Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Factors affecting DNA methylation in the sperm of men with male factor infertility Ng, Richard 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2018_february_ng_richard.pdf [ 2.69MB ]
Metadata
JSON: 24-1.0357954.json
JSON-LD: 24-1.0357954-ld.json
RDF/XML (Pretty): 24-1.0357954-rdf.xml
RDF/JSON: 24-1.0357954-rdf.json
Turtle: 24-1.0357954-turtle.txt
N-Triples: 24-1.0357954-rdf-ntriples.txt
Original Record: 24-1.0357954-source.json
Full Text
24-1.0357954-fulltext.txt
Citation
24-1.0357954.ris

Full Text

   FACTORS AFFECTING DNA METHYLATION IN THE SPERM OF MEN WITH MALE FACTOR INFERTILITY by Richard Ng B.Sc., The University of British Columbia, 2012  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Reproductive and Developmental Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) November 2017 © Richard Ng, 2017  ii   ABSTRACT Infertility currently affects an estimated 8-12% of couples globally and the male factor is suspected as the primary cause in about half of these cases. There are several identified genetic causes for male infertility such as Y chromosome microdeletions and chromosomal rearrangements; however, these known genetic causes only account for 15-30% of male infertility cases, with the majority being idiopathic. Recently, there has been a shift toward studying epigenetic mechanisms as a potential avenue to explain idiopathic cases of male infertility. Most commonly, DNA methylation of imprinted genes is studied as these genes are not erased from gamete to embryo and the function of these genes is critical for embryonic development. Furthermore, methylation of DNA is tightly linked to the folate cycle as the metabolism of folate generates the methyl donors used in methylation reactions. There is also evidence to suggest DNA damage may affect methylation by producing an inadequate substrate for methylation enzymes. We investigated DNA methylation, the folate cycle, and DNA damage in men with oligospermia and azoospermia and compared these to fertile control men. Similar to previously published studies, we found that the sperm of infertile men carries methylation defects in imprinted genes. To further the analysis, we investigated if any relationship with folate enzyme genotypes or folate concentration exists. The genotypes of folate enzymes were not different between groups and there were no effects on DNA methylation. We observed that the concentration of folate was positively correlated with GTL2 methylation, but no correlations were found for any other regions. Finally, GST genotypes and DNA fragmentation were examined for possible associations with DNA methylation. The frequency GST genotypes were iii  not different between groups and had no effect on DNA methylation. DNA fragmentation was also determined to have no associations with DNA methylation. We provide evidence that DNA methylation at imprinted genes in the sperm of infertile men may be affected by folate concentration. The modest association found here can be supplemented in future studies by expanding the coverage of DNA methylation to other regions, investigating homocysteine and methionine, or adding a measure of ROS.   iv   LAY ABSTRACT Infertility is becoming a concern for a growing number of couples. Since the cause of a majority of male infertility cases is unknown, there have been more research studies looking at factors that affect the expression of genes rather than the genes or DNA themselves. In this study, we investigate DNA methylation as the modulator of gene expression in the sperm of infertile men. Additionally, we examine the folate cycle and DNA damage as these factors are known to affect DNA methylation. We determined that our cohort of infertile men have abnormal DNA methylation profiles when compared with fertile control men. We observed that folate concentration was positively correlated with DNA methylation at only one gene, but DNA fragmentation was not associated with DNA methylation. Although we present few significant results, the study framework allows for future studies to build a comprehensive analysis of DNA methylation and related pathways.   v   PREFACE Ethical approval for all studies presented was obtained from the University of British Columbia prior to initiation (H06-03547). The studies were funded by Canadian Institutes of Health Research grants to Dr. Sai Ma.  The experiments presented below were conceived by Dr. Sai Ma. Oligospermic patients and testicular biopsies for our study were provided and performed by Dr. Victor Chow and Dr. Kenneth Poon, from the Department of Urologic Sciences at the University of British Columbia. With the exception of the 20 fertile control men, the data collection was performed by Richard Ng. Data for the 20 fertile control men was collected by Sanuja Pitigalaarachchige, including pyrosequencing, folate enzyme genotypes, DNA fragmentation, and GST enzyme genotypes. Peripheral blood for the analysis of folate concentration was prepared by Richard Ng, however, folate concentrations were determined by Roger Dyer at the BC Children’s Hospital Research Institute. Statistical analysis of the data and all figures and tables were performed and prepared by Richard Ng.   vi   TABLE OF CONTENTS ABSTRACT .................................................................................................................................... ii LAY ABSTRACT ......................................................................................................................... iv PREFACE ....................................................................................................................................... v TABLE OF CONTENTS ............................................................................................................... vi LIST OF TABLES ......................................................................................................................... xi LIST OF FIGURES ...................................................................................................................... xii LIST OF ABBREVIATIONS ...................................................................................................... xiv ACKNOWLEDGEMENTS ......................................................................................................... xix CHAPTER 1: INTRODUCTION ................................................................................................... 1 1.1 Development .................................................................................................................... 1 1.1.1 Induction of Primordial Germ Cells ......................................................................... 1 1.1.2 Sex Differentiation .................................................................................................... 4 1.2 Spermatogenesis ............................................................................................................... 5 1.2.1 Spermatogonia .......................................................................................................... 5 1.2.2 Meiosis I.................................................................................................................... 9 1.2.3 Meiosis II ................................................................................................................ 12 1.2.4 Spermatid Remodelling .......................................................................................... 13 1.3 Regulation ...................................................................................................................... 16 1.4 Male Infertility ............................................................................................................... 19 1.4.1 Assisted Reproductive Technologies ...................................................................... 19 1.4.2 Phenotypes of Male Infertility ................................................................................ 21 1.4.3 Genetic Causes of Male Infertility .......................................................................... 22 vii  1.4.3.1 Y Chromosome Microdeletions....................................................................... 22 1.4.3.2 Chromosomal Abnormalities ........................................................................... 25 1.4.3.3 Cystic Fibrosis Gene Mutations ...................................................................... 28 1.4.4 Epigenetics of Male Infertility ................................................................................ 29 1.4.4.1 DNA Methylation ............................................................................................ 29 1.4.4.1.1 Genomic Imprinting .................................................................................... 32 1.4.4.1.2 Epigenetic Inheritance ................................................................................. 36 1.4.4.2 Histone modifications ...................................................................................... 38 1.4.4.3 Non-coding RNAs ........................................................................................... 39 1.5 One-Carbon Metabolism ................................................................................................ 40 1.6 DNA Integrity ................................................................................................................ 43 1.7 Rationale ......................................................................................................................... 46 1.7.1 Hypotheses and Objectives ..................................................................................... 48 CHAPTER 2: INVESTIGATION OF DNA METHYLATION IN INFERTILE MEN .............. 50 2.1 Introduction .................................................................................................................... 50 2.2 Materials and Methods ................................................................................................... 51 2.2.1 Sample Collection ................................................................................................... 51 2.2.2 Sperm Cell Isolation ............................................................................................... 52 2.2.3 Cell Lysis ................................................................................................................ 54 2.2.4 Bisulfite Conversion and DNA Purification ........................................................... 54 2.2.5 PCR Amplification.................................................................................................. 55 2.2.6 Pyrosequencing ....................................................................................................... 56 2.2.7 Data Analysis .......................................................................................................... 57 viii  2.3 Results ............................................................................................................................ 58 2.3.1 H19 Gene ................................................................................................................ 58 2.3.2 GTL2 Gene .............................................................................................................. 60 2.3.3 MEST Gene ............................................................................................................. 62 2.3.4 LIT1 Gene ............................................................................................................... 64 2.3.5 LINE1 Transposable Element ................................................................................. 66 2.4 Discussion ...................................................................................................................... 68 CHAPTER 3: INVESTIGATION OF THE FOLATE CYCLE IN INFERTILE MEN ............... 72 3.1 Introduction .................................................................................................................... 72 3.2 Materials and Methods ................................................................................................... 73 3.2.1 Sample Collection ................................................................................................... 73 3.2.2 Folate Analysis........................................................................................................ 73 3.2.3 Genotyping .............................................................................................................. 74 3.2.3.1 DNA Purification ............................................................................................. 74 3.2.3.2 PCR Amplification .......................................................................................... 74 3.2.3.3 Restriction Fragment Length Polymorphism .................................................. 75 3.2.4 Data Analysis .......................................................................................................... 76 3.3 Results ............................................................................................................................ 77 3.3.1 Folate Concentration ............................................................................................... 77 3.3.2 Correlation of Folate Concentration and DNA Methylation .................................. 78 3.3.3 Folate Enzyme Genotypes ...................................................................................... 81 3.3.4 Effect of Folate Genotypes on DNA Methylation .................................................. 83 3.4 Discussion ...................................................................................................................... 86 ix  CHAPTER 4: INVESTIGATION OF SPERM DNA INTEGRITY IN INFERTILE MEN ........ 90 4.1 Introduction .................................................................................................................... 90 4.2 Materials and Methods ................................................................................................... 92 4.2.1 Sample Collection ................................................................................................... 92 4.2.2 DNA Fragmentation ................................................................................................ 92 4.2.2.1 Sample Preparation .......................................................................................... 92 4.2.2.2 Slide Preparation.............................................................................................. 92 4.2.2.3 TUNEL Assay ................................................................................................. 93 4.2.3 Genotyping .............................................................................................................. 94 4.2.3.1 DNA Purification ............................................................................................. 94 4.2.3.2 PCR Amplification .......................................................................................... 94 4.2.3.3 Agarose Gel Electrophoresis ........................................................................... 95 4.2.4 Data Analysis .......................................................................................................... 95 4.3 Results ............................................................................................................................ 95 4.3.1 DNA Fragmentation ................................................................................................ 95 4.3.2 Correlation of DNA Fragmentation and DNA Methylation ................................... 96 4.3.3 Genotypes of GST enzymes .................................................................................... 99 4.3.4 Effect of GST Genotypes on DNA Methylation ................................................... 101 4.4 Discussion .................................................................................................................... 104 CHAPTER 5: CONCLUSION ................................................................................................... 106 5.1 Summary ...................................................................................................................... 106 5.2 Limitations and Future Directions ................................................................................ 107 5.3 Significance .................................................................................................................. 108 x  BIBLIOGRAPHY ....................................................................................................................... 109    xi   LIST OF TABLES Table 1.1 WHO lower reference limits for common semen parameters. ..................................... 21 Table 2.1 Primer sequences used in methylation analysis. ........................................................... 56 Table 2.2 Summary of H19 methylation. ...................................................................................... 59 Table 2.3 Summary of GTL2 methylation. ................................................................................... 61 Table 2.4 Summary of MEST methylation. .................................................................................. 63 Table 2.5 Summary of LIT1 methylation. ..................................................................................... 65 Table 2.6 Summary of LINE1 methylation. .................................................................................. 67 Table 3.1 Primer sequences used in genotyping folate cycle genes. ............................................ 75 Table 3.2 Restriction enzymes used in RFLP genotyping. ........................................................... 76 Table 3.3 Expected sizes of bands in folate cycle genes. ............................................................. 77 Table 4.1 Primer sequences used in GST genotyping. .................................................................. 94   xii   LIST OF FIGURES Figure 1.1 Timeline of PGC development. ..................................................................................... 3 Figure 1.2 Gene signalling network for PGC specification. ........................................................... 3 Figure 1.3 Migration of spermatozoa through the seminiferous tubules during spermatogenesis. 7 Figure 1.4 Overview of spermatogenesis........................................................................................ 8 Figure 1.5 The meiotic cell cycle. ................................................................................................. 11 Figure 1.6 Major morphological features that appear during spermatid maturation. ................... 14 Figure 1.7 Feedback regulation of the HPG axis. ......................................................................... 18 Figure 1.8 The Y chromosome. .................................................................................................... 23 Figure 1.9 Products of non-disjunction during meiosis. ............................................................... 27 Figure 1.10 Mechanisms of genomic imprinting. ......................................................................... 33 Figure 1.11 Timeline of DNA methylation reprogramming. ........................................................ 36 Figure 1.12 Metabolic pathways of one-carbon metabolism. ....................................................... 42 Figure 2.1 Mean DNA methylation of H19 by pyrosequencing. .................................................. 60 Figure 2.2 Mean DNA methylation of GTL2 by pyrosequencing. ............................................... 62 Figure 2.3 Mean DNA methylation of MEST by pyrosequencing................................................ 64 Figure 2.4 Mean DNA methylation of LIT1 by pyrosequencing. ................................................. 66 Figure 2.5 Mean DNA methylation of LINE1 by pyrosequencing. .............................................. 68 Figure 3.1 Whole blood folate concentration in infertile men. ..................................................... 78 Figure 3.2 Correlation between whole blood folate and mean H19 methylation. ........................ 79 Figure 3.3 Correlation between whole blood folate and mean GTL2 methylation. ...................... 79 Figure 3.4 Correlation between whole blood folate and mean MEST methylation. ..................... 80 Figure 3.5 Correlation between whole blood folate and mean LIT1 methylation. ....................... 80 xiii  Figure 3.6 Correlation between whole blood folate and mean LINE1 methylation. .................... 81 Figure 3.7 Frequency of MTHFR genotypes. ............................................................................... 82 Figure 3.8 Frequency of MS genotypes. ....................................................................................... 82 Figure 3.9 Frequency of MTRR genotypes. .................................................................................. 83 Figure 3.10 DNA methylation of H19 separated by folate enzyme genotype. ............................. 84 Figure 3.11 DNA methylation of GTL2 separated by folate enzyme genotype. .......................... 84 Figure 3.12 DNA methylation of MEST separated by folate enzyme genotype. .......................... 85 Figure 3.13 DNA methylation of LIT1 separated by folate enzyme genotype. ............................ 85 Figure 3.14 DNA methylation of LINE1 separated by folate enzyme genotype. ......................... 86 Figure 4.1 DNA fragmentation in infertile men. .......................................................................... 96 Figure 4.2 Correlation between DNA fragmentation and mean H19 methylation. ...................... 97 Figure 4.3 Correlation between DNA fragmentation and mean GTL2 methylation. .................... 97 Figure 4.4 Correlation between DNA fragmentation and mean MEST methylation. ................... 98 Figure 4.5 Correlation between DNA fragmentation and mean LIT1 methylation. ..................... 98 Figure 4.6 Correlation between DNA fragmentation and mean LINE1 methylation. .................. 99 Figure 4.7 Frequency of GSTT1 genotypes. ............................................................................... 100 Figure 4.8 Frequency of GSTM1 genotypes. .............................................................................. 100 Figure 4.9 DNA methylation of H19 separated by GST genotype. ............................................ 101 Figure 4.10 DNA methylation of GTL2 separated by GST genotype. ........................................ 102 Figure 4.11 DNA methylation of MEST separated by GST genotype. ....................................... 102 Figure 4.12 DNA methylation of LIT1 separated by GST genotype. ......................................... 103 Figure 4.13 DNA methylation of LINE1 separated by GST genotype. ...................................... 103   xiv   LIST OF ABBREVIATIONS 5-caC    5-carboxylcytosine 5-fC     5-formylcytosine 5-hmC    5-hydroxymethylcytosine 5-mC    5-methylcytosine 5-mTHF   5-methylTHF 5,10-mTHF   5,10-methyleneTHF 8-OHdG   8-hydroxy-deoxyguanine αKB    α-ketobutyrate AID    Activation-induced deaminase APOBEC    Apolipoprotein B RNA-editing catalytic component ALB    Alkaline lysis buffer AMH    Anti-Müllerian hormone ANOVA   Analysis of variance ART    Assisted reproductive technology AS    Angelman syndrome ATP    Adenosine triphosphate AZF    Azoospermia factor BER    Base excision repair BMP    Bone morphogenetic protein BTB    Blood-testis barrier BWS    Beckwith-Wiedemann syndrome C    Control CBAVD   Congenital bilateral absence of vas deferens CBS    Cystathionine synthase CF    Cystic fibrosis CFTR    Cystic fibrosis transmembrane conductance regulator CGL    Cystathionine lyase CTCF    CCCTC binding factor xv  DAZ    Deleted in azoospermia DBY    DEAD-box helicase 3, Y-linked DHF    Dihydrofolate DHFR    Dihydrofolate reductase DMR    Differentially methylated region DNA    Deoxyribonucleic acid DNMT   DNA methyltransferase DOHaD   Developmental origins of health and disease DSB    Double-strand break DTT    Dithiothreitol EDTA    Ethylenediaminetetraacetic acid ES    Ectoplasmic specialization FDR    False discovery rate FOXL2   Forkhead box L2 FSH    Follicle-stimulating hormone fTHF    10-formyltetrahydrofolate GLDC    Glycine decarboxylase GnRH    Gonadotropin-releasing hormone GSH    Glutathione GSSG    Glutathione disulfide GST    Glutathione S-transferase hCYS    Homocysteine HPG    Hypothalamic-pituitary-gonadal ICR    Imprinting control region ICSI    Intracytoplasmic sperm injection IUGR    Intrauterine growth restriction IUI    Intrauterine insemination IVF    In vitro fertilization LH    Luteinizing hormone LINE    Long interspersed nuclear element xvi  MAT    Methionine adenyltransferase MBD4    Methylated DNA-binding protein MET    Methionine mHTF    Modified human tubal fluid miRNA   MicroRNA mRNA    Messenger RNA MS    Methionine synthase MTHFR   Methylenetetrahydrofolate reductase MTRR    Methionine synthase reductase NADPH   Nicotinamide adenine dinucleotide phosphate (reduced) ncRNA   Non-coding RNA NOA    Non-obstructive azoospermia O    Oligospermia OA    Obstructive azoospermia OAT    Oligoasthenoteratospermia P1    Protamine 1 P2    Protamine 2 PBS    Phosphate buffered saline PCR    Polymerase chain reaction PFA    Paraformaldehyde PGC    Primordial germ cell piRNA    Piwi-interacting RNA PIWI    P-element-induced wimpy PRDM1   PR domain zinc-finger protein 1 PWS    Prader-Willi syndrome RBC    Red blood cell RBMY   RNA binding motif protein, Y-linked REC8    Meiotic recombination protein 8 RFLP    Restriction fragment length polymorphism RISC    RNA-induced silencing complex xvii  RNA    Ribonucleic acid ROS    Reactive oxygen species RRBS    Reduced representation bisulfite sequencing RSPO1   R-spondin 1 SAH    S-adenosylhomocysteine SAHH    S-adenosylhomocysteine hydrolase SAM    S-adenosylmethionine SC    Synaptonemal complex SCOS    Sertoli cell-only syndrome SF1    Steroidogenic factor 1 SGO1    Shugoshin 1 SHMT    Serine hydroxymethyl transferase siRNA    Small-interfering RNA SNP    Single nucleotide polymorphism SO    Severe oligospermia SOX9    SRY-box 9 SOX17   SRY-box 17 SPO11    Meiotic recombination protein SPO11 SRS    Silver-Russell syndrome SRY    Sex-determining region Y StAR    Steroidogenic acute regulatory protein TBC    Tubulobulbar complex TDG    Thymine DNA glycosylase TET    Ten-eleven translocation THF    Tetrahydrofolate TP1    Transition protein 1 TP2    Transition protein 2 TUNEL   Terminal deoxynucleotidyl transferase dUTP nick-end labeling UPD    Uniparental disomy VR    Vasectomy reversal xviii  WHO    World Health Organization WNT3    Wingless-type MMTV integration site family, member 3 WNT4    Wingless-type MMTV integration site family, member 4   xix   ACKNOWLEDGEMENTS I would like to thank the faculty and staff of UBC who have helped me throughout my studies, I would not be here today without your support and dedication. In particular, I would like to thank my supervisor Dr. Sai Ma for her guidance and support during my time as a graduate student. I am especially grateful for the opportunity she has given me to complete a graduate degree in her lab. I would also like to thank the members of my supervisory committee Dr. Anthony Cheung, Dr. Wan Lam, and Dr. Kenneth Poon for their support and constructive feedback during my graduate studies. I would like to thank Dr. Victor Chow and Dr. Kenneth Poon for the patients and samples they have provided for the completion of this project. My graduate student experience would not be complete without the students of the RDS program and the members of the Ma lab for providing a wonderful community that fosters friendship and learning. In particular, I would like to acknowledge the past members of the Ma lab, especially Annie Ren, and the current members Luke Gooding, Kenny Louie, Samuel Schafer, and Kate Watt for their support, encouragement, and friendship throughout the years. Lastly, I own my deepest gratitude to my loving family for their unconditional support during my endeavours. I cannot express enough how much the loving and nurturing environment has meant to me throughout the years.  