The words you are searching are inside this book. To get more targeted content, please make full-text search by clicking here.


Discover the best professional documents and content resources in AnyFlip Document Base.
Published by soedito, 2018-12-29 07:06:42



Reproductive Genomics in
Domestic Animals

Edited by

Zhihua Jiang
Troy L. Ott

A John Wiley & Sons, Inc., Publication

Reproductive Genomics in
Domestic Animals

Reproductive Genomics in
Domestic Animals

Edited by

Zhihua Jiang
Troy L. Ott

A John Wiley & Sons, Inc., Publication

Edition first published 2010
© 2010 Blackwell Publishing

Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing program has
been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell.

Editorial Office
2121 State Avenue, Ames, Iowa 50014-8300, USA

For details of our global editorial offices, for customer services, and for information about how to apply
for permission to reuse the copyright material in this book, please see our website at

Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients,
is granted by Blackwell Publishing, provided that the base fee is paid directly to the Copyright Clearance Center,
222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license by

CCC, a separate system of payments has been arranged. The fee codes for users of the Transactional Reporting
Service are ISBN-13: 978-0-8138-1784-2/2010.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names
and product names used in this book are trade names, service marks, trademarks or registered trademarks of
their respective owners. The publisher is not associated with any product or vendor mentioned in this book.
This publication is designed to provide accurate and authoritative information in regard to the subject matter
covered. It is sold on the understanding that the publisher is not engaged in rendering professional services.
If professional advice or other expert assistance is required, the services of a competent professional should
be sought.

Library of Congress Cataloging-in-Publication Data

Reproductive genomics in domestic animals / editors, Zhihua Jiang, Troy L. Ott. – 1st ed.
p. cm.

Includes bibliographical references and index.
ISBN 978-0-8138-1784-2 (hardback : alk. paper) 1. Domestic animals–Genetics.
2. Domestic animals–Reproduction. 3. Livestock–Genetics. 4. Livestock–Reproduction.

5. Genomics–Research. I. Jiang, Zhihua, 1959– II. Ott, Troy L.
SF105.R45 2010

A catalog record for this book is available from the U.S. Library of Congress.

Set in 10 on 13 pt Trump Mediaeval by Toppan Best-set Premedia Limited
Printed in Singapore

1 2010


Contributors xi
Preface xv

Part I Quantitative Genomics of Reproduction 3
1 Reproductive Genomics: Genome, Transcriptome, and Proteome Resources
Noelle E. Cockett 5
1.1 Introduction 14
1.2 Discovery of underlying genetic influences 16
1.3 Characterization of gene expression 17
1.4 Resources for protein analysis 17
1.5 Future research directions
2 Quantitative Genomics of Female Reproduction 23
Jeffrey L. Vallet, Dan J. Nonneman, and Larry A. Kuehn 26
2.1 Introduction 37
2.2 Female reproductive phenotypes 41
2.3 Genetic markers and genotyping methods 43
2.4 Association of phenotypes with genotypes
2.5 Some illustrative examples of reproductive QTL 53
2.6 Future research directions
References 53
3 Quantitative Genomics of Male Reproduction 56
Eduardo Casas, J. Joe Ford, and Gary A. Rohrer 60
3.1 Introduction
3.2 Male reproduction phenotypes 67
3.3 Genetics, genomics, and quantitative trait loci (QTL)
3.4 QTL identified for male reproduction traits 67
3.5 Future research directions 68

References v

4 Genetics and Genomics of Reproductive Disorders
Peter Dovc, Tanja Kunej, and Galen A. Williams

4.1 Introduction
4.2 Reproductive disorders associated with the ovary

vi Contents 73
4.3 Reproductive disorders associated with the vagina and uterus 78
4.4 Reproductive disorders associated with pregnancy and placenta 85
4.5 Reproductive disorders associated with male reproductive organs 89
4.6 Reproductive disorders associated with embryos and fetuses 90
4.7 Future research directions
5 Genomics of Reproductive Diseases in Cattle and Swine 100
Holly Neibergs and Ricardo Zanella 102
5.1 Introduction 108
5.2 Bovine paratuberculosis 110
5.3 BRD 111
5.4 Brucellosis in cattle 113
5.5 Leptospirosis in swine 113
5.6 Aujeszky’s disease (pseudorabies)
5.7 PRRS 129
5.8 Future research directions
References 129
6 Comparative Genomics of the Y Chromosome and Male Fertility 136
Wansheng Liu 142
6.1 Introduction 146
6.2 Characteristics of the mammalian Y chromosome
6.3 Sequence and gene content of the Y chromosome 157
6.4 Function of Y chromosome genes in spermatogenesis and male fertility
6.5 Polymorphisms of the Y chromosome and male fertility 157
6.6 Future research directions 158
References 174
7 Mitochondriomics of Reproduction and Fertility
Zhihua Jiang, Galen A. Williams, Jie Chen, and Jennifer J. Michal 181

7.1 Introduction 183
7.2 Cytoplasm mitochondrial genomes in fertility and reproduction
7.3 Nuclear mitochondrial genomes in fertility and reproduction 183
7.4 Future research directions


Part II Physiological Genomics of Reproduction

8 Functional Genomics Studies of Ovarian Function in Livestock:
Physiological Insight Gained and Perspective for the Future
Beau Schilling and George W. Smith

8.1 Introduction

Contents vii

8.2 Transcriptomics of ovarian tissues: EST sequencing 184
8.3 Transcriptomics of ovarian tissues: Microarray studies 189
8.4 Proteomics of ovarian tissues 196
8.5 Future research directions 197
9 Physiological Genomics of Preimplantation Embryo Development in
Production Animals 205
Luc J. Peelman 206
9.1 Introduction 219
9.2 Preimplantation developmental stages and transcriptomics 220
9.3 Preimplantation developmental systems and transcriptomics
9.4 Future research directions 231

References 231
10 Physiological Genomics of Conceptus–Endometrial Interactions Mediating 235
Corpus Luteum Rescue 242
Troy L. Ott and Thomas E. Spencer 243

10.1 Introduction 251
10.2 Physiological genomics of luteal regression
10.3 Physiological genomics of blocking luteal regression 251
10.4 Future research directions 252
References 261
11 Physiological Genomics of Placental Growth and Development 263
Sukanta Mondal
11.1 Introduction
11.2 Placental development: Basics 269
11.3 Placental hormones and peptides 270
11.4 Transcriptomics of placental development 272
11.5 Future research directions 279
References 284

12 Cellular, Molecular, and Genomic Mechanisms Regulating Testis Function
in Livestock
Kyle Caires, Jon Oatley, and Derek McLean

12.1 Introduction
12.2 Spermatogenesis
12.3 Transcriptomics of testis in bulls
12.4 Reproductive genomics in boars
12.5 Future research directions


viii Contents 291

Part III Genomics and Reproductive Biotechnology 293

13 The Epigenome and Its Relevance to Somatic Cell Nuclear Transfer and 293
Nuclear Reprogramming 293
Jorge A. Piedrahita, Steve Bischoff, and Shengdar Tsai 297
13.1 Introduction 307
13.2 The epigenome 310
13.3 Epigenetic reprogramming 310
13.4 Genomic imprinting
13.5 SCNT and epigenetic abnormalities 317
13.6 Future research directions
References 318
14 Biotechnology and Fertility Regulation 323
Valéria Conforti 326
14.1 Introduction 332
14.2 Basic aspects in vaccine development 333
14.3 Specific aspects in vaccine development
14.4 Sperm antigens 339
14.5 Zona pellucida antigens
14.6 LHRH antigens 339
14.7 Future research directions 340
References 347
15 Proteomics of Male Seminal Plasma 352
Vera Jonakova, Jiri Jonak, and Marie Ticha 352

15.1 Introduction 367
15.2 Proteins of seminal plasma
15.3 Function of seminal plasma proteins 367
15.4 In vitro effects of seminal plasma proteins 369
15.5 Properties of major proteins of seminal plasma of domestic animals 374
15.6 Future research directions 380
References 383

16 Evolutionary Genomics of Sex Determination in Domestic Animals
Eric Pailhoux and Corinne Cotinot

16.1 Introduction
16.2 State of knowledge of sex differentiation
16.3 Sex differentiation in domestic mammals
16.4 Sex determination in nonmammal domestic species
16.5 Future research directions


Contents ix

17 Toxicogenomics of Reproductive Endocrine Disruption 397
Ulf Magnusson and Lennart Dencker
17.1 Introduction 398
17.2 Reproductive endocrine disruption 401
17.3 Reproductive endocrine disruptors 404
17.4 Toxicogenomics 408
17.5 Future research directions 408

References 413

18 Nutrigenomics for Improved Reproduction 413
John P. McNamara 414

18.1 Introduction 417
18.2 Nutritional physiology of reproduction: A brief view
18.3 Mechanistic connections between nutrient flux and reproductive 421
18.4 History of integration of physiological state, nutrient flux, 427
and reproduction 435
18.5 Nutritional physiology of pregnancy and lactation
18.6 Nutrigenetics and nutrigenomics approaches for improved fertility,

pregnancy, and lactation
18.7 Future research directions


Index 439


Steve Bischoff, Department of Molecular Peter Dovc, Department of Animal Science,
Biomedical Sciences and Center of Compara- University of Ljubljana, Groblje 3, SI-1230
tive Medicine and Translational Research, Domzale, Slovenia
College of Veterinary Medicine, North
Carolina State University, Raleigh, NC J. Joe Ford, USDA, Agricultural Research
27606 Service, U.S. Meat Animal Research Center,
Clay Center, NE 68933-166
Kyle Caires, Department of Animal Sciences,
Center for Reproductive Biology, Washington Zhihua Jiang, Department of Animal
State University, Pullman, WA Sciences, Center for Reproductive Biology,
Washington State University, Pullman, WA
Eduardo Casas, USDA, Agricultural Research 99164-6351
Service, U.S. Meat Animal Research Center,
Clay Center, NE 68933-0166 Jiri Jonak, Laboratory of Diagnostic for
Reproductive Medicine, Institute of Mole-
Jie Chen, Department of Animal Sciences, cular Genetics, Academy of Sciences of the
Center for Reproductive Biology, Washington Czech Republic, Videnska 1083, 142 20
State University, Pullman, WA 99164-6351; Prague 4, Czech Republic
and College of Animal Sciences and Tech-
nology, Nanjing Agricultural University, Vera Jonakova, Laboratory of Diagnostics
Nanjing 210095, China for Reproductive Medicine, Institute of
Molecular Genetics, Academy of Sciences of
Noelle E. Cockett, Department of Animal, the Czech Republic, Videnska 1083, 142 20
Dairy and Veterinary Sciences, Utah State Prague 4, Czech Republic
University, Logan, UT 84322-4900
Larry A. Kuehn, USDA, Agricultural Research
Valéria Conforti, Cincinnati Zoo & Botanical Service, U.S. Meat Animal Research Center,
Garden, 3400 Vine Street, Cincinnati, OH Clay Center, NE 68933
Tanja Kunej, Department of Animal Science,
Corinne Cotinot, CNRS, FRE 2857, F-78350, University of Ljubljana, Groblje 3, SI-1230
Jouy-en-Josas, France Domzale, Slovenia

Lennart Dencker, Department of Pharma- Wansheng Liu, Department of Dairy and
ceutical Sciences at Biomedical Centre, P.O. Animal Science, Center for Reproductive
Box 594, Uppsala University, SE-756 45 Biology and Health, The Pennsylvania State
Uppsala, Sweden University, University Park, PA 16802


xii Contributors

Ulf Magnusson, Department of Clinical Eric Pailhoux, INRA, UMR 1198 Biologie du
Sciences and Centre for Reproductive Développement et Reproduction, F-78350
Biology in Uppsala, P.O. Box 7054, Swedish Jouy-en-Josas, France
University of Agricultural Sciences, SE-750
07 Uppsala, Sweden Luc J. Peelman, Department of Nutrition,
Genetics and Ethology, Faculty of Veteri-
Derek McLean, Department of Animal nary Medicine, Ghent University, 9820
Sciences, Center for Reproductive Biology, Merelbeke, Belgium
Washington State University, Pullman, WA
Jorge A. Piedrahita, Department of Mole-
John P. McNamara, Department of Animal cular Biomedical Sciences and Center of
Sciences, Center for Reproductive Biology, Comparative Medicine and Translational
Washington State University, Pullman, WA Research, College of Veterinary Medicine,
99164-6351 North Carolina State University, Raleigh,
NC 27606
Jennifer J. Michal, Department of Animal
Sciences, Center for Reproductive Biology, Gary A. Rohrer, USDA, Agricultural Research
Washington State University, Pullman, WA Service, U.S. Meat Animal Research Center,
99164-6351 Clay Center, NE 68933-0166

