This page has been archived and is no longer updated
X-chromosome inactivation: a hypothesis linking ontogeny and phylogeny
Author: K. D. Huynh
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"8 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/genetics PERSPECTIVES 67. Daniels, R., Zuccotti, M., Kinis, T., Serhal, P. & Monk, M. XIST expression in human oocytes and preimplantation embryos. Am. J. Hum. Genet. 61, 33?39 (1997). 68. Fitzpatrick, G. V., Soloway, P. D. & Higgins, M. J. Regional loss of imprinting and growth deficiency in mice with a targeted deleltion of KvDMR1. Nature Genet. 32, 426?431 (2002). Acknowledgements The authors thank G. Kelsey, M. Constancia, M. Hemberger, W. Dean, G. Smits and D. Haig for advice and discussion. This work was supported by grants from the UK Biotechnology and Biological Sciences Research Council, the UK Medical Research Council and the European Union. Competing interests statement The authors declare no competing financial interests. Online links DATABASES The following terms in this article are linked online to: Entrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=gene Air | Cdkn1c | Igf2 | Kcnq1ot1 | Phlda2 | Sry | Xist FURTHER INFORMATION Epigenome Network of Excellence web site: http://www.epigenome-noe.net Geneimprint.com: http://www.geneimprint.com Harwell Mouse Imprinting web site: http://www.mgu.har.mrc.ac.uk/research/imprinting University of Otago Imprinted Gene Catalogue: http://cancer.otago.ac.nz/IGC/Web/home.html Wolf Reik?s laboratory: http://www.babraham.ac.uk/research/ developmental%20genetics/reik/ Access to this interactive links box is free online. 56. Grutzner, F. & Graves, J. A. A platypus? eye view of the mammalian genome. Curr. Opin. Genet. Dev. 14, 642?649 (2004). 57. Grutzner, F., Deakin, J., Rens, W., El-Mogharbel, N. & Marshall Graves, J. A. The monotreme genome: a patchwork of reptile, mammal and unique features? Comp. Biochem. Physiol. A 136, 867?881 (2003). 58. Rens, W. et al. Resolution and evolution of the duck-billed platypus karyotype with an X1Y1X2Y2X3Y3X4Y4X5Y5 male sex chromosome constitution. Proc. Natl Acad. Sci. USA 101, 16257?16261 (2004). 59. Grutzner, F. et al. In the platypus a meiotic chain of ten sex chromosomes shares genes with the bird Z and mammal X chromosomes. Nature 432, 913?917 (2004). 60. Ohlsson, R., Paldi, A. & Graves, J. A. Did genomic imprinting and X chromosome inactivation arise from stochastic expression? Trends Genet. 17, 136?141 (2001). 61. Plagge, A. et al. The imprinted signaling protein XL ? s is required for postnatal adaptation to feeding. Nature Genet. 36, 818?826 (2004). 62. Kaslow, D. C. & Migeon, B. R. 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. USA 84, 6210?6214 (1987). 63. Suzuki, S. et al. Genomic imprinting of IGF2, p57 KIP2 and PEG1/MEST in a marsupial, the tammar wallaby. Mech. Dev. 122, 213?222 (2005). 64. Zeng, S. M. & Yankowitz, J. X-inactivation patterns in human embryonic and extra-embryonic tissues. Placenta 24, 270?275 (2003). 65. Grati, F. R. et al. Biparental expression of ESX1L gene in placentas from normal and intrauterine growth-restricted pregnancies. Eur. J. Hum. Genet. 12, 272?278 (2004). 66. Ray, P. F., Winston, R. M. & Handyside, A. H. XIST expression from the maternal X chromosome in human male preimplantation embryos at the blastocyst stage. Hum. Mol. Genet. 6, 1323?1327 (1997). Abstract | In mammals, sex is determined by differential inheritance of a pair of dimorphic chromosomes: the gene-rich X chromosome and the gene-poor Y chromosome. To balance the unequal X-chromosome dosage between the XX female and XY male, mammals have adopted a unique form of dosage compensation in which one of the two X chromosomes is inactivated in the female. This mechanism involves a complex, highly coordinated sequence of events and is a very different strategy from those used by other organisms, such as the fruitfly and the worm. Why did mammals choose an inactivation mechanism when other, perhaps simpler, means could have been used? Recent data offer a compelling link between ontogeny and phylogeny. Here, we propose that X-chromosome inactivation and imprinting might have evolved from an ancient genome-defence mechanism that silences unpaired DNA. In most sexually dimorphic animals in which sex is genetically determined, the basis for sex lies in the differential constitution of the sex chromosomes. In mammals, males have one X and one Y chromosome (XY), whereas females carry two X chromosomes (XX) (reviewed in REF. 1), which results in an imbal- ance of sex-linked genes between the two sexes. The Y chromosome contains only a small number of functional genes, almost all of which can be classified into two groups: the Y-chromosome-specific genes with male- specific roles (such as testis determination) and genes that have homologues on the 27. Surani, M. A., Barton, S. C. & Norris, M. L. Development of reconstituted mouse eggs suggests imprinting of the genome during gametogenesis. Nature 308, 548?550 (1984). 28. Cattanach, B. M. & Kirk, M. Differential activity of maternally and paternally derived chromosome regions in mice. Nature 315, 496?498 (1985). 29. Constancia, M., Kelsey, G. & Reik, W. Resourceful imprinting. Nature 432, 53?57 (2004). 30. Wilkins, J. F. & Haig, D. What good is genomic imprinting: the function of parent-specific gene expression. Nature Rev. Genet. 4, 359?368 (2003). 31. Killian, J. K. et al. M6P/IGF2R imprinting evolution in mammals. Mol. Cell 5, 707?716 (2000). 32. Killian, J. K. et al. Monotreme IGF2 expression and ancestral origin of genomic imprinting. J. Exp. Zool. 291, 205?212 (2001). 33. Reik, W. & Walter, J. Genomic imprinting: parental influence on the genome. Nature Rev. Genet. 2, 21?32 (2001). 34. Ferguson-Smith, A. C. & Surani, M. A. Imprinting and the epigenetic asymmetry between parental genomes. Science 293, 1086?1089 (2001). 35. Sleutels, F. & Barlow, D. P. The origins of genomic imprinting in mammals. Adv. Genet. 46, 119?163 (2002). 36. Bartolomei, M. S. & Tilghman, S. M. Genomic imprinting in mammals. Annu. Rev. Genet. 31, 493?525 (1997). 37. Frank, D. et al. Placental overgrowth in mice lacking the imprinted gene Ipl. Proc. Natl Acad. Sci. USA 99, 7490?7495 (2002). 38. Takahashi, K., Kobayashi, T. & Kanayama, N. p57 Kip2 regulates the proper development of labyrinthine and spongiotrophoblasts. Mol. Hum. Reprod. 6, 1019?1025 (2000). 39. Takahashi, K. & Nakayama, K. Mice lacking a CDK inhibitor, p57 Kip2 , exhibit skeletal abnormalities and growth retardation. J. Biochem. (Tokyo) 127, 73?83 (2000). 40. Wutz, A. et al. Non-imprinted Igf2r expression decreases growth and rescues the Tme mutation in mice. Development 128, 1881?1887 (2001). 41. Engemann, S. et al. Sequence and functional comparison in the Beckwith?Wiedemann region: implications for a novel imprinting centre and extended imprinting. Hum. Mol. Genet. 9, 2691?2706 (2000). 42. Stoger, R. et al. Maternal-specific methylation of the imprinted mouse Igf2r locus identifies the expressed locus as carrying the imprinting signal. Cell 73, 61?71 (1993). 43. de Napoles, M. et al. Polycomb group proteins ring1A/B link ubiquitylation of histone H2A to heritable gene silencing and X inactivation. Dev. Cell 7, 663?676 (2004). 44. Wang, J. et al. Imprinted X inactivation maintained by a mouse Polycomb group gene. Nature Genet. 28, 371?375 (2001). 45. Braidotti, G. et al. The Air non-coding RNA ? an imprinted cis-silencing transcript. Cold Spring Harb. Symp. Quant. Biol. 69 (in the press). 46. Mager, J., Montgomery, N. D., de Villena, F. P. & Magnuson, T. Genome imprinting regulated by the mouse Polycomb group protein Eed. Nature Genet. 33, 502?507 (2003). 47. Sleutels, F., Zwart, R. & Barlow, D. P. The non-coding Air RNA is required for silencing autosomal imprinted genes. Nature 415, 810?813 (2002). 48. Sado, T. et al. X inactivation in the mouse embryo deficient for Dnmt1: distinct effect of hypomethylation on imprinted and random X inactivation. Dev. Biol. 225, 294?303 (2000). 49. Li, Y. & Behringer, R. R. Esx1 is an X-chromosome- imprinted regulator of placental development and fetal growth. Nature Genet. 20, 309?311 (1998). 