Evaluation of potential transfer of the pathogen Saprolegnia parasitica between farmed salmonids and wild fish

pathogens Article Evaluation of Potential Transfer of the Pathogen Saprolegnia parasitica between Farmed Salmonids and Wild Fish Perla Tedesco 1,*,†, Marcia Saraiva 2,†, Jose Vladimir Sandoval-Sierra 3,4 , Maria Letizia Fioravanti 1, Benedetto Morandi 1, Javier Dieguez-Uribeondo 3, Pieter van West 2and Roberta Galuppi 1   Citation: Tedesco, P.; Saraiva, M.; Sandoval-Sierra, J.V.; Fioravanti, M.L.; Morandi, B.; Dieguez-Uribeondo, J.; van West, P.; Galuppi, R. Evaluation of Potential Transfer of the Pathogen Saprolegnia parasitica between Farmed Salmonids and Wild Fish. Pathogens 2021,10, 926. https://doi.org/ 10.3390/pathogens10080926 Academic Editor: Dušan Pali´c Received: 15 June 2021 Accepted: 19 July 2021 Published: 22 July 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1Department of Veterinary Medical Sciences, Alma Mater Studiorum-University of Bologna, Ozzano Emilia, 40064 Bologna, Italy; [email protected] (M.L.F.); [email protected] (B.M.); [email protected] (R.G.) 2 Aberdeen Oomycete Laboratory, International Centre for Aquaculture Research and Development (ICARD), Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, UK; [email protected] (M.S.); [email protected] (P.v.W.) 3CSIC—Real Jardin Botanico, 28006 Madrid, Spain; [email protected]g.co (J.V.S.-S.); [email protected] (J.D.-U.) 4Instituto de Investigación de Recursos Biológicos Alexander von Humboldt, Bogotá111311, Colombia *Correspondence: [email protected]; Tel.: +39-051-209-7065 † The authors contributed equally as co-first authors. Abstract: Saprolegnia infections are among the main parasitic diseases affecting farmed salmonids. The distribution and potential transfer of Saprolegnia spp. between farms and the natural environment has been scarcely investigated. Therefore, this work aimed to study the diversity and abundance of oomycete species in salmonid farms, tributary water, and effluent water systems. Four trout farms in Italy and two Atlantic salmon farms in Scotland were considered. In Italian farms, 532 isolates of oomycetes were obtained from fish and water, at upstream, inside, and downstream the farms. In Scottish farms, 201 oomycetes isolates were obtained from water outside the farm and from fish and water inside the farming units. Isolates were identified to the species level through amplification and sequencing of the ITS rDNA region. In Italy, S. parasitica was significantly more present in farmed than in wild fish, while in water it was more frequently isolated from the wild, particularly in effluent systems, not associated with more frequent isolation of S. parasitica in wild fish downstream the farm. In Scotland, S. parasitica was the most prevalent species isolated from fish, while isolates from water were mostly Pythium spp. with few S. parasitica isolates from upstream and downstream the farms. Keywords: Saprolegnia parasitica; salmonid farms; Oncorhynchus mykiss;Salmo trutta;Salmo marmoratus; Salmo salar; Italy; Scotland 1. Introduction Parasites of wild fish pose a potential threat to aquaculture [ 1 ]. One of the main issues related to intensive aquaculture is the proliferation of parasites, particularly those with direct life cycles, and other infectious agents due to high farming densities [ 2 ], which can subsequently spread outside the farm into the natural environment. Therefore, the risk of diseases spreading from wild to farmed fish, with subsequent proliferation within the farm and transmission into the environment, generates great concern. In salmonid aquaculture, severe problems with respect to parasite exchange between farmed and wild fish have been reported [ 3 – 5 ]. Salmonid farming is mainly based on open aquaculture systems (open-net pens, raceways), which exchange water together with other materials, such as chemicals, waste, and a wide range of potential infectious agents with the natural