Investigation of Genome Variations As Potential Markers for Boar Fertility

Investigation of Genome Variations as Potential Markers for Boar Fertility

Anh Tuyet Quach

A Thesis Presented to The University of Guelph

In partial fulfilment of requirements for the degree of Doctor of Philosophy in Biomedical Sciences

Guelph, Ontario, Canada

ABSTRACT

INVESTIGATION OF GENOME VARIATIONS AS POTENTIAL MARKERS FOR BOAR FERTILITY

Anh Tuyet Quach Advisor: University of Guelph, 2015 Dr. W. Allan King

Boars can greatly affect the reproductive success of a herd, especially with the current widespread use of artificial insemination (AI). Low fertility boars in AI centers can cause enormous economic losses. Thus, identifying a method to detect low fertility boars at an early stage of their life will increase profitability for the swine industry. The purposes of this thesis were to investigate genomic variations including chromosome abnormalities, copy number variation (CNV) and a very special case of CNV on the Y chromosome, testis specific protein Y- encoded (TSPY) gene in pig populations to develop and assess significant tools for detecting low fertility boars. These assessments included karyotyping all young AI boars and evaluating whether CNV of some genes can be used as genetic markers of male fertility in pig. At the chromosome level, this study identified 12 different chromosome rearrangements including 9 reciprocal translocations, two inversions and one Robertsonian translocation. The large adverse effect on litter size of carriers of chromosome abnormalities together with the high prevalence of chromosome abnormalities (1.64%) and high percentage of de novo cases (2/5 cases) in the

Canadian pig populations highlights the significant value of having an imperative screening program for chromosome abnormalities in Canadian swine industry. At the DNA level, two

groups of animals with direct boar effects (DBE) values in the upper (high fertility) and lower

(low fertility) end of the distribution were used for CNV analysis and the identification of putative markers of fertility. This study identified 35 CNV regions (CNVRs) covering 36.5 Mb or ~1.3% of the porcine genome. The identified 35 CNVRs encompassed 50 genes. A functional analysis of several databases revealed that the genes found in CNVRs from the low fertility group have been significantly enriched in members of the innate immune system, Toll-like receptor and RIG-I-like receptor signaling and fatty acid oxidation pathways. Unlike humans and cattle, boars have only 3 copies of TSPY gene that do not vary among individuals thus copy number of TSPY gene represents a poor marker for fertility in pig.

DECLARATION OF WORK PERFORMED

I declare that all the work reported in this thesis was prepared by me with the exception of the items listed below.

Blood samples were collected by the veterinarians in the farms. Breed stock suppliers which provided samples for this thesis helped to collect reproductive data and calculated estimated breeding value (EBV) and direct boar effect (DBE) of carriers. Mariana Macedo,

Tamas Revay and Pradeep Balaraju helped in harvesting cell blood culture. Cell culture of progeny of rob(13;17) was performed by Mariana Macedo and Tamas Revay. C-banding and

DAPI staining for rob(13;17) was carried out with the assistance of Tamas Revay.

Laurence Maignel and Brian Sullivan organised and managed the genotype database.

CNV data analysis and qPCR in chapter 2 were performed by Tamas Revay.

Assistance was also provided for DNA extraction by Olutobi Oluwole and Primers design by Tamas Revay and Christine Hamilton. ddPCR, FISH and qPCR for low and high fertility boars in chapter 3 were performed by Tamas Revay.

ACKNOWLEDGEMENTS

First and foremost, I would like to express my sincere appreciation to my advisor, Dr. W.

Allan King, for your thoughtful, enthusiasm, immense knowledge and wise advice throughout my PhD program. I owe my deepest gratitude to your understanding, constant encouragements, continuous support and thoughtful guidance helped me to overcome all obstacles and challenges for the past nine years. I have been privileged to have you as an advisor and mentor since 2006.

You really changed my life. Thank you very much Dr. King. I would also like to extend my gratefulness to other members of my advisory committee, Dr. Pavneesh Madan, Dr. Cate Dewey and Dr. Bettina Kalisch for their great contributions to develop this project and productive comments for this thesis.

This research would have not been completed without the guidance and support of Dr.

Tamas Revay. I am indebted to your enthusiasm, whole-hearted instruction, persistent assistance, encouragement and tireless contributions to this thesis. Your experience and broad insights enriched my research and knowledge in the field of molecular genetics. I will never forget hundreds of hours you spent to help and teach me about this field. Out of science, your skills in computer, photography and Kata’s skill in face and belly painting are the great memories for me and my son.

I would like to thank Dr. Monica Antenos, Liz St. John, Ed Reyes, Allison MacKay for supporting me in many different faculties in the laboratory. A special thank is also given to Drs.

Christine Hamilton and Graham Gilchrist for their willingness to help familiarize me with qPCR and DNA extraction at the initial stage of my research. A big thank you goes to

Mariana Macedo, Pradeep Balaraju and Nayoung for helping me with blood culture harvesting which were long and hard working days.

Special thanks are offered to the managers and the veterinarians for collecting blood and data from their farms. I would like to give a special acknowledgement to Brian Sullivan,

Laurence Maignel and Stefanie Wyss for allowing using the genotype database and their multiple contributions to the completion of this thesis. I would also thank Dr. Daniel Villagomez for reviewing this thesis.

It is a pleasure to share wonderful moments with all members of King and Madan’s lab.

Thank you very much Jacky, Graham, Tim, Anja, Nayoung, Jeff and Kayla. A sincere gratitude and deeply engraved on my memory is offered to Ed Reyes, Tobi, Faz, Tamas and Laura for their friendship, sharing and invaluable supporting throughout my time in Canada. I would also like to thank Elizabeth, Lesli, Sarah and Teresa for sharing the office and I enjoyed a lot our discussions.

I would like to thank all my colleagues, friends, and administrative staff at the University of Guelph who made this research possible.

I am very grateful to the Vietnam government for sponsoring my Ph.D. studies and providing all necessary aid to complete this research. Lots of thanks to the financial support from

NSERC, Canada Research Chairs Program, Agricultural Adaptation Council and Canadian

Centre for Swine Improvement made this project possible.

Finally, I have to give a very special thank you to my parents, my mother-in-law, my son, my daughter and my husband. They have been an incredible source of support and encouragement.

vii

TABLE OF CONTENTS

DECLARATION OF WORK PERFORMED ...... iv ACKNOWLEDGEMENTS ...... v TABLE OF CONTENTS ...... vii LIST OF TABLES ...... x LIST OF FIGURES ...... xi LIST OF ABBREVIATIONS ...... xv INTRODUCTION ...... 1 LITERATURE REVIEW ...... 4 FERTILITY IN PIGS ...... 4 What affects fertility? ...... 4 Nutrition ...... 4 Diseases...... 5 Management and environmental factors ...... 6 Genetics...... 6 How is fertility measured? ...... 7 STRUCTURAL VARIATIONS OF THE GENOME ...... 10 Chromosome abnormalities ...... 11 Chromosomal abnormalities in pigs...... 13 The effects of chromosome aberrations ...... 15 Efforts to eliminate carriers ...... 16 Copy number variation ...... 18 HOW TO DETECT STRUCTURAL VARIATIONS ...... 23 Cytogenetics ...... 23 Molecular cytogenetics ...... 25 High resolution arrays for the investigation of Structural Variations ...... 27 CNV detection from whole genome sequence information ...... 30 Independent validation of high throughput assays with small-scale techniques ...... 31

RATIONALE ...... 34 CHAPTER ONE ...... 37

viii

The prevalence and consequences of chromosome abnormalities in Canadian and Vietnamese breeding pig populations ...... 37 INTRODUCTION ...... 38 MATERIALS AND METHODS ...... 40 Animals ...... 40 Chromosome analysis ...... 40 Reproductive Data Analysis ...... 42 RESULTS ...... 43 Identification of Rearrangement Carriers ...... 43 Origin and dissemination of rearrangements ...... 51 Reproductive performance of carriers ...... 52 The frequency of chromosome abnormalities in Canadian and Vietnamese pig populations ...... 54 DISCUSSION ...... 55

CHAPTER TWO ...... 66 Copy number variations in high and low fertility breeding boars ... 66 MATERIALS AND METHODS ...... 70 Animals and array genotyping ...... 70 CNV analysis ...... 71 qPCR validation ...... 73 Functional annotation of CNVRs ...... 73 RESULTS ...... 75 CNV analysis ...... 75 CNVRs overlapping reproduction QTL regions ...... 80 Functional annotation of CNVR gene content ...... 81 DISCUSSION ...... 83

CHAPTER THREE ...... 91 The porcine TSPY gene is tricopy but not a copy number variant . 91 INTRODUCTION ...... 92 MATERIALS AND METHODS ...... 95 Animals and DNA extraction ...... 95 Relative TSPY copy number determination by qRT-PCR ...... 96 Absolute copy number determination by droplet digital PCR ...... 97

Statistical analysis ...... 99 Fluorescence in situ hybridization (FISH) ...... 99 RESULTS ...... 101 TSPY copy number by qPCR...... 101 TSPY copy number by ddPCR...... 103 Chromosomal localization of TSPY by FISH...... 105

GENERAL DISCUSSION ...... 110 SUMMARY, CONCLUSIONS AND FUTURE DIRECTIONS .... 120 LITERATURE CITED ...... 123 APPENDIX ...... 146

LIST OF TABLES

Table 1. Fertility data of carriers. L=Landrace; Y=Yorkshire; Mat= maternal; Dn= de novo; ***=p<0.001; **=p<0.01; *=p<0.05; ns: no statistical difference when comparing with the same herd...... 53 Table 2. Comparison of cases 1, 2, 7 and 9 from the current study to the literature ...... 55 Table 3. Comparison of inversions reported in the current study (case 11 & 12) to results from previous studies ...... 56 Table 4. Functional enrichment analysis of genes encompassing the identified CNVRs. The results of the various tools in WebGestalt (http://bioinfo.vanderbilt.edu/webgestalt/option.php) are summarized...... 89 Table 5. Summary of micro RNAs found within CNVRs ...... 90

LIST OF FIGURES

Figure 1. Representative examples of genomic variations and their size differences: ...... 10 Figure 2. Types of chromosome abnormalities. (a) reciprocal translocation, (b) Robertsonian translocation, (c) paracentric inversion, (d) pericentric inversion, (e) duplication, (f) deletion (Modified from US National Library of Medicine) ...... 12 Figure 3. Spontaneous chromosome abnormalities occur in every swine population. The actual frequency of carriers is determined by the rate of de novo events, the presence of inherited abnormalities and successful eradication efforts...... 14 Figure 4. Chromosome abnormalities are a major cause of subfertility. The pink sow (normal) has a litter with 9 piglets while the red sow (carrier of chromosome abnormality) has a litter with only 5 piglets and 2 of them (red) carry the same chromosome abnormality...... 15 Figure 5. Meiotic segregation pattern of a reciprocal translocation leading to chromosomal unbalanced gametes and embryos (Farndon 2015) ...... 16 Figure 6. Chromosome abnormalities are transmissible to the progeny. The pink sow (normal) has a litter with 9 piglets while the red sow (carrier of chromosome abnormality) has a litter with only 5 piglets and 2 of them (red) carry the same chromosome abnormality. Again, if these two carrier piglets are used for breeding, the chromosome abnormality is transmitted to their progeny...... 17 Figure 7. CNVs could affect phenotype in various ways (from Feuk et al. 2006). Structural variants can be benign, can have subtle influences on phenotypes (for example, they can modify drug response), can predispose to or cause disease in the current generation (for example, owing to inversion, translocation or microdeletion that involves a disease-associated gene), or can predispose to disease in the next generation. On the basis of their proximity to structural variants, genes might be influenced in several ways. (a) Dosage-sensitive genes that are encompassed by a structural variant can cause disease through a duplication or deletion event (upper panel; a deletion is shown here). Dosage-insensitive genes can also cause disease if a deletion of the gene unmasks a recessive mutation on the homologous chromosome (lower panel). (b) Genes that overlap structural variants can be disrupted directly by inversion (upper panel), translocation or deletion (lower panel), or copy-number variant breakpoints (not shown), which leads to the reduced expression of dosage- sensitive genes. Breakpoints that disrupt gene structures can also lead to the formation of new transcripts through gene fusion or exon shuffling (not shown). (c) Structural variants that are located at a distance from dosage-sensitive genes can affect expression through position effects. An example is shown in the upper panel. A deletion of important regulatory elements can alter gene expression; similar effects could result from inversion or translocation of such elements. Alternatively, deletion of a functional element could unmask a functional polymorphism within an effector (lower panel), which could have consequences for gene function. (d) Structural variants can function as susceptibility alleles, where a combination of several genetic factors is required to produce the phenotype. This is illustrated by an example of two structural variants that, individually, do not produce a phenotype. However, in combination they contribute to a complex disease state...... 20

xii

Figure 8. CNVs on the human Y-chromosome (Jobling, 2008). Green spots mark copy number gains while red spots deletions. The height of signals reflect frequency of CNVs identified ...... 23 Figure 9. Standard karyotype of the pig (Gustavsson, 1988) ...... 24 Figure 10. CGH and aCGH (“Genetics of Auditory Disorders” 2015) ...... 27 Figure 11. Schematic view of the SNP genotyping array. The signal at each locus corresponds to the amount of DNA ...... 29 Figure 12. CNV detection from Whole Genome Sequencing data (Valsesia et al. 2013) ...... 31 Figure 13. Real-time quantitative PCR amplification curves ...... 32 Figure 14. Droplet digital PCR (Bio-Rad Bulletin #6407) ...... 33 Figure 15. GTG-banded karyotype of a boar carrying rcp(1; 5). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 43 Figure 16. GTG-banded karyotype of a boar carrying rcp(1; 15). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 44 Figure 17. GTG-banded karyotype of a boar carrying rcp(2; 5). Normal chromosomes are placed on the right, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the left in the karyotype ...... 45 Figure 18. GTG-banded karyotype of a boar carrying rcp(3; 4). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 45 Figure 19. GTG-banded karyotype of a boar carrying rcp(3; 12). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 46 Figure 20. GTG-banded karyotype of a boar carrying rcp(6;7). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 47 Figure 21. GTG-banded karyotype of a boar carrying rcp(7;15). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 47 Figure 22. GTG-banded karyotype of a boar carrying rcp(8;13). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 48 Figure 23. GTG-banded karyotype of a boar carrying rcp(12;14). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 49 Figure 24. Boars carrying rob(13;17). Arrows indicate the derived chromosomes. (a) and (b): GTG banded karyotypes of boars carrying heterozygous and homozygous rob(13;17), respectively. (c): C band of a boar carrying heterozygous rob(13;17). (d) DAPI stain of a boar carrying homozygous rob(13;17)...... 49 Figure 25. GTG-banded karyotype of a boar carrying inv(8). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 50 Figure 26. GTG-banded karyotype of a boar carrying inv(6)(q23;q35). Normal chromosomes are placed on the left, and their altered counterparts (derived)

xiii

chromosomes indicating the presumed break points (arrows) are on the right in the karyotype...... 51 Figure 27. GTG banded and schematic drawings with QTL and candidate genes influencing fecundity. Figure legend: bold solid lines = level of significance p < 0.05; dashed lines = level of significance p > 0.05; EPOR=erythropoietin receptor; FSHb=follicle stimulating hormone beta; FUT1=fucosyltransferase 1; GNRHR=gonadotropin releasing hormone receptor; LEPR=leptin receptor; SPP1 (OPN)=secreted phosphoprotein 1 (from Buske et al., 2006)...... 62 Figure 28. Schematic drawings with candidate genes influencing fecundity. ESR=estrogen receptor; EPOR=erythropoietin receptor; FSHb=follicle stimulating hormone beta; GNRHR=gonadotropin releasing hormone receptor; SPP1 (OPN)=secreted phosphoprotein 1 (from Buske et al., 2006). a= inv(1)(p21;q11) (Y-I. Miyake, T. Matsubara 1994); b= inv(1)(p221;q11)(Anielak-Czech B. et al., 1996); c= inv(1)(p24;q29); d= inv(1)(q12;q24); e= inv(1)(p21;q210); f= inv(2)(p11;q11); g= inv(2)(p11;q21); h= inv(2)(q13;q25)(K. Massip et al. 2009); i= inv(2)(p13;q11) (Alain Ducos et al., 2002); j=inv(4)(p14;q23)(Katia Massip et al. 2010; a Ducos et al. 1997); k= inv(8)(p11;p12) (Świtoński 1991)...... 63 Figure 29. Descriptive statistics (mean ± SD) of the direct boar effects (DBE) in the high and low fertile animals, the validating groups and all together...... 71 Figure 30. Examples for the two CNV identification options in SVS software. a) 21 individual animals genomes were subjected to the Univariate CNAM segmentation option. A red bar represents a segment with significantly lower log R ratio, as compared to its surrounding regions, thus identified as genomic loss. Slightly different length CNVs were identified in different low fertility samples, thus the overlapping region could be merged to a low fertility specific CNVR. b) The multivariate CNAM method segments a group of samples together and segments are called only if present in all samples. In order to identify CNVs specific for either the high fertility or the low fertility group, the samples were segmented together, as well as grouped according to the fertility status. Only those CNVs were accepted that were present in only one group, but neither in the other phenotypic group or in all samples together. Here the red bar identifies a CNVR specific for the high fertility group...... 76 Figure 31. Results of validation experiments for 8 CNVRs by qPCR. Relative quantity of target amplicons were calculated against the control sample (C) after normalization to the beta-actin locus...... 79 Figure 32. Location of the detected CNVRs on the porcine chromosome ideograms. The size of each ideogram is proportional to that of the chromosomes. The sex chromosomes were excluded from the analysis and no CNVR was detected on chromosome 4, 5, 7 and 15. The brown bars in the middle of each chromosome represent the positions of CNVRs. The purple columns in the left are the positions of QTLs and RefSeq genes are marked by red bars on the right side of the ideograms...... 81 Figure 33. Schematic view of the porcine TSPY gene (exons 1-5 in grey) and location of primers (green), TaqMan probes (red), PflMI enzyme cut positions (blue) and FISH probes (purple)...... 97 Figure 34. TSPY CN relative to the single copy AR gene, as determined by qPCR. CN was measured at three loci (Exon 1, 3, 5) in four breeds. Columns represent the average

xiv

value (±SD) of 5 animals from the same breed. Different letters within efficiency calculation methods (STD or LinRegPCR) represent statistical significance according to two-way ANOVA (p<0.05)...... 102 Figure 35. TSPY CN by ddPCR at exons 1, 3 and 5 in normal animals from four breeds, High-fertile and Low-fertile animals. Columns represent the average value (±SD) of 5 animals. Letters mark values that are statistically different (p<0.05)...... 104 Figure 36. Y-chromosome specific fluorescence in situ hybridization signal of FISH probe#1-5 with Tyramide signal detection on a boar metaphase. DAPI stained metaphase chromosomes are converted to grayscale and AF594 hybridization signal on the short arm of the Y-chromosome is red...... 105

LIST OF ABBREVIATIONS aCGH Array comparative genomic hybridization AD Aujeszky's disease AR Androgen receptor

AZF Azoospermia factors BLUP Best linear unbiased prediction Bp Base pair C-banding Centromeric heterochromatin staining CCSI Canadian Centre for Swine Improvement Inc CGH Comparative genomic hybridization CNA Copy number alteration CNAM Copy number analysis method in SVS CNP Common copy number polymorphism CNV Copy number variation CNVR Copy number variable region CSF Classical swine fever virus DAPI 4',6-diamidino-2-phenylindole DBE Direct boar effect ddPCR digital droplet PCR DGV Database of genomic variants DMEM Dulbecco modified eagle medium DNA Deoxyribonucleic acid EBV Estimated breeding value EPOR Erythropoietin receptor ESR Estrogen receptor FBS Fetal bovine serum FISH Fluorescence in situ hybridization FSHb Follicle stimulating hormone beta FUT1 Fucosyltransferase 1 GNRHR Gonadotropin releasing hormone receptor

xvi

HSFY Heat shock transcription factor Y Kb Kilobase LEPR Leptin receptor NAHR Non-allelic homologous recombination NBA New born alive NGS New generation sequencing NOR Nucleolar organizer region p Short arm of the chromosome PCA Principal component analysis PCR Polymerase chain reaction PHA Phytohemagglutinin PPV Porcine parvovirus PRRS Respiratory syndrome virus PRV Pseudorabies virus q Long arm of the chromosome QFQ Quinacrine fluorescence (Q)-banding qPCR Quantitative PCR QTL Quantitative trait locus RBA Reverse (R)-banding with acridine orange staining RBG Reverse (R)-banding with Giemsa staining SI Swine influenza virus SNA Single nucleotide alteration SNP Single nucleotide polymorphism SPP1 (OPN) Secreted phosphoprotein 1 SVS SNP and Variation Suite software (GoldenHelix Inc.) TB Total born TSPY Testis specific protein, Y encoded ZAR1 Zygote arrest 1

INTRODUCTION

Canada is the third largest pork exporter in the world and because of its outstanding quality is exported to more than 100 countries. In 2014, Canada exported about 1.2 million tons of pork products that generated 3.7 billion CAD revenue (“Canadian Pork Exports”). Hence, the swine industry plays a vital role in the Canadian economy. Moreover, as the global population is expected to reach 9 billion people by 2050, the demand for food continues to increase dramatically, which not only opens a promising opportunity for the Canadian pork industry, but also poses a challenge to fulfill. Maintaining a competitive edge in the highly competitive global swine industry requires that Canadian producers continuously improve traits that affect productivity. Therefore, one of the main emphases of Canadian swine research must be aimed at increasing productivity through improved health, nutrition, carcass yield, and fertility.

Heritable genetic variation is the basis for diversity within and between species and influences production parameters such as fertility, fecundity, growth, disease resistance, and health (Andersson 2001). From the beginning of the 20th century, genetics has undeniably dramatically changed many production schemes, resulting in swine improvement breeding programs throughout the world. Now in the era of large-scale sequencing technologies, genetics has been foreseen to change the performance of an individual animal even much more spectacularly. Since the completion of the pioneering human genome project in 2003 there has been a significant increase in our understanding of genomic variability resulting in benefits for many areas of study including evolution and disease (Sebat et al. 2004). This led to the sequencing of other domestic animal genomes such as bovine (Elsik et al. 2009) and pig (Schook et al. 2005; Archibald et al. 2010) with the hopes of providing an important resource for further genetic improvements of these important livestock species.

In the case of swine and most other food producing domestic animals, artificial insemination (AI) technique has been widely recognized as the best way to maximize genetic gain and productivity. The underlying principle of this selection tool is that a genetically superior individual male produces as many offspring as possible with a large number of females, making fertility a key component of successful breeding. Among Canadian pigs the number of piglets born per litter varies from 2 to 19 (Beaulieu et al. 2010). Individuals that produce small litters, particularly AI boars represent a financial burden necessitating their detection and elimination as soon as possible. Thus, looking for a method to detect these low fertility boars, or the ability to select high fertility boars, at an early stage of their life will significantly increase profitability for the swine industry. One approach to this is to use genomic variability as a marker of fertility.

Chromosome abnormalities, a type of genome variation that comprises large stretches of DNA, are broadly known as the cause of low fertility in about 50% of affected boars (Quach et al.

2010; a. Ducos et al. 2008). GTG-banding technique was proven as a very useful technique to detect changes in chromosome structure that constitute chromosome abnormalities (Basrur and

Stranzinger 2008). However, very tiny structural changes such as deletions and duplications that result in variation in the number of copies of DNA sequences known as copy number variations

(CNVs) cannot be detected by this technique. Genome-wide technologies have been developed to facilitate study of these small changes. However, in contrast to chromosome abnormalities, which usually are associated with fertility problems, studies of sequence CNVs in humans and cattle showed that depending on the number of copies of specific genes such as testis specific protein Y encoded (TSPY), is associated with an increase or decrease in the fertility of the carriers compared with other members of the population (Giachini et al. 2009; Hamilton et al.

2012; Nickkholgh et al. 2010). Thus, studies of CNVs with the hopes of finding genetic markers not only for low fertility but also for high fertility boars are in progress.

The aim of this thesis was to investigate genomic variation including chromosome abnormalities and CNV in the Canadian AI boar population with the hopes of providing breeding stock suppliers’ markers of boar fertility as well as a strategy for the detection of boars with chromosomal abnormalities and preventing the transmission of these aberrations to the subsequent generations. The effect of rearrangements on total born litter size was investigated to emphasize the cost of chromosome abnormalities for individual herds and the Canadian pig industry as a whole.

LITERATURE REVIEW

FERTILITY IN PIGS

Fertility in its broadest sense is the ability to produce progeny whereas reproduction refers to the production of offspring, which in mammals are derived from two organisms which each contribute almost equally to the offspring’s genome (Wells 2014). Infertility is the inability to produce progeny or subfertility is the below average chance of producing progeny. Pig reproduction is a very complex and delicate process that relies on the mating of fertile individuals. The potential reasons for infertility or fertility impairment in both sow and boar are numerous. It can be affected by genital malformations, physical dysfunctions, various diseases, nutrition, management, environmental stresses and genetics (Gordon 1997). In polytocous species such as pigs, total failure to reproduce is rare, and thus the degree of fertility is of greatest interest. This is reflected in the number of piglets born. Hence, understanding the causes of infertility or subfertility is important as these are among the most important components of profitable pig production.

What affects fertility? Nutrition Among the various costs of swine production, feed costs account for the most significant proportion ranging from 65% to 75% of overall cost. As a result, feeding strategies aim to maintain the animals in excellent condition for reproduction from minimal feed. Moreover, many studies have shown that excessively low or high feed intake may affect fertility indirectly

(Franco DA and Pearl G 2000; Dunn TG 1980). Hence, a balanced diet in both quantity and quality must strictly be followed to meet pigs’ nutritional needs at different stages of growth in order to have the highest fertility.

Sows in various stages of reproduction such as early pregnancy, late pregnancy and lactation need to have an optimum feeding strategy for each stage. Studies have shown that sows with low feed intake during lactation resulted in weight loss, extended re-mating intervals, reduced pregnancy rate and hence reduced embryo survival. In boars, low feed intake is known to diminish their libido (Louis et al. 1994). However, high feed intake is related to weight gain and the boars become lazy and libido decreases. Moreover, high levels of fat above the testicles interferes with spermatogenesis (Herrick and Self 1962).

Many vitamin and mineral supplements have been recommended as additives to the feed to maximise reproductive performance. Sows injected with vitamin A 6 days before mating or before ovulation had increasing the number of piglets born alive and early embryonic survival, respectively (Pusateri et al. 1999; Whaley et al. 1997). Supplementation of vitamin E for boar had been reported to raise sperm concentration in the ejaculate and better semen quality

(Kolodziej A and Jacyno E 2005; Brezezińska-Slebodzińska E, Slebodziński AB, Pietras B et al.

1995).

Diseases Most pathogens that cause fever in sows result in fertility problems. In pigs, these viruses and bacteria including aujeszky's disease (AD) virus or pseudorabies virus (PRV), porcine parvovirus (PPV), porcine reproductive and respiratory syndrome (PRRS) virus, classical swine fever virus (CSF), swine influenza virus (SI), porcine enteroviruses, leptospirosis, erysipelas and brucellosis can directly affect the fertility of animals. When sows are attacked by such infectious agents, the clinical signs include abortion, return-to-service, embryo death, fetal death, stillborn and mummified fetuses (Pozzi and Alborali 2012). Boars that are infected with AD virus, enterovirus, parvovirus and swine fever virus often do not have noticeable symptoms.

Unfortunately, their semen can contain these viruses and they can be transmitted to sows during breeding. Hence, high quality semen must to be carefully and routinely examined for those viruses (Herrick and Self 1962).

Management and environmental factors Many factors including too low or high temperature, humidity, short lactation and weaning, transportation, high stocking densities and grouping of unfamiliar individuals act as stressors for pigs. The effect of stress on reproduction varies depending on the type of stress, timing of stress, the duration of stress and the ability of the animals to cope with stress. Most forms of prolonged stress have negative effects on the fertility (Einarsson et al. 2008). High temperature, above 340C, in the summer has been reported to lower total sperm counts and decrease ejaculate volume (Stone 1981). High embryo mortality was noticed when sows were exposed to heat stress at the time of implantation (Ronnie E. Nelson et al. 1970). Further, short transportation (about one hour) delayed the puberty of most of the gilts, but it was noted to be a stimulation for estrous in sows (Dalin AM, Nyberg L 1988).

Genetics It is obvious that genetic improvement plays an important role in domestic animals in general and in pigs in particular. Genetic variability of reproductive traits varies considerably from breed to breed (Max F Rothschild and Ruvinsky 2011). A number of reproductive traits including age at puberty, total born litter size, pre-weaning viability, functional teats, ovulation rate, libido, sperm volume, and testis weight have been measured and selected. The heritability of reproduction traits have been estimated from low to moderate (Groenen et al. 2012). With the application of genetic selection by advanced technology to analyze whole genome variability, large gains in the efficiency of pig reproduction have been made (Max F Rothschild and

Ruvinsky 2011). Based on data from CCSI (Canadian Centre for Swine Improvement Inc) for different Canadian pig breeds, the average total pigs born of Yorkshire, Landrace and Duroc breeds in 1998 vs 2013 are 10.4 vs 13.4, 10.1 vs 12.5, 10 vs 10 piglets per litter, respectively

(CCSI 2014; CCSI 1999). After 4 years, genetic trends for total born litter size of Yorkshire and

Landrace showed much improvement, but that was not the case for Duroc. The economic values of genetic improvements for Canadian sows have also been estimated for the period 2008 to

2013. Because of genetic improvements influencing total born litter size, piglet survival, number of teats and other traits, sows are about $126/sow/year more productive today than 6 years ago

(CCSI 2014). Moreover, numerous studies have shown that genetic disorders resulting in fertility problems such as poor semen quality, defects of the genitalia and reduced total born litter size could be due to genome variations such as chromosome abnormalities (a. Ducos et al. 2008;

Barasc et al. 2012; Harvey 1967).

How is fertility measured? The complete interruption of fertility of individuals is of great concern for farmers.

However, it is usually not the case in pig farming where fertility is affected in different levels but rarely arrested. Low and moderate levels of fertility are harder to detect and take longer to identify and confirm. The longer the delay in detection is, the larger the overalls loss in profitability. Hence, in pig farming, measuring fertility efficiency is extremely important and should be done carefully and systematically. Fertility can be measured in many ways including weaning to estrus interval, anestrus, conception rate, farrowing interval, total number of pigs born, pigs born alive, stillborn, mummified, piglet perinatal survival, repeat breeding rate and semen quality (sperm concentrate, motility, color, volume). As a combination of total born litter size, piglet survival and farrowing interval, piglets weaned per sow per year is the most

8 important measure of swine fertility ([email protected] 2015). Piglets weaned/sow/year is calculated by total number of piglets weaned divided by total sows and mature gilts in the barn.

The goal should be 22 or more piglets raised per sow per year. In other words, potential problems should be considered if the target is not reached (King G 2002).

Currently, most Canadian breeding stock pig farms use EBV (estimated breeding value) for reproduction traits to measure and directly compare fertility of sows and boars in the same breed in the same herd or even across-herds (CCSI 2012). EBV of each boar and sow is obtained by using best linear unbiased prediction (BLUP). It is calculated based on the combination of many factors including the heritability of the trait, all information (individual animal and its relatives) available for each boar and sow, the genetic level of the herd, genetic trend, and non- genetic factors such as management groups. EBV, a statistical numerical prediction, predicts the genetic value of an animal that will be passed on to its offspring. In theory, only half of each animal’s genes will be passed on to its offspring. For example, a pig with EBV of +3 is expected to produce progeny that will have, on average, 1.5 more piglets than progeny of a pig with an

EBV of 0. However, EBVs are expressed in relative not absolute units. Therefore, the previous example only means prolificacy of offspring of the first pig is better than with the second’s if they are raised in the same conditions. EBVs are calculated based on the performance of the animal itself and its relatives including ancestors, siblings and progeny. The accuracy of an animal’s EBV depends on the amount of records available and the relationships of the animal being evaluated. Among relatives, record of parents and progeny that have half of their genes in the evaluated animals has higher percentage of contribution to estimated EBV when comparing with other relatives such as grandparents and siblings (CCSI 2012).

An alternative measure of boar fertility is the DBE (direct boar effects) on total born litter size of his mates and the value is calculated and used by CCSI (“34th Centralia Swine Research

Update, Kirkton Ontario 28 January 2015” 2015; Tribout et al. 2000). DBE is a corrected EBV for total born litter size and hence is based on farrowing records. The boar will have a DBE computed only once he has sired his own litters. To generate EBV for total born litter size, the factors included in the statistical model are covariate factors (litter inbreeding, sow inbreeding, farrowing age within parity and lactation length of previous parity), fixed factors (management group or herd-year-season, parity, mating type*number of services and service sire breed) and random factors (litter of sow’s birth effect, permanent environmental effect on the sow and additive genetic effect of the sow). DBE for total born litter size has the statistical modeled with all factors as seen in EBV and an additional factor as a fixed factor (service sire). DBE is an excellent indicator for fertility of a boar. However, it is usually only precise after at least 10 litters are born. Boars are usually already culled when their first DBE is computed. Therefore, it is not so helpful for detecting the fertility problems occurring in the farm. But it is still very valuable to detect high and low fertility boars for research or to track down relatives of carriers of chromosome abnormalities.

STRUCTURAL VARIATIONS OF THE GENOME

Genomic variations are heritable differences in DNA sequences between individuals and are fundamental to evolution, adaptation and phenotypic divergence. The range of this variation starts from sequence variants at the single nucleotide level (SNP = single nucleotide polymorphism) to structural variations, some large enough to be visible under the microscope.

The latter, chromosome abnormalities, are large-scale reorganizations of the genome structure that generally encompass 5-10 Mb long DNA segments. The ability to visualize mitotic chromosomes led to the early recognition of differences among species and extensive comparative evolutionary investigations (Graphodatsky et al. 2011). Also, clinical cases of various phenotypes were identified as carriers of chromosome abnormalities and initiated the establishment of both human and animal clinical cytogenetic laboratories worldwide (a. Ducos et al. 2008; Gersen and Keagle 2013). Recent development of microarrays and sequencing technologies enabled a closer view of the genome and led to the identification of sub- microscopic structural variants - called copy number variations or CNVs -, that can range from

50bp – to several Mb (Feuk et al. 2006). Figure 1 represents the major classes of genome variations.

Figure 1. Representative examples of genomic variations and their size differences: (A) Chromosomal abnormality > (B) Copy Number variation > (C) SNP

Although the estimated 10 million SNPs in the human genome were thought to be responsible for the majority of genetic variation (“The International HapMap Project” 2003), the accumulated almost 2.5 million CNV in the Database of Genomic Variants represent an even larger portion of variations in the genome and seem to have at least similar significance in human health and disease (Zarrei et al. 2015). From a population genetics point of view these variants are also classified according to their occurrence in the given population, thus called

“polymorphism” (SNP, CNP) if they exist in greater than 1% frequency and “alteration” (SNA,

CNA) in case of rare (<1%) variants.

Chromosome abnormalities

Chromosome abnormalities are microscopically visible genome rearrangements, the largest ones among structural variations (Basrur and Stranzinger 2008b). According to the traditional classification there are numerical chromosome abnormalities and structural ones. The latter are sub-divided into reciprocal translocations, Robertsonian translocations (or centric fusions), inversions, deletions and duplications. Although numerical abnormalities are not necessarily changing the structure of the genome only the number of chromosomes, they could be considered as gross insertions in case of chromosome surplus or deletions when chromosomes are missing from the complement. These aneuploid conditions (trisomic - an extra chromosome or monosomic - a missing chromosome) represent a large genomic imbalance and in most cases are lethal although there are exceptions such as the human Downs, Edwards, Patau syndromes and cases of sex chromosome aneuploidies (A. I. Taylor 2008; Hassold and Jacobs 1984).

Structural abnormalities could also be characterized according to their effect on genomic balance

(of dosage sensitive regions), thus translocations and inversions could keep genomic balance

12 while duplications and deletions obviously lead to imbalance. Figure 2 represents the different types of structural chromosome abnormalities.

Figure 2. Types of chromosome abnormalities. (a) reciprocal translocation, (b) Robertsonian translocation, (c) paracentric inversion, (d) pericentric inversion, (e) duplication, (f) deletion (Modified from US National Library of Medicine) Chromosomal abnormalities are identified to be major contributors to infertility, miscarriages, congenital malformations and mental retardations in human, thus routine clinical cytogenetic screening programs were implemented from the early 1980’s. In order to facilitate genetic counselling, several large-scale human clinical cytogenetic studies have investigated the frequency of chromosome abnormalities and attempted to estimate the risk of developing disease phenotypes (Warburton 1991; Tonk et al. 2003; Menasha et al. 2005; Jacobs et al. 1992).

Van Dyke et al. (1983) reported that 0.32% of 6,595 human newborns was carrier of chromosome abnormalities; among those 21% was de novo. The frequency of reciprocal translocation, inversion and Robertsonial translocation was 0.15%, 0.06% and 0.11%, respectively (Van Dyke et al. 1983). The study of almost 15,000 newborns analyzed in the same institute over 15 years (Jacobs et al. 1992) estimated the total frequency of all types of chromosomal abnormalities to be 0.9%. Approximately 1/200 humans were carriers of balanced structural abnormalities of which 7% were de novo. These values are slightly higher than the estimate from almost 90,000 amniocentesis tests, which resulted in 0.18% balanced reciprocal translocations, 0.09% inversions and 1/6,800 deletions. An even larger study of 380,000 amniocentesis analyses found 1/2,000 de novo reciprocal, 1/9,000 Robertsonial translocation and

1/10,000 de novo inversions (Warburton 1991). The proportions of these cases that are associated with clinical symptoms are estimated to be 3-10%.

Just as described above in humans, spontaneous chromosome abnormalities occur in every other species’ population as well. Not surprisingly, veterinary cytogeneticists have paid special attention to investigating the chromosomes of the economically most important food- producing animals; among those the studies involving pigs are presented here. These efforts were centered on three main areas: 1. The prevalence of chromosomal abnormalities. 2. The effects of chromosome abnormalities, especially from the producers view. 3. Efforts to eliminate carriers.

Chromosomal abnormalities in pigs.

The domestic pig (Sus scrofa domestica, L) has 38XX or 38XY chromosome constitution

(figure 9). Twelve among the 18 autosomes plus the sex chromosomes are bi-armed while six autosomes are acrocentric with one arm. The analysis of the karyotype is quite straightforward

14 due to the not so numerous and well differentiated chromosomes and available standard nomenclature (Gustavsson 1988a).

Pigs have been systematically screened in France for more than 20 years. Using high resolution G-band technique on more than 7,700 samples of young boars intended for AI use, the frequency of chromosome abnormality in this population has been estimated as 0.47%. This frequency of chromosomal aberration is sometimes higher in some particular populations and reaches as high as 1.5% in the Dutch pig population ( a. Ducos et al. 2008) and 2.5% in the

Canadian pig population (Quach et al. 2010). For that reason, pig is the only domestic animal for which a precise prevalence of balance structural chromosomal abnormalities is available.

Although not on the same scale as the study in France, scientists of some other countries such as

Poland, The Netherlands, Hungary and Canada have conducted smaller population sampling, together contributing to the more than 180 different chromosome rearrangements reported in the literature. Among the structural chromosome abnormalities the reciprocal translocations surpass by far any other type of abnormality. So far, over 150 cases of reciprocal translocations, 22 cases of inversions, four of Robertsonian translocations and two duplications have been described

(Appendix I and II).

Figure 3. Spontaneous chromosome abnormalities occur in every swine population. The actual frequency of carriers is determined by the rate of de novo events, the presence of inherited abnormalities and successful eradication efforts.

The effects of chromosome aberrations

Unless there is deletion or duplication associated with chromosome aberrations, carriers of balanced chromosome rearrangements are viable and phenotypically normal. However, their reproductive efficiency does not stay untouched and their mating usually results in reduced total born litter size, thus they are termed hypoprolific or subfertile (Figure 4). The fact that almost

50% of hypoprolific boars carry balanced rearrangements underlines that chromosome abnormalities are a major cause of reproductive failure (Popescu et al. 1984; Popescu and Tixier

1984; Gustavsson 1990; A Ducos et al. 2007; Quach et al. 2010).

