Characterizing the epidemiology of ophidiomycosis in North American snakes through field studies, modeling, pathogen genomic analysis, and treatment trials

Haynes, Ellen

Permalink

https://hdl.handle.net/2142/110675 Copy

Description

Title

Characterizing the epidemiology of ophidiomycosis in North American snakes through field studies, modeling, pathogen genomic analysis, and treatment trials

Author(s)

Haynes, Ellen

Issue Date

2021-04-15

Director of Research (if dissertation) or Advisor (if thesis)

Allender, Matthew

Doctoral Committee Chair(s)

Allender, Matthew

Committee Member(s)

• Allan, Brian
• Fan, Tim
• Smith, Rebecca
• Baker, Sarah
• Miller, Andrew

Department of Study

School of Integrative Biology

Discipline

Ecol, Evol, Conservation Biol

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• ophidiomycosis
• epidemiology
• snake
• fungus

Abstract

The enormous impacts of human activities on global ecosystems pose a threat to the health of people, animals, and the environment. It is critical to study wildlife diseases in order to protect global biodiversity and promote global health, as notably highlighted by the effects on society during the COVID-19 pandemic. Emerging fungal diseases have profound impacts on people, agricultural crops, domestic animals, and wildlife. Ophidiomycosis, caused by the keratinophilic fungus Ophidiomyces ophidiicola, is an emerging infectious disease of wild and managed snakes worldwide. All snake species appear to be susceptible and it is a particular threat to species of conservation concern. The purpose of this dissertation was to address the gaps in current knowledge of ophidiomycosis in the United States, including the full range of snake species and geographic areas impacted by the disease, how disease and pathogen prevalence vary based on host and environmental factors, the impact of O. ophidiicola genomic diversity on disease epidemiology, and the effectiveness of specific treatment options. The specific objectives were to use fieldwork, clinical trials, molecular approaches, and statistical modeling to examine host, agent, and environmental factors that impact development of disease and characterize the pharmacodynamic and pharmacokinetic profiles for a single medication and delivery system in diseased snakes. Ophidiomycosis surveillance was conducted in a variety of snake species in southeast Georgia, USA and in Lake Erie watersnakes (Nerodia sipedon insularum, LEWS) in Ottawa and Erie Counties, Ohio, USA. Snakes were captured during visual encounter surveys and visually inspected for skin lesions suggestive of ophidiomycosis, then a body swab was collected to detect the presence of O. ophidiicola using qPCR. Each snake was assigned to an ophidiomycosis category based on the presence of skin lesions and O. ophidiicola. Logistic regression models were used to predict lesion presence, O. ophidiicola detection, and ophidiomycosis category based on individual, spatial, and temporal factors. For LEWS, models were also used to test for associations between lesion and fungal presence and a variety of climate, demographic, and environmental factors. Next, a multiplex qPCR-based genotyping assay was designed to differentiate between clades of O. ophidiicola, which were defined based on whole-genome sequencing conducted at the National Wildlife Health Center. Primer-probes sets were designed to target conserved and divergent areas between clade-representative genomes, such that each clade had a specific amplification pattern. Using the Fluidigm Access Array to conduct multiplex qPCR, the amplification pattern of a particular O. ophidiicola DNA sample across these primer-probes was used to assign it to a clade. This approach was validated using samples of known clade and applied to swab-extracted O. ophidiicola DNA samples from multiple snake species, states, and years. Multinomial logistic regression modeling was used to predict clade based on snake taxonomic group, state of origin, and year of collection. Finally, a controlled clinical trial was conducted using terbinafine nebulization to treat wild-caught Lake Erie watersnakes with natural ophidiomycosis. Snakes received terbinafine nebulization, saline nebulization, or no treatment, and response to treatment was measured using physical exams to document clinical disease severity, as well as qPCR of body and lesion swabs to measure O. ophidiicola quantity. Mixed models were used to analyze response to treatment based on initial disease severity, treatment group, year of the study, and length of treatment. Plasma and skin concentrations of terbinafine were measured using high-performance liquid chromatography (HPLC). A total of 786 free-ranging snakes representing 34 species and four families were examined and sampled in southeast Georgia. Of these, 27.5% had skin lesions and 13.3% were positive for O. ophidiicola DNA. This was the first report of O. ophidiicola in five of the 22 species that were qPCR positive and multinomial logistic regression modeling indicated that eastern indigo snakes (Drymarchon couperi) had a higher relative risk of apparent ophidiomycosis (lesions present and qPCR positive). The best models predicting qPCR result and ophidiomycosis category included individual factors and excluded temporal and spatial factors. A total of 837 Lake Erie watersnakes were evaluated across five surveillance sites between 2017 and 2020. The prevalence and odds of skin lesion presence, qPCR detection of O. ophidiicola, and ophidiomycosis categories varied between sites and years. Snakes were more likely to have skin lesions at lower temperatures and O. ophidiicola was more likely to be detected in snakes at sites with open space development and low-intensity development. The presence of emerging herbaceous wetlands, urban land change, and certain soil types increased the odds of both lesion presence and qPCR detection of O. ophidiicola. The multiplex qPCR-based genotyping assay consisted of 11 primer-probe sets and, compared to full-genome sequencing, correctly assigned 77.3% of the samples with known clade (Cohen’s kappa = 0.682, 95% CI: 0.527 – 0.837). Swab-extracted O. ophidiicola DNA samples from across the United States were assigned to five different clades, including three of the six pre-established clades and two newly defined clades, which likely represent recombinant strains of O. ophidiicola. U.S.state of collection was the only significant predictor of clade in multinomial logistic regression models and clade was not associated with disease severity in LEWS. Finally, terbinafine nebulization resulted in external molecular resolution of disease in 29.2% of treated snakes following three to six months of daily nebulization, compared to 5% of snakes receiving saline nebulization and 11.1% of snakes receiving no treatment. Terbinafine nebulization did not significantly decrease clinical disease, as measured by disease severity scores, but significantly reduced fungal quantity after three or more 30-day courses of treatment. Daily terbinafine nebulization was found to produce concentrations above the minimum inhibitory concentration for O. ophidiicola in both plasma and skin. This treatment may be most useful in snakes from managed populations that can be treated for several months, rather than wild snakes who are not releasable after multiple months in captivity. This work provides a great deal of new information about the epidemiology of ophidiomycosis, but many questions remain. Additional work to improve surveillance, conduct experimental infection trials, and use molecular techniques to evaluate host and pathogen gene expression will support snake health and conservation, as well as informing the management of emerging fungal diseases affecting wildlife, humans, and domestic animals.

Graduation Semester

2021-05

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/110675

Copyright and License Information

Copyright 2021 Ellen Haynes

Owning Collections