Sukanta Mondal, Division of Animal Beau Schilling, Laboratory of Mammalian
Physiology, National Institute of Animal Reproductive Biology and Genomics, Depart-
Nutrition and Physiology (Indian Council of ment of Animal Science, Michigan State
Agricultural Research), Adugodi, Bangalore— University, East Lansing, MI 48824-1225
560 030, Karnataka, India
George W. Smith, Laboratory of Mammalian
Holly Neibergs, Department of Animal Reproductive Biology and Genomics,
Sciences, Center for Reproductive Biology, Department of Physiology, Michigan State
Washington State University, Pullman, WA University, East Lansing, MI 48824-1225
Thomas E. Spencer, Department of Animal
Dan J. Nonneman, USDA, Agricultural Science, Center for Animal Biotechnology
Research Service, U.S. Meat Animal and Genomics, Texas A&M University,
Research Center, Clay Center, NE 68933 College Station, TX 77843-2471

Jon Oatley, Department of Dairy and Marie Ticha, Laboratory of Diagnostic for
Animal Sciences, Center for Reproductive Reproductive Medicine, Institute of Mole-
Biology and Health, The Pennsylvania State cular Genetics, Academy of Sciences of the
University, University Park, PA Czech Republic, Videnska 1083, 142 20
Prague 4, Czech Republic
Troy L. Ott, Department of Dairy and
Animal Sciences, Center for Reproductive
Biology and Health, The Pennsylvania State
University, University Park, PA 16802

Contributors xiii

Shengdar Tsai, Department of Molecular Galen A. Williams, Devers Eye Institute,
Biomedical Sciences and Center of Compara- 1225 NE 2nd Ave., Portland, OR 97232
tive Medicine and Translational Research,
College of Veterinary Medicine, North Ricardo Zanella, Department of Animal
Carolina State University, Raleigh, NC Sciences, Center for Reproductive Biology,
27606 Washington State University, Pullman, WA
Jeffrey L. Vallet, USDA, Agricultural
Research Service, U.S. Meat Animal Research
Center, Clay Center, NE 68933


Reproductive efficiency has been considered loci associated with the traits. Chapter 3
one of the most critical factors affecting the defines the male reproductive phenotypes
productivity and profitability of the live- and updates the genes/quantitative trait loci
stock industries. Unfortunately, in spite of a associated with these traits. Chapters 2 and
significant improvement in growth, feed effi- 3 also include methods and technologies for
ciency, and carcass and meat quality due to the development and discovery of genomic
genetic selection and management advances, markers as well as their genotyping formats.
reproductive efficiency has declined in most Chapter 4 covers genetic and genomic
livestock species. Due to low heritabilities, aspects of reproductive disorders associated
and sex-limited complexity, it has been with the ovary, vagina, and uterus; preg-
very difficult to improve reproductive traits nancy and placenta; male reproductive
using traditional selection methods. The organs; and embryos and fetuses. Chapter 5
rapid development of molecular genetics deals with genetic and genomic aspects of
and genomics in recent years has, however, reproductive diseases, such as paratubercu-
enabled the identification, characterization, losis, respiratory disease, and brucellosis in
and utilization of genes and pathways that cattle, and leptospirosis, Aujeszky’s disease,
contribute to the genetic complexity of and porcine reproductive and respiratory
reproduction in domestic animals. This book syndrome in swine. Chapter 6 focuses on the
reviews the current status of reproductive structure, function, and evolution of the Y
genomics, transcriptomics, and proteomics chromosome and its effect on male fertility.
and highlights the current and potential Chapter 7 describes both mitochondrial
genomics tools and reagents for improving genomes in the cytoplasm and nucleus and
reproductive efficiency in domestic animals. their involvements in male reproduction,
It is our goal to have in the book a broad female reproduction, embryo development,
coverage on genome sciences and bio- and reproductive aging.
technologies that can help address and
understand various aspects of fertility and Part II of this book possesses five chapters
infertility in domestic animals. that target transcriptomics and physiologi-
cal genomics of reproduction, which link
The book consists of three main parts. genes to physiology and pathways critical
Part I has seven chapters that focus on for reproductive success. Chapter 8 deals
genome resources and quantitative genom- with transcriptomics of ovarian tissues
ics of reproduction. Chapter 1 demonstrates involved in follicular growth and develop-
genome resources specifically available to ment, luteinization of the dominant follicle,
livestock species, such as well-characterized corpus luteum regression, oocyte matura-
genome maps, whole genome and cDNA tion, and oocyte competence. Chapter 9
sequences, expression arrays, and high- focuses on transcriptomics related to differ-
density genetic marker chips. Chapter 2 ent preimplantation development stages and
defines the female reproductive phenotypes systems. Chapter 10 focuses on the genom-
and updates the genes/quantitative trait ics endometrial responses to conceptus


xvi Preface

signals mediating corpus luteum rescue. focuses on the mechanistic connections
Chapter 11 reviews hormones and peptides, between nutrient flux and reproductive
and transcriptomics involved in placental processes with emphasis on nutritional
development. Chapter 12 targets cellular, physiology of pregnancy and lactation and
molecular, and genomic mechanisms regu- demonstrates how nutrigenetics and nutrig-
lating testis function in livestock with enomics approaches can improve fertility,
emphasis on transcriptomics of the testis in pregnancy, and lactation.
bulls and reproductive genomics in boars.
This book is for researchers, instructors,
Part III has six chapters that deal with extension experts, and students in animal,
genomics of reproductive biotechnology and veterinary, and biomedical sciences who
their applications. Chapter 13 discusses the are interested in quantitative genomics,
importance of nuclear reprogramming during physiological genomics, mitochondriomics,
somatic cell nuclear transfer and its implica- pathological genomics, epigenomics, nutrig-
tions for normal fetal and placental develop- enomics, evolutionary genomics, and pro-
ment. Chapter 14 describes how to use teomics of reproduction. The 37 contributors
immunocontraception and immunosteril- to the book are all internationally recognized
ization as methods of fertility control in experts in their field, and they represent 15
animals. Chapter 15 deals with the structure different institutions from seven different
and properties of seminal plasma proteins countries. We thank them for their contri-
and their potential roles in fertilization butions to this first book on reproductive
affecting the oviductal reservoir, and as genomics of domestic animals. Support from
capacitation modulators, gamete interaction both families has been essential for us to
enhancers, and enzyme inhibitors. Chapter finish the book project, and we are grateful for
16 reports the state of knowledge on sex dif- their patience. Thanks also to Justin Jeffryes,
ferentiation in domestic mammals and sex Susan Engelken, Shelby Allen, the Wiley-
determination in nonmammal domestic Blackwell publishing team, and the team at
species. Chapter 17 addresses the disruption Toppan Best-set Premedia for their extra care
of the reproductive endocrine systems and and patience in publishing the book.
the mechanisms of action of endocrine dis-
rupting chemicals that exert hormone-like Zhihua Jiang
activity in humans and animals. Chapter 18 Troy L. Ott

Reproductive Genomics in
Domestic Animals

Part I

Quantitative Genomics of


Reproductive Genomics: Genome, Transcriptome,
and Proteome Resources

Noelle E. Cockett

1.1 Introduction and from hundreds of genetic markers to
tens of thousands markers—all assayable in
Genomic resources, tools, and technologies a few weeks to months as opposed to years.
that can be applied to studies in livestock
species, including investigations related to These resources and technologies can be
reproduction, have been under development combined in innovative ways to advance
for the last decade. While many of the two areas of research on reproductive traits,
genomic approaches were originally devel- specifically the identification of genes or
oped for use in humans or laboratory model genetic regions influencing phenotypes and
animals, they have been successfully applied the characterization of expression of genes
to studies in livestock. There are now a that are associated with traits.
myriad of resources specific to livestock
species, such as well-characterized genome 1.2 Discovery of underlying
maps, high-resolution genome, and comple- genetic influences
mentary DNA (cDNA) sequences, expres-
sion arrays, and high-density genetic marker The first area of interest for researchers
chips. In addition, there is an explosion studying reproductive traits is the character-
of high-throughput technology that will ization of genetic variation among animals
enhance these investigations, increasing the or populations underlying a phenotypic
scope and accuracy of the results beyond trait, leading to the identification of the
anything that was imagined just 5 years ago. genetic cause of the phenotype. Two general
These technologies advance studies of single approaches have been successfully used over
gene expression to full gene networks, from the last 10–15 years, with a third approach
single gene sequences to whole genomes, now on the horizon. In the first approach,


6 Quantitative Genomics of Reproduction

polymorphisms in a candidate gene likely to animals are detected by whether or not a
be involved in the phenotype are tested for restriction enzyme cuts, resulting in differ-
associations with different manifestations or ent-sized fragments. The genetic differences
phenotypes of the trait. The candidate genes are usually due to an SNP within the restric-
are selected for analysis based on an under- tion enzyme recognition site, although there
standing of trait physiology and/or because might be genetic differences due to inser-
of their involvement in similar traits in tions/deletions (in/del) in the gene, which
other species. In the second approach, genetic will also result in fragment size differences,
markers are analyzed for linkage with the although there is no variation in the restric-
phenotype using pedigrees of animals segre- tion enzyme recognition site. Animals are
gating for the trait and the markers. This expected to have two alleles for every gene
analysis identifies genetic regions that except those on the X and Y chromosomes
contain associated genes. By testing addi- in males, so that the presence of one frag-
tional markers through the families, the ment on the electrophoresis gel would indi-
interval is narrowed so candidate genes can cate that an animal is homozygous for the
be selected. The third approach, referred to PCR-RFLP allele whereas the presence of
as whole genome associations, will soon be two different-sized fragments would suggest
possible for livestock species now that the that an animal is heterozygous. However, an
development of high-density single nucleo- animal might be misclassified as a homozy-
tide polymorphism (SNP) arrays are readily gote if there is a polymorphism in the PCR
available. However, the application of whole primer sequence, which prevents that allele
genome associations requires very large from being amplified and therefore, not
numbers of phenotyped animals, which is a detected on the electrophoresis gel—referred
limitation for most research projects. to as a “null” allele. A null allele will often
be detected when misparentages are rou-
1.2.1 Candidate gene associations tinely found for a marker system. An animal
might also be misclassified if another, non-
As mentioned, the candidate gene approach allelic form of the gene is amplified with
uses information of the trait to determine the PCR primers and digestion with the
likely candidates for the underlying gene(s). restriction enzyme results in a different-
The choice of the gene is strengthened by its sized fragment. A nonallelic form is revealed
involvement in comparable traits in other by sequencing the fragments contained
species or its location in a region previously within the electrophoretic bands, which is a
identified as containing a quantitative trait recommended step when establishing any
loci (QTL) with similar attributes. marker system.