Figure 4. Chromosome abnormalities are a major cause of subfertility. The pink sow (normal) has a litter with 9 piglets while the red sow (carrier of chromosome abnormality) has a litter with only 5 piglets and 2 of them (red) carry the same chromosome abnormality. The explanation of total born litter size reduction lies in the gametogenesis of the carrier animals. The pairing of homologous chromosomes during meiosis is more complicated in case of chromosome abnormality carriers (especially in case of translocation or inversion). The sophisticated and well-regulated formation of bivalents should be modified to accommodate the reorganized and mostly heterozygous chromosome structures. This gives rise to alternative segregation products; some of them leading to gametes with unbalanced DNA composition

(Gustavsson 1990; D. a F. Villagómez et al. 2008). Carriers of a reciprocal translocation generally produce more chromosomally unbalanced sperm than normal or balanced ones (figure

5). Robertsonian translocations have little effect on the normal segregation pattern, while inversions could show a great variability in the percentages of unbalanced gametes (K. Massip et al. 2008; a. Pinton et al. 2005; K. Massip et al. 2009). This is in agreement with the general effect on total born litter size reduction of these scenarios. A normal egg fertilized by a chromosomally unbalanced sperm - or vice versa -, leads to the formation of non-viable unbalanced embryos that die early in gestation and thus result in reduced total born litter size.

Figure 5. Meiotic segregation pattern of a reciprocal translocation leading to chromosomal unbalanced gametes and embryos (Farndon 2015) Efforts to eliminate carriers

As the balanced carriers usually express normal physical, growth characteristics, phenotypic clues do not help their elimination before breeding decisions (A Ducos et al. 2007).

However, the carried chromosome rearrangement reveals its adverse effects when a carrier boar is used as a semen donor in AI. With wide application of the AI technique, a boar can serve hundreds of sows per month and consequently thousands of small size litters could be produced

17 per year with the inevitable economic loss. Depending on the chromosomes involved in the translocation, the average reduction of the total born litter size of a sow mated with a carrier boar ranges from 10% - 100% (A Ducos et al. 2007). It is estimated that an affected boar costs

$30,000-$60,000 until the reduction in total born litter size is first reliably detected on farm.

Within that period of time, the boar will have, on average, produced more than 1000 piglets and

50% of them will in turn carry the rearrangement (Gustavsson et al. 1983; a. Ducos et al. 2008).

We have calculated the projected economic loss due to the total born litter size reduction to reach as high $1.6M in Ontario and $12M in Canada, yearly. Although, the AI favourably hastens the spread of superior genetics of boars, it also causes detrimental impacts through the rapid dissemination of chromosome abnormalities from a carrier boar to its numerous progeny (Figure

6). For all of these reasons, development of a program for systematic screening of chromosome abnormalities with appropriate follow up studies of relatives (especially parents to determine de novo status) and counselling the optimal eradication strategy would be beneficial in Canada where the prevalence of alterations is still very high (>2%).

Figure 6. Chromosome abnormalities are transmissible to the progeny. The pink sow (normal) has a litter with 9 piglets while the red sow (carrier of chromosome abnormality) has a litter with only 5 piglets and 2 of them (red) carry the same chromosome abnormality. Again, if these two carrier piglets are used for breeding, the chromosome abnormality is transmitted to their progeny.

Copy number variation

Copy number variations (CNVs) are structurally unbalanced genomic variations

(duplication and/or deletion) in the size range below that of chromosomal abnormality (50 bp –

Mb). CNVs have been discovered in most mammalian species e.g. cover about 0.68-1.07% of the cattle genome, ~4% of the dog genome and 4.8-9.5% of the human genome (Zarrei et al.

2015). In the case of humans, thousands of individuals were investigated with the highest resolution analytical techniques and the most precisely mapped reference genome. Data is constantly deposited to the Database of Genomic Variants (DGV) (MacDonald et al. 2014), that now contains more than 6.5 million entries. The curated number of CNVs reaches 2.5 million and about 200,000 CNVRs (that refers to CNV Regions where the CNV is detected in at least two individuals). Chromosomal distribution of CNVs is uneven and varies between 5-55% being the lowest on chromosome 3 and 18 and the highest on chromosome 19 and 22, however in overall they cover ~78% of the human genome. Also, annotating CNV positions with the key functional elements of the genome shows that 94% of transcripts and 86.5% of microRNAs overlap with CNVs (MacDonald et al. 2014), thus representing a larger proportion of the genome than that involved in SNPs.

Deletions are rare and under negative selection in healthy individuals within genes with essential functions or in the ones showing strong evolutionary conservation (Lin et al. 2015).

However, they are frequent in promoters, non-coding genes and genes with non-essential functions. Many of these non-essential genes are members of gene families, where the other members can substitute the deleted function or genes that act late in the lifespan and affect age- related phenotypes such as osteoporosis, multiple sclerosis or autoimmune disease (McLean-

Tooke et al. 2007; Barcellos et al. 2000).

Duplications on the other hand, more frequently affect genes with critical functions, they seem to be less pathogenic and often under positive selection that could lead to the fixation of multicopy genes or gene families (Zarrei et al. 2015; Hurles 2004). An example is the human salivary amylase (AMY1) gene that hydrolyzes starch and was identified to exist in higher copies in populations with high-starch diet than in low-starch consuming ones. Most interestingly the amylase protein levels correlated with the changes of gene copy, thus could be considered as an example for positive selection and adaptation to the change of human diet (Carpenter et al.

2015).

The gene functions of common CNV (that exist in healthy individuals) covered regions could provide insights into their effects on normal phenotypic variation. These results show that olfactory receptors, pathways of drug, steroid, sucrose and starch metabolism are frequently involved in CNVs (Radke and Lee 2015; Zarrei et al. 2015). Moreover, CNVs are rare in regions that function in signal transduction, regulation of gene expression, cell cycle, development and differentiation, thus these rare variants likely contribute to serious pathologic conditions

(Stankiewicz and Lupski 2010). Rare or de novo CNVs are often behind the onset of genomic disorders, such as Prader-Willi syndrome, Duchenne muscular dystrophy, Charcot-Mary-Tooth disease or Williams’s syndrome, to name a few. DNA copy number alterations are signatures of cancer development and interestingly high correlation was found between the copy number and gene expression that suggests functional roles of CNV in tumor progression (Stankiewicz and

Lupski 2010; B. S. Taylor et al. 2008). The various mechanisms by which CNVs could affect gene function are visualized in Figure 7.

Figure 7. CNVs could affect phenotype in various ways (from Feuk et al. 2006). Structural variants can be benign, can have subtle influences on phenotypes (for example, they can modify drug response), can predispose to or cause disease in the current generation (for example, owing to inversion, translocation or microdeletion that involves a disease-associated gene), or can predispose to disease in the next generation. On the basis of their proximity to structural variants, genes might be influenced in several ways. (a) Dosage-sensitive genes that are encompassed by a structural variant can cause disease through a duplication or deletion event (upper panel; a deletion is shown here). Dosage-insensitive genes can also cause disease if a deletion of the gene unmasks a recessive mutation on the homologous chromosome (lower panel). (b) Genes that overlap structural variants can be disrupted directly by inversion (upper panel), translocation or deletion (lower panel), or copy-number variant breakpoints (not shown), which leads to the reduced expression of dosage-sensitive genes. Breakpoints that disrupt gene structures can also lead to the formation of new transcripts through gene fusion or exon shuffling (not shown). (c) Structural variants that are located at a distance from dosage-sensitive genes can affect expression through position effects. An example is shown in the upper panel. A deletion of important regulatory elements can alter gene expression; similar effects could result from inversion or translocation of such elements. Alternatively, deletion of a functional element could unmask a functional polymorphism within an effector (lower panel), which could have consequences for gene function. (d) Structural variants can function as susceptibility alleles, where a combination of several genetic factors is required to produce the phenotype. This is illustrated by an example of two structural variants that, individually, do not produce a phenotype. However, in combination they contribute to a complex disease state.

The first insight into CNV content of the porcine genome resulted from a study of only four chromosomes, as genome-wide array tools were not yet available (Fadista et al. 2008).

Recently, a genome-wide 720k resolution comparative genomic hybridization array (aCGH) was used to investigate 12 pigs from many different breeds and found 259 CNVRs (Y. Li, Mei, et al.

2012). Various Chinese and European breeds were tested on a 2.1M high-density arrayCGH by

Wang et al. (2014), describing 1344 CNVRs with 1.7% genome coverage. Ramayo-Caldas et al.

(2010) have identified 49 CNVRs in 55 pigs. A study of nearly 1700 pigs from 18 populations described 565 CNVRs and 1.8% genome coverage (C. Chen et al. 2012). CNVRs covering the largest fraction of the genome (560Mb or 26.2%) were identified in 585 Large White x Minzhu pigs by Wang et al. (2013). Fernandez et al. (2014) investigated Iberian pigs and found 65

CNVRs with a total length of 9.7Mb that equals 0.33% of the genome.

As there is not a collective porcine CNV database – similar to the human DGV – it is not easy to estimate the overall CNV-covered portion of the pig genome. However, based on the studies using the SNP60k chip, 16.08% of the porcine genome is involved (Fernández et al.

2014). Also, there appear to be breed-specific differences in CNVs in pigs which are consistent with a report in cattle (G. E. Liu et al. 2010). Specifically, Chinese indigenous breeds show more genetic variations than western breeds (Y. Wang et al. 2014).

Additionally, descriptive investigations of this level of genome variability, have been assessed in some studies and are associated with quantitative traits (QTL) related to carcass length, backfat thickness, abdominal fat weight, length of scapula, intramuscular fat content of longissimus muscle, body weight at 240 days, glycolytic potential of LD, mean corpuscular hemoglobin, mean corpuscular volume and humerus diameter with CNVRs and seven candidate genes potentially affecting these traits (Jiying Wang et al. 2014; Jiying Wang et al. 2013; Y.

Wang et al. 2014). The BTN1A1 and F5, genes related to milk-lipid droplets and preterm delivery in humans were found in CNVRs of the pig genome (Hao et al. 2004; Ogg et al. 2004).

Fowler et al. (2013) attempted to identify candidate CNVs associated with fatness (Fowler et al.

2013).

Most importantly, as mentioned above ~50% of the hypoprolific boars are carriers of chromosome abnormalities but the limited resolution of cytogenetic investigations might have left minor rearrangements hidden. It can be argued that the remaining 50% of the hypoprolific boars may potentially carry smaller genome rearrangements, such as deletions or duplications that result in CNVs. According to our knowledge there is no study that investigated the possible role of CNVs in low fertile boars.

Considering the scope of this thesis in investigating genomic variations related to fertility, this chapter cannot be complete without mentioning the role of the Y-chromosome and the unparalleled situation of multicopy genes within it. The male specific Y-chromosome due to its restricted recombination capability - with the X-chromosome -, evolved into a specialized

DNA structure that contains essential and unique genes for maleness, but it is also enriched in highly variable repeated elements (Skaletsky et al. 2003). Intrachromosomal recombination events generate a remarkable variability in structure, content and sometimes express serious consequences on fertility (Figure 8) (Jobling 2008). One such variable region is the

Azoospermia Factors (AZF) region built from almost identical sequence elements that surrounds genes that function in spermatogenesis. Frequent deletion within this region causes the loss of these genes and results in infertility (Vogt et al. 1996). Also, there are amplified genes that exist in many apparently full length and functional copies, such as HSFY and TSPY. While both of these genes have predicted functions in spermatogenesis, the former keeps its high but stable

23 copy number and TSPY is most probably the gene having the highest and most variable copy number among all mammalian genes (Hamilton et al. 2009b; Hamilton et al. 2012). Investigation of these barely known Y-chromosomal CNVs in relation to their effect on boar fertility is certainly a challenging opportunity.

Figure 8. CNVs on the human Y-chromosome (Jobling, 2008). Green spots mark copy number gains while red spots deletions. The height of signals reflect frequency of CNVs identified

HOW TO DETECT STRUCTURAL VARIATIONS

A vast array of methods have been developed to detect genomic variation of different sizes, however, the following review will be limited to techniques that are commonly used and/or have been applied to detect genomic variations in pigs.

Cytogenetics Classical cytogenetic techniques visualize the chromosomes under light microscopy, thus providing a genome-wide snapshot, although with limited resolution. The chromosomes are

24 prepared from dividing cells arrested at the metaphase stage of the mitotic cell cycle. Even the simplest Giemsa staining allow the experienced observer to recognize numerical abnormalities, such as sex chromosome aneuploidy or mosaicism that are of immediate clinical use as the most frequent abnormalities in humans or horses (Lear and Bailey 2008). Identification of the individual chromosomes has been made possible by the development of various banding techniques, among the most frequently used is G-banding. Here a controlled digestion of the chromatin with trypsin followed by Giemsa staining generates the light and dark banding pattern

(Seabright 1971). During the 1980s and ‘90s international expert groups were formed to determine the standard karyotypes of most domestic animals, including the pig (Gustavsson

1988a) (Figure 9).

Figure 9. Standard karyotype of the pig (Gustavsson, 1988)

Comparison of the G-banding pattern to the standard karyotype can help with the identification of numerical or structural chromosomal abnormalities including reciprocal translocation, Robersonian translocation, inversion, duplication and deletion (Iannuzzi L, King

WA 2009; Bayani and Squire 2004). Thus, the G-banding - due to its simplicity, low cost and the available standard karyotypes – became the essential routine cytogenetic method to investigate human or some domestic animal cases (a. Ducos et al. 2008). Many other chromosome banding methods were developed, providing insight into different structural elements of chromosomes, but these are used less frequently for routine clinical investigations. These include QFQ

(quinacrine fluorescence (Q)-banding) (Caspersson et al. 1970), RBG (reverse (R)-banding with

Giemsa staining) (Dutrillaux and Lejeune 1971), RBA (reverse (R)-banding with acridine orange staining) (Dutrillaux 1973), C-banding (centromeric heterochromatin staining) (Sumner 1972) and NOR staining (nucleolar organizer region) (Mikelsaar and Schwarzacher 1978).

Although the complete view of the chromosomal organization of the genome and its alterations represent an inevitable advantage of investigating cytogenetic preparations under the light microscope, the size of the recognizable G-bands limits the resolution of this analysis. At the routine G-band level, the chromatin is so condensed that only deletions or duplications larger than 5-10Mb DNA segment can be recognized, as well as translocation breakpoints mapped at this precision (Shaffer and Bejjani 2004; Carter 2007).

Molecular cytogenetics

To overcome the limited resolution of classical cytogenetic methods and to detect smaller genomic variations, molecular cytogenetics was born. The use of labeled DNA probes hybridized to complementary chromosome segments and visualized by fluorescent signal amplification techniques allowed the detection of sub-microscopic targets. Thus, fluorescent in situ

26 hybridization (FISH) became the tool for gene or translocation breakpoint mapping, although the sometimes-laborious generation of appropriate and specific probes restricts its routine use except for human clinical cases (Basrur and Stranzinger 2008a; Shaffer and Bejjani 2004).

A modification of the general FISH technique is called comparative genomic hybridization (CGH) that detects the structural alteration of a case genome relative to the reference DNA of a healthy subject (Figure 10). CGH was originally developed to investigate the frequent and large scale genome rearrangements representative of tumors, especially in cases of solid tumors where chromosome preparation is particularly difficult (Kallioniemi et al., 1992).

During CGH the test DNA (e.g. tumor tissue) and the reference DNA (e.g. normal tissue) are differentially labeled (green/red), mixed in equal ratio and simultaneously hybridized onto normal metaphase spreads. The ratio of green vs. red fluorescence along the length of each chromosome reflects the DNA copy number (loss or gain) in the test sample in relation to the reference one. Chromosomal regions without copy number changes fluoresce in yellow, while an excess of green fluorescence is indicative of a DNA gain, and excess of red fluorescence is indicative of DNA loss. The resolution of CGH is in the range of the highest resolution (and very cumbersome) banding techniques, typically 3-5Mb, as it is limited by the capabilities of the detection camera systems (Carter 2007). Moreover, CGH can only detect copy number alterations, but not translocations and inversions.

Figure 10. CGH and aCGH (“Genetics of Auditory Disorders” 2015)

High resolution arrays for the investigation of Structural Variations

Increased resolution of the CGH could be achieved by replacing the reference metaphase chromosomes with genome wide microarrays in the so-called arrayCGH or aCGH technique

(Ishkanian et al. 2004; Oostlander, Meijer, and Ylstra 2004; Freeman et al. 2006). Thus, the use of a large number of mapped marker points on the glass assay surface and the short DNA sequences served as hybridization targets both resulted in a highly capable platform, able to detect small CNVs (Figure 11).

Various DNA probes sources were used to build arrays from large-insert clones such as bacterial artificial chromosomes (BACs) (80-200kb in size), fosmid or cosmid (1.5-4.5kb) to small genomic PCR products (100bp-1.5kb) and oligonucleotides (25-80bp) (Solinas-Toldo et al.

1997; Pinkel et al. 1998; Mantripragada et al. 2004; Dhami et al. 2005; Fiegler et al. 2007).

Hence, oligonucleotides offer the highest potential resolution for aCGH. Over the past several years, one could witness a tremendous development of the array preparation technologies.

Agilent and NimbleGen (Roche) are the two biggest companies providing commercial oligonucleotide arrays. Agilent uses ink-jet technology and can achieve 1 M (million) oligonucleotides per array, whereas NimbleGen uses photolithography technology and can achieve 6.3 M oligonucleotides per array. Przybytkowski et al. (2010) reported that 1 M arrays of

Agilent could detect copy number alterations as small as ~8 kb. While Fadista et al. (2010) reported that 6.3 M arrays of NimbleGen could identify the CNVR in the cattle genome as small as 1.7 kb. The highest resolution aCGH for the pig is a custom designed 2.1 M aCGH by

NimbleGen. It covered all 20 pig chromosomes, containing 2,167,769 oligonucleotide probes with median and average intervals of 900bp and 889 bp, respectively, and can detect CNVs as small as 3.37kb (Jiying Wang et al. 2014).

In brief, aCGH is a powerful technique able to screen the whole genome at kb resolution and sensitive enough to detect single copy changes (Ishkanian et al. 2004; Carter 2007; J. H. Lee and Jeon 2008).

Recently, high-density SNP genotyping arrays became an alternative tool to detect genome-wide CNVs. Especially, as the SNP genotypes are widely used for genome wide association studies (GWAS) in human or animal models and it is fundamental for implementing genome selection for breeding improvement in dairy cattle with similarly promising future

29 applications in other farm animals too. Thus, SNP genotypes of large populations are already available and the same experimental data could be simultaneously mined for CNV information.

Either one or both oligonucleotide probes detect the di-allelic SNP at a given locus generating red, green or yellow signals, representing AA, BB or AB genotypes. The feature that makes

CNV detection possible is that the normalized total probe fluorescence intensity is proportional to the amount of DNA bound to that locus, thus the copy number (Gregory M. Cooper and

Heather C. Mefford 2011) (Figure 11).

Signal intensity

Genotype A Genotype B Genotype AB

Figure 11. Schematic view of the SNP genotyping array. The signal at each locus corresponds to the amount of DNA

The higher the density of the SNP array, the higher the resolution coverage that could be achieved. At present, the commercial human whole genome SNP chip contains over 2.7 million

SNP markers (Halper-Stromberg et al. 2011). For pig, quite far behind the human, the highest density chip in the market used for CNV analysis is the PorcineSNP60 BeadChip, released by

Illumina in 2009 (Ramos et al. 2009). It contains more than 64,232 SNPs across the entire porcine genome and its mean and median gap size is 43.4 kb and 27.9 kb, respectively. Due to the uneven distribution of SNPs within the pig genome, although SNP 60k chip can detect some small CNVRs as of 5.03kb, other areas of the genome are not covered that densely, several

CNVs could be missed using this technique. Recently a 700k Porcine SNP array was released and welcomed by the research community interested in exploring genome-wide CNVs.

CNV detection from whole genome sequence information

The development of new generation sequencing (NGS) technologies allows one to sequence full-length individual genomes and assemble the raw sequence to the reference within days. The tremendous information content of the sequence reads could also be used to deduct copy number status in various ways. Figure 12 shows two scenarios. The read-depth analysis compares the observed read coverage to the expected read distributions, thus insertions are identified where significantly more reads are generated, and similarly, deletions detected in regions lacking read coverage. The paired-reads in the second scenario could identify deletion or insertion events based on how they map to the reference genome (Valsesia et al. 2013).

Although the cost of sequencing is rapidly decreasing, thus become more affordable for large-scale studies, currently it is the method of choice for only human investigations. This is not only the case due to the overall cost of experiments, but also by the limited precision of the available reference genome assembly. Nonetheless some pioneer NGS studies were performed in pigs to provide insights into evolutionary and population genetics aspects of this species (Rubin et al. 2012; Paudel et al. 2013).

Figure 12. CNV detection from Whole Genome Sequencing data (Valsesia et al. 2013)

Independent validation of high throughput assays with small-scale techniques

Since the above methodologies target the whole genome and predict CNV status from differential signal intensity values, the small-scale validation using an independent method is a general requirement (J. H. Lee and Jeon 2008).

This can be done using quantitative real time polymerase chain reaction (qPCR) that goes beyond studying gene expression and is frequently used to verify the aCGH and SNP genotyping array results as well (Ginzinger 2002; Tse et al. 2005; Weaver et al. 2010). qPCR differs from the traditional PCR in that it allows real-time detection of the amplification curve by measuring the amount of DNA in each PCR cycle (Ginzinger 2002)(Figure 13).

Figure 13. Real-time quantitative PCR amplification curves

The quantity of target amplicon can be determined as the ratio to a single copy reference gene (Ginzinger 2002; Tse et al. 2005; Weaver et al. 2010). qPCR has been shown to be sensitive enough to detect single copy changes (Weaver et al. 2010). As qPCR cannot be used to examine

CNVs in a genome-wide fashion, it is usually applied to validate a selected number of loci in large-scale studies. Although qPCR in ideal conditions is an optimal tool to validate CNVs, its relative nature often requires challenging calculations.

Recently, another technique called digital droplet PCR (ddPCR) has been developed by Bio-

Rad and provides absolute quantity measurements. The method is similar to a qPCR assay,

33 although the reaction mix is partitioned into 20,000 uniform nanoliter volume droplets where the amplification is spatially separated. Depending on the concentration of the target DNA more or less droplets become amplification positive and are precisely detected by a microfluidic device. ddPCR carries the advantage of absolute quantitation using end-point detection instead of the efficiency dependent amplification curve-based relative calculation of qPCR (Manoj 2014).

Many publications used ddPCR as a diagnostic tool for conditions such as hepatitis B virus copy number (J.-T. Huang et al. 2015), 22q11.2 deletion syndrome (Pretto et al. 2015), various human

CNVs (Marques et al. 2014; Bharuthram et al. 2014) etc. but it has not been applied to detect porcine gene copy numbers.

Figure 14. Droplet digital PCR (Bio-Rad Bulletin #6407)

RATIONALE

A previous study of chromosome abnormalities in Canadian breeding boars revealed that about 50% of the hypoprolific boars raised for service in AI were carriers of chromosome abnormalities (Quach et al. 2010). This study depended heavily on the GTG-banding technique for the detection of chromosome abnormalities. However, GTG-banding alone is not sensitive enough to detect minor changes in the genome of boars (Basrur and Stranzinger 2008a). It can be argued that the remaining 50% of the hypoprolific boars may potentially carry very tiny chromosome abnormalities such as deletions, duplications that result in CNVs or small translocations which cannot be detected by the GTG-banding technique.

Genome-wide selection is currently revolutionizing livestock breeding programs around the world (Meuwissen, Hayes, and Goddard 2001). Numerous molecular genetic markers have been identified as valuable aids for selecting prolific boars. Up to now, most of genetic markers are based on SNPs, but no CNVs have been found. Although CNV studies on pigs are quite well established (C. Chen et al. 2012; J. Wang et al. 2014; J. Wang et al. 2012; Clop et al. 2012;

Fadista et al. 2008; Y. Ramayo-Caldas et al. 2010; Jiang et al. 2014), they are still just on the beginning stage. Studies found lots of CNVRs, and many of them contain various genes. Some studies tried to associate some traits such as meat, carcass quality and fat to CNVRs and then identified candidate genes which potential affect those traits (Fowler et al. 2013). No attempt has been made to associate fertility traits to CNVRs for potential genes related to fertility. No further studies have been reported to characterize the number of copies of any genes on CNVRs or compare copy number with related traits in pigs.

TSPY is a gene on the male specific region of the Y chromosome (MSY). It is a multi- copies gene in many species and also found in pig (Quilter et al. 2002). It was reported copy number differs between species with two copies in rats, 6 copies in chimpanzee, 20-60 copies in human and as much as 200 copies in cattle. Even more interestingly, copy number of this gene has a positive correlation with fertility in cattle. Hamilton et al., 2011 showed that when bulls were grouped as high or low fertility, those in the high fertility group had a significantly higher average TSPY copy number than bulls in the low fertility group. How many copies of TSPY that pig have and whether there any variations of TSPY copies associated with fertility in pig is still unknown.

The hypothesis of our study is as follows: subfertile or infertile boars are carriers of chromosome abnormalities and/or CNVs. Also, specific large (>1000 bp) genomic variations

(CNV) are associated with total born litter size. The following objectives will be used to address this hypothesis:

Objectives 1: Screen for and identify chromosomal abnormalities in boars using cytogenetic

techniques

Objectives 2: Use the SNP 60K array technology to compare boars of known low or high fertility

to identify consistent differences related to CNVs

Objectives 3: Investigate the copy number of TSPY (testis specific protein Y encoded) gene

In Canada, previous estimates of the frequency of chromosome abnormalities was 2.5% based on a small group of young boars (40 samples) (Quach et al. 2010). In Vietnam, no cytogenetic analysis has ever been applied in this pig population. Therefore, for the first objective, screening for chromosome abnormalities was carried out in over 700 young boars

36 from many different farms across Canada and 3 farms in the south of Vietnam to have a more accurate number to estimate the frequency of chromosome abnormalities in those pig populations. All available fertility data for carriers were also collected to estimate the effect of each rearrangement on the reproduction of that carrier.

The Canadian Centre for Swine Improvement Inc. has genotyping projects underway and provided access to their database of thousands of 60k Porcine SNP panel genotypes of breeding animals and corresponding fertility data of each animal, which facilitated our second objective.

Genome-wide studies on CNVs were be made on low and high fertility boars which assumed without chromosome abnormalities to find the associate between copy number variation regions and fertility.

Finally, special attention was be paid to CNV of the TSPY gene on the Y chromosome due to its essential roles in gametogenesis, and hence may relate to fertility of a boar.

CHAPTER ONE The prevalence and consequences of chromosome abnormalities in Canadian and Vietnamese breeding pig populations

INTRODUCTION

Pork is the most consumed meat in the world (FAO 2014) and with the expansion of the global population, the need for efficient large-scale breeding is more important than ever. Factors such as fertility and total born litter size have major impacts on the economics of pork production

(Max F. Rothschild 1996). Minimizing embryonic loss during pregnancy is key to improving total born litter size and it is well known that chromosome abnormalities are considered to be among the major etiologic factors in the induction of malformations and early embryo mortality in the domestic pig (Gustavsson and Settergren 1984). The negative impact of chromosome abnormalities on farm animal reproduction has led to the establishment of cytogenetic screening programs in many countries (a. Ducos et al. 2008). The largest of these programs was initiated in

France over 15 years ago and led to the regular testing of all young AI boars in the country. This resulted in not only the most precise estimate of the prevalence of structural chromosomal abnormalities in a farm animal species, but also in the appreciation of economic benefits of the eradication of carriers by the breeding companies. The overall rate of chromosomal rearrangements in hypoprolific boars in service is close to 50% and its prevalence in young untested AI boar candidates is 0.47% (Alain Ducos et al. 2007). This most probably reflects the rate of “de novo” abnormalities, as in other less intensively tested populations their frequency is usually much higher according to reports from Poland (1%), the Netherland (1.5%) and Spain

(3.4%) (Danielak-Czech et al. 1997; Rodriguez et al. 2010). In Canada, screening for chromosome abnormalities has not yet been seriously considered by farmers. Only a few cases identified in Canadian pig populations were published including rcp(Xp+;14q-) (Singh et al.

1994; Neal et al. 1998), rcp(1;6)(p22,q12), rcp(10;13), and rcp(9;14)(p24;q27) (Quach et al.

2010). Furthermore, cytogenetic testing has never been applied to Vietnamese pig populations before.

Therefore, the aims of the present study were to estimate the frequency of chromosome abnormalities in pig populations in Canada and Vietnam, where cytogenetic screening programs have not been broadly implemented, and evaluate the effect of each chromosome abnormality on the fertility of the carriers.

MATERIALS AND METHODS Animals A total of 732 young boars from various Canadian farms and 43 randomly chosen breeding pigs from Vietnamese farms were included in this study. The tested population consisted of 4 different breeds: Duroc (344), Landrace (245), Yorkshire (162) and Pietrain (24).

All Canadian samples were young boars intended to be used for AI. However, Vietnamese samples were breeding boars and sows that were used on the farms, but their fertility data were unavailable at the time of blood collection. In addition, relatives and offspring of boars that were identified as carriers of chromosomal abnormalities were also included in analyses.

Chromosome analysis Whole blood cultures were set up as previously described (Quach et al. 2010). Briefly,

1.0ml of heparinized blood was cultured at 37°C in a medium consisting of 8.0 ml of RPMI 1640 medium (Gibco) with Lglutamine (0.3g/l), 75μl phytohaemagglutinin (PHA; M form, Gibco),

1.5 ml fetal bovine serum (FBS; Gibco) and 20μl of an antibiotic solution (10,000 IU penicillin,

10 mg/ml streptomycin sulphate). Seventy-two hours after the initiation of culture, 40μl colcemid (10μg/ml, Gibco) was added to the culture medium and incubated for 20 minutes at

37°C to arrest mitotic cells in metaphase. Then cells were then treated with a hypotonic solution

(0.075 M KCl) for 20 min at 370C, followed by fixation in methanol: acetic acid (3:1).

For some offspring that were mummified and stillborn, fibroblast cultures from the tails were also initiated as soon as the tails arrived in our laboratory, no later than 72 hours after birth.

The cells were cultured in a T25 flask containing medium consisting of DMEM (Gibco) with

20% fetal bovine serum (FBS; Gibco), L-glutamine (0.3g/l) and 0.4% of an antibiotic solution

(10,000 IU penicillin, 10 mg/ml streptomycin sulphate) until completely confluent. After confluence was reached, the cells were passaged to T75 flasks for 14 hours. Cells were observed

41 from 2 to 4 hours for the accumulating presence of dividing cells, and then 10μl/ml of colcemid

(Gibco) was added to the culture. After 3 hours, the cells were trypsinized and centrifuged at

350g for 6 minutes and harvested as described above. Fixed cells were spread on wet slides and air dried for one week. Slides were placed in 0.0125% trypsin for 1 to 3 minutes at room temperature to generate GTG banding (Seabright 1971) and stained with 5% Giemsa for 10 minutes.

For confirming rob(13;17), DAPI staining and C-banding were used. In DAPI staining, metaphase chromosome spreads were stained with 1ug/ml DAPI in PBS for 5 minutes. C- banding was performed according to the procedure described by (Sumner 1972). Briefly, slides were placed in 0.1N HCl for 30 minutes at room temperature, followed by a rinse with distilled

0 water then placed in 5% Ba(OH)2 at 50 C for 5-15 minutes. After incubation in Ba(OH)2, slides were removed and rinsed with distilled water. Then, they were placed in 2 X SSC at room temperature for 15 seconds, removed, rinsed with distilled water and again placed in 2 X SSC at

600C for 1 hour. Slides were placed in 2 X SSC for a third rinse at room temperature for 15 seconds and dehydrated by serial ethanol washing 30%, 70%, 96% (2 minutes; 2 minutes; 4 minutes, respectively). Finally, slides were stained with Giemsa 5% for 10-20 minutes, removed, rinsed with tap water and air dried.

Image capture and karyotyping: the stained slides were examined and digitally photographed individually using a light microscope at 100X objective and CCD camera connected to a computer equipped with Openlab 5.02 software. For each animal, 10-20 metaphase spreads were examined and at least three good quality metaphases per animal were karyotyped according to the international standard for the domestic pig (Gustavsson 1988a) using SmartType software.

Reproductive Data Analysis

Reproductive data for each identified carrier boar and their relatives (if available) were collected and compared with their herd averages, which were determined for the same period when the carrier boar was in service. Additionally, estimated breeding value (EBV) on total born litter size (CCSI 2012) or the direct boar effect (DBE) (Tribout et al. 2000) on total born litter size was also used for comparison, when available. DBE could also be obtained as a by-product of the national genetic evaluation for total born litter size (BLUP) by computing the average total born litter size of its mates, followed by correction for different factors, such as herd, period and breeding values of the mates. A Student’s t test was used to compare total born litter size, born alive, stillbirths and mummified fetuses for each translocation carrier boar with the corresponding data for the herd whenever the data was available.

RESULTS

Identification of Rearrangement Carriers

Among the 775 pig samples that were sent for karyotyping, 14 animals were identified as carriers of one of 12 different chromosome abnormalities. These rearrangements fell into 3 different types of structural abnormalities including reciprocal translocations (9), Robertsonian translocation (1) and inversions (2).

Below is a description of the rearrangements with their presumptive breakpoints based on the observed GTG banding pattern:

Case 1: a Landrace boar with rcp(1;5)(q21;q23)

GTG banding revealed two altered chromosomes: a metacentric corresponding to chromosome 1 shortened on its long arm (1q-) and a modified telocentric created by lengthening of the long arm of chromosome 5 (5q+) confirming that the exchange was reciprocal. The presumptive break points for this interchange at bands 1q21 and at 5q13 defined this as rcp(1;5)(q21; q13) (fig. 15).

Figure 15. GTG-banded karyotype of a boar carrying rcp(1; 5). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype

Case 2: A Yorkshire boar with rcp(1;15)(q211;q13)

GTG banding revealed a reciprocal exchange between chromosome 1 and 15. Most of

15q was transferred to chromosome 1 (1q+), while chromosome 15 remains as a small element

(15q-) which had the size similar to chromosome 18. The presumptive break points occurred in bands 1q211 and 15q13 defined this as rcp(1;15)(q211;q13) (fig. 16).

Figure 16. GTG-banded karyotype of a boar carrying rcp(1; 15). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype

Case 3: A Landrace boar with rcp(2;5)(p16;p11) A reciprocal exchange between chromosomes 2 and 5 was discernible mainly from the changes of the tips of these chromosomes: as chromosome 2 displayed a gain of a small band

(2p+) and the short arm of chromosome 5 missing one dark band (5p-). The remaining dark band on the tip of 5p was smaller than normal and seemed closer to the centromere than normal. Thus, we believe it was not 5p12 rather than half of 2p16. Therefore, the presumptive break points are at the middle of dark band of the short arm of 2 (2p16), and at the negative band of chromosome

5 (5p11), very close to the centromere (fig. 17).

Figure 17. GTG-banded karyotype of a boar carrying rcp(2; 5). Normal chromosomes are placed on the right, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the left in the karyotype

Case 4: A Landrace boar with rcp(3;4)(p15;q13) GTG banding revealed a break on chromosome 4 at q13, the fractured segment was translocated to the short arm of chromosome number 3, forming a large metacentric chromosome. A tentative breakage in the terminal G-negative band of the short arm of chromosome 3 (3p15) resulted a minute element which was translocated to the shortened arm of chromosome 4 making this a small metacentric chromosome similar in size to chromosome 12

(fig. 18).

Figure 18. GTG-banded karyotype of a boar carrying rcp(3; 4). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype

Case 5: A Yorkshire boar rcp(3;12)(p13;q15) Similar to case 3, this abnormality could not be detected by simple Giemsa staining technique. Examination of the chromosomes after GTG banding revealed alteration of chromosome 3 and 12. The breakpoints involved in this translocation seemed to be situated at

3p13 and 12q15 (fig. 19).

Figure 19. GTG-banded karyotype of a boar carrying rcp(3; 12). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype

Case 6: A Duroc boar with rcp(6;7)(p15; q13) This abnormality was very easy to observe by GTG banding pattern since the derivative chromosome 6 was much different from any chromosome showing the standard banding pattern.

However, the derivative chromosome 7 was much more difficult to recognize because its size and its pattern was very similar to chromosome 12. Break points presumably occurred in bands

6p15 and 7q13 defined this as rcp(6;7)(p15; q13) (fig. 20).

Figure 20. GTG-banded karyotype of a boar carrying rcp(6;7). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype

Case 7: A Duroc boar with rcp(7;15)(q13;q13) In this boar a large translocation between chromosomes 7 and 15 was observed, almost the whole long arms of both chromosomes. The breakpoints involved in this translocation seemed to be situated on bands 7q13 and 15q13 (fig. 21).

Figure 21. GTG-banded karyotype of a boar carrying rcp(7;15). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype

Case 8: A Landrace boar with rcp(8;13)(p21;q41) By analyzing the GTG banding the breakages tentatively occurred in a GTG negative bands in the central part of the short arm of chromosome 8 (p21), and in a GTG negative bands in the long arm of chromosome 13 (q41). The translocation resulted a derivative chromosome 8 that is longer than normal with two prominent dark bands on the tip; the derivative chromosome

13 is slightly shorter and lacking two dark bands at the end of the q arm (fig. 22).

Figure 22. GTG-banded karyotype of a boar carrying rcp(8;13). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype

Case 9: Two Duroc boars with rcp(12;14)(q15;q23) In two boars, most of chromosome 14 was translocated to the q arm of chromosome 12.

The breakage occurred at the very end (q15) of chromosome 12. While chromosome 14 broke at the region just below the dark G-band of this proximal segment (14q23). This translocation appears to be rcp(12;14)(q15;q23) (fig. 23).

Figure 23. GTG-banded karyotype of a boar carrying rcp(12;14). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype

Case 10: A Landrace boar with rob(13;17) The centromeric fusion of two acrocentric chromosomes 13 and 17 resulted in one submetacentric chromosome with a size similar to chromosome 1 (fig. 24).

Figure 24. Boars carrying rob(13;17). Arrows indicate the derived chromosomes. (a) and (b): GTG banded karyotypes of boars carrying heterozygous and homozygous rob(13;17), respectively. (c): C band of a boar carrying heterozygous rob(13;17). (d) DAPI stain of a boar carrying homozygous rob(13;17)

Case 11: A Duroc boar with inv(8)(q11;q25) GTG banding showed that there was a peculiar appearance of chromosome 8. The structure of the other chromosomes was normal. On the long arm of the abnormal chromosome

8, close to the centromere, the presence of two dark bands instead of one dark band (8q12) was observed. We also observed the presence of two dark bands at the tip of the derivative chromosome 8 instead of one dark band (8q26). The tentative double breaks occurred in the light band of the long arm (8q11) and in 8q25 light band, followed by a paracentric inversion. The chromosomal rearrangement could be described as inv(8)(q11;q25) (fig. 25).

Figure 25. GTG-banded karyotype of a boar carrying inv(8). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype Case 12: Two Landrace sows with inv(6)(q23;q35) In two sows, the structure of one arm of chromosome 6 was normal, but the two dark bands normally situated at the end of the q arm were moved proximal to the centromere on the derivative chromosome 6. It was hypothesized that one break on the long arm close to the centromere (q23) and another break at the end of the long arm (q35) followed by an inversion of the paracentric part of the chromosome had occurred. The chromosomal rearrangement could be described as inv(6)(q23;q35)(fig. 26).

Figure 26. GTG-banded karyotype of a boar carrying inv(6)(q23;q35). Normal chromosomes are placed on the left, and their altered counterparts (derived) chromosomes indicating the presumed break points (arrows) are on the right in the karyotype Origin and dissemination of rearrangements The dam and/or sire of 5 rearrangements (case 4, 6, 8, 9 and 10) were available for cytogenetic investigation to determine whether the translocation was inherited or had occurred

“de novo”. In 3 out of 5 cases (case 4, 9 and 10), the abnormality was inherited from the dam, while the other 2 rearrangements (case 6 and 8) were a de novo event that was confirmed by the presence of only normal karyotypes in both parents.

In case 4, rcp(3;4), 1 full sister and 3 half-sisters of the carrier boar were identified as carriers as well. In case 7, rcp(7;15), we were able to karyotype 1 full-sister and 2 paternal half- sisters and all 3 sows were normal. Nevertheless, the sire and dam of rcp(7;15) carrier were no longer available and were therefore not tested and the inheritance or “de novo” status could not be determined. In case 12, surprisingly, the same previously unreported rearrangement, an inversion in chromosome 6, was found in two unrelated (different parents and grandparents)

Vietnamese Landrace sows.

