Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Consequences of mitotic loss of heterozygosity on genomic imprinting in mouse embryonic stem cells Elves, Rachel Leigh 2008

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

Item Metadata


24-ubc_2008_fall_elves_rachel.pdf [ 12.82MB ]
JSON: 24-1.0066579.json
JSON-LD: 24-1.0066579-ld.json
RDF/XML (Pretty): 24-1.0066579-rdf.xml
RDF/JSON: 24-1.0066579-rdf.json
Turtle: 24-1.0066579-turtle.txt
N-Triples: 24-1.0066579-rdf-ntriples.txt
Original Record: 24-1.0066579-source.json
Full Text

Full Text

CONSEQUENCES OF MITOTIC LOSS OF HETEROZYGOSITY ON GENOMIC IMPRINTING IN MOUSE EMBRYONIC STEM CELLS  by RACHEL LEIGH ELVES B.Sc., University of British Columbia, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIRMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Medical Genetics)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2008  © Rachel Leigh Elves, 2008  Abstract Epigenetic differences between maternally inherited and paternally inherited chromosomes, such as CpG methylation, render the maternal and paternal genome functionally inequivalent, a phenomenon called genomic imprinting. This functional inequivalence is exemplified with imprinted genes, whose expression is parent-of-origin specific. The dosage of imprinted gene expression is disrupted in cells with uniparental disomy (UPD), which is an unequal parental contribution to the genome. I have derived mouse embryonic stem (ES) cell sub-lines with maternal UPD (mUPD) for mouse chromosome 6 (MMU6) to characterize regulation and maintenance of imprinted gene expression. The main finding from this study is that maintenance of imprinting in mitotic UPD is extremely variable. Imprint maintenance was shown to vary from gene to gene, and to vary between ES cell lines depending on the mechanism of loss of heterozygosity (LOH) in that cell line. Certain genes analyzed, such as Peg10, Sgce, Peg1, and Mit1 showed abnormal expression in ES cell lines for which they were mUPD. These abnormal expression levels are similar to that observed in ES cells with meiotically-derived full genome mUPD (parthenogenetic ES cells). Imprinted CpG methylation at the Peg1 promoter was found to be abnormal in all sub-lines with mUPD for Peg1. Two cell sub-lines which incurred LOH through mitotic recombination showed hypermethylation of Peg1, consistent with the presence of two maternal alleles. Surprisingly, a cell sub-line which incurred LOH through full chromosome duplication/loss showed hypomethylation of Peg1. The levels of methylation observed in these sub-lines correlates with expression, as the first two sub-lines showed a near-consistent reduction of Peg1, while the latter showed Peg1 levels close to wild-type.  ii  Altogether these results suggest that certain imprinted genes, like Peg1 and Peg10, have stricter imprinting maintenance, and as a result show abnormal expression in UPD. This strict imprint maintenance is disrupted, however, in UPD incurred through full chromosome duplication/loss, possibly because of the trisomic intermediate stage which occurs in this mechanism.  iii  Table of contents Abstract ...............................................................................................................................ii Table of contents ................................................................................................................iv List of tables List of figures.....................................................................................................................vii Abbreviations ...................................................................................................................viii Acknowledgements ............................................................................................................ix CHAPTER 1 Introduction ................................................................................................. 1 1.1 General introduction ................................................................................................... 1 1.2 Genomic imprinting and DNA CpG methylation....................................................... 2 1.3 Imprinting control centres........................................................................................... 4 1.4 Parent-of-origin gene expression................................................................................ 5 1.5 Parthenogenetic, gynogenetic, and androgenetic uniparental embryos...................... 6 1.6 Discovery of imprinted genes..................................................................................... 7 1.7 Mouse as a model for human imprinting disorders .................................................... 9 1.8 Rationale and experimental approach....................................................................... 11 CHAPTER 2 Materials and methods ............................................................................. 14 2.1 Nucleic acid work ..................................................................................................... 14 2.2 Southern hybridization.............................................................................................. 15 2.3 Polymerase chain reaction and agarose gel electrophoresis ..................................... 16 2.4 Mice maintenance and conditions ............................................................................ 18 2.5 R3/R3 mouse stocks ................................................................................................. 18 2.6 Murine embryonic stem cell culture ......................................................................... 19 2.7 High G418 selection ................................................................................................. 19 2.8 LacZ staining ............................................................................................................ 21 2.9 Embryoid body differentiation ................................................................................. 21 CHAPTER 3 Derivation and characterization of embryonic stem cell lines with lossof-heterozygosity on mouse chromosome 6 .................................................................... 25 3.1 Introduction............................................................................................................... 25 3.2 Results ...................................................................................................................... 30 3.3 Discussion................................................................................................................. 33 iv  CHAPTER 4 Maintenance of imprinted expression and methylation in LOH ES cell sub-lines ............................................................................................................................. 50 4.1 Introduction............................................................................................................... 50 4.2 Results ...................................................................................................................... 58 4.3 Discussion................................................................................................................. 62 CHAPTER 5 Discussion................................................................................................... 76 5.1 Summary of results ................................................................................................... 78 5.2 Future directions ....................................................................................................... 79 References.......................................................................................................................... 81  v  List of tables Table 2.1. Laboratory solutions....................................................................................... 22 Table 2.2. Oligonucleotide primers used in genotyping and RT-PCR ........................ 23 Table 2.3. Tissue culture solutions .................................................................................. 24 Table 3.1. Sex chromosome complement of newly derived R129 and 129R ES cell lines..................................................................................................................................... 42 Table 3.2. Selection for spontaneous LOH at ROSA26 locus through high G418 selection of R129-E2 ES cells ........................................................................................... 43 Table 4.1. Imprinted genes on MMU6, imprinted domain 1 ........................................ 66 Table 4.2. Imprinted genes on MMU6, imprinted domain 2 ........................................ 67 Table 4.3. Imprinted gene expression in parthenogenetic and androgenetic embryos ............................................................................................................................................ 68 Table 4.4. Expected and observed imprinted gene expression in murine PGES cells and AGES cells.................................................................................................................. 69  vi  List of figures Figure 1.1. Lifecycle of genomic imprinting................................................................... 13 Figure 3.1. R26R insert on mouse chromosome 6 (MMU6) at Gt(ROSA)26Sor locus (ROSA26)........................................................................................................................... 38 Figure 3.2. LacZ staining of the Cre-electroporated ES cell line R26r3(+/-)-Cre. ..... 39 Figure 3.3. X chromosome PCR genotyping in ES cells................................................ 40 Figure 3.4. XY chromosome PCR genotyping in ES cells............................................. 41 Figure 3.5. Detection of LOH at the ROSA26 locus by Southern blot analysis of ES cell genomic DNA.............................................................................................................. 44 Figure 3.6. Characterization of MMU6 LOH in ES cells.............................................. 45 Figure 3.7. SNP marker analysis on MMU6 .................................................................. 47 Figure 3.8. Summary of regions of mUPD for MMU6 in 3 high G418-selected ES cells LOH for the ROSA26 locus ............................................................................................. 49 Figure 4.1. MMU6 proximal imprinted gene domains 1 and 2 .................................... 70 Figure 4.2. RT-PCR analysis of ID1 genes (25 cycles) .................................................. 71 Figure 4.3. RT-PCR analysis of ID1 genes (30 cycles) .................................................. 72 Figure 4.4. RT-PCR analysis of ID2 genes (25 cycles) .................................................. 73 Figure 4.5. RT-PCR analysis of ID2 genes (30 cycles) .................................................. 74 Figure 4.6. Peg1 methylation-sensitive Southern blot of wild-type embryo and R129E2 parental and R129*-E2 sub-lines LOH at ROSA26 locus....................................... 75  vii  Abbreviations AGES cells B6 CpG DMR Dnmt dpc EB ES cells ICR ID1 ID2 indel lacZ LOH MatDi MD6 MD6-2/dp1 MD6-2/p1 MEG mUPD neo pA PBS PCR PEFFs PEG PGES cells pUPD SA TS cells UPD  Androgenetic embryonic stem cells C57BL/6J mouse line 5’-CG-3’ Differentially methylated region DNA methyltransferase Days post coitum Embryoid bodies Embryonic stem cells Imprinting control region Imprinted domain 1 Imprinted domain 2 insertion/deletion β-galactosidase Loss of heterozygosity Maternal disomy mUPD6 mUPD6-ID2/dup-ID1 mUPD6-ID2/partial mUPD6-ID1 Maternally expressed gene Maternal uniparental disomy Neomycin resistance Polyadenylation sequence Phosphate-buffered saline Polymerase chain reaction Primary mouse embryonic fibroblast feeder cells Paternally expressed gene Parthenogenetic embryonic stem cells Paternal uniparental disomy Splice acceptor Trophectoderm stem cells Uniparental disomy  viii  Acknowledgements First and foremost I would like to thank my supervisor, Louis Lefebvre, for his patience, enthusiasm, and all the help he’s given me. I sincerely appreciate the time and effort he spent creating new ES cell lines, without which my project would not have been possible. I would like to thank the entire Lefebvre lab team, past and present, for their help, in particular Aaron Bogutz, for his help with Southern Blot hybridization and mouse care, and Tracy Weng, for her hard work genotyping MMU6 markers and designing spreadsheets. Thanks to Jacob Hodgson who was a big help with trouble-shooting many different techniques. A big thank you to my friends Connie Lee, Onkar Bains, Cheryl Bishop, Omid Toub, and Pam Kalas, for your support and advice. Thank you to my parents, Edward and Vicki Schmuland, and to my “adopted” parents Jean and David Elves, for their unwavering belief in me. Thank you to Helen Burston, for your love and support, and helping me through many hard times. And most of all, thank you to Robert Elves, for your love, encouragement, advise, and the huge amount of support you’ve given me, without which I would never had made it.  ix  CHAPTER 1 Introduction 1.1 General introduction Most genes in our genome are present in two copies, one from each of our parents. For most of these genes both copies, or alleles, are equivalent and used the same way in our cells. A small proportion of genes, however, are only expressed from one allele in a parent-of-origin-specific manner (Reik, et al, 2001). This phenomenon called genomic imprinting is characterized by epigenetic differences between maternally inherited and paternally inherited chromosomes, such as histone modifications and DNA methylation at specific cytosine residues (CpG sites) (de la Casa-Esperon, et al, 2003). Mammalian imprinting is driven by the establishment of parent-of-origin specific imprints during gametogenesis, which are then maintained in embryonic somatic cells after fertilization (Paoloni-Giacobino, et al, 2006; Reik, et al, 2001) (Figure 1.1). Maintenance of imprinting from generation to generation, however, depends upon erasure of biallelic imprints in the next generation of primordial gametes, so that sex-specific imprints can be established (Reik, et al, 2001; Tucker, et al, 1996). If both chromosomal homologs are inherited from the same parent an abnormal situation results, termed uniparental disomy (UPD). Imprinted genes are potentially misregulated, often being over-expressed or under-expressed, because of the associated imprinted epigenetic abnormalities. In this thesis a newly developed experimental system is described to select for spontaneous UPD in mitotically dividing embryonic stem (ES) cells, in order to study imprinting maintenance and regulation (Lefebvre, et al, 2001). This is accomplished through selection for spontaneous mitotic loss of heterozygosity (LOH), which not only generates marker homozygosity for the chromosomal region involved, but  1  also causes loss of alleles from one parent, rendering the chromosome monoallelic and with UPD. This research analyzes the nature of the mitotic events involved in LOH in ES cells, and studies the consequences for these spontaneous mitotic rearrangements on the maintenance of imprinted methylation and gene expression. This new tool provides a method to study LOH in ES cells, and provides a new experimental system to study the mechanistic aspects of imprinting and identify new imprinted genes. This research could also shed light on mechanisms for LOH and/or UPD in vivo, events causative for cancer (Cooper, et al, 2005; Holm, et al, 2005). The work presented in this thesis centers on mouse chromosome 6 (MMU6), the proximal region of which is known to contain at least two independent imprinted domains. Part of this region is syntenic homology with human 7q32, a region associated with imprinted effects leading to growth retardation and some forms of Silver-Russell syndrome (Abu-Amero, et al, 2008; Kobayashi, et al, 1997). Our approach could lead to new models systems, or to the identification of new candidate genes for these syndromes.  1.2 Genomic imprinting and DNA CpG methylation Of the three main types of epigenetic marks that could act as imprinting marks, covalent histone modifications, histone variants, and DNA methylation, the latter has been determined to be the most crucial in mammals. The only known methylation target in mammals is cytosine, specifically cytosines in CpG dinucleotides (5’-CG-3’) (Fazzari, et al, 2004). The palindromic nature of CpGs allow cytosine methylation to be maintained after DNA replication by methylation of the CpG on the newly synthesized opposite strand (Paoloni-Giacobino, et al, 2006).  2  Methylated cytosine is easily mutated to thymidine through spontaneous deamination, which has lead to a genome-wide depletion of both CpG sites, and G+C content overall (in humans there is 40% G+C content). Remaining regions high in both G+C content (≥50%) and CpG sites (0.6 observed to expected CpG ratio) are appropriately named CpG islands, since they differ substantially in G+C and CpG content from both surrounding genomic sequence and from the rest of the genome as a whole (Fazzari, et al, 2004; Wang, et al, 2004). CpG islands are 200bp or more in length and found near promoters in 50% of all known human genes, most of which are housekeeping genes (Wang, et al, 2004). They are also found at the 3’ end of many tissue-specific genes (Wang, et al, 2004). CpG islands are almost always unmethylated, possibly because active transcription of housekeeping genes and frequent occupation of the promoter-CpG island region of transcriptional machinery prevents access of methyltransferase enzymes. One exception to CpG island hypomethylation is seen with imprinted genes. For some of these genes CpG islands methylation is seen in a parent-of-origin specific manner, such that the repressed parental allele is typically hypermethylated, while the expressed parental allele is hypomethylated (Luedi, et al, 2005; Reik, et al, 2001). These differentially methylated CpG islands, or DMRs (differentially methylated regions), are maintained in all tissues, regardless of whether the imprinted gene is expressed and whether or not there is imprinted or biallelic expression. Genes that display tissue-specific imprinting (with biallelic expression in some tissues and monoparental expression in others) tend to have multiple promoters, usually with one of the promoters imprinted, and the remainder active from both alleles (e.g. IGF2, GNAS1, and Grb10) (Ekstrom, et al, 1995; Hayward, et al, 1998; Yamasaki-Ishizaki, et al, 2007).  