Rett Syndrome Symposium and Workshop

October 28 and 29, 2019

Prepared by Dr. Sameer Bajikar, postdoctoral fellow in the lab of Dr. Huda Zoghbi and co-chair of the young investigator workshop.

Rett syndrome is a debilitating neurological disorder, affecting approximately 1 in 10,000 girls. These girls experience developmental regression, repetitive movement, loss of speech, motor difficulties, breathing abnormalities, and seizures. In 1999, Dr. Huda Zoghbi and colleagues discovered that mutations in the gene encoding methyl CpG binding protein 2 (MECP2) cause Rett syndrome. To commemorate the 20^th anniversary of this discovery, Rettsyndrome.org (formerly the International Rett Syndrome Foundation) and Rett Syndrome Research Trust supported a special, two-day symposium and workshop on Rett syndrome biology that was hosted at the Jan and Dan Duncan Neurological Research Institute (Duncan NRI) at Texas Children’s Hospital. The meeting was co-organized by Dr. Adrian Bird, Buchanan Professor of Genetics and Wellcome Centre for Cell Biology at the University of Edinburgh, UK, and the discoverer of MeCP2 as a methyl CpG binding protein, and Dr. Zoghbi, Founding Director, Duncan NRI, Ralph D. Feigin Professor at Baylor College of Medicine, and Howard Hughes Medical Institute Investigator. The purpose of the meeting was to evaluate and synthesize the knowledge we have gained into the molecular function of MeCP2 and pathogenesis of Rett syndrome, to identify the key barriers in understanding and treating the disorder, and to develop strategies to overcome these barriers.

To achieve this goal, the symposium brought together world-class academics from a broad spectrum of fields, including physicians, scientists, trainees, patient advocates, and leaders from the pharmaceutical industry and the National Institutes of Health (NIH) (see List of Attendees). Altogether, over 300 people convened in the brand-new auditorium within the Duncan NRI in the last week of October, providing a compelling end to Rett Syndrome Awareness Month.

This symposium was structured into three scientific sessions: 1) Molecular function of MeCP2 and DNA methylation; 2) Pathogenesis and neuronal and circuit alterations; and 3) Therapeutics. In each session, eminent scientists shared both published and unpublished work (see Meeting Agenda). These diverse research presentations laid the foundation for the open discussions at the end of each session, as well as the organic discussions occurring throughout the symposium.

CURRENT AREAS OF BASIC EXPLORATION

The predominant molecular function of MeCP2 is to bind methylated cytosines. Further work has elucidated several other roles, like recruitment of repressor proteins and global gene regulation. Now, we must understand how MeCP2 works in different contexts. The basic science research talks centered around studying how loss of MeCP2 function disrupts the normal biology in several contexts:

• Genomic contexts: the gradation of severity observed in Rett patients raises the possibility that some patients may have additional mutations that worsen or improve the outlook of the disorder. By introducing random mutations in Rett mice, it was discovered that mutations in several components of the same pathway improved overall survival. This work not only revealed the importance of a new pathway that could be targeted therapeutically, but also demonstrated that other genes tangibly affect the Rett phenotype (Dr. Monica Justice).
• Molecular contexts: the human brain is composed of a variety of cells with varying methylation patterns (Drs. Joseph Ecker and Margarita Behrens), and these local methylation contexts critically determine where the MeCP2 protein resides. Though methylation establishes the pattern of where MeCP2 can bind, MeCP2 proteins are not static within the nucleus of the cell. MeCP2 proteins are in constant movement, both in and out of contact with DNA, and the diffusion pattern of MeCP2 is governed through its methyl-DNA binding domain, the NCoR/SMRT interaction domain and AT-hook domains. (Dr. Nathaniel Heintz). Last, we know MeCP2 mildly changes the expression of thousands of genes. Additionally, MeCP2 was also shown to affect the translation of many genes. For example. the translation of several genes in the ubiquitin pathway is reduced in Rett syndrome neurons derived from induced pluripotent cells (iPSCs). The reduced ubiquitination leads to increased levels of target proteins. Interestingly, none of these changes in protein levels are predicted by RNA changes (Dr. James Ellis).
• Cellular and circuit contexts: Rett brains are comprised of cells with normal MeCP2 and cells with mutant MeCP2, and this heterogeneity occurs across the various cell types in the brain. We are now starting to understand how mutant MeCP2 cells communicate with wild-type MeCP2 cells, and this communication has demonstrated that defective MeCP2 causes a signature of expression changes that are unique within the mutant cells (Dr. Zhaolan Zhou). Additionally, by testing the function of MeCP2 in non-neuronal cells, it was shown that mutant MeCP2 expression in astrocytes causes calcium misregulation, which in turn disrupts the activity of neuronal networks. Importantly, correction of calcium misregulation in MeCP2-deficient astrocytes could partially alleviate those neuronal defects (Dr. Qiang Chang). Lastly, neuronal activity leads to gene expression changes, which were altered in MeCP2 mutant neurons. These differences may partially contribute to neuronal and circuit-level dysfunction (Dr. Nurit Ballas).