In the past, polymorphisms in a candidate However, new technologies have sig-
gene were routinely detected by polymerase nificantly advanced our ability to identify
chain reaction—restriction fragment length SNPs and then explore multiple candidate
polymorphisms (PCR-RFLP), which involves genes at one time at a much lower cost/
steps of amplifying the gene, digesting the polymorphism than the PCR-RFLP method.
amplicon with a restriction enzyme, and The identification of SNPs within a gene
then using gel electrophoresis to separate or genetic region is now relatively easy. To
the resulting fragments. In the PCR-RFLP do this, the genomic DNA of key animals
technique, gene sequence differences among within a population is sequenced using high-

Genome, Transcriptome, and Proteome Resources 7

throughput automatic sequencing and then mine whether a custom-built SNP array is
compared with other sequences within the economical.
population or to sequences in publically
available databases. The later approach is Emerging technology is now allowing the
referred to as in silico SNP detection. detection of differences in copy number
Regardless of the approach, confidence of the variant (CNV) among animals. For some
SNP is dependent on the quality of the time, copy number variation has been associ-
sequence across the multiple sources of data. ated with diseases (McCarroll 2008; Schaschi
et al. 2009), while the ongoing analyses
Once an SNP is identified, the polymor- of livestock whole genome sequences
phism can be detected by establishing a has revealed the presence of CNV in multi-
PCR-RFLP assay. However, allele-specific ple gene systems involved with innate
PCR using allele-specific oligonucleotides immunity, including milk composition
(ASOs) is an emerging technique for detect- traits (Rijnkels et al. 2009; Tellam and
ing genetic variation created by the SNP Bovine Genome Sequencing and Analysis
(Saiki et al. 1986). The 3′ ends of the primers Consortium 2009). Detection of differences
used in the PCR amplification step of the among animals for genes that are known to
ASO technique are designed to include the be present in the genome in multiple copies
polymorphic site so that amplification of the is now possible using microarray technology
animal’s DNA is dependent on the absence (Baumbusch et al. 2008), with higher copy
or presence of the polymorphism within the number resulting in greater intensity for that
primer sequence. Allele-specific primers spot on the array.
can be combined into a single amplification
reaction and the presence of the specific Once the polymorphism is detected within
allele detected by the melting temperature a population, the genotypes are usually ana-
of the alleles (Papp et al. 2003; Wang et al. lyzed for association with the trait by com-
2005). Appropriate controls and design of the paring the trait means among the marker
primers (e.g., Strerath et al. 2007) are critical genotypes (Rocha et al. 1992). Appropriate
in the allele-specific amplification assay so statistical models are needed in order to
that absence of amplification is due to the account for additive, dominant, and epistatic
polymorphism and not because of technical effects. In addition, the selection of animals
problems. used in the analysis must be sufficiently
broad; otherwise, the marker alleles will
SNP arrays are an extension of the ASO merely serve as a trace of unique families,
method, but by spotting multiple ASOs onto particularly when one of the alleles is at a
a membrane or bead, multiple alleles or even very low frequency in the population and
multiple genetic markers can be assayed in present only in one family in the analysis.
a single run. Custom-built SNP chips spe- This situation can result in a spurious sig-
cific to a trait are usually designed in a 92-, nificant association, simply because the
384-, or 1534-SNP format. While the cost/ family differs for the trait and not because
SNP is lower for the SNP chip than with the the allele itself is associated.
PCR-RFLP or allele-specific amplification
techniques, the initial setup for the chip is The choice of the candidate gene(s) can be
substantially higher. Thus, the number of strengthened by its association with similar
SNPs that are tested and the number of traits in the same or other species. Possible
animals included in the analysis will deter- candidate genes can be found through litera-
ture searches using key words based on

8 Quantitative Genomics of Reproduction

Table 1.1 Websites containing genomics information in livestock species.

Species Website Information

Cattle QTL QTL
Goat Genome sequence
Horse Genome project
Sheep Genome project
Pig Genome project Genome sequence∼jillm/jill.htm Primary Web source Virtual sheep genome NCBI resources International Sheep Genome Consortium QTL Genome sequence Genome project

the trait physiology or through searches of traits is based on identifying and character-
databases devoted to genetic abnormalities. izing genetic variation that is found in
One such database for livestock traits is pedigrees of animals. This approach has
called Online Mendelian Inheritance in most commonly been done using linkage
Animals (OMIA; analysis, which examines the segregation
The OMIA database contains details on of marker alleles through animal families
genes, inherited disorders, and traits for a with known phenotypes (Nejati-Javaremi
large range of animals species, similar to and Smith 1995; Knott and Haley 2000; de
what is found within Online Mendelian in Koning et al. 2003) and subsequent refine-
Man (OMIM; ment of the genetic interval containing the
entrez?db=omim). trait locus (Riquet et al. 1999; Farnir et al.
2002). The data are analyzed to determine
There are also databases that describe the coinheritance of marker alleles with the
the location of QTLs for traits of interest in causative genetic mutation, presumably
livestock species (Table 1.1). Additional can- because they are closely located within the
didate genes can be identified by searching genome.
genetic sequences that lie within QTL inter-
vals and have involvement in the physiology Linkage mapping requires pedigrees with
of the trait, providing not only functional specific family structures; these pedigrees
evidence but also positional evidence for are most commonly reciprocal backcrosses
inclusion in the candidate gene analysis. or F2 crosses developed from lines or breeds
These genes are therefore referred to as of animals that significantly differ for the
“positional candidate genes” (see below). trait. The analysis can include families
within a single breed or line but the key
1.2.2 Analysis of genetic variation parents must be heterozygous for both the
markers and the trait in order for linkage to
The second approach for detecting genes or, be detected. As with the association analy-
more commonly, genetic regions involved in ses, appropriate statistical models are needed

Genome, Transcriptome, and Proteome Resources 9

to detect genetic mutations that are con- produced from a whole genome sequence
trolled by complex gene actions, such as the that has been assembled and annotated.
imprinted callipyge (Cockett et al. 1994, Assembled whole genome sequences are
1996) and IGF2 (Van Laere et al. 2003) loci. now publically available for cattle, swine,
The effects of these loci would not have and horses (Table 1.2). Millions of bases of
been detected without the appropriate sta- sequences can be accessed for the analysis of
tistical model (see Sandor and Georges 2008). genes, SNPs, regulatory features, and so on.
Comparisons across species, including non-
To perform a screen of markers across livestock species, are now possible using
the complete genome (i.e., a genome scan), “landmark” loci that anchor segments of the
markers are typically selected about one genome from species to species. International
every 10–20 centimorgans (cM). Because consortiums of experts have been organized
the typical mammalian genome is about for annotation of the sequences; for example,
3000 cM, around 150–300 markers are needed “the Horse Genome Project is a cooperative
for a genome scan. The availability of international effort by over 100 scientists in
genome-wide maps in livestock species 20 countries to define the genome, the DNA
provides the information needed to select sequence, of the domestic horse” (www.uky.
markers at appropriate intervals, which is edu/Ag/Horsemap/welcome.html). A wealth
dependent on the number of informative of knowledge from the analysis of these
offspring in the families and the genetic sequences is now being released.
variability in the trait. Several reviews on
conducting a genome scan and subsequent In addition, assembled whole genome
analysis are available, including Schwerin sequences can serve as the “reference”
(2001), Rocha et al. (2002), Andersson and for comparison of individual animal sequ-
Georges (2004), and Georges (2007). ences generated with state-of-the-art high-
throughput platforms such as ABI’s SOLiD™
1.2.3 Whole genome sequence (Carlsbad, CA), Roche 454 FLX Titanium™
(Branford, CT), and Illumina’s Solexa™ (San
Genetic markers for a genome scan are Diego, CA) systems. These technologies
usually selected from a genome map. The produce millions of reads of short sequences
most complete genome map for a species is (50–400 bases) in a single run at relatively

Table 1.2 Whole genome sequence assemblies in livestock species.1

Species Reference or website Sequenced animal Method Coverage
Cattle2 Baylor School of Medicine Hereford male L1 WGS 6.8X
Domino 99375 WGS 4X
Horse3 Broad Institute Thoroughbred female BAC by BAC tile path
Pig4 Sanger Institute Duroc female

1 As of February 1, 2009.
WGS, whole genome shotgun.

10 Quantitative Genomics of Reproduction

Table 1.3 Most recent published linkage maps in livestock species that
do not have a whole genome sequence.

Species Population No. of loci Reference

Goat INRA 307 Schibler et al. (1998a)
Deer Red deer × Pere 621 Slate et al. (2002)
David’s deer
Sheep IMF 1062 Maddox et al. (2001)

low cost ($10,000/10 Gb) from either single Table 1.4 Physical map in livestock species that
or pooled DNA samples. The sequences for do not have a whole genome sequence.
each run can then be compared back to the
reference genome sequences, allowing detec- Species No. of loci Reference
tion of genetic differences across animals.
River buffalo 388 Di Meo et al. (2008)
1.2.4 Linkage or genetic maps
Goat 202 Schibler et al. (1998a)
For those species without a whole genome
sequence, linkage and physical maps are Deer 59 Bonnet et al. (2001)
critical for the orientation of loci as well as
comparisons across species. Linkage maps Sheep 452 Di Meo et al. (2007)
usually contain a preponderance of highly
polymorphic anonymous markers, primarily somal fragment. Physical mapping is usually
microsatellites, and relatively few expressed done by in situ hybridization, somatic cell
genes, which have very limited genetic vari- hybrid analysis, or radiation hybrid (RH)
ability. Also, multiple linkage maps may mapping. Because a genetic variant within
exist for a species because different reference the locus is not necessary for physical
families were used to create the linkage mapping, these maps contain a relatively
maps. The various maps are often combined large number of expressed genes.
into a consensus linkage map, which is
anchored by common markers genotyped in One of the first reports assigning genes to
the different reference families (Table 1.3). physical locations within the genome was
The distances between loci on linkage maps performed by hybridizing a radioactively
are given in centimorgans (cM), with 1 cM labeled probe to a spread of metaphase chro-
representing 1% recombination between mosomes in a technique referred to as in situ
two loci. hybridization. A significant adaptation of
this method entailed labeling the probe with
1.2.5 Physical map assignments fluorophores, leading to the moniker of
fluorescent in situ hybridization (FISH).
In addition to the linkage maps, physical Hundreds of genes and genetic markers have
maps exist for each species. These maps are now been assigned to specific chromosomes
created by direct assignment of a gene or in livestock species using in situ hybridiza-
marker to an intact chromosome or chromo- tion techniques (Table 1.4).

Chromosome painting is an approach for
evaluating the conservation of chromosomal
segments across species. In this technique,
chromosomes of one species are fluores-
cently labeled and hybridized to metaphase

Genome, Transcriptome, and Proteome Resources 11

chromosomes of another species. Reciprocal RH mapping provides a higher level of
chromosome painting has been performed resolution of gene location and gene order
between humans and farm animal species than those produced by in situ hybridization
including pigs, cattle, sheep, and horses and somatic cell hybrids. This technique is
(Chowdhary et al. 1996; Chowdhary and based on detecting the presence/absence of
Raudsepp 2001). These studies have defined loci that are contained on fragments of DNA
the borders of conserved syntenies among maintained in a panel of hybrid clones,
the species, but because of insufficient reso- similar to the somatic cell hybrid approach,
lution, they do not allow the study of gene but the rodent cells are fused with target
order. cells that have been irradiated. Varying the
radiation dose on the target species cells will
Somatic cell hybrid panels were used create different-sized fragments and there-
frequently in the 1970s–1990s to assign fore, vary the resolution between two loci.
genes to specific chromosomes in livestock, The higher the radiation dose, the smaller
but this method is now used much less the fragments and the better resolution
frequently than other physical mapping between tightly linked loci. Thus, high rad
approaches that have a much better resolu- panels are suitable when fine-mapping
tion. A somatic cell hybrid panel is gener- markers within a specific region, but large
ated by fusing cells of the target species with numbers of random markers must be
cells of a rodent species, such as hamsters. screened in order to detect linkage of loci
The rodent cells randomly eject chromo- across the genome. Whole genome maps,
somes of the target species, until at some which do not require a saturation of markers,
point the cells are immortalized, leaving a are best done with a lower rad RH panel.
complement of target chromosomes that is RH panels have been generated for several
the signature of each somatic cell clone. livestock species (Table 1.5) and used for
DNA is harvested from each of the clones in generating chromosome and whole genome
the panel (usually around 30 clones) and RH maps.
then amplified with primers specific to a
gene or genetic marker. Those clones that Distances on an RH map are measured in
contain a piece of the chromosome harbor- centiRay (cR), with a distance of 1 cRrad
ing the gene amplify with the primers, as between two markers corresponding to a 1%
well as other “concordant” or linked genes. frequency of breakage between these two
The somatic cell hybrid approach can be markers after exposure to a specific radia-
used to identify genes found within long seg- tion (rad) dose. Statistical programs have
ments of the chromosome, but the order of been developed to analyze the RH panel data
the genes along the chromosomal segment to give the most likely order based on the
cannot be determined. least number of break points (e.g., Boehnke

Table 1.5 Radiation hybrid maps in livestock species that do not have a whole genome sequence.

Species RH panel Rad No. of loci1 Reference

River buffalo BBURH5000 5,000 3,990 Amaral et al. (2009)
Sheep USUoRH5000 5,000 2,300 Wu et al. (2007, 2008, 2009)
INRA 12,000 Laurent et al. (2007)

1 As of February 1, 2009.

12 Quantitative Genomics of Reproduction

1992; Lange et al. 1995; Lunetta et al. 1996). comparisons of genes contained within
A measure of relative likelihood of one order common genetic regions. These compara-
versus another is given for each map devel- tive maps can also be used to localize a
oped with the RH data. single gene across multiple species.

