Marta Teperek Clare Hall College University of Cambridge March 2014 This dissertation is submitted for the degree of Doctor of Philosophy Supervisor: Dr Jerome Jullien Co-supervisor: Dr John Gurdon Programming of the paternal nucleus for embryonic development Preface The work described in this dissertation has been conducted in the Gurdon Institute, University of Cambridge, Cambridge, United Kingdom, under the supervision of Dr Jerome Jullien and Dr John Gurdon. This dissertation is the result of my own work and includes nothing which is the outcome of work done in collaboration except where specifically indicated in the text. The work described here has not been submitted to any other university. Marta Teperek Statement of length This dissertation does not exceed the prescribed word limit (i.e., it is below 60 000 words, excluding bibliography, figures and appendices). Marta Teperek Acknowledgements I am extremely grateful to Dr Jerome Jullien, my supervisor, who provided me with his constant, day-to-day support during my rotation in the Gurdon laboratory and then later on through the whole period of my PhD. I am grateful not only for his scientific input, creative ideas, but also for being a patient, understanding and motivating supervisor. I am also immensely grateful for Dr Jullien’s sense of humour, which made my days in the lab enjoyable and entertaining. I am very grateful to Dr John Gurdon, my co-supervisor, who allowed me to do my rotation project as well as my PhD in his laboratory. I am grateful for his input into my work and all his valuable advice. I am also extremely grateful to Dr Gurdon for giving me a lot of freedom in my research, which stimulated my development as an independent scientist. I would like to thank Dr Vincent Gaggioli, Dr Angela Simeone and Dr Kei Miyamoto for their great help with performing the experiments as well as with the experimental design and analysis. Their input was vital for the progress of this project. I am also grateful to them for being wonderful friends, on top of being excellent scientific collaborators. I would like to thank all my colleagues from the Gurdon laboratory for creating a nice working atmosphere in the lab, and who were never tired of my questions and who were also fantastic colleagues. I am thankful to Dr Charles Bradshaw and Dr George Allen for performing vital bioinformatic analyses for this project. I am extremely grateful to all my friends from the Gurdon Institute, and particularly to Beata Wyspianska, for their endless support and who made my time in the Gurdon Institute such a great experience. I would like to thank my collaborators from outside the Institute for all the scientific help and input into this project. I am especially grateful to Dr Antoine Peters for hosting me in his laboratory and for teaching me some of his expert techniques. I would like to thank David Simpson for taking care of our frogs; Renata Feret and other member of the proteomics facility for performing 2-DIGE and mass spectrometry analyses of the data; and Dr Rachael Walker for performing the flow cytometry analysis for me. I am grateful to my dear friend, Jerzy Wilczynski, for his expert advice on the statistical analysis conducted in Chapters 3 and 5 of this thesis. I would also like to thank him for being a never-ending source of motivation and inspiration for my career and life choices. I am immensely grateful to my parents and to my sister for their enormous support during all my time in Cambridge and beyond. I would like to thank my parents for their guidance in life and for always telling me (and being a wonderful example of this) that education and self- development is important in one’s life. I would like to thank them and my sister for providing me with constant motivation and for never giving up on me. I could have never gone so far without them. Finally, I would like to thank the funding bodies: the Wellcome Trust and the Medical Research Council UK for their financial support. Summary Historically, sperm has been considered merely as a carrier of genetic material at fertilisation. However, it is known that sperm supports embryonic development better than other cell types, suggesting that it might also have additional important, non-genetic contributions to embryonic development. The work described in this dissertation focuses on identifying the molecular determinants of developmental programming of sperm. First, the development of embryos derived from sperm and spermatids, immature precursors of sperm was compared. Sperm-derived embryos developed significantly better than spermatid- derived embryos. Further research aiming to identify the reasons for the developmental advantage of sperm led to the identification of proteins that are present specifically in sperm and not in spermatids. Moreover, egg factors which are preferentially incorporated into the sperm, but not into the spermatid chromatin were identified with the use of egg extracts, suggesting that the chromatin of sperm could be programmed to interact with the components of the egg. Subsequently, the reasons for developmental failure of spermatid-derived embryos were investigated. By comparing the sperm with spermatids it was shown that the programming of sperm to support efficient development is linked to its special ability to regulate expression of developmentally-important embryonic genes, and not to its ability to support DNA replication or rRNA production. Further characterisation of the sperm and spermatid chromatin with the use of genome-wide sequencing allowed me to link the correct regulation of gene expression in the embryo with a certain combination of epigenetic marks in the sperm, but not in the spermatid chromatin. Finally, it is shown that enzymatic removal of epigenetic modifications at fertilisation leads to misregulation of gene expression. This therefore suggests that epigenetic information contained in parental genomes at fertilisation is required for a proper regulation of embryonic transcription. My results support the hypothesis that the sperm is not only a carrier of genetic material, but also provides the embryo with epigenetic information for regulation of transcription after fertilisation. I believe that these findings advance our current understanding of the nature and mechanisms of sperm programming for embryonic development, and are important contributions to the emerging field of transgenerational inheritance of epigenetic traits in general. Table of Contents Chapter 1 .................................................................................................................................... 1 Introduction ................................................................................................................................ 1 1.1. Stabilisation of the differentiated state by epigenetic mechanisms ............................ 1 1.1.1. DNA methylation ................................................................................................. 2 1.1.2. Post-translational histone modifications .............................................................. 3 1.1.3. Histone variants ................................................................................................... 4 1.1.4. Nucleosome density ............................................................................................. 6 1.1.5. Cooperative action of different epigenetic mechanism to achieve stable maintenance of a differentiated state .................................................................................. 7 1.2. Reprogramming as the reversal of differentiation..................................................... 10 1.3. Sperm is a highly specialised cell ............................................................................. 13 1.3.1. Global changes occurring during spermiogenesis ............................................. 13 1.3.2. Epigenetic marks are retained in the chromatin of mature sperm ..................... 17 1.3.3. Chromatin composition of the mature sperm in Xenopus laevis ....................... 19 1.4. Why is the sperm better than a somatic cell at supporting embryonic development? 20 1.4.1. Protamines.......................................................................................................... 22 1.4.2. Sperm-derived transcriptional regulators and RNAs .............................................. 23 1.4.3. Sperm-derived epigenetic marks ............................................................................ 25 1.5. Evidence for the sperm developmental advantages .................................................. 26 1.5.1. DNA replication ................................................................................................. 26 1.5.2. rRNA synthesis .................................................................................................. 27 1.5.3. mRNA synthesis ................................................................................................ 28 1.6. Work described in this thesis .................................................................................... 30 Chapter 2 .................................................................................................................................. 33 Experimental procedures ......................................................................................................... 33 2.1. Separation of sperm and spermatids.............................................................................. 33 2.2. FACS analysis ............................................................................................................... 34 2.3. Sperm and spermatid nuclei preparation, intra-cytoplasmic sperm injections (ICSI) to non-enucleated and to enucleated eggs, and embryo culture ............................................... 35 2.4. Nuclear transfer ............................................................................................................. 36 2.5. Protein preparation for the mass spectrometry analysis ................................................ 36 2.6. Interphase egg extract preparation ................................................................................ 37 2.7. Egg extract treatment and protein isolation for mass spectrometry analysis of egg proteins incorporated into sperm or spermatid chromatin. .................................................. 38 2.8. Immunoblotting analysis. .............................................................................................. 39 2.9. Molecular cloning of candidate sperm factors and mRNA synthesis ........................... 41 2.10. Cell culture, transfection, immunostaining and microscopic analysis ........................ 42 2.11. Cell squashing for ploidy assessment.......................................................................... 43 2.12. Injection of 1-cell embryos with mRNA ..................................................................... 43 2.13. Injection of mRNA or oligonucleotides into the oocytes ............................................ 44 2.14. RNA extraction and qRT-PCR analysis ...................................................................... 47 2.15. Analysis of DNA replication ....................................................................................... 48 2.16 Pulldown of BrUTP-labelled RNA from haploid sperm- and spermatid-derived embryos ................................................................................................................................ 50 2.17. Preparation of cDNA library for sequencing .............................................................. 51 2.18. Preparation of ChIP-seq samples ................................................................................ 51 2.19. Bioinformatic analyses ................................................................................................ 52 2.19.1. Sequencing of libraries ......................................................................................... 52 2.19.2. Xenopus laevis transcriptome ............................................................................... 52 2.19.3. Filtering sequencing data ...................................................................................... 53 2.19.4. Genome based RNA-Seq mapping ....................................................................... 53 2.19.5. Differential expression ......................................................................................... 54 2.19.6. Heatmaps for differentially expressed genes ........................................................ 54 2.19.7. Genome-wide correlation analysis of ChIP-seq data ........................................... 55 2.19.8. Histone methylation level analysis ....................................................................... 55 2.19.9. Peak calling for histone marks .............................................................................. 56 2.19.10. Statistical testing of ChIP-seq data ..................................................................... 56 Chapter 3 .................................................................................................................................. 57 Sperm-derived embryos develop better than nuclear transfer-derived and spermatid-derived embryos .................................................................................................................................... 57 3.1. Introduction ................................................................................................................... 57 3.2. Separation of Xenopus laevis sperm and spermatids .................................................... 58 3.3. Sperm-derived embryos develop better than spermatid-derived embryos .................... 60 3.4. Nuclear transfer-derived embryos develop with a similarly low efficiency as spermatid-derived embryos and worse than sperm-derived embryos. ................................. 64 3.5. Summary ....................................................................................................................... 66 Chapter 4 .................................................................................................................................. 67 Identification of proteins present in sperm, spermatids and incorporated into sperm and spermatids from the egg extract ............................................................................................... 67 4.1. Introduction ................................................................................................................... 67 4.2. Sperm and spermatids differ in their nuclear protein composition ............................... 68 4.3. Sperm and spermatids bind distinct egg factors ............................................................ 75 4.4. Validation of mass spectrometry results by immunobloting. ........................................ 82 4.5. Summary ....................................................................................................................... 84 Chapter 5 .................................................................................................................................. 85 Functional assessment of candidate reprogramming factors ................................................... 85 5.1. Introduction ................................................................................................................... 85 5.2. Functional assessment of sperm-specific proteins ........................................................ 87 5.2.1. Experimental design ............................................................................................... 87 5.2.2. Cloning and ectopic expression of sperm-specific factors ..................................... 88 5.2.3. Validation of UV treatment length required for the nuclear transfer procedure .... 89 5.2.4. Overexpression of candidate sperm factors does not increase the efficiency of nuclear transfer. ................................................................................................................ 96 5.3. Functional assessment of egg factors preferentially associating with the sperm chromatin .............................................................................................................................. 98 5.3.1. Experimental design ............................................................................................... 98 5.3.2. Validation of the antibodies .................................................................................. 100 5.3.3. Downregulation of the mRNA encoding the selected factors .............................. 101 5.3.4. Protein levels of the candidate egg factors are not reproducibly downregulated by the antisense oligonucleotides ........................................................................................ 105 5.4. Summary and discussion ............................................................................................. 108 Chapter 6 ................................................................................................................................ 110 Characterisation of the developmental defects of spermatid-derived embryos ..................... 110 6.1. Introduction ................................................................................................................. 110 6.2. Spermatids replicate their DNA as efficiently as sperm ............................................. 112 6.3. Spermatid-derived RNA is not deleterious for embryonic development .................... 119 6.4. Haploid paternal embryos as a tool for a specific assessment of transcription from the paternally-derived chromatin ............................................................................................. 122 6.5. rRNA synthesis occurs normally in spermatid-derived embryos................................ 124 6.6. Developmentally-important mRNAs are misexpressed in spermatid-derived embryos ............................................................................................................................................ 126 6.7. mRNAs misexpressed in spermatid-derived embryos are Polycomb targets in human sperm .................................................................................................................................. 128 6.8. Summary ..................................................................................................................... 132 Chapter 7 ................................................................................................................................ 133 Epigenetic profiling of sperm and spermatids ....................................................................... 133 7.1. Introduction ................................................................................................................. 133 7.2. Parentally-derived H3K27me3 is necessary for correct gene expression in embryos 135 7.3. General characterisation of chromatin structure in sperm and spermatids ................. 140 7.4. ChIP-seq analysis for H3K27me3 does not reveal differences between sperm and spermatids........................................................................................................................... 148 7.4.1. Overall methylation levels analysis ...................................................................... 149 7.4.2. Peak analysis......................................................................................................... 150 7.4.3. Summary – H3K27me3 marks are not different between sperm and spermatids 151 7.5. Misregulated genes have more H3K4me2/3 activating marks in spermatids than in sperm .................................................................................................................................. 157 7.5.1. Overall methylation levels analysis ...................................................................... 157 7.5.2. Peak analysis......................................................................................................... 158 7.5.3. Summary – H3K4me2/3 marks are more abundant in spermatids ....................... 159 7.6. Summary ..................................................................................................................... 163 Chapter 8 ................................................................................................................................ 164 Discussion .............................................................................................................................. 164 8.1. Are the defects of spermatid-derived embryos a consequence of gene misexpression? ............................................................................................................................................ 165 8.2. Model for epigenetic programming of the sperm nucleus .......................................... 167 8.3. How are the epigenetic marks transmitted to the embryo? ......................................... 171 8.4. What are the epigenetic changes occurring during spermiogenesis in Xenopus laevis? ............................................................................................................................................ 177 8.5. How the results obtained add to the current knowledge on sperm programming? ..... 179 8.6. Results described in this thesis in the context of the current knowledge on transgenerational inheritance of epigenetic information. ................................................... 180 Chapter 9 ................................................................................................................................ 183 References .............................................................................................................................. 183 Chapter 10 .............................................................................................................................. 201 Appendices ............................................................................................................................. 201 Appendix 1 ......................................................................................................................... 202 Review article: ‘Epigenetic reprogramming: is deamination key to active DNA demethylation?’ .............................................................................................................. 202 Appendix 2 ......................................................................................................................... 215 Review article: ‘Nuclear reprogramming of sperm and somatic nuclei in eggs and oocytes.’ .......................................................................................................................... 215 Appendix 3 ......................................................................................................................... 233 Book chapter: ‘Cloning of Amphibia’ (part of ‘Principles of Cloning’) ........................ 233 Appendix 4 ......................................................................................................................... 244 Research article: ‘Nuclear Wave1 is required for reprogramming transcription in oocytes and for normal development.’ ........................................................................................ 244 Chapter 1: Introduction 1 Chapter 1 Introduction 1.1. Stabilisation of the differentiated state by epigenetic mechanisms Vertebrate development starts when the sperm fertilises the egg. This leads to the formation of a totipotent zygote, which subsequently replicates its DNA and undergoes series of cell divisions. Later on embryonic cells start to differentiate to form multiple tissues and organs constituting the adult body. It is very important that such differentiated cells forming particular tissues do not de-differentiate and do not change their fates to other cell types, as this could lead to tissue malfunction and tumour formation. That is why a fully differentiated state is thought to be safe-guarded by several layers of stabilising mechanisms. Virtually all the cells in the body have the same genetic content, as evidenced by nuclear transfer experiments (see below) (Gurdon 1962, Gurdon & Uehlinger 1966, Hochedlinger & Jaenisch 2002, Eggan et al. 2004, Sung et al. 2006). This means that differential expression of genes as cells specialise in their developmental pathways is not the result of gene loss or gain. In other words, features that make cells in a body different from each other are epigenetic. The word ‘epigenetic’ is derived from the Greek preposition ‘epi’ meaning ‘above, outside, besides’ combined with the word ‘genetic’, therefore meaning ‘outside genetic’. Currently this definition is used to describe heritable, non-genetic changes in cellular states (Bonasio et al. 2010). Interestingly, it has been observed that the DNA on chromosome (and in the nucleus) tends to be clustered according to its transcriptional activity: regions of chromatin which are transcriptionally active and decondensed are termed ‘euchromatin’ and regions which are transcriptionally repressed and densely packed are termed ‘heterochromatin’. Epigenetic mechanisms were proposed to be involved in the stabilisation and separation of Chapter 1: Introduction 2 these two different states (Lamond & Earnshaw 1998, Noma et al. 2001). There are various epigenetic mechanisms that can contribute to transcriptional activation and silencing. The best characterised epigenetic mechanisms stabilising the expression states of genes are DNA methylation, post-translational histone modifications, presence of histone variants or density of nucleosomes itself. Such mechanisms can act cooperatively to achieve multiple layers of epigenetic mechanisms ensuring stable maintenance of chromatin states. All these mechanisms are briefly reviewed below. 1.1.1. DNA methylation Methylation of cytosine is the best described epigenetic modification occurring on the DNA itself. Cytosines in DNA can be unmethylated or methylated (or be an intermediate between the two states, for example 5-hydroxymethylcytosine, which is however not discussed here). Methylation occurs by covalent modification of cytosine: by an addition of a methyl group to its fifth carbon, which creates 5-methyl-cytosine (5meC). Methylation can occur on cytosines in various contexts (Lister et al. 2009), but 5meC in CpG dinucleotides is the best described (Doerfler 2008). CpG dinucleotides are usually overrepresented in gene promoters. Methylation of cytosines in such CpG-rich promoters often leads to gene silencing (Boyes & Bird 1991, Boyes & Bird 1992, Hsieh 1994, Schubeler et al. 2000, Chen et al. 2001a, Song et al. 2005). Conversely, the removal of methyl group from 5meC (DNA demethylation), leads to gene expression (Benvenuto et al. 1996, Papageorgis et al. 2010, Stengel et al. 2010). DNA methylation is implied in the establishment and/or the maintenance of correct gene expression patterns during development, differentiation and tissue specification (Maatouk et al. 2006, Song et al. 2009). Interestingly, comparison of DNA methylation profiles from 17 distinct adult mouse tissues led to identification of tissue- Chapter 1: Introduction 3 specific DNA methylation sites (Hon et al. 2013). This finding points towards important roles of DNA methylation during organism development and specialisation. It is therefore not surprising that alteration of these methylation patterns, on a genome-wide, or on a loci- specific level, is associated with genomic instability and with cancers (Gaudet et al. 2003, Reddington et al. 2013). In support of this hypothesis, it has been well documented that during the germline specification in mouse, DNA methylation patterns are globally erased to allow reprogramming of the germ cells and to allow them to re-gain the pluripotency (Surani et al. 2007, Hajkova et al. 2010, Popp et al. 2010). 1.1.2. Post-translational histone modifications Post-translational histone modifications are on the other hand not related to changes to DNA bases itself, but to histones – small basic proteins packing the DNA within the nucleus. There are four major types of histones which form the nucleosome – H2A, H2B, H3 and H4. Furthermore, inter-nucleosomal regions of DNA are bound by linker histone H1. All these histones can be subjected to post-translational modifications, which can affect the chromatin compaction and in turn lead to gene expression or repression or to stabilisation/destabilisation of these states (Strahl & Allis 2000, Jenuwein & Allis 2001, Godde & Ura 2008). For example, it is known that dimethylation or trimethylation of lysine 4 on histone 3 (H3K4me2/3), methylation of argininie 17 or arginine 26 of histone H3 or histone acetylation are associated with transcriptional activation (Simpson 1978, Chahal et al. 1980, Turner et al. 1992, Jeppesen & Turner 1993, Lee et al. 1993, Ogryzko et al. 1996, Noma et al. 2001, Bernstein et al. 2002, Santos-Rosa et al. 2002, Torres-Padilla et al. 2007, Wu et al. 2009). On the other hand, ubiquitination of lysine 119 of histone H2A (H2AK119Ub), di- or trimethylation of lysine 9 of histone H3 (H3K9me2/3) or trimethylation of lysine 27 of Chapter 1: Introduction 4 histone H3 (H3K27me3) are associated with transcriptional repression (Lachner et al. 2001, Nakayama et al. 2001, Noma et al. 2001, Cao et al. 2002, Czermin et al. 2002, Muller et al. 2002, Snowden et al. 2002, Plath et al. 2003, Wang et al. 2004, Richly et al. 2010). The effect of post-translational histone modifications on the chromatin state can be direct or indirect. It has been demonstrated that acetylation directly decreases the affinity of histones for DNA and therefore makes DNA more accessible to transcription factors (Cary et al. 1982). Non-modified histones are positively charged, which facilitates their binding to negatively charged DNA. Acetylation brings a negative charge to the histone, which therefore decreases its affinity to DNA. Interestingly, post-translational histone marks can also indirectly affect the chromatin structure, by the recruitment of proteins that recognize such marks, known as chromatin ‘readers’. Acetylated histones are recognized by proteins containing bromodomains, for example Brd2, Brd3 or Brd4 (Bromodomain proteins 2, 3, 4), which, by various mechanisms, can in turn facilitate transcription (Dey et al. 2003, Kanno et al. 2004, LeRoy et al. 2008, Dey et al. 2009, Umehara et al. 2010a, Umehara et al. 2010b, Zhao et al. 2011, Draker et al. 2012). It has been shown that Brd4-bound acetylated histones can recruit transcription elongation factor P-TEFb, which leads to transcriptional elongation and gene activation (Jang et al. 2005, Yang et al. 2005, Hargreaves et al. 2009, Zippo et al. 2009, Liu et al. 2013). Conversely, recognition and binding of repressive H3K9me2/3 mark by heterochromatin protein 1 (HP1) leads to heterochromatin formation and transcriptional repression (Bannister et al. 2001, Lachner et al. 2001, Motamedi et al. 2008). 1.1.3. Histone variants There are also numerous histone variants that can compose the chromatin and presence of these variants instead of canonical histones also affects the chromatin Chapter 1: Introduction 5 accessibility (Talbert et al. 2012). For example, H3 variant, histone H3.3 is associated with open chromatin state and active transcription and is preferentially incorporated into the paternal pronucleus after fertilisation (McKittrick et al. 2004, Loppin et al. 2005, Torres- Padilla et al. 2006, Jullien et al. 2012). Interestingly, genome-wide analysis of H3.3 binding revealed that on top of being enriched at actively transcribed genes (transcriptional start sites and gene bodies), H3.3 was also enriched at transcriptional start sites (but not gene bodies) of repressed genes and was also shown to be important for the establishment of the pericentromeric heterochromatin mouse embryos (Goldberg et al. 2010, Santenard et al. 2010). Another histone variant, H2A.Z, was shown to occupy developmentally-important gene promoters in embryonic stem cells (ES cells) and its downregulation led to overexpression of some of its target genes, suggesting that it is important for gene repression (Creyghton et al. 2008). Interestingly, presence of the very same histone variant H2A.Z in a combination with acetylated forms of histone H4, leads to recruitment of Brd2 and, in turn, to transcriptional activation (Draker et al. 2012). It has been also shown that presence of H2A.Z histone variant is essential for the removal of nucleosomes from the transcriptional start sites (TSSs) of tissue-specific genes during ES cells differentiation (Li et al. 2012) and that it generally facilitates transcription-coupled nucleosome removal from gene TSSs (Schones et al. 2008). These suggest that the presence of the very same histone variant can have different effects on transcription, depending on the chromatin context. Furthermore, combination of different histone variants can have additive effects on chromatin structure. It was shown that nucleosomes assembled with histone variants H3.3 and H2A.Z were more unstable than nucleosomes containing H3.3 and H2A or H3 and H2A.Z alone (Jin & Felsenfeld 2007). Instability of nucleosomes can then positively affect transcription, as discussed above. Besides, not only core histones have their variants. Linker histone H1 also has multiple variants. For example, there are 11 variants of histone H1 in humans and mice (Izzo et al. Chapter 1: Introduction 6 2008), some of which are oocyte/early embryo-specific (H1foo) or testis-specific (H1t, H1t2 or Hils1) (Seyedin & Kistler 1980, Tanaka et al. 2001, Yan et al. 2003, Martianov et al. 2005, Tanaka et al. 2005). Interestingly, it has been shown that early embryonic variant of histone H1 (dBigH1) in Drosophila is crucial for regulation of zygotic gene activation (ZGA): deletion of this histone variant resulted in pre-mature ZGA and embryo death (Perez- Montero et al. 2013). Oocyte and embryo-specific linker histone variant, B4, a homologue of the mammalian H1foo, has been also identified in frogs (Smith et al. 1988, Tanaka et al. 2001). It was shown that the affinity of B4 histone for DNA is 6 times lower than that of a canonical, somatic-type H1. The incorporation of B4 instead of canonical H1 into chromatin facilitated transcription (Ura et al. 1996). Interestingly, the very same histone B4 was shown to be incorporated into chromatin of transplanted nuclei during the nuclear transfer procedure and to be necessary for transcriptional reprogramming (Teranishi et al. 2004, Jullien et al. 2010). It is therefore not surprising that the embryonic-like linker histone variants are lost as embryos develop, probably in order to prevent promiscuous transcription and enable tissue specification (Smith et al. 1988). All these suggest that the mere composition of the nucleosomes can affect the susceptibility of chromatin to transcription. 1.1.4. Nucleosome density The positioning of nucleosomes on DNA and, specifically, the density of nucleosomes at particular genomic sites, is also known to affect the accessibility of the chromatin. In general, the higher the nucleosome density, the more compact the chromatin structure and the less accessible for transcriptional machinery (Kwon et al. 1994, Bai & Morozov 2010). Comparison of genome-wide profiles of nucleosome occupancy in seven different human cell lines revealed the presence of nucleosome-free regions at TSSs of Chapter 1: Introduction 7 transcriptionally active genes (Ozsolak et al. 2007). It was also shown that during ES cell differentiation, the activation of genes correlated with nucleosome depletion around their transcriptional start sites (TSSs) (Li et al. 2012). Furthermore, in yeast nucleosome occupancy is inversely correlated with transcriptional activity not only at gene promoters, but also across the gene body: nucleosomes are depleted at the promoters of highly transcribed genes and are more abundant at promoters and at gene bodies of repressed genes (Lee et al. 2004). Interestingly, it was shown in zebrafish embryos that the canonical organisation of nucleosomes on chromatin is detected only from around the time of zygotic genome activation, suggesting that proper positioning of nucleosomes on chromatin is important for the regulation of gene expression (Zhang et al. 2014). All these results support the model that the mere presence of nucleosome modulates the accessibility of the underlying DNA sequence. 1.1.5. Cooperative action of different epigenetic mechanism to achieve stable maintenance of a differentiated state It was demonstrated that various epigenetic mechanisms mentioned before can act cooperatively to achieve multiple layers of epigenetic stabilisation of chromatin states. For example, it has been shown that cytosine methylation is recognized by proteins that have methylated-cytosine binding domains (MBD): MeCP2, Mbd1, Mbd2, Mbd3 and Mbd4 (Bogdanovic & Veenstra 2009). These proteins then recruit other factors, which can further induce heterochromatinization of the DNA-methylated region. For example, it has been shown that MBD domain-containing proteins can recruit histone deacetylases (HDACs) to chromatin. These are responsible for removing the acetyl groups from histones and therefore lead to chromatin compaction and transcriptional repression (Jones et al. 1998, Wade et al. Chapter 1: Introduction 8 1999, Zhang et al. 1999, Le Guezennec et al. 2006). DNA methylation can be also reciprocally induced by the presence of certain histone marks. For example, HP1 protein that binds to H3K9me2/3 interacts with Dnmt1 and Dnmt3b, which are maintenance and de novo DNA methylases, respectively. This interaction is important for the induction of DNA methylation and gene repression (Lehnertz et al. 2003, Smallwood et al. 2007). Presence of DNA methylation can regulate the distribution of repressive histone marks also by restricting them. It has been shown that DNA methylation is mutually exclusive to H3K27me3 mark (Lindroth et al. 2008). Removal of DNA methylation leads to widespread accumulation of H3K27me3 at illegitimate regions (Hagarman et al. 2013, Reddington et al. 2013). In contrast, the removal of H3K27me3 results only in mild changes in 5meC distribution across the genome (Hagarman et al. 2013). Interestingly, it was shown that unmethylated CpG islands can be recognized and bound by Kdm2b, which is a histone H3 lysine 36 demethylase. Kdm2b then recruits Polycomb repressive complex 1 (PRC1), which deposits H2AK119Ub, leading to gene repression (Farcas et al. 2012, He et al. 2013). Presence of certain histone variants in the nucleosome can also influence the deposition of epigenetic marks. For example, in mouse ES cells, the presence of H3.3 is necessary for proper establishment of H3K27me3 mark at promoters of developmentally-important genes (Banaszynski et al. 2013). Concluding, various epigenetic mechanisms can act independently or synergistically in several layers of regulation to ensure a stable maintenance of a differentiated state (Fig. 1). Chapter 1: Introduction 9 Fig. 1. Epigenetic mechanisms stabilising a differentiated state. Example of a gene which is expressed in an undifferentiated cell and repressed in a differentiated cell. Repression of this gene is achieved and stabilised by several layers of epigenetic mechanisms: DNA becomes methylated, nucleosomes occupancy increases, histone variants associated with active transcription are replaced with canonical histone variants, activating post-translational histone marks are removed and repressive marks established. Chapter 1: Introduction 10 1.2. Reprogramming as the reversal of differentiation Interestingly, epigenetically-stabilised differentiated states can be experimentally reversed back to embryonic-like pluripotent states. It has been shown that if a nucleus of a somatic cell is transplanted to an unfertilised, enucleated egg, this can lead to a development of a new embryo, which can subsequently differentiate into a an adult organism (Gurdon et al. 1958, Gurdon & Uehlinger 1966). It was also proved with the use of nuclear markers that the genetic material in such cloned adult organisms was derived exclusively from the donor cell nucleus and not from the recipient egg (Elsdale et al. 1960, Hochedlinger & Jaenisch 2002). It means that a differentiated cell was able to de-differentiate to a totipotent state. However, it has been also shown that this process of de-differentiation is limited: the more differentiated the donor cell nucleus, the less frequent is the normal development of resulting nuclear transfer embryos. For example, it was shown in frogs that if donor cells for the nuclear transfer were derived from blastula or early gastrula stage embryos, up to 36% of successfully reconstructed embryos could reach a feeding tadpole stage, whereas if intestinal epithelium cell nuclei were used as donors, only 1.5% of embryos reached a feeding tadpole stage (Gurdon 1962). It has been further explained that the decreased susceptibility of differentiated cells to undergo reprogramming is related to the fact that multiple layers of epigenetic stabilisation of the differentiated state are acquired sequentially as the development progresses: the more differentiated the cell, the more epigenetic mechanisms safeguard the differentiated state and the more difficult it is to successfully de-differentiate such cells (Pasque et al. 2010, Pasque et al. 2011b). It was reported in mammalian systems that nuclear transfer-derived embryos exhibit abnormally high DNA methylation levels as compared with control, fertilised embryos. Moreover, these non-reprogrammable methylation abnormalities were memories of the methylation states in the donor cells used for the nuclear transfer: similarly high methylation pattern were also detected in the donor cells (Bourc'his et Chapter 1: Introduction 11 al. 2001, Kang et al. 2001, Chan et al. 2012). This suggests that repressive marks acquired during differentiation were a barrier for reprogramming. It has been also shown that histone variants can be a barrier for reprogramming. Experiments using the transcriptional reprogramming system of nuclear transfer to Xenopus prophase-arrested oocytes (Fig. 2) demonstrated that mouse embryonic fibroblasts (MEFs) reprogram worse than less differentiated epiblast stem cells due to the presence of repressive histone variant macroH2A in MEFs (Pasque et al. 2011a). Post-translational histone marks themselves were also shown to be refractory to correct reprogramming after nuclear transfer and to be responsible for the retention of the donor cell-like characteristics in nuclear transfer embryos. For example, cloned bovine embryos ‘remembered’ the histone H4 lysine 5 acetylation levels from the donor cells, which was also correlated with levels of gene expression in such embryos. Interestingly, this memory of histone acetylation level was retained even when blastomeres derived from the first round of nuclear transfers were used as donors for the second round of nuclear transfers, suggesting persistence of the memory through generations (Wee et al. 2006). The best documented and molecularly characterised phenomenon of the epigenetic memory in nuclear transfer embryos comes from experiments in frogs. It was shown that embryos derived from nuclear transfer with somite cells (muscle precursors) showed expression of MyoD gene (which is a gene normally expressed in somite tissues) in illegitimate regions of the embryo, in which this gene is not normally expressed (Ng & Gurdon 2008). This memory was dependent on the presence of lysine 4 of H3.3 histone variant at MyoD gene promoter (Ng & Gurdon 2008). This suggests that problems with resetting the epigenetic landscape apply not only to genes which are repressed and cannot be properly activated in the embryo, but also to those for which activating epigenetic environment inherited from the donor cell causes illegitimate gene expression in the resulting embryo. Chapter 1: Introduction 12 Concluding, even though the process of cell differentiation can be reversed experimentally, nuclear reprogramming has its limitations and in general is inefficient; especially when highly differentiated cells are used as nuclear donors. Fig. 2. Transcriptional reprogramming system of the nuclear transfer to Xenopus prophase- arrested oocytes. Somatic cells are first permeabilised and subsequently transplanted into the nucleus (germinal vesicle – GV) of a prophase-arrested Xenopus laevis oocyte. After the transplantation, nuclei of somatic cells reactivate gene transcription. This transcription can be measured by RT- qPCR. Chapter 1: Introduction 13 1.3. Sperm is a highly specialised cell 1.3.1. Global changes occurring during