One of the greatest proximate threats facing biodiversity is habitat change as a result of the combined effects of agricultural development and livestock grazing. Extensive livestock systems are increasingly competing with wildlife for access to land and natural resources in African rangelands. Ethiopia has the highest numbers of livestock in Africa with most livestock production taking place in highland systems. The Bale Mountains contain the largest extent of afroalpine habitat in Africa and are the most important conservation area in Ethiopia as one of thirty-four Conservation International Biodiversity Hotspots. The Bale Mountains National Park was established forty years ago to protect the endemic, and rodent-specialist Ethiopian wolf (Canis simensis) and the afroalpine habitats upon which they depend. I use a 21 year time-series of livestock counts in the park to detect changes in the Oromo traditional livestock production system or godantu. I show that a seasonal transhumant livestock production regime, akin to godantu, only appears to persist in the Web valley today and I also find some evidence that the type of livestock is changing with smaller stock, such as sheep and goats, being more frequently kept. Despite reportedly growing numbers of households in the Web valley, I do not detect an overall increase in livestock numbers, suggesting that the area may have reached its carrying capacity in terms of livestock. In contrast, the number of livestock on the Sanetti Plateau increased over the monitoring period, including the remote and inhospitable western section of the plateau. Furthermore, I detect an increase in the risk of contact and disease transmission between Ethiopian wolves and free-roaming dogs in wolf optimal habitats. I subsequently estimate, through the use of transects, the densities of livestock in the Web valley, Morebawa and Sanetti and calculate the ratio of observed to maximum sustainable livestock density (based on rainfall and vegetation productivity). The conservative results suggest that the Web valley and Morebawa are overstocked given the rangeland predicted productivity for those areas, and that the problem of overstocking is at its worst during the dry season. I then establish critical relationships between vegetation conditions, livestock grazing pressure and rodent populations (Tachyoryctes macrocephalus, Lophuromys melanonyx and Arvicanthis blicki) in the afroalpine, under both natural and controlled experimental conditions. Rodent biomass declines as the livestock index increases along a natural grazing gradient, but rodents‟ body condition, use of the habitat or breeding ecology and most vegetation variables measured do not significantly vary across the livestock grazing gradient. Rodent populations do respond to the experimental removal of livestock inside exclosures, although this response is site and species-specific, and I find some evidence that these responses are concurrent with changes in the vegetation. Grazed plots have higher plant species richness and diversity compared to the exclosure sites which harbour reduced bareground cover and taller vegetation. Differences among rodent species in their responses to grazing may be mediated by interaction between the direct effects of grazing on habitat quality and species-specific habitat requirements, although the specific mechanism of this interaction could not be tested. Livestock may negatively affect rodents by increasing their predation risk (through removing vegetation cover), by reducing the soil suitability for maintaining burrow systems (through trampling) or, less likely, by competing for food resources. A series of simple dynamic food chain models are developed to explore the interactions between Ethiopian wolves, rodents and vegetation and how they may be affected by increasing levels of livestock grazing. I also explore how predictions made about these trophic dynamics are affected by the type of functional response linking the different levels. The models suggest that the pastures of the Web valley and Morebawa are likely to be incapable of maintaining wolves at current livestock densities, while the Sanetti plateau may be able to sustain only slight increases in livestock density before tipping into a trophic configuration unable to sustain wolves. This model is a first step in assessing the seriousness of conflict between pastoralists and wildlife in BMNP. Resource selection functions are developed and validated for cattle and sheep/goats grazing in the Bale afroalpine in an attempt to understand some of the drivers behind the heterogeneous use of the landscape by livestock. Habitat use by livestock is focused on lower-lying pastures in the vicinity of water sources (rivers or mineral springs). Only cattle strongly select for/against particular vegetation types. The probability of habitat use is also linked to the distance from the nearest villages. The models developed highlight a strong association between livestock use and rodent biomass in Morebawa and Eastern Sanetti, suggesting that livestock grazing poses a threat to Ethiopian wolves‟ persistence in marginal habitats in which rodent availability is already limited. Furthermore, the concentration of livestock around water sources has serious implications for the degradation of the park‟s hydrological system and the livelihoods and food security of the millions of people living in the dependent lowlands. Integrating research and practice is a fundamental challenge for conservation. I discuss how the methodological tools developed and the insights gained into the dynamics of the afroalpine system can contribute to the management of livestock grazing pressure in Bale Mountains National Park and highlight gaps in the knowledge of the afroalpine ecosystem where more research is needed.

