Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2024 Dec 5;48(3):e14061. doi: 10.1111/jfd.14061

18S rRNA Metagenomic Analysis of Nodular Gill Disease in Swiss Rainbow Trout (Oncorhynchus mykiss)

James W Wynne 1, Chloe J English 2,✉, Stefania M Vannetti 3,4, Megan Rigby 1, Petra R Quezada‐Rodriguez 1, Ralph Knüsel 4, Christine Huynh 5, Heike Schmidt‐Posthaus 3
PMCID: PMC11837460  PMID: 39639673

ABSTRACT

Nodular gill disease (NGD) is a serious proliferative gill condition that affects farmed salmonids, particularly in Europe. While the cause of NGD remains unknown (and maybe multifactorial), various amoebae are often isolated from the gills of affected fish and can in some cases be seen associated with lesions by histopathology. The present study aimed to quantify the abundance of different amoeba species directly from the gills of rainbow trout affected by NGD and healthy controls. An 18S rRNA amplicon metagenomic approach was employed to profile the diversity and abundance of micro‐eukaryotes (including amoebae) while suppressing the amplification of host DNA using a salmonid‐specific C3 spacer blocking primer. The 18S rRNA metagenomics approach identified a diversity of micro‐eukaryotes on the gills of rainbow trout, including the phylum's Amoebozoa, Diatomea, Platyhelminthes and Ciliophora. Rainbow trout clinically affected by NGD had a significantly higher abundance of a specific sequence (zOTU2) classified as Vannella sp. compared to healthy controls. A quantitative PCR assay was then developed and validated which accurately quantified the abundance of this Vannella sp. sequence from a NGD outbreak in a Swiss rainbow trout farm. Additional PCR and Sanger sequencing analysis of the zOTU2 sequence demonstrated that this sequence is most likely derived from Vannella mustalahtiana. Our study highlights the potential role of Vannella mustalahtiana in NGD in Switzerland and further describes a specific and validated diagnostic PCR assay for accurate detection of this Vannella species.

Keywords: aetiology, Amoebozoa, aquaculture, gill health, micro‐eukaryotes, Vannella

1. Introduction

Nodular gill disease (NGD) is an emerging proliferative gill disease of farmed freshwater salmonids. Clinically, fish affected by NGD show clear signs of respiratory distress including apathy, tachypnoea, dyspnoea and abnormal swimming behaviour at the water surface with flared opercula (Vannetti et al. 2023). If left un‐treated NGD can cause significant mortality exceeding 60% of the population (Cocco et al. 2023; Vannetti et al. 2023). Gross signs of NGD include proliferation of white nodules on gill filaments with increased mucus production (Buchmann et al. 2004; Dyková et al. 2010; Quaglio et al. 2016). Histologically these proliferative nodules are characterised by extensive epithelial hyperplasia with fusion of lamellae and adjacent filaments associated with the presence of amoebae (Kudryavtsev et al. 2022; Quaglio et al. 2016; Speare 1999).

Nodular gill disease has now been reported across several regions including Canada (Speare 1999), Italy (Cocco et al. 2023; Perolo et al. 2019), Russia (Kudryavtsev et al. 2022), Germany (Dyková et al. 2010), New Zealand (Tubbs, Wybourne, and Lumsden 2010), Denmark (Buchmann et al. 2004; Jensen et al. 2020), Poland (Antychowicz 2007), Czech Republic (Dykova and Tyml 2016) and most recently Switzerland (Vannetti et al. 2023). Within these regions, NGD appears to be a seasonal disease, occurring most often in spring (Buchmann et al. 2004; Dykova and Tyml 2016; Kudryavtsev et al. 2022; Quaglio et al. 2016; Tubbs, Wybourne, and Lumsden 2010; Vannetti et al. 2023). Most reports of NGD have been in farmed rainbow trout ( Oncorhynchus mykiss ); however, this disease has also been reported in arctic char ( Salvelinus alpinus ) (Speare 1999), Chinook salmon ( Oncorhynchus tshawytscha ) (Tubbs, Wybourne, and Lumsden 2010) and brown trout ( Salmo trutta ) (Perolo et al. 2019).

While the aetiological agent(s) of NGD remain unknown, it is generally accepted that amoebae play a role in this disease (Padros and Constenla 2021). Wet mounts or smears of NGD‐affected gill tissue/mucus frequently identify amoebae of unknown identity associated with the margins of proliferative gill filaments (Perolo et al. 2019; Vannetti et al. 2023). Furthermore, histological examination of NGD‐affected gills has also identified amoebae with varying morphology associated with regions of hyperplasia and lamellar fusion (Brocca et al. 2024; Dyková et al. 2010; Kudryavtsev et al. 2022). Amoebae have also been successfully isolated and cultured from the gills of NGD‐affected rainbow trout. Indeed, Dyková et al. (2010) isolated nine amoeba strains classified into the genera Acanthamoeba, Hartmannella, Naegleria, Protacanthamoeba and Vannella from NGD‐affected rainbow trout in Germany. Similarly, in Switzerland Vannetti et al. (2023) isolated a diversity of genera from NGD‐affected rainbow trout including Cochliopodium sp., Naegleria sp., Vannella sp., Ripella sp., Saccamoeba sp. and Mycamoeba sp. Recently a proposed new species of Vannella designated V. mustalahtiana sp. nov was isolated and classified from NGD‐affected rainbow trout in Northwestern Russia (Kudryavtsev et al. 2022) and a study of NGD in Northeastern Italy isolated several amoebae strains which were classified as Rosculus, Copromyxa, Ptolemeba, Naegleria and Ripella based on morphology and phylogenetic analysis (Brocca et al. 2024). While all these studies provide valuable information concerning the diversity of amoebae that colonise the gills of NGD‐affected trout, they cannot identify a causative agent.

Usually, the prevalence of causal agents is higher in diseased rather than healthy hosts and the abundance of the agent positively correlates with the severity of disease. For instance, the abundance of Neoparamoeba perurans, the causal agent of amoebic gill disease (AGD) in marine salmonids globally (Crosbie et al. 2012; Oldham, Rodger, and Nowak 2016), is consistently associated with greater severity of AGD pathology (Bridle et al. 2010; English, Swords, et al. 2019). The present study utilised an 18S rRNA metagenomics approach to profile the diversity and abundance of micro‐eukaryotes (including amoebae) directly from the gills of rainbow trout affected by varying severity of NGD‐associated pathology and clinically healthy controls from a single farm in Switzerland. A strain‐specific qPCR assay was then developed to validate the trends identified in the amplicon sequencing.

