Stephanie A. White - US grants

DESCRIPTION (provided by applicant): The long term objective of this proposal is to discover the neural basis for socially-learned vocal communication, a form of implicit learning. Deficits in implicit learning, including language disorders, have devastating consequences for social integration. To treat or prevent these deficits, the neural mechanisms for learned vocal communication, currently unknown, must be understood. Language is uniquely human, but other species possess subcomponents of language enabling controlled molecular, physiological and behavioral studies. Songbirds are a useful model because they learn their songs through social interactions in a manner that exhibits significant parallels to human speech development. We use songbirds to investigate FoxP2- a conserved transcription factor whose mutation causes a severe language disorder as an entry point into the neuromolecular networks for vocal learning. Humans and songbirds possess full length and truncated FoxP2 isoforms. The latter lacks the DNA binding domain but includes a dimerization domain whereby it could interfere with transcriptional activity. These forms will serve as tools to augment or decrease FoxP2 function and determine bidirectional changes in gene coregulatory networks, neurophysiology and behavior. In addition to organizing brain structures, FoxP2 has post-organizational roles, as observed within area X, the basal ganglia sub-region dedicated to song. Area X FoxP2 levels are robust early in development, but when juvenile or adult birds practice their songs, FoxP2 is acutely down-regulated. We hypothesize that FoxP2 acts as a molecular gate of neural and behavioral plasticity even in the adult: Behaviorally driven reduction of FoxP2 during song learning and adult practice enables vocal adjustments. Conversely, high FoxP2 levels promote brain organization and reinforce optimal neural activity and behavior later in life. To test this, FoxP2 function will be augmented and reduced via viral driven expression of the two isoforms during periods of brain organization, and during song learning and adult maintenance. Molecular networks will be identified using RNAseq and a powerful systems level technique known as weighted gene co-expression network analysis, to highlight behaviorally significant relationships. Electrophysiological recordings from cultured neurons and acute brain slices will be used to examine emergent neurophysiological changes. Behavioral effects on song learning and on deafening-induced song deterioration in adulthood will be tested. This work is relevant to the NIMH's programmatic goals of developing and exploiting animal models for mental disorders in which social-learning deficits are a major component, including but not limited to autism spectrum disorder. By investigating an animal that learns its vocalizations, we can illuminate how molecules linked to human language disorders disrupt neural function, providing critical insight for the development of be- havioral and pharmacological interventions. These studies will provide basic but critical information about the neural processes underlying a complex socially-learned behavior.

DESCRIPTION (provided by applicant): Impairments in language, social interaction and behavioral flexibility that together occur in young children comprise the hallmarks of autism spectrum disorder (ASD). Language, and thus autism, is a uniquely human phenotype, but studies of its neurophysiological and molecular basis require animal models. The broad long-term objective of this proposal is to develop a novel model of ASD using the zebra finch songbird. While no single model will capture all features of ASD, songbirds are one of the few in which the language subcomponent comprised by learned vocal communication can be studied. This is because songbirds, like humans and unlike traditional laboratory animals, learn their vocalizations through social interactions with conspecifics. Support for this idea comes from comparing the expression patterns of the autism susceptibility gene, contactin-associated protein-like 2 (Cntnap2) in the brains of vocal learners and non-learners. In both humans and finches, Cntnap2 is enriched in regions that are functionally specified for learned vocal communication. In contrast, transcript distribution in rodent brain shows no region-specific enrichment. As humans with CNTNAP2 mutations exhibit features of ASD and SLI, we will develop small hairpin RNA constructs that decrease Cntnap2 levels, first in cultures of zebra finch telencephalic neurons, then in ovo in the developing embryo. We will determine the effects of Cntnap2 reduction on electrophysiological and neuroanatomical properties in vitro and in vivo, and on vocal learning, other social and repetitive behaviors. Not only will this work illuminate Cntnap2's role in ASD, it will additionally provide a proof-of- principle for use of songbirds in understanding the role of other autism susceptibility genes on socially-learned vocal communication. A songbird model of ASD promises to provide critical information about cellular and circuit effects, and will be useful for screening therapeutic interventions. Our studies aim to inform novel approaches to improve social interactions, and thus the quality of life, of autistic children. PUBLIC HEALTH RELEVANCE: Children diagnosed with autism fail to develop language, have other social difficulties and overly repetitive behaviors. To understand the neural basis for these deficits, we will develop a novel model of autism using songbirds, arguably the only practical laboratory model for probing the vocal learning subcomponent of language. We will investigate the role of the autism susceptibility gene, Cntnap2, on vocal learning and other social and repetitive behaviors, with the goal of applying these findings to humans and formulating novel approaches to improve social interactions, speech learning, and the quality of life of autistic children.

