Factors that can influence the transmission of aquatic viruses in fish production facilities and natural environment are the immune defense of host species, the ability of viruses to infect host cells, and the environmental persistence of viruses. In this thesis, fish cell lines were used to study different aspects of these factors. Five viruses were used in this study: viral hemorrhagic septicemia virus (VHSV) from the Rhabdoviridae family; chum salmon reovirus (CSV) from the Reoviridae family; infectious pancreatic necrosis virus (IPNV) from the Birnaviridae family; and grouper iridovirus (GIV) and frog virus-3 (FV3) from the Iridoviridae family. The first factor affecting the transmission of fish viruses examined in this thesis is the immune defense of host species. In this work, infections of marine VHSV-IVa and freshwater VHSV-IVb were studied in two rainbow trout cell lines, RTgill-W1 from the gill epithelium, and RTS11 from spleen macrophages. RTgill-W1 produced infectious progeny of both VHSV-IVa and -IVb. However, VHSV-IVa was more infectious than IVb toward RTgill-W1: IVa caused cytopathic effects (CPE) at a lower viral titre, elicited CPE earlier, and yielded higher titres. By contrast, no CPE and no increase in viral titre were observed in RTS11 cultures infected with either genotype. Yet in RTS11 all six VHSV genes were expressed and antiviral genes, Mx2 and Mx3, were up regulated by VHSV-IVb and -IVa. However, replication appeared to terminate at the translational stage as viral N protein, presumably the most abundant of the VHSV proteins, was not detected in either infected RTS11 cultures. In RTgill-W1, Mx2 and Mx3 were up regulated to similar levels by both viral genotypes, while VHSV-IVa induced higher levels of IFN1, IFN2 and LGP2A than VHSV-IVb. The second part of the thesis examined the ability of two Ranaviruses, GIV and FV3, to infect non-host fish cells. This is referred to as cellular tropism and is one of many host-virus interaction events required to established successful infection in new organisms. Grouper iridovirus (GIV), belonging to the Ranavirus genus of the Iridoviridae family, was demonstrated to differentially express viral genes and induce apoptosis in three non-host fish cell lines rainbow trout monocyte/macrophage (RTS11), Chinook salmon embryon (CHSE-214) and fathead minnow Epithelioma papulosum cyprinii (EPC). These cells were challenged with GIV and virus entry into all three cell lines was confirmed by the expression of viral immediate early genes. The expression of the late major capsid protein gene was detected in CHSE-214 and EPC, but not in RTS11, suggesting an earlier termination in the viral replication cycle in RTS11. Approximately 12 h after infection with GIV, cell death was prominent in all three non-host cell lines. Death was later confirmed to be apoptosis by the presence of chromosomal DNA fragmentation and phosphatidylserine externalization. To determine whether apoptosis was protein related or gene expression related, the three cell lines were infected with heat-inactivated GIV and UV-treated GIV (GIVUV). The heat inactivation abolished apoptosis in all three cell lines, but each cell line responded differently to GIVUV. Relative to GIV, GIVUV caused no apoptosis in CHSE-214, decreased apoptosis in RTS11, and increased apoptosis in EPC. These results suggest that early GIV gene expression was needed for apoptosis in CHSE-214 but impeded apoptosis in EPC. At the cellular level, only EPC was a permissive host as EPC was the only cell line of the three capable of producing a moderate increase in virus titre. The three non-host cell lines present a good system for potentially identifying different components of GIV-induced apoptotic pathways in future studies. Rainbow trout are not highly susceptible to frog virus 3 (FV3) induced diseases, and had been suggested to be a potential carrier for the virus. To determine which rainbow trout cell types are permissive for FV3 and act as a potential source for virus replication in vivo, the ability of rainbow trout cell lines from gonads (RTG-2), skin (RTHDF), liver (RTL-W1), gills (RTgill-W1), intestine (RTgut-GC) and spleen (RTS11), and primary leukocyte cultures from peripheral blood (PBL) and head kidney (HKL) to support FV3 infection was examined. RTG-2 supported a moderate level of FV3 replication while viral replication in RTL-W1 was minimal. The rest of the cell lines did not support viral replication but all succumbed to the infection and were killed by FV3. Lymphocyte-like cells from PBL and HKL were not killed by FV3 while macrophage-like cells were. Most of the cell lines died by an apoptosis-independent mechanism, presumably necrosis, while the monocyte/macrophage cell line, RTS11, died by an apoptosis-dependent mechanism. In addition, neoplastic macrophage-like human U937 cell line, and T lymphocyte-like PEER cell line were also infected with FV3 to compare their response to that of rainbow trout immune cells. U937 cells were killed by FV3 in an apoptosis-dependent manner; however, PEER T cells did not die from FV3 infection, a result similar to the lymphocyte-like fraction of rainbow trout PBL and HKL. In summary, most rainbow trout cell lines do not support significant FV3 replication; furthermore, cells of the lymphocyte origin appeared refractory to FV3 induced cell death while those of macrophage origin underwent apoptosis as a response to FV3. The last factor affecting the transmission of aquatic viruses examined in this thesis is the persistence of viruses in the aquatic environment. Virus persistence is influenced by natural environmental factors such as temperature, pH, desiccation and salinity, but the often unexplored anthropogenic factors can play a role. Therefore, the focus of this section was on the effect of one particular anthropogenic substance, Corexit 9500, on the infectivity of aquatic viruses with different physical characteristics. The effect of Corexit 9500, a dispersant used to clean up oil spills, on invertebrates, lower vertebrates, birds and human health have been examined but there is a significant lack of study on the effect of this dispersant on aquatic viruses. In this study, the effect of Corexit 9500 on four aquatic viruses of different structural composition was examined. Corexit 9500 reduced the titre of the enveloped viral hemorrhagic septicemia virus (VHSV) at all concentrations (10% to 0.001%) examined. The titre of frog virus 3 (FV3), a virus with both enveloped and non-enveloped virions, was only reduced at the high Corexit 9500 concentrations (10% to 0.1%). Corexit 9500 was unable to reduce the titre of non-enveloped infectious pancreatic necrosis virus (IPNV), but enhanced the titre of chum salmon reovirus (CSV) by 2-4 logs. With the ability to inactivate enveloped viruses and possibly enhance some non-enveloped viruses, Corexit 9500 has the potential to alter the aquatic virosphere.