3  Targeted mutations in the de novo DNA and maintenance methylation enzymes, DNA methyltransferases 3a and 3b (Dnmt3a, Dnmt3b), have been shown to prevent establishment of imprinted methylation in DMRs and cause biallelic expression of imprinted genes (Okano, et al, 1999). Loss of allele-specific methylation is seen in targeted mutations in the DNA methylation maintenance enzyme, Dnmt1. Like Dnmt3a and 3b mutations, Dnmt1 mutation also results in biallelic expression of imprinted genes (Tucker, et al, 1996). An additional line of evidence for the importance of methylation in imprinting was that addition of unmethylatable cytosine analogs, which covalently interact with Dnmts, causes passive loss of methylation accompanied by biallelic expression of imprinted genes (Lynch, et al, 2002).  1.3 Imprinting control centres The majority of imprinted genes are located together in clusters called imprinted domains, which contain a mixture of maternally expressed genes (MEGs) and paternally expressed genes (PEGs), with some interspersed non-imprinted genes (Wood, et al, 2006). It is hypothesized that each imprinted domain contains a cis-acting regulatory centre, called an imprinting control region (ICR), which acquires differential germline imprints. This initial DMR at the ICR starts a cascade of cellular events that establishes the imprinting for the rest of the imprinted genes within the imprinted domain. The ICR DMR may also be involved in imprint maintenance. Known ICRs are CpG-rich sequences which can act as promoters for proteincoding genes or non-coding RNAs, or as chromatin boundary elements. On MMU6, the focus of this thesis, the proposed imprinted control centers are located at the Peg1 and Peg10 loci. The Peg10 promoter contains the only CpG island that is differentially  4  methylated in this imprinted domain, and Peg10 is the only gene in the cluster with conserved imprinted expression and a DMR among eutherian mammals and marsupials (Suzuki, et al, 2007). Peg1 also has conserved imprinted expression between eutherian mammals and marsupials, but its promoter lacks a DMR in marsupials (Suzuki, et al, 2005). Peg1 is the only imprinted gene with a DMR within 30Mb of the imprinted genes in the second imprinted domain on MMU6 (Lefebvre, et al, 1997).  1.4 Parent-of-origin gene expression Imprinted genes are unusual in that they have different epigenetic marks, or a different epigenotype, between parental alleles. While maternal and paternal alleles of imprinted genes have an identical genotype, their difference in epigenotype causes the imprinted gene to be expressed from only one parental allele. Also unusual is the fact that because imprinted genes only have one active, or functional, allele, loss of this active allele results in loss of gene function. Several different model systems, including UPD, have been exploited to reveal and study these properties of imprinted genes. UPD, an unequal parental contribution to the genome, results from an entire homologous chromosome, or portion of a homolog being inherited solely from one parent, instead of one copy inherited from each parent (Allen, et al, 1994). Although the term UPD is used to denote every situation where part of the genome comes from a single parent, the mechanisms leading to UPD can be different, occurring in either meiosis or mitosis. Meiotic UPD can be caused in a trisomic embryo early after fertilization by loss of a chromosome (trisomic rescue), or by fertilization of a disomic oocyte by a sperm nullisomic for the same chromosome (Kotzot, 2004).  5  1.5 Parthenogenetic, gynogenetic, and androgenetic uniparental embryos Genome-wide, meiotically-derived UPD found in parthenogenetic and androgenetic embryos (uniparental embryos) has been important not only in uncovering the existence of genomic imprinting in mammals, but has also been a useful tool in discovering and studying imprinted genes. Research with uniparental embryos was first carried out to answer the essential questions, “Are parental genomes functionally equivalent or different? Are both the maternal and paternal genome required for development in mammals?” While plants and some reptiles can produce parthenogenetic offspring, having a fully maternal genome with no paternal contribution, there are no known examples of viable parthenogenotes in mammals. Diploid mouse parthenogenetic embryos can be produced artificially by preventing second polar body extrusion from the oocyte by chemical means, thus rendering the oocyte diploid and able to begin cell division (McGrath, et al, 1984; Ogawa, et al, 2006; Surani, et al, 1984). When implanted into a surrogate mother parthenogenetic embryos are growth retarded, having poor trophectoderm growth, and reaching at most mid-gestational development (McGrath, et al, 1984; Ogawa, et al, 2006; Surani, et al, 1984). In contrast, androgenotes, constructed from fertilized oocytes by substitution of the maternal pronucleus by a second male pronucleus, have abundant trophectoderm growth, and very little development of the embryo (McGrath, et al, 1984; Ogawa, et al, 2006; Surani, et al, 1984). This experimental result left some skepticism that developmental failure could be due to a necessary paternal cytoplasmic contribution to the oocyte, or for a fully homozygous genome uncovering lethal recessive alleles leading to the developmental failure rather than a genomic imprinting phenomenon. To disprove these alternate theories  6  a different experimental paradigm was used, dubbed gynogenetic to distinguish it from parthenogenetic, because the starting oocyte in the former was activated using normal fertilization by spermatozoa, whereas the latter was activated by artificial means and had no contact with spermatozoa (Surani, et al, 1984). Gynogenetic embryos constructed using maternal pronuclei from two different mouse strains were used to show that even with heterozygous alleles embryos with full maternal UPD (mUPD) could not survive past mid-gestation (Surani, et al, 1984). These experiments demonstrated that both parental genomes were indeed necessary for development. This observation sparked the hypothesis of genomic imprinting, which purported that maternal and paternal genomes are not functionally equivalent (Barton, et al, 1984; Surani, et al, 1984). It was then suggested that specific genes, imprinted genes, may have parent-of-origin specific expression, resulting in some genes only being expressed from the maternal chromosome, while others were expressed only from the paternal chromosome. If true, genomic imprinting would mean that expression of PEGs would be absent in parthenogenotes and gynogenotes, while expression of MEGs would be absent in androgenotes.  1.6 Discovery of imprinted genes Several different methods of imprinted gene identification have been employed including use of full genome UPD embryos, parthenogenetic and androgenetic, as well as reciprocal translocation embryos with UPD for one chromosome, or UPD for a portion of one chromosome (Ferguson-Smith, et al, 1991; Kaneko-Ishino, et al, 1995; Peters, et al, 2004). Reciprocal translocation embryos involve tedious crosses either between naturally occurring mouse populations with different Robertsonian translocations, or crosses  7  between mice with rare translocation mutations (Cattanach, et al, 1994). UPD embryos are recovered at a rare frequency when an oocyte nullisomic for a particular chromosome or chromosomal region is fertilized with a sperm disomic for the same genomic region, or vice versa. These reciprocal translocation embryos, although hard to obtain, have been key in identifying portions of the genome with UPD phenotypes (such as embryonic lethality, growth abnormalities, and behavioral abnormalities) thus helping to locate candidate imprinted gene clusters (Cattanach, et al, 1994; Cattanach, et al, 1997; Cattanach, et al, 1998; Lefebvre, et al, 1998). Uniparental material can be used to perform cDNA subtraction, where cDNA from full or partial UPD embryos is subtracted from cDNA of fertilized embryo controls (Kaneko-Ishino, et al, 1995). This technique can identify PEGs when subtracting mUPD from normal material, or MEGs when subtracting paternal UPD (pUPD) from normal material. These techniques were successful because of the observation that PEGs are not expressed in parthenogenetic embryos, and MEGs are not expressed in androgenetic embryos (Kono, et al, 2004; Ogawa, et al, 2006; Walsh, et al, 1994). Conversely, one would expect MEGs to be over-expressed (about 2-fold) in parthenogenetic embryos, and PEGs should be over-expressed in androgenetic embryos. With recent analysis, however, it seems that uniparental embryos do not strictly follow this expected expression pattern: many PEGs show expression (termed reactivation of PEGs) and many MEGs are seen at wild-type or only slightly increased expression levels, in parthenogenetic embryos (Jiang, et al, 2007; Szabo, et al, 1994). It seems that the cell can somewhat adjust or normalize imprinted gene expression to become closer to wild-type levels. Despite these possible compensatory mechanisms many imprinted genes that play critical roles during  8  development are still not at sufficient levels to allow for completion of development, as seen by parthenogenetic and androgenetic embryo death by mid-gestation (McGrath, et al, 1984; Surani, et al, 1984).  1.7 Mouse as a model for human imprinting disorders In general, mouse models have several advantages for studying human disease. The short generation time of the mouse, 21 days vs. 25 years in humans, makes multigenerational studies on the impact of mutations or epimutations feasible. Unlike humans mice are raised in a homogenous environment, and inbred lines provide a homogenous genetic background, both of which help filter out heterogeneous ‘noise’ from phenotypic analysis of genetic or epigenetic mutations. Mouse is an ideal model organism for genetic studies, because of the ease of genetic manipulation through ES cells which can then be used to derive new mouse lines. Finally, common inbred lines, such as 129S1 and C57BL/6J used in this study, have a large number of categorized SNPs, making allelespecific analysis possible, which is crucial in genomic imprinting research. Ease of gene targeting in ES cells, combined with the availability of a large number of mouse strains, and the ability of simple mouse crosses to bring mutations to homozygosity, have made genetic study in mouse an incredibly useful tool to study human disease.  1.7.1 Conservation of imprinting between mouse and human In addition to the benefits listed above, mouse is an important model organism because it is the closest living relative to human that combines short generation time, small size, and ease of genetic manipulation. The most common recent ancestor between mouse and human is 75 (±15) million years ago, the genomes of mouse and human are  9  40% identical, with coding regions on average 85% identical (Batzoglou, et al, 2000; Guenet, 2005; Pennacchio, 2003). More importantly there are 340-350 blocks of conserved homologous synteny, which suggests common mechanisms of gene regulation for many genes shared between the two organisms (Paulsen, et al, 2000; Pennacchio, 2003). This is especially relevant for imprinted genes, which tend to be found in clusters, the imprinting of each cluster controlled by one major imprinting control region (Paulsen, et al, 2000; Peters, et al, 2004). Furthermore, 80% of genes in humans have a one-to-one relationship with genes in mouse, meaning that the best alignment for a given mouse gene against human is the best alignment for a given human gene in mouse (Guenet, 2005; Pennacchio, 2003). This high level of conservation is also true for imprinted genes, where 29 of a total 71 genes confirmed imprinted in mouse have been shown also to be imprinted in human (Morison, et al, 2005).  1.7.2 Human imprinting disorders Some examples of genes with conserved imprinting between human and mouse are UE3A and IGF2, which cause Angelman’s and Beckwith-Wiedemann syndrome, respectively. Consistent with imprinted genes known role in growth, development, and behavior individuals with these disorders display fetal growth abnormalities, as well as marked behavioral phenotypes such as mental retardation, and obsessive-compulsive behavior (Cerrato, et al, 2008; Kent, et al, 2008; Lalande, et al, 2007; Peters, et al, 2004). Silver-Russell syndrome (SRS) is an imprinting disorder which results in severe pre- and/or post-natal growth retardation, can commonly present with triangular face shape, macrocephaly, pointed chin, and body asymmetry, and in some cases can include learning and/or developmental delays (Eggermann, et al, 2008; Zeschnigk, et al, 2008).  10  Candidate regions for this syndrome include human chromosome 7q32 and 11p15, which share synteny with proximal MMU6 and mouse chromosome 11, respectively (AbuAmero, et al, 2008). mUPD for human chromosome 7 is seen in 7-10% of cases, and microsatellite marker heterogeneity on chromosome 7 is consistent with trisomic rescue. Although persistence of trisomy has yet to be seen in patients, a relatively small number of tissues have been analyzed (solely leukocytes and fibroblasts). The main candidate, Peg1, shared with mouse, currently has no observed point mutations or epimutations in the SRS population, but other important candidates such as Peg10 and Klf14 have yet to be evaluated. An isolated case of partial mUPD for ICR2 on human chromosome 11 has been seen, as well as hypomethylation of ICR1 in 38-63% of SRS patients, the latter regulating H19 and IGF2. Because of the involvement of both human chromosome 7 and 11, and because of noted interactions between imprinted regions on separate chromosomes, it has been proposed that chromosomes 7 and 11 are part of an interactive imprinted gene network (Eggermann, et al, 2008). Future work to investigate the influence of change in imprinted expression on human chromosome 7q31 on human chromosome 11p15 (and vice versa) will be critical in understanding the extent and mechanism of this complex interaction.  1.8 Rationale and experimental approach The mechanism of normal imprint maintenance is currently unclear, but important for understanding developmental imprinting disorders, as well as somatic diseases such as cancer. Previous work done in parthenogenetic and androgenetic embryos suggests that UPD can cause mis-regulation of genomic imprinting, resulting in abnormal changes in imprinted gene expression and developmental failure. This research hopes to build on  11  previous work using a novel tool for studying maintenance of mitotically-derived UPD in ES cells. Mitotic UPD, which has never before been studied, is an important area of research that can be used to better understand somatic abnormalities in imprinting, as well as add to the body of work on imprint maintenance.  1.8.1 Research hypothesis UPD caused by mitotic abnormalities lead to disruption in genomic imprinting, which results in abnormal expression of imprinted genes and disruption of imprinted CpG methylation.  1.8.2 Research objectives My research aims to use UPD as a tool to characterize regulation and maintenance of imprinted gene expression and CpG methylation on MMU6, which contains two proximal imprinted gene domains. Since imprinting is highly conserved between human and mouse this research has the potential to provide a model for human diseases caused by imprinting abnormalities, particularly for carcinogenesis, and Silver-Russell syndrome cases caused by mUPD of human chromosome 7 (Eggermann, et al, 2005), which shares syntenic homology with proximal MMU6. It is important to learn what the consequences are of mitotic UPD on imprinting maintenance, since LOH ES cells derived by mitotic events are often used in research for purposes other than imprint investigation. In addition, a comparison between previous work on meiotic UPD to mitotic UPD will be made to assess whether mitotic LOH can provide a new experimental system to study imprinting in ES cells.  12  Figure 1.1. Lifecycle of genomic imprinting. The three stages of genomic imprinting are maintenance, erasure, and re-establishment. DNA methylation is maintained at imprinting control centers (IC) in somatic cells, but is erased in primordial germ cells. Sometime during maturation of gametes methylation imprints at ICs are re-established according to the sex of the embryo. A paternally imprinted gamete, with an imprinted methylation pattern characteristic of a paternal allele is passed on from the sperm, while a maternally imprinted gamete, with an imprinted methylation pattern characteristic of a maternal allele is passed on from the oocyte. These differential methylation patterns, rendering specific imprinted genes active and specific genes inactive, results in imprinted gene expression being passed on to the next generation. Reprinted by permission from Macmillan Publishers Ltd: Nature Review Genetics (Reik, et al, 2001), copyright (2001).  