PLOS ONE RESEARCH ARTICLE Development and diagnostic validation of a one-step multiplex RT-PCR assay as a rapid method to detect and identify Nervous Necrosis Virus (NNV) and its variants circulating in the Mediterranean Francesca Errani1☯, Enrico Volpe1☯, Enrique Riera-Ferrer2, Monica Caffara ID3, a1111111111 Francesc Padrós2, Andrea Gustinelli3, Marialetizia Fioravanti3, Sara Ciulli ID1* a1111111111 1 Department of Veterinary Medical Sciences, Alma Mater Studiorum, University of Bologna, Cesenatico a1111111111 (FC), Italy, 2 Facultat de Veterinària, Universitat Autònoma de Barcelona, Barcelona, Spain, 3 Department of a1111111111 Veterinary Medical Sciences, Alma Mater Studiorum, University of Bologna, Ozzano dell’Emilia (BO), Italy a1111111111 ☯ These authors contributed equally to this work. * sara.ciulli@unibo.it OPEN ACCESS Abstract Citation: Errani F, Volpe E, Riera-Ferrer E, Caffara M, Padrós F, Gustinelli A, et al. (2022) Nervous Necrosis Virus (NNV) represents one of the most threatening pathogens for Medi- Development and diagnostic validation of a one- terranean aquaculture. Several NNV strains are currently co-circulating in the Mediterra- step multiplex RT-PCR assay as a rapid method to nean Basin with a high prevalence of the RGNNV genotype and the RGNNV/SJNNV detect and identify Nervous Necrosis Virus (NNV) and its variants circulating in the Mediterranean. reassortant strain and a more limited diffusion of the SJNNV genotype and the SJNNV/ PLoS ONE 17(8): e0273802. https://doi.org/ RGNNV reassortant. In the present study, a one-step multiplex RT-PCR (mRT-PCR) assay 10.1371/journal.pone.0273802 was developed as an easy, cost-effective and rapid diagnostic technique to detect RGNNV Editor: Ahmed S. Abdel-Moneim, Taif University, and the reassortant RGNNV/SJNNV strain and to distinguish them from SJNNV and the SAUDI ARABIA reassortant SJNNV/RGNNV strain in a single RT-PCR reaction. A unique amplification pro- Received: April 5, 2022 file was obtained for each genotype/reassortant enabling their rapid identification from cell Accepted: August 15, 2022 culture lysates or directly from brain tissues of suspected fish. The method’s detection limit varied between 102.3 and 103.4 TCID50 ml-1 depending on viral strains. No cross-reacitivty Published: August 26, 2022 with viruses and bacteria frequently associated with gilthead seabream, European seabass Copyright: © 2022 Errani et al. This is an open and marine environment was observed. The mRT-PCR was shown to be an accurate, rapid access article distributed under the terms of the Creative Commons Attribution License, which and affordable method to support traditional diagnostic techniques in the diagnosis of VNN, permits unrestricted use, distribution, and being able to reduce considerably the time to identify the viral genotype or the involvement reproduction in any medium, provided the original of reassortant strains. author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files including GenBank accession numbers. Funding: This study has received funding from the European Union’s Horizon 2020 research and Introduction innovation programme under Grant Agreement No 727610 (PerformFISH). This publication reflects Nervous necrosis virus (NNV), within the Betanodavirus genus, Nodaviridae family [1], is the the views only of the authors and the European aetiological agent for viral nervous necrosis (VNN), otherwise known as viral encephalopathy Union cannot be held responsible for any use and retinopathy (VER), a neuropathological condition reported in at least 62 marine species [2]. PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 1 / 16 PLOS ONE A new quick method to identify NNV variants which may be made of the information contained Betanodaviruses are small, non-enveloped, single-stranded positive-sense RNA viruses, and therein. The funders had no role in study design, their genome consists of two segments named RNA1 (3.1 kb) and RNA2 (1.4 kb), which data collection and analysis, decision to publish, or encode for the RNA-dependent RNA polymerase (RdRp) and the coat protein, respectively preparation of the manuscript. [1]. Betanodaviruses are currently classified into four different genotypes officially recognised Competing interests: The authors have declared by the ICTV—International Committee on Taxonomy of Viruses: Striped jack nervous necrosis that no competing interests exist. virus (SJNNV), Redspotted grouper nervous necrosis virus (RGNNV), Barfin flounder nervous necrosis virus (BFNNV) and Tiger puffer nervous necrosis virus (TPNNV) [1]. At first, genotyp- ing was based on the analysis of RNA2; however later studies confirmed the same clustering for RNA1 analysis [3–5]. Furthermore, reassortant strains have emerged from the reassort- ment between the RGNNV and the SJNNV genotypes and named RGNNV/SJNNV (contain- ing the RNA1 deriving from the RGNNV-type and the RNA2 originating from the SJNNV- type) and SJNNV/RGNNV (containing the RNA1 deriving from SJNNV-type and the RNA2 originating from RGNNV-type) [5,6]. Nervous necrosis virus infections are causing severe problems, mainly in the Mediterra- nean, Asian and Australian marine aquaculture industry [2]. Historically, the RGNNV strain was considered the most important viral pathogen affecting Mediterranean aquaculture, causing severe mortality outbreaks in European sea- bass (Dicentrarchus labrax) and being widely spread in both farmed and wild aquatic ani- mals [7–10]. However, the unpredictable and tremendous ability to evolve of betanodaviruses has led to the co-circulation of different NNV strains in the Mediterranean Basin. Currently, RGNNV genotype and the reassortant RGNNV/SJNNV are the dominant strains infecting both gilthead seabream (Sparus aurata) and European seabass, posing a severe threat to the Mediterranean marine finfish industry [11–13]. The RGNNV/SJNNV reassortant strain is widespread in the Mediterranean and has been isolated from farmed European sea bass, common sole (Solea solea), Senegalese sole (Solea senegalensis) and gilt- head sea bream. In contrast, SJNNV/RGNNV reassortants have only been isolated from European sea bass in Italy and Croatia [5,6,11,14], while SJNNV was detected in fish reared in Spain and Portugal [15–17]. Several diagnostic techniques are available for NNV detection i.e. electron microscopy, cell culture isolation, immunodetection and molecular methods. However, according to the World Organization for Animal Health (OIE), the ‘Gold Standard’ method to detect NNV is the isola- tion of viral agents in cell culture, followed by immunological or molecular identification [18]. Furthermore, given the wide co-circulation of different betanodaviruses