Catching the fish with the worm: a case study on eDNA detection of the monogenean parasite Gyrodactylus salaris and two of its hosts, Atlantic salmon (Salmo salar) and rainbow trout (Oncorhynchus mykiss)

Catching the fish with the worm: a case study on eDNA detection of the monogenean parasite... Background: Environmental DNA (eDNA) monitoring is growing increasingly popular in aquatic systems as a valuable complementary method to conventional monitoring. However, such tools have not yet been extensively applied for metazoan fish parasite monitoring. The fish ectoparasite Gyrodactylus salaris, introduced into Norway in 1975, has caused severe damage to Atlantic salmon populations and fisheries. Successful eradication of the parasite has been carried out in several river systems in Norway, and Atlantic salmon remain infected in only seven rivers, including three in the Drammen region. In this particular infection region, a prerequisite for treatment is to establish whether G. salaris is also present on rainbow trout upstream of the salmon migration barrier. Here, we developed and tested eDNA approaches to complement conventional surveillance methods. Methods: Water samples (2 × 5 l) were filtered on-site through glass fibre filters from nine locations in the Drammen watercourse, and DNA was extracted with a CTAB protocol. We developed a qPCR assay for G. salaris targeting the nuclear ribosomal ITS1 region, and we implemented published assays targeting the mitochondrial cytochrome-b and NADH-regions for Atlantic salmon and rainbow trout, respectively. All assays were transferred successfully to droplet digital PCR (ddPCR). Results: All qPCR/ddPCR assays performed well both on tissue samples and on field samples, demonstrating the applicability of eDNA detection for G. salaris, rainbow trout and Atlantic salmon in natural water systems. With ddPCR we eliminated a low cross-amplification of Gyrodactylus derjavinoides observed using qPCR, thus increasing specificity and sensitivity substantially. Duplex ddPCR for G. salaris and Atlantic salmon was successfully implemented and can be used as a method in future surveillance programs. The presence of G. salaris eDNA in the infected River Lierelva was documented, while not elsewhere. Rainbow trout eDNA was only detected at localities where the positives could be attributed to eDNA release from upstream land-based rainbow trout farms. Electrofishing supported the absence of rainbow trout in all of the localities. (Continued on next page) * Correspondence: haakon.hansen@vetinst.no Norwegian Veterinary Institute, P.O. Box 750, Sentrum, NO-0106 Oslo, Norway Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Rusch et al. Parasites & Vectors (2018) 11:333 Page 2 of 12 (Continued from previous page) Conclusions: We provide a reliable field and laboratory protocol for eDNA detection of G. salaris,Atlantic salmon and rainbow trout, that can complement conventional surveillance programs and substantially reduce the sacrifice of live fish. We also show that ddPCR outperforms qPCR with respect to the specific detection of G. salaris. Keywords: Environmental DNA, Multiplex PCR, Droplet digital PCR (ddPCR), Internal transcribed spacer (ITS), Mitochondrial DNA (mtDNA), Invasive species Background including a control area is described in the Norwegian le- Gyrodactylus salaris Malmberg, 1957 (Monogenea) is an gislation [14]. A strategy to implement treatment of this ectoparasite first described on the skin of Atlantic region has not yet been conclusively devised by the salmon Salmo salar (L. 1758), where it attaches itself to Norwegian authorities, as this watercourse in many as- the host with a haptor, a specialized attachment organ pects is more complicated than previously treated systems. consisting of a large disc with 16 peripheral articulated This results from three basic factors. First, rainbow trout in marginal hooks and a single pair of ventrally orientated the system upstream of the current migration barriers for hamuli [1]. This ~500 μm long parasite [2] has also been salmon have a history of infection with G. salaris [8]. Sec- found on other salmonids such as rainbow trout Onco- ondly, Drammenselva contains a much higher fish species rhynchus mykiss (Walbaum, 1792) [3], brown trout diversity than other treated rivers, which mainly contain Salmo trutta (L., 1758) and Arctic charr Salvelinus alpi- salmonids. Thirdly, this river discharges into a large estuary nus (Linnaeus, 1758) [4]. While most species and popu- with surface water containing low salinity (< 2%) where G. lations of fish which act as hosts, including Baltic salaris can survive for longer periods [15]. In order to de- populations of Atlantic salmon, do not experience cide on measures regarding treatment of this water system, serious consequences of a G. salaris infection [1, 5], At- exact knowledge of the status of infections with G. salaris lantic populations of salmon are highly susceptible to G. in the area is a prerequisite. Rainbow trout farms in the salaris resulting in high mortality rates in mainly northern parts of the Drammen watercourse were infected Norwegian populations (see below). Rainbow trout is with G. salaris in the mid-1980s and later there have been less susceptible, and can sustain infections for long pe- both documented [16, 17] and anecdotal reports of riods, often at low intensities making it an important free-living rainbow trout in the system. There is thus a pos- host when considering spreading between fish farms in sibility that free-living rainbow trout are still present in the Europe [6]. system and these might have sustained the introduced G. In 1975, G. salaris was detected in Norway for the first salaris infection from the 1980s. Therefore, a surveillance time [7–9]. The parasite has since caused severe damage to program [18, 19] has been established to detect any pos- several Atlantic salmon populations [1]. Altogether, fish in sible presence of G. salaris on free-living populations of 50 rivers in Norway have been infected by G. salaris and rainbow trout upstream of the anadromous parts of the extensive eradication programs, mostly using pesticides Drammenselva catchment. such as rotenone, have been carried out in several of these Standard surveillance for fish parasites, including the sur- watercourses [10] since 1981 [11]. Over the last 15 years veillance programs for G. salaris in Norway, involves cap- [12], the eradication programs have been highly successful ture and euthanasia of fish, prior to manual examination and to date the parasite is present only in seven rivers [10]. for the presence of parasites. This is both costly and To document the absence of G. salaris in Norwegian river labour-intensive, and results in the sacrifice of a large num- systems and to detect new infections at an early stage, ber of usually infection-free healthy fish. In recent years, large-scale national surveillance programs are carried out capturing, amplifying and detecting species-specific DNA every year [10, 13]. Present surveillance is based on the fragments of several species in water samples has been catching and killing of numerous Atlantic salmon juveniles established as an accurate low-cost alternative or a comple- in rivers and farms, as well as rainbow trout reared in ment to traditional monitoring [20–23]. This approach, farms, for morphological screening for the presence or ab- harnessing so-called environmental DNA (eDNA), makes sence of G. salaris. In 2016 alone, 6981 fish were killed and useof the knowledgethatall organisms shed cells into their examined [10, 13]. surroundings (excretion, mucus layers, abrasions of epithe- One of the remaining regions where G. salaris is still lial tissue, gametes) [24, 25]. For eDNA monitoring of nat- present is the Drammen region (Buskerud and Vestfold ural waters, the eDNA content represents to a large extent County) in southern Norway, consisting of the rivers a snap-shot of the present living species, with a time lag of Drammenselva, Lierelva and Sandeelva (hereafter referred only some weeks after a species has disappeared from the to by their Norwegian names). The infection region system until eDNA is no longer detectable [26]. Results are Rusch et al. Parasites & Vectors (2018) 11:333 Page 3 of 12 delivered relatively fast and efficiently [27], often at lower (Buskerud County), a small river in the Drammen infec- agent-prevalence than through traditional methods [28]. tion region where Atlantic salmonhasbeen infected with To complement conventional surveillance methods for G. salaris since 1987 [1]. Drammenselva drains from the G. salaris, we aimed at developing an eDNA approach Jotunheimen Mountains in the north, down to for targeted detection of the parasite-host combination Drammensfjorden (Buskerud and Vestfold Counties) in water samples. We applied this method in a which connects the watercourse with the Atlantic Ocean case-study, where eDNA detection by means of species (Fig. 1). The infection region in Drammen incorporates specific quantitative PCR (qPCR) and droplet digital three of the remaining seven rivers in Norway where G. PCR (ddPCR) was used as a supplement to standard sur- salaris is still present. These are: Drammenselva, Lierelva veillance methods for G. salaris, Atlantic salmon and (both Buskerud County) and Sandeelva (Vestfold County) rainbow trout in the Drammen infection region, (Fig. 1), in all of which Atlantic salmon is present. Lierelva Norway. and Sandeelva are smaller rivers with catchment sizes of 309.6 and 193.4 km , respectively, while Drammenselva Methods drains from a much larger area (17,110.8 km ). In the Description of the study area northern part of the Drammen watercourse (see Fig. 1), One part of this study was conducted in the northern several rainbow trout producers can be found. Fish in part of the Drammenselva watercourse (Oppland farms in this area were infected by G. salaris in the County) where a presence of wild rainbow trout popula- mid-1980s and there were many reports of escaped fish tions is possible and the status of G. salaris is unknown. from the farms [14]. However, the fish populations in the The other part of the study was conducted in Lierelva farms were eradicated and all these farms were declared Fig. 1 Map of the Drammen watercourse region with all sampling locations and its location within Norway. Green points represent localities sampled. The thick blue line represents the Drammen watersystem, the thin blue lines represent the main rivers, the red lines indicate rivers where G. salaris is present and the black lines outline the Drammenselva drainage basin. The numbers refer to the sampling sites in Table 1. Pie charts: blue colour indicates detection of Atlantic salmon, red indicates detection of Gyrodactylus salaris and yellow indicates detection of rainbow trout. Rivers flow north to south Rusch et al. Parasites & Vectors (2018) 11:333 Page 4 of 12 free from G. salaris in 1987 [29]. In 1986, G. salaris was according to local conditions (stream size, depth, water also diagnosed from farmed rainbow trout and salmon in flow). Electrofishing was also conducted in Lierelva to the Lake Tyrifjorden which is a part of the Drammen collect salmon juveniles for estimation of the infection watercourse [8, 30]. The fish populations in these farms prevalence and intensity of G. salaris in the same locality were also eradicated, but a short time later, the parasite as water samples were taken. Fish captured for further was detected on salmon juveniles from Drammenselva examination were euthanised following the strict codes and Lierelva [30]. of practice in force in Europe, preserved intact in 96% ethanol and later examined for the presence of Gyrodac- Sample locations tylus spp. using a stereo microscope (Leica MZ 7.5, Leica The sampling sites included eight localities in the northern microsystems, St. Gallen, Switzerland). part of the Drammenselva watercourse, (Fig. 1,Table 1). These sampling sites were chosen as part of a monitoring Water filtering for eDNA sampling program [19] and with the intention of both declaring this At each sampling location, duplicate water samples of 5 region free from G. salaris and mapping the presence of L (2 × 5 l) were collected and filtered on site on to glass free-living rainbow trout. One of these eight sampling sites fibre filters (47 mm AP25 Millipore, 2 μm pore size, was a fish pond at a local trout farm that served as a Millipore, Billerica, USA) using a portable peristaltic rainbow trout positive field control sample. The ninth pump (Masterflex E/S portable sampler, Masterflex, sample was chosen as a positive field control sample for Gelsenkirchen, Germany), tygon tubing (Masterflex) and only G. salaris and Atlantic salmon and collected from a an in-line filter holder (Millipore) according to Strand et stretch in Lierelva (Fig. 1), a river with a confirmed al. [31]. At Lierelva, four samples were taken instead of presence of Atlantic salmon infected with G. salaris. two as this river was intended as a positive field control Within the area where rainbow trout farms can be for G. salaris and Atlantic salmon. Filters were placed in found, the locations of sample nos. 4 and 5 were chosen separate 15 ml Falcon tubes containing cetyl trimethyl based on information obtained from the local authorities ammonium bromide (CTAB) buffer and stored on ice prior to the field work. These samples were taken in directly after filtration. Upon arrival at the laboratory the streams flowing into the main watercourse in order to samples were stored at -20 °C until further analysis. As a avoid positive detections due to outlet water from farms safety precaution and part of the filtering protocol, the situated upstream in the main watercourse. For an indi- entire equipment was disinfected with a 10% bleach so- cation of the sensitivity of the rainbow trout eDNA assay lution after use at each location. Thus, any residual for detection in the field, three samples (nos. 6, 7 and 8) eDNA was broken down and contamination was pre- were taken from the main watercourse at different dis- vented. Before further sampling, the