The effect of exposure to farmed salmon on piscine orthoreovirus infection and fitness in wild Pacific salmon in British Columbia, Canada

Abstract

The disease Heart and Skeletal Muscle Inflammation (HSMI) is causing substantial economic losses to the Norwegian salmon farming industry where the causative agent, piscine orthoreovirus (PRV), is reportedly spreading from farmed to wild Atlantic salmon (Salmo salar) with as yet undetermined impacts. To assess if PRV infection is epidemiologically linked between wild and farmed salmon in the eastern Pacific, wild Pacific salmon (Oncorhynchus sp.) from regions designated as high or low exposure to salmon farms and farmed Atlantic salmon reared in British Columbia (BC) were tested for PRV. The proportion of PRV infection in wild fish was related to exposure to salmon farms (p = 0.0097). PRV was detected in: 95% of farmed Atlantic salmon, 37-45% of wild salmon from regions highly exposed to salmon farms and 5% of wild salmon from the regions furthest from salmon farms. The proportion of PRV infection was also significantly lower (p = 0.0008) where wild salmon had been challenged by an arduous return migration into high-elevation spawning habitat. Inter-annual PRV infection declined in both wild and farmed salmon from 2012-2013 (p ≤ 0.002). These results suggest that PRV transfer is occurring from farmed Atlantic salmon to wild Pacific salmon, that infection in farmed salmon may be influencing infection rates in wild salmon, and that this may pose a risk of reduced fitness in wild salmon impacting their survival and reproduction.

Fig 1. Map of BC with salmon farms and regions sampled.

This map shows the following: (1) locations of salmon farms (red dots), (2) the 9 regions where wild salmon were sampled, (3) three lakes discussed in the text, (a) Oweekeno (elevation 15 m) (b) Chilko (elevation 1172 m) and (c) Cultus (elevation 47 m), and three river systems also discussed, the Fraser, Skeena, and Nass. Region color corresponds to the cluster analysis in Fig 4. The blue arrow represents the major Fraser River salmon migration route [38].

Fig 2. Internal control ELF-1α Ct values indicating sample quality.

Ct values <30 are considered of sufficient quality for RTqPCR viral screening.

Fig 3. Proportions of PRV RT-qPCR-positive results.

Results are arranged in decreasing order. The “Wild” designations reflect the Region numbers in Fig 1; i.e., Wild 1 is from Region 1. Numbers of fish sampled are provided in parentheses on the horizontal axis labels. Relevant estimates and confidence limits for key differences in this figure were generated by the logistic regression modelling where the effects of potential confounding variables could be filtered out (Fig 5).

Fig 4. Hierarchical cluster analysis of test results by region and farmed categories.

Fig 5. Least-squares mean proportions for RT-qPCR positive test results.

The black squares provide a reference estimate for fish at high farm exposure level, and low migration challenge level in 2012 (the ‘common reference’). The blue circles are least-squares mean proportions with each of these factors switched in turn to the opposite level, with the other factors left at the common reference level. The vertical bars cover approximately 2 standard errors. Where a vertical bar does not span the gap between the two estimates, the difference is significant at approximately the 5% level in a test against a two-sided alternative.