Final Report Summary - 3SR (Sustainable Solutions for Small Ruminants)
Executive Summary:
With 100 million sheep and 11 million goats in the EU and a combined global population of around 1.94 billion heads, these two species are of considerable economic, social and environmental importance.
Mastitis and nematode worm infections are the two biggest health and welfare threats to sheep and goats in Europe and beyond, causing huge economic and productivity losses. Previous research has indicated that susceptibility to both infections is at least partly under genetic control. Understanding the basis of genetic control to susceptibility will help direct breeding programmes to produce more resistant and, therefore, healthier and more productive animals. Ovulation rate is the key factor determining the number of lambs per litter for sheep. Some particularly prolific breeds have been identified along with some important genes that affect fertility but further work has been required to fully understand the genetic basis of control.
Co-funded by the EC’s FP7 programme, the Sustainable Solutions for Small Ruminants (3SR) project set out in 2010 to:
1. Develop and verify (in commercial populations) selectable genetic markers for traits with high importance for health and sustainability in sheep and goats, and to identify the causative mutation(s) where possible, for a) Mastitis susceptibility; b) Resistance to nematodes, and c) Ovulation rate.
2. Develop tools and resources that are needed to achieve Objective 1 and provide a lasting resource for future small ruminant genomics research, specifically: a) Improved assembly and comparative and functional annotation of the sheep genome, b) Tailored SNP sets suitable for fine-mapping of polymorphisms relevant to sustainability traits in sheep and goats, and c) Improved resources for goat genomics, especially a higher resolution genome map
3. Disseminate results from the project to ensure a broad and long-term beneficial impact on European competitiveness and EU policy on animal health, welfare and sustainable agriculture.
The 3SR consortium worked closely with the complementary global efforts on-going through the International Sheep Genomics Consortium (ISGC) and the International Goat Genome Consortium (IGGC), and played a key part in the development of a well-annotated sheep reference genome sequence (OARv3.1) that will provide the international sheep genomics community a valuable tool for underpinning further research and a caprine 50K SNP chip which provides unparalleled opportunities to explore and understand nematode resistance in goats, to detect QTL for resistance, and to derive novel breeding strategies.
The consortium has furthermore developed and used cutting-edge genomic technologies such as an open-access genome browser, and produced large amounts of genome-related data (genotypes for 7,500 animals, genome sequences for 50 sheep, and over a terabyte of RNA-seq data). This has already led to the identification of causative mutations, candidate genes, SNPs and genetic markers for regions that affect the production of offspring in sheep, susceptibility to mastitis and nematode resistance in sheep and goats.
These tools and findings may assist breeders to produce animals with enhanced health and prolificacy, delivering significant benefits to productivity within the sheep and goat farming industries.
Project Context and Objectives:
Importance of the sheep and goat industries
Small ruminants (namely sheep, Ovis aries, and goats, Capra hircus) were the first livestock species to be domesticated around 10,000 years ago in the margin of the Fertile Crescent. They have since diffused all over the world following migratory and exploratory movements of humans and thereafter routes of commercial trade. Now, small ruminants are a major component of farm-animal production all over the world, with about 0.9 billion goats and 1.1 billion sheep.
The sheep and goat sectors play an important role in the economy of many EU countries, often underpinning the economic viability of rural communities. Their ability to make good use of low quality pasture as well as being ideally suited to being kept on small family farms makes them extremely valuable. Due to their small size and grazing preferences, and through often being highly adapted to local conditions, they also play an essential role in landscape maintenance and nature conservation.
Challenges being faced
The future sustainability of the sheep and goat sectors relies on breeding animals that are resilient against disease challenges, as well as being productive and efficient. Understanding the mechanisms of resistance and developing genomic tools and technologies is likely to be crucial in helping to address these challenges.
Mastitis, paratuberculosis and parasitic infection from nematodes are currently ubiquitous chronic diseases within large parts of the European sheep flock and goat herd. Many cases go largely undiagnosed, due to the extensive nature of many farming systems and the subclinical nature of the diseases, however their significance should not be underestimated.
Mastitis
The impact of mastitis is significant in both the dairy and meat industries. For dairy sheep it is estimated that up to 55% of ewes have sub-clinical mastitis while less than 5% have clinical mastitis. Losses in milk production have been reported to range from 20-37% in affected ewes. Making conservative assumptions of a 10% incidence of mastitis in the EU sheep flock as a whole, with a mean reduction in milk yield of 20% and using €650/tonne as the wholesale price of sheep milk, it can be conservatively estimated that the cost to the EU is €37million per annum.
Mastitis is the single largest reason for the culling of ewes (up to 50%) and results in increased veterinary expenses as well as impacting on the welfare of the animals. For meat sheep, lamb growth from affected ewes is less than lambs from healthy ewes. One study demonstrated that increasing mastitis resistance by 10% in the Texel breed in the UK would save £2.7million per annum. There is less data available for goats but it is expected that reducing mastitis susceptibility in this species could also lead to significant savings.
Nematode infection
On a global scale, ruminant diseases caused by gastrointestinal nematode parasite infections are the diseases with the greatest impact upon animal health and sustainability. Within Europe, nematode infections are the most costly endemic disease in meat production systems, and are close behind mastitis in milk production systems.
Welfare and cost implications of parasitism in sheep and goats are significant with even sub-clinical infections causing marked decreases in productivity; for example, the annual cost of nematode challenge in the sheep industry is estimated at £84million in the British sheep industry and $500m in Australia. Cost savings have been shown to be proportional to the severity of nematode infection therefore a 10% improvement in resistance, in this case, would reduce annual costs by £8.4million.
Ovulation rate
Ovulation rate is an indicator of prolificacy. For ovulation in sheep, increasing our knowledge of the ovulation rate genes that have been identified so far could lead to improvements in flock productivity and profit. Better knowledge of these genes will lead to better control and more informed decision making by the sheep industry.
Climate change
In the context of climate change, characterising the molecular mechanisms underlying the adaptation of small ruminants to harsh and different environments is of particular interest, both from a theoretical point of view and for identifying new valuable genetic resources for regions that will be subject to climate change. Characterising the genetic diversity of small ruminants requires the availability of genomic tools such as complete genome sequences.
Sheep and goat farming will also help address the EU core objective of sustainable development , as defined by the Brundtland Report (1987): “Development that meets the needs of the present without compromising the ability of future generations to meet their own needs.” Better use of selective breeding of livestock is arguably the single best way to improve the sustainability of animal agriculture, with the added benefit that improvements made are permanent and cumulative. Moreover, addressing these issues through selective breeding will contribute to the EU strategy for sustainable development launched in 2001 and the EU Sixth Environment Action Programme (6EAP) , which have set the overall objective of decoupling economic growth from environmental degradation.
Potential benefits from genetic and genomics
For a variety of reasons, most notably the fragmented nature of the goat and sheep industries globally and their lower economic worth in the United States, sheep and goats have lagged behind other major livestock species in the development of comprehensive genomic information that can inform selective breeding.
It is only recently (and aided by the work of the 3SR consortium) that high quality, draft reference genome sequences for sheep and goats have been published. With 100 million sheep and 11 million goats in the EU and a combined global population of around 1.94 billion heads, these two species are of considerable economic, social and environmental importance. This is particularly the case in marginal areas, and the benefits that can be delivered by better understanding the genetic basis of health and welfare traits are highly significant.
To remain sustainable, sheep and goat producers will need to continue to address the challenges of disease and improved efficiency. Genomic tools and technologies are likely to play a crucial role in helping to address these challenges.
Mastitis has a profound, negative effect on sheep and goat health and welfare, causing udder pain and swelling, lethargy, anorexia and even death. Current mastitis management is based on prevention, through good hygiene, nutrition and treatment, using intravenous antibiotics. None of these is entirely successful and the use of antibiotics raises concerns about the sustainable management of the condition. Increasing genetic resistance to mastitis offers a truly sustainable and complementary adjunct to existing strategies.
For the effective control of nematode parasites, farmers have come to rely almost exclusively on broad-spectrum anthelmintics. Issues relating to resistance, residues and eco-toxicity are increasingly challenging the sustainability of chemotherapy. Anthelmintic resistance is widespread and escalating, threatening to erode the efficacy of these drugs. All too often, a structured regime of anthelmintic treatment is not followed, resulting in ineffective control and inadvertent increased selection for resistance. There are also environmental concerns associated with anthelmintic use. The Ivermectin class of drugs, currently the most effective anthelmintics, are excreted at high concentrations in the faeces of treated animals and have a potent effect on the insects that feed on these faeces . Clearly the current situation is unsustainable - genetic improvement is a sustainable, cost effective alternative strategy that has yet to be fully exploited but which 3SR has aimed to help to redress.
Paratuberculosis is a debilitating disease of both sheep and goats. Currently there is no effective treatment for its control, and breeding for resistance to paratuberculosis may be the only sustainable method for its control.
Sheep account for 9% of EU enteric methane emissions from livestock . Increasing the efficiency of sheep and goat farming therefore reduces the amount of methane emitted for a given level of production, improving the carbon footprint of the sector.
An international challenge
As small ruminant genomics research is a relatively small field, the leading experts in these different areas are dispersed throughout Europe but also further afield in the USA, Australia, New Zealand, China and Argentina.
At the same time, genomics is big science requiring expensive large-scale activities, such as sequencing, assembly, curation and annotation of a genome sequence for each species. The costs and work for the development of these resources therefore need to be shared across as many users as possible and duplication of effort needs to be strenuously avoided.
A project to tackle the genomics of sheep and goats, thus, required an international and inter-disciplinary framework to effectively integrate the varying expertise, and to ensure the maximum progress.
The international sheep and goat research communities are collaborative rather than competitive. It is therefore not surprising that the leading European expertise in small ruminant genomics and key players in the International Sheep Genomics Consortium and the International Goat Genome Consortium collaborated in the 3SR project. This has created a unique collaboration to ensure that the EU is well positioned to take advantage of the genomics revolution in breeding.
The major EU countries within which sheep and goat production occurs in Marginal Areas, and where capabilities for genomics research exist, are France, Spain, Italy, Greece, UK and Ireland, and each of these countries is represented in the 3SR Consortium by at least one organisation, enabling the greatest access to tools, expertise and relevant resources, and short lines to dissemination and implementation of results in commercial populations.
The involvement of partners from outside the EU also aids the development of European expertise in small ruminant genomics, ensuring that the EU is well positioned to take advantage of the genomics revolution in breeding. This allowed the project to be much more efficient than it could be at a national level; it avoided duplication of effort and has produced a common approach with agreed protocols.
Aims and objectives
The aim of the 3SR project was to discover selectable genetic markers (single nucleotide polymorphisms (SNPs)) and causative mutations – where possible in commercial populations – for traits that are critical to sustainable farming; namely mastitis susceptibility, nematode resistance and ovulation rate. Once identified, the consortium sought to verify the markers in additional commercial populations.
To achieve this, 3SR had set itself three high-level objectives:
1. To develop and verify (in commercial populations) selectable genetic markers for traits with high importance for health and sustainability in sheep and goats, and to identify the causative mutation(s) where possible, for:
a. Mastitis susceptibility
b. Resistance to nematodes
c. Ovulation rate (sheep only)
2. To develop tools and resources that were needed to achieve Objective 1 and provide a lasting resource for future small ruminant genomics research, specifically:
a. Improved assembly and comparative and functional annotation of the sheep genome
b. Tailored SNP sets suitable for fine-mapping of polymorphisms relevant to sustainability traits in sheep and goats
c. Improved resources for goat genomics, especially a higher resolution genome map
3. To disseminate results from the project to ensure a broad and long-term beneficial impact on European competitiveness and EU policy on animal health, welfare and sustainable agriculture.
In order to deliver on these aims and objectives, the 3SR consortium set out to apply the latest high throughput genomics technologies, comparative and functional genomics, combined with targeted genome sequencing and extensive in silico analyses to dissect important genetic components controlling these traits. In order to deliver significant improvements in available genomic information and technologies for these species and have a lasting impact on European research capacity.
To make this feasible, the consortium worked closely with the complementary global efforts on-going through the ISGC and the IGGC, and made use of complementary resources provided by major research projects in Europe, Australasia, USA, Argentina and China.
The 3SR project has brought together an international consortium, to deliver a step-change in the understanding of the genetic basis of traits underlying sustainable production and health.
Project Results:
Project overview
The overall aim as part of the 3SR project was to develop molecular markers that can be used to select for small ruminants that are healthier and more productive, and can make a higher contribution to sustainable production in the EU.
In order for this to be achieved, the 3SR project:
i. Included the full breadth of complementary research needed to develop useful markers, consisting of “discovery” through to “validation” in commercial populations.
ii. Focused primarily on markers for susceptibility to mastitis, resistance to nematodes and ovulation rate, but also included analysis of data for production traits and susceptibility to paratuberculosis.
iii. Made best possible use of the wealth of resources available through the 3SR Consortium members to ensure the best value for the requested funding.
iv. Maintained very close links with the ISGC and IGGC to ensure full advantage of the most up-to-date resources available, and also to ensure that the resources developed by 3SR contribute positively to the international effort to develop genomic resources for sheep and goats.
The project activity was divided into seven interlinked work packages, as shown in the project overview (Figure 1 in Appendix 1).
Work Package (WP) 1 focused on developing the genomic resources and tools needed for the project as a whole. The tools, in terms of genome-wide or region-specific SNP arrays, sequences for specific genome regions or radiation hybrid maps, were produced in close collaboration with the trait-specific work packages. WP1 also ensured that state-of-the-art genomic tools and techniques were used throughout the project.
Three trait-specific WPs subsequently utilised the various outputs of WP1, focusing on mastitis susceptibility (WP2), nematode resistance (WP3) and ovulation rate (WP4), while also making full use of additional data recorded at no additional cost to the project, including standard performance traits and paratuberculosis resistance.
The SNP genotyping and functional studies performed in these three WP2, WP3 and WP4 were used by WP5, where the genomics data were fully mined and candidate genes nominated for inclusion on validation SNP arrays developed in WP1.
Thus the research iterated between these five WPs, with greater precision of results arising from each iteration, resulting in:
1. Selectable markers for use in sustainable breeding programmes;
2. A significant contribution to the activities of the ISGC and IGGC.
As research outputs are of little value without effective communication and dissemination, WP6 was trusted with the task to ensure that knowledge gained is optimally disseminated, translated and exploited, with close contact maintained with all potential end users.
Lastly, the task of effective management at all levels of a large and complex project as 3SR was placed within WP7. In addition, WP7 ensured that quality standards were maintained; and that reporting and auditing requirements were met. IPR arising from 3SR was managed jointly by WP6 and WP7. Careful ethical monitoring throughout the project ensured that any (potential) public concerns were met and that 3SR aligned with the “3Rs” (Replacement, Refinement and Reduction) of animals in research.
Key results
3SR’s close collaboration with the ISGC and IGGC has led to significant improvements in the genomics resources available for sheep and goats. The 3SR project has contributed to improving the quality of the reference sheep genome by sequencing animals, it also contributed to the annotation of the sheep genome by performing large scale RNA sequencing.
Activity highlights:
• Sequencing of around 50 animals has been carried out to help improve the quality of the reference sheep genome (coordinated by the ISGC), while large scale RNA sequencing has been performed to help with the sheep genome annotation.
• Helped the IGGC to develop a high density (50K) SNP chip for goats.
• Genotyped over 4000 sheep and 2000 goats using the respective SNP50K chips.
