Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of chromatin assembly in murine embryos Jansen, Hailey Janice 2013

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

Item Metadata

Download

Media
24-ubc_2013_fall_jansen_hailey.pdf [ 2.18MB ]
Metadata
JSON: 24-1.0074032.json
JSON-LD: 24-1.0074032-ld.json
RDF/XML (Pretty): 24-1.0074032-rdf.xml
RDF/JSON: 24-1.0074032-rdf.json
Turtle: 24-1.0074032-turtle.txt
N-Triples: 24-1.0074032-rdf-ntriples.txt
Original Record: 24-1.0074032-source.json
Full Text
24-1.0074032-fulltext.txt
Citation
24-1.0074032.ris

Full Text

 CHARACTERIZATION OF CHROMATIN  ASSEMBLY IN MURINE EMBRYOS   by  Hailey Janice Jansen  B.Sc., The University of British Columbia, 2010   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in   THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Zoology)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   August 2013    ? Hailey Janice Jansen, 2013 ii  Abstract  During differentiation, changes in chromatin proteins lead to the establishment and maintenance of gene expression patterns. Histone H3 trimethylated at lysine 4 (H3K4me3) by the trithorax group (trxG) gene family Mixed lineage leukemia (MLL) is associated with active genes.  H3K27me3 is trimethylated by the Polycomb group (PcG) Enhancer of Zeste (EZH2) at repressed genes. In Drosophila  embryos, trxG and PcG proteins but not H3K4me3 or H3K27me3 are stable to DNA replication. In contrast, methylated histones are detected on nascent DNA in Drosophila  and murine cell lines.  Therefore some aspect of chromatin assembly or histone trimethylation must differ in different cells. My first aim was to determine if there is a change in the abundance of methylated histones at the replication fork in undifferentiated versus differentiated murine ES cells using two novel in vivo assays.  Most undifferentiated ES cells lack early H3K4me3 and H3K27me3, but after 4 days of differentiation, most cells have early trimethylation of H3K4 and H3K27.  I propose that the change in kinetics of histone methylation correlates with differentiation. To test this hypothesis, I carried out similar experiments on cells dissociated from day 9.5 (E9.5) and 14.5 (E14.5) murine embryos. In E9.5 cells there are two populations of cells, one that lacks methylated histones and the other contains methylated histones on nascent DNA. By E14.5 most cells exhibit H3K4me3 and H3K27me3 on nascent DNA. To determine if the presence of histone methyltransferases could account for the changes in histone methylation, I tested MLL1 and a subunit of the EZH2 complex, Su(z)12. Both are present continuously on nascent DNA, suggesting that their activity is regulated. Methylation and acetylation antagonize each other at the same residue. However I showed that the presence of acetylated H3K27 is not anticorrelated with H3K27me3 in most murine embryos cells. My results using inhibitors of the appropriate histone acetyltransferase were not conclusive owing to toxicity of the inhibitors. Overall, my results support the hypothesis that trimethylation of H3K4 and H3K27 on nascent DNA is developmentally regulated.      iii  Preface  The candidate (H. Jansen) performed all the experiments and analysis. Murine ES cells and murine embryos were gifts from the Lefebvre laboratory. K. Jacob from the Lefebvre laboratory did all mice matings for E9.5 and E14.5 embryos using the CD-1 mouse strain. Ethics approval was obtained from the Animal Care Committee at the University of British Columbia for doing research on mice. The ethics protocol number is A11-0293.   iv  Table of Contents  Abstract........................................................................................................................................ii Preface.........................................................................................................................................iii Table of Contents.......................................................................................................................iv List of Tables..............................................................................................................................vi List of Figures............................................................................................................................vii List of Abbreviations..................................................................................................................ix Acknowledgements.....................................................................................................................x 1. Introduction......................................................................................................................1 1.1 DNA replication............................................................................................................1 1.2 Histone modifications...................................................................................................4 1.3 Changes in chromatin during development.................................................................8 1.4 Inheritance of chromatin marks...................................................................................9 1.5 Thesis objectives.......................................................................................................16 2. Materials and Methods...................................................................................................17 2.1 Preparation of ES cells..............................................................................................17 2.2 Preparation of trypsinized embryo cells.....................................................................17 2.3 Preparation of cells for PLA and CAA experiments...................................................17 2.4  Proximity Ligation Assay...........................................................................................18 2.5 Chromatin Assembly Assay.......................................................................................18 2.6 PLA and CAA data analysis.......................................................................................18 2.7 Treatment of embryo cells with inhibitors...................................................................19 3. Results.............................................................................................................................20 3.1 Active and repressive histone marks at the DNA replication fork in ES cells............20 3.2 Embryo work.............................................................................................................25 3.2.1 Histone modifications are present at the replication fork in embryo cells........................................................................................................25 3.2.2 Modified histones are present on nascent DNA in embryo cells..............27 v  3.2.3 Histone methyltransferases at the replication fork in embryo cells..........32 3.2.4 Experiments to determine a mechanism of H3K27me3 assembly...........36 4. Discussion.......................................................................................................................46 References.................................................................................................................................52       vi  List of Tables  Table 3.1   Amount of EdU positive cells following treatment with inhibitors.............................41 vii  List of Figures  Figure 1.1.  Schematic diagram of DNA replication and some of the key molecules required......2  Figure 1.2. Schematic of nucleosome assembly...........................................................................4 Figure 1.3. Schematic diagram of modified histone tails and chromatin proteins bound to the modifications..................................................................................................................................5 Figure 1.4.  Schematic showing the model alternating between active and repressed gene states.............................................................................................................................................8 Figure 1.5 Overview of the Proximity Ligation Assay..................................................................13 Figure 1.6. H3K4me3, H3K27me3, Ph, and Trx are detected on nascent DNA in S2 cells........14 Figure 1.7. H3K4me3, H3K27me3, Su(z)12 and MLL1 are detected in MEFs...........................15 Figure 3.1. H3K4me3 associates with PCNA in ES cells at two stages of differentiation.......... 21 Figure 3.2. H3K27me3 associates with PCNA in ES cells of two stages of differentiation.........24 Figure 3.3.  H3K4me3 is located at the replication fork in embryos............................................26 Figure 3.4.  H3K27me3 is located at the replication fork in embryos..........................................29 Figure 3.5. H3K4me3 is located distal to the replication fork in embryos....................................30 Figure 3.6. H3K27me3 is located distal to the replication fork in embryos..................................31 Figure 3.7. MLL1 is located at and distal to the replication fork..................................................33 Figure 3.8. Su(z)12 is located at and distal to the replication fork...............................................35 Figure 3.9. H3K27ac is located on nascent DNA in embryo cells...............................................37 Figure 3.10. Comparison of H3K27me3 and H3K27ac distribution in embryo cells....................39 Figure 3.11. Effect of a H3K27 demethylase inhibitor on E14.5 embryo cells.............................40 Figure 3.12. GSK J4 PLA and CAA average graphs using H3K4me3, H3K27me3, and H3K27ac antibodies on E9.5 embryo cells..................................................................................................43 Figure 3.13. C646 PLA and CAA average graphs using H3K4me3, H3K27me3, and H3K27ac antibodies on E14.5 embryo cells................................................................................................45 Figure 4.1. Schematic of the proposed models of chromatin assembly in undifferentiated and terminally differentiated cells.......................................................................................................51 viii  List of Abbreviations  ac - acetyl ATP ? Adenosine triphosphate CAA ? Chromatin Assembly Assay CAF ? 1 - Chromatin assembly factor 1  CBP - CREB binding protein CBP ? CREB binding protein ChIP ? Chromatin Immunoprecipitation DNA ? deoxyribonucleic acid E ? Embryonic stage E(z) ? Enhancer-of-Zest  EdU ? 5-ethynyl-2?-deoxyuridine Eed ? Embryonic ectoderm development ES ? Embryonic Stem cell ES ? embryonic stem cell FEN-1 ? Flap endonuclease  H3K27 ? Histone 3 lysine 27 H3K4 ? Histone 3 lysine 4 HAT ? Histone acetyltransferase HDAC ? histone deacetyltransferase LIF ? Leukemia Inhibitory Factor LSD1 ? lysine specific demethylase 1 ME ? Maintenance Element me3 ? trimethyl MEF ? mouse embryonic fibroblast MLL ? Mixed-lineage leukemia MME ? mouse mammary epithelial  ix  Pc - Polycomb  PcG ? Polycomb Group PCNA ? Proliferating Cell Nuclear Antigen PhoRC ? Pho Repressive Complex   PLA ? Proximity Ligation Assay PRC ? Polycomb Repressive Complex  PRE ? PcG response element PTM ? Post translational modification SILAC ? Stable isotope labeling by amino acids in cell culture Su(z)12 ? Suppressor of zeste 12 TRE ? trxG response element Trx ?  trithorax trxG ? trithorax Group UTX ? ubiquitously transcribed tetratricopeptide repeat gene on X chromosome   x  Acknowledgements   I would like to thank my mentor and supervisor, Dr. Hugh Brock for his guidance, support, and patience throughout the years. I am forever grateful for the opportunity to perform research in his laboratory. To my committee members, Dr. Matt Lorincz and Dr. Don Moerman, thank you for advice, insight, and assistance. To the past and current members of the Brock laboratory, thank you for your assistance and guidance in the lab and with experimental procedures. In particular, to Sina Kovermann, whose project was fundamental in the formation of my own project. To Dr. Jacob Hodgson, thank you for the numerous, thought provoking discussions, all of the valuable advice, and training over the past number of years. To the Lefebvre lab, without which my project would not have been possible. In particular, I would like to thank Aaron Bogutz and Karen Jacob for all their help with ES cells and mouse work. To my friends, thank you for always being there to share successes and tackle the challenges, and making the past couple years so memorable. Lastly, to my family, thank you for your unwavering support and encouragement throughout these years. To my parents, brothers, sister in law, and niece thank you so much for being there for me during every step of this journey. 1  1. Introduction  1.1. DNA replication   The eukaryotic genome is packaged into the nucleus. Within differentiated cells, there are distinct regions of tightly and loosely packed chromatin. Euchromatin is open and accessible to the binding of transcription complexes and therefore is transcriptionally active (Quina et al. 2006). On the other hand, heterochromatin is tightly condensed and the majority of genes in this region are less active (Hsu, 1962; Quina et al. 2006). The structural differences between euchromatin and heterochromatin are due to changes in the post translational modifications (PTM) on the core histones and the variety of protein complexes bound to chromatin (Quina et al. 2006).   Replication of the eukaryotic genome is a highly coordinated process that occurs in the S phase of each cell cycle. In eukaryotes, there are many origins of replication throughout the genome (Okada, 1968; Waga and Stillman, 1998). The presence of multiple origins of replication spread throughout the genome ensures DNA replication occurs quickly and that the number of origins is proportional to the length of S phase (Bogan et al. 2000).  In embryos, S phase takes minutes, whereas DNA replication takes 7.3 hours to complete in differentiated cells (Okada, 1968; Bogan et al. 2000). The rate of DNA replication remains constant, so variation in S phase length depends on the number of active origins and replicon size. The replisome is a multiprotein complex which contains the proteins required for DNA replication (Waga and Stillman, 1998; Benkovic et al. 2001). The replisome ensures that replication of the genome occurs quickly, very few errors per base pair, and in a highly coordinated and regulated fashion.  The replication fork is located at the junction between double stranded parental DNA and newly synthesized DNA strands (Waga and Stillman, 1998) (See Fig 1.1).    2   Figure 1.1 Schematic diagram of DNA replication and some of the key molecules required. (http://sharonchoi--bi4u.blogspot.ca/2010/09/dna-replication.html)   Double stranded parental DNA is unwound in the replisome by Replication Protein A and used as a template for synthesizing nascent DNA (Waga and Stillman, 1998). Next, hydrogen bonds binding complementary base pairs are broken, and DNA is unwound by DNA helicase as progression through the protein complexes occurs. Nucleic acids complementary to the parental strand are added by DNA polymerases on the leading strand (Benkovic et al. 2001). On the lagging strand, short regions of double stranded DNA called Okasaki fragments are formed by Polymerase III and primed by RNA (Benkovic et al. 2001). DNA ligase joins the 1 to 3 kilo base Okasaki fragments together, resulting in one continuous DNA strand (Benkovic et al. 2001).  