environment (sea, natural streams). Many oomycetes seem to be ubiquitous in freshwater environments, where they contribute to the structural and functional organization of aquatic ecosystems [ 6 ]. However, Pathogens 2021,10, 926. https://doi.org/10.3390/pathogens10080926 https://www.mdpi.com/journal/pathogens Pathogens 2021,10, 926 2 of 16 the introduction of potentially pathogenic oomycete species into lakes and ponds through fish stocking and other anthropogenic activities has been associated with the decline of amphibian [7] and crustacean [8,9] populations. Infections with oomycetes of the genus Saprolegnia represent one of the main parasitic diseases affecting freshwater-farmed salmonids. Saprolegnia parasitica is of primary importance with respect to infections in fish, while other Saprolegnia species such as S. diclina and S. australis infect fish and their eggs [ 10 – 13 ]. Saprolegnia species can cause heavy losses in salmonid farming [ 10 , 14 ] and are responsible for the ‘winter kill’ in catfish aquaculture [ 15 ]. Although infections are usually more severe in farmed fish than in wild fish [ 16 ], Saprolegnia outbreaks have also been reported in the latter [ 17 – 20 ]. Particularly, high mortality rates were reported in wild populations of brown trout, Salmo trutta, in Spanish rivers [ 18 ]. This evidence raised concern about the possible spread of potentially pathogenic Saprolegnia strains from farmed fish to wild populations. Despite the widespread occurrence of Saprolegnia spp. in wild and aquaculture environments [ 21 ], little attention has been directed to understand the distribution and potential transfer of these agents between the farm and its tributary and effluent systems. A recent study [ 22 ] on the distribution of Saprolegnia in selected trout farms in Croatia indicated a possible role of trout farms as a source of spreading Saprolegnia spp. into the environment. Until recently, difficulties in identifying oomycetes up to species level with morphological methods contributed to the lack of knowledge concerning their circulation in different farmed and wild fish [ 13 , 23 ]. The use of molecular techniques, based on amplification and sequencing of the ITS region, and phylogenetic approaches allowed to resolve problems in the identification of oomycetes and to establish DNA-based molecular operational taxonomic units (MOTU) for species delimitation [13]. The aim of this work is to study the distribution and species composition of oomycete assemblage through molecular identification of strains isolated in two different types of salmonid breeding and in their tributary and effluent water systems (trout farms located in Italy and Atlantic salmon farms located in Scotland), in order to: (i) study the diversity and abundance of oomycete species; (ii) evaluate the potential transfer of pathogenic species of Saprolegnia, specifically Saprolegnia parasitica, between wild and farmed fish, and (iii) determine whether there is an impact of salmonid farms on the diffusion of S. parasitica to wild fish in two different types of fish farming. 2. Results Due to differences in the farmed species and the farming system/environment, the results will be described separately for Italy and Scotland. 2.1. Environmental Parameters In Italian trout farms, recorded values of Dissolved Oxygen (DO) ranged from 3.9 to 16.8, with a median of 9.7 ppm (IQRs: 7.7–10.6), water temperatures ranged from a minimum of 5.1 to a maximum of 14.2 ◦ C with a median of 11.6 ◦ C (IQRs: 9.5–12.5). Finally, the pH observation revealed a median of 7.79, ranging from 7.05 to 8.41 (IQRs: 7.43–7.96). (Table 1; Figure 1a). A negative correlation was found between Oxygen saturation and water’s temperature (rho = 0.6152, p< 0.0001) (Figure 1b). From the 239 collected fish, 458 single inocula were obtained, and 120 oomycetes strains were isolated. Additionally, 412 isolates were obtained from baits. In Scottish salmon farms, water temperature