DNA methylation status of nuclear transfer and twin cows
Beatrice de Montera, INRA, France
Successful somatic cell nuclear transfer (SCNT) cloning requires epigenetic reprogramming of a differentiated donor cell nucleus. Incorrect reprogramming of epigenetic marks such as DNA methylation occurs frequently and is associated with compromised prenatal development and postnatal abnormalities. Clones that survive into adulthood, in contrast, are assumed to be normal and are expected to be used in global food production in the near future. However, the epigenetic status of such healthy adult clones has never been investigated. Here we show that global DNA methylation levels of 38 healthy adult bovine clones are highly variable between individuals of the same genotype in comparison with monozygotic twins. We used a capillary electrophoresis technique to measure 5methyl-cytosine (5mC) content in white blood cells DNA from 38 healthy female clones (Holtein genotypes, n=4, Simmental genotypes n=5) at 1-8 years of age. Strikingly we found that the estimated variance in 5mC level within clone genotypes from both breeds (0.104) was higher than between clone genotypes (< 10^-10). To quantify the contribution of SCNT to the observed variability between clones, we compared Simmental clones with female Simmental monozygotic twins of similar age that were generated by bisection of fertilised embryos. In twins, the variability within genotypes (0.0091) was lower than between genotypes (0.0136). The DNA methylation levels in the great majority of clones were also higher than in twins. The comparison with twins confirmed a SCNT cloning effect illustrated by a clone-specific variability of DNA methylation and provided new evidence of DNA hypermethylation by SCNT. These data clearly demonstrate that even apparently healthy SCNT clones must be considered as epigenome variants.

Genetics and genomics of behavioural traits related to domestication and welfare
Per Jensen, IFM Biology, Linköping University, Sweden
The dominant goal for animal breeding has been increased production, and this has caused a number of undesired side-effects on welfare related traits. Correlated selection, which has a huge interest not only for welfare science, but also for domestication research, may in principle occur from one or all of three major mechanisms: pleiotropy, linkage or common regulatory pathways. In this review, examples are given on all three from our own research on selection effects on poultry behaviour and welfare. Pleiotropy is exemplified by the gene PMEL17, which affects plumage melanisation and simultaneously increases the risk of becoming a victim of feather pecking. Pleiotropy or linkage may explain the occurrence of QTLs affecting several traits simultaneously, for example growth, egg production and fear-related behaviour. Modified gene regulation can explain stress-related behaviour changes which can be transferred via the egg to the next generation. Genetic mechanisms such as these may help increasing our understanding of correlated selection responses and thereby allow more welfare-friendly breeding in the future.

Genetics and aggressive behaviour, welfare and product quality issues
Richard B. D'Eath, Simon P. Turner, Sarah H. Ison, Alistair B. Lawrence (SAC, UK); Gary Evans (Genus, UK), Ludger Thölking, Holger Looft, Esra Kurt (Genus (PIC), Germany); Klaus Wimmers, Eduard Murani (FBN Dummerstorf, Germany); Ronald Klont (Vion Foods, Netherlands); Aline Foury, Pierre Mormède (INRA, France)
Pre-slaughter stress has a negative impact on animal welfare but also on meat quality in various livestock species. In pigs, we hypothesise that a major factor in pre-slaughter stress is aggressive behaviour when pigs are mixed together for transportation to, or on arrival at, the abattoir. Aggressive temperament (aggressiveness) in the pig population is variable, and partly under genetic control.
In an experiment for SABRE WP8.2, we mixed pigs at a young age to identify individuals of high (H) or low (L) aggressiveness. Prior to transport and slaughter, single-sex groups of eight pigs were then mixed according to their aggressiveness in HH, HL or LL combinations or left unmixed. Each treatment was replicated in at least two groups in each of four slaughter batches. This design aimed to contrast extremes of social stress. Additionally, 'commercially normal' levels of social stress were studied in four further slaughter batches with no manipulation of group composition.
At slaughter we collected blood and loin muscle samples and counted carcass lesions to indicate involvement in aggression. Stress was measured by quantifying glucose, lactate, cortisol and creatine kinase in blood plasma, and meat quality measured through pH, drip loss and colour. The effect of treatment on, and relationships between, these measures will be presented.
Tissue relevant to the stress response (adrenal), immune response (leukocytes), metabolic balance and acute phase response (liver) were collected in addition to muscle and are being used for gene expression profiling by microarray to identify functional networks of genes and further elucidate the relationship between aggression, stress and meat quality.

