The Discovery of Asian Fish Tapeworm (Schyzocotyle Acheilognathi) and Schyzocotyle Sp. In Western Australia May Pose a Threat to the Health of Endemic Native Fishes

Purpose The Asian sh tapeworm (Schyzocotyle acheilognathi) is an important sh parasite with a wide host range that infects over 300 species of sh worldwide. Schyzocotyle acheilognathi has been reported from eastern coastal areas of Australia, but has not been previously reported in Western Australia (WA). Methods During a control program for invasive freshwater shes in south-western WA, a region with a unique and highly endangered freshwater sh fauna, tapeworms identied as S. acheilognathi from their distinctive scolex morphology were found at a prevalence of 3.3% in goldsh (Carassius auratus), 37.0% in koi carp (Cyprinus carpio haematopterus), and 65.0% in eastern mosquitosh (Gambusia holbrooki). For molecular conrmation, the 18S rRNA gene was targeted at hypervariable region V4 using conventional PCR and Sanger sequencing.

Unusually for a tapeworm, S.acheilognathi has a very broad de nitive host range and has been found in at least 312 freshwater sh species, belonging to 38 families and 14 orders, worldwide (Kuchta et al. 2018). The species has also been identi ed in non-piscine vertebrate hosts, including amphibians, reptiles, and birds (García-Prieto and Osorio-Sarabia 1991; Scholz 1999; Kuchta et al. 2018), although these are considered accidental hosts (Kuchta et al. 2018). In a study conducted by Prigli (1975), S. acheilognathi was transmitted to ducks when they fed on young infected carp. Viable eggs were found in the faeces 3 h after ingestion, suggesting that birds could act as potential vectors of AFT (Prigli 1975). Schyzocotyle acheilognathi can also infect humans, a male patient from France, who regularly ate raw sh, had persistent abdominal pain and AFT eggs were found in his stool (Yera et al. 2013).
The global spread of invasive sh species and their parasites, such as S. acheilognathi, is likely attributable to the exotic sh trade (Košuthová et al. 2015;Kuchta et al. 2018). Transmission of S. acheilognathi is facilitated by its simple lifecycle which can be completed in 14 days at 25 °C. The de nitive host is a sh, the eggs are shed from the host and hatch into free-swimming coracidia, which are consumed by the intermediate host, cyclopoid copepods (Arthropoda, Crustacea). Infected copepods are then consumed by the sh and larval stages progress into mature tapeworms (Kuchta et al. 2018).
During the 1960's and 1970's, S. acheilognathi was imported from East Asia to Europe and North America in grass carp, which was introduced to reduce the growth of vegetation in freshwater ecosystems (Matey et al. 2015). The parasite has since spread to all continents, except Antarctica, and has been the cause of numerous devastating infections and high mortalities of naïve hosts in sh hatcheries, with serious economic consequences for the aquaculture industry (Han et al. 2010;Xi et al. 2016). In addition, the invasion of waterways outside of its natural range by S. acheilognathi may threaten native shes. Brouder and Hoffnagle (1997), for example, studied shes from the Colorado River, USA, and found that three of the rivers' four endangered sh species host S. acheilognathi, with an introduced alien sh species in the same river having a prevalence of 22.5% (Brouder and Hoffnagle 1997).
A subsequent study in 2004 found S. acheilognathi parasitising all sh species in the Little Colorado River (Choudhury et al. 2004).
Farming and breeding sh is a major source of global income (Shelton and Rothbard 2006;Olsen and Hasan 2012;Tacon and Metian 2013), particularly for developing countries (Mulokozi et al. 2020). However, studies have shown S. acheilognathi can have devastating impacts on the industry. Koi carp (Cyprinus carpio haematopterus) are bred for consumption and aesthetic purposes (Rahaman et al. 2012). During a routine inspection, Oros et al. (2015) examined 10 market-sized koi carp from a breeding pond in Eastern Slovakia. Schyzocotyle acheilognathi was morphologically identi ed in 40.0% of the sh, representing a signi cant threat to traditional sheries . In a more recent study, the live bait sh trade was emphasised as an important transmission vector for S. acheilognathi, to infect native sh communities (Boonthai et al. 2017). The use of live bait in the USA can compromise aquaculture facilities (Kilian et al. 2012) and AFT can spread to novel areas via bait-bucket introductions (e.g. minnows) with direct release of infected sh. Samples collected in a study from 78 retail stores in Michigan showed that bait sh had a prevalence of tapeworms as high as 58.0% (Boonthai et al. 2017).
Infection with S. acheilognathi can cause bothriocephalosis in its host, with clinical signs including blockage of the gastrointestinal tract, destruction of intestinal mucosa, intestinal rupturing, distended abdomen, weight loss, protein depletion, intestinal in ammation, anaemia, diminished swimming ability, and eventual mortality (Davydov 1978;Scott and Grizzle 1979;Brouder 1999;Hansen et al. 2006;Matey et al. 2015). The clinical signs also lead to reduced growth in the host, which can result in higher predation rates (Choudhury et al. 2013). As a result, S. acheilognathi has been implicated in the decline of native sh species in many countries (Brouder 1999; Pérez-Ponce de León and Poulin 2018).
In Australia, S. acheilognathi has been previously reported from the eastern coastal states including New South Wales (NSW), Queensland, and Victoria (Dove and Fletcher 2000). Only three studies have molecularly con rmed the species identity of the parasite in Australia, and all are from NSW (Luo et al. 2002;Xi et al. 2016;Kuchta et al. 2018;Rochat et al. 2020). With the advent of molecular technology there has been an increasing need for universal genus-speci c primers that amplify hypervariable regions to enable species identi cation. Hatziavdic et al. (2014) compared all of the regions within the 18S rRNA gene of S. acheilognathi and found that the hypervariable region V4 had the longest variable region with the greatest length of polymorphisms, and also included a 70 bp conserved region for primer annealing. Primers targeting the 18S rRNA V4 region in S. acheilognathi have been successful in identifying the species (Nickrent and Sargent 1991;Bean et al. 2007;Hadziavdic et al. 2014).
In the present study, we report the rst evidence of S. acheilognathi from Western Australia (WA), within the South-Western Province, a region of extreme endemism, using morphological and molecular identi cation at the 18S rRNA V4 locus.

