Scientific Meetings:
     Symposium Summaries

June 27-29, 2005
Itasca, IL

View photos of the Symposium.

On June 27-29, 2005, RSRF hosted the 6th Annual Rett Syndrome Symposium, the premiere scientific meeting devoted to this disorder. Scientists from around the world gathered to exchange the latest information and discuss future research directions.

In the first talk of the symposium, Daniel Glaze from Baylor College of Medicine introduced the incidence and clinical features of Rett Syndrome (RTT) to an audience of researchers from diverse disciplines. Dr. Glaze outlined five major challenges for clinical and basic research on RTT: to seek to understand, regression, seizures, autonomic dysfunction, motor dysfunction, and growth failure. As a highlight of his presentation, several local girls with RTT and their families were introduced to the audience, allowing basic researchers to obtain important personal context of the disorder.

Dr. Richard Goodman of the Vollum Institute at Oregon Health Sciences University discussed a novel genome-wide screening technique for identifying transcription factor binding sites. Dr. Goodman's research has focused on finding genes regulated by CREB (a transcription factor involved in learning and memory), which has been implicated in the pathogenesis of a different mental retardation disorder termed the Rubinstein-Taybi Syndrome. Dr. Goodman's laboratory discovered that CREB regulates a large number of brain-specific small regulatory RNA molecules encoded by micro RNA genes. Interestingly, expression of one of these microRNA genes promoted the formation of neurites (the processes of neurons in cell culture). In an interesting potential application of his research to RTT, Dr. Goodman showed that MECP2 may also be regulated by CREB-induced microRNAs.

Art Beaudet of Baylor College of Medicine presented autism as a model for understanding how both genetic and epigenetic defects can contribute to a complex disease. Dr. Beaudet has coined the abbreviation "MEGDI" that stands for "mixed epigenetic and genetic and mixed de novo and inherited" for a model to explain the difficulty in uncovering the genetic basis of most cases of autism. Parental imprinting (the phenomenon in which there is differential expression of a gene depending on whether it was maternally or paternally inherited) of human chromosome 15q11-13 and its effect on the related neurodevelopmental disorders Prader-Willi and Angelman Syndromes was explained. Maternal 15q11-13 duplications are observed in rare cases of autism, while other cases are hypothesized to have a mixture of genetic and epigenetic defects in 15q11-13. In support of this model, several autism brain samples exhibited abnormally low UBE3A, similar to Angelman Syndrome.

Reiko Fitzsimonds of Yale University and Huda Zoghbi of Baylor College of Medicine both presented data from their respective laboratories on alterations in the function of connections between neurons (synapses) in the Mecp2 knockout mice (Fitzsimonds used the Jaenisch and Bird mice strains; Zoghbi used the Mecp2308/y mice). The hippocampus is a region of the brain where excitatory neuronal communication, and use-dependent changes in the strength of synaptic communication ("synaptic plasticity) have been very well studied due to their likely involvement in certain forms of learning and memory. Both the Fitzsimonds and Zoghbi groups found that two long-lasting forms of synaptic plasticity, known as long-term potentiation (LTP) and long-term depression (LTD) are abnormal in symptomatic Mecp2-deficient mice. Subtle but significant changes in short-term plasticity suggested that the ability of presynaptic neurons to release neurotransmitter in response to action potentials might be impaired in symptomatic Mecp2-deficient mice. In mice overexpressing Mecp2, the Zoghbi lab found the opposite: that PPF (a form of short-term plasticity) and LTP were in fact enhanced. The Fitzsimonds group report that the underlying molecular mechanism for these impairments in LTP and LTD may be due to age-dependent abnormalities in the expression of a class of neurotransmitter receptors, the NMDA receptors, known to be critical for the induction of LTP and LTD in this region of the hippocampus. The Zoghbi lab also examined several tests for locomotor activity and spatial learning (which are thought to involve hippocampal LTP/D mechanisms) in the Mecp2-deficient and Mecp2-overexpressing mice. Interestingly, both groups did not detect alterations in many synaptic markers, including pre-, postsynaptic and cytoskeletal. Furthermore, the Zoghbi group did not detect significant changes in neuronal morphology. These findings raise the important question of whether the earlier reports from human post-mortem tissues were reflective of compensatory or secondary changes following years of neurodevelopmental arrest. The ability to perform in-depth morphological, biochemical and electrophysiological analyses of the Mecp2-deficient mouse model of RTT, and correlate changes at the molecular and cellular level with age-dependent symptomatic progression is of central importance to understanding RTT.