Since semen from three carrier boars were used in services prior to the discovery of the abnormalities, it was possible, in those instances, to sample the progeny and study the

52 dissemination of chromosomal abnormalities to the offspring of these carrier boars. In case 5, rcp(3;12), karyotyping of 23 (5 males and 18 females) randomly selected offspring resulted in the identification of 7 females, which inherited the abnormality from their sire; In this instance the transmission rate to progeny was 30%. In case 7, rcp(7;15), karyotyping of 15 (7 males and 8 females) randomly selected offspring resulted in the identification of 2 males and 2 females, as carriers of the abnormality, and resulted in a 27% transmission rate to progeny. These transmission rates were based on partial litters and had statistical difference from the expected rate (50%).

In case 10, rob(13;17), an average of 36% of the piglets from two litters were identified as carriers when carrier boars were mated with normal sows. A total of 55% of the carrier piglets were female and 45% were male. Screening of two litters (29 piglets) resulting from the mating of carrier boars mated with carrier sows, 77% of the animals were carriers; 49% were heterozygous and 28% were homozygous. Among carrier piglets, 58% were female and 42% were male. These matings yielded surprisingly high total born litter sizes (13 and 16 piglets).

Reproductive performance of carriers All carrier boars were destined for AI centers and the sows were being used in farms as breeding sows; they were healthy and had a normal phenotype. Semen quality and quantity of identified carrier boars had been tested and ranged in the average or above average of the herd.

Among 12 cases, 10 cases (except case 1 and 11) had been used and had some fertility data available. Average total born litter size for cases 2 to 10 were as follow: 10.8, 7.2, 9.2, 10.7,

10, 6, 8.4, 6.8 and 12.7, respectively, and of case 12 was 7.4 piglets per litter. Nine cases showed a reduction in total born litter size ranging from 4% to 46% in comparison to the herd average.

Surprisingly, case 6 showed a slight increase in total born litter size of +7.5%, but it was

53 calculated based on 4 litters only. Reductions on born alive was also seen in all cases (5 cases) for which data was available. Moreover, the numbers of stillborn and mummified piglets were also lower in cases 2, 4, 5 and 8 (reciprocal translocations) herd averages for these 4 reciprocal translocations (table 1). These differences were statistically significant in most cases (Table 1).

In contrast to reciprocal translocations, the number of stillborn and mummified piglets born from animals with Robertsonian translocations (rob(13;17)) were increased, although not significantly.

DBEs were also available for cases 2, 4, 5 and 8 which were -4.82, -2.67, -4.4 and -4.39, respectively.

Table 1. Fertility data of carriers. L=Landrace; Y=Yorkshire; Mat= maternal; Dn= de novo; ***=p<0.001; **=p<0.01; *=p<0.05; ns: no statistical difference when comparing with the same herd.

Case Abnormality Breed Origin Number Average % % % % DBE/EBV # of litters little different different different different size TB litter NBA For TB size Stillborn mummies little size

1 Rcp(1;5) L

2 Rcp(1;15) Y 13 10.8 -27*** -28*** -14ns -51ns -4.82

3 Rcp(2;5) L 25 7.2 -46

4 Rcp(3;4) L Mat 73 9.2 -22*** -21*** -66*** -85*** -2.67

5 Rcp(3;12) Y 15 10.7 -17* -17* -21ns -63*** -4.4

6 Rcp(6;7) D Dn 4 10 +7.5

7 Rcp(7;15) D 54 6 -36

8 Rcp(8;13) L Dn 61 8.4 -28*** -27*** -38*** -22** -4.39

9 Rcp(12;14) D Mat 6 6.8 -27

10 Rob(13;17) L Mat 68 12.7 -4ns -6ns +19ns +14ns

11 Inv(8) D

12 Inv(6) L 14 7.4 -24 -1.46; - 0.37

In case 4, boar with rcp(3;4), 4 sows (dam, 1 full sister and two maternal half-sister of the carrier boar) were identified as carriers of the same translocation. The dam had farrowed 6 litters and the other 3 sisters had 7 litters in total. The average total born litter size, number born alive

(NBA), stillborn and mummified piglets for carrier sows were 11.5, 10.5, 1 and 0.38 piglets/litter, respectively. That represented a 10% reduction in prolificacy, when compared with the herd average of 12.8 piglets per litter. The number of stillborn piglets per litter was similar to the herd average (1 vs. 0.9 piglets/litter) and number of mummified piglets was slightly lower

(0.38 vs. 0.5 piglets/litter). Furthermore, the EBV for total born litter size of these sows (-0.37, -

0.25, +0.33, and +0.22) were in the range of the herd average. This suggests that this translocation has adverse effect on male fertility, but affected female carriers to a lower extent.

For case 12, inv(6), these two sows had different parents and grandparents. Sow #1 produced 9 litters (6.9 piglets/litter) and sow #2 had 5 litters (8.4 piglets/litter). Those represent a

29% and 13% reduction in total born litter size (herd average 9.68 piglets/litter). EBVs were -

1.46 and -0.37, respectively.

The frequency of chromosome abnormalities in Canadian and Vietnamese pig populations

Among the screened 732 young Canadian AI boars, 12 were identified as carriers of a chromosomal abnormality, which indicates a 1.64% prevalence. For Vietnamese farms, among the 43 randomly chosen boars and sows, two unrelated sows carrying the same inversion were found. Therefore, the frequency of chromosome abnormalities in this study for the Vietnamese pig population was 4.7%.

DISCUSSION

In this study, the 9 reciprocal translocations and 2 inversions presented are novel and have not been reported to date. There have been publications about the exchanges between 2 chromosomes 1/5, 1/15, 7/15 and 12/14, which were similar but not identical to cases 1, 2, 7 and

9 (table 2) of our study. Inversion 6 and 8 had also been described in the past (table 3). The breakpoints reported in those papers were different from the ones described here. Moreover, the two inversions described in this study are paracentric inversions, hence different from the pericentric inversions inv(8)(p21;q11), inv(8)(p11;q25) and inv(6)(p14;q12) reported by Ducos et al.,(2007). Appendix VI to XIII show the comparison of the cases described here with those reported in the literature. Some rearrangements have one of the breakpoints in common with some of the reported cases but the second breakpoint is different. Therefore, our cases reported here are clearly distinctive.

Table 2. Comparison of cases 1, 2, 7 and 9 from the current study to the literature

Rearrangements Fragile sites Breed Reference t(1;5)(q21;q23) 1q21 Landrace This thesis (case 1) t(1;5)(q21;q21) 1q21 Danielak-Czech et al., 1994 t(1;15)(q211;q13) 1q211 Duroc This thesis (case 2) t(1;15)(p25;q13) 1p25 Koukkanen and Makinen, 1988 t(1;15)(q17;q22) 1q17 Synthetic Ducos et al., 2007 t(1;15)(q26;qter) 1q26 Popescu et al., 1988 t(7;15)(q13;q13) Duroc This thesis (case 7) t(7;15)(7q-;15q+) Popescu et al., 1983 t(7;15)(q24;q12) Large White Konfortovaet al., 1995 t(12;14)(q15;q23) Duroc This thesis (case 9) t(12,14)(q13;q15) 14q15 French Duroc Ducos et al., 2007 t(12;14)(q15;q13) French Pietrain Ducos et al., 2007

Table 3. Comparison of inversions reported in the current study (case 11 & 12) to results from previous studies

Rearrangement Breed Reference

Inv(8)(q11;q25) Duroc This thesis (case 11) Inv(8)(p21;q11) Pietrain Ducos et al., 2007 Inv(8)(p11;q25) Large white Ducos et al., 2007 Inv(8)(p11;p12) Polish Landrace M. Switonski 1991 Inv(6)(q23;q35) Landrace This thesis (case 12) Inv(6)(p14;q12) Sino European Ducos et al., 2007

Traditionally, the detection of hypoprolific boars was usually only possible after a minimum of 10 litters were born. By that time, an AI boar has been used for at least 4 months and served between 100 and 300 sows. In most of cases, substantially more than 10 litters born are needed for detection of a hypoprolific boar. There are many factors that can affect total born litter size including management practices, time of the year, breeding values of sows and using semen from multiple boars at each breeding…; this requires that the farmer be cautious in the classification of a boar as hypoprolific. Recently, DBE were extracted based on routine computations in national databases and have been used increasingly as an enhanced detection tool for hypoprolific boars (Tribout et al. 2000; CCSI 2004). The effect of service sire is usually included in the evaluation model, as a fixed or a random effect. The main advantage of DBEs is to provide direct effects of service sires on total born litter size, with these effects being corrected for various effects, such as herd, season and breeding values of sows. Therefore, DBEs are expected to be more accurate than using the average total born litter size alone to compare with the herd average (Tribout et al. 2000). The four carrier boars described in this study had

DBEs considerably lower than 0 (-4.82, -4.4, -4.39 and -2.67), indicating that DBEs are a good measurement for total born litter size in boars. Despite DBEs being an advanced tool used to detect hypoprolific boars, enough litters sired by a suspect boar have to be born and sow litter

57 data entered to obtain an accurate DBE. Again, that boar could be used for services for almost 4 months before having its first litter born. Therefore, although a DBE value is a good measurement for fertility, it is not as helpful for detecting fertility problems that are happening currently on a farm. Nevertheless, it is valuable to screen relatives of an animal previously identified as a carrier of a chromosome abnormality, especially in cases where parents of the carrier are no longer available for karyotype analysis to investigate the origin of the abnormality.

Examples for the utility of DBE in this study are outlined below. In only 5 cases, the parents were still available for karyotyping to determine the origin of the rearrangements. Case 2,

5 and 11 are 3 cases where the origin of rearrangements could not be confirmed. By looking at the DBEs of related males, it is possible to obtain some information as to the origin of the abnormality. For case 2, rcp(1;15), the karyotyped Yorkshire carrier boar had a very low DBE, -

4.82, indicating a large adverse effect of this translocation on total born litter size. This boar had a sire with DBE of -6.0294 with 19 litters and a paternal grand sire with DBE of -4.7 from 63 litters while the dam, maternal grandparents had average total born litter sizes above herd average. The origin of this rcp(1;15) abnormality is therefore most probably from the paternal line. For case 5, rcp(3;12), the carrier boar had a negative DBE value (-4.4) based on data from

11 litters, which indicated that this translocation had a large adverse effect on total born litter size. However, the parents and grandparents of the carrier boar all had an average total born litter size around/above line average. In detail, the sire has DBE of +0.8213 (19 litters). The affected animal’s dam had an EBV of +1.22 (14 piglets/litter), which was based on data from 3 litters.

The paternal grand-sire had a slightly negative DBE value (-0.57) from 94 litters, the paternal grand-dam had an EBV of +3.18 (18.7 piglets/litter) based on 5 litter records. The maternal grand-sire had a positive DBE value of +0.452 based on 10 litter records and the maternal grand-

58 dam had an EBV of +0.51 (13piglets/litter) from 3 litters. Therefore, rcp(3;12) seemed to have occurred spontaneously. For case 11, inv(8), the boar has no litters, hence we cannot confirm the adverse effect of this inversion on total born litter size. However, this carrier boar has a dam and maternal grand-sire with average total born litter sizes below line average. Specifically, his dam had an EBV of -0.06 (6.5 piglets/litter, herd average for Duroc was 9.64 piglets/litter) from 4 litters and his maternal grand-sire had a DBE value of -2.78 based on 8 litter records. Thus, the origin of this inversion 8 was most probably maternal.

The 12 rearrangements reported here involved 12 out of 18 autosomal chromosomes in pig. In addition, they seemed to occur randomly although specific chromosomes are thought to be at greater risk for translocation than others are. Our case 1 and 2 are reciprocal translocations related to chromosome number 1. Chromosome 1, the largest chromosome in the pig karyotype, covers 10.4% of the whole genome; It is involved in 44 or more rearrangements (Appendix III) of approximately 180 chromosomal abnormalities (Appendix I and II) reported so far in domestic pigs. This indicates that chromosome 1 most frequently participates in chromosomal defects presumably due to its large size (Fries and Stranzinger 1982; Gustavsson et al. 1988).

Fragile sites are heritable points on chromosomes at which the chromosomes are prone to breakage under condition such as aphidicolin, folate-sensitive sites and BrdU (Riggs and

Chrisman 1991; Rønne 1995; M. Y. Yang and Long 1993). In the pig, 60 fragile sites have been reported in GTG banded chromosome in total (Rønne 1995). The high number of unstable regions (14 fragile sites) existing on chromosome 1 and identified as the breakpoints in 32 different translocations in vivo (Appendix III) so far suggests that this chromosome is particularly fragile as well. Review of 37 reciprocal translocations involving chromosome 1 are also interesting since the majority of chromosomes exchanged with chromosome 1 are one-

59 armed chromosomes (chromosome 13 to 18: 16 different reciprocal translocations) or a subtelocentric chromosome (chromosome 6 and 7: 9 different reciprocal translocations), which might predispose a close orientation and location of these chromosomes in the nuclei. The frequency of exchange of two specific chromosomes has been explained by the mechanism(s) of formation of chromosomal translocations. The combination of broken chromosome ends located near each other in the interphase nucleus or may brought together in an active process lead to establish a new translocation (Gollin 2007). Chromosome 15 is also an acrocentric chromosome.

Including the one described in this study, often 4 reciprocal translocations between chromosome

1 and 15 have been reported in the literature (table 2). Similarly to chromosome 1, chromosome

13 (our case 8 and 10 are related to this chromosome) is the largest acrocentric chromosome, covers 7.2% of the whole genome and contains 7 fragile sites, which makes it the most fragile among acrocentric chromosomes. However, it is involved in only 22 different in vivo rearrangements, while 35 rearrangements in vivo have been reported for the acrocentric chromosome 14 (Appendix IV and V). Chromosome 14, is a medium size chromosome and covers 5% of the whole genome; the number of fragile sites reported on this chromosome was few (5 fragile sites). The break point 14q23 in case 9 of this study is not a fragile site and has been reported in 4 different translocations (Appendix IX). In addition, 7q13 and 15q13 in cases 7 and 2 are also not fragile sites, but they have been involved in 6 and 8 different translocations, respectively, while 1q211 (in case 2) is a fragile site and is the first time described in this study

(Appendix X). Briefly, it is clear that specific chromosomes are at a higher risk for rearrangements than others are. The large size of the chromosome and fragile sites are factors known to increase the occurrence of rearrangements. However, in the pig, there are more factors that we still do not know that can enable this process.

In this study, 10 out of 12 boars with chromosome rearrangements had total born litter size data available for analysis. All of them, except case 6, rcp(6;7), showed a reduction in total born litter sizes and number of piglets born alive (NBA). However, only 4 litters were recorded for case 6, and the slight increase (+7.5%) in comparison to the herd average could be due to few records (bias resulting from small data sets). Regardless the type of translocation, the effect of chromosome abnormalities was already well known due to unbalance gametes which were produced by carriers. When those unbalanced gametes were fertilized with normal gametes, unbalanced zygotes were produced and were usually eliminated before implantation due to an imbalance in the number of copies of genes located in the affected chromosomes, resulting in small total born litter sizes (King 1981; Popescu and Boscher 1982). The actual extent of the reduction varied in different rearrangements. In this study, they varied from 4% to 46%.

Robertsonian translocations had the smallest effect with only 4% reduction, while reciprocal translocations resulted in a reduction of total born litter sizes ranging from 17% to 46%. Even some cases were reported with 100% reduction such as rcp(6p+;14q-) (Madan et al. 1978) and rcp(6p+;15q-) (Bouter R et al., 1974). Experiments using sperm FISH to estimate the percentages of different types of balanced and unbalanced sperm have also shown that they vary from translocation to translocation (Kociucka et al. 2014; A. Pinton et al. 2004). Carriers of translocations in the pig usually have normal sperm parameters, which indicated that the synapses during meiosis are resolved in different way such as quadrivalent or heterosynapses (D. a Villagómez et al. 1995; D. a F. Villagómez and Pinton 2008). However, the type of the chromosomes involved in rearrangements (e.g. metacentric, submetacentric and acrocentric), location of the breakpoints involved in chromosome arrangement (central, terminal and interstitial location), size of the chromosome segments involved and candidate genes for total

61 born litter size which are located at the breakpoints as well as chromosomes involved can contribute to the percentage of unbalanced gametes and possibly the ability to fertilize those gametes due to the gene for total born litter size were altered. More detailed experiments for each specific translocation need to be carried out to determine the main cause for the level of the reduction in total born litter size. Case 3 has the highest reduction of total born litter size (-46%).

This reciprocal translocation involved chromosome 2 and 5. Although it has a small exchange segments on the tip of each chromosome, the presumptive breakpoint is in the genes FSHb on chromosome 2 (figure 27), which was reported associated with total born litter size in the pig

(Pripwai and Mekchay 2011; Du et al. 2002). Moreover, among the 12 cases, only the presumed breakpoint of case 3 was on a dark band (2p16) while others were located on a light G band. In mouse, it is hypothesized that the break occurs in a dark G band, more suppression of crossing over later on synapsis (Ashley 1988). Besides chromosome 2, other chromosomes in pigs have been reported to contain genes related to total born litter size. Chromosome 1 (case 1 and 2) contains ESR (Buske et al., 2006). Chromosome 6 (case 6) has FUT1 and LEPR (Buske et al.,

2006). Chromosome 7 (case 7) has BF and VEGFA(Buske et al., 2006; Chen et al. 2015).

Chromosome 8 (case 8) contains SPP1, GNRHR and EGF (Buske et al. 2006; Distl 2007).

Chromosome 12 (case 5 and 9) enclose NAT9 and IBP4 (Jiugang et al., 2012; Shifei and

Yonggang 2012). Chromosome 14 (case 9) has RBP4, GADD45G and LIF (Distl 2007; Liu et al.

2015; M F Rothschild et al. 2000).

Figure 27. GTG banded and schematic drawings with QTL and candidate genes influencing fecundity. Figure legend: bold solid lines = level of significance p < 0.05; dashed lines = level of significance p > 0.05; EPOR=erythropoietin receptor; FSHb=follicle stimulating hormone beta; FUT1=fucosyltransferase 1; GNRHR=gonadotropin releasing hormone receptor; LEPR=leptin receptor; SPP1 (OPN)=secreted phosphoprotein 1 (from Buske et al., 2006). Similar to reciprocal translocations, inversions are thought to cause fertility problems due to unbalanced gametes (Morel et al. 2007) and associated gene copy number. Moreover, if the size of the inverted segment constitutes >30% of the chromosome, crossing-over may likely occurs within inverted synapsed regions, resulting in the risk of having unbalanced offspring

(Iwarsson et al. 1998). In our study, two paracentric inversions were found with inverted segment of ~35% and ~55% for chromosome 8 and 6, respectively. In addition, the breakpoints on those inversions are located in fertility genes and QTL for total born litter size (figure 27).

Hence, those inversion were expected to cause large adverse effect on reduce total born litter size of carriers. Unfortunately, the boar that is a carrier of the inv(8)(q11;q25) was young and sired no litters. Thus, we are not able to identify any effects of this inversion on fertility. Two sows, carriers of the inv(6)(q23;q35), had 13% and 29% reduced total born litter size. Although the size of inverted segment is quite large (55%) and both breakpoints occur right in fertility genes and QTL for total born litter size, the effect of the abnormality is not as serious (figure 27). To

63 date, most of the studies on inversions in the pig show that the inversions had little or no effect on the fertility of carriers (Y-I. Miyake, T. Matsubara 1994; Anielak-Czech B., A. Kozubska-

Sobocinska, E. Slota 1996; Katia Massip et al. 2010; K. Massip et al. 2009; a Ducos et al. 1997;

Świtoński 1991). Only inv(2)(p13;q11) was found in a hypoprolific boar but no reproductive data was mentioned in detail (Alain Ducos et al. 2002). The inversions in those studies including both small and large inversion segments, some breakpoints occurred right in fertility genes

(figure 28). Hence, inversions in pigs may have some mechanism to resolve the problem in synapses to minimize the effect of this rearrangement.

Figure 28. Schematic drawings with candidate genes influencing fecundity. ESR=estrogen receptor; EPOR=erythropoietin receptor; FSHb=follicle stimulating hormone beta; GNRHR=gonadotropin releasing hormone receptor; SPP1 (OPN)=secreted phosphoprotein 1 (from Buske et al., 2006). a= inv(1)(p21;q11) (Y-I. Miyake, T. Matsubara 1994); b= inv(1)(p221;q11)(Anielak-Czech B. et al., 1996); c= inv(1)(p24;q29); d= inv(1)(q12;q24); e= inv(1)(p21;q210); f= inv(2)(p11;q11); g= inv(2)(p11;q21); h= inv(2)(q13;q25)(K. Massip et al. 2009); i= inv(2)(p13;q11) (Alain Ducos et al., 2002); j=inv(4)(p14;q23)(Katia Massip et al. 2010; a Ducos et al. 1997); k= inv(8)(p11;p12) (Świtoński 1991).

The rob(13;17) translocation has been described in countries like China (Shan et al.

1994), France (Ducos et al., 2007) and Germany (Schwerin et al. 1986). Studies on the rob(13;17) translocation showed a reduction of fertility in pigs between 10-20% (Golisch 1989;

R. McFeely et al. 1988; Wnuk and Kotylak 2005). However, in our study, carrier boars of the rob(13;17) translocation had only a slight reduction of 4% in total born litter size. This result is in agreement with the findings of Pinton et al., 2009. In that study, dual colour fluorescence in situ hybridization of decondensed sperm nuclei was used and found that in males, the rate of unbalanced spermatozoa ranged from 2.96% to 3.83% ( a. Pinton et al. 2009). Unbalanced sperms fertilize with normal eggs resulting in unbalanced zygotes which are eliminated during pregnancy (King 1981; Popescu et al. 1984). Similarly, if a carrier boar mates with a carrier sow, more unbalanced zygotes are expected to be produced, and lead to a higher reduction in total born litter size. In a study pertaining to the rcp(4;14) translocation by Popescu et al., (1984), the authors reported that the mating of one carrier parent resulted in 45% reduction in total born litter size while mating of two carrier parents resulting 52% reduction of total born litter size.

Surprisingly, when a carrier sow and boar of rob(13;17) were mated together, a higher total born litter size (14.5piglets/litter) than the herd average (13 piglets/litter) were produced. However, the results are based on only two litters, more data and experiments are needed to confirm and explain for this observation. Although the rob(13;17) translocation leads to a minor reduction in the total born litter size of carrier boars, it caused a high dissemination of the translocation especially when a carrier boar was bred with a carrier sow, a total of 77% of progeny had inherited the translocation. Due to the small effect on total born litter size, boars that are carriers of rob(13;17) translocation are not easily recognized by breeders and therefore are usually used

65 for services for a long time. In addition with high dissemination, it is easily wide spread to many countries as a result.

In conclusion, in agreement with previous reports, the results of the present study underline the high incidence and variability of chromosome abnormalities in domestic pig populations. The prevalence of reciprocal translocation in the Canadian pig populations (1.64%) seems to be much higher than other frequencies reported. Reciprocal translocations seem to be the most frequent chromosomal abnormality occurring in domestic pig populations; they have a large negative effect on the fertility of carrier animals, which leads to lower total born litter sizes, decreased number of stillbirth and malformed piglets as common symptoms (Quach et al. 2010).

Nevertheless, inversion and especially Robertsonian translocations also have a small effect on reducing total born litter size of animal carriers. The findings here emphasize the relevance of cytogenetic screening programs to systematically test all breeding boars as an essential tool for swine improvement in any country with an intensive pork industry.

CHAPTER TWO Copy number variations in high and low fertility breeding boars

INTRODUCTION

Pork is the most consumed meat in the world (USDA and Foreign Agricultural Service

2014), thus high prolificacy of breeding animals represent a very important economic factor for the industry. As pigs are polytocous species, total born litter size is a direct measure of efficient fertilization and successful breeding. As a consequence, various total born litter size related traits are incorporated into genetic improvement programs with high economic importance. Genetic variability in genes with predicted reproductive functions and genotypes of linked SNP markers have been explored to identify hundreds of QTLs (Hernandez et al. 2014) and these markers have been successfully used to increase the rate of genetic gains. It is also known that large structural variations, such as chromosome rearrangements are major etiologic factors behind reproductive dysfunction and eradication of carriers could help in efficient and economical breeding (A Ducos et al. 2008). Smaller sized genome rearrangements, such as deletions or duplications that disrupt the balance in genome integrity and result in copy number variations

(CNVs) represent a novel type of molecular marker (Bickhart and Liu 2014). This class of structural variations have become the focus of research since its discovery (Iafrate et al. 2004;

Sebat et al. 2004) and in particular the recognition that a surprisingly high proportion of the human genome is involved in CNVs, potentially affecting gene expression and phenotype

(Haraksingh and Snyder 2013). Since then, numerous studies described CNV in human populations and the current Database of Genomic Variants contains ~110,000 CNVs, covering

71.5% of the human genome. Also, 90% of transcripts and 79% of microRNA loci are overlapped by CNVs (MacDonald et al. 2014). Association of CNVs to disease states have also been attempted leading to the identification of putative markers involved in the development of

68 various cancers, neurological disorders, recessive diseases, etc. (Almal and Padh 2012; Boone et al. 2013).

Recently, this level of genome variability has also been investigated in domestic animals including cattle, pig, goat, sheep, horse, dog and chicken (Clop, Vidal, and Amills 2012) and hold promise to become useful markers for genetic selection (Gurgul et al. 2014; Bickhart and

Liu 2014). The first insight into CNV content of the porcine genome was from a study that involved only four chromosomes due to the difficulty of array CGH (aCGH) platform design

(Fadista et al. 2008). Recently, with the availability of a more refined genome assembly, genome-wide high density oligonucleotide CGH arrays could be designed and used to investigate pigs from many different breeds(Y. Li, Mei, et al. 2012; Jiying Wang et al. 2014), but it also opened the possibility of investigating CNVs from individual whole genome sequences (Rubin et al. 2012; Paudel et al. 2013).

The applicability of SNP genotyping arrays for the estimation of DNA copy numbers have made the Porcine SNP60k chip (Ramos et al. 2009) the method of choice for several other research projects. Ramayo-Caldas et al. (Yuliaxis Ramayo-Caldas et al. 2010) have identified 49

CNVRs in 55 pigs, while 565 CNVRs have been described in a study of nearly 1700 pigs from

18 populations (C. Chen et al. 2012). Wang et al. (L. Wang et al. 2013) investigated a large population of Large White x Minzhu pigs and described 249 CNVRs, while Fernandez et al.

(Fernández et al. 2014) investigated a highly inbred Iberian strain and found 65 CNVRs. Based on the studies using the SNP60k chip, CNVs cover 16.08% of the porcine genome (Fernández et al. 2014). This is a fraction of CNVR length reported in humans, most probably due to the smaller number of animals investigated and the less refined genome assembly and screening tools available, leaving much to discover.

Most of the available porcine CNV studies contain functional annotations of the gene content of identified CNVRs and provide important descriptions of new individual or breed specific variants with slightly different estimates of this level of genome variation in pig populations. Furthermore, Chen et al. (C. Chen et al. 2012) has associated several meat and carcass quality traits (QTL) with CNVRs and identified seven candidate genes potentially affecting these traits. Also, six CNVRs contained significant SNPs for several meat quality traits after merging genome-wide SNP association data with the copy number variation map (L. Wang et al. 2013). To our knowledge, only one study has initiated CNV discovery in pigs that were selected from the two ends of the fat/lean estimated breeding value spectrum, in an attempt to identify candidate CNVs associated with fatness (Fowler et al. 2013).

The goal of this study is to investigate the feasibility of identifying candidate CNVs related to fertility in a selected population of high and low fertility boars. Gene content and reproduction QTLs that are mapped to the positions of identified CNVs were analyzed.

MATERIALS AND METHODS

Animals and array genotyping

The Canadian Centre for Swine Improvement Inc. (“Canadian Centre for Swine

Improvement Inc.”), as a non-profit organization, collects and manages pedigree, breeding and performance information, as well as SNP genotypes of breeding animals from numerous major breeders across Canada to calculate and provide estimated breeding values for various traits. This integrated database was screened to identify boars with exceptional high and low fertility. The fertility indicator was defined as the calculated direct boar effect on total born litter size (DBE) that can be obtained as a by-product of the national genetic evaluation for total born litter size

(BLUP). DBEs are more accurate than only considering total born litter size averages in mates, since the estimated boar effects are then corrected for all identified environmental effects and breeding values of their mates (Tribout et al. 2000). The DBE value precisely shows how many more or less piglets a given boar produces per litter on average, as compared to the overall average of the population. The database contained 16,959 Yorkshire, 14,188 Landrace and 7,366

Duroc boars with calculated DBE values. From these more than 38,500 boars we catalogued animals from the most extreme 10% on both sides of the distribution, that would be equal to approximately 2 more or less piglets than herd average. Among these high and low fertile boars we have selected the ones which had the Porcine SNP60k array genotypes available, generated at the genomics facility in DNA LandMarks Inc. (QC, Canada), as part of large genotyping projects. For the purpose of CNV prediction 11 high fertility boars having 2.83±0.61 (mean±SD)

DBE value and 10 low fertile boars with -2.72±0.79 DBE value were randomly selected.

Moreover, eight high fertile (DBE 2.38±0.36) and 9 low fertile boars (DBE -3.38±0.97) were randomly chosen to be used for validation of the CNV predictions. Distribution of DBE values of

71 the various groups are presented in Figure 29. All together, the selected 19 high fertile and 19 low fertile animals represent various breeds, such as Yorkshire (21), Landrace (14) and Duroc

(3).

Figure 29. Descriptive statistics (mean ± SD) of the direct boar effects (DBE) in the high and low fertile animals, the validating groups and all together.

CNV analysis

Each probe-pair on the Illumina SNP60k genotyping array (for alleles A and B) marks a specific location in the porcine genome and its signal intensity could be related to the amount of

DNA at that locus. In order to estimate DNA copy numbers, the observed normalized probe signals in each sample were compared to an expected signal intensity calculated from the

Illumina defined reference sample cluster, thus generating a log R ratio value

(log2(Robserved/Rexpected), as described by Peiffer et al., (Peiffer et al. 2006)). This procedure was done using the Illumina GenomeStudio software, before being transferred into the SNP and

Variation Suite version 7.7.8 (SVS, GoldenHelix) for quality control and CNV analysis.

Noise in the log R ratio values, inherent from the sample preparation or genotyping procedure could cause faulty identification of CNVs or confounding associations. Thus, our quality assurance workflow consisted of several different steps to identify potentially low quality samples. The X and Y chromosomal data were excluded from the data. Initial quality controls for noisy samples were done by testing for outliers in the median derivative log ratio values.

Genomic wave factors were detected using the correction algorithm developed by (Diskin et al.

2008) Diskin et al., as implemented in SVS. Potential batch effects were tested by principal component analysis (PCA) of logR values. SNP marker locations were annotated on the latest genome build, Sscrofa 10.2/susScr3 (2011).

There are two different copy number analysis methods (CNAM) in SVS that implement the same segmenting algorithm two different ways. The univariate CNAM scans each sample separately and is ideal for identifying larger segments in individual genomes, while the multivariate CNAM segments all samples simultaneously, thus generally smaller but common

CNVs could be identified. We have applied both segmentation methods on our dataset to predict

CNVs with maximum pairwise segment p value being 0.005, the min number of markers/segment value either 1 or 3 and the segment means were filtered to be <-0.15 for losses or >0.1 for gains. Overlapping CNVs were then merged to a CNV region (CNVR). The high and low fertility groups were also separated for the multivariate CNAM, thus facilitating the identification of CNVs specific for only one phenotype. The validation set of high and low fertile animals were segmented with the univariate and multivariate methods using the same conditions.

The two-sided Mann-Whitney U-test were used to detect significant (p<0.05) differences between the high and low fertile groups.

73 qPCR validation

Eight CNVRs among the ones present in the largest number of animals (four CNVRs specific for the high and four CNVRs for the low fertility group) were validated by quantitative real-time PCR (qPCR). The DNA samples of the animals - in which the CNVRs were predicted - were retrieved from the DNA collection at our industry partner. No experiments were carried out on animals, thus no ethical approval was required. Primers were designed using the Primer3 plug-in of Geneious software. The primers by Chen et al. (C. Chen et al. 2012) were used for the beta-actin control region. Primer sequences and product sizes are in appendix XIX. qPCR was performed using a CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad) under the following thermal profile: 98°C, 2 min; 45×(98°C, 10 sec; 59°C, 10 sec) followed by the registration of a melting curve between 68°C to 95°C in 0.5°C/sec increments. The 10µl reaction was composed of 1× SsoFast EvaGreen Supermix (Bio-Rad), 3mM primers and 20ng genomic

DNA and samples were run in triplicate. Primer efficiencies were calculated as the average of individual well efficiencies determined by linear regression of amplification curves using the

LinRegPCR software (Ruijter et al. 2009). The relative quantity of each locus was determined against the control sample after normalization to the beta-actin signal using the formula described by Pfaffl et al. (Pfaffl 2001).

Functional annotation of CNVRs

Genomic locations of QTLs and the ones involved in reproductive traits (reproduction

QTLs) were downloaded from the Animal Genome Database (Hu et al. 2013). Enrichment of the latter in CNVR regions were tested using the Chi-square test with Yates correction. The RefSeq gene list was downloaded from the UCSC Table browser (Karolchik et al. 2014). The resulting list of annotated genes was further analyzed for functional enrichment in Gene Onthology (GO)

74 terms using the various tools implemented in the WEB-based Gene SeT AnaLysis Toolkit

(WebGestalt, (Jing Wang et al. 2013)). The porcine gene names were converted to the corresponding human ones and the resulting list was contrasted against the human genome as a reference set for the default statistical test (Benjamini-Hochberg, adjusted p-value <0.01).

RESULTS CNV analysis

Prior to CNV analysis several quality control steps were carried out. We have not identified any samples with outlier noise in the log R ratio values. We also checked another type of noise, the so called “genomic waves”, that are variations of the signal intensity related to the genomic position of the probe, thus the composition of the DNA (Diskin et al. 2008), and found no outlier wavy sample. The animals were selected from a large set of samples, which were not all genotyped at the same time, thus we performed principal component analysis to investigate potential batch effects. The PCA identified clear stratification of the data based on the date of array procedure. The effect of the fertility status and breed of animals were also PCA tested and no clustering was observed with any of these two parameters.

We chose to apply the two available segmentation options in the SVS software to explore putative CNVs. This algorithm - although widely used in human studies (GoldenHelix) - has not yet been applied to any data set generated on the Porcine SNP60k platform.

The first segmentation method (univariate CNAM) searches individual genomes.

Segments with significantly different log R ratios from their neighbors are identified as CNVs, which were then sorted according to the fertility status. Figure 30a shows a region of the genome where four CNVs of slightly different lengths were identified in four low fertility animals and marked by red bars representing losses. The overlapping region could then be merged to a low fertility specific CNVR. This method identified 48 CNVs in individual genomes, which were then compared to their fertility status and merged into 24 overlapping CNVRs. Among these, 10 were specific to the high fertility and 12 to the low fertility group, while two CNVRs - although present in both groups - showed the opposite copy number status (loss in low fertility and gain in

76 high fertility, or vice versa).

Figure 30. Examples for the two CNV identification options in SVS software. a) 21 individual animals genomes were subjected to the Univariate CNAM segmentation option. A red bar represents a segment with significantly lower log R ratio, as compared to its surrounding regions, thus identified as genomic loss. Slightly different length CNVs were identified in different low fertility samples, thus the overlapping region could be merged to a low fertility specific CNVR. b) The multivariate CNAM method segments a group of samples together and segments are called only if present in all samples. In order to identify CNVs specific for either the high fertility or the low fertility group, the samples were segmented together, as well as grouped according to the fertility status. Only those CNVs were accepted that were present in only one group, but neither in the other phenotypic group or in all samples together. Here the red bar identifies a CNVR specific for the high fertility group.

The second multivariate CNAM segmentation does not scan individual genomes, but rather checks if the segment cut-point is present in all samples for successful CNVR calling. We have used this approach to identify CNVRs specific for each fertility status by grouping the samples into low fertility or high fertility group and a 3rd control group containing all samples.

As it is represented in Figure 30b, an acceptable high fertility group specific CNVR would be identified in all members of the given group, but should not be present in the low fertility group, neither in the control group. As the PCA identified samples clustering according to the date of genotyping, those two clusters were separated before multivariate CNAM to provide the maximal homogenous set of samples without confounding batch effects. We have discovered 11

CNVRs that fulfilled these criteria of multivariate CNAM and among those, four were specific to the high fertility group and seven to the low fertility group.

All together the two different strategies of CNAM resulted in the identification of 35

CNVRs. Fourteen CNVRs were specific to the high fertility and 19 to the low fertility boars

(appendix XIV). Only 14 of the 18 autosomes harbour CNVRs, as none was identified on chromosomes 4, 5, 7 and 15 (Figure 32). The name of each region, such as CNVR18L, is composed of “CNVR” followed by a number and “L” for being specific to the low fertility group or “H” in case of the high fertility group.

Chromosome 2 had the highest number of CNVRs (8) and the largest region involved in them (~12Mb), while 5 other chromosomes had only 1-1 CNVR (Appendix XV). We observed an excess of copy losses (28) and five gains and two regions where both gains and losses were found (Appendix XVI). The total length of CNVRs is approximately 36.5 Mb, which is distributed in a ratio of 4:1.5:1 among losses, gains and gain/loss regions.

Quantitative real-time PCR (qPCR) was used to validate the identification of CNVRs.

The results for all eight tested loci were in agreement with our predictions (Figure 31). Every animal from the group showed the predicted CNVR status in five CNVRs (CNVR18L,

CNVR27L, CNVR28H, CNVR34L, CNVR36H), while five out of six samples tested positive at two loci (CNVR15H, CNVR37H) and two out of four samples were confirmed at CNVR38L.

Another aspect of validating the predicted CNVRs was the investigation of an independent set of high and low fertile animals to determine whether the same regions could be identified. In fact, 26 CNVRs were present and 7 maintained significant association with the fertility status (CNVR5L, 7L, 10L, 18L, 45L, 50L, 62L). Interestingly, we found two regions

(CNVR7L, 45L) where samples with both gain and loss status were present, although the original predictor set of animals represented only one of them.

Figure 31. Results of validation experiments for 8 CNVRs by qPCR. Relative quantity of target amplicons were calculated against the control sample (C) after normalization to the beta-actin locus.

CNVRs overlapping reproduction QTL regions

The genomic positions of the 35 identified CNVRs were used to search for reproduction

QTLs mapped to the same positions in the Animal QTLdb including the endocrine, total born litter size, reproductive organ and reproductive trait categories. The majority of CNVRs (30) overlapped with 137 QTLs from 16 traits and only 5 CNVRs are situated in regions of the porcine genome that have no reproduction QTL mapped (Appendix XVII). The chi-squared test with Yates correction (p<0.05) showed significant enrichment of reproduction QTLs among all

QTL categories within the boundaries of the identified CNVRs.

The most abundant QTLs were the teat number “TNUM” and age at puberty “AGEP”.

Thirty-eight TNUM and 26 AGEP were mapped to regions where CNVRs were detected.

Twelve traits had QTLs mapped to chromosome regions where either low or high fertility group specific CNVRs were found, however the following 4 QTLs were found to be specific for only one of the fertility groups. A QTL for “plasma FSH concentration” (QTL #646, (Rohrer et al.

2001)) was found to overlap with CNVR43H and CNVR44H from high fertility group animals.

Two low fertility group specific CNVRs (CNVR13L and CNVR39L) overlapped with two different QTLs for “gestation length” (#21837, (C. Y. Chen et al. 2010); and #452, (Wilkie et al.

1999)). One QTL for “testicular weight” (#6527, (Ren et al. 2009)) harboured CNVR5L, a low fertility group CNVR. And at last, a “uterine capacity” QTL (#523, (Rohrer et al. 1999)) lie together with the low fertility group specific CNVR34L.

Figure 32. Location of the detected CNVRs on the porcine chromosome ideograms. The size of each ideogram is proportional to that of the chromosomes. The sex chromosomes were excluded from the analysis and no CNVR was detected on chromosome 4, 5, 7 and 15. The brown bars in the middle of each chromosome represent the positions of CNVRs. The purple columns in the left are the positions of QTLs and RefSeq genes are marked by red bars on the right side of the ideograms.