in the same geo- graphic region and considering their different features in pathogenicity and host and tempera- ture tropism, laboratories are expected not only to detect, but also to genotype isolated or detected viruses [19,20]. However, proficiency tests performed by the OIE reference laboratory pointed out a limited capability to properly identify the NNV variants [19]. In particular, in the 2nd VER interlaboratory proficiency test (VER-IPT) less than half of the participants com- pleted the viral identification and only 2 out of 29 laboratories correctly and completely identi- fied all the viruses included in the panel test [19]. In the 3rd VER-IPT an increasing number of participants completed the viral identification, although their number was still too low to be considered satisfactory [19]. The low capacity to identify the viral genome is partially due to the lack of a easy and fast method widely implementable also in non-specialised laboratories. So far, identification of NNV variants is based on amplification, sequencing and phylogenetic analysis of both viral genome RNA molecules. At present, some approaches regarding PCR- based and blot reactions have been developed in order to discriminate between RGNNV and SJNNV genotypes, reflecting the need for an easy and affordable diagnostic tool to identify the different NNV variants circulating in the Mediterranean. However, these assays remain lim- ited to independent detection of RNA1 and RNA2 viral genome segments, rendering them PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 2 / 16 PLOS ONE A new quick method to identify NNV variants unable to distinguish parental RGNNV and SJNNV strains from reassortant RGNNV/SJNNV and SJNNV/RGNNV strains in a single reaction [21–28]. As viral identification is an essential information for the management of the disease in the field [19], the development of an easy and affordable method to genotype betanodaviruses was included upon request of the Mediterranean Marine Fish Farming sector in the H2020 Peformfish Project, an industry-driven project targeting industry defined priorities. In recent years, multiplex PCR assays have gained popularity due to their convenience in terms of cost and time and the ability to simultaneously detect several pathogens or genotyping them [29,30]. In particular, both conventional and real time PCR have successfully been multi- plexed to discriminate genetic variants of fish pathogens [29–31]. In general terms, real time PCR shows higher sensitivity than conventional PCR, further- more it can reduce cross-contamination associated to post-amplification steps and may pro- vide quantitative information [32]. However, real time PCR requests a higher investment in terms of cost and expertise compared to conventional PCR. On the other hand, conventional PCR, despite less performing, can be more affordable on a large scale to diagnostic laboratories including small laboratories supporting aquaculture industry. The aim of this study was to develop and validate a one-step multiplex conventional RT-PCR assay as an easy and affordable diagnostic technique to support Mediterranean fish production detecting NNV and simultaneously discriminating between RGNNV, SJNNV genotypes, and RGNNV/SJNNV and SJNNV/RGNNV reassortant strains. To achieve this, both RNA1 and RNA2 from RGNNV and RNA2 from SJNNV were targeted in a single RT-PCR reaction. Materials and methods Multiplex RT-PCR primer design In order to design primer candidates, two alignments were generated, one for RNA1 and another one for RNA2 viral genome segments. Both alignments included sequences from RGNNV and SJNNV isolates and the reassortant strains (RGNNV/SJNNV and SJNNV/ RGNNV) (S1 and S2 Tables). Sequences were obtained from the GenBank database (https:// www.ncbi.nlm.nih.gov/genbank/) and aligned with Clustal W implemented in the BioEdit software [33]. RGNNV-RNA1 and RGNNV-RNA2 primers were designed based on conserved regions within each genotype’s sequence and diverging from SJNNV-RNA1 and SJNNV-RNA2, respectively. In addition, SJNNV-RNA1 and SJNNV-RNA2 primers were designed based on conserved regions within each genotype’s sequence and diverging from RGNNV-RNA1 and RGNNV-RNA2, respectively. Several primers targeting RGNNV-RNA1, RGNNV-RNA2, SJNNV-RNA1 and SJNNV-RNA2 were designed and combined together and with other several primers already described in literature [5,34]. Primer sets were selected according to their position within each genotype’s RNA sequence, the melting temperature (Tm) and the predicted amplicon size to amplify fragments of differ- ent sizes specific for RGNNV-RNA1, RGNNV-RNA2 and SJNNV-RNA2. Newly designed primers were first checked for specificity using NCBI-Nucleotide blast (https://blast.ncbi.nlm.nih.gov/Blast.cgi) as well as for self-dimers and primer dimers via Thermo Fisher Scientific1 Multiple Primer Analyzer online software (https://www. thermofisher.com/) and then tested with NNV reference strains representing the RGNNV and SJNNV genotypes and the reassortant strains (RGNNV/SJNNV and SJNNV/RGNNV). PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 3 / 16 PLOS ONE A new quick method to identify NNV variants Viruses The optimisation of the multiplex PCR assay was carried out using already characterised NNV strains: It/351/Sb (RGNNV) isolated from D. labrax [35]; 416-Dec17 (RGNNV/SJNNV) iso- lated from S. aurata larvae [11]; 389/I96 (SJNNV/RGNNV) isolated from D. labrax and 484.2.2009 (SJNNV) isolated from S. senegalensis [14,36]. Viruses were cultured on SSN-1 cell monolayers using 25cm2 culture flasks [37]. Infected monolayers were incubated at 20 ± 1˚C (SJNNV and SJNNV/RGNNV) and 25 ± 1˚C (RGNNV and RGNVV/SJNNV) until cytopathic effects were observed. RNA from each NNV strain was extracted from cell lysates using NucleoSpin1 RNA (Macherey-Nagel GmbH & Co., Germany) according to manufacturer’s instructions and stored at -80˚C until use. Optimisation of the multiplex RT-PCR assay First, the individual amplification assays (singleplex) to detect each of the three viral targets (RGNNV-RNA1, RGNNV-RNA2 and SJNNV-RNA2) were conducted under various conditions of primer combination and concentration, and with various annealing temperatures, in order to choose the most suitable primer pair to each target. Then, the PCR primers were multiplexed and PCR conditions, such as primer concentrations, annealing temperature and annealing time, were optimised to amplify only specific targets for each strain. The reaction was tested first with a two step approach using the High Capacity cDNA Reverse Transcription Kit (Applied Biosys- tem, US) for cDNA synthesis and the Taq DNA Polymerase Recombinant (Invitrogen, USA) for