Spatial partitioning of the regulatory landscape of the X-inactivation centre

Abstract

In eukaryotes transcriptional regulation often involves multiple long-range elements and is influenced by the genomic environment1. A prime example of this concerns the mouse X-inactivation centre (Xic), which orchestrates the initiation of X-chromosome inactivation (XCI) by controlling the expression of the nonprotein- coding Xist transcript. The extent of Xic sequences required for the proper regulation of Xist remains unknown. Here we use chromosome conformation capture carbon-copy (5C)2 and super-resolution microscopy to analyse the spatial organization of a 4.5-megabases (Mb) region including Xist. We discover a series of discrete 200-kilobase to 1Mb topologically associating domains (TADs), present both before and after cell differentiation and on the active and inactive X. TADs align with, but do not rely on, several domain-wide features of the epigenome, such as H3K27me3 or H3K9me2 blocks and lamina-associated domains.TADs also align with coordinately regulated gene clusters. Disruption of a TAD boundary causes ectopic chromosomal contacts and long-range transcriptional misregulation. The Xist/Tsix sense/antisense unit illustrates how TADs enable the spatial segregation of oppositely regulated chromosomal neighbourhoods, with the respective promoters of Xist and Tsix lying in adjacent TADs, each containing their known positive regulators.Weidentify a novel distal regulatory region of Tsix within its TAD, which produces a long intervening RNA, Linx. In addition to uncovering a new principle ofcis-regulatory architecture ofmammalian chromosomes, our study sets the stage for the full genetic dissection of the X-inactivation centre. [PUBLICATION ABSTRACT]

In eukaryotes transcriptional regulation often involves multiple long-range elements and is influenced by the genomic environment1. A prime example of this concerns the mouse X-inactivation centre (Xic), which orchestrates the initiation of X-chromosome inactivation (XCI) by controlling the expression of the nonprotein- coding Xist transcript. The extent of Xic sequences required for the proper regulation of Xist remains unknown. Here we use chromosome conformation capture carbon-copy (5C)2 and super-resolution microscopy to analyse the spatial organization of a 4.5-megabases (Mb) region including Xist. We discover a series of discrete 200-kilobase to 1Mb topologically associating domains (TADs), present both before and after cell differentiation and on the active and inactive X. TADs align with, but do not rely on, several domain-wide features of the epigenome, such as H3K27me3 or H3K9me2 blocks and lamina-associated domains.TADs also align with coordinately regulated gene clusters. Disruption of a TAD boundary causes ectopic chromosomal contacts and long-range transcriptional misregulation. The Xist/Tsix sense/antisense unit illustrates how TADs enable the spatial segregation of oppositely regulated chromosomal neighbourhoods, with the respective promoters of Xist and Tsix lying in adjacent TADs, each containing their known positive regulators.Weidentify a novel distal regulatory region of Tsix within its TAD, which produces a long intervening RNA, Linx. In addition to uncovering a new principle ofcis-regulatory architecture ofmammalian chromosomes, our study sets the stage for the full genetic dissection of the X-inactivation centre.

The X-inactivation centre was originally defined by deletions and translocations as a region spanning several megabases3,4, and contains several elements known to affect Xist activity, including its repressive antisense transcript Tsix and its regulators Xite, DXPas34 and Tsx5,6. However, additional control elements must exist, as single-copy transgenes encompassing Xist and up to 460 kb of flanking sequences are unable to recapitulate proper Xist regulation7. To characterize the cisregulatory landscape of the Xic in an unbiased approach, we performed 5C2 across a 4.5-Mb region containing Xist. We designed 5C-Forward and 5C-Reverse oligonucleotides following an alternating scheme2, thereby simultaneously interrogating nearly 250,000 possible chromosomal contacts in parallel, with a mean resolution of 10-20 kb (Fig. 1a; see Supplementary Methods). Analysis of undifferentiated mouse embryonic stem cells (ESCs) revealed that long-range (.50 kb) contacts preferentially occur within a series of discrete genomic blocks, each covering 0.2-1Mb (Fig. 1b). These blocks differ from the higherorder organization recently observed by Hi-C8, corresponding to much larger domains of open or closed chromatin, that come together in the nucleus to form A and B types of compartments8. Instead, our 5C analysis shows self-associating chromosomal domains occurring at the sub-megabase scale. The size and location of these domains is identical in male and female mouse ESCs (Supplementary Fig. 1) and in different mouse ESC lines (Supplementary Fig. 2 and Supplementary Data 1).

