UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of a novel fluorescent reporter of genomic imprinting in the mouse Jones, Meaghan Jessica 2010

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


24-ubc_2010_spring_jones_meaghan.pdf [ 182.54MB ]
JSON: 24-1.0070921.json
JSON-LD: 24-1.0070921-ld.json
RDF/XML (Pretty): 24-1.0070921-rdf.xml
RDF/JSON: 24-1.0070921-rdf.json
Turtle: 24-1.0070921-turtle.txt
N-Triples: 24-1.0070921-rdf-ntriples.txt
Original Record: 24-1.0070921-source.json
Full Text

Full Text

CHARACTERIZATION OF A NOVEL FLUORESCENT REPORTER OF GENOMIC IMPRINTING IN THE MOUSE by Meaghan Jessica Jones B.Sc.H., Mount Allison University, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Medical Genetics)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2010  © Meaghan Jessica Jones, 2010  Abstract Regulation of inserted transcriptional units by epigenetic means has been reported for many years, and has been used to study characteristics of epigenetic regulation. Some of these transgenes have become regulated by genomic imprinting, and thus are expressed from only one of the two parental chromosomes and, occasionally, acquire parent-of-origin-specific epigenetic markings such as DNA methylation. These transgenes in particular have been useful in elucidating mechanisms of imprinted regulation. Here is described the first imprinted fluorescent transgene, a green fluorescent protein (GFP) gene inserted in the distal MMU7 imprinted domain between the imprinting centers 1 and 2 (IC1 and IC2) regulated regions. This transgene, called Tel7KI, exhibits imprinted expression only from the maternal allele, and is silenced and DNA methylated on the paternal allele in post-implantation embryos. In the embryo this allelespecificity is consistent throughout all tissues and developmental stages analyzed except the developing germ line, making Tel7KI a potential reporter of epigenetic reprogramming in that lineage. In the placenta, imprinted expression and DNA methylation of Tel7KI is lost, and both alleles are expressed and methylated at moderate levels. Finally, an analysis of the effect of IC2 on silencing of Tel7KI in an embryonic stem cell differentiation assay revealed a possible extension of the region of influence for that imprinting centre a further 300kb proximal. Thus, Tel7KI has the potential to be an extremely useful tool in the study of genomic imprinting.  ii  Table of Contents  Abstract
......................................................................................................................................ii Table of Contents
.....................................................................................................................iii List of Tables
.............................................................................................................................vi List of Figures
..........................................................................................................................vii List of Abbreviations
..............................................................................................................viii Acknowledgements
...................................................................................................................ix Chapter 1: Introduction
............................................................................................................1 1.1 Historical overview of genomic imprinting
..........................................................................1 1.1.1 Parthenogenesis/androgenesis and uniparental disomy
..............................................1 1.1.2 Imprinted genes
...........................................................................................................2 1.1.3 Imprinted transgenes
...................................................................................................4 1.2 Function of imprinting and imprinted genes
.........................................................................7 1.2.1 Evolutionary history of imprinting
.............................................................................7 1.2.2 Role of imprinted genes during development
.............................................................7 1.2.3 Proposed models for the acquisition of genomic imprinting in mammals.
................8 1.3 Epigenetic regulation of imprinting during mammalian development
................................10 1.3.1 Imprint status in mature gametes
..............................................................................10 1.3.2 Genomic imprinting at fertilization and during early development
..........................11 1.3.3 Imprint resetting during gametogenesis
....................................................................12 1.4 MMU7 distal imprinted domain
..........................................................................................14 1.4.1 IC1-regulated sub-domain
.........................................................................................14 1.4.2 IC2-regulated sub-domain
.........................................................................................17 1.4.3 Human homology and disease
..................................................................................19 1.5 Tel7KI transgenic line
..........................................................................................................21 1.6 Thesis objectives
..................................................................................................................23 Chapter 2: Materials and methods
........................................................................................24 2.1 Mice
.....................................................................................................................................24 2.1.1 Mouse strains
............................................................................................................24 2.1.2 Genotyping
................................................................................................................24  iii  2.1.3 Embryo dissection
.....................................................................................................25 2.2 Tissue culture
.......................................................................................................................25 2.2.1 ES cell line derivation and culture
............................................................................25 2.2.2 Electroporation and selection of clones
....................................................................26 2.2.3 Embryoid body differentiation
..................................................................................27 2.3 Cell separation and flow cytometry
.....................................................................................27 2.4 Isolation of nuclei from placentae
.......................................................................................27 2.5 Ectoplacental cone collection and culture
...........................................................................28 2.6 Tissue preparation and sectioning
........................................................................................28 2.7 Immunofluorescence
............................................................................................................29 2.8 RNA analysis
.......................................................................................................................29 2.8.1 RNA preparation and reverse transcription
...............................................................29 2.8.2 RT-PCR
.....................................................................................................................30 2.9 DNA analysis
.......................................................................................................................30 2.9.1 DNA preparation
.......................................................................................................30 2.9.2 Sodium bisulfite modification, PCR and COBRA
....................................................30 2.10 Deletion of IC2 in KIP ES cells
.........................................................................................31 2.10.1 Targeting vector construction
..................................................................................31 2.10.2 Screening of IC2KO clones
....................................................................................32 2.11 Microscopy and image analysis
.........................................................................................32 Chapter 3: Tel7KI expression and DNA methylation in the embryo ..................................35 3.1 Introduction..........................................................................................................................35 3.2 Results .................................................................................................................................36 3.2.1 GFP from Tel7KI is expressed in a parent-of-origin-specific manner in the embryo ............................................................................................................................................36 3.2.2 GFP silencing from Tel7KI in the embryo is associated with increased DNA methylation at the CAG promoter......................................................................................39 3.2.3 Maternal transmission of Tel7KI results in an expression pattern consistent embryoto-embryo with highest levels of GFP expression in the heart and CNS ...........................44 3.2.4 GFP expression in paternal transmission embryos is observed stochastically in the heart and brain and consistently in PGCs ..........................................................................47 3.3 Discussion............................................................................................................................52 Chapter 4: Characterization of Tel7KI expression in the placenta and implications of endoreduplication on imprinting in trophoblast giant cells (TGCs)
..................................58 iv  4.1 Introduction
..........................................................................................................................58 4.2 Results
.................................................................................................................................59 4.2.1 GFP expression of Tel7KI in the placenta is observed upon both maternal and paternal transmission
..........................................................................................................59 4.2.2 DNA methylation at Tel7KI in the placenta is moderate on both parental alleles
....62 4.2.3 Tissue-specific analysis of Tel7KI and comparison with the X-linked GFP D4 allele shows similar GFP expression in TGCs
.............................................................................62 4.2.4 Culture of paternal transmission Tel7KI EPCs gives rise to TGCs which show high levels of GFP expression
....................................................................................................64 4.2.5 TGCs do not show epigenetic defects at endogenous loci
........................................70 4.2.6 DNA methylation at Tel7KI is established in the placenta after E8.5
......................70 4.3 Discussion
............................................................................................................................75 Chapter 5: Tel7KI ES cells as a model for acquisition of imprinting during early postimplantation development
.......................................................................................................81 5.1 Introduction
..........................................................................................................................81 5.2 Results
.................................................................................................................................82 5.2.1 Tel7KI insertion is conditional in KIO ES cells
.......................................................82 5.2.2 Both paternal and maternal transmission of Tel7KI result in GFP+ blastocysts and ES cell lines
........................................................................................................................84 5.2.3 KIP and KIM ES cell lines silence GFP differently upon differentiation
.................84 5.2.4 Role of IC2 in epigenetic regulation of Tel7KI during ES cell differentiation
.........87 5.3 Discussion
............................................................................................................................94 Chapter 6: Discussion
..............................................................................................................98 6.1 Summary of results and conclusions
...................................................................................98 6.2 General discussion
.............................................................................................................100 6.3 Future directions
................................................................................................................106 References
...............................................................................................................................109 Appendix
.................................................................................................................................125  v  List of Tables Table 1.1 Table 2.1 Table 2.2 Table 2.3 Table 5.1  Characteristics of previously documented imprinted transgenes
.............................6 Primer sequences for genotyping, bisulfite PCR and RT-Q-PCR
..........................33 Primer sequences for allele-specific analysis
.........................................................34 Antibody details
.....................................................................................................34 ES cell lines used in this study
...............................................................................86  vi  List of Figures Figure 1.1 Regulation of imprinted expression in the IC1 sub-domain
..................................16 Figure 1.2 Regulation of imprinted expression in the IC2 sub-domain
..................................18 Figure 1.3 Cre-mediated insertion at the Ins2 locus and structure of the Tel7KI allele
..........22 Figure 3.1 
 Imprinted expression of Tel7KI in post-implantation embryo
...............................37 Figure 3.2 Loss of GFP silencing in Tel7KI embryos
.............................................................38 Figure 3.3 GFP expression from Tel7KI by flow cytometry
...................................................40 Figure 3.4 
 Q-RT-PCR analysis of expression of GFP from Tel7KI
........................................41 Figure 3.5 
 DNA methylation at Tel7KI in the embryo.
...........................................................43 Figure 3.6 
 GFP expression in 14μm sections of E12.5 KI/+ embryos
....................................45 Figure 3.7 
 DNA methylation on the maternal allele of Tel7KI
...............................................46 Figure 3.8 
 Loss of imprinting in +/KI embryos in specific tissues
.........................................48 Figure 3.9 
 GFP expression from Tel7KI in the germ line in cross-sections
............................49 Figure 3.10 
 GFP expression from Tel7KI in the germ line in sagittal sections
.........................50 Figure 3.11 
 Summary of GFP expression in germ cells of +/KI embryos
................................51 Figure 3.12 
 Models for long-range regulation of Tel7KI
..........................................................55 Figure 4.1 
 GFP expression in whole placentae of Tel7KI conceptuses
..................................60 Figure 4.2 Placental expression of GFP from Tel7KI upon maternal or paternal transmission

................................................................................................................................61 Figure 4.3 DNA methylation at Tel7KI in E10.5 placenta
.....................................................63 Figure 4.4 
 GFP colocalisation with CD-34 in Tel7KI and X-linked GFP D4 placentae
.........65 Figure 4.5 Ploidy of nuclei isolated from E10.5 placentae and parietal yolk sacs of +/KI or +/+ embryos

................................................................................................................................66 Figure 4.6 
 Trophoblast giant cell growth from cultured ectoplacental cones (EPC)
..............68 Figure 4.7 Expression of giant cell markers in cultured EPC
................................................69 Figure 4.8 
 Allelic gene expression of imprinted genes in +/KI TGC
......................................71 Figure 4.9 
 Methylation of IC1 and IC2 on MMU7 in +/KI TGC
...........................................72 Figure 4.10 
 Allelic analysis of relative parental DNA contributions after endoreduplication in TGC 
................................................................................................................................73 Figure 4.11 
 Methylation of Tel7KI upon paternal transmission in uncultured EPCs (d0) and cultured TGCs (d5)
.....................................................................................................................74 Figure 5.1 
 Tel7KI insertion is reversible with Cre
..................................................................83 Figure 5.2 
 GFP expression in blastocysts and ES cells made from transmission of Tel7KI through the maternal (KI/+) or paternal (+/KI) germlines
.........................................................85 Figure 5.3 
 Paternal (KIP) and maternal (KIM) transmission Tel7KI ES cells show differences in GFP silencing during embryoid body differentiation
.............................................................88 Figure 5.4 
 DNA methylation at the CAG promoter in Tel7KI ESC lines before and after differentiation
..............................................................................................................................89 Figure 5.5 
 Deletion of KvDMR1 in Tel7KI ESC
....................................................................90 Figure 5.6 
 Screening of clones for IC2 KO
.............................................................................92 Figure 5.7 
 A knockout of IC2 in paternal transmission ES cells impairs GFP silencing upon differentiation
..............................................................................................................................93 Figure 5.8 
 An IC2 knockout impairs acquisition of DNA methylation at Tel7KI during ES cell differentiation
.......................................................................................................................95 vii  List of Abbreviations bp CAG CNS CpG DMR DNMT DTA E EB ES EPC ExM FACS GFP GFP+ GFP– IC ICR kb KI KIM KIO KIP KO LIF lncNRA LOI LOM Mb MEG MI ncRNA ORF PBS PCR PEF PEG PGC RT-PCR TGC UPD XCI  base pairs cytomegalovirus/chicken β-actin/rabbit β-globin enhancer/promoter cassette central nervous system cytosine-phosphate-guanidine dinucleotide differentially methylated region DNA methyltransferase diptheria toxin embryonic day embryoid body embryonic stem (cell) ectoplacental cone extraembryonic mesoderm fluorescence-activated cell sorting green fluorescent protein green fluorescent protein-positive phenotype green fluorescent protein-negative phenotype imprinting centre imprinting control region kilobase pairs Tel7KI Tel7KI ES cell line, maternal transmission Tel7KI ES cell line, original insertion Tel7KI ES cell line, paternal transmission lnockout leukemia inhibitory factor large non-coding RNA loss of imprinting loss of methylation megabase pairs maternally expressed gene meiosis I non-coding RNA open reading frame phosphate-buffered saline polymerase chain reaction primary embryonic fibroblast paternally expressed gene primordial germ cell reverse transcriptase-polymerase chain reaction trophoblast giant cell uniparental disomy X-chromosome inactivation  viii  Acknowledgements    I have had a lot of help on this project over the past few years, so there are a lot of people  to thank.   Thank you Louis for always being helpful, interested, and involved while keeping a great  sense of humor and the ability to have fun. I appreciate all your advice and hope that I can be as good a supervisor to my hypothetical future students.   Thank you Lefebvre Lab members, particularly Julie, Rosemary, and Aaron. Louis has  done a wonderful job of assembling a group that is smart, fun, and passionate about their work. I love the atmosphere of cooperation, sharing, and assistance, it made everything more enjoyable. I will miss you all.   Thank you to my thesis advisory committee, Wendy Robinson, Jane Roskams, and Fabio  Rossi for your ability to make me look at my project in different ways and get excited about it anew every time.   Thank you to all those who helped me learn tricky techniques, particularly Andy Johnson,  Jeff Duenas, and Jessica MacDonald.   Thank you to my science sisters; Sharan, Sara, Desiree, and Allison. We went through a  lot together and I wish you all the best as we move on to new and interesting things.   Thank you to my friends, especially the ones who laugh at my nerdy science jokes even  when you don’t think they’re funny. I am very lucky to have friends like you.   Thank you to my parents, Elizabeth and Malcolm, and my sister Kathryn for being  supportive even when you didn’t really understand what I was doing.   Thank you David, for always being there. Making me cookies when I’m stressed,  cooking dinner if I’d had a long day, distracting me from the all-consuming world of science with the equally fascinating world of art. I don’t know what I would do without you.  ix  Chapter 1: Introduction 1.1 Historical overview of genomic imprinting   Genomic imprinting results in the functional non-equivalence of parental genomes,  caused by differential epigenetic marking of the parental chromosomes. One of the consequences of this germline imprinting is the allele-specific expression of so called “imprinted genes” (1). The first imprinted genes were discovered in 1991, but the concept of non-equivalence of the parental genomes had first been described years earlier (2-4). Since then, more than 80 genes in mice and humans have been found to be imprinted, and have been shown to be major contributors to human growth and disease (5). 1.1.1 Parthenogenesis/androgenesis and uniparental disomy   The initial evidence for non-equivalence of the parental genomes came from experiments  on parthenogenetic (two maternal nuclei) and androgenetic (two paternal nuclei) embryos. In both these situations, the resulting embryo does not survive to term, though the defects are very different (2; 4; 6). Parthenotes develop grossly normal but undersized embryos, with little extraembryonic support structure leading to the growth defect, while androgenotes can give rise to normal extraembryonic tissues but fail to develop embryonic tissues (2; 4; 6). Both are lethal before mid-gestation. Thus, both parental genomes are required for normal development. Recently, parthenogenetic mouse embryos carrying mutations on specific imprinted regions were cloned and able to develop to term, grow to adulthood, and produce offspring of their own (7; 8). The mutations in two important imprinting centres, at H19 and Dlk1/Gtl2, gave the pronucleus derived from a mutant female a more “paternal” epigenotype .   Experiments using mice carrying translocations or duplications segmented the genome  into regions for which there was a requirement for both parental genomes and those for which both genomes were equivalent. These experiments involved mice with balanced translocations being intercrossed, resulting in offspring with balanced chromosome numbers but with both copies of a chromosome or part of a chromosome being inherited from the same parent (9-11). This chromosomal composition is referred to as uniparental disomy (UPD). Some UPDs caused 1  visible phenotypes which were different from either parent, indicative of parent-specific gene expression in the uniparental segment. For example, early experiments with a Robertsonian translocation between chromosomes 11 and 13 showed that paternal duplication of chromosome 11 resulted in embryos which were larger than their biparental littermates, and embryos carrying a maternal duplication were smaller (9). The same study used a smaller translocation of chromosome 11 to map the region responsible for the parent-of-origin effect to the proximal part of the chromosome (9). As more translocations were discovered or generated, the regions with a requirement for both parental contribution became smaller and smaller and a map of imprinted domains within the mouse genome was drawn. A similar map was created for human diseases known to be caused by imbalance in parental chromosome contribution; for example paternal UPD of 11p results in a disease called Beckwith-Wiedeman Syndrome while maternal UPD of chromosome 15q results in Prader-Willi Syndrome (12-14). 1.1.2 Imprinted genes   Eventually the first imprinted genes were discovered in 1991 (3; 15-17). These genes are  expressed from only one of the two parental alleles and often carry epigenetic modifications which are also parent-of-origin specific. Eighty-three imprinted genes have been identified in human and mouse, though some prospective bioinformatic studies have predicted up to 600 genes will eventually be found to be imprinted (5; 18). Of these 83, 41 are imprinted in humans, 71 in mice, and 29 in both species (5). Other than allele-specific expression, many imprinted genes share other characteristics.   One of the common features of imprinted genes is their location in large clusters. Some  imprinted genes do not have known imprinted neighbours (Sfmbd2, Gatm, Impact, for example) but these are in the minority. This clustering in many cases has been hypothesized to occur due to (or result in) the presence of an element responsible for the imprinted expression of the other genes in the cluster (19). These elements are termed Imprinting Control Regions (ICRs) or Imprinting Centres (ICs), and are generally associated with parent-specific DNA methylation at a CpG island inherited from the germ line (19). For example, at the Igf2r/Air locus on mouse chromosome 17, an ICR resides in an intron of the maternally-expressed Igf2r gene, and is associated with production of the paternally-expressed ncRNA Air. Deletion of this element has 2  been shown to abolish imprinted expression of both these genes as well as others in the locus, a prerequisite for the “ICR” nomenclature (20).   ICRs are one of two kinds of differentially methylated regions (DMRs) normally  associated with imprinted genes, and allele-specific methylation at ICRs is sometimes referred to as a gametic imprint, or primary imprint (21; 22). This is to differentiate them from somatic or secondary DMRs, which are established after fertilization. Both kinds of DMRs have DNA methylation at a CpG island which is found on only one of the two parental alleles. DNA methylation at a gene most commonly has a silencing effect, by blocking the access of transcription factors or RNA polymerase, though it can have an activating effect if the DNA methylation inactivates a silencer or activates an enhancer, as is observed in the Igf2 gene (23-25). At the previously-mentioned Igf2r/Air locus, while the DMR at the Air locus is an ICR, Igf2r contains its own DMR, methylated on the silent paternal allele, but deletion of this element does not result in loss of imprinting at any of the genes in the region (26). Differential DNA methylation at this DMR is acquired later in development at approximately E6.5, not during gametogenesis as is the Air ICR (27). The Igf2r DMR, then, is a secondary or somatic DMR, dependent on the Air ICR.   Maintenance of DNA methylation at DMRs through cell divisions is accomplished  mainly by the maintenance DNA methyltransferase, Dnmt1 (28). Embryos null for Dnmt1 die at midgestation with substantial demethylation of genomic DNA (28). At imprinted genes, allelic expression of many imprinted genes is lost in the absence of differential DNA methylation, with some exceptions which will be discussed later (29). Overexpression of Dnmt1 results in unusual hypermethylation at only some imprinted genes; the ICR controlling H19 and Igf2 expression becomes methylated on the maternal allele, leading to biallelic expression of Igf2, while imprinted expression and DNA methylation of Snrpn, Peg3, Igf2r, and Rasgrf1 remain unchanged (30).   A third characteristic of imprinted genes is another epigenetic modification, allele-  specific modification of histone proteins to modulate gene expression (31; 32). Addition of methyl, acetyl, or some other small groups to the C-terminal tails of histone molecules has been shown to affect the amount of chromatin compaction and, thus, the accessibility of DNA to transcription factors, RNA polymerase, or DNA repair enzymes (reviewed in ref. 33). At imprinted genes, as with DNA methylation, these marks often appear in an allele-specific manner 3  (31; 34-36). Mutation or deletion of enzymes responsible for modification of histone tails has been shown to interfere with allele-specific expression of imprinted genes (37; 38). In addition, silencing histone marks are frequently associated with DNA methylation on specific alleles. For example there is an enrichment of H3 K9 methylation on the silent and DNA methylated promoter of the Air ncRNA (32; 39).   Imprinted genes have also been reported to replicate asynchronously (40-43). So far, all  imprinted clusters examined have replicated the paternal allele first as shown by in situ hybridization, including both imprinted domains on distal MMU7, the Prader-Willi/Angelman Syndrome region on MMU6, and the Igf2r/Air cluster on MMU11 (40; 43). There is some evidence that the differences in replication timing are dependent on cell type and gene expression level, though these data have not been reproduced (44). More recent analysis has shown replication asymmetry to be established very early in mouse development, erased along with allele-specific DNA methylation in primordial germ cells, and abolished in cases of UPD (41; 43; 45).   Parent-of-origin-specific expression and location of genes in an imprinted cluster have  frequently been used to identify novel imprinted genes, and allele-specific DNA methylation and histone modifications, or asymmetry of replication timing, have been tested to confirm their imprinted status (46-49). For a gene to be considered imprinted, it must be expressed in a parentof-origin-specific manner at some point during development or the life cycle. There is some debate as to whether a gene showing allelic bias fulfills the criteria to be called imprinted, and which other characteristics are necessary to call it truly imprinted. 1.1.3 Imprinted transgenes   Many early studies of imprinting were based on the characterization of transgenic  constructs. These mainly come in two categories; transgenes carrying parts of a known imprinted domain inserted elsewhere in the genome, which were used to determine the sequences sufficient for proper imprinting of a known locus at ectopic sites, and transgenes with no relationship to existing imprinted genes which nonetheless became imprinted. Both types have been extremely useful in elucidating the mechanisms by which imprinted genes are regulated, or by which inserted transcriptional units initially become imprinted (50-53). 4    A common way to confirm genetically the location of regulatory regions with respect to a  gene of interest is to create large transgenes containing the region and examine its expression at ectopic sites in the presence and absence of specific sequence motifs. This approach has been extensively used to identify imprinting centres and DMRs, and many examples exist of large transgenes carrying parts of an endogenous imprinted locus being inserted at other regions of the genome (51; 52; 54-58). These transgenes are then analyzed for characteristics of imprinted genes, to determine which parts of the endogenous locus are responsible for features such as allele-specific expression, DNA methylation, or replication asymmetry. This approach is not without complications; in one case a transgene has been shown to cause silencing at its endogenous locus in trans (59). Additionally, these transgenes generally examine regulatory regions of a single gene cluster at a time. So, while extremely useful for elucidating regulation of a particular gene of interest, these transgenes have limited usefulness for discovering novel features of imprinted expression in general.   Some transgenes show imprinted expression despite having no relationship to a known  imprinted gene. Some of these transgenes carry imprinting signals, such as the RSVIgmyc transgene, which contains approximately ten tandem copies of a construct containing a Rous Sarcoma virus (RSV) LTR followed by an immunoglobulin/c-myc fusion gene (53). It is expressed in the heart exclusively from the paternal allele, and acquires DNA methylation on both maternal and paternal alleles during gametogenesis, though DNA methylation on the paternal allele is lost in early development (53; 60). In this case, the DNA methylation acquired at the transgene is not found at the locus into which the transgene is inserted, indicating both that the locus does not normally become differentially methylated and that the methylation mark from the transgene does not spread beyond its boundaries (60). The resistance of these imprinting marks to exogenous factors was confirmed by the observation that multiple different insertions sites showed the same imprinted expression (61). This insertion does not recapitulate features of imprinted genes exactly. For example, the active paternal allele acquires some DNA methylation, though not to the extent of the silent maternal allele, during postimplantation development, and replication of the parental alleles of RSVIgmyc is synchronous (60-62). Another imprinted transgene was also inserted in multiple genomic locations, but was imprinted only in one of these, on MMU11, in a domain which had already been linked with UPD effects (63). This transgene, MPA434, contains the mouse metallothionein-I promoter fused to human 5  transthyretin, and is maternally methylated in the oocyte and paternally expressed in adult tissues (63; 64). Thus transgenes are capable of acquiring imprinted expression and allele-specific methylation established in the germline either through signals originating from the transgene itself, or through effects of insertion location. Other imprinted transgenes show some of the features of imprinted gene expression and not others, these and the two previously mentioned are summarized in Table 1.1 (65-67). Note that from these transgenes, only RSVIgmyc was convincingly shown to contain an internal imprinting signal, capable of directing paternal allelespecific expression at several different insertion sites in the genome. All the transgenes discussed here acquire maternal allele-specific DNA methylation marks, as primary germline DMRs or as somatic DMRs. Table 1.1: Characteristics of previously documented imprinted transgenes Name  Parental Expression  DNA methylation  Tissues analyzed  CAT17  n/a  embryo: maternal placenta: none  embryo, placenta, yolk sac  HBsAg  paternal  maternal  TN1  n/a  RSVIgmyc  MPA434  Method  Reversible  Imprinted at other loci  Reference  pronuclear injection  yes  no  (66)  embryonic liver #  pronuclear injection  no  no  (67)  maternal *  adult  pronuclear injection  yes*  no  (65)  paternal  maternal, germline  embryonic heart #  pronuclear injection  yes  yes  (53)  n/d  maternal, germline  adult tail  pronuclear injection  yes  no  (63)  * Passage through the maternal or paternal germlines resulted in an increase or decrease in DNA methylation level, respectively, as opposed to full methylation or demethylation. # Tissue-specific transgene expression n/a: not assayed n/d: not detected, no transcript expressed from this locus  6  1.2 Function of imprinting and imprinted genes 1.2.1 Evolutionary history of imprinting   Of the three mammalian lineages, monotremes (or prototherians, platypus and echidna  being the only known extant species) show no evidence of imprinting despite the fact that many genes which are imprinted in other mammalian species are conserved in monotremes, and organized in similar clusters (68). Marsupials (metatherians, such as wallaby and opossum) exhibit genomic imprinting on a somewhat smaller scale than placental mammals (eutherians, such as mouse and human) (69-71). To date, six genes known to be imprinted in humans and mice have been shown to be imprinted in marsupials; H19, Igf2, Ins2, Igf2r, Peg1/Mest, and Peg10 (69; 72-74). Of these, only H19, Igf2, and Peg10 show differential DNA methylation of regulatory regions, as is observed at their orthologues in placental mammals (69; 72-75). Thus the acquisition of imprinting appears to be correlated with the development of viviparity and maternal contribution to fetal nutrition: monotremes are egg-laying mammals, do not bear live young but do produce nutrients for them in early stages of life; marsupials bear live young and supply nutrients to the offspring for an extended extra-uterine developmental period; and eutherians have a highly developed maternal/fetal transfer system which continues after birth. This evidence suggests that the acquisition of imprinting as a regulatory mechanism may have evolved at the same time as increases in maternal/fetal nutrient exchange. 1.2.2 Role of imprinted genes during development    Now that more than 80 imprinted genes have been identified in mouse, some  generalizations can be made as to their functions. This is complicated by the fact that imprinted genes occur in clusters, and some members of the cluster may be imprinted simply due to their proximity and not because the gene has a function which would have made it susceptible to becoming imprinted. This phenomenon has been referred to as the “bystander effect” (76). Bearing this in mind, developmental genes are highly represented among all imprinted genes. In a census of imprinted genes, the most over-represented Gene Ontology (GO) terms were  7  organogenesis, morphogenesis, regulation of cell cycle, cell growth and/or maintenance, and neurogenesis, in that order (5).   Many imprinted genes have functions in the placenta (reviewed in refs 77-79). The fully  developed placenta is a temporary organ unique to therian mammals; metatherians (marsupials) have a primitive and very short-lived placenta while eutherians have a highly developed organ which lasts the entirety of fetal development. The mouse placenta has three main layers: the trophoblast giant cell layer, responsible for trophoblast invasion and found against the maternal decidua, the spongiotrophoblast layer, which appears to be structural though its actual function is unknown, and the labyrinth layer, comprised of blood vessels and support structures, where nutrient transfer occurs (80). Imbalances in imprinted genes have been reported to have striking effects on placental weight and morphology (81; 82, reviewed in 77). Interestingly, some maternally and paternally expressed genes have similar placental knockout phenotypes. For example, the maternally-expressed Ascl2 gene and the paternally-expressed Peg10 gene both show an absence of spongiotrophoblast cells and a disorganized (Ascl2) or underdeveloped (Peg10) labyrinth (83; 84). Knockout of the maternally-expressed Grb10 or Phlda2 genes shows an increase in embryonic and placental weights, while deletion of the paternally-expressed Igf2 gene results in smaller placentae (81; 85; 86). These and other examples show the vital role imprinted genes play in proper establishment of maternal-fetal nutrient transfer. 1.2.3 Proposed models for the acquisition of genomic imprinting in mammals.   Many theories have been proposed as to how genes became imprinted during evolution.  