Taken together, these studies highlight the need to dig deeper into MeCP2 biology in various DNA and cellular contexts, to study the consequences of its loss in various cell types, and to identify suppressors of such consequences. 

IMMEDIATE DIRECTIONS IN TREATMENT STRATEGIES

We know that MeCP2-related disorders are reversible in mouse models, giving hope that these disorders are treatable in humans. The next set of research talks discussed current proposed treatment strategies:

• Gene therapy: using viral approaches to express MeCP2 is a promising therapeutic strategy given that it substitutes for the mutant protein itself. The field has continued to iterate upon the approach to demonstrate that there is a tight window of therapeutic benefit with gene therapy. Importantly, introducing wild-type MeCP2 in the context of mutant MeCP2 was not shown to be detrimental (Dr. Stuart Cobb). Currently, further optimization of the viral delivery methods is being performed, as well as tweaks to the viral payload. For example, instead of introducing the entire MeCP2 gene, researchers are testing if introducing a smaller piece of the MeCP2 protein that will replace the mutant MeCP2 protein will revert the disease (Dr. Rebekah Tillotson, lab of Dr. Adrian Bird). However, others questioned the advantages of delivering this “mini-gene” over delivering the full-length MeCP2 gene. Further work is needed to directly compare the efficacy of treatment with the “mini-gene” and full-length gene. Others are testing if we can engineer smaller pieces of normal MECP2 RNA that will by-pass the mutated exons and allow for the production of normal MeCP2 protein. This approach is appealing because it will reduce the chances of producing too much MeCP2 protein since it simply replaces the mutated piece of the MECP2 RNA, which is produced at the correct levels.  

• Nucleic acid editing: CRISPR/Cas9 technology allows for precision genome editing. This approach could allow for mutations in MECP2 to be corrected. One challenge to this approach is that the therapy would need to be customized for each patient. New approaches now can insert larger fragments, which would allow for a single construct and therapy to repair entire mutated domains for multiple mutations in different patients (Dr. Feng Zhang). This approach would require effective gene therapy approaches to deliver the Cas9 and guide RNAs. Additionally, new approaches have allowed for RNA molecules to be edited. Editing MECP2 RNA is advantageous because it would not require a cut or permanent change to the genome. One approach uses CRISPR principles to recruit a protein called adenosine deaminase (ADAR) to a particular RNA molecule. This natural ADAR protein then makes precise edits to the targeted mRNA molecule. With this modality, the mutant MECP2 RNA can be corrected to produce healthy MECP2 protein (Dr. Phillip Reautschnig, lab of Dr. Thorsten Stafforst).
• Activation of the silenced wild-type allele: females with Rett syndrome carry a copy of healthy MeCP2 in all cells, though it is not produced in every cell due to X-chromosome inactivation. Efforts have been made to reactivate this silent copy. One approach is to use CRISPR/Cas9 to bring an enzyme to the MECP2 gene to remove silencing methyl marks. This approach was able to reactivate the healthy MeCP2 copy, but only transiently. These data suggest reactivation of the silenced locus is possible, but further optimization and investigation will be required to allow for sustained local X-reactivation and translational utility (Dr. Rudolf Jaenisch). 
• Pharmacologic treatments: many mutations in MECP2 cause the production of a shortened protein (truncating mutation). To treat this specific class of mutations, one can use drugs to allow the protein production machinery to continue translation beyond the stop codon to produce the missing part of the protein. With these drugs, a small amount of full-length MeCP2 was successfully produced in a mouse model that carried one such mutation; future work is needed to test if the recovery of some full-length MeCP2 is enough to ameliorate disease features (Dr. Jeffrey Neul).
• Non-pharmacologic therapy: Rett girls have a period of normal development, followed by regression. If we knew a child had Rett syndrome before the regression, could we intervene to improve disease outlook? To test this hypothesis, presymptomatic Rett mice were persistently trained on either a specific motor or learning task where we know the symptomatic Rett mice fail. Encouragingly, early training significantly improved the performance of the Rett mice for that specific task (Nathan Achilly, lab of Dr. Huda Zoghbi). These data demonstrate the need to screen for MECP2 mutations as early as possible so that any child identified with a mutation can receive early intervention to delay disease onset and progression. 

FUTURE DIRECTIONS AND BARRIERS TO THERAPY

After the research sessions were completed, workshop participants had open discussions over dinner to identify the key questions that are progress. The participants were divided into three groups: 1) Molecular Function – this group was comprised of primarily academic leaders tasked with defining the next key questions needed to understand MeCP2 function; 2) Therapeutic Strategies – this group was comprised of both academic and industry leaders tasked with identifying the remaining barriers to new treatments and translation of current preliminary treatments; and 3) Young Investigator Workshop – this group was comprised entirely of trainees tasked with asking longer-term and high-risk questions. The following morning, discussion leaders from each session summarized and presented the essence of the discussions.

