2025 Volume 60 Issue 3 Pages 138-144
Infectious hematopoietic necrosis virus (IHNV) was detected in female masu salmon Oncorhynchus masou broodstock in Hokkaido, Japan, in the fall of 2021 and 2022. In 2021, an outbreak of the disease occurred in yearling masu salmon at another nearby hatchery. We investigated the origin of the IHNV strains detected in 2021 and 2022 by genetic analysis and assessed the pathogenicity of the 2021 strain in fish ranging from fingerlings to juveniles weighing up to 50 g. The results showed that the strains detected in both years shared nearly identical nucleotide sequences and belonged to the JN lineage, which is prevalent in Honshu. Furthermore, the 2021 strain was found to be highly pathogenic even to juveniles over 10 g in weight. These findings revealed that a highly pathogenic strain, similar to those found in Honshu, has spread into rivers in Hokkaido. Until this virus is eradicated from Hokkaido, it is essential to implement strict measures such as treating wastewater during egg collection, disinfecting eggs and rearing equipment, and continuing surveillance for IHNV in rivers.
Large-scale stocking of chum salmon Oncorhynchus keta and other salmonid species for the purposes of stock enhancement has a long history in Hokkaido. As a major fishery species, stocking of masu salmon O. masou has been practiced for more than 100 years, beginning in the latter half of the 19th century (Miyakoshi, 2006). To prevent harmful consequences from the hatching and releasing of this species, the Japan Fisheries Research and Education Agency (FRA) conducts annual monitoring surveys to detect fish pathogens, mainly viruses, in the ovarian fluid of anadromous female broodstock of salmonid fishes captured in rivers (Yoshimizu and Nomura, 1989; Yoshimizu et al., 1993).
Infectious hematopoietic necrosis virus (IHNV) was isolated in 15% (10/65 fish) of masu salmon broodstock in a river in Hokkaido in the fall of 2021. Although three cases of IHNV in salmon broodstocks had been detected across five rivers in the previous 20 years (Kawana, personal communication), there had previously been no outbreaks of IHN in hatchery facilities or juvenile ponds except for two cases in fingerlings (age-0 fish) in hatchery ponds exposed to river water in the FRA stock enhancement program. However, IHN disease occurred in yearling masu salmon reared in a nearby region at the same time as the detection of the virus in broodstock in 2021. In Hokkaido, which is geographically isolated from Honshu, IHNV was first introduced from North America in 1971 and only the U genogroup, which mainly infects fingerlings, had been detected before the year 2000 (Hatakeyama, 2006). The IHNV strain detected in 2021 appears not to be the preexisting U genogroup strain but could be a strain from Honshu that is highly pathogenic and even infects large fish (>10 g). The subsequent detection of IHNV in the same river in the fall of 2022 added to fears that a Honshu strain of IHNV had reached a river in Hokkaido and begun to spread. If such a strain became established in the rivers of Hokkaido and spread widely, as has already occurred in Honshu, it would be quite difficult to eradicate the virus entirely. This would inflict serious damage to masu salmon stocks, because released salmon ranging from small juveniles to larger fish weighing more than 10 g could become infected and die in large numbers. Therefore, in the present study, the origins of the IHNV strains isolated from masu salmon broodstock in 2021 and 2022 were examined using genetic analysis. Additionally, the virulence of the 2021 IHNV strains on fingerlings and juvenile masu salmon was compared against that of the ChAb76 strain, an existing strain in Hokkaido, as a representative of the U genogroup.
The MsSb2021 strain of IHNV was used in the study for genetic analysis and infection testing, and the MsSb2022 strain was only used for genetic analysis. These strains were isolated from masu salmon ovarian fluid by using Chinook salmon embryo (CHSE-214) cells. The ovarian fluid was collected from masu salmon broodstock in a river in Hokkaido in the fall of 2021 and 2022 during monitoring surveys for the presence of fish pathogens. The strains were positively identified as IHNV by RT-PCR by using the method described by Arakawa et al. (1990). The ChAb76 strain (Yoshimizu et al., 1989), which was isolated from chum salmon in Hokkaido in 1976, was used as a control. The viruses were cultured and stored at −80°C until use.
