invasive host, leading to increased opportunities to infect native species [64]. In a previous study [11], researchers showed that there were only three reports of H. acuta shedding larval trematodes (cercariae) within its invasive range in Europe and the Middle East. However, due to a lack of genetic data for parasite larvae, they could not determine the origin of infection of invasive H. acuta (i.e., spillback versus spillover). As suggested by Ebbs et al. [11], including parasite genetic data, such as in the present study, is required to better understand the invasion dynamics. Parasite spillback from introduced species could potentially affect all host species in a parasite's life cycle and cause disease emergence [65]. Choanocotyle hobbsi is a parasite of freshwater turtles, many species of which are known to have had a massive decline in their population [66]. However, despite its significance, parasite spillback has been seriously neglected in the conservation plans of the ecologically fragile Murray Darling Basin in Australia. This should be brought to the attention of decision makers and conservation scientists in Australia, considering that over time, as invasive H. acuta populations increase, their role in local parasite transmission will also increase.
Parasite spillback might be a common occurrence in this region. Previously, a native nematode parasite, Contracaecum bancrofti, was found in several introduced fish hosts, Carassius auratus, Misgurnus anguillicaudatus, Cyprinus carpio, and Gambusia holbrooki [67,68]. Understanding the extent of parasite transmission between native and introduced species in the Murray Darling Basin is an important area for future research.
Another parasite found in the present study was Plagiorchis sp. found in Bullastra lessoni. We did not find an exact genetic match, nor fully developed cercaria, and therefore could not identify it to species level. The parasite belongs to the family Plagiorchiidae (Lühe, 1901), which is a very large family of digenean trematodes. Plagiorchis spp. parasitize the digestive system of many species of vertebrates, including humans [53,55,69,70]. In Australia, P. maculosus was reported in birds, including Hirundo neoxena, Rhipidura leucophrys, R. flabellifera, Gymnorhina hypoleuca, and Pomatostomus superrciliosus. Adult Plagiorchiids can be found in any part of the digestive system and can migrate throughout the digestive system of the vertebrate definitive host [55]. Although it is a large group of potentially dangerous parasites for many species, their taxonomy is poorly understood and in need of revision. There are currently 140 described species within the family, making it the largest family of digeneans [55]. Additionally, Johnston and Angel [71] studied the life history of Plagiorchis jaenschi and experimentally infected B. lessoni (= Lymnaea lessoni) with eggs collected from worms from a water rat in South Australia. They also reported a natural infection in the same species of snail.
Lymnaeid snails are known to be the intermediate host for Plagiorchiids [72]. In Angel's (1959) study, 2/55 snails were found to be infected with small cercaria. Mosquito larvae were experimentally infected with these cercaria and then fed to chickens once they developed into adult mosquitos. Two of the experimentally infected chickens were infected with adult trematodes of Plagiorchis maculosus. The eggs from these adult flukes were then successfully used to infect lab-raised snails. Sporocysts and some free cercaria were found in these snails. In the present study, snails were found naturally infected with Plagiorchis sp. Because no fully developed cercaria were found, it was not possible to compare the two species morphologically, and Angel [72] did not have genetic data available. It is important to note that many dipteran larvae were found living inside of the B. lessoni snail's shells, with 19 living inside of the infected snail. It is possible that this is how these larvae become infected with Plagiorchis. Observationally, many small adult midge-type flies were found in the present study after a few days of keeping the snails, possibly from these dipteran larvae. In future studies, it would be worth catching and identifying these flies and checking them for Plagiorchis spp. Additionally, a larger number of lymnaeid snails need to be collected from the same sampling site again in the future, and snails should be kept alive until cercaria are fully developed and are shed into water for the morphology to be completed.
