BSMN data is generated with funding through the following grants. See grant abstracts below

Project PI Institution Grant Title Grant Number
1 Christopher Walsh, Peter Park Boston Children’s Hospital, Harvard Medical School 1/2-Somatic mosaicism and autism spectrum disorder U01MH106883
1 Nenad Sestan Yale University 2/2-Somatic mosaicism and autism spectrum disorder U01MH106874
2 Daniel Weinberger Lieber Institute for Brain Development 1/3-Schizophrenia Genetics and Brain Somatic Mosaicism U01MH106893
2 John Moran University of Michigan 2/3-Schizophrenia Genetics and Brain Somatic Mosaicism U01MH106892
2 Fred Gage Salk Institute for Biological Studies 3/3-Schizophrenia Genetics and Brain Somatic Mosaicism U01MH106882
3 Flora Vaccarino, Alexej Abyzov, Alexander Urban Yale University, Mayo Clinic, Stanford University Somatic Mosaicism in the Brain of Tourette Syndrome U01MH106876
4 Joseph Gleeson Rockefeller University Mosaicism in focal cortical dysplasias spectrum seen in neuropsychiatric disease U01MH108898
5 Andrew Chess, Schahram Akbarian, Christopher Walsh Icahn School of Medicine at Mount Sinai, Boston Children’s Hospital Somatic Mosaicism in Schizophrenia and Control Brains U01MH106891
6 Jonathan Pevsner Kennedy Krieger Institute Role of brain somatic mosaicism in autism, schizophrenia, and bipolar disorder U01MH106884


Grant Abstracts

Project_1: Somatic mosaicism and autism spectrum disorder – U01MH106883U01MH106874

Somatic mutations are de novo mutations that occur after fertilization. Once a cell has acquired a somatic mutation, all of its progenitors will also carry that mutation. Thus, if a cell acquires a mutation early in embryonic development, the mutation will be carried by many of the cells in the body. However, if the mutation occurs late in development, then only a few cells might carry it. Thus, it is possible to have mutations that only occur in the brain, or a small region of the brain. It has been known for a while that somatic mutations can cause cancer, and recent studies are showing that somatic mutations are associated with neurodevelopmental disorders resembling autism spectrum disorders (ASDs) both in terms of their high de novo mutation rate and in terms of their associated symptoms such as intellectual disability and epilepsy. We hypothesize that somatic mutations represent a significant cause of (ASDs) because of the high rate of de novo mutations associated with ASDs, the importance of somatic mutations in some genes known to cause ASDs, and the importance of somatic mutations in other developmental brain disorders with features that overlap ASDs. The technical and resource limitations that had prevented a systematic study of the role of somatic mutations in ASDs have now been overcome thanks to 1] Next-Generation Sequencing (NGS), which allows for the deep sequencing of genes and their transcripts with the ability to analyze each sequence, and 2] tissue banks that have collected brain specimens from individuals who had ASD. In this collaborative UO1 we will employ complementary approaches to systematically identify and functionally characterize somatic brain mutations associated with ASD. For causative somatic mutations identified in ASD brain, we will use techniques developed in our labs to examine individual brain cells for the presence of somatic mutation. This will provide us with a map of what regions of the brain, and what cells types in the brain carry these somatic mutations. We will also model and functionally characterize ASD- associated brain mutations in induced pluripotent cells and mice. This study could 1] improve the genetic diagnosis of ASD; by assessing the prevalence of somatic mutations as a cause of ASD, 2] provide a paradigm that may apply to other complex neuropsychiatric diseases (such as schizophrenia), and 3] improve our understanding of the mechanisms underlying ASD by creating a map of brain regions and cell types involved in ASD.

Project_2: Schizophrenia Genetics and Brain Somatic Mosaicism – U01MH106893U01MH106892U01MH106882

Schizophrenia (SCZD) is a debilitating and typically incurable neuropsychiatric disease that affects 1% of the human population. Disease symptoms, which include hallucinations, paranoia, and impaired cognition, are thought to arise from impairments in neuronal connectivity and plasticity, but etiology of these defects remains unclear. Multiple lines of evidence suggest a strong genetic component to SCZD. Thus, identifying genetic variants associated with SCZD may provide critical tools for understanding and treating the disease. Indeed, recent genome wide association studies have identified >100 loci that are associated with SCZD, but these genetic variants account for only a small percentage of disease incidence. One potential explanation for this unsatisfying result is that SCZD risk alleles are not inherited through the germline, but instead arise through somatic mutations within neurons of affected individuals. Perhaps it is the propensity for somatic mosaicism that is inherited in patients with SCZD. It is now clear that somatic mosaicism of DNA sequence is much more common than previously thought (i.e., all cells within an individual do not contain the same genome), and that this phenomenon is particularly prevalent in the brain. These genomic differences may contribute to the diversity of neuronal function. However, dysregulation of processes that generate or control somatic mosaicism may lead to disease-related genomic instability. Our hypothesis, therefore, is that somatic mosaicism in neurons or their progenitors are a major contributor to SCZD pathogenesis. Aim 1 will use single-cell genomic sequencing techniques to identify somatic copy number variants (CNVs) in neuronal and non-neuronal cell types from patients with SCZD or neurotypic controls. These analyses will focus on the frontal cortex and hippocampus, two brain regions associated with SCZD pathogenesis. Results will determine whether somatic CNVs are overrepresented in SCZD brains, and whether SCZD risk alleles are disproportionately affected by these CNVs. Aim 2 will characterize somatic retrotransposon insertions within these same cell types, asking whether the frequency or location of retrotransposition events is altered in neurons from patients with SCZD compared with controls. A total of 8000 neurons will be analyzed in Aims 1 and 2, making this the most comprehensive analysis of neuronal somatic mosaicism to date. In Aim 3, genomic variants most overrepresented in patients with SCZD (identified in Aims 1 and 2) will be engineered into hESCs for functional validation tests. It has been shown that cultured neurons derived from patients with SCZD exhibit reduced levels of connectivity and have underdeveloped neurites compared with controls. Similar analyses will be performed using isogenic and mosaic cultures of neurons derived from engineered hESCs. Results from these studies will determine whether the level, pattern, or type of somatic mosaicism is altered in SCZD neurons, and potentially identify genes and gene networks most affected by these changes. Identifying causal disease factors will provide new therapeutic targets and move us closer to finding a cure for this devastating disease.

Project_3: Somatic Mosaicism in the Brain of Tourette Syndrome – U01MH106876

Tourette Syndrome (TS) is a disorder of the developing telencephalon for which no significant causative genetic variant has yet emerged through the examination of blood samples. In this proposal we investigate whether somatic mutations might underlie in part the pathogenesis of TS. Existing evidence suggests that cells accumulate somatic mutations after the formation of the zygote, implying that cells of the human body do not have identical DNA sequence. Besides single nucleotide variation (SNV) and small insertion/deletions (InDels), cells can accumulate copy number variations (CNVs, i.e., duplications and deletions), insertions of transposable elements, inversions and translocations, all involving from few hundred to several millions of nucleotides. Somatic mosaicism arising in brain cells could explain the failure to discover consistent, replicable genetic risk factors in neuropsychiatric disorders like TS, and underlie at least in part the frequently observed variability between blood genotype and overall phenotype. There is no estimate of somatic mosaicism in either normal development or in disease. To test the hypothesis that somatic mutations might underlie the emergence of TS, in this proposal we will discover and quantify somatic genome variation in TS and normal control brains, followed by exploration of potential functional consequences of this variation. In Aim 1, we will perform using advanced sequencing techniques comprehensive discovery of lineage-specific and region-specific somatic genomic variations: SNVs, InDels, CNVs, retrotransposon insertions, inversion and translocations. The analysis will involve 20 TS brains and matched 20 normal control brains. Mosaic variants will be discovered and validated in prefrontal cortex (PFC), premotor cortex (PMC) and striatum (STR), three regions strongly implicated in TS, as well as in specific cell lineages isolated from these regions, including pyramidal neurons, medium spiny neurons, interneurons and microglial cells. In Aim 2, we will select 10 genomic variants, engineer them into iPSCs and in transgenic mice using CRISPR technologies, and characterize their impact on the molecular, tissue and behavior level. Together, these specific aims will provide the first estimate of somatic genomic variation (number, type, frequency) in the brain of TS and will yield hypotheses about their significance for brain development.

Project_4: Mosaicism in focal cortical dysplasias spectrum seen in neuropsychiatric disease – U01MH108898

The recent discovery that de novo zygotic mutations in the form of CNVs and point mutations make major contributions to neuropsychiatric diseases such as schizophrenia and autism begs the question as to the degree to which de novo post-zygotic mosaic mutations also contribute to disease. In this model, a mutation that occurs post-zygotically can seed some percentage of cells in the brain, and is sufficient to lead to neuronal dysfunction and disease. The approach of sequencing only non-neural tissues such as blood may underpower the detection of mosaicism, because mutations may be restricted to neural tissue. This proposal brings together a highly productive and collaborative team, in which each member contributes a special resource to make this effort truly unique. Gleeson and Mathern recently identified among the first de novo somatic mutations in the developing brain in the condition known as `hemimegalencephaly’ (HME), a catastrophic neuropsychiatric condition associated with focal cortical disorganization (FCD). By comparing DNA from diseased brain vs. blood, we identified de novo somatic mutations in PIK3CA, AKT3 and MTOR, part of the mTOR pathway, in as few as 8% of brain cells, resulting in perturbations in an entire cerebral hemisphere. Courchesne and Roy recently identified focal patches of abnormal laminar cytoarchitecture in frontal and temporal cortex in the majority of available brain samples of children studied with autism (ASD), and we suggest focal patches akin to FCD may have similar mutations and contribute to disease. The goal of this application is to extend the discovery of mosaicism in patients with epilepsy and autism in which neurohistopathological evidence points to FCD, by sequencing dysplasias compared with adjacent normal tissue and/or blood at the DNA and RNA level at the single-cell level. We aim to uncover key sets of genes, in which specific de novo mutations, in specific locations, at specific mosaicism levels, is sufficient to produce clinically defined disease. We will combine next-generation sequencing of FCD brain from patients with neuropsychiatric disease with advanced bioinformatics, single-cell sequencing, complete clinical correlated neuroanatomy and mouse modeling. We will: 1] Test for de novo somatic mutations in a retrospective and prospective cohort of FCD presenting with autism or epilepsy. 2] Correlate genetic disease burden with clinical, imaging, histopathological and single-cell sequencing findings. 3] Test mechanisms by which uncovered de novo mutations alter progenitor cell functions in mammalian cerebral cortex.

Project_5: Somatic Mosaicism in Schizophrenia and Control Brains – U01MH106891

Schizophrenia (SCZ) is a generally devastating neuropsychiatric illness with considerable morbidity, mortality, and personal and societal cost. Genetic factors have been strongly implicated via family and twin data, and more recently directly through genome-wide association studies (GWAS) and sequencing studies. Epigenetic modifications play a well-accepted role in a variety of medical and neurological illnesses, and are also implicated in SCZ. Somatic mosaicism is an underexplored, but potentially very important contributor to SCZ. There have been some intriguing hints that somatic mosaicism may play a role in SCZ, but assessment of this possibility awaits rigorous experiments, and that is the overarching goal of this proposal. The primary objective of our project is to identify and characterize the extent of somatic variation in post-mortem human brain samples from individuals with SCZ and controls. Following on work of members of our team, we will rigorously assess the somatic mosaicism in a large cohort of post-mortem human brains from the Common Mind Consortium, which members of our group are already analyzing for genotype, mRNA-seq and epigenome mapping. These brains are from individuals with SCZ (250) and controls (50+). We will look for retrotransposition events, copy number variants (CNVs) and single nucleotide variants (SNVs). All data will be made available to the research community through the Sage Bionetworks Synapse Platform. We have assembled the critical personnel, sample resources, technological know-how, and analytic strategies to be able to assess the role of somatic variation in the brain as well as begin to unravel SCZ biology.

Project_6: Role of brain somatic mosaicism in autism, schizophrenia, and bipolar disorder – U01MH106884

The broad, long-term objective of the proposed research is to identify somatic genetic mutations that occur in brain regions of individuals with autism spectrum disorder (ASD), schizophrenia (SZ), and bipolar disorder (BD). Ultimately the discovery and characterization of somatic mutations may help us to understand the etiology of these disorders and eventually lead to improved therapeutic strategies and diagnostic markers. ASD, SZ, and BD are all commonly occurring neuropsychiatric disorders that have a large genetic basis (with estimated heritability of ~75% to 80% for each condition). However, relatively few DNA variants have been identified that have causal roles in the etiology. We hypothesize that somatic mosaicism – the occurrence of mutations in selected body regions after conception – occurs in brain and contributes to the etiology of ASD, SZ, and BD. Specific Aim 1 is to identify the nature and extent of somatic mosaic mutations across brain and body regions in postmortem samples from apparently normal individuals. We will assess four categories of somatic variation: (1) single nucleotide variants (SNVs), (2) structural variants (SVs) including copy number variants, (3) L1 retrotransposition events, and (4) mitochondrial heteroplasmy. These types of variation will be detected using whole genome sequencing (Years 1 and 2), single molecule imaging with DNA nanochannels (Year 2), and single nucleotide polymorphism (SNP) arrays. Samples include brain regions (e.g. prefrontal cortex and cerebellum) and organs (e.g. heart and kidney). After identifying somatic variants we will perform rigorous validation. Specific Aim 2 is to identify the nature and extent of somatic mosaicism in genomic DNA from individuals with ASD, SZ, and BD. The same approaches for discovery and validation will be applied as in Aim 1. Specific Aim 3 is to functionally categorize somatic variants, particularly those that are predicted to disrupt the functions of genes previously implicated in those disorders. One approach is single-cell RNA-seq (to determine the consequence of the mutation on transcription, and to infer the cell type of origin of the somatic variant). Another approach uses neurons (or glia) derived from induced pluripotent stem cells (iPSCs) and stably expressing the wildtype or mutant forms of the somatic variants. These studies will help to establish the role of somatic mutation in neuropsychiatric disorders, including the functional consequences of such variation.