spermiogenesis The sperm cell is highly specialised - designed to deliver genetic material to the egg at fertilisation. Spermatogenesis is the process in which a pluripotent germ cell differentiates into a mature spermatozoon. It starts when the germ cell precursors of the sperm, spermatogonia undergo a cell division, which can be either proliferative or differentiative. The first one produces more spermatogonial cells, the second one results in a formation of a spermatogonial cell and a more differentiated precursor – a primary spermatocyte (de Rooij 2001) (Fig.3). Primary spermatocytes reduce their chromosome and DNA content in a meiotic division and eventually, as a result of meiosis, give rise to four haploid spermatids (Roosen-Runge 1969). Spermatids are immediate precursors of sperm, but in order to be transformed into mature spermatozoa, they have to complete a series of substantial structural and morphological changes, called spermiogenesis (Toshimori 2003) (Fig. 3). One of the most dramatic changes occurring during spermiogenesis is a global compaction of the chromatin. The molecular basis behind the compaction of the sperm nucleus is the best described in mammalian systems, so in this section I will review what is known about this process in mammalian systems. It has been calculated that the volume of chromatin in mature sperm in mammals is around six times smaller than the volume of mitotic chromosomes (Ward & Coffey 1991). Such a high compaction of the sperm nucleus is possible due to the presence of protamines. Protamines are small basic proteins, which have higher affinity for DNA than histones, allowing a tight packaging of sperm DNA (Balhorn 2007). They are incorporated into the sperm chromatin in place of core histones in a multi-step process. Initially, the chromatin in the sperm is globally hyperacetylated (Hazzouri Chapter 1: Introduction 14 et al. 2000, Govin et al. 2004), which is thought to play at least two roles in promoting chromatin compaction. First, acetylated lysine residues of histones are recognized and bound by a protein called Brdt (Bromodomain-testis specific), which is suggested to be involved in histone eviction (Pivot-Pajot et al. 2003, Govin et al. 2006, Moriniere et al. 2009, Gaucher et al. 2012). Second, it has been recently shown that acetylation of core histones is also directly involved in their proteasome-mediated degradation during sperm maturation (Qian et al. 2013). Core histones are replaced with testis-specific histone variants, such as hTSH2B, H2AL2 or H3t, which are thought to further increase the instability of the nucleosome structure (Li et al. 2005, Syed et al. 2009, Tachiwana et al. 2010). Subsequently, histones are replaced with transition proteins (Yu et al. 2000, Meistrich et al. 2003, Zhao et al. 2004), which are ultimately exchanged for protamines in the final stages of sperm maturation (Chen et al. 2001b, Cho et al. 2003, Balhorn 2007, Ravel et al. 2007, Steger et al. 2008). These nuclear changes are accompanied by cessation of transcription and disappearance of the basal transcriptional machinery – as opposed to spermatids, mature spermatozoa are transcriptionally silent (Monesi et al. 1978, Martianov et al. 2001, Zheng et al. 2008) (Fig.4). Chapter 1: Introduction 15 Fig. 3. Spermatogenesis. The diagram depicts sequential stages of spermatogenesis. Spermatogenesis consists of meiotic maturation, in which the chromosome and DNA content is reduced and of spermiogenesis, in which immature spermatid maturates into spermatozoon. Chapter 1: Introduction 16 Fig. 4. Major nuclear changes during spermiogenesis (as described in mammalian systems) - adapted from (Teperek & Miyamoto 2013). The round spermatid undergoes a series of chromatin remodelling events that lead to the formation of a fully mature spermatozoon. First, canonical core histones packing the chromatin in the round spermatid are replaced by histone variants, which, together with global histone acetylation, leads to instability of the nucleosome structure. Subsequently, transition proteins are incorporated in place of unstable nucleosomes in the elongating spermatid. Finally, transition proteins are replaced with protamines. The mature sperm chromatin is mainly composed of protamines, with interspersed histones and with tightly associated mRNAs and transcription factors (see chapter 1.4.2). All these processes are occurring in parallel with the cessation of transcription – round spermatids are transcriptionally active, whereas no transcription is detected in the mature sperm. Chapter 1: Introduction 17 1.3.2. Epigenetic marks are retained in the chromatin of mature sperm Interestingly, it has been shown that not all the regions in the sperm chromatin are subjected to histone to protamine replacement. Regions which retain histones in the chromatin of mature sperm in human were shown to be endonuclease-sensitive and to be enriched for gene promoters and other regulatory sequences of the genome, such as enhancers (Arpanahi et al. 2009). It was also shown that among the genes that retain histones at their promoters in mature sperm in mouse and human, there is a significant overrepresentation of genes that are developmentally-important (Hammoud et al. 2009, Brykczynska et al. 2010). Furthermore, histones retained in sperm turned out to be post-translationally modified at specific regions of the chromatin in mouse and human. Some of the histone modifications, like acetylation of H4K12, (Paradowska et al. 2012), histone crotonylation (Tan et al. 2011) or H3K4me2 and H3K4me3 (Hammoud et al. 2009, Brykczynska et al. 2010) localise to genes which are highly active during spermiogenesis, suggesting that they are important for the regulation of expression of spermiogenesis-related genes. Therefore, the presence of these marks at the promoters of these genes in the chromatin of mature sperm likely reflects their expression at earlier stages. Strikingly however, other marks like H3K27me3, but also H3K4me2 and H3K4me3 (and a combination of H3K4me2/3 with H3K27me3), also marked promoters of genes that are developmentally-important, for example Hox genes, that are involved in embryo patterning (Hammoud et al. 2009, Brykczynska et al. 2010). Furthermore, comparison of sperm epigenetic landscapes with embryonic transcriptome revealed the following correlation: genes that have activating epigenetic marks in sperm tend to be expressed early in development, whereas those with repressive marks are activated late or silenced in early embryonic stages (Hammoud et al. 2009, Brykczynska et al. 2010). Surprisingly, similar observations were also made in zebrafish, which does not have Chapter 1: Introduction 18 protamines packing the DNA in the mature sperm. In zebrafish sperm, developmentally- important genes were enriched for histone variant H2AFV, for H3K4me2/3, H3K27me3 and for H3K36me3 at their promoter sequences (Wu et al. 2011). Genes bearing H3K4me3 together with H3K14 acetylation (H3K14ac) in sperm were enriched at promoters of genes related to spermiogenesis and also for housekeeping genes (Wu et al. 2011). Analogically to what was observed in mammals, genes with activating marks at their promoters in sperm (H3K4me2/3 and H3K14ac) tend to be expressed early in embryonic development, whereas genes with repressive mark H3K27me3 or enriched for histone variant H2AFV, tend to be repressed at the earliest developmental stages in the embryo (Wu et al. 2011). Interestingly, a mass-spectrometry based study of post-translational histone and protamine marks in mouse sperm, revealed that not only histones, but also the protamines, are subjected to numerous post-translational modifications (Brunner et al. 2014). Similarly to histone marks, DNA methylation patterns also show uneven distribution in the genome of mature sperm. Overall, cytosine methylation patterns in gene promoters of sperm are very similar to those of embryonic stem cells (Farthing et al. 2008). Interestingly, it was shown that developmentally-important genes were hypomethylated in sperm in mammals, as well as in zebrafish (Hammoud et al. 2009, Brykczynska et al. 2010, Wu et al. 2011). Importantly, experiments in zebrafish demonstrated that DNA methylation pattern of sperm, but not of oocytes, is recapitulated in the embryo at the time of zygotic genome activation. The oocyte methylation pattern had to be adjusted to match the sperm methylation pattern by a passive loss of DNA methylation (at loci methylated in an oocyte-specific manner) or by undergoing de novo methylation (for these loci, that were unmethylated in oocytes, but methylated in sperm) (Jiang et al. 2013, Potok et al. 2013). Experiments performed using in vitro egg extracts in Xenopus laevis showed that maintenance of the sperm-derived DNA methylation after replication is dependent on the recruitment of the Chapter 1: Introduction 19 maintenance DNA methylase via UHRF1-mediated ubiquitination of lysine 23 on histone H3 (Nishiyama et al. 2013). 1.3.3. Chromatin composition of the mature sperm in Xenopus laevis Global condensation of the mature sperm nucleus is also observed in Xenopus laevis. Six protamine-like sperm basic proteins (Sps) have been identified in Xenopus: Sp1-6 (Abe & Hiyoshi 1991, Hiyoshi et al. 1991, Yokota 1991, Frehlick et al. 2007). Furthermore, it has been shown that mature Xenopus sperm cells are devoid of somatic type of linker histone H1, and instead have linker histone variants composing their chromatin, like H1fx and H1.sp (Abe & Hiyoshi 1991, Shechter et al. 2009). Interestingly, it has been also shown that the levels of somatic histone variants H2A and H2B, but not H3 and H4, are significantly reduced in Xenopus sperm (Abe & Hiyoshi 1991, Yokota 1991, Shechter et al. 2009). Furthermore, mass spectrometry and immunoblotting analyses in Xenopus laevis sperm identified the presence of various post-translational histone marks on the sperm chromatin (Nicklay et al. 2009, Shechter et al. 2009) (Table 1). However, nothing is known about the positioning of these marks on the sperm chromatin or about the function (if any) of these marks in the future embryo development. Table1. Histone variants and post-translational histone marks identified in Xenopus laevis sperm * Histone variants H1fx, H1.sp, H2A.Z, Modifications on H3 Methylation: K4me1/2, K9me1/3, K27me1/2/3, K36me1/2, K79me1/2, R2me2a, R17me2a Acetylation: K18ac1, K23ac2, Modifications on H4 Acetylation: K5ac2/3, K8ac1/2/3, K12ac1/2/3, K16ac1/2/3 Phosphorylation: S1ph Methylation: K20me1/2/3, K79me2, R3me1, R3me2s * - information from (Nicklay et al. 2009, Shechter et al. 2009). Chapter 1: Introduction 20 1.4. Why is the sperm better than a somatic cell at supporting embryonic development? As mentioned before, the differentiation state of a somatic cell can be reversed back to pluripotency by reprogramming, for example by transferring the nucleus of such a differentiated cell to an unfertilised, enucleated egg, which in some cases can lead to a full- term embryonic development. However, it was also mentioned that the nuclear transfer procedure is generally highly inefficient. This inefficiency remains in contrast to the high rates of successful embryonic development when the sperm fertilises the egg. There are definitely some similarities between events occurring after somatic cell nuclear transfer and after fertilisation. First, upon introduction into the egg cytoplasm, both the chromatin of the sperm and the chromatin of the somatic cell nucleus decondense and increase in volume (Gurdon 1976, Lassalle & Testart 1991, McLay & Clarke 2003). Global chromatin changes accompany the decondensation process. During fertilisation, sperm-specific protamines are replaced with histones deposited in the egg (McLay & Clarke 2003). In nuclear transfer, the oocyte-specific histone variants are also incorporated into the chromatin of somatic cells (Jullien et al. 2010, Jullien et al. 2012). Interestingly, it was shown that in mice the same egg- specific linker histone variant – H1foo is incorporated into sperm chromatin upon fertilisation and into the nucleus of a somatic cell after somatic cell nuclear transfer (Gao et al. 2004, Becker et al. 2005). Moreover, it was shown that in Xenopus both processes are dependent on a protein called Nucleoplasmin, which is maternally deposited in the egg (Philpott et al. 1991, Philpott & Leno 1992, Gillespie & Blow 2000, Tamada et al. 2006, Ramos et al. 2010, Inoue et al. 2011, Okuwaki et al. 2012). After reprogramming, the DNA of the sperm or of the somatic cell can be replicated and used as a template for RNA synthesis (Aoki et al. 1997, Bouniol-Baly et al. 1997). These observations would suggest that the embryos generated by both ways undergo similar changes leading to the same outcome – embryonic development. Chapter 1: Introduction 21 However, embryos develop much better to the adulthood after fertilisation than after a somatic cells nuclear transfer. Does it mean that the sperm is somehow specialised to support embryonic development? There are numerous events occurring during spermiogenesis that transform the immature germ cell precursor into a mature sperm that is ready to fertilise the egg. It is possible that these events are important to somehow program the sperm to support embryonic development. Interestingly, ICSI (Intra Cytoplasmic Sperm Injection) experiments in mice showed that a spermatid (Fig. 3), when injected into an unfertilized egg, is a much less efficient donor cell at supporting normal embryonic development than a mature sperm cell (Kishigami et al. 2004). Moreover, experiments in mice also proved that when injected into eggs, spermatids that are more advanced in spermiogenesis are more efficient in producing normal embryos than the less advanced ones (Ohta et al. 2009). Both results could suggest that acquisition of specific sets of proteins and/or depletion of others may be responsible for the high efficiency of reprogramming and embryo generation upon fertilisation. Despite the fact that no factors directly responsible for differences in the efficiency of distinct sperm cell progenitors have been identified so far, there are some, which are more likely than the others to be involved in the sperm programming. Those include protamines, sperm-derived transcriptional regulators and sperm epigenetic marking. Each one of those is briefly discussed below in the context of its potential ability to confer to the sperm developmental advantage when compared with a somatic cell. Chapter 1: Introduction 22 1.4.1. Protamines A striking difference between the sperm and the somatic cell is the packaging of their DNA. DNA in sperm is wrapped around protamines (at least in the majority of the vertebrate species), whereas histones are the core unit organising chromatin in somatic cells. Having the protamines instead of canonical histones as the main components of chromatin can provide the sperm with at least two developmental advantages. Firstly, upon fertilisation, protamines from the sperm are removed and subsequently, maternally-derived histones are assembled onto the paternal DNA to allow its chromatinisation (McLay & Clarke 2003, van der Heijden et al. 2005). The chromatin of a somatic cell is packed with histones instead of protamines, and therefore there may be no need to replace the somatic cell histones with the oocyte-stored histones. As mentioned before, somatic cells are differentiated and during the course of specialisation they acquire many epigenetic mechanisms stabilising the expression states of their own differentiation-specific genes, for example in the form of post-translational histone marks. If such epigenetic marks are not correctly erased at fertilisation (as could be the case if the oocyte would not remove the somatic cell-specific histones), the epigenetic memory of a donor cell can be carried over to the embryos, as demonstrated in the examples above (chapter 1.2). In support of this hypothesis it was shown that embryos generated by the injection of round spermatids, which do not have protamines on their DNA, had elevated levels of DNA methylation and of H3K9me3, as compared to embryos generated by sperm injection (Kishigami et al. 2006). Secondly, protamines on sperm DNA ensure the tight packaging of its chromatin. This can protect the DNA from any physical damage. Interestingly, in rabbits no offspring could be derived after the round spermatid injection and the developmental arrest of the reconstructed embryos was correlated with their abnormal ploidy (Ogonuki et al. 2005). The Chapter 1: Introduction 23 nuclear transfer procedure involves numerous micromanipulations to the somatic cell nucleus. Since somatic cells do not have protamines, such procedures could lead to DNA damage. Indeed, it was suggested that a major cause of developmental failure of nuclear transfer embryos was DNA loss, which could be a result of DNA damage induced during the nuclear transfer procedure (Mizutani et al. 2012). On the other hand, experiments in mouse in which spermatids at different stages of spermiogenesis were used as the paternal DNA donors demonstrated that the highest difference in the ability of spermatids to support embryonic development was acquired during spermatid maturation from step 7-8 to step 9-10. Interestingly, at this stages protamines are not yet present in the spermatid nuclei, which argues against necessary roles of protamines in supporting embryonic development (Ohta et al. 2009). 1.4.2. Sperm-derived transcriptional regulators and RNAs Interestingly, even though mature sperm is transcriptionally silent and devoid of basal components of transcriptional machinery, some transcription factors, like Oct-1, Ets-1, C/EBP and TBP are retained in the chromatin of the mature sperm, and are associated with the hypersensitive regions of the sperm chromatin (Fig. 4) (Pittoggi et al. 2001, Zheng et al. 2008). Proteomic analysis of the mature sperm led to identification of several proteins involved in transcriptional regulation, for example, Bromodomain-containing protein 7 (Brd7) or Polycomb protein Suz12 (de Mateo et al. 2011). If such factors were delivered to the oocytes at fertilisation, they could be involved in the regulation of embryonic gene expression. On the other hand, a somatic cell has its own transcriptional regulators important for the maintenance of its differentiated state. When the nucleus of such a cell is transplanted Chapter 1: Introduction 24 to the oocyte, some of these differentiation-specific factors characteristic for a somatic cell transcriptional program would be delivered to the cytoplasm of the oocyte. This could be detrimental for the embryonic development, since it could interfere with the proper regulation of the embryonic gene expression. Indeed, it was shown that if fertilised embryos were injected with a cytoplasm derived from somatic cells, their development was impaired and such embryos had a decreased expression of pluripotency gene Oct4 (Van Thuan et al. 2006), suggesting that some components of a somatic cell cytoplasm may be indeed toxic for development. It was also hypothesised that sperm-derived RNAs may be needed for the future embryonic development. Despite the fact that mature sperm is transcriptionally silent and that it has almost no cytoplasm, it has been estimated that it contains about 10-100fg of RNA compared to 10-50pg of RNA typically found in a somatic cell (Pessot et al. 1989, Krawetz 2005), some of which were shown to be delivered to the oocyte at fertilisation (Ostermeier et al. 2004). Of the RNAs delivered by the sperm to the oocyte at fertilisation, microRNA miR- 34c was shown to be required for the first embryonic cleavage division in mouse (Liu et al. 2012). This led to the hypothesis that sperm-derived RNAs may be important for embryo development and that lack of such RNAs could explain the low developmental potential of somatic cells as donors in the nuclear transfer procedures (Miller 2007). However, the results of recent genome-wide analyses of sperm RNAs argue against this hypothesis. First, the great majority of the RNA contained in the sperm nucleus turned out to be fragmented to a variable extent (Sendler et al. 2013). Second, the comparison of RNA-seq profiles obtained from the purified sperm, from the whole testis and from various human tissues, revealed that there are only 102 transcripts (out of 37974 transcripts tested) that were expressed exclusively in the sperm and testis (or detected in other tissues at very low levels). All these transcripts were present throughout spermatogenesis and retained until the final stages of sperm maturation, Chapter 1: Introduction 25 so no sperm-specific transcripts were detected (Sendler et al. 2013). This is not surprising taken into account the fact that sperm is transcriptionally silent, and thus all transcripts present in mature sperm must be a result of transcription occurring at earlier stages of spermiogenesis. Therefore, the presence of transcripts specific for testicular cells could perhaps explain the developmental advantage of sperm as compared to somatic cells; however, it cannot explain the fact that the mature sperm is better than a spermatid in supporting embryonic development. 1.4.3. Sperm-derived epigenetic marks As mentioned before, it has been demonstrated in several species that mature sperm may also deliver to the oocyte information about the regulation of embryonic gene expression in the form of epigenetic marks on its chromatin. Promoters of developmentally-important genes in the sperm were shown to bear various post-translational epigenetic marks, which, if delivered to the oocyte at fertilisation, could instruct the embryo about the regulation of the future gene expression (Hammoud et al. 2009). Furthermore, it was also shown that the pattern of DNA methylation of the zebrafish sperm is retained in the embryo at the time of zygotic gene activation (Potok et al. 2013). Such epigenetic marking could therefore provide the sperm with a developmental advantage over a somatic cell nucleus. As discussed before, the chromatin of a somatic cell acquires various epigenetic marks in order to stabilise its gene expression. These multiple layers of epigenetic marks could be refractory to reprogramming after nuclear transfer and would result in the retention of the epigenetic memory of the previous state and interfere with embryonic development (Pasque et al. 2011b, Chan et al. 2012). Furthermore, in support of the hypothesis for the importance of sperm-derived epigenetic marks for successful embryonic development, it was shown that abnormal histone Chapter 1: Introduction 26 retention profiles or incorrect DNA methylation patterns in sperm correlated with idiopathic infertility cases in humans (Hammoud et al. 2010, Hammoud et al. 2011). However, due to the fact that the highly condensed sperm nucleus is not an easy material for experimental manipulations, so far there were no experiments proving this hypothesis. 1.5. Evidence for the sperm developmental advantages After fertilisation, the embryo has to successfully accomplish several major developmental processes. First, the DNA from both parents needs to be faithfully replicated. Second, the embryo has to undergo successful zygotic genome activation (ZGA) and initiate the synthesis of both zygotic rRNA and mRNA. All these events need to be tightly regulated, as otherwise they may create a barrier for successful embryonic development (Newport & Kirschner 1982). Interestingly, it was suggested that the sperm may be advantageous as compared with a somatic cell in undergoing these processes (see below). 1.5.1. DNA replication It was shown that a poor survival of nuclear transfer-derived embryos is correlated with chromosome loss during early embryonic division stages (Mizutani et al. 2012). It has been suggested that chromosome loss during cell division may result from incomplete DNA replication at the time of division (Laskey 2005). The use of egg extracts of Xenopus laevis, which are able to replicate DNA in vitro, demonstrated that the sperm, as opposed to a differentiated somatic cell (erythrocyte), can replicate DNA efficiently. DNA replication initiates from origins of replication. It was shown that during sperm replication, the median inter-origin distances were around 23.4kbp, whereas the median inter-origin distances for Chapter 1: Introduction 27 replicating erythrocytes were around 120.9kbp (Lemaitre et al. 2005). Since, at least in early Xenopus development, early cell cycle phases are very rapid, and the first cell cycle last for only around 120mins (at 18°C), too sparsely positioned origins of replication in somatic cells could prevent the replication from being timely completed before the onset of the first cell division. Incompletely replicated chromosomes could then be either stretched and broken at the division or pushed towards one of the two resulting blastomeres, resulting in DNA loss (Gurdon & Laskey 1970). Interestingly, it was shown that pre-treatment of mouse embryonic fibroblasts (MEF) nuclei with Xenopus egg extracts before the nuclear transfer procedure increased the efficiency of replication (making it similar to what is observed for sperm nuclei) (Ganier et al. 2011). This pre-treatment of MEFs with Xenopus egg extracts also increased the frequency of normal nuclear transfer embryo development in mouse (Ganier et al. 2011). Therefore, the ability of undergoing efficient DNA replication was suggested to be one of the developmental advantages of sperm. Conversely, inefficient replication was suggested to be the cause of frequent developmental failures of embryos derived from a somatic cell nuclear transfer (Laskey 2005). 1.5.2. rRNA synthesis Another challenge that the developing embryo has to face is a timely activation of rRNA synthesis. rRNA is a necessary component of ribosomes, which mediate protein translation. Series of experiments in mouse demonstrated that sperm-derived embryos, are significantly better at timely activating zygotic rRNA transcription than nuclear transfer- derived embryos. First, reverse transcription, quantitative PCR (RT-qPCR) measurements of ribosomal RNA content revealed that nuclear transfer embryos (NT-embryos) have significantly lower levels of 18S, 28S and 5.8S rRNA than control sperm-derived embryos Chapter 1: Introduction 28 (Suzuki et al. 2007). Second, inefficient rRNA activation, was also evidenced by a smaller amount of labelled RNA precursor incorporated into rDNA and by a significant delay in the appearance of markers of active nucleolar organising regions in NT-embryos, as compared to sperm-derived embryos (Bui et al. 2011). Finally, inefficient activation of rRNA synthesis was correlated with poor developmental outcomes of NT-embryos. First, NT-embryos developed worse than sperm-derived embryos, but also, NT-embryos which were more efficient at rRNA synthesis, developed better to adulthood than embryos less efficient at rRNA synthesis (Zheng et al. 2012). Interestingly, the ability to activate rRNA synthesis efficiently in NT-embryos was correlated with the levels of rDNA activity in the donor cells: the more active the donor cells at synthesising rRNA, the more efficient the resulting NT- embryos at activating rRNA synthesis and the better the embryonic development. Furthermore, this phenomenon was also related to the DNA methylation levels of the rDNA loci in the donor cells, which were shown to be the most methylated in cells which were the less active at rRNA synthesis, and the less methylated in cells which were the most active in the rRNA transcription (Zheng et al. 2012). This is an important finding, as it again points towards the epigenetic memory of a previous state being a barrier for a successful reprogramming, and suggests that the inability of NT-embryos to correctly activate rRNA transcription may be the cause of their developmental failures. 1.5.3. mRNA synthesis Finally, embryos need to initiate the synthesis of zygotic mRNAs in order to succeed in development (Newport & Kirschner 1982). There are reports indicating that NT-derived embryos do not properly initiate embryonic gene expression, as compared to sperm derived- embryos. It was shown in mouse that NT-embryos often fail to induce correct expression of a Chapter 1: Introduction 29 key pluripotency factor Oct4 and even if they succeed to induce its expression, they often do so abnormally and express it in ectopic tissues, in which Oct4 is not expressed in control sperm-derived embryos (Boiani et al. 2002). Again, correct activation of Oct4 correlated with successful development of NT-embryos (Bortvin et al. 2003). Microarray studies revealed that abnormal expression levels of 1633 genes was already detected at 2-cell stage in NT- embryos in mouse (as compared with fertilised embryos). Furthermore, the largest group within these early abnormally expressed transcripts were transcription factors (Vassena et al. 2007). Another microarray study in mouse in which 87 single blastocyst stage embryos derived from nuclear transfers with different donor cells were compared with control fertilised embryos revealed that gene expression profiles differed between nuclear transfer embryos and fertilised embryos and also that nuclear transfer embryos coming from the same donor cell type, were more similar to each other, than to the embryos coming from nuclear transfer of different donor cell types (Fukuda et al. 2010). The same conclusion has been drawn in an independent study, also using mouse nuclear transfer-derived embryos from different types of donor cells (Hirasawa et al. 2013). In the latter paper it was also showed that the donor-cell dependent aberrant gene expression was more pronounced in embryonic than in extraembryonic tissues (Hirasawa et al. 2013). All these results point towards the same conclusion, that the NT-derived embryos experience problems with initiation of embryonic gene expression and that the abnormalities in gene expression profiles are often correlated with the origin of the donor cell, suggesting persistent epigenetic memory in NT embryos. Concluding, the sperm, as opposed to a somatic cell, may be programmed for several aspects of embryo development: DNA replication, initiation of rRNA synthesis and zygotic mRNA transcription. Chapter 1: Introduction 30 1.6. Work described in this thesis The experiments described in this thesis aim to answer the question of what is the nature of developmental programming of sperm. I started from establishing a model system in which the developmental potential of sperm can be compared to the developmental potential of another cell type. I concluded that a suitable cell to compare with sperm is a spermatid. A spermatid is an immediate precursor of sperm, has already completed meiotic division, and differs from sperm only in the degree of specialisation. Furthermore, a spermatid can be compared to sperm in the same assay. I then showed that sperm is better at supporting embryonic development than spermatids, suggesting that sperm is programmed to support proper embryonic development. Subsequently, I aimed to identify the molecular basis for the sperm programming. I started from the identification of proteins present in sperm and spermatids by mass spectrometry analysis, reasoning that sperm-specific factors could be responsible for the developmental advantage of sperm. I also used the mass spectrometry- based approach to identify egg proteins which bind specifically to sperm or spermatid chromatin, since sperm programming could be also reflected in the ability to interact with the egg factors. Mass spectrometry analysis allowed me to identify several interesting candidate proteins, which I then functionally tested for their roles in the programming of the paternal nucleus. Unfortunately, none of these factors turned out to be beneficial for embryonic development. I have then changed my strategy for investigating the nature of sperm programming by characterising the developmental defects of spermatid-derived embryos. I showed that these defects are unlikely to be explained by inefficient DNA replication or by their inability to initiate zygotic rRNA transcription. Interestingly, RNA-seq analyses allowed me to identify 100 developmentally-important mRNAs which were misexpressed in spermatid-derived embryos as compared with sperm-derived embryos. Interestingly, the majority of them (82/100) were upregulated in spermatid-derived embryos. Further analysis Chapter 1: Introduction 31 revealed that orthologues of these genes are modified by a repressive H3K27me3 mark in human sperm. These results encouraged me to test the hypothesis that perhaps the sperm is epigenetically programmed to regulate embryonic gene expression after fertilisation. Indeed, the enzymatic removal of H3K27me3 marks from the parental chromatin in embryos at the time of fertilisation resulted in gene misexpression. This led to the hypothesis that perhaps spermatids lack the repressive H3K27me3, as compared with sperm, which causes upregulation of genes in spermatid-derived embryos. Surprisingly, ChIP-seq analyses on the chromatin isolated from sperm and spermatids revealed that spermatids, similarly to sperm, already have repressive H3K27me3 marks at misregulated genes. Therefore, H3K27me3 itself could not explain the difference in gene expression between sperm- and spermatid- derived embryos. Interestingly, further ChIP-seq analyses for activating histone marks revealed that spermatids have more of H3K4me2/3 marks than sperm. Interestingly, the presence of these activating marks at misregulated genes in spermatids correlated with their upregulation in spermatid-derived embryos. All these results support the hypothesis that the sperm, as opposed to the spermatid, is epigenetically programmed to regulate embryonic gene expression after fertilisation. The results presented in this thesis are divided into 5 separate chapters (Chapters 3-7). Each of these chapters has a brief introduction that explains the rationale for the experiments conducted, followed by a detailed report of the results and a subsequent short summary/discussion. Since all the chapters have their own short summary/discussion sections, when writing the final discussion of the thesis (Chapter 8) I took the advantage of the writer’s freedom and I do not discuss the results one by one in the order of appearance in the thesis. Instead, I took the pleasure of talking about the most exciting results; therefore I sometimes shuffle the results and discuss them according to the common connecting theme. In the discussion section I also propose new experiments, which are based on the results Chapter 1: Introduction 32 described, that are either currently conducted in the laboratory or are in the future plans. Finally, I put the obtained results in the broader context and I explain how they advance the current knowledge of the subject. In the appendices section of the thesis I attach publications that arose as a result of this work. There are two review articles, one book chapter and a research article by a colleague from my laboratory, Dr Kei Miyamoto, with whom I collaborated on some parts of his project (which is not connected to the work described in this thesis and therefore is not separately described here). The manuscript describing the work presented in this thesis is currently in preparation. Supplementary tables S1-S6 are Excel files with long protein/gene lists, and they are therefore provided electronically as separate attachments to this thesis. Chapter 2: Experimental procedures 33 Chapter 2 Experimental procedures All the experiments involving the use of animals comply with the Home Office regulations as set out in the Animals (Scientific Procedures) act 1986 (Establishment Code 8002802). All the animal care (tadpole and frog husbandry, hormonal injections (PMSG/hCG), sacrifice and dissections) was done by David Simpson at the Frog Facility at the Gurdon Institute. All the bioinformatic analyses described are written and kindly shared by Dr Angela Simeone, Dr George Allen and Dr Charles Bradshaw. Experimental procedures are described in order of their appearance in the ‘Results’ section of the thesis (Chapters 3-7), apart from all the bioinformatic analyses, which are described altogether as the last section. 2.1. Separation of sperm and spermatids Testes from 6 adult Xenopus laevis males were isolated and manually cleaned from blood vessels and fat in 1XMMR (100mM NaCl, 2mM KCl, 2mM CaCl2, 1mM MgCl2, 5mM HEPES pH 7.4) using forceps and paper tissues. It is crucial to clean the testes well from any non-testicular tissues, as otherwise the cells released from the tissues may negatively affect the final purity of the isolated cells. Subsequently, testes were torn into small pieces with Chapter 2: Experimental procedures 34 forceps and homogenised with 2-3 strokes of a Dounce homogenizer (tissue from 1 testis at a time). The cell suspension was then filtered (50um pore size, CellTrics, cat. 04-0042-2317) to remove tissue debris and cell clumps and spun down at 800rcf, 4°C, 20 minutes. Supernatant was discarded and the cell pellet was resuspended in 12ml of 1XMMR. If any red blood cells were visible at the bottom of the pellet (a result of incomplete removal of blood vessels), only the uncontaminated part of the pellet was recovered, taking extreme care not to disturb the red blood cells. Subsequently, six (1 per testes from each frog) step gradients of iodixanol (Optiprep – Sigma, D1556, is 60% iodixanol in water) in 1XMMR were manually prepared in pre-chilled 14ml ultra-clear centrifuge tubes (Beckman Coulter, 344060) in the following order from the bottom to the top of the tube: 4ml of 30% iodixanol, 1ml of 20% iodixanol, 5ml of 12% iodixanol and 2ml of cell suspension in 1XMMR. Gradients were spun down in pre-chilled SW40Ti rotor at 7500rpm (10000g), 4°C, 15 minutes, deceleration without break (Beckman Coulter Ultra-centrifuge, Optima L-100XP). The top interface fraction (between 1XMMR and 12% iodixanol), containing spermatids, and the pelleted fraction, containing mature sperm, were collected. Collected fractions were diluted six times with 1XMMR and collected by spinning first at 805 rcf, 4°C, 20 minutes and re-spinning at 3220 rcf, 4°C, 20 minutes to pellet the remaining cells. Pelleted cells were subjected to nuclei preparation (see below). 2.2. FACS analysis All the cell cycle analyses were performed by Dr Rachel Walker at the Flow Cytometry facility of Cambridge Stem Cell Institute, Cambridge, UK (samples were provided to Dr Walker after fixation and labelling). Chapter 2: Experimental procedures 35 Cells were fixed in cold 70% EtOH for 30mins at 4°C, washed twice in 1XMMR, spun at 850g and resuspended in 50ul of 1XMMR supplemented with 100ug/ml of DNase- free RNase. DNA was stained by the addition of 200ul of 50ug/ml propidium iodide solution. 2.3. Sperm and spermatid nuclei preparation, intra-cytoplasmic sperm injections (ICSI) to non-enucleated and to enucleated eggs, and embryo culture Sperm and spermatids nuclei have been permeabilised with digitonin (Smith et al. 2006) and stored at -80°C. Injections were performed using Drummond Nanoject microinjector (NanojectII Auto Nanolitre Injector, Biohit, 3-00-206A) and glass caplillaries (Biohit, 3-00-203-G/XL) pulled using a Flaming-Brown micropipette puller (settings: heat 700, pull 100, velocity 100, time 10). Cell suspension in sperm dilution buffer (SDB) (Smith et al. 2006) was sucked into the injection needle filled with mineral oil. Cell concentration was adjusted by doing mock injections on a microscope slide to deliver 1 cell per 4.6nl injection. The eggs were placed in batches of 20-25 on a blotting paper. If they were to be enucleated, they were placed with animal pole facing upwards, whereas if they were not subjected to enucleation, they were placed on a side (with the marginal zone upwards). For enucleation, eggs were treated for 30s with a UV mineralite lamp (Gurdon 1960) (this step was omitted for non-enucleated eggs). The jelly of the egg was removed by a minimal length of the Hanovia lamp treatment that allows for the needle penetration. The eggs were immediately injected with sperm or spermatid solution and moved to 1XMBS (Gurdon 1976) supplemented with 0.2% bovine serum albumin (BSA). Cell suspension in the needle was replaced every 20-25 eggs injected. At 4-cell stage embryos were sorted (all the non-cleaved embryos or those with irregular cleavage furrows were discarded) and the culture media was Chapter 2: Experimental procedures 36 replaced with 0.1MBS, 0.2% BSA. Embryos were cultured in 0.1MBS, 0.2% BSA (changed daily) at 16-18°C incubator. Assessment of developmental stages was performed according to Nieuwkoop and Faber (Nieuwkoop & Faber 1994). 2.4. Nuclear transfer Nuclear transfers were performed as described (Elsdale et al. 1960, Gurdon 1960), with a modification: donor cells were dissociated in 2mM EDTA. Donor cells were derived from animal caps of early gastrula (stage 10.5) embryos. Eggs were enucleated by a 30s treatment of Mineralite UV lamp and dejellied by a treatment with a Hanovia UV lamp (minimal length required to allow for the needle penetration). 2.5. Protein preparation for the mass spectrometry analysis All the methods were obtained from the Cambridge Centre for Proteomics (CCP), at the Biochemistry Department, University of Cambridge. I have performed all the steps of protein isolation, quantification and labelling under the supervision of Renata Feret from the CCP. All the subsequent steps of the analysis were carried by Renata Feret and colleagues at the CCP. Proteins from pelleted sperm and spermatids were extracted in Urea/Thiourea buffer (4% Chaps, 2M Thiourea, 6M Urea; Sigma: C9426, T7875, respectively and Fisher Chemicals, U/0500/53) supplemented with protease inhibitors (Roche, 11873580001) by Chapter 2: Experimental procedures 37 keeping on ice for 1h and vortexing from time to time until the pellet disappeared. Subsequently, the solution was sonicated with a probe sonicator (3s, Amplitude 40, sonication over an ice bath) in four separate rounds. The protein concentration was quantified with a Quick Start Bradford Protein Assay (Bio-Rad, 500-0202), following manufacturer’s recommendations. Fifty micrograms of protein lysate was subsequently labelled with Cy3 or Cy5 dye (Cy dyes label the lysine residues of proteins) by adding 1ul of 0.2mM dye. Labelling reaction was carried at 4ºC for 30mins in the dark and quenched by the addition of 1ul of 1mM lysine. Quenching was performed for 10mins at 4ºC in the dark. After the labelling was completed, labelled protein lysates were mixed. Reciprocal labelling was performed to rule out any abnormalities or biases in labelling (none were noticed). Further processing of the samples (1 st and 2 nd dimension electrophoresis, scanning, staining, spot excision, protein digestion and the mass spectrometry analysis) were performed by the staff at the CCP according to the methods used at the CCP. 2.6. Interphase egg extract preparation Eggs were collected in 1XMMR, dejellied with 0.2XMBS, 2% cysteine (pH 7.8-7.9) (Sigma, W326305) and washed with 0.2XMMR. Subsequently, eggs were activated for 3 minutes at RT with 0.2XMMR supplemented with 0.2ug/ml calcium ionophore (Sigma, C7522). Eggs were rinsed with 0.2XMMR and subsequently all abnormal or not activated eggs were removed. Eggs were washed with 50ml of ice-cold extraction buffer (EB) (5mM KCl, 0.5mM MgCl2, 0.2mM DTT, 5mM Hepes pH 7.5) supplemented with protease inhibtiors (PI) (Roche, 11873580001), transferred into centrifugation tube (Thinwall, Ultra- Clear™, 5mL, 13 x 51 mm tubes, Beckman, 344057) and supplemented with 1ml of EB buffer with PI and 100ug/ml of cytochalasin B (Sigma, C2743) and placed on ice for 10 Chapter 2: Experimental procedures 38 minutes. Subsequently, eggs were spun briefly at 350g for 1 minute at 4°C (SW55Ti rotor, Beckman Coulter Ultra-centrifuge, Optima L-100XP) and excess buffer was discarded. Eggs were then spun at 18000g for 10 minutes at 1°C, the extract was collected with a needle, transferred to a fresh, pre-chilled tube, supplemented with PI and 10ug/ml of cytochalasin B and re-spun using the same conditions. Extract was collected with a needle and used: frozen (on liquid nitrogen, in 100ul aliquots) for the analysis of proteins bound to the chromatin followed by the proteomic analysis; or fresh, for replication assays (see below). 