1  CHAPTER 1: INTRODUCTION The different tissues that make up the human body are made up of numerous cell types. These cell types, all of which carry an identical DNA complement, vary significantly in terms of physical and biological properties and, therefore, are functionally distinct. There are multiple mechanisms that allow cells containing identical DNA complements to become phenotypically different. Epigenetics is one such mechanism that can control the expression levels of genes contained in the DNA of different cell types that produces the phenotypic variations that we see. All epigenetic information is carried within each cell and can be reliably passed on to the next generation of cells through mitosis. This works for most cells; however, gametes are a special case because their genetic and epigenetic components are passed on to the offspring. The epigenetic process in gametogenesis, or the production of gametes, is also different between the two sexes. The following sections in this thesis will focus on male gametogenesis with references to female gametogenesis where necessary. 1.1 Development 1.1.1 Induction of Primordial Germ Cells The process of gametogenesis in males is also called spermatogenesis. This process converts diploid progenitor cells into haploid sperm cells with fertilization potential (Rato et al., 2012). The germline arises from the pluripotent pre-implantation epiblast cells of the blastocyst (Tang et al., 2016). Following implantation, the blastocyst undergoes gastrulation to form the trilaminar embryonic disc. At embryonic day 17, during the early stages of gastrulation, primordial germ cell (PGC) specification begins in the forming mesoderm layer (Figure 1.1). Bone morphogenetic protein (BMP) signalling from the epiblast and wingless-type MMTV integration site family, member 3 (WNT3) signalling from the hypoblast induces a few cells of 2  the epiblast to express SRY-box 17 (SOX17). PR domain zinc-finger protein 1 (PRDM1) is upregulated in response to SOX17 expression (Figure 1.2). Expression of both genes is critical for cells to become induced to the germ cell lineage. Fully specified PGCs begin to appear at approximately embryonic day 24. They are first found localized in the yolk sac wall near the allantois, before migrating through the hindgut to the forming genital ridges by embryonic day 37. It is during this migration period that epigenetic reprogramming takes place, which will be covered in section 1.4.4.1.1. PGCs that have colonized the genital ridge in males are termed gonocytes; in females, they are termed oogonia. Proliferation of gonadal PGCs continues in the genital ridges until about the tenth week of development when they become committed to one of two paths depending on the sex of the embryo: gonocytes of male embryos become mitotically quiescent, whereas oogonia of female embryos enter meiosis and become arrested at the first prophase.   3   Figure 1.1 Timeline of PGC development. Shortly after fertilization and the formation of the zygote, the blastocyst is formed with two distinct cell populations: the pluripotent epiblast and the hypoblast. After implantation, the embryo becomes a bilaminar disc. PGC specification begins shortly after gastrulation on embryonic day 17. A small group of cells in the newly forming mesoderm layer receive the appropriate signals to form the germline precursors. Specified PGCs are localized in the yolk sac wall near the allantois at embryonic day 24. PGCs then migrate through the hindgut to the genital ridges at embryonic day 37. Epigenetic reprogramming occurs during this migration period. After colonizing the genital ridges, the PGCs, now called gonocytes, remain proliferative until the tenth week when they become quiescent. Timeline not to scale. Figure adapted from Tang et al., 2016.   Figure 1.2 Gene signalling network for PGC specification. BMP-SMAD signals upregulate SOX17. SOX17 induces expression of its downstream target PRDM1. Both SOX17 and PRDM1 form a crucial signalling network for the specification of PGCs. PRDM1 represses somatic genes that were upregulated by BMP-SMAD and SOX17 to induce germ cell specification. WNT3-β-catenin signalling may also be involved but its importance remains unclear. Figure adapted from Tang et al., 2016. 4  1.1.2 Sex Differentiation Determination of the sex of the embryo begins with the chromosomal content inherited from the parents (She and Yang, 2017). The developing genital ridges during early embryogenesis begin undistinguished and are in a bipotential state, that is they are able to form either testes or ovaries depending on cellular signals (Shima and Morohashi, 2017). The master regulator in humans is the sex-determining region Y (SRY) gene on the Y chromosome. SRY directly upregulates SRY-box 9 (SOX9), which, together with steroidogenic factor 1 (SF1), begins a complex cascade of signalling networks that ultimately results in the differentiation of the supporting Sertoli cells and steroidogenic Leydig cells within the testes (She and Yang, 2017). The precise timing of SRY expression is also important for testis development. There is a narrow window in which SRY expression must reach a threshold in order for there to be proper regulation of SOX9. In females, lack of the Y chromosome, and therefore SRY, results in the formation of supporting granulosa cells and steroidogenic theca cells in the ovaries using a different network of genes including R-spondin 1 (RSPO1), wingless-type MMTV integration site family, member 4 (WNT4), and forkhead box L2 (FOXL2). Once the Sertoli and Leydig cells of the gonads have been established, they will begin to secrete anti-Müllerian hormone (AMH) and testosterone, respectively. AMH and testosterone further induce male sex differentiation and masculinization of the external genitalia. More specifically, AMH causes the regression of the Müllerian ducts, while testosterone stabilizes the Wolffian ducts, which will eventually develop into the vas deferens, epididymis, and seminal vesicle (Hughes, 2001). In females, the Wolffian ducts regress in the absence of testosterone, while the absence of AMH allows the Müllerian ducts to give rise to the female sex cords (uterus, fallopian tubes, and upper vagina). 5  The gonocytes specified earlier that have colonized the previously undifferentiated genital ridges will eventually become incorporated into the newly formed Sertoli cells within the testes (Shima and Morohashi, 2017). Gonocytes remain mitotically quiescent until shortly after birth when they migrate to the basal membrane of the seminiferous tubules and differentiate into spermatogonia, which then take part in spermatogenesis (Culty, 2013; Manku and Culty, 2015). 1.2 Spermatogenesis The process of spermatogenesis produces four haploid spermatozoa from a single spermatogonium. It takes an average of 74 days from spermatogonia to spermatozoa release and the seminiferous epithelium undergoes a new cycle every 16 days (Amann, 2008). A brief overview of the steps will be discussed in this section. 1.2.1 Spermatogonia Spermatogenesis takes place in the seminiferous tubules. Spermatogenesis is a process that takes a single diploid spermatogonium through a series of morphologically distinct stages that ultimately ends with mature spermatozoa capable of fertilization. The process begins on the basal membrane of seminiferous tubules which are lined with Sertoli cells (Figure 1.3). Type A spermatogonia are differentiated from gonocytes (Manku and Culty, 2015). Type A spermatogonia can be further divided into type Adark (Ad) and type Apale (Ap) based on their nuclear staining (Aponte et al., 2005). Type Ad spermatogonia are generally considered to be the spermatogonial stem cell reserve as they do not actively undergo mitosis (Figure 1.4; Amann, 2008; Aponte et al., 2005). Type Ad spermatogonia are only active under certain circumstances, namely, to repopulate after cell loss. Type Ap spermatogonia are the active spermatogonial stem cells that give rise to mature spermatozoa periodically. These cells are able to replenish their own pool, leaving the type Ad spermatogonia as reserves. Division of type Ap spermatogonia takes 6  place only once every 16 days when the two daughter type Ap spermatogonia are committed to become spermatozoa. Once committed, the type Ap spermatogonia differentiate and give rise to type B spermatogonia, which further differentiate into preleptotene primary spermatocytes (Figure 1.4; Lie et al., 2009).   7   Figure 1.3 Migration of spermatozoa through the seminiferous tubules during spermatogenesis. Seminiferous tubules are lined with Sertoli cells, which coordinate production of spermatozoa. Type A spermatogonia differentiate into type B spermatogonia on the basal membrane of the seminiferous tubules. Differentiation into preleptotene primary spermatocytes occurs prior to crossing the blood-testis barrier (BTB). Diploid spermatocytes then undergo meiosis to produce haploid spermatids. Spermiogenesis occurs during the final stages of spermatogenesis and consists of morphological changes that include tail and acrosome formation. Mature spermatozoa are then released into the seminiferous tubule lumen. Figure adapted from Rato et al., 2012. 8   Figure 1.4 Overview of spermatogenesis. Type Ad spermatogonia are the spermatogonial stem cell reserves. They only undergo mitosis under certain conditions (dotted arrows). Type Ap spermatogonia have self-renewing abilities but also differentiate to committed spermatogonia once every 16 days. Successive mitotic divisions and differentiation give rise to type B spermatogonia and preleptotene primary spermatocytes. Diploid primary spermatocytes cross the blood-testis barrier (red dotted line) before undergoing meiosis I to produce haploid secondary spermatocytes. Secondary spermatocytes undergo meiosis II to produce round spermatids. During spermiogenesis, round spermatids form tails and acrosomes. Finally, the spermatozoa are released from the Sertoli cells into the lumen of the seminiferous tubule. Process is only shown for one member of a pair of committed type Ap spermatogonia. Figure adapted from Amann, 2008 and Lie et al., 2009. 9  1.2.2 Meiosis I Before primary spermatocytes can undergo the first meiotic division, they must first cross the blood-testis barrier (BTB). The BTB is a specialized junction composed of tight, gap, and desmosome junctions as well as ectoplasmic specializations between adjacent Sertoli cells that separates the seminiferous epithelium into basal and adluminal compartments (Lie et al., 2009; Mruk and Cheng, 2015). The BTB functions to prevent toxins from entering into the lumen, where the developing sperm cells are located, and also allows fine control over the environmental conditions within the lumen. The BTB must undergo extensive restructuring as the spermatocytes cross compartments from the basal membrane of the seminiferous tubules to the lumen. As the spermatocytes cross the barrier, the BTB must disassemble in front of the spermatocytes and reassemble behind the spermatocytes in order to maintain the integrity of the barrier. This restructuring is a complex process that involves highly regulated pathways that must be coordinated with cell cycle progression and cell migration. After crossing the BTB, the spermatocytes can continue into prophase I of the first meiotic division. Meiosis I is a cell division process that results in the reduction of chromosome number or ploidy. During the synthesis phase of interphase, preleptotene primary spermatocytes replicate the DNA of each homologous chromosome to form sister chromatids that enter meiotic divisions. Homologous recombination occurs in the first phase of meiosis I called prophase I. Homologous recombination functions to produce new combinations of DNA in the gametes, which increases genetic variability in the offspring that aids in environmental adaptation and continues the process of evolution. Prophase I is divided into several substages that describe the events leading to recombination. Homologous chromosomes begin to assemble a protein core of lateral elements along their lengths during the leptotene stage (Figure 1.5; Heyting, 1996; 10  Morelli and Cohen, 2005; Pawlowski and Cande, 2005). It is at this stage that homologous recombination is initiated by the induction of DNA double-strand breaks (DSBs) by meiotic recombination protein SPO11 (SPO11). During the zygotene stage of prophase I, transverse filaments assemble between the lateral elements on the homologous chromosomes to form the central element of the synaptonemal complex (SC). SC formation allows homologous chromosomes to synapse in preparation for crossover events. Most DSBs are also repaired by this stage, however, these repairs do not result in crossovers. Complete synapsis is done by the pachytene stage. Once synapsed, a variety of proteins are recruited to repair DSBs between homologous chromosomes resulting in chiasmata at the crossover sites that tether the bivalents together. During the diplotene and diakinesis stages of prophase I, the SC disassembles and the chromosomes further condense, respectively. During metaphase I, the bivalents align themselves along the metaphase plate. 11   Figure 1.5 The meiotic cell cycle. Meiosis can be described as one round of replication followed by two rounds of division. Meiosis begins with DNA replication during the synthesis phase of interphase (top panel). Once replicated, sister chromatids are bound by two types of cohesins that allow sequential separation of homologous chromosomes (meiosis I) followed by sister chromatids (meiosis II). After replication, the cell moves on to prophase I of meiosis I (box, middle panel). During leptotene, the lateral elements begin to assemble (teal and purple). SPO11 begins to induce DSBs for homologous recombination. During zygotene, transverse elements begin to assemble between the lateral elements to form the central element (black). Complete SC formation and synapsis is complete by pachytene. During pachytene, a DSB is repaired between homologous chromosomes resulting in a chiasma at the crossover site. During diplotene and diakinesis, the SC disassembles and the chromosomes condense further. Meiosis I (middle panel) continues with the phosphorylation of arm cohesins, which are recognized and degraded by separase (pink blob) to allow homologous chromosomes to separate to opposite poles. In meiosis II (bottom panel), centromeric cohesins become phosphorylated to allow sister chromatids to separate. SC, synaptonemal complex; S, separase; P, phosphorylation. Figure adapted from Marston and Amon, 2004, Morelli and Cohen, 2005, Pawlowski and Cande, 2005, and Miller et al., 2013. 12  After the synthesis phase of interphase, the sister chromatids of each homologue are bound by protein complexes called cohesins. In order to separate homologous chromosomes during anaphase I, the cohesins must be removed. However, removing all cohesins at this stage would leave the sister chromatids unable to be faithfully separated at anaphase II. The solution is to have two types of cohesins that form during meiosis. Arm cohesins form along the length of the sister chromatids, while centromeric cohesins form near the centromeres. Meiotic cohesins differ from mitotic cohesins in that they substitute specific subunits for meiotic recombination protein 8 (REC8). Differential cohesin loss is accomplished by differential phosphorylation of REC8. REC8 in the arm cohesins become phosphorylated by various kinases resulting in their recognition and removal by the protease separase (Figure 1.5; Marston and Amon, 2004; Miller et al., 2013). Centromeric cohesins are protected from phosphorylation by shugoshin 1 (SGO1) recruitment of phosphatases. The sister kinetochores attach to the same pole allowing the homologous chromosomes to be separated to opposite poles and the crossovers to be resolved. Meiosis I concludes with telophase I and cytokinesis, which is the gathering of the chromosomes at the poles of the cell and division of the cytoplasm, respectively. 1.2.3 Meiosis II Meiosis II shares similarities with mitosis in that the resultant daughter cells retain the same chromosome number in an equational cell division process. Meiosis II is composed of prophase II, metaphase II, anaphase II, telophase II, and cytokinesis. The stages are similar to those in meiosis I; major differences will be briefly outlined in this section. First, SC formation and recombination events do not occur in prophase II. Instead, the chromosomes condense as they did at the end of prophase I. Second, during anaphase II, the remaining centromeric cohesins are degraded by separase due to either the inhibition or the 13  degradation of SGO1 (Figure 1.5; Marston and Amon, 2004; Miller et al., 2013). During anaphase II, the sister kinetochores are attached to opposite poles to allow for the separation of sister chromatids. The final result of meiosis is four haploid round spermatids. 1.2.4 Spermatid Remodelling In the last step of spermatogenesis, called spermiogenesis, the haploid round spermatids undergo extensive morphological changes before becoming mature spermatozoa capable of fertilization. The major changes that occur during the process of remodelling will be outlined below. The first morphologically distinct change is the formation of the acrosome. The acrosome is a membrane-bound organelle found attached to the front of the sperm head, between the nuclear and plasma membranes. It functions in fertilization by releasing various hydrolytic enzymes that digest the zona pellucida to allow the sperm to bind to the oocyte. The acrosome forms from the Golgi network, which produces hydrolytic vesicles that bud off and combine to cover the front half of the sperm head (O’Donnell, 2014). Various cytoskeletal elements, such as the perinuclear theca and the acroplaxome, anchor the acrosome in place between the nuclear and plasma membranes (Figure 1.6). 14   Figure 1.6 Major morphological features that appear during spermatid maturation. A. Diagram of a maturing spermatid. The acrosome (red) containing enzymes necessary for fertilization is anchored between the nuclear and plasma membranes by the acroplaxome (orange). The microtubule manchette (green) helps shape the nucleus into an oval shape characteristic of human sperm and positions the excess cytoplasm toward the back of the sperm for removal. The axoneme (blue) forms the core of the sperm flagellum that gives the sperm motility. B. Cross section of the axoneme. Microtubules are arranged in an outer ring of nine doublets (green and orange) with an additional doublet in the center to form a 9+2 arrangement. Sheaths and motor proteins not shown. C. During spermiogenesis, the ES anchors the sperm to the Sertoli cell. During spermiation, TBCs form to remove the ES in preparation for releasing the sperm into the lumen and may also play a role in removing excess cytoplasm. ES, ectoplasmic specialization; TBC, tubulobulbar complex. Figure adapted from Göb et al., 2010, O’Donnell et al., 2011, and O’Donnell, 2014.  The next major change during spermiogenesis is the shaping of the sperm head and compaction of the nucleus. At this stage, the nucleus and acrosome migrate to one side of the cell. The acroplaxome and the ectoplasmic specialization (ES) on the Sertoli cell transmit forces onto the nucleus that reshape it from spherical to the characteristic oval shape seen in humans. 15  The ES also functions to orient the nucleus toward the basal membrane and anchor the spermatid to the Sertoli cell throughout spermiogenesis and spermiation. The elongation phase is also facilitated by a related structure called the manchette, which is formed from numerous microtubule bundles. The manchette moves towards the back of the sperm and reshapes the nucleus as it goes; it may also play a role in positioning the cytoplasm towards the back of the sperm for later removal during spermiation (Figure 1.6; O’Donnell, 2014). Concurrently, the sperm chromatin must also be compacted to reduce the nuclear volume. The process disassembles nucleosomes and replaces histones with transition proteins which are subsequently replaced by protamines. Protamines allow for a higher level of compaction by using toroidal structures. Failure to properly replace histones and incorporate protamines can adversely affect head shape, since the acroplaxome, ES, and manchette are reshaping the nucleus concurrently. Protamines also have important implications on epigenetics and will be revisited in a later section. The final major change of spermiogenesis is the formation of the flagellum, which gives mature spermatozoa motility. Shortly after the completion of meiosis, the core of the flagellum called the axoneme is assembled. The axoneme’s structure is composed of a central pair of microtubules surrounded by nine additional pairs of microtubules to form a 9+2 arrangement (Figure 1.6; O’Donnell, 2014). There are also dynein motor proteins on the outer pairs of microtubules that produce movement of the flagellum. As the cell is no longer dividing, there is an unused pair of centrioles in the centrosome, which is now called the basal body. The basal body migrates to the opposite pole of the nucleus/acrosome to begin forming the axoneme. The outer dense fibers, fibrous sheath, and mitochondrial sheath are secondary structures that assemble around the axoneme during the elongation phase. These secondary structures provide 16  the flagellum with rigidity and facilitate motility. The formation of the flagellum relies on protein trafficking from the Golgi using the acroplaxome and manchette cytoskeletal elements. The final step before the Sertoli cells release mature spermatozoa is called spermiation. The ES junction and the Sertoli cell cytoskeleton transport the spermatids to the luminal edge of the Sertoli cell in preparation for release. There are major changes that take place in the adhesion complex between Sertoli cell and spermatid during spermiation. Modified endocytic structures called tubulobulbar complexes (TBCs) appear between the Sertoli cell and spermatid (Figure 1.6). These structures remove the tight ES adhesion between Sertoli cell and spermatid. The TBCs have a clathrin-coated pit and a bulbous region that buds off and proceeds to lysosomal degradation in the Sertoli cell. TBCs may also play a role in reducing the cytoplasm volume (O’Donnell et al., 2011; O’Donnell, 2014). Finally, the spermatid head is extended into the lumen by a cytoplasmic Sertoli cell stalk, while the spermatid cytoplasm remains anchored to the Sertoli cell. As the spermatids are release into the lumen, the cytoplasm is stripped off and remains within the Sertoli cell and becomes phagocytosed. The cytoplasm that is stripped away is termed the residual body and reduces the cytoplasm volume of the spermatids by up to 70%. The spermatozoa that are released from Sertoli cells are initially immotile but gain motility in the epididymis (Jones, 1999). 