Calico and tortoiseshell cats can be identified by the mosaic of orange and black patches appearing on their spotty fur coats. This blotchy color pattern is a visible manifestation of a genetic process known as X chromosome inactivation. In mammals, males receive one copy of the X chromosome while females receive two copies. The X chromosome contains more than 1000 genes that are essential for proper development and cell viability, so having 2 copies could lead to a toxic double dose of proteins produced by X-linked genes (1).
To correct this imbalance, mammalian females have evolved a unique mechanism of dosage compensation. One copy of the X chromosome in each female cell is transcriptionally silenced early in embryonic development, resulting in an inactivated X chromosome that condenses into a compact structure and remains inactive in all of the descendant cells. Inactivation is random: One X chromosome is turned off in some areas of the body while the other X chromosome is inactivated in other areas. In cats, fur pigmentation is X-linked, and coat color depends on which copy of the X chromosome is active in each cell.
The molecular mechanisms underlying X chromosome inactivation are still not completely understood, and a recent study published in Proceedings of the National Academy of Sciences (PNAS) now casts doubt on a widely accepted view that two molecules—X inactive specific transcript (Xist) RNA and polycomb repressive complex 2 (PRC2)—interact during X chromosome inactivation (2).
“A number of prior studies have correlated co-localization of these two molecules and proposed a direct interaction, but it is difficult to prove the two molecules are found at the same place in the genome at the same time,” said Eric Mendenhall, an expert in non-coding regions of the genome at the University of Alabama in Huntsville and the HudsonAlpha Institute for Biotechnology who was not involved in the new study.
The initiation and spread of X inactivation rely on the non-protein coding XIST gene, which is expressed on the inactive X chromosome. Xist RNA coats the chromosome and directly recruits chromatin modifying factors to inactivate it. But the molecular mechanisms linking Xist RNA to the silencing machinery are not completely clear.
In particular, studies of PRC2 recruitment to the inactive X chromosome have shown a close overlap with sites of Xist RNA localization. It has also been reported that PRC2 recruitment occurs at the same time as the onset of X chromosome inactivation and depends on continuous Xist RNA transcription, and that PRC2 proteins and Xist RNA interact directly. But the recent PNAS study calls this interaction into question.
In the new work, researchers analyzed Xist-mediated recruitment of PRC2 using microarray-based epigenomic mapping. Using a mouse embryonic stem cell line carrying an inducible Xist transgene, they showed that one day after synchronous induction of Xist expression, PRC2 binding sites mapped predominantly to gene-rich regions, but these new sites of PRC2 deposition did not correlate with Xist-mediated gene silencing. According to the authors, this observation argues against a role for PRC2 in the establishment of gene silencing in X chromosome inactivation.
To reexamine the relationship between Xist RNA and PRC2 recruitment, the researchers next applied superresolution 3D structured illumination microscopy (3D-SIM), along with immunofluorescence detection and immuno-RNA fluorescence in situ hybridization (FISH) to visualize protein chromosome interactions. The data showed a very clear spatial separation between Xist RNA and PRC2 proteins, estimated to be between 50 nm and 100 nm.
“The findings are very important as they can trigger a reexamination of the way Xist recruits PRC2,” said lead study author Andrea Cerase of the European Molecular Biology Laboratory. “The question for future research is still how Xist recruits these factors and how in general it exerts its function.”
Prior studies of Xist RNA-mediated recruitment of PRC2 used chromatin immunoprecipitation (ChIP) sequencing. “Genome-wide chromatin mapping in combination with next-generation sequencing (ChIP-seq) are very powerful methods to assess the occupancy of chromatin binding factors,” said study author Lothar Schermelleh, a cellular imaging expert at the University of Oxford. But these methods rely on pooling large cell populations and therefore only give an average of binding probabilities, he explained. “Some questions cannot be addressed with population-based approaches, such as the spatial relationship of two given factors.”
In contrast with conventional fluorescence light microscopy, which does not provide the spatial resolution needed to answer these questions, 3D-SIM allows multicolor 3D optical sectioning of whole cells with an eight-fold increase in volumetric resolution. “SIM is a superresolution technique that enables visual separation on the order of 100 nanometers, whereas conventional microscopy has a resolution of 200 to 500 nanometers at best,” explained Jeannie Lee, an expert in X chromosome inactivation at Harvard Medical School and Massachusetts General Hospital who was not involved in the new study. “The authors have chosen a good approach to studying an important question—the question of how Xist RNA relates structurally to the rest of the X chromosome and its known epitopes.”
The use of 3D microscopy to study the distance between a protein and RNA is novel, according to Mendenhall. “It allows for a view of individual molecules in individual cells, as opposed to looking at millions of molecules in millions of cells and inferring correlations and co-localizations. The epigenomics field is certainly moving towards a single-molecule direction and looking at protein or RNA localization in a single cell, and this paper describes a nice technique for doing that.”
But the technique is not without limitations. “The biological application of superresolution 3D-SIM imaging proves a challenging task, in particular when it comes to structures deep inside mammalian cell nuclei,” Schermelleh said. “This is mainly due to the contrast being compromised by light scattering and out-of-focus blur, as well as spherical and sample-induced aberrations.” These technical concerns impose significantly higher demands on sample preparation, system calibration, and quality control, he said. Indeed, the authors noted in the paper that they could not entirely rule out the potential influence of cell fixation on the observed spatial separation between Xist RNA and PRC2 proteins.
By the same token, Lee was surprised that the authors found a similarly large spatial separation (about 100 nm) between two different PRC2 subunits, SUZ12 and EED, in a control experiment. “This is very strange because EED and SUZ12 are directly interacting subunits of the polycomb complex and should be separated by only a few Angstroms to a few nanometers, depending on which epitope is examined,” she said.
Experts agree that future studies should attempt to confirm that Xist RNA and PRC2 do not directly interact. Although the authors cannot account for the discrepancy between their results and past findings, they suggest that sensitive biochemical assays could detect a low level of Xist RNA/PRC2 binding, which may be physiologically relevant or may reflect nonspecific binding of PRC2 proteins to RNA. They also speculate that PRC2 recruitment might occur as an indirect consequence of changes in chromatin configuration associated with Xist RNA–mediated silencing. The answers might take some time though. “The Xist and PRC2 story seems to get more complex with each paper published,” Mendenhall said.
1. Ahn, J. & Lee, J. (2008). X chromosome: X inactivation. Nature Education 1(1):24
2. Cerase A, Smeets D, Tang YA, Gdula M, Kraus F, Spivakov M, Moindrot B, Leleu M, Tattermusch A, Demmerle J, Nesterova TB, Green C, Otte AP, Schermelleh L, and Brockdorff N. (2014). Spatial separation of Xist RNA and polycomb proteins revealed by superresolution microscopy. Proc Natl Acad Sci U S A 111(6):2235-40. doi: 10.1073/pnas.1312951111. Epub 2014 Jan 27.