Protein arriving at the ruminant proximal duodenum consists of microbial protein, undegraded feed protein and endogenous protein. In this study, endogenous protein is defined as that fraction of the digesta derived from the animal itself (e.g. enzymes, plasma proteins, sloughed cells and mucus), not including any endogenous protein which may have been incorporated into the microorganisms. Recent feeding schemes (e.g. ARC 1980, 1984) require an accurate value of the degradability of feed in the rumen. When degradability is determined in vivo failure to account for a quantity of endogenous protein in the proximal duodenal digesta results in an underestimate of the degradability of a ration. Direct estimates of the endogenous protein content of proximal duodenal digesta are therefore required. This thesis describes the development of a method to do this and its application to sheep and cattle. The approach adopted, based on the concept of isotope dilution, involved a continuous intravenous infusion of L-[4,5-3H]-leucine. This resulted in all body protein becoming labelled. Any label detected in the duodenal digesta must therefore be derived from the animal itself. A comparison of the specific activity of duodenal digesta with that of suitable precursor proteins provided an estimate of the proportion of duodenal digesta of endogenous origin. Interestingly, the bacterial fraction of duodenal digesta was also labelled. This indicated that bacteria were utilising an endogenous source of leucine and circumstantial evidence suggested that this was derived largely from the rumen epithelium. The validity of the L-[4,5-3H]-leucine technique was investigated using three sheep fed an essentially protein free diet. Values of endogenous protein flow derived via the proposed technique (2.6 + 1.00 g N/day) were compared with those calculated by difference (2.1 + 0.92 g N/day). A possible dietary influence on the endogenous protein content of ruminant proximal duodenal digesta was examined. An estimate of 2.6 + 0.58 g N/day was derived for three sheep fed a concentrate diet and 2.1 + 0.22 g N/day for three sheep fed long hay. Thus, contrary to previous suggestions, no significant dietary effect was observed. Possible reasons for this are discussed. The proposed technique was also applied to derive estimates in cattle. A value of 11.3 + 1.73 g N/day was obtained for four steers (170 kg) fed silage supplemented with fishmeal. The continuous intravenous infusion of L-[4,5-3H]-leucine method is the first technique to provide a direct determination of the endogenous protein component of ruminant proximal duodenal digesta which can be applied to any dietary regime and used in sheep and cattle.

The wide-ranging potential impacts of climate change on both ecology and human infrastructure have led to a large amount of research; however, studies of the projected impacts on agricultural systems have so far focussed mainly on crops. Given the proven adverse effects of extreme weather conditions on the productivity and welfare of livestock, this thesis assesses the potential impact of such a change on the thermal balance of livestock in the UK. A series of mathematical models was designed to predict the metabolic rate and occurrence of thermal stress in sheep and cattle outdoors, and pigs and broiler chickens indoors by solution of the energy balance equations. The models run on commonly-available hourly weather data, and as far as possible were based on the physics of heat and mass transfer rather than empirical relationships. The animals were modelled as systems of geometrical shapes, incorporating the underlying tissue, a coat and the external environment. Physiological responses to hot and cold conditions, including panting, sweating, vasomotor action and shivering were parameterised. Validation of the model output showed good agreement with measured data. The climate predictions for the year 2050 were reduced to synthetic hourly weather data using a stochastic weather generator and several simple downscaling techniques. The climate change impact assessment was made for an upland and a dry lowland site in the UK. There are two main conclusions to the work. First, climate change is predicted to have little effect on ruminants outdoors, or on the suitability of a site for grazing livestock. Second, animals indoors will experience significantly more heat stress under climate change, probably since indoor animals are at greater risk of heat stress in the current climate than those outdoors. In the next fifty years, pig and broiler chicken farms will have to introduce methods for alleviation of heat stress to avoid economic and welfare problems. Future work will need to focus more on collection of accurate heat balance data rather than on more mathematical modelling.