target amplification and finally with a one-step approach using the SuperScript™ III One-Step RT-PCR System with Platinum™ Taq DNA polymerase (Invitrogen, USA). Positive and negative controls were run along with all reactions. Finally, a set of three optimal primer pairs was selected to conduct the one-step multiplex RT-PCR assay and further validation (Table 1). Analytical specificity of multiplex RT-PCR The analytical specificity of the multiplex RT-PCR was tested using several viruses and bacteria frequently associated with gilthead seabream, European seabass and marine environment. The assay was performed using total DNA extracted from gilthead sea bream affected by lympho- cystis disease virus (LCDV) [38] and DNA extracted from bacteria known to cause disease in gilthead sea bream and European sea bass or associated with the aquatic environment (Vibrio anguillarum, V. harveyi, V. alginolyticus, V. gigantis, Photobacterium damselae subsp. piscicida and Aeromonas veronii) [39,40]. Furthermore, the assay was performed using RNA extracted from viral isolates of viral hemorrhagic septicemia virus (VHSV), BFNNV and TPNNV and total RNA extracted from uninfected snakehead-fish cell line (SSN-1) [41]. Table 1. Selected primer pairs for the multiplex one-step RT-PCR assay. Primer name Sequences (5’!3’) Position1 Target Amplicon (bp) Reference RG_Fw1_RNA1 GACTCAGATCCAGCGGGAA 246–264a RGNNV-RNA1 169 This study RG_Rev6_RNA1 TCCAACCTCACGGGGTGAT 396–414a RG_SPCF1b_RNA2 CAATCGTCGGCGTAGTAATC 107–126a RGNNV-RNA2 647 This study RG_SPCFRev3b_RNA2 AGGAGGATGGACTTGAAGTC 735–754a SJ_F2_RNA2 ATTACTACCCAGGCGCCAC 691–709b SJNNV-RNA2 350 This study R3 CGAGTCAACACGGGTGAAGA 1022–1041b [30] a The position of primers is based on the sequences of strain SGWak97 (GenBank accession numbers for RNA1 AY324869 and RNA2 AY324870). b The position of primers is based on the sequences of strain SJNNV (GenBank accession number for RNA2 AB056572). https://doi.org/10.1371/journal.pone.0273802.t001 PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 4 / 16 PLOS ONE A new quick method to identify NNV variants Analytical sensitivity of multiplex RT-PCR In order to determine the limit of detection of the developed one-step multiplex RT-PCR method, the viral titre was calculated for each reference strain. Titrations of viruses were per- formed according to the end-point dilution method. Briefly, ten-fold serial dilutions (10−2 to 10−9) of viral solutions were inoculated onto 24h-old SSN-1 cell culture in 96-well plates and incubated at 20 ± 1˚C (SJNNV and SJNNV/RGNNV) and 25 ± 1˚C (RGNNV and RGNNV/ SJNNV). The viral adsorption period was 1h for each strain. Daily readings were carried out up to one-week post-inoculation. The titre was expressed as TCID50 ml-1 and calculated according to the Spearman-Karber method [42]. Limits of detection for the one-step multiplex RT-PCR were established individually for each reference strain using extracted RNA from at least five viral dilutions. RNA extracted from dilutions was also subjected to the real-time RT-PCR method validated by the OIE. Repeatability (intra- and inter-assay variance) To test for precision and robustness of the multiplex RT-PCR, RNA extracted from brains (n = 3) was analysed in triplicates in the same assay (intra-assay variance). RNA samples were subjected to both the real-time RT-PCR [25] and the novel mRT-PCR assay. According to the real time RT-PCR these samples were negative (sample 26/19), weakly positive (sample 156/ 18; Ct-value 29.6), and distinct positive (sample 173/18B; Ct-value 9.9). Furthermore, to test for variation in results between runs (inter-assay variance), the samples used for intra-assay variance were tested as single-preparations on two consecutive days. Diagnostic sensitivity and specificity Seventy-six samples including 50 European seabass and 23 gilthead sea bream brains and 3 Artemia salina (nauplii) samples were used to determine diagnostic sensitivity and specificity of the mRT-PCR (S3 Table). RNA was extracted from brain samples using PureLinkTM RNA mini kit (Thermo Fisher Scientific, USA) according to manufacturer’s instructions and stored at -80˚C until use. All samples were subjected to parallel testing using the one-step multiplex RT-PCR method and the real-time RT-PCR protocol validated by the OIE reference centre for viral encephalop- athy and retinopathy [25]. The one-step multiplex RT-PCR assay was conducted using the optimised protocol. The real-time RT-PCR assay was carried out according to Baud and colleagues [24] using the specific primers oPVP154 (5’TCCAAGCCGGTCCTAGTCAA3’), oPVP155 (5’CACGAA CGTKCGCATCTCGT 3’) and probe (Cy5-CGATCGATCAGCACCTSGTC-BHQ2). Briefly, the reaction mixture contained 1x of Quantitect RT-PCR master mix (Qiagen), 600 nM of each primer, 400 nM of the probe, 2.5 μl of RNA in 12.5 μl total volume. The thermal cycling condi- tions were 30 min at 50˚C, followed by 15 min at 95˚C, and 40 cycles of denaturation/exten- sion for 15 sec at 94˚C and 60 sec at 60˚C. mRT-PCR results were compared to the real-time RT-PCR [25] taken as the “Gold Stan- dard”. Specificity, sensitivity, positive predictive value (probability that a test-positive fish is true-positive), and negative predictive value (probability that a test-negative fish is true-nega- tive) of the mRT-PCR were calculated according to Thrusfield and colleagues [43]. Viral encephalo-retinopathy interlaboratory proficiency test To further test the robustness of the mRT-PCR, 10 freeze-dried samples (NNV-positive n = 5, NNV-negative n = 5), included in the 3rd VER-IPT organised by the OIE reference laboratory for PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 5 / 16 PLOS ONE A new quick method to identify NNV variants Viral Encephalopathy and Retinopathy [19], were tested. Positive samples consisted of different betanodavirus species, namely RGNNV (107.30 TCID50 ml-1), SJNNV (109.55 TCID50 ml-1), and the reassortant strains RGNNV/SJNNV (107.55 and 104.55 TCID50 ml-1) and SJNNV/RGNNV (108.30 TCID50 ml-1), whereas negative samples consisted in either MEM with and without 10% yeast extract or negative rainbow trout (Oncorhynchus mykiss) serum (S4 Table) [19]. Samples were resuspended in 0.5 ml of Leibovitz L-15 medium (Biowest, USA) according to the VER-IPT instructions, and RNA was extracted from 100μl using NucleoSpin1 RNA (Macherey-Nagel GmbH & Co., Germany) according to manufacturer’s instructions and stored at -80˚C until use. Genotyping To assess the accuracy of the genotyping performed via the new developed mRT-PCR a selec- tion of positive samples which reported a range of results in the mRT-PCR assay were sub- jected to conventional genotyping through amplification and