Ben A. Barres - US grants

DESCRIPTION (provided by applicant): We propose to investigate why central visual pathways fail to regenerate after injury and how their repair can be enhanced. We will focus on why retinal ganglion cell axons fail to regenerate after optic nerve injury. Many studies point to multifactorial cause of retinal ganglion cell axon regenerative failure. On the one hand adult retinal ganglion cells have an intrinsically limited axon growth potential. On the other, the optic nerve environment after nerve injury is strongly inhibitory to regenerating axons as a result of inhibitory cues deriving from both degenerating myelin and reactive astrocytes. We do not yet have an in depth understanding of the molecular basis of these phenomena. The proposed experiments address a longstanding central question in the field, which is why degenerating myelin, which is strongly inhibitory to axon regeneration, is cleared robustly by phagocytosis after PNS injury but not after CNS injury. Schwann cells play a critical role in clearning degenerating PNS myelin but the relevant phagocytic pathways are not yet known. We have recently discovered that Schwann cells in the PNS and astrocytes in the CNS express the same repertoire of phagocytic pathways including the Mertk/Axl and the Megf10/LRP1 pathways. This is surprising as only Schwann cells perform robust myelin clearance after nerve injury. Furthermore, we have established that at least one of these pathways, the Mertk pathway, is required for clearance of PNS myelin debris. The proposed experiments are designed to elucidate the molecular mechanisms of Schwann cell-mediated myelin debris clearance and to investigate why these same pathways do not mediate myelin clearance after CNS injury. Using conditional knockout mice, in which each pathway is deleted specifically in Schwann cells, we will test the hypothesis that the Mertk/Axl and Megf10/LRP1 phagocytic pathways are required for glial-mediated clearance of degenerating (inhibitory) PNS myelin. We will then test several hypotheses for why astrocyte phagocytosis of myelin is disabled in the injured optic nerve, either because the required phagocytic pathways are downregulated in astrocytes after optic nerve injury or because degenerating CNS myelin is resistant to glial-mediated phagocytosis. Our ultimate goal is to determine how phagocytic clearance of degenerating myelin can be activated in optic nerve astrocytes in order to develop new treatments to promote retinal ganglion cell axon regeneration after injury in ocular diseases including glaucoma, retinal ischemia, optic neuritis, and optic neuropathy.

DESCRIPTION (from applicant's abstract): Synapses throughout the brain are ensheathed by glial cells, which provide nutrients and help to terminate the action of neurotransmitters. It is not known whether glia also actively regulate the development or function of synapses. We have developed methods for the isolation and culture of highly purified populations of CNS neurons and glia. We are using these methods to ask two questions: Do synapses between neurons form in the absence of glial cells? Do glial cells regulate synaptic transmission? Our preliminary findings demonstrate that ultrastructurally normal synapses form in the absence of glia. However, in the absence of glia, several proteins necessary for vesicular release fail to accumulate in presynaptic terminals. Physiological studies and FM1-43 imaging indicate that there is also a very low presynaptic efficacy with little vesicle recycling occurring in response to presynaptic depolarization. Thus, in the absence of glia, synapses form that are biochemically and functionally immature. Glial cells secrete a protein that enhances synaptic protein accumulation and increases presynaptic efficacy by more than ten-fold, as shown by an increase in the frequency of miniature excitatory postsynaptic currents, evoked transmitter release, and vesicle recycling using FM1-43. Taken together, these preliminary findings indicate that glial cells strongly promote presynaptic maturation. In this proposal, we will ask: (1) By what type of biochemical mechanism do glia enhance presynaptic efficacy? (2) By what functional mechanism do glial cells enhance presynaptic efficacy? (3) Do glial cells enhance presynaptic calcium currents? (4) Which synaptic proteins are elevated by glial cells? and (5) What is the identity of the glial-derived protein that enhances presynaptic efficacy? Methods that we will use to answer these questions include patch-clamp recording, immunostaining, FM1-43 imaging, semiquantitative Western Blotting, gene chip profiling, biochemical purification and expression cloning. We will continue to use a culture system that allows highly purified retinal ganglion cells (RGCs) to be maintained for extended periods of culture under serum-free conditions in the presence or absence of glia. RGCs serve as a good model system for CNS glutamatergic neurons and are presently the only defined CNS neuron type that can be highly purified and cultured in the absence of glial cells with high survival. Our ultimate goal is to understand whether glial cells normally regulate synaptic function in vivo. If so, glial regulation of synaptic function may participate in learning and memory or be perturbed by reactive gliosis after brain injury.

[unreadable] DESCRIPTION (provided by applicant): We propose to investigate the functions of mature astrocytes, a major class of glial cells in the mammalian central nervous system. We presently know a great deal about the properties of immature astrocytes, which can be easily isolated from neonatal rodent brains and cultured. Study of more mature astrocytes has been hindered by the inability to purify them and because these more mature astrocytes rapidly die in culture. Recently we have developed a method to isolate mature, protoplasmic astrocytes from postnatal mouse cerebral cortex. We have found that unlike the immature astrocytes, the mature astrocytes rapidly undergo apoptosis unless they are cultured together with endothelial cells. Using this new culture preparation, we propose to begin to investigate the functions of mature astrocytes. First, we will investigate the signaling mechanisms by which endothelial cells promote protoplasmic astrocyte survival. Second, we will investigate the functions of mature astrocytes purified from healthy mouse brains, by asking whether they promote neuronal survival, synaptogenesis, or myelination. Third, we will ask whether mature astrocytes by themselves can induce and maintain the blood-brain barrier and, if not, whether other neural cell types such as pericytes or neurons collaborate with astrocytes. Finally, we will investigate the relationship between healthy and reactive astrocytes. We will purify reactive astrocytes and compare their functions as well as their gene expression patterns by microarray profiling. We will then investigate the mechanisms by which normal astrocytes are induced to become reactive and whether the reactive astrocyte phenotype is stabile or can be induced to revert to a normal phenotype. These studies have potential to shed new light on the mysterious role of astrocytes in health and disease. For instance, understanding the blood-brain barrier may help us to develop new ways to deliver drugs into the brain. Understanding reactive gliosis may also lead to new treatments, as it contributes to the pathophysiology of neurodegenerative diseases, stroke, epilepsy, and limits regenerative CNS repair. [unreadable] [unreadable]