Much remains unclear about the molecular pathogenesis of Rett syndrome, but several experiments could help shed critical light on this issue:

• Molecular interactome: MeCP2 has been shown to interact with several nuclear proteins, most notably the NCoR/SMRT complex. However, it is not known whether the constituents of this complex are identical across genomic loci and brain regions. Breaking this complex down further and understanding the contribution of this “black box” interacting component will further our understanding of how MeCP2 regulates gene expression. Additionally, identifying the components within the NCoR/SMRT complex may identify other candidate genes that cause neurologic diseases.
• Probing the proteome: the transcriptome in MECP2/Mecp2 mutant mice and cells has been measured in great detail. How these transcriptional changes propagate to the proteome have been understudied. Dissecting the changes to the proteome, which may be far more dramatic than the transcriptome (Dr. James Ellis), will help us understand the biology most proximal to cellular behavior. 
• Cell-type specificity: while MeCP2 is broadly expressed throughout the human and mouse brains, we do not have a comprehensive, cell-level map of its expression. These data would identify if certain cell-types express MeCP2 or if this expression changes during development. Next, this map would help drive which cell-populations critically require wild-type MeCP2 function, which can be tested with detailed conditional deletion in mice. These studies could also reveal which brain regions are most important to target therapeutically. 
• Influence of mutation on function: Mecp2 deletion models are the workhorse of the field. However, several other mouse models of human Rett-causing MECP2 mutations have also been generated. Using these mouse models for basic science and preclinical work will help preserve the critical molecular contexts that may have been underappreciated in a deletion model (see Current Areas of Basic Research). Additionally, using these point mutation models, especially female mice, will help increase the clinical translatability of future findings.

Furthermore, several recurrent barriers to clinical translation were raised from all of the discussion groups:

• Generalization of therapies to all Rett patients: the efficacy of potential treatments tends to be tested in Mecp2 deletion contexts. However, the majority of Rett-causing mutations still generate a MeCP2 protein. How therapeutic modalities, like gene therapy, behave with different mutations must be explored in greater detail. For example, is the dosing for a patient with a mutation that decreases DNA binding (e.g., T158M) the same as a patient with a mutation that still binds DNA but disrupts separate MeCP2 function (e.g., R306C)? We must strongly incorporate “personalized medicine” concepts for treating Rett syndrome, and these considerations will be important when designing a clinical trial.
• Improvement of viral delivery methods: many therapeutic strategies under investigation rely on viruses to deliver a payload. The efficacy of these therapies is limited by the ability of the virus to reach the intended cells. Further investigation is needed to identify newer and optimal viral strains that broadly, yet safely, target the correct cells in the brain.
• Large animal models: the mouse has been the workhorse for devising and testing preclinical strategies to treat MeCP2-related disorders. While mouse models have been invaluable, they cannot model the size of the human brain. This means all preclinical work in the mouse will not be able to account for issues of treatment distribution and efficacy that must be addressed for a clinical trial. Biodistribution issues are particularly important for gene therapy. Additionally, orthogonal large animal models will increase rigor and reproducibility of findings generated in mice. The central point is that larger animal models (for example, macaques) must be developed and used for preclinical work. 
• Biomarkers: clinical trials will be enabled by identifying biomarkers that are sensitive to MeCP2 levels. By measuring this proxy marker, we can identify the proper dosage of any potential therapy. These biomarkers may come from biofluids (blood, CSF), but also may be non-invasive, such as electroencephalography.
• Clinical trial readiness: as potential treatments appear on the horizon, it will be critical to properly define outcome measures. Rett syndrome is a nuanced disease, and we need to collectively think about the most relevant and sensitive outcome measures that will give a clinical trial the greatest likelihood of success.

During the closing remarks, Dr. Bird noted that the discovery of MECP2 as the causal gene for Rett syndrome provided a glimmer of hope for the patients with Rett syndrome. Since then, the field has advanced rapidly, and we have tangible paths forward for treatments and potential cures for Rett. Taken together, Dr. Zoghbi’s lasting charge of the meeting was to push the research to address the gaps so that a viable treatment strategy will be in place by the celebration of the 25^th anniversary of the discovery of MECP2 as the cause of Rett syndrome.

LIST OF ATTENDEES:

Organizers:

Adrian Bird, PhD
University of Edinburgh

Huda Y. Zoghbi, MD
Baylor College of Medicine / Texas Children’s Hospital

SPEAKERS, WORKSHOP PARTICIPANTS, AND SPECIAL GUESTS:

YOUNG INVESTIGATORS:

DOWNLOAD AGENDA