To examine this organization with an alternative approach, we performed three-dimensional DNA fluorescent in situ hybridization (FISH) in male mouse ESCs. Nuclear distances were found to be significantly shorter between probes lying in the same 5C domain than in different domains (Fig. 1c, d), and a strong correlation was found between three-dimensional distances and 5C counts (Supplementary Fig. 3a, b). Furthermore, using pools of tiled bacterial artificial chromosome (BAC) probes spanning up to 1Mb and structured illumination microscopy, we found that large DNA segments belonging to the same 5C domain colocalize to a greater extent than DNA segments located in adjacent domains (Fig. 1e), and this throughout the cell cycle (Supplementary Fig. 3c, d). Based on 5C and FISH data, we conclude that chromatin folding at the sub-megabase scale is not random, and partitions this chromosomal region into a succession of topologically associating domains (TADs).

We next investigated what might drive chromatin folding in TADs. We first noticed a striking alignment between TADs and the large blocks of H3K27me3 and H3K9me2 (ref. 9) that are known to exist throughout the mammalian genomes10-13 (for example, TAD E, Fig. 2 and Supplementary Fig. 4). We therefore examined 5C profiles of G9a2/2 (also known as Ehmt2) mouse ESCs, which lack H3K9me2, notably at the Xic14, and Eed2/2 mouse ESCs, which lack H3K27me3 (ref. 15). No obvious change in overall chromatin conformation was observed, andTADs were not affected either in size or position in these mutants (Fig. 2 and Supplementary Fig. 4b). Thus TAD formation is not due to domain-wide H3K27me3 or H3K9me2 enrichment. Instead, such segmental chromatin blocks might actually be delimited by the spatial partitioning of chromosomes into TADs.

We then addressed whether folding in TADs is driven by discrete boundary elements at their borders. 5C was performed in a mouse ESC line carrying a 58-kb deletion (DXTX16), encompassing the boundary between the Xist and Tsix TADs (D and E; Fig. 2b). We observed ectopic contacts between sequences in TADs D and E and an altered organization of TAD E. Boundary elements can thus mediate the spatial segregation of neighbouring chromosomal segments. Within the TAD D-E boundary, a CTCF-binding site was recently implicated in insulating Tsix from remote regulatory influences17. However, alignment ofCTCF- and cohesin-binding sites in mouse ESCs18 with our 5C data showed that, although these factors are present at most TAD boundaries (Supplementary Fig. 4), they are also frequently present within TADs, excluding them as the sole determinants of TAD positioning. Furthermore, the fact that the two neighbouring domains do not merge completely in DXTX cells (Fig. 2b) implies that additional elements, within TADs, can act as relays when a main boundary is removed. The factors underlying an element's capacity to act as a canonical or shadow boundary remain to be investigated.

Next we asked whether TAD organization changes during differentiation or XCI. Both male neuronal progenitors cells (NPCs) and male primary mouse embryonic fibroblasts (MEFs) show similar organization to mouse ESCs, with no obvious change in TAD positioning. However, consistent differences in the internal contacts within TADswere observed (Fig. 3a, Supplementary Figs 2 and 5). Noticeably, some TADs were found to become lamina-associated domains19 (LADs) at certain developmental stages (Fig. 3b). Thus chromosome segmentation into TADs reveals a modular framework where changes in chromatin structure or nuclear positioning can occur in a domainwide fashion during development.

We then assessed TAD organization on the inactive X, by combining Xist RNA FISH, to identify the inactive X, and super-resolution DNA FISH using BAC probe pools on female MEFs. We found that colocalization indices on the inactive X were still higher for sequences belonging to the sameTADthan for neighbouring TADs (Supplementary Fig. 6a). However, the difference was significantly lower for the inactive X than for the active X. Deconvolution of the respective contributions of the active X and inactive X in 5C data from female MEFs (see Supplementary Methods and Supplementary Fig. 6) similarly revealed that global organization in TADs remains on the inactive X, albeit in a much attenuated form, but that specific long-range contacts within TADs are lost. This, together with a recent report focused on longer-range interactions20, suggests that the inactive X has a more random chromosomal organization than its active homologue, even below the megabase scale.