Study site and sample collection
Between February and May, 2018, 117 introduced shes were collected from Blue Lake, a modi ed wetland in the City of Joondalup, WA (S 31.745001, E 115.766113) as part of an approved feral sh control program. The sh consisted of 91 gold sh (Carassius auratus) and 26 koi carp (Cyprinus rubrofuscus). Additionally, a small number (n = 17) of introduced eastern mosquito sh (Gambusia holbrooki) were also collected as part of the control program.
Fishes were dissected at Murdoch University's Fish Health Unit and tapeworms removed from the intestine. Half of the tapeworms from each sample were placed in formalin for microscopy and the remainder placed in 100% ethanol for molecular analysis.

Statistical analysis
Prevalence (% of infected sh, with 95% con dence intervals (CI) estimated assuming a binomial distribution) and mean intensity (number of parasites per infected sh, with 95% CI estimated by bootstrapping) were calculated using the program QP Web (Reiczigel et al. 2019).

Morphological analysis
For morphological analysis on samples from gold sh and koi carp, tapeworms were stained using a dilute Semichon's stain, dehydrated using a graded alcohol series, cleared in methyl salicylate, mounted whole in Canada balsam and viewed in bright eld on an Olympus BX50 compound microscope (10x -40x) using an Olympus DP71 universal camera. The eggs were viewed on the same microscope, but using a Nomarski lter. The tapeworms from G. holbrooki were examined using an Olympus microscope (10x -40x).
Total genomic DNA (gDNA) was extracted from ~25 mg of tapeworm tissue (one tapeworm per extraction) and sh intestinal tissue. The tapeworms and intestinal tissue were dissected in sterile petri dishes and extractions were conducted using a Qiagen DNeasy Blood and Tissue kit (Qiagen, St Louis, Missouri). The 18S rRNA V4 region was ampli ed using the forward primer Ces1 (5'-CCAGCAGCCGCGGTAACTCCA-3') and reverse primer Ces2 (5'-CCCCCGCCTGTCTCTTTTGAT-3') producing a ~420 bp product as previously described (Scholz et al. 2003;Bean et al. 2007), except that the annealing temperature was adjusted to 64 °C after optimisation. Conventional PCRs were conducted using 25 µL reaction volumes, which consisted of 0.02 U/µL Taq DNA polymerase, 1x reaction buffer, 1 mM MgCl 2 , 200 µM each of dNTP (dATP, dCTP, dGTP, and dTTP), 1 µM of each primer, 2 µL template gDNA, and PCR grade water to the nal volume. PCR cycling conditions for the Ces1/Ces2 primer set consisted of an initial denaturation at 95 °C for 15 min, followed by 45 cycles of 94 °C for 1 min, 64 °C for 1 min, and 72 °C for 2 min, with a nal elongation step at 68 °C for 10 min. PCR products were separated by gel electrophoresis using a 1.0% (w/v) agarose gel (Fisher Biotec, Australia) in Tris-Acetate (TAE) buffer (consisting of 40 mM Tris-HCl, 20 mM EDTA at pH 7.0). The DNA was stained with SYBR® safe DNA gel stain (Invitrogen, Oregon, USA). 