sequencing of the RNA1 (n = 5) and the RNA2 (n = 27) viral genome using previously published protocols [5,8]. Briefly, the amplification step was conducted through a one-step RT-PCR assay with primers S6 (5’-ATG GTACGCAAAGGTGATAAGAAA-3’) and S7 (5’-GTTTTCCGAGTCAACACGGGT-3’) targeting the whole coding sequence of RNA2 viral genome using the SuperScript III One-Step RT-PCR System (Invitrogen, Carlsbad, USA). The reaction mixture contained 1x Reaction Mix, 0.8 μM for each primer, 0.3 μl Superscript III/Platinum Taq enzyme mix and 3 μl RNA in 15 μl total volume. The thermal cycling condi- tions were 45˚C for 30 min, 95˚C for 2 min, followed by 40 amplification cycles of 94˚C for 60 sec, 58˚C for 60 sec and 72˚C for 60 sec. A final extension was performed at 72˚C for 7 min. To avoid cross-contaminations, the amplification reactions were all set up with negative controls. PCR products were purified using the Exosap reagent (Affymetrix, Santa Clara, CA) and then sequenced at Bio-Fab Research srl (Rome, Italy). The sequences obtained were corrected manually, aligned and compared with reference strain sequences available in GenBank using Clustal W implemented in the BioEdit software [33]. The phylogenetic analysis was conducted using the maximum likelihood (ML) method, with the general time-reversible (GTR) nucleotide substitution model [44], available in MEGA X software [45]. One thousand bootstrap replicates were performed to assess the robustness of individual nodes, and only values �70% were considered significant. Results Multiplex RT-PCR optimisation A one-step multiplex RT-PCR targeting RGNNV-RNA1, RGNNV-RNA2 and SJNNV-RNA2 was developed to detect and simultaneously identify the presence of RGNNV, SJNNV or one of their reassortant strains (RGNNV/SJNNV, SJNNV/RGNNV). Three primer pairs were selected and multiplexed (Table 1), optimal conditions were selected to obtain 1 to 2 specific bands for each viral strain. The optimised reaction consisted of a reaction mix containing 3 μl RNA, 7.5 μl of 2x Reac- tion Mix, 0.3 μl of Enzyme mix, 10 μM of each primer and DNase-free water to reach the final volume of 15 μl. The optimal thermal cycling conditions were 45˚C for 30 min, 2 min at 94˚C followed by 40 amplification cycles of 30 sec at 95˚C, 30 sec at 60˚C and 30 sec at 72˚C. Post- elongation was performed for 10 min at 72˚C. The reference strains used for mRT-PCR optimisation were correctly identified producing the specific bands for each genotype/reassortant strain. The It/351/Sb strain (RGNNV geno- type) was identified by two amplification bands one at 169 bp targeting the RGNNV-RNA1 and one at 647 bp targeting the RGNNV-RNA2, the 484.2.2009 strain (SJNNV genotype) was PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 6 / 16 PLOS ONE A new quick method to identify NNV variants identified by the 350 bp band targeting the SJNNV-RNA2. Regarding the reassortant strains, the 416-Dec17 strain (RGNNV/SJNNV) produced two specific bands one at 169 bp targeting RGNNV-RNA1 and one at 350 bp targeting SJNNV-RNA2 and the 389/l96 strain (SJNNV/ RGNNV) generated the 647 bp band targeting the RGNNV-RNA2 (Fig 1). Analytical specificity No false-positive signals were observed with cell-culture RNA sample nor with any viral and bac- terial DNA/RNA (LCDV, VHSV, V. anguillarum, V. harveyi, V. alginolyticus, V. gigantis, P. damselae subsp. piscicida, A. veronii). The mRT-PCR assay also excluded the other NNV geno- types (BFNNV and TPNNV). The presence of specific bands in the positive controls has demon- strated a valid test performance. No specific bands were present in the negative control (Fig 2). Analytical sensitivity of multiplex RT-PCR The infectious titres of the produced viral solutions were 108.4, 106.8, 107.3 and 105.2 TCID50 ml-1 for RGNNV, SJNNV, RGNNV/SJNNV and SJNNV/RGNNV, respectively. The analysis of tenfold dilutions of titrated viral solutions showed a sensitivity ranging from 102.3 to 103.4 TCID50 ml-1. The highest sensitivity was referred to the detection of the RGNNV/SJNNV strain: the mRT-PCR assay was able to detect the viral RNA extracted from the fifth tenfold dilution of a solution with a titre of 107.3 TCID50 ml-1 corresponding to an inferred titre of 102.3 TCID50 ml-1 and a Ct value of 33.7 ± 0.77 at the real time RT-PCR. For RGNNV, SJNNV and SJNNV/RGNNV, the mRT-PCR detected the virus into dilutions with inferred titres of 103.4, 102.8 and 103.2 TCID50 ml-1, respec- tively and Ct values of 35.4 ± 0.37, 30.5 ± 0.02 and 24.9 ± 0.11, respectively (Table 2). Intra and inter-assay variability No variation in multiplex RT-PCR results for positive and negative samples within and between runs have been observed. All replicates of both weakly (Ct-value 29.6), and distinct Fig 1. Multiplex RT-PCR. Agarose gel electrophoresis of PCR products amplified from RNA extracted from NNV reference strains. Lane M: 100 bp ladder (Invitrogen, Carlsbad, USA); Lane 1: RGNNV genotype; Lane 2: RGNNV/ SJNNV reassortant strain; Lane 3: SJNNV/RGNNV reassortant strain; Lane 4: SJNNV genotype; Lane 5: No template control; Molecular weight markers (bp) are indicated to the left of the gel. https://doi.org/10.1371/journal.pone.0273802.g001 PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 7 / 16 PLOS ONE A new quick method to identify NNV variants Fig 2. Multiplex RT-PCR specificity test. Agarose gel electrophoresis of PCR products amplified from cell-culture samples and viral and bacterial fish pathogens. A) Lane 1: VHSV; Lane 2: V. harveyi; Lane 3: V. alginolyticus; Lane 4: V. gigantis; Lane 5: RGNNV genotype; Lane 6: RGNNV/SJNNV reassortant strain; Lane 7: SJNNV/RGNNV reassortant strain; Lane 8: SJNNV genotype; Lane 9: No template control; B) Lane 1: BFNNV genotype; Lane 2: TPNNV genotype; Lane 3: RGNNV genotype; Lane 4: RGNNV/SJNNV reassortant strain; Lane 5: SJNNV/ RGNNV reassortant strain; Lane 6: SJNNV genotype; Lane 7: No template control; C) Lane 1: SSN1 cell line; Lane 2: LCDV; Lane 3: V. anguillarum; Lane 4: P. damselae subsp. piscicida; Lane 5: A. veronii; Lane 6: RGNNV genotype; Lane 7: RGNNV/SJNNV reassortant strain; Lane 8: SJNNV/RGNNV reassortant strain; Lane 9: SJNNV genotype; Lane 10: No template control. Lanes M: 100 bp ladder (Invitrogen, Carlsbad, USA). https://doi.org/10.1371/journal.pone.0273802.g002 positive Ct-value 9.9) samples showed 100% positive results when tested in triplicate within the same multiplex RT-PCR run (intra-assay variability; Fig 3) and when tested in multiplex RT-PCRs at two consecutive days (inter-assay variability). All replicates of negative sample were negative in both intra- and inter-assay variability tests. Diagnostic sensitivity and specificity Diagnostic sensitivity and specificity