First identified in dividing cells, Proliferating Cell Nuclear Antigen (PCNA) is a member of a family of DNA sliding clamps that coordinates activities occurring at the replication fork (Miyachi et al. 1978; Bravo and Celis, 1980; Gogal et al. 1992; Moldovan et al. 2007). PCNA is recruited to the initiation site of DNA replication where it encircles the double stranded DNA, followed by DNA polymerase recruitment (Kelman, 1997). PCNA serves as a sliding anchor for DNA polymerase and holds the two newly synthesized DNA strands together. PCNA stimulates Flap endonuclease (FEN-1), a protein involved in the maturation of Okasaki fragments (Li et al. 1995). PCNA directly interacts with a variety of proteins, including histone acetyltransferases, DNA methyltransferases, sister-chromatid cohesion factors, and cell cycle regulators (Moldovan et al. 2007).  Thus it has been suggested that PCNA may play a key role in transferring parental chromatin proteins to nascent DNA (Petruk et al. 2012).  During DNA replication, nucleosomes are removed from parental chromatin and deposited on nascent DNA. In Drosophila  embryos, nucleosomes are detected right at the replication fork (McKnight and Miller, 1976). In front of the replication fork, an average of two 3  nucleosomes are destabilized ahead of the replication complex and the first nucleosomes can be detected 260 base pairs from the replication fork (Gasser et al. 1996).  Deposition of parental histone H3 and H4 (H3/H4) tetramers on nascent DNA occurs within seconds following the progression through the replication fork (Gasser et al. 1996; Lucchini et al. 2001). The addition of histone H2A and H2B (H2A/H2B) dimers completes the formation of nucleosomes on nascent DNA. As the amount of parental histones is limiting, newly synthesized histones will also be deposited on nascent DNA.  The histone chaperone chromatin assembly factor 1 (CAF-1) plays an important role in histone deposition. CAF-1 is associated with PCNA and responsible for depositing H3/H4 tetramers onto the nascent DNA (Bavykin et al., 1993; Lucchini et al., 2001; Hoek and Stillman, 2003; Krude and Keller, 2001; Gunjan et al. 2005; Ladoux et al., 2000). Once H3/H4 tetramers are added to nascent DNA, H2A and H2B are deposited, independently of CAF-1, to complete the nucleosome formation (Bavykin et al., 1993) (See Fig 1.2).  Stable isotope labeling by amino acids in cell culture (SILAC) experiments in HeLa cells show that modified parental histones, including dimethyl histone H4 lysine 20 (H4K20me2), monomethyl histone H3 lysine 9, 4, and 27 (H3K9me1, H3K4me1, and H3K27me1 respectively) are deposited on nascent DNA (Scharf et al. 2009). Acetylated parental histones are deposited on newly replicated DNA and are deacetylated during the cell cycle (Scharf et al. 2009). SILAC experiments on HeLa cells show that histone methylation associated with active genes occurs faster than that associated with silenced genes (Zee et al. 2010).  Unmodified newly synthesized histones are incorporated into nucleosomes on nascent DNA. SILAC experiments in HeLa cells have shown that newly synthesized histones acquire histone methylation during the cell cycle (Schard et al. 2009). For example, H4K20 is monomethylated on newly synthesized histones following incorporation into chromatin. Rtt106 is a histone chaperone for newly synthesized histones in yeast (Fazly et al. 2012). It is thought that acetylation of histone H3 on lysine 56 (H3K56ac) on newly synthesized histones increases the binding affinity between CAF-1 and Rtt106, therefore increasing the deposition efficiency of H3/H4 (Masumoto et al. 2005; Han et al. 2007; Chan et al, 2008).   4    Figure 1.2. Schematic of nucleosome assembly. Modified with permission from Henikoff (2003) Nature 423: 814-817   1.2. Histone modifications Histones can be post-translationally modified in the core or the histone tails. The combination of covalent modifications on the histone tails projecting from the nucleosome forms the histone code. The histone code hypothesis suggests that post translational modifications work in combination and/or sequentially, and that the modifications are recognized by chromatin proteins that in turn change chromatin structure (Strahl and Allis, 2000; Yu et al. 2011). Histone modifications include methylation, acetylation, phosporylation, and ubiquitylation and many others (Lennartsson and Ekwall, 2009; Bannister and Kouzarides, 2011).  See Fig. 1.3. Histone modifications are correlated with the transcriptional status of a given gene. Histone modifications, including H3K4me3, H3K36me3, and H3K9ac are associated with active gene states and are abundant in euchromatic regions (Barski et al. 2007; Lennartsson and Ekwall, 2009). On the other hand, H3K9me2, H3K9me3, and H3K27me3 are associated with transcriptionally silenced genes (Huisinga et al. 2005; Barski et al. 2007; Lennartsson and Ekwall, 2009). Although some mutations in genes that alter the histone code have no phenotype, there are many mutations that result in alterations to the histone code that cause misexpression of genes, developmental defects, and disease (Lennartsson and Ekwall, 2009).   5   Figure 1.3. Schematic diagram of modified histone tails and chromatin proteins bound to the modifications. Modified with permission from Bannister and Kouzarides (2011) Cell Research 21:381-395   Histone acetylation is a labile modification deposited by histone acetyltransferases (HAT) and removed by histone deacetyltransferases (HDAC) (Wade et al. 1997; Kouzarides, 2000). Histone acetylation affects the overall charge of a histone tail, removing the negative charge of DNA and increasing nucleosome fluidity (Yu et al. 2011). Histone acetylation also recruits the SWI/SNF complex, a nucleosome remodelling complex (Wade et al. 1997; Bannister and Kouzarides, 2011). CREB binding protein (CBP) is a HAT targeting a variety of proteins including histones and transcription factors. Histone acetylation is associated with active gene states. CBP functions to stabilize the transcription complex (Martinez-Balbas et al., 1998; Partanen et al., 1999; Kalkhoven, 2004; Wang et al. 2009; Zenter et al., 2011). Sequencing of DNA fragments retained in chromatin immunoprecipitation (ChIP-seq) experiments have shown that HATs and H3K27ac are located at the enhancers and around the transcription start site of active genes (Wang et al. 2009; Zenter et al., 2011). H3K9ac and H3K14ac are located at transcriptionally active promoters.  HDACs oppose the effect of HATS.  Like HATS, HDACs are located at the promoter and in intergenic regions of active genes (Wang et al. 2009).  Multiple histone methyltransferases have been identified.  The trithorax group (trxG) of proteins maintain an active gene state.  Trx knockouts in flies exhibit severe phenotypic defects including the failure to activate Hox gene expression and anterior transformations (Eissenberg 6  and Shilatifard, 2010). Trx, ASH1, and ASH2 H3K4 trimethylases, and other members are subunits of nucleosome remodelling complexes (Fisher and Brock 2005). In mammals, there are at least six H3K4 trimethylases, including the Mixed lineage leukemia (MLL) family of proteins that are homologues of Trx  (Wang et al. 2009). MLL and Trx proteins contain the PHD and SET domains (Madan et al. 2009). PHD domains contain a zinc finger and bind to chromatin (Schindler et al. 1993). SET domains have specific lysine methyltransferase activity.   MLL1 is required for H3K4 trimethylation of less than 5% of promoters with H3K4me3 (Wang et al. 2009). MLL1 and H3K4me3 are localized with RNA polymerase II at the 5? end of active genes (Guenther et al. 2005). MLL2, created though a gene duplication of MLL1 and is required during development (Glaser et al. 2009). Embryonic mice that lose MLL2 following E10.5 do not display abnormal phenotypes, suggesting that MLL2 is not required following E10.5 or in adults (Glaser et al. 2009). MLL3/4 monomethylates H3K4me at the enhancer region of genes (Morgan and Shilarifard, 2013). MLL5 is required for hematopoesis and MLL5 knockouts have altered DNA methylation patterns (Heuser et al. 2005).  H3K4me3 is associated with transcriptionally active genes. The presence of H3K4me3 in the coding region results in a 10 fold increase in luciferase expression (Okitsu et al. 2010). ChIP-Seq data has shown that H3K4me3 is located at the transcriptional start site and in the 5? coding region of genes (Barski et al. 2007; Eissenberg and Shilatifard, 2010; Schneider et al., 2003; Koch et al., 2007). H3K4me3 is also detected at the distal enhancer regions of interferon-? in T cells (Barski et al. 2007). Using a genome wide approach, H3K4me3 modified histones are detected at the sites of transcriptional initiation in human cells (Guenther et al. 2007).  The Polycomb Group (PcG) of proteins are transcriptional silencers. Polycomb Repressive Complex (PRC) 2 members, including  Suppressor of zeste 12 (Su(z)12), Enhancer of Zeste Homolog 2 (EZH2), and Embryonic ectoderm development (Eed) trimethylate H3K27 (Cao et al., 2002; Pasini et al., 2004; Bernstein et al., 2006; Pasini et al., 2010;  Kim and Kim, 2011). EZH2 contains a SET domain for H3K27 trimethylation (Chase and Cross, 2011). H3K27me3 is associated with transcriptionally silenced genes and is thought to recruit and facilitate binding of the PcG proteins and chromatin compaction (Hansen et al, 2008; Chase and Cross, 2011).  Nevertheless, experiments in flies where the SET domains of Su(var)3-9 that methylates H3K9 and the EZH2 homologue E(z) are switched do not result in a change in recruitment of PcG proteins (Henikoff and Shilatifard 2011). PcG gene heterozygous mice show segmental abnormalities and knockout mice are embryonic lethal (Yu et al. 1995; Pasini et al. 2004). 7  H3K27me3 ChIP-Seq data has indicated that H3K27me3 is spread upwards of 10000 base pairs upstream and downstream of the transcriptional start site of transcriptionally inactive genes (Barski et al. 2007; Pan et al. 2007). H3K27me3 spreads throughout the Ubx locus and encompasses the PREs (Cao et al., 2002; Kahn et al., 2006), and maintains a silenced Ubx state (Papp and Muller, 2006). PcG mutants show derepression of Hox genes and undergo posterior transformations (Ingham, 1998; Birve et al. 2001; Chen et al. 2008; Eissenberg and Shilarifard, 2010).  Interestingly, there are domains that contain both H3K4me3 and H3K27me3. In these bivalent domains, H3K4me3 signals are highest around the promoter region of genes whereas H3K27me3 is distributed throughout the gene (Barski et al. 2007). In human ES cells, bivalent domains are detected at genes encoding lineage specific transcription factors (Pan et al. 2007). In undifferentiated cells, these bivalent domains may maintain pluripotency (Meshorer and Misteli, 2006; Rugg-Gunn et al. 2010; Abraham et al. 2013; Kim et al. 2011). In hematopoiesis, bivalency at the transcriptional start site is observed in undifferentiated cells (Abraham et al. 2013). Furthermore, RNA polymerase II is bound at genes with the bivalent marks, suggesting that these genes are in a paused gene state (Abraham et al. 2013). In human ES cells, 16% of genes have both H3K4me3 and H3K27me3 (Pan et al. 2007).  It was thought initially that histone methylation was a permanent modification and histone turnover was slow (Byvoet et al. 1972; Agger et al. 2008). The first histone demethylase, lysine-specific demethylase 1, LSD1, was identified 9 years ago (Shi et al. 2004). LSD1 demethylates H3K4 and thus is a transcriptional corepressor. As with histone methyltransferases, histone demethylases are residue-specific (Bannister and Kouzarides, 2011). Multiple demethylases have been identified that play a key role in development and diseases including cancer (Agger et al. 2008; Quina et al. 2006).  Histone modifications interact with each other. For example H2B and H2A ubiquitylation precede methylation of H3K4 or H3K27 respectively (Suganuma and Workman, 2008). Antagonism also exists between some histone modifications. H3K27 is modified by both acetylation and methylation. RNAi knockdown experiments for PRC2 members in S2 cells have shown that when H3K27me3 levels decrease, H3K27ac levels increase (Tie et al., 2009). CBP knockdown experiments showed the opposite result, namely H3K27ac levels decreased and H3K27me3 levels increased. Combined, these data show that H3K27ac and H3K27me3 are mutually exclusive. Another study using ES cells also showed that H3K27me3 and H3K27ac are 8  antagonistic marks at the global level (Pasini et al., 2010). Knockout ES cells for PRC2 members Su(z)12 and Ezh2 resulted in a lack of H3K27me3 and a significant increase in H3K27ac using western blot experiments.  Histone methyltransferases and demethylases are found in the same protein complexes and it is thought that the removal of one methyl mark is closely followed with the addition on a different residue (Fig. 1.4) (Agger et al. 2007; Shi 2007; Cloos et al. 2008; Pasini et al. 2008). For a gene to switch from an active to a repressed gene state, H3K4me3 will be removed by RBP2, followed by acetylation of H3K4 and trimethylation of H3K27 by PRC2 (Pasini et al. 2008). In contrast, the H3K4 methyltransferase MLL1 is found in the same complex as the H3K27 demethylase UTX (Cloos et al. 2008; Pasini et al. 2008). This allows for the simultaneous removal of H3K27 trimethylation followed by trimethylation of H3K4. Histone acetyltransferase acetylates H3K27, thereby blocking the site from methylation.   Figure 1.4.  Schematic showing the model alternating between active and repressed gene states. Modified with permission from Pasini et al. (2008) Genes and Development, 22: 1345-1355   1.3. Changes in chromatin during development During development, gene expression patterns become restricted as differentiation occurs down a particular cell lineage. Transcription factors initially determine whether or not a gene will be expressed. During each cell division, cytoplasmic determinants and transcription 9  factors are diluted and eventually, will no longer be present (Simon, 1995). Thus, maintenance of gene expression patterns must be established to ensure their inheritance in daughter cells.   There are changes in chromatin structure during development. In pluripotent cells, chromatin is open and DNA is freely accessible to transcription factors. During differentiation, cell fates become restricted and distinct heterochromatic and euchromatic regions can be seen (Meshorer et al. 2006; Golob et al. 2008). Particular loci will be accessible to transcription factors and RNA polymerase II when transcribed. During this time histone marks associated with an active gene state will also be present (Pasini et al. 2008). When a particular transcript is no longer required, active histone marks will be removed and replaced with repressive histone marks. The structure of chromatin will also change, and become less accessible to transcriptional machinery (Agger et al. 2007; Pasini et al. 2008; Abraham et al. 2011; Kim et al. 2011).  During development, there is a change in the types of histone modifications present in the genome. ChIP experiments at the promoter of a gene involved in mesodermal differentiation show that H3K4me3 is only present five days following induction of ES cell differentiation when the gene is active (Lee et al. 2004). H3K4me3 is absent when this gene is no longer expressed. H3K9me3 levels increased during ES cell differentiation (Meshorer et al. 2006). Human mesenchymal stem cell differentiation experiments have shown that H3K4me3 increases at the promoter and 5? end of genes involved in chondrogenesis, cartilage formation (Herlofsen et al 2013). Furthermore, these genes are also transcriptionally upregulated during this time. Genes which are downregulated during chrondogenesis showed an increase in H3K27me3 (Herlofsen et al. 2013).   