We next investigated how TAD organization relates to gene expression dynamics during early differentiation. A transcriptome analysis, consisting of microarray measurements at 17 time points over the first 84 h of female mouse ESC differentiation was performed (Fig. 4a). During this time window, most genes in the 5C region were either up- or downregulated. Statistical analysis demonstrated that expression profiles of genes with promoters located within the same TAD are correlated (Fig. 4b). This correlation (median correlation coefficient cc of 0.40) is significantly higher than for genes in different domains (cc of 0.03, P,1029) or for genes across the X chromosome in randomly selected, TAD-size regions (cc of 0.09, P,1027). The observed correlations within TADs seem not to depend on distance between genes, and are thus distinct from previously described correlations between neighbouring genes21 that decay on a length scale of approximately 100 kb (Supplementary Fig. 7). Our findings indicate that physical clustering within TADs may be used to coordinate gene expression patterns during development. Furthermore, deletion of the boundary between Xist and Tsix in DXTX cells was accompanied by long-range transcriptional misregulation (Supplementary Fig. 8), underlining the role that chromosome partitioning into TADscan play in long-range transcriptional control.

A more detailed analysis of each domain (Supplementary Fig. 7) revealed that co-expression is particularly pronounced in TADs D, E regulators Jpx, Ftx, Xpr/Xpct and Rnf125 (Jpx, Ftx, Xpct and Rnf12 are also known as Enox, B230206F22Rik, Slc16a2 and Rlim, respectively) is anti-correlated with most other genes in the 4.5Mb region, being upregulated during differentiation (Supplementary Fig. 7). The fact that these coordinately upregulated loci are located in the same TAD suggests that they are integrated into a similar cis-regulatory network, potentially sharing common cis-regulatory elements. We therefore predict that TAD E (,550 kb) represents the minimum 59 regulatory region required for accurate Xist expression, explaining why even the largest transgenes tested so far (covering 150 kb 59 to Xist, Fig. 5a) cannot recapitulate normal Xist expression7.

The respective promoters of Xist and Tsix lie in two neighbouring TADs with transcription crossing the intervening boundary (Fig. 2b), consistent with previous 3C experiments22. Whereas the Xist promoter and its positive regulators are located in TAD E, the promoter of its antisense repressor, Tsix, lies in TAD D, which extends up to Ppnx (also known as 4930519F16Rik)/Nap1l2, more than 200 kb away (Fig. 2b). Thus, in addition to the Xite enhancer, more distant elements within TAD D may participate in Tsix regulation. To test this we used two different single-copy transgenic mouse lines, Tg53 and Tg80 (ref. 23). Both transgenes contain Xist, Tsix and Xite (Fig. 5a). Tg53 encompasses the whole of TAD D, whereas Tg80 is truncated just 59 to Xite (Fig. 5a and Supplementary Fig. 9). In the inner cell mass of male mouse embryos at embryonic day 4.0 (E4.0), Tsix transcripts could be readily detected from Tg53, as well as from the endogenous X (Fig. 5b). However, no Tsix expression could be detected from Tg80, which lacks the distal portion of TAD D (Fig. 5b). Thus, sequences within TAD D must contain essential elements for the correct developmental regulation of Tsix.

