Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies of X-chromosome inactivation and the identification of the Xist gene in the insectivore Scapanus… Karalić, Sanja 2001

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

Item Metadata


831-ubc_2001-0420.pdf [ 17.13MB ]
JSON: 831-1.0090227.json
JSON-LD: 831-1.0090227-ld.json
RDF/XML (Pretty): 831-1.0090227-rdf.xml
RDF/JSON: 831-1.0090227-rdf.json
Turtle: 831-1.0090227-turtle.txt
N-Triples: 831-1.0090227-rdf-ntriples.txt
Original Record: 831-1.0090227-source.json
Full Text

Full Text

STUDIES OF X-CHROMOSOME INACTIVATION AND THE IDENTIFICATION OF THE XIST GENE IN THE INSECTIVORE SCAPANUS ORARIUS by SANJA KARALIC B:Sc. (Hon.) Simon Fraser University, 1998 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR T H E DEGREE OF MASTER OF SCIENCE in T H E FACULTY OF GRADUATE STUDIES DEPARTMENT OF MEDICAL GENETICS F A C U L T Y OF MEDICINE We accept this thesis as conforming to the required standard  T H E UNIVERSITY OF BRITISH COLUMBIA May 2001 © Sanja KaraliC, 2001  In  presenting  this  degree at the  thesis  in  University of  partial  fulfilment  of  of  department  this or  thesis for by  his  or  requirements  British Columbia, I agree that the  freely available for reference and study. I further copying  the  representatives.  an advanced  Library shall make it  agree that permission for extensive  scholarly purposes may be her  for  It  is  granted  by the  understood  that  head of copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  H f 3& ' /2j3Qyf uc  i 1  Abstract  X-chromosome inactivation is the transcriptional silencing of one of the two X chromosomes in the mammalian female and is thought to be a means of dosage compensation. The inactive X chromosome is condensed, late replicating, hypoacetylated and hypermethylated at CpG islands. The silenced chromosome also expresses Xist - a fascinating large, untranslated R N A involved in the inactivation process. Although the mechanism of Xist function remains unknown, it has been established that the R N A is necessary for X-chromosome inactivation in two eutherian mammals - humans and mice. Xist fragments have also been identified for several other eutherians, but the only complete sequences are those of the human, murine and vole genes. Interestingly, Xist has not yet been identified in marsupials, although these mammals undergo X-silencing as well. Imprinted inactivation and the lack of CpG methylation are two additional aspects of X inactivation in which the marsupials differ from the eutherians. X-chromosome inactivation studies in the eutherian mammals have mostly been limited to the human and murine model organisms and little is known about the process in other members of this mammalian subclass. This study examined the process of X-inactivation and identified the Xist gene in a virtually unstudied mammalian organism - the Coast mole, Scapanus orarius.  The mole  belongs to the order Insectivora, an ancient and possibly basal mammalian group, which was traditionally thought to be closest to the eutherian common ancestor. Prior to this study, X chromosome inactivation in the insectivores had not been investigated in detail and the Xist gene had not been identified. By using methylation-sensitive restriction digests of several X linked genes (ARA, FMR1 and ZFX) followed by P C R with primers flanking the restriction  sites, this study confirmed the presence of a methylated X chromosome, indicating that these animals undergo X-inactivation. Additionally, the ARA and FMR1 genes were found to be subject to silencing, while ZFX was shown to escape X-inactivation in the mole. The results of the methylation assay not only show that X-inactivation exists in the Coast mole, but that silencing through methylation is conserved within the Eutheria. Low stringency P C R with primers designed from sequence conserved among the available Xist genes was successful in amplifying a mole fragment.  Sequence analysis and  female-specific expression confirmed that this was the Coast mole Xist homologue. Additional Scapanus Xist fragments were obtained through long range PCR and inverse PCR resulting in a total of 2.5 kilobases of sequence, which show significant similarity to Xist fragments from other mammals. The presence of Xist in the mole, along with its exclusively female expression pattern and sequence similarity to the gene in other species, indicate that Xist is involved in the X-inactivation process in this insectivore. In conclusion, the study of X-chromosome inactivation in the Coast mole shows that the unique features of eutherian X-linked silencing have been conserved in the Order Insectivora.  iv  Table of contents Abstract  H  Table of contents  iv  List of tables  vi  List of figures  vii  Acknowledgements  viii  Chapter 1 An introduction to sex determination and sex chromosomes Mammalian sex chromosomes X-chromosome inactivation in the Eutheria The elements of X-chromosome inactivation Late replication timing Histone modifications Hypermethylation The X Inactivation Centre (XIQ The XIST/Xist gene Tsix Spread of inactivation Genes escaping inactivation Silencing in the Metatheria and Prototheria and theories about the origins of X-inactivation Beyond the human and mouse models of X-chromosome inactivation Chapter 2 Materials and Methods Mole samples and cell lines Extraction of genomic D N A R N A isolation DNAse treatment of R N A The Polymerase chain reaction (PCR) Inverse P C R Reverse transcription Methylation-sensitive restriction digests D N A purification for sequencing and Xist sequence analysis Chapter 3 X-chromosome inactivation in Scapanus orarius Introduction The Coast mole {Scapanus orarius) D N A methylation in the mammalian genome  1 5 13 16 16 16 19 20 21 26 31 32 34 40 45 45 48 48 49 50 51 53 57 57 59 59 59 63  Methylation on the inactive X chromosome Identification of male and female Coast moles Results Sex determination using conserved Sty primers Methylation analysis Discussion  67 70 71 71 71 77  Chapter 4 Xist in the Coast mole: identification, expression and sequence analysis 82 Introduction Results Identification of Coast mole Xist Expression analysis of the putative Coast mole Xist Further sequencing of Coast mole Xist Discussion Chapter 5 Discussion and future directions Bibliography Appendix I  fragment  82 87 87 89 92 116 125 132 145  vi  List of tables  Table 1.  Coast mole samples.  46  Table 2.  Primers designed from Coast mole Xist sequence.  55  Table 3.  Miscellaneous primers.  56  Table 4.  Summary of methylation analysis.  76  Table 5.  Conserved Xist primers.  90  vii  List of figures  Figure 1.  Phylogeny of the Class Mammalia.  6  Figure 2.  Sex chromosomes in the three mammalian subclasses.  10  Figure 3.  A diagrammatic representation of inverse PCR.  52  Figure 4.  Conserved Xist primers.  54  Figure 5.  The Coast mole, Scapanus orarius.  60  Figure 6.  Coast mole D N A s amplified with the T e S R Y l ,2 primer pair.  Figure 7.  Methylation analysis of the ZFX, ARA, FMR1 and SMCX genes in the Coast mole.  72  75  Figure 8.  Available Xist sequences as of April 10 2001.  85  Figure 9.  The first Coast mole Xist fragment identified.  91  th  Figure 10. Female-specific expression of the putative Coast mole Xist fragment.  94  Figure 11. Coast mole Xist fragments.  97  Figure 12a. Aligned sequences in the 5' region of exon one.  98  Figure 12b. Aligned sequences in the 5' region of exon one (without rodents and cow).  104  Figure 13a. Aligned sequences at the end of exon one.  107  Figure 13b. Aligned sequences at the end of exon one, excluding the rodents.  111  Figure 14. The 12kb Coast mole fragment produced by ExpandLong PCR.  114  Figure 15. The alignment of known Xist sequences as of April 10 2001, including th  the Coast mole fragments.  117  Vlll  Acknowledgements  I would like to thank my supervisor, Dr. Carolyn Brown, for her excellent guidance and support throughout my Master's programme. I would also like to thank everyone in the Brown Lab, my committee members (Dr. Wendy Robinson, Dr. Sally Otto, Dr. Marco Marra, Dr. Fred Dill, Dr. Dixie Mager and Dr. Kelly McNagny), my "mole connections", Dr. Kevin Campbell and Tim Sheehan, Dr. Jim Rupert and Jane Lee.  1  Chapter 1  An introduction to sex determination and sex chromosomes  X-chromosome inactivation is the transcriptional silencing of one of the two X chromosomes (XX) in the mammalian female in order to achieve dosage equivalency with the male and is thought to be a direct consequence of the mammalian system of chromosomal sex determination (Lyon, 1961). The choice of sex in the animal world takes many forms and can involve a range of chromosomal and environmental factors. Sex determination in invertebrates is varied and a complete discussion is beyond the scope of this manuscript. However, two invertebrates using the chromosomal system of sex determination, Drosophila melanogaster and Caenorhabditis elegans, are especially relevant to the understanding of dosage compensation and will be discussed later. Although the most familiar type of sex determination in vertebrates is the X X - X Y system, characteristic of most mammals, in reality the subphylum Vertebrata shows a great diversity when it comes to defining sex (for a review see Solari, 1993). Fish, the most basal and ancient of all vertebrates, show a large plasticity regarding sex determination - most species have undifferentiated sex chromosomes, while others use the X X - X Y , ZW-ZZ, X X - X O , or multiple gonosome (sex chromosome) systems. These gonosome combinations are found scattered among the various species, seemingly without any strong phylogenetic connections, implying that sex chromosomes arose independently several times in the Pisces. Additionally, fish are unique among vertebrates in the number of species that are naturally hermaphroditic, either simultaneously or sequentially. Furthermore, in some gonochoristic species (separate sexes) sex  2 change can be easily elicited by simple hormonal treatment, while in hermaphrodites even behavioural changes are sufficient for sex reversal. The most ancient group of extant reptiles, the turtles and crocodilians, show environmental sex determination in which the crucial switch seems to be temperature. Heterogamety seems to have arisen several times in reptiles and species exhibit simple male heteromorphism ( X X - X Y ) , simple female heteromorphism (ZW-ZZ) or depend on multiple sex chromosomes. The degree of differentiation of the sex chromosomes varies widely and ranges from the cytogenetically indistinguishable to the visibly different, as is the situation in most snakes.  While the  knowledge of reptilian linkage maps remains scarce, it does not seem that the X or Z chromosomes have been conserved among the various reptilian groups, possibly due to the lack of dosage compensation in the form of X inactivation (Solari, 1993). Compared to other vertebrates, the situation in the Classes Mammalia and Aves is relatively straightforward.  Both groups possess well-differentiated heteromorphic sex chromosomes,  which seem to have arisen from an ancestral autosomal pair (Pask and Graves, 1999). In mammals the male is the heterogametic sex ( X Y vs. X X ) , while in birds it is the female who carries two types of sex chromosomes (ZW vs. ZZ). The X and Z chromosomes are relatively large and conserved, while the Y and W seem to show a range of differentiation (Graves et al., 1998a). In eutherian and marsupial mammals and carinate (flying) birds, the Y and W are small, heterochromatic and may pair over a small region with the X or Z. On the other hand, the Y in the monotremes and W in the ratite (flightless) birds are large and may pair completely with their respective partner (Graves et al., 1998a). However, significant differences exist between these two systems - the most glaring discrepancy is the identity of the heterogametic sex. In most mammals sex is controlled by a  3 dominant testis determining factor, identified as SRY (sex determining region on the Y) by Sinclair et al. in 1990, while avian sex seems to depend on a gene-dosage effect of the Z-linked DMRT1 (doublesex and mab-3 related transcription factor) gene (Nanda et a l , 2000). A recent study in turtles has shown that DMRT1 is the first gene to show temperature-dependent expression prior to sexual differentiation in reptiles (Kettlewell et al., 2000). Interestingly, DMRT1 is related to genes involved in sex regulation in both Drosophila and C. elegans and the mammalian DMRT1 homologue is located on the sex-reversal region of chromosome 9 (Nanda et al. 2000). These observations have led Nanda and colleagues (2000) to speculate that DMRT1 may be the dosage-based sex determination gene in ancestral vertebrates. Furthermore, while mammals engage in X-chromosome inactivation to achieve dosage equivalency of essential genes between males and females, birds do not seem to compensate for the gene imbalances arising from the possession of heteromorphic  sex chromosomes  (Baverstock, 1982).  Comparative mapping of the avian Z chromosome and mammalian X chromosome has shown that most bird Z-linked genes map to mammalian autosomes, especially human chromosome 9 (Nanda et al. 2000), and vice versa, indicating that these two sex chromosome systems evolved independently  from different  autosomal  pairs (Ohno,  1967, Graves et  al., 1998a).  Unfortunately, it seems impossible to deduce the sex determining mechanism of a common reptilian ancestor that would have existed approximately 350 million years ago (Pask and Graves, 1999), since modern reptiles exhibit a wide range of chromosomal and environmental sex determination. As is evident from this overview of vertebrate sex determination, sex chromosomes have arisen many times in many different phyla. It is now widely believed that the vertebrate protogonosomes were a pair of autosomes in which one allele acquired a sex-determining function,  4 followed by a suppression of recombination and gradual degradation of most of the chromosome carrying the novel gene, eventually leading to a fully differentiated pair of sex chromosomes. Evidence of this process can be seen in the residual identities between the sex chromosomes and by the identification of species that seem to be at different stages in this differentiation process from a homomorphic to a heteromorphic chromosome pair, such as snakes or ratite and carinate birds (Solari, 1993). The problem of dosage compensation that may arise in organisms with heteromorphic sex chromosomes has been dealt with in three manners represented by the nematode Caenorhabditis elegans, the fly Drosophila melanogaster and the eutherian mammal, such as mouse or humans. In C. elegans the X X genotype defines a functional hermaphrodite, while an X O animal is a male.  Gene product equivalency has been achieved by down regulating gene expression  through a multi-protein complex that assembles on both X chromosomes of the hermaphrodite, presumably reducing the transcriptional activity by altering chromatin structure (Lucchesi, 1998, Meller, 2000). The mechanism by which this dosage compensation machinery recognizes the correct chromosome is unknown, but may have something to do with the organization of the C. elegans chromosomes.  Unlike the autosomes, the X chromosome lacks a central gene  cluster, and one type of repetitive element is less frequent and more uniformly distributed on the X in comparison to the autosomes (Meller, 2000). The fruit fly deals with the problem of dosage compensation in the opposite manner by upregulating expression from the X Y male's single X chromosome, while expression from both of the female's X ' s remains unchanged. This increase in transcription is achieved through the M S L (male specific lethal) complex, composed of five identified proteins and two non-coding  5 roX RNAs (roXl and roX2), which appears to remodel chromatin by site-specific acetylations of histone H4 (Kelley and Kuroda, 2000a). Mammals employ the third known alternative for dosage equivalency, which is to completely silence one of the female's two X chromosomes.  The inactivated chromosome is  condensed, hypoacetylated, hypermethylated, nuclease-insensitive and replicates late in the cell cycle (Heard et al., 1997). In some species the condensed chromatin is visible as a Barr body. The process of X inactivation is thought to be responsible for the widespread conservation of synteny and size of the mammalian X chromosome, especially among eutherian mammals, since fragmentation of the X would disrupt dosage compensation. This phenomenon was first observed by Susumu Ohno in 1967 and is now known as Ohno's Law.  Interestingly, dosage  compensation has not been observed in a large number of organisms with heteromorphic sex chromosomes, such as birds or reptiles (Solari, 1993). Another paradox is the observation that the content of the bird Z chromosome seems to be well conserved despite the lack of chromosome wide inactivation (Nanda et al., 2000).  Mammalian Sex Chromosomes  The Order Mammalia consists of two subclasses: the Prototheria (monotremes) and Theria, which is in turn divided into the Metatheria (marsupials) and Eutheria (incorrectly referred to as the "placentals", since all therians possess placentae) (Figure 1). The monotremes, represented by the platypus and two echidna species, are thought to have diverged from the Therian line approximately 170 million years ago ( M Y A ) (Air et al., 1971, Pask and Graves, 1999) and display a mixture of ancestral reptilian traits and derived mammalian characteristics.  Figure 1: Phylogeny of the Class Mammalia (adapted from Pask and Graves, 1999).  7  They are ovoparous yet possess hair and milk glands. In contrast to all other mammals, which maintain a constant body temperature of 30-39° C, the internal temperature of a platypus is only about 30° C (Campbell, 1993). A l l the Therian mammals, which split into the metatherian and eutherian line 130 M Y A (Pask and Graves, 1999), are viviparous, but differ in the placental type and the maturity of young at birth. The tiny marsupial embryo exits the uterus only about a month after fertilization and continues its development in the marsupium or pouch. The eutherians, the most numerous and familiar group of mammals, give birth to relatively welldeveloped live young and possess an allantoic placenta. While the chromosomal sex determination system, in which X X defines a female and X Y a male, is conserved within mammals there are differences in the way primary and secondary sex characteristics are controlled among the Prototheria, Metatheria and Eutheria. Additionally, there are slight differences within the eutherians themselves such as the loss of the Y in molevoles and multiple gonosome systems in lemmings and Akodon rodents (Solari, 1993), but these organisms seem to represent exceptions to the established eutherian system. In the vast majority of eutherians the Y chromosome carries a dominant "testis-determining factor" and defines the male phenotype; individuals with an X O genotype are female, while a Y with any number of X chromosomes ( X X Y , X X X Y ) will be phenotypically male. In marsupials the Y also controls testis determination, however it seems that the presence of the male gonad is not sufficient for the development of all the secondary male characteristics, as is the case in eutherian mammals. A n X X Y animal has no scrotum and possesses a pouch with mammary glands, while an X O individual lacks a pouch and mammaries, but has a scrotum (Pask and Graves 1999). The situation in monotremes is not known because the animals do not breed in captivity (Watson et al., 1991).  8  The eutherian X chromosome is large (165 Mb (megabases) in humans), representing 5% of the haploid genome, and may carry as many as 3000 genes (Strachan and Read, 1999). In contrast, the largely heterochromatic Y chromosome is estimated at a mere 60 Mb in length (in humans) and contains approximately 30 functional genes, some of which have homologues on the X chromosome, which are generally not inactivated, and others which are Y-specific and exclusively expressed in the testis (Strachan and Read, 1999). Both the X and Y can be divided into differential regions and pseudoautosomal regions - areas that are shared between the sex chromosomes and, as a result, do not appear sex-linked. The human gonosome pair possesses two pseuodoautosomal regions (PARs), P A R I at the tips of the short arms and PAR2 at the tips of the long arms, at which recombination still takes place (Graves et al., 1998b). The gene content and size of the eutherian X chromosome are remarkably conserved, constituting 5% of the haploid genome regardless of the size of the autosomes (Ohno, 1967). Ohno (1967) postulated that this was due to mammalian X inactivation and that any rearrangements of this chromosome would be selected against since they would disrupt dosage compensation. Unlike the differential region of the X , the size, content and number of the pseudoautosomal regions varies widely between species.  Comparisons with metatherian and monotreme  chromosomes have shown that the genes in the eutherian PARs are autosomal in these diverged mammals, indicating that there have been several independent additions of autosomal material to both the X and Y in various eutherians (Solari, 1993, Graves et al., 1998b). The P A R genes are not subject to inactivation and their dosage is equal in the two sexes (Graves et al., 1998b). While pairing and recombination at the large P A R in both mice and humans is essential for spermatogenesis and fertility, the difference in gene content between these two species indicates that this function is not sequence-dependent (Graves et al., 1998b). Additionally, in some  rodents and all marsupials the gonosomes do not synapse at all, challenging the notion that X Y pairing is essential for meiosis and male gametogenesis (Solari, 1993, Graves et al., 1998b). Monotreme sex chromosomes are relatively large and the X and Y pair at meiosis over the entire short arm of the X and the long arm of the Y (Pask and Graves, 1999). The marsupial X is smaller than its eutherian counterpart and does not pair with the extremely small Y chromosome (12 Mb) at meiosis (Pask and Graves, 1999). Despite the lack of a pairing PAR, the marsupial gonosomes form an unusual X Y body ("balloon" or "dense plate structure") during male meiosis (Solari, 1993). Comparative mapping studies have shown that the region corresponding to the current marsupial X chromosome, which maps to the monotreme X and includes only the genes of the long arm and pericentric region of the human X , may represent the ancestral X chromosome content (Figure 2) (Pask and Graves, 1999). The remainder of the genes on the short arm of the eutherian X , including the large PAR, lie on the autosomes in monotremes and marsupials, indicating that they were a later addition to the eutherian gonosomes.  Additionally, Pask and Graves (1999) believe that the large region of homology  between the monotreme X and Y is another, independent addition of autosomal material to the sex chromosomes in the Prototheria. Although the X and Y chromosome pair of extant mammals may differ drastically in their gene content and size, there is ample evidence that the sex chromosomes of modern mammals evolved from an ancient autosomal pair.  The strongest evidence comes in the form of  homologous X - Y genes present on the differential regions of these chromosomes.  The  UBE1X/Y (ubiquitin-activating enzyme E l ) and SMCX/Y (selected mouse cDNA on the X ) genes, as well as SRY and its ancestor SOX3 (SRY-box containing gene 3), map to the X and Y in eutherians and marsupials (Graves et al., 1998a). Additionally, there are several marsupial X -  10  MONOTREMES  MARSUPIALS  EUTHERIANS  (HUMAN)  EUTHERIAN PSEUDOAUTOSOMAL REGIONS PAIRING DURING MALE MEIOSIS  Figure 2. Sex chromosomes in the three mammalian subclasses.  11  specific genes that map to the monotreme X Y pairing region (Watson et al., 1991), indicating an ancient homology. While the exact sequence of events in the evolution of the sex chromosomes is unknown, it must have involved the acquisition of a dominant male-determining gene on the Y , probably SRY, suppression of recombination between the proto-X and proto-Y chromosomes followed by decay of most Y-linked genes and the inactivation of most X-linked genes.  A n interesting  exception is the case of the SYBL1 (synaptobrevin-like 1) and HSPRY3 (sprouty (Drosophila) homologue 3), two genes located on the smaller P A R of the human X and Y (Lahn et al., 2001). These genes appear to undergo inactivation on both the X and Y chromosomes, suggesting that there may have been a recent X to Y translocation in the human lineage, since they are not present on the Y in other species (Lahn et al., 2001).  The differentiation of the sex  chromosomes was driven by the lack of recombination of the Y , which led to an accumulation of deleterious mutations as a result of genetic drift (Muller's ratchet), by hitchhiking (being carried along with a beneficial mutation) or the effects of sexually antagonistic genes (Graves et al., 1998a, Strachan and Read, 1999). A study by Lahn and Page (1999) based on the rate of synonymous nucleotide divergence estimated the time of X - Y divergence to be 240-320 million years ago, shortly after the mammalian and avian lineages spilt. The authors were able to divide the human X chromosome into four strata that range from the oldest proto-X material to subsequent additions of autosomal material.  According to their research, the oldest gene pair  on the human X and Y is the SRY/SOX3 pair, which is in accordance with the proposed model of sex chromosome evolution. The XIST gene, which has a crucial role in X-chromosome inactivation, was found to be slightly younger than the oldest gene pairs, which argues that  12 dosage compensation evolved after sex chromosome differentiation, although the possibility remains that it was X-inactivation that lead to Y attrition (Graves et al., 1998a). The end point of Y degeneration may be a complete loss of the chromosome, a process that seems to have already taken place in two species of voles, Ellobius tancrei and Ellobius lutescens, which lack a Y chromosome and the Sry gene (Solari, 1993, Just et al., 1995, Vogel et al., 1998). Both male and female E. lutescens carry only one X chromosome, while in E. tancrei all individuals possess an X X pair (reviewed in Solari, 1993). A recent study has shown that SOX9, an autosomally located gene involved in testis differentiation, is not the testis determining factor in E. lutescens (Baumstark et al., 2000), leaving sex determination in these rodents a complete mystery. A consequence of the mammalian system of dosage compensation is the reduction of X linked gene products by half in both males and females - a condition that would probably be very deleterious to both sexes.  Therefore, it is expected that an increase in the rate of  transcription of genes on the X chromosome would accompany the loss of the X in the male and the functional hemizygosity of the female (Ohno, 1967). Adler and colleagues (1997) compared the transcription rate of the Clc4 (chloride channel 4) gene, which is autosomal in Mus spretus and X-linked in M. musculus. The activity of the autosomal gene was half that of the X chromosome copy, supporting Ohnos's prediction. The entire eutherian X , excluding the pseudoautosomal regions, is remarkably conserved within the subclass, but not within the Class Mammalia. At the class level, Ohno's law only applies to the original, conserved proto-X shared by all three mammalian subclasses.  Ohno  hypothesized that this conservation is due to the phenomenon of X-inactivation, which occurs in both eutherians and metatherians, but it is still unclear why the monotreme X-chromosome  13 would share this conserved region since these animals may not undergo X-inactivation (Cooper et al., 1993, Graves et al., 1998a, Watson et al., 1990). Graves and colleagues have proposed that the monotreme and marsupial genomes may be more stable than that of eutherians, since the three extant monotreme species, although they have been separated for 50 million years, retain a virtually unchanged karyotype (Solari, 1993).  Similarly, the ancestral marsupial  karyotype or some simple derivative is shared by virtually all the species (Solari 1993, Graves et al., 1998b). Additionally, recent studies of non-mammalian vertebrate genomes have shown a conserved vertebrate genome even in fish, indicating that the large variations in eutherian karyotypes may be the exception rather then the rule (Graves et al., 1998a).  X-chromosome inactivation in the Eutheria  Although there are many unknowns regarding the mechanism involved in the transcriptional silencing of the X , even among the extensively studied eutherian models, the characteristics of the inactive chromosome are well established. The inactive X chromosome in eutherians is late replicating, hypermethylated at the promoter and CpG islands at the 5' ends of genes, hypoacetylated at the lysine residues of core histones, accumulates a novel histone variant macroH2A (Avner and Heard, 2000) and expresses the XIST gene (X Inactive Specific Transcript, Xist in mouse). The process of X-chromosome inactivation is thought to consist of three distinct phases: initiation, spread and maintenance (Heard et al., 1997) and is under the control of the X-chromosome inactivation centre (XIC). Initiation involves the counting of X chromosomes in the cell and choice of which one to inactivate, or more probably, the choice of  14  which one will remain active. The inactivation signal then spreads in cis from the XIC, silencing most of the genes on the chosen X chromosome.  Maintenance mechanisms, such as D N A  methylation, ensure the inactivation of the chromosome is clonally transmitted to the future cell generations.  Based on observations in murine female embryonic stem (ES) cells, Xist  expression, late replication and gene silencing are the earliest signs of inactivation, occurring at 1.5-2 days of differentiation (Keohane et al., 1998). These changes are followed by histone deacetylation at four days and, finally, D N A methylation at 21 days after induction of differentiation (Keohane et al., 1998). X-inactivation takes place very early during embryogenesis, approximately at the late blastocyst stage, and is stably maintained within the somatic cells of eutherians (Heard et al., 1997). During oogenesis, however, the silenced chromosome will be reactivated. In the mouse the extraembryonic trophoectoderm initiates inactivation at 3.5-4.5 days post coitum (dpc) and the primitive endoderm (yolk sac) from 4.5 to 5.5 dpc. The embryonic (epiblast) cells do not inactivate until after implantation at day 6.5, at which point the parental imprint has been erased and inactivation is random (reviewed in Lee, 2000). In the somatic tissues of eutherians both the paternal or maternal X can be inactivated. However, in the extraembryonic tissues of some eutherian mammals (mouse, rat) and in the somatic tissues of marsupials it is always the paternal X that is silenced (Avner and Heard, 2000). Since X-chromosome inactivation is thought to be a means of dosage compensation it is not expected that silencing would take place in the male. Interestingly, transcriptional silencing of the X and Xist expression does take place in the male germline during spermatogenesis, albeit at much lower levels than in female somatic tissues (Kay et al., 1993, Ayoub et al., 1997). Both the X chromosome and its pairing partner, the Y chromosome, become condensed and late  15  replicating at the onset of meiosis and combine to form a sex vesicle or X Y body, which has been reported in most eutherian and marsupial species (Solari, 1993). This process is transient and restricted to the period of meiosis, as a number of X-linked genes show post-meioitic transcription in spermatids (Hendriksen et al, 1995).  The silencing of the X chromosome  during male gametogenesis may be a mechanism to avoid possible damage from illegitimate recombination between the two sex chromosomes or the formation of unpaired recombination sites on the single X chromosome (Heard et al., 1997) and is probably distinct from X inactivation in the female. Although the X Y body accumulates the novel histone macroH2A (Hoyer-Fender et al., 2000) like the inactive X chromosome, the male X lacks the histone hypoacetylation seen in female somatic cells (Armstrong et al., 1997). The most significant difference was detected by Marahrens et al. (1997) who found that male Xist null mice were healthy and fertile, indicating that the silencing of the X in the male germline may not be crucial or that it may proceed through an ^'^-independent mechanism. Another intriguing observation, obtained using an X-linked lacZ transgene, is the transient silencing of the X chromosome in the somatic cells of the urogenital ridge in early male mouse embryos (10.5-11.5 dpc) (Jamieson et al., 1997). The authors have suggested that this may inactivate dosage-sensitive sex reversal loci on the X chromosome and may allow Sry to more efficiently guide testis differentiation.  16 The elements of X-chromosome inactivation  Late replication timing  One of the hallmarks of the inactive X , and one of the earliest signs of inactivation, is the appearance of a late-replicating chromosome. Replication late in the S phase is frequently used as a criterion to determine whether X:autosome translocations have undergone silencing. Although the chromosome as a whole appears late-replicating, studies at the single gene level have shown that replication timing corresponds to the transcriptional activity of genes or chromosomal domains (Heard et al., 1997). Genes subject to inactivation will replicate later than their counterparts on the active X chromosome, however, genes that escape inactivation replicate synchronously with the rest of the genome.  Demethylating treatments with 5-  azacytidine, which lead to chromatin decondensation and earlier replication in gerbil fibroblasts, have implicated D N A methylation in the control of replication timing (Heard et al.,1997).  