Quantification of host-parasite interactions: sheep and their nematodes

The objective of this dissertation is to use modelling and statistical approaches to expand our knowledge of the immune responses against gastrointestinal nematode infections, to assess the impact of nematode infection, and to use our improved knowledge to examine novel means of selective breeding in farm animals (sheep) as a control strategy. To expand the knowledge of the host immune response against infection, Chapter 2 of this dissertation focuses on immunoglobulin A (IgA), an antibody that binds nematode molecules, and its transfer through the body from the abomasal mucus (i.e. site of infection, where it is produced) to the blood plasma, where it is typically measured. These findings have been published in Parasitology (Prada Jimenez de Cisneros et al., 2014a). The implications of low levels of infection in adult milking ewes, which are more resistant than lambs, were also studied. A relationship is generated between infection levels using parasitological data and production data. There were however limitations in the dataset, which are discussed at the end of Chapter 3. Parasite resistance in adults sheep at low levels of infection was also studied, especially since the most common parasitological marker of disease is the faecal egg count (i.e. number of nematode eggs in the animals faeces) which is subject to substantial measurement error, among other limitations. Chapter 4 analyses a dataset of adult animals with low infection levels using a zero inflated binomial model (ZINB) and extends the model by including other evidence of parasite resistance to discriminate between exposed and unexposed animals. To examine selective breeding, an individual-based data-driven immunogenetically explicit mathematical model was developed. One application of this model is to compare the efficacy of two selective breeding schemes, each based on a different marker for disease, namely faecal egg counts and plasma IgA. This work has been published in Journal of the Royal Society Interface (Prada Jimenez de Cisneros et al., 2014b). The model can be extended to create a distribution for the variation in larval intake that best fits the field data. This allows the partitioning of the variation in adult worm burden into different components. The purpose is to quantify the contribution of the immune response and larval intake to determine which of the two accounts for more of the variation in the level of infection. The model can be also extended to explore selection schemes in the two components of the immune response (i.e. namely the IgA mediated and IgE mediated immune response) and estimate animal size at the end of the grazing season