were tested using RNA from 76 tissue samples extracted from European sea bass, gilthead sea bream and Artemia salina (S3 Table). NNV positivity/ negativity was established based on the real-time RT-PCR protocol validated by the OIE refer- ence centre for viral encephalopathy and retinopathy [25] regarded as the ‘Gold Standard’ method [43]. According to this protocol, 34 NNV-positive and 42 NNV-negative samples were obtained. The mRT-PCR accurately detected no NNV-RNA in NNV-negative samples (n = 42), assuming a diagnostic specificity of 100%. Furthermore, all NNV-positive samples (n = 34) produced at least one specific amplification band being identified as positive (diagnostic sensitivity = 100%). PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 8 / 16 PLOS ONE A new quick method to identify NNV variants Table 2. Limit of detection of mRT-PCR determined on tenfold dilution of viral solutions; the titre is expressed as TCID50 ml-1; the real time RT-PCR results are expressed as Ct values ± standard deviation (SD). Dilution RGNNV SJNNV RGNNV/SJNNV SJNNV/RGNNV factor It/351/Sb 484.2.2009 Sa-416-Dec17 389/I96 mRT-PCR Virus Real time mRT-PCR Virus Real time mRT-PCR Virus Real time mRT-PCR Virus Real time titre RT-PCR titre RT-PCR titre RT-PCR titre RT-PCR 10−0 + 108.4 n.d + 106.8 n.d + 107.3 n.d + 105.2 n.d −1 7.4 5.8 6.3 4.2 10 + 10 n.d + 10 + (18.7 ± 0.17) + 10 n.d + 10 + (20.7 ± 0.35) 10−2 + 106.4 + (25.4 ± 0.04) + 104.8 + (23.2 ± 0.28) + 105.3 + (23.2 ± 0.11) + 103.2 + (24.9 ± 0.11) 10−3 + 105.4 + (29.4 ± 0.18) + 103.8 + (26.7 ± 0.20) + 104.3 + (26.7 ± 0.39) - 102.2 + (28.2 ± 0.09) 10−4 + 104.4 + (32.2 ± 0.37) + 102.8 + (30.5 ± 0.02) + 103.3 + (30.0 ± 0.32) - 101.2 + (31.5 ± 0.38) 10−5 + 103.4 + (35.4 ± 0.37) - 101.8 + (33.4 ± 0.02) + 102.3 + (33.7 ± 0.77) n.d n.d. n.d. 10−6 - 102.4 + (38.0 ± 0.04) - 100.8 + (36.0 ± 0.30) - 101.3 + (34.8 ± 0.65) n.d n.d. n.d. −7 1.4 10 - 10 - n.d n.d. n.d. - 100.3 - n.d n.d. n.d. 10−8 - 100.4 - n.d n.d. n.d. - 100.03 - n.d n.d. n.d. n.d. not determined. https://doi.org/10.1371/journal.pone.0273802.t002 Viral encephalo-retinopathy interlaboratory proficiency test The application of the mRT-PCR to the samples from the 3rd VER-IPT enabled to correctly detect their positivity/negativity (Fig 4). Furthermore, all positive samples included in the VER-IPT test covering all genotypes (RGNNV and SJNNV) and reassortant strains (RGNNV/ SJNNV and SJNNV/RGNNV) were correctly genotyped (Fig 4). Genotype identification Analysis conducted with mRT-PCR allowed us to assign the RNA1 of all the positive analysed samples (34/34) to a genotype (RGNNV-type). In contrast, the RNA2 was genotyped in 30 of Fig 3. Multiplex RT-PCR repeatability test. Agarose gel electrophoresis of PCR products amplified from RNA extracted from NNV-positive and NNV-negative samples. Lane M: 100 bp ladder (Invitrogen, Carlsbad, USA); Lane 1–3: NNV-negative sample; Lane 4–6: NNV weakly positive sample; Lane 7–9: NNV distinct positive sample; Lane 10: RGNNV genotype; Lane 11: RGNNV/SJNNV reassortant strain; Lane 12: No template control; Molecular weight markers (bp) are indicated to the left of the gel. https://doi.org/10.1371/journal.pone.0273802.g003 PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 9 / 16 PLOS ONE A new quick method to identify NNV variants Fig 4. Multiplex RT-PCR applied to VER-IPT samples. Agarose gel electrophoresis of PCR products amplified from RNA extracted from ten VER-IPT samples. Lane M: 100 bp ladder (Invitrogen, Carlsbad, USA); Lane 1–4 and 6: NNV- positive samples; Lane 5 and 7–10: NNV-negative samples; Lane 11: No template control; Molecular weight marker (bp) isindicated to the left of the gel. For sample details see S4 Table. https://doi.org/10.1371/journal.pone.0273802.g004 the 34 positive samples. Twenty-four samples belonged to RGNNV genotype showing the two specific bands at 169 and 647 bp, whereas 6 samples were RGNNV/SJNNV reassortant strains amplifying two bands of 169 and 350 bp. Four samples showed positivity to the RNA1 amplifi- cation band common to RGNNV and RGNNV/SJNNV reassortant strain (169 bp) but did not generate the specific band for RGNNV-RNA2 (647 bp) nor SJNNV-RNA2 (350 bp), not allow- ing their differentiation. Conventional genotyping through amplification and sequencing of the RNA1 and RNA2 fragments confirmed mRT-PCR genotype identification. All sequenced RNA1 fragments were attributed to RGNNV-type including parental RGNNV and reassortant strains (Fig 5). Sequencing of an RNA2 fragment from the positive samples confirmed that the strains identified as RGNNV by mRT-PCR had RGNNV-type RNA2, whereas samples identified as reassortants had SJNNV-type RNA2 (Fig 6). Regarding the samples not fully genotyped by the mRT-PCR, all of them were identified as RGNNV genotypes (Fig 6). Sequence analysis in the primer region targeting the RGNNV-RNA2 of three ungenotyped samples showed 1 nucleotide mismatch at the 3’ end of the forward primer (RG_SPCF1b_RNA2). However, this mismatch was present also in 6 out of 19 sequences of sam- ples correctly genotyped by the mRT-PCR (S1 Fig). For one sequence (sample 179/12) it was possible to obtain only the nested PCR product that does not include the forward primer region. Discussion Nervous necrosis virus is one of the most threatening pathogens of Mediterranean aquaculture. Since its emergence in the late 1980s, the virus has rapidly and broadly evolved, leading to the emergence of new viral variants, including two reassortant strains. Furthermore, the commer- cial trade of live aquatic animals has significantly contributed to the diffusion of the NNV and its variants together with maritime transport means and migration of the wild hosts [46]. Currently the Mediterranean Basin contains the majority of the viral variances, including the reassortant strains [2]. PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 10 / 16 PLOS ONE A new quick method to identify NNV variants Fig 5. Phylogenetic tree based on RNA1 sequences. Phylogenetic tree based on RNA1 nucleotide sequences. Reference strains are reported with genotype name and accession number. All sequences of tested samples (bold) including parental RGNNV (dot) and RGNNV/SJNNV reassortant strains (triangle) clustered within the RGNNV group. Bootstrap values>70% are shown. Branch lengths are scaled according to the number of nucleotide substitutions per site. The scale bar is reported. https://doi.org/10.1371/journal.pone.0273802.g005 The increasing variability of NNV strains circulating in aquaculture calls for a continuous upgrade of diagnostic methods. Currently, NNV diagnosis relies on a real-time RT-PCR method validated by the OIE reference laboratory, which provides a