Footrot (FR) is a highly infectious and debilitating disease of sheep, which has a significant economic impact on the sheep farming industry, in the UK and worldwide and causes significant suffering of sheep. Despite some recent advances, FR remains a scientifically challenging disease to understand. To help improve our understanding of disease pathogenesis, two culture-independent techniques were developed to examine the microbial succession events between the causative agent, Dichelobacter nodosus and an accessory agent, Fusobacterium necrophorum, the latter also postulated to be involved in disease initiation. The two populations were monitored in relation to disease initiation and progression during a longitudinal study and disease presentation in tissue biopsies (in situ). Finally, the distribution of these two species of bacteria in the environment was examined to highlight possible sources of infection. The work in this thesis has demonstrated that FR is a disease where expression is related to D. nodosus load present in the ovine interdigital space. D. nodosus (rpoD) load increased from that on a healthy foot to one presenting with interdigital dermatitis (ID) and feet with a higher D. nodosus (rpoD) load were more likely to go on to develop FR one week later. FISH analysis of the D. nodosus population present within the epidermis also revealed similar findings; D. nodosus cell counts increased during stages of ID, but the organism was less frequently detected in biopsies from feet with FR. Suggesting that ID might be the most infectious stage of the disease process. A fact that needs to be highlighted to farmers to encourage treatment at this stage of disease. In contrast, F. necrophorum (rpoB) load did not correlate with ID presentation or prior to the development of FR, but increased the week of FR onset. FISH analysis also revealed that F. necrophorum cell counts were higher in feet with FR than those with ID. It is possible therefore that F. necrophorum may thrive in the altered environment of a foot presenting with FR, possibly contributing to disease persistence and severity. Finally, both pathogens were detected in a range of environmental samples from a farm with endemic FR, highlighting possible sources of infection and material, which once contaminated with D. nodosus and F. necrophorum may contribute to the spread of FR. This study has provided an improved understanding of the microbial population dynamics involved in the development of ID and FR in sheep, which may have implications for control and treatment practices not only in the UK, but world-wide.

Four visits were made to 50 dairy farms in England and Wales between February 2003 and February 2004 and a further three visits to 42 of these farms between February 2005 and February 2006. At each visit the locomotion of all cows was scored (15,597 cows, 34,643 measures). Multivariable regression analyses were preformed to identify risk factors associated with increased lameness (poor locomotion). Factors associated with mean herd poor locomotion were dry cows kept in straw yards compared with cubicle houses, pregnant heifers kept with milking cows compared with dry cows in winter, passage way widths <3m compared with 2:,3m, a kerb height of ~15cm compared with >15cm, routine trimming of claws ofall cows by a claw trimmer or by- the farmer compared with no routine claw trimming, feeding maize silage to milking cows compared with other forage types, and the use of automatic scrapers compared with tractor scrapers in the cubicle house. Farmers recorded the lesions they observed while treating lame cows. Sole ulcer, white line disease and digital dermatitis were the three most frequently recorded lesions. The occurrence of sole ulcer, white line disease or digital dermatitis compared with having no lesion was used as the outcome variable in three multilevel binomial logistic regression models with month from calving nested within cow nested within farm. Risk factors associated with increased risk of sole ulcer were being housed on sparse bedding for four months or more, having grooved concrete floors on ~e farm, large herd sizes and parity numbers of four and above. Risk factors associated with increased white line disease were grooved concrete floors increasing parity number. Risk factors associated with increased digital dermatitis were grooved concrete floors and large herd sizes. A large clinical trial was used to assess the effect of intervening on known risk factors associated with lameness. Farmers in the treatment group received farm specific recommendations grouped under five targets areas which reflected the hypothesised aetiologies of sole ulcer, white line disease and digital dermatitis. There were small reductions in locomotion score and sole ulcer rate on treatment farms compared with control. These suggested that the hypotheses that reduced standing time and encouraged increased lying time through improved cow comfort may reduce sole ulcer were, at least in part, correct. There was no clear effect of treatment on the rate white line disease and digital dermatitis. It is likely that more than one of the following factors explain the lack of significant results from these initial investigations; poor recognition oflameness by farmers, insufficient uptake of recommendations, predisposition to lameness from previous lameness event and insufficiently specific lesion definitions.

