The thesis contains papers on both experimental and theoretical work carried out with the objective of maximizing genetic gain in breeding schemes under a variety of constraints. The experimental work is principally in the context of dairy cattle improvement schemes, whereas the theoretical development in more generally applicable to truncation selection. The work had its origins in the introduction of multiple ovulation and embryo transfer (MOET) in dairy cattle breeding schemes. The initial group of papers concerned the benefits and development of a juvenile predictor of dairy merit. The potential benefit of such a predictor was quantified and shown to offer substantial increases in gain (ΔG) without increasing rates of inbreeding (ΔF). The experimental work developed a new successful approach to the problem via growth hormone release. The association were replicated and an understanding of the association in terms of endogenous growth hormone profiles was demonstrated. The next group of papers developed methods to predict ΔF for selection populations. Initial work compared published and novel 1-generation methods as a means of predicting ΔF and the identification of their shortcomings as predictors. Further papers followed the approach of predicting squared contributions and showed that the method was equivalent to the variance of family size but with a correction for the expected proliferation of lines with selective advantage. Finally, other issues related to constrained breeding schemes were explored including defining acceptable ΔF for populations, the impact of inbreeding on the overall productivity, an examination of the benefit of SRY transgenes, and the relationship of genetic contributions to the selection criteria in commercial data.

This thesis addresses the issues of genetic improvement of carcass and maternal traits in hill sheep. It i) compares the performance of two genetic lines of Scottish Blackface lambs divergent for subcutaneous fat, ii) quantifies the genetic components of carcass traits in extensive hill environments, iii) explores the implications of selecting for reduced fatness in hill lambs, iv) develops and describes methods to include carcass traits in the breeding goals for hill sheep, and v) gives predicted results from index selection for maternal and carcass traits, using indexes of overall economic merit. For points i) to iii), approximately 2000 Scottish Blackface lambs were measured, sired by 32 rams divergent for subcutaneous fat depth, and born to 1660 unselected ewes in 1991 and 1992. They were reared under extensive conditions on two contrasting hill farms. Results showed that genetic differences in subcutaneous fatness arising from selection in an intensive environment are still expressed despite harsh rearing environments. Heritabilities for birth weight, marking weight (at approximately 6 weeks of age) and weaning weight (at 17 weeks) were 0.07±0.04, 0.02±0.03, and 0.14±0.05, respectively. Heritabilities for ultrasonic muscle and fat depth were 0.27±0.09 and 0.16±0.06, respectively. Heritability estimates for carcass traits were: pre-slaughter liveweight 0.36±0.13, cold carcass weight 0.39±0.14, fat class 0.13±0.08, conformation score 0.09±0.07, dissected lean weight 0.27±0.27, dissected bone weight 0.36±0.13 (constant subcutaneous fatness), dissected intermuscular fat weight 0.10±0.07, subcutaneous fat weight 0.20±0.09 (constant cold carcass weight). There was a strong maternal effect on live weight which declined with age. The rearing environment of the lambs was an important environmental effect on the heritability estimate for backfat thickness, being twice as large for animals reared on the improved pasture compared to those reared on hill pasture.

The purpose of this thesis is to quantify the effect of selection for disease resistance on the epidemiology of microparasitic infection in pigs, combining quantitative genetics and epidemiology using computer simulation. Two methods of selection are investigated: continuous selection at an arbitrary rate of progress and resistance gene introgression Selection is simulated on a 500 sow farrow-to-finish farm with selection taking place in the sires used to provide semen. A discrete-time model is introduced which expands that of De Jong (1994) allowing reduced susceptibility to infection to be included. The benefits of selection for disease resistance are measured by the effect on the basic reproductive ratio, R₀, which describes the expected number of secondary cases from a single infected individual. Two functions of R₀, the maximum and total proportions of pigs infected during simulated epidemics, are used to demonstrate the benefits of the different selection implementations. The model shows that although the reduction, under selection, in R₀ is linear, the reduction in the proportions of infected animals is not. For a highly infectious pathogen, it may take many years of selection before the benefits are seen on the farm. This is not the case for the gene by the time it takes to introgress the resistance alleles. The gene introgression model provides an indication of the proportion of a population that need to be resistant for that population to be protected from epidemics. Unless the pathogen is extremely infectious it is not necessary that the whole population is resistant. For example, when R₀=5.0 for the population prior to selection, the proportion that need to be resistant to protect the population is 0.82 and when R₀= 10.0, the proportion is 0.95.

