Eukaryote cells are the products of a complex history of interspecies interactions, with some organelles now known to have arisen through the endosymbiosis of prokaryotic cells. Whilst these organelles 'evolved' from the original endosymbionts, their evolution has not stopped within the modern eukaryote. Trichomonas vaginalis is protozoan parasite, with an unusual cellular biology, this species appears to lack peroxisomes, and instead of mitochondria has divergent organelles called hydrogenosomes. The hydrogenosomes of Trichomonas represent one of a growing number of highly divergent organelles, which are present in species throughout the eukaryotic kingdom. In this investigation the hydrogenosomes of Trichomonas are investigated with respect to their preprotein membrane translocases, a multi-membraned molecular system essential for the organelle's maintenance and biogenesis. The complexity and components of such a system are unlikely to arise duplicated from a separate organellogenesis, and thus the system's architecture is expected to indicate this organelle's descent. To reveal the structure of this system two different practical approaches were used to determine the biology of the hydrogenosome. The first builds upon work to characterise translocase kinetics and probes the nature of the hydrogenosomal membrane translocon directly. The second explores the use of candidate translocases determined from the Trichomonas genome through bioinformatic analysis and their development into practical investigation through the expression of tagged proteins in transformant T. vaginalis. These transformants were used to visualise a population of membrane proteins in situ within the hydrogenosome by immunofluorescence microscopy, and further to identify their associations and interactions within the hydrogenosomal membrane using protein biochemical methods. The data produced within this study are finally brought together to present a model for the Trichomonas hydrogenosomal preprotein import system, as well as the first molecular characterization of its translocase components.

Virulence (parasite- induced host fitness reduction) is thought to correlate positively with pathogen reproduction rates, but its relationship with pathogen transmission is likely to be determined by a trade- off between the costs and benefits of harming hosts. This project aims to investigate factors which affect host-parasite interactions and particularly those which may play a role in virulence evolution. In doing so, it describes experiments carried out using a monogenean ectoparasitic flatworm (Gyrodactylus arcuatus) and its three-spined stickleback (Gasterosteus aculeatus) host. Populations of this fish species experience a range of both environmental and ecological conditions. Such environmental heterogeneity has been found to drive changes in fish phenotypic traits such as morphology, behaviour, life history and physiology which may consequently affect parasite fitness. I demonstrate that under these conditions, parasite strains from different host populations evolve variation in virulence levels. I also reveal that coevolution in this host parasite system is likely to lead to local adaptation of G. arcuatus at the host population level. I show that competition between parasite species sharing a single host leads to increased parasite reproduction rates, but it shortens the infection time which may be due to earlier stimulation of host immune responses. I show that virulence is neither influenced by the population density, immunity (epidermal mucus), social behaviour of fish hosts nor the natural parasite life expectancy. Lastly, I find that virulence in this system is negatively influenced by the density of stickleback predators and positively associated with loch water pH. Taken together, these results suggest that in this host parasite system, both ecological and environmental factors which drive phenotypic changes in fish hosts may evolutionarily feedback to affect parasite virulence.