There are numerous vaccines available for the control of infectious bronchitis virus (IBV) in poultry, however protection is short-lived and poorly cross-protective between strains. The vaccines must currently be grown in embryonated eggs, a cumbersome and expensive process. The ability to grow vaccines on a cell-line such as Vero cells would be highly advantageous. The spike (S) glycoprotein of IBV is comprised of two subunits, S1 and S2, has a vital role in virulence in vivo and is responsible for cellular tropism in vitro. This project aims to identify the amino acids present in the S glycoprotein involved in determination of cellular tropism and cell-to-cell fusion. The IBV Beaudette strain is able to replicate in both primary chick kidney (CK) cells and Vero cells, whereas the IBV M41 strain replicates in primary cells only. Recombinant IBVs with chimaeric S genes were generated using a reverse genetics system with the genomic background of Beaudette and part of the S gene from M41. Their growth characteristics and cellular tropism were investigated. The S2 subunit of Beaudette was found to be sufficient to confer the ability to grow on Vero cells and swapping just three amino acids with corresponding ones from M41 was sufficient to remove the ability of the Beaudette S glycoprotein for growth on Vero cells. Beaudette was further adapted to syncytia formation on Vero cells by serial passage and isolates were sequenced to identify amino acid changes between parent and Vero-adapted viruses that are potentially involved in cell-to-cell fusion. Understanding the way in which IBV infects host cells is vital in order to rationally design better vaccination and treatment strategies and help to reduce the prevalence of IBV infection in poultry worldwide. Using the IBV reverse genetics system, we now have the potential to grow IBV vaccines on Vero cells.

The avian coronavirus infectious bronchitis virus (IBV) expresses four nonstructural, non-gene 1 proteins (3a, 3b, 5a and 5b) which have been shown to be dispensable for virus replication in cell culture. These IBV accessory proteins have no sequence homology to any of the accessory proteins of the group I and II coronaviruses but are highly conserved among the group III coronaviruses. Characterisation of naturally occurring strains of IBV which do not express two or more of the accessory proteins and of genetically modified recombinant IBVs has demonstrated that these accessory proteins contribute at most a minor role to the pathogenicity of the virus. To understand the relevance of these proteins for IBV the subcellular location of the 3a, 5a and 5b proteins have been characterised along with the identification of potential protein-protein interactions for the 3a protein. The subcellular location and protein-protein interactions of the 3b protein were attempted but specific problems were encountered. Indirect immunofluorescence with confocal microscopy of IBV-infected chick kidney cells was used to study the subcellular location of the accessory proteins. The 3a protein displayed a punctate, cytoplasmic distribution pattern which colocalised with virally-induced double-stranded RNA. A diffuse, cytoplasmic distribution was observed for the 5b protein which produced limited colocalisation with an IBV structural protein. Expression of a FLAG-tagged 5a protein in transfected Vero cells resulted in a punctate, cytoplasmic pattern. The protein interactions of the 3a protein were identified using FLAG-tag pull-down experiments with tandem mass spectrometry. Six cellular proteins were identified as interacting with the FLAG/3a protein within transfected Vero cells, three of which, GCN1, PP2A and Exportin-1, may interact with native 3a protein in IBV-infected cells. The 3a protein could sequester the viral dsRNA to hide it from the innate immune system and the potential interactions with three cellular proteins indicate that the IBV 3a protein may contribute to attenuation of host cell translation, induce cell cycle arrest and/or attenuate the nuclear export of a specific subset of mRNAs.