tubes were rinsed tances from the rainbow trout farms. Samples 1 and 2 with sodium thiosulphate to neutralise the bleach solu- were taken upstream of the area containing rainbow tion, and then flushed with ambient river water directly trout farms. before sampling. Electrofishing and Gyrodactylus counts DNA extraction Electrofishing was carried out in rivers and tributaries in DNA was extracted from the filters according to a CTAB the Drammen watercourse to reveal the possible pres- protocol described in Strand et al. [31], with the excep- ence of rainbow trout, using this standard surveillance tion that the CTAB buffer contained no added 1% method. The area examined was chosen on site 2-mercapto-ethanol. During extraction each filter was Table 1 List of sampling sites including location, sampling date and amount of water filtered Site no. Site name Water filtered (l) Coordinates Date 1 Storåne at Ala camping 5 (×2) 61.1473N, 8.7121E 26.06.2017 2 Storåne at Tørpegårdsvegen/bru 5 (×2) 61.1522N, 8.7250E 26.06.2017 3 Trout farm 5 (×2) 61.0379N, 9.0466E 14.11.2016 4 Leireelvi at Leira/Garlivegen 5 (×2) 60.9742N, 9.2936E 26.06.2017 5 Leireelvi at Leira camping 5 (×2) 60.9680N, 9.2884E 26.06.2017 6 Lake Strondafjorden at Faslefoss 5 (×2) 60.9671N, 9.2889E 26.06.2017 7 River Begna at Bagn 5 (×2) 60.8198N, 9.5612E 26.06.2017 8 River Begna at Nes 5 (×2) 60.5628N, 9.9929E 26.06.2017 9 Lierelva at Sjåstad 5 (×4) 59.8580N, 10.2213E 31.08.2017 Rusch et al. Parasites & Vectors (2018) 11:333 Page 5 of 12 split into two subsamples (A and B) due to volume matching them against the database of the National restrictions imposed by centrifuge size and extracted Centre for Biotechnology Information (NCBI, http:// separately. An environmental control and a blank extrac- www.ncbi.nlm.nih.gov/) nucleotide database using the tion control were included as a precaution to detect any Basic Local Alignment Search Tool (BLASTn). The aim of possible contamination during DNA extraction. The the new qPCR assay was to attempt to obtain the best pos- blank extraction control consisted of a Falcon tube con- sible sensitivity and specificity for eDNA applications. taining the CTAB buffer but no filter, which was then Similar to the assay Collins et al. [32] designed, the newly processed in the same way as all other tubes containing designed assay is not able to distinguish between G. salaris buffer and filters. The environmental control used in this and G. thymalli Žitňan, 1960 as these two species have in- study consisted of an Eppendorf tube containing 200 μl distinguishable ITS sequences [33]. PCR-grade water. This tube remained open in the fume The assays used for eDNA-detection of Atlantic hood throughout the entire process of extraction. salmon and rainbow trout (Table 2) follow Matejusova et al. [34] and Wilcox et al. [35], respectively. These were successfully tested on DNA extracts from tissue of PCR-based assays for eDNA detection of G. salaris, Atlantic salmon and rainbow trout before use in the rainbow trout and Atlantic salmon current study (data not shown). The ddPCR assay for G. A quantitative PCR assay (qPCR) using species-specific salaris, rainbow trout and Atlantic salmon applied the primers and a minor groove binder (MGB) probe targeting same primers and probes as the qPCR. the G. salaris internal transcribed spacer region 1 (ITS1) was developed. The ITS1 sequence published as GenBank accession no. DQ898302 was used as template and the spe- Evaluation of qPCR and ddPCR assay specificity cificity of the designed primers and probe was checked The specificity of the assay was tested on DNA extracts of against closely related species and other species that might G. salaris collected from three different locations in be present in Norwegian watercourses: G. salmonis Yin & Norway in addition to DNA extracts from the following Sproston, 1948 (GQ368233), G. truttae Gläser, 1974 other species present in the collection at the NVI: G. thy- (AJ132260), G. lucii Kulakovskaya, 1952 (EU304811), G. malli, G. salmonis, G. arcuatus, G. lucii and G. derjavi- arcuatus Bychowsky, 1933 (JN703797) and G. derjavi- noides. Species identification of these samples had been noides Malmberg, Collins, Cunningham & Jalali, 2007 done previously by sequencing of ITS (results not shown). (EU304810). Multiple sequences were aligned using AlignX We also ran the same samples with the previously pub- (Vector NTI Advance 11.5, Invitrogen, Carlsbad, USA). lished assay [32] to compare the specificity and sensitivity The design of primers and probe was performed manually, of the assays. ddPCR applies the same primers and probes targeting ITS1 sequence regions displaying the highest se- as qPCR and the specificity was tested on G. derjavinoides quence diversity between G. salaris and the species listed due to the low level of cross amplification shown in a pre- above. The final primer and probe sequences (Table 2) viously published assay [32]. The ddPCR assay was also partly overlap with those previously published for this tested on isolates of G. salaris obtained from fish from parasite [32] and their specificity was confirmed through Lierelva to determine optimal annealing temperature. Table 2 Primers and probes for Gyrodactylus salaris, rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar) used in the present study. The probes used are TaqMan MGB probes with either Fam or Hex reporter dyes Target species/gene Name Primer/probe Sequence (5'-3') Reference G. salaris/ITS G.sal208F Forward GGTGGTGGCGCACCTATTC Present study G.sal149R Reverse ACGATCGTCACTCGGAATCGAT Present study G.sal188P Probe (FAM)CAAGCAGAACTGGTTAAT(MGBNFQ) Present study G. salaris/ITS F Forward CGATCGTCACTCGGAATCG Collins et al. [32] R Reverse GGTGGCGCACCTATTCTACA Collins et al.[32] Gsal2 Probe (FAM)TCTTATTAACCAGTTCTGC(MGBNFQ) Collins et al. [32] O. mykiss/cytb RBTF Forward AGTCTCTCCCTGTATATCGTC Wilcox et al. [35] RBTR Reverse GATTTAGTTCATGAAGTTGCGTGAGTA Wilcox et al. [35] RBTP Probe (FAM)CCAACAACTCTTTAACCATC(MGBNFQ) Wilcox et al. [35] S. salar/cytb Salmonid Cyt B FOR Forward CGGAGCATCTTTCTTCTTTATCTGT Matejusova et al. [34] S. salar REV Reverse ACTCCGATATTTCAGGTTTCTTTATATAGA Matejusova et al. [34] S. salar Cyt B Probe Probe (HEX)CCAACAACTCTTTAACCATC-(MGBNFQ) Matejusova et al. [34] Rusch et al. Parasites & Vectors (2018) 11:333 Page 6 of 12 qPCR and ddPCR protocols for G. salaris eDNA detection the same primers and probes as for the singleplex reac- All qPCR analyses were carried out on an Mx3005P tions. This duplex method was set up by running the fol- qPCR system (Stratagene, San Diego, USA). Droplet lowing 22 μl reactions for each eDNA extract in digital PCR was performed on a QX200 AutoDG Drop- duplicates: 11 μl ddPCR Supermix for probes - no dUTP let Digital PCR System (Bio-Rad, Hercules, USA). (Bio-Rad), 0.99 μlof20 μM of Salmonid Cyt B FOR and For qPCR detection of G. salaris, three qPCR repli- S. salar REV primers, 0.55 μlof10 μM S. salar Cyt B cates were run for each eDNA extract in the following Probe, 0.99 μlof 20 μM of G.sal208F and G.sal149R 25 μl reactions: 1.25 μl of PCR-grade water, 12.5 μlof primers, 0.275 μlof 20 μM G.sal188P probe, 0.215 μl ExTaq mastermix (Takara Biotechnology, Dalian, China), PCR-grade water and 1 μl of restriction-enzyme mix 1.5 μl of each 10 μM primer (forward and reverse), 0.75 consisting of 0.2 μl HindIII, 0.1 μl buffer (10×), 0.7 μl μlof10 μM probe, 0.5 μl of Rox II reference dye and 5 PCR-grade water and 5 μl of DNA sample. The optimal μl of eDNA template. The qPCR cycling conditions were primer-probe concentration for both assays had been as follows: an initial denaturation at 95 °C for 15 min; 45 determined to be 900:250. The same cycling conditions cycles of denaturation at 94 °C for 30 s, annealing at 54 ° were used as in the G. salaris singleplex reaction. C for 45 s, and extension at 72 °C for 1 min; followed by For qPCR detection of O. mykiss, three qPCR repli- a final elongation phase at 72 °C for 10 min. cates were run for each eDNA extract in the following The following 22 μl reactions were run for each eDNA 12 μl reactions: 2.35 μl of PCR-grade water, 6.25 μlof extract on ddPCR: 11 μl ddPCR Supermix for probes - ExTaq mastermix (Takara), 0.3 μlof10 μM RBTF no dUTP (Bio-Rad), 1.98 μl of each 10 μM primer, 0.55 forward primer and 0.6 μlof 10 μM RBTR reverse pri- μlof10 μM probe, 0.49 μl PCR-grade water and 1 μlof mer, 0.25 μlof10 μM RBTP probe, 0.25 μl of Rox II restriction-enzyme mix consisting of 0.2 μl HindIII, 0.1 reference dye and 2 μl of DNA template. The qPCR μl buffer (10×) and 0.7 μl PCR-grade water and 5 μlof (Stratagene) cycling conditions were as follows: an initial DNA sample. The optimal primer-probe concentration denaturation at 95 °C for 1 min; 45 cycles of denatur- was determined to be 900:250 and the optimal annealing ationat94°Cfor 30 s, annealingat54°Cfor 45 s temperature of 58 °C was confirmed through amplifica- and extension at 72 °C for 1 min; followed by a final tion tests along a temperature gradient. Here, we used elongation phase at 72 °C for 10 min. We used a the HindIII restriction enzyme to fragment the repetitive cut-off at Cq 41 for the rainbow trout-assay, similar multi-copy ITS regions within the nuclear ribosomal to the suggestion for eDNA qPCR detection cut-off in DNA in order to ensure that the targeted DNA copies Agersnap et al. [37]. were distributed among different droplets for accurate For the singleplex ddPCR detection of rainbow trout, quantification. the following 22 μl reactions for each eDNA extract To allow for sufficient time for the restriction enzymes were run in duplicates: 11 μl ddPCR Supermix for to digest, the plate was sealed using Microseal ‘B’ plate Probes - no dUTP (Bio-Rad), 0.99 μl of RBTF 10 μM sealing film (Bio-Rad), wrapped in tin foil and left on the forward primer, 1.98 μlof10 μM RBTR reverse primer, bench for 20 min. Droplet generation in the QX200 0.55 μlof 10 μM RBTP probe, 2.48 μl PCR-grade water AutoDG Droplet Digital PCR System (Bio-Rad) creates and 5 μl of DNA template. The optimal primer-probe an emulsion with 20 μl of the 22 μl originally pipetted concentration for both assays had been determined to into each well, resulting in a 10% loss of template and be 450:900:250 for forward primer, reverse primer and mastermix. After generation of the droplets, the new probe, respectively, which follows the suggestions in plate was immediately transferred to a TM100 thermo- Wilcox et al. [35]. The same cycling conditions were cycler (Bio-Rad) and the QX200 Droplet Digital PCR used as in all other ddPCR reactions. system (Bio-Rad) with the following cycling conditions: An initial denaturation at 95 °C for 10 min; 45 cycles of Calculation of eDNA copies denaturation at 94 °C for 30 s, annealing at 58 °C for 60 The number of eDNA copies (for each target species) s; followed by a final elongation phase at 58 °C for 10 per litre of water for each sample is calculated min. The threshold for a positive sample was set at three according to the following formula, also used by positive droplets per well according to Dobnik et al. Agersnap et al. [37]: [36]. To ensure the validity of each run, positive and blank PCR controls containing G. salaris DNA and rdd distilled water, respectively, were run on each plate for C ¼ both qPCR and ddPCR. To be able to detect G. salaris and Atlantic salmon simultaneously in future surveillance programmes in where C is the number of target-eDNA copies per litre Norwegian rivers, we also tested a duplex method using of filtered water, C is the ddPCR calculation of eDNA rdd Rusch et al. Parasites & Vectors (2018) 11:333 Page 7 of 12 copy numbers per reaction volume (20 μl), adjusted locations 3, 4, 5, 6 and 7 (see Table 3). One of the five for a 10% loss during droplet generation, V is the positive sampling sites (no. 6) was at the outlet of the total elution volume after extraction, V is the volume lake into which all the rainbow trout farms drain, while of eluted extract used in the ddPCR reaction, V is another (no. 7) was found in the main river 25 km the volume of filtered water. The copy numbers of downstream of the outlet. According to new information both subsamples (A and B) were added together, thus from local authorities we received upon enquiry after revealing the number of eDNA copies per litre of any our analyses detected rainbow trout DNA in samples 4 given sample. Calculation of eDNA copy numbers per and 5, these locations were indeed also situated roughly reaction volume was performed by the QuantaSoft 400 and 1200 m, respectively, downstream of a trout software (v.1.7.4, Bio-Rad) and was estimated using farm (see Table 3). None of the field samples in the the ratio between positive and negative droplets northern part of the Drammenselva watercourse yielded within a sample, using Poisson statistics. a positive result when tested against G. salaris, neither did the rainbow trout positive control at the trout farm. Results All extraction blank controls and environmental blank qPCR assay optimisation and specificity tests controls were negative, both in qPCR and ddPCR. The current assay proved slightly more sensitive (by ~0.5 Cq) towards G. salaris than the assay in Collins Conventional monitoring methods 10-1/slope et al. [31]. The PCR efficiency ([E = ]-1)× At location 1, electrofishing of an area of roughly 300 100 calculated from triplicates of non-diluted and m yielded seven juvenile brown trout. Two juvenile three 10-fold dilutions of a DNA extract originating brown trout were caught at location 2 after electrofish- from a single parasite, was 100 % (Cq = 20.5 to 30.6, ing an area of c.200 m . At location 3, electrofishing was slope = 3.312) (not shown). The qPCR assay for G. carried out in selected pot-holes along a stretch of 150 salaris yielded negative qPCR results for all other m. A high density of brown trout with sizes ranging species except G. salaris (and G. thymalli as previ- from juveniles up to 500 g adults was observed. At the ously explained), except for a low level of fourth location, electrofishing was carried out along a cross-reaction towards the tested specimen of G. der- stretch of 