• Over 7000 50K genotypes analysed from 11 sheep populations and over 2000 50K genotypes analysed for goats
• Controlled challenge studies were carried out for mastitis (in sheep and goats) and nematodes (sheep) to explore the mechanisms of resistance
• Functional studies of genes involved in ovulation rate in several sheep breeds were carried out to support the biological understanding of the mutations discovered.
Key results:
• Identified causative mutations affecting ovulation rate in Grivette (French) and Olkuska (Polish) sheep, and 2 candidate genes for sterility were identified in Cambridge sheep.
• In both sheep and goats, small regions of the genome were identified that have a significant effect on mastitis susceptibility and nematode resistance. Genetic markers for these regions were validated in commercial populations.
• The 3SR consortium helped the ISGC deliver sheep reference genome v3.1
• The 3SR consortium produced over a terabyte of RNA-seq data to assist with the annotation of the sheep genome
• A 50K SNP chip has been developed for goats. This enabled over 2000 animals to be genotyped for selectable genetic markers of interest within the 3SR project.
• Confirmed genetic variation to paratuberculosis in sheep (heritability 0.28±0.10)
• An open access Genome Browser for sheep and goats was developed http://genome.itb.cnr.it/gb2/gbrowse/oarv3.1/
• A whole-genome radiation hybrid (RH) map was produced for goats, containing more than 30,000 markers ordered across the chromosomes
• Led the establishment of the goat AdaptMap project which aims to coordinate and collate the research data across a number of projects (including 3SR)
• As well as the reference genomes, the genomic tools (for example genotyping arrays) have been greatly advanced throughout the duration of the project
• Around 4000 sheep from 11 populations with relevant phenotypic data have been genotyped, using the 50K ovine SNP chip
• A further 3400 genotypes from the same populations and funded by other national projects, have been made available for analysis in 3SR.
• Technical bulletin providing an easy to understand overview of the key activities and results in 3SR
State of the art
When the 3SR project was first proposed in 2009, the genomic resources for sheep and goats were limited and falling behind those of the other major livestock species. An International Sheep Genomics Consortium (ISGC) had previously been formed, and by early 2009 the IGCS’s work had led to the availability of improved sheep maps, a sheep Bacterial Artificial Chromosome (BAC) library and associated end sequences, the first Ovine 50K SNP Beadchip (“SNP50K chip”) and a first draft of the sheep genome assembly. This first draft was estimated to cover around 80% of the unique sequence in the sheep genome, indicating that there were still significant gaps in the sequence, and therefore quite a bit of work needed to improve the quality of the reference genome for sheep.
The situation for sheep was better than that for goats. In 2009, no concerted international research effort to develop caprine genomic resources had been established, and consequently no reference genome sequence was available for the goat, although there was a low density (1536) SNP chip.
A number of external events occurred in 2010 and throughout the duration of the 3SR project that provided a valuable opportunity to increase the potential impact and the value for money that could be achieved through the 3SR project. These opportunities resulted mostly through the close links between the partners in the 3SR consortium and the wider international communities for both sheep and goats.
The main changes and opportunities that were harnessed were as follows:
1. In March 2010 the Beijing Genomic Institute (BGI) announced that they had sequenced the entire genome of a female Texel sheep. During the remainder of 2010 BGI worked with the ISGC and 3SR partners to use the information generated to improve the quality of the ovine reference assembly.
2. In the summer of 2010 AgResearch in New Zealand coordinated an international order for the ovine SNP50K chip. This provided an opportunity for 3SR partners to purchase the chips at a substantially lower price than usual. By delaying the SNP chip order in 3SR by a few months we were able to take advantage of this opportunity and dramatically increase the number of SNP50K chips that could be purchased through the project, rather than developing targeted, low density 1536 chips. Participation on the international order also provided an additional benefit. Some of the consortium partners were able to change their genotyping strategy to use SNP50K chips in other projects that were funded through national sources, and these genotypes were also made available for inclusion in the analyses for 3SR.
3. An International Goat Genome Consortium (IGGC) was formed in March 2010. A number of the 3SR partners are part of this consortium. Coordinated international activity was initiated in three areas: de novo goat genome assembly and re-sequencing; development of a goat Radiation Hybrid (RH) panel and mapping; and production of a caprine SNP50K chip. 3SR was closely involved in the delivery of these activities and was able to use the caprine SNP50K chip for the genotyping, delivering greater value compared to the original plan.
4. These international developments also allowed more 3SR staff resources to be focused on developing tools and resources for goats, which further increased the impact that could be achieved by the 3SR project.
Tools and resources
To achieve the aims and objectives, 3SR project partners have applied the latest high-throughput genomics technologies, comparative and functional genomics, combined with targeted genome sequencing and extensive in silico analyses to dissect important genetic components controlling their traits of interest (and where possible the causal mutations).
By working closely with the ISGC and the IGGC, the project has been able to deliver a significant improvement in the available genomic information and technologies for these species, whilst also having a lasting impact on European research capacity.
In addition, 3SR has benefited from the use of complementary resources provided by major research projects in Europe, Australasia, USA, Argentina and China through key partners within the consortium that are involved in these projects.
The tools and resources that have been developed include an improved assembly, comparative and functional annotation of the sheep genome, and improved resources for goat genomics.
An overview of the most significant resources and tools developed and used are provided below.
Sheep and goat research populations
The project made use of data from over 10,000 sheep from 15 diverse breeds in 10 countries and over 4,000 goats from 3 breeds in 3 countries.
One of the strengths of the 3SR project is the large number and variety of breeds that are available for use through the partners. These include over 10,000 sheep and 4,000 goats, all of which have been individually recorded. Many of these animals have been genotyped as part of the project, primarily with the relevant SNP50K chip. In addition, specific 1K SNP chips were developed in the project for the purposes of validation. Details are provided in Figure 2 (in Appendix 1).
The sheep and goat populations used in the 3SR project are involved in research applied to one of the three sustainability traits considered. There is one exception amongst the sheep: a Sarda-Lacaune backcross research flock from Sardinia has been sufficiently intensively recorded to allow the data to make valuable contributions to work on all three traits.
Sarda-Lacaune Backcross sheep population
This experimental population was established in 1999 by INRA and AGRIS for the purpose of aiding the detection of QTL associated with milk production and functional traits including somatic cell count (SCC) and faecal egg count (FEC). The population was formed by mating animals from the two most numerous dairy breeds from the respective countries, the Lacaune (France) and the Sarda breed which is predominantly found in Sardinia (Italy). Although two dairy breeds, they differ phenotypically for many traits, such as body size, growth rate, wool, prolificacy, milk yield and milkability.
The mating of 14 elite Lacaune rams to 184 Sarda ewes formed the first generation of the cross. Ten sons from each of ten founder sires were then mated to 2,719 ewes to generate 968 backcross females, which included approximately 98 from each of ten families. A further four structured matings were subsequently conducted and all retained ewes were fully recorded. DNA and records were also collected on more recent descendants from the initial backcross animals.
Tools for mining genomic information
Activities of the 3SR partners have resulted in large amounts of genome-related information. In order to facilitate collaboration on sheep and goat genome studies, a web platform was created with a repository to collect results from the sheep genome annotation with tools for customisable data extraction, visualisation and analysis. This has been updated and maintained consistent with the state-of-the-art of the 3SR project, and has been designed such that new functionalities may be added to facilitate results integration and analysis.
The Sheep and Goat Genome Browser
The genome browser allows users to visually inspect genomic annotations. Small amounts of custom annotation can be uploaded, thus adding a custom track on the genome viewer. Annotation can either be inserted manually or by file upload, and visualisation configuration can be customised.
It relies on the generic genome browser database, GBrowse (http://gmod.org/wiki/GBrowse) containing all the data produced by the 3SR members and used as a viewer for the sheep genome, with new tracks added as soon as they become available.
Available tracks include mapped bovine RefSeq genes, sheep gene predictions and eight tracks of SNPs from different submitters. The GBrowse database is, and will continue to be, the repository for all the sheep-related information that will be made available during the project. Browsing via the GBrowse interface allows extraction of data related to genomic regions of interest and downloading of partial or complete datasets. Data can also be downloaded as gff3 files or exported as images (suitable for publication).
QTL regions are visualized in GBrowse and QTL-driven searches can be performed via the GoSh interface to retrieve genes and other features included in the QTL of interest. Every additional feature (i.e. promoters, alternative splicing sites, etc.) can be displayed in GBrowse tracks, and dedicated web pages, based on the structure of the GoSh interface, can be added, containing full information for the selected feature.
The 3SR GBrowse includes GLEAN and GeneWise gene predictions, mapped bovine RefSeqs, mapped ESTs and EST clusters, SNPs, tRNA predictions, tandem repeats and mapped TFBS and canonical promoter elements. MiRNA hairpins and mature sequences positioned by BLAST are also available.
To incorporate the new reference sequence data available for goat, a genome synteny browser was developed, based on the GMOD GBrowse syn structure (http://www.gmod.org/wiki/GBrowse_syn). GBrowse syn can be used to view multiple sequence alignment data, synteny or co-linearity data from other sources against genome annotations provided by GBrowse. The sheep and goat synteny browser includes the sheep genome, version 3.1 (Oar-v3.1) and the goat genome, version 1.0 (http://goat.kiz.ac.cn/GGD/).
The Genome Browser (open access):
http://genome.itb.cnr.it/gb2/gbrowse/oarv3.1/ http://www.livestockgenomics.csiro.au/sheep/
http://genome.itb.cnr.it/gb2/gbrowse_syn/sheep3.1-goat/
Genome Mining Tool (GMT)
The Genome Mining Tool (GMT, http://155.253.6.237/3SR_gmt) is a web interface dedicated to 3SR partners that focuses on defined genomic regions. It includes pre-defined regions relevant to the study of mastitis and nematode resistance, and also allows user-defined regions to be selected. Registered users can add, delete or modify regions. Once a region is selected, a list of the RefSeq gene models and SNPs is given, containing links to related resources. Every region or feature is also linked to the GBrowse (see above).
The GMT also hosts a page for a pathway-oriented view of the trait of interest. In this page, tentative QTLs for mastitis or nematode resistance are analysed as a whole, and a list of pathways that involve genes from those regions is given, together with the number of genes for each pathway. This number is linked to the full list of genes.
The bioinformatic structure of the GMT is based on a MySQL database. Each genomic feature can be inserted in this database, provided it can be defined by an absolute position on the sheep genome. Custom scripts allow data conversion from the GBrowse or Pathway Commons formats to the input formats suitable for upload in the GMT database. Selection of features in regions is performed dynamically via the web interface.
Genome Mining Tool (GMT) (restricted access)
(http://155.253.6.237/3SR_gmt)
The 3SR GBrowse and GMT web interfaces are dedicated to the collection and display of data produced within the activities of the 3SR Project. Both interfaces can contain data from any genomic feature, provided an absolute position on sheep reference genome (version 3.1) is given. The retrieval system offered by the GMT offers the possibility to focus on pre-defined genomic regions, including all tentative QTLs for mastitis and nematode resistance, or on user-defined test intervals, and to obtain pathway-oriented views of the traits under investigation.
The following describes some of the features of the tools developed.
• GMT users can explore and visualise all the records included in putative QTL regions.
• Records can be modified by authorised users or displayed by project partners.
• All regions are linked to the genomic browser: by clicking on the “Open Gbrowse” link, the selected region will be displayed in GBrowse in a separate web page.
• Users can retrieve all the genes and all the SNPs within the selected region by clicking on the “Genes” and “SNPs” links.
• Featured annotations are also available by following links.
• Data can be downloaded in an Excel- (or other spreadsheet) compliant format.
Sequence data for sheep
The proposed sequencing strategy in 3SR was updated throughout the project as the sheep reference genome was improved so that the quality of the new assembly in the genomic regions of interest to 3SR could be assessed. As the quality of assembly in those regions was found to be good, the sequencing strategy was amended to increase the benefit that could be achieved through use of the project resources.
In order to leverage the activities and resources available outside the project, activities in 3SR were concentrated around the following areas:
• Genotyping using SNP50K chip
• Improving the genome assembly using available sequencing information
• Re-sequencing of whole genomes for individual animals
• BAC sequencing
• CNV detection
• Sequencing of RNA form a wide range of tissue to help with annotating the reference genome sequence
• Construction of high density RH maps of the sheep and goat genomes
• Generation of validation chip.
This has provided a big impact in terms of information to help improve the quality of the reference assembly for sheep but also to understand what the sequence means.
Details on a number of these activities are provided in the following paragraphs.
Sheep genome v3.1
Version 3 of the sheep genome (“Oar v3.1”) was built by the ISGC in close collaboration with 3SR and the USDA, with a Post Doc at CSIRO, supported in part by 3SR, undertaking most of the work in building Oar v3.1. The sequencing was undertaken by 3SR partner The Roslin Institute (UK), Beijing Genomics Institute (China) and Baylor College of Medicine (USA).
Oar v3.1 was built using next generation sequence derived from one female and one male Texel. The primary de novo assembly was performed using ~75 fold Illumina GA sequence from the female with several rounds of filling gap steps undertaken using both the female and the male Texel.
The major improvements are the large reduction in gaps, including in GC rich regions around the start sites of genes. Many of the remaining problematic areas are in regions orthologous to regions of the bovine genome with copy number variation, and cannot be solved using currently available resources.
The work on Oar v3.1 has been published in the journal Science:
Yu Jiang, Min Xie, et al., and Brian P. Dalrymple. (2014). The sheep genome illuminates biology of the rumen and lipid metabolism. Science 344, 1168-1173
DOI: 10.1126/science.1252806
http://www.sciencemag.org/content/344/6188/1168.abstract?sid=93cc6e14-c2c5-428d-b9a8-72957a5d1d61
Improvements continue to be carried out beyond the end of the 3SR project, with the first priorities for improvements to reduce the number of large gaps and provide greater coverage of segmental duplications and copy number variation regions.
Oar v3.1 can be downloaded from: www.livestockgenomics.csiro.au/sheep/oar3.1.php and www.ncbi.nlm.nih.gov/bioproject/PRJNA169880.
Genotyping using ovine SNP50K chip
Largely as the result of the participation in a large international order, the number of SNP50K genotypes that were available for analysis as part of the 3SR project increased by eight-fold compared to the original plan, increasing from 900 to nearly 7,500 across 12 populations representing 11 different breeds/crosses. More detail on the numbers of each population of interest is shown in Figure 3 (in Appendix 1).
Re-sequencing of whole genomes for individual animals
Sequencing of key animals has been carried out to help improve the quality of the reference sheep genome (coordinated by the ISGC), while large-scale RNA sequencing has been performed to help with the sheep genome annotation (Figure 4 in Appendix 1).
Construction of high density radiation hybrid (RH) maps for sheep and goats
Construction of a high-density RH map of the sheep genome has been completed to assist with assembly of the goat genome. The whole-genome RH map produced for goats contains more than 30,000 markers ordered across the chromosomes.
INRA and USU constructed a whole-genome high-density RH map of the ovine genome using two ovine RH panels that were developed at both institutes. A specific genotype calling algorithm, based on raw fluorescence intensities provided by the Illumina platform, was developed because the genomes of RH clones differ in many ways from the genomic DNA of diploid individuals.
From the 49,035 SNPs of the ovine SNP50K BeadChip, 41,999 could be reliably called on both RH panels using the specific new calling algorithm. RH maps were constructed for each chromosome using the comparative approach developed at INRA and were constructed using the sheep assembly as the reference. At the completion of the analysis, RH maps for all the sheep autosomes were developed, containing a total of 31,162 markers. This high-density RH map provides an orientation of these ovine SNPs along the goat genome, thereby linking the genome assemblies of goat and sheep. This information will be useful as detected QTL regions in the goat are comparatively mapped in the sheep.