The benefits of diploidy are eliminated for genes which become functionally haploid when one allele is inactivated, so there must exist selective pressure to become imprinted. One of these, the host defense model, proposes that genomic imprinting is an evolutionary hijacking of epigenetic modifications which were developed to protect the organism from invasion by foreign DNA (87). This model is based on the observation that some transgenes become DNA methylated in a parent-of-origin-specific manner, which may be an effect of a defensive strategy to silence invading DNA. It is possible that the reason some genes become DNA methylated is because they resemble foreign DNA (87). Given that the great majority of imprinted genes have functions in growth and development, two of the more popular models involve growth; the ovarian 8  teratoma model and the parental conflict hypothesis. The ovarian teratoma model, or “ovarian timebomb hypothesis”, suggests that the reason paternally-expressed genes exist and are required for development of extraembryonic tissues is in order to protect female mammals from activation of eggs and subsequent cancerous invasion of trophoblast-like tissue (76). Major criticisms of this model include that it does not propose a selection pressure for maternally-expressed genes, nor does it apply to several imprinting-like phenomena observed in plants and sciarid flies. (88-90).   Perhaps the most widely accepted theory as to the development of genomic imprinting in  mammals is the parental conflict hypothesis. This model proposes that the fundamental selection pressure leading to parent-of-origin-specific expression of genes is the difference in reproductive strategies which exist between male and female mammals (91; 92). This theory states that upon the evolution of viviparity, female mammals were required to invest large amounts of energy into each offspring, to the theoretical detriment of future offspring. Male mammals, on the other hand, contributed relatively little to each offspring during its developmental period. Thus, female mammals would have an advantage to limit the total amount of resources attributed to each offspring, to maximize the total number of offspring they were able to have. The advantage for males, on the other hand, would be to have the largest and fittest offspring, regardless of the potential impact on the mother’s future reproductive success. This conflict may have resulted in what has been termed an “epigenetic tug-of-war” with respect to genes involved in nutrient allocation in the prenatal and immediate postnatal period (93).   Often cited as an example of this process is the regulation of the Igf2 gene. Igf2 is an  embryonic mitogen which promotes fetal growth. Deletion of this gene in the mouse results in pups which are comparatively small, and over-expression results in large pups (3; 85). Igf2 is paternally-expressed, fitting very well with the parental conflict model which predicts that, over evolutionary time, the paternal germline would preferentially express genes which enhance growth (16). Even more interestingly, there are two antagonists of Igf2 expression, H19 and Igf2r, both of which are maternally expressed (15; 94). H19 competes with Igf2 for enhancers, while IGF2R is a non-signaling receptor for IGF2, showing that both genetically and biochemically, the maternal and paternal genomes appear to be in competition at this locus (71; 95; 96).  9    From this model, certain predictions can be made. Since clusters of imprinted genes  contain genes not known to be involved in embryonic growth or nutrient allocation, it follows that some of these genes became imprinted by the bystander effect. An example of this is the acquisition of imprinting at the Frat3/Peg12 gene in rodents. This retrotransposon-derived gene is located on the telomeric end of the subproximal MMU7 imprinted domain, homologous to the human Prader-Willi Syndrome region on 15q11-15q13 (47). The gene is present in only rodent lineages, and has acquired paternal-specific expression due to its proximity to an imprinted domain (47; 97). This implies that novel rearrangements or insertions near imprinted domains can result in spreading of imprinted expression from a known domain to novel sequences, assuming the newly-inserted or translocated genes are capable of becoming regulated in this manner. 1.3 Epigenetic regulation of imprinting during mammalian development   Since imprinted genes are known to have important functions in mammalian  development, it stands to reason that their expression is tightly controlled. Many levels of regulation come to bear on imprinted genes, and epigenetic modifications during mammalian development affect timing and allele-specificity of imprinted gene expression. 1.3.1 Imprint status in mature gametes    The initial parent-specific differences in epigenetic modification of imprinted genes are  inherited from mature gametes. Haploid germ cells carry all the genetic and epigenetic information necessary to engender the next generation, including the modifications necessary for correct expression of parent-of-origin-specific genes (98; 99). One important feature of imprinted genes found in germ cells is DNA methylation of ICRs; each ICR (or primary DMR) is found methylated in either sperm or egg, but not both (19; 100-102). Thus, mature gametes carry with them only a single parental epigenotype, different from somatic cells which normally contain genetic and epigenetic contributions from both parents (100).   Male and female gametes show some differences in their modifications at imprinted  genes. The most obvious being the stripping of most histones from DNA in mature sperm head 10  and replacement with protamines (103). Those few histone nucleosomes which are retained in mature sperm have been shown to be enriched at key early development genes and imprinted genes (104). In addition, many more ICRs are maternally methylated than paternally methylated, despite the relative equality in numbers of MEGs and PEGs. This implies that either paternal ICRs have a tendency to control expression of more genes than maternal ICRs, or that MEGs and PEGs tend to be regulated by different pathways. Only three ICRs are methylated in sperm; at H19, at the IG-DMR of the Dlk1/Gtl2 cluster, and at Rasgrf1, whereas more than 15 ICRs are known to be methylated in in eggs (105; 106). 1.3.2 Genomic imprinting at fertilization and during early development    At fertilization, the differences in degree of chromatin packaging between egg and sperm  result in an asymmetry between male and female chromatin. Maternal DNA maintains its histones and accompanying modifications while the paternal pronucleus retains few or no modified histones (103; 107). Immediately post-fertilization, the paternal pronucleus undergoes an egg cytoplasm-mediated chromatin decondensation and is globally demethylated at all sites except the three paternal ICRs previously mentioned (108-110). This active demethylation is hypothesized to occur in order to erase any established lineage-specific marks and re-establish totipotency (108; 109). Once the cleavage divisions begin, the maternal genome also becomes gradually demethylated until both genomes have reached an equivalent low methylation level at the time of implantation (108; 110).   The maintenance of DNA methylation at imprinted genes during this phase of  demethylation has been extensively studied. Three genes have been shown to be essential to maintenance of imprints in the immediate post-fertilization period: Dnmt1o, PGC7, and Zfp57. Dnmt1o is a maintenance DNA methyltransferase 1 variant expressed from an oocyte-specific promoter and deposited in mature oocytes (111). It has been shown to translocate to the nucleus of the embryo at the 8-cell stage, and mutants for this gene lose methylation on the maternallysilent Snrpn and paternally-silent H19 alleles (112). PGC7, a putative DNA binding protein of unknown function, has been shown to protect DMRs from active demethylation by egg cytoplasm. In the absence of PGC7, H19 and Snrpn become demethylated on the silent allele, though the Dlk1/Gtl2 IG-DMR remains properly methylated (113). Conversely, in the absence of 11  ZFP57, a KRAB zinc finger protein, paternal methylation of the Dlk1/Gtl2 IG-DMR is lost, and the H19 DMR is unaffected (114). Interestingly, this deletion also impairs the establishment of maternal methylation marks at the Snrpn DMR (114). These results imply that a combination of factors maintain DNA methylation during this phase, perhaps differing at different imprinted domains.   At the blastocyst stage, de novo DNA methylation occurs in developing inner cell mass  but not trophoblast cells (110). At imprinted genes, secondary DMRs begin to become methylated in an allele-specific manner also at this stage (27; 115). In many cases, this is believed to be due to the actions of large ncRNAs regulated by ICRs and causing downstream regulation of other imprinted genes in the cluster (116). Many imprinted gene clusters also contain smaller ncRNAs, including microRNAs, snoRNAs, and snRNAs, some of which still have unknown function. Examples of imprinting clusters with lncRNAs associated with ICRs include the Igf2r/Air locus on MMU17, the H19 and KvDMR1 loci on MMU7, and the Dlk1/Gtl2 locus on MMU12 (117-120).   Once the embryo has implanted and the embryonic and extraembryonic lineages begin to  differentiate further, the epigenetic differences between these lineages are maintained (121). The de novo methylation which occurs in the ICM is maintained in the embryo, but the trophoblast never acquires the same amount of DNA methylation and remains globally hypomethylated during development (121; 122). Perhaps because of this general hypomethylation, some imprinted genes in the placenta rely more heavily on histone modifications to maintain silencing of one allele than on DNA methylation. This is manifested in the distal MMU7 imprinted domain, for example, by maintenance of imprinted expression of some genes in the placenta even in the absence of the DNA methyltransferase Dnmt1 (31). 1.3.3 Imprint resetting during gametogenesis   The germ line is set apart from what will become the extraembryonic mesoderm by  BMP4 signaling from the extraembryonic ectoderm at or around E6.5 (123; 124). The primordial germ cells (PGCs) first arise at the base of the allantois, and are carried into the hindgut during invagination at E8.5 (125). From there, PGCs migrate anteriorly along the hindgut until they reach the location of the developing gonadal ridges, at E9.5 when they migrate dorsally through 12  the mesentery and then laterally into the new genital ridges (126). They reach the ridges at approximately E10.5-11.5. During this migration process, epigenetic reprogramming begins (reviewed in ref. 127). The presumptive germ cells must erase their current epigenetic modifications in order to facilitate resetting of their imprinting marks based on whether the germ cell will be an egg or a sperm. It is unknown and somewhat controversial exactly when this reprogramming begins; some investigators believe it begins when the PGCs enter the gonadal ridge and some believe it begins earlier, during the migration period (128-131). Regardless of when the process begins, by E12.5 it is mostly complete and the PGCs are now in what is referred to as a “naive” state, with no parent-specific DNA methylation (128; 129). From here the dynamics of imprint re-establishment vary depending on whether male or female gametes are being reprogrammed.   Male and female gametes arrest at approximately E13.5-E15.5, male gametes in mitosis  and female in prophase of meiosis I (MI), and DNA methylation is acquired in a sperm- or eggspecific pattern at different stages. In eggs, methylation at the maternally-methylated DMRs is acquired during oocyte maturation, in the period immediately after birth while the eggs are in the growth phase arrested in MI and is complete before the eggs complete the first meiotic division (99; 101; 132; 133). This de novo methylation is variable between genes; some genes initiate de novo methylation at earlier stages than others, though all genes show nearly 100% methylation in mature fully grown oocytes (101; 133). In sperm, DNA methylation at the paternal H19, Rasgrf1 and Dlk1/Gtl2 DMRs begins before birth in early spermatogonia. This process is mostly complete before entry into meiosis, though 100% methylation at these DMRs is not observed until mature sperm are formed (102; 132). Interestingly, in spermatogenesis the phenomena of epigenetic memory has been noted at the H19 locus. In this situation, the allele in the newlyerased prospective sperm which was inherited from the father becomes methylated before the allele which was inherited from the mother (102; 134). This implies some sort of mark exists which is not completely erased during the reprogramming phase, maintaining an “epigenetic memory”. The existence of this phenomenon suggests the possibility for trans-generational epigenetic effects, or “grandparental effects”, though so far no confirmed events have supported this theory.  13  1.4 MMU7 distal imprinted domain   A region of approximately 1 Mb on distal MMU7 contains two clusters of imprinted  genes. The region was identified early on as one for which parental duplications result in abnormal phenotype and, was one of the regions to show lethality in UPD embryos (10). Each cluster is controlled by an allele-specific germline DNA methylation mark at an imprinting centre (IC). The region between the two clusters is approximately 300 Mb in the mouse, is gene-poor, and is extremely rich in repeats. While the interval itself is not conserved between mammalian species, the two imprinted clusters are closely linked throughout eutherian mammals. Studies deleting this region have shown no major phenotypes associated with its loss, though subtle changes remain to be elucidated (ref. 135 and R. Oh-McGinnis and A. Bogutz, unpublished data). 1.4.1 IC1-regulated sub-domain   The proximal imprinted domain on distal MMU7 contains three imprinted genes; H19,  Igf2, and Ins2. H19, the most proximal of the three, is a maternally-expressed untranslated RNA which is expressed and imprinted in all mouse tissues with the exception of brain (15). It is imprinted in both embryonic and extraembryonic tissues (134). The H19 gene is regulated by a CpG-rich promoter, which acquires DNA methylation preferentially on the silent paternal allele peri-implantation, acting as a secondary DMR (136). The H19 transcript has been shown to function as a primary microRNA precursor for mi-675, a microRNA expressed in neonatal mice and conserved in rodents, and to function as a tumour suppressor gene (137; 138). Mice carrying deletions of the H19 RNA are viable and fertile, though they have a small growth phenotype linked to mis-expression of Igf2 (117).   IGF2 is a paternally-expressed insulin-like growth factor also not essential for embryonic  development; deletions of Igf2 result in progeny with a significant growth defect, which is rescued by a further deletion of the maternally-expressed IGF2-receptor, Igf2r (85; 139). Igf2 regulation is complicated by the presence of three DMRs in addition to the H19 ICR, as well as multiple promoters. Deletion of the silencer element in the paternally-methylated DMR1, located upstream of promoter 1, results in biallelic Igf2 expression with no effect on H19 expression 14  (24). DMR2, also paternally methylated and located in the last exon, contains a methylationsensitive activator, deletion or demethylation of which results in decreased Igf2 expression (25). DMRs 1 and 2 do not fall neatly into the primary or secondary DMR categories. They are both methylated in sperm, but lose their allele-specific DNA methylation during the wave of preimplantation demethylation, and re-acquire it in early postimplantation stages (94; 140). The third DMR, DMR0, regulates a placental-specific Igf2 promoter, and is maternally methylated but has an as yet unknown function (141). The H19 ICR is required to protect the Igf2 DMRs 1 and 2 from methylation on the maternal allele, and DMR1 protects DMR2 from methylation (142).   The third gene, Ins2, is one of two insulin genes in mouse, and its human orthologue INS  is the sole insulin gene found in humans. Its imprinting pattern is not well characterized, since it is only known to be imprinted in yolk sac endoderm during late embryonic development (143; 144). However, it is known to respond to imprinting signals from the H19 DMR, as deletion of the DMR has been shown to abolish paternal-specific expression of the gene (145). Little recent characterization of the imprinted pattern of Ins2 has been performed.   All three genes in this cluster are controlled by an ICR located upstream of the H19  promoter, referred to in this study as IC1 (Figure 1.1). Paternal deletion of IC1 results in hypomethylation of the H19 promoter and subsequent activation from the paternal allele, and reduction of Igf2 expression (146). IC1 is methylated on the paternal allele in sperm, unmethylated in egg and contains four binding sites for the CTCF protein (94; 101; 147). CTCF, a zinc-finger protein, blocks access to downstream enhancers by promoters and its binding at the H19 ICR is regulated by DNA methylation (148; 149). Proximal to the H19 gene are endodermal enhancers which can act on both H19 and Igf2 gene expression (95). When CTCF is bound to the unmethylated maternal IC1, the enhancers are blocked from acting on the Igf2 promoters and so Igf2 is expressed only from the paternal allele where CTCF binding is eliminated by DNA methylation. Absence of CTCF binding on the maternal allele leads to decreased H19 expression and acquisition of DNA methylation at the H19 promoter (150; 151). Interestingly, this enhancer competition model for H19 and Igf2 imprinting does not appear to apply to Ins2, as deletion of the endodermal enhancers does not change its expression levels (95).   The allele-specific binding of the CTCF protein to the DMR and subsequent access to  enhancers is part of the greater “chromatin-looping” model for imprinted regulation of H19 and 15  A  IC1 D0  D1  D2  ?  pat  p0  p1  Ins2  Igf2  H19  B  H19  pat  ? D1  D2  Ins2  Igf2  IC1  C mat  D0  D1  p0  p1  Ins2  D2  CTCF  Igf2  H19  D  mat  F CTC  H19  D1  2 Ins  D2  Igf2  Figure imprinted expression in the IC1 sub-domain. Figure 1.1: 1.V: Regulation Regulation ofofimprinted expression of the proximal MMU7 imprinted domain. Genes inin grey grey are are paternally paternally expressed, expressed,inin white white are are maternally maternally expressed. expressed.White White lollipops Genes lolipops represent unmethylated DMRs, black represent methylated DMRs. Large greyrepresent represent unmethylated DMRs, black represent methylated DMRs. Large grey ellipses ellipses represent silenced heterochromatic regions. A) and C) Linear representations of Linear silenced heterochromatic regions and shaded circles represent enhancers. (A) and (C) paternal (A) or maternal (C) regions showing Igf2 DMRs (d), and promoters (p), IC1, and representations of paternal (A) or maternal (C) regions showing Igf2 DMRs (D), and promoters gene structure orientation. B) and D)(B)Model of Model chromatin looping showing (p), IC1, and geneand structure and orientation. and (D) of chromatin looping showing interactions between DMRs resulting in sequestering of Igf2 into heterochromatic loop interactions between DMRs resulting in sequestering of Igf2 into heterochromatic loop mediated mdiated by binding proteins CTCF and an unknown paternal factor (?). Adapted from by binding proteins CTCF and an unknown paternal factor (?). Adapted from (153). Murrell et al, (2004).  16  Igf2 (Figure 1.1). This models has been explored in detail using chromatin conformation capture (3C), a technique which detects genomic DNA sequences located in close proximity in the nucleus (152). In this model, the proximal endodermal enhancers are one end of a silent-state chromatin loop, and fold back to make contact with DMRs on the distal side of H19. On the maternal allele, the DMR is unmethylated and CTCF is bound. This CTCF complex then binds to Igf2’s DMR1, partitioning its p1 and p2 into the silent chromatin loop and resulting in silencing of the maternal allele of Igf2 (Figure 1.1 C and D) (142; 153). H19, however, is outside the loop and free to interact with the enhancers. On the paternal allele, where DNA methylation is observed, other as yet unidentified proteins are bound, and form an interaction with Igf2’s DMR2, leaving DMR1 and its associated promoters p1 and p2 outside the silent chromatin loop and able to access the enhancers and be expressed (Figure 1.1 A and B) (142; 153). The secondary DMR at H19 acquires DNA methylation spreading from the ICR and thus H19 is silenced (136).  1.4.2 IC2-regulated sub-domain   The distal imprinted sub-domain on MMU7 is regulated by a maternally-methylated IC  R(IC2) which serves also as the promoter for an 80-120kb large nuclear ncRNA, Kcnq1ot1 (154). The DMR and its associated transcript are located in intron 10 of the Kcnq1 gene, a potassium ion channel expressed maternally in the embryo but showing biallelic expression in adult mouse and human (155-157). Deletion of IC2 or truncation of the Kcnq1ot1 transcript both result in loss of imprinted expression of eight genes located in the sub-domain; Ascl2, Cd81, Tssc4, Kcnq1, Cdkn1c, Slc22a18l, Phlda2, and Osbpl5 (listed in order - proximal to distal, Figure 1.2) (119; 155; 158).   Interestingly, the genes found in the IC2 sub-domain fall into two categories with respect  to their imprinted expression during development: those which are imprinted only in the placenta and those that are imprinted in both embryo and placenta. The genes located further from IC2, Ascl2, Cd81, Tssc4, and Osbpl5, are imprinted only in the placenta and are biallelic or not expressed in the embryo (48; 159). Analysis of blastocysts, ES cells and trophoblast stem (TS) cells suggests that the signals from the ICR travel along the chromosome slowly, such that by the time the embryonic and trophoblast lineages split, these signals have not yet reached the genes 17  IC2  Th  Ascl2  Tspan32  Trmp5 Tssc4 Cd81  Kcnq1  Kcnq1ot1  Nap1l4 Phlda2 Slc22a18 Cdkn1c  Cars  Osbpl5  pat  50Kb IC2  Th  Ascl2  Tspan32  Trmp5 Tssc4 Cd81  Kcnq1  Kcnq1ot1  Nap1l4 Phlda2 Slc22a18 Cdkn1c  Cars  Osbpl5  mat  Figure Figure 1.2 Regulation of imprinted expression in the IC2 sub-domain. 1.W: Regulation of imprinted expression of the distal MMU7 imprinted domain. Linear Linear representation of gene order, distribution and transcriptional orientationorientation for maternally repreentation of gene order, distribution and transcriptional for expressed (white) expressed or paternally expressed (grey)expressed imprinted(grey) genes. Location of non-imprinted maternally (white) or paternally imprinted genes. Location of genes are shown in black. On theinpaternal chromosome top), the(pat, ncRNA non-imprinted genes shown black. On the paternal (pat, chromosome top), Kcnq1ot1 the ncRNAis expressed and represses the maternally-expressed in cis, including genes the acquisition of DNA Kcnq1ot1 is expressed and represses the genes maternally-expressed in cis, including the acquisition of DNA at thelines DMR represent of Cdkn1c.signals Solid lines represent signals and in methylation at the DMR of methylation Cdkn1c. Solid in both embryonic both embryonic and extraembryonic lineages, dotted lines represent placental-specific extraembryonic lineages, dotted lines represent placental-specific signals. On the maternal signals.(mat, On the maternalIC2 chromosome (mat, and bottom), IC2 is methylated and Kcnq1ot1 is chromosome bottom), is methylated Kcnq1ot1 is not expressed, allowing not expressed, allowing expression of the maternal genes. expression of the maternal genes.  18  located further from the IC (34). Thus, only the proximal “ubiquitously-imprinted” genes are imprinted in both lineages. After the lineage split, spreading of the silencing signal on the paternal chromosome occurs only in extraembryonic lineages, resulting in the distal “placentally imprinted” genes (34). This is hypothesized to reflect differences in the way epigenetic regulation occurs in these two lineages; imprinting in this domain in the placenta relies more heavily on histone modifications than DNA methylation, and “placentally imprinted” genes retain their imprinted expression in the absence of the maintenance methyltransferase Dnmt1 (31; 160).   Unlike IC1, the regulatory signals for this sub-domain are not well established. Other  than the presence of the ICR and the transcript from the Kcnq1ot1 locus, little is known about required features for regulation of the domain. CTCF binding sites near IC2 have been identified and shown to be paternal allele-specific, though their exact function remains unknown (161). A recent study found that paternally-expressed Kcnq1ot1 transcript is associated in cis with a silent nuclear compartment covering at least part of the domain in a Dicer-independent manner (154). A fascinating observation from that study was that the silent domain was larger in the placenta than in the embryo, consistent with spreading of the silencing signal over longer genomic distance in that tissue (154). One of the complications to studying this domain is that differences exist in imprinted regulation of expression pattern between genes in the region, in addition to the two types of imprinted genes regulated by IC2. For example, Cdkn1c is the only gene in the region that contains its own functional secondary DMR, methylated on the silent paternal allele (115). In a mouse line carrying a deletion of the chromatin remodeling complex member gene Lsh, Cdkn1c is the only gene in the whole distal MMU7 imprinted region which shows loss of imprinted expression along with loss of its DMR methylation (162). Additionally, Cdkn1c imprinting is not completely abolished in truncations of Kcnq1ot1, again unlike the other genes in the region (119). 1.4.3 Human homology and disease   The MMU7 distal imprinted domain is orthologous to a region on human chromosome  11p15.5 (163). The gene order, transcriptional orientation, and, to a certain extent, gene size and exon distribution are very well conserved between human and mouse (48; 159). The major 19  difference in this region between human and mouse is their inversion relative to the telomere; in mouse the IC2 region is distal, but in humans, the IC1 subdomain is closer to the telomere. This region in humans has been linked to Silver-Russell Syndrome (SRS) and Beckwith-Wiedemann Syndrome (BWS). The etiology of SRS is heterogeneous and poorly understood; it has been associated with cytogenetic abnormalities of chromosomes 7, 8, 15, 17, and 18, as well as maternal UPD of chromosome 7 and mis-regulation of genes in the 11p15 region, including maternal UPD11 (164). Clinical features of SRS are varied and include intra-uterine and postnatal growth restriction, relative macrocephaly, triangular faces, and skeletal asymmetries (165; 166).   BWS also shows variable clinical presentations and causes (reviewed in refs. 167, 168).  More common clinical findings include pre- and/or post-natal gigantism, macroglossia, abdominal wall defects, and propensity for childhood tumours (169). Molecular causes of BWS fall into three main categories; paternal duplication or UPD of 11p15, loss of imprinting at one of the two ICs, or gene mutations. All of these involve incorrect expression of IGF2 or CDKN1C, the genes responsible for the BWS phenotype (167). Maternal UPD 11p15 results in biallelic CDKN1C, lack of IGF2, and SRS-related growth retardation, whereas paternal UPD results in biallelic IGF2, lack of CDKN1C, and BWS-related overgrowth (170). Loss of maternal methylation of IC2 and subsequent loss of CDKN1C expression is also associated with variable loss of imprinting at IGF2, and accounts for the majority (50-60%) of BWS cases (171; 172). In the mouse, this has been modeled by abolishing maternal germ line expression of the de novo DNA methyltransferases Dnmt3a, Dnmt3b, and/or Dnmt3l. Early in development these deletions mimic BWS gene expression, with loss of Cdkn1c expression, though a total lack of all maternally methylated sequences results in mid-gestation lethality (173; 174). Approximately 20% of BWS cases show paternal UPD of 11p15 with similar loss of CDKN1C (14; 175). In the mouse, paternal UPD of the distal MMU7 imprinted region results in mid-gestational lethality possibly due to gross placental abnormalities (10). This lethality may be due to loss of expression of Ascl2, deletion of which also results in mid-gestational lethality and lack of spongiotrophoblast (83). In humans, ASCL2 is not imprinted, which may explain why UPD of 11p15 results in BWS and not embryonic lethality (176). Maternally inherited CDKN1C mutations are responsible for a further 5% of BWS cases (177) and paternal duplications,  20  translocations or inversions another 3% (178). IC1 defects, including mutations of CTCF binding sites or loss of imprinting at the IC itself, have been found in 5% of BWS patients.   The association of loss of IC2 methylation with biallelic IGF2 expression in addition to  loss of CDKN1C in BWS patients implies a mechanistic link between IC1 and IC2. Studies in the Lefebvre lab have investigated both the independence of IC1 in the absence of IC2 as well as the effect of deletions of the IC1-IC2 interval in mice (135; 179). In both these cases, imprinted expression and DNA methylation is normal, but further investigation of this domain may reveal spreading of imprinting effects between the two ICs.  1.5 Tel7KI transgenic line    The Tel7KI transgenic mouse line was generated as a by-product of a strategy to truncate  MMU7 between the IC1- and IC2-regulated regions (179). The purpose of this experiment was to determine whether IC1 would continue to function normally in the absence of the entirety of the IC2 domain. A construct was designed to take advantage of a previous insertion 2.6 kb distal to Ins2 on MMU7 called I2loxP (179). I2loxP consists of a single loxP site 5’ of a promoter-less neomycin gene, which confers resistance to the drug G418 when a promoter is introduced to drive neo expression (Figure 1.3 A).   The construct designed to truncate MMU7 consisted of 1.6kb of telomeric repeats  (T2AG3), a CAG-EGFP reporter construct, and a Pgk promoter followed by a single loxP site (Figure 1.3 B). When linearized, the construct containing a single loxP site was predicted to recombine with the single loxP site at I2loxP in the presence of Cre recombinase and result in substitution of the telomere-EGFP sequence for the distal ~2.6Mb of MMU7, including the IC2 region and the approximately 18 genes distal to it. This ES cell line was indeed recovered and has been previously published (179).   In addition, a different variant of the line without the deletion of distal sequences was  also recovered (Figure 1.3 C). Screening and confirmation of transgene structure found that the truncation construct had inserted at the I2loxP site, resulting in neomycin expression but without deleting the distal sequences, essentially generating a “pop-in” allele (179). A possible model for the direct insertion of the vector involves multimerization of the originally-linear deletion vector, resulting in a molecule containing multiple loxP sites, from which a circular molecule containing 21  Tel 7q  IC2  B  pCAG-EGFP multimerization  cen  Kcnq1ot1  Th  Cre tel  loxP-neopA  I2loxP  Pgk-loxP  tel  H19  Ins2 Igf2  IC1  Kcnq1 Tssc4 Cd81 Ascl2 Th  Nap1l4 Phlda2 Slc22a18 Cdkn1c  A  +  Ins2  pCAG-EGFP Pgk-loxP  Cre  C loxP  Tel7KI Th  tel  Pgk-loxP-neopA pCAG-EGFP  Ins2  Figure 1.3 Cre-mediated insertion at the Ins2 locus and structure of the Tel7KI allele. (A) Diagram of the imprinted domain on distal Chr7 showing the location of the insertion upstream of Cre-mediated Ins2 (arrowhead), relative to the imprinting centersof(IC). Maternally Figure 1.X: insertion at the Ins2two locus and structure the Tel7KI allele.and (A) Diagram of thegenes imprinted distal Chr7 showing location of the paternally expressed are in domain white andongray, non-imprinted genesthe in black; arrows show insertion upstream of Ins2 (arrowhead), relativeI2loxP to theallele, two imprinting centers (IC). transcriptional orientation. Structure of the targeted with its promoter-less loxPMaternally and paternally expressed genes are in white and gray, non-imprinted genes neo-polyA (loxP-neopA) cassette inserted 2. 6 kb telomeric to Ins2 (below). (B) Proposed in black; arrows showatranscriptional orientation. Structure of the targeted I2loxP allele, mechanism to generate circular intermediate in I2loxP/+ ES cells. The linear targeting vector is with its promoteress loxP-neo-polyA (loxP-neopA) cassette inserted 2. 6 kb telomeric postulated to have multimerized, then a circular intermediate was resolved by Cre from this to Ins2 Proposed mechanism to generate a circular intermediate in of array. This(below). provided(B) a substrate for Cre-mediated insertion at I2loxP. (C) Resulting structure I2loxP/+ ES cells. The linear targeting vector is postulated to have multimerized, then the loxP-flanked Tel7KI allele showing the telomeric repeats (tel), the ubiquitous EGFP reporter a circular intermediate resolved by Cre from this array. This provided a substrate (pCAG-EGFP), and activewas Pgk-loxP-neopA marker. for Cre-mediated insertion at I2loxP. (C) Resulting structure of the loxP-flanked Tel7KI allele showing the telomeric repeats (tel), the ubiquitous EGFP reporter (pCX-EGFP), and active Pgk-loxP-neopA marker.  22  a single loxP site would be readily excised in the presence of Cre (Figure 1.3 B). This circular molecule could then “pop-in” at the loxP site present at I2loxP, resulting in reconstitution of a Pgk-loxP-neo cassette and G418 resistance, but all sequences distal to I2loxP would remain intact. This line was named Tel7KI (official name Ins2tm1lef). Note that Tel7KI is frequently abbreviated KI in this thesis, and that with reference to genotypes, the maternal allele is written first (ie. KI/+ refers to maternal transmission of the Tel7KI allele). 1.6 Thesis objectives   The initial observations of the Tel7KI transgenic mouse line (performed by L. Lefebvre)  suggested that expression of GFP from the transgene occurred in a parent-of-origin-specific manner. The availability of an imprinted fluorescent protein-based reporter in vivo in the mouse would allow real-time visualization and isolation of cells undergoing dynamic changes in imprinted expression or DNA methylation. It has been expressed by others in the field of genomic imprinting that such a reporter, which would allow more detailed analysis of cells undergoing malfunctions in imprinted expression, would be useful to the study of genomic imprinting (180). Based on preliminary observations, the main hypothesis tested in this thesis is that the Tel7KI mouse line is regulated by genomic imprinting. The experiments detailed here will test i) whether this regulation results in consistent parent-of-origin specific expression and epigenetic modification, and ii) whether this line is useful in the analysis of acquisition and loss of genomic imprinting in the mouse. Specific objectives of this thesis are: 1) To characterize the expression pattern of the Tel7KI mouse line when passed through the maternal or paternal germlines. 2) To determine whether this parent-of-origin-specific expression is correlated with acquisition of allele-specific DNA methylation. 3) To investigate specific cell types showing unusual expression of this imprinted transgene, specifically germ cells and trophoblast giant cells, for loss of imprinted gene expression. 4) To determine whether the long-range imprinting signals causing imprinted expression at Tel7KI originate at the IC1 or IC2 imprinting centres. 23  Chapter 2: Materials and methods 2.1 Mice 2.1.1 Mouse strains   The derivation of the Tel7KI allele on Chr7 by a recombinase-mediated insertion at the  Ins2 locus in ES cells, followed by derivation of chimeric mice and subsequent breeding was previously described (179). The line has been maintained in a homozygous state on the 129S1/ SvImJ background for more than eight generations. Inbred mice from strain 129S1/SvImJ were obtained from The Jackson Laboratory (stock number: 002448) and the outbred CD-1 mice were from the UBC Animal Care Centre. The congenic mouse line with distal Chr 7 M. m. castaneus SNPs on the 129S1 background (129S1cCAST7/Lef, hereafter referred to as 129c) was derived in the Lefebvre Lab by R. Oh-McGinnis. All animal experimentation were carried following the guidelines from the Canadian Council on Animal Care (CCAC) under UBC animal care license numbers A03-0289 and A03-0292. 2.1.2 Genotyping   Animals were genotyped at weaning by ear punch lysis and PCR, as described (181).  When post-implantation embryos and placentae were collected, a small sample of yolk sac was retained and used for genotyping as well. Briefly, ear punches or yolk sacs were lysed overnight at 50oC in proteinase K (proK) buffer (50mM KCL, 10mM Tris HCL, 2mM MgCl2, 0.1 mg/ml gelatin, 0.45% NP-40 substitute, and 0.45% Tween 20) with 5mg/ml ProK (Roche). The resulting lysate was used in a PCR reaction as follows: 10 minutes at 95oC (to inactivate the ProK), followed by a pause at 85oC to add the PCR master mix (1x PCR buffer, 2mM MgCl2 200μM dNTPs, 2.5mM each primer, and 0.04U Tsg polymerase) then 35 cycles of 95oC for 30 s, 60oC for 30s, and 72oC for 30s and final analysis by agarose gel electrophoresis. Primer sequences are found in Table 2.1.   Genotyping of the Tel7KI allele was performed with the Δ5’ PCR reaction, which  amplifies from the flanking Ascl2 sequences to the 5’end of the Pgk promoter (179). 24  Homozygous mutant embryos were detected by a positive Δ5’ PCR reaction and absence of a wild-type band (I2wt) for a PCR reaction amplifying across the genomic SpeI site defining the I2loxP insertion site. Castaneus congenic F1 embryos were identified using a SNP in the IN2F1/726R PCR product which was identified by HpaII digest post-PCR. Primer sequences and enzyme analysis for genotyping are given in Tables 2.1 and 2.2. 2.1.3 Embryo dissection   Heterozygous male and/or female TelKI mice were mated to 129S1, CD-1 or to 129c  congenics to generate embryos of the appropriate parental inheritance of the transgene. No difference in phenotype was observed on any strain background, so data from both 129S1 and CD-1 are combined. Wild-type littermates were used as controls. For timed mating, the day of the vaginal plug is E0.5. Pre-implantation embryos were collected from the uterine horns of pregnant females at E3.5 according to published procedures (181). Females were sacrificed at the desired day from E7.5 to E18.5, embryos were scored for phenotype and a sample of yolk sac was taken for genotyping by PCR. Embryos and placentae collected were treated in one of four ways: frozen at -80oC for subsequent RNA and/or DNA extraction, disaggregated for flow cytometry, prepared for further culture, or fixed for immunofluorescence. 2.2 Tissue culture 2.2.1 ES cell line derivation and culture   ES cell lines were derived by L. Lefebvre (KIP cells) or A. Bogutz (KIM cells) from  preimplantation embryos isolated from uterine horns at E3.5 as described (181). Individual blastocysts were plated in gelatin-coated 4-well dishes and allowed to attach. Once an inner cell mass outgrowth was formed, the outgrowth was manually detached from the trophoblast cells adhered to the plate, trypsinized and plated in a new well. The resulting cells were grown for further analysis and experiments. Newly-derived and previously-established ES cell lines were grown in DMEM (Specialty Media) supplemented with 15% Fetal Bovine Serum (Hyclone), 2 mM L-glutamine (Gibco), 0.1 mM 2-mercaptoethanol (Sigma), 0.1 mM MEM non-essential 25  amino acids (Gibco), 1 mM sodium pyruvate (Gibco), 50μg/ml each penicillin and streptomycin (Gibco), and 100μl LIF concentrate (generously provided by H. Ding). ES cells were thawed on mitomycin-C-inactivated primary embryonic fibroblasts (PEF) and passaged onto gelatin-coated plates for FACS experiments, cell collection for DNA and/or RNA, and embryoid body differentiation. 2.2.2 Electroporation and selection of clones   ES cells grown on gelatin for a minimum of 2 passages were trypsinized (0.25% trypsin,  Gibco), collected, and resuspended at approximately 1x107 cells in 0.8 ml electroporation buffer (Specialty Media). Cuvettes containing cells and plasmid DNA were incubated on ice for 20 minutes, then electroporated using a Bio-Rad Gene Pulser Xcell set at 250V and 500uF in a 4mm cuvette. For targeted deletion of IC2, 20 μg targeting vector was used. After 30 minutes on ice, cells were plated on G418 and puro-resistant mitomycin-C-treated PEFs in ES media without selection.   