Survival of Frog Virus 3 in Freshwater and Sediment from an English Lake

Ranaviruses (Iridoviridae: Ranavirus) are emerging pathogens of fish, amphibians, and reptiles (Whittington et al. 2010; Marschang 2011; Miller et al. 2011). Frog virus 3 (FV3), the type species of the genus, was first isolated from an asymptomatic leopard frog (Rana pipiens) and not initially considered a primary pathogen of frogs (Granoff et al. 1966). Frog virus 3 and similar ranaviruses have since been recovered from free-living or farmed, healthy, and diseased frogs in the Americas, Europe, and Asia (Chinchar et al. 2009), and a FV3-like virus caused disease in fish and reptiles following laboratory exposure (Brenes et al. 2014). Because of their wide distribution and high virulence, amphibian ranaviruses are a significant risk to global amphibian populations (Miller et al. 2011), leading to the listing of amphibian ranavirus by the World Organization for Animal Health (Schloegel et al. 2010).

Ranaviruses can be transmitted by contaminated water and sediment (Brunner et al. 2007). The minimum infective dose using an immersion infection model was 100 plaque-forming units/mL in larval tiger salamanders (Ambystoma tigrinum; Brunner et al. 2005); however, the likelihood of becoming infected increased with virus titer. For successful transmission, sufficient virus must retain infectivity long enough to infect a susceptible host. Knowledge of viral persistence in the environment can be used with infectious dose data to make epidemiologic inferences to provide management and control strategies. We evaluated the persistence of FV3 in freshwater and sediment from a lacustrine environment in England, complementing similar studies in Germany (Nazir et al. 2012) and the US (Johnson and Brunner 2014). Comparison of results helps to elucidate whether factors such as geographic location and habitat affect ranavirus survival. We additionally compared survival in the two matrices at a range of temperatures applicable to the known geographic range of FV3.

Freshwater and sediment were collected from Radipole Lake, Weymouth, Dorset, UK (50°38′N, 02°28′W) in January 2010. Sediment was taken approximately 1.5 m from the water’s edge. Water parameters were pH 8.16, NH₃⁺ 0.0 ppm, NO₂⁻ 0.0 ppm, NO₃⁻ 10.0 ppm, general hardness (calcium carbonate) 280 ppm, carbonate hardness 210 ppm, conductivity 458 μs, conductivity normalized to 25 C 643 μs, and dissolved oxygen 11.34 mg/L. Water was used untreated or passed through a 0.2-μm nitrocellulose syringe filter (Sartorius, Epsom, Surrey, UK). Sediment was placed into 16 sealable bottles in 20-g aliquots. Four bottles each were left untreated, dried at 32 C, heat sterilized at 120 C for 30 min, and dried at 32 C and heat sterilized at 120 C for 30 min.

We propagated FV3 by using Epithelioma papulosum cyprini (EPC) cells (Fijan et al. 1983) and Glasgow modification of Eagle’s minimal essential medium (GMEM; Sigma-Aldrich, Poole, Dorset, UK), supplemented with 10% fetal calf serum, penicillin (100 IU/mL), and streptomycin (100 μg/mL). Cells were grown in 75 cm² flasks to confluent monolayers, infected with FV3, and reincubated at 25 C. Virus was recovered on day 7 by decanting the cell culture media.

The decanted FV3 was diluted 1/10 in untreated and filtered freshwater and GMEM. An aliquot from each was taken immediately (day 0), and the virus titrated. Each virus-containing matrix was divided into four bottles and stored at 4, 15, 20, and 30 C. An aliquot was taken from each bottle weekly for 98 d, and any virus titrated. All aliquots were filtered (0.2 μm) before titration.

Decanted FV3 (diluted 1/10 in sterile water) was added to each of the 16 sediment samples in volumes of 60 mL. Immediately prior to addition to the sediment, the virus was titrated (day 0). A sample from each treatment group was stored at 4, 15, 20, and 30 C. Aliquots were taken from each bottle every 24 h for 4 d (days 1–4) and then every 7 d for 49 d (days 11–53), and any virus titrated. Prior to taking aliquots, bottles were shaken and left to settle for a few minutes. Approximately 1 mL of liquid was added to an Eppendorf tube and centrifuged at 16,100 × G for 15 min at 4 C and filtered (0.2-μm) before titration.

The virus was titrated by using EPC cells in 96-well microtiter plates. Endpoint dilution assays to determine the 50% tissue culture infectious dose (TCID₅₀) titer (Karber 1931) were performed by inoculating serial log₁₀ dilutions of virus onto the cells (six wells per dilution). Cells were incubated at 25 C for 7 d and observed for cytopathic effect. Minimum detectable titer was 17.6 TCID₅₀/mL.