Distance between loci on an RH map is Numerous causative mutations for single
directly proportional to physical distance, gene traits in livestock have been identified.
measured as the frequency of retention of a In contrast, although numerous QTL have
given pair of markers. The more times two been identified for economically important
loci are retained together, the closer they traits in livestock (see Table 1.1), very few
are found on a chromosome. Retention fre- of the causative mutations for QTL (referred
quency is calculated as the percentage of to as quantitative trait nucleotides or QTNs)
clones that retain a given marker and is have been characterized. There are numer-
usually between 18% and 30% for whole ous challenges in identifying the mutation
genome RH panels. for a quantitative trait, including a limita-
tion on animals and/or families suitable
1.2.6 Positional candidate genes for narrowing the QTL interval, an often
unwieldy number of candidate genes and
Once the location of a trait within the mutations within the genetic region, diffi-
genome is determined because of linkage to culty in estimating the interactions of other
previously mapped genetic markers, possi- QTL on the trait, and technological and bio-
ble candidate genes controlling the trait logical limitations when establishing the
can be inferred because of their proximity to functionality of the candidate mutations.
the linked markers. A typical genome scan However, step-by-step approaches for estab-
usually assigns the trait locus or QTL to a lishing the causality of mutations involved
∼20-cM interval, which can contain hun- in QTL have been proposed (Grisart et al.
dreds of genes. However, it is not necessary 2001, 2004; Andersson and Georges 2004; de
to have map locations of all possible genes Koning et al. 2007; Georges 2007; Ron and
in a single livestock species. Rather, a subset Weller 2007; Sellner et al. 2007).
of genes that are mapped in well-studied
species, such as humans and mice, are also 1.2.7 Analysis of genetic fragments
mapped in farm animals; these genes serve
as “anchors” across the comparative maps Several large insert libraries, including bac-
and allow inference of the locations of other terial artificial chromosome (BAC), yeast
genes within a region, based on what is artificial chromosome (YAC), and fosmid
known within the well-mapped species (Burt vectors, exist for each livestock species, with
2002). the vast majority being BAC libraries (Table
1.6). Most of the BAC libraries for livestock
Positional candidate genes can be identi- species have been prepared by Pieter de
fied for traits mapped by linkage analysis Jong’s group at BACPAC Resources Center
once markers used in the linkage analysis ( and contain inserts with
are located on the comparative map, either an average size of 90–200 Mb. These libraries
by direct mapping or because a gene linked can be screened by PCR amplification of
to the marker is placed on the comparative plate, row, and column pools or by probe
map. Several online comparative map data- hybridization of high-density filters. The
bases have been established, which allow

Genome, Transcriptome, and Proteome Resources 13

Table 1.6 Large insert libraries in livestock species.

Species Library Genome coverage Reference for BAC map

Cattle CHORI-240 10.7X Snelling et al. (2007)
RPCI-42 10X
Horse CHORI-241 11.8X Gustafson et al. (2003)
Sheep CHORI-243 12X Dalrymple et al. (2007)
Goat 6:15 translocation 3.3X Schibler et al. (1998b)
Pig CHORI-242 11.4X Humphray et al. (2007)
RPCI-44 10.2X

Table 1.7 High-density SNP chips in livestock species.

Species SNP chip Reference

Cattle Affymetrix 25K GeneChip Khatkar et al. (2007)
Illumina 50K BeadChip Van Tassell et al. (2008)
Horse Illumina 50K BeadChip Chowdhary and Raudsepp (2008)
Sheep Illumina 50K BeadChip Kijas J. et al. (unpublished)
Pig Illumina 60K BeadChip Schook L. et al. (unpublished)

screening provides an exact clone address 2008). Assuming that the trait allele of inter-
that contains the DNA sequence or gene of est has descended from one or a few ances-
interest, and the clones can be purchased for tral chromosomes, animals displaying the
about $20 from the BACPAC Resources trait of interest will possess the ancestral
Center. haplotype that contains the allele of interest
surrounded by closely linked marker alleles.
Large overlapping segments of DNA called These haplotypes may be fixed in a popula-
contigs can be generated by chromosome tion or breed because of natural or artificial
walking. To do this, the ends of isolated selection for the favorable allele, which is
clones are sequenced, and then primers known as a “selective sweep” (Kim and
designed from the new sequence and used to Nielsen 2004; Pollinger et al. 2005; Voight
screen the library in subsequent rounds. et al. 2006; McVean 2007). The trait haplo-
Sequential screenings will provide overlap- type will be identified from all the wild-type
ping clones that can be pieced together into haplotypes through the GWA analysis.
a single continuous fragment. Because very large numbers of markers are
used in the analysis (around one marker
1.2.8 Whole genome association every 100–500 kb), the resulting LD maps
are typically of higher resolution than
In contrast to genetic linkage methods that genetic linkage scans, which helps to limit
use pedigrees segregating for the trait, the size of the interval that contains posi-
genome-wide association (GWA) mapping is tional candidate genes.
an approach that tests for allelic association
at the population level through an analysis High-density SNP chips are now under
of linkage disequilibrium (LD), using large construction for all major livestock species
samples of unrelated individuals (Meuwissen (Table 1.7). These chips usually include
et al. 2001; Amos 2007; McCarthy et al. between 30,000 and 60,000 SNPs, suitable for

14 Quantitative Genomics of Reproduction

large-scale genotyping applications such which can then be examined in a variety
as the GWA analysis. The availability of of ways. Because RNases, enzymes that
whole genome high-density SNP chips at a break down RNA, are quite ubiquitous and
relatively low price per animal ($150–$300) difficult to degrade, care must be taken
means that GWA analyses in livestock to preserve the tissue without delay after
become a more common approach for local- collection, such as snap freezing the tissue
izing a trait locus within the genome. An in liquid nitrogen or immersing the tissue
application of GWA has recently been illus- in a preservative/RNase inhibitor such as
trated by the fine mapping of five recessive RNALater™ (QIAGEN, Inc., Valencia, CA).
disorders in cattle using the high-density After extracting the RNA from the tissue,
bovine BeadChip (Charlier et al. 2008) with another enzyme, reverse transcriptase, con-
less than 20 affected animals and 20 controls. verts the RNA into the first strand of cDNA
However, the number of unrelated animals followed by second strand synthesis using
needed for mapping quantitative traits is pre- DNA polymerase. The cDNA does not
dicted to be greater than 1000 (McCarthy et directly correspond to genomic DNA because
al. 2008; Orr and Chanock 2008; Tian et al. intronic segments have been spliced out
2008). Unfortunately, very few livestock when the RNA molecule was produced and
populations of that size currently exist. only the exonic sequences are contained
with the cDNA strand. The cDNA mixture
1.3 Characterization of can be analyzed for transcript content using
gene expression various techniques or used in the creation of
a cDNA library by cloning into vectors,
Researchers are often interested in charac- usually plasmids, which are then trans-
terizing the expression of genes in a specific formed into competent Escherichia coli
tissue at a specific time or under a specific cells. Replication of the host cells in the
set of circumstances. The expression of cDNA library results in the replication of
these genes can be “captured” by examining the plasmid as well as the unique cDNA
the messenger RNA (mRNA) within the sequence contained within the plasmid.
tissue sample. The abundance of a particular
mRNA within a tissue can now be measured Numerous kits for synthesizing cDNA
with relative ease using techniques devel- from mRNA followed by analysis are now
oped within the last decade, such as Northern commercially available. Kits for the con-
blots, and newly developed techniques, such struction of cDNA libraries are also avail-
as real-time PCR. It is also possible to deter- able, or library construction can be contracted
mine a “profile” of mRNAs within a tissue for a relatively modest price or a library
using an analysis system such as expression made from a particular tissue can be pur-
microarrays or serial analysis of gene expres- chased from a commercial company. The
sion (SAGE). bacterial library can be gridded onto filters
and then screened for a particular gene by
hybridization with a probe.

1.3.1 Synthesis and analysis of cDNA 1.3.2 Analysis of gene expression

The array of mRNAs found within a tissue Quantitative real-time reverse transcription
is often converted into a cDNA library, PCR (qRT-PCR; see Logan et al. 2009) is a

Genome, Transcriptome, and Proteome Resources 15

method for measuring levels of specific on the array in higher density and are cheaper
mRNA transcripts within a sample. After to synthesize. Oligonucleotide arrays are
the RNA sample is treated with reverse tran- usually preferred to the cDNA arrays because
scriptase, the resulting cDNA is amplified of more uniform hybridization and ease
in a PCR reaction using primers specific to of probe synthesis (Barrett and Kawasaki
a transcript and the amount of transcript 2003; Hardiman 2004). Detection of a spe-
quantified in “real time” after each amplifi- cific transcript within a sample is based on
cation cycle. hybridization to the probes on the array.
Annotation of the probes on the arrays is
Detection of the transcripts is usually a key consideration for their usefulness.
done with fluorescent dyes that intercalate Statistical analysis of the data is challenging
with double-stranded DNA, although non- and requires appropriate controls, normal-
specific binding can occur, which decreases ization of the signals, and adjustments for
the accuracy of the quantification. Another multiple comparisons. Significant results
method of detection in qRT-PCR uses DNA from an expression array experiment are
oligonucleotide probes specific to the tran- often verified by qRT-PCR.
script that fluoresce when hybridized with a
cDNA molecule. This method is more accu- SAGE allows whole genome analysis of
rate than the double-stranded dyes, but the gene expression (i.e., mRNA) within a sample
synthesis of fluorescent reporter probes is (Velculescu et al. 1995, 1997). Based on the
expensive. concept that 10–14 bp of sequence provides
“sufficient information to uniquely identify
Relative concentration of the transcript is a transcript” within a sequence database,
determined in the qRT-PCR by plotting flu- “quantification of the number of times a
orescence (dependent on the number of particular tag is observed provides the expres-
copies of the transcript within the sample) sion level of the corresponding transcript”
against cycle number on a logarithmic scale. (
The quantity of a control, such as a house- Previously unreported genes can also be
keeping gene, is also measured on each detected through the generation of tags that
sample so as to normalize for possible varia- are not contained within the databases.
tion in the amount and quality of RNA Subsequent adaptations of SAGE, such as
between different samples, with the assump- SuperSAGE (Matsumura et al. 2005), allow
tion that the expression of the control is precise annotation of existing and new genes
similar across all samples. because of an increased tag length of 25–
27 bp. However, SAGE is relatively much
“Global” expression of genes within a more expensive than DNA microarrays, so
sample is commonly analyzed using expres- large-scale projects are typically not per-
sion microarrays, which allow simultaneous formed with SAGE.
analysis of hundreds to thousands of genes.
Probes spotted on the arrays were originally 1.3.3 cDNA libraries and reproductive
cDNAs but more recently, expression arrays transcriptomes
contain oligonucleotides, usually in the
range of 25–75 mers, designed from cDNA To date, there are at least 270 publically
sequences. The longer the oligonucleotide, available cDNA libraries that were derived
the more specific the detection, especially in from different reproductive tissues/organs in
cross-species experiments (Walker et al.
2006), but the shorter probes can be spotted

16 Quantitative Genomics of Reproduction

Table 1.8 cDNA libraries and EST sequences for reproductive tissues/organs in livestock species.