Ulrike Nuber of the Max Planck Institute for Molecular Genetics compared transcriptomes of Mecp2-null and normal mouse brain using a cDNA microarray approach. Sgk and Fkbp5 genes were identified to be up-regulated in mutant mice and confirmed by real time PCR and Northern blot analysis. These genes are known to belong to the HPA stress pathway and can be induced by glucocorticoids. Circulating glucocorticoids were measured in Mecp2-null mouse blood and no significant deviation from wild-type animals was observed, proposing that up-regulation of Sgk and Fkbp5 expression is not due to the hormone over-production. Chromatin IP showed the presence of MeCP2 in the vicinity of Sgk1 and Fkbp5 promoters suggesting that these are likely to be the primary targets of MeCP2.

Dr. Bruce Cohen of the Cleveland Clinic is a clinician interested in mitochondrial diseases. Patients diagnosed with mitochondrial dysfunction share many striking similarities with RTT patients-these similarities are of interest for (1) exploring the possibilities of genetic relatedness of mitochondrial dysfunction and RTT (e.g., does Mecp2-deficiency lead to mitochondrial dysfunction?), (2) asking whether mitochondrial dysfunction plays a role in the major pathophysiologies of RTT, and if so (3) exploring useful therapeutic approaches used in the treatment of mitochondrial dysfunction in RTT patients. Dr. Cohen presented a case study of a patient clinically presenting with suspected mitochondrial disorders who was ultimately diagnosed to not only have biochemically confirmed mitochondrial dysfunction, but ultimately diagnosed also as having MECP2 mutation and classical RTT. To date, three patients of Dr. Cohen with clinically diagnosed mitochondrial disease also have confirmed MECP2 mutations. Dr. Cohen is interested in further exploring the possible relationship between mitochondrial dysfunction and RTT.

Dr. Peter Scheiffele of Columbia University presented data from his work on the neuroligin-neurexin adhesion complex which has been shown to play an important role in the formation of synaptic connections between neurons. Neuroligins are proteins that are concentrated specifically at the synaptic junctions where neuronal communication occurs. Mutations in neuroligin isoforms have been linked to autism spectrum disorders and mental retardation disorders. How neuroligin-neurexin complex formation and the cellular signals that the interaction induces are involved in the normal assembly and ultimately the function of excitatory and inhibitory synaptic connections will hopefully shed light on how loss of MeCP2 might lead to abnormalities in synapse and brain development, and more importantly on the delicate balance of excitation and inhibition that occurs at various synapses within complex neural circuits.

John Greer of the University of Alberta described the neuroanatomical and physiological basis of breathing rhythm generation, with particular emphasis on rodent models. A small area of the ventrolateral medulla, known as the pre-Bötzinger complex, has been clearly established as the source of the breathing rhythm. Dr. Greer discussed what is known about the development of the preBötC in rodents and how studying respiratory neural control can elucidate the pathophysiology of mutant mouse models of diseases with respiratory dysfunction.

Jean-Christophe Roux in collaboration between the INSERM and CNRS in Marseille described respiratory pathology in the Mecp2 knockout mouse. The animals show an increase in the number of apnea episodes as development progresses, and breathing failure is a common, if not uniform, cause of death in the animals. Dr. Roux showed data that mild anesthesia normalizes breathing of Mecp2 knockout mice. To begin understanding the mechanistic basis of this dysfunction and its alleviation. By using biochemical tools, their data showed lower levels of catecholamine and serotonin contents specifically in the medulla of Mecp2 knockout mice. Moreover, Dr. Roux and colleagues described a significant decrease in catecholaminergic neurons in the same area, which can explain in part the decrease in bioaminergic contents.

