DISSERTATION THE USE OF METAGENOMIC SEQUENCING AS A TOOL FOR PATHOGEN DISCOVERY WITH FURTHER INVESTIGATION OF NOVEL REPTILIAN SERPENTOVIRUSES Submitted by Laura L. Hoon-Hanks Department of Microbiology, Immunology, and Pathology In partial fulfillment of the requirements For the Degree of Doctor of Philosophy Colorado State University Fort Collins, Colorado Summer 2019 Doctoral Committee: Advisor: Mark Stenglein Gregory Ebel Daniel Sloan Sushan Han Copyright by Laura L. Hoon-Hanks 2019 All Rights Reserved ABSTRACT THE USE OF METAGENOMIC SEQUENCING AS A TOOL FOR PATHOGEN DISCOVERY WITH FURTHER INVESTIGATION OF NOVEL REPTILIAN SERPENTOVIRUSES Infectious diseases play a significant role in the health of all organisms, from the largest mammals to the smallest bacteria. The consistent identification of pathogens has been propelled by generations of perceptive scientists and the development of important biological tools for microbe detection. In recent years, a new technology has surfaced that has changed the face of pathogen discovery: next generation sequencing and metagenomics. This sequencing technology allows for unbiased sampling of genetic material from all organisms within a given sample, independent of our pre-conceived suspicions of the pathogens most likely to be present. Therefore, metagenomic sequencing has become a powerful tool for investigating both known and newly emerging infectious diseases that evade classical methods of diagnosis. The preliminary goal of this dissertation was to present metagenomic-based pathogen discovery projects that highlight the benefits and limitations to this technology. First, we assessed meningoencephalitis of unknown origin (MUO) in dogs, a suspected autoimmune disease, but for which an underlying infectious agent remains a possibility. We did not detect any infectious agents associated with disease, further supporting the hypothesis of an autoimmune pathogenesis. Second, we assessed two cases of malignant catarrhal fever (MCF)-like disease in free-ranging mule deer. In the U.S., MCF is most commonly associated with ovine herpesvirus-2 infection, however, we identified caprine herpesvirus-2 (CpHV-2) in both cases. CpHV-2 is an uncommon cause of MCF and has not been previously described in mule deer. Therefore, metagenomics aided in the identification of a known but unexpected pathogen of disease. Lastly, we investigated an outbreak of granulomatous nephritis in a collection of seahorses. We identified a novel paramyxovirus and parvovirus, but no clear association between infection and disease was drawn for case and control animals. Therefore, additional examination of these viruses will be necessary to determine their association with disease. In this case, metagenomics provided a new thread of ii investigation but did not provide a definitive answer. This is a common outcome with metagenomic projects and highlights its use an investigatory tool upon which future experimentation can be based. The second goal of this dissertation was to further characterize a virus discovered by metagenomic sequencing through confirmation of disease causality and epidemiologic investigation of species susceptibility and disease. Recently, a novel serpentovirus was discovered by metagenomic sequencing in association with fatal respiratory disease in pythons. Investigations of this virus included genome characterization, tissue tropisms, viral nucleic acid localization to areas of disease, and potential mechanisms of pathogenesis. However, these studies only provided circumstantial evidence linking serpentovirus infection to disease. In attempt to confirm disease causality, we conducted experimental infections in ball pythons (Python regius) with a serpentovirus known as ball python nidovirus (BPNV). Three ball pythons were inoculated orally and intratracheally with cell culture isolated BPNV and two were sham inoculated. Antemortem choanal, oroesophageal, and cloacal swabs and postmortem tissues of infected snakes were positive for viral RNA, protein, and infectious virus by qRT-PCR, immunohistochemistry, western blot and virus isolation. Clinical signs included oral mucosal reddening, abundant mucus secretions, open-mouthed breathing, and anorexia. Histologic lesions included chronic- active mucinous rhinitis, stomatitis, tracheitis, esophagitis and proliferative interstitial pneumonia. Control snakes remained negative and free of clinical and histologic disease throughout the experiment. Our findings establish a causal relationship between serpentovirus infection and respiratory disease in ball pythons and shed light on disease progression and transmission. Following an established disease causation, we sought to characterize the epidemiologic features of serpentovirus infection in captive