values ranged from 6.1 to 16.7 ◦ C with a median of 10.3 ◦ C (IQRs: 7.3–14.3), pH ranged from 4.5 to 7.9 with a median of 6.9 ( IQRs: 6.6–7.2 ), being more constant at farm G (6.5–7.6), and dissolved oxygen values ranged from 90 to 94 mg/L (IQRs: 91–93) with a median of 92mg/L (Table 2). A negative correlation was found between DO and water temperature (p= 0.04) and between DO and pH (p= 0.02) (Figure 2). From the 200 fish samples collected, 128 oomycetes were obtained. Moreover, 73 oomycetes were collected from hempseeds. Pathogens 2021,10, 926 3 of 16 Table 1. Characteristic of the farms considered in Italy and number of samples collected. Coordinates: A = 45 ◦ 42 0 57.2 00 N, 11 ◦ 40 0 49.2 00 E, 84 m a.s.l.; B = 46 ◦ 32 0 644 00 N, 11 ◦ 12 0 448 00 E, 250 m a.s.l; C = 45 ◦ 54 0 53 00 N, 13 ◦ 4 0 35 00 E, 30 m a.s.l.; D = 46◦003400 N , 12 ◦ 29 0 38 00 E, 42 m a.s.l. farmed species: 1 = Oncorhynchus mykiss; 2 = Salmo marmoratus and Salmo trutta water supply: 3 = river; 4 = mixed (river/spring). Environmental Parameters (Min–Max) n◦Samples Collected Farm ID (Farmed Species/Water Supply T (◦C) pH O2 (mg/L) Upstream Farm Downstream Tot A (1/3) 5.2–9.8 7.25–8.41 10–16.8 fish baits 20 34 30 32 16 38 66 105 B (1/3) 11.7–13 7.05–7.45 3.9–10.1 fish baits 1 38 30 45 42 30 73 113 C (2/4) 9.5–13.2 7.04–8.21 8.2–11.2 fish baits 15 39 25 27 20 35 60 101 D (2/3) 9.4–14.2 7.45–7.85 7.3–9.7 fish baits 10 39 20 30 10 24 40 93 Figure 1. ( a ) distribution of Dissolved Oxygen, temperature, and pH values collected during the visits in Italian farms; (b) Correlation between Oxygen saturation and water temperature in Italian farms (rho = 0.6152, p< 0.0001). Table 2. Characteristics of the farms considered in Scotland and number of samples collected. Coordinates: F = 56.966 ◦ N 5.134◦W, G = 56◦5800700 N 4◦5403800 W; Farmed species: Atlantic salmon (Salmo salar). Environmental Parameters (Min–Max) n◦Samples Collected upstream farm downstream tot Farm ID T (◦C) pH O2(mg/L) F 6.1–16.7 4.5–7.9 91–93 fish baits 0 240 200 240 0 240 200 720 G 6.5–14.7 6.5–7.6 90–94 fish baits 0 240 200 240 0 240 200 720 2.2. Oomycetes Diversity A total of 532 oomycetes isolates from Italy, and 201 from Scotland were identified by phylogenetic and molecular taxonomic analyses; the sequences obtained corresponded to the following MOTUs (Figure 3): Achlya colorata,Saprolegnia australis,Saprolegnia delica, Saprolegnia diclina,Saprolegnia ferax,Saprolegnia parasitica,Pythium aquatile,Pythium dissimile, Pythium rhizooryzae,P. dissotocum, P. oopapillum,P. pectinolyticum and other could only be identified to genus level, e.g., Saprolegnia sp., Leptolegnia sp., Pythium sp. Pathogens 2021,10, 926 4 of 16 Figure 2. Pearson’s correlation coefficients calculated considering environmental data from Scottish farms. In Italian farms and connected aquatic systems, S. parasitica was isolated from 156 samples (17.9%). The frequency of S. parasitica, both from fish and water, was significantly different among the four surveyed farms (including their surrounding waters) ( χ2(3) = 87.5 8; p< 0.001). Detection proportion ranged from 0.39% (1 positive sample) in farm B to a maximum of 32.4% (58 positive samples) in farm D. In particular, in farm B, where saprolegniosis is not considered a significant problem in farmed O. mykiss,S. parasitica was not isolated from fish upstream and inside the farm, but only from a wild S. trutta without lesions collected about 1 km downstream. Figure 4shows the distribution of frequencies by farm. DO did not significantly affect the probability of detecting S. parasitica. On the contrary, it was more frequently detected when water temperatures were colder (H = 10.48; p= 0.001) and with higher pH values (H = 45.58; p< 0.001). Quantitative analysis showed that there was 13% more chance to isolate S. parasitica for each Celsius degree decreased (p= 0.001). S. parasitica was significantly more frequently isolated from fish (21.2% of the samples) compared