For excision of the Tel7KI construct, +/Tel7KI ES cells were electroporated with 10 μg of  the Cre-expression vector pCX-nlsCre-puro (A. Nagy). After 24 hours in drug-free media, puromycin was added (1.5mg/ml, Sigma) for 48 hours. The colonies were allowed to grow in selection-free media for a further 7 days, after which GFP– clones were picked and expanded on a 96-well plate. 10 lines were established and screened by PCR for deletion of Tel7KI with assistance from T. Ngai. For deletion of IC2 in KIP ES cells, puromycin selection was also begun after 24 hours for a total of 48 hours, after which the selection was changed to G418 (150ug/ml, Sigma) for 14 days. At the end of this time period, 72 colonies were picked and expanded on a 96-well plate, one plate was frozen for a stock, and colonies were screened by PCR. Once likely colonies were identified, the stock plate was thawed and 24 candidates were expanded on a 24well dish, then re-frozen and larger cell pellets collected for more detailed analysis. When a final positive clone was identified, it was thawed and expanded.  26  2.2.3 Embryoid body differentiation   ES cells grown for at least 2 passages in the absence of PEFs were collected and seeded  at low density on untreated bacterial dishes in ES media without LIF. Half the media was changed every other day, and samples of cells were taken for flow cytometry and RNA/DNA extraction every 5 days for a total of 20-30 days. 2.3 Cell separation and flow cytometry   For flow cytometry analysis of embryos and placentae, whole tissues were washed in  PBS and passed through a p200 pipette tip in the presence of trypsin (0.25%, Gibco). Suspensions were pipetted up and down to disaggregate cells after 2 and 5 minutes at 37oC, then the trypsin was inactivated by the addition of FACS buffer (PBS with 2mM EDTA, 2% FCS (Gibco), and 2mg/ml propidium iodide (PI, Gibco)), and finally passed through a 26-gauge needle before analysis. For sorting cells from E10.5 embryos, the procedure was identical, though collagenase/dispase (Roche, 1mg/ml) was used in place of trypsin.   Flow cytometry analysis of ES cells and embryoid bodies was performed by normal  trypsinization (5 minutes at 37oC in 0.25% trypsin (Gibco)) followed by trypsin inactivation by addition of FACS buffer. The cell suspension was pipetted up and down repeatedly with a p200 pipette tip and passed through a cell sieve (Fisher) before flow cytometry analysis. All analyses were performed on a BD LSRII and sorting on a BDFacsARIA, both running BD FACS DIVA. Analysis was performed in FlowJo 8.0, the negative population was gated based on the profile for wildtype (+/+) embryos or ES cells, and PI-positive staining was used to eliminate dead cells. 2.4 Isolation of nuclei from placentae   Nuclei were isolated from placentae and parietal yolk sacks (PYS) either freshly collected  or previously frozen at -20oC. Frozen tissues were thawed to room temperature, after which the procedure was identical. Tissues were placed on a flat surface and minced with a razor blade into a fine paste. The paste was transferred to a 1.5ml tube with 500μl of nuclei extraction buffer (15mM Tris-HCl, 60mM KCl, 15mM NaCl, 5mMMgCl2, 0.5M EGTA pH8.0, 0.3M sucrose), 27  inverted 5 times, then spun at maximum speed (14000 RPM) for 2 minutes. The supernatant was carefully removed then the pellet was resuspended again in 500μl nuclei extraction buffer, pipetted up and down repeatedly, and homogenized with a pellet pestle (Kontes). A small sample was removed and stained with Trypan Blue (0.5% Sigma) then examined for cellular disruption. If blue-staining nuclei were observed, samples were spun down and resuspended in fresh nuclei buffer containing 1mg/ml Hoescht 33342 (Sigma), then passed through a cell sieve (Fisher) before flow cytometry analysis. If further disruption was necessary, the sample was rehomogenized and re-examined with Trypan Blue staining, then processed as described. 2.5 Ectoplacental cone collection and culture   Ectoplacental cones (EPCs) were isolated from E8.5 conceptuses resulting from a cross  between a 129S1/129c female and a +/KI male as follows. Conceptuses were removed from the uterus and placed in PBS containing 10% FCS. Each conceptus was split lengthwise to reveal the embryo and EPC, and the EPC was removed from the decidua with fine forceps and treated in one of two ways. For culture, EPCs were placed in a 4-well dish and cultured as previously described (182) and for immediate RNA and DNA preparation, 5-6 EPCs were pooled and snap frozen on dry ice. Yolk sack samples were taken for genotyping. EPCs were cultured on fibronectin-coated (100μg/ml, Roche) 12 mm glass coverslips (VWR) in DMEM supplemented with 20% FCS, penicillin/streptomycin (Gibco), and 2mM L-glutamine (Gibco) for 5 days before either being collected with trypsin, pooled, and snap-frozen for RNA and DNA extraction, or fixed in 4% PFA at room temperature for 30 minutes for immunofluorescence (IF). Wild-type (129/129;+/+) EPCs were discarded, uninformative mutant (129/129;+/KI) EPCs were fixed for staining, and all informative (129/129c;+/+ and 129/129c;+/KI) EPCs were collected for DNA and RNA. 2.6 Tissue preparation and sectioning   Embryos and placentae for sectioning were rinsed in PBS, then fixed in 4%  paraformaldehyde (PFA, Sigma) for 10-30 minutes at room temperature. Next, tissues were cryopreserved overnight in 10% then 30% sucrose at 4oC, rinsed in PBS, and embedded in OCT 28  (Tissue-Tek) on liquid nitrogen, then stored at –80oC until sectioned. Frozen blocks were sectioned on a Leica CM3050 cryostat and 14um sections placed on Superfrost slides (Fisher). Slides were stored at -20oC until needed. 2.7 Immunofluorescence   For detection of epitopes in embryo and placenta, slides were warmed to room  temperature, rinsed twice with PBS, and permeabilized with 0.2% Triton-X (Fisher). Sections were then blocked with 4% normal serum (Chemicon), and incubated with primary antibodies diluted in PBS overnight at 4oC (dilutions in Table 2.3). The following morning slides were washed again with PBS, incubated with labeled secondary antibody for 1h at room temperature, washed in water and counterstained with DAPI (2 μg/ml, Sigma) for 5 minutes. An exception to this is the SSEA-1 monoclonal hybridoma antibody. For this antibody, the permeabilization step was eliminated, sections were blocked with the M.O.M blocking reagent (Vector Labs) in addition to 4% serum and secondary antibody staining was reduced to 10 minutes. All antibody types, suppliers, and concentrations are given in Table 2.3. Coverslips were mounted on the stained sections with vectashield (Vector Labs) and slides were imaged as described below.   Whole EPCs cultured on coverslips were treated similarly with a few modifications.  Coverslips were blocked in 3% BSA (Sigma) with 0.2% Triton-X for 30 minutes, then incubated with chicken anti-GFP (1:500, AbCam) and rabbit anti-placental lactogen I (Pl-I, or PRL3D1,1:250, Chemicon) for 2 hours at room temperature. After washing with PBS-T and reblocking for 10 minutes, they were incubated with secondary antibodies, washed, counterstained, mounted and imaged as above. 2.8 RNA analysis 2.8.1 RNA preparation and reverse transcription   For RNA purification, cells, embryos, or placentae were snap-frozen on dry ice and stored  at –80oC. RNA was extracted from snap-frozen tissues by a single-step isolation using Trizol (Invitrogen Life Technologies, CA) according to manufacturer’s directions. Approximately 1μg 29  of RNA was reverse transcribed with random pentadecamers (N15) as per Invitrogen SSII protocol. 2.8.2 RT-PCR   Q-RT-PCR on 1μl of cDNA was performed as follows on an Bio-Rad Opticon 2: 95oC 2  minutes followed by 35 cycles of 95oC 30 sec, 58oC 30 sec, 72oC 30 sec, 85oC 1sec with plate read. Ct values of triplicate samples were averaged and used to calculate relative amounts of transcript, normalized to Gapdh. EGFP transcript from the Tel7KI transgene was detected with a forward primer (BAE1F) in the short chicken β-actin exon 1 of the CAG promoter, and a reverse primer (BAE1R) in exon 2 (third exon of β-globin gene) of the same construct, which is fused to GFP. For allele-specific analysis, RT-PCR products of H19, Igf2, Cdkn1c, and Tssc4 were digested overnight and run on a 4% NuSieve gel (Cambrex). The primer names, sequences, and expected RFLP for each allele are given in Table 2.1. 2.9 DNA analysis 2.9.1 DNA preparation   DNA was extracted either from the remaining Trizol fraction after removal of the  aqueous phase containing RNA as per manufacturer’s instructions, or from tissues digested with proteinase K as described (183). 2.9.2 Sodium bisulfite modification, PCR and COBRA   Approximately 1μg of genomic DNA was treated with sodium bisulfite as previously  described (134; 179; 184). 1 μl of the treated DNA was used in the first round of nested PCR with conditions as follows: for the CAG promoter, GFP ORF, and H19 DMR: 40 cycles of 95oC for 30s, 52oC for 30s and 72oC for 30s, for KvDMR1: 35 cycles of 94oC for 1m, 52oC for 1m30s, an 72oC for 1m. Of the first round product, 1 μl was used in a 50-μl reaction for a second round PCR with an annealing temperature of 55oC for 35 cycles. Bisulfite PCR for the CAG promoter 30  (36 CpGs) was a semi-nested reaction, using primers BABF6 and BABR5d in round 1 (299 bp), followed by round 2 with BABF6 and BABR4c (253 bp). For COBRA analysis, 5μl of the bisulfite PCR product for the β-actin promoter was digested overnight with BstBI at 65oC, then run on a 4% NuSieve gel. BstBI cuts only the methylated CpG at 38bp of the 257bp PCR product, resulting in a methylated band at 219bp and an unmethylated band at 257bp. The band intensities were measured with ImageJ Gel analysis tool by plotting intensities and integrating the areas under the resulting peaks. Bisulfite sequencing primers for the GFP open-reading-frame were a gift from M. Lorincz and analyze 37 CpGs in a 430-bp fragment. H19 DMR (16 CpGs in 473 bp) and KvDMR1 (31 CpGs in 335 bp) analysis has been previously described (134; 179; 184). For COBRA analysis of KvDMR1 (IC2), the 335bp fragment is cut with BccI which cuts at 45 and 165bp, as well as RsaI which cuts only the methylated CpG at 185. Fragments are digested overnight and run on a 2% agarose gel. For each sequencing assay, four independent PCR reactions were performed on the sodium bisulfite-modified genomic DNA sample and three to four cloned strands were sequenced per PCR. All primer sequences are in Table 2.1. 2.10 Deletion of IC2 in KIP ES cells 2.10.1 Targeting vector construction   The IC2KO targeting vector was constructed from the 64lox (pKvDMRKO) and 11kb-  loxppA-neo (pKOT11pA) plasmids generously given by M. Higgins (119; 155). It was necessary to modify these targeting vectors to change the existing G418 resistance gene to a puromycin resistance gene, and to add a DTA negative-selectable marker. This was accomplished by cloning the small arm of homology from pKvDMRKO cut with XhoI and EcoRI next to the DTA negative selectable marker in the ploxPDTA plasmid cut with SalI and EcoRI, resulting in pIC2sDTA. Next, this plasmid was cut with NheI and XhoI to release the small arm-DTA fragment, which was triple-ligated with a KpnI/XbaI PGK-puro-pA fragment flanked by loxP sites from pfpuro.1, and the backbone plus the long arm of homology, from pKOT11pA cut with XhoI and KpnI. The resulting vector contains a PGK-puro-pA flanked by loxP sites between the two arms of homology, with a PGK-DTA-pA marker outside the small arm, and is called  31  pIC2KOPD. This vector was linearized with NotI before electroporation into KIP ES cells. See Figure 5.5 for final vector structure. 2.10.2 Screening of IC2KO clones   Electroporation, selection, picking and expansion of clones is described above. After 1  passage in the absence of PEFs, one 96-well plate was washed with PBS and incubated with 10μl ProK buffer, as described for genotyping above at 50oC overnight. Resulting lysates were used as above in section 2.1.2 in long-range PCR for screening of IC2 deletion. The only modification to the above protocol is the length of PCR steps: 35 cycles of 95oC for 30 s, 60oC for 1 m, and 72oC for 2m30s. Each clone was screened with three PCR reactions; the I2wt reaction as a positive control, an internal transgene-specific PCR with primers IC2sF2 and IC2ScR1, and an insertionspecific PCR with primers IC2ScF1 and IC2ScR1. The transgene-specific PCR would amplify any clones which had integrated the IC2KOPD vector, regardless of insertion site, while the insertion-specific PCR would amplify only the clones which had correctly recombined the targeting vector and deleted IC2. Four clones were positive for the insertion-specific PCR, and these were analyzed by COBRA analysis to determine which parental IC2 had been deleted. See Figure 5.5 for primer locations and Table 2.1 for primer sequences. 2.11 Microscopy and image analysis   Photographs at low magnification of whole mount embryos and placentae were taken on  a Leica MS5 dissecting microscope equipped with a Q-imaging Micropublisher 3.3 RTC colour camera and the fluorescent light source MAA-03 (BLS Ltd.). For tissue sections or cultured tissues and cells, high magnification analysis was performed with a Leica DMI6000B inverted fluorescent microscope, and images were captured with a Q-imaging Retiga 4000R monochrome camera controlled by Openlab (Improvision). Images captured in black and white were colorized and overlapped in Photoshop 8.0. Photos showing sections of whole embryos or placentae were stitched together from 6-8 overlapping photos in Photoshop. Overlay analysis on placental sections was performed with “Colocalization Highlighter” plugin for ImageJ (created by Pierre  32  Bourdoncle, Institut Jacques Monod, Service Imagerie, Paris, bourdoncle@ijm.jussieu.fr) with default values. Table 2.1 Primer sequences for genotyping, bisulfite, RT-Q-PCR, and screening Purpose  Sequence  Name  Reference  Genotyping (I2wt)  AGCACAGTCCCCTGTGTTCT  I2wt F  This study  GTCTTCAACCCCATGTGACC  I2wt R  This study  CCAAAGAACGGAGCCGGTTG  PGK4  (179)  TGAATGGGAAATGTGGTCCTTGG  M2G  (179)  GGAGAGGTGYGGYGGTAGTTAATTAGAG  BABF6  This study  TCATTAAACCAAACRCTAATTACAACCC  BABR4c  This study  AAACCCCTCAAAACTTTCACRCAACCACAA  BABR5d  This study  ATTATTTTCTAGATTGTTATGGTGAGTAAGGG  1870Fbis  (185)  GAGGAGTTGTTTATYGGGGTGGTGTTT  1903Fbis  (185)  TAACTATTATAATTATACTCCAACTTATACC  2303Rbis  (185)  GAGTATTTAGGAGGTATAAGAATT  BMsp2t1  (102)  ATCAAAAACTAACATAAACCCCT  BHha1t3  (102)  TGTAAGGAGATTATGTTTTATTTTTGGA  BMsp2t2.2  (179)  AACCTCATAAAACCCATAACTATA  BHha1t4.2  (179)  GGTTATAAAGTTTAGGGGTTTTTAGATTTG  Kcnq1ot1 OF  (184)  AAAACTTTTCTATTCAACTTAATTCCCAAC  Kcnq1ot1 OR  (184)  GGTTTTAAGATTATTTTTGTTTTTGTAAGT  Kcnq1ot1 IF  (184)  AATTCTCCTAAATATAATTTTTTTCTCAAC  Kcnq1ot1 IR  (184)  GCTCTGACTGACCGCGTTACT  BAE1F  This study  GGACACGCTGAACTTGTGG  BAE1R  This study  ACCACAGTCCATGCCATCAC  G3PF  (179)  TCCACCACCCTGTTGCTGTA  G3PR  (179)  GAGCTTCCTGGCTGTTTTTG  IC2sF2  This study  AAAGAACTGGGGGTTCCACT  IC2scF1  This study  ACCCGGTAGAATTTCGAGGT  IC2scR1  This study  Genotyping (Δ5’)  Bisufite (CAG)  Bisulfite (GFP ORF)  Bisulfite (H19 DMR)  Bisulfite (KvDMR1)  RT-Q-PCR (GFP)  RT-Q-PCR (G3PDH)  IC2 KO screening  33  Table 2.2: Primer sequences used for allele-specific analysis Name  Sequence  H19rt1 CCTCAAGATGAAAGAAATGGT  Size  Enzyme  Cast  129  Ref  641  SmaI/  244, 44,  244, 44,  (186)  Cac81  222, 130  352  Tsp5091  170,  183,  multiple  multiple  H19rt2 AACACTTTATGATGGAACTGC Igf2F CCATCAATCTGTGACCTCCTCTTG  200  Igf2R GGGTGTCAATTGGGTTGTTT P57S GCCAATGCGAACGACTTC P574  364  Taqα1  206  HaeIII  257, 58, 49 306, 58  (179)  (187)  TACACCTTGGGACCAGCGTACTCC  Tssc4F GACAGCCAGAGGTGGTGGTA  55, 151 55, 63, 88  This study  Tssc4R TAGCAGGTGAGGGGTTAGAGTGATAG Lqt2  ATTGAGATCCCAGGGCTGAGG  465  SspI, StuI 252, 150,  (179)  96, 198,  (179)  63  Lqt18 GGCACACGGTATGAGAAAAGATTG IN2F1 CGGTGTCCGTACACATTGCAC  40,263  511  HpaII  198, 313  217  726R ACAGCAGGGTTCCCACACTGG  Table 2.3 Antibody details Antibody name  Species  Type  Supplier  Concentration  Catalog number  anti-GFP  chicken  primary polyclonal  AbCam  1:200  ab13970-100  DSHB*  1:5  MC-480  Chemicon  1:100  AB1289  Molecular  1:250  A-21203  1:250  A-11039  1:250  A-21207  concentrate MC-480 (anti  mouse  SSEA-1) anti-PL-II  primary monoclonal hybridoma supernatant  rabbit  primary polyclonal concentrate  anti-mouse  donkey  secondary  Alexa594 anti-chicken  Probes goat  secondary  Alexa488 anti-rabbit Alexa594  Molecular Probes  donkey  secondary  Molecular Probes  * Developmental Studies Hybridoma Bank, University of Iowa.  34  Chapter 3: Tel7KI expression and DNA methylation in the embryo1 3.1 Introduction   The study of the dynamics of genomic imprinting and imprinted gene expression during  development has been complicated by the difficulty in localizing and isolating tissues undergoing these changes in epigenetic state in vivo. In particular, this difficulty arises because these events frequently occur in very small cell populations. Transgenes which become expressed in an imprinted manner but are not related to known imprinted genes have the potential to be useful in these studies, as modifiable elements for genetic screens, for example. To be a candidate for this type of study, a transgene would preferably show parent-of-origin-specific expression in a wide range of tissues and developmental stages, show some allele-specific epigenetic modifications, and ideally be shown to be regulated by endogenous imprinting signals. Another useful characteristic would be the ability to use the transgene as a marker for cell localization and isolation. Unfortunately, most known imprinted transgenes are not ideal in some manner. In some cases imprinted expression of these transgenes is due to the transgene itself, and is not reflective of endogenous signals (53). Tissue-specificity of some transgenes means that imprinted expression can be examined in few cells, and that the mechanisms regulating these transgenes may include position effects (53; 67). In addition, generally these transgenes are transcriptional units which do not confer the ability to easily visualize or isolate the cells in which they are expressed. As existing transgenes do not fulfill all the desired characteristics for this reporter, the development of such an imprinted transgene would be extremely useful to the study of genomic imprinting.   The construct used to create both the Chr7 truncation described elsewhere and Tel7KI is  based on the ubiquitous CAG enhancer-promoter, which has been used in the past to create many transgenic reporter lines, including some driving GFP expression (179; 188; 189). Since the broadly and strongly expressed GFP reporter construct of the Tel7KI allele was previously shown to be controlled by the epigenetic phenomena of X chromosome inactivation in the D4/ 1  Parts and/or versions of Figures 3.1, 3.3, 3.4, 3.5, 3.6, and 3.12 have been published in Jones, M and Lefebvre, L. (2009) An imprinted GFP insertion reveals long-range epigenetic regulation in embryonic lineages. Developmental Biology. 336: 42-52. 35  XEGFP transgenic line (190), it was hypothesized that the same cassette might be regulated by long-range imprinting signals in the context of its insertion site within the IC1- and IC2regulated imprinted domain in the Tel7KI line. This hypothesis was tested by examination of embryos carrying paternally- or maternally-inherited Tel7KI transgenes for expression of the GFP gene. Further experiments attempted to determine more features of the expression and regulation of Tel7KI on MMU7 and to examine specific tissues which show unusual or interesting imprinted expression. It is predicted that the Tel7KI mouse line will prove to be a useful reporter of genomic imprinting on MMU7 in the embryo. 3.2 Results 3.2.1 GFP from Tel7KI is expressed in a parent-of-origin-specific manner in the embryo    Initially, Tel7KI was examined by whole mount fluorescence analysis of postimplantation  stage embryos upon either maternal or paternal transmission of the allele. Beginning at E7.5, the earliest stage examined, the CAG-EGFP reporter was expressed in a parent-of-origin-specific manner. GFP fluorescence from Tel7KI was observed only in the embryos inheriting Tel7KI from the maternal germline (Figure 3.1). The widespread GFP activity of the maternally inherited allele has been consistently observed at all stages examined, from E7.5 to E18.5 (Figure 3.1 and data not shown), though little GFP expression is observed in transgenic KI/+ neonates or adult tissues (data not shown).   Upon paternal transmission, the GFP reporter of Tel7KI is silenced in most embryonic  tissues in the majority of transgenic embryos analyzed at all stages from E7.5 to E18.5 (Figure 3.1 and data not shown). Exceptions to this will be discussed in section 3.2.4 of the current chapter. In extraembryonic membranes, approximately 24% of paternal transmission conceptuses analyzed (55/225), show significant GFP expression in the yolk sac. This expression takes the form of a branching pattern which appears to be exclusive to the extraembryonic mesoderm layer of the yolk sac (Figure 3.2 A). Interestingly, this stochastic yolk sac patch phenotype has also been observed occasionally upon maternal transmission of Tel7KI, as GFP is otherwise not expressed in maternal transmission yolk sac. More than half of the paternal transmission  36  E8.5 KI/+  E10.5 +/KI  KI/+  E12.5 KI/+  +/KI  E14.5 +/KI  KI/+  +/KI  Figure 3.1 Imprinted expression of Tel7KI in post-implantation embryo. E8.5 E10.5, E12.5, and E14.5 embryos carrying a maternal (KI/+) or a paternal (+/KI) Tel7KI FIgure 3.1: Imprinted Tel7KI in E10.5,GFP were visualized underexpression bright fieldof(above) or post-implantation GFP fluorescence embryo. (below). E8.5 Occasional E12.5, and E14.5 embryos carrying a maternal (KI/+) or a paternal (+/KI) Tel7KI were fluorescence is observed in hearts of +/KI embryos (white arrowheads), and GFP expression visualized under of bright (above) orseen GFPthrough fluorescence GFP from the gonads +/KI field embryos can be the body(bleow). wall afterOccasional E13.5 (white arrow). fluorescence is observed in hearts of +/KI embryos (white arrowheads), and GFP Scale bar: 1 mm. expression from the gonads of +/KI embryos can be seen through the body wall after E13.5 (white arrow). Scale bar: 1 mm. 37  A  B  +)NORMAL  +)  +)  +),/)  C  +)  Figure Loss of GFP silencing in Tel7KI embryos. Figure 3.2 3.2: Loss of GFP silencing in Tel7KI embryos. A) Yolk sac patch phenotype (A) Yolk sac patch phenotype observed in both +/KI and more KI/+ embryos, though more B) frequently observed in both +/KI and KI/+ embryos, though frequently in +/KI. +/KI in +/KI. (B) E13.5 +/KI embryo with loss of silencing manifesting as GFP patches in the embryo embryo with loss of silencing manifesting as GFP patches in the embryo (right, (right, observed in 28/225 embryos).C)(C) Amnionfrom fromsame same conceptus in A), showing bright observed in 28/225 embryos). Allanois conceptusasas in A), showing foci of GFP expression in the amnion. bright foci of GFP expression in the allanois.  38  conceptuses showing these yolk sac patches (28/55) also show mosaic GFP expression in the embryo proper in a pattern which differs between embryos, and a subset also show bright GFP foci in the amnion (Figure 3.2 B and C). This GFP reactivation in the embryo has been difficult to assess upon maternal transmission, as these embryos already express widespread GFP.   Importantly, the parent-of-origin specific expression of GFP from Tel7KI is reversible.  Female mice inheriting a silent allele from their fathers give rise to embryos which show high levels of GFP expression, and in the converse, male mice with an active maternal allele give rise to progeny which show no GFP expression (data not shown). In addition, further generations in a silent or active state have no effect on the ability to silence or activate the allele, such that an allele that has been passed down through the silent paternal line for multiple generations can be fully reactivated by a single passage through a female germ line. This indicates that the parentof-origin specific expression of Tel7KI is appropriately reset at each generation as is observed at endogenous imprinted loci.   To determine more precisely the differences in GFP expression between maternal and  paternal transmission of Tel7KI in the embryo proper, E9.5 embryos of all four possible genotypes from a +/KI x +/KI intercross (wild-type (+/+), paternal hemizygous (+/KI), maternal hemizygous (KI/+), or homozygous mutant (KI/KI)) were disaggregated and flow cytometry analysis was performed on single-cell suspensions. While no cells expressing GFP above wildtype backgrounds were detected in +/KI embryos, approximately 40% of cells analyzed from KI/ + and KI/KI embryos express GFP at high levels (Figure 3.3). In addition, GFP expression was analyzed by quantitative RT-PCR (Q-RT-PCR) for two individual embryos of each genotype at E10.5. These results confirmed the absence of GFP expression in wild-type or paternal transmission embryos, and detected a high but variable level of expression in maternal transmission and homozygous embryos (Figure 3.4). 3.2.2 GFP silencing from Tel7KI in the embryo is associated with increased DNA methylation at the CAG promoter   DNA methylation at promoter regions has been associated with silencing at imprinted  genes, genes on the inactive X chromosome as well as during programmed or pathological silencing of gene expression in mammals. Since the CAG-EGFP reporter is CpG rich and was 39  100  % of max  B  +/+  80 60  +/KI  KI/KI KI/+ +/KI +/+  40 20 0  KI/KI  KI/+  C 100 80  % cells  A  60 40 20  0  102 103 104 105 GFP fluorescence  0  +/+ +/KI KI/+ KI/KI genotype  Figure 3.3 GFP expression from Tel7KI by flow cytometry. (A) E9.5 wild-type (+/+), paternal transmission (+/KI), maternal transmission (KI/+), and homozygous (KI/KI) embryos show imprinted expression of GFP viewed in bright field (above) Figure 3.3:GFP (A) E9.5 wild-type (+/+), Scale paternal (+/KI),cytometry maternaldata transmisand under fluorescence (below). bar:transmission 1 mm. (B) Flow of single sion (KI/+), and homozygous (KI/KI) embryos show imprinted expression of GFP disaggregated E9.5 mouse embryos. GFP expression profiles are from single +/+ (red), +/KI viewedKI/+ in bright field and under GFP (C) fluorescence (below). Scale bar: 1 mm. (blue), (green), or(above) KI/KI (orange) embryos. Summary of flow cytometry analysis over (B) Flow cytometry data of single disaggregated E9.5 mouse embryos. GFP expresmultiple litters showing the percentage of cells expressing GFP in E9.5 embryos. Bars indicate sion profiles are from single (red), +/KI from (blue), KI/+ (green), and KI/KI (orange) mean and standard deviation of +/+ results obtained multiple individual embryos: +/+ n=3, +/ embryos. (C) Summary of flow cytometry analysis over multiple litters showing the KI n=9, KI/ + n=14, and KI/KI, n=4. percentage of cells expressing GFP in E9.5 embryos. Bars indicate mean and standard deviation of results obtained from multiple individual embryos: +/+ n=3, +/KI n=9, KI/ + n=14, and KI/KI, n=4.  40  -2 relative levels (10 )  1.4  1.05 0.7  0.35 0 +/+  +/KI  KI/+  KI/KI  genotype  Figure 3.4: 3.4 Q-RT-PCR GFP expression from Tel7KI. Figure Q-RT-PCRanalysis analysisofof GFP expression from Tel7KI on RNA purified from RNA purified from whole E12.5 was a analyzed for transcription of GFP from the of CAG whole E12.5 embryos. Each barembryos represents single embryo, and shows expression promoter by quantitative RT-PCR. Each bar represents a single embryo, and shows expression EGFP transcript relative to Gapdh (10-2 scale). Error bars indicate standard deviations of EGFP transcript relative to Gapdh (10-2 scale). Error bars indicate standard deviations for for technical triplicates. technical triplicates.  41  previously shown to be responsive to position-related epigenetic mechanisms, it is possible that DNA methylation might be implicated in the regulation of its expression (190; 191). A sodium bisulfite sequencing assay was devised to examine 36 CpG dinucleotides from the 5’ portion of the reporter, including part of the chicken β-actin promoter, the transcription start site, exon 1 and a small portion of intron 1 (Figure 3.5). The entire CAG enhancer-promoter cassette is highly CpG-rich.   Two different developmental stages (E10.5 and E14.5) were analyzed for DNA  methylation status at the promoter of Tel7KI. Both of these stages show high levels of GFP expression from the maternal allele and essentially no GFP in embryos carrying a paternally inherited allele in most cells, as described above (Figure 3.1). Since these embryos are hemizygous for the GFP reporter, a single parental allele is analyzed in each amplification and sequencing reaction. At E10.5 there is a difference in the overall level of DNA methylation at the CAG promoter region between maternal (16%) and paternal (69%) transmission of the allele (Figure 3.5). This methylation difference is maintained at E14.5, where the paternal allele is methylated at 87% but the maternal allele at only 32%. During this period both +/KI and KI/+ embryos show an increase in the methylation level, and the expressed maternal allele is not fully unmethylated despite the high expression levels.   In order to determine whether the methylation of Tel7KI constitutes a germline imprint,  DNA methylation was examined in mature sperm collected from a 1-year old transgenic +/KI male. No methylation was detected in the epididymal sperm sample, meaning that paternal DNA methylation at Tel7KI is not acquired directly from the male germline (Figure 3.5, bottom left panel). Finally, CpG methylation at the GFP open reading frame (ORF) was also examined, within the transcribed portion of the reporter, at E14.5 to see if the observed differential methylation extended past the promoter region. The GFP ORF is also CpG-rich (51 CpGs in 732bp), but no differences in DNA methylation levels were observed between maternal and paternal transmission of Tel7KI as both alleles were highly methylated (89% for +/KI and 94% for KI/+; Figure 3.5).  42  CMV enhancer  b-actin promoter  EGFP  ex 1  E10.5  +/KI KI/+ KI/+  E14.5  32%  94%  87%  89%  +/KI mature sperm  16%  +/KI  //  69%  0%  Figure 3.5: DNA methylation at Tel7KI in the embryo.Structure of the CAG-EGFP reporter of Tel7KI (above) showing the CMV enhancer, the chicken !-actin promoter, and a 5’ intron from the chicken !-globin gene, fused to EGFP. Sequencing Figure 3.5 DNA methylation at Tel7KI in the embryo. of sodium bisulfite-modified genomic DNA (below) purified from paternal (+/KI) and maternal (KI/+) transmission embryos Structure ofsperm the CAG-EGFP reporter of Tel7KI showing CMV enhancer, theabsent chicken and mature from a +/KI adult male. Circles represent(above) methylated (filled) orthe unmethylated (open) CpGs; circles indicate ambiguous sites. β-actin promoter, and a 5’ intron from the rabbit β-globin gene, fused to EGFP. Sequencing of  sodium bisulfite-modified genomic DNA (below) purified from paternal (+/KI) and maternal (KI/ +) transmission embryos and mature sperm from a +/KI adult male. Circles represent methylated (filled) or unmethylated (open) CpGs; absent circles indicate ambiguous sites. Percent methylation is indicated for each set and is calculated as total methylated CpGs over total assayed CpGs.  43  3.2.3 Maternal transmission of Tel7KI results in an expression pattern consistent embryo-toembryo with highest levels of GFP expression in the heart and CNS   Initial observations of whole-mount and FACS-analyzed maternal Tel7KI embryos  indicated that some tissues did not express GFP (Figures 3.1 and 3.3). To examine the tissuespecificity of GFP expression following maternal transmission of Tel7KI, immunofluorescence was performed on frozen sections of E12.5 embryos using an antibody against GFP. Two independent embryos from two different E12.5 litters were sectioned, stained, and visualized (only two representative littermates are shown in Figure 3.6). Production of GFP protein was observed most strongly in the heart and central nervous system (CNS), both exhibiting consistently high levels of expression in all embryos examined. In other tissues, GFP expression was lower but is still detectable above background. This broad zone of lower expression also exhibits more cell-to-cell variability (Figure 3.6). Most tissues contained at least some cells expressing GFP, but no tissues outside the heart and CNS showed consistent high-level GFP expression in the sagittal sections analyzed. GFP expression was particularly low in the liver.   Finally there remained the possibility that the small amount of DNA methylation  observed on the active maternal allele in whole embryos was related to the variable expression observed in KI/+ embryos. More specifically, it was asked whether the GFP-negative cells found in maternal transmission embryos (Figures 3.3 and 3.6) are the cells showing DNA methylation of Tel7KI (Figure 3.5). To test this hypothesis, KI/+ E10.5 embryos were disaggregated and pooled, then the cells were sorted based on GFP fluorescence levels and the genomic DNA purified from GFP+ and GFP– fractions analyzed for methylation at Tel7KI, using the promoter sodium bisulfite assay. Most molecules analyzed were fully unmethylated, showing an absence of DNA methylation at most maternal Tel7KI alleles, irrespective of the GFP expression levels (Figure 3.7). A few methylated sites were detected but these were in the pool of expressing cells (3%, Figure 3.7). Thus, there appears to be no relationship between GFP expression and DNA methylation on the active maternal allele.  44  A  $!0) '&0  $!0)  '&0  B  $!0) '&0  $!0)  '&0  Figure 3.6 GFP expression in 14-μm frozen sections of E12.5 maternal transmission Tel7KI embryos. Two KI/+ (A and B) were stained with a polyclonal anti-GFP transmission antibody and Figure 3.6:embryos GFP expression in 14-µm frozen sections of E12.5 maternal counterstained with DAPI to(Aexamine and reproducibility of GFPanti-GFP expressionantibody from the embryos. Two embryos and B) extent were stained with a polyclonal maternal allele of Tel7KI. Scale bars: 1 mm. and counterstained with DAPI to examine extent and reproducibility of GFP expression from the maternal allele of Tel7KI. Scale bars: 1 mm.  45  GFP+  3%  GFP-  0%  Figure 3.7 DNA methylation on the maternal allele of Tel7KI. E10.5 embryos were disaggregated, sorted into GFP+ and GFP– cell populations, and DNA Figure 3.7: DNA methylation on the maternal allele of Tel7KI in E10.5 embryos methylation at the CAG promoter of Tel7KI was analyzed. Percent methylation is indicated at disaggregated and sorted into GFP+ and GFP- cell populations. No methylation is right and is calculated as total methylated CpGs over total assayed CpGs. No methylation is observed in the GFP- cells, and only a small amount (3%) in GFP+ cells. observed in the GFP– cells, and only a small amount (3%) in GFP+ cells.  46  3.2.4 GFP expression in paternal transmission embryos is observed stochastically in the heart and brain and consistently in PGCs   The paternal transmission of Tel7KI results in nearly complete silencing of the GFP  reporter in the embryo, with a few exceptions. In some paternal transmission embryos, particularly at later stages (E14.5-E18.5), localized foci of GFP expressing cells can be observed in the heart and, much less frequently and in a more variable pattern, in the brain (Figure 3.8). These patterns overlap with the highest levels of expression observed from the active maternal allele, potentially indicating escape from epigenetic silencing in these highly-expressing tissues.   The most consistent tissue to show loss of silencing of GFP in +/KI embryos is the  developing germ line. After observations of GFP-expressing germ cells in later stage embryos (see arrow, Figure 3.1), +/KI embryos at E9.5, E10.5, E11.5, and E12.5 were collected and sectioned transversally and sagittally, then doubly stained by immunofluorescence for GFP and the germ cell marker SSEA-1 (also known as CD-15) (192). In transverse sections, germ cells can be tracked as they migrate from the hindgut (E9.5) through the dorsal mesentery (E10.5) to the genital ridges (E11.5) where they colonize the developing gonad (E12.5, Figure 3.9). At E9.5 and E10.5, very few primordial germ cells (PGCs) show GFP expression (Figure 3.9 A and B). However, once these cells have entered the genital ridge at E11.5, and later in developing gonads at E12.5, significant amounts of GFP expression can be observed in 14μm transverse cryosections (Figure 3.9 C and D). Several sagittal sections at each stage were stained in order to show higher numbers of germ cells than can be observed in transverse sections (one of each stage shown in Figure 3.10) and the percentage of GFP+ germ cells as a fraction of the total number of germ cells was calculated. 5.2% (4/77) and 8.2% (8/97) of the germ cells detected at E9.5 and E10.5, respectively, show GFP expression (Figure 3.11). At E11.5, the percentage increases to 22% (83/384), and at E12.5 to 66% (467/703) potentially indicating a stage-specific re-activation of GFP from the silent paternal allele in the germ line (Figure 3.11).  47  A  B  E11.5  E12.5  E13.5  Figure 3.8 Loss of imprinting in +/KI embryos in specific tissues. (A) Head of E14.5 +/KI embryo shown under bright field (right) and GFP fluorescence (left) with patches of GFP expressioninin+/KI the brain (white as observed in very (1%) of Figure 3.8: Loss of imprinting embryos in arrowheads), specific tissues. (A) In very fewfew (1%) Hearts dissected from E11.5, E12.5, and E13.5 +/KI also show spots ofcases. cases(B)GFP expression can be observed in patches in embryos the brain, shown here of at GFP expression of variable sizes and locations. This is observed in 30% of embryos at all stages from E14.5. (B) Much more frequently (30% of embryos) spots of GFP expression are E9.5 to E18.5. observed in the heart, at all stages from E9.5 to E18.5 (only E 11.5, E12.5, and E13.5 are shown).  48  A  DAPI SSEA-1 GFP  E9.5  DAPI SSEA-1 GFP  SSEA-1 GFP  DAPI SSEA-1 GFP  DAPI SSEA-1 GFP  SSEA-1 GFP  N DAPI SSEA-1 GFP  DAPI SSEA-1 GFP  SSEA-1 GFP  DAPI SSEA-1 GFP  DAPI SSEA-1 GFP  SSEA-1 GFP  A  N  B  E10.5  A  C  E11.5  A  D  E12.5  A  N  Figure in the the germ in cross-sections cross-sections. of +/KI Figure 3.9 3.9:GFP GFPexpression expressionfrom from Tel7KI Tel7KI in germ line line in 14μm cryosections of +/KI embryos at (A) E9.5, (B) E10.5,stained (C) E11.5, and (D) E12.5 were embryos at E9.5, E10.5, E11.5, and E12.5. 14um cryosections with antibodies stained with (green) antibodies against (red), GFP counterstained (green) and the SSEA-1 (red), against GFP and SSEA-1 withgerm DAPI cell (blue).marker A= dorsal aorta, N= notochord. Panel on right magnified in panels on left. counterstained with DAPI (blue). A= dorsal aorta, N= notochord. Panel on left magnified in panels on right.  49  A  E9.5  DAPI SSEA-1 GFP  DAPI SSEA-1 GFP  SSEA-1 GFP  E10.5  DAPI SSEA-1 GFP  DAPI SSEA-1 GFP  SSEA-1 GFP  C  E11.5  DAPI SSEA-1 GFP  DAPI SSEA-1 GFP  SSEA-1 GFP  D  E12.5  DAPI SSEA-1 GFP  DAPI SSEA-1 GFP  SSEA-1 GFP  A  B  A  Figure 3.10: GFP expression expression from in in thethe germ lineline in saggital sections of +/KI Figure 3.10 GFP fromTel7KI Tel7KI germ in sagittal sections. embryos at A) E9.5, B) E10.5, C)E11.5, and D) E12.5. 14um cryosections stained 14μm cryosections of +/KI embryos at (A) E9.5, (B) E10.5, (C) E11.5, and with (D) E12.5 were antibodies against GFP (green) and SSEA-1 (red), counterstained with DAPI (blue). A=DAPI (blue). stained with antibodies against GFP (green) and SSEA-1 (red), counterstained with dorsal aorta. Panel on right magnified in panels on left. Arrowheads in E9.5 and E10.5 A= dorsal aorta. Panel on left magnified in panels on right. Arrowheads in E9.5 and E10.5 panels panels indicate GFP+ germ cells. indicate GFP+ germ cells. 