Histone modifications  Histone modification plays an essential part in mammalian dosage compensation, as it does in Drosophila, and is probably an ancient and conserved feature of inert chromatin. Both the constitutive heterochromatin and inactive X chromatin of mammals are enriched for the hypoacetylated forms of histones H2A, H3 and H4 (Heard et al., 1997). The acetylation of histones at specific lysine residues is thought to reduce their affinity for D N A and each other,  17 leading to a open chromatin structure that is more conducive to transcription (Strachan and Read, 1999). According to this model, hypoacetylation of histones results in a closed chromatin conformation. Recent studies have linked the silencing ability of the methyl-binding protein MeCP2 to the recruitment of histone deacetylases (Ng and Bird, 1999).  Additionally, the  maintenance methyltransferase, D N M T 1 , can interact with histone deacetylases and repress transcription (Robertson and Wolffe, 2000). These interactions indicate that histone acetylation patterns might target methylation and that, conversely, D N A methylation can attract histone deacetylases through MeCP2. Until recently it was not possible to examine the histone acetylation status of individual genes and the globally deacetylated status of the X chromosome was inferred from immunofluorescence studies (Heard et al., 1997). This large scale deacetylation happens after four days of differentiation in female ES cells and does not represent an early event in the X inactivation process (Keohane et al., 1998).  However, recent work using a chromatin  immunoprecipitation technique has shown that, contrary to the established view, the levels of histone acetylation were comparable on the active and inactive X chromosome. A significant difference in acetylation status was only seen in the promoters of inactivated genes, which showed a dramatic increase in hypoacetylated histones and correlated with the methylated status of the adjacent CpG islands (Gilbert and Sharp, 1999).  Since deacetylation precedes  methylation, at least in female ES cells (Keohane et al., 1998), this promoter-specific deacetylation may not be due to the deacetylase-recruiting capabilities of MeCP2, but may precede it (Gilbert and Sharp, 1999, Ng and Bird, 1999). More support for a crucial role for histone deacetylation in the process of X-inactivation comes from experiments in which cells were treated with the deacetylase inhibitor trichostatin A  18 (TSA). The treated cells not only had increased levels of acetylated histones, but may also be defective in Xist localization and gene silencing (O'Neill et al., 1999). O'Neill and colleagues also reported the existence of a differentially hyperacetylated region upstream of the main Xist somatic promoter (1999).  The hyperacetylated domain is only present in undifferentiated  female ES cells and disappears upon differentiation, indicating a role in the choice of chromosome to be inactivated, possibly facilitating the upregulation of Xist expression through the maintenance of an open chromatin conformation (O'Neill et al., 1999). Since male ES cells and female cells with only one functional Xist allele do not show this hyperacetylation mark, the authors suggest it may also have a role in chromosome counting. The combined results of these recent experiments (Gilbert and Sharp, 1999, O'Neill et a l , 1999) imply that histone modifications may play a crucial role not only in the maintenance of the silenced state, but also in the initiation of the X-chromosome inactivation. Another role for histones in X-chromosome inactivation was revealed by the discovery of a novel histone variant, macroH2A, which associates with the inactive X chromosome to form a macrochromatin body (MCB) in humans, mice and dogs (Costanzi and Pehrson, 1998). Conditional deletions of the Xist gene in mouse embryonic fibroblasts result in the disruption of the M C B , indicating that the Xist R N A is necessary for histone macroH2A deposition on the inactive X chromosome. However, the lack of both the Xist R N A and the M C B do not affect X silencing once it has been established and will not lead to the reactivation of the chromosome (Csankovski et al., 1999). Previous results had indicated that XIST is not associated with the nucleosome, as it remained attached to the nuclear matrix when the chromosomal D N A and histones were removed  (Clemson et al., 1996).  However, using a more sensitive method  Gilbert et al. (2000) showed that XIST R N A can be co-precipitated by antiserum for macroH2A  19  and by antisera that recognize hypoacetylated histones H3 and H4. This discrepancy can be explained by the fact that the older study used relatively harsh conditions when treating the nuclei, which may have lead to the disruption of the putative Z75T-histone complexes (Gilbert et al., 2000). These results, along with an indication that the novel domain of macroH2A may have an RNA-binding function (Pehrson and Fuji, 1998), support a role for XIST/Xist in chromatin remodeling, possibly by nucleating ribonucleoproetin chromatin modifying complexes as is the case in Drosophila (Gilbert et al., 2000, Kelley and Kuroda, 2000a, b).  Hypermethylation  Although the global methylation levels on the inactive and active X chromosomes seem to be comparable, methylation of CpG islands at the 5' end of genes on the inactive X chromosome is thought to be a crucial factor in the stable maintenance of inactivation (Heard et al., 1997).  Methylation patterns on the inactive X chromosome clearly differ between  individual genes and correlate with expression - genes that are transcriptionally repressed are methylated at the CpG islands, while those that escape X-inactivation are not (Heard et al., 1997, Hansen et al., 1998). Lack of such methylation on the marsupial X chromosome (Kaslow and Migeon, 1987) and hypomethylation in the extraembryonic tissues is thought to be the reason for the relative instability of silencing in these two systems (Heard et al., 1997). Despite its late appearance in the silencing process, methylation has been the prime candidate for a epigenetic mark that would be involved in counting, chromosome choice or imprinted X inactivation in the extraembryonic tissues, based on its crucial role in the regulation of  20 autosomal imprinted genes (Heard et al., 1997, Sado et al., 2000, Lee, 2000).  However, as of  yet, no differential methylation has been identified that fulfills the criteria of an epigenetic mark involved in the initiation of the inactivation process. The majority of the evidence indicates that methylation may not be an initial event of X-inactivation, but a later stabilizing mechanism (Heard et al., 1997).  The XInactivation Centre (XIC)  The XlC/Xic (human/mouse), defined as an 80 kilobase (kb) fragment in the mouse and a one megabase region in humans, is located on the X chromosome and contains several elements involved in the inactivation process (reviewed in Heard et al., 1997, Avner and Heard, 2001). It is essential for silencing to occur in cis, and an X chromosome lacking the XIC cannot be inactivated. This region is also thought to be necessary for the counting of the ratio of X chromosomes to autosomes. The putative counting mechanism is unknown, although the most favoured hypothesis is that of an autosomal blocking factor or nuclear attachment site that would be present in limited quantities and bind to only one X chromosome per autosomal complement.  While this mechanism remains elusive, it is clear that it exists, as cells that are  aneuploid for the X (47, X X X and 48, X X X X ) inactivate all but one chromosome (Brown and Willard, 1993). Tetraploid cells contain two inactive and two active X chromosomes, while the situation in triploids is less clear and 69, X X X and 69, X X Y triploids have been found to have one or two active X chromosomes (Brown and Willard, 1993). Although X-chromosome inactivation is random in eutherian somatic tissues, in the absence of selection against a particular X at the cellular level, a locus named the X controlling element  21  (Xce) was found to influence the probability that a particular chromosome would be silenced (Avner and Heard, 2000). A chromosome with a "strong" Xce allele, contained within the Xic in mouse, but distinct from Xist, would remain active more often than one with a weaker allele (approximately 70%:30% ratio).  The XIST/Xist gene  In recent years significant advances have been made in the understanding of X-chromosome inactivation, however, the function of the most significant element in the silencing process, the XIST/Xist gene, is still a mystery. The XIST/Xist gene is contained within the XlC/Xic and is the 1  only gene expressed exclusively from the inactive X chromosome, resulting in a large (17 to 19 kb in human, 15 to 18 kb in mouse) spliced, polyadenylated transcript (Brown et al., 1991, Brockdorff et al., 1992, Brown et al., 1992, Hong et al., 2000). The XIST/Xist R N A is not translated and is retained in the nucleus where it associates, through a yet unknown mechanism, with the chromosome from which it originated (Clemson et al., 1996). XIST/Xist expression is thought to be an early, possibly the primary, step in the inactivation process and chromosomes lacking the gene cannot inactivate (Penny et al., 1996, Avner and Heard 2001). The XIST/Xist R N A coating the inactive X dissociates from the chromosome as the cells enter mitosis and is re-transcribed in the daughter cells (Lee and Jaenisch, 1997). Based on an experiment in which a transgene carrying Xist and 15 kb of surrounding sequence was sufficient for inactivation, Herzing et al. (1997) have even suggested that Xist  "By convention, the human X Inactive Specific Transcript gene is capitalized (XIST), while homologous genes in other species are not (Xist).  22  may be the XIC and is the only sequence necessary for silencing. However, various transgenes, females with deletions downstream of ATSTand mice with induced deletions 3' of Xist, suggest that the counting and choice function of the XlC/Xic lies outside of XIST/Xist (reviewed in Avner and Heard, 2001, Wutz and Jaenisch, 2001). Penny et al. (1996) demonstrated, through the use of a deleted Xist allele in which the minimal promoter was left intact, that the presence of the Xist transcript, rather than just the act of transcription in the region of the gene, is essential for silencing to occur. A n upregulation of the Xist gene and increased transcript stability seems to be the trigger for inactivation (Avner and Heard, 2001, Wutz and Jaenisch, 2001). While some authors (Johnston et al., 1998) have suggested that this increased half-life is due a promoter switch, those results have not been substantiated (Warshawsky et al., 1999). A n alternative theory holds that the increased R N A stability depends on the accumulation of a threshold amount of the transcript and the instability of Xist in normal ES cells may be a result of poor expression (Wutz and Jaenisch, 2001). It is still unclear what leads to the increase in transcription, but a differentially acetylated region upstream of the promoter is a candidate (O'Neill et a l , 1999). The histone H4 hyperacetylation, which is seen only in undifferentiated female cells and disappears upon differentiation, may lead to transcriptional upregulation through the maintenance of an open chromatin conformation. Additionally, the Tsix antisense transcript seems to be involved in Xist expression in both random and imprinted inactivation (see below). When the XIST/Xist promoter is methylated, as it is on the active X chromosome, the gene is not expressed (Brown, 1992, Norris et al., 1994, Beard et al., 1995).  In mouse embryos  deficient for the D N A methyltransferase gene, Xist is hypomethylated and expressed in both males and females (Heard et al., 1997). This expression may simply be due to the global  23 hypomethylation of the genome in these cells, rather than a result of the specific loss of Xist methylation.  Initial results indicating that parental differences in the methylation of the  promoter could be responsible for the imprinted inactivation seen in extraembryonic tissues (Ariel et al., 1995, Zucotti and Monk, 1995) have been contradicted by more recent experiments that have shown that the parental alleles do not exhibit differential methylation in the region (McDonald et al., 1998). Thus, the methylation pattern seen in the promoter and first exon of XIST/Xist seems to be a result of its transcriptional status. Experiments involving transgenes containing the XIST/Xist genes and surrounding sequence have shown successful transcriptional silencing of autosomes into which they have integrated, even in male cells (Migeon et al., 1999), and confirm the observations from Xrautosome translocations, which show that some autosomes are more amenable to the spread of inactivation than others (Lyon, 1998). Interestingly, once the inactivated state is established, as it is somatic cells, deletion of XIST or the XIC does not seem to lead to reactivation (Brown and Willard, 1994). Likewise, reactivation of XIST through demethylation in rodent-human somatic cell hybrids, while it leads to XIST expression, does not lead to the correct XIST localization nor X-inactivation (Hansen et al., 1998, Tinker and Brown, 1998). Subsequent experiments have indicated that even with Xist expression and correct localization, X-chromosome inactivation cannot be induced in adult somatic cells, indicating that the developmental timing of XIST expression is crucial (Clemson et al., 1998). Experiments by Wutz and Jaenisch (2001) involving an inducible Xist cDNA transgene in male ES cells support these observations. The authors found that they could induce reversible A7sr-dependent repression only in undifferentiated ES cells.  Once the cells had begun to  differentiate, however, the inactivation became irreversible and independent of ongoing Xist  24  expression. The transgenic Xist, although it lacked the counting function, led to late replication and histone hypoacetylation of the inactive chromosome once the cells had differentiated. A n intriguing observation, which may shed light on the mechanism of Xist function, was the Xistinduced silencing in undifferentiated ES cells that proceeded without any detectable change in replication timing or histone hypoacetylation. There is no doubt that the XIST/Xist R N A is an essential part of eutherian X inactivation, yet it does not exhibit a high degree of sequence conservation. The only two genes to have been sequenced in their entirety, the mouse and human homologues, exhibit only 70% sequence identity (Brockdorff et al., 1992, Brown et al., 1992). A study by Hendrich and colleagues (1993) examined the evolutionary conservation of the gene in a number of mammals (several primates, fox, cat, rabbit, cow, hamster and horse). By performing Southern blots with probes from various regions of the human XIST, the authors were able to identify a region of increased sequence homology in the 5' region of the gene. Sequencing of a subset of the animals revealed a series of eight to nine repeats, encompassing approximately one kilobase, with a highly conserved core. However, alignments of the surrounding sequenced regions indicated a modest homology. The level of conservation was less than what is seen in protein-coding exons, yet significantly higher than found in introns, suggesting a selective pressure.  Since XIST/Xist  probably functions at the R N A level, appropriate secondary structure, rather than strict sequence conservation, would be the expected target of natural selection. Additionally, methylation analysis in the putative promoter region of the gene showed a similar methylation pattern in human, murine and bovine samples. While the authors note that the 5' conserved repeats represent the strongest candidate for a functional domain, this hypothesis has not been confirmed. Despite the low conservation, crucial regions seem to be conserved between the  25  mouse and human homologues since a human XIST Y A C has been shown to induce inactivation in mouse autosomes, although without the accompanying late replication and histone hypoacetylation (Heard et al., 1999, Migeon et al., 1999). The absence of histone modifications and late replication in this system, along with the observation that in hamster-human somatic cell hybrids XIST is not properly localized and cannot induce repression, point to the existence of species-specific autosomal factors that contribute to the inactivation process (Hansen et al., 1998). Although evidence is currently lacking, most authors believe that XIST/Xist acts as part of a ribonucleoprotein complex that is capable of chromatin remodeling and gene silencing (Mlynarczyk and Panning, 2000) in a manner analogous to the roX RNAs in Drosophila (reviewed in Kelley and Kuroda, 2000a,b).  This theory is supported by the co-  immunoprecipitation of Xist and hypoacetylated histones H3 and H4, and the requirement of Xist for the deposition of macroH2A, containing a putative RNA-binding domain, on the inactive X chromosome (Pehrson and Fuji, 1998, Gilbert et al., 2000). Both XIST/Xist and the roX 1 and roX 2 R N A are expressed from the X chromosome and spread in cis from their site of synthesis, although the extent of Xist spread is much greater, and both involve a change in histone acetylation.  A l l three RNAs are spliced and seem to coat the X chromosome in a  banded pattern (Duthie et al., 1999, Kelley and Kuroda, 2000a,b). While both systems involve a non-coding R N A expressed from the X chromosome and are the only RNAs to spread from their point of origin, no indications of a common evolutionary origin have been reported (Kelley and Kuroda, 2000b). The roX RNAs associate with a six-protein complex that mediates malespecific hypertranscription of the X-linked genes. Mutations in five of the genes forming this group, known as the male-specific-lethal (MSL) complex, have been isolated and all cause male  26 death.  Five of the protein components have been characterized so far and fall into two  categories.  The M L E component is an R N A helicase, the M O F is a histone H4  acetyltransferase, while JIL-1 can phosphorylate histone H3 in vitro. proteins seem to bind R N A through their chromodomains.  The remaining two  The M S L group seems to recruit  nascent roX R N A at two chromatin entry sites to form a mature complex.  Unlike XIST/Xist,  which only coats its chromosome of origin, the roX RNAs seem to be able to locate the X in trans and appear to diffuse freely between the 35 chromatin entry sites.  Not much is known  about the sex-specific regulation of roX expression, although there are no indications that an antisense R N A , analogous to Tsix, is involved (see below).  Tsix  Another crucial element of the Xic, at least in the mouse, seems to be the Tsix antisense R N A , which is transcribed in undifferentiated ES cells and early embryos and spans the entire Xist gene (Lee et al., 1999). Very early in embryogenesis low levels of the Xist and Tsix R N A are expressed from every Xic present. The Xist R N A is unstable during this pre-inactivation stage and does not advance from its site of transcription (Mlynarczyk and Panning, 2000). Once inactivation is initiated, through an unknown mechanism, Xist R N A is seen to spread over the chromosome that will be silenced, while Tsix expression stops.  On the remaining X  chromosome, low levels of both Xist and Tsix transcripts persist for some time and then cease. The 5' end and promoter of Tsix lie within the DXPas34 locus - a short CpG rich region that is hypermethylated on the active X chromosome. The degree of methylation at this locus seems to  27 correlate with the tendency of a chromosome to avoid inactivation and is reminiscent of the Xce, although mapping studies show the two loci to be separate (Avner and Heard, 2000). Deletions of the 5' end of Tsix and DXPas34 lead to the exclusive inactivation of the deleted chromosome in ES cells, indicating that these elements influence X-chromosome choice and imprinting, but not counting. Recent research has shown that Tsix is not only involved in random silencing, but is vital for imprinted inactivation in mouse extraembryonic tissues (Lee, 2000). In normal mouse embryos the maternal chromosome is thought to be resistant to the inactivation signal or shows a higher affinity for an activating factor, and it is the paternal chromosome that is silenced. Sometime after the differentiation of the extraembryonic lineages, the parental mark is lost and inactivation is random in the embryo proper. Mice inheriting a Tsix gene deletion at the 5' CpG island from their fathers were born at the expected frequency, while there was a marked shortage (18% survival rate) of pups carrying the deletion on the maternal X chromosome, indicating a severe parent-of-origin effect. In the mutant male embryos the deleted maternal X was inactivated, while both the paternal and maternal X chromosomes were silenced in mutant female embryos. This embryonal lethality could be due to the inability to reactivate the deleted X chromosome during oogenesis or to the failure of imprinted inactivation in the extraembryonic tissues. Lee (2000) found that the cause of death was due to the incorrect inactivation of the maternal X chromosome in the extraembryonic tissues, rather than a failure in oogenesis. These results are consistent with previous work, which showed that the maternal X chromosome in rodent extraembryonic tissues is normally not silenced (Marahrens et al., 1997, Goto and Takagi, 1999). Lee proposes that the probable imprint differentiating between the maternal and paternal chromosome, with respect to Tsix expression, is the differentially methylated CpG region of  28  DXPas34.  Studies on other imprinted genes have shown that the putative imprinting centres  contain differentially methylated CpG islands and are located within or near genes coding for antisense and noncoding RNAs (reviewed in Lee, 2000). Abolition of methylation through targeted knockouts of the Dnmt 1 maintenance methyltransferase results in the loss of imprinted gene expression (reviewed in Robertson and Wolffe, 2001). The identification of differential methylation in the 4kb CpG-rich putative promoter and transcriptional start site domain of Tsix, also known as the DXPas34 region, strengthens this hypothesis (Courtier et al., 1995). However, more recent experiments by Prissette and colleagues (2001) have not revealed a differential methylation pattern in extra-embryonic tissues by bisulfite sequencing. Additional evidence that methylation may not be the imprint is given by the experiments of Sado and colleagues (2000) in which X-inactivation was examined in mice lacking the maintenance methyltransferase Dnmtl.  Previous work had shown that targeted disruption of Dnmt I led to  the lethal ectopic Xist expression in male embryos and ES cells, and expression from both X chromosomes in female ES cells. This expression was presumably due to the loss of methylation at the 5'region of Xist and it was thought that it may lead to inappropriate X-inactivation in vivo (Beard et al., 1995, Panning and Jaenisch, 1996). However, the results of Sado et al (2000) show that the lack of Dnmtl does not seem to disrupt imprinted X-inactivation in the extraembryonic tissues of the mutant mice embryos indicating that methylation may not be the imprinting mark in this case (2000).  However, they report a disruption of inactivation and  reactivation of the silenced X chromosome in the embryo proper, supporting a role in the maintenance and stabilization of X-chromosome inactivation. The authors hypothesize that histone deacetylation may have a dominant role in the imprinted silencing seen in extraembryonic tissues and that histone deacetylases, which have a universal role in gene  29 silencing regardless of whether an organism possesses methylation, may have a more fundamental role than previously thought. Although a differentially hyperacetylated region was identified upstream of the Xist promoter in undifferentiated female ES cells, it has not been examined in extraembryonic tissues (O'Neill et al., 1999). Further work needs to be done to determine whether a sex-specific mark, possibly due to differential hypoacetylation, is responsible for the inactivation of the paternal X chromosome in rodent extraembryonic tissues. In conclusion, the Tsix deletion in ES cells leads to the virtually exclusive inactivation of the mutant chromosome. In extraembryonic tissues the parental imprint overrides the counting mechanism and leads to paternal inactivation in the vast majority of cells, despite the fact that the maternal chromosome is already inactivated due to the Tsix deletion.  The combined  evidence points to a role for Tsix in chromosome choice and imprinting, but not in counting. However, neither the inactivation of the paternal X (Xp) in ES cells, nor the lethality of the deleted X p X m (maternal) and X m Y embryos was absolute. Similarly, X p O mice, although growth-retarded and rare, are viable (Solter and Wei, 1997). These results imply that imprinting must not be complete in extraembryonic tissues, a conclusion supported by previous research estimating that only 85%-90% of rodent extraembryonic cells actually show paternal X inactivation (Tagaki and Sasaki, 1975, Lee, 2000). Furthermore, since male mice carrying an Xist deletion were fertile (Marahrens et al., 1997) and females with a paternal Tsix deletion had normal oogenesis (Lee, 2000), it seems that X-inactivation and reactivation in gametogenesis may be proceed through a Xist-Tsix independent mechanism or may not be as crucial as in other tissues. Although expression downstream of human XIST, in the putative region of a human Tsix equivalent, has been detected, it does not seem to result in the large transcript seen in mouse (J.  30 Chow, personal communication). Due to the difficulty inherent in studying early human development it remains unclear i f a Tsix homologue exists, or is necessary, in humans. Unlike mice, humans with supernumerary  maternal X chromosomes have relatively normal  development ( X X X females and X X Y Klinefelter males). Most research indicates that humans do not possess a parental imprint on one of the X chromosomes and that either can be inactivated.  However, recent results by Looijenga et al. (1999) have shown heterogenous  expression of the androgen receptor (ARA) gene in human placenta. Some placentae exhibited preferentially maternal inactivation, some preferentially paternal, while others had biallelic expression. While these observations may be due to some sort of unstable imprinting, they may just be a result of variable inactivation seen for several X-linked genes. While it seems apparent that Tsix is a negative regulator of Xist, the mechanism by which this regulation is achieved is unknown. The proposed mechanisms can be divided into two classes: those that require the Tsix transcript and those in which the act of transcription is a byproduct of Tsix function. In the first group of mechanisms interaction between the Tsix and Xist transcripts could result in the destabilization of Xist secondary structure or might target the putative Tsix.Xist R N A duplex for destruction by an R N A interference mechanism (Mlynarczyk and Panning, 2000). In the second class of mechanisms the very act of Tsix transcription could regulate Xist expression, as is the situation in the imprinted H19-Igf2 gene pair which are transcribed in a mutually exclusive fashion. The Tsix R N A , like the H19 transcript, may be nonfunctional (Mlynarczyk and Panning, 2000). Differential methylation of an imprinting centre allows an enhancer element, necessary for both H19 and Ig/2, to act only on one of the genes depending on the chromosomal parent-of-origin. In contrast to what was initially thought about Xist regulation, the available results indicate that it is the differential epigenetic mark of the  31  Tsix, rather then the Xist, promoter which may be responsible for the imprinted inactivation in the extraembryonic tissues of eutherian mammals.  Spread of inactivation  Although X:autosome translocations are capable of inactivation, it is clear that the spread of inactivation is not as extensive as it would be on the X chromosome (Heard et al., 1997). Additionally, some autosomes and translocations seem more amenable to inactivation than others. Mary Lyon (1998), drawing on the earlier Gartler-Riggs model (1983), proposed that an explanation for this phenomenon could be the presence of certain repeated elements that are ubiquitous in the genome, but are enriched on the X chromosome. These repeats would act as "booster elements" or "way stations" allowing the inactivation signal to spread more efficiently. A good candidate for these repeat elements are the LINE 1 sequences (long interspersed repeat elements), which are particularly concentrated on the X chromosome in both humans and mouse (Lyon, 1998). LINE elements are retrotransposons that tend to cluster in dark G-bands and are present in all three subclasses of mammals, unlike other repeat candidates, which are specific to certain groups (for example, Alu elements are primate-specific) (Lyon, 1998). More recent research by Bailey et al. (2000) using available genomic sequence confirmed that the LINE 1 composition of the human X chromosome is distinct from that of the autosomes and that the highest concentration of LINES is found in Xql3-Xq21, which is the location of the human XIC. Furthermore, regions of the X chromosome carrying genes that escape inactivation tended to  32 harbour less L I elements than regions that were silenced. The authors also found that the greatest L I enrichment on the human X was due to a particular subset of eutherian-specific LINEs, which may account for the more stable silencing process in eutherians in comparison with marsupials. Another tantalizing clue regarding the presence and identity of "booster elements" is the observation that Xist R N A coating the inactive X in the mouse, rat and vole appears to have a banded appearance (Duthie et al.,1999), although the pattern seemed to correlate with the LINEpoor G-light bands. This observation may indicate that Xist does not directly associate with the repeats, but with the gene rich portions of the chromosome. Lyon (1998) proposes that the LINE elements may aid the spread of inactivation through repeat-induced silencing, a putative celldefense mechanism present in most eukaryotes. In this model the XIST/Xist R N A would act to bind and bring the LINE elements into contact with each other, thus masking intervening nonrepetitive sequences and triggering silencing.  Genes escaping inactivation  Despite the various factors involved in inactivation, a number of genes on the X chromosome manage to escape silencing. A recent study has determined that 34 genes out of 224 X-linked genes studied, three of which are pseudoautosomal, escape inactivation in humans (Carrel et al., 1999). The genes that avoid inactivation have acetylated promoter regions (Gilbert and Sharp, 1999), are not methylated at their CpG islands and do not replicate late (Heard et al., 1997).  Although genes that escape inactivation are dispersed throughout the human X  33 chromosome, a significant number are located around the large P A R (Heard et al., 1997). Since X-inactivation is thought to be a means of dosage compensation, it is logical that genes on the pseudoautosomal regions are not inactivated. Likewise, genes with functional Y homologues are not expected to be silent. However, a large proportion of genes adjacent to, but not part of the P A R escape inactivation in humans.  This failure to inactivate may be a result of the  relatively recent autosomal origin of most of the X p . Various authors have suggested that a portion of the X p sequences have not had time to acquire the putative repeat elements necessary for the propagation of the inactivation signal and that they behave more like autosomal material, which is less conducive to silencing (Lyon, 1998, Carrel et al., 1999, Bailey et al., 2000). A n additional evolutionary problem is presented by the several genes with no detectable Y homologues that also escape inactivation, since the presence of Y-linked copy is a strong indicator that a gene will escape inactivation (Jegalian and Page, 1998). It is possible that the recent loss of the Y locus has not allowed sufficient time and selective pressure to include the X-linked homologue into the dosage compensation machinery (Graves et al., 1998b). Carrel et al. (1999) have suggested that the presence of these genes may indicate that strict dosage compensation may not be necessary for all genes on the chromosome. The existence of genes, which are expressed from the inactive X chromosome in only some women, such as the Tissue Inhibitor of Metalloproteinases 1 gene (TIMP1) (Anderson and Brown, 1999), along with the observation that some genes in mouse reactivate with age, support this hypothesis (Gartler and Goldman, 1994, Brown et al., 1997). Not surprisingly, the genes that were shown to reactivate in older mice, did not have extensive CpG islands (Gartler and Goldman, 1994). In contrast to the situation on the human X chromosome, there seem to be remarkably few escapees in the only other well-studied eutherian - the mouse.  This difference may be  34  attributable to limited number of genes that were examined in mouse. The results of Carrel et al. (1999) indicate that the position of a genes (Xp versus Xq) has a statistically significant influence on whether a gene is inactivated or not, and since there are many rearrangements separating the mouse and human X , one would not expect that genes escaping in one species will do so in another. However, the phenotype of an X O (39,X) mouse is much less severe than that in humans with Turner syndrome (45,X) supporting the notion that fewer genes escape inactivation in the mouse.  While females with Turner syndrome show a greatly reduced  viability in utero and a number of phenotypic abnormalities, their murine counterparts only show a slight reduction in developmental viability and fertility (Banzai et al., 1995).  