Title: The X Factor: X chromosome dosage compensation in the evolutionarily divergent monotremes and marsupials

Transcription

1 Accepted Manuscript Title: The X Factor: X chromosome dosage compensation in the evolutionarily divergent monotremes and marsupials Author: Deanne J. Whitworth Andrew J. Pask PII: S (16) DOI: Reference: YSCDB 1917 To appear in: Seminars in Cell & Developmental Biology Received date: Revised date: Accepted date: Please cite this article as: Whitworth Deanne J, Pask Andrew J.The X Factor: X chromosome dosage compensation in the evolutionarily divergent monotremes and marsupials.seminars in Cell and Developmental Biology This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

2 The X Factor: X chromosome dosage compensation in the evolutionarily divergent monotremes and marsupials Deanne J. Whitworth a* d.whitworth@uq.edu.au, Andrew J. Pask b a School of Veterinary Science, University of Queensland, Gatton, Queensland, 4343, Australia b School of BioSciences, The University of Melbourne, Parkville, Victoria, 3010, Australia * Corresponding author. Tel

3 Abstract Marsupials and monotremes represent evolutionarily divergent lineages from the majority of extant mammals which are eutherian, or placental, mammals. Monotremes possess multiple X and Y chromosomes that appear to have arisen independently of eutherian and marsupial sex chromosomes. Dosage compensation of X linked genes occurs in monotremes on a gene by gene basis, rather than through chromosome wide silencing, as is the case in eutherians and marsupials. Specifically, studies in the platypus have shown that for any given X linked gene, a specific proportion of nuclei within a cell population will silence one locus, with the percentage of cells undergoing inactivation at that locus being highly genespecific. Hence, it is perhaps not surprising that the expression level of X linked genes in female platypus is almost double that in males. This is in contrast to the situation in marsupials where one of the two X chromosomes is inactivated in females by the long noncoding RNA RSX, a functional analogue of the eutherian XIST. However, marsupial X chromosome inactivation differs from that seen in eutherians in that it is exclusively the paternal X chromosome that is silenced. In addition, marsupials appear to have globally upregulated X linked gene expression in both sexes, thus balancing their expression levels with those of the autosomes, a process initially proposed by Ohno in 1967 as being a fundamental component of the X chromosome dosage compensation mechanism but which may not have evolved in eutherians. Keywords: X chromosome inactivation; monotreme; marsupial 2

4 1. Introduction In mammals, sex is determined by heteromorphic sex chromosomes with male heterogamety (XY male, XX female). The sex chromosomes evolved from ancestral autosomes, and in the common therian ancestor of both the placental mammals (eutherians) and marsupials (metatherians), it is estimated that the X and Y chromosomes emerged approximately 181 million years ago, just prior to the divergence of the two lineages [1] (Figure 1). In contrast, monotremes (prototherians) possess multiple X and Y chromosomes that appear to have arisen independently of, and yet concomitantly with, the therian sex chromosomes around 175 million years ago [1] and which display partial homology to the sex chromosomes of birds [2, 3] (Figure 1). Given that monotremes and ancestral therians evolved their sex chromosomes independently of each other, between 175 and 181 million years ago, the mechanism by which their common mammalian ancestor determined sex, around 200 million years ago, is a fascinating enigma. A defining feature of the evolution of the sex determining Y chromosome is its progressive degradation, such that in humans it contains less than 3% of its original gene content [4], in contrast to the X chromosome which has recruited and retained large numbers of genes [5]. A consequence of Y chromosome decay is an imbalance between the expression of X linked genes in the male (equivalent to a monosomy) and those in the female. Hence, in 1967, Ohno [6] proposed a mechanism whereby a two fold increase in the expression of X linked genes in both sexes, followed by an inactivation of one of the X chromosomes in females, would both balance the expression of X linked genes between males and females in addition to restoring the level of expression of genes on the X chromosome relative to those on the autosomes. While in eutherians the phenomenon of X chromosome inactivation (XCI) has been extensively characterised, studies into Ohno s premise of X chromosome upregulation have yielded data that are conflicting and the results inconclusive [7 14]. Given their evolutionary divergence from eutherians, monotremes and marsupials are well placed to provide insight into the evolution of X chromosome dosage compensation. 3