Nucleotide sequencingOnce the cryopreserved virus was thawed, RNA was extracted by using the MasterPure DNA and RNA purification kit (LGC Biosearch Technologies) according to the manufacturer’s protocol. Viral sequencing methods followed the procedure described by Namba et al. (2021) to synthesize cDNA and prepare DNA data for analysis. 3-1F RT and 3-4R RT primers were used as the primer sets. The 50 μL of reaction mixture contained 1 μM of each primer, 25 μL of 2× Reaction Mix, 2 μL of SuperScript III RT/Platinum Taq Mix (Invitrogen), 1 μL of RNA, and distilled water. RT-PCR was performed with a programmed temperature control system, and amplification conditions were 55°C for 30 min and 94°C for 2 min, followed by 40 cycles of 94°C for 15 s, 55°C for 30 s, 68°C for 2.5 min, and 68°C for 5 min. PCR products were purified by gel extraction and used for DNA sequencing. Sequencing was performed with 11 primers (3-1F RT, 3-4F RT, 2701F, IG577, 4182R, HG, 3395F RT, fG 1150, 3-1R, 3-2R, and NV Ext) covering the viral coat protein (G) gene described in Namba et al. (2021) and was performed by using the BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific) on a 3730xl DNA Analyzer (Thermo Fisher Scientific).
Molecular phylogenetic treeTo perform phylogenetic analysis, the genetic sequences from both MsSb strains as well as ChAb76 (accession number AB250927) were analyzed along with those from a total of 28 strains representing all of the IHNV lineages isolated in Japan and North America as described by Nishizawa et al. (2006), Mochizuki et al. (2009), and Namba et al. (2021). The base sequences encoding the G proteins of these strains were obtained from the GenBank database. The 28 strains (accession numbers in parentheses) included six strains of the U genogroup from North America (L40879, M16023, U15170, U50401) and Japan (AB250928, AB250931); four strains of the Jnk lineage (LC148366, LC159155, LC159156, LC159190); six strains of the JS lineage (AB250933, AB250935, AB510193, AF244128, LC159186, LC159187); and twelve strains of the JN lineage from masu salmon (LC148363, LC159139, LC159152, LC159158, LC159171) and the closely-related RtNag96 strains (AB250932, AB288207, LC159166, LC159168, LC159169, LC159175, LC159176).
Multiple sequence alignment was performed on these sequences by using Molecular Evolutionary Genetic Analysis (MEGA) version 12.0 (Kumar et al., 2024) with the MUSCLE interface. Phylogenetic trees were constructed by using default parameters and 1000 bootstrap iterations and estimated by using the maximum likelihood method and a general time-reversible model using G (range of variance).
Fish studiedWe examined 223 masu salmon. These were comprised of 75 fingerlings (average body weight 1.04 g), 70 age-0 juveniles (1.90 g), 48 age-1 juveniles (9.63 g), and 30 age-2 juveniles (46.2 g). These fish were reared in running water at 11°C and fed dry pellets twice daily at a facility in the Sapporo Field Station, Fisheries Resources Institute, FRA.
Methods of infection with IHNVImmersion method: Seventy-five fingerlings weighing an average of 1.04 g were divided into 3 groups of 25 fish each. Two of the groups were used for virus infection, and the third was used for the control. Fish in each test group were infected by immersion in 2 L of viral diluted solution of ChAb76 or MsSb2021 strains at 103.80 TCID50 (50% tissue culture infectious dose)/mL for 60 min with aeration at 11°C. After infection, the fish from each test group were transferred to a 30 L tank and reared in running water. Minimum Essential Medium (MEM, Gibco) with 5% serum was used for the control group instead of the virus dilution.
Intraperitoneal (IP) injection: IP injection tests were conducted three times according to body weight.
Seventy juveniles weighing an average of 1.90 g were divided into seven groups of 10 fish each. Three groups each were injected with the ChAb76 or MsSb2021 strain, and the remaining group was used as the control. The experimental fish were injected intraperitoneally to achieve virus concentrations of 102.80, 103.80, or 104.80 TCID50/fish.
Forty-eight age-1 juveniles weighing an average of 9.63 g were divided into 8 groups of 6 fish each. Three groups were injected with the ChAb76 strain (doses: 104.80, 105.80, or 106.80 TCID50/fish), four groups with the MsSb2021 strain (doses: 102.80, 103.80, 104.80, or 105.80 TCID50/fish), and the remaining group was used as the control.
Thirty age-2 juveniles weighing an average of 46.2 g were divided into six groups of 5 fish each. Two of the groups were injected with ChAb76 (doses: 105.80 or 106.80 TCID50/fish), three groups with MsSb2021 (doses: 103.80, 104.80, or 105.80 TCID50/fish), and the remaining group was used as the control. For all IP injection tests, the control group was injected with MEM medium with 5% serum instead of the virus dilution. The fish body weights and virus doses for injection are shown in Table 1.
| IHNV Strain | B.W.(g) | Cumulative mortality (%)** IP Injection dose (TCID50/fish) | ||||
|---|---|---|---|---|---|---|
| 102.80 | 103.80 | 104.80 | 105.80 | 106.80 | ||
| ChAb76 | 1.90 | 60 | 70 | 100 | ― | ― |
| 9.63 | ―* | ― | 0 | 50 | 50 | |
| 46.2 | ― | ― | ― | 0 | 0 | |
| MsSb 2021 | 1.90 | 100 | 100 | 100 | ― | ― |
| 9.63 | 50 | 100 | 100 | 100 | ― | |
| 46.2 | ― | 80 | 100 | 100 | ― | |
*:Not tested. **At the end of the observation period. Statistical processing methods were performed by Fisher’s exact tests with Bonferroni correction.