In piglets, the post-weaning growth check is commonly associated with the weaning process that occurs within the European Union at approximately 3-4 weeks of age. The aims of the studies reported here were to investigate the contribution and importance of nutritional influences on the complex and multi-factorial problem of the post-weaning growth check in the piglet. Multi-disciplinary investigations focussed upon the relationship between post-weaning nutrition and the gut ecosystem with specific emphasis on gut physiology, immunity and microflora. The influence of dietary acid binding-capacity (ABC) on gut morphology characteristics was investigated in 20 newly-weaned piglets up to 14 days post-weaning. Piglets offered the low ABC diet displayed more rapid recovery of villus height (after a degree of villus atrophy) than control animals (P<0.001). Prior to the initial villus atrophy, villus height increased significantly throughout the experimental period for both dietary treatment groups (P<0.001; <0.001 (L); <0.001 (Q)). In control animals, villus width was greater (P=0.006) compared with treatment animals, and villus width increased over time for both groups (P<0.001; <0.001 (L); 0.014 (Q)). Crypt depth also increased temporally (P<0.001; <0.001 (L)) for both dietary groups, with treatment animals exhibiting the greatest overall dietary mean (P=0.009). No significant differences between ileal digesta pH and feed intake levels were determined. Despite the improvements in intestinal structure post-weaning, these effects were not manifested in increased performance, i.e. DLWG. The improvements in intestinal structure may not have been of significant magnitude to influence performance parameters. Dietary zinc oxide (ZnO) and avilamycin supplementation was found to exert a beneficial (although non-significant: P>0.05) effect on gut morphology; villus atrophy occurred over the initial 2 days post-weaning for animals fed ZnO, avilamycin or ZnO plus avilamycin (diets 2-4 respectively), compared to 4 days for control animals. No significant differences between intestinal coliform and lactobacilli load were established with respect to dietary treatment. Any differences observed in microflora load are most likely to be age-dependent. A positive relationship was established between dietary treatment (ZnO, avilamycin and ZnO plus avilamycin) on daily live weight gain post-weaning (P<0.001). Although not significant (P>0.05), a positive influence of dietary ZnO supplementation on feed intake levels was apparent, which may account, in part, for the enhanced growth performance. This finding was not however manifested through modifications of intestinal morphology or the lactobacilli and coliform populations studied. This further suggests that dietary ZnO may exert an effect either luminally or systemically. Further research is required to determine the mechanism responsible for the enhanced feed intake and DLWG response. The effects of feeding a yeast-based nucleotide source pre- and post-weaning revealed no significant differences with respect to villus height and width. Crypt depth was significantly greater in animals fed the treatment diet post-weaning (P<0.001). Post-weaning nucleotide-supplemented diets were found to significantly reduce intestinal coliform load (P=0.033). Such an effect was not evident in animals fed the diets pre-weaning, suggesting that the gut microflora may have adapted to the dietary regimen throughout the pre-weaning period. Lymphocyte blastogenesis assays revealed that piglets fed a yeast-based nucleotide source post-weaning might be immunosuppressed at the time weaning. Conversely, when the same diets (in terms of composition) were fed from 14 days pre-weaning (study 4), no indication of immunosuppression was evident. Since no dietary effects were apparent in either study, it is postulated that this could be a general effect of the diet per se and not the actual dietary composition. It is however also possible that the animals involved in study 4 were experiencing hypersensitivity reactions to the pre-weaning dietary antigens. These animals were also combating an E. coli infection. Additional studies are however required to identify conclusively a cause and effect relationship, and elucidate the complicated interactions between nutrition or feed intake and immunobiology in the post-weaned piglet. Implementation of dietary nucleotide-supplementation from 7 days pre-weaning through to 25 days post-weaning within a commercial environment was found to enhance significantly DLWG (P<0.001). In summary, the current work extends current knowledge and offers a greater understanding of the factors and complex process that influence the gut ecosystem and physiology in the post-weaned piglet. This thesis confirms the crucial role feed intake or, more specifically luminal nutrition, in post-weaned piglets and has highlighted key areas for future investigation.