Functional annotation of CNVR gene content

Sequences - with RefSeq IDs - mapped to positions of CNVRs were retrieved from the

UCSC Table browser. The identified 35 CNVRs encompassed 50 genes (Appendix XIV and

XVIII). The majority of these were specific to the low fertility group (40), 7 to the high fertility group, while 3 were found in regions that were present in both groups. Most of the genes, 27 and

10 respectively, were found on chromosomes 2 and 12, not surprisingly these two chromosomes are covered with the largest regions, approximately half of the total size of CNVRs. CNVRs identified on chromosomes 3, 6, 11, 13 contain no genes.

A functional analysis of several databases revealed that the genes found in CNVRs from the low fertility group have been significantly enriched in members of the innate immune

82 system, toll-like receptor (TLR) and RIG-I-like receptor signaling and fatty acid oxidation pathways (Table 4). The seven genes from the high fertility group CNVRs and the ones present in both groups do not specify any pathways with significant enrichment p-value.

Five micro RNAs (miRNAs) were also found to position within CNVRs: miR-21, miR-

142, miR-143, miR-145, miR-202 (Table 5). The latter was detected in both high and low fertility groups with opposite copy number status (deletion and gain, respectively), while the other four were only found in the low fertility group CNVRs, as deletions.

DISCUSSION

In this study we applied the extreme groups / selective genotyping approach (Darvasi and

Soller 1992) for identifying copy number variations in high and low fertility breeding boars.

These two groups of animals representing approximately 10% of both the upper and lower ends of the distribution from a large population of boars had mean DBE values of -2.7 and 2.8. One represents outstanding high fertility, while the others having high negative DBE values are considered low fertility. Animals from these two diverse phenotypes were compared by several approaches in order to prove the feasibility of our CNV analysis and to identify putative markers of fertility.

It should be noted that using a small subset of animals from the extreme ends of the phenotypic distribution not only reduced the cost of genotyping, but could retain the power of analysis as proven by simulation (Darvasi and Soller 1992) and numerous QTL mapping studies

(H. Lee et al. 2014). Recently, it was also applied for CNV discovery as well, based on animals sampled from the distribution of fatness EBV (Fowler et al. 2013).

We have identified 35 CNVRs covering 36.5Mb or ~1.3% of the 2800Mb porcine genome. The size range distribution of CNVRs is similar to that of other publications using the same SNP60k chip. There are numerous software tools available, such as PennCNV, cnvPartition, QuantiSNP, GADA to name a few which employ very different algorithms for the identification of CNVs from SNP array data (K. Wang et al. 2007; Colella et al. 2007; Pique-

Regi, Cáceres, and González 2010). A comparative analysis of several of these has found highly variable CNV calls due to this inherent difference (Xu et al. 2013). Previous studies into porcine

CNVs from SNP array data have chosen from these four softwares with slight preference given to PennCNV (Yuliaxis Ramayo-Caldas et al. 2010; L. Wang et al. 2013; C. Chen et al. 2012;

Fowler et al. 2013). However, we opted for the SNP and Variation Suite (SVS) (GoldenHelix) mainly because of the extensive tools available for quality assurance and the unique multivariate segmentation option providing the detection of associated regions across samples. Our motivation was to combine the advantages of using the extreme groups for the DBE phenotype and the increased analytical power of marker-level CNV test in detecting smaller common

CNVs, as the latter was tested by Breheny et al. (Breheny et al. 2012).

In fact, among the identified 35 CNVRs, 14 were specific to the high fertility animals, while 19 CNVRs were specific to the low fertility group, thus worth investigating their putative roles in fertility.

The quality of CNVR analysis was assessed by qPCR validation of four CNVRs specific for the low fertility and four for the high fertility group. All of the qPCR assays confirmed the

CNV call and 90% of the tested animals gave results in agreement with the prediction, that represents among the highest validation rate published to date in pigs (C. Chen et al. 2012). We have also validated the identification of CNVRs using an additional independent set of high and low fertile boars. Our further analysis steps involved the comparison of CNVRs to already mapped reproduction QTLs, then the functional characterization of transcript content. Visual representation of these two comparisons are given in Figure 32, where the chromosomal positions of discovered CNVRs are aligned with QTLs and genes.

The identified CNVRs overlap with 137 QTLs of various reproductive traits (Appendix

XVII). QTLs generally represent the first estimation of the link between the genetic component of an important phenotypic variation and a smaller or larger segment of the genome (Ernst and

Steibel 2013). These QTLs were identified and mapped with statistical significance by using various methods from microsatellite markers to whole genome scan on very different

85 populations. The experiments led to the mapping of these 137 reproductive QTLs were published in more than one hundred studies, that could not be cited in this article, but could be accessed from the PigQTLdb (Hu et al. 2013). Appendix XIV contains all corresponding CNVR IDs and

QTL IDs. The described CNVRs fall into the 7kb to 1.6Mb size range, that is in many cases much smaller than the current QTL region and thus could facilitate narrowing down the real functional locus and help the identification of the causative gene. It should also be mentioned that CNVRs described here were not tested for statistical association with QTLs, simply the overlapping genomic positions of the latter was used as one indicator of the potential function.

Nonetheless we found that reproduction QTLs were over-represented within CNVR boundaries.

As the porcine genome sequence and annotation are available in public databases

(Groenen et al. 2012), we attempted to characterize the functional content of CNVRs. One of the common results of pathway analysis using the various databases was the significant enrichment of elements of the innate immune system in low fertility samples (Table 4). A well-known connection exists between infections of either the female or male reproductive tract and impaired fertility. The innate immune system exhibits the non-specific response against pathogens, as the first-line of defense and then helps to activate the adaptive immune system. It is comprised of specific cell types, pattern recognition receptors and antimicrobial peptides, etc. Among these, we have identified CNVRs containing various components of the TLR signaling and RIG-

I/MDA5 mediated induction of interferon signaling pathways. TLRs are transmembrane proteins that recognize pathogen associated molecular patterns. TLR2 binds those of microbes, while

TLR3 is involved in cytoplasmic binding of viral nucleic acids (Saeidi et al. 2013), as well as

RIG-I and MDA5 receptors (Meylan, Tschopp, and Karin 2006). These proteins are all localized throughout the male and female reproductive tract in humans and domestic species (Bermejo-

Alvarez et al. 2008; Saeidi et al. 2013). The balance of TLR response is required for physiological function of the reproductive organs - in order to protect against infections, and alterations in function have been associated with adverse effects on endometritis, ovulation, pregnancy outcome and sperm production (Bhushan et al. 2009; Kannaki, Shanmugam, and

Verma 2011).

Another significant pathway among the genes localized within CNVRs was the fatty acid metabolism. The presence and balanced metabolism of fatty acids are connected to a plethora of cellular functions, including mitochondrial energy production, oxidative stress, cytoplasmic and membrane functions. These biological processes all affect fertility through the development of germ cells and their ability for successful fertilization. Fatty acids are metabolized in the mitochondria to produce acetyl-coA that enters the citric acid cycle and thus result in ATP.

Motility of spermatozoa requires substantial energy resources (Amaral et al. 2013) but the ATP level of the matured oocyte is also essential for providing energy for the developing embryo

(McKeegan and Sturmey 2012). The cellular availability of different types of fatty acids contribute to the fluidity of plasma membrane, that is essential for cell fusion events, such as fertilization (Wathes, Abayasekara, and Aitken 2007), but is also key to protect the cellular integrity from oxydative damage (Tremellen 2008). Three different CNVRs contain the following three members of this metabolic pathway. PNPLA2 encodes an enzyme in the initial steps of lipid metabolism by catalyzing the hydrolysis of triglycerids and its impaired function was shown to result in decreased plasma fatty acid levels (Kienesberger et al. 2009). Similarly, the product of the CPT1A gene is key to the mitochondrial transport of long-chained fatty acids

(K. Lee, Kerner, and Hoppel 2011), while ECHS1 is the hydrolase in the second step of the beta

87 oxidation, thus their functional imbalance affects the rate of fatty acid metabolism (Janssen et al.

1997).

MicroRNAs (miRNAs) are key players in gene expression regulatory networks and as such, they might be strong candidates for disease-causing non-coding sequences. The variable dosage of miRNA genes due to CNVs are affecting their expression profile and regulatory role

(Marcinkowska et al. 2011). Wu et al. (X. Wu et al. 2012) suggested an evolutionary mechanism that could correct for this by increasing the diversity of acting miRNAs on their targets and/or adjusting the copy numbers of their major target genes according to the CNV of the miRNA. The

CNVRs found here harbor five miRNA genes (miR-21, miR-142, miR-143, miR-145, miR-202), the first four of them specific for the low fertility animals. Interestingly, none of their predicted target genes are situated within the boundaries of the 35 CNVRs described in this study. This would theoretically suggest the impaired function of these miRNAs and their putative role in the phenotype, although laboratory validation of their expression level is necessary to prove this. It is also interesting that these are among the most abundant miRNAs expressed in the male and female reproductive tissues (Donadeu, Schauer, and Sontakke 2012; J. Huang et al. 2011). miR-

21 is present in testicular germ cells (Kotaja et al. 2006) and linked to the maintenance of spermatogenic stem cell population (Niu et al. 2011). Furthermore, it is also localized in granulosa cells of pre-ovulatory follicles and plays a role in the follicular-luteal transmission, proven by its increasing level of expression (McBride et al. 2012). Similarly, miR-142 shows variable expression levels between follicular and luteal phases (Donadeu, Schauer, and Sontakke

2012). miR-143 and miR-145 are found to co-express and function in the regulation of cell proliferation (Ohlsson Teague, Print, and Hull 2010), smooth muscle (Cordes et al. 2009) and adipocyte development (T. Wang et al. 2011). Some studies found these to be preferentially

88 expressed in the male gonads (Takada et al. 2009) and epididymis (Y. Li, Wang, et al. 2012) while others reported abundant expression in the ovary (J. Huang et al. 2011) and functions related to endometriosis (Ohlsson Teague, Print, and Hull 2010).

miR-202 was identified as copy number gain in low fertility and deletion in the high fertility group, which would imply a negative role in fertility regulation. This is in agreement with the observations of its marked upregulation in various testicular hystopathologic conditions

(Abu-Halima et al. 2014) and also in premature ovarian failure patients (X. Yang et al. 2012).

Similar to miR-202, we have found one gene with gain/deletion copy number status in

CNVR1HL. However, it is a deletion in the low fertility and a gain in the high fertility group.

Although this status distribution would make it an optimal marker for fertility, it was only found in 1 animal from each group. Furthermore one gene, the gluthatione S-transferase mu2

(LOC780435, NM_001078684), is mapped to this region of chromosome 1. The superfamily of these metabolic enzymes functions as important players in protecting the cells from oxidative damage and endogenous toxicity (Strange et al. 2001). Interestingly, in humans it lies in a hypervariable region, where structural rearrangements and deletions are frequent. The resulting variability in gene copy number, structure and enzyme activity thought to contribute to the individual’s stress response and strong association has been found with sperm production and male infertility (W. Wu et al. 2013).

Table 4. Functional enrichment analysis of genes encompassing the identified CNVRs. The results of the various tools in WebGestalt (http://bioinfo.vanderbilt.edu/webgestalt/option.php) are summarized.

adjusted p- Database PathwayName ID Statistics value Toll-like receptor signaling pathway 04620 C=102;O=3;E=0.10;R=28.83;rawP=0.0002 0.0014 KEGG Fatty acid metabolism 00071 C=43;O=2;E=0.04;R=45.59;rawP=0.0009 0.0032 RIG-I-like receptor signaling pathway 04622 C=71;O=2;E=0.07;R=27.61;rawP=0.0024 0.0036

C=73;O=3;E=0.07;R=40.28;rawP=5.86e- Fatty Acid Beta Oxidation WP143 0.0005 05 WikiPathways Toll-like receptor signaling pathway WP75 C=116;O=3;E=0.12;R=25.35;rawP=0.0002 0.0008 Regulation of toll-like receptor signaling pathway WP1449 C=154;O=3;E=0.16;R=19.09;rawP=0.0005 0.0013

C=532;O=6;E=0.54;R=11.05;rawP=1.62e- Immune System 522 0.0012 05 C=190;O=4;E=0.19;R=20.63;rawP=4.32e- Innate Immune System 1094 0.0012 05 RIG-I/MDA5 mediated induction of IFN- C=67;O=3;E=0.07;R=43.89;rawP=4.53e- 1115 0.0012 alpha/beta pathways 05 Pathway Interferon Signaling 1123 C=98;O=3;E=0.10;R=30.00;rawP=0.0001 0.0013 Commons Toll Receptor Cascades 1095 C=90;O=3;E=0.09;R=32.67;rawP=0.0001 0.0013 C=77;O=3;E=0.08;R=38.19;rawP=6.87e- Interferon alpha/beta signaling 1122 0.0013 05 Toll Like Receptor 9 (TLR9) Cascade 1084 C=65;O=2;E=0.07;R=30.16;rawP=0.0020 0.0089 Toll Like Receptor 2 (TLR2) Cascade 1136 C=65;O=2;E=0.07;R=30.16;rawP=0.0020 0.0089 TRIF mediated TLR3 signaling 1074 C=56;O=2;E=0.06;R=35.01;rawP=0.0015 0.0089 *where C= number of reference genes in the category, O= observed number of genes in the gene set from the category, E= expected number in the category, R= Ratio of enrichment, rawP= p value from hypergeometric test, adjusted p-value= p value adjusted by the multiple test adjustment.

Table 5. Summary of micro RNAs found within CNVRs

Name Transcript ID Chromosome CNVR ID Fertility CNV status miR-21 NR_038508 12 CNVR50L Low DEL miR-142 NR_038555 12 CNVR50L Low DEL miR-143 NR_038529 2 CNVR16L Low GAIN miR-145 NR_038484 2 CNVR16L Low GAIN miR-202 NR_035399 14 CNVR59HL High & Low GAIN/DEL

CHAPTER THREE The porcine TSPY gene is tricopy but not a copy number variant

INTRODUCTION

The majority of the Y-chromosome represents a special part of the mammalian genome, that is unique to males. A multitude of evolutionary rearrangements, deletions, inversions, transpositions and amplifications, are thought to have eroded its size and shaped its structure

(Jennifer A Marshall Graves 2006). It is not surprising that this resulted in highly variable Y- chromosomal structures, as showed by the comparison of few completely sequenced, assembled and mapped examples of Y-chromosomes which include human, chimpanzee, rhesus monkey and mouse (Skaletsky et al. 2003; Soh et al. 2014; J. Hughes et al. 2010; J. F. Hughes et al.

2012). On the other hand, it also contains the SRY gene, the essential sex determination locus, as well as several genes expressed specifically in the testis and potentially with substantial functions in gametogenesis and fertility (J A M Graves 2006; Skaletsky et al. 2003).

The testis-specific protein Y-encoded (TSPY) gene is situated on the male specific portion of the mammalian Y-chromosome and exhibits some remarkable biological characteristics. Foremost, TSPY has the highest copy number (CN) of all protein coding genes in the human genome (~35, (Xue and Tyler-Smith 2011)). This exceptional CN is surpassed in domestic cattle bulls which have on average 94 copies of TSPY (Hamilton et al. 2009a). While most mammalian genes exist and function in two copies, recent discovery of CN variations

(CNV) identified that those genes could be affected by deletions or duplications resulting in ± 1-

2 copies (Redon et al. 2006). This is not the simple case for TSPY, as the variation among men and bulls ranges from 11-76 (Krausz, Giachini, and Forti 2010), and 50 to 200, respectively (C.

K. Hamilton et al. 2009). This variation could be explained by the tandem repeated array structure within the Y-chromosome that this gene lies in, that could provide opportunities for imperfect pairing and recombination between the sister chromatids, resulting in large deletions or

93 additional copies (Jobling 2008). This tandem repeated genomic structure – named simply as the

TSPY cluster (although individual copies have an additional numerical identifier) represents the largest and most homogenous protein coding cluster in the human genome (Skaletsky et al.

2003).

TSPY orthologs were identified in many other species with strikingly different copy numbers from the human and bovine examples. The chimpanzee and rhesus monkey Y- chromosomes contain six copies of TSPY, while multi copy status has also been suggested in other great apes, horses, cats and dogs (J. Hughes et al. 2010; J. F. Hughes et al. 2012; Paria et al.

2011; G. Li et al. 2013). Interestingly rats have only one functional copy, while mice have lost the functional TSPY locus (Vogel et al. 1998; Dechend et al. 1998). The structure of the porcine

TSPY locus is not known, however a PCR identified BAC clone has been used as a FISH probe and mapped to a single Y-chromosomal locus (Quilter et al. 2002) but its copy number was not yet determined. There is only one predicted TSPY gene sequence identified in the various gene banks, interestingly named as TSPY4. Although the current Sscrofa10.2 genome does not contain a Y-chromosome assembly, this TSPY4 mRNA aligns perfectly to the whole Y- chromosomal draft sequence.

Neither the biological function of TSPY, nor the effect of its variable copy number has been clearly elucidated. The conservation of the protein sequence across many mammals and its specific expression in the testis suggests roles in male reproduction. Molecular analysis identified several binding domains to interact with cyclins and EEF1A, so functions in regulation of cell cycle, renewal of spermatogonia and also as proto-oncogene in various testicular cancers have been proposed (Schubert and Schmidtke 2010; Kido et al. 2014). It is also mapped at the putative gonadoblastoma locus (Lau, Li, and Kido 2009). The impact of variation in TSPY copy

94 number is not known, however association with fertility has been identified in cattle and humans

(Krausz, Giachini, and Forti 2010; Hamilton et al. 2012; Mukherjee et al. 2015).

There are TSPY paralogs identified outside the Y-chromosomal TSPY locus. The human genome contains six TSPY-like genes, the five autosomal copies are most probably derived by retrotransposition, while the one copy on the X-chromosome marks the common evolutionary origin of the sex chromosome pair (Krausz, Giachini, and Forti 2010). Interestingly studies suggest contrasting functions of TSPX as tumor supressor compared to TSPY, based on its expression pattern and molecular interactions (Krausz, Giachini, and Forti 2010). The role of the autosomal TSPY-like genes is not clear, however a mutation of TSPY-L1 identified in a case of sudden infant death syndrome, suggests involvement in testis differentiation (Puffenberger et al.

2004). There are two autosomal (TSPYL1, TSPYL4) and one X-chromosomal (TSPYL2) paralogs identified in the current porcine genome, but their functions are unknown.

The purpose of this study was to characterize the porcine Y-chromosomal TSPY gene through the detection of several loci along the gene, determine its copy number and CNV status.

Here we report, for the first time the copy number of TSPY in four breeds of domestic pig.

MATERIALS AND METHODS Animals and DNA extraction

Peripheral blood samples from 20 unrelated boars of normal reproductive health from four breeds (5 from each of Piétrain, Landrace, Duroc and Yorkshire) and a female control animal were obtained from local producers. Sampling was done as part of the general animal health check and mandatory sampling for CFIA (Canadian Food Inspection Agency) tests, according to the Canadian Council on Animal Care and University of Guelph’s Animal Care

Committee guidelines by licensed veterinarians. These animals were not selected for research purposes, but regular breeding animals at various Canadian farms and leftover blood samples were used for DNA extraction using standard phenol-chloroform technique (Miller et al. 1988) and concentrations were determined by a NanoDrop Spectrophotometer.

In order to extend the panel of the 21 tested animals, of which we had no specific fertility information available other than their general good reproductive health, 10 additional unrelated boars with known fertility status (five high-fertility and five low-fertility) were chosen from a separate population (Revay et al. 2015). The fertility indicator parameter was the direct boar effect on litter size (DBE) that is the number of piglets the given boar produces in average per litter (appendix XXI) as compared to the overall average of the population and corrected for all identified environmental effects and breeding values of their mates (Tribout et al. 2000). DNA samples of these animals were retrieved from the owner’s DNA bank. It should be noted that the method of DNA extraction and the age of the DNA samples were variable and could not be controlled in this study.

Relative TSPY copy number determination by qRT-PCR

Quantitative real-time PCR (qPCR) was used to determine TSPY CN relative to the androgen receptor gene (AR, NM_214314.2, localized on the X-chromosome: NC_010461.4), a single copy reference gene. The porcine TSPY gene sequence was identified using the

Sscrofa10.2 genome assembly, that contains the raw Y-chromosome sequence (NC_010462) and is built from seven scaffolds. One of the scaffolds (NW_003536871.2) contains the predicted

TSPY4 gene (testis-specific Y-encoded 4-like, Gene ID: 100625034). The predicted mRNA sequence (XM_003360532.2) was aligned to the Y-chromosome DNA sequence and used as source for PCR assay development. Primers were designed to amplify exons 1, 3 and 5 of the

TSPY gene (Fig. 33) and exon 2 of the AR using the Primer3 plug-in of Geneious software.

Primer sequences, annealing conditions and product sizes are in appendix XXII. qPCR was performed using a CFX96 Touch Real-Time PCR Detection System (Bio-Rad) under the following thermal profile: 98°C, 2 min; 49×(98°C, 15 sec; 60°C, 15 sec). A melting curve was then generated between 72°C to 95°C in 0.5°C/sec increments. The 10µl reaction mix consisted of 1× SsoFast EvaGreen Supermix (Bio-Rad), 2.25mM primers and 5ng genomic DNA. Samples were run in triplicate and an inter-run calibrator sample was selected (one Duroc boar) and included in all runs to minimize technical variations. The TSPY CN was determined relative to the calibrator sample (Eq 0.1) after normalization against the single copy reference gene using the efficiency corrected ΔΔCq formula (Eq 0.2-0.3), as described in detail previously (Hamilton et al. 2009).

where ΔCq=(CqCAL-CqSample) is the difference of the threshold cycle values for the calibrator and the sample, while E is the primer efficiency for each amplicon. The efficiency values were calculated by two different ways. Method STD: from STD curves created from 2- fold serial dilutions (16ng - 0.25ng) of DNA mixed from the 20 samples and analyzed in CFX manager (Bio-Rad). Method LinReg: the average of individual well efficiencies were calculated by linear regression of amplification curves using the LinRegPCR software (Ramakers et al.

2003; Ruijter et al. 2009).

Figure 33. Schematic view of the porcine TSPY gene (exons 1-5 in grey) and location of primers (green), TaqMan probes (red), PflMI enzyme cut positions (blue) and FISH probes (purple).

Absolute copy number determination by droplet digital PCR

Droplet digital PCR (ddPCR) was performed using the QX100 Droplet Digital PCR

System (Bio-Rad). The TaqMan chemistry (5’ FAM – internal ZEN – 3’ IowaBlack FQ

98 quenchers) was used to detect the amplicons instead of the EvaGreen dye, used in qPCR described above. TaqMan probes were designed to target TSPY exon 1, 3 and 5, as well as AR

(Fig. 33) using the PrimerQuest software (IDT). Prior to ddPCR 5g of DNA from each of 21 animals (the same 20 boars + one female control used for qPCR) were digested with FastDigest

PfIMI restriction endonuclease (Thermo Scientific) in 50l 1× FastDigest buffer for 5 min at

37°C, followed by 10 min inactivation at 65°C. This enzyme was chosen, as the simulated restriction digestion with all commercially available restriction endonuclease in Geneious software showed cut positions surrounding the PCR amplicons (TSPY, AR), but not within any of them (Fig. 33). This digestion step helps to prevent tandem repeats from being encapsulated into the same droplet, thus causing false CN measurement. The ddPCR (Bio-Rad) is a three-step procedure (Hindson et al. 2011). First, the reaction mix (50ng digested DNA, 900nM primers,

250nM TaqMan probe, 1×ddPCR supermix in 25l) was partitioned into 20,000 uniform 1-nl volume droplets using the QX100 droplet generator by mixing 20l of reaction mix and 70l of droplet oil (Bio-Rad) in the generator cartridge. Then 40l of the resulting emulsion was transferred to a PCR plate and DNA was amplified under the following thermal cycle conditions:

94°C, 10 min; 45×(94°C, 30 sec; 60°C, 1 min with 2.5°C/sec ramp); 98°C, 10 min. The third step, the automatic signal detection was performed using the robotic droplet reader (Bio-Rad) that serially aspirated each sample, separated the droplets and detected the fluorescence in each droplet. Each sample was run in duplicate. The TSPY CN was calculated by dividing the measured target concentration (copies/l,) with the concentration of the AR single copy reference gene (Eq 0.4, Quantasoft detection software, Bio-Rad).

Statistical analysis

For statistical analysis all data was subjected to D’Agostino-Pearson normality test, then comparisons of efficiency values were done using the Kruskal-Wallis test and CN values of different breed and exon groups were done using two-way ANOVA. A p-value of less than 0.05 was considered significant. The data was analyzed using GraphPad Prism6 software (GraphPad

Software).

Fluorescence in situ hybridization (FISH)

Metaphase chromosome spreads were prepared from short term lymphocyte culture according to standard cytogenetic techniques. Whole blood (1.2ml) from the same boar used as calibrator for PCR was cultivated in 10ml RPMI 1640 (Invitrogen) medium containing 10% FBS

(Invitrogen), 0.15% penstrep (Invitrogen) and 0.06% phytohemagglutinin (Invitrogen) for 3 days at 37°C. KaryoMax colcemid (0.025g/ml, Invitrogen) was added to the culture for the last 25 minutes. Slides were prepared by incubating in 0.075M KCl for 20 minutes at 37°C followed by dipping three times in a MeOH/AcOH (3:1) suspension and aging of the slides at room temperature for 3 days before using for FISH.

Four different biotinilated hybridization probes were prepared by PCR. The primers used for the copy number determination were combined across exons to result in different sized probes (Fig. 33). Probes were named according to the exons content, as FISH#1-5 covered the whole length (1846bp), while FISH#1-2 (1187bp), FISH#2-3 (337bp) and FISH#4-5 (404bp), targeted only a part of the gene. PCR mixes (50l) contained 80 ng male genomic DNA, 5 units

AmpliTaq Gold DNA Polymerase (Applied Biosystems), 1PCR Buffer II, 1.5 mM MgCl2, 200

M each dATP, dCTP, dGTP, 133 M dTTP, 67 M biotin-11-dUTP (Metkinen) and 0.5 M of each primer. All PCR reactions were performed in the MJ Research PTC-200 Thermo Cycler

100 using the following thermal profile: 94°C, 10 min; 35×(94°C, 30 sec; 60°C, 30 sec; 72°C, 30 sec), 72°C, 10 min. PCR products were purified and concentrated to 20l final volume using a

Amicon Ultra 30K centrifugal filters (Millipore) before adding to the hybridization mixture, that was composed of 50% (v/v) formamide (Fisher), 10% (w/v) dextran-sulfate (Sigma-Aldrich),

15g salmon sperm DNA (Life Technologies) and 2×SSC.

The FISH experiments were performed with standard protocols (Raap 2001), briefly: slides were treated with pepsin, dehydrated in ethanol series, denatured at 72°C for 2 minutes in

70% formamide (Fisher)/2×SSC, then quenched in ice-cold ethanol. The hybridization mix was denatured at 72°C for 10 min before applying to the slide. After overnight hybridization at 37°C, slides were washed twice in formamide/2×SSC (50:50), then twice in 0.2×SSC at 47°C and

PBS/0.5% Tween 20 at room temperature for 3 minutes each. Slides were blocked in 1% BSA in

PBS for 30 minutes at 37°C before detection. Biotinylated probe signals were visualized by one of the following two methods: 1. alternating layers of FITC-avidin (Vector, 1:400 in PBS/0.5%

Blocking Reagent (Roche)) and anti-avidin-FITC antibody (Cambio, 1:250 in PBS/0.5%

Blocking Reagent) incubated for 30 minutes at 37°C and washed by 3×(PBS/0.5% Tween 20) at room temperature. 2. Alternatively, the Tyramide signal amplification method was applied using streptavidin-HRP and AlexaFluor 594 tyramide (LifeTechnologies) according to the manufacturer’s protocol. Chromosomes were counterstained with DAPI (Sigma-Aldrich) and slides were mounted with Vectashield (Vector). Images were captured using a Leica DM5500B fluorescence microscope (Leica), equipped with a Retiga Exi Fast (QImaging) cooled digital camera and the OpenLab imaging software (Perkin Elmer).

101

RESULTS TSPY copy number by qPCR

We have measured the TSPY CN at three loci along the gene relative to AR by designing primers for exons 1, 3 and 5 (Fig. 33). PCR amplification efficiencies were calculated for all three loci by two different methods (STD curve and LinRegPCR). PCR efficiencies of the

TSPY-E1 assay (TSPY-E1 STD E=1.89 and TSPY-E1 LinReg E=1.82) were found to be different (Kruskal-Wallis test, p<0.01) from the other amplicons (AR, TSPY-E3, TSPY-E5), which were all very close to the theoretical ideal value of 2.00 (appendix XXII).

The relative qPCR results in general, showed a highly variable TSPY CN ranging from 2-

5 copies depending on the locus (E1, 3 or 5) or the method used for the calculation of PCR efficiency (STD or LinReg, Fig. 34, appendix XX).

Significantly higher average copy numbers were calculated at all three exons (TSPY-E1

STD CN=4.89±1.01, TSPY-E3 STD CN=3.85±0.42, TSPY-E5 STD CN=3.80±0.61) when the

STD curves were used to determine PCR efficiency values, as compared to the TSPY copy number at the same locus calculated from LinRegPCR efficiencies (TSPY-E1 LinReg

CN=1.66±0.60, TSPY-E3 LinReg CN=2.06±0.23, TSPY-E5 LinReg CN=2.69±0.43).

The next level of comparison was among exons using either the STD or LinRegPCR efficiency calculation method. TSPY-E1 STD CN was significantly higher (4.89±1.01), than

TSPY CN at exon 3 or 5 (TSPY-E3 STD CN=3.85±0.42, TSPY-E5 STD CN=3.80±0.61).

However, TSPY-E5 was significantly higher (2.69±0.43) than the copy number at exons 1 and 3

(TSPY-E1 LinReg CN=1.66±0.60, TSPY-E3 LinReg CN=2.06±0.23) when the LinReg PCR was used to calculate efficiencies. This also suggests the dependency of CN calculation on efficiency values.

102

Figure 34. TSPY CN relative to the single copy AR gene, as determined by qPCR. CN was measured at three loci (Exon 1, 3, 5) in four breeds. Columns represent the average value (±SD) of 5 animals from the same breed. Different letters within efficiency calculation methods (STD or LinRegPCR) represent statistical significance according to two-way ANOVA (p<0.05).

103

We have included Duroc, Landrace, Pietrain and Yorkshire boars in the investigation of potential breed specific CN variability. We found that the breeds showed very similar copy numbers at the same exon locus within STD or LinRegPCR calculations, with only a few exceptions, as labeled in Fig. 34 A and C.

TSPY copy number by ddPCR

The ddPCR experiments determined the porcine TSPY as a tricopy gene. The average

TSPY CN among all boars tested was 3.01±0.08 and there were no differences among the three exons measured (TSPY-E1 CN=2.99±0.08, TSPY-E3 CN=3.06±0.07, TSPY-E5 CN=2.98±0.10,

Fig. 35, appendix XXI). The 20 boars, regardless of breed, showed no differences in TSPY CN

(Fig. 35).

We detected no positive droplets in the control female samples for all three exon specific

TSPY primers, which clearly confirmed their specificity for only male specific Y-chromosomal sequences.

We found no CN difference among the high-fertile and low-fertile animals at any of the three exons, however the values were generally slightly lower than that of normal animals, as measured by ddPCR (Fig. 35, appendix XXI). The average CN values for the high-fertile group were TSPY-E1: 2.56±0.16, TSPY-E3: 2.76±0.29, TSPY-E5: 2.62±0.2 which were similar to the low-fertile group averages at TSPY-E1: 2.57±0.16, TSPY-E3: 2.85±0.19, TSPY-E5: 2.70±0.21.

The average CN at TSPY-E1 for both high- and low-fertile groups were statistically lower than the average CN of normal animals at all three loci (Fig. 35). The CN of TSPY-E5 in both high- and low-fertile groups were also significantly lower than TSPY-E3 CN in normal animals.

104

Figure 35. TSPY CN by ddPCR at exons 1, 3 and 5 in normal animals from four breeds, High-fertile and Low-fertile animals. Columns represent the average value (±SD) of 5 animals. Letters mark values that are statistically different (p<0.05).

105

Chromosomal localization of TSPY by FISH

We performed FISH using 4 different sized probes (Fig. 33). The two longest probes

(FISH#1-5 (Fig. 36) and FISH#1-2) resulted in clear, intense and specific signals on the short arm of the Y-chromosome when the Tyramide signal detection method was applied. The same probes with the conventional avidin-FITC detection method and the two shorter probes (FISH#2-

3, FISH#4-5) did not produce reliably detectable signals (data not shown).

Figure 36. Y-chromosome specific fluorescence in situ hybridization signal of FISH probe#1-5 with Tyramide signal detection on a boar metaphase. DAPI stained metaphase chromosomes are converted to grayscale and AF594 hybridization signal on the short arm of the Y-chromosome is red.

106

DISCUSSION

The structural arrangement of the porcine Y-chromosome is not fully characterized. The

Sscrofa 10.2 genome assembly (Groenen et al. 2012) contains the Y-chromosome sequence that is built from seven partially annotated scaffolds. The aim of our study was to determine the previously unreported Y-chromosomal TSPY gene copy number in pigs and investigate whether there is CNV between breeds and individual animals.

To ensure the male specificity of the GenBank predicted TSPY sequence that served as the template for assay designs, we localized it by FISH on metaphase chromosomes. We chose to generate probes from combinations of the actual qPCR primers in order to avoid potential inclusion and detection of repetitive DNA from large-insert clones (Fig. 33). Two of the four different probes produced signals specific for the short arm of the Y-chromosome (Yp, Fig. 36), but an enhanced signal amplification method was needed (Speel, Hopman, and Komminoth

2006). This is still a surprising result, especially in light of the low TSPY copy number, as 1-2 kb long probes are generally suitable for physical mapping of highly repetitive sequences, in order to produce signals of detectable strength (Raap 2001). The specific localization of TSPY on Yp is in agreement with a previous study by Quilter et al. (Quilter et al. 2002), who mapped TSPY among other genes by radiation hybrid panels and BAC clone based FISH.

The qPCR assays were designed to amplify three different exons (E1, 3, 5) from the currently known TSPY sequences assigned to the porcine Y-chromosome. The investigation of three different loci along the TSPY gene allowed the detection of the potential partial amplification/deletion of the gene. This process occurs frequently on various loci of the Y- chromosome e.g. AZF regions, due to intrachromosomal recombination of repeated elements

(Repping et al. 2002). Primers were also tested and proved to be negative against female DNA to

107 confirm that only Y-chromosomal TSPY gene copies and not autosomal or X-linked paralogs were detected.

We characterized the TSPY CN by qPCR, which is a frequently used technique for accurate copy number profiling, especially if the calculations are relative to a standard sample with known target CN (D’haene et al. 2010). To apply qPCR for CN discovery, thus without having any prior information on the TSPY CN, we used the methodology described by Hamilton et al. (Hamilton et al. 2009b). We calculated the CN of a random selected animal, dedicated as calibrator, and the CN of all samples were calculated relative to the calibrator. The TSPY CN determined by relative qPCR showed surprisingly low CN that varied between 2 and 5 copies at the individual, breed and exon levels. It is likely that the main source of this variability is the dependency of calculations on PCR efficiency values. This has been clearly shown when minimally different efficiency values determined by two methods for efficiency calculation

(STD curve vs. LinReg PCR) resulted in significantly different CNs (Fig. 34). Observations by others also pointed to the lower precision of qPCR for a low CN range (Pinheiro et al. 2012;

Huggett et al. 2013) and the impact of DNA quality variations among the samples could also not be ruled out.

To further investigate TSPY CN and its CNV status we applied ddPCR, a recently developed technique for absolute quantitation of nucleic acids, to the same samples that were used for qPCR (Hindson et al. 2011). In contrast to qPCR, where the quantitative information is calculated from the parameters of the real-time registered amplification curve (threshold cycle or

Ct value and efficiency), ddPCR measures the absolute quantity of DNA in thousands of nanolitre volume reaction partitions (droplets) generated from the homogenous reaction mix.

This method employs end-point detection (positive or negative PCR amplifications in each

108 droplet), thus independent of reaction efficiency, which represents a major advantage in our

TSPY CN discovery study (Huggett and Whale 2013; Pinheiro et al. 2012). The ddPCR experiments clearly identified the porcine TSPY as a multicopy gene with three genomic copies

(CN=3). We did not observe any copy number variations among the 20 individual samples or the four breeds investigated. Moreover, no difference was found among the three exons (Fig. 35).

Although qPCR is the standard and proven technique for relative quantitation, its applicability for DNA CN discovery, especially in the low CN range, is limited by the lack of a standard sample with known CN (D’haene et al. 2010).

The low copy non-CNV status makes the pig TSPY gene similar to that of the chimpanzee and rhesus monkey, which also have low CNs (CN=6, although CNV status has not been reported), and very different from humans and cattle, both of which have high TSPY CNs and show inter-individual variation (Hamilton et al. 2009a; Xue and Tyler-Smith 2011). This

CNV has also been associated with various reproductive phenotypes. Low TSPY CN was found in men with impaired sperm production (Giachini et al. 2009; Shen et al. 2013; Krausz, Giachini, and Forti 2010), although other human studies with smaller or more heterogeneous donor population reported contradictory results. Nickkholgh et al. (2010) found no association between

TSPY copy numbers and severe spermatogenic failure while Vodicka et al. (2007) found a significantly higher number of TSPY copies in infertile men compared with the controls

(Nickkholgh et al. 2010; Vodicka et al. 2007). Also, a positive correlation was found between

TSPY CN and field fertility (adjusted non-return rates) in Holstein and multiple seminal quality parameters of crossbred (Bos taurus × Bos indicus) bulls (Mukherjee et al. 2015; Hamilton et al.

2012). To further investigate the speculation that the porcine TSPY is not a CNV, we have included five low-fertility boars with negative DBE scores and five boars with exceptionally

109 high-fertility records. Interestingly, there was no difference between the high and low-fertile groups, but we detected slightly lower TSPY CN at all three exons (2.56-2.85) in both groups, as compared to normal animals (Fig. 35) Most probably this difference would not show up as an alteration in clinical practice and would still be interpreted as CN=3 (D’haene et al. 2010). We attributed the difference in our study to technical variability of the tests resulting from the age and quality of the latter set of DNA samples that had been banked for several years.

Additionally, we could not exclude the biological possibility that the fractional and in some cases statistically different (exons 1, 5) copy numbers are due to somatic mosaicism. Furthermore, although our PCR analyses were designed to amplify all TSPY sequences in the currently unfinished porcine Y-chromosome, the presence of additional as yet to be discovered sequences cannot be ruled out.

Evolutionary investigation in humans has suggested that the highly amplified TSPY copy number has been affected by positive selection and might result in selection advantage through some functions in male reproduction (Xue and Tyler-Smith 2011) the non-variable, tricopy

TSPY in boars that we report here for the first time might represent the minimal number of functional copies to maintain fertility.

110

GENERAL DISCUSSION

The purpose of this thesis was to evaluate genome variation at different levels in relationship to boar fertility. The large genome variation, so called chromosome abnormalities and medium genome variation known as copy number variations (CNVs) were studied by different whole genome snapshot techniques. Specifically, 775 pigs were karyotyped using the

G-band technique to detect chromosomal abnormalities, which are known to be related to the production of significantly fewer piglets per litter, resulting in hypoprolific boars. Then, from a population of more than 38000 boars, high-density genotypes of 60K SNP of 21 boars with extremely low or high fertility were chosen and analyzed for CNVs specifically related to fertility. Finally, the number of copies of TSPY, a well-known multi-copy Y gene associated with fertility in bulls and humans, was examined in 30 boars by qPCR and droplet digital PCR

(ddPCR) techniques.

Firstly, we used G-band technique, as a simple means of detecting very large genomic variations (> 5Mb, chromosome level) including not only unbalance variations such as deletion and duplication but also balance structural variations such as reciprocal translocations, inversions and Robertsonian translocations, which usually cannot be detected by genome-wide techniques including SNP genotyping. In total, 732 young Canadian AI boars (340 Duroc, 222 Landrace,

146 Yorkshire and 24 Pietrain) and 43 randomly chosen Vietnamese boars and sows (4 Duroc,

23 Landrace and 16 Yorkshire) were karyotyped. This makes our study the largest systematic screening program for chromosomal abnormalities among young breeding boars in Canada to date. For the Vietnamese pig population, this is the first time screening for chromosome abnormalities was carried out in this pig industry. Twelve carriers of chromosome abnormalities were found in Canadian pigs, thus the frequency of chromosome abnormalities in this pig

111 population is 1.64% (0.76 to 2.56%). For the Vietnamese pig population, 2 carrier sows were found in 43 samples that defines 4.7% (1.6% to 10.9%) prevalence. More samples need to be karyotyped to have a more accurate estimate of the frequency of chromosome abnormalities for the Vietnamese pig population. Although numbers were small, the data indicated that in

Vietnam, the pig population could have a very high frequency of chromosome abnormalities, even higher than the Canadian pig population and that this could result from this being the first time this pig population was screened for chromosome abnormalities and to date no selective elimination has been practiced.