50  70  % GFP+ germ cells  60 50 40 30 20 10 0  E9.5  E10.5  E11.5  E12.5  Embryonic stage  Figure 3.11: 3.11 Summary of GFP expression in germ cells embryos. of +/KI embryos. Figure GFP expression in germ cells of +/KI Two embryos at each Two embryos at each stage were from E9.5 to E12.5 were sectioned stained for a germ cell marker stage from E9.5 to E12.5 sectioned and stained for aand germ cell marker (SSEA-1) (SSEA-1) and GFP. Cells were counted and scored for GFP expression, and percent GFPand GFP. Cells were counted and scored for GFP expression, and percent + expressing germ cells (GFP SSEA1+) over total over germ total cells germ (SSEA1+) calculated. Total GFP-expressing germ cellsand (GFP+ and SSEA1+) cellswas (SSEA1+) was germ cell numbers at E9.5 n= 77, E10.5 n= 97, E11.5 n= 384, E12.5 n= 703. calculated. Total germ cell numbers at E9.5 n= 77, E10.5 n= 97, E11.5 n= 384, E12.5 n= 703.  51  3.3 Discussion    The present study reports the first imprinted fluorescent transgene inserted near a known  imprinted domain which appears to be regulated by long-range signals from that domain. Tel7KI is based on sequences not known to result in imprinted expression at other genomic loci, but which have been shown to respond to external epigenetic signals in the context of Xchromosome inactivation (190; 191). From its inserted position between two imprinted domains on MMU7, Tel7KI expresses GFP reliably in the embryo from only the maternal chromosome, and thus it is likely a downstream target of endogenous imprinting mechanisms operating on MMU7.   Parental-allele-specific gene expression is one of the characteristics of endogenous  imprinted genes observed at Tel7KI. In addition, Tel7KI also shows correlation between that allele-specific expression and allele-specific DNA methylation. This transgene generally follows the pattern of a secondary DMR; it does not carry a primary imprint inherited from the paternal germ line, but differential DNA methylation is found later in development, and the active maternal allele does not become hyper-methylated in those tissues where it is not expressed. One major difference between the Tel7KI DMR and other endogenous DMRs is that while the maternal allele is active and expresses GFP, it is not completely unmethylated. This is not usual for imprinted genes, but can be found in reports of other imprinted transgenes (60). The DNA methylation observed in the ORF of EGFP is consistent with previous reports of high levels of DNA methylation in the body of genes, regardless of their expression status (193-195).   The tissue-specificity of the promoter used by Tel7KI is unexpected. For many years the  CAG promoter construct has been used to create “ubiquitously-expressed” transgenes. Here it is shown that in the case of the Tel7KI insertion this promoter is not in fact ubiquitous in the embryo, but follows a very precise tissue-specific pattern. It appears that Tel7KI is highly active in most cells of the CNS and heart, expressed in a cell-to-cell variable manner in most tissues of the embryo, and expressed at very low levels in some tissues, such as the liver. This indicates that the CAG promoter is susceptible to position effect regulation and should be noted in the future when “ubiquitous” transgenes using this system are reported. Close examination of examples where the CAG-EGFP transgene has been used reveal little in the way of tissuespecific experiments on embryos, even in those lines where high levels of GFP expression are 52  reported (189). There are examples of studies reporting tissue-specific effects from these constructs, but generally in-depth analysis was not performed (196). Thus it is difficult to say whether the pattern of expression observed from Tel7KI is common in transgenes using the CAG enhancer/promoter or whether it is an effect from its inserted location.   One puzzling aspect of the phenotype exhibited by Tel7KI is the patchy expression  observed in some embryos upon paternal transmission. This reactivation is observed most frequently in yolk sac, which is often associated with patches of expression in the embryo, but is also observed in the heart and brain. Expression in these latter tissues may be due to high promoter usage, as evidenced by high levels of GFP in these tissues upon maternal transmission, but the expression in yolk sac and embryo patches is mysterious. The GFP from Tel7KI is not normally expressed in yolk sac upon maternal transmission of Tel7KI, indicating that this is a tissue-specific silencing as opposed to an imprinted silencing. However, the link between yolk sac patches upon paternal transmission and reactivation of the GFP in patches of the corresponding embryo suggests that there may be an epigenetic link, since silencing in the embryo is epigenetically regulated. It is possible that reactivation of GFP in the embryo is demonstrative of loss of imprinting (LOI) at this locus, though neither the mechanism nor the extent of this loss is known. Expression in yolk sac, however, may be due to either of these mechanisms; LOI in a subset of cells or promoter over-expression in a tissue where it is normally silent. It would be interesting to determine whether the GFP reactivation is indicative of changes in epigenetic status of the cells; if, for example, these cells show loss of DNA methylation (LOM) or imprinted expression at other imprinted genes. This can be examined by disaggregating and sorting GFP-expressing from non-expressing cells and determining whether LOI or LOM has occurred.   In the context of our current understanding of imprinting on distal Chr 7, the regulation of  Tel7KI suggests that the effects of existing imprinting centres can reach loci located further than previous appreciated. Based on the ontogeny of allele-specific methylation described here at the Tel7KI allele, it is possible that either the H19 DMR (IC1) or KvDMR1 (IC2) could be responsible for the imprinted expression observed at Tel7KI, thus two models for its imprinted behavior are described here (Figure 3.12).   In the IC1 domain, the DNA methylation patterns at Igf2 are reminiscent of what is  observed at Tel7KI, though the parental expression of these two genes is opposite. Both Tel7KI 53  and Igf2 are paternally methylated but Igf2 is also paternally expressed, the hypothesis being that the paternal methylation on this gene represses a silencer element (24). The methylation acquired on the paternal allele of Tel7KI could mimic the situation at Igf2, but in the case of Tel7KI the promoter DNA methylation would result in silencing of GFP transcription (Figure 3.12B). Not only is the timing of acquisition of DNA methylation similar for Igf2 and Tel7KI (142), but there is also a parallel with regard to imprinted transcription, the Igf2 gene being biallelically expressed in blastocysts, as is observed for Tel7KI (See Chapter 5) (197). For Igf2, this is likely to reflect a basal rather than activated biallelic transcription. Whether or not a similar basal transcription is responsible for the observed biallelic expression of Tel7KI in blastocyst remains to be determined, though preliminary evidence suggests that GFP expression from Tel7KI is much lower in undifferentiated than differentiated ES cells (data not shown). A potential flaw in this hypothesis is the lack of DNA methylation on Tel7KI in sperm. Igf2 DMRs 1 and 2 become DNA methylated in mature sperm, and this DNA methylation is lost during the wave of preimplantation demethylation and re-established after implantation (94; 140). If Tel7KI is simply mimicking methylation at Igf2, it would be expected to see DNA methylation in sperm as well.   The current model of how the maternal DMRs of Igf2 remain unmethylated involves  chromatin looping, CTCF binding, and epigenetically-mediated contact between distant sites (Figure 1.1) (142; 153). The Tel7KI allele is found more than 20 kb away from the Igf2 CpG-rich region involved in this looping. Furthermore, the gene found in between Igf2 and Tel7KI, Ins2, is imprinted only in embryonic yolk sac endoderm, does not have a DMR, and has not been implicated in this looping model. However, circular chromosome conformation capture experiments designed to identify genomic regions physically associated with the CTCF complex at IC1 have uncovered several interacting regions on distal Chr 7, including 3 sites immediately distal of the Tel7KI insertion site and two other sites proximal of Th, located ~300 kb telomeric of Ins2 (198). Our model therefore raises the possibility that the allele-specific regulation mediated by IC1 extends distally beyond Ins2, perhaps as far as the Th locus, which is consistent with the recent finding that Th is preferentially expressed from the maternal allele in the placenta (199) (A. Bogutz, unpublished data). A prediction from this model would be that absence of CTCF binding from the maternal IC1 should lead to acquisition of DNA methylation at the maternal Tel7KI and silencing of the GFP. 54  A  IC1  IC2  B  H19  Tel7KI  Th  Ascl2  Cd81  Tssc4  Kcnq1  Cdkn1c  Slc22a18  Phlda2  Nap1l4  Mat  IC2  IC1  C  Igf2  Kcnq1ot1  Ins2  Pat  IC2  IC1  Igf2  Kcnq1ot1  Ins2  Pat  Figure 3.12 Models for long-range regulation of Tel7KI. 3.12: Models for long-range regulation of Tel7KI. (A) The maternal (A)Figure The maternal chromosome is marked by a DNA methylation imprint atchromosome IC2. The Tel7KI is marked by a DNA methylation imprint at IC2. The Tel7KI promoter is unmethylated promoter is unmethylated and active. (B) In the first model, the germline imprint on the paternal active.not (B)only In the the to germline on the IC1and spreads to first Igf2model, but also Tel7KI.imprint Whereas thispaternal inhibitsIC1 thespreads functionnot of Igf2 Igf2 but also to Tel7KI. Whereas this inhibits the function of Igf2 silencers, the only to silencers, the promoter methylation leads to silencing of the paternal Tel7KI allele. (C) Another promoter methylation leads to silencingonof the allele. (C)byAnother modelof the model implicates long-range silencing the paternal paternalTel7KI chromosome formation implicates long-range silencing on the paternal chromosome formation Kcnq1ot1 non-coding RNA, which bidirectionally spreads repressiveby histone marksofinthe the IC2 Kcnq1ot1 non-coding RNA, which bidirectionally spreads repressive histone marks in cluster (dotted lines) and silencing DNA methylation at Cdkn1c and Tel7KI (solid lines). the IC2 cluster (dotted lines) and silencing DNA methylation at Cdkn1c and Tel7KI (solid lines).  55    The post-fertilization acquisition of DNA methylation on the silent paternal Tel7KI allele  is also reminiscent of that observed at the IC2-regulated maternally expressed Cdkn1c, the only imprinted gene regulated by IC2 which contains its own DMR (31). The pattern of Cdkn1c methylation is similar to that observed at Tel7KI, with paternal methylation acquired between E6.5 and E8.5, though the GFP from Tel7KI becomes monoallelically expressed between E4.5 and E7.5 (see Chapter 5), while Cdkn1c is already preferentially maternally expressed at E4.5 (31; 37; 115; 184; 200). Interestingly, other genes regulated by IC2 are biallelically expressed in blastocysts and acquire their monoallelic expression during post-implantation development (34).   Unlike the situation at IC1 where long-range effects involve an epigenetically regulated  insulator, imprinting in the IC2 sub-domain is dependent on the cis-spreading of repressive chromatin via the action of a large non-coding RNA, Kcnq1ot1 (155; 172; 184). In the second model, Tel7KI is regulated by IC2 through the action of Kcnq1ot1 which would spread a further 300 kb towards the proximal IC1 sub-domain (Figure 3.11C). A main difference between Tel7KI and most of the endogenous genes of the IC2 cluster is that Tel7KI contains a CpG island capable of undergoing differential DNA methylation (31). Thus, it is possible that IC2 can act on Tel7KI in the embryo through the presence of sites capable of acquiring DNA methylation. According to this second model, a prediction would be that deletion of IC2 in cis of Tel7KI or truncation of the Kcnq1ot1 ncRNA should prevent silencing of the paternally inherited allele, as indeed observed for endogenous IC2-regulated genes (119; 155; 158). See Chapter 5 for further investigation of this model.   Regardless of whether IC1 or IC2 is responsible for the imprinting of Tel7KI, the  discovery of a fluorescent reporter regulated by genomic imprinting in the embryo has the potential to be an extremely useful tool. Both maternal (active) and paternal (silent) alleles of Tel7KI could be used as reporters for mis-regulation of imprinting at least on MMU7. For example, mutagenesis screens using male Tel7KI animals would look for resulting embryos that show GFP expression where normally none is seen. The genes identified could include modifiers of epigenetic regulation like DNA methyltransferases (or hypothetical demethylases), histone modification enzymes, methylated-DNA binding proteins, and others.   Perhaps the most exciting phenotype of Tel7KI for use as a potential tool is the timing of  increased expression of GFP in the germ line from the paternal allele. Before further experiments are undertaken, it will be very important to examine GFP expression in germ cells of the same 56  stages upon maternal transmission, to ensure that epigenetic and not stage-specific processes are responsible for the increasing GFP expression in paternal-transmission embryonic germ cells during development. E11.5, when the GFP expression begins to increase in germ cells of +/KI embryos, is an important stage for the developing germ line. At this time, DNA methylation at DMRs from imprinted genes are becoming erased in preparation for germline reprogramming, a process which is complete by E12.5 (128). It is hypothesized that the entry of germ cells into the ridge triggers this reprogramming process; if true, the GFP reactivation from Tel7KI could very well prove to be a reporter for this process. It will be important to determine what exactly the GFP reactivation is a reporter of; this can be accomplished by isolating genital ridges from E10.5, E11.5, and E12.5 embryos and disaggregating and sorting the germ cells into GFP+ and GFP– fractions. These fractions can then be analyzed for DNA methylation at the Tel7KI promoter, as well as endogenous imprinted genes and genomic repeats. If, as hypothesized, Tel7KI is functioning as a reporter for imprint erasure, the GFP+ and GFP– pools of germ cells should differ in some characteristic of imprinted genes or epigenetic regulation. For example, perhaps DNA demethylation has begun in those cells which are GFP+ and not in those that are GFP–. If global DNA methylation levels are similar, differences may still be observed at only a single gene or set of genes. Regardless of what is found to be correlated with GFP-reactivation of the germ line, hopefully it will give insight into this rather mysterious process.   In conclusion, the Tel7KI transgene exhibits three of the common characteristics of  imprinted genes i) allele-specific gene expression ii) allele-specific DNA methylation, and iii) appropriate intergenerational resetting. It is as yet unknown where the signals causing this imprinted expression originate, but given the position of the transgene between two imprinted centres and the fact that other insertions of similar transgenes do not become imprinted, it appears that Tel7KI is being regulated by long-range signals from one of these centres. It is hoped that this transgenic mouse line will prove to be a useful tool to study the maintenance of imprinted silencing in mice. In order to further characterize imprinted expression of Tel7KI, the following chapter will investigate Tel7KI expression and DNA methylation in the placenta.  57  Chapter 4: Characterization of Tel7KI expression in the placenta and implications of endoreduplication on imprinting in trophoblast giant cells (TGCs)2 4.1 Introduction    As Tel7KI is regulated by genomic imprinting and expressed in a parent-of-origin-  specific manner in the embryo, it was hypothesized that extraembryonic tissues would also exhibit imprinting. All endogenous imprinted genes which are imprinted in the embryo are also imprinted in the placenta, assuming they are expressed in the placenta at all. In the MMU7 imprinted domain, for example, H19, Igf2, Ascl2, Tssc4, Cd81, Kcnq1, Kcnq1ot1, Phlda2 and Osbpl5 are all imprinted in the placenta (3; 16; 31; 48; 134; 155; 159; 184; 201-203). Of these, Cd81, Tssc4, and Osbpl5 are expressed biallelically in the embryo (48; 159; 201). Ascl2 has also been reported to be biallelically expressed in the embryo, but its levels are extremely low (34). Ins2, the remaining imprinted gene in this domain, is imprinted only in yolk sac mesoderm, and biallelic in embryo (144). Proper imprinted expression of these genes is critical for placental development and thus it is important to examine imprinted expression of Tel7KI in this tissue to gather more data for the long-range regulation of this transgene.   Tel7KI was compared to the X-linked GFP transgene D4, which contains the same  regulatory elements driving GFP expression and has been previously hypothesized to show loss of epigenetic regulation in trophoblast giant cells (TGCs) of the placenta (191). When paternally inherited in females, the transgene was inactivated in almost all trophoblast-derived cells, as would be expected for imprinted paternal X-chromosome inactivation (190; 204). The exception to this was TGCs, which showed high levels of GFP expression upon paternal inheritance of D4 (190). TGCs arise from the ectoplacental cone (EPC) and are responsible for invasion of the uterus. They produce large amounts of hormones and secretions, and are known to be highly polyploid. DNA contents of 1000N have been reported, though approximately 600N is much 2  Parts and/or versions of Figures 4.1, 4.2, 4.3, 4.4, 4.7, 4.8, 4.9, 4.10, and 4.11 have been published in Jones, M and Lefebvre, L. (2009) An imprinted GFP insertion reveals long-range epigenetic regulation in embryonic lineages. Developmental Biology. 336: 42-52. 58  more commonly observed (205; 206). The observation of what appeared to be loss of imprinted X-inactivation in TGCs led to the hypothesis that these cells show an epigenetic defect resulting in reactivation of the silent X chromosome. The cause of this defect could very possibly be the extremely high ploidy these cells reach; if they continue to replicate their DNA without dividing, some factor important for epigenetic stability may become limiting. The current chapter will examine the epigenetic stability of these cells as a part of the general characterization of expression from and DNA methylation of the Tel7KI transgene in mouse placentae. 4.2 Results 4.2.1 GFP expression of Tel7KI in the placenta is observed upon both maternal and paternal transmission   Unlike the pattern observed in the embryo (see Chapter 3), in the placenta Tel7KI shows  no obvious differences in GFP expression when inherited from either parent. Both paternal and maternal transmission placentae show punctuate GFP expression over the placenta at all stages examined from E8.5 to E17.5 (Figure 4.1 and data not shown). On the fetal surface, this punctuate pattern is brighter around the rim of the placenta, and GFP expression appears fainter in the middle where blood exchange is occurring. However, when blood is removed from the placenta, the level of GFP expression becomes uniform across the surface of the placenta, likely due to the removal of the blood masking GFP fluorescence (data not shown). The level of expression is much more variable in the placenta than is observed in the embryo, however observations have noted a tendency for slightly brighter GFP expression upon paternal than maternal transmission in the placenta, particularly at later stages (Figure 4.1 and data not shown).   To examine more closely the cell-to-cell expression pattern in the placenta, maternal and  paternal transmission E12.5 placentae were sectioned and stained with an anti-GFP antibody and counterstained with DAPI nuclear stain. Expression patterns of GFP upon maternal or paternal transmission were similar; a punctuate pattern of expression throughout the labyrinth and spongiotrophoblast (Figure 4.2 A and B). High levels of GFP expression were also observed in cells with large nuclei at the fetal-maternal junction (marked in Figure 4.2 G and H with a dotted  59  E11.5  E10.5 KI/+  +/KI  KI/+  +/KI  E12.5 KI/+  +/KI  Figure 4.1 GFP expression in whole placentae of hemizygous Tel7KI conceptuses. FigureE11.5, 4.1: GFP expression in transmission whole E10.5, and E12.5 maternal (KI/+) and paternal transmission (+/KI) placentae placentae E10.5, E11.5, E12.5 This expression pattern is observed at all stages are shown infrom bright field and GFPand fluorescence. Tel7KI embryos. Maternal transmission until E18.5. (KI/+, left) and paternal transmission (+/KI, right) are shown in bright field and GFP fluorescence. This expression pattern is observed at all stages until E18.5.  60  KI/+  +/KI  A  GFP B  C  DAPI D  E  GFP DAPI  G  F  H  4.2 Placental of GFP from Tel7KI upon(KI/+, maternal left) or paternal FIgure 4.2:Figure Placental expressionexpression of GFP from Tel7KI upon maternal left) (KI/+, or paternal right) transmission. (+/KI, right)(+/KI, transmission. 14um cryosections of E12.5 placentae were stained with an 14μm cryosections E12.5 were stained with an GFP (A and B) and antibody against GFP (A andofB) and placentae counterstained with DAPI (C antibody and D). against Six individual counterstained with DAPI (C and D). the Six images individual photos were taken photos were taken and stitched together to form in A-F. Scale bar: 1cm.and (E stitched and F) together to images A-F. Scale bar:high 1cm.levels (E and Overlay of DAPI and giant GFP signals showing Overlay of form DAPIthe and GFP in signals showing of F) GFP expression in the cell highH) levels of GFP expression cellshowing layer. (Gthe andlabyrinth/giant H) Magnification boxes in E and layer. (G and Magnification of boxesininthe E giant and F, cell of layer F, showingline) the labyrinth/giant cell layer boundary (dashed line) and as well as trophoblast boundary (dashed as well as trophoblast giant cells (arrows) glycogen cells giant cells (arrowheads). Scaleand bar: 100um.cells (arrowheads). Scale bar: 100μm. (arrows) glycogen 61  line), predicted to be trophoblast giant cells (TGCs) (Figure 4.2 G and H, white arrows). In addition, cells showing GFP expression were seen beyond the giant cell layer, which are likely trophoblast glycogen cells invading the decidua (Figure 4.2 G and H, white arrowheads). These cells are not likely to be maternal decidual cells, since they are observed upon paternal transmission, and the cross used to generate these embryos involved a wild-type female. 4.2.2 DNA methylation at Tel7KI in the placenta is moderate on both parental alleles   To examine the molecular basis for this deviation of imprinted GFP expression from the  embryonic pattern, DNA methylation at Tel7KI was examined by sodium bisulfite sequencing of the CAG promoter in these placentae. The objective was to determine whether the biallelic expression of Tel7KI in the placenta is correlated with lack of DNA methylation on both parental alleles. At E14.5, while paternal DNA is methylated at a much higher level than maternal in the embryo (Figure 3.5), these differences are not observed in the placenta (Figure 4.3). Both maternal and paternal transmission placentae are only moderately methylated (38% paternal, 46% maternal). Interestingly, the increased expression of GFP observed upon paternal versus maternal transmission in the placenta is associated here with slightly decreased DNA methylation at the promoter. 4.2.3 Tissue-specific analysis of Tel7KI and comparison with the X-linked GFP D4 allele shows similar GFP expression in TGCs   It was important to examine more specifically the expression pattern of GFP from Tel7KI  in the extraembryonic mesoderm (ExM). This tissue is derived from the epiblast, as are embryonic tissues, such that if Tel7KI is imprinted in all epiblast derivatives, GFP expression should be visible in ExM only upon maternal transmission, as it is in the embryo. As a control, GFP expression in Tel7KI placentae was compared to expression from the X-linked GFP D4 transgene, which shows imprinted X-inactivation in trophoblast tissues and random Xinactivation in embryonic tissues and ExM (190). Thus, in the placenta, GFP expression from a paternally-inherited D4 allele should be visible solely in the ExM. 14μm cryosections of  62  CMV enhancer  ß-actin promoter  ex 1  //  KI/+ placenta  49%  +/KI placenta  38%  Figure 4.3 DNA methylation at Tel7KI in E10.5 placenta. Methylation was analyzed by sodium bisulfite sequencing of DNA from whole E10.5 placentae with the maternal decidua removed. Each circle represents a single CpG; open circles represent unmethylated CpGs and filled circles represent methylated CpGs. Each line represents a single DNA strand, and the total percentatmethylation is notedplacenta. to the right, calculated as total methylated Figure 4.3: DNA methylation Tel7KI in E10.5 Methylation was analyzed CpGs over total assayedsequencing CpGs. by sodium bisulfite of DNA from whole E10.5 placentae with the maternal decidua removed. Each circle represents a single CpG; open circles represent unmethylated CpGs and filled circles represent methylated CpGs. Each line represents a single DNA strand, and the total precent methylation is noted to the right. Both maternal (K/+) and paternal (+/K) transmission show moderate levels of DNA methylation at Tel7KI.  63  placentae carrying Tel7KI or the X-linked GFP were stained for GFP and CD-34, a marker of the ExM lineage in the placenta (207). CD-34 stained heavily in mesodermal cells of the labyrinth, and little co-localization was observed between CD-34 and GFP in Tel7KI, indicating that Tel7KI is not highly expressed in extraembryonic mesoderm, irrespective of its parental origin (Figure 4.4). Therefore, analysis of parent-specific expression in the ExM cannot be performed by immunofluoresence. In contrast, a placenta from a female embryo carrying a paternallyderived D4 shows high co-localization between these two markers, as expected for random Xinactivation in this epiblast-derived tissue (Figure 4.4).   An interesting commonality of both these transgenes is that they show high levels of GFP  expression in trophoblast giant cells (TGCs) (Figure 4.4, arrowheads). In Tel7KI this GFP expression is seen at all stages, however, in D4, GFP expression in TGCs is much brighter at E13.5 than at E12.5 (data not shown). As previously mentioned, this reactivation has been hypothesized to be indicative of X-reactivation in TGC. The observation of high levels of GFP expression in Tel7KI upon paternal transmission leads to the possibility of general epigenetic instability in these highly-polyploid cells. 4.2.4 Culture of paternal transmission Tel7KI EPCs gives rise to TGCs which show high levels of GFP expression   In order to analyze gene expression and DNA methylation from TGCs, an attempt was  made to isolate cells from whole placentae and sort them by nuclear size and GFP expression status. The working hypothesis was that the larger a TGC nucleus grew, the more endoreduplication cycles it had undergone, and the more likely it would be that it would suffer some sort of epigenetic instability resulting in reactivation of the silent paternal GFP. Previous experiments had shown that giant cells did not appear to be recovered in a disaggregation of a whole placenta, since no cells larger than 8N were observed with Hoescht staining upon flow analysis (data not shown and J. Cross, personal communication). To solve the problem of cellular size, nuclei were instead isolated from E10.5 placentae and parietal yolk sacs (PYS), then analyzed by flow cytometry. Nuclei resulting from this isolation did show the characteristic large size of TGCs, and peaks of 2N, 4N, 8N, 16N, and 32N were visible upon Hoescht staining on the flow cytometer (Figure 4.5). Interestingly, differences in the profile of ploidy recovered were 64  KI/+  XXgfp  +/KI  A  GFP  B  C  D  CD-34  E  F  G  GFP CD-34  H  I  J  GFP CD-34 Overlap  K  L  Figure 4.4 GFP co-localization with CD-34, a marker of extraembryonic mesoderm, in Tel7KI and X-linked GFP D4 placentae. Figure 4.4: GFP colocalization with CD-34, a marker of extraembryonic mesoderm, in 14um cryosections of E12.5 placentae were stained with an antibody against GFP (A-C, green) Tel7KI X-linked GFPOverlay D4 placentae.14um cryosections of E12.5 placentae and and CD-34 (D-F, red). (G-I) shows co-localization between the two markers.were Scale bars: stained an magnification antibody against GFPin(A-C, and CD-34 (D-F, red). 1mm.with (J-K) of boxes G-I, green) co-localization between GFP andOverlay CD-34 (G-I) is shown in shows colocalization between the two markers.Scale bars: (J-K) magnification of white. Also indicated are GFP-positive trophoblast giant cells1mm. (arrowheads). Scale bar: 500μm. boxes in G-I, colocalization between GFP and CD-34 is shown in white. Also indicated are GFP-positive Trophoblast Giant Cells. Scale bar: 500um.  65  Placenta 250  4n  2n  150 100 16n  50  4n  200  OFNUCLEI  OFNUCLEI  2n  8n  200  +/KI  Parietal yolk sac  150 100  8n 16n  50  32n  32n 0 10  0  10  1  2  10 10 Hoescht  3  10  0 100  4  2n  4n 250  150 100 16n 50 0  10  1  2  10 10 Hoescht  3  104  4n  150 100  8n 16n  50  32n  32n  0 10  102 103 Hoescht  200  OFNUCLEI  OFNUCLEI  200  +/+  8n  2n  101  10  4  0  100  101  102 103 Hoescht  104  Figure 4.5 Ploidy of nuclei isolated from E10.5 placentae and parietal yolk sacs of paternal transmission (+/KI) or wild-type (+/+) embryos. Figure 4.5:form Ploidy nuclei isolated from E10.5 placentae and parietallarger yolk sacs 2n nuclei the ofmajority, with declining percentages of progressively sizes, ofto a paternal transmission (+/KI) or wildtype (+/+) embryos. 2n nuclei form the majority, with maximum of 32n before the nuclei become too large to fit in the cytometer. declining percentages of progressively larger sizes, to a maximum of 32n before the nuclei become too large to fit in the cytometer.  66  noticed between placentae and parietal yolk sacs. The 8n peak was significantly reduced in the PYS, with a concomitant increase in the 16n and 32n peaks. This likely is due to the difference in TGC types present in these two tissues; the PYS contains primary TGCs which are the first to arise and are thus more mature, and the TGCs present in the placenta are secondary TGCs which arise later in development (208). Unfortunately, it was not possible to obtain information about the GFP status of any of these nuclei, since the GFP from Tel7KI is cytoplasmic. An attempt was made to isolate nuclear RNA from these samples, which could have been used to determine the GFP expression status. Unfortunately, however, the sorting was not performed in an RNAse-free environment, and no RNA was ever recovered. In addition, it was difficult to obtain enough nuclei to isolate sufficient DNA for sodium bisulfite analysis of Tel7KI and other imprinting centers. Thus, it was decided that this approach was not the ideal method for enriching a population of TGCs of high ploidy, and an in vitro approach was established.   The in vitro approach for enrichment of TGCs is based on the fact that culturing  ectoplacental cones (EPCs) from early post-implantation embryos results in >90% differentiation into TGCs (209). EPCs from E8.5 paternal transmission Tel7KI embryos carrying Mus musculus castaneus polymorphisms on their maternal MMU7 and wild-type littermates were collected and cultured. Upon culture, extremely large cells appeared at the edges of the attached EPC from 24 hours onward (Figure 4.6), and by 120 hours of culture, very large cells had accumulated and the cells in the center of the culture where the EPC had originally attached had begun to die. Thus, 120 hours, or 5 days of culture was set as the end point of this study. In addition, many of the large cells observed in +/KI EPCs showed GFP expression (Figure 4.6B).   Staining of cultured EPCs with antibodies against GFP and placental lactogen I (Pl-I), a  TGC marker, showed that most of the cells derived from the EPC were in fact TGCs (Figure 4.7). The criteria used were giant nuclei greater than 2x the size of a fibroblast nucleus, and positive staining for Pl-I. The GFP staining also showed that most of the cells staining positive for GFP were also TGCs, though some diploid cells remaining in the center of the EPC were also GFP-positive. In this case, however, no correlation was observed between increased nuclear size and GFP expression, likely indicating no relationship between Tel7KI loss of silencing and increasing TGC ploidy.  67  A 60 hrs  48 hrs  *  76 hrs  * *  B  +/+  *  +/KI  *  Figure Figure 4.6 Trophoblast giant giant cell growth from cultured ectoplacental (EPC). 4.6: Trophoblast cell growth from cultured Ectoplacentalcones Cones (EPC). (A) (A) EPC culture results in growth of cells with large nuclei outwards from the original point of EPC culture results in growth of cells with nuclei outwards from the original point attachment (marked with *). Cultures were stained with Hoescht and imaged at 48, at 60,48, and 76 of attachment (marked with *). Cultures were stained with Hoescht and imaged hours post-dissection. Note that at 60 hours, culture did not stain well, cells are 60, and 76 hours post-dissection. Note this that particular at 60 hours, this particular culture did not well, faintly visible in the center.(B) Scale bars: 100um. After 48 hours faintly stain visible in cells the are center. Scale bars: 100μm. Wild-type (+/+)(B)and paternal Tel7KI of culture, large nuclei for are48observed in wildtype (+/+) and by paternal Tel7KI transmission (+/KI)very EPCs cultured hours show very large nuclei Hoescht staining. transmission EPCs,only but in GFP is observed in Tel7KI tissues. GFP fluorescence is (+/KI) also visible thefluorescence +/KI cells (bottom). Boxonly in middle panel magnified Box in middle panel magnified in right panel. in right panel.  68  DAPI  EGFP  PRL3D1  DAPI EGFP PRL3D1  merge  Figure 4.7 Expression of giant cell markers in cultured EPC. Trophoblast giant cells (TGCs) grown from +/KI ectoplacental cones stained with antibodies Figure 4.7: Trophoblast cellsmarker grown PRL3D1 from +/KI(red), Ectoplacental cones stained with(blue). against GFP (green) andgiant the TGC and counterstained with DAPI antibodies againsinGFP (green) and Pl-I (red), and counterstained DAPI (blue). Overlay shown bottom right panel. Most PRL3D1-positive cells alsowith stain positive for GFP, Overlay shown in bottom right panel. and contain large nuclei, though some GFP-positive, PRL3D1-negative cells with smaller nuclei are visible (white arrowheads).  69  4.2.5 TGCs do not show epigenetic defects at endogenous loci   Though there appeared to be no connection between ploidy and loss of silencing of GFP  from Tel7KI, it remained possible that TGCs have an inherent defect in maintaining epigenetic stability. To assess this, the allelic expression status of four imprinted genes expressed in TGCs was examined with castaneus polymorphisms on MMU7; H19, Igf2, Cdkn1c, and Tssc4. H19 and Igf2 are regulated by IC1, and Cdkn1c and Tssc4 are regulated by IC2. In addition, Tssc4 has been shown in the past to be biallelically expressed in TGCs grown from trophoblast stem (TS) cells, though it is monoallelic in uncultured EPCs (34). This analysis showed normal allelic expression of H19, Igf2 and Cdkn1c in 5-day TGCs (Figure 4.8), all three genes were expressed in a monoallelic fashion with the correct parent of origin. Tssc4 was indeed expressed biallelically, as shown previously.   Next sodium bisulfite sequencing was performed on the DMRs at IC1 and IC2 (Figure  4.9). M.m. castaneus polymorphisms allow the determination of the parent of origin of each strand examined. Methylation patterns at both IC1 and IC2 are normal, with the maternal allele unmethylated and the paternal allele methylated at IC1, and vice versa at IC2. Unusually, at IC2, very few maternal strands were recovered. In order to determine that there was no bias in the endoreduplication process resulting in more paternal strands being replicated, PCR on the genomic DNA followed by an allele-specific restriction digest at both Kcnq1ot1 (the transcript associated with IC2) and Ascl2 was performed. Both genes showed equal amounts of each parental strand, indicating no preferential replication of the DNA (Figure 4.10). 4.2.6 DNA methylation at Tel7KI is established in the placenta after E8.5   Finally, to determine the timing of DNA methylation acquisition at Tel7KI in  extraembryonic tissues, methylation on the paternal allele was examined in both undifferentiated EPCs and 5-day cultured TGCs. No methylation was observed in E8.5 EPCs indicating that Tel7KI is not as highly methylated in extraembryonic tissues in early development as it is in the embryo (Figure 4.11 - compare to Figure 3.5, E9.5 embryo data). Some (18%) CpGs were methylated in TGCs, which implies that DNA methylation acquisition can occur after E8.5 in trophoblast lineages. 70  C/+ TGC  +/C embryo  +/C plcenta  C/+ embryo  C/+ placenta  C/+ TGC  +/C embryo  C/+ embryo H19  Tssc4 Igf2  Cdkn1c  Figure 4.8 Allelic gene expression of imprinted genes in +/KI TGC. Figure from 4.8: TGCs Allelicgrown gene in expresison imprintedbygenes in TGC grown paternal cDNA culture wasofamplified PCR and products werefrom digested with an transmission (+/KI) EPC. cDNA from TGCs was 129 amplified by PCRstrains. and products were appropriate restriction enzyme polymorphic between and castaneus 