5 µL of 100 bp ladder (Promega, Madison, USA) was used as a reference marker in the rst well. PCR products were electrophoresed at 85 V for 30-40 min in a gel tank (BioRAD, USA). Gels were then visualised under a transilluminator (FisherBiotech, Australia). PCR products of the expected size were excised from the agarose gel with sterile blades, puri ed using an in-house lter tip-based method (Yang et al. 2013) and sequenced using a Big Dye version 3.1 Terminator Cycle Sequencing Kit (Applied Biosystems, Massachusetts, USA) on a 96-capillary 3730xl DNA Analyzer (Thermo Fisher Scienti c, Waltham, Massachusetts, USA) at Murdoch University and at the Australian Genome Research Facility (AGRF), Perth, WA. Sequence chromatograms in the forward and reverse directions were aligned to produce a consensus sequence, assessed for quality, and trimmed of primers using Geneious v10.2.2 (Kearse et al. 2012). Consensus sequences were checked against sequences in GenBank® using the Basic Local Alignment and Search Tool (BLAST) and aligned with related sequences using the MUSCLE alignment tool (Edgar 2004). The 364 bp alignment (including gaps) was imported into the PhyML program (Guindon et al. 2010) to assess nucleotide substitution models based on Bayesian Information Criterion (BIC). The model GTR+G was used to construct a Bayesian phylogenetic tree using Geneious v8.1.9 and the MrBayes plugin v3.2.6 (Huelsenbeck and Ronquist 2001).

Morphological identi cation
The tapeworms from the koi carp and gold sh were identi ed as S. acheilognathi based on the following morphological characters: eshy heart-shaped scolex with unarmed bothria that were short and deep; proglottids had rounded edges with an absent neck; the rst proglottids were immediately posterior to the scolex and much narrower than the scolex; and in mature proglottids, genital pores and vitelline follicles were clearly visible (Figs. 1,  2A, B). All tapeworms examined had gravid proglottids (Fig. 2C), with eggs being both operculated and unembryonated (Fig. 2D).

Prevalence and intensity of infection
The gold sh ranged in total length between 106-498 mm and in weight from between 30.27-2350 g, whilst the koi carp ranged in total length between 448-768 mm and in weight ranged between 1466-4780 g, and the eastern mosquito sh ranged in total length between 38-84 mm and in weight ranged between 0.67g and 4.03gTapeworms were found in 3 of the 91 gold sh (3.3%, 95% CI 3.0-4.3%) and 7 of the 26 koi carp (37.0%, 95% CI 36.0-38.0%), with an overall tapeworm prevalence of 8.5% (95% CI 7.5-9.5%). In addition, 11 of 17 eastern mosquito sh examined (65.0%, 95% CI 64.0-66.0%) were found to be infected.
The intensity of infection, measured in gold sh and koi carp was high, ranging from 203-607 parasites per infected sh (mean = 386, 95% CI 296-486) (Table 1), with evidence of intestinal perforation, blockage, and ischemia (Fig.  3). Several tapeworms exceeded 400 mm in length, which was almost equal to the length of the intestines in some sh. Eastern mosquito sh also had high parasite burdens, although intensity of infection was not measured (Fig.   4A-C).

