Genomic Selection for Bovine Tuberculosis Resistance in Dairy Cows

The bacterium, Mycobacterium bovis, has a major economic, trade, health and welfare impact on the cattle industry worldwide as well as posing a risk to humans, other domesticated, feral and wild animal populations. This pathogen causes the chronic respiratory disease, bovine tuberculosis (bTB), which remains an increasing problem in cattle herds in the UK and Republic of Ireland despite over sixty years of costly eradication programmes. These programmes included the slaughter of animals which are positive for a skin test which indicates that the animal has become infected. Alternative control strategies are urgently needed. Previous studies have suggested that cattle differ genetically in their risk of bTB, opening up the possibility of genetic selection for decreased risk of bTB. Breeding livestock for more favourable traits is becoming faster and more accurate through advances in genomic resources and information, including new genotyping tools such as high density single nucleotide polymorphism (SNP) chips. These 'chips' consist of thousands of SNP markers which can relate variation across the genome to variation seen in traits. In our previous BBSRC CEDFAS grant using bTB cases (defined as skin test and lesion positive) and herd-matched Holstein-Friesian controls from Northern Ireland, we could account for ca. 25% of the observed variation in bTB status using chips comprising over 700,000 SNPs. We now propose to build on this information and combine it with datasets derived from other cattle populations in the UK and Republic of Ireland (for which we have obtained reciprocal permission in principle) to develop robust genomic predictors of bTB risk. These could then be directly applied by the cattle industry to select for bTB resistance.

First we will enhance the power of our original study by genotyping a further set of cattle which are skin test positive but lesion negative. This will also enable us to clarify the genetic relationship between this phenotype and our more strict definition of a bTB case. We will then conduct a meta-analysis combining this enriched dataset with that from other sources, and use this large dataset to develop genomic predictors of bTB risk. The results will provide a direct tool to the dairy industry, enabling it to select for increased bTB resistance without a continuing requirement to collect bTB phenotypes from cattle in currently infected herds. In addition, we will ensure that selection for bTB resistance is not detrimental to production traits by determining the genetic relationship of bTB resistance with milk production and other economically important traits in the UK breeding goal.

We also want to find the actual SNPs that lead to the genetic differences we detect in bTB resistance. To do this, we propose to take advantage of the fact that it has now become feasible to sequence whole genomes of individual animals; this usually reveals many novel SNPs. Furthermore, genetic changes due to insertions and deletions (indels and copy number variants) in the DNA sequence are also increasingly associated with variation in traits and may underpin disease resistance as well. In order to investigate this type of variation and also identify SNPs closer to the actual causative SNPs, we propose to resequence animals with the most extreme bTB risk, as determined by the genomic predictors developed in the earlier part of the grant. This information, i.e. identification of the actual DNA changes associated with increased resistance, would improve the accuracy of the genomic predictors across generations and potentially have utility in other breeds. These results will also enable us to explore the underlying basis for resistance to M. bovis infection, which could advance our ability to design further control strategies for this intractable disease.

Bovine tuberculosis (bTB) is a chronic respiratory disease caused by the pathogen, Mycobacterium bovis. Despite many years of statutory testing and slaughter, bTB still has major impacts on the health and welfare of UK cattle and the livestock economy. Substantial evidence now exists that the risk of a herd contracting this disease is variable and has a strong genetic component. We propose to build on our recent results from a genome wide association study (GWAS) in which we genotyped ca. 1200 cases and controls for 750,000 SNPs in the major UK dairy breed, the Holstein. We identified a marker-based heritability of 0.24 and chromosome wide significance for specific loci associated with bTB resistance. We now plan to provide accurate genomic predictors for bTB risk which can be directly applied by the dairy industry by combining our current data with additional datasets in a meta-analysis. We will exploit our unique biobank of DNA samples collected from bTB phenotyped cattle from high prevalence herds, to explore whether skin test positive, lesion negative animals have similar or distinct genotypes to our previous case definition of skin test positive, lesion positive animals. Further datasets available from collaborators consist of Holstein sires with estimated breeding values for bTB risk, and another case/control study. These have been genotyped at lower density compared to our samples. The meta-analysis of 5 distinct datasets will enable us to develop enhanced genomic predictors of bTB resistance calibrated by cross-validation, an essential requirement for GWAS studies. Predictors for complex traits in livestock based on whole genome sequencing are now on the horizon; we plan to whole genome sequence selected high and low bTB risk sires to provide novel information that could lead to further accuracy of the predictors as well as identifying novel SNPs closer to the causal loci, potentially identifying putative candidate genes for bTB resistance in cattle.

The goal of this proposed project is to help tackle the seemingly intractable problem of bovine tuberculosis (bTB) in the UK, caused by infection with the bacterium Mycobacterium bovis. This disease has shown a year on year increases in incidence, in cattle herds in both the UK and Republic of Ireland (ROI), especially over the last decade. We aim to take a novel approach to help combat the issue, by providing essential science-led information to the dairy industry, enabling it to utilise advanced genomic breeding tools for selecting for increased resistance to bTB. The major output from this project will be an enhancement of the health, welfare and productivity of UK cattle, adding to food security and contributing toward evidence based policy-making at a national and international level. This leads to both economic and societal impacts, as bTB is a costly and controversial problem for the UK government. It has major economic impacts on the UK livestock industry, especially as the UK does not enjoy bTB-free status, unlike the rest of the EU. Furthermore, the impact of bTB on the farmers dealing with this problem is immensely distressing, with many farms being under statutory restrictions on animal movement for considerable periods of time. Thus, any measures to lessen the disease impact will be beneficial at many levels. Our project will provide information and tools directly applicable to the dairy industry through DairyCo who are supportive and will facilitate the necessary technology transfer. The information provided will enable farmers to avoid using highly bTB susceptible sires and cows, and choose bulls which are expected by genomic prediction to be more bTB resistant. Uptake may be expected to be regional and dependent on individual circumstances, in that use by breeders may well be greatest in high bTB prevalence areas such as Northern Ireland, Wales and South West of England. Utilising accurate genomic predictors would be expected to also have a positive feed-back in that as ever-more resistant bulls are employed in a herd, the overall level of infection should be reduced, resulting in greater herd protection. Furthermore genomic selection for bTB resistance can be used as a complementary control measure alongside other strategies, and could have a positive benefit to other control efforts. Adoption of genomic predictors for bTB resistance would move towards providing a green and sustainable solution to the issue, and thus make major contributions to environmental sustainability, protection and impact reduction. Our proposal should also be seen as public-friendly as it should reduce the emphasis on badger controls and reduce the impacts of the hugely costly test and slaughter schemes. The project also has other benefits. It would have impact on the UK academic community by enhancing the knowledge economy, particularly for those engaged in animal health and agricultural/food production research. The project has potentially even greater benefit to the worldwide farming fraternity as our information will be invaluable in other countries where bTB is rife. It will provide information that could be applicable directly to Holstein-Friesian cattle herds globally, as well as potentially to other breeds of cattle. State-of-the-art technologies will be employed to tackle an important health trait and thus it will represent a proof-of-principle that could be applied to other infectious diseases of livestock. Furthermore, the project will provide training in the latest genomic technologies, and the PDRA employed on the project will become a highly skilled researcher. In summary, this is a proposed project with many direct and indirect benefits. The potential returns on investment for the research should be very high.