200 m. Several minnows Phoxinus phoxinus javinoides (Cq = 35.6). (L., 1758) were observed and many brown trout (juve- niles to 300 g) were captured in the stream while elec- Optimisation of ddPCR assay and specificity tests trofishing. No electrofishing was carried out at locations Both the qPCR assay (primers and probes) for G. salaris 5, 6 and 7 as none of these locations were suitable for developed in this study and the assays for rainbow trout electrofishing. A total of 21 Atlantic salmon with a and Atlantic salmon [32, 34] were transferable to the length of 9.6 cm (± SD 3.6 cm) were caught in Lierelva. ddPCR platform without further optimisation, using an The parasite prevalence and intensity on these fish was annealing temperature of 58 °C. Unlike the qPCR assay determined to be 85.7% and 83 parasites (± SD 63), however, the ddPCR assay showed no signs of cross respectively. Throughout the entire electrofishing, no amplification of G. derjavinoides. rainbow trout were caught. eDNA monitoring of G. salaris, Atlantic salmon and Discussion rainbow trout In the present study, eDNA monitoring is used for the The positive control field samples for G. salaris taken first time to detect the monogenean parasite G. salaris from Lierelva all yielded positive results in qPCR with along with two of its hosts, Atlantic salmon and rainbow Cq-values ranging from 24.76 to 35.86, and in ddPCR trout. Detections were successfully obtained both in all with eDNA copies/l ranging from 371,440 to 560, re- singleplex reactions (qPCR and ddPCR) and in a duplex spectively. For Atlantic salmon, the eDNA copy numbers reaction (ddPCR) targeting both G. salaris and Atlantic ranged from 10,160 (sample 9/2) to 7520 (sample 9/4) salmon. The prevalence in susceptible Atlantic salmon (Table 3) at an average of 8948 copies (± SD = 945). populations most often reaches 100 % [11]. In general, The two positive control field samples for rainbow the infection grows exponentially on non-responding trout obtained at the trout farm in 2016 tested positive hosts and may reach several thousand individuals per for rainbow trout (Cq 17.48 and Cq 17.50; 8,800,000 fish [5]. In our study, the G. salaris infected Atlantic eDNA copies/l and 7,848,000 eDNA copies/l, respect- salmon individuals caught in Lierelva were only moder- ively) (see Table 3). Of the other 18 water samples that ately infected (prevalence of 85.7%, mean parasite abun- were collected at the eight sampling points in June and dance of 83 parasites). Here G. salaris eDNA was August 2017, five were positive for rainbow trout. Posi- detected in amounts ranging from 500 to > 350,000 tive samples for rainbow trout were obtained from copies per litre of water in the same river stretch. These Rusch et al. Parasites & Vectors (2018) 11:333 Page 8 of 12 Table 3 Overview of results from qPCR and ddPCR analyses for Gyrodactylus salaris (ITS), Oncorhynchus mykiss (CytB) and Salmo salar (CytB) at each sampling site. List of sampling sites including amount of water filtered, number of samples per site (each sample constitutes of one filter), the Cq value (from qPCR) and number of eDNA copies per litre (ddPCR) from all filters taken at each point, respectively. eDNA copies per litre are abbreviated as eDNA/l. No detection is indicated with a minus (-) for qPCR and a zero for ddPCR and those samples where analysis was not applicable are indicated with NT Site Site name Sample Volume Gyrodactylus salaris Oncorhynchus mykiss Salmo salar no. (l) a b a a b, a qPCR ddPCR qPCR ddPCR qPCR ddPCR 1 Storåne at Ala camping 1 1 - - - 0 - 0 21 - - - 0 - 0 2 Storåne at Tørpegårdsvegen/bru 1 1 - - - 0 - 0 21 - - - 0 - 0 3 Trout farm 1 1 - - 17.48 7,848,000 - 0 2 1 - - 17.50 8,800,000 - 0 4 Leireelvi at Leira/Garlivegen 1 1 - - 29.62 1624 - 0 2 1 - - 29.09 3816 - 0 5 Leireelvi at Leira camping 1 1 - - 30.05 2240 - 0 2 1 - - 30.02 2124 - 0 6 Lake Strondafjorden at Faslefoss 1 1 - - 32.3 560 - 0 2 1 - - 31.68 576 - 0 7 River Begna at Bagn 1 1 - - > cut-off 0- 0 2 1 - - 36.91 22 - 0 8 River Begna at Nes 1 1 - - - 0 - 0 21 - - - 0 - 0 9 River Lierelva at Sjåstad 1 1 34.52 560 - NT NT 9200 2 1 33.56 840 - NT NT 10,160 3 1 33.94 864 - NT NT 7520 4 1 24.89 371,440 - NT NT 8912 Run as singleplex Run as duplex Cut-off value was set at Cq 41 results strongly indicate that eDNA analysis of samples simultaneous detection of parasite and host. Using the obtained by water filtering can indeed be used for moni- protocol for filtration, DNA-extraction and the analysis toring the occurrence of G. salaris in freshwater ecosys- we describe here, it is not only possible to detect the tems containing natural Atlantic salmon populations. parasite G. salaris but also two of its hosts within on Environmental DNA-detection is a promising tool that single sample. With the use of other assays, the presence can be used to supplement or even replace classical of virtually any aquatic host-pathogen complex can be surveillance where it produces fast and robust results. detected and monitored, provided that the filter size is This is reflected in the ever growing number of assays appropriate to capture eDNA from the target organism. being developed to monitor parasites which infect fish. The aim of the G. salaris qPCR assay designed in the These include both ectoparasites like Amyloodinium present study was to achieve an optimal combination of ocellatum Brown, 1931 [38], Chilodonella hexasticha both specificity and sensitivity, and the assay was chosen Kiernik, 1909 [39]or Neobenedenia girellae Hargis, 1955 over the one previously published by Collins et al. [32] [40] and endoparasites such as Opisthorchis viverrini due to its slightly higher sensitivity. Both the qPCR assay Poirier, 1886 [41], Ichthyophonus spp. [42] and myxozo- presented in this paper and the qPCR assay designed by ans [43, 44]. Unlike traditional monitoring, there is no Collins et al. [32] display a low-level amplification of need to kill large numbers of fish or to carry out Gyrodactylus derjavinoides. However, this issue was not time-consuming manual examinations. Thus, the eDNA observed when applying the newly designed primers and monitoring method has far-reaching potential as it re- probe in ddPCR. Any assay for Gyrodactylus salaris duces the time and cost of sampling and improves fish targeting the ITS1 region will yield positive results for G. welfare. A further advantage of this method is the thymalli since these two species have nearly identical Rusch et al. Parasites & Vectors (2018) 11:333 Page 9 of 12 sequences [33] and it is therefore impossible to differen- 1000-fold higher amount of eDNA in sample 9/4. This tiate between them in this way. This does not affect the calculation is also reflected in the ddPCR results where monitoring of G. salaris in systems uninhabited by gray- an increase from 560 copies/l to 371,440 copies/l was ling, the host for G. thymalli. In systems where grayling observed. This assumption is substantiated by the fact occur, negative samples would still indicate the absence that Gyrodactylids are reported to consist of roughly of the parasite. A positive detection would certainly 1000 cells [1]. The possibility that one sometimes might require additional examination and attention. Here, one catch a whole parasite specimen in the filter does not option would be to design assays targeting the more vari- pose a problem for a simple proof of presence detection, able mitochondrial cytochrome oxidase gene (see, e.g. but in fact increases the certainty of the results. How- Meinilä et al. [45], Hansen et al. [46]). ever, while some studies have demonstrated a correlation In the present eDNA study, as well as for most other between biomass and eDNA concentration [51], quanti- applications, the low level of cross-reaction against G. fication of parasites and establishing an agent-level derjavinoides when using qPCR poses no problem. If a would, in this case, result in an overestimation of para- population of fish were infected with a high number of site numbers. The use of a pre-filter such as fitting a G. derjavinoides and a low number of G. salaris, analysis plankton net in front of the filter with a mesh size small with qPCR could yield ambiguous results. We therefore enough to prevent an entire specimen to pass on to the recommend the use of ddPCR analysis since this method filter may solve this problem of overestimation. In com- bypasses the problem of cross-amplification. Alterna- parison to the results for G. salaris, the copy number for tively, sampling by electrofishing followed by manual Atlantic salmon eDNA was fairly constant in all four examination and standard species identification could be samples at an average of 8948 copies (± SD = 945) as carried out in this particular case. displayed in Fig. 2. This indicates a constant emission We detected rainbow trout eDNA at four locations in rate of eDNA into the water by Atlantic salmon which the northern part of the Drammen watercourse in has also been observed in other studies [52]. addition to the sample taken at the trout farm (sample no. 3). We observed an apparent decline in eDNA Comparison of qPCR/ddPCR monitoring concentration with increasing distance from the source Quantitative real-time PCR (qPCR) offers the possibility (area with trout farms, sample nos. 6 and 7). This corre- to measure the rate of generation of the amplified product sponds with data from studies that examine the dilution after each cycle, thus making it possible to calculate the effects of eDNA in river ecosystems [47, 48]. However, amount of copies in the original sample. Previous studies the number and the distribution of sampling points in have demonstrated that quantification of biomass and cal- this study were not comprehensive enough to examine a culation of population size through using qPCR is possible gradient thoroughly. Extensive electrofishing at each [22, 53]. ddPCR, which now allows the user to operate on sampling point produced no evidence for the presence a nanolitre rather than on a microlitre scale, enables even of rainbow trout in the streams. We therefore attribute more precise detection and absolute quantification of tar- all positive samples to eDNA discharge/emission from get molecules while simultaneously removing the need for trout farms and assume the areas and streams of the standard curves [51, 54]. Our results demonstrate this pre- northern part of the Drammenselva watercourse that cision by detecting both rainbow trout and G. salaris at were tested to be free from wild populations of rainbow very low copy levels with 22 eDNA copies/l and 560 trout. The occurrence of these positive samples reveals eDNA copies/l, respectively. Furthermore, this technology one of the pitfalls of the eDNA methodology, as it sim- has been proven to perform better on inhibition prone ply points out the presence of eDNA from the targeted samples than the predecessor qPCR [55]. This is a particu- organism without verifying the actual presence of the or- lar advantage when analysing environmental samples ganism within the examined body of water [20, 49, 50]. which often tend to include PCR inhibitors [56–58]. Our It does, however, also highlight the sensitivity of this study also shows that ddPCR seems to surpass qPCR re- method. garding specificity, as there was no cross-amplification of One of the four filter samples taken at Lierelva, the G. derjavinoides in the G. salaris assay although the same river with a known presence of G. salaris, displayed a primer-probe combinations were used. We speculate that significantly higher signal than the other three filters, this is due to the lower copy numbers of both target and even though the very same location was sampled. These non-target DNA per reaction (droplets) in the ddPCR results were observed in qPCR, and both the singleplex system. Ideally, one droplet contains only one copy of the and multiplex ddPCR reactions. We presume that this is target DNA and only a few non-target copies, thus redu- due to one or more whole specimens of G. salaris being cing the possibility of unspecific amplification. picked up on this particular filter. The signal difference For a more precise monitoring of G. salaris and its in qPCR is roughly ten cycles which would suggest a hosts, further research and development is needed in Rusch et al. Parasites & Vectors (2018) 11:333 Page 10 of 12 Fig. 2 Visual output from the duplex ddPCR for G. salaris in Channel 1 (blue) and Atlantic salmon in Channel 2 (green) on the samples taken at Lierelva. Wells containing samples 9/1B and 9/3B are not displayed and were excluded due to insufficient droplet generation. Each blue and green point represents a positive amplification of respective DNA template. The horizontal purple line represents the threshold and the black points represent negative droplets. The eDNA copy number for G. salaris is markedly higher in two of the wells containing samples 9/4A and 9/4B. However, the copy number of Atlantic salmon eDNA remains relatively stable in all four samples order to improve the specificity of the G. salaris assay qPCR when screening samples for G.salaris.Further to distinguish from G. thymalli, as well as to determine studies are needed to determine the limit of detection when it is no longer possible to obtain a positive eDNA regarding eDNA and to compare the eDNA signal result (limit of detection) when the parasite load per against fish parasite load in experimental and natural fish drops. settings. Abbrevations Conclusions BLASTn: Basic Local Alignment Search Tool; Cq-Value: quantification cycle; We have successfully designed and implemented a CTAB: cetyl trimethyl ammonium bromide; ddPCR: droplet digital PCR; eDNA: environmental DNA; ITS1: internal transcribed spacer 1; MGB: minor method for eDNA detection of an aquatic groove binder; mtDNA: mitochondrial DNA; NADH: nicotinamide adenine host-parasite system, specifically G. salaris and its dinucleotide dehydrogenase; qPCR: quantitative PCR two hosts Atlantic salmon and rainbow trout. Thus, Acknowledgements we demonstrate for the first time that eDNA moni- We thank Saima Nasrin Mohammad and Mari Darrud, Norwegian Veterinary toring canbeusedfor thedetection of G. salaris and Institute (NVI), for technical assistance with molecular work and field work, its host Atlantic salmon in natural freshwater systems respectively. with a moderately infected salmon population. Fur- Funding thermore,wehavedeterminedthe assaywedesigned This study is part of the “eDNAqua-Fresh” PhD project, funded by the NVI. to be species-specific and demonstrated the usefulness No external grants were received for this work. of eDNA methodology when examining a river system Availability of data and materials for the presence of G. salaris. Within the paper we All data generated or analysed during this study are included in this present a protocol, both field and laboratory, on how published article. to conduct eDNA monitoring of G. salaris and Atlan- Authors’ contributions tic salmon successfully, using a duplex ddPCR. We HH, JR, SH and TV planned the study. SH was in charge of the field work and show that ddPCR appears to be a better tool than performed the electrofishing and JR and HH participated in the field work. JR Rusch et al. Parasites & Vectors (2018) 11:333 Page 11 of 12 carried out water filtering, qPCR analyses, ddPCR analyses and optimisation kommuner, Akershus, Buskerud og Vestfold. 