Tissue/organ Cattle Swine Sheep

No. of libraries No. of ESTs No. of libraries No. of ESTs No. of libraries No. of ESTs

Embryo 27 62,951 14 89,916 — —
Fetus — —
Mammary 14 72,914 16 6,468 1 2,309
Ovary 6 2,899
Oviduct 56 65,227 3 16,656 — —
Pituitary — —
Placenta 17 13,813 28 75,026 — —
Testes — —
Uterus 2 70 2 3,556 1 2,722

Total 5 2,102 7 12,404 8 7,930

10 23,665 6 21,307

7 15,033 11 42,494

12 31,380 15 43,392

150 287,155 102 311,219

cattle, swine, and sheep (Table 1.8). The 1.4 Resources for protein analysis
library names, tissue/organ/cell line sources,
physiological or reproductive stages, and A unique complement of proteins is present
contributors can be retrieved from either the in the cells of an organism at any one time
GenBank database at NCBI (www.ncbi.nlm. under any one condition. This complement or the Gene Index database at of proteins does not necessarily match to the
Harvard University (compbio.dfci.harvard. complement of mRNA transcripts within
edu/tgi/). As seen in Table 1.8, cattle have the cells because of posttranslational modi-
a slight edge over swine in the number of fications, splicing variants, and protein
constructed libraries (105 and 102, respec- and RNA degradation. A recently defined
tively), but swine lead cattle in the number area of research called proteomics encom-
of expressed sequence tags (ESTs) that passes large-scale studies of proteins, includ-
have been placed in the public databases ing their structure, function, and quantity
(311,219 and 287,155, respectively). To (Anderson and Anderson 1998; Blackstock
date, only eight libraries have been estab- and Weir 1999). Two-dimensional (2D) gel
lished in sheep, and less than 8000 ovine electrophoresis is a well-established method
ESTs for reproductive tissues/organs have commonly used to analyze proteins (Berth
been released. et al. 2007), although there are challenges
in automatic analysis software. Technologies
These resources have been widely used in that allow high-throughput analysis of
the survey of reproductive transcriptomes, proteins within a tissue are now available,
identification of some breed- and develop- such as high-performance chromatography
mental-stage-specific genes or gene clusters, and mass spectrometry, but these approaches
and investigations of the genetic and physi- require highly specialized equipment.
ological mechanisms underlying reproduc-
tion quantitative traits in livestock species. Because of increasing emphasis on systems
In addition, comparisons of livestock ESTs biology, databases have been created that
with sequences from other species have present whole biological systems of intercon-
served as a valuable resource for compara- nected proteins, with access to underlying
tive map development. genes, their sequences, and the background

Genome, Transcriptome, and Proteome Resources 17

studies with a click of a mouse (see www. Amos, C.I. 2007. Successful design and and www. conduct of genome-wide association studies. Human Molecular Genetics 16:
querying). R220–R225.

1.5 Future research directions Anderson, N.L. and Anderson, N.G. 1998.
Proteome and proteomics: New technolo-
Genomic resources are now available for all gies, new concepts, and new words.
major livestock species. These resources Electrophoresis 19: 1853–1861.
will allow researchers to identify regions
within the genome that influence reproduc- Andersson, L. and Georges, M. 2004.
tive traits with relative ease. While the Domestic-animal genomics: Deciphering
pursuit of the causative mutation control- the genetics of complex traits. Nature
ling a quantitative trait may be complicated, Reviews in Genetics 5: 202–212.
combining knowledge from several lines of
investigations should lead to the successful Barrett, J.C. and Kawasaki, E.S. 2003.
identification of the responsible gene. There Microarrays: The use of oligonucleotides
are also multiple approaches for estimating and cDNA for the analysis of gene expres-
gene expression in livestock species at sion. Drug Discovery Today 8: 134–
both the single and whole genome levels. 141.
However, resources for the study of proteins
are much less developed in livestock species, Baumbusch, L.O., Aarøe, J., Johansen, F.E.,
and therefore, researchers will need to Hicks, J., Sun, H., Bruhn, L., Gunderson,
exploit available information from humans K., Naume, B., Kristensen, V.N., Liestøl,
and biomedical animal models. K., Børresen-Dale, A.L., and Lingjaerde,
O.C. 2008 Comparison of the Agilent,
References ROMA/NimbleGen and Illumina plat-
forms for classification of copy number
Amaral, M.E., Grant, J.R., Riggs, P.K., alterations in human breast tumors. BMC
Stafuzza, N.B., Filho, E.A., Goldammer, Genomics 9: 379.
T., Weikard, R., Brunner, R.M., Kochan,
K.J., Greco, A.J., Jeong, J., Cai, Z., Lin, G., Berth, M., Moser, F.M., and Kolbe, M. 2007.
Prasad, A., Kumar, S., Saradhi, G.P., The state of the art in the analysis of two-
Mathew, B., Kumar, M.A., Miziara, M.N., dimensional gel electrophoresis images.
Mariani, P., Caetano, A.R., Galvão, S.R., Applied Microbiology and Biotechnology
Tantia, M.S., Vijh, R.K., Mishra, B., Kumar, 76: 1223–1243.
S.T., Pelai, V.A., Santana, A.M., Fornitano,
L.C., Jones, B.C., Tonhati, H., Moore, S., Blackstock, W.P. and Weir, M.P. 1999.
Stothard, P., and Womack, J.E. 2009. A Proteomics: Quantitative and physical
first generation whole genome RH map mapping of cellular proteins. Trends in
of the river buffalo with comparison to Biotechnology 17: 121–127.
domestic cattle. BMC Genomics 9: 631.
Boehnke, M. 1992. Multipoint analysis for
radiation hybrid mapping. Annuals in
Medicine 24: 383–386.

Bonnet, A., Thevenon, S., Claro, F., Gautier,
M., and Hayes, H. 2001. Cytogenetic
comparison between Vietnamese sika
deer and cattle: R-banded karyotypes and
FISH mapping. Chromosome Research 9:

18 Quantitative Genomics of Reproduction

Burt, D.W. 2002. Comparative mapping in Barris, W., Zhao, S., Shetty, J., Maddox,
farm animals. Briefings in Functional J.F., O’Grady, M., Nicholas, F., Crawford,
Genomics and Proteomics 1: 159–168. A.M., Smith, T., de Jong, P., McEwan,
J.C., Oddy, V.H., and Cockett, N.E. 2007.
Charlier, C., Coppieters, W., Rollin, F., Constructing the virtual sheep genome.
Desmecht, D., Agerholm, J.S., Cambisano, Genome Research 8: R152.
N., Carta, E., Dardano, S., Dive, M., de Koning, D.J., Pong-Wong, R., Varona, L.,
Fasquelle, C., Frennet, J.C., Hanset, R., Evans, G.J., Giuffra, E., Sanchez, A.,
Hubin, X., Jorgensen, C., Karim, L., Kent, Plastow, G., Noguera, J.L., Andersson, L.,
M., Harvey, K., Pearce, B.R., Simon, P., and Haley, C.S. 2003. Full pedigree quan-
Tama, N., Nie, H., Vandeputte, S., Lien, titative trait locus analysis in commercial
S., Longeri, M., Fredholm, M., Harvey, pigs using variance components. Journal
R.J., and Georges, M. 2008. Highly effec- of Animal Science 81: 2155–2163.
tive SNP-based association mapping and de Koning, D.J., Archibald, A., and Haley,
management of recessive defects in live- C.S. 2007. Livestock genomics: Bridging
stock. Nature Genetics 40: 449–454. the gap between mice and men. Trends in
Biotechnology 25: 483–489.
Chowdhary, B.P., Fronicke, L., Gustavsson, Di Meo, G.P., Perucatti, A., Floriot, S.,
I., and Scherthan, H. 1996. Comparative Hayes, H., Schibler, L., Incarnato, D., Di
analysis of the cattle and human genomes: Berardino, D., Williams, J., Cribiu, E.,
detection of ZOO-FISH and gene map- Eggen, A., and Iannuzzi, L. 2008. An
ping-based chromosomal homologies. extended river buffalo (Bubalus bubalis,
Mammalian Genome 7: 297–302. 2n = 50) cytogenetic map: Assignment of
68 autosomal loci by FISH-mapping and
Chowdhary, B.P. and Raudsepp, T. 2001. R-banding and comparison with human
Chromosome painting in farm, pet and chromosomes. Chromosome Research
wild animal species. Methods in Cell 16: 827–837.
Science 23: 37–55. Di Meo, G.P., Perucatti, A., Floriot, S., Hayes,
H., Schibler, L., Rullo, R., Incarnato, D.,
Chowdhary, B.P. and Raudsepp, T. Ferretti, L., Cockett, N., Cribuiu, E.,
2008. The horse genome derby: Racing Williams, J.L., Eggen, A., and Iannuzzi, L.
from map to whole genome sequence. 2007. An advanced sheep (Ovis aries, 2n
Chromosome Research 16: 109–127. = 54) cytogenetic map and assignment of
88 new autosomal loci by fluorescence in
Cockett, N.E., Jackson, S.P., Shay, T.L., situ hybridization and R-banding. Animal
Nielsen, D., Steele, M.R., Green, R.D., Genetics 38: 233–240.
and Georges, M. 1994. Chromosomal Farnir, F., Grisart, B., Coppieters, W., Riquet,
localization of the callipyge gene in sheep J., Berzi, P., Cambisano, N., Karim, L.,
(Ovis aries) using bovine DNA markers. Mni, M., Moisio, S., Simon, P., Wagenaar,
Proceedings of the National Academy of D., Vilkki, J., and Georges, M. 2002.
Sciences of the United States of America Simultaneous mining of linkage and
91: 3019–3023. linkage disequilibrium to fine map quan-
titative trait loci in outbred half-sib
Cockett, N.E, Jackson, S.P., Shay, T.L., pedigrees: Revisiting the location of a
Farnir, F., Berghmans, S., Snowder, G.,
Nielsen, D., and Georges, M. 1996. Polar
overdominance at the ovine callipyge
locus. Science 273: 236–238.

Dalrymple, B.P., Kirkness, E.F., Nefodov,
M., McWilliam, S., Ratnakumar, A.,

Genome, Transcriptome, and Proteome Resources 19

quantitative trait locus with major effect 2007. A high utility integrated map of the
on milk production on bovine chromo- pig genome. Genome Biology 8: R139.
some 14. Genetics 161: 275–287. Khatkar, M.S., Zenger, K.R., Hobbs, M.,
Georges, M. 2007. Mapping, fine mapping, Hawken, R.J., Cavanagh, J.A., Barris,
and molecular dissection of quantitative W., McClintock, A.E., McClintock, S.,
trait loci in domestic animals. Annual Thomson, P.C., Tier, B., Nicholas, F.W.,
Reviews of Genomics and Human and Raadsma, H.W. 2007. A primary
Genetics 8: 131–162. assembly of a bovine haplotype block map
Grisart, B., Coppieters, W., Farnir, F., Karim, based on a 15,036-single-nucleotide poly-
L., Ford, C., Berzi, P., Cambisano, N., morphism panel genotyped in Holstein-
Mni, M., Reid, S., Simon, P., Spelman, R., Friesian cattle. Genetics 176: 763–772.
Georges, M., and Snell, R. 2001. Positional Kim, Y. and Nielsen, R. 2004. Linkage dis-
candidate cloning of a QTL in dairy cattle: equilibrium as a signature of selective
Identification of a missense mutation in sweeps. Genetics 167: 1513–1524.
the bovine DGAT1 gene with major effect Knott, S.A. and Haley, C.S. 2000. Multitrait
on milk yield and composition. Genome least squares for quantitative trait loci
Research 12: 222–231. detection. Genetics 156: 899–911.
Grisart, B., Farnir, F., Karim, L., Cambisano, Lange, K., Boehnke, M., Cox, D.R., and
N., Kim, J.J., Kvasz, A., Mni, M., Simon, Lunetta, K.L. 1995. Statistical methods
P., Frère, J.M., Coppieters, W., and for polyploid radiation hybrid mapping.
Georges, M. 2004. Genetic and functional Genome Research 5: 136–150.
confirmation of the causality of the Laurent, P., Schibler, L., Vaiman, A., Laubier,
DGAT1 K232A quantitative trait nucleo- J., Delcros, C., Cosseddu, G., Vaiman, D.,
tide in affecting milk yield and com- Cribiu, E.P., and Yerle, M. 2007. A 12,000-
position. Proceedings of the National rad whole-genome radiation hybrid panel
Academy of Sciences in the United States in sheep: Application to the study of the
of America 101: 2398–2403. ovine chromosome 18 region containing
Gustafson, A.L., Tallmadge, R.L., Ramlachan, a QTL for scrapie susceptibility. Animal
N., Miller, D., Bird, H., Antczak, D.F., Genetics 38: 358–363.
Raudsepp, T., Chowdhary, B.P., and Logan, J., Edwards, K., and Saunders, N. (eds.).
Skow, L.C. 2003. An ordered BAC contig 2009. Real-Time PCR: Current Techno-
map of the equine major histocompatibil- logy and Applications. Wymondham, UK:
ity complex. Cytogenetics and Genome Caister Academic Press.
Research 102: 189–195. Lunetta, K.L., Boehnke, M., Lange, L., and
Hardiman, G. 2004. Microarray platforms— Cox, D.R. 1996. Selected locus and mul-
Comparisons and contrasts. Pharmaco- tiple panel models for radiation hybrid
genomics 5: 487–502. mapping. American Journal of Human
Humphray, S.J., Scott, C.E., Clark, R., Genetics 59: 717–725.
Marron, B., Bender, C., Camm, N., Davis, Maddox, J.F., Davies, K.P., Crawford, A.M.,
J., Jenks, A., Noon, A., Patel, M., Sehra, Hulme, D.J., Vaiman, D., Cribiu, E.P.,
H., Yang, F., Rogatcheva, M.B., Milan, D., Freking, B.A., Beh, K.J., Cockett, N.E.,
Chardon, P., Rohrer, G., Nonneman, D., Kang, N., Riffkin, C.D., Drinkwater, R.,
de Jong, P., Meyers, S.N., Archibald, A., Moore, S.S., Dodds, K., Lumsden, J.K.,
Beever, J.E., Schook, L.B., and Rogers, J. Adelson, D., Birkin, H., Broom, J.E.,

