Viviana Gradinaru - US grants

DESCRIPTION (provided by applicant): Neurodegeneration has proven notoriously difficult to study and there is currently no proven or acceptable method to prevent or slow down the course of disease in humans. A very successful intervention for Parkinson's disorder (PD) is neuromodulation via deep brain stimulation (DBS). DBS successfully restores motor function but what DBS does to the course of the disease is very poorly understood. Intriguingly, a few rodent model studies and clinical observations suggest that DBS could be neuroprotective, but because current practice is to implant the electrodes late in the progression of the disease, neuroprotective effects of electrical stimulation have been challenging to document. It is therefore vital that we are not missing on a crucial opportunity for neurological patients and research the causal links, mechanisms, and timelines associated with neuroprotection via neuromodulation. I propose an interdisciplinary approach for which I am uniquely trained that uses optogenetics, electrophysiology, biochemistry, and collaborative device engineering to study the interplay between neuronal health and brain circuit activity in intact behaving rodents. Specifically, I propose to study the factors influencing the function and health of dopaminergic neurons in the brain and their role in animal behavior. Our findings could allow us to positively interfere with cells such as the dopaminergic neurons in the substantia nigra pars compacta (SNc) that degenerate and die in PD. Below I list 3 specific challenges that I will tackle using innovative, interdisciplinary, approaches. 1. Are all SNc dopaminergic neurons equally impactful on behavior or are there hotspots where cells, due to their heterogeneous electrical and neurochemical characteristics and connectivity, can maximally interfere with behavior when degenerated? 2. Once dopaminergic degeneration starts, can neurodegeneration be halted or slowed down by altering the activity of defined brain circuits? I will test this intriguing hypotheis by performing chronic optogenetic control of inputs to the SNc and measure changes in the degeneration rate. 3. Is growth factor signaling directly contributing to dopaminergic neuroprotection and what are the timelines needed for neuroprotection? Previous experiments applied growth factors liberally in a non-specific fashion and/or with poor temporal resolution. I will develop optogenetic methods to achieve cell-type specific control of growth factor signaling so I can directly probe the protective role of growth factors in defined cell types, and especially cells prone to degeneration. These tools could also be applied to research beyond the nervous system since growth factor signaling is involved in key cellular phenomena such gene transcription that can impact the cell survival, differentiation, and function. Together these innovative projects will contribute to my long-term goals of building cellular resilience via neuromodulation and have a paradigm-shifting impact in neurodegeneration research.

DESCRIPTION (provided by applicant): Microbial opsins are light-sensitive proteins that can be expressed in specified cells via targeting promoters and turned on/off with millisecond speed, thus providing genetic and optical ('optogenetic') control of cell function with high spatial and temporal specificity. Their ability to control the electrical activity of neural circuits and confe reversible gain and loss of function of specific neuronal phenotypes allows us to study neural systems and diseases in unprecedented manner. Optogenetic research today, however, relies on a limited set of natural microbial opsins with broad activation spectra, limited ion selectivity and a narrow range of kinetics. We are proposing a novel approach to opsin engineering that capitalizes on the power of structure-guided protein engineering and directed evolution. In parallel, we will also search for novel naturally occurring opsins in niche environments. The expanded optogenetic toolkit will facilitate the investigation of neuronal circuits in health and disease.

DESCRIPTION (provided by applicant): During aging, motor function declines, with deficits in fine and fast movement and coordination. Experimental studies associate age-dependent motor deficits with the malfunction of dopaminergic (DA) pathway, which originates in the substantia nigra pars compacta (SNc). However we do not understand how the activity of DA neurons varies throughout aging in the different tiers of nigral neurons in vivo, what type of activity changes precede neurodegeneration, how these activity changes affect behavior, and whether restoring perturbed activity can delay neurodegeneration and/or behavioral deficits. To characterize, for the first time in the intact circuit, the function and anatomy of aging nigral dopaminergic circuits, we propose to use two powerful technological advances in neuroscience: one for cell-type specific bidirectional control of neuronal activity in vivo with high temporal precision (optogenetics); and one for intact brain circuit mapping and phenotyping, slicing-free (CLARITY). Optogenetics uses microbial opsins, light-sensitive proteins that can be expressed in specified cells via targeting promoters and turned on/off with millisecond speed, thus providing control of cell function with high spatial, temporal, and genetic specificity. Their abilty to control the electrical activity of neural circuits and confer reversible gain and loss of functin of specific neuronal phenotypes allows us to study neural systems and diseases in unprecedented manner. To target subsets of SNc DA neurons we will take advantage of the TH- Cre transgenic lines as well as localized stereotaxic opsin delivery and targeted light application We hypothesize that throughout aging, DA neurons in different SNc tiers have distinct behavioral contributions (Aim 1), which is due to differences in their intrinsic excitability (Aim 2) and changes in synaptic inputs (Aim 3). This proposal combines powerful complementary techniques (optogenetics, electrophysiology, and neuroanatomy by CLARITY) to advance our understanding of dopaminergic function and contribution to behavior throughout aging by performing studies in the intact circuit. The PI has been involved in the development of both techniques and our laboratory is ideally positioned to apply these techniques to the aging brain with a focus on the DA system. A better understanding of the properties of DA neurons in the aging SNc can aid in identifying circuit targets and/or behavioral/nutritional methods to delay/reverse age-related alterations in these neurons and in motor functions.