Temperature represents a major driving force in biology as it influences essential functions across multiple levels of biological organization. The role of temperature is especially important for ectothermic animals, whose biotic processes are dependent on both body and environmental temperature. Assessing the relationship between temperature and organismal performance represents an important research direction as temperatures continue to warm under anthropogenic climate change. Chapters two and three are focused on a recently colonized population of the invasive Mediterranean House Geckos at the northern edge of their invasion front. These chapters examine the ecological and physiological factors that enable these lizards to persist in a cooler and more temperate environment than their native range. The thermal breadth of a reptile greatly influences its ability to tolerate a thermally variable environment, particularly when environmental options are limited for behavioral thermoregulation. These chapters explore the thermal performance of this species, and the results show that the eurythermality of these geckos promotes their rapid colonization of novel environments despite experiencing prolonged periods of cool temperatures. Chapters four, five, and six, by contrast, shift focus to larval amphibians to explore the constraints and factors underlying plasticity in acclimation to temperature extremes. As habitats continue to warm with climate change, ectotherms with limited capacity to thermoregulate, such as larval amphibians in shallow ponds, will be under a heightened threat of heat stress and mortality. Resultantly, identifying different factors that can increase organismal heat tolerance would reduce the risk of overheating and promote survival. Chapters four, five, and six explore this topic by measuring the critical thermal maximum (CTmax) of larval wood frogs. Chapter four focuses on the tradeoff between basal CTmax and plasticity of CTmax and its consequences for how a larval anuran responds to an acute heat shock. Chapter five examines the role a viral pathogen, ranavirus, has on larval CTmax. Surprisingly, a lethal dose of ranavirus did not reduce CTmax which goes against the common pattern of pathogenic infections lowering host heat tolerance. Lastly, chapter six explores the relationship between the gut microbiota and host CTmax with a particular focus on cross-species microbiota transplants. In line with our prediction, transplanting the gut microbiota of a heat-tolerant donor species promoted greater CTmax in the heat-sensitive recipient species.

World-wide amphibian declines sparked concern and encouraged investigation into potential causes beginning in the 1980’s. Infectious disease has been identified as one of the major potential contributors to amphibian declines. For example, Ranavirus has caused amphibian die-offs throughout the United States. Investigators isolated Ranavirus from dead or moribund amphibians during large-scale die-offs of amphibians in the Cades Cove area of Great Smoky Mountains National Park in 1999-2001. In 2009, after nearly a decade without follow-up monitoring, I undertook an investigation to determine if the virus persisted in the area, and if so, to assess spatial, temporal, and taxonomic patterns in prevalence. Three amphibian breeding ponds, including Gourley Pond, the site of these earlier mortality events, were monitored for Ranavirus during the 2009 amphibian breeding season. A peak in prevalence occurred at Gourley Pond corresponding to a massive amphibian die-off. Prevalence varied among three different taxonomic groups during this mortality event with the highest prevalence, 84%, detected in larval Ambystomatids, 44.4% prevalence in adult Newts, and no virus detected in adult Plethodontids. I did not detect virus at either of the other two breeding ponds despite equivalent sampling effort, similar community composition, and close proximity to Gourley Pond. These results suggest that the severity and spatial extent of Ranavirus in Cades Cove remains unchanged since its initial detection a decade ago. Also, despite the observed massive die-offs there is no evidence of local amphibian extinction at Gourley Pond.