2016. https://lovdata.no/ of ddPCR assays. DS took part in the design and optimisation of ddPCR dokument/FV/forskrift/2016-07-07-919. Accessed 30 Apr 2018. assays. TM and HH designed the qPCR assay for G. salaris and TM optimized 15. Soleng A, Bakke TA. Salinity tolerance of Gyrodactylus salaris the qPCR assay. JR and HH wrote the first draft of the manuscript and all (Platyhelminthes, Monogenea): laboratory studies. Can J Fish Aquat Sci. authors contributed in the writing of the manuscript. All authors read and 1997;54:1837–45. approved the final manuscript. 16. Brabrand Å. Fiskeribiologiske undersøkelser i Slidrefjorden. Oppland Fylke: Vurdering av tilslag på settefisk: Laboratorium for ferskvannsøkologi og Ethics approval and consent to participate innlandsfiske (LFI), Zoologisk Museum, Universitetet i Oslo; 1988. p. 44. No approval from Institutional Animal Care and Use Committee (IACUC) or 17. Thomassen G, Norum I. Fiskeundersøkelse i Strondafjorden. Fylkesmannen i ethics committee was necessary. No experiments that involved fish were Oppland, miljøvernavdelingen, Rapport nummer 17; 2012. p. 18. performed. All fish were euthanised following the strict codes of practice in 18. Hytterød S, Mo TA, Hansen H, Tavornpanich S. The post-treatment force in Europe. surveillance programme to ascertain freedom from infection with Gyrodactylus salaris in Atlantic salmon, Annual report 2013. Norwegian Competing interests Veterinary Institute: Oslo; 2014. The authors declare that they have no competing interests. 19. Hytterød S, Rusch JC, Darrud M, Nasrin Mohammad S, Hansen H. Mapping the occurrence of Gyrodactylus salaris upstream of the natural anadromous region of the Drammenselva catchment. In: Annual Report 2016. Norwegian Publisher’sNote Veterinary Institute: Oslo; 2017. Springer Nature remains neutral with regard to jurisdictional claims in 20. Evans NT, Shirey PD, Wieringa JG, Mahon AR, Lamberti GA. Comparative published maps and institutional affiliations. cost and effort of fish distribution detection via environmental DNA analysis and electrofishing. Fisheries. 2017;42:90–9. Author details 21. Jerde CL, Chadderton WL, Mahon AR, Renshaw MA, Corush J, Budny ML, Norwegian Veterinary Institute, P.O. Box 750, Sentrum, NO-0106 Oslo, et al. Detection of Asian carp DNA as part of a Great Lakes basin-wide Norway. Department of Biosciences, University of Oslo, P.O. Box 1066, surveillance program. Can J Fish Aquat Sci. 2013;70:522–6. Blindern, NO-0316 Oslo, Norway. Faculty of Veterinary Medicine, Norwegian 22. Rees HC, Maddison BC, Middleditch DJ, Patmore JRM, Gough KC. The University of Life Sciences, P.O. Box 8146, Dep, NO-0033 Oslo, Norway. detection of aquatic animal species using environmental DNA - a review of eDNA as a survey tool in ecology. J Appl Ecol. 2014;51:1450–9. Received: 31 January 2018 Accepted: 25 May 2018 23. Smart AS, Tingley R, Weeks AR, van Rooyen AR, McCarthy MA. Environmental DNA sampling is more sensitive than a traditional survey technique for detecting an aquatic invader. Ecol Appl. 2015;25:1944–52. References 24. Mächler E, Deiner K, Spahn F, Altermatt F. Fishing in the water: effect of 1. Bakke TA, Cable J, Harris PD. The biology of gyrodactylid monogeneans: The sampled water volume on environmental DNA-based detection of “Russian-doll killers”. Adv Parasitol. 2007;64:161–376. macroinvertebrates. Environ Sci Technol. 2016;50:305–12. 2. Malmberg G. Om förekomsten av Gyrodactylus på svenska fiskar. Skrifter 25. Deiner K, Fronhofer EA, Mächler E, Walser JC, Altermatt F. Environmental utgivna av Södra Sveriges Fiskeriförening, Årsskrift. 1957;1956:19–76. DNA reveals that rivers are conveyer belts of biodiversity information. Nat 3. Malmberg G, Malmberg M. Species of Gyrodactylus (Platyhelminthes, Commun. 2016;7:12544. Monogenea) on salmonids in Sweden. Fish Res. 1993;17:59–68. 26. Dejean T, Valentini A, Duparc A, Pellier-Cuit S, Pompanon F, Taberlet P, 4. Bakke TA, Harris PD, Jansen PA, Hansen LP. Host specificity and dispersal Miaud C. Persistence of environmental DNA in freshwater ecosystems. PLoS strategy in gyrodactylid monogeneans, with particular reference to One. 2011;6:e23398. Gyrodactylus salaris (Platyhelminthes, Monogenea). Dis Aquat Organ. 1992; 27. Gustavson MS, Collins PC, Finarelli JA, Egan D, Conchuir RO, Wightman GD, 13:63–74. et al. An eDNA assay for Irish Petromyzon marinus and Salmo trutta and field 5. Bakke TA, Harris PD, Cable J. Host specificity dynamics: observations on validation in running water. J Fish Biol. 2015;87:1254–62. gyrodactylid monogeneans. Int J Parasitol. 2002;32:281–308. 28. Wittwer C, Stoll S, Strand D, Vrålstad T, Nowak C, Thines M. eDNA-based 6. Hansen H, Cojocaru CD, Mo TA. Infections with Gyrodactylus spp. crayfish plague monitoring is superior to conventional trap-based (Monogenea) in Romanian fish farms: Gyrodactylus salaris Malmberg, 1957 assessments in year-round detection probability. Hydrobiologia. 2018;807: extends its range. Parasit Vectors. 2016;9:444. 87–97. 7. Johnsen BO. The effect of an attack by the parasite Gyrodactylus salaris on 29. Mo TA. Virksomheten i 1987 og program for virksomheten i 1988: the population of salmon parr in the river Lakselva, Misvaer in northern Gyrodactylus-undersøkelsene ved Zoologisk Museum, Universitetet i Oslo; Norway. Astarte. 1978;11:7–9. 8. Johnsen BO, Jensen AJ. Infestations of Atlantic salmon, Salmo salar,by 30. Mo TA. Variations of opisthaptoral hard parts of Gyrodactylus salaris Gyrodactylus salaris in Norwegian rivers. J Fish Biol. 1986;29:233–41. Malmberg, 1957 (Monogenea, Gyrodactylidae) on rainbow trout 9. Hansen H, Bachmann L, Bakke TA. Mitochondrial DNA variation of Oncorhynchus mykiss (Walbaum, 1792) in a fish farm, with comments Gyrodactylus spp. (Monogenea, Gyrodactylidae) populations infecting on the spreading of the parasite in south-eastern Norway. Syst Atlantic salmon, grayling, and rainbow trout in Norway and Sweden. Int J Parasitol. 1991;20:1–9. Parasitol. 2003;33:1471–8. 31. Strand DA, Jussila J, Johnsen SI, Viljamaa-Dirks S, Edsman L, Wiik-Nielsen J, 10. Hytterød S, Hansen H, Johansen K, Larsen S. The surveillance programme et al. Detection of crayfish plague spores in large freshwater systems. J Appl for Gyrodactylus salaris in Atlantic salmon and rainbow trout in Norway Ecol. 2014;51:544–53. 2016. In: Annual Report 2016. Oslo: Norwegian Veterinary Institute; 2017. 32. Collins CM, Kerr R, McIntosh R, Snow M. Development of a real-time PCR 11. Johnsen BO, Møkkelgjerd PI. The parasite Gyrodactylus salaris on salmon assay for the identification of Gyrodactylus parasites infecting salmonids in parr in Norwegian rivers, status report at the beginning of year 2000, vol. northern Europe. Dis Aquat Organ. 2010;90:135–42. 617. Trondheim: NINA Oppdragsmelding; 1999. (In Norwegian) 33. Ziętara MS, Lumme J. Speciation by host switch and adaptive radiation in a 12. Sandodden R, Brazier M, Sandvik M, Moen A, Wist AN, Adolfsen P. fish parasite genus Gyrodactylus (Monogenea, Gyrodactylidae). Evolution. Eradication of Gyrodactylus salaris infested Atlantic salmon (Salmo salar)in 2002;56:2445–58. the Rauma River, Norway, using rotenone. Manage Biol Inv. 2018;9:10. 13. Hytterød S, Darrud M, Nasrin Mohammad S, Hansen H. The post-treatment 34. Matejusova I, Doig F, Middlemas SJ, Mackay S, Douglas A, Armstrong JD, control programme to ascertain freedom from infection with Gyrodactylus et al. Using quantitative real-time PCR to detect salmonid prey in scats of salaris in Atlantic salmon 2016. In: Annual Report 2016. Oslo: Norwegian grey Halichoerus grypus and harbour Phoca vitulina seals in Scotland - an Veterinary Institute; 2017. experimental and field study. J Appl Ecol. 2008;45:632–40. 14. Mattilsynet (Norwegian Food Safety Authority). Forskrift om kontrollområde 35. Wilcox TM, Carim KJ, McKelvey KS, Young MK, Schwartz MK. The dual for å forebygge, begrense og utrydde lakseparasitten Gyrodactylus salaris challenges of generality and specificity when developing environmental hos akvatiske dyr, Asker, Drammen, Hof, Holmestrand, Hurum, Krødsherad, DNA markers for species and subspecies of Oncorhynchus. PLoS One. 2015; Lier, Modum, Nedre Eiker, Røyken, Sande, Sigdal, Svelvik og Øvre Eiker 10:e0142008. Rusch et al. Parasites & Vectors (2018) 11:333 Page 12 of 12 36. Dobnik D, Spilsberg B, Kosir AB, Holst-Jensen A, Zel J. Multiplex quantification of 12 European union authorized genetically modified maize lines with droplet digital polymerase chain reaction. Anal Chem. 2015;87: 8218–26. 37. Agersnap S, Larsen WB, Knudsen SW, Strand D, Thomsen PF, Hesselsøe M, et al. Monitoring of noble, signal and narrow-clawed crayfish using environmental DNA from freshwater samples. PLoS One. 2017;12:e0179261. 38. Levy MG, Poore MF, Colorni A, Noga EJ, Vandersea MW, Litaker RW. A highly specific PCR assay for detecting the fish ectoparasite Amyloodinium ocellatum. Dis Aquat Organ. 2007;73:219–26. 39. Gomes GB, Hutson KS, Domingos JA, Chung C, Hayward S, Miller TL, Jerry DR. Use of environmental DNA (eDNA) and water quality data to predict protozoan parasites outbreaks in fish farms. Aquaculture. 2017;479:467–73. 40. Agawa Y, Tani K, Yamamoto S, Hirano C, Shirakashi S. Development of a quantitative polymerase chain reaction assay for the detection of skin fluke Neobenedenia girellae larvae from environmental water. Fisheries Sci. 2016; 82:827–33. 41. Hashizume H, Sato M, Sato MO, Ikeda S, Yoonuan T, Sanguankiat S, et al. Application of environmental DNA analysis for the detection of Opisthorchis viverrini DNA in water samples. Acta Trop. 2017;169:1–7. 42. White VC, Morado JF, Crosson LM, Vadopalas B, Friedman CS. Development and validation of a quantitative PCR assay for Ichthyophonus spp. Dis Aquat Organ. 2013;104:69–81. 43. Hartikainen H, Bass D, Briscoe AG, Knipe H, Green AJ, Okamura B. Assessing myxozoan presence and diversity using environmental DNA. Int J Parasitol. 2016;46:781–92. 44. Hallett SL, Ray RA, Hurst CN, Holt RA, Buckles GR, Atkinson SD, Bartholomew JL. Density of the waterborne parasite Ceratomyxa shasta and its biological effects on salmon. Appl Environ Microb. 2012;78:3724–31. 45. Meinilä M, Kuusela J, Zietara MS, Lumme J. Initial steps of speciation by geographic isolation and host switch in salmonid pathogen Gyrodactylus salaris (Monogenea: Gyrodactylidae). Int J Parasitol. 2004;34:515–26. 46. Hansen H, Bakke TA, Bachmann L. DNA taxonomy and barcoding of monogenean parasites: lessons from Gyrodactylus. Trends Parasitol. 2007; 23:363–7. 47. Deiner K, Altermatt F. Transport distance of invertebrate environmental DNA in a natural river. PLoS One. 2014;9:e88786. 48. Balasingham KD, Walter RP, Heath DD. Residual eDNA detection sensitivity assessed by quantitative real-time PCR in a river ecosystem. Mol Ecol Resour. 2017;17:523–32. 49. Darling JA, Mahon AR. From molecules to management: Adopting DNA-based methods for monitoring biological invasions in aquatic environments. Environ Res. 2011;111:978–88. 50. Goldberg CS, Turner CR, Deiner K, Klymus KE, Thomsen PF, Murphy MA, et al. Critical considerations for the application of environmental DNA methods to detect aquatic species. Methods Ecol Evol. 2016;7:1299–307. 51. Doi H, Uchii K, Takahara T, Matsuhashi S, Yamanaka H, Minamoto T. Use of droplet digital PCR for estimation of fish abundance and biomass in environmental DNA surveys. PLoS One. 2015;10:e0122763. 52. Maruyama A, Nakamura K, Yamanaka H, Kondoh M, Minamoto T. The release rate of environmental DNA from juvenile and adult fish. PLoS One. 2015;10:e114639. 53. Tellenbach C, Grunig CR, Sieber TN. Suitability of quantitative real-time PCR to estimate the biomass of fungal root rndophytes. Appl Environ Microb. 2010;76:5764–72. 54. Baker M. Digital PCR hits its stride. Nat Methods. 2012;9:541–4. 55. Doi H, Takahara T, Minamoto T, Matsuhashi S, Uchii K, Yamanaka H. Droplet digital polymerase chain reaction (PCR) outperforms real-time PCR in the detection of environmental DNA from an invasive fish species. Environ Sci Technol. 2015;49:5601–8. 56. McKee AM, Spear SF, Pierson TW. The effect of dilution and the use of a post-extraction nucleic acid purification column on the accuracy, precision, and inhibition of environmental DNA samples. Biol Conserv. 2015;183:70–6. 57. Tsai YL, Olson BH. Rapid method for separation of bacterial-DNA from humic substances in sediments for polymerase chain-reaction. Appl Environ Microb. 1992;58:2292–5. 58. Hoshino T, Inagaki F. Molecular quantification of environmental DNA using microfluidics and digital PCR. Syst Appl Microbiol. 2012;35:390–5. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Parasites & Vectors Springer Journals

Catching the fish with the worm: a case study on eDNA detection of the monogenean parasite Gyrodactylus salaris and two of its hosts, Atlantic salmon (Salmo salar) and rainbow trout (Oncorhynchus mykiss)

Free
12 pages
Loading next page...