13  CHAPTER 2 Materials and methods 2.1 Nucleic acid work DNA was prepared from embryonic stem cell pellets (106-107 cells) lysed overnight using 4µL proteinase K (Roche) in 400µL proteinase K lysis buffer (Table 2.1). This was followed by phenol-chloroform DNA extraction repeated twice, using 500µL phenol-chloroform (1:1), 2-10min. of shaking, 1-10min. centrifugation, and transfer of aqueous phase to a clean microtube (Sambrook, et al, 2001). The 100% chloroform step was omitted. The DNA was then precipitated with 50µL sodium acetate, 3µL glycoblue (to make pellet more visible), and 500µL 100% ethanol. All samples were washed with 1mL 70% ethanol, and DNA samples used in Peg1 methylation-sensitive Southern hybridization were washed with an additional 70% ethanol step to remove excess salt. All DNA was re-suspended in 20-50µL TE (pH 7.4). RNA was prepared from embryonic stem cell or day 14/15 embryoid body pellets (~106-107 cells) using TRIzol Reagent (Invitrogen) according to manufacturer’s protocols using RNase-free reagents and DEPC-dH2O. Glycogen blue was added prior to 100% ethanol precipitation. RNA was dissolved in ~20µL DEPC-dH2O and kept on ice shortterm (up to 3 hours), -70°C long-term. 25µL DNase reactions with ~10µL RNA were carried out according to Promega manufacturer’s protocols. DNase-treated RNA was run in an RNase-decontaminated gel tank on 1% agarose gel alongside control non-DNase-treated RNA to assess success of DNase treatment. DNase-treated RNA was stored short-term (up to 3 hours) on ice, -70°C long-term.  14  First-strand cDNA synthesis was carried out with 10µL DNase-treated RNA. 9µL of master mix 1 was added to each reaction, consisting of 2µL N15 primers (10 ng/µL), 2µL 10mM dNTPs, 5µL DEPC-dH2O. Reactions were held at 65°C 5min., stored at 4°C if needed, then centrifuged. 19µL of master mix 2 was then added, consisting of 8µL 5x first-strand buffer (Invitrogen), 4µL 0.1M DTT (Invitrogen), 6.5µL DEPC-dH2O, and 0.5µL RNasin (Promega). After mixing, reactions were divided in 2 (19µL x 2), adding 1µL SuperScript II Reverse Transcription (Invitrogen) to one of these aliquots (RT+ reaction) and not the other (RT− reaction). All reactions (RT+ and RT−) were kept at 42°C 60min., then heat inactivated at 70°C for 15min. Synthesized cDNA was kept at room temperature (~25°C) short-term up to 3 hours, 4°C long-term.  2.2 Southern hybridization Approximately 10µg genomic DNA was run on 0.8% agarose in 1x TBE at 35V, depurinated with 0.1N HCl until loading dye fronts changed colour, and blotted overnight onto Amersham Hybond-N+ nitrocellulose membranes using 0.4M NaOH for neutral transfer according to manufacturer’s protocols. Prehybridization was carried out in 10mL PerfectHyb buffer (Sigma-Aldrich Inc.) at 65°C for 1 hour-overnight. Approximately 25ng of gel-purified genomic DNA probe was labeled with α32P-dCTP by random primed labeling using Roche High Prime and unincorporated label was removed using GE Healthcare nick column purification (according to manufacturer’s protocols). Overnight hybridization was done in 10mL PerfectHyb buffer at 65°C. Wash stringency and duration were adjusted depending on periodic Gieger counter observations, but in general were 1 hour at low stringency (2x SSC, 0.1% SDS) room temperature, 1 hour at low stringency  15  65°C, and 30min. high stringency (0.1x SSC, 0.1% SDS) 65°C. Blots were exposed to Kodak scientific imaging film for 1 week – month depending on signal strength.  2.3 Polymerase chain reaction and agarose gel electrophoresis Dinucleotide repeat polymorphisms and SNPs were used as genetic markers to genotype DNA extracted from ES cells (Table 2.2). PCR primers to detect these microsatellites were obtained from Mouse Genome Informatics (MGI), and some of these were redesigned using the website Primer3. PCR primers to detect SNPs were designed using SNPs between C57B6/6J and 129S1/SvImJ from the MGI website, and were designed using Primer3. Gt(ROSA)26Sor locus primers (Table 2.2) were obtained from Dr. Philippe Soriano from his laboratory website (Soriano), and wild-type reverse primer (R2I) was redesigned using Primer3. Reverse-transcriptase PCR primers were obtained from Dr. Fumitoshi Ishino (Ono, et al, 2006), Julie Hoscheit, and designed using Primer3 (Table 2.2). 1µL embryo-derived cDNA, 2µL embryoid body-derived cDNA, and 2µL ES cells-derived cDNA was used in reverse-transcriptase PCR reaction. Each PCR reaction was carried out in a 25µL reaction volume using 2.5µL MgSO4, 2.5µL 10x PCR buffer, 2µL 2mM dNTPs, 1µL of each primer, 0.1µL Fermentas Tsg-Taq DNA polymerase, and 15.9µL dH2O (autoclaved and UV light-treated 30min.). Primer concentrations were reduced by ½ (and amount of dH2O adjusted accordingly) in reactions with excessive primer dimers. General PCR reaction conditions were 94°C 2min. (denaturation), followed by 25-40 cycles of 94°C 30sec. (denaturation), 54-64°C 30sec. (annealing), 72°C 30sec. (extension), ending with 72°C 3min. (extension). PCR products were run on 1% agarose gels (Invitrogen) in 0.5x TBE, except for microsatellite expansion PCR reactions which were run on 4% NuSieve 3:1 agarose (Cambrex), and 16  post-stained in SYBR Green (Invitrogen) for 10-30min. at a concentration of 3-5µL SYBR Green in 250µL 0.5x TBE.. Unprocessed RT-PCR gel pictures taken with a digital camera were analyzed using ImageJ (National Institute of Mental Health). Plot profiles of each lane were used to display a column horizontally-averaged plot of pixel intensity. The background intensity for each lane was determined separately for each band by drawing a trend-line of the average background. The background immediately below the highest peak of intesity was determined, subtracted from the peak intensity value, and this number was used as the corrected value for band intensity. Theses values were made semi-quantitative by normalizing each sample’s band intensity to Gapdh. One replicate was analyzed as above for each sample at both 25 and 30 cycles of RT-PCR, although PCR reactions were performed between 3-8 times using 1-3 separate RNA extractions, and 1-5 separate cDNA synthesis RT reactions. Approximately 15% variability was seen in PCR amplication for identical samples under identical PCR conditions. One major source of variability could come from determining background levels in image analysis: although every effort was made to be consistant, background measurements are subjective to a certain degree, and were harder to determine for specific lanes, particularly for lowly expressed genes with high background in the lane. Varaibility could also come from signal saturation that has probably occurred for Gapdh intensity measurements. This control is much more highly expressed than any of the genes analyzed, making it diffucult to find a PCR cycle number where the gene of interest could be detected without saturating the Gapdh control. Some normalization data was thus lost from the analysis, making it possibly that some of the expression differences between cell lines was attributable to differences in starting cDNA material, rather than there actual  17  relative expression level. Even with the variablilty seen, however, consistent differences between specific samples and the parental control were seen for specific genes such as Peg1, Peg10, Sgce, and Mit1.  2.3.1 Sex chromosome genotyping of newly derived F1 ES cells New ES cell lines derived from reciprocal F1 crosses between 129S1 and Gt(ROSA)26Sortm1Sor (C57BL/6J background) were typed for sex chromosome complement using PCR primers designed against a known insertion/deletion (indel) polymorphism between the X and Y chromosome within the Jarid1c gene on the X chromosome and its homolog Jarid1d on the Y chromosome. A microsatellite repeat expansion polymorphism between 129S1 and C57BL/6J, DXMit210, was used to distinguish between one and two X chromosomes, and to confirm X chromosome inheritance from either 129S1 or C57BL/6J.  2.4 Mice maintenance and conditions The mice were maintained in windowless rooms in the animal unit of the Department of Medical Genetics at the University of British Columbia. Mice were housed in 5” x 11” x 7” polycarbonate cages with stainless lids and supplied food and water ad libitum. The room was maintained at a temperature of approximately 70°F and on a light cycle from 6:00 AM to 6:00 PM. Density of mice was 2-5 per cage.  2.5 R3/R3 mouse stocks B6.129S4-Gt(ROSA)26Sortm1Sor/J (R3/R3) (JAX Stock#003474) is an inbred strain that was generated by Dr. Philippe Soriano on a 129S4/SvJaeSor background and  18  subsequently crossed to C57B6/6J utilizing a marker-assisted protocol ( JAX® Mice Database). R3/R3 was derived from a 129S4/SvJaeSor ES cell line targeted at the Gt(ROSA)26Sor locus, which produces a non-coding RNA transcript, ubiquitously expressed in all tissues throughout development and adulthood (Soriano, 1999). The targeting vector includes a splice acceptor sequence, a PGK promoter-driven neomycin resistance (neo) expression cassette terminated by a triple polyadenylation sequence and flanked by loxP sites, and a conditional lacZ (β-galactosidase) gene. R3/R3 is maintained in a homozygous state and has no known phenotype except for below average litter sizes (typically 3-6 pups/litter).  2.6 Murine embryonic stem cell culture ES cells culture solutions and culture conditions were maintained according to Isolation and Culture of Blastocyst-Derived Stem Cell Lines (Chapter 8) from Nagy, et al. (Nagy, 2003) using solutions detailed in Table 2.3. ES cells were grown on Mitomycin Cinactivated primary mouse embyronic fibroblast feeders (PEFFs) on 0.1% gelatin, except when grown solely on gelatin in order to collect ES cells to be used for RNA or genomic DNA extraction, prior to G418 selection, or prior to embyroid body differentiation. 0.025% trypsin was used for ES cells, while 0.005% Trypsin (diluted in PBS) was used for PEFF passaging.  2.7 High G418 selection G418 solutions were prepared from G418 powder dissolved in autoclaved, sterile dH2O and filter sterilized. Parental lines (R3/+ or +/R3) are hemizygous for the neomycinresistance gene at the Gt(ROSA)26Sor locus and are resistant to standard/low levels of  19  G418 (0.15mg/mL) (Geneticin, Invitrogen #11811-031). High G418 selection was carried out using G418 concentrations between 0.75mg/mL and 7mg/mL. Selection for LOH at the Gt(ROSA)26Sor locus was only successful at G418 concentrations between 5mg/mL and 7mg/mL. Cells were plated at 5x105 cells/100mm plate and 1x106 cells/100mm plate. Two controls were used: R129E2 (R3/+) parental line and Oct4 SLR1 EGFP (+/+ with no neo resistance gene) were plated at 3.6x105 cells/60mm plate with both 0.150mg/mL (standard/low G418 concentration) and 0mg/mL. This provided a comparison for no cell death, and no growth with complete cell death, in order to determine the success of high G418 selection. Selection was carried out for 10-15 days, until colonies were large enough to pick, with G418-ES cells DMEM changes every day until cell death was obvious, followed by changes every 2-3 days. Colonies were picked 48 colonies at a time, placing ½ of the colony into 96-well conical bottom plates with 100uL trypsin/well and ½ of the colony into PCR tubes with 20uL Proteinase K PCR lysis buffer. This was accomplished by taking about 10uL of PBS with a whole picked colony, pipetting up and down on the side of the PCR tube (not coming into contact with the lysis buffer) to homogenize the colony, leaving 5uL of homogenized colony in the PCR tube for PCR screening, and transferring 5uL to trypsin in 96-well plates. ½-colonies for culture were trypsinized, plated, and cultured according to Isolation and Culture of Blastocyst-Derived Stem Cell Lines (Chapter 8) from Nagy, et al. (Nagy, 2003). ½-colonies for PCR screening were immersed in Proteinase K PCR lysis buffer, 1uL Proteinase K was added, and cell lysis reactions were carried out by incubation at 50°C for 90min. up to overnight. Multiplex PCR screening was carried out using LF-2, R2I-2, R1295-2 primers for the Gt(ROSA)26Sor (Table 2.2) with the following PCR conditions: 1-2uL cell lysate + 13-14uL dH2O at 94°C 10 min. (proteinase  20  K inactivation), 85°C pause (PCR master mix addition), 94°C 2 min. (denaturation), 40 cycles of 94°C 30sec. (denaturation), 59°C 30sec. (annealing), 72°C 30 sec. (extension), followed by 72°C 3 min. (extension). Colonies in which amplification of the wild-type Gt(ROSA)26Sor allele was identified as probable LOH were expanded in tissue culture.  2.8 LacZ staining 1 x 108 R26r3(+/-) ES cells were electroporated with the Cre-expressing vector, pCx-Cre-Puro while suspended in 800µL electroporation buffer in a BioRad 4.0mm Gap electroporation cuvette. Electroporation parameters were the following: set voltage: 250V, observed voltage: 280V, time constant: 7.5ms. Cre-electroporated cells, R26r3(+/-)-Cre, were then placed on ice approximately 10 minutes, then plated on 3 x 100mm tissue culture-grade plates and incubated at 37°C, 5%CO2. After 3 days of puromycin selection, cells were washed twice with PBS Ca2+Mg2+ (Table 2.1), and fresh fixative solution was added. After incubating 10min. at 25°C cells were washed three times with PBS Ca2+Mg2+, and 400µL fresh X-Gal stain was added. Cells were incubated in stain overnight in the dark, and pictures of the stained cells were taken the next day.  2.9 Embryoid body differentiation ES Cells were grown 1-2 passages on gelatinized plates in order to prevent contamination with fibroblast feeder cells. Cells were then passaged 1:3 onto ungelatinized bacterial-grade petri dishes by trypsinization followed by plating cells using gentle pipetting in order to promote cell aggregation. ES cell media without LIF was changed 3 days later, then every 2-3 days, and the resulting cystic embryoid bodies were collected on day 15. 21  Table 2.1 Laboratory solutions. Solution Proteinase K Lysis Buffer  lacZ Stain  Fixative Solution PBS Ca2+Mg2+  Components 100mM NaCl 10mM Tris-Cl 25mM EDTA 0.5% SDS in dH2O 8.78mL PBS Ca2+Mg2+ 500µL 100mM K4Fe(CN)6 500µL 100mM K3Fe(CN)6 200µL 20mg/mL X-Gal in DMSO 1M MgCl2 20mL PBS Ca2+Mg2+ 160µL 25% grade II glutaraldehyde 200mL PBS 100µL 1M MgCl2 2mL 0.1M CaCl2  Sterilization N/A  N/A  N/A Autoclaved  22  23  Primer Name Col1a2-F5: Col1a2-R5: Copg2-ex1-R1 Copg2-ex5-F2 D6Mit15L D6Mit15R D6Rp2-F2 D6Rp2-R2 DXMit210-F DXMit210-R G3PDH F G3PDH R Klf14-F3 Klf14-R3 KT44 KT45 Mit1-F1 Mit1-R3 Nap1l5-F Nap1l5-R Peg1 5'U-1F Peg1-2Ra Peg10-7700-F1 Peg10-7700-R1 Pon2-F7 Pon2-R7 Ppp1r9a-F4 Ppp1r9a-R4 R1295-2(mt) R2I-2(wt) RosaLF-2 Sgce-F3 Sgce-R3 SMCX-1 SMCY-1 Tsga14-F Tsga14-R  Sequence 5’-3’ AACAGCATTGCGTACCTGG GTGCAATGTCAAGGAACGG TCCAATCCTTTCCAGCATTT TCCTTGATGG TGAGGTAGCA CACTGACCCTAGCACAGCAG TCCTGGCTTCCACAGGTACT CAATGGAGGTAGAAAACTGAAGC TTCTATGAAATTAGTCCTTAATAAATCAGT GGGATAAAGTCTGAACTGTAGAAAGG AATGATGATT ACTGACTTGC TCTCC ACCACAGTCCATGCCATCAC TCCACCACCCTGTTGCTGTA ACACCCTCTCCAAAGTCCGCCCT CAAGCGACATCAGTGCTCCTTCCAT CAAGCTGCCTTTGCACATGGC CCATGGGTCCATAGCTCGGGC AACAAAACTAGCTTTACTTGAGAG CCACTTGGATCTGTAACTGTA CACACTACACCAGAACATCCAA AACAAACACTACCAAGAAAACAGG TGAGAGAGTGGTGGGTCCAAGTAG CTTCCATGAG TGCAGAGCAG G ACTTACAATTGCCGAGCTCC GGCACAACGATTATTCGTCC ATGGACAGAGGCTCTTCGTG AGCGCATCAGAATTGCAAG ACTCTCCTGCCGAGGATG CAGTTTCAGGGGCTCTCACT GCGAAGAGTTTGTCCTCAAC CACACACCAGGTTAGCCTTTA AAAGTCGCTCTGAGTTGTTATCAGT AGCCTGGGGAGGTTAGTAATGA TGCATCTTAACTGATTCTGTGGATT CCGCTGCCAAATTCTTTGG TGAAGCTTTTGGCTTTGAG GTAGACAAAGGGCTCGTAAAGAA GGTTCATTGTCCTGGATAGAGTT  Function RT RT RT RT G G G G G G RT RT RT RT G G RT RT RT RT RT RT RT RT RT RT RT RT G G G RT RT G G RT RT  Reaction Col1a2 Col1a2 Copg2 Copg2 D6Mit15 D6Mit15 D6Rp2 D6Rp2 DXMit210 DXMit210 Gapdh Gapdh Klf14 Klf14 Ptn Ptn Mit1/Lb9 Mit1/Lb9 Nap1l5 Nap1l5 Peg1 Peg1 Peg10 Peg10 Pon2 Pon2 Ppp1r9a Ppp1r9a Gt(ROSA)26Sor Gt(ROSA)26Sor Gt(ROSA)26Sor Sgce Sgce Jarid1c Jarid1d Tsga14 Tsga14 (Ono, et al, 2006) (Ono, et al, 2006) (Ono, et al, 2006) (Ono, et al, 2006) (Ono, et al, 2006) (Ono, et al, 2006) (Soriano, 1999) (Soriano, 1999) (Soriano, 1999) (Ono, et al, 2006) (Ono, et al, 2006) (Mouse Genome Informatics) (Mouse Genome Informatics)  (Parker-Katiraee, et al, 2007) (Parker-Katiraee, et al, 2007) (Mouse Genome Informatics) (Mouse Genome Informatics)  (Mouse Genome Informatics) (Mouse Genome Informatics) (Mouse Genome Informatics) (Mouse Genome Informatics) (Mouse Genome Informatics) (Mouse Genome Informatics)  References (Ono, et al, 2006) (Ono, et al, 2006)  Table 2.2. Oligonucleotide primers used in genotyping and RT-PCR. G: genotyping primer. RT: RT-PCR primer.  