The corpus luteum (CL) is a tissue formed from the remnants of an ovulated follicle in the ovary, and it produces the progesterone needed for a healthy pregnancy. CL growth is highly dependent on a growing nutrient supply, and can be compared with the most aggressive vascular tumours. Angiogenesis, the growth of new blood vessels from existing ones, plays a key role in the growth and function of the CL. Inadequate angiogenesis has been linked to infertility in cows. The CL is composed of several vascular(e.g. endothelial cells (ECs), pericytes (PCs)), and avascular (e.g. luteal cells (LCs), immune cells) cell types, and several pro-angiogenic factors (e.g. Fibroblast Growth Factor 2, FGF2) found to be important in the angiogenic process. The objective of this thesis is to shed light on the cellular and extracellular level determinants of angiogenesis in the bovine CL. We begin with the relevant biological and mathematical literature in Chapter 1. In Chapter 2, an ordinary differential equation model of CL growth is introduced. We assume that the CL volume is a continuum of three cell types, ECs, LCs, and stromal cells (such as PCs). The fourth variable in the model, FGF2, enhances the EC proliferation rate. The model is able, by varying parameters such as the maximal proliferation rate of the ECs, to distinguish cases where the CL shifts from a ‘normal’ to a ‘pathological’ growth. In Chapter 3, we present in vitro CL published and novel studies from Robinson’s Lab. Preliminary results demonstrate interesting endothelial and pericyte behaviours regarding cell aggregation and sprout formation, which are the motivation for the next two Chapters. In these experimental studies, all the CL cell types were incorporated in the same in vitro culture, hence providing a closer approximation to the in vivo environment compared to other in vitro cultures which use only a single cell type (mainly ECs). However, this complicates matters in terms of distinguishing cell behaviours and factors which contribute on the overall cell dynamics. Therefore, in the Chapters 4 and 5 we use data from literature. In Chapter 4, by using the Cellular Potts Model (CPM) framework, we focus on EC-PC interactions, and particularly on the mechanism which is responsible for the EC growth inhibition. Our model incorporates two possible mechanisms for inhibition. That is, the mechanical cell-cell contact inhibition, and the inhibition mediated from diffusive TGF-b secreted once the two cell types come in contact. Interestingly, our model results suggest that the effective range of TGF-b is a crucial determinant of the degree of EC growth inhibition. Chapter 5, by using a CPM, is devoted to sprouting angiogenesis (the formation of new blood vessel). The dynamic interchange between stalk and tip EC phenotype is incorporated through the Notch signalling pathway, with the leading tip cell moving up macrophage-mediated VEGFA gradients in a non-uniform matrix environment. The model reproduces phenomena in sprouting angiogenesis, including sprout morphology, tip competition, and explains knockout experiments on the Notch signalling pathway. Finally, we close with Chapter 6 where we summarise the ain results from each chapter and propose model extensions for future directions.

Recent published work indicates a role for pre-slaughter dietary vitamin D supplementation to promote post mortem meat tenderization in cattle (Montgomery et al., 2000; Foote et al., 2004; Montogmery et al., 2004), pigs (Wilborn et al., 2004) and sheep (Wiegand et al., 2001; Boleman et al., 2004). The hypothesis being that vitamin D supplementation at supra-nutritional levels is able to cause increases in the calcium status of the animals, increasing the activity of the calcium-dependant proteolytic enzymes, the calpains, which are responsible for meat tenderization (Koohmaraie & Geesink, 2006). Muscle fibre type is a variable factor in muscle and is related to meat quality (Klont et al., 1998). Vitamin D has been suggested to play a role in regulating skeletal muscle function through the creation of vitamin D receptor knockout mice models (Endo et al., 2003) and observations that muscle weakness and falling risk in vitamin D deficient patients is linked to a loss of fast muscle fibres (Aniansson et al., 1986; Larsson et al., 1979; Sorenson et al., 1979; Sato et al., 2005). This thesis investigated two vitamin D pre-slaughter diet regimes on their effects on meat quality of the most economically important cut of the carcass, the longissimus dorsi (LD) muscle (Molina et al., 2005) and the expression of the calpain system. Trial 1 fed sheep vitamin D at 2.0 X 10[superscript]6 IU/day for four days prior to slaughter and found that this had no effect on