 
/lp/springer_journal/catching-the-fish-with-the-worm-a-case-study-on-edna-detection-of-the-d64hbZ18AE
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Parasitology; Entomology; Tropical Medicine; Infectious Diseases; Veterinary Medicine/Veterinary Science; Virology
eISSN
1756-3305
D.O.I.
10.1186/s13071-018-2916-3
Publisher site
See Article on Publisher Site

Abstract

Background: Environmental DNA (eDNA) monitoring is growing increasingly popular in aquatic systems as a valuable complementary method to conventional monitoring. However, such tools have not yet been extensively applied for metazoan fish parasite monitoring. The fish ectoparasite Gyrodactylus salaris, introduced into Norway in 1975, has caused severe damage to Atlantic salmon populations and fisheries. Successful eradication of the parasite has been carried out in several river systems in Norway, and Atlantic salmon remain infected in only seven rivers, including three in the Drammen region. In this particular infection region, a prerequisite for treatment is to establish whether G. salaris is also present on rainbow trout upstream of the salmon migration barrier. Here, we developed and tested eDNA approaches to complement conventional surveillance methods. Methods: Water samples (2 × 5 l) were filtered on-site through glass fibre filters from nine locations in the Drammen watercourse, and DNA was extracted with a CTAB protocol. We developed a qPCR assay for G. salaris targeting the nuclear ribosomal ITS1 region, and we implemented published assays targeting the mitochondrial cytochrome-b and NADH-regions for Atlantic salmon and rainbow trout, respectively. All assays were transferred successfully to droplet digital PCR (ddPCR). Results: All qPCR/ddPCR assays performed well both on tissue samples and on field samples, demonstrating the applicability of eDNA detection for G. salaris, rainbow trout and Atlantic salmon in natural water systems. With ddPCR we eliminated a low cross-amplification of Gyrodactylus derjavinoides observed using qPCR, thus increasing specificity and sensitivity substantially. Duplex ddPCR for G. salaris and Atlantic salmon was successfully implemented and can be used as a method in future surveillance programs. The presence of G. salaris eDNA in the infected River Lierelva was documented, while not elsewhere. Rainbow trout eDNA was only detected at localities where the positives could be attributed to eDNA release from upstream land-based rainbow trout farms. Electrofishing supported the absence of rainbow trout in all of the localities. (Continued on next page) * Correspondence: haakon.hansen@vetinst.no Norwegian Veterinary Institute, P.O. Box 750, Sentrum, NO-0106 Oslo, Norway Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Rusch et al. Parasites & Vectors (2018) 11:333 Page 2 of 12 (Continued from previous page) Conclusions: We provide a reliable field and laboratory protocol for eDNA detection of G. salaris,Atlantic salmon and rainbow trout, that can complement conventional surveillance programs and substantially reduce the sacrifice of live fish. We also show that ddPCR outperforms qPCR with respect to the specific detection of G. salaris. Keywords: Environmental DNA, Multiplex PCR, Droplet digital PCR (ddPCR), Internal transcribed spacer (ITS), Mitochondrial DNA (mtDNA), Invasive species Background including a control area is described in the Norwegian le- Gyrodactylus salaris Malmberg, 1957 (Monogenea) is an gislation [14]. A strategy to implement treatment of this ectoparasite first described on the skin of Atlantic region has not yet been conclusively devised by the salmon Salmo salar (L. 1758), where it attaches itself to Norwegian authorities, as this watercourse in many as- the host with a haptor, a specialized attachment organ pects is more complicated than previously treated systems. consisting of a large disc with 16 peripheral articulated This results from three basic factors. First, rainbow trout in marginal hooks and a single pair of ventrally orientated the system upstream of the current migration barriers for hamuli [1]. This ~500 μm long parasite [2] has also been salmon have a history of infection with G. salaris [8]. Sec- found on other salmonids such as rainbow trout Onco- ondly, Drammenselva contains a much higher fish species rhynchus mykiss (Walbaum, 1792) [3], brown trout diversity than other treated rivers, which mainly contain Salmo trutta (L., 1758) and Arctic charr Salvelinus alpi- salmonids. Thirdly, this river discharges into a large estuary nus (Linnaeus, 1758) [4]. While most species and popu- with surface water containing low salinity (< 2%) where G. lations of fish which act as hosts, including Baltic salaris can survive for longer periods [15]. In order to de- populations of Atlantic salmon, do not experience cide on measures regarding treatment of this water system, serious consequences of a G. salaris infection [1, 5], At- exact knowledge of the status of infections with G. salaris lantic populations of salmon are highly susceptible to G. in the area is a prerequisite. Rainbow trout farms in the salaris resulting in high mortality rates in mainly northern parts of the Drammen watercourse were infected Norwegian populations (see below). Rainbow trout is with G. salaris in the mid-1980s and later there have been less susceptible, and can sustain infections for long pe- both documented [16, 17] and anecdotal reports of riods, often at low intensities making it an important free-living rainbow trout in the system. There is thus a pos- host when considering spreading between fish farms in sibility that free-living rainbow trout are still present in the Europe [6]. system and these might have sustained the introduced G. In 1975, G. salaris was detected in Norway for the first salaris infection from the 1980s. Therefore, a surveillance time [7–9]. The parasite has since caused severe damage to program [18, 19] has been established to detect any pos- several Atlantic salmon populations [1]. Altogether, fish in sible presence of G. salaris on free-living populations of 50 rivers in Norway have been infected by G. salaris and rainbow trout upstream of the anadromous parts of the extensive eradication programs, mostly using pesticides Drammenselva catchment. such as rotenone, have been carried out in several of these Standard surveillance for fish parasites, including the sur- watercourses [10] since 1981 [11]. Over the last 15 years veillance programs for G. salaris in Norway, involves cap- [12], the eradication programs have been highly successful ture and euthanasia of fish, prior to manual examination and to date the parasite is present only in seven rivers [10]. for the presence of parasites. This is both costly and To document the absence of G. salaris in Norwegian river labour-intensive, and results in the sacrifice of a large num- systems and to detect new infections at an early stage, ber of usually infection-free healthy fish. In recent years, large-scale national surveillance programs are carried out capturing, amplifying and detecting species-specific DNA every year [10, 13]. Present surveillance is based on the fragments of several species in water samples has been catching and killing of numerous Atlantic salmon juveniles established as an accurate low-cost alternative or a comple- in rivers and farms, as well as rainbow trout reared in ment to traditional monitoring [20–23]. This approach, farms, for morphological screening for the presence or ab- harnessing so-called environmental DNA (eDNA), makes sence of G. salaris. In 2016 alone, 6981 fish were killed and useof the knowledgethatall organisms shed cells into their examined [10, 13]. surroundings (excretion, mucus layers, abrasions of epithe- One of the remaining regions where G. salaris is still lial tissue, gametes) [24, 25]. For eDNA monitoring of nat- present is the Drammen region (Buskerud and Vestfold ural waters, the eDNA content represents to a large extent County) in southern Norway, consisting of the rivers a snap-shot of the present living species, with a time lag of Drammenselva, Lierelva and Sandeelva (hereafter referred only some weeks after a species has disappeared from the to by their Norwegian names). The infection region system until eDNA is no longer detectable [26]. Results are Rusch et al. Parasites & Vectors (2018) 11:333 Page 3 of 12 delivered relatively fast and efficiently [27], often at lower (Buskerud County), a small river in the Drammen infec- agent-prevalence than through traditional methods [28]. tion region where Atlantic salmonhasbeen infected with To complement conventional surveillance methods for G. salaris since 1987 [1]. Drammenselva drains from the G. salaris, we aimed at developing an eDNA approach Jotunheimen Mountains in the north, down to for targeted detection of the parasite-host combination Drammensfjorden (Buskerud and Vestfold Counties) in water samples. We applied this method in a which connects the watercourse with the Atlantic Ocean case-study, where eDNA detection by means of species (Fig. 1). The infection region in Drammen incorporates specific quantitative PCR (qPCR) and droplet digital three of the remaining seven rivers in Norway where G. PCR (ddPCR) was used as a supplement to standard sur- salaris is still present. These are: Drammenselva, Lierelva veillance methods for G. salaris, Atlantic salmon and (both Buskerud County) and Sandeelva (Vestfold County) rainbow trout in the Drammen infection region, (Fig. 1), in all of which Atlantic salmon is present. Lierelva Norway. and Sandeelva are smaller rivers with catchment sizes of 309.6 and 193.4 km , respectively, while Drammenselva Methods drains from a much larger area (17,110.8 km ). In the Description of the study area northern part of the Drammen watercourse (see Fig. 1), One part of this study was conducted in the northern several rainbow trout producers can be found. Fish in part of the Drammenselva watercourse (Oppland farms in this area were infected by G. salaris in the County) where a presence of wild rainbow trout popula- mid-1980s and there were many reports of escaped fish tions is possible and the status of G. salaris is unknown. from the farms [14]. However, the fish populations in the The other part of the study was conducted in Lierelva farms were eradicated and all these farms were declared Fig. 1 Map of the Drammen watercourse region with all sampling locations and its location within Norway. Green points represent localities sampled. The thick blue line represents the Drammen watersystem, the thin blue lines represent the main rivers, the red lines indicate rivers where G. salaris is present and the black lines outline the Drammenselva drainage basin. The numbers refer to the sampling sites in Table 1. Pie charts: blue colour indicates detection of Atlantic salmon, red indicates detection of Gyrodactylus salaris and yellow indicates detection of rainbow trout. Rivers flow north to south Rusch et al. Parasites & Vectors (2018) 11:333 Page 4 of 12 free from G. salaris in 1987 [29]. In 1986, G. salaris was according to local conditions (stream size, depth, water also diagnosed from farmed rainbow trout and salmon in flow). Electrofishing was also conducted in Lierelva to the Lake Tyrifjorden which is a part of the Drammen collect salmon juveniles for estimation of the infection watercourse [8, 30]. The fish populations in these farms prevalence and intensity of G. salaris in the same locality were also eradicated, but a short time later, the parasite as water samples were taken. Fish captured for further was detected on salmon juveniles from Drammenselva examination were euthanised following the strict codes and Lierelva [30]. of practice in force in Europe, preserved intact in 96% ethanol and later examined for the presence of Gyrodac- Sample locations tylus spp. using a stereo microscope (Leica MZ 7.5, Leica The sampling sites included eight localities in the northern microsystems, St. Gallen, Switzerland). part of the Drammenselva watercourse, (Fig. 1,Table 1). These sampling sites were chosen as part of a monitoring Water filtering for eDNA sampling program [19] and with the intention of both declaring this At each sampling location, duplicate water samples of 5 region free from G. salaris and mapping the presence of L (2 × 5 l) were collected and filtered on site on to glass free-living rainbow trout. One of these eight sampling sites fibre filters (47 mm AP25 Millipore, 2 μm pore size, was a fish pond at a local trout farm that served as a Millipore, Billerica, USA) using a portable peristaltic rainbow trout positive field control sample. The ninth pump (Masterflex E/S portable sampler, Masterflex, sample was chosen as a positive field control sample for Gelsenkirchen, Germany), tygon tubing (Masterflex) and only G. salaris and Atlantic salmon and collected from a an in-line filter holder (Millipore) according to Strand et stretch in Lierelva (Fig. 1), a river with a confirmed al. [31]. At Lierelva, four samples were taken instead of presence of Atlantic salmon infected with G. salaris. two as this river was intended as a positive field control Within the area where rainbow trout farms can be for G. salaris and Atlantic salmon. Filters were placed in found, the locations of sample nos. 4 and 5 were chosen separate 15 ml Falcon tubes containing cetyl trimethyl based on information obtained from the local authorities ammonium bromide (CTAB) buffer and stored on ice prior to the field work. These samples were taken in directly after filtration. Upon arrival at the laboratory the streams flowing into the main watercourse in order to samples were stored at -20 °C until further analysis. As a avoid positive detections due to outlet water from farms safety precaution and part of the filtering protocol, the situated upstream in the main watercourse. For an indi- entire equipment was disinfected with a 10% bleach so- cation of the sensitivity of the rainbow trout eDNA assay lution after use at each location. Thus, any residual for detection in the field, three samples (nos. 6, 7 and 8) eDNA was broken down and