UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Transcriptional regulation of the XIST locus Chapman, Andrew Glen 2013

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

Item Metadata

Download

Media
24-ubc_2013_fall_chapman_andrew.pdf [ 3.68MB ]
Metadata
JSON: 24-1.0074252.json
JSON-LD: 24-1.0074252-ld.json
RDF/XML (Pretty): 24-1.0074252-rdf.xml
RDF/JSON: 24-1.0074252-rdf.json
Turtle: 24-1.0074252-turtle.txt
N-Triples: 24-1.0074252-rdf-ntriples.txt
Original Record: 24-1.0074252-source.json
Full Text
24-1.0074252-fulltext.txt
Citation
24-1.0074252.ris

Full Text

  Transcriptional regulation of the human XIST locus   by  Andrew Glen Chapman B.Sc., Memorial University of Newfoundland, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  Master of Science in The Faculty of Graduate Studies and Postdoctoral Studies (Medical Genetics) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2013  ? Andrew Glen Chapman, 2013  ii  Abstract X-chromosome inactivation is a mechanism that has evolved in mammalian females allowing dosage compensation of X-linked genes.  A region of the X chromosome called the X-inactivation centre (XIC) is required for X inactivation to occur.  Within this region is a long non-coding RNA, XIST/Xist, which is upregulated on the future inactive X and initiates silencing.  A major questions in the field of X inactivation is how XIST/Xist is regulated, becoming expressed on the inactive X and silenced on the active X.  Much of what we currently know about XIST/Xist regulation comes from studies using mice, however, differences in conservation of the XIC and expression patterns of the major mouse Xist regulator, Tsix, indicate that humans and mice may regulate XIST/Xist differently.  The objective of this thesis was to identify regulatory elements that are important for regulation of XIST in humans.   Since regulatory elements controlling XIST are believed to reside within the XIC, we searched the XIC and identified two inactive X specific regulatory elements within the 5? end of XIST using DNase I hypersensitivity mapping.  We found one of the hypersensitive sites to be acting as an alternative P2 promoter for XIST which contains an upstream antisense promoter, P2as.  The second hypersensitive site was associated with alternative splicing and inclusion of two novel exons for XIST. Interestingly, both P2 and the novel alternative splicing result in transcripts that lack functional domains of XIST.  An additional candidate regulator is the region 3? of XIST due to the importance of Tsix in mice.  We found that transcription 3? of XIST in somatic cells is low level sense transcription so we believe this to be leaky XIST rather than TSIX. In human embryonic stem cells we found an antisense transcript that extends the full length of XIST providing the first evidence for mouse-like TSIX in humans but very low-levels of this transcript argue against regulatory ability.  Taken together, our results highlight the differences between mouse and human X inactivation and indicate that XIST transcription is more complex than previously thought, generating XIST molecules that lack functional domains.        iii  Preface This thesis is a product of original and independent intellectual work by the author.   iv  Table of Contents Abstract ................................................................................................................................................................. ii Preface ................................................................................................................................................................. iii Table of Contents ................................................................................................................................................. iv List of Tables ......................................................................................................................................................... vi List of Figures ....................................................................................................................................................... vii List of Abbreviations ........................................................................................................................................... viii List of Gene Names ............................................................................................................................................... ix Acknowledgements ............................................................................................................................................... x 1 Introduction .................................................................................................................................................. 1 1.1 Thesis overview ................................................................................................................................. 2 1.2 Dosage compensation of sex-linked genes ....................................................................................... 2 1.3 X-inactivation centre ......................................................................................................................... 3 1.4 X inactive specific transcript (XIST/Xist) ............................................................................................ 4 1.4.1 Evidence for XIST/Xist role in silencing ............................................................................ 5 1.4.2 Localization of XIST/Xist ................................................................................................... 5 1.5 Initiation and establishment of XCI ................................................................................................... 6 1.6 Counting and choice.......................................................................................................................... 7 1.7 XIST/Xist regulators: modulators of counting and choice ................................................................. 9 1.7.1 X-controlling element (Xce) .............................................................................................. 9 1.7.2 TSIX/Tsix ......................................................................................................................... 10 1.7.3 CTCF and YY1 .................................................................................................................. 12 1.7.4 Pluripotency factors ....................................................................................................... 12 1.7.5 Non-coding RNAs: JPX/Jpx, FTX/Ftx, RepA, XACT ........................................................... 13 1.7.6 RNF12/Rnf12 .................................................................................................................. 14 1.8 Developmental models for human XCI and XIST regulation ........................................................... 14 1.9 Genomic approaches to identifying regulatory elements in the XIC .............................................. 16 1.10 Thesis objective ............................................................................................................................... 17 2 Materials and Methods ............................................................................................................................... 19 2.1 Tissue culture and cell lines ............................................................................................................ 20 2.2 PCR and quantitative PCR ............................................................................................................... 20 2.3 RNA extraction and reverse transcription ....................................................................................... 21 2.4 DNase I hypersensitivity .................................................................................................................. 21 2.5 5? and 3? rapid amplification of cDNA ends ..................................................................................... 22 2.6 siRNA-mediated knockdown ........................................................................................................... 22 2.7 Statistical analysis ........................................................................................................................... 22 3 Results ........................................................................................................................................................ 28 3.1 Refining the boundaries of the XIC ................................................................................................. 29 3.2 Xi-specific DNase I hypersensitivity site at the 5? end of XIST ......................................................... 29 3.3 DNase I hypersensitivity at DHS 200b.1 is associated with an actively transcribing promoter ...... 32 3.4 Regulatory features of the P2 promoter ......................................................................................... 35 3.5 siRNA-mediated knockdown of YY1 diminishes XIST expression but not exclusively at P2 ............ 39 3.6 DNase I hypersensitivity at DHS 200a.1 is associated with an alternative splice site of XIST. ........ 39 3.7 Transcription 3? of XIST ................................................................................................................... 42 3.8 Male hES cell line, CA1S, transcribes in antisense orientation across XIST locus ........................... 42 v   4 Discussion ................................................................................................................................................... 46 References ........................................................................................................................................................... 55     vi  List of Tables Table 2.1: List of Primers (Shown 5? to 3?) ...................................................................................... 23   vii  List of Figures Figure 1.1 - Regulation of XCI and XIST/Xist by the XIC/Xic ........................................................................18 Figure 3.1 ? Refining the boundaries of the XIC ..........................................................................................30 Figure 3.2 - Xi-specific DNase I hypersensitive site ~1.5kb within XIST. .....................................................31 Figure 3.3 - DHS 200b.1 show enrichment of promoter associated proteins .............................................33 Figure 3.4 - Xi specific DHS site corresponds to an actively transcribing P2 promoter ..............................34 Figure 3.5 - Protection of P2 from transcriptional interference is not due to separate transcripts or imprinted promoters ...................................................................................................................................36 Figure 3.6 - Antisense transcription from P2 promoter. .............................................................................38 Figure 3.7 - Knockdown of YY1 causes depletion XIST upstream and downstream of P2 ..........................40 Figure 3.8 - DHS 200a.1 is  a site of alternative splicing .............................................................................41 Figure 3.9 - Transcription beyond the 3? end of XIST is leaky XIST transcription ........................................43 Figure 3.10 - Transcription at the XIST locus in CA1S male hES cells ..........................................................45 Figure 4.1 ? Summary of findings and proposed model .............................................................................54     viii  List of Abbreviations bp = base pairs cDNA = complementary deoxyribonucleic acid ChIP = chromatin immunoprecipitation DNA = deoxyribonucleic acid DNMT/Dnmt = DNA methyltransferase DMEM = Dulbecco Modified Eagle Medium dNTPs = deoxynucleosides ES = embryonic stem H = Histone K = Lysine kb = kilo base pairs LNA = locked nucleic acid Mb = mega base pairs MEM = minimal essential media PBS = phosphate buffered saline PCR = polymerase chain reaction qPCR = quantitative-polymerase chain reaction RNA = ribonucleic acid RNA Pol II = RNA polymerase II RT = reverse transcription RT-PCR = reverse transcription polymerase chain reaction RPMI = Roswell Parks Memorial Institute TSS = transcription start site UCSC = University of California ? Santa Cruz Xa = active X chromosome Xi = inactive X chromosome XCI = X chromosome inactivation XIC/Xic = X inactivation centre ix  List of Gene Names Genes in all capital letters refer to human gene names whereas mouse gene names are in lowercase letters.  CTCF/Ctcf = CCCTC-binding factor (zinc finger protein) DNMT1/Dnmt1 = DNA methyltransferase 1 DNMT3A/Dnmt3a = DNA methyltransferase 3 alpha DNMT3B/Dnmt3b = DNA (cytosine-5)-methyltransferase 3 beta FTX/Ftx = five prime to XIST/Xist RNF12/Rnf12 = ring finger protein 12 XACT = Active X coating transcript XIST/Xist = Xi specific transcript Xite = X inactivation intergenic transcript  YY1/Yy1 = Yin Yang 1                x  Acknowledgements  I am very thankful to have gotten the opportunity to hit the frontier of science and to write this thesis.  I would like to give genuine thanks to my supervisor, Dr. Carolyn Brown for providing me with this opportunity, along with plenty of scientific-wiggle room and patience.  I would also like to thank Sarah Baldry, Allison Cotton, Jakub Minks, Angela Kelsey, Sam Peeters and Christine Yang in the Brown lab whose technical and intellectual mentorship have contributed greatly to the completion of this thesis.     My parents and sister deserve more thanks, for their faith in my choices and for their constant support, than I can give.  Lastly, I would like to thank Abby for her presence throughout this process; her daily support has been valued immeasurably.  xi  Dedication       To Mom and Dad1           1 Introduction2  1.1 Thesis overview  Long non-coding RNAs have emerged in recent years as a surprising addition to the transcriptomes of many species and have been found to play crucial roles in gene regulation and development.  Specifically, long non-coding RNAs have been tied into regulatory networks that control chromatin and cellular differentiation.  One of the most famous long non coding RNAs is XIST which is capable of initiating chromosome-wide epigenetic modifications in a crucial gene regulatory process in development known as X inactivation.   There are two major questions in the field of X inactivation: 1) how does XIST fulfill its role in initiating chromosome-wide silencing and 2) how does XIST become upregulated on one of two essentially identical X chromosomes,  and remain silenced on the other.   Mice have provided an excellent model system for elucidating the question of how XIST is regulated but apparent differences between the mouse and human system suggest that humans may have evolved a unique mechanism of XIST regulation.  