20 Large White x Landrace sows were kept throughout 4 parities on 1 of 2 exercise treatments, free (F) or restricted (R), imposed at 2 stages of the production cycle (namely gestation and farrowing/lactation), thus giving 4 treatments: FF, FR, RF and RR. During gestation, the sows were housed in cubicles - group a being tethered while group F were allowed unrestricted movement within the cubicle and dunging area. From day 110 of gestation to weaning, all sows were kept in the same farrowing house, group R in conventional crates, group F in strawed pens. Management of the sows was otherwise identical. All farrowings were monitored with regard to the duration of parturition and stillbirth rate, while various behavioural and physiological measurements were recorded as indicators of piglet viability. Observations were also made of now behaviour at various stages of the production cycle. There were significant (P< 0.001) behavioural differences in gestation with groups R and F spending 45% v. 25% of the time lying and 27% v. 42% in manipulating straw. There were also significant (P<0.001) differences in amount and type of locomotor activity with the restricted sows making more minor movements while the free sows made more pace movements. At farrowing, all sows showed increased restlessness but group a made more leg movements and exhibited significantly (P<0.01) more straining and quivering (both pre- and intrapartum) than group F, although the latter stood and nested more frequently during the farrowing process (P< 0.001). There were no significant differences between treatments during lactation. In terms of reproductive performance, the differences between treatments were non significant, but sows in pens (group F) farrowed more quickly (a mean birth interval of 21 v. 39 minutes), produced more live pigs (11.3 v. 10.5) and fewer stillborn pigs (0.5 v. 0.8) per litter, than sows in crates (group a). Group F sows also produced piglets which had been subjected to less hypoxia as evidenced by their lower (P <0-05) serum lactic acid levels at birth (140 v. 158 m.I.U./ml) although there were no significant differences between piglets of groups F and R in times taken to breathe following birth or (53 v. 45 seconds), to suckle (36 v. 32 minutes) or in plasma immunoglobulin levels at 36 hours post partum (41.5 v. 41.7 mg/ml). Neither were there any significant treatment effects on piglet growth rate and pre-weaning mortality.

Leptin is produced primarily by white adipose tissue but also by brown adipose tissue (BAT) and the placenta. It has a major physiological role in the control of energy balance. BAT is critically important for the initiation of non-shivering thermogenesis in precocial newborns through the uncoupling protein, UCP1 unique to BAT. This is particularly important in neonatal lambs and infants in which levels of UCP1 peak at birth concomitant with a peak in plasma leptin and a surge in Cortisol. Other mitochondrial proteins may contribute to this effect e.g. VDAC which is located in the outer mitochondrial membrane. The aim of this thesis was to: 1. investigate the role of the late gestation Cortisol surge on BAT development 2. investigate the effect of route of delivery on subsequent neonatal plasma leptin concentrations 3. assess the effect of leptin administration to neonatal lambs on thermoregulation Late gestation fetuses were infused with Cortisol or saline or underwent adrenalectomy or sham operation. BAT was sampled at 129 and 144 days of gestation, respectively. UCPl abundance was significantly increased in the animals receiving Cortisol treatment compared to controls and was significantly reduced in adrenalectomised animals. Plasma leptin was analysed from lambs delivered vaginally or near term by caesarean section (CS). Cord plasma leptin decreased significantly after birth, an adaptation that was delayed by CS delivery. Acute and chronic administration of leptin to neonatal lambs improved thermoregulation by preventing a decline in body temperature. Chronic leptin treatment over 7 days (lOOp-g daily) promoted the loss of UCP1 mRNA and protein, but had no deleterious effects on body temperature. Polyclonal antibodies were developed against mitchondrial voltage-dependent anion channel (VDAC). The postnatal ontogeny of VDAC was found to be similar to that of UCP1 and cytochrome c, with abundance peaking around one day of age. VDAC was found in high abundance in organs with high metabolic requirements such as heart, muscle and BAT. These results suggest that VDAC is involved in ensuring BAT maintains a maximal rate of thermogenesis after birth. In conclusion, I have shown for the first time that leptin has an important role in thermogenesis during the transition from fetal to neonatal life. This is attenuated in CS animals, possibly linked to a reduced sympathetic nervous system activity. The rapid loss of UCP1 mRNA, which occurs within the first few days of life, appears to be modulated by leptin, possibly stimulating the development of white adipose tissue and generation of body heat through mechanisms other than nonshivering thermogenesis in BAT. VDAC may be important during this period in ensuring adequate substrate delivery to BAT. Intact adrenal glands are also necessary for the increase in UCP1 abundance during late gestation, an effect mediated in part by Cortisol. NB. This ethesis has been created by scanning the typescript original and may contain inaccuracies. In case of difficulty, please refer to the original text.