23  Table 2.3. Tissue culture solutions. Solution ES cells-DMEM  PEFM  0.025% Trypsin  Mitomycin C (Sigma) PBS (Ca2+, Mg2+ free)  2x Freezing Solution  0.1% Gelatin  Components 400mL Dulbecco's modified Eagle's medium (DMEM) (Chemicon Intl. Embryomax) 5mL 200mM L-glutamine (Gibco) 5mL 10mM 2-mercaptoethanol (Sigma) 5mL Penicillin/Streptomycin (10,000 U Pen; µg Strep) (Gibco) 5mL 10mM MEM Non-essential Amino-acids (Gibco) 5mL 100mM Sodium pyruvate (Gibco) 75mL FBS (Hyclone or Wisent Inc. Multicell) 1 x 106 units Lif ( Chemicon Intl. #LIF2010) 500mL DMEM, w/o L-glutamine (Gibco) 6mL Penicillin/Streptomycin (10,000 U Pen; µg Strep) (Gibco) 56mL FBS (Gibco) 0.35g/L NaHCO3 0.4g/L KCl 0.0g/L Phenol red 1.0g/L glucose 8.0g/L NaCl 0.048g/L Na2HPO4 (anhydrous) 0.2g/L EDTA dH2O to 90mL 10mL 0.25% Trypsin (Gibco) 1.0mg/mL in dH2O 16g NaCl  Sterilization N/A  N/A  Autoclaved before addition of trypsin  Filtersterilized Autoclaved  0.4g KCl 2.88g Na2HPO4 0.48g KH2PO4 dH2O to 2L 60% ES cells -DMEM 20% FBS (Hyclone or Wisent Inc. Multicell) 20% DMSO (Sigma) Gelatin dissolved in dH2O to 0.1%  Autoclaved  24  CHAPTER 3 Derivation and characterization of embryonic stem cell lines with loss-ofheterozygosity on mouse chromosome 6 3.1 Introduction Imprinting studies have been typically carried out on uniparental mice or on fertilized mice with a uniparental disomy (UPD) for a specific chromosome. These UPDs originate from a meiotic non-disjunction event involving reciprocal or Robertsonian translocations, such that the uniparental chromosomal regions are present at fertilization. In this thesis, I present the first known attempt at using mitotic recombination or nondisjunction to induce UPD in ES cells to study its consequence on imprinted gene expression. This could be an important tool for studying early and late maintenance of imprints in a UPD environment, as adjustments to imprinted gene methylation and expression may occur within an early window of time that would be missed with meiotic events. Spontaneous loss of heterozygosity (LOH) in ES cells not only results in marker homozygosity or hemizygosity, but can also cause UPD. Therefore mitotic LOH in somatic cells could have important consequences for the expression of imprinted genes located on the affected chromosome. The three different mechanisms of LOH: full chromosome duplication and loss; mitotic recombination; and gene conversion, each result in full UPD, partial UPD, and local UPD respectively. LOH can also occur by deletion, which leads to hemizygosity, and thus formally LOH. This mechanism will not be discussed further, since the selection for spontaneous LOH in this thesis is based on gain of an extra copy of a marker, not loss of a marker.  25  3.1.1 Mechanisms of mitotic LOH Full chromosome duplication in a diploid cell can occur via a mitotic nondisjunction event, followed by chromosome loss. Prior to chromosome loss there should be a trisomic intermediate stage, which may exist for varying amounts of time depending on how well trisomy for the particular chromosome and/or homolog is tolerated in the specific cell type. During chromosomal loss there is a 1/3 chance of losing the monosomic parental homolog, which will result in LOH and UPD for the entire chromosome. For example, for a trisomy of MMU6 with two maternal and one paternal homologs, LOH and mUPD6 will only result upon loss of the paternal homolog. This method has been thought to be the most common mechanism of LOH in ES cells, and the 2nd most common in somatic cells such as fibroblasts (Cervantes, et al, 2002; Lefebvre, et al, 2001). Mitotic recombination has been shown to be the 2nd most common mechanism of LOH in ES cells, and the most common mechanism in fibroblasts (Cervantes, et al, 2002; Lefebvre, et al, 2001). This apparent suppression of mitotic recombination in pluripotent cells such as ES cells and other stem cells is thought to be critical in maintaining genome integrity. LOH could result in unmasking of deleterious recessive alleles and the potential for cancer, which would be passed down to more differentiated lineages. Gene conversion is the least common mechanism in both ES cells and fibroblasts, and is hypothesized to occur by a strand-invasion mechanism resulting in a double-crossover event that contains a relatively small area i.e. 5Mb or less. This work shows that, with the appropriate selection, it is possible to recover ES cell line variants in which a mitotic event such as chromosome duplication or mitotic recombination has occurred, leading to LOH. In both of these cases, we expect not only genetic consequences (LOH), but also epigenetic consequences, if the region involved  26  happens to contain imprinted genes (UPD) (Lefebvre, et al, 2001). In this study, the main goal is to assess the consequences, if any, of LOH and “mitotic UPD” on the maintenance of epigenetic imprints and the expression of imprinted genes.  3.1.2 Rationale for using chromosome 6 and the ROSA26 locus MMU6 has two proximal imprinted gene domains separated by 23.8Mb. Imprinted domain 1 (ID1) includes 9 imprinted genes (7 MEGs and 2 PEGs) and spans 3.2Mb (3.6Mb – 6.9Mb from the centromere) (Hoshiya, et al, 2003; Kile, et al, 2000; Monk, et al, 2008; Ono, et al, 2003), while imprinted domain 2 (ID2) includes 6 imprinted genes (2 MEGs and 4 PEGs) and spans 28.1Mb (30.6Mb – 58.8Mb from the centromere) on the 147.9Mb chromosome 6 (Beechey, 2004; Kaneko-Ishino, et al, 1995; Lee, et al, 2000; Parker-Katiraee, et al, 2007). Of 5 known CpG islands overlapping with these imprinted genes, three are known DMRs. The DMRs of Peg10 and Peg1 have been shown to be established in the germline (Lefebvre, et al, 1997; Ono, et al, 2001), while the Nap1l5 DMR is known to be established by E10 or earlier (Davies, et al, 2004; Smith, et al, 2003). They are located at Peg10 in ID1, and at Peg1 and Nap1l5 in ID2 (Ono, et al, 2003; Smith, et al, 2003). As mentioned earlier (Section 1.3, page 4), Peg10 and Peg1 are thought to be ICRs that control their respective imprinted domains. While MMU6 imprinted genes have been fairly well characterized in terms of knock-out phenotypes (Table 4.1, Table 4.2), the maintenance and control of imprinting on this chromosome has received little attention. This combined with the presence of possible candidate genes for Silver-Russell syndrome (Eggermann, et al, 2008; Kobayashi, et al, 1997) call for further studies of MMU6. The ability to derive ES cell sub-lines with different regions of UPD for MMU6 provides a unique tool where imprinted regions can be dissected and studied in  27  near isolation, which should allow candidates for both Silver-Russell syndrome and ICRs to be identified. The system also allows for investigation of imprinting maintenance with mitotic UPD, and for the study of interactions between imprinted domains when one domain is UPD while the other is biparental. The Gt(ROSA)26Sor locus, ROSA26 for short, is an ideal candidate to work with for study of LOH, and thus UPD, on MMU6. ROSA26, on distal MMU6 at 113Mb, has already been targeted in ES cells with a mutant neo gene, allowing for selection of spontaneous mitotic LOH. Targeted ES cells, as well as a mouse line, are readily available. This saves the time and effort of targeting the locus, and allows us to derive early passage ES cell lines from the mouse line.  3.1.3 Starting ES cell lines R26R ES cell line on the 129S4 strain background The mouse line known as Gt(ROSA)26Sortm1Sor (or R26R) carries an insertion at the Gt(ROSA)26Sor locus (ROSA26) on MMU6 originally targeted in AK7 ES cells of the 129S4 strain background. This allele is the targeted knock-in of a cassette which included a 5’ splice acceptor (SA), mutant neomycin gene (neo) driven by a PGK promoter, loxP sites flanking neo, a promoter-less lacZ, and a 3’ polyadenylation sequence (pA) (Figure 3.1) (Soriano, 1999). It was generated as a reporter mouse strain for Cre excision, which is monitored by activation of the conditional lacZ gene. The original ES cell line heterozygous for the R26R insertion allele was donated by Dr. Philippe Soriano for our use, and is referred to hereafter as the R26r3(+/-) ES cell line. This mutant neo allele, unlike the wild-type neo, is a hypomorphic allele which confers dose-sensitive resistance to the drug G418 (Yenofsky, et al, 1990). Mammalian  28  cells hemizygous for this mutant form of neo are resistant to low, or standard, doses of G418 (0.15mg/mL), while cells homozygous, or with two doses of mutant neo, are resistant to high doses of G418 (from 0.75mg/mL potentially up to 8mg/mL). This feature makes it possible to select for spontaneous LOH for the locus where neo (mutant) has been targeted, which on average occurs at a frequency of 10-5 per cell generation (Lefebvre, et al, 2001). The loxP sites flanking the neo marker can be used for Cre-mediated excision of neo, moving the lacZ next to the endogenous ROSA26 promoter, which then drives lacZ expression. In cells homozygous for the R26R insert this characteristic of the cassette could be used to reintroduce a polymorphism between the two homologs, rendering the cell line resistant to low doses, but sensitive to high doses of G418. It is then possible to perform a second round of high G148 selection to select for a second spontaneous LOH event at the targeted locus. Derivation of reciprocal F1 ES cell lines heterozygous for the R26R insert. After mice were derived from R26R ES cells this mouse line was acquired by Jackson Laboratories and was then backcrossed to C57BL/6J for 5 generations, followed by intercrossing. The resulting homozygous mouse line, Gt(ROSA)26Sortm1Sor (referred to as R26r3(-/-)), is therefore a congenic line, on a nearly homozygous C57BL/6J background, but with small and uncharacterized regions of 129S4 DNA, including an undefined region flanking the insert at the ROSA26 locus. The lack of heterozygosity in the pre-existing R26r3(+/-) ES cell line (pure 129S4 background) would not permit one to analyze the mechanism of LOH, and thus the extent of UPD on MMU6. To make more useful ES cell lines for studying imprinting where we would be able to determine the  29  extent of UPD our laboratory derived new ES cell lines from blastocysts obtained from reciprocal F1 crosses between R26r3(-/-) and 129S1 mice. R26R-/- female x male crosses were named R129 (R26R insert inherited from female) while 129S1 female x R26R-/male crosses were named 129R (R26R insert inherited from male). The results below describe the characterization of the newly derived F1 ES cell lines as well as the F1 ES cell sub-lines derived from high G418 selection.  3.2 Results 3.2.1 LacZ staining of Cre-electroporated R26r3(+/-) ES cells Previous work in the lab showed that the R26r3(+/-) ES cells are resistant to low doses of G418. At the start of my thesis project I wanted to confirm that, as expected, the R26R allele could be transformed by Cre-mediated excision of the neo cassette (Figure 3.1). This should activate the lacZ gene, a phenotype which can easily be detected by staining the cell with the colorimetric substrate X-Gal. Overnight lacZ staining of these R26r3(+/)-Cre ES cells demonstrated lacZ expression in the majority, but not all ES cells (Figure 3.2).  3.2.2 Sex chromosome complement in newly derived F1 ES cell lines Ten ES cell lines were successfully derived and were typed for sex chromosome complement as described in Materials and Methods. The derived ES cell lines were determined to be XX for 4 lines (129R-B5, R129-B1, R129-D3, R129-E1) by the presence of both a 129 and C57BL/6J X chromosome polymorphism (Figure 3.3), as summarized in Table 3.1. The remaining 6 lines (R129-E2, R129-E3, R129-E4, R129-F1, R129-F2, R129-F4) were determined to be XY by the presence of a C57BL/6J and absence of a 129  30  X chromosome polymorphism (Figure 3.3), as well as the presence of a Y chromosome polymorphism (Figure 3.4), as summarized in Table 3.1. Because of the known global hypomethylation and epigenetic instability of XX ES cell lines (Zvetkova, et al, 2005), a male XY line, R129-E2, was chosen to proceed with selection of spontaneous LOH through high G418 selection, using the protocol as previously described in Chapter 2.  3.2.3 Screening for spontaneous LOH at the ROSA26 locus through high G418 selection Four separate attempts to obtain LOH at the ROSA26 locus were made using high G418 selection (Table 3.2). Concentrations of G418 ranged from 0.75mg/mL to 7mg/mL during the four separate attempts of high G418 selection. A total of four hundred sixtyfive colonies were screened for loss of the wild-type ROSA26 allele using a genomic PCR assay and confirmed with Southern blot analysis (Figure 3.5). Seven ES cells sub-lines (referred to as R129*-E2- 1.4.6, 4.4.4, 4.5.2, 4.5.6, 4.6.3, 4.10.7, 4.11.6) were recovered using G418 concentrations of either 5mg/mL or 6mg/mL, which showed loss of the wildtype ROSA26 allele through PCR and Southern blot analysis.  3.2.4 Characterization of the extent of LOH in seven ES cell sub-lines Characterization of LOH with microsatellite markers Figure 3.6 shows that four sub-lines, 1.4.6, 4.4.4, 4.5.2, and 4.11.6, had lost the wild-type ROSA26 allele, but were heterozygous for C57BL/6J (B6) and 129S1 alleles at both a distal (D6mit15) and a proximal marker (D6Rp2, Ptn). Sub-line 4.5.6 had loss of wild-type ROSA26 and loss of the 129 distal marker D6mit15, but was heterozygous at  31  the proximal markers. Two sub-lines, 4.6.3 and 4.10.7, had loss of the 129S1 alleles at all markers tested, along with loss of wild-type ROSA26. Illumina genome-wide SNP analysis of R129-E2 and select R129*-E2 sub-lines After characterization using microsatellite markers the parental line R129-E2 and 4 sub-lines, 4.5.6, 4.6.3, 4.10.7, and 4.11.6, were analyzed using the Illumina Inc. (Toronto, Ontario) mouse medium density (md) linkage panel for whole genome SNP genotyping (Figure 3.7). R129-E2 and the four sub-lines show a region of 129S4 homozygosity for a 67.7Mb – 75.9Mb region on distal MMU6 (starting between 49.8Mb to 53.7Mb and ending between 121.3Mb to 125.7Mb). Line 4.6.3 showed loss of 129S1 markers for all typed MMU6 SNPs, except for the distal region of 129S4 homozygosity present in all tested lines. Line 4.5.6 showed loss of distal 129 markers starting between 46.1 – 63.9Mb. Line 4.10.7 is similar to 4.5.6, but with loss of 129S1 markers starting between 27.9Mb – 29.4Mb, and a proximal neighbouring region of 129/B6/B6 trisomy between 24.8Mb – 23.3Mb in size. Line 4.11.6 showed no loss of 129S1 markers, and was 129/B6 for the entire typed length of MMU6. Aside from MMU6, all lines shared a 17.2Mb – 26.1Mb region of 129S4 homozygosity on MMU15 (starting between 4.2Mb to 8.3Mb and ending between 25.5Mb to 30.4Mb), as well as B6 homozygosity on the X chromosome. The former represents a region of 129S4 that remained from the original 129S4 mouse line after backcrossing to C57BL/6J. The latter confirms the sex-chromosome genotyping, in that R129-E2 is an XY cell line with the X chromosome being inherited from the female R26r3(-/-), which is C57BL/6J background.  32  Line 4.6.3 has a unique region of 129 homozygosity starting between 20.4Mb – 28.3Mb continuing to the distal-most marker on MMU2, and a unique region of 129/129/B6 trisomy starting between 59.9Mb – 69.8Mb continuing to the distal-most marker on MMU13. These rearrangements are not shared in any of the other lines, which means they had to occur in the original 4.6.3 colony picked after high G418 selection. Line 4.6.3 was the only line to have incurred rearrangements, which makes it likely that the majority of cell lines, including R129-E2, are genetically stable in ES cell culture.  3.3 Discussion 3.3.1 Verification of R26R insert and consequences of high G418 selection Before beginning work on the newly-derived R129-E2 ES cell line, the insertion on the original ES cell line, R26r3(+/-) was evaluated for integrity of loxP sites surrounding the mutant neo gene. LacZ staining of R26r3(+/-)-Cre ES cells, which were electroporated with a Cre-expressing vector, provided evidence of the integrity of the R26R cassette construct in R26r3(+/-) ES cells. The cassette was shown to contain loxP sites flanking the neo, which would be necessary for Cre-mediated excision, relocating lacZ and allowing the endogenous ROSA26 promoter to drive lacZ expression. Not all ES cells stained with lacZ (Figure 3.2), which is expected as Cre-mediated loxP excision has between 2-9% efficiency in mouse cell lines (Sauer, et al, 1989). After validating the original R26R3(+/-) ES cell line R26R insert, high G418 selection was performed on the R129-E2 ES cell line, made possible by dose-sensitive G418 resistance conferred by a mutant neo on the R26R insert. Surprisingly, a large number of ES cell colonies, 458, survived high G418 selection (Table 3.2) that did not have loss of the wild-type ROSA26 allele and were not LOH. This significant amount of  33  background G418 resistance was surprising given that LOH has been able to be selected for with G418 levels between 1 – 2mg/mL at other loci (Paludan, et al, 1989; Yenofsky, et al, 1990), whereas selection for LOH at ROSA26 was only successful at 5 – 6mg/mL. LOH is expected at a frequency of 10-5 per generation on average per locus (Lefebvre, et al, 2001). ES cells survived at a much higher frequencies than expected under high G418 (101 – 104 times the expected rate, with negligible cell death at G418 <1.5mg/mL). Even at the highest G418 concentrations of 5mg/mL – 7mg/mL ROSA26 LOH was observed ten-fold lower than the reported genome-wide average rate of LOH (Lefebvre, et al, 2001), at1.5 x 10-6. The large number of high G418-resistant non-LOH ES cells could be a result of the highly-expressing ROSA26 locus having an influence on neo expression, through nearby enhancers or a transcriptionally-permissive epigenetic chromosomal environment. The strong promoter in combination with the high level of expression normally seen at ROSA26 could make these ES cells resistant to higher than normal levels of G418 than other loci. Another reason for high background could be because of local epigenetic changes that up-regulate neo, or a local duplication of the neo. To distinguish between these possibilities, quantitative PCR on neo could be performed using genomic DNA from non-LOH lines to assess for presence of local duplications. Analysis of histone modifications at the site of R26R integration could be used to look for an increase in active histone marks associated with non-LOH high G418 resistant ES cells.  