Vol. 106: 117–127, 2013 DISEASES OF AQUATIC ORGANISMS Published October 11 doi: 10.3354/dao02654 Dis Aquat Org FREEREE ACCESSCCESS Characterization of perch rhabdovirus (PRV) in farmed grayling Thymallus thymallus Tuija Gadd1,*, Satu Viljamaa-Dirks2, Riikka Holopainen1, Perttu Koski3, Miia Jakava-Viljanen1,4 1Finnish Food Safety Authority Evira, Mustialankatu 3, 00790 Helsinki, Finland 2Finnish Food Safety Authority Evira, Neulaniementie, 70210 Kuopio, Finland 3Finnish Food Safety Authority Evira, Elektroniikkatie 3, 90590 Oulu, Finland 4Ministry of Agriculture and Forestry, PO Box 30, 00023 Government, Finland ABSTRACT: Two Finnish fish farms experienced elevated mortality rates in farmed grayling Thy- mallus thymallus fry during the summer months, most typically in July. The mortalities occurred during several years and were connected with a few neurological disorders and peritonitis. Viro- logical investigation detected an infection with an unknown rhabdovirus. Based on the entire gly- coprotein (G) and partial RNA polymerase (L) gene sequences, the virus was classified as a perch rhabdovirus (PRV). Pairwise comparisons of the G and L gene regions of grayling isolates revealed that all isolates were very closely related, with 99 to 100% nucleotide identity, which suggests the same origin of infection. Phylogenetic analysis demonstrated that they were closely related to the strain isolated from perch Perca fluviatilis and sea trout Salmo trutta trutta caught from the Baltic Sea. The entire G gene sequences revealed that all Finnish grayling isolates, and both the perch and sea trout isolates, were most closely related to a PRV isolated in France in 2004. According to the partial L gene sequences, all of the Finnish grayling isolates were most closely related to the Danish isolate DK5533 from pike. The genetic analysis of entire G gene and partial L gene sequences showed that the Finnish brown trout isolate ka907_87 shared only approximately 67 and 78% identity, respectively, with our grayling isolates. The grayling isolates were also ana- lysed by an immunofluorescence antibody test. This is the first report of a PRV causing disease in grayling in Finland. KEY WORDS: Grayling · Perch rhabdovirus · Phylogenetic · Summer mortality · Freshwater Resale or republication not permitted without written consent of the publisher INTRODUCTION Rhabdoviruses are negative-stranded RNA viruses of the order Mononegavirales and have been isolated Of all piscine viruses, rhabdoviruses have the most from vertebrates, insects and plants (Wagner 1987). serious socio-economic impact. More than 14 fish The rhabdovirus genome, with a length of approxi- rhabdovirus species from both freshwater and mar- mately 11 to 16 kb, consists of 5 genes in the order 3’- ine fishes have been described (Mork et al. 2004, N-P-M-G-L-5’, encoding 5 structural proteins: nucle- Hoffmann et al. 2005, Kuzmin et al. 2009, Talbi et al. ocapsid (N), phospho- (P), matrix (M), glyco- (G) and 2011). The rhabdoviruses causing major epizootics RNA polymerase (L) protein (Dietzgen et al. 2011). are the novirhabdoviruses viral haemorrhagic septi- Within the family Rhabdoviridae, 9 genera are now caemia virus (VHSV) and infectious haematopoietic recognised, of which 3, Novirhabdovirus, Perhabdo - necrosis virus (IHNV) and the vesiculovirus causing virus and Vesiculovirus, infect fish worldwide (ICTV spring viraemia of carp (SVCV). All of these viruses 2013a). All novirhabdoviruses have a sixth non- are notifiable to the World Organisation for Animal virion gene between the G and L genes (Kurath et al. Health (OIE). 1997, Schütze et al. 1999, Johnson et al. 2000). The *Email: [email protected] © Inter-Research 2013 · www.int-res.com 118 Dis Aquat Org 106: 117–127, 2013 genera Perhabdovirus and Vesiculovirus comprise cases, fish samples were sent to the laboratory alive an ecologically diverse but genetically similar group in oxyge na ted water and processed on the same day. of viruses. Perhabdovirus, the new genus of fish Virological and bacteriological samples were col- rhabdoviruses, includes 3 species: Perch rhabdovirus lected during the pathological examination. Details (PRV), Anguillid rhabdovirus (AngRV) and Sea trout of grayling sam pled for virus isolation are given in rhabdovirus (STRV). Perhabdoviruses share morpho- Table 1. logical characteristics, genome organisation and sequence similarities with vesiculoviruses and with viruses in the newly proposed genus Sprivivirus (pro- Grayling fry mortality posal under revision; ICTV 2013a). SVCV (Ahne et al. 2002) is officially classified as a member of Farm A (2002) the genus Vesiculovirus, and pike fry rhabdovirus (PFRV) is classified as a tentative member of the A group of grayling fry showed elevated mortality genus Vesiculovirus according to the International after transfer to larger basins on 2 July. The mortal- Committee on Taxonomy of Viruses (ICTV) (Carstens ity rate was about 1% d−1. A sample of 20 fish was 2010); however, the newly proposed genus Spriv- sent to the laboratory on 10 July. Antibiotic