50. Shi, W. et al. Choroideremia gene product affects trophoblast development and vascularization in mouse extra-embryonic tissues. Dev. Biol. 272, 53?65 (2004). 51. Chiao, E. et al. Overgrowth of a mouse model of the Simpson?Golabi?Behmel syndrome is independent of IGF signaling. Dev. Biol. 243, 185?206 (2002). 52. Tada, T., Takagi, N. & Adler, I. D. Parental imprinting on the mouse X chromosome: effects on the early development of X0, XXY and XXX embryos. Genet. Res. 62, 139?148 (1993). 53. Reik, W. et al. Chromosome loops, insulators and histone methylation: new insights into regulation of imprinting in clusters. Cold Spring Harb. Symp. Quant. Biol. 69 (in the press). 54. Wilkins, J. F. & Haig, D. Genomic imprinting of two antagonistic loci. Proc. R. Soc. Lond. B 268, 1861?1867 (2001). 55. McQueen, H. A., McBride, D., Miele, G., Bird, A. P. & Clinton, M. Dosage compensation in birds. Curr. Biol. 11, 253?257 (2001). X-chromosome inactivation: a hypothesis linking ontogeny and phylogeny Khanh D. Huynh and Jeannie T. Lee OPINION Nature Reviews Genetics | AOP, published online 8 April 2005; doi:10.1038/nrg1604 PERSPECTIVES in transcriptional levels are not cis-limited along one X chromosome. So, the fly and worm mechanisms are (at least at a superficial level) easier to imagine and execute. Recent findings provide hints as to why mammals might have favoured a cis-limited silencing mechanism for dosage compensa- tion or, more intriguingly, might even have been forced into adopting this mechanism . Here, we discuss recent insights into the ontogeny of X-chromosome inactivation and argue that they provide a glimpse into its phylogeny. The ontogeny of X inactivation It has long been known that X-chromosome inactivation in mice and humans is com- pleted during embryogenesis, so that in females dosage compensation has taken place in all cells at birth (reviewed in REFS 26?28). But when is dosage compensation necessary and when does X-chromosome inactivation take place? After Mary Lyon?s first description of X-chromosome inactivation in the mouse, some debate ensued over the transcriptional status of the X chromosomes in the early embryo and the exact timing of inactivation 29 . One view held that both X chromosomes are active in the XX zygote and silencing of one X chromosome occurs at a later stage. Another view held that one X chromosome might be silenced very early on, so that XX and XY embryos would be dosage compensated in the pre-implantation stage. The possibility was also raised that both X chromosomes might be inactive in the zygote and dosage compensation could involve reactivation of one of them. A series of elegant experiments in the 1970s and 1980s solidified the first view ? specifically, that the XX zygote begins devel- opment with two active X chromosomes and X-chromosome inactivation does not take place until implantation (around day 3.5?5.5 of embryonic development in the mouse) 30?36 . For example, a late-replicating X chromosome does not appear until the BLASTOCYST stage, implying that the two X chromosomes are transcriptionally equal in the pre-implantation embryo 36 . Further- more, expression analysis of the X-encoded hypoxanthine guanine phosphoribosyl transferase (HPRT) 32,33 and ?-galactosidase (GLA) 30 proteins indicated that XX embryos have twice the activity level of these proteins than XY embryos have. So, in the consensus view, the differentiating TROPHECTODERM of the blastocyst goes through imprinted X-chromosome inactivation for the first time at the implantation stage, and the EPIBLAST lineage undergoes random X-chromosome X chromosome and are involved in ?house- keeping? functions 2 (also reviewed in REFS 3,4). These characteristics of Y-linked genes pre- clude a need for reconciling Y-linked dosage disparity between males and females. On the other hand, the X chromosome contains approximately 1,500 genes 4,5 that have a diverse range of functions. Most of these genes have no Y-chromosome homologue and so are present at twice the dosage in females than in males. This imbalance has led to the evolution of X-linked dosage compensation. In mammals, dosage compensation occurs by X-chromosome inactivation, whereby one of the two X chromosomes is transcriptionally silenced in the female 6 . X-chromosome inacti- vation occurs in two forms, both of which are intricate epigenetic processes that are tightly controlled at many levels. In the ?random? form of X-chromsome inactivation, the two X chromosomes in each female cell have an equal chance of being inactivated 6 . This form of X-chromosome inactivation is a multi-step process and takes place in somatic lineages of EUTHERIAN mammals (reviewed in REF. 7). As a first step, a counting mechanism in the zygote determines whether one or two X chromo- somes are present in each cell. In the pres- ence of two or more X chromosomes, the X-chromosome inactivation pathway is set in motion. The cell then randomly chooses between its X chromosomes for silencing. Silencing is strictly cis-limited, involving only those genes that lie on the same X chromo- some. The choice of chromosome and the silencing mechanism require competing interactions between three non-coding loci: X (inactive)-specific transcript (Xist) 8,9 , its antisense partner Tsix 10 , and the intergenic locus Xite 11 . Whereas Xite and Tsix preserve the transcriptional competence of the future active X chromosome (Xa) 10?13 , Xist initiates silenc- ing on the future inactive X chromosome (Xi) through propagation of its RNA along the X chromosome 14,15 . How X-chromosome counting and choice are regulated and how silencing is propagated exclusively in cis remain unsolved mysteries in epigenetics. The imprinted form of X-chromosome inactivation is equally complex. In imprinted X-chromosome inactivation, the paternal X chromosome (X p ) is inactivated 16 . Imprinted X-chromosome inactivation is found in cells that comprise the extra-embryonic tissues of certain eutherians, notably the rodent 17 and the cow 18 , but remains controversial in humans 19?21 . Interestingly, this form of X-chromosome inactivation is also found in some of the earliest mammals, including the kangaroos and opossums (metatherians), a fact that has led to the popular view that imprinted X-chromosome inactivation evolved first and was transformed into ran- dom X-chromosome inactivation in euthe- rians 22 . Although the rules of counting and choice do not seem to apply, imprinted X-chromosome inactivation also requires Tsix 12,13 and Xist 23 . However, their expression patterns are pre-determined in the parental germline, rather than by the zygote during early development. Why do mammals carry out dosage com- pensation through such complicated means? Drosophila melanogaster and Caenorhabditis elegans have evolved an ostensibly more straightforward approach to dosage compen- sation (FIG. 1). As in THERIAN mammals, fruitfly sex is also determined by X and Y chromo- somes: XX individuals are female and XY indi- viduals are male (reviewed in REFS 24,25). However, unlike mammals, equalization of X-linked gene expression between the sexes is achieved by global upregulation of the single X chromosome in males 24,25 . In the worm, sex is determined by X-chromosome number (worms do not have a male-specific chromo- some): XX individuals are hermaphrodites and XO individuals are males 24 . In this system, dosage compensation occurs by a twofold reduction of activity on each X chromosome in the XX hermaphrodite 24 . As in mammals, dosage compensation in the fruitfly and the worm is initiated by a count- ing mechanism; but unlike mammals, there is no epigenetic choice, as all X chromosomes in the nucleus are treated equally and changes NATURE REVIEWS | GENETICS ADVANCE ONLINE PUBLICATION | 9 Mammals 0 1 X X X Y Flies 0 1 2 X X X Y Worms 0 0.5 1 X X X O T ranscription level Figure 1 | Strategies of dosage compensation. In Drosophila melanogaster, male (XY) X-linked expression increases twofold. In Caenorhabditis elegans, hermaphrodite (XX) transcription from each of the two X chromosomes decreases twofold. In mammals, one of the two X chromosomes in females is inactivated. 