5 2. Monotremes Monotremes are egg laying mammals whose extant representatives include two genera of echidna (the Australian species Tachyglossus aculeatus and several species of Zaglossus that are endemic to Papua New Guinea) and the Australian platypus (Ornithorhynchus anatinus). Monotremes diverged from the therian lineage approximately 200 million years ago [1] (Figure 1) and sequencing of the platypus genome [15] provided genetic evidence in support of the observational suspicion that they are indeed an extraordinary amalgam of ancestral reptilian and derived mammalian traits, perhaps best exemplified by the fact that although young hatch from eggs, they are nourished post hatching with milk. The female platypus and echidna have five pairs of distinct X chromosomes (X 1 X 5 ), while male platypus have five X chromosomes (X 1 X 5 ) and five Y chromosomes (Y 1 Y 5 ); the male echidna similarly has five X chromosomes (X 1 X 5 ) but only four Y chromosomes (Y 1 Y 4 ) due to a fusion of Y 3 and Y 5 [2, 16, 17]. During meiosis in the male, X and Y chromosomes form an alternating chain (X 1 Y 1 X 2 Y 2 X 5 Y 5 ) with each chromosome linked to its neighbours via a pseudoautosomal region, eventually segregating to form spermatozoa bearing five X chromosomes or five Y chromosomes [2] (Figure 1). It has been estimated that the X chromosomes constitute around 15% of the monotreme haploid genome, which would seem to present a strong incentive for X chromosome dosage compensation [18]. However, DNA methylation and histone modification analyses of female platypus fibroblasts found no evidence of a chromosome wide dosage compensation mechanism [19]. This is in keeping with an earlier study by Deakin et al. [20] that showed that for a given X linked gene some, but not all, nuclei will undergo silencing of one locus. Thus, in monotremes, X linked genes appear to be dosage compensated on a gene by gene, rather than chromosome wide, basis [19]. Using RNA fluorescence in situ hybridisation (RNA FISH) on female platypus fibroblasts, Livernois and colleagues [21] expanded upon the earlier study of Deakin et al. [20] and demonstrated the inactivation of 28 genes across the various X chromosomes. Significantly, when they examined multiple fibroblast nuclei for each of these genes they discovered that for any given gene the proportion of nuclei that had undergone silencing ranged from 25% 4

6 to 62%, with no gene displaying silencing, or lack thereof, in 100% of nuclei [21]. But perhaps the most significant observation is that each gene is expressed from one or both loci at a characteristic, and reproducible, frequency [21]. The authors also noted that if a particular gene was expressed at higher levels in female fibroblasts as compared to male fibroblasts, a greater proportion of nuclei would show silencing for that gene in female cells than in male cells [21]. Furthermore, for a given female fibroblast, the activity status of an X linked gene is not clonally inherited; that is, if the mother cell had inactivated one locus of a particular X linked gene, the daughter cells may or may not undergo silencing for that same gene [21]. Subsequent analyses by Julien and colleagues [8] lend further support to a variable frequency, gene by gene, process of X chromosome dosage compensation in monotremes. Using RNA seq data from a range of tissue types in addition to cultured fibroblasts, they showed that for genes on platypus chromosome X 5, median expression levels in females were almost double those in males (X 5 :X 5 X 5 = 0.59) [8].This is in contrast to the situation in marsupials, where chromosome wide XCI occurs and where the level of X linked gene expression is very similar between the sexes [8]. Although the platypus doesn t initiate XCI per se, it does still silence genes on its X chromosomes which raises the question of whether for each chromosome pair all inactive loci are localised to the one X chromosome, essentially creating a partially inactivated X chromosome and an active X chromosome. Livernois et al. [21] addressed this possibility by examining the silencing of neighbouring genes on chromosome X 5. They found that the inactivated loci did, in fact, co localise to the same X 5 suggesting that for each X chromosome pair, there may be one X chromosome which is fully transcriptionally active (X 5 a), and one which is partially inactive (X 5 i) [21]. Further studies, looking at a greater number of genes across all five X chromosome pairs, should help to define if various silenced regions are consistently restricted to the one X chromosome in a pair or if different regions are silenced on each of the two X chromosomes, which also begs the question of whether silencing is imprinted, with either genes on the paternal or maternal X chromosome preferentially silenced. 5