20 Quantitative Genomics of Reproduction

Buitkamp, J., Cambridge, E., Cushwa, Wayne, R.K. 2005. Selective sweep
W.T., Gerard, G., Galloway, S., Harrison, mapping of genes with large phenotypic
B., Hawken, R.J., Hiendleder, S., Henry, effects. Genome Research 15: 1809–
H., Medrano, J., Paterson, K., Phua, S.H., 1819.
Schibler, L., Stone, R.T., and van Hest, B. Rijnkels, M., Lemay, D.G., Barris, W.C.,
2001. An enhanced linkage map of the Casey, T.M., German, J.B., Hinrichs, A.S.,
sheep genome comprising more than 1000 Kriventseva, E.V., Lynn, D.J., Martin,
loci. Genome Research 11: 1275–1289. W.F., Maqbool, N.J., Medrano, J.F.,
Matsumura, H., Ito, A., Saitoh, H., Winter, Molenaar, A.J., Neville, M.C., Pollard,
P., Kahl G., Reuter, M., Kruger, D.H., and K.S., Rincon, G., Zdobnov, E.M., Tellam,
Terauchi, R. 2005. SuperSAGE. Cellular R.L., and Bovine Genome Sequencing and
Microbiology 7: 11–18. Analysis Consortium. 2009. Milking the
McCarroll, S.A. 2008. Extending genome- bovine genome: insights in role of milk
wide associations studies to copy-number and variation of milk composition. Plant
variation. Human Molecular Genetics 17: and Animal Genomes XVII Conference,
R135–R142. W088, January 10–14, San Diego, CA.
McCarthy, M.I., Abecasis, G.R., Cardon, Riquet, J., Coppieters, W., Cambisano, N.,
L.R., Goldstein, D.B., Little, J., Ioannidis, Arranz, J.J., Berzi, P., Davis, S.K., Grisart,
J.P., and Hirschhorn, J.N. 2008. Genome- B., Farnir, F., Karim, L., Mni, M., Simon,
wide association studies for complex P., Taylor, J.F., Vanmanshoven, P.,
traits: Consensus, uncertainty and chal- Wagenaar, D., Womack, J.E., and Georges,
lenges. Nature Reviews Genetics 9: M. 1999. Fine-mapping of quantitative
356–369. trait loci by identify by descent in outbred
McVean, G. 2007. The structure of linkage populations: Application to milk pro-
disequilibrium around a selective sweep. duction in dairy cattle. Proceedings of
Genetics 175: 1395–1406. the National Academy of Sciences of the
Meuwissen, T.H., Hayes, B.J., and Goddard, United States of America 96: 9252–
M.E. 2001. Prediction of total genetic 9257.
value using genome-wide dense marker Rocha, J.L., Baker, J.F., Womack, J.E.,
maps. Genetics 157: 1819–1829. Sanders, J.O., and Taylor, J.F. 1992.
Nejati-Javaremi, A. and Smith, C. 1995. Statistical associations between restric-
Assigning linkage haplotypes from parent tion fragment length polymorphisms and
and progeny genotypes. Genetics 142: quantitative traits in beef cattle. Journal
1363–1367. of Animal Science 70: 3360–3370.
Orr, N. and Chanock, S. 2008. Common Rocha, J.L., Pomp, D., and Van Vleck, L.D.
genetic variation and human diseases. 2002. QTL analysis in livestock. Methods
Advances in Genetics 62: 1–32. in Molecular Biology 195: 311–346.
Papp, A.C., Pinsonneault, J.K., Cooke, G., Ron, M. and Weller, J.I. 2007. From QTL
and Sadée, W. 2003. Single nucleotide to QTN identification in livestock—
polymorphism genotyping using allele- Winning by points rather than knockout:
specific PCR and fluorescence melting A review. Animal Genetics 38: 429–439.
curves. Biotechniques 34: 1068–1072. Saiki, R.K., Bugawan, T.L., Horn, G.T.,
Pollinger, J.P., Bustamante, C.D., Fiedel- Mullis, K.B., and Erlich, H.E. 1986.
Alon, A., Schmutz, S., Gray, M.M., and Analysis of enzymatically amplified

Genome, Transcriptome, and Proteome Resources 21

beta-globin and HLA-DQ DNA with Snelling, W.M., Chiu, R., Schein, J.E., Hobbs,
allele-specific oligonucleotide probes. M., Abbey, C.A., Adelson, D.L., Aerts, J.,
Nature 324: 163–166. Bennett, G.L., Bosdet, I.E., Boussaha, M.,
Sandor, C. and Georges, M. 2008. On the Brauning, R., Caetano, A.R., Costa, M.M.,
detection of imprinted quantitative trait Crawford, A.M., Dalrymple, B.P., Eggen,
loci in line crosses: Effect of linkage dis- A., Everts-van der Wind, A., Floriot, S.,
equilibrium. Genetics 180: 1167–1175. Gautier, M., Gill, C.A., Green, R.D., Holt,
Schaschi, H., Aitman, T.J., and Vyse, T.J. R., Jann, O., Jones, S.J., Kappes, S.M.,
2009. Copy number variation in the Keele, J.W., de Jong, P.J., Larkin, D.M.,
human genome and its implication in Lewin, H.A., McEwan, J.C., McKay,
autoimmunity. Clinical and Experimental S., Marra, M.A., Mathewson, C.A.,
Immunology. February 11 [Epub ahead of Matukumalli, L.K., Moore, S.S., Murdoch,
print]. B., Nicholas, F.W., Osoegawa, K., Roy, A.,
Schibler, L., Vaiman, D., Oustry, A., Giraud- Salih, H., Schibler, L., Schnabel, R.D.,
Delville, C., and Cribiu, E.P. 1998a. Silveri, L., Skow L.C., Smith, T.P.,
Comparative gene mapping a fine-scale Sonstegard, T.S., Taylor, J.F., Tellam, R.,
survey of chromosome rearrangements Van Tassell, C.P., Williams, J.L., Womack,
between ruminants and humans. Genome J.E., Wye, N.H., Yang, G., Zhao, S., and
Research 8: 901–915. the International Bovine BAC Mapping
Schibler, L., Vaiman, D., Oustry, A., Guinec, Consortium. 2007. A physical map of the
N., Dangy-Caye, A.L., Billault, A., Cribiu, bovine genome. Genome Biology 8: R165.
E.P. 1998b. Construction and extensive Strerath, M., Detmer, I., Gaster, J., and
characterization of a goat bacterial artifi- Marx, A. 2007. Modified oligonucleotides
cial chromosome library with threefold as tools for allele-specific amplification.
genome coverage. Mammalian Genome Methods in Molecular Biology 402:
9: 119–124. 317–328.
Schwerin, M. 2001. Molecular genome anal-
ysis in livestock at the beginning of a new Tellam, R.L. and the Bovine Genome
millennium. Reproduction in Domestic Sequencing and Analysis Consortium.
Animals 36: 133–138. 2009. What does analysis of the bovine
Sellner, E.M., Kim, J.W., McClure, M.C., genome sequence say about innate immu-
Taylor, K.H., Schnabel, R.D., and nity? Plant and Animal Genome XVII
Taylor, J.F. 2007. Board-invited review: Conference, W087, January 10–14, San
Applications of genomic information in Diego, CA..
livestock. Journal of Animal Science 85:
3148–3158. Tian, C., Gregersen, P.K., and Seldin, M.F.
Slate, J., Van Stijn, T.C., Anderson, R.M., 2008. Accounting for ancestry: Population
McEwan, K.M., Maqbool, N.J., Mathias, substructure and genome-wide associa-
H.C., Bixley, J.J., Stevens, D.R., Molenaar, tion studies. Human Molecular Genetics
A.J., Beever, J.E., Galloway, S.M., and 17: R143–150.
Tate, M.L. 2002. A deer (subfamily
Cervinae) genetic linkage map and the Van Laere, A.S., Nguyen, M., Braunschweig,
evolution of ruminant genomes. Genetics M., Nezer, C., Collette, C., Moreau, L.,
160: 1587–1597. Archibald, A.L., Haley, C.S., Buys, N.,
Tally, M., Andersson, G., Georges, M.,
and Andersson, L. 2003. A regulatory
mutation in IGF2 causes a major QTL

22 Quantitative Genomics of Reproduction

effect on muscle growth in the pig. Nature Wang, J., Chuang, K., Ahluwalia, M., Patel,
425: 832–836. S., Umblas, N., Mirel, D., Higuchi, R.,
Van Tassell, C.P., Smith, T.P., Matukumalli, Germer, S. 2005. High-throughput SNP
L.K., Taylor, J.F., Schnabel, R.D., Lawley, genotyping by single-tube PCR with TM-
C.T., Haudenschild, C.D., Moore, S.S., shift primers. Biotechniques 39: 885–
Warren, W.C., and Sonstegard, T.S. 2008. 893.
SNP discovery and allele frequency esti-
mation by deep sequencing of reduced Wu, C.H., Nomura, K., Goldammer, T.,
representation libraries. Nature Methods Hadfield, T., Dalrymple, B.P., McWilliam,
5: 247–252. S., Maddox, J.F., Womack, J.E., and
Velculescu, V.E., Zhang, L., Vogelstein, B., Cockett, N.E. 2008. A high-resolution
and Kinzler, K.W. 1995. Serial analysis of comparative radiation hybrid map of ovine
gene expression. Science 270: 484–487. chromosomal regions that are homolo-
Velculescu, V.E., Zhang, L., Zhou, W., gous to human chromosome 6 (HSA6).
Vogelstein, J., Basrai, M.A., Bassett, D.E., Animal Genetics 39: 459–467.
Hieter, P., Vogelstein, B., and Kinzler,
K.W. 1997. Characterization of the yeast Wu, C.H., Nomura, K., Goldammer,
transcriptome. Cell 88: 243–251. T., Hadfield, T., Dalrymple, B.P.,
Voight, B.F., Kudaravalli, S., Wen, X., and McWilliam, S., Maddox, J.F., Womack,
Pritchard, J.K. 2006. A map of recent posi- J.E., and Cockett, N.E. 2009. A radiation
tive selection in the human genome. PLoS hybrid comparative map of ovine chro-
Biology 4: e72. mosome 1 aligned to the virtual sheep
Walker, S.J., Wang, Y., Grant, K.A., Chan, F., genome. Animal Genetics 40: 435–
and Hellmann, G.M. 2006. Long versus 455.
short oligonucleotide microarrays for
the study of gene expression in nonhu- Wu, C.H., Nomura, K., Goldammer, T.,
man primates. Journal of Neuroscience Hadfield, T., Womack, J.E., and Cockett,
Methods 152: 179–189. N.E. 2007. An ovine whole-genome radia-
tion hybrid panel used to construct an RH
map of ovine chromosome 9. Animal
Genetics 38: 534–536.


