Reproductive Genomics in Domestic Animals

Zhihua Jiang, Troy L. Ott, Zhihua Jiang, Troy L. Ott

• Publisher: Wiley-Blackwell
• Year: 2011
• ISBN:
  9780470961827
• Edition: 1
• Topic: Biowissenschaften
• Subtopic: Genetik & Genomik

Reproductive Genomics in Domestic Animals is a thorough examination of genomics in the livestock industry, encompassing genome sciences, genome biotechnology, and reproduction. Recent developments in molecular genetics and genomics have enabled scientists to identify and characterize genes contributing to the complexity of reproduction in domestic animals, allowing scientists to improve reproductive traits. Providing the livestock industry with essential tools for enhancing reproductive efficiency, Reproductive Genomics in Domestic Animals surveys the current status of reproductive genomes and looks to the future direction of research.

Part I

Quantitative Genomics of Reproduction

1

Reproductive Genomics: Genome, Transcriptome, and Proteome Resources

Noelle E. Cockett

1.1 Introduction

Genomic resources, tools, and technologies that can be applied to studies in livestock species, including investigations related to reproduction, have been under development for the last decade. While many of the genomic approaches were originally developed for use in humans or laboratory model animals, they have been successfully applied to studies in livestock. There are now a myriad of resources specific to livestock species, such as well-characterized genome maps, high-resolution genome, and complementary DNA (cDNA) sequences, expression arrays, and high-density genetic marker chips. In addition, there is an explosion of high-throughput technology that will enhance these investigations, increasing the scope and accuracy of the results beyond anything that was imagined just 5 years ago. These technologies advance studies of single gene expression to full gene networks, from single gene sequences to whole genomes,and from hundreds of genetic markers to tens of thousands markers—all assayable in a few weeks to months as opposed to years.

These resources and technologies can be combined in innovative ways to advance two areas of research on reproductive traits, specifically the identification of genes or genetic regions influencing phenotypes and the characterization of expression of genes that are associated with traits.

1.2 Discovery of underlying genetic influences

The first area of interest for researchers studying reproductive traits is the characterization of genetic variation among animals or populations underlying a phenotypic trait, leading to the identification of the genetic cause of the phenotype. Two general approaches have been successfully used over the last 10–15 years, with a third approach now on the horizon. In the first approach,polymorphisms in a candidate gene likely to be involved in the phenotype are tested for associations with different manifestations or phenotypes of the trait. The candidate genes are selected for analysis based on an understanding of trait physiology and/or because of their involvement in similar traits in other species. In the second approach, genetic markers are analyzed for linkage with the phenotype using pedigrees of animals segregating for the trait and the markers. This analysis identifies genetic regions that contain associated genes. By testing additional markers through the families, the interval is narrowed so candidate genes can be selected. The third approach, referred to as whole genome associations, will soon be possible for livestock species now that the development of high-density single nucleotide polymorphism (SNP) arrays are readily available. However, the application of whole genome associations requires very large numbers of phenotyped animals, which is a limitation for most research projects.

1.2.1 Candidate gene associations

As mentioned, the candidate gene approach uses information of the trait to determine likely candidates for the underlying gene(s). The choice of the gene is strengthened by its involvement in comparable traits in other species or its location in a region previously identified as containing a quantitative trait loci (QTL) with similar attributes.

In the past, polymorphisms in a candidate gene were routinely detected by polymerase chain reaction—restriction fragment length polymorphisms (PCR-RFLP), which involves steps of amplifying the gene, digesting the amplicon with a restriction enzyme, and then using gel electrophoresis to separate the resulting fragments. In the PCR-RFLP technique, gene sequence differences among animals are detected by whether or not a restriction enzyme cuts, resulting in different-sized fragments. The genetic differences are usually due to an SNP within the restriction enzyme recognition site, although there might be genetic differences due to insertions/deletions (in/del) in the gene, which will also result in fragment size differences, although there is no variation in the restriction enzyme recognition site. Animals are expected to have two alleles for every gene except those on the X and Y chromosomes in males, so that the presence of one fragment on the electrophoresis gel would indicate that an animal is homozygous for the PCR-RFLP allele whereas the presence of two different-sized fragments would suggest that an animal is heterozygous. However, an animal might be misclassified as a homozygote if there is a polymorphism in the PCR primer sequence, which prevents that allele from being amplified and therefore, not detected on the electrophoresis gel—referred to as a “null” allele. A null allele will often be detected when misparentages are routinely found for a marker system. An animal might also be misclassified if another, nonallelic form of the gene is amplified with the PCR primers and digestion with the restriction enzyme results in a different-sized fragment. A nonallelic form is revealed by sequencing the fragments contained within the electrophoretic bands, which is a recommended step when establishing any marker system.

However, new technologies have significantly advanced our ability to identify SNPs and then explore multiple candidate genes at one time at a much lower cost/ polymorphism than the PCR-RFLP method. The identification of SNPs within a gene or genetic region is now relatively easy. To do this, the genomic DNA of key animals within a population is sequenced using high throughput automatic sequencing and then compared with other sequences within the population or to sequences in publically available databases. The later approach is referred to as in silico SNP detection. Regardless of the approach, confidence of the SNP is dependent on the quality of the sequence across the multiple sources of data.