In this thesis, we examine the role of cis-regulatory elements in controlling XIST expression in human cells in an effort to elucidate XIST?s complex role in XCI and as a model for understanding the control of long non-coding RNAs and their role in developmental processes.    1.2 Dosage compensation of sex-linked genes Chromosomes in diploid genomes are ordered into homologous pairs with the exception being the sex-determining chromosomes, known as the X and Y chromosomes in mammals.  The Y chromosome evolved when one allele of a gene developed a mutation to become the sex-determining gene SRY.  Clustering of sex-determining genes near the SRY locus resulted from limited ability for homologous recombination to occur between the sex chromosomes leading to the degradation of the Y chromosome [1].  The divergence of the sex chromosomes caused a substantial dosage imbalance of the genes on the X chromosome not only with respect to the sexes, but also relative to the autosomes in the genome.  To overcome the differences in gene dosage between males and females many species have evolved 3  mechanisms of dosage compensation.  In mammals, dosage compensation between XX females and XY males occurs through a process called X-chromosome inactivation (XCI) which allows transcriptional silencing of one of the two X chromosomes in females to generate one inactive X chromosome (Xi) and one active X chromosome (Xa) [2].  Once an Xi and Xa have been chosen, the status is maintained in all subsequent daughter cells.    1.3 X-inactivation centre The X-inactivation centre (Xic) was first described as a region of the X chromosome required for cis inactivation to occur.  In humans, this region was determined through studies of X:autosome (X:A) translocations in which spreading of silencing into autosomal DNA occurred only when the full XIC was present [3].  DNA was used from patients containing structurally abnormal chromosomes that are still capable of XCI for molecular characterization of the XIC boundaries.  An X:14 translocation [4] and a rearranged X chromosome [5, 6] were used to delineate the proximal and distal boundaries of the XIC, respectively, and the boundaries on both ends were narrowed to a region of ~260kb.  From these studies the XIC is estimated to be between 680-1200 kb in size.  Similarly, the mouse Xic which maps to a syntenic region, was uncovered based on the observation that X chromosomes containing disruptions of an X-linked locus (the Xic), either by X:autosome translocations or truncated X chromosomes, were incapable of initiating XCI [7].   The loss of XCI ability in animals with an incomplete Xic suggested that within this locus there must be genetic elements that are required for chromosome wide silencing.  Indeed, several genes within the Xic have been implicated in the XCI process and will be outlined in detail in the following sections. Overall, there is poor sequence conservation within the XIC/Xic.  The mouse Xic is smaller than the human Xic with much less intergenic sequence between described genes (Figure 1.1).   Chureau et al. [8] identified 72kb of conserved sequence in the XIC/Xic and 62kb of the conserved sequence was located in known genes.  It is therefore likely that beyond the known genes, mouse and human do not share many regulatory elements that would be found in intergenic regions of the XIC/Xic.     4   1.4 X inactive specific transcript (XIST/Xist)  XIST/Xist is a gene found within the XIC/Xic that was originally described as having a unique expression pattern which is exclusive to the Xi [9, 10] and has since been shown to be essential for XCI.  The transcribed product of XIST/Xist is polyadenylated but contains limited protein coding potential and is therefore believed to be a long non-coding RNA of 17kb in humans and 15kb in mice[11, 12].  Overall, XIST/Xist is poorly conserved with only 49% homology between mouse and human [13] but throughout the gene are six tandem repeat elements named Repeat A-F  that, despite differences in size, are conserved between species [14] (Figure 1.1).  In fact, Repeat A as well as a region at the boundary of exon 1 and intron 1 and exon 4 were identified as being very highly conserved in a comparison of 10 mammal species [14].   Johnston et al. [15] first described an alternative promoter for Xist, designated P2, which lies ~1.5kb within Xist overlapping Repeat F.  By RNase Protection Assay, transcripts initiating from P2 were found to be much more abundant than transcripts originating from the canonical P1 promoter.  Another report however, noted 10 fold more binding of TFIIB, a general transcription factor of the pre-initiation complex, at P1 than P2 and 2 fold more binding of RNA polymerase II at P1 than P2 suggesting that of the two Xist promoters P1 is considerably more primed for transcription initiation [16].  Interestingly, transcripts originating from the P2 promoter are also deficient of Repeat A which is crucial to the function of Xist (discussed in Section 1.4.1) but the function of P2 transcripts in mouse has never been directly assayed.  In humans, P1 driving a promoterless reporter showed limited reporter expression but when a region at a similar location to mouse P2 was included in combination with P1, increased reporter transcription was observed providing evidence for P2 conservation in humans [17].     XIST RNA is spliced into 8 exons and shows variable splicing patterns in which some transcripts omit exons 3, 4 and 7 and the 3? half of exon 6 [11].  Regions of XIST/Xist are generally believed to have two functional abilities: 1) localize to the chromosome that it is expressed from and 2) initiate silencing of the X chromosome.   5   1.4.1 Evidence for XIST/Xist role in silencing When XIST expression is lost in somatic cells the X chromosome does not reactivate [18] indicating that there is a specific developmental window during the time of XCI initiation where XIST functions.  Therefore, to study Xist function at an earlier time in development, mouse embryonic stem cells (ES cells) have proven invaluable due to their ability to initiate XCI upon differentiation [19].  Penny et al. used mouse ES cells with X chromosomes from two different backgrounds to show that wild-type cells underwent random XCI but cells that contained a targeted deletion of a portion of Xist showed complete non-random inactivation of the wild-type chromosome [20].  In the reverse experiment, a 450kb transgene containing Xist was inserted into autosomes and initiation of long-range silencing was seen [21], moreover when Xist alone was inserted into an autosome it was found to be sufficient for silencing in cis [22].  Since the discovery that XIST/Xist is required for XCI, several groups have attempted to delineate critical domains of XIST/Xist that are required to confer silencing ability.  In a study of multiple DOX inducible Xist transgenes inserted into the X chromosome in male mouse ES cells it was shown that when a 900bp region containing Repeat A is deleted the silencing ability of Xist is abolished [23] and in a similar model, human HT1080 cells containing DOX inducible Repeat A alone have been shown to be able to induce silencing of a nearby GFP reporter [24].     1.4.2 Localization of XIST/Xist Beyond XIST/Xist RNA?s function in initiating silencing of the X chromosome is its unique ability to coat the Xi once it has been transcribed.  Fluorescent in situ hybridization has revealed densely stained foci of XIST/Xist at the Xi in interphase nuclei [25] and it has been proposed that this allows XIST/Xist to act as a link between Xi chromatin and factors involved in silencing but attempts to delineate the sequences involved in XIST/Xist localization have been largely inconclusive.  A study using Locked Nucleic Acid (LNA) inhibition of Xist sequences found Repeat C to be crucial for localization of Xist [26] while in a panel of Xist deletion constructs generated by Wutz et al. [23] the 5? region of Xist, including Repeat A and several regions of the 3? end Xist were required for normal silencing suggesting several elements of Xist may work in concert to 6  achieve localization.  In humans, DOX-inducible XIST constructs in HT1080 cells confirmed that the 5? region of XIST was required for proper localization but no loss of localization was seen when a region containing Repeat C was removed [27].   Several proteins have also been implicated in proper localization of XIST/Xist.  Female cells depleted of the matrix protein hnRNP U show diffuse Xist throughout their nuclei and ES cells lacking hnRNP U fail to initiate X inactivation [28].  Similarly cells that have undergone siRNA knockdown of Yin Yang 1 (YY1), a transcription factor with both activating and repressing abilities (reviewed in [29]),  fail to maintain XIST foci on the Xi.  YY1 appears to be binding XIST RNA and DNA to act as a tether for the spread of XIST localization across the Xi [30].   1.5 Initiation and establishment of XCI Upregulation of XIST/Xist marks the onset of XCI initiation occurring at variable time points in the early embryo in different species.  In mice, XCI initiates in two phases; imprinted XCI occurs at the 4-8 cell stage in which the paternal X chromosome is selected for inactivation but is then reactivated in the inner cell mass of the blastocyst and random X inactivation occurs while imprinted XCI persists in extra-embryonic tissues (reviewed in [31]).  In humans, XIST is expressed at the 8-cell stage and may be starting to accumulate on the X chromosome at the 8-cell stage but there are conflicting reports concerning the XCI status.  One report noted XCI in single cell analysis of pre-implantation embryos based on the monoallelic expression of X-linked genes whereas another report did not observe XCI in complete embryos at the blastocyst stage despite the upregulation of XIST [32, 33].  Notably, humans undergo random XCI and do not appear to have the imprinted form of XCI (reviewed in [34]).        Soon after the upregulation of XIST/Xist several unique chromatin features become associated with the Xi.  Broadly, the Xi forms a densely compact chromatin structure known as heterochromatin while the Xa consists of mostly euchromatin which is less compact.  DNA wraps around proteins called histones to form chromatin which can undergo specific post-translational modifications.  The Xi accumulates a wide variety of histone modifications that are associated with silenced chromatin, such as histone H3 lysine 27 trimethylation (H3K27me3), 7  H3K9me3 and the histone variant macro H2A, and loses modifications that are associated with active chromatin, such as acetylation and H3K4me3.    The Xi also acquires high levels of DNA methylation, the addition of a methylation group onto cytosine nucleotides in the context of CpG dinucleotides, in regions of dense CpG dinucleotides called CpG islands.  The methylation of CpG islands found at promoters is associated with gene silencing therefore the increased DNA methylation of CpG island methylation on the Xi acts as another layer of control of gene silencing [35]  whereas on the Xa, a CpG island approximately 1.5kb into the XIST gene becomes methylated on the Xa to maintain XIST repression [36] and demethylation leads to ectopic XIST expression [37, 38].  Indeed, knockdown of the DNA methyltransferase, DNMT1, leads to aberrant XCI [39] and inhibition of methylation causes reactivation of some genes on the Xi [40].   Importantly, the histone modifications and DNA methylation act in concert to achieve silencing [41] on the Xi and once histone modifications and DNA methylation are in place they are stably transmitted through generations providing an explanation for how the Xi remains inactivated throughout cell divisions.     1.6 Counting and choice Before XIST/Xist can recruit epigenetic changes to the Xi and induce silencing, a female cell must complete the remarkable task of ensuring monoallelic upregulation of XIST while a male cell must fully repress XIST/Xist expression.  The regulation of XIST/Xist expression is believed to be controlled by two main stages; X chromosome counting and choice.      Studies of anueploidies have hinted at X-chromosome counting, an early event in the regulation of XCI resulting in one X chromosome remaining active per diploid set of chromosomes.  Similar to males, females with a 45, XO karyotype show no XCI whereas 47, XXX show inactivation of two chromosomes [42, 43].  In 69, XXX triploid cases, the XCI ratios become more variable with respect to the number of Xis, not only between individuals but also between cells of individuals [44].  The variability in the number of Xis in triploids compared to 8  the consistent XCI of all but one X in cells with diploid genomes has underlined the importance of the autosomes in the counting process. The importance of the XIC/Xic in the counting process is evident by experiments using transgenes of the human XIC inserted on autosomes in male mouse ES cells, called ES-10 cells.  ES-10 cells were capable of initiating XCI on the endogenous X chromosome as well as local silencing on the autosome carrying the transgene pointing to transgene initiated counting which is still possible between species [45].  XCI was only observed after implantation into a blastocyst, not after differentiation into embryoid bodies therefore proper counting and choice between the human and mouse XIC/Xic seems to require a complete developmental environment [45].  Furthermore, XCI was only observed in transgenic clones containing multiple tandem transgene copies [46], a result also seen in mouse transgenic Xic insertions into ES cells [47], which may be a reflection of overall XIST/Xist abundance or differential spreading onto the autosome .   Several molecular models have been proposed to explain how counting and choice may be facilitated.  One model assumes the presence of a limited autosomal blocking factor that is only present in enough copies to block one X chromosome from XCI, therefore inactivating one chromosome in females, blocking XCI on the  X chromosome in males [7] and integrating the counting and choice processes.  The alternative state model suggests that each X chromosome?s likelihood to become the Xi is pre-determined by the chromatin state and may be a result of cohesion differences between sister chromatids and is supported by the evidence that the X chromosomes whose sister chromatids are closest to each other is more likely to become the Xi [48].  Recently, Monkhorst et al. proposed a stochastic model for XCI in which each chromosome has the same probability of inactivation therefore allowing the possibility of cells with two Xas or two Xis which are removed by selection [49].  In this model, counting is performed by a combination of X-linked activators and autosomal repressors and a threshold level of these regulators determines the initiation of XCI while the need for a choice step is removed.     The observation that the two XICs transiently pair before the initiation of XCI, during differentiation, in mouse ES cells [50-52]  also suggested that direct communication between 9  the X chromosomes may be involved in proper counting and choice to allow monoallelic upregulation of XIST/Xist.  