Nine of the chromosome abnormalities reported here were associated with smaller litter size when compared with the herd average, a feature characteristic of chromosome rearrangements that has previously been reported in the literature ( a Pinton et al. 2000; a. Ducos et al. 2008). Among reciprocal translocations detected in this study, case 3, rcp(2;5), is associated with the highest adverse effect on fertility while case 5, rcp(3;12), has the lowest association, with a 46% and 17% reduction in litter size, respectively. The lower fertility of Case

4, rcp(3;4), was not noticed by the breeder and as a result, 4 of its relatives (dam and 3 sisters) were used for breeding and carried the same translocation and the carrier boar himself had the highest use recorded in our study, with 73 litters. Chromosomal abnormalities are well known to cause large economic consequences for the swine industry (Popescu and Tixier 1984; Popescu et al. 1984; Popescu 1990). This study was a first step towards showing the Canadian pork industry the important value of implementing a screening protocol for boars in Canadian AI centers. Early screening would be beneficial as it would take several months before low fertility was observed on the farms for the following reasons.

112

How much money is lost by using a carrier during the minimum 4 months before litter numbers can be calculated? For example, in our case 4, the boar had an average litter size of 9.2 piglets/litter, which represent 2.6 piglets less/litter compared with the herd’s average litter size of

11.82 piglets/litter. Most producers will not notice this small loss because many litters are born at the same time. However, the total piglets lost which this farm had already had due to this translocation was 2.6piglets/litter x 73 litters = 189.8 piglets. At an estimated price of $30 per piglet, that represents more than $5000 loss for this particular carrier boar until it was removed from service as a result of the karyotype information that was sent to the farm. Luckily, this rcp(3;4) translocation only caused a small reduction in prolificacy (22%). In contrast, the boar in case 3 rcp(2;5), had a 46% reduction in prolificacy or 6.1 piglets loss/litter, thus the total piglets reduction for the farm due to this translocation was much more dramatic. However, because of the overall small reduction in piglet numbers due to multiple litter births, the rcp (3;4) translocation was less noticeable to the farmers. the dam was the carrier and her carrier son and 4 carrier daughters were kept for breeding on the farm. If this rcp(3;4) boar had not been karyotyped, this translocation would have been much more wide spread and as a consequence the efforts to completely remove this translocation and the associated decline in litter size in this herd would be much more difficult.

Another important question that must be addressed is whether the cost for routine screening for chromosomal abnormalities outweighs the cost of testing? In France, even with a lower frequency of chromosomal abnormalities, only 1/200 (0.47%), they proved that the cost of using one translocation boar after deducting for the testing fee is at least 11500 CAD (8000 euro)

(Ducos et al., 2008). In Canada and Vietnam, the translocation frequency is more than three times higher (1.64%, 4.7%, respectively) and initiating screening for chromosome abnormalities

113 in AI boars will be even more profitable for the Canadian swine industry. Karyotype analysis, based on subsidized price of $250 per test, would eliminate carriers of chromosomal abnormalities and reduced litter sizes before substantial breeding occurred and would far outweigh the $250 invested for testing. The boar in case 4, sired 73 litters, if the average litter size is 9 for this boar and half (4.5) of the progeny are carriers, this would mean that 328 progeny would carry the chromosomal translocation and potentially be capable of being selected for breeding for future generations. If even only 10% (32) of these pigs are selected as breeding stock, the abnormality can still be transmitted for many more generations.

Among 5 of the cases for which we could determine the origin, 3 of them were inherited from either the maternal or paternal line. Because these pig populations were not broadly involved in a screening program before, a high number of inherited abnormalities among the population are expected. In addition, among 3 cases for which we have fertility data for the parents and grandparents for predicting the origin, 2 are suspected to be inherited and 1 is suspected to be de novo (see chapter 1). If this assumption is correct, among 8 rearrangements, 3 were de novo and 5 were inherited. It is important to emphasize that 3 cases out of 8 are de novo, which is a rate of almost 40% in pig. That means the chance of having new chromosome abnormalities is very high even if the parents were in a screening program and the more than 180 different rearrangements detected in pigs emphasizes that the pig genome is fragile and prone to change. Therefore, screening for chromosome abnormalities in the pig is important and needs to be broadly implement and continued in all AI boars because even though all existing rearrangements will eventually be eliminated, the de novo rearrangements will continue to spontaneously occur.

114

How does one minimize the cost and get the most benefit from the test for chromosome abnormalities? First, all breeding pig farms should persistently screen for chromosome abnormalities in all boars destined for use in AI to detect all the inherited and de novo chromosome abnormalities. For the inherited carriers, in some cases, they can be suspected via average litter size and DBE. Nevertheless, rearrangements and mutations can arise spontaneously in animals, especially in pigs and can to a certain extent, be undetected in a herd. Therefore, screening for chromosomal abnormalities in boars entering AI centers is imperative. As soon as a carrier boar is detected, the parents of the carrier should be sampled to identify how widely the spread of that translocation is in the herd. In the absence of available samples from the parents, it is not possible to determine whether the translocation was inherited or occurred de novo. The optimal strategy then is to sample all available full-sibs and the detection of the abnormality in any other littermates would confirm its inherited status. Hence, strategies of identifying and eliminating further animals with abnormal karyotypes from the herd of origin were given and such follow-up samples were collected.

Secondly, many hypoprolific boars without any identifiable chromosome abnormalities possibility contain certain patterns of CNVs which are related to low prolificacy. Those CNVs are small DNA segments (DNA level) that were not able to detect by the G-band technique.

Hence, the second step presented in chapter 2 is a complement for our first step of chromosomal abnormality assessment bydetecting genome variations by using SNP 60k beadchip, a technique that can detect only small deletion or duplication variations known as CNVs, but is not able to detect translocations such as inversion, reciprocal translocation and Robertsonian translocation that we were able to detect in chapter 1.

115

The data presented in chapter 2 and 3 of this thesis focus on a type of genetic variation,

CNV. The importance of CNVs has been reported in hundreds of studies and is linked with a wide range of diseases in humans (Amar et al. 2010; Anney et al. 2010; Bassett et al. 2010; de

Vries et al. 2005; Rovelet-Lecrux et al. 2006). In cattle, several CNVRs contained genes related to environmental stress (ie. MHC complex genes) and sensory function (ie. olfactory receptors)

(Fadista et al., 2010, Lemay et al., 2009, Liu et al., 2008, Liu et al., 2010) have been identified.

Similar to cattle, olfactory receptors and sensory perception of smell are highly involved in

CNVRs of most large-scale CNV studies in pigs (Paudel et al. 2015; Jiying Wang et al. 2012).

Moreover, genes associated with neurological processes, response to stimulus, some basic metabolic processes, drug metabolism and immunity have also been reported to be localized within CNVRs (Jiying Wang et al. 2014; Jiying Wang et al. 2012; Paudel et al. 2013). All of these genes found in CNVRs were discovered in general pig populations assessing different breeds. The highlight of our study about pig CNV is that it was studied different breeds that were divided into two groups - high and low fertility. Therefore, the list of genes found on CNVRs

(appendix XIV) in this thesis is focused especially on the association with fertility. Our functional analysis of several databases revealed that the genes found in CNVRs from the low fertility group have been significantly enriched in members of the innate immune system, Toll- like receptor and RIG-I-like receptor signaling and fatty acid oxidation pathways (chapter 2).

As for most large-scale CNV studies, the Y chromosome was omitted from the swine

CNV projects due to the peculiar highly repetitive nature of this chromosome resulting in difficulties in sequencing. Similarly, in chapter 2, we used SNP 60k beadchip which also did not contain the Y chromosome. However, it is obvious that the Y chromosome is the most important chromosome for fertility in male, which is the focus of this thesis. In relationship with CNV,

116

TSPY, a gene on the Y chromosome has shown links to male infertility in humans (Vodicka et al. 2007). More interestingly, in cattle, the variations in the number of copies of this gene are associated with fertility (de Vries et al., 2002, Giachini et al., 2009, Nickkholgh et al., 2010,

Vodicka et al., 2007, Hamilton et al. 2012). Thus, TSPY, the most interesting candidate gene on the Y chromosome, was chosen to study in detail as presented in chapter 3.

The TSPY gene has the highest known CN of all protein coding genes in the human and bovine genomes (up to 74 and 200, respectively) and also shows high individual variability

(Hamilton et al. 2009). In pig, the number of copies of TSPY has never been reported in the literature. However, not as expected, our result showed that porcine TSPY is a multicopy gene with three copies located on the short arm of the Y chromosome with no variation at three exon loci among 20 animals of normal reproductive health from four breeds of domestic pigs

(Piétrain, Landrace, Duroc and Yorkshire). To further investigate the speculation that porcine

TSPY is not a CNV, we included samples from five low-fertility boars and five boars with exceptionally high-fertility records. Interestingly, there was no difference in TSPY CN between the high- and low-fertile groups, but we detected slightly lower TSPY CN at all three exons

(2.56-2.85) in both of these groups, as compared to the normal animals tested previously, which could be attributed to technical variability or somatic mosaicism.

Data presented in chapter 1 and 2 of this thesis showed that chromosome number 2 in pig is a very interesting chromosome for fertility. In this study, chromosome abnormality, rcp(2;5), which had the break on the tip of chromosome 2, had the highest adverse effect on fertility among the carrier animals. Moreover, this chromosome 2 in the CNVs study related to fertility is outstanding for the highest number of CNVRs (8) and the largest region involved in them

(CNVRs occupy 7.6% of this chromosome). In addition, CNVRs in chromosome number 2

117 contained 30 out of 50 total genes found in this study; making it the chromosome containing the highest number of genes on CNVRs. Most of these genes are located on both tips of this chromosome including 16 genes on the tip of thr p arm and 12 genes on the tip of the q arm

(figure 32 in chapter 2). Among the genes located on the tip of the p arm, 14 genes located on the

CNVR7L with gain status were found in boars with low fertility while the other 2 genes located on the CNVR8H with loss status were found in boars with high fertility (appendix XIV).

Although we were not able to map the breakpoint on rcp(2;5)(p16;p11) to its accurate position on chromosome 2 to know whether it overlaps with these CNVRs or not, the high number of genes located in the CNVR around this breakpoint (2p16) may be somehow affected by the reciprocal translocation resulting in a more severe effect of this translocation on litter size of the carrier boar. Reviewing 16 reciprocal translocations that have been found on chromosome 2

(appendix I and II), the majority of the breakpoints are located on the tips, especially the tip of the p arm of this chromosome. The majority of reproduction QTL regions overlapping with

CNVRs was also found in the tip of p arm of chromosome 2 (figure 32 in chapter 2), suggesting that these regions are important for fertility. More studies about genes related to CNV found on chromosome 2 (especially on the tip of p arm) hold the promise of explaining the high reduction of litter size of rcp(2;5) and the importance of this special region for fertility in the pig. Another very interesting aspect about chromosome 2 is its frequent involvement in complex structural translocations, such as a reciprocal translocation between 3 or more chromosomes which usually causes a very severe reduction in litter size. Only 3 out of 180 rearrangements reported in the literature are complex translocation including (1;2;17)(1q11;1q17;2p16;17q13) (Perucatti et al.

2014), (2;15;4)(p13;q26;p15) (Makinen et al. 1987) and (2;9;14)(q23;q22;q25) (Mäkinen et al.

1997) and all have chromosome 2 involved. Animals with two or more rearrangements in the

118 same individual are also rare, for examples, (13;14)(q31;q21) + inv(2)(p13;q12) and

(10;18)(p11;q24) +(6;8)(p15;q27) (Alain Ducos et al. 2007). Only 2 cases have been reported, interestingly, one of them is related to chromosome 2.

Chromosome 12 is also a chromosome brought to attention in this thesis. It was found in relation to cases 5, rcp(3;12)(p13;q15), and 9, rcp(12;14)(q15;q23). The breakpoints of both cases were on the tip of the q arm (12q15) on chromosome 12. With regard to CNVs, this chromosome had the largest percentage region involved per chromosome (8%) and a high number of genes (10) located on those CNVRs (appendix XVI). There are 4 genes (UBB,

SLC5A10, COPS3 and PEMT) located on the tip of this q arm (on CNVR51L) with gain status that were found in samples from low fertility boars. Similarly, we were not able to map the break

12q15 in cases 5 and 9 to the accurate position to determine whether they are overlapped with those CNVRs or genes. However, case 5 which is contrary to case 2, has the least effect on reduction of litter size (-17%) comparing among 7 reciprocal translocations. In brief, more studies on CNVRs and genes on chromosome 2 and 12, specific on translocation carriers (case 2,

5 and 9) may help to explain the relationship between CNV, gene, reciprocal translocation and fertility in pig.

One important limitation of this study is that the 19 high fertile and 19 low fertile animals that were used for the CNV study (chapter 2) were not able to be karyotyped to be confirmed free of chromosome abnormalities. Because, as mentioned in chapter 2, these animals had been genotyped for other projects several years ago, screening for chromosome abnormalities in this project was started only two years ago. Moreover, to have DBE available for samples selecting for CNV related to fertility, the boar needs to have sized at least 10 litters or the boar has to have been used for at least 4 months. By that time most low fertility boars have been culled and as a

119 result it is very difficult to obtain samples of those boars for karyotyping. For this reason, in this preliminary study, we picked only 36 boars for the CNV study. The option for studying CNV of boars confirmed free of chromosome abnormalities is possible in the future. Screening for chromosome abnormalities is still going on and karyotype status of a boar has been suggested to be recorded at the same time with all other information about fertility of that boar in the CCSI database. Hence, it will facilitate the choice of animals for CNV study related to fertility in the future.

120

SUMMARY, CONCLUSIONS AND FUTURE DIRECTIONS

Screening 775 breeding pigs allowed the identification of 11 new different rearrangements that have never been reported in the literature including rcp(1;5)(q21;q23), rcp(1;15)(q211;q13), rcp(2;5)(p16;p11), rcp(3;4)(p15;q13), rcp(3;12)(p13;q15), rcp(6;7)(p15;q13), rcp(7;15)(q13;q13), rcp(8;13)(p21;q41), rcp(12;14)(q15;q23), inv(8)(q11;q25) and inv(6)(q23;q35). Moreover, one previously reported Robertsonian translocation rob(13;17) was also found in this study.

This study was significant for breeders because it allowed for the identification of 12 boars in the AI centers and 2 breeding sows with chromosome abnormalities. If using one translocation boar costs 11,500 CAD, for 12 boars, this is a saving of nearly 138,000 CAD for the Canadian pig industry. Moreover, it helps to prevent the transmission of these abnormalities and associated deleterious effects to subsequent generations.

Comparing with the herd average, average total born litter size of rcp(1;15), rcp(2;5), rcp(3;4), rcp(3;12), rcp(7;15), rcp(8;13), rcp(12;14) and inv(6) translocation carrier boars were noted to be reduced (27%, 46%, 22%, 17%, 36%, 28%, 27% and 24%, respectively) while for carriers of rob(13;17), it was only slightly reduced (4%). Stillborn and mummified foetuses of all reciprocal translocations that had fertility data available were also reduced when compared to the herd average. Reciprocal translocations had a wide range of adverse effects on fertility while

Robertsonian translocations had very small effects but high prevalence of transmission rates to their progeny.

The obvious reduction in total born litter size of carriers, the high prevalence of chromosome abnormalities in Canadian pig populations (1.64%) and high percentage of de novo

121 cases (2/5 cases) highlight the importance of an obligate screening of all AI boars before entering the breeding stock at AI centers.

We have identified 35 CNVRs covering 36.5 Mb or ~1.3% of the porcine genome.

Among these 35 CNVRs, 14 were specific to the high fertility group, while 19 CNVRs were specific to the low fertility group and these CNVRs overlap with 137 QTLs of various reproductive traits of which the majority were teat number (TNUM) and age at puberty (AGEP).

The identified 35 CNVRs encompassed 50 genes, among them 40 were specific to the low fertility group, seven to the high fertility group and three were found in regions that were present in both groups but with opposite gain/loss status.

Three databases (KEGG, WikiPathways and Pathway Commons) were used to analyze for the genes in CNVRs. The genes found in CNVRs from the low fertility group have been significantly enriched in members of the innate immune system, Toll-like receptor and RIG-I- like receptor signalling and fatty acid oxidation pathways.

This study demonstrated that our analysis pipeline could identify putative CNV markers of fertility, especially in the case of subfertile boars. Their relevance was demonstrated by analyzing the nature of co-localized reproductive QTLs and genes.

These studies also showed that chromosomes 2 and 12 were very interesting chromosomes from a fertility perspective due to the high number of CNVRs, genes and QTL for fertility located on those chromosomes. Further studies such as sequencing carriers of rcp(2;5), rcp(3;12) and rcp(12;14) to locate the breakpoints in order to determine whether they are on the

CNVRs or genes found in this study should be completed. Moreover, detail studies on each of these genes to see if they might represent genetic markers of fertility in pigs would be beneficial.

122

Investigation of the copy number of the TSPY gene in the pig showed that although it was a multi-copies gene with 3 copies situated on the short arm of the Y chromosome, it did not vary among exons (E1, E3 and E5), individuals and breeds, and its copy number did not correlate with fertility. Hence it cannot be used as a marker of fertility.

In conclusion, this thesis emphasizes the importance of karyotyping boars entering A.I. centers for the Canadian pig breeders and breeding stock suppliers and could be used as part of the criterion for entry of a boar into an A.I. center. Moreover, from this thesis, a list of potential genetic markers of male fertility in the pig was found. However, more studies on each CNVR or gene on this list need to be carried out to elucidate the function of it on the fertility of boars.

Later, if association analyses reveal that a given CNV is associated with reduced or increased prolificacy, then the findings could lead to a commercial service to the industry where breeding stock suppliers could test for that marker in order to make sure that they have the best swine genetics available for the industry. The industry and the consumer will benefit from such an initiative.

123

LITERATURE CITED

“34th Centralia Swine Research Update, Kirkton Ontario 28 January 2015.” 2015. In . Kirkton - Woodham Community Centre. http://www.centraliaswineresearch.ca/proceedings/2015/centralia2015proceedings.pdf.

Abu-Halima, Masood, Christina Backes, Petra Leidinger, Andreas Keller, Abdel Monem Lubbad, Mohamad Hammadeh, and Eckart Meese. 2014. “MicroRNA Expression Profiles in Human Testicular Tissues of Infertile Men with Different Histopathologic Patterns.” Fertility and Sterility 101 (1). Elsevier Inc.: 78–86.e2. doi:10.1016/j.fertnstert.2013.09.009.

Alain Ducos, Alain Pinton, Martine Yerle, Anne Seguela, Helene Marie Berland, Corinne Brun-Baronnat, Nathalie Bonnet, Roland Darre. 2002. “Cytogenetic and Molecular Characterization of Eight New Reciprocal Translocations in the Pig Species. Estimation of Their Incidence in French Populations.” Genet Sel Evol 34: 389–406. doi:10.1051/gse.

Almal, Suhani H, and Harish Padh. 2012. “Implications of Gene Copy-Number Variation in Health and Diseases.” Journal of Human Genetics 57 (1): 6–13. doi:10.1038/jhg.2011.108.

Amar, Shirly, Ofer Ovadia, Wolfgang Maier, Richard Ebstein, R H Belmaker, Dan Mishmar, and Galila Agam. 2010. “Copy Number Variation of the SELENBP1 Gene in Schizophrenia.” Behavioral and Brain Functions : BBF 6 (January): 40. doi:10.1186/1744-9081-6-40.

Amaral, Alexandra, Judit Castillo, Josep Maria Estanyol, José Luís Ballescà, João Ramalho-Santos, and Rafael Oliva. 2013. “Human Sperm Tail Proteome Suggests New Endogenous Metabolic Pathways.” Molecular & Cellular Proteomics : MCP 12: 330–42. doi:10.1074/mcp.M112.020552.

Andersson, L. 2001. “Genetic Dissection of Phenotypic Diversity in Farm Animals.” Nature Reviews. Genetics 2 (2): 130–38. doi:10.1038/35052563.

Anielak-Czech B., A. Kozubska-Sobocinska, E. Slota, B. Rejduch and A. Kwaczynska. 1996. “Preliminary Identification of Pair 1 Chromosome Rearrangement in the Polish Landrace Sow.pdf.” Arch. Zootec 45: 215– 19.

Anita Kolodziej and Eugenia Jacyno. 2005. “Effect of Selenium and Vitamin E Supplementation on Reproductive Performance of Young Boars.” Arch. Tierz., Dummerstorf 48 (1): 68–75.

Anney, Richard, Lambertus Klei, Dalila Pinto, Regina Regan, Judith Conroy, Tiago R Magalhaes, Catarina Correia, et al. 2010. “A Genome-Wide Scan for Common Alleles Affecting Risk for Autism.” Human Molecular Genetics 19 (20): 4072–82. doi:10.1093/hmg/ddq307.

Archibald, Alan L, Lars Bolund, Carol Churcher, Merete Fredholm, Martien A M Groenen, Barbara Harlizius, Kyung-Tai Lee, et al. 2010. “Pig Genome Sequence--Analysis and Publication Strategy.” BMC Genomics 11 (January): 438. doi:10.1186/1471-2164-11-438.

Ashley, T. 1988. “G-Band Position Effects on Meiotic Synapsis and Crossing Over.” Genetics 118 (2): 307–17. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1203283&tool=pmcentrez&rendertype=abstract.

Astachova, N M, L V Vysotskaya, and A S Graphodatsky. 1991. “Detailed Analysis of a New Translocation in Pig [reciprocal Translocation].” Genetics Selection Evolution 23, suppl . INRA Paris (France): Genet Sel Evol: 65s – 69s.

124

Barasc, H, N Mary, R Letron, a Calgaro, a M Dudez, N Bonnet, Y Lahbib-Mansais, M Yerle, a Ducos, and a Pinton. 2012. “Y-Autosome Translocation Interferes with Meiotic Sex Inactivation and Expression of Autosomal Genes: A Case Study in the Pig.” Sexual Development : Genetics, Molecular Biology, Evolution, Endocrinology, Embryology, and Pathology of Sex Determination and Differentiation 6 (1-3): 143–50. doi:10.1159/000331477.

Barcellos, L F, A M Schito, J B Rimmler, E Vittinghoff, A Shih, R Lincoln, S Callier, et al. 2000. “CC-Chemokine Receptor 5 Polymorphism and Age of Onset in Familial Multiple Sclerosis. Multiple Sclerosis Genetics Group.” Immunogenetics 51 (4-5): 281–88. http://www.ncbi.nlm.nih.gov/pubmed/10803840.

Basrur, P K, and G Stranzinger. 2008a. “Veterinary Cytogenetics: Past and Perspective.” Cytogenetic and Genome Research 120 (1-2). Department of Biomedical Sciences, Ontario Veterinary College, University of Guelph, Guelph, ON, Canada. [email protected]: S. Karger AG, Basel: 11–25. doi:10.1159/000118737.

Bassett, Anne S, Gregory Costain, Wai Lun Alan Fung, Kathryn J Russell, Laura Pierce, Ronak Kapadia, Ronald F Carter, Eva W C Chow, and Pamela J Forsythe. 2010. “Clinically Detectable Copy Number Variations in a Canadian Catchment Population of Schizophrenia.” Journal of Psychiatric Research 44 (15): 1005–9. doi:10.1016/j.jpsychires.2010.06.013.

Bayani, Jane, and Jeremy a Squire. 2004. “Traditional Banding of Chromosomes for Cytogenetic Analysis.” Current Protocols in Cell Biology Chapter 22: Unit 22.3. doi:10.1002/0471143030.cb2203s23.

Beaulieu, A. D., J. L. Aalhus, N. H. Williams, and J. F. Patience. 2010. “Impact of Piglet Birth Weight, Birth Order, and Litter Size on Subsequent Growth Performance, Carcass Quality, Muscle Composition, and Eating Quality of Pork.” Journal of Animal Science 88 (8): 2767–78. doi:10.2527/jas.2009-2222.

Bermejo-Alvarez, P, D Rizos, D Rath, P Lonergan, and A Gutierrez-Adan. 2008. “Epigenetic Differences between Male and Female Bovine Blastocysts Produced in Vitro.” Physiological Genomics 32: 264–72. doi:10.1152/physiolgenomics.00234.2007.

Bharuthram, Avani, Maria Paximadis, Anabela C P Picton, and Caroline T. Tiemessen. 2014. “Comparison of a Quantitative Real-Time PCR Assay and Droplet Digital PCR for Copy Number Analysis of the CCL4L Genes.” Infection, Genetics and Evolution 25. Elsevier B.V.: 28–35. doi:10.1016/j.meegid.2014.03.028.

Bhushan, Sudhanshu, Hans-Christian Schuppe, Monika Fijak, and Andreas Meinhardt. 2009. “Testicular Infection: Microorganisms, Clinical Implications and Host-Pathogen Interaction.” Journal of Reproductive Immunology 83 (1-2): 164–67. doi:10.1016/j.jri.2009.07.007.

Bickhart, Derek M, and George E Liu. 2014. “The Challenges and Importance of Structural Variation Detection in Livestock.” Frontiers in Genetics 5: 37. doi:10.3389/fgene.2014.00037.

Bonneau, M, J Boscher, J J Delate, C Legault, and C P Popescu. 1991. “Two Novel Reciprocal Translocations Involving Chromosome 6 in the Domestic Pig.” Annales de Genetique 34 (2). INRA, CRJ Installations Experimentales Porcines de la Miniere, Jouy-en-Josas, France.: 65–69.

Boone, Philip M, Ian M Campbell, Brett C Baggett, Zachry T Soens, Mitchell M Rao, Patricia M Hixson, Ankita Patel, et al. 2013. “Deletions of Recessive Disease Genes: CNV Contribution to Carrier States and Disease- Causing Alleles.” Genome Research 23 (9): 1383–94. doi:10.1101/gr.156075.113.

BOUTERS, R., P. BONTE, J.SPINCEMAILLE, and M. VANDEPLASSCHE. 1974. “Les Chromosomes Des Animaux Domestiques. II-Les Anomalies Chromosomiques Associées Aux Troubles de Reproduction (en Néerlandais).” Vlaams Diergeneeskd. Tijdschr 43: 85–91. https://scholar.google.ca/scholar?hl=en&q=Les+chromosomes+des+an-

125

+imaux+domestiqucs.+II.+Les+anomalies+chromo- +somiques+associees+aux+troubles+de+reproduction&btnG=&as_sdt=1%2C5&as_sdtp=#0.

Breheny, Patrick, Prabhakar Chalise, Anthony Batzler, Liewei Wang, and Brooke L. Fridley. 2012. “Genetic Association Studies of Copy-Number Variation: Should Assignment of Copy Number States Precede Testing?” PLoS ONE 7 (4). doi:10.1371/journal.pone.0034262.

Brezezińska-Slebodzińska E, Slebodziński AB, Pietras B, Wieczorek G., E. Brzezińska-Ślebodzińska, A. B. Ślebodziński, B. Pietras, and G. Wieczorek. 1995. “Antioxidant Effect of Vitamin E and Glutathione on Lipid Peroxidation in Boar Semen Plasma.” Biol Trace Elem Res 47 (1-3): 69–74. doi:10.1007/BF02790102.

Buske, Bernd, Ina Sternstein, and Gudrun Brockmann. 2006. “QTL and Candidate Genes for Fecundity in Sows.” Animal Reproduction Science 95 (3-4). Humboldt-University of Berlin, Agricultural and Horticultural Faculty, Institute of Animal Sciences, Department of Breeding Biology and Molecular Genetics, 10115 Berlin, Invalidenstr. 42, Germany. [email protected]: 167–83. doi:10.1016/j.anireprosci.2005.12.015.

“Canadian Pork Exports.” http://www.canadapork.com/en/industry-information/canadian-pork-exports.

Carpenter, Danielle, Sugandha Dhar, Laura M Mitchell, Beiyuan Fu, Jess Tyson, Nzar A A Shwan, Fengtang Yang, Mark G Thomas, and John A L Armour. 2015. “Obesity, Starch Digestion and Amylase: Association between Copy Number Variants at Human Salivary (AMY1) and Pancreatic (AMY2) Amylase Genes.” Human Molecular Genetics 24 (12): 3472–80. doi:10.1093/hmg/ddv098.

Carter, N P. 2007. “Methods and Strategies for Analyzing Copy Number Variation Using DNA Microarrays.” Nature Genetics 39 (7 Suppl). Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge, CB10 1SA, UK. [email protected]: S16–21. doi:10.1038/ng2028.

Caspersson, T, L Zech, and C Johansson. 1970. “Differential Binding of Alkylating Fluorochromes in Human Chromosomes.” Experimental Cell Research 60 (3): 315–19. http://www.ncbi.nlm.nih.gov/pubmed/5422961.

CCSI. 1999. Canadian Centre for Swine Improvement Inc. Anuual Meeting 1999. Toronto, Ontario. https://www.ccsi.ca/main.cfm?target_page=service.

CCSI. 2004. “Detection of Hypoprolific Boars Potential Ly Carrying Chromo Somal Abnormalities.” In Genetics Committee. https://www.ccsi.ca/reports_2004/414_Hypoprolific_boars.pdf.

CCSI. 2012. “Questions and Answers About Swine EBVs.” https://www.ccsi.ca/main.cfm?target_page=qa.

CCSI. 2014. Canadian Centre for Swine Improvement Inc. 2014 Annual Meeting. Charlottetown, Prince Edward Island. https://www.ccsi.ca/meetings/annual/Annualreport_2014_EN.pdf.

Chen, C Y, Y M Guo, Z Y Zhang, J Ren, and L S Huang. 2010. “A Whole Genome Scan to Detect Quantitative Trait Loci for Gestation Length and Sow Maternal Ability Related Traits in a White Duroc × Erhualian F2 Resource Population.” Animal 4 (6): 861–66. doi:10.1017/S1751731110000169.

Chen, Congying, Ruimin Qiao, Rongxing Wei, Yuanmei Guo, Huashui Ai, Junwu Ma, Jun Ren, and Lusheng Huang. 2012. “A Comprehensive Survey of Copy Number Variation in 18 Diverse Pig Populations and Identification of Candidate Copy Number Variable Genes Associated with Complex Traits.” BMC Genomics 13 (1). BMC Genomics: 733. doi:10.1186/1471-2164-13-733.

Chen, Xiao, Aiyun Li, Wencheng Chen, Julong Wei, Jinluan Fu, and Aiguo Wang. 2014. “Differential Gene Expression in Uterine Endometrium During Implantation in Pigs.” Biology of Reproduction 92 (2): 52–52. doi:10.1095/biolreprod.114.123075.

126

Clop, a., O. Vidal, and M. Amills. 2012. “Copy Number Variation in the Genomes of Domestic Animals.” Animal Genetics 43 (5): 503–17. doi:10.1111/j.1365-2052.2012.02317.x.

Colella, Stefano, Christopher Yau, Jennifer M. Taylor, Ghazala Mirza, Helen Butler, Penny Clouston, Anne S. Bassett, Anneke Seller, Christopher C. Holmes, and Jiannis Ragoussis. 2007. “QuantiSNP: An Objective Bayes Hidden-Markov Model to Detect and Accurately Map Copy Number Variation Using SNP Genotyping Data.” Nucleic Acids Research 35 (6): 2013–25. doi:10.1093/nar/gkm076.

Cordes, Kimberly R, Neil T Sheehy, Mark P White, Emily C Berry, Sarah U Morton, Alecia N Muth, Ting-Hein Lee, Joseph M Miano, Kathryn N Ivey, and Deepak Srivastava. 2009. “miR-145 and miR-143 Regulate Smooth Muscle Cell Fate and Plasticity.” Nature 460 (7256). Nature Publishing Group: 705–10. doi:10.1038/nature08195.

D’haene, Barbara, Jo Vandesompele, and Jan Hellemans. 2010. “Accurate and Objective Copy Number Profiling Using Real-Time Quantitative PCR.” Methods (San Diego, Calif.) 50 (4): 262–70. doi:10.1016/j.ymeth.2009.12.007.

Dalin AM, Nyberg L, Eliasson L. 1988. “The Effect of Transportation/relocation on Cortisol, CBG and Induction of Puberty in Gilts with Delayed Puberty.” Acta Veterinaria Scandinavica 29 (2): 207–18.

Danielak-Czech, B, E Slota, and M Switonski. 1994. “Identification of the First Reciprocal Translocations in the Pig Population Bred in Poland.” KVL Frederiksberg (Denmark).

Danielak-Czech, B, M Switoński, and E Slota. 1997. “First Identification of Reciprocal Translocations in Polish Pigs.” Journal of Animal Breeding and Genetics = Zeitschrift Für Tierzüchtung Und Züchtungsbiologie 114 (1-6): 69–78. doi:10.1111/j.1439-0388.1997.tb00493.x.

Darvasi, a., and M. Soller. 1992. “Selective Genotyping for Determination of Linkage between a Marker Locus and a Quantitative Trait Locus.” Theoretical and Applied Genetics 85: 353–59. doi:10.1007/BF00222881.

De Vries, Bert B A, Rolph Pfundt, Martijn Leisink, David A Koolen, Lisenka E L M Vissers, Irene M Janssen, Simon van Reijmersdal, et al. 2005. “Diagnostic Genome Profiling in Mental Retardation.” American Journal of Human Genetics 77 (4): 606–16. doi:10.1086/491719.

Dechend, F, S Schubert, I Nanda, T Vogel, M Schmid, and J Schmidtke. 1998. “Organization and Expression of Rat Tspy.” Cytogenet Cell Genet 83: 270–74.

Dhami, P, A J Coffey, S Abbs, J R Vermeesch, J P Dumanski, K J Woodward, R M Andrews, C Langford, and D Vetrie. 2005. “Exon Array CGH: Detection of Copy-Number Changes at the Resolution of Individual Exons in the Human Genome.” American Journal of Human Genetics 76 (5). Human Genetics, The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, United Kingdom.: 750–62. doi:10.1086/429588.

Diskin, Sharon J., Mingyao Li, Cuiping Hou, Shuzhang Yang, Joseph Glessner, Hakon Hakonarson, Maja Bucan, John M. Maris, and Kai Wang. 2008. “Adjustment of Genomic Waves in Signal Intensities from Whole- Genome SNP Genotyping Platforms.” Nucleic Acids Research 36. doi:10.1093/nar/gkn556.

Distl, O. 2007. “Mechanisms of Regulation of Litter Size in Pigs on the Genome Level.” Reproduction in Domestic Animals = Zuchthygiene 42 Suppl 2 (September): 10–16. doi:10.1111/j.1439-0531.2007.00887.x.

Don A. Franco and Gary Pearl. 2000. “Rendering Food Waste.” In Food Waste to Animal Feed, edited by Westendorf L. Michael, 241–48.

127

Donadeu, F X, S N Schauer, and S D Sontakke. 2012. “Involvement of miRNAs in Ovarian Follicular and Luteal Development.” The Journal of Endocrinology 215 (3): 323–34. doi:10.1530/JOE-12-0252.

Du, Li-Xin, Shu-Fang Liu, Yan-Chun Yan, and Yun-Liang Jiang. 2002. “[Research on Alu Element Inserted Mutation in Porcine FSH Beta Subunit Gene].” Yi Chuan Xue Bao = Acta Genetica Sinica 29 (11): 977–82. http://www.ncbi.nlm.nih.gov/pubmed/12645260.

Ducos, A, H M Berland, N Bonnet, A Calgaro, S Billoux, N Mary, A Garnier-Bonnet, R Darre, and A Pinton. 2007. “Chromosomal Control of Pig Populations in France: 2002-2006 Survey.” Genetics, Selection, Evolution.: GSE 39 (5). UMR INRA-ENVT 444 Genetique cellulaire, Ecole nationale veterinaire de Toulouse, 23 chemin des Capelles - BP 87614, 31076 Toulouse cedex 3, France. [email protected]: 583–97. doi:10.1051/gse:2007023.

Ducos, a, a Pinton, a Séguéla, Hm Berland, Mf Blanc, a Darré, P Pinton, M Yerle, and R Darré. 1997. “A Pericentric Inversion of Chromosome 4 in Pigs.” Genetics, Selection, Evolution : GSE 29 (4): 383–94. doi:10.1186/1297- 9686-29-3-383.

Ducos, A, T Revay, A Kovacs, A Hidas, A Pinton, A Bonnet-Garnier, L Molteni, E Slota, M Switonski, and M Arruga. 2008. “Cytogenetic Screening of Livestock Populations in Europe: An Overview.” Cytogenetic and Genome Research 120 (1-2): 26–41.

Ducos, a., T. Revay, a. Kovacs, a. Hidas, a. Pinton, a. Bonnet-Garnier, L. Molteni, et al. 2008. “Cytogenetic Screening of Livestock Populations in Europe: An Overview.” Cytogenetic and Genome Research 120 (1-2). INRA-ENVT, UMR 444 Genetique Cellulaire, Toulouse, France. [email protected]: S. Karger AG, Basel: 26–41. doi:10.1159/000118738.

Ducos, Alain, Hélène-Marie Berland, Nathalie Bonnet, Anne Calgaro, Sébastien Billoux, Nicolas Mary, Amélie Garnier-Bonnet, Roland Darré, and Alain Pinton. 2007. “Chromosomal Control of Pig Populations in France: 2002–2006 Survey.” Genetics, Selection, Evolution : GSE 39 (5): 583–97. doi:10.1186/1297-9686-39-5-583.

Ducos, Alain, Alain Pinton, Hélène-marie Berland, Anne Séguéla, Corinne Brun-baronnat, Nathalie Bonnet, Roland Darré, and U M R Inra-envt De Cytogénétique. 2002. “Contrôle Chromosomique Des Populations Porcines En France : Bilan de Cinq Années D ’ Activité.” Structure, 269–75.

Dunn TG, Kaltenbach CC. 1980. “Nutrition and the Postpartum Interval of the Ewe, Sow and Cow.” Journal of Animal Science 51 (Suppl 2): 29–39.

Dutrillaux, B. 1973. “[New Techniques for Analysing the Human Karyotype].” Revue Française de Gynécologie et D’obstétrique 68 (11): 619–24. http://www.ncbi.nlm.nih.gov/pubmed/17474220.

Dutrillaux, B, and J Lejeune. 1971. “[A New Technic of Analysis of the Human Karyotype].” Comptes Rendus Hebdomadaires Des Séances de l’Académie Des Sciences. Série D: Sciences Naturelles 272 (20): 2638–40. http://www.ncbi.nlm.nih.gov/pubmed/4104656.

Einarsson, Stig, Ylva Brandt, Nils Lundeheim, and Andrzej Madej. 2008. “Stress and Its Influence on Reproduction in Pigs: A Review.” Acta Veterinaria Scandinavica 50: 48. doi:10.1186/1751-0147-50-48.

Elsik, Christine G, Ross L Tellam, Kim C Worley, Richard A Gibbs, Donna M Muzny, George M Weinstock, David L Adelson, et al. 2009. “The Genome Sequence of Taurine Cattle: A Window to Ruminant Biology and Evolution.” Science (New York, N.Y.) 324 (5926): 522–28. doi:10.1126/science.1169588.

Ernst, Catherine W, and Juan P Steibel. 2013. “Molecular Advances in QTL Discovery and Application in Pig Breeding.” Trends in Genetics : TIG 29 (4). Elsevier Ltd: 215–24. doi:10.1016/j.tig.2013.02.002.

128

Fadista, J, M Nygaard, L E Holm, B Thomsen, and C Bendixen. 2008. “A Snapshot of CNVs in the Pig Genome.” PLoS ONE 3 (12). Group of Molecular Genetics and Systems Biology, Department of Genetics and Biotechnology, Faculty of Agricultural Sciences, Aarhus University, Tjele, Denmark.: e3916. doi:10.1371/journal.pone.0003916.

FAO. 2014. “Sources of Meat.” http://www.fao.org/ag/againfo/themes/en/meat/backgr_sources.html.