to 14.3% of isolates from water baits (χ2= 6.933, p= 0.008). Indeed, S. parasitica represented 80.8% of the oomycetes isolates from fish, while other species of oomycetes (e.g., S. ferax,S. delica, and S. australis) were more rarely found from different fish species (O. mykiss,S. trutta, and Sq. cephalus) alone or in association with S. parasitica. In eight farmed and wild fish (O. mykiss,S. trutta,Sq. cephalus, and P. fluviatilis), species of the genus Pythium were obtained. Particularly, in one wild S. trutta sampled downstream the farm, Pythium sp. was isolated from fin lesions. Isolates identified as Leptolegnia sp. were obtained in wild fish without lesions (1 O. mykiss, 4 S. trutta, and 1 S. cephalus) sampled downstream in one of the monitored farms. Using baits, a more heterogeneous variety of species was isolated (either from farm, upstream, and downstream water), and S. ferax was predominant (31.6% of the isolates), followed by S. delica and S. australis, while S. parasitica represented only 14.3% of the isolates). In particular, in farm B, Pythium spp. and S. ferax seemed to be prevalent in the aquatic environment, and Pythium sp. was also isolated from fish without lesions (2 farmed O. mykiss, 3 S. cephalus, and 1 P. fluviatilis downstream the farm) (Table 3; Figure 5a). Pathogens 2021,10, 926 5 of 16 Figure 3. Maximum Likelihood tree based on ITS nrDNA sequences of oomycetes isolated during the present study with indication of the MOTUs (in bold) identified. Pathogens 2021,10, 926 6 of 16 Figure 4. Frequencies of S. parasitica detected within the four Italian farms and the relative 95% CIs. Pairwise comparisons showed differences between farm A and B; B and C and B and D (Fischer’s exact p< 0.001), and between farm C and D (p= 0.002). Different letters indicate significant difference. Table 3. Oomycete isolates in Italy from water and fish samples. Farm ID n◦Isolates Identified Species Identified Upstream Farm Downstream Tot Upstream Farm Downstream A fish 4 35 6 45 2S. parasitica 1Pythium sp. 1S delica 31 S. parasitica 4S delica 6Leptolegnia sp. baits 35 32 38 105 1S. australis 11 S. parasitica 13 S. delica 9Pythium sp. 1S. ferax 24 S. delica 6S. australis 1S. ferax 1S. parasitica 13 S. parasitica 11 S. delica 7Pythium sp. 4S. australis 1S. ferax 2S. hypogyna B fish 0 2 5 7 0 2 Pythium sp. 3Pythium sp. 1S. parasitica 1S.ferax baits 38 45 30 113 19 Pythium sp. 19 S. ferax 29 S. ferax 6Pythium sp. 10 S. delica 18 S. ferax 4S. australis 5Pythium sp. 3S. delica C fish 4 12 10 26 4 S. parasitica 12 S. parasitica 9S. parasitica 1S. australis baits 39 27 35 101 23 S. ferax 3S. parasitica 6Pythium sp. 5S. australis 2S. delica 6S. parasitica 11 S. ferax 7S. australis 2S. delica 1Pythium sp. 22 S. australis 7S. parasitica 4Pythium sp. 1S. delica 1S. ferax D fish 3 32 7 42 2S. parasitica 1Pythium sp. 30 S. parasitica 2S. ferax 6S. parasitica 1S. australis baits 39 30 24 93 11 S. australis 7S. parasitica 13 S. ferax 6Pythium sp. 1Saprolegnia sp. 1Achlya colorata 11 S. australis 10 S. ferax 5S. delica 4S. parasitica 7S. parasitica 7S. australis 3 S. ferax 6 S. delica 1 Pythium sp. Pathogens 2021,10, 926 7 of 16 Figure 5. Species composition of oomycetes isolated from fish and water in Italian (a) and Scottish (b) farms. In Scottish farms and connected aquatic systems, S. parasitica was significantly more frequently isolated from fish compared to isolates from water baits (p< 0.0001). Indeed, S. parasitica represented 62.5% of the oomycetes isolates from fish, while other species of oomycetes, such as S. diclina or S. australis, were less found (9.3% and 13.54%, respectively). Pythium spp. were also found in fish (10.4%). In contrast with the Italian farms, no S. delica or S. ferax were isolated from fish. Farm F, although without reported saprolegniosis problems, presented a statistically significantly higher number of S. parasitica isolates (41) when compared to Farm G (20) with p= 0.006. This is probably the result