2.7. Egg extract treatment and protein isolation for mass spectrometry analysis of egg proteins incorporated into sperm or spermatid chromatin. Sperm and spermatids were collected and permeabilised as described above. Egg extracts were prepared as described above. Twenty millions of permeabilised sperm or spermatids were treated with 3650ul of egg extract, supplemented with the 1X final energy regeneration mix (20X energy regeneration mix is prepared and stored in aliquots at -80ºC: 2mg/ml Creatine Kinase (Roche, 10127566001), 150mM Creatine Phosphate (Roche, 10621714001), 20mM ATP (Roche, 10519979001), 2mM EGTA, 20mM MgCl2). Control permeabilised cells were treated with EB buffer and the energy regeneration mix alone. Cells were incubated in the extract/buffer for 1h at room temperature with frequent tapping. After that, cells were washed with 15 volumes of Buffer D (10mM Hepes pH 7.7, 10mM KCl, 1.5mM MgCl2, 1mM DTT) (spin at 3220 rcf, 4°C, 20mins). Pellets of black colour were observed in the egg extract-treated samples. Subsequently, pellets were resuspended in 15ml of Buffer E (250mM Sucrose, 10mM Tris-HCl pH 8.0, 5mM MgCl2, 1mM DTT, 0.1% Triton X-100) and spun at 3220 rcf, 4°C, 20mins. Subsequently, pellets were resuspended in 1ml of Buffer E each and transferred to a 1.5ml tubes and washed 6 more times with Buffer E (each Chapter 2: Experimental procedures 39 time spun at 1000rcf, 5mins, 4°C). The final pellet contained proteins bound to chromatin, which were isolated and processed for 2-DIGE and proteomics analysis as described in chapter 2.5. 2.8. Immunoblotting analysis. All the immunoblotting analyses were performed according to standard immunoblotting protocols (Green & Sambrook 2012). For the validation of mass spectrometry results, equal amount of proteins isolated in Urea/Thiourea buffer were loaded on each blot (see chapter 2.5). For the immunoblotting on oocytes, embryos or GVs, equivalents of 1 oocyte, cytoplasm or embryo or an indicated number of GVs were loaded on each lane of a gel. Isolations of cytoplasmic proteins from oocytes/embryos were performed using a mild lysis buffer (10mM Tris pH 8.0, 150mM NaCl, 1%NP-40, 1mM EDTA and protease inhibitors): oocytes/embryos were disrupted in the lysis buffer, kept on ice for 10mins, vortexed from time to time, spun at 16100 rcf, 4ºC, 1min and the supernatant was used for immunoblotting. Isolations of nuclear proteins (from somatic cells, sperm, spermatids or from embryos) were performed using Emilie’s Buffer (500mM Tris pH 6.8, 500mM NaCl, 1% NP40, 0.1% SDS, 1% b-Mercapthoethanol and protease inhibitors). If nuclear proteins were isolated from embryos, then first the nuclei were isolated (to remove contaminating yolk, see below) and only then the Emilie’s buffer was used to extract the proteins from the pelleted nuclei. Nuclei were isolated from embryos with the following method: up to 5 embryos were homogenised by pipetting up and down in 50ul of buffer E1 (50mM Hepes-KOH pH 7.5, 140mM NaCl, 1mM EDTA pH 8.0, 10% glycerol, 0.5% Igepal CA-630, 0.25% Triton X-100, 1mM DTT and protease inhibitors) and spun at 1100rcf, 2mins, 4ºC (all the following centrifugations were performed in the same conditions). The Chapter 2: Experimental procedures 40 supernatant and the lipids attached to the walls of the tube were discarded and the pellet was resuspended in 2ml of buffer E1. The spin was repeated and the pellet was again resuspended in 2ml of buffer E1 and incubated on ice for 10mins. After another spin, the pellet was resuspended in 2ml of buffer E2 (10mM Tris pH 8.0, 200mM NaCl, 1mM EDTA pH 8.0, 0.5mM EGTA pH 8.0 and protease inhibitors) and centrifuged. This procedure was repeated twice (the last wash in E2 was preceded by 10mins incubation in E2 on ice). Finally, the pellet (containing the isolated nuclei) was resuspended in the desired volume of Emilie’s buffer). Polyacrylamide gels were cast at the percentage appropriate for the separation of the desired protein size (Green & Sambrook 2012). Gel electrophoresis, transfer of proteins, immunoblotting and washes were performed according to the standard protocols (Green & Sambrook 2012). PVDF membranes with 0.45uM pore size were used (Immobilon, Millipore, IPFL00010) with a semi-dry transfer system (Trans-Blot® SD Semi-Dry Transfer Cell, BioRad, 170-3940) transferring for 30mins, at room temperature (25V). All the protein detections were performed using immunofluorescence with the use of the LI-COR Imaging System. Primary antibodies were used at 1:1000 dilutions, unless stated otherwise and blots were incubated with the primary antibodies overnight at 4°C. Primary antibodies against the following proteins were used in this thesis: Hdac1 (rabbit polyclonal; Abcam, ab33278), Hdac2 (rabbit polyclonal; Epitomics S2398), Mbd3 (1:250; mouse monoclonal; Abcam, ab45027), Hp1γ (goat polyclonal; Abcam, ab40827), HA-tag (1:2500, mouse monoclonal; Sigma, H9658), Rbbp4 (rabbit polyclonal; Abcam, ab1765), Rbbp7 (rabbit polyclonal; Abcam, ab3535), Lsf (rabbit polyclonal; Abcam, ab80445), actin (1:2500, rabbit polyclonal; Sigma, A2103), beta-actin (1:2500; mouse monoclonal; Abcam, ab6276), H3K27me3 (rabbit monoclonal; Cell Signalling Technology, 9733), H2A (rabbit polyclonal; Millipore, 07-146), H2B (1:2000, rabbit polyclonal; Abcam, ab1790), H3 (rabbit polyclonal; Chapter 2: Experimental procedures 41 Abcam, ab18521) and H4 (1:500, rabbit polyclonal; Abcam, ab10158). To detect the primary antibodies, the following secondary antibodies were used (all at 1:25000 dilution): goat anti- mouse IRDye 800 (Licor, 926-32210), Goat anti-Rabbit IRDye 800 (Licor, 926-32211), goat anti-rabbit Alexa 680 (Invitrogen, A-21109), goat anti-mouse Alexa 680 (Invitrogen, A- 21057), donkey anti-goat Alexa 680 (Invitrogen, A-21084). Blots were incubated with the secondary antibody solution for 1h at RT. To reveal the proteins, blots were scanned using a LI-COR detection system (Odyssey laser scanner, LI-COR Biosciences). 2.9. Molecular cloning of candidate sperm factors and mRNA synthesis cDNA from testis isolated from Xenopus laevis was generated with the use of oligodT(15) primer. NCBI-deposited sequences were then used to design primers to amplify the sequences of candidate sperm factors: Sp1 (NM_001137586.1), Sp4 (NM_001087761.1), Sp5 (S71764.1), H1fx (BC041758.1), Mlf1 (NM_001095375.1). Primers used to amplify the candidate factors cDNA sequences and to allow their cloning into the pEntry vector are listed in Table 2. Clonings were performed using pENTR™/D-TOPO® Cloning Kit (Life Technologies, K2400) and Gateway® LR Clonase® II enzyme mix (Life Technologies, 11791-020) according to the manufacturer’s instructions. Directional cloning was performed into the C-terminus (destination vector used was pCS2+ with a C-terminal HA-tag). Clones were first checked by a directional colony PCR screen (using M13 forward primer: GTAAAACGACGGCCAGT and the insert-specific reverse primer) and second, by restriction enzyme digestion and Sanger sequencing (Sanger sequencing reactions were performed at the sequencing facility at the Department of Biochemistry, University of Cambridge). mRNA was synthesized in vitro using MEGAscript® SP6 Kit (Ambion, AM1330M) following the manufacturer’s instructions. mRNA concentration was measured Chapter 2: Experimental procedures 42 with a Nanodrop, adjusted to a final concentration of 1mg/ml with DEPC H2O, aliquoted (aliquots of 2ul) and frozen at -80ºC. Clones of mouse K6B (amino acids 1025-1642) and its catalytically inactive mutant (K6B-mut) were in pCS2+ destination vector with a C-terminal HA-tag and NLS-tag and were a kind gift from Dr Jerome Jullien. Table 2. Sequences of primers used for the cloning of candidate sperm factors Name of primer Sequence XL_SP1_Entry_F CACCATGGCACTGCCCTCCGAGACC XL_SP1_Entry_R CACTATCATGGTTCTGGGAACCCTGCGCTTG XL_SP4_Entry_F CACCATGAGCAAAGTGAGTGGCGGG XL_SP4_Entry_R ACTGCGATAATCTGAGCCATAGTCTCTTGCC XL_SP5_Entry_F CACCATGAGCAAAATGAGAGGCGGG XL_SP5_Entry_R ACTGCGATATTCTGACCCATAGGC XL_H1fx_Entry_F CACCATGGCTCTAGAGCTGGAAGAGAATTTACACAGC XL_H1fx_Entry_R CGCTTTCTTGGATTTAGGCGCTTTGCGGACGC XL_MLF1_Entry_F CACCATGTTCCGCAGTTTGCTGAGAGACTTTGACG XL_MLF1_Entry_R TTTCTCCCTTGCCGGCAACTGCAGCTG 2.10. Cell culture, transfection, immunostaining and microscopic analysis Mouse C2C12 cells were cultured in DMEM (GIBCO, 41965-062) supplemented with 10% FBS and 100u/ml of penicillin/streptomycin at 37ºC, 5% CO2. Xenopus laevis XL 177 cells were cultured in the same medium, but diluted to 60% with a ddH2O at 23ºC. Cells were split whenever the confluence was reached. Transfections were performed with lipofectamine 2000 (Invitrogen, 11668027), following the manufacturer’s protocol. For the immunostaining procedure, cells were fixed with 4% PFA in1XPBS, permeabilised with 0.1% Triton X-100 in 1XPBS and blocked with 5% BSA in 1XPBS. Antibody against the HA-tag (mouse monoclonal; Sigma, H9658) was diluted 1:500 and detected with a secondary antibody conjugated with Alexa 488, diluted 1:500 (donkey anti-mouse, Invitrogen, A21202). Samples were mounted on a microscope slide in Vectashield mounting medium with DAPI Chapter 2: Experimental procedures 43 (Vector Labs, H-1200) and sealed with nail polish. Microscopic analyses were performed using Zeiss 510 META confocal LSM microscope. 2.11. Cell squashing for ploidy assessment Early tadpoles (stage 30-36) were anaesthetised and decapitated. Subsequently, tadpoles were transferred on a microscope slide and the yolky tissues were removed. The remaining tissues of the tadpole were then squashed with the coverslip and observed under a phase contrast microscope. 2.12. Injection of 1-cell embryos with mRNA Eggs were in vitro fertilised and dejellied using 2% cysteine solution in 0.1XMMR. Injections into 1-cell stage embryos were performed in injection solution (Smith et al. 2006) using a Drummond Nanoject microinjector, delivering 9.2ng of mRNA per injection (mRNA at 1mg/ml in DEPC H2O). Embryos were cultured at 18°C and collected for qRT-PCR analysis (chapter 2.14) at the stage indicated in the text (see Chapter 3.7). Expression of K6B and K6B-mut proteins in embryos, as well as the removal of H3K27me3 by K6B were confirmed by immunoblotting. Chapter 2: Experimental procedures 44 2.13. Injection of mRNA or oligonucleotides into the oocytes Oligonucleotides and qPCR primers for HP1γ were designed by Matthew Jones, a 4-year PhD rotation student during his rotation in the Gurdon laboratory. Ovaries were isolated from frogs pre-primed with PMSG at least 48h before use. Oocyte defolliculation was performed enzymatically with a liberase treatment (Halley-Stott et al. 2010). For in vitro maturation experiments, oocytes were liberated for a maximum time of 1h 15mins and the remaining defolliculation was done manually (Miyamoto et al. 2013). Oocytes were cultured in 1XMBS medium (88mM NaCl, 1mM KCl, 2.4mM NaHCO3, 0.82mM MgSO4, 0.41mM CaCl2, 0.33mM Ca(NO3)2, 10mM HEPES pH7.4, 10ug/ml streptomycin sulfate and 10μg/ml penicillin). mRNA was injected as described above, with the difference that the use of injection solution was omitted (oocytes were injected directly in the culture medium). Oligonucleotide design was performed according to the guidelines published (Hulstrand et al. 2010). Sequences of the oligonucleotides used are listed in Table 3. The oligonucleotide injection and in vitro maturation and ICSI (for the K6B/K6B-mut) experiments were conducted as described before (Miyamoto et al. 2013). Sequences of primers used for the assessment of the knockdown are listed in Table 4. Chapter 2: Experimental procedures 45 Table 3. Sequences of oligonucleotides used for the knockdown of candidate egg factors Name of oligonucleotide Sequence HP1g #1 C*T*T*CCTCCACCTTCT*T*G*C HP1g #2 A*C*C*TTTCCATTTACTAC*A*C*G HP1g #3 T*T*C*AATCAACTCTGGACA*G*T*C Hdac1 #1 C*C*A*ACATCACCATCA*T*A*G Hdac1 #2 C*C*A*TAGTTGAGCAGC*A*G*G Hdac1 #3 T*G*T*CTGGTCGTATGG*A*G*C Hdac2 #1 G*A*C*CTTCTTCTTGGC*A*C*C Hdac2 #2 C*A*C*CATCATAATAGT*A*G*C Hdac2 #3 C*G*G*ATTCTGTGAGGC*T*T*C Mbd2/3 #1 C*T*G*AGGCTTACTGCG*G*A*A Mbd2/3 #2 T*C*C*TAAGTAACGAGC*A*A*G Mbd2/3 #3 C*C*G*CATTCGCTGTCT*G*T*T LSF #1 T*G*C*CAGCGGTAACTT*C*A*G LSF #2 T*C*C*ACCTCCATCCTT*C*T*A LSF #3 T*A*T*TCTGTCTCCAGG*T*C*T Rbbp4 #1 T*A*T*GACTCGTTCCTC*C*A*C Rbbp4 #2 G*A*G*CATCATCATTAG*G*A*A Rbbp4 #3 A*G*G*TTGCCACTTAGG*T*T*G Rbbp7 #1 C*T*T*CCACTGTATCCT*C*A*A Rbbp7 #2 C*A*T*AACCAGGTCATA*C*A*G Rbbp7 #3 G*G*A*ACCTGGACACGA*G*C*C * denotes a phosphorothioate bond Table 4. Sequences of primers used for qRT-PCR assays Name of the primer Sequence Hdac1_F1 CGCTCCATACGACCAGACA Hdac1_R1 GCCATCAAACACAGGACAGT Hdac2_F1 TCTGTAGCTGGTGCTGTAAAACTCA Hdac2_R1 CCTGCCCAGTTAACAGCCATA HP1g_F1 GGGAGCCTGAGGAAAACTTAG HP1g_R1 CAAATCCCCGTGGTTTATCA Rbbp4_F1 GGAGAGTTTGGAGGCTTTGG Rbbp4_R1 TGGAGTTTTGGTGGCAATAA Rbbp7_F2 GTGTCCAGGTTCCCAATGAT Rbbp7_R2 CAAAGCCACCAAACTCTCCT PWP1_F GACTTCGAAAATCTGGCATCTCA PWP1_R GGGACTTTCACCATTGACTTAAACA GATA3_F2 CACAGGATCTCCATTGGCATT GATA3_R2 CCTGTGCAAACTGTCAAACCA SFRP2_F1 GGAATAAGAAGAGACAGGCCCAAT Chapter 2: Experimental procedures 46 SFRP2_R1 TTACCAAAGCCACCCCAGAA C19ORF26_tr_F2 GCCATCAACCCCTTCTTCATC C19ORF26_tr_R2 ACACGTTACCACAGCACTTTGC PLOD2_F2 CACATTCTTTATTCTGCCGACAGAT PLOD2_R2 AAGAGCCAATCACGCAAGCT NA_F1 GATGCTCAGCTTTGGATCTTGA NA_R1 CCACACGGGCCTGATCTG MN1_F2 TGCCTTCAGCTAGGGACACA MN1_R2 CACCCAGTCGTGATAAAGCAGTAG XL_HES1_F1 TGAGCAATACCCCGGATAAG XL_HES1_R1 TCCAGGATGAGGGTTTTGAG WAVE1_F2 AGGAATCCAGCTTCGCAAAG WAVE1_R2 ACGCGCTCGTGTTTTGCT HOXB1_F CCCCACAAAATTGCAACCA HOXB1_R TCTGCTTCTTGGCTGGCATA HES7_F1 TCCTCTCCCTCCGCCTTTT HES7_R1 CCATGGAAACCCATAGAAAGCT DOLPP1_F GGGCATTCGCTATGCTCTCT DOLPP1_R GCCTGAAATCCCTCAACCAA ZNF33A_F2 GGTCTGTCTCATCCTGAATGCTT ZNF33A_R2 AATAGGTGTGGATTCTGCTGTTGA SOX21_F1 CCCACATTGGGTTCCAACTG SOX21_R1 GGCATGACAGCCCGACTAAG XL_GJB1_F1 GCATCAGCAAAGAGCATCAA XL_GJB1_R1 CAGGGAGCCGTGAGAGTTAG FOSL2_F1 TGTGTGATAAAGTAGACCAGAGGATTTT FOSL2_R1 GTCGCTTGTTTCCTTTTCAACA FOS_F2 AGTCCTGGATCGCCGAGTT FOS_R2 TCACAGTAACCGCAACGATCTATT CHD3_F2 GTTCCCACGCACGTTTGTT CHD3_R2 TGGCTGCTGCATCCATAATG MIX1b_F2 AGGATGGAGTGGAGGATCTGAA MIX1b_R2 GGTTCTCCGGGAAGGTAAAGG XL_18srRNA_F1 ATGGCCGTTCTTAGTTGGTG XL_18srRNA_R1 TATTGCTCGATCTCGTGTGG XL_28srRNA_F1 TCATCAGACCCCAGAAAAGG XL_28srRNA_R1 GATTCGGCAGGTGAGTTGTT Chapter 2: Experimental procedures 47 2.14. RNA extraction and qRT-PCR analysis Oocytes or embryos were collected and frozen at -80°C. RNA extractions were performed using Qiagen RNeasy Mini kit (Qiagen, 74106) according to the manufacturer’s protocol, unless stated otherwise. RNA was eluted in 50ul DEPC H2O and was used for cDNA synthesis and for RT-qPCRs, as described before (Halley-Stott et al. 2010). For all the experiments described, apart from rRNA levels assessment, cDNA synthesis was performed using oligo dT(15) primer (0.5ul of 100uM primer/reaction). For rRNA levels measurement, gene-specific primers were used for cDNA synthesis. qPCRs were performed with gene specific primers (Table 4) using a SybrGreen detection system (Sigma, S9194) and ABI 7300 machine (Applied Biosystems) as detailed before (Halley-Stott et al. 2010). Results were exported to Microsoft Excel for analysis. Gene expression was normalized to pwp1 or dolpp1 transcripts. Subsequently, a Grubb’s test was used to identify and exclude any potential outliers in the datasets with a p-value cut-off < 0.05: 1 out of 15 samples for spermatid- derived embryos for rRNA expression analysis was identified as outlying, 1 out of 6 sperm/spermatid embryos was outlying for mn1 and 1 out of 6 for chd3 transcript for RNA- seq validation analysis. Statistical significance was assessed using a t-test. Chapter 2: Experimental procedures 48 2.15. Analysis of DNA replication The protocol described here is a modification of the original protocol kindly provided by Dr Vincent Gaggioli. Replication on single DNA fibres was performed as described before (Gaggioli et al. 2013) with slight modifications. Freshly prepared egg extracts were supplemented with energy regeneration mix (components as mentioned in chapter 2.7) and with 20uM biotin-16- dUTP (Roche, 11093070910). Permeabilised cells were added to a final concentration of 200 nuclei/ul of extract and incubated at RT for 30mins, 40mins or for 2h (tapping every 10 minutes). Reaction was stopped by adding 10 volumes of ice-cold 1XPBS (Phosphate Buffer Saline: 137mM NaCl, 2.7mM KCl, 10mM Na2HPO4x2H2O, 2mM KH2PO4) and cells were spun down at 1000g, 4°C, 7 minutes. Cells were resuspended in 50ul of 1XPBS and mixed immediately with 50ul of melted (at 65°C) 2% low melting point agarose (Invitrogen, 16520050) in 1XPBS. After solidification, the agarose plug was incubated overnight (O/N) at 50°C with 1ml 0.5M EDTA pH8.0, 100uL 10% sarkosyl (Sigma, L5125), 1mg/mL Proteinase K (New England Biolabs, P8102S) followed by three washes in TE pH 6.5. Subsequently, the plug was incubated twice in TE supplemented with 0.1mM PMSF (Sigma, 93482) for 30 minutes at 50°C and washed four times with 1ml of 50mM MES (Sigma, 69889) pH 6.35, 1mM EDTA (1h at RT each wash). Then the solution was removed; the plug was melted in 400ul of MES pH 6.35, 1mM EDTA at 68°C for 20 minutes and the agarose was digested with 2 units of β-agarase (New England Biolabs, M0392S) O/N at 42°C. Silanised coverslips were prepared as described before (Labit et al. 2008). Thirty microliters of replicated DNA solution was pipetted onto a silanised coverslip, covered with a Chapter 2: Experimental procedures 49 non-silanised coverslip and incubated for 5 minutes at RT. Subsequently, the top coverslip was slid away to stretch DNA fibres and the silanised coverslip with stretched fibres was fixed in 3:1 solution of Methanol:Glacial Acetid Acid for 10 minutes, RT. The fibres were then denatured with 2.5M HCl (1h, RT) and dehydrated by washes in 70% ethanol, 90% ethanol and 100% ethanol (1 minute for each wash). Subsequently, the coverslip was dried, washed 3 times in 1XPBS, 0.1% Tween (Sigma, P5927) (5 minutes for each wash) and blocked in 3%BSA in 1XPBS (1h, RT). All antibodies were diluted in 1XPBS, 3%BSA, 0.1% Tween. Total DNA and replicated DNA were simultaneously detected using the following primary antibodies: anti-DNA antibody (Millipore, MAB3034) 1:300 dilution and streptavidin-Alexa 594 antibody 1:50 to detect biotin (Invitrogen, S-11227) for 30 minutes at 37°C. Slides were washed with 1XPBS, 0.1% Tween (4 washes) and secondary antibodies (diluted 1:50) were added: chicken anti-mouse Alexa 488 (Invitrogen, A-21200) and biotinylated antibody anti-streptavidin (Vector Labs, BA-0500) (incubation for 30 minutes, 37°C). After four washes in 1XPBS, 0.1% Tween, a tertiary detection was performed with antibodies diluted 1:50: goat anti-chicken Alexa 488 (Invitrogen, A-11039) and streptavidin- Alexa 594 (for 30 minutes, 37°C). The coverslip was washed three times with 1XPBS 0.1% Tween, three times in 1XPBS, mounted on a microscope slide with a mounting medium (50% glycerol in 1XPBS), and sealed with nail polish. Images were acquired with a Zeiss 510 META confocal LSM microscope. Image analysis was performed in ImageJ; the amount of replicated DNA and total DNA was measured individually on single DNA fibres. The data was analysed in Microsoft Excel with a macro kindly provided by Dr Vincent Gaggioli. Chapter 2: Experimental procedures 50 2.16 Pulldown of BrUTP-labelled RNA from haploid sperm- and spermatid-derived embryos Sperm- and spermatid-derived haploid embryos were generated as described above, but using SDB buffer containing 100mM BrUTP (Sigma, B7166). Single embryos were harvested at stage 10.5-11.5 and total RNA was isolated with Qiagen RNeasy Mini kit (Qiagen, 74106) according to the manufacturer’s protocol (elution in 50ul DEPC H2O). For each embryo 20ul of agarose-conjugated beads with antibody against BrdUTP (Santa Cruz Bioechnology, sc-32323) were washed twice in 1ml of 0.5XSSPE (wash buffer) (1XSSPE: 300mM NaCl, 20mM NaH2PO4xH2O, 2mM EDTA), 0.05% Tween 20, 0.1% PVP (Sigma, P5288) by centrifugation for 2 minutes at 2000g, RT. Subsequently, beads were blocked O/N on a rotating wheel at 4°C in wash buffer supplemented with 1mg/ml of RNase-free BSA final (New England Biolabs, B9000S). Next, 20ul of beads were resuspended in 500ul of 0.5XSSPE supplemented with 0.05% Tween and RNase Inhibitor (New England Biolabs, M0314S), mixed with 30ul of purified RNA and incubated for 4 hours on a rotating wheel at 4°C. Subsequently, beads were washed for 10 minutes on a rotating wheel at 4°C with the following solutions: once with 0.2XSSPE, 0.05% Tween; twice with 0.5XSSPE, 0.05% Tween, 150mM NaCl and once with 1XTE (10mM Tris, 1mM EDTA, pH 8.0), 0.05% Tween (spins 2000g, RT for 2 minutes). RNA was eluted four times for 5 minutes at RT with 100ul of elution buffer (300mM NaCl, 5mM Tris, 1mM EDTA, 0.1%SDS, 2mM DTT); the eluates were pooled, extracted with phenol:chloroform, precipitated by adding 1ml of EtOH, 40ul of 3M NaAcetate and 1ul of 10mg/ml tRNA (Sigma, R8759) and resuspended in 15ul DEPC H2O. Chapter 2: Experimental procedures 51 2.17. Preparation of cDNA library for sequencing Embryos were collected and frozen at -80°C (pools of 5 embryos) at stage 10.5-11.5. RNA extractions were performed using Qiagen RNeasy Mini kit (Qiagen, 74106) according to the manufacturer’s protocol. RNA was eluted in 50ul DEPC H2O and was used to generate a cDNA sequencing library with Illumina TrueSeq kit (RS-122-2001), according to the manufacturer’s protocol. Quality and the size of libraries obtained were validated using Tapestation equipment and software (Agilent). 2.18. Preparation of ChIP-seq samples Sperm and spermatids were separated as described above. Chromatin fractionation and chromatin immunoprecipitation (ChIP) were performed as described before (Erkek et al. 2013, Hisano et al. 2013) with slight modifications. Pre-treatment of cells with DTT was omitted and chromatin was digested with 2.5U of MNase/1 million of cells (Roche, 12533700) for 30 minutes at 37°C. Following the digestion, antibodies against histone marks were added: anti-H3K4me2 (Millipore, 07-030), anti-H3K4me3 (Abcam, ab8580), anti- H3K4me3 (Millipore CS200580), anti-H3K27me3 (Millipore, 07-449), anti-H3K27me3 (kind gift from Dr Thomas Jenuwein). Before ChIP, primary antibodies were bound to magnetic beads conjugated with secondary antibodies (Invitrogen, 11204D) according to the manufacturer’s protocol and all wash steps in the protocol were performed using a magnet, instead of centrifugation. Bound DNA was isolated, separated by electrophoresis and mononucleosomal bands from sperm and spermatids were excised (Hisano et al. 2013) and subjected to library preparation with TruSeq DNA kit (Illumina, FC-121-2001), according to Chapter 2: Experimental procedures 52 the manufacturer’s protocol. Quality and the size of libraries obtained were validated with the use of Tapestation equipment and software (Agilent). 2.19. Bioinformatic analyses All the methods in this section were written and kindly shared by Dr Angela Simeone, Dr George Allen and Dr Charles Bradshaw. 2.19.1. Sequencing of libraries RNA-Seq and ChIP-Seq libraries were sequenced on an Illumina HiSeq 2000 instrument in single read mode at 36 base length or paired read mode at 50 or 100 base length. The resulting fastq files were filtered and mapped against the Xenopus laevis genome (JGI version 6.1) using BWA 0.6.2 (ChIP-Seq) or TopHat 2.0.6 (RNA-Seq) (Li & Durbin 2009, Trapnell et al. 2009, Kim et al. 2013). 2.19.2. Xenopus laevis transcriptome The 553,960 assembled transcripts were provided by the International Xenopus Genome Project (http://www.marcottelab.org/index.php/Xenopus_Genome_Project) in October 2012. This assembly was augmented with Xenopus laevis sequences from the NCBI RefSeq database downloaded in Feburary 2012 (30,611 sequences). The combined transcript sequences were filtered with cd-hit-est 4.5.7 (Li & Godzik 2006) with a similarity score of Chapter 2: Experimental procedures 53 95% to remove redundant sequences. This resulted in a final set of 39,384 transcripts. To provide gene names, orthologues were found against the M. musculus proteome (downloaded in January 2013 – NCBI RefSeq) using Inparanoid 4 (Alexeyenko et al. 2006) on predicted ORFs from the Trinity Suite (Grabherr et al. 2011). The sequences were further annotated using InterProScan 4.8 (Zdobnov & Apweiler 2001) to provide both InterPro Domains (Release 35) and Panther 7.2 ontology terms (Thomas et al. 2003). Xenopus laevis NCBI descriptions were provided for transcripts that originated from the NCBI. 2.19.3. Filtering sequencing data Fastq files were filtered for low quality reads ( 100kb. This resulted in 34,373 transcripts mapping to the genome. This mapping was used as a junction file for Tophat 2.0.6 Chapter 2: Experimental procedures 54 (Trapnell et al. 2009, Kim et al. 2013) which was used to map the RNA-Seq reads to the genome. Read counts were then generated for each of the transcripts. 2.19.5. Differential expression RPKMs were calculated by normalizing read counts for each transcript by the transcript length and the total number of reads in the corresponding sample. Zeros were replaced with values obtained by randomly sampling from all RPKM values greater than zero and less than 0.2. These were converted back to raw counts, rounding up to the nearest integer, and then normalized using the Bioconductor package EdgeR (Robinson et al. 2010). Transcripts were kept in the analysis if they had at least one count per million in all of the sperm-derived embryo samples or all of the spermatid-derived embryo samples, leaving 18,340 transcripts post-filter. Differentially expressed transcripts were then called using EdgeR, taking into account the pairing of sperm- and spermatid-embryos in the design matrix of the model. Gene ontology terms over-represented among the differentially expressed genes were found using topGO (Alexa et al. 2006). 2.19.6. Heatmaps for differentially expressed genes For each differentially expressed gene, log2 fold changes were calculated pairwise for each spermatid/sperm-derived embryo pair. Genes were filtered out if they were not consistently upregulated in at least 6 of 7 pairs or consistently downregulated in 6 of 7 pairs. Chapter 2: Experimental procedures 55 These log2 fold change values were then plotted for each remaining gene, ordered by mean fold change, using heatmap.2 from the gplots library in R. 2.19.7. Genome-wide correlation analysis of ChIP-seq data For each ChIP experiment, reads in the bound (IP) and in the input samples (input) were normalised to the total number of reads aligned and scaled by a factor of 10 6 (i.e. values represent count per million, cpm). The entire genome was binned into 200bp wide windows. The coverage was computed as the number of reads in each window normalised by the total number of reads in the experiment. For each mark in each cell type the reproducibility was evaluated by estimating the Pearson correlation coefficient between ChIP-Seq replicates. 2.19.8. Histone methylation level analysis The methylation level was computed as: where NIP is the total number of aligned reads in the IP experiment and Ninput is the total number of aligned reads in the input experiment. Chapter 2: Experimental procedures 56 For H3K4me2/3 the methylation level was computed in a window around the TSS [TSS-10kb, TSS+2kb]. For H3K27me3 the window considered for estimating the methylation level included the 10 kb upstream region together with the gene body. Methylation levels across replicates were averaged. Heatmaps for methylation levels at misregulated genes were generated in the same way as for differentially expressed genes. 2.19.9. Peak calling for histone marks The detection of highly methylated histone regions (peaks) was performed with MACS2 2.0.9 (Zhang et al. 2008) using the broad-region option and a q-value of < 0.01. The list of confirmed peaks for each histone mark analyzed consisted of the peaks with a p-value of < 0.01 detected in at least two out of three replicates. 2.19.10. Statistical testing of ChIP-seq data Statistical analysis was conducted in R (http://www.R-project.org). The comparison of the methylation levels between promoter regions of 100 misregulated genes and promoter regions in the entire genome was performed using Kolmogorov-Smirnov test (R function ks.test()). The difference between methylation levels of promoter regions between cell types (sperm, spermatids) was tested in the same way. The enrichment in the proportion of misregulated genes positive for H3K27me3 in sperm and spermatid was tested by the non-parametric Chi-squared test for proportions (R function prop.test()). Chapter 3: Results 57 Chapter 3 Sperm-derived embryos develop better than nuclear transfer-derived and spermatid-derived embryos 3.1. Introduction The goal of my PhD is to understand the mechanisms of sperm programming for embryonic development. To be able to investigate these mechanisms, firstly a good system is needed in which the developmental capacity of sperm, and of other cells, could be tested and compared. One would ideally want to compare the developmental potential of embryos generated with sperm or with a somatic cell. However, a typical somatic cell is not appropriate for such comparison, for the reason that it is diploid. Therefore, if embryos are to be generated with a diploid cell (by a nuclear transfer to unfertilised egg), the egg would need to be enucleated. By contrast, fertilisation of an egg by a sperm naturally gives rise to a diploid embryo with no need for egg enucleation. Thus, comparison of somatic cell nuclear transfer with fertilisation is limited by technical differences in embryo generation; one would need another haploid cell for fair comparisons with sperm. Therefore, it was reasoned that spermatids, immature precursors of sperm, could be appropriate for such comparisons. First, the same ploidy would enable their comparison in the same assay and second, the fact that they come from the same lineage, reduces the differences between the two cells. Too many differences, for example between sperm and somatic cells, would make it difficult to identify those differences, which are developmentally-relevant. Chapter 3: Results 58 In this chapter I first described how I isolated sperm and spermatids from Xenopus laevis testicular cells. Second, I described the results of the comparison of the developmental potential of sperm- and of spermatid-derived embryos. Last, I described the results of the comparison of the developmental ability of spermatid-derived embryos and nuclear transfer- derived embryos. 3.2. Separation of Xenopus laevis sperm and spermatids To compare the abilities of sperm and spermatids to support embryonic development I first needed to establish a method allowing their separation from Xenopus leavis testes. I have achieved this by adapting and modifying a previously described method using a density gradient centrifugation in a Metrizamide gradient (Risley & Eckhardt 1979). Metrizamide was unfortunately no longer commercially available, therefore I have found another non-ionic solution, Optiprep (Sigma, D1556), with a similar density. The density of 60% Metrizamide solution used in (Risley & Eckhardt 1979) was 1.33g/cm 3 , whereas the density of Optiprep solution (which is 60% solution of iodixanol in water) was 1.32g/cm 3 . I have therefore used Optiprep in my attempts to separate Xenopus testicular cells. Briefly, a mixture of testicular cells was isolated from testes (see Experimental procedures) and overlaid on an Optiprep step gradient prepared in a centrifuge tube. A step gradient was prepared as described (Risley & Eckhardt 1979), with a slight modification: an additional 1ml of 20% step was introduced between the 30% and 12% steps (Fig. 5). Centrifugation was performed in the same conditions as described in (Risley & Eckhardt 1979) and this allowed separation of testicular cells into two different fractions: spermatids (collected from the top interface fraction) and mature sperm (pelleted at the bottom of the tube) (Fig. 5). Chapter 3: Results 59 Fig. 5. Separation of sperm and spermatid by density gradient centrifugation The diagram explains the procedure of separating sperm and spermatids from Xenopus laevis testis. First, step gradients of Optiprep are prepared in a centrifuge tube. Percentages of Optiprep in 1XMMR in each step are indicated in the diagram (see also Experimental procedures). Testicular cells isolated from the testis are overlaid on top of the prepared gradient. After centrifugation, spermatids are recovered from the interface between the 1XMMR and 12% Optiprep fraction, and the sperm is recovered from the bottom of the tube (pelleted cells). Morphological observations under a phase contrast microscope, as well as DAPI staining of these two cell populations confirmed their successful separation (Fig. 6A and 6B). The mature sperm population was 95-99% pure, and it was possible to assess the purity of the sperm by microscopic observations, due to the fact that mature sperm in Xenopus laevis have a characteristic, snake-like shape (Risley & Eckhardt 1979, Risley et al. 1982). It is however not possible to judge the purity of the spermatid population by simple microscopic observations, due to the fact that spermatids at many stages of their maturation, have a round or elongated shape. Spermatids can therefore be misassigned as another cell types by an inexperienced researcher (Abe 1988, Abe & Hiyoshi 1991). To circumvent this problem, the purity of the spermatid fraction was assessed by flow cytometry analysis. Spermatids have already completed meiosis and are haploid with a reduced DNA content, so they can be Chapter 3: Results 60 distinguished from other round-shape diploid and tetraploid testicular cells by DNA staining and ploidy assessment. Flow cytometry assessment estimated the spermatid content to be around 80% (Fig. 6C). These two purified populations were used in all the subsequent experiments in which sperm and spermatids are compared. 3.3. Sperm-derived embryos develop better than spermatid-derived embryos Next, I needed to develop an experimental setup in which the developmental potential of sperm and spermatids can be compared. As opposed to sperm, spermatids are not motile and therefore cannot swim to the egg for fertilisation. I therefore reasoned that a fair comparison between the two cells would be to inject them directly into the cytoplasm of the egg. For that, a protocol for intra-cytoplasmic sperm injection (ICSI) was adapted from (Smith et al. 2006). The technique relays on injecting single permeabilised sperm cells into unfertilised eggs (Fig. 7A). I confirmed that embryos obtained in such way can develop into healthy adult organisms (Fig. 7B). For the purpose of comparing the developmental capacity of sperm and spermatids, sperm and spermatids were permeabilised and subsequently injected into unfertilised Xenopus eggs. The embryo development was assessed at two different stages: at an early gastrula stage and at a swimming tadpole stage and the embryos were scored as the percentage of those reaching a gastrula/swimming tadpole stage to the total number of cleaved embryos (Fig. 7C). Chapter 3: Results 61 Both sperm- and spermatid-derived embryos reached the gastrula stage with a similar efficiency. However, sperm-derived embryos developed significantly better to the swimming tadpole stage than spermatid-derived embryos (p-value < 0.05) (Fig. 8A and 8B). Fig. 6. Purity assessment of the spermatids and sperm populations (A) Observations of sperm and spermatids population with a phase contrast microscope reveals that the two populations isolated are morphologically different. Sperm cells have a characteristic snake-like shape (left panel), whereas spermatids are round cells with a dense interior structures (right panel). Scale bar = 10um. (B) DNA staining with DAPI of sperm and spermatids confirms their morphological (nuclear shape) differences. Scale bar = 10um. Due to the fact that the sperm cells have a distinct morphology, microscopic observations (A and B) allow the estimation of the purity of sperm population to be around 95-99%. (C) Flow cytometry-based assessment of the purity of the spermatid population. Spermatids (haploid cells) consist around 80% of all the cells. Chapter 3: Results 62 Fig. 7. ICSI with sperm and spermatids (A) Diagram explaining the principles of the ICSI procedure. The sperm is injected directly into the cytoplasm of the unfertilised egg. Successfully injected eggs can subsequently develop into tadpoles. (B) Healthy frogs obtained by ICSI procedure. (C) Diagram explaining the design of ICSI experiment to compare the developmental potential of sperm and spermatids. Sperm or spermatids are injected into unfertilised egg. Such egg develops and embryos are scored at two stages: at an early gastrula stage and at a swimming tadpole stage. Chapter 3: Results 63 Fig. 8. Sperm-derived embryos develop better than spermatid-derived embryos (A) Graph summarising the results of ICSI experiments with sperm and spermatids. Embryos were scored at two different developmental stages: at a gastrula stage and at a swimming tadpole stage. Sperm- and spermatid-derived embryos developed similarly to the gastrula stage, but the sperm-derived embryos developed significantly better to the swimming tadpole stage. * - p-value = 0.000002 (z-test). N = 6 independent experiments. Error bars show ±SEM. Numbers above each bar represent the number of embryos tested. (B) Shows representative images of sperm-derived and spermatid-derived embryos. Note that developmental abnormalities of spermatid-derived embryos are not visible before stage 12. Timescale below the images indicates average developmental time for embryos cultured at 18ºC. Scale bar = 1mm. Chapter 3: Results 64 3.4. Nuclear transfer-derived embryos develop with a similarly low efficiency as spermatid-derived embryos and worse than sperm-derived embryos. Developmental potential of somatic cells cannot be directly compared to that of sperm and spermatids, due to differences in ploidy of these cells and therefore different technical manipulations required to generate the embryos. However, to get an idea as to whether development of somatic cell-derived embryos is more similar to the development of sperm- derived or spermatid-derived embryos, nuclear transfer experiments were performed and compared to the results of ICSI experiments obtained above. Late blastula/early gastrula stage cells were used as nuclear donors for the somatic cell nuclear transfer experiments, as these cells proved to be efficient at supporting embryonic development (Gurdon 1962). Optimisation of the nuclear transfer procedure is described in the chapter 5.2.3. Cell membranes were mechanically disrupted and single nuclei were then transferred into enucleated eggs (Fig. 9A). Resulting embryos were scored as before: at an early gastrula and at a swimming tadpole stage (as a percentage of the cleaved embryos). Subsequently, the results were compared to the results obtained with ICSI with sperm and spermatids. Nuclear transfer-derived embryos developed to the gastrula stage with a similar frequency to sperm- and spermatid-derived embryos. Interestingly, nuclear transfer-derived embryos developed to a swimming tadpole stage less efficiently than sperm-derived embryos (p-value < 0.05) and with a similar efficiency to spermatid-derived embryos (p-value > 0.05) (Fig. 9B). Chapter 3: Results 65 Fig. 9. Sperm-derived embryos develop better than nuclear transfer-derived embryos (A) Diagram explaining the procedure of nuclear transfer to Xenopus laevis eggs. The egg is enucleated by UV radiation and subsequently injected with an embryonic cell nucleus. Successfully reconstructed embryos can develop to a swimming tadpole stage. (B) Comparison of the developmental potential of sperm-, spermatid- and nuclear transfer- derived embryos. Scoring was performed as described in Fig. 8. Note that the data for sperm- and spermatid-derived embryos come from the experiments described in Fig. 8. Nuclear transfer embryos are from 4 independent experiments. Error bars show ±SEM. Numbers above each bar represent the number of embryos tested. * indicates p-value=0.002; ** indicates p-value=0.000002 (z-test). Chapter 3: Results 66 3.5. Summary The results obtained so far show that sperm-derived embryos developed significantly better than spermatid-derived embryos. Furthermore, the developmental potential of nuclear transfer-derived embryos was as low as that of spermatid-derived embryos (as compared with sperm-derived emrbyos). This suggests that the sperm is better suited to support embryonic development than a spermatid and a somatic cell. Since a direct comparison between sperm and somatic cells is not possible and since spermatids were similarly inefficient at supporting development as somatic cells, in all subsequent analysis the sperm is compared with spermatids. Chapter 4: Results 67 Chapter 4 Identification of proteins present in sperm, spermatids and incorporated into sperm and spermatids from the egg extract 4.1. Introduction Results obtained so far suggest that the sperm, as opposed to the spermatid, is programmed to support efficient embryonic development. Spermiogenesis is a complex, multistep process, involving numerous molecular changes to the maturing spermatid nucleus. Many proteins are lost and gained during sperm maturation (Gaucher et al. 2010). It is therefore possible, that the loss or gain of particular proteins in the course of spermiogenesis is responsible for the acquisition of the developmental advantage of sperm, as compared to the spermatid. Such proteins could have a direct or indirect effect on embryonic development. For example, if sperm was delivering transcription factors to the embryo, they could directly influence the embryonic development. On the other hand, sperm factors could also have indirect effect on development if they were recognised by egg-derived effector proteins. For example, sperm-derived protamines are recognised and processed by egg-derived Nucleoplasmin (Philpott et al. 1991, Philpott & Leno 1992), which could help the sperm nucleus to acquire a chromatin state compatible with early development. I therefore aimed to identify proteins that: 1) are present in the sperm nucleus itself, or 2) are egg factors that are specifically attracted to the sperm chromatin. To identify the first factors, I have compared proteins present in sperm and spermatids. To identify the second type of factors, I have incubated sperm and spermatids in egg extracts and compared the proteins bound to each type Chapter 4: Results 68 of chromatin. In both cases the identification of differences in protein composition between the two samples was performed with the use of 2-DIGE electrophoresis (2-D Fluorescence Difference Gel Electrophoresis) followed by mass-spectrometry analysis of selected protein spots. This approach led to the identification of 51 sperm-specific proteins, 47 spermatid- specific proteins and also 107 egg proteins binding specifically to sperm upon egg-extract treatment and 20 egg proteins incorporated specifically into spermatid chromatin upon egg extract incubation. 