shear force of LD chops. Calcium levels were unchanged but mRNA levels of calpain I and II were increased 3.7 and 10% respectively (P=0.099 and P=0.014) but there was no effect on calpastatin mRNA nor changes in the calpain system at the protein level. Trial 2 fed sheep the same dose of vitamin D for 7 days with an additional calcium bolus, resulting in a 10% increase in calcium concentrations of both serum and LD. Toughness of LD chops was increased (P<0.01), there was no effect on mRNA of the calpain system but there was an increase in the protein levels of calpain II and calpastatin by 16 and 17% respectively (P=0.05 and P=0.087). A microarray study of rat primary myoblasts treated with 1,25(OH)[subscript]2D[subscript]3 for 24 hours highlighted a number of responsive genes significantly up and down regulated 1.5 fold or more (P<0.05). Pathway analysis identified novel targets of 1,25(OH)[subscript]2D[subscript]3 with a possible relationship to muscle growth and function; these included C/EBPβ metallothionein 2A and the MAPK, ERK. Three muscle cell strains, the rat primary muscle cells, L6 Aston and C2C12, were assessed for myosin heavy chain (MHC) gene expression using semi-quantitative PCR and western blotting analysis. The muscle cell line demonstrating the broadest range of MHC genes relevant to mature muscle tissue was used for the final experiments; this was the C2C12 cell line demonstrating expression of the slow MHC 1/β, an isoform which was absent or showed much lower expression in the other cells. C2C12 cells treated with 1α(OH)D[subscript]3 for 48 hours at varying stages of development responded in changes in myogenic regulatory factors (MRFs), MHCs and novel target gene expression. Real time PCR analysis of C2C12 cells treated with the active vitamin D metabolite 1α(OH)D[subscript]3 affirmed C/EBPβ mRNA expression to be upregulated (P<0.001) and MAPK ERK 1/2 phosphorylation to be down regulated (P<0.001) by 1α(OH)D[subscript]3 in muscle cells. The effect of 1α(OH)D[subscript]3 in myoblasts was reduce proliferation and promote differentiation, as myotubes formed the effect of 1α(OH)D[subscript]3 was to promote MHC gene expression of an intermediate oxidative fibre type, increasing expression of MHC 1/β and 2A, decreasing MHC 2B. In conclusion, there is no apparent benefit of a pre-slaughter dietary vitamin D feeding regime on meat quality, but the active metabolites of vitamin D, 1α(OH)D[subscript]3 and 1,25(OH)[subscript]2D[subscript]3, exert changes in gene expression and MAPK signalling which are likely to affect muscle growth and fibre type, and is of relevance in terms of both meat quality and muscle function in the elderly.

A poor maternal diet leads to offspring with a greater risk of developing chronic diseases later in life. This thesis considered whether a low protein diet during pregnancy in sheep affected the development of the fetal kidney, and how this impacted upon adult renal function when challenged by obesity. Pregnant ewes were fed either a control diet or a diet that was isocaloric but contained only 50% of the protein, in either early or late gestation. The effects of the diet were assessed on the ewe, day 65 fetuses (0.44 gestation), and two year old offspring which had been subjected to an obesogenic environment (ab libitum feed and reduced exercise). Few effects were observed on the ewe, confirming that the nutritional insult was relatively mild. Fetal renal vasculature (assessed by vascular corrosion casts) was not different between groups, although the microvasculature was significantly reduced in the early protein group, as evidenced by CD34+ staining of endothelial cells. This was accompanied by a reduction in angiogenic factors compared to control animals. Protein-energy malnutrition in the ewe led to reduced urea in maternal and fetal plasma, along with a concomitant reduction in ornithine in the fetal plasma and amniotic fluid. Other amino acids were relatively unaffected. In the adult sheep there were no effects on long-term renal function in the group fed low protein late in pregnancy, despite the lambs having lower birth weights and a period of postnatal catch up growth compared to the other groups. However, the group fed low protein during early pregnancy had reduced nephron number, microalbumuria and reduced renal function as assessed by gamma scintigraphy. There was also evidence of microvascular rarefaction which may have been exacerbated by obesity. This study did not reveal any consistent sex-specific effects of the maternal low protein diet. This study emphasises the importance of diet quality rather than quantity, and the importance of consuming a well-balanced diet during pregnancy to protect against future chronic diseases.