Nathaniel Heintz of Rockefeller University described his work aimed at understanding the transcriptional basis of specific cell type function in the central nervous system (CNS). Dr. Heintz pointed out that out of the ~30,000 genes in the human genome, up to 20,000 are expressed in the CNS. Importantly, not every gene is active in every cell. As part of the GENSAT project (www.gensat.org), Dr. Heintz has developed a system which allows identification of various neuronal subtypes as well as access to functionally uniform cell types -- an important goal, given the extensive functional heterogeneity in cells found in the mammalian CNS. The system involves modification of BAC clones either to study the neuronal distribution of specific genes or to isolate RNAs from a select neuronal population "BACarray". Application of this technology has yielded two fundamental advances: first, an accurate spatial expression profile for over 2,000 genes in the CNS is now available; second, it is now possible to profile gene expression in specific cell types in the CNS. Application of this approach to Purkinje cells in the cerebellum yielded the surprising observation that a very considerable number of genes are expressed exclusively in this cell type. With RSRF funding, Dr. Heintz, in collaboration with Dr. Zoghbi, are using the BACarray technology to identify neuron-specific changes in gene expression in the Mecp2 mutant mice in the hopes of identifying pathways that may be targeted pharmacologically.

Dr. Zhaolan (Joe) Zhou of the Children's Hospital of Boston/Harvard Medical School presented his study in understanding how BDNF transcription is regulated by neuronal activity-dependent phosphorylation of MeCP2. He first used a combination of immunoprecipitation and mass spectrometric analysis to identify four potential phosphorylation sites in the MeCP2 protein. He then used site-directed mutagenesis and phosphorylation-/phosphorylation site-specific antibodies to confirm that one of the four sites was indeed phosphorylated upon neural activity. Furthermore, he demonstrated that mutation at this site affected BDNF transcription. Finally, he showed an approach to block the MeCP2 expression from its endogenous locus while at the same time express MeCP2 from an exogenous viral vector. This strategy could potentially be used to silence the endogenous mutant MeCP2 and supplement wild type MeCP2 in Mecp2 deficient animals.

Dr. Qiang Chang of the Whitehead Institute discussed his work in defining the functional role of BDNF in modulating disease progression in Mecp2 mutant mice. He showed that there was less BDNF protein in the Mecp2 mutant brain, and that the Bdnf conditional knockout mice exhibited similar symptoms as the Mecp2 mutant mice. More interestingly, he demonstrated that the manifestation of some of the defects in the Mecp2 mutant mice were sensitive to the level of BDNF. Specifically, BDNF overexpression in the Mecp2 mutant brain expanded the lifespan, reversed a locomotor deficit, and rescued an electrophysiological defect in these mice. In contrast, BDNF deletion in the Mecp2 mutant brain caused an earlier onset of RTT in these mice. These results suggest a functional interaction between MeCP2 and BDNF, and indicate BDNF/BDNF signaling as potential therapeutic targets for treating RTT.

Dr. Helen Scharfman from the Helen Hayes Hospital/Columbia University talked about the various effects of acute and chronic BDNF overexpression in rats and mice and their implications for the etiology of RTT. She focused on the morphological and physiological characterization of BDNF overexpression in a transgenic mouse line. She stressed that it is important to define when, where, how much BDNF level is changed, and whether compensatory changes occur in TrkB receptors before one can accurately interpret the relationship between BDNF and RTT. In addition, she pointed out that BDNF expression can be regulated by fluctuations in reproductive steroids, and there may be implications for such a regulation in RTT girls.

Weidong Wang of the National Institute of Aging (NIH) questioned the recent data in the literature regarding whether MeCP2 is in a stable complex with Brahma, a protein involved in regulating chromatin architecture. The presented data clearly indicated that MeCP2 is unlikely to associate stably with Brahma, because Brahma does not co-purify with MeCP2, and the gel-filtration peaks of MeCP2 and Brahma do not overlap. In the second part of the talk Weidong showed purification of the phosphorylated MeCP2 form and identification of phosphorylated sites. A specific antibody was raised recognizing the phosphorylated serine 80 in MeCP2, and the presence of the modified MeCP2 form was confirmed in brains of mouse, rat and human. In vitro phosphorylation mimic of serine 80 had no affect on MeCP2 association with methylated DNA, however, serine 80 to alanine mutation reduced the induction of BDNF (brain-derived neurotrophic factor) in depolarized neurons. The data suggest that dynamic modification of Ser80 within MeCP2 plays critical roles in the gene regulation function of this protein.

