MeCP2 nuclear dynamics in live neurons results from low and high affinity chromatin interactions

Abstract

Loss of function mutations in Methyl-CpGbinding Protein 2 (MeCP2) cause the severe neurological disorder Rett Syndrome. MeCP2 is a highly abundant nuclear protein particularly enriched in neurons. Although biochemical and genomic analyses of MeCP2-DNA interaction and genomic distribution demonstrate that MeCP2 binding on chromatin is dependent on DNA modification state, the dynamic behavior of individual MeCP2 proteins in live neurons has not been explored. Here we use live-cell singlemolecule imaging to assess the detailed kinetic features of MeCP2 in distinct sub-nuclear regions at high spatial and temporal resolution. Surprisingly, we found that, in granule cell nuclei, MeCP2 has unique diffusion and chromatin binding kinetics that are distinct from highly mobile sequence-specific transcription factors (TF) and immobile histone proteins. Approximately, half of MeCP2 is bound to DNA in a transiently stable mode that is similar to TF binding to their cognate sites. The binding of meCP2 to DNA requires its methyl-binding domain (MBD) and is sensitive to the levels of both DNA methylation and hydroxymethylation. However, when not stably bound, MeCP2 moves slowly in the nucleus most closely resembling histone H1.0. The rate of MeCP2 diffusion in compact, granule cell nuclei is determined by weak, transient DNA interactions mediated primarily by the MBD and three AT-hook domains located in the C-terminal portion of the protein. Both the fraction of stably bound MeCP2 and its rate of diffusion depend on the level of chromatin compaction and neuronal cell type.Our data reveal new features of MeCP2 that dictate its dynamic behavior in neuronal nuclei and suggest that the limited nuclear diffusion of MeCP2 in live neurons may contribute to its local impact on chromatin structure and gene expression.

Buffering of transcription rate by mRNA half-life is a conserved feature of Rett syndrome models

“…At the DNA level, our classifier model discovered that the direction of transcription rate shifts in human and mouse RTT models are best predicted by combinations of three dinucleotides that include the canonical MECP2-binding sites CA and CG, together with other dinucleotides including AT. The MECP2 AT-hook domain contributes to low-affinity transient interactions with AT-rich DNA that influence the dynamics of MECP2 binding to local high affinity methylated DNA sites 30 . Tellingly, the classifier predictions are low when modelling only CA/CG dinucleotide frequencies without accounting for low-affinity sites, and the accurate predictions are unaffected when low-affinity sites are included but the CA/CG dinucleotide frequencies are omitted.…”

Epigenetic Factors that Control Pericentric Heterochromatin Organization in Mammals

“…It was recently demonstrated that in neurons the fraction of MeCP2 that is stably bound to chromatin is higher in comparison with the distribution of other transcription factors [ 113 ]. MeCP2 behavior in this context depends on both the integrity of its MBD and the DNA methylation [ 113 ].…”

“…It was recently demonstrated that in neurons the fraction of MeCP2 that is stably bound to chromatin is higher in comparison with the distribution of other transcription factors [ 113 ]. MeCP2 behavior in this context depends on both the integrity of its MBD and the DNA methylation [ 113 ]. Indeed, several mutations in the MBD of MeCP2 that can cause Rett syndrome, such as R106W, result in reduced binding of MeCP2 to methylated DNA [ 112 , 114 ], and an increased rate of MeCP2 diffusion in the nucleus [ 113 ].…”

“…MeCP2 behavior in this context depends on both the integrity of its MBD and the DNA methylation [ 113 ]. Indeed, several mutations in the MBD of MeCP2 that can cause Rett syndrome, such as R106W, result in reduced binding of MeCP2 to methylated DNA [ 112 , 114 ], and an increased rate of MeCP2 diffusion in the nucleus [ 113 ]. MeCP2 has also a transcriptional repression domain (TRD) and a C-terminal region.…”

Single-molecule tracking (SMT) and localization of SRF and MRTF transcription factors during neuronal stimulation and differentiation

“…SMT has been particularly successful in the nucleus to identify dynamic properties of chromatin-associated molecules such as transcription factors [2,3]. So far, several parameters of TF nuclear kinetics have been determined, including exact DNA binding times (ranging from a few seconds to 1–2 min), the TF fraction bound to DNA and localizations of transcriptional ‘hot-spots’ [4–11]. Previous studies have analysed the impact of cell stimulation by, for example, growth factor, hormone administration or DNA damage on TF-DNA interaction, and demonstrated that DNA occupancy and bound fraction of TFs such as p53, SRF, GR, ER, CREB and SOX2 are enhanced by cell activation [4–10].…”