Co-opted transposons help perpetuate conserved higher-order chromosomal structures

Abstract

Transposable elements (TEs) make up half of mammalian genomes and shape genome regulation by harboring binding sites for regulatory factors. These include architectural proteins—such as CTCF, RAD21 and SMC3—that are involved in tethering chromatin loops and marking domain boundaries. The 3D organization of the mammalian genome is intimately linked to its function and is remarkably conserved. However, the mechanisms by which these structural intricacies emerge and evolve have not been thoroughly probed. Here we show that TEs contribute extensively to both the formation of species-specific loops in humans and mice via deposition of novel anchoring motifs, as well as to the maintenance of conserved loops across both species via CTCF binding site turnover. The latter function demonstrates the ability of TEs to contribute to genome plasticity and reinforce conserved genome architecture as redundant loop anchors. Deleting such candidate TEs in human cells leads to a collapse of such conserved loop and domain structures. These TEs are also marked by reduced DNA methylation and bear mutational signatures of hypomethylation through evolutionary time. TEs have long been considered a source of genetic innovation; by examining their contribution to genome topology, we show that TEs can contribute to regulatory plasticity by inducing redundancy and potentiating genetic drift locally while conserving genome architecture globally, revealing a paradigm for defining regulatory conservation in the noncoding genome beyond classic sequence-level conservation.One-sentence summaryCo-option of transposable elements maintains conserved 3D genome structures via CTCF binding site turnover in human and mouse.

CTCF-dependent chromatin boundaries formed by asymmetric nucleosome arrays with decreased linker length

“…We showed here that these regions may be linked to transcription of transposons such as Pol III-dependent SINE repeats. Several publications suggested important roles of transposons in the evolution of CTCF sites (57)(58)(59)(60)(61), and also it is known that mouse SINE B2 repeats can act as insulators (domain boundaries) per se (62). In addition, our data suggests that CTCF may play active role in transposon functioning as transcribed units separating nucleosome arrays.…”

Comparing 3D genome organization in multiple species using Phylo-HMRF

“…To fully understand 3D genome organization evolution, it will be crucial to explore the underlying mechanisms of the different evolutionary patterns in 3D genome structure across species, e.g., in concert with the evolution of particular types of DNA sequence features that play key roles in the formation and maintenance of genome architecture and function (Sima et al, 2018;Choudhary et al, 2018;Zhang et al, 2019), which may in turn inform us about the principles of 3D genome organization. For example, we previously showed that more conserved CTCF motifs in mammalian evolution (considering motif turnover) are more likely to be involved in CTCF mediated chromatin loops .…”

“…We use the Expectation-Maximization (EM) algorithm (Dempster et al, 1977;Zhang et al, 2001) for parameter estimation in our model. Zhang et al (2001) developed the HMRF-EM algorithm where EM is adapted to estimate a HMRF model with several justified assumptions and approximations, including the pseudo-likelihood assumption (Geman and Graffigne, 1986) and mean-field approximation (Celeux et al, 2003;Zhang, 1992).…”

Clustered CTCF binding is an evolutionary mechanism to maintain topologically associating domains

“…These clusters suggest a mechanism by which local turnover events can largely preserve TAD structure and function. Indeed, a recent study has demonstrated CTCF binding site turnover at loop anchors mediated by TEs, and it suggested that this is a common mechanism of 10 contributing to conserved genome folding events between human and mouse (Choudhary et al, 2018). Based on these observations, we conclude that the formation of CTCF binding site clusters serves as an additional evolutionary buffering mechanism to preserve the CTCF binding potential of TAD boundaries and ensure resilience of higher order chromatin structure by maintaining a dynamic redundancy of CTCF binding sites.…”