Discussion
Asian sh tapeworm identi cation has historically been based on the unique morphology of the scolex, which is inadequate for detecting closely related subtypes (Brabec et al. 2016;Xi et al. 2016). Indeed, in this study, the gross morphology of the AFT and Schyzocotyle sp. genotype Blue Lake (GenBank® accession numbers MT898664-MT898667) was indistinguishable. Schyzocotyle sp. genotype Blue Lake was evident by the low sequence homology (91.9% sequence similarity) to AFT sequences in GenBank® (Table 2), and the phylogenetic grouping of this tapeworm was distinct from other Schyzocotyle species, with strong support from (pp = 1.0) (Fig. 5). This emphasises the need for molecular methods for species identi cation of Schyzocotyle sp. genotype Blue Lake and the AFT in future studies. As this study targeted a partial fragment of the 18S rRNA gene, characterisation of the complete or near full-length of this gene is required to con rm that Schyzocotyle sp. genotype Blue Lake represents a novel species from an unknown origin. Future studies should also aim to morphologically and genetically characterise Schyzocotyle species identi ed by this study, which will also aid our understanding of the potential origins of the AFT in south-western WA (Luo et al. 2002;Xi et al. 2016).
In the present study, S. acheilognathi was identi ed by microscopy and molecular characterisation for the rst time in WA, from introduced gold sh, koi carp and eastern mosquito sh in a modi ed urban wetland. If translocated from this locality (e.g. by humans, birds, oods etc) to adjacent lentic or lotic systems, this parasite would pose a serious threat to the unique freshwater (and possibly estuarine) sh fauna of south-western WA. Several studies have already shown that AFT can infect sh in environments with low to moderate salinity (Ozturk et al. 2002;Bean 2008;Bean and Bonner 2010;İnnal et al. 2016;Sara et al. 2016;McAllister et al. 2017 Anthropogenic stressors, such as habitat loss, global warming, river regulation, secondary salinisation, decreased rainfall, and increased agricultural use has already threatened the survival of many of these endemic species. Further invasion of introduced species can lead to additional stressors, homogenisation, and possible extinction of native freshwater species (Olden et al. 2008;Beatty and Morgan 2013;Morgan et al. 2014). A recent study of the distribution of non-native shes in the southwest of WA found that there has been an increase of up to 63.0% of introduced shes since 1970, with an increase of 44.0% in the previous decade. Of the introduced species, up to 54.0% (seven species, including gold sh and koi carp) have established natural populations in the south-west of WA (Beatty and Morgan 2013).
Invasive sh species may predate or compete with native shes, or alter the habitat to the detriment of native species. An additional threat posed by alien shes, and one that is often under-appreciated, is the introduction of new parasites and pathogens (co-invaders) (Lymbery et al. 2014). Co-invading parasites can be more pathogenic to native shes than to their natural hosts, possibly due to the lack of coevolutionary existence (Lymbery et al. 2014) and as such can cause morbidity and mortality on already declining native populations. In south-western WA, many river systems have seasonally intermittent ow, and sh communities must survive the dry season in small, disconnected refuge pools which could amplify the transmission and effects of co-invading parasites (Lymbery et al. 2020). At least two other species of co-invading parasites, including Ligula intestinalis (Morgan 2003;Chapman et al. 2006) and Lernaea cyprinacea have recently been identi ed in native shes in south-western WA and they also infect diadromous and estuarine shes that venture into freshwaters (Hassan et al. 2008). Lernaea cyprinacea was introduced to WA's native waterways as a co-invader with gold sh and now infects at least six native freshwater species (Hassan et al. 2008). The original source of the S. acheilognathi in WA is unknown, but it appears that either goldish or koi carp are likely to be the source as these are the natural hosts for this particular parasite (Dove and Fletcher 2000;Košuthová et al. 2015;Oros et al. 2015;Salgado-Maldonado et al. 2015;Kuchta et al. 2018).
The histopathological signs of S. acheilognathi in sh hosts are frequently extreme, with the parasite eventually killing its host (Brouder 1999). In the present study there was evidence of intestinal perforation and a large number (>600) of S. acheilognathi detected in each sh; far greater than previous studies, which have reported infection intensities of between 2 and 45 tapeworms (Scholz 1997;Brouder 1999;Košuthová et al. 2015). In this study, the high intensity of tapeworms caused blockage and ischemia in parts of the intestines, resulting in intestinal perforations. The size of the tapeworms varied according to the size of the host and in one koi carp sample a single tapeworm measured over 400 mm in length. Over one quarter of the native shes in south-western WA are listed as endangered with many very small in size (<100 mm total length) . As with many parasitic interactions, the pathogenicity of S. acheilognathi increases in smaller hosts and with the potential to impact native shes further. Gambusia holbrooki has been previously identi ed as a host for S. acheilognathi (Dove et al. 1997;İnnal et al. 2016;McAllister et al. 2017). Gambusia holbrooki is already widespread throughout Australia and because of its very large population sizes represents a potential vector for further spread of the parasite (Morgan and Buttemer 1996;Reynolds 2009). The discovery of S. acheilognathi in WA waters reinforces the importance of invasive sh control programs. Communication and education programs to the wider community are needed to help reduce the release of alien shes into Australian waterways. It should also be noted that S. acheilognathi poses a potential health risk to humans. Although not generally considered zoonotic, there is one case study that identi ed a 32-year-old male who was initially diagnosed with Diphyllobothrium, however, molecular analysis identi ed them as S. acheilognathi, highlighting the lack of morphological distinction in some parasites (Yera et al. 2013). Future research investigations should also study other potential vectors, such as frogs, reptiles, and birds, (Kuchta et al. 2018), for the transmission of Schyzocotyle spp. in south-western WA. It is imperative that the potential spread and impacts of the invasion of S. acheilognathi be identi ed in future research to effectively control and manage the transmission of the parasite.  Table 2 18S rRNA V4 isolates from shes with closest matches to sequences in GenBank®.