10 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/genetics PERSPECTIVES hypermethylated at lysine 9 and 27 (H3-K9; H3-K27), histone H3 is hypomethylated at lysine 4 (H3-K4), and polycomb group proteins such as embryonic ectoderm devel- opment (EED) and enhancer of zeste homo- logue 2 (EZH2) localize to that X chromosome. These changes begin at the 4- to 8-cell stages and accumulate as pre-implantation develop- ment proceeds. These results indicate that one X chromosome is silenced as early as the 4-cell stage. Using reverse-transcriptase PCR (RT-PCR) and FISH analyses to directly assess X-linked expression, our laboratory showed that one X chromosome is also silenced early, and that it is the X P that is affected 43 . Allele-specific RT-PCR showed that silencing does not occur uniformly along the chromosome 43 . Genes that are close to the X-inactivation centre (Xic) 45 , from which Xist RNA originates, are more completely silenced than genes that are farther away, indicating that silencing occurs in a graded fashion from the Xic. The degree to which genes can escape silencing and the variability between embryos shows that the pre-implantation form of X-chromosome inactivation is leaky and imperfect in many respects. Interestingly, X-chromosome inactivation becomes more complete after implantation in both the pla- cental and somatic lineages 43 . Consistent with this, Xist RNA seems to spread less extensively along the X chromosome in the pre-implantation embryo than in placental and somatic cells. Although both studies are largely in agree- ment that X-chromosome inactivation takes place much earlier than the conventionally held view, they differ significantly on the issue of the exact timing of X-chromosome inacti- vation and, therefore, how it is initiated from a mechanistic standpoint. Okamoto et al. reported that RNA polymerase II localizes to the X P at the 2-cell stage and that transcripts from the cysteine-rich hydrophobic domain 1 (Chic1) locus can be detected at this stage, implying that at least some domains of the X P are actively transcribed at this stage 44 .So,the authors conclude that X-chromosome inacti- vation initiates at the 4- to 8-cell stages in some cells and is generalized to all cells by the time of implantation (FIG 2a). Direct analysis of gene expression has led us to propose a different model in which dosage compensa- tion has already occurred at the time of con- ception 43 (FIG. 2b). COT1 FISH, a technique that allows the visualization of nascent transcrip- tion, shows that the Xist RNA domain is poorly transcribed in XX embryos. This is observed first at the 2-cell stage (when gen- eral zygotic transcription occurs for the first By combining FLUORESCENCE IN SITU HYBRI- DIZATION (FISH) and immunofluoresence on pre-implantation embryos, Okamoto et al. 44 showed that the X chromosome that expresses a high level of Xist RNA carries chromatin modifications that are typically found on heterochromatin. Histone H3 is inactivation as it differentiates into various germ lineages in the post-implantation period 34,36 . Supported by a wealth of evi- dence, this was the accepted model for almost 30 years. With the advantage of new technology, recent studies have inspired the need to revise the original model. For example, it has been known for some time that Xist RNA, the agent of silencing, is expressed in pre-implantation XX embryos from the 2-cell stage onwards 37?40 . It is also known that some X-linked genes, such as phospho- glycerate kinase 1 (Pgk1), seem to be silent on the X P until after implantation 37,41,42 .Two recent papers provide evidence to indicate that X-chromosome inactivation takes place much earlier than has been conventionally accepted 43,44 . ?With MSCI also being present in the marsupial germline, the idea that MSUD might be the original imprinting mechanism for X-chromosome inactivation becomes rather attractive.? a De novo inactivation Blastocyst X P reactivation Blastocyst X P pre-inactivation Incomplete X P silencing Complete X P silencing b Pre-inactivation Xi Xa Xi Zygote Egg Sperm Xi Xa Xi Zygote Egg Sperm Imprinted X-chromosome inactivation Blastocyst Blastocyst X P reactivation X-chromosome reactivation Xa Xa Xa Xi Xi Random X-chromosome inactivation X P inactivation X P inactivation Imprinted X-chromosome inactivation Random X-chromosome inactivation Figure 2 | The ontogeny of X-chromosome inactivation in the mouse: two current views. a | The de novo inactivation model requires many rounds of inactivation and reactivation: the paternal germline initiates meiotic sex-chromosome inactivation, but the X chromosome is completely reactivated after meiosis. The zygote inherits two fully active X chromosomes and begins re-inactivation of the paternal X chromosome (X P ) at the 4- to 8-cell stage. In the trophectoderm (extra-embryonic cells, shown in blue), X P silencing is maintained, therefore accounting for the imprinted form of X-chromosome inactivation. By contrast, in the epiblast (green cells), yet another round of reactivation takes place in preparation for a final round of inactivation in the form of random X-chromosome inactivation. b | In the pre-inactivation model, the female zygote inherits a partially silent X P and maintains the silent state throughout pre-implantation development. Silencing becomes globalized and complete in extra-embryonic tissues. This accounts for the imprinted form of X-chromosome inactivation. By contrast, the epiblast cells of the inner cell mass (ICM) undergo a single round of reactivation followed by a random form of X-chromosome inactivation. PERSPECTIVES time 46 ) and persists throughout the pre- implantation stage. Our results lead us to conclude that the mouse embryo is dosage compensated from the beginning, using a primitive form of X-chromosome inactiva- tion that is leaky and incomplete. With the further recruitment of heterochromatic fac- tors during pre-implantation development, X-chromosome inactivation is gradually improved and culminates in the ?complete? forms that are found in the placenta (imprinted X-chromosome inactivation) and the soma (random X-chromosome inactivation). Although the exact timing of X-chromo- some inactivation is subject to further investi- gation, the model above is consistent with what is currently known about the evolution of mammals, their sex chromosomes and X-chromosome inactivation. Assuming the above model is correct, what are the impli- cations for the origin and mechanism of X-chromosome inactivation? The importance of the father How is the X P silenced at conception? One possibility is that it is silenced de novo at the 2-cell stage when zygotic Xist RNA is expressed for the first time 43 . The 1- to 2-cell period represents a crucial step in the transi- tion from maternal to zygotic gene expres- sion 46 . At this time, the Xist RNA domain is small, but continues to grow as the 2-cell embryo divides to produce a 4-cell embryo 43,44 . The initially small Xist RNA domain indi- cates that silencing at the 2-cell stage is not as effective as it is later on. If correct, this could account for the reported leakiness of Chic1 expression in the 2-cell embryo and explain how immunostaining for chromatin changes might give the impression that X-chromosome inactivation is not initiated until the 4- to 8-cell stage 44 . How the zygote would establish X-chromosome inactivation at the 2-cell stage is currently unknown; includ- ing whether the mechanism is Xist-dependent. Perhaps the mechanism of silencing might be likened to other imprinted chromosomal domains (review in REF. 47), whereby the X P is ?marked? during spermatogenesis and prefer- ential silencing is subsequently executed in the zygote from the 2-cell stage onwards. Can X P silencing take place even earlier than the 2-cell stage? Indeed, a more parsimo- nious solution is that the X P might arrive in the zygote in a ?pre-inactivated? state 43 . This view fits elegantly with X-chromosome dynamics in the male germline. Unique to the male germline, the process of meiotic sex- chromosome inactivation (MSCI) silences the X and Y chromosomes during the first meiotic prophase 48 . The raison d?