Erich David Jarvis - US grants

Jarvis. Lay Abstract


Vocal learning is the process by which young animals learn to imitate the vocal sounds made by their parents or a tutor. This rare trait has only been found in humans, whales, dolphins, bats and 3 groups of birds (hummingbirds, parrots and songbirds). The goal of this project is to identify and characterize the brain areas that control vocal communication in hummingbirds, as well as test whether the brains of birds that do not have vocal learning, such as chicken and pigeons, contain similar structures. The main method used is analysis of a gene that is expressed in the brain when neuronal cells fire. Utilizing a molecular technique called in situ hybridization, one can determine the exact brain cells that are involved in the perception and production of vocalizations. By comparing hearing and vocalizing animals, one can then generate high-resolution maps of brain areas that control the production of learned vocalizations.

Language acquisition, both in terms of speech production and semantics, is a fundamental aspect of the human experience and depends on vocal learning. Why is it though that so few animals have vocal learning? Do only these animals have the necessary brain structures and connections? If so, what are these structures, and how did they arise during evolution? By addressing such questions, this study may help reveal what are the brain mechanisms required for vocal learning, and potentially help understand how humans learn speech. The results should also provide a framework for studying the neurobiology of learned vocal communication in other animal groups. If the evolution of vocal learning is under strong epigenetic constraints, it is possible that humans, cetaceans and bats have also evolved similar brain structures. Alternatively, brain areas for vocal learning may have evolved as a specialization of structures present in a vocal non-learning ancestor. Results from this project are expected to throw considerable light on these questions.

DESCRIPTION (provided by applicant): Many scientific projects funded by NIH investigate neural functions in the avian brain. However, there have been very few efforts by NIH projects to directly link discoveries in the avian brain with mammalian brain. This is due in part, because of a current dilemma on nomenclature of the avian brain. Nearly 100 years ago, scientists decided that the avian brain, above the level of the thalamus, is one large basal ganglia, a ventral portion of the mammalian telecephalon. In the 1960s, using several molecular markers, this conclusion was shown to be incorrect. However, since then there has been very little consensus on the correct correspondences between the avian and mammalian brains. As a result, the 100-year basal ganglia terminology is still in effect, creating confusion and inaccurate comparisons in basic and biomedical research. The objective of the Avian Brain Nomenclature Forum is to bring together experts in the field of avian neuroscience and comparative neuroanatomy for the purpose of making historical revisions of avian brain nomenclature. The Forum will take place July 18th, 19th, and 20th, of 2002, at the Duke University Medical Center, Searle Center. There will be a web site for forum participants to prepare themselves with hypotheses and knowledge, as well as post-forum publication of nomenclature changes in established scientific journals. The participants include professors, post-docs, and graduate students. The forum is expected to lead to significant changes in avian brain research and impact neurobiology research in general. Since it is the goal of NIH to improve public health, a mammalian group, it is imperative that the relationship between the avian and mammalian be updated and that serious comparative errors be removed.

This is an Alan T. Waterman's Award.

Lack of consensus on terminology and disparate uses of terms have stymied efforts to directly link discoveries in the avian brain with research on the brains of other vertebrates. Nearly 100 years ago, scientists decided that the avian brain, above the level of the thalamus, consists solely of the basal ganglia, which is a ventral part of the mammalian forebrain. In the 1960's, using several molecular markers, this conclusion was shown to be incorrect. However, since then there has been very little agreement on how avian and mammalian brains correspond with one another. As a result, the 100-year old basal ganglia terminology is still used, creating confusion and inaccurate comparisons in scientific research. The purpose of the Avian Brain Nomenclature Forum is to bring together experts in the fields of avian neuroscience and comparative neuroanatomy for the purpose of revising avian brain nomenclature. Participants will prepare for the forum by using a website for information before they attend, and then meet for three days for presentations and discussions. Professors, postdoctoral students and graduate students will try to reach a consensus by the end of the conference. Afterwards, the results will be published and made available at the web site. Popular avian brain atlases will be revised using the consensus normenclature, and will be available via web and CD formats.
The impact of this work will go far beyond simply comparative neuroanatomy, because of the strong interest across neuroscience in using birds as models for learning, for development and for studies of behavior including migration and social behavior; additionally, some of the conference output may be interesting to a broader public that is interested in birds.