quick, sensitive and reliable method to detect NNV being able to amplify and then detect all variants so far described. However, identify- ing the genotypes and reassortant strains is so far based on conventional genotyping that consists in amplifying and sequencing of both viral genome RNA molecules. Only the analysis of both viral genome segments (RNA1 and RNA2) led to reassortant identification. This method is time-con- suming and requires highly specialised staff and equipment, resulting in a limited number of labo- ratories able to perform a complete and correct viral species identification [19]. For this reason, the availability of a quick, cheap and practical method to distinguish NNV variants, including reas- sortant strains, could significantly contribute to improve NNV diagnosis capabilities. In this study, a one-step multiplex RT-PCR targeting RGNNV-RNA1, RGNNV-RNA2 and SJNNV-RNA2 was successfully developed. The method detects the presence and identifies RGNNV genotype and reassortant RGNNV/SJNNV strain and differentiates them from SJNNV genotype and the reassortant SJNNV/RGNNV strain in a single PCR reaction. The method showed optimal specificity without cross-reaction against viruses and bacteria frequently associated with gilthead seabream, European seabass and marine environment such as LCDV, VHSV, V. anguillarum, V. harveyi, V. alginolyticus, V. gigantis, P. damselae subsp. piscicida and A. veronii. Furthermore, the method showed a fairly high sensitivity according to the detection limit reported for other mRT-PCR assays [29,47] and can be applied to confirm VNN field outbreaks. Moreover, the method showed a high identification capacity. Most of the positive samples (30/34) were detected and fully characterised at RNA1 and RNA2 level. On the other hand, few samples (4/34) were detected and partially characterised being identified only at RNA1 level. The lack of amplification of the RGNNV-RNA2 specific band in these samples could be due to a combination of causes: lower efficiency of the primer pair amplifying this fragment compared to the other primer pairs, long size of the product, presence of 1 mismatch in some sequences and low amount of target RNA in some samples. PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 11 / 16 PLOS ONE A new quick method to identify NNV variants Fig 6. Phylogenetic tree based of RNA2 sequences. Phylogenetic tree based on RNA2 nucleotide sequences. Reference strain are reported with genotype name and accession number. Sequences of tested samples (bold) clustered within the RGNNV or SJNNV group according to their parental RGNNV (dot) or RGNNV/SJNNV reassortant (triangle) nature. Bootstrap values>70% are shown. Branch lengths are scaled according to the number of nucleotide substitutions per site. The scale bar is reported. https://doi.org/10.1371/journal.pone.0273802.g006 The simultaneous detection of several viral targets with a multiplex PCR approach is relatively rapid and cost-effective compared to standard virological and other molecular assays. The mRT-PCR assay developed was carried out in a mix reaction volume equivalent to that of a sin- gleplex; in this way, the cost and the time of execution of the test for the simultaneous detection and identification of NNV and its variants circulating in the Mediterranean are equivalent to those necessary to test only one target leading to a considerable saving of time and costs. More- over, the developed assay using a so widespread technology as conventional PCR can found a wide implementation in a large number of laboratories with different tecnological levels. Six finfish species are currently being reared and commercialised in the Mediterranean. Farms in the region commonly operate rearing several species in the same on-growing site, being most of the times gilthead seabream and European seas bass, and most recently meagre (Argyrosomus PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 12 / 16 PLOS ONE A new quick method to identify NNV variants regius), in which has also been detected NNV [23]. Due to the wide range of NNV-susceptible spe- cies, the proposed multiplex RT-PCR test could be potentially helpful to understand inter-species transmissions in farming conditions. In fact, the novel assay can be applied to detect and simulta- neously genotype viral strains causing an outbreak which involves different species. Moreover, the proposed method could be applied sequentially with the already OIE vali- dated RT-qPCR method in case of positivity providing NNV genotype information and saving time in comparison to conventional genotyping approach. The need to discriminate between NNV variants lies in the differences in virulence and pathogenicity of each variant. Furthermore, the developed one-step multiplex RT-PCR is suit- able for genotyping a numerous of samples unlike gene sequencing analyses. Therefore, it could support epidemiological studies on NNV variants circulating in the Mediterranean, expanding the knowledge on this virus with reference to its behaviour and distribution. This approach was successfully applied on a large scale survey to investigate the distribution of NNV genotype in Korean shellfish [28]. In conclusion, the developed multiplex RT-PCR represents an easy, rapid, and affordable method to support NNV detection and identification that could find wide application in sev- eral sectors of the study of the VNN contributing to improve its knowledge and therefore lead- ing to a better disease control. The method can find wide application for confirmatory diagnosis in NNV outbreaks in the Mediterranean increasing laboratories’ capacity to cor- rectly identify the betanodavirus genotype. Supporting information S1 Fig. RNA2 sequence alignment of samples belonging to RGNNV genotype and the primer RG_SPCF1b_RNA2. Primer overlap are shown in bold, mismatches with primer are highlighted in blue. Sequences presenting the mismatch and correctly genotyped by the devel- oped mRT-PCR are in green, whereas sequences presenting the mismatch but ungenotyped by the developed mRT-PCR are in red. (PDF) S1 Table. RNA1 sequences used for primer design. The Table reports the details of the RNA1 sequences used for primer design: isolate name, genotype, GenBank accession number and the host species. (DOCX) S2 Table. RNA2 sequences used for primer design. The Table reports the details of the RNA2 sequences used for primer design: isolate name, genotype, GenBank accession number and the host species. (DOCX) S3 Table. Samples used to validate the multiplex RT-PCR assay. The Table reports the details of the samples used to validate the multiplexRT-PCR assay. (XLSX) S4 Table. VER-IPT samples tested with the multiplex