contamination was pre- were taken from the main watercourse at different dis- vented. Before further sampling, the tubes were rinsed tances from the rainbow trout farms. Samples 1 and 2 with sodium thiosulphate to neutralise the bleach solu- were taken upstream of the area containing rainbow tion, and then flushed with ambient river water directly trout farms. before sampling. Electrofishing and Gyrodactylus counts DNA extraction Electrofishing was carried out in rivers and tributaries in DNA was extracted from the filters according to a CTAB the Drammen watercourse to reveal the possible pres- protocol described in Strand et al. [31], with the excep- ence of rainbow trout, using this standard surveillance tion that the CTAB buffer contained no added 1% method. The area examined was chosen on site 2-mercapto-ethanol. During extraction each filter was Table 1 List of sampling sites including location, sampling date and amount of water filtered Site no. Site name Water filtered (l) Coordinates Date 1 Storåne at Ala camping 5 (×2) 61.1473N, 8.7121E 26.06.2017 2 Storåne at Tørpegårdsvegen/bru 5 (×2) 61.1522N, 8.7250E 26.06.2017 3 Trout farm 5 (×2) 61.0379N, 9.0466E 14.11.2016 4 Leireelvi at Leira/Garlivegen 5 (×2) 60.9742N, 9.2936E 26.06.2017 5 Leireelvi at Leira camping 5 (×2) 60.9680N, 9.2884E 26.06.2017 6 Lake Strondafjorden at Faslefoss 5 (×2) 60.9671N, 9.2889E 26.06.2017 7 River Begna at Bagn 5 (×2) 60.8198N, 9.5612E 26.06.2017 8 River Begna at Nes 5 (×2) 60.5628N, 9.9929E 26.06.2017 9 Lierelva at Sjåstad 5 (×4) 59.8580N, 10.2213E 31.08.2017 Rusch et al. Parasites & Vectors (2018) 11:333 Page 5 of 12 split into two subsamples (A and B) due to volume matching them against the database of the National restrictions imposed by centrifuge size and extracted Centre for Biotechnology Information (NCBI, http:// separately. An environmental control and a blank extrac- www.ncbi.nlm.nih.gov/) nucleotide database using the tion control were included as a precaution to detect any Basic Local Alignment Search Tool (BLASTn). The aim of possible contamination during DNA extraction. The the new qPCR assay was to attempt to obtain the best pos- blank extraction control consisted of a Falcon tube con- sible sensitivity and specificity for eDNA applications. taining the CTAB buffer but no filter, which was then Similar to the assay Collins et al. [32] designed, the newly processed in the same way as all other tubes containing designed assay is not able to distinguish between G. salaris buffer and filters. The environmental control used in this and G. thymalli Žitňan, 1960 as these two species have in- study consisted of an Eppendorf tube containing 200 μl distinguishable ITS sequences [33]. PCR-grade water. This tube remained open in the fume The assays used for eDNA-detection of Atlantic hood throughout the entire process of extraction. salmon and rainbow trout (Table 2) follow Matejusova et al. [34] and Wilcox et al. [35], respectively. These were successfully tested on DNA extracts from tissue of PCR-based assays for eDNA detection of G. salaris, Atlantic salmon and rainbow trout before use in the rainbow trout and Atlantic salmon current study (data not shown). The ddPCR assay for G. A quantitative PCR assay (qPCR) using species-specific salaris, rainbow trout and Atlantic salmon applied the primers and a minor groove binder (MGB) probe targeting same primers and probes as the qPCR. the G. salaris internal transcribed spacer region 1 (ITS1) was developed. The ITS1 sequence published as GenBank accession no. DQ898302 was used as template and the spe- Evaluation of qPCR and ddPCR assay specificity cificity of the designed primers and probe was checked The specificity of the assay was tested on DNA extracts of against closely related species and other species that might G. salaris collected from three different locations in be present in Norwegian watercourses: G. salmonis Yin & Norway in addition to DNA extracts from the following Sproston, 1948 (GQ368233), G. truttae Gläser, 1974 other species present in the collection at the NVI: G. thy- (AJ132260), G. lucii Kulakovskaya, 1952 (EU304811), G. malli, G. salmonis, G. arcuatus, G. lucii and G. derjavi- arcuatus Bychowsky, 1933 (JN703797) and G. derjavi- noides. Species identification of these samples had been noides Malmberg, Collins, Cunningham & Jalali, 2007 done previously by sequencing of ITS (results not shown). (EU304810). Multiple sequences were aligned using AlignX We also ran the same samples with the previously pub- (Vector NTI Advance 11.5, Invitrogen, Carlsbad, USA). lished assay [32] to compare the specificity and sensitivity The design of primers and probe was performed manually, of the assays. ddPCR applies the same primers and probes targeting ITS1 sequence regions displaying the highest se- as qPCR and the specificity was tested on G. derjavinoides quence diversity between G. salaris and the species listed due to the low level of cross amplification shown in a pre- above. The final primer and probe sequences (Table 2) viously published assay [32]. The ddPCR assay was also partly overlap with those previously published for this tested on isolates of G. salaris obtained from fish from parasite [32] and their specificity was confirmed through Lierelva to determine optimal annealing temperature. Table 2 Primers and probes for Gyrodactylus salaris, rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar) used in the present study. The probes used are TaqMan MGB probes with either Fam or Hex reporter dyes Target species/gene Name Primer/probe Sequence (5'-3') Reference G. salaris/ITS G.sal208F Forward GGTGGTGGCGCACCTATTC Present study G.sal149R Reverse ACGATCGTCACTCGGAATCGAT Present study G.sal188P Probe (FAM)CAAGCAGAACTGGTTAAT(MGBNFQ) Present study G. salaris/ITS F Forward CGATCGTCACTCGGAATCG Collins et al. [32] R Reverse GGTGGCGCACCTATTCTACA Collins et al.[32] Gsal2 Probe (FAM)TCTTATTAACCAGTTCTGC(MGBNFQ) Collins et al. [32] O. mykiss/cytb RBTF Forward AGTCTCTCCCTGTATATCGTC Wilcox et al. [35] RBTR Reverse GATTTAGTTCATGAAGTTGCGTGAGTA Wilcox et al. [35] RBTP Probe (FAM)CCAACAACTCTTTAACCATC(MGBNFQ) Wilcox et al. [35] S. salar/cytb Salmonid Cyt B FOR Forward CGGAGCATCTTTCTTCTTTATCTGT Matejusova et al. [34] S. salar REV Reverse ACTCCGATATTTCAGGTTTCTTTATATAGA Matejusova et al. [34] S. salar Cyt B Probe Probe (HEX)CCAACAACTCTTTAACCATC-(MGBNFQ) Matejusova et al. [34] Rusch et al. Parasites & Vectors (2018) 11:333 Page 6 of 12 qPCR and ddPCR protocols for G. salaris eDNA detection the same primers and probes as for the singleplex reac- All qPCR analyses were carried out on an Mx3005P tions. This duplex method was set up by running the fol- qPCR system (Stratagene, San Diego, USA). Droplet lowing 22 μl reactions for each eDNA extract in digital PCR was performed on a QX200 AutoDG Drop- duplicates: 11 μl ddPCR Supermix for probes - no dUTP let Digital PCR System (Bio-Rad, Hercules, USA). (Bio-Rad), 0.99 μlof20 μM of Salmonid Cyt B FOR and For qPCR detection of G. salaris, three qPCR repli- S. salar REV primers, 0.55 μlof10 μM S. salar Cyt B cates were run for each eDNA extract in the following Probe, 0.99 μlof 20 μM of G.sal208F and G.sal149R 25 μl reactions: 1.25 μl of PCR-grade water, 12.5 μlof primers, 0.275 μlof 20 μM G.sal188P probe, 0.215 μl ExTaq mastermix (Takara Biotechnology, Dalian, China), PCR-grade water and 1 μl of restriction-enzyme mix 1.5 μl of each 10 μM primer (forward and reverse), 0.75 consisting of 0.2 μl HindIII, 0.1 μl buffer (10×), 0.7 μl μlof10 μM probe, 0.5 μl of Rox II reference dye and 5 PCR-grade water and 5 μl of DNA sample. The optimal μl of eDNA template. The qPCR cycling conditions were primer-probe concentration for both assays had been as follows: an initial denaturation at 95 °C for 15 min; 45 determined to be 900:250. The same cycling conditions cycles of denaturation at 94 °C for 30 s, annealing at 54 ° were used as in the G. salaris singleplex reaction. C for 45 s, and extension at 72 °C for 1 min; followed by For qPCR detection of O. mykiss, three qPCR repli- a final elongation phase at 72 °C for 10 min. cates were run for each eDNA extract in the following The following 22 μl reactions were run for each eDNA 12 μl reactions: 2.35 μl of PCR-grade water, 6.25 μlof extract on ddPCR: 11 μl ddPCR Supermix for probes - ExTaq mastermix (Takara), 0.3 μlof10 μM RBTF no dUTP (Bio-Rad), 1.98 μl of each 10 μM primer, 0.55 forward primer and 0.6 μlof 10 μM RBTR reverse pri- μlof10 μM probe, 0.49 μl PCR-grade water and 1 μlof mer, 0.25 μlof10 μM RBTP probe, 0.25 μl of Rox II restriction-enzyme mix consisting of 0.2 μl HindIII, 0.1 reference dye and 2 μl of DNA template. The qPCR μl buffer (10×) and 0.7 μl PCR-grade water and 5 μlof (Stratagene) cycling conditions were as follows: an initial DNA sample. The optimal primer-probe concentration denaturation at 95 °C for 1 min; 45 cycles of denatur- was determined to be 900:250 and the optimal annealing ationat94°Cfor 30 s, annealingat54°Cfor 45 s temperature of 58 °C was confirmed through amplifica- and extension at 72 °C for 1 min; followed by a final tion tests along a temperature gradient. Here, we used elongation phase at 72 °C for 10 min. We used a the HindIII restriction enzyme to fragment the repetitive cut-off at Cq 41 for the rainbow trout-assay, similar multi-copy ITS regions within the nuclear ribosomal to the suggestion for eDNA qPCR detection cut-off in DNA in order to ensure that the targeted DNA copies Agersnap et al. [37]. were distributed among different droplets for accurate For the singleplex ddPCR detection of rainbow trout, quantification. the following 22 μl reactions for each eDNA extract To allow for sufficient time for the restriction enzymes were run in duplicates: 11 μl ddPCR Supermix for to digest, the plate was sealed using Microseal ‘B’ plate Probes - no dUTP (Bio-Rad), 0.99 μl of RBTF 10 μM sealing film (Bio-Rad), wrapped in tin foil and left on the forward primer, 1.98 μlof10 μM RBTR reverse primer, bench for 20 min. Droplet generation in the QX200 0.55 μlof 10 μM RBTP probe, 2.48 μl PCR-grade water AutoDG Droplet Digital PCR System (Bio-Rad) creates and 5 μl of DNA template. The optimal primer-probe an emulsion with 20 μl of the 22 μl originally pipetted concentration for both assays had been determined to into each well, resulting in a 10% loss of template and be 450:900:250 for forward primer, reverse primer and mastermix. After generation of the droplets, the new probe, respectively, which follows the suggestions in plate was immediately transferred to a TM100 thermo- Wilcox et al. [35]. The same cycling conditions were cycler (Bio-Rad) and the QX200 Droplet Digital PCR used as in all other ddPCR reactions. system (Bio-Rad) with the following cycling conditions: An initial denaturation at 95 °C for 10 min; 45 cycles of Calculation of eDNA copies denaturation at 94 °C for 30 s, annealing at 58 °C for 60 The number of eDNA copies (for each target species) s; followed by a final elongation phase at 58 °C for 10 per litre of water for each sample is calculated min. The threshold for a positive sample was set at three according to the following formula, also used by positive droplets per well according to Dobnik et al. Agersnap et al. [37]: [36]. To ensure the validity of each run, positive and blank PCR controls containing G. salaris DNA and rdd distilled water, respectively, were run on each plate for C ¼ both qPCR and ddPCR. To be able to detect G. salaris and Atlantic salmon simultaneously in future surveillance programmes in where C is the number of target-eDNA copies per litre Norwegian rivers, we also tested a duplex method using of filtered water, C is the ddPCR calculation of eDNA rdd Rusch et al. Parasites & Vectors (2018) 11:333 Page 7 of 12 copy numbers per reaction volume (20 μl), adjusted locations 3, 4, 5, 6 and 7 (see Table 3). One of the five for a 10% loss during droplet generation, V is the positive sampling sites (no. 6) was at the outlet of the total elution volume after extraction, V is the volume lake into which all the rainbow trout farms drain, while of eluted extract used in the ddPCR reaction, V is another (no. 7) was found in the main river 25 km the volume of filtered water. The copy numbers of downstream of the outlet. According to new information both subsamples (A and B) were added together, thus from local authorities we received upon enquiry after revealing the number of eDNA copies per litre of any our analyses detected rainbow trout DNA in samples 4 given sample. Calculation of eDNA copy numbers per and 5, these locations were indeed also situated roughly reaction volume was performed by the QuantaSoft 400 and 1200 m, respectively, downstream of a trout software (v.1.7.4, Bio-Rad) and was estimated using farm (see Table 3). None of the field samples in the the ratio between positive and negative droplets northern part of the Drammenselva watercourse yielded within a sample, using Poisson statistics. a positive result when tested against G. salaris, neither did the rainbow trout positive control at the trout farm. Results All extraction blank controls and environmental blank qPCR assay optimisation and specificity tests controls were negative, both in qPCR and ddPCR. The current assay proved slightly more sensitive (by ~0.5 Cq) towards G. salaris than the assay in Collins Conventional monitoring methods 10-1/slope et al. [31]. The PCR efficiency ([E = ]-1)× At location 1, electrofishing of an area of roughly 300 100 calculated from triplicates of non-diluted and m yielded seven juvenile brown trout. Two juvenile three 10-fold dilutions of a DNA extract originating brown trout were caught at location 2 after electrofish- from a single parasite, was 100 % (Cq = 20.5 to 30.6, ing an area of c.200 m . At location 3, electrofishing was slope = 3.312) (not shown). The qPCR assay for G. carried out in selected pot-holes along a stretch of 150 salaris yielded negative qPCR results for all other m. A high density of brown trout with sizes ranging species except G. salaris (and G. thymalli as previ- from juveniles up to 500 g adults was observed. At the ously explained), except