Silencing in the Methatheria and Prototheria and theories about the origins of Xinactivation  While many aspects of X-chromosome inactivation in eutherians remain a mystery, the situation in the marsupial and monotreme subclasses presents an even greater unknown. Late replication and histone hypoacetylation, both hallmarks of inactive chromatin, are conserved in the Metatheria (Cooper et al., 1993, Wakefield et al., 1997). However, marsupial X-inactivation differs from the eutherian model in several crucial ways. In the somatic and extraembryonic tissues of marsupials it is always the paternal X that is silenced, but a Barr body is observed in only a few species (Solari, 1993). Furthermore, this inactivation is often partial and genes may reactivate in particular tissues. Methylation of CpG islands on the inactive X in marsupials has not been detected by restriction enzyme digestion nor bisulfite sequencing (Kaslow and Migeon, 1987, Loebel and Johnstone, 1996). This lack of methylation as a maintenance system has been  35 hypothesized to be one of the reasons that inactivation in this subclass is unstable. The imprinted nature and instability of silencing in these ancient mammals may represent the ancestral state of X-inactivation. Perhaps the most significant difference is the apparent lack of an XIST homologue in the metatherians, despite intense efforts to locate one (Wakefield et al., 1997, Wakefield, 1998). The investigation of R N A secondary structure identified only two small regions of conservation between the human and mouse genes. In addition to the 5' conserved repeat region identified by Hendrich et al. (1993), Wakefield located another 94 base pair structure in exon four that seemed conserved at the secondary structure level between mouse and humans. The general lack of conservation led the author to speculate that a large portion of XIST/Xist may not be crucial for its function, is capable of performing the same function with a very different sequence and structure, or interacts with very different species-specific factors (Wakefield, 1998). Extensive screening of genomic and c D N A libraries from Australian and South American marsupials using a variety of probes from the two conserved regions proved unsuccessful, as were PCR screens involving degenerate oligonucleotide primers designed from the same sequence regions (Wakefield, 1998). Given the crucial importance of XIST/Xist in eutherian inactivation, this absence is truly baffling. X-chromosome inactivation in monotremes is the least studied of all three subclasses due to the fact that these animals do not breed in captivity and their cells are very difficult to culture. The monotreme sex chromosomes, as discussed earlier, share a large region of homology and the short arm of the X is thought to correspond to the marsupial and conserved eutherian X chromosomes. In addition to the incomplete differentiation of the gonosomes, the X and Y chromosomes are involved in a multivalent translocation chain with several small chromosomes  36 that are unpaired in both sexes (reviewed in Solari, 1993).  The only evidence that the  Prototheria undergo silencing at all is indirect. Wrigley and Graves demonstrated replication asynchrony of the short, but not the long arm in echidna lymphocytes (1988). This replication timing difference could not be duplicated in fibroblasts, indicating that the inactivation, i f it exists at all, is tissue specific. Lack of markers did not allow the authors to determine the parental origin of the late replicating chromosome. Significant differences exist between the imprinted silencing of the paternal chromosome found in eutherian extra-embryonic lineages and marsupial somatic tissue and random X inactivation present in eutherian somatic cells. Hypermethylation of CpG islands, an indicator of silencing in eutherians, is not seen in imprinted inactivation and may account for its relative instability (Kaslow and Migeon, 1987). Furthermore, a counting mechanism that determines the sex chromosome:autosome ratio and ensures that only one X per autosome set is active, is only present in random inactivation. In imprinted inactivation, all the paternal and none of the maternal X chromosomes will be inactivated, irregardless of the X:autosome ratio. The mark that protects the murine maternal chromosome from inactivation seems to lie within a CpG rich domain of Tsix and was thought to consist of a differential methylation pattern (Lee, 2000), although that is very much in doubt in light of the findings of Prissette et al. (2001). These findings raise interesting questions about the similarities of eutherian and marsupial imprinted inactivation, since marsupials do not seem to methylate X-linked sequences, but do employ histone modification (Kaslow and Migeon, 1987, Wakefield, 1997).  However, it is possible  that the eutherian parental imprint is in the form of differential acetylation (O'Neill et al., 1999, Sado et a l , 2000).  37 The exclusive occurrence of imprinted X-inactivation in marsupials has lead to the prevailing view that this may represent the ancestral form of mammalian X-chromosome silencing. However, based on new data regarding the origins of autosomal genomic imprinting Ohlsson, Paldi and Graves (2001) investigated the possibility that random inactivation may in fact represent the original situation.  Autosomal imprinting, in which a gene's expression is  dependent on the parent of origin, shows remarkable similarities to X-chromosome inactivation that indicate a common evolutionary origin. chromosome is late replicating, hypermethylated  In both processes the silenced allele or and hypoacetylated.  Furthermore,  phenomena result in functional hemizygosity and may involve a differentially  both  methylated  inactivation centre that produces an untranslated R N A associated with an antisense transcript. The authors postulate that imprinting arose from random monoallelic expression, which by introducing phenotypic variation may have presented a selective advantage, followed by the skewing of the inactivation process by the appearance of a parental mark.  They hypothesize  that monoallelic expression began as a stochastic event due to the low transcriptional activity of some genes.  In this stochastic inactivation scenario, some cells would have monoallelic  expression, others biallelic, while a number would be expected to be functionally null for the gene in question. Eventually, a counting mechanism ensuring proper gene dosage would lead to counted random inactivation. Imprinted inactivation would develop from the random model possibly due to the intrinsic evolutionary competition between the maternal and paternal genomes. A n example of stochastic inactivation is found in the IL2, IL3, IL4, IL5, and IL13 cytokine genes expressed from one or both parental alleles in T-lymphocytes or in the olfactory receptor genes. This stochastic inactivation becomes epigenetically fixed producing a stable monoallelic  38 expression pattern in the lymphocyte clones. New evidence showing that imprinted Igf2 gene is transcribed randomly or biallelically in some mouse fetal liver cells, hydatidiform moles and human tumours, indicates that random expression may underlie the normally imprinted status of this gene. X-chromosome inactivation, however, affects an entire chromosome and almost certainly arose as a means of dosage compensation.  The evolutionary progression from stochastic  through counted random to imprinted inactivation could not be tolerated since ensuring proper dosage, through counting or imprinting, must have been present from the very beginning. In combination with the complete absence of a random mechanism of inactivation in marsupials, it seems that the inactivation of the paternal chromosome was the ancestral condition. The functional quasi-heterozygosity for X-linked genes in the eutherian female may have been the selective force that led to the evolution of random inactivation in this mammalian subclass. Whether or not imprinted inactivation is the ancestral state of X-chromosome silencing, it is not clear why the paternal chromosome is exclusively affected. From an evolutionary point of view, paternal inactivation may be the result of the genetic conflict of interest between the maternal and paternal genes within the embryo (Hurst, 1997). A mechanistic explanation for paternal X-chromosome inactivation in the extraembryonic tissues of eutherians may be found in the fact that the XIST/Xist is expressed and the X chromosome silenced during male meiosis (Heard et al., 1997). Since the trophoblast is the first to develop and undergoes X inactivation before the epiblast, perhaps there is insufficient time to erase the inactivating epigenetic pattern acquired during spermatogenesis and the paternal X is silenced by default.  This may be  especially significant in rapidly developing animals such as the mouse and rat, which also show paternal inactivation in the extraembryonic tissues (Wake et al., 1976). The human embryo and  39 supporting tissues may simply have more time to develop and therefore undergo random inactivation.  Unfortunately, there are no data concerning the parental origin of the  extraembryonically inactivated chromosome in other eutherian mammals, which may shed light on the problem. Further support for this theory comes from studies on the appearance of macroH2A and its association with the inactive X to form the macrochromatin body (MCB). In imprinted inactivation in the trophoblast the M C B is formed at a much earlier stage than in ES cells, perhaps due to its association with the sex vesicle formed by the X and Y chromosome in the eutherian male meioses (Solari, 1993, Avner and Heard, 2001). Interesting evidence from cloned mouse embryos seems to support this theory (Eggan et al., 2000). Mice cloned from adult somatic cells exhibit random inactivation in the embryo proper, indicating that the epigenetic mark of somatic cell inactivation has been erased during the cloning process. However, the silencing pattern in the extra-embryonic cells is not random. The X chromosome in the female somatic cells that was induced to be inactive through cell culture techniques will be inactive in the trophoblast, possibly due to its inactive chromatin configuration at the time of transfer.  In contrast, mice cloned from ES cells, in which the two  X chromosomes are equivalent and theoretically lack the epigenetic mark, since they have not yet undergone silencing, will have a random inactivation pattern in both the trophoblast and epiblast tissues.  These results point to the existence of a gametic mark on the paternal X  chromosome, possibly in the promoter of Tsix (Lee, 2000), which predisposes it to inactivation and may be a result of its silenced status during gametogenesis. In the absence of the somatic or embryonic epigenetic mark, inactivation occurs randomly.  40  Beyond the human and mouse models of X-chromosome inactivation  It has now been well established that the XIST/Xist gene is essential for X-chromosome inactivation in mouse and humans, in contrast to the situation in marsupials where its very existence is in doubt. Considerably less research has been done on other eutherian mammals, but the few reports in cows, voles and rats seem to conform with the mouse/human model. The Xist R N A has been seen to coat the vole and rat inactive X chromosome (Duthie et al., 1999). The only non-rodent organism studied has been the cow and recent evidence from bovine embryos shows Xist expression in female embryos (De L a Fuente et al., 1999). Apart from the 5' fragments of lepine, equine, primate and bovine Xist sequenced for the evolutionary conservation studies of Hendrich et al., (1993), and a few small and scattered bovine, feline, porcine and rat ESTs, the only full XIST/Xist sequence is for the human, mouse and three voles (Brockdorff et al., 1992, Brown et al., 1992, Nesterova et al., 2001 in press). With the exception of the gene sequences, which have been deposited in GenBank, no other information regarding vole Xist is available to the public at this time (Nesterova et al., 2001 in press).  The  identification of conserved regions of the gene through mouse - human comparisons have not been helpful in the search for functional domains nor a marsupial homologue, indicating that it is time to look beyond the traditional human and mouse models. Although, Xist sequences have been recovered from all the eutherians examined so far, no studies have been done in the Insectivora, an order that is traditionally thought to represent an early eutherian lineage (Krettek et al., 1995). The only indication that X-inactivation occurs in this order comes from a study by  41  Jegalian and Page (1998), in which differential methylation was seen for three out four X-linked genes studied in the male and female European hedgehog. While the evolutionary relationships between the mammalian subclasses  are well  established, the phylogeny within the Eutheria is still a subject of considerable debate (Novacek, 1992, Cooper and Fortey, 1998, Cao et al., 2000, Madsen et al., 2001, Murphy et al., 2001). The eutherians are thought to have had two periods of rapid radiation, which have resulted in a starlike or bush-like phylogeny, which effectively means that all of the lineages are more or less equidistant from each other (Cao et al., 2000).  However, some authors have challenged this  star-shaped phylogeny and maintain that there are deeper branches and divisions than was previously thought (Cooper and Fortey, 1998). The oldest radiation occurred around 102 M Y A which coincides with the separation of the African and South American continents (Cao et al., 2000). The second diversification at the Cretaceous-Tertiary boundary (65 M Y A ) is thought to be the result of the demise of the dinosaurs and the availability of new environmental niches for the extant mammals (Cao et al., 2000). Despite the intense interest in the field, the eutherian phylogeny, especially at the deeper levels, still remains unresolved (Cooper and Fortey, 1998, Cao et al., 2000). An especially contentious problem has been the monophyly and position of the order Insectivora (Liptophyla) within the higher level relationships of the placental mammals. On the basis of morphology the insectivores occupied a basal position in traditional eutherian phylogenies and were thought to represent the closest living relatives to the ancestral placental stock (Novacek, 1992). According to this morphological view, which was the prevalent model until recently, the insectivores were a monophyletic group consisting of the Soricidae (shrews), Talpidae (moles), Erinaceidae (hedgehogs and gymnures), Solenodontidae  (solenodons),  42  Tenrecidae (tenrecs) and Chrysochloridae (golden moles) (Stanhope et al., 1998).  Three  independent studies based on nuclear and mitochondrial sequence have effectively shattered the monophyly of the Insectivora (Stanhope et al., 1998, Madsen et a l , 2001, Murphy et al., 2001). A l l of the studies have consistently placed the African insectivores, the tenrecs and golden moles, within the new order Afrotheria. Despite these significant results, the position and grouping of the remaining insectivores (shrews, moles, solenodons and hedgehogs), which now seem to be called the Insectivora, Liptophyla or Euliptophyla, still remains uncertain. Analysis of the complete mitochondrial D N A of the European hedgehog placed the insectivores basal to all the other eutherian sequences (Krettek et al., 1995).  However,  Mouchaty et al. (2000) rejected a monophyletic relationship between the hedgehog and the mole, based on the analysis of complete mitochondrial genome of the European mole. They showed the mole and shrew to be a sister group to the Chiroptera and proposed the dissolution of the Order Liptophyla. This group was supported by another study of complete mitochondrial genomes by Cao et al. (2000). On the other hand, the extensive studies of Murphy et al. (2001), which used 10,000 base pairs of nuclear and mitochondrial sequence, support the monophyly of the shrews, moles and hedgehogs (the Euliptophyla) and place the group in a superordinal clade (Laurasiatheria) along with the Cetariodactyla (even-toed animals and whales), Perissodactyla (cloven-toed animals), Carnivora, Chiroptera (bats) and Pholidota (pangolin). Laurasiatheria, meaning "from the area of Laurasia" or Europe, Asia and North America was suggested for this new clade, although no statistical support was provided. Similarly, Madsen and colleagues (2001) place the hedgehog, mole and shrew together and position them within a comparable superordinal group, based on approximately 8000 base pairs of nuclear and mitochondrial sequence. The results of Stanhope  43 et al. (1998) derived from several combinations of mitochondrial and nuclear genes also support the Liptophylan clade of shrews, moles and hedgehogs.  Interestingly, the results of both  Madsen et al. (2001) and Murphy et al. (2001) place the primates and rodents into the same superordinal clade, which never includes the core insectivores. Additionally, both studies find the core insectivores and the African insectivores (tenrecs, golden moles) at the base of each of their respective superordinal clades, an observation that is consistent with the traditional view that insectivore-like forms gave rise to the various eutherian taxa.  In a very recent, and  probably the largest, phylogenetic study concerning the eutherian radiation the authors used a novel approach and combined all the known molecular and morphological phylogenies into several parsimonius "supertrees" (Liu et al., 2001). In all three "supertrees" presented, the Soricomorpha and Erinaceomorpha formed a monophyletic group that was distinct and distant from both primates and rodents. A review of all the available phylogenetic studies concerning the core insectivores (Erinaceidea, Talpidae, Soricidae and Solenodontidae) reveals that the grouping is rejected by two analyses of complete mitochondrial genomes (Cao et al., 2000, Mouchaty et al., 2000), yet it is supported by morphological (Novacek, 1992) and nuclear data and combined nuclear and mitochondrial data (Liu et al., 2001, Madsen et al., 2001 Murphy et al., 2001). The higher rate of substitution (more informative sites) and smaller effective population size (shorter coalescence times) in comparison to nuclear protein-coding genes, has made mitochondrial D N A the genome of choice for studying closely related species (Springer et al., 2001). However, a recent paper by Springer and colleagues (2001) showed that nuclear exons are more efficient and consistently achieve greater resolving power on a per-residue basis in comparison with mitochondrial protein-coding or rRNA genes.  In an attempt to recreate several well-  44 established eutherian phylogenies using various sets of nuclear and mitochondrial data sets it was found that the nuclear genes were more successful in recovering the benchmark clades. Although mitochondrial D N A sequence are invaluable in resolving more recent divergences, it seems that nuclear genes should be used when recovering deep mammalian divergences, especially if the mitochondrial data does not agree with the morphological evidence. In conclusion, the authors argue that the most valuable data set would be one composed of both nuclear and mitochondrial sequences, since they provide two independent estimates of the phylogenetic relationship. In light of this study, the monophyly of the core insectivores, which is supported by all the current phylogenetic studies, except those relying exclusively on mitochondrial data, may prove to be correct. Although the phylogenetic position and monophyly of the Insectivora has not been unequivocally established, molecular evidence seems to support a basal position for the group, at least within the Laurasiatheria. More importantly for the purposes of this study, none of the analyses show a close phylogenetic relationship between the Euliptophylans and primates or rodents, indicating that Xist sequence from a member of this order may provide new information regarding the evolution of the gene. The identification of the X-chromosome inactivation process and the Xist gene in an insectivore would indicate that Xz'sr-mediated silencing evolved before the eutherian radiation and is shared by all members of this subclass.  Furthermore,  insectivore Xist sequence may provide insight into the function of the gene and may indicate areas of conservation that may facilitate the identification of the elusive marsupial homologue.  45  Chapter 2  Materials and Methods  Mole samples and cell lines  Prior to the establishment of Coast mole cell lines I had been using mole D N A kindly given to me by Dr. Kevin Campbell (at the Department of Zoology at the University of British Columbia) and had some frozen organs from several moles found dead on the road or in mole traps. The R N A extracted from these samples was either nonexistent or of very poor quality. Fresh mole samples were finally obtained from three female moles sacrificed in a study by Dr. Campbell. Additional animals were acquired in the spring of 2000 during a short mole hunting "field season" - an endeavor which would have been impossible without the help of Dr. Campbell and Tim Sheehan. R N A was extracted from these new samples, which were also used to create Coast mole cell lines that served as additional source of R N A . Unfortunately, most of these moles were females and only one male animal, found in a trap, was sufficiently preserved to allow the establishment of a cell line. Information on the identity and history of all the mole samples is presented in Table 1. The mole cell lines, all of which were primary cultures, were prepared from fresh kidneys, which were placed in Minimal Essential Medium (a M E M ) (Gibco/BRL) immediately after collection. The organs were then rinsed with 70 % ethanol in the tissue culture hood to reduce bacterial contamination, followed by a rinse in a M E M . The tissue was then diced on a  46  Comments  Coast mole  Source  Date  CM1 (female)  Dr. K. Campbell  May 1998  -  Received DNA only  CM3 (male)  Dr. K. Campbell  May 1999  -  Found dead on a road  CM4 (male)  Dr. K. Campbell  Feb. 2000  -  CM5 (male)  Dr. K. Campbell  May 1999  -  CM6 (male)  Dr. K. Campbell  May 1999  -  CM7 (female)  Dr. K. Campbell  May 1999  -  CM8 (male)  Captured *  March 2000  -  Accidentally killed in the field  CM9 (male)  Captured *  March 2000  -  Accidentally killed in the field  CM 10 (female)  Captured *  March 2000  -  Drowned  CM 11 (female)  Captured *  March 2000  +  Sacrificed  CM 12 (female)  Captured *  March 2000  +  Sacrificed  CM 13 (female)  Captured *  March 2000  +  Sacrificed  CM 14 (male)  Tim Sheehan  May 2000  -  Found in trap (dead <24 hours)  CM 15 (male)  Tim Sheehan  May 2000  +  Found in trap (dead <1 hour)  CM 16 (female)  Tim Sheehan  June 2000  +  Found in trap  CM 17 (female)  Tim Sheehan  June 2000  +  Found in trap  C e l l Line  * Captured in the field with Dr. K . Campbell and T. Sheehan  Table 1: Coast mole samples  47 coverslip using a sterile razor blade and transferred to a fresh tube with lOOOunits/mg collagenase (Sigma), an enzyme, which destroys the tissue matrix, and releases the individual cells. After 20 minutes of incubation at 37° C, the sample was briefly centrifuged (3500 rpm). Once the supernatant had been removed, the cell pellet was rinsed in a M E M and centrifuged once again.  The cell pellet was then incubated with 0.1 ml of 0.25 % trypsin-EDTA  (Gibco/BRL) at 37° C for 5 minutes. Following this incubation, the cells were centrifuged, rinsed with a M E M and resuspended in 200-500 p:l of a M E M supplemented with 7.5 % fetal calf serum (Cansera), 1 % M E M nonessential amino acids, 1% L-glutamine and 1% penicillin and streptomycin (all from Gibco/BRL). Before the cells were plated trituration was used to break up any remaining cell clumps. The cells were incubated at 37° C, 5 % C O 2 and 90 % humidity. To obtain cell pellets for the extraction of genomic D N A , the medium was removed, the cells were washed with phosphate saline buffer (PBS), trypsinized (0.25 % trypsin-EDTA) and resuspended in fibroblast medium. The Coast mole cells proved very hardy and had to be scraped off the plates after trypsinization. The cell suspension is then centrifuged and the supernatant removed leaving the cell pellet. The protocol for freezing cell lines was very similar to that for producing cell pellets, except that the cells were resuspended in fibroblast medium enriched with 15 % fetal calf serum and 10 % D M S O (dimethyl sulfoxide) after trypsinization. The cell suspension was then placed in an isopropanol filled container at -70 °C and allowed to cool slowly, after which they are transferred to liquid nitrogen for long term storage.  48  Extraction of genomic DNA  In order to extract genomic D N A , cell pellets or homogenized tissue were placed in TrisE D T A (TE) buffer (pH 7.5-8) and incubated over night at room temperature with l/20 volume th  of 20 % SDS (sodium dodecyl sulfate) and a pipette tip of proteinase K - a proteolytic enzyme active against a broad spectrum of proteins.  These solutions serve to rupture the cell  membranes and digest proteins, thus releasing the nucleic acids. The following day 1/10  th  volume 5 M NaCl was added, the tube was shaken vigorously and spun for 15 minutes (2500 rpm). The supernatant was then removed from the pellet of denatured proteins. Additional salt (5 M NaCl, one third of the original volume) and SDS (l/30 of original volume) were added, th  the preparation was again shaken vigorously and spun at 2500rpm for 15 minutes.  The  supernatant was then transferred to a clean tube and two volumes of 95 % ethanol were added to precipitate the D N A , which was carefully spooled, resuspended in double distilled water (ddH20) and stored at 4° C.  RNA isolation  Total R N A was isolated using the acid-guanadinium-phenol-chlorofom R N A protocol, also known as the RNAzol B method, developed by Chomczynski and Sacchi (1987). The guanadinium and water molecules form complexes with the R N A , abolishing the hydrophilic interactions with the D N A and proteins, thus the D N A and proteins are removed from the aqueous phase, in which the R N A remains. Phenol acts as a denaturing and deproteinating  49  agent, while chloroform removed any traces of phenol from the preparation. The first step in the protocol involved the homogenization and vortexing of tissue in Solution D (4 M guanadinium thiocyanate, 25 m M sodium citrate pH7, 0.5 % sarcosyl, 0.1 M 2-mercaptoethanol (added just before use).  Normally 0.6 ml of Solution D were used per 100 mg tissue or a  confluent 60 mm tissue culture dish. Homogenization was not necessary for tissue culture cells since vortexing is enough to disrupt the cell membranes. After the addition of equal volume of diethyl pyrocarbonate-treated (DEPC) ddH20 saturated phenol and 1/10 volume 2 M sodium th  acetate (pH4), the sample was mixed by inversion and vortexing.  Chloroform (0.4 volume)  was added to the solution and the mixture was vortexed and placed on ice for 5 to 15 minutes. After centrifugation (10 minutes, 13500 rpm) the RNA-containing supernatant was removed, while the proteins and D N A remained in the lower organic layers. After the addition of an equal volume of isopropanol, the solution was left at -20° C overnight. A 10 minute centrifugation at 13500 rpm precipitated the R N A , which could then be washed with 70% ethanol and reprecipitated (10 min, 13500rpm). The air-dried R N A pellet was resuspended in 10-100 |il of DEPC-ddH20 and stored at -20° C.  DNAse treatment of RNA  R N A in DEPC-ddH20 was combined with porcine RNAsin (l/20 Amersham Pharmacia Biotech) and RNAse-free DNAse (1/10 technologies) and incubated at 37° C for one hour.  th  lh  total volume,  total volume, Epicentre  The volume of the solution was then  increased to 200 | l l with DEPC-ddH20 and an equal volume of 1:1 phenol-chloroform was  50 added to denature any proteins. The sample was then vortexed, placed on ice and centrifuged for 5-10 minutes at 12500 rpm. The aqueous upper layer was transferred to a new tube where an equal volume of chloroform was added to remove any traces of phenol. The sample was vortexed and centrifuged once again. The aqueous phase was transferred to a new tube to which 0.15 volume of 2 M sodium acetate in DEPC-ddH20 and an equal volume of isopropanol were added to precipitate the D N A (-20° C, overnight). The following day, after centrifuging, the supernatant is discarded and the R N A pellet was resuspended in DEPC-ddH20.  The Polymerase chain reaction (PCR)  Approximately 100 ng of template were added to a 25 | i l PCR reaction containing 20 u M each dNTP, 1.5 m M M g C l , 1 ul 10X PCR buffer, 1 u M primer and 0.625 units Taq D N A 2  polymerase (all reagents , with the exception of the primers, are from Gibco/BRL). The reactions were overlayed with 20 ul mineral oil and forty cycles of 1 minute at 94° C, 1 minute at 50° C and 2 minutes at 72° C were performed in a Techne Genius Thermal Cycler. Unless otherwise specified, these conditions were used in all the PCRs. The amplification of the CpG islands in the methylation assay required a change in the MgCl2 concentration and the addition of betaine (Sigma), which improves the PCR amplification of GC-rich sequences by reducing the formation of secondary structure (Henke et al., 1997). Long PCR, using the ExpandLong Template system (Boehringer Mannheim), was performed on fragments greater than 1 kilobase. Two 25 ul reaction mixes, one containing the enzyme mix (2.5 units), 5ul 5x Buffer 3 (0.75 m M MgCb.) and ddH20, and the other composed  51  of 0.5 m M of each dNTP, 0.3 u M of each primer and approximately 200 ng of genomic D N A , were combined and overlayed with 30 ul mineral oil. The mix was incubated at 94° C for two minutes and then cycled forty times (30 seconds at 94° C, 30 seconds at a 50-54° C, elongation at 68° C). The elongation times were varied according to the manufacturer's recommendations and ranged from three to eight minutes. A final elongation step was performed at 68° C for seven minutes.  Inverse PCR  Inverse PCR is a useful method for obtaining additional sequence flanking a known fragment. In this technique, genomic D N A is digested with a restriction enzyme that does not cut within the known area and then ligated. Primers are designed from the known sequence, but pointing out of the region, and are used for P C R amplification of the ligated D N A (see Figure 3). Coast mole 10 genomic D N A (1 ug) was digested with the appropriate enzymes overnight in a 10 or 20 pi reaction. Restriction enzymes with a four base pair recognition site, such as Alul, are particularly useful since they produce relatively small fragments that can be PCR amplified. The following day the digest was combined with 4000 units of T4 D N A ligase (NEB), 50 ul 10X ligase buffer and ddH20 was added increase the volume to 500 ul. The ligation reaction was left at room temperature overnight. The D N A was precipitated with the addition of salt and two volumes of 95 % ethanol, resuspended in 10-20 u.1 of ddH20 and 1-2 ul of this suspension were used in a PCR reaction (see above for details on ethanol precipitation).  52  53 A l l the PCR products were run on 1-2.5 % agarose gels, stained with ethidium bromide and visualized under U V light. P C R primers were selected manually and the oligonucleotides were made at the N A P S facility at the University of British Columbia. The primer sequences, sources and amplification conditions for the primers designed from conserved XIST/Xist sequence, mole-specific primers and non-XIST/Xist primers are presented in Figure 4 and Table 5 (Chapter 4), and Tables 2 and 3, respectively.  Reverse transcription  Five nanograms of R N A in DEPC-ddH20 were mixed with 5X first strand buffer (1 /5th volume, Gibco/BRL), 0.1 M DTT (dithiothreitol, l/10 volume), 1.25 m M deoxynucleotide th  triphosphates (dNTPS, 1 /20 volume, Gibco/BRL), l/20 volume random hexamer primers and th  l/20  th  th  volume M - M L V (Moloney Murine Leukemia virus) reverse transcriptase (MLV-RT,  Gibco/BRL) and incubated at 42° C for two hours. The RT enzyme uses the presence of R N A to synthesize a complementary D N A strand, which can serve as a template in PCR. Although the principle is the same, the Superscript (Gibco/BRL) reverse transcriptase requires slightly different conditions, which are described in detail in the Gibco/BRL product manual (Cat No. 18064-014).  54  COW(AF10496)  TTGCTGCAGGGACAATATGGC  COW(AF10496)  ATCCGACCCCAGCATTAGCC  HORSE(U50911)  TTGCCGCAGGGACAAIATCGC  HORSE(U50911)  ATCCGACCCCAGCATTAGCC  RABBIT (U50910)  TTGCCGCAGGGACAATATGGC  RABBIT (U50910)  ATCCGACCCCAGCATTAGCC  HUMAN (M97168)  TTGCCGCAGGGACAATATGGC  HUMAN ( M 9 7 1 6 8 )  ATCCGACCCCAGCATTAGCC  SKXIST2  ATCCGACCCCAGCATTAGCC  COW(AF10496)  CCTAATATCCAGCACTGCAAG  SKXIST1  TTGC (T/C) GCAGGGACAATATGGC  COW(AF10496)  GCCACTTGCAGTGCTGGATA  HORSE (U50911)  GCCTCTTGCAGTGCTGGATA  RABBIT (U50910) HUMAN (M97168)  HORSE (U50911)  CCAGCTATCCAGCACTGCAAG  GCCTCTTGCAGTGCTGGATA  RABBIT (U50910)  CCAGATATCCAGCACTGCAAG  GCCTCTTGCAGTGCTGGATA  HUMAN(M97168)  CCAGATATCCAGCACTGCAAG  SKXIST3  GCC ( A / T ) CTTGCAGTGCTGG  COW(BE483406)  TAATAGCTACAAAGTAGTACTTAC  HUMAN(M97168)  TAATAGCTACGAAGTAGTACTTAC  SKXIST6  TAATAGCTAC (G/A) AAGTAGTACTTAC  SKXIST4  CC(T/A)  ( A / G )A T A T C C A G C A C T G C A A G  COW(BE483406)  ACCACACTCAAGTGAGGACTT  HUMAN (M97168)  ACCACACTGAGGTGAGGACTT  SKXIST7  ACCACACT ( G / C )A (G/A) GTGAGGACTT  COW(AV617038)  TTTNACGATTCCTAGGTGGA  COW(AV617038)  TAGCATTGCAGG-CATGTTGA  H U M A N ( M 9 7 1 6 8)  TTTGACGATCCCTAGGTGGA  HUMAN (M97168)  TAGAGTGCCAGG-CATGTTGA  MOUSE(L04961)  TTTGACGATCCCTAGGTGGA  MOUSE ( L 0 4 9 6 1 )  TAGGGTAGCAGTGCATCTTGA  SKXIST9  CAT(AF197966) HUMAN (M97168) SKXIST11  TTTGACGAT  (T/OCCTAGGTGGA  ACAAGAG-TAC-CAGATGTGCC ACAAGAAATACACAGATGTGCC  ACAAGA (A/G)ATACACAGATGTGCC  HORSE (U50911)  TCACTTCTTAAAGCGCTGCA  RABBIT(U50910)  TCAATTCTTAAAGCGCTGCA  HUMAN(M97168)  TCAGTTCTTAAAGCGCTGCA  SKXIST13  TCANTTCTTAAAGCGCTGCA  SKXIST10  TAGN (G/A)T N (G/C) CAGGCATGTTGA  CAT(AF197966)  CCTTGGGGCCTCACTCTGTC  HUMAN(M97168)  CCTTGGGACCTCGCTTTGTC  SKXIST12  CCTTGGG(G/A)CCTC(A/G)CT(C/T)TGTC  Figure 4: Conserved Xist primers. The locations of the primers are shown in Figures 8,12,13 and 15. The exact locations and amplification conditions of the primers are presented in Table 5 (Chapter 4).  55  Primer  Sequence  Conditions  CMXIST1  cattgctgaagtggcctgagg  50-54° C, 40X  CMXISTREV1  cctcaggccacttcagcaatg  50-54° C, 40X  CMXIST2  gttcctcttgagaaaggcaagg  50-54° C, 40X  CMXISTREV2  ccttgcctttctcaagaggaac  50-54° C, 40X  CMXISTREV3  aaggccaattaatgagttca  50-54° C, 40X  CMXISTREV4  tccatactgccttatcactg  50-54° C, 40X  CMXISTREV5  atcaggcaacaactcactgc  50-54° C, 40X  CMXISTREV6  gaacagcagttctttgtaatc  50-54° C, 40X  CMXISTREV7  cctcaattccgtgactgtag  50-54° C, 40X  CMXISTREV8  gatactagagtaactgcagcg  50-54° C, 40X  CMXIST10  caggtggagttgataacctgg  50-54° C, 40X  CMXIST11  agcactgctcagaagcaatgc  50-54° C, 40X  Comments Positive for expression with CMXIST2  Positive for expression with CMXIST1  Positive for expression with CMXIST10  Positive for expression with CMXISTREV5  Table 2: Primers designed from Coast mole Xist sequence. The locations of the primers are indicated in Figures 9, 11, 12 and 13. The " R E V E R S E " primers are referred to in the text as C M X I S T R E V E R S E . For example, CMXIST1 R E V E R S E is abbreviated to C M X I S T R E V 1 .  