�tre of MSCI represents yet another mystery in the field of sex-chromosome biology, although com- pelling rationales have been proposed. These include the need to silence X-linked genes that inhibit spermatogenesis 49 (also reviewed in REF. 50), to suppress recombination between non-homologous portions of the X and the Y chromosomes 51 , or to prevent ASYNAPSED XY domains from triggering the MEIOTIC CHECKPOINT and apoptosis 52 . MSCI might also have been usurped by mammals for dosage compensation 43,53 , indicating a continuity of X-chromosome silence from paternal PACHYTENE to the early embryo. The transcriptional fate of the X and Y chromosomes after pachytene has remained highly debated, with some studies arguing for a reactivation of X- and Y-linked genes 54,55 and other genes that demonstrate persistent silence 56?60 . An analysis of the sparse literature on this subject indicates that 6 of the 9 genes that have been examined are not expressed after MSCI (FIG. 3). This supports the idea that silencing might extend beyond meiosis. Further support comes from the existence of ?retrogenes?, a growing class of spermato- genesis genes that transposed from an X-linked to an autosomal location during the course of mammalian evolution 60?66 . Retrogenes include such spermatogenesis- specific genes as phosphoglycerate kinase 2 (Pgk2) 64 , pyruvate dehydrogenase E1-?-2 (Pdha2) 63 , glucose-6-phosphate dehydroge- nase 2 (G6pd2) 60 , and zinc-finger autosomal protein (Zfa) 61 . The extensive gene traffick- ing from the X chromosome to autosomes might have been in response to MSCI, which would have otherwise silenced those genes required for spermatogenesis 62,64 . Although a more comprehensive analysis of post-MSCI gene expression is required, the available data raise an interesting point: X-linked genes that remain silent tend to be centrally located, whereas those that become reactivated are farther from the Xic (FIG. 3). This pattern is strikingly similar to what is observed in the pre-implantation embryo 43 . We therefore pro- pose that MSCI and zygotic X-chromosome inactivation form a continuum. The silencing that is initiated at MSCI in the father is propa- gated as an inactive X P in the daughter 43,67 .In this way, imprinted X-chromosome inactiva- tion in the zygote might trace its origins to MSCI in the paternal germline. In this hypothesis, fathers have a signifi- cant role in X-chromosome inactivation ? a departure from the dogma that dosage compensation is restricted to females in mammals. Furthermore, the ?pre-inactiva- tion hypothesis? greatly simplifies the NATURE REVIEWS | GENETICS ADVANCE ONLINE PUBLICATION | 11 cM 10 20 30 40 50 60 Ube1x Ube2a Pgk1 Pdha1 G6pd Zfx Phka1 Fmr1 Mage1/2 Pdha1 Pgk1 XistXist G6pd Fmr1 HprtHprt Pctk Mecp2 Chic1 Xnp Atp7a Rnf12 Gla Post-meiotic round spermatid Pre-implantation embryo Xic Figure 3 | Paternal X-chromosome silencing from sperm to zygote. The diagram shows an expression profile of the paternal X chromosome (X p ) in post-meiotic round spermatids and pre-implantation embryos 37,41?43,91,92 . Genes labelled in green are expressed; those labelled in red are either not expressed or show very low levels of expression; the expression state of those labelled in yellow is not conclusively known (interpretation is complicated by the fact that fragile X mental retardation 1 (Fmr1) is not inactivated during pachytene and the melanoma antigen gene Mage1/2 is part of a whole family of homologous genes that might have resulted in non-specific PCR detection 54,58 ). Atp7a, ATPase, Cu 2+ transporting, ?-polypeptide; Chic1, cysteine-rich hydrophobic domain 1; Gla, ?-galactosidase; G6pd, glucose-6-phosphate dehydrogenase X-linked 60 ; Hprt, hypoxanthine guanine phosphoribosyl transferase 57?59 ; Mecp2, methyl CpG binding protein 2; Pctk, PCTAIRE-motif protein kinase 1; Pdha1, pyruvate dehydrogenase E1-?-1 (REFS 54,57,58); Pgk1, phosphoglycerate kinase 1 (REFS 57?59); Phka1, phosphorylase kinase ?-1 (REF. 57); Rnf12, ring finger protein 12; Ube1x, ubiquitin-activating enzyme E1, Chr X (REF. 54); Ube2a, ubiquitin-conjugating enzyme E2A, RAD6 homologue 54 ; Xist, X (inactive)- specific transcript 88?90 ; Xnp, X-linked nuclear protein; Zfx, zinc finger protein X-linked 57 . 12 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/genetics PERSPECTIVES ?primitive? form of X-chromosome inactivation in the pre-implantation mouse embryo 43,44 . Specifically, marsupial X-chromosome inacti- vation is incomplete in the sense that many genes escape inactivation and the extent of inactivation varies widely between cells and tissues (reviewed in REFS 1,70). So, the early mammal and the early mouse embryo share an imperfect mechanism of dosage compensation. Between 150 and 80 Mya, X-chromosome inactivation evolved from an imperfect imprinted form, as seen in marsupials, to coexist with a more complete random form that is seen in eutherians today. As if mirroring its evolutionary history, the early mouse embryo begins development with an imperfect imprinted form that is later replaced with a more complete random form after implanta- tion 43 . This parallel progression through evo- lution and development captures the idea that ontogeny recapitulates phylogeny. The recapit- ulation indicates that the pre-implantation mouse embryo might in fact be an excellent retrospective model for studying primordial forms of X-chromosome inactivation 43 . But are the molecular details of imprinted X-chromosome inactivation similar in marsu- pials and in the early mouse embryo? One important question is whether inheritance of a pre-inactivated X P might also underlie marsu- pial X-chromosome inactivation. Variations on such a possibility have been entertained 1,22,71 . The hypothesis predicts that MSCI would also take place in the metatherian clade. In fact, there is compelling evidence to indicate that MSCI occurs in the marsupial male germline (reviewed in REF. 50). How can MSCI be a source of imprint- ing in the germline? This type of imprinting mechanism might have ancient origins in the form of meiotic silencing by unpaired DNA (MSUD). In the filamentous fungus, Neurospora crassa, non-homologous DNA fails to pair at meiosis and the lack of pairing results in silencing of the unpaired region 72 . MSUD is believed to have evolved originally as a defence mechanism against invading foreign DNA and transposons, but the phenomenon has now been observed more generally. In C. elegans, males have an XO constitution, leaving the X chromosome without a pairing partner during meiosis. The absence of pair- ing is associated with silencing, a result that is reminiscent of MSCI in the mammalian germline 73 . Intriguingly, lack of pairing and silencing are coincident with accumulation of chromatin imprints that include methylation of Lys9 on histone H3 (REF. 73). So, the original imprinting mechanism might have involved MSUD-induced chromatin modifications. Ontogeny recapitulates phylogeny? Ernest Haeckel?s original nineteenth century aphorism invokes the idea that the embryonic development of a species mirrors the stages of its evolutionary past. The ontogeny and phy- logeny of X-chromosome inactivation might be an excellent example of this idea (FIG. 4). Imprinted X-chromosome inactivation is thought to have arisen either in metatherians (marsupials; ~150 million years ago (Mya)) or even earlier in prototherians (monotremes; ~200 Mya) 68,69 . As is evident from extant mar- supials, this form of imprinted X-chromosome inactivation bears a striking resemblance to the ontogeny of X-chromosome inactivation. It posits a single round of inactivation in the paternal germline, followed by a single round of reactivation in the epiblast in preparation for random inactivation in the soma (FIG. 2b). By contrast, the classical model advocates many rounds of inactiva- tion and reactivation from the paternal germline to the zygote and beyond (FIG. 2a). We therefore suggest that a pre-inactivation mechanism is not only consistent with the available evidence but is also the most par- simonious solution for the ontogeny of X-chromosome inactivation. Pre-implantation embryo Multilayered blastocyst Unilaminar