Several loci have been implicated in the pairing process including Tsix, Xite (described further in section 1.7.1) and a region within the Xic, upstream of Xist, called the X-pairing region (Xpr) [51, 53].  Deletion of Tsix and Xite disturbed proper pairing of the Xic?s and insertion of an autosomal transgene containing Tsix or Xite resulted in the autosome pairing with the X chromosomes [51].  Pairing of the transgenic autosome with X chromosomes was also found to result in aberrant XCI [54] suggesting that Tsix and Xite are required for pairing and that pairing mediates X chromosome counting.  The involvement of Xpr is more contentious; one report noted the Xpr region could drive interactions between transgenic and endogenous Xic [53] whereas another study was unable to observe this effect  [55].  Masui et al. speculate that pairing may result in asymmetric distribution of either activators or repressors of XIST providing the choice of X chromosome to become the Xi [52].  1.7 XIST/Xist regulators: modulators of counting and choice Accomplishing monoallelic upregulation of XIST/Xist, all within a very specific developmental window unsurprisingly appears to involve a combination of many participating factors that reside both in the XIC/Xic and on autosomes (Figure 1.1).  Much like our knowledge of XIST/Xist, our understanding of factors involved in regulation of Xist comes from studies in mouse models while regulation of human XIST remains largely unstudied.  Key regulators of Xist that modulate the counting and choice processes are outlined below along with partial insights into human regulation.      1.7.1 X-controlling element (Xce) X chromosomes, in inbred mice, have an equal probability of initiating XCI but the probability becomes unequal in heterozygous mice leading to non-random XCI.  The non-random choice observed in heterozygotes is controlled by a locus 3? of XIST  within the Xic known as the X-controlling element (Xce) [56-58].  Three major Xce alleles have been identified; Xcea has the highest probability of inactivating followed by Xceb and then Xcec [59, 60].  The exact location of 10  the Xce and the mechanism causing allele-specific modulation of choice are unknown but studies have identified candidate polymorphic regions 3? of XIST that correlate with the strength of the Xce allele [61, 62].  Recently, Thorvaldsen et al. [63] examined recombinant progeny from crosses of a mouse line containing and Xcea and one containing Xcec and found there to be multiple regions both proximal and distal to Xist that contribute to the Xce.    1.7.2 TSIX/Tsix A landmark in the study of Xist regulation was the discovery, in mice, of another alternatively spliced non-coding gene that is transcribed antisense to Xist, through the Xist promoter, called Tsix [64-66].   Tsix is transcribed on both chromosomes prior to the initiation of XCI but during differentiation Tsix expression is monoallelic and limited to the future Xa [64].  A 3.7kb deletion of the Tsix promoter on one X chromosome caused an increase in XCI on the mutated X chromosome [67].  Transcription was later found to be the functional requirement for Tsix action since an insertion of a stop signal within Tsix resulted in complete non-random upregulation of Xist and XCI on the mutated X chromosome.  Conversely, induction of Tsix transcription during ES cell differentiation resulted in the mutated X chromosome always becoming the Xa [68].  Moreover, termination of Tsix transcription just before Repeat A and the Xist promoter still results in complete non random XCI [69].  This line of evidence implies that complete transcription of Tsix through Xist is required for negative regulation of Xist expression and the choice step of XCI (Figure 1.1).  Interestingly, Tsix deletion in undifferentiated cells does not result in premature activation of Xist and during this time in development other mechanisms are repressing Xist expression [67]. Tsix appears to be regulated by two enhancer elements known as DXPas34 and X-Inactivation Intergenic Transcription (Xite).  DXPas34 is a tandem repeat that lies 750bp downstream of the Tsix promoter [70] which itself shows bidirectional promoter ability and is believed to be both a positive regulator of Tsix, since its deletion results in downregulation of Tsix, and also a negative regulator because the same deletion results in derepression of Tsix 11  after XCI has taken place [71].  Similarly, Xite is also a region with bidirectional promoter ability upstream of the Tsix promoter which, when deleted, causes diminished Tsix transcription and skewed XCI indicating Xite regulation of Tsix in cis [72].  Truncating Xite transcripts does not result in diminished Tsix suggesting that Xite`s regulatory role is independent of transcription [72].    Several mechanisms have been proposed for Tsix driven repression of Xist.  RNAi-mediated repression of Xist through direct binding of Tsix:Xist is a possibility since mutations of Dicer caused upregulation of Xist expression and small RNAs matching the Xist promoter have been observed [73], however a subsequent study has suggested that Xist upregulation in this situation was a result of perturbed regulation of microRNAs that control DNMT3A [74].  Alternatively, Tsix may regulate Xist expression by altering chromatin conformation, inducing DNA methylation and histone modifications, at the Xist locus [69, 75, 76].    While the importance of Tsix in mouse XCI is undisputed, the role of human TSIX is much less clear.  First of all, there is limited sequence homology between human and mouse TSIX/Tsix and greater than 50% of homology lies within XIST/Xist sequence.  Further sequence analysis by Migeon et al. [77] indicated that substantial evolutionary divergence has occurred in the 5? regions of TSIX/Tsix indicative of large genomic rearrangements along with a high influx of repetitive elements into the human TSIX locus.  The transcription of human TSIX also shows remarkable differences from its mouse counterpart.  TSIX is transcribed in antisense orientation to XIST and has been observed in human embryonic stem (hES) cells , ES-10 cells, human placenta and an HT-1080 male somatic line with an XIST containing transgene [77-79].  Notably, TSIX transcription does not appear to be capable of repressing XIST nor does TSIX proceed across the XIST promoter [77], a property crucial for mouse Tsix repression of Xist [68].  Surprisingly, fluorescent in situ hybridization of TSIX transcripts indicate transcription is originating from the Xi rather than the Xa [43].  Taken together, this evidence suggests that TSIX function may not be conserved and that humans may have evolved a different method of repressing XIST during XCI initiation.  12  1.7.3 CTCF and YY1 CCCTC binding factor (CTCF) is a well conserved protein with a wide variety of functions in gene regulation (reviewed in [80]).  In mice, CTCF binds within Tsix and has been proposed to confer enhancer blocking ability and act with Tsix to prevent Xist upregulation [81].  Moreover, CTCF binding is frequently paired with YY1 binding through clustered binding sites for both proteins and these two proteins together may be activating Tsix expression [82].  YY1 is found binding adjacent to CTCF ~1.5kb downstream of the XIST/Xist promoter adjacent to high levels of CTCF binding in humans (Figure 1.1) [29, 83] and YY1 has also been suggested to be crucial for high level XIST transcription [84] which may mean the pairing of these two proteins is significant to both human and mouse XCI in regions  other than Tsix.    A single base pair mutation, C(-43bp)G, in the human promoter was identified in a family showing skewed XCI towards inactivation of the mutant X chromosome [85] whereas a similar mutation, C(-43)A, resulted in preferential inactivation of the wild-type X chromosome [86].  In vitro experiments implicated CTCF as the factor that is affected by these two mutations since the C(-43)A mutation inhibited binding of CTCF whereas C(-43)G enhanced binding of CTCF suggesting that binding of CTCF at the XIST promoter is involved in the choice of XCI and interacts with the Xa [87].  However, genome-wide ChIP-seq data does not indicate CTCF is binding to -43bp of the XIST promoter in female or male cell lines [83] suggesting that in vivo, CTCF may not be the factor responsible for XCI skewing in -43bp mutants.    1.7.4 Pluripotency factors  Since Tsix mutation does not appear to result in ectopic Xist expression in undifferentiated ES cells it is believed that other factors are responsible for Xist repression prior to differentiation.  Pluripotency factors, transcription factors responsible for maintaining pluripotency of ES cells, are logical candidates for Xist repression in the early embryo because they are downregulated during differentiation at the time of XCI and would therefore release repression of Xist.   13  Several pluripotency factors, namely Oct4, Sox2 and Nanog, bind to a region within the first intron of Xist in mouse ES cells and induced repression of Oct4 and deletion of Nanog both result in upregulation of Xist expression [88].  Oct4, Sox2 and Nanog also exert a repressive effect on Rnf12 [89].  Reporter assays have indicated that the Xist promoter alone can become activated during female ES cell differentiation [90] and a recent study by Minkovsky et al. [91] deleted the pluripotency binding site within intron 1 and saw no effect on XCI in male or female ES cells pointing to indirect regulation of Xist by pluripotency factors.    1.7.5 Non-coding RNAs: JPX/Jpx, FTX/Ftx, RepA, XACT In addition to widely studied XIST/Xist and Tsix several other non-coding RNAs that lie within the XIC/Xic including JPX/Jpx, FTX/Ftx and RepA and one non-coding RNA telomeric to the XIC, called XACT, in humans may be implicated in regulation of XCI (Figure 1.1).    JPX/Jpx is believed to be expressed from both the Xi and Xa in both humans and mice [92-94].  In mice, deletion of Jpx results in inhibition of XCI and is female lethal, a phenotype that is rescued by trans-addition of Jpx [93] but overexpression of Jpx in male ES cells, however, does not result in initiation of XCI [95].  Similarly, FTX/Ftx escapes XCI and is upregulated during mouse ES cell differentiation.  Ftx transcription is believed to alter the chromatin landscape of the Xic and affect the transcription levels of genes within the Xic, more specifically, Ftx null cells showed a decrease in transcription of Xist, possibly through methylation of the Xist promoter [96].  RepA is another non coding RNA within the 5? region of the Xist gene itself that has also been proposed to activate Xist.  RepA is found in both male and female ES cells and upregulates slightly in only females to bind Polycomb Repressive Complex 2 (PRC2) and aid in the upregulation of Xist [97].   From this evidence it has been proposed that Jpx, Ftx and RepA are all non-coding RNA activators of Xist.   An interesting recent study has proposed a human specific long non-coding RNA called X Active Coating Transcript (XACT) that is expressed exclusively in pluripotent cells and uniquely coats the active X chromosome(s) [98].  The function of XACT remains unknown but the RNAs ability to associate with Xa chromatin may be due to involvement in the early XCI program.   14   1.7.6 RNF12/Rnf12 Jonkers et al. screened several BAC transgenes across the Xic to look for X-linked activators of Xist and found that a region containing a gene called Rnf12 was capable of increasing the percentage of male ES cells that initiate XCI and female cells that initiate XCI on both X chromosomes [95].  Furthermore, heterozygous deletion of Rnf12 results in decreased levels of XCI [95] and homozygous deletion of Rnf12 abolished XCI completely implicating Rnf12 as a dosage-dependent X-linked activator of XCI that is critical in X chromosome counting [99] (Figure 1.1).   Rnf12 is an E3 ubiquitin ligase and has been found to polyubiquitinate Rex1, a transcription factor that correlates strongly with pluripotency, targeting Rex1 for degradation [100-102]; during differentiation Rnf12 is upregulated leading to more rapid degradation of Rex1.  Male ES cells homozygous for a Rex1 deletion showed ectopic XCI whereas female ES cells overexpressing Rex1 showed inhibition of XCI linking the downstream effects of Rnf12 and Rex1 function to XCI regulation.  Indeed, further analysis revealed that Rex1 directly binds to regulatory elements of both Xist and Tsix and that transient overexpression of Rex1 causes downregulation of Xist suggesting that Rex1 may repress Xist directly [101].  Initiation of XCI in mouse ES cells is also seen with overexpression of human RNF12 indicating that the function of Rnf12 in XCI may be conserved in humans but the action of human RNF12 in an endogenous system remains to be seen.  Interestingly, the cell line containing a rearranged X, used to map the distal boundary of the XIC, is missing a copy of RNF12 but is still capable of initiating XCI [5] despite the dosage sensitivity observed for Rnf12 action in mouse.     1.8 Developmental models for human XCI and XIST regulation The usage of mouse ES cells has clearly been invaluable to the study of mechanisms of XCI.  Unfortunately, the clear differences between human and mouse XCI may not make mice the 15  best model for humans but the available models for studying the early events of human XCI bring a number of limitations.   Human ES (hES) cells appear to be from a different developmental time point than their mouse counterparts, showing many similarities to mouse epiblast stem cells (EpiS cells) in morphology, growth factors utilized, and overall gene expression  [103].  The XCI status has been characterized for a number of female hES cell lines showing variable patterns of XCI and XIST expression that broadly fit into three categories: Class I hES cells show low levels of XIST expression which upregulates upon differentiation, akin to mouse ES cells, whereas lines classified as Class II have upregulated XIST and undergone XCI and Class III have undergone XCI but have lost expression of XIST [104-106].  To add to the complexity of XCI in hES cells, the class of any given hES cell may change depending on the time in culture and the growth conditions [105, 107].   Due to the anomalous XCI patterns in hES cells, reprogramming of Class II and Class III to a ?na?ve? epigenetic state has been explored by altering culture conditions.  Deriving hES cells in physiological oxygen concentration (5%) rather than ambient oxygen concentration (20%) results in hES cells that are class I, retaining two Xa?s and upregulate XIST upon differentiation [107].  The addition of a small molecule cocktail that reinforces the use of specific signaling pathways and transgenes that ectopically express pluripotency factors helps to maintain na?ve ES cells derived from, epigenetically unstable, non-obese diabetic mice and rats [108-110].  When these culture modifications are adapted to Class II hES cells, the cells revert to Class I after 10 days in culture and show gene expression and growth factor profiles similar to mouse ES cells [111].  