Collaborative Seed Grants
The HBI Collaborative Seed Grants Program stimulates and supports research projects linking laboratories at Harvard Medical School (HMS) and the Harvard University Faculty of Arts and Sciences (FAS). Such creative collaborations are central to HBI’s mission. This program is possible through generous support from the Alice and Rodman W. Moorhead III Collaborative Grants Fund for HBI, the Andrew B. Suzman Research Fund for HBI, the Thomas H. Lee Fund, and an allocation from the President’s Discretionary Fund. The grants make possible a broad range of fundamental neuroscience research that may yield fundamental insights into the mechanisms underlying brain disorders.
2018-2019 Awardees:
Developmental machine learning and the origins of inferotemporal organization
Talia Konkle, PhD
Assistant Professor of Psychology, Harvard University
Margaret Livingstone, PhD
Takeda Professor of Neurobiology, Harvard Medical School
In adult humans and macaques, inferotemporal cortex (IT) is systematically organized into functional regions that represent specific kinds of entities, such as faces, objects, and places. However, there is active debate about the nature of the representations underlying this organization, and how this organization develops. For example, does tuning in face-selective regions represent only faces or does it participate in a broader shape-coding scheme for all other kinds of objects? Do regions that will later become face-selective have any pre-specified biases are they entirely learned through visual experience? The present work will take a computational approach to these questions, leveraging deep neural networks to examine the nature of face-selective neural responses and the developmental trajectory of their tuning. By controlling what is built in to the network structure, and what the deep net gets as input, this work aims to provide a computational existence proof for the kinds of functional tuning properties that can manifest with and without specific visual experience.
Regulated proteome remodeling during neurogenesis
Lee Rubin, PhD
Professor, Department of Stem Cell and Regenerative Biology, Harvard University; Director of Translational Medicine, Harvard Stem Cell Institute
Wade Harper, PhD
Chair, Department of Cell Biology, Harvard Medical School
Bert and Natalie Vallee Professor of Molecular Pathology, Harvard Medical School
Neuronal development involves lineage-dependent alterations in cellular states to create cells with distinct morphological, functional and synaptic properties. These states emerge from pluripotent stem cells often through a series of steps, which drive particular lineages to their final differentiated state. Changes in cell state can be viewed as being orchestrated through alterations in the key signaling networks that determine the identity of individual cell lineages. Fundamentally, a change from state A to B can be viewed as an alteration in the constellation of proteins that support the identity of each state – generally termed “proteome remodeling”. While a state change may be initiated through alterations in the transcriptome, mRNA abundance itself is typically a poor predictor of protein abundance and the extent to which proteomes are remodeled. In this collaborative pilot project, we will explore the role of proteome remodeling via the ubiquitin-proteasome system during the conversion of embryonic stem cells to individual neuronal cell types. We will focus on a superfamily of ubiquitin modifying enzymes in order to elucidate how these enzymes are regulated during neurogenesis in vitro and how the activity of these enzymes control proteome content. This work will use advanced gene-editing approaches to create cells lacking individual ubiquitin modifying enzymes and will employ quantitative mass spectrometry methods measure proteome content and modification events in distinct cell states, thereby revealing the extent to which proteome remodeling contributes to individual cell states. We anticipate that this work will inform numerous biochemical processes that may go awry in neurodevelopmental disorders.
Topological and geometrical order in neuronal bundles
L. Mahadevan, PhD
Lola England de Valpine Professor of Applied Mathematics, Harvard University
Professor of Physics and of Organismic and Evolutionary Biology, Harvard University
Maxwell Heiman, PhD
Assistant Professor of Genetics, Harvard Medical School
Principal Investigator, Boston Children’s Hospital
The cells of our nervous system communicate with each other through wire-like projections called axons and dendrites that are braided together into nerve bundles. While extensive attention has been paid to the “wiring problem”, i.e. how nerve cells project to the correct targets, much less consideration has been given to the “tangling problem”, i.e. how projections are arranged within each bundle. The latter is important because, if nerve bundles are not carefully organized, the nerve cells may become tangled and twisted and thus be prone to constrictions, breakage and dysfunction. This HBI Seed Award supports a collaboration between an applied mathematician and physicist (Mahadevan) and a developmental neurobiologist (Heiman) to develop methods to quantify the organization of nerve bundles using concepts from topology and geometry such as Twist, Link, and Writhe. The outcome of the project will be a quantitative framework for analyzing tangling within nerve bundles that can eventually be applied to the large
datasets that are beginning to emerge from the brain imaging initiative.
