Lineage-specific function of the noncoding Tsix RNA for Xist repression and Xi reactivation in mice

The noncoding Tsix RNA is an antisense repressor of Xist and regulates X inactivation in mice. Tsix is essential for preventing the inactivation of the maternally inherited X chromosome in extraembryonic lineages where imprinted X-chromosome inactivation (XCI) occurs. Here we establish an inducible Tsix expression system for investigating Tsix function in development. We show that Tsix has a clear functional window in extraembryonic development. Within this window, Tsix can repress Xist, which is accompanied by DNA methylation of the Xist promoter. As a consequence of Xist repression, reactivation of the inactive X chromosome (Xi) is widely observed. In the parietal endoderm, Tsix represses Xist and causes reactivation of an Xi-linked GFP transgene throughout development, whereas Tsix progressively loses its Xist-repressing function from embryonic day 9.5 (E9.5) onward in trophoblast giant cells and spongiotrophoblast, suggesting that Tsix function depends on a lineage-specific environment. Our data also demonstrate that the maintenance of imprinted XCI requires Xist expression in specific extraembryonic tissues throughout development. This finding shows that reversible XCI is not exclusive to pluripotent cells, and that in some lineages cell differentiation is not accompanied by a stabilization of the Xi.     

Characterization of the genomic Xist locus in rodents reveals conservation of overall gene structure and tandem repeats but rapid evolution of unique sequence

The Xist locus plays a central role in the regulation of X chromosome inactivation in mammals, although its exact mode of action remains to be elucidated. Evolutionary studies are important in identifying conserved genomic regions and defining their possible function. Here we report cloning, sequence analysis, and detailed characterization of the Xist gene from four closely related species of common vole (field mouse), Microtus arvalis. Our analysis reveals that there is overall conservation of Xist gene structure both between different vole species and relative to mouse and human Xist/XIST. Within transcribed sequence, there is significant conservation over five short regions of unique sequence and also over Xist-specific tandem repeats. The majority of unique sequences, however, are evolving at an unexpectedly high rate. This is also evident from analysis of flanking sequences, which reveals a very high rate of rearrangement and invasion of dispersed repeats. We discuss these results in the context of Xist gene function and evolution.     

Transgenic mice carrying an Xist-containing YAC

The initiation of X-chromosome inactivation in female mammals is controlled by a key locus, the X-inactivation centre (Xic). The Xist gene, which maps to the candidate region for Xic and is expressed exclusively from the inactive X chromosome, is thought to be an essential component of the Xic. To test whether sequences spanning several hundred kilobases and including Xist from the Xic region are capable of initiating inactivation, we have created a series of transgenic mice using a 460 kb yeast artificial chromosome (YAC). Analysis in these mice of the expression of Xist, of a LacZ reporter gene and of two genes in the region that are normally silent on the inactive X chromosome, suggests that essential sequences for Xist expression and X-inactivation may be absent in these transgenic animals.      

Familial cases of point mutations in the XIST promoter reveal a correlation between CTCF binding and pre-emptive choices of X chromosome inactivation

The choice mechanisms that determine the future inactive X chromosome in somatic cells of female mammals involve the regulated expression of the XIST gene. A familial C(-43)G mutation in the XIST promoter results in skewing of X chromosome inactivation (XCI) towards the inactive X chromosome of heterozygous females, whereas a C(-43)A mutation found primarily in the active X chromosome results in the opposite skewing pattern. Both mutations point to the existence of a factor that might be responsible for the skewed patterns. Here we identify this factor as CTCF, a conserved protein with a 11 Zn-finger (ZF) domain that can mediate multiple sequence-specificity and interactions between DNA-bound CTCF molecules. We show that mouse and human Xist/XIST promoters contain one homologous CTCF-binding sequence with the matching dG-contacts, which in the human XIST include the -43 position within the DNase I footprint of CTCF. While the C(-43)A mutation abrogates CTCF binding, the C(-43)G mutation results in a dramatic increase in CTCF-binding efficiency by altering ZF-usage mode required for recognition of the altered dG-contacts of the mutant site. Thus, the skewing effect of the two -43C mutations correlates with their effects on CTCF binding. Finally, CTCF interacts with the XIST/Xist promoter only in female human and mouse cells. The interpretation that this reflected a preferential interaction with the promoter of the active Xist allele was confirmed in mouse fetal placenta. These observations are in keeping with the possibility that the choice of X chromosome inactivation reflects stabilization of a higher order chromatin conformation impinging on the CTCF-XIST promoter complex.     

Unbalanced X;autosome translocations provide evidence for sequence specificity in the association of XIST RNA with chromatin

Whether XIST RNA is indifferent to the sequence content of the chromosome is fundamental to understanding its mechanism of chromosomal inactivation. Transgenic Xist RNA appears to associate with and inactivate an entire autosome. However, the behavior of XIST RNA on naturally occurring human X;autosome translocations has not been thoroughly investigated. Here, the relationship of human XIST RNA to autosomal chromatin is investigated in cells from two patients carrying X;autosome translocations in the context of almost complete trisomy for the involved autosome. Since trisomies of either 14 or 9 are lethal in early development, the lack of serious phenotypic consequences of the trisomy demonstrates that the translocated autosomes had been inactivated. Surprisingly, our analyses show that in primary fibroblasts from adult patients, XIST RNA does not associate with most of the involved autosome even though the bulk of it exhibits other hallmarks of inactivation beyond the region associated with XIST RNA. While results show that XIST RNA can associate with human autosomal chromatin to some degree, several observations indicate that this interaction may be unstable, with progressive loss over time. Thus, even where autosomal inactivation is selected for rather than against, there is a fundamental difference in the affinity of XIST RNA for autosomal versus X chromatin. Based on these results we propose that even autosomal chromatin that had been inactivated earlier in development may undergo a stepwise loss of inactivation hallmarks, beginning with XIST RNA. Hence compromised interaction with XIST RNA may be a primary cause of incomplete or unstable autosomal inactivation.     

Comparative mapping of X chromosomes in vole species of the genus Microtus

Comparative mapping of X-linked genes has progressed rapidly since Ohno's prediction that genes on the X chromosome should be conserved as a syntenic group in all mammals. Although several conserved blocks of homology between human and mouse have been discovered, rearrangements within the X chromosome have also been characterized. More recently, some exceptions to Ohno's law have been reported. We have used fluorescence in situ hybridization (FISH) to map five genes, Gla, G6pd, Hprt, Pgk1 and Xist, to two of the largest conserved segments of X material in five members of the genus Microtus (grey vole) and show that vole X chromosomes demonstrate greater homology to human than to mouse. Cytogenetic analysis indicates a relatively high frequency of rearrangement during vole evolution, although certain blocks of homology appear to be highly conserved in all species studied to date. On this basis we were able to predict the probable location of the rat X inactivation centre (Xic) based solely on high-resolution G-banding. Our prediction was then confirmed by mapping the rat Xist gene by FISH. The possible significance of conserving long-range chromosome structure in the vicinity of the Xic is discussed with respect to the mechanism of X inactivation.