The growth of ovarian follicles is well documented in terms of hormonal control, however the fluid dynamics of antral follicle growth is less well understood. Aquaporins (AQP) are transmembrane water channels which facilitate the passive movement of water. In mammals 13 AQPs have been identified in a vast range of tissue types. In terms of ovarian AQPs there is currently a paucity of information. Recent studies in rat, pig and human have revealed the presence of ovarian AQPs, but in doing so have also highlighted a lack of consensus on AQP-type and location. The main aim of this study was to investigate the potential role of AQP in antral follicle growth. The first objective was to identify tissue expression and localisation of AQP proteins in the bovine ovary. This required the characterisation of a panel of polyclonal serum antibodies. Immunohistochemistry (IHC) was then used to identify AQPs and to detect changes in protein expression during follicular growth. Aquaporin 1 was found in most vascular endothelium; it was plentiful in capillaries surrounding antral follicles and increased in abundance as vasculature increased with follicle development. Aquaporin 2 was not found in bovine ovarian tissue and the remaining antibodies were deemed too nonspecific to permit reliable conclusions. The second objective was to investigate, via RT-qPCR, mRNA levels of AQPs in granulosa and theca cells isolated from preantral, through to large preovulatory follicles. Transcripts of AQP 1, -3, -4, -5, -7 and -9 were detected in both the granulosa and theca of antral follicles with expression levels generally higher in theca. The expression of AQP 1, -5, -7 and -9 was initiated in the theca cells of early antral follicles. Finally, swelling assays using bovine and porcine granulosa cells demonstrated the ability of granulosa to swell. This was inhibited by HgCb which is characteristic of AQP function. Porcine granulosa cells incubated with androgen swelled by 27%, this effect was inhibited by hydroxyflutamide. Protein analysis of AQP5 via IHC and Western blotting showed possible up-regulation in porcine follicles. RTqPCR did not reveal AQP5 transcript, the reasons for this currently remain unclear. In conclusion, this study has revealed for the first time the involvement of AQPs in bovine ovarian follicle development, with AQPI, -5, -7 and -9 potentially playing a pivotal role in antrum formation. The AQP system in porcine granulosa cells is androgen sensitive however identification of the AQP/s responsible needs further investigation. The evidence from this investigation suggests a role for AQPs in facilitating follicle growth. The stage-dependent expression of certain AQPs and the androgen sensitive porcine granulosa cells reveals the possibility that AQPs may be modulated by follicle-regulating hormones.

Very little is known about cattle vocalizations. The few studies available in the literature have been conducted using animals under stress or very intensive husbandry conditions. Similarly, the individual consistency of behaviour in cattle has rarely been considered except in applied studies of constrained and isolated animals, and no previous research has attempted to address a possible association between vocal communication and temperament in cattle. The studies reported here address these gaps in our knowledge. I found that cattle contact calls have acoustic characteristics that give them individualized distinctiveness, in both adult cows and calves. These results were confirmed using playback experiments, where I found that there is bidirectional mother-offspring recognition, as has been recorded in other “weak hider” ungulates. Additionally, using visual and acoustic stimuli, I assessed individual cattle temperament. The results showed that there was no individual behavioural consistency in responses to a novel object presentations. However, calves behaved consistently more boldly than cows. Furthermore, there was significant individual consistency in responses to vocalisations of heterospecifics, when they were played back through a speaker in the field. Surprisingly, no correlations were found between the ability of cattle to identify their own mother/offspring and the acoustic features of their vocalisations, or behavioural responses in any other context. There were, however, significant correlations between one characteristic of vocalisations in adult cows (formant spacing) and the boldness of behavioural responses to both novel objects and auditory stimuli. Additionally, higher F0 in calf contact vocalizations correlated with boldness in the auditory stimuli experiment. These relationships imply that vocalisations may encode information about individual temperament, something which has rarely been documented. Surprisingly, no strong correlations were found between the behavioural responses to visual and acoustic stimuli, suggesting that individual consistency in behaviour across contexts was limited, and that behavioural plasticity could play an important role in determining responses in different environmental contexts. Overall, my results contribute to our knowledge of animal communication in mammals from a bioacoustic point of view, and they are also potentially relevant to studies of vocalizations as indicators of cattle welfare.