for a low level of fourth location, electrofishing was carried out along a cross-reaction towards the tested specimen of G. der- stretch of 200 m. Several minnows Phoxinus phoxinus javinoides (Cq = 35.6). (L., 1758) were observed and many brown trout (juve- niles to 300 g) were captured in the stream while elec- Optimisation of ddPCR assay and specificity tests trofishing. No electrofishing was carried out at locations Both the qPCR assay (primers and probes) for G. salaris 5, 6 and 7 as none of these locations were suitable for developed in this study and the assays for rainbow trout electrofishing. A total of 21 Atlantic salmon with a and Atlantic salmon [32, 34] were transferable to the length of 9.6 cm (± SD 3.6 cm) were caught in Lierelva. ddPCR platform without further optimisation, using an The parasite prevalence and intensity on these fish was annealing temperature of 58 °C. Unlike the qPCR assay determined to be 85.7% and 83 parasites (± SD 63), however, the ddPCR assay showed no signs of cross respectively. Throughout the entire electrofishing, no amplification of G. derjavinoides. rainbow trout were caught. eDNA monitoring of G. salaris, Atlantic salmon and Discussion rainbow trout In the present study, eDNA monitoring is used for the The positive control field samples for G. salaris taken first time to detect the monogenean parasite G. salaris from Lierelva all yielded positive results in qPCR with along with two of its hosts, Atlantic salmon and rainbow Cq-values ranging from 24.76 to 35.86, and in ddPCR trout. Detections were successfully obtained both in all with eDNA copies/l ranging from 371,440 to 560, re- singleplex reactions (qPCR and ddPCR) and in a duplex spectively. For Atlantic salmon, the eDNA copy numbers reaction (ddPCR) targeting both G. salaris and Atlantic ranged from 10,160 (sample 9/2) to 7520 (sample 9/4) salmon. The prevalence in susceptible Atlantic salmon (Table 3) at an average of 8948 copies (± SD = 945). populations most often reaches 100 % [11]. In general, The two positive control field samples for rainbow the infection grows exponentially on non-responding trout obtained at the trout farm in 2016 tested positive hosts and may reach several thousand individuals per for rainbow trout (Cq 17.48 and Cq 17.50; 8,800,000 fish [5]. In our study, the G. salaris infected Atlantic eDNA copies/l and 7,848,000 eDNA copies/l, respect- salmon individuals caught in Lierelva were only moder- ively) (see Table 3). Of the other 18 water samples that ately infected (prevalence of 85.7%, mean parasite abun- were collected at the eight sampling points in June and dance of 83 parasites). Here G. salaris eDNA was August 2017, five were positive for rainbow trout. Posi- detected in amounts ranging from 500 to > 350,000 tive samples for rainbow trout were obtained from copies per litre of water in the same river stretch. These Rusch et al. Parasites & Vectors (2018) 11:333 Page 8 of 12 Table 3 Overview of results from qPCR and ddPCR analyses for Gyrodactylus salaris (ITS), Oncorhynchus mykiss (CytB) and Salmo salar (CytB) at each sampling site. List of sampling sites including amount of water filtered, number of samples per site (each sample constitutes of one filter), the Cq value (from qPCR) and number of eDNA copies per litre (ddPCR) from all filters taken at each point, respectively. eDNA copies per litre are abbreviated as eDNA/l. No detection is indicated with a minus (-) for qPCR and a zero for ddPCR and those samples where analysis was not applicable are indicated with NT Site Site name Sample Volume Gyrodactylus salaris Oncorhynchus mykiss Salmo salar no. (l) a b a a b, a qPCR ddPCR qPCR ddPCR qPCR ddPCR 1 Storåne at Ala camping 1 1 - - - 0 - 0 21 - - - 0 - 0 2 Storåne at Tørpegårdsvegen/bru 1 1 - - - 0 - 0 21 - - - 0 - 0 3 Trout farm 1 1 - - 17.48 7,848,000 - 0 2 1 - - 17.50 8,800,000 - 0 4 Leireelvi at Leira/Garlivegen 1 1 - - 29.62 1624 - 0 2 1 - - 29.09 3816 - 0 5 Leireelvi at Leira camping 1 1 - - 30.05 2240 - 0 2 1 - - 30.02 2124 - 0 6 Lake Strondafjorden at Faslefoss 1 1 - - 32.3 560 - 0 2 1 - - 31.68 576 - 0 7 River Begna at Bagn 1 1 - - > cut-off 0- 0 2 1 - - 36.91 22 - 0 8 River Begna at Nes 1 1 - - - 0 - 0 21 - - - 0 - 0 9 River Lierelva at Sjåstad 1 1 34.52 560 - NT NT 9200 2 1 33.56 840 - NT NT 10,160 3 1 33.94 864 - NT NT 7520 4 1 24.89 371,440 - NT NT 8912 Run as singleplex Run as duplex Cut-off value was set at Cq 41 results strongly indicate that eDNA analysis of samples simultaneous detection of parasite and host. Using the obtained by water filtering can indeed be used for moni- protocol for filtration, DNA-extraction and the analysis toring the occurrence of G. salaris in freshwater ecosys- we describe here, it is not only possible to detect the tems containing natural Atlantic salmon populations. parasite G. salaris but also two of its hosts within on Environmental DNA-detection is a promising tool that single sample. With the use of other assays, the presence can be used to supplement or even replace classical of virtually any aquatic host-pathogen complex can be surveillance where it produces fast and robust results. detected and monitored, provided that the filter size is This is reflected in the ever growing number of assays appropriate to capture eDNA from the target organism. being developed to monitor parasites which infect fish. The aim of the G. salaris qPCR assay designed in the These include both ectoparasites like Amyloodinium present study was to achieve an optimal combination of ocellatum Brown, 1931 [38], Chilodonella hexasticha both specificity and sensitivity, and the assay was chosen Kiernik, 1909 [39]or Neobenedenia girellae Hargis, 1955 over the one previously published by Collins et al. [32] [40] and endoparasites such as Opisthorchis viverrini due to its slightly higher sensitivity. Both the qPCR assay Poirier, 1886 [41], Ichthyophonus spp. [42] and myxozo- presented in this paper and the qPCR assay designed by ans [43, 44]. Unlike traditional monitoring, there is no Collins et al. [32] display a low-level amplification of need to kill large numbers of fish or to carry out Gyrodactylus derjavinoides. However, this issue was not time-consuming manual examinations. Thus, the eDNA observed when applying the newly designed primers and monitoring method has far-reaching potential as it re- probe in ddPCR. Any assay for Gyrodactylus salaris duces the time and cost of sampling and improves fish targeting the ITS1 region will yield positive results for G. welfare. A further advantage of this method is the thymalli since these two species have nearly identical Rusch et al. Parasites & Vectors (2018) 11:333 Page 9 of 12 sequences [33] and it is therefore impossible to differen- 1000-fold higher amount of eDNA in sample 9/4. This tiate between them in this way. This does not affect the calculation is also reflected in the ddPCR results where monitoring of G. salaris in systems uninhabited by gray- an increase from 560 copies/l to 371,440 copies/l was ling, the host for G. thymalli. In systems where grayling observed. This assumption is substantiated by the fact occur, negative samples would still indicate the absence that Gyrodactylids are reported to consist of roughly of the parasite. A positive detection would certainly 1000 cells [1]. The possibility that one sometimes might require additional examination and attention. Here, one catch a whole parasite specimen in the filter does not option would be to design assays targeting the more vari- pose a problem for a simple proof of presence detection, able mitochondrial cytochrome oxidase gene (see, e.g. but in fact increases the certainty of the results. How- Meinilä et al. [45], Hansen et al. [46]). ever, while some studies have demonstrated a correlation In the present eDNA study, as well as for most other between biomass and eDNA concentration [51], quanti- applications, the low level of cross-reaction against G. fication of parasites and establishing an agent-level derjavinoides when using qPCR poses no problem. If a would, in this case, result in an overestimation of para- population of fish were infected with a high number of site numbers. The use of a pre-filter such as fitting a G. derjavinoides and a low number of G. salaris, analysis plankton net in front of the filter with a mesh size small with qPCR could yield ambiguous results. We therefore enough to prevent an entire specimen to pass on to the recommend the use of ddPCR analysis since this method filter may solve this problem of overestimation. In com- bypasses the problem of cross-amplification. Alterna- parison to the results for G. salaris, the copy number for tively, sampling by electrofishing followed by manual Atlantic salmon eDNA was fairly constant in all four examination and standard species identification could be samples at an average of 8948 copies (± SD = 945) as carried out in this particular case. displayed in Fig. 2. This indicates a constant emission We detected rainbow trout eDNA at four locations in rate of eDNA into the water by Atlantic salmon which the northern part of the Drammen watercourse in has also been observed in other studies [52]. addition to the sample taken at the trout farm (sample no. 3). We observed an apparent decline in eDNA Comparison of qPCR/ddPCR monitoring concentration with increasing distance from the source Quantitative real-time PCR (qPCR) offers the possibility (area with trout farms, sample nos. 6 and 7). This corre- to measure the rate of generation of the amplified product sponds with data from studies that examine the dilution after each cycle, thus making it possible to calculate the effects of eDNA in river ecosystems [47, 48]. However, amount of copies in the original sample. Previous studies the number and the distribution of sampling points in have demonstrated that quantification of biomass and cal- this study were not comprehensive enough to examine a culation of population size through using qPCR is possible gradient thoroughly. Extensive electrofishing at each [22, 53]. ddPCR, which now allows the user to operate on sampling point produced no evidence for the presence a nanolitre rather than on a microlitre scale, enables even of rainbow trout in the streams. We therefore attribute more precise detection and absolute quantification of tar- all positive samples to eDNA discharge/emission from get molecules while simultaneously removing the need for trout farms and assume the areas and streams of the standard curves [51, 54]. Our results demonstrate this pre- northern part of the Drammenselva watercourse that cision by detecting both rainbow trout and G. salaris at were tested to be free from wild populations of rainbow very low copy levels with 22 eDNA copies/l and 560 trout. The occurrence of these positive samples reveals eDNA copies/l, respectively. Furthermore, this technology one of the pitfalls of the eDNA methodology, as it sim- has been proven to perform better on inhibition prone ply points out the presence of eDNA from the targeted samples than the predecessor qPCR [55]. This is a particu- organism without verifying the actual presence of the or- lar advantage when analysing environmental samples ganism within the examined body of water [20, 49, 50]. which often tend to include PCR inhibitors [56–58]. Our It does, however, also highlight the sensitivity of this study also shows that ddPCR seems to surpass qPCR re- method. garding specificity, as there was no cross-amplification of One of the four filter samples taken at Lierelva, the G. derjavinoides in the G. salaris assay although the same river with a known presence of G. salaris, displayed a primer-probe combinations were used. We speculate that significantly higher signal than the other three filters, this is due to the lower copy numbers of both target and even though the very same location was sampled. These non-target DNA per reaction (droplets) in the ddPCR results were observed in qPCR, and both the singleplex system. Ideally, one droplet contains only one copy of the and multiplex ddPCR reactions. We presume that this is target DNA and only a few non-target copies, thus redu- due to one or more whole specimens of G. salaris being cing the possibility of unspecific amplification. picked up on this particular filter. The signal difference For a more precise monitoring of G. salaris and its in qPCR is roughly ten cycles which would suggest a hosts, further research and development is needed in Rusch et al. Parasites & Vectors (2018) 11:333 Page 10 of 12 Fig. 2 Visual output from the duplex ddPCR for G. salaris in Channel 1 (blue) and Atlantic salmon in Channel 2 (green) on the samples taken at Lierelva. Wells containing samples 9/1B and 9/3B are not displayed and were excluded due to insufficient droplet generation. Each blue and green point represents a positive amplification of respective DNA template. The horizontal purple line represents the threshold and the black points represent negative droplets. The eDNA copy number for G. salaris is markedly higher in two of the wells containing samples 9/4A and 9/4B. However, the copy number of Atlantic salmon eDNA remains relatively stable in all four samples order to improve the specificity of the G. salaris assay qPCR when screening samples for G.salaris.Further to distinguish from G. thymalli, as well as to determine studies are needed to determine the limit of detection when it is no longer possible to obtain a positive eDNA regarding eDNA and to compare the eDNA signal result (limit of detection) when the parasite load per against fish parasite load in experimental and natural fish drops. settings. Abbrevations Conclusions BLASTn: Basic Local Alignment Search Tool; Cq-Value: quantification cycle; We have successfully designed and implemented a CTAB: cetyl trimethyl ammonium bromide; ddPCR: droplet digital PCR; eDNA: environmental DNA; ITS1: internal transcribed spacer 1; MGB: minor method for eDNA detection of an aquatic groove binder; mtDNA: mitochondrial DNA; NADH: nicotinamide adenine host-parasite system, specifically G. salaris and its dinucleotide dehydrogenase; qPCR: quantitative PCR two hosts Atlantic salmon and rainbow trout. Thus, Acknowledgements we demonstrate for the first time that eDNA moni- We thank Saima Nasrin Mohammad and Mari Darrud, Norwegian Veterinary toring canbeusedfor thedetection of G. salaris and Institute (NVI), for technical assistance with molecular work and field work, its host Atlantic salmon in natural freshwater systems respectively. with a moderately infected salmon population. Fur- Funding thermore,wehavedeterminedthe assaywedesigned This study is part of the “eDNAqua-Fresh” PhD project, funded by the NVI. to be species-specific and demonstrated the usefulness No external grants were received for this work. of eDNA methodology when examining a river system Availability of data and materials for the presence of G. salaris. Within the paper we All data generated or analysed during this study are included in this present a protocol, both field and laboratory, on how published article. to conduct eDNA monitoring of G. salaris and Atlan- Authors’ contributions tic salmon successfully, using a duplex ddPCR. We HH, JR, SH and TV planned the study. SH was in charge of the field work and show that ddPCR appears to be a better tool than performed the electrofishing and JR and HH participated in the field work. JR Rusch et al. Parasites & Vectors (2018) 11:333 Page 11 of 12 carried out water filtering, qPCR analyses, ddPCR analyses and optimisation kommuner, Akershus, Buskerud og Vestfold. 