Animals are adapted to live in fluctuating environments. Some stimuli to which they are exposed will be ignored, some will be avoided and others will be approached. Stimuli perceived as a threat or associated with a painful stimulation will tend to be avoided. Therefore to understand more fully how an animal copes with a particular situation, e.g. transportation, its perception of all stimuli needs to be determined. The aim of the study reported in this thesis was to determine how auditory stimuli, to which pigs are exposed during production, are perceived by individual pigs. A field study was carried out to characterise the sounds to which pigs are exposed during production and studies were made of pig responses to sound under experimental conditions. The sound pressure level in artificially ventilated fattening units was quite loud (70 to 80 dB(Lin)), but relatively constant. In contrast, naturally ventilated units were quieter (60 to 70 dB(Lin)), but more variable. Sound pressure levels during transport were more than 88 dB(Lin) and highly variable. Similar levels were measured in articulated transporters and small livestock trailers. Sound pressure levels measured in abattoir lairages varied from 77 dB(Lin) to 89 dB(Lin). Equivalent sound pressure levels (Leq 20 min) of 97 dB(Lin) were measured in the stun pen of one abattoir that used electric stunning. Pigs' perception of mechanical sounds between 85 and 100 dB(Lin) was assessed. The onset of sound activity and visual searching. Stronger responses were measured for louder sounds. Over a constant exposure period of 15 to 20 minutes the responses observed decreased towards basal levels.

Three-generation pedigrees, in which the founders were Meishan and Large White purebred pigs, were used to identify QTL for female reproductive performance. The Chinese Meishan is one of the most prolific pig breeds known, farrowing up to five more piglets per litter than the European commercial Large White breed. However, the Meishan is not commercially viable in Europe due to its poor growth rate and high carcass fat content. Therefore, including the beneficial alleles from the Meishan into the Large White breed is of commercial relevance. QTL for ovulation rate, teat number, age at puberty and uterine capacity have been mapped to chromosome 8 (SSC8) in earlier studies. The aim of this study was to focus specifically on SSC8 and to identify and test candidate genes for the QTL. The genotypes of twenty markers on SSC8 were combined with data collected on the reproductive performance of 220 F2 females. QTL for the related traits of litter size and prenatal survival were identified at the distal end of the q arm. The beneficial alleles at these QTL seem to be from the Meishan. The QTL for prenatal survival was defined as a region of about 30 cM and therefore contains many positional candidate genes, but only a few of these have been mapped in pigs. Human chromosome 4 (HSA4) shares extensive homology to SSC8. A gene map of SSC8 was developed by radiation hybrid mapping in order to align the human genome sequence of HSA4 with the QTL and thus identify comparative positional candidate genes. The resulting comparative map revealed extensive conservation of synteny and gene order. One positional candidate gene, secreted phosphoprotein 1 (<i>SPP1</i>), is also a physiological candidate gene as the protein is involved in porcine embryo implantation and maintenance of pregnancy. A copy of <i>SPP1 </i>(~10 kb) from both Meishan and Large White origin was sequenced to identify candidate causal variation. A total of 97 variants, including single nucleotide polymorphisms (SNPs) and indels, were found. Three of the SNPs would result in non-synonymous amino acid changes. For one of these SNPs the Meishan allele encodes a serine, which would be phosphorylated and the Large White allele encodes a proline, which would not. To test whether the variation at this particular locus resulted in differences in litter sizes, the SNP was genotyped over 2974 pigs of varying breed origin. No association was identified between genotypes at this locus and litter size.

Transgenic sheep 229 carries one copy of a mouse metallothionein-I promoter HSV-tk fusion gene (pMK), Poly A<SUP>+</SUP> RNA from a liver biopsy revealed no HSV-tk message. A skin fibroblast cell-line was established from sheep 229 and is the subject of this thesis. The cell-line is non-immortal and senesces after 30-40 population doublings <i>in vitro</i>. The cell-line carries one intact copy of pMK which is not rearranged as detected by southern blot analysis of genomic DNA. The cells do not display any HSV-tk activity and no HSV-tk mRNA is detected by northern analysis. The expression of the sheep metallothionein gene family in the cell-line was investigated using gene specific probes. All four known active members are expressed and induced in response to zinc. The two major transcripts, from the Ia and II genes, are also induced by copper. The basal level and zinc induced expression of the gene family is not significantly altered by transformation of the cells by transfecting plasmids encoding the SV40 early region. When mouse metallothionein-I promoter fusion genes (including pMK) are introduced into the cell-line (by electroporation) both expression and zinc regulation is noticed. Based upon stable transformation efficiency the mouse metallothionein-I promoter is 3-4 times less efficient than the SV40 early region promoter. Immuno-staining was used to detect the transient expression of the SV40 large T antigen (Tag) driven by either the SV40 early region (pSV3gpt) or the mouse metallothionein promoter (pMTLT). In the absence of zinc the number of cells detectably expressing pMTLT is 2% of that expressing pSV3gpt, following zinc induction the number of cells expressing pMTLT increases to 20% of that expressing pSV3gpt. The mMT-I promoter is therefore deficient in the establishment of expression compared to the SV40 early region promoter in sheep cells. Since both the endogenous sheep metallothionein genes and mouse metallothionein-I promoter fusion genes are active in the cell-line, itis concluded that an overriding, negative position effect is the most likely explanation for the failure of the pMK transgene to be expressed. The ability of viral and cellular oncogenes to transform the cell-line was investigated. Expression of SV40 Tag causes the cells to divide more rapidly, loose contact inhibition and anchorage dependent growth but does not lead to immortalisation. Clones established which express SV40 Tag have a vastly increased proliferative capacity. However, after 60-70 population doublings all clones enter crisis and die. Failure to express SV40 Tag is not the cause of crisis. No stable phenotypic effect is noticed when activated Ha-ras and c-myc (singly or in conjunction) or the bovine papilloma virus transforming region are introduced. A transforming murine p53 clone severely inhibits the plating efficiency of the cells. Attempts were made to re-claim the transgene for further analysis. Screening of a random genomic library failed to detect any positive clones. Attempts to clone a Hind III fragment containing the whole transgene from size fractionated DNA enriched for the fragment, were also unsuccessful.

