Molecular Roots of the Social Brain

This work has been conducted as a multi-disciplinary collaboration within the “Gene Networks in Neural & Developmental Plasticity” (GNDP) research theme at the Institute for Genomic Biology (IGB).  The Stubbs lab is working with other GNDP faculty to test the hypothesis that conserved genomic mechanisms underlie the brain response to salient social interactions across the animal kingdom.  To test this hypothesis, we have used a diverse range of established animal behavioral models –honey bees, stickleback fish, and mice – and the stringent filter of comparative genomics to uncover conserved neurogenomic signatures of adaptive social responses. Our core focus is on networks of transcription factors (TFs) and other chromatin interactors that may determine individual and interspecies variation in social behavior.

The goal of the GNDP theme’s efforts is to tie the evidence we extract from each species together into a fundamental model of social responsiveness, and to follow these leads to elaborate conserved neurogenomic mechanisms that are directly relevant to human disorders and disease.  Indeed, our studies so far have uncovered conserved genes, pathways, and TF networks that are activated by salient social stimuli in all three species, including many associated with human disease.

Regulatory mutations that disrupt “neighborhoods” of co-expressed neurodevelopmental genes

Many critical developmental regulators are involved in differentiation of multipotent progenitor cell types and display especially dynamic expression patterns, as their differentiation roles are deployed across in different spatial locations over developmental time.  In partcular, many genes associated with neurodevelopmental disorders are involved in basic processes – the timely replication of neuron progenitor cells (NPC), their differentiation into neurons or glia, the outgrowth of neurites, or formation of synapses – events that occur at different times in each region of the developing brain.  The deployment of these factors at the right time and place requires the precise combinatorial action of enhancers, silencers, and other regulatory elements (REs) including some located significant genomic distances from the gene.  These REs interact with each other and target promoters through the formation of chromatin loops to specify the location, time, and circumstances of the genes’ expression.

The basic mechanisms of long-distance regulation are just now beginning to be revealed, and only a few distant enhancers have been described in detail so far.  However, genetic studies in humans and mice made it clear that distal REs play a much more prevalent role.

Classically, the relationship between essential genes and long-distance REs has been identified through mutations that either involve the REs themselves, or that separate REs from their cognate promoters via genome rearrangements.  Our own studies have leveraged a unique collection of mouse translocation mutants for this same purpose, focusing on mutations that occur far from genes, but give rise to dramatic developmental effects.  Most of the translocations we have identified are associated with neurological phenotypes, and closely model certain types of human neurodevelopmental disease.

A key example is the intense current focus of our group, a translocation called 16Gso that breaks between and dysregulates the neighboring Galnt17 and Auts2 genes. Mutations in this human region have been associated with intellectual disability (ID), autistic spectrum disorder (ASD), epilepsy, as well as depression, bipolar disorder and wide range of neurological diseases, and the AUTS2 gene has been the major focus of attention.  AUTS2 is a complex gene with isoforms that are involved in different aspects of basic neuron development.  Like other essential developmental genes, AUTS2 is also expressed dynamically across the brain throughout development, reflecting the gene’s essential roles in neuron development. GALNT17 had not been characterized previously, but our data show that the gene is expressed very dynamically across the developing brain, and highly co-expressed with AUTS2. We hypothesize that the two genes operate within the same functional pathways and that their expression is coordinated by a shared system of REs spread across the genomic region. We are focused on determining the details of gene function in conditional-mutant mice and in cultured neurons, and also in the characterization of the regulatory domain that controls this “neighborhood” of co-expressed neurodevelopmental genes.

We are convinced that this research has broader implications, in that regulatory mutations within co-expressed gene “neighborhoods” (more formally called topologically associating domains, or TADs) could explain many aspects of phenotypic variation associated with human neurological disease.

Developmental Regulation

The vertebrate T-box transcription factor gene Tbx18 performs a vital role in development of multiple organ systems. Tbx18 insufficiency manifests as recessive phenotypes in the upper urinary system, cardiac venous pole, inner ear, and axial skeleton; homozygous null mutant animals die perinatally.


Eukaryotic gene expression is orchestrated by a complex interplay between cis-regulatory elements (CREs) and the proteins that bind to and interact with those DNA sites, including transcription factors (TFs) and various types of chromatin proteins. In metazoan species the cis-elements that regulate gene expression can be found within the gene, directly upstream near the transcription start site, or long distances away from the regulated gene sequence.  These long-distance CREs interact with each other and with gene promoters through chromatin loops, dynamic structures that are unique to specific cell types and cell states.

Many TFs act by recruiting specific chromatin proteins or chromatin remodeling complexes to the CREs, thereby altering their activity and accessibility.  In mammals, one class of chromatin-modifying TFs, called KRAB-ZNFs, dominate the genomic landscape. The KRAB-ZNF family of genes is relatively ancient, but has expanded to a large family, through series of tandem segmental duplications, only in mammals.