Chinese partner HZAU constructed a 5,000 rad caprine radiation hybrid (RH) panel and genotyped the goat panel with approximately 108,000 SNP markers on the Illumina bovine SNP50K and ovine SNP50K chips. A robust RH map was constructed in collaboration with INRA, containing 30,988 SNP markers covering all 30 caprine autosomes and the X chromosome at an average spacing of approximately 85 Kilobases (Kb). This high-density RH map has been used to verify the assembly of the de novo goat genome sequence.
The marker order in the caprine RH map is 96.1% consistent with the sheep genome and 90.9% with the cattle genome. A comprehensive caprine–mammalian comparative map has been generated by identifying the locations of caprine RH-mapped loci within the genomes of sheep, cattle, human and other mammalian species.
Development and employment of caprine 50K SNP chip
A notable success of 3SR has been the development of the caprine 50K SNP chip (Tosser-Klopp et al., 2014), which provides unparalleled opportunities to explore and understand nematode resistance in goats, to detect QTL for resistance, and to derive novel breeding strategies.
The release of the goat genome sequence in 2010 by the Beijing Genomics Institute paved the way for an international effort towards constructing projects and tools for goat genomics. In the frame of the International Goat Genome Consortium (IGGC), and with the support of 3SR and the French dairy goat industry, a caprine 50K SNP chip was released by Illumina at the end of 2011.
The success of Genome Wide Association Studies (GWAS) in the discovery of sequence variation responsible for, or closely linked to, complex traits in humans has increased interest in high throughput SNP genotyping assays in livestock species. The primary goals with this technology are QTL detection and genomic selection.
The success of a moderate density SNP assay depends on the reliability of bioinformatic SNP detection procedures, the technological success rate of the SNP design, even spacing of SNPs on the genome and selection of Minor Allele Frequencies (MAF) suitable to use in diverse breeds.
Data collected within the 3SR project and other international resources were combined within the IGGC (www.goatgenome.org) to design of a 50-60K SNP chip for goats.
Approximately 12 millions high quality SNP variants were identified within and between 6 different breeds (meat, milk and dual purpose: Alpine, Boer, Creole, Katjang, Saanen and Savanna), using a total of 97 animals.
Whole genome and Reduced Representation Library sequences were aligned on >10kb scaffolds of the de novo goat genome assembly, and then evenly spaced on the goat genome.
60,000 SNPs were selected and submitted to Illumina for oligo manufacturing. The final manufactured bead-pool consists of 53,347 probes that passed manufacturing quality control. Ten different breeds were then used to validate the SNP content and 52,295 loci could be successfully genotyped and used to generate a cluster file.
The caprine SNP 50K chips were successfully used for QTL detection in 3SR for mastitis susceptibility and nematode resistance.
The 60,000 SNPs were also submitted to the dbSNP database (http://www.ncbi.nlm.nih.gov/projects/SNP) as a resource for international academic research, along with flanking sequences and map position on goat assemblies (i.e. scaffolds and pseudo-chromosomes), sheep genome V2 and cattle UMD3 assembly.
Inclusion of the goat genome in a comparative genomic database
An efficient exploitation of complete genomic sequences requires the availability of dedicated bioinformatic tools. One particularly powerful approach to analyse genomic sequences is the comparative genomic approach, which takes advantage of the existing whole genome sequences of evolutionary-related species.
In order to enable a straightforward exploitation of the newly sequenced goat genome in a comparative genomic perspective, 3SR partner Utah State University (USU) compared the goat genome to 15 mammalian completely sequenced genomes (Human, cattle, sheep, pig, horse, dog, cat, mouse, rat, rabbit, chimpanzee, gorilla, orang-utan, marmoset and platypus). The results are available through the Narcisse comparative genome browser (http://narcisse.toulouse.inra.fr) providing a view of the difference in the evolution of the various species.
Traits: Mastitis susceptibility
Introduction
Amongst infectious diseases in dairy ruminants, mastitis is of major importance because of its high frequency and related costs. In small dairy ruminants, mastitis is the primary health reason for involuntary culling, and prevalence of subclinical mastitis is 20-30% per lactation.
Accumulating research results over the last decades give strong evidence that the host’s response to control udder health is under genetic control in dairy ruminants. Existence of QTL for mastitis resistance in sheep and cattle had been previously identified. Among studies, the common mastitis resistance phenotype was milk somatic cell count (SCC), measured periodically over lactation.
Mastitis susceptibility (sheep)
Project partners used the ovine SNP50K chip and subsequently performed Genome Wide QTL analyses to identify regions of significant genetic variation (“QTL”) in mastitis susceptibility in four sheep populations. Twenty-nine QTL
s associated with SCC were found in the Churra (7 regions), Lacaune (7) and Sardinian-Lacaune Backcross populations (15). QTLs had rather moderate effects and were mostly population specific, in agreement with the polygenic nature of the trait and the complexity of the resistance phenotype. However, a few commonalties among populations were found, for example on chromosomes 3, 14 and 20. One
on chromosome 5 was also associated with mastitis resistance (intra mammary abscesses) in the divergent lines of Lacaune sheep.
One of the major findings was a highly significant QTL located on a narrow region (<0.5 Mb) of chromosome 3 in the Lacaune population. This QTL was detected at the same location by all analysis methods that were performed. Subsequently, whole genome sequencing of a trio of animals led to the identification of a putative causal mutation on chromosome 3 associated with increased susceptibility to mastitis in the Lacaune breed. A functional assay for evaluating the protein binding affinity, suggests a functional knockout of the socs2 gene in homozygous susceptible sheep.
Mastitis validation studies in sheep
Results obtained in the project, and described above were used to develop and verify (in commercial populations) selectable genetic markers for mastitis susceptibility. Thereto a custom-made ovine SNP chip of 960 SNPs was designed and developed. This array was used to genotype four independent sheep populations: Chios, Churra, Manech Red Faced and Lacaune. This highlighted a number of regions and SNPs significantly associated with mastitis, of which the most significant results are described below.
Chios sheep
In Chios sheep, a total of 18 SNPs were found to affect one of the mastitis related traits (either clinical mastitis or indicator traits such as milk SCC, total bacterial count in milk and CMT). Importantly, none of these SNPs was associated with milk production, implying that this information can be directly used to genetically control mastitis without compromising the productivity of the animals.
Lacaune sheep
Three SNPs, located in the narrow region of a QTL found on chromosome 3, were found to be significantly associated with the LSCS mastitis trait in the Lacaune population. The two most significant ones were located in the SOCS2 gene (R96C non synonymous) and CRADD genes (intron). This reinforces the hypothesis that the SNP in the SOCS2 gene is likely to be causative, although a direct effect of other polymorphisms in this narrow region still cannot be excluded.
Mastitis susceptibility (goats)
Genotyping of the Saanen and Alpine breed populations with the newly released caprine SNP50K chip resulted in the identification of significant QTL for traits related to mastitis resistance.
At the start of the 3SR project, there was no information available on chromosomal regions controlling mastitis resistance in goats, nor was there a commercial SNP tool for goats.
The previously described release of a caprine SNP50K chip enabled the project to undertake the first genome–wide association studies (GWAS) to map loci describing significant genetic variation in mastitis susceptibility based on the SCC trait. The number of loci are shown in Figure 5 (in Appendix 1). Results are presented for significant QTL found in LA and LD, with many more significant QTLs found with the LD method compared to the LA method, and a higher number of QTLs found in the Saanen breed than in the Alpine breed.
3SR partners were able to genotype an additional population of 192 Alpine goats from a divergent selection experiment with the caprine SNP50K chip in order to further confirm and fine map those QTLs.
Among the numerous regions detected, eleven were of particular interest because results were consistent across at least two methods or two populations. In most of the cases, a QTL detected in the Saanen population was confirmed in the total population (on chromosomes 1, 17, 18, 22, 26, 2 and 28). In addition, other regions on chromosome 16, 19 and 21 were of high interest because they were detected in the two breeds and because the significance level was especially high (chromosome 19). Those regions are of primary interest for further fine mapping and validation in independent populations.
Mastitis validation studies in Skopelos goats
From the QTL discovery in Alpine and Saanen goats, three QTL regions on OAR16, 19 and 21 that were common in the two breeds were selected for designing a custom-made chip (576 SNPs) in order to validate the QTL in a third breed: a total of 415 Skopelos dairy goats from Greece were genotyped with this caprine array. Although these goats had monthly phenotypic records for SCC, along with milk yield, fat and protein % at each measurement time, the SNPs could not be found. This was not unexpected, as the heritabilities for all mentioned traits were low, even though all traits were strongly repeatable within and across lactations.
Conclusions
The methods used in this study have been successful in identifying QTL for mastitis resistance in sheep and goats using their respective ovine and caprine SNP50K chips. The results are consistent with the complex biology controlling mastitis susceptibility.
One of the major findings for sheep was the identification of a putative causal mutation on chromosome 3 associated with increased susceptibility to mastitis in the Lacaune breed, likely to be a functional knockout of the socs2 gene in homozygous susceptible sheep.
Comparative mapping between sheep and goat mastitis furthermore showed that a QTL on goat chromosome 19 was in the homologous sheep region carrying a QTL for Lactation Somatic Cell Score (LSCS) in the Lacaune dairy sheep population OAR11.
Traits: Nematode resistance
Introduction
On a global scale, ruminant diseases caused by gastrointestinal nematode parasite infections are the diseases with the greatest impact upon animal health and sustainability. Within Europe, nematode infections are the most costly endemic disease in meat production systems, and are close behind mastitis in milk production systems.
Grazing ruminants invariably face nematode challenges, although actual infestation levels vary according to the environment and climactic conditions. The widespread development of anthelmintic resistance in parasite populations means that alternative control strategies are required, with genetic selection being a long-term sustainable option. Many studies have shown that resistance to nematode parasites is a heritable trait in both sheep and goats, irrespective of the species of nematode infecting the host.
Nematode resistance (sheep)
Analyses have been performed to identify QTL for nematode resistance in Blackface lambs, Martinique Blackbelly x Romane backcross lambs and in the Sardinia-Lacaune Backcross population.
QTL regions associated with nematode resistance were found in 7 regions in the Blackface lamb population, in 7 regions in the Martinique Blackbelly x Romane backcross lambs and in 12 regions in the Sardinia-Lacaune backcross population.
The most significant QTL, validated across several families, was on chromosome 12, which was highly significant at the genome-wide level in the Martinique Blackbelly x Romane backcross, and also appeared in the Sardinian-Lacaune backcross population. In general however, the QTL were spaced throughout the genome with little commonality across studies. Only three common regions were observed:
• OAR9 (16-29 Mb) in the Sardinia-Lacaune backcross and Blackface populations
• OAR12 (40-46 Mb) in the Martinique Blackbelly x Romane and Sardinia-Lacaune backcross
• OAR21 (21-40 Mb) in the Martinique Blackbelly x Romane and Blackface populations
These results provided the basis for further exploration of nematode resistance, for meta-analyses combining diverse experiments, and for detailed genome exploration using full sequencing.
Meta-analyses (or, more specifically, joint analyses) of sheep nematode resistance SNP chip data suggested that genomic prediction for nematode resistance and body weight may be of value in closely related animals, but that with the current SNP chip genomic predictions are unlikely to work across breeds.
Nematode resistance (goats)
Although much of the 3SR focus has been on resistance traits in sheep, it is resistance to nematodes in goats that is of paramount importance, as goats are especially susceptible to gastrointestinal nematodes. Furthermore, it is often observed that anthelmintic resistance develops faster in goats than sheep. This may be related to the evolutionary history of goats, in which they have been preferential browsers rather than grazers, and hence throughout much of their domestication history have been able to avoid parasites. Nematode infections become particularly important, when goats are forced to be obligate grazers. This is especially the case in tropical environments such as Guadaloupe, as in these environments Haemonchus contortus infections are particularly severe and constant. Hence, survival of goats under such conditions requires them to be relatively resistant. Breeding strategies to improve nematode resistance under such conditions are therefore particularly important.
The caprine SNP50K chip, developed by joint efforts of the IGGC and 3SR, was employed to explore and quantify the genetic and genomic control of nematode resistance in goats with a view to derive enhanced breeding strategies. Successfully detecting and quantifying genetic variation for nematode resistance traits in Creole goats under tropical conditions in Guadeloupe verified the utility of this SNP50K chip (Figure 6 in Appendix 1). Indications of significant genetic regions were also found, some of which lay on the same chromosomes of significant QTL found in previous studies in the same population, albeit on previous generation animals.
A long term programme of work on this Creole goat population has now verified beyond doubt the wide genetic variation in nematode resistance, with various appropriate breeding strategies developed. The SNP50K chip results strengthen these results and give additional options for genetic improvement. However, it appears that there are no major QTL effects detectable in this population, and that the considerable genetic variation in this population is largely polygenic (i.e. influenced by more than one gene). This means that simple marker-assisted selection is unlikely to be effective, with the optimal strategy being a combination of phenotypic and targeted SNP-chip-assisted (i.e. genomic) selection.
Conclusions
The results for the sheep nematode resistance SNP chip data suggested that genomic prediction for nematode resistance and body weight may be of value in closely related animals, but that, with the current SNP chip, genomic predictions are unlikely to work across breeds.
Genetic variation for nematode resistance traits in Creole goats under tropical conditions in Guadeloupe has been detected and quantified with the use of the caprine SNP50K chip that was developed by joint efforts of the IGGC and 3SR. The caprine SNP50K chip may be suitable for detecting the presence of major QTL effects for nematode resistance or other traits in other populations/breed. This gives additional options for breeding strategies to address sustainability traits in many goat populations worldwide.
Traits: Ovulation rate (sheep)
The number of lambs born per ewe (“litter size”) is underpinned by ovulation rate and is an important measure of productivity on sheep farms and therefore the economic viability of a sheep production system. The identification of major genes controlling ovulation rate is therefore of interest to sheep farmers.
Within 3SR ovulation rate was only researched in sheep and followed two general approaches:
i) The identification of new mutations that control with ovulation rate in breeds with a reputation for high prolificacy, and
ii) The identification of genes/polymorphisms in the Cambridge breed that are associated with streak ovaries other than those in the BMP15 and GDF9 genes that have been previously identified.
Changes in the genomic sequence of the BMP15 (Bone Morphogenetic Protein 15) gene have been long known to be associated with effects on fecundity. A total of six different changes with effects on fecundity have been previously identified in various breeds. In each case the base change was found to result in higher fecundity in females that carried one copy, but resulted in sterility when a female carried two copies.
Grivette and Olkuska:
Grivette and Olkuska sheep are highly prolific and each breed carries a major gene affecting ovulation rate. A total of 40 Grivette and 63 Olkuska animals were genotyped with the ovine SNP50K chip. A genome-wide association study mapped the gene close to the previously identified BMP15 gene. Sequencing the BMP15 gene in the Grivette and Olkuska populations identified a new mutation in each breed. Both mutations result in a non-conservative amino acid substitution; T317I in Grivette and N337H in Olkuska.
In order to investigate the effect of the T317I mutation further and to verify its effect in the population, 151 adult Grivette ewes chosen at random in the population were genotyped for this mutation. The mean prolificacy of wild type, heterozygous and homozygous carriers was 1.93 1.98 and 2.31 respectively, demonstrating that animals homozygous for the mutation had an increased ovulation rate. The functional effect of this mutation was also tested in vitro after directed mutagenesis.