Another parasite found in the present study is Petasiger sp. Members of this genus are known to be cosmopolitan and to be found in snails belonging to the family Planorbidae as cercariae, in the esophagus or pharynx of freshwater teleosts as metacercariae, and in the intestine of fish-eating birds (Anhingidae, Phalacrocoracidae, Phoenicopteridae, Podicipedidae, and occasionally Anatidae, and Laridae) in the adult form [73]. Few species of Petasiger have been reported from Australian birds [74], with P. australis reported from grebes in South Australia [71], P. exaeretus from cormorants and shags in South Australia, NSW, and Queensland, although not from the Murrumbidgee catchment area [75], and a Petasiger sp. from a barn owl in South Australia [74]. Johnston and Angel [71] described a cercaria (Cercaria gigantura), presumed to be the larval stage of P. australis, to have a total of 19 collar spines and a "relatively huge tail" that affected the swimming motion of the cercaria. A comparison of the measurements presented for C. gigantura with the cercaria collected in this study showed that although the tail lengths were approximately equal, the body length for C. gigantura was shorter (105-267 µm) compared to the cercaria collected in this study. Both P. exaeretus and the Petasiger sp., however, have 27 collar spines; this former species has also been reported from cormorants from Europe and Japan [75]. As there are no genetic sequences for adult specimens of Petasiger spp. collected in Australia for comparison, whether the Petasiger sp. collected in this study is the larval stage of one of the previously described species or is a new, undescribed species cannot yet be determined.
In the present study, Petasiger sp. could not be identified to species level due to the absence of any identical and comparable sequence data from adult specimens. The cercaria found in our study had similar morphology to those reported by Našincová et al. [76], including similarly located posterior and collar spines; however, the staining procedure in our study did not allow for a clear enough visualization of the exact number of collar spines present. Additionally, some of the cercaria collected in our study possessed a small group of large granules posterior to the oral sucker, similar to that described by Laidemitt et al. [53] for Petasiger sp. 3 and sp. 4, collected from snails in Kenya. The results of the 28S analysis found the sequences collected in this study to be very close to those for Petasiger sp. 4 ( Figure 4B). In the tree presented by Laidemitt, Brant, Mutuku, Mkoji, and Loker [53], Petasiger sp. 4 matched an adult worm collected from Microcarbo africanus in Kenya and was grouped with an undescribed Echinostoma sp., collected in Australia by Morgan and Blair [77]. Petasiger sp. 4 possessed 27 collar spines [53], whereas the undescribed Echinostoma sp. possessed over 40 collar spines [77]; the number of collar spines could not be determined in the specimens collected in this study, potentially due to their young stage of development and being dissected from the snails.
When studying P. radiatus, Našincová, Scholz, and Moravec [76] did not find sporocysts in any of the naturally or experimentally infected snails, but rediae were found in both, similar to our results. In Europe, the cercarial stage of Petasiger has been found in freshwater pulmonate snails Gyraulus albus and Segmentina nitida, both of which belong to the family Planorbidae, and Radix auricularia, a pulmonate Lymanaeid [76]. In our study, the cercarial stage was found in Isidorella hainesii, a native Australian snail, also from the family Planorbidae. Pulmonates have air sacs to enable them to breathe air, meaning they must go to the surface of the water from time to time. This could explain why the cercaria of many Petasiger spp. have long tails with fins, as they must move through the water to find snails that may be near the surface of the water. The Petasiger sp. cercaria found in the present study had these morphological characteristics and were also observed to be highly motile for a number of hours after exiting the snail host.
In the study by Našincová, Scholz, and Moravec [76], experimentally infected fish had metacercaria encysted around the mouth and gills, eyes, nasal hollows, and in the skin. Metacercaria from the Echinostomatidae family are frequently found in fish and, close to where snails were collected in the present study, various fish were found to be infected with metacercaria of Trematoda [78,79]. However, they did not belong to Petasiger sp. Therefore, it is important for parasites found in wild and farmed fish to be examined properly for specific identification and to inform subsequent management decisions. Petasiger spp. are a commonly found trematode parasite in the intestine of piscivorous birds (particularly cormorants) in Europe, Asia, and Africa [48,76]. In Australia, Petasiger australis has been reported from Hoary-headed Grebe, Poliocephalus poliocephalus [71].
Aquaculture ponds are known to favor populations of predators that could be potential definitive hosts, such as aquatic birds [80]. Although our sampling sites were from both natural reservoirs and aquaculture farms, due to significant differences in the number of snails collected, no reliable conclusion can be drawn about any significant difference in the population of the infected snails between different sites. An interesting area for future study would be to investigate this matter.

Conclusions
The knowledge of parasites in Australian wildlife is poor, with most host species, especially those that act as intermediate hosts, unstudied. The documentation of this fauna, including both morphological and molecular characterization, is important to ensure an understanding of biodiversity, parasite transmission, and ecosystem impacts.