Farndon, P. 2015. “Reciprocal Translocations.” Genetics. Accessed August 21. http://mymds.bham.ac.uk/genetics/reciptrans.htm.

Fernández, a I, C Barragán, a Fernández, M C Rodríguez, and B Villanueva. 2014. “Copy Number Variants in a Highly Inbred Iberian Porcine Strain.” Animal Genetics. doi:10.1111/age.12137.

Feuk, L, A R Carson, and S W Scherer. 2006. “Structural Variation in the Human Genome.” Nature reviews.Genetics 7 (2). The Centre for Applied Genomics and Program in Genetics and Genomic Biology, The Hospital for Sick Children, Department of Molecular and Medical Genetics, University of Toronto, Ontario, Canada.: 85–97. doi:10.1038/nrg1767.

Fiegler, H, J B Geigl, S Langer, D Rigler, K Porter, K Unger, N P Carter, and M R Speicher. 2007. “High Resolution Array-CGH Analysis of Single Cells.” Nucleic Acids Research 35 (3). The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SA, UK. [email protected]: e15. doi:10.1093/nar/gkl1030.

Forster, M, H Willeke, and L Richter. 1981. “Eine Autosomale, Reziproke 1/16 Translokation Bei Deutschen Landrasse Schweinen.” Reproduction in Domestic Animals 16 (2). Blackwell Synergy: 54–57.

Fowler, Katie E, Ricardo Pong-Wong, Julien Bauer, Emily J Clemente, Christopher P Reitter, Nabeel a Affara, Stephen Waite, Grant a Walling, and Darren K Griffin. 2013. “Genome Wide Analysis Reveals Single Nucleotide Polymorphisms Associated with Fatness and Putative Novel Copy Number Variants in Three Pig Breeds.” BMC Genomics 14 (1): 784. doi:10.1186/1471-2164-14-784.

Freeman, J L, G H Perry, L Feuk, R Redon, S A McCarroll, D M Altshuler, H Aburatani, et al. 2006. “Copy Number Variation: New Insights in Genome Diversity.” Genome Research 16 (8). Department of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts 02115, USA.: 949–61. doi:10.1101/gr.3677206.

Fries, R, and G Stranzinger. 1982. “Chromosomal Mutations in Pigs Derived from X-Irradiated Semen.” Cytogenetics and Cell Genetics 34 (1-2). SWITZERLAND: 55–66.

Gabriel-Robez, O., H. Jaafar, C. Ratomponirina, J. Boscher, J. Bonneau, C. P. Popescu, and Y. Rumpler. 1988. “Heterosynapsis in a Heterozygous Fertile Boar Carrier of a 3;7 Translocation.” Chromosoma 97 (1). Institut d’Embryologie, Faculte de Medecine, Strasbourg, France.: 26–32. doi:10.1007/BF00331792.

“Genetics of Auditory Disorders.” 2015. American Speech-Language-Hearing Association. Accessed August 21. http://www.asha.org/aud/Articles/Genetics-of-Auditory-Disorders/.

Gersen, Steven L., and Martha B. Keagle, eds. 2013. The Principles of Clinical Cytogenetics. New York, NY: Springer New York. doi:10.1007/978-1-4419-1688-4.

Giachini, Claudia, Francesca Nuti, Daniel J. Turner, Ilaria Laface, Yali Xue, Fabrice Daguin, Gianni Forti, Chris Tyler-Smith, and Csilla Krausz. 2009. “TSPY1 Copy Number Variation Influences Spermatogenesis and Shows Differences among Y Lineages.” Journal of Clinical Endocrinology and Metabolism 94 (10): 4016–22. doi:10.1210/jc.2009-1029.

129

Ginzinger, D G. 2002. “Gene Quantification Using Real-Time Quantitative PCR: An Emerging Technology Hits the Mainstream.” Experimental Hematology 30 (6). Genome Analysis Core Facility, Comprehensive Cancer Center, School of Medicine, University of California at San Francisco, 94143-0808, USA. [email protected]: 503–12.

GoldenHelix. “Published Articles Using SVS.” www.goldenhelix.com/SNP_Variation/published_articles.htm.

Golisch, D. 1989. “CYTOGENETIC ANALYSIS OF A BREEDING PIG-POPULATION FOR OCCURRENCE OF 13 17 ROBERTSONIAN TRANSLOCATIONS.” ARCHIV FUR TIERZUCHT- …. https://scholar.google.ca/scholar?q=Cytogenetic+analysis+of+a+breeding+pig+population+for+occurrence+of +13%3B17+Robertsonian+translocation&btnG=&hl=en&as_sdt=0%2C5#0.

Golisch, D, E Ritter, and M Schwerin. 1982. “Cytogenetic Examination of Spontaneous Aberrations in Boars.” Archiv Fuer Tierzucht 25 (6): 537–47.

Gollin, S M. 2007. “Mechanisms Leading to Nonrandom, Nonhomologous Chromosomal Translocations in Leukemia.” Seminars in Cancer Biology 17 (1). Department of Human Genetics, University of Pittsburgh Graduate School of Public Health, 130 DeSoto Street, Room A302 Crabtree Hall, Pittsburgh, PA 15261, USA. [email protected]: 74–79. doi:10.1016/j.semcancer.2006.10.002.

Gordon, Ian. 1997. Controlled Reproduction in Pigs, Volume 3. CAB International. https://books.google.ca/books/about/Controlled_Reproduction_in_Pigs.html?id=0Hd2QgAACAAJ&pgis=1.

Graphodatsky, Alexander S, Vladimir A Trifonov, and Roscoe Stanyon. 2011. “The Genome Diversity and Karyotype Evolution of Mammals.” Molecular Cytogenetics 4 (January): 22. doi:10.1186/1755-8166-4-22.

Graves, Jennifer A Marshall. 2006. “Sex Chromosome Specialization and Degeneration in Mammals.” Cell 124 (5): 901–14. doi:10.1016/j.cell.2006.02.024.

Gregory M. Cooper and Heather C. Mefford. 2011. “Detection of Copy Number Variation Using SNP Genotyping.” Methods in Molecular Biology 767: 243–52. doi:10.1007/978-1-61779-201-4.

Groenen, Martien a M, Alan L Archibald, Hirohide Uenishi, Christopher K Tuggle, Yasuhiro Takeuchi, Max F Rothschild, Claire Rogel-Gaillard, et al. 2012. “Analyses of Pig Genomes Provide Insight into Porcine Demography and Evolution.” Nature 491 (7424). Nature Publishing Group: 393–98. doi:10.1038/nature11622.

Gurgul, Artur, Ewelina Semik, Klaudia Pawlina, Tomasz Szmatoła, Igor Jasielczuk, and Monika Bugno- Poniewierska. 2014. “The Application of Genome-Wide SNP Genotyping Methods in Studies on Livestock Genomes.” Journal of Applied Genetics 55 (2): 197–208. doi:10.1007/s13353-014-0202-4.

Gustavsson, I. 1988a. “Committee for the Standardized Karyotype of the Domestic Pig.” Hereditas 109 (2). Blackwell Synergy: 151–57.

———. 1988b. “Reciprocal Translocation in Four Boars Producing Decreased Litter Size.” Hereditas 109 (2). SWEDEN: 159–68.

Gustavsson, I, L Jönsson, and L Jonsson. 1992. “Stillborns, Partially Monosomic and Partially Trisomic, in the Offspring of a Boar Carrying a Translocation: rcp(14;15)(q29;q24).” Hereditas 117 (1). Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, Uppsala.: 31–37.

Gustavsson, I, and I Settergren. 1984. “Reciprocal Chromosome Translocation with Transfer of Centromeric Heterochromatin in the Domestic Pig Karyotype.” Hereditas 100 (1). SWEDEN: 1–5.

130

Gustavsson, I, I Settergren, and W A King. 1983. “Occurrence of Two Different Reciprocal Translocations in the Same Litter of Domestic Pigs.” Hereditas 99 (2). SWEDEN: 257–67.

Gustavsson, I, M Switonski, K Larsson, L Ploen, K Hojer, M Switoński, K Larsson, L Plöen, and K Höjer. 1988. “Chromosome Banding Studies and Synaptonemal Complex Analyses of Four Reciprocal Translocations in the Domestic Pig.” Hereditas 109 (2). SWEDEN: 169–84.

Gustavsson, I. 1990. “Chromosomes of the Pig.” In Domestic Animal Cytogenetics, edited by Richard A McFeely, 34:73. Academic Press, Inc.: Adv. Vet. Sci. Comp. Med. http://www.cabdirect.org/abstracts/19910187653.html;jsessionid=BD4ADB804B100C13BA13DAC02116D1 81.

Hageltorn, M, I Gustavsson, and L Zech. 1973. “The Q- and G-Banding Patterns of a t(11p+; 15q-) in the Domestic Pig.” Hereditas 75 (1). SWEDEN: 147–51.

———. 1976. “Detailed Analysis of a Reciprocal Translocation (13q-; 14q+) in the Domestic Pig by G- and Q- Staining Techniques.” Hereditas 83 (2). SWEDEN: 268–72.

Halper-Stromberg, Eitan, Laurence Frelin, Ingo Ruczinski, Robert Scharpf, Chunfa Jie, Benilton Carvalho, Haiping Hao, et al. 2011. “Performance Assessment of Copy Number Microarray Platforms Using a Spike-in Experiment.” Bioinformatics 27 (8): 1052–60. doi:10.1093/bioinformatics/btr106.

Hamilton, C. K., a. R. Verduzco-Gmez, L. a. Favetta, P. Blondin, and W. a. King. 2012. “Testis-Specific Protein Y- Encoded Copy Number Is Correlated to Its Expression and the Field Fertility of Canadian Holstein Bulls.” Sexual Development 6 (5): 231–39. doi:10.1159/000338938.

Hamilton, C.K., L.a. Favetta, G.P. Di Meo, S. Floriot, a. Perucatti, J. Peippo, J. Kantanen, a. Eggen, L. Iannuzzi, and W.a. King. 2009b. “Copy Number Variation of Testis-Specific Protein, Y-Encoded (TSPY) in 14 Different Breeds of Cattle (Bos Taurus).” Sexual Development 3 (4): 205–13. doi:10.1159/000228721.

Hao, Ke, Xiaobin Wang, Tianhua Niu, Xin Xu, Ang Li, Weili Chang, Lin Wang, Guang Li, Nan Laird, and Xiping Xu. 2004. “A Candidate Gene Association Study on Preterm Delivery: Application of High-Throughput Genotyping Technology and Advanced Statistical Methods.” Human Molecular Genetics 13 (7): 683–91. doi:10.1093/hmg/ddh091.

Haraksingh, Rajini R, and Michael P Snyder. 2013. “Impacts of Variation in the Human Genome on Gene Regulation.” Journal of Molecular Biology 425 (21). The Authors: 3970–77. doi:10.1016/j.jmb.2013.07.015.

Harvey, A. 1967. “A MALE PIG WITH AN XXY / XXXY SEX CHROMOSOME Department of Veterinary Reproduction , University of Glasgow Veterinary Hospital ,.”

Hassold, T J, and P A Jacobs. 1984. “Trisomy in Man.” Annual Review of Genetics 18 (January). Annual Reviews 4139 El Camino Way, P.O. Box 10139, Palo Alto, CA 94303-0139, USA: 69–97. doi:10.1146/annurev.ge.18.120184.000441.

Henricson, B, L Backstrom, and L Bäckström. 1964. “Translocation Heterozygosity in a Boar.” Hereditas 52 (166.170): 166–70.

Hernandez, S C, H a Finlayson, C J Ashworth, C S Haley, and A L Archibald. 2014. “A Genome-Wide Linkage Analysis for Reproductive Traits in F2 Large White × Meishan Cross Gilts.” Animal Genetics 45 (2): 191–97. doi:10.1111/age.12123.

131

Herrick, John B., and H. L. Self. 1962. Evaluation of Fertility in the Bull and Boar. Iowa State University Press. https://books.google.com/books?id=ASJBAAAAYAAJ&pgis=1.

Hindson, Benjamin J, Kevin D Ness, Donald a Masquelier, Phillip Belgrader, Nicholas J Heredia, Anthony J Makarewicz, Isaac J Bright, et al. 2011. “High-Throughput Droplet Digital PCR System for Absolute Quantitation of DNA Copy Number.” Analytical Chemistry 83 (22): 8604–10. doi:10.1021/ac202028g.

Hu, Zhi-Liang, Carissa a Park, Xiao-Lin Wu, and James M Reecy. 2013. “Animal QTLdb: An Improved Database Tool for Livestock Animal QTL/association Data Dissemination in the Post-Genome Era.” Nucleic Acids Research 41 (Database issue): D871–79. doi:10.1093/nar/gks1150.

Huang, J.-T., Y.-J. Liu, J. Wang, Z.-G. Xu, Y. Yang, F. Shen, X.-h. Liu, X. Zhou, and S.-M. Liu. 2015. “Next Generation Digital PCR Measurement of Hepatitis B Virus Copy Number in Formalin-Fixed Paraffin- Embedded Hepatocellular Carcinoma Tissue.” Clinical Chemistry 61 (1): 290–96. doi:10.1373/clinchem.2014.230227.

Huang, Jinming, Zhihua Ju, Qiuling Li, Qinlei Hou, Changfa Wang, Jianbin Li, Rongling Li, et al. 2011. “Solexa Sequencing of Novel and Differentially Expressed microRNAs in Testicular and Ovarian Tissues in Holstein Cattle.” International Journal of Biological Sciences 7 (7): 1016–26.

Huggett, Jim F, Carole a Foy, Vladimir Benes, Kerry Emslie, Jeremy a Garson, Ross Haynes, Jan Hellemans, et al. 2013. “The Digital MIQE Guidelines: Minimum Information for Publication of Quantitative Digital PCR Experiments.” Clinical Chemistry 59 (6): 892–902. doi:10.1373/clinchem.2013.206375.

Huggett, Jim F, and Alexandra Whale. 2013. “Digital PCR as a Novel Technology and Its Potential Implications for Molecular Diagnostics.” Clinical Chemistry 59 (12): 1691–93. doi:10.1373/clinchem.2013.214742.

Hughes, J, H Skaletsky, T Pyntikova, T Graves, S Van Daalen, P Minx, R Fulton, et al. 2010. “Chimpanzee and Human Y Chromosomes Are Remarkably Divergent in Structure and Gene Content.” Nature 463 (January): 536–39.

Hughes, Jennifer F, Helen Skaletsky, Laura G Brown, Tatyana Pyntikova, Tina Graves, Robert S Fulton, Shannon Dugan, et al. 2012. “Strict Evolutionary Conservation Followed Rapid Gene Loss on Human and Rhesus Y Chromosomes.” Nature 483 (7387): 82–86. doi:10.1038/nature10843.

Hurles, Matthew. 2004. “Gene Duplication: The Genomic Trade in Spare Parts.” PLoS Biology 2 (7): E206. doi:10.1371/journal.pbio.0020206.

Iafrate, A John, Lars Feuk, Miguel N Rivera, Marc L Listewnik, Patricia K Donahoe, Ying Qi, Stephen W Scherer, and Charles Lee. 2004. “Detection of Large-Scale Variation in the Human Genome.” Nature Genetics 36: 949–51. doi:10.1038/ng1416.

Iannuzzi L, King WA, Di Berardino D. 2009. “Chromosome Evolution in Domestic Bovids as Revealed by Chromosome Banding and FISH-Mapping Techniques.” Cytogenetic and Genome Research 126 (1-2): 49–62. [email protected]. 2015. “Fertility Measures | Department of Agriculture and Rural Development.” [email protected]. Accessed August 10. http://www.dardni.gov.uk/index/farming/livestock/pigs/pigs- husbandry/fertility-measures.htm.

Ishkanian, A S, C A Malloff, S K Watson, R J DeLeeuw, B Chi, B P Coe, A Snijders, et al. 2004. “A Tiling Resolution DNA Microarray with Complete Coverage of the Human Genome.” Nature Genetics 36 (3). British Columbia Cancer Research Centre, 601 West 10th Avenue, Vancouver, British Columbia V5Z 1L3, Canada.: 299–303. doi:10.1038/ng1307.

132

Iwarsson, E, L Ahrlund-Richter, J Inzunza, B Rosenlund, M Fridström, T Hillensjö, P Sjöblom, M Nordenskjöld, and E Blennow. 1998. “Preimplantation Genetic Diagnosis of a Large Pericentric Inversion of Chromosome 5.” Molecular Human Reproduction 4 (7): 719–23. http://www.ncbi.nlm.nih.gov/pubmed/9701795.

Jaafar, H, A Bouvet, O Gabriel-Robez, M Bonneau, J Boscher, J L Maetz, C P Popescu, and Y Rumpler. 1992. “Heterosynapsis in Two Fertile but Hypoprolific Boars Carriers of Reciprocal Translocations.” Annales de Genetique 35 (1). Institut d’Embryologie, Faculte de Medecine, Strasbourg, France.: 3–7.

Jacobs, P a, C Browne, N Gregson, C Joyce, and H White. 1992. “Estimates of the Frequency of Chromosome Abnormalities Detectable in Unselected Newborns Using Moderate Levels of Banding.” Journal of Medical Genetics 29 (2). Wessex Regional Genetics Laboratory, Salisbury General Infirmary, Wiltshire.: 103–8. doi:10.1136/jmg.29.2.103.

Janssen, U, E M Davis, M M Le Beau, and W Stoffel. 1997. “Human Mitochondrial Enoyl-CoA Hydratase Gene (ECHS1): Structural Organization and Assignment to Chromosome 10q26.2-q26.3.” Genomics 40: 470–75.

Jiang, Jicai, Jiying Wang, Haifei Wang, Yan Zhang, Huimin Kang, Xiaotian Feng, Jiafu Wang, et al. 2014. “Global Copy Number Analyses by next Generation Sequencing Provide Insight into Pig Genome Variation.” BMC Genomics 15 (1): 593. doi:10.1186/1471-2164-15-593.

Jiugang, Zhao, Lan Jing, and Liu Yonggang. 2012. “Characterization, Expression Profile, Polymorphism and Association of Porcine NAT9 Gene.” Molecular Biology Reports 39 (3): 3137–42. doi:10.1007/s11033-011- 1079-5.

Jobling, M. 2008. “Copy Number Variation on the Human Y Chromosome.” Cytogenetic and Genome Research 123 (1-4): 253–62.

Kannaki, T. R., M. Shanmugam, and P. C. Verma. 2011. “Toll-like Receptors and Their Role in Animal Reproduction.” Animal Reproduction Science. doi:10.1016/j.anireprosci.2011.03.008.

Karolchik, Donna, Galt P. Barber, Jonathan Casper, Hiram Clawson, Melissa S. Cline, Mark Diekhans, Timothy R. Dreszer, et al. 2014. “The UCSC Genome Browser Database: 2014 Update.” Nucleic Acids Research 42. doi:10.1093/nar/gkt1168.

Kido, Tatsuo, Regina Cheuk-Lam Lo, Yunmin Li, Joyce Lee, Z Laura Tabatabai, Irene Oi-lin Ng, and Yun-Fai Chris Lau. 2014. “The Potential Contributions of a Y-Located Protooncogene and Its X Homologue in Sexual Dimorphisms in Hepatocellular Carcinoma.” Human Pathology 45 (9): 1847–58. doi:10.1016/j.humpath.2014.05.002.

Kienesberger, Petra C, Monika Oberer, Achim Lass, and Rudolf Zechner. 2009. “Mammalian Patatin Domain Containing Proteins: A Family with Diverse Lipolytic Activities Involved in Multiple Biological Functions.” Journal of Lipid Research 50 Suppl (April): S63–68. doi:10.1194/jlr.R800082-JLR200.

King G. 2002. “Reproductive Management of Pigs Points to Consider: The Importance of Reproductive Performance.” Animal Science 1 (8).

King, W a. 1981. “Meiotic Behaviour of a Rcp (13q-; 14q+) Translocation in Heterozygous Pigs.” Hereditas 94: 235–40.

Kociucka, B, I Szczerbal, S Bugaj, M Orsztynowicz, and M Switonski. 2014. “A High Incidence of Adjacent-1 Meiotic Segregation Pattern, Revealed by Multicolor Sperm FISH, in a Carrier Boar of a New Reciprocal Translocation t(6;16)(p13;q23).” Cytogenetic and Genome Research 142 (1): 21–27. doi:10.1159/000355213.

133

Kotaja, Noora, Suvendra N Bhattacharyya, Lukasz Jaskiewicz, Sarah Kimmins, Martti Parvinen, Witold Filipowicz, and Paolo Sassone-Corsi. 2006. “The Chromatoid Body of Male Germ Cells: Similarity with Processing Bodies and Presence of Dicer and microRNA Pathway Components.” Proceedings of the National Academy of Sciences of the United States of America 103: 2647–52. doi:10.1073/pnas.0509333103.

Krausz, Csilla, Claudia Giachini, and Gianni Forti. 2010. “TSPY and Male Fertility.” Genes 1 (2): 308–16. doi:10.3390/genes1020308.

KUOKKANEN, MARJA-TERTTU, and AULI MÄKINEN. 1988. “Reciprocal Chromosome Translocations, (1p-; 11q+) and (1p+; 15q-), in Domestic Pigs with Reduced Litter Size.” Hereditas 109 (1): 69–73. doi:10.1111/j.1601-5223.1988.tb00184.x.

Lau, Yun Fai Chris, Yunmin Li, and Tatsuo Kido. 2009. “Gonadoblastoma Locus and the TSPY Gene on the Human Y Chromosome.” Birth Defects Research Part C - Embryo Today: Reviews. doi:10.1002/bdrc.20144.

Lear, T L, and E Bailey. 2008. “Equine Clinical Cytogenetics: The Past and Future.” Cytogenetic and Genome Research 120 (1-2): 42–49. doi:10.1159/000118739.

Lee, Hsin-i, Hsiang-an Ho, and Chen-hung Kao. 2014. “A New Simple Method for Improving QTL Mapping under Selective Genotyping.” Genetics 198 (December): 1685–98. doi:10.1534/genetics.114.168385.

Lee, J H, and J T Jeon. 2008. “Methods to Detect and Analyze Copy Number Variations at the Genome-Wide and Locus-Specific Levels.” Cytogenetic and Genome Research 123 (1-4). Division of Animal Science and Resources, College of Agriculture and Life Sciences, Chungnam National University, Daejeon, Korea. [email protected]: S. Karger AG, Basel: 333–42. doi:10.1159/000184725.

Lee, Kwangwon, Janos Kerner, and Charles L Hoppel. 2011. “Mitochondrial Carnitine Palmitoyltransferase 1a (CPT1a) Is Part of an Outer Membrane Fatty Acid Transfer Complex.” The Journal of Biological Chemistry 286: 25655–62. doi:10.1074/jbc.M111.228692.

Li, Gang, Brian W Davis, Terje Raudsepp, Alison J Pearks Wilkerson, Victor C Mason, Malcolm Ferguson-Smith, Patricia C O’Brien, Paul D Waters, and William J Murphy. 2013. “Comparative Analysis of Mammalian Y Chromosomes Illuminates Ancestral Structure and Lineage-Specific Evolution.” Genome Research 23 (9): 1486–95. doi:10.1101/gr.154286.112.

Li, Yan, Shuqi Mei, Xuying Zhang, Xianwen Peng, Gang Liu, Hu Tao, Huayu Wu, Siwen Jiang, Yuanzhu Xiong, and Fenge Li. 2012. “Identification of Genome-Wide Copy Number Variations among Diverse Pig Breeds by Array CGH.” BMC Genomics 13 (1): 725. doi:10.1186/1471-2164-13-725.

Li, Yan, Hai-Yan Wang, Feng-Chun Wan, Fu-Jun Liu, Juan Liu, Na Zhang, Shao-Hua Jin, and Jian-Yuan Li. 2012. “Deep Sequencing Analysis of Small Non-Coding RNAs Reveals the Diversity of microRNAs and piRNAs in the Human Epididymis.” Gene 497 (2). Elsevier B.V.: 330–35. doi:10.1016/j.gene.2012.01.038.

Lin, Yen-Lung, Pavlos Pavlidis, Emre Karakoc, Jerry Ajay, and Omer Gokcumen. 2015. “The Evolution and Functional Impact of Human Deletion Variants Shared with Archaic Hominin Genomes.” Molecular Biology and Evolution 32 (4): 1008–19. doi:10.1093/molbev/msu405.

Liu, Chong, Wenyang Zhang, Deming Yang, and Yonggang Liu. 2015. “Molecular Characterization, Polymorphism, and Association of Porcine GADD45G Gene.” Animal Biotechnology 26 (3): 230–36. doi:10.1080/10495398.2015.1005216.

134

Liu, George E, Yali Hou, Bin Zhu, Maria Francesca Cardone, Lu Jiang, Angelo Cellamare, Apratim Mitra, et al. 2010. “Analysis of Copy Number Variations among Diverse Cattle Breeds Analysis of Copy Number Variations among Diverse Cattle Breeds,” 693–703. doi:10.1101/gr.105403.110.

Locniskar, F, I Gustavsson, M Hageltorn, and L Zech. 1976. “Cytological Origin and Points of Exchange of a Reciprocal Chromosome Translocation (1p-; 6q+) in the Domestic Pig.” Hereditas 83 (2). SWEDEN: 273–75.

Louis, G F, a J Lewis, W C Weldon, P S Miller, R J Kittok, and W W Stroup. 1994. “The Effect of Protein Intake on Boar Libido, Semen Characteristics, and Plasma Hormone Concentrations.” J Anim Sci 72 (8): 2038–50. 7982833.

MacDonald, Jeffrey R., Robert Ziman, Ryan K C Yuen, Lars Feuk, and Stephen W. Scherer. 2014. “The Database of Genomic Variants: A Curated Collection of Structural Variation in the Human Genome.” Nucleic Acids Research 42 (D1): 986–92. doi:10.1093/nar/gkt958.

Madan, Kamlesh, C. E. Ford, and C. Polge. 1978. “A Reciprocal Translocation, T (6p+; 14q−), in the Pig.” Journal of Reproduction and Fertility 53: 395–98.

Makinen, A, M T Kuokkanen, T Niini, and L Perttola. 1987. “A Complex Three Breakpoint Translocation in the Domestic Pig.” Acta Veterinaria Scandinavica 28 (2). DENMARK: 189–96.

Mäkinen, A, T Pitkänen, and M Andersson. 1997. “Two Cases of Reciprocal Translocations in Domestic Pigs Producing Small Litters.” Journal of Animal Breeding and Genetics = Zeitschrift Fur Tierzuchtung Und Zuchtungsbiologie 114 (1-6): 377–84. doi:10.1111/j.1439-0388.1997.tb00523.x.

MÄKINEN, AULI, and EERO REMES. 1986. “Low Fertility in Pigs with rcp(4q+;13q-) Translocation.” Hereditas 104 (February): 223–29. doi:10.1111/j.1601-5223.1986.tb00536.x.

Manoj, P. 2014. “Droplet Digital PCR Technology Promises New Applications and Research Areas.” Mitochondrial DNA 1736 (Table 1): 1–5. doi:10.3109/19401736.2014.913168.

Mantripragada, K K, I Tapia-Paez, E Blennow, P Nilsson, A Wedell, and J P Dumanski. 2004. “DNA Copy-Number Analysis of the 22q11 Deletion-Syndrome Region Using Array-CGH with Genomic and PCR-Based Targets.” International Journal of Molecular Medicine 13 (2). Department of Genetics and Pathology, Rudbeck Laboratory, Uppsala University Hospital, SE-751 85 Uppsala, Sweden.: 273–79.

Marcinkowska, Malgorzata, Maciej Szymanski, Wlodzimierz J Krzyzosiak, and Piotr Kozlowski. 2011. “Copy Number Variation of microRNA Genes in the Human Genome.” BMC Genomics 12 (1). BioMed Central Ltd: 183. doi:10.1186/1471-2164-12-183.

Marques, Francine Z, Priscilla R Prestes, Leonardo B Pinheiro, Katrina Scurrah, Kerry R Emslie, Maciej Tomaszewski, Stephen B Harrap, and Fadi J Charchar. 2014. “Measurement of Absolute Copy Number Variation Reveals Association with Essential Hypertension.” BMC Medical Genomics 7 (1): 44. doi:10.1186/1755-8794-7-44.

Massip, K., H. Berland, N. Bonnet, a. Calgaro, S. Billoux, V. Baquié, N. Mary, et al. 2008. “Study of Inter- and Intra-Individual Variation of Meiotic Segregation Patterns in t(3;15)(q27;q13) Boars.” Theriogenology 70 (4). UMR 444 INRA-ENVT Genetique Cellulaire, Ecole Nationale Veterinaire de Toulouse, BP 87614, 31076 Toulouse Cedex 3, France.: 655–61. doi:10.1016/j.theriogenology.2008.04.026.

Massip, K., N. Bonnet, a. Calgaro, S. Billoux, V Baquie, N. Mary, a. Bonnet-Garnier, et al. 2009. “Male Meiotic Segregation Analyses of Peri- and Paracentric Inversions in the Pig Species.” Cytogenetic and Genome

135

Research 125 (2). UMR INRA-ENVT Genetique Cellulaire, Ecole Nationale Veterinaire de Toulouse, Toulouse, France.: S. Karger AG, Basel: 117–24. doi:10.1159/000227836.

Massip, Katia, Martine Yerle, Yvon Billon, Stéphane Ferchaud, Nathalie Bonnet, Anne Calgaro, Nicolas Mary, et al. 2010. “Studies of Male and Female Meiosis in inv(4)(p1.4;q2.3) Pig Carriers.” Chromosome Research : An International Journal on the Molecular, Supramolecular and Evolutionary Aspects of Chromosome Biology 18 (8). Ecole Nationale Veterinaire de Toulouse, UMR 444 INRA-ENVT Genetique Cellulaire, 23 chemin des Capelles, BP 87614, 31076 Toulouse Cedex 3, France.: 925–38. doi:10.1007/s10577-010-9162-7.

McBride, D., W. Carre, S. D. Sontakke, C. O. Hogg, A. Law, F. X. Donadeu, and M. Clinton. 2012. “Identification of miRNAs Associated with the Follicular-Luteal Transition in the Ruminant Ovary.” Reproduction. doi:10.1530/REP-12-0025.

McFeely, RA, LR Klunder, and JB Goldman. 1988. “A Robertsonian Translocation in a Sow with Reduced Litter Size.” Proc 8th Eur Colloq Cytogenet Domest Anim. https://scholar.google.ca/scholar?q=A+Robertsonian+translocation+in+a+sow+with+reduced+litter+size.&btn G=&hl=en&as_sdt=0%2C5#0.

McKeegan, Paul J., and Roger G. Sturmey. 2012. “The Role of Fatty Acids in Oocyte and Early Embryo Development.” Reproduction, Fertility and Development. doi:10.1071/RD11907.

McLean-Tooke, A, G P Spickett, and A R Gennery. 2007. “Immunodeficiency and Autoimmunity in 22q11.2 Deletion Syndrome.” Scandinavian Journal of Immunology 66 (1): 1–7. doi:10.1111/j.1365- 3083.2007.01949.x.

Menasha, Joshua, Brynn Levy, Kurt Hirschhorn, and Nataline B Kardon. 2005. “Incidence and Spectrum of Chromosome Abnormalities in Spontaneous Abortions: New Insights from a 12-Year Study.” Genetics in Medicine : Official Journal of the American College of Medical Genetics 7 (4): 251–63. doi:10.1097/01.GIM.0000160075.96707.04.

Meuwissen, T H, B J Hayes, and M E Goddard. 2001. “Prediction of Total Genetic Value Using Genome-Wide Dense Marker Maps.” Genetics 157 (4): 1819–29. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1461589&tool=pmcentrez&rendertype=abstract.

Meylan, Etienne, Jürg Tschopp, and Michael Karin. 2006. “Intracellular Pattern Recognition Receptors in the Host Response.” Nature 442: 39–44. doi:10.1038/nature04946.

Mikelsaar, A V, and H G Schwarzacher. 1978. “Comparison of Silver Staining of Nucleolus Organizer Regions in Human Lymphocytes and Fibroblasts.” Human Genetics 42 (3): 291–99. http://www.ncbi.nlm.nih.gov/pubmed/352909.

Miller, S. A., D. D. Dykes, and H. F. Polesky. 1988. “A Simple Salting out Procedure for Extracting DNA from Human Nucleated Cells.” Nucleic Acids Research 16 (3): 1215. doi:10.1093/nar/16.3.1215.

Miyake, Y I, T Matsubara, M Hata, Y Kaneda, and M. Hata and Y. Kaneda Y-I. Miyake, T. Matsubara. 1994. “Chromosomal Pericentric Inversion Detected in a Sow and Her Piglets.” Theriogenology 42 (2). Laboratory of Theriogenology, Faculty of Agriculture Iwate University, Morioka 020, Japan.: 241–46.

Morel, F, B Laudier, F Guérif, M L Couet, D Royère, C Roux, J L Bresson, V Amice, M De Braekeleer, and N Douet-Guilbert. 2007. “Meiotic Segregation Analysis in Spermatozoa of Pericentric Inversion Carriers Using Fluorescence in-Situ Hybridization.” Human Reproduction (Oxford, England) 22 (1): 136–41. doi:10.1093/humrep/del317.

136

Mukherjee, Ayan, Gulshan Dass, Jagan Mohanarao G, Vinay Kumar Katneni, Dipak Banerjee, Tapan Kumar Das, Moloya Gohain, a K Chakrabarty, Tirtha Kumar Datta, and Sachinandan De. 2015. “Copy Number Differences of Y Chromosomal Genes Between Superior and Inferior Quality Semen Producing Crossbred (Bos Taurus × Bos Indicus) Bulls.” Animal Biotechnology 26 (1): 65–72. doi:10.1080/10495398.2014.887020.

Neal, Michael S., Ed R. Reyes, Ken S. Fisher, W. Allan King, and Parvathi K. Basrur. 1998. “Reproductive Consequences of an X-Autosome Translocation in a Swine Herd.” The Canadian Veterinary journal.La Revue Veterinaire Canadienne 39 (4). Department of Biomedical Sciences, Ontario Veterinary College, University of Guelph, Ontario.: 232–37.

Nickkholgh, Bita, Michiel J. Noordam, Suzanne E. Hovingh, a. M M Van Pelt, Fulco Van Der Veen, and Sjoerd Repping. 2010. “Y Chromosome TSPY Copy Numbers and Semen Quality.” Fertility and Sterility 94 (5). Elsevier Ltd: 1744–47. doi:10.1016/j.fertnstert.2009.09.051.

Niu, Zhiyv, Shaun M Goodyear, Shilpa Rao, Xin Wu, John W Tobias, Mary R Avarbock, and Ralph L Brinster. 2011. “MicroRNA-21 Regulates the Self-Renewal of Mouse Spermatogonial Stem Cells.” Proceedings of the National Academy of Sciences of the United States of America 108: 12740–45. doi:10.1073/pnas.1109987108.

Ogg, Sherry L, Anne K Weldon, Lorraine Dobbie, Andrew J H Smith, and Ian H Mather. 2004. “Expression of Butyrophilin (Btn1a1) in Lactating Mammary Gland Is Essential for the Regulated Secretion of Milk-Lipid Droplets.” Proceedings of the National Academy of Sciences of the United States of America 101 (27): 10084– 89. doi:10.1073/pnas.0402930101.

Ohlsson Teague, E Maria C, Cristin G Print, and M Louise Hull. 2010. “The Role of microRNAs in Endometriosis and Associated Reproductive Conditions.” Human Reproduction Update 16 (2): 142–65. doi:10.1093/humupd/dmp034.

Oostlander, A E, G A Meijer, and B Ylstra. 2004. “Microarray-Based Comparative Genomic Hybridization and Its Applications in Human Genetics.” Clinical Genetics 66 (6). Microarray Core Facility, Vu University Medical Center, Amsterdam, The Netherlands.: 488–95. doi:10.1111/j.1399-0004.2004.00322.x.

Paria, N, T Raudsepp, A Pearks Wilkerson, P O’brien, M Ferguson-Smith, C Love, C Arnold, P Rakestraw, W Murphy, and B Chowdhary. 2011. “A Gene Catalogue of the Euchromatic Male-Specific Region of the Horse Y Chromosome: Comparison with Human and Other Mammals.” PLoS ONE 6 (7): e21374.

Paudel, Yogesh, Ole Madsen, Hendrik-Jan Megens, Laurent a F Frantz, Mirte Bosse, John W M Bastiaansen, Richard P M a Crooijmans, and Martien a M Groenen. 2013. “Evolutionary Dynamics of Copy Number Variation in Pig Genomes in the Context of Adaptation and Domestication.” BMC Genomics 14: 449. doi:10.1186/1471-2164-14-449.

Paudel, Yogesh, Ole Madsen, Hendrik-Jan Megens, Laurent a F Frantz, Mirte Bosse, Richard P M a Crooijmans, and Martien a M Groenen. 2015. “Copy Number Variation in the Speciation of Pigs: A Possible Prominent Role for Olfactory Receptors.” BMC Genomics 16 (1). ??? 1–14. doi:10.1186/s12864-015-1449-9.

Peiffer, Daniel A, Jennie M Le, Frank J Steemers, Weihua Chang, Tony Jenniges, Francisco Garcia, Kirt Haden, et al. 2006. “High-Resolution Genomic Profiling of Chromosomal Aberrations Using Infinium Whole-Genome Genotyping.” Genome Research 16 (9): 1136–48. doi:10.1101/gr.5402306.

Perucatti, A, A Iannuzzi, V Genualdo, D Incarnato, Plenary Session, Madingley Road, and Biomedical Sciences. 2014. “21st International Colloquium on Animal Cytogenetics and Gene Mapping.” Chromosome Research 22 (3): 393–437. doi:10.1007/s10577-014-9435-7.

137

Pfaffl, M W. 2001. “A New Mathematical Model for Relative Quantification in Real-Time RT-PCR.” Nucleic Acids Research 29 (9). Institute of Physiology, FML-Weihenstephan, Center of Life and Food Sciences, Technical University of Munich, Germany. [email protected]: e45.

Pinheiro, Leonardo B., Victoria A. Coleman, Christopher M. Hindson, Jan Herrmann, Benjamin J. Hindson, Somanath Bhat, and Kerry R. Emslie. 2012. “Evaluation of a Droplet Digital Polymerase Chain Reaction Format for DNA Copy Number Quantification.” Analytical Chemistry 84: 1003–11. doi:10.1021/ac202578x.

Pinkel, D, R Segraves, D Sudar, S Clark, I Poole, D Kowbel, C Collins, et al. 1998. “High Resolution Analysis of DNA Copy Number Variation Using Comparative Genomic Hybridization to Microarrays.” Nature Genetics 20 (2). Cancer Genetics Program, UCSF Cancer Center, University of California San Francisco, 94143-0808, USA. [email protected]: 207–11. doi:10.1038/2524.

Pinton, a, a Ducos, H Berland, a Seguela, C Brun-Baronnat, A Darre, R Darre, et al. 2000. “Chromosomal Abnormalities in Hypoprolific Boars.” Hereditas 132 (1). URA INRA-ENVT de Cytogenetique des Populations Animales, Toulouse, France.: 55–62.

Pinton, a., a. Calgaro, N. Bonnet, S. Ferchaud, S. Billoux, a. M. Dudez, N. Mary, et al. 2009. “Influence of Sex on the Meiotic Segregation of a t(13;17) Robertsonian Translocation: A Case Study in the Pig.” Human Reproduction 24 (8): 2034–43. doi:10.1093/humrep/dep118.

Pinton, a., T. Faraut, M. Yerle, J. Gruand, F. Pellestor, and a. Ducos. 2005. “Comparison of Male and Female Meiotic Segregation Patterns in Translocation Heterozygotes: A Case Study in an Animal Model (Sus Scrofa Domestica L.).” Human Reproduction (Oxford, England) 20 (9). UMR INRA-ENVT Cytogenetique des Populations Animales, Toulouse, France. [email protected]: 2476–82. doi:10.1093/humrep/dei067.

Pinton, Alain, Alain Ducos, and Martine Yerle. 2004. “Estimation of the Proportion of Genetically Unbalanced Spermatozoa in the Semen of Boars Carrying Chromosomal Rearrangements Using FISH on Sperm Nuclei.” Genetics, Selection, Evolution.: GSE 36 (1). UMR INRA-ENVT Cytogenetique des populations animales, Ecole nationale veterinaire de Toulouse, 23, chemin des Capelles, 31076 Toulouse 3, France.: 123–37. doi:10.1051/gse:2003055.

