Study of gastrointestinal nematodes co-infecting feral Soay sheep on St Kilda

The unmanaged, feral Soay sheep population on St Kilda has survived for hundreds of years, despite enduring potentially deleterious gastrointestinal nematode co-infections. Co-infections with multiple nematode species are ubiquitous in feral and managed ruminants. Within these mixed burdens, the different species may vary in their pathogenicity, epidemiology and clinical presentation. Elucidating the diversity of different parasite species in a host, rather than studying them as a homogenous group, is a prerequisite to understanding host-parasite interactions. The primary aim of this thesis is to explore and validate non-invasive conventional and next generation molecular parasitological methods to identify and quantify mixed-species infections in feral hosts. Seasonal patterns of gastrointestinal nematode parasitism in the Soay sheep were investigated by faecal egg counts (FEC). Two FEC datasets were analysed: a large dataset, collected between 1988 and 2014, and counted by a modified McMaster method where each egg count represents 100 eggs per gram (epg); and a smaller dataset collected over one year between April 2015 and April 2016, counted by cuvette salt floatation method with a lower detection limit of 1 epg. FEC generally declined with increasing sheep age, until the animals became geriatric (8 years +). Seasonal FEC patterns in females generally followed a decline over the year starting in Spring. FEC were generally higher in males, regardless of their age category, with little seasonal variation between summer and winter once they became adults (3 years +). Monte Carlo simulations were run in order to compare the effects of different detection limits between the FEC methods detecting eggs in faeces to 100 epg and to 1 epg. The simulations suggest that the method with the detection limit of 50 – 100 epg over-estimates the true egg counts within the samples (at high counts), resulting in data that is highly negatively skewed and with an inflated mean. Despite within-year variation of egg counts, both datasets resulted in overall similar seasonal and host patterns. A shorter study with fewer replicates may benefit from a FEC method with a lower detection threshold. The eggs of most strongyle nematodes are morphologically similar, hence whilst FEC can identify general trends in parasitology, the method provides limited information about the proportions of mixed gastrointestinal nematode species burdens of the Soay sheep. The development of advanced molecular methods for the in-direct genus or species-specific diagnosis of strongyle infections in ruminants has negated many of the issues arising from traditional parasitological techniques. Chapter 3 compared two molecular methods; a semi automated multiplex-tandem PCR (AusDiagnostics™) with ITS-2 rDNA next-generation amplicon sequencing (nemabiome assay), to identify species (presence/absence), and quantify the relative proportion (%) of ovine strongyle species in naturally infected samples. This chapter provides the first, and preliminary, comparison of the sensitivity and quantitative ability of both methods. There was good agreement between both molecular tests in determining the presence/absence of species within a sample, but the correlation in their ability to quantify relative proportions of the species present was moderate (C. ovina, R2 = 0.6096) to poor (Trichostrongylus spp., R2 = 0.2334). AusDiagnostics™ characterises a priori strongyle species present, which partially suits a diagnostic tool for managed ruminant systems. The nemabiome assay proved to be the most effective method for identifying and quantifying previously unknown strongylid species, such as those associated with a feral host. The sensitivity, bias, and repeatability of the nemabiome assay was tested for the strongylid species identified in the Soay sheep of St Kilda (Chapter 4). A correction factor was calculated for each species in order to reduce potential species-specific sequencing bias. This was subsequently applied to the field data presented in Chapter 5. Chapter 5 provides an epidemiological survey of the strongyle nematode species co-infecting the Soay sheep of St Kilda. Teladorsagia circumcincta, Trichostrongylus axei, Trichostrongylus vitrinus, Chabertia ovina and Bunostomum trigonocephalum were identified by the nemabiome assay. The study highlights epidemiological trends in the Soay sheep that had not previously been identified using conventional gross and molecular parasitological methods. There were seasonal, age and sex differences in species proportions; whereby trends appeared to correspond with the dynamic life-history of the sheep. Trends included a year round prevalence of all species in lambs (males and females) to the age of one year old, high levels of T. circumcincta over summer and winter in females older than 8 years, and no clear seasonality to T. axei which persisted, at low levels, over all months in all sex/age groups sampled. However, when relative species proportions were adapted to account for average FEC, the impact of the trends seen was minimised; the adequate representation for individual nematode species relied on a high faecal egg count. This is the first study to use the nemabiome deep amplicon sequencing approach to characterise seasonal patterns in different co-infecting gastrointestinal nematodes in feral sheep. Additionally, it highlights the flexibility of the nemabiome assay as a viable non-invasive tool for parasitological surveys of wild animals. Better knowledge of the epidemiology of different co-infecting gastrointestinal nematodes in the ancient feral St Kilda Soay sheep population could help to explain the impact of grazing management and anthelmintic drug treatments in managed sheep flocks, aiding in the development of sustainable control strategies.