Genetics and genomics of biological stress responses as related to product quality and welfare
Pierre Mormède, PsyNuGen, INRA, Université de Bordeaux, France
Evidence for genetic influences on HPA axis activity has been well documented in farm animal species. Differences among genetically defined stocks have been described in pigs or chickens and divergent lines have been obtained by genetic selection in poultry (in response to ACTH or to social stress), Japanese quail (response to immobilization), turkey (cold stress), trout (confinement stress). Functional consequences of HPA axis variability are diverse on production traits, resistance to disease, product quality, and newborn survival for instance.
It is therefore of interest to explore the molecular mechanisms underlying individual differences for further genetic selection of the optimal phenotype. Adrenal production of glucocorticoid hormones in response to ACTH is a major source of variation. Global gene expression in adrenals from Large White (low responders) and Meishan (high responders) piglets to ACTH showed a number of genes differentially expressed in several pathways contributing to cortisol synthesis. Bioavailability of cortisol is the second source of variation. We found by QTL analysis in pigs that polymorphisms of corticosteroid binding globulin (CBG) that carries cortisol in plasma, is involved in the higher cortisol levels measured in Meishan pigs and influences carcass composition and meat quality. Finally, the efficiency of receptors and post-receptor pathways is also subject to genetic variation. Further experiments on differential selection based on genetic markers will allow a more precise characterization of functional consequences of variations in HPA axis activity and the definition of selection objectives.

Multidimensional study of behavioural reactivity in sheep
Ivan Dimitrov, Agricultural Institute, Stara Zagora, Bulgaria
The purpose of the study was to examine the influence of behavioural reactivity (temperament) on the stress-induced physiological reactions in dairy sheep subjected to acute and chronic stressors during normal farm rearing practices.
Temperament was measured using a Complex Score (CS), reflecting the main behavioural traits of each animal in group situation during machine milking and then confirmed by different fear inducing and learning tests. A Principal Component Analysis for behavioural traits produced three factors: First - “Feeding”; Second - “Milking”; Third – “Social interaction”. On the basis of the CS, three temperament groups reflecting significantly different behavioural responses to milking environment (p<0.001) were established: Calm (C) – 75 ewes, Intermediate – 89 ewes and Nervous (N) – 71 ewes.
Endocrine responses of C (n=20) and N (n=20) ewes were measured after three treatments: 1) Machine milking; 2) Shearing; 3) Machine milking 24h after shearing. Plasma cortisol levels were significantly higher in N ewes after machine milking (C-8.1±1.4nmol/l; N-23.7±6.1nmol/l; P<0.05) and shearing (C-19.9±5.2nmol/l; N-38.5±6.1nmol/l, P<0.05). Plasma lysozyme levels were higher after shearing for C ewes than for N ewes (C – 0.102±0.014µg/ml; N – 0.065±0.005µg/ml; P<0.05) but did not differ after the other treatments. This study shows that animals with different temperaments also expressed different physiological responses to stressful events commonly encountered during modern sheep management. Sheep displaying high behavioural reactivity also showed higher physiological responses to stressful events than sheep displaying lower behavioural reactivity. These results may have implications for selection and management of dairy sheep to improve welfare and milk production.