7 The first step in Ohno s hypothesis of X chromosome dosage compensation posits that in order to compensate for the two fold reduction in expression of X linked genes in the male, as a consequence of the demise of their Y linked homologues, there has been an upregulation in the expression of X linked genes in both males and females in order to restore their ancestral expression levels (ie before their differentiation into sex chromosomes) and to bring them back into parity with the expression levels of the autosomes. It is this part of Ohno s hypothesis that has proved the most contentious and difficult to resolve. However, in an elegant study of eutherians, marsupials and the platypus, Julien et al. [8] compared current expression levels between the extant platypus X 5 chromosomes and the predicted ancestral, or proto, X 5 chromosomes in females and found the ratio of expression to be very close to 1.0, indicating that expression of X 5 linked genes has remained unchanged throughout evolution which, given the lack of chromosome wide XCI, does not support an argument for X chromosome upregulation in monotremes. The authors also compared extant X 5 linked expression with transcription from the autosomes (X 5 X 5 :AA) and similarly found no evidence for an upregulation of genes on X 5 [8]. We have recently generated induced pluripotent stem cells (ipscs) from female platypus fibroblasts (Whitworth, unpublished data) and used RNA seq to examine the transcriptome of both cell types. Within both the platypus ipscs and fibroblasts, the ratio between the median transcription levels for all annotated X linked genes and those identified as autosomal (X 1 5X 1 5 :AA) is approximately 1.0 (Whitworth, unpublished data), in keeping with the findings of Julien et al. [8] that there has been no global upregulation of X linked genes. Thus, in monotremes, while chromosome wide XCI does not occur, for each of the genes studied to date one locus is silenced on a gene by gene basis, and at gene specific frequencies, within a population of cells. It also appears that silenced loci within a region are inactivated on the same X chromosome. Monotremes also show no evidence of having undergone X chromosome upregulation and so in spite of the X chromosomes constituting around 15% of the haploid genome, there is no dosage compensation of X linked genes between males and females, or between X linked genes and the autosomes in males. 6

8 3. Marsupials In contrast to the monotremes, marsupials share orthologous sex chromosomes with eutherian mammals. Thus, our X and Y chromosomes originated in the therian (marsupial and eutherian) ancestor around 180 million years ago and around 20 million years after the divergence of monotremes [1]. The sex chromosomes started life as an autosomal pair. However, the separate evolution of the marsupials and eutherians has seen diversification of the X and Y chromosome content in each lineage. Similarly, while eutherians and marsupials both use an X inactivation dosage compensation mechanism, the control and regulation of chromosome silencing are quite diverse between both groups. The eutherian X chromosome has a unique gene content, showing an enrichment of genes involved in testis and brain function [22]. This is likely due to the X chromosome being a unique evolutionary environment, stemming from its hemizygosity in males. The marsupial X chromosome is significantly smaller than its orthologue in eutherians which has grown in size due to an autosomal addition [23 27] (Figure 1). Interestingly, studies in marsupials showed that the autosomal blocks acquired by the growing eutherian X chromosome contained genes that already had functions in brain and testis development [27 30]. In contrast, the marsupial X chromosome appears to have remained without large scale additions and it is thought to more closely resemble the ancestral therian X chromosome. The male sex determining SRY gene evolved, for the first time, on the therian Y chromosome and has been maintained on the Y in both the eutherian and marsupial lineages. The critical role for this gene in sex determination led to restricted recombination of the X and Y, sealing the fate of the Y chromosome on a rapidly evolving trajectory [1]. Over time, different subsets of genes have been lost from the Y chromosome leading to dosage differences between these genes in males and females. This led to a requirement for dosage compensation mechanisms to evolve, to maintain an equal dosage of X linked genes in both male and female cells. In eutherians, equality is achieved through XCI which is primarily mediated by the long non coding RNA XIST (X inactive specific transcript), which is expressed from, and coats, the inactive X chromosome in cis. Eutherian XCI is a thorough mechanism, although around 15% of alleles in humans, and 3% of alleles in mice, escape its 7