World-wide amphibian declines sparked concern and encouraged investigation into potential causes beginning in the 1980’s. Infectious disease has been identified as one of the major potential contributors to amphibian declines. For example, Ranavirus has caused amphibian die-offs throughout the United States. Investigators isolated Ranavirus from dead or moribund amphibians during large-scale die-offs of amphibians in the Cades Cove area of Great Smoky Mountains National Park in 1999-2001. In 2009, after nearly a decade without follow-up monitoring, I undertook an investigation to determine if the virus persisted in the area, and if so, to assess spatial, temporal, and taxonomic patterns in prevalence. Three amphibian breeding ponds, including Gourley Pond, the site of these earlier mortality events, were monitored for Ranavirus during the 2009 amphibian breeding season. A peak in prevalence occurred at Gourley Pond corresponding to a massive amphibian die-off. Prevalence varied among three different taxonomic groups during this mortality event with the highest prevalence, 84%, detected in larval Ambystomatids, 44.4% prevalence in adult Newts, and no virus detected in adult Plethodontids. I did not detect virus at either of the other two breeding ponds despite equivalent sampling effort, similar community composition, and close proximity to Gourley Pond. These results suggest that the severity and spatial extent of Ranavirus in Cades Cove remains unchanged since its initial detection a decade ago. Also, despite the observed massive die-offs there is no evidence of local amphibian extinction at Gourley Pond.

No description available.

<p><a>Natural enemy ecology strives to integrate the fields of disease ecology and community ecology to forge a broader understanding of how pathogens and predators structure communities. To advance this field, we need a greater emphasis on: 1) quantifying pathogen-mediated effects on community structure and comparing these effects to those observed with predators and 2) determining the interactive effects of combined natural enemies on communities. I conducted a mesocosm experiment designed to assess the individual and combined effects of predators (dragonfly larvae and adult water bugs) and a pathogen (ranavirus) on a larval amphibian community. Additionally, I conducted laboratory experiments to assess whether ranavirus exposure increases the vulnerability of tadpoles to predation. In my laboratory experiments, I found that virus exposure increased predation rates with dragonflies, but not water bugs. For tadpoles in the dragonfly treatments, the probability of survival for virus-exposed tadpoles was 66-77% lower compared to unexposed tadpoles. This data suggests that predators may selectively remove infected individuals from the population, which can enhance the magnitude of the healthy herds effect. I found that the risk level of the predators largely explained effects on the community. For instance, high-risk dragonflies reduce overall survival to 30% whereas low-risk water bugs only reduced survival to 67%. Additionally, I found that virus reduce survival to 62%, which was comparable to effect of the low-risk predator. Interestingly, all three natural enemies influenced community structure (i.e. species relative abundance) in unique ways. These results demonstrate that pathogens can have effects similar to predators on communities, and that natural enemy identity is important when considering impacts on community structure. When predators were combined with the virus, I found that mortality was relatively unchanged from the predator-only treatments suggesting less than additive effects of combined natural enemies. This result was driven by the healthy herds effect; the presence of dragonflies reduced overall infection prevalence in the community to 7% compared to 30% in the virus-only treatment. This effect was observed in the water bug treatments, to a lesser degree, suggesting that predator risk or efficiency contributes the magnitude of the effect. Collectively, my work demonstrates the importance of examining the individual and combined effects of natural enemies on ecological communities.</a></p>

For over three decades, amphibian populations have been declining across the globe. Emerging infectious diseases are responsible for some of these declines. Ranaviruses have caused die-offs in wild amphibian populations on 4 continents, in 5 Canadian provinces, and in over 25 U.S. states. In order to understand host-pathogen dynamics, it is critical to establish baseline information on species susceptibility and the effects of natural stressors. The goal of my thesis research was to quantify the effects of anuran development and exposure to invertebrate predators on species-specific susceptibility to ranavirus. My experiments were designed in factorial arrangements, and consisted of exposure to ranavirus during different developmental stages or with and without predator cues in a controlled environment. I found that exposure to invertebrate predator cues did not increase susceptibility to ranavirus for 4 anuran species tested. Susceptibility differed among embryo, hatchling, larval and metamorph stages, but trends differed among species and did not follow predictions based on Xenopus laevis immune function. Low susceptibility during the embryo stage was the only consistent development result among species, perhaps owing to protective qualities of the vitelline membrane or mucoidal capsules surrounding the embryo. Across 7 anuran species tested, mean mortality rates ranged from 5 – 100%, with Lithobates sylvaticus and Scaphiopus holbrookii most susceptible. I found that infection rates and viral load were correlated with mortality rates, thus these variables are good indicators of susceptibility to ranavirus. My results indicate that ranaviruses can cause catastrophic natural mortality in some anuran species, and likely play a significant role in local population dynamics. For highly susceptible species, ranaviruses could cause local extirpations that lead to species declines. More information is needed on the role of natural (e.g., co-infection, competition) and anthropogenic stressors in driving ranavirus epizootic events. I encourage natural resource agencies to initiate ranavirus surveillance programs, especially for rare species and fragmented populations. Future studies should take an immunogenetic approach to identifying mechanisms driving susceptibility. Identifying mechanisms associated with ranavirus emergence is fundamental to developing science-based conservation strategies.