Within TADD, several significant looping events involving the Tsix promoter or its enhancer Xite were detected (Figs 2b and 5a, Supplementary Fig. 10). Alignment of 5C maps with chromatin signatures of enhancers in mouse ESCs (Supplementary Fig. 11) suggested the existence of multiple regulatory elements within this region. We also identified a transcript initiating approximately 50 kb upstream of the Ppnx promoter (Fig. 5a), from a region bound by pluripotency factors and corresponding to a predicted promoter for a large (80 kb) intervening non-coding RNA (lincRNA24, Supplementary Fig. 12) which we termed Linx (large intervening transcript in the Xic). Linx RNA shares several features with non-coding RNAs, such as accumulation around its transcription site25 (Fig. 5c), nuclear enrichment and abundance of the unspliced form26 (Supplementary Fig. 12 and 13). Linx and Tsix are co-expressed in the inner cell mass of blastocysts from E3.5-4.0 onwards, as well as in male and female mouse ESCs (Fig. 5c). LinxRNAis not detected earlier in embryogenesis, nor in extra-embryonic lineages, implying an epiblast-specific function (Supplementary Fig. 9). Triple RNA FISH for Linx, Tsix and Xist in differentiating female mouse ESCs (Supplementary Fig. 14) revealed that before Xist upregulation, the probability of Tsix expression from alleles co-expressing Linx is significantly higher than from alleles that do not express Linx (Fig. 5d). Furthermore, Linx expression is frequently monoallelic, even before Xist upregulation (Supplementary Fig. 14), revealing a transcriptional asymmetry of the two Xic alleles beforeXCI. Taken together, our experiments based on 5C, transgenesis and RNA FISH, point towards a role for Linx in the long-range transcriptional regulation of Tsix-either through its chromosomal association with Xite and/or via the RNA it produces. This analysis of the Xist/Tsix region illustrates how spatial compartmentalization of chromosomal neighbourhoods in TADs partitions the Xic into two large regulatory domains, with opposite transcriptional fates (Supplementary Fig. 15).

In conclusion, our study reveals that sub-megabase folding of mammalian chromosomes results in the self-association of large chromosomal neighbourhoods in the three-dimensional space of the nucleus. The stability of such partitioning throughout differentiation, X inactivation and in cell lines with impaired histone-modifying machineries, indicates that this level of chromosomal organization may provide a basic framework onto which other domain-wide features, such as lamina association and blocks of histone modification, can be dynamically overlaid. Our data also point to a role for TADs in shaping regulatory landscapes, by defining the extent of sequences that belong to the same regulatory neighbourhood. We anticipate that TADs may underlie regulatory domains previously proposed on the basis of functional and synteny conservation studies27,28. We believe that the principles we have revealed here will not be restricted to the Xic, as spatial partitioning of chromosomal neighbourhoods occurs throughout the genome of mouse and human29, as well as Drosophila30 and E. coli31. We have shown that TAD boundaries can have a critical role in high-order chromatin folding and proper longrange transcriptional control. Future work will clarify the mechanisms driving this level of chromosomal organization, and to what extent it generally contributes to transcriptional regulation. In summary, our study provides new insights into the cis-regulatory architecture of chromosomes that orchestrates transcriptional dynamics during development, and paves the way to dissecting the constellation of control elements of Xist and its regulators within the Xic.

METHODS SUMMARY

5C was performed on mouse ESCs, mouse NPCs and primary MEFs following a previously described protocol2 with modifications, and sequenced on one lane of an Illumina GAIIx. RNA and DNA FISH were performed on mouse ESCs and inner cell masses extracted from pre-implantation embryos as previously described7, with modifications. Full experimental and bioinformatic methods are detailed in Supplementary Information.

Received 3 October 2011; accepted 22 March 2012.

Published online 11 April 2012.

References

1. Kleinjan, D. A. & Lettice, L. A. Long-range gene control and genetic disease. Adv. Genet. 61, 339-388 (2008).

2. Dostie, J. et al. Chromosome conformation capture carbon copy (5C): a massively parallel solution for mapping interactions between genomic elements. Genome Res. 16, 1299-1309 (2006).

3. Rastan, S. Non-random X-chromosome inactivation in mouse X-autosome translocationembryos-location of the inactivation centre. J. Embryol. Exp. Morphol. 78, 1-22 (1983).

4. Rastan, S. & Robertson, E. J. X-chromosome deletions in embryo-derived (EK) cell lines associated with lack of X-chromosome inactivation. J. Embryol. Exp. Morphol. 90, 379-388 (1985).

5. Augui, S., Nora, E. P. & Heard, E. Regulation of X-chromosome inactivation by the X-inactivation centre. Nature Rev. Genet. 12, 429-442 (2011).

6. Anguera, M. C. et al. Tsx produces a long noncodingRNA and has general functions in the germline, stem cells, and brain. PLoS Genet. 7, e1002248 (2011).

7. Heard, E., Mongelard, F., Arnaud, D. & Avner, P. Xist yeast artificial chromosome transgenes function as X-inactivation centers only in multicopy arrays and not as single copies. Mol. Cell. Biol. 19, 3156-3156 (1999).