digested with an apropriate restriction enzyme polymorphic between 129 and castaneus strains.  71  pat  G G G G G G G  A A A A A A A  80%  mat  A A A A A A  G G G G G G  11%  IC1  mat  A A A A A A A  97%  pat  T T T T T T T T  0.03%  IC2 29  Figure 4.9 Methylation of IC1 and IC2 on MMU7 in +/KI TGC grown from cultured EPC. Figure 4.9: Methylation of IC1 and IC2 on MMU7 in +/KI TGC grown from cultured DNA methylation was analyzed by sequencing of sodium bisulfite-modified genomic DNA. Each EPC. represents DNA methylation was analyzed by represent sequencing of sodiumCpGs bisulfite-modified circle a single CpG; open circles unmethylated and filled circles genomic DNA. Each circle represents a single CpG; open circles represent represent methylated CpGs. Polymorphisms by which parental strands were identified are noted. unmethylated CpGsa and circles methylated CpGs. Polymorphisms Each line represents singlefilled DNA strand,represent and the total percent methylation is noted to theby right, which parental strands were identified are noted. Each line represents a single DNA calculated as total methylated CpGs over total assayed CpGs. strand, and the total percent methylation is noted to the right.  72  cast/129 F1 TGC  cast/129 F1  cast  129  Kcnq1ot1  Ascl2  Figure Allelic analysis analysisoftorelative determine relative contributions of after parental DNA after Figure 4.10: 4.10 Allelic parental DNA contributions endoreduplication endoreduplication of TGC. Genomic DNA from pure 129, pure castaneus, or in TGC. 129/castanues F1 embryos F1 TGCs were amplifiedand by 129/castaneus PCR and Genomic DNA from pure 129,and pure129/castaneus castaneus, or 129/castaneus F1 embryos digested with amplified a polymorphic enzyme. No bias is observed in the relative F1 TGCs were by PCRrestriction and digested with a polymorphic restriction enzyme. No bias is allelic amplification of gDNA in TGCs. of gDNA in TGCs. observed in the relative allelic amplification  73  +/KI 0d EPC  +/KI 5d TGC  0%  18%  Figure 4.11 Methylation of Tel7KI upon paternal transmission in uncultured EPCs and FigureTGCs. 4.11: Methyaltion of Tel7KI upon paternal tranmission in uncultured EPCs (d0 cultured EPC) and EPCs cultured by in sequencing vitro for 5ofdays to bisulfite-modified generate TGCsgenomic (d5 TGC). DNA methylation was analyzed sodium DNADNA at the methylation was analyzed by sequencing of sodium bisulfite-modified genomic CAG promoter. Each circle represents a single CpG; open circles represent unmethylatedDNA. CpGs a single CpG; open representa unmethylated CpGs and andEach filled circle circlesrepresents represent methylated CpGs. Eachcircles line represents single DNA strand, and the filled circles represent methylated CpGs. Each line represents a single DNA strand, total percent methylation is noted to the right, calculated as total methylated CpGs over total and the total percent methylation is noted to the right. assayed CpGs.   74  4.3 Discussion   The analysis of Tel7KI expression and DNA methylation in the placenta gave unexpected  results given the behavior of this transgene in the embryo. Both maternal and paternal transmission of the Tel7KI transgene result in expression of GFP throughout the placenta, with the highest levels of expression in TGCs. The lack of imprinting of Tel7KI in the placenta is puzzling. While differences in the epigenetic regulation of genes between the placenta and embryo have been observed, there are no examples of endogenous imprinted genes which are imprinted in the embryo and not in the placenta. None of the known imprinted genes in the MMU7 distal imprinted domain region show a similar pattern of imprinting to Tel7KI, indicating that the mechanisms regulating Tel7KI may be a combination of effects.   The fundamental differences in epigenetic regulation between embryo and placenta could  have an effect on the acquisition of imprinting at an inserted transcriptional unit such as Tel7KI. It has been shown at specific imprinted genes in this region that the placenta does not rely as heavily on DNA methylation to regulate its gene expression as does the embryo (31; 122), so it is not unexpected that a novel transgene showing monoallelic expression associated with differential DNA methylation would show differences between these two lineages. In the case of Tel7KI, clearly multiple types of regulation are involved; in the embryo tissue-specific regulation of the CAG promoter as well as epigenetic regulation by DNA methylation were discussed in Chapter 3. The analysis presented in this chapter has revealed that lineage-specific differences in regulation between epiblast- and trophoblast-derived tissues are also involved. It would be extremely interesting to examine expression of Tel7KI in extraembryonic mesoderm, since this tissue is embryo-derived but located in the placenta. An attempt was made to determine the GFP expression by immunofluorescence, but no visible GFP signal was observed (Figure 4.4). An alternative could be to disaggregate the placenta and sort cells based on staining for CD-34, to isolate extraembryonic mesoderm and perform molecular analysis of expression and DNA methylation at Tel7KI. This might give some insight as to whether the imprinting at Tel7KI is truly lineage-specific.   High levels of GFP expression in TGCs have been reported previously with the X-linked  D4 allele, which carries a multi-copy insertion of the the CAG-EGFP construct on the X chromosome (190; 191). TGCs have been shown to reach ploidies of at least 500N and have 75  been previously hypothesized to have at least partial polytenization of their chromosomes (209-211). The combination of accumulation of massive amounts of DNA in the nucleus and atypical packaging make these cells interesting candidates for epigenetic instability. It is not inconceivable that something in the epigenetic maintenance machinery becomes limiting under these conditions and results in the gradual loss of proper epigenetic modifications and thus misregulation of normally imprinted genes or genes on the normally inactive X chromosome. In this study, initial immunofluorescent analysis of the placental phenotype of both D4 and Tel7KI (Figure 4.4) suggested that it may be the case that TGCs exhibit epigenetic instability in both X inactivation and genomic imprinting. Our studies did not examine X-inactivation in TGCs and in fact a similar investigation of transgene expression and methylation in the D4 line would be complicated by the fact that the CAG-EGFP inserted in that case is multi-copy. In addition, the Mus musculus castaneus polymorphisms used for analysis of MMU7 imprinted gene expression are from a congenic line containing castaneus DNA only on chromosome 7, and are not useful for X chromosome analysis. In the future, this experiment could be performed by repeating the TGC in vitro culture and analysis with the proper X-chromosome markers.   A major challenge of this experiment was the isolation of TGCs from placental tissue. A  typical cell isolation protocol by flow cytometry was unfeasible due to the large size of the cells, and an adapted protocol using only nuclei experienced similar problems. TGCs are too large and too fragile to sort on their own, and while their nuclei can be sorted up to 32N, GFP expression from the Tel7KI allele cannot be assayed in these nuclei. This left an in vitro culture method as the most efficient way to produce an enriched population of TGCs. There are two main ways of doing this; differentiation of trophoblast stem (TS) cells in vitro, or culture of ectoplacental cones (EPCs). EPC differentiation was chosen despite the fact that this method reduces the total number of cells recovered for two reasons. One is the fact that EPCs are relatively easy to recover from E8.5 conceptuses. The other is the fact that many long-term cultured cell lines show epigenetic instability (212-215). This potential for instability would confound the analysis of DNA methylation at ICs and at Tel7KI, so a primary culture system with minimized culture time of 5 days was used.   As reported, this culture system produced mainly TGCs growing out from the periphery  of the attached EPC. Qualitatively, by observing the size of nuclei and their staining with PL-I, it is estimated that >90% of the cells produced by this in vitro culture are TGCs, which correlates 76  with previous reports (216). Most of the remaining smaller, possibly diploid cells remain in the center of the colony where the EPC originally attached. These are likely TS cells, the progenitors of TGCs or other trophoblast derivatives.   Since reliable imprinted gene expression and DNA methylation had been observed in  Tel7KI in the embryo, the hypothesis of loss of imprinting due to high ploidy seemed a likely candidate for GFP expression in both Tel7KI and D4. However, here DNA methylation of the promoter driving GFP expression as well as expression and DNA methylation on endogenously imprinted genes was examined, and no defect in imprinted gene expression in TGCs was found. The only gene to show biallelic expression in TGCs was Tssc4, and this loss of imprinting in TGCs has been previously reported, though the mechanism is not understood (34). In addition, DNA methylation analysis shows that both imprinting centres (ICs) are normally methylated, thus no loss of imprinting in these highly-polyploid cells was observed. This analysis has conclusively shown that the GFP expression from the paternal allele in TGCs is not an indication of loss of imprinting at imprinted loci, but potentially caused by increased promoter usage in this cell type.   Interestingly, the D4 line was used to analyze the effect of the polycomb group gene Eed  on maintenance of paternal X-chromosome inactivation in TGCs (217). The researchers found that at E7.5-E9.5, GFP expression was observed in TGCs only from Eed-null embryos and concluded that Eed is responsible for maintaining paternal X-chromosome inactivation in the placenta. The D4 line has not been examined in the present study at these early stages, but the reactivation of GFP in TGCs seen at E12.5 and E13.5 in both this analysis and others does raise questions about these conclusions, at least as they apply to TGCs. If the CAG-EGFP transgene shows high promoter usage in TGCs at later stages, perhaps the reactivation observed at earlier stages in Eed-null embryos is not in fact indicative of X-chromosome reactivation, but merely suboptimal transgene silencing resulting in earlier reactivation of the GFP transgene in these cells.   An unusual bias was observed during the analysis of IC2 in +/KI cultured TGCs. Strand  bias is a known confounder of sodium bisulfite sequencing, which is why polymorphisms are often used to determine the parent of origin of each strand or, when possible, analysis is done on cells in a hemizygous state. In this case, the unmethylated paternal allele was over-represented by a factor of 5 (36 vs 7, see Figure 4.9). The assay in question has been used in the lab for many 77  years, and this bias has not before been observed. How the bias was introduced has not been determined, but it does not change the interpretation of the results since the parent of origin of each strand is known. There was a possibility that the bias actually represented a bias in the parental dosage of chromosome 7 homologues in TGCs, but this was ruled out by the allelic analysis of Kcnq1ot1 and Ascl2 genomic DNA showing equal amounts of maternal and paternal alleles in the polyploid gDNA of TGCs.   DNA methylation in EPC-derived TGCs carrying a paternally-inherited Tel7KI allele was  also examined, since in the embryo this allele acquires high levels of DNA methylation. In the EPC at E8.5, no DNA methylation has been acquired whereas in the embryo just one day later at E9.5, significant DNA methylation is already present (Figure 3.5). Upon culture, TGCs acquire DNA methylation at Tel7KI, such that after 5 days, 18% of CpGs are methylated. The increase in methylation during EPC maturation to TGC is further supported by the moderate level of methylation (37%) observed on the paternal allele at E14.5 in whole placenta (see Figure 4.3). It appears that DNA methylation in extraembryonic tissues at Tel7KI is acquired, at least on the paternal allele, during early post-implantation development. The relatively equal amount of DNA methylation on the paternal alleles observed in Figure 4.3 and the similar levels and patterns of GFP expression between the alleles implies that this is not acquired in a parent-oforigin-dependent manner as it is in the embryo, though the analysis has not been performed on maternal Tel7KI TGCs.   Based on the models of IC1 vs IC2 regulation of Tel7KI proposed in Chapter 3, there are  two different possibilities to account for the differences in imprinted expression of Tel7KI between the embryo and placenta. As previously mentioned, the H19 ICR is the control element for imprinted expression of Igf2, which contains three DMRs, DMR0, DMR1, and DMR2. DMR0, the closest DMR to Tel7KI, surrounds a placental-specific promoter driving Igf2 expression in the placenta (141). Unlike DMRs 1 and 2, DMR0 is maternally methylated, and its function is currently unknown, though it has been shown to be responsive to signals from the H19 DMR (141). Interestingly, though it is known to have an effect on placental expression of Igf2, the DMR0 is not implicated in the now well-defined “chromatin looping model” for imprinted regulation of Igf2 (Figure 1.1). In that model, the position of DMR0 in a particular chromatin loop does not change depending on the methylation of the H19 DMR, unlike DMRs 1 and 2 (142; 153). This is of particular interest in terms of modeling possible reasons for the 78  difference in epigenetic regulation of Tel7KI in the embryo and placenta. Igf2, after all, is the closest gene that is imprinted in more than a single tissue, since Ins2, the gene between the Tel7KI insertion and Igf2, is imprinted only in extraembryonic endoderm (144). The fact that Igf2 contains DMRs which show differential regulation by the H19 control element in embryo and placenta may be supportive of a role for that imprinting centre in the regulation of Tel7KI. However, DMR0 still shows allele-specific modifications in the placenta, though they are opposite of the patterns observed at the embryonic DMRs, whereas Tel7KI shows an absence of parent-of-origin-specific epigenetic marks (141). In addition, DMR0 is heavily methylated on both parental alleles in the embryo, while Tel7KI shows preferential paternal methylation (141). While evidence of lineage-specific differences in regulation is encouraging, the actual patterns of DNA methylation and allelic expression are inconsistent with IC1 regulation at Tel7KI.   The relationship with IC2 as it relates to differential imprinting in embryo and placenta is  a similar combination of similarities and contradictions. Here, the differences between embryo and placenta with respect to imprinting of the genes in this domain are more clear, but how this could relate to Tel7KI is uncertain. In this domain, the noncoding RNA Kcnq1ot1 is produced from the unmethylated paternal allele and represses in cis the remainder of the maternallyexpressed genes (155). There has been shown to be a gradient of influence of Kcnq1ot1 when it comes to genes imprinted in both embryonic and extraembryonic lineages versus those expressed solely in extraembryonic tissues. Genes located close to the IC (which also serves as the promoter for the ncRNA) are ubiquitously imprinted in both lineages, and those further away are imprinted only in the placenta (184). It is predicted that the signals from IC2 have not yet reached these distal genes by the time the trophoblast lineage has differentiated, as evinced by the fact that ubiquitously imprinted genes show monoallelic expression in blastocysts, while placentally imprinted genes do not (34). Since Tel7KI is located almost 300kb further than Ascl2, the most centromeric gene known to be regulated by IC2, it would follow that, if Tel7KI was under the regulation of IC2, it would resemble these distal placentally-imprinted genes. A potential confounding factor in this comparison, however, is the presence of a CpG island in the promoter of Tel7KI. Cdkn1c contains a CpG island, but is located very close to IC2 and is ubiquitously imprinted (115). The only placentally imprinted gene in the IC2 domain which has been reported to have a CpG island is Tssc4, and reports are conflicting whether or not it is capable of undergoing allele-specific modification, so it is reasonable to assume that the CpG 79  island at Tel7KI will change its response to the IC2-mediated signal. It is possible that the combination of being located at a distance from IC2 and containing a CpG island has resulted in a unique combination of mechanisms regulating Tel7KI.   There are two possible reasons for the embryonic/extraembryonic differences under IC2  regulation. The first may be differences in chromatin spreading along MMU7, it is possible that the signal to silence the paternal allele simply never reaches Tel7KI in the placenta. Alternatively, in early extraembryonic development, once the paternal-allele-specific signal from IC2 has spread as far as Tel7KI, it is not able to be maintained by histone modifications as a normal placentally-imprinted gene would be, and the general hypomethylation in this tissue does not allow for allele-specific marks to be established at the correct time.   The evidence for either IC being involved in the regulation of Tel7KI is mixed. The  models both have features which are inconsistent with the established models of imprinted regulation on MMU7. However, the fact that the IC1-regulated gene Ins2 is not imprinted, Igf2 is imprinted in the opposite manner, and that the accepted chromatin looping model for IC1 is incompatible with observations at Tel7KI lead to a slightly higher probability that IC2 is responsible for imprinting at Tel7KI. The next chapter will examine this possibility using an EScell based approach.  80  Chapter 5: Tel7KI ES cells as a model for acquisition of imprinting during early post-implantation development3 5.1 Introduction   Embryonic stem (ES) cells are derived from the inner cell mass of the blastocyst, and  lines have been derived for many mammals including mouse and human (218; 219). ES cells are pluripotent, can be maintained in culture with the presence of Leukemia Inhibitory Factor (LIF), and are generally believed to be a model for early embryonic cells (220). When injected into the cavity of a blastocyst or combined with an early cleavage embryo, mouse ES cells can contribute to the resulting embryo, forming chimeras (221). If ES cells are combined with tetraploid embryos, which are capable of forming trophoblast but do not usually make significant contribution to embryonic lineages, completely-ES cell derived animals can be obtained (222; 223). Additionally, ES cells can be made to differentiate into many lineages in vitro by withdrawal of LIF and/or addition of other factors (ie. retinoic acid) which can encourage differentiation (224). In vivo or vitro, markers of pluripotency such as Nanog and Oct-4 are down-regulated in the initial differentiation steps (225).   ES cells have been used to study early steps in X-chromosome inactivation. Like  genomic imprinting, X chromosome inactivation (XCI) in females is an epigenetic process and the two phenomena have many similarities including differential DNA methylation, histone modifications, and replication timing (226-228). Undifferentiated female mouse ES cells have two active X chromosomes, and upon differentiation one of the two is randomly inactivated (229; 230). This process has been used to study timing and order of XCI silencing events, tissueand cell-specific mechanisms of XCI, as well as chromosome counting and parental choice of XCI (229-231). One advantage to these ES-cell based experiments is that, in vivo, these events occur shortly after implantation, a stage which is not amenable to experimentation or analysis due to limiting cell numbers (232).   A recent study examined allelic expression of the imprinted Igf2r gene and its ncRNA  regulator, Air, in undifferentiated and differentiated ES cells. Igf2r is biallelically expressed in  Figure 5.1 and part of Figure 5.2 have been published in Jones, M and Lefebvre, L. (2009) An imprinted GFP insertion reveals long-range epigenetic regulation in embryonic lineages. Developmental Biology. 336: 42-52. 3  81  blastocysts and ES cells and becomes monoallelic in embryos later in development (197; 233). It is regulated by a primary germline DNA methylation imprint on the maternal Air locus, which results in Air ncRNA expression from the paternal allele, and post-implantation repression of the paternal Igf2r with subsequent acquisition of DNA methylation at its promoter (27; 234). Differentiation of ES cells by LIF withdrawal and retinoic acid treatment resulted in recapitulation of the embryonic program; Igf2r became preferentially maternally expressed and acquired DNA methylation at its promoter, while Air maintained its primary DNA methylation imprint and paternal expression (235). This approach will be used to determine the usefulness of Tel7KI ES cells as a tool to study acquisition of imprinting and to assay the role of IC2 in regulation of imprinted expression of Tel7KI. 5.2 Results 5.2.1 Tel7KI insertion is conditional in KIO ES cells   The original Tel7KI allele was inserted as described in R1 ES cells on the maternal  chromosome, where the original I2loxP allele was inserted (179). In this study, this original Tel7KI insertion ES cells from which the mouse line was created are referred to as KIO. The I2loxP allele contained a single loxP site, as did the targeting vector used to generate both the DelTel7 and Tel7KI alleles (179). The insertion thus resulted in a loxP-flanked allele (Figure 1.3). To confirm this structure, these KIO cells were electroporated with a transient Creexpressing vector, and GFP– colonies were picked and sub-cloned. Many of these cells have been shown to have reverted to the I2loxP genotype, indicating that the insertion at Tel7KI is easily reversible by expression of Cre recombinase and therefore constitutes a conditional insertion (Fig. 5.1).  82  A I2loxP  B  I2 5’ !3’ loxP-neopA Ins2  I2 5’ 2x  Cre  Tel7KI  1 2 3 4 c!3’  2x Th  loxP  tel  !3’ pCX-EGFP Pgk-loxP-neopA  Ins2  Figure 5.1 Tel7KI insertion is reversible with Cre. (A) Schematic of Tel7KI reversible targeting showing genotyping PCR reactions. (B) Analysis of Figure Tel7KI insertion is reversible Cre. from (A) Schematic of Tel7KI reversible Tel7KI5.1: clones excised with Cre. Two cloneswith resulting Cre expression in Tel7KI ES cells (1 targeting showing genotyping PCR(3)reactions. (B) Analysis of Tel7KI clones excised and 2), Tel7KI ES cells without Cre and the parental I2loxP cells in which Tel7KI targeting with Cre. Two clones resulting from Cre expression in Tel7KI ES cells (1 and 2), Tel7KI was performed (4) were analyzed. No template control (c-) shown at right. ES cells without Cre (3) and the parental I2loxP cells in which Tel7KI targeting was performed (4) were analyzed. No template control (c-) shown at right.  83  5.2.2 Both paternal and maternal transmission of Tel7KI result in GFP+ blastocysts and ES cell lines   Given the fact that Tel7KI is imprinted and maternally expressed in post-implantation  embryos, preimplantation embryos were examined to determine the timing of epigenetic silencing of Tel7KI. Unlike the pattern observed in postimplantation embryos, both maternal and paternal transmission of Tel7KI result in GFP fluorescence in inner cell mass and trophectoderm cells in E3.5 pre-implantation embryos (Fig. 5.2 A). Next, male or female Tel7KI animals (maintained on the 129/S1 strain background) were mated to wild-type C57BL/6J animals to generate paternal and maternal transmission Tel7KI ES cell lines, which are named KIP and KIM, respectively. Six KIP lines were derived by L. Lefebvre and one KIM line was derived by A. Bogutz. All cell lines were analyzed with respect to sex chromosome composition; the sole KIM line (KIM A2) and three of the six KIP lines (KIP B1, B2, C3) were found to be XX, while two of the other KIP lines (KIP B3, C4) were XY and one (KIP D4) was found to be XO (data not shown). Not surprisingly given that the ICM of +/KI and KI/+ blastocysts were GFPpositive, all KI ES cell lines showed GFP expression by microscopy, and nearly 100% GFPexpressing cells by flow cytometry (data not shown and Fig. 5.2 B). 5.2.3 KIP and KIM ES cell lines silence GFP differently upon differentiation   Many imprinted genes are expressed biallelically in blastocysts but monoallelically in the  embryo, for example Igf2 and Igf2r (197). Thus it was possible that Tel7KI followed this same pattern. Tel7KI ES cells were differentiated into embryoid bodies (EBs) by withdrawal of LIF and culture on bacterial dishes to determine whether the in vivo silencing on the paternal allele of Tel7KI was recapitulated in vitro. Six cell lines were used (see Table 5.1 for details); the originally-targeted Tel7KI cells from which the mouse line was created (KIO), the maternaltransmission Tel7KI line (KIM A2), and two paternal-transmission Tel7KI lines (KIP C3 and C4). Two KIP lines were used to determine whether sex chromosome composition had an effect on epigenetic modifications, as KIP C3 is XX and KIP C4 is XY. In addition, two control lines were used, one GFP– parental line, I2loxP, and a control line which expresses the GFP gene under the control of the Oct-4 promoter, Oct-4 EGFP. This line has been shown to silence GFP 84  A  KI/+  +/KI  B KI/+ (KIM A2)  +/KI (KIP C3) 600 # Cells  # Cells  300 200 100 0  4.17  95.8  0 102 103 104 105 GFP fluorescence  400 200 0  1.2  98.8  0 102 103 104 105 GFP fluorescence  Figure 5.2 GFP expression in blastocysts and ES cells made from transmission of Tel7KI through theGFP maternal (KI/+) or (+/KI) Figure 5.2 expression in paternal blastocysts andgermline. ES cells made from transmission of (A) E3.5 embryos from a cross between wild-type C57BL/6J and A) Tel7KI female Tel7KI through the maternal (KI/+) or paternal (+/KI) germlines. E3.5heterozygous maternal (KI/+) (KI/+) or male (+/KI) mice. Pre-implantation embryos were observed by microscopy under and paternal (+/KI) hemizygous embryos were collected from crosses between bright fieldC57BL/6J (left) or GFP (right). Scalemice. bar: 20 μm. (B) Flow cytometry wild-type andfluorescence Tel7KI heterozygous Pre-implantation embryosprofiles were for ES cell lines derived from Tel7KI-carrying blastocysts. KIM A2 (KI/+) and KIP C3 (+/KI) observed by microscopy under bright field (left) or GFP fluorescence (right). Scale bar: ES cells were and passaged on gelatin before trypsinization 20 µm. B)derived Flow cytometry profiles for ES cell lines derived and fromanalysis. blastocysts shown in A). KIMA2 (KI/+) and KIP C3 (+/KI) ES cells were derived and passaged on gelatin before trypsinization and analysis.  85   Table 5.1: ES cell lines used in this study Cell line  Genotype/ transgene  Germ line transmission?  Sex  Origin  Reference  I2loxP  I2loxP/+  no, targeted  XX  Nagy lab  (179)  Oct-4 EGFP  Oct-4 EGFP multicopy insertion  no, targeted  XY  Nagy lab  (236)  KIM A2  Tel7KI/+  yes, maternal  XX  Lefebvre lab  this study  KIO  Tel7KI/+  no, targeted  XX  Nagy lab  (179)  KIP C3  +/Tel7KI  yes, paternal  XX  Lefebvre lab  this study  KIP C4  +/Tel7KI  yes, paternal  XY  Lefebvre lab  this study  KIP C4 IC2KO  +/Tel7KI; +/IC2KO  Tel7KI: yes, paternal IC2KO: no, targeted  XY  M. Jones  this study  86  efficiently upon ES cell differentiation (236; 237).   All six cell lines were counted and seeded, and samples were examined by flow  cytometry on day 0 and every 5 days for a total of 30 days. At day 0 all cell lines showed close to 100% GFP expressing cells, with the exception of Oct-4 EGFP, which began with approximately 30% of cells with a silent EGFP, and the GFP– I2loxP (Fig. 5.3A). By day 10, the Oct-4 EGFP cell line had effectively completely silenced its transgene, indicating efficient differentiation of the ES cells, but the experiment was allowed to continue to examine the kinetics of silencing at Tel7KI, which appeared to be slower (Fig. 5.3). By day 15 and onward to day 30, both KIP lines silenced their GFP to a greater extent than either the KIM or KIO line (Fig. 5.3B). The amount of silencing observed in the KIM line (35% GFP+ by day 30) was consistent with observations in the embryo, where approximately 40% of the cells at E9.5 were GFP+ (Fig. 3.3). This tissuespecific silencing of GFP may also be responsible for the silencing in the KIO cells, which also carry the Tel7KI allele on the maternal chromosome (179).   Next, DNA methylation at Tel7KI was examined to determine whether the differences in  GFP silencing were reflected in differences in acquisition of DNA methylation. Bisulfitemodified genomic DNA from days 0 and 25 of differentiation were analyzed by COBRA assay. Day 25 was used as an end point because of an unusual dip in GFP-expressing cells in the KIO line at day 30, which implied that the cells may be under some distress due to extended culture. Ratios of methylated/unmethylated band density were calculated by ImageJ (Figure 5.4). Both the paternal KIP C3 and KIP C4 showed higher ratios of methylated/unmethylated bands after 25 days in culture than the maternal KIM A2 line, consistent with DNA methylation being inversely correlated with GFP expression (Figure 5.4). KIP C3 had a much higher ratio than KIP C4, again reflecting its much lower GFP expression. 5.2.4 Role of IC2 in epigenetic regulation of Tel7KI during ES cell differentiation   Once an assay to examine acquisition of imprinted expression in Tel7KI ES cells was  established, it was possible to perform experiments aimed at discovering the mechanism of imprinted regulation at this allele. In order to determine whether IC1 or IC2 signals are resulting in the imprinted expression of Tel7KI, a deletion was created of KvDMR1, which serves both as  87  d0  d5  d10  d15  80  80  80  80  60 40 20  60 40  0  0 102 103 104 105 GFP Fluorescence  d20  60 40 20  20  0  0  0 102 103 104 105 GFP Fluorescence  80  80  % of max  80 % of max  100  40 20  40 20  0  0 102 103 104 105 GFP Fluorescence  0  40  0 102 103 104 105 GFP Fluorescence  0  0 102 103 104 105 GFP Fluorescence  d30  100  60  60  20  d25  100  60  % of max  100  % of max  100  % of max  100  % of max  100  % of max  A  Oct4 EGFP KIP C4 KIP C3 KIO KIM A2 I2loxP  60 40 20 0  0 102 103 104 105 GFP Fluorescence  0 102 103 104 105 GFP Fluorescence  B 100  % of cells GFP+  80 60 40 20 0 0  5  10 15 20 days of differentiation  25  30  Figure 5.3 Paternal (KIP) and maternal (KIM) transmission Tel7KI ES cells show Figure 5.3: silencing Paternal (KIP) and maternalbody (KIM)differentiation. transmission Tel7KI ES cells show differences in GFP during embryoid differences in GFP silencing during embryoid body differentiation. were was ES cells were cultured on bacterial dishes in the absence of LIF for 30 days,ES andcells a sample bacterial dishes in the absence of LIF for 30 of days, sample was30 taken taken forcultured FACS on analysis every 5 days. (A) FACS analysis sixand cella lines over days of for FACS analysis every 5 days. A) FACS analysis of six cell lines over 30 days of differentiation showing GFP expression profiles each day. See Table 5.1 for cell line details. (B) differentiation showing GFP expression profiles each day. See table 5.1 for cell line Graphical representation of percent GFP-positive cells for each cell line over the 30 days of details. B) Graphical representation of percent GFP-positive cells for each cell line differentiation. inof A differentiation. is consistent with colours over theLegend 30 days Legend in Ain is B. consistent with colours in B.  88  A KIP C4 d0  KIP C3  d25  d0  d25  KIM A2 d0  d25  u m  B ratio methylated/unmethylated  1.6 d0 d25  1.4 1.2 1 0.8 0.6 0.4 0.2 0 KIP C4  KIP C3  KIM A2  Cell line  Figure 5.4 DNA methylation at the CAG promoter in Tel7KI ESC lines before and after differentiation. Figure 5.4: DNA methylation at the CAG promoter in Tel7KI ESC lines before and after (A) COBRA A) analysis of bisulfite-treated gDNA from KIP C4,from KIP C3, A2 ESC differentiation. COBRA analysis of bisulfite-treated gDNA KIP and C4, KIM KIP C3, and lines before (d0) and after (d25) 25 days of embryoid body differentiation by LIF withdrawal. KIM A2 ESC lines before (d0) and after (d25) 25 days of embryoid body differentiation PCR products digested with and run on a digested 4% nusieve gel.BstBI BstBIand cuts run the methylated through LIFwere withdrawal. PCRBstBI products were with on a 4% CpG at 38bp, unmethylated (u: 257bp) and methylated (m: 219bp) bands are indicated. (B) Ratio of nusieve gel. BstBI cuts the methylated CpG at ??, unmethylated (u) and methylated to unmethylated density measured bytoImageJ for each cell linedensity at both time (m)methylated bands are indicated. B)band Ratio of methylated unmethylated band points. measured by ImageJ for each cell line at both time points.  89  Tel 7q  B  H19  IC1  Igf2  Tel7KI Ins2  Th  Cd81 Ascl2  Tssc4  IC2  Kcnq1  Nap1l4 Phlda2 Slc22a18 Cdkn1c  A  cen  Kcnq1ot1  BstBI  BsrGI CpG  CpG  KvDMR1  C  S targeting vector  PGK-puro-pA Transgene-specific PCR Insertion-specific PCR  Figure 5.5 Deletion of KvDMR1 in Tel7KI ESC. Figure 5.4: Deletion of showing KvDMR1relative in Tel7KI ESC.ofA)Tel7KI Schematic of (B) MMU7 showing of (A) Schematic of MMU7 locations and IC2. Magnification relative of Tel7KI IC2. B) Magnification intron 10islands of Kcnq1 showing intron 10locations of Kcnq1 showingand KvDMR1, which containsoftwo CpG as well as the KvDMR1, which contains two CpG islands as well as the transcription start transcription start site for Kcnq1ot1 (arrow). (C) Targeting vector for deletion of site bothfor CpG Kcnq1ot1 (arrow). C) Targeting vector for deletion of both CpG islands in KvDMR1 and islands in KvDMR1 and replacement with a puromycin resistance gene cassette flanked by loxP replacement with a puromycin resistance genescreening casette PCR flanked loxP sites (black A sites (black triangles). Primer sites for knockout (halfbyarrows) are shown. triangles). Primer sites knockout site screening (halfa 1.8kb arrows) are internal shown.toAthe common reverse at the Pgkfor promoter/loxP junctionPCR produces product common reverse in the PGK promoter 1.8kb specific productfor internal to theIC2 targeting vector (transgene-specific PCR) orproduces a 2.3kb aproduct the correct targeting vector (transgene-specific PCR) or a 2.3kb product specific for the correct recombination site (insertion-specific). IC2 recombination site (insertion-specific).  90  IC2 and as the promoter for the non-coding RNA Kcnq1ot1 (Fig. 5.5). This deletion results in lack of expression of Kcnq1ot1 from the paternal allele and biallelic expression of the linked maternally expressed genes, and has been previously published (155). A targeting construct was made based on the published construct (with generous donations of starting material from M. Higgins) but with the substitution of a puromycin resistance gene in place of the existing G418 resistance, since Tel7KI ES cells already contain a G418 resistance gene (Figure 1.3). Paternal transmission ES cells (KIP C4) were electroporated with this construct and puromycin-resistant colonies were screened for deletions of KvDMR1 on the paternal chromosome. This particular line was chosen for two reasons; one, because it is an XY cell line, which has shown to be more epigenetically stable and should result in male chimeras when the time comes to create a mouse line based on this deletion (215). Two, of the XY KIP lines, it had the best morphology and grew in culture quickly. The screening for the deletion was done by PCR and COBRA (Figure 5.6). Four clones (1A, 3D, 7E, and 12G) were identified on the basis of an insertion-specific PCR assay (Figures 5.5, 5.6). COBRA analysis of these four clones was extremely useful, as it helped confirm hemizygosity at IC2 and determine exactly which parental DMR was missing in the deleted cells. Two of these clones were determined to have deleted KvDMR1, one on the maternal allele (12G) as determined by COBRA, and one on the paternal allele (3D) (Figure 5.6). Clone 3D was selected for further investigation as it had successfully deleted IC2 in cis to Tel7KI on the paternal chromosome.   Next, the EB differentiation experiment outlined in section 5.2.3 was repeated with the  addition of this new cell line, called KIP C4 IC2KO. This line contained the IC2 knockout on the same chromosome as the paternally-inherited Tel7KI which had been shown to silence GFP upon differentiation. This had the advantage of replicating the original differentiation study while examining the IC2 knockout for any effects on the silencing of Tel7KI on the paternal allele. As determined by the first experiment, after day 20 of differentiation little difference was noted in GFP expression profiles and the cells appeared to be under some stress, as determined by poor morphology and increased cell death, so 20 days was used as the end point for differentiation. This time, the Oct-4 EGFP line did not completely silence its GFP until day 15, meaning that the differentiation conditions were not as stringent the second time. Otherwise, similar results were observed with respect to the KIO, KIM A2, and KIP C4 lines (compare Fig 5.3 to Fig. 5.7) up to day 10, where the GFP signal leveled off at a slightly higher percentage 91  A 3  L a  b  c  d  4 e  f  g  h  a  b  c  d  LL e  f  g  ch c-  I2wt  3  L a  b  c  4 e  d  f  g  h  a  b  c  d  L e  f  g  h c-  transgenespecific PCR  3  L a  b  c  d  4 e  f  g  h  a  b  c  d  L e  f  g  h c-  Insertionspecific PCR  *  B  +/+  +/"  "/+  3D  1A  7E  12G pat (unmethylated) mat (methylated) common  Figure 5.6 Screening of clones for IC2 KO. (A) PCR screening of clones from groups 3 and 4. Each clone was screened with an I2wt PCR as a positive control, a transgene-specific PCR to determine whether the clone carried the IC2KO Figure 5.5: Screening of clones for IC2 KO. A) PCR screening of clones from groups 3 targeting vector, and an insertion-specific PCR which would amplify only those clones who had and 4. Each clone was screened with an I2wt PCR as a positive control, a correctly recombined and deleted IC2. Of 72 clones in 12 groups tested, four were positive by transgene-specific PCR to determine whether the clone carried the IC2KO targeting these criteria (1A, 3D, 7E, and 12G), only one (3D, asterisk) is shown here. Ladders (L) and vector, and an insertion-specific PCR which would amplify only those clones who had negative control (c-) are also shown. (B) COBRA analysis on bisulfite-modified gDNA from correctly recombined andindeleted IC2. Of 72 screening clones tested, four these as potential clones identified preliminary PCR as well as were distalpositive MMU7 by deletions criteria (1A, 3D, 7E, AND 12G), only one (3D, asterisk) is shown here. Ladders (L) and controls. Bisulfite PCR products for KvDMR1 were digested with RsaI and BccI to generate negative (c-) unmethylated are also shown. COBRA analysis on bisulfite-modified gDNA methylatedcontrol (mat) and (pat) B) bands, as well as a common non-methylation-sensitive from clones identified preliminary PCRonly screening asatwell as distalasMMU7 band. potential DelTel7 reciprocal (+/∆ and in ∆/+) samples show one allele KvDMR1, do clones deletions as controls. Bisulfite PCR products for KvDMR1 were digested with RsaI and 3D and 12G. BccI to generate methylated (mat) and unmethylated (pat) bands, as well as a common non-methylation-sensitive band. Deletions (+/! and !/+) samples show only one allele at KvDMR1, as do clones 3D and 12G.  92  A  d0  d5  d10  80  80  80  60 40 20  % of max  100  % of max  100  % of max  100  60 40 20  0  0 0 10 2 10 3 10 4 10 5 GFP fluorescence  0 10 2 10 3 10 4 10 5 GFP fluorescence  d20  d15 100  80  80  % of max  100  % of max  40 20  0 0 10 2 10 3 10 4 10 5 GFP fluorescence  60  60 40 20  KIPC4 KIPC4IC2KO KIPC3 KIO KIMA2 Oct4 EGFP I2loxP  60 40 20  0  0 0 10 2 10 3 10 4 10 5 GFP fluorescence  0 10 2 10 3 10 4 10 5 GFP fluorescence  B 100  % of cells GFP +  80 60 40 20 0  0  5  10 day of differentiation  15  20  Figure 5.7 A knockout of IC2 in paternal transmission (KIP) ES cells (KIP C4 IC2KO) impairs upon differentiation. FigureGFP 5.7: silencing A knockout of IC2 in paternal transmission (KIP) ES cells (KIP C4IC2KO) ESresults cells were in the absenceGFP of LIF fordifferentiation.ES 20 days, and a sample in a cultured decreaseoninbacterial abililty ofdishes the cells to silence upon cells was taken forcultured FACS analysis every 5 days. (A)absence FACS analysis of 30 seven celland lines over 20was days of were on bacterial dishes in the of LIF for days, a sample taken for FACS analysis 5 days. A) FACS of seven cellcell lines over 20 (B) differentiation showing GFP every expression profiles each analysis day. See table 5.1 for line details. days of differentiation GFP expression cells profiles Graphical representation showing of percent GFP-positive foreach eachday. cell See line table over 5.1 the for 20cell days of line details. B) Graphical representation of percent GFP-positive cells for each cell line differentiation. Legend in A is consistent with colours in B. over the 20 days of differentiation. Legend in A is consistent with colours in B.  93  than the first experiment. The KIP C3 line silenced its GFP slower and to a lesser extent during this second differentiation experiment, but was still silenced more than the KIO or KIM lines (Fig 5.7). Most striking in this experiment was the difference in differentiation profiles between the parental KIP C4 (purple) and the KIP C4 IC2KO (orange) carrying the IC2 deletion. Between days 10 and 15, when the GFP signals from the cell lines are quite stable, the KIP C4 line shows less than 40% GFP+ cells, while the IC2 KO exhibits a percentage of almost 70% (Fig. 5.7).   Finally, the effect of the IC2 KO on the ability of the KIP C4 line to establish DNA  methylation marks at Tel7KI was assayed. DNA methylation at the CAG promoter driving GFP expression in KIM A2, KIP C4, and KIP C4 IC2KO cells was examined in day 0 ES cells and day 15 EBs by bisulfite sequencing. In undifferentiated cells, the KIP C4 IC2KO cells and the KIM cells showed low amounts of DNA methylation (4.9% and 6.7%, respectively) while the KIP C4 cells showed a higher level of methylation, at 14.5%. After differentiation, both the KIP C4 IC2KO and KIM A2 acquired a small amount of DNA methylation, increasing to 12.9% and 16.4% respectively, whereas the KIP C4 line acquired more methylation, with 36.2% of CpGs methylated in day 15 EBs (Figure 5.8). Thus, the IC2 knockout changed the dynamics of acquisition of DNA methylation at Tel7KI in cis. The strand distribution of methylated CpGs is interesting here; upon differentiation, the KIP C4 IC2KO cells acquired DNA methylation densely only on a single strand, which accounts for a large percentage of its methylated CpGs (Figure 5.8 A). It may be that the DNA methylation in this sample is over-represented due to that strand, and thus that the difference caused by the IC2 KO in cis in paternal transmission Tel7KI ES cells is larger than it appears. 5.3 Discussion   In later stage post-implantation embryos, imprinting of Tel7KI is consistent and reliable  with extensive maternal allele-specific GFP expression and very little expression from the paternal allele (See Chapter 3). The analysis of preimplantation embryos and ES cells has shown evidence for a post-implantation acquisition of imprinting, as GFP expression from Tel7KI is observed in both blastocysts and ES cells derived from them upon both parental transmission (Fig. 5.2). This pattern is observed at other imprinted genes, for example Igf2r and Igf2. Igf2r  94  A 0 days  4.9%  15 days  12.9%  0 days  14.5%  15 days  36.2%  0 days  6.7%  15 days  16.4%  KIM A2  KIP C4  KIP C4 IC2KO  B % methylated CpGs  40  day 0 day 15  35 30 25 20 15 10 5 0  KIMA2  KIPC4  KIPKO  Cell line Figure 5.8 An IC2 knockout impairs acquisition of DNA methylation at Tel7KI during ES Figure 5.8: An IC2 knockout impairs acquisition of DNA methylation at Tel7KI during ES cell differentiation. cellSodium differentiation. A) Sodium analysis bisulfiteofsequencing analysis of DNA methylation at (A) bisulfite sequencing DNA methylation at CpGs in the CAG promoter of CpGs in the CAG promoter of Tel7KI. Filled circles represent methylated CpGs, unfilled Tel7KI. Filled circles represent methylated CpGs, unfilled circles represent unmethylated CpGs, circles represent and absent indicate CpGs for cells whichand data and absent circles unmethylated indicate CpGsCpGs, for which data is circles unavailable. Day 0 ES day 15 is unavailable. Day 0 ES cells and day 15 Embryoid Bodies of three cell lines (KIM A2, embryoid bodies of three cell lines (KIM A2, KIP C4, and KIP C4 IC2KO) were analyzed. (B) KIP C4, and KIP C4 IC2KO) were analyzed. B) Graphical representation of percentage Graphical representation of percentage methylated CpGs over total CpG number for samples methylated CpGs over total CpG number for samples shown in A). shown in A). 95  begins to be expressed at the 4-cell stage, but both parental alleles are expressed until implantation, when the maternal allele becomes the active allele (197). Igf2 is not activated until the blastocyst stage, but is biallelically expressed until after gastrulation, when it becomes paternally expressed (197). Biallelic expression of both of these genes has been hypothesized to be basal, and monoallelic expression is established once full expression levels are reached (235). Additionally, while Cdkn1c is monoallelically expressed from early cleavage stages, it does not acquire allele-specific DNA methylation until after implantation (115). Thus, a precedent exists for postimplantation acquisition of allele-specific expression and DNA methylation. The fact that Tel7KI appears to follow this pattern has allowed for an in vitro analysis of the effect of IC2 on the ability of Tel7KI to acquire its imprinted phenotype.   This postimplantation acquisition of imprinting also contributes to the models of IC1  versus IC2 regulation of Tel7KI. As mentioned, Igf2 acquires its monoallelic expression early in preimplantation development, similar to what is observed for Tel7KI (197). This could indicate a spreading of signals from IC1 to Tel7KI at the blastocyst stage. However, there remain two factors which confound this theory. The first is the fact that Igf2 is paternally expressed, while Tel7KI is maternally expressed (16). The second is Ins2 paternal tissue-specific expression (144). Neither of these patterns are consistent with a simple spreading of imprinting signals from IC1 to Tel7KI. Should Tel7KI prove to be regulated by IC2, its preimplantation expression pattern could be explained by its distance from that regulatory region. It is likely that in the blastocyst, the imprinting signal from IC2 has not yet reached Tel7KI, as is observed by biallelic expression of distal or “placentally-imprinted” IC2-regulated genes (31; 238). As previously observed, these placentally imprinted genes do not contain CpG islands, and this may be a reason why they do not become imprinted in the embryo. Tel7KI, however, with its highly-CpG rich island, becomes DNA methylated on one allele in the embryo in response to the signals from IC2 similarly to the “ubiquitously-imprinted” Cdkn1c (115).   This was one of the reasons IC2 was chosen to be investigated more thoroughly as a  regulatory factor in Tel7KI imprinting. Another is that deletion of IC1, the H19 DMR, does not result in consequences which are as clearly defined as the IC2 deletion. Maternal deletion of the H19 DMR results in reduced H19 expression and biallelic Igf2, whereas paternal transmission results in biallelic H19 and reduced Igf2 (146). With respect to DNA methylation, maternal transmission of an H19 DMR deletion results in increased maternal DNA methylation of Igf2, 96  while paternal transmission of the same deletion has little effect (142). Since Tel7KI is expressed from the opposite parental chromosome from Igf2 but methylated on the same chromosome, it is difficult to formulate hypotheses about how it would behave in this context. It is possible that maternal deletion of the H19 DMR would result in increased DNA methylation at Tel7KI and thus silencing, if the DNA methylation is the sole feature responsible for imprinting of Tel7KI.   Deletion of IC2 on the paternal allele