o  3 O  CN  Os  o fi oo  s© OS  O o Os D. m W3  cu  E S  •2 >n  |5  •a CN CN  u-s  "5. E-s  rt  <L>  o  to  -  3  £  ~  IS.'* c _ IS  •a u  00  60  o  J  &E  <C eti T3 cj ^  •El  c J3 H  c  i5  o  E  >>  ~o  CJ  cj  PH  PH  E  c  o O o O  e  a.  ° S E |  <u o  a.  U  2 o\  8 X  u  cj  3  cd os  111  <S1 - H  U  2" X 2 X  •S o cj  f  2  CN  O  o  & XJ O lO CN  o  ^7  CJ  "3  I* OS  ^  o  CU CG  a. CA  — i > CM  X o  r  CJ  c CJ a CN <n os  & 00 M CJ =8  32  'o  a u  a  T3 C OO sj Os j - Os  .2  J 3 !/->  &  o u  e  T3 CJ  ID  o  B-  E •c  Co  £3 oo cd Os  e o < uo  X2  o 3  E  •a  a.  CN g CN j *  E S c  cu  w s-  -  CO CJ  CL  T3  a o  E  sj  ID  •a e  a o. u  O -H in M tS O CN m u-> m  E  s E  o  §  w  IS  a*  c  cj  o  a «cj s)  e  w  " .23 u u u >•  T3 CJ  M  SO  oo o r~ o •- — c -o CJ u o c o u C  JD  •o  cj  s § .§  O ^  ID  X>  1? CO  CJ  CO CJ  aa u  C  CJ U  •5 > S9 t~ "8 .2?oo x> a  <L>  o >-/>  'E a  S §  <o -S >  3 O  r  r  CN  X o U o o in  o  ^ oo  CN  IH  m -  (  C  (  oo H CJ  60 o o 00 u bo O f • • eg  B CU  a c ^  oo o oo oo £ oo S o o  ca  £ Q  3 t5 CB 03 o  Co  00 tt  u o  CJ  CO  CJ  o  o  00  S S> |)|>  CJ  o  CJ  oo  a  CJ  tj  CJ  CJ  £ S  bo o oo ea o cj  SB u c5  £ oo o o 00 °  CU  £  i IE  » CN »  X X  o  00 o o to CO 00 o bp cS° •&  < m  —< CN  —i CN  X X u. PH N  N  X X  oo  —i CN  o oo oo  00  & g8 8>i - H CN  Pd  QQ  on CJ  PH PH  a-a  " °  bo 00  I-  oo ti  C/) CJ  H H  1^  57 Methylation-sensitive restriction digests  A reaction mix containing 20-200 ng of genomic D N A , 5X DBS (5 m M DTT, 0.5 mg/ml bovine serum albumin and 10 m M spermidine), 1 | i l 10X of the appropriate buffer and the restriction enzyme (10 Units) was incubated at 37° C overnight. The enzyme was heatinactivated in a 10 minute incubation at 65° C the following day. One microliter of the digest was used in P C R amplifications using primers flanking CpG sites and the products were visualized under U V on an agarose gel stained with ethidium-bromide. The methylation sensitive enzyme Hpall (NEB) was used to evaluate the methylation status of the CpG sites, while Hindlll (Gibco/BRL) served as a negative cutting control.  DNA purification for sequencing and Xist sequence analysis  Gel extraction and purification of P C R fragments was performed using the QIAquick PCR purification kit (QIAGEN Cat. No.28704), according to the enclosed protocol (page 24). Briefly, the P C R band was excised from an agarose gel stained in ethidium bromide and combined with a solubilization buffer (Buffer QG). The liquefied agarose was then placed in a silica-gel membrane spin column, to which the D N A should adsorb, and centrifuged. Additional buffer Q G was added to the column and centrifuged to remove all traces of agarose. The column was washed several times in an ethanol based buffer (buffer PE) to remove any salts and the D N A was eventually eluted in 30 to 100 | i l ddH20. The purified fragment was sent to the NAPS laboratory at U B C for sequencing.  58 Homology searches were performed using the B L A S T programs (Altschul et al., 1990), while the sequence alignments were done in C L U S T A L W v. 1.6 set to D N A and using IUB matrix (Thompson et al., 1994) and available at the Canadian Bioinformatics Resource site (  The alignments  were  not  manually improved. The  ORF-Finder  ( and GRAIL programmes (Uberbacher et al., 1991) were used to search for open reading frames.  59 Chapter 3  X-chromosome inactivation in Scapanus orarius  Introduction  The Coast mole (Scapanus orarius)  The focus of this thesis is the study of X-chromosome inactivation and the Xist gene in the Coast mole, Scapanus orarius (Mammalia, Insectivora, Talpidae, Talpinae) (Figure 5). Before any attempts at identifying an Xist homologue were made and in order to evaluate any future results, it was necessary to determine whether these animals underwent X-chromosome inactivation. The detection of X-chromosome silencing in Scapanus is the topic of this chapter. The Coast mole (Figure 5) is a representative of the family Talpidae whose members are distributed widely throughout North America and Eurasia and include moles, shrew-moles and desmans (Yates and Moore, 1990). The unique pectoral anatomy necessary for the burrowing lifestyle and other traits have provided morphological evidence for the monophyly of the talpid group whose members enjoy fossorial, terrestrial and aquatic lifestyles. The Coast mole, also known as the Pacific mole, belongs to an exclusively terrestrial genus and is found from the Fraser Delta in British Columbia in the north to the coastal areas of northwest California in the south (Nagorsen, 1996). Like all fossorial moles, the Coast moles spend most of their lives underground, preferably in well-drained and soft soil, and are solitary except during the mating season (Yates and Moore, 1990).  Figure 5: The Coast mole, Scapanus orarius (courtesy of S. T i m Sheehan).  61 Although superficially similar to the only other underground mammals, the fossorial rodents, the talpids are a much older and more widely distributed lineage. The family was widespread and diverse during the Oligocene, while the oldest fossils date to the late Eocene (approx. 36 M Y A ) in Europe and the Miocene (approx. 24 M Y A ) in North America (Yates and Moore, 1990). Additionally, unlike the subterranean rodents, which are well known to exhibit high levels of chromosomal variation and polymorphisms, the talpids have a remarkable degree of chromosomal conservation despite their antiquity. The vast majority of examined species, including S. orarius, have a diploid number of 34 including an X X or X Y pair (unpublished data in Yates and Moore, 1990).  The Y chromosome is minute (Yates and Moore, 1990) and  synapses with the X chromosome during spermatogenesis to form the X Y body (Solari, 1993). Sex determination mechanisms in the mole, which have only been extensively studied in the European mole (Talpa europea) and other members of the genus Talpa, conform to the mammalian X X , X Y system. The reason for the intense interest in sex determination of the European talpids was due to a large number of apparently hermaphroditic X X individuals. Initial studies indicated that these animals had functional ovotestes (Jimenez et al., 1993, Sanchez et al., 1996a), but these results were later challenged after no evidence of spermatogenesis was found (Whitworth et al., 1999). Further experiments showed that all the male European moles carry the Sty gene (Sanchez et al., 1996b) and that the "hermaphrodites" were chromosomally normal females. Although the testicular tissue of the ovotestis does not produce viable gametes, the presence of bilateral ovotestes in all the females of this species makes them unique among mammals. Additionally, the ovarian and testicular components of the female ovotestis seem to fluctuate in size seasonally. The ovarian portion is increased during mating season, while the testicular region, which produces the androgens responsible for the  62  observed masculinization, is at its largest during the autumn when the increased aggressiveness and territoriality of the female may carry significant evolutionary benefit. Sanchez et al. (1996b) obtained the Sty DNA-binding H M G box sequence from other mole species, as well as from the Algerian hedgehog and several shrew species, and found that the insectivore sequences bear more resemblance to the human than to the mouse homologue. This analysis also confirmed the long-standing view that the moles and shrews are very closely related (Suborder Soricomorpha), while the hedgehog may represent a more distant group (Suborder Erinaceomorpha) within the Euliptophyla (Sanchez et al., 1996b). Although the karyotypes of several mole species have been published (Yates and Moore, 1990), there have been no reports regarding the presence of a Barr body or any other indications of X-chromosome inactivation. The absence of the Barr body is not surprising since it is not seen in many mammals, especially those with large blocks of heterochromatin at other locations of the genome, and is not a very reliable indicator of X silencing (Solari, 1993). In the absence of a Barr body, X-chromosome inactivation can be established through molecular methods such as assaying the expression of various X-linked genes in cell culture, determining the methylation status of genes on the X , observing replication timing differences between the two X chromosomes or performing fluorescence in situ hybridization (FISH) for Xist or other X linked RNAs.  To establish the existence of X inactivation in a somatic cell culture it is  necessary to obtain sequence polymorphisms to distinguish between the two alleles present in the genomic D N A and determine which one is expressed. In the absence of a polymorphism, one can use somatic cell hybrids in which the X chromosome of the species of interest is retained in the chromosomal background of the host species.  Gene expression can then be  assayed through RT-PCR (see Materials and Methods) to determine whether the retained  63 chromosome is active or inactive.  To perform FISH studies, gene sequence has to be known in  order to design the nucleic acid probes.  Since neither mole somatic-cell hybrids, polymorphic  markers nor X-linked gene sequence were available, we decided to investigate the existence of X-chromosome inactivation in the Coast mole by examining the methylation status of X-linked genes in the male and female animals.  DNA methylation in the mammalian genome  D N A methylation is an epigenetic modification that only occurs at the 5' cytosine in a CpG dinucleotide in the mammalian genome in which approximately 70 % of all CpGs are methylated (for review see Goto and Monk, 1998, Robertson and Jones, 2000, Robertson and Wolffe, 2000). The CpG dinucleotide is significantly under-represented in the genome due to the increased spontaneous deamination rate of the 5-methyl-cytosine to thymine, a naturally occurring D N A base, and the inefficient repair of this deamination product. However, the resulting CpG depletion is not randomly distributed and there are regions of the genome in which the dinucleotide can be found at the expected frequency. These regions (known as CpG islands) occur at the 5' end of genes and, although they represent only 15 % of the total genomic CpG sites, account for 50 % of the unmethylated CpG dinucleotides. The approximately 45,000 CpG islands identified are normally unmethylated and are accompanied by an open chromatin structure and gene expression. Apart from the imprinted genes, the only exceptions to this rule are the CpG islands of inactivated genes on the silenced X chromosome, which are methylated.  64  Methylation was not thought to be present in the genomes of the two invertebrate model organisms, C. elegans and D. melanogaster, however a small number methylated cytosines have recently been identified in the fly (reviewed in Lyko, 2001). Most of the methylation was in the context of CpT and CpA dinucleotides and was restricted to a very early stage in development. Although it is not known whether these methylation patterns have any functional significance, they certainly explain the presence of a D N A methyltransferase (Dnmt2) and methyl-binding protein homologues (dMBD2/3), which had previously baffled scientists. Methylation is of vital importance in mammals, although the reasons for its existence are not completely understood.  One view, that may have been put in jeopardy by the recent  discoveries in Drosophila, holds that methylation serves to control transcriptional noise and is only necessary in large, complex genomes. Another theory proposes that methylation arose as a silencing defense mechanism against invading D N A sequences. This interpretation is bolstered by the observation that most of the hypermethylated CpG dinucleotides reside within parasitic DNA  elements, such as retrotransposons,  and lead to their transcriptional silencing.  Additionally, kangaroo hybrids show a dramatic expansion of endogenous retroviral-like elements, which is thought to result from the marked hypomethylation of the hybrid genome, lending support to this hypothesis (Waugh O'Neill et al., 1988). Apart from containing the parasitic D N A elements, there is evidence that methylation might stabilize the genome by masking large numbers of homologous repeats and preventing inappropriate recombination through a currently unknown mechanism (reviewed in Roberston and Wolffe, 2000). Although mouse primordial germ cells and undifferentiated ES cells seem to function without D N A methylation (Lei et al., 1996), several targeted mutations of the genes involved in methylation, the D N A methyltranferases, have shown that methylation is necessary for survival  65 once cellular differentiation begins (Li et al., 1992, Okano et al., 1999). It is believed that D N A methylation, along with other chromatin modifications, functions to divide the genome into active and inactive compartments, thus enabling the transcription machinery to focus on a limited number of genes necessary for a particular cell's survival (reviewed in Robertson and Wolffe, 2000).  Methylation levels and patterns are tissue-specific and vary developmentally.  During gametogenesis most parental methylation marks are erased and the gamete D N A is generally hypomethylated, although sperm methylation levels are higher than those of the ovum (reviewed in Goto and Monk, 1998). In pre implantation embryos, there is a further loss of methylation leading to an almost globally demethylated state. However, some methylation marks, such as those on the CpG islands of imprinted genes, seem to be more resistant to demethylation and can persist until the blastocyst stage.  Upon implantation and during  gastrulation new epigenetic patterns are set down by de novo methylation and maintained throughout the lifetime of the organism by the maintenance methyltransferases. In mammals, D N A methylation is achieved through the action of the D N A methyltransferases (DNMTs), some of which are capable of de novo methylation, while others, the maintenance methyltransferases, just duplicate the set methylation patterns. So far, three independent methyltransferases have been identified: D N M T 1 , D M N T 3 A and DNMT3B. DNMT1 is the most abundant methyltransferase in somatic cells and is necessary for proper embryonic development, imprinting and X-chromosome inactivation (Li et al., 1992, Beard et al., 1995, Panning and Jaenisch, 1996, Sado et al., 2000).  Based on the observations that this  enzyme has a great affinity for hemimethylated D N A and that it is associated with replication foci, D N M T l i s believed to be the principal maintenance methylase.  66 The DMNT3 group of methyltransferases (reviewed in Robertson and Wolffe, 2000) show an equal preference for hemi- and unmethylated D N A and are thought to be the de novo methylases responsible for the wave of methylation following embryo implantation, as well as for the de novo methylation of newly integrated retroviral sequences in mouse ES cells. Mouse knockouts of Dnmt3b are lethal, while Dnmt3a minus mutants are runted and survive only a few days.  New evidence, however, suggest that all three methyltransferases may possess both de  novo and maintenance capabilities and that the division into de novo and maintenance categories may not be valid. Although the connection between methylation and silencing has been well established, the mechanism is not completely understood (reviewed in Robertson and Wolffe, 2000). Early experiments showed that methylated CpG dinucleotides directly interfered with transcription factor binding sites. However, this mechanism could not account for all the observed silencing, since only a subset of transcription factors contain CpGs in their recognition sequence. The second mechanism is mediated through the methyl-binding protein, MeCP2 (Lewis et al., 1992). Jones et al. (1998) have shown that methylated D N A , through MeCP2, has the ability to recruit histone deacetylases, thus providing a link between methylation and inert chromatin structure (see Chapter 1 for more detail).  67 Methylation on the inactive X chromosome  The methylation at the CpG islands of inactivated X-linked genes is a well- established feature of the inactive X chromosome in eutherian mammals (Heard et al., 1997) and is frequently used to determine whether a gene is inactivated (Luoh et al., 1995, Carrel and Willard, 1996). The methylation status of a gene can be determined by bisulfite sequencing or by a combination of methylation-sensitive restriction-enzyme digestion followed by PCR (or Southern analysis), which is the method used in this study. Genomic D N A is digested with methylation-sensitive enzymes, such as Hpall or Hhal, which are able to cleave D N A only i f their recognition sequence is not methylated at the CpG dinucleotide. In a subsequent step, the polymerase chain reaction (PCR) is performed using primers that flank the enzyme cut site. Only methylated D N A will remain undigested and will provide a template for the PCR, resulting in a band visualized by agarose gel electrophoresis.  Thus, the PCR will only be  successful i f a gene is methylated and inactive. In the course of this assay it is necessary to perform a "mock" restriction digest using an enzyme that does not cut within the sequence of interest, to ensure that the lack or presence of a PCR fragment is due to the restriction digest and not a PCR artifact. Of special significance to the study of inactivation in moles is the observation that CpG island methylation correlates perfectly with the transcriptional state of a gene in a broad range of eutherians, including an insectivore representative, the European hedgehog (Jegalian and Page, 1998). Jegalian and Page (1998) studied the inactivation status of four X-linked genes (SMCX, RPS4X, ALD and ZFX) in large group of eutherians mammals that included one insectivore, the 2  Upper case notation will be used for the genes in the methylation analysis, with the exception of the murine genes.  2  68 hedgehog. They found that silencing through methylation at CpG-islands was conserved in all the species observed. Additionally, they observed that genes escaping X-inactivation, such as ZFX, were very reliable indicators of the presence of a Y-linked homologue.  The authors  concluded that the decay of the Y homologue seems to be the necessary step for a gene to be included in the X dosage compensation pathway, and seems to proceed on a gene-by-gene, rather than regional, basis.  O f the four genes studied, ALD was inactivated in all species  observed, while SMCX, ZFX and RPS4X were silenced in some groups, but not in others. Zfx appeared silenced in only in the myomorph rodents (mouse, rat, lemming, hamster), probably due to the restricted testis-limited expression of the Y-linked Zfyl and Zfy2 genes, which led to the inactivation of the X chromosome homologue (Luoh et al., 1995). This is unlike the situation in all other eutherians where the ZFY gene is well expressed and ZFX escapes inactivation (Jegalian and Page, 1998). To determine i f X-inactivation is present in the Coast mole, I examined the methylation status of ZFX (zinc finger on the X ) , FMR1 (FRAX, fragile X mental retardation; and ARA (androgen receptor), which are known to possess CpG islands and are X-linked in humans and rodents (Cutler Allen et al., 1992, Eichler et al., 1995, Luoh et al., 1995, Carrel and Willard, 1996, Heard et al., 1997). Attempts were made to perform this assay using two other X-linked genes, SMCX  (selected  mouse  cDNA  on  X  and  human  homologue)  and  ALD1  (adrenoleukodystrophy), but proved unsuccessful. In humans ARA, ALD1, FMR1 and S M C X are located on the conserved region of the X , while ZFX lies on a region of the chromosome that is X-linked in eutherians, but autosomal in marsupials (Pask and Graves, 1998). Although no data exist regarding the chromosomal location of these genes in the Coast mole, according to Ohno's Law, genes present on the non-pseudoautosomal region of the X chromosome in one eutherian,  69 should be present on the X chromosome of all eutherians (Ohno, 1967). Studies examining the methylation status of 5' CpG islands have determined that FMR1 and ARA are subject to X inactivation in both rodents and humans (Fu et al., 1991, Norris et al., 1991, Cutler Allen et al., 1992), ZFX escapes silencing in every eutherian order studied, except myomorph rodents (Jegalian and Page, 1998). The human and mouse ZFX/Zfx CpG islands each contain 19 CpG methylation sites and are almost identical at the sequence levels, indicating a selective pressure for conservation, despite the fact that in one species the gene is silent, while in the other it escapes X-inactivation (Luoh et al., 1995). The inactivation status of SMCX is less clear. In general, species that possess the Y linked homologue (SMCY), such as primates and rodents, do not inactivate SMCX (Jegalian and Page, 1998). In contrast, cows and guinea pigs, which lack SMCY, silence SMCX.  However,  rabbits and goats seem to possess SMCY, yet still inactivate SMCX most probably due to the very low expression of the Y-linked homologue (Jegalian and Page, 1998).  As mentioned  previously, Smcx is thought to escape X-inactivation in rodents, but a more detailed analysis of its expression patterns in the mouse have revealed a more complicated picture. It seems that the murine Smcx actually exhibits developmental and tissue-specific variation in escape from X silencing (Carrel et al., 1996b, Sheardown et al., 1996). The degree of expression from the inactive X allele is highly variable and ranges from 20 % to 70 % of the active X allele levels. Imprinted extraembryonic tissues showed the most complete repression of the inactive X Smcx expression, while the highest levels of expression was seen in the heart (Carrel et al., 1996b).  70 Identification of male and female Coast moles  The success of the methylation-based X-inactivation assay, or any other method for that matter, depends on the reliable sexing of individual animals. The determination of sex in the Coast mole is complicated by the lack of any significant sexual dimorphism and the rather similar appearance of the external genitalia. The Coast mole is not unique regarding this aspect of their anatomy - the masculine appearance of the female European moles led researchers to believe that the animals were hermaphrodites (Jimenez et al., 1993, Sanchez et al., 1996a). However, there have been no indications that any North American mole possesses an ovotestis or exhibits masculinization to the degree seen in the European talpids. While it is possible to unequivocally determine the sex of the individual upon dissection, for many of our animals we were only provided with a D N A or tissue sample. Therefore, it was necessary to devise a D N A based sexing method, such as one relying on the presence of the Sry gene, which is malespecific in all eutherians tested, including European moles (Sanchez et al., 1996b).  The  sequence of the Sry H M G box had been previously determined for the Algerian hedgehog (Erinaceus algirus), two shrew species (Crocidura suaveolens, Neomys anomalus) and three mole representative (Talpa romana, T. europea, T. occidentalis) (Sanchez et al., 1996b). The TeSRYl,2 primers was designed from the insectivore Sry H M G box sequence determined by Sanchez and colleagues (1996b) and were tested as indicators of sex in the Coast mole samples (Table 1).  71  Results  Sex determination using conserved Sry primers  PCR with Sry primers (TeSRYl,2) designed from the sequence of Sanchez et al. (1996b) was performed on all the available moles samples, some of which had already been sexed by K . Cambpell during dissections. Although slightly relaxed P C R conditions (50° C annealing, 40 cycles) were necessary to obtain strong bands, the primers worked extremely well and the P C R assay confirmed the sex of most of the animals, although in two cases the identification based on external morphology did not agree with the molecular results and was presumably incorrect (Figure 6).  Methylation analysis  Methylation-sensitive Hpall restriction digests followed by PCR with conserved primers spanning the CpG sites were performed for all three genes. Hindlll digests were performed as negative controls and all the reactions were duplicated on murine D N A , which served as a positive control. Since the primers were heterologous with respect to the Coast mole template and the degree of sequence conservation was unknown, low stringency P C R conditions (50° C annealing, 40 cycles) were used.  In the S M C X PCR, however, it was possible to use an  annealing temperature of 60° C. Additionally, due to the C G richness of the template, increased concentrations of M g C b and the addition of betaine were required (see Materials and Methods).  -tO  Qj  Qj  40  40  158 bp  Figure 6: Coast mole DNAs amplified with the TeSRYl,2 primer pair The 40 cycle PCR reaction was performed at an annealing temperature of 50C. The primers do not amplify mouse DNA, which served as a negative control in this reaction.  73  The ZFX primers, which were designed from conserved sequence at the human and murine CpG island (Luoh et al., 1995), amplify a 105 base pair fragment spanning two C C G G (Hpall) sites in mouse and humans (Table 3). Human CpG-island sequence was used to derive the ARA primers (Cutler Allen et al., 1992), which span two Hpall sites in both human and mouse. The FMR1 primers, surrounding one Hpall cut site in humans and three in mouse, were similarly derived from human sequence (Fu et al., 1991). Both the ARA and FMR1 CpG islands are associated with lengthy trinucleotide repeats (CAG in ARA and C G G in FMR1) that, when expanded, lead to the development of neurodegenerative disorders in humans (Strachan and Read, 1999). ARA and FMR1 fragments are only seen in the female Coast mole and female mouse control (see Figure 7), indicating that the CpG sites spanned by the primers were methylated in genomic DNA. The lack of a male ARA or FMR1 signal is due to an absence of methylation at the same sites, which allowed the methylation-sensitive enzyme Hpall to cut the genomic template. The presence of bands in both male and female moles and mice in the Hindlll lane confirms that the pattern seen in the samples digested by Hpall was not to lack of template or nonspecific digestion, but to a real methylation difference between the male and female animals at the Hpall sites. Thus, the female-specific FMR1 and ARA bands seen in the Coast mole indicate that these genes are methylated and inactivated only in the females and not in the male mole and mouse. ZFX, on the other hand, shows a different pattern (Figure 7). A band is seen only in the female mouse samples, while the male mouse and male and female mole lanes are blank. Therefore, Zfx is methylated and inactivated only in the female mouse. The gene is unmethylated and, therefore presumed to be transcribed from the male mouse and from both sexes in the mole.  74  Attempts were made to perform this analysis on two other X-linked genes, ALD1 and SMCX, but were unsuccessful. Although the ALD1 primers (from Jegalian and Page, 1998) were chosen from CpG-island sequence conserved between human and mouse and appeared to work well in various eutherians (Jegalian and Page, 1998), it proved impossible to amplify a Coast mole fragment (data not shown). The reasons for the failure of SMCX assay, on the other hand, were not due to the inability to amplify the Coast mole product.  The SMCX primers, again designed from a  conserved CpG island in mouse and humans (Jegalian and Page, 1998), not only amplified a female band following digestion with Hpall, but amplified a male band as well (Figure 7). Increasing the stringency of the PCR did not manage to eliminate this band. The inability of the methylation-sensitive restriction enzyme, Hpall, to cut the mole sequence was not due to the lack of Hpall sites since successful digests of the P C R products (which do not carry any methylation mark) demonstrated that these sites were present (data not shown). A n additional problem with SMCX PCR was the persistence of a band in female mouse - a result that was at odds with published data (Jegalian and Page, 1998). The results of the methylation assay are summarized in Table 4.  75  Hind III  o o C D C D f c g s CD 3  0J_ CD  01 CD  CD g Q)_ CD  Hpa II  2  o  f c C D C D f c f c CD CD ID CD  3  3 ffi.CD CD 3  3  t£  CD  M_ CD CD  CD  ZFX 200 bp 100 bp FMR1 300 bp 200 bp  ARA  300 bp 200 bp  SMCX  200 bp 100 bp  Figure 7. Methylation analysis of the ZFX, ARA, FMR1 and SMCX genes in the Coast mole Coast mole genomic D N A was digested with the methylation-sensitive enzyme Hpall, followed by PCR primers flanking the potentially methylated CpG sites. In the presence of methylation Hpall cannot cut and a PCR product is amplified. Since the methylation status of these genes in the mouse was known, mouse DNA and digestion with Hindffl, which does not cut within the region amplified, served as controls.  76  ARA  FMR1  SMCX  ZFX  Amplification in methylation assay  Amplification in methylation assay  Amplification in methylation assay  Amplification in methylation assay  Reference  This study. Coast mole  Human  Mouse  F  M  F  M  F  M  F  M  +  -  +  _  +  +  _  _  Inactivated  Inactivated  Inconclusive  F  M  F  M  F  M  F  M  +  -  +  -  -  -  -  -  Inactivated  Inactivated  F  M  F  M  F  M  F  M  +  -  +  -  +*  -  +  -  Inactivated  •  Escapes  Inactivated  Escapes  Jegalian and Page, 1998; Carrel and Willard, 1996a, Cutler-Allen et al., 1992.  Escapes  Inactivated *  Inactivated  Data are in contrast to published results (Jegalian and Page, 1998). F - female, M - male  Table 4. Summary of methylation analysis.  Confirmed in this study (except*); Jegalian and Page, 1998, Carrel et al., 1996b; Sheardown et al., 1996.  77 Discussion  Sry has been identified as a male-specific gene, in a wide range of eutherians, including European moles and has been proven to be the male switch in the sex determination cascade in humans and mice (Sanchez et al., 1996b, Pask et al., 2000).  Sry is also found on the Y  chromosome of several marsupial species and sequence conservation at the H M G box has been noted (Foster et al., 1992). However, there is no functional evidence that Sry is involved in sex determination in the Metatheria as it seems to be ubiquitously expressed (Foster et al., 1992, Pask et al., 2000). The absence of an Sry homologue in the monotremes and the loss of Sry from several eutherian species (mole-voles (Ellobius), spiny rats) imply that other genes may control testis determination in some mammals. The recent identification of X and Y-linked marsupial homologues the of the human sex-reversing gene ATRX (alpha thalassemia, mental retardation on the X ) is especially interesting (Pask et al., 2000). In humans and mice the ancestral Y homologue of ATRX is no longer present in the Y chromosome, while in marsupials the ATRY is not only present, it is also expressed.  The  marsupial ATRX is expressed everywhere except in the developing male gonad, while the ATRY product is testis-specific. The ATRXJATRY gene pair does not conform to the current model of evolution for the sex chromosomes (Lahn and Page, 1997). According to this theory, the Y carries two classes of genes with different evolutionary origins. The Class I genes arose from shared X / Y sequences and are expressed ubiquitously, while Class II is comprised of malespecific genes transposed onto the Y chromosome and exclusively expressed in the testis. Recent evidence has shown that even the genes with testis-specific expression may actually trace their origin to ancient X Y gene pairs (Pask et a l , 2000). Based on the fact that human  78  ATRX is involved in the sex-determination cascade downstream of Sry and several other genes and its expression pattern in marsupials, Pask and colleagues (2000) propose that ATRX/ATRY may represent the ancestral testis-determining mechanism. However, Sry is present only in the Coast mole males, indicating that it is borne on the Y chromosome in these organisms. Although in the absence of functional data it is impossible to say whether Sry is the testisdetermining gene in moles, its presence on the Y chromosome is a very strong indicator that Scapanus conforms to the usual eutherian method of sex determination. Despite the demanding P C R conditions necessary in the methylation analysis, it is still surprising that murine or human primers worked at all on an evolutionarily distant insectivore. However, the high level of conservation, the authors used the term "unprecedented", seen at the human and mouse ZFX CpG island indicates that there may be a selective pressure to conserve the 5' regulatory sequences of X-linked genes to ensure proper and stable inactivation (Luoh et al., 1995). The CpG-islands in the ARA and FMR1 genes also seem to be relatively well conserved in the Eutheria, although the size of the associated trinucleotide repeats differs between species and between individuals (Cutler Allen et al., 1992, Eichler et al., 1995, Krongrad et al., 1995). In light of these results, it is unclear why the ALD P C R primers, designed from sequence conserved between humans and mice, was unsuccessful in moles, even though the gene is located on the conserved region of the X chromosome. It is possible that the region from which the primers had been selected was not as conserved as was initially thought and had diverged sufficiently in the mole to impede amplification. The problems with the SMCX assay, in which a male and female P C R fragment were seen after digestion, did not allow us to make any firm conclusions regarding the inactivation status of the gene in the Coast mole. The existence of a SMCY, a housekeeping gene, has been  79 established in many eutherians, although it is not clear if the gene is functional in all cases, but its presence has not been studied in any insectivore (Jegalian and Page, 1998). Jegalian and Page (1998) reported CpG island methylation at the SMCX gene in the hedgehog, but did not comment on the presence of a Y-linked homologue. It is possible that moles have several SMCY copies, one of which may be functional, and that one of the gene copies is methylated, giving rise to the observed P C R fragment.  