DESCRIPTION (provided by applicant): Communication deficits accompany a wide array of developmental disorders and adult-acquired neurological diseases. The neurobehavioral genetic bases for these deficits are poorly understood, rendering treatment challenging yet motivating experimental investigation. Songbirds are advantageous models for uncovering the neural basis for human vocal communication given their structural and functional similarities to humans and the ability to conduct molecular, physiological, and behavioral manipulations not feasible in humans. This proposal focuses on dopaminergic (DA) regulation of a key candidate molecule, Synaptotagmin 4 (Syt4), in basal ganglia circuitry dedicated to learned vocalizations using the songbird model. In both songbird and human basal ganglia, DA regulates pathways important for behavior; when dopamine is lost as occurs in Parkinson's Disease (PD), vocal and non-vocal motor symptoms arise. The molecular pathways that mediate the vocal changes, currently unknown, must be determined in order to remediate this facet of the disease. Recent converging evidence highlights the importance of Syt4 in these pathways. Our studies on Syt4 gene expression show that its levels within the song-dedicated sub-region of the songbird basal ganglia are tightly linked to singing. Bioinformatic studies from the lab predict that Syt4 interacs with other genes in a DA pathway supporting learned vocal behavior. Additional findings implicate Syt4 in human cognitive specializations that distinguish our species from other primates. I thus hypothesize that dopaminergic regulation of Syt4 is functionally specific to vocal pathways, and that loss of DA converts Syt4 regulation from being driven by patterned activity associated with vocalizing to generalized non-specific activity. To test this, I will first determie whether Syt4 is regulated by natural fluctuations in DA that occur during vocal behavior under different social contexts and in a Parkinsonian-like state. Follow up experiments will then test whether loss of DA, such as occurs in PD, switches Syt 4 regulation to that found in non-vocal areas. Results from these aims will provide insight into molecular mechanisms operating in the basal ganglia to support vocal behavior in songbirds and potentially, humans, with the promise of new therapeutic targets to treat vocal disorders.

Project Summary Klinefelter Syndrome is a common chromosomal abnormality of males who have two X chromosomes XXY, rather than one, XY. XXY males experience a variety of congenital developmental problems, including infertility, lower levels of androgens, increased risk for obesity and metabolic disease, increased risk for autoimmune diseases, and cognitive features including alterations in executive function and delayed language development. The long-term objectives of this project are to identify X chromosome genes that cause behavioral features of Klinefelter Syndrome, using novel mouse models. An overarching question is to separate the direct effects of X chromosome genes that cause Klinefelter Syndrome traits, from those caused by lower testosterone levels in XXY individuals. A novel mouse model produces XXY, XY, and XX mice that have either testes or ovaries, so that sex chromosomal effects can be identified that do not require testicular secretions, or occur when testicular secretions do not explain differences. Mice will be compared in a series of Klinefelter Syndrome-relevant behavioral measurements that assess executive functions, development of vocalizations, and partner preference. The expression levels of six specific X chromosome genes, which are the major candidates for causing the features of Klinefelter Syndrome, will be directly manipulated to assess which is/are likely the causal genes in the mouse model.

Summary The basal ganglia comprise key brain structures for generating and refining motor sequences necessary for a variety of complex behaviors that are acquired through procedural learning. These skills are often best learned during early developmental critical periods. Prior work has shown that practicing these skills drives acute changes in gene expression within the underlying basal ganglia microcircuit. This behavior-linked transcriptional activation is observed in juveniles during the sensorimotor critical period but also occurs in adults after the critical period has closed, suggesting that it is not specific to learning. Remarkably, a separate transcriptional profile is only found in juveniles and correlates with the quality of the learned skill. These observations suggest that the spatiotemporal overlap of the behavior-linked and learning-related changes in juveniles constitute a transcriptional program that is permissive for learning. To test this idea, the individual basal ganglia cell types in which these programs occur, currently unknown, must be resolved. Understanding these ?transcriptional fingerprints? will be key to deciphering molecular signaling pathways that support sensorimotor learning. This project leverages a well-characterized vertebrate model, the zebra finch, in which basal ganglia transcriptional changes linked to both practice and learning have been demonstrated via bulk sequencing of the entire region, but have not yet been traced to distinct basal ganglia cell types. Thus, one major aim is to identify and compare single-cell gene transcripts from behaviorally activated and non-activated basal ganglia, during and after the critical period, in order to identify specific cell types and cell signaling pathways undergoing behaviorally regulated changes, including those that support learning. In this species, only males undergo sensorimotor learning so comparison to the analogous regions in female brains will highlight the most relevant changes. The second goal is to select key cell type identifiers as well as molecules implicated in the sensorimotor learning process and develop probes to map their spatial expression in samples of the intact microcircuit using multiplexed error-robust fluorescence in situ hybridization (MERFISH). Together, these two integrated aims will illuminate how the basal ganglia changes over the course of repeated behavioral refinement to enable optimal sensorimotor learning. This work has direct implications for better understanding of the mechanisms that underlie the effectiveness of human behavioral therapies and may highlight pharmaco-therapeutic targets to improve treatment efficacy in brain disorders ranging from autism to stroke to cerebral palsy.