4.2. Sperm and spermatids differ in their nuclear protein composition First, the nuclear composition of sperm and spermatids was compared. For that, the cells were prepared in the same way as for ICSI experiments (see Experimental procedures). Proteins were extracted with Urea/Thiourea buffer (see Experimental procedures). Subsequently, equal amount of proteins isolated from sperm and spermatids were labelled with fluorescent Cy3 and Cy5 dyes and separated in two dimensions (Fig. 10A and B). During spermiogenesis, the protein composition of the nucleus of the maturing spermatid undergoes numerous changes (Gaucher et al. 2010). For example many proteins that become incoportated into the sperm chromatin in Xenopus laevis, are highly basic (Abe & Hiyoshi 1991, Hiyoshi et al. 1991). Therefore, in order to allow an appropriate separation of all proteins and of nuclear proteins, amongst which many are highly basic, the first dimension electrophoresis (separating proteins according to their isoelectric point) was performed separately in two different pH ranges: 3-10 (to better separate the majority of the proteins) and in pH range 7-11 (to specifically separate the basic proteins). Subsequently, all the proteins were separated by electrophoresis in the second dimension, according to their molecular mass. Gels were imaged using a laser scanner to identify protein spots which were Chapter 4: Results 69 present specifically in sperm, spermatids and those which were common between the two samples (manual identification) (Fig. 11A and B). Afterwards, gels were silver-stained in order to allow visualisation and excision of selected protein spots (Fig. 12A and B). Subsequently, proteins isolated from each spot were subjected to mass spectrometry analysis. Only those proteins which had an overall protein probability score (calculated by Mascot: http://www.matrixscience.com/help/scoring_help.html) above 100 (the higher the score, the more probable the correct identification of the protein), or a score below 100, but at least two different peptides confirming their identity, were included in the final list of identified proteins. In total 51 sperm-specific, 47 spermatid-specific and 38 proteins present in both cell types were identified (Table S1). Amongst the sperm-specific proteins identified, proteins previously reported to be present in Xenopus laevis sperm were found, for example sperm basic protein 1 (Sp1), sperm basic protein 4 (Sp4) (Sp1-6 proteins are functional orthologues of mammalian protamines) and histone 1 variant H1fx (Shechter et al. 2009), which confirms that the approach used to identify these proteins is valid. Similarly, in the list of spermatid- specific proteins a homologue of a human spermatid-specific protein, Rsb-66, was found (Yang et al. 2003, Chen et al. 2008), which confirms a successful separation of a spermatid population and also indicates that the approach used can successfully identify spermatid- specific proteins. Proteins that were found in both cell types contained mainly basic metabolism and structural proteins, for example actin, tubulin or ATP synthase (Table S1). Within the sperm-specific proteins there are several proteins which can be implicated in rendering the sperm susceptible to egg reprogramming. For example, I have identified Wdr5 protein in the sperm nuclei. Wdr5 has been shown to recognise and bind to dimethylated and trimethylated lysine 4 of histone H3 (H3K4me2/3). Wdr5 then recruits histone H3K4 methylase (via a direct interaction with the methylase), which results in further spreading of the activating H3K4me2/3 epigenetic mark and leads to a transcriptional Chapter 4: Results 70 activation (Wysocka et al. 2005, Wysocka et al. 2006, Zhang et al. 2012). It has been shown that in Xenopus laevis embryos, knockdown of Wdr5 leads to abnormal expression patterns of developmentally important Hox genes. Furthermore, Wdr5 was shown to be required for the self-renewal of embryonic stem cells (ES cells) and also for the induction of pluripotency during the derivation of induced pluripotent stem cells (iPS cells) (Ang et al. 2011). Therefore, the presence of Wdr5 protein in the sperm, but not in the spermatid nucleus, could be advantageous for the sperm, as Wdr5 protein was directly shown to be involved in the regulation of transcription. Another example of a sperm-specific protein identified in this study, which could also explain the developmental advantage of the sperm is Mlf1. Mlf1 is a transcription factor that has been shown to regulate both gene transcription and cell cycle progression (Winteringham et al. 2004, Yoneda-Kato et al. 2005, Yoneda-Kato & Kato 2008). Therefore, Mlf1 could be important for the early phases of the embryonic development, and could potentially explain the developmental advantage of sperm over the spermatids. Conversely, proteins named Prohibitin and Prohibitin 2 were identified as being present specifically in spermatids. They are highly conserved proteins, with 90% and 88% amino acids identity between Mus musculus and Xenopus laevis, respectively. They were shown to have anti-proliferative functions, but also to be involved in differentiation and morphogenesis (Chowdhury et al. 2013). It is thus also possible that the presence of certain proteins in the spermatids, but not in sperm, such as anti-proliferative Prohibitins, could impair the development of spermatid-derived embryos. I also tested whether any particular types of proteins are enriched in sperm and spermatids. For that I have performed gene ontology (GO) analysis using the online Database for Annotation, Visualization and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov/). Gene ontology analysis indeed indicated that distinct classes of proteins are overrepresented in sperm and spermatids (see Table S2 for terms enriched Chapter 4: Results 71 with a p-value < 0.05). GO analysis for biological processes (BP) for sperm-specific proteins showed a significant enrichment for terms associated with chromatin and nuclear changes, for example nucleosome organisation, nucleosome assembly or DNA packaging. This likely reflects the high degree of sperm nucleus specialisation that occurs during spermiogenesis (incorporation of protamines and global chromatin remodelling). In agreement with that, such terms were not enriched in the list of the spermatid proteins. Among the spermatid proteins the only significantly enriched BP terms were related to cellular homeostasis. Similarly, the BP terms enriched among the proteins present in both cell types were related to basic metabolism, such as glycolysis, oxidation reduction or anion transport. Chapter 4: Results 72 Fig. 10. Experimental design for 2-D Fluorescence Difference Gel Electrophoresis (2-DIGE) analysis (A) The same quantity of proteins isolated from sperm or spermatids was labelled with Cy5 (red) or Cy3 (green) dye. (B) After labelling, proteins were mixed and separated in the first dimension electrophoresis in the pH range, according to isoelectric points of proteins. Subsequently, the proteins were run in the second dimension electrophoresis, according to the molecular weight of proteins. These two runs allowed separation of proteins into spots of three colours: in this example the red spots were sperm-specific; the green ones spermatid- specific and the yellow ones were proteins present in both cell types. Chapter 4: Results 73 Fig. 11. 2-DIGE electrophoresis of proteins isolated from sperm and spermatids Proteins isolated from spermatids and sperm were labelled with Cy3 (green) and Cy5 (red) dyes, respectively, and subsequently separated in two dimensions. Examples of sperm- and spermatid-specific proteins identified are indicated with arrows: red arrows for sperm- specific proteins and green arrows for spermatid-specific proteins. (A) Laser scanned image of a gel with the proteins separated in pH range 7-11 (basic range). (B) Laser scanned image of gel with proteins separated in the pH range 3-10 (broad range). Chapter 4: Results 74 Fig. 12. Silver staining of 2-DIGE gels for spot excision Gels from figure 11 were silver-stained to allow protein visualisation and spot excision. Protein examples from Fig. 11 are also indicated with arrows. (A) proteins separated in pH 7- 11; (B) proteins separated in pH 3-10. Chapter 4: Results 75 4.3. Sperm and spermatids bind distinct egg factors I next tested whether differences in the reprogramming potential between sperm and spermatids could be a result of their differential ability to attract and bind egg factors after fertilisation. One cell embryo contains only one set of maternal and paternal chromosomes, making the analysis of chromatin-bound proteins challenging for two reasons. First, there is an equimolar amount of paternal and maternal chromosomes, which therefore does not allow the analysis of the factors bound only to the paternal chromosomes. Second, just one set of chromosomes per embryo limits the material availability. Therefore, I took advantage of the availability of egg extracts in Xenopus laevis. Extracts from activated eggs can recapitulate the whole first embryonic cell cycle: global protamine to histone exchange in the sperm nucleus, DNA synthesis and chromosome condensation for mitosis (Lemaitre et al. 2001, Gillespie et al. 2012). Incubating large number of sperm/spermatids in extracts prepared from activated egg overcomes the problem of limited amount of material in 1-cell stage embryos and mimics the events happening after fertilisation. Therefore, to test whether sperm and the spermatids attract different egg factors, I incubated permeabilised sperm and spermatids in egg extracts. Subsequently, chromatin and chromatin-bound proteins were isolated, followed by extensive washes to enrich for proteins bound to chromatin (Fig. 13 and Experimental procedures). Proteins were then isolated, labelled with Cy3 and Cy5 dyes and subjected to 2- DIGE analysis. Since only two samples can be simultaneously compared on one gel, and there were four samples to be compared (sperm, sperm-extract treated, spermatid, spermatid- extract treated), I first compared sperm with sperm-extract treated and spermatid with spermatid-extract treated (Fig. 14A and 14B, respectively). Interestingly, it turned out that virtually all the detectable proteins changed after the egg extract treatment (almost no ‘yellow’ spots, see fig. 14A and 14B). This can be explained by two possibilities: 1. the great majority of the donor nuclear proteins are removed (and/or modified post-translationally) Chapter 4: Results 76 upon the incubation with egg extract or 2. egg-derived proteins are in such excess over the donor cell-derived proteins that the donor cell-derived proteins become undetectable after the egg extract treatment. Therefore, I decided to perform a direct comparison of sperm-extract treated sample with the spermatid-extract treated sample to identify egg proteins that are specifically incorporated from extracts to the sperm or spermatid nuclei. 2-DIGE analysis identified numerous protein spots present specifically in the sperm extract-treated or in the spermatids extract-treated sample (Fig. 15A). Selected spots were excised and subjected to mass spectrometry-based identification (Fig. 15B). Mass-spectrometry based analysis of peptides isolated from the selected spots, led to the identification of 107 proteins bound specifically to sperm nuclei, 20 proteins bound to spermatid nuclei and 108 proteins incorporated from the egg into both cell types (Table S3). Fig. 13. Experimental design for mass spectrometry analysis of extract-treated sperm or spermatids Sperm or spermatids are separately treated with egg extracts. Subsequently, sperm or spermatid chromatin is purified and chromatin-bound proteins are isolated. Isolated proteins are then subjected to 2-DIGE and mass spectrometry identification. Chapter 4: Results 77 Fig. 14. 2-DIGE electrophoresis of proteins from sperm and sperm-extract treated and of spermatid and spermatid-extract treated. (A) Proteins isolated from sperm (red) were run on 2-D gel together with proteins bound to the sperm chromatin after egg extract treatment (green). (B) Proteins isolated from spermatid (red) were run on 2-D gel together with proteins bound to the spermatid chromatin after egg extract treatment (green). Note the presence of numerous red or green spots on both gels (A and B) and the low number of yellow spots. The first dimension electrophoresis for both gels shown was carried in the pH 3-10 (broad range). Chapter 4: Results 78 Fig. 15. 2-DIGE electrophoresis of sperm-extract treated and spermatid-extract treated. (A) Proteins bound to sperm chromatin after egg extract treatment (green) were run on a pH 3-10 gel together with proteins bound to spermatid chromatin after egg extract treatment (red). Examples of proteins binding specifically to sperm, spermatid or to both types of cells are indicated with arrows. (B) The same gel as in (A) silver-stained to allow protein spot visualisation and excision. Examples of proteins are indicated with arrows. Chapter 4: Results 79 Among the egg proteins that were bound to both types of nuclei, many proteins involved in global chromatin remodelling, for example, chromodomain helicase DNA binding protein 4 (Chd4) and Imitation Switch (ISWI) were identified. Both Chd4 and ISWI are ATP-dependent chromatin remodelling complexes that move nucleosomes. Nucleosome sliding can change the accessibility of the DNA and plays important roles in multiple biological process, like regulation of transcription mediated by polymerases I-III, DNA replication or DNA repair (Tong et al. 1998, Poot et al. 2005, Erdel & Rippe 2011). ISWI has been already reported to be recruited to chromatin of somatic cell nuclei incubated in egg extracts (Kikyo et al. 2000), thus its identification here validates the experimental setup applied. Another class of proteins abundantly detected in both types of cells after extract treatment are structural proteins that are important for the maintenance of the cell cytoskeleton and of the nucleus and nuclear structures, like actin, nucleoporin or nuclear pore complex proteins Nup98-Nup96 (Fontoura et al. 1999, Fontoura et al. 2001, Enninga et al. 2003, Loiodice et al. 2004), which likely reflect pronuclear formation activities induced by the extract. An example of another highly represented group of proteins incorporated from the egg extract into sperm and spermatid nuclei are proteins directly involved in DNA replication: origin recognition proteins (Orc1), protein involved in unwinding and remodelling of the DNA (topoisomerase I and II, FACT complex subunit Spt16, DNA replication helicase Mcm2), but also proteins involved in stabilising the single stranded DNA, necessary for replication, like Replication protein A (RPA) (Henricksen et al. 1996, Bochkarev et al. 1997, Rowles & Blow 1997, Sible et al. 1998, Wang et al. 2002, VanDemark et al. 2006, Han et al. 2010). Identification of the types of proteins mentioned above confirms that the extracts used were functional and that they were able to recapitulate at least some of the events occurring in the first embryonic cell cycle. Last, a protein named Sal-like protein 4 (Sall4) was also identified as bound to both cell types after egg extract Chapter 4: Results 80 treatment. This finding is a bit surprising, as Sall4 is a transcription factor, and transcription factors are usually not present at very high concentrations. Sall4 was highly abundant among the proteins incorporated into chromatin after the egg extract treatment, as it was identified as the fourth most abundant protein in all the groups analysed. Interestingly, Sall4 has been reported as a master regulator of the core pluripotency network, necessary for the early embryonic development (Elling et al. 2006, Zhang et al. 2006, Tan et al. 2013). This finding could suggest that factors important for the pluripotency in the early embryo are incorporated into the paternal chromatin immediately after fertilisation. Interestingly, some proteins that were identified in spots originating uniquely from extract-treated sperm or from extract-treated spermatids, turned out to be isoforms sharing a high degree of similarity, for example, a protein identified within sperm-extract treated (gi|27881711) was just 3 amino acid different from protein identified in spermatids-extract treated (gi|639691) and both of them were isoforms of High mobility group protein X (HMG- X) (Kinoshita et al. 1994). Another such example is a protein name Hira (histone cell cycle regulation defective homolog A) (Ray-Gallet et al. 2002). One isoform of this protein (Hira- A) was identified as specific for sperm-extract treated (gi|50416397) and another one (Hira) as present in both cell types after egg extract treatment (gi|14330670). There were also some cases in which the same protein was identified separately in sperm- and spermatid-extract treated. For example, Metastasis associated 1-like protein (mta2) (gi|5901733) was identified independently in sperm- and in spermatid-specific protein spots. This is likely a result of post-translational modification of the protein; however, due to the fact that the analysis performed did not discriminate between different post-translational modifications, such proteins were classified as present in both cell types after the egg extract treatment. Chapter 4: Results 81 There were however also egg proteins that were bound exclusively to one chromatin type and not to the other. Many of the protein identified exclusively in sperm extract-treated were structural chromatin proteins, for example core histones: H2A, H2B, H3 and H4. Presence of core histones incorporated into the sperm chromatin likely reflects the remodelling of the paternal chromatin after fertilisation and the exchange of sperm-derived protamine-like proteins (sp1-6) to canonical type of histones derived from the egg. Another class of egg proteins binding specifically to the sperm chromatin are transcriptional repressors, for example heterochromatin protein 1 gamma (HP1γ), methyl-CpG binding domain protein 3 (Mbd3), histone deacetylase 1 and histone deactylase 2 (Hdac1 and Hdac2, respectively). HP1γ was shown to recognise and bind methylated lysine 9 of histone H3 (Lachner et al. 2001). This binding is important for the regulation of gene expression (Kwon & Workman 2011, Smallwood et al. 2012) and also for cell reprogramming to pluripotency (Sridharan et al. 2013). Mbd3 does not recognise post-translational marks on histones, but binds to methylated DNA (Wade et al. 1999). Mbd3 was shown to be necessary for embryonic development in Xenopus (Iwano et al. 2004) and to be a roadblock for reprogramming to pluripotency (Rais et al. 2013). Hdac1 and Hdac2 are enzymes responsible for removal of acetyl marks from histones, which were shown to be involved in transcriptional repression (Laherty et al. 1997, Hassig et al. 1998). Furthermore, both Hdac1 and Hdac2 are involved in DNA replication, for example by stabilising newly formed nucleosomes and also by directly interacting with topoisomerase II (Tsai et al. 2000, Bhaskara et al. 2013). Identification of many repressive egg proteins binding specifically to the sperm chromatin may seem somewhat surprising. However, during the rapid cell cycle phases of early Xenopus development, no transcription is observed (Newport & Kirschner 1982). Therefore, the ability to recruit all the repressive proteins from the egg may reflect Chapter 4: Results 82 programming of sperm to participate in the earliest phases of embryonic development – to support efficient replication and prevent premature transcription. Such a wide variety of chromatin remodelling proteins were not identified in spermatid-extract treated, with the exception of Baf57/Smarce1. Baf57 was shown to be important for cell cycle progression via transcriptional regulation of cell-cycle related genes (Hah et al. 2010). Furthermore, a couple of unique isoforms of structural proteins – tubulin and vimentin were identified as binding specifically to spermatid extract-treated (Table S3). GO analysis of egg proteins binding specifically to sperm, to spermatids and to both cell types confirmed the observations made by looking at examples of proteins (Table S4). First, among the egg proteins incorporated specifically to the sperm chromatin, there was a significant overrepresentation of those belonging to BP (biological process) terms connected with chromosome and chromatin organization, chromatin modification, chromatin assembly and disassembly. Such BP terms were not enriched among proteins incorporated into spermatid-nuclei, and instead terms related to protein polymerisation and protein complex assembly were identified as significantly enriched. As expected, many cell cycle related BP terms were enriched among egg proteins incorporated into both cell types, such as DNA replication, mitosis, cell division, spindle assembly etc. This likely reflects the functionality of the extracts used for the experiments, and their ability to support the events happening during the first embryonic cell cycle. 4.4. Validation of mass spectrometry results by immunobloting. Next I wanted to validate the mass spectrometry approach. One possibility to validate the mass spectrometry results is to perform immunoblotting analysis for candidate, mass- Chapter 4: Results 83 spectrometry-identified proteins. For that I have chosen four proteins, Hdac1, Hdac2, Hp1γ and Mbd3, identified as binding preferentially from the egg to the sperm chromatin. Those particular proteins were chosen due to the availability of antibodies that recognise the Xenopus laevis proteins. Immunoblotting analysis confirmed that those proteins are preferentially incorporated into the sperm chromatin upon egg extract treatment (Fig. 16A and B), therefore validating the use of mass spectrometry approach. Fig. 16. Validation of mass spectrometry results by immunoblotting. (A) Immunoblotting results on proteins isolated from sperm and spermatids after extract treatment. ‘sperm’ - proteins bound to sperm chromatin after egg extract treatment; ‘spermatid’ - proteins bound to spermatid chromatin after egg extract treatment. The antibody used is indicated on the left hand side of the blot inset. The same quantity of proteins was loaded on each lane. (B) Quantification of the blots shown in (A). Results are shown as fold difference between the band intensity in sperm-extract treated to spermatid extract-treated. Names of the proteins are indicated below the x axis. Chapter 4: Results 84 4.5. Summary To conclude, mass spectrometry analysis of sperm, spermatids and of egg extract- treated sperm and spermatids allowed the identification of numerous proteins present specifically in these cell types. Some of these proteins were previously identified as present in these cell types, therefore validating the approach used, for example the presence of Sp1 basic protein in sperm, Rsb-66 in spermatids or the incorporation of ISWI into chromatin after egg extract treatment. On the other hand, interesting novel findings were also made. For example, Wdr5 protein was found as present exclusively in sperm nuclei, which could indicate sperm programming for efficient embryonic development. Also, several proteins were identified as incorporated specifically to the sperm chromatin, for example HP1γ. This may reflect interplay between the sperm chromatin and the egg cytoplasm – unique features of sperm, but not of spermatid chromatin, may allow the binding of specific egg factors. It is also important to note that the approach used here to identify the proteins has some limitations. First, proteins need to be sufficiently abundant in order to be identified by 2-DIGE/mass spectrometry approach. Therefore, less abundant proteins may be missed in this approach. Second, only some of the spots (not all of them) were excised and analysed. This means that not all the proteins that are different between sperm and spermatids were analysed and also, not all the egg proteins preferentially binding to the sperm or spermatid chromatin were identified. Therefore, even though there are many interesting candidate proteins identified in the analysis (see above), one has to remember that this is only a subset of all the proteins changing between these different conditions. Chapter 5: Results 85 Chapter 5 Functional assessment of candidate reprogramming factors 5.1. Introduction Mass spectrometry based approach allowed the identification of several factors which could potentially explain the developmental advantage of sperm over spermatids. My next aim was to functionally validate these factors. To do this, I first selected several sperm- specific factors identified by mass spectrometry, as well as several egg factors specifically incorporated from the egg extract into the sperm chromatin. For the functional validation of the sperm-specific factors I have selected these factors which were abundantly detected in sperm, and which are implicated in the most dramatic nuclear changes between the spermatid and sperm nucleus. These were: sperm-specific basic nuclear proteins Sp1, Sp4 and Sp5 and sperm linker histone variant H1fx. Additionally, I also included a transcription factor Mlf1 for the following reasons. First, it was very abundant specifically in the sperm nucleus, and such high abundance is somewhat surprising for a transcription factor, especially in the transcriptionally silent sperm nucleus. Second, recent proteomic studies of human sperm also identified MLF1 as present specifically in the sperm (Wang et al. 2013), and the conservation of the presence of this protein in the sperm nucleus between Xenopus and human suggests that it could be functionally relevant. Third, microarray analysis identified mRNA encoding an interacting partner of Mlf1, Mlf1IP (Mlf1 Interacting Protein) as present in the oocytes/eggs of three different animal species: mouse, Chapter 5: Results 86 bovine and Xenopus laevis (Vallee et al. 2005). All these suggest that the presence of Mlf1 in the sperm could be important for the embryonic development, perhaps due to interaction of Mlf1 with its oocyte counterpart. I have additionally chosen another sperm factor, Brdt, based on published research indicating that Brdt was necessary during spermiogenesis to remodel the maturing sperm nucleus (Gaucher et al. 2012). Furthermore, chemical inhibition of Brdt caused infertility in mouse, suggesting that the presence of Brdt could be important for the developmental potential of sperm (Matzuk et al. 2012). For the functional validation of egg factors identified as binding specifically to the sperm chromatin I have selected 7 proteins. Four of them were previously reported as transcriptional repressors (Laherty et al. 1997, Hassig et al. 1998, Jiang et al. 2004, du Chene et al. 2007): Hdac1, Hdac2, Mbd3 and Hp1γ and their recruitment to the sperm chromatin at fertilisation could be important to maintain a transcriptionally silent state during the earliest phases of embryonic development, before the onsets of zygotic genome activation. Two other proteins incorporated to the sperm chromatin from the egg extract: Rbbp4 and Rbbp7, were selected for a functional validation due to their presence in repressive complexes with Hdac1 and Hdac2 proteins, but also with Polycomb group proteins (they are both present in Polycomb Repressive Complex 2) and also because of their reported interactions with histones and roles in chromatin assembly (Vermaak et al. 1999, Nicolas et al. 2000, Kuzmichev et al. 2002, Yao & Yang 2003). Lastly, Lsf protein (Late SV40 Protein, also known as Cp2 or Tfcp2) was also selected for a functional validation, as it is an egg-derived transcription factor incorporated specifically to the sperm chromatin and it was reported to have oncogenic properties and to be important for the cell cycle entry and progression (Saxena et al. 2009, Yoo et al. 2010). These therefore suggest that binding of Lsf1 to the sperm chromatin could potentially facilitate the rapid cell cycles of early Xenopus laevis embryos and thus explain the developmental advantage of sperm over spermatids. Chapter 5: Results 87 The functional tests of sperm-specific proteins were performed by a somatic cell nuclear transfer of donor cells ectopically overexpressing these factors, whereas tests of the functional importance of egg-derived factors were attempted by a knockdown of the egg factors. Unfortunately, none of the sperm-specific factors exogenously expressed in the donor cells led to an increase in the efficiency of nuclear transfer. Similarly unsuccessful was the attempt to knock down the egg factors, as even though the strategy used allowed to downregulate the expression measured at the mRNA levels, the protein levels of none of these factors were reproducibly downregulated. 5.2. Functional assessment of sperm-specific proteins 5.2.1. Experimental design High condensation of the sperm nucleus makes it very inaccessible for any technical manipulations, for example for selective protein depletion. In order to test whether the presence of candidate sperm-specific factors makes the sperm better at supporting development, I have therefore chosen to use a different strategy than depleting these factors in the sperm itself and instead I decided to ectopically overexpress these factors in somatic cells and examine the ability of such cells to support embryonic development. To achieve this, mRNA encoding a factor of interest is first injected into a 1-cell stage Xenopus embryo. Such embryo is then allowed to develop and as the embryo develops, the injected mRNA becomes translated into the corresponding protein. Subsequently, when the embryo reaches a late blastula/early gastrula stage it is collected and disaggregated to obtain single cells that Chapter 5: Results 88 overexpress the protein of interest and which can be used as donors for nuclear transfer experiments (Fig. 17). Late blastula/early gastrula stage embryos were used as cell donors, since cells coming from such early embryos proved before to be efficient donors in nuclear transfer experiments (Gurdon 1962). Successfully reconstructed embryos can develop into tadpoles and their developmental potential can be compared to the developmental potential of control reconstructed embryos. I have therefore used this experimental setup to functionally test the candidate sperm-specific factors. Fig. 17. Overexpression of sperm factors in donor cells for a nuclear transfer experiment. To test whether a sperm factor can increase the developmental potential of somatic cells, mRNA encoding the factor is first injected into a 1-cell stage embryo. During embryonic development, the mRNA is translated into protein. The resulting embryo expressing the protein of interest is then disaggregated and the cells, pre-loaded with the factor of interest are used as donors for nuclear transfer experiments. 5.2.2. Cloning and ectopic expression of sperm-specific factors Candidate sperm-specific factors (Sp1, Sp4, Sp5, H1fx and Mlf1) were first cloned from cDNA into pCS2 vectors that allow in vitro mRNA synthesis. They were additionally tagged with a hemagglutinin tag (HA-tag) to allow monitoring of protein expression in the absence of available antibodies. The construct encoding Brdt was tagged with a green fluorescent protein (GFP) and was a kind gift of Dr Saadi Khochbin and was sub-cloned into pCS2 vectors. At this stage, pre-testing of some of the constructs (Sp1, Sp4, Sp5 and Brdt) was performed by transiently transfecting cultured cells and checking whether the proteins Chapter 5: Results 89 are targeted to the nucleus, since the nuclear localisation of these proteins was reported previously (Risley & Eckhardt 1981, Abe & Hiyoshi 1991, Pivot-Pajot et al. 2003). Microscopic observations of the transfected cells confirmed that all the tested proteins localised to the nuclei (Fig. 18A - D). Interestingly, ectopic expression of Brdt combined with TSA (Trichostatin A) treatment led to chromatin compaction (Fig. 19). TSA is an inhibitor of histone deacetylases, and therefore TSA treatment leads to an increase of histone acetylation levels (Yoshida et al. 1990). It was reported previously that Brdt protein can recognise and bind to acetylated histones via its bromodomains and that this leads to chromatin compaction (Pivot-Pajot et al. 2003, Govin et al. 2006, Moriniere et al. 2009). Therefore, my observation that expression of Brdt combined with TSA treatment led to the chromatin compaction validates the functionality of Brdt exogenously expressed in the transfected cells. Subsequently, mRNAs encoding the candidate proteins were in vitro transcribed. The size and the purity of the synthesised mRNA was confirmed by an agarose gel electrophoresis and mRNAs were subsequently injected into 1-cell stage embryos to test whether they can be efficiently translated into proteins. Expression of Sp1, Sp4, Sp5, H1fx and Mlf1 was tested by immunoblotting (staining against the HA tag of the proteins), whereas the expression of Brdt was assessed by microscopic observations of whole embryos (Brdt protein was tagged with GFP). All the mRNAs tested allowed efficient protein synthesis in the embryos (Fig. 20). 5.2.3. Validation of UV treatment length required for the nuclear transfer procedure Since the embryonic cells used as donors for the nuclear transfer procedure are diploid, the recipient egg needs to be enucleated to allow the development of a diploid embryo. Enucleation is performed by first placing the eggs on a small piece of a blotting paper soaked in water and mounted on a microscope slide. The eggs are oriented with their Chapter 5: Results 90 animal poles and the white spot (which indicates the position of the meiotic spindle and egg chromosomes) upwards. Subsequently, the eggs are placed for 30s under a Mineralite UV lamp for enucleation and finally, for 3-6s under a Hanovia UV lamp for dejellination (Gurdon 1962) (Fig. 21). The latter treatment is used to soften the jelly that coats the egg and to make it penetrable by the injection needle. The length of the Hanovia lamp treatment has to be optimized every time a new batch of eggs is used. In general, too short a treatment does not allow the insertion of a needle into the egg, whereas too long a treatment is detrimental for the development of embryos (Fig. 22), therefore each time the nuclear transfer procedure is performed, the researcher has to determine the shortest length of Hanovia lamp treatment that allows the needle to penetrate the egg. Next I validated whether 30s treatment of eggs with Mineralite UV lamp, which has been used in the nuclear transfer procedure in the past (Gurdon 1962), is sufficient to enucleate the eggs. If the eggs are successfully enucleated, the resulting embryos should be haploid (since the sperm that fertilises the egg is haploid and the maternal genetic content is destroyed). Haploid embryos in Xenopus laevis are viable, but differ from the diploid ones morphologically: haploid tadpoles are more vegetalised (shorter and thicker) than the diploid ones. It is also possible to assess the ploidy of the embryo by squashing its cells on a microscope slide and looking at their nuclei with a phase contrast microscope. In Xenopus laevis each set of parental chromosomes give rise to one nucleolus, visible as a black dot in the nucleus. Therefore, diploid cells have two nucleoli in each nucleus (two black dots), whereas haploid cell have just one nucleolus in each nucleus (one black dot). To test whether 30s treatment with Mineralite UV lamp is sufficient to enucleate the eggs, I first enucleated them as described above (Fig. 21) and then fertilised them. The control embryos were fertilised without enucleation. Morphological observations of the embryos revealed that the Mineralite-treated eggs gave rise to haploid embryos (they were vegetalised as compared to Chapter 5: Results 91 control, diploid embryos) (Fig. 23A). Furthermore, microscopic observations of squashed cells also confirmed that embryos obtained from Mineralite-treated eggs were haploid (evidenced by the presence of only one nucleolus/nucleus) (Fig. 23B). Fig. 18. Nuclear localisation of Sp1, Sp4, Sp5 and Brdt in transfected C2C12 cells. C2C12 myoblast cells were transfected with plasmids encoding selected sperm factors: Sp1 (A), Sp4 (B), Sp5 (C) and Brdt (D). 48h after transfection cells were fixed and subjected to immunostaining revealing DNA (staining with DAPI, left column) and the overexpressed sperm factor (middle panel). Merge images (right column) show that the overexpressed proteins (green) localise to the nuclei (blue). Scale bars = 10um. Chapter 5: Results 92 Fig. 19. Brdt-dependent compaction of chromatin upon TSA treatment. C2C12 myoblast cells were transfected with a plasmid encoding Brdt and treated with 120ng/ml TSA or with an equivalent concentration of DMSO (control). 48h after transfection cells were fixed and subjected to immunostaining revealing DNA (staining with DAPI, left column) and Brdt (middle panel). Merge images (right column) show that control, DMSO- treated cells do not compact chromatin upon Brdt overexpression (A), but cells treated with TSA and overexpressing Brdt, do compact the chromatin (B). Scale bars = 10um. Chapter 5: Results 93 Fig. 20. mRNA injection into 1-cell stage embryos allows protein synthesis. One cell stage embryos were injected with mRNAs encoding various sperm-specific proteins. At gastrula stage the embryos were either collected for immunoblotting analysis against the HA-tag (A) or were photographed under the microscope equipped with a fluorescent lamp (B). (A) Immunoblotting analysis confirmed that all proteins encoded by the injected mRNAs (Sp1, Sp4, Sp5, H1fx and Mlf1) were translated in the embryos. Note that the predicted molecular weights are slightly different from the observed molecular weights of proteins, which could be due to their post-translational modifications. (B) Microscopic observations under fluorescent light of embryos injected with mRNA encoding Brdt-GFP, revealed that the protein is expressed in the embryos (merge image). Scale bar = 1mm. Chapter 5: Results 94 Fig. 21. Diagram explaining recipient egg preparation for the nuclear transfer procedure. Eggs are first immobilised with the animal pole (white spot) facing upwards on a wet blotting paper. Subsequently, eggs are enucleated with a 30s treatment with a Mineralite lamp UV light and then the jelly coat of the eggs is softened by a 3-6s treatment with a Hanovia lamp UV light. Fig. 22. Prolonged Hanovia lamp treatment leads to abnormal development. Control fertilised embryos were treated with a Hanovia lamp for 4s - the shortest time allowing the needle penetration (A) or for a prolonged time – 8s (B). Note that a prolonged treatment with Hanovia lamp results in abnormal development of the embryos. Chapter 5: Results 95 Fig. 23. Mineralite UV lamp treatment for 30s destroys the genetic material of the egg. Eggs were subjected to 30s Mineralite UV lamp treatment for enucleation and subsequently fertilised. Control eggs were directly fertilised, omitting the Mineralite treatment. (A) Microscopic observations of control embryos (non-enucleated) (left panel, ‘diploid embryos’) and of embryos obtained from Mineralite-treated eggs (enucleated) (right panel, ‘haploid, enucleated embryos’) reveals that haploid embryos are vegetalised and therefore confirms the successful egg enucleation. Note that the vegetalised phenotype is not apparent at the gastrula stage (upper panel) and only becomes visible when the embryo elongates (bottom panel). Scale bars = 1mm. (B) Microscopic observations of cells from control embryos (left panel) and from embryos obtained from Mineralite-treated eggs (right panel) confirms a successful egg enucleation. Note that in the nuclei of cells from diploid embryos two nucleoli (black dots) are visible (left panel), whereas in the nuclei of cells from haploid embryos only one nucleolus per nucleus can be detected (single black dots). Examples of nuclei in each cell preparation are inside the red circles. Chapter 5: Results 96 5.2.4. Overexpression of candidate sperm factors does not increase the efficiency of nuclear transfer. I next tested the effect of overexpression of candidate sperm factors in the donor cells on the efficiency of nuclear transfer. In this experimental setup the developmental capacity of embryos reconstructed with cells overexpressing the factor of interest was compared with the developmental capacity of embryos reconstructed with control cells, which did not overexpress the sperm factor. Reconstructed embryos were scored as the number of swimming tadpoles obtained to the total number of cleaved embryos. The reasoning behind this experimental design was that if the presence of some sperm-specific factors is beneficial for the sperm to support the embryonic development, its ectopic expression in a donor cell should increase the efficiency of the nuclear transfer. Some of the candidate sperm factors were therefore expressed as single factors: Brdt, Mlf1, H1fx. A combination of factors was also used for those which have similar functions: Sp1, Sp4 and Sp5 or Sp1, Sp4, Sp5 and H1fx altogether (as those factors are structuring the chromatin in the mature sperm). These experiments, which initially gave promising outcomes (tendency of sperm factor- overexpressing cells to support higher efficiency of nuclear transfer than control cells) were independently repeated to validate whether initially promising outcomes are reproducible. Unfortunately, none of the sperm factors tested, alone or in combinations, reproducibly increased the efficiency of the nuclear transfer. Even those factors which initially gave a promising outcome (Brdt and the mixture of Sp1, Sp4, Sp5 and H1fx) did not show reproducible effects in the following experiments (Table 5). Chapter 5: Results 97 Table 5. Summary of nuclear transfer experiments using cells overexpressing candidate sperm factors. Factor/combination of factors tested Number of embryos tested: total number of swimming tadpoles obtained (ST)/cleaved embryos (CE) Control cells (% ST/CE) Overexpressing cells (% ST/CE) Brdt – experiment 1 2/42 (4.8%) 3/46 (6.5%) Brdt – experiment 2 10/36 (27.8%) 8/31 (25.8%) Brdt – experiment 3 5/48 (10.4%) 1/49 (2.0%) Brdt – experiment 4 7/61 (11.5%) 1/30 (3.3%) Brdt – total from 4 experiments 24/187 (12.9%) 13/156 (8.3%) Mlf1 6/35 (17.1%) 5/42 (11.9%) H1fx 2/18 (11.1%) 2/19 (10.5%) Sp1, Sp4, Sp5 3/31 (9.7%) 2/28 (7.1%) Sp1, Sp4, Sp5 and H1fx – experiment 1 0/44 (0%) 4/44 (9.1%) Sp1, Sp4, Sp5 and H1fx – experiment 2 1/20 (5%) 6/27 (22.2%) Sp1, Sp4, Sp5 and H1fx – experiment 3 2/21 (9.5%) 1/22 (4.5%) Sp1, Sp4, Sp5 and H1fx – experiment 4 3/26 (11.5%) 2/38 (5.3%) Sp1, Sp4, Sp5 and H1fx – total from 4 experiments 6/111 (5.4%) 13/131 (9.9%) Chapter 5: Results 98 5.3. Functional assessment of egg factors preferentially associating with the sperm chromatin 5.3.1. Experimental design In order to test whether egg factors preferentially associating with the sperm chromatin have a function in early development, one would ideally remove such factors from the early embryos. The fact that the egg is a much more accessible cell for any type of manipulations than sperm makes it possible to try to downregulate the selected factors. The best characterised way of downregulating proteins in Xenopus laevis oocytes is by the injection of antisense deoxy-oligonucleotides (Hulstrand et al. 2010). In this approach, oligonucleotides complimentary to mRNA encoding the protein of interest are designed and injected into the GV stage oocyte. Such oligonucleotides form DNA-RNA heteroduplexes with the target mRNA, which are recognised and cleaved by endogenous oocyte-derived RNase-H activity. Cleaved mRNAs are subsequently degraded by oocyte-derived exonucleases. If the protein of interest is sufficiently unstable, then downregulation of the mRNA can lead to the reduction of the desired protein level (Fig. 24A). Such oocytes in which the protein is downregulated can be subsequently in vitro matured to eggs and injected with sperm (in ICSI procedure), which allows the assessment of the effects of the protein downregulation on embryonic development (Fig. 24B). Here I aimed to use this approach to assess the developmental function of egg-derived candidate factors binding specifically to the sperm chromatin after the egg extract treatment. Chapter 5: Results 99 Fig. 24. Diagram explaining oligonucleotide-mediated knockdown of proteins in Xenopus laevis oocytes. (A) Antisense deoxy-oligonucleotides against mRNA encoding the protein X are injected into the GV stage oocyte. These oligonucleotides form DNA-RNA heteroduplexes with the mRNA in the region of the base pair complementarity. DNA-RNA heteroduplexes are recognised and cleaved by endogenous, oocyte-derived RNase-H. The cleaved mRNA is then degraded by endogenous exonucleases. In the absence of mRNA, protein X becomes degraded. (B) GV stage oocytes depleted for protein X can be in vitro matured into eggs and injected with sperm (by ICSI procedure) to generate embryos depleted for the protein X. The effect of protein X depletion on embryonic development can be subsequently assessed. Chapter 5: Results 100 5.3.2. Validation of the antibodies I have first tested whether the proteins I want to downregulate (Hdac1, Hdac2, Rbbp4, Rbbp7, Mbd3, Hp1γ and Lsf) are detectable with commercially available antibodies. To do this I have treated sperm with egg extracts to test whether bands of correct sizes are detected. All antibodies recognised bands of approximately the expected size (Fig. 25), therefore validating their use for assessing the knockdown efficiency of the oligonucleotides. Next, I tested whether the proteins I want to downregulate are also detectable by immunoblotting in the oocytes, as the knockdown itself is performed in the oocytes. All proteins apart from Mbd3 and Lsf were detected in the oocyte lysates. Due to the fact that the oocyte is pre- loaded with a lot of proteins, it is difficult to load more than one oocyte per a gel lane. The ability to detect the protein in the oocyte is crucial for the assessment of knockdown efficiency. Therefore, I tested whether I could detect these proteins if instead of the whole oocyte lysate I would use germinal vesicles (nuclei) isolated from the oocytes. The advantage of using the nuclei is that one can load many nuclei per one lane and in this way focus the analysis on nuclear proteins. I have therefore tried to detect Mbd3 and Lsf proteins in lysates from 20 germinal vesicles. Unfortunately, even the use of this approach did not allow me to detect the proteins of the correct size (Fig. 26). The anti-Mbd3 antibody recognised a band of around 45kDa instead of 33kDa. This antibody was reported to also recognise Mbd2, which was reported to migrate at 45kDa (Zhu et al. 2011), therefore the observed band could be Mbd2 and not Mbd3. Alternatively; the 45kDa band could be a post-translationally modified form of Mbd3. For Lsf protein a band of around 35kDa was detected instead of expected 57kDa. This again could be the effect of post-translational clipping of the protein in the oocyte and not in the egg, or of unspecific recognition of another protein by anti-Lsf antibody. Due to the uncertainty about the detection of Mbd3 and Lsf proteins in the oocyte, which could impede the validation of the knockdown effect, I have decided to exclude these Chapter 5: Results 101 two proteins from the list of candidate egg factors to be downregulated and for the further steps I focused on the downregulation of the five remaining proteins: Hdac1, Hdac2, Hp1γ, Rbbp4 and Rbbp7. 