? DESCRIPTION (provided by applicant): Our bodies appear optically opaque because biological tissue scatters light strongly. Although advances such as multiphoton excitation have enabled deeper access for optical imaging by gating out scattered light, these strategies are still fundamentally limited to superficial depths (~ 1 mm). Yang's group at Caltech has pioneered time-reversal symmetry of optical scattering as a direct strategy to 'turn off' tissue scattering. n 2012, Yang's group demonstrated a time-reversal ultrasound-encoded (TRUE) focusing strategy based on the use of digital optical phase conjugation to flexibly and controllably deliver high optical power in ex vivo tissues. Here we propose to realize a digital TRUE focusing in vivo with rapid wavefront sensing and wavefront modulation. If successful, this novel approach will enable light focusing up to a depth of 4 mm in a living rodent brain, with a focal minimal width close to single-cell level (30 ¿m). This ability to render a tight laser focus within biological tissues can be translated into powerful new methods for functional imaging and manipulation of the brain. We can scan the focus spot to perform fluorescence, Raman, and other types of imaging. We can also use the focus spot to selectively ablate tissues with high precision. This technology will also enable non-invasive focused light delivery for optogenetics - a key application area that is the focus of our proposed research. The use of digital TRUE would enable the extension of optogenetic techniques to the deep brain for non-invasive, spatially specific, excitation/inhibition. For this project, we will complement the power of optogenetic control of defined brain circuits with real-time circuit activity feedback, via in vivo anaesthetizd recordings, to establish digital TRUE as a new, noninvasive optical tool for optogenetic studies. This proposed work represents a powerful enabling technology for optogenetics - potentially opening up new applications and new methods for optogenetics. In addition to optogenetics, digital TRUE promises broader impacts on biomedical research and diagnosis. Digital TRUE's unique capability to focus light in deep tissues holds tremendous potential in enabling in vivo deep tissue optical imaging and biochemical analysis. Although there are significant technical challenges to be tackled, our proposed project is an important and necessary step in advancing TRUE to reach its full potential.

? DESCRIPTION (provided by applicant): Type 2 diabetes (T2DM) results from impaired insulin secretion, insulin resistance, and increased hepatic glucose production with deficits in insulin secretion likely the key determinant of whether T2DM develops. However, we do not understand the cause of reduced ß cell mass and/or a/ß cell dysfunction in human T2DM. This is because models and hypotheses about T2DM ß or a cells generally arise from studies of rodent models or have not been confirmed in human samples. Furthermore, most studies on the T2DM human pancreas do not adequately incorporate the clinical phenotype of the patient or the disease stage and, consequently, combine profiles from different stages. In addition, previously studied T2DM pancreatic specimens have been collected in ways that do not completely allow newly available molecular analyses. To overcome these deficits in our knowledge, we have established a new infrastructure for procuring human T2DM pancreatic specimens for analysis by new technologies and isolating islets from the same pancreas. In addition, we have assembled an interdisciplinary team of scientists with human islet biology expertise and investigators who bring new technologies for tissue and cell profiling from the neuroscience and cancer arenas. Using this new infrastructure and technologies, we propose to: 1) Identify unique protein and RNA signatures in the pancreatic islets and isolated a and ß cells from clinically phenotyped T2DM donors of short (i.e. <5 years)- and long (i.e. >10 years)-duration with technologies such as RNA-sequencing, tissue-clearing, and single cell phenotyping. 2) Using tissue imaging mass spectrometry, identify differentially expressed lipids, proteins, and metabolites in T2DM islets. 3) Integrate molecular signatures with functional T2DM islet profiles and test the impact of candidate molecules from the lipid, metabolite, and/or mRNA/protein datasets on a and ß cell activity and viability. In keeping with the goals of the R24 mechanism, our discovery-based approaches will generate new resources and datasets, create new paradigms for a/ß cell dysfunction in human T2DM, and foster fundamental discoveries that will not only improve our understanding of how ß cell dysfunction/loss occurs, but potentially lead to therapeutic interventions.