2016. https://lovdata.no/ of ddPCR assays. DS took part in the design and optimisation of ddPCR dokument/FV/forskrift/2016-07-07-919. Accessed 30 Apr 2018. assays. TM and HH designed the qPCR assay for G. salaris and TM optimized 15. Soleng A, Bakke TA. Salinity tolerance of Gyrodactylus salaris the qPCR assay. JR and HH wrote the first draft of the manuscript and all (Platyhelminthes, Monogenea): laboratory studies. Can J Fish Aquat Sci. authors contributed in the writing of the manuscript. All authors read and 1997;54:1837–45. approved the final manuscript. 16. Brabrand Å. Fiskeribiologiske undersøkelser i Slidrefjorden. Oppland Fylke: Vurdering av tilslag på settefisk: Laboratorium for ferskvannsøkologi og Ethics approval and consent to participate innlandsfiske (LFI), Zoologisk Museum, Universitetet i Oslo; 1988. p. 44. No approval from Institutional Animal Care and Use Committee (IACUC) or 17. Thomassen G, Norum I. Fiskeundersøkelse i Strondafjorden. Fylkesmannen i ethics committee was necessary. No experiments that involved fish were Oppland, miljøvernavdelingen, Rapport nummer 17; 2012. p. 18. performed. All fish were euthanised following the strict codes of practice in 18. Hytterød S, Mo TA, Hansen H, Tavornpanich S. The post-treatment force in Europe. surveillance programme to ascertain freedom from infection with Gyrodactylus salaris in Atlantic salmon, Annual report 2013. Norwegian Competing interests Veterinary Institute: Oslo; 2014. The authors declare that they have no competing interests. 19. Hytterød S, Rusch JC, Darrud M, Nasrin Mohammad S, Hansen H. Mapping the occurrence of Gyrodactylus salaris upstream of the natural anadromous region of the Drammenselva catchment. In: Annual Report 2016. Norwegian Publisher’sNote Veterinary Institute: Oslo; 2017. Springer Nature remains neutral with regard to jurisdictional claims in 20. Evans NT, Shirey PD, Wieringa JG, Mahon AR, Lamberti GA. Comparative published maps and institutional affiliations. cost and effort of fish distribution detection via environmental DNA analysis and electrofishing. Fisheries. 2017;42:90–9. Author details 21. Jerde CL, Chadderton WL, Mahon AR, Renshaw MA, Corush J, Budny ML, Norwegian Veterinary Institute, P.O. Box 750, Sentrum, NO-0106 Oslo, et al. Detection of Asian carp DNA as part of a Great Lakes basin-wide Norway. Department of Biosciences, University of Oslo, P.O. Box 1066, surveillance program. Can J Fish Aquat Sci. 2013;70:522–6. Blindern, NO-0316 Oslo, Norway. Faculty of Veterinary Medicine, Norwegian 22. Rees HC, Maddison BC, Middleditch DJ, Patmore JRM, Gough KC. The University of Life Sciences, P.O. Box 8146, Dep, NO-0033 Oslo, Norway. detection of aquatic animal species using environmental DNA - a review of eDNA as a survey tool in ecology. J Appl Ecol. 2014;51:1450–9. Received: 31 January 2018 Accepted: 25 May 2018 23. Smart AS, Tingley R, Weeks AR, van Rooyen AR, McCarthy MA. Environmental DNA sampling is more sensitive than a traditional survey technique for detecting an aquatic invader. Ecol Appl. 2015;25:1944–52. References 24. Mächler E, Deiner K, Spahn F, Altermatt F. Fishing in the water: effect of 1. Bakke TA, Cable J, Harris PD. The biology of gyrodactylid monogeneans: The sampled water volume on environmental DNA-based detection of “Russian-doll killers”. Adv Parasitol. 2007;64:161–376. macroinvertebrates. Environ Sci Technol. 2016;50:305–12. 2. Malmberg G. Om förekomsten av Gyrodactylus på svenska fiskar. Skrifter 25. Deiner K, Fronhofer EA, Mächler E, Walser JC, Altermatt F. Environmental utgivna av Södra Sveriges Fiskeriförening, Årsskrift. 1957;1956:19–76. DNA reveals that rivers are conveyer belts of biodiversity information. Nat 3. Malmberg G, Malmberg M. Species of Gyrodactylus (Platyhelminthes, Commun. 2016;7:12544. Monogenea) on salmonids in Sweden. Fish Res. 1993;17:59–68. 26. Dejean T, Valentini A, Duparc A, Pellier-Cuit S, Pompanon F, Taberlet P, 4. Bakke TA, Harris PD, Jansen PA, Hansen LP. Host specificity and dispersal Miaud C. Persistence of environmental DNA in freshwater ecosystems. PLoS strategy in gyrodactylid monogeneans, with particular reference to One. 2011;6:e23398. Gyrodactylus salaris (Platyhelminthes, Monogenea). Dis Aquat Organ. 1992; 27. Gustavson MS, Collins PC, Finarelli JA, Egan D, Conchuir RO, Wightman GD, 13:63–74. et al. An eDNA assay for Irish Petromyzon marinus and Salmo trutta and field 5. Bakke TA, Harris PD, Cable J. Host specificity dynamics: observations on validation in running water. J Fish Biol. 2015;87:1254–62. gyrodactylid monogeneans. Int J Parasitol. 2002;32:281–308. 28. Wittwer C, Stoll S, Strand D, Vrålstad T, Nowak C, Thines M. eDNA-based 6. Hansen H, Cojocaru CD, Mo TA. Infections with Gyrodactylus spp. crayfish plague monitoring is superior to conventional trap-based (Monogenea) in Romanian fish farms: Gyrodactylus salaris Malmberg, 1957 assessments in year-round detection probability. Hydrobiologia. 2018;807: extends its range. Parasit Vectors. 2016;9:444. 87–97. 7. Johnsen BO. The effect of an attack by the parasite Gyrodactylus salaris on 29. Mo TA. Virksomheten i 1987 og program for virksomheten i 1988: the population of salmon parr in the river Lakselva, Misvaer in northern Gyrodactylus-undersøkelsene ved Zoologisk Museum, Universitetet i Oslo; Norway. Astarte. 1978;11:7–9. 8. Johnsen BO, Jensen AJ. Infestations of Atlantic salmon, Salmo salar,by 30. Mo TA. Variations of opisthaptoral hard parts of Gyrodactylus salaris Gyrodactylus salaris in Norwegian rivers. J Fish Biol. 1986;29:233–41. Malmberg, 1957 (Monogenea, Gyrodactylidae) on rainbow trout 9. Hansen H, Bachmann L, Bakke TA. Mitochondrial DNA variation of Oncorhynchus mykiss (Walbaum, 1792) in a fish farm, with comments Gyrodactylus spp. (Monogenea, Gyrodactylidae) populations infecting on the spreading of the parasite in south-eastern Norway. Syst Atlantic salmon, grayling, and rainbow trout in Norway and Sweden. Int J Parasitol. 1991;20:1–9. Parasitol. 2003;33:1471–8. 31. Strand DA, Jussila J, Johnsen SI, Viljamaa-Dirks S, Edsman L, Wiik-Nielsen J, 10. Hytterød S, Hansen H, Johansen K, Larsen S. The surveillance programme et al. Detection of crayfish plague spores in large freshwater systems. J Appl for Gyrodactylus salaris in Atlantic salmon and rainbow trout in Norway Ecol. 2014;51:544–53. 2016. In: Annual Report 2016. Oslo: Norwegian Veterinary Institute; 2017. 32. Collins CM, Kerr R, McIntosh R, Snow M. Development of a real-time PCR 11. Johnsen BO, Møkkelgjerd PI. The parasite Gyrodactylus salaris on salmon assay for the identification of Gyrodactylus parasites infecting salmonids in parr in Norwegian rivers, status report at the beginning of year 2000, vol. northern Europe. Dis Aquat Organ. 2010;90:135–42. 617. Trondheim: NINA Oppdragsmelding; 1999. (In Norwegian) 33. Ziętara MS, Lumme J. Speciation by host switch and adaptive radiation in a 12. Sandodden R, Brazier M, Sandvik M, Moen A, Wist AN, Adolfsen P. fish parasite genus Gyrodactylus (Monogenea, Gyrodactylidae). Evolution. Eradication of Gyrodactylus salaris infested Atlantic salmon (Salmo salar)in 2002;56:2445–58. the Rauma River, Norway, using rotenone. Manage Biol Inv. 2018;9:10. 13. Hytterød S, Darrud M, Nasrin Mohammad S, Hansen H. The post-treatment 34. Matejusova I, Doig F, Middlemas SJ, Mackay S, Douglas A, Armstrong JD, control programme to ascertain freedom from infection with Gyrodactylus et al. Using quantitative real-time PCR to detect salmonid prey in scats of salaris in Atlantic salmon 2016. In: Annual Report 2016. Oslo: Norwegian grey Halichoerus grypus and harbour Phoca vitulina seals in Scotland - an Veterinary Institute; 2017. experimental and field study. J Appl Ecol. 2008;45:632–40. 14. Mattilsynet (Norwegian Food Safety Authority). Forskrift om kontrollområde 35. Wilcox TM, Carim KJ, McKelvey KS, Young MK, Schwartz MK. The dual for å forebygge, begrense og utrydde lakseparasitten Gyrodactylus salaris challenges of generality and specificity when developing environmental hos akvatiske dyr, Asker, Drammen, Hof, Holmestrand, Hurum, Krødsherad, DNA markers for species and subspecies of Oncorhynchus. PLoS One. 2015; Lier, Modum, Nedre Eiker, Røyken, Sande, Sigdal, Svelvik og Øvre Eiker 10:e0142008. Rusch et al. Parasites & Vectors (2018) 11:333 Page 12 of 12 36. Dobnik D, Spilsberg B, Kosir AB, Holst-Jensen A, Zel J. Multiplex quantification of 12 European union authorized genetically modified maize lines with droplet digital polymerase chain reaction. Anal Chem. 2015;87: 8218–26. 37. Agersnap S, Larsen WB, Knudsen SW, Strand D, Thomsen PF, Hesselsøe M, et al. Monitoring of noble, signal and narrow-clawed crayfish using environmental DNA from freshwater samples. PLoS One. 2017;12:e0179261. 38. Levy MG, Poore MF, Colorni A, Noga EJ, Vandersea MW, Litaker RW. A highly specific PCR assay for detecting the fish ectoparasite Amyloodinium ocellatum. Dis Aquat Organ. 2007;73:219–26. 39. Gomes GB, Hutson KS, Domingos JA, Chung C, Hayward S, Miller TL, Jerry DR. Use of environmental DNA (eDNA) and water quality data to predict protozoan parasites outbreaks in fish farms. Aquaculture. 2017;479:467–73. 40. Agawa Y, Tani K, Yamamoto S, Hirano C, Shirakashi S. Development of a quantitative polymerase chain reaction assay for the detection of skin fluke Neobenedenia girellae larvae from environmental water. Fisheries Sci. 2016; 82:827–33. 41. Hashizume H, Sato M, Sato MO, Ikeda S, Yoonuan T, Sanguankiat S, et al. Application of environmental DNA analysis for the detection of Opisthorchis viverrini DNA in water samples. Acta Trop. 2017;169:1–7. 42. White VC, Morado JF, Crosson LM, Vadopalas B, Friedman CS. Development and validation of a quantitative PCR assay for Ichthyophonus spp. Dis Aquat Organ. 2013;104:69–81. 43. Hartikainen H, Bass D, Briscoe AG, Knipe H, Green AJ, Okamura B. Assessing myxozoan presence and diversity using environmental DNA. Int J Parasitol. 2016;46:781–92. 44. Hallett SL, Ray RA, Hurst CN, Holt RA, Buckles GR, Atkinson SD, Bartholomew JL. Density of the waterborne parasite Ceratomyxa shasta and its biological effects on salmon. Appl Environ Microb. 2012;78:3724–31. 45. Meinilä M, Kuusela J, Zietara MS, Lumme J. Initial steps of speciation by geographic isolation and host switch in salmonid pathogen Gyrodactylus salaris (Monogenea: Gyrodactylidae). Int J Parasitol. 2004;34:515–26. 46. Hansen H, Bakke TA, Bachmann L. DNA taxonomy and barcoding of monogenean parasites: lessons from Gyrodactylus. Trends Parasitol. 2007; 23:363–7. 47. Deiner K, Altermatt F. Transport distance of invertebrate environmental DNA in a natural river. PLoS One. 2014;9:e88786. 48. Balasingham KD, Walter RP, Heath DD. Residual eDNA detection sensitivity assessed by quantitative real-time PCR in a river ecosystem. Mol Ecol Resour. 2017;17:523–32. 49. Darling JA, Mahon AR. From molecules to management: Adopting DNA-based methods for monitoring biological invasions in aquatic environments. Environ Res. 2011;111:978–88. 50. Goldberg CS, Turner CR, Deiner K, Klymus KE, Thomsen PF, Murphy MA, et al. Critical considerations for the application of environmental DNA methods to detect aquatic species. Methods Ecol Evol. 2016;7:1299–307. 51. Doi H, Uchii K, Takahara T, Matsuhashi S, Yamanaka H, Minamoto T. Use of droplet digital PCR for estimation of fish abundance and biomass in environmental DNA surveys. PLoS One. 2015;10:e0122763. 52. Maruyama A, Nakamura K, Yamanaka H, Kondoh M, Minamoto T. The release rate of environmental DNA from juvenile and adult fish. PLoS One. 2015;10:e114639. 53. Tellenbach C, Grunig CR, Sieber TN. Suitability of quantitative real-time PCR to estimate the biomass of fungal root rndophytes. Appl Environ Microb. 2010;76:5764–72. 54. Baker M. Digital PCR hits its stride. Nat Methods. 2012;9:541–4. 55. Doi H, Takahara T, Minamoto T, Matsuhashi S, Uchii K, Yamanaka H. Droplet digital polymerase chain reaction (PCR) outperforms real-time PCR in the detection of environmental DNA from an invasive fish species. Environ Sci Technol. 2015;49:5601–8. 56. McKee AM, Spear SF, Pierson TW. The effect of dilution and the use of a post-extraction nucleic acid purification column on the accuracy, precision, and inhibition of environmental DNA samples. Biol Conserv. 2015;183:70–6. 57. Tsai YL, Olson BH. Rapid method for separation of bacterial-DNA from humic substances in sediments for polymerase chain-reaction. Appl Environ Microb. 1992;58:2292–5. 58. Hoshino T, Inagaki F. Molecular quantification of environmental DNA using microfluidics and digital PCR. Syst Appl Microbiol. 2012;35:390–5.

Journal

Parasites & VectorsSpringer Journals

Published: Jun 4, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off