In fact, a large proportion of Y chromosome is  heterochromatic and consists of repeated sequences and pseudogenes, consistent with its degeneration after recombination with the X chromosome was suppressed (Chapter 1). However, multiple copies of SMCY have not been seen in other eutherians (Lahn et al., 2001) and one would not expect the inactivation of the X-linked gene i f the Y homologue is functional. Alternatively, mole SMCY may have become nonfunctional and methylated, which in turn led of the silencing of the X-linked copy. In these two scenarios, the presence of a female band, despite the mysterious origin of the male fragment would still indicate that the gene is subject to X-inactivation. However, a third possibility is that the PCR bands were due to duplicated and silenced SMCX/Y pseudogenes that are autosomally located, in which case the results of the methylation assay are invalid and do not provide any information with respect to the inactivation of this gene. Nevertheless, the data allow us to conclude that at least one methylated SMCA77-like sequence exists in both male and female moles. The presence of a Smcx PCR fragment in the female mouse does not agree with the data of Jegalian and Page (1998). This inconsistency may be due to several reasons, such as the tissue or strain source of the D N A or P C R conditions. Jegalian and Page (1998) used D N A extracted from the liver and spleen of a B A L B / c mouse, while our sample originated from the spleen of a C57Black6 mouse.  Since Smcx shows only partial escape from X-chromosome  80 inactivation, which differs between tissues and seems to change through development (Carrel et al., 1996, Sheardown et al., 1996), the D N A source tissue could have made a significant difference. On the other hand, since the mouse tissues used do not appear to differ strikingly, it is also possible that the degree and location of escape varies between the mouse strains. Additionally, the P C R conditions were not identical in the two experiments. For example, Jegalian and Page (1998) amplified only 50 ng of D N A using an annealing temperature of 62° C and 30 reaction cycles, while I used 100 ng of D N A , 60° C for annealing and 30 to 40 cycles of amplification. It is possible that the PCR under the conditions employed in this study was capable of amplifying very low levels of methylated Smcx sequences that were not detected by Jegalian and Page (1998). The successful methylation analysis of ARA, FMR1 and ZFX showed that ARA and FMR1 are subject to silencing in the Coast mole, while ZFX escapes inactivation. In the mouse samples all three genes are inactivated, an observation that is in accordance with what is known about murine X-inactivation (Heard et al., 1997, Jegalian and Page, 1998). The fact that the inactivation states of the mouse genes, which served as a positive control, match those reported in the literature, confirms that the methylation assay is reliable method for investigating the question of silencing.  The presence of sex-specific methylation in the Coast mole is a very  strong indicator that the observed genes are sex-linked and that these animals undergo X chromosome inactivation. The results of this assay, in combination with the methylation observed in the hedgehog (Jegalian and Page, 1998) show that silencing through the methylation of CpG islands is present in both of the lineages comprising the core Insectivores, the Soricomorphs and the Erinaceomorphs. Thus, methylation seems to be an integral part of X inactivation in the Eutheria and must have been recruited into the dosage compensation  81 mechanism after the split with the Metatheria, in which methylation of CpG islands on the inactive X chromosome has not been observed (Kaslow and Migeon, 1987, Loebel and Johnstone, 1996). Additionally, the methylation analysis of ZFX presented in this thesis in combination with the data indicating ZFX is not silenced in hedgehogs (Jegalian and Page, 1998), confirms that this gene escapes inactivation in insectivores of the Erinaceomorph and Soricomorph lineage. In conclusion, the results of the methylation-sensitive restriction digest and PCR assay not only reveal that X-chromosome inactivation exists in Scapanus orarius, but also confirm that silencing through methylation is conserved in the insectivores and the Eutheria.  82  Chapter 4  Xist  in the Coast mole: identification, expression and sequence analysis  Introduction  The XIST/Xist gene is the only gene expressed exclusively from the inactive X chromosome and encodes a large polyadenylated functional R N A (Brown et al., 1991, Brockdorff et al., 1992, Brown et al., 1992,). While the mechanism of XIST/Xist action remains unclear, there is no doubt that this R N A is absolutely essential for X-chromosome inactivation in eutherians. X chromosomes lacking the X inactivation centre (XIC), and thus the entire XIST/Xist gene cannot inactivate and neither can chromosomes with partial deletions of the gene (Marahrens et al., 1996, Penny et al., 1997). Gain-of-function experiments involving transgene constructs have confirmed that Xist is necessary and sufficient for inactivation, although the counting and choice aspect of X-chromosome inactivation seem to lie outside of the Xist gene (Avner and Heard, 2001) (for more detail refer to Chapter 1). A n XIST homologue has been identified in every eutherian species examined (cow, horse, rabbit, cat, rat, pig, various primates - see Appendix 1 for GenBank accession numbers), although expression studies have only been done in humans, mice, voles and cows. At this time there are no indications that a marsupial Xist genes has been identified (Wakefield, 1998). Until very recently the only complete sequences of the gene were those of human and mouse (Brown et al., 1991, Brockdorff et al., 1992, Brown et al., 1992), however on March 23 , 2001 rd  Nesterova et al. submitted the complete Xist sequences for three species of voles (Microtinae,  83 Rodentia): Microtinus rossiameridionalis, M. argavis and M. transcaspicus. Partial Xist sequences are also available for the horse, cow, rabbit, cat, pig, rat, rhesus monkey, African green monkey, howler monkey and chimpanzee.  A l l of the partial sequences, except the rat,  pig, cat and two bovine ESTs, encompass the 5' repeat and promoter region and were obtained by screening of cDNA libraries with a human XIST c D N A probe (Hendrich et al., 1993). The remaining cat, rat and bovine sequences are very short fragments obtained as ESTs or in genomic surveys. The available Xist sequences are shown in Figure 8. Sequence conservation of the XIST gene amongst the examined species is generally low. The overall sequence similarity between the human and murine genes is only 60-70 %, although some regions show up to 80 % identity (Brown et al., 1992, Hendrich et al., 1993). Although several open reading frames (ORFs) were identified in the murine and human sequences, they were all quite small (483 bp in humans, 576 bp in mouse) and unlikely to be encoded by such a large transcript as XIST/Xist (Brockdorff et al., 1992, Brown et al., 1992). None of the ORFs were shared between the two species nor spanned an entire exon.  Additionally, the best  candidate for a protein-coding region in the human gene (ORF#4), was not well conserved in other primates (Brown et al., 1992, Hendrich et al., 1993). The combination of the lack of a conserved ORF and nuclear localization, has led to the widely accepted view that XIST/Xist is not translated and functions at the R N A level. A n unusual feature of XIST is the presence of several regions of direct tandem repeats. Two regions of these repeats are relatively conserved between mouse and human (the 5' repeats and downstream C-rich repeats), while others have been differentially amplified in the two genes. By performing Southern analysis with a variety of human XIST probes, Hendrich et al. (1993) identified the region of conserved repeats in the 5' region of exon one and located another 380  84  bp region of conservation approximately one kilobase downstream of the repeats within the same exon in several eutherian species. In the 380 bp area the human, lepine and bovine sequences, showed approximately 70 % identity. Mouse, on the other hand, exhibited only 47 % identity to the human sequence. In general, the alignment shows small regions of high similarity interspersed by areas of high divergence, as could be expected for a functional R N A in which secondary structure rather than primary sequence similarity is selected for. The primate Xist fragments were very similar, in some cases identical, to the human sequence. Similarly, the vole Xist genes, which also contain the 5' repeats described by Hendrich et al. (1993), and the small rat fragment seem to be very similar to the mouse, although the detailed vole sequence analysis is not yet available to the public (Nesterova et al., 2001 in press). Thus, despite its low sequence conservation, Xist is present in all eutherians studied, supporting its essential role in X-chromosome inactivation in this mammalian subclass. The Coast mole is a eutherian mammal and, based on the results of the methylation assay presented in Chapter 3, this animal seems to utilize D N A methylation in X-linked gene silencing just like other eutherians. Although methylation of the inactive X chromosome is not seen in the Metatheria, methylated CpGs are present in other regions of the marsupial genome (Loebel et al., 1996, O'Neill et al., 1998). The lack of marsupial X-chromosome methylation has generally been thought of as an ancestral mammalian characteristic (Heard et al., 1997). However, since neither methylation nor X-chromosome inactivation have been extensively studied in the monotremes and are generally unknown (Cooper et al., 1993), the unlikely possibility remains that methylation of CpG islands on the X chromosome is the ancestral mammalian condition that was lost in the metatherians.  In this scenario, the CpG island methylation of X-linked  85  <t>  1-  «  o>  <u  >£ S | .SP i 4> 25 J>  co <L>  a. g fi  9  O  4)  u  -5  §  B  o go  2 §•'•£  H  P o .5, =  on £  Qo — 08  ES  O  ,P  i Is § 03 -  E  *  E  03 HQ  ¥  o  P  £ * fe 5.  u  O CJ  e 3  > S 3 ot<0  J •* S3  _Q .£>  — *3  a tS  00 fi _ =3 _g T3  fc- T3 1) O - O 03 —  03  O c  s  -a  a> <u > o 03 S  T3 cu  .e a.' « C  O on 08  CD —  oo w  •= 8 o —  1  .v2i "Op  — o o  P  <u x , o 03  5 fi < fi § s Q. © on  03 C  11  c a> c JS 8 e £ •2 e«g 3  f=  Oa-  03  O  >  cu  ID  LU H LU  o  O CL  cr o  CD CQ  £ i w on on 03  03 =» C a> R « a> e ^ "St e) nt * o £3 o « s on cu <— co flj o f Q C8 O cu G •2 « v CU o S P 03  •I LU CO  c  on CD • -  WCL  o o  < o  <>  cr  18  LU hLU LU  cu  CU  fi *  -s »  E n.  on 00 cu a Q. 0> ,** g on on i> m i- on S O < -  Sf —  S -=  > on  1§  86  sequences in Scapanus may represent an ancestral condition and may not mean that other aspects of the inactivation process in this species match those seen in mouse and humans. Apart from the methylation differences, another distinction between the inactive X in the two mammalian infraclasses is the presence of Xist, a gene that has been identified in all eutherians examined, but remains elusive in the marsupials (Hendrich et al., 1993, Wakefield, 1998). A crucial factor that would confirm that elements of X-chromosome inactivation are indeed conserved within the Eutheria would be the identification of a functional Xist homologue in the Coast mole. Previous studies trying to detect Xist in species other than mouse and humans have used Southern analysis (Hendrich et al., 1993) or have localized the R N A by probing with heterologous probes, as was done by Duthie et al. (1999) in their study of vole chromosomes using a mouse Xist probe. PCR was not attempted or was unsuccessful, except when amplifying sequences from very closely related organisms, such as using human primers on primate sequences (Hendrich et al., 1993). However, in order to identify a gene by Southern analysis it is necessary to obtain a genomic or cDNA library of the organism of interest, neither of which exist for the mole nor any insectivore. A lack of high quality mole R N A early in the study prevented us from constructing our own cDNA library and led to the decision to attempt the identification of Coast mole Xist using PCR-based techniques - an approach that ultimately proved fruitful.  87 Results  Identification of Coast mole Xist  Before attempts were made to identify mole Xist through P C R it was necessary to obtain a positive PCR control. Since no suitable mole sequences were available in the database, primers had to be designed using sequence information from other species.  HPRT (hypoxanthine-  guanine phosphoribosyltransferae) c D N A sequences from a broad range of eutherian and one marsupial species, all available from GenBank, were aligned and regions of high conservation were identified. The primers SKHPRT1, 2 were picked manually and were expected to amplify both genomic and c D N A (see Table 3). However, although this primer pair works on human, bovine and mouse D N A and cDNA, it only seems to amplify mole cDNA. These results could indicate the presence of an intron between the primers in the mole. After this unexpected discovery, the SKHPRT1,2 primers were used to check for presence of mole cDNA. The rapid identification of a mole Xist P C R fragment from genomic D N A and the mole-specific primers designed from this fragment served as a positive D N A control. Initially, I attempted, without any success, to detect mole Xist by using human and murine primers available in the laboratory in very low stringency PCR reactions (annealing temperature as low as 47° C, 40 cycles). Even the mXistl,2 primer pair, designed from exon 6 sequence conserved between mouse and human, failed to amplify a band in Scapanus. Keeping in mind the low identity score seen for mouse in the alignment of the 380 bp conserved fragment described by Hendrich et al. (1993), it seemed necessary to include more than the human and mouse sequences when designing primers expected to work across a range of species. Even i f  88 mouse had not shown such poor alignment with the other sequences, it would still have been beneficial to expand the number of sequences aligned when designing the inter-specific primers to maximize the chances of identifying a region that was truly conserved. The 5' region of exon one was aligned between human, bovine, lepine and equine sequences and regions of high similarity were identified. The mouse sequence was initially included in the alignment, but proved to be somewhat of an outlier showing significantly less conservation (between 11 % and 60 %) with respect to the other sequences in this region, which showed similarities of about 70 %. The removal of the murine sequence dramatically increased the quality of the alignment and facilitated primer choice.  The locations and amplification  conditions of the four primers selected, SKXIST1, SKXIST2, SKXIST3, SKXIST4, and other conserved primers are shown in Table 5 (for primer sequences see Figure 4 in Materials and Methods). Although all four primers appeared to be well conserved between the species examined, only the SKXIST2 and SKXIST3 primer combination produced a 286 bp fragment in Coast mole D N A using relatively relaxed PCR conditions (50° C annealing temperature, 40 cycles), which was then gel-purified and sequenced. B L A S T searches using this mole sequence against the non-redundant database detected good matches with all the known Xist sequences in this area of the gene, with the exception of mouse. C L U S T A L W (v. 1.8) alignments with the 245 base pair mole fragment (the 286 base pair fragment without the primer sequences), showed significant similarity to the human, cow, horse and rabbit Xist regions from which the primers were derived (Figure 9). The highest identity was with the horse sequence (81 %), the lowest with human at 59 %, while the cow and rabbit were 77 % and 76 % identical to the Coast mole  89 fragment. The comparatively low alignment score between the human gene and the remaining sequences is primarily due to stretch of C-rich repeats not present in the other species examined. The mouse sequence showed very poor alignment in this region with 22 -59 % identity, while the voles, which were not included in the figure since they were not available at the time of this experiment, scored even lower. However, amongst themselves, the four rodent species exhibited a higher sequence identity. The three Microtus species were 98 % identity to each other and shared 70 % homology with the mouse.  Expression analysis of the putative Coast mole Xist fragment  The sequence analysis of the newly identified 245 bp Coast mole band was a very good indicator that the fragment was indeed Xist. However, this sequence similarity, especially in a gene that is not especially conserved at the primary level, did not provide sufficient evidence for the existence of a Coast mole homologue.  Additionally, even if the Scapanus fragment  represented an Xist-re\ated sequence, there was no evidence that it was functional and involved in X-chromosome inactivation in this species. If this Coast mole fragment was indeed Xist, one would expect it to follow the female-specific expression pattern seen for its homologues in mice, humans and cow. Mole-specific primers CMXIST1 and 2 were designed from the 245 bp fragment, which allowed an increase in the stringency of the P C R conditions and more certainty when determining the expression status of the Scapanus sequence. However, before the expression of  90  Primer name  Location in human XIST (M97168)  Amplification in mole  SKXIST1  1446-1470  NO (with SKXIST2, SKXIST4)  SKXIST2  2255-2236  YES (with SKXIST3); 50° C, 40X  SKXIST3  1898-1908  YES (with SKXIST2); 50° C, 40X  SKXIST4  1913-1891  NO(withSKXISTl)  SKXIST6  29711-29688 (U80460 PAC, intronic)  YES (with SKXIST7); 50° C, 40X  SKXIST7  11010-11030  YES (with SKXIST6); 50° C, 40X  SKXIST9  11609-11628  NO (with SKXIST10)  SKXIST10  11915-11895  NO (with SKXIST9)  SKXIST11  13738-13760  NO(withSKXIST12)  SKXIST12  13892-13872  NO(withSKXISTll)  SKXIST13  3-23  NO (with CMXISTREV1, CMXIST2, CMXISTREV4, CMXIST7  Table 5. Conserved Xist primers. The locations and sequences of the primers are shown in Figures 4, 11, 12 and 13.  91  Coastmole Cow Horse Human Rabbit Mouse  CMXISTl-» GACTGTGCTGGCTGAACCT-ACAC AATTCCTCTTCATTGCTGAAGTGGCCTGAGG GGTTGTGCAGGCAGAACCTCACGC CATTCCTCTGCATTGGTGCCGTGGCCGAAGC GGCTGTGCGGGCTGAACCTCACCC CATTCCTCTGCGTTGCTGCTGTGGCCTCAGG GGCTGTGTGGTCTGAACCTCCCTC CATTCCTCTGTATTGGTGCC-TCACCTAAGG GGCTATGTGGGCTGAACCCCACCCA CATTCCTCTGTATTGATGCAATGGCCTTTGG GGCGGTGCAAACTAAAACTCAGCCCGTTCCATTCCTTTGTATTGTT CAGTGG ******  Coastmole Cow Horse Human Rabbit Mouse  TTAG CCTT CTAG CTAAGTATACCTCCCCCCCCACCCCCCAACCCCCCCAACTCCCCACCCCCACCCCCCACC CTAA CTAG  Coastmole Cow Horse Human Rabbit Mouse  ACTTACCTG GCCCCCTTGCCCC AGGCCC CGCTCTTC— CCCCACTT CCCCCTT-CCCC GCCCCC TGCTCTTC— TCT-ACT CCCCCCT GGCTTC TGCTCTTC— CCCCACCTCCCCACCCCCCTACCCCCCTACCCCCCTACCCCCCTCTGGTCTGCCCTGCAC -TCTACTC CCGCCCT GGCCTG TGGTCCTCCC -TCTACTTA CACCTT GGCCTC TATTTAGC-C  Coastmole Cow Horse Human Rabbit Mouse  CACTCTGTGGCCCCCGGAGCAGTGCTCCATGCTGGCCAAGTGTGAACATGGTGATCAGTC TGCGTTGTGGCC—AGGGGCAGTGCGCCATGCCTGCTAAGTGTGAACTTGGCGGTGAGTC CGCACTGTGGCC—AGGGGCAGTGCTCCATGCCTGCCAAGTATGAATATGGCGGTGAGTC TGCACTGTTGCC—ATGGGCAGTGCTCCAGGCCTGCTTGGTGTGGACATGGTGGTGAGCC TGCACTGTGGCC—ACCGGCAGTGCTCCAGGCCTGCCTGGTGTGGACATGGTGGTGAGCT AGCACTGATCTC AAGCGGTTCTCTAAGCCTACTGGGTATAA GTGGTGACTT  Coastmole Cow Horse Human Rabbit Mouse  GAGGCAAGGACCAGAATGGATCGCAGATGATCGTTGACCA-CAGGTGGCAGAAGAGGAAT GTGGCAAGGACCAGAATGGATCGCAGATGATCGTTGGCCAACTGTTGGCAGAAGAGGAAT GTGGCCAAGACCAGAATGGATCGCAGATGATCGTTGGCCAACAGGTGGCAGAAGAGGAAT GTGGCAAGGACCAGAATGGATCACAGATGATCGTTGGCCAACAGGTGGCAGAAGAGGAAT GTGGCAAGGACCAGAATGGATCACAGATGATCG AAGGGAAAT -TGGCCAGAGTCATAGTGGATCACAAATCACTGGTGA AGAGGTAGAAT  Coastmole Cow Horse Human Rabbit Mouse  •-CMXIST2 CCTTGCCTTTCTCAAGAGGAACATCTACCCCG CCCTGCCTTCCTCAAGAGGAACACCTACCCCG CCCTGTCTTCCTCAAGAGGAACACCTACCCCG TCCTGCCTTCCTCAAGAGGAACACCTACCCCT -CCTGCCTTTCTCAGGAGGAACATCTACCCCG -CCTACCTTCTTCCAAA ATCTACCCCA  ***  +  ** * * * * * * *  ** ** *  *  * *  ** +  PERCENT SEQUENCE SIMILARITY Coast mole Cow Horse Human Rabbit Mouse  79  Cow  S3  Hb  60 78 29  77 Tl  S6  Horse  64  82 b"l  Human  Rabbit  53 4U  59  Figure 9: The first Coast mole Xist fragment identified The fragment is aligned with the human (M97168; bases 1913- 2234), bovine (AF104906; 1912-2153), lepine(U50910; 2782-3002), equine (U50911; 5679-5914) and murine genes (L04961; 2029-2237). The mole gene was amplified using the SKXIST2, 3 conserved primers, which lie outside of this alignment. The CMXIST1 and CMXIST2 mole-specific primers are indicated. The vole sequences were not available at the time of this experiment and are not included in the figure.  92 this putative Xist fragment could be examined, it was necessary to obtain undegraded R N A from the mole samples (Table 1). Although reasonably intact D N A could be obtained from frozen mole organs, animals found dead in traps or as "roadkill", the R N A from these specimens was of very poor quality. Undegraded R N A had to be obtained from recently sacrificed animals and from cell lines. Since no mole cell lines were available, kidneys from recently deceased Coast moles were used to establish the cell cultures. The cells, probably fibroblasts, initially grew well, but eventually started to senesce and differentiate, which is to be expected from a nontransformed cell line with a finite lifespan. The five female cultures (CM11, C M 12, CM13, C M 16, C M 17) derived from sacrificed moles or animals killed in the field proliferated well. Unfortunately, the lone male cell line ( C M 15) did not grow as well, probably due to the fact that the mole from which the cell line was derived had been found in a trap and had been dead for an uncertain amount of time. RT-PCR reactions on the tissue sample and cell line RNAs were performed to create cDNAs, which could then be used as a template for PCR. Following P C R (54° C annealing, 40 cycles) with the CMXIST1 and CMXIST2 primers, the 245 base pair product was present in all the female cDNAs, but absent in the male C M 15 samples (Figure 10).  Further sequencing of Coast mole Xist  The results of the previous two sections had established that Xist is present and probably functional in the Coast mole, since it is only expressed in the female. M y next goal was to obtain as much Scapanus Xist sequence as possible to determine the size of the gene and to evaluate the global sequence conservation.  Several more interspecific Xist primers were  93 designed from areas of the gene conserved between the available sequences - of these, only one primer pair was successful (Figure 4, Table 5). The identification of a short bovine EST at the very 3' end of human exon one and extending by 81 base pairs into the intron led to the design of the SKXIST6 and SKXIST7 primers, which managed to produce a 343 base pair band in the mole. This fragment was analyzed through B L A S T and it showed significant identity to human and cow (74% and 75%). On the other hand, the SKXIST9 and 10 pair derived from a bovine EST corresponding to human exon four, amplified a band that, upon sequence analysis, proved to be either a P C R artifact or due to an unknown contaminant.  B L A S T searches with the  SKXIST9,10 fragment failed to match any Xist sequences in the database, but did show a high similarity to a variety of cloning vectors. Additionally, it was impossible to align this fragment with the bovine EST and human sequence from which the SKXIST9,10 primers were derived (data not shown). Other primer pairs designed were SKXIST11 and 12, made from a short feline sequence fragment located in human exon six, while the alignment of murine, human, lepine and equine sequences served as a guide for the design of SKXIST13 at the start of exon one. None of these primers were successful in yielding a product after PCR. Once the method of designing conserved primers was exhausted as a means for obtaining more Coast mole Xist sequence, I turned to inverse PCR. This technique, described in greater detail in Materials and Methods and Figure 3, allows for the amplification of sequence lying outside of the primers, which face away from each other, in contrast to conventional PCR. Amplification is achieved through restriction digestion and ligation reactions that result in a circular D N A molecule containing the two primers now facing each other and spanning the area of interest. Initially, two Coast mole-specific primer pairs facing away from each other were designed from the sequenced fragments (CMXIST1,2REVERSE and CMXIST5,6REVERSE).  94  Figure 10: Female-specific expression of the putative Coast mole Xist fragment P C R amplification with the CMXIST1 and C M X I S T 2 primer pair was successful only in c D N A derived from female animals or cell lines. P C R with SKHPRT1,2 primers was performed and confirmed the presence of R N A in both the male and female samples (data not shown).  95 Genomic D N A of Coast mole 10, was digested with enzymes that do not cut within or between the chosen primer sequences. The Alul and Rsal enzymes, both of which have a four base pair recognition site, were deemed as good choices for the digest since they should produce fragments of moderate size that would be amenable to PCR. Additionally, both enzymes left blunt ends upon digestion, which allowed double digests to be performed. Mole genomic D N A (Coast mole 10, a female) was then digested with each enzyme separately and in combination, ligated and ethanol precipitated. Initially the P C R reaction using Taq polymerase (Gibco) did not result in a band, however when the Expand Long PCR (Roche) enzyme kit was used, which is optimized for the creation of long P C R templates, large (1.2 kb and 1.0 kb, approximately) intensely-staining fragments  were recovered for both the  CMXIST1,2REVERSE  and  CMXIST5,6REVERSE primer pairs. These bands were gel purified, sequenced and contigs were assembled consisting of the original mole fragments through the use of conserved primers and the sequences obtained from inverse PCR. The Coast mole fragment at the 5' region of the gene had been increased to almost 1.4 kb, while the region at the end of exon one extended over 1.2 kb. Both of the sequenced regions of mole Xist have shown female-specific expression (see Figure 10 and data not shown). Additional inverse PCRs aimed at recovering the transcription initiation site at the beginning of exon one using the CMXIST6,7REVERSE primers and a variety of restriction enzymes have so far been unsuccessful. A schematic of all the Coast mole fragments obtained is presented in Figure 11. Although the mole Xist sequence has been enlarged considerably, not much has changed with respect to sequence conservation. The mole fragment corresponding to the 5' region of exon one, shows between 68 % to 84 % homology with the rabbit, human, horse and bovine genes (Figure 12a,b), while the mouse and voles remain outliers with only about 20 % identity  96 to any of the sequences in this area. Different fragments in this region of the murine gene were used in order to optimize the alignment, but the scores remained very low. Similarly, the Coast mole fragment at the end of human exon one is 77 % similar to the human gene and shows poor alignment with the mouse and vole (approximately 40 %), while the identity between the murine and human sequence is 54 % (Figure 13a,b). Thus, the enlarged Scapanus fragments exhibit almost identical levels of homology as did the initial 286 and 343 base pair sequences. Note the while the alignments with and without the rodents in Figures 12 and 13 differ, the distance matrices (percent similarity) do not. This is a result of the C L U S T A L algorithm, which aligns the most closely related sequences first, which are then kept as a group, while more distantly related sequences (i.e. rodents) are added on without disrupting the relative alignments within the first group. Even though the mouse and human XistlXIST genes do not code for a protein product (Brockdorff et al., 1992, Brown et al., 1992, Hendrich et al., 1993), the available mole sequences were nevertheless checked for any ORFs using the G R A I L program (Uberbacher and Mural, 1991). The G R A I L algorithm requires one to choose a model organism in which to identify ORFs, and since mole was not available, the program was run twice, once using the human model and once with mouse model. The choice of organisms did not make a substantial difference in the detection of potential polypeptides. G R A I L identified seven very short ORFs, three of which were classified as having excellent coding potential, while three had good potential and one was of marginal quality. The size of the polypeptides encoded by these ORFs ranged between 7 and 35 amino acids. Five of the seven open reading frames overlapped with each other and encompassed the same area as ORF4 in the human sequence (Brown et al., 1992, Hendrich et al., 1993). No significant ORFs were identified in the 3' region of exon one.  97  C3  cj  co O  CO CO X CO  X CO  CO  >  "3"  c £  HI CC  its  < CL  hco X  CC LU  >  CC CL in > LU CC — I CO  LU QC ICO X  s  a  cfc  | 00  >  LU CC  o  LU >  a: „ LU CO Z  CC O CL LU  JS  JS  ._  CO  co  e3 2 6 S S Js  CM  CC CL  CN  12 18  o o  LU  5  m  EC  .22  X  to  T3  a cs — in  3  h-  CO  X Q X  S £  co  c  Q.  3  hO o co cc o -r CC  3B  5  o ca  CC  o  LU > Z  O  CL  o  o  |&  o  -a 2 o o  X  CL  o O o CC  CO X CO  > LU CC  h-  co X  2 O  >  eg x CO  LU CC H  CO  D LU  CC  I  Q LU  > >  CC LU Q LU O  CC LU Q LU O  z  z  LU 3  o  LU CO  LU  CC Li. Q LU > CC LLJ Q LU O  z  LU  ^  or  UJ  LU  CC CL  CC CL  W  Q  O  £  £ co  •o 2  c c  • « •£ J2 co DH  — I LU CO CO  3 < z o o O LU O LU o CO o CO  E JS  a  i  £  D C =LcoS e 2s CCDT C  C  co  S  X  CJ  cs  © 3 E  E  ?  A  ®  O  «8  •  a a  ss  >  C  B a 1  98  ••CMXISTREV8  Cow Horse coastmole Human Rabbit VoleM. r . VoleM. a. VoleM. t . Mouse  ATCATGTTAGGAGC-TTGGCATGAATTGTGGTATCATGAGGTGGGAAAACATAGGATCAT GGCTCGTTAGGAGC-TTTGTATGAATTAGGGTATCGTGAGGTGGAGAAACA—GGATGGT AGCTTGTTAGGAA GCGAGGTCTCGAAACG--GGGTTGT GGCTTGTTAGGAGC-TTTGCGTG-ATTGTTGTATCGGGAGGCAGTAAGA ATCAT TGCTTGTTAGCAGC-TTTGCATGAATTGTGGTACTGGGATGCCGGAAAACR—AGATTGT GACTCAT-GGTAAC TG-GAATC AGAAAATCAAG CTCT GACTCAT-GGTAAC CG-GAATC AGAAAATCAAG CTCT GACTCAT-GGTAAC TG-GAATC AGAAAATCAAG CTCT GATTTGT-GGTAGCATTTGCG-GGGTTGTGCTAGCCGGAAG-AGAAAGCCAAGGAGTGCT * * * * * *  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM. a. VoleM. t . Mouse  CCTATGTCATATTACAAGGGTCTAA-TGGAAAATGAGCGG--GAGAAGAATTA--GGCGC CCTGTGTTATGTTAAAAGAGGCTAA-TCGAAAATGGGAGG—GAGGAAACTTA—GGCAC GCTGT CACTACCAGAGGCACG-TTGAGAATGACAAG--GAGG CTTTTAT--CAGTACAAGGGACTAG-TTAAAAATGGAAGGTTAGGAAAGACTAA-GGTGC CCTGCAG—CAGCACGAGGCGTTAACTTGAAAATGA GACAGGACTTAAAAGTGC CGTGT TCATGTGTGGTGTTAT-GTGAAAGGAGCTTT ATAAGACTTC--AGGAT CGTGT TCATGTGTGGTGTTAT-GTGAAAGGAGCTTT ATAAGACTTC--AGGAT CGTGT TCATGTGTGGTGTTAT-GTGAAAGGAGCTTT ATAAGACTTC--AGGAC CGTAT TAGTGTGCGGTGTTGC-GCGGA AGCCGC AGAGGACT AG - - GGGAT  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  AGTGTTCAAAATGGCGAT TTTG ACCTTG CAGCATT AGGGTCCGAAATGGCGAT TTTG ACTTTG CGACATT •OATCAAAATGGCGGC GTTA ACTTTG TGGCGTT AGGGCTTAAAATGGCGAT TTTG ACATTG CGGCATT TGGGATCAAAATGGCGCT TTTG CCATTG CAGCGTT AGGGCTTGAAATCGATGT TTCG ATATTG CTGG-GTG AGGGC T T GAAAT C GAT GT TTCG ATATTG CTGG-G-G AGGGCTTGAAATCGATGT TTCG ATATTG CTGG-GGG AGGGCTCAGCGTGGGTGTGGGGATTGGGCAGGGTGTGTGTGCATATGGACCCCTGGCGCG  Cow Horse Coastmole Human Rabbit VoleM. r. VoleM. a . VoleM. t . Mouse  GCTTAGCATGGCTCTCTG—CT--TTGTTAGAGTGTTCAAAATGGCGGATCCA--TTTTG GCTCAGCATGGCTGCCTG--CT--TTGTTAG--TGTCCAAAATGGCGGACACAG-TTTTG GCTCAGCATGGCTGCCTG--TT—TTGTGAA—TGTCCAAAATGGCGGATTCATTTTTTT GCTCAGCATGGCGGGCTGTGCT—TTGTTAGGTTGTCCAAAATGGCGGATCCAG-TTCTG TCTCAGCATGGCTCCCTGTGCT--TTGTTAAATCGTCTAAAATGGCGGATTCAG-TTCGG CTCTGCCCAGATTCTAAGGGGTGCTTGTTAGGGTTTATCAAATGGTGC CTCGG CTCTGCCCAGATTCTAAGGGGTGCTTGTTAGGGTTTATCAAATGGTGT CTCGG CTCTGCCCAGATTCTAAGGGGTGCTTGTTAGGGTTTATCAAATGGTGT CTCGG GTCCCCCGTGGCTTTAAGGGCTGCTCA--GAAGTCTATAAAATGGCGG CTCGG  CMXISTREV7-JV  *  *  *  *  *  *  ******  *  *  Cow Horse Coastmole Human Rabbit VoleM. r. VoleM. a. VoleM. t . Mouse  CCGCAGTGTTCCAATGGCGGGAAG-CCACATCATG GGT GTCTTTGTTC-TAGTG CCGCAGTGTTCAAGTGGCGGGAAGGCCACATCATGGT-GGT GTCTTTGTTC-TAGTG CCGCAGTGTCCAAGTGGCGGGAAGGCCACATCCTGGT-GGC GTCTTTGTTC-TAATG TCGCAGTGTTCAAGTGGCGGGAAGGCCACATCATGAT-GGGCGAGGCTTTGTTA--AGTG CCGCAACGTTAAAGTGGCGGGAAGCCCACATCATGATGGGTGGGGTCTTTTTTTGTAATG CTGCTCTACCCAA--GGCTCGACAACCCCACC TTTATTC-TAACA CTGCTCTACCCAA—GGCTCGACAACCCCACC TTTATTC-TAACA CTGCTCTACCCAA—GGCTCGACAACCCCACC TTTATTC-TAACA GGGCTCCACCCGA--GGCTCGACAGCCCAATC TTTGTTC-TGGTG  Horse Coastmole Human Rabbit Vo1eM.r. VoleM.a. VoleM.t. Mouse  TGCAGCA-TGGCG-GTAGAAATATTCTGTTACATAGTAAAAGATGGCCGCTTAAGTACTT CGCAGCA-TGGCG-GTGGAAATATTCTGTTACATGGCAAAAGATGGCGGCTGAAGTACTT TGCAGCA-TGGCG-GTGGAAATATTCTGTTACATGCTAAAAGATGGCGGCTAGAGCACTT GTTAGCA-TGGTG-GTGGACATGTGCGGTCACACAGGAAAAGATGGCGGCTGAAGGTCTT CGCAGCA-TGATG-GTGRCAGAGTTCGGTCACACAGTGAAAGATGGCGGCTGAAACCCTT TGTAATG-TGGAA-GTAGGGCATTTAGGTTATACAAGAGAAGATGGCGTCTCGAGCTCTT TGTAGCG-TGGAA-GTAGGGCATTTAGGTTATACAAGAGAAGATGGCGTCTCGAGCTCTT TGTAGCG-TGGAA-GTAGGGCATTTAGGTTATACAAGAGAAGATGGCGTCTCGAGCTCTT TGTAGCAATGGATTATAGGACATTTAGGTCGTACAGGAAAAGATGGCGGCTCAAGTTCTT  **  *  ***  **  **  * *  ***  ********  **  99  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  GCCGCAGTCTAAAACATGGCGGGCTT—TTGTCTCTGCCGTGTGCATTTCCTGA-TAGGT GCCGCAGTATAAAATATGGCGGGCTT—TTGTGTTTGCCGTGTGCATTTCCTGA-CAGGT G TGTAACATGGCGG-CCT--TTGTCTTTGCTGTGTGTGTTTCCTGG-CAGGT GCCGCAGTGTAAAACATGGCGGGCCTCTTTGTCTTTGCTGTGTGCTTTTCGTGT-TGGGT GCCGCAGTGTAAAATATGGCGGGCGTCTTTGTCTTTGCCGTGTGCATTTCGTGG-CGGGT GCTGCAG--TATAAGGAAGCTGGCTT TGTTTTTCACAGGCCAGTCTGGTGGCCAAAG GCTGCAG—TATAAGGTAGCTGGCTT TGTTTTTCACAGGCCAGTCTGGTGGCCAAAG GCTGCAG--TATAAGGTAGCTGGCTT TGTTTTTCACAGGCCAGTCTGGTGGCCAAAG GGTGCGG —TATAACGCAAAGGGCTT TGTGTGTCACATGTCAGCTTCATG-TCTGAG  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mous e  TTTGCTGCAGGGACAATATGGCTGACCTTGTCA—TGTGGATATCATGGCAGTTTGTCAC TTTGCCGCAGGGACAATATGGCAGACCTTGTCA—TGTGGATATCATGGCAGTTCGTCAC TTTGCCTCGGGGATAGTATGGCAGGCCTTATTA—TGTGGAAGTCTTGGCAATTTGTCAC TTTGCCGCAGGGACAATATGGCAGGCGTTGTCA—TATGTATATCATGGCT-TTTGTCAC TTTGCCGCAGGGACAATATGGCTGACGATATCA—TGTGGTTATCATGGAGATTTGTCAC TTAGCCTAGAGGACAATGTAGCACATGCTTTGGG—GTGGCCAAGATGGCG—ATGTCAC TTAGCCTAGAGGACAATGTAGCACATGCTTTGGG—GTGGCCAAGATGGCG—ATGTCAC TTAGCCTAGAGGACAATGTAGCACATGCTTTGGG—GTGGCCAAGATGGCG—ATGTCAC TTAGCCTGGAGAG GTGGCACATGCTCTTGAATGTGTCTAAGATGGCGG-AAGTCAT  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  GTGGATATCGTGGCAGGTGGACGTCATGGCGGGTGGATGTCTTGGCAGA GTGGACATCATGGC GTGGACGTCATGGCGCGTGATTGTCCCTTGAGGG-TTTTTTTCTGATATATTCTTGGCGGGCTTAGTCTTGGCGGG GTGATTGTCCCTTGAGGG-TTTTTTTCTGATATATTCTTGGCGGGCTTAGTCTTGGCGGG GTGATTGTCCCTTGAGGGCTTTTTTTCTGATATAGTCTTGGCGGGCTTAGTCTTGGCGGG GTGACCTGCCCTCTAGTGGTTTCTTTCAG TGATTTT  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM. a. VoleM.t. Mouse  GGGTGTTT-GACCGTTACATTCTTGGCGGGCTTTG GGGGATTTTGCCCGGTACATTCTTGGCGGGCTTTG GGGTC CACAGGGGCGTTC-GTCTG-CATGTTCTTGGCGGGCTCAG GGGCT GTTTTTGCTCGGTAAATT GGCGGGCTTTG CTTAGTCTTGGCGGGCTTAATCTTGGCGGGCTTTGCTTAGC-CACACATGGCGGGCTTTG CTTA ATCTTGGCGGGCTTTGCTTAGC-CACGCATGGCGGGCTTTG CTTAATCTTGGCGGGCTTAATCTTGGCGGGCTTTGCTTAGC-CACACATGGCGGGCTTTG -TT TTTTGGCGGGCTT TAGC-TAC--TTGGCGGGCTTTG  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  CATC CACC  AGGAGGGCCTGCCGCATT—GTTAAAG-ATGGCG AGGTGGGCCTGCCGCATT—GTTAAAG-ATGGCGGGGTTTGGCGCGTTTT  CACC  TGGTGGTATTGCCGCATT—GTTAAAG-ATGGCGG  TGCCGCATT--GTTAAAG-ATGGCGG CACA AGGTGGACCTACGGCATT —GTTAAAG-ATGGCGG TCCTTGTCCAAGATGTACTTAGTGCATTAAGATGTGA-GTGCTTG TCCTTGTCCAAGATGTACTTAGTGCATTAAGATGTGA-GTGCTTG TCCTTGTCCAAGATGTACTTAGTGCATTAAGATGTGA-GTGGTTG CCC GAGGGTACACTTGGTGCATTATGGTAGGGTGTGGTTG  Cow Horse Coastmole Human Rabbit VoleM. r. VoleM. a. VoleM. t . Mouse  TGCTTTGCCGCG GAC AAAGTG AAAGGA GGGA GGCGCGTTTTGCCGCGTTTGCTGCGTTTTGCCCCGAAAAGTG GAAGAAAGGTGGGG GCTTTGCCGCGTTTG TGGC—GGAAAGCNCGAAGGAGAAACATGGGG GTTTTGCCGCCTAG TGCCACGCAGAGCG GGAGAAAAGGTGGGA GCGTTGCCGCATTT TGCCGGTCAAAGCA GGAGAAAACCGGGGTTTTACCTTGT GCCACTCGAAGC CCAGT GTTTTACCTTGT GCCACTCGAAGC CCAGT GTTTTACCTTGT GCCACTCGAAGC CCAGT GTCCTACCTTGT GCCACTCGAAGCT GAGGCAAGGCTAAG  100  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  TTGGCAATGTTAGATTGCCGCGTGTC CCA CCCAATC-AGAAAG TGGACAGTGTTGGACTGCCGCACGAC CCA CCCAATCGAGAAGG TGGGGAGTGNTCAATTGCTGCTTTAC ATA CCCAATC-AGAGAG TGGACAGTGCTGGATTGCTGCATAAC CCA ACCAATT-AGAAAT TGCTGGATTGCCGCATGAC TTA ACCAATC-AGAAAT TGGA-AAT TTGCCACTTAA-; TCAGAAATGTGTACTACTTTGGAAGT TGGA-AAT TTGCCACTTAA TCAGAAATGTGTACTACTTTGGAAGT TGGA-AAT TTGCCACTTAA TCAGAAATGTGTACTACTTTGAAAGT TGGA-AGTGTTGG-TTGCCACTTGACGTAACTCGTCAGAAATGGGCACAAGTGTGAAAGT  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  GGTGGT-AGAATCGGTCACAGCCAGTTAG--TGGAGGATG GGTGGT-AGAATTGGGAACAACCAATTAG--TAGAGGATG CATGTT-AGAAAAGAGTATGACCAATTAG--TGGAAGATG GGGGGT-GGAATTGATCACAGCCAATTAGAGCAGAAGATG GAGGGT-GGATTAGA-CACAACCAATTAGCATGGAGGATG ATTGGTTGCCTCTTGACTTAACTAGTCAAAAATGGGCACTTCTACTTGGTGTAGAGGATT ATTGGTTGTCTCTTGACTTAACTAGTCAAAAATGGGCACTTCTACTTGGTGTAGAGGATT ATTGGTTGCCTCTTGACTTAACTAGTCAAAAATGGGCACTTCTACTTGGTGTAGAGGATT GTTGGTGTTTGCTTGACTT—CCAGTTAGAAATGTGCATTATTGCTTGGTGGCCAGGATG  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  GAATTAGA -TGAGTTA-GC GAATTAGACTA -TGAGGTA-GT GAATTACANTGG -TGAGTTA-GC GAATTAGACTGATGACACACTGTCCAGCTACTCAGCGAAGACCTGGG- -TGAATTA-GC GAATTAGACTGTCGATGCACTGTCCAGGTACTGAGCATGGTCATGTGGGCTGAGTTA-GC GAAGTAGACTGT GAAGACTCACTGTCCTACAAGGATTTGAGTTAC-C GAAGTAGACTGT GAAGACTCACTGTCCTACAAGGATTTGAGTTAC-C GAAGTAGACTGT GAAGACTCACTGTCCTACAAGGATTTGAGTTAC-C GAATTAGACTGT GATGAGTCACTGTCCCATAAGGACGTGAGTTTCGC  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  ATAGCACCTCGC-TAC CGTCTCTATT CAGCCAGTCAGG GTAGCACTGCGC-TGT GGTCTTTATT CAGCCAGTCAGGTGGAGGA ATGGCTCCTCAC-TGA AGTCTTTTTT ' X»GI ATGGCACTTCGC-AGC TGTCTTT AGCCAGTCAGGAGAAAGA ATGGCTCTTTGC-AGC TTTCTTTTTT CAGCCAGTCAGGACGAAGA TTGGTACCTCTTATGC CATCGTTTT AAGTGACTACCCAGGT-G TTGGTACCTCTTATGC CATCGTTTT AAGTGACTACCCAGGT-G TTGGTACCTCTTATGC CATCGTTTT AAGTGACTACCCAGGT-G TTGGTACTTCACGTGTGTCTTTAGTCATCATTTTTTCGAAGTGCCTGCCCAGGTCG  Cow Horse Coastmole Human Rabbit  ATTGGCCACAT-TTGTACT AGCAGAGAGGTCACGT-CTCTACGTCG .GCAGCCACGT-TTCTAAGTCG AGTGGAGGGGCCACGT -GTATGT AGTGGAGCGGCCATGTAGTGTGCCC  CMXISTREV4-*  AATCTCAG GCTCCCAG GCTCCCCG CTCCCAG CTCCCAG  TGGGTGGTA TGGGTGGTAA TGGGTGGTGC TGGGCGGTACACCAGG TGGGTGGTATTAAAGT  VoleM.r.  