8. Lieberman-Aiden, E. et al. Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326, 289-293 (2009).

9. Marks, H. et al. High-resolution analysis of epigenetic changes associated with X inactivation. Genome Res. 19, 1361-1373 (2009).

10. Pauler, F. M. et al. H3K27me3 forms BLOCs over silent genes and intergenic regions and specifies a histone banding pattern on a mouse autosomal chromosome. Genome Res. 19, 221-233 (2009).

11. Wen, B.,Wu, H., Shinkai, Y., Irizarry, R. A. & Feinberg, A. P. Large histone H3 lysine 9 dimethylated chromatin blocks distinguish differentiated from embryonic stem cells. Nature Genet. 41, 246-250 (2009).

12. Lienert, F. et al. Genomic prevalence of heterochromatic H3K9me2 and transcription do not discriminate pluripotent from terminally differentiated cells. PLoS Genet. 7, e1002090 (2011).

13. Hawkins, R. D. et al. Distinct epigenomic landscapes of pluripotent and lineagecommitted human cells. Cell Stem Cell 6, 479-491 (2010).

14. Rougeulle, C. et al. Differential histoneH3Lys-9 and Lys-27methylation profileson the X chromosome. Mol. Cell. Biol. 24, 5475-5484 (2004).

15. Montgomery, N. D. et al. The murine polycomb group protein Eed is required for global histone H3 lysine-27 methylation. Curr. Biol. 15, 942-947 (2005).

16. Monkhorst, K., Jonkers, I., Rentmeester, E., Grosveld, F. & Gribnau, J. X Inactivation counting and choice is a stochastic process: evidence for involvement of an X-linked activator. Cell 132, 410-421 (2008).

17. Spencer, R. J. et al. A boundary element between Tsix and Xist binds the chromatin insulator Ctcf and contributes to initiation of X chromosome inactivation. Genetics CrossRef (2011).

18. Kagey, M. H. et al. Mediator and cohesin connect gene expression and chromatin architecture. Nature 467, 430-435 (2010).

19. Peric-Hupkes, D. et al. Molecular maps of the reorganization of genome-nuclear lamina interactions during differentiation. Mol. Cell 38, 603-613 (2010).

20. Splinter, E. et al. The inactive X chromosome adopts a unique three-dimensional conformation that is dependent on Xist RNA. Genes Dev. 25, 1371-1383 (2011).

21. Caron, H. et al. The human transcriptome map: clustering of highly expressed genes in chromosomal domains. Science 291, 1289-1292 (2001).

22. Tsai, C.-L.,Rowntree, R. K.,Cohen, D. E.&Lee, J. T. Higher orderchromatin structure at the X-inactivation center via looping DNA. Dev. Biol. 319, 416-425 (2008).

23. Heard, E. et al. Transgenic mice carrying an Xist-containing YAC. Hum. Mol. Genet. 5, 441-450 (1996).

24. Guttman, M. et al. Chromatin signature reveals over a thousand highly conserved large non-coding RNAs in mammals. Nature 458, 223-227 (2009).

25. Khalil, A. M. et al. Many human large intergenic noncoding RNAs associate with chromatin-modifying complexes and affect gene expression. Proc. Natl Acad. Sci. USA 106, 11667-11672 (2009).

26. Seidl, C. I. M., Stricker, S. H. & Barlow, D. P. The imprinted Air ncRNA is an atypical RNAPII transcript that evades splicing and escapes nuclear export. EMBO J. 25, 3565-3575 (2006).

27. Ruf, S. et al. Large-scale analysis of the regulatory architecture of the mouse genome with a transposon-associated sensor. Nature Genet. 43, 379-386 (2011).

28. Kikuta, H. et al. Genomicregulatory blocks encompassmultiple neighboring genes and maintain conserved synteny in vertebrates. Genome Res. 17, 545-555 (2007).

29. Dixon, J. R. et al. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature doi:10.1038/nature11082 (this issue).

30. Sexton, T. et al. Three-dimensional folding and functional organization principles of the Drosophila genome. Cell 148, 458-472 (2012).

31. Mercier, R. et al.TheMatP/matS site-specific systemorganizes the terminus region of the E. coli chromosome into a macrodomain. Cell 135, 475-485 (2008).