Histone deaceylase inhibitors have also been shown to repress XIST expression and revert the XCI status in Class II hES cells thus converting them to a na?ve, Class I state [112].  Na?ve hES cells represent the ideal model for studying the early events in human XCI but unfortunately the reprogramming processes outlined make them difficult to generate and maintain in many labs.  Non-na?ve hES cells are often the most feasible option for many human developmental studies and despite the anomalies of XCI they still represent an earlier developmental time point and may therefore show unique regulatory elements that could be controlling expression of XIST which would not be found in somatic cells.   16   1.9 Genomic approaches to identifying regulatory elements in the XIC Regulatory elements within the XIC that may be controlling XIST expression can be identified using several approaches in combination.  Using the DNA sequence alone, regions of the genome that are highly conserved between species but do not fall within any known genes are likely to have regulatory function.  Chureau et al. [8] used this approach to compare the XIC?s of human, mouse and cow and uncovered novel conserved genes and CpG islands as well as conserved pseudogenes which may be acting to control expression of neighbouring genes.  The poor conservation of the XIC/Xic, however, limits the value of this technique.  Sequence analysis is also capable of locating clusters of transcription factors based on their predicted binding sequence which is indicative of the presence of a regulatory element.    Experimentally, the use of a technique known as DNase I hypersensitivity (DHS) has been invaluable to the discovery of regulatory elements in the genome due to their sensitivity to digestion by DNase I.  DHS sites are present as a result of the open chromatin structure at regulatory elements that is needed to allow the binding of transcription factors and other DNA binding proteins.  Within the XIC, DHS mapping has revealed three putative regulatory elements, one downstream of XIST  two upstream of XIST, that do not appear to be conserved between humans and mice [113].  A high throughput genome sequencing approach to DHS mapping, called DNase-seq, is now commonly used for studies of regulatory elements and allows for easy screening of the XIC for potential candidates [114].  Perhaps the biggest addition to our ability to identify regulatory elements has been our increasing understanding of the association of specific histone modifications with different types of regulatory elements.  ChIP combined with microarray analysis (ChIP-chip) or whole genome sequencing (ChIP-Seq) can easily identify regions of the genome enriched for various histone modifications.  The chromatin marks associated with specific regions corroborated with ChIP-seq data for transcription factors can now allow for the prediction of regulatory elements as promoters, enhancers, silencers or insulators.  Acetylation of H3 and H4 marks both promoters and enhancers, strong enrichment of H3K4me3 is found at promoters and H3K4me1 is found at enhancers and insulators [115].  In addition to the histone modifications, general 17  transcription factor enrichment can be found at promoters and enhancers, pre-initiation complex components are enriched at promoters [116] and the co-activator p300 is known to localize to enhancers [117] increasing the predictive strength for the type of regulatory element.  CTCF binding is considered a hallmark of insulator function [80, 118] but CTCF also has a wide variety of other functions throughout the genome and therefore binding is not exclusive to insulators [119].    Together, sequence analysis and experimental approaches such as DHS and genome-wide ChIP-seq for modified histones and transcription factors within the XIC can help to pick candidate regulatory elements for XIST expression and predict regulatory element function.     1.10 Thesis objective  Essentially all of our understanding of XIST regulation during counting, choice and initiation of XCI comes from studies in mice but poor conservation of the XIC/Xic and the Xist repressor, Tsix points towards differences between mouse and humans in regulation of XIST.   To help uncover human specific mechanisms, the objective of this thesis was to further refine the boundaries of the XIC and examine candidate cis-regulatory elements within the XIC.  We show evidence for an Xi specific DNase I hypersensitivity site, and an alternative promoter and alternative splicing in the 5? end of XIST and beyond the 3? end of XIST, show evidence against TSIX expression in placenta but find TSIX-like antisense transcription in human male ES cells.          HumanMouseSLC16A2/XPCT50kbRNF12CNBP2FTXJPXTSIXXISTCHIC1CDX4NAP1L2Centromere TelomereRnf12Xpct/XprCnbp2FtxJpxXistTsixTsxChic1Nap1l250kbCentromere TelomereHumanCdx4X-inactivation centreMouseXiteDxpas34TsixOct4Sox2NanogRex1Yy1 CtcfYY1 CTCFYy1 Ctcf5kbP1P2RepAL1.10.1 TSIX TSS ES-10 TSIX TSSFigure 1.1 - Regulation of XCI  and XIST/Xist by the XIC/Xic Schematic representation of the human XIC on the top.  Above the gene map is the current location of the XIC where the span containing the XIC boundaries are shown in grey.  The lower gene map shows the mouse Xic where green arrows represent activators of Xist and red blocked arrows represent repressors of Xist.  Proteins are indicated as ovals.   The grey box contains a magnied image of the XIST/Xist locus and orientation is reversed in relation to the genomic context to show XIST/Xist transcribing left to right.  Black boxes in the schematic represent exons except for human Tsix which has not been surveyed for alternative splicing so is represented by one large transcriptional block.  Red, Pink, blue, green, yellow and brown boxes indicate Repeats A, F, B, C, D, E, respectively.  18XIST Xist 19           2 Materials and Methods              20  2.1 Tissue culture and cell lines Mouse-human somatic cell hybrid cell lines, t75-2maz 34-1a (containing a human Xi) and t60-12 (containing a human active Xa) [120] were cultured at 37?C in alpha Minimum Essential Medium (MEM) supplemented with 7.5% fetal calf serum (PAA Laboratories Inc), 1% penicillin/streptomycin (Life Technologies) and 1% L-glutamine (Life Technologies).  The GM11200 male lymphoblast cell line and GM11201 and GM7350 female lymphoblast cells lines (Coriell cell repository) were maintained in Roswell Park Memorial Institute (RPMI) medium supplemented with 15% fetal calf serum (PAA Laboratories Inc), 1% penicillin/streptomycin (Life Technologies) and 1% L-glutamine (Life Technologies) at 37?C.   The HT1080 transgenic cell line, L1.10.1, and the HT1080 cell line with containing a DOX-inducible XIST were cultured in Dulbecco Modified Eagle Medium (DMEM) supplemented with 10% fetal calf serum (PAA Laboratories Inc),, 1% penicillin/streptomycin (Life Technologies) and 1% L-glutamine (Life Technologies) at 37?C.   HEK293 cells were cultured in DMEM, supplemented with 10% fetal calf serum (PAA Laboratories Inc), 1% penicillin/streptomycin and 1% L-glutamine (all from Invitrogen) at 37?C.  All above cell lines were passaged using 0.25% Trypsin EDTA. CA1S cells were cultured as described [121] on Matrigel (BD Biosciences) coated 6-well plates in mTeSR1 basal medium (STEMCELL) supplemented with mTeSR1 5x supplement (STEMCELL) and passaged using Accutase (STEMCELL).  2.2 PCR and quantitative PCR PCR was performed with 100 ng of genomic DNA template, 1U Taq polymerase, 0.2 mM dNTPs, 1.5 mM MgCl2, 1X PCR buffer (all from Invitrogen) and 0.5 ?M of both forward and reverse primers.  PCR was performed using 30-40 cycles of [95? for 30 s, 55?-60? for 30 s, 72? for 1 min].  Quantitative PCR (qPCR) was performed using the StepOnePlusTM Real-Time PCR System (Applied Biosystems) and the qPCR reaction mix was composed of 0.2 mM dNTP mix, 2.5 mM MgCl2, 1X HS reaction buffer, 1X EvaGreen dye (Biotum), 0.25 ?M forward and reverse primer, and 0.8 U Maxima Hot Start Taq (Fermentas) and cycling conditions were as follows: 95? for 5 21  min, followed by 40 cycles of [95? for 15 s, 60? for 30 s, 72? for 1 min].  Each sample and negative control was assayed in triplicate.    2.3 RNA extraction and reverse transcription RNA was extracted using Trizol (Invitrogen) following the manufacturer?s instructions.  RNA was treated using the DNA-free kit (Ambion) to remove genomic DNA contamination according to the manufacturer?s instructions and RNA concentrations were determined using a spectrophotometer.  Reverse transcription (RT) of RNA was carried out using 2?g of RNA, 1x first strand buffer (Invitrogen), 0.01 mM Dithiothreitol (DTT) (Invitrogen), 0.125mM dNTPs, 1 ?L random hexamers, 1?l RNAse Inhibitor (Fermentas) and 1 ?L (1U) of Moloney Murine Leukemia Virus reverse transcriptase (M-MLV) and water was added to a total volume of 20 ?L.  Reactions without RT were also carried to out to test for complete removal of genomic DNA contamination.  The reaction was incubated for 1 h at 42?C and heat inactivated by incubating at 95?C for 5 min.   For strand specific RT reactions 2 ?g of RNA was mixed with 0.50 ?M dNTPs and 2 pmol of sense or antisense gene specific primer with a T7 sequence tag on the 5? end of the primer and then heated to 70?C for 5 min.  Following this incubation the tubes were placed on ice for 1 min and then mixed with 1x first strand buffer (Invitrogen), 0.005 DTT, 1 ?L (1U) RNase Inhibitor (Fermentas) and 1 ?L (1U) Superscript III.  Reactions minus RT were also carried to out to test for complete removal of genomic DNA contamination.  The reaction was incubated for 1 h at 55?C and heat inactivated by incubating at 95?C for 5 min.    2.4 DNase I hypersensitivity 2,000,000 cells were harvested and washed twice in ice cold PBS.  Cells were then lysed using 0.1% NP40 in resuspension buffer (RSB) (10mM Tris pH 7.4, 10mM NaCl, 3mM MgCl2) and spun at 1500 rpm for 10 min to pellet nuclei.  Nuclei were resuspended in 720 ?L RSB and digested with different concentrations of DNase I.  Mouse human hybrid cell lines, t75 2maz 34-1a and t60-12 and human lymphoblast cells, GM11201, and GM11200 and HT1080 cells were digested 22  with 0U, 20U, 40U and 60U of DNase I at 37?C for 10 min.  Digestion was stopped and DNA was extracted using 0.8 mL of DNAzol (Invitrogen) followed by ethanol precipitation.  The DNA was diluted to a final concentration of 20 ng/?l to be used in qPCR.  Primers for qPCR were designed to span the test hypersensitive site (200b.1 and 200a.1) as well as a positive control region (JPX) and an insensitive region (XIST3?5?).  Hypersensitivity was calculated by normalizing each DNase I concentration to the insensitive region and then each DNase I concentration was plotted as a fold difference from the untreated sample.    2.5 5? and 3? rapid amplification of cDNA ends 5? RACE was performed using the First Choice RLM-RACE kit (Ambion) as per the manufacturer?s instructions.  5? RACE cDNA was amplified using nested PCR reactions (components as above) with 35 cycles each and annealing temperatures of 57?C, for the outer PCR reaction, and 57?C for the inner PCR reaction.  3? RACE was performed using the First Choice RLM-RACE kit (Ambion) as per the manufacturer?s instructions.  PCR products from both 5? and 3? RACE were analyzed using 2% gel electrophoresis and PCR products purified with the QIAquickGel extraction kit (Qiagen) for Sanger sequencing.    2.6 siRNA-mediated knockdown Knockdown was performed according to the manufacturer?s instructions.  Briefly, 100 000 cells were seeded into each well of a 24 well tissue culture plate.  After 24 hrs the cells were transfected with 2 ?L of Dharmafect 4 transfection reagent (Themo Scientific) and 0.05 ?M of siGenome SMARTpool siRNA and harvested after a further 72 hrs.    2.7 Statistical analysis  Statistical analysis was performed using GraphPad Prism 5.02.  One-way ANOVA was used to test for significance in DHS experiments between different concentrations of DNase I.   23  Table 2.1: List of Primers (Shown 5? to 3?) Name  Sequence 1F CCTCACAAGCAACAGAACGA 1R CGAGCCTTGGTTTACAGCTC 2F TGCATTTTCGACTGAAGCAC 2R CACGGTTTCCCTTGGTTAGA 3F AATCCTGTGGGCCTGTAGTG 3R TCATTTAGGAGCCAGCGACT 4F CTGCCAAGGGCTAGTGAGAC 4R ACTCCTACTTGGGGGCCTTA 5F CCCACAAGTAAGCCCTGGTA 5R ACAATGTTGATGGTGCTGGA 6F AGCTAAAATGAGGGCTGCAA 6R TAGGGAGCCTTGATGATTGG 7F AAGCGTCTCTGGGTGAGAAA 7R CGCCAAGCTGAGAGATAACC 8F ACTCCAAAAGAGGGGAAGGA 8R CATGACACCACACCTGGAAG 9F GGGCTGCTAGAGAACACCAG 9R ATGCCCTGTGGTAAGTCCTG 10F CACCATTATTTCCCCACAGG 10R TTCAGGTTCGTAGCCAGCTT   24  Name  Sequence  200b.1F  TGTCCATCCCACCTTTTCTC  200b.1R TCTTTGCTGTGTGCTTTTCG 200a.1F ATCCGACCCCAGCATTAGC 200a.1R GCTCCAGGCCTGCTTGGT JPX F GCGGAGGCATTTAGGTAGTG JPX R GGCGAGTTTCTGGACTTTTG XIST 5? GAAGTCTCAAGGCTTGAGTTAGAAG XIST 3? TTGGGTCCTCTATCCATCTAGGTAG P1-1F TGTCAACCAAAAATGATTCCA P1-1R TCTCTGCACTTGGGGTTCTT 19F AACTGATCCACAAAAAACAGAGATGT 19R TCTTCTTGACACGTCCTCCATATTT 200DF AGAGGACACCAGACCACAGC  200DR TGTGCTGGTCATTTTCTTTGA  200D.2F GGATTCTCCAGAAGCACAGC 200D.2R AGCACTCTGAACCCCATTTG qXIST5F CCTAGTTCAGGCCTGCTTTTCAT  qXIST5R TCAGCCCATCAGTCCAAGATC  YY1F ACCTGGCATTGACCTCTCAGA YY1R TTTTTCTTGGCTTCATTCTAGCAA 25  Name Sequence Exon3F TGTTTGCAGTCCTCAGGTCTCA  5?AF TGACTTCCTCTGCCTGACC 5?AR GATTCCCTTCCCCTCTGAAC  XE1BF AGTGCCAAATGCCAGGATAC  XE1BR AATGCTGGTAAAGCCCACAC  TSIX7 CAGTACCAGCATTCTCAGTG  TSIX8 CCACTCTCATTGTCATTGCG TSIX11 CCAGCTGCAACTCAGATGTA TSIX12 CCTTCTTCTCAGAGACTCCT TSIX13 CTGATAAGTGACCAGTCACC TSIX14 TGAAGACACTGGCCTTGACA TSIX15 TGGCACACGTATGTGGTTCT TSIX16 CTCTGAGTCTTCCTATGACC TSIX5 TTGGGGATGGAGAATAGGTG TSIX6 CCTGATCTGAGTTATGGCAC T7.Intron 4F TAATACGACTCACTATAGGGAGA Intron4R TAATACGACTCACTATAGGGAGACTGTCATCAGGCAGGAGCTA  T7.Intron 4R TAATACGACTCACTATAGGGAGACTGTCATCAGGCAGGAGCTA  Exon5F TAGAGTGCCAGGCATGTTGA   26  Name Sequence T7.Exon5F TAATACGACTCACTATAGGGAGATAGAGTGCCAGGCATGTTGA  Exon5R ACAAGCAGTGCAGAGAGCTG T7.Exon5R TAATACGACTCACTATAGGGAGAACAAGCAGTGCAGAGAGCTG  T7.XIST5? TAATACGACTCACTATAGGGAGATTGGGTCCTCTATCCATCTAGGTAG T7.XIST3? TAATACGACTCACTATAGGGAGAGAAGTCTCAAGGCTTGAGTTAGAAG  Exon7F CCTTGTAAATGCACTTCAAAACC T7.Exon7F TAATACGACTCACTATAGGGAGACCTTGTAAATGCACTTCAAAACC Exon7R AGGAGGGATGATGACCAACT T7.Exon7R TAATACGACTCACTATAGGGAGAAGGAGGGATGATGACCAACT Exon8F CCAACTCCCCAGTTTGTTTC T7.Exon8F TAATACGACTCACTATAGGGAGACCAACTCCCCAGTTTGTTTC Exon8R TGAGTCTTTGCTGTTTGGAAGA T7.Exon8R TAATACGACTCACTATAGGGAGATGAGTCTTTGCTGTTTGGAAGA T7.TSIX12 TAATACGACTCACTATAGGGAGACCAGCTGCAACTCAGATGTA T7.TSIX11 TAATACGACTCACTATAGGGAGACCTTCTTCTCAGAGACTCCT P1 outer TGTCAACCAAAAATGATTCCA P1 inner TGGAGGACGTGTCAAGAAGA P2a outer ATCTGAACACGCCCTTAGCTTAA P2a inner TGACTTCCTCTGCCTGACCT P2b outer AACACTGCGACAGAACTGGA 27  Name Sequence P2b inner CTGCCTCCCGATACAACAAT P2c outer CAATTCCACCCCCATTTCTA P2c inner TGTCCATCCCACCTTTTCTC P2d outer AGAGGACACCAGACCACAGC P2d inner ATCCGACCCCAGCATTAGC P2asa outer TGCACTCTCTGGAATATCTACACTTTTT P2asa inner TGCTGATCATTTGGTGGTGTGT P2asb outer TGCTGATCATTTGGTGGTGTGT P2asb inner GATAGCAGGTCAGGCAGAGG P2asc outer GATAGCAGGTCAGGCAGAGG P2asc inner TTGATTTGGGGCTTGTTAGG P2asd outer TCTTTGCTGTGTGCTTTTCG P2asd inner GTGCTTTTCGTGTTGGGTTT JPX control outer GGCGAGTTTCTGGACTTTTG JPX control inner AGTTAGGCGATCAGCGAGAA 5? RACE outer GCTGATGGCGATGAATGAACACTG  5? RACE inner CGCGGATCCGAACACTGCGTTTGCTGGCTTTGATG    28            3 Results             29  3.1 Refining the boundaries of the XIC The XIC was characterized using X:autosome translocations[3] and later refined to 1 MB of DNA at Xq13 [4, 5].  