TGAAGGCTCGTGCTTTTTAATTCTTTTTAAT-TTGAGGTGTTGAATGACG  VoleM.a. VoleM.t. Mouse  TGAAGGCTCGTGCTTTTTAATTCTTTTTAAT-TTGAGGCGTTGAATGACC TGAAGGCTCGTGCTTTTTAATTCTTTTTAAT-TTGAGGTGTTGAATGACG GGAGAGCGCATGCTTGC-AATTCT AACACTGAAGTGTTGGATGATGT  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  CCAA-AGTCTTT-CCAAGGCCAA-GGTCTTT-CCAAGGCCAA-GGTCCTT-CCAAGGTGTTTTCAA-GGTCTTT-TCAAGGTCTTTCCAA-GGTCTTT-TCAAGGATCCGATCTGAGAGACTTGAGGGTGGAGGTTCTTGAGGTTGAGGTTGATGTGGATTG ATCCGATCTGAGAGACTTGAGGGTGGAGGTTCTTGAGGTTGAGGTCGATGTGGATTG ATCCGATCTGAGAGACTTGAGGGTGGAGGTTCTTGAGGTTGAGGTTGGTGTGGATTG CGGATCCGATTCGAGAGACC-GAGGCTGCGGGTTCTTG GTCGATGTAAAT —  101  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  -ACATTTG-GCCTTTCC ACC-TCCTCC -ACATCTG-GCCTTTCC ACC— TCC -ACATGTG-GCCTTTCC ACC-TCC -ACATTTA-GCCTTTCC ACC — TCT -ACATTTG-GCCTTTCT ACC-TCC CACATTGAAGCCTCACCTATTAAAAAGAAAGAAGAGAATCAGCGATGAGGTCAGGTTAAC CACATTGAAGCCTCACCTATTAAAAAGAAAGAAGAGAATCAGCGATGAGGTCAGGTTAAC CACATTGAAGCCTCACCTATTAAAAAGAAAGAAGAGAATCAGCGATGAGGTCAGGTTAAA --CATTGAAACCTCACCTATTAAAA-GAAAGAAAAGTATC TAAGGCCA--  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  CTCCCC T CTCACTGGCT — CCCTCCCCTCCCT-CA TTGCCAC CTCCCC T CTCACTTGCT--CCCTCCCCTCCGG-CA TTGACTC CTCCCT C-—CTTCCCTGCT--CCCTCCGCTCCAG-CA—TTACCTC GTCCCC T CTTATTTGTC--CCCTCCTGTCCAG-TG CT CTTCCC T CTTGTTTGTC — CTCTCCTCTCCAGACA TGGTCTC GTATCTCTCCGAAGAGAGT-AATTTATCTGCTTGCTCTCCTATTTGTCTAAATTGGTCTT GTATCTCTCCGAGGAGAGT-AATTTATCTGCTTGCTCTCCTATTTGTCTAAATTGGTCTT GTATCTCTCCGAGGAGAGT-AATTTATCTGCTTGCTCTCCTATTTGTCTAAATTGGTCTT TTTCAAGGACATTTGACTCATCCGCTTGCGTTCATA GTCTC *  *  *  *  *  *  SKXIST4-5>  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  TTGCAGTGCTGGATATTAGGTTGTGCAGGCAGAACCTCA-CGC CATTCCTCTGCAT TTGCAGTGCTGGATAGCTGGCTGTGCGGGCTGAACCTCA-CCC CATTCCTCTGCGT TTGCACTGCTGGGT-TTTGACTGTGCTGGCTGAACCT-A-CAC AATTCCTCTT TCTGGCTGTGTGGTCTGAACCTCC-CTC CATTCCTCTGTAT TTGCAGTGCTGGATATCTGGCTATGTGGGCTGAACCCCA-CCCA CATTCCTCTGTAT TTGCAGTGCTCTATA--TGGTGGTACAAGCTAAGCCTCA--CCG CATTTCTTTGTAT TTGCAATGCTCTATA--TGGTGGTACAAGCTAAGCCTCA—CCG CATTTCTTTGTAT TTGCAGTGCTCTATA—TGGTGGTACAAGCTAAGCCTCA—CCG CATTTCTTTGTAT TTACAGTGCTCTATACGTGGCGGTGCAAACTAAAACTCAGCCCGTTCCATTCCTTTGTAT ** ** **** * * * * * * * *** ** * *  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  TGGTGCCGTGGCCGAAGCCCTTCCCCACTT CCCCC TGCTGCTGTGGCCTCAGGC-TAGTCTACTC CCCCC TTAGACTTACCTGGCCCCCT TGGTGCC-TCACCTAAGGCTAAGTATACCTCCCCCCCCACCCCCCAACCCCCCCAACTCC TGATGCAATGGCCTTTGGC-TAATCTACTC CCGCC TGGTTCAGTGGCCTAAAAC-TTGTCTACTCA--CTCCC TGGTTCAGTGGCCTAAAAC-TTGTCTACTCA--CTCCC TGGTTCAGTGGCCTAAAAC-TTGTCTACTCA--CTCCC TG-TTCAGTGGC TAGTCTACTTA--CACCT  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  TTCCCCGCCCCC-TGCTCTTCT GGCTTC-TGCTCTTCC TGCCCCAGGCCC-CGCTCTTCC CCACCCCCACCCCCCACCCCC-CACCTCCCCACCCCCCTAC CTGGCCTG-TGGTCCTCC -TGGCCTTGGATTTAGTCCGCACTGCCGTATTTAGTC-TGGCCTTGGATTTAGTCCGCACTGCCGTATTTAGTC-TGGCCTTGGATTTAGTCCGCACTGCCGTATTTAGTC-TGGCCTC TATTTAGCC-  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  GCGTTGTGGCCA--GGGGCAG GCACTGTGGCCA--GGGGCAG ACTCTGTGGCCCCCGGAGCAG CCCCCTACCCCCCTACCCCCCTCTGGTCTGCCCTGCACTGCACTGTTGCCAT—GGGCAG CTGCACTGTGGCCAC—CGGCAG CACACTGCTGTCTT--AAGCAG CACACTGCCGTCTT—AAGCAG CACACTGCTGTCTT—AAGCAG AGCACTG--ATCTC--AAGCGG  «-CMXISTREVl/CMXISTl->  T  *  **  *  ** *  102  Cow Horse Coastmole Human Rabbit VoleM. r. VoleM.a. VoleM.t. Mouse  TGCGCCATGCCTGCTAAGTGTGAACTTGGCGGTGAGTCGTGGCAAGGACCAGAATGGATC TGCTCCATGCCTGCCAAGTATGAATATGGCGGTGAGTCGTGGCCAAGACCAGAATGGATC TGCTCCATGCTGGCCAAGTGTGAACATGGTGATCAGTCGAGGCAAGGACCAGAATGGATC TGCTCCAGGCCTGCTTGGTGTGGACATGGTGGTGAGCCGTGGCAAGGACCAGAATGGATC TGCTCCAGGCCTGCCTGGTGTGGACATGGTGGTGAGCTGTGGCAAGGACCAGAATGGATC TTCTCCAGGCCTGC TGTGGGCTTAGTGCTGACAC-TGGCCAGAAT TTCTCCAGGCCTGC TGTGGGCTTAGTGCTGACAC-TGGCCAGAGT TTCTCCAGGCCTGC TGTGGGCTTAGTGCTGACAC-TGGCCAGAGT TTCTCTAAGCCTAC TG-GGTATAAGTGGTGACTT-TGGCCAGAGTCATAGTGGATC  Cow Horse Coastmole Human Rabbit VoleM. r . VoleM. a. VoleM. t . Mouse  «-CMXIST2/CMXISTREV2-> GCAGATGATCGTTGGCCAACTGTTGGCAGAAGAGGAATCCCTGCCTTCCTCAAGAGGAAC GCAGATGATCGTTGGCCAACAGGTGGCAGAAGAGGAATCCCTGTCTTCCTCAAGAGGAAC GCAGATGATCGTTGACCA-CAGGTGGCAGAAGAGGAAT ACAGATGATCGTTGGCCAACAGGTGGCAGAAGAGGAATTCCTGCCTTCCTCAAGAGGAAC ACAGATGATCG AAGGGAAATCC-TGCCTTTCTCAGGAGGAAC --AAA GCCTTGGCCA GAGGTGGAATCC-TGCCTCCTTCCACAGGCTC --AAA GCCCTGGCCA GAGGTGGAATCC-TGCCTCCTTCCACAGGCTC --AAA GCCCTGGCCA GAGGTGGAATCC-TGCCTCCTTCCACAGGCTC ACAAA TCACTGGTGAA GAGGTAGAATCC-TACCTTCTTCCAAAA  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  ACCTACCCCGTGGCTAATGCTGGGGTCGG ACCTACCCCGTGGCTAATGCTGGGGTCGG ATCTACCCCGTGGCTAATGCTGGGGTCGGGATATCTACCCNTGGCTATGGCTGGGGTCGG ACCTACCCCTT ATCTACCCCGTGGCTAATGCTGGGGTCGG GTCTACCCCTTGGCTAATGCTGGGGTTGA GTCTACCCCATGGCTAATGCTGGGGTTGA GTCTACCCCATGGCTAATGCTGGGGTTGA -TCTACCCCATGACTATTGCTGGGGTTGC  * * * * **  *  *  *  * * * *  *** *  IHIIHiHHHHHI  *SKXIST2  r* * * * * *  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM. t . Mouse Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  **  t r * * * * * * *  ATTTT TGATA ATTTTGATTTCTATTT CTTGGATGTCAGTCATATATAGTTTGATTATGTGGTTTGC ATTTTGATTTCTATTTATTTCTTGGATGTCAGTCATATATAGTTTGATTGTGTGGTTTGC TTTGATTTATATTTATCTTTTGGATGTCAGTCATACA—GTCTGATTTTGTGGTTTGC ATTTTGATTTATATTATTTTTTTGGATGTCAGTCTTA-ATAGTCTGACTATGTGGTTTGC AATTTCATTTATGTGAAGGTTTTGGATGTCAGTGAA GTTTGATAGTGTGGTTTGC AATTTCATTTATGTGAAGGTTTTGGATGTCAATGAA GTTTGATAGTATGGTTTGC AATTTGATTTATGTGAAGGTTTTGGATGTAAGTGAA GTTTGATAGTGTGGTTTGC ATTTTGATTTCAATGAATATTTTGGATGCCAACGACAC GTCTGATAGTGTGCTTTGC  TAGTGTTCGA-TTTAAGCCTTAAGTGACTAT-ATGGAAATGTATTTAGGGGCTTTATTTAAGTATTCTA-CTTAAACCTTAAGTGACTGA-ATGGAAATGTATTTAGGGACTTTGTTTA TAGTGTTTGAATTTAAGTCTTAAGTGACTATTATAGAAATGTATTAAGAGGCTTTATTTTAGTGTTCCAATTTAAGTCTTAAGTGACTAGTATAGAAATGTATTTTGTGACTTTTTTAA TAGTGCTTGAATCTAAAGCTGGAGTGATTATA  AAAATGTATTTAGGAGCTTTACTG-  TAGTGCTTGAATCTAAAGCTGGAGTGATTATA AAAATGTATTTAGGAGCTTTACTGTAGTGCTTGAATCTAAAGCTGGAGTGATTATA—AAAATGTATTTAGGAGCTTTACTGTAGTGTTTGAATTTAAAACCGAAGTGATTGTTTTCAAAATGTATTTACGGATTTGCTTACMXISTREV3-*  GTAGAATTC TCTGTAGGATTCATTTT ' 7 T-TTTCGGTGTCCCTAAAATTTC GTAGAATTCACTTTAATTACATTTAATGAGTTTTTGTTTTGAGTTCCTTAAAATTCC TTTGTAGAATTCATTTTAGTCGCACTTAATGAGCTCTCATTGCGCATTCCATAGAATCCC CTTGTAGAATTAATTTGAATCACC GTATGCTTTCACTCCAGAGTCCTTAGTGTTTCTTGTAGAATTCAATTGAATCACC GTATGCTTTCACTCCAGAGTCCTTAGTGTTTCTTGTAGAATTCAATTGAATCACC GTATGCTTTCACTCCAGAGTCCTTAGTGTTTCTTGTTGAATTCATTTTAATTACCTTTAGTGAATTGTTACTTTGGAGTCCTTAAAGTTT-  103  Cow Horse Coastmole Human Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  TTAATGTTTTCCATTTCTTTTTTGTAAATTGCTTAAAAGCACTTTGGCTT-TAGTTATTT TTAAAGTTTTTAGCTTCTCATT-ACAAATTCCTTAACCTTTTTTTGGCAG-TAGATAGTC TTAAATTTCTCTAT AAATACATAAACTAGTTGTGGCAGGCAGATAGCC TTAGCTTCTTTAA AATTCATTAA—CTTTTTTGGCAGAT-GATACTC TTAGCTTCTTTAA AATTCATTAA—CTTTTTTGGCAGAT-GATACTC TTAGCTTCTTTAA AATTCATTAA—CTTTTTTGGCAGAT-GATACTC TCAA TAA—TTTTTTTGGCAGAT-GATACTC  Cow Horse Coastmole Rabbit VoleM.r. VoleM.a. VoleM.t. Mouse  A--GT CAAATAATTT — GACATTTAA TAAGCT AAAGT CAAATCATTTCTAATGTTTTA AAAATG AAATC CACCTCATCT —GACATTTAA AAACTT AAGT CACTTAAATTT—ACGTTCTT TCAAACT AAGT CACTTAAATTT—ACGTTCTT TCAAACT AAGT CACTTAAATTT—ACGTTCTTCGTTCTTTCAATCAAACT AAATTACTTGGCACTTAAATGT—ACTTTCTT TCAAACT  PERCENT SEQUENCE SIMILARITY  Cow  Cow  Coast mole 75  Horse  77  84  Human  70  76  77  Rabbit  71  74  77  77  Vole M. r . Vole M.a. Vole M.t. Mouse  12  16  21  20  24  12  16  21  19  23  98  12  24  21  19  24  98  98  11  16  21  17  22  68  69  Horse  Human  Rabbit  Vole M. r .  Vole M.a.  Vole M.t.  68  Figure 12a: Aligned sequences in the 5'region of exon one. The Coast mole Xist fragment corresponding to the 5' region of exon one is aligned with the Xist seqeunces from human (M97168; bases 1074-2519), cow(AF 104906:1102-2182), rabbit (U50910;1899-3278), horse (U50911 ;4810-6055), mouse (L04961;2029-2237) and the three vole species, M. arvalis (AJ310129; 1031-2507), M. rossiameridionalis (AJ310130; 1041-2534), and M. transcaspicus (AJ310127:1034-2538). Primers designed from conserved sequence are green, mole-specific primers are red and the two Alul sites are highlighted in turquoise.  104  Rabbit Human Horse Coastmole  TGCTTGTTAGCAGCTTTGCATGAATTGTGGTACTGGGATGCCGGAAAACRAGATTGTCCT GGCTTGTTAGGAGCTTTGCGTGA-TTGTTGTATCGGGAGGCAGTAAGA ATCATCTT GGCTCGTTAGGAGCTTTGTATGAATTAGGGTATCGTGAGGTGGAGAAACAGGATGGTCCT AGCTTGTTAGGAACGCTGCAGTTACTCTAGTATCGCGAGGTCTCGAAACGGGGTTGTGCT *** * * * * * * * ** * * * * * ** * * * * * *  Rabbit Human Horse Coastmole  GCAGCAG--CACGAGGCGTTAACTTGAAAATG AGACAGGACTTAAAAGTGCTG TTATCAG--TACAAGGGACTAG-TTAAAAATGGAAGGTTAGGAAAGACT—AAGGTGCAG GTGTTATGTTAAAAGAGGCTAA-TCGAAAATGGGAG GGAGGAAACTT—AGGCACAG GT--CAC--TACCAGAGGCACG-TTGAGAATGACAA GGAGGGGAGTT-AAGGCACTG *  *  **  *  * ****  *  * *  * *  * *  Rabbit Human Horse Coastmole  GGATCAAAATGGCGCTTTTGCCATTGCAGCGTTTCTCAGCATGGCTCCCTGTGCTTTGTT GGCTTAAAATGGCGATTTTGACATTGCGGCATTGCTCAGCATGGCGGGCTGTGCTTTGTT GGTCCGAAATGGCGATTTTGACTTTGCGACATTGCTCAGCATGGCTGCCTG—CTTTGTT ACATCAAAATGGCGGCGTTAACTTTGTGGCGTTGCTCAGCATGGCTGCCTGT—TTTGTG  Rabbit Human Horse Coastmole  AAATCGTCTAAAATGGCGGATTCAGTTCGG-CCGCAACGTTAAAGTGGCGGGAAGCCCAC AGGTTGTCCAAAATGGCGGATCCAGTTCTG-TCGCAGTGTTCAAGTGGCGGGAAGGCCAC AG--TGTCCAAAATGGCGGACACAGTTTTG-CCGCAGTGTTCAAGTGGCGGGAAGGCCAC AA—TGTCCAAAATGGCGGATTCATTTTTTTCCGCAGTGTCCAAGTGGCGGGAAGGCCAC * *** *********** ** ** **** ** ************* ****  Rabbit Human Horse Coastmole  ATCATGATGGGTGGGGTCTTTTTTTGTAATGCGCAGCATGATGGTGRCAGAGTTCGGTCA ATCATGATGGGCGAGGCTTTGTTAAGTGGT TAGCATGGTGGTGGACATGTGCGGTCA ATCATGGTGG TGTCTTTGTTC—TAGTGCGCAGCATGGCGGTGGAAATATTCTGTTA ATCCTGGTGG CGTCTTTGTTC—TAATGTGCAGCATGGCGGTGGAAATATTCTGTTA  Rabbit Human Horse Coastmole  CACAGTGAAAGATGGCGGCTGAAACCCTTGCCGCAGTGTAAAATATGGCGGGCGTCTTTG CACAGGAAAAGATGGCGGCTGAAGGTCTTGCCGCAGTGTAAAACATGGCGGGCCTCTTTG CATGGCAAAAGATGGCGGCTGAAGTACTTGCCGCAGTATAAAATATGGCGGGC--TTTTG CATGCTAAAAGATGGCGGCTAGAGCACTTG TGTAA--CATGGCGGCC TTTG  Rabbit Human Horse Coastmole  TCTTTGCCGTGTGCATTTCGTGGCGGGTTTTGCCGCAGGGACAATATGGCTGACGATATC TCTTTGCTGTGTGCTTTTCGTGTTGGGTTTTGCCGCAGGGACAATATGGCAGGCGTTGTC TGTTTGCCGTGTGCATTTCCTGACAGGTTTTGCCGCAGGGACAATATGGCAGACCTTGTC TCTTTGCTGTGTGTGTTTCCTGGCAGGTTTTGCCTCGGGGATAGTATGGCAGGCCTTATT  ********  ***  ** * * *  **  **  *  * ***  ** **  *************  * ***** *****  *  * * * * **  * ** * * * * * * * * * * *  *  *  ******  ****  ****  ***  ****  *****  * + ** *  ********  ****  ********* * **** * ****** * *  * *  Rabbit Human Horse Coastmole  ATGTGGTTATCATGGAGATTTGTCACGTGGACGTCATGGCGCGT TTTT ATATGTATATCATGGCT-TTTGTCACGTGGACATCATGGCG ATGTGGATATCATGGCAGTTCGTCACGTGGACGTCATGGCGGGGGG ATTTT ATGTGGAAGTCTTGGCAATTTGTCACGTGGATGTCTTGGCAGAGGGTCCACAGGGGCGTT  Rabbit Human Horse Coastmole  GCTCGGTAAATT  ** **  ** **********  GGCGGGCTTTGCACAAGGTGGACCTACGGCATTGTTAAAGATGGC GGCTTGCCGCATTGTTAAAGATGGC GCCCGGTACATTCTTGGCGGGCTTTGCACCAGGTGGGCCTGCCGCATTGTTAAAGATGGC CGTCTGCATGTTCTTGGCGGGCTCAGCACCTGGTGGTATTGCCGCATTGTTAAAGATGGC *********!  Rabbit Human Horse Coastmole  GGG CGTTGCCGCATTTTGCCGGTC AAAGCAGGG TTTTGCCGCCTAGTGCCACGC AGAGCGGGGGTTTGGCGCGTTTTGGCGCGTTTTGCCGCGTTTGCTGCGTTTTGCCCCGAAAAGTGGGG CTTTGCCGCGTT-TGTGGCGG AAAGCNC  ***  Rabbit Human H  °  r S e  Coastmole  *******  **  , v*  GGAGAAAAC CGG- —GTGCTGGATTGCCGCATGACTTAACCAATCAGAAAT GGAGAAAAGGTGGGATGGACAGTGCTGGATTGCTGCATAACCCAACCAATTAGAAAT J—GAAGAAAGG-TGGGGTGGACAGTGTTGGACTGCCGCACGACCCACCCAATCGAGAAG GAAGGAGAAACA-TGGGGTGGGGAGTGNTCAATTGCTGCTTTACATACCCAATCAGAGA-  * ****i Rabbit Human Horse Coastmole  GAG-GGTGGA-TTAGACACAACCAATTAGCATGGAGGATGGAATTAGACTGTCGATGCAC GGG-GGTGGAATTGATCACAGCCAATTAGAGCAGAAGATGGAATTAGACTGATGACACAC GGGTGGTAGAATTGGGAACAACCAATTAG--TAGAGGATGGAATTAGACTA GCATGTTAGAAAAGAGTATGACCAATTAG—TGGAAGATGGAATTACANTGG * * * * *  *  ********  **  **********  * *  105  Rabbit Human Horse Coastmole  TGTCCAGGTACTGAGCATGGTCATGTGGGCTGAGTTAGCATGGCTCTTTGCAGCTTTCTT TGTCCAGCTACTCAGCGAAG—ACCTGGG-TGAATTAGCATGGCACTTCGCAGCTGTCTT TGAGGTAGTGTAGCACTGCGCTGTGGTCTT TGAGTTAGCATGGCTCCTCACTGAAGTCTT *** *** * ** * * * ****  Rabbit Human Horse Coastmole  TTTTCAGCCAGTCAGGACGAAGAAGTGGAGCGGCCATGTA-GTGTGCCC-CTCCCAGTGG T AGCCAGTCAGGAGAAAGAAGTGGAGGGGCCACGT--GTATGTC TCCCAGTGG TATTCAGCCAGTCAGGTGGAGGAAGCAGAGAGGTCACGTCTCTACGTCGGCTCCCAGTGG TTTTCAGCCAGTGAGAAGGCAGTA-TGGAGCAGCCACGTTTCTAAGTCGGCTCCCCGTGG  Rabbit Human Horse Coastmole  GTGGTATTAAAGTTCTTTCCAAGGTCTTTTCAAGGACATTTGGCCTTTCTACCTCCCTTC GCGGTACACCAGGTGTTTTCAAGGTCTTTTCAAGGACATTTAGCCTTTCCACCTCTGTCC GTGGTAA CCAAGGTCTTTCCAAGGACATCTGGCCTTTCCACCTCCCTCC GTGGTGC CCAAGGTCCTTCCAAGGACATGTGGCCTTTCCACCTCCCTCC  Rabbit Human Horse Coastmole  CCTCTTGTTTGTCCTCTCCTCTCCAGACATGGTCTCTTGCAGTGCTGGATATCTGGCTAT CCTCTTATTTGTCCCCTCCTGTCCAG-TGCTGCCTCTTGCAGTGCTGGATATCTGGCTGT CCTCTCACTTGCTCCCTCCCCTCCGG-CATTGACTCTTGCAGTGCTGGATAGCTGGCTGT CTCCTTCCCTGCTCCCTCCGCTCCAG-CATTACCTCTTGCACTGCTGGGT-TTTGACTGT * ** * * * ** * * *** * ******** ****** * ** ** *  Rabbi t Human Horse Coastmole  GTGGGCTGAACCCCACCCACATTCCTCTGTATTGATGCAATGGCCTTTGGCTAA GTGGTCTGAACCTCCCTC-CATTCCTCTGTATTGGTGCC-TCACCTAAGGCTAAGTATAC GCGGGCTGAACCTCACCC-CATTCCTCTGCGTTGCTGCTGTGGCCTCAGGCTAG GCTGGCTGAACCT-ACAC-AATTCCTCTTCATTGCTGAAGTGGCCTGAGGTTAGACTTA* * ******* * * ******** * * * ** * *** ** **  p . Human Horse Coastmole  TCTACTCCC GCCCTGG CTCCCCCCCCACCCCCCAACCCCCCCAACTCCCCACCCCC ACCCCCCACCCCCCACC TCTACTCCC CCCTGG CCTGGCCCCCTTGCCCCAGG  Rabbit Human Horse Coastmole  CCTGTGGTCCTCCCTGC ACTG TCCCCACCCCCCTACCCCCCTACCCCCCTACCCCCCTCTGGTCTGCCCTGCACTGCACTG CTTCTGCTCTTCC—GC ACTG CCCCGCTCTTCC- - AC TCTG  Rabbit Human Horse Coastmole  TGGCCACCGG--CAGTGCTCCAGGCCTGCCTGGTGTGGACATGGTGGTGAGCTGTGGCAA TTGCCATGGG—CAGTGCTCCAGGCCTGCTTGGTGTGGACATGGTGGTGAGCCGTGGCAA TGGCCAGGGG--CAGTGCTCCATGCCTGCCAAGTATGAATATGGCGGTGAGTCGTGGCCA TGGCCCCCGGAGCAGTGCTCCATGCTGGCCAAGTGTGAACATGGTGATCAGTCGAGGCAA  a b b  t  *  *******  * ***  **  * *  *******  ***  * **  **  *  **  * * * * * * * * * * **  ****  ** * * * * * * * * * * * * * * * * * * * * * *  ***  * ***  * *  **  **  *  ** ** * * * * * * * **  ****  * *  ***  * *** *  Rabbit Human Horse Coastmole  GGACCAGAATGGATCACAGATGATCG AAGGGAAATCCTGCC GGACCAGAATGGATCACAGATGATCGTTGGCCAACAGGTGGCAGAAGAGGAATTCCTGCC AGACCAGAATGGATCGCAGATGATCGTTGGCCAACAGGTGGCAGAAGAGGAATCCCTGTC GGACCAGAATGGATCGCAGATGATCGTTGACCA-CAGGTGGCAGAAGAGGAATCCTTGCC ************** ********** * **++ * ** *  Rabbit Human Horse Coastmole  TTTCTCAGGAGGAACATCTACCCCGTGGCTAATGCTGGGGTCGG TTCCTCAAGAGGAACACCTACCCCTTGGCTAATGCTGGGGTCGG TTCCTCAAGAGGAACACCTACCCCGTGGCTAATGCTGGGGTCGG TTTCTCAAGAGGAACATCTACCCCGTGGCTAATGCTGGGGTCGGGATATCTACCCNTGGC ** **** ******** ******* *******************  Rabbit Human Horse Coastmole  ATTTTGATTTATATTATTTTTTTGGATGTCAGTCTTAA-TAGTCT ATTTTGATTTATATTTATCTTTTGGATGTCAGTCATA—CAGTCT ATTTTGATTTCTATT TCTTGGATGTCAGTCATATATAGTTT TATGGCTGGGGTCGGATTTTGATTTCTATTTATTTCTTGGATGTCAGTCATATATAGTTT ********** **** * ************* ** *** *  Rabbit Human Horse Coastmole  GACTATGTGGTTTGCTAGTGTTCCAATTTAAGTCTTAAGTGACTAGTATAGAAATGTATT GATTTTGTGGTTTGCTAGTGTTTGAATTTAAGTCTTAAGTGACTATTATAGAAATGTATT GATTATGTGGTTTGCTAGTGTTCGA-TTTAAGCCTTAAGTGACTA-TATGGAAATGTATT GATTGTGTGGTTTGCAAGTATTCTA-CTTAAACCTTAAGTGACTG-AATGGAAATGTATT ** * ********** *** ** * **** *********** ** **********  106  Rabbit Human Horse Coastmole  TTGTGACTTTTTTAATTTGTAGAATTCATTTTAGTCGCACTTAATGAGCTCTCATTGCGC AAGAGGCTTT ATTTGTAGAATTCACTTTAATTACATTTAATGAGTTTTTGTTTTGA TAGGGGCTTT ATTTGTAGAATTC TAGGGACTTTGTTTATCTGTAGGATTCATTTTAAGGCCAATTAATGAGTTCAT-TTTCGG * * **** ** ***** ****  Rabbit Human Horse Coastmole  ATTCCATAGAATCCCTTAAA TTTCTC TATAAATACATAAAC-TAGTTGT GTTCCTTAAAATTCCTTAAAGTTTTTAGCTTCTCAT-TACAAATTCCTTAACCTTTTTTT  Rabbit Human Horse Coastmole  GGCAGGCAGATAGCCAAATCCACCTCA--TCTGACATTTAAAAACTT GGCAG-TAGATAGTCAAAGTCAAATCATTTCTAATGTTTTAAAAATG  TGTCCCTAAAATTTCTTAATGTTTTCCATTTCTTTTTTGTAAATTGCTTAAAAGCACTTT  GGCTT-TAGTTATTTAG—TCAAATAA—TTTGACATTTAATAAGCT  Figure 12b: Aligned sequences in the 5'region of exon one (without rodents, cow) The Coast mole Xist fragment corresponding to the 5' region of exon one is aligned with the human, equine and lepine sequences. The rodent sequences were omitted so that the high level of homology among the genes of the remaining species could be visualized. Although the bovine gene also shows high homology to the sequences in this figure, it was not included due to its small size. The locations of the various primers and the sequence similarities are shown in Part A of this figure.  107  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  AGCTCCA-TCACCCTGTCCCGTCTGCCTCTTTG CTGGGC C AGTTCCA-TCACCCTGTCCCGTCTGCCTCTTTG CTGGGC C AGTTCCA-TCACCCTGTCCCGTCTGCCTCTTTG CTGGGC C AATTCCA-TCATGCCATCCCTTTTGCCTCTTTGTTTGAACAGTATTGACTGGGCAAAGCC AGCTGCACTTATCTTCATCC TGCCCATCAGAC ATGGAC C AGTTGCACTTTCCTTGGTCC ACCCATTATAC ATGAAC C  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  CTTCTCTTGACTTAAAGTCAACAATTCCAGACGTCATCAGCTTACCTGCTGCACATGGCT CTTCTCTTGACTTAAAGTCAACAATTTCAGACGTCATCAGCTTACCTGCTGCACATGGCT CTTCTCTTGACTTAAAGTC AATTCCAGACGTCATCAGCTTACCTGCTGCACATGGCT CTTCTCTTGACTTAAAGTCAACAA CACCAGTTTACTCACTTCATATGGCT CTTCC ACGT CCTTCTTT CCTCT ACTT CCTT--TCGCATTGCT * ** ** * * *****  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  GCAGTGTTTCAGTTGCCCTCTGCTCGGTCCTCCCTGTTGAATGGAAACACCTCCACCGCA GCAGTGTTTCAGTTGCCCTCTGCTCGGTCCTCCCTGTTGAATGGAAACACCTTCACCGCA GCAGTGTTTCAGTTGCCCTCTGCTTGGTCCTCCCTGTTGAATGGAAACACCTCCACCGCA ACAGTGTCTCAGTTGCCTTCTCCTTG CTCCC-ACTGAACAGAGACACCTCGAATTCGA CTACCCA AAGCCCC TCTGAGTA TGCTGA CTACCCA AAGCCCCTT  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  CTACACTGTCAATTGCCTGTGTGTCATTAATGAACACAGCACCGATGGACAGTCCCACTT CTACACTGTCAATTGCCTGTGTGTTATTAATGAACACAGCACCGATGGACAGTCCCACTT CTACACTGTCAATTGCCTGTGTGTCATTAATGAACACAGCACCGATGGACAGTCCCACTT TTACATTATT CTGGGTAATGTTAATTACCCCAA CTGTGT—TCTTAATTAACACAGTAATG TCCCATTT CTGTGT—TATTAATAAACACAGTACTGATTG TCCCATTT  4-CMXIST11  *  *  **  *  ***  **  **  **  *****  **  *  *  *  *  **  **  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  TGCAGCCCATGTGTCCCAGAGCTCCCTGCAACTTTGATGCATTTGTACA-TGTAAAATTT TTCAGCCCATGTGTCCCAGAGCTCCCTGCAACTTTGATGCATTTGTACA-TGTAAAATTT TTCAGCCCATGTGTCCCAGAGCTCCCTGCAACTTTGATGCATTTGTACA-TGTAAAATTT -ACACCCTATGTGTCATTAA TA-AATTTTGGTGTATTTATACA-CT—GAATAG TTCAGCCCATCGGCCCAAGATCTCCATGT TGTGTATGTGCGGTGTTGACTAC TTCAGCCCATCAGTCCAAGATCTCCCTACCACTTTGGTGTGTTGGTGCAGTGTTGACTAT ******** * * ** * * * * *  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  CAAAAGCGGGCCAAAGCTAGGTGGATGAAGCATCGCTCTTCAGCCT-CATTTTTTAATTA CAAAAGCGGGCCAAAGCTAGGTGGATGAAGCATCGCTCTTCAGCCT-CATTTTTTAATTA CAAAAGCGGGCCAAAGCTAGGTGGATGAAGCAACGCTCTTCAGCCT-CATTTTTTAATTA CAAAAGCAGGCCAAAACTAGGTGGATGAGCCTTCAATCTTTAACTTGCACTTCTAAATTA CAAAAGCAGGCCGAAACTCGGTGGATGGACCCTCCTTCCTTTTCCTGCATTTGTTAATGA GAAAAGCAGGCCTGAACTAGGTGGATAAGCCTTCACTCATTTTCTTTCATTTATTAATGA ****** **** * ** ******* * * ** * * * * * * * * *** *  VoleM.r. VoleM.t. Vo1eM.a. Mouse Coastmole Human  CTCCAATTCCAATTGCTGGCATATTCTAGGGCCAGGAATCATTCCTGCCCACCTTTATTA TTCCAATTCCAATTGCTGGCATACTCTAGGGCCAGGAATCATTCCTGCCCACCTTTATTA CTCCAATTCCAATTGCTGGCATATTCTAGGGCCAGGAATCATTCCTGCCCACCTTTATTA TTCCAATTCCAACTGCTGGCACATTCTAGGGCCAGGAACCATTCTTGCCTACCTTTATTA TTCCAACACCAGTC GTCAAATTCTGGGGGCAGGAACCATCCTTGCCCACCTCTGTTA TCCTAGTTTCAATTATTGTCAGATTCTGGGGACAAGAACCATTCTTGCCCACCTGTGTTA * * ** * ** * *** *** ** *** *** * **** **** * ***  VoleM.r. VoleM.t. VoleM. a. Mouse Coastmole Human  ATGCTTTATTGTGCAAAAAAAATATTTCAGGCAACTT--GCTCAGGGAGCTGGATTGCCA AT-CTTTATTGTGCAAAAAAA-TATTTCAGGCAACTT—GCTCAGGGAGCTGGATTGCCA AT-CTTTATTGTGTAAAAAAA-TATTTCAGGCAACTT—GCTCAGGGAGCTGGATTGCCA ATGCTTTATTGTGCAAAA TATTGCAGGCAAGTA--GCTCAGGGAGTTGGATTGCCA CTGCTTTACTGGGCAAAA TGCCCAAGATAAGCCAGGCCCACAGAACTGGATTGCTG CTGCTTTACTGTGCAAAA TACTGAAGGCAAGTCAGACCCAGGGAGCTGGATTGCCA * ***** ** * **** * ** ** * ** ** ********  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  CCTTTTATTTGAGGCTTTCCTTA--TAC-AGTATCAACTGAAAACTGTCTCCCAAAGAAG CCTTTTATTTGGGGCTTTCCTTA--TAC-AGTATCAACTGAAAACTGTCTCCCAAAGAAG CCTTTTATTTGGGGCTTTCCTTA--TAC-AGTATCAACTGAAAACTGTCTCCCAAAGAAG CCTTTTACTTGGGGCTTTCCTT TAC-AGTATGAACTGAAAATTGTCTTCCTGAGAAG TCTTTTATTTTGGATGTTCAGTAGTTACCATTATCAGTTGAAAATTGTCTCCCCAGGAAA TCCTTTATTTTGTGTTTCCAGTG--TAC-ACTAT AAAATTGTCTCCCCAGGAAG * ****  **  * *  *  *** * ***  ****  *****  **  ***  108  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  GAGGATGAGCTAGTACTTT-CTTTACACCCTTCCTCCAGGA GCTGACCGCCTGCTT GAGGATGAGCTAGTACTTT-CTTTACACCCTTCCTCCAGGA GCTGACCGCCTGCTT GAGGATGAGCTAGTACTTT-CTTTACACCCTTCCTCCAGGA GCTGACCGCCTGCTT GAAGCT TAGCACTTTTCTTTCCATTCTTCCTCCAGGAAGGAGCCAACTGTCTGCTT GAAG-T TAGCGTTGT-CTCT TTCCTGA GCAGAGTGCCTGGCT GAAGGT TGGCACTTT-CTCTGCATTCTTCTTTCCAGA GCAGATTGCCTGGTT ** * * * * * * ** * * * ** ** * * * * * *  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  AACAAACTC—AACTTGTTTCTC--TTGGTATATTGCTAC-GTACAGTGCCAACTGCCAG AACAAACTC—AACTTGTTTCTC—TTGGTATATTGCTAC-GTACAGTGCCAACTGCCAG AACAAACTC—AACTTGTTTCTC—TTGGTATATTGCTAC-GTACAGTGCCAACTGCCAG AAGAAACTTTAAGCCCGATTTT GTATATTGCTACTGTACAGGACCAACTGCCAG GTAAATCTC TCATC TTATATATTGCTATTGTAAAGTGCCAGTTGCCAG AAGAATCTC TTGTTGTCCCTTCTGTATATTGTTATTGTAAAGTGCCAAATGCCAG * * * * * * * ** * * * ** **; ******  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  -AAAGTTATTGATAGTTTATGCCTTTAAGAAAGGCCTCTGG -AAAGTTATTGATAGTTTATGCCTTTAAGAAAGGCCTCTGG -AAAGTTATTGATAGTTTATGCCTTTAAGAAAGGCCTCTGG -AAAGTTATTGATAATTTTATTCCTTAAGAAAGGCATTTGG AG-GCAACCA—AAAGTTGCTATTAATTTTTTA TTTTCTAAAGAAAGGTATCTGG GATACAGCCAGAAAAATTGCTTATTATTATTAAAAAAATTTTTTTAAGAAAGACATCTGG *** ** * * ** * ******* * ***  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  CMXIST11-* CTTGCCAGGTGGAATTGATTATGTGATCATTAGCTTTTCTGAAGTAAAAATATCTTGCTT CTTGCCAGGTGGAATTGATTATGTGATCATTAGCTTTTCTGAAGTAAAAATATCTTGCTT CTTGCCAGGTGGAATTGATTATGTGATCATTAGCTTTTCTGAAGTAAAAATATCTTGCTT ATTGCAAGGTGGA-TTGACTGTGAGATCATTAGCTTTTGTGAAGTAAAAATAGCC-ATTT AATAT ' TCATTAAATATT-TGAAGTCCTAAAACCC-ATTT ATTGTAGGGTGGACTCGATAACCTGGTCATTATTTTTT-TGAAGCCAAAATATCC-ATTT * ****** * *: ** * *  VoleM.r. VoleM.t. VoleM. a . Mouse Coastmole Human  CTACTATATATCTGATGACCAAGTAGTGTCTCTCATTTTA-CTGAGGGTGGCGAGTCTGT CTACTATATATCTGATGACCAAGCAGTGTCTCTCATTTTA-CTGAGGGTGGCGAGTCTGT CTACTATATATCTGATGACCAAGCAGTGTCTCTCATTTTA-CTGAGGGTGGCGAGTCTGT GTGTCATGTTTCTGAAGACTAAGCAGTGTCTC—AGTGTA-CTGAGGGTGATGAGTCTGT ATATCATGTACCTAATGAC CAGTGTCTCTCATTTTA-CTAAGGGTGGTGGGTCTGN ATACTATGTACCTGGTGAC CAGTGTCTCTCATTTTAACTGAGGGTGGTGGGTCTGT ** * * * * * * * ***** ** * ** ******** * *  Vo1eM.r. Vo 1 eM. t . Vo1eM.a. Mouse Coastmole Human  GGAGAAACCACTGTAACTGTTATTCTAATACCGTCAAAGTGGTTGTAGAATGGCCTGCTT GGAGAAACCACTGTAACTGTTATTCTAATACCGTCAAAGTGGTTGTAGAATGGCCTGCTT GGAGAAACCACTGTAACTGTTATTCTAATACCGTCAAAGTGGTTGTAGAATGGCCTGCTT GGAAAGATCAGTGCAACTATT GCAGAAT GTTT GGATAGACCACTGTGACTCTTGCT ACTTTATTAT TGCTC GGATAGAACACTG—ACTCTTGCT ATTTTAATAT CA *** * * ** ** * * * ** * **  VoleM. r . VoleM. t . Vo1eM.a. Mous e Coastmole Human  AAGGCCACGATGTATGTTTGGTCTGTACTACCAGTTTAA ATACAAAAGTCAATCTT AAGGCCAGGATGTATGTTTGGTCTGTACTACCAGTTTAA ATACAAAAGACAATCTT AAGGCCAGGATGTATGTTTGGTCTGTACTACCAGTTTAA ATACAAAAGTCAATCTT AAGACAAGTATCTTTGCTTGGTCTTTACTACAAGTTTAACAAAACGAAAAAGTCAATCTT AAGTTAGTGA GTTTTCAAGTGAAACT CTTAA AGACCGGT—ACCTT AAGATA TTCTAGAGTGGAACT CTTAA GACCAGT—ATCTT  ***  *  *  ***  ****  *  *  * ***  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  TGTGTGGCCTTCAGTATCCTTATCTTTAAGGAAAATGACC-AAACATTTTAATATTTAAA TGTGTGGCCTTTAGTATCCTTATCTTTAAGGAAAATGACC-AAACATTTTAATATTTAAA TGTGTGGCCTTTAGTATCCTTATCTTTAAGGAAAATGACC-AAACATTTTAATATTTAAA TGTGTGGCCTTTAGTATGATTAACTTTTTGGAAGATGACCTAAGCCTTCTAATCATTATA TG GGCTCTACCACCATTCATTTTTAGAAAAACTATGGAAACTTTATAGTCCTTAAA TGTGTGGGCTTTACCAGCATTCACTTTTAGAAAAACTACCTAAATTTTATAATCCTTTAA ** * * * * * ** * * * * ** * * ** * * * * * * * *  VoleM. r . VoleM. t . VoleM. a. Mouse Coastmole Human  TTTGGCCTGACATTGAACGCCAGCC CCTACTGAT GGGCTAC TTTGGCCTGACATTGAACGCCAGCC CCTACTGAT GGGCTGC TTTGGCCTGACATTGAACGCCAGCC CCTACTGAT GG-CTGC TTTTGTCTGACATTGGTCACCAGTC CTTGCTTATTTTTAAAAGGTGACT — TTT—CTTCATCTGGAGCACCAACCAGTCCAGTCCTTATTTATTTCAATAAGATTGTTAT TTT--CTTCATCTGGAGCACCTGCC CCTACTTATTTCAAGAAGATTGCAGT  ***  * * * * * * * *  * * * * *  +  109  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  GAAAGGATTAAGTGAGA GGACATGT—CAAGTTGTCTTTGAG TAGGCCA GAAAGGATTAAGTGAGA GGACATGT—CAAGTTGTCTTTGAG TAGGCCA GAAAGGGTTAAGTGAGA GGACATGT—CAAGTTGTCTTTGAG TAGGCCA GGATGGATTAAATTTGA GAACATGT—CAAGTCGCCTTTGAAAATTATATAGGCCA AAAACAATTATATGAGAGAATGAACATATGCTGAGGTGCTTTTGAAAACCA—TAGATCA AAA AC G AT T AAAT GAGG GAACATATGCAGAGGTGCTTTTAAAAAGCA—TATGCCA * *** * * * **** * ** * * * * * ** **  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  CCATGT--GATTAATTCATTCTATC TGCC- -ATTAAATTTGGGCAATGATTTGA CCATGT--GATTAATTCATTCTATC TGCC- -ATTAAATTTGGGCAATGATTTGA CCATGT--GATTAATTCATTCTATC TGCC- -ATTAAATTTGGGCAATGATTTGA TCACATTTAATTAATTCATTCTATC CACC- -ATTAAACTCTGGCAATAATTTGA CCTTCTATGATTAATCTATTACATAAAATGCACCCTGTTTTAATTATGCTAAT--TNTAA CCTTTTTT-ATTAAT-TATTATATAAAATGAAGC ATTTAATTATAGTAATAATTTGA ******  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  SKXIST7-JV AGTAGCTTGAAAA AAAAAA CCCACCCTAAAG-TGG—GAACTT AGTAGCTTGAAAA AAAAAAAAAACCCACCCTAAAG-TGG—GAACTT AGTAGCTTGAAAA AAAAAA CCCACCCTAAAA-TGG--GAACTT AGTAGCTTGAAAATTCCTAAAGTGGGAATTTATTTTAGAGATGATAGAACCTGTTTCCCC AGTATTCCGAA GCA CCACACTCAAC-TGA—GGATTTAG AGTAGTTTGAA G A  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  AAATAATAT ATATATATATCCAGATCTAATCATTC—TTGAGAGTACACTTCAA AAATAATAT ATATATATATCCAGATCTAATCATTC—TTGAGAGTACACTTCAA AAATAATAT ATATATA—TCCAGATCTAATCATTC—TTGAGAGTACACTTCAA ACTTTACATTTTAAAATATGTCTGCCAGGATCTAATCATTCCTTTAAACGTACACTTCAA AACTGATG GGA-CCAGTTTTTC TTTGTTATAT AAATGATA AGA-CGAGTTCC-C TATTTTATAA **  * *  VoleM. r . VoleM. t . VoleM. a . Mouse Coastmole Human  AGAGAGATTCGCCTATTAAGAAAATATCT GTCTCTATTGGCGCCGAGTATTTA—A AGAGAGATCCGCCTATTAAGAAAATATCT GTCTCTATTGGCGCCGAGTATTTA—A AGAGAGATCCGCCTATTAAGAAAATATCT GTCTCTATTGGCGCCGAGTATTTA—A AGAGAGATTTTCCTAGTAAGAAAAGAGCT TTCTCTAGTG-TGAAGGGTGCTTTGTA TAAAAGA AAG ATC TTTTTG GGGTAT GAAAAAT AAGCCAAAATTAAATATTCTTTT GGATAT  VoleM. r . VoleM.t. VoleM.a. Mouse Coastmole Human  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole  GCTTTCTGGT-TTTCGTTCTTT AATTATTGTGCATAAATGAGTGTGTA GCTTTCTGGT-TTTCGTTCTTT AATTATTGTGCATAAATGAGTGTG GCTTTCTGGT-TTTCGTTCTTT AATTATTGTGCATAAATGAGTGTGTGTGTGTG GCCGCCGAGTACTTAGGTCTTTTTTGGGAGCTATTGTGTA TGAGTGTA AATTTTTAAAATTTCA-ACA GTGAGA-TA * * * * * * * * * ••CMXISTREV5 TGTGTGTGTGTGTGTGTGTGCATGTTTGCATGTGCACTTGAGTATCTGTACTCGTG TGTGTGTGTGTGTGTGTGTGCATGTTTGCATGTGCACTTGAGTATCTGTACTCGTG TGTGTGTGTGTGTGTGTGTGTGTGCATGTTTGCATGTGCACTTGAGTATCTGTACTCGTG TGTATGTGTGTGTGTACATGCATGTT-GC-TGCGCGC AGTCATT TAAAAGTGAAT CATATTC CAGCCTCT GC-TGCCTA-GTGGAAATGAAT AATATCC CAGCCACT  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  CATTTGCATTGGTGACCAGAGAGCAATGGGAACTGGTTC TACCTTGTGGGTCCCTGG CATTTGCATTGGTGACCAGAGAGCAATGGGAACTGGTTC TACCTTGTGGGTCCCTGG CATTTGCATTGGTGACCAGAGAGCAATGGGAACTGGTTC TACCTTGTGGGTCCCTGG CATTCACAT-GGTGCTCAGACAACAATGGGAGCTGGTTCGTCTATCTTGTGGGTCC-TGG TGTATACAG-GCTGTTTGG-TAGCA—GAGAAA ATTTAGTAAG AGTGTACAG-GGTGTTTTG-TGGCA--CAGGATT ATGTAATATG  Vo1eM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  AGATAAAAACTCAGATCGTCAGGTTTGGCAGCAAGTGCCTTCACTCTCTGAG--CCATCT AGATAAAAACTCAGATCGTCAGGTTTGGCAGCAAGTGCCTTCACTCTCTGAG--CCATCT AGATAAAAACTCAGATCGTCAGGTTTGGCAGCAAGTGCCTTCACTCTCTGAG—CCATCT AGATCAAAG-TGAGATCATCAGGCTTGGCAGCAAGTGCCTTTACCCTCCGCGTGCCATCT GAACT GCTGA—AACAAGTACCT AG--TTATCGAACT GCTCA--AGCAAATAACT AG--TCATC*  ***  *  110 VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  TGCCAGCCC ATGTGACATTGCTGATGAAACCA TGCCAGCCC ATGTGACATTGCTGATGAAACCA TGCCAGCCC ATGTGACATTGCTGATGAAACCA TGCCATCCCGCTGCTGAGTGTTTGATATGACATTGCTGATGAAAATAATCATCACAACAG , ACAACA G  VoleM. r. VoleM. t . VoleM.a. Mouse Coastmole Human  CMXISTREV6-* CAGTTCTTC-AA-GTTTCTGAAAAAAAAAAAAAACAATATTTTTCTGAGAAGGATGT-CA CAGTTCTTC-AAAGTTTCTGGAAAAAAAAAAAA—AATATTTTTCTGAGAAGGATGT-CA CAGTTCTTC-AAAGTTTCTGAAAAAAAAAAAAC—AATATTTTTCTGAGAAGGATGT-CA CAGTTCTCCCAGCATTACTGAGAAATGA TACTATTTTTCTGAGGAGGATGTTCA . CNCTGAAAAAGGA TACTATTCCTCTNAGAAGGATGC-CN CAGTTCTTT-GTAATAACTGAAAAAGAA TATTGTTTCTCGGAGAAGGATGT-CA  VoleM.r. VoleM.t. VoleM.a. Mouse Coastmole Human  AGAAA-TCGACCCAGTT AGAAA-TCGACCCAGTT AGAAA-TCGACCCAGTT AGTAACTCATCCCAGTG AAANA-TCCGCCeMHI AAAGA-TCGGCCCAGCT  A  *******  Coast mole  *  Human  ***  ***  *  Mouse  * * * * * * * *  V o l e M.r.  Human  77  Mouse  44  54  V o l e M.r.  40  51  72  V o l e M.a.  39  51  72  98  V o l e M.t.  40  51  73  98  ******  *  V o l e M.a.  98  Figure 13a: Aligned sequences at the end of exon one. The alignment of Coast mole Xist sequence at the end of exon one with the corresponding Xist fragments from human (M97168; bases 10011-11295), mouse (L04961; 7861-9422), vole Microtus rossiameridionalis (AJ310130; 6325-7889), vole M. arvalis (AJ310129; 6516-8080) and vole M. transcaspicus (AJ310127; 6331-7893). Primers derived from sequence conserved between the human gene and a short bovine EST (BE483406; not shown) are green, while those that are mole-specific are red. Note the flanking Alul sites highlighted in turquoise.  