snakes. We performed a large longitudinal study of serpentovirus infection with the aim of generating an improved understanding of the course and clinical outcome of infection, assessing the susceptibility of different types of snakes to infection and disease, surveying viral genetic diversity, and defining effective management practices. We collected 777 antemortem choanal or oral swabs from 639 snakes from 12 collections across the United States. These included snakes from 62 species, 28 genera, and 6 families: Pythonidae (N=414 snakes), Boidae (79), Colubridae (116), iii Lamprophiidae (4), Elapidae (12), and Viperidae (14). Infection was more common in pythons (40% infected), and in boas (10% infected) than in other types of snakes, which had infection rates of 0-1%. Pythons were more likely to exhibit signs of respiratory disease: 85 of 144 infected pythons displayed signs of mild to severe respiratory disease, but only 1 of 8 infected boas did. Infected snakes had increased mortality: the percentage of infected pythons that died of respiratory disease during the course of this study ranged from 45 to 75% in two different collections. Genetically distinct serpentoviruses, including divergent genotypes, were detected in pythons, boas, and colubrids, suggesting a potential species barrier between snakes of different families. In some cases, nearly identical viruses were found to infect pythons of different genera, indicating host plasticity amongst python serpentoviruses. Older snakes were more likely to be serpentovirus-positive, but males and females were equally likely to be infected. Neither age nor sex were statistically associated with disease. Longitudinal sampling of forty pythons over 28 months revealed serpentovirus infection is chronic, and definitive evidence of viral clearance was not observed. We followed offspring of infected parents and found that vertical transmission either does not occur or occurs at a much lower efficiency than horizontal transmission. Management strategies that could reduce the spread of serpentoviruses include rapid physical separation of infected snakes, testing snakes multiple times over several months to ensure true negative results, and rigorous quarantine when introducing pythons into a collection. Our findings regarding serpentovirus infection in snakes provide a solid foundation for future investigations of related serpentoviruses. In addition to snakes, serpentoviruses have also been identified in lizards and turtles in Australia. Serpentovirus infection in lizards has been associated with a respiratory disease similar to that seen in snakes, but disease causation is only circumstantial. We further develop our understanding of serpentovirus-associated disease in lizards by describing novel serpentoviruses in a collection of veiled chameleons with respiratory disease-associated mortalities. Over the course of 1 year, a reptile collection lost 24 of 30 veiled chameleons. Clinical signs included wheezing and vertical head tilting with gasping, increased mucus in the oral cavity, anorexia, and reduced water intake. Histopathology included severe proliferative and mucinous interstitial pneumonia, tracheitis, and rhinitis. Tissue pools from iv 3 chameleons were evaluated by metagenomic sequencing and two novel serpentoviruses were detected; coinfection with both serpentoviruses was found in 1 chameleon. Antemortem oral swabs were sampled from the remaining 6 live chameleons in the collection and were tested for serpentovirus infection by PCR. Five of 6 chameleons were serpentovirus positive, one of which was exhibiting mild signs of respiratory disease. Genomic assessment revealed approximately 50% nucleotide identity between each of the chameleon serpentoviruses, as well as to other reptile serpentoviruses in snakes, lizards, and turtles. These findings further support the hypothesis that serpentoviruses cause disease in lizard species. In the future, serpentovirus should be considered a possible differential for respiratory disease in chameleons. v ACKNOWLEDGMENTS There is an immense community of people that I would like to thank for the support and encouragement I have received throughout this process. First, I would like to thank Craig Kuchel for instilling in me a love of science at an early age. I would like to thank the Samuels’ laboratory at the University of Montana for creating such a wonderful and encouraging environment in which to learn and appreciate research. Their continued support has been immensely important to my growth as a scientist. I would like to thank Dr. Paula Schaffer, Dr. Chad Frank, Dr. EJ Ehrhart, Dr. Gary Mason, Dr. Colleen Duncan, and the pathology residents at CSU for making my residency a fruitful and fulfilling experience. I wouldn’t