1.4. Inheritance of chromatin marks Epigenetics refers to heritable changes in chromatin structure that do not affect the DNA sequence required to maintain gene expression patterns in daughter cells. This definition is conceptually useful, but operationally unhelpful.  No change in chromatin structure affects DNA sequence, so all chromatin changes that affect gene regulation are ?epigenetic?.  In this sense ?epigenetic? is a synonym for gene regulation.  Unfortunately, at this time there is no functional test for heritability.  It is often overlooked that any change in transcriptional regulation will affect maintenance, and any change in maintenance will affect gene regulation.  Therefore mutational analyses that are not cell-cycle stage-specific do not allow us to distinguish chromatin changes that affect transcription and maintenance. Until it is possible to make changes in chromatin that 10  act only during DNA replication or mitosis, and it can be shown that these changes affect transcription in the next cell cycle, it will not be possible to define which (if any) events are epigenetic.   It is widely assumed that DNA replication destabilizes chromatin, and thus that there must be a molecule or molecules that identify the transcriptional state of a gene, that are stable to replication and mitosis, and that are sufficient to reconstitute the chromatin state of active or repressed genes (Corpet and Almouzni 2009).  These putative molecules are called epigenetic marks.  So far, DNA methylation is the best analogue for an epigenetic mark because it is associated with transcriptional repression, is stable to replication, and DNA hemimethylase provides a mechanism to propagate the mark in nascent chromatin (Jones and Takai, 2001; Bird, 2002; Lorincz et al 2004;  Lande-Diner et al. 2007).  There are multiple theories in the field on how chromatin is assembled during DNA replication. An influential theory proposes that histone modifications are the epigenetic mark deposited on nascent DNA (Jackson and Chalkley, 1985; Corpet and Almounzie, 2009; Jasencakova and Groth, 2010). These modifications will serve as recruiters for modifying proteins to bind. As a convenient shorthand, I will refer to this as the ?modification first? model. Another related model suggests that chromatin assembly is dynamic and it is the combination of histone methylation, methyltransferases, and demethylases that affect chromatin assembly (Agger et al. 2007; Shi 2007; Cloos et al. 2008; Pasini et al. 2008). Lastly, a recent study in Drosophila  embryos showed that modifying enzymes are recruited to nascent DNA and serve as the epigenetic mark (Petruk et al. 2012). I will refer to this as the ?modifier first? model.  I will explain these models below.    The histone modification first model assumes that nucleosomes on nascent DNA are composed of a mixture of modified and unmodified histones. The model proposes that histone modifications on parental histones are transferred to nascent DNA, are recognized by histone modifying enzymes, which in turn modify the newly synthesized histones that lack post-translational modifications, to replicate the patterns established in the parental cell. Parental histones with their modifications, including H3K4me3 and H3K27me3, are thought to be randomly deposited on nascent DNA, (Jackson and Chalkley, 1985; Corpet and Almounzie, 2009; Jasencakova and Groth, 2010). SILAC experiments on HeLa cells have shown that histone methylation associated with active genes occurs faster than that associated with silenced genes (Zee et al. 2010). SILAC experiments in HeLa cells have also shown that modified histones, including H4K20me2, H3K9me1, H3K4me1, and H3K27me1 are deposited on nascent DNA (Scharf et al. 2009). Methylated histones and PcG proteins are recruited to ME 11  prior to their replication in late S phase (Lanzuolo et al. 2011). However, the experiments cited do not examine events which occur soon after replication. Experiments in HeLa cells have shown that PRC2 is recruited to and binds to H3K27me3 which results in trimethylation of newly synthesized histones (Hansen et al. 2008; Jasencakova and Groth, 2010). The conclusions drawn by the authors of these studies provide interesting examples of why it is difficult to identify an epigenetic mark. Neither study identifies when recruitment of EZH2 is important. If the modifier is recruited during transcription, and is stable to replication, then newly synthesized histones will be modified, not because the me3 mark is important in nascent DNA, but because it is important in transcription.  Thus these studies do not show that H3K27me3 is or is not an epigenetic mark.  On the other hand, there is growing evidence that histone modifying enzymes are stable to DNA replication. In HeLa cells,H3K9 is monomethylated prior to deposition on nascent DNA (Jasencakova and Groth, 2010). SETDB1 is in a complex with CAF-1 and HP1?, and HP1? is thought to recruit the histone methyltransferase Suv39, resulting in H3K9me3 on nucleosomes of nascent DNA (Jasencakova and Groth, 2010). In mammalian cell lines, the H3K27 histone methyltransferase EZH2 co-localizes with both PCNA and BrU (Hansen et al. 2008).  Using the SV40 in vitro replicating system, the Psc and Pc subunits of the PRC1 PcG complex were detected on nascent DNA (Francis et al. 2009). Furthermore, Francis et al. (2009) propose that PcG complexes are rapidly bound to nascent DNA immediately following progression through the replication fork. In this model of chromatin assembly, the modifying enzyme is recruited to nascent DNA first, and histone modification occurs afterwards.  Previous studies have looked at chromatin assembly distal to the replication fork. Long DNA labeling times of nascent DNA mean that events at the replication fork or immediately following the passage through the replication fork are not detected (Scharf et al. 2009; Hansen et al. 2008). Western blots and ChIP experiments use material from a mixed population. Synchronized cells ensure the cells are in the same stage of the cell cycle, however, there are many replication origins and therefore one cannot distinguish between events occurring at the replication fork, immediately following the passage through the replication fork, or much later.  These experiments make it difficult to study the activities at the replication fork and shortly after the passage through the replication fork.  Petruk et al. (2012) used reChIP experiments, using material following an IP with PCNA, to separate chromatin fragments that contain the replication fork from those that don?t, followed 12  by sequential immunoprecipitation with antibodies to the protein or histone modification of interest, with PCR analysis.  This procedure allows analysis of what proteins are present at the replication fork for specific loci.  This procedure was modified by labelling embryos for short periods with BrdU, followed by immunoprecipitation with antibodies to BrdU to identify newly synthesized DNA.  Proteins or modifications of interest bound to the newly synthesized DNA are identified using western blotting (Francis et al. 2009; Petruk et al. 2012). The disadvantages of these methods is that a lot of tissue is required to ensure that enough PCNA positive fragments are isolated in the IP for the ChIP assay, and that unless they are performed in cell lines, the results are the aggregate of all the cell types in the sample.  However, Petruk et al. 2012 have developed two new experimental techniques to investigate how histones and chromatin proteins are assembled at the replication fork and on nascent DNA immediately following the progression through the replication fork that overcome these limitations. The first experimental technique used was the proximity ligation assay (PLA) described in the Materials and Methods. In PLA experiments, antibodies raised in different organisms directed against any 2 epitopes are used in fixed tissue to detect the epitopes of interest (Fig. 1.5). Species-specific secondary antibodies coupled to DNA oligomers are allowed to bind the primary antibodies. Proximity of the two secondary antibodies is detected by adding primers that link the oligomers on each secondary antibody.  Rolling circle amplification in the presence of a fluorescently tagged nucleotide occurs when the secondary antibodies are less than 40nM apart.  The sensitivity is estimated to be 10-42M, which is enables detection of transient single-molecules interacting with each other in vivo.  Petruk et al. (2012) used an antibody to PCNA in combination with antibodies to histone modifications or chromatin proteins to identify proteins found close to the replisome. Thus, the presence of modified or unmodified histones as well as chromatin proteins can be studied at the global level in cells, tissue sections, or embryos using the PLA assay. In the second technique, the Chromatin Assembly Assay (CAA), EdU is incorporated into newly synthesized DNA (Petruk et al. 2012). Biotin azides are conjugated onto EdU through a Click-IT? reaction. Then an ?-biotin antibody in combination with an antibody to modified histones or chromatin proteins can be used in the PLA reaction to determine what molecules are present on nascent DNA. The average rate of progression of DNA polymerase is 1-3 kb/min, so a five minute labeling time will result in a 10-15 kb of labelled DNA. In practice, it takes some time for the EdU to be taken up and to reach the nucleus in sufficient concentrations, although we have successfully labelled for 2.5 min, so this assay likely 13  analyzes less than 10 kb of DNA. Both the PLA and CAA assay use a genome wide approach to studying chromatin assembly.    Figure 1.5. Overview of the Proximity Ligation Assay. Modified with permission from Olink Biosciences  Using these techniques in 0-12 hour Drosophila embryos, Petruk et al. (2012) detected the histone methyltransferases Trx, Pc, and E(z) at the replication fork and on nascent DNA but H3K4me3 or H3K27me3 were not detected. This result supports the modifying enzyme first model of chromatin assembly. These observations were confirmed in re-ChIP experiments where input material was precipitated with antibodies to PCNA to isolate chromatin fragments in S phase.  The trxG and PcG proteins Trx and E(z) are recruited to their binding sites (maintenance elements (ME) also known as trxG response elements (TRE) or PcG response elements (PRE)) following passage through the replication fork (Petruk et al. 2012). Unmodified H3 was detected in PLA and CAA experiments, but methylated histones were not detected until the following G2 phase, an hour following progression through the replication fork (Petruk et al. 2012). These results are surprising and raise a number of questions with respect to the mechanism of chromatin assembly in Drosophila , in other species, and during development. Does Drosophila  have a unique form of chromatin assembly or is this a universal form? Is there a difference in chromatin assembly in embryos and differentiated cells? What types of 14  enzymatic regulation could account for these observations? Is the activity of histone methyltransferases regulated? Are demethylases functioning to remove the methyl marks? Is there antagonism between histone modifications? To determine if Drosophila embryos have a unique form of chromatin assembly, Sina Kovermann in our lab performed some of the same experiments using Drosophila S2 cells. S2 cells are a primary cell line derived from 20-24 hour embryos (Schneider 1972). She showed  that H3K4me3, H3K27me3, Trx, and Ph could be detected on nascent DNA. (Figure 1.6).   Figure 1.6. H3K4me3, H3K27me3, Ph, and Trx are detected on nascent DNA in S2 cells.  (A) Representative cell images for CAA experiments using the indicated antibodies. Top panel is the merged image. Bottom panel is the CAA signal onlbut . DAPI in blue, EdU positive cells in green, CAA signal in red. (B) Average CAA values for the indicated antibodies. EdU-IgG control in blue. Experimental antibodies in red. All values are mean + SEM.   Combined, the results in Drosophila  embryos and S2 cells show that there is more than one form of chromatin assembly in Drosophila . The first form is seen in embryos, where histone methyltransferases are present but modified histones are absent. The second form is seen in 15  S2 cells, where both histone methyltransferases and methylated histones are detected.  These results raise a number of questions. Is there a difference in chromatin assembly in cell lines and embryos? Do immortalized cells behave differently to primary cell cultures? Can the Drosophila  embryo and S2 cell differences be a result of differentiation? To determine if immortalized cell lines have a unique form of chromatin assembly, the same PLA and CAA experiments were performed in mouse embryonic fibroblasts (MEFs) and mouse mammary epithelial (MME) cell lines. There are two MME cell lines that were tested, one that is Ras transformed and one that is not. MEF PLA and CAA data is shown in Figure 1.7. All cell lines, whether they are Drosophila or mouse, behave the same. Methylated histones and histone methyltransferases are detected at the replication fork in PLA experiments. In CAA experiments, H3K4me3, H3K27me3, MLL1, and Su(z)12 are detected on nascent DNA.    Figure 1.7. H3K4me3, H3K27me3, Su(z)12 and MLL1 are detected in MEFs.  (A) Average PLA values for the indicated antibodies. PCN-IgG control in blue. Experimental antibodies in red. All values are mean + SEM. (B) Average CAA values for the indicated antibodies. EdU-IgG control in blue. Experimental antibodies in red. All values are mean + SEM. 0204060H3K4me3 H3K27me3 Su(z)12 MLL1Average PLA signal per PCNA positive cellExperimental antibodyMEF PLAIgGExperimental0246810H3K4me3 H3K27me3 Suz12 MLL1 Average CAA signal per EdU positive cellExperimental antibodyMEF CAAIgGExperimentalA B 16  1.5. Thesis Objectives  The mechanism of chromatin assembly is poorly understood. Furthermore, experiments have indicated that chromatin changes during development, but changes in chromatin assembly during development remains uninvestigated. Chromatin assembly at the replication fork and on nascent DNA can now be studied using PLA and CAA as outlined by Petruk et al. (2012).  The goal of my thesis are first to determine if the kinetics of H3K4 and H3K27 methylations change during murine development using two differentiating systems, embryonic stem (ES) cells, and murine embryos. To determine if the changes in histone methylation are due to presence of absence of histone methyltransferases, I will determine if the trxG protein MLL1, and the PcG protein Su(z)12 binding to the replisome or to nascent DNA is regulated. Third, I will determine if H3K27 methylation is antagonized by H3K27 acetylation. Together, these experiments will allow me to determine if H3K4 and H3K27 methylation kinetics during differentiation, and to test 2 models for regulation of this process.       17  2.  Materials and Methods  2.1    Preparation of ES cells   Murine R.1 ES cells were kindly given to us from the Lefebvre lab. Undifferentiated ES cells were grown in EmbryoMax DMEM (Millipore SLM-120-B) media supplemented with 2mM L-glutamine (Thermo Scientific SH3003401), 0.1mM MEM nonessential amino acids (Gibco 11140), 0.1mM ?-mercaptoehtanol (Sigma M3148), 1mM sodium pyruvate (Gibco 11360-070), penicillin-streptomycin (Gibco 15140-122), 15% fetal bovine serum (Sigma F1051), and Leukemia inhibitory factor (LIF) (courtesy of the Lefebvre laboratory).  Cells were grown in a humid incubator set at 37?C and 5% CO 2. ES cell differentiation was by removing LIF from cell media.  The ES cells were trypsinized and plated onto gelatinized 8 well chamber slides (Fisher 1256518), and allowed to attach to the slides overnight prior to use in experiments.  