To investigate the effect of the N337H mutation in Olkuska, ovulation rate was compared in ewes that were wild-type, heterozygous and homozygous carriers for this mutation. Mean ovulation rates were 1.63 1.93 and 4.34 for the respective 3 genotypes, demonstrating that ewes homozygous for the mutation had a significantly higher ovulation rate.
Cambridge:
Cambridge sheep are also highly prolific and are characterised by high ovulation rate with occasional sterility due to ovarian hypoplasia. Extreme variation between individuals is consistent with segregation of a major gene controlling ovarian function. The inheritance pattern of sterile cases suggests an autosomal recessive gene unlinked to the known GDF9 mutation.
A total of 21 Cambridge animals (9 sterile ewes and 12 carriers) were genotyped with the ovine SNP50K chip (Figure 7 in Appendix 1). Homozygosity mapping identified genomic areas for which sterile ewes presented large homozygous segments. For a region of homozygosity on chromosome 2 it was possible to determine 2 regions (112Kb and 1.2Mb) putatively carrying the causative mutation. The smaller region encodes 2 strong candidate genes that have been sequenced to further investigate this mutation.
Localisation of major genes for ovulation rate
The number of SNPs and genomic regions that were found to be significantly associated with ovulation rate in each of the sheep breed investigated were: Cambridge (2 regions), Lacaune (5 SNPs), Davidsdale (2 SNPs), Grivette (one region and one SNP) and Olkuska (10 SNPs).
In the Lacaune, two markers on chromosome 7 and 8 were found to be significant (P<0.05) at the genome level, while 5 markers on chromosome 15, 21 and 25 were significant at the chromosome level.
No significant markers were found (P<0.05) at the genome level in the Davisdale. However, 2 markers on chromosome 24 and X were found to be significant at the chromosome level.
In the Grivette, no significant (P<0.05) markers were found at the genome level, but 15 were significant at the chromosome level. Five ovulation rate-related SNPs on chromosome X are very close to the BMP15 gene (312 Kb).
After sequencing the BMP15 gene, 3SR partners detected a new polymorphism (change of nucleotide C→T) at position 45935166. This non-conservative mutation is located in the regulatory region of the gene, and is strongly associated with the high prolificacy phenotype (P<0.001). This new SNP is also highly significant at the chromosome level when added to the other ovine SNP50K. If this SNP is the causal mutation, it would lead to a totally new phenotype for this gene: in contrast with all other known mutations of BMP15, homozygous animals would not be sterile and would present even higher ovulation than heterozygous animals.
The highly prolific Cambridge breed contains known mutations in the BMP15 and GDF9 genes, and females homozygous for either mutation display ovarian hypoplasia, commonly referred to as streak ovaries. However, some cases of streak ovaries in the breed cannot be explained by these mutations. Previous pedigree analysis has suggested the presence of a polymorphism in another autosomal recessive gene, unlinked to either BMP15 or GDF9, which can lead to sterility. The location and action of this gene and polymorphism is as yet unknown.
Analysis of the data generated by 3SR identified 2 highly relevant SNPs on chromosome 2, which could be the causative mutation for the Cambridge sterile phenotype. Further genotyping will need to confirm their causality; although they are located in the small region that was identified at the beginning of the project, other potential genomic areas have to be definitively excluded.
Traits: Added-value traits
Genome-wide QTL analyses of added-value traits (milk production, udder type traits, milking ease), measured in the three discovery populations (Churra, Sardinian-Lacaune Backcross and Lacaune), were performed. Several traits were highly heritable and although they did not show convincing SNP associations, the results are strongly suggestive of many of these traits being polygenic in nature. This being the case, genomic selection may be a route to exploitation.
Paratuberculosis (sheep)
Paratuberculosis, also known as Johne’s disease, is caused by Mycobacterium avium subsp. paratuberculosis (Map) that infects the digestive tract of young animals during the first few weeks of life and can prove fatal.
The population used to study mastitis and parasitism resistance (Sardinia-Lacaune Backcross population was reared in a flock where paratuberculosis was endemic for at least 10 years. Several phenotypic traits (detailed in Figure 8 in Appendix 1) related to Map infection were recorded.
3SR partners were the first to confirm the existence of a genetic variability in the antibody response against Map in sheep by estimating the heritability for this trait as 0.28±0.10. This heritability is higher than that for Map in cattle (0.10-0.15) and is also higher than previously estimated for sheep (0.15) which was done using post-mortem analysis rather than antibody response (as was done in 3SR). This is consistent with the existence of genetic variation in susceptibility to Map and forms an important basis for future studies on the identification of genetic loci linked to susceptibility to Map.
The existence of genetic variation in susceptibility to paratuberculosis may lead to the identification of disease resistance markers and consequently selection may be used as an additional tool for control of the disease.
Potential Impact:
With 100 million sheep and 11 million goats in the EU and a combined global population of around 1.94 billion heads, these two species are of considerable economic, social and environmental importance. The sheep and goat sectors play an important role in the economy of many EU countries, often underpinning the economic viability of rural communities. Their ability to make good use of low quality pasture as well as being ideally suited to being kept on small family farms makes them extremely valuable. Due to their small size and grazing preferences and through often being highly adapted to local conditions, they also play an essential role in landscape maintenance and nature conservation.
The future sustainability of the sheep and goat sectors relies on breeding animals that are resilient to disease challenges, as well as being productive and efficient. Understanding the mechanisms of resistance and developing genomic tools and technologies is likely to be crucial in helping address these challenges.
Mastitis and nematode worm infections are the two biggest health and welfare threats to sheep and goats in Europe and beyond, causing huge economic and productivity losses. Previous research has indicated that susceptibility to both infections is at least partly under genetic control. Understanding the basis of genetic control to susceptibility will help direct breeding programmes to produce more resistant and, therefore, healthier and more productive animals.
Ovulation rate is the key factor determining the number of lambs per litter for sheep. Some particularly prolific breeds have been identified along with some important genes that affect fertility.
Prior to the 3SR project, the genomic resources for sheep and goats lagged behind the other major livestock species, putting the industry at a disadvantage in terms of the potential to implement selective breeding.
It is only recently (and aided by the work of the 3SR consortium) that high quality, draft reference genome sequences for sheep and goats have been published. The benefits that can be delivered by better understanding the genetic basis of health and welfare traits are highly significant.
An International Sheep Genomics Consortium (ISGC) had previously been formed, and by early 2009 the ISGC’s work had led to the availability of improved sheep maps, a sheep Bacterial Artificial Chromosome (BAC) library and associated end sequences, the first Ovine 50k SNP Beadchip and a first draft of the sheep genome assembly. This first draft was estimated to cover around 80% of the unique sequence in the sheep genome, indicating that there were still significant gaps in the sequence, and therefore quite a bit of work needed to improve the quality of the reference genome for sheep.
The situation for sheep was better than that for goats. In 2009, no concerted international research effort to develop genomic resources had been established, and consequently no reference genome sequence was available; although there was a low density (1536) SNP chip.
A number of external events occurred in 2010 and throughout the duration of the 3SR project that provided a valuable opportunity to increase the potential impact and the value for money that could be achieved through the 3SR project. These opportunities resulted mostly through the close links between the partners in the 3SR consortium and the wider international communities for both sheep and goats.
The 3SR consortium worked closely with the complementary global efforts on-going through the ISGC and the International Goat Genome Consortium (IGGC). By doing so, 3SR has been able to make available to European researchers complementary resources provided by major research projects outside Europe, and has been involved in developing resources which will be able to enhance the livelihood of goat and sheep farmers worldwide;
• 3SR played a key part in the development of a well-annotated sheep reference genome sequence. This will provide the international sheep genomics community a valuable tool for underpinning further research
• 3SR partners were also central to development of a caprine 50K SNP chip which provides unparalleled opportunities to explore and understand nematode resistance in goats, to detect QTL for resistance, and to derive novel breeding strategies.
A key strength of 3SR has been that because commercial populations were used in the project, the results are directly relevant to the sheep and goat industry. Partners in the project have used populations of sheep drawn from a range of divergent breeds from throughout Europe to maximise the relevance of the polymorphisms found to the EU sheep and goat industries. The project consortium has strong links and a wide network with commercial breeders within the EU to ensure that effective dissemination of project outcomes is achieved.
Genetic variation in sustainability traits, such as mastitis susceptibility and nematode resistance, has been found. A putative causal mutation on chromosome 3 associated with increased susceptibility to mastitis was found in the Lacaune breed.
Both an ovine and a caprine custom-made DNA array of 500-1000 SNPs have been designed and developed on the basis of the results achieved in the search for genetic variation in mastitis susceptibility and nematode resistance. These have been used in commercial sheep and goat populations to validate the existence of predicted molecular markers for mastitis susceptibility and nematode resistance. Whilst the cross population analysis was inconclusive, the results may provide a valuable source of information for within breed breeding programmes.
Thus, the 3SR project has succeeded in providing selectable genetic markers that can be affordably applied in specific sheep and goat breeds to make contributions to improving animal health, welfare, sustainability and the long-term competitiveness of small ruminant production in the EU.
None of the SNPs for traits related to mastitis in Chios sheep was associated with milk production, implying that this information can be directly used to genetically control mastitis without compromising the productivity of the animals.
The caprine SNP50K chip, developed by joint efforts of the International Goat Genome Consortium and 3SR, may be further used by researchers and breeders to identify specific markers for other traits of interest. Knowing which animals are particularly susceptible or resistant to various health challenges will help breeders make decisions about how to breed animals in the future.
The consortium has furthermore developed and used cutting-edge genomic technologies such as an open-access genome browser, and produced large amounts of genome-related data (genotypes for 7,500 animals, genome sequences for 50 sheep, and over a terabyte of RNA-seq data). This has already led to the identification of causative mutations, candidate genes, SNPs and genetic markers for regions that affect the production of offspring in sheep, susceptibility to mastitis and nematode resistance in sheep and goats.
3SR has contributed to significant improvements in the genomics resources available for sheep and goats:
• Identified causative mutations affecting ovulation rate in Grivette (French) and Olkuska (Polish) sheep, and 2 candidate genes for sterility were identified in Cambridge sheep.
• In both sheep and goats, small regions of the genome were identified that have a significant effect on mastitis susceptibility and nematode resistance. Genetic markers for these regions were validated in commercial populations.
• The 3SR consortium helped the ISGC deliver sheep reference genome v3.1
• The 3SR consortium produced over a terabyte of RNA-seq data to assist with the annotation of the sheep genome
• A 50K SNP chip has been developed for goats. This enabled over 2000 animals to be genotyped for selectable genetic markers of interest within the 3SR project.
• Confirmed genetic variation to paratuberculosis in sheep (heritability 0.28±0.10)
• An open access Genome Browser for sheep and goats was developed http://genome.itb.cnr.it/gb2/gbrowse/oarv3.1/
• A whole-genome radiation hybrid (RH) map was produced for goats, containing more than 30,000 markers ordered across the chromosomes
• As well as the reference genomes, the genomic tools (for example genotyping arrays) have been greatly advanced throughout the duration of the project
Some of the results generated can be used immediately by breeding groups to increase selection accuracy, either through adopting a marker-assisted section approach or by prioritising polymorphisms for genomic selection. The exploitation plan will need to be breed specific, as optimal utilisation varies according to the breed structure and size. Numerically larger breeds will be in a better position to exploit genomic selection and hence make greater gains, whereas smaller breeds will realistically only be able to utilise individual markers and hence realise smaller benefits.
Other results provide a basis for better understanding the underpinning biology of the traits studied. The availability of such knowledge, combined with the identification of causative mutations, will play an important part in allowing the management and monitoring of future breeding programmes that will be needed to tackle future challenges, including these caused by climate change.
These tools and findings may assist breeders to produce animals with enhanced health and prolificacy, delivering significant benefits to productivity within the sheep and goat farming industries.
Besides the core beneficial impact on animal health, welfare and sustainability and the long-term competitiveness of small ruminant production in the EU delivered by the 3SR project, a wider and long-term impact may be expected from the project. The substantial new genomic resources generated (which are freely and publicly available) and the advanced basic understanding of sheep and goat genomes will guide future research efforts and boost our knowledge of these species. This helps accelerate and reduce the costs of future genetics/genomics research into traits of importance to animal health, welfare, sustainability and competitiveness – a substantial lasting impact of this timely project.
All in all, the project has been able to deliver a significant improvement in the available genomic information and technologies for goats and sheep, whilst also having a lasting impact on European research capacity.
The legacy of 3SR will hopefully be an improvement in the health, welfare and productivity of sheep and goat breeds in Europe and globally, underpinned by the better understanding of the genetics controlling their resilience to disease generally and their productivity in a farming context.
Dissemination Activities
In addition to an exceptionally strong research dimension, dissemination and knowledge management featured prominently in the project objectives, activities and outcomes, as summarised below.
The dissemination and knowledge management objectives of the 3SR project were:
• To create awareness and disseminate, manage and exploit knowledge generated in this Collaborative Project.
• To ensure transfer of knowledge from the project consortium to potential end-users and stakeholders, and promote exploitation of results.
• To ensure smooth communication and transfer of knowledge amongst participating partners.
• To link this project with other relevant research programmes and explore potential synergies in the dissemination and exploitation of results.
• To provide the platform for protection of Intellectual Property that may arise during the project.
The work towards meeting these objectives was organised around five tasks and eight deliverables involving all project partners drawn from research, industry and academia.
The key activities and outcomes are highlighted below:
1. A project website (http://www.3srbreeding.eu/) was set-up and was being continuously updated throughout this period as new information and material were made available. The website will continue being updated at least until August 2014.
2. A total of 12 peer-reviewed scientific articles were published in the international literature; a complete list with appropriate links can be found on the project website.
3. A total of 66 presentations featured in 33 national and international scientific conferences; these presentations are listed on the project website.
4. Four popular-press articles were written in non-scientific terms for farmers and the industry.
5. Two project Newsletters were prepared and distributed to more than 5,500 recipients, and also posted on the project website.
6. A Technical Bulleting summarising results for the industry was prepared. The scope of the bulletin was to increase awareness among the industry representatives and facilitate the uptake of research results at the farm, breed and population levels.
7. On seven occasions project partners exchanged visits to enhance intra-project collaboration and communication.
8. A total of 17 dissemination and industry briefing meetings took place in seven countries where project partners explained results and discussed implementation strategies with farmers, advisors, breed associations, breeding companies and industry representatives.
9. A sheep genomics seminar was organised in conjunction with the annual meeting of the European Federation of Animal Science (26-30 August 2013, Nantes, France). This event was linked to a similar event organised by another EC-funded research project (“Quantomics: From sequence to consequence – tools for the exploitation of livestock genomes”; Agreement no. 222664) thereby forging closer links between these two relevant projects and promoting complementary outcomes in the area of genomics for the improvement of livestock species.
10. Technology translation initiatives were developed including a user-friendly poster presenting the main concepts, goals and outcomes of the project, and a mastitis monitoring and improvement programme for small ruminant farms.
11. A plan for exploitation of the project results was completed based on a partners’ survey and industry feedback.
12. A Consortium Agreement was produced, agreed and signed by all partners at the start of the project. In addition, a Standard Operating Procedure (SOP) was developed outlining the process that should be followed by all partners wishing to publish results of 3SR. This SOP was followed throughout the project.
The above outcomes are expected to have a long-term impact on science as well as practical applications that will last well beyond the completion of the project.