34  3.3.2 Characterization of R129-E2 ES cell line uncovers multiple chromosomal rearrangements Seven ES cell sub-lines, with LOH of ROSA26, were obtained through high G418 selection. These R129*-E2 sub-lines, as well as the parental R129-E2 ES line, were then genotyped to characterize the original parental line, as well as the extent of LOH in R129*-E2 sub-lines. The R26R mouse strain had been previously uncharacterized in terms of residual 129 regions remaining in the genome. In particular, the extent of 129 homozygosity surrounding the ROSA26 locus was unknown and could pose problems for determining the extent and/or mechanism of LOH after high G418 selection. From the Illumina SNP analysis it was discovered that the 129 homozygosity was started between 49.8Mb – 53.7Mb and ended between 121.3Mb – 125.7Mb (Figure 3.7, Figure 3.8). Four R129*-E2 sub-lines, 1.4.6, 4.4.4, 4.5.2, 4.11.6, renamed GC-MMU6ROSA26 (GC6.1, GC6.2, GC6.3, GC6.4 respectively), are thought to have LOH caused by gene conversion, due to the presence of biallelic markers both proximal and distal to the region of 129 homozygosity. This could be confirmed by genotyping SNPs between 129S1 and 129S4. An alternate explanation for the loss of wild-type ROSA26 in these sub-lines could be a local deletion of this locus, or a double cross-over event within this sole region of homozygosity on an otherwise F1 heterozygous chromosome. R129*-E2 sub-line 4.6.3 was shown to be a full chromosome loss / duplication because of the complete loss of 129S4 markers along MMU6 (outside of the pre-existing region of 129 homozygosity in the parental line) (Figure 3.7, Figure 3.8). Therefore, this sub-line has not only LOH, but is also mUPD for both proximal imprinting domains on MMU6, and was renamed mUPD6 (MD6) accordingly. Sub-line 4.10.7 had undergone mitotic recombination with a breakpoint somewhere between Peg10 and Asb4 in  35  imprinting domain 1 (Figure 3.7, Figure 3.8), and was renamed mUPD6-ID2/partial mUPD6-ID1 (MD6-2/p1). While imprinted domain 2 has full mUPD it is unclear to what extent imprinted domain 1 has mUPD. Sub-line 4.5.6 is also a mitotic recombination sub-line with a breakpoint between imprinting domain 1 and 2, but has a regional duplication within imprinting domain 1 (Figure 3.7, Figure 3.8), and was renamed mUPD6-ID2/dup-ID1 (MD6-2/dp1). This regional duplication and neighboring distal LOH most likely occurred through an unequal mitotic recombination event. It is unclear to what extent this regional duplication will affect the imprinted methylation and gene expression in imprinted domain 1, given that chromosomal trisomies do not normally have the expected 1.5-fold increase in gene expression (Sommer, et al, 2008). The duplication of a region of the C57BL/6J, maternal homolog, results in two active and one inactive allele for MEGs, and one active and two inactive alleles for PEGs. MEG expression would be expected to increase roughly twofold. PEG expression, however, would be similar to expression from a normal, nonduplicated region unless there is leaky expression of PEGs from the maternal allele. Outside of MMU6 a number of genomic rearrangements took place in some sublines. None of the ES cells sub-lines had trisomy for MMU8, which has been reported to be quite common, and to potentially provide a growth advantage for ES cells (Park, et al, 1998; Sugawara, et al, 2006). Sub-line 4.6.3 seems to have had an unequal translocation event between MMU2 and MMU13, resulting in distal 129S1 homozygosity for the majority of MMU2 and a large region of distal B6/129/129 trisomy for MMU13. The chromosomal rearrangements on MMU6, MMU2, and MMU13 in 2/7 sublines suggests that mitotic recombination in ES cells may be more common than previously reported (Cervantes, et al, 2002; Lefebvre, et al, 2001). Possible explanations  36  are that some LOH studies lacked the necessary proximal markers able to distinguish between mitotic recombination and full chromosome duplication, or that that different LOH mechanisms predominate depending on the specific chromosome or chromosomal region. Level of homology between chromosomes has been shown to play a major role in the frequency of mitotic recombination events (Shao, et al, 2001), so differences in frequency may represent differences between mouse genetic backgrounds, and differences in degrees of homology between non-sister chromatids.  37  Figure 3.1. R26R insert on mouse chromosome 6 (MMU6) at Gt(ROSA)26Sor locus (ROSA26). Distances shown are Mb from centromere. Chromosome is to scale, R26R insert is not to scale. R26R insert includes (from 5’ to 3’) a splice acceptor (SA), loxP site, PGK promoter-driven mutant neomycin (neo) resistance gene, loxP site, promoter-less lacZ gene, polyadenylation sequence (pA). ID1: Imprinted domain 1. ID2: Imprinted domain 2. 38  Figure 3.2. LacZ staining of the Cre-electroporated ES cell line R26r3(+/-)-Cre.  39  Figure 3.3. X chromosome PCR genotyping in ES cells. PCR genotyping of the X chromosome in ES cell lines using PCR primers flanking the DXMit210 microsatellite repeat expansion polymorphism between 129S1 and C57BL/6J. Expected sizes: 129S1 (129) X chromosome polymorphism: 124bp. C57BL/6J (B6) X chromosome polymorphism: 114bp. ES cell lines: R129-B1 (B1): B6/129, R129-D3 (D3): B6/129, 129R-B5 (B5): B6/129, R129-E1 (E1): B6/129, R129-E2 (E2): B6, R129-E3 (E3): B6, R129-E4 (E4): B6, R129-F1 (F1): B6, R129-F2 (F2): B6, R129-F4 (F4): B6. B6/129 results indicate XX ES cell lines. B6 results indicate either XO or XY ES cell lines. 40  Figure 3.4. XY chromosome PCR genotyping in ES cells. PCR primers were designed flanking an indel polymorphism between the X and Y chromosome homologs Jarid1c and Jarid1d. Expected sizes: X chromosome polymorphism: 170bp. Y chromosome polymorphism: 150bp. R129-E1 (E1): X, R129-E2 (E2): X/Y, R129-E3 (E3): X/Y, R129E4 (E4): X/Y, R129-F1 (F1): X/Y, R129-F2 (F2): X/Y, R129-F4 (F4): X/Y. Note: This PCR reaction is unable to distinguish between one and two X chromosomes. 41  Table 3.1 Sex chromosome complement of newly derived R129 and 129R ES cell lines. XY genotyping was performed using an indel in the X and Y chromosome homologs, Jarid1c and Jarid1d. X chromosome genotyping was performed using a microsatellite expansion polymorphism, DXMit210, to differentiate between the 129S1 (129) and C57BL/6J (B6) X chromosomes. ES cell line  XY genotyping  129R-B5 R129-B1 R129-D3 R129-E1 R129-E2 R129-E3 R129-E4 R129-F1 R129-F2 R129-F4  X X X X X/Y X/Y X/Y X/Y X/Y X/Y  X chromosome 129/B6 genotyping 129/B6 129/B6 129/B6 129/B6 B6 B6 B6 B6 B6 B6  Sex chromosome complement XX XX XX XX XY XY XY XY XY XY  42  Table 3.2. Selection for spontaneous LOH at ROSA26 locus through high G418 selection of R129-E2 ES cells. Four separate attempts at high G418 selection were made with increasing concentrations of G418. ES cells were plated at listed amounts onto large (10mm) gelatinized tissue-culture-grade plates. Concentration of G418 for “typical” selection is 0.15mg/mL (compared with 0.75-7mg/mL used in this experiment for “high” selection). # Cells Plated and # High G418 resistant cells obtained are number averaged/per large plate. # screened is number of total colonies screened for ROSA26 LOH.  43  A  B  C  Figure 3.5. Detection of LOH at the ROSA26 locus by Southern blot analysis of ES cell genomic DNA. A. Restriction map of the ROSA26 locus, targeting vectors, and location of probe used for Southern blot (Soriano, 1999). Small arrows: location of ROSA26 genotyping primers. B. EcoRV Southern blot analysis of ES cells selected with high G418 for spontaneous mitotic LOH at the ROSA26 locus using dose-sensitive Neo-mediated resistance. Wild-type ROSA26 locus: 11kb. Mutant R26R insert at ROSA26: 3.8kb. R3/R3: Homozygous for R26R insert. R3/+: R26R/wild-type. C. LOH in seven ES cell sub-lines shown by PCR for wild-type and mutant (R26R) ROSA26 locus. Expected sizes: Wild-type ROSA26: 250bp. Mutant R26R: 300bp.  44  Figure 3.6. Characterization of MMU6 LOH in ES cells. Seven high G418-selected ES cell sub-lines were analyzed for LOH using PCR for D6Mit marker polymorphisms between 129S1 (129) and C57BL/6J (B6). Sub-lines tested include 1.4.6, 4.4.4, 4.5.2, 4.5.6, 4.6.3, 4.10.7, 4.11.6. MMU6 chromosome is to scale. 45  Figure 3.7. SNP marker analysis on MMU6. Illumina SNP microarray genotyping was used to type 5 different ES cell sub-lines: 4 R129*-E2 sub-lines (4.5.6, 4.6.3, 4.10.7, 4.116) and the parental R3/+ (R129-E2). A: 129 mouse strain genotype. B: C57Bl/6J mouse strain genotype. Pink cells: Homozygous C57Bl/6J genotype. Yellow: Heterozygous C57Bl/6J / 129S1 genotype. Grey: Homozygous 129S1 genotype. Green: C57Bl/6J / C57Bl/6J / 129S1 genotype (trisomic). Red: maternally expressed genes. Blue: paternally expressed genes. Note: Nap1l5 may not actually be located within imprinted domain 2, and instead may represent a third imprinted domain on MMU6 (Beechey, 2004).  46  47  Figure 3.8. Summary of regions of mUPD for MMU6 in 3 high G418-selected ES cells LOH for the ROSA26 locus. LOH: Loss of heterozygosity, mUPD. Trisomic: B6/B6/129. ?: No informative markers tested. UI: Uninformative (region of 129 homozygosity in parental R3/+ ES cell line). MD6-2/dp1: mUPD6-ID2/dup-ID1 (aka 4.5.6). MD6:mUPD6 (aka 4.6.3). MD6-2/p1: mUPD6-ID2/partial mUPD6-ID1 (aka 4.10.7). R3/+: R129-E2 parental line.  48  49  CHAPTER 4 Maintenance of imprinted expression and methylation in LOH ES cell sub-lines 4.1 Introduction The primary mark of imprinting is differential CpG methylation depending on the parent-of-origin, or DMRs, which are consistently maintained in all somatic cells throughout development, regardless of whether there is expression in any given cell type (de la Casa-Esperon, et al, 2003; Pardo-Manuel de Villena, et al, 2000). This imprinted methylation is particularly important at ICRs, which are hypothesized to be Peg10 and Peg1 on MMU6 (Suzuki, et al, 2005; Suzuki, et al, 2007). These ICRs are instrumental in cis-regulating proper imprinted expression within their respective imprinted domains. UPD is thought to disrupt normal imprinted expression, as these regulatory DMRs are differentially marked on the parental alleles (Horii, et al, 2008; Kono, et al, 2004). As a result, the expression of a MEG is increased roughly two-fold in mUPD, and that of a PEG is significantly decreased or absent, while the opposite is true of pUPD. By studying the abnormal regulation and expression of imprinted genes in UPD we hope to learn about normal imprinting regulation and maintenance. MMU6 is an ideal chromosome for studying imprinting due to the presence of two proximal imprinting domains, both of which have candidate ICRs (Figure 4.1). Further study of MMU6 is warranted, since new imprinted genes on MMU6 are still being discovered, and MMU6 is also a candidate for the orthologous region causative for Silver-Russell syndrome (Cuisset, et al, 1997; Eggermann, et al, 2008; Kobayashi, et al, 1997).  50  4.1.1 MMU6 imprinted domain 1 Imprinted domain 1 (ID1), the proximal-most domain, extends from 3.6Mb – 6.8Mb and contains 7 MEGs and 2 PEGs (Figure 4.1, Table 4.1) (Monk, et al, 2008; Ono, et al, 2003). mUPD for ID1, or for the entire MMU6, results in embryonic lethality before E11.5 (Beechey, 2000). The promoter region of the paternally expressed gene 10 (Peg10) is the candidate ICR for the domain (Monk, et al, 2008; Suzuki, et al, 2007). This retrotransposon-derived gene inserted into its current locus sometime after the most common recent ancestor between marsupials and placental mammals diverged from the prototherian mammal (egg-laying mammal) lineage (Suzuki, et al, 2007). Its CpG island overlapping the promoter region is shared with sarcoglycan, epsilon (Sgce) in mouse and human, and is a DMR which is established in the oocyte (Ono, et al, 2003). In marsupials, while Sgce and Peg10 share a CpG island, a DMR is present only within the Peg10 promoter region, consistent with the observation that Peg10 is imprinted in marsupials, while Sgce is biallelically expressed (Suzuki, et al, 2007). In fact, Peg10 is the only gene from the domain seen to be imprinted in mouse and human that is imprinted in marsupials, which suggests that Peg10 control of the present imprinted domain 1 in mouse may have instigated regulatory control over some nearby genes, causing these other genes to have imprinted expression. An additional line of evidence for Peg10 as an ICR for ID1 comes from Dnmt3l–/+ mice, where loss of methylation at Peg10 resulted in loss of imprinted expression of ID1. PEGs Peg10 and Sgce were biallelically expressed, while the MEGs in ID1 were repressed (Monk, et al, 2008). Peg10 has two protein isoforms, one shorter isoform that resembles the retrovirus pol protein with a DNA-binding domain, and a longer isoform produced from a frameshift read through, that codes for a retrovirus gag-like protein, with a protease domain  51  (Clark, et al, 2007; Ono, et al, 2001). While the specific functions of these isoforms is unknown, they are known to be involved in placental development, as the Peg10 knockout mouse has severe placental defects and is unable to survive past E10.5 (Ono, et al, 2006). Peg10 is a strictly imprinted gene, imprinted in all tissues examined, including ES cells and TS cells, with no leaky biallelic expression (Monk, et al, 2008). It shows strong expression in embryonic and extra-embryonic lineages, as well as neonatal brain (Ono, et al, 2001; Ono, et al, 2003; Ono, et al, 2006). Nearby Sgce functions as a component of the dystrophin-sarcoglycan complex (Ono, et al, 2003; Piras, et al, 2000) . Relaxed imprinting, with a strong paternal bias, of Sgce is seen in adult brain, while all other adult mouse tissues, as well as ES cells and TS cells, display strict imprinting (Monk, et al, 2008; Piras, et al, 2000). Three paroxonase genes, Pon1, Pon2, and Pon3 are located within ID1, although only Pon2 and Pon3 show imprinted expression (Ono, et al, 2003). All three gene products have antioxidant activities, while only Pon1 and Pon3 are known to be associated with high-density lipoprotein (Ng, et al, 2006). Knockout mice lacking Pon2 develop larger artherosclerotic lesions when fed high-fat, high-cholesterol diets than wildtype mice, most likely because of increased oxidative stress and heightened inflammatory response (Ng, et al, 2006). Pon2 and Pon3 have been shown to be imprinted, having a strong maternal bias of expression in extra-embryonic lineages in E10 and E13, while they are biallelically expressed in all neonatal tissues studied (Ono, et al, 2003). Pon2 has also been recently shown to be biallelically expressed in ES cells and TS cells (Monk, et al, 2008). Calcr (calcitonin receptor), the proximal-most imprinted gene in ID1 has tissuespecific imprinting in E15.5 and adult brain, showing a maternal bias in expression, while  52  it is biallelic in all other tissues examined (Hoshiya, et al, 2003; Ono, et al, 2003). Known to function as a G protein-coupled calcitonin receptor, Calcr suppresses appetite and gastric acid secretion, as well as inhibiting bone re-absorption (Dacquin, et al, 2004; Hoshiya, et al, 2003). Ankyrin repeat and suppressor of cytokine signaling (Asb4) imprinting is thought to be established before implantation (≤E3.5), because of it’s maternal-biased expression in ES cells and trophoblast stem cells (TS cells) (Monk, et al, 2008). Asb4 also shows a maternal bias in E9.5 and E15.5 embryo and placenta. Strongly expressed in testes, Asb4 functions to suppress cytokine signaling (Kile, et al, 2000). Coding for a DNA-binding homeobox protein, distal-less homeobox 5 (Dlx5) mouse knockouts cause multiple craniofacial structural defects (Kimura, et al, 2004). Dlx5 is strongly expressed in brain and testis and shows maternal bias in the former, but biallelic expression in the latter (Kimura, et al, 2004). Protein phosphatase 1, regulatory (inhibitor) subunit 9A (Ppp1r9a, also known as neurabin) codes for an actin filament-binding protein that regulates synapse formation (Ono, et al, 2003). Ppp1r9a has a strong maternal bias in expression in E10 and E13 extraembryonic tissues, while it is biallelic in neonatal and adult brain, as well as ES cells and TS cells (Monk, et al, 2008; Ono, et al, 2003). The expression data suggests that imprinted expression is established sometime after E3.5, but before, or at E10. For the newly discovered MEG Tfpi2 (tissue factor pathway inhibitor 2) imprinted expression is established before implantation (≤E3.5), as hypothesized by maternalspecific expression in ES cells and TS cells (Monk, et al, 2008). Imprinting is also seen in E12.5 and E14.5 extra-embryonic tissues, while biallelic expression in embryo is seen at these stages.  