2.2     Preparation of trypsinized embryo cells Karen Jacob from the Lefebvre lab set up all mouse matings using the CD-1 mouse strain. E9.5 and E14.5 embryos were dissected out of the conceptus and placed in ice cold PBS. E9.5 embryos were treated with 0.25% trypsin-EDTA (Gibco 25200-056) for 30 minutes at 37?C, resuspended in media and plated into 8 well gelatin coated chamber plates.  E14.5 embryos were treated with 0.25% trypsin-EDTA for 1 hour at 37?C, homogenized, and then incubated for an additional 30 minutes prior to resuspension and plating. DMEM media (Gibco 11965-092) was supplemented with 1% L-Glutamine, 10% FBS and 1% pen/strep. Cells were allowed to settle overnight in a humid incubator set at 37?C a nd 5% CO2 prior to use in PLA and CAA experiments.  2.3     Preparation of cells for PLA and CAA experiments  Cells were rinsed with PBS, fixed for 20 minutes in 4% formaldehyde in PBS at 4?C, followed by 2 1xPBS washes. Cells were cracked open using ice cold 100% methanol for 10 minutes at -20?C then rehydrated using 1% bovine se rum albumin in PBS for 30 minutes at room temperature. All blocking was performed using 2% casein in maleate acid buffer and primary antibodies were incubated overnight at 4?C.     18  2.4    Proximity Ligation Assay All proximity ligation assay (PLA) experiments were performed using the Duolink? kit from Olink Bioscience according to the manufacturer?s protocol. The following dilutions of antibodies were used: 1/20 000 PCNA (Sigma P-8825); 1/50 000 H3K4me3 (Active motif 39159); 1/5000 H3K27me3 (Millipore 07-449); 1/10 000 H3K27ac (Active Motif 39133); 1/50 MLL1 (Bethyl A300-086A); or 1/5000 Suz(12) (Abcam ab12073).  IgG controls were performed using PCNA with IgG (Abcam ab27478) at the same dilution as the experimental antibody. Counterstaining for PCNA positive cells used 1/250 Alexa Fluor? 448 goat ? mouse (Invitrogen A11029) for 1 hour at room temperature. Slides were air dried and mounted in mounting media with DAPI. 2.5     Chromatin Assembly Assay DNA was labeled with 60uM of EdU in media for 5 minutes prior to fixation for trypsinized mouse embryo experiments. 30uM of EdU was used for labeling ES cells.  Biotin azides (Invitrogen B10184) were incorporated into the EdU labeled DNA using the Click-IT reaction (Invitrogen C10269) as per the manufacturers? protocol.  The following dilutions of primary antibodies (sources as listed above) were used: 1/1000 Biotin (Sigma B7653); 1/50 000 H3K4me3; 1/5000 H3K27me3; 1/10 000 H3K27ac; 1/50 MLL1; or 1/5000 Suz(12).  Counterstaining for EdU positive was done using 1/250 Alexa Fluor? 448 goat ? mouse (Invitrogen A11029) for 1 hour at room temperature.   2.6     PLA and CAA data analysis A Leica DMI 6000 B inverted fluorescent microscope was used to capture images using Openlab 5.0. ES cell images were taken at 40x magnification and embryo cell images were taken at 100x magnification. A minimum of 100 PCNA or EdU positive murine embryos cells were counted. To avoid bias, counting was done in a blinded fashion where random slides were counted at any given time. I determined the signal per cell for a given well and marked down the well location. Once the counting was completed, I used the well location to determine which set of antibodies was used. Each experiment was replicated 2 times, and for each, I carried out 2 technical replicates. For ES cells, poor attachment of the cells to the slides led to very high loss during the PLA and CAA, so for ES cells I counted as many cells as possible. For inhibitor experiments, all possible cells were counted to a maximum of 100 cells per treatment, but toxicity of the inhibitor made it difficult to obtain uniform cell numbers. I recorded data in PLA 19  experiments only in cells in S phase as determined by immunohistochemical detection of PCNA.  In CAA experiment, I recorded data only in cells that counterstained for EdU. For each cell, I counted the number of PLA signals lying over the nucleus (determined by DAPI staining of DNA). These values were imported into Excel. The data were plotted two ways.  First I made bar graphs generated based on the average number of signals per positive cell + the SEM to determine if the overall PLA or CAA signal frequency varied. To determine if there was one or more populations of cells in a samples, graphs of the frequency distribution were made by plotting the number of cells against the PLA or CAA signal. If average signal levels were below 5 per cell, distributions were plotted for signal in individual cells. If average signal levels were above 5 per cell, signal was clustered into groups of 5. The control values were plotted on the same graph to simplify comparison.  The number of signals per cell was highly variable, so I did not calculate any measure of variance.  This variability could come from many causes: 1) there are many different cell types in embryos; 2) the cells can be pluripotent to unipotent, and thus be at very different stages of differentiation; 3) S phase is about 7-8 hours to the PLA and CAA experiments can capture cells early, middle, or late in S phase.  Thus different amounts of the genome can be already replicated; 4) following the logic of the previous point, euchromatin tends to replicate early, whereas heterochromatin replicates later, so different cells will have different ratios of euchromatin vs heterochromatin, which likely have different relative amounts of H3K4me3 and H3K27me3. 2.7     Treatment of embryo cells with inhibitors Two inhibitors were used to treat E9.5 and E14.5 cells. GSK J4 (Tocris bioscience 4594) inhibits the H3K27me3 demethylase UTX. E14.5 cells were treated for 1 hour with 25uM of GSK J4 dissolved in DMSO (Sigma 472301-100mL). E9.5 cells were treated for 1 hour with 4uM, 10uM, and 25uM GSK J4. All treatments occurred in the humid incubator at 37?C and 5% CO 2. C646 (Sigma SML 0002-5mg) is a competitive inhibitor for the histone acetyl transferase CBP that acetylates H3K27. E14.5 cells were treated for 30 minutes using 2.5uM or 1uM C646 dissolved in DMSO. Controls for these inhibitors used the DMSO alone. EdU labeling for CAA experiments was done in the presence of the inhibitors.     20  3.  Results  3.1. Active and repressive histone marks are located at the replication fork in ES cells In Drosophila  embryos, the H3K4me3 modification associated with active genes is not found at the replication fork, as determined using PLA assays that monitor colocalization of the PCNA and H3K4me3 (Petruk et al. 2012).  This result was completely unexpected because it was assumed that parental histone modifications were transferred to nascent DNA.  One explanation for these results is that chromatin assembly on nascent DNA in Drosophila embryos is exceptional.  To test this possibility, Sina Kovermann in our lab carried out similar PLA experiments with PCNA and H3K4me3 in the S2 embryonic cell line.  This immortalized line is aneuploid, and its germ line of origin is uncertain.  In contrast to Drosophila embryos, in Drosophila S2 cells, PCNA and H3K4me3 colocalize in PLA assays.  This result raised two interesting possibilities.  One is that chromatin assembly on nascent DNA changes during development.  Another is that assembly of nascent chromatin in Drosophila differs from that in other organisms.  To test these ideas, I chose to use a murine ES system.   In this system ES cell differentiation can easily be induced by the removal of LIFs from cell culture media.  I chose 0 and 4 days of differentiation, reasoning that the first time point should mimic embryonic state, and 4 days a differentiated state. Many aspects of the experimental design are common to most of the experiments below. I outline these here.  In all experiments, cells were labelled with DAPI to identify the nucleus, and counterstained with antibodies to PCNA, which marks cells in S phase, and PLA was carried out with antibodies to PCNA and an IgG control.  Together, these controls allowed me to accurately determine the non-specific background signals. I show images of the data, and then quantitate the data by plotting a frequency distribution showing the number of cells with a given number of PLA signals. While most frequency distributions approximate a normal distribution, in some cases I obtained bimodal distributions that suggest that there are different populations of cells in the sample.   I counted the number of PLA signals over the nucleus from 50 cells in S phase.  On the same plot, I also show the frequency distribution for the control. This allows me to determine with reasonable sensitivity if a signal is present.    In undifferentiated ES cells, I first determined the average number of PLA signals in S phase cells and obtained a value of 3.36 + 0.38 (Fig. 3.1B).  Four days after inducing     C      Figure 3.1. H3K4me3 associates with PCNA in ES cells at two stages of differentiation. (A) Representative cell 40x  images for PLA experiments at zero and four days of differentiation. PLA signal is in red, PCNA in green, and DAPI in band 4 days of differentiation. Values are mean days of differentiation. PCNA-IgG signal in blue, PCNAof PLA signal at four days of differentiation. red. 024680 daysAverage PLA signalper  cell01020300 5Number of cellsPLA signal 0 daysPLA          Merge A       B           D lue. (B) Average PCNA-H3K4me3signal0 + SEM.  (C) Distribution of PLA signal at zero -H3K4me3 signal in red. (D) Distribution PCNA-IgG signal in blue, PCNA-H3K4me3 signal in 4 daysDays of differentiationIgGH3K4me310IgGH3K4me302040600 10Number of cellsPLA signal 4 days0 days     4 days 21    20IgGH3K4me322  differentiation, the average number of PLA signals per cell was 6.62 +0.54 (Fig 3.1B).  Therefore the average number of PCNA-H3K4me3 signals increases with differentiation of ES cells. However, I noticed that at zero days, half the cells in S phase had background levels of PLA signal, and the remainder had signal.  I reasoned that there may be two populations of cells, one that behaved like Drosophila embryos (no or very low PCNA-H3K4me3 PLA), and others that behaved like S2 cells (PCNA-H3K4me3 signals present).  To test this idea, I plotted the number of cells against the number of PLA signals per cell.  As can be seen in Fig. 3.1C, the distribution is bimodal.  25 cells have 1 or fewer PLA signals, and 25 cells have 2 or more signals.  I propose that like Drosophila, murine embryonic and differentiated cells have different modes of H3K4me3 assembly on nascent DNA.  The distribution of cells at 4 days may be bimodal, but the small sample size for the cells with low PLA signals means that I cannot rule out the possibility that there distribution is normal, and the apparent bimodality is an artifact.  If the population is bimodal, the medians of each population are shifted to the right compared to the frequency distribution at 0 days.  Combining these data, I propose that the amount of H3K4me3 found on nascent DNA increases with differentiation.  These data refute the hypothesis that Drosophila embryos have an unusual or exceptional assembly of nascent chromatin because both undifferentiated ESC and Drosophila embryos have very low amounts of H3K4me3 in nascent chromatin.  These data also suggest that Drosophila and mammals share features of chromatin assembly.   The H3K27me3 modification is associated with repressed genes and in Drosophila  embryos, and is not found at the replication fork using the PCNA-H3K27me3 PLA assay (Petruk et al. 2012), suggesting that trimethyl modifications associated with repression behave similarly to those associated with activation.  To test the hypothesis that chromatin assembly in Drosophila embryos is unusual, Sina Kovermann performed similar PLA experiments on S2 cells using PCNA and H3K27me3. H3K27me3 and PCNA do colocalize in PLA assays. This result indicates that chromatin assembly on nascent DNA in embryos and cell lines is different. The results from the previous experiment predict that in the murine ES system, H3K27me3 should behave like H3K4me3.   In undifferentiated ES cells, the average number of PLA signals in 50 S phase cells is 2.72 + 0.32 (Fig. 3.1B). Four days after inducing differentiation, the average number of PLA signals per cell is 6.16 + 0.52 (Fig.2.1B). Therefore, the average number of PCNA-H3K27me3 signals increase during ES cell differentiation. Like the results with H3K4me3, there appeared to be two populations of cells, one with very low amounts of signal, and one with more. After 23  plotting the number of cells against the number of PLA signals per cell, as seen in Fig. 3.2C, there is a bimodal distribution of PCNA-H3K27me3 signal present. Sixteen cells had background PLA signals, and the remainder had 1 or more signals. I propose that undifferentiated and differentiated murine ES cells have different modes of H3K27me3assembly on nascent DNA.  However, the data are much less clear for H3K27me3 than for H3K4me3.  This may be because there are fewer signals with the H3K27me3 antibody, or because the kinetics of developmental change in H3K27me3 assembly into nascent chromatin differs from those of H3K4me3. In general, experiments using ES cells were technically difficult. I found that cells did not adhere to gelatine coated chamber slides or glass cover slips very well, so I lost a lot of cells during PLA experiments. Ideally, I would like to have counted 100 PCNA positive cells per experiment to increase the sensitivity and statistical confidence in the results. However, I was only able to count 50 cells using murine ES cells. Differentiated ES cells form embryoid bodies.  The clumped and stacked cells made counting PLA signal in individual cells extremely difficult. Therefore, for the remainder of my work I switched to analysis of murine embryos at two different embryonic ages, E9.5 and E14.5, so I can determine how chromatin assembly changes during development.  Despite the technical difficulties, the ES experiments gave me the confidence to investigate how chromatin assembly on nascent DNA changes during development.    A B   C      Figure 3.2. H3K27me3 associates with PCNA in ES cells of two stages of differentiation. (A) Representative cell 40x images for PLA experiments at 0 and 4 days of differentiation.  PLA signal is in red, PCNA in green, and DAPI in blue. (B) Average H3K27me3 PLA signal in PCNA positive cells at zero and four days of differentiation. Values are mean of PLA signal at zero days of differentiation. in red. (D) Distribution of PLA signal at four days of differentiation. PCNA-H3K27me3 signal in red. 02468Average PLA signal per cell0102030400 5Number of cellsPLA signal0 daysPLA          Merge            D + SEM.  (C) DistributionPCNA-IgG signal in blue, PCNA-H3K27me3 signal PCNA-IgG signal in blue, 0 days 4 daysDays of differentiationIgGH3K27me310IgGH3K27me3010203040500 10Number of cellsPLA signal4 days0 days      4 days 24    20IgGH3K27me325  3.2. Embryo work  The experiments I performed using murine ES cells showed that there are different modes of chromatin assembly in undifferentiated than differentiated cells. E9.5 embryos have completed gastrulation, and organogenesis of the brain, central nervous system, heart, and circulatory system is underway, but development of endodermal derivatives, muscle, and skeleton, and reproductive system is much less advanced.  E14.5 embryos, like 6 month human embryos have organogenesis underway in all organ systems. Because different organ systems develop at different rates,  murine embryos contain mixtures of cells at different stages of differentiation, Nevertheless, older embryos should on average have proportionately more differentiated relative to undifferentiated cells relative to younger embryos, and vice versa. To simplify analysis, I trypsinized whole embryos, and then plated out the cells.  