111  Coastmole Human  AGCTGCACTTATCTTCATCCTGCCCATCAGACATGGACCCTTCCACGTCCTTCTTTGCAT AGTTGCACTTTCCTTGGTCC-ACCCATTATACATGAACCCCTCTACTTCCTT—TCGCAT ** ******* *** *** ***** * ***** **** ** ** ***** * ****  Coastmole Human  TGCTTCTGAGCAGTGCTGACTACCCAAAGCCCC—CTGTGTTCTTAATTAACACAGTAAT TGCTTCTGAGTA-TGCTGACTACCCAAAGCCCCTTCTGTGTTATTAATAAACACAGTACT ********** * ******************** ******* ***** ********* *  Coastmole Human  G TCCCATTTTTCAGCCCATCGGCCCAAGATCTCCATGT TGTGTATGTG GATTGTCCCATTTTTCAGCCCATCAGTCCAAGATCTCCCTACCACTTTGGTGTGTTGGTG *  ******************* * *********** *  *****  ***  Coastmole Human  CGGTGTTGACTACCAAAAGCAGGCCGAAACTCGGTGGATGGACCCTCCTTCCTTTTCCTG CAGTGTTGACTATGAAAAGCAGGCCTGAACTAGGTGGATAAGCCTTCACTCATTTTCTTT * ********** *********** **** ******* ** ** ** ***** *  Coastmole Human  C AT T T GT T AATG ATTCC AACACCAGT C GTCAAATTCTGGGGGCAGGAACCATCCTTG CATTTATTAATGATCCTAGTTTCAATTATTGTCAGATTCTGGGGACAAGAACCATTCTTG ***** ******** * * ** * **** ********* ** ******* ****  Coastmole Human  CCCACCTCTGTTACTGCTTTACTGGGCAAAATGCCCAAGATAAGCCAGGCCCACAGAACT CCCACCTGTGTTACTGCTTTACTGTGCAAAATACTGAAGGCAAGTCAGACCCAGGGAGCT ******* **************** ******* * *** *** *** **** ** **  Coastmole Human  GGATTGCTGTCTTTTATTTTGGATGTTCAGTAGTTACCATTATCAGTTGAAAATTGTCTC GGATTGCCATCCTTTATTTTGTGTTTCCAGTG--TAC-ACTAT AAAATTGTCTC *******  Coastmole Human  ** * * * * * * * * *  * * ****  *** * ***  ***********  CCCAGGAAAGAAG-TTAGCGTTGTCTCT TTCCTGAGCAGAGTGCCTGGCT CCCAGGAAGGAAGGTTGGCACTTTCTCTGCATTCTTCTTTCCAGAGCAGATTGCCTGGTT ********  ****  ** **  * *****  ****  *******  *******  *  Coastmole Human  GTAAATCTCTCATC TTATATATTGCTATTGTAAAGTGCCAGTTGCCAGAG-GC AAGAATCTCTTGTTGTCCCTTCTGTATATTGTTATTGTAAAGTGCCAAATGCCAGGATAC ******* * * ******* *************** ****** *  Coastmole Human  AACCA—AAAGTTGCTATTAATTTTTTA TTTTCTAAAGAAAGGTATCTGGAATAT AGCCAGAAAAATTGCTTATTATTATTAAAAAAATTTTTTTAAGAAAGACATCTGGATTGT * ***  ***  *****  * * * * ** *  ****  * *******  *******  * *  Coastmole Human  CAGGTGGAGTTGATAACCTGGTCATTAAATATTTGAAGTCCTAAAACCCATTTATATCAT AGGGTGGACTCGATAACCTGGTCATTATTTTTTTGAAGCCAAAATATCCATTTATACTAT ****** * **************** * ******* * ** * ********* **  Coastmole Human  GTACCTAATGACCAGTGTCTCTCATTTTA-CTAAGGGTGGTGGGTCTGNGGATAGACCAC GTACCTGGTGACCAGTGTCTCTCATTTTAACTGAGGGTGGTGGGTCTGTGGATAGAACAC ****** ********************* ** *************** ******* ***  Coastmole Human  TGTGACTCTTGCTACTTTATTATTGCTCAAGTTAGTGAGTTTTCAAGTGAAACTCTTAAA TG—ACTCTTGCTATTTTAATATCA AAGATA TTCT AGAGTGGAACT CTTAA** ********** **** *** *** ** ** * **** *********  Coastmole Human  GACCGGTACCTTTG GGCTCTACCACCATTCATTTTTAGAAAAACTATGGAAACTTT GACCAGTATCTTTGTGTGGGCTTTACCAGCATTCACTTTTAGAAAAACTACCTAAATTTT **** *** ***** **** ***** ****** ************** *** ***  Coastmole Human  ATAGTCCTTAAATTTCTTCATCTGGAGCACCAACCAGTCCAGTCCTTATTTATTTCAATA ATAATCCTTTAATTTCTTCATCTGGAGCACCTGCC CCTACTTATTTCAAGA ***  *****  *********************  **  * ** * * * * * * * * *  *  Coastmole Human  AGATTGTTATAAAACAATTATATGAGAGAATGAACATATGCTGAGGTGCTTTTGAAAACC AGATTGCAGTAAAACGATTAAATGAGG GAACATATGCAGAGGTGCTTTTAAAAAGC ****** ****** **** ***** ********** *********** **** *  Coastmole Human  ATAGATCACCTTCTATGATTAATCTATTACATAAAATGCACCCTGTTTTAATTATGCTAA ATATGCCACCTTTTTT-ATTAAT-TATTATATAAAATGAAGC ATTTAATTATAGTAA ***  ******  * * ******  *****  ********  * *  *********  ***  112  Coastmole Human  T--TNTAAAGTATTCCGAAGCACCACACTCAACTGAGGATTTAGAACTGATGGGACCAGT TAATTTGAAGTAGTTTGAAGTACCACACTGAGGTGAGGACTTAAAAATGATAAGACGAGT * * * ***** * **** ******** * ****** *** ** **** *** ***  Coastmole Human  TTTTCTTTGTTATATTAAAAGAAAG ATC TTTTTGGGGTATAATTTTTAG TCC-CTATTTTATAAGAAAAATAAGCCAAAATTAAATATTCTTTTGGATATAAATTTCAA * ** * ***** **** *** ** * *** ** ***** *** *  Coastmole Human  CAGTGAGTT-GTTGCCTGATAAAAGTGAATCATATTCCAGCCTCTTGTATACAGGCTGTT CAGTGAGATAGCTGCCTAGTGGAAATGAATAATATCCCAGCCACTAGTGTACAGGGTGTT ******* * * ***** * ** ***** **** ****** ** ** ****** ****  Coastmole Human  TGGTAGCAGAGAAA-ATTTAGTAAGGAACTGCTGAAACAAGTACCTAGTTATCAGAACAG TTGTGGCACAGGATTATGTAATATGGAACTGCTCAAGCAAATAACTAGTCATCACAACAG * ** *** ** * ** ** ** ********* ** *** ** ***** **** *****  Coastmole Human  CAGTTCTTTGTAATCNCTGAAAAAGGATACTATTCCTCTNAGAAGGATGCCNAAANATCC CAGTTCTTTGTAATAACTGAAAAAGAATATTGTTTCTCGGAGAAGGATGTCAAAAGATCG ************** ********* *** * ** *** ********* * *** ***  Coastmole Human  GCCCAGCT GCCCAGCT ********  Figure 13b: Aligned sequences at the end of exon one, excluding the rodents. An alignment of the Coast mole Xist fragment located at the 3' end of exon one with the corresponding human sequence. The rodent genes were removed to highlight the conservation between the Coast mole and human genes. For primer locations and levels of sequence similarity see Part A of this figure.  112  Coastmole Human  TTTTCTTTGTTATATTAAAAGAAAG  ATC  TTTTTGGGGTATAATTTTTAG  Coastmole Human  CAGTGAGTT-GTTGCCTGATAAAAGTGAATCATATTCCAGCCTCTTGTATACAGGCTGTT  TCC-CTATTTTATAAGAAAAATAAGCCAAAATTAAATATTCTTTTGGATATAAATTTCAA * ** * * * * * * **** *** ** * *** ** * * * * * *** *  CAGTGAGATAGCTGCCTAGTGGAAATGAATAATATCCCAGCCACTAGTGTACAGGGTGTT  ******* * * *****  *  ** * *  Coastmole Human  TGGTAGCAGAGAAA-ATTTAGTAAGGAACTGCTGAAACAAGTACCTAGTTATCAGAACAG  Coastmole Human  CAGTTCTTTGTAATCNCTGAAAAAGGATACTATTCCTCTNAGAAGGATGCCNAAANATCC  Coastmole Human  GCCC§GCT  TTGTGGCACAGGATTATGTAATATGGAACTGCTCAAGCAAATAACTAGTCATCACAACAG * * * * * * ** * ** ** * * * * * * * * * * * * * * * * ** * * * * * * * * * * * * * *  CAGTTCTTTGTAATAACTGAAAAAGAATATTGTTTCTCGGAGAAGGATGTCAAAAGATCG ************** * * * * * * * * * * * * * ** * * * * * * * * * * * * * *** ***  GCCCAGCT  ********  Figure 13b: Aligned sequences at the end of exon one, excluding the rodents. A n alignment of the Coast mole Xist fragment located at the 3' end of exon one with the corresponding human sequence. The rodent genes were removed to highlight the conservation between the Coast mole and human genes. For primer locations and levels of sequence similarity see Part A of this figure.  113  Based on the organization of the human XIST gene, the distance between the two mole fragments should be approximately 8 kilobases (kb), a size that is within the range of the ExpandLong PCR system. PCR using this kit and several Coast mole primer combinations succeeded in amplifying a band in the 12 kilobase range (Figure 11, Figure 14) in genomic D N A (54° C annealing, 8 minutes extension, 40 cycles). I tried to directly sequence this band, but the amount of template necessary to sequence a product of this size proved impossible to obtain through PCR and sequencing will require cloning of the product.  Once the known  sequence separating the primers (1.5 kb) was accounted for, the length between the end of the ' two fragments was reduced from over 12 kilobases to approximately 10.5 kilobases. Thus, barring the presence of an intron, the length of mole exon is at least 13 kilobases. Keeping in mind that neither the start of exon one, nor the 5' repeat region identified by Hendrich et al. (1993), have been reached and that these regions may extend another kilobase or two, the first exon of mole Xist may extend over 15 kilobases. Since this exon, which forms the majority of the Xist R N A , in mouse, human and voles ranges between 8 and 11 kilobases, the theoretical size of the mole equivalent is very unusual. Although there is a possibility that the 12 kb fragment is a PCR artifact, it is doubtful since the reaction has been replicated a number of times, with different primer pairs in the region and under relatively stringent conditions (54° C annealing). Another explanation for this increased size may be due the presence of an intron in this region of the mole gene, which is not present in other species. To address this problem Superscript reverse transcription (RT) reactions were performed on Coast mole R N A of good quality ( C M 16, C M 17) in the hope that very large c D N A molecules may be produced, although the upper limit of the enzyme is only 12.3 kb. If an intron were present, a change in size of the PCR fragment would be noticeable depending on whether D N A or c D N A was used as a  114  CT  aT Q.  CL CD  o  o  o o  o  o  I  X CO  x  x  X CO  X CO  X CO  CD  —k —  —i —  —i —»•  —\ 73  m  <  CO H I  O  m <  H  73 m  m  0 1  H  H  H  m  o  < oo  6  6  6  O  X CO —\ 73  X CO  X CO  I  I id co  CO  X CO — 73 m  m  <  H  -»  c 3  CD 3  cn i -si I  —i 73 m  <  <  Figure 14: The 12kb Coast mole fragment produced by ExpandLong PCR. The PCR was performed with various primers from the two fragments of Coast mole Xist using an annealing temperature of 54C and 40 cycles of amplification.The sequences and locations of the Coast mole primers are presented in Table 2 and Figures 12 and 13. Human DNA (GM0773) in combination with the E5-C7-6 primer pair (Table 3) was used as a positive control.  115  template.  However, ExpandLong P C R on Superscript RT cDNA template using the  CMXIST5REVERSE and CMXIST2REVERSE primers, which gave the strongest 12 kb fragment (see Figure 14), did not result in a product.  The possible interpretations of this result  will be addressed in the Discussion of this chapter. Another approach that was considered in the attempt to determine the approximate size of the Coast mole Xist gene, was northern analysis with a newly generated mole XIST probe. R N A of high quality was available from the female cell lines, but the male cell line grew very poorly and could not provide the amount of R N A necessary for this method. Since the male R N A was essential as a negative control, I did not proceed with the northern blot.  116 Discussion  PCR with primers derived from conserved regions of the human, bovine, equine and lepine Xist genes was successful in detecting the Coast mole homologue. This PCR method was quite efficient in amplifying mole sequences, especially since Xist is an untranslated R N A and does not exhibit the high levels of conservation seen in protein-coding genes (Hendrich et al., 1993). However, while conservation between cow, horse, human and rabbit was a good indicator of identity in moles, there were primer pairs that, despite being perfectly conserved between these species, failed in mole.  When the sequence around one of these pairs (SKXIST1,2) was  obtained by inverse PCR, it was shown that the mole gene showed several mismatches in the region explaining the PCR failure in this case. Using conserved primers, along with inverse PCR over 2.5 kilobases of the gene were identified and sequenced. The alignment of all known genes and gene fragments are presented in Figure 15. The Coast mole Xist fragments show significant similarity (around 70-80%) to their counterparts in all the examined species, with the exception of the rodents, represented by the mouse and three vole species, in which the sequence identity falls to 20-50%. The initial sequences obtained are biased towards conserved regions of the gene, since they were derived from conserved primers. However, the inverse PCR fragments should not be affected by this bias, since they do not rely on conserved primers. The strikingly low identity with the rodent sequences agrees with the work of Hendrich et al. (1993), where mouse showed the least sequence conservation compared to the human, bovine, equine and lepine genes. The similarity scores seen for the Coast mole fragment fall within the range seen for the other sequenced species in my alignments and those reported in the literature (Brockdorff et al., 1992, Brown et al., 1992, Hendrich et al.,1993).  Thus, the new mole  117  on 4>  O  I«D X§*;C  CD  -a  co >  on §  *"  CD  o -  c  O  <u o  •S '5b 9 " CD  -I B -  • 3O IJ  ^ "J  ~ CD w on  —« C «w e „ a  .5 ^ J>  o o  IS  © H  rv  E CD  «  >  *  CJ  o B  C M  on  «  £  a. CN  « o c on 4» TD W B B aj  CD on CD  S>  CD  O.  fi  on . B B  S I  u  X I B  xi  1 «» 5  p-S « M  I--  2  B  •  |  a, CD  B «  LU  UJ  il  i g  LU ILU  —1  UJ  si  So:  o LU CO  or:  o X  m  CO <  or  C O < O  o  Ml  o o  CD *-  2 °o E £ O  CJ  CD  «£> "on  co  CD  on  «  on a> S3  si h  *  i f .2 •s «* "3 01  LU  OJD  — I  h-  <  <  —I  co 3O O 0  LU — I LU  -I  aAO  LU  a> o =3 u  fa s-  CD >  CD  p  cc c3 on CD E B 1 0  7 3  CD  2 —  Xi  ">  T3  CS  CO  L.  118 sequence is similar enough to be an Xist homologue, yet shows significant differences and unique base changes to prove that its presence was not due to contamination, by human D N A for example. Expression of both of the fragments of Coast mole Xist was only detected in female tissue and cell lines and not in the male. This mode of expression is typical of Xist, as the gene is only expressed from the inactive X chromosome, and confirms that the Coast mole fragment is an expressed, and probably functional homologue of this gene. Xist is expressed in the human and murine testis, but the situation in moles could not be established since R N A from this tissue was not available. While the complete Xist gene in Scapanus has not yet been sequenced, it seems that it will be of similar, i f not greater, size than the mouse, human and vole homologues. The 12 kilobase fragment obtained through long PCR indicates that, in the absence of an intron, Coast mole exon one may extend over 15 kilobases, which is 4 kilobases greater than its 11 kilobase human counterpart (Brown et al.,1992). Exon one in mouse and voles is even smaller at approximately 8 kilobases (Brockdorff et al., 1992, Nesterova et al., 2001 in press). This unexpected size of the mole exon could be due to an intron in this species.  Unfortunately, the long PCR on  Superscript cDNAs, which would have provided an answer to this question, did not work. The failure of this PCR can be due to one of two reasons: i f the exon is indeed 12 kilobases, it is not surprising that a cDNA of that size was not produced, since the empirical upper limit of the Superscript reverse transcriptase is 12.3 kilobases. Even i f a small intron were present, reducing the cDNA size to 8 to 10 kilobases, it is still doubtful that the reverse transcriptase would have been able to copy a fragment of that length since we have only been able to produce  119  a five kilobase transcript in our laboratory (J. Chow, personal communication).  The  combination of these uncertainties renders the absence of a cDNA band in this experiment inconclusive. The Coast mole Xist fragments do not contain any significant protein-coding potential, although a number of small open reading frames were scattered throughout the sequence, as was the case in the murine and human genes (Brockdorff et al., 1992, Brown et al., 1992, Hendrich et al., 1993). The three Coast mole ORFs with excellent coding potential in the 5' region of exon one all overlapped with each other and with human ORF4 (Brown et al., 1992, Hendrich et al, 1993) in different reading frames, which would not be expected i f there was a conserved polypeptide in the region. Based on the negative results of the exhaustive searches in human and mouse Xist, which failed to identify a functional ORF, I did not perform in depth analyses of the ORFs in the mole Xist. The pattern of Xist sequence conservation is unlike that seen for most protein coding genes in which large domains are well conserved. Instead, small blocks of identity are surrounded by sequence that has been poorly conserved (Brockdorff et al., 1992, Brown et al., 1992, Nesterova et al., 2001 in press). The functional significance of these conserved areas, such as the 5' repeats has not been established, since the Xist knockouts to date have not been targeted to small regions, but have involved rather large portions of the gene (Penny et al., 1996, Marahrens et al., 1997). Since neither the putative region of the 5' conserved repeat region, nor any of the other major repeat areas (Brockdorff et al., 1992, Brown et a l , 1992), have been sequenced in Scapanus, it is not known if these motifs are present. However, even in the small fragments that have been identified, the unique pattern of Xist conservation can be seen. In both fragments, we  120 see small regions of conservation, some of which seem to be imperfect repeats, interspersed with divergent sequence. Xist is almost certainly a functional R N A (Brockdorff et al., 1992, Brown et al., 1992) as it contains no significant open reading frame and does not leave the nucleus - a requirement for translation. Although the function and secondary structure of the transfer RNAs (tRNAs) and ribosomal RNAs have been extensively studied, the presence and significance of other noncoding RNAs has become apparent only recently and not much is known about how they perform their tasks. RNAs are now known to be involved in the regulation of gene expression by mediating mammalian X-inactivation, dosage compensation in Drosophila and the developmental regulation of certain genes in C. elegans. Functional RNAs are also involved in processing and modification of tRNA, rRNA and mRNA and include the RNAs involved in splicing, R N A editing and tRNA maturation. Additionally, there are the "housekeeping" RNAs, such as the telomerase R N A and the signal recognition particle R N A (SRP) involved in protein sorting in the cell (reviewed in Caprara and Nilsen, 2000).  Although R N A folding is still  poorly understood, it is now clear that functional RNAs adopt a remarkably diverse set of secondary and even tertiary structure interactions that surpass the stem-loop interactions envisioned previously (reviewed in Caprara and Nilsen, 2000). While the mode of action of some functional RNAs is becoming clearer, the mechanism through which the Xist R N A mediates X-chromosome inactivation is still a complete mystery, although the LINE interaction hypothesis is a good candidate (see Chapter 1). The elucidation of Xist function is hampered by the fact that, at least at the primary sequence level, the R N A doesn't show homology to almost any sequences in the database.  One  interesting exception are the Xlsirt RNAs (Xenopus laevis short interspersed repeat transcripts)  121  in the Xenopus species, which are translocated with the germ plasm in the oocyte and may contribute to the establishment of the germ cell lineage (Kloc et al., 1993). The region of homology between Xist and the Xlsirt RNAs is very small (26 nucleotides) and consists of a repeated sequence in the Xenopus gene that is similar to the core of the 5' conserved repeats in human, mouse, rabbit, horse and voles. The occurrence of these sequence motifs in two classes of possibly functional RNAs is probably not a coincidence, but at this time the function of the repeats remains unclear. The view that XIST/Xist functions at the R N A level is strengthened by recent analyses of the non-coding H19 R N A and the telomerase RNA.  The function of the H19 R N A is still an  enigma, although it is known to be involved in the negative regulation of the imprinted Ig/2 gene through the competition for a shared enhancer (reviewed in Juan et al., 2000). Additionally, induced expression of H19 leads to the suppression of cellular proliferation and tumourigenicity in certain malignant cell lines, prompting the idea of a new role as a tumour suppressor (reviewed in Juan et al., 2000). Since no substantial open reading frames have been identified in HI 9, it was initially thought that the R N A was a non-functional by-product of enhancer competition with Ig/2. However, a study by Hurst and Smith (1999) examining the sequence conservation of H19 between rats and mice showed that the R N A is subject to stabilizing selection and probably functional. A recent analysis of the secondary structure of H19 R N A in a number of mammals (cat, lynx, gopher, elephant, orangutan, cat, mouse, rat, rabbit and human), observed evolutionary conservation of the secondary structure, despite low sequence identity (Juan et al., 2000). A virtually identical conclusion was reached by Chen et al. (2000) in their examination of telomerase R N A , which serves as the R N A template for the telomerase ribonucleoprotein enzyme that maintains telomere length in organisms with linear  122 chromosomes. The authors compared the telomerase R N A from the five classes of vertebrates (mammals, reptiles, birds, amphibians and fish) and found that all the R N A molecules folded into similar secondary structures even though primary sequence conservation was negligible (for example human and shark only showed 44 % nucleotide identity).  The structure of the  vertebrate telomerase R N A also showed a remarkable similarity to the architecture of the ciliate telomerase R N A , although some domains are different.  The secondary structure of the  XIST/Xist molecule has not been established, primarily due to the large size of the molecule and insufficient computing power. However, it is interesting to note that the pattern of primary sequence conservation seen for HI 9 and the telomerase RNAs, in which blocks of high identity are surrounded by areas of low conservation, is very similar to that seen for Xist. (this study, Brockdorff et al., 1992, Brown et al., 1992, Hendrich et al., 1993, Nesterova et al., 2001 in press). Thus the mode of Xist sequence conservation may simply reflect the fact that it is the secondary structure, rather than the primary nucleotide sequence, that is the subject of natural selection. If the presence of strategically placed repeats or blocks of conservation is sufficient to ensure the proper secondary structure, the intervening sequence need not be conserved. However, since all known Xist genes seem to be quite large, it may be that the approximate size of the R N A is also under selective pressure.  It is possible that a long R N A molecule is  necessary to be able to effectively bind the hypothetical anchoring repeat sequences and then coat an entire chromosome. Although the Xist gene does not seem particularly conserved amongst the eutherians, the extremely low homology exhibited by the rodent sequences with respect to the other species in the regions studied is striking. While the human, mole, rabbit, horse and cow fragments show 70-80 % identity, the mouse and vole scores are as low as 20 % for the 5' region of exon one  123 and are slightly better at 40-50 % for the end of exon one. Even amongst themselves these muroid rodents do not show a lot of conservation in the tested regions; although the vole genes are 98 % identical, the mouse has only 70% similarity to the voles. While rodents have some blocks of conservation shared by other species (see Figures 9, 12, 13) and share the conserved 5' repeats with humans, cows, horses and rabbits (Brockdorff et al., 1992, Brown et al., 1992, Hendrich et al., 1993, Nesterova et al., 2001 in press), the surrounding sequence shows a high degree of divergence.  The functional significance, if any, of this lack of conservation is  unknown since Xist function may only depend on a small proportion of the entire sequence, such as the conserved 5' repeats or other conserved blocks. For example, a Y A C containing human XIST can induce inactivation in mouse autosomes, although without the accompanying late replication and histone hypoacetylation, indicating that, despite the general lack of conservation, crucial motifs may have been preserved (Heard et al., 1999, Migeon et al., 1999). A possible source of the sequence divergence seen in the mouse and voles may be the higher rate of nucleotide substitutions reported in rodents in comparison to primates (WenHsiung et al., 1996). According to this hypothesis, which still remains controversial, organisms with a short generation time should go through more cycles of D N A replication per unit time and should have a higher mutation rate than organisms with longer generation times. Insectivores, especially the shrews, may be superficially similar to the muroid rodents, but they do not share their prolific breeding habits and thus have longer generation times (Gorman, 1990, Nagorsen, 1996). Although small body size has been correlated with an increased nucleotide substitution rate, due to a higher metabolic rate and generation time, a phylogenetic study in shrews based on complete mitochondrial D N A does not show an increased rate in nucleotide substitution (Fumagalli et al.,1999). Since all of the organisms examined in our study of Xist  124  have longer generation times than rodents, the extreme divergence seen in the mouse and vole is consistent with an increase in nucleotide substitution in this lineage. Of course, it must be kept in mind that I have analyzed a very small region (2.5 kb) out of a very large R N A and that the pattern of conservation may differ in other regions of the gene. Before any firm conclusions regarding the levels of conservation and can be made, it is necessary to completely sequence Xist in several organisms, including the mole. Additionally, more precise deletions in the gene will be necessary to identify functionally significant regions. The data presented in this chapter are the first to show the presence and female-specific expression of the Xist gene in an insectivore.  Although functional studies have not been  performed in Scapanus, the female-specific expression pattern in combination with the relatively high sequence conservation are good indicators that this gene is involved in X chromosome inactivation in the Coast mole.  125 Chapter 5  Discussion and future directions  The methylation analysis of Coast mole genes, along with results of Jegalian and Page (1998), demonstrate that both X-chroniosome inactivation and silencing of X-linked sequences through methylation have been conserved in the Insectivora. While the ZFX, ARA and FMR1 genes were presumed to be on the X chromosome of the mole on the basis of Ohno's law (1967), the consistent methylation differences between the male and female alleles, a situation only seen for X-inactivated genes, confirm this prediction. The results also highlight the extraordinary conservation of the CpG island regulatory regions located in the 5' regions of ZFX, ARA and FMR1 and suggest the importance of strict dosage compensation for the studied genes. The status of the ZFX gene, which escapes inactivation in the Coast mole and European hedgehog (Jegalian and Page, 1998), supports the observation that this gene became silenced only in the rodent lineage (Jegalian and Page, 1998).  It would be interesting to establish  whether ZFY is present and functional in the mole and thus strengthen the hypothesis that it is the loss of the Y homologue that led to the inactivation of ZFX in the muroid rodents. A full appreciation of the nucleotide conservation of the mole Xist gene will be gained once the entire gene is sequenced. At the moment I am attempting to clone the 12 kilobase fragment (Chapter 3) into a plasmid vector. Due to the rather large size of this product, it may be necessary to fragment it by performing restriction digests and then clone the smaller pieces. A parallel strategy is the application of inverse PCR (Fig 3), which may allow the sequencing of the 5' end of the gene and the 12 kilobase central fragment. This strategy will probably not be  126  successful in isolating the 3' terminus of the gene, since it is reasonable to expect that the coding sequence will be interrupted by several introns. In fact, the Coast mole fragment at the 3' region of exon one seems to be at very end of that exon. Therefore, the 3' region of the gene will probably have to be amplified by 3' R A C E (rapid amplification of cDNA ends) using polyT and internal gene primers. The polyT primer anneals to the polyA tail of polyadenylated RNAs, such as XIST/Xist (Brockdorff et al., 1992, Brown et al., 1992), and in combination with a primer within the known gene sequence, the intervening R N A is amplified and the lengthy introns are avoided. I have attempted this technique once, without any conclusive results (data not shown). Additionally, EST sequencing of various species may allow the design of more conserved primers in the future. In the case that all of these PCR-based methods fail, it will be necessary to construct a mole cDNA library if Coast mole Xist is to be sequenced. Preliminary analyses of the Coast mole Xist fragments, show that the gene is quite similar to the available sequences, with the exception of the rodents, although it will be necessary to analyze the entire sequence in order to establish global homology levels. A n encouraging result of the analyses presented in this work is the detection of conserved regions that are not apparent when only the human and murine genes are compared. Confirmation that these are functional domains, along with the previously identified 5' conserved repeats, would require deletions of these regions. It is unclear whether Coast mole Xist will aid in the search for a marsupial homologue at this point, precisely because of the relatively high levels of sequence similarity seen between most of the genes. However, the complete sequence of the Scapanus gene will certainly be helpful in determining the presumably conserved secondary structure of the XIST/Xist RNA. The elucidation of this structure could clarify the mechanism of XIST/Xist function, which may  127 involve the formation of a ribonucleoprotein, possibly analogous to the M S L complex in Drosophila, that establishes the closed chromatin conformation of the mammalian inactive X (Kelley and Kuroda, 2000a,b). Despite the continuing ignorance regarding Xist function, it is becoming increasingly clear in the case of non-coding RNAs that the secondary structure is the phenotype acted on by natural selection. Studies of the HI 9 R N A in several eutherian mammals has identified a secondary structure conserved in all the species examined (Juan et al., 2000). Even more striking are the results of Chen and colleagues (2000) who identified a conserved secondary structure of the telomerase R N A structure that was conserved not only within mammals and vertebrate, but down to the protozoan ciliates. Perhaps searches of the marsupial genomes, which have yet to be sequenced, for a conserved secondary, rather that primary, structure of Xist will yield a functional homologue of the eutherian gene. While I believe that it is necessary to expand our studies of X-inactivation beyond the human and mouse models to fully understand the mechanism and evolution of this process, it is doubtful that most of the studies will be done in the mole. As a result of the mostly underground existence of the majority of moles, little is known about the breeding habits and genetics of these animals. Moles have not been bred in captivity and no embryonic stem cell lines, which would be necessary for performing Xist knockout experiments, are available. Although it is possible to keep a mole in captivity, where they get used to handling and make "gentle, but demanding" pets, breeding programs are almost certainly impossible (Gorman and Stone, 1990). Moles are solitary territorial animals and two captured individuals placed in the same container have been known to fight to the death (Gorman and Stone, 1990). Only during the short mating season can the moles tolerate each other's presence for limited periods of time. Furthermore, it is unlikely the genetics of this species will be intensively studied in the near future since most of  128  the attention moles get from humans seems to be negative. The common moles are usually viewed as pests and since they have little or no commercial value, they are eliminated in various inhumane ways, including traps, poison and "mole-vacuums", which suck the animals out of their tunnels. A temporary, and rather unfortunate, increased interest in these little insectivores occurred at the beginning of the last century when the normally miniscule European mole fur trade, burgeoned and nearly drove the European moles to extinction (Gorman and Stone, 1990). Fortunately, fashion has changed and the moles have once again retreated to relative obscurity. However, despite the lack of mole breeding programs and genetic maps, it is still possible to answer some relevant questions in X-inactivation in this organism. We have already established mortal Coast mole cell lines and it should not be too difficult to create immortal cells lines through viral transformation. Once a cell lineage is established it becomes easier to conduct a whole range of experiments without the time limitations inherent in normal, mortal cells. RNA-FISH with available human and murine probes may be successful in determining whether Coast mole Xist coats the X chromosome, as it does in the other studied eutherians (Clemson et al., 1996, Heard et al., 1997, Duthie et al., 1999). If an association of the Xist R N A with the X chromosome were found, it would strengthen the results presented in this thesis, which indicate that Xist is probably functional in the mole. Additionally, DNA-FISH could be used to locate Coast mole Xist and confirm that is on the X chromosome. It would also be possible to examine the acetylation status of histones on the Coast mole X chromosome and determine i f replication-timing differences exist between the two homologues.  Another avenue of interest are the LINE (LI) repeats, which are very strong  candidates for the role of the booster elements that facilitate the spread of inactivation along the mouse and human X chromosome, both of which are exceptionally rich in these sequences  129 (Bailey et al., 1998, Lyon, 1998). Although L I elements have been identified in a large number of eutherians and in a marsupial, the core insectivores (Euliptophyla) have not been examined for the presence of these repeats (Dorner and Paabo, 1995).  One could perform FISH  experiments with a conserved eutherian L I probe to determine whether the Coast mole X chromosome is enriched for these elements as is the case in humans and mice (Lyon, 1998). The extent and identity of genes that escape X-chromosome inactivation is another matter that could, at least theoretically, be examined in the mole. However, for this study to be feasible one must either identify X-linked Coast mole polymorphisms for each of the genes or establish a somatic cell hybrid line in which the mole X is retained in the background of another species. It may be difficult to locate polymorphisms in the mole sequences since underground mammals tend to exhibit low levels of genetic variation, presumably due to the stable and less complex underground environment (Yates and Morre, 1990). Indeed, heterozygosity values for the highly fossorial moles, such as the Coast mole, are quite low, with an average of 0.0144 (Yates and Moore, 1990). Based on a sample of 13 animals and a mean number of loci of 21.2, the heterozygosity of a Scapanus orarius population was estimated at 0.003 (Yates and Moore, 1990). On the other hand, the creation of somatic cell hybrids is hampered by the fact that the cells of some organisms simply fragment in a foreign background, are completely expelled or expel the chromosomes of the other species, making it impossible to maintain a stable culture. Another very interesting question which, unfortunately may be impossible to study in the mole system, is the presence of imprinted X-inactivation. This question has only been addressed in the mouse, rat and human and results showed paternal inactivation only in the extraembryonic tissues of the two rodents (Heard et al., 1997). Since one primate and two closely related rodents hardly constitute a significant data set, it is crucial that other organisms  130  be examined to determine whether random inactivation in the extraembryonic tissues is a primate-specific phenomenon, possibly due to the selective advantage of psuedo-heterozygosity, or i f it is the rodents that are the exception to the rule. The answer to this question would greatly illuminate the evolutionary origins of X-inactivation not only in eutherians, but also in the marsupials in which it is also the paternal chromosome that is always silenced (Cooper et al., 1993). One hypothesis for imprinted inactivation in the mouse trophoblast holds that the short gestation time and rapid development in rodents simply do not allow enough time for the erasure of epigenetic marks acquired during the transcriptional silencing of the X in spermatogenesis (Heard et al., 1997). The situation in moles is particularly interesting, not only because of their possibly basal phylogenetic position, but because, like mice, moles are small mammals with short gestation times. The average laboratory mouse lives up to three years and has a gestation period of 21 days (Green, 1966). Although moles seem to have longer life spans (one trapped European mole was thought to be seven years old!), their gestation times of approximately four weeks are comparable to those of the mouse (Gorman and Stone, 1990). If paternal inactivation were to be observed in mole extraembryonic tissues, it would support the argument that imprinted silencing is simply a consequence of rapid development. However, i f this were not the case in moles, other evolutionary reasons for the situation in rodents would have to be considered. However, to determine the parental origin of the two X chromosomes it is necessary to identify X-linked polymorphisms and to genotype both parents. Unfortunately, since the animals do not breed in captivity, it is doubtful that this sort of study will be feasible in the mole in the near future. However, the question of imprinted inactivation can be addressed in other, more studied eutherian model systems such as the cow, sheep or the pig. There is considerable interest, mostly motivated by economics, in the genetics of these organisms and  131 genome projects are underway for the cow and pig. Furthermore, advanced cloning technology in these animals would allow the researchers to address other questions regarding X-inactivation in early mammalian development. This study of X-chromosome inactivation in the Coast mole shows that the unique features of eutherian X-linked silencing have been conserved in the Order Insectivora. Methylation analysis of X-linked CpG islands not only shows that the Coast mole undergoes X chromosome inactivation, but that it employs CpG methylation to achieve silencing. Furthermore, a homologue of the, at the moment, uniquely eutherian XIST/Xist gene has been identified in Scapanus. The Coast mole gene shows significant similarity to the available Xist sequences and is only expressed in the female animals.  