1.3 Regulation Sex differentiation, development of secondary sexual characteristics, and spermatogenesis all rely on a steady supply of steroid hormones. The hormones are produced in collaboration with the central nervous system. The hypothalamic-pituitary-gonadal (HPG) axis is composed of the hypothalamus, anterior pituitary, and testes in males. The HPG axis is the main regulator of sex 17  hormones and reproduction. The HPG axis also contains many levels of regulation through multiple feedback mechanisms. The hypothalamus releases gonadotropin-releasing hormone (GnRH) in periodic pulses every 30-120 minutes (Figure 1.7; Millar et al., 2004; Vadakkadath Meethal and Atwood, 2005; Jin and Yang, 2014). GnRH is a neuropeptide secreted from the axons of hypothalamic neurons. The pulses of GnRH are released into the hypophyseal portal system to the anterior pituitary to directly stimulate the production and release of two gonadotropins from the anterior pituitary: luteinizing hormone (LH) and follicle-stimulating hormone (FSH). The pulsatile release of GnRH also confers pulsatile secretion of the gonadotropins from the anterior pituitary, however, FSH pulses are not as distinct because it is constitutively released and it is regulated by other feedback mechanisms (see below). The gonadotropins enter into the circulatory system where they travel to their effector organs: the testes. LH binds to and activates LH receptors on Leydig cells to activate a cellular signalling pathway that causes the steroidogenic acute regulatory protein (StAR) to import cholesterol into the mitochondria (Isidori et al., 2008). Cholesterol then undergoes a series of enzymatic reactions catalyzed by cytochrome P450 to produce testosterone. A small amount of testosterone undergoes aromatase conversion to estrogen, which is also important for male reproductive function. Both testosterone and estrogen negatively regulate GnRH secretion from the hypothalamus and LH secretion from the anterior pituitary. FSH binds to and activates FSH receptors on Sertoli cells which activates pathways leading to transcription and expression of genes important for spermatogenesis. One important product of FSH binding to Sertoli cells is the production of the glycoprotein inhibin B. Inhibin B negatively regulates the secretion of FSH from the anterior pituitary.  18  Additionally, another feedback loop involving activin and follistatin further adds complexity. Activin is produced by peripheral tissues and upregulates GnRH and gonadotropin secretion (Vadakkadath Meethal and Atwood, 2005). Inhibin B produced by the Sertoli cells can also act to inhibit activin from stimulating the hypothalamus and anterior pituitary. Peripheral tissues also produce follistatin which negatively regulates activin (Figure 1.7).  Figure 1.7 Feedback regulation of the HPG axis. Peripheral tissues initiate the HPG axis through the production of activin. Activin begins by stimulating the hypothalamus to secrete GnRH, which acts on the anterior pituitary. In response to GnRH, the anterior pituitary produces and releases two gonadotropins: LH and FSH. LH acts on the Leydig cells in the testes to stimulate the production of testosterone and estrogen from cholesterol. The sex steroids provide negative feedback on the hypothalamus and anterior pituitary resulting in a decrease in GnRH and LH. FSH stimulates the Sertoli cells in the testes to express genes necessary for spermatogenesis. The Sertoli cells also produce inhibin B which has negative feedback on FSH and activin. Peripheral tissues also produce follistatin that inhibits activin. Green arrows show positive feedback or stimulating interactions. Red blunted lines show negative feedback or inhibiting interactions. Figure adapted from Vadakkadath Meethal and Atwood, 2005 and Isidori et al., 2008. 19  1.4 Male Infertility The World Health Organization (WHO) defines infertility as couples failing to achieve a pregnancy within 12 months of unprotected sexual intercourse. Since there is no reliable method of estimating infertility rates, the current prevalence of infertility reported in the literature is inconsistent. However, many groups agree the global incidence of infertility among couples is estimated to be 8-12%, with some suggesting the rate to be as high as 20% (Zhu et al., 2006; Bisht et al., 2017). There also appears to be an increasing trend in these infertility rates due to various factors including obesity, diabetes, environmental factors, and socioeconomic factors (Miyamoto et al., 2015; Neto et al., 2016; Craig et al., 2017). The effect of some of these factors on male fertility will be discussed below. Among the different sources, there is general agreement that male factor infertility affects approximately half of these cases (Miyamoto et al., 2015). Most cases in which a cause can be identified involve genetic factors, however, this only accounts for a small portion of cases; most cases are diagnosed as idiopathic (Georgiou et al., 2006). This section will focus on the types and causes of male factor infertility. 1.4.1 Assisted Reproductive Technologies As technology advances, assisted reproductive technologies (ARTs) are becoming a common treatment for infertile couples seeking medical treatment. There are three commonly used ARTs which all involve the in vitro handling of gametes. In vitro fertilization (IVF) is generally used for female factor infertility because the sperm must be motile and relatively concentrated. In IVF, the female is treated with hormones to induce superovulation of multiple oocytes, instead of just a single oocyte per cycle. Oocytes are then aspirated out of the ovary and placed in a petri dish containing culture medium. Sperm from the male partner is washed before being incubated with the oocyte. After fertilization occurs, the embryo is cultured to either the 20  cleavage stage or blastocyst stage before being transferred back into the uterus for implantation. The second of the commonly used ARTs is called intracytoplasmic sperm injection (ICSI) and is used mainly for severe male factor infertility. ICSI can be used for couples with low sperm counts and lack of motile sperm because only one sperm is manually inserted into an oocyte. The process follows similar procedures to IVF except that a single sperm is manually isolated and injected into an oocyte under a microscope. Lastly, intrauterine insemination (IUI) is a less invasive procedure used for more mild forms of infertility. IUI can also be used with donor sperm for surrogacy or with same-sex couples. Instead of aspirating the oocytes out of the ovary, washed sperm is just injected into the uterus to facilitate fertilization.  Infertility arises from some genetic or other defects in the reproductive system which would prevent the transmission of the defect to future generations. Therefore, a caveat of using ART is that couples are bypassing this natural biological mechanism. Although the majority of children conceived via ART grow to be healthy adults, several reports suggest that there are greater incidences of many adverse outcomes and defects, such as preterm birth, low birth weight, and congenital malformations, seen in ART children compared to naturally conceived children (Hansen et al., 2005; Zhu et al., 2006; Rimm et al., 2011; Davies et al., 2012). Although the relative risks reported are higher in the ART population, the absolute number of cases is still very low. Nevertheless, these findings continue to question the safety of ART as a viable treatment for infertility. A major problem in assessing the safety of ART is that there is considerable difficulty in separating the effects of using infertile patients’ gametes and any effects the ART procedure itself may be causing; however, there are some aspects of ARTs are already known to affect the embryo (Manipalviratn et al., 2009). 21  1.4.2 Phenotypes of Male Infertility Male infertility has many different clinical presentations that often do not adhere to strict definitions. In general, a routine semen analysis will be able to provide basic information regarding several sperm parameters that are critical for diagnosing several phenotypes of infertility. Common characteristics reported from a semen analysis include sperm count, motility, and morphology. Using WHO’s 5th centile lower reference limits, phenotypic diagnoses can be made regarding these characteristics. Oligospermia is defined as sperm count below 15 million sperm/mL, asthenospermia is defined as progressive sperm motility below 32%, and teratospermia is defined as normal sperm form below 4% (Table 1.1; WHO, 2010). More commonly, semen characteristics of infertile men will be below several reference limits and, therefore, a combination of phenotypes are assigned. For example, a semen sample with low sperm count, motility, and morphology would be diagnosed as oligoasthenoteratospermia (OAT). Fortunately, men with any of these diagnoses still have a chance to be able to conceive through the use of ARTs. Table 1.1 WHO lower reference limits for common semen parameters. Parameter 5th centile lower reference limit Semen volume (mL) 1.5 Total sperm number (million per ejaculate) 39 Sperm concentration (million per mL) 15 Total motility (%) 40 Progressive motility (%) 32 Vitality (%) 58 Morphology (%) 4  When there is a complete absence of sperm from the ejaculate, the patient would be diagnosed with azoospermia. Azoospermia can be further categorized depending on whether the 22  defect involves the transport of sperm through the reproductive tract or the production of sperm in the testes. Obstructive azoospermia involves a blockage in the reproductive tract that prevents sperm from travelling from the testes and into the ejaculate. However, these men will generally have normal spermatogenesis and will be able to retrieve sperm from the testes through testicular sperm extraction, testicular sperm aspiration, or other sperm retrieval techniques for use in ICSI. Men with non-obstructive azoospermia have abnormal spermatogenesis which leads to a reduced capacity for normal sperm formation. This can be caused by a congenital defect or defects in the HPG axis. For example, hypogonadotropic hypogonadism is characterized by impaired production of gonadotropins which leads to decreased sex steroid production and ultimately defects in spermatogenesis. Men with this condition will have a lower chance of retrieving sperm for ICSI. Therefore, it is common for these men to have difficulty in isolating sperm for ICSI. 1.4.3 Genetic Causes of Male Infertility Genetics can be identified as a cause of male infertility in 15-30% of cases (Ferlin et al., 2006; Georgiou et al., 2006; Neto et al., 2016). There are three common genetic related factors that are known to cause infertility in males: Y chromosome microdeletions, chromosomal abnormalities, and cystic fibrosis gene mutations. Since these factors are common genetic causes of male infertility, it is routine practice to investigate these during treatment for infertility in the clinic. A brief overview of each genetic factor will be discussed in this section. 1.4.3.1 Y Chromosome Microdeletions The Y chromosome carries several genes that are important for proper spermatogenesis. Y chromosome microdeletions have a prevalence of between 10% - 15% in azoospermic men, whereas a lower prevalence of between 5% - 10% is seen in oligospermic men (Georgiou et al., 2006; O’Flynn O’Brien et al., 2010). Y chromosome microdeletions are chromosomal deletions 23  of genes on the Y chromosome that cannot be detected with cytogenetics. The Y chromosome is composed of multiple repeats that share sequence identity, some of which are inverted (Lange et al., 2013). These repeats make the sequences on the Y chromosome more prone to mismatches during homologous recombination (Figure 1.8). As a result, interchromosomal or intrachromosomal rearrangements may occur that can introduce deletions, duplication, or inversions. Depending on the gene or genes involved, there may or may not be impaired spermatogenesis.   Figure 1.8 The Y chromosome. A. Mechanism of Y chromosome microdeletion. During homologous recombination, pairing may occur between different repeat sequence (pink). After resolution of double-strand breaks, one product will have a duplication and the other product will have a deletion. B. Structure of the Y chromosome. AZF regions are highlighted (red, green, blue). Important genes found within each region are indicated. Figure adapted from O’Flynn O’Brien et al., 2010.  Important genes for spermatogenesis on the Y chromosome are found in a region of the long arm called the azoospermia factor (AZF) region. The AZF region makes up most of the long arm and is subdivided into AZFa, AZFb, and AZFc regions (Figure 1.8). Deletions in these 24  AZF regions have been known to cause phenotypes ranging from oligospermia to azoospermia. The AZFa region contains the DEAD-box helicase 3, Y-linked (DBY) gene, which results in Sertoli cell-only syndrome (SCOS) when deleted (O’Flynn O’Brien et al., 2010). SCOS is characterized by azoospermia and a lack of germ cells lining the seminiferous tubules. There would be very little chance of men with SCOS to obtain sperm for ICSI through sperm retrieval methods, however, there have been cases in which sperm was retrieved from the seminiferous tubules (Georgiou et al., 2006). Although AZFa region deletions produce the most severe phenotype, these deletions only have a prevalence of up to 1% (O’Flynn O’Brien et al., 2010). The main gene within the AZFb region is the RNA binding motif protein, Y-linked (RBMY) gene. Deletions in this region have the same prevalence as AZFa deletions but are less severe than deletions in the AZFa region. Deletions will often result in spermatogenic arrest at the primary spermatocyte stage which will lead to azoospermia (Georgiou et al., 2006; O’Flynn O’Brien et al., 2010). The most common of the Y chromosome microdeletions occurs in the AZFc region, accounting for 60% of all Y chromosome deletions (Neto et al., 2016). Phenotypes of AZFc deletions are much more varied than AZFa or AZFb deletions. A reduction in semen parameters is commonly seen in AZFc deletion cases, however, deletions can cause azoospermia or have no effect (normozoospermia). The deleted in azoospermia (DAZ) gene is central to the AZFc region. Since there are variable phenotypes upon deletion, the DAZ gene may not be necessary for the initiation of spermatogenesis, instead may only be necessary for the completion of spermatogenesis. As AZFc deletions are considered to be the least severe with regards to phenotype, men with these deletions are often able to conceive using ARTs. It is important to test for Y chromosome microdeletions in infertile men seeking treatment for their infertility. The use of ART can artificially propagate any Y chromosome microdeletions 25  to male newborns, who will invariably suffer from the same infertility status as the father (Georgiou et al., 2006). Additionally, some studies have reported that partial deletions of the Y chromosome predispose existing deletions to further expansion and possibly entire chromosome loss (O’Flynn O’Brien et al., 2010). Finally, reported incidences and phenotypes can vary greatly due to both geographical and ethnic interactions (O’Flynn O’Brien et al., 2010; Neto et al., 2016). 1.4.3.2 Chromosomal Abnormalities Large-scale chromosomal abnormalities are also a commonly seen cause of male infertility with a prevalence of 5% of infertile men (O’Flynn O’Brien et al., 2010). There are two general categories of chromosomal abnormalities that will be discussed in this section: translocations and aneuploidy. Autosomal translocations are 4-10 times more likely to be found in men with infertility than in normal fertile men (Georgiou et al., 2006; O’Flynn O’Brien et al., 2010). Reciprocal translocations are also called balanced translocations because DNA fragments are exchanged between chromosomes resulting in no gain or loss of fragments. Depending on the exact location of the breakpoints, normal gene function can be disrupted in any number of pathways including reproductive function (O’Flynn O’Brien et al., 2010). It is also possible for reciprocal translocations to introduce genes into a new regulatory network (Harewood and Fraser, 2014). Depending on the combination of gene and regulatory network, the gene may be turned on or off at inappropriate times resulting in a disruption of the gene’s function. Regardless of the genes involved in a translocation, there is generally some effect on the fertility of the individual. Translocating fragments of one chromosome onto another chromosome makes the pairing of homologous chromosomes during meiosis difficult. Consequently, segregation of chromosomes 26  may lead to an imbalance of chromosomes in the daughter cells (Georgiou et al., 2006). There is a special class of translocations called Robertsonian translocations which involves fusion of two acrocentric chromosomes (chromosomes with the centromere close to one end of the chromosome). Robertsonian translocations have a prevalence of 0.8% in infertile men, however, this is substantially higher than normal fertile men (O’Flynn O’Brien et al., 2010). These translocations can affect fertility in a similar manner as reciprocal translocations. Robertsonian translocations that are passed on to offspring may also result in one of many trisomies (Georgiou et al., 2006). The second major chromosomal abnormality that affects fertility is aneuploidy. Aneuploidy is the deviation from the correct chromosome number; in humans, the correct chromosome number is 46. Oocytes and embryos can have an aneuploidy rate up to 20% and 25% respectively, however, only about 2% of sperm is reported to contain aneuploid chromosomes. This high level of aneuploidy means that meiotic defects are quite common (Hassold and Hunt, 2001). There are two main mechanisms that give rise to aneuploidy: non-disjunction and premature separation of sister chromatids. Since there are two cell division phases during meiosis, there is a chance that non-disjunction may occur at either meiosis I or meiosis II. Non-disjunction at meiosis I occurs when the homologous chromosomes migrate to the same pole. Alternatively, if chromosomes fail to pair and recombine to form chiasma, the homologous chromosomes may end up migrating to the same pole by chance. The result in both cases is all four daughter cells being aneuploid (Figure 1.9). Indeed, previous studies by our lab provide evidence that decreased recombination events are associated with the production of aneuploid sperm (Ma et al., 2006; Ferguson et al., 2007; Kirkpatrick et al., 2015). More recently, we also show that the distribution of recombination events along a chromosome may also play a 27  role in sperm aneuploidy (Ren et al., 2016). Non-disjunction at meiosis II involves the failure of sister chromatids to separate. The sister chromatids will migrate to the same pole, resulting in 50% of the daughter cells being aneuploid (Calogero et al., 2003). The premature loss of cohesin during segregation will result in the premature separation of sister chromatids. A variety of products are possible depending on if one or both homologues are affected and whether it occurs in meiosis I or meiosis II (Nagaoka et al., 2012).  Figure 1.9 Products of non-disjunction during meiosis. Left panel, non-disjunction occurring at meiosis I causes both homologous chromosomes to migrate to the same pole. Segregation after normal meiosis II results in all daughter cells being aneuploid: 50% will be disomic and 50% will be nullisomic. Right panel, homologous chromosomes successfully segregate during meiosis I, but non-disjunction occurs in one of the daughter cells. In this case, sister chromatids migrate to the same pole resulting in normal gametes from the unaffected daughter cell and aneuploid gametes from the non-disjunction affected daughter cell. The final outcome will be 50% normal and 50% aneuploid, half of which are disomic and the other half are nullisomic. Figure adapted from Calogero et al., 2003.  A common case of aneuploidy that affects male fertility results from an extra X chromosome called Klinefelter syndrome. The extra X chromosome is inactivated as in females, 28  but the residual activity from the silent chromosome interferes with spermatogenesis (Tüttelmann and Gromoll, 2010; Neto et al., 2016). Klinefelter syndrome can be seen in up to 5% of oligospermic men and 10% of azoospermic men and generally results in spermatogenic arrest at the primary spermatocyte stage (Georgiou et al., 2006; O’Flynn O’Brien et al., 2010). Klinefelter syndrome patients most commonly have a 47, XXY non-mosaic karyotype, but mosaic karyotypes such as 46, XY/47, XXY are also observed. In either case, patients are affected with hypogonadism and infertility; nevertheless, sperm retrieval methods are often able to extract sperm from either the ejaculate or testicular biopsies for use in ICSI. However, studies have observed that Klinefelter syndrome patients possess a greater portion of gametes with aneuploidy, which increases the risk of producing children with chromosomal abnormalities when using ART. 1.4.3.3 Cystic Fibrosis Gene Mutations Cystic fibrosis (CF) is a common genetic disease that can affect various exocrine tissues found throughout the body such as in the respiratory, digestive, or reproductive systems (Cuppens and Cassiman, 2004; Chen et al., 2012). Although more than 1000 mutations have been described, the recessive disease most commonly originates from a single point mutation called ΔF508 in the cystic fibrosis transmembrane conductance regulator (CFTR) gene, which results in misfolding of the protein (Rowe et al., 2005). This one mutation can be found in approximately 70% of all CFTR alleles with defects and is present in approximately 90% of CF patients in the United States (Rowe et al., 2005). CFTR is a transmembrane regulator of chloride permeability. Without chloride ion transport, epithelial secretions become thick and viscous, causing obstruction within the ducts. It is estimated that 4% of Caucasians are carriers of CFTR gene mutations, leading to approximately 1 in 3500 newborns inheriting the disease. Depending 29  on which mutations are involved, CF can be caused in several ways, including complete lack of protein synthesis, protein misfolding and degradation, or defective regulation and functioning of the ion channel (Rowe et al., 2005). With respect to male infertility, CFTR mutations cause thick epithelial secretions in the vas deferens in utero which results in atresia. Eventually the vas deferens and associated ducts undergo atrophy resulting in congenital bilateral absence of vas deferens (CBAVD), a form of obstructive azoospermia (Lissens et al., 1996). Moreover, CBAVD and infertility is consistently observed in greater than 95% of men with CF (Lissens et al., 1996). 1.4.4 Epigenetics of Male Infertility As mentioned in an earlier section, the aetiologies for male infertility are unknown in the majority of cases. However, recently there has been growing interest in looking for potential explanations that do not involve the chromosomes themselves. Many groups have begun to study epigenetics to elucidate mechanisms of infertility. Epigenetics is the study of reversible and heritable modifications to the DNA that induce alterations to gene expression without affecting the underlying DNA sequence. Furthermore, differentiation of cells with the same DNA composition can be accomplished with epigenetic mechanisms (Bird, 2002). Additionally, many epigenetic mechanisms have been shown to be necessary for proper development (O’Flynn O’Brien et al., 2010). Thus, epigenetics is an attractive field of study that may explain idiopathic cases of infertility. This section will focus on DNA methylation as it relates to male infertility followed by a brief discussion of other epigenetic mechanisms. 1.4.4.1 DNA Methylation DNA methylation is by far the most well studied of all the epigenetic mechanisms and usually confers gene silencing. The modification process involves the addition of a methyl (CH3) 30  group to cytosine residues of DNA. Methyl group addition to the carbon at the fifth position in the cytosine ring produces 5-methylcytosine (5-mC). Cytosine residues that are immediately followed by a guanine form a CG dinucleotide or CpG site. Clusters of CpG-rich regions can often be found upstream of genes and are termed CpG islands (Bird, 2002). DNA methylation may silence genes by physically interfering with transcription factor binding, or associating with other epigenetic mechanisms to form chromatin (Jin et al.,2011). The process of methylation is catalyzed by a family of enzymes called DNA methyltransferases (DNMTs) which recognize CpG sites. There are several members of the DNMT family that all regulate DNA methylation (Jin et al.,2011). DNMT1 is localized to sites of replication during S phase and preferentially recognizes and methylates hemimethylated DNA. Therefore, DNMT1 is suggested to be the maintenance methyltransferase that confers an identical methylation pattern to daughter strands after replication. DNMT3A and DNMT3B are proposed to be the de novo methyltransferases because they preferentially methylate unmethylated DNA. Although DNMT3L shares homology with DNMT3A and DNMT3B, it has no catalytic activity by itself, instead it is involved in assisting with de novo methylation by increasing activity of the de novo methyltransferases and increasing methyl donor binding. More recently, however, there is growing evidence to suggest that the DNMTs described may cooperate together for both maintenance and de novo methylation. In addition to the activities of DNMTs, there is also a necessity to have demethylation of DNA. DNA demethylation may also be just as important during embryogenesis or gametogenesis as DNA methylation. Demethylation also plays a role in correcting methylation errors and remodelling methylation as a response to environmental stimuli (Zhu, 2009). The demethylation 31  process is not as well understood, but there are many proposed mechanisms. Demethylation can occur through both active or passive mechanisms. Active demethylation mechanisms are mediated by enzymatic reactions to remove 5-mCs, while passive mechanisms rely on DNA replication to dilute 5-mCs. One proposed active mechanism in humans is through base excision repair (BER) pathways. In order for 5-mCs to be recognized and replaced through the BER pathway, there needs to be a change in the base that would necessitate BER activity. This has been proposed to occur in at least two ways. First, the 5-mC undergoes deamination by 5-mC deaminases such as activation-induced deaminase (AID) or apolipoprotein B RNA-editing catalytic component (APOBEC) to produce thymine, resulting in a mismatch with the opposite DNA strand. Consequently, this is followed by glycosylase activity which recognizes the mismatch and cleaves the N-glycosidic bond between the ribose ring and thymine to produce an abasic site. This process is performed by TG mismatch glycosylases such as methylated DNA-binding protein (MBD4) or thymine DNA glycosylase (TDG). The abasic site is then replaced with an unmodified cytosine via the BER pathway (Zhu, 2009; Chen and Riggs, 2011). The second method involves the ten-eleven translocation (TET) family of enzymes, which are Fe(II)/α-ketoglutarate-dependent dioxygenases. The three members, TET1, TET2, and TET3, are involved in iterative oxidation of 5-mC into higher oxidation intermediates. 