This results in a mixed population of cells, and cell representatives from all lineages are present. But I cannot rule out the possibility that only subsets of the cells attached to the substrates, and thus that my data do not reflect the behaviour of the embryo as a whole.  3.2.1 Histone modifications are present at the replication fork in murine embryo cells  The experiments I performed on murine ES cells indicate that there is a change in how H3K4me3 is assembled at the replication fork during development. To confirm these results in E9.5 cells, I determined the average PCNA-H3K4me3 PLA signals per cell in S phase to be 4.38 + 0.35 (Fig. 3.3B). In older embryos, the average number of PLA signals per cell was 15.97 + 0.8 (Fig. 3.3B). Therefore, the average number of PCNA-H3K4me3 signals increases with embryonic age. The results in ES cells predict that I should see more cells with low numbers of H3K4me3-PCNA PLA in cells derived from E9.5 embryos, so I plotted the number of cells against the number of PLA signals per cell, in groupings of 5 signals per cell (Fig 3.3C). As the average H3K4me3-PCNA signal is very low in E9.5 embryos, I became concerned that I was missing the distribution of signal when grouping signal levels. Thus, I decided to plot the distribution of signal per individual signal level rather than groups of 5 (Fig. 3.3D). As seen in Fig. 3.3D, a bimodal distribution of signal is present in E 9.5 cells relative to E14.5 cells. In E9.5 cells 48 cells have background levels of signal and 52 cells contained signal, whereas in E14.5 cells, essentially all cells have H3K4me3-PCNA PLA signals. I conclude that in young embryos, like undifferentiated ES cells, there are two modes of H3K4me3 assembly on nascent DNA. Furthermore, there is a change in how H3K4me3 assembly occurs as embryos age.      A   B     D      Figure 3.3.  H3K4me3 is located at the replication fork in embryos. (A) Representative E9.5 and E14.5 cell images for PLA experiments using PCNA with H3K4me3. PLA signal is in red, PCNA in green, and DAPI in blue. (B) Average PLA signal per PCNA positive cell. Data are represented as mean cells in groupings. (D) PLA signal in E9.5 cells IgG (blue) and PCNAsignal in E14.5 cells IgG (blue) and PCNA05101520E9.5 E14.5Average PLA signalper cellEmbryonic stage0102030400 10Number of cellsPLA signalE9.5E9.5PLA            Merge      C  E   + SEM. (C) Distribution of PLA si-H3K4me3 (red). (E) PLA -H3K4me3 (red). IgGH3K4me30204060800 10Number of cellsPLA signalE9.5IgGH3K4me30204060800 20Number of cellsPLA signalE14.5 E14.5 26      gnal in E9.5 20IgGH3K4me340IgGH3K4me327   The average number of PCNA-H3K27me3 signals in S phase cells was 8.8 + 0.56 in E9.5 cells (Fig. 3.4B) but increases to 26.9 + 0.56 (Fig 3.4B) in E14.5 cells. Therefore, as murine embryos age there is an increase in the number of PCNA-H3K27me3 signals. As before, I plotted the number of cells against the number of PLA signals in bins of 5, as seen in Fig. 3.4C, and as single cells as seen in Fig. 3.4D. In E9.5 embryo cells there is a bimodal distribution of PCNA-H3K27me3 signal. 35 cells had background levels of signal. In contrast, all E14.5 cells had PLA signals above background levels (Fig. 3.4D).  Thus, H3K27me3 and H3K4me3 assembly into chromatin on nascent DNA are similar in younger versus older embryos.  3.2.2 Modified histones are present on nascent DNA in embryo cells   The PLA with PCNA and H3K4m3 and H3K27me3 experiments measure proximity of the modified histone to the replication fork. To examine events further from the replication fork (and thus that occur later in assembly of nascent chromatin) we carried out PLA with antibodies to DNA labelled with EdU and modified histones.  We have called this assay the chromatin assembly assay (CAA). Nascent DNA can be labeled through the incorporation of the thymidine analogue EdU into newly synthesized DNA. Following a short EdU labeling time of 5 minutes, cells were washed with PBS to remove any residual EdU in the media and then fixed. Biotin molecules were conjugated to EdU through a Click-IT? reaction to allow their detection with anti-biotin antibodies. I chose to label DNA for 5 minutes.  The average rate of progression of DNA polymerase is 1-3 kb/min, so I am examining about 5-15 kb of DNA in the CAA assays in the remainder of this thesis.  The CAA assay can be extended to determine if any protein or protein modification for which an antibody is available associates with nascent DNA. I labeled embryo cells with EdU for 5 minutes and performed CAA experiments with H3K4me3. In E9.5 embryo cells the average EdU-H3K4me3 signal is 7.73 + 0.58 (Fig. 3.5B). In E14.5 embryo cells, the average EdU-H3K4me3 signal increased to 17.35 + 0.99 (Fig 3.5B). Therefore, EdU-H3K4me3 signal on nascent DNA increases during murine embryonic development. To determine if more than one population of cells is detected in CAA experiments, I plotted number of cells against the amount of EdU-H3K4me3 signal per cell. As shown in Fig. 3.5C, most E9.5 cells have signal levels that overlap with the IgG control. There is a small population of cells that have signal levels above the IgG control. In contrast, the majority of E14.5 cells have EdU-H3K4me3 signal levels that are higher than the IgG control levels. Therefore, there is a difference in how H3K4me3 is assembled further from the replication fork in young and older embryos.  28  Using the same logic as for H3K4me3, I also performed CAA experiments with H3K27me3 to determine if H3K27me3 is stable on nascent DNA following passage through the replication fork. There was an average EdU-H3K27me3 signal of 7.78 + 0.72 in E9.5 cells (Fig 3.6B). The amount of signal increased to 18.9 + 1.08 in E14.5 cells (Fig. 3.6B).  As before, a plot of the frequency distribution of the CAA signals in E9.5 cells shows a significant number of unlabelled cells, whereas in E14.5 cells, the majority of EdU-H3K27me3 signal is higher than the background levels (Fig. 3.6D). Therefore, there is a shift towards higher H3K27me3 levels on nascent DNA as development progresses. These data show that differences in chromatin assembly are not confined to chromatin in the immediate vicinity of the replication fork, but extend several kb from the fork.   A      B     D     Figure 3.4.  H3K27me3 is located at the replication fork in embryos. (A) Representative E9.5 and E14.5 cell images for PLA experiments using PCNA with H3K27me3. PLA signal is in red, PCNA in green, and DAPI in blue. (B) Average PLA signal per PCNA positive cell. Data are represented as mean in E9.5 cells. Signal is grouped in bins of 5. PCNA(D) Distribution of H3K27me3 signal in E9.5 cells. (E) Distribution of H3K27me3 signal in E14.5 cells. IgG in blue, H3K27me3 in red. Signals are in group051015202530E9.5 E14.5Avereage PLA signal per cellEmbryonic stage01020300 10 20Number of cellsPLA signalE9.5PLA            Merge      C    E  + SEM. (C) Distribution of H3K27m-IgG in blue, PCNA-H3K27me3 signal in red s of 5.    IgGH3K27me30204060800 10 20Number of cellsPLA signalE9.530IgGH3K27me30204060800 50Number of cellsPLA signalE14.5E14.5 E9.5 29     e3 signal 30IgGH3K27me3IgGH3K27me3A  B   C     Figure 3.5. H3K4me3 is located distal to the replication fork in embryos.(A) Representative E9.5 and E14.5 cell images for CAA experiments with H3K4me3. CAA signal is in red, EdU in green, and DAPI in blue. (B) Average CAA signal per EdU positive cell. Data are represented as mean +signal in blue, EdU-H3K4me3 in red. Signal is in bins of 5. (E14.5 cells. IgG in blue, H3K4me3 in red. Signal is in bins of 5. 05101520Average CAA signalper cell0204060801000 10 20Number of cellsCAA signalE9.5E9.5CAA            Merge       D  SEM. (C) Distribution of H3K4me3 CAA signal D) Distribution of H3K4me3 signal in  E9.5 E14.5Embryonic stageIgGH3K4me330IgGH3K4me30204060801000 10 20Number of cellsCAA signalE14.5 E14.5 30     in E9.5 IgG 30 40IgGH3K4me3A   B  C    Figure 3.6. H3K27me3 is located distal to the replication fork in embryos.(A) Representative E9.5 and E14.5 cell images for CAA experiments with H3K27me3. CAA signal is in red, biotin in green, and DAPI in blue. (B) Average CAA signal per EdU positive cell. Data are represented as mean +signal in blue, EdU-H3K27me3 in red. Signal is in bins of 5. (in E14.5 cells. IgG in blue, H3K27me3 in red. Signal is in bins of 5.0510152025Average CAA signal per cell0204060801000 10 20Number of cellsCAA signalE9.5CAA            Merge      D  SEM. (C) Distribution of H3K27me3 CAA signal D) Distribution of H3K27me3 signal  E9.5 E14.5Embryonic stageIgGH3K27me330 40IgGH3K27me30204060801000 10 20Number of cellsCAA signalE14.5E14.5 E9.5 31     in E9.5 IgG 30 40IgGH3K27me332  3.2.3 Histone methyltransferases at the replication fork in embryo cells  Proteins that create histone modifications are potential epigenetic marks if they are deposited before histones are modified. Thus, I focused on MLL1, a trxG protein with histone methyltransferase activity for H3K4 and Su(z)12 a subunit of the PRC2 complex that has the E(Z)H2 histone methyltransferase activity for H3K27. In Drosophila embryos, trxG and PcG histone methyltrasferases are located at the replication fork, but methylated histones are not (Petruk et al. 2012). The experiments I have performed above show that there are two modes of chromatin assembly in undifferentiated ES cells and E9.5 embryo cells. One population of cells behaves like Drosophila and do not have H3K4me3 or H3K27me3 at the replication fork. I wanted to determine if MLL and Su(z)12  are assembled on nascent DNA, and if this changes during development.  To determine if MLL1 is recruited to nascent DNA near the replication complex, I performed PLA experiments with PCNA and MLL1. The average PCNA-MLL1 signal was 2.80 + 0.22 in E9.5 embryos (Fig. 3.7C). In E14.5 embryo cells, the average number of PLA signals was 9.93 + 0.60 (Fig 3.7C).  I plotted the frequency distribution of PLA signals for individual cells because the average PCNA-MLL1 levels are very low in E9.5 embryo cells. In E9.5 embryo cells, the majority of cells had background levels of signal, and some cells had higher levels of signal (Fig. 3.7E). In E14.5 cells, half of the cells had signal levels that overlap with the IgG control (Fig. 3.7F). The remainder of cells had higher levels of signal. These data indicate that the majority of cells do not have MLL1 located at the DNA replication fork. Furthermore, there is a difference in how MLL1 is assembled at the replication fork in young and old embryos.  This is significantly different than the Drosophila  embryo result, in which the MLL homolog Trx is always present (Petruk et al. 2012), and this does not change in development (Kovermann, unpublished).  However, I needed to eliminate the possibility that MLL1 is recruited to nascent DNA after passage of the replication machinery. To test this, I EdU labeled cells for 5 minutes and performed CAA experiments with MLL1. The average EdU-MLL1 signal in E9.5 is 2.87 + 0.54 signals per cell (Fig 3.7D). The average signal increased to 4.87 + 0.54 in E14.5 cells (Fig 3.7D). As above, I plotted the frequency distribution for individual cells given the low level of signal. The majority of E9.5 cells contained signal levels that were the same as the IgG control,   A              C    E     G    Figure 3.7. MLL1 is located at and distal to the replication fork.(A) Representative E9.5 and E14.5 cell images for PLA experiments between PCNA and MLL1. PLA signal is in red, PCNA in green, and DAPI in blue. (B)images for CAA experiments using EdU and MLL1. CAA signal is in red, EdU in green, and DAPI in blue. (C) Average PLA signal in E9.5 and E14.5 embryo cells. Data is mean Average CAA data in embryo cells. Data icells. (F) Distribution of signal in E14.5 cells. PCNAred. (G) Distribution of CAA signal in E9.5 cells. (H) Distribution of CAA signal in E14.5 cells. EdU-IgG in blue, EdU-MLL1 in red. 01020E9.5 E14.5Average PLA signal per cellEmbryonic stageMLL1 PLA020400 5Number of cellsPLA signalE9.5 0500 5 10Number of cellsCAA signalE9.5PLA    Merge PLA E9.5  E14.5          B                  D   F  H  Representative E9.5 and E14.5 cell s mean + SEM. (E) Distribution of PLA signal in E9.5 -IgG signal in blue, PCNA-MLL1 signal in   IgGMLL1 0510E9.5 E14.5Average CAAsignal per cellEmbryonic stageMLL1 CAA10IgGMLL10501000 10 20Number of cellsPLA signalE14.515IgGMLL10501000 10 20 30Number of cellsCAA signalE14.5CAA       Merge CAA E9.5  E14.5 33     + SEM. (D) IgGMLL130IgGMLL140IgGMLL134  yet there are a small subset of cells that contain EdU-MLL1 signal (Fig. 3.7G). In E14.5 cells, the majority of cells have background levels of MLL1, yet there are some cells with higher levels of MLL1 present, as seen in Fig. 3.7H.  I conclude that the differences between Drosophila  and mouse embryos are not accounted for because of differences in kinetics of chromatin assembly very near versus further from the replication fork. To determine if Su(z)12, a subunit of the H3K27 methyltransferase complex behaves similarly to MLL1, I carried out similar PCNA and CAA assays. There is an average of 3.29 + 0.34 PCNA-Su(z)12 signal in E9.5 embryo cells (Fig. 3.8C). In E14.5 embryo cells, this decreases to 2.91 + 0.24 (Fig. 3.8C). Therefore, there is not a significant difference between PCNA-Su(z)12 signal in young and old embryos, similar to results in Drosophila . A plot of the PCNA-Su(z)12 frequency distribution shows that in E9.5 cells, the majority of PCNA-Su(z)12 signal overlaps with the IgG control, yet there is a small population of cells that contain higher levels of signal (Fig. 3.8E). Therefore, in early embryos there is a bimodal distribution of signal. In older embryos, there is no difference between the control and EdU-Su(z)12 signal (Fig. 3.8F). Therefore, either there is no Su(z)12 present at the replication fork in older embryos, or the antibody does not recognize the Su(z)12 epitope in older embryos. For example, the epitope might be blocked after assembly into mature chromatin. To confirm this result I determined if Su(z)12 is located distal to the replication fork using the CAA assay. The data show that there is no significant signal detected in these experiments. No conclusion can be drawn from negative data in the Su(z)12 CAA assay.     A        C    E     G    Figure 3.8. Su(z)12 is located at and (A) Representative E9.5 and E14.5 cell images for PLA experiments between PCNA and Su(z)12. PLA signal is in red, PCNA in green, and DAPI in blue. (B) Representative E9.5 and E14.5 cell images for CAA experiments using EdU ain blue. (C) Average PLA signal in E9.5 and E14.5 embryo cells. Data is mean CAA data in embryo cells. Data is mean Distribution of signal in E14.5 cells. Blue is PCNACAA signal in E9.5 cells. (H) Distribution of CAA signal in E14.5 cells. Blue is IgG, red is Su(z)12. 024E9.5 E14.5Average PLA signal per cellEmbryonic stageSu(z)12 PLA020400 5 10Number of cellsPLA signalE9.5 Su(z)120501000 2 4Number of cellsCAA signalE9.5 Su(z)12PLA E9.5       E14.5PLA Merge    B                  D  F  H distal to the replication fork. nd Su(z)12. CAA signal is in red, EdU in green, and DAPI + SEM. (D) Average + SEM. (E) Distribution of signal in E9.5 cells. (F) -IgG. Red is PCNA-Su(z)12. (G) Distribution of IgGSu(z)12 012E9.5 E14.5Average CAA signal per cellEmbryonic stageSu(z)12 CAA15IgGSu(z)1201020300 5Number of cellsPLA signalE14.5 Su(z)126IgGSu(z)120501000 5Number of cellsCAA signalE14.5 Su(z)12 CAA E9.5           E14.5 CAA Merge 35     IgGSu(z)1210IgGSu(z)12IgGSu(z)1236  3.2.4 Experiments to determine a mechanism of H3K27me3 assembly   My experiments have shown that there are two modes of chromatin assembly in embryos. The first mode of assembly is the same as Drosophila  embryos and lacks histone modifications at the replication fork. The second resembles S2 cells and has histone modifications at the replication fork. Previous studies have shown that H3K27ac prevents deposition of the H3K27me3 mark (Pasini et al. 2010). This could explain how the HMT is present, but the histone modification is not. My goal was to determine if H3K27ac is preventing methylation of H3K27.  