9 effects and continue to be expressed from the inactive X [31 33]. In humans, genes in the recently added regions of the eutherian X chromosome are more likely to escape XCI than those from the ancient therian conserved portion, indicating the progressive recruitment of alleles into the X inactivation system over evolutionary time [34, 35]. The choice of which X chromosome to inactivate is a random process in all eutherians (but not all tissues) and the mechanisms by which one X chromosome is selected appear to differ across species [36, 37]. Marsupials, similarly, show a silencing of one of their X chromosomes but lack the XIST gene [38]. While the syntenic neighbours of XIST can be found on the marsupial X chromosome, they are in distant locations [38]. Thus, it appears that XIST evolved independently in the eutherian lineage [38]. Furthermore, XCI is not random like that in eutherians, but instead is paternally imprinted, incomplete, and tissue specific [39, 40]. Despite these fundamental differences, the overall mechanism of XCI appears related [19]. Histone underacetylation (of histone H4) is a common feature of XCI in marsupial, as well as eutherian, mammals [41]. Thus, it is likely such a mechanism of dosage compensation existed in the therian ancestor [42 44]. In eutherian mammals, histone underacetylation on the X chromosome is accompanied by high levels of DNA methylation. However, the marsupial X chromosome is not hypermethylated [45]. Much conjecture existed over how modified epigenetic silencing of the X chromosome was initiated in marsupials in the absence of XIST [18] and this was finally resolved in 2012 with the discovery of the RSX gene [31]. RSX (RNA on the Silent X) was discovered while performing RNA FISH using an opossum BAC which contained HPRT1 [31]. A novel long non coding RNA was discovered on the BAC that coated the inactive X chromosome in a distribution similar to that seen for XIST in eutherians. RSX produces a 27 Kb mature transcript that is repeat rich, similar to XIST, but does not share any significant homology with XIST [31]. Thus, it would appear that both marsupials and eutherians independently evolved long non coding RNA mediated X chromosome inactivation [31]. Despite the sequence differences, RSX is active only in female tissues and is silenced in female germ cells (where both X chromosomes are active), identical to XIST in eutherian mammals [31]. Definitive evidence for RSX as the key mediator of marsupial X chromosome silencing came from experiments artificially expressing RSX 8

10 from a mouse autosome. RSX coated the transgenic chromosome and silenced the genes contained on it [31]. XCI in marsupials has been reported to be incomplete and that a large proportion of genes escape RSX mediated X inactivation. In the South American grey short tailed opossum (Monodelphis domestica), a comprehensive study of XCI reported that around 14% of genes failed to be silenced [46]. This figure is similar to that reported for XIST mediated silencing in some eutherians, including humans [34]. However, studies in the Australian tammar wallaby (Macropus eugenii) found that 32% of genes escape XCI [47]. Furthermore, unlike the pattern seen in humans where genes in the more recently added regions of the X chromosome are more likely to escape XCI, the marsupial escapees span all the different evolutionary regions on the X [48]. Marsupial X chromosome inactivation also differs from that seen in eutherians in that it is exclusively the paternal X that is silenced. Paternal X chromosome (Xp) silencing is not unique to marsupials; in fact, many eutherian species exhibit tissue dependent Xp silencing as well as random XCI [18, 46]. Mouse, rat and cow all display Xp inactivation in the trophectoderm lineage of the early embryo (which will later give rise to the placenta), while random X inactivation is seen for the rest of the embryo. In contrast, random XCI is seen throughout the trophectoderm of human, rabbit, horse and mule [36]. While the choice of which X chromosome to inactivate varies between tissues and species, selective inactivation of the paternal X chromosome is a common feature of marsupials and eutherians and appears to be the ancestral mechanism from which random X inactivation arose. However, the precise mechanism by which the marsupial Xp is marked for silencing is still unknown. Given that XCI in marsupials has been shown to be incomplete and that a significant proportion of genes escape inactivation, it is perhaps surprising that the median expression levels of X linked genes between male and female opossums are very similar, giving an XY:XX ratio close to 1.0 [11]. Thus, in spite of the incomplete and leaky XCI in this species, dosage compensation for X linked genes between males and females appears to be very efficient. Furthermore, comparisons of extant X chromosome expression levels with those 9

