Molecular convergence of neurodevelopmental disorders

Transcriptomics analysis of iPSC-derived neurons and modeling of neuropsychiatric disorders

Induced pluripotent stem cell (iPSC)-derived neurons and neural progenitors are great resources for studying neural development and differentiation and their disruptions in disease conditions, and hold the promise of future cell therapy. In general, iPSC lines can be established either specifically from patients with neuropsychiatric disorders or from healthy subjects. The iPSCs can then be induced to differentiate into neural lineages and the iPSC-derived neurons are valuable for various types of cell-based assays that seek to understand disease mechanisms and identify and test novel therapies. In addition, it is an ideal system for gene expression profiling (i.e., transcriptomic analysis), an efficient and cost-effective way to explore the genetic programs regulating neurodevelopment. Moreover, transcriptomic comparison, which can be performed between patient-derived samples and controls, or in control lines in which the expression of specific genes has been disrupted, can uncover convergent gene targets and pathways that are downstream of the hundreds of candidate genes that have been associated with neuropsychiatric disorders. The results, especially after integration with spatiotemporal transcriptomic profiles of normal human brain development, have indeed helped to uncover gene networks, molecular pathways, and cellular signaling that likely play critical roles in disease development and progression. On the other hand, despite the great promise, many challenges remain in the usage of iPSC-derived neurons for modeling neuropsychiatric disorders, for example, how to generate relatively homogenous populations of specific neuronal subtypes that are affected in a particular disorder and how to better address the genetic heterogeneity that exists in the patient population.

CRISPR/Cas9-mediated heterozygous knockout of the autism gene CHD8 and characterization of its transcriptional networks in neurodevelopment

Background

Disruptive mutation in the CHD8 gene is one of the top genetic risk factors in autism spectrum disorders (ASDs). Previous analyses of genome-wide CHD8 occupancy and reduced expression of CHD8 by shRNA knockdown in committed neural cells showed that CHD8 regulates multiple cell processes critical for neural functions, and its targets are enriched with ASD-associated genes.

Methods

To further understand the molecular links between CHD8 functions and ASD, we have applied the CRISPR/Cas9 technology to knockout one copy of CHD8 in induced pluripotent stem cells (iPSCs) to better mimic the loss-of-function status that would exist in the developing human embryo prior to neuronal differentiation. We then carried out transcriptomic and bioinformatic analyses of neural progenitors and neurons derived from the CHD8 mutant iPSCs.

Results

Transcriptome profiling revealed that CHD8 hemizygosity (CHD8^+/−) affected the expression of several thousands of genes in neural progenitors and early differentiating neurons. The differentially expressed genes were enriched for functions of neural development, β-catenin/Wnt signaling, extracellular matrix, and skeletal system development. They also exhibited significant overlap with genes previously associated with autism and schizophrenia, as well as the downstream transcriptional targets of multiple genes implicated in autism. Providing important insight into how CHD8 mutations might give rise to macrocephaly, we found that seven of the twelve genes associated with human brain volume or head size by genome-wide association studies (e.g., HGMA2) were dysregulated in CHD8^+/− neural progenitors or neurons.

Conclusions

We have established a renewable source of CHD8^+/− iPSC lines that would be valuable for investigating the molecular and cellular functions of CHD8. Transcriptomic profiling showed that CHD8 regulates multiple genes implicated in ASD pathogenesis and genes associated with brain volume.

Electronic supplementary material

The online version of this article (doi:10.1186/s13229-015-0048-6) contains supplementary material, which is available to authorized users.

A molecular model for neurodevelopmental disorders

Genes implicated in neurodevelopmental disorders (NDDs) important in cognition and behavior may have convergent function and several cellular pathways have been implicated, including protein translational control, chromatin modification, and synapse assembly and maintenance. Here, we test the convergent effects of methyl-CpG binding domain 5 (MBD5) and special AT-rich binding protein 2 (SATB2) reduced dosage in human neural stem cells (NSCs), two genes implicated in 2q23.1 and 2q33.1 deletion syndromes, respectively, to develop a generalized model for NDDs. We used short hairpin RNA stably incorporated into healthy neural stem cells to supress MBD5 and SATB2 expression, and massively parallel RNA sequencing, DNA methylation sequencing and microRNA arrays to test the hypothesis that a primary etiology of NDDs is the disruption of the balance of NSC proliferation and differentiation. We show that reduced dosage of either gene leads to significant overlap of gene-expression patterns, microRNA patterns and DNA methylation states with control NSCs in a differentiating state, suggesting that a unifying feature of 2q23.1 and 2q33.1 deletion syndrome may be a lack of regulation between proliferation and differentiation in NSCs, as we observed previously for TCF4 and EHMT1 suppression following a similar experimental paradigm. We propose a model of NDDs whereby the balance of NSC proliferation and differentiation is affected, but where the molecules that drive this effect are largely specific to disease-causing genetic variation. NDDs are diverse, complex and unique, but the optimal balance of factors that determine when and where neural stem cells differentiate may be a major feature underlying the diverse phenotypic spectrum of NDDs.