blastocyst Globalized imprinted X-chromosome inactivation Leaky imprinted X-chromosome inactivation Primitive X-chromosome inactivation mechanism 150?200 Mya Leaky imprinted X-chromosome inactivation b Eutherians (placentals) Metatherians (marsupials) Prototherians (monotremes) Birds 310 Mya 210 Mya 150?180 Mya 80?100 Mya Evolutionary timeline X and Y chromosomes X-chromosome inactivation Xist? a Mammals Imprinted X-chromosome inactivation Xi Zygote Xa Egg Sperm Xa Xi Random X-chromosome inactivation Pre-implantation embryo Xa Xi Eutherians ~80?100 Mya ~150 Mya metatherians Random X-chromosome inactivation Figure 4 | Parallels through ontogeny and phylogeny. a | The time-frame of mammalian evolution. Dimorphic X and Y chromosomes evolved after mammals split off from the avian lineage. X-chromosome inactivation might have evolved concurrently. The timing of the appearance of Xist is currently unknown. b | A form of meiotic sex-chromosome inactivation-driven imprinted X-chromosome inactivation might have been the original mechanism of dosage compensation in mammals, evolving some 150?200 Mya (million years ago). This imprinted form is present in somatic cells of metatherians, which are some of the earliest mammals to have evolved (~150 Mya), and is mirrored in the pre-implantation embryo of the mouse, a eutherian mammal. The extra-embryonic tissues of some eutherians maintain the original imprinted mechanism of X-chromosome inactivation but show a more complete form of silencing. Random X-chromosome inactivation subsequently evolved in eutherian mammals, such as the mouse, and we can therefore trace its time of origin to 80?100 Mya. PERSPECTIVES genome-wide balance of gene dosage, it is easy to imagine that the activation of the sin- gle X chromosome in the male would be turned up twofold, as this would ensure that the X-to-autosome gene dosage remained the same 81 . This is a relatively frugal approach to dosage compensation because, conceptually, this mechanism requires only one significant step in evolution. By contrast, dosage com- pensation in mammals is an inactivation mechanism that seems complicated from two perspectives. First, as argued above, X-chromosome inactivation is a multi-step process that involves counting, epigenetic choice and cis-limited silencing ? a system that seems difficult to evolve simultaneously. The D. melanogaster system is also regulated by a chromosome-counting mechanism, but this mechanism was presumably already in place for the determination of sex 24,25 . Epigenetic choice does not take place in D. melanogaster, nor is dosage compensation cis-limited. Second, X-chromosome inactivation is complicated from the perspective that if female X-chromosome inactivation was concurrent with male Y-chromosome degen- eration it would have led to a temporary imbalance between X-chromosome and autosomal gene dosages, owing to a sudden halving of X-linked gene transcription. This is very different from fruitflies, in which concur- rent male X-chromosome upregulation and male Y-chromosome degeneration would have preserved the overall X-to-autosome gene balance. According to one model, early mammals responded to this problem by a general increase in X-linked transcription as the Y chromosome decayed 81 . But because Might a similar imprinting mechanism be used by mammals during MSCI? Recent studies show that the state of asynapsis in the mammalian germline directly results in silen- cing 74?76 (see also REF. 53). This unpaired silencing depends on the coordinated actions of the breast cancer predisposition gene, Brca1, various modifications of the histone variant H2AX (including phosphorylation and ubiquitylation), and the ATR kinase, all of which localize to the asynapsed sex chromo- somes 74?76 (FIG. 5). Taken together, these find- ings potentially provide a logical mechanism by which MSCI might lead to X-chromosome imprinting in the mammalian germline. With MSCI also being present in the mar- supial germline, the idea that MSUD might be the original imprinting mechanism for X-chromosome inactivation becomes rather attractive. This idea also fits nicely with a recent hypothesis that genomic imprinting evolved first on the X chromosome for the purposes of X-linked dosage compensa- tion 77 . Once fixed and further adapted on the X chromosome, a modified imprinting mechanism could have easily spread to autosomal locations by transposition of the X-linked imprinting cassette. The role of Xist in the ontogeny and phy- logeny of X-chromosome inactivation is of particular interest (FIG. 4a). Although Xist is clearly required for somatic and placental X-chromosome inactivation, MSCI does not seem to require an intact Xist gene 23,58,78 . Whether the pre-implantation form of imprinted X-chromosome inactivation involves Xist is currently not known. Because Xist is expressed in early embryos and the spread of Xist RNA loosely correlates with the gradient of silencing on the X p 43 , it seems likely that Xist will be necessary. However, the role of Xist in pre-implantation silencing in the mouse might strictly be to maintain X P inactivation. In this model, X P inactiva- tion is initiated by the paternal germline through an Xist-independent MSUD mecha- nism, but silencing is maintained thereafter in the zygote in an Xist-dependent fashion. The idea of heterochromatin that is induced by different mechanisms in two stages is interesting in light of a recent report on spa- tially distinct classes of heterochromatin on the Xi 79 . These classes of heterochromatin have distinct epigenetic marks. Class I is enriched for heterochromatin protein 1 (HP1), trimethylation at H3-K9 and tri- methylation at H4-K20, but is not associ- ated with Xist RNA. By contrast, class II is enriched for Xist RNA, trimethylation at H3-K27 and the histone variant macroH2A. We could speculate that the inherited X P resembles the class I heterochromatin of the somatic Xi and that the expression of Xist RNA at the 2-cell stage leads to the transition to class II heterochromatin, which would gradually provide a more complete silenc- ing, as is observed in the trophectoderm. In this sense, X-chromosome inactivation in pre-implantation embryos can be viewed as proceeding in a progressive manner. Is there a primordial Xist orthologue in marsupials? So far, such a gene has not been described. Although it might just be a mat- ter of time before an Xist equivalent is found, it is formally possible that marsupial X-chromosome inactivation might be Xist- independent. Consistent with the idea that MSUD might have been the ancestral mech- anism of imprinting, it is possible that the earliest form of X-chromosome inactivation was entirely MSCI-driven (FIG. 4b). In this model, Xist could have been acquired subse- quently by eutherians to carry out a more global and complete form of inactivation, as observed in the post-implantation tissues (placenta and soma) 43 . On the other hand, given the obvious similarities between euther- ian and metatherian X-chromosome inacti- vation, the ancestral form of X-chromosome inactivation might have also required Xist RNA, but perhaps in a form that was more localized and more variable in its degree of silencing ? similar to the form that is seen in the pre-implantation mouse embryo 43 . Why X inactivation in mammals? Mammalian sex chromosomes are believed to have descended from a pair of identical auto- somes (reviewed in REFS 80,81). Approximately 200 Mya, the X and the Y chromosomes began to differentiate into the dimorphic chromosomes that they are in mammals today, with the Y chromosome losing almost all of its genetic material (except what is required for sex determination) owing to sup- pression of recombination between the X and Y chromosomes during male meiosis (for reviews of Y-chromosome degeneration see REFS 3,80,81). The progressive loss of genetic material from the Y chromosome would have generated an increasing dosage imbalance between XX and XY individuals. This imbal- ance is thought to have given rise to the cur- rent mechanism of dosage compensation in mammals (reviewed in REF. 81). But why did mammals evolve an inactivation mechanism? Sex chromosomes in D. melanogaster are also thought to have evolved from a pair of autosomes, and to have been subject to the forces that shaped the dimorphic features of the X and Y chromosomes (reviewed in REF. 81). In the interest of maintaining a NATURE REVIEWS | GENETICS ADVANCE ONLINE PUBLICATION | 13 XY X Y ?