Pique-Regi, Roger, Alejandro Cáceres, and Juan R González. 2010. “R-Gada: A Fast and Flexible Pipeline for Copy Number Analysis in Association Studies.” BMC Bioinformatics 11: 380. doi:10.1186/1471-2105-11-380.

Popescu, C P. 1990. “Consequences of Abnormalities of Chromosome Structure in Domestic Animals.” Reproduction, Nutrition, Development Suppl 1. INRA, CRJ, Laboratoire de Cytogenetique, Jouy-en-Josas, France.: 105s – 116s.

Popescu, C P, M Bonneau, M Tixier, I Bahri, and J Boscher. 1984. “Reciprocal Translocations in Pigs: Their Detection and Consequences on Animal Performance and Economic Losses.” The Journal of Heredity 75 (experiment 1): 448–52. http://www.ncbi.nlm.nih.gov/pubmed/6542578.

Popescu, C P, and J Boscher. 1982. “Cytogenetics of Preimplantation Embryos Produced by Pigs Heterozygous for the Reciprocal Translocation (4q+;14q-).” Cytogenetics and Cell Genetics 34 (1-2): 119–23. http://www.ncbi.nlm.nih.gov/pubmed/7151483.

POPESCU, C P, J BOSCHER, and M TIXIER. 1983. “Une Nouvelle Translocation Réciproque T, rcp(7q−; 15q−) Chez Un Verrat «hypoprolifique».” Génétique, Sélection, Évolution 15 (4): 479–88.

Popescu, C P, J Boscher, and S Q Zhang. 1988. “Cytogenetic Evaluation of Boars with a Low Prolificacy: Two New Types of Chromosome Translocation.” Annales de Genetique 31 (2). I.N.R.A. C.R.J. Laboratory of Cytogenetics, Jouy-en-Josas.: 75–80.

138

Popescu, C P, and C Legault. 1979. “Une Nouvelle Translocation Reciproque t(4q; 14q-) Chez Le Porc Domestique (Sus Scrofa Domesticus).” Ann.Genet.Sel.Anim. 11: 361–69.

Popescu, C P, and M Tixier. 1984. “The Frequency of Chromosome Abnormalities in Farm Animals and Their Economic Consequences.” Annales de Genetique 27 (2). FRANCE: 69–72.

Pozzi, PS, and GL Alborali. 2012. “Reproductive Diseases in Sows (Sus Scrofa Domestica): A Review.” Israel Journal of Veterinary Medicine. http://www.vetserveng.moag.gov.il/NR/rdonlyres/CEF47B91-1D76-410B- B958-AA629CC75810/0/2012Reproductivediseasesinsows.pdf.

Pretto, D., D. Maar, C. M. Yrigollen, J. Regan, and F. Tassone. 2015. “Screening Newborn Blood Spots for 22q11.2 Deletion Syndrome Using Multiplex Droplet Digital PCR.” Clinical Chemistry 61 (1): 182–90. doi:10.1373/clinchem.2014.230086.

Pripwai, Nakarin, and Supamit Mekchay. 2011. “Novel BsuRI-c.930A>G-FSHb Associated with Litter Size Traits on Large White X Landrace Crossbred Sows.” Journal of Agricultural Science 4 (1): 104–13. doi:10.5539/jas.v4n1p104.

Puffenberger, Erik G, Diane Hu-Lince, Jennifer M Parod, David W Craig, Seth E Dobrin, Andrew R Conway, Elizabeth A Donarum, et al. 2004. “Mapping of Sudden Infant Death with Dysgenesis of the Testes Syndrome (SIDDT) by a SNP Genome Scan and Identification of TSPYL Loss of Function.” Proceedings of the National Academy of Sciences of the United States of America 101 (32): 11689–94. doi:10.1073/pnas.0401194101.

Pusateri, A E, M A Diekman, and W L Singleton. 1999. “Failure of Vitamin A to Increase Litter Size in Sows Receiving Injections at Various Stages of Gestation.” Journal of Animal Science 77 (6): 1532–35. http://www.ncbi.nlm.nih.gov/pubmed/10375230.

Quach, T. A., D. A F Villagómez, G Coppola, A Pinton, E. J. Hart, E. R. Reyes, P. K. Basrur, and W. A. King. 2010. “A Cytogenetic Study of Breeding Boars in Canada.” Cytogenetic and Genome Research 126 (3). Department of Biomedical Sciences, University of Guelph, Guelph, Ont, Canada.: S. Karger AG, Basel: 271– 80. doi:10.1159/000251964.

Quilter, Claire R., Sarah C. Blott, Alan J. Mileham, Nabeel a. Affara, Carole a. Sargent, and Darren K. Griffin. 2002. “A Mapping and Evolutionary Study of Porcine Sex Chromosome Genes.” Mammalian Genome 13 (10): 588–94. doi:10.1007/s00335-002-3026-1.

Raap, A K. 2001. “Overview of Fluorescence in Situ Hybridization Techniques for Molecular Cytogenetics.” Current Protocols in Cytometry / Editorial Board, J. Paul Robinson, Managing Editor ... [et Al.] Chapter 8: Unit 8.1. doi:10.1002/0471142956.cy0801s00.

Radke, David W, and Charles Lee. 2015. “Adaptive Potential of Genomic Structural Variation in Human and Mammalian Evolution.” Briefings in Functional Genomics, May. doi:10.1093/bfgp/elv019.

Ramakers, Christian, Jan M Ruijter, Ronald H.Lekanne Deprez, and Antoon F.M Moorman. 2003. “Assumption- Free Analysis of Quantitative Real-Time Polymerase Chain Reaction (PCR) Data.” Neuroscience Letters. doi:10.1016/S0304-3940(02)01423-4.

Ramayo-Caldas, Y, A Castello, R N Pena, E Alves, A Mercade, C A Souza, A I Fernandez, M Perez-Enciso, and J M Folch. 2010. “Copy Number Variation in the Porcine Genome Inferred from a 60 K SNP BeadChip.” BMC Genomics 11 (October). Departament de Ciencia Animal i dels Aliments, Facultat de Veterinaria, Universitat Autonoma de Barcelona, 08193 Bellaterra, Spain.: 593. doi:10.1186/1471-2164-11-593.

139

Ramayo-Caldas, Yuliaxis, Anna Castelló, Romi N Pena, Estefania Alves, Anna Mercadé, Carla a Souza, Ana I Fernández, Miguel Perez-Enciso, and Josep M Folch. 2010. “Copy Number Variation in the Porcine Genome Inferred from a 60 K SNP BeadChip.” BMC Genomics 11: 593.

Ramos, Antonio M., Richard P M a Crooijmans, Nabeel a. Affara, Andreia J. Amaral, Alan L. Archibald, Jonathan E. Beever, Christian Bendixen, et al. 2009. “Design of a High Density SNP Genotyping Assay in the Pig Using SNPs Identified and Characterized by next Generation Sequencing Technology.” PLoS ONE 4 (8). doi:10.1371/journal.pone.0006524.

Redon, Richard, Shumpei Ishikawa, Karen R Fitch, Lars Feuk, George H Perry, T Daniel Andrews, Heike Fiegler, et al. 2006. “Global Variation in Copy Number in the Human Genome.” Nature 444 (7118): 444–54. doi:10.1038/nature05329.

Ren, D R, J Ren, Y Y Xing, Y M Guo, Y B Wu, G C Yang, H R Mao, and L-S Huang. 2009. “A Genome Scan for Quantitative Trait Loci Affecting Male Reproductive Traits in a White Duroc X Chinese Erhualian Resource Population.” Journal of Animal Science 87 (1): 17–23. doi:10.2527/jas.2008-0923.

Repping, S, H Skaletsky, J Lange, S Silber, Fvd Veen, R D Oates, D C Page, and S Rozen. 2002. “Recombination between Palindromes P5 and P1 on the Human Y Chromosome Causes Massive Deletions and Spermatogenic Failure.” The American Journal of Human Genetics 71 (January): 906–22.

Revay, Tamas, Anh T Quach, Laurence Maignel, Brian Sullivan, and W Allan King. 2015. “Copy Number Variations in High and Low Fertility Breeding Boars.” BMC Genomics 16 (1). ??? 280. doi:10.1186/s12864- 015-1473-9.

Riggs, Pk, and Cl Chrisman. 1991. “Identification of Aphidicolin-Induced Fragile Sites in Domestic Pig Chromosomes.” Genetics Selection Evolution 23 (Suppl 1): S187. doi:10.1186/1297-9686-23-S1-S187.

Rodriguez, A, E Sanz, E De Mercado, E Gomez, M Martin, C Carrascosa, E Gomez-Fidalgo, et al. 2010. “Reproductive Consequences of a Reciprocal Chromosomal Translocation in Two Duroc Boars Used to Provide Semen for Artificial Insemination.” Theriogenology 74 (1). Pig Research Center, Instituto Tecnologico Agrario de Castilla y Leon, Consejeria de Agricultura y Ganaderia, Junta de Castilla y Leon, Ctra. Riaza-Toro s/n., Hontalbilla (Segovia), Spain.: 67–74. doi:10.1016/j.theriogenology.2009.12.020.

Rohrer, G A, J J Ford, T H Wise, J L Vallet, and R K Christenson. 1999. “Identification of Quantitative Trait Loci Affecting Female Reproductive Traits in a Multigeneration Meishan-White Composite Swine Population.” J Anim Sci 77 (6): 1385–91.

Rohrer, G A, T H Wise, D D Lunstra, and J J Ford. 2001. “Identification of Genomic Regions Controlling Plasma FSH Concentrations in Meishan-White Composite Boars.” Physiol Genomics 6: 145–51.

Rønne, M. 1995. “Localization of Fragile Sites in the Karyotype of Sus Scrofa Domestica: Present Status.” Hereditas 122 (2): 153–62.

Ronnie E. Nelson, IT Omtvedt, EJ Turman, DF Stephens and GWA Mahoney. 1970. “Effects of Heat Stress at Various Stages of Pregnancy on Sow Productivity.” Animal Science Research, 71–77.

Rothschild, M F, L Messer, A Day, R Wales, T Short, O Southwood, and G Plastow. 2000. “Investigation of the Retinol-Binding Protein 4 (RBP4) Gene as a Candidate Gene for Increased Litter Size in Pigs.” Mammalian Genome : Official Journal of the International Mammalian Genome Society 11 (1): 75–77. http://www.ncbi.nlm.nih.gov/pubmed/10602998.

Rothschild, Max F, and Anatoly Ruvinsky. 2011. The Genetics of the Pig, 2nd Edition.

140

Rothschild, Max F. 1996. “Genetics and Reproduction in the Pig.” Anim.Reprod.Sci 42 (96): 143–51. doi:10.1016/0378-4320(96)01486-8.

Rovelet-Lecrux, Anne, Didier Hannequin, Gregory Raux, Nathalie Le Meur, Annie Laquerrière, Anne Vital, Cécile Dumanchin, et al. 2006. “APP Locus Duplication Causes Autosomal Dominant Early-Onset Alzheimer Disease with Cerebral Amyloid Angiopathy.” Nature Genetics 38 (1): 24–26. doi:10.1038/ng1718.

Rubin, Carl-Johan, Hendrik-Jan Megens, Alvaro Martinez Barrio, Khurram Maqbool, Shumaila Sayyab, Doreen Schwochow, Chao Wang, et al. 2012. “Strong Signatures of Selection in the Domestic Pig Genome.” Proceedings of the National Academy of Sciences of the United States of America 109 (48): 19529–36. doi:10.1073/pnas.1217149109.

Ruijter, J. M., C. Ramakers, W. M H Hoogaars, Y. Karlen, O. Bakker, M. J B van den hoff, and A. F M Moorman. 2009. “Amplification Efficiency: Linking Baseline and Bias in the Analysis of Quantitative PCR Data.” Nucleic Acids Research 37. doi:10.1093/nar/gkp045.

Saeidi, S, F Shapouri, E Amirchaghmaghi, H Hoseinifar, M Sabbaghian, M a Sadighi Gilani, a a Pacey, and R Aflatoonian. 2013. “Sperm Protection in the Male Reproductive Tract by Toll-like Receptors.” Andrologia 10 (September): 1–7. doi:10.1111/and.12149.

Schook, Lawrence B, Jonathan E Beever, Jane Rogers, Sean Humphray, Alan Archibald, Patrick Chardon, Denis Milan, Gary Rohrer, and Kellye Eversole. 2005. “Swine Genome Sequencing Consortium (SGSC): A Strategic Roadmap for Sequencing the Pig Genome.” Comparative and Functional Genomics 6 (4): 251–55. doi:10.1002/cfg.479.

Schubert, Stephanie, and Jörg Schmidtke. 2010. “Transgenic Mouse Studies to Understand the Regulation, Expression and Function of the Testis-Specific Protein Y-Encoded (TSPY) Gene.” Genes 1 (2): 244–62. doi:10.3390/genes1020244.

Schwerin, M, D Golisch, and E Ritter. 1986. “A Robertsonian Translocation in Swine.” Génét. Sél. Evol 18: 367–74. doi:10.1186/1297-9686-18-4-367.

Seabright, M. 1971. “A Rapid Banding Technique for Human Chromosomes.” Lancet 2 (7731). ENGLAND: 971– 72.

Sebat, J, B Lakshmi, J Troge, J Alexander, J Young, P Lundin, S Maner, et al. 2004. “Large-Scale Copy Number Polymorphism in the Human Genome.” Science 305: 525–28. doi:10.1126/science.1098918.

Shaffer, Lisa G, and Bassem A Bejjani. 2004. “A Cytogeneticist’s Perspective on Genomic Microarrays.” Human Reproduction Update 10 (3): 221–26. doi:10.1093/humupd/dmh022.

Shan, J, C Zhang, J Sun, and L Li. 1994. “[Synaptonemal Complex Analysis of 13/17 Robertsonian Translocation in the Domestic Pig].” Yi Chuan Xue Bao = Acta Genetica Sinica 21 (2): 96–103. http://www.ncbi.nlm.nih.gov/pubmed/8080664.

Shen, Ying, Yuanlong Yan, Yunqiang Liu, Sizhong Zhang, Dong Yang, Peng Zhang, Lei Li, et al. 2013. “A Significant Effect of the TSPY1 Copy Number on Spermatogenesis Efficiency and the Phenotypic Expression of the Gr/gr Deletion.” Human Molecular Genetics 22 (8): 1679–95. doi:10.1093/hmg/ddt004.

Shifei, P, and L Yonggang. 2012. “Molecular Characterization, Polymorphism and Association of Porcine IBP4 Gene.” Folia Biologica 58 (1): 30–36. http://www.ncbi.nlm.nih.gov/pubmed/22464822.

141

Singh, B, K R Fisher, B R Yadav, and P K Basrur. 1994. “Characterization of a Translocation and Its Impact on Fertility in the Pig.” Genome / National Research Council Canada = Genome / Conseil National de Recherches Canada 37 (2). Department of Biomedical Sciences, Ontario Veterinary College, University of Guelph, Canada.: 280–88. doi:10.1139/g94-039.

Skaletsky, H, T Kuroda-Kawaguchi, P J Minx, H S Cordum, L Hillier, L G Brown, S Repping, et al. 2003. “The Male-Specific Region of the Human Y Chromosome Is a Mosaic of Discrete Sequence Classes.” Nature 423 (6942). Howard Hughes Medical Institute, Whitehead Institute, and Department of Biology, Massachusetts Institute of Technology, 9 Cambridge Center, Cambridge, Massachusetts 02142, USA.: 825–37. doi:10.1038/nature01722.

Soh, Y.Q. Shirleen, Jessica Alföldi, Tatyana Pyntikova, Laura G. Brown, Tina Graves, Patrick J. Minx, Robert S. Fulton, et al. 2014. “Sequencing the Mouse Y Chromosome Reveals Convergent Gene Acquisition and Amplification on Both Sex Chromosomes.” Cell 159 (4): 800–813. doi:10.1016/j.cell.2014.09.052.

Solinas-Toldo, S, S Lampel, S Stilgenbauer, J Nickolenko, A Benner, H Dohner, T Cremer, and P Lichter. 1997. “Matrix-Based Comparative Genomic Hybridization: Biochips to Screen for Genomic Imbalances.” Genes, Chromosomes & Cancer 20 (4). Organisation komplexer Genome, Deutsches Krebsforschungszentrum, Heidelberg, Germany.: 399–407.

Speel, Ernst J M, Anton H N Hopman, and Paul Komminoth. 2006. “Tyramide Signal Amplification for DNA and mRNA in Situ Hybridization.” Methods in Molecular Biology (Clifton, N.J.) 326: 33–60. doi:10.1385/1- 59745-007-3:33.

Stankiewicz, Paweł, and James R Lupski. 2010. “Structural Variation in the Human Genome and Its Role in Disease.” Annual Review of Medicine 61 (January): 437–55. doi:10.1146/annurev-med-100708-204735.

Stone, B a. 1981. “Thermal Characteristics of the Testis and Epididymis of the Boar.” Journal of Reproduction and Fertility 63 (2): 551–57. doi:10.1530/jrf.0.0630551.

Strange, R C, M A Spiteri, S Ramachandran, and A A Fryer. 2001. “Glutathione-S-Transferase Family of Enzymes.” Mutation Research 482: 21–26. doi:10.1016/S0027-5107(01)00206-8.

Sumner, A T. 1972. “A Simple Technique for Demonstrating Centromeric Heterochromatin. Expl.” Cell Res 75 (1): 304–6. http://www.ncbi.nlm.nih.gov/pubmed/4117921.

Świtoński, M. 1991. “Paracentric Inversion Involving NOR of Chromosome 8 in a Boar: Studies of Synaptonemal Complexes under a Light Microscope.” Genetics, Selection, Evolution : GSE 23 (3): 181–89. doi:10.1186/1297-9686-23-3-181.

Takada, Shuji, Eugene Berezikov, Young Lim Choi, Yoshihiro Yamashita, and Hiroyuki Mano. 2009. “Potential Role of miR-29b in Modulation of Dnmt3a and Dnmt3b Expression in Primordial Germ Cells of Female Mouse Embryos.” RNA (New York, N.Y.) 15 (8): 1507–14. doi:10.1261/rna.1418309.

Tarocco, C, F Franchi, and G Croci. 1987. “A New Reciprocal Translocation Involving Chromosomes 1/14 in a Boar [chromosome Abnormality, Breakpoint].” Genetique Selection Evolution 19 (3): 381–86.

Taylor, Angela I. 2008. “Patau’s, Edwards' and Cri Du Chat Syndromes: A Tabulated Summary of Current Findings.” Developmental Medicine & Child Neurology 9 (1): 78–86. doi:10.1111/j.1469- 8749.1967.tb02213.x.

142

Taylor, Barry S, Jordi Barretina, Nicholas D Socci, Penelope Decarolis, Marc Ladanyi, Matthew Meyerson, Samuel Singer, and Chris Sander. 2008. “Functional Copy-Number Alterations in Cancer.” PloS One 3 (9): e3179. doi:10.1371/journal.pone.0003179.

“The International HapMap Project.” 2003 426 (6968). Macmillian Magazines Ltd.: 789–96.

Tonk, V S, H E Wyandt, X Huang, N Patel, D L Morgan, M Kukolich, L H Lockhart, and G V Velagaleti. 2003. “Disease Associated Balanced Chromosome Rearrangements (DBCR): Report of Two New Cases.” Annales de Genetique 46 (1). Department of Pediatrics, Texas Tech University Health Sciences Center, TX, Lubbock, USA.: 37–43.

Tremellen, Kelton. 2008. “Oxidative Stress and Male Infertility--a Clinical Perspective.” Human Reproduction Update 14 (3): 243–58. doi:10.1093/humupd/dmn004.

Tribout, T, A Ducos, L Maignel, and JP Bidanel. 2000. “La Detection de Verrats Porteurs D’anomalies Chromosomiques Utilisation Du Systeme D'information BLUP.” TECHNIPORC 23: 19–24. https://scholar.google.ca/scholar?hl=en&q=La+detection+de+verrats+porteurs+d%E2%80%99anomalies+chr omosomiques+Utilisation+du+systeme+d%27information+BLUP&btnG=&as_sdt=1%2C5&as_sdtp=#0.

Tse, C, D Brault, J Gligorov, M Antoine, R Neumann, J P Lotz, and J Capeau. 2005. “Evaluation of the Quantitative Analytical Methods Real-Time PCR for HER-2 Gene Quantification and ELISA of Serum HER-2 Protein and Comparison with Fluorescence in Situ Hybridization and Immunohistochemistry for Determining HER-2 Status in Breast Cancer Pa.” Clinical Chemistry 51 (7). Laboratoire de Biochimie, Hopital Tenon, Paris, France. [email protected]: 1093–1101. doi:10.1373/clinchem.2004.044305.

Tzocheva, K. 1994. “A New Reciprocal Translocation rcp(1p-;11p+) in the Karyotype of the Boar.” Hereditas 120 (2): 187–89. http://www.ncbi.nlm.nih.gov/pubmed/8083061.

USDA, and Foreign Agricultural Service. 2014. Livestock and Poultry: World Markets and Trade.

Valsesia, Armand, Aurélien Macé, Sébastien Jacquemont, Jacques S Beckmann, and Zoltán Kutalik. 2013. “The Growing Importance of CNVs: New Insights for Detection and Clinical Interpretation.” Frontiers in Genetics 4 (January): 92. doi:10.3389/fgene.2013.00092.

Van Dyke, D L, L Weiss, J R Roberson, and V R Babu. 1983. “The Frequency and Mutation Rate of Balanced Autosomal Rearrangements in Man Estimated from Prenatal Genetic Studies for Advanced Maternal Age.” American Journal of Human Genetics 35 (2): 301–8.

Villagómez, D a F, and a Pinton. 2008. “Chromosomal Abnormalities, Meiotic Behavior and Fertility in Domestic Animals.” Cytogenetic and Genome Research 120 (1-2): 69–80. doi:10.1159/000118742.

Villagómez, D a F, a Pinton, D A Villagomez, and a Pinton. 2008. “Chromosomal Abnormalities, Meiotic Behavior and Fertility in Domestic Animals.” Cytogenetic and Genome Research 120 (1-2). Department of Biomedical Sciences, University of Guelph, Guelph, ON, Canada. [email protected]: S. Karger AG, Basel: 69–80. doi:10.1159/000118742.

Villagomez, D A, I Gustavsson, B Alabay, and L Ploen. 1993. “Meiotic Chromosome Asynapsis in a Boar with a Reciprocal Translocation and Acquired Testicular Degeneration.” Hereditas 118 (2). Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, Uppsala.: 101–11.

Villagómez, D a, I Gustavsson, L Jönsson, and L Plöen. 1995. “Reciprocal Chromosome Translocation, rcp(7;17)(q26;q11), in a Boar Giving Reduced Litter Size and Increased Rate of Piglets Dying in the Early Life.” Hereditas 122: 257–67.

143

Vodicka, Radek, Radek Vrtel, Ladislav Dusek, Arvind R Singh, Katerina Krizova, Veronika Svacinova, Vera Horinova, et al. 2007. “TSPY Gene Copy Number as a Potential New Risk Factor for Male Infertility.” Reproductive Biomedicine Online 14: 579–87. doi:10.1016/S1472-6483(10)61049-8.

Vogel, Tanja, Holly Boettger-Tong, Indrajit Nanda, Frank Dechend, Alexander I. Agulnik, Colin E. Bishop, Michael Schmid, and Jörg Schmidtke. 1998. “A Murine TSPY.” Chromosome Research 6: 35–40. doi:10.1023/A:1009214307764.

Vogt, P H, A Edelmann, S Kirsch, O Henegariu, P Hirschmann, F Kiesewetter, F M Köhn, et al. 1996. “Human Y Chromosome Azoospermia Factors (AZF) Mapped to Different Subregions in Yq11.” Human Molecular Genetics 5 (7): 933–43. http://www.ncbi.nlm.nih.gov/pubmed/8817327.

Wang, Jing, Dexter Duncan, Zhiao Shi, and Bing Zhang. 2013. “WEB-Based GEne SeT AnaLysis Toolkit (WebGestalt): Update 2013.” Nucleic Acids Research 41 (Web Server issue): W77–83. doi:10.1093/nar/gkt439.

Wang, Jiying, Jicai Jiang, Weixuan Fu, Li Jiang, Xiangdong Ding, Jian-Feng Liu, and Qin Zhang. 2012. “A Genome-Wide Detection of Copy Number Variations Using SNP Genotyping Arrays in Swine.” BMC Genomics. doi:10.1186/1471-2164-13-273.

Wang, Jiying, Jicai Jiang, Haifei Wang, Huimin Kang, Qin Zhang, and Jian-Feng Liu. 2014. “Enhancing Genome- Wide Copy Number Variation Identification by High Density Array CGH Using Diverse Resources of Pig Breeds.” PloS One 9: e87571. doi:10.1371/journal.pone.0087571.

Wang, Jiying, Haifei Wang, Jicai Jiang, Huimin Kang, Xiaotian Feng, Qin Zhang, and Jian Feng Liu. 2013. “Identification of Genome-Wide Copy Number Variations among Diverse Pig Breeds Using SNP Genotyping Arrays.” PLoS ONE 8 (7). doi:10.1371/journal.pone.0068683.

Wang, Kai, Mingyao Li, Dexter Hadley, Rui Liu, Joseph Glessner, Struan F a Grant, Hakon Hakonarson, and Maja Bucan. 2007. “PennCNV: An Integrated Hidden Markov Model Designed for High-Resolution Copy Number Variation Detection in Whole-Genome SNP Genotyping Data.” Genome Research 17: 1665–74. doi:10.1101/gr.6861907.

Wang, Ligang, Xin Liu, Longchao Zhang, Hua Yan, Weizhen Luo, Jing Liang, Duxue Cheng, et al. 2013. “Genome- Wide Copy Number Variations Inferred from SNP Genotyping Arrays Using a Large White and Minzhu Intercross Population.” PLoS ONE 8 (10): 1–8. doi:10.1371/journal.pone.0074879.

Wang, Tao, Mingzhou Li, Jiuqiang Guan, Penghao Li, Huiyu Wang, Yanqin Guo, Surong Shuai, and Xuewei Li. 2011. “MicroRNAs miR-27a and miR-143 Regulate Porcine Adipocyte Lipid Metabolism.” International Journal of Molecular Sciences 12 (11): 7950–59. doi:10.3390/ijms12117950.

Wang, Yanan, Zhonglin Tang, Yaqi Sun, Hongyang Wang, Chao Wang, Shaobo Yu, Jing Liu, et al. 2014. “Analysis of Genome-Wide Copy Number Variations in Chinese Indigenous and Western Pig Breeds by 60 K SNP Genotyping Arrays.” PLoS ONE 9 (9): e106780. doi:10.1371/journal.pone.0106780.

Warburton, D. 1991. “De Novo Balanced Chromosome Rearrangements and Extra Marker Chromosomes Identified at Prenatal Diagnosis: Clinical Significance and Distribution of Breakpoints.” American Journal of Human Genetics 49 (5). Department of Genetics, Columbia University, New York, NY.: 995–1013.

Wathes, D Claire, D Robert E Abayasekara, and R John Aitken. 2007. “Polyunsaturated Fatty Acids in Male and Female Reproduction.” Biology of Reproduction 77 (2): 190–201. doi:10.1095/biolreprod.107.060558.

144

Weaver, S, S Dube, A Mir, J Qin, G Sun, R Ramakrishnan, R C Jones, and K J Livak. 2010. “Taking qPCR to a Higher Level: Analysis of CNV Reveals the Power of High Throughput qPCR to Enhance Quantitative Resolution.” Methods (San Diego, Calif.) 50 (4). Fluidigm Corporation, 7000 Shoreline Court, Suite 100, South San Francisco, CA 94080, USA.: Elsevier Inc: 271–76. doi:10.1016/j.ymeth.2010.01.003.

Wells, Jonathan C K. 2014. “Commentary: Paternal and Maternal Influences on Offspring Phenotype: The Same, Only Different.” International Journal of Epidemiology 43 (3): 772–74. doi:10.1093/ije/dyu055.

Whaley, S L, V S Hedgpeth, and J H Britt. 1997. “Evidence That Injection of Vitamin A before Mating May Improve Embryo Survival in Gilts Fed Normal or High-Energy Diets.” Journal of Animal Science 75 (4): 1071–77. http://www.ncbi.nlm.nih.gov/pubmed/9110222.

Wilkie, P J, A A Paszek, C W Beattie, L J Alexander, M B Wheeler, and L B Schook. 1999. “A Genomic Scan of Porcine Reproductive Traits Reveals Possible Quantitative Trait Loci (QTLs) for Number of Corpora Lutea.” Mammalian Genome : Official Journal of the International Mammalian Genome Society 10 (6): 573–78.

Wnuk, M, and Z Kotylak. 2005. “KARYOTYPE ANALYSIS IN PIGS—HYBRIDS OF EUROPEAN WILD BOAR (SUS SCROFA SCROFA) AND DOMESTIC PIG (SUS SCROFA DOMESTICA).” … INSTITUTE OF ANIMAL …. http://www.researchgate.net/profile/Barbara_Niwinska/publication/264536390_EFFECTS_OF_TYPE_OF_LI QUID_FEED_AND_FEEDING_FREQUENCY_ON_RUMEN_DEVELOPMENT_AND_REARING_PERF ORMANCE_OF_CALVES/links/53e32f600cf2b9d0d832f2fe.pdf#page=10.

Wu, Wei, Jing Lu, Qiuqin Tang, Simin Zhang, Beilei Yuan, Jing Li, Di Wu, et al. 2013. “GSTM1 and GSTT1 Null Polymorphisms and Male Infertility Risk: An Updated Meta-Analysis Encompassing 6934 Subjects.” Scientific Reports 3 (January): 2258. doi:10.1038/srep02258.

Wu, Xudong, Dinglin Zhang, and Guohui Li. 2012. “Insights into the Regulation of Human CNV-miRNAs from the View of Their Target Genes.” BMC Genomics 13 (1). BMC Genomics: 707. doi:10.1186/1471-2164-13-707.

Xu, Lingyang, Yali Hou, Derek Bickhart, Jiuzhou Song, and George Liu. 2013. “Comparative Analysis of CNV Calling Algorithms: Literature Survey and a Case Study Using Bovine High-Density SNP Data.” Microarrays 2: 171–85. doi:10.3390/microarrays2030171.

Xue, Yali, and Chris Tyler-Smith. 2011. “An Exceptional Gene: Evolution of the TSPY Gene Family in Humans and Other Great Apes.” Genes. doi:10.3390/genes2010036.

Yang, H, H R Jung, S Solinas-Toldo, E Lang, R Bolt, R Fries, and G Stranzinger. 1992. “A Reciprocal Whole-Arm Translocation, rcp(1;6)(1p6p;1q6q) in a Boar, Localization of the Breakpoints, and Reassignment of the Genes for Glucose Phosphate Isomerase (GPI) and Calcium Release Channel (CRC).” Cytogenetics and Cell Genetics 61 (1). Department of Animal Sciences, Swiss Federal Institute of Technology, ETH-Zentrum, Zurich.: 67–74.

Yang, M Y, and S E Long. 1993. “Folate Sensitive Common Fragile Sites in Chromosomes of the Domestic Pig (Sus Scrofa).” Research in Veterinary Science 55 (2). Department of Animal Husbandry, University of Bristol.: 231–35. doi:10.1016/0034-5288(93)90086-U.

Yang, Xiaokui, Ying Zhou, Sha Peng, Liang Wu, Hai-Yan Lin, Shuyu Wang, and Hongmei Wang. 2012. “Differentially Expressed Plasma microRNAs in Premature Ovarian Failure Patients and the Potential Regulatory Function of Mir-23a in Granulosa Cell Apoptosis.” Reproduction (Cambridge, England) 144 (2): 235–44. doi:10.1530/REP-11-0371.

145

Y-I. Miyake, T. Matsubara, M. Hata and Y. Kaneda. 1994. “Chromosomal Pericentric Inversion Detected in a Sow and Her Piglets.” Theriogenology 42: 241–46.

Zarrei, Mehdi, Jeffrey R Macdonald, Daniele Merico, and Stephen W Scherer. 2015. “A Copy Number Variation Map of the Human Genome.” Nature Publishing Group 16 (February). Nature Publishing Group: 172–83. doi:10.1038/nrg3871.

Zhang, T Q, L C Buoen, G R Ruth, A F Weber, R Morrison, B Christianson, W Marsh, and G R Anderson. 1991. “Two Different Types of Chromosomal Translocations Found in U.S. Boars.” In Proc. 7th North American Colloq. Domest. Anim. Cytogenet. Gene Mapp., Philadelph:2–5.

146

APPENDIX

Appendix I: List of reciprocal translocations in pigs reported in the literature

No Reciprocal Breed Reference Translocation 1 (1;2;17)(1q11;1q17;2p16 Large White (Perucatti et al. 2014) ;17q13) 2 (1;4)(q27;q21) Germany Pietrain (Alain Ducos et al. 2007) 3 (1;5)(q21;q21) (Danielak-Czech et al. 1994) 4 (1;5)(q21;q23) Landrace This thesis (case 1) 5 (1;6) (p13;q35) Large white (Locniskar et al. 1976) 6 (1;6)(p11;q11) Swiss large white (H. Yang et al. 1992) 7 (1;6)(q12;q22) Gascon (Alain Ducos et al. 2002) 8 (1;6)(q17;p11) Synthetic (large (Perucatti et al. 2014) white+pietrain) 9 (1;6)(q17;q35) Synthetic Ducos et al., 2002 10 (1; 6)(p22,q12) Synthetic TA Quach et al., 2009 11 (1;7)(q17;q13) Not indicated (Alain Ducos et al. 2007) 12 (1;7)(q17;q26) Large white Ducos et al., 2002 13 (1;7)(q213;q24) Swedish landrace (Gustavsson 1988b) 14 (1;8)(1p-;8q+) Gustavsson et al., 1976 15 (1;8)(p13;q27) Swedish yorkshire (Gustavsson et al. 1983) 16 (1;9)(1p-;9p+) Large white Ducos et al., 2002 17 (1;11)(p21;q15) (Tzocheva 1994) 18 (1;11)(p23;q15) (KUOKKANEN and MÄKINEN 1988) 19 (1;11)(q-;p+) (Rodriguez et al. 2010) 20 (1;11)(q11;q11) French Pietrain (Alain Ducos et al. 2007) 21 (1;11)(q24;p13) Not indicated (Alain Ducos et al. 2007) 22 (1;13)(q27;q41) Duroc (Alain Ducos et al. 2007) 23 (1;14)(1q+;14q-) (Golisch et al. 1982) 24 (1;14)(p25;q15) Gustavsson, 1988 25 (1;14)(q17;q21) (Tarocco et al. 1987) 26 (1;14)(q212;q22) (Zhang et al. 1991) 27 (1;14)(q23;q21) Golisch et al., 1982 28 (1;15)(p25;q13) Koukkanen and Makinen, 1988 29 (1;15)(q17;q22) Synthetic Ducos et al., 2007 30 (1;15)(q211;q13) Duroc This thesis (case 2) 31 (1;15)(q26;qter) (Popescu et al. 1988) 32 (1;16)(p11;q22) Fries and Stranzinger, 1982 33 (1;16)(q11;q11) Germany landrace (Forster, Willeke, and Richter 1981) 34 (1;17)(p11;q11) Pietrain (Alain Ducos et al. 2007) 35 (1;17)(q21;q11) Gustavsson, 1988 36 (1;18)(q213;q21) (HENRICSON and BÄCKSTRÖM 1964) 37 (2;3) Large white Ducos et al., 2002 38 (2;5)(p16;p11) Landrace This thesis (case 3) 39 (2;6)(p15;q27) Yorkshire Ducos et al., 2002 40 (2;8)(p11;p13) French Large white (Alain Ducos et al. 2007) 41 (2;9)(q13;q24) French Large white (Alain Ducos et al. 2007) 42 (2;9;14)(q23;q22;q25) Finnish Yorkshire (Mäkinen et al. 1997) 43 (2;14)(p14;q23) Hampolish (Villagomez et al. 1993) synthetichire 44 (2;14)(p15;q26) crossbred (Alain Ducos et al. 2007) 45 (2;14)(q13;q27) French landrace Ducos et al., 2002

147

46 (2;14)(q21;q24) Not indicated (Alain Ducos et al. 2007) 47 (2;15)(q28;q24) Not indicated (A Ducos et al. 2007) 48 (2;15;4)(p13;q26;p15) (Makinen et al. 1987) 49 (2;16)(q28;q21) sino European (Alain Ducos et al. 2007) 50 (2;17)(p12;q14) French Duroc (A Ducos et al. 2007) 51 (3;4)(p15;q13) Landrace This thesis (case 4) 52 (3;5)(p13;q23) French landrace Ducos et al., 2002 53 (3;7)(p15;q21) Large white (Gabriel-Robez et al. 1988) 54 (3;8)(q25;p21) Synthetic (A Ducos et al. 2007) 55 (3;11)(q13;p11) Synthetic (A Ducos et al. 2007) 56 (3;12)(p13;q15) Yorkshire This thesis (case 5) 57 (3;13)(p15;q31) Large white Ducos et al., 2002 58 (3;15)(q27;q13) Large white K. Massip et al., 2008 59 (3;16)(q23;q22) Crossbred Ducos et al., 2007 60 (3;17)(p15;q21) Popescu et al., 1983 61 (4;5)(p13;q21) Crossbred (A Ducos et al. 2007) 62 (4;6)(q21;q14-15) Pietrain Ducos et al., 2002 63 (4;6)(q21;q28) Large white Ducos et al., 2002 64 (4;12)(p13;q13) Synthetic Ducos et al., 2002 65 (4;12)(q21;q13) Crossbred (A Ducos et al. 2007) 66 (4;13)(p15;q41) Not indicated (A Ducos et al. 2007) 67 (4;13)(q25;q41) (MÄKINEN and REMES 1986) 68 (4;14)(p11;q11) Large white x (Popescu and Legault 1979) French Landrace 69 (4;14)(q25;q22) (Popescu and Boscher 1982) 70 (4;15)(q11;qter) (Popescu et al. 1988) 71 (4;15)(q25;q11) French landrace (A Ducos et al. 2007) 72 (4;16)(q25;q21) Synthetic (A Ducos et al. 2007) 73 (5;7)(q23;p11) sino European (A Ducos et al. 2007) 74 (5;8)(p11;p23) Synthetic Ducos et al., 2002 75 (5;8)(p11;q11) Large white Ducos et al., 2002 76 (5;8)(q12;q27) (Gustavsson 1988b) 77 (5;9)(q21;p13) French landrace (A Ducos et al. 2007) 78 (5,9)(p11;p24) French landrace (A Ducos et al. 2007) 79 (5,14)(q21;q12) French Duroc (A Ducos et al. 2007) 80 (5;14)(p11;q11) English landrace x (Popescu et al. 1984) Duroc 81 (5;17)(p12;q13) Yorkshire Ducos et al., 2002 82 (6;7)(p15;q13) Duroc This thesis (case 6) 83 (6;8)(q33;q26) (Bonneau et al. 1991; Jaafar et al. 1992) 84 (6;13)(p13;q49) French Large white (A Ducos et al. 2007) 85 (6;13)(p15;q41) Large white x Ducos et al., 2002 Pietrain 86 (6;14)(p11;q11) England Large white (Madan et al. 1978) x Essex 87 (6;14)(q27;q21) Large white x Ducos et al., 2002 Pietrain 88 (6;15)(6p+;15q-) Belgian landrace (BOUTERS et al. 1974) 89 (6;15)(p15;q12) (Bonneau et al. 1991) 90 (6;16)(p13;q23) (Kociucka et al. 2014) 91 (6;16)(q11;q11) Synthetic Ducos et al., 2002 92 (7;8)(q24;p23) French landrace x Ducos et al., 2002 Meishan 93 (7;9)(q11;q26) French Large white (A Ducos et al. 2007) 94 (7;9)(q15;q15) French Large white (A Ducos et al. 2007) 95 (7;10)(q13;q11) Not indicated (A Ducos et al. 2007)