2. Materials and Methods

2.1. Sampling

A total of 206 rainbow trout were sampled in July 2022 from a common holding tank experiencing an NGD outbreak at a commercial rainbow trout farm in northwest Switzerland. The fish were maintained at a stocking density of 25 kg/m2 in a partial recirculation system with inflow water originating from spring water, ground water and surface water. The average water quality during Spring consisted of temperature 11.2°C ± 2.2°C, pH 7.7 ± 0.2, dissolved oxygen 8.2 ± 2.4 mg/L, salinity 0.1 ± 0.6 ppt, ammonia 0.003 ± 0.002 mg/L and nitrite 0.4 ± 0.9 mg/L. All fish were sampled from a single tank with an average weight of 61 g. A proportion of the fish within the holding tank were showing clear clinical signs of NGD, including thickened filaments, discoloration and white mucoid spots (Figure 1). Ninety‐seven clinically healthy (no gross sign of NGD) and 109 clinically affected (gill proliferation, discoloration and high amount of mucus) fish were sampled. Fish were sedated with 0.15 g/L of MS222 (tricaine methanesulphonate; MS 222, Pharmaq) and the gill score [from 0 (no lesion) to 5 (severe lesion, thickened filaments, discoloration, white mucoid spots)] was determined using the scoring system described by Vannetti et al. (2023). A sterile cotton swab (Henry Schein) was used to sample gill mucus from all gill arches and stored in RNAlater (ThermoFisher Scientific). After swabbing, the fish were placed in a freshwater recovery tank before returning to their original tank. Twenty fish with high severity of NGD‐like pathology and showing gross clinical signs (high respiratory rate, swimming at the surface, weak) from the same tank were also euthanised for gill histology to support gill disease diagnosis. Fish were humanely killed with a high dose of MS222 and then the gills cut and bleed. The gill swab was taken and a piece of gill arch with lesion was fixed in 10% neutral buffered formalin for 24 h for histology. All fish used in this study were approved for sampling by the ethical commission (permission number BE102/2022).

FIGURE 1.

FIGURE 1

Open in a new tab

Macroscopic and histologic appearance of healthy and affected rainbow trout. (a) Macroscopic appearance of gills of a clinically healthy rainbow trout. Filaments are clearly distinguished, without proliferation. (b) Histologic appearance of gills of a clinically healthy rainbow trout. Lamellae are clearly distinguished, without proliferation; H&E stain, bar = 50 μm. (c) Macroscopic appearance of gills of an NGD‐affected rainbow trout. Filaments are proliferated, mainly at the tips of the filaments, with whitish discoloration. (d) Histologic appearance of gills of a clinically affected rainbow trout. Lamellae show severe epithelial proliferation with fusion of lamellae (stars) and squamous metaplasia (open arrowheads), multifocal necrosis of epithelial cells (arrow with open arrowhead); H&E stain, bar = 100 μm. (e) Native smear of affected gill tissue, amoeba (closed arrowhead) present between lamellae. (f) Histologic picture of affected gill tissue, showing severe epithelial proliferation and infiltration with eosinophilic granular cells (open arrowhead) and overlying amoeba (closed arrowhead); H&E stain, bar = 50 μm. (g) Higher magnification of affected filament tips showing epithelial necrosis, infiltration with macrophages and lymphocytes (open arrowheads) as well as amoebae (closed arrowheads); H&E stain, bar = 10 μm.

2.2. DNA Extraction and 18S PCR

Genomic DNA was extracted from swabs using the DNeasy Blood and Tissue Kit (QIAGEN) according to the manufacturer's instructions. DNA was quantified with NanoDrop One (ThermoFisher Scientific) and stored at −20°C until further examination. 18S rRNA amplicon sequencing was performed on each sample using methods described by Patchett, Rigby, and Wynne (2024). Briefly, the universal 18S rRNA amplification primers 1391F (Amaral‐Zettler et al. 2009) and EukBr (Stoeck et al. 2010) were used to amplify a ~260 base‐pair amplicon containing Illumina adapter binding sequences. A salmonid C3 spacer blocking primer (Salmonid_block_I‐short_1391f) was utilised to suppress amplification of rainbow trout 18S rRNA (Patchett, Rigby, and Wynne 2024). Amplification of 18S rRNA was performed from 20 to 30 ng of appropriate template DNA in 20 μL reactions using Platinum Taq DNA Polymerase (Bio‐Rad, USA), according to the manufacturer's directions. Controls with no DNA template were included. Forward and reverse primers were added at a final concentration of 0.25 μM each. Reactions were performed with blocking primer at 1.6 μM. Thermocycling was performed with an initial melting step at 94°C for 3 min, followed by 25 cycles of 94°C for 45 s, 65°C for 15 s, 57°C for 30 s and 72°C for 90 s, and a final extension at 72°C for 3 min. PCR amplicons were visualised by gel electrophoresis to confirm 18S rRNA amplification.

2.3. Amplicon Sequencing and Analysis

18S rRNA amplicons prepared by end‐point PCR were sequenced by 300 base‐pair, paired‐end sequencing on the Illumina MiSeq platform, by the Australian Genome Research Facility (AGRF, Australia). Raw sequencing data is available in the NCBI Sequence Read Archive (SRA) at BioProject PRJNA1161435. Paired‐end sequencing reads were merged using FLASH version 2.2.0, based on a minimum overlap of 60 base‐pairs (Magoc and Salzberg 2011). Primer sequences were trimmed, and low‐quality reads were removed based on a quality score of 20 using USEARCH version 11.0.667 (Edgar 2010). Reads with a sequence length greater than 160 base‐pair were removed using mothur version 1.43.5 (Schloss et al. 2009). Unique sequences were identified using VSEARCH version 2.21.1 (Rognes et al. 2016). Taxa were extracted from the SILVA 18S database NR v138 using QIIME version 2021.4 (Caporaso et al. 2010) and assigned to unique sequences with a sintax cut‐off of 0.5 using VSEARCH. Detected taxa were rarefied, and the most abundant unique sequences were plotted at the genus level using phyloseq version 1.40.0 in R version 4.2.1 (McMurdie and Holmes 2013). Other plots were generated using ggplot2 version 3.3.6 (Wickham 2009) and statistics calculated by paired Student's t‐test using the R version 4.2.1 base statistics package.

2.4. Histology

Gill histology was performed on 20 severely NGD‐affected rainbow trout. Formalin‐fixed gill tissue was dehydrated in ethanol series, embedded in paraffin, sectioned and stained with haematoxylin–eosin as previously described by Vannetti et al. (2023).

2.5. Quantitative PCR Assay Design and Optimisation

A multiple sequence alignment was prepared to identify regions suitable for designing a quantitative PCR (qPCR) assay specific to the zOTU2 Vannella strain identified during the amplicon sequence analysis. All the amoeba 18S rRNA sequences identified in the metagenomic analysis were within Vannellida and Dactylopodida, two closely related families according to Smirnov et al. (2020) 18S rRNA Discosea phylogeny. Hence, a Vannellida and Dactylopodida sequence dataset were collated using publicly available databases (NCBI; www.ncbi.nlm.nih.gov/) and included relevant high blast hits and fish‐associated amoeba strains (particularly gill‐associated) from nine targeted genera. The 46 reference sequences chosen from the databases were as long as possible, around 2000 base‐pairs, and were derived from well‐characterised strains from recognised culture collections such as the American Type Culture Collection (ATCC; www.atcc.org/), the Culture Collection of Algae and Protozoa (CCAP; www.ccap.ac.uk/) or the Institute of Parasitology, Czech Republic (Dykova and Kostka 2013). A multiple sequence alignment was performed with Geneious 2023.2 (https://www.geneious.com) and is included as Data S1.

A TaqMan qPCR assay was designed to amplify 71 base‐pairs of the zOTU2 Vannella. Primers and probes were designed with SnapGene and the assay including the forward primer Val_1826F (5′‐ATCCGGTGAAATCCTCGGATC‐3′), reverse primer Val_1875R (5′‐AATCAACTTCTCTCAGCAATCTC‐3′) and probe Val_1849P (5′‐AGCTTTCATCTCTAATCTTTGGT‐3′) with a 5′ FAM reporter and 3′ TAMRA quencher. The concentration of forward and reverse primers (between 100 and 400 nM) and probe (between 100 and 300 nM) was optimised with 1.5 μL of the Vannella zOTU2‐specific gBlock gene fragment (Integrated DNA Technologies) diluted 10−7 as the standard template. The optimal probe and primer concentrations were selected based on the combination that amplified the lowest ct value and had the lowest variation between technical replicates (Table S1). After primer and probe optimisation, all PCRs were performed in 10 μL reactions with 200 nM forward primer, 200 nM reverse primer, 200 nM probe, x2 TaqMan Universal PCR Master Mix (Applied Biosystems, Warrington, UK) and 1.5 μL of template. PCR reactions underwent the following thermal cycle conditions: 95°C for 10 min, then 95°C for 15 s and 57°C for 1 min for 40 cycles with a ViiA 7 Real‐Time PCR Machine (Applied Biosystems). To generate a standard curve and determine the efficiency of the assay, the amplicon‐specific gBlock gene fragment was serially diluted 10‐fold and amplified in triplicate reactions as detailed above. Amplification efficiency was calculated based on the ct slope method (Efficiency = [10(−1/slope)] – 1), and linearity was demonstrated with the coefficient of determination (R 2). The equivalent number of 18S rRNA copies corresponding to each point in the standard curve was calculated with the formula (C) (M) (1 × 10−15 mol/fmol) (Avogadro's number) = copy number/μL, where C was the current concentration (ng/μL), and M was the molecular weight in fmol/ng.

The specificity of probe and primers was theoretically assessed using the alignment generated during the design phase and with NCBI nucleotide Basic Local Alignment Search Tool (BLASTn) to identify potential non‐target amplification. The assay specificity was assessed by amplifying triplicate reactions with 15 ng of DNA from seven gill‐associated amoeba strains characterised in English, Swords, et al. (2019), English, Tyml, et al. (2019) (Vannella sp. MV3, Vannella sp. MV4, Neoparamoeba perurans MP1, Nolandella sp. MX5, Pseudoparamoeb asp. MX1, Vexillifera sp. MX6) and four amoeba cultures from American Type Culture Collection (ATCC) (Neoparamoeba pemaquidensis ATCC 50172 TM, Hartmannella vermiformis ATCC, Acanthamoeba jacobsi ATCC 30732 TM, Nolandella sp. ATCC PRA‐27 TM and 30967 TM). The Vannella‐ zOTU2 gBlock fragment was used as the positive control.

The 10‐fold dilution series of amplicon‐specific gBlock gene fragment was then used to determine the limit of detection (LOD). The lowest dilution that provided a ct in all replicates underwent a 2‐fold dilution and was tested in quadruplicate. The mean of the lowest concentration from the 2‐fold dilution that provided a ct value in all replicates was deemed the LOD. LOD was then expressed as copy numbers per PCR reaction.

The reproducibility of the assay was assessed by amplifying 10 gill swab DNA samples (5 with high Vannella zOTU2 relative abundance and 5 with low relative abundance based on sequence data) in triplicate on three consecutive days. Assay reproducibility was assessed in terms of variation in copy number between technical replicates (intra‐assay variance) and between different days (inter‐assay variance) and was expressed as the mean coefficient of variation (CV). The standard curve was included on each plate to determine copy number. All validation assays used the appropriate positive and negative controls.

2.6. Quantitative PCR Analysis of Gill Swab

The concentration of each rainbow trout gill swab DNA sample was measured with a Nanodrop ND‐1000 spectrophotometer (LifeTechnologies) and then normalised to 10 ng/μL. Each sample (15 ng) was tested in triplicate reactions as described above. The standard curve was used to determine copy number as a measure of zOTU2 abundance. Each plate contained a negative no‐template control and a positive control with zOTU2 plasmid as template.

The 18S rRNA copy number determined by qPCR was compared with the relative abundance of the zOTU2 determined by metagenomic analysis. The linear correlation was measured by the Pearson correlation coefficient using R version 4.3.2 base statistics package. Vannella sp. abundance determined by qPCR was also compared between rainbow trout classed with healthy or NGD‐affected gills with a Welch t‐test using the R version 4.3.2 base statistics package.

2.7. Characterisation of Full‐Length 18S rRNA for Vannella sp.

Given the zOTU2 sequence generated from the metagenomic analysis was relatively small at ~190 bp (following removal of primers and Illumina adapters), we attempted to amplify and sequence a larger region of this 18S rRNA gene to provide a better taxonomic classification. To achieve this, we designed two forward primers (Val_143F and Val_1162F, Table S2) located at the start and middle of the 18S rRNA gene that were specific to the Vannella genus based on the multiple sequence alignment in Data S1. Two PCRs were performed using either Val_143F or Val_1162F as the forward primer, combined with the reverse primer Val_1875R. Note that this was the same reverse primer used in the qPCR assay and was specific to zOTU2 (and three closely related Vannella sequences). PCR was performed in a total volume of 50 μL containing 2.5 μL of each forward and reverse primer at 10 μM, 25 μL Q5 Hot Start High‐Fidelity 2x Master Mix (New England Biolabs), 16 μL water and 4 μL of 10 ng/μL DNA template. The PCR products visualised on a 2.5% agarose gel and purified with QIAquick PCR Purification Kit (Qiagen, Germany). The resulting amplicons were then sequenced using Sanger sequencing and poor‐quality bases trimmed from the start and end of the sequences. The trimmed sequences were then classified based on BLASTn analysis and aligned to the zOTU2 sequence.

3. Results

3.1. Gross Gill Pathology and Histopathology of NGD

This study aimed to sample rainbow trout with and without gross signs of NGD from a common production environment (i.e., single tank). In total, 206 rainbow trout were sampled, 97 with anatomically normal gills (Figure 1a,b) and 109 showing gross clinical signs of NGD (Figure 1c). The severity of NGD was determined using the gross gill scoring method for the 109 affected fish (Table 1). We successfully sampled fish from all gill score levels. The gross NGD pathologies observed were consistent with those previously described by past NGD studies (Buchmann et al. 2004; Dyková et al. 2010; Quaglio et al. 2016; Vannetti et al. 2023). Grossly, gills displayed thickened filaments and discoloration with multiple nodules on the tips of filaments with extensive mucus deposition (Figure 1c). Histology of NGD cases revealed extensive epithelial hyperplasia and hypertrophy (Figure 1d,f,g). Amoebae could be identified in gill smears of affected tissue (Figure 1e) and along the margins of lesions (Figure 1f,g).

TABLE 1.

Frequency and description of NGD gross gill scores from 206 rainbow trout sampled.

Gill score Sampled fish Gill score description (from Vannetti et al. 2023)
0 97 No lesion
1 15 Scattered focal area with mild proliferation and/or mild discoloration at the filament tips
2 21 ≥ 10% of gills showing mild proliferation and/or discoloration at the filament tips
3 28 ≤ 20% of gills showing proliferation of filament tips and mucus patches
4 21 ≤ 50% of gills showing proliferation of filament tips and/or discoloration, with mucus patches
5 23 > 50% of gills affected by serve lesions
Open in a new tab

3.2. 18S rRNA Amplicon Sequencing

Metagenomic analysis of the swab samples was successful for 203 of the 206 samples. Two samples failed initial PCR amplification, while one sample failed sequencing analysis (< 2000 total reads). The average number of raw paired sequence reads for each swab sample was 23,156 (range 49,087–3709). We observed no significant difference in the number of raw sequence reads from rainbow trout samples with and without NGD (t = 0.6321, df = 222, p = 0.529, Figure 2a). Complete read count statistics are provided as Data S3.

FIGURE 2.

FIGURE 2

Open in a new tab

(a) Raw and vertebrate excluded 18S rRNA amplicon read counts for swab samples derived from rainbow trout with (NGD) and without (Healthy) nodular gill disease. (b) Proportion of zOTUs classified at the domain and phylum level. Only Eukaryote classification at the phylum level is shown. The number of zOTUs assigned to each domain and phylum is illustrated in parathesis.

3.3. Assembly and Exclusion of Host zOTUs

Across the entire dataset a total of 80 zOTUs were identified using the UNOISE algorithm. At the domain level the zOTUs were classified as either Eukaryota (67) or Bacteria (11). Two zOTUs remained unclassified at the domain level (Figure 2b). A total of 67 Eukaryote zOTUs were further classified to the phylum level representing major eukaryotic microbiota lineages (Figure 2b). Figure 2b shows the proportion of unique zOTU (sequences) that were classified to different phyla. Indeed, from the 67 unique zOTUs classified at the phylum level, 12 were classified to Diatomea and 12 to Ciliophora. Therefore, more zOTU sequences were classified to Diatomea and Ciliophora compared to all phyla. All bacterial zOTUs were classified as Proteobacteria at the phylum level (data not shown).

Despite utilising a C3 spacer blocking primer that has been shown to reduce amplification of salmonid host‐derived 18S rRNA (Patchett, Rigby, and Wynne 2024), a considerable proportion of total reads for each sample were taxonomically assigned to ‘Vertebrate’ zOTUs at the phylum level. Indeed, a total of six zOTUs were taxonomically assigned to Vertebrate at the phylum level, with the majority of these further classified to the Teleostei order. By excluding all zOTUs classified as Vertebrate the total number of reads for each sample was reduced significantly (Figure 2a). The average number of reads remaining for the swab samples was 1151 (Figure 2a).

3.4. Non‐Metric Dimensional Scaling

A non‐metric dimensional scaling (NMDS) analysis was performed to examine the relationship between samples in the context of NGD (Figure 3a). Distances between samples were calculated using a Bray‐Curtis distance matrix with a stress of 0.22. Statistical significance was tested using PERMDISP and PERMANOVA analysis. PERMDISP analysis was performed to determine if the dispersions (variance) within the healthy and NGD groups were statistically different. The distance of the healthy and NGD group members to their group centroid is presented in Figure 3b and was subjected to ANOVA. The dispersal of samples within the healthy and NGD groups was not significantly different (df = 1114, F = 0.795, p = 0.374, Figure 3b). Utilising a PERMANOVA we observed a statistically significant difference in the distribution of samples derived from healthy and NGD fish (df = 1115, F = 17.330, p = 0.01, Figure 3a).

FIGURE 3.

FIGURE 3

Open in a new tab

(a) Non‐metric dimensional scaling plot of Bray‐Curtis distance matrix of eukaryotic microbiota between healthy and NGD rainbow trout. (b) Dispersion of microbiota Bray‐Curtis distances within the healthy and NGD.

3.5. Taxonomy Abundance

The abundance of non‐vertebrate classified zOTUs was assessed at the phylum and genus level. The abundance for each zOTU was calculated as the percentage of reads assigned to that zOTU relative to the total reads (non‐rarefied) for that sample. At the phylum level the mean relative abundance was calculated for the healthy and NGD groups (Figure 4). The four most abundant phylum of eukaryotic microorganisms were Amoebozoa, Diatomea, Platyhelminthes and Cillophora. Significant differences in the abundance of four specific phylum were observed between the healthy and NGD groups, most notably the increase in abundance of Amoebozoa within the NGD group (log2FC = 5.95, p adj < 0.001). The abundance of Ciliophora (log2FC = 0.36, p adj < 0.05), Diatomea (log2FC = 0.51, p adj < 0.05) and Euglenozoa (log2FC = 0.32, p adj < 0.05) was all statistically lower in the NGD samples compared to healthy controls.

FIGURE 4.

FIGURE 4

Open in a new tab

Relative abundance of taxa at the phylum level between the healthy and NGD‐affected rainbow trout. Note zOTUs classified as ‘Vertebrate’ were excluded from this analysis.

The relative abundance at the genus level was assessed for individual animals from both the healthy and NGD groups (Figure 5). The most abundant eukaryote genera were Vannella (phylum—Amoebozoa), Gyrodactylidae (phylum Platyhelminthes) and Cymbella (phylum—Diatomea). We also identified several other genera that have previously been shown to colonise freshwater salmonids, including ciliates (Peritrichia, Vorticella, Ichthyobodo), flagellates (Cercomonas, kinetoplastids Neobodo) and fungi (Cladosporium). Several algae taxa were also identified, including diatoms (Achnanthidium) and green algae (Chlorophyceae). Statistical analysis demonstrated that the abundance of only two genera was significantly different between the healthy and NGD groups. This includes Vannella (log2FC = 6.05, p adj < 0.001) and Cymbella (log2FC = 0.85, p adj = 0.038). The relative abundance of Vannella sp. was further examined in the context of gill score and was found to be significantly more abundant in NGD samples with high gill score (df = 5, 218, F = 23.87, p < 0.0001, Figure 6). Tukey's multiple comparison test showed that the Vannella sp. abundance was significantly different between the gill score extremes (0 and 5). The abundance of Vannella sp. was not significantly different between gill scores 1 and 4.

FIGURE 5.

FIGURE 5

Open in a new tab

Relative abundance of non‐vertebrate zOTUs classified to the genus level from rainbow trout gill swabs with different severity of NGD. Each stacked bar represents a single swab sample.

FIGURE 6.

FIGURE 6

Open in a new tab

Relative abundance of zOTU2 classified as Vannella sp. for rainbow trout with different severities of NGD illustrated as box and whisker plot with individual values shown. One‐way ANOVA and Tukey's post hoc test were used to compare the relative abundance between gill scores. Significant differences between gill scores are illustrated by compact letter display (in red).

3.6. Quantitative PCR

A specific quantitative Taqman PCR assay was designed and validated to amplify the zOTU sequence (zOTU2) which was classified as Vannella sp. (Table 2). The assay was deemed specific based on no complete matches to existing reference sequences in BLASTn database, three base‐pair differences in the probe sequence compared to the closest amoeba strain (Figure 7) and no amplification of non‐target amoeba DNA (Table S3). A six‐point linear standard curve was generated with an R 2 close to 1 and slope that establishes the assay had a 93.8% amplification efficiency (Figure S1). The limit of detection of the assay was determined to be greater than 4.59 copies per PCR. The assay reproducibility expressed as coefficient of variation (CV) of log10 copies was 8.2% for intra‐assay variance and 4.7% for inter‐assay variance.

TABLE 2.

Summary of validation metrics for the Taqman qPCR assay specific to the Vannella sp. (zOTU2).

qPCR validation metric
Amplification efficiency (%) 93.8
Standard curve R 2 0.99
Limit of detection (Ct mean ± SD) 37.72 ± 0.4
Limit of detection (18S rRNA copies/PCR) 4.59
Log10 copy intra‐assay variance (mean CV %) 8.2
Log10 copy inter‐assay variance (mean CV %) 4.7
Open in a new tab

FIGURE 7.

FIGURE 7

Open in a new tab

Sequence alignments of amoeba 18S rRNA gene fragments with zOTU2 set as the reference sequence highlighted in yellow. The position of the qPCR assay forward and reverser primers are shown in green and the probe in red. Different base‐pairs to zOTU2 in the non‐reference sequences are in colour. This figure shows the more closely related amoeba strains to zOTU2, which were all Vannella strains. The full alignment consisting of sequences from the families Vannellida and Dactylopodida is available in Data S1.

The validated qPCR assay was then applied to 109 NGD and 97 healthy control swab samples to determine if the abundance of Vannella sp. was different between NGD and control samples. The qPCR assay was performed using the absolute standard curve method which calculates 18 s rRNA copy number. Across the entire data set we compared the 18S copy number determined by qPCR with the relative abundance of the zOTU2 determined by metagenomic analysis. A positive relationship was observed between the two measures (Pearson correlation coefficient = 0.78, p < 0.001, Figure 8A). Furthermore, using the qPCR assay we observed a significant difference in the abundance of Vannella sp. 18s rRNA copy number between the healthy and NGD samples (t = −10.56, df = 193.6, p < 0.001, Figure 8B). Similar to that observed for the metagenomic analysis, there was a significant increase in Vannella sp. abundance with increasing gill score (df = 5, 192, F = 19.14, p < 0.0001, Figure 8C). Tukey's multiple comparison test showed that the Vannella sp. abundance was significantly different between the gill score extremes (0, 1 and 5) but was not significantly different between gill scores 1 and 4.

FIGURE 8.

FIGURE 8

Open in a new tab

(A) Correlation between Vannella sp. (zOTU2) 18S rRNA copy number determined by qPCR with the relative abundance of the zOTU2 determined by metagenomic analysis on the gills of each rainbow trout. (B) Vannella sp. (zOTU2) abundance expressed as 18S copy number for rainbow trout classed with healthy or NGD‐affected gills. (C) Vannella sp. (zOTU2) abundance expressed as 18S rRNA copy number for rainbow trout with different gill scores. One‐way ANOVA and Tukey's post hoc test were used to compare the abundance between gill scores. Significant differences between gill scores are illustrated by compact letter display (in red).

3.7. Sequence Analysis zOTU2

Both our metagenomic analysis and qPCR assay amplify a relatively small region of the Vannella sp. 18S rRNA gene. To provide a more definitive taxonomic classification of this sequence we designed primers to amplify a larger section of the 18S rRNA. A multiple sequence alignment was used to compare the zOTU2 sequence to 50 published 18S rRNA sequences derived from different amoeba taxa. This included 31 sequences classified as Vannella sp. (Data S1). A reverse primer (Val_1875R) specific to the zOTU2 and three closely related Vannella sequences was then designed based on this alignment. Note this was the same reverse primer used in the qPCR assay. Two forward primers which were specific to the Vannella genus were also designed (Val_143F and Val_1162F). Using these two sets of primers (Table S2), two PCR amplicons were generated and sequenced. Blastn analysis showed that these two sequences (named 3_G_F_B12 and 6_I_R_C03_RC) were 99% and 100% identical to the Vannella mustalahtiana, respectively. A multiple sequence alignment was also used to compare 3_G_F_B12 and 6_I_R_C03_RC with 18S rRNA sequence from Vannella mustalahtiana and zOTU2 (Data S2). Together these results show that zOTU2 which was classified as Vannella sp. is most likely a sequence derived from V. mustalahtiana.

4. Discussion

The role that amoebae play in NGD has been the focus of several studies in recent years. Across these studies a diversity of amoebae have been isolated and cultured from the gills of NGD‐affected fish. This includes Acanthamoeba, Hartmannella, Naegleria, Protacanthamoeba, Ripella, Saccamoeba, Cochliopodium, Mycamoeba, Rosculus, Copromyxa, Ptolemeba and Vannella (Brocca et al. 2024; Dyková et al. 2010; Vannetti et al. 2023). While it is possible that some of these species may play a causative role in NGD, it is equally plausible that many of these amoebae are simply bystanders and are just part of the gill's normal micro‐eukaryotic community. An example of this was demonstrated by English, Swords, et al. (2019), English, Tyml, et al. (2019) where a diversity of amoeba species could be isolated and cultured from AGD‐affected gills, but only Neoparamoeba perurans, the proven causative agent of AGD (Crosbie et al. 2012), was able to cause significant AGD pathology following disease challenge trials (English et al. 2021). Culture‐based studies are ultimately biased towards those species that can adapt and proliferate within a laboratory culture environment. It is therefore important to recognise that while culture‐based studies can provide valuable insights concerning the diversity of ‘culturable’ amoebae that may colonise the gills of NGD‐affected fish, their presence within a culture does not necessarily reflect their abundance on the gills nor prove their role as a true pathogenic agent. To overcome this challenge the present study utilised an 18S rRNA metagenomic approach to quantify the diversity and abundance of micro‐eukaryotes (including amoebae) directly on the gills of fish affected by NGD. Our study focused on sampling fish from a single NGD outbreak on a Swiss rainbow trout farm. At the populational level we showed that the gill eukaryotic microbiota was different between NGD and healthy fish. However, it is important to note that the large dispersal within the groups suggests considerable variation in gill eukaryotic microbiota exists between individuals. At the individual level we found that trout with clinical signs of NGD had a significantly higher abundance of a zOTU sequence that was classified as Vannella sp. Furthermore, developing a qPCR assay specific to this zOTU confirmed this sequence was significantly more abundant in trout with severe NGD and provides an important diagnostic tool for future investigations. Finally, we demonstrated that this Vannella sequence is likely derived from Vannella mustalahtiana, a recently described species that was associated with NGD in Northwestern Russia (Kudryavtsev et al. 2022).

The genus Vannella is a highly abundant and diverse group of amoebae commonly associated with aquatic environments (Dyková et al. 2005). As described above Vannella species have been isolated from NGD‐affected fish from Germany (Dyková et al. 2010), Switzerland (Vannetti et al. 2023), Czech Republic (Dykova and Tyml 2016) and Northwestern Russia (Kudryavtsev et al. 2022). Furthermore, Vannella spp. have also been isolated from cases of amoebic gill disease in marine cultured Atlantic salmon (English, Tyml, et al. 2019). Taken together, these studies suggest that the presence of Vannella may be associated with clinical NGD in at least some cases. However, Vannella was also present in many of the healthy trout. In the instance Vannella does play a causative role, its presence in healthy gills (gross pathology score of 0) may indicate early stages of disease (pre‐clinical), resistant hosts or absence of abiotic or biotic factor promoting Vannella proliferation. Irrespective of these hypotheses, it is important to acknowledge that the data presented here does not provide adequate evidence to conclude that Vannella is a causative agent of NGD. While it is possible that Vannella may play a causative role in NGD, it is also plausible that Vannella may act as a commensal or secondary pathogen that colonise previously compromised gills, as described in AGD (English, Tyml, et al. 2019).

This study describes the design and analytical validation of a quantitative PCR assay for detection and quantification of a sequence classified as V. mustalahtiana. The assay was shown to be highly specific when tested against genomic DNA from several ‘non‐target’ amoebae. When applied to the same DNA samples used for the 18S metagenomic analysis, the assay confirmed the increased abundance of V. mustalahtiana with increasing severity of NGD gross pathology. This assay provides a useful tool for investigating the role of V. mustalahtiana directly on the gills of NGD cases across different farming regions. Specifically, it will be important to apply this assay across multiple disease outbreaks from different farming regions to determine if NGD cases are always associated with the presence of and an increased abundance of V. mustalahtiana. Furthermore, quantifying the abundance of V. mustalahtiana in healthy and diseased rainbow trout in association with other abiotic factors within the production environment will provide valuable insights into the role of this potential pathogen.

While Amoebozoa was the only phylum with higher abundance in diseased gills, this coincided with the reduction of several other components of micro‐eukaryotic community such as Ciliophora, Diatomea and Euglenozoa, highlighting that NGD may relate to community shifts rather than the flux of a single agent. When considering the community at genus‐level, aside from Vannella, Cymbella was the only other genus with significantly different abundance between healthy and NGD trout, with higher abundance in healthy gills. Cymbella, a diverse genus of diatoms (Agardh 2023), has not been previously associated with gill condition, but it is commonly found in marine and freshwater aquaculture production systems with some proposed benefits to wastewater remediation (Mishra and Tiwari 2022; Tikue and Workagegn 2023). It is also important to acknowledge that a proportion of gill hyperplasia could be caused by other eukaryotic parasites and could explain some level of variation seen between the Vannella abundance and gill score correlation. Indeed, we observed a number of genera that contain parasitic members including ciliates (Ichthyobodo and Peritrichia), monogenean ectoparasite (Gyrodactylidae) and water moulds (Peronosporomycetes).

Unlike 16S rRNA sequencing of bacterial communities which is increasingly applied to study bacterial dysbiosis in aquaculture (Reid et al. 2017; Slinger, Adams, and Wynne 2020; Zamparo et al. 2024), 18S rRNA metagenomic analysis is more challenging due to the preferential amplification of the host 18S rRNA. Our previous study has shown that including a salmonid‐specific C3 spacer blocking primer can reduce host amplification and allow parasite communities to be observed (Patchett, Rigby, and Wynne 2024). In the current study we show that even with the blocking primer the vast majority of sequence reads are derived from the host. Nevertheless, despite the over‐representation of host reads our approach was still capable of identifying several micro‐eukaryotes that were colonising the gills of rainbow trout, some of which appeared bona fide parasitic agents. We suggest that in its current form, 18S rRNA metagenomics can be a useful tool for profiling parasite diversity, and there is merit in repeating this method on future NGD outbreaks throughout different farming regions to test the previously proposed hypothesis that NGD has a multi‐amoeba aetiology (Dykova and Tyml 2016; English and Lima 2020). Metagenomic analysis of 16S rRNA, as recently done during a gill disease outbreak in Italian farmed rainbow trout, could also be performed concurrently with 18S NGS to begin to elucidate possible links between NGD and bacteria gill disease (BGD) in freshwater salmonids (Zamparo et al. 2024).

Due to their intimate contact with the environment, gills are exposed to a diverse range of organisms, some of which may be pathogenic or capable of causing injury under certain suboptimal conditions. Complex gill disease (CGD) is a term used to describe a diverse range of clinical gill presentations within marine salmonid aquaculture that may be caused by environmental insults (including zooplankton exposure or high suspended solids), pathogens (such as parasites, bacteria and viruses) and farm management practices (Herrero et al. 2018). Given this complexity, we believe that metagenomic methods (such as described here) are well‐suited additional tools for investigating unknown causes of gill disease in aquaculture systems. One downside however with metagenomic analysis of the gills is that the spatial context of pathogens associated with lesions is lost. For this reason, histology and in situ hybridisation (ISH) remain an important part of the tool kit when investigating the role, a novel agent may play in disease. In the case of NGD, in situ hybridisation was used to show several Naegleria trophozoites attached to hyperplastic gill epithelium from farmed NGD‐affected trout in Germany (Dyková et al. 2010). Developing and evaluating an ISH assay for the zOTU2 Vannella sp. sequence described here was beyond the scope of our study, but we strongly recommend these studies be performed in the future.

5. Conclusion

Using 18S rRNA metagenomic analysis and species‐specific qPCR, we demonstrated a significant increase in Vannella mustalahtiana abundance in NGD‐affected rainbow trout compared to healthy controls on one farm in Switzerland. We also observed that Vannella mustalahtiana increased with severity of gill lesions suggesting this species may be associated with clinical NGD in at least some cases.

Author Contributions

James W. Wynne: conceptualization, methodology, investigation, funding acquisition, writing – original draft. Chloe English: conceptualization, investigation, funding acquisition, writing – original draft, methodology. Stefania M. Vannetti: methodology, writing – review and editing, investigation. Megan Rigby: investigation, writing – review and editing. Petra R. Quezada‐Rodriguez: investigation, writing – review and editing. Ralph Knüsel: project administration, resources. Christine Huynh: project administration, resources. Heike Schmidt‐Posthaus: project administration, supervision, funding acquisition, writing – original draft, writing – review and editing, investigation.

Ethics Statement

All fish used in this study were approved for sampling by the ethical commission (permission number BE102/2022).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1.

JFD-48-e14061-s002.aln (212.9KB, aln)

Data S2.

JFD-48-e14061-s004.aln (12KB, aln)

Data S3.

JFD-48-e14061-s003.csv (5.7KB, csv)

Data S4.

JFD-48-e14061-s001.docx (147.5KB, docx)

Acknowledgements

We would like to thank the Swiss fish farmers for their sponsorship, access to their facility and help during sampling. We would also like to thank the FIWI Team for helping with sample preparation and Dr. Richard Taylor for his valuable suggestions to the manuscript.

Funding: The study was in part funded by the Swiss Food Safety and Veterinary Office, grant number 714001610, the Swiss fish farmers and the European Association of Fish Pathologists Small Grant Scheme provided to C.E. Funding was also provided to J.W.W. for the CSIRO Julius Career Award.

James W. Wynne and Chloe English contributed equally to this manuscript.

Contributor Information

James W. Wynne, Email: James.Wynne@csiro.au.

Chloe J. English, Email: chloe.english@uq.edu.au.

Data Availability Statement

The raw data that supports the findings of this study is available from the corresponding authors on request. Sequence data is deposited in the Sequence Read Archive under BioProjectID: PRJNA1161435.

References

  1. Agardh, C. A. 2023. “GBIF Backbone Taxonomy, Checklist Dataset.” 10.15468/39omei. [DOI]
  2. Amaral‐Zettler, L. A. , McCliment E. A., Ducklow H. W., and Huse S. M.. 2009. “A Method for Studying Protistan Diversity Using Massively Parallel Sequencing of V9 Hypervariable Regions of Small‐Subunit Ribosomal RNA Genes.” PLoS One 4, no. 7: e6372. 10.1371/journal.pone.0006372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Antychowicz, J. 2007. “Study on Rainbow Trout Nodular Gill Disease Detected in Poland.” Bulletin of the Veterinary Institute in Pulawy 51: 547–551. [Google Scholar]
  4. Bridle, A. R. , Crosbie P. B. B., Cadoret K., and Nowak B. F.. 2010. “Rapid Detection and Quantification of Neoparamoeba perurans in the Marine Environment.” Aquaculture 309, no. 1–4: 56–61. 10.1016/j.aquaculture.2010.09.018. [DOI] [Google Scholar]
  5. Brocca, G. , Truant A., Peckova H., et al. 2024. “Identification of New Amoebae Strains in Rainbow Trout ( Oncorhynchus mykiss , Walbaum) Farms Affected by Nodular Gill Disease (NGD) in Northeastern Italy.” Journal of Fish Diseases 47, no. 6: e13933. 10.1111/jfd.13933. [DOI] [PubMed] [Google Scholar]
  6. Buchmann, K. , Nielsen T., Sigh J., and Bresciani J.. 2004. “Amoebic Gill Infections of Rainbow Trout in Freshwater Ponds.” Bulletin of the European Association of Fish Pathologists 24, no. 2: 87–91. [Google Scholar]
  7. Caporaso, J. G. , Kuczynski J., Stombaugh J., et al. 2010. “QIIME Allows Analysis of High‐Throughput Community Sequencing Data.” Nature Methods 7, no. 5: 335–336. 10.1038/nmeth.f.303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cocco, A. , Toson M., Perolo A., et al. 2023. “Nodular Gill Disease in Northeastern Italy: An Investigation on the Prevalence of the Disease and the Risks of Introduction in Rainbow Trout Farms.” Journal of Fish Diseases 46, no. 10: 1021–1028. 10.1111/jfd.13821. [DOI] [PubMed] [Google Scholar]
  9. Crosbie, P. B. , Bridle A. R., Cadoret K., and Nowak B. F.. 2012. “In Vitro Cultured Neoparamoeba perurans Causes Amoebic Gill Disease in Atlantic Salmon and Fulfils Koch's Postulates.” International Journal for Parasitology 42, no. 5: 511–515. 10.1016/j.ijpara.2012.04.002. [DOI] [PubMed] [Google Scholar]
  10. Dyková, I. , Bohácová L., Fiala I., Machácková B., Pecková H., and Dvoráková H.. 2005. “Amoebae of the Genera Bovee, 1965 and Page, 1969 Isolated From Fish and Their Phylogeny Inferred From SSU rRNA Gene and ITS Sequences.” European Journal of Protistology 41, no. 3: 219–230. 10.1016/j.ejop.2005.05.004. [DOI] [Google Scholar]
  11. Dykova, I. , and Kostka M.. 2013. Illustrated Guide to Culture Collection of Free‐Living Amoebae. Vol. 363. Prague, Czechia: Academia. [Google Scholar]
  12. Dyková, I. , Kostka M., Wortberg F., Nardy E., and Pecková H.. 2010. “New Data on Aetiology of Nodular Gill Disease in Rainbow Trout.” Folia Parasitologica 57, no. 3: 157–163. 10.14411/fp.2010.021. [DOI] [PubMed] [Google Scholar]
  13. Dykova, I. , and Tyml T.. 2016. “Testate Amoeba Rhogostoma minus Belar, 1921, Associated With Nodular Gill Disease of Rainbow Trout, Oncorhynchus mykiss (Walbaum).” Journal of Fish Diseases 39, no. 5: 539–546. 10.1111/jfd.12384. [DOI] [PubMed] [Google Scholar]
  14. Edgar, R. C. 2010. “Search and Clustering Orders of Magnitude Faster Than BLAST.” Bioinformatics 26, no. 19: 2460–2461. 10.1093/bioinformatics/btq461. [DOI] [PubMed] [Google Scholar]
  15. English, C. J. , Botwright N. A., Adams M. B., et al. 2021. “Immersion Challenge of Naive Atlantic Salmon With Cultured Nolandella sp. and Pseudoparamoeba sp. Did Not Increase the Severity of Neoparamoeba perurans‐Induced Amoebic Gill Disease (AGD).” Journal of Fish Diseases 44, no. 2: 149–160. 10.1111/jfd.13319. [DOI] [PubMed] [Google Scholar]
  16. English, C. J. , and Lima P. C.. 2020. “Defining the Aetiology of Amoebic Diseases of Aquatic Animals: Trends, Hurdles and Best Practices.” Diseases of Aquatic Organisms 142: 125–143. 10.3354/dao03537. [DOI] [PubMed] [Google Scholar]
  17. English, C. J. , Swords F., Downes J. K., et al. 2019. “Prevalence of Six Amoeba Species Colonising the Gills of Farmed Atlantic Salmon With Amoebic Gill Disease (AGD) Using qPCR.” Aquaculture Environment Interactions 11: 405–415. 10.3354/aei00325. [DOI] [Google Scholar]
  18. English, C. J. , Tyml T., Botwright N. A., et al. 2019. “A Diversity of Amoebae Colonise the Gills of Farmed Atlantic Salmon ( Salmo salar ) With Amoebic Gill Disease (AGD).” European Journal of Protistology 67: 27–45. 10.1016/j.ejop.2018.10.003. [DOI] [PubMed] [Google Scholar]
  19. Herrero, A. , Thompson K. D., Ashby A., Rodger H. D., and Dagleish M. P.. 2018. “Complex Gill Disease: An Emerging Syndrome in Farmed Atlantic Salmon ( Salmo salar L.).” Journal of Comparative Pathology 163: 23–28. 10.1016/j.jcpa.2018.07.004. [DOI] [PubMed] [Google Scholar]
  20. Jensen, H. M. , Karami A. M., Mathiessen H., Al‐Jubury A., Kania P. W., and Buchmann K.. 2020. “Gill Amoebae From Freshwater Rainbow Trout ( Oncorhynchus mykiss ): In Vitro Evaluation of Antiparasitic Compounds Against Vannella sp.” Journal of Fish Diseases 43, no. 6: 665–672. 10.1111/jfd.13162. [DOI] [PubMed] [Google Scholar]