RT-PCR assay. Details of VER-IPT samples. The table has been modified from Toffan and colleagues [19]. (DOCX) S1 Raw images. (PDF) PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 13 / 16 PLOS ONE A new quick method to identify NNV variants Acknowledgments The Authors thank Dr. Anna Toffan (IZSVe, Italy) for providing 389/I96 (SJNNV/RGNNV), 484.2.2009 (SJNNV), BFNNV and TPNNV strains. Author Contributions Conceptualization: Enrico Volpe, Sara Ciulli. Data curation: Sara Ciulli. Formal analysis: Francesca Errani, Enrico Volpe, Sara Ciulli. Funding acquisition: Francesc Padrós, Marialetizia Fioravanti. Investigation: Francesca Errani, Enrico Volpe, Enrique Riera-Ferrer, Monica Caffara, Andrea Gustinelli. Methodology: Enrico Volpe, Enrique Riera-Ferrer, Sara Ciulli. Project administration: Sara Ciulli. Resources: Andrea Gustinelli, Marialetizia Fioravanti. Visualization: Francesca Errani, Enrico Volpe. Writing – original draft: Francesca Errani, Enrique Riera-Ferrer, Sara Ciulli. Writing – review & editing: Francesca Errani, Enrico Volpe, Enrique Riera-Ferrer, Monica Caffara, Francesc Padrós, Andrea Gustinelli, Marialetizia Fioravanti, Sara Ciulli. References 1. Sahul Hameed AS, Ninawe AS, Nakai T, Chi SC, Johnson KL, ICTV Report Consortium. ICTV Virus Taxonomy Profile: Nodaviridae. J Gen Virol. 2019; 100: 3–4. https://doi.org/10.1099/jgv.0.001170 PMID: 30431412 2. Bandı́n I, Souto S. Betanodavirus and VER Disease: A 30-year Research Review. Pathogens. 2020; 9: 106. https://doi.org/10.3390/pathogens9020106 PMID: 32050492 3. Nishizawa T, Mori K -i., Furuhashi M, Nakai T, Furusawa I, Muroga K. Comparison of the coat protein genes of five fish nodaviruses, the causative agents of viral nervous necrosis in marine fish. J Gen Virol. 1995; 76: 1563–1569. https://doi.org/10.1099/0022-1317-76-7-1563 PMID: 9049363 4. Nishizawa T, Furuhashi M, Nagai T, Nakai T, Muroga K. Genomic classification of fish nodaviruses by molecular phylogenetic analysis of the coat protein gene. Appl Environ Microbiol. 1997; 63: 1633–1636. https://doi.org/10.1128/aem.63.4.1633-1636.1997 PMID: 9097459 5. Toffolo V, Negrisolo E, Maltese C, Bovo G, Belvedere P, Colombo L, et al. Phylogeny of betanoda- viruses and molecular evolution of their RNA polymerase and coat proteins. Mol Phylogenet Evol. 2007; 43: 298–308. https://doi.org/10.1016/j.ympev.2006.08.003 PMID: 16990016 6. Olveira JG, Souto S, Dopazo CP, Thiéry R, Barja JL, Bandı́n I. Comparative analysis of both genomic segments of betanodaviruses isolated from epizootic outbreaks in farmed fish species provides evi- dence for genetic reassortment. J Gen Virol. 2009; 90: 2940–2951. https://doi.org/10.1099/vir.0. 013912-0 PMID: 19710256 7. Ciulli S, Galletti E, Grodzki M, Alessi A, Battilani M, Prosperi S. Isolation and Genetic Characterization of Betanodavirus from Wild Marine Fish from the Adriatic Sea. Vet Res Commun. 2007; 31: 221–224. https://doi.org/10.1007/s11259-007-0010-y PMID: 17682880 8. Volpe E, Grodzki M, Panzarin V, Guercio A, Purpari G, Serratore P, et al. Detection and molecular char- acterization of betanodaviruses retrieved from bivalve molluscs. J Fish Dis. 2018; 41: 603–611. https:// doi.org/10.1111/jfd.12759 PMID: 29205402 9. Bitchava K, Chassalevris T, Lampou E, Athanassopoulou F, Economou V, Dovas CI. Occurrence and molecular characterization of betanodaviruses in fish and invertebrates of the Greek territorial waters. J Fish Dis. 2019; 42: 1773–1783. https://doi.org/10.1111/jfd.13098 PMID: 31637759 PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 14 / 16 PLOS ONE A new quick method to identify NNV variants 10. Errani F, Ponti M, Volpe E, Ciulli S. Spatial and seasonal variability of human and fish viruses in mussels inside and offshore of Ravenna’s harbour (Adriatic Sea, Italy). J Appl Microbiol. 2021; 130: 994–1008. https://doi.org/10.1111/jam.14806 PMID: 32743895 11. Volpe E, Gustinelli A, Caffara M, Errani F, Quaglio F, Fioravanti ML, et al. Viral nervous necrosis out- breaks caused by the RGNNV/SJNNV reassortant betanodavirus in gilthead sea bream (Sparus aur- ata) and European sea bass (Dicentrarchus labrax). Aquaculture. 2020; 523: 735155. https://doi.org/ 10.1016/j.aquaculture.2020.735155 12. Padrós F, Caggiano M, Toffan A, Constenla M, Zarza C, Ciulli S. Integrated Management Strategies for Viral Nervous Necrosis (VNN) Disease Control in Marine Fish Farming in the Mediterranean. Patho- gens. 2022; 11: 330. https://doi.org/10.3390/pathogens11030330 PMID: 35335654 13. Chérif N, El Jeni R, Amdouni F, Zreilli S, Djabou H, Khemiri S, et al. Phylogeography of betanodavirus genotypes circulating in Tunisian aquaculture sites, 2012–2019. Dis Aquat Organ. 2021; 146: 53–63. https://doi.org/10.3354/dao03614 PMID: 34553693 14. Panzarin V, Fusaro A, Monne I, Cappellozza E, Patarnello P, Bovo G, et al. Molecular epidemiology and evolutionary dynamics of betanodavirus in southern Europe. Infect Genet Evol. 2012; 12: 63–70. https://doi.org/10.1016/j.meegid.2011.10.007 PMID: 22036789 15. Cutrı́n JM, Dopazo CP, Thiéry R, Leao P, Olveira JG, Barja JL, et al. Emergence of pathogenic betano- daviruses belonging to the SJNNV genogroup in farmed fish species from the Iberian Peninsula. J Fish Dis. 2007; 30: 225–232. https://doi.org/10.1111/j.1365-2761.2007.00803.x PMID: 17394524 16. Garcı́a-Rosado E, Cano I, Martı́n-Antonio B, Labella A, Manchado M, Alonso MC, et al. Co-occurrence of viral and bacterial pathogens in disease outbreaks affecting newly cultured sparid fish. Int Microbiol Off J Span Soc Microbiol. 2007; 10: 193–199. PMID: 18076001 17. Thiéry R, Cozien J, de Boisséson C, Kerbart-Boscher S, Névarez L. Genomic classification of new betanodavirus isolates by phylogenetic analysis of the coat protein gene suggests a low host-fish species specificity. J Gen Virol. 2004; 85: 3079–3087. https://doi.org/10.1099/vir.0.80264-0 PMID: 15448371 18. OIE. Chapter 2.3.12. Viral encephalopathy and retinopathy. Manual of Diagnostic Tests for Aquatic Ani- mal. Paris, France: OIE; 2019. Available: https://www.woah.org/fileadmin/Home/eng/Health_ standards/aahm/current/chapitre_viral_encephalopathy_retinopathy.pdf. 19. Toffan A, Buratin A, Toson M, Abbadi M, Leardini S, Pascoli F. Viral Encephalo-Retinopathy Interlaboratory Proficiency Test (VER-IPT): results from two years’ experience. Bull Eur Fish Pathol. 2021; 41: 59–69. 20. Zrnčić S, Padros F, Mladineo I, Fioravanti M-L, Gustinelli A, Palenzuela O, et al. Bottlenecks in diagnos- tics of Mediterranean fish diseases. Bull Eur Fish Pathol. 2020; 40: 70–80. 