5-mC is first oxidized to 5-hydroxymethylcytosine (5-hmC), then to 5-formylcytosine (5-fC), and finally to 5-carboxylcytosine (5-caC) (Hill et al., 2014; Rasmussen and Helin, 2016). The two higher oxidation products, 5-fC and 5-caC, are recognized by TDG and replaced with an abasic site, which is subsequently replaced with cytosine through the BER pathway (Wu and Zhang, 2011; Kohli and Zhang, 2013; Weber et al., 2016). Additionally, passive replicative dilution may also be promoted because of the low activity of the maintenance DMNT1 on the higher oxidation 32  products of the TET enzymes (Kohli and Zhang, 2013; Hill et al., 2014; Xu and Wong, 2015; Rasmussen and Helin, 2016). 1.4.4.1.1 Genomic Imprinting A further complexity of DNA methylation involves imprinted genes. While most genes are expressed from both maternally and paternally inherited chromosomes, imprinted genes are only expressed from one allele. Although there are relatively few imprinted genes, they mostly function in fetal development, which makes them of particular interest when studying infertility. Imprinting is thought to have evolved from a conflict of interest between the maternal and paternal genomes in the offspring (Amor and Halliday, 2008; Peters, 2014). For the evolutionary fitness of the father, the paternal genome is interested in maximizing resource consumption of the offspring at any expense to produce larger offspring that will survive to reproduce. Whereas for the evolutionary fitness of the mother, the maternal genome is interested in conserving resources for the mother and all of her offspring (Amor and Halliday, 2008; Peters, 2014). Expression of just one of the two alleles is accomplished by differential methylation patterns at specific regions known as differentially methylated regions (DMRs). When a DMR controls several genes, it is called an imprinting control region (ICR). The monoallelic expression of imprinted genes is proposed to occur through the insulator model or the non-coding RNA (ncRNA) model (Plasschaert and Bartolomei, 2014; Weaver and Bartolomei, 2014). A representative domain for the insulator model is the IGF2/H19 locus, which possesses an intergenic ICR (Figure 1.10). In the IGF2/H19 locus, the paternal ICR is methylated and the maternal ICR is unmethylated. The methylation on the paternal allele physically blocks CCCTC binding factor (CTCF) from binding to the ICR, resulting in transcription enhancers skipping over H19 and instead expressing IGF2. On the maternal allele, lack of methylation allows CTCF 33  to bind to the ICR and prevents enhancers from accessing IGF2, resulting in H19 expression (Plasschaert and Bartolomei, 2014; Weaver and Bartolomei, 2014). The ncRNA model is best represented by the KCNQ1 locus. In this locus, the paternal ICR is unmethylated and the maternal ICR is methylated (Figure 1.10). An unmethylated ICR on the paternal allele allows transcription of the ncRNA KCNQ1OT1 that then silences all adjacent genes controlled by the ICR. Methylation of the KCNQ1 ICR on the maternal allele prevents transcription of the ncRNA and results in the expression of all adjacent genes (Plasschaert and Bartolomei, 2014; Weaver and Bartolomei, 2014). Thus, from these mechanisms, it is clear that monoallelic expression makes parental DNA functionally non-equivalent.  Figure 1.10 Mechanisms of genomic imprinting. A. Insulator model of genomic imprinting. DNA methylation (filled lollipops) on the ICR (green) of the paternal allele prevents CTCF from binding and results in IGF2 expression from the paternal allele. A lack of methylation on the maternal allele allows binding of CTCF to the ICR which blocks enhancers from accessing IGF2 and results in H19 expression from the maternal allele. B. ncRNA model of genomic imprinting. DNA methylation at the ICR of the maternal allele prevents transcription of the KCNQ1OT1 ncRNA and results in expression of all genes controlled by the ICR on the maternal allele (red). The lack of methylation on the paternal allele allows transcription of the KCNQ1OT1 ncRNA which represses expression of all genes controlled by the ICR on the paternal allele (grey). Red genes are expressed from the maternal allele, blue genes are expressed from the paternal allele, grey genes are repressed genes due to imprinting, white genes are constitutively expressed. Figure adapted from Weaver and Bartolomei, 2014.  34  Cellular differentiation is a hallmark of epigenetic mechanisms; however, the germ cells must undergo an epigenomic reset to erase any epigenetic memories through a process called epigenetic reprogramming (Messerschmidt et al., 2014; von Meyenn and Reik, 2015). Reprogramming is necessary to restore totipotency to the PGCs, since they arise sometime after fertilization from an already differentiated embryonic lineage which contains its own epigenetic signature (Smallwood and Kelsey, 2012). Regaining totipotency is a stepwise process that occurs in two major phases (Figure 1.11; Messerschmidt et al., 2014; Gkountela et al., 2015; Tang et al., 2015). First, PGCs undergo global DNA demethylation as they migrate through the hindgut to the genital ridge (section 1.1.1). This restores totipotency to the PGCs and allows them to be remethylated starting from a mostly unmodified genome (Smallwood and Kelsey, 2012). There are some regions that can escape reprogramming such that PGCs are never completely devoid of methylation; instead, they retain a <10% basal level of methylation compared to >70% overall in the embryo (Smallwood and Kelsey, 2012; Gkountela et al., 2015). After the events of sex differentiation of the gonads (section 1.1.2), PGCs acquire de novo methylation marks. At this point the PGCs are developing in different fetal environments due to the gonads now being differentiated. This leads to asymmetric remethylation patterns of germ cells that are specific to the sex of the embryo. In male embryos, de novo methylation begins in the mitotically arrested gonocytes. The process is completed before birth and therefore spermatogenesis uses germ cells which have been completely reprogrammed. In contrast, the process is substantially more complex in female embryos. Female embryos begin de novo methylation after birth when the oocytes have become meiotically arrested in prophase I. The complexity arises because the oocytes have already replicated their chromosomes and completed homologous recombination. Thus, to ensure fidelity of reprogramming, methylation must take place on replicated 35  chromosomes with chiasmata. The process is completed at puberty when oocytes resume meiosis I. The second major phase of epigenetic reprogramming concerns the new zygote post-fertilization. There is another wave of demethylation that occurs which is now also asymmetric. As the zygote replicates to form the blastocyst, DNA methylation is passively lost due to replicative dilution. In female zygotes, this is due to a lack of maintenance methylation, however, in male zygotes, the TET enzymes assist the process by oxidizing 5-mC (Wu and Zhang, 2011; Smallwood and Kelsey, 2012; Kohli and Zhang, 2013; Hill et al., 2014; Xu and Wong, 2015). This results in demethylation occurring much quicker in males than in females. Additionally, imprinted genes are protected from demethylation in both sexes. This results in imprinted gene methylation and expression patterns in the embryo being inherited from the gametes of the previous generation. Therefore, the consequences of abnormalities in imprinted genes can directly affect the offspring and will be discussed in the next section. Finally, after implantation, the embryo regains tissue-specific methylation as cell lineages begin to form.   36   Figure 1.11 Timeline of DNA methylation reprogramming. Human PGCs begin to appear around embryonic day 17 and are fully specified by embryonic day 24. PGCs undergo global DNA demethylation as they migrate to the forming genital ridges by embryonic day 37. After sex differentiation, male and female germ cells are reprogrammed asymmetrically. Male germ cells (blue line) are remethylated before birth while the cells are mitotically quiescent. Female germ cells (red line) are remethylated after birth during the time that the cells are meiotically arrested at prophase I. After fertilization, the zygote is demethylated through replicative dilution, which also occurs asymmetrically. The maternal complement of the genome is passively demethylated due to lack of maintenance methylation. The paternal genome is also passively demethylated but is assisted by TET oxidation of 5-mC. Imprinted genes (purple dotted line) are protected from demethylation and are a direct consequence of inheritance. After implantation of the blastocyst, remethylation occurs according to the new cell lineages being formed. Timeline not drawn to scale. Figure adapted from Smallwood and Kelsey, 2012 and Xu and Wong, 2015.  1.4.4.1.2  Epigenetic Inheritance Bypassing the natural barriers of infertility using ART may result in the direct transmission of epigenetic abnormalities since methylation of imprinted genes is not erased in the zygote. For example, inherited abnormalities in the IGF2/H19 locus can result in Beckwith-Wiedemann syndrome (BWS) or Silver-Russell syndrome (SRS). These imprinting disorders are broadly growth related. Other common imprinting disorders include Angelman syndrome (AS) and Prader-Willi syndrome (PWS), which are broadly development and intellect related, which are 37  caused by epigenetic abnormalities in the SNRPN locus (Peters, 2014; Plasschaert and Bartolomei, 2014). Defects in imprinted genes cause these disorders because the offspring receives either two active or two silenced copies of these genes when only one should be active and the other should be silenced. This may also result from uniparental disomy (UPD) which is a consequence of aneuploid gametes and causes both copies of the gene to come from one parent resulting in the same gene dosage as imprinted gene defects (Amor and Halliday, 2008). Defects in imprinted genes can also cause non-syndrome disorders such as intrauterine growth restriction (IUGR), which adversely affects fetal development (Peters, 2014). Some ART procedures (for example, superovulation or culture media) have been shown to affect imprinting in the gametes or embryo and can lead to imprinting defects (Doornbos et al., 2007; Manipalviratn et al., 2009; Chiba et al., 2013). Exposure of the fetus to any adverse environment, such as IUGR, malnutrition, or pollutants, can influence the development of later life morbidities such as obesity, diabetes, and cancer as proposed in the developmental origins of health and disease (DOHaD) theory (Heard and Martienssen, 2014). Investigations into the mechanisms of DOHaD must differentiate between intergenerational versus transgenerational effects to assess the risk of transmission. In intergenerational transmission, both the individual and the individual’s gametes are exposed to the insult, resulting in the F1 generation being affected because of the original insult to the parental gametes (Heard and Martienssen, 2014; Klengel et al., 2016). If the female is pregnant, the insult is extended to the gametes of the fetus in utero as well. Thus, true transgenerational effects are not seen until the F2 generation or F3 generation if the female is pregnant (Heard and Martienssen, 2014; Klengel et al., 2016). 38  1.4.4.2 Histone modifications Recent studies have also begun to investigate the significance of histones in infertility. Like any epigenetic mechanism, histone tail modifications are heritable and regulate gene expression. Modifications are both subunit and amino acid specific and most commonly involve acetylation, phosphorylation, and methylation of the N-terminal tails (Jin et al.,2011; Stuppia et al., 2015). Modifications that neutralize the positive charge of histones can induce structural changes in the chromatin (Vaissière et al., 2008; Bannister and Kouzarides, 2011). As a consequence, the electrostatic interaction between the histones and the negatively charged DNA can be regulated to increase or decrease access to transcriptional machinery as necessary. Alternatively, modifications can also modulate interactions with chromatin binding factors to elicit transcriptional activation or inhibition (Bannister and Kouzarides, 2011). For example, histone modifications have been proposed to directly interact with DNA methylation machinery. Activating histone marks such as H3K4me2/3 inhibit DNMT interaction and therefore prevents DNA methylation. Conversely, repressive histone marks such as H3K9me2/3 or H3K27me3 recruit DNMTs resulting in subsequent DNA methylation (Lennartsson and Ekwall, 2009; Jin et al.,2011; Oliva and Luís Ballescà, 2012; Smallwood and Kelsey, 2012). Additionally, as mentioned previously, histones are replaced by protamines during spermiogenesis. Protamines are more positively charged than histones due to the high arginine content, which allows a greater level of compaction required for sperm motility, protection from oxidation, and inhibition of transcription (Miller et al., 2010; Oliva and Luís Ballescà, 2012; Stuppia et al., 2015; Gunes et al., 2016). The process begins with the hyperacetylation of histone tails that loosens the chromatin structure. This allows histones to be replaced by the transition proteins TP1 and TP2, which assist in the removal of histones and subsequently the compaction 39  by protamines. The transition proteins are then replaced by two different protamines P1 and P2 that result in the final condensed sperm genome in the mature spermatozoa (Stuppia et al., 2015; Gunes et al., 2016). The ratio of P1/P2 protamines in fertile men is around one and deviations from this ratio have been observed in infertile men. Abnormal ratios have also been associated with higher levels of DNA damage, suggesting inadequate DNA condensation (Miller et al., 2010; Stuppia et al., 2015; Gunes et al., 2016). Furthermore, only 85-95% of histones are replaced by protamines (Oliva and Luís Ballescà, 2012; Stuppia et al., 2015; Gunes et al., 2016). The regions of DNA which retain histones are not random and are believed to confer epigenetic information to the embryo (Oliva and Luís Ballescà, 2012). 1.4.4.3 Non-coding RNAs There are many varieties of ncRNAs whose functions are the subject of ongoing research. Some classifications of ncRNAs have been found to have gene regulatory functions and therefore are of interest in the context of epigenetics. These include microRNAs (miRNAs), small-interfering RNAs (siRNAs), and piwi-interacting RNAs (piRNAs) (Nanassy and Carrell, 2008). In the nucleus, miRNAs are first transcribed from DNA into pri-miRNAs which are then processed by the type III RNase Drosha into pre-miRNAs. Upon export into the cytoplasm, pre-miRNAs are processed by another type III RNase Dicer before one of the strands is incorporated into the RNA-induced silencing complex (RISC). Gene regulation is accomplished by complementary base pairing of RISC and the target messenger RNAs (mRNAs). Depending on the level of complementarity, the mRNA is either degraded or translationally repressed (Gunes et al., 2016). siRNAs are processed and function in a similar manner, however, piRNAs do not go through Dicer processing and also interact with P-element-induced wimpy (PIWI) proteins to produce their gene regulatory function (Peschansky and Wahlestedt, 2014; Gunes et al., 2016). 40  All of these ncRNAs have been found to be expressed in male germ cells and are reported to have important functions during spermatogenesis. Microarray studies have shown that these ncRNAs are dysregulated in infertile men compared to fertile controls. More specifically, several miRNAs have been shown to upregulated or downregulated in the testes, spermatozoa, and even in the seminal fluid (Gunes et al., 2016). 1.5 One-Carbon Metabolism The methyl donors used as substrates in the methylation of DNA or histones are derived from the folate and methionine cycles (Figure 1.12). The folate and methionine cycles also provide substrates for purine synthesis and glutathione synthesis, respectively. Folic acid enters the cell and is sequentially reduced by dihydrofolate reductase (DHFR) to dihydrofolate (DHF) and then tetrahydrofolate (THF). THF can then receive a methyl group from either serine or glycine to form 5,10-methyleneTHF (5,10-mTHF). Serine donates a methyl group to THF by using serine hydroxymethyl transferase (SHMT), whereas glycine donates through glycine decarboxylase (GLDC). Through a series of reactions, 5,10-mTHF can be converted to 10-formyltetrahydrofolate (fTHF) for purine synthesis. For methylation, 5,10-mTHF is converted to 5-methylTHF (5-mTHF) by methylenetetrahydrofolate reductase (MTHFR). Completion of the folate cycle occurs when the methyl group is donated to homocysteine (hCYS) to generate the original THF and methionine (MET). This reaction is the result of methionine synthase (MS) and its vitamin B12 cofactor. MET goes on to produce S-adenosylmethionine (SAM) after transfer of adenosine from adenosine triphosphate (ATP) by methionine adenyltransferase (MAT). SAM is the methyl donor used for methylation reactions. Once the methyl group is donated, S-adenosylhomocysteine (SAH) is formed. SAH can then be deadenylated by S-adenosylhomocysteine hydrolase (SAHH) to form the original hCYS and complete the 41  methionine cycle. Additionally, hCYS can be shunted to the transsulfuration pathway when it is condensed with serine by cystathionine synthase (CBS) to form cystathionine. Cystathionine undergoes cleavage by cystathionine lyase (CGL) to produce α-ketobutyrate (αKB) and cysteine. Finally, cysteine goes on through a series of reactions to produce glutathione which is important for regulating the redox status of cells (see next section) (Locasale, 2013; Yang and Vousden, 2016).   42   Figure 1.12 Metabolic pathways of one-carbon metabolism. In the folate cycle, folic acid enters the cell and is converted to THF through the DHFR enzyme. Serine can donate a methyl group using SHMT or glycine can donate using GLDC. Methyl group donation produces 5,10-mTHF which is a precursor to purine synthesis. 5,10-mTHF is reduced by MTHFR to produce 5-mTHF. MS/B12 takes the methyl group from 5-mTHF and transfers it to hCYS to generate THF and MET. In the methionine cycle, MAT uses ATP to adenylate MET to produce the methyl donor SAM. Upon methyl donor reactions, SAH is generated. SAH undergoes deadenylation by SAHH to produce hCYS. In the transsulfuration pathway, hCYS is condensed with serine by CBS to produce cystathionine. Cleavage of cystathionine by CGL produces αKB and cysteine, which is a precursor to glutathione. Green boxes indicate inputs, red boxes indicate outputs, gray boxes indicate pathway intermediates. Figure adapted from Locasale, 2013 and Yang and Vousden, 2016.  Important enzymes in the generation of methyl donors are MTHFR, MS, and methionine synthase reductase (MTRR). The function of MTRR is closely associated with MS. MS uses a vitamin B12 cofactor which contains a cobalt ion to transfer methyl groups. The reduced form of 43  cobalt [Co(I)] is used to accept the methyl group from 5-mTHF and consequently forms MS-Co(III)-CH3. The MS-Co(III)-CH3 can then donate the methyl group to hCYS to form MET and return the cofactor to Co(I) to be used in the next cycle (Leclerc et al., 1998; Olteanu and Banerjee, 2001; Olteanu et al., 2002; Wolthers and Scrutton, 2007). However, after several hundred cycles, Co(I) can spontaneously lose an electron resulting in its oxidation to Co(II) and, consequently, MS will lose its activity. The function of MTRR is to catalyze the reductive activation of MS/B12. MTRR uses the reduced nicotinamide adenine dinucleotide phosphate (NADPH) as an electron donor and SAM to reform MS-Co(III)-CH3, which can then be used to transfer methyl groups once again. Single nucleotide polymorphisms (SNPs) are known to adversely affect the activity of these enzymes and may also be a risk factor for male infertility (Liu et al., 2015). 1.6 DNA Integrity Transmission of intact DNA from the sperm to the embryo is an essential requirement for normal fertilization and development. Sperm with high levels of damaged DNA is associated with infertility and recurrent pregnancy loss as well as adverse pregnancy outcomes and childhood morbidity (Bisht et al., 2017). Evidence suggests that DNA damage in the sperm may account for up to 80% of the pathology of infertility in men. DNA damage has been suggested to be a result of abortive apoptosis in which the process began in early spermatogenesis but did not run to completion by the time that the major remodelling during spermiogenesis occurred and the necessary apoptotic machinery was removed as a result (Koppers et al., 2011). Alternatively, oxidative stress can also be a source of DNA damage (Aitken et al., 2003). Oxidative stress occurs when cells no longer have the antioxidant capacity to manage the excess production of reactive oxygen species (ROS), such as superoxide anions (O2•), hydrogen 44  peroxide (H2O2), and hydroxyl radicals (OH•) (Bisht et al., 2017). ROS can come from exogenous sources or they can be generated intracellularly. Exogenous sources include smoking, excessive alcohol consumption, ionizing radiation, and environmental chemicals such as solvents and pesticides (Aitken et al., 2003; Bisht et al., 2017; Fullston et al., 2017). Endogenously, ROS generation occurs in leukocytes from seminal plasma and the mitochondria in the sperm. Physiologically, a low level of ROS generation is necessary for various redox pathways such as oxidative phosphorylation used in the production of ATP. Additionally, capacitation and hyperactivation are ROS-dependent processes specific to the sperm, which are both requirements for normal oocyte fusion (Aitken et al., 2003; Bisht et al., 2017). Leukocytes in the seminal plasma are capable of generating levels of ROS that are orders of magnitude higher than in spermatozoa as a response to infection or inflammation, thus rapid treatment will minimize the effects of oxidative stress. One pathological mechanism that could explain excess ROS and oxidative stress is the retention of excess cytoplasm during spermiogenesis. Proper maturation of the spermatozoa requires the removal of excess cytoplasm by the Sertoli cell during the final stages of spermiogenesis. However, if the process fails, the defective spermatozoa are released with an excess of cytoplasm, which contains enzymes that generate ROS (Aitken et al., 2003). The spermatozoa also have limited antioxidant capacities and DNA repair mechanisms, which leaves them vulnerable to the effects of ROS damage and oxidative stress. ROS can affect sperm cells by oxidizing the proteins and lipids present on the plasma membrane, resulting in decreased plasma membrane fluidity and sperm motility, ultimately resulting in reduced fertilization potential (Aitken et al., 2003; Bisht et al., 2017). More importantly, ROS can also damage the DNA of the sperm, both in the nucleus and mitochondria. Mitochondrial DNA is far more susceptible to damage than nuclear DNA due to the lack of 45  protective histones/protamines and base repair mechanisms. With respect to sperm nuclear DNA, the protamines associated with the majority of the genomes offers some protection against ROS-mediated damage, however, the regions that remain associated with histones are most vulnerable to damage. ROS damage to DNA can manifest in several ways, including DNA strand breaks, DNA fragmentation, and base modifications, all of which have the potential to disrupt transcription and contribute to genomic instability. ROS damaged DNA has been suggested to be an inadequate substrate for DNMTs leading to a disruption in methylation (Tunc and Tremellen, 2009). ROS can also decrease the number of mature sperm available due to apoptosis resulting in oligospermia and infertility. Telomeres are made up of TTAGGG repeats which are bound by a multiprotein complex called shelterin to form a protective cap on the ends of chromosome arms (Blasco, 2007; Kalmbach et al., 2013; Bisht et al., 2017). During every round of cell division, the ends of the chromosomes become a little shorter because the replication machinery fails to replicate the very ends of the chromosomes. The telomeres function to protect the gene coding portion of the DNA by acting as a buffer to the end-replication loss. However, if telomeres become shorter than a critical length, the cells will undergo apoptosis. Telomere length can be maintained or lengthened by telomerase which uses a reverse transcriptase and RNA template to add de novo repeats to the chromosome ends. In cells which do not express telomerase, telomeres are maintained using homologous recombination between telomeric regions. In addition to the end-replication loss of telomeric repeats, ROS can also result in shortened telomeres by oxidizing the guanine-rich repeats. Guanine is particularly susceptible to oxidation because of its low oxidation potential and will result in 8-hydroxy-deoxyguanine (8-OHdG) adducts which leads to removal of the telomeric repeat by cellular repair mechanisms. 