Quantitative Genomics of Female Reproduction

Jeffrey L. Vallet, Dan J. Nonneman, and Larry A. Kuehn

2.1 Introduction more QTL affecting a trait. In practice, each
step presents challenges in the successful
The purpose of this chapter is to review quan- implementation of this technology.
titative trait loci (QTL) and the development
and discovery of genomic markers for female 2.2 Female reproductive
reproductive traits of domestic livestock phenotypes
species. For this chapter, we define a quanti-
tative trait as one for which the phenotypes 2.2.1 Complexity of reproduction
of individual animals in a population form a
continuum, and the trait itself is influenced Females contribute most of the complexity
by the function of numerous genes in concert. of reproduction in livestock. In order to
The utility of the QTL is that they can be produce offspring, females must efficiently
assessed in individual animals using DNA reach puberty; display estrus; shed one or
markers to determine which alleles of the more competent ova; create the appropriate
QTL are present. Individual alleles of a QTL oviductal environment for fertilization to
are associated with genetic variation in the take place; undergo the necessary systemic,
trait of interest. Discovery of QTL relies ovarian, and uterine modifications to support
on collecting appropriate phenotypes, accu- pregnancy; deliver the offspring; lactate; and
rately determining genotypes within the area successfully return to estrus after offspring
of the genome affecting the trait, and dem- are weaned. Puberty, estrous cyclicity, oocyte
onstrating a statistical association between competence, and the oviductal contribution
genotypes and phenotypes. Making use of to fertilization are all controlled by the
QTL requires one more step, the appropriate female. The conceptus contributes to implan-
incorporation of the QTL in selection tation, placental development and function,
schemes, which is reliant on accurately pre- fetal survivability during pregnancy, suscep-
dicting the effect of different alleles of one or tibility to stillbirth, and preweaning survival


24 Quantitative Genomics of Reproduction

and growth rate. Thus, both the conceptus genetic influence contributing to later preg-
and mother contribute genetically to the nancy success may not be manifested if ovu-
success of pregnancy, parturition, lactation, lation or fertilization rates are low. These
and postweaning return to estrus. In litter- issues are relevant to the search for QTL.
bearing livestock like the pig, the genetic Traits that are the result of multiple inter-
contribution to success of traits like litter acting processes will be affected by many
size is the combined effect of many interact- genes each with small effects on the overall
ing genotypes (sow and all piglets in the litter) trait. In traits where genes influence the
and thus is very complex. Negative correla- outcome sequentially, poor performance of
tions between the mother’s and piglet’s genes early in the process renders later gene
genetic contributions for birth weight, still- effects undetectable. Accurate detection and
birth, and preweaning mortality have been estimation of these interdependent genetic
reported (Roehe 1999; Arango et al. 2006). effects require the collection of phenotypes
Therefore, the effect of genes of the sow on a of component traits to isolate gene effects
trait may have an opposing effect to the same on specific subcomponents of these complex
genes of the piglet. Despite this potentially traits.
complex antagonism, most female reproduc-
tive QTL analyses have focused on the influ- 2.2.3 Genetic correlation and
ence of genetic differences between dams, pleiotropic effects
and ignored the influence of genes in the
offspring. Since each offspring inherits half A further complexity is that many genes
its genome from the dam, genetic effects contribute to more than one phenotypic
attributed to the dam in QTL analyses for trait. This situation is known as pleiotropy
pregnancy traits are in fact some combina- and results in genetic correlations between
tion of maternal and fetal genetic effects. traits. Numerous genetic correlations
between reproductive traits and other traits
2.2.2 Complex phenotypes have been reported. Genetic correlations can
result in beneficial or detrimental changes
Many reproductive traits are combinations in traits other than the trait of interest. For
of several traits. For example, litter size in example in pigs, litter size is negatively
pigs combines ovulation rate; oocyte com- genetically correlated with lean meat
petence; oviductal factors necessary for content (Holm et al. 2004), birth weight
fertilization; systemic, ovarian, and uterine (Mesa et al. 2005), and average birth interval
factors needed to maintain pregnancy; the (Canario et al. 2006) and positively geneti-
fertilization rate of the sperm; implantation; cally correlated with percent stillborn
placental formation and function; fetal sur- (Canario et al. 2006). Given these correla-
vivability; and piglet susceptibility to still- tions, selection for litter size results in det-
birth. In this case, genomes of the dam, sire, rimental effects on the piglets, unless other
and fetuses contribute to the trait, and each traits are also considered in the selection
subphenotype is typically controlled by program (e.g., birth weight). In contrast, far-
numerous genes. This genetic complexity rowing survival and preweaning survival are
contributes to the low heritability of repro- positively genetically correlated (Mesa et al.
ductive traits (Table 2.1). In addition for 2006). Thus, selecting for decreased still-
some traits such as litter size in pigs, the birth rate would also improve preweaning

Female Reproduction 25

Table 2.1 Heritabilities for reproductive traits in livestock.

Trait Heritability Species Reference

Age at puberty 0.31–0.40 Swine Holm et al. (2005); Sterning et al. (1998)
Weaning to Swine Holm et al. (2005); Sterning et al. (1998)
estrus interval 0.02–0.24
Ovulation rate Swine Rosendo et al. (2007)
0.33 Cattle Gregory et al. (1997)
Pregnancy rate 0.10 (single obs.) Cattle Gregory et al. (1997)
Litter size 0.35 (six obs.)
Cattle Bormann et al. (2006); MacNeil et al. (2006)
Twinning 0.07–0.13
Birth weight Swine Canario et al. (2006); Holm et al. (2005); van der Steen (1985)
0.10–0.20 (dam only) Swine Mesa et al. (2005)
Stillbirth 0.08 (direct) Swine Mesa et al. (2005)
Calving difficulty 0.08 (maternal) Sheep Janssens et al. (2004); Okut et al. (1999)
Preweaning 0.06–0.17
mortality Cattle Gregory et al. (1997); Komisarek and Dorynek (2002)
Length of 0.01–0.10
productive life Swine Arango et al. (2006); Roehe (1999)
Stayability 0.03–0.09 (direct) Swine Arango et al. (2006); Roehe (1999)
0.19–0.26 (maternal) Cattle Gregory et al. (1997); Gutierrez et al. (2007)
0.39–0.42 (direct) Cattle Gregory et al. (1997); Gutierrez et al. (2007)
0.11–0.21 (maternal) Sheep Hanford et al. (2002)
0.27 (direct) Sheep Hanford et al. (2002)
0.25 (maternal)
Swine Arango et al. (2006); Mesa et al. (2006); White et al. (2006)
0.001–0.14 (direct) Swine Arango et al. (2006); Mesa et al. (2006); White et al. (2006)
0.002–0.16 (maternal)
Cattle Bennett and Gregory (2001); Gutierrez et al. (2007)
0.19–0.43 (direct) Cattle Bennett and Gregory (2001); Gutierrez et al. (2007)
0.14–0.23 (maternal)
Swine Arango et al. (2006); Mesa et al. (2005)
0.05–0.18 (direct) Swine Arango et al. (2006); Mesa et al. (2005)
0.08–0.10 (maternal)
Swine Serenius and Stalder (2006)
Cattle Martinez et al. (2005)

survival. Age at first service is positively Any QTL with pleiotropic effects could
correlated with weaning to first service thus be determined, and this information is
interval after first parity (Sterning et al. used in selection schemes. Hence, collecting
1998; Holm et al. 2005) and negatively a variety of economically relevant pheno-
genetically correlated with sow lifetime types on every animal in a QTL population,
productivity (Serenius and Stalder 2004). not just the phenotype(s) of interest, is
Selection for early puberty would improve important.
early return to estrus after weaning and
would be associated with improved sow pro- 2.2.4 Trait measurement
ductive lifetime. Because of these genetic
correlations, QTL effects on a broad variety Another consideration regarding collection
of health and economic traits should be of female reproductive phenotypes is the
determined before they are used to manipu- ease with which phenotypes can be obtained.
late any one particular trait of interest. Traits like ovulation rate (i.e., by laparoscopy

26 Quantitative Genomics of Reproduction

or ultrasound) and speed of parturition (i.e., nucleotide polymorphisms (SNPs) and
by video surveillance or constant monitor- insertions/deletions (indel). Detection of an
ing) can be measured, but are time-consum- SNP relies on strategies to detect individual
ing and therefore costly. Factors involving nucleotides within a sequence. Detection of
oocyte competence, oviductal influences on indels depends on the size of the indel. Small
fertilization, and the systemic, ovarian, and indels can be detected using methods useful
uterine changes required to maintain preg- for either SNP detection or detection of
nancy are still being defined, such that the DNA fragment sizes. Larger indels are typi-
appropriate measurements for these traits are cally detected by differences in fragment
the subject of ongoing research. Generally sizes. Indels can range in size up to the inser-
speaking, traits that are easily and externally tion or deletion of the entire genes (Redon
measured currently form the basis of routine et al. 2006; Beckmann et al. 2007; McCarroll
genetic selection for female reproductive and Altshuler 2007), resulting in differences
traits. These include age at puberty, preg- in copy number of specific genes. Detection
nancy rate, litter size and nipple number of copy number differences is a special case
(for polytocous species), calving difficulty/ and is carried out using strategies that differ
stillbirth rate, preweaning mortality, return from the detection of SNP and smaller frag-
to estrus after parturition, and stayability. ments (see below).
Although a single incidence of pregnancy or
stillbirth of a particular animal is clearly not 2.3.1 SNPs and genotyping
a continuously distributed trait, this trait can
be converted to a rate or probability after SNPs are differences in one nucleotide base
multiple observations on the same individ- at a specific position in the DNA between
ual or on multiple female progeny of a given members of a population. The rate of con-
parent (either by calculating a rate of occur- version or mutation of adenine (A), cytosine
rence [i.e., number of pregnancies divided by (C), guanine (G), and thymidine (T) nucleo-
the number of services or the number of tides to the other nucleotides between gen-
stillborns divided by the number of offspring] erations is low (∼2 × 10−8 per nucleotide;
or using other methods of statistical model- Nachman and Crowell 2000). The low
ing of categorical traits). Thus, the chance of rate means that the incidence of reverse
a particular outcome forms a continuum mutation is extremely low; thus, SNPs are
among individuals and can be thought of as thought to be permanent changes in the
a continuously distributed trait (Gregory DNA, distinguishing them from microsatel-
et al. 1997). lites, whose mutation rate is much higher
(∼1 × 10−4 per gamete × locus; Crawford and
2.3 Genetic markers and Cuthbertson 1996; see below). Detection
genotyping methods of SNP typically relies on the ability to
distinguish individual nucleotides in a
The goal of genotyping is to detect DNA sequence. Methods include restriction endo-
sequence variation between individuals in a nuclease susceptibility, hybridization dif-
population. Methods differ according to the ferences, or detection of incorporation of
type of genetic variation to be detected. Two different nucleotides. Restriction fragment
general types of variation are found: single length polymorphism or RFLP analysis relies
on the introduction of a new restriction

Female Reproduction 27

endonuclease site within the DNA caused available from a variety of companies. These
by the SNP. The DNA is amplified with methods all have in common multiplexing
specific primers flanking a genetic polymor- strategies that allow simultaneous genotyp-
phism; the resulting amplified product is ing from a variety of loci. These strategies
digested with restriction endonuclease and reduce the average cost of a single genotype
subjected to gel electrophoresis. The new and allow the possibility of whole genome
restriction endonuclease site introduced by association studies, which will be discussed
the SNP is detected as digestion of the later.
amplified product, allowing visualization of
genotypes by differences in DNA banding 2.3.2 Indels/microsatellites
patterns after electrophoresis. and genotyping

Although RFLP analysis of SNP is an Small indels can also be detected using the
effective detection method, analysis of many above strategies, or by direct detection of
individual SNP using this method is time- the size of amplified fragments of DNA.
consuming and costly. Emphasis has been Microsatellites, which fall into this cate-
placed on multiplexing of genotype collec- gory, have been used extensively in QTL
tion, such that genotypes from multiple analyses and are typically detected by elec-
sites are collected simultaneously. These trophoresis after polymerase chain reaction
higher-throughput genotyping platforms are (PCR) amplification of specific DNA regions
based on either hybridization or primer in the presence of isotope or fluorescent
extension/nucleotide incorporation to dif- dye-labeled nucleotides. Microsatellites are
ferentiate alleles. Sequenom genotyping regions of nucleotide repeats (e.g., (CA)n) of
technology employs detection of incorpora- varying length n within genomic DNA.
tion of different nucleotides by mass spec- They occur most often in noncoding regions
trometry, since each incorporated nucleotide of the genome, with some exceptions, and
differs in molecular weight. Amplification this results in a random distribution through-
and detection of numerous fragments can be out the genome. The number of repeats is
performed simultaneously, allowing dozens highly polymorphic (mutation rate ∼1 × 10−4
of genotypes to be determined from a single per gamete × locus) and is greater for repeats
genomic DNA sample (www.sequenom. of large n. Microsatellites have several
com/). Affymetrix genotyping chips rely advantages over SNP for genetic analysis.
on differences in hybridization of genomic Because they cause differences in DNA frag-
DNA to thousands of oligonucleotide probes ment size, they are easy to detect. Rather
immobilized on the chip, allowing for the than two alternative alleles at a given locus,
simultaneous detection of thousands of microsatellites may have several alleles
SNPs from a single sample (www.affyme- (e.g., CA10, CA12, CA14) within the popula- Illumina genotyping tion. The larger number of alleles allows
methods detect incorporation of specific better tracking of genetic variation through
labeled nucleotides into oligonucleotides a pedigree during linkage analysis (see
linked to beads and also allow for simultane- below). This advantage, which is due to the
ous detection of thousands of genotypes high mutation rate between generations, is
from a single sample ( also a disadvantage, in that the rate of inter-
These platforms are provided as examples conversion between alternate alleles is high.
of currently available strategies; others are