No description available.

No description available.

The aim of this Thesis was to increase understanding of the regulation of gonadotrophin (LH and FSH) secretion in poultry with particular reference to ageing broiler breeder hens (<i>Gallus domesticus</i>). In broiler breeder hens, plasma FSH and LH were higher in peak-of-lay than in end-of-lay birds and this was correlated with higher numbers of yellow yolky ovarian follicles and pituitary common α-subunit mRNA. Old out-of-lay hens had higher pituitary FSH β mRNA and circulating FSH plasma levels than laying hens. This higher plasma FSH in out-of-lay hens was suggested to be due to a lifting of the inhibitory feedback effects of ovarian factors, probably oestrogen. Lifting feed restriction in laying broiler breeder hens increased the number of yellow yolky ovarian follicles. This was associated with an increase in hypothalamic GnRH-I and pituitary common α subunit and follistatin mRNA and plasma LH, while pituitary activin β_B mRNA and plasma FSH were depressed. There was no change in other mRNAs measured, including GnIH mRNA. The expression of incubation behaviour was associated with depressed plasma LH, decreased hypothalamic GnRH-I, common α and LH β subunit mRNAs and increased GnIH mRNA. In photosensitive short day female quail, exposure to 3 long days increased ovarian weight in association with increases in common α and FSH β subunit mRNAs with no change in LH β mRNA. It is concluded that the decrease in egg laying in broiler breeders is a consequence of decreased LH and FSH secretion. The decrease in LH secretion in older hens is associated with a reduction in common α-subunit mRNA with no change in LH β subunit mRNA. In contrast to mammals, common α subunit mRNA synthesis may be a rate-limiting step in LH release in birds. The decrease in FSH secretion in older hens may be a consequence of reduced FSH β mRNA and a reduction in constitutive release.

<I>Theileria annulata</I>, a protozoan parasite, causes a lymphoproliferative disease of cattle known as tropical theileriosis. This disease is a serious constraint to the improvement of indigenous cattle using crossbreeding in developing countries and is usually fatal in exotic and cross-bred animals if not treated. <I>T. annulata</I> infected tissue culture cell lines are used as a vaccine in most parts of the affected world. The draining lymph node (DLN) is believed to be the site for immune responses but the mechanism of responses in the DLN and afferent lymph was unknown when this work began. This thesis set out to investigate responses in the DLN and afferent lymph after cell line immunisation by immunohistological, cytokine and pseudoafferent cannulation studies. Immune responses in the DLN were examined by immunohistological and reverse transcription polymerase chain reaction (RT-PCR of cytokines) methods. Initial activation of T cells was observed in the paracortex (the normal anatomical site for activation) by day 4 after immunisation. Primary follicles (PF) were observed by day 4, developed germinal centres (GC) by day 12 which increased in size and number by day 16. A high, 1:10240, schizont serum antibody titer was recorded on day 16. A few T cells (T cells are critically important for GC formation) were observed in the PF. These increased by day 16 when well developed GCs were observed. Similarly, a few proliferating cells were noticed in PF by day 4. These had increased by day 9 and were numerous by day 16, by which time PF had became large GCs. VPM30<SUP>+</SUP> cells were found in the light zone in GCs on day 16 only. This molecule is expressed on differentiated B cells in the GC. Schizonts were observed in CD 11b<SUP>+</SUP> cells, this molecule is expressed upon monocyte/macrophages and granulocytes. Proliferating cells were observed by day 4 in the paracortex and by day 9 & 16 in the medulla. Results presented in this thesis showed that a normal immune response developed in the DLN with T cell activation and formation of GCs following cell line immunisation. Cytokine studies showed production of IFNγ followed by IL-4. This was in complete contrast to studies by other workers on lethal sporozoite infections where GCs were lost and only IFNγ was detected. It seemed as if the production of both IFNγ and IL-4 by T cells could serve as a marker for cell lines of low virulence. Cannulation studies indicated DLN to be the site of parasite transfer and showed: the circulation of naive T cells through the afferent lymphatics, the site of immunisation as the initial T cell activation site and NK-like activity at the site of immunisation.