in embryos results in lack of Kcnq1ot1 expression  and subsequent de-repression of Ascl2, Tssc4, Kcnq1, Cdkn1c, Slc22a18, and Phlda2 (155). This deletion confers a “maternal” epigenotype to this region, as evidenced by a rescue of maternallyinherited deletion of the entire distal MMU7 (179). In that study, paternal inheritance of the IC2 knockout resulted in sufficient expression of the normally paternally silent genes to fully rescue a lethal phenotype associated with complete loss of maternal gene expression upon inheritance of a maternal DelTel7 deletion. In effect, the deletion of IC2 mimics a maternal, methylated allele, with no expression of the ncRNA Kncq1ot1, which would have specific and easily-testable effects on Tel7KI given our models (155). The deletion described in this chapter is identical to the published knockout with the exception of the resistance marker used for selection, and the existence of the Tel7KI gene in cis. Eventually a mouse line carrying this new knockout, KIP C4 IC2KO, will be created, but for now the analysis has been done on differentiating ES cells. An ES-cell differentiation-based experiment of this kind has not been attempted with the original IC2 knockout line, and it is unfortunate that the KIP C4 IC2KO ES-cells do not contain polymorphisms which would allow examination of imprinting at other IC2-regulated genes during differentiation of these ES cells.   If Tel7KI is under the control of IC2, when IC2 is deleted in cis on the paternal allele one  would expect a switch from a paternal/off to a maternal/on expression pattern due to the lack of Kcnq1ot1 expression from the deleted paternal allele. The described observations are consistent with this model, with the KIP C4 IC2KO cell line resembling the KIM A2 cell line much more than its parental KIP C4 cell line in both GFP silencing and acquisition of DNA methylation at Tel7KI. However, since it is not possible to examine imprinted expression at known IC2regulated genes to confirm the effect of the IC2 deletion, this data must be considered suggestive and not conclusive. To confirm the possibility that Tel7KI is regulated by IC2, the KIP C4 IC2KO cell line will be used to make chimeric mice, and a mouse line containing both alleles in cis will be created. 97  Chapter 6: Discussion 6.1 Summary of results and conclusions    The Tel7KI transgene is inserted between imprinting centres 1 and 2 in the distal MMU7  imprinted domain. It is located 2.6kb distal to the Ins2 gene, which is paternally expressed only in yolk sac and thought to be regulated by IC1 (143; 145). Expression of a GFP gene within Tel7KI was shown to exhibit allele-specific expression in the embryo, with an active maternal allele and an inactive paternal allele, acting as a maternally-expressed imprinted gene. The inactive paternal allele also acquired a high level of DNA methylation marks at its promoter, another characteristic of imprinted genes. A low level of methylation was also observed on the maternal allele, not a common feature of endogenous imprinted genes, but occasionally observed at imprinted transgenes (60). This imprinted pattern was, however, found only in the embryo; in the placenta both maternal and paternal alleles showed expression of the GFP gene.   In the embryo, the extent of GFP expression was analyzed. The construct upon which the  Tel7KI construct was based has been used for “ubiquitous” expression, but that was not observed at this locus. Instead, GFP from the maternal allele was expressed stochastically in most tissues, and both the level and frequency of expression varied from tissue to tissue. The highestexpressing cells were found in the heart and CNS upon maternal transmission. Interestingly, these are the same tissues where loss of silencing of GFP is occasionally observed upon paternal transmission, potentially indicating high promoter usage can overcome the epigenetic silencing in a minority of cells. The yolk sac also showed GFP activation from both alleles, though more frequently from the paternal allele than the maternal. It is unknown whether this activation is due to epigenetic factors but the correlation between yolk sac patches of GFP expression and patches in paternal transmission embryos may imply an epigenetic link.   The only embryonic tissue to show consistent GFP expression upon paternal transmission  of Tel7KI is the developing germ line. Specialized primordial germ cells (PGCs) undergo a period of imprint erasure and reactivation during their specification, resulting in a homogeneous pattern of imprints to pass on to the next generation. The exact timing of this erasure phase is unknown, but it appears to occur between E10.5 and E12.5, and may possibly be different for different genes (128; 129; 131). At these stages, the number of GFP expressing germ cells from 98  Tel7KI increases from 10% to 66%, suggesting a possible relationship between imprint erasure and reactivation of the transgene. Further experiments will examine this possibility.   Another tissue of interest for expression of Tel7KI is the placenta, which shows a  complete lack of imprinting at this locus. Both parental alleles show GFP expression in a similar pattern, characterized by punctate expression throughout the labyrinth and spongiotrophoblast, with bright foci of GFP expression in TGCs. Reflecting this biallelic expression, in DNA preparations from whole E10.5 placentae, moderate methylation is found on both alleles. This lack of imprinting in the placenta is unusual for an endogenously imprinted gene, and not observed at any of the imprinted genes near the insertion site. Analysis of lineage-specification of imprinting was not able to be completed, as Tel7KI is not expressed in epiblast-derived tissues in the placenta, thus it is still unknown whether imprinting is found in all epiblast derivatives.   High levels of GFP expression in TGCs as seen in Tel7KI have been observed previously  from a related X-linked GFP line and been hypothesized to be due to loss of imprinted Xinactivation and, possibly, general epigenetic instability (190). An in vitro culture system to derive TGCs from EPCs was established when it was determined that isolating TGCs from placentae was unfeasible by flow cytometry. This culture system gave rise to a large population of high-ploidy cells staining positively for a TGC marker. Though both the Tel7KI and X-linked D4 lines showed loss of paternal transgene silencing in these cells in vivo no defects of imprinted expression or DNA methylation at any MMU7 imprinted genes were discovered in TGCs grown in vitro. No analysis of X-inactivation was possible in our system, though future experiments may address this issue.   ES cells carrying maternally- or paternally-inherited Tel7KI transgenes (called KIM and  KIP, respectively) were derived, both of which show GFP expression, indicating a postimplantation acquisition of imprinting. To attempt to recapitulate this in vitro and assay the usefulness of the Tel7KI line as a tool to study the acquisition of silencing, the ES cell lines were differentiated into embryoid bodies by LIF withdrawal. KIP ES cells were able to silence their GFP more quickly and to a greater extent than KIM ES cells, and acquired a greater amount of DNA methylation at the promoter driving GFP expression, mimicking the situation in vivo.   Tel7KI is located much closer to IC1 than IC2, however the maternal expression and  paternal DNA methylation observed at the transgene is inconsistent with current models of regulation of IC1-linked genes. Thus, a deletion of IC2 in cis to Tel7KI was created and the ES99  cell differentiation system was used to analyze its effect on the silencing of Tel7KI on the deleted paternal allele. KIP ES cells carrying the deletion in cis as well as the parental KIP cells and KIM cells were differentiated, and their GFP expression pattern and DNA methylation at Tel7KI were examined. The deletion of IC2 impaired the ability of paternal-transmission Tel7KI ES cells to silence their GFP and to acquire DNA methylation marks, implying a role for IC2 in the imprinted regulation of this transgene. This extends the known region of influence for this IC more than 300kb proximal and introduces a potential tool to study acquisition of imprinted expression in the mouse. 6.2 General discussion   Historically, analysis of imprinted transgenes has been crucial in our understanding of the  mechanisms of imprinted genes. These transgenes generally fall into two main categories; alleles created by relocating known imprinted genes and parts of their regulatory sequence to other regions of the genome and transgenes which are found to be imprinted though they bear no relation to nor are they inserted near known imprinted genes (51; 53; 55; 63). While useful, neither of these types of inserted alleles is capable of reporting the epigenetic status of a known imprinted domain in a manner which is non-invasive and allows for easy visualization and isolation of cells. The results reported in this thesis show that Tel7KI fulfills the desired criteria for an imprinted reporter gene; parent-of-origin-specific expression and DNA methylation in a wide range of embryonic tissues, with evidence to suggest that this is a downstream effect of a known imprinted domain.   No evidence exists that the sequences present in the Tel7KI allele contain regulatory  regions of endogenously imprinted genes, nor are they capable of becoming imprinted at other loci. While it is true that sequences flanking one of the loxP sites in the original construct are derived from genomic sequence distal to the Ascl2 gene, it is highly unlikely that these sequences are causing imprinted expression at Tel7KI. Ascl2 is imprinted in the exact opposite manner to Tel7KI, maternally expressed in the placenta and biallelic, though at an extremely low level, in the embryo (239). It is possible that the signals from IC2 which are reaching Tel7KI can do so because of the presence of these Ascl2 sequences, and future experiments are planned to repeat the Tel7KI insertion without these sequences. Alternatively, telomeres have been shown to cause 100  position effects, and Tel7KI contains 1.6kb of telomeric repeats originally intended to function as a seed to recreate a stable telomeric end to the truncated MMU7 (179). Small (50-700bp) remnant interstitial telomeric sequences have been located in the mouse and been hypothesized to be remainders of chromosomal rearrangements or leftovers of telomerase-mediated chromosome repair (240-243). None of the seven interstitial telomeric repeats found in the mouse have been associated with regions showing parent-of-origin effects (240-244). Other than the Ascl2-linked sequences and the telomeric repeats, Tel7KI is based, in the main, on the CAGEGFP reporter which has been used extensively in mouse transgenes (189; 191; 196; 245). The only parent-of-origin effect reported in these studies was a single line in which the CAG-EGFP reporter was inserted on the X chromosome, in which case it became regulated by XCI (190). In one study, 142 mouse lines were generated from CAG-EGFP. Only paternal transmission was ever examined in this line, so it is not known whether any of these lines show imprinted expression, but since GFP expression was observed from all of them upon paternal transmission, it is clear that the paternal silencing observed at Tel7KI is unique to this particular insertion site (ref. 245, Masaru Okabe, personal communication).   The tissue-specificity of Tel7KI, both in the embryo where high levels are found in the  heart and CNS and in the placenta, where GFP is expressed highest in TGCs, has not before been reported with CAG-EGFP-derived lines. Other studies using this construct or the related CAGEYFP to generate mouse lines with randomly-integrated copies of the transgene have found variable levels of expression and tissue-specificity of GFP and YFP in the embryo, but the tissues involved were different from those showing high levels of expression from Tel7KI (196; 245; 246). In addition, little global expression analysis has been done on these lines at later postimplantation stages, so the actual expression pattern in the embryo from these lines is unknown. In the placenta, as mentioned the GFP expression from the D4 X-linked line in TGCs was hypothesized to be due to epigenetic instability rather than transgene-specific effects (190). The analysis presented here shows that these giant cells are in fact completely normal at imprinted loci. It remains possible that X-inactivation is defective, but given the results presented here, it is doubtful. Instead, what appears to be occurring is that the CAG-EGFP transgene is expressed at high levels in TGCs, regardless of the parent-of-origin from which it is inherited. This phenotype may mimic the GFP reactivation observed in paternal Tel7KI embryonic brain and heart, where high levels of promoter activity can overcome epigenetic marks which may be 101  present, or it may be that these epigenetic marks are not acquired in these cells. In support of the first hypothesis, other groups have examined maternal transmission of the D4 line and shown extremely high levels of GFP expression in TGCs (Satoshi Namekawa, personal communication). To determine whether the remainder of the placental Tel7KI phenotype is recapitulated in this line, it would be necessary to examine maternal-transmission placentae, since the paternal X chromosome is preferentially inactivated. Neither this study nor other published studies have examined the extent of GFP expression upon maternal transmission in the placenta and so this remains unknown. Of the other CAG-EGFP and -EYFP lines, little analysis has been performed in the placenta and so no comparisons can be made. Considering both the embryonic and placental patterns and examination of similar transgenic lines, it seems that the expression pattern of Tel7KI may be a compound effect of activity of the CAG promoter and its inserted position. Further analysis of the D4 transgene may lead to a better understanding of this pattern.   If, as this work suggests, Tel7KI is under the regulation of genomic imprinting, GFP  expression from the paternal allele of Tel7KI in PGCs may be due to erasure of those imprints during the imprint reprogramming phase (102; 247). Analysis of imprint erasure during this stage has shown that imprinted genes become demethylated at different stages of this process; for example Igf2 DMR2 and the Igf2r/Air DMR are completely unmethylated by E11.5, while the Peg3 DMR, KvDMR1 (IC2), and the Snrpn DMR still carry some DNA methylation at that stage and are not completely unmethylated until E12.5 (128; 129). The H19 DMR (IC1) appears to be demethyated even later, as it still maintains some DNA methylation at E12.5 (128). Similarly, for another characteristic of epigenetic reprogramming at this stage, reactivation of the inactive X chromosome in female embryos has been shown to occur progressively during germ cell development, rather than in a single concerted step (248; 249). Since the data presented here on Tel7KI expression in germ cells is preliminary, it will be important to perform further experiments to determine whether reactivation of GFP from the paternal allele is indicative of imprint erasure. See the following section 6.3 for details of these proposed experiments.   Based on all correlations between DNA methylation at the β-actin promoter and  expression of GFP, it is possible to estimate the percent methylation required for silencing in different tissues. This is complicated, of course, given that measurement of DNA methylation in whole embryos and placentae are by necessity done in mixed populations, and thus overall 102  percentage methylation is difficult to relate to expression level. In the embryo, KI/+ embryos at E14.5 show 32% methylation and still express GFP at high levels, while at E9.5, silent +/KI embryos have 69% methylation. Sorting of maternal transmission embryos into expressing and non-expressing cells did not clarify this question, as little DNA methylation was present in either population, but it appears that 32% methylation still allows for GFP expression from Tel7KI, while 69% does not. This is supported by the fact that in the placenta up to 49% methylation still allows for GFP expression from the paternal allele, though it is possible that these two tissues have different methylation thresholds. In undifferentiated ES cells, up to 14% methylation is observed with close to 100% GFP-positive cells. Upon differentiation, it is more difficult to make an estimate of the amount of methylation required for silencing, given that the cell population is very heterogeneous. In the future, sorting of GFP-positive from GFP-negative differentiated cells might help to answer this question.   Other studies which have examined the relationship between methylation and expression  levels have found varying results. A study examining the status of MMP-9 expression in lymphoma cell lines found that cells showing 85% methylation do not express the gene, while a reduction to 65% methylation results in a detectable level of gene expression (250). In another study examining expression of a human β-globin transgene inserted in mice, tissues expressing the transgene at high levels showed up to 12% methylation, while tissues with very low transgene expression levels showed as little as 34% methylation (251). These results combined with the data shown here clearly show that the effect of methylation level on expression level varies widely depending on the gene examined.   There are no examples of endogenous genes which are imprinted only in the embryo  proper and not in the placenta. Some transgenes are imprinted only in specific tissues of the embryo, but these phenotypes appear to be related to transgene sequence and not epigenetic factors acting on the alleles (61). One reported transgene showed allele-specific DNA methylation in the embryo, but equivalent DNA methylation on both parental alleles in the placenta and yolk sac (66). This is very similar to what is observed at Tel7KI, with the exception that this transgene was maternally methylated and paternally expressed, opposite to the pattern at Tel7KI.   A possible scenario for the difference in imprinting status of Tel7KI between embryo and  placenta involves the difference between these tissues in the acquisition of DNA methylation and 103  histone modification marks throughout the IC2 region. If, as suggested by the ES cell differentiation experiment, Tel7KI is indeed regulated by IC2, the mechanism by which the signal from this domain spreads a further 300kb proximal to the Tel7KI insertion site may be a factor in this question. Biallelic expression of Tel7KI in the blastocyst, as mentioned, implies a similar regulation to that observed at placentally-imprinted genes, where it is proposed that the signal from IC2 has not yet spread to these distantly-located genes (34). Once the lineages have split into embryonic and extraembryonic components, however, Tel7KI differs from the established program. In the embryo, Tel7KI acquires DNA methylation on its paternal allele and becomes expressed only from the maternal allele, in a similar fashion to Cdkn1c. It is likely that the presence of a CpG island capable of undergoing differential DNA methylation results in the imprinted embryonic expression, since this tissue relies heavily on DNA methylation to silence genes. In the placenta, however, DNA methylation is not relied upon as heavily in this region, though the secondary DMR at Cdkn1c remains differentially methylated in this tissue (119). The lack of imprinted expression of Tel7KI in the placenta also implies a lack of differential histone modifications, which are normally observed at IC2-regulated imprinted genes (31). A possible explanation for this apparent incongruence would be if Tel7KI is capable of acquiring imprinted expression only through differential DNA methylation, and either does not or cannot become differently marked by histone modifications. This will have to be examined to be certain, but if indeed these marks are lacking, either the distance from IC2 or a characteristic of the insertion site may be responsible.   Another possible mechanism by which Tel7KI could acquire differential imprinted  expression in embryo and placenta under the regulation of IC2 involves recent data on subnuclear localization of the paternal MMU7 chromosome. In this study, Kncq1ot1 expressed from the paternal allele was found to localize to a silent chromatin domain (154). This domain is larger in the placenta than in the embryo, consistent with more of the IC2 domain being silenced in the placenta (154). Of particular interest is the localization of Ascl2, the proximal-most imprinted gene in the domain, compared to that of Igf2, located more than 300kb away, on the other side of Tel7KI. In the embryo, Ascl2 and Igf2 are located at approximately the same distance from the edge of the silent domain, while in the placenta, Ascl2 is located significantly closer to the edge (154). This is consistent with the fact that Ascl2 is silenced on the paternal allele in the placenta, and not expressed in the embryo. The implications of this localization for Tel7KI expression are 104  important: in the embryo, Tel7KI is located closer to the silent domain of Kcnq1ot1 than in the placenta. This is another possible explanation for the lack of imprinting of Tel7KI in the placenta; the paternal allele is located too far away to be affected by the silencing signal from IC2.   The extension of the IC2-regulated domain proximal to the location of Tel7KI has  implications for the regulation of other genes in the region. Between the well-defined targets for IC2 regulation and Tel7KI lies the Th gene. This gene has been shown to have preferential maternal expression in E13.5 placenta, but biallelic expression in embryo, fitting with the zone model of IC2 regulation (199). However, since the paternal allele of Th is expressed at detectable levels and this result has not been replicated in the published record, the inclusion of Th as an imprinted gene is not universally accepted (199). Experiments in the Lefebvre lab have indicated that a maternal-specific LTR-driven alternatively spliced transcript specific to the placenta is maternally expressed and possibly responsible for the maternal bias in the placenta (A. Bogutz, unpublished data). Experiments analyzing the expression of this alternative transcript, called Thα, in placentae carrying a deletion of IC2 have so far been inconclusive. If an effect of IC2 on this transcript can be confirmed, this extends the possible region of influence of IC2 a further 75kb closer to Tel7KI. No other known genes are located between Th and Tel7KI.   Further proximal to Tel7KI, the imprinted expression of the Ins2 gene has never been  fully explained by the enhancer competition and chromatin looping model for the regulation of H19 and Igf2. How this paternally-expressed gene imprinted in only a single tissue could be regulated by IC2 is unclear, but given the data implying regulation of Tel7KI by IC2, it is important to analyze the expression of this gene located a mere 2.6kb proximal. The expression of Ins2 in the embryo and yolk sac has been well documented, but expression levels and parentof-origin in the placenta have not been analyzed in as much depth. Current experiments in the Lefebvre lab are examining this question.   This study has conclusively shown that Tel7KI is acting as a maternally-expressed gene,  and that it appears to be regulated from endogenous imprinted loci, most likely by a spreading of paternal inactivation signal from IC2. Tel7KI carries many features of imprinted genes, and its imprinting has the potential to allow us to learn more about the interval between IC1 and IC2. The final section of this thesis will detail some of the experiments to be completed in the future  105  to further characterize Tel7KI and confirm its use as a reporter for genomic imprinting in the mouse.  6.3 Future directions   Here many features of imprinted genes have been analyzed with respect to the imprinted  transgene Tel7KI. There are two other characteristics which have not yet been examined, and further work could give insight into its imprinted regulation. Firstly, no analysis of acquisition of histone modifications has been performed. This could be accomplished through the use of chromatin immunoprecipitation (ChIP) assays, which isolate DNA fragments associated with specific histone modifications. Based on the DNA methylation data presented in Chapters 3 and 4, the hypothesis would be that in the embryo, differential histone modifications would be observed, for example enrichment of active marks on the maternal allele and silencing marks on the paternal allele. In the placenta, however, these differences would be expected to be missing. Secondly, differences in replication timing have been reported for many imprinted genes and transgenes. This can be assayed most easily by fluorescence in-situ hybridization (FISH). By necessity, this analysis must be performed on cells homozygous for Tel7KI. Interphase G2 nuclei are hybridized to a transgene-specific probe, and the number of resulting signals are counted. If asynchronous replication is occurring, then three signals (for one homolog replicated and one not) should be observed more frequently than four or two signals (both replicated or both not).   One intriguing phenotype of Tel7KI which would be interesting to further develop is the  stochastic loss of silencing observed in some cells in the embryo upon paternal transmission. As shown in Figures 3.2 and 3.8, +/KI embryos occasionally show some GFP expression in particular tissues. As discussed, the expression observed in brain and heart is perhaps due to high promoter usage, but loss of silencing in yolk sac and embryo patches is puzzling, particularly given that yolk sac patches are occasionally observed in KI/+ embryos. It would, therefore, be extremely interesting to isolate GFP+ and GFP– cells from these embryos or yolk sacs and examine DNA methylation at Tel7KI as well as imprinted gene expression at other genes on MMU7. As a temporary tissue, imprint stability in the yolk sac is perhaps not as critical as the same analysis in the embryo. It is possible that Tel7KI undergoes stochastic reactivation in some embryos, or it is possible that it is acting as a reporter for loss of imprinting which occurs 106  naturally during development. Since loss of silencing of imprinted genes is generally associated with developmental abnormalities, it seems unlikely that the latter possibility is occurring, but the fact that the cells showing reactivation are not concentrated in a single tissue and appear to be in the minority may allow for wild-type cells to rescue the cells showing loss of imprinting.   Another tissue of interest with respect to this mouse line is the germ line. Preliminary  experiments will determine whether epigenetic or expression differences exist between GFPpositive and GFP-negative germ cell populations of paternal-transmission Tel7KI during the germ cell reprogramming phase. Once it has been determined if the reactivation of GFP is due to DNA demethylation at Tel7KI and if that demethylation is indicative of loss of DNA methylation genome-wide, several potential lines of investigation present themselves. Firstly, it will be helpful in determining the exact timing of imprint erasure in the germ line and the location of germ cells when they undergo erasure. This can be accomplished by crossing the Tel7KI line to a line expressing a different fluorescent marker under, for example, the Oct-4 promoter. Then, thick sagittal sections of embryos can be cultured and the migration of germ cells monitored by two-colour fluorescence microscopy, as previously done by others (252), and the position of these cells when they become GFP+ after imprint erasure can be determined. If, as hypothesized, these cells do not begin erasure until their arrival in the genital ridge, then Tel7KI can be used to examine what signals are required to cause germ cells to begin reprogramming. Next, it should be possible to cross Tel7KI with a mutant line whose germ cells fail to migrate properly to the genital ridge and test whether these cells still erase their imprints. By this model, cells that do not reach the ridge should maintain their silent state of Tel7KI and not erase their imprints. Another experiment might entail culturing paternal transmission Tel7KI germ cells just prior to their reactivation time with genital ridges of different embryonic stages and determining which stages are capable of causing reactivation. A similar strategy has recently been used to study the reactivation of the inactive X chromosome in female embryos during germ cell specification (248). Expression profiling comparing genital ridge cells capable and incapable of causing erasure can be used to determine potential signals. Also, by comparing the germ cells themselves just prior to and after reactivation, it might be possible to identify factors present in these cells that are involved in the reprogramming process.   The ES-cell differentiation-based system used in this study to analyze the effects of an  IC2 deletion on silencing of paternally-inherited Tel7KI is useful, but perhaps not representative 107  of the situation in vivo. To do this analysis more thoroughly, the KIP C4 IC2KO ES cell line will be aggregated with wild-type preimplantation embryos to form chimeric animals. These chimeras will be used to create an animal strain carrying Tel7KI and the IC2KO in cis. Both these strains exist independently as mouse lines, but it was not practical to obtain them in cis by breeding, since they are a mere 0.2cM apart, meaning that 500 F2 offspring would have to be screened to recover an appropriate recombination product. Once a male carrying both alleles in cis is born, he will be mated to a female wild-type animal carrying polymorphisms on MMU7, and GFP expression as well as expression of IC2-regulated imprinted genes will be examined in their progeny. If Tel7KI is indeed regulated by IC2, as suggested by the ES cell differentiation experiment, the normally silent paternal allele should be reactivated, resulting in GFP+ progeny, and this should also be observed at the other IC2-regulated genes, as has been shown previously (155; 179).   Another genetic method to analyze the effect of IC2 on Tel7KI is available. Animals null  for the de novo DNA methyltransferase cofactor gene Dnmt3l are viable, though the males are infertile and the females do not methylate maternally-imprinted ICs in their oocytes (173). This is manifest as loss of methylation at IC2 and biallelic expression of Kcnq1ot1, which in turn suppresses all maternally-expressed genes in cis in the embryo (31). If a female null for Dnmt3l and carrying Tel7KI is mated to a wild-type male, the resulting offspring will all have biallelic Kcnq1ot1, which will result in silencing of Tel7KI in cis if Tel7KI is regulated by IC2. These resulting embryos will therefore not express GFP as normal KI/+ embryos would.   It is also possible to envision an in vitro system using Tel7KI which would allow for  screening of factors involved in imprint maintenance or establishment. To study imprint maintenance, it would be desirable to establish culture of a cell type known to express GFP at high levels in maternal transmission Tel7KI embryos, for instance cells of the heart or CNS. If these cell lines were established from a paternal transmission embryo, they would likely be GFPnegative. Then, the cells could be treated with factors or constructs designed to up- or downregulate a gene of interest, and GFP-reactivation would indicate a role for that gene in maintenance of silencing. Additionally, a similar screen done with ES cells which would then be differentiated as performed in Chapter 5 might identify potential modifiers of imprint establishment. The use of Tel7KI with its imprinted fluorescent reporter in these assays would greatly simplify identification of epigenetic modifiers. 108  References  1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.  12. 13. 14. 15. 16.  Reik, W., and Walter, J. (2001) Genomic imprinting: parental influence on the genome, Nat Rev Genet, 2:21-32. Barton, S. C., Surani, M. A., and Norris, M. L. (1984) Role of paternal and maternal genomes in mouse development, Nature, 311:374-376. Ferguson-Smith, A. C., Cattanach, B. M., Barton, S. C., Beechey, C. V., and Surani, M. A. (1991) Embryological and molecular investigations of parental imprinting on mouse chromosome 7, Nature, 351:667-670. Surani, M. A., Barton, S. C., and Norris, M. L. (1984) Development of reconstituted mouse eggs suggests imprinting of the genome during gametogenesis, Nature, 308:548-550. Morison, I., Ramsay, J. P., and Spencer, H. G. (2005) A census of mammalian imprinting, Trends Genet, 21:457-465. McGrath, J., and Solter, D. (1984) Completion of mouse embryogenesis requires both the maternal and paternal genomes, Cell, 37:179-183. Kawahara, M., Wu, Q., Takahashi, N., Morita, S., Yamada, K., Ito, M., Ferguson-Smith, A. C., and Kono, T. (2007) High-frequency generation of viable mice from engineered bimaternal embryos, Nat Biotechnol, 25:1045-1050. Kono, T., Obata, Y., Wu, Q., Niwa, K., Ono, Y., Yamamoto, Y., Park, E. S., Seo, J. S., and Ogawa, H. (2004) Birth of parthenogenetic mice that can develop to adulthood, Nature, 428:860-864. Cattanach, B. M., and Kirk, M. (1985) Differential activity of maternally and paternally derived chromosome regions in mice, Nature, 315:496-498. McLaughlin, K. J., Szabó, P., Haegel, H., and Mann, J. R. (1996) Mouse embryos with paternal duplication of an imprinted chromosome 7 region die at midgestation and lack placental spongiotrophoblast, Development, 122:265-270. Kikyo, N., Williamson, C. M., John, R. M., Barton, S. C., Beechey, C. V., Ball, S. T., Cattanach, B. M., Surani, M. A., and Peters, J. (1997) Genetic and functional analysis of neuronatin in mice with maternal or paternal duplication of distal Chr 2, Dev Biol, 190:66-77. Nicholls, R. D. (1993) Genomic imprinting and uniparental disomy in Angelman and Prader-Willi syndromes: a review, Am J Med Genet, 46:16-25. Ledbetter, D. H., and Engel, E. (1995) Uniparental disomy in humans: development of an imprinting map and its implications for prenatal diagnosis, Human Molecular Genetics, 4 Spec No:1757-1764. Slatter, R. E., Elliott, M., Welham, K., Carrera, M., Schofield, P. N., Barton, D. E., and Maher, E. R. (1994) Mosaic uniparental disomy in Beckwith-Wiedemann syndrome, J Med Genet, 31:749-753. Bartolomei, M., Zemel, S., and Tilghman, S. (1991) Parental imprinting of the mouse H19 gene, Nature, 351:153-155. DeChiara, T. M., Robertson, E. J., and Efstratiadis, A. (1991) Parental imprinting of the mouse insulin-like growth factor II gene, Cell, 64:849-859.  109  17.