53  4.1.2 MMU6 imprinted domain 2 The sub-proximal imprinted domain 2 (ID2) contains 4 PEGs and 2 MEGs, and is separated from ID1 by 23.9Mb (Figure 4.1, Table 4.2). mUPD for ID2 causes growth retardation in mice (Beechey, 2000), while pUPD causes fetal overgrowth (Beechey, 2004). There are two DMRs located within ID1: one within the promoter region at paternally expressed gene 1 (Peg1, also known as Mest, mesoderm specific transcript) (Lefebvre, et al, 1997), and one within the promoter region of nucleosome assembly protein 1, like 5 (Nap1l5) (Smith, et al, 2003). Peg1 is the proposed ICR of ID2 (Suzuki, et al, 2005), while Nap1l5 may actually represent a separate imprinted domain (Beechey, 2004). Thus, ID2 extends either 0.23Mb (excluding Nap1l5), or 28.2Mb (including Nap1l5). Peg1 has conserved imprinted expression between eutherian mammals and marsupials, lending support to its status as a proposed ICR (Suzuki, et al, 2005). In marsupials there is a lack of the maternally methylated DMR at the Peg1 promoter, that exists in human and mouse. Marsupials, however, may have alternate methods to control imprinting in addition to differential CpG methylation. Upon activation of Peg1 expression at E9.5, Peg1 is strictly paternally expressed in all tissues in which it is expressed, which are mesodermal in origin (Kaneko-Ishino, et al, 1995; Lefebvre, et al, 1997). The imprinting status of Coatomer protein complex subunit gamma 2 (Copg2) is currently disputed, as a maternal bias in expression has been seen in E11.5-E17.5 embryo and adult brain in reciprocal crosses of C57BL/6J x M. m. molossinus (Lee, et al, 2000), whereas biallelic expression has been seen in neonate tissues in C57BL/6J x M. spretus (no reciprocal cross) (Yun, et al, 2003). A similar situation has been seen with human  54  COPG2, where separate groups have disputed the imprinting status of this locus (Blagitko, et al, 1999; Yamasaki, et al, 2000). This may reflect polymorphic imprinting depending on genetic background, or could be a difference in the opinion of the definition of imprinting (whether to include genes as imprinted with a parental bias, but leaky biallelic expression). The CpG island overlapping the promoter of Copg2 has been reported to be a DMR (Yun, et al, 2003), but careful examination of the methylation data shows not only methylation of the paternal allele, but also partial methylation of the maternal allele. Antisense to Copg2 (Copg2as1) is a non-coding, paternally expressed transcript, which overlaps with the 3’ UTRs of Peg1 and Copg2, and has an unknown function (Lee, et al, 2000). It has been shown to have imprinted expression in whole embryo from E11.5 – E17.5, whole neonate, and adult brain. Another antisense transcript in the region, Mit1 (mest linked imprinted transcript 1) is also non-coding and paternally expressed, and is located within intron 20 of Copg2. Mit1 is imprinted in E13.5 – E17.5 embryo, neonates, and adult brain (Lee, et al, 2000). The newly discovered Klf14 (kruppel-like factor 14) is a MEG derived from a retrotransposon insertion event, and has been shown to be imprinted in all embryonic and extra-embryonic tissues examined in E9.5, E15.5, neonate, and adult (Parker-Katiraee, et al, 2007). PEG Nap1l5 shows tissue-specific imprinting starting at E10 in brain and adrenal glands (Beechey, 2004; Davies, et al, 2004; Smith, et al, 2003).  55  4.1.3 Imprinted methylation and gene expression during parthenogenesis and androgenesis Imprinting in parthenogenetic and androgenetic development Embryo expression of imprinted genes for the most part follows the trend of increased MEG and absent PEG expression in parthenogenetic (PG) embryos, and absent MEG and increased PEG expression in androgenetic (AG) embryos (Table 4.3) (KanekoIshino, et al, 1995; Obata, et al, 1998; Ogawa, et al, 2006; Sotomaru, et al, 2001; Sotomaru, et al, 2002). Imprinting in general seems to be tightly controlled in embryos with very low leaky biallelic expression in a minority of genes (U2af1, Dlk1, Ata3, Impact, Asb4, H19, Tssc3, Dcn, Grb10) (Ogawa, et al, 2006), which is consistent with the postulated functions of imprinted genes in embryo and placental development (Tycko, et al, 2002). Imprinting in parthenogenetic and androgenetic ES cells Imprinted gene expression in parthenogenetic and androgenetic ES cells (PGES cells and AGES cells, respectively) tends to have a lot more variation, and to deviate more from expected levels of expression compared to PG and AG embryo (Jiang, et al, 2007; Szabo, et al, 1994) (Table 4.3). Seven/nine imprinted genes surveyed followed expected trends, but failed to reach expected levels in PGES cells and AGES cells: absent PEG expression, 2-fold MEG expression in PGES cells and 2-fold PEG expression, absent MEG expression in AGES cells (Table 4.4). Embryoid bodies (EB), which are derived from differentiated ES cells, representing a large number of tissues, had a more unpredictable gene expression pattern than ES cells. There was a large amount of gene expression variation in EB, showing differences between different cell lines. When  56  present, expression from normally inactive parental alleles showed expression that was unlikely to be simple leaky expression, as expression was comparable to normal ES cells (Szabo, et al, 1994). EB tended to show a large difference in expression between high passage and low passage number cells, with increased loss of imprinting and expression closer to normal with higher passage number. ES cells showed smaller changes between high and low passage, with some imprinted genes showing no difference in expression with higher passage. Experimental approach ES cells are an ideal model for imprinting as they are readily available at low passage number by derivation from inner cell mass from E3.5 blastocysts, they are quick to divide and easy to culture with the proper equipment, and unlike mouse colonies, don’t require a lot of space to maintain. An XY-ES cell line, R129-E2, newly derived in our lab from an F1 129S1 x C57BL/6J cross, was selected to undergo high G418 selection. The R26R insert on the C57BL/6J MMU6 chromosome makes it possible to select for spontaneous mitotic LOH, as ES cells with only one copy of the neo on the R26R insert will die under high G418, while ES cells with two copies of neo are able to survive. Seven ES cells sub-lines were recovered with LOH for the site of R26R insertion, the ROSA26 locus. After detailed genotyping using polymorphisms including dinucleotide repeat expansion (Mit markers) and SNPs through a linkage panel from Illumina Inc., 3/7 sublines, MD6, MD6-2/p1, and MD6-2/dp1, were found to be LOH, and thus UPD for one or both imprinted domains. These three cell sub-lines all have different regions of UPD for MMU6: all have UPD for ID2, MD6-2/p1 has UPD for part of ID1, and MD6 has UPD for the entire ID1,  57  and MD6-2/dp1 is biallelic for ID1. These sub-lines were chosen to carry out expression analysis of imprinted genes for both imprinted domains. This research is unique in that consequences of mitotically-derived UPD on imprinting have never been studied before. This work has the added benefit of being able to study UPD for MMU6 in isolation and compare the results to work previously done with meiotically-derived UPD for the entire genome, as with PG and AG embryos. To analyze maintenance of imprinting, the expression of 2 MEGs, 2 PEGs, and one non-imprinted control from each imprinted domain was analyzed for MD6, MD62/p1, and MD6-2dp1 using semi-quantitative RT-PCR. All seven LOH sub-lines were used to evaluate the DNA methylation status of Peg1, the postulated ICR for ID2, to investigate whether or not the DNA methylation status of this gene reflected expression levels of imprinted genes in ID2 in any given sub-line.  4.2 Results RT-PCR was tested using a range of cycle numbers: 20, 25, 30, 35, 40. Expression for many genes was undetectable at 20 cycles, while at 35 and 40 cycles the PCR amplification was saturated, and many expression differences between ES cell lines present at lower cycle numbers were not detectible. Therefore, the following data represents RT-PCR using 25 and 30 cycles.  4.2.1 Expression of ID1 imprinted genes in mUPD6 cells RT-PCR results for ID1 are presented in Figure 4.2 and Figure 4.3. ES cells used included the parental line, R3/+ (R129-E2), MD6, MD6-2/p1, and MD6-2dp1. R3/+ is normal, biparental for ID1, MD6 is mUPD for ID1, MD6-2/p1 is UPD for a part of ID1  58  including at least Dlx5 that may extend as far as Sgce, and MD6-2dp1 is B6/B6/129 trisomic for all of ID1 including Sgce and possibly extending to Tfpi2 (Figure 3.8). RTPCR reactions were repeated 2-6 times, using 2-4 separate RNA extractions and 2-4 separate reverse-transcriptase reactions, with varying PCR cycle numbers (25 or 30). Using ImageJ the intensity of Sybr green-stained reverse transcriptase PCR products was measured for specific imprinted genes and for non-imprinted controls Col1a2 and Gapdh. After subtracting the background intensity a ratio was taken of specific gene expression level over control Gapdh expression level. The ratio for each UPD line was then compared to the ratio for the parental line, in order to directly compare gene expression levels to normal biallelic patterns. Sub-line MD6 consistently showed decreased expression of PEGs Peg10 and Sgce in EB and ES cells (with 5 replicates of varying PCR cycle number), less than half the amount of parental line R129-E2 Peg10 and Sgce expression levels. MEG Ppp1r9a was increased in ES cells 2-fold and decreased in EB by half. MEG Pon2 expression was shown to be decreased by half in ES cells and EB at 25 cycles RT-PCR, but was close to levels of R129-E2 in ES cells and EB at 30 cycles RT-PCR. The non-imprinted control, Col1a2, had levels near R129-E2 in EB and ES cells, with more variation in EB. In MD6-2/p1 ES cells Sgce showed normal or reduced levels compared to R129E2, with ES cells showing consistently lowered expression. Peg10 was close to R129-E2 levels, between 0.9 – 1.5 in EB and ES cells except for EB at 25 cycles RT-PCR which was reduced by half. Ppp1r9a was decreased in EB to less than half, but increased in ES cells over three-fold. Comparable levels to R129-E2 were seen of Pon2. Control Col1a2 was close to normal parental levels in ES cells, but reduced by half in EB.  59  MD6-2/dp1 had near wild-type levels of Sgce in ES cells and EB using 25 cycles RT-PCR, but with 30 cycles showed a decrease by half in EB (0.42) and a 1.5-fold increase in ES cells. Similar levels of Peg10 were seen in ES cells and EB to parental except for a 1.5-fold increase in EB at 30 cycles. Pon2 was also similar to parental in ES cells and EB, with a 1.5-fold in ES cells at 30 cycles being an exception. Ppp1r9a followed the same pattern as in the other two sub-lines, being decreased by half in EB and increased 2-fold in ES cells. Col1a2 was mostly near R129-E2 levels in ES cells and EB, except for a decrease to less than half in EB at 25 cycles.  4.2.2 Expression of ID2 imprinted genes in mUPD6 cells Expression analysis was done as for ID1 (section 4.2.1) and is presented in Figure 4.4 and Figure 4.5. ID2 is biparental only in the ES cell line R3/+, whereas it is mUPD in MD6, MD6-2/p1, and MD6-2dp1. PEG Peg1 in MD6 ES cells and EB was at levels comparable to R129-E2 or above, ranging from a slight decrease to a nearly 1.5-fold increase in ES cells and EB. Wild-type levels or lower were seen with the MEG Copg2, with ES cells ranging from half to the full amount of parental expression and EB much closer to parental expression. PEG Mit1 was shown to be near or above wild-type levels in ES cells and EB, except for ES cells at 30 cycles, which was slightly decreased. Non-imprinted control Tsga14 was near or above R129-E2, ranging from normal levels to a 1.5-fold increase in EB and ES cells. MD6-2/p1 showed Peg1 expression at or half the amount of normal parental levels in ES cells and EB. Copg2 was also reduced or at wild-type levels ranging from less than half up to normal levels in ES cells and EB. A large decrease in Mit1 was seen in both ES  60  cells and EB. Control Tsga14 was decreased in ES cells, while in EB had a large range of expression from slightly below normal up to nearly a 1.5-fold increase. MD6-2/dp1 showed a decrease of half of Peg1 in ES cells and EB. Copg2 was close to normal levels in both ES cells and EB. Expression of Mit1 in EB was decreased slightly, while it was increased slightly in ES cells. Tsga14 expression varied largely in ES cells and EB from one-half up to two-fold normal levels. Klf14 was expressed at low levels (compared to control Gapdh) or was absent in all sub-lines. R129-E2 did not express Klf14 at detectable levels in ES cells or EB, except for EB at 30 cycles RT-PCR, where it was expressed at one-fifth the level of Gapdh. MD6 expression of Klf14 was detected only at 30 cycles RT-PCR, where it was approximately one-fifth the level of Gapdh in ES cells and EB. MD6-2/p1 Klf14 was higher than wildtype in EB, where it was expressed up to one-fourth the level of Gapdh. MD6-2/dp1 did not express Klf14 in either ES cells or EB.  4.2.3 Peg1 methylation status in LOH ES cell sub-lines DNA methylation at the Peg1 locus was examined using Southern blot analysis with the methylation-sensitive enzyme, BssHII, with a recognition site located within the Peg1 promoter DMR (Figure 4.6). The CpG within this recognition site is normally maternally methylated and paternally unmethylated (Lefebvre, et al, 1997). When used in combination with XbaI, this results in a 2.5kb band when the CpG site is unmethylated and BssHII can cut, and a 3.0kb band when the CpG site is methylated and BssHII is unable to cut. Three out of four gene conversion sub-lines, GC6.1, GC6.3 and GC6.4 had both a methylated (3.0kb) and an unmethylated allele (2.5kb), while gene conversion subline GC6.2 was completely unmethylated. MD6-2/dp1 and MD6-2/p1 were completely  61  methylated with one 3.0kb band, while MD6 was completely unmethylated with one 2.5kb band.  4.3 Discussion 4.3.1 Reactivation of PEGs from MMU6 ID1 In cells with mUPD, PEGs are expected to have no or severely reduced expression, while MEGs are expected to be over-expressed roughly two-fold. The R129*-E2 sub-line MD6, the only sub-line fully mUPD for MMU6 was the only sub-line confirmed to be UPD for the entire ID1, and was also the only sub-line to consistently show lowered expression of ID1 PEGs, Peg10 and Sgce, in both ES cells and EB with 5-6 replications. This leaky expression of Peg10 and Sgce from MD6 suggests that full chromosomal mUPD caused by mitotic non-disjunction mildly disrupts maintenance at ID1, but that the maternal imprinting expression pattern remains largely intact. Even though sub-line MD6-2/dp1 has a local duplication that encompasses the Peg10 and Sgce region, these PEGs were expressed at levels comparable to the parental line R129-E2, and only rarely increased at roughly 1.5-fold. This is to be expected because while this sub-line contains three copies of the PEGs in ID1, two copies are maternal alleles, which are expected to be silent, while only one copy is a paternal allele, which is expected to be active. This sub-line is similar to a wild-type cell line in that it has only one active allele of Peg10 and Sgce. Since expression was only rarely increased it can be inferred that there was very little leaky expression of PEGs from the maternal allele, and that proper imprinting was maintained. It is unknown whether MD6-2/dp1 is UPD for the Peg10 and Sgce region, and expression studies did not offer solid evidence one way or the other. While Sgce was  62  moderately reduced in the majority of reactions, Peg10 was near wild-type levels in the majority of reactions. This variability in PCR results was seen throughout 5-6 replications, with 2 separate RNA extractions and 3 separate reverse-transcriptase PCR reactions. This suggests that low expression levels (relative to G3PDH) may be difficult to quantify using my technique, and/or that there is some saturation of signal with increased PCR cycle number. Both of these issues could be solved by analyzing these genes through quantitative real-time PCR. Pon2 and Ppp1r9a did not display the expected increase of up to two-fold in any of the sub-lines for the great majority of reactions, suggesting that these genes do not show imprinted expression in ES cells. This confirms the recent report by Monk et al. (Monk, et al, 2008), who showed biallelic expression of Pon2 and Ppp1r9a in both ES cells and TS cells with an M.m.castaneus x C57BL/6J background. Col1a2, a non-imprinted gene within ID1 showed a large variability in expression patterns, which could be due to limitations in the experimental technique as discussed above. Alternatively, Col12 could be affected by the change in expression and/or epigenetic profile of its neighboring imprinted genes.  4.3.2 ID2 imprinting maintained in partial mUPD but lost in full mUPD Peg1, whose promoter is the purported ICR of ID2, was expressed at normal levels or lower in two/three ES cells sub-lines, MD6-2/dp1 and MD6-2/p1. It is unclear why Peg1 was not completely absent from these lines when they are known to be UPD for all of ID2. This is similar to the results obtained using PGES cells, but different from PG embryos, where Peg1 expression is absent, or at very low levels. That ES cells, with either mitotically-derived (MD6-2/dp1, MD6-2/p1, MD6), or meiotically-derived (PGES cells)  63  show a common expression pattern suggests that the mechanism of imprint maintenance in these two distinct types of cells may be similar. ES cells, in general, may have more flexible epigenetic gene regulation than embryos, which allows ES cells to more readily adapt to perturbations in gene dosage. MD6 showed less reduction in Peg1 expression compared to MD6-2/dp1 and MD6-2/p1, and was sometimes higher than the normal parental line. ID2 imprinting may be more unstable, because this sub-line has full UPD for MMU6, which may result in chromosomal-wide instability of gene expression. It is unclear why ID2 would be more affected by this gene expression instability than ID1. This could represent innate differences in regulation of imprinted expression from the two imprinted domains. The other PEG, Mit1 had a more unpredictable expression pattern. While usually slightly or moderately reduced in MD6-2/dp1 and MD6-2/p1, Mit1 was sometimes at or slightly above wild-type levels in these two sub-lines. MD6 showed nearly consistent increase in Mit1 expression above wild-type, reminiscent of its Peg1 expression pattern. MEG Copg2 also showed unexpected expression patterns, being at or below wildtype levels in all three sub-lines. While not shown to be biallelic in ES cells, Copg2 has been shown to have polymorphic imprinting, showing biased maternal expression on some mouse backgrounds and biallelic on others (Lee, et al, 2000; Yun, et al, 2003). MEG Klf14 was only expressed at low levels in 2/3 sub-lines, MD6-2/p1 and MD6, although expression was slightly higher than wild-type. Klf14 has been shown to be imprinted in a number of tissues, but nothing is known earlier than E9.5 so far (Parker-Katiraee, et al, 2007), so it is possible that its imprint is not established by E3.5, in which case it would not imprinted in ES cells.  