28 Quantitative Genomics of Reproduction

While SNP alleles are assumed to be identi- copy number are detected as differences
cal by descent at some point in the history in individual probe hybridization signals
of the species, the high mutation rate of mic- between genomic DNA from different
rosatellites does not allow this assumption. individuals. An alternative approach is to
The same microsatellite allele could have use quantitative polymerase chain reaction
been generated numerous times in a popula- (qPCR) using genomic DNA as template.
tion. Thus, although microsatellites have Correction of the results with qPCR values
been useful in detection and selection of of a known single copy gene yields an accu-
QTL in defined pedigrees by exploiting rate assessment of copy number of target
linkage, they are of limited value as genetic regions. Although allelic duplication of large
markers across unrelated populations. regions of the genome is much rarer than the
incidence of SNP, the chance of these poly-
2.3.3 Gene copy number morphisms having an effect on the animal
and genotyping is much greater than individual SNP, most
of which have no functional effect on the
A more recent development in genetic analy- animal. Because of the greater chance that
sis has been the detection of differences these differences will result in phenotypic
in gene copy numbers between individuals differences, this is a growing area of research.
within a population (Redon et al. 2006;
Beckmann et al. 2007; McCarroll and 2.4 Association of phenotypes
Altshuler 2007). Although differences in with genotypes
reproductive traits associated with differ-
ences in gene copy numbers have not yet Once phenotypes and a method of genotyp-
been described in livestock, in mice, it ing are available, it is possible to associate
appears that the number of copies of Qa-2 differences in genotype with differences
genes in the major histocompatibility locus in phenotype, and there are two broad
influences embryo development and sur- approaches: candidate gene and genome
vival (Byrne et al. 2007). Although this region scan. The candidate gene approach relies on
is deleted in swine (Renard et al. 2006), there prior knowledge of the role of a specific gene
is evidence that major histocompatibility in some aspect of the phenotype measured.
genes are associated with litter size (Conley The genome scan approach assumes no prior
et al. 1988). In addition, the number of major knowledge of the genes involved in the trait
histocompatibility genes that are expressed of interest, and the whole genome or indi-
appears to vary in cattle (Ellis et al. 1999), vidual chromosomes are surveyed for regions
which could be the result of differences in associated with differences in the trait. The
copy number. Differences in gene copy genome scan approach falls into two general
numbers can be detected by analyzing rela- categories: linkage analysis and linkage
tive abundance of a particular sequence in disequilibrium (LD) analysis. In livestock,
the genome. Strategies include hybridizing linkage analysis has historically dominated
genomic DNA with arrays of probes tiling for surveys of the whole genome, but tech-
the genome (e.g., arrays of bacterial artificial nology is rapidly becoming available to
chromosomes [BACs] sufficient to cover the make LD analysis of the whole genome
whole genome, or cDNA arrays designed for more feasible. LD analysis has historically
transcription analysis). Differences in gene

Female Reproduction 29

been used to fine-map regions associated end result of the pathway, and thus those
with a trait that was discovered using linkage nearer the end result would have more
analysis. predictable effects on reproduction. Genes
whose effects are essential in a particular
2.4.1 Candidate genes process would seem to make bad candidates,
because significant changes in the function
Available scientific literature describes the of these genes have a good chance of being
role of numerous genes in female reproduc- lethal. Thus, although selection of candidate
tive traits, and it is not difficult to create a genes for female reproduction traits, or
list of potential genes that could have sig- indeed any trait, would seem to be straight-
nificant effects on reproduction. Estrogen forward given current scientific knowledge,
receptor was one of the first genes investi- more knowledge of the regulation of specific
gated in livestock for association between female reproductive traits and their relation
genetic variation and a female reproductive to other traits would improve the selection
trait, specifically litter size (Rothschild et al. of candidate genes. Furthermore, this strat-
1996). Since then, other genes known to egy is unlikely to discover all the genetic
play roles in female reproduction have been variation influencing reproductive traits,
investigated for association with female because despite our extensive knowledge of
reproductive traits. Genetic variation in the the role of various genes in reproduction, the
prolactin receptor (Drogemuller et al. 2001), roles of numerous other genes is not yet
retinol-binding protein (Rothschild et al. known.
2000), folate-binding protein (Vallet et al.
2005a), and erythropoietin receptor (Vallet Once the candidate gene is selected,
et al. 2005b) genes have been associated with the gene is searched for DNA sequence vari-
litter size; and genetic variation in the genes ation in a population of interest. A typical
for insulin-like growth factor and its binding approach is to amplify a region of the gene
proteins and receptors have been associated in several animals by PCR and sequence the
with sow productive lifetime (Mote et al. product. Animals to be sequenced should
2006). Superficially, any gene with a role in be as genetically diverse as possible but
female reproduction is a legitimate candi- still represent the population of interest.
date. However, some genes represent better Diversity increases the probability that QTL
candidates than others. For example, genes explaining portions of the genetic variation
that have broad effects on traits other than within the population will be detected,
reproduction (i.e., pleiotropy), such as tran- while maintaining representation of the
scription factors, are likely to be poor candi- population ensures that the variation found
date genes, because changes in the function will be useful in the population of interest.
of the gene would have broad effects beyond How many animals to sequence will be
female reproductive performance, unless the determined by the lowest allele frequency of
effect of the genetic variation within the interest. In a population in equilibrium, if
gene modified expression only in repro- the frequency of a given SNP is 90% for one
ductive tissues. Related to this, genetic allele and 10% for the other, then assuming
variation in genes at the top of metabolic random mating under Hardy–Weinberg
pathways that affect reproduction would equilibrium, 81% of animals will be homo-
have broader effects than those nearer to the zygous for the major allele, 18% heterozy-
gous, and 1% homozygous for the minor

30 Quantitative Genomics of Reproduction

allele. Analysis of 10 animals is expected especially relevant to early embryo develop-
to yield 2 heterozygous animals, and these ment (Schier 2007; Stitzel and Seydoux
may be difficult to distinguish from PCR 2007). Thus, due to our current lack of ability
or sequencing-induced errors. Additionally, to predict whether a specific polymorphism
if only 10 animals are genotyped, the prob- does or does not have an effect on gene func-
ability of having no heterozygous animals tion, it should be assumed that any genetic
in the group is relatively high (∼0.12, the variation in the proximity of a gene could
probability of a homozygous animal raised alter its function in some way, and a better
to the 10th power, 0.8110). Even if the approach to this problem is to perform a
low-frequency polymorphism is discovered, comprehensive search for genetic variation
association with differences in a trait in and around the gene, and perform asso-
requires sufficient observations to reach sta- ciation analyses using all of the SNP
tistical significance; therefore, low frequency discovered.
limits statistical power as well. These two
problems make detection of associations in 2.4.2 Genome scans
low-frequency alleles much more difficult.
Genome scans can be done either by linkage
It has been a common practice to sequence analysis or by LD analysis, although cur-
a gene among animals only within the exons, rently all genome scans in livestock have
and after the examination of the polymor- been done by linkage analysis. Linkage anal-
phisms detected concludes that no useful ysis is performed on a population of animals
polymorphisms are present because differ- in which the pedigree and relationships
ences observed did not alter the amino acid between animals is known. The number of
sequence of the resulting protein. Aside from generations between animals in the popula-
altering the amino acid coding sequence, tion is typically limited, creating artificially
genetic polymorphisms can have a variety of large regions of LD between parents and off-
other effects on gene function. The DNA spring. These large regions allow tracking of
sequence upstream, downstream, and within specific chromosomal regions from parents
the exons and introns has been shown in to offspring within the population using
numerous reports to be involved in the markers spaced every 10 or 20 million bases;
control of gene function (Fedorova and thus, the inheritance of the entire genome
Fedorov 2003). Effects of DNA elements on of livestock can be examined with 150–300
gene transcription through promoter and genetic loci. Microsatellites have typically
enhancer elements (Roeder 1991; Maniatis been used for these analyses because each
and Reed 2002), on the efficiency of mRNA locus may have numerous alleles, improving
splicing (Maniatis and Reed 2002), on trans- the information content of each locus and
lation of the mRNA through transcription increasing the ability to track regions from
initiation (Iida and Masuda 1996) or blockade parent to offspring in the analysis. This
or through codon bias (Kurland 1991; Akashi approach has several advantages. No prior
2001), and on mRNA degradation and/or knowledge of the genes responsible is
storage through mRNA–protein (Ruiz- needed. Relatively few markers are needed
Echevarria et al. 2001) or mRNA–microRNA to assess the effects of the entire genome.
interactions (Wienholds and Plasterk 2005; Founder parent animals can be selected from
Kiefer 2006; Zhao and Srivastava 2007) have breeds or lines that are divergent in the trait
been described. MicroRNAs are likely to be

Female Reproduction 31

of interest, making the discovery of genomic chromosomal segments are inherited in rela-
regions having a major influence on the phe- tively large segments between generations.
notype more likely. Most of the QTL for Linkage analysis often results in associa-
reproductive traits for livestock generated tions with genomic regions that are millions
by genome scans currently reported in the of bases in length harboring hundreds of
literature have employed variations of this genes. Our knowledge of individual gene
strategy. function is typically insufficient to allow
the effective selection of candidate genes
Although linkage analysis is very good within regions of this size. Finally, because
at finding QTL regions, it has several disad- the linkage between markers and QTL
vantages. Because a pedigreed population is regions within the experimental population
needed for this type of analysis, generation has been artificially created by the limited
of a suitable population may take years to number of generations in the pedigree, and
accomplish for livestock and is expensive. results are reliant on the founder animals
Dairy cattle have had a significant advantage used, marker associations with QTL found
over other livestock species in this regard, for the experimental population are specific
in that large pedigrees were already available to the population and typically do not trans-
within the dairy industry. Similar resources fer to livestock populations at large without
have been used in other countries for swine further research using LD analysis. Linkage
(Tribout et al. 2008) but have not been analysis will detect regions associated with
exploited in the United States. Instead, differences in traits, but strategies related to
QTL analysis in swine has relied on the gen- those discussed for candidate genes must
eration of specific populations. A related then be used to obtain markers that are of
disadvantage is that the only genomic use to livestock populations beyond that
regions that will be found to be associated used to discover the QTL. The LD analysis
with the trait of interest will be those that is needed to reduce the size of the associated
differed among the original founder animals region. Combined linkage, LD analysis of
of the population used in the QTL analysis. specific genome regions has been used suc-
Thus, a different population with different cessfully to fine-map QTL regions that were
founder animals may yield different results. previously identified by linkage analysis
Therefore, selection of sufficient founder (Olsen et al. 2005; Schnabel et al. 2005b).
animals to represent the inference popula-
tion is important, because sampling of small 2.4.3 LD
numbers of animals will limit the number
of QTL regions that are identifiable within Discovery of the actual genetic variation
the population. On the other hand, selection responsible for the difference in a trait is not
of too many founder animals increases the necessary for the association to be of use,
number of total animals needed in the popu- because of LD. LD is defined as the simulta-
lation to provide sufficient statistical power neous inheritance of adjacent regions on the
to be able to detect the effect of the genetic same chromosome. LD in a randomly breed-
contribution of each founder animal. The ing population is removed by genetic recom-
extent of newly created LD among the bination, in which corresponding regions
animals within the artificially generated of an individual animal’s two chromosomes
population limits the ability to reduce the switch places during gamete production.
size of any QTL region discovered, because

Click to View FlipBook Version
Previous Book
Next Book