The boundaries of the XIC however were mapped before the completion of the human genome project when probes were more difficult to generate so they span a region of approximately 260kb both distally and proximally.  We aimed to further refine the boundaries of the XIC so that candidate regulatory elements within the XIC can be more precisely chosen.  To map the distal end of the XIC we used the somatic cell hybrid cell lines tAG-1Baz1b, which contains an isodicentric X chromosome, and tSA70-D1-34az1f which contains a rearranged X chromosome.  The X chromosome in both of cell lines have a distal breakpoint at Xq13 and showed late replication indicating they are capable of XCI [4-6].  To map the proximal boundary we used a somatic cell hybrid cell line which contained an X:14 translocation, called t4-1a-az1.  The portion of the X chromosome in t4-1A lies proximal to the XIC breakpoint but the reciprocal translocation containing X chromosome material distally, is capable of XCI [4].  We refined the XIC boundaries in these somatic cell hybrid lines using PCR with primers spanning the previously defined boundaries (Figure 3.1A).   Primer amplification was lost between primers 4 and 5 in t4-1a-az1 cells, narrowing the proximal boundary of the XIC to a region of 24, 381bp.  In tSA70-D1-34az1f cells, amplification was lost between primers 8 and 9 which narrows the distal boundary of the XIC to 62, 332bp (Figure 3.1A,B).  The new proximal and distal boundaries put the XIC genomic distance at a maximum of 910,426bp.     3.2 Xi-specific DNase I hypersensitivity site at the 5? end of XIST We next looked within the newly refined X-inactivation centre for candidate regulatory elements that may be involved in regulating XIST expression.  By surveying genome-wide DHS-seq data within the XIC we located one strong DHS site as a candidate for XIST regulation that was located ~1.5kb within the exon 1 of XIST (Figure 3.2A).  We carried out a DNase I digestion combined with qPCR at two regions, DHS 200b.1 and DHS 200a.1 and found that DHS 200b.1 was hypersensitive to digestion by increasing concentrations of DNase I in female lymphoblast cells but not in male lymphoblast cells while DHS 200a.1 was not hypersensitive in female or male lymphoblast cells (Figure 3.2B).  The difference between male and female hypersensitivity  Figure 3.1 - Rening the boundaries of the XIC. A) Schematic of XIC indicating primer positions (*) used in PCR .  B) Representative PCR in cell lines containing rearrangements of the XIC to determine the smallest region capable of XCI.  t4-1a az1tAG -1Baz1bMouse DNAtSA70-D1-34az1f Human DNA1   2    3   4   5 6   7   8   9   10NAP1L20k50kbSLC16A2/XPCTCNBP2FTXJPXTSIXXISTCHIC1CDX4Centromere TelomeretSA70-D1-34az1f t4-1a az1tAG-1Baz1b* * ** *** * **1    2      3       4 5 6    7     ABX-inactivation centre8  9  10  RNF12300246810JPX 200a.1 200b.10246810JPX 200a.1 200b.10510152025JPX 200a.1 200b.102468101214JPX 200a.1 200b.102468101214JPX 200a.1 200b.102468101214JPX 200a.1 200b.10510152025JPX 200a.1 200b.1Female Male Xi HybridL1.10.1 HT1080No DOXHT108010D DOXXa Hybrid* ** * ** * *** ** **n.s.n.s.n.s.n.s.n.s.n.s.*n.s.n.s. n.s.n.s.n.s.500bp* **JPX 200b.1 200a.1Figure 3.2 - Xi-specic DNase I hypersensitive site ~1.5kb within XIST. A) Schematic indicating DHS sites from ENCODE genome wide survey of the XIC and locations of primers used for qPCR-based DHS assay. B) qPCR-based DHS assay on female and male lymphoblasts, Xi and Xa mouse-human hybrids and male brosarcoma cells containing an XIC integration (L1.10.1) and a DOX-inducible XIST.  Sensitivity of biological triplicates ? standard deviation is shown relative to an unsenstitive region.  *p=0.05-0.01, **p=0.01-0.001, ***p=0.001, n.s.=not signicant Relative sensitivityRelative sensitivityDNase IDNase IX-inactivation centreAB3132  at DHS 200b.1 hinted at an Xi to Xa hypersensitivity difference so we used Xi and Xa mouse-human hybrid cells to assess the chromatin of the Xa and Xi separately.  We saw a >10 fold increase in sensitivity at DHS 200b.1 in an Xi containing mouse-human hybrid cell line and a modest DHS 200a.1 while no hypersensitivity was seen at either region in an Xa hybrid cell line indicating that the DHS site was specific to Xi.  The difference in sensitivity at region DHS 200a.1 between female lymphoblast cells and Xi mouse human hybrid cells is likely a reflection of the Xa in female cells masking weak sensitivity at DHS 200a.1.  Interestingly, HT-1080 cells transfected with a 450kb, XIST containing, portion of the XIC and HT1080 cells with a DOX inducible XIST cDNA clone, before or after DOX induction showed no hypersensitivity at either DHS 200b.1 and DHS 200a.1 (Figure 3.2B) indicating that a full genomic context or progression through normal development may be required to confer hypersensitivity.      3.3 DNase I hypersensitivity at DHS 200b.1 is associated with an actively transcribing promoter  By analysing genome-wide ChIP-seq data accessible on the UCSC genome browser, we identified several promoter associated proteins in female cell lines that overlap the DHS site such as RNA polII, H3K4me3, H3K27ac, H3K9ac and over a dozen transcription factors including YY1, a protein implicated in regulation of XCI in mouse [82] but H3K4me1, a histone modification considered to be highly associated with active enhancers [115], shows very low levels across the 5? end of XIST (Figure 3.3).  Moreover, studies in mouse have suggested the presence of an alternative promoter approximately 1.5kb within the Xist gene [15].  The combination of proteins binding to the DHS site and the precedence for alternative promoter usage in other species implied that the DHS site may be an active promoter rather than an enhancer.   To address the question of promoter activity at the DHS site we performed 5? Rapid Amplification of cDNA Ends (RACE) assays on a region overlapping the DHS site and CpG island in XIST. (Figure 3.4A).  This analysis revealed three transcription start sites in the sense orientation, which we designated P2, spanning a region of 404bp (Figure 3.4A, B) in female   XISTDHS 200b.1DHS 200a.1CpG islandPolIIH3K9acH3K27acH3K4me1H3K36me3H2A.ZYY1CTCFH3K4me3MaleFemalePolIIH3K9acH3K27acH3K4me1H3K36me3H2A.ZCTCFH3K4me3Figure 3.3- DHS 200b.1 show enrichment of promoter associated proteins.   Genome-wide ChIP-seq data for promoter and enhancer associated histone modications, YY1 and CTCF.  Peaks are shown for female (GM12878 shown) in red and male (HepG2 shown) in pink YY133F emal eMal e5?R A CE pr imersqR T -PCR pr imers*** * *500bpXIST exon 1P1 P2a P2b P2c P2d19 200a.1 200D 200D.2 qXIST5C pG islandAB C     Relative RN A  levels       (Relative to 1 9 )exon 1P2a P2b P2c P2dP1+ ControlP2bP2c (1)P2c (2)2000RNA-seq00.511.522.53P1-1 19 200a.1 200D 200d.2 qXIST500.511.522.53P1-1 19 200a.1 200D 200d.2 qXIST5    n.d.DHS 200a.lDHS 200b.lFigure 3.4 - Xi specic DHS site corresponds to an actively transcribing P2 promoter A) Schematic of XIST exon 1 indicating the locations of primers used for 5?RACE and qRT-PCR analysis, RNA-seq data for the region (outlined box) and the transcription start sites found by 5? RACE. B) 5?RACE spanning ~1.5kb at DHS 200b.1 and DHS 200a.1.  +control is JPX 5? end.  Sequenced products using P2c products are indicated by white arrows C) qRT-PCR upstream and downstream of P2 shown as biological triplicates ? standar d deviation. n.d. = no data  *P1-1Female lymphoblast HT1080 5D Dox34Repeat AP2c(1)P2c(2)35  lymphoblasts but not male lymphoblasts.  While multiple transcription start sites are often found at CpG island promoters it is probable that the dominant transcription start sites are P2c (1) and P2c (2) because of their high correlation with the ChIP-seq peak in PolII binding and their location within a valley of H3K4me3 peaks which is a common feature of transcription start sites due to a nucleosome free region surrounded by well positioned H3K4me3 modified nucleosomes [115, 122-124]. P2b on the other hand, lies outside of the region of these ChIP-seq peaks and the CpG island.          Since Repeat A lies between P1 and P2, the functional significance of P2 transcripts is unclear.  We therefore addressed the abundance of P2 transcripts to determine the quantity of XIST that is lacking the silencing capabilities of Repeat A.  We used qPCR with primers upstream and downstream of P2 and found this region to be very sensitive to differences in RT temperature indicating that secondary structure may be affecting RT efficiency [16].  Using a high temperature RT we found there to be increases and decreases in overall transcript abundance in female lymphoblast both upstream and downstream of P2 when genomic DNA was used as a standard curve (Figure 3.4C).  We believe the variability to be a result of differences in RT efficiency at different loci because male HT1080 cells containing a DOX inducible XIST transgene, which are likely to have consistent transcription across XIST, and female RNA-seq data [125] show a similar trend in transcript abundance.  Despite the limitation of RT efficiency, the fact that there is not a consistent increase in transcript abundance downstream of P2 in female lymphoblasts suggests that P2 is not grossly expressed in relation to P1.    3.4 Regulatory features of the P2 promoter Elsewhere in the genome, tandem alternative promoters arranged like P1 and P2 are commonly transcribed individually because of cellular and physiological signals and frequently the upstream promoter regulates silencing of the downstream promoter by transcriptional interference [126].  Clearly, both P1 and P2 are transcribed from female somatic cells, whether simultaneously or in different cells (Figure 3.4B), so we believed that there must be a regulatory  P1-1 200b.1 200a.1 XIST 3?5?G M01 7 3 0500bpXIST exon 1C pG island**19 200a.1*200b.1 Xi=XpBAXISTXIST1kb18s EXON 8EXON 8 18s3?RACE primersFemale lymphoblast Xi=XmFigure 3.5 - Protection of P2 from transcriptional interference is not due to separate transcripts or imprinted promoters A) 3? RACE between P1 and P2 promoters suggests there are no polyadenylated 3? ends. B) RNA-seq data from paternal Xi skewed GM12878 cells and RT-PCR analysis upstream and downstream of P2 in maternal Xi containing GM01730 cells indicating P1 and P2 are not exclusively transcribed specic to parent-of-origin 3637  feature of P2 that prevents transcriptional interference.  We first asked if XIST was actually composed of two transcripts, one initiating from P1 and terminating before P2 and another initiating from P2 but 3? RACE assays between P1 and P2 did not identify any polyadenylated transcripts terminating upstream of P2 (Figure 3.5A).   Mouse Xist has been implicated as containing an imprinting control region that acts during imprinted XCI and Kcnq1ot1 provides an example for imprinted tandem alternative promoters so we hypothesized that P1 and P2 may be differentially expressed depending on the parent-of-origin of the Xi.  We investigated two cell lines; one contained an X:A translocation leading to complete inactivation of the maternal X chromosome and the other cell line shows >95% skewing towards inactivation of the paternal X chromosome [127] and if P2 was imprinted we expected to see P1 and P2 transcription separately.  An RNA-seq track from the UCSC genome browser in GM12878 [125] and RT-PCR analysis in GM01730 indicated that transcription in these lines is occurring from P1 and therefore exclusive parent-of-origin expression is not occurring (Figure 3.5B). Since antisense regulation by Tsix is critical to proper mouse XCI and could prevent transcriptional interference on P2, we next asked whether P2 was capable of bidirectional transcription to generate transcripts antisense to XIST.  5? RACE revealed a transcript expressed in the antisense orientation ~350bp upstream of P2 and 7bp upstream of the CpG island, in female lymphoblast cells which we designated P2as (Figure 3.6A,B).  Several amplicons were found in 5? RACE reactions on male lymphoblast but sequence analysis showed that none of these products aligned to the X chromosome.  We performed 3?RACE using a poly(T) primer to locate the 3? end of P2as and found that transcription extends into Repeat A but the poly(T) primer appears to be priming from an encoded poly(A) rather than a poly(A) tail.  P2as is, therefore, extending at least into Repeat A.  Strand specific qRT-PCR of P2as relative to sense expression of XIST revealed that P2as is at extremely low abundance, ~1% of XIST levels, which roughly equates to 2 molecules per cell based on the previously described 2000 molecules of XIST [128].   5?RACE primers500bpXIST exon 1P2asaC pG islandAB CF emal eMal e00.0010.002Female Male00.511.5P2asb P2asc P2asdD F emal e Mal eP2asaP2asbP2ascP2asd~~~Strand specic qRT-PCR primers*P2as2Figure 3.6 - Antisense transcription from P2 promoter. A) Schematic of XIST exon 1 indicating the locations of primers used for 5?  and 3? RACE and strand specic qRT-PCR analysis and the antisensetranscription start sites found by 5?RACE.  B) 5?RACE spanning the sense P2 promoter in female and male lymphoblasts. ~ indicates amplicon did not align to the X chromosome C)  Strand specic qRT-PCR indicating low level transcription of P2as. D) 3?RACE of P2as.Sense P2as2Antisense P2as23?RACE primers5?A           RNA levels (relative to sense XIST)  3839  3.5 siRNA-mediated knockdown of YY1 diminishes XIST expression but not exclusively at P2 Since YY1 is so highly enriched at P2 we hypothesized that with siRNA-mediated knockdown of YY1 we may see downregulation of P2 transcription and may therefore be able to address P2 function.  We achieved >80% YY1 knockdown efficiency (Figure 3.7A) and using qRT-PCR saw a 50% drop in XIST transcription both upstream and downstream of P2 indicating that YY1 is not exclusively regulating expression of the P2 promoter (Figure 3.7B).   3.6 DNase I hypersensitivity at DHS 200a.1 is associated with an alternative splice site of XIST. Genome-wide splice junction mapping by RNA-seq indicated that DHS 200a.1 may be a splice donor site for alternative splicing of XIST.  Using primers upstream of the splice donor sequence paired with a primer within XIST exon 3 we observed two PCR products indicative of splicing from exon 1 (Figure 3.8A).  When we sequenced the amplicons we did indeed observe splicing out of XIST exon 1 at DHS 200a.1 which removes ~9kb of the 3? end of exon 1 (Figure 3.8A).  More surprisingly, sequencing of the smaller amplicon showed a spliced product that includes a 33bp novel exon 2, exon 2.1, located 61bp within XIST intron 1, and sequencing of the larger amplicon showed inclusion of a 92bp exon, exon 2.2 at the immediate 3? end of canonical XIST exon 1.  