21. Dalla Valle L, Zanella L, Patarnello P, Paolucci L, Belvedere P, Colombo L. Development of a sensitive diagnostic assay for fish nervous necrosis virus based on RT-PCR plus nested PCR. J Fish Dis. 2000; 23: 321–327. https://doi.org/10.1046/j.1365-2761.2000.00255.x 22. Dallavalle L, Toffolo V, Lamprecht M, Maltese C, Bovo G, Belvedere P, et al. Development of a sensitive and quantitative diagnostic assay for fish nervous necrosis virus based on two-target real-time PCR. Vet Microbiol. 2005; 110: 167–179. https://doi.org/10.1016/j.vetmic.2005.07.014 PMID: 16146676 23. Lopez-Jimena B, Cherif N, Garcia-Rosado E, Infante C, Cano I, Castro D, et al. A combined RT-PCR and dot-blot hybridization method reveals the coexistence of SJNNV and RGNNV betanodavirus genotypes in wild meagre (Argyrosomus regius): SJNNV and RGNNV nodavirus in wild Argyroso- mus regius. J Appl Microbiol. 2010; 109: 1361–1369. https://doi.org/10.1111/j.1365-2672.2010. 04759.x PMID: 20497277 24. Panzarin V, Patarnello P, Mori A, Rampazzo E, Cappellozza E, Bovo G, et al. Development and valida- tion of a real-time TaqMan PCR assay for the detection of betanodavirus in clinical specimens. Arch Virol. 2010; 155: 1193–1203. https://doi.org/10.1007/s00705-010-0701-5 PMID: 20532929 25. Baud M, Cabon J, Salomoni A, Toffan A, Panzarin V, Bigarré L. First generic one step real-time Taqman RT-PCR targeting the RNA1 of betanodaviruses. J Virol Methods. 2015; 211: 1–7. https://doi.org/10. 1016/j.jviromet.2014.09.016 PMID: 25311184 26. Toubanaki DK, Margaroni M, Karagouni E. Development of a Novel Allele-Specific PCR Method for Rapid Assessment of Nervous Necrosis Virus Genotypes. Curr Microbiol. 2015; 71: 529–539. https:// doi.org/10.1007/s00284-015-0880-0 PMID: 26210900 27. Toubanaki DK, Karagouni E. Genotype-specific real-time PCR combined with high-resolution melting analysis for rapid identification of red-spotted grouper nervous necrosis virus. Arch Virol. 2017; 162: 2315–2328. https://doi.org/10.1007/s00705-017-3375-4 PMID: 28451901 28. Kim YC, Kwon WJ, Kim MS, Kim KI, Min JG, Jeong HD. High prevalence of betanodavirus barfin floun- der nervous necrosis virus as well as red-spotted grouper nervous necrosis virus genotype in shellfish. J Fish Dis. 2018; 41: 233–246. https://doi.org/10.1111/jfd.12702 PMID: 29027679 PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 15 / 16 PLOS ONE A new quick method to identify NNV variants 29. Pinheiro ACAS, Volpe E, Principi D, Prosperi S, Ciulli S. Development of a multiplex RT-PCR assay for simultaneous detection of the major viruses that affect rainbow trout (Oncorhynchus mykiss). Aquac Int. 2016; 24: 115–125. https://doi.org/10.1007/s10499-015-9912-9 30. LaFrentz BR, Garcı́a JC, Shelley JP. Multiplex PCR for genotyping Flavobacterium columnare. J Fish Dis. 2019; 42: 1531–1542. https://doi.org/10.1111/jfd.13068 PMID: 31469439 31. Munir K, Kibenge FSB. Detection of infectious salmon anaemia virus by real-time RT–PCR. J Virol Methods. 2004; 117: 37–47. https://doi.org/10.1016/j.jviromet.2003.11.020 PMID: 15019258 32. Purcell MK, Getchell RG, McClure CA, Garver KA. Quantitative Polymerase Chain Reaction (PCR) for Detection of Aquatic Animal Pathogens in a Diagnostic Laboratory Setting. J Aquat Anim Health. 2011; 23: 148–161. https://doi.org/10.1080/08997659.2011.620217 PMID: 22216714 33. Hall TA. BioEdit: a user-friendly biological sequence alignment editor andanalysis program for Windows 95/98/NT. Nucleic Acids Symp Ser. 1999; 41: 95–98. 34. Nishizawa T, Mori K, Nakai T, Furusawa I, Muroga K. Polymerase chain reaction (PCR) amplification of RNA of striped jack nervous necrosis virus (SJNNV). Dis Aquat Organ. 1994; 18: 103–107. https://doi. org/10.3354/dao018103 35. Ciulli S, Gallardi D, Scagliarini A, Battilani M, Hedrick R, Prosperi S. Temperature-dependency of Beta- nodavirus infection in SSN-1 cell line. Dis Aquat Organ. 2006; 68: 261–265. https://doi.org/10.3354/ dao068261 PMID: 16610592 36. Vendramin N, Toffan A, Mancin M, Cappellozza E, Panzarin V, Bovo G, et al. Comparative pathogenic- ity study of ten different betanodavirus strains in experimentally infected European sea bass, Dicen- trarchus labrax (L.). J Fish Dis. 2014; 37: 371–383. https://doi.org/10.1111/jfd.12117 PMID: 23662921 37. Frerichs GN, Rodger HD, Peric Z. Cell Culture Isolation of Piscine Neuropathy Nodavirus from Juvenile Sea Bass, Dicentrarchus labrax. J Gen Virol. 1996; 77: 2067–2071. https://doi.org/10.1099/0022-1317- 77-9-2067 PMID: 8811004 38. Ciulli S, Pinheiro AC de AS, Volpe E, Moscato M, Jung TS, Galeotti M, et al. Development and applica- tion of a real-time PCR assay for the detection and quantitation of lymphocystis disease virus. J Virol Methods. 2015; 213: 164–173. https://doi.org/10.1016/j.jviromet.2014.11.011 PMID: 25522921 39. Mosca F, Ciulli S, Volpatti D, Romano N, Volpe E, Bulfon C, et al. Defensive response of European sea bass (Dicentrarchus labrax) against Listonella anguillarum or Photobacterium damselae subsp. pisci- cida experimental infection. Vet Immunol Immunopathol. 2014; 162: 83–95. https://doi.org/10.1016/j. vetimm.2014.10.002 PMID: 25454083 40. Volpe E, Mandrioli L, Errani F, Serratore P, Zavatta E, Rigillo A, et al. Evidence of fish and human patho- gens associated with doctor fish (Garra rufa, Heckel, 1843) used for cosmetic treatment. J Fish Dis. 2019; 42: 1637–1644. https://doi.org/10.1111/jfd.13087 PMID: 31578759 41. Frerichs GN, Morgan D, Hart D, Skerrow C, Roberts RJ, Onions DE. Spontaneously productive C-type retrovirus infection of fish cell lines. J Gen Virol. 1991; 72: 2537–2539. https://doi.org/10.1099/0022- 1317-72-10-2537 PMID: 1717644 42. Hierholzer JC, Killington RA. Virus isolation and quantitation. Academic Press Limited. Virology meth- ods manual. Academic Press Limited. London: Mahy BWJ, Kangro HO; 1996. pp. 25–46. 43. Thrusfield M. Veterinary epidemiology. 4th ed. Chichester: John Wiley & Sons, Ltd. 4th ed. Chichester: John Wiley & Sons, Ltd. 2018. 44. Darriba D, Taboada GL, Doallo R, Posada D. jModelTest 2: more models, new heuristics and parallel computing. Nat Methods. 2012; 9: 772–772. https://doi.org/10.1038/nmeth.2109 PMID: 22847109 45. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Battistuzzi FU, editor. Mol Biol Evol. 2018; 35: 1547–1549. https://doi.org/ 10.1093/molbev/msy096 PMID: 29722887 46. Doan QK, Vandeputte M, Chatain B, Morin T, Allal F. Viral encephalopathy and retinopathy in aquacul- ture: a review. J Fish Dis. 2017; 40: 717–742. https://doi.org/10.1111/jfd.12541 PMID: 27633881 47. Yoshinaka T, Yoshimizu M, Ezura Y. Simultaneous Detection of Infectious Hematopoietic Necrosis Virus (IHNV) and Infectious Pancreatic Necrosis Virus (IPNV) by Reverse Transcription (RT)-Polymer- ase Chain Reaction (PCR). Fish Sci. 1998; 64: 650–651. https://doi.org/10.2331/fishsci.64.650 PLOS ONE | https://doi.org/10.1371/journal.pone.0273802 August 26, 2022 16 / 16