List of Websites:
www.3srbreeding.eu
With 100 million sheep and 11 million goats in the EU and a combined global population of around 1.94 billion heads, these two species are of considerable economic, social and environmental importance.
Mastitis and nematode worm infections are the two biggest health and welfare threats to sheep and goats in Europe and beyond, causing huge economic and productivity losses. Previous research has indicated that susceptibility to both infections is at least partly under genetic control. Understanding the basis of genetic control to susceptibility will help direct breeding programmes to produce more resistant and, therefore, healthier and more productive animals. Ovulation rate is the key factor determining the number of lambs per litter for sheep. Some particularly prolific breeds have been identified along with some important genes that affect fertility but further work has been required to fully understand the genetic basis of control.
Co-funded by the EC’s FP7 programme, the Sustainable Solutions for Small Ruminants (3SR) project set out in 2010 to:
1. Develop and verify (in commercial populations) selectable genetic markers for traits with high importance for health and sustainability in sheep and goats, and to identify the causative mutation(s) where possible, for a) Mastitis susceptibility; b) Resistance to nematodes, and c) Ovulation rate.
2. Develop tools and resources that are needed to achieve Objective 1 and provide a lasting resource for future small ruminant genomics research, specifically: a) Improved assembly and comparative and functional annotation of the sheep genome, b) Tailored SNP sets suitable for fine-mapping of polymorphisms relevant to sustainability traits in sheep and goats, and c) Improved resources for goat genomics, especially a higher resolution genome map
3. Disseminate results from the project to ensure a broad and long-term beneficial impact on European competitiveness and EU policy on animal health, welfare and sustainable agriculture.
The 3SR consortium worked closely with the complementary global efforts on-going through the International Sheep Genomics Consortium (ISGC) and the International Goat Genome Consortium (IGGC), and played a key part in the development of a well-annotated sheep reference genome sequence (OARv3.1) that will provide the international sheep genomics community a valuable tool for underpinning further research and a caprine 50K SNP chip which provides unparalleled opportunities to explore and understand nematode resistance in goats, to detect QTL for resistance, and to derive novel breeding strategies.
The consortium has furthermore developed and used cutting-edge genomic technologies such as an open-access genome browser, and produced large amounts of genome-related data (genotypes for 7,500 animals, genome sequences for 50 sheep, and over a terabyte of RNA-seq data). This has already led to the identification of causative mutations, candidate genes, SNPs and genetic markers for regions that affect the production of offspring in sheep, susceptibility to mastitis and nematode resistance in sheep and goats.
These tools and findings may assist breeders to produce animals with enhanced health and prolificacy, delivering significant benefits to productivity within the sheep and goat farming industries.
Project Context and Objectives:
Importance of the sheep and goat industries
Small ruminants (namely sheep, Ovis aries, and goats, Capra hircus) were the first livestock species to be domesticated around 10,000 years ago in the margin of the Fertile Crescent. They have since diffused all over the world following migratory and exploratory movements of humans and thereafter routes of commercial trade. Now, small ruminants are a major component of farm-animal production all over the world, with about 0.9 billion goats and 1.1 billion sheep.
The sheep and goat sectors play an important role in the economy of many EU countries, often underpinning the economic viability of rural communities. Their ability to make good use of low quality pasture as well as being ideally suited to being kept on small family farms makes them extremely valuable. Due to their small size and grazing preferences, and through often being highly adapted to local conditions, they also play an essential role in landscape maintenance and nature conservation.
Challenges being faced
The future sustainability of the sheep and goat sectors relies on breeding animals that are resilient against disease challenges, as well as being productive and efficient. Understanding the mechanisms of resistance and developing genomic tools and technologies is likely to be crucial in helping to address these challenges.
Mastitis, paratuberculosis and parasitic infection from nematodes are currently ubiquitous chronic diseases within large parts of the European sheep flock and goat herd. Many cases go largely undiagnosed, due to the extensive nature of many farming systems and the subclinical nature of the diseases, however their significance should not be underestimated.
Mastitis
The impact of mastitis is significant in both the dairy and meat industries. For dairy sheep it is estimated that up to 55% of ewes have sub-clinical mastitis while less than 5% have clinical mastitis. Losses in milk production have been reported to range from 20-37% in affected ewes. Making conservative assumptions of a 10% incidence of mastitis in the EU sheep flock as a whole, with a mean reduction in milk yield of 20% and using €650/tonne as the wholesale price of sheep milk, it can be conservatively estimated that the cost to the EU is €37million per annum.
Mastitis is the single largest reason for the culling of ewes (up to 50%) and results in increased veterinary expenses as well as impacting on the welfare of the animals. For meat sheep, lamb growth from affected ewes is less than lambs from healthy ewes. One study demonstrated that increasing mastitis resistance by 10% in the Texel breed in the UK would save £2.7million per annum. There is less data available for goats but it is expected that reducing mastitis susceptibility in this species could also lead to significant savings.
Nematode infection
On a global scale, ruminant diseases caused by gastrointestinal nematode parasite infections are the diseases with the greatest impact upon animal health and sustainability. Within Europe, nematode infections are the most costly endemic disease in meat production systems, and are close behind mastitis in milk production systems.
Welfare and cost implications of parasitism in sheep and goats are significant with even sub-clinical infections causing marked decreases in productivity; for example, the annual cost of nematode challenge in the sheep industry is estimated at £84million in the British sheep industry and $500m in Australia. Cost savings have been shown to be proportional to the severity of nematode infection therefore a 10% improvement in resistance, in this case, would reduce annual costs by £8.4million.
Ovulation rate
Ovulation rate is an indicator of prolificacy. For ovulation in sheep, increasing our knowledge of the ovulation rate genes that have been identified so far could lead to improvements in flock productivity and profit. Better knowledge of these genes will lead to better control and more informed decision making by the sheep industry.
Climate change
In the context of climate change, characterising the molecular mechanisms underlying the adaptation of small ruminants to harsh and different environments is of particular interest, both from a theoretical point of view and for identifying new valuable genetic resources for regions that will be subject to climate change. Characterising the genetic diversity of small ruminants requires the availability of genomic tools such as complete genome sequences.
Sheep and goat farming will also help address the EU core objective of sustainable development , as defined by the Brundtland Report (1987): “Development that meets the needs of the present without compromising the ability of future generations to meet their own needs.” Better use of selective breeding of livestock is arguably the single best way to improve the sustainability of animal agriculture, with the added benefit that improvements made are permanent and cumulative. Moreover, addressing these issues through selective breeding will contribute to the EU strategy for sustainable development launched in 2001 and the EU Sixth Environment Action Programme (6EAP) , which have set the overall objective of decoupling economic growth from environmental degradation.
Potential benefits from genetic and genomics
For a variety of reasons, most notably the fragmented nature of the goat and sheep industries globally and their lower economic worth in the United States, sheep and goats have lagged behind other major livestock species in the development of comprehensive genomic information that can inform selective breeding.
It is only recently (and aided by the work of the 3SR consortium) that high quality, draft reference genome sequences for sheep and goats have been published. With 100 million sheep and 11 million goats in the EU and a combined global population of around 1.94 billion heads, these two species are of considerable economic, social and environmental importance. This is particularly the case in marginal areas, and the benefits that can be delivered by better understanding the genetic basis of health and welfare traits are highly significant.
To remain sustainable, sheep and goat producers will need to continue to address the challenges of disease and improved efficiency. Genomic tools and technologies are likely to play a crucial role in helping to address these challenges.
Mastitis has a profound, negative effect on sheep and goat health and welfare, causing udder pain and swelling, lethargy, anorexia and even death. Current mastitis management is based on prevention, through good hygiene, nutrition and treatment, using intravenous antibiotics. None of these is entirely successful and the use of antibiotics raises concerns about the sustainable management of the condition. Increasing genetic resistance to mastitis offers a truly sustainable and complementary adjunct to existing strategies.
For the effective control of nematode parasites, farmers have come to rely almost exclusively on broad-spectrum anthelmintics. Issues relating to resistance, residues and eco-toxicity are increasingly challenging the sustainability of chemotherapy. Anthelmintic resistance is widespread and escalating, threatening to erode the efficacy of these drugs. All too often, a structured regime of anthelmintic treatment is not followed, resulting in ineffective control and inadvertent increased selection for resistance. There are also environmental concerns associated with anthelmintic use. The Ivermectin class of drugs, currently the most effective anthelmintics, are excreted at high concentrations in the faeces of treated animals and have a potent effect on the insects that feed on these faeces . Clearly the current situation is unsustainable - genetic improvement is a sustainable, cost effective alternative strategy that has yet to be fully exploited but which 3SR has aimed to help to redress.
Paratuberculosis is a debilitating disease of both sheep and goats. Currently there is no effective treatment for its control, and breeding for resistance to paratuberculosis may be the only sustainable method for its control.
Sheep account for 9% of EU enteric methane emissions from livestock . Increasing the efficiency of sheep and goat farming therefore reduces the amount of methane emitted for a given level of production, improving the carbon footprint of the sector.
An international challenge
As small ruminant genomics research is a relatively small field, the leading experts in these different areas are dispersed throughout Europe but also further afield in the USA, Australia, New Zealand, China and Argentina.
At the same time, genomics is big science requiring expensive large-scale activities, such as sequencing, assembly, curation and annotation of a genome sequence for each species. The costs and work for the development of these resources therefore need to be shared across as many users as possible and duplication of effort needs to be strenuously avoided.
A project to tackle the genomics of sheep and goats, thus, required an international and inter-disciplinary framework to effectively integrate the varying expertise, and to ensure the maximum progress.
The international sheep and goat research communities are collaborative rather than competitive. It is therefore not surprising that the leading European expertise in small ruminant genomics and key players in the International Sheep Genomics Consortium and the International Goat Genome Consortium collaborated in the 3SR project. This has created a unique collaboration to ensure that the EU is well positioned to take advantage of the genomics revolution in breeding.
The major EU countries within which sheep and goat production occurs in Marginal Areas, and where capabilities for genomics research exist, are France, Spain, Italy, Greece, UK and Ireland, and each of these countries is represented in the 3SR Consortium by at least one organisation, enabling the greatest access to tools, expertise and relevant resources, and short lines to dissemination and implementation of results in commercial populations.
The involvement of partners from outside the EU also aids the development of European expertise in small ruminant genomics, ensuring that the EU is well positioned to take advantage of the genomics revolution in breeding. This allowed the project to be much more efficient than it could be at a national level; it avoided duplication of effort and has produced a common approach with agreed protocols.
Aims and objectives
The aim of the 3SR project was to discover selectable genetic markers (single nucleotide polymorphisms (SNPs)) and causative mutations – where possible in commercial populations – for traits that are critical to sustainable farming; namely mastitis susceptibility, nematode resistance and ovulation rate. Once identified, the consortium sought to verify the markers in additional commercial populations.
To achieve this, 3SR had set itself three high-level objectives:
1. To develop and verify (in commercial populations) selectable genetic markers for traits with high importance for health and sustainability in sheep and goats, and to identify the causative mutation(s) where possible, for:
a. Mastitis susceptibility
b. Resistance to nematodes
c. Ovulation rate (sheep only)
2. To develop tools and resources that were needed to achieve Objective 1 and provide a lasting resource for future small ruminant genomics research, specifically:
a. Improved assembly and comparative and functional annotation of the sheep genome
b. Tailored SNP sets suitable for fine-mapping of polymorphisms relevant to sustainability traits in sheep and goats
c. Improved resources for goat genomics, especially a higher resolution genome map
3. To disseminate results from the project to ensure a broad and long-term beneficial impact on European competitiveness and EU policy on animal health, welfare and sustainable agriculture.
In order to deliver on these aims and objectives, the 3SR consortium set out to apply the latest high throughput genomics technologies, comparative and functional genomics, combined with targeted genome sequencing and extensive in silico analyses to dissect important genetic components controlling these traits. In order to deliver significant improvements in available genomic information and technologies for these species and have a lasting impact on European research capacity.
To make this feasible, the consortium worked closely with the complementary global efforts on-going through the ISGC and the IGGC, and made use of complementary resources provided by major research projects in Europe, Australasia, USA, Argentina and China.
The 3SR project has brought together an international consortium, to deliver a step-change in the understanding of the genetic basis of traits underlying sustainable production and health.
Project Results:
Project overview
The overall aim as part of the 3SR project was to develop molecular markers that can be used to select for small ruminants that are healthier and more productive, and can make a higher contribution to sustainable production in the EU.
In order for this to be achieved, the 3SR project:
i. Included the full breadth of complementary research needed to develop useful markers, consisting of “discovery” through to “validation” in commercial populations.
ii. Focused primarily on markers for susceptibility to mastitis, resistance to nematodes and ovulation rate, but also included analysis of data for production traits and susceptibility to paratuberculosis.
iii. Made best possible use of the wealth of resources available through the 3SR Consortium members to ensure the best value for the requested funding.
iv. Maintained very close links with the ISGC and IGGC to ensure full advantage of the most up-to-date resources available, and also to ensure that the resources developed by 3SR contribute positively to the international effort to develop genomic resources for sheep and goats.
The project activity was divided into seven interlinked work packages, as shown in the project overview (Figure 1 in Appendix 1).
Work Package (WP) 1 focused on developing the genomic resources and tools needed for the project as a whole. The tools, in terms of genome-wide or region-specific SNP arrays, sequences for specific genome regions or radiation hybrid maps, were produced in close collaboration with the trait-specific work packages. WP1 also ensured that state-of-the-art genomic tools and techniques were used throughout the project.
Three trait-specific WPs subsequently utilised the various outputs of WP1, focusing on mastitis susceptibility (WP2), nematode resistance (WP3) and ovulation rate (WP4), while also making full use of additional data recorded at no additional cost to the project, including standard performance traits and paratuberculosis resistance.
The SNP genotyping and functional studies performed in these three WP2, WP3 and WP4 were used by WP5, where the genomics data were fully mined and candidate genes nominated for inclusion on validation SNP arrays developed in WP1.
Thus the research iterated between these five WPs, with greater precision of results arising from each iteration, resulting in:
1. Selectable markers for use in sustainable breeding programmes;
2. A significant contribution to the activities of the ISGC and IGGC.
As research outputs are of little value without effective communication and dissemination, WP6 was trusted with the task to ensure that knowledge gained is optimally disseminated, translated and exploited, with close contact maintained with all potential end users.
Lastly, the task of effective management at all levels of a large and complex project as 3SR was placed within WP7. In addition, WP7 ensured that quality standards were maintained; and that reporting and auditing requirements were met. IPR arising from 3SR was managed jointly by WP6 and WP7. Careful ethical monitoring throughout the project ensured that any (potential) public concerns were met and that 3SR aligned with the “3Rs” (Replacement, Refinement and Reduction) of animals in research.
Key results
3SR’s close collaboration with the ISGC and IGGC has led to significant improvements in the genomics resources available for sheep and goats. The 3SR project has contributed to improving the quality of the reference sheep genome by sequencing animals, it also contributed to the annotation of the sheep genome by performing large scale RNA sequencing.
Activity highlights:
• Sequencing of around 50 animals has been carried out to help improve the quality of the reference sheep genome (coordinated by the ISGC), while large scale RNA sequencing has been performed to help with the sheep genome annotation.
• Helped the IGGC to develop a high density (50K) SNP chip for goats.
• Genotyped over 4000 sheep and 2000 goats using the respective SNP50K chips.