46  The cell’s defence against ROS includes several different antioxidant enzymes and biomolecules. For example, the antioxidant enzymes superoxide dismutase and catalase are able to reduce superoxide and hydrogen peroxide into water (Gorrini et al., 2013; Bisht et al., 2017). There is also a class of antioxidant enzymes called glutathione S-transferases (GSTs) which use the abundant biomolecule glutathione (GSH) as an antioxidant. GSH is composed of the amino acids glutamate and cysteine and becomes oxidized to glutathione disulfide (GSSG) upon ROS reduction. GSH is also used as a detoxification enzyme in the metabolism of drugs and environmental toxins (Buchard et al., 2007). 1.7 Rationale Infertility is becoming a growing concern in modern society with an estimated 8-12% of couples globally having problems conceiving (Bisht et al., 2017). Reports consistently attribute the cause of infertility being of male origin in 50% of the cases (Bisht et al., 2017; Miyamoto et al., 2015). Although there are several genetic alterations known to cause male factor infertility, this only accounts for approximately 15-30% of cases with the remaining cases of male infertility being of unknown origin (Ferlin et al., 2006; Georgiou et al., 2006; Neto et al., 2016). Recently, there has been great interest in studying epigenetics for a potential mechanism to explain idiopathic cases of male infertility. A common symptom of male infertility is the reduced quality of sperm which usually involves impaired spermatogenesis. Sperm quality is assessed as part of infertility treatment and several parameters, including sperm count, motility, and morphology, are often reduced in infertile men. Several studies by our lab and others have also reported that the sperm of infertile men show DNA methylation profiles that differ significantly compared to fertile men, particularly at imprinted genes (Marques et al. 2004; Kobayashi et al., 2007; Marques et al. 47  2008; Filipponi and Feil, 2009; Li et al., 2013; Urdinguio et al., 2015; Louie et al., 2016). These findings may be particularly worrisome for infertile couples choosing to undergo ICSI to conceive as the chosen sperm will artificially bypass natural selection barriers and potentially transmit methylation defects to the offspring. Indeed, several studies have looked at DNA methylation and ART use, however, the findings are generally contradicting and thus no concrete conclusions can be made regarding the use of aberrantly methylated sperm (Kobayashi et al., 2007; Montjean et al., 2013). Despite the lack of conclusions, these studies continue to add evidence for an association between DNA methylation and infertility. However, whether DNA methylation is a cause or a secondary effect of some other mechanism is unknown. Here, we aim to investigate DNA methylation and factors that may affect aberrant DNA methylation in infertile men. As described earlier, DNA methylation is dependent on DNMT enzymes and the cellular availability of methyl donors, which are primarily produced via one-carbon metabolism. There are several enzymes of the folate cycle that convert dietary folate into the methyl donor SAM through a series of enzymatic steps. These enzymes are known to have SNPs that can affect enzyme activity and there have been several studies that have shown disrupted methylation as a result (Chen et al., 2001; Friso et al., 2002; Castro et al., 2004). However, these studies only focused on SNPs in MTHFR and its effect on global DNA methylation. Although there are studies which show an association between SNPs in the various enzymes and male infertility (Mfady et al., 2014; Liu et al., 2015; Karimian and Hosseinzadeh Colagar, 2016), there is only one group to study the effects of an MTHFR SNP on the sperm DNA methylation of infertile men (Aarabi et al., 2015). Here, we intend to extend the analysis to genotypes of MS and MTRR and their effects on sperm DNA methylation of infertile men. 48  The association between DNA methylation and DNA damage is less clear. As mentioned above, DNA strand breaks or base adducts can interfere with binding to DNMTs and thus produce an adverse effect on DNA methylation. Of the few studies that have been done, there is generally an inverse relationship between DNA methylation and DNA damage (Tavalaee et al., 2009; Tunc and Tremellen, 2009; Montjean et al., 2015). However, these studies only looked at global methylation level and did not assess this relationship in imprinted genes. Additionally, the antioxidant GST enzymes use GSH to reduce ROS levels and potentially protect DNA from damage. The effect of GST null genotypes on the level of DNA damage and DNA methylation has not previously been studied. 1.7.1 Hypotheses and Objectives Investigation of DNA methylation in infertile men Hypothesis 1a. Men with oligospermia will display greater abnormalities in DNA methylation compared to fertile control men. Hypothesis 1b. Men with azoospermia will display greater abnormalities in DNA methylation compared to fertile men undergoing vasectomy reversals. Objective 1. Determine the mean DNA methylation of four imprinted genes H19, GTL2, MEST, and LIT1, and global DNA methylation using LINE1 by bisulfite pyrosequencing the sperm and compare the values to the relevant control groups. Investigation of the folate cycle in infertile men Hypothesis 2a. Men with lower folate concentrations will have lower levels of DNA methylation. 49  Hypothesis 2b. Men with SNPs in the folate cycle genes will have lower levels of DNA methylation. Objective 2a. Determine the folate concentration in peripheral blood. Objective 2b. Determine the genotypes of MTHFR, MS, and MTRR genes of the folate cycle. Objective 2c. Assess any effects folate concentration and genotype may have on DNA methylation. Investigation of sperm DNA integrity in infertile men Hypothesis 3a. Men with GST null alleles will have higher levels of DNA fragmentation. Hypothesis 3b. Men with higher levels of DNA fragmentation will have lower levels of DNA methylation. Objective 3a. Determine the level of DNA fragmentation in the sperm of infertile men. Objective 3b. Determine the genotypes of GST genes. Objective 3c. Assess any effects DNA fragmentation and GST genotypes may have on DNA methylation.   50  CHAPTER 2: INVESTIGATION OF DNA METHYLATION IN INFERTILE MEN 2.1 Introduction Infertility is becoming increasingly problematic and is suspected to affect 8-12% of couples worldwide, with male factor being implicated in approximately 50% of these cases (Bisht et al., 2017). A large majority of cases of male infertility are diagnosed as idiopathic (Georgiou et al., 2006). DNA methylation is an epigenetic mechanism that has long been studied for a potential explanation of idiopathic male infertility. DNA methylation is an epigenetic mechanism that allows a cell to differentially express its complement of genes to give rise to specific tissue lineages (Bird, 2002). DNA methylation also plays an important role in imprinted genes. These genes are uniquely expressed from only one parental allele and are critical for the proper growth and development of the embryo and placenta (Peters, 2014). Although there are at least 100 imprinted genes identified in humans, the functions of only a handful have been characterized and are consequently the most consistently studied (Peters, 2014). Sperm parameters are commonly reduced in infertile men, leading many to associate abnormal sperm parameters and DNA methylation (Filipponi and Feil, 2009). Early studies by Marques et al. showed that abnormal imprinting of H19 and MEST was associated with oligospermia (Marques et al., 2004; Marques et al., 2008). Normally in the sperm, H19 is methylated and MEST is unmethylated, but Marques et al. found hypomethylation of H19 and hypermethylation of MEST in men with oligospermia. These findings were supported by our previous study and others (Kobayashi et al., 2007; Kobayashi et al., 2009; Poplinski et al., 2010; Sato et al., 2011; Li et al., 2013; Montjean et al., 2013; Laurentino et al., 2015; Louie et al., 2016). Most of these studies, however, used molecular cloning techniques to look at only a limited number of individual sperm in each patient. Few groups also looked into the paternally 51  imprinted GTL2 and maternally imprinted LIT1 genes; furthermore, of the studies available, the results are conflicting. Even fewer studies looked at men with azoospermia due to the difficulty in retrieving sperm from these men (Marques et al., 2010; Minor et al., 2011). Using a proper control group for these men is also challenging. In our previous study, we observed hypomethylation of H19, but no methylation defects in MEST or GTL2 (Minor et al., 2011). Thus, among these studies of imprinted gene defects, only H19 hypomethylation is observed consistently. In the present study, we measure methylation at two paternally methylated genes, H19 and GTL2, and two paternally unmethylated genes, MEST and LIT1 using bisulfite pyrosequencing. We will also measure methylation at LINE1 repetitive elements for a surrogate measure of global methylation level. The groups of infertile men will include oligospermia, subcategorized by sperm count, and azoospermia, subcategorized by pathology. From the data obtained, we can determine whether aberrant methylation affects the whole genome or is isolated to imprinted genes. Additionally, we can determine if abnormalities occur equally between paternal and maternal imprinted genes. 2.2 Materials and Methods 2.2.1 Sample Collection We collected samples from infertile men of varying infertile phenotypes. These men were undergoing fertility evaluation at fertility clinics in the Metro Vancouver Regional District. Ejaculate samples were donated after three days of abstinence from patients diagnosed with low sperm concentration according to the WHO’s 5th centile lower reference limit (WHO, 2010). We further assessed each semen sample for concentration and motility in our lab using the Makler counting chamber (Sefi Medical Instruments, Haifa, Israel). Semen samples were allowed to 52  liquefy for 30 minutes in a 37°C incubator before assessment. A total of 29 ejaculate samples were determined to be below the WHO reference limit and were included in this study. Eighteen of these men were further categorized as oligospermia (O; 5-15 million sperm/mL) and eleven men were categorized as severe oligospermia (SO; <5 million sperm/mL). The ejaculate from twenty proven fertile men was used as controls; these men have naturally conceived a pregnancy within twelve months of recruitment. Testicular samples were collected from eleven men undergoing testicular biopsies for sperm retrieval to use in ARTs. A small portion of the biopsied testicular tissue was donated for our study. These men were further categorized based on their infertility diagnosis as reported by their physicians: seven men were diagnosed with obstructive azoospermia (OA) and four men were diagnosed with non-obstructive azoospermia (NOA). Additionally, since there are inherent differences in the source of the ejaculate and testicular sperm, we used the testicular tissue of eight men undergoing vasectomy reversals (VR) as a control group for the testicular samples. These men have previously fathered a child prior to getting a vasectomy. All patients included in this study had normal 46, XY karyotypes, an absence of microdeletions on the Y chromosome, and no CFTR mutations. 2.2.2 Sperm Cell Isolation After assessing semen parameters, ejaculate samples were divided into 1.5 mL microcentrifuge tubes. For testicular samples, testicular tissue was minced in modified human tubal fluid (mHTF) (Vitrolife, San Diego, CA, USA) in a petri dish using fine forceps to separate the seminiferous tubules. Tubules were then minced using the forceps to extract the contents. The tubules and contents in the mHTF were then transferred to a 1.5 mL microcentrifuge tube. 53  All samples, whether ejaculate or testicular in origin, were then washed in mHTF to remove any seminal plasma or testicular debris. Samples were then centrifuged for two minutes and the supernatant was removed. Samples were washed in this manner an additional two times and finally resuspended in mHTF. A 10 µL droplet of the washed sample was placed in the center of a small petri dish and two drops of mHTF were added using a glass Pasteur pipette to dilute. Using a pipette tip, the sample was spread around in the center of the dish to flatten the droplet. Several 10 µL of clean mHTF droplets were placed around the main sample and the entire petri dish was filled with enough mineral oil to submerge all droplets to prevent evaporation. Sperm cells from each sample were isolated using an inverted microscope (Nikon, Tokyo, Japan) and micromanipulators (Narishige, Tokyo, Japan). The microscope was equipped with Hoffman modulating optics to assist in identifying sperm cells and a thermal stage to keep the sample at 37°C. Custom made glass micropipettes were produced from glass capillary tubes for use in the micromanipulators using a microforge (Narishige). Isolated sperm cells were transferred to one of the clean droplets of mHTF surrounding the sample. Sperm cells were carefully isolated such that other cells, debris, or other contaminants were not transferred. After approximately 200 sperm cells were isolated per sample, the mHTF droplet containing the pure sperm cells was transferred to a 0.5 mL microcentrifuge tube using a 10 µL micropipette. A thorough transfer was confirmed afterward by visually checking the remnants of the transferred droplet. Alternatively, if the sample was determined to have adequate concentration and motility, the sample set up for swim-up procedure to isolate the sperm cells. After washing and resuspending in mHTF, the sample was centrifuged for two minutes to pellet the sperm cells. 54  The tube was then placed in a 37°C incubator at an approximately 45° angle and incubated for at least one hour. Swim-up was checked by taking 10 µL of medium slightly above the pellet and placing on a microscope slide to visually assess for concentration and absence of debris or other contaminants. Care was taken to keep the angle of the tube constant and not to disturb the pellet. Another 10 µL of medium was taken above the pellet and transferred to a 0.5 mL microcentrifuge tube if the sample was determined to be clean and had adequate sperm cells present on the microscope slide. Samples were left to incubate longer before isolation if required. 2.2.3 Cell Lysis Sperm cells were lysed using an alkaline lysis method (Manning, 2001). 10 µL of alkaline lysis buffer (ALB) containing 200 mM KOH (Sigma-Aldrich, Oakville, ON, Canada) and 50mM of dithiothreitol (DTT) (Sigma-Aldrich) was added to the tubes of pure sperm cells. The tubes were immediately frozen at -20°C for two days. Tubes were then placed in a heat block at 80°C for 15 minutes to finish cell lysis. 10 µL of neutralization buffer containing 22.5 mM Tris-HCl (Invitrogen, Burlington, ON, Canada), 0.3 M KCl (Fisher Scientific, Ottawa, ON, Canada), and 0.2 M HCl (Fisher Scientific) was then added to neutralize the ALB. 2.2.4 Bisulfite Conversion and DNA Purification Released DNA was then subjected to bisulfite modification using the EZ DNA Methylation-Gold Kit (Zymo Research, Irvine, CA, USA), according to the manufacturer’s protocol. The protocol also included DNA purification via spin columns as part of the kit. Purified bisulfite treated DNA was then frozen at -20°C until later usage. Sodium bisulfite deaminates cytosine residues to form uracil, while 5-methylcytosines are essentially non-reactive (Frommer, 1992). Following bisulfite treatment, PCR is performed to amplify the region(s) of interest, where the uracil residues (formerly cytosine residues) are paired with thymine. 55  Therefore, downstream sequencing and analysis will be able to identify and differentiate cytosines and 5-methylcytosines in the original DNA sequence. 2.2.5 PCR Amplification In this study, we analysed four genomic regions and one transposable element found throughout the genome. For all regions, PCR was carried out in 25 µL reactions as follows: 1x PCR buffer, containing 15 mM MgCl2 (Qiagen, Toronto, ON, Canada), 0.1 mM dNTPs (Invitrogen), 0.4 µM forward and reverse primers, 1 U HotStarTaq DNA polymerase (Qiagen), and 1 µL bisulfite modified DNA. The primers used differ depending on the region analysed and are shown in Table 2.1. One primer of each pair is biotinylated (/5Biosg/) in order to be captured for pyrosequencing in the following section. PCR reactions were run using the following cycling conditions: initial denaturation at 95°C for 15 min; 49 cycles of 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min; and a final extension at 72°C for 10 min. Each region analysed also included an amplification using the 100% methylated EpiTect PCR Control DNA (Qiagen) as a positive control and a DNA free amplification as a negative control.   56  Table 2.1 Primer sequences used in methylation analysis. Region Type Sequence (5′ - 3′) H19 Forward TGAGTGTTTTATTTTTAGATGATTTT  Reverse /5Biosg/ACAATACAAACTCACACATCACAAC  Sequencing GTGGTTTGGGTGATT GTL2 Forward ATTGAATTGGGTTTGTTAGTAGT  Reverse /5Biosg/TCAAAACAACTCAAATCCTTTATA  Sequencing TGAATTGGGTTTGTTAGTAG MEST Forward /5Biosg/TAGGTGAGTGTGYGGTGGG  Reverse TAACACCCCCTCCTCAAATAAACA  Sequencing AAAAACAACCCCAACT LIT1 Forward /5Biosg/GTTGTYGTTTAATTAGTAGGTGG  Reverse AAATCTTACTAAAAAACTCCCTAA  Sequencing CTAAAAAACTCCCTAAAAAT LINE1 Forward TTTTGAGTTAGGTGTGGGATATA  Reverse /5Biosg/AAAATCAAAAAATTCCCTTTC  Sequencing AGTTAGGTGTGGGATATAGT /5Biosg/ = 5′ biotin modification  2.2.6 Pyrosequencing Amplicons of the correct size were verified before running on the pyrosequencer by agarose gel electrophoresis. 5 µL of PCR product was mixed with 6x loading dye (Thermo Fisher Scientific, Burlington, ON, Canada) before running on a 2% agarose gel (Invitrogen) in 1x TAE [containing 40 mM Tris (Invitrogen), 20 mM acetic acid (Fisher Scientific), and 1 mM EDTA (Sigma-Aldrich)] with SYBR Safe DNA staining (Invitrogen) and 5 µL of 100bp DNA ladder (Invitrogen). Gels were run for one hour at 100 V before being visualized using a UV transilluminator.  57  Sequencing was performed on the PyroMark Q96 MD instrument (Qiagen) according to the manufacturer’s protocol. Briefly, PCR products were added to a mixture of binding buffer (Qiagen) and streptavidin sepharose high-performance beads (GE Healthcare, Mississauga, Ontario, Canada) in a semi-skirted 96-well PCR plate (Sarstedt, Montréal, QC, Canada). The beads will only capture biotin labelled PCR products. Another negative control containing only the bead solution and no PCR product was also added to the plate. The PCR plate was placed on a plate shaker to prevent beads from settling. 0.3 µM of the appropriate sequencing primers (Table 2.1) were added together with annealing buffer (Qiagen) into a pyrosequencing plate (Qiagen). Using the vacuum tool, the beads were aspirated out of the PCR plate and washed with 70% ethanol, denatured with NaOH solution (Fisher Scientific), and finally washed with wash buffer (Qiagen). Turning off the vacuum releases the beads into the pyrosequencing plate containing the primers. The plate was then heated to 80°C for two minutes and then allowed to cool to room temperature to denature and reanneal primers and amplicons. The plate was then placed in the PyroMark Q96 MD instrument for processing. PyroMark Gold Q96 Reagents (Qiagen) were reconstituted and the appropriate volumes were added to either the PyroMark Q96 HS Reagent Tips or the PyroMark Q96 HS Capillary Tips (Qiagen). Dispensation tips were tested before every run to ensure proper alignment with the wells of the plate. Assays designed specifically for each analysed region were then run on the instrument. 2.2.7 Data Analysis Pyrograms were analysed with the included Pyro Q-CpG Software (Qiagen). Only samples that passed the software’s quality control checks were included in the analysis. Average percentage methylation values given were based on the methylation of CpG sites within each 58  analysed region. These values were exported from the software to be analysed in R: A language and environment for statistical computing (v3.3.3). Mean methylation values were determined to be not normally distributed (p < 0.05; Shapiro-Wilk Normality test) or not equal in variances (p < 0.05; Levene’s test). Therefore, the mean methylation values were compared overall within a gene using the non-parametric Kruskal-Wallis test and between each group using the non-parametric Mann-Whitney U test with the false discovery rate (FDR) correction for multiple comparisons. FDR corrections were used as they are less stringent than other corrections such as the Bonferroni method. 2.3 Results 2.3.1 H19 Gene We investigated DNA methylation of the paternally imprinted H19 gene by bisulfite pyrosequencing. There were four CpG sites studied within this gene. The values shown in Table 2.2 represent the average of these CpG sites within each patient studied. In this gene, there was no significant difference in the overall comparison of groups (p > 0.05; Kruskal-Wallis test), however, men with severe oligospermia were found to be significantly different from fertile control men after pairwise group comparisons (p = 0.01378; Mann-Whitney test; Figure 2.1).  59  Table 2.2 Summary of H19 methylation. Control Oligospermia Severe Oligospermia Vasectomy Reversal Obstructive Azoospermia Non-Obstructive Azoospermia C01 90.80 O01 81.56 SO01 57.17 VR01 93.16 OA01 84.82 NOA01 97.29 C02 71.43 O02 92.88 SO02 62.05 VR02 92.74 OA02 91.16 NOA02 71.30 C03 85.24 O03 95.76 SO03 90.64 VR03 65.16 OA03 34.79 NOA03 96.82 C04 75.60 O04 78.25 SO04 53.69 VR04 96.05 OA04 86.91 NOA04 53.74 C05 89.49 O05 97.27 SO05 71.09 VR05 55.67 OA05 60.23   C06 86.78 O06 36.05 SO06 81.39 VR06 73.98 OA06 94.63   C07 83.32 O07 58.48 SO07 65.93 VR07 83.45 OA07 96.56   C08 88.98 O08 68.67 SO08 85.76 VR08 82.31     C09 89.18 O09 49.69 SO09 81.75       C10 90.54 O10 53.92 SO10 71.90       C11 93.83 O11 97.28 SO11 91.22       C12 93.93 O12 90.49         C13 91.65 O13 97.58         C14 91.29 O14 96.33         C15 93.98 O15 75.39         C16 91.52 O16 78.37         C17 89.00 O17 90.23         C18 91.14 O18 77.99         C19 90.96           C20 90.85           Mean 88.48  78.68  73.87  80.32  78.44  79.79 Median 90.67  79.97  71.90  82.88  86.91  84.06  60   Figure 2.1 Mean DNA methylation of H19 by pyrosequencing. Mean DNA methylation of the paternally imprinted H19 gene. C, control men (n = 20); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 11); VR, vasectomy reversal men (n = 8); OA, obstructive azoospermic men (n = 7); NOA, non-obstructive azoospermic men (n = 4). Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range. Asterisks represent significant differences compared to controls (p < 0.05; Mann-Whitney test).  2.3.2 GTL2 Gene For the paternally imprinted GTL2 gene, we studied five CpG sites. The values shown in Table 2.3 represent the average of these CpG sites within each patient studied. In this gene, we did not find any groups to be significantly different when compared overall (p > 0.05; Kruskal-Wallis test) or pairwise (p > 0.05; Mann-Whitney test; Figure 2.2). 61  Table 2.3 Summary of GTL2 methylation. Control Oligospermia Severe Oligospermia Vasectomy Reversal Obstructive Azoospermia Non-Obstructive Azoospermia C01 95.72 O01 86.76 SO01 83.66 VR01 82.66 OA01 40.00 NOA01 94.15 C02 95.76 O02 87.11 SO02 67.16 VR02 96.60 OA02 96.93 NOA02 92.64 C03 95.07 O03 94.53 SO03 92.11 VR03 68.93 OA03 47.01 NOA03 62.87 C04 64.72 O04 97.05 SO04 90.18 VR04 69.53 OA04 95.04 NOA04 76.52 C05 94.38 O05 92.73 SO05 93.24 VR05 1.88 OA05 97.76   C06 84.74 O06 75.28 SO06 98.04 VR06 92.80 OA06 97.59   C07 95.42 O07 76.59 SO07 92.04 VR07 98.11 OA07 96.71   C08 95.48 O08 97.14 SO08 77.06 VR08 89.34     C09 94.97 O09 89.25 SO09 80.30       C10 91.53 O10 87.63 SO10 96.70       C11 95.81 O11 77.57 SO11 58.62       C12 94.48 O12 87.29         C13 93.90 O13 88.62         C14 90.37 O14 83.73         C15 91.96 O15 80.00         C16 94.14 O16 87.74         C17 90.30 O17 92.70         C18 93.46 O18 83.45         C19 94.11           C20 91.17           Mean 91.87  86.95  84.46  74.98  81.58  81.55 Median 94.13  87.46  90.18  86.00  96.71  84.58  62   Figure 2.2 Mean DNA methylation of GTL2 by pyrosequencing. Mean DNA methylation of the paternally imprinted GTL2 gene. C, control men (n = 20); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 11); VR, vasectomy reversal men (n = 8); OA, obstructive azoospermic men (n = 7); NOA, non-obstructive azoospermic men (n = 4). Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  2.3.3 MEST Gene For the maternally imprinted MEST gene, we studied eight CpG sites. The values shown in Table 2.4 represent the average of these CpG sites within each patient studied. In this gene, we found that methylation was significantly different in an overall comparison of groups (p = 0.000004827; Kruskal-Wallis test) and also all groups were significantly different when compared to fertile control men in a pairwise comparison (C vs. O, p = 0.001870; C vs. SO, p = 0.0002466; C vs. VR, p = 0.000009652; C vs. OA, p = 0.0007432; C vs. NOA, p = 0.003388; Mann-Whitney test; Figure 2.3). However, there were no significant differences found between the VR group and either azoospermia groups (VR vs. OA or NOA; p > 0.05). Some samples repeatedly failed pyrosequencing at some CpG sites for this gene and were excluded in the analysis (Table 2.4). 