-H2AX ATR BRCA1 PAR pairing Meiotic silencing by unpaired DNA Figure 5 | Meiotic sex-chromosome inactivation by unpaired silencing in mammals. Unpaired regions of the X (blue) and Y (yellow) chromosomes are transcriptionally silenced at the pachytene stage of spermatogenesis. This silencing is dependent on the localization of BRCA1 (breast cancer 1, early onset), ?-H2AX (the phosphorylated form of H2A histone family, member X), and ATR (ataxia telangiectasia and rad3 related) proteins (red circles) to the unpaired regions. Dotted lines indicate limited pairing at the pseudoautosomal regions (PARs). 14 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/genetics PERSPECTIVES This logic also naturally raises the oppo- site question: why do we not see a mecha- nism similar to X-chromosome inactivation in the fruitfly, the worm and other organ- isms with heterogametic sex determination? Indeed, meiotic sex-chromosome silencing is well described in both flies 48 and worms 82 . In worms, MSUD has also recently been shown to be the mechanism that is respon- sible for male germline silencing 73 . We argue that these other organisms were locked into disparate mechanisms of dosage compensa- tion because genetic circumstances limited their evolutionary choices. Specifically, in the fruitfly and the worm, sex is determined by the X-to-autosome ratio that is mea- sured during the first few cell divisions (reviewed in REF. 24). The X-to-autosome ratio is quantified by relative expression of specific X-linked and autosomal genes. So, although MSUD might occur in the male germline, dosage compensation cannot occur by maintenance of a pre-inactivated X p in the zygote, as this would interfere with measure- ment of the number of X chromosomes. In mammals, sex is not influenced by the X-to- autosome ratio, but is determined by the presence or absence of the male-determining Y chromosome. So, although X-chromosome inactivation might have been the natural choice for mammals, it would not have been a logical choice for D. melanogaster or C. elegans. It seems that the strategy of dosage compen- sation for a given organism is directly affected by two factors ? first, how sex is determined, and second, how their sex chromosomes evolved (FIG. 6). A corollary to this hypothesis is that X-chromosome inactivation need not be unique to mammals and might arise inde- pendently in other organisms with a com- patible sex-determination system. Indeed, an epigenetic phenomenon that is rem- iniscent of mammalian X-chromosome inactivation has been described in a dioe- cious plant species, Silene latifolia (also known as Melandrium album) (reviewed in REF. 83). Similar to mammals, S. latifolia has an XX?XY sex-determination system with the Y chromosome carrying a domi- nant male-determining gene. One of two X chromosomes in the female seems to be hypermethylated and late-replicating 84,85 , two tell-tale signs of inert transcription. Furthermore, as in mammals, the X and Y chromosomes synapse only at their ends during meiosis in S. latifolia 86 . These obser- vations indicate that MSUD might also operate in plant meiosis and that MSCI might be more widespread than is currently appreciated. Although the mammalian system might seem unnecessarily complex, we suggest that it might in fact have been the easiest solution available. First, complete silencing might be safer than a mechanism that requires very fine tuning to allow for the precise doubling of gene expression. Second, as the Y chromo- some began to lose genetic material, a simple solution for dosage compensation would have been to silence the regions of the X chromo- some that no longer had pairing partners on the Y chromosome ? and what better way to achieve this than through an already avail- able, ancient mechanism of MSUD? As the Y chromosome continued to degenerate, MSUD would progressively affect more and more of the X chromosome, culminating in what we see in rodents today, where nearly all of the X chromosome is subject to silenc- ing. Because only female offspring can inherit the X chromosome from the father, an MSUD-driven imprinted process would have immediately solved the problem of dosage compensation in the early mammal, without the need to evolve counting or choice mechanisms at the same time. X-chromosome upregulation occurred in both XY and XX individuals, females faced the problem of double X-chromosome dosage and were therefore subject to considerable pressure to downregulate their X chromosomes. So, in mammals, the responsibility for dosage compensation ultimately rested with the female. As argued above, MSUD-mediated silencing in the male germline would have been a simple first solution to ensure that XX zygotes started out without unwanted X-linked gene dosage. In summary, the evolution of X-chromo- some inactivation is thought to have involved a series of significant evolutionary steps in response to Y-chromosome degeneration, including generalized X-chromosome upreg- ulation in both sexes, followed by female- specific X-chromosome inactivation 81 . This seems more complicated than the process that is thought to have occurred in the fruit- fly. Using this type of logic, it might be dif- ficult to imagine why mammals did not simply adopt the male-specific upregulation of the X chromosome � la D. melanogaster. Dosage- compensation scheme Egg Sperm Mammals Caenorhabditis elegans Drosophila melanogaster XX Zygote X-chromosome reactivation X-chromosome inactivation maintenance Dominant sex-determining factor (Y-linked) Sex- determination system X:A ratio X-chromosome inactivation X-chromosome downregulation in females X-chromosome upregulation in males Figure 6 | A generalized model for how the nature of the dosage-compensation mechanism depends on the method of sex determination. In this model, heteromorphic sex chromosomes are subjected to meiotic silencing by unpaired DNA during gametogenesis, and the X chromosome is transmitted to the zygote as an inactive chromosome. Dosage compensation in the zygote will then depend on an organism?s method of sex determination: sex determination that is based on a dominant sex-determining factor(s) (for example, the Y-linked Sry) is conducive to preserving the inherited inactive X chromosome. But sex-determining mechanisms that depend on the X:A ratio (the ratio of X chromosomes to autosomes) must reactivate the X chromosome to obtain an accurate X-chromosome number measurement. So, a non-X-chromosome inactivation type of dosage compensation can evolve in this system (for example, downregulation of the female X chromosomes in Caenorhabditis elegans or upregulation of the male X chromosome in Drosophila melanogaster). Red circles, heterochromatin. PERSPECTIVES 15. Penny, G. D., Kay, G. F., Sheardown, S. A., Rastan, S. & Brockdorff, N. Requirement for Xist in X chromosome inactivation. Nature 379, 131?137 (1996). 16. Sharman, G. B. Late DNA replication in the paternally derived X chromosome of female kangaroos. Nature 230, 231?232 (1971). 17. Takagi, N. & Sasaki, M. Preferential inactivation of the paternally derived X chromosome in the extraembryonic membranes of the mouse. Nature 256, 640?642 (1975). 18. Xue, F. et al. Aberrant patterns of X chromosome inactivation in bovine clones. Nature Genet. 31, 216?220 (2002). 19. Migeon, B. R. & Do, T. T. In search of non-random X inactivation: studies of fetal membranes heterozygous for glucose-6-phosphate dehydrogenase. Am. J. Hum. Genet. 31, 581?585 (1979). 20. Ropers, H. H., Wolff, G. & Hitzeroth, H. W. Preferential X inactivation in human placenta membranes: is the paternal X inactive in early embryonic development of female mammals? Hum. Genet. 43, 265?273 (1978). 21. Zeng, S. M. & Yankowitz, J. X-inactivation patterns in human embryonic and extra-embryonic tissues. Placenta 24, 270?275 (2003). 