Logistic regression was performed for each matrix-treatment-temperature combination by using the Ime4 package (Bates et al. 2015) in R (R Core Team 2014). The dose.p function (Venables et al. 2002) was used to predict the time required for 90% reduction in virus titer (T-90 values) and calculate 95% confidence intervals. In some cases, virus titer increased before declining; the maximal titer was used as the initial value for the regression. Values below the limit of detection were treated as zeros in the analysis.

The data and fitted values for survival of FV3 in untreated lake water and sediment at four temperatures are plotted in Figure 1. The T-90 values and 95% confidence intervals of FV3 in treated and untreated lake water and sediment and culture media are presented in Table 1. The virus survived in untreated water and sediment; however, survival times were lower in sediment. Viruses lost infectivity in both matrices with increasing temperature. In untreated water, T-90 values decreased from 34 d at 4 C to 5 d at 30 C. In untreated sediment, T-90 values decreased from 10 d at 4 C to 1 d at 30 C.

Similar trends for persistence of FV3 were observed by Nazir et al. (2012) who evaluated survival of FV3 in water and soil from a small pond; however, in that study, T-90 values for pond water were 22 d at 20 C and 62 d at 4 C, and T-90 values for soil were 18 d at 20 C and 33 d at 4 C. Shorter survival times in our study may be due to the initial titer of virus being lower; filtration of samples prior to titration also likely reduced titer. Filtration removed approximately 0.75 log₁₀ units of FV3 when spiked into sterile water in pilot experiments using the same methods (A.E.B. unpubl. data). Filtration was done to minimize contamination, which could have damaged cell growth and masked the presence of virus, especially at low titers. Our addition of virus and culture media suspension to nontreated water and sediment samples may have affected virus survival by introducing novel components that may have enhanced the growth of other microorganisms and reduced virus survival (Nazir et al. 2012). The use of virus adsorbed to germ carriers by Nazir et al. minimized the introduction of novel components to water and sediment samples. However, the protective filter membrane would have prevented microorganisms in the sample from gaining access to the virus (Johnson and Brunner 2014). Another difference between the two studies was the source of water and sediment. Persistence of a FV3-like ranavirus varied substantially in water samples from five ponds, although results were qualitatively similar (Johnson and Brunner 2014). Their study was limited to detection of viral DNA rather than live virus past 24 h, due to rapid decline of virus titer on addition to pond water, so comparison with our results should be made with caution. Ranavirus titers, as detected by quantitative real-time PCR, initially declined quickly, with T-90 values across ponds of <1 d at 22–24 C, but then declined much more slowly, remaining detectable for at least 78 d. Similar variation in decline was not observed in our study in which the viral titer was measured in water samples immediately after inoculation and again after 7 d. Johnson and Brunner (2014) measured titer up to 30 min postinoculation, and then every 24 h, with the breakpoint observed around 3 d after inoculation, so any similar trend in our study may have been missed.

Survival of FV3 was consistently higher in filtered than untreated water, perhaps due to organic matter or microorganisms in untreated water. Bacteria can have antiviral activity (Kamei et al. 1988); however, attempts were not made to isolate bacteria or other components from the water in this trial, so no conclusions can be made. Survival of FV3 tended to be reduced in filtered water compared with GMEM, possibly because serum in GMEM aided survival as occurs with other viruses (Frost and Wellhausen 1974; Pietsch et al. 1977). Alternatively, virus-inactivating substances in the water may have passed through the filter (Yoshimizu et al. 2005).

In sediment, FV3 survival was consistently higher in sterilized than nonsterilized samples. Although persistence in the two nonsterilized samples was similar, there appeared, particularly at 4 C, to be variation in virus inactivation with respect to sediment being dried at 32 C or untreated prior to addition of virus, with more virus surviving in dried samples. It is unlikely that lack of moisture affected virus survival because the addition of the virus suspension to the dried and nondried sediment samples at the start of the experiment caused both to become moist. Also, for the two sterilized samples, virus survival was higher in the predried sample, indicating that drying the sediment affected substances responsible for inactivating the virus (or that heat sterilization was more effective with dry sediment).

There are few data for survival of other ranaviruses in the natural environment. Epizootic hematopoietic necrosis virus persisted in distilled water for 97 d with no decrease in titer and survived in frozen infected tissue culture medium and fish tissue for 2 yr (Langdon 1989). That study was not representative of environmental samples and cannot realistically be used to determine virus survival in the environment; however, it suggests ranavirus may survive in the environment for extended periods when frozen.

Our results indicate that FV3 would not remain infectious outside of the host for long at high temperatures. However, in temperate locations during winter, sediment or, more likely, water could retain sufficient virus to remain infectious weeks after contamination.

This work was funded by Department for Environment, Food and Rural Affairs contract FC1194. We thank W. Ahne (University of Munich, Germany) for providing the FV3 isolate.