DESCRIPTION:(provided by applicant) In the past several years, songbirds have increasingly become a useful model system for studying the functional role of the basal ganglia pathway loop in complex behaviors such as learned vocal communication. In young birds, the vocal part of the basal ganglia pathway is required for song learning. In adults who have learned their songs, it is not necessary for production of learned vocalizations, but shows dramatic changes in gene activation depending upon the social context in which vocalizing occurs. This context-dependent vocal gene activation lead to a number of intriguing testable hypotheses of the basal ganglia's functional role in adult vocal communication. These ranged from 1) regulating gene expression and activity of the vocal motor pathway to 2) generating on-line complexity of the singer's vocalizations. The goals of the experiments in this proposal are to test these hypotheses and, in doing so, to determine the basic function of the vocal basal ganglia loop in learned vocal communication. Since the loop is built within a circuit that is conserved in the vertebrate brain, it is believed that the underlying mechanisms discovered will generally apply to complex social behaviors such as learned vocal communication in humans. However, since vocal learning is a very rare trait, as it is only found in 3 groups of mammals (humans, dolphins, and bats) and 3 groups of birds (songbirds, hummingbirds and parrots), with songbirds being the most studied animal model, the results of this proposal are expected to generate unique insight into higher order brain function, and insight into diseases that affect speech, language, and cognitive processes.

Abstract Not Provided

DESCRIPTION (provided by applicant): The goal of this proposal is to identify sensory- and motor-driven genes activated by learned vocal communication in normal and deafened subjects. The animal model we will utilize is a songbird. Songbirds are one of the only accessible non-human animals where learned vocal communication, the substrate for human language, can be studied. Non-human primates, rodents, and other commonly studied animals do not have; this ability. In songbirds, hearing species-specific vocalizations (sensory activity) induces large increases of gene expression throughout the auditory pathway. This sensory-driven expression is experience-dependent, as induction requires that the birds be raised with adult conspecifics and induction is highest when the birds hear novel species-specific songs. The expression is blocked by deafening. The act of producing imitated vocalizations (motor activity) induces large increases of gene expression throughout the cerebral vocal pathway, and this motor-driven expression is not blocked by deafening. However, when birds are deafened, like humans, their learned vocalizations deteriorate. This deterioration, in songbirds, is an active process that requires the basal ganglia cortical-like vocal pathway, in which vocalizing-driven gene expression is found. To date, few genes have been identified with such sensory- and motor-driven regulation during behavior and none have yet been identified that change with deafening-induced deterioration of learned vocalizations. It is believed that an entire gene regulatory network is activated in these behavioral processes. We will use behavioral, neuroanatomical, and high throughput molecular biological approaches to identify and characterize sensory and motor-driven genes activated by normal vocal communication and by deafened-induced deterioration of learned vocalizations. Since most songbird genes have significant homology to known mammalian genes our experiments will enable us to identify avian brain genes with humain/mammalian homologues amenable to experimental characterization in the songbird system.

[unreadable] DESCRIPTION (provided by applicant): About 1% of the US population are deaf, and about 29.1% over the age of 65 have hearing loss problems. These problems cause deterioration in learned speech. The goal of this proposal is to identify proteins involved in active deterioration of learned vocalizations, using songbirds as a model system. Songbirds are one of the only accessible non-human animals where learned vocal communication, the substrate for human language, can be studied. Other commonly studied animals do not have this ability. In songbirds, hearing oneself vocalize induces large increases of gene and protein expression in parts of the auditory pathway. The expression is blocked by deafening. The act of vocalizing also induces large increases of gene and protein expression in the vocal pathway. When birds are deafened, like humans, their learned vocalizations deteriorate. This deterioration in songbirds is an active process involving the basal ganglia cortical-like part of the vocal pathway, in which vocalizing-driven gene expression is found. That is, the prevention of deterioration in intact animals requires that they hear themselves vocalize by auditory feedback. To date, few proteins have been identified with such sensory- and motor-driven regulation and none have been identified that change with deafening-induced deterioration of learned vocalizations. It is believed that an entire gene regulatory network is activated in these behavioral processes. We will use behavioral, neuroanatomical, and high throughput proteomic approaches to identify and characterize proteins activated by normal hearing of oneself vocalize as a control group and by deafened-induced deterioration of learned vocalizations as an experimental group. Since most songbird proteins have significant homology to known mammalian proteins, our experiments will enable us to identify avian brain proteins with human homologues amenable to experimental characterization in the songbird system. Our long-term goal is to manipulate such proteins to prevent deafened-induced vocal deterioration. [unreadable] [unreadable] [unreadable]

[unreadable] DESCRIPTION (provided by applicant): While studying the role the basal ganglia pathway in learned vocal communication, we discovered an unexpected phenomenon: following neurotoxic lesions, the avian striatum recovered itself. This phenomenon, as far as we know, is unprecedented in the mammalian brain. Here we propose to investigate the mechanisms of this recovery: whether this is in fact new neuron regeneration or neuron invasion from the surrounding areas. Further we will determine the time course of this recovery accompanied with the behavioral (song) recovery. We will identify whether the cellular organization in the recovered striatum is the same as in intact striatum. Finally, we will test whether the recovery is specific to the neurotoxic lesion or if it is a more general aspect of the avian striatum. As the avian striatum contains neurons similar in their electrophysiological and molecular properties to their mammalian counterparts, the project is expected to impact our understanding of brain regeneration. The relevance of this research to public health is that we would find potential ways to repair damaged basal ganglia brain areas, and in particular for correcting speech deficits. This research will be done primarily in Slovakia at the Institute of Animal Biochemistry and Genetics at the Slovak Academy of Sciences in collaboration with Lubica Kubikova, as an extension of NIH grant #R01 DC007218-01. [unreadable] [unreadable] [unreadable]