11 of the predicted therian proto X chromosomes, and the autosomes, suggest that in contrast to monotremes (and possibly eutherians), marsupials have upregulated X linked genes in both males and females thus restoring their ancestral expression levels and balancing their transcriptional output with that of the autosomes [12], thus supporting Ohno s hypothesis [6]. Although XIST and RSX do not share any sequence homology, their strikingly similar mode of action suggests long non coding RNA silencing of genomic loci was a feature of XCI in the common therian ancestor (Figure 1). Different loci in both eutherians and marsupials escape X inactivation but, unlike the situation in eutherians, in marsupials this is not correlated with their evolutionary history. The ancestral XCI mechanism was also likely to have been paternally imprinted, with eutherians developing a more sophisticated random X inactivation mechanism after their divergence from marsupials (Figure 1). In addition, marsupials appear to have globally upregulated X linked gene expression in both sexes (Figure 1). 4. Conclusions The sex chromosomes of therians evolved after the divergence of therians and monotremes, but before the eutherian marsupial split [1], and so they share orthologous sex chromosomes. Intriguingly, both have independently evolved an effective XCI mechanism that is dependent upon a long non coding RNA, and while eutherians display random XCI in the embryo, with paternal XCI restricted to the trophectoderm of some species, marsupials show strict paternal XCI throughout the embryo. Thus, it is possible that ancestral therians similarly employed long non coding RNA dependent XCI that was paternally imprinted. Given that monotremes appear to have diverged from the therian lineage before the differentiation of sex chromosomes in either monotremes or therians [1], monotremes are very much the outliers in studies of mammalian X chromosome dosage compensation and so it is perhaps not at all surprising that they have their own unique strategy of silencing specific loci, at gene specific frequencies, rather than entire chromosomes, and show no need for dosage compensation of X linked genes between males and females. 10

12 Acknowledgements We thank Dr Dmitry Ovchinnikov for his critical reading of the manuscript. 11

13 References [1] Cortez D, Marin R, Toledo Flores D, Froidevaux L, Liechti A, Waters PD, et al. Origins and functional evolution of Y chromosomes across mammals. Nature 2014;508: [2] Grutzner F, Rens W, Tsend Ayush E, El Mogharbel N, O'Brien PC, Jones RC, et al. In the platypus a meiotic chain of ten sex chromosomes shares genes with the bird Z and mammal X chromosomes. Nature 2004;432: [3] Veyrunes F, Waters PD, Miethke P, Rens W, McMillan D, Alsop AE, et al. Bird like sex chromosomes of platypus imply recent origin of mammal sex chromosomes. Genome Res 2008;18: [4] Skaletsky H, Kuroda Kawaguchi T, Minx PJ, Cordum HS, Hillier L, Brown LG, et al. The male specific region of the human Y chromosome is a mosaic of discrete sequence classes. Nature 2003;423: [5] Ross MT, Grafham DV, Coffey AJ, Scherer S, McLay K, Muzny D, et al. The DNA sequence of the human X chromosome. Nature 2005;434: [6] Ohno S. Sex chromosomes and sex linked genes. Berlin, New York,: Springer; [7] Deng X, Hiatt JB, Nguyen DK, Ercan S, Sturgill D, Hillier LW, et al. Evidence for compensatory upregulation of expressed X linked genes in mammals, Caenorhabditis elegans and Drosophila melanogaster. Nat Genet 2011;43: [8] Julien P, Brawand D, Soumillon M, Necsulea A, Liechti A, Schutz F, et al. Mechanisms and evolutionary patterns of mammalian and avian dosage compensation. PLoS Biol 2012;10:e [9] Kharchenko PV, Xi R, Park PJ. Evidence for dosage compensation between the X chromosome and autosomes in mammals. Nat Genet 2011;43:1167 9; author reply [10] Lin F, Xing K, Zhang J, He X. Expression reduction in mammalian X chromosome evolution refutes Ohno's hypothesis of dosage compensation. Proc Natl Acad Sci U S A 2012;109: [11] Pessia E, Makino T, Bailly Bechet M, McLysaght A, Marais GA. Mammalian X chromosome inactivation evolved as a dosage compensation mechanism for dosagesensitive genes on the X chromosome. Proc Natl Acad Sci U S A 2012;109: [12] Xiong Y, Chen X, Chen Z, Wang X, Shi S, Wang X, et al. RNA sequencing shows no dosage compensation of the active X chromosome. Nat Genet 2010;42: [13] Yildirim E, Sadreyev RI, Pinter SF, Lee JT. X chromosome hyperactivation in mammals via nonlinear relationships between chromatin states and transcription. Nat Struct Mol Biol 2012;19: [14] Lin H, Halsall JA, Antczak P, O'Neill LP, Falciani F, Turner BM. Relative overexpression of X linked genes in mouse embryonic stem cells is consistent with Ohno's hypothesis. Nat Genet 2011;43: ; author reply [15] Warren WC, Hillier LW, Marshall Graves JA, Birney E, Ponting CP, Grutzner F, et al. Genome analysis of the platypus reveals unique signatures of evolution. Nature 2008;453: [16] Rens W, Grutzner F, O'Brien P C, Fairclough H, Graves JA, Ferguson Smith MA. Resolution and evolution of the duck billed platypus karyotype with an X1Y1X2Y2X3Y3X4Y4X5Y5 male sex chromosome constitution. Proc Natl Acad Sci U S A 2004;101:

14 [17] Rens W, O'Brien PC, Grutzner F, Clarke O, Graphodatskaya D, Tsend Ayush E, et al. The multiple sex chromosomes of platypus and echidna are not completely identical and several share homology with the avian Z. Genome Biol 2007;8:R243. [18] Deakin JE, Chaumeil J, Hore TA, Marshall Graves JA. Unravelling the evolutionary origins of X chromosome inactivation in mammals: insights from marsupials and monotremes. Chromosome Res 2009;17: [19] Rens W, Wallduck MS, Lovell FL, Ferguson Smith MA, Ferguson Smith AC. Epigenetic modifications on X chromosomes in marsupial and monotreme mammals and implications for evolution of dosage compensation. Proc Natl Acad Sci U S A 2010;107: [20] Deakin JE, Hore TA, Koina E, Marshall Graves JA. The status of dosage compensation in the multiple X chromosomes of the platypus. PLoS Genet 2008;4:e [21] Livernois AM, Waters SA, Deakin JE, Marshall Graves JA, Waters PD. Independent evolution of transcriptional inactivation on sex chromosomes in birds and mammals. PLoS Genet 2013;9:e [22] Graves JA, Gecz J, Hameister H. Evolution of the human X a smart and sexy chromosome that controls speciation and development. Cytogenet Genome Res 2002;99: [23] Delbridge ML, Patel HR, Waters PD, McMillan DA, Marshall Graves JA. Does the human X contain a third evolutionary block? Origin of genes on human Xp11 and Xq28. Genome Res 2009;19: [24] Pask A, Renfree MB. Sex determining genes and sexual differentiation in a marsupial. J Exp Zool 2001;290: [25] Pask A, Graves JA. Sex chromosomes and sex determining genes: insights from marsupials and monotremes. Cell Mol Life Sci 1999;55: [26] Pask AJ, Harry JL, Renfree MB, Marshall Graves JA. Absence of SOX3 in the developing marsupial gonad is not consistent with a conserved role in mammalian sex determination. Genesis 2000;27: [27] Pask A, Toder R, Wilcox SA, Camerino G, Graves JA. The candidate sex reversing DAX1 gene is autosomal in marsupials: implications for the evolution of sex determination in mammals. Genomics 1997;41: [28] Hu Y, Yu H, Pask AJ, O'Brien DA, Shaw G, Renfree MB. A kinase anchoring protein 4 has a conserved role in mammalian spermatogenesis. Reproduction 2009;137: [29] Hu Y, Yu H, Shaw G, Pask AJ, Renfree MB. Kallmann syndrome 1 gene is expressed in the marsupial gonad. Biol Reprod 2011;84: [30] Hu Y, Yu H, Shaw G, Renfree MB, Pask AJ. Differential roles of TGIF family genes in mammalian reproduction. BMC Dev Biol 2011;11:58. [31] Grant J, Mahadevaiah SK, Khil P, Sangrithi MN, Royo H, Duckworth J, et al. Rsx is a metatherian RNA with Xist like properties in X chromosome inactivation. Nature 2012;487: [32] Berletch JB, Ma W, Yang F, Shendure J, Noble WS, Disteche CM, et al. Escape from X inactivation varies in mouse tissues. PLoS Genet 2015;11:e [33] Yang F, Babak T, Shendure J, Disteche CM. Global survey of escape from X inactivation by RNA sequencing in mouse. Genome Res 2010;20: [34] Johnston CM, Lovell FL, Leongamornlert DA, Stranger BE, Dermitzakis ET, Ross MT. Large scale population study of human cell lines indicates that dosage compensation is virtually complete. PLoS Genet 2008;4:e9. 13

