Imaging the binding of MECP2 to DNA

In this Outlook, West discusses a study in this issue of Genes & Development by Zhou et al., putting into context the consequences of a novel, loss-of-function mutation in methyl-CpG binding protein MECP2 with those driven by other MECP2 mutations causative of Rett syndrome. West further expands on the existing methods to study the functional and molecular effects of MECP2 mutations and their application in the screening of potential therapeutic strategies.

Abstract

Mutations in the methyl-DNA binding domain of MECP2 cause Rett syndrome; however, distinct mutations are associated with different severity of the disease. Live-cell imaging and single-molecule tracking are sensitive methods to quantify the DNA binding affinity and diffusion dynamics of nuclear proteins. In this issue of Genes & Development, Zhou and colleagues (pp. 883–900) used these imaging methods to quantitatively describe the partial loss of DNA binding resulting from a novel pathological MECP2 mutation with intermediate disease severity. These data demonstrate how single-molecule tracking can advance understanding of the molecular mechanisms connecting MECP2 mutations with Rett syndrome pathophysiology.

Loss-of-function mutations in the methyl-DNA binding protein MECP2 cause Rett syndrome (RTT), a progressive and severe neurodevelopmental disorder (Amir et al. 1999). Two decades of research on RTT-associated MECP2 mutations have established that impaired binding of MECP2 to methylated regions of genomic DNA is key to the cellular pathophysiology of RTT. Some mutations in the methyl-DNA binding domain of MECP2 cause more severe disease phenotypes than others; however, standard biochemical assays lack the resolution to assess the differential effects of these mutations on genomic DNA binding. To overcome this limitation, in a new study, Zhou et al. (2023) showed that they can use live, single-molecule imaging to quantify the partial loss of DNA binding activity associated with a novel MECP2 mutation.

Mutations in MECP2 mostly arise de novo, and hundreds of different loss-of-function mutations have been associated with RTT, including a very large number of missense mutations scattered across the full coding sequence of MECP2 (Krishnaraj et al. 2017). Interpreting genotype–phenotype correlations of this complex genetic landscape is challenging because MEPC2 is an X-chromosome gene and RTT most commonly occurs in females. This means the severity of the disorder is influenced by the percent and distribution of cells with random inactivation of the X chromosome bearing the MECP2 mutation. However, in a small number of cases, pathological mutations in MECP2 have been identified in males, in whom the relationship between the mutation, MECP2 function, and disease severity can be more easily assessed (Neul et al. 2019).

Zhou et al. (2023) found a previously unreported G118E mutation in the methyl-binding domain of MECP2 in a boy with neurodevelopmental delay and motor impairment. The investigators developed both iPSC-derived neurons and a knock-in mouse model to determine the consequences of this mutation on both the MECP2 protein and physiological function. Male mice bearing the G118E mutation showed RTT-like features including motor impairments, microcephaly, and impaired learning and memory, as well as impaired survival compared with Mecp2 wild-type mice. Importantly, comparing survival across an allelic series of strains of mice with RTT-causing Mecp2 missense mutations (Collins and Neul 2022) offers a strategy to predict the severity of new mutations. These data suggest that G118E may cause partial disruption of the DNA binding domain because these mice live longer than those expressing the severe loss-of-function mutations T158M and R111G; about the same as mice expressing the R306C mutation, which is outside the DNA binding domain; and somewhat less long than mice with the R133C mutation, which is known to be only a partial loss of function.

DNA binding of chromatin factors is often studied using biochemical methods such as chromatin immunoprecipitation (ChIP) followed by sequencing. These methods have the advantage of showing protein binding across the entire genome, giving insight into possible target genes as well as clues to mechanisms of action. However, ChIP protocols have been optimized to capture even weak protein–DNA interactions, making them poorly suited to estimate the consequences of modulatory DNA binding domain mutations, including RTT-associated MECP2 mutations (Schmiedeberg et al. 2009).

A more quantitative way to measure a protein's DNA binding affinity is to use live-cell imaging to capture molecular dynamics (Dahal et al. 2023). The simple idea is that what a protein binds determines how it diffuses. Within the nucleus, a protein that tightly binds DNA, such as the nucleosomal histone H2B, will diffuse slowly, whereas a protein that lacks DNA binding will diffuse quickly. To test whether this method can detect MECP2 binding to DNA, a previous group generated a mouse strain in which they knocked a Halo tag into the endogenous Mecp2 locus (Piccolo et al. 2019). A protein bearing this tag can be fluorescently labeled in live cells by the addition of a cell-permeable, dye-conjugated Halo ligand. As expected, Piccolo et al. (2019) found that MECP2 diffused more slowly than the free Halo tag but faster than H2B. They also showed that introducing the severe R106W DNA binding domain mutation into MECP2 substantially enhanced the rate of diffusion, whereas the mild R133C mutation had no effect.

Given this evidence that live-cell imaging can distinguish very mild from very severe MECP2 binding mutations, Zhou et al. (2023) asked whether these methods were sensitive enough to detect a partial loss of DNA binding when the G118E mutation was knocked into the Halo-tagged Mecp2 allele. Both wild-type and G118E MECP2 were enriched in heterochromatic chromocenters, but following bleaching at these sites, the Halo signal returned significantly faster in G118E neurons compared with wild-type neurons, suggesting that the G118E MECP2 molecules exchange more freely than wild-type with MECP2 elsewhere in the nucleus. When the investigators used photoactivatable dyes to tag single molecules of MECP2 for live tracking, they found that G118E led to a significant increase in the jump length between frames and a significant decrease in the bound fraction of molecules, indicating that the mutation causes the mutant protein to be both less bound and more mobile than wild-type MECP2.

In addition to resolving subtle differences in DNA binding between MECP2 mutants, the investigators showed that their live-imaging analysis is able to assess DNA binding of MECP2 independent of any effects these mutations have on protein expression levels. This is important because several of the RTT-associated mutations in MECP2, including G118E, result in substantially reduced protein expression. Rescuing the expression levels of MECP2 has been suggested as a possible therapy in cases in which missense or nonsense mutations are associated with reduced protein levels (Lamonica et al. 2017). Zhou et al. (2023) point out that expression above wild-type levels may be required to rescue DNA binding. In addition, in the future, the live-cell imaging platform developed by Zhou et al. (2023) and Piccolo et al. (2019) may be useful as a platform for screening therapeutic strategies that directly modulate MECP2 DNA binding.