of an intensive treatment regime from Farm G. In addition, the species distribution pattern was different between farms, with unique isolates of Saprolegnia sp1 being isolated from farm F and unique S. delica and S. parahypogyna isolates from farm G. Furthermore, a significant number of Pythium spp. isolates was obtained at farm G compared to farm F. In water, a diversity of Pythium spp. was isolated, representing 92.30%, followed by S. parasitica,S. delica and S. diclina. A reduced number of isolates was obtained at fish farms with only Pythium spp.; this is possibly due to the treatments carried out at the farm and the daily removal of mort’s. Upstream the fish farms, the only Saprolegnia species isolated was S. parasitica, while downstream only S. delica and S. diclina were isolated (Figure 5b, Table 4). No statistically significant difference was found between the water collection sites. 2.3. Transfer of Saprolegnia parasitica between Wild and Farmed Fish and Impact of Salmonid Farms on the Spread of S. parasitica to Wild Fish In Italian farms, the isolation of S. parasitica was significantly higher in farmed fish than in wild fish from both upstream and downstream waters ( χ 2(2) = 55.37, p< 0.001), while no differences were found between wild fish captured upstream and downstream. On the contrary, the presence of S. parasitica in water, assessed through the use of baits, was significantly lower in farms compared to outside ( χ2 (2) = 9.085; p= 0.011) with particular references to downstream (Figure 6). Pathogens 2021,10, 926 8 of 16 Table 4. Oomycete isolates in Scotland from water and fish samples. Farm ID n◦Isolates Identified Species Identified Upstream Farm Downstream Tot Upstream Farm Downstream F fish 0 61 0 61 0 2P. dissotocum 2P. flevoense 2P. monospermum 1P. pachycaule 5S. australis 6S. diclina 41 S. parasitica 2Saprolegnia sp1 0 baits 14 8 18 40 2Phy. gonapodyides 3P. dissotocum 6P. dissotocum 7P. dissotocum 2P. pachycaule 1P. pachycaule 2P. pachycaule 2P. pyrilobum 7P. pyrilobum 2P. pyrilobum 1S. parasitica 2Pythium sp03 1S. parasitica 1Pythium sp05 1S. parasitica G fish 0 67 0 67 0 17 P. dissotocum 1P. flevoense 1P. monospermum 1P. pachycaule 7P. pyrilobum 1Pythium sp01 2Pythium sp03 1Pythium sp05 1Pythium sp08 8S. australis 1S. delica 4S. diclina 2S. parahypogyna 20 S. parasitica. 0 baits 18 9 14 41 6P. dissotocum 5P. dissotocum 1P. pachycaule 3P. pyrilobum 10 P. dissotocum 1S. diclina 1S. delica 2Pythium spp. 1P. pachycaule 7P. pyrilobum 2Pythium sp03 1Pythium sp05 1S. parasitica Figure 6. Isolation of S. parasitica in fish and water from upstream (before), in-farm (inside) or downstream (after) sampling sites in Italy. Pathogens 2021,10, 926 9 of 16 The quantitative analysis remarks these differences, showing six times less chance to isolate S. parasitica in wild fish from upstream and downstream compared to farmed fish ( p< 0.001 ), while when the isolation was made from waters using baits, it was three times more likely to detect S. parasitica downstream compared to in-farm (OR: 3.02; p= 0.004), whereas no significant difference was found when in-farm and upstream were considered (p= 0.132). These results are also summarized in Table 5. Table 5. Probability to isolate S. parasitica upstream or downstream the Italian farms compared to the in-farms baseline based on different sources and how the chance to isolate S. parasitica changes as a function of the water temperature. * Odds ratio is reported as the inverse (1/OR). Categories OR p-Value 95% CI fish * in-farm baseline - - upstream 6.39 <0.001 2.92–13.95 downstream 5.97 <0.001 3.31–10.78 baits in-farm baseline - - upstream 1.8 0.132 0.26–1.19 downstream 3.02 0.004 1.43–6.39 T (◦C) - 1.13 0.001 1.05–1.21 In Scottish farms, the presence of S. parasitica in water, assessed using hempseeds as baits, was not statistically significant compared to upstream and downstream. Nevertheless, S. parasitica was isolated from farmed fish at the farms, with significant differences when compared to water (p= 0.0001). No wild Atlantic salmon was sampled due to conservation restrictions. 