1. Introduction
The development of marine aquaculture over the last decades has been accompanied by several sanitary issues of importance that have caused economic losses owing to reduced growth performances, mortalities, or associated treatment costs at various stages of fish development [
1].
The global burden of pathogens on European sea bass (
Dicentrarchus labrax) production units in the Mediterranean Sea has recently been evaluated through the European project MedAid. Tenacibaculosis, vibriosis, and photobacteriosis were the most common bacteriosis reported, whereas the viral encephalopathy and retinopathy induced by the nervous necrosis virus (NNV) remained the main virosis [
2].
If the on-growing stages are carried out in an open environment, reproduction and larval development are performed in hatcheries with high biosecurity levels, including management of water quality, regular sanitary controls, and vaccinations. Larvae are fragile, and their growth is associated with a significant level of mortality, especially during the first few days post-hatching (dph). Despite all the precautions taken, these mortality episodes can sometimes be more intense, with no identification of the exact cause(s) of this excess mortality. Viruses belonging to the
Totiviridae family are known to infect a wide range of hosts, mainly unicellular-like protozoans or fungi [
3,
4]. Five genera have been recognized by the ICTV:
Giardiavirus,
Leishmaniavirus,
Trichomonasvirus,
Totivirus, and
Victorivirus. The first three genera contain viruses infecting protozoa, whereas
Totivirus and
Victorivirus target yeast and fungi [
3,
5]. Numerous other viruses, qualified as “toti-like viruses” belonging to Totiviridae family, have been described in recent decades, but they have not been assigned to a specific genus. Among these, the infectious myonecrosis virus (IMNV), described in shrimp, clusters with some species of
Giardiavirus but could represent a new genus within the family
Totiviridae [
6,
7]. Several unassigned arthropod viruses have also been proposed to form a new genus called
Artivirus [
8], and totivirus-like viruses have also been described in ants, with
Camponotus yamaokai virus being the first
Totiviridae member reported in a hymenopteran [
9]. Piscine myocarditis virus (PMCV), identified as the causative agent of cardiomyopathy syndrome (CMS) and responsible for high losses in salmon farms, notably in Scotland and Ireland, is another unclassified member of the
Totiviridae family [
10,
11,
12,
13]. Recently, several new toti-like viruses have been identified in fish in the context of natural mortality events, notably owing to the generalization of high-throughput sequencing. This was the case for common carp toti-like virus-1 (CCTLV-1) in common carp, bluegill toti-like virus-1 (BGTLV-1) in bluegill,
Cyclopterus lupus toti-like virus (CLuTLV) in lumpsucker fingerlings, or golden shiner toti-like virus-1 (GSTLV-1) in golden shiner
Notemigonus crysoleucas baitfish [
7,
14]. Interestingly, none of the animal toti-like viruses reported in recent years could be assigned to one of the previously described genera of the family
Totiviridae, suggesting the existence of other potentially unknown genera [
4].
Totiviridae are non-enveloped double-stranded RNA viruses with an icosahedral capsid. Their genomes vary from 4600 to more than 8000 nucleotides long, coding at minimum for a capsid protein and an RNA-dependent RNA polymerase (RdRp) and, according to the genus, may contain variable number (2 to 4) of potential additional open reading frames (ORFs) [
15]. Totiviridae translation of capsid and RdRp is characterized by the typical −1, −2, or +1 ribosomal frameshifting described in giardiaviruses, totiviruses, trichomonasviruses, and leishmaniaviruses, respectively; viruses belonging to these genera usually display the capsid–RdRp fusion protein. In contrast, capsid and RdRp are expressed separately after a termination/reinitiation process in Victoriviruses [
16].
In the context of an increase in mortality/morbidity in several batches of a sea bass hatchery in winter 2018, we used a panel of cellular, serological, and molecular methods associated with in vivo experimental infections to characterize a virus isolated from symptomatic larvae. For this virus, which has never been described to date and has a 6818-nucleotide-long RNA genome structurally and phylogenetically assigned to the Totiviridae family, we propose the name sea bass toti-like virus (SBTLV).
2. Materials and Methods
2.1. Biological Samples
2.1.1. Larvae and Juveniles
The first samples of clinically diseased sea bass larvae (3 batches, 22–27 dph), raised at a water temperature ranging from 18.5 °C to 20 °C, were collected in February 2018 (identified as QQ40, QQ41, and QQ42). Additional samples of sea bass or seabream larvae, sea bass eggs, rotifers, Artemia, or isolated internal sea bass organs with a healthy or moribund status were obtained at various times from 2018 to the end of 2019 (
Table 1, samples used for initial diagnosis (
n = 10);
Table S1, samples used for epidemiological monitoring (
n = 86)). Twenty-eight additional batches corresponding to 250 samples were also regularly analyzed from the beginning of 2021 to spring 2022. All samples were stored at a temperature lower than 10 °C until cell culture analysis (with a maximum of 72 h between sampling and analysis) or in an RNA stabilization reagent (RNAlater™, Invitrogen, Waltham, MA, USA) before being processed for RNA analyses.
For cell culture applications, approximately 1 g of larvae was homogenized in a mortar with pestles and sand and then transferred into 9 mL of Eagle’s solution containing antibiotics (200 IU·mL−1 penicillin G, 0.2 mg·mL−1 streptomycin, and 0.2 mg·mL−1 kanamycin). Samples were then centrifuged (2000× g) for 15 min at 5 °C +/− 3 °C and the supernatants were collected. For molecular applications, samples were homogenized in phosphate-buffered saline (PBS, pH 7.4, Merck, Darmstadt, Germany) using FastPrep-24™ 5 (MP Biomedicals™, Seven Hills, NSW, Australia) with a 10% weight/volume ratio.
2.1.2. Genitors
Eggs and sperm or gonads from sea bass or seabream were sampled from various batches of mature fish. In order to improve the level of viral detection and the yield of extraction, samples were cleaned through successive steps of centrifugation and then concentrated. Briefly, 1 to 15 g of tissue was processed by adding 10 to 15 mL of PBS (0.2 M NaCl, 50 mM Tris-HCl, 5 mMCaCl2, 5 mM MgCl2, pH 7.5) per gram of tissue. Samples were frozen/thawed 3 times, centrifuged at 2000× g, and the supernatant mixed with 10 volumes of cold PBS. Samples were then serially centrifuged to remove tissue debris at 1000× g, 3000× g, 5000× g, 8000× g, 10,000× g, and 12,000× g for 5 min at 5 °C +/− 3 °C, discarding pellet and centrifuging the supernatant again for each step. The final supernatant was transferred into an ice bath for 10 min before laying it down on a sucrose cushion at 28% (w/w) and ultracentrifugation at 300,000× g for 2 h at 5 °C +/− 3 °C. The supernatant was discarded and the pellet mixed with 100 to 200 µL of cold PBS before use for nucleic acid extraction. Internal organs (kidney, spleen, heart) were processed in parallel (homogenization with mortar and pestle) without applying the purification/concentration protocol.
2.2. Cell Culture Isolation
For each sample analyzed, 100 µL of supernatant was inoculated onto 24-well-plate cultures of a total of 11 cell lines: bluegill fry (BF-2) [
17], epithelioma papulosum cyprini (EPC) [
18], chinook salmon embryo (CHSE-214) [
19], rainbow trout gonad (RTG-2) [
20], white sturgeon skin (WSSK-1) [
21], koi carp fin (KF-1) [
22], eel kidney-1 (EK-1) [
23], striped snakehead (SSN-1) [
24], sea bass larvae (SBL) [
25] and larvae of baeri sturgeon (LEB, homemade). Inoculated cell cultures were incubated at 14 °C ± 2 °C and checked regularly for the development of a cytopathic effect (CPE). A blind passage was performed after 7 days on the same cell lines. When a CPE was observed, the cell supernatant was filtered and a chloroform test was performed to check for the presence of a lipid-bilayer-enveloped virus, as well as a specific detection for viral RNA using the double-stranded (ds) RNA antibody J2, as described by Louboutin et al. [
26].
2.3. Production of Specific Rabbit Antibodies
Antibodies specific to the isolated virus were produced on specific pathogen-free rabbits immunized by four subcutaneous injections of 1 mL (days 0, 7, 14, and 36) of a 25% glycerol-cushion-purified virus from homogenized infected sea bass larvae (viral amount of around 6 × 1011 copies RNA·mL−1; rabbit n°183) or from clarified positive cell supernatant (viral amount of around 5 × 1012 copies RNA·mL−1; rabbit n°184; Biotem Company, Isère, France). Antibody titers of sera collected on D28, 42, and 56 were assessed using a homemade ELISA coated with shredded and purified infected larvae (Biotem protocol). Rabbit n°183 D42 sera was selected for the immunofluorescence antigen test (IFAT) and rabbit n°183 D56 sera was used for ELISA development.
2.4. Indirect Fluorescent Antibody Test (IFAT)
A specific IFAT was performed in cell cultures with CPEs using a specific polyclonal rabbit antibody n°183 D42 diluted 1:2000 in PBS-T20 (phosphate-buffered saline, 0.02% Tween 20).
2.5. Transmission Electronic Microscopy (TEM)
The isolated virus was propagated in a 25 cm3 flask (Clipmax, TPP, Geneva, Switzerland) on the LEB cell line at 14 °C. At day 2 pi, the medium was removed, infected as well as non-infected cells were washed with 0.1 M pH 7.2 cacodylate buffer, and then a fixation buffer (glutaraldehyde 2%–paraformaldehyde 2%–tampon cacodylate 0.1 M, pH 7.2) was added for 1 h at room temperature (RT). The cells were washed 3 times with cacodylate buffer and stored at 5 °C until processing. TEM was performed using the Merimage platform (Roscoff, France) and examined with a Jeol 1400 transmission electron microscope operating at 80 kV. Images were obtained using a Gatan (Pleasanton, CA, USA) Orius camera.
2.6. Nucleic Acid Extraction, Library Preparation, and High-Throughput Sequencing
RNA was extracted from either 100 to 200 µL larvae homogenates (RR200) or cell culture supernatants (QQ40 1st passage and QQ40 10th passage; see
Table 1) using, respectively, a NucleoSpin virus kit (Macherey-Nagel, Hoerdt, France) or an Adiamag kit (Bio-X Diagnostics, Belgium) coupled with a KingFisher Duo Prime instrument (ThermoFisher Scientific Inc., Worcester, MA, USA) following the manufacturer’s instructions. cDNA libraries were prepared with an Ion Total RNA-Seq Kit (Life Technologies, Carlsbad, CA, USA) following supplier’s instructions. The cDNA libraries were sequenced using an Ion Proton Sequencer and Ion PI Chip v3 (Life Technologies, Carlsbad, CA, USA). The reads were first mapped onto the host’s genome using Bowtie 2 [
27]. Unmapped reads were then assembled with SPades (v3.10.0, option-careful) [
28] and the de novo contigs were BLASTed against the NCBI nr nucleotide or protein databases for identification of viral sequences.
2.7. Phylogenetic Analyses
The predicted ORFs of the contig coding for viral sequences were analyzed with Geneious (Geneious Prime
® 2020.2.4, Boston, MA, USA). To investigate the relationship between the isolated virus and other known viruses, protein sequences coding for RdRp were collected on Genbank and aligned with sequences of piscine totiviruses and other totiviruses or toti-like viruses using MAFFT version 7.221
https://doi.org/10.1093/molbev/mst010 (accessed on 24 February 2023); alignments were cleaned with Trimal (version 1.4.1;
https://doi.org/10.1093/bioinformatics/btp348). Pairwise identity matrices were obtained using BioEdit (version 7.2.5.). Phylogenetic reconstructions were generated using maximum likelihood (ML) after selecting the best-fit model with IQ-TREE 2.1.2 COVID-edition built 30 March 2021 [
29]. The LG + F + R9 substitution model was selected [
30] and the ultrafast bootstrap approximation was used [
31]. A phylogenic analysis with ORF3 SBTLV protein was also performed using the same strategy. The best-fit model according to AIC was WAG + F + G4. The phylogenetic trees were represented and annotated using iTOL [
32].
2.8. PCR Tools for Viral Detection and Quantification
Several pairs of primers were designed along the viral genome to control the consistency of the sequences obtained by RNA high-throughput sequencing or after amplification of fragments and Sanger sequencing or to perform real-time RT-PCR (RT-qPCR, see
Table 2 for details). RNA was extracted using the protocol described in
Section 2.6. The fragments were amplified with a one-step conventional RT-PCR (RT-cPCR) reaction performed with the SuperScript III One-Step RT-PCR System with Platinum Taq High Fidelity (Invitrogen, Waltham, MA, USA) using the following mix: approximately 1 µg of RNA was added to 20 µM of each primer, 1 µL of Taq, 25 µL of reaction mix, and water in a final volume of 50 µL. The RT-cPCRs were conducted in a thermocycler T100 (Biorad, Hercules, CA, USA) with an initial step of 52 °C (30 min), followed by one step at 94 °C (2 min), then 40 cycles of 94 °C (15 s), 55 °C to 62.5 °C (30 s) depending on the hybridization temperature of the primers, and 68 °C for the elongation step (20 s to 60 s depending the fragment length; see
Table 2). The sizes of the PCR products were assessed on agarose gel (2%). 5′ and 3′ rapid amplifications of cDNA end (RACE) PCRs using the 5′/3′ RACE Kit—2nd generation (Roche) were performed on a QQ40 sample to confirm the 5′ and 3′ sequences of the virus genomes following the manufacturer’s instructions. Briefly, a specific primer oPVP612 was used to produce complementary cDNA with a polyadenylated tail necessary for an oligo dT anchor (contained in the kit). Two successive PCRs using oPVP615 and then oPVP616 with the anchor primer were finally performed to obtain a fragment of 257 nt. 3′RACE was performed through the addition of a poly-A tail on viral RNA with DNA polymerase I (
E. coli) (NEB, Hitchin, UK), cDNA synthesis, and a final PCR using the primer oPVP606 (position 6377,
Table 2). All PCR products generated were purified using a NucleoSpin gel and PCR clean-up kit (Macherey-Nagel) and cloned using the TOPO TA cloning kit (Invitrogen). Three clones were selected and sequenced in both directions, and all nucleotide differences were checked visually using VectorNTI v.11.5 software.
An additional set of primers, oPVP559 5′-CCGAGGCTATCAAAGTCAGC-3′ and oPVP560 5′-GCAGTCCATCTCCAACCACT-3′, located at position 3920 to 4029 (fragment length 110 bp) of the virus genome was designed (Primer3plus) for a specific RT-qPCR. Reverse transcription (RT) and amplification were performed using the SuperScript III One-Step RT-qPCR System (Invitrogen, France) using the following mix: 5 µL of extracted RNA was added to 800 nM of each primer, 300 nM of probe, 0.5 μL of Invitrogen RT mix, 12.5 μL of reaction mix (2×), and water in a final volume of 25 μL. To assess the effectiveness of the whole analytical process, a known amount of external exogenous control (RNA bacteriophage MS2) was spiked into the samples before lysis. Nucleic acid extraction and specific RT-qPCR [
33] were performed in parallel with the targeted virus. The RT-PCRs were conducted in a QuantStudio5 (Applied, Cleveland, OH, USA) after an initial step of denaturation at 95 °C (5 min), with an initial step at 50 °C (30 min), followed by one step at 95 °C (15 min), then 40 cycles at 94 °C (15 s) and 60 °C (60 s). Results were expressed as cycle threshold (Ct), and the Ct values obtained for MS2 were systematically reported in predefined control cards with low and high values to validate the analyses. A PCR product of 286 nucleotides long, covering the RT-qPCR zone, and containing the specific T7 sequence, was obtained with the primers oPVP 608 and 609. This product was used as a template for in vitro synthesis of an RNA transcript using T7 RNA polymerase (T7 RiboMAX™ Express Large Scale RNA Production System, Promega, Madison, WI, USA) following the manufacturer’s recommendations. Quantification via spectrophotometry (Nanodrop One, Thermofisher, Waltham, MA, USA) gave a titer of 2 × 10
13 RNA copies.µL
−1. Aliquots were stored at −80 °C and diluted in 10-fold dilutions for standard curve use and absolute quantification of the viral amount in samples (expressed as number of copies·mg tissue
−1) and also included in each run as a positive amplification control.
2.9. Data Availability
The virus genome sequences are available in GenBank under the following nucleotide accession numbers: OQ791576 for QQ40-1st passage on BF
2, OQ791577 for RR200-MINOR, OQ791578 for RR200-MAJOR, and OQ791581 for QQ40-10th passage on LEB cells. Sequencing data were submitted to the NCBI Sequence Read Archive and are available under the BioProject accession numbers listed in
Table 3.
2.10. Experimental Trials
2.10.1. Fish
A total of 125 sea bass fingerlings (3.0 to 8.5 g) and 20 adults (mean weight 400 g) were used through 3 different protocols. Before the experiments, some individuals from each batch underwent virological (inoculation on EPC, BF2, and SSN-1 cell lines) and bacteriological (nonspecific medium) analyses to confirm the absence of cultivable and pathogenic agents.
2.10.2. Experimental Design
Authorization to Conduct Research and Ethical Aspects
All the protocols were conducted in accordance with European Commission Recommendation 2007/526/EC on revised guidelines for the accommodation and care of animals used for experimental and other scientific purposes. The ANSES Plouzané site has authorization to conduct experiments on fish in its facilities according to Administrative Order No. C29-212-3 issued by the Prefecture of the Finistère. Furthermore, the procedure was approved by the national ethics committee on animal experimentation (COMETH ANSES/ENVA/UPC No. 16) and was authorized by the French Ministry for Education, Higher Education and Research under No. 20-023 #24166. Euthanasia involved the addition of a lethal dose of 100 ppm of Eugenol into the tank water. Animals were kept in contact with Eugenol until the complete disappearance of any respiratory activity. In case of evident signs of suffering during the experiments, compassionate euthanasia was performed.
Experimental System
Assays were carried out in filtered and UV-treated seawater continuously flowing into each tank (open circuit) to maintain optimal water conditions for the fish throughout the experiments (oxygen saturation > 80%, pH close to 8, water free of nitrates and nitrites). The tanks were maintained in a natural light/dark cycle (14 h/10 h in winter) in a room with air volume renewed every hour. The water temperature was either unregulated (first assay) or regulated at 18 °C ± 2 °C (second assay) or 20 °C ± 2 °C (third assay) and recorded with a wireless probe (Cobalt, Oceasoft®, Dickson, Addison, USA) coupled with an acquisition system (ThermoClient 4.1.0.24). For the whole experiments, fish were fed twice a day with commercial mini-pellets NeoGrower extra Marin (Le Gouessant Aquaculture, Lamballe, France).
Viral Challenge Conditions
In a first assay, 70 fingerlings were infected in a 50 L (L) tank. A total of 50 were immersed for three hours in 3 L of hyperoxygenated seawater containing 10 mL of the isolated virus (9th passage on LEB, with notable CPEs), whereas 20 were injected intraperitoneally (IP) with 50 µL of the same viral suspension and positioned in a basket in the same tank. Fifty negative controls were positioned in another 50 L tank after being immersed in Eagle’s solution containing antibiotics instead of the virus suspension. General behavior, the presence of clinical signs, and mortalities were recorded daily. At 16, 23, and 30 days post-infection (dpi), 5 fish per condition (negative controls were only sampled at 16 dpi) were sacrificed and the following organs were sampled: blood, brain, kidney, heart, spleen, dorsal fin, and gills. At the end of the experiment (60 dpi), three fish (immersed or injected) were sampled. Survivors were euthanized following the procedure described before.
For the second assay, 55 fingerlings, reared in a 50 L tank, were IP infected with viral supernatant (100 µL of an 11th passage on an LEB cell line with CPEs). The pelvic fins, dorsal fins, gills, hearts, spleens, and kidneys from five fish were sampled at 1, 4, 7, 12, 19, 26, and 35 dpi. In parallel, blood from the same fish was individually sampled for ELISA assay. Final blood sampling was performed at 4 months p.i.
The last assay included 20 adults. Seven fish were IP injected with 1 mL of the virus supernatant (23rd passage on the LEB cell line, controlled via RT-qPCR, Ct = 16.06) and seven were injected with a larvae homogenate (controlled by RT-qPCR, Ct = 20.44). Six fish were used as negative controls. Fish were positioned in three 400 L tanks and fed and monitored for more than 2 months. At 17 dpi, fish were anesthetized with Eugenol (20 ppm) and a small amount of blood was withdrawn from the caudal vein with a lithium heparinized vacutainer (BD Vacutainer LH 85 IU). At 73 dpi, a final sampling was performed with a maximum of blood withdrawn. Blood was centrifuged at 1200× g for 10 min, and the plasma samples were stored at −80 °C for further ELISA applications.
2.11. ELISA Test
Plasma samples were obtained either after several experimental trials conducted at Anses (see Section Viral Challenge Conditions) or directly from the infected farm.
Maxisorb® 96-well polystyrene plates (Nunc Immunoplates, Rochester, USA) coated with 100 µL·well−1 of non-purified rabbit anti-virus IgG n°183 D56 diluted 1/1000 in PBS were incubated for 16 h at 5 °C ± 3 °C. Twin plates, i.e., one with the virus and one as a control in which cell culture supernatant was used instead of the viral suspension, were prepared. Plates were washed 3 times with washing buffer (PBS-T20; phosphate-buffered saline, 0.02% Tween 20). Non-specific sites were blocked with homemade blocking buffer BB (5% skimmed milk, 0.2% Tween 20, 0.01% thimerosal mixed in PBS) and the plates incubated for 1 h at 37 °C before washing twice with PBS-T20. Then, 100 μL of viral suspension diluted to 1/2 in diluting buffer (DB; BB diluted to 1/10 in PBS) was distributed into each well and the plates incubated for 1 h at 37 °C before being washed three times with PBS-T20. A volume of 100 μL of sea bass plasma diluted to 1/100 in diluting buffer was distributed into each well, and the plates were then incubated for 2 h at 20 °C. Positive and negative control plasma obtained from in vivo assays were added to each plate. After washing three times with PBS-T20, the plates were incubated with 100 μL of purified and biotinylated rabbit anti-sea bass immunoglobulin M (IgG anti-IgM) diluted to 1/2000 in DB. After 1 h incubation at 37 °C, the plates were washed four times with PBS-T20. The wells were then incubated for 1 h at 37 °C with 100 μL of ExtrAvidin®-Peroxydase (Sigma, Burbank, CA, USA) diluted to 1/1000 in DB and washed six times with PBS-T20. A volume of 100 μL of a solution containing 4 mg of O-phenylenediamine (Sigma) and 4 μL of 30% hydrogen peroxide (Sigma) in 10 mL phosphate citrate buffer (pH 5) were distributed in each well for the revelation of the binding, and the plates were incubated for 20 min at RT. Finally, 25 μL of H2SO4 3 M (ChemLab) was added to each well to stop the reaction, and absorbance was measured spectrophotometrically at 492 nm in a Spark 10 M Tecan analyzer (Tecan Trading AG, Männedorf, Switzerland). The absorbance differences (ΔOD) between the wells with and without the virus were calculated, and the relative quantity of antibodies was expressed as the percentage of the relative value compared with the ΔOD of the positive control. The plasma was considered positive when the difference was greater than the threshold of positivity (TP) calculated as: TP = mean percentage of the negative control of the test + three standard deviations (SD) defined from 38 plasma samples from virus-free sea bass.
2.12. Statistical Analysis
All statistical analyses were performed using XLSTAT statistical software (2020), with a significance level of 5%. Viral load values were compared between healthy and dying larvae in the first instance and regarding the age of larvae in the second instance using a Mann–Whitney non-parametric test. All data are expressed as the mean ± standard deviation (SD).
4. Discussion
The purpose of this project was to identify the etiology of an abnormal mortality event observed in 20–35-day-post-hatching sea bass larvae from a hatchery. Cell culture inoculation with larval homogenates produced a CPE in cell lines, allowing the successful isolation of a virus, its re-inoculation into sea bass, and the investigation of its potential pathogeny in the larval and adult stages. The full-length genome of the virus was obtained using NGS and identified as a new virus related to the family Totiviridae. We propose the name sea bass toti-like virus (SBTLV) for the virus, a potential new member of the Totiviridae family.
Viral amplification in the LEB and BF2 cell lines was a several week process, and infected plates had to be trypsinized and replated with uninfected cells several times to increase the CPE and viral multiplication rate. Finally, the virus was isolated from healthy larvae after a long cultivation period in the BF2 and LEB cell lines. The presence of dsRNA, a hallmark of RNA virus replication, was confirmed via IFAT using the J2 antibody. Furthermore, a CPE-positive cell culture with a chloroform-treated inoculum was a feature of the presence of non-enveloped virions. The difficulty of viral amplification has also been reported for PMCV, another member of the
Totiviridae family, which uses an extracellular transmission route for replication to spread to uninfected cells during cell division, sporogenesis, or cell fusion [
11,
34].
Our analysis, including NGS performed on the positive cell culture supernatant, revealed that the virus obtained from the QQ40 sample was an RNA virus with a genome size of 6.8 kb. Bioinformatics analysis of the sequence enabled the determination of six potential ORFs on the positive strand of viral RNA. The genomic organization is compatible with those of most of the
Totiviridae family members, with a programmed ribosome frameshift (PRF) between the capsid (ORF1) and the RdRp (ORF2) leading to the production of a fusion protein that is thought to contribute to the correct encapsidation of the newly produced positive-strand RNA for the virion of most of the
Totiviridae family members [
35]. The phylogenetic analysis of the SBTLV RdRp attributes this protein to the Pistolvirus cluster proposed by Sandlund et al. [
14], which represents viruses from fish species; however, it should be noted that not all fish toti-like viruses are members of this cluster, the golden shiner toti-like virus and the bluegill toti-like virus RdRps cluster with the toti-like viruses from arthropods.
Although Totiviridae have been initially described with only two proteins, the capsid and RdRp, additional proteins have been described in Toti–like viruses from various species, particularly from fish, with the number of additional proteins varying from one in PMCV, GSTLV-1, and CCTLV-1 to three in CluTLV [
14] and four in SBTLV. The expression of these additional proteins and the mechanisms by which they are expressed deserves to be analyzed further. PMCV and SBTLV both display potential ribosome slippery sites upstream of the stop codon of ORF2; however, the downstream sequences have several additional stop codons in either of the three different frames, rendering this mechanism of translation unlikely.
In contrast with the ORF3s from PMCV, GSTLV-1, and CCTLV-1, which share almost no identity with other known proteins but might share structural properties and possibly have a chemokine-like domain [
14], the SBTLV-ORF3 protein has a significant identity (45%) with the CLuTLV-ORF3 protein but also, to a lesser extent, with small p10 proteins found in some bat coronaviruses. CLuTLV-ORF3 displays a 40 to 50% identity with p10 proteins found in avian reoviruses. All of these p10 proteins, either from reovirus or coronavirus, are related to the fusion-associated small transmembrane (FAST) protein family, which mediates cell–cell fusion and syncytia formation [
36,
37,
38]. The formation of syncytia observed in SBTLV-infected cells is probably mediated by the proteins coded by ORF3. The phylogenetic analysis of the FAST proteins from Reovirales highlights four distinct clades corresponding to fish, reptiles, and avian Orthoreovirus and the fourth clade corresponding to rotaviruses. The FAST proteins found in fish totiviruses belong to the avian clade, suggesting a recombination event in the FAST sequence from an avian reovirus and not from a fish reovirus, as might be expected. The PMCV FAST protein does not cluster with SBTLV or CLuTLV, suggesting distinct events leading to acquisitions of these FAST proteins in fish totiviruses.
Experimental infections were performed to assess the tropism and pathogeny of the newly isolated SBTLV and to characterize the development of an immune response towards SBTLV. Even if SBTLV could be detected in sampled fish using RT-qPCR after infection via immersion or intraperitoneal injection, no clinical signs nor mortality were observed in the fingerlings. The SBTLV-positive fish from the immersion condition indicated that the virus could enter fish through a natural route of infection and persist in infected specimens for at least two months. In the second assay, where only the IP route was tested, the spleens and hearts of infected fish displayed higher viral amounts than the other organs tested. This suggests preferential tropism of the virus for these two organs, which could be compared (at least for the heart) with PMCV [
39,
40], even if the mid-kidney also appears to be involved in the early amplification of the virus before it is transferred to the target organ, the heart [
13].
The immune response following toti-like virus infection in fish has not yet been described in detail in the literature. Even with the more described PMCV infection, only immune gene expression has been reported through transcriptomic studies, which provide evidence of early IFN-dependent gene activation, followed by B cell and MHC pathway activation [
40]. Here, the humoral immune response was evaluated using ELISA performed on sera from infected fish sampled from 1 to 35 dpi (second experimental trial). The results provide evidence of the development of a humoral-specific response in experimentally infected sea bass as early as at 7 dpi for sea bass fingerlings infected intraperitoneally and that is sustained over time (high number of seropositive samples at 73 days and 4 months dpi). Several samples collected directly from the hatchery at different times also revealed the presence of various levels of specific anti-SBTLV antibodies.
Thus, even if no clinical signs are observed in infected fingerlings or mature fish, SBTLV triggers an adaptive immune response, as demonstrated by the production of specific antibodies. The apparent stable amount of virus in the internal organs until 35 dpi, a period during which the humoral response has developed, suggests that this specific response is not efficient enough to control the viral load. Other specific immune mechanisms, particularly the cytotoxic T lymphocyte response mediated by the MHC, may be involved in viral control, as demonstrated for PMCV infection in salmon, for which upregulation of complement-associated genes was observed immediately before an upregulation of the genes involved in the T-cell response [
34].
Some gonad samples from mature sea bass exhibited positive SBTLV signals following a purification and concentration protocol. This result, with the absence of virus detection in internal organs from the same genitors, could suggest potential vertical transmission, even if strict analysis of sperm and eggs instead of gonads would give more insight into this hypothesis. A recent study focusing on the vertical transmission of PMCV, showed that the virus was detected in all the studied stages (eggs, larvae, fingerlings, and pre-smolt salmon) of progeny from PMCV-positive genitors [
41]. Nevertheless, no clinical signs of CMS were observed in hatcheries or in later intermediate stages, such as juveniles or pre-smolts. Mikalsen et al. [
42] failed to detect viral particles after reproduction by PMCV-positive genitors and artificial fertilization of salmon eggs [
42]. This could be consistent with the maintenance of the virus at very low levels (below LOD) in the very early stages of development and the inefficacy of the immune response at these stages to clear the virus, opening the way to its persistence until the adult stage.
This study describes the isolation and characterization of a novel toti-like virus from sea bass larvae, SBTLV. Even if the virus was regularly and widely detected between 2018 and 2022 in the hatchery, inducing the development of sustainable specific humoral responses, the pathogenicity of this virus remains to be elucidated. The very narrow time window when the virus can be detected (larvae) makes it difficult to assess its pathogenicity through in vivo trials. Our experimental trials were performed with the SBTLV strain QQ40, which was isolated after long-term culture of BF2 cells. This strain was pure; however, the initial sample from the larvae contained a mix of two strains (RR200-MINOR and RR200-MAJOR). These strains differed by 147 mutations, among which one resulted in a potentially extended ORF2, which seems highly unlikely if RdRp is produced as a fusion protein with the capsid. Another mutation results in the loss of a potential ribosome slippery site at the 3′ end of ORF2. If this slippery site plays a functional role during the infection process, the possibility that the pathogenicity is linked either to the mix of strains or to the minor strain cannot be excluded. Recently, another toti-like virus, with a high level of similarity to the minor variant we describe (98.8% nt identity), was detected in another sea bass hatchery after a serious mortality event; once again this was in the early larval stages. This observation suggests that SBTLV is present in the production environment and could, depending on the circumstances and potential co-factors, become a health and economic burden in the same way PMCV has.