The biological activity and genomic variation of five Kenyan, three Asian and two sublethal isolates of Plutella xylostella granulovirus (PlxyGV) were determined and compared in P. xylostella larvae reared on artificial diet. Optimal production of PlxyGV occlusion bodies (occluded virus, OB) was observed in larvae reared in isolation rather than communally. Incorporation of a juvenile hormone analogue in the artificial diet increased OB production seven times. The pathogenicity and infectivity of the isolates was compared using neonate P. xylostella. A single Kenyan isolate (PlxyGV-2.18) was shown to have a significantly shorter median lethal time to death (LT50) than the other nine isolates in other geographical populations of P. xylostella. No synergism was observed between the isolates in mixed inoculations of P. xylostella and a slower killing isolate predominated over a faster killing isolate in mixed inoculations. The presence and nature of “sub-lethal” infections of PlxyGV observed in stock cultures of P. xylostella collected in Kenya and England was studied. Sub-lethal infections were observed in approximately 23% of apparently healthy stock larvae from two of the stocks of P. xylostella (one from England and one from Kenya) but was absent from a third stock originating in England. It was possible to isolate OBs from apparently healthy P. xylostella larvae in the two virus positive stocks, indicating some productive infection. The sub-lethal isolates had similar pathogenicity and infectivity to the other Kenyan and Asian isolates (excluding PlxyGV-2.18). The genomes of the isolates were compared by restriction fragment length profiling and by sequencing specific regions of the viral genome.

Clinical audit was adopted from the medical profession in the late 1990s and is now widely discussed as a quality improvement tool in veterinary medicine. It is suggested by the Royal College of Veterinary Surgeons as a method to improve the care provided to veterinary patients and many opinion pieces offering advice have been published. However, there is no evidence to support its use in farm animal veterinary practice and no understanding of how it is currently used among farm animal practitioners. This PhD thesis aimed to determine if and how clinical audit can be used efficiently and effectively in farm animal veterinary practice in the United Kingdom (UK). A review of the literature found that the translation of clinical audit from human to veterinary medicine had resulted in confusing and conflicting advice for practitioners working with all species. The definition of clinical audit seemed unclear, advice on how to conduct the process differed between authors and the relationship with evidence and guidelines was ambitious. Very few publications were based on research evidence and most centred on expert opinion. There was also no information on how clinical audit was used or understood by farm animal practitioners. A cross-sectional study to collect experiences and attitudes of farm animal veterinary surgeons in the UK towards clinical audit was conducted using a nationwide survey. Despite the emphasis put on the process, the survey revealed that many veterinary surgeons had never heard of clinical audit, or never been involved in clinical audit in farm animal practice. The participants’ knowledge of clinical audit varied widely, supporting the findings of the literature review that clarification of the process was required. A case-series using qualitative data gathering methods and analysis, including a nominal group technique (Priority Setting Partnership approach) and thematic analysis, was carried out with three farm animal veterinary practices in the UK. Prospective and retrospective clinical audits were implemented in the practices using a clearly defined framework and detailed feedback was collected on the process through a number of routes. Attempting clinical audit in the three practices highlighted many challenges to conducting clinical audit in farm animal veterinary practice. This included the difficulty of differentiating between research and clinical audit, the logistics of data collection on farm, the limitations to conducting retrospective audits in farm practice, and the time taken for the process. A number of benefits of the process were also appreciated. The practitioners found clinical audit to be an interesting exercise that increased communication within the practice and improved knowledge on how procedures were conducted. Protocols and guidelines are a suggested component of clinical audit but little information was found about their veterinary application in the literature review. To investigate the use of protocols and guidelines in farm animal practice a survey-based case study using structured interviews and a questionnaire was completed examining the understanding of, and attitudes towards, a surgical protocol used by veterinary surgeons within a practice. The case study found that guidance can have a place in practice, but highlighted that it should always allow for clinical findings in each case to be considered and acted on appropriately. Suggestions were made that checking all staff follow a guideline may not be the most beneficial way to conduct clinical audit. Following the work of this thesis, the definition of clinical audit and suggestions for its use in the veterinary setting have been simplified and defined. Discussions about the suitability of clinical guidelines and protocols and how each relate to clinical audit have been started and may provide a baseline for future research. Recommendations for farm animal practitioners wishing to conduct clinical audit in practice have been produced with an emphasis on attempting the process for the first time. This thesis has modified the process of clinical audit to ensure that it is accessible and practicable for veterinary surgeons wishing to improve the quality of care delivered in practice.