 Barlow, D. P., Stöger, R., Herrmann, B. G., Saito, K., and Schweifer, N. (1991) The mouse insulin-like growth factor type-2 receptor is imprinted and closely linked to the Tme locus, Nature, 349:84-87. 18.
 Luedi, P. P., Hartemink, A. J., and Jirtle, R. L. (2005) Genome-wide prediction of imprinted murine genes, Genome Research, 15:875-884. 19.
 Spahn, L., and Barlow, D. P. (2003) An ICE pattern crystallizes, Nat Genet, 35:11-12. 20.
 Wutz, A., Smrzka, O. W., Schweifer, N., Schellander, K., Wagner, E. F., and Barlow, D. P. (1997) Imprinted expression of the Igf2r gene depends on an intronic CpG island, Nature, 389:745-749. 21.
 Morgan, H. D., Santos, F., Green, K., Dean, W., and Reik, W. (2005) Epigenetic reprogramming in mammals, Human Molecular Genetics, 14 Spec No 1:R47-58. 22.
 Surani, M. A., Kothary, R., Allen, N. D., Singh, P. B., Fundele, R., Ferguson-Smith, A. C., and Barton, S. C. (1990) Genome imprinting and development in the mouse, Dev Suppl, 89-98. 23.
 Wilkins, J. F. (2005) Genomic imprinting and methylation: epigenetic canalization and conflict, Trends Genet, 21:356-365. 24.
 Constância, M., Dean, W., Lopes, S., Moore, T., Kelsey, G., and Reik, W. (2000) Deletion of a silencer element in Igf2 results in loss of imprinting independent of H19, Nat Genet, 26:203-206. 25.
 Murrell, A., Heeson, S., Bowden, L., Constância, M., Dean, W., Kelsey, G., and Reik, W. (2001) An intragenic methylated region in the imprinted Igf2 gene augments transcription, EMBO Rep, 2:1101-1106. 26.
 Sleutels, F., Tjon, G., Ludwig, T., and Barlow, D. P. (2003) Imprinted silencing of Slc22a2 and Slc22a3 does not need transcriptional overlap between Igf2r and Air, EMBO J, 22:3696-3704. 27.
 Stöger, R., Kubicka, P., Liu, C. G., Kafri, T., Razin, A., Cedar, H., and Barlow, D. P. (1993) Maternal-specific methylation of the imprinted mouse Igf2r locus identifies the expressed locus as carrying the imprinting signal, Cell, 73:61-71. 28.
 Li, E., Bestor, T. H., and Jaenisch, R. (1992) Targeted mutation of the DNA methyltransferase gene results in embryonic lethality, Cell, 69:915-926. 29.
 Li, E., Beard, C., and Jaenisch, R. (1993) Role for DNA methylation in genomic imprinting, Nature, 366:362-365. 30.
 Biniszkiewicz, D., Gribnau, J., Ramsahoye, B., Gaudet, F., Eggan, K., Humpherys, D., Mastrangelo, M. A., Jun, Z., Walter, J., and Jaenisch, R. (2002) Dnmt1 overexpression causes genomic hypermethylation, loss of imprinting, and embryonic lethality, Mol Cell Biol, 22:2124-2135. 31.
 Lewis, A., Mitsuya, K., Umlauf, D., Smith, P., Dean, W., Walter, J., Higgins, M., Feil, R., and Reik, W. (2004) Imprinting on distal chromosome 7 in the placenta involves repressive histone methylation independent of DNA methylation, Nat Genet, 36:1291-1295. 32.
 Fournier, C., Goto, Y., Ballestar, E., Delaval, K., Hever, A. M., Esteller, M., and Feil, R. (2002) Allele-specific histone lysine methylation marks regulatory regions at imprinted mouse genes, EMBO J, 21:6560-6570. 33.
 Kouzarides, T. (2007) Chromatin modifications and their function, Cell, 128:693-705. 34.
 Lewis, A., Green, K., Dawson, C., Redrup, L., Huynh, K., Lee, J., Hemberger, M., and Reik, W. (2006) Epigenetic dynamics of the Kcnq1 imprinted domain in the early embryo, Development, 133:4203-4210. 110  35.
 Grandjean, V., O'Neill, L., Sado, T., Turner, B., and Ferguson-Smith, A. (2001) Relationship between DNA methylation, histone H4 acetylation and gene expression in the mouse imprinted Igf2-H19 domain, FEBS Letters, 488:165-169. 36.
 Yang, Y., Li, T., Vu, T. H., Ulaner, G. A., Hu, J., and Hoffman, A. R. (2003) The histone code regulating expression of the imprinted mouse Igf2r gene, Endocrinology, 144:5658-5670. 37.
 Mager, J., Montgomery, N., De Villena, F., and Magnuson, T. (2003) Genome imprinting regulated by the mouse Polycomb group protein Eed, Nat Genet, 33:502-507. 38.
 Xin, Z., Tachibana, M., Guggiari, M., Heard, E., Shinkai, Y., and Wagstaff, J. (2003) Role of histone methyltransferase G9a in CpG methylation of the Prader-Willi syndrome imprinting center, J Biol Chem, 278:14996-15000. 39.
 Regha, K., Sloane, M. A., Huang, R., Pauler, F. M., Warczok, K. E., Melikant, B., Radolf, M., Martens, J. H., Schotta, G., Jenuwein, T., and Barlow, D. P. (2007) Active and repressive chromatin are interspersed without spreading in an imprinted gene cluster in the mammalian genome, Molecular Cell, 27:353-366. 40.
 Kitsberg, D., Selig, S., Brandeis, M., Simon, I., Keshet, I., Driscoll, D. J., Nicholls, R. D., and Cedar, H. (1993) Allele-specific replication timing of imprinted gene regions, Nature, 364:459-463. 41.
 Knoll, J. H., Cheng, S. D., and Lalande, M. (1994) Allele specificity of DNA replication timing in the Angelman/Prader-Willi syndrome imprinted chromosomal region, Nat Genet, 6:41-46. 42.
 Windham, C. Q., and Jones, P. A. (1997) Expression of H19 does not influence the timing of replication of the Igf2/H19 imprinted region, Dev Genet, 20:29-35. 43.
 Kagotani, K., Takebayashi, S., Kohda, A., Taguchi, H., Paulsen, M., Walter, J., Reik, W., and Okumura, K. (2002) Replication timing properties within the mouse distal chromosome 7 imprinting cluster, Biosci Biotechnol Biochem, 66:1046-1051. 44.
 Kawame, H., Gartler, S. M., and Hansen, R. S. (1995) Allele-specific replication timing in imprinted domains: absence of asynchrony at several loci, Human Molecular Genetics, 4:2287-2293. 45.
 Simon, I., Tenzen, T., Reubinoff, B. E., Hillman, D., McCarrey, J. R., and Cedar, H. (1999) Asynchronous replication of imprinted genes is established in the gametes and maintained during development, Nature, 401:929-932. 46.
 Wood, A., Roberts, R., Monk, D., Moore, G., Schulz, R., and Oakey, R. (2007) A screen for retrotransposed imprinted genes reveals an association between X chromosome homology and maternal germ-line methylation, PLoS Genetics, 3:e20. 47.
 Kobayashi, S., Kohda, T., Ichikawa, H., Ogura, A., Ohki, M., Kaneko-Ishino, T., and Ishino, F. (2002) Paternal expression of a novel imprinted gene, Peg12/Frat3, in the mouse 7C region homologous to the Prader-Willi syndrome region, Biochemical and Biophysical Research Communications, 290:403-408. 48.
 Paulsen, M., El-Maarri, O., Engemann, S., Strödicke, M., Franck, O., Davies, K., Reinhardt, R., Reik, W., and Walter, J. (2000) Sequence conservation and variability of imprinting in the Beckwith-Wiedemann syndrome gene cluster in human and mouse, Hum Mol Genet, 9:1829-1841. 49.
 Monk, D., Wagschal, A., Arnaud, P., Müller, P. S., Parker-Katiraee, L., Bourc'his, D., Scherer, S. W., Feil, R., Stanier, P., and Moore, G. E. (2008) Comparative analysis of human chromosome 7q21 and mouse proximal chromosome 6 reveals a placental-specific 111  50. 51. 52. 53. 54. 55.  56. 57. 58. 59. 60. 61. 62. 63. 64.  imprinted gene, TFPI2/Tfpi2, which requires EHMT2 and EED for allelic-silencing, Genome Research, 18:1270-1281. Chaillet, J. (1994) Genomic imprinting: lessons from mouse transgenes, Mutat Res, 307:441-449. Cranston, M. J., Spinka, T. L., Elson, D. A., and Bartolomei, M. S. (2001) Elucidation of the minimal sequence required to imprint H19 transgenes, Genomics, 73:98-107. Kantor, B., Makedonski, K., Green-Finberg, Y., Shemer, R., and Razin, A. (2004) Control elements within the PWS/AS imprinting box and their function in the imprinting process, Human Molecular Genetics, 13:751-762. Swain, J. L., Stewart, T. A., and Leder, P. (1987) Parental legacy determines methylation and expression of an autosomal transgene: a molecular mechanism for parental imprinting, Cell, 50:719-727. Arima, T., Yamasaki, K., John, R. M., Kato, K., Sakumi, K., Nakabeppu, Y., Wake, N., and Kono, T. (2006) The human HYMAI/PLAGL1 differentially methylated region acts as an imprint control region in mice, Genomics, 88:650-658. John, R. M., Aparicio, S. A., Ainscough, J. F., Arney, K. L., Khosla, S., Hawker, K., Hilton, K. J., Barton, S. C., and Surani, M. A. (2001) Imprinted expression of neuronatin from modified BAC transgenes reveals regulation by distinct and distant enhancers, Dev Biol, 236:387-399. Jones, B. K., Levorse, J., and Tilghman, S. M. (2002) A human H19 transgene exhibits impaired paternal-specific imprint acquisition and maintenance in mice, Human Molecular Genetics, 11:411-418. Kantor, B., Kaufman, Y., Makedonski, K., Razin, A., and Shemer, R. (2004) Establishing the epigenetic status of the Prader-Willi/Angelman imprinting center in the gametes and embryo, Human Molecular Genetics, 13:2767-2779. Yevtodiyenko, A., Steshina, E., Farner, S. C., Levorse, J. M., and Schmidt, J. (2004) A 178kb BAC transgene imprints the mouse Gtl2 gene and localizes tissue-specific regulatory elements, Genomics, 84:277-287. Hatada, I., Nabetani, A., Arai, Y., Ohishi, S., Suzuki, M., Miyabara, S., Nishimune, Y., and Mukai, T. (1997) Aberrant methylation of an imprinted gene U2af1-rs1(SP2) caused by its own transgene, J Biol Chem, 272:9120-9122. Chaillet, J. R., Vogt, T. F., Beier, D. R., and Leder, P. (1991) Parental-specific methylation of an imprinted transgene is established during gametogenesis and progressively changes during embryogenesis, Cell, 66:77-83. Chaillet, J. R., Bader, D. S., and Leder, P. (1995) Regulation of genomic imprinting by gametic and embryonic processes, Genes & Development, 9:1177-1187. Shuster, M., Dhar, M. S., Olins, A. L., Olins, D. E., Howell, C. Y., Gollin, S. M., and Chaillet, J. (1998) Parental alleles of an imprinted mouse transgene replicate synchronously, Dev Genet, 23:275-284. Sasaki, H., Hamada, T., Ueda, T., Seki, R., Higashinakagawa, T., and Sakaki, Y. (1991) Inherited type of allelic methylation variations in a mouse chromosome region where an integrated transgene shows methylation imprinting, Development, 111:573-581. Ueda, T., Yamazaki, K., Suzuki, R., Fujimoto, H., Sasaki, H., Sakaki, Y., and Higashinakagawa, T. (1992) Parental methylation patterns of a transgenic locus in adult somatic tissues are imprinted during gametogenesis, Development, 116:831-839.  112  65.
 Sapienza, C., Peterson, A. C., Rossant, J., and Balling, R. (1987) Degree of methylation of transgenes is dependent on gamete of origin, Nature, 328:251-254. 66.
 Reik, W., Collick, A., Norris, M. L., Barton, S. C., and Surani, M. A. (1987) Genomic imprinting determines methylation of parental alleles in transgenic mice, Nature, 328:248-251. 67.
 Hadchouel, M., Farza, H., Simon, D., Tiollais, P., and Pourcel, C. (1987) Maternal inhibition of hepatitis B surface antigen gene expression in transgenic mice correlates with de novo methylation, Nature, 329:454-456. 68.
 Pask, A., Papenfuss, A. T., Ager, E. I., McColl, K. A., Speed, T. P., and Renfree, M. (2009) Analysis of the platypus genome suggests a transposon origin for mammalian imprinting, Genome Biol, 10:R1. 69.
 Killian, J. K., Byrd, J. C., Jirtle, J. V., Munday, B. L., Stoskopf, M. K., MacDonald, R. G., and Jirtle, R. L. (2000) M6P/IGF2R imprinting evolution in mammals, Molecular Cell, 5:707-716. 70.
 Killian, J. K., Nolan, C. M., Stewart, N., Munday, B. L., Andersen, N. A., Nicol, S., and Jirtle, R. L. (2001) Monotreme IGF2 expression and ancestral origin of genomic imprinting, J Exp Zool, 291:205-212. 71.
 Renfree, M., Hore, T. A., Shaw, G., Graves, J. A., and Pask, A. (2009) Evolution of genomic imprinting: insights from marsupials and monotremes, Annual review of genomics and human genetics, 10:241-262. 72.
 Ager, E., Suzuki, S., Pask, A., Shaw, G., Ishino, F., and Renfree, M. (2007) Insulin is imprinted in the placenta of the marsupial, Macropus eugenii, Dev Biol, 309:317-328. 73.
 Smits, G., Mungall, A., Griffiths-Jones, S., Smith, P., Beury, D., Matthews, L., Rogers, J., Pask, A., Shaw, G., Vandeberg, J., Mccarrey, J., Consortium, S. A. V. O. I. R., Renfree, M., Reik, W., and Dunham, I. (2008) Conservation of the H19 noncoding RNA and H19-IGF2 imprinting mechanism in therians, Nat Genet, 40:971-976. 74.
 Suzuki, S., Renfree, M., Pask, A., Shaw, G., Kobayashi, S., Kohda, T., Kaneko-Ishino, T., and Ishino, F. (2005) Genomic imprinting of IGF2, p57(KIP2) and PEG1/MEST in a marsupial, the tammar wallaby, Mech Dev, 122:213-222. 75.
 Suzuki, S., Ono, R., Narita, T., Pask, A., Shaw, G., Wang, C., Kohda, T., Alsop, A. E., Marshall Graves, J. A., Kohara, Y., Ishino, F., Renfree, M., and Kaneko-Ishino, T. (2007) Retrotransposon silencing by DNA methylation can drive mammalian genomic imprinting, PLoS Genet, 3:e55. 76.
 Varmuza, S., and Mann, M. (1994) Genomic imprinting--defusing the ovarian time bomb, Trends Genet, 10:118-123. 77.
 Bressan, F. F., De Bem, T. H., Perecin, F., Lopes, F. L., Ambrosio, C. E., Meirelles, F. V., and Miglino, M. A. (2009) Unearthing the roles of imprinted genes in the placenta, Placenta, 30:823-834. 78.
 Charalambous, M., da Rocha, S. T., and Ferguson-Smith, A. C. (2007) Genomic imprinting, growth control and the allocation of nutritional resources: consequences for postnatal life, Current opinion in endocrinology, diabetes, and obesity, 14:3-12. 79.
 Coan, P. M., Burton, G. J., and Ferguson-Smith, A. C. (2005) Imprinted genes in the placenta--a review, Placenta, 26 Suppl A:S10-20. 80.
 Georgiades, P., Ferguson-Smith, A. C., and Burton, G. J. (2002) Comparative developmental anatomy of the murine and human definitive placentae, Placenta, 23:3-19.  113  81.
 Charalambous, M., Smith, F. M., Bennett, W. R., Crew, T. E., Mackenzie, F., and Ward, A. (2003) Disruption of the imprinted Grb10 gene leads to disproportionate overgrowth by an Igf2-independent mechanism, Proc Natl Acad Sci USA, 100:8292-8297. 82.
 Salas, M., John, R., Saxena, A., Barton, S., Frank, D., Fitzpatrick, G., Higgins, M., and Tycko, B. (2004) Placental growth retardation due to loss of imprinting of Phlda2, Mech Dev, 121:1199-1210. 83.
 Guillemot, F., Nagy, A., Auerbach, A., Rossant, J., and Joyner, A. L. (1994) Essential role of Mash-2 in extraembryonic development, Nature, 371:333-336. 84.
 Ono, R., Nakamura, K., Inoue, K., Naruse, M., Usami, T., Wakisaka-Saito, N., Hino, T., Suzuki-Migishima, R., Ogonuki, N., Miki, H., Kohda, T., Ogura, A., Yokoyama, M., Kaneko-Ishino, T., and Ishino, F. (2006) Deletion of Peg10, an imprinted gene acquired from a retrotransposon, causes early embryonic lethality, Nat Genet, 38:101-106. 85.
 DeChiara, T. M., Efstratiadis, A., and Robertson, E. J. (1990) A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting, Nature, 345:78-80. 86.
 Frank, D., Fortino, W., Clark, L., Musalo, R., Wang, W., Saxena, A., Li, C. M., Reik, W., Ludwig, T., and Tycko, B. (2002) Placental overgrowth in mice lacking the imprinted gene Ipl, Proc Natl Acad Sci USA, 99:7490-7495. 87.
 Barlow, D. P. (1993) Methylation and imprinting: from host defense to gene regulation?, Science, 260:309-310. 88.
 Haig, D. (1994) Refusing the ovarian time bomb, Trends Genet, 10:346-7; author reply 348-9. 89.
 Moore, T. (1994) Refusing the ovarian time bomb, Trends Genet, 10:347-349. 90.
 Solter, D. (1994) Refusing the ovarian time bomb, Trends Genet, 10:346; author reply 348-346; author reply 349. 91.
 Haig, D. (1996) Altercation of generations: genetic conflicts of pregnancy, Am J Reprod Immunol, 35:226-232. 92.
 Moore, T., and Reik, W. (1996) Genetic conflict in early development: parental imprinting in normal and abnormal growth, Rev Reprod, 1:73-77. 93.
 Moore, T., and Haig, D. (1991) Genomic imprinting in mammalian development: a parental tug-of-war, Trends Genet, 7:45-49. 94.
 Brandeis, M., Kafri, T., Ariel, M., Chaillet, J., McCarrey, J., Razin, A., and Cedar, H. (1993) The ontogeny of allele-specific methylation associated with imprinted genes in the mouse, EMBO J, 12:3669-3677. 95.
 Leighton, P. A., Saam, J. R., Ingram, R. S., Stewart, C., and Tilghman, S. (1995) An enhancer deletion affects both H19 and Igf2 expression, Genes & Development, 9:2079-2089. 96.
 Rechler, M. M., and Nissley, S. P. (1985) The nature and regulation of the receptors for insulin-like growth factors, Annu Rev Physiol, 47:425-442. 97.
 Chai, J. H., Locke, D. P., Ohta, T., Greally, J. M., and Nicholls, R. D. (2001) Retrotransposed genes such as Frat3 in the mouse Chromosome 7C Prader-Willi syndrome region acquire the imprinted status of their insertion site, Mamm Genome, 12:813-821. 98.
 Oakes, C. C., La Salle, S., Smiraglia, D. J., Robaire, B., and Trasler, J. M. (2007) Developmental acquisition of genome-wide DNA methylation occurs prior to meiosis in male germ cells, Dev Biol, 307:368-379.  114  99.
 Lucifero, D., Mertineit, C., Clarke, H. J., Bestor, T. H., and Trasler, J. M. (2002) Methylation dynamics of imprinted genes in mouse germ cells, Genomics, 79:530-538. 100.
 Kafri, T., Ariel, M., Brandeis, M., Shemer, R., Urven, L., McCarrey, J., Cedar, H., and Razin, A. (1992) Developmental pattern of gene-specific DNA methylation in the mouse embryo and germ line, Genes & Development, 6:705-714. 101.
 Hiura, H., Obata, Y., Komiyama, J., Shirai, M., and Kono, T. (2006) Oocyte growthdependent progression of maternal imprinting in mice, Genes Cells, 11:353-361. 102.
 Davis, T. L., Trasler, J. M., Moss, S. B., Yang, G. J., and Bartolomei, M. S. (1999) Acquisition of the H19 methylation imprint occurs differentially on the parental alleles during spermatogenesis, Genomics, 58:18-28. 103.
 Ward, W. S., and Coffey, D. S. (1991) DNA packaging and organization in mammalian spermatozoa: comparison with somatic cells, Biol Reprod, 44:569-574. 104.
 Hammoud, S. S., Nix, D. A., Zhang, H., Purwar, J., Carrell, D. T., and Cairns, B. R. (2009) Distinctive chromatin in human sperm packages genes for embryo development, Nature, 460:473-478. 105.
 Glaser, R. L., Ramsay, J. P., and Morison, I. (2006) The imprinted gene and parent-oforigin effect database now includes parental origin of de novo mutations, Nucleic Acids Research, 34:D29-31. 106.
 Morison, I. M., Paton, C. J., and Cleverley, S. D. (2001) The imprinted gene and parent-oforigin effect database, Nucleic Acids Research, 29:275-276. 107.
 Wykes, S. M., and Krawetz, S. A. (2003) The structural organization of sperm chromatin, J Biol Chem, 278:29471-29477. 108.
 Mayer, W., Niveleau, A., Walter, J., Fundele, R., and Haaf, T. (2000) Demethylation of the zygotic paternal genome, Nature, 403:501-502. 109.
 Oswald, J., Engemann, S., Lane, N., Mayer, W., Olek, A., Fundele, R., Dean, W., Reik, W., and Walter, J. (2000) Active demethylation of the paternal genome in the mouse zygote, Curr Biol, 10:475-478. 110.
 Santos, F., Hendrich, B., Reik, W., and Dean, W. (2002) Dynamic reprogramming of DNA methylation in the early mouse embryo, Dev Biol, 241:172-182. 111.
 Mertineit, C., Yoder, J. A., Taketo, T., Laird, D. W., Trasler, J. M., and Bestor, T. H. (1998) Sex-specific exons control DNA methyltransferase in mammalian germ cells, Development, 125:889-897. 112.
 Cirio, M. C., Martel, J., Mann, M., Toppings, M., Bartolomei, M., Trasler, J., and Chaillet, J. (2008) DNA methyltransferase 1o functions during preimplantation development to preclude a profound level of epigenetic variation, Dev Biol, 324:139-150. 113.
 Nakamura, T., Arai, Y., Umehara, H., Masuhara, M., Kimura, T., Taniguchi, H., Sekimoto, T., Ikawa, M., Yoneda, Y., Okabe, M., Tanaka, S., Shiota, K., and Nakano, T. (2007) PGC7/ Stella protects against DNA demethylation in early embryogenesis, Nat Cell Biol, 9:64-71. 114.
 Li, X., Ito, M., Zhou, F., Youngson, N., Zuo, X., Leder, P., and Ferguson-Smith, A. C. (2008) A maternal-zygotic effect gene, Zfp57, maintains both maternal and paternal imprints, Dev Cell, 15:547-557. 115.
 Bhogal, B., Arnaudo, A., Dymkowski, A., Best, A., and Davis, T. L. (2004) Methylation at mouse Cdkn1c is acquired during postimplantation development and functions to maintain imprinted expression, Genomics, 84:961-970. 116.
 Pauler, F. M., Koerner, M. V., and Barlow, D. P. (2007) Silencing by imprinted noncoding RNAs: is transcription the answer?, Trends Genet, 23:284-292. 115  117.
 Ripoche, M. A., Kress, C., Poirier, F., and Dandolo, L. (1997) Deletion of the H19 transcription unit reveals the existence of a putative imprinting control element, Genes & Development, 11:1596-1604. 118.
 Schuster-Gossler, K., Bilinski, P., Sado, T., Ferguson-Smith, A., and Gossler, A. (1998) The mouse Gtl2 gene is differentially expressed during embryonic development, encodes multiple alternatively spliced transcripts, and may act as an RNA, Dev Dyn, 212:214-228. 119.
 Shin, J. Y., Fitzpatrick, G. V., and Higgins, M. J. (2008) Two distinct mechanisms of silencing by the KvDMR1 imprinting control region, EMBO J, 27:168-178. 120.
 Sleutels, F., Zwart, R., and Barlow, D. P. (2002) The non-coding Air RNA is required for silencing autosomal imprinted genes, Nature, 415:810-813. 121.
 Schütt, S., Florl, A. R., Shi, W., Hemberger, M., Orth, A., Otto, S., Schulz, W. A., and Fundele, R. H. (2003) DNA methylation in placentas of interspecies mouse hybrids, Genetics, 165:223-228. 122.
 Rossant, J., Sanford, J. P., Chapman, V. M., and Andrews, G. K. (1986) Undermethylation of structural gene sequences in extraembryonic lineages of the mouse, Dev Biol, 117:567-573. 123.
 Lawson, K. A., Dunn, N. R., Roelen, B. A., Zeinstra, L. M., Davis, A. M., Wright, C. V., Korving, J. P., and Hogan, B. L. (1999) Bmp4 is required for the generation of primordial germ cells in the mouse embryo, Genes & Development, 13:424-436. 124.