• Over 7000 50K genotypes analysed from 11 sheep populations and over 2000 50K genotypes analysed for goats
• Controlled challenge studies were carried out for mastitis (in sheep and goats) and nematodes (sheep) to explore the mechanisms of resistance
• Functional studies of genes involved in ovulation rate in several sheep breeds were carried out to support the biological understanding of the mutations discovered.
Key results:
• Identified causative mutations affecting ovulation rate in Grivette (French) and Olkuska (Polish) sheep, and 2 candidate genes for sterility were identified in Cambridge sheep.
• In both sheep and goats, small regions of the genome were identified that have a significant effect on mastitis susceptibility and nematode resistance. Genetic markers for these regions were validated in commercial populations.
• The 3SR consortium helped the ISGC deliver sheep reference genome v3.1
• The 3SR consortium produced over a terabyte of RNA-seq data to assist with the annotation of the sheep genome
• A 50K SNP chip has been developed for goats. This enabled over 2000 animals to be genotyped for selectable genetic markers of interest within the 3SR project.
• Confirmed genetic variation to paratuberculosis in sheep (heritability 0.28±0.10)
• An open access Genome Browser for sheep and goats was developed http://genome.itb.cnr.it/gb2/gbrowse/oarv3.1/
• A whole-genome radiation hybrid (RH) map was produced for goats, containing more than 30,000 markers ordered across the chromosomes
• Led the establishment of the goat AdaptMap project which aims to coordinate and collate the research data across a number of projects (including 3SR)
• As well as the reference genomes, the genomic tools (for example genotyping arrays) have been greatly advanced throughout the duration of the project
• Around 4000 sheep from 11 populations with relevant phenotypic data have been genotyped, using the 50K ovine SNP chip
• A further 3400 genotypes from the same populations and funded by other national projects, have been made available for analysis in 3SR.
• Technical bulletin providing an easy to understand overview of the key activities and results in 3SR
State of the art
When the 3SR project was first proposed in 2009, the genomic resources for sheep and goats were limited and falling behind those of the other major livestock species. An International Sheep Genomics Consortium (ISGC) had previously been formed, and by early 2009 the IGCS’s work had led to the availability of improved sheep maps, a sheep Bacterial Artificial Chromosome (BAC) library and associated end sequences, the first Ovine 50K SNP Beadchip (“SNP50K chip”) and a first draft of the sheep genome assembly. This first draft was estimated to cover around 80% of the unique sequence in the sheep genome, indicating that there were still significant gaps in the sequence, and therefore quite a bit of work needed to improve the quality of the reference genome for sheep.
The situation for sheep was better than that for goats. In 2009, no concerted international research effort to develop caprine genomic resources had been established, and consequently no reference genome sequence was available for the goat, although there was a low density (1536) SNP chip.
A number of external events occurred in 2010 and throughout the duration of the 3SR project that provided a valuable opportunity to increase the potential impact and the value for money that could be achieved through the 3SR project. These opportunities resulted mostly through the close links between the partners in the 3SR consortium and the wider international communities for both sheep and goats.
The main changes and opportunities that were harnessed were as follows:
1. In March 2010 the Beijing Genomic Institute (BGI) announced that they had sequenced the entire genome of a female Texel sheep. During the remainder of 2010 BGI worked with the ISGC and 3SR partners to use the information generated to improve the quality of the ovine reference assembly.
2. In the summer of 2010 AgResearch in New Zealand coordinated an international order for the ovine SNP50K chip. This provided an opportunity for 3SR partners to purchase the chips at a substantially lower price than usual. By delaying the SNP chip order in 3SR by a few months we were able to take advantage of this opportunity and dramatically increase the number of SNP50K chips that could be purchased through the project, rather than developing targeted, low density 1536 chips. Participation on the international order also provided an additional benefit. Some of the consortium partners were able to change their genotyping strategy to use SNP50K chips in other projects that were funded through national sources, and these genotypes were also made available for inclusion in the analyses for 3SR.
3. An International Goat Genome Consortium (IGGC) was formed in March 2010. A number of the 3SR partners are part of this consortium. Coordinated international activity was initiated in three areas: de novo goat genome assembly and re-sequencing; development of a goat Radiation Hybrid (RH) panel and mapping; and production of a caprine SNP50K chip. 3SR was closely involved in the delivery of these activities and was able to use the caprine SNP50K chip for the genotyping, delivering greater value compared to the original plan.
4. These international developments also allowed more 3SR staff resources to be focused on developing tools and resources for goats, which further increased the impact that could be achieved by the 3SR project.
Tools and resources
To achieve the aims and objectives, 3SR project partners have applied the latest high-throughput genomics technologies, comparative and functional genomics, combined with targeted genome sequencing and extensive in silico analyses to dissect important genetic components controlling their traits of interest (and where possible the causal mutations).
By working closely with the ISGC and the IGGC, the project has been able to deliver a significant improvement in the available genomic information and technologies for these species, whilst also having a lasting impact on European research capacity.
In addition, 3SR has benefited from the use of complementary resources provided by major research projects in Europe, Australasia, USA, Argentina and China through key partners within the consortium that are involved in these projects.
The tools and resources that have been developed include an improved assembly, comparative and functional annotation of the sheep genome, and improved resources for goat genomics.
An overview of the most significant resources and tools developed and used are provided below.
Sheep and goat research populations
The project made use of data from over 10,000 sheep from 15 diverse breeds in 10 countries and over 4,000 goats from 3 breeds in 3 countries.
One of the strengths of the 3SR project is the large number and variety of breeds that are available for use through the partners. These include over 10,000 sheep and 4,000 goats, all of which have been individually recorded. Many of these animals have been genotyped as part of the project, primarily with the relevant SNP50K chip. In addition, specific 1K SNP chips were developed in the project for the purposes of validation. Details are provided in Figure 2 (in Appendix 1).
The sheep and goat populations used in the 3SR project are involved in research applied to one of the three sustainability traits considered. There is one exception amongst the sheep: a Sarda-Lacaune backcross research flock from Sardinia has been sufficiently intensively recorded to allow the data to make valuable contributions to work on all three traits.
Sarda-Lacaune Backcross sheep population
This experimental population was established in 1999 by INRA and AGRIS for the purpose of aiding the detection of QTL associated with milk production and functional traits including somatic cell count (SCC) and faecal egg count (FEC). The population was formed by mating animals from the two most numerous dairy breeds from the respective countries, the Lacaune (France) and the Sarda breed which is predominantly found in Sardinia (Italy). Although two dairy breeds, they differ phenotypically for many traits, such as body size, growth rate, wool, prolificacy, milk yield and milkability.
The mating of 14 elite Lacaune rams to 184 Sarda ewes formed the first generation of the cross. Ten sons from each of ten founder sires were then mated to 2,719 ewes to generate 968 backcross females, which included approximately 98 from each of ten families. A further four structured matings were subsequently conducted and all retained ewes were fully recorded. DNA and records were also collected on more recent descendants from the initial backcross animals.
Tools for mining genomic information
Activities of the 3SR partners have resulted in large amounts of genome-related information. In order to facilitate collaboration on sheep and goat genome studies, a web platform was created with a repository to collect results from the sheep genome annotation with tools for customisable data extraction, visualisation and analysis. This has been updated and maintained consistent with the state-of-the-art of the 3SR project, and has been designed such that new functionalities may be added to facilitate results integration and analysis.
The Sheep and Goat Genome Browser
The genome browser allows users to visually inspect genomic annotations. Small amounts of custom annotation can be uploaded, thus adding a custom track on the genome viewer. Annotation can either be inserted manually or by file upload, and visualisation configuration can be customised.
It relies on the generic genome browser database, GBrowse (http://gmod.org/wiki/GBrowse) containing all the data produced by the 3SR members and used as a viewer for the sheep genome, with new tracks added as soon as they become available.
Available tracks include mapped bovine RefSeq genes, sheep gene predictions and eight tracks of SNPs from different submitters. The GBrowse database is, and will continue to be, the repository for all the sheep-related information that will be made available during the project. Browsing via the GBrowse interface allows extraction of data related to genomic regions of interest and downloading of partial or complete datasets. Data can also be downloaded as gff3 files or exported as images (suitable for publication).
QTL regions are visualized in GBrowse and QTL-driven searches can be performed via the GoSh interface to retrieve genes and other features included in the QTL of interest. Every additional feature (i.e. promoters, alternative splicing sites, etc.) can be displayed in GBrowse tracks, and dedicated web pages, based on the structure of the GoSh interface, can be added, containing full information for the selected feature.
The 3SR GBrowse includes GLEAN and GeneWise gene predictions, mapped bovine RefSeqs, mapped ESTs and EST clusters, SNPs, tRNA predictions, tandem repeats and mapped TFBS and canonical promoter elements. MiRNA hairpins and mature sequences positioned by BLAST are also available.
To incorporate the new reference sequence data available for goat, a genome synteny browser was developed, based on the GMOD GBrowse syn structure (http://www.gmod.org/wiki/GBrowse_syn). GBrowse syn can be used to view multiple sequence alignment data, synteny or co-linearity data from other sources against genome annotations provided by GBrowse. The sheep and goat synteny browser includes the sheep genome, version 3.1 (Oar-v3.1) and the goat genome, version 1.0 (http://goat.kiz.ac.cn/GGD/).
The Genome Browser (open access):
http://genome.itb.cnr.it/gb2/gbrowse/oarv3.1/ http://www.livestockgenomics.csiro.au/sheep/
http://genome.itb.cnr.it/gb2/gbrowse_syn/sheep3.1-goat/
Genome Mining Tool (GMT)
The Genome Mining Tool (GMT, http://155.253.6.237/3SR_gmt) is a web interface dedicated to 3SR partners that focuses on defined genomic regions. It includes pre-defined regions relevant to the study of mastitis and nematode resistance, and also allows user-defined regions to be selected. Registered users can add, delete or modify regions. Once a region is selected, a list of the RefSeq gene models and SNPs is given, containing links to related resources. Every region or feature is also linked to the GBrowse (see above).
The GMT also hosts a page for a pathway-oriented view of the trait of interest. In this page, tentative QTLs for mastitis or nematode resistance are analysed as a whole, and a list of pathways that involve genes from those regions is given, together with the number of genes for each pathway. This number is linked to the full list of genes.
The bioinformatic structure of the GMT is based on a MySQL database. Each genomic feature can be inserted in this database, provided it can be defined by an absolute position on the sheep genome. Custom scripts allow data conversion from the GBrowse or Pathway Commons formats to the input formats suitable for upload in the GMT database. Selection of features in regions is performed dynamically via the web interface.
Genome Mining Tool (GMT) (restricted access)
(http://155.253.6.237/3SR_gmt)
The 3SR GBrowse and GMT web interfaces are dedicated to the collection and display of data produced within the activities of the 3SR Project. Both interfaces can contain data from any genomic feature, provided an absolute position on sheep reference genome (version 3.1) is given. The retrieval system offered by the GMT offers the possibility to focus on pre-defined genomic regions, including all tentative QTLs for mastitis and nematode resistance, or on user-defined test intervals, and to obtain pathway-oriented views of the traits under investigation.
The following describes some of the features of the tools developed.
• GMT users can explore and visualise all the records included in putative QTL regions.
• Records can be modified by authorised users or displayed by project partners.
• All regions are linked to the genomic browser: by clicking on the “Open Gbrowse” link, the selected region will be displayed in GBrowse in a separate web page.
• Users can retrieve all the genes and all the SNPs within the selected region by clicking on the “Genes” and “SNPs” links.
• Featured annotations are also available by following links.
• Data can be downloaded in an Excel- (or other spreadsheet) compliant format.
Sequence data for sheep
The proposed sequencing strategy in 3SR was updated throughout the project as the sheep reference genome was improved so that the quality of the new assembly in the genomic regions of interest to 3SR could be assessed. As the quality of assembly in those regions was found to be good, the sequencing strategy was amended to increase the benefit that could be achieved through use of the project resources.
In order to leverage the activities and resources available outside the project, activities in 3SR were concentrated around the following areas:
• Genotyping using SNP50K chip
• Improving the genome assembly using available sequencing information
• Re-sequencing of whole genomes for individual animals
• BAC sequencing
• CNV detection
• Sequencing of RNA form a wide range of tissue to help with annotating the reference genome sequence
• Construction of high density RH maps of the sheep and goat genomes
• Generation of validation chip.
This has provided a big impact in terms of information to help improve the quality of the reference assembly for sheep but also to understand what the sequence means.
Details on a number of these activities are provided in the following paragraphs.
Sheep genome v3.1
Version 3 of the sheep genome (“Oar v3.1”) was built by the ISGC in close collaboration with 3SR and the USDA, with a Post Doc at CSIRO, supported in part by 3SR, undertaking most of the work in building Oar v3.1. The sequencing was undertaken by 3SR partner The Roslin Institute (UK), Beijing Genomics Institute (China) and Baylor College of Medicine (USA).
Oar v3.1 was built using next generation sequence derived from one female and one male Texel. The primary de novo assembly was performed using ~75 fold Illumina GA sequence from the female with several rounds of filling gap steps undertaken using both the female and the male Texel.
The major improvements are the large reduction in gaps, including in GC rich regions around the start sites of genes. Many of the remaining problematic areas are in regions orthologous to regions of the bovine genome with copy number variation, and cannot be solved using currently available resources.
The work on Oar v3.1 has been published in the journal Science:
Yu Jiang, Min Xie, et al., and Brian P. Dalrymple. (2014). The sheep genome illuminates biology of the rumen and lipid metabolism. Science 344, 1168-1173
DOI: 10.1126/science.1252806
http://www.sciencemag.org/content/344/6188/1168.abstract?sid=93cc6e14-c2c5-428d-b9a8-72957a5d1d61
Improvements continue to be carried out beyond the end of the 3SR project, with the first priorities for improvements to reduce the number of large gaps and provide greater coverage of segmental duplications and copy number variation regions.
Oar v3.1 can be downloaded from: www.livestockgenomics.csiro.au/sheep/oar3.1.php and www.ncbi.nlm.nih.gov/bioproject/PRJNA169880.
Genotyping using ovine SNP50K chip
Largely as the result of the participation in a large international order, the number of SNP50K genotypes that were available for analysis as part of the 3SR project increased by eight-fold compared to the original plan, increasing from 900 to nearly 7,500 across 12 populations representing 11 different breeds/crosses. More detail on the numbers of each population of interest is shown in Figure 3 (in Appendix 1).
Re-sequencing of whole genomes for individual animals
Sequencing of key animals has been carried out to help improve the quality of the reference sheep genome (coordinated by the ISGC), while large-scale RNA sequencing has been performed to help with the sheep genome annotation (Figure 4 in Appendix 1).
Construction of high density radiation hybrid (RH) maps for sheep and goats
Construction of a high-density RH map of the sheep genome has been completed to assist with assembly of the goat genome. The whole-genome RH map produced for goats contains more than 30,000 markers ordered across the chromosomes.
INRA and USU constructed a whole-genome high-density RH map of the ovine genome using two ovine RH panels that were developed at both institutes. A specific genotype calling algorithm, based on raw fluorescence intensities provided by the Illumina platform, was developed because the genomes of RH clones differ in many ways from the genomic DNA of diploid individuals.
From the 49,035 SNPs of the ovine SNP50K BeadChip, 41,999 could be reliably called on both RH panels using the specific new calling algorithm. RH maps were constructed for each chromosome using the comparative approach developed at INRA and were constructed using the sheep assembly as the reference. At the completion of the analysis, RH maps for all the sheep autosomes were developed, containing a total of 31,162 markers. This high-density RH map provides an orientation of these ovine SNPs along the goat genome, thereby linking the genome assemblies of goat and sheep. This information will be useful as detected QTL regions in the goat are comparatively mapped in the sheep.