I performed PCNA-H3K27ac PLA in murine embryo cells to determine if H3K27ac was present. The average H3K27ac PCNA-H3K27ac signal in E9.5 cells is 5.4 + 0.56 (Fig. 3.9C), whereas in E14.5 cells, the average PLA signal increased to 10.6 + 0.59 (Fig 3.9C). The plot of the PLA signal frequency distribution showed that the majority of E9.5 cells had background levels of PCNA-H3K27ac signal (Fig. 3.9E). In contrast, the majority of E14.5 cells had signal levels higher than the IgG background (Fig. 3.9F). CAA with H3K27ac in murine embryo cells showed an average signal in E9.5 cells of 4.54 + 0.40 (Fig. 3.9D), whereas E14.5 cells had an average of 18.31 signals per cell (Fig. 3.9D). These results are contrary to the hypothesis that H3K27ac blocks methylation of H3K27, because this model predicts that H3K27ac PLA signals should be high in E9.5, and low in E14.5 embryos, and the opposite was observed. PLA and CAA experiments were performed on a mixed population of embryo cells. To confirm the conclusion that H3K27ac does not block methylation of H3K27, I wanted to determine how H3K27me3 signal compares to H3K27ac signals in embryos of the same age. To do this, I plotted the frequency distributions for H3K27me3 and H3K27ac in E9.5 or E14.5 cells. In E9.5 cells, PCNA-H3K27me3 and PCNA-H3K27ac PLA signal both have the same distribution (Fig. 3.10B). Therefore, in E9.5 embryos, it appears be that H3K27ac does not antagonize H3K27me3.  Interestingly, in E14.5 cells, the distribution of PCNA-H3K27me3 and PCNA-H3K27ac signal shows that there are two types of cells (Fig. 3.10C). The first type of cell have higher levels of H3K27me3 and low levels of H3K27ac, and the other type of cell shows the opposite result. This suggests that in E14.5 embryo cells, there could be antagonism between these marks at the replication fork. To determine if there was any possible antagonism further from the replication fork, I plotted the CAA values in E14.5 embryos, as seen in Fig. 3.10D. The distribution of H3K27me3 and H3K27ac signals is random and fluctuates for all CAA signal  A                C    E    G    Figure 3.9. H3K27ac is located on nascent DNA in embryo cells. (A) E9.5 and E14.5 cell images showing PCNAgreen, DAPI in blue. (B) E9.5 and E14.5 cell images showing EdUin green, DAPI in blue. (C) Average PLA signal in E9.5 and E14.5 cells. Data is represented as mean + SEM. (D). Average CAA signal in embryo cells. Data is mean H3K27ac PLA signal in E9.5 cells H3K27ac. (G) Distribution of CAA signal in E9.5 cells. (H) Distribution of CAA cells. Blue is IgG, red is H3K27ac. 01020E9.5 E14.5Average PLA signalper cellEmbryonic stageH3K27ac PLA0501000 10 20 30Number of cellsPLA signalE9.5 H3K27ac0501000 10Number of cellsCAA signalE9.5 H3K27acPLA  E9.5         E14.5PLA Merge             B                        D           F  H -H3K27ac signal. PLA signal is in red, PCNA in -H3K27ac signal (red). PCNA + SEM. (E) Distribution of (F) Distribution of PLA signal in E14.5 cells. Blue is IgG, red is signal  IgGH3K27ac02040E9.5 E14.5Average CAA signal per cellEmbryonic stageH3K27ac CAA40IgGH3K27ac0501000 10 20Number of cellsPLA signalE14.5 H3K27ac20IgGH3K27ac0501000 10 20 30Number of cellsCAA signalE14.5 H3K27acCAA E9.5      E14.5  CAA    Merge 37    in E14.5 IgGH3K27ac30IgGH3K27ac40IgGH3K27ac38  groupings. Thus, these data alone cannot determine if antagonism is present between these marks.  If there is antagonism, it can occur only very near the replication fork. There is a population of cells with high levels of H3K27ac and low levels of H3K27me3 and another population of cells with the opposite, high levels of H3K27me3 and low levels of H3K27ac, as indicated in E14.5 PLA experiments. Thus, it is possible that these two modifications are antagonistic. Unfortunately, due to the limitations of the assay, it is not possible to determine if an individual cell has high levels of one modification and low levels of another. Treating cells with a H3K27 demethylase or a histone acetyltransferase inhibitor would allow me to determine if global changes in H3K27me3 or H3K27ac affect the opposing K27 modification. Therefore I decided to treat embryo cells with GSK J4, a UTX inhibitor. UTX is a H3K27 demethylase. Treating cells with this inhibitor will allow me to determine if H3K27me3 antagonizes H3K27ac in embryo cells.  I used 25uM GSK J4 inhibitor on E14.5 cells for 1 hour in the incubator. A DMSO control was also performed under the same conditions as the cells. H3K4me3 serves as an internal control because GSK J4 is a specific UTX inhibitor for H3K27 demethylation and therefore will not directly affect H3K4me3. I first determined the average EdU-H3K4me3 signal in cells treated with the inhibitor and those treated with the DMSO control and found that the signal levels were constant (Fig. 3.11D). Next, I wanted to know the efficacy of the inhibitor on UTX activity. If UTX is inhibited, there will be an increase in EdU-H3K27me3 signal in cells treated with GSK J4 compared to those treated with DMSO alone (Fig. 3.11D). On average there were 8.40 + 0.90 EdU-H3K27me3 signals in cells treated with the UTX inhibitor, compared to 5.04 + 0.66 signals in cells treated with DMSO alone (Fig. 3.11D). Therefore, UTX is inhibited under these conditions. However, the H3K27ac was unaffected in cells treated with the UTX inhibitor. As seen in Fig. 3.11D, there is no significant difference in EdU-H3K27ac signal between cells treated with a UTX and the control. Therefore, H3K27me3 does not antagonize H3K27ac on nascent DNA in E14.5 cells, as suggested by the previous experiments. I repeated the UTX inhibitor experiments using PCNA PLA in case the antagonism is only detected near the replication fork. As expected, the inhibitor had no effect on PCNA-H3K4me3 signals because cells treated with the UTX inhibitor averaged of 15.32 + 1.37 (Fig. 3.11C) PLA signals, and cells treated with DMSO alone averaged 18.90 + 0.94 signals (Fig 3.11C).   39  A  B  C  D  Figure 3.10. Comparison of H3K27me3 and H3K27ac distribution in embryo cells.  (A) Comparative distribution of PLA signal with PCNA and H3K27ac (red) or H3K27me3 (blue) in E9.5 cells. (B) Comparative distribution of CAA signal in E9.5 cells between H3K27ac (red) and H3K27me3 (blue). (C) E14.5 PLA distribution of H3K27ac signal in red and H3K27me3 signal in blue. (D) E14.5 CAA distribution of H3K27ac in red and H3K27me3 in blue signal.  0501000 10 20 30 40Number of cellsPLA signalE9.5 PLAH3K27me3H3K27ac0501000 10 20 30 40Number of cellsCAA signalE9.5 CAAH3K27me3H3K27ac020400 20 40 60Number of cellsPLA signalE14.5 PLAH3K27me3H3K27ac01020300 10 20 30 40Number of cellsCAA signalE14.5 CAAH3K27me3H3K27ac A                         C    Figure 3.11. Effect of a H3K27 demethylase inhibitor on E14.5 embryo cells(A) Representative E14.5 cell images for PLA experiments between PCNA and the indicated antibody. The top two panels are treated with GSK J4 and the bottom 2 with DMSO. PLA signal is in red, PCNA in green, DAPI in blue. (B) Representative E14.5 cell imagesexperiments between PCNA and the indicated antibody. The top two panels are treated with GSK J4 and the bottom 2 with DMSO. CAA signal is in red, PCNA in green, DAPI in blue. (C) Average PLA signal in cells treated with DMSO in blue and GSK J4 in (D) Average CAA signal in cells treated with DMSO in blue or GSK J4 in red. Data is mean SEM.  0510152025H3K4me3 H3K27me3 H3K27ac Average PLA signal per cellAntibodyE14.5 GSK J4 25uM PLAPLA H3K4me3     H3K27me3    H3K27acGSK J4 Merge PLA DMSO PLA Merge    B           D  for CAA red. Data is mean DMSOGSK J4 25uM0246810H3K4me3 H3K27me3 H3K27acAverage CAA dots per cellAntibodyE14.5 GSK J4 25uM CAACAAH3K4me3     H3K27me3    H3K27ac Merge CAA Merge CAA 40        + SEM. + DMSOGSK J4 25uM  41   Embryonic stage Treatment Number of EdU positive cells Total number of cells E9.5 GSK J4 25uM 13 6570 GSK J4 25uM control 17 9975 GSK J4 10uM 108 5295 GSK J4 10uM control 107 8145 GSK J4 4uM 110 6075 GSK J4 4uM control 101 8475 control 733 10860 E14.5 GSK J4 25uM 63 8385 GSK J4 25uM control 114 12450 C646 2.5uM 49 7125 C646 2.5uM control 88 7785 C646 1.0uM 80 13965 C646 1.0uM control 80 10650 control 1590 22965     Table 3.1. Amount of EdU positive cells following treatment with inhibitors The number of EdU positive cells out of the total number of cells for E9.5 and E14.5 cells treated with inhibitors.  42  Next, I showed that the average PCNA-H3K27me3 signal for cells treated with DMSO was 12.46 + 0.91 signals (Fig 3.11C). This value increased to 15.32 + 1.37 signals in cells treated with the UTX inhibitor (Fig. 3.11C). Therefore, UTX is being inhibited. Lastly, I determined the effect of UTX on PCNA-H3K27ac signal. In cells treated with DMSO there was an average of 10.54 + 0.66 signals (Fig 3.11C). In cells treated with the inhibitor, there was an average of 7.59 + 0.50 signals (Fig. 3.11C). As PCNA-H3K4me3 signals were also lower in cells treated with the UTX inhibitor, I think the decrease in signal is an artefact of treating cells with the inhibitor. PLA experiments use a PCNA antibody to determine if cells are in S phase, independent of whether or not the cell is alive.  When I did the counting of signal in my CAA inhibitor experiments, I only found 63 EdU positive cells (Table 3.1). Furthermore, when I was doing the PLA signal counting, I noticed that the PCNA counterstaining was very weak. This provides additional support to the idea that cells that were still alive and cells which died due to the presence of the inhibitor stained were counted. Thus, I think that the decrease in PCNA-H3K4me3 and PCNA-H3K27ac signal is a result of the effect of the UTX inhibitor on cell survival.  Experiments using H3K27me3 and H3K27ac in E9.5 embryo cells showed that there were two populations of cells, one containing high signal levels and the second containing background signals. I decided to carry out the UTX inhibitor experiments on E9.5 cells.  E14.5 experiments showed that treating cells 25uM of GSK J4 for 1 hour inhibited UTX activity. Therefore, I treated E9.5 cells with the same conditions.  However, on average only 13 EdU positive cells treated with GSK J4 survived (Table 3.1). Therefore, under these conditions, GSK J4 is toxic to cells. I lowered the concentration of GSK J4 to 10uM and 4uM in attempt to reduce toxicity . An average of 108 EdU positive cells were found when treating cells with 10uM, and 110 when treating with 4uM of GSK J4. However the CAA and PLA signal were variable between treatments, as seen in Fig. 3.12. Therefore, UTX inhibitor experiments on E9.5 cells are not conclusive.      43     Figure 3.12. GSK J4 PLA and CAA average graphs using H3K4me3, H3K27me3, and H3K27ac antibodies on E9.5 embryo cells. PLA and CAA data for E9.5 cells treated with GSK J4, a H3K27 demethylase inhibitor, using H3K4me3, H3K27me3, or H3K27ac antibodies. Treatment conditions are as indicated on the graph. Values are mean + SEM.   0510152025303540H3K4me3 H3K27me3 H3K27ac Average PLA signal per PCNA positive cellAntibodyGSK J4 PLA 25uMDMSOGSK J4 25uM024681012141618H3K4me3 H3K27me3 H3K27ac Average signal per EdU positive cellAntibodyGSK J4 25uM CAADMSOGSK J4 25uM0510152025H3K4me3 H3K27me3 H3K27acAverage PLA signal per PCNA positive cellAntibodyGSK J4 PLA 10uMDMSOGSK J4 10uM0510152025H3K4me3 H3K27me3 H3K27ac Average CAA signal per EdU positive cellAntibodyGSK J4 CAA 10uMDMSOGSK J4 10uM051015202530H3K4me3 H3K27me3 H3K27acAverage PLA signal per PCNA positive cellAntibodyGSK J4 4uM PLADMSOGSK J4 4uM024681012141618H3K4me3 H3K27me3 H3K27ac Average CAA signal per EdU positive cellAntibodyGSK J4 4uM CAADMSO GSK J4 4uM44  Given the toxicity of the UTX inhibitor, I decided to try an alternative approach. CBP is a H3K27 histone acetyltransferase and therefore in the presence of a CBP inhibitor C646, H3K27ac levels will decrease. If H3K27ac is blocking methylation of H3K27, then the CBP inhibitor should increase H3K27 methylation. I treated E14.5 cells with 2.5uM and 1.0uM  inhibitor, and compared them to cells treated with DMSO alone. When I treated cells with the 2.5uM of C646, I noticed many cells instantly detached from the substrate and died. By the end of experiment, 49 EdU positive cells were present (Table 3.1). I determined the EdU-H3K4me3 signal, and noticed there was a decrease in cells treated with the inhibitor (Fig. 3.13). Therefore, under these conditions, C646 is toxic. I also performed experiments using 1.0uM of C646 and under these conditions, an average of 80 EdU positive cells could be found (Table 3.1). When I determined the average EdU-H3K4me3, I found that the levels were significantly higher in cells treated with the inhibitor (Fig. 3.13). Furthermore, when I determined the average EdU-H3K27ac signal in cells, I found that there was no change in H3K27ac signal (Fig. 3.14). Therefore, under these conditions C646 is not only toxic to cells, but does not inhibit CBP activity. Due to the fact that the CAA experiments show that C646 is toxic to cells under these conditions, the PLA results are not interpretable. Due to the effect of C646 on E14.5 cells, I decided not to perform experiments using this inhibitor on E9.5 cells.               45             Figure 3.13. C646 PLA and CAA average graphs using H3K4me3, H3K27me3, and H3K27ac antibodies on E14.5 embryo cells.  PLA and CAA data for E14.5 cells treated with C646, a HAT inhibitor, using H3K4me3, H3K27me3, or H3K27ac antibodies. Treatment conditions are as indicated on the graph. Values are mean + SEM.  05101520253035H3K4me3 H3K27me3 H3K27ac Average dots per PCNA positive cellAntibodyC646 2.5uM PLADMSOC646 2.5uM0246810121416H3K4me3 H3K27me3 H3K27acAverage CAA dots per EdU positive cellAntibodyE14.5 C646 2.5uM CAA DMSOC646 2.5uM051015202530H3K4me3 H3K27me3 H3K27ac Average PLA dots per PCNA positive cellAntibodyE14.5 C646 1uM PLADMSOC646 1uM02468101214161820H3K4me3 H3K27me3 H3K27acAverage CAA signal per EdU positive cellAntibodyE14.5 C646 1uM CAADMSOC646 1uM46  4. Discussion  Most of what we know about epigenetic marks is at the level of transcription or DNA repair. Relatively little is known about how epigenetic marks are deposited on nascent DNA. In this study, I used two experimental techniques to determine how chromatin is assembled during S phase in murine embryo cells. PLA experiments allow me to determine if a specific histone modification or protein is associated with PCNA at the replication fork. CAA experiments allow me to determine if a particular molecule is detected on nascent DNA up to 15kb from the replication fork.   During differentiation, chromatin changes structure as chromatin becomes reorganised. The chromatin in pluripotent cells is open and the level of chromatin compaction is low (Ho and Crabtree, 2010; Ahmed et al. 2010). The first lineage commitments occur following the eight cell stage in murine embryos, and compact chromatin regions can be observed. At E5.5, highly compact chromatin regions can be detected throughout the nucleus. In differentiated cells, chromatin is organised into euchromatic and heterochromatic regions.  Thus, the changes I observed between E9.5 and E14.5 cells could be indicative of the changes in chromatin structure during differentiation of cells down a particular cell lineage.  Furthermore, the distribution of signal within cells at a given embryonic stage is representative of the cells from varying cell lineages in trypsinized whole embryos. Interestingly, I found there are two populations of cells in E9.5 murine embryos. The first type of cell lacks histone methylation on nascent DNA and is characteristic of the results seen in Drosophila  embryos.  