Despite the absence of functional  studies, the combined evidence of the methylation analysis, sequence conservation and expression pattern points to a role for Coast mole Xist in the X-chromosome inactivation in this animal. Although the complete sequence of mole Xist will be very useful, it is unlikely that the mole system will ever be as amenable for the study of early development as other model organisms. However, while the mole does not seem suited to genetic research, the remaining Liptophylans, the hedgehogs and shrews, may be better candidates. Hedgehogs are increasingly popular as pets and can breed in captivity (Smith, 1999), while the population structure and genetic polymorphims of the shrews are well studied (Boutellier and Perrin, 2000). In addition, shrews do not seem to have problems reproducing in captivity and numerous breeding colonies have been established (Ohno et al., 1998). The hedgehog, and especially the shrew, may represent realistic candidates for the establishment of an insectivore model of X-chromosome inactivation, which I believe, would be very helpful in the elucidation of this fascinating process.  132 Bibliography Adler, D.A., Rugarli, E.I., Lingenfelter, P.A., Tsuchiya, K . , Poslinkski, D., Liggit, H.D., Chapman, V . M . , Elliot, R.W., Ballabio, A . and Disteche, C M . Evidence of evolutionary upregulation of the single active X chromosome in mammals based on CIC4 expression levels in Mus spretus and Mus musculus. Proceedings of the National Academy of Sciences U.S.A. 94:9244-9248, 1997. Air, G . M , Thompson, E.O.P., Richardson, B.J. and Sharman, G.B. Amino-acid sequences of kangaroo myoglobin and haemoglobin and the date of marsupial-eutherian divergence. Nature 229:391-394, 1971. Altschul, S.F., Gish, W., Miller W., Myer, E.W. and Lipman, D.J. Basic local alignment search tool. Journal of Molecular Biology, 215:403-410, 1990. Anderson, C L . and Brown, C.J. Polymorphic X-chromosome inactivation of the human TIMP1 gene. American Journal of Human Genetics 65:699-708, 1999. Ariel, M . , Robinson, E., McCarrey, J.R. and Cedar, H . Gamete-specific methylation correlates with imprinting of the murine Xist gene. Nature Genetics 9:312-315, 1995. Armstrong S.J., Hulten, M . A . , Keohane, A . M . , and Turner, B . M . Different strategies of X inactivation in germinal and somatic cells: Histone H4 underacetylation does not mark the inactive X in the mouse male germline. Experimental Cell Research 230, 399-402, 1997. Avner, P. and Heard, E. X-chromosome inactivation: Counting, choice and initiation. Nature Reviews Genetics, 2:59-67, 2001. Ayoub, N . , Richler, C , and Wahrman, J. Xist R N A is associated with the transcriptionally inactive X Y body in mammalian male meiosis. Chromosoma 106, 1-10, 1997. Bailey, J.A., Carrel, L . , Chakravati, A . and Eichler, E. Molecular evidence for a relationship between LINE-1 elements and X chromosome inactivation: The Lyon repeat hypothesis. Proceedings of the National Academy of Sciences U.S.A. 97(12):6634-6639, 2000. Banzai M . , Omoe K . , Ishikawa H . , Endo A . Viability, development and incidence of chromosome anomalies of preimplantation embryos from X O mice. Cytogenetics and Cell Genetics 70:273-277, 1995. Baumstark, A., Akhverdyan, Schulze, A., Reisert, I., Vogel, W. and Just, W. Exclusion of SOX9 as the testis determining factor in Ellobius lutescens: evidence for another testis gene besides SRY and SOX9. Molecular Genetics and Metabolism, 72:61-66, 2001. Baverstock, P.R., Adams, M . , Polkinhorne, R.W., and Gelder, M . A sex-linked enzyme in birds - Z chromosome conservation but no dosage compensation. Nature 296:763, 1982.  133 Beard, C , L i , E., and Jaenisch, R. Loss of methylation activates Xist in somatic but not in embryonic cells. Genes and Development 9:2325-2334, 1995. Bouteiller, C. and Perrin, N . Individual reproductive success and effective population size in the greater white-toothed shrew Crocidura russula. Proceedings of the Royal Society of London B. Biological Sciences, 267(1444):701-705, 2000. Brockdorff, N . , Ashworth, A., Kay, G.F., McCabe, V . M . , Norris, D.P., Cooper, P.J., Swift, S. and Rastan, S. The product of the mouse Xist gene is a 15 kb inactive X-specific transcript containing no conserved ORF and located in the nucleus. Cell, 71:515-526, 1992. Brown, C.J., Ballabio, A . , Rupert, J.L., Lafreniere, R.G., Grompe, M . , Tonlorenzi, R. and Willard, H.F. A gene from the region the human X inactivation centre is expressed exclusively from the inactive X chromosome. Nature, 349:38-44, 1991.Brown, C.J., Hendrich, B.D., Rupert, J.L., Lafreniere, R.G., Xing, Y . , Lawrence, J. and Willard, H.F. The Human XIST Gene: analysis of a 17 kb inactive X-specific R N A that contains conserved repeats and is highly localized within the nucleus. Cell 71:527-542, 1992. Brown, C.J and Willard, H.F. Molecular and genetic studies of human X chromosome inactivation. Advances in Developmental Biology, 2:37-72, 1993. Brown, C.J. and Willard, H.F. The human X-inactivation centre is not required for maintenance of X-chromosome inactivation. Nature, 368:154-156, 1994. Brown, C.J., Carrel, L . and Willard, H.F. Expression of genes from the human active and inactive X chromosomes. American Journal of Human Genetics, 60:1333-1343, 1997. Campbell, N . A . Biology, Third edition. The Benjamin/Cummings Publishing Company, 1993. Cao, Y., Fujiwara, M . , Nikaido, M . , Okada, N . and Hasegawa, M . Interordinal relationships and timescale of eutherian evolution as inferred from mitochondrial genome data. Gene, 259:149158, 2000. Caprara, M . G . and Nilsen, T.W. R N A : versatility in form and function. Nature Structural Biology, 7(10):831-833, 2000. Carrel, L . and Willard, H.F. A n assay for X inactivation based on differential methylation at the fragile X locus, FMR1. American Journal of Medical Genetics, 64:27-30, 1996a. Carrel, L , Hunt, P.A. and Willard, H.F. Tissue and lineage-specific variation in inactive X chromosome expression of the murine Smcx gene. Human Molecular Genetics 5(9): 1361-1366, 1996b.  134 Carrel. L., Cottle, A . A . , Goglin, K . C . and Willard, H.F. A first-generation X-inactivation profile of the human X chromosome. Proceedings of the National Academy of Sciences U.S.A. 96(25): 14440-14444, 1999. Chen, J-L., Blasco, M . and Greider, C.W. Secondary structure of vertebrate telomerase R N A . Cell 100:503-514, 2000. Chomczynski, P. and Sacchi, N . Single-step method of R N A isolation by acid guanadiniumthiocyanate-phenol-chloroform extraction. Analytical Biochemistry,\62:\56-\59, 1987. Ciccodicola, A , D'Esposito, M , Esposito, T, Gianfranco, F, Migliaccio, C, Miano, M . G . , Matarazzo, M.R., Vacca, M . , Franze, A . , Cuccurese, M . , Cocchia, M . , Curci, A . , Terracciano, A., Torino A . , Cocchia S., Mercadente, G., Pannone, E., Archidiacono, N . , Rocchi, M . , Schlessinger, D. and D'Urso, M . Differentially regulated and evolved genes in the fully sequenced Xq/Yq pseudoautosomal pseudoautosomal region. Human Molecular Genetics 9:395-401, 2000. Clemson,C.M., McNeill, J.A., Willard, H.F. and Lawrence, J.B. XIST R N A paints the inactive X chromosome at interphase: evidence for a novel R N A involved in nuclear chromosome structure. The Journal of Cell Biology, 132:259-275, 1996. Clemson Moulton, C , Chow, J.C., Brown, C.J. and Lawrence Bentlet, J. Stabilization and localization of Xist R N A are controlled by separate mechanisms and are not sufficient for X inactivation. The Journal of Cell Biology 142(l):13-23, 1998. Clerc, P; and Avner, P. Role of the region 3' to Xist exon 6 in the counting process of X chromosome inactivation. Nature Genetics 19:249-253, 1998. Cooper, D.W, Johnston, P.G., Watson, J.M. and Graves, J . A . M . X-inactivation in marsupials and monotremes. Seminars in Developmental Biology, 4:117-128, 1993. Cooper, A . and Fortey, R. Evolutionary explosions and the phylogenetic fuse. Trends in Ecology and Evolution, 113:151-156, 1998. Costanzi, C. and Pehrson, J.R. Histone macroH2Al is concentrated in the inactive X chromosome of female mammals. Nature, 393:599-601, 1998. Courtier, B., Heard, E. and Avner. P. Xce haplotypes show modified methylation in a region of the active X chromosome lying 3' to Xist. Proceedings of the National Academy of Sciences U.S.A. 92:3531-3535, 1995. Csankovszki, G., Panning, B., Bates, B., Pehrson, J.R. and Jaenisch, R. Conditional deletion of Xist disrupts histone macroH2A localization but not maintenance of X inactivation. Nature Genetics, 22:3230324, 1999.  135 Cutler-Allen, R., Zoghbi, H.Y., Moseley, A . B . , Rosenblatt, H . M and Belmont, J.W. Methylation of Hpall and Hhal sites near the polymorphic C A G repeat in human androgen-receptor gene correlates with X chromosome inactivation. American Journal of Human Genetics, 51:12291239, 1992. Debrand, E., Heard, E. and Avner, P. Cloning and localization of the murine Xpct gene: evidence for complex rearrangements during the evolution of the region around the Xist gene. Genomics, 48:296-303, 1998. De L a Fuente, R., Hahnel, A., Basrur, P.K. and King, W.A. X inactive-specific transcript (Xist) expression and X chromosome inactivation in the preattachment bovine embryo. Biology of Reproduction, 60:769-775, 1999. Disteche, C M . The great escape. American Journal of Human Genetics 60:1312-1315, 1997. Dorner, M . and Paabo, S. Nucleotide sequence of a marsupial LINE-1 element and the evolution of placental mammals. Molecular Biology and Evolution, 12(5):944-948, 1995. Duthie, S.M., Nesterova, T.B., Formstone, E.J., Keohane, A . M . , Turner, B . M . , Zakian, S.M. and Brockdorff, N . Xist R N A exhibits a banded localization on the inactive X chromosome and is excluded from autosomal material in cis. Human Molecular Genetics, 8:195-204, 1999. Eggan, K., Akutsu, H., Hochedlinger, K., Rideout III, W., Tanagimachi, R. and Jaenisch, R. X chromosome inactivation in cloned mouse embryos. Science, 290(5496): 1578-1581, 2000. Eichler, E.E., Kunst, C.B., Lugenbeel, K . A . , Ryder, O.A., Davison, D., Warren, S.T. and Nelson, D.L. Evolution of the cryptic FMR1 C G G repeat. Nature Genetics, 11:301-308, 1995. Foster, J.W., Breman, F.E., Hampikian, G.K., Goodfellow, P.N., Sinclair, A . H . , Lovell-Badges, R., Selwood, L., Renfree, M . B . , Cooper, D.W. and Graves, J.A.M. Evolution of sex determination and the Y chromosome: 57?7-related sequences in marsupials. Nature, 359:531533, 1992. Fu, Y - H , Kuhl, D.P., Pizzuti, A . , Pieretti, M . , Sutcliffe, J.S., Richards, Verkerk, A.J.M.H., Holde, J.J.A., Fenwick, R . G , Warren, S.T., Oostra, B . A , Nelson, D . A . and Cakey, T.C. Variation of the C G G repeat at the fragile X site results in genetic instability: resolution of the Sherman paradox. Ce//,67:1047-1058, 1991. Fumagalli, L . , Taberlet, P., Stewart, D.T., Gielly, L . , Hausser, J and Vogel, P. Molecular phylogeny and evolution of Sorex shrews (Soricidae: Insectivora) inferred from mitochondrial D N A sequence data. Molecular Phylogenetics and Evolution, 1l(2):222-235, 1999. Gartler, S.M. and Riggs, A . D . Mammalian X-chromosome inactivation. Annual Review of Genetics 17, 155-190, 1983.  136  Gartler, S.M. and Goldman, M . Reactivation of inactive X-linked genes. Developmental Genetics, 15:504-514, 1994. Gilbert, S.L and Sharp, P.A. Promoter-specific hypoacetylation of X-inactivated genes. Proceedings of the National Academy of Sciences U.S.A., 96(24): 13825-13830, 1999. Gilbert, S.L, Pehrson, J.R. and Sharp, P.A. XIST R N A associates with specific regions of the inactive X chromatin. The Journal of Biological Chemistry, 275(47):36491-36494, 2000. Gorman, M . L . and Stone, R.D. The natural history of moles. Comstock Publishing, Cornell University Press, New York, 1990. Goto, T. and Monk, M . Regulation of X-chromosome inactivation in development in mice and humans. Microbiology and Molecular Biology Reviews, 62(2):362-378, 1998. Goto, Y . and Takagi, N . Maternally inherited X chromosome is not inactivated in mouse blastocysts due to parental imprinting. Chromosome Research 7:101-109, 1999. Graves, J.A.M., Disteche, C M . and Toder R. Gene dosage in the evolution and function of mammalian sex chromosomes. Cytogenetics and Cell Genetics 80:94-103, 1998a. Graves, J.A.M., Wakiefield, M . J . and Toder, R. The origin and evolution of the pseudoautosomal regions of human sex chromosomes. Human Molecular Genetics, 7(13): 11911196,1998b. Green, E.L.ed. The staff of the Jackson laboratory. Biology of the laboratory mouse. Dover Publications, New York, 1966. Hansen, S.R, Canfield, T.K, Stanke, A . M . , Keitges, E.A. and Gartler, S.M. Reactivation of XIST in normal fibroblasts and a somatic cell hybrid: Abnormal localization of XIST R N A in hybrid cells. Proceedings of the National Academy of Sciences U.S.A. 95:5133-5138, 1998. Heard, E., Clerc, P. and Avner, P. X-chromosome inactivation in mammals. Annual Review of Genetics, 31:571-610, 1997. Heard, E., Mongetard, F., Arnaud, D., Chureau, C , Vourc'h, C. and Avner, P. Human XIST yeast artificial chromosome transgenes show partial X inactivation center function in mouse embryonic stem cells. Proceedings of the National Academy of Sciences U.S.A. 96:6841-6846, 1999. Hendrich, B.D., Brown, C J . and Willard, H.F. Evolutionary conservation of possible functional domains of the human and murine XIST genes. Human Molecular Genetics 2(6):663-672, 1993.  137 Hendriksen. P.M., Hoogerbrugge, J.W., Thernmen, A.P., Koken, M . H . M . , Hoeijmakers, J.H.J., Oostra, B.A., Van Der Lende, T. and Grootegoed, A . Postmeiotic transcription of X and Y chromosomal genes during spermatogenesis in the mouse. Developmental Biology 170:730-733, 1995. Henke, W., Herdel, K . , Jung, K . , Schnorr, D. and Loening, S. Betaine improves the PCR amplification of GC-rich D N A sequences, Nucleic Acids Research 25(19): 3957-3958, 1997. Herzing. L . B . , Romer, J.T., Horn, J . M . and Ashworth, A . Xist has properties of the X chromosome inactivation centre. Nature 386:272-275, 1997. Hong, Y - K . , Ontiveros, S.D. and Strauss, W . M . A revision of the human XIST gene organization and structural comparison with mouse Xist. Mammalian Genome 11:220-224, 2000. Hoyer-Fender, S., Costanzi, C. and Pehrson, J.R. Histone MacroH2A1.2 is concentrated in the XY-body by the early pachytene stage of spermatogenesis. Experimental Cell Research, 258:254-260, 2000. Hurst, L.D. Evolutionary theories of genomic imprinting. Genomic Imprinting. Reik W, Surani A. (Eds.). Oxford, Oxford University Press, 1997. Hurst, L.D. and Smith, N.G.C. Molecular evidence that H19 mRNA is functional. Trends in Genetics, 4:134-135, 1999. Jamieson, R.V., Zhou S.X., Tan S.S. X-chromosome inactivation during the development of the male urogenital ridge in the mouse. International Journal of Developmental Biology 41:49-55, 1997. Jegalian, K . and Page, D.C. A proposed path by which genes common to mammalian X and Y chromosomes evolve to become X inactivated. Nature, 394:776-780, 1998. Jimenez, R., Burgos, M . , Sanchez, A , Sinclair, A . H . , Alarcon , F.J., Marin, J.J., Ortega, E. and Diaz de la Guardia, R. Fertile females of the mole Talpa occidentalis are phenotypiclntersexes with ovotestes. Development, 118:1303-1311, 1993. Johnston, C M . , Nesterova, T.B., Formstone, E.J., Newall, A.E.T., Duthie, S.M., Sheardown, S.A. and Brockdorff, N . Develpomentally regulated Xist promoter switch mediates initiation of X inactivation. Cell 94:809-817, 1998. Jones, P.L., Veenstra, G.J.C, Veenstra, G.J.C, Wade, P.A., Vermaak, D., Kass, S.U., Landsberger, N . , Strouboulis, J. and Wolffe, A.P. Methylated D N A and MeCP2 recruit histone deacetylase to repress transcription. Nature Genetics, 19:187191, 1998. Juan, V , Crain, C. and Wilson, C. Evidence for evolutionarily conserved secondary structure in the H19 tumor suppressor R N A . Nucleic Acids Research 28(5):1221-1227, 2000.  138  Just, W., Rau, W., Vogel, W., Akhverdian, M . , Fredga, K., Graves, J.A.M. and Lyapunova, E. Absence of Sry in species of the vole Ellobins. Nature Genetics, 11:117-118, 1995. Kaslow, D.C. and Migeon, B.R. D N A methylation stabilizes X chromosome inactivation in eutherians but not in marsupials: Evidence for multistep maintenance of mammalian X dosage compensation. Proceedings of the National Academy ofSciences U.S.A. 84:6210-6214, 1987. Kay, G.F., Penny, G.D., Patel, D., Ashworth, A., Brockdorff, N . , and Rastan, S. Expression of Xist during mouse development suggests a role in the initiation of X chromosome inactivation. Cellll, 171-182, 1993. Kelley, R.L. and Kuroda, M . I. The role of chromosomal RNAs in marking the X for dosage compensation. Current Opinion in Genetics and Development, 10:555-561, 2000a. Kelley, R.L. and Kuroda, M . I. Noncoding R N A genes in dosage compensation and imprinting. Cell, 103:9-12, 2000b. Keohane, A . M . , Lavender, J.S., O'Neill, L.P. and Turner, B . M . Histone acetylation and X inactivation. Developmental Genetics, 22:65-73, 1998. Kettlewell, J.R., Raymond, C.S. and Zarkower, D. Temperature-dependent expression of turtle Dmrtl prior to sexual differentiation. Genesis, 26:174-178, 2000. Kloc, M . , Spohr, G. and Etkin, L.D. Translocation of repetitive R N A sequences with the germ plasm in Xenopus oocytes. Science, 262:1712-1714, 1993. Krettek, A . , Gullberg, A . and Arnason, U . Sequence analysis of the complete mitochondrial D N A molecule of the Hedgehog, Erinaceus europaeus, and the phylogenetic position of the Liptophyla. Journal of Molecular Evolution, 41:952-957, 1995. Lahn, B. and Page, D.C. Functional coherence of the human Y chromosome. Science, 278:675680, 1997. Lahn, B.T. and Page, D.C. Four evolutionary strata on the human X chromosome. Science, 286:964-967, 1999. Lahn, B.T., Pearson, N . M . and Jegalian, K. The human Y chromosome, in the light of evolution. Nature Reviews Genetics, 2:207-216, 2001. Lee, J.T. and Jaenisch, R. Long-range cis effects of ectopic X-inactivation centres on a mouse autosome. Nature 386:275-X, 1997. Lee, J.T. and Lu, N . Targeted mutagenesis of Tsix leads to nonrandom X inactivation. Cell 99:47-57, 1999a.  139  Lee, J.T., Davidow, L.S. and Warshawsky, D. Tsix, a gene antisense to Xist at the X-inactivation centre. Nature Genetics 21:400-404, 1999b. Lee, J.T. Disruption of imprinted X inactivation by parent-of-origin effects at Tsix. Cell 103:1727, 2000. Lei, H., Oh, S.P., Okano, M . , Jutterman, R., Goss, K . A . , Jaenisch, R. and L i , E. De novo D N A cytosine methyltransferase activities in mouse embryonic stem cells. Development, 122:31953205, 1996. Lewis, J.D., Meehan, R.R., Henzel, W.J., Maurer-Fogy, I., Jeppesen, P., Klein, F. and Bird, A . Purification, sequence and cellular localization of a novel chromosomal protein that binds to methylated D N A . Cell 69:905-914, 1992. L i , E., Bestor, T.H. and Jaenisch, R. Targeted mutation of the D N A methyltransferase gene results in embryonic lethality. Cell, 69:915-926, 1992. L i , W-H., Ellsworth, D.L., Krushkal, J., Chang, B.H-J. and Hewett-Emmett, D. Rates of nucleotide substitution in primates and rodents and the generation-time effect hypothesis. Molecular Phylogenetics and Evolution, 5(1): 182-187, 1996. Liu, F.R., Miyamoto, M . M . , Freire, N.P., Ong, P.Q., Tennat, M.R., Young, T.S. and Gugel, K.F. Molecular and morphological supertrees for eutherian (placental) mammals. Science, 291:17861789, 2001. Loebel, D.A.F. and Johnston, P.G. Methylation analysis of a marsupial X-linked CpG island by bisulfite genomic sequencing. Genome Research, 6:114-123, 1996. Looijenga, L . H . , Gillis, A . J . , Verkerk, A . J . , Van Putten, W.L. and Oosterhuis, J.W. Heterogeneous X-inactivation in human full-term female placentas. American Journal of Human Genetics, 64(5): 1445-1452, 1999. Lucchesi, J.C. Dosage compensation in flies and worms: the ups and downs of X-chromosome regulation. Current Opinion in Genetics and Development, 8:179-184, 1998. Luoh, S-W., Jegalian, K., Lee, A., Chen, E.Y., Ridley, A . and Page, D.C. CpG islands in human ZFX and ZFY and mouse Zfx genes: sequence similarities and methylation differences. Genomics, 29:353-363, 1995. Lyko, F. D N A methylation learns to fly. Trends in Genetics, 17(4): 169-172, 2001. Lyon, M . F . Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature, 190:372-373, 1961. Lyon, M.F. X-Chromosome inactivation: a repeat hypothesis. Cytogenetics and Cell Genetics, 80:133-137, 1998.  140 Madsen, O., Scally, M . , Douady, C.J., Kao, D.J., DeBry, R.W., Adkins, R., Arnrine, H . M . , Stanhope, M.J., de Jong, W.W. and Springer, M . Parallel adaptive radiations in two major clades of placental mammals. Nature, 409:610-614, 2001. Marahrens, Y . , Panning, B., Dausman, J., Strauss, W. and Jaenisch, R. X/st-deficient mice are defective in dosage compensation but not spermatogenesis. Genes and Development 11:156166, 1997. McDonald, L.E., Paterson, C A . and Kay, G.F. Bisulfite genomic sequencing-derived methylation profile of the Xist gene throughout early mouse development. Genomics 54:379386, 1998. Meller, V . Dosage compensation: making I X equal 2X. Trends in Cell Biology, 10(2):54-59, 2000. Migeon, B.R. Non-random X chromosome inactivation in mammalian cells. Cytogenetics and Cell Genetics 80:142-148, 1998. Migeon, B.R., Kazi, E., Haisley-Royster, C , Hu, J., Reeves, R., Call, L., Lawler, A . , Moore, C.S., Morison, H . and Jeppesen, P. Human X inactivation center induces random X chromosome inactivation in male transgenic mice. Genomics 59:113-121, 1999. Migeon, B.R., Winter, H . , Kazi, E., Chowdhury, A . K . , Hughes, A . , Haisley-Royster, C , Morison, H . and Jeppesen, P. Low-copy-number human transgene is recognized as an X inactivation center in mouse ES cells, but fails to induce m-inactivation in chimeric mice. Genomics 71:156-162, 2001. Mlynarczyk and Panning, B. X Inactivation: Tsix and Xist as yin and yang. Current Biology, 10:R899-R903,2000. Mouchaty, S.K., Gullberg, A . , Janke, A . and Aranson, U . The phylogenetic position of the Talpidae within Eutheria based on analysis of complete mitochondrial sequences. Molecular Biology and Evolution, 17(l):60-67, 2000. Murphy, W.J., Eizirik, E., Johnson, w.E., Zhang, J.P., Ryder, O. and O'Brien, S.J. Molecular phylogenetics and the origin of placental mammals. Nature, 409:614-618, 2001. Nagorsen, D. Opossums, shrews and moles of British Columbia. Vancouver, U B C Press, 1996. Nanda, I., Zend-Ajusch, E., Shan, Z., Grutzner, F., Schartl, M . , Burt, D.W., Koehler, M . , Fowler, V . M . , Goodwin, G., Schneider, W.J., Mizuno, S., Dechant, G., Haaf, T. and Schmid, M . Conserved synteny between the chicken Z sex chromosome and human chromosome 9 includes the male regulatory gene DMRTT. a comparative (re)view on avian sex determination. Cytogenetics and Cell Genetics, 89:67-78, 2000.  141  Nesterova, T.B., Duthie, S.M., Mazoruk, N.A., Isaenko, A . , Rubtsova, N . , Zakian, S.M. and Brockdorff, N . Comparative mapping of X chromosomes in vole species of the genus Microtus. Chromosome Research, 6:41-48, 1998. Nesterova, T.B., Slobodyanyuk, S.Y., Elisaphenko, E.A., Shevchenko, A . I., Johnston, C , Pavlova, M.E., Rogozin, I.B., Kolesnikov, N . N . , Brockdorff, N . and Zakian, S.M. Characterization of the genomic Xist locus in rodents reveals conservation of overall gene structure and tandem repeats but rapid evolution of unique sequence. Genome Research, 2001, in press. Ng, H . H . and Bird, A . D N A methylation and chromatin modification. Current Opinion in Genetics andDevelpoment 9:158-163, 1999. Norris, D.P., Brockdorff, N . and Rastan. S. Methylation status of CpG-rich islands on active and inactive X-chromosomes. Mammalian Genome, l(2):78-83, 1991. Novacek, M.J. Mammalian phylogeny: shaking the tree. Nature, 356:121-125, 1992. Ohlsson, R., Paldi, A . and Graves, J . A . M . Did genomic imprinting X chromosome inactivation arise from stochastic expression? Trends in Genetics, 17(3): 136-141, 2001. Ohno, S. Sex Chromosome and Sex-linked Genes. Monographs on endocrinology. Berlin, Heidelberg, New York. Springer-Verlag, 1967 Ohno, T., Yoshida F., Ichikawa Y., Matsuo S., Hotta N . , Terada M . , Tanaka S., Yamashita K , Namikawa T. and Kitoh, J. A new spontaneous animal model of N I D D M without obesity in the musk shrew. Life Sciences, 62(11):995-1006, 1998. Okano, M . , Bell, D.W., Daniel, A . H . and L i , E. D N A methylatransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell, 99:247-257, 1999. O'Neill, L.P., Keohane, A . M , Lavender, J.S., McCabe, V., Heard, E., Avner, P., Brockdorff, N . and Turner, B . M . A developmental switch in H4 acetylation upstream of Xist plays a role in X chromosome inactivation. The EMBO Journal, 18(10):2897-2907, 1999. Panning, B . and Jaenisch, R. D N A hypomethylation can activate Xist expression and silence X linked genes. Genes and Development 10:1991-2002, 1996. Panning, B., Dausman, J. and Jaenisch, R. X chromosome inactivation is mediated by Xist R N A stabilization. Cell 90:907-916, 1997. Panning, B. and Jaenisch, R. R N A and the epigenetic regulation of X chromosome inactivation. Cell, 93:305-308, 1998. Pask, A . and Graves, J . M . Sex chromosomes and sex-determining genes: insights from marsupials and monotremes. Cellular and Molecular Life Sciences 55:864-875, 1999.  142  Pask, A., Renfree, M . B . and Marshall Graves, J.A. The human sex-reversing ATRX gene has a homologue on the marsupial Y chromosome, ATRY: implications for the evolution of mammalian sex determination. Proceedings of the National Academy of Sciences U.S.A. 97(24): 13198-13202, 2000. Pehrson, J.R. and Fuji, R . N . Evolutionary conservation of histone macroH2A subtypes and domains. Nucleic Acids Research, 26(12);2837-2842, 1998. Penny, G.D., Kay, G.F, Sheardown, S.A., Rastan, S. and Brockdorff, N . Requirement for Xist in X chromosome inactivation. Nature 379:131-137, 1996. Prissette, M . , El-Maari, O., Arnaud, D., Walter, J. and Avner, P. Methylation profile of DXPAS34 during the onset of X-inactivation. Human Molecular Genetics, 10(1):31-38, 2001. Robertson, K . D . and Jones. P.A. D N A methylation: past, present and future directions. Carcinogenesis, 21(3):461-4667, 2000. Robertson, K . D . and Wolffe, A.P. D N A methylation in health and disease. Nature Reviews Genetics, 1:11-19, 2000. Sado, T., Fenner, M . H . , Tan, S-S., Tarn, P., Shioda, T. and L i , E. X Inactivation in the mouse embryo deficient for Dnmt P. distinct effect of hypomethylation on imprinted and random X inactivation. Developmental Biology, 225:294-303, 2000. Sanchez, A., Bullejos, M . , Burgos, M . , Hera, C., Stamatopoulos, C., Diaz de la Guardia, R., and Jimenez, R. Females of four mole species of genus Talpa (Insectivora, Mammalia) are true hermaphrodites with ovotestes. Molecular Reproduction and Development, 44:289-294, 1996a. Sanchez, A., Bullejos, M . , Burgos, M . , Hera, C., Jimenez, R. and Diaz de la Guardia, R. High sequence identity between the S R Y H M G box from humans and insectivores. Mammalian Genome, 7:536-538, 1996b. Sheardown, S., Norris, D., Fisher, A . and Brockdorff, N . The mouse Smcx gene exhibits developmental and tissue specific variation in degree of escape from X inactivation. Human Molecular Genetics 5(9): 1355-1360, 1996. Sinclair, A . H , Berta, P., Palmer, M.S., Hawkins, J.R., Grffiths, B.L., Smith, M.J., Foster, M.J., Frischauf, A . M . , Lovell-Badge, R. And Goodfellow, P.N. A gene from the human sexdetermining region encodes a protein with homology to a conserved DNA-binding motif. Nature 346:240-244, 1990. Smith, A. J. Husbandry and nutrition of hedgehogs. Veterinary Clinics of North America. Exotic Animal Practice, 1:127-141, 1999. Solari, A . Sex Chromosomes and Sex Determination in Vertebrates. Boca Raton, Ann Arbor, London, Tokyo. C R C Press, 1993.  143 Solter, D. and Wei, G. Ends Xist, but where are the beginings? Genes and Development 11:153155, 1997. Springer, M.S., De Bry, R.W., Douady, C., Amrine, H . M . , Madsen, 0., de Jong, W. and Stanhope, M . J . Mitochondrial versus nuclear gene sequences in deep-level mammalian phylogeny reconstruction. Molecular Biology and Evolution, 18(2): 132-143, 2001. Stanhope, M.J., Waddel, V . G . , Madsen, O., De Jong, W., Hedges, S.B., Cleven, G.C., Kao, D. and Springer, M.S. Molecular evidence for multiple origins of Insectivora and for a new order of endemic African insectivore mammals. Proceedings of the National Academy of Sciences U.S.A., 95:9967-9972, 1998. Strachan, T. and Read, A.P. Human Molecular Genetics 2. BIOS Scientific Publishers, 1999. Takagi, N . And Sasaki, M . Preferential inactivation of the paternally derived X-chromosome in the extraembryonic membranes of the mouse. Nature 256:640-642, 1975. Thompson, J.D., Higgins, D.G. and Gibson, T.J. C L U S T A L W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position specific gap penalties and weight matrix choice. Nucleic Acids Research, 22(22):4673-4680, 1994. Tinker, A . V . and Brown, C.J. Induction of XIST expression from the human active X chromosome in mouse/human somatic cell hybrids by D N A demethylation. Nucleic Acids Research 26(12):2935-2940, 1998. Uberbacher, E.C. and Mural, R.J. Locating protein-coding regions in human D N A sequences by a multiple sensor-neural network approach. Proceedings of the National Academy of Sciences U.S.A. 88:11261-11265, 1991 Vogel, W., Jainta, S., Rau, W., Geerkens, C , Baumstark, A., Correa-Cerro, L.-S., Ebenhoch, C. and Just, W. Sex determination in Ellobius lutescens: the story of an enigma. Cytogenetics and Cell Genetics 80:214-221, 1998. Wachtel, S.S. X-linked sex-reversing genes. Cytogenetics and Cell Genetics, 80:222-225, 1998. Wake, N . , Tagaki, N . and Sasaki, M . Non-random inactivation of X chromosome in the rat yolk sac. Nature, 262:580-581, 1976. Wakefield, M.J., Keohane, A . M . , Turner, B . M . and Graves, J . A . M . Histone underacetylation is an ancient component of mammalian X chromosome inactivation. Proceedings of the National Academy of Sciences U.S.A., 94:9665-9668, 1997. Warshawsky, D., Stavropoulos, N . and Lee, J.T. Further examination of the Xist promoterswitch hypothesis in X inactivation: evidence against the existence and function of a Po promoter. Proceedings of the National Academy of Sciences U.S.A. 96(25): 14424-14429, 1999.  144  Watson, J.M., Spencer, J.A., Riggs, A . D . and Marshall Graves, J.A. The X chromosome of monotremes shares a highly conserved region with the eutherian and marsupial X chromosomes despite the absence of X chromosome inactivation. Proceedings of the National Academy of Sciences U.S.A. 87:7125-7129, 1990. Watson, J . M . , Spencer, J.A., Riggs, A . D . and Graves, J.A.M. Sex chromosome evolution: Platypus gene mapping suggests that part of the human X chromosome was originally autosomal. Proceedings of the National Academy of Sciences U.S.A., 88:11256-11260, 1991. Waugh O'Neill, R.J., O'Neill, M . and Graves, J . A . M . Undermethylation associated with retroelement activation and chromosome remodelling in an interspecific mammalian hybrid. Nature, 393:68-72, 1998. Wen-Hsiung. L., Ellsworth, D.L., Krushkal, J., Chang, B. H-J. and Hewett-Emmett, D. Rates of nucleotide substitution in primates and rodents and the generation-time effect hypothesis. Molecular Phylogenetics and Evolution 5(1): 182-187, 1996. Whitworth, D.J., Licht, P, Racey, P.A. and Glickman, S.E. Testis-like steroidogenesis in the ovotestis of the European mole, Talpa europea. Biology of Reproduction, 60:413-418, 1999. Wutz, A . and Jaenisch, R. A shift from reversible to irreversible X inactivation is triggered during ES cell differentiation. Molecular Cell 5:695-705, 2000. Wrigley, J.M. and Graves, J . A . M . Sex chromosome homology and incomplete, tissue-specific X-inactivation suggest that monotremes represent an intermediate stage of mammalian sex chromosome evolution. Journal of Heredity 79:115-118, 1988. Yates, T.L. and Moore, D.W. Speciation and evolution in the family Talpidae (Mammalia: Insectivora), Progress in Clinical and Biological Research, 335:1-22, 1990. Zarkower, D. Establishing sexual dimorphism: conservation amidst diversity? Nature Reviews Genetics, 2:175-185, 2001. Zuccotti, M . and Monk, M . Methylation of the mouse Xist gene in sperm and eggs correlates with imprinted Xist expression and paternal X-inactivation. Nature Genetics 9:316-320, 1995.  145  Appendix 1: XIST/Xist sequences used in the creation of Figures 8 and 15  Primate  Human  M97168  Rhesus monkey  L10733, AF140294 (EST), BE483406 (EST)  Chimpanzee  LI 0729  Howler monkey  L10731  African green monkey  LI 0723  Rodent  Mouse  L04961  Vole ( M rossiameridionalis) AJ310130 Vole (M. arvalis)  AJ310129  Vole (M. transcaspicus)  AJ310127  Rat (ESTs)  BF419051, All45414.1, BF415216  Others Horse  U50911  Rabbit  U50910  Cat  AF197966  Cow  AF104906, AV617038 (EST)  Pig  BF702572  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items