During brain development, neurons become fully functional when they are correctly integrated into neural networks, thereby establishing proper connections (synapses) with other neurons, a process known as neuronal maturation. Maturation of neurons involves complex controls at both cellular and molecular levels. One of the critical steps in neuronal maturation is synaptogenesis, which can be evaluated by expression of synapse-specific proteins and by the density and shape of synaptic spines (tiny protrusions on neuronal dendrites that are responsible for communication with other neurons). Several recent publications suggest that the MeCP2 protein is involved in neuronal maturation. Drs. Pavel V. Belichenko (Stanford University) and Xinyu Zhao (University of New Mexico) presented data that further confirmed that mutations in MeCP2 significantly alter neuronal maturation. Dr. Belichenko's findings clearly indicated that, in Mecp2 mutant mice brains, neurons display severe morphological alterations like reduced dendritic spine density, abnormal spine distribution and reduced spine size. These dysfunctional neurons also displayed more dendritic swellings, a phenomenon that may be associated with neuronal toxicity. Dr. Zhao's group found that Mecp2 was not necessary for the generation of new neurons in adult mice brains. However, the synaptic distribution was abnormal in both young and adult Mecp2 mutant mice. Using a retrovirus to label dividing hippocampal stem cells and consequently follow their differentiation and maturation in postnatal brains, preliminary work from Dr. Zhao's group indicated that neurons generated in the postnatal hippocampus of Mecp2 mutant mice also had reduced synaptic spine density as compared to wild-type mice.

Dr. Gregory Pelka of Sydney University presented another Mecp2 mutant mouse line that exhibited similar phenotypes as the existing three Mecp2 mice models. In addition, these mutant mice not only had motor deficits, but also displayed learning and emotional deficits, and interestingly, reduced anxiety. In addition, Pelka et al generated female Mecp2-deficient mice that had only one mutant X chromosome (39XO) and demonstrated that these mice had similar deficits, onset of illness and lifespan as male mutant mice (40XY), suggesting that the Y chromosome does not contribute to deficits in boys with RTT. Using gene array analyses, Pelka et al found that Mecp2 mutant brains exhibited abnormal expression levels of several neuronal genes. For example, mutant brains had up regulation of Sgk, a kinase that plays important roles in neuronal excitation; and down regulation of Gap43, a key molecule involved in axon growth and targeting. In addition, Pelka et al found that the level of these expression changes correlated with the severity of neurological deficits of mice, such that "well" (presymptomatic) male mutant mice showed abnormal expression of these genes when compared with normal male littermates, and "sick" (symptomatic) male mutant mice showed even more severe alterations in gene expression. Pelka et al also generated mutant mice that expressed a reporter gene under the endogenous Mecp2 promoter. These mice will be very useful in analyzing the expression patterns of mutant Mecp2 in the brain, and will allow him to accurately map specific behavioral abnormalities in the mutant mouse to the regions in the brain where normal Mecp2 expression has been disrupted. The results from Pelka et al further support the model that MeCP2 is involved in regulating neuronal maturation, but its deficiency is likely affecting mammalian development even before birth.




RSRF SPONSORS 5TH ANNUAL RETT SYNDROME SYMPOSIUM

View photos of the Symposium.

For the 5th consecutive year RSRF gathered scientists from around the world for its annual Rett Syndrome Symposium. For 3 days at the end of June 94 scientists met in Baltimore, MD to share their latest data, form collaborations and set research directions for the coming year.

The Rett Syndrome Symposium is unique in that it brings together clinicians and scientists from very diverse fields. This convergence of expertise allows for an energizing flow of ideas and ñout of the boxî thinking.


We are indebted to Huda Zoghbi and Adrian Bird, co-chairs, for providing the participants with a stimulating program and for spearheading the informative discussion sessions.

In the first talk Jeffrey Neul of Baylor College of Medicine presented a comprehensive description of the clinical features of Rett Syndrome. The videos of children with RTT embedded in his presentation were very helpful to the audience comprised mostly of basic scientists with limited clinical knowledge of RTT.

Michael Greenberg of ChildrenÍs Hospital Boston and Harvard Medical School discussed the importance of the interplay between certain enzymes and neuronal activity in the healthy development of the nervous system. He focused on the formation of synapses and how they are maintained and/or changed by neural activity. Of particular relevance to RTT is the possibility that stimuli may influence the release of MeCP2 from brain derived neurotrophic factor (BDNF) thereby allowing its production. Dr. GreenbergÍs talk provided some insights on how synapse development may be misregulated in Rett Syndrome.

Yi Sun of UCLA discussed potential roles of DNA methylation and methyl-DNA binding proteins in neuronal function and plasticity in early embryonic development in mouse. She touched on several transcription factors and one signaling pathway, and described how DNA methylation at these loci would affect the outcome of early neuron/glia differentiation.

Lisa Monteggia of University of Texas Southwestern Medical Center presented behavioral analysis of Mecp2 conditional mutant mice including locomotion test, open field test, fear conditioning, pain sensitivity, and social interaction test. She showed that these mice have RTT-like symptoms and can be used as an animal model to study the behavior abnormalities.

Michael G. Rosenfeld of University of California San Diego and Howard Hughes Medical Institute presented work using a variety of state-of-the-art methodologies aimed at elucidating global gene expression patterns. This work will allow an understanding of normal developmental events as well as gain insight into neurological diseases, including RTT by identifying direct MeCP2 binding targets in the genome.

Skirmantas Kriaucionis of the University of Edinburgh used a variant of differential display to find a dozen genes which are mis-regulated in Mecp2-null mice brain. Some of these genes are mis-expressed early in symptom progression and may contribute to neurological defects in this mouse model of Rett Syndrome. Future work will investigate this issue further and attempt to link the findings in mouse to the human condition.

David Amaral of the University of California Davis and The M.I.N.D. Institute spoke about a specific brain region, the amygdala, and its involvement in autism. Through MRI, Dr. Amaral demonstrated an abnormal growth trajectory of the amydala in boys with autism, with an initial enlargement followed by a premature cessation of growth. To examine the functional role of the amygdala, Dr. Amaral has used primates in whom the amygdala has been selectively disrupted. These animals, while exhibiting substantial alterations in emotional behavior, do not show dramatic change in social behavior. Amygdala abnormalities in autism, therefore, likely contribute to the abnormal fears and anxieties seen in the majority of children with autism, but not to the social impairments.

Features of Rett syndrome also overlap with those of Angelman syndrome (AS), caused by loss of the UBE3A gene, and Laura Herzing of Northwestern University has found that expression of UBE3A is also decreased in brain tissue from girls with RTT. To further investigate how MeCP2 mediates UBE3A expression, she is examining gene expression patterns in the Mecp2-knockout mouse model. Dr. Herzing has found that in regions of the brain where Ube3a expression most closely matches that seen in humans, animals lacking Mecp2 also show a decrease in Ube3a expression. However, this change occurs over a period of months during and after the development of symptoms, rather than during the weeks prior to mouse adulthood, thus correlating with the actual time-period of regression in RTT instead of with developmental (weaning) stage.

Luca Santarelli, of Columbia University, uses mouse models, pharmacology and gene replacement to study anxiety and depression. He showed that knocking out the serotonin 1A receptor in mice results in increased anxiety-like behaviors. Using a tissue-specific conditional rescue strategy, he was able to provide evidence that such an anxious phenotype is mediated by the absence of the 1A receptor during the early postnatal period, but not during adulthood. These changes in behavior are paralleled by morphological and physiological changes in the hippocampus, where the serotonin 1A receptor is prominently expressed. These findings underscore the importance of the early postnatal period in establishing the normal function of limbic structures that underlie higher brain functions such as mood and anxiety responses.

Gail Mandel of Stonybrook and Howard Hughes Institute is investigating the repressor REST and its function in neuronal development. REST was shown to bind RE1 elements and recruit transcriptional co-repressors coREST and Sin3A. During ES cell differentiation REST is down-regulated in a post-transcriptional way by digestion with proteases. REST represses calbindin and synaptogonin genes, however the treatment with de-methylating agent (aza-C) and deacetylase inhibitor (TSA) relieved transcriptional repression. Methylated regions were identified near the calbindin promoter, which is bound by MeCP2. The proposed model suggests that MeCP2 and REST interact by a common co-repressor, Sin3A, and the presence of MeCP2 is necessary to maintain repression when REST is absent. SACO (Serial Analysis of Chromatin Occupancy) was introduced as a global method to identify protein binding sites of REST. Surprisingly, for some of the genes the RE1 site is not in close proximity to the promoter, suggesting that there may be long-range repression events.

Gabriele Ronnett of Johns Hopkins discussed using the olfactory system to learn about Rett. She found that the balance between immature and mature neurons is abnormal in RTT patients. In the knockout mice, Dr. Ronnett observed that, as in the human tissue, the balance between immature and mature neurons was abnormal, with many more immature neurons in the MeCP2-null mice. Experiments by Dr. Ronnett point to the conclusion that this imbalance is a consequence of the failure of neurons to mature (as opposed to the failure of mature neuronal survival). This is consistent with the prevalent hypothesis that RTT is a disorder of developmental arrest, and not neurodegeneration. Analysis off the olfactory system of Mecp2 null revealed that neuronal maturation was transiently blocked at the time of peak axon outgrowth and synaptogenesis. To investigate the possible molecular bases of these findings, proteomic analysis was performed on the olfactory epithelium and olfactory bulbs of wild type and null mice. Mecp2 null mice displayed abnormal expression of proteins that function in cytoskeletal arrangement, neuroprotection, chromatin modeling, energy metabolism, and cell signaling.

Denis Jugloff of the University of Toronto over-expressed MeCP2 in neurons along with GFP and found that neurons had longer axons and dendrites and were overall more complex than control neurons. No change in the short-term survival of the neurons was observed. This data supports the hypothesis that MeCP2 plays a role in the growth and branching pattern of neurons as they develop. It corroborates the striking finding that in human RTT brains dendritic arborizations are dramatically simplified.

Nino Ramirez of the University of Chicago studies the function of neural networks that are important for respiration. He gave 3 examples of neuromodulators that are found to be abnormal in RTT patients: norepinephrine, serotonin and dopamine. Dr. Ramirez and colleagues found that brain slices prepared from MeCP2-mull mice had abnormal patterns of activity, consistent with behavioral findings of irregular breathing patterns in the mice. Moreover, when neuromodulators such as norepinephrine, serotonin and substance P were applied to the null slices, the expected (normal) patterns of activity were expressed. This is an important observation that suggests that the basic neural network wiring of the null mice are capable of generating normal rhythmic firing patterns. This points to an exciting hypothesis, albeit preliminary, that irregular breathing patterns observed in RTT patients may be a consequence of an abnormal neuromodulatory milieu.

Mutations in MeCP2 have now been identified in the majority of girls diagnosed with RTT, along with some patients with atypical RTT, Angelman Syndrome or other features. However, too many girls remain for whom no cause has been identified. Dr. Eva Gak has found that patients with C-terminal deletions of MeCP2 have lower gene expression levels. At the Danek Gertner Institute of Human Genetics at the Sheba Medical Center, she has developed a screen using peripheral blood to identify individuals with lower MeCP2 levels, and has found that 3 of 10 patients with no identified MeCP2 coding mutations have decreased MeCP2 expression. Dr. Gak has identified a single nucleotide change in the non-coding end of MeCP2 in one of these patients, and suggests that the C-terminus of MeCP2 may be an important factor for RNA stability or transcription rate.

Carl Wu of NIH, is one of the foremost experts in the world on chromatin. MeCP2 is believed to control chromatin structure, and therefore RTT may be a disorder in which some poorly understood aspect of chromatin biology goes awry. Dr. Wu described recent important discoveries from his lab on how chromatin structure is changed during gene regulation by highly intricate molecular machines known as "chromatin remodeling ATPases." He drew the attention of the audience to possible effects of MeCP2 on dynamic changes in chromatin folding and organization. This is an insufficiently investigated subject, and careful study of this problem may provide insights on how to restore that process to normalcy in children with RTT.

Megumi Adachi of the Neurosciences Institute described her work on how the MeCP2 gene itself is controlled. Adachi discovered a mechanism that, in her words, provide a "link between physiological stimuli (such as stress and neurotrophic factors) and MeCP2 production by the brain." Further study of this mechanism may help develop strategies to overcome the lack of MeCP2 function in RTT.