General principles of olfactory processing from insects to rodents
Aravinthan Samuel, PhD
Professor of Physics, Harvard University
Sandeep Robert Datta, MD, PhD
Associate Professor of Neurobiology, Harvard Medical School
Sensory systems gather up information about the outside world, and then package that information so the brain can usefully interact with the environment. In representing the outside world, the brain has to balance two fundamental and competing computational demands. On the one hand, it needs to be able to tell objects in the world apart. Survival demands that fresh meat be distinguished from rotten, that potential mates are distinguished from potential predators. On the other hand, it is essential that the brain be able to identify similarities amongst objects to be able to generalize — to know, for example, that because a fresh orange is delicious it is likely that a fresh tangerine is as well. But these two imperatives stand in conflict — the harder your brain works to tell things apart, the more difficult it is for your brain to understand how objects are related. How does the brain balance these two competing demands to allow us to distinguish objects in the world while at the same time understanding how objects might be usefully related? We propose to explore how the neural wiring charged with processing information about smells accomplishes this task. We will do so in two different model organisms — the fruit fly and the mouse — where we will use complementary tools, ranging from high-resolution electron microscopy to live circuit imaging to explore how smells are encoded in the parts of the brain responsible for perception and behavior. Because the networks that encode smells are emblematic of networks responsible for memory and for multimodal integration, it is likely this work will lead to general lessons about how the brain addresses the problem of understanding the external world.
Identification of a nitric oxide receptor and the underlying sensory mechanisms
Yun Zhang, PhD
Professor of Organismic and Evolutionary Biology, Harvard University
Joshua Kaplan, PhD
Professor of Neurobiology, Harvard Medical School
Department of Molecular Biology, Massachusetts General Hospital
Nitric Oxide (NO) is one of the most common environmental cues that terrestrial animals encounter. NO in the air is produced by lightning or heat. It can also be released into the air by animals that produce NO. In mammals, NO regulates key physiological events, such as vasodilation, inflammatory response, and neurotransmission. Defects in these processes are associated with various human diseases. While the function of NO in regulating physiological events has been well characterized, before our recent study it was not known whether animals detect NO in their environment as a sensory cue and generate behavioral responses to it.
Using a worm (C. elegans) as a model, the Kaplan and Zhang labs showed, for the first time, that NO in the environment is detected by sensory neurons to mediate behavioral response and that the sensing of NO is mediated by a receptor guanylate cyclase (rGC) DAF-11 and a cyclic nucleotide-gated (CNG) channel TAX-2/4 (Hao et al., 2018). The proteins that comprise the worm’s NO sensing pathway are strongly conserved in other organisms, including humans. Thus, our results suggest that NO produced endogenously or by bacterial pathogens may also elicit potent sensory responses, which could exacerbate inflammatory responses associated with infection. In this project, the Kaplan and Zhang labs test the idea that the DAF-11 rGC is directly activated by NO –and if so, we will determine how NO-activates this receptor.
2016-2017 Awards:
The Role of DNA Breakage and Repair in Generating Neural Diversity and Disease
Frederick Alt, Department of Genetics, HMS
Xiaoliang Sunney Xie, Department of Chemistry and Chemical Biology, FAS
The 80 billion neurons of the adult human brain develop from rapidly dividing neuronal progenitors cells. Recently, the Alt lab found that the DNA of numerous genes that encode proteins critical for mouse brain circuitry is highly susceptible to breakage in the progenitor cells. To maintain chromosomal integrity as they proliferate, neuronal progenitors repair breaks in their genes by rejoining their ends or, sometimes, by joining together two breaks in the same gene and deleting DNA in between. Based on the distribution of breaks across this set of neural genes, we hypothesize that joining two breaks within them could functionally alter encoded proteins, potentially contributing to diversification of brain circuitry and function. Also, most frequently breaking neural genes also have been implicated in neuropsychiatric disorders and/or cancer. In this regard, we propose that improperly joining breaks in these genes could contribute to such diseases by inactivating the genes or adversely altering their function. The Alt and Xie labs will combine their complementary expertise to test these hypotheses by elucidating the genes that similarly break and join in human neuronal progenitors and determining whether diversity generated by this process is developmentally propagated to mature brain neurons.
Genes that Regulate Sleep Across Species
Dragana Rogulja, Department of Neurobiology, HMS
Alexander Schier, Department of Molecular and Cellular Biology, FAS
When we fall asleep, our relationship with the world changes dramatically: we no longer respond to most external stimuli, unless they are particularly strong or salient. Our goal is to understand how the brain builds a gating system that prevents most of the sensory information from arousing us during sleep. Specifically, we will look for genes that control information flow. Because sleep is a conserved process, we will use flies and zebrafish in our approach. Flies are an excellent system for identifying new genes, and zebrafish are vertebrates whose brains are very similar to our own. We will do rapid screening for new gating genes in the fly, and test their function in the zebrafish. Our combined approach will reveal new and conserved principles of sleep regulation.
Neurocomputational discovery of visual structure
Jan Drugowitsch, Department of Neurobiology, HMS
Samuel Gershman, Department of Psychology, FAS
Visual experience is alive with motion, but little is known about how the brain transforms patterns on the retina into such richly structured object motion. We approach this problem by combining behavioral, computational and clinical techniques. Structured object motion should eases tracking multiple objects simultaneously, which we will first study in healthy subjects. At the same time, we will explore theoretically how neurons could represent and operate with such structure. Finally, we will investigate if and how structured object tracking breaks down in autism, a disorder that has been previously associated with object tracking deficits. This work will provide initial insights into the neural representation of visual motion structure and open the door towards understanding how the brain represents structure in general.