64  4.3.3 Peg1 hypomethylation with full mUPD Three mUPD sub-lines, MD6-2/dp1, MD6, and MD6-2/p1, have mUPD for ID2. While MD6-2/dp1 and MD6-2/p1 were shown to be fully methylated by methylationsensitive Southern blot, MD6 was shown to be fully unmethylated (Figure 4.6). It seems that with mUPD of ID2 alone methylation imprints are able to be maintained, but full chromosomal mUPD somehow causes a disruption in imprint maintenance. This may be due in part to the different mitotic mechanisms that result in partial vs. full UPD: mitotic recombination and chromosome duplication. The first is a byproduct of a normal repair mechanism that exists in the cell to repair mutations on non-sister chromatids (Vrieling, 2001), whereas the other is a result of mitotic aneuploidy, causing a trisomic intermediate that is then resolved by chromosome loss (Lefebvre, et al, 2001). A trisomic intermediate stage may in fact induce loss of imprinting at specific loci, causing a change in expression levels closer to that of wild-type (Vacik, et al, 2003).  65  66 66  1. (Hoshiya, et al, 2003). 2. (Dacquin, et al, 2004). 3. (Monk, et al, 2008). 4. (Ono, et al, 2003). 5. (Yokoi, et al, 2005). 6. (Clark, et al, 2007). 7. (Ono, et al, 2006). 8. (Shih, et al, 2007). 9. (Ng, et al, 2006). 10. (Kile, et al, 2000). 11. (Kimura, et al, 2004).  Table 4.1. Imprinted genes on MMU6, imprinted domain 1. Imprinted genes characterized by time of imprint establishment, tissue-specific imprinting, presence of CpG island and DMR. Red: MEGs. Blue: PEGs.  67  1. (Lefebvre, et al, 1997). 2. (Kaneko-Ishino, et al, 1995). 3. (Lefebvre, et al, 1998). 4. (Yun, et al, 2003). 5. (Blagitko, et al, 1999). 6. (Lee, et al, 2000). 7. (ParkerKatiraee, et al, 2007). 8. (Smith, et al, 2003). 9. (Davies, et al, 2004). 10. (Beechey, 2004).  67  Table 4.2. Imprinted genes on MMU6, imprinted domain 2. Imprinted genes characterized by time of imprint establishment, tissue-specific imprinting, presence of CpG island and DMR. Red: MEGs. Blue: PEGs.  Table 4.3. Imprinted gene expression in parthenogenetic and androgenetic embryos. Ratio of expression seen compared to normal fertilized embryos over expected expression. Expected expression in PG embryo is 2-fold MEG, absent PEG, AG embryo is absent MEG, 2-fold PEG. Expression not quantified is listed as absent or present. PG: parthenogenetic. AG: Androgenetic. PEG: paternally expressed gene. MEG: Maternally expressed gene.  1. (Ogawa, et al, 2006). 2. (Obata, et al, 1998). 3. (Sotomaru, et al, 2001). 4. (Kaneko-Ishino, et al, 1995).  1. (Ogawa, et al, 2006). 2. (Obata, et al, 1998). 3. (Sotomaru, et al, 2001). 4. (Kaneko-Ishino, et al, 1995).  68  69  *Maternal methylation lost after blastocyst stage. **Monoallelic methylation – unable to determine parental origin. 1. (Jiang, et al, 2007). 2.(Szabo, et al, 1994). 3.(Obata, et al, 1998). 4.(Hershko, et al, 1999). 5.(Hanel, et al, 2001). 6.(Song, et al, 2008). 7.(Murphy, et al, 2001). 8.(Sunahara, et al, 2000). 9.(Okamura, et al, 2000).  Table 4.4. Expected and observed imprinted gene expression in murine PGES cells and AGES cells.  69  Figure 4.1. MMU6 proximal imprinted gene domains 1 and 2. Locations of imprinted domains on MMU6 and imprinted genes found within them. ROSA26 locus, site of R26R insert, is shown on distal MMU6. Chromosome is to scale. Red: MEGs. Blue: PEGs. Black: non-imprinted control genes. 70  Figure 4.2. RT-PCR analysis of ID1 genes (25 cycles). Gapdh control reactions are shown below the reactions for each gene. A) 25 cycle semi-quantitative reverse-transcriptase PCR (RT-PCR) of select ID1 genes in differentiated embryoid bodies (day 14) and undifferentiated ES cells in R129-E2 parental cell line and 3 mUPD6 ES cells sub-lines. Numbers below lanes are gene expression levels normalized to Gapdh control. B) Reverse transcriptase + and reverse transcriptase – PCR reactions of genes and cell lines in A. C) Comparison of gene expression levels measured by RT-PCR in 3 mUPD6 cell sub-lines. Values represent ratio of gene expression level normalized to Gapdh divided by ratio of parental (P) cell line gene expression normalized to Gapdh. P: Parental. MD6-2/dp1: mUPD6-ID2/dup-ID1. MD6:mUPD6. MD6-2/p1: mUPD6ID2/partial mUPD6-ID1. Black: non-imprinted control. Blue: PEG. Red: MEG.  71  Figure 4.3. RT-PCR analysis of ID1 genes (30 cycles). Gapdh control reactions are shown below the reactions for each gene. A) 30 cycle semi-quantitative reverse-transcriptase PCR (RT-PCR) of select ID1 genes in differentiated embryoid bodies (day 14) and undifferentiated ES cells in R129-E2 parental cell line and 3 mUPD6 ES cells sub-lines. Numbers below lanes are gene expression levels normalized to Gapdh control. B) Reverse transcriptase + and reverse transcriptase – PCR reactions of genes and cell lines in A. C) Comparison of gene expression levels measured by RT-PCR in 3 mUPD6 cell sub-lines. Values represent ratio of gene expression level normalized to Gapdh divided by ratio of parental (P) cell line gene expression normalized to Gapdh. P: Parental. MD6-2/dp1: mUPD6-ID2/dup-ID1. MD6:mUPD6. MD6-2/p1: mUPD6ID2/partial mUPD6-ID1. Black: non-imprinted control. Blue: PEG. Red: MEG. 72  Figure 4.4. RT-PCR analysis of ID2 genes (25 cycles). Gapdh control reactions are shown below the reactions for each gene. A) 25 cycle semi-quantitative reverse-transcriptase PCR (RT-PCR) of select ID2 genes in differentiated embryoid bodies (day 14) and undifferentiated ES cells in R129-E2 parental cell line and 3 mUPD6 ES cells sub-lines. Numbers below lanes are gene expression levels normalized to Gapdh control. B) Reverse transcriptase + and reverse transcriptase – PCR reactions of genes and cell lines in A. C) Comparison of gene expression levels measured by RT-PCR in 3 mUPD6 cell sub-lines. Values represent ratio of gene expression level normalized to Gapdh divided by ratio of parental (P) cell line gene expression normalized to Gapdh. P: Parental. MD6-2/dp1: mUPD6-ID2/dup-ID1. MD6:mUPD6. MD6-2/p1: mUPD6ID2/partial mUPD6-ID1. Black: non-imprinted control. Blue: PEG. Red: MEG. 73  Figure 4.5. RT-PCR analysis of ID2 genes (30 cycles). Gapdh control reactions are shown below the reactions for each gene. A) 30 cycle semi-quantitative reverse-transcriptase PCR (RT-PCR) of select ID2 genes in differentiated embryoid bodies (day 14) and undifferentiated ES cells in R129-E2 parental cell line and 3 mUPD6 ES cells sub-lines. Numbers below lanes are gene expression levels normalized to Gapdh control. B) Reverse transcriptase + and reverse transcriptase – PCR reactions of genes and cell lines in A. C) Comparison of gene expression levels measured by RT-PCR in 3 mUPD6 cell sub-lines. Values represent ratio of gene expression level normalized to Gapdh divided by ratio of parental (P) cell line gene expression normalized to Gapdh. P: Parental. MD6-2/dp1: mUPD6-ID2/dup-ID1. MD6:mUPD6. MD6-2/p1: mUPD6ID2/partial mUPD6-ID1. Black: non-imprinted control. Blue: PEG. Red: MEG. 74  A  B  Figure 4.6. Peg1 methylation-sensitive Southern blot of wild-type embryo and R129-E2 parental and R129*-E2 sub-lines LOH at ROSA26 locus. A. Restriction map of Peg1 locus and Southern blot probe location. Samples were digested with XbaI (insensitive to methylated CpG), and BssHII (sensitive to methylation CpG). Maternal, methylated allele: 3.0kb. Paternal, unmethylated allele: 2.5kb. Xb: XbaI. B: BssHII. R3/+: Parental, R129E2. E: E14.5 Embryo. GC6.1: GC-MMU6-ROSA26-1. GC6.2: GC-MMU6-ROSA26-2. GC6.3: GC-MMU6-ROSA26-3. MD6-2/dp1: mUPD6-ID2/dup-ID1. MD6: mUPD6. MD6-2/p1: mUPD6-ID2/partial mUPD6-ID1. GC6.4: GC-MMU6-ROSA26-4.  75  CHAPTER 5 Discussion While a considerable body of work has been done on maintenance of imprinting in embryos and ES cells derived from meiotic UPD (Allen, et al, 1994; Jiang, et al, 2007; Ogawa, et al, 2006; Szabo, et al, 1994), UPD derived from mitotic events, the subject of this thesis, has not been studied before now. The study of single chromosome UPD, as in this thesis, is crucial for dissecting and understanding imprinting regulation and maintenance, and for understanding imprinting in human disease, caused by single chromosomal UPD or mis-regulation of single imprinted genes. UPD in mouse represents an ideal model for human disease in that mouse is closely related, having a most recent common ancestor 75 million years ago (Guenet, 2005). Ninety-nine percent of genes are homologous to human genes, and about 90% of the genome has conserved synteny with human, making mouse genes likely to share common mechanisms of gene regulation with human (Guenet, 2005; Pennacchio, 2003). Murine ES cells in particular are a unique experimental model able to differentiate into any tissue, excepting trophectoderm, which makes it possible to study genetics and epigenetics of both stem cells and differentiated lineages (Jiang, et al, 2007). Murine ES cells can be obtained readily on nearly any mouse background, homozygous, F1 heterozygous, or outbred, many of which are well-characterized in terms of phenotype and markers. Contrary to human ES cells, low-passage, chromosomally normal lines of murine ES cells can be readily obtained. The ability to genetically manipulate ES cells, derive mouse lines, and direct the differentiation of these cells down specific developmental pathways allow these cells to be used for a variety of studies including knockout, developmental, and tissue-specific assays.  76  Genomic imprinting studies using ES cells are potentially hampered by concerns of epigenetic stability, particularly with extended passage or XX chromosome content (Humpherys, et al, 2001; Zvetkova, et al, 2005). Imprint instability has been observed with human in vitro fertilization, and derivation of mouse lines from ES cells, which also highlights the impact of cell culture on imprint maintenance (Cox, et al, 2002; Deng, et al, 2007; Rossignol, et al, 2006). Because of the dynamic and sensitive nature of epigenetic imprinting marks it is necessary to work with low passage ES cells (1-20 passages), to maintain proper culture conditions, especially ideal cell density, which are the same for all cells grown, to include non-imprinted control genes in analysis, and to use two or more comparable ES cell lines whenever possible to control for natural variation in methylation and gene expression levels. Epigenetic stability in ES cells is of importance to research on imprinting disorders in somatic cells, such as loss of imprinting in cancer (Holm, et al, 2005; Jackson-Grusby, et al, 1996; Sakatani, et al, 2005). Genomic imprinting has been shown to have a strong influence on colorectal cancer, where loss of imprinting causing biallelic expression of the PEG Igf2 results in increased number and size of colorectal tumors (Sakatani, et al, 2005). Holm et al. (Holm, et al, 2005) demonstrated that aside from aggravating carcinogenesis, loss of imprinting can cause cancer. By transiently demethylating ES cells Holm was able to establish a line with normal genome-wide methylation, but lacking methylation at DMRs. Tumors are formed in SCID mice injected with imprint-free fibroblasts derived from ES cells, as well as in chimeric mice derived from imprint-free ES cells, showing that loss of imprinting alone can cause tumorigenesis. The following discussion summarizes a new model for studying imprint maintenance with mitotic-mUPD in mouse ES cells, which may be a valuable tool for  77  dissecting the role of imprinting in human disease. The genetic characterization of the newly developed ES cell line, R129-E2, derived from an F1 cross between C57BL/6J (female) with the R26R insert at ROSA26 and 129S1 (male). The characterization of the seven R129*-E2 LOH sub-lines generated through high G418 selection helps shed light for the first time on the changes in imprinted gene expression and methylation in terms of imprinting maintenance in mitotic UPD. Finally, imprint maintenance in mitotic-UPD is compared to that of meiotic-UPD, and the hypothesis, which UPD caused by mitotic abnormalities leads to a disruption in imprinting, is re-examined.  5.1 Summary of results The newly-derived MMU6 mUPD sub-lines, MD6-2/dp1, MD6, and MD6-2/p1 are a novel tool to investigate maintenance of imprinting in UPD caused by mitotic aberrations. Investigations using these sub-lines have the potential to shed light on somatic diseases such as cancer, as well as developmental imprinting disorders caused by inheritance of full or partial UPD for a single chromosome. The main finding from this study is that maintenance or loss of imprinting with mitotic UPD is extremely variable, depending on the specific gene involved, and the mechanism under which UPD was incurred. Imprinting maintenance with mitotic UPD may also vary depending on which imprinted domains are involved, and upon each imprinted domain’s specific mechanism of normal imprint maintenance. Certain genes analyzed, such as Peg10, Sgce, Peg1, and Mit1 showed abnormal expression in UPD ES cell lines for which they were UPD, similar to expression levels in PG and AG ES cells (Jiang, et al, 2007; Szabo, et al, 1994). This lends support to the hypothesis that UPD causes a disruption in imprinted expression.  78  Imprinted methylation was found to be abnormal in all sub-lines with UPD for one or more imprinted domain on MMU6: MD6-2/dp1 and MD6-2/p1 showed hypermethylation of Peg1, consistent with the presence of two maternal alleles, while MD6 surprisingly showed hypomethylation of Peg1. The levels of methylation observed in these sub-lines correlates with the Peg1 expression data, as the first two sub-lines showed a near-consistent reduction of Peg1 in both ES cells and EB. The latter sub-line, however, showed either a slight decrease from wild-type, or even slightly higher levels than wild-type expression. Since these sub-lines have had relatively few passages after spontaneous mitotic LOH, it is possible that they are still in a period of epigenetic and gene expression adjustment to compensate for the loss of a paternal allele. The large difference in expression and methylation of the Peg1 locus between sub-lines may reflect the difference in mechanism of UPD, mitotic recombination vs. full chromosome duplication followed by chromosome loss.  5.2 Future directions While all three sub-lines have shown abnormal imprinted expression, primarily in one or more of Peg10, Sgce, Peg1, and Mit1 it is important to quantify these expression changes using either northern blot analysis, or quantitative real-time reverse transcriptase PCR. Equally important is the evaluation of the methylation status of the putative ICR overlapping the promoted of Peg10 (Suzuki, et al, 2007). Analysis of Peg10 methylation would allow for a comparison in imprint maintenance between two different imprinted domains, and between ES cell lines with varying regions of mUPD. To directly compare mitotically-derived UPD in ES cells with the previous research on meiotically-derived UPD in embryos, UPD ES cells could be used to derive  79  mice. When aggregated with tetraploid morula and transferred to pseudopregnant females, UPD ES cells would contribute solely to embryonic tissues, while tetraploid cells would contribute solely to the placenta (Eakin, et al, 2006). This would allow examination of imprint maintenance in mUPD MMU6 cells at various developmental time-points and tissues.  80  References Abu-Amero, S., Monk, D., Frost, J., Preece, M., Stanier, P., and Moore, G.E. (2008). The genetic aetiology of Silver-Russell syndrome. J. Med. Genet. 45, 193-199. Allen, N.D., Barton, S.C., Hilton, K., Norris, M.L., and Surani, M.A. (1994). A functional analysis of imprinting in parthenogenetic embryonic stem cells. Development 120, 14731482. Barton, S.C., Surani, M.A., and Norris, M.L. (1984). Role of paternal and maternal genomes in mouse development. Nature 311, 374-376. Batzoglou, S., Pachter, L., Mesirov, J.P., Berger, B., and Lander, E.S. (2000). Human and mouse gene structure: comparative analysis and application to exon prediction. Genome Res. 10, 950-958. Beechey, C.V. (2004). A reassessment of imprinting regions and phenotypes on mouse chromosome 6: Nap1l5 locates within the currently defined sub-proximal imprinting region. Cytogenet. Genome Res. 107, 108-114. Beechey, C.V. (2000). Peg1/Mest locates distal to the currently defined imprinting region on mouse proximal chromosome 6 and identifies a new imprinting region affecting growth. Cytogenet. Cell Genet. 90, 309-314. Blagitko, N., Schulz, U., Schinzel, A.A., Ropers, H.H., and Kalscheuer, V.M. (1999). gamma2-COP, a novel imprinted gene on chromosome 7q32, defines a new imprinting cluster in the human genome. Hum. Mol. Genet. 8, 2387-2396. Cattanach, B.M., Barr, J.A., Beechey, C.V., Martin, J., Noebels, J., and Jones, J. (1997). A candidate model for Angelman syndrome in the mouse. Mamm. Genome 8, 472-478. Cattanach, B.M., and Jones, J. (1994). Genetic imprinting in the mouse: implications for gene regulation. J. Inherit. Metab. Dis. 17, 403-420. Cattanach, B.M., Shibata, H., Hayashizaki, Y., Townsend, K.M., Ball, S., and Beechey, C.V. (1998). Association of a redefined proximal mouse chromosome 11 imprinting region and U2afbp-rs/U2af1-rs1 expression. Cytogenet. Cell Genet. 80, 41-47. Cerrato, F., Sparago, A., Verde, G., De Crescenzo, A., Citro, V., Cubellis, M.V., Rinaldi, M.M., Boccuto, L., Neri, G., Magnani, C. et al. (2008). Different mechanisms cause imprinting defects at the IGF2/H19 locus in Beckwith-Wiedemann syndrome and Wilms' tumour. Hum. Mol. Genet. 17, 1427-1435.  