63  Table 2.4 Summary of MEST methylation. Control Oligospermia Severe Oligospermia Vasectomy Reversal Obstructive Azoospermia Non-Obstructive Azoospermia C01 4.57 O01 38.81 SO01 8.40 VR01 48.01 OA01 46.53 NOA01 47.24 C02 8.55 O02 38.52 SO02 44.47 VR02 32.61 OA02 64.22 NOA02 61.91 C03 8.44 O03 29.62 SO03 32.57 VR03 44.20 OA03 9.86 NOA03 47.47 C04 10.30 O04 83.29 SO04 18.72 VR04 32.05 OA04 43.09 NOA04 - C05 19.75 O05 65.13 SO05 37.71 VR05 57.18 OA05 69.41   C06 7.82 O06 5.22 SO06 36.42 VR06 32.98 OA06 7.32   C07 7.52 O07 5.58 SO07 5.71 VR07 26.54 OA07 47.37   C08 5.04 O08 5.95 SO08 50.66 VR08 41.89     C09 3.89 O09 10.47 SO09 72.48       C10 6.82 O10 53.14 SO10 14.72       C11 3.31 O11 1.58 SO11 -       C12 3.47 O12 26.42         C13 4.22 O13 30.84         C14 5.28 O14 45.67         C15 3.51 O15 5.17         C16 6.93 O16 22.23         C17 5.36 O17 5.49         C18 4.23 O18 35.87         C19 4.29           C20 4.63           Mean 6.40  28.28  32.19  39.43  41.11  52.21 Median 5.16  28.02  34.50  37.44  46.53  47.47  64   Figure 2.3 Mean DNA methylation of MEST by pyrosequencing. Mean DNA methylation of the maternally imprinted MEST gene. C, control men (n = 20); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 8); OA, obstructive azoospermic men (n = 7); NOA, non-obstructive azoospermic men (n = 3). Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range. Asterisks represent significant differences compared to controls (p < 0.05; Mann-Whitney test).  2.3.4 LIT1 Gene For the maternally imprinted LIT1 gene, we studied thirteen CpG sites. The values shown in Table 2.5 represent the average of these CpG sites within each patient studied. In this gene, we did not find methylation to be significantly different when groups were compared overall (p > 0.05; Kruskal-Wallis test) or when compared pairwise (p > 0.05; Mann-Whitney test; Figure 2.4). Some samples repeatedly failed pyrosequencing at some CpG sites for this gene and were excluded in the analysis (Table 2.5). 65  Table 2.5 Summary of LIT1 methylation. Control Oligospermia Severe Oligospermia Vasectomy Reversal Obstructive Azoospermia Non-Obstructive Azoospermia C01 2.93 O01 4.81 SO01 26.76 VR01 62.90 OA01 - NOA01 20.76 C02 3.60 O02 6.17 SO02 24.84 VR02 8.32 OA02 1.41 NOA02 75.27 C03 2.04 O03 10.70 SO03 2.06 VR03 19.22 OA03 64.88 NOA03 19.90 C04 - O04 1.58 SO04 5.36 VR04 4.71 OA04 18.88 NOA04 - C05 5.85 O05 14.80 SO05 2.11 VR05 1.51 OA05 2.93   C06 7.45 O06 1.42 SO06 16.66 VR06 7.13 OA06 11.66   C07 6.90 O07 7.64 SO07 23.01 VR07 1.25 OA07 -   C08 5.81 O08 19.70 SO08 25.28 VR08 7.76     C09 2.18 O09 17.50 SO09 1.28       C10 4.99 O10 4.03 SO10 2.26       C11 3.99 O11 47.00 SO11 -       C12 4.13 O12 3.12         C13 4.98 O13 3.74         C14 5.63 O14 21.55         C15 2.66 O15 7.61         C16 7.45 O16 3.02         C17 7.14 O17 10.35         C18 10.61 O18 30.88         C19 3.49           C20 1.73           Mean 4.92  11.98  12.96  14.10  19.95  38.64 Median 4.98  7.63  11.01  7.45  11.66  20.76  66   Figure 2.4 Mean DNA methylation of LIT1 by pyrosequencing. Mean DNA methylation of the maternally imprinted LIT1 gene. C, control men (n = 19); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 8); OA, obstructive azoospermic men (n = 5); NOA, non-obstructive azoospermic men (n = 3). Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  2.3.5 LINE1 Transposable Element For the repetitive element LINE1, we studied five CpG sites. The values shown in Table 2.6 represent the average of these CpG sites within each patient studied. We did not find methylation to be significantly different when groups were compared overall (p > 0.05; Kruskal-Wallis test) or pairwise (p > 0.05; Mann-Whitney test; Figure 2.5). 67  Table 2.6 Summary of LINE1 methylation. Control Oligospermia Severe Oligospermia Vasectomy Reversal Obstructive Azoospermia Non-Obstructive Azoospermia C01 75.73 O01 49.89 SO01 86.60 VR01 61.19 OA01 80.67 NOA01 84.25 C02 88.67 O02 64.67 SO02 84.34 VR02 66.32 OA02 68.70 NOA02 70.58 C03 70.52 O03 86.69 SO03 69.32 VR03 61.01 OA03 90.49 NOA03 57.52 C04 69.83 O04 76.02 SO04 83.22 VR04 70.46 OA04 72.02 NOA04 65.51 C05 76.68 O05 62.14 SO05 76.70 VR05 61.55 OA05 60.36   C06 73.04 O06 54.04 SO06 65.74 VR06 63.98 OA06 72.20   C07 86.06 O07 90.23 SO07 65.16 VR07 63.90 OA07 68.74   C08 76.69 O08 72.86 SO08 67.73 VR08 70.59     C09 83.01 O09 79.79 SO09 64.00       C10 79.89 O10 69.00 SO10 17.33       C11 69.50 O11 56.93 SO11 54.11       C12 81.61 O12 75.60         C13 78.32 O13 60.32         C14 71.36 O14 65.10         C15 82.93 O15 69.12         C16 61.38 O16 70.59         C17 56.05 O17 68.03         C18 81.98 O18 68.13         C19 74.90           C20 61.26           Mean 74.97  68.84  66.75  64.88  73.31  69.47 Median 76.21  68.57  67.73  63.94  72.02  68.05  68   Figure 2.5 Mean DNA methylation of LINE1 by pyrosequencing. Mean DNA methylation of the global transposable element LINE1. C, control men (n = 20); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 11); VR, vasectomy reversal men (n = 8); OA, obstructive azoospermic men (n = 7); NOA, non-obstructive azoospermic men (n = 4). Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  2.4 Discussion In this study of DNA methylation at imprinted genes, we studied two paternally imprinted genes and two maternally imprinted genes in the sperm of infertile men. While most previous studies used cloning techniques, we used a bisulfite pyrosequencing method to allows us to get a better overall picture of the methylation in the sperm of infertile men. Previous studies only looked at methylation of only a small number of sperm per patient, while pyrosequencing theoretically enables us to look at the average methylation within the entire available sample. However, the pyrosequencing technology only allows for sequencing of small fragments of DNA that are ~100 bp or less. Consequently, the methylation values obtained were based on only small numbers of CpG sites (4-13 CpG sites, see results section), but over a greater number of sperm per individual. Nevertheless, we obtained results similar to those already published 69  (Kobayashi et al., 2007; Kobayashi et al., 2009; Poplinski et al., 2010; Sato et al., 2011; Li et al., 2013; Montjean et al., 2013; Laurentino et al., 2015; Louie et al., 2016). Consistent with these studies, we found H19 hypomethylation among oligospermic men, particularly at very low sperm counts (severe oligospermia; sperm count <5 million/mL). We also found MEST hypermethylation in oligospermic men, which is consistently seen across studies. Interestingly, we found that MEST was also significantly hypermethylated in azoospermic men compared to fertile controls, which was not observed in our previous study of azoospermic men, despite the low sample sizes in both studies (Minor et al., 2011). It is possible that the cloning methodology used previously only preferentially amplified sperm with the proper methylation profile, as was suggested to be a limitation in the previous study. Indeed, a study by Laurentino et al. has shown that the sperm of infertile men displays mosaic methylation patterns that are mostly normal and only a minority are abnormal (Laurentino et al., 2015). With such a mosaic pattern in methylation, it is clear to see that cloning of only small numbers of sperm may not accurately reflect the whole sperm population in the individual. In the paternally imprinted GTL2 and maternally imprinted LIT1 genes, we found no significant methylation differences between any groups, which is in agreement with the limited number of studies looking at theses genes (Kobayashi et al., 2007; Kobayashi et al., 2009; Minor et al., 2011; Sato et al., 2011; Louie et al., 2016). It has been suggested that the repeat nature of the GTL2 gene may offer stability of the methylation level (Li et al., 2004; Minor et al., 2011), similar to LINE1 satellite repeats, which are also highly stable (Figure 2.5). It is unknown whether this may also be the case for the LIT1 gene. No differences in global DNA methylation was also observed and this is consistent with previous studies (Kobayashi et al., 2007; Marques et al., 2008; Boissonnas et al., 2009; Kobayashi et al., 2009; Li et al., 2013). Taken together, the 70  results highlight the differential nature of DNA methylation aberrancies: the defects are isolated to imprinted genes, but only some imprinted genes are affected and not others. There also does not seem to be discrimination between paternal and maternal imprinted genes. Studies by Kobayashi et al. found the same methylation defect in ART conceptuses and the originating sperm (Kobayashi et al. 2009). This has raised concerns about the safety of ART and the potential to transmit imprinted gene defects to the offspring. Indeed, several reports looked into whether ART increases malformations (Rycke et al., 2002; Zhu et al., 2006; Rimm et al., 2011; Farhi et al., 2013; Kelley-Quon et al., 2013; Heisey et al., 2015; Qin et al., 2015) or imprinting defects (Doornbos et al., 2007; Tierling et al., 2010; Zheng et al., 2011; Hiura et al., 2012; Oliver et al., 2012; Hiura et al., 2014; Sakian et al., 2015; Tenorio et al., 2016; Vincent et al., 2016; Mussa et al., 2017) in the offspring; however, the reports are conflicting. Regarding malformations, early studies suggested there was no increased risk for offspring conceived via ART (Rycke et al., 2002) and even a later meta-analysis conducted by Rimm et al. showed that malformations were likely overestimated (Rimm et al., 2011). More recent meta-analyses consistently report higher risks of malformations in the ART population, although the risk varies considerably among the studies (Farhi et al., 2013; Kelley-Quon et al., 2013; Heisey et al., 2015; Qin et al., 2015). From these studies, it is unclear whether the ART itself increases the risk or if the increased risk is inherent in the use of subfertile gametes.  The risk of imprinting defects in ART offspring is more difficult to determine. Previous studies by our lab and others have demonstrated differential imprint methylation in ART vs natural conception (Tierling et al., 2010; Vincent et al., 2016), but not others (Zheng et al., 2011; Oliver et al., 2012). Studies have also shown that imprinting disorders are increased when using ART and that methylation defects are a contributing factor in a greater number of those affected 71  (Hiura et al., 2012; Hiura et al., 2014; Tenorio et al., 2016; Mussa et al., 2017), however, some studies failed to find an association between ART and imprinting defects (Doornbos et al., 2007; Tierling et al., 2010). None of these studies looked at associating the sperm of infertile men with these imprinting defects. A study by Camprubí et al. found that there is little effect of using sperm with methylation defects in ART (Camprubí et al., 2012). There are some explanations as to why using the sperm of infertile men in ART should not be a cause for alarm. First, as suggested by Laurentino et al., the sperm population is mosaic with respect to methylation defects and there is a greater probability that a normal sperm will be chosen in the ART process. Second, the majority of the methylation defects across studies involve only a few CpG sites and this may not have any downstream biological effects especially when ART processes themselves are speculated to perturb methylation. Nevertheless, the study of methylation defects in the sperm of infertile men still holds value in that the mechanisms involved may elucidate potential causes of infertility.   72  CHAPTER 3: INVESTIGATION OF THE FOLATE CYCLE IN INFERTILE MEN 3.1 Introduction Potentially, aberrancies in methylation seen in the sperm of infertile men may be caused by disruptions in the generation of the universal methyl donor SAM. The metabolism of folate in the folate and methionine cycles generates the SAM necessary for methylation reactions but is also important for DNA synthesis. Three key enzymes in these cycles work together to produce SAM from folate: MTHFR, MS, and MTRR. All of these enzymes are known to have SNPs within their gene sequences that can affect the enzymes’ activity. The MTHFR C677T SNP is a mutation of the nucleotide at the 677th position which changes cytosine to thymine resulting in an amino acid substitution from alanine to valine at position 222 of the protein (Camprubí et al., 2013). There is another SNP in the MTHFR gene at position 1298 that changes adenosine to cytosine and an amino acid substitution from glutamate to alanine (Castro, 2004). Since the effect of the A1298C SNP on DNA methylation was observed to be lower than the C677T SNP (Castro, 2004), we chose to only look at the C677T SNP in our study. Similarly, the MS A2756G changes adenosine to guanine and results in glycine substituted for aspartic acid (Matsuo et al., 2001); while the MTRR A66G changes adenosine to guanine and corresponds to an isoleucine to methionine substitution (Gaughan et al., 2001). A recent meta-analysis concluded that the MTHFR C677T SNP was a risk factor for male infertility in men with azoospermia or OAT, but the MTHFR A1298C SNP was not related to male infertility (Liu et al., 2015). Associations between MS A2756G or MTRR A66G and male infertility could not be made or disappeared after statistical correction. The lack of association of MS and MTRR SNPs with male infertility was suggested to be caused by an inadequate number of appropriate studies included in the analysis. Additionally, the authors highlight the importance 73  that ethnicity plays in associations of these SNPs and male infertility, though inconsistencies within studies separated by ethnicity were also observed. Although these studies associate folate enzyme SNPs with male infertility, few studies have associated these SNPs with aberrant DNA methylation (Chen et al., 2001; Friso et al., 2002; Castro et al., 2004; Camprubí et al., 2013; Louie et al., 2016). Here, we determine the genotypes of the MTHFR C677T, MS A2756G, and MTRR A66G SNPs in infertile men with oligospermia and azoospermia using restriction fragment length polymorphism (RFLP) and compare the frequency distribution to that of fertile control men. We also measure blood folate concentrations to observe the effect of these SNPs on folate concentrations and correlate these measures with DNA methylation, if possible. Overall, the data will allow us to determine if reduced folate metabolism has an effect on DNA methylation of imprinted genes in infertile men. 3.2 Materials and Methods 3.2.1 Sample Collection Peripheral blood samples were collected from the same patients described in section 2.2.1. 6 mL of peripheral blood was collected in EDTA Vacutainer blood collection tubes (Becton Dickinson, Mississauga, ON, Canada) at the same time that ejaculate or testicular tissue was donated. 3.2.2 Folate Analysis Immediately after collection, 200 µL of peripheral blood was mixed with 1.8 mL of a 1% solution of ascorbic acid (Fisher Scientific) in a 5 mL round bottom tube (VWR, Mississauga, ON, Canada). The mixture was then incubated at 37°C for 30 minutes and 1 mL was aliquoted into each of two 1.6 mL CryoPure tubes (Sarstedt). The tubes were frozen at -80°C until further 74  analysis. Determination of folate concentration was done using the Architect i2000SR immunoassay analyser (Abbott, Mississauga, ON, Canada) according to the manufacturer’s protocol. Since neither the percentage of hematocrit or plasma folate were measured, the values reported are folate concentrations in whole blood. 3.2.3 Genotyping 3.2.3.1 DNA Purification Genomic DNA was extracted and purified from the remainder of each peripheral blood sample using the Gentra Puregene Blood kit (Qiagen) according to the manufacturer’s protocol. Briefly, the blood was divided into two 15 mL conical Falcon tubes (VWR) containing approximately 3 mL each. RBC lysis solution (Qiagen) was added to remove red blood cells, cell lysis solution (Qiagen) was added to lyse the white blood cells which contain the DNA, and protein precipitation solution (Qiagen) was added to remove proteins from the solution. The supernatant containing the DNA was isolated and mixed with isopropanol to precipitate the DNA. 70% ethanol was used to wash the DNA before dissolving in DNA hydration solution (Qiagen). DNA was stored at 4°C for short-term usage or -20°C for longer term storage. 3.2.3.2 PCR Amplification PCR was used to amplify the appropriate regions of interest within each folate cycle-related gene using the primer pairs shown in Table 3.1. PCRs were carried out in 25 µL reactions: 1x PCR buffer, containing 15 mM MgCl2 (Qiagen), 0.2 mM dNTPs (Invitrogen), 0.4 µM forward and reverse primers, 0.5 U HotStarTaq DNA polymerase (Qiagen), and 1 µL purified peripheral blood DNA. The following PCR cycling conditions were used for MTHFR: initial denaturation at 95°C for 15 min; 44 cycles of 95°C for 45 sec, 65°C for 1 min, and 72°C for 90 sec; and a final extension at 72°C for 5 min. The following PCR cycling conditions were 75  used for both MS and MTRR: initial denaturation at 95°C for 15 min; 49 cycles of 95°C for 1 min, 56°C for 1 min, and 72°C for 1 min; and a final extension at 72°C for 5 min. Table 3.1 Primer sequences used in genotyping folate cycle genes.  3.2.3.3 Restriction Fragment Length Polymorphism Patients were genotyped using RFLP assays. Restriction digest reactions were carried out in 20 µL reactions containing 17.5 µL of each amplified PCR product, 1x NEBuffer 4, and 5 U of the appropriate restriction enzyme (New England Biolabs, Ipswich, MA, USA) shown in Table 3.2. Reactions were incubated at 37°C overnight. The remaining 7.5 µL of PCR product was to be used as an undigested control for comparison and was stored at 4°C in the meantime. Digested and undigested amplicons were mixed with 6x loading dye (Thermo Fisher Scientific) before loading side by side to compare sizes on a 3% agarose gel (Invitrogen) in 1x TAE [containing 40 mM Tris (Invitrogen), 20 mM acetic acid (Fisher Scientific), and 1 mM EDTA (Sigma-Aldrich)] with SYBR Safe DNA staining (Invitrogen) and 5 µL of 100bp DNA ladder (Invitrogen). Gels were run for one hour at 140 V before being visualized using a UV transilluminator. Region Type Sequence (5′ - 3′) MTHFR Forward TGAAGGAGAAGGTGTCTGCGGGA  Reverse AGGACGGTGCGGTGAGAGTG MS Forward TGTTCCAGACAGTTAGATGAAAATC  Reverse GATCCAAAGCCTTTTACACTCCTC MTRR Forward GCAAAGGCCATCGCAGAAGAC  Reverse TGGTGGTATTAGTGTCCTTTT 76  Table 3.2 Restriction enzymes used in RFLP genotyping. N = A or C or G or T 3.2.4 Data Analysis Since whole blood folate concentrations were determined to be both normally distributed (p > 0.05; Shapiro-Wilk Normality test) and equal in variances (p > 0.05; Levene’s test), comparisons between groups were done using ANOVA with Tukey’s honest significant difference correction for multiple comparisons. Correlation analysis between DNA methylation and folate concentration was done using Spearman’s rank correlation coefficient because the relationship is predicted to be monotonic. All analyses were done using R: A language and environment for statistical computing (v3.3.3). Gel bands were compared for each digested and undigested pair and genotypes were assigned based on Table 3.3. Genotypes were compared between groups using Fisher’s exact test with FDR corrections. Comparisons of DNA methylation within each folate genotype were done using the non-parametric Kruskal-Wallis test. All statistical tests used a significance threshold of p < 0.05. Region Restriction Enzyme Restriction Site MTHFR HinfI 5′-G|ANT C-3′ 3′-C TNA|G-5′ MS HaeIII 5′-GG|CC-3′ 3′-CC|GG-5′ MTRR NdeI 5′-CA|TA TG-3′ 3′-GT AT|AC-5′ 77  Table 3.3 Expected sizes of bands in folate cycle genes. bp = base pair 3.3 Results 3.3.1 Folate Concentration We measured folate concentrations in whole blood of infertile men using the Architect i2000SR immunoassay analyser. Blood was drawn as part of the fertility evaluation upon recruitment of oligospermic men. However, azoospermic men and men undergoing vasectomy reversals rarely had blood drawn as part of their surgery. Therefore, the folate concentrations measured are missing a majority of the testicular samples (Figure 3.1). We were able to measure blood folate concentrations for a total of 28 men: one control man, 16 men with oligospermia, 10 men with severe oligospermia, and one man with non-obstructive azoospermia. Although we observed higher variability in the oligospermic men, there were no statistical differences between any groups (p > 0.05; ANOVA). Region Homozygous Wildtype Heterozygous Homozygous SNP MTHFR CC 198bp CT 198bp 175bp 23bp TT 175bp 23bp MS AA 211bp AG 211bp 131bp 80bp GG 131bp 80bp MTRR AA 179bp AG 179bp 157bp 22bp GG 157bp 22bp 78   Figure 3.1 Whole blood folate concentration in infertile men. Whole blood folate concentration measured in each patient group. C, control men (n = 1); O, oligospermic men (n = 16); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 0); OA, obstructive azoospermic men (n = 0); NOA, non-obstructive azoospermic men (n = 1). Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  3.3.2 Correlation of Folate Concentration and DNA Methylation Despite only having a subset of samples with folate concentration data, we are still able to calculate correlations with DNA methylation. Since folate data was mostly collected from infertile men with oligospermia, the correlations are heavily biased toward these men. Nevertheless, these correlations are still valid for oligospermic men (Figure 3.2 to Figure 3.6). Using Spearman’s rank correlation, we found that there is a significant positive relationship between folate concentration and GTL2 methylation (p = 0.0241; Figure 3.3). Folate concentration was not significantly correlated with methylation at any other genes (p > 0.05). 79   Figure 3.2 Correlation between whole blood folate and mean H19 methylation. Calculated using Spearman’s rank correlation (n = 28). Data represents individuals with oligospermia.   Figure 3.3 Correlation between whole blood folate and mean GTL2 methylation. Calculated using Spearman’s rank correlation (n = 28). Data represents individuals with oligospermia.  80   Figure 3.4 Correlation between whole blood folate and mean MEST methylation. Calculated using Spearman’s rank correlation (n = 27). Data represents individuals with oligospermia.   Figure 3.5 Correlation between whole blood folate and mean LIT1 methylation. Calculated using Spearman’s rank correlation (n = 27). Data represents individuals with oligospermia.  81   Figure 3.6 Correlation between whole blood folate and mean LINE1 methylation. Calculated using Spearman’s rank correlation (n = 28). Data represents individuals with oligospermia.  3.3.3 Folate Enzyme Genotypes Genotypes were determined in peripheral blood using RFLP. For the same reasons given above, genotypes of the majority of the testicular samples were not determined. For all genes, we observed no significant differences in frequency of SNPs between any groups (p > 0.05; Fisher’s exact test; Figure 3.7 to Figure 3.9). Interestingly, within our population, we only observed a total of five men who were homozygous for a SNP (four MTHFR and one MTRR), three of which were found in control men. 82   Figure 3.7 Frequency of MTHFR genotypes. Frequency of MTHFR genotypes determined using RFLP. C, control men (n = 10); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 4); OA, obstructive azoospermic men (n = 2); NOA, non-obstructive azoospermic men (n = 2).   Figure 3.8 Frequency of MS genotypes. Frequency of MS genotypes determined using RFLP. C, control men (n = 9); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 4); OA, obstructive azoospermic men (n = 2); NOA, non-obstructive azoospermic men (n = 2).  83   Figure 3.9 Frequency of MTRR genotypes. Frequency of MTRR genotypes determined using RFLP. C, control men (n = 10); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 4); OA, obstructive azoospermic men (n = 2); NOA, non-obstructive azoospermic men (n = 2).  3.3.4 Effect of Folate Genotypes on DNA Methylation Since enzyme activity is reported to be affected by SNPs, the pool of methyl donors can possibly be diminished resulting in abnormal methylation in men who have at least one SNP. To determine if SNPs in folate metabolism enzymes have an effect on DNA methylation, we separated the mean DNA methylation values measured previously for each gene according to the genotypes obtained (Figure 3.10 to Figure 3.14). There were no significant differences found between genotypes of any folate enzymes and DNA methylation of the genes studied (p > 0.05; Kruskal-Wallis test). 84   Figure 3.10 DNA methylation of H19 separated by folate enzyme genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.   Figure 3.11 DNA methylation of GTL2 separated by folate enzyme genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  85   Figure 3.12 DNA methylation of MEST separated by folate enzyme genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.   