 Lawson, K. A., and Hage, W. J. (1994) Clonal analysis of the origin of primordial germ cells in the mouse, Ciba Found Symp, 182:68-84; discussion 84-91. 125.
 Anderson, R., Copeland, T. K., Schöler, H., Heasman, J., and Wylie, C. (2000) The onset of germ cell migration in the mouse embryo, Mech Dev, 91:61-68. 126.
 Clark, J. M., and Eddy, E. M. (1975) Fine structural observations on the origin and associations of primordial germ cells of the mouse, Dev Biol, 47:136-155. 127.
 Sasaki, H., and Matsui, Y. (2008) Epigenetic events in mammalian germ-cell development: reprogramming and beyond, Nat Rev Genet, 9:129-140. 128.
 Hajkova, P., Erhardt, S., Lane, N., Haaf, T., El-Maarri, O., Reik, W., Walter, J., and Surani, M. A. (2002) Epigenetic reprogramming in mouse primordial germ cells, Mech Dev, 117:15-23. 129.
 Sato, S., Yoshimizu, T., Sato, E., and Matsui, Y. (2003) Erasure of methylation imprinting of Igf2r during mouse primordial germ-cell development, Mol. Reprod. Dev., 65:41-50. 130.
 Yamazaki, Y., Low, E. W., Marikawa, Y., Iwahashi, K., Bartolomei, M. S., Mccarrey, J., and Yanagimachi, R. (2005) Adult mice cloned from migrating primordial germ cells, Proc Natl Acad Sci USA, 102:11361-11366. 131.
 Yamazaki, Y., Mann, M. R., Lee, S. S., Marh, J., Mccarrey, J., Yanagimachi, R., and Bartolomei, M. S. (2003) Reprogramming of primordial germ cells begins before migration into the genital ridge, making these cells inadequate donors for reproductive cloning, Proc Natl Acad Sci USA, 100:12207-12212. 132.
 Li, J. Y., Lees-Murdock, D. J., Xu, G. L., and Walsh, C. P. (2004) Timing of establishment of paternal methylation imprints in the mouse, Genomics, 84:952-960. 133.
 Obata, Y., and Kono, T. (2002) Maternal primary imprinting is established at a specific time for each gene throughout oocyte growth, J Biol Chem, 277:5285-5289. 134.
 Davis, T. L., Tremblay, K. D., and Bartolomei, M. (1998) Imprinted expression and methylation of the mouse H19 gene are conserved in extraembryonic lineages, Dev Genet, 23:111-118. 116  135.
 Lefebvre, L., Mar, L., Bogutz, A., Oh-McGinnis, R., Mandegar, M. A., Paderova, J., Gertsenstein, M., Squire, J. A., and Nagy, A. (2009) The interval between Ins2 and Ascl2 is dispensable for imprinting centre function in the murine Beckwith-Wiedemann region, Human Molecular Genetics, 18:4255-4267. 136.
 Sasaki, H., Ferguson-Smith, A. C., Shum, A. S., Barton, S. C., and Surani, M. A. (1995) Temporal and spatial regulation of H19 imprinting in normal and uniparental mouse embryos, Development, 121:4195-4202. 137.
 Cai, X., and Cullen, B. R. (2007) The imprinted H19 noncoding RNA is a primary microRNA precursor, RNA, 13:313-316. 138.
 Yoshimizu, T., Miroglio, A., Ripoche, M. A., Gabory, A., Vernucci, M., Riccio, A., Colnot, S., Godard, C., Terris, B., Jammes, H., and Dandolo, L. (2008) The H19 locus acts in vivo as a tumor suppressor, Proc Natl Acad Sci USA, 105:12417-12422. 139.
 Filson, A. J., Louvi, A., Efstratiadis, A., and Robertson, E. J. (1993) Rescue of the Tassociated maternal effect in mice carrying null mutations in Igf-2 and Igf2r, two reciprocally imprinted genes, Development, 118:731-736. 140.
 Feil, R., Walter, J., Allen, N. D., and Reik, W. (1994) Developmental control of allelic methylation in the imprinted mouse Igf2 and H19 genes, Development, 120:2933-2943. 141.
 Moore, T., Constancia, M., Zubair, M., Bailleul, B., Feil, R., Sasaki, H., and Reik, W. (1997) Multiple imprinted sense and antisense transcripts, differential methylation and tandem repeats in a putative imprinting control region upstream of mouse Igf2, Proc Natl Acad Sci USA, 94:12509-12514. 142.
 Lopes, S., Lewis, A., Hajkova, P., Dean, W., Oswald, J., Forné, T., Murrell, A., Constância, M., Bartolomei, M., Walter, J., and Reik, W. (2003) Epigenetic modifications in an imprinting cluster are controlled by a hierarchy of DMRs suggesting long-range chromatin interactions, Human Molecular Genetics, 12:295-305. 143.
 Deltour, L., Montagutelli, X., Guenet, J. L., Jami, J., and Páldi, A. (1995) Tissue- and developmental stage-specific imprinting of the mouse proinsulin gene, Ins2, Developmental Biology, 168:686-688. 144.
 Giddings, S. J., King, C. D., Harman, K. W., Flood, J. F., and Carnaghi, L. R. (1994) Allele specific inactivation of insulin 1 and 2, in the mouse yolk sac, indicates imprinting, Nat Genet, 6:310-313. 145.
 Leighton, P. A., Ingram, R. S., Eggenschwiler, J., Efstratiadis, A., and Tilghman, S. M. (1995) Disruption of imprinting caused by deletion of the H19 gene region in mice, Nature, 375:34-39. 146.
 Thorvaldsen, J. L., Duran, K. L., and Bartolomei, M. S. (1998) Deletion of the H19 differentially methylated domain results in loss of imprinted expression of H19 and Igf2, Genes & Development, 12:3693-3702. 147.
 Bartolomei, M., Webber, A., Brunkow, M., and Tilghman, S. (1993) Epigenetic mechanisms underlying the imprinting of the mouse H19 gene, Genes & Development, 7:1663-1673. 148.
 Bell, A. C., and Felsenfeld, G. (2000) Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene, Nature, 405:482-485. 149.
 Hark, A. T., Schoenherr, C. J., Katz, D. J., Ingram, R. S., Levorse, J. M., and Tilghman, S. (2000) CTCF mediates methylation-sensitive enhancer-blocking activity at the H19/Igf2 locus, Nature, 405:486-489.  117  150.
 Engel, N., Thorvaldsen, J. L., and Bartolomei, M. (2006) CTCF binding sites promote transcription initiation and prevent DNA methylation on the maternal allele at the imprinted H19/Igf2 locus, Human Molecular Genetics, 15:2945-2954. 151.
 Fedoriw, A. M., Stein, P., Svoboda, P., Schultz, R. M., and Bartolomei, M. S. (2004) Transgenic RNAi reveals essential function for CTCF in H19 gene imprinting, Science, 303:238-240. 152.
 Dekker, J., Rippe, K., Dekker, M., and Kleckner, N. (2002) Capturing chromosome conformation, Science, 295:1306-1311. 153.
 Murrell, A., Heeson, S., and Reik, W. (2004) Interaction between differentially methylated regions partitions the imprinted genes Igf2 and H19 into parent-specific chromatin loops, Nat Genet, 36:889-893. 154.
 Redrup, L., Branco, M. R., Perdeaux, E. R., Krueger, C., Lewis, A., Santos, F., Nagano, T., Cobb, B. S., Fraser, P., and Reik, W. (2009) The long noncoding RNA Kcnq1ot1 organises a lineage-specific nuclear domain for epigenetic gene silencing, Development, 136:525-530. 155.
 Fitzpatrick, G., Soloway, P., and Higgins, M. (2002) Regional loss of imprinting and growth deficiency in mice with a targeted deletion of KvDMR1, Nat Genet, 32:426-431. 156.
 Gould, T. D., and Pfeifer, K. (1998) Imprinting of mouse Kvlqt1 is developmentally regulated, Human Molecular Genetics, 7:483-487. 157.
 Horike, S., Mitsuya, K., Meguro, M., Kotobuki, N., Kashiwagi, A., Notsu, T., Schulz, T. C., Shirayoshi, Y., and Oshimura, M. (2000) Targeted disruption of the human LIT1 locus defines a putative imprinting control element playing an essential role in BeckwithWiedemann syndrome, Hum Mol Genet, 9:2075-2083. 158.
 Mancini-Dinardo, D., Steele, S. J., Levorse, J., Ingram, R. S., and Tilghman, S. (2006) Elongation of the Kcnq1ot1 transcript is required for genomic imprinting of neighboring genes, Genes & Development, 20:1268-1282. 159.
 Engemann, S., Strödicke, M., Paulsen, M., Franck, O., Reinhardt, R., Lane, N., Reik, W., and Walter, J. (2000) Sequence and functional comparison in the Beckwith-Wiedemann region: implications for a novel imprinting centre and extended imprinting, Human Molecular Genetics, 9:2691-2706. 160.
 Tanaka, M., Puchyr, M., Gertsenstein, M., Harpal, K., Jaenisch, R., Rossant, J., and Nagy, A. (1999) Parental origin-specific expression of Mash2 is established at the time of implantation with its imprinting mechanism highly resistant to genome-wide demethylation, Mech Dev, 87:129-142. 161.
 Fitzpatrick, G., Pugacheva, E. M., Shin, J. Y., Abdullaev, Z., Yang, Y., Khatod, K., Lobanenkov, V. V., and Higgins, M. (2007) Allele-specific binding of CTCF to the multipartite imprinting control region KvDMR1, Mol Cell Biol, 27:2636-2647. 162.
 Fan, T., Hagan, J. P., Kozlov, S. V., Stewart, C. L., and Muegge, K. (2005) Lsh controls silencing of the imprinted Cdkn1c gene, Development, 132:635-644. 163.
 Paulsen, M., Davies, K. R., Bowden, L. M., Villar, A. J., Franck, O., Fuermann, M., Dean, W. L., Moore, T. F., Rodrigues, N., Davies, K. E., Hu, R. J., Feinberg, A. P., Maher, E. R., Reik, W., and Walter, J. (1998) Syntenic organization of the mouse distal chromosome 7 imprinting cluster and the Beckwith-Wiedemann syndrome region in chromosome 11p15.5, Human Molecular Genetics, 7:1149-1159. 164.
 Rossignol, S., Netchine, I., Le Bouc, Y., and Gicquel, C. (2008) Epigenetics in SilverRussell syndrome, Best Pract Res Clin Endocrinol Metab, 22:403-414. 118  165.
 Russell, A. (1954) A syndrome of intra-uterine dwarfism recognizable at birth with craniofacial dysostosis, disproportionately short arms, and other anomalies (5 examples), Proc R Soc Med, 47:1040-1044. 166.
 Silver, H. K., Kiyatsu, W., George, J., and Deamer, W. C. (1953) Syndrome of congenital hemihypertrophy, shortness of stature, and elevated urinary gonadotropins, Pediatrics, 12:368-376. 167.
 Enklaar, T., Zabel, B. U., and Prawitt, D. (2006) Beckwith-Wiedemann syndrome: multiple molecular mechanisms, Expert reviews in molecular medicine, 8:1-19. 168.
 Weksberg, R., Shuman, C., and Smith, A. C. (2005) Beckwith-Wiedemann syndrome, American journal of medical genetics Part C, Seminars in medical genetics, 137C:12-23. 169.
 Elliott, M., and Maher, E. R. (1994) Beckwith-Wiedemann syndrome, J Med Genet, 31:560-564. 170.
 Eggermann, T., Eggermann, K., and Schönherr, N. (2008) Growth retardation versus overgrowth: Silver-Russell syndrome is genetically opposite to Beckwith-Wiedemann syndrome, Trends Genet, 24:195-204. 171.
 Diaz-Meyer, N., Day, C. D., Khatod, K., Maher, E. R., Cooper, W., Reik, W., Junien, C., Graham, G., Algar, E., Der Kaloustian, V. M., and Higgins, M. J. (2003) Silencing of CDKN1C (p57KIP2) is associated with hypomethylation at KvDMR1 in BeckwithWiedemann syndrome, J Med Genet, 40:797-801. 172.
 Lee, M. P., DeBaun, M. R., Mitsuya, K., Galonek, H. L., Brandenburg, S., Oshimura, M., and Feinberg, A. (1999) Loss of imprinting of a paternally expressed transcript, with antisense orientation to KVLQT1, occurs frequently in Beckwith-Wiedemann syndrome and is independent of insulin-like growth factor II imprinting, Proc Natl Acad Sci USA, 96:5203-5208. 173.
 Bourc'his, D., Xu, G. L., Lin, C. S., Bollman, B., and Bestor, T. H. (2001) Dnmt3L and the establishment of maternal genomic imprints, Science, 294:2536-2539. 174.
 Kaneda, M., Okano, M., Hata, K., Sado, T., Tsujimoto, N., Li, E., and Sasaki, H. (2004) Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting, Nature, 429:900-903. 175.
 Catchpoole, D., Lam, W. W., Valler, D., Temple, I. K., Joyce, J. A., Reik, W., Schofield, P. N., and Maher, E. R. (1997) Epigenetic modification and uniparental inheritance of H19 in Beckwith-Wiedemann syndrome, J Med Genet, 34:353-359. 176.
 Miyamoto, T., Hasuike, S., Jinno, Y., Soejima, H., Yun, K., Miura, K., Ishikawa, M., and Niikawa, N. (2002) The human ASCL2 gene escaping genomic imprinting and its expression pattern, J Assist Reprod Genet, 19:240-244. 177.
 Lee, M. P., DeBaun, M., Randhawa, G., Reichard, B. A., Elledge, S. J., and Feinberg, A. P. (1997) Low frequency of p57KIP2 mutation in Beckwith-Wiedemann syndrome, Am J Hum Genet, 61:304-309. 178.
 Cooper, W. N., Luharia, A., Evans, G. A., Raza, H., Haire, A. C., Grundy, R., Bowdin, S. C., Riccio, A., Sebastio, G., Bliek, J., Schofield, P. N., Reik, W., Macdonald, F., and Maher, E. R. (2005) Molecular subtypes and phenotypic expression of Beckwith-Wiedemann syndrome, Eur J Hum Genet, 13:1025-1032. 179.
 Oh, R., Ho, R., Mar, L., Gertsenstein, M., Paderova, J., Hsien, J., Squire, J. A., Higgins, M., Nagy, A., and Lefebvre, L. (2008) Epigenetic and phenotypic consequences of a truncation disrupting the imprinted domain on distal mouse chromosome 7, Mol Cell Biol, 28:1092-1103. 119  180.
 Mager, J., and Bartolomei, M. S. (2005) Strategies for dissecting epigenetic mechanisms in the mouse, Nat Genet, 37:1194-1200. 181.
 Nagy, A., Gertsenstein, M., Vintersten, K., and Behringer, R. (2003) Manipulating the Mouse Embryo: A Laboratory Manual, 3rd ed, 182.
 El-Hashash, A. H., and Kimber, S. J. (2004) Trophoblast differentiation in vitro: establishment and characterisation of a serum-free culture model for murine secondary trophoblast giant cells, Reproduction, 128:53-71. 183.
 Lefebvre, L., Viville, S., Barton, S. C., Ishino, F., and Surani, M. A. (1997) Genomic structure and parent-of-origin-specific methylation of Peg1, Human Molecular Genetics, 6:1907-1915. 184.
 Umlauf, D., Goto, Y., Cao, R., Cerqueira, F., Wagschal, A., Zhang, Y., and Feil, R. (2004) Imprinting along the Kcnq1 domain on mouse chromosome 7 involves repressive histone methylation and recruitment of Polycomb group complexes, Nat Genet, 36:1296-1300. 185.
 Dong, K. B., Maksakova, I. A., Mohn, F., Leung, D., Appanah, R., Lee, S., Yang, H. W., Lam, L. L., Mager, D. L., Schübeler, D., Tachibana, M., Shinkai, Y., and Lorincz, M. C. (2008) DNA methylation in ES cells requires the lysine methyltransferase G9a but not its catalytic activity, EMBO J, 27:2691-2701. 186.
 Searle, A. G., and Beechey, C. V. (1990) Genome imprinting phenomena on mouse chromosome 7, Genet Res, 56:237-244. 187.
 Doherty, A. S., Mann, M. R., Tremblay, K. D., Bartolomei, M. S., and Schultz, R. M. (2000) Differential effects of culture on imprinted H19 expression in the preimplantation mouse embryo, Biol Reprod, 62:1526-1535. 188.
 Niwa, H., Yamamura, K., and Miyazaki, J. (1991) Efficient selection for high-expression transfectants with a novel eukaryotic vector, Gene, 108:193-199. 189.
 Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T., and Nishimune, Y. (1997) 'Green mice' as a source of ubiquitous green cells, FEBS Lett, 407:313-319. 190.
 Hadjantonakis, A. K., Cox, L. L., Tam, P. P., and Nagy, A. (2001) An X-linked GFP transgene reveals unexpected paternal X-chromosome activity in trophoblastic giant cells of the mouse placenta, genesis, 29:133-140. 191.
 Hadjantonakis, A. K., Gertsenstein, M., Ikawa, M., Okabe, M., and Nagy, A. (1998) Noninvasive sexing of preimplantation stage mammalian embryos, Nat Genet, 19:220-222. 192.
 Solter, D., and Knowles, B. B. (1978) Monoclonal antibody defining a stage-specific mouse embryonic antigen (SSEA-1), Proc Natl Acad Sci USA, 75:5565-5569. 193.
 Appanah, R., Dickerson, D. R., Goyal, P., Groudine, M., and Lorincz, M. C. (2007) An unmethylated 3' promoter-proximal region is required for efficient transcription initiation, PLoS Genetics, 3:e27. 194.
 Ball, M. P., Li, J. B., Gao, Y., Lee, J. H., LeProust, E. M., Park, I. H., Xie, B., Daley, G. Q., and Church, G. M. (2009) Targeted and genome-scale strategies reveal gene-body methylation signatures in human cells, Nat Biotechnol, 27:361-368. 195.
 Hellman, A., and Chess, A. (2007) Gene body-specific methylation on the active X chromosome, Science, 315:1141-1143. 196.
 Hadjantonakis, A. K., Gertsenstein, M., Ikawa, M., Okabe, M., and Nagy, A. (1998) Generating green fluorescent mice by germline transmission of green fluorescent ES cells, Mechanisms of Development, 76:79-90.  120  197.
 Szabó, P. E., and Mann, J. R. (1995) Allele-specific expression and total expression levels of imprinted genes during early mouse development: implications for imprinting mechanisms, Genes & Development, 9:3097-3108. 198.
 Zhao, Z., Tavoosidana, G., Sjölinder, M., Göndör, A., Mariano, P., Wang, S., Kanduri, C., Lezcano, M., Sandhu, K. S., Singh, U., Pant, V., Tiwari, V., Kurukuti, S., and Ohlsson, R. (2006) Circular chromosome conformation capture (4C) uncovers extensive networks of epigenetically regulated intra- and interchromosomal interactions, Nat Genet, 38:1341-1347. 199.
 Schulz, R., Menheniott, T. R., Woodfine, K., Wood, A., Choi, J., and Oakey, R. (2006) Chromosome-wide identification of novel imprinted genes using microarrays and uniparental disomies, Nucleic Acids Research, 34:e88. 200.
 Hatada, I., and Mukai, T. (1995) Genomic imprinting of p57KIP2, a cyclin-dependent kinase inhibitor, in mouse, Nat Genet, 11:204-206. 201.
 Caspary, T., Cleary, M. A., Baker, C. C., Guan, X. J., and Tilghman, S. (1998) Multiple mechanisms regulate imprinting of the mouse distal chromosome 7 gene cluster, Molecular and Cellular Biology, 18:3466-3474. 202.
 Dao, D., Frank, D., Qian, N., O'Keefe, D., Vosatka, R. J., Walsh, C. P., and Tycko, B. (1998) IMPT1, an imprinted gene similar to polyspecific transporter and multi-drug resistance genes, Human Molecular Genetics, 7:597-608. 203.
 Qian, N., Frank, D., O'Keefe, D., Dao, D., Zhao, L., Yuan, L., Wang, Q., Keating, M., Walsh, C., and Tycko, B. (1997) The IPL gene on chromosome 11p15.5 is imprinted in humans and mice and is similar to TDAG51, implicated in Fas expression and apoptosis, Human Molecular Genetics, 6:2021-2029. 204.
 Takagi, N., and Sasaki, M. (1975) Preferential inactivation of the paternally derived X chromosome in the extraembryonic membranes of the mouse, Nature, 256:640-642. 205.
 Barlow, P. W., and Sherman, M. I. (1972) The biochemistry of differentiation of mouse trophoblast: studies on polyploidy, Journal of embryology and experimental morphology, 27:447-465. 206.
 Johnson, M. H., and Rossant, J. (1981) Molecular studies on cells of the trophectodermal lineage of the postimplantation mouse embryo, Journal of embryology and experimental morphology, 61:103-116. 207.
 Sánchez, M. J., Holmes, A., Miles, C., and Dzierzak, E. (1996) Characterization of the first definitive hematopoietic stem cells in the AGM and liver of the mouse embryo, Immunity, 5:513-525. 208.
 Gardner, R. L., Papaioannou, V. E., and Barton, S. C. (1973) Origin of the ectoplacental cone and secondary giant cells in mouse blastocysts reconstituted from isolated trophoblast and inner cell mass, Journal of embryology and experimental morphology, 30:561-572. 209.
 Rossant, J., and Tamura-Lis, W. (1981) Effect of culture conditions on diploid to giant-cell transformation in postimplantation mouse trophoblast, Journal of embryology and experimental morphology, 62:217-227. 210.
 Varmuza, S., Prideaux, V., Kothary, R., and Rossant, J. (1988) Polytene chromosomes in mouse trophoblast giant cells, Development, 102:127-134. 211.
 Zybina, T. G., Zybina, E. V., and Shteĭn, G. I. (1985) [DNA content of the nuclei of secondary giant cells of the rat trophoblast at different phases of the polytene nucleus cycle], Tsitologiia, 27:957-960.  121  212.
 Dahl, J. A., Duggal, S., Coulston, N., Millar, D., Melki, J., Shahdadfar, A., Brinchmann, J. E., and Collas, P. (2008) Genetic and epigenetic instability of human bone marrow mesenchymal stem cells expanded in autologous serum or fetal bovine serum, Int J Dev Biol, 52:1033-1042. 213.
 Huntriss, J., and Picton, H. M. (2008) Stability of genomic imprinting in embryonic stem cells: lessons from assisted reproductive technology, Current stem cell research & therapy, 3:107-116. 214.
 Pannetier, M., and Feil, R. (2007) Epigenetic stability of embryonic stem cells and developmental potential, Trends Biotechnol, 25:556-562. 215.
 Zvetkova, I., Apedaile, A., Ramsahoye, B., Mermoud, J. E., Crompton, L. A., John, R., Feil, R., and Brockdorff, N. (2005) Global hypomethylation of the genome in XX embryonic stem cells, Nat Genet, 37:1274-1279. 216.
 Gonçalves, C. R., Antonini, S., Vianna-Morgante, A. M., Machado-Santelli, G. M., and Bevilacqua, E. (2003) Developmental changes in the ploidy of mouse implanting trophoblast cells in vitro, Histochem Cell Biol, 119:189-198. 217.
 Wang, J., Mager, J., Chen, Y., Schneider, E., Cross, J. C., Nagy, A., and Magnuson, T. (2001) Imprinted X inactivation maintained by a mouse Polycomb group gene, Nat Genet, 28:371-375. 218.
 Martin, G. R. (1981) Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells, Proc Natl Acad Sci USA, 78:7634. 219.
 Thomson, J. A., Itskovitz-Eldor, J., Shapiro, S. S., Waknitz, M. A., Swiergiel, J. J., Marshall, V. S., and Jones, J. M. (1998) Embryonic stem cell lines derived from human blastocysts, Science, 282:1145-1147. 220.
 Williams, R. L., Hilton, D. J., Pease, S., Willson, T. A., Stewart, C. L., Gearing, D. P., Wagner, E. F., Metcalf, D., Nicola, N. A., and Gough, N. M. (1988) Myeloid leukaemia inhibitory factor maintains the developmental potential of embryonic stem cells, Nature, 336:684. 221.
 Robertson, E., Bradley, A., Kuehn, M., and Evans, M. (1986) Germ-line transmission of genes introduced into cultured pluripotential cells by retroviral vector, Nature, 323:445-448. 222.
 Nagy, A., Gócza, E., Diaz, E. M., Prideaux, V. R., Iványi, E., Markkula, M., and Rossant, J. (1990) Embryonic stem cells alone are able to support fetal development in the mouse, Development, 110:815-821. 223.
 Nagy, A., Rossant, J., Nagy, R., Abramow-Newerly, W., and Roder, J. C. (1993) Derivation of completely cell culture-derived mice from early-passage embryonic stem cells, Proc Natl Acad Sci USA, 90:8424-8428. 224.
 Pikarsky, E., Sharir, H., Ben-Shushan, E., and Bergman, Y. (1994) Retinoic acid represses Oct-3/4 gene expression through several retinoic acid-responsive elements located in the promoter-enhancer region, Mol Cell Biol, 14:1026-1038. 225.
 Okamoto, K., Okazawa, H., Okuda, A., Sakai, M., Muramatsu, M., and Hamada, H. (1990) A novel octamer binding transcription factor is differentially expressed in mouse embryonic cells, Cell, 60:461-472. 226.
 Jeppesen, P., and Turner, B. M. (1993) The inactive X chromosome in female mammals is distinguished by a lack of histone H4 acetylation, a cytogenetic marker for gene expression, Cell, 74:281-289. 227.
 Monk, M. (1986) Methylation and the X chromosome, Bioessays, 4:204-208. 122  228.
 Takagi, N., and Oshimura, M. (1973) Fluorescence and Giemsa banding studies of the allocyclic X chromosome in embryonic and adult mouse cells, Exp Cell Res, 78:127-135. 229.
 Farivar, S., Yamaguchi, S., Sugimoto, M., and Takagi, N. (2004) X-chromosome inactivation in differentiating mouse embryonic stem cells carrying X-linked GFP and lacZ transgenes, Int J Dev Biol, 48:629-635. 230.
 Tada, T., Tada, M., and Takagi, N. (1993) X chromosome retains the memory of its parental origin in murine embryonic stem cells, Development, 119:813-821. 231.
 Kalantry, S., Mills, K. C., Yee, D., Otte, A., Panning, B., and Magnuson, T. (2006) The Polycomb group protein Eed protects the inactive X-chromosome from differentiationinduced reactivation, Nat Cell Biol, 8:195-202. 232.
 Sado, T., and Ferguson-Smith, A. C. (2005) Imprinted X inactivation and reprogramming in the preimplantation mouse embryo, Human Molecular Genetics, 14 Spec No 1:R59-64. 233.
 Wang, Z. Q., Fung, M. R., Barlow, D. P., and Wagner, E. F. (1994) Regulation of embryonic growth and lysosomal targeting by the imprinted Igf2/Mpr gene, Nature, 372:464-467. 234.
 Zwart, R., Sleutels, F., Wutz, A., Schinkel, A. H., and Barlow, D. P. (2001) Bidirectional action of the Igf2r imprint control element on upstream and downstream imprinted genes, Genes & Development, 15:2361-2366. 235.
 Latos, P. A., Stricker, S. H., Steenpass, L., Pauler, F. M., Huang, R., Senergin, B. H., Regha, K., Koerner, M. V., Warczok, K. E., Unger, C., and Barlow, D. P. (2009) An in vitro ES cell imprinting model shows that imprinted expression of the Igf2r gene arises from an allelespecific expression bias, Development, 136:437-448. 236.
 Viswanathan, S., Benatar, T., Mileikovsky, M., Lauffenburger, D. A., Nagy, A., and Zandstra, P. W. (2003) Supplementation-dependent differences in the rates of embryonic stem cell self-renewal, differentiation, and apoptosis, Biotechnol Bioeng, 84:505-517. 237.
 Yoshimizu, T., Sugiyama, N., De Felice, M., Yeom, Y. I., Ohbo, K., Masuko, K., Obinata, M., Abe, K., Schöler, H. R., and Matsui, Y. (1999) Germline-specific expression of the Oct-4/green fluorescent protein (GFP) transgene in mice, Dev Growth Differ, 41:675-684. 238.
 Terranova, R., Yokobayashi, S., Stadler, M., Otte, A., Van Lohuizen, M., Orkin, S., and Peters, A. (2008) Polycomb group proteins Ezh2 and Rnf2 direct genomic contraction and imprinted repression in early mouse embryos, Developmental Cell, 15:668-679. 239.
 Guillemot, F., Caspary, T., Tilghman, S., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., Anderson, D. J., Joyner, A. L., Rossant, J., and Nagy, A. (1995) Genomic imprinting of Mash2, a mouse gene required for trophoblast development, Nat Genet, 9:235-242. 240.
 Elliott, R. W., and Pazik, J. (1995) An interstitial telomere array near Hba on mouse Chr 11 is a candidate for the homolog of the telomere at human 16p, Genomics, 27:217-218. 241.
 Yen, C. H., Hohman, C., and Elliott, R. W. (1997) Mapping and characterization of three YAC clones containing TTAGGG arrays, Mamm Genome, 8:775-777. 242.
 Yen, C. H., Pazik, J., and Elliott, R. W. (1996) A polymorphic interstitial telomere array near the center of mouse chromosome 8, Mamm Genome, 7:218-221. 243.
 Yen, C. H., Pazik, J., Zhang, Y., and Elliott, R. W. (1997) An interstitial telomere array proximal to the distal telomere of mouse chromosome 13, Mamm Genome, 8:411-417. 244.
 Yen, C. H., Matsuda, Y., Chapman, V. M., and Elliott, R. W. (1995) A genomic clone containing a telomere array maps near the centromere of mouse chromosome 6, Mamm Genome, 6:96-102.  123  245.
 Nakanishi, T., Kuroiwa, A., Yamada, S., Isotani, A., Yamashita, A., Tairaka, A., Hayashi, T., Takagi, T., Ikawa, M., Matsuda, Y., and Okabe, M. (2002) FISH analysis of 142 EGFP transgene integration sites into the mouse genome, Genomics, 80:564-574. 246.
 Hadjantonakis, A. K., and Nagy, A. (2000) FACS for the isolation of individual cells from transgenic mice harboring a fluorescent protein reporter, Genesis, 27:95-98. 247.
 Shemer, R., Birger, Y., Riggs, A. D., and Razin, A. (1997) Structure of the imprinted mouse Snrpn gene and establishment of its parental-specific methylation pattern, Proc Natl Acad Sci USA, 94:10267-10272. 248.
 Chuva De Sousa Lopes, S., Hayashi, K., Shovlin, T., Mifsud, W., Surani, M. A., and McLaren, A. (2008) X chromosome activity in mouse XX primordial germ cells, PLoS Genet, 4:e30. 249.
 Sugimoto, M., and Abe, K. (2007) X chromosome reactivation initiates in nascent primordial germ cells in mice, PLoS Genet, 3:e116. 250.
 Chicoine, E., Estève, P. O., Robledo, O., Van Themsche, C., Potworowski, E. F., and StPierre, Y. (2002) Evidence for the role of promoter methylation in the regulation of MMP-9 gene expression, Biochemical and Biophysical Research Communications, 297:765-772. 251.
 Yan, Z. H., Gong, X. L., Guo, X. B., Xu, M., Ren, Z. R., and Zeng, Y. T. (2009) Association of differential and site-dependent CpG methylation and diverse expression of DNA methyltransferases with the tissue-specific expression of human beta-globin gene in transgenic mice, Int J Hematol, 89:414-421. 252. Molyneaux, K. A., Stallock, J., Schaible, K., and Wylie, C. (2001) Time-lapse analysis of living mouse germ cell migration, Dev Biol, 240:488-498.  124  Appendix  125  The University of British Columbia  Animal Care Certificate Application Number:  A03-0289  Investigator or Course Director: Louis Lefebvre Department:  Medical Genetics  Animals Approved:  Start Date:  Mice 445  2003-10-1  Approval Date: 2005-03-07  Funding Sources: Funding Agency: Funding Title: Funding Agency: Funding Title: Funding Agency: Funding Title: Funding Agency: Funding Title: Funding Agency: Funding Title: Funding Agency: Funding Title:  Michael Smith Foundation for Health Research Function and mechanism of genomic imprinting on mouse chromosome 6 Canadian Institutes of Health Research Function and mechanism of genomic imprinting in the mouse Michael Smith Foundation for Health Research Function and mechanism of genomic imprinting on mouse chromosome 6 Michael Smith Foundation for Health Research Function and mechanism of genomic imprinting on mouse chromosome 6  Canadian Institutes of Health Research Function and mechanism of genomic imprinting in the mouse Michael Smith Foundation for Health Research Function and mechanism of genomic imprinting on mouse chromosome 6  126  Unfunded title:  N/A  The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies.  A copy of this certificate must be displayed in your animal facility  Office of Research Services and Administration 102, Agronomy Road, Vancouver, V6T 1Z3 Phone: 604-827-5111 Fax: 604-822-5093  127  THE UNIVERSITY OF BRITISH COLUMBIA  ANIMAL CARE CERTIFICATE Application Number: A07-0160 Investigator or Course Director: Louis Lefebvre Department: Medical Genetics Animals:  Mice 129S1 10 Mice Dnmt3l KO 20 Mice IC2KO 15 Mice Del7AI 55 Mice C57BL/6J 15 Mice Peg1 KO 10 Mice DelTel7 50 Mice Mash2 KI 20 Mice C3H 15 Mice D4 (X-linked gfp) 10 Mice 129C7 15 Mice I2 15 Mice Dnmt3l KO 20 Mice Tel7KI 50 Mice CD-1 110 Mice Dp7AI 25  Start Date:  April 1, 2007  Approval Date:  July 13, 2009  Funding Sources: Funding Agency: Funding Title:  Canadian Institutes of Health Research (CIHR) Developmental regulation of genomic imprinting in the mouse  Funding Agency:  Michael Smith Foundation for Health Research  Funding Title:  Somatic and gametic loss of imprinting (LOI) in mammalian development: studies using a novel imprinted transgene on the mouse distal chromosome 7 (MMU7) imprinted region  Funding Agency:  Canadian Institutes of Health Research (CIHR)  Funding Title:  Gametic and somatic loss of imprinting (loi) in mammalian development: studies using a novel imprinted transgene on the mouse distal chromosome 7 imprinted region as an indicator of loi  Unfunded title:  N/A  The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies.  A copy of this certificate must be displayed in your animal facility.  Office of Research Services and Administration 102, 6190 Agronomy Road, Vancouver, BC V6T 1Z3 Phone: 604-827-5111 Fax: 604-822-5093  128  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items