5.3.3. Downregulation of the mRNA encoding the selected factors I next tested whether deoxy-oligonucleotides injected into the oocytes can degrade the mRNAs encoding the target proteins. I have designed antisense deoxy-oligonucleotides complimentary to Hdac1, Hdac2, Hp1γ, Rbbp4 and Rbbp7, following the guidelines described before (Hulstrand et al. 2010). Three different oligonucleotides targeting each of the mRNAs were designed, whereas scrambled oligonucleotides were designed for control experiments. Oligonucleotides were injected into GV stage oocytes. Injected oocytes were collected 48h after the oligonucleotide injection and processed for reverse-transcription quantitative PCR analysis (RT-qPCR). Amount of transcripts present in the oligonucleotide- injected oocytes were normalised to the amount of transcript present in the control, scrambled-oligonucleotide injected oocytes. qPCR results showed that all the oligonucleotides allowed a significant downregulation of the target mRNAs (Fig. 27). Subsequently, the most efficient oligonucleotides at degrading the mRNAs (oligonucleotide 3 for Hdac1, Hdac2 and Rbbp7, oligonucleotide 1 for Hp1γ and oligonucleotide 2 for Rbbp4), were chosen to assay the degradation of the target proteins. Chapter 5: Results 102 Fig. 25. Validation of the antibodies on sperm-extract treated samples. (A) Immunoblotting for Hdac1, Hdac2, Hp1γ (Hp1g), Rbbp7, Rbbp4, Lsf and Mbd3. ‘S-E’ indicates the position of sperm-extract treated sample and the arrow indicates the protein band of interest. (B) Table presenting the expected molecular weights of the proteins tested. Note that the protein bands detected in (A) are of similar weight to the expected molecular weight (B), suggesting that the antibodies recognised the correct proteins. Chapter 5: Results 103 Fig. 26. Mbd3 and Lsf proteins are not detected in the oocyte. Proteins isolated from 20 germinal vesicles (oocyte nuclei) were loaded on each gel. Subsequently gels were stained for Mbd3 (left panel, green) and for Lsf (right panel, red). Detected proteins were not migrating at the expected molecular weights (observed before in the lysate from egg extract-treated sperm chromatin): Mbd3 should be detected at 33kDa (not at 45kDa) and the expected molecular weight of Lsf is 57kDa (not 35kDa). Chapter 5: Results 104 Fig. 27. Assessment of mRNA degradation by oligonucleotide-mediated knockdown. Oocytes were injected either with scrambled oligonucleotides (control) or with three different oligonucleotides (oligo1, oligo2, oligo3) designed to target the mRNA of interest (name of the target mRNA indicated above the graph). All samples were normalised against a housekeeping mRNA pwp1 and the amount of mRNA in the control sample was set to 100% (the other samples were normalised accordingly). All oligonucleotides tested led to a significant downregulation of the mRNA levels (p < 0.05, t-tests). Error bars show ±SEM. Chapter 5: Results 105 5.3.4. Protein levels of the candidate egg factors are not reproducibly downregulated by the antisense oligonucleotides I have subsequently tested whether the injection of the most efficient oligonucleotide affects the target protein levels in the oocyte. For each candidate factor oocytes were injected either with the oligonucleotide against the mRNA or with the control scrambled oligonucleotide. Oocytes were collected for protein extraction and immunoblotting analysis 48h, 96h and 144h after the injection. Only the injection of the oligonucleotide against Hp1γ led to a downregulation of the protein level in the oocyte, whereas the expression level of other proteins tested was not affected (Fig. 28A). I have therefore repeated the experiments with the oligonucleotide against Hp1γ. Unfortunately; the initial downregulation of the protein level observed after injection of the oligonucleotide 1 was not reproduced in independently repeated experiments. One of the reasons for that could be a sequence polymorphism between different frogs (animals in our frog colony are not from an inbred line and can therefore have slightly different DNA sequence which could prevent targeting by the oligonucleotide). I have therefore tried to inject the two other oligonucleotides against Hp1γ: oligonucleotide 2 and 3; which however did not result in a protein knockdown (Fig. 28B). I have then also tried a higher dose of the oligonucleotide 1: 2 times more than initially, the same amount and 2 times less; however, none of the concentrations led to the protein downregulation (Fig. 28C). The lack of knockdown could have been also caused by the fact that the oocytes used in the initial experiments were fully grown and therefore already ceased RNA transcription, whereas the oocytes used for the repeat experiments were still transcribing mRNAs and therefore the knockdown was not efficient. Alternatively, oligonucleotides could have degraded during the storage period and could have become less efficient at the mRNA degradation. To shed light on the discrepancy in the results obtained, I have treated non-injected oocytes with cycloheximide, which inhibits protein translation. Chapter 5: Results 106 Treatment with cycloheximide therefore allows monitoring of the half-life of proteins. Immunoblotting of proteins from cycloheximide-treated oocytes revealed that the level of Hp1γ protein was not reduced during 48h of cycloheximide treatment (Fig. 28D) (longer treatment with cycloheximide is toxic to the oocytes and therefore cannot be applied). This result is contradictory to the initial result that demonstrated a knockdown of Hp1γ already after 48h from the oligonucleotide injection. The reason for the discrepancy between the results remains therefore unclear; however, since all the follow-up experiments suggest no knockdown of Hp1γ, it is unlikely that the oligonucleotide-mediated route can lead to a reproducible depletion of this protein. Chapter 5: Results 107 Fig. 28. Assessment of protein degradation by oligonucleotide-mediated knockdown. (A) Oocytes were injected either with scrambled oligonucleotides (s) or with oligonucleotides against the mRNA of interest to achieve a protein knockdown (k) (indicated below the lanes). Oocytes were collected 48h (2d), 96h (4d) or 144h (6d) after injection for immunoblotting analysis. The protein detected on each blot is indicated above the blot. Red rectangle in the Hp1γ blot indicates the position of the Hp1γ protein band. (B) Assessment of Hp1γ level upon injection of various oligonucleotides. Oocytes were injected with three different oligonucleotides against Hp1γ (lanes 1, 2, 3) or with scrambled oligonucleotides (s). Samples were collected 48h after injection. Red rectangle indicates the position of the Hp1γ protein band. (C) Assessment of Hp1γ level upon injection of various doses of oligonucleotide 1. Oocytes were injected with three different doses of oligonucleotide 1 against Hp1γ mRNA (2x, 1x and 0.5x times of the amount injected in panel A) or with scrambled oligonucleotides (s). Samples were collected 48h after injection. Red rectangle indicates the position of the Hp1γ protein band. (D). Assessment of Hp1γ level upon cycloheximide treatment of oocytes. Cycloheximide-treated oocytes are labelled with ‘cyc’ below the lane, whereas the control oocytes, are labelled ‘ctrl’. Samples were collected 48h after the addition of cycloheximide to the media. Red rectangle indicates the position of the Hp1γ protein band. Chapter 5: Results 108 5.4. Summary and discussion In this section I described the effect of the overexpression of candidate sperm factors on the efficiency of nuclear transfer and my attempts to downregulate the egg factors that bind specifically to the sperm chromatin upon egg extract treatment. In the first part I have tested the effect of several candidate sperm factors and several combinations of sperm factors on the efficiency of nuclear transfer. However, none of the factors/combinations of factors led to a reproducible improvement of the nuclear transfer efficiency. There could be several explanations for that, for example: wrong factors were selected, essential co-factors were not co-expressed with the candidate factors, over- expressed factors were not functional or they were expressed at inappropriate levels. Also, it is likely that introduction of sperm factors could have rescued some aspects of the development, but not all. In order to address this in the future a more detailed molecular analysis of the nuclear transfer embryos (for example, qPCR analysis for candidate gene expression) would be needed. I have also attempted to downregulate several candidate factors which were identified as binding specifically to the sperm chromatin upon egg extract treatment. Knockdown was performed with the use of oligonucleotides antisense to mRNA encoding the protein of interest. All the oligonucleotides tested led to a significant downregulation of mRNA targets, as evidenced by a qPCR. However, none of the candidates tested was reproducibly downregulated at the protein level. This suggests that even though mRNA levels decreased, the target proteins were stable and did not degrade. In the future it would be worth performing a simple test with cycloheximide treatment before choosing the candidate Chapter 5: Results 109 proteins to be downregulated. In this way one could eliminate those proteins, whose levels are unlikely to be affected by the oligonucleotide-mediated knockdown. Chapter 6: Results 110 Chapter 6 Characterisation of the developmental defects of spermatid-derived embryos Replication assays described in this chapter were performed in collaboration with Dr Vincent Gaggioli. Preparation of libraries and sequencing of two RNA-seq samples was done in collaboration with Dr Taejoon Kwon and Dr Edward Marcotte. Bioinformatic analyses described in this chapter were performed by Dr Angela Simeone, Dr George Allen and Dr Charles Bradshaw. 6.1. Introduction Comparison of the developmental potential of sperm- and spermatid-derived embryos revealed that sperm-derived embryos develop significantly better than spermatid-derived embryos to a swimming tadpole stage. Since the mass spectrometry analysis and further functional testing of sperm proteins and of sperm-binding factors failed to unravel the source of the developmental advantage of sperm, I modified my strategy. Instead of directly looking for potential factors conferring developmental benefits to sperm, I have decided to first characterise the developmental defects of spermatid-derived embryos. Understanding what processes occur abnormally in spermatid-derived embryos could help to identify in what respects sperm is better at supporting embryonic development. Chapter 6: Results 111 Early Xenopus embryo development starts with rapid cell divisions (fast DNA replication cycles) in the absence of transcription from the zygotic genome. Only around the 12 th cell cycle division, at the time called mid-blastula transition (MBT), cell divisions slow down and the zygotic genome activation occurs (Newport & Kirschner 1982, Kimelman et al. 1987) (Fig. 29). I have therefore tested whether these major events during embryonic development: DNA replication accompanying cell divisions and gene transcription needed for further embryonic development occur normally in spermatid-derived embryos. My results demonstrate that spermatids are equally good as sperm at supporting DNA replication. Furthermore, I showed that developmental failure of spermatid-derived embryos is not due to RNA carried over to the embryo. I also showed that zygotic rRNA transcription is initiated normally in spermatid-derived embryos. Interestingly, RNA sequencing (RNA-seq) analysis of spermatid- and of sperm-derived embryos revealed that one hundred developmentally- important mRNAs are misregulated in spermatid-derived embryos, which can reflect developmental programming of sperm to support correct regulation of gene expression in the embryo. Fig. 29. Rapid cell divisions precede zygotic genome activation in Xenopus. In the very early stages of embryonic development in Xenopus rapid cell divisions (fast DNA replication cycles) occur in the absence of zygotic transcription. Only when the embryo reaches the mid-blastula transition stage, the zygotic genome is activated, which is concurrent with slowing down of cell divisions. Chapter 6: Results 112 6.2. Spermatids replicate their DNA as efficiently as sperm All replication studies described below were performed with great help and supervision from Dr Vincent Gaggioli from the Gurdon Institute, at the University of Cambridge, UK. Cell divisions occurred normally in spermatid-derived embryos (Fig. 8); however, it is not known whether accompanying DNA replication was also normal. Experiments in which sperm and somatic cells (erythrocytes) were incubated in Xenopus egg extracts showed that the sperm chromatin was better suited for supporting DNA replication that the chromatin of somatic cells (Lemaitre et al. 2005). DNA replication starts from the origins of replication and spreads in both directions from the origins (Fig. 30A). The first cell cycle in Xenopus lasts for only about 2h (at 18ºC). During this 2h the sperm needs to remodel its chromatin (exchange the protamines for histones) and initiate and timely complete DNA replication before the onsets of the cell division. It was shown that DNA replication in the sperm chromatin is initiated from multiple origins of replication spaced on average every 23.4kb of DNA. This was in sharp contrast to what was seen in erythrocyte chromatin in which the origins of replication were much sparser and positioned on average every 120.9kb of DNA (Lemaitre et al. 2005). This could prevent the erythrocytes from completing the replication before the onset of the cell division (Fig. 30B). Indeed, it was shown that after 2h of egg extract treatment, erythrocytes replicated less than 10% of DNA, whereas 80% of sperm DNA was replicated in this time (Lemaitre et al. 2005). This led to the hypothesis that inefficient replication is a roadblock to embryo development after somatic cell nuclear transfer in frogs (Laskey 2005). It was also shown that pre-treatment of mammalian donor Chapter 6: Results 113 cells with Xenopus egg extracts increased the efficiency of mouse nuclear transfer and that it was correlated with an increased replication efficiency (Ganier et al. 2011). I have therefore hypothesised that the developmental failure of spermatid-derived embryos might be related to inefficient DNA replication. I have therefore decided to test the ability of sperm and spermatids to undergo replication with the use of in vitro Xenopus egg extracts. Fig. 30. Multiple origins of replication allow timely finishing of DNA replication in egg extracts. Figure based on results from (Lemaitre et al. 2005). (A) DNA replication occurs in both directions from the origin of replication. (B) Sperm has multiple origins of replication, densely positioned on the DNA fiber (left panel), whereas the erythrocyte has sparse origins of replication (right panel). Multiple origins of replication ensure timely finishing of replication after the egg extract treatment in sperm, but not in erythrocytes. To be able to assess the replication efficiency and at the same time to look into the initiation of replication I decided to use the technique of molecular combing. Help with this technique was kindly provided by Dr Vincent Gaggioli. I have prepared extracts from activated eggs, which are able to replicate DNA (Blow & Laskey 1986). Permeabilised sperm or spermatids nuclei were incubated in extracts at a concentration of 200 nuclei per a microliter of extract. Extracts were supplemented with biotin-dUTP, which is a DNA precursor incorporated into DNA upon replication (Fig. 31A). Newly synthesised (replicated) Chapter 6: Results 114 DNA can be then revealed by biotin detection with fluorescently tagged streptavidin. In molecular combing procedure DNA fibers are isolated, spread on a microscope slide and then stained for DNA, to visualise all DNA fibers, and for biotin, to reveal those fibers/fragments of fibers that were replicated (Fig. 31B). Fig. 31. Diagram explaining the experimental design for the replication assessment in sperm and spermatids. (A) Permeabilised sperm or spermatids are incubated in egg extracts supplemented with biotin-dUTP. Subsequently, replicated DNA is revealed by molecular combing. (B) Diagram explaining the results of molecular combing. DNA fibers are stretched on the microscope slide and revealed with anti-DNA staining (shown in green). Replicated DNA is revealed with anti-biotin staining (shown in red). Replicated DNA in the merged image is shown in yellow. Chapter 6: Results 115 The initial experiment was designed to get an idea as to whether the cells can replicate and how they initiate replication (what is the spacing of the origins of replication in each cell type). For that I have incubated sperm or spermatids in freshly prepared egg extracts, supplemented with biotin-dUTP, for short time periods: 30mins and 40mins. Then the DNA fibers were isolated and subjected to molecular combing and immunostaining. Interestingly, immunostaining results revealed that both samples were able to initiate replication with very similar spacing of origins of replication; however, initiation of replication in sperm was approximately 10mins delayed as compared with spermatids. At 30mins from the start of extract treatment some fibers in spermatids already started to initiate replication (Fig. 32). At the same time the replication was not yet detectable in sperm. At 40mins most of the fibers in spermatids were replicating, whereas only some fibers in the sperm sample initiated replication (Fig. 32). This suggests that sperm is not better than spermatids at initiating DNA replication. If anything, the sperm is delayed as compared with spermatids, which is likely due to the fact that it has to remodel its chromatin (replace protamines for egg-derived histones) before the onset of replication. No major differences between sperm and spermatids were found at the initiation of replication, which however does not mean that both cell types are capable of timely completing the replication. It could be that the speed of the replication fork is different in both cell types (for example due to differences in the chromatin structure, which could interfere with the replication progress). If that was the case it could be that even though the cells started the replication similarly, they would not be able to complete replication at the same time. I have therefore incubated sperm or spermatids in egg extracts, supplemented with biotin-dUTP) for 2h – the time which is equivalent to the length of the first embryonic cell cycle. Also, not to miss any potential differences between the samples, this time I have precisely quantified the extent of replication. Quantification of the replication extent was Chapter 6: Results 116 performed by measuring with ImageJ the length of replicated DNA in each fiber (staining anti-biotin) to the total length of the fiber (staining against DNA). Measurements were performed on at least 125 independent DNA fibres (22000kb of DNA for each sample) (Fig. 33A). Data acquired from all the measurements were then exported to Microsoft Excel for calculations with the use of a macro created and shared by Dr Vincent Gaggioli. At 120mins from the start of the egg extract treatment both cell types replicated more than 80% of the total DNA length. There was no difference in the replication extent between sperm and spermatid fibers, suggesting that replication problems are unlikely to be the explanation of the developmental defects of spermatid-derived embryos (Fig. 33B). Chapter 6: Results 117 Fig. 32. Spermatids initiate replication earlier than sperm and show similar spacing of origin of replication to that of sperm. Sperm or spermatids were incubated in egg extracts supplemented with biotin-dUTP for 30 or for 40mins. Antibody staining against DNA reveals the total length of the fiber (green) and antibody staining against biotin reveals the replicated DNA (red). Replicated regions of the fibers are yellow in the merged images. (A) Examples of fibers isolated from the sperm sample. (B) Examples of fibers isolated from the spermatid sample. Note that spermatids initiate replication earlier than sperm (30mins for spermatids and 40mins for sperm) and that the spacing of origins is similar in both samples. Chapter 6: Results 118 Fig. 33. Spermatids replicate DNA as efficiently as sperm (A) Examples of DNA fibers after immunostaining procedure. Antibody staining against DNA reveals the total length of the fibre (green) and antibody staining against biotin reveals the replicated DNA (red). The bottom panels show representative examples of replication staining from sperm and from spermatids incubated in egg extracts. (B) Replication extent measured as the proportion of DNA that incorporated biotin-dUTP to the total fiber length. Results are from at least 125 independent DNA fibers (22000kb of DNA for each sample). Error bars show ±SEM. Samples were not significantly different (p-value = 0.37, t-test). Panel ‘A’ of this figure was created and kindly shared by Dr Vincent Gaggioli. Chapter 6: Results 119 6.3. Spermatid-derived RNA is not deleterious for embryonic development Since my results showed that problems with DNA replication are unlikely to be the cause of developmental defects of spermatid-derived embryos, I next investigated other possible reasons which could explain their defects. I first tested whether carried-over RNA can be a problem for development of spermatid-derived embryos. All the RNAs present in sperm are also present in spermatids; however, during spermiogenesis the maturing spermatozoon reduces its cytoplasmic contents together with the vast majority of RNAs. It has been estimated that mature sperm contains only about 10-100fg of RNA compared to 10- 50pg of RNA typically found in a somatic cell (Pessot et al. 1989, Krawetz 2005), which is about 1000 times reduction in the RNA content. Since there is no transcription in the nucleus of the mature sperm, spermatids contain all RNAs which are required for sperm maturation, for example mRNAs encoding sperm basic proteins 1-6, which are the functional equivalents of mammalian protamines in Xenopus laevis (Abe & Hiyoshi 1991, Hiyoshi et al. 1991). Translation of these mRNAs into proteins and their further incorporation into chromatin is thought to be important for the acquisition of the highly specialised, almost crystalline structure of the sperm nucleus. One could imagine that delivery of all these spermiogenesis- specific RNAs by the spermatid to the egg at fertilisation could lead to the illegitimate translation of spermiogenesis-specific mRNAs in the embryo after fertilisation. Presence of such translated proteins could then interfere with the embryonic development, for example by altering the chromatin architecture of the spermatid-derived embryos. Interestingly, when mRNAs encoding the sperm-specific factors Sp4 or Sp5 were injected into 1-cell stage embryos at high concentrations (9.2ng of mRNA per embryo), embryos died around the gastrulation stage, suggesting that illegitimate expression of spermiogenesis-related proteins can indeed be deleterious for embryonic development (Fig. 34). Chapter 6: Results 120 I have therefore tested whether the potential carried-over RNA from spermatids can have any effects on embryonic development. I have injected fertilised embryos (at 1-cell stage) with 50pg (corresponding to the maximum amount of RNA found in a typical somatic cell) of either total RNA isolated from testis (isolated with Trizol, therefore recovering all different RNA types) or with 50pg of the mRNA encoding the mixture of sperm basic proteins (Sp1, Sp4 and Sp5 – mRNAs which were shown to be toxic for embryonic development when injected at high doses). None of the injections had any detrimental effects on embryonic development (Fig. 35) suggesting that carried-over RNA from spermatids is not the cause of developmental failure of spermatid-derived embryos. Chapter 6: Results 121 Fig. 34. Injection of high doses of mRNAs encoding spermiogenesis-related proteins is toxic for embryos Embryos were injected at 1-cell stage with water or with 9.2ng of mRNAs encoding Sperm basic protein 1 (Sp1), Sperm basic protein 4 (Sp4) or Sperm basic protein 5 (Sp5). Injection of Sp4 or Sp5 was toxic to the embryos. Fig. 35. RNA carry-over is not the cause of developmental defects of spermatid-derived embryos Fertilised embryos (at 1-cell stage) were injected either with water, with 50pg of mRNAs encoding sperm basic proteins (Sp1, Sp4 and Sp5) or with 50pg of total testicular RNA (‘Total RNA’). Injections did not affect the normality of embryonic development, since all embryos developed into normal swimming tadpoles. Chapter 6: Results 122 6.4. Haploid paternal embryos as a tool for a specific assessment of transcription from the paternally-derived chromatin Another major challenge that the embryo needs to accomplish to develop successfully is the zygotic genome activation (Newport & Kirschner 1982). Therefore, I decided to assess the ability of sperm- and spermatid-derived embryos to support embryonic transcription. After ICSI, embryonic transcription occurs from both the paternal and the maternal genome. In order to specifically assess the ability of the paternally-inherited chromatin to drive transcription, I decided to use haploid paternally-derived embryos. Haploid paternally- derived embryos are typically generated by first enucleating the egg with a Mineralite UV lamp treatment, followed by in vitro fertilisation. Developing embryos are haploid and their genetic material is inherited solely paternally (Gurdon 1960, Hamilton 1963) (Fig. 23). Since here I wanted to generate haploid sperm- or spermatid-derived embryos, I had to slightly modify this protocol to compensate for the fact that a spermatid cannot fertilise the egg: sperm or spermatids were injected into enucleated eggs (instead of performing the in vitro fertilisation) (Fig. 36A). I have first tested whether haploid embryos recapitulate the developmental phenotypes of diploid sperm- and spermatid-derived embryos. I have allowed the generated haploid embryos to develop and scored them in the same way as diploid embryos were scored: as a number of gastrula embryos and as a number of swimming tadpoles to the total number of cleaved embryos. The results obtained with haploid embryos agreed with the findings obtained with diploid embryos – there was no significant difference in the embryo development to the gastrula stage, but sperm-derived embryos developed significantly better to the swimming tadpole stage compared to spermatid-derived embryos (p-value < 0.05) (Fig. 36B). This therefore validates the use of haploid embryos for the assessment of transcription Chapter 6: Results 123 originating specifically from sperm- or spermatid-derived chromatin at the time of embryonic gene activation. Fig. 36. Developmental advantage of sperm over spermatid is maintained in haploid embryos (A) Diagram explaining ICSI into enucleated eggs. Eggs are first enucleated with the Mineralite UV lamp treatment and subsequently injected with sperm or spermatids. (B). Haploid sperm-derived embryos developed better than haploid spermatid-derived embryos. Embryos were scored as the % of embryos reaching a gastrula stage and a swimming tadpole stage to the total number of cleaved embryos. Numbers of embryos analysed are indicated above the bars. N = 3 independent experiments. Error bars show ± SEM. * indicates p-value = 0.008 (z-test). Chapter 6: Results 124 6.5. rRNA synthesis occurs normally in spermatid-derived embryos I next tested whether spermatid-derived embryos are equally suited to support embryonic transcription of rRNA as sperm-derived embryos. It was reported that mouse nuclear transfer-derived embryos aberrantly expressed rRNAs when compared to in vitro fertilised embryos (Suzuki et al. 2007) and that this correlated with their poor developmental outcomes (Zheng et al. 2012). I have therefore hypothesised that sperm is programmed to support efficient rRNA synthesis, whereas the spermatid is not. To test this hypothesis, enucleated eggs were injected with sperm or spermatids and with BrUTP. BrUTP was co- injected with sperm or spermatids in order to label only the newly synthesised transcripts (Core et al. 2008) and to distinguish them from rRNA maternally accumulated during oogenesis (Roger et al. 2002). This procedure allowed generation of haploid sperm- or spermatid-embryos which were collected at the gastrula stage. Subsequently, newly synthesised RNA was pulled down and quantified by reverse transcription quantitative PCR (RT-qPCR) for 18S and 28S rRNA (Fig. 37A). The results of the RT-qPCR analysis revealed that there are no significant differences in the amount of 18S or 28S rRNA synthesised between the sperm- and spermatid-derived haploid embryos (Fig. 37B). This suggests that problems with correct activation of embryonic rRNA synthesis are unlikely to be the cause of developmental defects of spermatid-derived embryos. Chapter 6: Results 125 Figure 37. Spermatid-derived embryos are as good as sperm-derived embryos at synthesising rRNAs. (A) Diagram explaining newly synthesised RNA isolation from sperm- and spermatid- embryos. Haploid sperm- and spermatid-embryos are obtained by ICSI to enucleated eggs and are co-injected with BrUTP to label newly synthesised RNA. Embryos develop to a gastrula stage when they are collected. Subsequently, BrUTP-labelled, newly synthesised RNA is pulled down. (B) Spermatid-derived embryos synthesised rRNA as efficiently as sperm-derived embryos, as evidenced by RT-qPCR quantification of 18S and 28S rRNAs. Values are shown as a percentage of pulled down RNA to the total input RNA. Error bars show ± SEM. N=20 sperm-derived embryos and N=14 spermatid-derived embryos. Samples were not significantly different (p-value = 0.82 for 18S rRNA and p-value = 0.36 for 28S rRNA, t-test). Chapter 6: Results 126 6.6. Developmentally-important mRNAs are misexpressed in spermatid- derived embryos Zygotic rRNA activation was not different between sperm- and spermatid-derived embryos, therefore I next hypothesised that spermatid-derived embryos may show aberrant mRNA transcription. In order to investigate the potential mRNA expression changes in a global way, I decided to perform RNA sequencing analysis of sperm- and spermatid-derived embryos. To focus my analysis on the transcription originating from the paternal chromatin, I again used haploid sperm- and spermatid-derived embryos. In order to eliminate technical variation between experiments and to facilitate the identification of the biologically meaningful differences, I collected the embryos (pools of 5 sperm- or spermatid-derived embryos) in seven independent experiments: experiments were conducted on different days, with eggs obtained from seven independent frogs and from three independent sperm and spermatid cell preparations. I generated haploid embryos as described above (Fig. 36A) and I collected them at the gastrula stage, before the onset of developmental defects. Subsequently, I isolated RNA from the embryos and generated sequencing libraries for five out of seven experimental replicates and sequenced them at the sequencing facility at the Cambridge Research Institute. Two remaining RNA samples were sent to our collaborators, Dr Taejoon Kwon and Dr Edward Marcotte at the University of Texas, USA, for independent library preparation and sequencing. Bioinformatic analyses performed by Dr Angela Simeone, Dr Charles Bradshaw and Dr George Allen identified 255 out of 18,340 transcripts as abnormally expressed in spermatid-derived embryos (compared to sperm-derived embryos) with a false discovery rate (FDR) below 0.05 (Table S5). When applying more stringent filtering criteria (selecting only those transcripts which were consistently up- or down-regulated in at least 6 out of 7 separate Chapter 6: Results 127 experiments), a final list of 100 transcripts differentially expressed in spermatid-derived embryos was obtained. From now on I refer to these 100 transcripts as ‘misregulated’ (Fig. 38A and Table S5). The majority of these misregulated transcripts (82 out of 100) were found to be upregulated, while only 18 out of 100 were downregulated in spermatid-derived embryos, as compared with sperm-derived embryos. The RNA-seq results were confirmed by RT-qPCR analysis (Fig. 38B). In order to further characterise the misregulated genes, gene ontology (GO) enrichment analysis was performed by Dr George Allen. The analysis revealed that several developmentally-important terms are significantly enriched in the misregulated gene list (p- value < 0.05) (Fig. 39A). Indeed, more than 25% of the misregulated transcripts are known transcriptional regulators essential for embryonic development, for example gata2, gata3, hes1 and fos (Zon et al. 1991, Kelley et al. 1994, Maeno et al. 1996, Kim et al. 1998, Read et al. 1998, Nardelli et al. 1999, Jouve et al. 2000, Friedle & Knochel 2002, Nakazaki et al. 2008, Lee et al. 2011). Furthermore, other misregulated transcripts, such as bmp2, bmp7 or dhh, are morphogens with crucial roles in the induction of germ layers and cell signalling (Bitgood & McMahon 1995, Reversade & De Robertis 2005, Reversade et al. 2005, Wills et al. 2008). Chapter 6: Results 128 6.7. mRNAs misexpressed in spermatid-derived embryos are Polycomb targets in human sperm Interestingly, I noticed that the GO terms enriched in the misregulated set of transcripts (Fig. 39A) are very similar to those enriched for genes bearing trimethylated lysine 27 on histone H3 (H3K27me3) in the mature sperm in human (Brykczynska et al. 2010) (Fig. 39B). I therefore tested whether human orthologues of the Xenopus misregulated transcripts identified here may also be H3K27me3-modified in human sperm. Orthology search was performed by Dr Charles Bradshaw and the cross comparison of H3K27me3- modifed genes in human sperm to their Xenopus orthologues was performed by Dr Angela Simeone. It was found that among the Xenopus misregulated genes that have human orthologues, 41% were enriched for H3K27me3 in human sperm (Fig. 39C). This is a significant overrepresentation of H3K27me3-modified genes (p-value < 0.05), since of all human orthologues of Xenopus genes, only 16% are enriched for H3K27me3 in sperm (Fig. 39C and Table S6). Concluding, 100 developmentally-important transcripts were identified as misregulated in spermatid-derived embryos, of which the majority of transcripts was upregulated. Furthermore, orthologues of these misregulated genes are enriched for the H3K27me3 mark in human sperm. This result supports the hypothesis that the nucleus of the sperm, but not of a spermatid, can be a subject of epigenetic programming to regulate the transcription of developmentally-important genes in the future embryo. Chapter 6: Results 129 Fig. 38. Legend on the subsequent page Chapter 6: Results 130 Fig. 38. Identification of 100 transcripts misregulated in spermatid-derived embryos (A) Heatmap of transcripts misregulated in spermatid-embryos. Haploid sperm- and spermatid-embryos were collected in seven experimental replicates and subjected to RNA- seq analysis. Heatmap shows 100 misregulated transcripts in spermatid-embryos: transcripts in red were upregulated, whereas transcripts in blue were downregulated (FDR < 0.05). Columns represent expression values in counts per million of reads (cpm) for each transcript (rows) obtained in seven independent experiments (columns 1-7). Transcripts are sorted by average log2 of fold difference in expression levels between spermatid- to sperm-embryos. Examples of interesting misregulated transcripts are indicated on the right hand side of the heatmap (B) RT-qPCR validation of misregulated transcripts identified by RNA-seq analysis. Ten randomly selected transcripts from the fifty lowest FDR, upregulated transcripts, and four randomly selected out of eighteen downregulated transcripts were selected for RT-qPCR validation. Expression values for each gene were normalised to the housekeeping gene pwp1. Bars show average log2 of fold change (FC) of expression values obtained for spermatid- derived embryos to values obtained for sperm-derived embryos. Red bars show transcripts which were identified as upregulated in spermatid-derived embryos in RNA-seq, and blue bars show transcripts identified as downregulated in RNA-seq. N=6 independent experiments for all transcripts, apart from Mn1 and Chd3 for which N=5 independent experiments. P- values below 0.1 are shown above the bars (t-test). Panel ‘A’ of this figure was created and kindly shared by Dr Angela Simeone and Dr George Allen. Chapter 6: Results 131 Fig. 39. Human orthologues of Xenopus misregulated genes are marked by H3K27me3 in sperm. (A) Developmentally-important gene ontology terms are enriched in the list of misregulated genes in spermatid-derived embryos (p-value < 0.05). (B) Top 6 gene ontology terms enriched within genes having H3K27me3 mark in human sperm (Brykczynska et al. 2010). (C) Number of Xenopus laevis orthologues of genes enriched for H3K27me3 and H3K4me2 in human sperm (Brykczynska et al. 2010). * - p-value=0.000009 and p-value=0.00006 (proportion test and hypergeometric test, respectively), demonstrating that H3K27me3- positive genes are significantly overrepresented in the list of human orthologues of Xenopus misregulated genes. Statistical analysis of the enrichment was performed by Dr Angela Simeone. Chapter 6: Results 132 6.8. Summary Results described in this chapter aim at the identification of abnormalities observed in spermatid-derived embryos, as compared with sperm-derived embryos. I showed that DNA replication problems, carry-over mRNA or problems with the activation of the zygotic rRNA transcription are unlikely to be the cause of developmental defects of spermatid-derived embryos. On the other hand, RNA-seq analysis of sperm- and spermatid-derived embryos allowed the identification of 100 developmentally-important mRNAs which are misexpressed in spermatid-derived embryos, as compared with sperm-derived embryos. Interestingly, the majority of these mRNAs (82/100) turned out to be upregulated in spermatid-derived embryos. Misregulation of these mRNAs is a plausible explanation for the developmental defects of spermatid-derived embryos, especially since many of them are important transcriptional regulators of embryonic development. Another interesting and unexpected finding is that the gene ontology terms enriched in the misregulated gene list in spermatid-derived embryos were strikingly similar to the gene ontology terms enriched within H3K27me3-positive genes in mature human sperm. More detailed analysis of human orthologues of Xenopus genes revealed a significant enrichment for genes positive for H3K27me3 in the list of human orthologues of the Xenopus misregulated genes. This finding suggests that perhaps sperm, but not the spermatid, is epigenetically programmed to regulate transcription of embryonic genes. Chapter 7: Results 133 Chapter 7 Epigenetic profiling of sperm and spermatids Experiments with enzymatic removal of H3K27me3 mark in in vitro matured/ICSI embryos were performed under the supervision and with a great help from Dr Kei Miyamoto and Dr Jerome Jullien. I was trained and supervised on how to perform MNase digestions and ChIP- seq analyses on the chromatin from sperm and spermatids by Dr Serap Erkek and Dr Antoine Peters (Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland), who collaborated with our laboratory on this project and who allowed me to visit their laboratory to obtain a technical training vital for this project. Bioinformatic analyses described in this chapter were performed by Dr Angela Simeone. 