Figure 3.13 DNA methylation of LIT1 separated by folate enzyme genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  86   Figure 3.14 DNA methylation of LINE1 separated by folate enzyme genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  3.4 Discussion In our study of the folate cycle, we measured folate concentration and determined the frequency of SNPs in folate metabolism enzymes in infertile men. Using the methylation data collected in the previous chapter, we correlated DNA methylation at all regions and folate concentration. Although we only had a subset of our samples with folate data, we still show that there is a significant, though modest, correlation between GTL2 methylation and folate concentration (r = 0.4251; Figure 3.3). This finding is in contrast to other studies which found that methylation of GTL2 does not change after low (Chan et al., 2017) or high (Aarabi et al., 2015) folic acid supplementation. Both groups also studied H19 and MEST methylation in response to folic acid supplementation and did not see any changes, which is similar to the findings presented here. The contradicting results highlight two issues in studying the effects of folate in infertile men. First, the study populations consist of different phenotypes of infertile 87  men. Our infertile population is composed primarily of men with oligospermia, whereas the study by Aarabi et al. used normozoospermic idiopathic infertile men and Chan et al. did not look at infertile men. From what we already know about the aetiologies of the different phenotypes of infertility, there can be great differences in the genes and mechanisms involved. Therefore, it is possible that folate concentration may affect infertile populations differently and further studies should ensure proper distinction between infertile phenotypes in order to confirm these findings. The second issue in folate studies is ethnic variability. None of the studies included information regarding the ethnicities of the studied populations. A recent meta-analysis studied the association between the MTHFR genotype and male infertility (Liu et al., 2015). The authors found that MTHFR C677T mutations were a risk factor for male infertility, however, the risk factor was higher for men of certain ethnicities. With this in mind, we suggest that it is possible for folate concentration to affect methylation more in certain ethnicities. When looking at the effect of genotypes on methylation, we do not see any significant differences within any gene. A similar result was also found by Aarabi et al. in imprinted genes. However, interestingly, they also found that DNA methylation was disrupted globally with high dose folic acid supplementation and the effect was exacerbated by MTHFR C677T mutations. The authors suggest high dose folic acid may disrupt the folate cycle which leads to reduced methylation due to a decrease in the methyl donor pool. The different findings could be a result of the high dose folic acid supplementation which our study population did not receive. Thus, the infertile men in our study did not obtain the high folate concentration seen in the study by Aarabi et al. and consequently the folate cycle and methylation were unaffected. Another possible explanation of the findings is the different methodologies used in each study. In the present study, we used pyrosequencing of LINE1 repetitive elements to determine the level global 88  methylation, while Aarabi et al. used reduced representation bisulfite sequencing (RRBS) to assess global methylation. RRBS is a genome-wide methylation technique that covers far more CpG sites than pyrosequencing and is also able to cover CpG sites from several different repeat regions. Indeed, the authors show that methylation of long interspersed nuclear elements (LINEs), such as LINE1, is disrupted to a much lesser extent compared to other repeat regions. Overall, a small sample size and a limited methodology may be the main cause for the results presented in this study. Although the results are still valuable, future studies should focus on extending the analysis of global methylation to all types of repeat regions rather than just one, as they are not all affected equally. There are few other studies which also investigated folate genotypes and methylation and found that mutations were associated with overall global hypomethylation (Chen et al., 2001; Friso et al., 2002; Castro et al., 2004). These studies, however, only focused on the relationship between mutations in MTHFR and global DNA methylation. Furthermore, since folate metabolism is important for a variety of biological mechanisms in addition to fertility, these studies investigated MTHFR genotypes and global methylation in different study populations or using an animal model. More recently, a study by Camprubí et al. used the sperm of infertile men to associate MTHFR polymorphisms with DNA methylation of imprinted genes (Camprubí et al., 2013). However, the authors only considered MTHFR genotype and two imprinted genes. Despite this, the study still provides further evidence against polymorphisms negatively impacting imprinted gene methylation, which is a similar finding to the present study. Taken together, there is growing evidence that global DNA methylation is decreased as a result of mutations in the folate cycle, but imprinted genes may not be directly affected. This effect is also not limited to infertile individuals, but may affect anyone with polymorphisms in folate genes. 89  To the best of our knowledge, the current study is the first to associate a greater number of folate-related genes with DNA methylation both globally and at imprinted genes in infertile men. Although we could not draw any conclusions due to the small sample size, the basic framework will provide a more in-depth analysis for future studies to build from by using more folate enzymes and both global and imprinted gene-specific DNA methylation. Additionally, future studies should also include a genome-wide methylation approach for a better representation of global methylation level.   90  CHAPTER 4: INVESTIGATION OF SPERM DNA INTEGRITY IN INFERTILE MEN 4.1 Introduction In contrast to the well-documented effects of folate on various biological mechanisms, including DNA methylation, the literature regarding damage to DNA and its effects on DNA methylation is lacking. Of the few reports available, there is general agreement that there is a negative relationship between DNA fragmentation and global DNA methylation (Tavalaee et al., 2009; Tunc and Tremellen, 2009; Montjean et al., 2015). The exact mechanistic link between the two is not yet known, however, there are two speculated mechanisms. The first mechanism involves chromatin compaction. There are studies that have included the analysis of chromatin integrity while investigating DNA fragmentation (Tavalaee et al., 2009; Montjean et al., 2015). The study by Montjean et al. found a negative relationship between DNA methylation and both DNA fragmentation and chromatin packaging. However, the study by Tavalaee et al. only found an association between chromatin packaging and DNA fragmentation. Thus, Montjean et al. suggest that DNA methylation precedes, and is necessary for, proper chromatin packaging which then ultimately provides protection of the DNA from any insults. However, the results from the other report by Tavalaee et al. suggests that DNA fragmentation results from abnormal chromatin packaging and that DNA methylation is an independent process affecting DNA fragmentation only. Although there is no consensus on these results regarding chromatin packaging, they do agree that there is a relationship between DNA methylation and DNA fragmentation. The second possible mechanistic link between DNA methylation and DNA fragmentation involves ROS. A study by Tunc and Tremellen included a measure of ROS concentration in addition to DNA fragmentation and found global DNA demethylation in samples with higher DNA fragmentation and ROS (Tunc and Tremellen, 2009). The authors 91  suggest that ROS oxidative damage to guanine residues of DNA can produce 8-OHdG adducts which are inadequately bound to DMNTs or methyl-binding proteins and results in impaired DNA methylation activity. It is important to note that the authors also measured ROS concentration which supports their claim that DNA fragmentation is a result of ROS. Recently, a study by Rajabi et al. found more evidence in support of this idea. The authors found that male pronuclei from sperm with DNA fragmentation had higher levels of DNA methylation (Rajabi et al., 2017). Since they were investigating pronuclei instead of sperm, an opposite relationship was found where DNA fragmentation increased DNA methylation. Similar to the previous study, the authors suggest that DNA fragmentation impairs demethylation activity. However, the study lacked any measure of ROS that would support the generation of adducts that interfere with demethylation reactions. As DNA damage is closely related to oxidative stress, it may also be useful to investigate antioxidants and their pathways in an effort remedy DNA fragmentation and possibly DNA methylation abnormalities. One of the more abundant antioxidants produced in cells is GSH, created by reacting glutamate with cysteine (Gorrini et al., 2013). GSH has the capacity to reduce ROS to prevent the negative consequences of oxidative stress, such as DNA damage, and is converted to GSSG in the process. This reaction is catalyzed by GST enzymes, with GSTT1, GSTM1, and GSTP1 being the most studied. In particular, polymorphisms in GSTT1 and GSTM1 can produce null variants resulting in between zero and two functional enzymes depending on the zygosity of the polymorphism (Buchard et al., 2007). The null genotypes produce non-functional enzymes that reduce the ROS processing rate. This ultimately reduces the antioxidant potential and renders the cell more vulnerable to oxidative stress. Since the functional nature of the GST enzymes is so broad, the literature on its relationship with infertility is relatively sparse. 92  However, a meta-analysis concluded that null genotypes of either gene is a risk factor for infertility and having both null genotypes further increases the risk (Ying et al., 2013). Additionally, studies investigating methylation and GST enzymes are lacking. The aim of this study is to determine the level of DNA fragmentation present in the sperm of infertile men. Fragmentation will be measured using the terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay in men with oligospermia and azoospermia. The level of fragmentation can then be correlated with the methylation data obtained previously. Genotypes of GSTT1 and GSTM1 will also be determined to evaluate its impact on DNA fragmentation and, if possible, DNA methylation. 4.2 Materials and Methods 4.2.1 Sample Collection Patients in this study were the same as described in section 2.2.1. 4.2.2 DNA Fragmentation 4.2.2.1 Sample Preparation Ejaculate or testicular samples were washed in mHTF as described in section 2.2.2. The washed samples were checked on a microscope slide for adequate concentration. If necessary, more mHTF was added to dilute or the sample was pelleted and some supernatant was removed to concentrate. 40 µL of the sample was pipetted onto a microscope slide and allowed to air dry. Microscope slides were frozen at -20°C until further analysis. 4.2.2.2 Slide Preparation Microscope slides were taken from the freezer to thaw and air dry. A fixation solution of 4% paraformaldehyde (PFA) was made by dissolving 1.6 g of PFA (Sigma-Aldrich) in 40 mL 1x 93  phosphate buffered saline (PBS) (Thermo Fisher Scientific). The solution was heated in a water bath and two drops of NaOH were added to help dissolve. Each slide was fixed with 100 µL of fixation solution and a coverslip for one hour at room temperature. A permeabilization solution of 1% Triton X-100 in 0.1% sodium citrate was made by mixing 400 µL of Triton X-100 (Sigma-Aldrich) with 0.04 g of sodium citrate (Fisher Scientific) and made to 40 mL with water. The solution was put on ice until use. After fixation, coverslips were removed and the slides were rinsed in 1x PBS in a coplin jar. After air drying, the slides were incubated in another coplin jar filled with the permeabilization solution for two minutes on ice. The slides were then rinsed twice in 1x PBS and left to air dry. 4.2.2.3 TUNEL Assay The In Situ Cell Death Detection Kit, Fluorescein (Sigma-Aldrich) was used to perform the TUNEL assay. The TUNEL reaction mixture was made using 5 µL enzyme solution and 45 µL label solution per slide. 50 µL of the TUNEL reaction mixture was then added to each slide and a coverslip was placed over top. Slides were then incubated in a dark humid chamber for 60 minutes at 37°C. Positive control slides were treated with a solution of 100 U of DNase I (New England Biolabs) in 50 mM Tris-HCl (Invitrogen) for ten minutes at room temperature prior to adding the TUNEL reaction mixture. The TUNEL reaction mixture for the negative control slides contained only label solution and no enzyme solution. After incubation, the coverslips were removed and the slides were rinsed three times in 1x PBS and air dried. One drop of Vectashield antifade mounting medium (Vector Laboratories, Burlingame, CA, USA) was added to prevent photobleaching of the fluorescent label and covered with a coverslip. Slides were stored at -20°C until analysis. 500 sperm cells were counted using fluorescence microscopy on the AxioPlan 2 microscope (Carl Zeiss Canada, Toronto, ON, Canada). Fragmented sperm cells 94  were labelled green, while unfragmented sperm cells were labelled red. Percentage DNA fragmentation was determined as the ratio of fragmented sperm cells counted to the total number of sperm cells observed. 4.2.3 Genotyping 4.2.3.1 DNA Purification DNA was purified from peripheral blood as described in section 3.2.3.1. 4.2.3.2 PCR Amplification Regions of interest were PCR amplified in a multiplex PCR assay using the primer pairs shown in Table 4.1. PCR amplifications were carried out in a single 25 µL multiplex reaction: 1x PCR buffer, containing 15 mM MgCl2 (Qiagen), 0.2 mM dNTPs (Invitrogen), 0.3 µM forward and reverse primers, 0.625 U HotStarTaq DNA polymerase (Qiagen), and 1 µL purified peripheral blood DNA. The following PCR cycling conditions were used: initial denaturation at 95°C for 15 min; 34 cycles of 95°C for 1 min, 59°C for 1 min, and 72°C for 1 min; and a final extension at 72°C for 10 min. β-globin is a subunit of the haemoglobin protein found in RBCs and therefore was used as an internal PCR control. Table 4.1 Primer sequences used in GST genotyping. Region Type Sequence (5′ - 3′) GSTT1 Forward TTCCTTACTGGTCCTCACATCTC  Reverse TCACCGGATCATGGCCAGCA GSTM1 Forward CTGCCCTACTTGATTGATGGG  Reverse CTGGATTGTAGCAGATCATGC β-globin Forward ACACAACTGTGTTCACTAGC  Reverse CAACTTCATCCACGTTCACC  95  4.2.3.3 Agarose Gel Electrophoresis Amplicons were mixed with 6x loading dye (Thermo Fisher Scientific) before loading on a 2% agarose gel (Invitrogen) in 1x TAE [containing 40 mM Tris (Invitrogen), 20 mM acetic acid (Fisher Scientific), and 1 mM EDTA (Sigma-Aldrich)] with SYBR Safe DNA staining (Invitrogen) and 5 µL of 100bp DNA ladder (Invitrogen). Gels were run for one hour at 140 V before being visualized using a UV transilluminator. The experiment was repeated two more times for every null genotype to confirm deletions. 4.2.4 Data Analysis Since DNA fragmentation as determined to be both normally distributed (p > 0.05; Shapiro-Wilk Normality test) and equal in variances (p > 0.05; Levene’s test), comparisons between groups were done using ANOVA with Tukey’s honest significant difference correction for multiple comparisons. Correlation analysis between DNA methylation and DNA fragmentation was done using Spearman’s rank correlation coefficient. All analyses were done using R: A language and environment for statistical computing (v3.3.3). Gel bands that were present were classified as GST positive and gel bands that were absent were classified as GST negative. Genotypes were compared between groups using Fisher’s exact test with FDR corrections. Comparisons of DNA methylation within each GST genotype were done using the non-parametric Kruskal-Wallis test. All statistical tests used a significance threshold of p < 0.05. 4.3 Results 4.3.1 DNA Fragmentation The level of DNA fragmentation in sperm was measured using the TUNEL assay. Similar to the previous study, men with azoospermia had few sperm to isolate which resulted in a lack of 96  data for these men (Figure 4.1). However, we were able to measure the DNA fragmentation level in a reasonable amount of ejaculate samples: 10 control men, 13 men with oligospermia, and 10 men with severe oligospermia. Although we saw the DNA fragmentation level increase with the severity of oligospermia, this was not significantly different from any other group (p > 0.05; ANOVA; Figure 4.1).  Figure 4.1 DNA fragmentation in infertile men. DNA fragmentation measured in each patient group. C, control men (n = 10); O, oligospermic men (n = 13); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 1); OA, obstructive azoospermic men (n = 1); NOA, non-obstructive azoospermic men (n = 0). Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  4.3.2 Correlation of DNA Fragmentation and DNA Methylation Correlations with DNA methylation were calculated using Spearman’s rank correlation (Figure 4.2 to Figure 4.6). Since we had a lack of DNA fragmentation data for azoospermic men, these correlations are valid for men with oligospermia only. We found that DNA fragmentation was not significantly correlated with DNA methylation at any of the studied genes, including global DNA methylation (p > 0.05). 97   Figure 4.2 Correlation between DNA fragmentation and mean H19 methylation. Calculated using Spearman’s rank correlation (n = 35).   Figure 4.3 Correlation between DNA fragmentation and mean GTL2 methylation. Calculated using Spearman’s rank correlation (n = 35).  98   Figure 4.4 Correlation between DNA fragmentation and mean MEST methylation. Calculated using Spearman’s rank correlation (n = 34).   Figure 4.5 Correlation between DNA fragmentation and mean LIT1 methylation. Calculated using Spearman’s rank correlation (n = 34).  99   Figure 4.6 Correlation between DNA fragmentation and mean LINE1 methylation. Calculated using Spearman’s rank correlation (n = 35).  4.3.3 Genotypes of GST enzymes Genotypes were determined in peripheral blood using a multiplex PCR assay. We did not observe any significant differences in the frequency of null genotypes between any groups (p > 0.05; Fisher’s exact test; Figure 4.7 and Figure 4.8).   100   Figure 4.7 Frequency of GSTT1 genotypes. Frequency of GSTT1 genotypes. C, control men (n = 10); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 4); OA, obstructive azoospermic men (n = 2); NOA, non-obstructive azoospermic men (n = 2).   Figure 4.8 Frequency of GSTM1 genotypes. Frequency of GSTM1 genotypes. C, control men (n = 10); O, oligospermic men (n = 18); SO, severe oligospermic men (n = 10); VR, vasectomy reversal men (n = 4); OA, obstructive azoospermic men (n = 2); NOA, non-obstructive azoospermic men (n = 2).  101  4.3.4 Effect of GST Genotypes on DNA Methylation ROS and environmental toxicants can cause DNA fragmentation and also changes in DNA methylation. The GST enzymes function to mitigate this damage, however, the relationship between GST genotypes and DNA methylation has not been previously studied. We separated each sample by GST genotype to assess if there are any effects on DNA methylation (Figure 4.9 to Figure 4.13). However, we did not observe any significant differences in DNA methylation for any genes due to GST null genotypes (p > 0.05; Kruskal-Wallis test).  Figure 4.9 DNA methylation of H19 separated by GST genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  102   Figure 4.10 DNA methylation of GTL2 separated by GST genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.   Figure 4.11 DNA methylation of MEST separated by GST genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  103   Figure 4.12 DNA methylation of LIT1 separated by GST genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.   Figure 4.13 DNA methylation of LINE1 separated by GST genotype. Points represent individual patients. Thick lines in the boxplots represent median values, top and bottom of the boxes represent 75th and 25th percentiles, respectively. Whiskers extending from the boxplots represent 1.5 times the interquartile range.  104  4.4 Discussion We determined the DNA fragmentation level in infertile men and compared it to control men. We did not find any significant differences in DNA fragmentation level between groups and there was no significant correlation with DNA methylation at any of the genes studied. To the best of our knowledge, this is the first study to investigate the relationship between DNA methylation at imprinted genes and DNA fragmentation. Few studies have looked at the relationship between DNA fragmentation and DNA methylation (Tavalaee et al., 2009; Tunc and Tremellen, 2009; Montjean et al., 2015; Rajabi et al., 2017). Currently, these studies only focus on the relationship between global DNA methylation and not specific imprinted genes. Interestingly, however, we did not find any significant correlations with global DNA methylation as previous studies have found. The lack of an adequate number of samples analysed could contribute to the discrepancy, however, we also note that the infertile populations studied in each report are also phenotypically different. This is likely to cause a large variation in the results since we know the aetiology of infertility can involve varied genes and pathways. The high variability in the aetiology of infertility highlights the difficulty in designing studies using this sample population. Alternatively, the variation in results may also be due to the difference in the methods used to measure DNA methylation. Previous studies used immunohistochemical techniques to determine the level of global DNA methylation, whereas, in the current study, we used pyrosequencing of LINE1 repetitive elements as a surrogate measure of global DNA methylation. Since pyrosequencing targets specific sequences, this method may not be best suited for this particular application. Immunohistochemical assays specifically targeting 5-mC are able to canvas more of the genome and a larger variety of sequences that may yield a better representation of global DNA methylation. In any case, DNA damage may be randomly 105  distributed throughout the genome and thus we may only see global methylation changes and not in specific imprinted genes. However, the imprinted genes of infertile men may be more vulnerable to insults, thus it may still be worth continuing to investigate these specific regions in this population. Studies have previously associated polymorphisms in GST enzymes with male infertility (Ying et al., 2013). The authors suggested that GST null genotypes are risk factors for male infertility. In the current study, we found that the frequency of GST null genotypes was not significantly different between fertile and infertile men. Thus, we cannot provide evidence in support of this claim. However, as the sample size is inadequate, we are unable to be confident in this conclusion. Additionally, our study, as well as previous studies, fail to discern the effects of the zygosity of the null genotype; that is, there is no discrimination between having zero, one, or two working copies of the gene. In the context of being a risk factor for male infertility, the effects of zygosity may not produce a detectable amount of variability, however, there is still a potential to reveal more detailed information, especially in regards to DNA methylation. Considering there are currently assays available (Buchard et al., 2007), it may be worthwhile to include this extra detail in future studies. Although we did not find a significant relationship, this is the first study to investigate an association between GST genotypes and DNA methylation. As mentioned previously, since GSTs can protect against DNA damage from ROS, and DNA fragmentation is correlated with global DNA methylation, then we may deduce that there may be a possible relationship between DNA methylation and GST genotype. Further studies with larger sample sizes are necessary to elucidate a true relationship or lack thereof.   106  CHAPTER 5: CONCLUSION 5.1 Summary The main objective of this project was to study DNA methylation defects in infertile men and elaborate on possible sources of these defects. Similar to previous studies by our lab and others, our current study also found that infertile men have a greater number of sperm carrying epigenetic defects. Using bisulfite pyrosequencing of sperm DNA, we found that SO men had DNA methylation patterns at the paternally imprinted H19 gene that were significantly different from fertile men. Additionally, we also found that DNA methylation at the maternally imprinted MEST gene was significantly different in all study populations when compared to control fertile men. With these results, we suggest that the methylation defects commonly seen in infertile men are not isolated to solely paternally or maternally imprinted genes; however, these defects do seem to have specific effects on imprinted genes and not global DNA methylation. After confirming that our cohort of infertile men carries DNA methylation defects, we explored the folate cycle as a promising mechanism which may affect methylation in these men. SNPs in folate metabolism enzymes were compared between groups, but no differences were found. We also measured folate concentrations in these men to see if a reduced activity of these enzymes would be reflected in the folate concentration, but we still did not find any differences. When assessing the effects of these SNPs or folate concentration on DNA methylation, we only found a significant positive correlation between folate concentration and GTL2 methylation. We also sought to look into DNA fragmentation, which may potentially affect DNA methylation. We found that DNA fragmentation and the frequency of the null genotypes of the antioxidant enzyme GST were not different between groups. Lastly, we evaluated the effect of both DNA fragmentation and GST genotypes on DNA methylation but found no significant associations. 107  5.2 Limitations and Future Directions The current study was the first, to the best of our knowledge, to evaluate DNA methylation and possible related mechanisms in different phenotypes of infertile men. Looking at many different phenotypes, however, caused the sample numbers in each group to be fairly low. Focusing our efforts on fewer phenotypes may have yielded a higher number of samples for each group. Since we studied men with azoospermia using a similar methodology to men with oligospermia, this study suffered from a lack of samples for these azoospermic men. More specifically, in chapters 3 and 4, the analysis of the folate cycle and DNA fragmentation included very few men with azoospermia. Therefore, these analyses mainly pertain to men with oligospermia and conclusions made for men with azoospermia should be made with caution. Additionally, the categorizations of oligospermia and severe oligospermia made in this study may be considered rather arbitrary as sperm concentrations can vary greatly within an individual and these categorizations were based on only one semen analysis. Therefore, multiple semen analyses may be necessary to adequately categorize individuals. It may also be useful to pool the men with oligospermia into a single group rather than subcategorize as in the present study. In addition to sample size issues, this study also lacked ethnic information. SNPs in folate metabolism enzymes and GST null alleles have previously been reported to be risk factors for male infertility only in certain ethnic populations (Ying et al., 2013; Liu et al., 2015). Future studies analysing these factors after segregation by ethnicity may yield more meaningful results. Since folate and DNA fragmentation may only induce subtle changes in DNA methylation, it may also be more useful to analyse these factors only in men which display the greatest defects in methylation compared to controls. 