Chinese partner HZAU constructed a 5,000 rad caprine radiation hybrid (RH) panel and genotyped the goat panel with approximately 108,000 SNP markers on the Illumina bovine SNP50K and ovine SNP50K chips. A robust RH map was constructed in collaboration with INRA, containing 30,988 SNP markers covering all 30 caprine autosomes and the X chromosome at an average spacing of approximately 85 Kilobases (Kb). This high-density RH map has been used to verify the assembly of the de novo goat genome sequence.
The marker order in the caprine RH map is 96.1% consistent with the sheep genome and 90.9% with the cattle genome. A comprehensive caprine–mammalian comparative map has been generated by identifying the locations of caprine RH-mapped loci within the genomes of sheep, cattle, human and other mammalian species.
Development and employment of caprine 50K SNP chip
A notable success of 3SR has been the development of the caprine 50K SNP chip (Tosser-Klopp et al., 2014), which provides unparalleled opportunities to explore and understand nematode resistance in goats, to detect QTL for resistance, and to derive novel breeding strategies.
The release of the goat genome sequence in 2010 by the Beijing Genomics Institute paved the way for an international effort towards constructing projects and tools for goat genomics. In the frame of the International Goat Genome Consortium (IGGC), and with the support of 3SR and the French dairy goat industry, a caprine 50K SNP chip was released by Illumina at the end of 2011.
The success of Genome Wide Association Studies (GWAS) in the discovery of sequence variation responsible for, or closely linked to, complex traits in humans has increased interest in high throughput SNP genotyping assays in livestock species. The primary goals with this technology are QTL detection and genomic selection.
The success of a moderate density SNP assay depends on the reliability of bioinformatic SNP detection procedures, the technological success rate of the SNP design, even spacing of SNPs on the genome and selection of Minor Allele Frequencies (MAF) suitable to use in diverse breeds.
Data collected within the 3SR project and other international resources were combined within the IGGC (www.goatgenome.org) to design of a 50-60K SNP chip for goats.
Approximately 12 millions high quality SNP variants were identified within and between 6 different breeds (meat, milk and dual purpose: Alpine, Boer, Creole, Katjang, Saanen and Savanna), using a total of 97 animals.
Whole genome and Reduced Representation Library sequences were aligned on >10kb scaffolds of the de novo goat genome assembly, and then evenly spaced on the goat genome.
60,000 SNPs were selected and submitted to Illumina for oligo manufacturing. The final manufactured bead-pool consists of 53,347 probes that passed manufacturing quality control. Ten different breeds were then used to validate the SNP content and 52,295 loci could be successfully genotyped and used to generate a cluster file.
The caprine SNP 50K chips were successfully used for QTL detection in 3SR for mastitis susceptibility and nematode resistance.
The 60,000 SNPs were also submitted to the dbSNP database (http://www.ncbi.nlm.nih.gov/projects/SNP) as a resource for international academic research, along with flanking sequences and map position on goat assemblies (i.e. scaffolds and pseudo-chromosomes), sheep genome V2 and cattle UMD3 assembly.
Inclusion of the goat genome in a comparative genomic database
An efficient exploitation of complete genomic sequences requires the availability of dedicated bioinformatic tools. One particularly powerful approach to analyse genomic sequences is the comparative genomic approach, which takes advantage of the existing whole genome sequences of evolutionary-related species.
In order to enable a straightforward exploitation of the newly sequenced goat genome in a comparative genomic perspective, 3SR partner Utah State University (USU) compared the goat genome to 15 mammalian completely sequenced genomes (Human, cattle, sheep, pig, horse, dog, cat, mouse, rat, rabbit, chimpanzee, gorilla, orang-utan, marmoset and platypus). The results are available through the Narcisse comparative genome browser (http://narcisse.toulouse.inra.fr) providing a view of the difference in the evolution of the various species.
Traits: Mastitis susceptibility
Introduction
Amongst infectious diseases in dairy ruminants, mastitis is of major importance because of its high frequency and related costs. In small dairy ruminants, mastitis is the primary health reason for involuntary culling, and prevalence of subclinical mastitis is 20-30% per lactation.
Accumulating research results over the last decades give strong evidence that the host’s response to control udder health is under genetic control in dairy ruminants. Existence of QTL for mastitis resistance in sheep and cattle had been previously identified. Among studies, the common mastitis resistance phenotype was milk somatic cell count (SCC), measured periodically over lactation.
Mastitis susceptibility (sheep)
Project partners used the ovine SNP50K chip and subsequently performed Genome Wide QTL analyses to identify regions of significant genetic variation (“QTL”) in mastitis susceptibility in four sheep populations. Twenty-nine QTL
s associated with SCC were found in the Churra (7 regions), Lacaune (7) and Sardinian-Lacaune Backcross populations (15). QTLs had rather moderate effects and were mostly population specific, in agreement with the polygenic nature of the trait and the complexity of the resistance phenotype. However, a few commonalties among populations were found, for example on chromosomes 3, 14 and 20. One
on chromosome 5 was also associated with mastitis resistance (intra mammary abscesses) in the divergent lines of Lacaune sheep.
One of the major findings was a highly significant QTL located on a narrow region (<0.5 Mb) of chromosome 3 in the Lacaune population. This QTL was detected at the same location by all analysis methods that were performed. Subsequently, whole genome sequencing of a trio of animals led to the identification of a putative causal mutation on chromosome 3 associated with increased susceptibility to mastitis in the Lacaune breed. A functional assay for evaluating the protein binding affinity, suggests a functional knockout of the socs2 gene in homozygous susceptible sheep.
Mastitis validation studies in sheep
Results obtained in the project, and described above were used to develop and verify (in commercial populations) selectable genetic markers for mastitis susceptibility. Thereto a custom-made ovine SNP chip of 960 SNPs was designed and developed. This array was used to genotype four independent sheep populations: Chios, Churra, Manech Red Faced and Lacaune. This highlighted a number of regions and SNPs significantly associated with mastitis, of which the most significant results are described below.
Chios sheep
In Chios sheep, a total of 18 SNPs were found to affect one of the mastitis related traits (either clinical mastitis or indicator traits such as milk SCC, total bacterial count in milk and CMT). Importantly, none of these SNPs was associated with milk production, implying that this information can be directly used to genetically control mastitis without compromising the productivity of the animals.
Lacaune sheep
Three SNPs, located in the narrow region of a QTL found on chromosome 3, were found to be significantly associated with the LSCS mastitis trait in the Lacaune population. The two most significant ones were located in the SOCS2 gene (R96C non synonymous) and CRADD genes (intron). This reinforces the hypothesis that the SNP in the SOCS2 gene is likely to be causative, although a direct effect of other polymorphisms in this narrow region still cannot be excluded.
Mastitis susceptibility (goats)
Genotyping of the Saanen and Alpine breed populations with the newly released caprine SNP50K chip resulted in the identification of significant QTL for traits related to mastitis resistance.
At the start of the 3SR project, there was no information available on chromosomal regions controlling mastitis resistance in goats, nor was there a commercial SNP tool for goats.
The previously described release of a caprine SNP50K chip enabled the project to undertake the first genome–wide association studies (GWAS) to map loci describing significant genetic variation in mastitis susceptibility based on the SCC trait. The number of loci are shown in Figure 5 (in Appendix 1). Results are presented for significant QTL found in LA and LD, with many more significant QTLs found with the LD method compared to the LA method, and a higher number of QTLs found in the Saanen breed than in the Alpine breed.
3SR partners were able to genotype an additional population of 192 Alpine goats from a divergent selection experiment with the caprine SNP50K chip in order to further confirm and fine map those QTLs.
Among the numerous regions detected, eleven were of particular interest because results were consistent across at least two methods or two populations. In most of the cases, a QTL detected in the Saanen population was confirmed in the total population (on chromosomes 1, 17, 18, 22, 26, 2 and 28). In addition, other regions on chromosome 16, 19 and 21 were of high interest because they were detected in the two breeds and because the significance level was especially high (chromosome 19). Those regions are of primary interest for further fine mapping and validation in independent populations.
Mastitis validation studies in Skopelos goats
From the QTL discovery in Alpine and Saanen goats, three QTL regions on OAR16, 19 and 21 that were common in the two breeds were selected for designing a custom-made chip (576 SNPs) in order to validate the QTL in a third breed: a total of 415 Skopelos dairy goats from Greece were genotyped with this caprine array. Although these goats had monthly phenotypic records for SCC, along with milk yield, fat and protein % at each measurement time, the SNPs could not be found. This was not unexpected, as the heritabilities for all mentioned traits were low, even though all traits were strongly repeatable within and across lactations.
Conclusions
The methods used in this study have been successful in identifying QTL for mastitis resistance in sheep and goats using their respective ovine and caprine SNP50K chips. The results are consistent with the complex biology controlling mastitis susceptibility.
One of the major findings for sheep was the identification of a putative causal mutation on chromosome 3 associated with increased susceptibility to mastitis in the Lacaune breed, likely to be a functional knockout of the socs2 gene in homozygous susceptible sheep.
Comparative mapping between sheep and goat mastitis furthermore showed that a QTL on goat chromosome 19 was in the homologous sheep region carrying a QTL for Lactation Somatic Cell Score (LSCS) in the Lacaune dairy sheep population OAR11.
Traits: Nematode resistance
Introduction
On a global scale, ruminant diseases caused by gastrointestinal nematode parasite infections are the diseases with the greatest impact upon animal health and sustainability. Within Europe, nematode infections are the most costly endemic disease in meat production systems, and are close behind mastitis in milk production systems.
Grazing ruminants invariably face nematode challenges, although actual infestation levels vary according to the environment and climactic conditions. The widespread development of anthelmintic resistance in parasite populations means that alternative control strategies are required, with genetic selection being a long-term sustainable option. Many studies have shown that resistance to nematode parasites is a heritable trait in both sheep and goats, irrespective of the species of nematode infecting the host.
Nematode resistance (sheep)
Analyses have been performed to identify QTL for nematode resistance in Blackface lambs, Martinique Blackbelly x Romane backcross lambs and in the Sardinia-Lacaune Backcross population.
QTL regions associated with nematode resistance were found in 7 regions in the Blackface lamb population, in 7 regions in the Martinique Blackbelly x Romane backcross lambs and in 12 regions in the Sardinia-Lacaune backcross population.
The most significant QTL, validated across several families, was on chromosome 12, which was highly significant at the genome-wide level in the Martinique Blackbelly x Romane backcross, and also appeared in the Sardinian-Lacaune backcross population. In general however, the QTL were spaced throughout the genome with little commonality across studies. Only three common regions were observed:
• OAR9 (16-29 Mb) in the Sardinia-Lacaune backcross and Blackface populations
• OAR12 (40-46 Mb) in the Martinique Blackbelly x Romane and Sardinia-Lacaune backcross
• OAR21 (21-40 Mb) in the Martinique Blackbelly x Romane and Blackface populations
These results provided the basis for further exploration of nematode resistance, for meta-analyses combining diverse experiments, and for detailed genome exploration using full sequencing.
Meta-analyses (or, more specifically, joint analyses) of sheep nematode resistance SNP chip data suggested that genomic prediction for nematode resistance and body weight may be of value in closely related animals, but that with the current SNP chip genomic predictions are unlikely to work across breeds.
Nematode resistance (goats)
Although much of the 3SR focus has been on resistance traits in sheep, it is resistance to nematodes in goats that is of paramount importance, as goats are especially susceptible to gastrointestinal nematodes. Furthermore, it is often observed that anthelmintic resistance develops faster in goats than sheep. This may be related to the evolutionary history of goats, in which they have been preferential browsers rather than grazers, and hence throughout much of their domestication history have been able to avoid parasites. Nematode infections become particularly important, when goats are forced to be obligate grazers. This is especially the case in tropical environments such as Guadaloupe, as in these environments Haemonchus contortus infections are particularly severe and constant. Hence, survival of goats under such conditions requires them to be relatively resistant. Breeding strategies to improve nematode resistance under such conditions are therefore particularly important.
The caprine SNP50K chip, developed by joint efforts of the IGGC and 3SR, was employed to explore and quantify the genetic and genomic control of nematode resistance in goats with a view to derive enhanced breeding strategies. Successfully detecting and quantifying genetic variation for nematode resistance traits in Creole goats under tropical conditions in Guadeloupe verified the utility of this SNP50K chip (Figure 6 in Appendix 1). Indications of significant genetic regions were also found, some of which lay on the same chromosomes of significant QTL found in previous studies in the same population, albeit on previous generation animals.
A long term programme of work on this Creole goat population has now verified beyond doubt the wide genetic variation in nematode resistance, with various appropriate breeding strategies developed. The SNP50K chip results strengthen these results and give additional options for genetic improvement. However, it appears that there are no major QTL effects detectable in this population, and that the considerable genetic variation in this population is largely polygenic (i.e. influenced by more than one gene). This means that simple marker-assisted selection is unlikely to be effective, with the optimal strategy being a combination of phenotypic and targeted SNP-chip-assisted (i.e. genomic) selection.
Conclusions
The results for the sheep nematode resistance SNP chip data suggested that genomic prediction for nematode resistance and body weight may be of value in closely related animals, but that, with the current SNP chip, genomic predictions are unlikely to work across breeds.
Genetic variation for nematode resistance traits in Creole goats under tropical conditions in Guadeloupe has been detected and quantified with the use of the caprine SNP50K chip that was developed by joint efforts of the IGGC and 3SR. The caprine SNP50K chip may be suitable for detecting the presence of major QTL effects for nematode resistance or other traits in other populations/breed. This gives additional options for breeding strategies to address sustainability traits in many goat populations worldwide.
Traits: Ovulation rate (sheep)
The number of lambs born per ewe (“litter size”) is underpinned by ovulation rate and is an important measure of productivity on sheep farms and therefore the economic viability of a sheep production system. The identification of major genes controlling ovulation rate is therefore of interest to sheep farmers.
Within 3SR ovulation rate was only researched in sheep and followed two general approaches:
i) The identification of new mutations that control with ovulation rate in breeds with a reputation for high prolificacy, and
ii) The identification of genes/polymorphisms in the Cambridge breed that are associated with streak ovaries other than those in the BMP15 and GDF9 genes that have been previously identified.
Changes in the genomic sequence of the BMP15 (Bone Morphogenetic Protein 15) gene have been long known to be associated with effects on fecundity. A total of six different changes with effects on fecundity have been previously identified in various breeds. In each case the base change was found to result in higher fecundity in females that carried one copy, but resulted in sterility when a female carried two copies.
Grivette and Olkuska:
Grivette and Olkuska sheep are highly prolific and each breed carries a major gene affecting ovulation rate. A total of 40 Grivette and 63 Olkuska animals were genotyped with the ovine SNP50K chip. A genome-wide association study mapped the gene close to the previously identified BMP15 gene. Sequencing the BMP15 gene in the Grivette and Olkuska populations identified a new mutation in each breed. Both mutations result in a non-conservative amino acid substitution; T317I in Grivette and N337H in Olkuska.
In order to investigate the effect of the T317I mutation further and to verify its effect in the population, 151 adult Grivette ewes chosen at random in the population were genotyped for this mutation. The mean prolificacy of wild type, heterozygous and homozygous carriers was 1.93 1.98 and 2.31 respectively, demonstrating that animals homozygous for the mutation had an increased ovulation rate. The functional effect of this mutation was also tested in vitro after directed mutagenesis.