7.1. Introduction The results discussed in the previous chapter show that the expression of one hundred developmentally-important genes is misregulated in spermatid-derived embryos, as compared with sperm-derived embryos. Interestingly, human orthologues of these genes are enriched for H3K27me3 mark in the mature sperm. This result prompted me to hypothesise that perhaps sperm, as opposed to spermatids, is epigenetically programmed to regulate expression of embryonic genes in the embryo after fertilisation. Indeed, it was shown in many species that mature sperm retains post-translationally modified histones on chromatin Chapter 7: Results 134 (Hammoud et al. 2009, Brykczynska et al. 2010, Wu et al. 2011). It has been suggested that such histones can be delivered to the embryo at fertilisation (van der Heijden et al. 2008). If the epigenetic marks on the chromatin could be delivered to the oocyte at fertilisation, then they could pattern the gene expression in the future embryo. Any differences in the chromatin structure between sperm and spermatids could potentially result in differences in the gene expression patterns. I have therefore decided to test the hypothesis that sperm, but not spermatids, may be epigenetically suitable to support embryonic transcription. In this chapter I describe my results that characterise the epigenetic status of the chromatin of sperm and spermatids in Xenopus laevis. I first tested whether the presence of H3K27me3 mark on the paternal chromatin at fertilisation is important for the regulation of the gene expression in the embryo. The results obtained showed that this is indeed the case – enzymatic removal of H3K27me3 marks from the parental chromatin at fertilisation led to gene misexpression in the embryo. This confirmed that the presence of epigenetic marks on the parental chromatin is important for the future regulation of embryonic gene expression. I therefore hypothesised that differential gene expression between sperm- and spermatid- derived embryos could result from differences in epigenetic marks between sperm and spermatids. To interrogate the possible epigenetic differences between sperm and spermatids I first performed a brief general characterisation of sperm and spermatids chromatin by micrococcal nuclease digestion, which allowed the identification of unique chromatin structures in sperm, but not in spermatids. Subsequently, I performed ChIP-seq analyses for H3K27me3 and for H3K4me2 and H3K4me3. Interestingly, my results suggest that the overexpression of genes in spermatid-derived embryos is not explained by the lack of H3K27me3 in spermatids, but instead by a higher abundance of H3K4me2/3 marks in spermatids than in sperm. Chapter 7: Results 135 7.2. Parentally-derived H3K27me3 is necessary for correct gene expression in embryos Cross-comparison of genes modified by H3K27me3 in human sperm with the misregulated genes suggested that sperm might be epigenetically programmed by H3K27me3 to support a proper embryonic gene transcription. Before embarking on a detailed characterisation of the epigenetic marks in sperm or spermatids, I wanted to test whether epigenetic marks on the parental chromatin are of any importance for the regulation of gene expression in the early embryo. In other words, I first wanted to test the functional importance of parentally-derived H3K27me3 in the regulation of embryonic gene expression. To interfere with H3K27me3, I used Kdm6b (K6B), an enzyme that specifically demethylates H3K27me3, and as a control I used K6B mutant, which is catalytically inactive (K6B-mut). Constructs encoding these enzymes were kindly provided by Dr Jerome Jullien. First, by injecting mRNA encoding K6B or K6B-mut into 1-cell stage embryos I confirmed that K6B enzyme, but not its mutant version, removed H3K27me3 from embryonic chromatin (Fig. 40). I designed my experiments in a way that allows H3K27me3 removal from both parental chromatin sets immediately at fertilisation. For that purpose I adapted the technique of in vitro maturation of prophase-arrested oocytes (Miyamoto et al. 2013) and I was supervised and assisted in performing these experiments by Dr Kei Miyamoto and Dr Jerome Jullien. Firstly, mRNA encoding K6B is injected into immature, GV stage oocytes to allow for protein overexpression. Such oocytes, pre-loaded with K6B or K6B-mut enzyme, are subsequently in vitro matured into eggs and injected with sperm (via ICSI procedure) to generate embryos (Figure 41A). In this experimental setup, with the currently available protocols, it is not technically possible to enucleate the oocyte and to obtain haploid embryos Chapter 7: Results 136 derived from paternal genome only. Instead, in the conditions used here, both maternal and paternal chromatins contribute to development of the embryo and the enzyme accumulated in the oocyte prior to fertilisation acts on both parental chromatin sets. Embryos derived in such way were allowed to develop and were collected at the gastrula stage for RT-qPCR analysis. I assessed the expression levels of five genes misregulated in spermatid-derived embryos (gata3, plod2, hes1, mn1 and c19orf26), and of two control genes: hoxb1 and wasf1. Expression of hoxb1 is regulated by H3K27me3 (Agger et al. 2007). Therefore, hoxb1 serves as a positive control for the experiment, since the expression level of this gene should be affected by the removal of H3K27me3. Wasf1 does not have H3K27me3 around its genomic region (Akkers et al. 2009), so its expression levels should not change upon H3K27me3 removal. As expected, in this experimental setup K6B overexpression had pronounced effects on gene expression (Figure 41B). Hoxb1 expression was upregulated 20-fold. Importantly, expression of 4 out of 5 misregulated genes tested was also upregulated (statistically significant upregulation was observed for plod2 and gata3, p-values < 0.05, t-test). I concluded from this experiment that the presence of H3K27me3 on parental chromatin is required for the proper regulation of embryonic gene expression. To determine whether the upregulation of gene expression observed was a result of H3K27me3 removal immediately at fertilisation or at a later point during development, I tested the effect of H3K27me3 removal during embryogenesis (after fertilisation). For that I injected fertilised 1-cell embryos with mRNA encoding K6B (Fig. 41C). In that way the enzyme is absent at fertilisation and becomes translated and modifies the chromatin only as the embryo develops. I again collected the embryos for RT-qPCR analysis at the gastrula stage and performed the analysis for the same genes as described above. Interestingly, in this experimental setup I observed that control genes, as well as genes identified as misregulated in spermatid-derived embryos were similarly transcribed in embryos expressing K6B or Chapter 7: Results 137 K6B-mut (Fig. 41D). I therefore concluded that H3K27me3 removal during embryogenesis does not affect early embryonic gene expression. Summarising, these two sets of experiments (the removal of H3K27me3 at fertilisation and the removal of H3K27me3 later during embryonic development) demonstrate that the presence of H3K27me3 on parental chromatin at fertilisation is required for the proper regulation of expression of embryonic genes. Removal of the H3K27me3 marks during embryogenesis does not affect the expression levels of the tested genes. Results of these experiments support the hypothesis that the epigenetic marking in the sperm may be required for a proper embryonic development. Chapter 7: Results 138 Fig. 40. Overexpression of K6B leads to a removal of H3K27me3 mark. (A) One-cell embryos were injected either with mRNA encoding K6B or K6B-mut (MUT). Non-injected embryos were used as a control (WT). Embryos were collected for a western blot analysis 1 day (1d, gastrula stage), 2 days (2d, tailbud stage) or 4 days (4d, early tadpole) after mRNA injection. Membrane is stained with an antibody against H3K27me3. H3K27me3 is removed upon overexpression of K6B, but not upon overexpression of K6B- mut. (B) Quantification of H3K27me3 removal based on the results of immunoblots from three independent experiments. Error bars show ±SEM. Chapter 7: Results 139 Fig. 41. Removal of H3K27me3 at fertilisation leads to gene misexpression. (A) H3K27me3 mark removal at fertilisation. Immature, prophase-arrested germinal vesicle (GV) stage oocytes are injected with mRNA encoding K6B or K6B-mut. mRNA is translated into protein and subsequently the oocytes, pre-loaded with K6B or K6B-mut, are in vitro matured into eggs. Sperm is then injected into such eggs and the resulting embryos are collected for RT-qPCR analysis at the gastrula stage. (B) Gene expression levels in in vitro matured and sperm injected (IVM/ICSI) embryos expressing K6B or K6B-mut. Gene expression was normalised to the housekeeping gene pwp1. Expression levels are shown as fold differences in gene expression between embryos expressing K6B to embryos expressing K6B-mut. Error bars show ± SEM. N=5 experimental replicates. * indicates p-value < 0.05 (t-test). (C) H3K27me3 mark removal during embryonic development. 1-cell stage embryos are injected with mRNA encoding K6B or K6B-mut. In this way the proteins are absent at fertilisation and they become expressed only as the embryo develops. Embryos are collected for RT-qPCR analysis at the gastrula stage. (D) Gene expression in in vitro fertilised (IVF) embryos expressing K6B or K6B-mut. Gene expression was normalised to the housekeeping gene pwp1. Expression levels are shown as fold differences in gene expression between embryos expressing K6B to embryos expressing K6B-mut. Error bars show ±SEM. N=3 experimental replicates. Chapter 7: Results 140 7.3. General characterisation of chromatin structure in sperm and spermatids Removal of H3K27me3 marks from the parental chromatin at the time of fertilisation showed that they are important for the correct regulation of embryonic gene expression. I have therefore decided to further characterise the structure of chromatin in sperm and spermatids. I initially decided to see whether the chromatin is any different in sperm or spermatids in terms of its accessibility or its global structure. One simple way to test this is to treat the isolated chromatin with micrococcal nuclease (MNase). MNase is an endo- exonuclease enzyme that digests DNA at regions which are not protected from digestion. Protection from digestion can be mediated for example by the presence of nucleosomes on DNA. The nuclesome protects DNA fragment of around 150bp, therefore after the MNase digestion of chromatin isolated from somatic cells, one usually obtains a so called ‘nucleosome ladder’ – a DNA ladder of band sizes which are a multiplication of 150bp, for mono-, di- trinucleosomes etc. (Fig. 42). To establish a protocol for Xenopus laevis sperm and spermatids chromatin preparation and MNase digestion, a collaboration with the lab of Dr Antoine Peters (FMI, Basel) has been initiated, in order to benefit from the expertise of Dr Peters in this methodology (Brykczynska et al. 2010, Erkek et al. 2013, Hisano et al. 2013). Chromatin from sperm and spermatids was isolated and subjected to MNase digestion for 30mins at 37ºC (using 2.5 units of MNase per chromatin isolated from 1 million of cells). As a control for MNase digestion, I have also prepared and digested chromatin from Xenopus laevis cell line XL177. Subsequently, DNA has been isolated and run on an agarose gel. Somatic cells and spermatids displayed a typical ‘nucleosome ladder’ digestion pattern (Fig. 43). Spermatids were more digested than XL177 cells (more mononucleosomes and less of the higher order chromatin structures observed after digestion). Unexpectedly, it turned out that the sperm has a very unique pattern of nuclease-protected DNA regions. Three fragments Chapter 7: Results 141 of different sizes were observed after MNase digestion: ~75bp, ~110bp and ~150bp long fragments (Fig. 43). Interestingly, mouse sperm digested with MNase displays a digestion pattern typical for any other somatic cells (150bp size nucleosomes) (Brykczynska et al. 2010), which therefore suggests that Xenopus sperm is very unique in its chromatin composition. Fig. 42. ‘Nucleosome ladder’ after MNase digestion Drawing showing an example of a ‘nucleosome ladder’ when DNA isolated from chromatin from somatic cells digested with MNase is run on a gel. Note that the size of the bands of the ‘ladder’ is approximately a multiplication of 150bp, which is a size of DNA protected by a single nucleosome. Chapter 7: Results 142 Fig. 43. MNase digestion reveals that sperm chromatin has a unique structure Chromatin isolated from 1 million of sperm, spermatids or XL177 cells was digested with 2.5 units of MNase for 30mins at 37ºC. Subsequently, DNA was isolated and run on the agarose gel. Spermatids and XL177 cells digestion pattern displays a typical ‘nucleosome ladder’ (mononucleosome band size indicated), whereas digestion of sperm revealed a unique structure of sperm chromatin, consisting of three bands of approximately 150bp, 110bp and 75bp (bands are respectively numbered 1, 2, 3 and indicated on the gel). The lowest band (4) is digested DNA, which was also observed when mouse or human sperm is digested with MNase (Brykczynska et al. 2010). Chapter 7: Results 143 It is known that mature sperm in Xenopus laevis retains core histones H3 and H4, but has a reduced amount of histones H2A and H2B (Abe & Hiyoshi 1991, Yokota 1991, Shechter et al. 2009). My immunoblotting results on Xenopus leavis sperm and spermatids agreed with the published observations – I could detect similar amounts of H3 and H4 by immunoblotting in sperm and spermatids and reduced amounts of histones H2A and H2B (Fig. 44). I next hypothesised that perhaps at least some of the chromatin structures that I observed in the sperm, but not in the spermatid chromatin after MNase digestion are due to the incorporation of sperm basic proteins into the sperm chromatin. Sperm basic proteins in Xenopus are more basic than histones, as evidenced by theoretical isoelectric point calculations (calculations performed using online Expasy tool http://web.expasy.org/cgi- bin/compute_pi/pi_tool): the least basic of the sperm basic proteins (Sp1) is still more basic than any canonical histone (Table 6). I have therefore reasoned that if any of the sperm chromatin structures observed after MNase digestion have sperm basic proteins as their components, they could be preferentially disrupted upon heparin treatment of chromatin. Heparin is a molecule of high negative charge and is therefore able to displace positively charged proteins from chromatin (Hildebrand et al. 1977). Treatment of permeabilised Xenopus sperm with heparin leads to chromatin dispersal and DNA release. Dispersal os sperm chromatin increased with the increasing concentration of heparin used (Fig. 45). Since the sperm basic proteins are more basic than histones, they should be displaced from sperm chromatin by the heparin treatment earlier than histones. Therefore, if some of the structures on sperm chromatin that protect the DNA from the MNase digestion are composed with the sperm basic proteins to a higher extent than the others, they could be displaced from the chromatin first. I have therefore decided to test the effect of sperm chromatin treatment with increasing doses of heparin on the pattern of MNase digestion. I have treated permeabilised sperm nuclei with increasing doses of heparin and subsequently performed MNase digestion. Chapter 7: Results 144 Interestingly, ~110bp band (number 2) generated by MNase digestion is the first one to disappear upon increased heparin concentration (Fig. 46). This suggests that the band number 2 (~110bp long) could be composed of sperm basic proteins or be a mixture of sperm basic proteins and histones. Alternatively, the second band could also be composed of unstable histones. Regardless of the reason for the preferential displacement of the 2 nd band, the important information is that these three chromatin structures in the sperm are not equally stable. The exact identity of these three bands and the nature of the DNA protected by these structures is a subject of follow-up studies currently conducted in the Gurdon laboratory. Table 6. Theoretical isoelectric point of nuclear proteins in Xenopus laevis. Name H2A H2B H3 H4 Sp1 Sp4 Sp5 Isoelectric point 10.17 10.31 11.27 11.36 11.89 12.57 12.62 Chapter 7: Results 145 Fig. 44. Immunoblotting analysis for histone H2A, H2B, H3 and H4 on sperm and spermatids. Immunoblotting analysis for core histones H2A, H2B, H3 and H4 on the protein lysates from sperm and spermatids. The number of cells from which the proteins were isolated is indicated above each lane (numbers represent thousands of cells, for example, ‘8’ stands for proteins isolated from 8000 cells). Note that sperm and spermatids contain similar amount of histones H3 and H4, and that mature sperm has a reduced amount of histones H2A and H2B. Chapter 7: Results 146 Fig. 45. Heparin treatment leads to chromatin dispersal in sperm. Permeabilised sperm were treated with increasing doses of heparin and stained with DAPI. Three representative images are shown for each of the heparin treatment doses. Note that the higher the heparin dose, the more decondensed the sperm. Scale bar = 20um. All images were taken with the same magnification. Chapter 7: Results 147 Fig. 46. Heparin treatment first releases the ~110bp structure from sperm chromatin. Sperm chromatin was treated with increasing concentrations of heparin (0.0025 – 0.125 mg/ml) and subsequently digested with 2.5U MNase for 30mins. DNA was isolated and run on the gel. Note that the chromatin structure protecting a ~110bp DNA fragments (number 2) disappears as the first one with the increasing dose of heparin treatment. This is concomitant with the appearance of digested DNA (band number 4). Further increase of the heparin concentration leads to a loss of all the protective chromatin structures and digestion of all DNA in sperm. Chapter 7: Results 148 7.4. ChIP-seq analysis for H3K27me3 does not reveal differences between sperm and spermatids ChIP experiments were performed according to the advice provided by Dr Serap Erkek and Dr Antoine Peters. All the bioinformatic analyses were performed by Dr Angela Simeone. My results obtained so far demonstrated that parentally-derived H3K27me3 marks regulate embryonic gene expression. Furthermore, I have shown that sperm and spermatids differ in their chromatin structure. I therefore reasoned that perhaps the misregulation of gene expression in spermatid-embryos, as compared to sperm-embryos, might be a consequence of improper epigenetic marking in the spermatid. As the majority of the misregulated genes are upregulated in spermatid-embryos, I hypothesised that the spermatid may lack the repressive H3K27me3 marks as compared with sperm. To address this, I performed chromatin immunopurification followed by a genome-wide sequencing (ChIP-seq) analysis for H3K27me3 mark on mononucleosomal chromatin isolated from Xenopus laevis sperm and spermatids. I have chosen the mononucleosomal chromatin fraction for the ChIP analysis, since this fraction is a canonical chromatin structure observed in both cell types and the identity of the other chromatin structures in sperm is currently unknown. I performed the ChIP experiments in three independent replicates (three different cell preparations from different frogs and ChIP experiments performed on different days). To assess the variability between the experiments, Dr Angela Simeone performed a Pearson correlation coefficient analysis for the replicates. The three biological replicates for the H3K27me3 mark showed an average Pearson correlation coefficient of 0.9 between replicates in sperm and 0.72 in spermatids. The ChIP-seq data was analysed first by looking at overall methylation levels and Chapter 7: Results 149 second, by looking at localised regions of enrichment for histone marks (peaks) (see Fig. 47 for a detailed explanation). 7.4.1. Overall methylation levels analysis I first wanted to test whether the misregulated genes have different histone H3K27me3 methylation levels than the genomic average. Quantification of methylation levels was performed by Dr Angela Simeone. Methylation levels were calculated as the total number of reads obtained for H3K27me3 (normalised to the number of reads obtained in the input sample) at the region around the transcriptional start site and also in the region encompassing the gene bodies, as H3K27me3 was shown before to spread in broad domains across the genes (Fig. 47) (Barski et al. 2007). The analysis of the H3K27me3 methylation levels at misregulated genes as compared with the genomic averages revealed that in Xenopus laevis misregulated genes are significantly more methylated (have more reads) for H3K27me3 in both sperm and in spermatids (p-values<0.05) (Fig. 48A). This confirms the previous finding made by cross-species comparison to human sperm (Fig. 39) – the misregulated genes in Xenopus indeed have more of H3K27me3 mark. Unexpectedly however, it seems that H3K27me3 cannot on its own explain the difference in gene expression between sperm- and spermatid-derived embryos, since it is enriched at misregulated genes in both cell types. To make sure that meaningful differences in H3K27me3 methylation levels are not overlooked in the bioinformatic analysis described, I also decided to compare H3K27me3 levels at misregulated genes between sperm and spermatids. Therefore, in addition to the comparison of the methylation levels between the misregulated genes and the genomic averages (see above), Dr Angela Simeone also directly compared the methylation levels at Chapter 7: Results 150 the misregulated genes between sperm and spermatids. The reasoning behind performing this additional analysis is that the mark could be overrepresented in both samples, but still be more abundant in sperm than in spermatids. The results of the comparison between sperm and spermatids showed H3K27me3 levels at misregulated genes was not significantly different between sperm and spermatids (p-value = 0.7) (Fig. 48B and Fig. 49A). The same was true also when looking at individual gene tracks – no significant differences in H3K27me3 methylation patterns were observed between sperm and spermatids (Fig. 49B). To conclude, the quantification of overall methylation levels at misregulated genes showed a similar level of enrichment over the genome-wide average for H3K27me3 in both cell types. 7.4.2. Peak analysis To further characterise any potential differences in H3K27me3 level between sperm and spermatids, regions of enrichment (peaks) for histone marks were identified in the gene regulatory regions and in the gene bodies by Dr Angela Simeone (Fig. 47). In agreement with the findings for the overall methylation levels, genes with H3K27me3 peaks were significantly enriched within the misregulated gene list in both cell types, as compared to the genomic averages (p-values < 0.05) (Fig. 50A). It was also assessed (analysis by Dr Angela Simeone) how broad were the H3K27me3 peaks, as it could be that peaks for H3K27me3 differ in size between sperm and spermatids. Again, H3K27me3 peaks were of similar size in sperm and in spermatids (Fig. 50B), therefore the peak size of H3K27me3 could not explain the difference in gene expression between sperm- and spermatid-derived embryos. Chapter 7: Results 151 7.4.3. Summary – H3K27me3 marks are not different between sperm and spermatids To summarise, both the quantitative assessment of methylation levels and the peak- oriented analysis led to the conclusion that H3K27me3 is enriched at misregulated genes above the genomic average both in sperm and spermatids. This suggests that, even though H3K27me3 is necessary for proper gene expression (Fig. 41), presence of this mark alone cannot explain differences in gene expression between sperm- and spermatid-derived embryos, as the mark is enriched in both cell types. Chapter 7: Results 152 Fig. 47. Legend on the subsequent page. Chapter 7: Results 153 Fig. 47. Schematic representation of ChIP-seq analysis of histone marks: overall methylation level versus localised enrichments (peaks). (A) Quantification procedure is explained on the example of tbx3 gene for H3K4me2 ChIP. Firstly, all the reads obtained from sequencing of ChIP samples were normalised to the reads obtained from sequencing of the corresponding input samples and to the total number of sequencing reads. Normalised methylation tracks are shown as ‘Sperm methylation’ and ‘Spermatid methylation’. Regions of read enrichment for each mark (‘peaks’) were identified with the use of MACS2 (Zhang et al., 2008) and are depicted as horizontal bars spanning the regions of significant read enrichment and are visualised as tracks named ‘Sperm peaks’ and ‘Spermatid peaks’. (B) Subsequently, overall methylation levels were quantified by summing up the normalised read number in the regions spanning -10kbp to +2kbp of gene transcriptional start sites (TSSs) for H3K4me2 and H3K4me3 and in the regions spanning -10kbp of TSS and the entire gene body for H3K27me3 (Bernstein et al. 2005, Barski et al. 2007, Akkers et al. 2009, van Heeringen et al. 2014). (C) Example of the results of quantification of normalised read number for tbx3. (D) The presence of peaks was evaluated in the same regions as above: -10kbp/+2kbp from TSS for H3K4me2/3 and -10kbp + gene body for H3K27me3. Table shows an example of the results of the peak presence assessment for tbx3. Chapter 7: Results 154 Fig. 48. Cumulative distribution curves for H3K27me3 in sperm and spermatids. (A) Cumulative distribution curves of overall methylation levels for H3K27me3 compared between misregulated genes and the genome-wide averages in sperm or spermatids. Curves for the misregulated gene methylation level in sperm are shown in blue, for the spermatid – in red, and for the genome-wide average – in grey. P-values for the difference between the methylation level at misregulated genes and the genome-wide average are indicated in the graphs (Kolmogorov-Smirnov, KS-test). (B) Cumulative distribution curves of H3K27me3 methylation level at misregulated genes between sperm (blue) and spermatid (red). P-values for the difference between the methylation levels are indicated in the graphs (KS-test). Analysis for this figure was performed by Dr Angela Simeone. The figure was generated and kindly shared by Dr Angela Simeone. Chapter 7: Results 155 Fig. 49. H3K27me3 at misregulated genes is not different between sperm and spermatids. (A) A heatmap showing the average normalised number of reads for misregulated genes in sperm and spermatids for H3K27me3. Genes (rows) are sorted from the most methylated to the least methylated in spermatids. (B) Methylation patterns of a representative misregulated gene in sperm and spermatids for H3K27me3. Track shows read numbers in the bound fraction, normalised to the input and to the total number of sequenced reads. Analysis for this figure was performed by Dr Angela Simeone. The panel ‘A’ of this figure was generated and kindly shared by Dr Angela Simeone. Chapter 7: Results 156 Fig. 50. Peak analysis for H3K27me3 reveals no differences between sperm and spermatids. (A) Peaks were counted in the upstream regions and in the gene bodies for H3K27me3. Statistical analysis shows a significant enrichment for H3K27me3-positive genes amongst misregulated ones in sperm and spermatids (as compared to genome-wide average) (B) Box plot analysis of the H3K27me3 peak width in sperm and spermatids reveals that peaks are of similar size in both cell types. The analyses for this figure were performed by Dr Angela Simeone, who also created and kindly shared the panel ‘B’ of this figure. Chapter 7: Results 157 7.5. Misregulated genes have more H3K4me2/3 activating marks in spermatids than in sperm My ChIP-seq results for H3K27me3 revealed that this repressive mark is not different between sperm and spermatids; therefore, it cannot explain the difference in gene expression between sperm- and spermatid-derived embryos. I next hypothesised that since the majority of misregulated genes are upregulated in spermatid-embryos, perhaps the spermatid has more activating epigenetic marks than the sperm. To address this, I performed ChIP-seq analysis on mononucleosomal chromatin isolated from sperm and spermatids for two histone marks associated with gene activation: histone H3 lysine 4 dimethylation (H3K4me2) and histone H3 lysine 4 trimethylation (H3K4me3). Average Pearson correlation coefficient analysis performed by Dr Angela Simeone confirmed the reproducibility of the results (the exact average Pearson correlation coefficient values were: 0.9 for sperm H3K4me2; 0.55 for sperm H3K4me3; 0.95 for spermatid H3K4me2 and 0.8 for spermatid H3K4me3). The ChIP-seq data were again analysed by two different approaches: first, looking at the overall methylation levels and second, by looking at the localised regions of enrichment for histone marks (peaks) (Fig. 47). 7.5.1. Overall methylation levels analysis In the initial analysis histone methylation levels were compared at the misregulated genes with the genomic average. Methylation levels were again quantified (by Dr Angela Simeone) as the total number of reads obtained for H3K4me2 and H3K4me3 in the regions around the transcriptional start sites (TSS) (Fig. 47). Interestingly, there was a slight enrichment of H3K4me2 reads at misregulated genes over the genome-wide average in Chapter 7: Results 158 spermatids (p-value = 0.1) (Fig. 51A), suggesting that the presence of high levels of H3K4me2 might be responsible for the gene upregulation in spermatid-embryos. To further characterise this difference, histone methylation levels were compared at the misregulated genes between sperm and spermatids. Interestingly, misregulated genes had a significantly higher methylation in spermatids than in sperm for H3K4me2 and H3K4me3, (p-values < 0.05) (Fig. 51B and Fig. 52A). At individual gene tracks, H3K4me2 and H3K4me3 showed mainly quantitative differences in the methylation levels between sperm and spermatids (Fig. 52B). Interestingly, for some genes H3K4me2 additionally showed a different distribution of methylation between sperm and spermatids: in such genes H3K4me2-positive regions in spermatids were broader than in sperm and often contained several enriched regions close to the TSS (Fig. 52B). To conclude, quantification of H3K4me2/3 levels at misregulated genes between spermatids and sperm revealed significantly higher levels of H3K4me2/3 in spermatids. Higher levels of H3K4me2/3 in spermatids correlate with the observed upregulation of genes in spermatid-derived embryos. 7.5.2. Peak analysis Subsequently, peak analysis for H3K4me2 and for H3K4me3 was performed (by Dr Angela Simeone). Interestingly, in sperm, but not in spermatids, there was a two-fold decrease in the number of H3K4me3-positive genes among the misregulated ones as compared to the genomic average (Fig. 53A). Next, the size of peaks for H3K4me2 and H3K4me3 was analysed in sperm and spermatids. Interestingly, peaks for H3K4me2 and for H3K4me3 in spermatids were broader than in sperm (Fig. 53B). Chapter 7: Results 159 7.5.3. Summary – H3K4me2/3 marks are more abundant in spermatids To summarise, the activating marks H3K4me2 and H3K4me3 are more abundant in spermatids and depleted in sperm at misregulated genes. Enrichment of activating H3K4me2/3 in spermatids, as compared sperm, correlates well with the fact that the majority of misregulated genes (82/100) are overexpressed in spermatid-derived embryos, as compared to sperm-derived embryos. This suggests that sperm, as opposed to the spermatid, is epigenetically programmed for proper embryonic gene expression. Chapter 7: Results 160 Fig. 51. Cumulative distribution curves for H3K4me2 and for H3K4me3 in sperm and spermatids. (A) Cumulative distribution curves of overall methylation levels for H3K4me2 and H3K4me3 compared between misregulated genes and the genome-wide averages in sperm or spermatids. Curves for the misregulated genes methylation levels in sperm are shown in blue, for the spermatid – in red, and for the genome-wide averages – in grey. P-values for the difference between the methylation level at misregulated genes and the genome-wide averages are indicated in the graphs (Kolmogorov-Smirnov, KS-test). (B) Cumulative distribution curves of H3K4me2 and H3K4me3 methylation levels at misregulated genes between sperm (blue) and spermatids (red). P-values for the difference between the methylation levels are indicated in the graphs (KS-test). Analysis for this figure was performed by Dr Angela Simeone. The figure was generated and kindly shared by Dr Angela Simeone. Chapter 7: Results 161 Fig. 52. H3K4me2 and H3K4me3 marks are more abundant in spermatids than in sperm at the misregulated genes. (A) Heatmaps showing the average normalised number of reads for the misregulated genes in sperm and spermatids for H3K4me2 and for H3K4me3. Genes (rows) are sorted from the most methylated to the least methylated in spermatids, separately for each mark. (B) Methylation patterns of representative misregulated genes in sperm and spermatids for H3K4me2 and for H3K4me3. Tracks show read numbers in the bound fraction, normalised to the input and to the total number of sequenced reads. Analysis for this figure was performed by Dr Angela Simeone. The panel ‘A’ of this figure was generated and kindly shared by Dr Angela Simeone. Chapter 7: Results 162 Fig. 53. Peak analysis for H3K4me2 and for H3K4me3 reveals depletion of H3K4me3 peaks in sperm and also shows that peaks for H3K4me2 and H3K4me3 are broader in spermatids. (A) Peaks were counted in the upstream regions for H3K4me2 and for H3K4me3. Note a two-fold depletion of H3K4me3-positive genes amongst misregulated ones in sperm (as compared to the genome-wide average) (B) Box plot analysis of the H3K4me2 and H3K4me3 peaks width in sperm and spermatids reveals that peaks are broader in spermatids than in sperm. The analyses for this figure were performed by Dr Angela Simeone, who also created and kindly shared the panel ‘B’ of this figure. Chapter 7: Results 163 7.6. Summary In this chapter I investigated the roles of epigenetic signatures in the parental chromatin for the embryonic development. I first showed that the presence of H3K27me3 on the parental chromatin at fertilisation (but not afterwards) is needed for a proper regulation of embryonic gene expression. Then I looked into the chromatin structure of sperm and spermatids. It turned out that the sperm of Xenopus laevis has a uniquely structured chromatin, which is evidenced by unusual patterns of DNA bands released after the MNase digestion (DNA bands pattern are different from the canonical ‘nucleosome ladder’ and they are also different from the patterns observed in mouse or human sperm). Subsequent ChIP- seq analysis of mononucleosomal chromatin from sperm and spermatids for the repressive H3K27me3 and activating H3K4me2 and H3K4me3 revealed that sperm and spermatids do not differ in the abundance of H3K27me3, but that H3K4me2/3 was higher in spermatids than in sperm at misregulated genes. Interestingly, the average width of peaks for H3K4me2/3 was also higher in spermatids than in sperm. These results support the hypothesis that the upregulation of developmentally-important genes in spermatid-derived embryos is a consequence of the presence of H3K4me2 and H3K4me3 at higher levels in spermatids than in sperm. These also suggest that during spermiogenesis the spermatid has to epigenetically mature (lose H3K4me2/3 marks) in order to be correctly programmed to support embryonic development. Chapter 8: Discussion 164 Chapter 8 Discussion In this thesis I described my results investigating the nature of sperm programming for embryonic development. To address this question, I compared the sperm with its precursor cell, a spermatid. The results obtained support the hypothesis that programming of sperm is related to the acquisition of chromatin signatures that support a correct gene expression in the embryo. Specifically, I showed that spermatids, as opposed to sperm, retain on the chromatin activating H3K4me2/3 marks. Presence of these activating marks at misregulated genes in spermatids correlates with upregulation of these genes in spermatid- derived embryos. Therefore, my results suggest that sperm, as opposed to spermatids, are epigenetically mature to support normal embryonic development. There are however certain limitations to the conclusions that can be drawn from the results presented. Also, there are experiments, which are now carried in the Gurdon laboratory, that can potentially extend the findings presented in this thesis. Therefore, below I thematically group these matters and discuss them. I finish this section by explaining how the results presented in this thesis broaden our current understanding of sperm programming and how they relate to the phenomena of transgenerational inheritance of the phenotypic changes through paternal epigenetic marks. Chapter 8: Discussion 165 8.1. Are the defects of spermatid-derived embryos a consequence of gene misexpression? In order to understand the nature of sperm programming, I investigated the developmental defects of spermatid-derived embryos. Identifying the defects of the spermatid-derived embryos, not observed in sperm-derived embryos, allowed me to narrow down the aspects of development in which the sperm-derived embryos are better. I showed that replication problems, carry-over RNA or inefficient activation of rRNA transcription are unlikely to explain the developmental defects of spermatid-derived embryos. Interestingly, I identified 100 developmentally-important mRNAs as misregulated in spermatid-derived embryos (compared to sperm-derived embryos). Is this however the real cause of the developmental defects of spermatid-derived embryos? The experiments presented in this thesis do not directly answer this question. Results shown allow correlating the developmental effects with the misexpression of these mRNAs. However, the presented results do not provide the functional evidence that could prove the hypothesis that misexpression of these mRNAs is indeed causative for the defects of spermatid-derived embryos. Experiments that could functionally test the relationship between the misexpression of these mRNAs and the occurrence of developmental defects in spermatid-derived embryos could be potentially designed in two different ways. First, one could try to overexpress the misregulated mRNAs in sperm-derived embryos to test whether developmental defects similar to those observed in spermatid-derived embryos would be induced. Alternatively, one could also downregulate proteins encoded by these mRNAs in spermatid-derived embryos to rescue their defects. Unfortunately however, none of these strategies are technically possible for multiple reasons. First, transcription factors need to be present at the correct concentrations and at very precise developmental time windows and Chapter 8: Discussion 166 their expression should be limited only to specific tissues. Currently available protocols aiming at overexpressing or downregulating the proteins utilise mRNA or morpholino oligonucleotides injection, respectively. Injections are performed into 1-cell stage embryos; therefore the effects would happen in all the embryonic tissues and would not be limited to the desired tissues only. Secondly, the majority of these misregulated mRNAs are embryonically expressed with a precisely controlled time of expression. Injecting mRNAs or morpholinos into 1-cell stage embryos does not allow one to have control over the developmental timing of the expression. Last, but not least, it is difficult to imagine how one would express all the factors of interest at the precisely desired concentration. This would be even more problematic in the morpholino knockdown strategy, as for this strategy one would also need to precisely measure how the mRNA concentration encoding a given factor is reflected in the final concentration of the translated protein. Interestingly, it has been demonstrated that about a 2-fold difference in the protein concentration of the transcription factor Xbra prevented nucleocytoplasmic hybrid embryos between Xenopus laevis and Xenopus tropicalis from the correct convergence/extension movements during gastrulation, which resulted in their failure to gastrulate successfully. These defects were partially rescued by correcting for the Xbra protein concentration (Narbonne et al. 2011). Therefore, if inappropriate concentration of a single transcription factor can be responsible for at least some gastrulation defects of the nucleocytoplasmic hybrid embryos (Narbonne et al. 2011), it is very likely that misregulation of 100 different mRNAs, which are developmentally-important and amongst which many are transcription factors, is indeed responsible for the developmental failure of spermatid-derived embryos. However, as mentioned before, the definite experimental evidence to support this statement is lacking and it is highly unlikely it would be technically possible to provide such evidence, at least with the currently available protocols. Chapter 8: Discussion 167 8.2. Model for epigenetic programming of the sperm nucleus My results show that expression of developmentally-important mRNAs is altered in spermatid-derived embryos as compared to sperm-derived embryos and that the majority of these mRNAs are upregulated in spermatid-derived embryos. With the help of Dr Kei Miyamoto and Dr Jerome Jullien I showed that the presence of H3K27me3 marks at fertilisation is necessary to prevent gene overexpression in embryos. This led me to the hypothesis that perhaps sperm has on its chromatin repressive H3K27me3 marks, which are absent on the spermatid chromatin, hence gene overexpression is observed in spermatid- derived, but not in sperm-derived embryos. However, ChIP-seq results showed that upregulation of genes in spermatid-derived embryos could not be explained by the absence of repressive H3K27me3 marks in spermatids, since these marks were enriched at misregulated genes in both sperm and spermatids. Instead, upregulation of misregulated genes correlated with the presence of higher levels of activating H3K4me2/3 marks in spermatid than in sperm. These findings suggest that proper regulation of embryonic gene expression is ensured in the sperm chromatin by two separate epigenetic layers: the presence of repressive H3K27me3 marks and a simultaneous depletion of activating H3K4me2/3 marks (Fig. 54). The spermatid already has the repressive H3K27me3, but does not yet lose activating marks from the misregulated genes (Fig. 54). Therefore, spermatids, as opposed to sperm, are not epigenetically mature to support correct gene expression in the embryo due to the retention of activating marks on their chromatin. These findings are in agreement with reports in mouse: in contrast to H3K27me3, which is similar between sperm and spermatids, H3K4me3 levels are reduced in sperm (Erkek et al. 2013). Chapter 8: Discussion 168 Curiously, the results described in this thesis show a similar low efficiency of normal development between spermatid-derived embryos and nuclear transfer-derived embryos (as compared with sperm-derived embryos). In others words, spermatid had a similarly low developmental potential as an early embryonic cell. Furthermore, it was shown in nuclear transfer experiments conducted in Xenopus laevis that ectopic expression of donor-cell genes in nuclear transfer-derived embryos was associated with the presence of H3K4me3 at these gene promoters in such embryos (Ng & Gurdon 2005, Ng & Gurdon 2008), as compared to control fertilised embryos. This is again similar to our observations that spermatids, as compared to sperm, had more of activating H3K4me2/3 marks on chromatin which correlated with the overexpression of embryonic genes in spermatid-derived embryos. All these results support the hypothesis that sperm is epigenetically-programmed to regulate the expression of embryonic genes. Chapter 8: Discussion 169 Fig. 54. Model for epigenetic programming of sperm nucleus for embryonic development. The presence of H3K27me3 and the absence of H3K4me2/3 in sperm is important for a proper regulation of embryonic gene expression. Model is explained on examples of several misregulated genes (gata3, hes1 and plod2), which are upregulated in spermatid-derived embryos. In spermatids, these genes have H3K27me3 and H3K4me2/3 marks. Repressive H3K27me3 marks are retained in the chromatin of mature sperm, but activating marks H3K4me2/3 are lost. Aberrant retention of activating marks at the misregulated genes in spermatids correlates with overexpression of these genes in spermatid-derived embryos and with abnormal development of spermatid-derived embryos. The remaining question is why the activating H3K4me2/3 marks were present at misregulated genes in the spermatid chromatin. As discussed earlier, misregulated genes were enriched for developmentally-important ones and many of them were transcription factors important for embryogenesis. Therefore, it is unlikely that those genes would be expressed in spermiogenesis. However, to exclude this possibility, I have performed a qRT-PCR analysis for selected misregulated genes (mn1, hes1, gata3 and sfrp2) to test whether they were expressed in spermatids. These genes were identified by ChIP-seq as having H3K4me3 peaks exclusively in spermatids and also as having lower quantity of H3K4me2 methylation level in sperm than in spermatids. As expected, the results of the qRT-PCR analysis confirmed that Chapter 8: Discussion 170 these genes were not expressed in spermatids (data not shown). Therefore, active expression of these genes in spermatids cannot explain the presence of H3K4me2/3 in spermatids. Interestingly, it was shown that in spermatids, a global acetylation of histones is observed (Hazzouri et al. 2000), which was suggested to be involved in chromatin rearrangements that allow its ultimate compaction in several of ways. First, it was shown that acetylated histones can be recognised and bound by Brdt – bromodomain, testis-specific protein, and it was shown that such binding by Brdt leads to the chromatin compaction (Pivot-Pajot et al. 2003, Govin et al. 2006, Moriniere et al. 2009). Secondly, acetylation (but not their polyubiquitination) triggered degradation of core histones during spermiogenesis by testis-specific proteasomes (Qian et al. 2013), presumably facilitating protamine deposition on chromatin, which are the main chromatin component of the mature mammalian sperm. Last, it was also suggested that global acetylation increases the instability of the canonical histone-structured chromatin during spermiogenesis, therefore facilitating the incorporation of protamines (Gaucher et al. 2010). Therefore, it is plausible to think that the presence of activating H3K4me2/3 marks at genes in spermatids, similarly to what was reported for histone acetylation, does not necessarily reflect their transcriptional activity, but that it is important for structural changes occurring during spermatid to sperm transition. For example, analogously to what has been suggested for histone acetylation, methylation of H3K4 could create a more open chromatin structure in the spermatid, which in turn could facilitate the access of sperm basic proteins to chromatin and as a result, lead to a chromatin remodelling which allows its compaction in sperm. Then, if such H3K4me2/3-modified chromatin of spermatid is delivered to an egg, the egg, which does not have sperm basic proteins in the cytoplasm and other machinery required for remodelling of the chromatin to the sperm-like state, could interpret the H3K4me2/3 as transcriptionally-activating marks. This in turn Chapter 8: Discussion 171 would lead to overexpression of genes having such marks, as indeed observed in spermatid- derived embryos. 