Methylated histones are detected in the second group of E9.5 cells. In E14.5 cells, a higher level of histone methylation was detected on nascent DNA.  Therefore, I propose that the second group of E9.5 cells is characteristic of a differentiated population of cells. Gastrulation, the formation of three germ layers, begins at E6.75 and organogenesis begins at E8.0 (Downs and Davies, 1993). Myocardial precursors begin the formation of the heart at E6.5, and a tube shape heart is formed by E8.0 (Evans et al., 2010). In E9.5 cells, many tissues are beginning to form, and therefore many cell precursors are present. The liver haematopoetic system begins formation at E9.5 as well as the endothelial lining required for blood vessel formation (Evans et al. 2010). Forelimbs begin developing at E9.5 and by E14.5 a fully formed limb is visible, yet digits are not fully separated. Although not fully functional, most central nervous system neurons have migrated towards their final location at E14.5, including 47  neurons innervating the brain and spinal cord (Diez-Roux et al., 2011). The heart begins to contract shortly thereafter and by E14.5, a four chambered heart is present (Evans et al., 2010). Thus, the increase in H3K4me3 and H3K27me3 abundance I observed correlates with the gross changes in chromatin structure during development. Furthermore, as cell fate becomes more restricted, there is an increase in the net amount of genes that are active or repressed which could explain the increase in average number of PLA signals per cell. For nearly 30 years it has been known that parental histones are present immediately preceding the replication fork, and are found on nascent DNA within a few hundred base pairs of the replication fork (Gasser et al. 1996). Research in vitro has clarified the role of histone chaperones in deposition of newly synthesized and parental histones on nascent DNA.  It has been widely assumed that parental histones with PTM are deposited on nascent DNA, and that these modified histones recruit HMT, which methylate newly deposited histones (Jackson and Chalkley, 1985; Corpet and Almounzie, 2009; Jasencakova and Groth, 2010).  Nevertheless, previous studies have lacked the resolution to determine in vivo if histones near the replication fork are methylated, or to determine if the nucleosomes on nascent chromatin near the replication fork are parental, or newly deposited.  The observations of Petruk et al. (2012) show that unmodified H3 is detected in association with PCNA, but H3K4me3 and H3K27me3 are not.  H3K4me3 and H3K27me3 were not associated with EdU-labeled DNA in Drosophila  embryos until between 30-60 min after replication.  These authors argued that the model proposed in the previous paragraph is incorrect. Instead, they suggest that unmodified histones are deposited immediately on nascent DNA, and that trimethylation is delayed in embryos. These results of (Petruk et al. 2012) in Drosophila  embryos do not rule out the possibility that parental histones are transferred to nascent DNA, but are demthylated prior to, during, or immediately after transfer. Reviewers of Petruk et al. (2012) suggested another possibility, namely that parental histones are transferred to nascent DNA with mono- or dimethylated H3K4 or H3K27, and that the trimethyl PTM is acquired after transfer. We have carried out similar PLA and CAA experiments in Drosophila  embryos using antibodies to H3K4me1 and H3K4me2, and shown that these modifications are detected with the same kinetics as H3K4me3 (Petruk et al. 2013, submitted). The levels of histone modifications change during the cell cycle (Bonenfant et al. 2007; Lanzuolo et al. 2011; Lanzuolo et al. 2012; Petruk et al. 2012). In Drosophila  embryos, unmodified H3 is detected on nascent DNA and histone methylation can be detected in the next G phase rather than S phase (Petruk et al. 2012). In S2 cells, H3K27me3 levels are higher in early S phase than late S phase, 48  and G1/S phase (Lanzuolo et al. 2011). At this time it is not known whether these cell cycle stage-specific changes in modification reflect transcriptional status, nuclear architecture, or other features of chromatin regulation. Histone methylation is a dynamic (Agger et al. 2007; Shi 2007; Cloos et al. 2008). Interestingly, methylases and demethylases are found in the same protein complexes. The H3K4 methyltransferase MLL1 is found in the same complex as the H3K27 demethylase UTX (Cloos et al. 2008; Pasini et al. 2008). This allows for the simultaneous removal of H3K27 trimethylation followed by trimethylation of H3K4. Histone acetyltransferases acetylate H3K27, thereby blocking the site from methylation. On the other hand, the H3K4 demethylase RBP2 is associated with the PRC2 complex and is involved in transcriptional repression (Pasini et al. 2008). These observations suggest that one complex can contain the enzymes necessary to switch from actively transcribed to repressed, and vice versa.  Before the discovery of histone demethylases, histone methylation was thought to be very stable (Byvoet et al. 1972; Annunziato et al. 1995) and to contrast with histone acetylation, which was labile (Wade et al. 1997). The existence of histone acetyl transferases (HAT) and histone deacetylases (HDAC) provide a mechanistic way to rapidly add or remove acetyl groups (Wade et al. 1997; Kouzarides, 2000; Yu et al. 2011). However, with the discovery of histone demethylases (Shi et al. 2004) it has become apparent that histone methylation is also labile. If the histone demethylase activity is higher in undifferentiated than differentiated cells, this could explain the difference in the presence H3K4 or H3K27 methylation in nascent chromatin in undifferentiated compared to differentiated cells. In this thesis, I show that in E9.5 embryos, there is a population of cells that behaves like Drosophila  embryo cells, in which H3K4me3 and H3K27me3 are not detected using PLA with PCNA or CAA. My observations disprove the model that modifications on parental histones deposited on nascent DNA are stable to replication. However they do not distinguish between the other two models proposed above, namely that unmodified H3 is deposited on nascent chromatin, or that parental trimethylated histones are demethylated prior to, during, or immediately following transfer to nascent DNA.   The most important finding of my thesis is that H3K4me3 and H3K27me3 are detected using PLA and CAA assays on nascent DNA in some cells of E9.5 embryos, and in most cells of E14.5 embryos.  One potential explanation is that chromatin assembly on nascent DNA differs in undifferentiated versus differentiated cells.  It could be that in Drosophila  embryos, and in 49  undifferentiated murine cells, parental trimethylated H3K4 and H3K27me3 are transferred to nascent DNA in differentiated cells whereas, but not undifferentiated cells.  This might come about by differential regulation of the deposition of newly synthesized unmodified in undifferentiated and differentiated cells. However, I think it is more likely that the process of chromatin assembly in undifferentiated and differentiated cells is the same, and that the key difference is regulation of histone demethylase activity in these cells.  This suggestion is consistent with my observations that in some cells from E9.5 embryos the H3K4 HMT (MLL1), and a subunit of the H3K27me3 HMT (Su)z)12 were present at low levels at the replication fork, but the corresponding H3K4me3 and H3K27me3 modifications were not.  These observations strongly suggest that the activity of the HMT is down-regulated, or that the activity of the demethylase is elevated, in undifferentiated cells. Several previous studies have shown that HMTs required for H3K4 or H3K27 trimethlylation are stable to DNA replication. In mammalian cell lines, the H3K27 HMT EZH2 co-localizes with both PCNA and BrU (Hansen et al. 2008).  Using the SV40 in vitro replicating system, the Psc and Pc subunits of the PRC1 PcG complex were detected on nascent DNA (Francis et al. 2009). Furthermore, Francis et al. (2009) propose that PcG complexes are rapidly bound to nascent DNA immediately following progression through the replication fork. In Drosophila  embryos the H3K4 HMT  Trx and the H3K27 HMT E(z) are stable to DNA replication (Petruk et al. 2012).  Thus an important question for the future is to determine how the activity of H3K4 and H3K27 HMT and their corresponding demethylases are regulated immediately following DNA replication. Cross-talk occurs between histone modifications. For example, H3K27 is modified by both acetylation and methylation and it has been proposed that these histone modifications antagonize each other (Pasini et al. 2010).  CREB binding protein, CBP, is a histone acetyltransferase that targets a variety of proteins including histones and transcription factors. Histone acetylation is associated with active gene states, and is thought that CBP functions to stabilize the transcription complex (Martinez-Balbas et al., 1998; Partanen et al., 1999; Zenter et al., 2011; Kalkhoven, 2004). H3K27ac is located at the enhancers of active genes, around the transcription start site of active genes (Zenter et al., 2011). In early Drosophila  embryos, H3K27ac was high in early stages of development and H3K27me3 was low at this time (Tie et al., 2009). As development progressed and PcG silencing started, H3K27ac levels decreased and H3K27me3 levels increased. E(z) RNAi knockdown in S2 cells caused a decrease in H3K27me3 levels, and a ~3 fold increase in H3K27ac levels. CBP knockdown experiments 50  showed the opposite result, namely H3K27ac levels decreased and H3K27me3 levels increased. Combined, these data suggest that H3K27ac and H3K27me3 are mutually exclusive. Another study using ES cells also showed that H3K27me3 and H3K27ac are antagonistic marks at the global level (Pasini et al., 2010). Knockout ES cells for PRC2 members Su(z)12 and Ezh2 resulted in a lack of H3K27me3 and a significant increase in H3K27ac using western blot experiments. I performed experiments to determine if the possible antagonism of these two marks could account for the changes in histone methylation between E9.5 and E14.5 embryos on nascent DNA. When I treated E14.5 cells with a UTX inhibitor, I concluded H3K27me3 does not antagonize H3K27ac at this stage of murine development. Instead their presence is independent, unlike results from previous studies.  I propose the following model of chromatin assembly in murine embryo cells, as seen in Fig. 4.1. In undifferentiated cells, parental histones are transferred to nascent DNA and H3K4me3 or H3K27me3 are removed by demethylases. One or both of these sites may be subsequently acetylated to prevent histone methylation. When a particular nucleosome requires histone methylation, histone deacetyltransferases will remove the acetylation mark and HMTs will methylate it at a later stage in the cell cycle. In terminally differentiated cells, when cell fates and gene expression patterns are established, demethylation of parental histones would not occur. I tested the role of a demethylase inhibitor in E14.5 cells. In these cells, my results indicate that demethylation of parental histones is developmentally regulated. Previous studies have indicated that demethylases are important in development (Terndrup et al. 2010; Oh and Janknecht 2012; Lan et al. 2007; Agger et al. 2007). Demethylase knockouts are lethal and show developmental defects (Lan et al. 2007; Agger et al. 2007; Terndrup et al. 2010; Oh and Janknecht 2012). The following experiments would test this model. PLA and CAA experiments using antibodies to histone demethylases, histone acetyltransferases, and histone deacteyltransferases will show higher levels in younger embryo cells. Inhibition of demethylase activity in younger embryo cells will result in chromatin assembly patterns seen in older embryos.  If demethylases are upregulated in older embryos, they will be driven to a less differentiated state and less histone methylation will be detected in PLA and CAA experiments. Furthermore, I would predict that tissues isolated from very early embryos will have no histone modifications at the replication fork and the amount will increase as development progresses. Adult tissues would have the highest levels of the modifications.              Undifferentiated cells  Figure 4.1. Schematic of the proposed models of chromatin assembly in undifferentiated and terminally differentiated cells.              Differentiated cells 51  52  References  Abraham, B., Cui, K., Tang, Q., Zhao, K. (2013) Dynamic regulation of epigenomic landscapes during hematopoiesis. BMC Gemonics 14:193-208 Agger. K., Cloos, P., Christensen, J., Pasini, D., Rose, S., Rappsilber, J., Issaeva, I., Canaani, E., Salcini, A., Helin, K. (2007) UTX and JMJD3 are histone H3K27 demethylases involved in HOX gene regulation and development. Nature 449: 731-734 Ahmed, K,. Dehghani, H., Rugg-Gunn, P., Fussner, E., Rossant, J., Bazett-Jones, D. (2010) Global chromatin architecture reflects pluripotenct and lineage commitment in the early mouse embryo PLoS ONE 5:1-13 Bannister, A., Kouzarides, T. (2011) Regulation of chromatin by histone modifications. Cell Research 21: 381-395 Barski, A., Cuddapah, S., Cui, K., Roh, T., Schones, D., Wang, Z., Wei, G., Chepelev, I., Zhao, K. (2007) High-resolution profiling of histone methylations in the human genome. Cell 129: 823-837 Bavykin, S., Srebreva, L., Banchev, T., Tsanev, R., Zlatanova, J., Mirzabekov, A. (1993) Histone H1 deposition and histone-DNA interactions in replicating chromatin. Proc. Natl, Acada, Sci. 90: 3918-3922 Benkovic, S., Valentine, A., Salinas, F. (2001) Replisome-Mediated DNA replication Annual Reviews in Biochemistry 70: 181-208 Bernstein, B., Mikkelsen, T., Xie, X., Kamal, M., Heubert, D., Cuff, J., Fry, B., Meissner, A., Wernig, M., Plath, K., Jaenisch, R., Wagschal, A., Feil, R. Schreiber, S., Lander, E. (2006) A bivalent chromatin structure marks key developmental genes in embryonic stem cells Cell 125:315-326 Bird, A. (2002). DNA methylation patterns and epigenetic memory. Genes and Development 16: 6-21 Birve, A., Sengupta, A., Beuchle, D., Larsson, J., Kennison, J., Rasmuson-Lestander, A., Muller, J. (2001) Su(z)12, a novel Drosophila  Polycomb group gene that is conserved in vertebrates and plants Development 128:3371-3379 53  Bogan, J., Natale, D., Depamphilis, M. (2000) Initiation of Eukaryotic DNA replication: conservative or liberal? Journal of Cellular Physiology 184: 139-150 Boheler, K., Czyz, J., Tweedie, D., Yang, H., Anisimov, S., Wobus, A. (2002) Differentiation of pluripotent embryonic stem cells into cardiomyocytes. Circulation Research 91:189-201 Bonenfant, D., Towbin, H., Coulot, M., Schindler, P., Mueller, D., van Oostrum, J. (2007) Analysis of dynamic changes in post translation modifications of human histones during cell cycle by mass spectrometry. Molecular & Cellular Proteomics 6: 1917-1932. Bravo, R., Celis, J. (1980) A search for differential polypeptide synthesis throughout the cell cycle of HeLa cells. Journal of Cellular Biology 84: 795-802 Byvoet, P., Shepherd, G., Hardin, J., Noland, B. (1972) The distribution and turnover of labelled methyl groups in histone fractions of cultured mammalian cells Arch. Biochem. Biophys. 148: 558-567 Cao, R., Wang, L., Wang, H., Xia, L., Erdjument-Bromage, H., Tempst, P., Jones, R., Zhang, Y. (2002) Role of histone H3 lysine 27 methylation in Polycomb-Group silencing. Science 298:1039 ? 