22. Cooper, D. W. Directed genetic change model for X chromosome inactivation in eutherian mammals. Nature 230, 292?294 (1971). 23. Marahrens, Y., Panning, B., Dausman, J., Strauss, W. & Jaenisch, R. Xist-deficient mice are defective in dosage compensation but not spermatogenesis. Genes Dev. 11, 156?166 (1997). 24. Cline, T. W. & Meyer, B. J. Vive la difference: males vs females in flies vs worms. Annu. Rev. Genet. 30, 637?702 (1996). 25. Park, Y. & Kuroda, M. I. Epigenetic aspects of X-chromosome dosage compensation. Science 293, 1083?1085 (2001). 26. Takagi, N. Imprinted X-chromosome inactivation: enlightenment from embryos in vivo. Semin. Cell Dev. Biol. 14, 319?329 (2003). 27. Goto, T. & Monk, M. Regulation of X-chromosome inactivation in development in mice and humans. Microbiol. Mol. Biol. Rev. 62, 362?378 (1998). 28. Heard, E., Clerc, P. & Avner, P. X-chromosome inactivation in mammals. Annu. Rev. Genet. 31, 571?610 (1997). 29. Lyon, M. F. X-chromosome inactivation and developmental patterns in mammals. Biol. Rev. Camb. Philos. Soc. 47, 1?35 (1972). 30. Adler, D. A., West, J. D. & Chapman, V. M. Expression of ?-galactosidase in preimplantation mouse embryos. Nature 267, 838?839 (1977). 31. Gardner, R. L. & Lyon, M. F. X chromosome inactivation studied by injection of a single cell into the mouse blastocyst. Nature 231, 385?386 (1971). 32. Epstein, C. J., Smith, S., Travis, B. & Tucker, G. Both X chromosomes function before visible X-chromosome inactivation in female mouse embryos. Nature 274, 500?503 (1978). 33. Kratzer, P. G. & Gartler, S. M. HGPRT activity changes in preimplantation mouse embryos. Nature 274, 503?504 (1978). 34. Monk, M. & Harper, M. I. Sequential X chromosome inactivation coupled with cellular differentiation in early mouse embryos. Nature 281, 311?313 (1979). 35. Mukherjee, A. B. Cell cycle analysis and X-chromosome inactivation in the developing mouse. Proc. Natl Acad. Sci. USA 73, 1608?1611 (1976). 36. Sugawara, O., Takagi, N. & Sasaki, M. Correlation between X-chromosome inactivation and cell differentiation in female preimplantation mouse embryos. Cytogenet. Cell Genet. 39, 210?219 (1985). 37. Latham, K. E. & Rambhatla, L. Expression of X-linked genes in androgenetic, gynogenetic, and normal mouse preimplantation embryos. Dev. Genet. 17, 212?222 (1995). 38. Matsui, J., Goto, Y. & Takagi, N. Control of Xist expression for imprinted and random X chromosome inactivation in mice. Hum. Mol. Genet. 10, 1393?1401 (2001). 39. Sheardown, S. A. et al. Stabilization of Xist RNA mediates initiation of X chromosome inactivation. Cell 91, 99?107 (1997). 40. Nesterova, T. B., Barton, S. C., Surani, M. A. & Brockdorff, N. Loss of Xist imprinting in diploid parthenogenetic preimplantation embryos. Dev. Biol. 235, 343?350 (2001). 41. Krietsch, W. K. et al. The expression of X-linked phosphoglycerate kinase in the early mouse embryo. Differentiation 23, 141?144 (1982). 42. Pravtcheva, D. D., Adra, C. N. & Ruddle, F. H. Timing of paternal Pgk-1 expression in embryos of transgenic mice. Development 111, 1109?1120 (1991). 43. Huynh, K. D. & Lee, J. T. Inheritance of a pre-inactivated paternal X chromosome in early mouse embryos. Nature 426, 857?862 (2003). Conclusions We propose that there is a link between the ontogeny and phylogeny of X-chromosome inactivation. We have suggested that our recent finding and those of others indi- cate that X-chromosome inactivation in mammals evolved from an ancient silenc- ing mechanism (MSUD) that was first described in N. crassa. As MSUD is con- served throughout evolution, from N. crassa 72 to C. elegans 73 to mammals 75 , it will be interesting to learn how widespread the MSUD-driven X-chromosome inactivation type of mechanism might be. In fact, the spectre of X-chromosome inactivation in the plant species S. latifolia 84,85 indicates that it might be more widespread than is currently appreciated. There are several important questions that remain to be answered. To what extent is the X P -expression pattern heritable from father to zygote in mice? How well is this mirrored by X-chromosome inactiva- tion in marsupials and other organisms that undergo X-chromosome inactivation? Does Xist have a role? It will also be crucial to determine to what extent MSUD and dosage compensation can be applied to vertebrates, particularly the earliest mam- mals, the prototherians, and their close cousins, the birds. Preliminary evidence seems to favour dosage-compensation mechanisms in prototherians and birds 69,87 . If present, is it an inactivation type of mechanism and is it imprinted? The answers to these questions might further support the link between ontogeny and phylogeny. Khanh D. Huynh and Jeannie T. Lee are at the Howard Hughes Medical Institute, Department of Molecular Biology, Massachusetts General Hospital and Department of Genetics, Harvard Medical School, Boston, Massachusetts 02114, USA. Correspondence to J.T.L. e-mail: lee@frodo.mgh.harvard.edu doi:10.1038/nrg1604 Published online 8 April 2005 1. Graves, J. A. Mammals that break the rules: genetics of marsupials and monotremes. Annu. Rev. Genet. 30, 233?260 (1996). 2. Lahn, B. T. & Page, D. C. Functional coherence of the human Y chromosome. Science 278, 675?680 (1997). 3. Lahn, B. T., Pearson, N. M. & Jegalian, K. The human Y chromosome, in the light of evolution. Nature Rev. Genet. 2, 207?216 (2001). 4. Vallender, E. J. & Lahn, B. T. How mammalian sex chromosomes acquired their peculiar gene content. Bioessays 26, 159?169 (2004). 5. Spatz, A., Borg, C. & Feunteun, J. X-chromosome genetics and human cancer. Nature Rev. Cancer 4, 617?629 (2004). 6. Lyon, M. F. Gene action in the X chromosome of the mouse (Mus musculus L.). Nature 190, 372?373 (1961). 7. Avner, P. & Heard, E. X-chromosome inactivation: counting, choice and initiation. Nature Rev. Genet. 2, 59?67 (2001). 8. Brockdorff, N. et al. 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). 9. Brown, C. J. et al. The human XIST gene: analysis of a 17-kb inactive X-specific RNA that contains conserved repeats and is highly localized within the nucleus. Cell 71, 527?542 (1992). 10. Lee, J. T., Davidow, L. S. & Warshawsky, D. Tsix, a gene antisense to Xist at the X-inactivation centre. Nature Genet. 21, 400?404 (1999). 11. Ogawa, Y. & Lee, J. T. Xite, X-inactivation intergenic transcription elements that regulate the probability of choice. Mol. Cell 11, 731?743 (2003). 12. Lee, J. T. Disruption of imprinted X inactivation by parent-of-origin effects at Tsix. Cell 103, 17?27 (2000). 13. Sado, T., Wang, Z., Sasaki, H. & Li, E. Regulation of imprinted X-chromosome inactivation in mice by Tsix. Development 128, 1275?1286 (2001). 14. Clemson, C. M., McNeil, J. A., Willard, H. F. & Lawrence, J. B. XIST RNA paints the inactive X chromosome at interphase: evidence for a novel RNA involved in nuclear/chromosome structure. J. Cell Biol. 132, 259?275 (1996). NATURE REVIEWS | GENETICS ADVANCE ONLINE PUBLICATION | 15 Glossary ASYNAPSIS Failure of chromosomes to pair during meiosis. BLASTOCYST An early stage of mammalian embryonic development at which the first cell lineages become established. COT1 FISH A technique to visualize nascent transcription that uses the Cot1 fraction of DNA that is rich in repetitive elements that often occur in introns and 3? untranslated regions. EPIBLAST An embryonic lineage that is derived from the inner cell mass of the blastocyst, which gives rise to the body of the fetus. EUTHERIANS Mammals that give birth to live offspring (that is, they are viviparous) and possess an allantoic placenta ? the allantois is the fetal membrane that facilitates nutrient and waste exchange between the fetus and the mother. FLUORESCENCE IN SITU HYBRIDIZATION A technique in which a fluorescently labelled DNA probe is used to detect a particular chromosome or gene with the help of fluorescence microscopy. MEIOTIC CHECKPOINT A surveillance mechanism specific to meiosis that ensures proper chromosome segregation. PACHYTENE The third phase of prophase I in meiosis. THERIAN MAMMALS A group of mammals that includes the eutherians and marsupials. TROPHECTODERM The outer layer of the blastocyst-stage embryo; the precursor to the bulk of the embryonic part of the placenta. 