Genetic improvement of robustness
Roel F. Veerkamp, Animal Breeding and Genomics Centre, Animal Sciences Group, Wageningen UR, Lelystad, the Netherlands
Dairy farmers call for more robust cows as exemplified by the interest in cross-breeding and other breeds. It can be questioned whether selection for health, fertility and longevity in an average environment addresses this request for robustness sufficiently. Especially when robustness is defined as “the capacity to handle disturbances in common, sustainable and economical farming systems”. Several genetic concepts exists that may allow selection for robustness more directly, including or not the interaction with environment. A first concept is to select against (or for) environmental sensitivity using reaction norms. Some studies indicate that in a continuously improving environment, selection for increased performance leads to increased sensitivity (i.e. reduced robustness). The second concept that may enable for direct selection for robustness is to account for genetic heterogeneity of environmental variance. Genetic heterogeneity of environmental variance can be considered as genetic differences in environmental sensitivity to random micro environmental fluctuations. A third concept that might enhance selection for robustness directly might be to take into account the social effect that an animal has on its herd mates. Such social effects appear to be important for improving robustness in other species (Bijma et al. 2007), and might be of interest in dairy cattle, for example, to avoid spreading of mastitis within a herd or improving feeding and grazing behaviour at herd level. These aspects of robustness in interaction with selection for improved milk quality, will be investigate din EU project RobustMilk.

Combining QTL and gene expression studies to understand the genetics of mastitis susceptibility
Peter Sørensen, University of Aarhus, Denmark
The major goal of SABRE's WP5 is to understand the genetics of mastitis. Insights into the genetic background of complex traits such as mastitis are gained from two complementary experimental approaches. The first approach (QTL/association studies) identifies variations in genomic regions that are associated with the trait. These genomic regions may contain tens or even hundreds of genes and identifying the causative genes is difficult. The second approach (expression studies) on the other hand identifies changes in expression levels of genes/proteins/pathways that associate with the trait, but this approach does not provide information about the genetic variation underlying the trait. In this talk we will present a study that combines the two experimental approaches. We identified a QTL for mastitis susceptibility in Danish Holstein cattle. From the QTL region we determined haplotypes with high or low high susceptibility (H/L) to mastitis. A pathogen specificity test showed that the QTL was associated with E-coli mastitis. In an expression study we then compared the expression profile in liver and udder tissues of cows with high or low haplotypes in the resting state and in response to E-coli pathogen challenge.

Functional genome analysis of fertility in cattle reproduction
Dawit Tesfaye, Institute of Animal Science, Animal Breeding and Husbandry Group, University of Bonn, Germany
With increasing productivity, especially in milk yield, there is a parallel decline in fertility of the animals which is manifested as abnormal ovarian cyclicity, poor estrus expression and reduced oocyte quality, which is leading to fertilization failure and early embryo loss. As direct genetic selection for fertility is hampered by a relatively low heritability, knowledge on the regulation of cascade of processes underlying successful reproduction will pave the way to increased utilization of genetic merit of dairy cattle by breeders. In SABRE’s WP6, partners are working with the objectives of investigating the genomics of important aspects of fertility mainly female sexual behaviour (oestrus expression and follicle development), embryo survival and endometrium receptivity. Moreover, it is aimed to identify segregating QTL (BTA7) affecting conception rate in dairy cows. Together with the competent blastocyst, endometrium plays a central role for successful pregnancy. In addition, the level and type transcripts orchestrated in endometrium during the pre-transfer period of the cow may have a profound effect on the conceptus to come to term. Large-scale transcriptome analysis of endometrium biopsies derived at day 7 and 14 of the pretransfer cycles based on subsequent pregnancy success showed significant differences in the relative abundance of genes involved in various pathways. Moreover, blastocyst biopsies obtained prior to transfer to recipients were used for transcriptome analysis based on the pregnancy success after transfer to recipients. This could lead to identification of potential candidate genes directly related to the fate of the embryo till term.

Interaction of pigs and chickens with their environment and zoonotic disease
Tom Humphrey, University of Bristol, UK
Most pigs and chickens in the EU are produced intensively with the animals being reared at high stocking densities. Such conditions may be regarded as “unnatural” and widespread concern has been expressed about the welfare of these animals in intensive production, with pressure groups being particularly concerned about chicken production. Such concerns will only increase when final broiler stocking densities are permitted to rise from 38 to 42Kg/M2.
Contaminated chicken and pork are major vehicles for Salmonella infection and chicken is the principal means by which people acquire Campylobacter. Both pathogens can be invasive in foods, being isolated from liver and muscle. Such internal contamination poses a significant threat to public health, as the bacteria may survive cooking better than ones on meat surfaces. Under-cooked chicken meat is an important vehicle for human Campylobacter infection.
Research in Bristol has examined the impact of social and production-related stresses in pigs and chickens on the behaviour of the above two pathogens. Our work on Campylobacter and chickens encompasses field and laboratory studies. We have found, from study of 789 commercial housed broiler flocks, that carriage of Campylobacter, especially at high frequency, is significantly more likely if flocks are infected with APEC and/or there are high levels of hock and/or pad burn. In laboratory studies, Campylobacter jejuni was highly invasive in broilers when it had been pre-treated with Noradrenaline, simulating spread from “stressed” animals, or when chronic stress was simulated in these animals. These and other food safety-related issues will be discussed. 

Genetics and genomics of boar taint – ongoing research in Norway
Eli Grindflek, Norsvin, Norway
Male slaughter pigs are normally castrated early in life to prevent boar taint in the meat. In the future, castration is likely to be prohibited due to animal welfare concerns. We are therefore encouraged to find (genetic) solutions to reduce the incidence of boar taint in uncastrated males. Boar taint is primarily caused by elevated levels of androstenone and/or skatole in the adipose tissue. Androstenone is a steroid produced in testis and degraded in liver, and skatole is a metabolic product of the amino acid tryptophan and is absorbed from gut and metabolised in liver. Selective breeding is likely to be a realistic approach for reducing levels of boar taint. Selecting for low androstenone, however, generally coincides with selection for low androgen production, which might influence reproduction and fertility of the boars. Before starting selection it is therefore necessary to understand the complex genetic system controlling boar taint, and to detect genetic markers useful for selection. Comprehensive studies of gene expression profiles of testis and liver from boars with extreme high and low levels of androstenone have been performed. Moreover, a genome scan using the 7.6K Illumina chip has been done in 2800 purebred Norwegian Landrace and Duroc boars, with the phenotypic records androstenone (fat and plasma), skatole, indole, testosterone, 17β-estradiol, estrone sulphate, length of bulbo urethralis gland, as well as several economic traits. The same animal material is used in examination of associations between single nucleotide polymorphisms (SNPs) in candidate genes and all the available traits. A number of genes involved in biosynthesis and metabolism of androstenone were detected and will be useful in the future selection program.

The genomic region responsible for E. coli F4ab/ac susceptibility in pigs: Characterization and Application
Mette Juul Jacobsen, Alan Archibald, Susanna Cirera, Gloria Esteso, Bjarne Nielsen, Inger Edfors, Leif Andersson, Merete Fredholm and Claus B. Jørgensen
Approximately, 170.000 pigs in Denmark succumb each year to infection by enterotoxigenic Escherichia coli (ETEC F4ab/ac). An effort to reduce the prevalence of this infection will thus have a significant impact on pig welfare and greatly diminish the use of antibiotic treatments in pig production.
In 1975 Sellwood and coworkers published a paper that described the existence of two pig phenotypes in relation to ETEC F4, namely resistant pigs and susceptible pigs, respectively. In 1977, Gibbons et al. showed that ETEC F4ac resistance was inherited as an autosomal recessive Mendelian trait and linkage to the transferrin locus (TF) was suggested and later confirmed (Guérin et al. 1993).  Linkage mapping of the porcine loci responsible for susceptibility towards ETEC F4ac and F4ab confirmed the segregation with TF and indicated the regional localization on pig chromosome 13 (Edfors-Lilja et al., 1995), but no candidate gene was found or suggested. This means that the only available diagnostic test for this type of ETEC F4 resistance is the adhesion test developed by Sellwood et al. in 1975. Since the adhesion test either demands major intestinal surgery or slaughter of the pig and the fact that it is very laborious makes it difficult to incorporate it into breeding programs. The prerequisites that need to be fulfilled in order to incorporate a diagnostic test into a breeding program are that it needs to be quick, easy to use and allows precise genotyping of living animals. In this respect a DNA-based test is by far the most preferable.
By the use of a family material phenotyped using an ETECF4 adhesion assay and linkage analysis the genomic region containing the gene responsible for ETEC F4 resistance/susceptibility has been narrowed down to pig chromosome 13q41 (SSC13q41). SNPs segregating with susceptibility/resistance have been identified in intronic regions of MUC4 and one of these SNPs is presently used in selection programmes by the pig breeding industry. The current status of the breeding programme will be addressed along with the impact of the breeding programme on other traits in the Danish breeding index.
In order to elucidate the molecular mechanisms involved in ETECF4ab/ac infection in pigs a detailed haplotype map of SSC13q41 has been constructed. By a targeted approach, the SSC13q41 region was subjected to SNP screening mainly focusing on intronic sequences. A total of 25 genes were partially sequenced and SNPs were identified in ACK1, APOD, CENTB2, GP5, KIAA0226, LRCH3, MUC13, MUC20, MUC4, OSTa, PCYT1a and ZDDHC19. Overall, 163 SNPs were discovered in the parental generation. This detailed haplotype map of the SNPs serves as valuable resources for further fine mapping of the locus involved in ETECF4ab/ac adhesion and ultimately for the identification of the responsible mutation.

QTL for eggshell quality
Johanna Vilkki, MTT Agrifood Research Finland
The objective of SABRE WP7 is to develop genomic tools to improve eggshell quality by selection to enhance resistance to structural failure and bacterial penetration. One aim is to scan the chicken genome for eggshell quality QTL. The information on the QTL will be combined with the information from functional analyses of gene expression in the shell gland. Candidate QTN or markers tightly linked to the QTL will be tested in independent population samples to verify the effect. A scan of 27 chromosomes has been completed using 162 microsatellite markers. Autosomal linkage groups were analysed with QTLExpress. The analyses were extended using a new epistasis module developed by WP1 for GridQTL. Additive effects on the Z chromosome were analysed by linear regression. In all, 23 QTL affecting eggshell quality were found on the autosomes. Each QTL explains 2-5 % of the phenotypic variance of the trait. Genome-wide significant QTL were found on chromosomes 2, 6 and 14, and additional suggestive QTL cluster on these chromosomes as well as on chromosome 3. On the Z-chromosome, a cluster of 5 QTL affecting both eggshell breaking strength and deformation was found within one marker interval. One genome-wide suggestive epistatic pair was detected. Overall, it seems that epistatic effects do not play a significant role in the genetics of eggshell quality in this mapping population, and consequently do not affect the fine-mapping strategy.

Quality Improvements: implementation in practice
Gary Evans, PIC/Genus plc.
One of the first molecular genetics discoveries to be used in animal breeding was relevant to both welfare and quality. A DNA point mutation discovered by Fujii et al, in 1991, was found to be responsible for porcine stress syndrome – a group of conditions in pigs which includes acute stress, sudden death and undesirable pale, soft and exudative (PSE) meat characteristics. This discovery led to subsequent commercial availability of a DNA test (HAL-1843™) to determine genotype for the “halothane” gene.  The use of this DNA test by the pig breeding industry has resulted in the ability improve pig welfare by selecting against the recessive mutation to eradicate porcine stress syndrome and also improve meat quality by reducing the occurrence of PSE meat.
Since then, QTL mapping and candidate gene analysis has led to the discovery of many more causative mutations and linked marker polymorphisms that are related to welfare and quality traits e.g., RN, PRKAG3, CAST (all meat quality) and FUT1, MUC4 (disease resistance). Some of these DNA markers have been used in commercial pig breeding programs by integrating the molecular genotype information with quantitative genetic selection decisions through Marker Assisted Selection (MAS) to improve quality and welfare related traits.
This presentation will review some commercial aspects of quality and how current and future quantitative genetic and genomic developments may be applied to improve these traits in commercial pig breeding programs.