Figure 1
Semichon stained heart-shaped scolex of Schyzocotyle acheilognathi showing unarmed bothria and proglottids much narrower than the scolex with rounded edges (Photo: Aileen Elliot). Image taken at 40x magni cation on an Olympus stereo microscope SZX7, at a magni cation range of 0.8-5.6x, with an Olympus DP27 camera using cellSens software.

Figure 2
Semichon stained mature and gravid proglottids and of Schyzocotyle acheilognathi. A Arrows indicate genital pores and B vitelline follicles (Photos: Aileen Elliot). C Gravid proglottids ready for dispersal and D operculated and unembryonated eggs, with arrow indicating operculum (Photos: Aileen Elliot). Images A-C taken at 10-40x magni cation, on an Olympus stereo microscope SZX7, at a magni cation of 0.8-5.6x , with an Olympus DP27 camera using cellSens software. Image D taken at 10-40x magni cation in Bright eld, on an Olympus BX50 compound microscope, using a DP1 universal camera and a Nomarski lter.

Figure 3
Intensity of the Schyzocotyle acheilognathi infection in carp (Photo: Aileen Elliot). Image taken on an Olympus stereo microscope SZX7, at a magni cation of 0.8-5.6x , with an Olympus DP27 camera using cellSens software.

Figure 4
A-C Photos showing Schyzocotyle tapeworms (species uncon rmed) within the intestines of Gambusia holbrooki (Photos: Cindy Palermo). The scolex and bothria were morphologically similar to those found in the koi carp and gold sh. Taken at 10-40x magni cation in Bright eld, on an Olympus BX50 compound microscope, with a Nikon DS-L4 universal camera.

Figure 5
Bayesian phylogenetic tree of a 354 bp alignment (including gaps) of 18S rRNA V4 sequences of known Schyzocotyle acheilognathi species and Schyzocotyle sequences derived from this study. The tree was built using the following parameters: HKY85+GTR+G model; 1,100,000 Markov chain Monte Carlo (MCMC) length; ' burn-in' length of 10,000; subsampling frequency of 200. The tree was rooted with the outgroup sequence Parabothriocephalus gracilis (KR780945) (not shown). Scale-bar indicates the number of nucleotide substitutions per site. Sequences from this study are indicated by bold typeface.