The extensive use of antibiotics has led to the emergence of methicillin-resistant Staphylococcus pseudintermedius, a bacterium causing difficult-to-treat canine skin infection (pyoderma). The administration of bacteriophages (phage therapy) can be an alternative to antibiotic therapy. Lytic phages, which lyse their host, are considered the only appropriate type of phages for phage therapy as opposed to temperate phages, which can survive within their host (lysogeny). However, it is possible to mutate temperate phages so that they cannot establish lysogeny anymore. Phage λ virulent (Vir) mutants have lost the operator to which the CI repressor binds to inhibit the expression of lytic genes. As a result, Vir mutants are strictly lytic. The work presented in this thesis was undertaken to isolate S. pseudintermedius phages and gain knowledge about their biology with the aim to develop phage therapy to treat pyoderma. The work was novel; very few data were available on S. pseudintermedius phages and no data have been published on phage therapy to treat canine skin infection. Four temperate phage candidates were selected after phenotypic and genotypic characterisation. No lytic phages were found. Random mutagenesis approaches were unsuccessful for the isolation of Vir mutants. An operator and three point mutations leading to the absence of CI repressor binding to this operator were identified through gel shift assay. These mutations should lead to a virulent phenotype if introduced in the relevant phage genome through site-directed mutagenesis. A PCR-based assay was performed to explore how widespread lysogeny was in S. pseudintermedius: 11 out of 45 tested strains were positive for the presence of prophage genes. Bioinformatic analyses revealed some of the genetic characteristics of S. pseudintermedius phages: genomic circular permutation and the presence of a genetic switch similar to that of phage λ. The work reported in this thesis represents a first step towards understanding the biology of S. pseudintermedius phages and developing phage therapy.

The hypothesis that midge sex pheromones could be chemically more diverse in structure than previously thought was tested in the context of four midge species of importance to UK horticulture: pear leaf midge, Dasineura pyri; pear midge, Contarinia pyrivora; blackcurrant midge, D. tetensi; and blackberry midge, D. plicatrix. The major component of the pheromone of D. pyri was identified as (2R, 13R, 8Z)-2, 13-diacetoxy-8-heptadecene. Four isomers were separated by HPLC and in field tests the first eluting isomer only was attractive to male midges. Analysis of volatile collections from female C. pyrivora showed two consistent responses from male midges and they were identified as 2,7-diacetoxyundecane and 7-acetoxyundecane-2-one. The field testing with isomers of 2,7-diacetoxyundecane separated by HPLC revealed that the first and the third eluting isomers were attractive. The racemic 7-acetoxyundecane-2-one was active as well as the first eluting isomer from HPLC. Two EAG active components were detected in D. tetensi female volatile collections. The major component was identified as (Z)-2,12-diacetoxy-8-heptadecene and after separation of stereoisomers by HPLC the third eluting isomer has shown to be attractive to male D. tetensi in the field. The structure for the minor component was proposed as a keto-acetate homologue of the corresponding major component. Preliminary work carried out on identification of the female sex pheromone of D. plicatrix indicated two responses from conspecific males. These were shown to be 15-carbon acetates with the acetate function at C-2, probably with two and one double bonds respectively.