? DESCRIPTION (provided by applicant): This application to join the Consortium on Beta Cell Death and Survival (CBDS) within the HIRN outlines new technologies, of value to all HIRN investigators and specifically requested in the RFA, to provide unprecedented detail in our understanding of the development, plasticity, and molecular signatures of the human pancreatic islet. While it is increasingly clear that rodent and human islets and ? cells have major structurl and expression differences, new data from our groups and others indicate that young (juvenile) human islets are substantially different from adult human islets. New data in humans suggests that the juvenile period (<5 years of age) is an incredibly important developmental period of substantial islet plasticity, but our knowledge of the human pancreatic islet in this tim period is quite limited. For example, the fetal and juvenile determinants of an individual's ?-cel mass, which varies by three- to five-fold in adult humans, are unknown, yet a person's initial ?-cell mass is likely deterministic of if or when one develops T1D. We know little about how the human islet becomes vascularized and innervated, but both are essential for normal islet and ?-cell microenvironment and function. Recent studies demonstrate that ?-cell-directed autoimmunity in individuals with genetic susceptibility to type 1 diabetes (T1D) appears within this same period of the first 5 years of life. The temporal overlap of these two processes generates our over-arching hypothesis that the onset of ?-cell- directed autoimmunity is causally related to ongoing alterations in islet or pancreas architecture, or individual expression states o endocrine cells, including ? cells, during this period of maturation. Previous limitations in the analysis of juvenile tissue result partly from difficulty in procuring human pancreas tissue, but also because standard methods are limited in their resolution and number of markers that can be concurrently analyzed. To address these limitations, our interdisciplinary team brings advanced tissue-clearing technology and multiplexed RNA in situ tissue-analysis approaches, plus substantial experience in studying the human pancreas over the fetal to human juvenile periods, wherein we made essential discoveries about human- specific aspects of the birth and differentiation of endocrine cells, and their assembly into functional islets. The assembled research group incorporates expertise in human and mouse pancreas developmental biology, islet function, and brings in superior methods of analysis towards the production of an atlas of structure and function of the human pancreas over the juvenile period. We propose to: (1) Define the expression of key cell-surface, physiological markers, transcription factors, and other factors to define the sequence of events in human juvenile pancreatic islet development; (2) Define the intra- and inter-islet vascular and neural networks, and these 3D structure-function relationships at the cellular and organ-wide level; (3) Integrate data from Aims #1 and #2 to create a comprehensive profile of the human juvenile pancreatic islet during the first 5 years of life.

ABSTRACT Cardiovascular diseases such as heart failure, arrhythmias, and hypertension are leading causes of morbidity and mortality in the United States and world-wide. The autonomic nervous system plays a critical role in the pathophysiology of these diseases and neuraxial modulation provides an important avenue for therapeutic intervention. The major goal of our research team is to precisely define the cardiac neural hierarchy and develop circuit diagrams from the macroscopic to cellular and molecular levels and share these data on an ongoing basis with the scientific community. This effort will also provide verified methods and tools for assessing neuromodulation. The research team will make them freely available to the scientific community. A multiscale, multidisciplinary approach across various species, highly relevant to human disease, will be used to define the anatomy of cardiac innervation in high definition. Neural structure will be linked to cardiac function. The complexity of cardiac neural control necessitates an integrative approach that will represent a tour de force in this field. State-of-the-art anatomical, physiological, and pharmacological approaches from `cells to man' must be combined in order to achieve the above goals. This approach will be utilized at each level of the neuraxis (heart, extracardiac intrathoracic neural structures and extrathoracic neural structures). The techniques proposed will allow, for the first time, a detailed description of the anatomical and molecular interactions at the synaptic and cell body levels in cardiac and extracardiac ganglia. The techniques used and the integration of these pathways represents the most innovative attempt to understand cardiac neural control ever undertaken. Understanding these pathways has the potential to accelerate development of therapies that will be able to precisely target neural structures and also guide methods to re-purpose already available therapies (e.g. nerve stimulators) for therapeutic purposes. Ultimately, these approaches are required to develop novel, effective, and affordable interventions for the management and prevention of heart disease and sudden cardiac death.