81  Cervantes, R.B., Stringer, J.R., Shao, C., Tischfield, J.A., and Stambrook, P.J. (2002). Embryonic stem cells and somatic cells differ in mutation frequency and type. Proc. Natl. Acad. Sci. U. S. A. 99, 3586-3590. Clark, M.B., Janicke, M., Gottesbuhren, U., Kleffmann, T., Legge, M., Poole, E.S., and Tate, W.P. (2007). Mammalian gene PEG10 expresses two reading frames by high efficiency -1 frameshifting in embryonic-associated tissues. J. Biol. Chem. 282, 3735937369. Cooper, W.N., Luharia, A., Evans, G.A., Raza, H., Haire, A.C., Grundy, R., Bowdin, S.C., Riccio, A., Sebastio, G., Bliek, J. et al. (2005). Molecular subtypes and phenotypic expression of Beckwith-Wiedemann syndrome. Eur. J. Hum. Genet. 13, 1025-1032. Cox, G.F., Burger, J., Lip, V., Mau, U.A., Sperling, K., Wu, B.L., and Horsthemke, B. (2002). Intracytoplasmic sperm injection may increase the risk of imprinting defects. Am. J. Hum. Genet. 71, 162-164. Cuisset, L., Le Stunff, C., Dupont, J.M., Vasseur, C., Cartigny, M., Despert, F., Delpech, M., Bougnere, P., and Jeanpierre, M. (1997). PEG1 expression in maternal uniparental disomy 7. Ann. Genet. 40, 211-215. Dacquin, R., Davey, R.A., Laplace, C., Levasseur, R., Morris, H.A., Goldring, S.R., Gebre-Medhin, S., Galson, D.L., Zajac, J.D., and Karsenty, G. (2004). Amylin inhibits bone resorption while the calcitonin receptor controls bone formation in vivo. J. Cell Biol. 164, 509-514. Davies, W., Smith, R.J., Kelsey, G., and Wilkinson, L.S. (2004). Expression patterns of the novel imprinted genes Nap1l5 and Peg13 and their non-imprinted host genes in the adult mouse brain. Gene Expr. Patterns 4, 741-747. de la Casa-Esperon, E., and Sapienza, C. (2003). Natural selection and the evolution of genome imprinting. Annu. Rev. Genet. 37, 349-370. Deng, T., Kuang, Y., Zhang, D., Wang, L., Sun, R., Xu, G., Wang, Z., and Fei, J. (2007). Disruption of imprinting and aberrant embryo development in completely inbred embryonic stem cell-derived mice. Dev. Growth Differ. 49, 603-610. Eakin, G.S., and Hadjantonakis, A.K. (2006). Production of chimeras by aggregation of embryonic stem cells with diploid or tetraploid mouse embryos. Nat. Protoc. 1, 11451153.  82  Eggermann, T., Eggermann, K., and Schonherr, N. (2008). Growth retardation versus overgrowth: Silver-Russell syndrome is genetically opposite to Beckwith-Wiedemann syndrome. Trends Genet. 24, 195-204. Eggermann, T., Meyer, E., Ranke, M.B., Holder, M., Spranger, S., Zerres, K., and Wollmann, H.A. (2005). Diagnostic proceeding in Silver-Russell syndrome. Mol. Diagn. 9, 205-209. Ekstrom, T.J., Cui, H., Li, X., and Ohlsson, R. (1995). Promoter-specific IGF2 imprinting status and its plasticity during human liver development. Development 121, 309-316. Fazzari, M.J., and Greally, J.M. (2004). Epigenomics: beyond CpG islands. Nat. Rev. Genet. 5, 446-455. 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. Guenet, J.L. (2005). The mouse genome. Genome Res. 15, 1729-1740. Hanel, M.L., and Wevrick, R. (2001). Establishment and maintenance of DNA methylation patterns in mouse Ndn: implications for maintenance of imprinting in target genes of the imprinting center. Mol. Cell. Biol. 21, 2384-2392. Hayward, B.E., Kamiya, M., Strain, L., Moran, V., Campbell, R., Hayashizaki, Y., and Bonthron, D.T. (1998). The human GNAS1 gene is imprinted and encodes distinct paternally and biallelically expressed G proteins. Proc. Natl. Acad. Sci. U. S. A. 95, 10038-10043. Hershko, A., Razin, A., and Shemer, R. (1999). Imprinted methylation and its effect on expression of the mouse Zfp127 gene. Gene 234, 323-327. Holm, T.M., Jackson-Grusby, L., Brambrink, T., Yamada, Y., Rideout, W.M.,3rd, and Jaenisch, R. (2005). Global loss of imprinting leads to widespread tumorigenesis in adult mice. Cancer. Cell. 8, 275-285. Horii, T., Kimura, M., Morita, S., Nagao, Y., and Hatada, I. (2008). Loss of genomic imprinting in mouse parthenogenetic embryonic stem cells. Stem Cells 26, 79-88. Hoshiya, H., Meguro, M., Kashiwagi, A., Okita, C., and Oshimura, M. (2003). Calcr, a brain-specific imprinted mouse calcitonin receptor gene in the imprinted cluster of the proximal region of chromosome 6. J. Hum. Genet. 48, 208-211. 83  Humpherys, D., Eggan, K., Akutsu, H., Hochedlinger, K., Rideout, W.M.,3rd, Biniszkiewicz, D., Yanagimachi, R., and Jaenisch, R. (2001). Epigenetic instability in ES cells and cloned mice. Science 293, 95-97. Jackson-Grusby, L., and Jaenisch, R. (1996). Experimental manipulation of genomic methylation. Semin. Cancer Biol. 7, 261-268. JAX® Mice Database - 003474 B6.129S4-Gt(ROSA)26Sor<tm1Sor>/J.2007. Jiang, H., Sun, B., Wang, W., Zhang, Z., Gao, F., Shi, G., Cui, B., Kong, X., He, Z., Ding, X. et al. (2007). Activation of paternally expressed imprinted genes in newly derived germline-competent mouse parthenogenetic embryonic stem cell lines. Cell Res. 17, 792803. Kaneko-Ishino, T., Kuroiwa, Y., Miyoshi, N., Kohda, T., Suzuki, R., Yokoyama, M., Viville, S., Barton, S.C., Ishino, F., and Surani, M.A. (1995). Peg1/Mest imprinted gene on chromosome 6 identified by cDNA subtraction hybridization. Nat. Genet. 11, 52-59. Kent, L., Bowdin, S., Kirby, G.A., Cooper, W.N., and Maher, E.R. (2008). Beckwith Weidemann syndrome: A behavioral phenotype-genotype study. Am. J. Med. Genet. B. Neuropsychiatr. Genet. [Epub ahead of print]. Kile, B.T., Viney, E.M., Willson, T.A., Brodnicki, T.C., Cancilla, M.R., Herlihy, A.S., Croker, B.A., Baca, M., Nicola, N.A., Hilton, D.J., and Alexander, W.S. (2000). Cloning and characterization of the genes encoding the ankyrin repeat and SOCS box-containing proteins Asb-1, Asb-2, Asb-3 and Asb-4. Gene 258, 31-41. Kimura, M.I., Kazuki, Y., Kashiwagi, A., Kai, Y., Abe, S., Barbieri, O., Levi, G., and Oshimura, M. (2004). Dlx5, the mouse homologue of the human-imprinted DLX5 gene, is biallelically expressed in the mouse brain. J. Hum. Genet. 49, 273-277. Kobayashi, S., Kohda, T., Miyoshi, N., Kuroiwa, Y., Aisaka, K., Tsutsumi, O., KanekoIshino, T., and Ishino, F. (1997). Human PEG1/MEST, an imprinted gene on chromosome 7. Hum. Mol. Genet. 6, 781-786. 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. Kotzot, D. (2004). Advanced parental age in maternal uniparental disomy (UPD): implications for the mechanism of formation. Eur. J. Hum. Genet. 12, 343-346.  84  Lalande, M., and Calciano, M.A. (2007). Molecular epigenetics of Angelman syndrome. Cell Mol. Life Sci. 64, 947-960. Lee, Y.J., Park, C.W., Hahn, Y., Park, J., Lee, J., Yun, J.H., Hyun, B., and Chung, J.H. (2000). Mit1/Lb9 and Copg2, new members of mouse imprinted genes closely linked to Peg1/Mest(1). FEBS Lett. 472, 230-234. Lefebvre, L., Dionne, N., Karaskova, J., Squire, J.A., and Nagy, A. (2001). Selection for transgene homozygosity in embryonic stem cells results in extensive loss of heterozygosity. Nat. Genet. 27, 257-258. Lefebvre, L., Viville, S., Barton, S.C., Ishino, F., Keverne, E.B., and Surani, M.A. (1998). Abnormal maternal behaviour and growth retardation associated with loss of the imprinted gene Mest. Nat. Genet. 20, 163-169. Lefebvre, L., Viville, S., Barton, S.C., Ishino, F., and Surani, M.A. (1997). Genomic structure and parent-of-origin-specific methylation of Peg1. Hum. Mol. Genet. 6, 19071915. Luedi, P.P., Hartemink, A.J., and Jirtle, R.L. (2005). Genome-wide prediction of imprinted murine genes. Genome Res. 15, 875-884. Lynch, C.A., Tycko, B., Bestor, T.H., and Walsh, C.P. (2002). Reactivation of a silenced H19 gene in human rhabdomyosarcoma by demethylation of DNA but not by histone hyperacetylation. Mol. Cancer. 1, 2. McGrath, J., and Solter, D. (1984). Completion of mouse embryogenesis requires both the maternal and paternal genomes. Cell 37, 179-183. Monk, D., Wagschal, A., Arnaud, P., Muller, P., Parker-Katiraee, L., Bourchis, 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 imprinted gene, TFPI2/Tfpi2, which requires G9a and Eed for allelic-silencing. Genome Res. 18(8):1270-81. Morison, I.M., Ramsay, J.P., and Spencer, H.G. (2005). A census of mammalian imprinting. Trends Genet. 21, 457-465. Murphy, S.K., Wylie, A.A., and Jirtle, R.L. (2001). Imprinting of PEG3, the human homologue of a mouse gene involved in nurturing behavior. Genomics 71, 110-117.  85  Nagy, A. (2003). Manipulating the mouse embryo : a laboratory manual (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press). Ng, C.J., Bourquard, N., Grijalva, V., Hama, S., Shih, D.M., Navab, M., Fogelman, A.M., Lusis, A.J., Young, S., and Reddy, S.T. (2006). Paraoxonase-2 deficiency aggravates atherosclerosis in mice despite lower apolipoprotein-B-containing lipoproteins: antiatherogenic role for paraoxonase-2. J. Biol. Chem. 281, 29491-29500. Obata, Y., Kaneko-Ishino, T., Koide, T., Takai, Y., Ueda, T., Domeki, I., Shiroishi, T., Ishino, F., and Kono, T. (1998). Disruption of primary imprinting during oocyte growth leads to the modified expression of imprinted genes during embryogenesis. Development 125, 1553-1560. Ogawa, H., Wu, Q., Komiyama, J., Obata, Y., and Kono, T. (2006). Disruption of parental-specific expression of imprinted genes in uniparental fetuses. FEBS Lett. 580, 5377-5384. Okamura, K., Hagiwara-Takeuchi, Y., Li, T., Vu, T.H., Hirai, M., Hattori, M., Sakaki, Y., Hoffman, A.R., and Ito, T. (2000). Comparative genome analysis of the mouse imprinted gene impact and its nonimprinted human homolog IMPACT: toward the structural basis for species-specific imprinting. Genome Res. 10, 1878-1889. Okano, M., Bell, D.W., Haber, D.A., and Li, E. (1999). DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99, 247-257. Ono, R., Kobayashi, S., Wagatsuma, H., Aisaka, K., Kohda, T., Kaneko-Ishino, T., and Ishino, F. (2001). A retrotransposon-derived gene, PEG10, is a novel imprinted gene located on human chromosome 7q21. Genomics 73, 232-237. Ono, R., Nakamura, K., Inoue, K., Naruse, M., Usami, T., Wakisaka-Saito, N., Hino, T., Suzuki-Migishima, R., Ogonuki, N., Miki, H. et al. (2006). Deletion of Peg10, an imprinted gene acquired from a retrotransposon, causes early embryonic lethality. Nat. Genet. 38, 101-106. Ono, R., Shiura, H., Aburatani, H., Kohda, T., Kaneko-Ishino, T., and Ishino, F. (2003). Identification of a large novel imprinted gene cluster on mouse proximal chromosome 6. Genome Res. 13, 1696-1705. Paludan, K., Duch, M., Jorgensen, P., Kjeldgaard, N.O., and Pedersen, F.S. (1989). Graduated resistance to G418 leads to differential selection of cultured mammalian cells expressing the neo gene. Gene 85, 421-426. 86  Paoloni-Giacobino, A., and Chaillet, J.R. (2006). The role of DMDs in the maintenance of epigenetic states. Cytogenet. Genome Res. 113, 116-121. Pardo-Manuel de Villena, F., de la Casa-Esperon, E., and Sapienza, C. (2000). Natural selection and the function of genome imprinting: beyond the silenced minority. Trends Genet. 16, 573-579. Park, J.I., Yoshida, I., Tada, T., Takagi, N., Takahashi, Y., and Kanagawa, H. (1998). Trisomy 8 does not affect differentiative potential in a murine parthenogenetic embryonic stem cell line. Jpn. J. Vet. Res. 46, 29-35. Parker-Katiraee, L., Carson, A.R., Yamada, T., Arnaud, P., Feil, R., Abu-Amero, S.N., Moore, G.E., Kaneda, M., Perry, G.H., Stone, A.C. et al. (2007). Identification of the imprinted KLF14 transcription factor undergoing human-specific accelerated evolution. PLoS Genet. 3, e65. Paulsen, M., El-Maarri, O., Engemann, S., Strodicke, 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. Pennacchio, L.A. (2003). Insights from human/mouse genome comparisons. Mamm. Genome 14, 429-436. Peters, J., and Beechey, C. (2004). Identification and characterisation of imprinted genes in the mouse. Brief Funct. Genomic Proteomic 2, 320-333. Piras, G., El Kharroubi, A., Kozlov, S., Escalante-Alcalde, D., Hernandez, L., Copeland, N.G., Gilbert, D.J., Jenkins, N.A., and Stewart, C.L. (2000). Zac1 (Lot1), a potential tumor suppressor gene, and the gene for epsilon-sarcoglycan are maternally imprinted genes: identification by a subtractive screen of novel uniparental fibroblast lines. Mol. Cell. Biol. 20, 3308-3315. Reik, W., and Walter, J. (2001). Genomic imprinting: parental influence on the genome. Nat. Rev. Genet. 2, 21-32. Rossignol, S., Steunou, V., Chalas, C., Kerjean, A., Rigolet, M., Viegas-Pequignot, E., Jouannet, P., Le Bouc, Y., and Gicquel, C. (2006). The epigenetic imprinting defect of patients with Beckwith-Wiedemann syndrome born after assisted reproductive technology is not restricted to the 11p15 region. J. Med. Genet. 43, 902-907.  87  Sakatani, T., Kaneda, A., Iacobuzio-Donahue, C.A., Carter, M.G., de Boom Witzel, S., Okano, H., Ko, M.S., Ohlsson, R., Longo, D.L., and Feinberg, A.P. (2005). Loss of imprinting of Igf2 alters intestinal maturation and tumorigenesis in mice. Science 307, 1976-1978. Sambrook, J., and Russell, D.W. (2001). Molecular cloning : a laboratory manual (Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press). Sauer, B., and Henderson, N. (1989). Cre-stimulated recombination at loxP-containing DNA sequences placed into the mammalian genome. Nucleic Acids Res. 17, 147-161. Shao, C., Stambrook, P.J., and Tischfield, J.A. (2001). Mitotic recombination is suppressed by chromosomal divergence in hybrids of distantly related mouse strains. Nat. Genet. 28, 169-172. Shih, D.M., Xia, Y.R., Wang, X.P., Wang, S.S., Bourquard, N., Fogelman, A.M., Lusis, A.J., and Reddy, S.T. (2007). Decreased obesity and atherosclerosis in human paraoxonase 3 transgenic mice. Circ. Res. 100, 1200-1207. Smith, R.J., Dean, W., Konfortova, G., and Kelsey, G. (2003). Identification of novel imprinted genes in a genome-wide screen for maternal methylation. Genome Res. 13, 558569. Sommer, C.A., Pavarino-Bertelli, E.C., Goloni-Bertollo, E.M., and Henrique-Silva, F. (2008). Identification of dysregulated genes in lymphocytes from children with Down syndrome. Genome 51, 19-29. Song, J., Lu, Y., Giang, A., Pang, S., and Chiu, R. (2008). Promoter analysis of the mouse Peg3 gene. Biochim. Biophys. Acta 1779, 134-138. Soriano, P. (1999). Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70-71. Soriano, P. Soriano lab home page.2007, Sotomaru, Y., Katsuzawa, Y., Hatada, I., Obata, Y., Sasaki, H., and Kono, T. (2002). Unregulated expression of the imprinted genes H19 and Igf2r in mouse uniparental fetuses. J. Biol. Chem. 277, 12474-12478. Sotomaru, Y., Kawase, Y., Ueda, T., Obata, Y., Suzuki, H., Domeki, I., Hatada, I., and Kono, T. (2001). Disruption of imprinted expression of U2afbp-rs/U2af1-rs1 gene in mouse parthenogenetic fetuses. J. Biol. Chem. 276, 26694-26698. 88  Sugawara, A., Goto, K., Sotomaru, Y., Sofuni, T., and Ito, T. (2006). Current status of chromosomal abnormalities in mouse embryonic stem cell lines used in Japan. Comp. Med. 56, 31-34. Sunahara, S., Nakamura, K., Nakao, K., Gondo, Y., Nagata, Y., and Katsuki, M. (2000). The oocyte-specific methylated region of the U2afbp-rs/U2af1-rs1 gene is dispensable for its imprinted methylation. Biochem. Biophys. Res. Commun. 268, 590-595. 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. Suzuki, S., Ono, R., Narita, T., Pask, A.J., Shaw, G., Wang, C., Kohda, T., Alsop, A.E., Marshall Graves, J.A., Kohara, Y. et al. (2007). Retrotransposon silencing by DNA methylation can drive mammalian genomic imprinting. PLoS Genet. 3, e55. Suzuki, S., Renfree, M.B., Pask, A.J., 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. Szabo, P., and Mann, J.R. (1994). Expression and methylation of imprinted genes during in vitro differentiation of mouse parthenogenetic and androgenetic embryonic stem cell lines. Development 120, 1651-1660. Tucker, K.L., Beard, C., Dausmann, J., Jackson-Grusby, L., Laird, P.W., Lei, H., Li, E., and Jaenisch, R. (1996). Germ-line passage is required for establishment of methylation and expression patterns of imprinted but not of nonimprinted genes. Genes Dev. 10, 10081020. Tycko, B., and Morison, I.M. (2002). Physiological functions of imprinted genes. J. Cell. Physiol. 192, 245-258. Vacik, T., and Forejt, J. (2003). Quantification of expression and methylation of the Igf2r imprinted gene in segmental trisomic mouse model. Genomics 82, 261-268. Vrieling, H. (2001). Mitotic maneuvers in the light. Nat. Genet. 28, 101-102. Walsh, C., Glaser, A., Fundele, R., Ferguson-Smith, A., Barton, S., Surani, M.A., and Ohlsson, R. (1994). The non-viability of uniparental mouse conceptuses correlates with the loss of the products of imprinted genes. Mech. Dev. 46, 55-62. Wang, Y., and Leung, F.C. (2004). An evaluation of new criteria for CpG islands in the human genome as gene markers. Bioinformatics 20, 1170-1177. 89  Wood, A.J., and Oakey, R.J. (2006). Genomic imprinting in mammals: emerging themes and established theories. PLoS Genet. 2, e147. Yamasaki, K., Hayashida, S., Miura, K., Masuzaki, H., Ishimaru, T., Niikawa, N., and Kishino, T. (2000). The novel gene, gamma2-COP (COPG2), in the 7q32 imprinted domain escapes genomic imprinting. Genomics 68, 330-335. Yamasaki-Ishizaki, Y., Kayashima, T., Mapendano, C.K., Soejima, H., Ohta, T., Masuzaki, H., Kinoshita, A., Urano, T., Yoshiura, K., Matsumoto, N. et al. (2007). Role of DNA methylation and histone H3 lysine 27 methylation in tissue-specific imprinting of mouse Grb10. Mol. Cell. Biol. 27, 732-742. Yenofsky, R.L., Fine, M., and Pellow, J.W. (1990). A mutant neomycin phosphotransferase II gene reduces the resistance of transformants to antibiotic selection pressure. Proc. Natl. Acad. Sci. U. S. A. 87, 3435-3439. Yokoi, F., Dang, M.T., Mitsui, S., and Li, Y. (2005). Exclusive paternal expression and novel alternatively spliced variants of epsilon-sarcoglycan mRNA in mouse brain. FEBS Lett. 579, 4822-4828. Yun, J., Park, C.W., Lee, Y.J., and Chung, J.H. (2003). Allele-specific methylation at the promoter-associated CpG island of mouse Copg2. Mamm. Genome 14, 376-382. Zeschnigk, M., Albrecht, B., Buiting, K., Kanber, D., Eggermann, T., Binder, G., Gromoll, J., Prott, E.C., Seland, S., and Horsthemke, B. (2008). IGF2/H19 hypomethylation in Silver-Russell syndrome and isolated hemihypoplasia. Eur. J. Hum. Genet. 16, 328-334. 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.  90  


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