We call the XIST molecules containing exon 2.1 and exon 2.2, novel spliced XIST 1 and novel spliced XIST 2, respectively.      Using competitive PCR with a mixture of primers that amplify both canonical exon 1 and the novel splice products we saw that canonical exon 1 still amplified with a template dilution of 1:125 whereas novel spliced XIST 2 only amplified down to a template dilution of 1:5 and novel spliced XIST indicating that canonical XIST exon 1 is ~25 fold more abundant than the novel spliced XIST 1 and ~125 fold more abundant than novel spliced XIST 2.   The splice donor site in exon 1 appears to be well conserved in 5 eutherian species but the splice donor and acceptor sequences for exon 2.1 are not conserved in any of the examined species indicating that splice sequences are likely to have been lost in other lineages (Figure 3.8B).  Exon 2.2, however, shows conservation of the splice acceptor sequence between all 5   ABCompetetive PCR00.20.40.60.811.21.4EGFP siRNAYY1 siRNAqRT-PCR primers*** * *500bpXIST exon 119 200a.1 200D 200D.2 qXIST5C pG island00.20.40.60.811.21.419 200a.1 200D qXIST5 Exon1-3Relative YY1 RNA levels(Normalized to B-actin)Relative XIST RNA levels(Normalized to B-actin)EGFP siRNAYY1 siRNAFigure 3.7 - Knockdown of YY1 causes depletion XIST upstream and downstream of P2 transcriptionfrom P2 promoter. A) YY1 RNA levels in HEK293 cells treated with YY1 siRNA after 72hrs relative to HEK293 cells treated with EGFP siRNA after 72hrs knockdown indicating ecient knockdown of YY1 RNA.  B) XIST RNA levels in HEK293 cells after YY1 knockdown relative to HEK293 cells after EGFP knockdown.40Figure 3.8 - DHS 200a.1 is  a site of alternative splicing A) PCR  and competitive PCR performed on female lymphoblast cDNA using primers within exon 1 and exon 3.  Primers are indicated by arrows.  Canonical XIST is shown in the upper schematic and the sequenced spliced PCR product is shown in thelower  two schematics.  B) XIST  sequence alignment in 6 species.  Black boxes represent exon sequence.  Orange bars represent  a insertion of the number of nucleotides indicated in orangeBAGT AG GTCanonical XISTNovel spliced XIST 1200a.1R 1kbFemale Lymphoblast200DF Exon 3F 200a.1R:200DF 200a.1R: Exon 3F 200a.1R: 200DF: Exon3F 1 1:5 1:25 1:125 1:625 1:3125GT AG GTNovel spliced XIST 2HumanDogMouseCowRabbitHumanDogMouseCowRabbitHumanDogMouseCowRabbitExon 1 Splice Donor Exon 2.1Exon 2.24142  species and the splice donor site is also conserved as expected given that it is the canonical XIST exon 1 splice donor site (Figure 3.8B).     3.7 Transcription 3? of XIST The importance of antisense regulation by Tsix generates interest in investigating the role of TSIX in humans.  Specifically, we wished to explore the unexpected finding that TSIX is colocalized with the Xi rather than the Xa by analyzing the extent of transcription in several cell lines across XIST and beyond the 3? end of XIST.   Using strand-specific RT-PCR across the XIST gene body (Figure 3.9A) we observed transcription is in the sense orientation as previously reported in both female lymphoblast and female placenta cells (Figure 3.9B).  Sense transcription was also found in the sense orientation using strand specific RT-PCR primers beyond the 3? end of XIST at a locus in both female lymphoblast cells and female placenta cells (Figure 3.9B), in which antisense transcription was believed to be occurring [77].  Sense transcription beyond the 3? end of XIST suggested that in this region there was either a cryptic promoter initiating transcription or there was run-on transcription of XIST.  The abundance of transcript beyond the 3? end of XIST declines >10 fold relative to exon 8 of XIST in two female lymphoblast cell lines and three placental cell lines (Figure 3.9C) and  5?RACE analysis of the region in the GM7350 female lymphoblast cell line showed no transcription initiating beyond the 3? end of XIST (Figure 3.9D).  The combination of very low level transcription and the lack of novel transcription start sites argue that transcription 3? to the end of XIST is a result of run-on transcription of XIST.   3.8 Male hES cell line, CA1S, transcribes in antisense orientation across XIST locus We suspected that since XCI and the bulk of XIST regulation must occur early in human development that our studies of somatic cells may be missing regulatory element function.  To address the usage of P2, P2as and the region 3? of XIST we used the male hES cell line CA1S.  The obvious limitation of using CA1S cells is the inability to undergo XCI and therefore activation of XIST cannot be completely studied, however the repression of XIST can still be  Female lymphoblastFemale Placenta S         A S   S        A S  S         A S   S        A S  S        A S S         A SABC00.020.040.060.080.10.12Femalelymphoblast 1Femalelymphoblast 2female placenta1Female placenta2Terminatedfemale placentaRelative to XIST exon 8678? ly1  ? ly2 ? pl1 ? pl2 ? pl3Relative to primer 5+ C o n t r o lD1kbFigure 3.9 - Transcription beyond the 3? end of XIST is leaky XIST transcription. A) Schematic of XIST indicating primer positions (*) used in RT-PCR .  B) Representative strand specic RT-PCR to determine the orientation of transcription at the XIST locus.  C) qRT-PCR in female lymphoblast (ly) and placenta (pl).  Averages of biological triplicates ? standard deviation are shown.  D) 5?RACE beyond the 3? end of XIST to locate any unknown TSSs . + control is JPX 5?end.    5?RACE* * * * * ** *intron 4XISTexon 5exon  6exon  7exon  83?XIST13?XIST2TSIX11:12intron 4exon 5exon  6exon  7exon  8TSIX11:123?XIST13?XIST23?XIST13?XIST2TSIX11:124344  assayed in male cells and they avoid the confounding factor of XCI variability in female hES cells. We assayed both the region 3? of XIST and the 5? end of XIST by strand-specific RT-PCR and found antisense transcription in both locations (Figure 3.10A,C).  Stepping primers across the XIST locus pointed to consistent transcription throughout XIST and 3? of XIST suggesting that the antisense transcription is likely from one transcript rather than separate transcripts at the 3? and 5? regions (Figure 3.10B).  Amplification was lost between primers TSIX 13:14 and TSIX 15:16 (Figures 3.10A,B) signifying that the transcription start site is located between these two primers, 8683bp to 9934bp beyond the 3? end of XIST.  The region between and 5? of these two primers contains an enrichment of endogenous retrovirus (ERV) in the antisense orientation and an Alu element so it is possible that transcription is initiating from an ERV long terminal repeat promoter.  The faint RT-PCR bands hinted at low level transcription which was verified by qRT-PCR, indicating that the antisense transcript in this region is at 0.00012% of XIST expression in a somatic female lymphoblast (Figure 3.9D).                   * *TSIX7:8XIST3? of XIST * ** * * *1kbP1-1 5?A 200 b.1XE1bXIST ABC DTSIX11:12* *TSIX13:14TSIX15:16TSIX5:6Strand specic RT-PCRS              A SAs.trxS              A STSIX 11:12TSIX7:8 TSIX11:12TSIX13:14TSIX15:16TSIX5:6P1-1 5?A XE1bXIST 3?5?200b.1CA1S cDNAgDNA*As.trxXIST 3?5?Figure 3.10 - Transcription at the XIST locus in CA1S male hES cells A) Schematic of XIST indicating primer positions (*) used in RT-PCR .  B) RT-PCR in CA1S cells at the XIST locus.  C)  Stand specic RT-PCR to detemine orientation of transcription. D) qRT-PCR to determine abundance of CA1S antisense transcripts. CA1S cDNA levels are normalized to B-actin and shown relative to sense XIST in a female lymphoblast9E-050.0001050.000120.000135P1-1 RN A levels relative to female XIST     ( N ormaleiz ed to B- Actin)CA1S4546           4 Discussion             47  The motivation for this study was the lack of consensus about mouse and human XCI due to differences in sequence conservation of the XIC/Xic and transcription and conservation differences between mouse and human in the crucial mouse Xist regulator, TSIX/Tsix. We aimed to directly examine potential regulators of human XIST  and found regulation of XIST, in the form of alternative promoter usage and alternative splicing, allows for several XIST isoforms to lack domains involved in both silencing and localization (Figure 4.1A).    Refinement of the XIC to a 910,426bp region allowed for a more accurate picture of the location of cis-regulatory elements for XIST.  On the proximal end of the XIC our new mapping has excluded a gene previously considered a part of the XIC called NAP1L2.  The two remaining genes in the XIC region downstream of XIST are CHIC1 and CDX4.  These two genes are unlikely candidates for regulators of XIST since CHIC1 is expressed exclusively in brain [129] and CDX4 expression is activated in the primitive streak [130] which occurs after XCI is believed to take place.  We therefore believe that if regulators are located in the XIC region downstream of XIST they would be intergenic elements.  On the distal end of the XIC, the previous boundary did exclude the murine dosage-sensitive Xist activator, RNF12 but our new mapping provides the first direct evidence that FTX is not required intact for XCI to occur in humans.  We have not ignored the possibility that an alternative downstream promoter for FTX is compensating for the truncation but an examination of genome-wide DHS data does not indicate any elements nearby that would be capable of fulfilling this role.  The dispensability of FTX is in contrast to findings in mice that implicate Ftx as an Xist activator since Ftx null mutants showed alterations in chromatin environment within the Xic and decreases in XIST transcription [96].  The gene, XPCT, is also not found within the XIC and in mice overlaps the Xpr locus which may be important for X chromosome pairing.  Taken together, these findings provide further evidence for differences between mouse and human XIST/Xist regulation.    Looking within the newly refined XIC we determined a good candidate for XIST regulation to be located within the 5? end of XIST and found two DHS sites, 200b.1 and 200a.1.  These two DHS sites were specific to the inactive X chromosome based on the presence of sensitivity in 48  female and Xi hybrid cells but not in male or Xa hybrid cells.  In females, DHS 200b.1 overlaps binding of transcription factors and 200a.1 overlaps binding of CTCF so it seems likely that these factors are contributing to the observed sensitivity to DNase I.  Interestingly, these sites were not recapitulated in male HT1080 cells with XIST containing transgenes.  L1.10.1, a transgenic clone containing six tandem copies of a 460kb transgene of the XIC, shows a mild increase in sensitivity that was not statistically significant so it may be possible that there is variability in sensitivity between the six insertions.  A single-copy DOX-inducible XIST containing line also showed no sensitivity at this region before or after induction of XIST.  It is possible that long range cis-regulatory elements that lie outside of the L1.10.1 transgene are required but it seems more plausible that this DHS site requires passage through development to be fully established.  Interestingly, YY1 binding in a syntenic region in mouse has been found to be required for proper Xist localization [30], whereas we expect that the lack of DHS in our transgenic cells may be associated with limited YY1 binding indicating that YY1s role in localization is more complex than strictly acting as a tether at DHS 200b.1.    The finding that DHS 200b.1 is associated with an actively transcribing P2 promoter presents an interesting paradox; since P2 has been conserved between mouse and humans we propose that the region containing P2 must hold functionality yet transcripts originating from P2 lack the functional silencing element, Repeat A.  It was previously proposed, using RNAse Protection Assay (RPA), that mouse P2 might even be the major XIST promoter, with more than two fold higher expression than P1 but Navarro et al [16] suggested that secondary structure differences between probe locations may be responsible for quantity differences.  Our data supports the possibility of secondary structure effects, seeing that we observed large differences in RT efficiency at several regions but did not see a large increase in transcription downstream of P2.  Interestingly, a genome-wide survey of genes with long first exons, like XIST which has an 11kb first exon, demonstrated multiple transcription initiation sites and transcription factor binding throughout exon 1 so it may be true that P2 transcription is actually a product of first exon length.  Regardless of P2 origination the question of P2 function remains unanswered.        49  A major limitation in determining P2 function was the difficulty in directly analyzing P1 versus P2 transcripts.  Ideally, downregulation of P1 or P2 transcripts individually could determine functional capabilities of the two transcripts and other studies examining transcripts from alternative promoters have deleted the upstream promoter to be able to more effectively examine the downstream one [131].  We attempted to indirectly downregulate P2 via siRNA knockdown of YY1 which strongly binds to P2 and found decreased transcription upstream and downstream of P2 implying that both P1 and P2 are downregulated with YY1 knockdown.  It is unclear whether depletion of YY1 results in XIST downregulation due to YY1 directly binding at P2 or if YY1 acts in a pathway that is controlling XIST expression; removal of the YY1 binding sites at DHS 200b.1 would be an interesting experiment to address this question.  It seems plausible that direct binding of YY1 at DHS 200b.1 may somehow be controlling both P1 and P2 since methylation of the CpG island that overlaps DHS 200b.1 plays a major role in silencing of XIST on the Xa in males and females.   In any case, our findings conflict a previous report in mice that suggested downregulation of YY1 did not affect XIST transcription [30].  This discrepancy could reflect knockdown timing since Jeon et al. used a 48hr knockdown and we used a 72hr knockdown, or it could reflect species differences in the role of YY1 in XCI.  Another oddity of P2 is its ability to circumvent transcriptional interference by P1 transcription.  We did not find any evidence suggesting that P2 is protected from transcription interference because P1 transcripts are terminating upstream of P2.  This meant that P1 transcripts are indeed passing through P2 and likely extending the full length of XIST.  Another interesting possibility was that P1 and P2 transcription was imprinted, with the CpG island reflecting an imprinted control element but we found that P1 and P2 are not exclusively transcribed based on parent-of-origin, in line with evidence suggesting that human XCI does not show an imprinted pattern [34].  Lastly, we tested whether P2 was bidirectional since antisense transcription could inhibit complete interference of P2 transcription and uncovered P2as, an antisense transcript.  Several features of P2as including orientation, the distance upstream of P2, the location at the 5? edge of a CpG island and the very low level transcription all match characteristics of most active sense promoters [132, 133] so it appears that P2as is likely a 50  natural property of the sense P2 promoter.  