Claw horn disruption lesions (CHDLs: sole haemorrhage, sole ulcer and white line disease) cause a large proportion of lameness in dairy cattle and have a high rate of recurrence, yet their aetiopathogenesis remains poorly understood. Untreated CHDLs appear to be associated with trauma within and damage to the internal anatomy of the foot. Chapter 2 explored associations between abnormal bone modelling on the flexor tuberosity of the distal phalanx of cull cows and lameness during life, using a retrospective cohort study design. The hind claws of 72 Holstein dairy cows culled from a research herd were imaged using μ-computed tomography (CT) and lameness and lesion incidence data were available on cows throughout life. Four measures of bone modelling were taken from CT images from the flexor tuberosity of each distal phalanx, in plantar, distal and dorsal planes, and combined within claw. Bone modelling was greater in older cows, in cows with history of CHDL and in cows that had been lame at an increased proportion of locomotion scores during the 12 months preceding slaughter. Further, histological study demonstrated that the bone modelling resembled heterotopic ossification, also termed osteoma, which could have been due to either inappropriate force transfer through the distal phalanx or pathology in the soft tissues with lesion presence. Anatomical damage within the foot does appear to be associated with lameness and CHDLs, and may further predispose lameness. Preventing lameness constitutes a critical component of lameness control, and prophylactic foot trimming is a common management strategy for maintaining claw structure and function. However, over-trimming can cause damage to the foot architecture and lameness. Step 1 of the widely used Dutch Method of foot trimming states to cut the dorsal wall of the hoof to 75 mm. A vertical 5 mm step is left at the toe, therefore based on these recommendations, dorsal wall length would be 82 mm if the toe were trimmed to a point and the dorsal wall extended to the floor (at a toe angle of 50°). Chapter 3 used the CT data to assess the minimum dorsal wall length that would be suitable for trimming each claw. The median length was 76 mm (83 if the toe were trimmed to a point) and ranged from 59 to 86 mm; trimming all claws to 75 mm would have over-trimmed 55 % of claws. In a linear regression model, minimum dorsal wall length increased with age and carcass weight; older and larger cows had bigger claws. However, the vast majority of variation in claw length remained unexplained (only 22 % of the null variance was explained). In order to minimise the number of claws that are over-trimmed, recommendations for foot trimming dimensions should be based on the proportion of claws for which a measurement is suitable, rather than on population means. The minimum lengths that would have been suitable for all claws were 93 mm for cows aged ≥4 years and 86 for cows aged < 4 years; 7 mm could be taken from these measurements if a step is left at the toe. CHDLs appear to initially occur through trauma to the germinal epithelium of the sole, and Chapters 4, 5 and 6 present a longitudinal study of how the sole soft tissues (SST; i.e. the digital cushion and corium) alter throughout lactation. The digital cushion is a modified layer of the subcutis that is situated beneath the plantar and distal aspects of the distal phalanx and is considered to be important in dissipating forces during foot strike and to protect the germinal epithelium. The digital cushion contains depots of adipose tissue and recent work has identified that body condition loss is a risk factor for lameness. Previous work found that fatter cows had thicker SST and suggested that fat could be mobilized from the digital cushion and causes it to have decreased biomechanical function. The prospective cohort study assessed the SST of 179 parity 1, 2 3 or 4 cows at 5 assessment points, between 8 weeks pre- and 29 weeks post-calving of one lactation. Lesions present on claws and measures of body fat were recorded at each assessment point, and mobility scoring was performed fortnightly from calving. SST thickness at two sites beneath the distal phalanx were used as outcomes in 4-level mixed effects linear regression models (Chapter 5), and was positively correlated with back fat thickness. However, the effect size was much smaller than reported in previous cross-sectional work and only apparent under some circumstance. SST was thicker when a sole ulcer was present on a claw and was thinner immediately after calving (during the 4-10 days post-calving). The final model left 61 % of the null variance unexplained, of which 48 % remained between repeated measures of the same claw at different assessment points. Chapter 6 presents a series of logistic regression models of survival to first lesion or to first lameness (repeated lameness events were initially tested, but models were discarded due to the high rates of recurrence of both lameness and lesions). Lesion models demonstrate that claws were more likely to develop a lesion if SST was thin, and there was an additional effect of having thin back fat (all animals) or having lost back fat between previous assessment points (parity >1 animals only). Lameness models demonstrated that thin SST on the lateral claw increased the odds of a leg becoming lame, but SST on the medial claw had no effect on lameness. Change in SST thickness did not predispose lesions or lameness; only absolute thinness did. The work suggests that whilst loss of body condition loss may be one variable that contributes towards thinning of SST and subsequent claw horn disruption, many other variables also had a large effect on SST thickness, CHDL and lameness. This thesis presents a sequence of studies of how the anatomy of the foot is related to CHDL incidence, addressing recurrent lameness, mechanisms for the onset of new lameness and the appropriateness of prophylactic foot trimming guidelines as a management tool for lameness. The research literature is deficient in work demonstrating beneficial effects of interventions on lameness, and work throughout this thesis provides novel insights into the aetiopathogenesis of the claw horn disruption lesions. Based on this work, targeted interventions to reduce lameness can be tested.