The Effect of In Vitro Passage of Infectious Pancreatic Necrosis Virus (IPNV) on Virulence and Sensitivity of the Virus to Rainbow Trout Serum

—Three infectious pancreatic necrosis virus (IPNV) isolates (A1 serotype, Buhl subtype) were passaged five times in rainbow trout gonad (RTG)-2 cells with either minimum essential medium supplemented with 10% fetal bovine serum (MEM-10) or MEM-10 supplemented with 1% rainbow trout Oncorhynchus mykiss serum (MEM2RTS) to determine the effect of passage of the virus on its virulence and sensitivity to RTS. Mortalities in brook trout Salvelinus fontinalis fry were highly variable during viral passages; however, in general, most of the IPNV isolates were virulent under all conditions. The IPNV mortalities ranged between 30% and 89%. Maintenance of virulence during the five passages was dependent on the IPNV isolate and culture conditions. Rainbow trout serum did not always facilitate maintenance of viral virulence during the passages. Isolate mortalities after multiple passages were virulence dependent: the highly virulent isolate showed peak mortality from 4 to 11 d postexposure, while the low-virulence isolate showed delayed peak mortality from 8 to 14 d postexposure. The virus isolates passaged in MEM10 or MEM2RTS were generally resistant to in vitro inactivation by RTS except for the isolate from noble crayfish Astacus astacus passaged in MEM-10. This isolate became highly sensitive to RTS, demonstrating a reduction in tissue culture infectious dose with 50% endpoint of 1028.0/ mL in virus titer. However, increased RTS sensitivity of the virus did not correlate with decreased virulence. Overall, no relationship was observed between virulence and the RTS sensitivity of the virus. All isolates passaged showed identical monoclonal reaction patterns with a panel of 11 monoclonal antibodies, irrespective of viral virulence and sensitivity to RTS. Infectious pancreatic necrosis virus (IPNV), a member of the family Birnaviridae, was the first virus isolated from teleosts (Wolf et al. 1960) and is capable of mortalities as high as 90–100% in rainbow trout Oncorhynchus mykiss 1–4 months old (Frantsi and Savan 1971; McAllister 1983). Serum from unexposed rainbow trout (RTS) has been demonstrated to inhibit IPNV replication in vitro (Vestergard-Jörgensen 1973; Dorson and de Kinkelin 1974). The serum inhibitor was estimated to have a sedimentation coefficient of approximately 6S and thus is different from fish antibody immunoglobulin M, which has a sedimentation coefficient of 14–16S (Dorson and de Kinkelin 1974). Cell-culture-adapted virus strains were more susceptible to RTS than wild-type viruses (Dorson and de Kinkelin 1974; Hill and Dixon 1977; Dorson et al. 1978). Kelly and Nielson (1985) determined that virus adsorption to cells is * Corresponding author: [email protected] 1 Present address: Institute for Marine Biosciences, National Research Council of Canada, 1411 Oxford Street, Halifax, Nova Scotia B3H 3Z1, Canada. Received October 24, 2002; accepted April 4, 2003 reduced in the presence of RTS. Approximately 97% of virus preincubated with serum remained unadsorbed, compared with about 55% of control virus. Ögüt (1995) tested the ability of RTS to inactivate 109 IPNV isolates, all from serogroup A (Hill and Way 1995), and found that 65 were inactivated (a reduction in tissue culture infectious dose with 50% endpoint [TCID50] of more than 1021/mL). However, even within one serotype, not all of the virus isolates tested were inhibited by RTS. Since the first report relating the ability of RTS to inhibit replication of IPNV, it has been suggested by some researchers that the ability of IPNV strains to replicate in the presence of normal RTS is an important aspect of virus pathogenesis and can be considered as a virulence factor (Hill and Dixon 1977; Hill 1982). It has been reported that RTS sensitivity was inversely correlated with the virulence of IPNV and that avirulent strains could become virulent by passaging the virus in the presence of RTS (Hill and Dixon 1977; Hill 1982; Ögüt 1995). However, others could not substantiate these findings (Dorson et al. 1975, 1978; McAllister and Owens 1986). Together, these data indicate that RTS resistance does not always cor129 EFFECT OF IN VITRO PASSAGE OF IPNV relate with virulence and that RTS does not always sustain viral virulence during in vitro passage. One of major weaknesses of the study on RTS inhibition was that there was no standardization or quantitation of the RTS inhibitor (6S; Dorson and de Kinkelin 1974; Hill and Dixon 1977; McAllister and Owens 1986; Ögüt 1995). In the previous studies, researchers used a pool of RTS from a different strain of rainbow trout that might have differing amounts of RTS inhibitor. These could significantly affect the results of in vivo or in vitro experiments. Although earlier reports compared the changes in RTS sensitivity and virulence during viral passages with RTS or minimum essential medium supplemented with 10% fetal bovine serum (MEM-10) (Dorson et al. 1975; Hill and Dixon 1977; McAllister and Owens 1986; Ögüt 1995), the information had some limitations because serologically different IPNV strains (WB versus Sp) were used in various studies or limited numbers of isolates and passages were compared. McAllister and Owens (1986) used one isolate and compared viral virulence at passages 1, 5, 10, and 15. Ögüt (1995) used three IPNV isolates but only compared passage 1 and passage 11. A recent study (Bruslind and Reno 2000) showed that even one passage of IPNV (91-137 isolate, Buhl subtype) in Chinook salmon embryo (CHSE)-214 cells could result in a dramatic change in virulence. This demonstrates that the relationship between RTS sensitivity and virulence is more complex than previous work had indicated. Currently, no information is available comparing the relationship between RTS sensitivity and virulence at each passage during multiple passages of closely related IPNV isolates. The present study was conducted to clarify the possible effects of in vitro passage of IPNV on RTS sensitivity and virulence. We compared the relationship between RTS sensitivity and virulence at each of five passages of three closely related Buhl subtype strains that were either RTS sensitive or RTS resistant. We also attempted to determine whether the change in RTS susceptibility during viral passages in vitro was related to epitope alteration.