Probing Cell Fate in the Central Nervous System
Constance Cepko, Department of Genetics, HMS
Sharad Ramanathan, Department of Molecular and Cellular Biology and School of Engineering and Applied Sciences, FAS
Our goal is to understand how cells in the developing central nervous system acquire their identities. How exactly does a cell integrate information from a variety of sources to ultimately choose what type of cell to become, i.e. a motor neuron or a photoreceptor? We will use as a test case the choice of developing neurons in the retina to become a rod vs. a cone photoreceptor cell. This is a key choice for cells to make, as cones detect and process light during the day to give us our high acuity, color vision, and rods detect dim light to give us great sensitivity at night. Our collaboration will bring together the generation and computational analyses of genome wide data and high throughput functional experiments to understand these fate choices. These methods and the resulting findings will provide a road map for understanding other cell fate decisions within the CNS, as well as a protocol for the generation of photoreceptors from stem cells. In particular, given the importance of cones for daylight and color vision, and their loss in many forms of retinal degeneration, the ability to create cones vs. rods may enable new, potentially more robust treatment strategies for certain eye diseases.
2015 Awards:
Molecular Profiling of Neuronal Diversity in the Human Central and Peripheral Nervous Systems
Paola Arlotta, Department of Stem Cell and Regenerative Biology, FAS and HMS
Lisa Goodrich, Department of Neurobiology, HMS
Our long term goal is to understand how neurons differ at the level of their genetic material. This will point us towards specific drugs that might prevent neuronal loss, and provide us with the blueprints needed to engineer the exact type of neuron needed to repair the brain and restore function. We will devise and optimize a new method for purifying and characterizing defined types of neurons from the central and peripheral nervous systems. This method will enable researchers to access clinically relevant populations of neurons from any region of the nervous system, thereby laying a valuable foundation for the development of new drugs and therapies to prevent or treat diseases like Parkinson’s, ALS, and a wide range of other neurodevelopmental and neurodegenerative disorders.
The Role of the Dopaminergic System in Interval Timing
John Assad, Department of Neurobiology, HMS
Nao Uchida, Department of Molecular and Cellular Biology, FAS
Timing is central to many aspects of our behavior, such as associative learning and decision-making. We hypothesize that altering the effect of the neurotransmitter dopamine should affect the ability to keep track of time. We will test this hypothesis using mice that are trained to estimate short intervals of a few seconds. We will use a new technique, optogenetics, to use light to specifically alter the function of neurons in the brain that release dopamine, thus altering the critical balance between the direct and indirect pathways. These experiments will shed light on the normal function of cortical-basal ganglia brain circuits, which will in turn inform our understanding of the brain-circuit aberrations in psychiatric disorders such as schizophrenia.
Behavioral Neurogenetics: Deep Phenotyping of Genes Regulating Social Approach
Sandeep Robert Datta, Department of Neurobiology, HMS
Hopi Hoekstra, Departments of Organismic and Evolutionary Biology and Molecular and Cellular Biology, FAS
We have recently developed a new technology that gives us objective insight into the structure of mouse behavior, which we call “Deep Phenotyping”. We propose to apply this technique — coupled with modern advances in gene sequencing — to characterize complex patterns of social behavior in a variety of mouse strains and to identify individual genes responsible for a particular behavior called “social approach”. These experiments have the potential to fundamentally transform both the theory and practice of behavioral neurogenetics, and will significantly impact our understanding of the mechanisms that underlie social interactions.
Oxygen, Odors and Brain Serotonin
Susan Dymecki, Department of Genetics, HMS
Venkatesh Murthy, Department of Molecular and Cellular Biology, FAS
Breathing not only supplies life-sustaining oxygen to the body, but also brings in smells (odorants) that modify critical behaviors such as feeding and mating. Yet, these demands for oxygen versus odor information are rarely in sync. How are these demands processed by our sensory system and brain so that the necessary, even life-sustaining, outcome is achieved? Results of this project are expected to inform not only basic neurophysiology but also disorders ranging from apnea syndromes (e.g. apneas of prematurity, the sudden infant death syndrome, and adult sleep apnea syndromes) to asthma to infant and adolescent eating and feeding disorders.
Investigating the Maturation of Connectivity within a Neural Circuit
Aravinthan Samuel, PhD, Professor of Physics, Department of Physics, FAS, and Center for Brain Science
Rachel Wilson, PhD, Professor of Neurobiology, HMS; Investigator, HHMI
A basic question in neuroscience is how a young brain develops into a mature brain. This problem is relevant to all developmental disorders of the nervous system. Previous studies have addressed this problem from many perspectives; however, no studies have ever examined how individual identifiable neurons and synaptic connections are altered during development. We aim to completely reconstruct the connectivity of a particular region of the fly brain (the antennal lobe). At the end of one year, we aim to have maps of the adult and juvenile circuits, as well as electrical activity patterns from both circuits. This one-year collaborative pilot project should serve as a template for efforts to understand brain development in larger and more complex organisms, as the technology in this field allows us to map ever larger volumes.
Banner image shows Brainbow staining of a section of brain tissue, courtesy of the labs of Takao Hensch and Jeff Lichtman