1043 Chase, A., Cross, N. (2011) Aberrations of EZH2 in Cancer. Clinical Cancer Research 17: 2613-2618 Chen, C., Carson, J., Feser, J., Tamburini, B., Zabaronick, S., Linger, J., Tyler, J. (2008). Acetylated lysine 56 histone H3 drives chromatin assembly after repair and signals for the completion of repair. Cell 134: 231-243 Choi, S. (2012) DNA replication. Retrieved May 29, 2012, from http://sharonchoi--bi4u.blogspot.ca/2010/09/dna-replication.html  Cloos, P., Christensen, J., Agger, K., Helin, K. (2008) Erasing the methyl mark: histone demethylases at the center of cellular differentiation and disease Genes and Development 22:1115-1140 Corpet, A., Almouzni, G. (2009) Making copies of chromatin: the challenge of nucleosomal organization and epigenetic information. Cell19: 29-41 Diez-Roux, G., Banfi, S., Sulta, M., Geffers, L, Anand, S., Rozado, D., Magen, A., Canidio, E., 54  Pagani, M., Peluso, I., Lin-Marq, N., Koch, M., Bilio, M., Caniello, I., Vrde, R., De Masi, C., Bianchi, S., Cicchini, J., Perroud, E., Mehmeti, S., Dagand, E., Schrinner, S., Nurnberger, A., Schmidt, K., Metz, K., Zwingmann, C., Brieske, N., Springer, C., Martinez Hernandez, A., Herzog, S., Grabbe, F., Sieverding, C., Fisher, B., Schrader, K., Brockmeyer, M., Dettmer, S., Helbig, C., Alunni, V., Battaini, M,. Mura, C., Henrichsen, C., Garcia-Lopez, R., Echevarria, D., Puelles, E., Garcia-Calero, E., Kruse, S., Uhr, M., Kauck, C., Feng, G., Milyaev, N., Ong, C., Kumar, L., Lam, M., Semple, C., Gyenesei, A., Mundlos, S., Radelof, U., Lehrach, H., Sarmientos, P., Reymond, A., Davisdon, D., Dolle, P., Antonarakis, S., Yaspo, M., Martinez, S., Baldock, R., Eichele, G., Ballabio, A. (2011) A high-resolution anatomical atlas of the transcriptome in the mouse embryo. PLoS Biology 9:1-13   Downs, K., Davies, T. (1993) Staging of gastrulating mouse embryos by morophological landmarks in the dissecting microscope. Development 118:1255-1266 Eissenberg, J. and Shilarifard, A. (2010) Histone H3 lysine 4 (H3K4) methylation in development and differentiation. Developmental Biology 339:240-249 Evans, S., Yelon, D., Conlon, F., Kirby, M. (2010) Myocardial lineage development Circulatory Research 107: 1428-1444 Fazly, A., Li, Q., Hu, Q., Mer, G., Horazdovsky, B., Zhang, Z. (2012) Histone Chaperone Rtt106 Promotes Nucleosome Formation Using (H3-H4)2 Tetramers. Journal of Biological Chemistry 287: 10753-10760 Francis, N., Follmer, N., Simon, M., Aghia, G., Butler, J. (2009) Polycomb proteins remain bound to chromatin and DNA during DNA replication in vitro Cell 137: 110-122 Gasser, R., Koller, T., Sogo, J. (1996) The stability of nucleosomes at the replication fork. Journal of Molecular Biology 258: 224-239 Glaser, S., Lubitz, S., Loveland, K., Ohbo, K., Robb, L., Schwenk, F., Seibler, J., Roellig, D., Kranz, A., Anastassiadis, K., Stweart, F. (2009) The histone 3 lysine 4 methyltransferase, Mll2, is only required briefly in development and spermatogenesis. Epigenetics and Chromatin 2:5 Gogol, E., Young, M., Kubasek, W., Jarvis, T., von Hippel, P. (1992) Cryoelectron microscopic visualization of functional subassemblies of the bacteriophage T4 DNA replication complex. Journal of Molecular Biology 224: , Issue 2, 20 March 1992, Pages 395?412 55  Golob, J., Paige, S., Muskheli, V., Pabon, L., Murry, C. (2008) Chromatin remodeling during mouse and human embryonic stem cell differentiation. Developmental Dynamics 237:1389-1398 Guenther, M., Jenner, R., Chevalier, B., Nakamura, T., Croce, C., Canaani, E., Young, R. (2005) Global and Hox-specific roles for the MLL1 methyltransferase. PNAS 102: 8603-8608 Guenther, M., Levine, S., Boyer, L., Jaenisch, R., Young, R. (2007) A chromatin landmark and transcriptional initiation at most promoters in human cells. Cell130:77-88 Gunjan, A., Paik, J., Verreault, A. (2005) Regulation of histone synthesis and nuclesome assembly. Biochemie 87: 625-635 Han, J., Zhou, H., Horazdovsky, B., Zhang, K., Xu, R., Zhang, Z. (2007) Rtt109 acetylates histone H3 lysine 56 and functions in DNA replication. Science 315: 653-655 Hansen, K., Bracken, A., Pasini, D., Dietrich, N., Gehani, S., Monrad, A., Rappsilber, J., Lerdrup, M., Helin, K. (2008) A model for transmission of the H3K27me3 epigenetic mark Nature Cell Biology10: 1291-1300 Henikoff, S., Shilatifard, A. (2011) Histone modification: cause or cog? Trends in Genetics 10: 389-396  Herlofsen, S., Bryne, J., Hoiby, T., Wang, L., Issner, R., Zhang, X., Coyne, M., Boyle, P., Gu, H., Mezza-Zepeda, L., Collas, P., Mikkelsen, T., Brinchmann, J. (2013). Genome-wide map of quantified epigenetic changes during in vitro chongrogenic differentiation of primary human mesenchymal stem cells. BMC Gemonics 14: 105 Heuser, M., Yap, D., Ruiz de Algara, T., Tafech, A., McKinney, S., Dixon, J., Thresher, R,. Colledge, B., Carlton, M., Humphries, K., Aparicio, S. (2009) Loss of Mll5 results in pleiotropic hematopoietic defects, reduced neutrophil immune function, and extreme sensitivity to DNA demethylation. Blood113: 1432-1443 Ho, L., Crabtree, G. (2010) Chromatin remodelling during development. Nature 463:474-484 Hoek, M., Stillman, B. (2003) Chromatin assembly factor 1 is essential and couples chromatin assembly to DNA replication in vivo. PNAS 100:12183-12188 56  Hsu, T. (1962) Differential rate in RNA synthesis between euchroamtin and heterochromatin. Experimental Cell Research 27: 332-334 Huisinga K., Brower-Toland B., Elgin S. (2006) The contradictory definitions of heterochromatin: Transcription and silencing. Chromosoma 115: 110?122. Ingham, P. (1998) trithorax and the regualtion of homeotic gene expression in Drosophila : a historical perspective. Int. J. Dev. Biol 42:423-429 Jackson, V.m Chalkley, R. (1985) Histone segregation on replicating chromatin. Biochemistry 24: 6930-6938 Jasencakova, Z., Groth, A. (2010) Restoring chromatin after replication: How new and old histone marks come together. Seminars in Cell & Developmental Biology 21: 231-237 Jones, P., Takai, D. (2001) Role of DNA Methylation in Mammalian Epigenetics. Science 10: 1068-1070 Kahn, T., Schwartz, Y., Dellino, G., Pirrotta, V. (2006). Polycomb complexes and the propagation of the methylation mark at the Drosophila  Ubx gene. J of Bio Chem 281: 29064 - 29075 Kalkhoven, E. (2004) CBP and p300: HATs for different occasions Biochemical Pharmacology 68:1145-1155 Kelman, Z. (1997) PCNA: structure, functions  and interactions. Oncogene 14: 629-640 Kim, H., Jang, M., Kang, M., Han, Y. (2011) Epigenetic signatures and temporal expression of lineage-specific genes in hESCs during differentiation of hepatocytes in vitro. Human Molecular Genetics 20: 401-412 Kim, Y., and Kim, A. (2011) Characterization of histone H3K27 modifications in the ?-globin locus Biochemical and Biophysical Research Communications 405:210-215 Koch, C., Andrews, R., Flicek, P., Dillon, S., Karaoz, U., Clelland, G., Wilcox, S., Beare, D., Fowler, J., Couttet, P., James, K., Lefebvre, G., Bruce, A., Dovey, O., Ellis, P., Dhami, P., Langford, C., Weng, Z., Birney, E., Carter, N., Vetrie, D., Dunham. (2007) The landscape of histone modifications across 1% of the human genome in five human cell lines. Genome Research 6:691-707 57  Kouzarides, T. (2000). Acetylation: a regulatory modification to rival phosphorylation? EMBO Journal 19: 1176-1179 Krude, T., Keller, C. (2001) Chromatin assembly during S phase: contributions from histone deposition, DNA replication and the cell division cycle. CMLS 58: 665-672 Ladoux, B., Quivy, J., Doyle, P., du Roure, O., Almouzni, G., Viovy, J. (2000) Fast kinetics of chromatin assembly revealed by single-molecule videomicroscopy and scanning force microscopy. PNAS 97: 14251-14256 Lande-Diner, L., Zhang, J., Ben-Porath, I., Amariglio, N., Keshet, I., Hechet, M., Azuara, V., Fisher, A., Rechavi, G., Cedar, H. (2007) Role of DNA methyation in stable gene repression. The Journal of Biological Chemistry 282: 12194-12200 Lanzuolo, C., Lo Sardo, F., Orlando, V. (2012) Concerted epigenetic signatures inheritance at PcG targets through replication. Cell Cycle 11: 1296-1300 Lanzuolo, C., Sardo, F., Diamantini, A., Orlando, V. (2011). PcG complexes set the stage for epigenetic inheritance of gene silencing in early S phase before replication. PLoS Genetics 7:1-11 Lee, J., Hart, S., Skalnik, D. (2003) Histone deacetylase activity is required for embryonic stem cell differentiation. Genetics 38: 32-38 Lennartsson, A., Ekwall, K. (2009) Histone modification patterns and epigenetic codes Biochemica et Biophysica Acta 1790: 863-868 Li X., Li,J., Harrington,J., Lieber,M.R. and Burgers,P.M. (1995) Lagging strand DNA synthesis at the eukaryotic replication fork involves binding and stimulation of FEN-1 by proliferating cell nuclear antigen. Journal of Bioogical Chemistry 270: 22109?22112. Lorincz, M., Dickerson, D., Schmitt, M., Groudine, M. (2004) Intragenic DNA methylation alters chromatin structure and elongation in mammalian cells. Nature Structural & Molecular Biology 11: 1068-1075. Lucchini, R., Wellinger, R., Sogo, J. (2001) Nucleosome positioning at the replication fork. EMBO 20:7294-7302 58  Madan, V., Mada, B., Brykczynska, U., Zibermann, F., Hogeveen, K., Dohner, K., Dohner, H., Weber, O., Blum, C., Rodwald, H., Sassone-Corsi, P., Peters, A., Fehling, H. (2009) Impaired function of primitive hematopoietic cells in mice lacking Mixed-Lineage-Leukemia homolog Mll5. Blood 113:1444-1454 Martinez-Balbas, M., Bannister, A., Martin, K,. Haus-Seuffert, P., Meisterernst, M., Kouzarides, T. (1998) The acetyltransferase activity of CBP stimulates transcription The EMBO Journal 17:2886-2893 Masumoto, H., Hawke, K., Verreault, A. (2005) A role for cell-cycle-regulated histone H3 lysine 56 acetylation in the DNA damage response. Nature 436: 294-298 McKnight, S., Miller, O. (1977) Electron microscopic analysis of chromatin replication in the cellular blastoderm Drosophila  melanogaster embryo. Cell 12: 795-804 Meshorer, E., Misteli, T., (2006) Chromatin in pluripotent embryonic stem cells and differentiation. Nature Reviews Molecular Cell Biology 7: 540-546 Meshorer, E., Yellajoshula, D., George, E., Scambler, P., Brown, D., Misteli, T. (2006) Hyperdynamic plasticity of chromatin proteins in pluriporent embryonic stem cells Developmental Cell 10: 105-116 Miyachi, K.,  Fritzler, M,. Tan. E. (1978) Autoantibody to a nuclear antigen in proliferating cells. Journal of Immunology 121: 2228-2234  Moldovan, G., Pfander, B., Jentsch, S. (2007) PCNA, the maestro of the replication fork. Cell 129:665-679 Morgan, M., Shilatifard, A. (2013) Drosophila  SETs its sights on cancer: Trr/MLL3/4 COMPASS-Like complexes in development and disease. Molecular Cell Biology 33: 1698-1701 Oh, S., and Janknecht, R. (2012). Histone demethylase JMJD5 is essential for embryonic development. Biochemical and Biophysical Research Communications 420: 61-65 Okada, S. (1968) Replicating Units (Replicons) of DNA in Cultured Mammalian Cells. Biophysical Journal 8: 650-664 59  Okitsu, C., Hsieh, J., Hsieh, C. (2010) Transcriptional activity affects the H3K4me3 level and distribution in the coding region. Molecular and Cellular Biology 30: 2933-2946   Olink Biosciences (2012) Duolink Application Note [Brochure]. Sweden. Olink Biosciences  Pan, G., Tian, S., Nie, J., Yang, C., Ruotti, V., Wei, H., Jonsdottir, G., Stewart, R., Thomson, J. (2007) Whole-genome analysis of histone H3 lysine 4 and lysine 27 methylation in human embryonic stem cells. Cell Stem Cell 1: 299-312 Papp, B., Muller, J. (2006) Histone trimethylation and the maintenance of transcriptional ON and OFF states by trxG and PcG proteins. Genes and Dev 20:2041 ? 2054 Partanen, M., Motoyama, J., Hui, C. (1999) Developmentally regulated expression of the transcriptional cofactors/histone acetyltransferases CBP and p300 during mouse embryogenesis Int. J. Dev. Biol. 43:487-494 Pasini, D., Bracken, A., Hansen, J,. Capillo, M., Helin, K. (2007) The Polycomb Group Protein Suz12 is required for embryonic stem cell differentiation. Molecular and Cellular Biology 27:3769 ? 3779 Pasini, D., Malatesta, M., Jung, H., Walfridsson, J., Willer, A., Olsson, L., Skotte, J., Porse, B., Jensen, O., Helin, K. (2010) Characterization of an antagonistic switch between histone H3 lysine 27 methylation and acetylation in the transcriptional regulation of Polycomb group target genes Nucleic Acids Research 38:4958-4969 Petruk, S., Sedkov, Y., Johnston, D., Hodgson, J., Black, K., Kovermann, S., Beck, S., Canaani, E., Brock, H., Mazo, A. (2012) TrxG and PcG proteins but not methylated histones remain associated with DNA through replication Cell 150:922-933  Petruk, S., Sedkov, Y., Riley, K., Hodgson, J., Schweisguth, F., Hirose, S., Jaynes, J., Brock, H., Mazo, A. (2006). Transcriptional elongation of non-coding bxd RNAs promoted by the Trithorax TAC1 complex represses Ubx by a transcriptional interference mechanism. Cell: 125: 1209 ? 1221 Quina, A., Bushbeck, M., Croce, L. (2006) Chromatin structure and epigenetics. Biochemical Pharmacology 72: 1563-1569 Rugg-Gunn, P., Cox, B., Ralston, A., Rossant, J. (2010). Distinct histone modifications in stem cell lines and tissue lineages from the early mouse embryo. PNAS 107: 10783-10790 60  Scharf, A., Barth, T., Imhof, A. (2009) Establishment of histone modifications after chromatin assembly. Nucleic Acids Research 37: 5032-5040 Schindler, U., Beckmann, H., Cashmore, A. (1993) HAT3.1, a novel Arabidopsis homeodomain protein containing a conserved cysteine-rich region. The Plant Journal 4: 137-150 Schneider, E. (1972). Cell lines derived from late embryonic stages of Drosophila  melanogaster J. Embryol. exp. Morph 27:353-365 Schneider, R., Bannister, A., Myers, F., Thorne, A., Crane-Robinson, C., Kouzarides, T. (2003) Histone H3 lysine 4 methylation patterns in higher eukaryotic genes. Nature Cell Biology 6:73-77 Shi, Y. (2007) Histone lysine demethylases: emerging roles in development, physiology, and disease. Nature Reviews in Genetics 8: 829-833 Simon, J. (1995) Locking in stable states of gene expression: transcriptional control during Drosophila  development. Current Opinion in Cell Biology7: 376-385 Strahl, B., Allis, C. (2000). The language of covalent histone modifications. Nature 403: 41-45 Suganuma, Y., Workman, J. (2008). Crosstalk among histone modifiactions. Cell 135: 604-607 Terndrup Pedersen, M. and Helin, K. (2010) Histone demethylases in development and disease. Cell 20:662-671 Tie, F., Banerjee, R., Stratton, C., Prasad-Sinha, J., Stepanik, V., Zlobin, A., Diaz, M., Scacheri, P., Harte, P. (2009) CBP-mediated acetylation of histone H3 lysine 27 antagonizes Drosophila  Polycomb silencing Development 136:3131-3141 Wade, P., Pruss, D., Wolfe, A. (1997) Histone acetylation: chromatin in action. Cell 22: 128-132 Waga, S., Stillman, B. (1998). The DNA replication fork in eukaryotic cells. Annual Review in Biochemistry 67: 721-751 Wang, P., Lin, C., Smith, E., Guo, H., Sanderson, B., Wu, <., Gogol, M., Alexander, T., Seidel, C., Wiedemann, L., Ge, K., Krumlauf, R., Shilatifard, A. (2009) Global analysis of H3K4 methylation defines MLL family member targets and points to a role for MLL1-mediated H3K4me methylation in the regulation of transcriptional initiation by RNA polymerase II. Molecular and Cellular Biology 29: 6074-6085  61  Wang, Z., Zhang, C., Cui, K., Schones, D., Barski, A., Peng, W., Zhao, K. (2009) Genome-wide mapping of HATs and HDACs reveals distinct functions in active and inactive genes. Cell 138: 1019-1031  Yu, B., Hess, J., Horing, S., Brown, G., Korsmeyer, S. (1995) Altered Hox expression and segmental identity in Mll-mutant mice. Nature 378:505-508 Zee, B., Levin, R., Xu, B., LeRoy, G., Wingreen, N., Garcia, B. (2010) In vivo residue-specific histone methylation dynamics. Journalof Biological Chemistry 285: 3341-3350 Zenter, G., Tesar, P., Scacheri, P. (2011) Epigenetic signatures distinguish multiple classes of enhancers with distinct cellular functions Genome Research 21:1273-1283     

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0074032/manifest

Comment

Related Items