8.3. How are the epigenetic marks transmitted to the embryo? Another interesting aspect emerging from this study, which was only partially addressed in this thesis, is how the epigenetic marks present on sperm are transmitted to the embryo. In collaboration with other members of the Gurdon laboratory, I showed that the experimental removal of H3K27me3 marks at fertilisation had a pronounced effect on gene expression, whereas removal of the very same mark later during embryonic development did not have such effect (Fig. 41). In early Xenopus embryos the first twelve cell cycles are rapid, with no gene transcription. Only after reaching the mid-blastula transition stage, is the zygotic genome activated (Newport & Kirschner 1982). It was shown that in species in which such rapid cell cycle phases precede zygotic genome activation, histone post-translational modifications, such as H3K4me3 or H3K27me3 were not detected on chromatin during these rapid cell cycles (Akkers et al. 2009, Vastenhouw et al. 2010, Lindeman et al. 2011). Post- translational epigenetic marks were established again on chromatin only around the time of zygotic genome activation (Lindeman et al. 2011). It is possible that the presence of post- translational histone marks and the need for their re-establishment after each cell cycle would impede rapid DNA replication phases. During S-phase in Drosophila embryos H3K4me3 and H3K27me3 marks disappear completely from the chromatin; to re-appear again only in G2 phase. Interestingly, in Drosophila the enzymes that carry out these modifications: Trithorax and Enhancer-of-Zeste, respectively, remain associated with chromatin during S-phase. This suggests that the enzymes that modify histone tails (chromatin ‘writers’) also act as Chapter 8: Discussion 172 placeholders for the actual histone marks (Petruk et al. 2012). Nascent chromatin capture experiments (performed by a pull down of chromatin at different phases of replication) showed that the enzymatic machinery associated with deposition of H3K27me3 mark on chromatin is indeed stably bound to chromatin in all cell cycle phases (Alabert et al. 2014). Interestingly, it was reported that Ezh2 (enhancer of zeste homolog 2), the enzyme that catalyses H3K27me3 deposition, was associated with chromatin in Xenopus tropicalis embryos and that this preceded H3K27me3 deposition (van Heeringen et al. 2014). This could suggest that writers of epigenetic marks could be, at the same time, their placeholders during the rapid cell cycle phases in the early embryo. A placeholder model fits well with the experimental data reported in this thesis – removal of epigenetic marks at fertilisation, but not after the mark has been already recognised by the placeholder, leads to the misregulation of gene expression (Fig. 41). It is also important to note that the removal of H3K27me3 marks late in embryogenesis, does not affect gene expression (Fig. 41). It is likely that in those conditions egg-derived Ezh2 is correctly targeted to genes, via binding to parentally-inherited H3K27me3, and can later on exert repression of these genes through an H3K27me3- independent mechanism. An alternative possibility to the one described above is that the marks are present in the embryo all the time, faithfully recapitulated from the parental chromatin even in the rapid cell cycle phases, and the only time when they are absent from the chromatin is the S-phase itself. Then, due to the fact that the cell cycle phases in the early embryos are rapid, with almost no G1 and G2 phases (almost exclusively S-phases followed by mitoses) (Newport & Kirschner 1982), by probing the chromatin isolated from the early embryo, one would almost always look into the chromatin undergoing active DNA replication, and therefore devoid of epigenetic marks. Furthermore, it has been also shown that during the rapid cell cycle phases in zebrafish, the nucleosomes are not well positioned. Canonical nucleosome organisation Chapter 8: Discussion 173 was only achieved around the time of zygotic genome activation (Zhang et al. 2014). Therefore, a globally low amount of well-positioned nucleosomes, plus the fact that early embryos are most of the time in S-phase, would prevent post-translational epigenetic marks from being detected at the earliest stages of embryogenesis, even if they were indeed present on the chromatin. How to distinguish between the two hypotheses? One would ideally probe for the presence of the marks at different phases of the cell cycle. In vitro egg extracts in Xenopus laevis make such experiments possible. With the use of such egg extracts it was demonstrated that DNA methylation from the sperm chromatin is faithfully recapitulated after DNA replication and the molecular mechanism of DNA methylase targeting to newly replicated, hemimethylated DNA was revealed (Nishiyama et al. 2013). It would be important to perform such experiments to examine what is happening with histone marks. The use of egg extracts would allow one to test for the first time what happens to sperm-derived epigenetically marked histones after fertilisation. So far the only experimental data providing hints that sperm-derived modified histones may be retained after fertilisation come from immunostainings in mammalian systems (van der Heijden et al. 2006, van der Heijden et al. 2008), which however do not provide any information about the localisation of the retained marks or how global is this retention. A substantial advantage of using the Xenopus laevis egg extracts is the possibility of precisely controlling and monitoring the progression of the cell cycle – one could collect the samples at various time points from the start of the egg extract treatment (various phases of the cell cycle). Also, with the use of egg extracts it is possible to perform ChIP-qPCR and ChIP-seq analyses, to identify the localisation of the retained marks on the paternal chromatin. Currently it is not known whether at the time of fertilisation the egg replaces all sperm-derived nuclear proteins (including post-translationally histones) with its own, maternally-stored histones, or whether sperm-derived, epigenetically Chapter 8: Discussion 174 marked histones are retained after this global chromatin remodelling. Also, even if such sperm-derived modified histones would survive the chromatin remodelling after the fertilisation, it is not known whether their marks would be reproduced after the DNA replication. Therefore, a precise timing of egg extract treatment would not only allow one to provide answers to these questions but also would enable discrimination between the two processes. Furthermore, the use of egg extracts creates the opportunity of a relatively easy examination of mechanisms allowing the possible retention of histone marks. One could perform immunodepletion of certain candidate factors from the extracts and examine the effects of their removal on the retention of histone marks. Experiments with the use of egg extracts aiming to determine what happens to H3K4me3 and H3K27me3 marks after fertilisation and during the first cell cycle, together with uncovering the underlying mechanisms of possible histone mark retention, are currently being designed in the Gurdon laboratory as follow-up studies on the results described in this thesis. It would be also extremely important to track the epigenetic marks during the early phases of embryonic development in vivo. Even though egg extracts were shown to be able to recapitulate the early cell cycle events, a demonstration that the same events occur in embryos in vivo would be still required. Using Xenopus as a model system provides the advantage of almost unlimited material available for ChIP (or ChIP-seq) experiments. My preliminary experiments with ChIP for H3K4me3 followed by qPCR using Xenopus laevis gastrula stage embryos revealed that as little at 10 embryos give one a sufficient amount of material to reliably detect this histone mark (data not shown). Therefore, it would be interesting to test whether the marks observed on the chromatin of sperm and spermatids are indeed retained in the developing embryos (and whether the differences observed between sperm and spermatids between them are maintained between sperm- and spermatid-derived embryos). Chapter 8: Discussion 175 Last, in the experiments that were designed to functionally test the importance of H3K27me3 by a K6B-mediated removal of these marks at fertilisation, the enzyme removed the marks from both the maternal and the paternal chromatin sets. Furthermore, the effects of H3K27me3 removal had apparent effects on gene expression only when it was present in the egg already at the time of fertilisation (Fig. 41). The mRNA was injected into fully-grown oocytes (isolated from PMSG-treated frogs), which are transcriptionally silent (Gilbert 2010), therefore it is unlikely that any major transcriptional events in the oocyte were disturbed by the H3K27me3 removal from the maternal chromatin before the sperm injection. However, this possibility cannot be excluded. Therefore, in order to be able to state precisely that the observed changes in transcription are due to the epigenetic mark removal at fertilisation, one would ideally examine the effects of delivering to the egg a paternal chromatin devoid of such marks. Also, in an ideal situation, such experiment would be performed in haploid embryos. As mentioned in the results section (Chapter 7.2), with the currently available protocols involving in vitro maturation of mRNA-injected oocytes, followed by ICSI, it is not possible to generate haploid, paternally-derived embryos. Therefore, the use of IVM/ICSI procedure does not allow the assessment of the effects of the histone mark removal at fertilisation solely on the paternal chromatin. However, a new experimental strategy is currently being designed in the Gurdon laboratory as a follow up of these experiments, in which recombinant histone demethylases are used. Such enzymes are able to remove histone marks from synthetic methylated peptides in vitro, therefore in theory they should be also able to remove such marks from a chromatin template. It is still technically impossible to remove the marks from the sperm chromatin, as it is highly condensed and any attempts of chromatin loosening lead to DNA dispersal (data not shown). However, Dr Jerome Jullien from the Gurdon laboratory is currently optimising the treatment of spermatids with histone demethylases in vitro. If such approach gives promising results, it should be possible to test Chapter 8: Discussion 176 in the future whether the removal of histone marks solely on the paternal (spermatid) chromatin affects gene expression in the embryo. It would be especially interesting to test whether removal of H3K4me2/3 activating marks from the spermatid chromatin would be able to rescue (downregulate) the upregulation of misregulated genes in spermatid-derived embryos. In the future (when a better annotation of the Xenopus laevis genome is available) it would be worth trying to target the enzymes that modify the chromatin (for example demethylases that remove H3K4me2/3 marks) to specific loci in the genome with the use of CRISPR-Cas9 system (Mali et al. 2013). This system is naturally used by bacteria as a defence mechanism against foreign nucleic acids. It is based on guide RNAs that bring the Cas9 nuclease to the foreign nucleotides for their degradation (Mali et al. 2013). Such guide RNAs can be custom designed to target any desired sequence in the genome and therefore this strategy has been widely applied to obtain gene knock-outs in many different systems (Cho et al. 2013, Cong et al. 2013, Friedland et al. 2013). Interestingly, it has been shown that using custom-designed targeting RNAs and nuclease-dead Cas9 fused to the proteins of interest it was possible to modify the levels of gene expression of the desired gene. For example, it was shown that by creating a fusion between a nuclease-dead Cas9 and a transcriptional transactivator protein and by simultaneously expressing in human cells appropriate guide RNAs, it was possible to selectively activate desired endogenous target genes (Perez-Pinera et al. 2013). Conversely, it was also shown that a another type of Cas9 modification can result in gene repression in bacteria, human and yeast by blocking transcriptional initiation and/or elongation (Gilbert et al. 2013, Qi et al. 2013), confirming that Cas9 system can be successfully applied not only for genome editing, but also for modulating gene expression. It would be therefore interesting to fuse epigenetic modifiers (for example H3K4me2/3 demethylase) to nuclease-dead Cas9 to try rescuing Chapter 8: Discussion 177 (downregulating) the gene expression in spermatid-derived embryos. Doing such experiment in a targeted way (targeting all the 100 misregulated genes by injecting appropriate guide mRNAs into 1-cell stage embryos), would eliminate side effects which would occur if the enzyme would act on the whole chromatin. 8.4. What are the epigenetic changes occurring during spermiogenesis in Xenopus laevis? ChIP-seq experiments described in this thesis probed for repressive H3K27me3 and for activating H3K4me2/3 marks in sperm and spermatids. Even though this experiments provided interesting and unexpected findings (H3K27me3 did not change between spermatids and sperm, whereas the activating H3K4me2/3 marks were more abundant in spermatids, which correlated well with the gene upregulation observed in spermatid-derived embryos), one has to realise that the picture obtained is far from complete. As mentioned in the introduction chapter, there are multiple epigenetic modifications known on DNA: DNA methylation, DNA hydroxymethylation, DNA formylation, DNA carboxylation; and even more histone modifications: methylation, acetylation, phosphorylation, ubiquitylation, sumoylation, ADP ribosylation, crotonylation and many others. Moreover, histone modifications occur on many different residues and different histones, for example both serine and threonine can become phosphorylated, or histones H3 and H4 can both be acetylated (Dawson & Kouzarides 2012). Plus, there are many histone variants existing and also, the positioning of histones itself can affect the chromatin structure. All these epigenetic changes affect each other and they can also attract different binding partners, which can further alter the transcriptional outcome of the target gene (Kouzarides 2007). On top of that, Chapter 8: Discussion 178 it has been shown that protamines in mammalian sperm are also subjected to post- translational modifications (Brunner et al. 2014). All these suggest that the possible combination of various epigenetic marks that could affect the chromatin and the transcriptional state of a given gene is enormous. It is for sure not possible to probe for all the known DNA modifications, histone marks and histone variants. However, the more epigenetic profiles of spermatids and sperm are obtained, the more data can be utilised as an input into the model aiming at predicting gene expression status in the embryo based on the epigenetic states of the paternal chromatin. Experiments currently conducted in the Gurdon laboratory aim to obtain epigenetic profiles of DNA methylation and also trimethylation of lysine 9 of histone H3 (H3K9me3) in sperm and spermatids. Investigations into these marks are of a particular interest, as the proteomic analysis of egg factors binding specifically to sperm, and not to the spermatid chromatin, identified HP1γ and Mbd3 proteins (Fig. 15). These proteins recognise and bind to H3K9me2/3 modification and to methylated cytosine in DNA, respectively. This therefore suggests that those two epigenetic marks: H3K9me3 and DNA methylation are likely to be different between sperm and spermatids. Furthermore, finding the functional connection between the epigenetic marks in sperm and spermatids and their readers from the egg side will provide a valuable addition to the current understanding of the inheritance of the epigenetic marks via the gametes. The results described in this thesis also identify a unique chromatin structure in Xenopus laevis sperm. I showed that MNase digestion of chromatin leads to a release of DNA of three different sizes: ~75bp, ~110bp and ~150bp (Fig. 43). I also showed that ~110bp structure is the most unstable one, as it disappears first upon a treatment with an increasing concentration of a heparin (Fig. 46). What is the identity of these three different structures? What are they composed of? Are the main components of these structures histones, sperm Chapter 8: Discussion 179 basic proteins or perhaps a mixture of both? Do they contain any specific sequence features? For example, the canonical nucleosome size, ~150bp structure could contain gene promoters or developmentally-important genes; or, the ~75bp structure could have mostly intronic sequences or housekeeping genes. The experiments currently ongoing in the Gurdon laboratory (in collaboration with Dr Angela Simeone and Dr Jerome Jullien) aim to address these questions by performing sequence analysis of DNA contained in each of these fragments and by chromatin immunopurifications against histones. 8.5. How the results obtained add to the current knowledge on sperm programming? It is currently known that sperm of different species can bear epigenetic marks on DNA and also on histones (Hammoud et al. 2009, Brykczynska et al. 2010, Wu et al. 2011). It was correlated that genes bearing the repressive marks tend to be repressed during the earliest developmental stages (Brykczynska et al. 2010), whereas those with activating marks tend to be expressed early in embryogenesis (Wu et al. 2011). What is the novelty of the findings described in this thesis? First, it is not possible to manipulate the sperm nucleus to change the histone marks on chromatin. Therefore, comparing sperm and spermatids in the same type of assay allows bypassing the problem of inaccessibility of the sperm chromatin. By using these two different cell types, coming from the same lineage, having the same DNA and chromosome content, but differing in their developmental potential I was able to assess for the first time what is the source of developmental advantage of the sperm. I showed that the developmental advantage of sperm over the spermatids is likely related to the ability to correctly regulate embryonic gene expression. I also managed to link this ability to underlying differences in histone marks between sperm and spermatids: spermatids, which Chapter 8: Discussion 180 overexpress the misregulated genes, have more of activating H3K4me2/3 marks. These results largely extend the currently available knowledge in the field, since even though I was not able to directly manipulate the epigenetic status of the sperm chromatin, I was able to compare the sperm to its direct precursor, that has a different epigenetic state. Furthermore, enzymatic removal of H3K27me3 marks from the parental chromatin at fertilisation proved for the first time that the presence of an epigenetic mark on the parental chromatin is indeed required for a correct regulation of embryonic gene expression. As such, the experiments presented in this thesis significantly advance our current understanding of the sperm programming. Furthermore, presented experiments became a basis for many follow-up projects currently carried in the Gurdon laboratory (as discussed above), therefore I strongly hope that the obtained results will also have interesting future implications. 8.6. Results described in this thesis in the context of the current knowledge on transgenerational inheritance of epigenetic information. Concluding, the results described in this thesis provide experimental evidence supporting the hypothesis that the sperm is epigenetically programmed to regulate embryonic gene expression. This hypothesis is further strengthened by the fact that an incorrect pattern of histone modifications or DNA methylation was associated with cases of idiopathic infertility in humans (Hammoud et al. 2010, Hammoud et al. 2011). Furthermore, it was reported in mammals that epigenetic traits can be transgenerationally inherited from the father to the offspring (Braunschweig et al. 2012, Daxinger & Whitelaw 2012, Lambrot et al. 2013, Padmanabhan et al. 2013, Vassoler et al. 2013, Dias & Ressler 2014). For example, it was shown in mice that offspring of males fed on a low-protein diet had elevated expression of many hepatic genes, which was linked to changes in DNA methylation at the promoters of Chapter 8: Discussion 181 these genes (Carone et al. 2010). In another example it was shown that paternal obesity in mice is transgenerationally inherited up to F2 generation and that this phenomenon was correlated with global changes in DNA methylation patterns and with altered expression profiles of mRNAs and microRNAs in the testes of the F0 obese male mouse (Fullston et al. 2013). There is therefore growing evidence for the existence and importance of transgenerational inheritance of epigenetic traits via the gametes. Not surprisingly, this subject receives more and more of the media attention, as the possibility that the environmental cues acting on the sperm/sperm progenitors can be transmitted and affect the phenotype of the offspring is revolutionising the current view based on genetic mutations as the main source of evolutionary adaptation (Grossniklaus et al. 2013, Hughes 2014, Kaiser 2014, Szyf 2014). Further research into the subject is needed not only because of its impact on the basic biology and our understanding of the non-genetic inheritance, but also because it is necessary to get insight into the potential effects of various environmental cues on the future fitness of the offspring from the perspective of human health. This is especially important as many of the reports on the transgenerational inheritance of epigenetic traits in animals point to the inheritance of metabolic adaptations. For example, it is crucial to understand the consequences of the diet on the health of the future offspring, especially in light of the growing numbers of obesity cases in humans. Furthermore, it was also reported that exposure of pregnant female rats to commonly used environmental toxins: vinclozin (fungicide) or methoxychlor (insecticide) induced transgenerational inheritance of reduced male fertility (complete infertility in 8% of the cases) in male mice in all subsequent generations tested, which was correlated with changes in global DNA methylation patterns transmitted through the male germline (Anway et al. 2005). Therefore, a better understanding of agents causing the heritable changes in the germline epigenome, as well as the mechanisms of the inheritance of such changes is clearly needed. Chapter 8: Discussion 182 I hope that the results presented in this thesis will help our understanding of the mechanisms of transgenerational inheritance of epigenetic traits in general. I believe that Xenopus, in which it is possible to generate haploid sperm- and spermatid-derived embryos, to assess their developmental potential, transcriptional capacity, and manipulate the epigenetic state of chromatin, is a well suited model organism for further investigations of the nature of the sperm programming. Chapter 9: References 183 Chapter 9 References Abe S 1988 Cell-Culture of Spermatogenic Cells from Amphibians. Development Growth & Differentiation 30 209-218. Abe S & Hiyoshi H 1991 Synthesis of sperm-specific basic nuclear proteins (SPs) in cultured spermatids from Xenopus laevis. Exp Cell Res 194 90-94. Agger K, Cloos PA, Christensen J, Pasini D, Rose S, Rappsilber J, Issaeva I, Canaani E, Salcini AE & Helin K 2007 UTX and JMJD3 are histone H3K27 demethylases involved in HOX gene regulation and development. Nature 449 731-734. Akkers RC, van Heeringen SJ, Jacobi UG, Janssen-Megens EM, Francoijs KJ, Stunnenberg HG & Veenstra GJ 2009 A hierarchy of H3K4me3 and H3K27me3 acquisition in spatial gene regulation in Xenopus embryos. Dev Cell 17 425-434. Alabert C, Bukowski-Wills JC, Lee SB, Kustatscher G, Nakamura K, de Lima Alves F, Menard P, Mejlvang J, Rappsilber J & Groth A 2014 Nascent chromatin capture proteomics determines chromatin dynamics during DNA replication and identifies unknown fork components. Nat Cell Biol 16 281-293. Alexa A, Rahnenfuhrer J & Lengauer T 2006 Improved scoring of functional groups from gene expression data by decorrelating GO graph structure. Bioinformatics 22 1600-1607. Alexeyenko A, Tamas I, Liu G & Sonnhammer EL 2006 Automatic clustering of orthologs and inparalogs shared by multiple proteomes. Bioinformatics 22 e9-15. Ang YS, Tsai SY, Lee DF, Monk J, Su J, Ratnakumar K, Ding J, Ge Y, Darr H, Chang B, Wang J, Rendl M, Bernstein E, Schaniel C & Lemischka IR 2011 Wdr5 mediates self-renewal and reprogramming via the embryonic stem cell core transcriptional network. Cell 145 183-197. Anway MD, Cupp AS, Uzumcu M & Skinner MK 2005 Epigenetic transgenerational actions of endocrine disruptors and male fertility. Science 308 1466-1469. Aoki F, Worrad DM & Schultz RM 1997 Regulation of transcriptional activity during the first and second cell cycles in the preimplantation mouse embryo. Dev Biol 181 296-307. Arpanahi A, Brinkworth M, Iles D, Krawetz SA, Paradowska A, Platts AE, Saida M, Steger K, Tedder P & Miller D 2009 Endonuclease-sensitive regions of human spermatozoal chromatin are highly enriched in promoter and CTCF binding sequences. Genome Res 19 1338-1349. Bai L & Morozov AV 2010 Gene regulation by nucleosome positioning. Trends Genet 26 476-483. Balhorn R 2007 The protamine family of sperm nuclear proteins. Genome Biol 8 227. Banaszynski LA, Wen D, Dewell S, Whitcomb SJ, Lin M, Diaz N, Elsasser SJ, Chapgier A, Goldberg AD, Canaani E, Rafii S, Zheng D & Allis CD 2013 Hira-dependent histone H3.3 deposition facilitates PRC2 recruitment at developmental loci in ES cells. Cell 155 107-120. Bannister AJ, Zegerman P, Partridge JF, Miska EA, Thomas JO, Allshire RC & Kouzarides T 2001 Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410 120-124. Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, Wei G, Chepelev I & Zhao K 2007 High- resolution profiling of histone methylations in the human genome. Cell 129 823-837. Becker M, Becker A, Miyara F, Han Z, Kihara M, Brown DT, Hager GL, Latham K, Adashi EY & Misteli T 2005 Differential in vivo binding dynamics of somatic and oocyte-specific linker histones in oocytes and during ES cell nuclear transfer. Mol Biol Cell 16 3887-3895. Chapter 9: References 184 Benvenuto G, Carpentieri ML, Salvatore P, Cindolo L, Bruni CB & Chiariotti L 1996 Cell-specific transcriptional regulation and reactivation of galectin-1 gene expression are controlled by DNA methylation of the promoter region. Mol Cell Biol 16 2736-2743. Bernstein BE, Humphrey EL, Erlich RL, Schneider R, Bouman P, Liu JS, Kouzarides T & Schreiber SL 2002 Methylation of histone H3 Lys 4 in coding regions of active genes. Proc Natl Acad Sci U S A 99 8695-8700. Bernstein BE, Kamal M, Lindblad-Toh K, Bekiranov S, Bailey DK, Huebert DJ, McMahon S, Karlsson EK, Kulbokas EJ, 3rd, Gingeras TR, Schreiber SL & Lander ES 2005 Genomic maps and comparative analysis of histone modifications in human and mouse. Cell 120 169-181. Bhaskara S, Jacques V, Rusche JR, Olson EN, Cairns BR & Chandrasekharan MB 2013 Histone deacetylases 1 and 2 maintain S-phase chromatin and DNA replication fork progression. Epigenetics Chromatin 6 27. Bitgood MJ & McMahon AP 1995 Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. Dev Biol 172 126-138. Blow JJ & Laskey RA 1986 Initiation of DNA replication in nuclei and purified DNA by a cell-free extract of Xenopus eggs. Cell 47 577-587. Bochkarev A, Pfuetzner RA, Edwards AM & Frappier L 1997 Structure of the single-stranded-DNA- binding domain of replication protein A bound to DNA. Nature 385 176-181. Bogdanovic O & Veenstra GJ 2009 DNA methylation and methyl-CpG binding proteins: developmental requirements and function. Chromosoma 118 549-565. Boiani M, Eckardt S, Scholer HR & McLaughlin KJ 2002 Oct4 distribution and level in mouse clones: consequences for pluripotency. Genes Dev 16 1209-1219. Bonasio R, Tu S & Reinberg D 2010 Molecular signals of epigenetic states. Science 330 612-616. Bortvin A, Eggan K, Skaletsky H, Akutsu H, Berry DL, Yanagimachi R, Page DC & Jaenisch R 2003 Incomplete reactivation of Oct4-related genes in mouse embryos cloned from somatic nuclei. Development 130 1673-1680. Bouniol-Baly C, Nguyen E, Besombes D & Debey P 1997 Dynamic organization of DNA replication in one-cell mouse embryos: relationship to transcriptional activation. Exp Cell Res 236 201-211. Bourc'his D, Le Bourhis D, Patin D, Niveleau A, Comizzoli P, Renard JP & Viegas-Pequignot E 2001 Delayed and incomplete reprogramming of chromosome methylation patterns in bovine cloned embryos. Curr Biol 11 1542-1546. Boyes J & Bird A 1991 DNA methylation inhibits transcription indirectly via a methyl-CpG binding protein. Cell 64 1123-1134. Boyes J & Bird A 1992 Repression of genes by DNA methylation depends on CpG density and promoter strength: evidence for involvement of a methyl-CpG binding protein. EMBO J 11 327-333. Braunschweig M, Jagannathan V, Gutzwiller A & Bee G 2012 Investigations on transgenerational epigenetic response down the male line in F2 pigs. PLoS One 7 e30583. Brunner AM, Nanni P & Mansuy IM 2014 Epigenetic marking of sperm by post-translational modification of histones and protamines. Epigenetics Chromatin 7 2. Brykczynska U, Hisano M, Erkek S, Ramos L, Oakeley EJ, Roloff TC, Beisel C, Schubeler D, Stadler MB & Peters AH 2010 Repressive and active histone methylation mark distinct promoters in human and mouse spermatozoa. Nat Struct Mol Biol 17 679-687. Bui HT, Seo HJ, Park MR, Park JY, Thuan NV, Wakayama T & Kim JH 2011 Histone deacetylase inhibition improves activation of ribosomal RNA genes and embryonic nucleolar reprogramming in cloned mouse embryos. Biol Reprod 85 1048-1056. Cao R, Wang L, Wang H, Xia L, Erdjument-Bromage H, Tempst P, Jones RS & Zhang Y 2002 Role of histone H3 lysine 27 methylation in Polycomb-group silencing. Science 298 1039-1043. Carone BR, Fauquier L, Habib N, Shea JM, Hart CE, Li R, Bock C, Li C, Gu H, Zamore PD, Meissner A, Weng Z, Hofmann HA, Friedman N & Rando OJ 2010 Paternally induced transgenerational Chapter 9: References 185 environmental reprogramming of metabolic gene expression in mammals. Cell 143 1084- 1096. Cary PD, Crane-Robinson C, Bradbury EM & Dixon GH 1982 Effect of acetylation on the binding of N-terminal peptides of histone H4 to DNA. Eur J Biochem 127 137-143. Chahal SS, Matthews HR & Bradbury EM 1980 Acetylation of histone H4 and its role in chromatin structure and function. Nature 287 76-79. Chan MM, Smith ZD, Egli D, Regev A & Meissner A 2012 Mouse ooplasm confers context-specific reprogramming capacity. Nat Genet 44 978-980. Chen C, Yang MC & Yang TP 2001a Evidence that silencing of the HPRT promoter by DNA methylation is mediated by critical CpG sites. J Biol Chem 276 320-328. Chen GC, Guan LS, Yu JH, Li GC, Choi Kim HR & Wang ZY 2001b Rb-associated protein 46 (RbAp46) inhibits transcriptional transactivation mediated by BRCA1. Biochem Biophys Res Commun 284 507-514. Chen X, Hu T, Liang G, Yang M, Zong S, Miao S, Koide SS & Wang L 2008 A novel testis protein, RSB- 66, interacting with INCA1 (inhibitor of Cdk interacting with cyclin A1). Biochem Cell Biol 86 345-351. Cho C, Jung-Ha H, Willis WD, Goulding EH, Stein P, Xu Z, Schultz RM, Hecht NB & Eddy EM 2003 Protamine 2 deficiency leads to sperm DNA damage and embryo death in mice. Biol Reprod 69 211-217. Cho SW, Kim S, Kim JM & Kim JS 2013 Targeted genome engineering in human cells with the Cas9 RNA-guided endonuclease. Nature biotechnology 31 230-232. Chowdhury I, Thompson WE, Welch C, Thomas K & Matthews R 2013 Prohibitin (PHB) inhibits apoptosis in rat granulosa cells (GCs) through the extracellular signal-regulated kinase 1/2 (ERK1/2) and the Bcl family of proteins. Apoptosis 18 1513-1525. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, Marraffini LA & Zhang F 2013 Multiplex genome engineering using CRISPR/Cas systems. Science 339 819-823. Core LJ, Waterfall JJ & Lis JT 2008 Nascent RNA sequencing reveals widespread pausing and divergent initiation at human promoters. Science 322 1845-1848. Creyghton MP, Markoulaki S, Levine SS, Hanna J, Lodato MA, Sha K, Young RA, Jaenisch R & Boyer LA 2008 H2AZ is enriched at polycomb complex target genes in ES cells and is necessary for lineage commitment. Cell 135 649-661. Czermin B, Melfi R, McCabe D, Seitz V, Imhof A & Pirrotta V 2002 Drosophila enhancer of Zeste/ESC complexes have a histone H3 methyltransferase activity that marks chromosomal Polycomb sites. Cell 111 185-196. Dawson MA & Kouzarides T 2012 Cancer epigenetics: from mechanism to therapy. Cell 150 12-27. Daxinger L & Whitelaw E 2012 Understanding transgenerational epigenetic inheritance via the gametes in mammals. Nat Rev Genet 13 153-162. de Mateo S, Castillo J, Estanyol JM, Ballesca JL & Oliva R 2011 Proteomic characterization of the human sperm nucleus. Proteomics 11 2714-2726. de Rooij DG 2001 Proliferation and differentiation of spermatogonial stem cells. Reproduction 121 347-354. Dey A, Chitsaz F, Abbasi A, Misteli T & Ozato K 2003 The double bromodomain protein Brd4 binds to acetylated chromatin during interphase and mitosis. Proc Natl Acad Sci U S A 100 8758- 8763. Dey A, Nishiyama A, Karpova T, McNally J & Ozato K 2009 Brd4 marks select genes on mitotic chromatin and directs postmitotic transcription. Mol Biol Cell 20 4899-4909. Dias BG & Ressler KJ 2014 Parental olfactory experience influences behavior and neural structure in subsequent generations. Nat Neurosci 17 89-96. Doerfler W 2008 In pursuit of the first recognized epigenetic signal--DNA methylation: a 1976 to 2008 synopsis. Epigenetics 3 125-133. Chapter 9: References 186 Draker R, Ng MK, Sarcinella E, Ignatchenko V, Kislinger T & Cheung P 2012 A combination of H2A.Z and H4 acetylation recruits Brd2 to chromatin during transcriptional activation. PLoS Genet 8 e1003047. du Chene I, Basyuk E, Lin YL, Triboulet R, Knezevich A, Chable-Bessia C, Mettling C, Baillat V, Reynes J, Corbeau P, Bertrand E, Marcello A, Emiliani S, Kiernan R & Benkirane M 2007 Suv39H1 and HP1gamma are responsible for chromatin-mediated HIV-1 transcriptional silencing and post-integration latency. EMBO J 26 424-435. Eggan K, Baldwin K, Tackett M, Osborne J, Gogos J, Chess A, Axel R & Jaenisch R 2004 Mice cloned from olfactory sensory neurons. Nature 428 44-49. Elling U, Klasen C, Eisenberger T, Anlag K & Treier M 2006 Murine inner cell mass-derived lineages depend on Sall4 function. Proc Natl Acad Sci U S A 103 16319-16324. Elsdale TR, Gurdon JB & Fischberg M 1960 A description of the technique for nuclear transplantation in Xenopus laevis. J Embryol Exp Morphol 8 437-444. Enninga J, Levay A & Fontoura BM 2003 Sec13 shuttles between the nucleus and the cytoplasm and stably interacts with Nup96 at the nuclear pore complex. Mol Cell Biol 23 7271-7284. Erdel F & Rippe K 2011 Chromatin remodelling in mammalian cells by ISWI-type complexes--where, when and why? FEBS J 278 3608-3618. Erkek S, Hisano M, Liang CY, Gill M, Murr R, Dieker J, Schubeler D, Vlag J, Stadler MB & Peters AH 2013 Molecular determinants of nucleosome retention at CpG-rich sequences in mouse spermatozoa. Nat Struct Mol Biol 20 868-875. Farcas AM, Blackledge NP, Sudbery I, Long HK, McGouran JF, Rose NR, Lee S, Sims D, Cerase A, Sheahan TW, Koseki H, Brockdorff N, Ponting CP, Kessler BM & Klose RJ 2012 KDM2B links the Polycomb Repressive Complex 1 (PRC1) to recognition of CpG islands. Elife 1 e00205. Farthing CR, Ficz G, Ng RK, Chan CF, Andrews S, Dean W, Hemberger M & Reik W 2008 Global mapping of DNA methylation in mouse promoters reveals epigenetic reprogramming of pluripotency genes. PLoS Genet 4 e1000116. Fontoura BM, Blobel G & Matunis MJ 1999 A conserved biogenesis pathway for nucleoporins: proteolytic processing of a 186-kilodalton precursor generates Nup98 and the novel nucleoporin, Nup96. J Cell Biol 144 1097-1112. Fontoura BM, Dales S, Blobel G & Zhong H 2001 The nucleoporin Nup98 associates with the intranuclear filamentous protein network of TPR. Proc Natl Acad Sci U S A 98 3208-3213. Frehlick LJ, Prado A, Calestagne-Morelli A & Ausio J 2007 Characterization of the PL-I-related SP2 protein from Xenopus. Biochemistry 46 12700-12708. Friedland AE, Tzur YB, Esvelt KM, Colaiacovo MP, Church GM & Calarco JA 2013 Heritable genome editing in C. elegans via a CRISPR-Cas9 system. Nature methods 10 741-743. Friedle H & Knochel W 2002 Cooperative interaction of Xvent-2 and GATA-2 in the activation of the ventral homeobox gene Xvent-1B. J Biol Chem 277 23872-23881. Fukuda A, Cao F, Morita S, Yamada K, Jincho Y, Tane S, Sotomaru Y & Kono T 2010 Identification of inappropriately reprogrammed genes by large-scale transcriptome analysis of individual cloned mouse blastocysts. PLoS One 5 e11274. Fullston T, Ohlsson Teague EM, Palmer NO, DeBlasio MJ, Mitchell M, Corbett M, Print CG, Owens JA & Lane M 2013 Paternal obesity initiates metabolic disturbances in two generations of mice with incomplete penetrance to the F2 generation and alters the transcriptional profile of testis and sperm microRNA content. FASEB J 27 4226-4243. Gaggioli V, Le Viet B, Germe T & Hyrien O 2013 DNA topoisomerase IIalpha controls replication origin cluster licensing and firing time in Xenopus egg extracts. Nucleic Acids Res. Ganier O, Bocquet S, Peiffer I, Brochard V, Arnaud P, Puy A, Jouneau A, Feil R, Renard JP & Mechali M 2011 Synergic reprogramming of mammalian cells by combined exposure to mitotic Xenopus egg extracts and transcription factors. Proc Natl Acad Sci U S A 108 17331-17336. Chapter 9: References 187 Gao S, Chung YG, Parseghian MH, King GJ, Adashi EY & Latham KE 2004 Rapid H1 linker histone transitions following fertilization or somatic cell nuclear transfer: evidence for a uniform developmental program in mice. Dev Biol 266 62-75. Gaucher J, Boussouar F, Montellier E, Curtet S, Buchou T, Bertrand S, Hery P, Jounier S, Depaux A, Vitte AL, Guardiola P, Pernet K, Debernardi A, Lopez F, Holota H, Imbert J, Wolgemuth DJ, Gerard M, Rousseaux S & Khochbin S 2012 Bromodomain-dependent stage-specific male genome programming by Brdt. EMBO J 31 3809-3820. Gaucher J, Reynoird N, Montellier E, Boussouar F, Rousseaux S & Khochbin S 2010 From meiosis to postmeiotic events: the secrets of histone disappearance. FEBS J 277 599-604. Gaudet F, Hodgson JG, Eden A, Jackson-Grusby L, Dausman J, Gray JW, Leonhardt H & Jaenisch R 2003 Induction of tumors in mice by genomic hypomethylation. Science 300 489-492. Gilbert LA, Larson MH, Morsut L, Liu Z, Brar GA, Torres SE, Stern-Ginossar N, Brandman O, Whitehead EH, Doudna JA, Lim WA, Weissman JS & Qi LS 2013 CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 154 442-451. Gilbert SF 2010 Developmental biology. Sunderland, Mass.: Sinauer Associates. Gillespie PJ & Blow JJ 2000 Nucleoplasmin-mediated chromatin remodelling is required for Xenopus sperm nuclei to become licensed for DNA replication. Nucleic Acids Res 28 472-480. Gillespie PJ, Gambus A & Blow JJ 2012 Preparation and use of Xenopus egg extracts to study DNA replication and chromatin associated proteins. Methods 57 203-213. Godde JS & Ura K 2008 Cracking the enigmatic linker histone code. J Biochem 143 287-293. Goldberg AD, Banaszynski LA, Noh KM, Lewis PW, Elsaesser SJ, Stadler S, Dewell S, Law M, Guo X, Li X, Wen D, Chapgier A, DeKelver RC, Miller JC, Lee YL, Boydston EA, Holmes MC, Gregory PD, Greally JM, Rafii S, Yang C, Scambler PJ, Garrick D, Gibbons RJ, Higgs DR, Cristea IM, Urnov FD, Zheng D & Allis CD 2010 Distinct factors control histone variant H3.3 localization at specific genomic regions. Cell 140 678-691. Govin J, Caron C, Lestrat C, Rousseaux S & Khochbin S 2004 The role of histones in chromatin remodelling during mammalian spermiogenesis. Eur J Biochem 271 3459-3469. Govin J, Lestrat C, Caron C, Pivot-Pajot C, Rousseaux S & Khochbin S 2006 Histone acetylation- mediated chromatin compaction during mouse spermatogenesis. Ernst Schering Res Found Workshop155-172. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q, Chen Z, Mauceli E, Hacohen N, Gnirke A, Rhind N, di Palma F, Birren BW, Nusbaum C, Lindblad-Toh K, Friedman N & Regev A 2011 Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol 29 644-652. Green MR & Sambrook J 2012 Molecular cloning : a laboratory manual. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press. Grossniklaus U, Kelly WG, Ferguson-Smith AC, Pembrey M & Lindquist S 2013 Transgenerational epigenetic inheritance: how important is it? Nat Rev Genet 14 228-235. Gurdon JB 1960 The Effects of Ultraviolet Irradiation on Uncleaved Eggs of Xenopus laevis. Quarterly Journal of Microscopical Science 101 299-311. Gurdon JB 1962 The developmental capacity of nuclei taken from intestinal epithelium cells of feeding tadpoles. J Embryol Exp Morphol 10 622-640. Gurdon JB 1976 Injected nuclei in frog oocytes: fate, enlargement, and chromatin dispersal. J Embryol Exp Morphol 36 523-540. Gurdon JB, Elsdale TR & Fischberg M 1958 Sexually mature individuals of Xenopus laevis from the transplantation of single somatic nuclei. Nature 182 64-65. Gurdon JB & Laskey RA 1970 The transplantation of nuclei from single cultured cells into enucleate frogs' eggs. J Embryol Exp Morphol 24 227-248. Gurdon JB & Uehlinger V 1966 "Fertile" intestine nuclei. Nature 210 1240-1241. Chapter 9: References 188 Hagarman JA, Motley MP, Kristjansdottir K & Soloway PD 2013 Coordinate regulation of DNA methylation and H3K27me3 in mouse embryonic stem cells. PLoS One 8 e53880. Hah N, Kolkman A, Ruhl DD, Pijnappel WW, Heck AJ, Timmers HT & Kraus WL 2010 A role for BAF57 in cell cycle-dependent transcriptional regulation by the SWI/SNF chromatin remodeling complex. Cancer Res 70 4402-4411. Hajkova P, Jeffries SJ, Lee C, Miller N, Jackson SP & Surani MA 2010 Genome-wide reprogramming in the mouse germ line entails the base excision repair pathway. Science 329 78-82. Halley-Stott RP, Pasque V, Astrand C, Miyamoto K, Simeoni I, Jullien J & Gurdon JB 2010 Mammalian nuclear transplantation to Germinal Vesicle stage Xenopus oocytes - a method for quantitative transcriptional reprogramming. Methods 51 56-65. Hamilton L 1963 An experimental analysis of the development of the haploid syndrome in embryos of Xenopus laevis. J Embryol Exp Morphol 11 267-278. Hammoud SS, Nix DA, Hammoud AO, Gibson M, Cairns BR & Carrell DT 2011 Genome-wide analysis identifies changes in histone retention and epigenetic modifications at developmental and imprinted gene loci in the sperm of infertile men. Hum Reprod 26 2558-2569. Hammoud SS, Nix DA, Zhang H, Purwar J, Carrell DT & Cairns BR 2009 Distinctive chromatin in human sperm packages genes for embryo development. Nature 460 473-478. Hammoud SS, Purwar J, Pflueger C, Cairns BR & Carrell DT 2010 Alterations in sperm DNA methylation patterns at imprinted loci in two classes of infertility. Fertil Steril 94 1728-1733. Han J, Li Q, McCullough L, Kettelkamp C, Formosa T & Zhang Z 2010 Ubiquitylation of FACT by the cullin-E3 ligase Rtt101 connects FACT to DNA replication. Genes Dev 24 1485-1490. Hargreaves DC, Horng T & Medzhitov R 2009 Control of inducible gene expression by signal- dependent transcriptional elongation. Cell 138 129-145. Hassig CA, Tong JK, Fleischer TC, Owa T, Grable PG, Ayer DE & Schreiber SL 1998 A role for histone deacetylase activity in HDAC1-mediated transcriptional repression. Proc Natl Acad Sci U S A 95 3519-3524. Hazzouri M, Pivot-Pajot C, Faure AK, Usson Y, Pelletier R, Sele B, Khochbin S & Rousseaux S 2000 Regulated hyperacetylation of core histones during mouse spermatogenesis: involvement of histone deacetylases. Eur J Cell Biol 79 950-960. He J, Shen L, Wan M, Taranova O, Wu H & Zhang Y 2013 Kdm2b maintains murine embryonic stem cell status by recruiting PRC1 complex to CpG islands of developmental genes. Nat Cell Biol 15 373-384. Henricksen LA, Carter T, Dutta A & Wold MS 1996 Phosphorylation of human replication protein A by the DNA-dependent protein kinase is involved in the modulation of DNA replication. Nucleic Acids Res 24 3107-3112. Hildebrand CE, Gurley LR, Tobey RA & Walters RA 1977 Action of heparin on mammalian nuclei. I. Differential extraction of histone H1 and cooperative removal of histones from chromatin. Biochim Biophys Acta 477 295-311. Hirasawa R, Matoba S, Inoue K & Ogura A 2013 Somatic donor cell type correlates with embryonic, but not extra-embryonic, gene expression in postimplantation cloned embryos. PLoS One 8 e76422. Hisano M, Erkek S, Dessus-Babus S, Ramos L, Stadler MB & Peters AH 2013 Genome-wide chromatin analysis in mature mouse and human spermatozoa. Nat Protoc 8 2449-2470. Hiyoshi H, Uno S, Yokota T, Katagiri C, Nishida H, Takai M, Agata K, Eguchi G & Abe S 1991 Isolation of cDNA for a Xenopus sperm-specific basic nuclear protein (SP4) and evidence for expression of SP4 mRNA in primary spermatocytes. Exp Cell Res 194 95-99. Hochedlinger K & Jaenisch R 2002 Monoclonal mice generated by nuclear transfer from mature B and T donor cells. Nature 415 1035-1038. Hon GC, Rajagopal N, Shen Y, McCleary DF, Yue F, Dang MD & Ren B 2013 Epigenetic memory at embryonic enhancers identified in DNA methylation maps from adult mouse tissues. Nat Genet 45 1198-1206. Chapter 9: References 189 Hsieh CL 1994 Dependence of transcriptional repression on CpG methylation density. Mol Cell Biol 14 5487-5494. Hughes V 2014 Epigenetics: The sins of the father. Nature 507 22-24. Hulstrand AM, Schneider PN & Houston DW 2010 The use of antisense oligonucleotides in Xenopus oocytes. Methods 51 75-81. Inoue A, Ogushi S, Saitou M, Suzuki MG & Aoki F 2011 Involvement of mouse nucleoplasmin 2 in the decondensation of sperm chromatin after fertilization. Biol Reprod 85 70-77. Iwano H, Nakamura M & Tajima S 2004 Xenopus MBD3 plays a crucial role in an early stage of development. Dev Biol 268 416-428. Izzo A, Kamieniarz K & Schneider R 2008 The histone H1 family: specific members, specific functions? Biol Chem 389 333-343. Jang MK, Mochizuki K, Zhou M, Jeong HS, Brady JN & Ozato K 2005 The bromodomain protein Brd4 is a positive regulatory component of P-TEFb and stimulates RNA polymerase II-dependent transcription. Mol Cell 19 523-534. Jenuwein T & Allis CD 2001 Translating the histone code. Science 293 1074-1080. Jeppesen P & Turner BM 1993 The inactive X chromosome in female mammals is distinguished by a lack of histone H4 acetylation, a cytogenetic marker for gene expression. Cell 74 281-289. Jiang CL, Jin SG & Pfeifer GP 2004 MBD3L1 is a transcriptional repressor that interacts with methyl- CpG-binding protein 2 (MBD2) and components of the NuRD complex. J Biol Chem 279 52456-52464. Jiang L, Zhang J, Wang JJ, Wang L, Zhang L, Li G, Yang X, Ma X, Sun X, Cai J, Huang X, Yu M, Wang X, Liu F, Wu CI, He C, Zhang B, Ci W & Liu J 2013 Sperm, but not oocyte, DNA methylome is inherited by zebrafish early embryos. Cell 153 773-784. Jin C & Felsenfeld G 2007 Nucleosome stability mediated by histone variants H3.3 and H2A.Z. Genes Dev 21 1519-1529. Jones PL, Veenstra GJ, Wade PA, Vermaak D, Kass SU, Landsberger N, Strouboulis J & Wolffe AP 1998 Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat Genet 19 187-191. Jouve C, Palmeirim I, Henrique D, Beckers J, Gossler A, Ish-Horowicz D & Pourquie O 2000 Notch signalling is required for cyclic expression of the hairy-like gene HES1 in the presomitic mesoderm. Development 127 1421-1429. Jullien J, Astrand C, Halley-Stott RP, Garrett N & Gurdon JB 2010 Characterization of somatic cell nuclear reprogramming by oocytes in which a linker histone is required for pluripotency gene reactivation. Proc Natl Acad Sci U S A 107 5483-5488. Jullien J, Astrand C, Szenker E, Garrett N, Almouzni G & Gurdon JB 2012 HIRA dependent H3.3 deposition is required for transcriptional reprogramming following nuclear transfer to Xenopus oocytes. Epigenetics Chromatin 5 17. Kaiser J 2014 The epigenetics heretic. Science 343 361-363. Kang YK, Koo DB, Park JS, Choi YH, Chung AS, Lee KK & Han YM 2001 Aberrant methylation of donor genome in cloned bovine embryos. Nat Genet 28 173-177. Kanno T, Kanno Y, Siegel RM, Jang MK, Lenardo MJ & Ozato K 2004 Selective recognition of acetylated histones by bromodomain proteins visualized in living cells. Mol Cell 13 33-43. Kelley C, Yee K, Harland R & Zon LI 1994 Ventral expression of GATA-1 and GATA-2 in the Xenopus embryo defines induction of hematopoietic mesoderm. Dev Biol 165 193-205. Kent WJ 2002 BLAT--the BLAST-like alignment tool. Genome Res 12 656-664. Kikyo N, Wade PA, Guschin D, Ge H & Wolffe AP 2000 Active remodeling of somatic nuclei in egg cytoplasm by the nucleosomal ATPase ISWI. Science 289 2360-2362. Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R & Salzberg SL 2013 TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol 14 R36. Chapter 9: References 190 Kim J, Lin JJ, Xu RH & Kung HF 1998 Mesoderm induction by heterodimeric AP-1 (c-Jun and c-Fos) and its involvement in mesoderm formation through the embryonic fibroblast growth factor/Xbra autocatalytic loop during the early development of Xenopus embryos. J Biol Chem 273 1542-1550. Kimelman D, Kirschner M & Scherson T 1987 The events of the midblastula transition in Xenopus are regulated by changes in the cell cycle. Cell 48 399-407. Kinoshita M, Hatada S, Asashima M & Noda M 1994 HMG-X, a Xenopus gene encoding an HMG1 homolog, is abundantly expressed in the developing nervous system. FEBS Lett 352 191-196. Kishigami S, Van Thuan N, Hikichi T, Ohta H, Wakayama S, Mizutani E & Wakayama T 2006 Epigenetic abnormalities of the mouse paternal zygotic genome associated with microinsemination of round spermatids. Dev Biol 289 195-205. Kishigami S, Wakayama S, Nguyen VT & Wakayama T 2004 Similar time restriction for intracytoplasmic sperm injection and round spermatid injection into activated oocytes for efficient offspring production. Biol Reprod 70 1863-1869. Kouzarides T 2007 Chromatin modifications and their function. Cell 128 693-705. Krawetz SA 2005 Paternal contribution: new insights and future challenges. Nat Rev Genet 6 633- 642. Kuzmichev A, Nishioka K, Erdjument-Bromage H, Tempst P & Reinberg D 2002 Histone methyltransferase activity associated with a human multiprotein complex containing the Enhancer of Zeste protein. Genes Dev 16 2893-2905. Kwon H, Imbalzano AN, Khavari PA, Kingston RE & Green MR 1994 Nucleosome disruption and enhancement of activator binding by a human SW1/SNF complex. Nature 370 477-481. Kwon SH & Workman JL 2011 The changing faces of HP1: From heterochromatin formation and gene silencing to euchromatic gene expression: HP1 acts as a positive regulator of transcription. Bioessays 33 280-289. Labit H, Goldar A, Guilbaud G, Douarche C, Hyrien O & Marheineke K 2008 A simple and optimized method of producing silanized surfaces for FISH and replication mapping on combed DNA fibers. Biotechniques 45 649-652, 654, 656-648. Lachner M, O'Carroll D, Rea S, Mechtler K & Jenuwein T 2001 Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins. Nature 410 116-120. Laherty CD, Yang WM, Sun JM, Davie JR, Seto E & Eisenman RN 1997 Histone deacetylases associated with the mSin3 corepressor mediate mad transcriptional repression. Cell 89 349- 356. Lambrot R, Xu C, Saint-Phar S, Chountalos G, Cohen T, Paquet M, Suderman M, Hallett M & Kimmins S 2013 Low paternal dietary folate alters the mouse sperm epigenome and is associated with negative pregnancy outcomes. Nat Commun 4 2889. Lamond AI & Earnshaw WC 1998 Structure and function in the nucleus. Science 280 547-553. Laskey R 2005 Solving mysteries of DNA replication and frog cloning. Cell 123 760-762. Lassalle B & Testart J 1991 Sequential transformations of human sperm nucleus in human egg. J Reprod Fertil 91 393-402. Le Guezennec X, Vermeulen M, Brinkman AB, Hoeijmakers WA, Cohen A, Lasonder E & Stunnenberg HG 2006 MBD2/NuRD and MBD3/NuRD, two distinct complexes with different biochemical and functional properties. Mol Cell Biol 26 843-851. Lee CK, Shibata Y, Rao B, Strahl BD & Lieb JD 2004 Evidence for nucleosome depletion at active regulatory regions genome-wide. Nat Genet 36 900-905. Lee DY, Hayes JJ, Pruss D & Wolffe AP 1993 A positive role for histone acetylation in transcription factor access to nucleosomal DNA. Cell 72 73-84. Lee SY, Yoon J, Lee HS, Hwang YS, Cha SW, Jeong CH, Kim JI, Park JB, Lee JY, Kim S, Park MJ, Dong Z & Kim J 2011 The function of heterodimeric AP-1 comprised of c-Jun and c-Fos in activin mediated Spemann organizer gene expression. PLoS One 6 e21796. Chapter 9: References 191 Lehnertz B, Ueda Y, Derijck AA, Braunschweig U, Perez-Burgos L, Kubicek S, Chen T, Li E, Jenuwein T & Peters AH 2003 Suv39h-mediated histone H3 lysine 9 methylation directs DNA methylation to major satellite repeats at pericentric heterochromatin. Curr Biol 13 1192- 1200. Lemaitre JM, Danis E, Pasero P, Vassetzky Y & Mechali M 2005 Mitotic remodeling of the replicon and chromosome structure. Cell 123 787-801. Lemaitre JM, Vassetzky Y & Mechali M 2001 Analysis of Chromatin Assembly, Chromatin Domains, and DNA Replication Using Xenopus Systems. LeRoy G, Rickards B & Flint SJ 2008 The double bromodomain proteins Brd2 and Brd3 couple histone acetylation to transcription. Mol Cell 30 51-60. Li A, Maffey AH, Abbott WD, Conde e Silva N, Prunell A, Siino J, Churikov D, Zalensky AO & Ausio J 2005 Characterization of nucleosomes consisting of the human testis/sperm-specific histone H2B variant (hTSH2B). Biochemistry 44 2529-2535. Li H & Durbin R 2009 Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25 1754-1760. Li W & Godzik A 2006 Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22 1658-1659. Li Z, Gadue P, Chen K, Jiao Y, Tuteja G, Schug J, Li W & Kaestner KH 2012 Foxa2 and H2A.Z mediate nucleosome depletion during embryonic stem cell differentiation. Cell 151 1608-1616. Lindeman LC, Andersen IS, Reiner AH, Li N, Aanes H, Ostrup O, Winata C, Mathavan S, Muller F, Alestrom P & Collas P 2011 Prepatterning of developmental gene expression by modified histones before zygotic genome activation. Dev Cell 21 993-1004. Lindroth AM, Park YJ, McLean CM, Dokshin GA, Persson JM, Herman H, Pasini D, Miro X, Donohoe ME, Lee JT, Helin K & Soloway PD 2008 Antagonism between DNA and H3K27 methylation at the imprinted Rasgrf1 locus. PLoS Genet 4 e1000145. Lister R, Pelizzola M, Dowen RH, Hawkins RD, Hon G, Tonti-Filippini J, Nery JR, Lee L, Ye Z, Ngo QM, Edsall L, Antosiewicz-Bourget J, Stewart R, Ruotti V, Millar AH, Thomson JA, Ren B & Ecker JR 2009 Human DNA methylomes at base resolution show widespread epigenomic differences. Nature 462 315-322. Liu W, Ma Q, Wong K, Li W, Ohgi K, Zhang J, Aggarwal AK & Rosenfeld MG 2013 Brd4 and JMJD6- associated anti-pause enhancers in regulation of transcriptional pause release. Cell 155 1581-1595. Liu WM, Pang RT, Chiu PC, Wong BP, Lao K, Lee KF & Yeung WS 2012 Sperm-borne microRNA-34c is required for the first cleavage division in mouse. Proc Natl Acad Sci U S A 109 490-494. Loiodice I, Alves A, Rabut G, Van Overbeek M, Ellenberg J, Sibarita JB & Doye V 2004 The entire Nup107-160 complex, including three new members, is targeted as one entity to kinetochores in mitosis. Mol Biol Cell 15 3333-3344. Loppin B, Bonnefoy E, Anselme C, Laurencon A, Karr TL & Couble P 2005 The histone H3.3 chaperone HIRA is essential for chromatin assembly in the male pronucleus. Nature 437 1386-1390. Maatouk DM, Kellam LD, Mann MR, Lei H, Li E, Bartolomei MS & Resnick JL 2006 DNA methylation is a primary mechanism for silencing postmigratory primordial germ cell genes in both germ cell and somatic cell lineages. Development 133 3411-3418. Maeno M, Mead PE, Kelley C, Xu RH, Kung HF, Suzuki A, Ueno N & Zon LI 1996 The role of BMP-4 and GATA-2 in the induction and differentiation of hematopoietic mesoderm in Xenopus laevis. Blood 88 1965-1972. Mali P, Esvelt KM & Church GM 2013 Cas9 as a versatile tool for engineering biology. Nature methods 10 957-963. Martianov I, Brancorsini S, Catena R, Gansmuller A, Kotaja N, Parvinen M, Sassone-Corsi P & Davidson I 2005 Polar nuclear localization of H1T2, a histone H1 variant, required for Chapter 9: References 192 spermatid elongation and DNA condensation during spermiogenesis. Proc Natl Acad Sci U S A 102 2808-2813. Martianov I, Fimia GM, Dierich A, Parvinen M, Sassone-Corsi P & Davidson I 2001 Late arrest of spermiogenesis and germ cell apoptosis in mice lacking the TBP-like TLF/TRF2 gene. Mol Cell 7 509-515. Martin M 2011 Cutadapt removes adapter sequences from high-throughput sequencing reads. Matzuk MM, McKeown MR, Filippakopoulos P, Li Q, Ma L, Agno JE, Lemieux ME, Picaud S, Yu RN, Qi J, Knapp S & Bradner JE 2012 Small-molecule inhibition of BRDT for male contraception. Cell 150 673-684. McKittrick E, Gafken PR, Ahmad K & Henikoff S 2004 Histone H3.3 is enriched in covalent modifications associated with active chromatin. Proc Natl Acad Sci U S A 101 1525-1530. McLay DW & Clarke HJ 2003 Remodelling the paternal chromatin at fertilization in mammals. Reproduction 125 625-633. Meistrich ML, Mohapatra B, Shirley CR & Zhao M 2003 Roles of transition nuclear proteins in spermiogenesis. Chromosoma 111 483-488. Miller D 2007 Spermatozoal RNA as reservoir, marker and carrier of epigenetic information: implications for cloning. Reprod Domest Anim 42 Suppl 2 2-9. Miyamoto K, Teperek M, Yusa K, Allen GE, Bradshaw CR & Gurdon JB 2013 Nuclear Wave1 is required for reprogramming transcription in oocytes and for normal development. Science 341 1002-1005. Mizutani E, Yamagata K, Ono T, Akagi S, Geshi M & Wakayama T 2012 Abnormal chromosome segregation at early cleavage is a major cause of the full-term developmental failure of mouse clones. Dev Biol 364 56-65. Monesi V, Geremia R, D'Agostino A & Boitani C 1978 Biochemistry of male germ cell differentiation in mammals: RNA synthesis in meiotic and postmeiotic cells. Curr Top Dev Biol 12 11-36. Moriniere J, Rousseaux S, Steuerwald U, Soler-Lopez M, Curtet S, Vitte AL, Govin J, Gaucher J, Sadoul K, Hart DJ, Krijgsveld J, Khochbin S, Muller CW & Petosa C 2009 Cooperative binding of two acetylation marks on a histone tail by a single bromodomain. Nature 461 664-668. Motamedi MR, Hong EJ, Li X, Gerber S, Denison C, Gygi S & Moazed D 2008 HP1 proteins form distinct complexes and mediate heterochromatic gene silencing by nonoverlapping mechanisms. Mol Cell 32 778-790. Muller J, Hart CM, Francis NJ, Vargas ML, Sengupta A, Wild B, Miller EL, O'Connor MB, Kingston RE & Simon JA 2002 Histone methyltransferase activity of a Drosophila Polycomb group repressor complex. Cell 111 197-208. Nakayama J, Rice JC, Strahl BD, Allis CD & Grewal SI 2001 Role of histone H3 lysine 9 methylation in epigenetic control of heterochromatin assembly. Science 292 110-113. Nakazaki H, Reddy AC, Mania-Farnell BL, Shen YW, Ichi S, McCabe C, George D, McLone DG, Tomita T & Mayanil CS 2008 Key basic helix-loop-helix transcription factor genes Hes1 and Ngn2 are regulated by Pax3 during mouse embryonic development. Dev Biol 316 510-523. Narbonne P, Simpson DE & Gurdon JB 2011 Deficient induction response in a Xenopus nucleocytoplasmic hybrid. PLoS Biol 9 e1001197. Nardelli J, Thiesson D, Fujiwara Y, Tsai FY & Orkin SH 1999 Expression and genetic interaction of transcription factors GATA-2 and GATA-3 during development of the mouse central nervous system. Dev Biol 210 305-321. Newport J & Kirschner M 1982 A major developmental transition in early Xenopus embryos: I. characterization and timing of cellular changes at the midblastula stage. Cell 30 675-686. Ng RK & Gurdon JB 2005 Epigenetic memory of active gene transcription is inherited through somatic cell nuclear transfer. Proc Natl Acad Sci U S A 102 1957-1962. Ng RK & Gurdon JB 2008 Epigenetic memory of an active gene state depends on histone H3.3 incorporation into chromatin in the absence of transcription. Nat Cell Biol 10 102-109. Chapter 9: References 193 Nicklay JJ, Shechter D, Chitta RK, Garcia BA, Shabanowitz J, Allis CD & Hunt DF 2009 Analysis of histones in Xenopus laevis. II. mass spectrometry reveals an index of cell type-specific modifications on H3 and H4. J Biol Chem 284 1075-1085. Nicolas E, Morales V, Magnaghi-Jaulin L, Harel-Bellan A, Richard-Foy H & Trouche D 2000 RbAp48 belongs to the histone deacetylase complex that associates with the retinoblastoma protein. J Biol Chem 275 9797-9804. Nieuwkoop PD & Faber J 1994 Normal table of Xenopus laevis (Daudin) : a systematical and chronological survey of the development from the fertilized egg till the end of metamorphosis. New York: Garland Pub. Nishiyama A, Yamaguchi L, Sharif J, Johmura Y, Kawamura T, Nakanishi K, Shimamura S, Arita K, Kodama T, Ishikawa F, Koseki H & Nakanishi M 2013 Uhrf1-dependent H3K23 ubiquitylation couples maintenance DNA methylation and replication. Nature 502 249-253. Noma K, Allis CD & Grewal SI 2001 Transitions in distinct histone H3 methylation patterns at the heterochromatin domain boundaries. Science 293 1150-1155. Ogonuki N, Inoue K, Miki H, Mochida K, Hatori M, Okada H, Takeiri S, Shimozawa N, Nagashima H, Sankai T & Ogura A 2005 Differential development of rabbit embryos following microinsemination with sperm and spermatids. Mol Reprod Dev 72 411-417. Ogryzko VV, Schiltz RL, Russanova V, Howard BH & Nakatani Y 1996 The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87 953-959. Ohta H, Sakaide Y & Wakayama T 2009 Functional analysis of male mouse haploid germ cells of various differentiation stages: early and late round spermatids are functionally equivalent in producing progeny. Biol Reprod 80 511-517. Okuwaki M, Sumi A, Hisaoka M, Saotome-Nakamura A, Akashi S, Nishimura Y & Nagata K 2012 Function of homo- and hetero-oligomers of human nucleoplasmin/nucleophosmin family proteins NPM1, NPM2 and NPM3 during sperm chromatin remodeling. Nucleic Acids Res 40 4861-4878. Ostermeier GC, Miller D, Huntriss JD, Diamond MP & Krawetz SA 2004 Reproductive biology: delivering spermatozoan RNA to the oocyte. Nature 429 154. Ozsolak F, Song JS, Liu XS & Fisher DE 2007 High-throughput mapping of the chromatin structure of human promoters. Nat Biotechnol 25 244-248. Padmanabhan N, Jia D, Geary-Joo C, Wu X, Ferguson-Smith AC, Fung E, Bieda MC, Snyder FF, Gravel RA, Cross JC & Watson ED 2013 Mutation in folate metabolism causes epigenetic instability and transgenerational effects on development. Cell 155 81-93. Papageorgis P, Lambert AW, Ozturk S, Gao F, Pan H, Manne U, Alekseyev YO, Thiagalingam A, Abdolmaleky HM, Lenburg M & Thiagalingam S 2010 Smad signaling is required to maintain epigenetic silencing during breast cancer progression. Cancer Res 70 968-978. Paradowska AS, Miller D, Spiess AN, Vieweg M, Cerna M, Dvorakova-Hortova K, Bartkuhn M, Schuppe HC, Weidner W & Steger K 2012 Genome wide identification of promoter binding sites for H4K12ac in human sperm and its relevance for early embryonic development. Epigenetics 7 1057-1070. Pasque V, Gillich A, Garrett N & Gurdon JB 2011a Histone variant macroH2A confers resistance to nuclear reprogramming. EMBO J 30 2373-2387. Pasque V, Jullien J, Miyamoto K, Halley-Stott RP & Gurdon JB 2011b Epigenetic factors influencing resistance to nuclear reprogramming. Trends Genet 27 516-525. Pasque V, Miyamoto K & Gurdon JB 2010 Efficiencies and mechanisms of nuclear reprogramming. Cold Spring Harb Symp Quant Biol 75 189-200. Perez-Montero S, Carbonell A, Moran T, Vaquero A & Azorin F 2013 The embryonic linker histone H1 variant of Drosophila, dBigH1, regulates zygotic genome activation. Dev Cell 26 578-590. Perez-Pinera P, Kocak DD, Vockley CM, Adler AF, Kabadi AM, Polstein LR, Thakore PI, Glass KA, Ousterout DG, Leong KW, Guilak F, Crawford GE, Reddy TE & Gersbach CA 2013 RNA- Chapter 9: References 194 guided gene activation by CRISPR-Cas9-based transcription factors. Nature methods 10 973- 976. Pessot CA, Brito M, Figueroa J, Concha, II, Yanez A & Burzio LO 1989 Presence of RNA in the sperm nucleus. Biochem Biophys Res Commun 158 272-278. Petruk S, Sedkov Y, Johnston DM, Hodgson JW, Black KL, Kovermann SK, Beck S, Canaani E, Brock HW & Mazo A 2012 TrxG and PcG proteins but not methylated histones remain associated with DNA through replication. Cell 150 922-933. Philpott A & Leno GH 1992 Nucleoplasmin remodels sperm chromatin in Xenopus egg extracts. Cell 69 759-767. Philpott A, Leno GH & Laskey RA 1991 Sperm decondensation in Xenopus egg cytoplasm is mediated by nucleoplasmin. Cell 65 569-578. Pittoggi C, Magnano AR, Sciamanna I, Giordano R, Lorenzini R & Spadafora C 2001 Specific localization of transcription factors in the chromatin of mouse mature spermatozoa. Mol Reprod Dev 60 97-106. Pivot-Pajot C, Caron C, Govin J, Vion A, Rousseaux S & Khochbin S 2003 Acetylation-dependent chromatin reorganization by BRDT, a testis-specific bromodomain-containing protein. Mol Cell Biol 23 5354-5365. Plath K, Fang J, Mlynarczyk-Evans SK, Cao R, Worringer KA, Wang H, de la Cruz CC, Otte AP, Panning B & Zhang Y 2003 Role of histone H3 lysine 27 methylation in X inactivation. Science 300 131-135. Poot RA, Bozhenok L, van den Berg DL, Hawkes N & Varga-Weisz PD 2005 Chromatin remodeling by WSTF-ISWI at the replication site: opening a window of opportunity for epigenetic inheritance? Cell Cycle 4 543-546. Popp C, Dean W, Feng S, Cokus SJ, Andrews S, Pellegrini M, Jacobsen SE & Reik W 2010 Genome- wide erasure of DNA methylation in mouse primordial germ cells is affected by AID deficiency. Nature 463 1101-1105. Potok ME, Nix DA, Parnell TJ & Cairns BR 2013 Reprogramming the maternal zebrafish genome after fertilization to match the paternal methylation pattern. Cell 153 759-772. Qi LS, Larson MH, Gilbert LA, Doudna JA, Weissman JS, Arkin AP & Lim WA 2013 Repurposing CRISPR as an RNA-guided platform for sequence-specific control of gene expression. Cell 152 1173-1183. Qian MX, Pang Y, Liu CH, Haratake K, Du BY, Ji DY, Wang GF, Zhu QQ, Song W, Yu Y, Zhang XX, Huang HT, Miao S, Chen LB, Zhang ZH, Liang YN, Liu S, Cha H, Yang D, Zhai Y, Komatsu T, Tsuruta F, Li H, Cao C, Li W, Li GH, Cheng Y, Chiba T, Wang L, Goldberg AL, Shen Y & Qiu XB 2013 Acetylation-mediated proteasomal degradation of core histones during DNA repair and spermatogenesis. Cell 153 1012-1024. Rais Y, Zviran A, Geula S, Gafni O, Chomsky E, Viukov S, Mansour AA, Caspi I, Krupalnik V, Zerbib M, Maza I, Mor N, Baran D, Weinberger L, Jaitin DA, Lara-Astiaso D, Blecher-Gonen R, Shipony Z, Mukamel Z, Hagai T, Gilad S, Amann-Zalcenstein D, Tanay A, Amit I, Novershtern N & Hanna JH 2013 Deterministic direct reprogramming of somatic cells to pluripotency. Nature 502 65-70. Ramos I, Martin-Benito J, Finn R, Bretana L, Aloria K, Arizmendi JM, Ausio J, Muga A, Valpuesta JM & Prado A 2010 Nucleoplasmin binds histone H2A-H2B dimers through its distal face. J Biol Chem 285 33771-33778. Ravel C, Chantot-Bastaraud S, El Houate B, Berthaut I, Verstraete L, De Larouziere V, Lourenco D, Dumaine A, Antoine JM, Mandelbaum J, Siffroi JP & McElreavey K 2007 Mutations in the protamine 1 gene associated with male infertility. Mol Hum Reprod 13 461-464. Ray-Gallet D, Quivy JP, Scamps C, Martini EM, Lipinski M & Almouzni G 2002 HIRA is critical for a nucleosome assembly pathway independent of DNA synthesis. Mol Cell 9 1091-1100. Chapter 9: References 195 Read EM, Rodaway AR, Neave B, Brandon N, Holder N, Patient RK & Walmsley ME 1998 Evidence for non-axial A/P patterning in the nonneural ectoderm of Xenopus and zebrafish pregastrula embryos. Int J Dev Biol 42 763-774. Reddington JP, Perricone SM, Nestor CE, Reichmann J, Youngson NA, Suzuki M, Reinhardt D, Dunican DS, Prendergast JG, Mjoseng H, Ramsahoye BH, Whitelaw E, Greally JM, Adams IR, Bickmore WA & Meehan RR 2013 Redistribution of H3K27me3 upon DNA hypomethylation results in de-repression of Polycomb target genes. Genome Biol 14 R25. Reversade B & De Robertis EM 2005 Regulation of ADMP and BMP2/4/7 at opposite embryonic poles generates a self-regulating morphogenetic field. Cell 123 1147-1160. Reversade B, Kuroda H, Lee H, Mays A & De Robertis EM 2005 Depletion of Bmp2, Bmp4, Bmp7 and Spemann organizer signals induces massive brain formation in Xenopus embryos. Development 132 3381-3392. Richly H, Rocha-Viegas L, Ribeiro JD, Demajo S, Gundem G, Lopez-Bigas N, Nakagawa T, Rospert S, Ito T & Di Croce L 2010 Transcriptional activation of polycomb-repressed genes by ZRF1. Nature 468 1124-1128. Risley MS & Eckhardt RA 1979 Dissociation and separation of Xenopus laevis spermatogenic cells. Journal of Experimental Zoology 207 93-106. Risley MS & Eckhardt RA 1981 H1 histone variants in Xenopus laevis. Dev Biol 84 79-87. Risley MS, Eckhardt RA, Mann M & Kasinsky HE 1982 Determinants of sperm nuclear shaping in the genus Xenopus. Chromosoma 84 557-569. Robinson MD, McCarthy DJ & Smyth GK 2010 edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26 139-140. Roger B, Moisand A, Amalric F & Bouvet P 2002 rDNA transcription during Xenopus laevis oogenesis. Biochem Biophys Res Commun 290 1151-1160. Roosen-Runge EC 1969 Comparative aspects of spermatogenesis. Biol Reprod 1 Suppl 1:24-31. Rowles A & Blow JJ 1997 Chromatin proteins involved in the initiation of DNA replication. Curr Opin Genet Dev 7 152-157. Santenard A, Ziegler-Birling C, Koch M, Tora L, Bannister AJ & Torres-Padilla ME 2010 Heterochromatin formation in the mouse embryo requires critical residues of the histone variant H3.3. Nature cell biology 12 853-862. Santos-Rosa H, Schneider R, Bannister AJ, Sherriff J, Bernstein BE, Emre NC, Schreiber SL, Mellor J & Kouzarides T 2002 Active genes are tri-methylated at K4 of histone H3. Nature 419 407- 411. Saxena UH, Powell CM, Fecko JK, Cacioppo R, Chou HS, Cooper GM & Hansen U 2009 Phosphorylation by cyclin C/cyclin-dependent kinase 2 following mitogenic stimulation of murine fibroblasts inhibits transcriptional activity of LSF during G1 progression. Mol Cell Biol 29 2335-2345. Schones DE, Cui K, Cuddapah S, Roh TY, Barski A, Wang Z, Wei G & Zhao K 2008 Dynamic regulation of nucleosome positioning in the human genome. Cell 132 887-898. Schubeler D, Lorincz MC, Cimbora DM, Telling A, Feng YQ, Bouhassira EE & Groudine M 2000 Genomic targeting of methylated DNA: influence of methylation on transcription, replication, chromatin structure, and histone acetylation. Mol Cell Biol 20 9103-9112. Sendler E, Johnson GD, Mao S, Goodrich RJ, Diamond MP, Hauser R & Krawetz SA 2013 Stability, delivery and functions of human sperm RNAs at fertilization. Nucleic Acids Res 41 4104-4117. Seyedin SM & Kistler WS 1980 Isolation and characterization of rat testis H1t. An H1 histone variant associated with spermatogenesis. J Biol Chem 255 5949-5954. Shechter D, Nicklay JJ, Chitta RK, Shabanowitz J, Hunt DF & Allis CD 2009 Analysis of histones in Xenopus laevis. I. A distinct index of enriched variants and modifications exists in each cell type and is remodeled during developmental transitions. J Biol Chem 284 1064-1074. Sible JC, Erikson E, Hendrickson M, Maller JL & Gautier J 1998 Developmental regulation of MCM replication factors in Xenopus laevis. Curr Biol 8 347-350. Chapter 9: References 196 Simpson RT 1978 Structure of chromatin containing extensively acetylated H3 and H4. Cell 13 691- 699. Smallwood A, Esteve PO, Pradhan S & Carey M 2007 Functional cooperation between HP1 and DNMT1 mediates gene silencing. Genes Dev 21 1169-1178. Smallwood A, Hon GC, Jin F, Henry RE, Espinosa JM & Ren B 2012 CBX3 regulates efficient RNA processing genome-wide. Genome Res 22 1426-1436. Smith RC, Dworkin-Rastl E & Dworkin MB 1988 Expression of a histone H1-like protein is restricted to early Xenopus development. Genes Dev 2 1284-1295. Smith SJ, Fairclough L, Latinkic BV, Sparrow DB & Mohun TJ 2006 Xenopus laevis transgenesis by sperm nuclear injection. Nat Protoc 1 2195-2203. Snowden AW, Gregory PD, Case CC & Pabo CO 2002 Gene-specific targeting of H3K9 methylation is sufficient for initiating repression in vivo. Curr Biol 12 2159-2166. Song F, Mahmood S, Ghosh S, Liang P, Smiraglia DJ, Nagase H & Held WA 2009 Tissue specific differentially methylated regions (TDMR): Changes in DNA methylation during development. Genomics 93 130-139. Song F, Smith JF, Kimura MT, Morrow AD, Matsuyama T, Nagase H & Held WA 2005 Association of tissue-specific differentially methylated regions (TDMs) with differential gene expression. Proc Natl Acad Sci U S A 102 3336-3341. Sridharan R, Gonzales-Cope M, Chronis C, Bonora G, McKee R, Huang C, Patel S, Lopez D, Mishra N, Pellegrini M, Carey M, Garcia BA & Plath K 2013 Proteomic and genomic approaches reveal critical functions of H3K9 methylation and heterochromatin protein-1gamma in reprogramming to pluripotency. Nat Cell Biol 15 872-882. Steger K, Wilhelm J, Konrad L, Stalf T, Greb R, Diemer T, Kliesch S, Bergmann M & Weidner W 2008 Both protamine-1 to protamine-2 mRNA ratio and Bcl2 mRNA content in testicular spermatids and ejaculated spermatozoa discriminate between fertile and infertile men. Hum Reprod 23 11-16. Stengel S, Fiebig U, Kurth R & Denner J 2010 Regulation of human endogenous retrovirus-K expression in melanomas by CpG methylation. Genes Chromosomes Cancer 49 401-411. Strahl BD & Allis CD 2000 The language of covalent histone modifications. Nature 403 41-45. Sung LY, Gao S, Shen H, Yu H, Song Y, Smith SL, Chang CC, Inoue K, Kuo L, Lian J, Li A, Tian XC, Tuck DP, Weissman SM, Yang X & Cheng T 2006 Differentiated cells are more efficient than adult stem cells for cloning by somatic cell nuclear transfer. Nat Genet 38 1323-1328. Surani MA, Hayashi K & Hajkova P 2007 Genetic and epigenetic regulators of pluripotency. Cell 128 747-762. Suzuki T, Minami N, Kono T & Imai H 2007 Comparison of the RNA polymerase I-, II- and III- dependent transcript levels between nuclear transfer and in vitro fertilized embryos at the blastocyst stage. J Reprod Dev 53 663-671. Syed SH, Boulard M, Shukla MS, Gautier T, Travers A, Bednar J, Faivre-Moskalenko C, Dimitrov S & Angelov D 2009 The incorporation of the novel histone variant H2AL2 confers unusual structural and functional properties of the nucleosome. Nucleic Acids Res 37 4684-4695. Szyf M 2014 Lamarck revisited: epigenetic inheritance of ancestral odor fear conditioning. Nat Neurosci 17 2-4. Tachiwana H, Kagawa W, Osakabe A, Kawaguchi K, Shiga T, Hayashi-Takanaka Y, Kimura H & Kurumizaka H 2010 Structural basis of instability of the nucleosome containing a testis- specific histone variant, human H3T. Proc Natl Acad Sci U S A 107 10454-10459. Talbert PB, Ahmad K, Almouzni G, Ausio J, Berger F, Bhalla PL, Bonner WM, Cande WZ, Chadwick BP, Chan SW, Cross GA, Cui L, Dimitrov SI, Doenecke D, Eirin-Lopez JM, Gorovsky MA, Hake SB, Hamkalo BA, Holec S, Jacobsen SE, Kamieniarz K, Khochbin S, Ladurner AG, Landsman D, Latham JA, Loppin B, Malik HS, Marzluff WF, Pehrson JR, Postberg J, Schneider R, Singh MB, Smith MM, Thompson E, Torres-Padilla ME, Tremethick DJ, Turner BM, Waterborg JH, Chapter 9: References 197 Wollmann H, Yelagandula R, Zhu B & Henikoff S 2012 A unified phylogeny-based nomenclature for histone variants. Epigenetics Chromatin 5 7. Tamada H, Van Thuan N, Reed P, Nelson D, Katoku-Kikyo N, Wudel J, Wakayama T & Kikyo N 2006 Chromatin decondensation and nuclear reprogramming by nucleoplasmin. Mol Cell Biol 26 1259-1271. Tan M, Luo H, Lee S, Jin F, Yang JS, Montellier E, Buchou T, Cheng Z, Rousseaux S, Rajagopal N, Lu Z, Ye Z, Zhu Q, Wysocka J, Ye Y, Khochbin S, Ren B & Zhao Y 2011 Identification of 67 histone marks and histone lysine crotonylation as a new type of histone modification. Cell 146 1016-1028. Tan MH, Au KF, Leong DE, Foygel K, Wong WH & Yao MW 2013 An Oct4-Sall4-Nanog network controls developmental progression in the pre-implantation mouse embryo. Mol Syst Biol 9 632. Tanaka H, Iguchi N, Isotani A, Kitamura K, Toyama Y, Matsuoka Y, Onishi M, Masai K, Maekawa M, Toshimori K, Okabe M & Nishimune Y 2005 HANP1/H1T2, a novel histone H1-like protein involved in nuclear formation and sperm fertility. Mol Cell Biol 25 7107-7119. Tanaka M, Hennebold JD, Macfarlane J & Adashi EY 2001 A mammalian oocyte-specific linker histone gene H1oo: homology with the genes for the oocyte-specific cleavage stage histone (cs-H1) of sea urchin and the B4/H1M histone of the frog. Development 128 655-664. Teperek M & Miyamoto K 2013 Nuclear reprogramming of sperm and somatic nuclei in eggs and oocytes. Reprod Med Biol 12 133-149. Teranishi T, Tanaka M, Kimoto S, Ono Y, Miyakoshi K, Kono T & Yoshimura Y 2004 Rapid replacement of somatic linker histones with the oocyte-specific linker histone H1foo in nuclear transfer. Dev Biol 266 76-86. Thomas PD, Campbell MJ, Kejariwal A, Mi H, Karlak B, Daverman R, Diemer K, Muruganujan A & Narechania A 2003 PANTHER: a library of protein families and subfamilies indexed by function. Genome Res 13 2129-2141. Tong JK, Hassig CA, Schnitzler GR, Kingston RE & Schreiber SL 1998 Chromatin deacetylation by an ATP-dependent nucleosome remodelling complex. Nature 395 917-921. Torres-Padilla ME, Bannister AJ, Hurd PJ, Kouzarides T & Zernicka-Goetz M 2006 Dynamic distribution of the replacement histone variant H3.3 in the mouse oocyte and preimplantation embryos. The International journal of developmental biology 50 455-461. Torres-Padilla ME, Parfitt DE, Kouzarides T & Zernicka-Goetz M 2007 Histone arginine methylation regulates pluripotency in the early mouse embryo. Nature 445 214-218. Toshimori K 2003 Biology of spermatozoa maturation: an overview with an introduction to this issue. Microsc Res Tech 61 1-6. Trapnell C, Pachter L & Salzberg SL 2009 TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25 1105-1111. Tsai SC, Valkov N, Yang WM, Gump J, Sullivan D & Seto E 2000 Histone deacetylase interacts directly with DNA topoisomerase II. Nat Genet 26 349-353. Turner BM, Birley AJ & Lavender J 1992 Histone H4 isoforms acetylated at specific lysine residues define individual chromosomes and chromatin domains in Drosophila polytene nuclei. Cell 69 375-384. Umehara T, Nakamura Y, Jang MK, Nakano K, Tanaka A, Ozato K, Padmanabhan B & Yokoyama S 2010a Structural basis for acetylated histone H4 recognition by the human BRD2 bromodomain. J Biol Chem 285 7610-7618. Umehara T, Nakamura Y, Wakamori M, Ozato K, Yokoyama S & Padmanabhan B 2010b Structural implications for K5/K12-di-acetylated histone H4 recognition by the second bromodomain of BRD2. FEBS Lett 584 3901-3908. Ura K, Nightingale K & Wolffe AP 1996 Differential association of HMG1 and linker histones B4 and H1 with dinucleosomal DNA: structural transitions and transcriptional repression. EMBO J 15 4959-4969. Chapter 9: References 198 Vallee M, Gravel C, Palin MF, Reghenas H, Stothard P, Wishart DS & Sirard MA 2005 Identification of novel and known oocyte-specific genes using complementary DNA subtraction and microarray analysis in three different species. Biol Reprod 73 63-71. van der Heijden GW, Derijck AA, Ramos L, Giele M, van der Vlag J & de Boer P 2006 Transmission of modified nucleosomes from the mouse male germline to the zygote and subsequent remodeling of paternal chromatin. Dev Biol 298 458-469. van der Heijden GW, Dieker JW, Derijck AA, Muller S, Berden JH, Braat DD, van der Vlag J & de Boer P 2005 Asymmetry in histone H3 variants and lysine methylation between paternal and maternal chromatin of the early mouse zygote. Mech Dev 122 1008-1022. van der Heijden GW, Ramos L, Baart EB, van den Berg IM, Derijck AA, van der Vlag J, Martini E & de Boer P 2008 Sperm-derived histones contribute to zygotic chromatin in humans. BMC Dev Biol 8 34. van Heeringen SJ, Akkers RC, van Kruijsbergen I, Arif MA, Hanssen LL, Sharifi N & Veenstra GJ 2014 Principles of nucleation of H3K27 methylation during embryonic development. Genome Res 24 401-410. Van Thuan N, Wakayama S, Kishigami S, Ohta H, Hikichi T, Mizutani E, Bui HT & Wakayama T 2006 Injection of somatic cell cytoplasm into oocytes before intracytoplasmic sperm injection impairs full-term development and increases placental weight in mice. Biol Reprod 74 865- 873. VanDemark AP, Blanksma M, Ferris E, Heroux A, Hill CP & Formosa T 2006 The structure of the yFACT Pob3-M domain, its interaction with the DNA replication factor RPA, and a potential role in nucleosome deposition. Mol Cell 22 363-374. Vassena R, Han Z, Gao S, Baldwin DA, Schultz RM & Latham KE 2007 Tough beginnings: alterations in the transcriptome of cloned embryos during the first two cell cycles. Dev Biol 304 75-89. Vassoler FM, White SL, Schmidt HD, Sadri-Vakili G & Pierce RC 2013 Epigenetic inheritance of a cocaine-resistance phenotype. Nat Neurosci 16 42-47. Vastenhouw NL, Zhang Y, Woods IG, Imam F, Regev A, Liu XS, Rinn J & Schier AF 2010 Chromatin signature of embryonic pluripotency is established during genome activation. Nature 464 922-926. Vermaak D, Wade PA, Jones PL, Shi YB & Wolffe AP 1999 Functional analysis of the SIN3-histone deacetylase RPD3-RbAp48-histone H4 connection in the Xenopus oocyte. Mol Cell Biol 19 5847-5860. Wade PA, Gegonne A, Jones PL, Ballestar E, Aubry F & Wolffe AP 1999 Mi-2 complex couples DNA methylation to chromatin remodelling and histone deacetylation. Nat Genet 23 62-66. Wang G, Guo Y, Zhou T, Shi X, Yu J, Yang Y, Wu Y, Wang J, Liu M, Chen X, Tu W, Zeng Y, Jiang M, Li S, Zhang P, Zhou Q, Zheng B, Yu C, Zhou Z, Guo X & Sha J 2013 In-depth proteomic analysis of the human sperm reveals complex protein compositions. J Proteomics 79 114-122. Wang H, Wang L, Erdjument-Bromage H, Vidal M, Tempst P, Jones RS & Zhang Y 2004 Role of histone H2A ubiquitination in Polycomb silencing. Nature 431 873-878. Wang S, Zhang B & Faller DV 2002 Prohibitin requires Brg-1 and Brm for the repression of E2F and cell growth. EMBO J 21 3019-3028. Ward WS & Coffey DS 1991 DNA packaging and organization in mammalian spermatozoa: comparison with somatic cells. Biol Reprod 44 569-574. Wee G, Koo DB, Song BS, Kim JS, Kang MJ, Moon SJ, Kang YK, Lee KK & Han YM 2006 Inheritable histone H4 acetylation of somatic chromatins in cloned embryos. J Biol Chem 281 6048- 6057. Wills A, Dickinson K, Khokha M & Baker JC 2008 Bmp signaling is necessary and sufficient for ventrolateral endoderm specification in Xenopus. Dev Dyn 237 2177-2186. Winteringham LN, Kobelke S, Williams JH, Ingley E & Klinken SP 2004 Myeloid Leukemia Factor 1 inhibits erythropoietin-induced differentiation, cell cycle exit and p27Kip1 accumulation. Oncogene 23 5105-5109. Chapter 9: References 199 Wu Q, Bruce AW, Jedrusik A, Ellis PD, Andrews RM, Langford CF, Glover DM & Zernicka-Goetz M 2009 CARM1 is required in embryonic stem cells to maintain pluripotency and resist differentiation. Stem cells 27 2637-2645. Wu SF, Zhang H & Cairns BR 2011 Genes for embryo development are packaged in blocks of multivalent chromatin in zebrafish sperm. Genome Res 21 578-589. Wysocka J, Swigut T, Milne TA, Dou Y, Zhang X, Burlingame AL, Roeder RG, Brivanlou AH & Allis CD 2005 WDR5 associates with histone H3 methylated at K4 and is essential for H3 K4 methylation and vertebrate development. Cell 121 859-872. Wysocka J, Swigut T, Xiao H, Milne TA, Kwon SY, Landry J, Kauer M, Tackett AJ, Chait BT, Badenhorst P, Wu C & Allis CD 2006 A PHD finger of NURF couples histone H3 lysine 4 trimethylation with chromatin remodelling. Nature 442 86-90. Yan W, Ma L, Burns KH & Matzuk MM 2003 HILS1 is a spermatid-specific linker histone H1-like protein implicated in chromatin remodeling during mammalian spermiogenesis. Proc Natl Acad Sci U S A 100 10546-10551. Yang M, Li S, Liang G, Sun L, Zhang X, Liu N, Pang H, Miao S, Wang L & Rao Z 2003 Crystallization and preliminary crystallographic analysis of RSB-66, a novel round spermatid-specific protein. Acta Crystallogr D Biol Crystallogr 59 1853-1855. Yang Z, Yik JH, Chen R, He N, Jang MK, Ozato K & Zhou Q 2005 Recruitment of P-TEFb for stimulation of transcriptional elongation by the bromodomain protein Brd4. Mol Cell 19 535- 545. Yao YL & Yang WM 2003 The metastasis-associated proteins 1 and 2 form distinct protein complexes with histone deacetylase activity. J Biol Chem 278 42560-42568. Yokota T 1991 Nuclear Basic Proteins of Xenopus laevis Sperm: Their Characterization and Synthesis during Spermatogenesis. Development, Growth & Differentiation 33 9-17. Yoneda-Kato N & Kato JY 2008 Shuttling imbalance of MLF1 results in p53 instability and increases susceptibility to oncogenic transformation. Mol Cell Biol 28 422-434. Yoneda-Kato N, Tomoda K, Umehara M, Arata Y & Kato JY 2005 Myeloid leukemia factor 1 regulates p53 by suppressing COP1 via COP9 signalosome subunit 3. EMBO J 24 1739-1749. Yoo BK, Emdad L, Gredler R, Fuller C, Dumur CI, Jones KH, Jackson-Cook C, Su ZZ, Chen D, Saxena UH, Hansen U, Fisher PB & Sarkar D 2010 Transcription factor Late SV40 Factor (LSF) functions as an oncogene in hepatocellular carcinoma. Proc Natl Acad Sci U S A 107 8357- 8362. Yoshida M, Kijima M, Akita M & Beppu T 1990 Potent and specific inhibition of mammalian histone deacetylase both in vivo and in vitro by trichostatin A. J Biol Chem 265 17174-17179. Yu YE, Zhang Y, Unni E, Shirley CR, Deng JM, Russell LD, Weil MM, Behringer RR & Meistrich ML 2000 Abnormal spermatogenesis and reduced fertility in transition nuclear protein 1- deficient mice. Proc Natl Acad Sci U S A 97 4683-4688. Zdobnov EM & Apweiler R 2001 InterProScan--an integration platform for the signature-recognition methods in InterPro. Bioinformatics 17 847-848. Zhang J, Tam WL, Tong GQ, Wu Q, Chan HY, Soh BS, Lou Y, Yang J, Ma Y, Chai L, Ng HH, Lufkin T, Robson P & Lim B 2006 Sall4 modulates embryonic stem cell pluripotency and early embryonic development by the transcriptional regulation of Pou5f1. Nat Cell Biol 8 1114- 1123. Zhang P, Lee H, Brunzelle JS & Couture JF 2012 The plasticity of WDR5 peptide-binding cleft enables the binding of the SET1 family of histone methyltransferases. Nucleic Acids Res 40 4237- 4246. Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C, Myers RM, Brown M, Li W & Liu XS 2008 Model-based analysis of ChIP-Seq (MACS). Genome Biol 9 R137. Zhang Y, Ng HH, Erdjument-Bromage H, Tempst P, Bird A & Reinberg D 1999 Analysis of the NuRD subunits reveals a histone deacetylase core complex and a connection with DNA methylation. Genes Dev 13 1924-1935. Chapter 9: References 200 Zhang Y, Vastenhouw NL, Feng J, Fu K, Wang C, Ge Y, Pauli A, van Hummelen P, Schier AF & Liu XS 2014 Canonical nucleosome organization at promoters forms during genome activation. Genome Res 24 260-266. Zhao M, Shirley CR, Hayashi S, Marcon L, Mohapatra B, Suganuma R, Behringer RR, Boissonneault G, Yanagimachi R & Meistrich ML 2004 Transition nuclear proteins are required for normal chromatin condensation and functional sperm development. Genesis 38 200-213. Zhao R, Nakamura T, Fu Y, Lazar Z & Spector DL 2011 Gene bookmarking accelerates the kinetics of post-mitotic transcriptional re-activation. Nat Cell Biol 13 1295-1304. Zheng J, Xia X, Ding H, Yan A, Hu S, Gong X, Zong S, Zhang Y & Sheng HZ 2008 Erasure of the paternal transcription program during spermiogenesis: the first step in the reprogramming of sperm chromatin for zygotic development. Dev Dyn 237 1463-1476. Zheng Z, Jia JL, Bou G, Hu LL, Wang ZD, Shen XH, Shan ZY, Shen JL, Liu ZH & Lei L 2012 rRNA genes are not fully activated in mouse somatic cell nuclear transfer embryos. J Biol Chem 287 19949-19960. Zhu D, Hunter SB, Vertino PM & Van Meir EG 2011 Overexpression of MBD2 in glioblastoma maintains epigenetic silencing and inhibits the antiangiogenic function of the tumor suppressor gene BAI1. Cancer Res 71 5859-5870. Zippo A, Serafini R, Rocchigiani M, Pennacchini S, Krepelova A & Oliviero S 2009 Histone crosstalk between H3S10ph and H4K16ac generates a histone code that mediates transcription elongation. Cell 138 1122-1136. Zon LI, Mather C, Burgess S, Bolce ME, Harland RM & Orkin SH 1991 Expression of GATA-binding proteins during embryonic development in Xenopus laevis. Proc Natl Acad Sci U S A 88 10642-10646. Chapter 10: Appendices 201 Chapter 10 Appendices List of Appendices: - Appendix 1: review article: ‘Epigenetic reprogramming: is deamination key to active DNA demethylation?’, Teperek-Tkacz M., Pasque V., Gentsch G., Ferguson-Smith AC., Reproduction, 2011 - Appendix 2: review article: ‘Nuclear reprogramming of sperm and somatic nuclei in eggs and oocytes.’, Teperek M., Miyamoto K., Reprod Med Biol., 2013 - Appendix 3: book chapter: ‘Cloning of Amphibia’ (part of ‘Principles of Cloning’), Teperek-Tkacz M., Byrne JA., Gurdon JB., Elsevier, 2014 - Appendix 4: research article: ‘Nuclear Wave1 is required for reprogramming transcription in oocytes and for normal development.’, Miyamoto K., Teperek M., Yusa K., Allen GE., Bradshaw CR., Gurdon JB., Science, 2013 Chapter 10: Appendices 202 Appendix 1 Review article: ‘Epigenetic reprogramming: is deamination key to active DNA demethylation?’ Teperek-Tkacz M., Pasque V., Gentsch G., Ferguson-Smith AC., Reproduction, 2011 Chapter 10: Appendices 203 Chapter 10: Appendices 204 Chapter 10: Appendices 205 Chapter 10: Appendices 206 Chapter 10: Appendices 207 Chapter 10: Appendices 208 Chapter 10: Appendices 209 Chapter 10: Appendices 210 Chapter 10: Appendices 211 Chapter 10: Appendices 212 Chapter 10: Appendices 213 Chapter 10: Appendices 214 Chapter 10: Appendices 215 Appendix 2 Review article: ‘Nuclear reprogramming of sperm and somatic nuclei in eggs and oocytes.’ Teperek M., Miyamoto K., Reprod Med Biol., 2013 Chapter 10: Appendices 216 Chapter 10: Appendices 217 Chapter 10: Appendices 218 Chapter 10: Appendices 219 Chapter 10: Appendices 220 Chapter 10: Appendices 221 Chapter 10: Appendices 222 Chapter 10: Appendices 223 Chapter 10: Appendices 224 Chapter 10: Appendices 225 Chapter 10: Appendices 226 Chapter 10: Appendices 227 Chapter 10: Appendices 228 Chapter 10: Appendices 229 Chapter 10: Appendices 230 Chapter 10: Appendices 231 Chapter 10: Appendices 232 Chapter 10: Appendices 233 Appendix 3 Book chapter: ‘Cloning of Amphibia’ (part of ‘Principles of Cloning’) Teperek-Tkacz M., Byrne JA., Gurdon JB., Elsevier, 2014 Chapter 10: Appendices 234 Chapter 10: Appendices 235 Chapter 10: Appendices 236 Chapter 10: Appendices 237 Chapter 10: Appendices 238 Chapter 10: Appendices 239 Chapter 10: Appendices 240 Chapter 10: Appendices 241 Chapter 10: Appendices 242 Chapter 10: Appendices 243 Chapter 10: Appendices 244 Appendix 4 Research article: ‘Nuclear Wave1 is required for reprogramming transcription in oocytes and for normal development.’ Miyamoto K., Teperek M., Yusa K., Allen GE., Bradshaw CR., Gurdon JB., Science, 2013 Chapter 10: Appendices 245 Chapter 10: Appendices 246 Chapter 10: Appendices 247 Chapter 10: Appendices 248