108  Future studies may benefit from next-generation sequencing methods. More recent studies of DNA methylation have migrated to using methylation microarrays which can give information on methylation throughout the entire genome. With new advanced technologies such as the Infinium Methylation EPIC array, some of the older technologies, such as the HumanMethylation450 BeadChip array or RRBS, may become a more affordable alternative that still generates a wealth of information. However, this is not necessarily a replacement for the techniques used in the present study as these are still used today for validation of results. Additionally, future studies should include concentrations of homocysteine and methionine for investigations of the folate cycle, as these biomolecules are more directly connected to DNMTs and DNA methylation. Finally, in evaluating DNA fragmentation, it is important to also include measures of ROS to provide evidence for the cause of the damage and concentrations of antioxidants, such as glutathione, may also support these claims. 5.3 Significance Although we could not provide evidence for DNA methylation aberrancies being caused by specific mechanisms, this study still adds to the growing wealth of data we have on the characteristics of infertile men. More specifically, the novel finding of a significant correlation between methylation at specific imprinted genes and folate concentration warrants further exploration. With the lack of significant findings, there are limited clinical applications for our study. However, the inclusion of folate and antioxidants in our study design can potentially be translated to clinics and patients in the future, as the mitigation of adverse epigenetic effects of infertility can occur through dietary supplements. Overall, the framework of the study presented here may provide more potential avenues for studying DNA methylation and male infertility in the future.  109   BIBLIOGRAPHY Aarabi, M. et al. High-dose folic acid supplementation alters the human sperm methylome and is influenced by the MTHFR C677T polymorphism. Hum Mol Genet 24, 6301–6313 (2015).  Aitken, R. J., Baker, M. A. & Sawyer, D. Oxidative stress in the male germ line and its role in the aetiology of male infertility and genetic disease. Reproductive BioMedicine Online 7, 65–70 (2003).  Amann, R. P. The Cycle of the Seminiferous Epithelium in Humans: A Need to Revisit? Journal of Andrology 29, 469–487 (2008).  Amor, D. J. & Halliday, J. A review of known imprinting syndromes and their association with assisted reproduction technologies. Hum. Reprod. 23, 2826–2834 (2008).  Aponte, P. M., VAN BRAGT, M., De Rooij, D. G. & Van Pelt, A. M. Spermatogonial stem cells: characteristics and experimental possibilities. Apmis 113, 727–742 (2005).  Bannister, A. J. & Kouzarides, T. Regulation of chromatin by histone modifications. Cell Res 21, 381–395 (2011).  Bird, A. DNA methylation patterns and epigenetic memory. Genes Dev. 16, 6–21 (2002).  Bisht, S., Faiq, M., Tolahunase, M. & Dada, R. Oxidative stress and male infertility. Nat Rev Urol advance online publication, (2017).  Blasco, M. A. The epigenetic regulation of mammalian telomeres. Nat Rev Genet 8, 299–309 (2007).  Boissonnas, C. C. et al. Specific epigenetic alterations of IGF2-H19 locus in spermatozoa from infertile men. Eur J Hum Genet 18, 73–80 (2009).  Buchard, A., Sanchez, J. J., Dalhoff, K. & Morling, N. Multiplex PCR Detection of GSTM1, GSTT1, and GSTP1 Gene Variants. The Journal of Molecular Diagnostics 9, 612–617 (2007).  Calogero, A. et al. Sperm aneuploidy in infertile men. Reproductive BioMedicine Online 6, 310–317 (2003).  Camprubí, C. et al. Semen samples showing an increased rate of spermatozoa with imprinting errors have a negligible effect in the outcome of assisted reproduction techniques. Epigenetics 7, 1115–1124 (2012).  Camprubí, C. et al. Lack of association of MTHFR rs1801133 polymorphism and CTCFL mutations with sperm methylation errors in infertile patients. J Assist Reprod Genet 30, 1125–1131 (2013).  110  Castro, R. et al. 5,10-methylenetetrahydrofolate reductase (MTHFR) 677C→T and 1298A→C mutations are associated with DNA hypomethylation. Journal of Medical Genetics 41, 454–458 (2004).  Chan, D. et al. Stability of the human sperm DNA methylome to folic acid fortification and short-term supplementation. Hum Reprod 32, 272–283 (2017).  Chen, Z. et al. Mice deficient in methylenetetrahydrofolate reductase exhibit hyperhomocysteinemia and decreased methylation capacity, with neuropathology and aortic lipid deposition. Hum Mol Genet 10, 433–444 (2001).  Chen, Z. & Riggs, A. D. DNA Methylation and Demethylation in Mammals. J. Biol. Chem. 286, 18347–18353 (2011).  Chen, H., Ruan, Y. C., Xu, W. M., Chen, J. & Chan, H. C. Regulation of male fertility by CFTR and implications in male infertility. Hum Reprod Update 18, 703–713 (2012).  Chiba, H. et al. DNA methylation errors in imprinting disorders and assisted reproductive technology. Pediatr Int 55, 542–549 (2013).  Craig, J. R., Jenkins, T. G., Carrell, D. T. & Hotaling, J. M. Obesity, male infertility, and the sperm epigenome. Fertility and Sterility 107, 848–859 (2017).  Culty, M. Gonocytes, from the Fifties to the Present: Is There a Reason to Change the Name?1. Biology of Reproduction 89, (2013).  Cuppens, H. & Cassiman, J.-J. CFTR mutations and polymorphisms in male infertility. International Journal of Andrology 27, 251–256 (2004).  Davies, M. J. et al. Reproductive Technologies and the Risk of Birth Defects. New England Journal of Medicine 366, 1803–1813 (2012).  Doornbos, M. E., Maas, S. M., McDonnell, J., Vermeiden, J. P. W. & Hennekam, R. C. M. Infertility, assisted reproduction technologies and imprinting disturbances: a Dutch study. Hum. Reprod. 22, 2476–2480 (2007).  Farhi, A. et al. Congenital malformations in infants conceived following assisted reproductive technology in comparison with spontaneously conceived infants. The Journal of Maternal-Fetal & Neonatal Medicine 26, 1171–1179 (2013).  Ferguson, K. A., Wong, E. C., Chow, V., Nigro, M. & Ma, S. Abnormal meiotic recombination in infertile men and its association with sperm aneuploidy. Hum Mol Genet 16, 2870–2879 (2007).  Ferlin, A., Arredi, B. & Foresta, C. Genetic causes of male infertility. Reproductive Toxicology 22, 133–141 (2006). 111  Filipponi, D. & Feil, R. Perturbation of genomic imprinting in oligozoospermia. Epigenetics 4, 27–30 (2009).  Friso, S. et al. A common mutation in the 5,10-methylenetetrahydrofolate reductase gene affects genomic DNA methylation through an interaction with folate status. Proc Natl Acad Sci U S A 99, 5606–5611 (2002).  Frommer, M. et al. A genomic sequencing protocol that yields a positive display of 5-methylcytosine residues in individual DNA strands. Proc. Natl. Acad. Sci. U.S.A. 89, 1827–1831 (1992).  Fullston, T., McPherson, N. O., Zander-Fox, D. & Lane, M. The most common vices of men can damage fertility and the health of the next generation. J Endocrinol JOE-16-0382 (2017). doi:10.1530/JOE-16-0382  Gaughan, D. J. et al. The methionine synthase reductase (MTRR) A66G polymorphism is a novel genetic determinant of plasma homocysteine concentrations. Atherosclerosis 157, 451–456 (2001).  Georgiou, I. et al. Genetic and epigenetic risks of intracytoplasmic sperm injection method. Asian J. Androl. 8, 643–673 (2006).  Gkountela, S. et al. DNA Demethylation Dynamics in the Human Prenatal Germline. Cell 161, 1425–1436 (2015).  Göb, E., Schmitt, J., Benavente, R. & Alsheimer, M. Mammalian Sperm Head Formation Involves Different Polarization of Two Novel LINC Complexes. PLOS ONE 5, e12072 (2010).  Gorrini, C., Harris, I. S. & Mak, T. W. Modulation of oxidative stress as an anticancer strategy. Nat Rev Drug Discov 12, 931–947 (2013).  Gunes, S., Arslan, M. A., Hekim, G. N. T. & Asci, R. The role of epigenetics in idiopathic male infertility. J Assist Reprod Genet 33, 553–569 (2016).  Hansen, M., Bower, C., Milne, E., de Klerk, N. & J. Kurinczuk, J. Assisted reproductive technologies and the risk of birth defects—a systematic review. Hum Reprod 20, 328–338 (2005).  Harewood, L. & Fraser, P. The impact of chromosomal rearrangements on regulation of gene expression. Hum Mol Genet 23, R76–R82 (2014).  Hassold, T. & Hunt, P. To err (meiotically) is human: the genesis of human aneuploidy. Nat Rev Genet 2, 280–291 (2001).  Heard, E. & Martienssen, R. A. Transgenerational Epigenetic Inheritance: Myths and Mechanisms. Cell 157, 95–109 (2014). 112   Heisey, A. S., Bell, E. M., Herdt-Losavio, M. L. & Druschel, C. Surveillance of congenital malformations in infants conceived through assisted reproductive technology or other fertility treatments. Birth Defects Research Part A: Clinical and Molecular Teratology 103, 119–126 (2015).  Heyting, C. Synaptonemal complexes: structure and function. Current Opinion in Cell Biology 8, 389–396 (1996).  Hill, P. W. S., Amouroux, R. & Hajkova, P. DNA demethylation, Tet proteins and 5-hydroxymethylcytosine in epigenetic reprogramming: An emerging complex story. Genomics 104, 324–333 (2014).  Hiura, H. et al. Characterization of DNA methylation errors in patients with imprinting disorders conceived by assisted reproduction technologies. Hum. Reprod. 27, 2541–2548 (2012).  Hiura, H. et al. Imprinting methylation errors in ART. Reprod Med Biol 13, 193–202 (2014).  Hughes, I. A. Minireview: Sex Differentiation. Endocrinology 142, 3281–3287 (2001).  Isidori, A. M., Giannetta, E. & Lenzi, A. Male hypogonadism. Pituitary 11, 171–180 (2008).  Jin, B., Li, Y. & Robertson, K. D. DNA Methylation. Genes Cancer 2, 607–617 (2011).  Jin, J.-M. & Yang, W.-X. Molecular regulation of hypothalamus–pituitary–gonads axis in males. Gene 551, 15–25 (2014).  Jones. To store or mature spermatozoa? The primary role of the epididymis. International Journal of Andrology 22, 57–67 (1999).  Kalmbach, K. H. et al. Telomeres and human reproduction. Fertility and Sterility 99, 23–29 (2013).  Karimian, M. & Hosseinzadeh Colagar, A. Methionine synthase A2756G transition might be a risk factor for male infertility: Evidences from seven case-control studies. Molecular and Cellular Endocrinology 425, 1–10 (2016).  Kelley-Quon, L. I., Tseng, C.-H., Janzen, C. & Shew, S. B. Congenital malformations associated with assisted reproductive technology: A California statewide analysis. Journal of Pediatric Surgery 48, 1218–1224 (2013).  Kirkpatrick, G., Ren, H., Liehr, T., Chow, V. & Ma, S. Meiotic and sperm aneuploidy studies in three carriers of Robertsonian translocations and small supernumerary marker chromosomes. Fertility and Sterility 103, 1162–1169.e7 (2015).  113  Klengel, T., Dias, B. G. & Ressler, K. J. Models of Intergenerational and Transgenerational Transmission of Risk for Psychopathology in Mice. Neuropsychopharmacology 41, 219–231 (2016).  Kobayashi, H. et al. Aberrant DNA methylation of imprinted loci in sperm from oligospermic patients. Hum. Mol. Genet. 16, 2542–2551 (2007).  Kobayashi, H. et al. DNA methylation errors at imprinted loci after assisted conception originate in the parental sperm. Eur J Hum Genet 17, 1582–1591 (2009).  Kohli, R. M. & Zhang, Y. TET enzymes, TDG and the dynamics of DNA demethylation. Nature 502, 472–479 (2013).  Koppers, A. J., Mitchell, L. A., Wang, P., Lin, M. & Aitken, R. J. Phosphoinositide 3-kinase signalling pathway involvement in a truncated apoptotic cascade associated with motility loss and oxidative DNA damage in human spermatozoa. Biochemical Journal 436, 687–698 (2011).  Lange, J. et al. Intrachromosomal homologous recombination between inverted amplicons on opposing Y-chromosome arms. Genomics 102, 257–264 (2013).  Laurentino, S. et al. Epigenetic germline mosaicism in infertile men. Hum. Mol. Genet. 24, 1295–1304 (2015).  Leclerc, D. et al. Cloning and mapping of a cDNA for methionine synthase reductase, a flavoprotein defective in patients with homocystinuria. PNAS 95, 3059–3064 (1998).  Lennartsson, A. & Ekwall, K. Histone modification patterns and epigenetic codes. Biochimica et Biophysica Acta (BBA) - General Subjects 1790, 863–868 (2009).  Li, J.-Y., Lees-Murdock, D. J., Xu, G.-L. & Walsh, C. P. Timing of establishment of paternal methylation imprints in the mouse. Genomics 84, 952–960 (2004).  Li, B. et al. Altered DNA Methylation Patterns of the H19 Differentially Methylated Region and the DAZL Gene Promoter Are Associated with Defective Human Sperm. PLoS ONE 8, e71215 (2013).  Lie, P. P. Y., Cheng, C. Y. & Mruk, D. D. Coordinating cellular events during spermatogenesis: a biochemical model. Trends in Biochemical Sciences 34, 366–373 (2009).  Lissens, W. et al. Cystic fibrosis and infertility caused by congenital bilateral absence of the vas deferens and related clinical entities. Hum Reprod 11, 55–80 (1996).  Liu, K., Zhao, R., Shen, M., Ye, J., Li, X., Huang, Y., Hua, L., Wang, Z. & Li, J. Role of genetic mutations in folate-related enzyme genes on Male Infertility. Scientific Reports 5, 15548 (2015).  114  Locasale, J. W. Serine, glycine and one-carbon units: cancer metabolism in full circle. Nat Rev Cancer 13, 572–583 (2013).  Louie, K., Minor, A., Ng, R., Poon, K., Chow, V. & Ma, S. Evaluation of DNA methylation at imprinted DMRs in the spermatozoa of oligozoospermic men in association with MTHFR C677T genotype. Andrology 4(5), 825–831 (2016).  Ma, S., Ferguson, K. A., Arsovska, S., Moens, P. & Chow, V. Reduced recombination associated with the production of aneuploid sperm in an infertile man: a case report. Hum Reprod 21, 980–985 (2006).  Manipalviratn, S., DeCherney, A. & Segars, J. Imprinting disorders and assisted reproductive technology. Fertility and Sterility 91, 305–315 (2009).  Manku, G. & Culty, M. Mammalian gonocyte and spermatogonia differentiation: recent advances and remaining challenges. Reproduction 149, R139–R157 (2015).  Manning, M., Lissens, W., Weidner, W. & Liebaers, I. DNA methylation analysis in immature testicular sperm cells at different developmental stages. Urologia Internationalis 67, 151–155 (2001).  Marques, C. J., Carvalho, F., Sousa, M. & Barros, A. Genomic imprinting in disruptive spermatogenesis. The Lancet 363, 1700–1702 (2004).  Marques, C. J. et al. Abnormal methylation of imprinted genes in human sperm is associated with oligozoospermia. Mol. Hum. Reprod. 14, 67–74 (2008).  Marques, C. J. et al. Methylation defects of imprinted genes in human testicular spermatozoa. Fertility and Sterility 94, 585–594 (2010).  Marston, A. L. & Amon, A. Meiosis: cell-cycle controls shuffle and deal. Nature Reviews Molecular Cell Biology 5, 983–997 (2004).  Matsuo, K. et al. Association between polymorphisms of folate- and methionine-metabolizing enzymes and susceptibility to malignant lymphoma. Blood 97, 3205–3209 (2001).  Messerschmidt, D. M., Knowles, B. B. & Solter, D. DNA methylation dynamics during epigenetic reprogramming in the germline and preimplantation embryos. Genes Dev. 28, 812–828 (2014).  Mfady, D. S. et al. Associations of variants in MTHFR and MTRR genes with male infertility in the Jordanian population. Gene 536, 40–44 (2014).  Millar, R. P. et al. Gonadotropin-Releasing Hormone Receptors. Endocrine Reviews 25, 235–275 (2004).  115  Miller, D., Brinkworth, M. & Iles, D. Paternal DNA packaging in spermatozoa: more than the sum of its parts? DNA, histones, protamines and epigenetics. Reproduction 139, 287–301 (2010).  Miller, M. P., Amon, A. & Ünal, E. Meiosis I: when chromosomes undergo extreme makeover. Current Opinion in Cell Biology 25, 687–696 (2013).  Minor, A., Chow, V. & Ma, S. Aberrant DNA methylation at imprinted genes in testicular sperm retrieved from men with obstructive azoospermia and undergoing vasectomy reversal. Reproduction 141, 749–757 (2011).  Miyamoto, T., Minase, G., Okabe, K., Ueda, H. & Sengoku, K. Male infertility and its genetic causes. J. Obstet. Gynaecol. Res. 41, 1501–1505 (2015).  Montjean, D. et al. Methylation changes in mature sperm deoxyribonucleic acid from oligozoospermic men: assessment of genetic variants and assisted reproductive technology outcome. Fertility and Sterility 100, 1241–1247.e2 (2013).  Montjean, D. et al. Sperm global DNA methylation level: association with semen parameters and genome integrity. Andrology (2015). doi:10.1111/andr.12001  Morelli, M. A. & Cohen, P. E. Not all germ cells are created equal: Aspects of sexual dimorphism in mammalian meiosis. Reproduction 130, 761–781 (2005).  Mruk, D. D. & Cheng, C. Y. The Mammalian Blood-Testis Barrier: Its Biology and Regulation. Endocr Rev 36, 564–591 (2015).  Mussa, A. et al. Assisted Reproductive Techniques and Risk of Beckwith-Wiedemann Syndrome. Pediatrics 140, e20164311 (2017).  Nagaoka, S. I., Hassold, T. J. & Hunt, P. A. Human aneuploidy: mechanisms and new insights into an age-old problem. Nat Rev Genet 13, 493–504 (2012).  Nanassy, L. & Carrell, D. T. Paternal effects on early embryogenesis. J Exp Clin Assist Reprod 5, 2 (2008).  Neto, F. T. L., Bach, P. V., Najari, B. B., Li, P. S. & Goldstein, M. Spermatogenesis in humans and its affecting factors. Seminars in Cell & Developmental Biology 59, 10–26 (2016).  O’Donnell, L., Nicholls, P. K., O’Bryan, M. K., McLachlan, R. I. & Stanton, P. G. Spermiation: The process of sperm release. Spermatogenesis 1, 14–35 (2011).  O’Donnell, L. Mechanisms of spermiogenesis and spermiation and how they are disturbed. Spermatogenesis 4, e979623 (2014).  116  O’Flynn O’Brien, K. L., Varghese, A. C. & Agarwal, A. The genetic causes of male factor infertility: A review. Fertility and Sterility 93, 1–12 (2010).  Oliva, R. & Luís Ballescà, J. Altered histone retention and epigenetic modifications in the sperm of infertile men. Asian J Androl 14, 239–240 (2012).  Oliver, V. F. et al. Defects in imprinting and genome-wide DNA methylation are not common in the in vitro fertilization population. Fertility and Sterility 97, 147–153.e7 (2012).  Olteanu, H. & Banerjee, R. Human Methionine Synthase Reductase, a Soluble P-450 Reductase-like Dual Flavoprotein, Is Sufficient for NADPH-dependent Methionine Synthase Activation. J. Biol. Chem. 276, 35558–35563 (2001).  Olteanu, H., Munson, T. & Banerjee, R. Differences in the Efficiency of Reductive Activation of Methionine Synthase and Exogenous Electron Acceptors between the Common Polymorphic Variants of Human Methionine Synthase Reductase. Biochemistry 41, 13378–13385 (2002).  Pawlowski, W. P. & Cande, W. Z. Coordinating the events of the meiotic prophase. Trends in Cell Biology 15, 674–681 (2005).  Peschansky, V. J. & Wahlestedt, C. Non-coding RNAs as direct and indirect modulators of epigenetic regulation. Epigenetics 9, 3–12 (2014).  Peters, J. The role of genomic imprinting in biology and disease: an expanding view. Nat Rev Genet 15, 517–530 (2014).  Plasschaert, R. N. & Bartolomei, M. S. Genomic imprinting in development, growth, behavior and stem cells. Development 141, 1805–1813 (2014).  Poplinski, A., Tüttelmann, F., Kanber, D., Horsthemke, B. & Gromoll, J. Idiopathic male infertility is strongly associated with aberrant methylation of MEST and IGF2/H19 ICR1. International Journal of Andrology 33, 642–649 (2010).  Qin, J. et al. Assisted reproductive technology and risk of congenital malformations: a meta-analysis based on cohort studies. Arch Gynecol Obstet 292, 777–798 (2015).  Rajabi, H., Mohseni-kouchesfehani, H., Eslami-Arshaghi, T. & Salehi, M. Sperm DNA fragmentation affects epigenetic feature in human male pronucleus. Andrologia (2017). doi:10.1111/and.12800  Rasmussen, K. D. & Helin, K. Role of TET enzymes in DNA methylation, development, and cancer. Genes Dev. 30, 733–750 (2016).  Rato, L. et al. Metabolic regulation is important for spermatogenesis. Nature Reviews Urology 9, 330–338 (2012).  117  Ren, H. et al. Altered Crossover Distribution and Frequency in Spermatocytes of Infertile Men with Azoospermia. PLOS ONE 11, e0156817 (2016).  Rimm, A. A., Katayama, A. C. & Katayama, K. P. A meta-analysis of the impact of IVF and ICSI on major malformations after adjusting for the effect of subfertility. J Assist Reprod Genet 28, 699–705 (2011).  Rowe, S. M., Miller, S. & Sorscher, E. J. Cystic Fibrosis. New England Journal of Medicine 352, 1992–2001 (2005).  Rycke, M. D., Liebaers, I. & Steirteghem, A. V. Epigenetic risks related to assisted reproductive technologies Risk analysis and epigenetic inheritance. Hum. Reprod. 17, 2487–2494 (2002).  Sakian, S. et al. Altered gene expression of H19 and IGF2 in placentas from ART pregnancies. Placenta 36, 1100–1105 (2015).  Sato, A. et al. Assessing loss of imprint methylation in sperm from subfertile men using novel methylation polymerase chain reaction Luminex analysis. Fertility and Sterility 95, 129–134.e4 (2011).  She, Z.-Y. & Yang, W.-X. Sry and SoxE genes: How they participate in mammalian sex determination and gonadal development? Seminars in Cell & Developmental Biology 63, 13–22 (2017).  Shima, Y. & Morohashi, K. Leydig progenitor cells in fetal testis. Molecular and Cellular Endocrinology 445, 55–64 (2017).  Smallwood, S. A. & Kelsey, G. De novo DNA methylation: a germ cell perspective. Trends in Genetics 28, 33–42 (2012).  Stuppia, L., Franzago, M., Ballerini, P., Gatta, V. & Antonucci, I. Epigenetics and male reproduction: the consequences of paternal lifestyle on fertility, embryo development, and children lifetime health. Clin Epigenetics 7, (2015).  Tang, W. W. C. et al. A Unique Gene Regulatory Network Resets the Human Germline Epigenome for Development. Cell 161, 1453–1467 (2015).  Tang, W. W. C., Kobayashi, T., Irie, N., Dietmann, S. & Surani, M. A. Specification and epigenetic programming of the human germ line. Nature Reviews Genetics 17, 585–600 (2016).  Tavalaee, M., Razavi, S. & Nasr-Esfahani, M. H. Influence of sperm chromatin anomalies on assisted reproductive technology outcome. Fertility and Sterility 91, 1119–1126 (2009).  Tenorio, J. et al. Clinical and molecular analyses of Beckwith–Wiedemann syndrome: Comparison between spontaneous conception and assisted reproduction techniques. Am. J. Med. Genet. 170, 2740–2749 (2016). 118   Tierling, S. et al. Assisted reproductive technologies do not enhance the variability of DNA methylation imprints in human. J Med Genet 47, 371–376 (2010).  Tunc, O. & Tremellen, K. Oxidative DNA damage impairs global sperm DNA methylation in infertile men. J Assist Reprod Genet 26, 537–544 (2009).  Tüttelmann, F. & Gromoll, J. Novel genetic aspects of Klinefelter’s syndrome. Mol Hum Reprod 16, 386–395 (2010).  Urdinguio, R. G. et al. Aberrant DNA methylation patterns of spermatozoa in men with unexplained infertility. Hum. Reprod. 30, 1014–1028 (2015).  Vadakkadath Meethal, S. & Atwood, C. S. Alzheimer’s disease: the impact of age-related changes in reproductive hormones: The role of hypothalamic-pituitary-gonadal hormones in the normal structure and functioning of the brain. CMLS Cellular and Molecular Life Sciences 62, 257–270 (2005).  Vaissière, T., Sawan, C. & Herceg, Z. Epigenetic interplay between histone modifications and DNA methylation in gene silencing. Mutation Research/Reviews in Mutation Research 659, 40–48 (2008).  Vincent, R. N., Gooding, L. D., Louie, K., Chan Wong, E. & Ma, S. Altered DNA methylation and expression of PLAGL1 in cord blood from assisted reproductive technology pregnancies compared with natural conceptions. Fertility and Sterility 106, 739–748.e3 (2016).  von Meyenn, F. & Reik, W. Forget the Parents: Epigenetic Reprogramming in Human Germ Cells. Cell 161, 1248–1251 (2015).  Weaver, J. R. & Bartolomei, M. S. Chromatin regulators of genomic imprinting. Biochimica et Biophysica Acta (BBA) - Gene Regulatory Mechanisms 1839, 169–177 (2014).  Weber, A. R. et al. Biochemical reconstitution of TET1–TDG–BER-dependent active DNA demethylation reveals a highly coordinated mechanism. Nature Communications 7, ncomms10806 (2016).  WHO laboratory manual for the examination and processing of human semen. (World Health Organization, 2010).  Wolthers, K. R. & Scrutton, N. S. Protein Interactions in the Human Methionine Synthase−Methionine Synthase Reductase Complex and Implications for the Mechanism of Enzyme Reactivation. Biochemistry 46, 6696–6709 (2007).  Wu, H. & Zhang, Y. Mechanisms and functions of Tet protein-mediated 5-methylcytosine oxidation. Genes Dev. 25, 2436–2452 (2011).  119  Xu, G.-L. & Wong, J. Oxidative DNA demethylation mediated by Tet enzymes. Natl Sci Rev 2, 318–328 (2015).  Yang, M. & Vousden, K. H. Serine and one-carbon metabolism in cancer. Nat Rev Cancer 16, 650–662 (2016).  Ying, H.-Q., Qi, Y., Pu, X.-Y., Liu, S.-R. & A, Z.-C. Association of GSTM1 and GSTT1 Genes with the Susceptibility to Male Infertility: Result from a Meta-Analysis. Genet Test Mol Biomarkers 17, 535–542 (2013).  Zheng, H.-Y. et al. Study of DNA methylation patterns of imprinted genes in children born after assisted reproductive technologies reveals no imprinting errors: A pilot study. Exp Ther Med 2, 751–755 (2011).  Zhu, J. L., Basso, O., Obel, C., Bille, C. & Olsen, J. Infertility, infertility treatment, and congenital malformations: Danish national birth cohort. BMJ 333, 679 (2006).  Zhu, J.-K. Active DNA Demethylation Mediated by DNA Glycosylases. Annual Review of Genetics 43, 143–166 (2009).  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0357954/manifest

Comment

Related Items