In order to understand how neural signals propagate to conduct specific functions in behaving animals and how individual neurons are physically connected in the context of behavior, advanced tools should be available at the hands of neuroscientists. The Multimodal Integrated Neural Technologies (MINT) hub aims to develop and provide tools that are able to read from and modulate neurons at multiple sites independently at high spatial and temporal resolutions. The hub will disseminate tools and methods to correlate the recorded cell activity with the structural connection. In this way, the connectivity of active cells can be visualized, labeled, and traced for detailed functional mapping. The mission of the MINT hub is to provide a collection of tools, synergistically developed, integrated, and available to the neuroscience community, to address one theme: connecting neurophysiology and structural analysis with a greater scale and resolution. The synergistic integration of these neurotechnology tools at the MINT Hub would accelerate the rate of discovery in neuroscience. This in turn can be expected to pave the way to improved treatments for neurological disorders and to breakthroughs in artificial intelligence, especially neuromorphic computing. The MINT hub will provide annual training workshops for new users to be familiar with new technologies and able to use them effectively. To achieve sustainability, the hardware tools will be actively marketed to the community and those with sustainable volume will be transitioned to commercialization partners. Importantly, this program will cross-train neuroscience and technology personnel during the course of this program, resulting in preparation of a new generation of multi-disciplinary engineers and scientists.

This hub uniquely combines high-density electrodes, chemical sensing, optical stimulation, and cell labeling. Fiberless high-density optoelectrodes can allow optical stimulation of individual or few neurons with high specificity and selectivity using monolithically integrated micro-LEDs or optical waveguides on multi-shank silicon probes. Carbon microthreads will be used to create advanced arrays that will dramatically increase the ability to record from interconnected neurons and label those cells with high accuracy. Advanced metal alloys will also be used to greatly enhance the signal-to-noise ratio of miniaturized electrodes. The MINT hub will innovate viral vector delivery and tissue clearing in the nervous system and combine these with multispectral labeling for intact cell phenotyping. Furthermore, an open-source software will be developed to improve the accuracy and efficiency of anatomical reconstruction for creating connectivity maps. The MINT hub will validate the developed tools and methods in three in-vivo experiments to exemplify what can be accomplished when the proposed modalities and methods are synergistically integrated. This NeuroTechnology Hub award is co-funded by the Division of Emerging Frontiers within the Directorate for Biological Sciences, and the Division of Chemical, Bioengineering, Environmental & Transport Systems within the Directorate for Engineering as part of the BRAIN Initiative and NSF's Understanding the Brain activities.

SUMMARY In order to understand how the CNS encodes, modifies, stores and retrieves information it is necessary to explore the diverse cell populations that comprise the CNS. There is an emerging consensus that the CNS cannot be satisfactorily understood solely as a collection of circuits1. One significant missing aspect in our collective strategy to comprehensively understand the CNS is the largely unmet need to understand additional cell types such as astrocytes1. Astrocytes represent around 40% of all CNS cells and are found throughout the brain. Their close proximity to neurons has been known for over a century. It is now well established that astrocytes serve vital support roles including buffering of K+ around neurons, clearing neurotransmitters from synapses as well as providing nutrients. Astrocytes may also regulate blood flow to meet demands set by neuronal activity. In addition to these varied supportive roles, increasing evidence suggests that astrocytes regulate neuronal function via synapse formation, synapse removal, and regulation of synaptic function through uptake and release of neuromodulators and neurotransmitters. In addition, astrocytes are proposed to engage in bidirectional communication with neurons in a Ca2+-dependent manner, which in some circumstances involves bidirectional ATP signaling. However, despite progress, experimental studies of astrocytes have lagged behind those of neurons by decades, largely because twentieth century neuroscience was dominated by the emergent field of electrophysiology that provided a precise and valuable way to study electrical activity in neurons and its relationship to neural circuit function and behavior. In contrast, astrocytes do not fire action potentials or display any other type of propagated electrical signals, and thus electrophysiology was ill suited to study these cells. As a result, our understanding of astrocytes, their identity, diversity and dynamics is still in its infancy. We seek to capitalize on recent breakthroughs in our laboratories to advance tools that will allow neuroscientists to study in detail the molecular make-up of astrocytes in different brain areas at multiple levels from gene expression, to proteins (Aim 1), to physiology within neural circuit functions in vivo (Aim 2). We will also provide tools to target astrocytes in a selective and non-invasive manner by gene delivery across the blood-brain-barrier (Aim 3). Our overarching hypothesis is that the availability and open dissemination of new, selective tools to study astrocytes at molecular, cellular and circuit levels of investigation may reveal insights about the CNS as striking and as influential as those revealed by early measurements of electrical signals in neurons. Furthermore, the free dissemination of such tools will catalyze additional advances in the context of physiology and brain disease.