3. Discussion 3.1. Oomycetes Diversity Overall, a high number of oomycete isolates were obtained from fish and baits in the Italian and Scottish farms under study. Particularly, for Italian farms, the results obtained by culture/isolation of Oomycetes from water showed that hempseed baits allow the isolation of Saprolegnia spp., including S. parasitica, and other oomycete species from water in a very efficient way. Although S. parasitica is more frequently isolated from fish, our findings highlight the usefulness of environmental baits to isolate Saprolegnia spp. and other oomycetes during epidemiological studies, allowing to reduce the sampling of fish. Besides representing a useful tool to avoid invasive techniques in fish, the use of hempseed baits allows overcoming problems deriving from difficulties in sampling wild fish. For example, in Italy, in the upstream water system of farm B, only a few specimens of protected fish species were present; in farm A, in rainy periods the use of electrofishing was not feasible, and angling is difficult; in farm C, that receives the water from a spring, very few wild fish were present upstream. Saprolegnia species other than S. parasitica isolated in the present study (S. delica, S. australis,S. ferax, and S. turfosa), mainly through the use of baits, are often considered as part of the aquatic ecosystem and of lower importance for farming activities [ 21 ]. However, these species may also affect the aquatic fauna: particularly, S. ferax was involved, together with S. diclina, in mortalities of amphibian embryos [ 24 ]. Results of an experimental study [ 25 ] suggested that the transfer of S. ferax from restocked rainbow trout to wild amphibians was a possible cause of the decline of Bufo boreas populations, although few scientific data are available to support this hypothesis. Leptolegnia species are often isolated from arthropods exuviae or from mosquito larvae [ 26 ]. Interestingly, in the present study, Leptolegnia was only found in six wild fish collected downstream the farm A; its isolation might be linked to the abundant presence of amphipod crustaceans in this environment. These amphipods are intermediate hosts of the swim-bladder nematode Cystidicola farionis and several acanthocephalans, parasites often encountered in trout cultured in farm A and may also be vectors of Leptolegnia spp. for other crustaceans and fish. However, no lesions associated with Leptolegnia were detected in the six positive fish. Pathogens 2021,10, 926 16 of 16 34. White, T.J.; Bruns, T.D.; Lee, S.B.; Taylor, J.W. Amplification and Direct Sequencing of Fungal Ribosomal RNA Genes for Phylogenetics. In PCR Protocols: A Guide to Methods and Applications; Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J., Eds.; Academic Press: New York, NY, USA, 1990; pp. 315–322. 35. Kearse, M.; Moir, R.; Wilson, A.; Stones-Havas, S.; Cheung, M.; Sturrock, S.; Buxton, S.; Cooper, A.; Markowitz, S.; Duran, C.; et al. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 2012,28, 1647–1649. [CrossRef] [PubMed] 36. Katoh, K.; Standley, D.M. MAFFT multiple sequence alignment software version 7: Improvements in performance and usability. Mol. Biol. Evol. 2013,30, 772–780. [CrossRef] [PubMed] 37. Katoh, K.; Toh, H. Improved accuracy of multiple ncRNA alignment by incorporating structural information into a MAFFT-based framework. BMC Bioinform. 2008,9, 212. [CrossRef] 38. Katoh, K.; Toh, H. Recent developments in the MAFFT multiple sequence alignment program. Brief. Bioinform. 2008 ,9, 286–298. [CrossRef] 39. Stamatakis, A. RAxML version 8: A tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 2014 ,30, 1312–1313. [CrossRef] 40. Kim, H.Y. Statistical notes for clinical researchers: Chi-squared test and Fisher’s exact test. Restor. Dent. Endod. 2017 ,42, 152–155. [CrossRef] [PubMed]