After immersion and IP injection, the test fish were reared in running water at 11°C and fed dry pellets twice daily for 30 days, and the number of dead fish was recorded daily. Significant differences in final cumulative percent mortality were determined with a Fisher’s exact test, and survival plots were estimated with the Kaplan–Meier method and compared with a log-rank test. Bonferroni correction was used for these tests. In the case of IP injection, these statistical procedures were performed for fish groups of each body weight. EZR software was used to perform the statistical tests (Kanda, 2013).
Virus assays were conducted on all dead fish in all test groups, as well as on all surviving fish in cases in which there were 10 or fewer survivors in a group or on 10 randomly selected surviving fish in cases in which there were more than 10 survivors, including for the control group. The head and internal organs (including the kidney and spleen) were sampled from fish of the 1.04 and 1.90 g body weight groups, and only the kidney and spleen were sampled from fish of the 9.63 and 46.2 g groups. All samples were stored at −80°C until use. Each organ sampled from control groups of the 9.63 and 46.2 g fish was tested for bacteria isolation using TSA and modified cytophaga agar media to confirm the absence of bacterial infection.
Virus titration assaySamples frozen at −80°C were homogenized by adding 9× MEM supplemented with 10× antibiotic-antimycotic mixture (Gibco), centrifuged at 15,000 ×g for 5 min, and only the supernatant was kept at 4°C overnight. A 10-fold step dilution series of the supernatant was prepared with MEM and inoculated into 96-well plates cultured with EPC cells at 50 μL/well. Virus titer was shown by the TCID50.
Nucleotide sequences of 1,527 base pairs, including one open reading frame encoding the G protein, were obtained for IHN viruses isolated in 2021 and 2022 from masu salmon broodstock, and the sequences of the two strains were found to be almost identical (there was a one-base difference). The molecular phylogenetic tree created by adding the G protein sequences of IHNV obtained from the GenBank database to these sequences is shown in Fig. 1. The ChAb76 strain, a virus strain previously isolated in Hokkaido, was classified to the U genogroup, whereas both MsSb strains were found to belong to the JN lineage found in Honshu.
No mortality and no viral presence in surviving fish was observed after immersion infection with the ChAb76 strain, as was the case with the control group (Fig. 2). These results are consistent with previous reports on the ChAb76 strain in rainbow trout (Mochizuki et al., 2009). In contrast, mortality of the group exposed to the MsSb2021 strain began after one week, and the cumulative mortality reached 92% by the end of the experiments. The cumulative mortality results showed that the MsSb2021 strain was significantly different (P < 0.05) from both the ChAb76 strain and the control. The dead fish showed characteristic symptoms of IHN, such as darkening of the body, abdominal distention, ascites accumulation, and hemorrhage in the lateral muscles. The virus was detected in all survivors and dead fish, and some survivors showed a virus titer of more than 106.0 TCID50/g (Fig. 2).
A, C: ChAb76; B, D: MsSb2021.
Arrowhead indicates the lower limit of quantification for viral titer: 102.80 TCID50/g.
The results of the IP injection tests were as follows (Figs. 3, 4 and Table 1).
A: BW 1.90 g, ChAb76. B: BW 1.90 g, MsSb2021. C: BW 9.63 g, ChAb76. D: BW 9.63 g, MsSb2021. E: BW 46.2 g, ChAb76. F: BW 46.2 g, MsSb2021.
A: BW 1.90 g, ChAb76. B: BW 1.90 g, MsSb2021. C: BW 9.63 g, ChAb76. D: BW 9.63 g, MsSb2021. E: BW 46.2 g, ChAb76. F: BW 46.2 g, MsSb2021.
Arrowhead indicates the lower limit of quantification for viral titer: 102.80 TCID50/g.
For the age-0 juveniles of average weight 1.90 g, all MsSb2021 strain injection groups showed high pathogenicity (100% cumulative mortality), and the ChAb76 strain injection group also showed high pathogenicity at a high injection dose (104.80 TCID50/fish). The Fisher’s exact test showed significant differences (P < 0.05) between the control group and the ChAb76 strain injection group at a dose of 104.80 TCID50/fish and between the control group and all MsSb2021 injection groups, and the log-rank test showed significant differences between the control group and the ChAb76 strain injection group at a dose of 103.80 or 104.80 TCID50/fish and between the control group and all MsSb2021 injection groups as well as between the ChAb76 strain injection group (102.80 TCID50/fish) and all MsSb2021 injection groups and between the ChAb76 strain injection group (103.80 TCID50/fish) and the MsSb2021 strain injection group (104.80 TCID50/fish). The virus was detected in some surviving fish in the ChAb76 injection groups.