It remains a possibility that P2 is exploiting associated P2as transcription to inhibit transcriptional interference but it seems improbable given the low level expression of P2as.    Our analysis of DHS 200a.1 uncovered a splice donor within DHS 200a.1 that removes ~9kb of the 3?end of XIST exon 1 and inclusion of two novel exons; exon 2.1 and exon 2.2 that are located at the 3? end of canonical exon 1 and within intron 1, respectively.  We surmise that alternative splicing at DHS 200a.1 is facilitated by CTCF binding since CTCF has been found to sufficiently pause PolII to allow co-transcriptional assembly of the splicing machinery on RNA  [134].  However, why the splice out at DHS 200a.1 is never directly spliced to canonical exon 2 but seems to always be via exon 2.1 or exon 2.2 is unknown.  Secondary structure of RNA is believed to play a role in mutually exclusive exon usage [135] and given the strong secondary structure that we and others have found, this seems like a plausible mechanism regulating XIST splicing.  Similar to P2 transcripts, this isoform of XIST also lacks functional elements since Repeats C and D may contribute to proper localization of XIST [23, 26, 30] .  Considering P2 transcription and alternative splicing together, it is tempting to speculate that the bulk of the novel spliced XIST that we describe could have initiated from P2 since P2 transcripts lack Repeat A and it?s been shown that when Repeat A is deleted, proper splicing of Xist is impaired [136].  This hypothesis could be tested using long PCR and a forward primer upstream of P2 and a reverse primer in a downstream exon.      An examination of splice site conservation indicated that exon 2.1 is unlikely to be conserved between species whereas exon 2.2 may be conserved.  The poor conservation of splice sequences found for exon 2.1 is interesting given that Horvath et al. describe a 194bp region, called CNS2, containing exon 2.1 to be highly conserved in 10 species including eight primates, mouse, dog and cow.  CNS2 does not fall within any of the exons that are conserved from Lnx3, the gene from which XIST is derived so it seems that it may be important to XIST function but maybe not through exon 2.1 usage.   51  When considering exon 2.2, its location within canonical XIST exon 1 is a good argument for why the splice sequences may be conserved but we have yet to test whether exon 2.2 is used in other species.     In the region 3? of XIST we have shown that the transcription, previously described as TSIX, in placental cells [78] may reflect run through XIST transcription.  Such a result could provide an explanation for the unanticipated finding that transcription 3? of XIST was colocalized with the Xi.  Migeon et al. also saw transcription 3? of XIST in fetal and neonatal cells [78]; it is logical to assume that this transcription is also due to run-on XIST.  Furthermore, transcription 3? of XIST in female fibroblasts has not been found and our quantitation of transcription indicated lower levels in this region in female lymphoblasts than in female placenta.  This warrants the question, why is run-through transcription more prominent at earlier developmental time points? It may be that XIST transcription itself diminishes with age and therefore leaky transcription does as well or that DNA binding factors that regulate transcription termination become more stringent.  Importantly, our finding that TSIX is not  transcribed in placenta also provides further support for the hypothesis that mouse and human TSIX/Tsix are not used equivocally since mouse extra-embryonic tissue shows persistence of Tsix expression after implantation [137].      To provide developmental context to our unique findings about human XIST we chose to look in male human ES cell line, CA1S for activity of P2, P2as and TSIX and transcriptional analysis identified an antisense transcript originating 3? of XIST and extending the full length of the XIST gene body.  This discovery is surprising given that all other descriptions of antisense transcripts at XIST did not extend that far into XIST, either terminating in XIST intron 4 or the 3? end of XIST exon 1 [77, 79].  A common argument against conservation of mouse-like TSIX function in humans is that transcription does not extend across the XIST promoter but we suggest that in human ES cells this is likely not the case.  However, we find it unlikely that this antisense transcript could be acting as an XIST repressor in these cells given its extremely low 52  abundance when compared to mouse TSIX which, based on RNA levels in male mouse ES cells [66], is ~300 fold more abundant.  A restriction in this interpretation is that hES cells do have a physiological propensity to advance towards a more epiblast-like state rather than remain mouse ES cell-like so the cells we have examined may have progressed past the point in which antisense regulation is required and have silenced transcription.  One way to investigate this possibility it to examine ?na?ve? hES cells which are more similar to mouse ES cells in terms of their gene expression profiles pathways employed.  Another enticing argument for the low levels of antisense transcription in CA1S cells is the observation that transcription is initiating from a very repetitive element rich region.  Since cells of the blastocyst show low levels of global DNA methylation levels (reviewed in [138]), it is possible that repetitive element transcription is not as tightly silenced and that a leaky repetitive element promoter is responsible for the antisense transcription in CA1S.   To integrate our findings and put them into the context of XIST regulation we propose that both P2 transcription and the presented alternative splicing of XIST could be playing a role in the initial choice and initiation of XCI.   Royce-Tolland et al. [136] proposed a model in which stochastic differences in properly spliced Xist levels during differentiation result in choice of Xi.  To elaborate on this model, we propose that in the early events of XCI the choice of the Xa and Xi is based on stochastic regulatory differences that allow one X chromosome to express basal levels of functional XIST more abundantly and become the Xi while the Xa transcribes basal levels of the alternatively spliced P2 transcripts that lacks both silencing and localization domains until repressive histone modifications are deposited and the P2-associated CpG island is methylated later in development (Figure 4.1B).  Asymmetric binding of trans-activating factors could determine whether P1 or P2 are used while asymmetric binding of CTCF may facilitate alternative splicing of non-functional XIST.  Asymmetric binding could be a result of pairing especially since CTCF binding has been suggested to facilitate pairing [139].  This model may also eliminate the requirement for TSIX since repression of XCI is fulfilled on the Xa by expression of non-functional XIST molecules.  In the future, this model and the function of other aspects of the presented findings could be tested using ?na?ve? hES cells that undergo XCI 53  upon differentiation to uncover the specific mechanisms regulating human XIST expression and XCI.                                   NAP1L20kSLC16A2/CNBP2FTXJPXTSIXXISTCHIC1CDX4Centromere TelomereDHS 200b.1DHS 200a.1P1 P2P2asX-inactivation centreExon 2.2Exon 2.1ABYY1DHS 200b.1DHS 200a.1P2P2asExon 2.2Exon 2.1DHS 200b.1DHS 200a.1P1 P2P2asFuture XaFuture XiAsymmetric distribution of trans- activators and CTCFCTCFNon-Functional XISTFunctional XISTTSIX in hES cellsFigure 4.1 - Summary of ndings and proposed model A) Summary of ndings indicating the rened XIC boundaries, DHS sites, P2 and P2as, YY1 activation of P1/P2, novel alternative splicing of XIST, and mouse-like TSIX in CA1S hES cells.  The orientation of XIST in the magnied image is reversed compared to the genomic context of the XIC B) XCI model in which asymmetric distribution of P1 and P2 trans- activators and CTCF causes dierences in basal transcription of functional XIST (P1 initated and properly spliced) and non-functional XIST (P2 and novel alternatively splicing).  The X chromosome which transcribes more functional XIST becomes the Xi whereas the X chromosome which trascribes more non-functional XIST becomes the Xa5455  References  1. Graves, J.A., Evolution of the testis-determining gene--the rise and fall of SRY. Novartis Found Symp, 2002. 244: p. 86-97; discussion 97-101, 203-6, 253-7. 2. Lyon, M.F., Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature, 1961. 190: p. 372-373. 3. Russell, L.B., Mammalian X-chromosome action: inactivation limited in spread and in region of origin. Science, 1963. 140: p. 976-978. 4. Brown, C.J., et al., Localization of the X inactivation centre on the human X chromosome in Xq13. Nature, 1991. 349(6304): p. 82-84. 5. Leppig, K.A., et al., Mapping of the distal boundary of the X-inactivation center in a rearranged X chromosome from a female expressing XIST. Hum. Mol. Genet., 1993. 2(7): p. 883-888. 6. Lafreniere, R.G., et al., 2.6 Mb YAC contig of the human X inactivation center region in Xq13: physical linkage of the RPS4X, PHKA1, XIST and DXS128E genes. Hum. Mol. Genet., 1993. 2(8): p. 1105-1115. 7. Rastan, S., Non-random X-chromosome inactivation in mouse X-autosome translocation embryos-location of the inactivation centre. J. Embryol. Exp. Morph., 1983. 78: p. 1-22. 8. Chureau, C., et al., Comparative sequence analysis of the X-inactivation center region in mouse, human and bovine. Genome Res., 2002. 12: p. 894-908. 9. Brown, C.J., et al., A gene from the region of the human X inactivation centre is expressed exclusively from the inactive X chromosome. Nature, 1991. 349: p. 38-44. 10. Brockdorff, N., et al., Conservation of position and exclusive expression of mouse Xist from the inactive X chromosome. Nature, 1991. 351: p. 329-331. 11. Brown, C.J., et al., The human XIST gene: analysis of a 17 kb inactive X-specific RNA that contains conserved repeats and is highly localized within the nucleus. Cell, 1992. 71: p. 527-542. 12. Brockdorff, N., et al., The product of the mouse Xist gene is a 15kb inactive X-specific transcript containing no conserved ORF and located in the nucleus. Cell, 1992. 71: p. 515-526. 13. Nesterova, T.B., et al., Characterization of the genomic Xist locus in rodents reveals conservation of overall gene structure and tandem repeats but rapid evolution of unique sequence. Genome Res, 2001. 11(5): p. 833-49. 14. Horvath, J.E., et al., Comparative analysis of the primate X-inactivation center region and reconstruction of the ancestral primate XIST locus. Genome Res, 2011. 21(6): p. 850-62. 15. Johnston, C.M., et al., Developmentally regulated Xist promoter switch mediates initiation of X inactivation. Cell, 1998. 94: p. 809-817. 16. Navarro, P., et al., Tsix transcription across the Xist gene alters chromatin conformation without affecting Xist transcription: implications for X-chromosome inactivation. Genes Dev, 2005. 19(12): p. 1474-84. 17. Kim, J.D., et al., Identification of clustered YY1 binding sites in imprinting control regions. Genome Res, 2006. 16(7): p. 901-11. 18. Brown, C.J. and H.F. Willard, The human X inactivation center is not required for maintenance of X inactivation. Nature, 1994. 368: p. 154-156. 19. Rastan, S. and E.J. Robertson, X-chromosome deletions in embryo-derived (EK) cell lines associated with lack of X-chromosome inactivation. J. Embryol. Exp. Morph., 1985. 90: p. 379-388. 56  20. Penny, G.D., et al., Requirement for Xist in X chromosome inactivation. Nature, 1996. 379: p. 131-137. 21. Lee, J.T. and R. Jaenisch, Long-range cis effects of ectopic X-inactivation centres on a mouse autosome. Nature, 1997. 386: p. 275-279. 22. Herzing, L.B.K., et al., Xist has properties of the X-chromosome inactivation centre. Nature, 1997. 386: p. 272-275. 23. Wutz, A., T.P. Rasmussen, and R. Jaenisch, Chromosomal silencing and localization are mediated by different domains of Xist RNA. Nat Genet, 2002. 30(2): p. 167-74. 24. Minks, J., et al., XIST-induced silencing of flanking genes is achieved by additive action of repeat a monomers in human somatic cells. Epigenetics Chromatin, 2013. 6(1): p. 23. 25. Clemson, C.M., et al., XIST RNA paints the inactive X chromosome at interphase: evidence for a novel RNA involved in nuclear/chromosome structure. J. Cell Biol., 1996. 132(3): p. 259-275. 26. Sarma, K., et al., Locked nucleic acids (LNAs) reveal sequence requirements and kinetics of Xist RNA localization to the X chromosome. Proc Natl Acad Sci U S A, 2010. 107(51): p. 22196-201. 27. Chow, J.C., et al., Inducible XIST-dependent X-chromosome inactivation in human somatic cells is reversible. Proc Natl Acad Sci U S A, 2007. 104(24): p. 10104-9. 28. Hasegawa, Y., et al., The matrix protein hnRNP U is required for chromosomal localization of Xist RNA. Dev Cell, 2010. 19(3): p. 469-76. 29. Gordon, S., et al., Transcription factor YY1: structure, function, and therapeutic implications in cancer biology. Oncogene, 2006. 25(8): p. 1125-42. 30. Jeon, Y. and J.T. Lee, YY1 tethers Xist RNA to the inactive X nucleation center. Cell, 2011. 146(1): p. 119-33. 31. Augui, S., E.P. Nora, and E. Heard, Regulation of X-chromosome inactivation by the X-inactivation centre. Nat Rev Genet, 2011. 12(6): p. 429-42. 32. van den Berg, I.M., et al., X chromosome inactivation is initiated in human preimplantation embryos. Am J Hum Genet, 2009. 84(6): p. 771-9. 33. Okamoto, I., et al., Eutherian mammals use diverse strategies to initiate X-chromosome inactivation during development. Nature, 2011. 472(7343): p. 370-4. 34. Migeon, B.R., X chromosome inactivation: theme and variations. Cytogenet Genome Res, 2002. 99(1-4): p. 8-16. 35. Sharp, A.J., et al., DNA methylation profiles of human active and inactive X chromosomes. Genome Res, 2011. 21(10): p. 1592-600. 36. Hendrich, B.D., C.J. Brown, and H.F. Willard, Evolutionary conservation of possible functional domains of the human and murine XIST genes. Hum. Mol. Genet., 1993. 2(6): p. 663-672. 37. Tinker, A.V. and C.J. Brown, Induction of XIST expression from the human active X chromosome in mouse/human somatic cell hybrids by DNA demethylation. Nucl. Acids Res., 1998. 26: p. 2935-2940. 38. Beard, C., E. Li, and R. Jaenisch, Loss of methylation activates Xist in somatic but not in embryonic cells. Genes & Dev., 1995. 9: p. 2325-2334. 39. Panning, B. and R. Jaenisch, DNA hypomethylation can activate Xist expression and silence X-linked genes. Genes & Dev., 1996. 10: p. 1991-2002. 40. Hansen, R.S., et al., Role of late replication timing in the silencing of X-linked genes. Hum. Mol. Genet., 1996. 5: p. 1345-1353. 41. Csankovszki, G., A. Nagy, and R. Jaenisch, Synergism of Xist RNA, DNA methylation, and histone hypoacetylation in maintaining X chromosome inactivation. J. Cell Biol., 2001. 153: p. 773-783. 42. Gartler, S.M., et al., Abnormal X: autosome ratio, but normal X chromosome inactivation in human triploid cultures. BMC Genet, 2006. 7: p. 41. 