  21. Kudryavtsev, A. , Parshukov A., Kondakova E., and Volkova E.. 2022. “ Vannella mustalahtiana sp. Nov. (Amoebozoa, Vannellida) and Rainbow Trout Nodular Gill Disease (NGD) in Russia.” Diseases of Aquatic Organisms 148: 29–41. 10.3354/dao03641. [DOI] [PubMed] [Google Scholar]
  22. Magoc, T. , and Salzberg S. L.. 2011. “FLASH: Fast Length Adjustment of Short Reads to Improve Genome Assemblies.” Bioinformatics 27, no. 21: 2957–2963. 10.1093/bioinformatics/btr507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. McMurdie, P. J. , and Holmes S.. 2013. “Phyloseq: An R Package for Reproducible Interactive Analysis and Graphics of Microbiome Census Data.” PLoS One 8, no. 4: e61217. 10.1371/journal.pone.0061217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mishra, B. , and Tiwari A.. 2022. “Chapter 8—Sustainable Aquaculture Wastewater Remediation Through Diatom and Biomass Valorization.” In Biomass, Biofuels, Biochemicals, edited by Varjani S., Pandey A., Taherzadeh M., Ngo H. H., and Tyagi R. D.. Amsterdam, Netherlands: Elsevier. [Google Scholar]
  25. Oldham, T. , Rodger H., and Nowak B. F.. 2016. “Incidence and Distribution of Amoebic Gill Disease (AGD)—An Epidemiological Review.” Aquaculture 457: 35–42. 10.1016/j.aquaculture.2016.02.013. [DOI] [Google Scholar]
  26. Padros, F. , and Constenla M.. 2021. “Diseases Caused by Amoebae in Fish: An Overview.” Animals 11, no. 4: 991. 10.3390/ani11040991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Patchett, A. L. , Rigby M. L., and Wynne J. W.. 2024. “Improved 18S rDNA Profiling of Parasite Communities in Salmonid Tissues Using a Host Blocking Primer.” Parasitology Research 123, no. 2: 124. 10.1007/s00436-024-08136-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Perolo, A. , Gustinelli A., Fioravanti M. L., et al. 2019. “Occurrence of Nodular Gill Disease in Farmed Brown Trout ( Salmo trutta L.).” Journal of Fish Diseases 42, no. 9: 1315–1320. 10.1111/jfd.13027. [DOI] [PubMed] [Google Scholar]
  29. Quaglio, F. , Perolo A., Bronzatti P., et al. 2016. “Nodular Gill Disease in Farmed Rainbow Trout in Italy.” Journal of Fish Diseases 39, no. 9: 1139–1142. 10.1111/jfd.12446. [DOI] [PubMed] [Google Scholar]
  30. Reid, K. M. , Patel S., Robinson A. J., et al. 2017. “Salmonid Alphavirus Infection Causes Skin Dysbiosis in Atlantic Salmon ( Salmo salar L.) Post‐Smolts.” PLoS One 12, no. 3: e0172856. 10.1371/journal.pone.0172856. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Rognes, T. , Flouri T., Nichols B., Quince C., and Mahe F.. 2016. “VSEARCH: A Versatile Open Source Tool for Metagenomics.” PeerJ 4: e2584. 10.7717/peerj.2584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Schloss, P. D. , Westcott S. L., Ryabin T., et al. 2009. “Introducing Mothur: Open‐Source, Platform‐Independent, Community‐Supported Software for Describing and Comparing Microbial Communities.” Applied and Environmental Microbiology 75, no. 23: 7537–7541. 10.1128/Aem.01541-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Slinger, J. , Adams M. B., and Wynne J. W.. 2020. “Bacteriomic Profiling of Branchial Lesions Induced by Neoparamoeba perurans Challenge Reveals Commensal Dysbiosis and an Association With Tenacibaculum dicentrarchi in AGD‐Affected Atlantic Salmon ( Salmo salar L.).” Microorganisms 8, no. 8: 1189. 10.3390/microorganisms8081189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Smirnov, A. V. , Kulishkin N. S., Surkova A. A., et al. 2020. “Molecular Phylogeny of Paradermamoeba valamo (Amoebozoa, Discosea, Dermamoebida).” Protistology 14, no. 4: 219–226. [Google Scholar]
  35. Speare, D. J. 1999. “Nodular Gill Disease (Amoebic Gill Infestation) in Arctic Char.” Journal of Comparative Pathology 121, no. 3: 277–282. 10.1053/jcpa.1999.0317. [DOI] [PubMed] [Google Scholar]
  36. Stoeck, T. , Bass D., Nebel M., et al. 2010. “Multiple Marker Parallel Tag Environmental DNA Sequencing Reveals a Highly Complex Eukaryotic Community in Marine Anoxic Water.” Molecular Ecology 19, no. Suppl 1: 21–31. 10.1111/j.1365-294X.2009.04480.x. [DOI] [PubMed] [Google Scholar]
  37. Tikue, G. T. , and Workagegn B. K.. 2023. “Plankton Composition and Abundance in Semi Intensive Aquaculture Pond and Their Preference by Nile Tilapia.” Aquaculture Studies 23, no. 2: 1032. [Google Scholar]
  38. Tubbs, L. , Wybourne B. A., and Lumsden J. S.. 2010. “Nodular Gill Disease Causing Proliferative Branchitis and Mortality in Chinook Salmon ( Oncorhynchus tshawytscha ).” New Zealand Veterinary Journal 58, no. 1: 59–61. 10.1080/00480169.2010.65061. [DOI] [PubMed] [Google Scholar]
  39. Vannetti, S. M. , Wynne J. W., English C., et al. 2023. “Amoeba Species Colonizing the Gills of Rainbow Trout ( Oncorhynchus mykiss ) in Swiss Aquaculture.” Journal of Fish Diseases 46, no. 9: 987–999. 10.1111/jfd.13819. [DOI] [PubMed] [Google Scholar]
  40. Wickham, H. 2009. ggplot2: Elegant Graphics for Data Analysis, 1–212. New York, NY: Taylor & Francis. 10.1007/978-0-387-98141-3. [DOI] [Google Scholar]
  41. Zamparo, S. , Orioles M., Brocca G., et al. 2024. “Novel Insights on Microbiome Dynamics During a Gill Disease Outbreak in Farmed Rainbow Trout ( Oncorhynchus mykiss ).” Scientific Reports 14, no. 1: 17791. 10.1038/s41598-024-68287-w. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1.

JFD-48-e14061-s002.aln (212.9KB, aln)

Data S2.

JFD-48-e14061-s004.aln (12KB, aln)

Data S3.

JFD-48-e14061-s003.csv (5.7KB, csv)

Data S4.

JFD-48-e14061-s001.docx (147.5KB, docx)

Data Availability Statement

The raw data that supports the findings of this study is available from the corresponding authors on request. Sequence data is deposited in the Sequence Read Archive under BioProjectID: PRJNA1161435.

Articles from Journal of Fish Diseases are provided here courtesy of Wiley

RESOURCES