148

96 (7;11)(q21;q11) Swedish yorkshire (Gustavsson and Settergren 1984) 97 (7;12)(q11;p15) French Pietrain (A Ducos et al. 2007) 98 (7;12)(q23;q15) Finnish Yorkshire Koukkanen and Makinen, 1987 99 (7;13)(p13;q21) Hampolish Gustavsson et al., 1988 synthetichire 100 (7;13)(q13;q46), Danielak-Czech et al., 1997 101 (7;14)(q15;q27) crossbred (A Ducos et al. 2007) 102 (7;14)(q26;q25) Not indicated (A Ducos et al. 2007) 103 (7;15)(q25;q25) Large White (POPESCU et al. 1983) 104 (7;15)(q13;q13) Duroc This thesis (case 7) 105 (7;15)(q24;q12) Large White Konfortovaet al., 1995 106 (7;17)(q26;q11) Hampolish Villagomez et al., 1995 synthetichire 107 (8;10)(p11;q13) Finnish Yorkshire Makinen et al., 1999 108 (8;12)(p11;p11) Not indicated (A Ducos et al. 2007) 109 (8;13)(p21;q41) Landrace This thesis (case 8) 110 (8;14)(p21;q25) Danielak-Czech et al., 1997 111 (9;11)(9p+;11q-) Gustavsson et al., 1976 112 (9;11)(q14;p13) French Meishan (A Ducos et al. 2007) 113 (9;11)(q24;q11) Swedish yorkshire Gustavsson et al., 1983 114 (9;14)(p24;q15) Synthetic (A Ducos et al. 2007) 115 (9;14)(p24;q27) Yorkshire TA Quach et al., 2009 116 (9;14)(q14;q23) Polish synthetic Rejduch et al., 2003 117 (9;15)(p24;q13) Large white x french Ducos et al., 2002 landrace 118 (9;17)(p24;q23) crossbred (A Ducos et al. 2007) 119 (10;11)(q16;q13) French Pietrain (A Ducos et al. 2007) 120 (10;13)(p13,q45) Synthetic TA Quach et al., 2009 121 (10;13)(q13;q22) Synthetic (A Ducos et al. 2007) 122 (10;13)(q16;q21) Danielak-Czech et al., 2007 123 (10;17)(q11;q21) Large white (A Ducos et al. 2007) 124 (10;18)(p11;q24) + French Pietrain (A Ducos et al. 2007) (6;8)(p15;q27) 125 (11;13)(11q+;13q-) Large white Ducos et al., 2002 126 (11;15) (p15;q13) Swedish Landrace (Henricson and Bäckström 1964; Hageltorn et al. 1973) 127 (11;16)(p14;q14) Large white x Ducos et al., 2002 Pietrain 128 (11;17)(p13;q21) sino European (A Ducos et al. 2007) 129 (12;13)(q11;q11) Astachova et al., 1991 130 (12,14)(q13;q15) French Duroc (A Ducos et al. 2007) 131 (12;14)(q15;q13) French Pietrain (A Ducos et al. 2007) 132 (12;14)(q15;q21) Duroc This thesis (case 9) 133 (13;14)(13q-;14q+) Swedish yorkshire (Hageltorn et al. 1976) 134 (13;14)(q21;q27) Hageltorn et al., 1976 (King 1981) 135 (13;14)(q31;q21) + Synthetic (A Ducos et al. 2007) inv(2)(p13;q12) 136 (13;15)(q31;q26) French Pietrain (A Ducos et al. 2007) 137 (13;16)(q41;q21) Not indicated (A Ducos et al. 2007) 138 (13;17)(q41;q11) Large white x Ducos et al., 2002 Pietrain 139 (14;15)(q28;q13) sino European (A Ducos et al. 2007) 140 (14;15)(q29;q24) (Gustavsson and Jonsson 1992) 141 (14;16)(q13;q21) French Pietrain (A Ducos et al. 2007) 142 (15;16)(q26;q21) Swedish Yorkshire Gustavsson et al., 1988

149

143 (15;17) French landrace Ducos et al., 2002 144 (15;17)(q13;q21) Large white Ducos et al., 2002 145 (16;17)(q23;q14) (Jaafar et al. 1992) 146 (16;17)(q23;q21) Popescu and Boscher, 1986 147 (17,18)(q21;q11) French Pietrain (A Ducos et al. 2007) 148 (X;13)(q24;q21) Gustavsson et al., 1989 149 (X;14)(p21;q11) (Singh et al. 1994) 150 (Y;1)(q16;qter) (Barasc et al. 2012) 151 (Y;14)(q10;q11) French Duroc (A Ducos et al. 2007)

Appendix II: List of rearrangements in pigs different from reciprocal translocations reported in the literature

No Rearrangement Breed Reference 1 dup(9p+) Large white Ducos et al., 2002 2 dup(14q+) Large white Ducos et al., 2002 3 Inv(1)(p21;q11) (Miyake et al. 1994) 4 inv(1)(p21;q21) Synthetic Ducos et al., 2002 5 Inv(1)(p21;q210) Massip K et al., 2009 6 Inv(1)(p22;q11) Polish Landrace Danielak-Crech et al.,1996 7 Inv(1)(p24;q29) French landrace Ducos et al., 2007 8 Inv(1)(q12;q24) Massip K et al., 2009 9 inv(1)(q18;q24) Large white Ducos et al., 2002 10 Inv(2)(p11;q11) Massip K et al., 2009 11 Inv(2)(p11;q21) French Pietrain Ducos et al., 2007 12 inv(2)(p13;q11) Synthetic Ducos et al., 2002 13 Inv(2)(p13;q12) Synthetic Ducos et al., 2007 14 Inv(2)(q13;q25) Duroc Ducos et al., 2007 15 inv(4)(p14;q23) Large white Ducos et al., 2002 16 Inv(4)(p15;q24) Synthetic (Perucatti et al. 2014) 17 Inv(6)(q13;q35) Landrace This thesis (case 12) 18 Inv(6)(p14;q12) Sino European Ducos et al., 2007 19 Inv(8)(p11;p12) Polish Landrace (Świtoński 1991) 20 Inv(8)(p11;q25) Large white Ducos et al., 2007 21 Inv(8)(p21;q11) Pietrain Ducos et al., 2007 22 Inv(8)(q11;q25) Duroc This thesis (case 11) 23 inv(9)(p12;p22) Synthetic Ducos et al., 2002 24 inv(16) Synthetic Ducos et al., 2002 25 Rob(13;17)=der(13;17)(q10;q10) French landrace Ducos et al., 2007 Alonso and Cautu, 1982 Miyake et al., 1977 This thesis (case 10) 26 Rob(14;15)=der(14;15)(q10;q10) French landrace Ducos et al., 2007 27 Rob(14;17)=der(14;17)(q10;q10) Not indicated Ducos et al., 2007 28 Rob(16/17) (Astachova et al. 1991)

150

Appendix III: chromosomal rearrangements reported in in vivo and fragile sites related to chromosome number 1

No Rearrangement Fragile Site Breed Reference 1 (1;2;17)(1q11;1q17;2p16 1q11, 1q17 Large White P. Cruz-Vigo el al., 2014 ;17q13) 2 (1;4)(q27;q21) 1q27 Germany Pietrain Ducos et al., 2007 3 (1;5)(q21;q21) 1q21 Danielak-Czech et al., 1994 4 (1;5)(q21;q23) 1q21 Landrace This thesis (case 1) 5 (1;6) (p13;q35) 1p13 Large white Locniskar et al., 1976 6 (1;6)(1p6q;1q6q) 1p11 Swiss large white Yang et al., 1992 7 (1;6)(q12;q22) Gascon Ducos et al., 2002 8 (1;6)(q17;p11) 1q17 Synthetic (large M. Martine-Lluch et al., 2014 white+pietrain) 9 (1;6)(q17;q35) 1q17 Synthetic Ducos et al., 2002 10 (1; 6)(p22,q12) 1p22 Synthetic TA Quach et al., 2009 11 (1;7)(q17;q13) 1q17 Not indicated Ducos et al., 2007 12 (1;7)(q17;q26) 1q17 Large white Ducos et al., 2002 13 (1;7)(q213;q24) Swedish landrace Gustavsson et al. 1988 14 (1;8)(1p-;8q+) Gustavsson et al., 1976 15 (1;8)(p13;q27) 1p13 Swedish Gustavsson et al., 1983 yorkshire 16 (1;9)(1p-;9p+) Large white Ducos et al., 2002 17 (1;11)(p21;q15) Tzocheva, 1994 18 (1;11)(p23;q15) 1p23 Koukkanen and Makinen, 1988 19 (1;11)(q-;p+) A. Rodriguez et al., 2009 20 (1;11)(q11;q11) 1q11 French Pietrain Ducos et al., 2007 21 (1;11)(q24;p13) Not indicated Ducos et al., 2007 22 (1;13)(q27;q41) 1q27 Duroc Ducos et al., 2007 23 (1;14)(1q+;14q-) Golisch et al., 1982 24 (1;14)(p25;q15) 1p25, 14q15 Gustavsson, 1988 25 (1;14)(q17;q21) 1q17, 14q21 Tarocco et al., 1987 26 (1;14)(q212;q22) Zhang et al., 1991 27 (1;14)(q23;q21) 14q21 Golisch et al., 1982 28 (1;15)(p25;q13) 1p25 Koukkanen and Makinen, 1988 29 (1;15)(q17;q22) 1q17 Synthetic Ducos et al., 2007 30 (1;15)(q211;q13) 1q211 Duroc This thesis (case 2) 31 (1;15)(q26;qter) 1q26 Popescu et al., 1988 32 (1;16)(p11;q22) 1p11 Fries and Stranzinger, 1982 33 (1;16)(q11;q11) 1q11 Germany Forster et al., 1981 landrace 34 (1;17)(p11;q11) 1p11 Pietrain Ducos et al., 2007 35 (1;17)(q21;q11) 1q21 Gustavsson, 1988 36 (1;18)(q213;q21) Henricson and Backstrom, 1964 37 Inv(1)(p21;q11) q11 Miyake et al., 1994 38 inv(1)(p21;q21) q21 Synthetic Ducos et al., 2002 39 Inv(1)(p21;q210) q21 Massip K et al., 2009 40 Inv(1)(p22;q11) p22 Polish Landrace Danielak-Crech et al.,1996 41 Inv(1)(p24;q29) p24, q29 French landrace Ducos et al., 2007 42 Inv(1)(q12;q24) Massip K et al., 2009 43 inv(1)(q18;q24) Large white Ducos et al., 2002 44 (Y;1)(q16;qter) H. Barasc et al., 2011

151

Appendix IV: chromosomal rearrangements reported in in vivo and fragile sites related to chromosome number 14 No Rearrangement Fragile Site Breed Reference 1 (1;14)(1q+;14q-) Golisch et al., 1982 2 (1;14)(p25;q15) 1p25, 14q15 Gustavsson, 1988 3 (1;14)(q17;q21) 1q17, 14q21 Tarocco et al., 1987 4 (1;14)(q212;q22) Zhang et al., 1991 5 (1;14)(q23;q21) 14q21 Golisch et al., 1982 6 (2;14)(p14;q23) Hampolish Villagomez et al., 1993 synthetichire 7 (2;14)(p15;q26) 14q26 crossbred Ducos et al., 2007 8 (2;14)(q13;q27) 14q27 French landrace Ducos et al., 2002 9 (2;14)(q21;q24) 1q21 Not indicated Ducos et al., 2007 10 (4;14)(p11;q11) Large white x Popescu and Legault, 1979 French Landrace 11 (2;9;14)(q23;q22;q25) Mäkinen et al. 1997 12 (4;14)(q25;q22) Popescu and Boscher, 1982 13 (5,14)(q21;q12) French Duroc Ducos et al., 2007 14 (5;14)(p11;q11) English landrace x Popescu et al., 1984 Duroc 15 (6;14)(p11;q11) England Large Madan et al., 1978 white x Essex 16 (6;14)(q27;q21) 14q21 Large white x Ducos et al., 2002 Pietrain 17 (7;14)(q15;q27) 14q27 crossbred Ducos et al., 2007 18 (7;14)(q26;q25) 14q25 Not indicated Ducos et al., 2007 19 (8;14)(p21;q25) 14q25 Danielak-Czech et al., 1997 20 (9;14)(p24;q15) 14q15 Synthetic Ducos et al., 2007 21 (9;14)(p24;q27) 14q27 Yorkshire TA Quach et al., 2009 22 (9;14)(q14;q23) Polish synthetic Rejduch et al., 2003 23 (12,14)(q13;q15) 14q15 French Duroc Ducos et al., 2007 24 (12;14)(q15;q13) French Pietrain Ducos et al., 2007 25 (12;14)(q15;q21) 14q21 Duroc This thesis (case 9) 26 (13;14)(13q-;14q+) Swedish yorkshire Hageltorn et al., 1976 27 (13;14)(q21;q27) 14q27 Hageltorn et al., 1976 King et al., 1981 28 (13;14)(q31;q21) + 14q21 Synthetic Ducos et al., 2007 inv(2)(p13;q12) 29 (14;15)(q28;q13) sino European Ducos et al., 2007 30 (14;15)(q29;q24) Gustavsson and Jonsson, 1992 31 (14;16)(q13;q21) French Pietrain Ducos et al., 2007 32 (X;14)(p21;q11) Singh et al., 1994 33 (Y;14)(q10;q11) French Duroc Ducos et al., 2007 34 dup(14q+) Large white Ducos et al., 2002 35 Rob(14;15)=der(14;15)(q10;q10) French landrace Ducos et al., 2007 36 Rob(14;17)=der(14;17)(q10;q10) Not indicated Ducos et al., 2007

152

Appendix V: chromosomal rearrangements reported in in vivo and fragile sites related to chromosome number 13 No Rearrangement Fragile Site Breed Reference 1 (1;13)(q27;q41) 1q27 Duroc Ducos et al., 2007 2 (3;13)(p15;q31) Large white Ducos et al., 2002 3 (4;13)(p15;q41) Not indicated Ducos et al., 2007 4 (4;13)(q25;q41) Mekinen and Remes, 1986 5 (6;13)(p13;q49) French Large white Ducos et al., 2007 6 (6;13)(p15;q41) Large white x Ducos et al., 2002 Pietrain 7 (7;13)(p13;q21) Hampolish Gustavsson et al., 1988 synthetichire 8 (7;13)(q13;q46), Danielak-Czech et al., 1997 9 (8;13)(p21;q41) 13q41 Landrace This thesis (case 8) 10 (10;13)(p13,q45) Synthetic TA Quach et al., 2009 11 (10;13)(q13;q22) Synthetic Ducos et al., 2007 12 (10;13)(q16;q21) Danielak-Czech et al., 2007 13 (11;13)(11q+;13q-) Large white Ducos et al., 2002 14 (12;13)(q11;q11) Astachova et al., 1991 15 (13;14)(13q-;14q+) Swedish yorkshire Hageltorn et al., 1976 16 (13;14)(q21;q27) 14q27 Hageltorn et al., 1976 King et al., 1981 17 (13;14)(q31;q21) + 14q21 Synthetic Ducos et al., 2007 inv(2)(p13;q12) 18 (13;15)(q31;q26) French Pietrain Ducos et al., 2007 19 (13;16)(q41;q21) Not indicated Ducos et al., 2007 20 (13;17)(q41;q11) Large white x Ducos et al., 2002 Pietrain 21 (X;13)(q24;q21) Gustavsson et al., 1989 22 Rob(13;17)=der(13;17)(q10;q10) French landrace Ducos et al., 2007 Alonso and Cautu, 1982 Miyake et al., 1977 This thesis (case 10)

Appendix VI: the comparison of our case 1 to the literature Case 1 Fragile sites 1q21 5q23 Breed Reference t(1;5)(q21;q23) 1q21 x x Landrace This thesis (case 1) t(1;5)(q21;q21) 1q21 x Danielak-Czech et al., 1994 t(1;17)(q21;q11) 1q21 x Gustavsson, 1988 inv(1)(p21;q21) 1q21 x Synthetic Ducos et al., 2002

Appendix VII: the comparison of our case 3 to the literature Case 3 Fragile sites 2p16 5p11 Breed Reference t(2;5)(p16;p11) x x Landrace This thesis (case 3) t(1;2;17)(1q11;1q17;2p16;17q13) 1q11, 1q17 x x Large White P. Cruz-Vigo el al., 2014 t(5;8)(p11;p23) x Synthetic Ducos et al., 2002 t(5;8)(p11;q11) x Large white Ducos et al., 2002 t(5,9)(p11;p24) x French landrace Ducos et al., 2007 t(5;14)(p11;q11) x English landrace x Popescu et al., 1984 Duroc

153

Appendix VIII: the comparison of our case 4 to the literature Case 4 Fragile sites 3p15 4q13 Breed Reference t(3;4)(p15;q13) x x Landrace This thesis (case 4) t(3;7)(p15;q21) x Large white Bahri et al., 1984 Gabriel-Robez et al., 1988 t(3;13)(p15;q31) x Large white Ducos et al., 2002 t(3;17)(p15;q21) x Popescu et al., 1983

Appendix IX: the comparison of our cases 5 and 9 to the literature

Case 5 & 9 Fragile 3p13 12q15 14q23 Breed Reference sites t(3;12)(p13;q15) x x Yorkshire This thesis (case 5) t(3;5)(p13;q23) x French landrace Ducos et al., 2002 t(12;14)(q15;q13) X French Pietrain Ducos et al., 2007 t(12;14)(q15;q23) x X Duroc This thesis (case 9) t(12,14)(q13;q15) 14q15 French Duroc Ducos et al., 2007 t(12;14)(q15;q13) x French Pietrain Ducos et al., 2007 t(9;14)(q14;q23) X Polish synthetic Rejduch et al., 2003 t(2;14)(p14;q23) X Hampolish Villagomez et al., synthetichire 1993

Appendix X: the comparison of our cases 2, 6 and 7 to the literature

Case 2, 6 & 7 Fragile sites 1q211 6p15 7q13 15q13 Breed Reference t(1;15)(q211;q13) 1q211 x x Duroc This thesis (case 2) t(1;15)(p25;q13) 1p25 x Koukkanen and Makinen, 1988 t(1;15)(q17;q22) 1q17 Synthetic Ducos et al., 2007 t(1;15)(q26;qter) 1q26 Popescu et al., 1988 t(7;15)(q13;q13) x x Duroc This thesis (case 7) t(7;15)(q25;q25) Popescu et al., 1983 t(7;15)(q24;q12) Large White Konfortovaet al., 1995 t(6;7)(p15;q13) 6p15 X x Duroc This thesis (case 6) t(1;7)(q17;q13) 1q17 x Not indicated Ducos et al., 2007 t(1;7)(q17;q13) 1q17 x Not indicated Ducos et al., 2007 t(3;15)(q27;q13) x Large white K. Massip et al., 2008 t(6;13)(p15;q41) X Large white x Ducos et al., 2002 Pietrain t(6;15)(p15;q12) X Bonneau et al., 1991 t(7;10)(q13;q11) x Not indicated Ducos et al., 2007 t(7;13)(q13;q46), x Danielak-Czech et al., 1997 t(9;15)(p24;q13) x Large white x Ducos et al., 2002 french landrace t(10;18)(p11;q24) + X French Pietrain Ducos et al., 2007 (6;8)(p15;q27) t(11;15) (p15;q13) x Swedish Henricson and Landrace Backstrom, 1964 Hageltorn et al., 1973 t(14;15)(q28;q13) x sino European Ducos et al., 2007 t(15;17)(q13;q21) x Large white Ducos et al., 2002

154

Appendix XI: the comparison of our case 8 to the literature

Case 8 Fragile sites 8p21 13q41 Breed Reference t(8;13)(p21;q41) 13q41 x x Landrace This thesis (case 8) t(3;8)(q25;p21) x Synthetic Ducos et al., 2007 t(8;14)(p21;q25) 14q25 x Danielak-Czech et al., 1997 Inv(8)(p21;q11) x Pietrain Ducos et al., 2007 t(1;13)(q27;q41) 1q27 x Duroc Ducos et al., 2007 t(4;13)(p15;q41) x Not indicated Ducos et al., 2007 t(4;13)(q25;q41) x Mekinen and Remes, 1986 t(6;13)(p15;q41) x Large white x Pietrain Ducos et al., 2002 t(13;16)(q41;q21) x Not indicated Ducos et al., 2007 t(13;17)(q41;q11) x Large white x Pietrain Ducos et al., 2002

Appendix XII: the comparison of our case 11 to the literature

Case 11 Fragile sites 8q11 8q25 Breed Reference Inv(8)(q11;q25) x x Duroc This thesis (case 11) Inv(8)(p21;q11) x Pietrain Ducos et al., 2007 t(5;8)(p11;q11) x Large white Ducos et al., 2002 Inv(8)(p11;q25) x Large white Ducos et al., 2007

Appendix XIII: the comparison of our case 12 to the literature

Case 12 Fragile sites 6q13 6q35 Breed Reference Inv(6)(q13;q35) x x Landrace This thesis (case 12) Inv(6)(p14;q12) Sino European Ducos et al., 2007 t(1;6) (p13;q35) 1p13 x Large white Locniskar et al., 1976 t(1;6)(q17;q35) 1q17 x Synthetic Ducos et al., 2002

155

Appendix XIV: Details of the identified CNVRs

Length No. of # CNVR ID Chr* Start End State FERT* Reproduction QTLs Gene Acc # Gene name (bp) Samples 1 CNVR1HL 1 67,164,859 70,920,152 3,755,294 GAIN/DEL H&L 2 TNUM QTL #5223 GSTM2 NM_001078684 TNUM QTL #5255,TNUM QTL 2 CNVR2L 1 170,095,561 173,141,931 3,046,371 DEL L 2 A NM_213799 #5223,TNUM QTL #6481 TNUM QTL #822,TNUM QTL #845,TNUM QTL #5255,TNUM QTL 3 CNVR5L 1 294,847,613 297,384,003 2,536,391 DEL L 2 #5223,TESTIWT QTL #6527,AGEP QTL #8815,TNUM QTL #1250 NM_001243810, NM_001243865, NM_001243838, NM_001098605, CHID1, CD151, NM_001244866, EFCAB4A, PNPLA2, NM_213866, RPLP2, CEND1, NM_001244935, 4 CNVR7L 2 16,466 3,571,261 3,554,796 GAIN L 2 NSB QTL #18095,NSB QTL #18094 TALDO1, TMEM80, NM_001244431, IRF7, PHLDA2, NM_001097428, NAP1L4, FADD, NM_001174057, CPT1A, GAL NM_001285976, NM_001031797, NM_001129805, NM_214234 NM_001204770, 5 CNVR8H 2 7,190,712 7,633,891 443,180 DEL H 2 NSB QTL #18095,NSB QTL #18094 MACROD1, OTUB1 NM_001162403 NSB QTL #18094,NSB QTL 6 CNVR10L 2 13,448,974 13,601,640 152,667 DEL L 2 #18095,NSB QTL #18096,NSB QTL #18097,TNUM QTL #909 NSB QTL #18094,NSB QTL 7 CNVR11L 2 23,647,220 25,134,932 1,487,713 DEL L 2 #18095,NSB QTL #18096,NSB QTL #18097,TNUM QTL #909 NM_214233, 8 CNVR13L 2 106,268,518 107,324,961 1,056,443 DEL L 2 GEST QTL #21837 GLRX, CAST NM_214067 9 CNVR15H 2 130,991,927 131,016,807 24,880 DEL H 7 - NM_001244817, AFAP1L1, GRPEL2, NM_001244870, 10 CNVR16L 2 156,677,791 159,938,857 3,261,066 GAIN L 2 - LOC100514340, NM_001244954, MIR143, MIR145 NR_038529, NR_038484

156

NM_001078668, NM_001246214, MGAT1, IFITM2, NM_001201382, 11 CNVR17L 2 159,938,857 162,298,136 2,359,280 DEL L 2 - IFITM3, COX8H, NM_001097500, SIRT3, ODF3 NM_001110057, NM_001193448 TNUM QTL #5224,OVRATE QTL 12 CNVR18L 3 48,316,588 48,324,106 7,519 DEL L 5 #515,NNIP QTL #7455,NNIP QTL #7472 TNUM QTL #5224,TNUM QTL 13 CNVR19H 3 122,043,048 122,082,701 39,654 DEL H 2 #8797,TNUM QTL #8798 NSB QTL #18129,TLWT QTL 14 CNVR27L 6 18,410,765 18,418,236 7,472 DEL L 5 #22946,TLWT QTL #22947 TNB QTL #10620,TNUM QTL 15 CNVR28H 6 75,455,400 75,464,703 9,304 DEL H 7 #5226,NNIP QTL #7459,NNIP QTL #7476,TNUM QTL #8665 TNUM QTL #4253,OVRATE QTL #492,NNIP QTL #7477,AGEP QTL NM_001244395, 16 CNVR34L 8 79,177,886 80,427,348 1,249,463 DEL L 5 SFRP2, TLR2 #589,UC QTL #523,TNUM QTL NM_213761 #1100 AGEP QTL #22087,AGEP QTL #22088,AGEP QTL #22089,AGEP NM_214207, 17 CNVR36H 9 36,855,330 37,305,674 450,345 DEL H 7 QTL #22090,AGEP QTL MMP7, MMP25 NM_213905 #22091,AGEP QTL #22092,NNIP QTL #7462,AGEP QTL #22086 18 CNVR37H 9 81,691,177 81,750,330 59,153 DEL H 7 OVRATE QTL #517 SGCE NM_001144124 19 CNVR38L 9 111,425,418 111,714,858 289,441 DEL L 5 OVRATE QTL #517 20 CNVR39L 9 143,837,762 143,878,023 40,261 DEL L 5 GEST QTL #452 21 CNVR41L 10 6,048,737 6,093,590 44,854 DEL L 2 NSB QTL #18122,NSB QTL #18123 22 CNVR42H 10 55,903,766 56,046,451 142,686 DEL H 2 NSB QTL #18122,NSB QTL #18123 FSH QTL #646,TNUM QTL 23 CNVR43H 10 60,051,402 61,491,332 1,439,931 DEL H 2 #1107,TNUM QTL #2928,OVRATE ITGB1 NM_213968 QTL #518,TNUM QTL #5258 FSH QTL #646,TNUM QTL 24 CNVR44H 10 64,593,871 64,711,288 117,418 DEL H 2 #1107,TNUM QTL #2928,OVRATE QTL #518 TNUM QTL #5260,AGEP QTL 25 CNVR45L 11 39,122,418 40,052,374 929,957 DEL L 5 #22093,TNUM QTL #586,AGEP QTL #22094 26 CNVR48H 11 75,996,131 76,165,143 169,013 GAIN H 2 NSB QTL #7534,NNIP QTL #7478

157

REMPAR QTL #15998,REMPAR QTL #15999,LNPDR QTL #16006,NBA QTL #18068,NBA QTL #18075,NBA QTL #18091,TNB QTL #18197,TNB 27 CNVR49H 12 31,229,775 31,975,983 746,209 DEL H 2 STAT2 NM_213889 QTL #18230,TNUM QTL #5261,TNUM QTL #5227,RTNUM QTL #6479,TNUM QTL #2929,TNUM QTL #595,LTNUM QTL #6472 REMPAR QTL #15998,REMPAR QTL #15999,LNPDR QTL #16006,NBA QTL #18068,NBA QTL #18075,NBA NM_001243919, QTL #18091,TNB QTL #18197,TNB NM_001243813, CUEDC1, MRPS23, QTL #18230,AGEP QTL NM_001038007, 28 CNVR50L 12 34,975,703 37,778,877 2,803,175 DEL L 5 SRSF1, MIR142, #22095,TNUM QTL #5261,TNUM NR_038555, CLTC, MIR21 QTL #5227,AGEP QTL #22096,AGEP NM_001146127, QTL #22097,RTNUM QTL NR_038508 #6479,TNUM QTL #2929,TNUM QTL #595,LTNUM QTL #6472 NM_001105309, UBB, SLC5A10, NM_001012297, 29 CNVR51L 12 61,949,320 63,499,875 1,550,556 GAIN L 2 TNUM QTL #5261,TNUM QTL #5227 COPS3, PEMT NM_001244149, NM_214365 30 CNVR56L 13 142,228,823 142,572,923 344,101 DEL L 2 NNIP QTL #7479,NNIP QTL #7466 NR_035399, MIR202, CYP2E1, 31 CNVR59HL 14 151,808,821 153,786,811 1,977,991 GAIN/DEL H&L 2 - NM_214421, ECHS1 NM_001190175 32 CNVR62L 16 84,891,022 86,875,405 1,984,384 DEL L 2 - TERT NM_001244300 TNB QTL #22920,TLWT QTL #22965,TNUM QTL #5229,AGEP QTL 33 CNVR63H 17 40,116,835 40,197,983 81,149 DEL H 2 ID1 NM_001244700 #22122,AGEP QTL #22123,AGEP QTL #22124,AGEP QTL #22125 TNB QTL #22920,TLWT QTL #22965,TNUM QTL #5229,AGEP QTL 34 CNVR64H 17 41,544,230 41,558,361 14,132 GAIN H 2 #22122,AGEP QTL #22123,AGEP QTL #22124,AGEP QTL #22125 NBA QTL #7538,AGEP QTL #21988,NNIP QTL #7470,AGEP QTL 35 CNVR65H 18 35,903,061 36,257,473 354,413 DEL H 2 IFRD1 NM_001007519 #22127,AGEP QTL #22128,AGEP QTL #22126 *Chr = Chromosome; Fert.= Fertility

158

Appendix XV: Numbers and length of CNVRs are represented by chromosome HIGH-Fertile HIGH & LOW LOW-Fertile Total Total Chromosome #CNVR #CNVR #CNVR #CNVR length 1 1 2 3 3,755,294 5,582,762 9,338,056 2 2 468,060 6 11,871,965 8 12,340,025 3 1 1 2 39,654 7,519 47,173 6 1 1 2 9,304 7,472 16,776 8 1 1,249,463 1 1,249,463

9 2 2 4 509,498 329,702 839,200 10 3 1 4 1,700,035 44,854 1,744,889 11 1 1 2 169,013 929,957 1,098,970 12 1 746,209 2 4,353,731 3 5,099,940 13 1 1 344,101 344,101 14 1 1 1,977,991 1,977,991 16 1 1 1,984,384 1,984,384 17 2 2 95,281 95,281 18 1 1 354,413 354,413 Total 14 4,091,467 2 5,733,285 19 26,705,910 35 36,530,662

159

Appendix XVI: Summary of CNVRs according to the copy number states. The number and length of CNVRs identified as Gain, Deletion or Gain/Del are given.

GAIN DEL GAIN/DEL Chromosome #CNVR Length #CNVR Length #CNVR Length Total #CNVR Total length % CNVR/Chromosome 1 5,582,762 2 1 3,755,294 3 9,338,056 2.96 2 2 5,524,163 6 6,815,862 8 12,340,025 7.59 3 47,173 2 2 47,173 0.03 6 16,776 2 2 16,776 0.01 8 1,249,463 1 1 1,249,463 0.84 9 839,200 4 4 839,200 0.55 10 1,744,889 4 4 1,744,889 2.21 11 1 929,957 1 169,013 2 1,098,970 1.25 12 1 3,549,384 2 1,550,556 3 5,099,940 8.02 13 344,101 1 1 344,101 0.16 14 1 1,977,991 1 1,977,991 1.29 16 1,984,384 1 1 1,984,384 2.28 17 1 81,149 1 14,132 2 95,281 0.14 18 354,413 1 1 354,413 0.58 Total Length 5 22,247,814 28 8,549,563 2 5,733,285 35 36,530,662

160

Appendix XVII: Summary of the number of reproduction QTLs overlapping with CNVRs.

QTL High Fertile High & Low Fertile Low Fertile Total AGEP 19 7 26 FSH 2 2 GEST 2 2 LNPDR 1 1 2 LTNUM 1 1 2 NBA 4 3 7 NNIP 5 5 10 NSB 5 13 18 OVRATE 3 3 6 REMPAR 2 2 4 RTNUM 1 1 2 TESTIWT 1 1 TLWT 2 2 4 TNB 5 2 7 TNUM 16 1 21 38 UC 1 1 No QTL 1 1 3 5 Total 67 2 68 137

Appendix XVIII: Summary of gene content (RefSeq genes) within CNVRs specific for high fertility, low fertility animals or present in both.

Chromosome GAIN DEL GAIN/DEL Total 1 1 1 2

2 17 10 27 8 2 2

9 3 3

10 1 1

12 4 6 10 14 2 2

16 1 1

17 1 1

18 1 1

Total 21 26 3 50

161

Appendix XIX: Primer squences used for the qPCR validation experiments

Name Sequence 5'-3' Product size (bp) BACT-F* TCGATCATGAAGTGCGACGTG 113 BACT-R* GTGATCTCCTTCTGCATCCTGTC CNVR15H-F TTCTCTCTGCCCCAAACACT 100 CNVR15H-R TCCAAAAGTTTGCAGAAAGGA CNVR18L-F TCTGCAACTGCAGCTCTGAT 114 CNVR18L-R TGGAAGGAATTACTGCCCTG CNVR27L-F CAGCTTGCAGCAGCTTGAT 107 CNVR27L-R GTTCCCTGGTGGTGTAGTGG CNVR28H-F TTCACGATGGCATCTTTCTG 197 CNVR28H-R AGGGCCTCAGAGGGTTTCTA CNVR34L-F CAGAGAAGAAGCCTGATGGG 101 CNVR34L-R CAACTCTCTGGACCATGCC CNVR36L-F TTCCACGGCAGCATCTATTTG 168 CNVR36L-R GAATAGCACATTGTCCTCTACAAGC CNVR37L-F TCCCACCAACGACATACCTG 126 CNVR37L-R TCCAAGGTGCACTCTGATCG CNVR38L-F TTTTGCATGGAGCTGGAAAT 167 CNVR38L-R TGCTGCCTAGCAAACAACTG * from Chen et al. BMC Genomics 2012, 13:733.

162

Appendix XX. TSPY CN at three exons by qPCR in 20 boars. At each exon calculations were made using two different primer efficiencies, as determined from either STD curve or as the average of individual LinRegPCR efficiencies.

TSPY-E1 STD TSPY-E1 LinReg TSPY-E3 STD TSPY-E3 LinReg TSPY-E5 STD TSPY-E5 LinReg DA Cali 5.27 2.16 4.33 2.33 4.47 3.17 DB 4.58 2.31 4.06 2.18 4.27 3.03 DC 6.06 1.89 4.21 2.25 4.56 3.23 DD 5.68 2.26 3.81 2.03 4.15 2.94 DE 4.67 2.69 4.15 2.21 4.68 3.32 LA 4.11 1.46 4.45 2.38 4.21 2.98 LB 3.82 0.88 3.00 1.59 2.84 2.00 LC 3.94 1.31 3.94 2.11 3.70 2.62 LD 5.44 1.23 3.69 1.97 3.19 2.25 LE 8.02 1.11 3.77 2.02 3.43 2.42 PA 4.41 2.02 3.41 1.80 3.67 2.57 PB 4.13 2.16 4.12 2.20 4.42 3.13 PC 6.15 2.26 3.90 2.08 4.19 2.96 PD 4.83 2.28 4.09 2.19 4.36 3.09 PE 3.95 1.90 3.31 1.76 3.35 2.36 YA 4.52 0.78 3.62 1.93 3.27 2.31 YB 4.74 1.06 3.37 1.79 3.27 2.31 YC 4.84 1.02 4.27 2.29 3.53 2.49 YD 4.73 1.68 4.32 2.32 3.95 2.80 YE 3.97 0.79 3.23 1.72 2.57 1.81 Duroc 5.25 2.26 4.11 2.20 4.42 3.13 Landrace 5.07 1.20 3.77 2.01 3.47 2.45 Pietrain 4.69 2.12 3.77 2.01 4.00 2.82 Yorkshire 4.56 1.07 3.76 2.01 3.32 2.34 All 4.89 1.66 3.85 2.06 3.80 2.69

SD Duroc 0.64 0.29 0.19 0.11 0.21 0.15 SD 1.78 0.22 0.52 0.29 0.52 0.37 Landrace SD Pietrain 0.88 0.16 0.38 0.21 0.47 0.34 SD 0.35 0.37 0.51 0.28 0.50 0.36 Yorkshire SD All 1.01 0.60 0.42 0.23 0.61 0.43

163

Appendix XXI: TSPY ddPCR CN ratios at three exons in 20 normal, 5 Low- and 5 High- fertility boars. Direct Boar Effect (DBE) values are included where available.

TSPY-E1 TSPY-E3 TSPY-E5 breed fertility DBE DA Cali 3.11 3.20 3.11 Duroc normal N/A DB 2.93 3.04 2.94 Duroc normal N/A DC 2.94 3.03 2.89 Duroc normal N/A DD 2.94 3.09 3.10 Duroc normal N/A DE 2.99 3.01 2.92 Duroc normal N/A LA 2.96 3.07 2.82 Landrace normal N/A LB 2.97 3.02 2.89 Landrace normal N/A LC 2.90 3.00 2.96 Landrace normal N/A LD 2.90 3.06 2.89 Landrace normal N/A LE 2.94 3.06 3.11 Landrace normal N/A PA 3.03 3.12 3.02 Pietrain normal N/A PB 2.92 3.04 2.87 Pietrain normal N/A PC 3.05 3.06 2.94 Pietrain normal N/A PD 3.13 3.20 3.04 Pietrain normal N/A PE 2.93 2.99 2.97 Pietrain normal N/A YA 3.05 3.10 3.18 Yorkshire normal N/A YB 2.99 2.98 2.94 Yorkshire normal N/A YC 2.89 2.95 2.94 Yorkshire normal N/A YD 3.17 3.21 3.11 Yorkshire normal N/A YE 3.01 3.06 2.96 Yorkshire normal N/A H1 2.42 2.53 2.42 Landrace high 2.25 H2 2.48 2.58 2.48 Landrace high 2.72 H3 2.80 3.25 2.99 Yorkshire high 3.28 H4 2.65 2.85 2.77 Yorkshire high 3.79 H5 2.44 2.57 2.46 Landrace high 3.11 L1 2.47 2.59 2.41 Yorkshire low -2.27 L2 2.64 2.81 2.69 Landrace low -2.31 L3 2.33 2.78 2.64 Yorkshire low -2.33 L4 2.73 3.00 2.98 Landrace low -2.78 L5 2.66 3.08 2.76 Landrace low -2.41 Duroc 2.98 3.07 2.99

Landrace 2.93 3.04 2.94

Pietrain 3.01 3.08 2.97

Yorkshire 3.02 3.06 3.03

All 2.99 3.06 2.98

AVE-High 2.56 2.76 2.62

AVE-Low 2.57 2.85 2.70

164

SD Duroc 0.07 0.08 0.10

SD Landrace 0.03 0.03 0.11

SD Pietrain 0.09 0.08 0.07

SD Yorkshire 0.10 0.10 0.11

SD All 0.08 0.07 0.10

SD High 0.16 0.29 0.20

SD Low 0.16 0.19 0.21

165

Appendix XXII: Primer and probe sequences, annealing temperature, product sizes and primer efficiency values determined from either STD curve or as the average of individual LinRegPCR efficiencies. Gene name GenBank # Primer name Primer and probe sequence (5'-3') Product (bp) Efficiency STD curve Efficiency LinReg

pAR-F TGTGCAGTTCACTCCTGAAGA Androgen receptor NM_214314.2 pAR-R GGAGAACATGGTCCCTGGTA 115 2.01 2.03 (AR) pAR-TM /FAM/TCTGCCAAC/ZEN/TACTGACATGTGTTTCATTGT/IABkFQ/ pTSPYE1-F TAGGCCTCAGTCGGTAGTTG

TSPY exon 1 pTSPYE1-R TGCTCCTCCTCCTCAAACTC 198 1.89* 1.82*

pTSPYE1-TM /FAM/TGCTGCTGA/ZEN/TGGAGGACGTCGT/IABkFQ/

pTSPYE3-F CTGATTGCTGCAAAATCATGT Gene ID: TSPY exon 3 pTSPYE3-R CCCTCCTTCTCACAAACCAT 125 2.02 2.09 100625034 pTSPYE3-TM /FAM/TGTTAACCT/ZEN/CAATGGTAAGAGGAGGCTCC/IABkFQ/

pTSPYE5-F CAACCCGCTGCACTATTACC

TSPY exon 5 pTSPYE5-R GACTCGGTTTCTCCATGAGC 132 2.02 2.07

pTSPYE5-TM /FAM/AGTTCCAGG/ZEN/GAGGAGGACAGGTGA/IABkFQ/ The TaqMan probes were labeled with 5’-FAM, internal ZEN and 3’-Iowa Black FQ quencher molecules (IDT). Primer efficiencies, as determined from either STD curve or as the average of individual LinRegPCR efficiencies. *different (p<0.01) from unmarked according to Kruskal-Wallis test