16 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/genetics PERSPECTIVES 82. Kelly, W. G. et al. X-chromosome silencing in the germline of C. elegans. Development 129, 479?492 (2002). 83. Charlesworth, D. Plant sex determination and sex chromosomes. Heredity 88, 94?101 (2002). 84. Siroky, J., Castiglione, M. R. & Vyskot, B. DNA methylation patterns of Melandrium album chromosomes. Chromosome Res. 6, 441?446 (1998). 85. Vyskot, B., Siroky, J., Hladilova, R., Belyaev, N. D. & Turner, B. M. Euchromatic domains in plant chromosomes as revealed by H4 histone acetylation and early DNA replication. Genome 42, 343?350 (1999). 86. Lengerova, M., Moore, R. C., Grant, S. R. & Vyskot, B. The sex chromosomes of Silene latifolia revisited and revised. Genetics 165, 935?938 (2003). 87. McQueen, H. A., McBride, D., Miele, G., Bird, A. P. & Clinton, M. Dosage compensation in birds. Curr. Biol. 11, 253?257 (2001). 88. McCarrey, J. R. & Dilworth, D. D. Expression of Xist in mouse germ cells correlates with X-chromosome inactivation. Nature Genet. 2, 200?203 (1992). 89. Salido, E. C., Yen, P. H., Mohandas, T. K. & Shapiro, L. J. Expression of the X-inactivation-associated gene XIST during spermatogenesis. Nature Genet. 2, 196?199 (1992). 90. Richler, C., Soreq, H. & Wahrman, J. X inactivation in mammalian testis is correlated with inactive X-specific transcription. Nature Genet. 2, 192?195 (1992). 91. Singer-Sam, J., Chapman, V., LeBon, J. M. & Riggs, A. D. Parental imprinting studied by allele-specific primer extension after PCR: paternal X chromosome-linked genes are transcribed prior to preferential paternal X chromosome inactivation. Proc. Natl Acad. Sci. USA 89, 10469?10473 (1992). 92. Mak, W. et al. Reactivation of the paternal X chromosome in early mouse embryos. Science 303, 666?669 (2004). Acknowledgements We thank all the members of our laboratory for insightful discussion and R. Spencer, S. Namekawa and M. Anguera for critical reading of the manuscript. Competing interests statement The authors declare no competing financial interests. Online links DATABASES The following terms in this article are linked online to: Entrez: http://www.ncbi.nlm.nih.gov/entrez Chic1 | EED | EZH2 | G6pd2 | Pdha2 | Pgk1 | Pgk2 | Tsix | Xist | Zfa FURTHER INFORMATION Jeannie Lee?s web site: http://genetics.mgh.harvard.edu/LeeWeb/lab%20personnel/ jeannielee.html Access to this interactive links box is free online pyruvate dehydrogenase E1-? subunit is coded for by an intronless gene on chromosome 4. Genomics 8, 225?232 (1990). 64. McCarrey, J. R. & Thomas, K. Human testis-specific PGK gene lacks introns and possesses characteristics of a processed gene. Nature 326, 501?505 (1987). 65. Emerson, J. J., Kaessmann, H., Betran, E. & Long, M. Extensive gene traffic on the mammalian X chromosome. Science 303, 537?540 (2004). 66. Wang, P. J. X chromosomes, retrogenes and their role in male reproduction. Trends Endocrinol. Metab. 15, 79?83 (2004). 67. Huynh, K. D. & Lee, J. T. Imprinted X inactivation in eutherians: a model of gametic execution and zygotic relaxation. Curr. Opin. Cell Biol. 13, 690?697 (2001). 68. Grutzner, F. et al. In the platypus a meiotic chain of ten sex chromosomes shares genes with the bird Z and mammal X chromosomes. Nature 432, 913?917 (2004). 69. Grutzner, F. & Graves, J. A. A platypus? eye view of the mammalian genome. Curr. Opin. Genet. Dev. 14, 642?649 (2004). 70. VandeBerg, J. L., Johnston, P. G., Cooper, D. W. & Robinson, E. S. X-chromosome inactivation and evolution in marsupials and other mammals. Isozymes Curr. Top. Biol. Med. Res. 9, 201?218 (1983). 71. Lyon, M. F. Imprinting and X-chromosome inactivation. Results Probl. Cell Differ. 25, 73?90 (1999). 72. Shiu, P. K., Raju, N. B., Zickler, D. & Metzenberg, R. L. Meiotic silencing by unpaired DNA. Cell 107, 905?916 (2001). 73. Bean, C. J., Schaner, C. E. & Kelly, W. G. Meiotic pairing and imprinted X chromatin assembly in Caenorhabditis elegans. Nature Genet. 36, 100?105 (2004). 74. Turner, J. M. et al. BRCA1, histone H2AX phosphorylation, and male meiotic sex chromosome inactivation. Curr. Biol. 14, 2135?2142 (2004). 75. Turner, J. M. et al. Silencing of unsynapsed meiotic chromosomes in the mouse. Nature Genet. 37, 41?47 (2005). 76. Baarends, W. M. et al. Silencing of unpaired chromatin and histone H2A ubiquitination in mammalian meiosis. Mol. Cell. Biol. 25, 1041?1053 (2005). 77. Lee, J. T. Molecular links between X-inactivation and autosomal imprinting: X-inactivation as a driving force for the evolution of imprinting? Curr. Biol. 13, R242?R254 (2003). 78. Turner, J. M. et al. Meiotic sex chromosome inactivation in male mice with targeted disruptions of Xist. J. Cell Sci. 115, 4097?4105 (2002). 79. Chadwick, B. P. & Willard, H. F. Multiple spatially distinct types of facultative heterochromatin on the human inactive X chromosome. Proc. Natl Acad. Sci. USA 101, 17450?17455 (2004). 80. Charlesworth, B. The evolution of sex chromosomes. Science 251, 1030?1033 (1991). 81. Charlesworth, B. The evolution of chromosomal sex determination and dosage compensation. Curr. Biol. 6, 149?162 (1996). 44. Okamoto, I., Otte, A. P., Allis, C. D., Reinberg, D. & Heard, E. Epigenetic dynamics of imprinted X inactivation during early mouse development. Science 303, 644?649 (2004). 45. Brown, C. J. et al. Localization of the X inactivation centre on the human X chromosome in Xq13. Nature 349, 82?84 (1991). 46. Schultz, R. M. Regulation of zygotic gene activation in the mouse. Bioessays 15, 531?538 (1993). 47. Reik, W. & Walter, J. Genomic imprinting: parental influence on the genome. Nature Rev. Genet. 2, 21?32 (2001). 48. Lifschytz, E. & Lindsley, D. L. The role of X-chromosome inactivation during spermatogenesis. Proc. Natl Acad. Sci. USA 69, 182?186 (1972). 49. Lifschytz, E. & Lindsley, D. I. Sex chromosome activation during spermatogenesis. Genetics 78, 323?331 (1974). 50. Hoyer-Fender, S. Molecular aspects of XY body formation. Cytogenet. Genome Res. 103, 245?255 (2003). 51. McKee, B. D. & Handel, M. A. Sex chromosomes, recombination, and chromatin conformation. Chromosoma 102, 71?80 (1993). 52. Miklos, G. L. Sex-chromosome pairing and male fertility. Cytogenet. Cell Genet. 13, 558?577 (1974). 53. Lee, J. T. Sex chromosome inactivation: the importance of pairing. Curr. Biol. (in the press). 54. Hendriksen, P. J. et al. Postmeiotic transcription of X and Y chromosomal genes during spermatogenesis in the mouse. Dev. Biol. 170, 730?733 (1995). 55. Khalil, A. M., Boyar, F. Z. & Driscoll, D. J. Dynamic histone modifications mark sex chromosome inactivation and reactivation during mammalian spermatogenesis. Proc. Natl Acad. Sci. USA 101, 16583?16587 (2004). 56. McCarrey, J. R. et al. Differential transcription of Pgk genes during spermatogenesis in the mouse. Dev. Biol. 154, 160?168 (1992). 57. McCarrey, J. R., Dilworth, D. D. & Sharp, R. M. Semiquantitative analysis of X-linked gene expression during spermatogenesis in the mouse: ethidium-bromide staining of RT-PCR products. Genet. Anal. Tech. Appl. 9, 117?123 (1992). 58. McCarrey, J. R. et al. X-chromosome inactivation during spermatogenesis is regulated by an Xist/Tsix-independent mechanism in the mouse. Genesis 34, 257?266 (2002). 59. Singer-Sam, J., Robinson, M. O., Bellve, A. R., Simon, M. I. & Riggs, A. D. Measurement by quantitative PCR of changes in HPRT, PGK-1, PGK-2, APRT, MTase, and Zfy gene transcripts during mouse spermatogenesis. Nucleic Acids Res. 18, 1255?1259 (1990). 60. Hendriksen, P. J. et al. Testis-specific expression of a functional retroposon encoding glucose-6-phosphate dehydrogenase in the mouse. Genomics 41, 350?359 (1997). 61. Ashworth, A., Skene, B., Swift, S. & Lovell-Badge, R. Zfa is an expressed retroposon derived from an alternative transcript of the Zfx gene. EMBO J. 9, 1529?1534 (1990). 62. Bradley, J. et al. An X-to-autosome retrogene is required for spermatogenesis in mice. Nature Genet. 36, 872?876 (2004). 63. Dahl, H. H., Brown, R. M., Hutchison, W. M., Maragos, C. & Brown, G. K. A testis-specific form of the human "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
No comments yet.
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment