Defining Cell-Type Specific Signatures and Dysregulated Pathways from Blood and Brain in PWS (Year 2)

Funding Summary

In year 1 of this project we found increased UBE3A levels in white blood cells was linked to more severe autism features, but only in non-deletion PWS (most matUPD). In year 2, we will analyze the dataset created in year 1 to help us understand how activity of UBE3A and other key genes (related to inflammatory and other dysregulated pathways) influence expression of behavioral issues and other key clinical features in PWS. We will examine if and how the genetic findings from blood may be specifically linked to genetic changes in the related brain immune cells (glial cells) which support the proper functioning of neurons in cortex and hypothalamus. Together these will: (i) define genetic pathways dysregulated in PWS blood and brain, and (ii) identify existing medications used for other conditions to target these pathways in future PWS clinical trials to address the needs of people with PWS, their families, and the clinicians caring for them.

Dr. Theresa Strong, Director of Research Programs, shares details on this project in this short video clip. 

Lay Abstract

For people with PWS, their families, and the medical professionals caring for them, the pressing needs are identification of early predictors (clinical and/or biological) of symptoms including serious mental illness that may be targeted with specific treatments. To date, this has not been comprehensively investigated particularly in individuals with PWS due to maternal uniparental disomy (matUPD), who are especially vulnerable to both psychosis and autism features. The underlying biological mechanisms in key areas of the human brain, linked to these phenotypes, have also not been comprehensively studied.

This project is the 2nd year renewal aimed at characterizing biological mechanisms in blood and brain underlying key symptoms of PWS to address these needs. It will bring together an interdisciplinary team of investigators with an outstanding record in the field, established state-of-the-art technologies, deep clinical expertise and access to the world’s largest collection of post-mortem brain tissues from individuals with PWS. In year 1 we found increased UBE3A levels in white blood cells was linked to more severe autism features, but only in non-deletion PWS (most matUPD) (Baker et al. Godler 2020 Translational Psychiatry 10:362). We also analyzed brain tissues from 15 PWS and 39 typically developing individuals and processed 8 PWS (4 deletion and 4 non-deletion) and 4 control brain samples using a state-of-the-art single nucleus RNA sequencing technology. This provided information about the activity of most genes in the human body (transcriptome) at unparalleled resolution in different types of cells of the brain, and confirmed that UBE3A expression was significantly elevated in the brain tissues, but only in the non-neuronal cell types, and only in the non-deletion subtypes. The non-deletion brain tissues also showed significant dysregulation of specific inflammatory pathways (e.g. neutrophil degranulation). While these pathways have been previously implicated in autism, this is the first time they have been identified to be dysregulated in PWS non-deletion brain tissues.

In year 2, we will analyze this rich dataset to help us understand how activity of UBE3A and other key genes (related to inflammatory and other dysregulated pathways) influence expression of behavioral issues and other key clinical features in PWS. We will examine if and how the genetic findings from blood may be specifically linked to genetic changes in the related brain immune cells (glial cells) which support the proper functioning of neurons in cortex and hypothalamus. We will use artificial intelligence to identify treatments from global drug databases, to target specific pathways identified to be dysregulated in blood and brain tissues in different types of cells. Together these will: (i) define genetic pathways dysregulated in PWS blood and brain, and (ii) identify existing medications used for other conditions to target these pathways in future PWS clinical trials to address the needs of people with PWS, their families, and the clinicians caring for them.

Research Outcomes: Public Summary 

Abnormal cell-type specific expression of genes at 15q11-q13 and outside this region has not been mapped in the human brain in PWS. This is despite an important role 15q11-q13 genes play in brain connectivity and related behavioural issues. Understanding how different genes and related pathways are affected in different cell types in brain and peripheral tissues, and how these underly different PWS phenotypes, are needed for development of prognostic tests and early interventions using targeted therapeutics.
This study used state-of-the-art technologies, deep phenotyping and unique materials, including the world’s largest collection of post-mortem PWS brain tissues from the the NICHD Brain and Tissue Bank for Developmental Disorders, University of Maryland, Baltimore, Maryland, U.S.A. to address these gaps in knowledge.
The first phase of this study examined expression of 8,338 long non-coding RNAs and 17,079 protein-coding genes using single-nucleus RNA-sequencing (snRNA-seq) in a brain region called the prefrontal cortex of donors with PWS and matched neurotypical controls. The shortlisted differentially expressed genes (DEG) were examined using targeted analysis using a test called droplet digital PCR (ddPCR) in blood from an independent group of living individuals with PWS, with measures of PWS symptoms severity.
Glial cells (cell that support function of neurons in the brain) showed the highest number of DEGs and dysregulated pathways as compared to all other cell types in PWS groups, with 54 genes and related pathways dysregulated in brain tissues of donors with PWS across all cell types. RPS18 was the only gene upregulated across all comparisons in the brain, with increases in its expression in blood from an independent group of 38 individuals with PWS being associated with intellectual functioning and measures of challenging behaviors in PWS.
The second phase of this study examined different cell types in brain tissues only focusing on genes from 15q11-q13 – the PWS critical region. Significant differences were observed between PWS deletion and non-deletion groups in the proportion of non-neuronal cells expressing UBE3A and HERC2 as compared to control tissues. Expression levels of UBE3A and HERC2 were then examined in blood from an independent cohort of 38 individuals with PWS, aged 1 to 45 years. Increased HERC2 expression in blood was associated with better intellectual functioning and reduced number of behavioral issues in the deletion group. Decreased expression of UBE3A in blood was associated with better intellectual functioning, but only in the non-deletion group.
Together, these studies used intergroup comparisons in brain tissues of individuals with PWS and genotype-phenotype studies to identify cell types specific pathways dysregulated in the brain and in blood that are linked to clinical severity of PWS.
In phase 3 of this project, we used established machine learning approaches in drug re-purposing studies utilizing data from phases 1 and 2. Studies utilizing DEGs conserved between all cell types in all PWS subtypes as compared to control tissues identified only 2 genes out of 54 DEGs across cell types that encoded potential drug targets with one of these being RPS18. RPS18 was also identified as a druggable target using data from subtype specific analyses. Three drugs were identified from further comparisons which included an anti-inflammatory medication and a small molecule used for cancer therapeutics. Drug repurposing studies using data from differentially expressed in glial cells identified many more possible targets and approved compounds targeting these. There were 185 such targets and compounds from microglial data and 48 for oligodendrocyte data. With top compounds including specific vitamins and specific drugs targeting ribosomal function, used in settings other than PWS, as well as GABA-A receptor, previously implicated as a contributor to PWS clinical features.
In summary this study will result in paradigm shift and impact including: (1) Improved understanding of PWS molecular causes which can be reversed. (2) New targets identified for novel therapeutics. (3) Evidence to support development of new human in vitro disease models (i.e. glial cells) identified that reflect cell-type specific changes in vivo in the human brain. (4) Novel single cell brain transcriptomic and clinical databank established for the wider community to address key issues beyond the scope of this project. (5) Subtype, mechanism and phenotype specific drugs selected for clinical trials (drug repurposing). Together this knowledge will enable provision of more accurate prognostic information, evidence-based choice of already available targeted treatments previously used for other conditions (for PWS as a whole and/or specific subtypes), more effective participant stratification in clinical trials, and identification of new targets for novel therapeutics based on omics-driven medical care. Due to the cost, long lead-time and lack of access to such unique resources, this project provided one of a kind scenario to achieving much needed ready-to-translate outcomes.