The chromatin state is one of the main determinants of transcription rate in eukaryotes. The chromatin is described as a fiber made up of an array of nucleosomes which consist of core histone proteins wrapped around by DNA double helix (1). Dynamic chromatin movements and interactions play a crucial role in gene regulation. Adjacent genes can be packed together in distinct chromatin domains that ensure coordinated gene expression. Evidence is emerging that distant genes located on different chromosomes also can interact, and their proximity might be essential for coordinated regulation. However, it has been unclear whether these interchromosomal associations are exceptional, or occur frequently.

We systematically analyzed genome-wide interchromosomal interactions in the nuclei of human cells. 3D data from (2) were associated with the results of several high-throughput studies of the chromatin functional state (3). All pairs of regions from different chromosomes were divided into groups according to their proximity, and the distribution of various chromatin marks was calculated within these groups and then compared between the groups.

The results show that, indeed, gene regions that are spatially close tend to have similar patterns of histone modifications, methylation state, open or closed chromatin state, and expression level. Spatially close genome domains tend to have similar chromatin state and to be coregulated and coexpressed.

Moreover, we found that interacting domains may produce transcripts composed of sequence segments coming from two different chromosomes. We analyzed chimeric transcripts as determined by genome mapping of paired-read RNA-Seq data (4,5) and observed that the frequency of pairs mapping to two different genome loci is higher among spatially proximal regions. We suggest that these transcripts might be formed by trans-splicing.

In summary, interchromosomal associations seem to be much more common than previously believed, and they likely play important roles in the regulation of gene expression.

This research has been supported by State contracts, grants of the Russian Foundation of Basic Research, programs ”Molecular and Cellular Biology” and ”Basic Science for Medicine” of the Russian Academy of Sciences.

References

1. A. Travers, E. Hiriart, M. Churcher, M. Caserta, and E. Di Mauro. J Biomol Struct Dyn 27, 713-724 (2010).
2. E. Lieberman-Aiden, N. L. van Berkum, L. Williams, M. Imakaev, T. Ragoczy, et al. Science 326, 289–293 (2009).
3. ENCODE Project Consortium Science 306, 636–640 (2004).
4. A. G. Xu, L. He, Z. Li, Y. Xu, M. Li, et al. PLoS Comp. Biol., 6(7), e1000843 (2010).
5. M.F. Berger, J.Z. Levin, K. Vijayendran, A. Sivachenko, X. Adiconis, et al. Genome Res. 20, 413–427 (2010).