For the age-1 juveniles of average weight 9.63 g, the cumulative mortality after inoculation with 104.80 TCID50/fish of the ChAb76 strain was 0%, and when even higher titer viruses were injected, the cumulative mortality was 50%. Conversely, all MsSb2021 strain injection groups except for the lowest injection dose (102.80 TCID50/fish) had 100% mortality. The Fisher’s exact test indicated a statistically significant overall difference; however, this difference was not observed between each group. Furthermore, the log-rank test revealed a significant difference between the control group or the ChAb76 strain injection group at a dose of 104.80 TCID50/fish and all MsSb2021 injection groups with the exception of the lowest injection dose group (102.80 TCID50/fish). The presence of the virus could not be detected, or was at a low level, in the surviving fish of the ChAb76 strain injection group.
For the age-2 juveniles of average weight 46.2 g, no mortality was observed and no virus was detected in any fish after inoculation with the ChAb76 strain at any dose. Conversely, the cumulative mortality rates of all MsSb2021 strain injection groups reached or exceeded 80%. The Fisher’s exact test didn’t indicate a statistically significant between each group, though overall difference. Furthermore, the log-rank test revealed a significant difference was observed between the control group or ChAb76 strain injection group at a dose of 105.80 or 106.80 TCID50/fish and all MsSb2021 injection groups with the exception of the lowest injection dose group (103.80 TCID50/fish). The virus was undetectable or present at low titer in surviving fish of the ChAb76 strain injection groups.
These results indicate that the ChAb76 strain was only pathogenic to juveniles weighing less than about 10 g, and was especially deadly to fish of around 2 g, whereas the MsSb2021 strain had strong pathogenicity regardless of fish size.
Fukuda et al. (1992) compared the pathogenicity of IHNV strains isolated in the 1970s and 1980s from rainbow trout in Honshu, Japan, and showed that the strains became highly pathogenic over time not only to juveniles but also to larger fish. Nishizawa et al. (2006) reported variations in the G gene of IHNV isolated from all over Japan. The IHNV strains isolated mainly from Hokkaido in the 1970s belong to genogroup U, whereas those isolated from 1980 onward as the virus was spreading and rapidly mutating in Honshu formed the new gene lineages JRt Sizuoka and JRt Nagano. These lineages are significantly more virulent to rainbow trout than group U (Mochizuki et al., 2009) and include group J, which includes three subgroups: JS, JN, and Jnk (Namba et al., 2021). Hatakeyama (2006) analyzed the NV genomes of IHNV strains from salmon farms impacted by IHN in Hokkaido and found that all IHNV isolates from the 1980s to the 2000s were very similar to the RB-76 strain, which belongs to group U. Therefore, strongly pathogenic virus strains such as those of group J, which are prevalent in Honshu, had not intruded into Hokkaido until the 2000s, although IHN has been occurring in young masu salmon (body weight about 50 g) in salmon culture farms in Hokkaido since the 1980s (Suzuki and Sakai, 1991).
However, in the present study, it is clear that the IHNV strains isolated from masu salmon broodstock in 2021 belong to gene lineage JN of group J from Honshu and show high pathogenicity even to large juveniles. Furthermore, during the monitoring survey of masu salmon broodstock in the fall of 2022, IHNV was detected again in the same river in which IHNV was observed in 2021, although the detection rate was lower (Kawana, personal communication). This indicates that virus strains of the J genogroup have entered rivers in Hokkaido.
In masu salmon stock enhancement programs, eggs are disinfected twice with iodine just after water absorption at fertilization and during the eyed period, and if eggs and larvae are reared in virus-free spring water, viral diseases do not occur during these periods. The risk of IHN increases with the introduction of river water as the fry grow, but since they are no longer in the larval stage at this point, almost no IHN was previously observed. However, if a new strain like MsSb were to become established in the rivers of Hokkaido and spread widely, it will be quite difficult to eradicate, as has been the case in Honshu, and the virus will impact masu salmon resources seriously. Fortunately, isolation rates of the virus in masu salmon broodstock are still low at present as described in this paper, and it remains possible that the virus will disappear from the river. Therefore, to prevent the spread of IHNV, it is important to thoroughly treat wastewater at the fertilization stage and sterilize eggs and rearing gear, and also to continue to monitor viral presence through pathogen monitoring surveys of masu salmon broodstock.
We thank the staff of Field Stations in Salmon Research Department, FRA, for their cooperation in collecting samples and conducting the survey.