To investigate the effect of the N337H mutation in Olkuska, ovulation rate was compared in ewes that were wild-type, heterozygous and homozygous carriers for this mutation. Mean ovulation rates were 1.63 1.93 and 4.34 for the respective 3 genotypes, demonstrating that ewes homozygous for the mutation had a significantly higher ovulation rate.
Cambridge:
Cambridge sheep are also highly prolific and are characterised by high ovulation rate with occasional sterility due to ovarian hypoplasia. Extreme variation between individuals is consistent with segregation of a major gene controlling ovarian function. The inheritance pattern of sterile cases suggests an autosomal recessive gene unlinked to the known GDF9 mutation.
A total of 21 Cambridge animals (9 sterile ewes and 12 carriers) were genotyped with the ovine SNP50K chip (Figure 7 in Appendix 1). Homozygosity mapping identified genomic areas for which sterile ewes presented large homozygous segments. For a region of homozygosity on chromosome 2 it was possible to determine 2 regions (112Kb and 1.2Mb) putatively carrying the causative mutation. The smaller region encodes 2 strong candidate genes that have been sequenced to further investigate this mutation.
Localisation of major genes for ovulation rate
The number of SNPs and genomic regions that were found to be significantly associated with ovulation rate in each of the sheep breed investigated were: Cambridge (2 regions), Lacaune (5 SNPs), Davidsdale (2 SNPs), Grivette (one region and one SNP) and Olkuska (10 SNPs).
In the Lacaune, two markers on chromosome 7 and 8 were found to be significant (P<0.05) at the genome level, while 5 markers on chromosome 15, 21 and 25 were significant at the chromosome level.
No significant markers were found (P<0.05) at the genome level in the Davisdale. However, 2 markers on chromosome 24 and X were found to be significant at the chromosome level.
In the Grivette, no significant (P<0.05) markers were found at the genome level, but 15 were significant at the chromosome level. Five ovulation rate-related SNPs on chromosome X are very close to the BMP15 gene (312 Kb).
After sequencing the BMP15 gene, 3SR partners detected a new polymorphism (change of nucleotide C→T) at position 45935166. This non-conservative mutation is located in the regulatory region of the gene, and is strongly associated with the high prolificacy phenotype (P<0.001). This new SNP is also highly significant at the chromosome level when added to the other ovine SNP50K. If this SNP is the causal mutation, it would lead to a totally new phenotype for this gene: in contrast with all other known mutations of BMP15, homozygous animals would not be sterile and would present even higher ovulation than heterozygous animals.
The highly prolific Cambridge breed contains known mutations in the BMP15 and GDF9 genes, and females homozygous for either mutation display ovarian hypoplasia, commonly referred to as streak ovaries. However, some cases of streak ovaries in the breed cannot be explained by these mutations. Previous pedigree analysis has suggested the presence of a polymorphism in another autosomal recessive gene, unlinked to either BMP15 or GDF9, which can lead to sterility. The location and action of this gene and polymorphism is as yet unknown.
Analysis of the data generated by 3SR identified 2 highly relevant SNPs on chromosome 2, which could be the causative mutation for the Cambridge sterile phenotype. Further genotyping will need to confirm their causality; although they are located in the small region that was identified at the beginning of the project, other potential genomic areas have to be definitively excluded.
Traits: Added-value traits
Genome-wide QTL analyses of added-value traits (milk production, udder type traits, milking ease), measured in the three discovery populations (Churra, Sardinian-Lacaune Backcross and Lacaune), were performed. Several traits were highly heritable and although they did not show convincing SNP associations, the results are strongly suggestive of many of these traits being polygenic in nature. This being the case, genomic selection may be a route to exploitation.
Paratuberculosis (sheep)
Paratuberculosis, also known as Johne’s disease, is caused by Mycobacterium avium subsp. paratuberculosis (Map) that infects the digestive tract of young animals during the first few weeks of life and can prove fatal.
The population used to study mastitis and parasitism resistance (Sardinia-Lacaune Backcross population was reared in a flock where paratuberculosis was endemic for at least 10 years. Several phenotypic traits (detailed in Figure 8 in Appendix 1) related to Map infection were recorded.
3SR partners were the first to confirm the existence of a genetic variability in the antibody response against Map in sheep by estimating the heritability for this trait as 0.28±0.10. This heritability is higher than that for Map in cattle (0.10-0.15) and is also higher than previously estimated for sheep (0.15) which was done using post-mortem analysis rather than antibody response (as was done in 3SR). This is consistent with the existence of genetic variation in susceptibility to Map and forms an important basis for future studies on the identification of genetic loci linked to susceptibility to Map.
The existence of genetic variation in susceptibility to paratuberculosis may lead to the identification of disease resistance markers and consequently selection may be used as an additional tool for control of the disease.
Potential Impact:
With 100 million sheep and 11 million goats in the EU and a combined global population of around 1.94 billion heads, these two species are of considerable economic, social and environmental importance. The sheep and goat sectors play an important role in the economy of many EU countries, often underpinning the economic viability of rural communities. Their ability to make good use of low quality pasture as well as being ideally suited to being kept on small family farms makes them extremely valuable. Due to their small size and grazing preferences and through often being highly adapted to local conditions, they also play an essential role in landscape maintenance and nature conservation.
The future sustainability of the sheep and goat sectors relies on breeding animals that are resilient to disease challenges, as well as being productive and efficient. Understanding the mechanisms of resistance and developing genomic tools and technologies is likely to be crucial in helping address these challenges.
Mastitis and nematode worm infections are the two biggest health and welfare threats to sheep and goats in Europe and beyond, causing huge economic and productivity losses. Previous research has indicated that susceptibility to both infections is at least partly under genetic control. Understanding the basis of genetic control to susceptibility will help direct breeding programmes to produce more resistant and, therefore, healthier and more productive animals.
Ovulation rate is the key factor determining the number of lambs per litter for sheep. Some particularly prolific breeds have been identified along with some important genes that affect fertility.
Prior to the 3SR project, the genomic resources for sheep and goats lagged behind the other major livestock species, putting the industry at a disadvantage in terms of the potential to implement selective breeding.
It is only recently (and aided by the work of the 3SR consortium) that high quality, draft reference genome sequences for sheep and goats have been published. The benefits that can be delivered by better understanding the genetic basis of health and welfare traits are highly significant.
An International Sheep Genomics Consortium (ISGC) had previously been formed, and by early 2009 the ISGC’s work had led to the availability of improved sheep maps, a sheep Bacterial Artificial Chromosome (BAC) library and associated end sequences, the first Ovine 50k SNP Beadchip and a first draft of the sheep genome assembly. This first draft was estimated to cover around 80% of the unique sequence in the sheep genome, indicating that there were still significant gaps in the sequence, and therefore quite a bit of work needed to improve the quality of the reference genome for sheep.
The situation for sheep was better than that for goats. In 2009, no concerted international research effort to develop genomic resources had been established, and consequently no reference genome sequence was available; although there was a low density (1536) SNP chip.
A number of external events occurred in 2010 and throughout the duration of the 3SR project that provided a valuable opportunity to increase the potential impact and the value for money that could be achieved through the 3SR project. These opportunities resulted mostly through the close links between the partners in the 3SR consortium and the wider international communities for both sheep and goats.
The 3SR consortium worked closely with the complementary global efforts on-going through the ISGC and the International Goat Genome Consortium (IGGC). By doing so, 3SR has been able to make available to European researchers complementary resources provided by major research projects outside Europe, and has been involved in developing resources which will be able to enhance the livelihood of goat and sheep farmers worldwide;
• 3SR played a key part in the development of a well-annotated sheep reference genome sequence. This will provide the international sheep genomics community a valuable tool for underpinning further research
• 3SR partners were also central to development of a caprine 50K SNP chip which provides unparalleled opportunities to explore and understand nematode resistance in goats, to detect QTL for resistance, and to derive novel breeding strategies.
A key strength of 3SR has been that because commercial populations were used in the project, the results are directly relevant to the sheep and goat industry. Partners in the project have used populations of sheep drawn from a range of divergent breeds from throughout Europe to maximise the relevance of the polymorphisms found to the EU sheep and goat industries. The project consortium has strong links and a wide network with commercial breeders within the EU to ensure that effective dissemination of project outcomes is achieved.
Genetic variation in sustainability traits, such as mastitis susceptibility and nematode resistance, has been found. A putative causal mutation on chromosome 3 associated with increased susceptibility to mastitis was found in the Lacaune breed.
Both an ovine and a caprine custom-made DNA array of 500-1000 SNPs have been designed and developed on the basis of the results achieved in the search for genetic variation in mastitis susceptibility and nematode resistance. These have been used in commercial sheep and goat populations to validate the existence of predicted molecular markers for mastitis susceptibility and nematode resistance. Whilst the cross population analysis was inconclusive, the results may provide a valuable source of information for within breed breeding programmes.
Thus, the 3SR project has succeeded in providing selectable genetic markers that can be affordably applied in specific sheep and goat breeds to make contributions to improving animal health, welfare, sustainability and the long-term competitiveness of small ruminant production in the EU.
None of the SNPs for traits related to mastitis in Chios sheep was associated with milk production, implying that this information can be directly used to genetically control mastitis without compromising the productivity of the animals.
The caprine SNP50K chip, developed by joint efforts of the International Goat Genome Consortium and 3SR, may be further used by researchers and breeders to identify specific markers for other traits of interest. Knowing which animals are particularly susceptible or resistant to various health challenges will help breeders make decisions about how to breed animals in the future.
The consortium has furthermore developed and used cutting-edge genomic technologies such as an open-access genome browser, and produced large amounts of genome-related data (genotypes for 7,500 animals, genome sequences for 50 sheep, and over a terabyte of RNA-seq data). This has already led to the identification of causative mutations, candidate genes, SNPs and genetic markers for regions that affect the production of offspring in sheep, susceptibility to mastitis and nematode resistance in sheep and goats.
3SR has contributed to significant improvements in the genomics resources available for sheep and goats:
• Identified causative mutations affecting ovulation rate in Grivette (French) and Olkuska (Polish) sheep, and 2 candidate genes for sterility were identified in Cambridge sheep.
• In both sheep and goats, small regions of the genome were identified that have a significant effect on mastitis susceptibility and nematode resistance. Genetic markers for these regions were validated in commercial populations.
• The 3SR consortium helped the ISGC deliver sheep reference genome v3.1
• The 3SR consortium produced over a terabyte of RNA-seq data to assist with the annotation of the sheep genome
• A 50K SNP chip has been developed for goats. This enabled over 2000 animals to be genotyped for selectable genetic markers of interest within the 3SR project.
• Confirmed genetic variation to paratuberculosis in sheep (heritability 0.28±0.10)
• An open access Genome Browser for sheep and goats was developed http://genome.itb.cnr.it/gb2/gbrowse/oarv3.1/
• A whole-genome radiation hybrid (RH) map was produced for goats, containing more than 30,000 markers ordered across the chromosomes
• As well as the reference genomes, the genomic tools (for example genotyping arrays) have been greatly advanced throughout the duration of the project
Some of the results generated can be used immediately by breeding groups to increase selection accuracy, either through adopting a marker-assisted section approach or by prioritising polymorphisms for genomic selection. The exploitation plan will need to be breed specific, as optimal utilisation varies according to the breed structure and size. Numerically larger breeds will be in a better position to exploit genomic selection and hence make greater gains, whereas smaller breeds will realistically only be able to utilise individual markers and hence realise smaller benefits.
Other results provide a basis for better understanding the underpinning biology of the traits studied. The availability of such knowledge, combined with the identification of causative mutations, will play an important part in allowing the management and monitoring of future breeding programmes that will be needed to tackle future challenges, including these caused by climate change.
These tools and findings may assist breeders to produce animals with enhanced health and prolificacy, delivering significant benefits to productivity within the sheep and goat farming industries.
Besides the core beneficial impact on animal health, welfare and sustainability and the long-term competitiveness of small ruminant production in the EU delivered by the 3SR project, a wider and long-term impact may be expected from the project. The substantial new genomic resources generated (which are freely and publicly available) and the advanced basic understanding of sheep and goat genomes will guide future research efforts and boost our knowledge of these species. This helps accelerate and reduce the costs of future genetics/genomics research into traits of importance to animal health, welfare, sustainability and competitiveness – a substantial lasting impact of this timely project.
All in all, the project has been able to deliver a significant improvement in the available genomic information and technologies for goats and sheep, whilst also having a lasting impact on European research capacity.
The legacy of 3SR will hopefully be an improvement in the health, welfare and productivity of sheep and goat breeds in Europe and globally, underpinned by the better understanding of the genetics controlling their resilience to disease generally and their productivity in a farming context.
Dissemination Activities
In addition to an exceptionally strong research dimension, dissemination and knowledge management featured prominently in the project objectives, activities and outcomes, as summarised below.
The dissemination and knowledge management objectives of the 3SR project were:
• To create awareness and disseminate, manage and exploit knowledge generated in this Collaborative Project.
• To ensure transfer of knowledge from the project consortium to potential end-users and stakeholders, and promote exploitation of results.
• To ensure smooth communication and transfer of knowledge amongst participating partners.
• To link this project with other relevant research programmes and explore potential synergies in the dissemination and exploitation of results.
• To provide the platform for protection of Intellectual Property that may arise during the project.
The work towards meeting these objectives was organised around five tasks and eight deliverables involving all project partners drawn from research, industry and academia.
The key activities and outcomes are highlighted below:
1. A project website (http://www.3srbreeding.eu/) was set-up and was being continuously updated throughout this period as new information and material were made available. The website will continue being updated at least until August 2014.
2. A total of 12 peer-reviewed scientific articles were published in the international literature; a complete list with appropriate links can be found on the project website.
3. A total of 66 presentations featured in 33 national and international scientific conferences; these presentations are listed on the project website.
4. Four popular-press articles were written in non-scientific terms for farmers and the industry.
5. Two project Newsletters were prepared and distributed to more than 5,500 recipients, and also posted on the project website.
6. A Technical Bulleting summarising results for the industry was prepared. The scope of the bulletin was to increase awareness among the industry representatives and facilitate the uptake of research results at the farm, breed and population levels.
7. On seven occasions project partners exchanged visits to enhance intra-project collaboration and communication.
8. A total of 17 dissemination and industry briefing meetings took place in seven countries where project partners explained results and discussed implementation strategies with farmers, advisors, breed associations, breeding companies and industry representatives.
9. A sheep genomics seminar was organised in conjunction with the annual meeting of the European Federation of Animal Science (26-30 August 2013, Nantes, France). This event was linked to a similar event organised by another EC-funded research project (“Quantomics: From sequence to consequence – tools for the exploitation of livestock genomes”; Agreement no. 222664) thereby forging closer links between these two relevant projects and promoting complementary outcomes in the area of genomics for the improvement of livestock species.
10. Technology translation initiatives were developed including a user-friendly poster presenting the main concepts, goals and outcomes of the project, and a mastitis monitoring and improvement programme for small ruminant farms.
11. A plan for exploitation of the project results was completed based on a partners’ survey and industry feedback.
12. A Consortium Agreement was produced, agreed and signed by all partners at the start of the project. In addition, a Standard Operating Procedure (SOP) was developed outlining the process that should be followed by all partners wishing to publish results of 3SR. This SOP was followed throughout the project.
The above outcomes are expected to have a long-term impact on science as well as practical applications that will last well beyond the completion of the project.
List of Websites:
www.3srbreeding.eu