15 [35] Carrel L, Willard HF. X inactivation profile reveals extensive variability in X linked gene expression in females. Nature 2005;434: [36] Okamoto I, Patrat C, Thepot D, Peynot N, Fauque P, Daniel N, et al. Eutherian mammals use diverse strategies to initiate X chromosome inactivation during development. Nature 2011;472: [37] Lyon MF. Gene action in the X chromosome of the mouse (Mus musculus L.). Nature 1961;190: [38] Hore TA, Koina E, Wakefield MJ, Marshall Graves JA. The region homologous to the X chromosome inactivation centre has been disrupted in marsupial and monotreme mammals. Chromosome Res 2007;15: [39] Cooper DW, Woolley PA, Maynes GM, Sherman FS, Poole WE. Studies on metatherian sex chromosomes. XII. Sex linked inheritance and probable paternal X inactivation of alphagalactosidase A in Australian marsupials. Aust J Biol Sci 1983;36: [40] VandeBerg JL, Johnston PG, Cooper DW, Robinson ES. X chromosome inactivation and evolution in marsupials and other mammals. Isozymes 1983;9: [41] Wakefield MJ, Keohane AM, Turner BM, Graves JA. Histone underacetylation is an ancient component of mammalian X chromosome inactivation. Proc Natl Acad Sci U S A 1997;94: [42] Koina E, Chaumeil J, Greaves IK, Tremethick DJ, Graves JA. Specific patterns of histone marks accompany X chromosome inactivation in a marsupial. Chromosome Res 2009;17: [43] Chaumeil J, Waters PD, Koina E, Gilbert C, Robinson TJ, Graves JA. Evolution from XISTindependent to XIST controlled X chromosome inactivation: epigenetic modifications in distantly related mammals. PLoS One 2011;6:e [44] Mahadevaiah SK, Royo H, VandeBerg JL, McCarrey JR, Mackay S, Turner JM. Key features of the X inactivation process are conserved between marsupials and eutherians. Curr Biol 2009;19: [45] Kaslow DC, Migeon BR. DNA methylation stabilizes X chromosome inactivation in eutherians but not in marsupials: evidence for multistep maintenance of mammalian X dosage compensation. Proc Natl Acad Sci U S A 1987;84: [46] Wang X, Douglas KC, Vandeberg JL, Clark AG, Samollow PB. Chromosome wide profiling of X chromosome inactivation and epigenetic states in fetal brain and placenta of the opossum, Monodelphis domestica. Genome Res 2014;24: [47] Al Nadaf S, Waters PD, Koina E, Deakin JE, Jordan KS, Graves JA. Activity map of the tammar X chromosome shows that marsupial X inactivation is incomplete and escape is stochastic. Genome Biol 2010;11:R122. [48] Rodriguez Delgado CL, Waters SA, Waters PD. Paternal X inactivation does not correlate with X chromosome evolutionary strata in marsupials. BMC Evol Biol 2014;14:

16 Figure Captions 15

17 Figure 1: Evolution of X chromosome inactivation in mammals. Monotremes have a 5X, 5Y sex chromosome system that forms a translocation chain in meiosis (homologous regions are denoted with crossed lines). The X chromosomes contain regions of homology to the bird Z chromosome (green shading). Therians evolved an XY sex chromosome system. In eutherians, the X contains a large added region derived from an autosomal gene in the therian ancestor (blue shading) and that has remained autosomal in marsupials (blue; A). The monotreme lineage (purple box) does not show chromosome wide X chromosome inactivation (XCI); rather, for each gene, one locus is silenced at a specific frequency within a population of cells. Monotremes show no indication of global X upregulation. XCI evolved in the therian ancestor (blue box) and was likely paternal, incomplete and mediated by a long non coding RNA. This mechanism of XCI has persisted in marsupials (green box) and is controlled by RSX expression. Marsupials display global upregulation of X linked genes, restoring parity with expression levels of the autosomes. In eutherians (red box), XCI has become random, involves DNA methylation and is controlled by the XIST gene. Eutherians appear not to have upregulated the expression of X linked genes, although the data are still conflicting on this point. (Myr: million years ago). 16