Scientific problem to be addressed, and why it is important The fundamental scientific problem we propose to address is to determine the basic neural and molecular requirements for vocal learning, the behavioral substrate for spoken language. Language is one of the essential behaviors that make us human. With it, we are able to communicate complex concepts, pass on knowledge culturally, and advance human civilization. Without it [unreadable] due to brain damage, trauma, or developmental diseases - we live a life of impoverished social communication and life dependency on others. Studying this fundamental problem requires that we compare the vocal behavior and associated brain pathways of the few rare groups that have vocal learning - four groups of distantly related mammals (humans, cetaceans, elephants, and bats) and three groups of distantly related birds (parrots, hummingbirds, and songbirds) [unreadable] with the vast majority of species that do not have it - non-human primates, rodents, suboscine songbirds, pigeons, chickens, etc.1,2. Remarkably, although vocal learners are distantly related to each other, of those whose brains that have been studied (humans, parrots, hummingbirds, and songbirds), evidence suggests that they share a similar vocal pathway forebrain organization: a premotor or anterior vocal pathway (AVP) necessary for vocal learning, including syntax learning, and a motor or posterior vocal pathway (PVP) necessary for production of learned vocalizations1. These forebrain pathways are not found in vocal non-learners. Yet, vocal non-learners appear to possess similar brain pathways for learning and production of non-vocal motor behaviors. Given these findings, we have proposed that the fundamental difference between vocal learners and non-learners is a genetic difference or several genetic differences that control the connection of forebrain motor learning pathways onto brainstem motor neurons that normally control the production of innate vocalizations1. In this essay, I outline the following proposal for testing this novel idea: 1. Discover molecular differences in the motor learning pathways between vocal learners and non-learners. 2. Manipulate their network connectivity by developing novel gene manipulation tools. 3. Use these tools to modify vocal nuclei connectivity and thus vocal behavior of a vocal non-learner, potentially allowing other species to modify and imitate sounds and allowing correction of damaged vocal learning brain pathways in vocal learners. Inducing such connectivity and behavioral changes in vocal non-learners would have profound impact towards understanding molecular mechanisms of vocal learning and evolution of language. Repairing the pathway in vocal learners, when damaged, would have profound impact for correcting neurological disorders of speech.

Abstract The goal of this Transformative R01 project is to develop genetic strategies for neuroengineering a robust vocal learning phenotype in mice, which may yield the first mammalian model for treating human vocal communication disorders. Up to 10% of humans have some sort of communication dysfunction in their lifetimes (Speech and Language Impairments, NICHCY, 2011), yet there is no genetically tractable system for enhancing or repairing brain circuits involved in speech. We recently discovered that mice, which are highly tractable, show evidence of a rudimentary vocal learning phenotype. Specifically, mice have some features once thought unique to humans and other vocal learning species, including the ability modify ultrasonic vocalizations (USVs) based on context; a forebrain vocal circuit that is active during vocalizing, is required for frequency modulation and organization of syllables, and that directly connects to brainstem motor neurons that control the larynx; and syllable sequencing deficits when given a FoxP2 mutation known to cause phoneme sequencing dyspraxia in humans. However, compared to humans and songbirds, these phenotypes are much more limited in mice. These and other findings led us to hypothesize that similar to natural variation in ability among vocal learners, presumed vocal non-learners may exhibit vocal learning-like phenotypes along a continuum of complexity across species. In this context, given the presence of the basic neuroarchitecture in mice considered obligate for vocal learning in categorical species, we postulate that the mouse vocal system and associated behaviors may be liable to enhancement, thereby providing a foundation for the development of novel and effective strategies for ameliorating disorders of human vocal communication. To accomplish this, we will exploit recent findings from our laboratory where we discovered convergent specialized gene expression of ~50 genes in vocal brain regions of several vocal learning species, including humans and songbirds, many of which are involved in brain pathway development. We hypothesize that evolutionary changes in the regulation of trait-specialized genes are responsible for the emergence of more advanced vocal plasticity and other complex behavioral traits. Our objective is to recapitulate the unique expression patterns of these genes in mice to enhance the vocal learning phenotype at the level of connectivity, in vivo electrophysiology, and behavior. We will do so using viral strategies, introduction of human neural stem cells, and the generation of transgenic animals. If successful, our studies are expected to impact the field by: 1) Establishing how vocal-learning specialized genes shape the neurocircuitry and physiology for this complex behavior; 2) Developing a novel, genetically tractable mammalian model system for unveiling the neurobiological details of human language and treatments for its dysfunction; and 3) Serving as a platform for neuroengineering complex behavioral traits in general.