Once an SNP is identified, the polymorphism can be detected by establishing a PCR-RFLP assay. However, allele-specific PCR using allele-specific oligonucleotides (ASOs) is an emerging technique for detecting genetic variation created by the SNP (Saiki et al. 1986). The 3′ ends of the primers used in the PCR amplification step of the ASO technique are designed to include the polymorphic site so that amplification of the animal’s DNA is dependent on the absence or presence of the polymorphism within the primer sequence. Allele-specific primers can be combined into a single amplification reaction and the presence of the specific allele detected by the melting temperature of the alleles (Papp et al. 2003; Wang et al. 2005). Appropriate controls and design of the primers (e.g., Strerath et al. 2007) are critical in the allele-specific amplification assay so that absence of amplification is due to the polymorphism and not because of technical problems.

SNP arrays are an extension of the ASO method, but by spotting multiple ASOs onto a membrane or bead, multiple alleles or even multiple genetic markers can be assayed in a single run. Custom-built SNP chips specific to a trait are usually designed in a 92-, 384-, or 1534-SNP format. While the cost/ SNP is lower for the SNP chip than with the PCR-RFLP or allele-specific amplification techniques, the initial setup for the chip is substantially higher. Thus, the number of SNPs that are tested and the number of animals included in the analysis will determine whether a custom-built SNP array is economical.

Emerging technology is now allowing the detection of differences in copy number variant (CNV) among animals. For some time, copy number variation has been associated with diseases (McCarroll 2008; Schaschi et al. 2009), while the ongoing analyses of livestock whole genome sequences has revealed the presence of CNV in multiple gene systems involved with innate immunity, including milk composition traits (Rijnkels et al. 2009; Tellam and Bovine Genome Sequencing and Analysis Consortium 2009). Detection of differences among animals for genes that are known to be present in the genome in multiple copies is now possible using microarray technology (Baumbusch et al. 2008), with higher copy number resulting in greater intensity for that spot on the array.

Once the polymorphism is detected within a population, the genotypes are usually analyzed for association with the trait by comparing the trait means among the marker genotypes (Rocha et al. 1992). Appropriate statistical models are needed in order to account for additive, dominant, and epistatic effects. In addition, the selection of animals used in the analysis must be sufficiently broad; otherwise, the marker alleles will merely serve as a trace of unique families, particularly when one of the alleles is at a very low frequency in the population and present only in one family in the analysis. This situation can result in a spurious significant association, simply because the family differs for the trait and not because the allele itself is associated.

The choice of the candidate gene(s) can be strengthened by its association with similar traits in the same or other species. Possible candidate genes can be found through literature searches using key words based on the trait physiology or through searches of databases devoted to genetic abnormalities. One such database for livestock traits is called Online Mendelian Inheritance in Animals (OMIA; www.omia.angis.org.au/). The OMIA database contains details on genes, inherited disorders, and traits for a large range of animals species, similar to what is found within Online Mendelian in Man (OMIM; www.ncbi.nlm.nih.gov/sites/entrez?db=omim).

There are also databases that describe the location of QTLs for traits of interest in livestock species (Table 1.1). Additional candidate genes can be identified by searching genetic sequences that lie within QTL intervals and have involvement in the physiology of the trait, providing not only functional evidence but also positional evidence for inclusion in the candidate gene analysis. These genes are therefore referred to as “positional candidate genes” (see below).

Table 1.1 Websites containing genomics information in livestock species.

Species	Website	Information
Cattle	www.animalgenome.org/QTLdb/cattle.xhtml	QTL
	www.vetsci.usyd.edu.au/reprogen/QTL_Map/	QTL
	www.hgsc.bcm.tmc.edu/projects/bovine/	Genome sequence
	bovinegenome.org	Genome project
Goat	dga.jouy.inra.fr/cgi-bin/lgbc/main.pl?BASE=goat	Genome project
Horse	www.uky.edu/Ag/Horsemap/welcome.xhtml	Genome project
	www.broad.mit.edu/mammals/horse	Genome sequence
Sheep	rubens.its.unimelb.edu.au/~jillm/jill.htm	Primary Web source
	www.livestockgenomics.csiro.au/perl/gbrowse.cgi/vsheep2/	Virtual sheep genome
	www.ncbi.nlm.nih.gov/genome/guide/sheep/index.xhtml	NCBI resources
	www.sheephapmap.org/	International Sheep Genome Consortium
Pig	www.animalgenome.org/QTLdb/pig.xhtml	QTL
	www.sanger.ac.uk/Projects/S_scrofa/	Genome sequence
	www.piggenome.org/index.php	Genome project

1.2.2 Analysis of genetic variation

The second approach for detecting genes or, more commonly, genetic regions involved in traits is based on identifying and characterizing genetic variation that is found in pedigrees of animals. This approach has most commonly been done using linkage analysis, which examines the segregation of marker alleles through animal families with known phenotypes (Nejati-Javaremi and Smith 1995; Knott and Haley 2000; de Koning et al. 2003) and subsequent refinement of the genetic interval containing the trait locus (Riquet et al. 1999; Farnir et al. 2002). The data are analyzed to determine the coinheritance of marker alleles with the causative genetic mutation, presumably because they are closely located within the genome.

Linkage mapping requires pedigrees with specific family structures; these pedigrees are most commonly reciprocal backcrosses or F2 crosses developed from lines or breeds of animals that significantly differ for the trait. The analysis can include families within a single breed or line but the key parents must be heterozygous for both the markers and the trait in order for linkage to be detected. As with the association analyses, appropriate statistical models are needed to detect genetic mutations that are controlled by complex gene actions, such as the imprinted callipyge (Cockett et al. 1994, 1996) and IGF2 (Van Laere et al. 2003) loci. The effects of these loci would not h...

Table of contents

• Cover
• Title page
• Copyright page
• Contributors
• Preface
• Part I: Quantitative Genomics of Reproduction
• Part II Physiological Genomics of Reproduction
• Part III Genomics and Reproductive Biotechnology
• Index