treat- ivirus (ICTV 2013b) includes both SVCV and PFRV. ment was started with oxytetracycline, but it had no For other important fish pathogens such as Siniperca effect. The mortality rate was 13.5% between 10 chuatsi rhabdovirus (SCRV), the classification is still and 16 July, and a new sample of 20 fish was sent to under discussion, although their full-length genome the laboratory on 16 July. The water temperature at sequences are clearly more closely related to those of that time was 15.9°C. The total mortality of this fish perhabdoviruses than of members of any other viral group remains uncertain, since the affected fish genus (ICTV 2013a). groups were destroyed after the detection of a viral Until 1981, PFRV and SVCV were the only known infection. vesiculo-type viruses responsible for disease epi- zootics in cultured fish. A rhabdovirus pathogenic for perch Perca fluviatilis was reported in France in 1980 Farm B (2003) from individuals exhibiting a loss of equilibrium and swimming disturbances (Dorson et al. 1984) and later Eyed grayling eggs, disinfected with iodo - also from Denmark, Ireland, Germany and Norway phores, were transferred from the first farm to (Jørgensen et al. 1993, Dannevig et al. 2001). A PRV- another fish farm in the lake area. This farm like virus has been associated with diseased grayling reported elevated mortality in the grayling fry dur- Thymallus thymallus in France (Betts et al. 2003). In ing the period from 10 to 16 July. The average northern Finland, the brown trout rhabdovirus virus water temperature during this period was 20°C. ka903_87 was isolated from farmed brown trout The fish were treated with oxytetracycline, but Salmo trutta m. lacustris in 1987 (Koski et al. 1992, without effect, and a sample of 14 fish was sent to Jørgensen et al. 1993). the laboratory on 16 July. The mortality rate of this Here, we describe the isolation and characteriza- fish group remains uncertain, since the fish were tion of a PRV that was connected with the summer destroyed. mortalities of grayling fry in 2 freshwater fish farms. Table 1. Farms, dates, sample numbers and isolate codes for grayling Thymallus thymallus sampled for virus isolation MATERIALS AND METHODS Farm Sampling Sample No. of No. of Isolate code date code samples positive of positive Fish sampling (dd/mm/yy) samples samples A 10/07/2002 ka706−707_02 2 2 ka706−707_02 Fish samples were received A 16/07/2002 ka712_02 1 1 ka712_02 from 2 fish farms, A and B, both B 16/07/2003 ka672_03 1 1 ka672_03 situated in a different water catch- A 01/07/2009 ka495−496_09 2 1 ka495_09 ment area in Finland, but sharing A 13/07/2009 ka512−524_09 13 7 ka512_09, ka519−524_09 A 10/09/2009 ka 626−655_09 30 1 ka 636_09 an epidemiological link through A 27/07/2010 ka501_10 1 1 ka501_10 transfers of disinfected eggs. In all Gadd et al.: Perch rhabdovirus in grayling 119 Farm A (2009) to standard virological procedures as described by Raja-Halli et al. (2006) and inoculated onto subcon- During the start of the feeding period in June, fluent monolayer cell cultures of bluegill fry (BF-2, exceptionally high mortality was noted in groups of Wolf et al. 1966) and epithelioma papulosum cyprini grayling fry. Mortality normally remains around (EPC, Fijan et al. 1983) cell lines. 10%, but reached up to 80% at this time. The first Originally the samples were incubated at 16°C sample was sent to the laboratory on the first day of in CO2. Starting in 2009, the samples were buf - July, when the fish were transferred to 5 larger fered with sodium carbonate 7.5% (Gibco 25080) basins. The water temperature at this time was and in cubated without CO2. The samples were in - 15.2°C. spected regularly under a microscope for the Follow-up samples were sent to the laboratory on occurrence of cytopathic effect (CPE). After 7 d of 13 July, when the mortality in on-growing basins incubation, supernatant from samples without CPE had reached about 5%. At the same time, the fry was diluted 1:100 and 1:1000, sub-cultured onto from the 3 most affected basins were destroyed, fresh cells and incubated for a further 7 d. In cases because of the suspicion of a viral infection. In the 2 where total CPE was evident, the cell culture remaining groups, the mortality rates remained medium was collected and stored at −70°C for high (5 and 13%), and new samples were sent to future studies. the laboratory in September. At the same time, antibiotic treatment was started with oxytetracy- cline. This treatment had no effect on mortality, and ELISA for IHNV, infectious pancreatic necrosis the fish were destroyed on 15 October, when the virus (IPNV), SVCV and VHSV total mortality reached 20%. Aliquots of 50 µl of culture medium from cell cul- tures showing evidence of CPE were analysed with Farm A (2010) commercial ELISA kits according to the manufac- turer’s instructions to test for the presence of IPNV, The mortality rate in a group of grayling fry SVCV, VHSV (Test-Line), and since 2008, also for the increased from the daily 0.1−0.4% to 2% at the end presence of IHNV (Bio-X Diagnostics S.P.R.L).