57  43. Migeon, B.R., et al., Species differences in TSIX/Tsix reveal the roles of these genes in X-chromosome inactivation. Am J Hum Genet, 2002. 71: p. 286-293. 44. Jacobs, P.A., et al., Late replicating X chromosomes in human triploidy. Am. J. Hum. Genet., 1979. 31: p. 446-457. 45. Migeon, B.R., et al., Human X inactivation center induces random X chromosome inactivation in male transgenic mice. Genomics, 1999. 59: p. 113-121. 46. Migeon, B.R., et al., Low-copy-number human transgene is recognized as an X inactivation center in mouse ES cells, but fails to induce cis-inactivation in chimeric mice. Genomics, 2001. 71(2): p. 156-62. 47. Heard, E., et al., Xist yeast artificial chromosome transgenes function as X-inactivation centers only in multicopy arrays and not as single copies. Mol. Cell. Biol., 1999. 19: p. 3156-3166. 48. Mlynarczyk-Evans, S., et al., X chromosomes alternate between two states prior to random X-inactivation. PLoS Biol, 2006. 4(6): p. e159. 49. Monkhorst, K., et al., X inactivation counting and choice is a stochastic process: evidence for involvement of an X-linked activator. Cell, 2008. 132(3): p. 410-21. 50. Bacher, C.P., et al., Transient colocalization of X-inactivation centres accompanies the initiation of X inactivation. Nat Cell Biol, 2006. 8(3): p. 293-9. 51. Xu, N., C.L. Tsai, and J.T. Lee, Transient homologous chromosome pairing marks the onset of X inactivation. Science, 2006. 311(5764): p. 1149-52. 52. Masui, O., et al., Live-cell chromosome dynamics and outcome of X chromosome pairing events during ES cell differentiation. Cell, 2011. 145(3): p. 447-58. 53. Augui, S., et al., Sensing X chromosome pairs before X inactivation via a novel X-pairing region of the Xic. Science, 2007. 318(5856): p. 1632-6. 54. Lee, J.T., Regulation of X-chromosome counting by Tsix and Xite sequences. Science, 2005. 309(5735): p. 768-71. 55. Sun, S., et al., Characterization of Xpr (Xpct) reveals instability but no effects on X-chromosome pairing or Xist expression. Transcription, 2010. 1(1): p. 46-56. 56. Cattanach, B.M. and J.H. Issacson, Contolling elements in the mouse X chromosome. Genetics, 1967. 57: p. 331-346. 57. Cattanach, B.M. and P. Johnston, Evidence of non-random X-inactivation in the mouse. Hereditas, 1981. 94: p. 5. 58. Simmler, M.-C., et al., Mapping the murine Xce locus with (CA)n repeats. Mammalian Genome, 1993. 4: p. 523-530. 59. Cattanach, B.M. and C.E. Williams, Evidence of non-random X chromosome activity in the mouse. Genet. Res., 1972. 19: p. 229-240. 60. Johnston, P.G. and B.M. Cattanach, Controlling elements in the mouse.  IV.  Evidence of non-random X-inactivation. Genet. Res., 1981. 37: p. 151-160. 61. Courtier, B., E. Heard, and P. Avner, Xce haplotypes show modified methylation in a region of the active X chromosome lying 3' to Xist. Proc. Natl. Adad. Sci., 1995. 92: p. 3531-3535. 62. Boumil, R.M., et al., Differential methylation of Xite and CTCF sites in Tsix mirrors the pattern of X-inactivation choice in mice. Mol Cell Biol, 2006. 26(6): p. 2109-17. 63. Thorvaldsen, J.L., et al., Nonrandom X chromosome inactivation is influenced by multiple regions on the murine X chromosome. Genetics, 2012. 192(3): p. 1095-107. 64. Lee, J.T., L.S. Davidow, and D. Warshawsky, Tsix, a gene antisense to Xist at the X-inactivation center. Nat. Genet., 1999. 21: p. 400-404. 65. Sado, T., et al., Regulation of imprinted X-chromosome inactivation in mice by Tsix. Development, 2001. 128: p. 1275-1286. 66. Shibata, S. and J.T. Lee, Characterization and quantitation of differential Tsix transcripts: implications for Tsix function. Hum. Mol. Genet., 2003. 12: p. 125-136. 58  67. Lee, J.T. and N. Lu, Targeted mutagenesis of Tsix leads to nonrandom X inactivation. Cell, 1999. 99: p. 47-57. 68. Luikenhuis, S., A. Wutz, and R. Jaenisch, Antisense transcription through the Xist locus mediates Tsix function in embryonic stem cells. Mol. Cell. Biol., 2001. 21: p. 8512-8520. 69. Ohhata, T., et al., Crucial role of antisense transcription across the Xist promoter in Tsix-mediated Xist chromatin modification. Development, 2008. 135(2): p. 227-35. 70. Avner, P., et al., Molecular correlates of the murine Xce locus. Genet. Res. Camb., 1998. 72: p. 217-224. 71. Cohen, D.E., et al., The DXPas34 repeat regulates random and imprinted X inactivation. Dev Cell, 2007. 12(1): p. 57-71. 72. Ogawa, Y. and J.T. Lee, Xite, X-inactivation intergenic transcription elements that regulate the probability of choice. Mol. Cell, 2003. 11: p. 731-743. 73. Ogawa, Y., B.K. Sun, and J.T. Lee, Intersection of the RNA interference and X-inactivation pathways. Science, 2008. 320(5881): p. 1336-41. 74. Nesterova, T.B., et al., Dicer regulates Xist promoter methylation in ES cells indirectly through transcriptional control of Dnmt3a. Epigenetics Chromatin, 2008. 1(1): p. 2. 75. Sado, T., Y. Hoki, and H. Sasaki, Tsix silences Xist through modification of chromatin structure. Dev Cell, 2005. 9(1): p. 159-65. 76. Navarro, P., et al., A role for non-coding Tsix transcription in partitioning chromatin domains within the mouse X-inactivation centre. Epigenetics Chromatin, 2009. 2(1): p. 8. 77. Migeon, B.R., et al., Identification of TSIX, encoding an RNA antisense to human XIST, reveals differences from its murine counterpart: implications for X inactivation. Am J Hum Genet, 2001. 69(5): p. 951-60. 78. Migeon, B.R., et al., Species differences in TSIX/Tsix reveal the roles of these genes in X-chromosome inactivation. Am J Hum Genet, 2002. 71(2): p. 286-93. 79. Chow, J.C., et al., Characterization of expression at the human XIST locus in somatic, embryonal carcinoma, and transgenic cell lines. Genomics, 2003. 82(3): p. 309-22. 80. Phillips, J.E. and V.G. Corces, CTCF: master weaver of the genome. Cell, 2009. 137(7): p. 1194-211. 81. Chao, W., et al., CTCF, a candidate trans-acting factor for X-inactivation choice. Science, 2002. 295(5553): p. 345-7. 82. Donohoe, M.E., et al., Identification of a Ctcf cofactor, Yy1, for the X chromosome binary switch. Mol Cell, 2007. 25(1): p. 43-56. 83. Ernst, J., et al., Mapping and analysis of chromatin state dynamics in nine human cell types. Nature, 2011. 473(7345): p. 43-9. 84. Hendrich, B.D., R.M. Plenge, and H.F. Willard, Identification and characterization of the human XIST gene promoter: implications for models of X chromosome inactivation. Nucl. Acids Res., 1997. 25: p. 2661-2671. 85. Plenge, R.M., et al., A promoter mutation in the XIST gene in two unrelated families with skewed X-chromosome inactivation. Nat. Genet., 1997. 17: p. 353-356. 86. Tomkins, D.J., et al., Lack of expression of XIST from a small ring X chromosome containing the XIST locus in a girl with short stature, facial dysmorphism and developmental delay. Eur J Hum Genet, 2002. 10(1): p. 44-51. 87. Pugacheva, E.M., et al., Familial cases of point mutations in the XIST promoter reveal a correlation between CTCF binding and pre-emptive choices of X chromosome inactivation. Hum Mol Genet, 2005. 14(7): p. 953-65. 88. Navarro, P., et al., Molecular coupling of Xist regulation and pluripotency. Science, 2008. 321(5896): p. 1693-5. 89. Navarro, P., et al., The X-inactivation trans-activator Rnf12 is negatively regulated by pluripotency factors in embryonic stem cells. Hum Genet, 2011. 130(2): p. 255-64. 59  90. Sun, B.K., A.M. Deaton, and J.T. Lee, A transient heterochromatic state in Xist preempts X inactivation choice without RNA stabilization. Mol Cell, 2006. 21(5): p. 617-28. 91. Minkovsky, A., et al., The pluripotency factor-bound intron 1 of xist is dispensable for x chromosome inactivation and reactivation in vitro and in vivo. Cell Rep, 2013. 3(3): p. 905-18. 92. Johnston, C.M., et al., Enox, a novel gene that maps 10 kb upstream of Xist and parially escapes X inactivation. Genomics, 2002. 80: p. 236-244. 93. Tian, D., S. Sun, and J.T. Lee, The long noncoding RNA, Jpx, is a molecular switch for X chromosome inactivation. Cell, 2010. 143(3): p. 390-403. 94. Chow, J.C. and C.J. Brown, Forming facultative heterochromatin: silencing of an X chromosome in mammalian females. Cell Mol Life Sci, 2003. 60(12): p. 2586-603. 95. Jonkers, I., et al., RNF12 is an X-Encoded dose-dependent activator of X chromosome inactivation. Cell, 2009. 139(5): p. 999-1011. 96. Chureau, C., et al., Ftx is a non-coding RNA which affects Xist expression and chromatin structure within the X-inactivation center region. Hum Mol Genet, 2011. 20(4): p. 705-18. 97. Zhao, J., et al., Polycomb proteins targeted by a short repeat RNA to the mouse X chromosome. Science, 2008. 322(5902): p. 750-6. 98. Vallot, C., et al., XACT, a long noncoding transcript coating the active X chromosome in human pluripotent cells. Nat Genet, 2013. 45(3): p. 239-41. 99. Barakat, T.S., et al., RNF12 activates Xist and is essential for X chromosome inactivation. PLoS Genet, 2011. 7(1): p. e1002001. 100. Ostendorff, H.P., et al., Ubiquitination-dependent cofactor exchange on LIM homeodomain transcription factors. Nature, 2002. 416(6876): p. 99-103. 101. Gontan, C., et al., RNF12 initiates X-chromosome inactivation by targeting REX1 for degradation. Nature, 2012. 485(7398): p. 386-90. 102. Hosler, B.A., et al., Expression of REX-1, a gene containing zinc finger motifs, is rapidly reduced by retinoic acid in F9 teratocarcinoma cells. Mol Cell Biol, 1989. 9(12): p. 5623-9. 103. Vallier, L., et al., Early cell fate decisions of human embryonic stem cells and mouse epiblast stem cells are controlled by the same signalling pathways. PLoS One, 2009. 4(6): p. e6082. 104. Hall, L.L., et al., X-inactivation reveals epigenetic anomalies in most hESC but identifies sublines that initiate as expected. J Cell Physiol, 2008. 216(2): p. 445-52. 105. Silva, S.S., et al., X-chromosome inactivation and epigenetic fluidity in human embryonic stem cells. Proc Natl Acad Sci U S A, 2008. 105(12): p. 4820-5. 106. Dvash, T., N. Lavon, and G. Fan, Variations of X chromosome inactivation occur in early passages of female human embryonic stem cells. PLoS One, 2010. 5(6): p. e11330. 107. Lengner, C.J., et al., Derivation of pre-X inactivation human embryonic stem cells under physiological oxygen concentrations. Cell, 2010. 141(5): p. 872-83. 108. Buehr, M., et al., Capture of authentic embryonic stem cells from rat blastocysts. Cell, 2008. 135(7): p. 1287-98. 109. Hanna, J., et al., Metastable pluripotent states in NOD-mouse-derived ESCs. Cell Stem Cell, 2009. 4(6): p. 513-24. 110. Nichols, J., et al., Validated germline-competent embryonic stem cell lines from nonobese diabetic mice. Nat Med, 2009. 15(7): p. 814-8. 111. Hanna, J., et al., Human embryonic stem cells with biological and epigenetic characteristics similar to those of mouse ESCs. Proc Natl Acad Sci U S A, 2010. 107(20): p. 9222-7. 112. Ware, C.B., et al., Histone deacetylase inhibition elicits an evolutionarily conserved self-renewal program in embryonic stem cells. Cell Stem Cell, 2009. 4(4): p. 359-69. 60  113. Chang, S.C. and C.J. Brown, Identification of regulatory elements flanking human XIST reveals species differences. BMC Mol Biol, 2010. 11: p. 20. 114. Song, L. and G.E. Crawford, DNase-seq: a high-resolution technique for mapping active gene regulatory elements across the genome from mammalian cells. Cold Spring Harb Protoc, 2010. 2010(2): p. pdb prot5384. 115. Heintzman, N.D., et al., Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat Genet, 2007. 39(3): p. 311-8. 116. Barrera, L.O., et al., Genome-wide mapping and analysis of active promoters in mouse embryonic stem cells and adult organs. Genome Res, 2008. 18(1): p. 46-59. 117. Wang, Q., J.S. Carroll, and M. Brown, Spatial and temporal recruitment of androgen receptor and its coactivators involves chromosomal looping and polymerase tracking. Mol Cell, 2005. 19(5): p. 631-42. 118. Bell, A.C., A.G. West, and G. Felsenfeld, The protein CTCF is required for the enhancer blocking activity of vertebrate insulators. Cell, 1999. 98(3): p. 387-96. 119. Kim, T.H., et al., Analysis of the vertebrate insulator protein CTCF-binding sites in the human genome. Cell, 2007. 128(6): p. 1231-45. 120. Brown, C.J., et al., A gene on the short arm of the human X chromosome complements the murine tsA1S9 DNA synthesis mutation. Som. Cell Mol. Genet., 1989. 15: p. 173-178. 121. Caron, N.J., et al., A human embryonic stem cell line adapted for high throughput screening. Biotechnol Bioeng, 2013. 122. Ozsolak, F., et al., High-throughput mapping of the chromatin structure of human promoters. Nat Biotechnol, 2007. 25(2): p. 244-8. 123. Schones, D.E., et al., Dynamic regulation of nucleosome positioning in the human genome. Cell, 2008. 132(5): p. 887-98. 124. Guenther, M.G., et al., A chromatin landmark and transcription initiation at most promoters in human cells. Cell, 2007. 130(1): p. 77-88. 125. The ENCODE (ENCyclopedia Of DNA Elements) Project. Science, 2004. 306(5696): p. 636-40. 126. Mazo, A., et al., Transcriptional interference: an unexpected layer of complexity in gene regulation. J Cell Sci, 2007. 120(Pt 16): p. 2755-61. 127. McDaniell, R., et al., Heritable individual-specific and allele-specific chromatin signatures in humans. Science, 2010. 328(5975): p. 235-9. 128. Buzin, C.H., J.R. Mann, and J. Singer-Sam, Quantitative RT-PCR assays show Xist RNA levels are low in mouse female adult tissue, embryos and embryoid bodies. Development, 1994. 120: p. 3529-3536. 129. Simmler, M.C., et al., Localization and expression analysis of a novel conserved brain expressed transcript, Brx/BRX, lying within the Xic/XIC candidate region. Mamm Genome, 1997. 8(10): p. 760-6. 130. Gamer, L.W. and C.V. Wright, Murine Cdx-4 bears striking similarities to the Drosophila caudal gene in its homeodomain sequence and early expression pattern. Mech Dev, 1993. 43(1): p. 71-81. 131. Kowalczyk, M.S., et al., Intragenic enhancers act as alternative promoters. Mol Cell, 2012. 45(4): p. 447-58. 132. Core, L.J., J.J. Waterfall, and J.T. Lis, Nascent RNA sequencing reveals widespread pausing and divergent initiation at human promoters. Science, 2008. 322(5909): p. 1845-8. 133. Seila, A.C., et al., Divergent Transcription from Active Promoters. Science, 2008. 322(5909): p. 1849-1851. 134. Shukla, S., et al., CTCF-promoted RNA polymerase II pausing links DNA methylation to splicing. Nature, 2011. 479(7371): p. 74-U99. 61  135. Yang, Y., et al., RNA secondary structure in mutually exclusive splicing. Nat Struct Mol Biol, 2011. 18(2): p. 159-68. 136. Royce-Tolland, M.E., et al., The A-repeat links ASF/SF2-dependent Xist RNA processing with random choice during X inactivation. Nat Struct Mol Biol, 2010. 17(8): p. 948-54. 137. Sado, T., et al., Regulation of imprinted X-chromosome inactivation in mice by Tsix. Development, 2001. 128(8): p. 1275-86. 138. Hackett, J.A. and M.A. Surani, DNA methylation dynamics during the mammalian life cycle. Philos Trans R Soc Lond B Biol Sci, 2013. 368(1609): p. 20110328. 139. Xu, N., et al., Evidence that homologous X-chromosome pairing requires transcription and Ctcf protein. Nat Genet, 2007. 39(11): p. 1390-6.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0074252/manifest

Comment

Related Items