Multi-level analysis of the Shank3 mutation in autism: translating genetic studies into pathophysiological mechanisms and therapeutic targets

Recent large scale genetic studies strongly suggest that hundreds of genes are associated with autism spectrum disorder (ASD). This heterogeneity of etiology, together with variable penetrance of genetic contributions, has made it very difficult to dissect the neural mechanisms underlying the core defects and symptoms in ASD. One approach, which has been successful in studying many other complex disorders, is to focus on a few rare and highly penetrant genetic mutations to gain a deep mechanistic understanding of the disorder and to derive novel strategies for treatment. The hope is that knowledge obtained from the study of rare cases will likely be applicable to a broader group or subgroup of patients with genetically diverse but mechanistically related etiology.

Human genetic studies have identified deletions/mutations of the Shank3 gene as the cause of Phelan- McDermid Syndrome (PMS), a developmental disorder with more than 50% of the patients having a diagnosis of autism or ASD. Shank3 is a member of the Shank family of proteins which functions as a postsynaptic scaffolding protein to orchestrate the assembly of the macromolecular postsynaptic complex at excitatory synapses. Recent animal model studies from several laboratories show that disruptions of the Shank3 gene in mice lead to synaptic dysfunction and autistic-like behaviors. Together, these data suggest that disruption of Shank3 is a monogenic cause of ASD.

The major challenge now is how to translate the results from animal models and human genetic studies into a deeper mechanistic understanding of the pathophysiology and into novel therapeutic targets and strategies. Meeting this challenge requires coordinated efforts from a group of scientists with unique and complementary expertise. The establishment of the Simons Center for the Social Brain at MIT, with a collection of outstanding neuroscientists from MIT and collaborating Harvard hospitals, provides an unprecedented opportunity and resource to take on this challenge. In particular, the Center provides a much needed platform to bring scientists across the MIT campus together to work on a single common topic and functions as a command center to set strategic directions allowing for big questions which are beyond the scope of any single laboratory to be addressed. It is with this vision that we propose to launch a Shank3-targeted program project. We will use Drosophila, mouse, and ES cell/iPS cell-derived human neurons as model systems. Each model system has its unique strength and together they will help to identify Shank3-regulated molecular and cellular pathways and circuitry mechanisms specific to ASD. If successful, these studies will pave the way for developing mechanism-based effective treatments.


Component Projects:

Use novel human mutation-based mouse models to identify molecular, cellular, and circuitry mechanisms specific to ASD (Guoping Feng)

Aim 1: To characterize behavioral and synaptic phenotypes of mutant mice containing the 3680Gins autism mutation and make these mice available to the research community.
Aim 2: Using proteomic approaches to identify molecular network and signaling pathway defects at the postsynaptic complex in the 3680Gins mutant mice.
Aim 3: To examine circuitry and network level defects in 3680Gins autism mutant mice using high-density multi-electrode recordings.
Aim 4: To compare synaptic and circuit dysfunctions, as well as behavioral differences between R1117X and 3680Gins knock-in mouse lines and identify mutation-specific defects.

Use Drosophila as a model system to identify modifiers of Shank3 function at the synapse (Troy Littleton)

Aim 1: Generation and characterization of DShank mutants.We will generate null mutations in the sole Drosophila Shank homolog using P-element mediated excision of the locus to make intragenic deletions. We will assay synaptic development, structure, and function in the mutants, and determine if Shank is required pre- versus post-synaptically using transgenic rescue with DShank wildtype transgenes. These experiments will allow us to determine the normal biological role of Shank at the Drosophila NMJ, a prerequisite for examining conservation of Shank function and performing genetic interaction screens.
Aim 2: Characterization of conservation of Shank function and the effects of autism-linked mutations.We have localized DShank to the postsynaptic compartment at the Drosophila NMJ, similar to its localization at mammalian synapses. We will determine whether human Shank 1-3 proteins can rescue any observed phenotypes when expressed either pre or postsynaptically. It will be interesting to determine if Shank3 has unique properties compared to Shanks 1 and 2, or whether each protein performs similar function. We will also assay autism-associated human Shank3 point mutants in transgenic rescue experiments to define how these autism-causing alleles disrupt Shank localization or function.
Aim 3: Genetic Analysis of Shank Interacting Proteins. The objective of this aim is to perform unbiased genetic screens to identify regulators of Shank localization and function. We will take advantage of the postsynaptic localization of Shank and any phentoypes we identify to screen for 2nd site loci that can genetically enhance or suppress Shank function, or that can alter its localization. We will use an RNAi-based screening approach using a custon built UAS-RNAi transgenic collection that includes known synaptic proteins. We will examine any identfied interactors in detail to determine how they function independently of Shank, and how they interface with Shank to control synaptic development and function.

Use human ESC and iPSC-derived neurons to identify molecular, cellular, and functional readouts that are specific for ASD (Rudolf Jaenisch)

Aim 1: Generation of SHANK3 mutant and control human pluripotent stem cells.
a. We will introduce the 3680Gins point mutation into SHANK3 using TALEN-mediated gene editing in established human ES lines. Our goal is to generate isogenic pairs of mutant and control cells that differ exclusively at the disease-causing mutation. Both alleles of SHANK3 will be targeted consecutively to derive homozygous mutant cells.
b. We will derive patient iPSCs employing an mRNA-mediated reprogramming strategy. Fibroblasts of Phelan-McDermid Syndrome patients carrying a small deletion or point mutation in the SHANK3 gene will be used as starting material. Isogenic control cells will be generated by either repairing the point mutation of the mutant allele or by inserting a full-length SHANK3 cDNA into the cells carrying a deletion.
Aim 2: Differentiation of SHANK3 mutant and control cells into excitatory and inhibitory neurons.
a. Established protocols will be used to differentiate pluripotent stem cells into neural precursors and into mature excitatory and inhibitory neurons.
b. To facilitate the derivation of particular subtypes of neurons, we will use TALENs to generate lineage-specific reporter cell lines by inserting a GFP reporter into key transcription factors.
Aim 3: Phenotypic characterization of mutant and control neurons.
a. Mutant and control neurons will be compared using various cellular markers to identify SHANK3 deficiency induced morphological changes, such as dendritic arborization and spine density.
b. We will assess whether key postsynaptic scaffolding signaling pathways are affected by SHANK3 deficiency. This will be aided by the use of specific calcium sensors.
c. We will characterize the electrophysiological properties of mutant and control neurons to investigate whether abnormalities seen in mutant mice can be detected in cultured human mutant neurons.
d. To perform global transcriptome and proteome analysis on mutant and control we will identify mutant-specific gene expression and post-transcriptional alterations.
Aim 4: Transplantation of SHANK3 mutant human cells into the developing mouse CNS
a. We will generate chimeric mice carrying human SHANK3 mutant neurons, using ultrasound-guided in utero transplantation of GFP-labeled human neural precursors into the developing CNS of mouse embryos at different stages of gestation.
b. If chimeric mice can be obtained, we will characterize the cellular and electrophysiological properties of the human neurons utilizing similar approaches as being used for the characterization of Shank3 mutant mice.