PROJECT SUMMARY The use of current and emerging genetically encoded tools could greatly benefit from advanced methods for gene delivery to the desired cell population. When used in conjunction with transgenic animals to restrict expression to cell populations of interest, adeno-associated viruses (AAVs) can provide well-tolerated and targeted transgene expression that enables long-term behavioral, in vivo imaging, and physiological experiments. Lacking from the current suite of vector tools is a way to achieve cell- or circuit-specificity with AAVs without the use of transgenic animals. We pioneered a powerful strategy that allows for the generation and selection of viral vectors with optimized properties by Cre-recombination-dependent AAV targeted evolution (CREATE). We have used CREATE to evolve AAVs that are capable of crossing the blood?brain barrier (BBB) and transducing most cells in the adult brain. These systemically delivered AAVs enable noninvasive CNS-wide transduction of specific cell types and regions in rodents when used with gene regulatory elements. We propose to build upon our success with the CREATE method to develop a suite of systemic viral vector tools and approaches that will enable cell-state monitoring in defined cell populations/circuits and noninvasive modulation of complex behaviors in wild-type animals. Methods for noninvasive modulation of specific circuits need to couple actuators gated by highly penetrant moieties (BBB-permeant ligands, ultrasound, etc) with noninvasive delivery methods that have a brain-wide reach, yet can confine actuator expression to specific circuits. The systemic AAVs we developed provide a solution to the latter. We will: develop AAV genomes for expression of genetically encoded calcium indicators in specific cell types without transgenesis (Aim 1); engineer AAV capsids capable of anterograde trans-synaptic trafficking (Aim 2); and provide validation for noninvasive control of behavior in transgenic and wild-type animals (Aim 3). These novel vector reagents and protocols will be distributed, upon validation, to the neuroscience research community by our established resource center (www.clover.caltech.edu). We expect that these vector reagents will provide new avenues for studying and ultimately treating the vertebrate nervous system. Potential uses for the systemic viruses are: circuit mapping; fast screening of gene regulatory elements; and genome editing with CRISPR-Cas9. In addition, when paired with appropriate activity modulator genes, the new AAVs could enable noninvasive deep-brain modulation. Importantly, the broadly efficient and mosaic expression strategies developed and validated in this project will be compatible with other sensors and actuators developed for circuit studies, even, long-term, across species.

Viviana Gradinaru, Caltech With the advent of technologies such as CRISPR/Cas9, genome engineering for both basic research and therapeutic applications is becoming reality. An outstanding challenge is the mean to safely and efficiently transfer large genomes to desired cells across life span. We have developed an in vivo Cre-based selection platform (CREATE) for identifying adeno-associated viruses (AAVs) that efficiently transduce genetically defined populations. We used CREATE to select for viruses that transduce the brain after intravascular delivery and found a vector that nonspecifically transduces most cells across the adult brain. Since the restrictive nature of the blood brain barrier presents a major impediment toward treating CNS disorders our discovery has the potential to enable exciting advances in gene editing/replacement via CRISPR-Cas or RNA interference to restore diseased CNS circuits if the needed level of efficiency and specificity can be engineered for diseased targets. We plan to enable such efforts by creating viral-based solutions to non-invasive whole- brain large cargo delivery across the blood-brain barrier from embryo to adult by: 1. Generating AAVs for cell-type and region specific gene delivery across the blood-brain- barrier, noninvasively via the bloodstream in the adult rodent for neurodegeneration applications. 2. Generate AAVs capable of transducing the developing brain in utero with a simple systemic injection to the pregnant dam for neurodevelopment research and therapy. 3. Increase the packaging capability of AAVs by about 2-fold to enable delivery of large genomes for gene therapy and research. 4. Enable non-invasive circuit specific deep brain modulation by the use of systemic vectors and genetically encoded activity modulators (e.g. by chemogenetics or others in development now). Longer term we plan, in our laboratory and also with collaborators, to contribute our neurotechnologies (including, in addition to viral vectors, tissue clearing and optogenetic control and imaging) towards elucidating maladaptive neural circuits that contribute to brain pathology in neurodegeneration and neurodevelopment.