[unreadable] DESCRIPTION (provided by applicant): The Gordon Conference in Neural Development has become a key meeting in the larger field of neuroscience. From its inception in 1981, this biannual meeting has attracted superb speakers, excellent students, postdoctoral fellows, and a wide range of active scientists from junior faculty to senior leaders in the field. Sessions featured in this year's meeting include: neural stem cells, cell specification, neuronglial interactions, brain patterning, polarity, axon guidance, synaptogenesis, and development of neural circuits underlying behavior. Studies from both invertebrate and vertebrate model systems will be presented. A diverse range of speakers in terms of age, sex, race, and ethnicity will participate. About 30% of the speakers are women. We will continue the tradition started last meeting of reserving an entire section for late-breaking news. Several of the speakers work in areas of direct clinical relevance, including studies on stern cells, antidepressant mechanisms, axon growth, critical period plasticity, and neural circuits underlying behavior. The talks will be held in morning and evening sessions daily for 5 days. Speakers will be given about 40 minutes for their presentations, which includes a 15 minute discussion period. Afternoons will be open for informal interactions. Poster sessions will be scheduled for afternoons and will remain posted during the social periods following the evening presentations. A diverse group of participants will be selected. We will actively recruit minority students to attend the meeting and let them know that funds are available to support their attendence. Participants will be encouraged to present posters, which serves as an excellent training function, and further facilitates the interaction of students and postdoctoral fellows with faculty. To further stimulate the participation of students and fellows, we will ask that the first question after every talk be from a student or fellow rather than a faculty member. This mechanism helps to encourage participation by otherwise shy students. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Synapses are specialized cell adhesions that are the fundamental functional units of the nervous system, but the extracellular signals that induce CNS synapse formation are poorly understood. By using highly purified populations of neurons and glia, we have recently observed that the number of synapses on retinal ganglion cells (RGCs), as well as other types of CNS neurons, in culture is enhanced 7-fold by soluble signals released by astrocytes. These data add to the growing evidence that extracellular signals powerfully regulate CNS synaptogenesis. In our preliminary studies, we have identified 2 proteins that appear to play a crucial role in astrocyte-enhanced synaptogenesis. First, we have found that thrombospondin (TSP), a large matrix-associated protein released by astrocytes in vitro and in vivo, is sufficient to induce the astrocytes, and is also sufficient to induce the formation of structural synapses in vitro, and is necessary for astrocytes to induce synaptogenesis in vitro. Second, we have found that the complement protein C1q is highly upregulated in neurons by astrocytes, and is also sufficient to induce the formation of structural synapses in vitro. Both proteins are localized to synapses throughout the developing brain. TSP and C1q are thus among the first few identified soluble proteins sufficient to trigger formation of structural synapses between CNS neurons in vitro. We propose to further investigate the molecular basis of TSP and C1q in inducing synaptogenesis in vitro and then use this knowledge to directly test the hypothesis that they also help to control synaptogenesis in the developing visual system. Understanding the mechanisms that regulate synaptogenesis and synaptic plasticity are crucial to understanding the neural basis of learning and memory, Alzheimer's disease, drug addiction, and epilepsy.

[unreadable] DESCRIPTION (provided by applicant): What are the molecular mechanisms that control the formation of visual circuits? Throughout the mammalian central nervous system, the most salient structural correlate of synaptic specificity is laminar specificity: neurons confine their axonal and dendritic arbors to particular layers and thereby, synaptic partners, within a given target. In this proposal, we will focus on understanding how alpha and beta retinal ganglion cell (RGC) types form their precise synaptic connections in the deep and superficial layers of the superior colliculus respectively. Precise wiring of RGCs is critical for proper motion and pattern visual perception, but the molecular mechanisms that dictate how these two major retinal ganglion cell classes connect to their appropriate laminar target neurons are still mysterious. We hypothesize that the genes responsible for laminar-specific synaptic choices of RGCs will be selectively expressed by these two main classes of RGCs during the developmental period when their specific laminar connections are forming. In our preliminary studies, we have identified specific markers of alpha and beta RGCs as well as two new mouse strains that express green fluorescent protein specifically in each of these RGC classes. We will use these mice to address the following questions: (1) What is the genetic profile associated with these two major classes of functionally distinct RGCs?, and (2) What are the molecular cues that direct axons arising from these functionally distinct classes of RGCs, into anatomically distinct layers within their major target, the superior colliculus? We will then use this information to address a longstanding question about the development of laminar specificity: do RGCs form connections in their target laminae that are initially precise or instead do they form connections that are initially diffuse and then eliminate inappropriate connections? Our ultimate goal is to understand how precise visual synaptic connections form during development and to extend those findings into an understanding of how to induce visual system connections to regenerate properly after injury in ocular diseases including glaucoma, retinal ischemia, and optic neuritis. An understanding of the molecular mechanisms that control the formation of visual circuits will allow us to develop new treatments to promote their repair and regeneration in order to restore vision in glaucoma, optic neuropathy, and after injury. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Synapses are specialized cell adhesions that are the fundamental functional units of the nervous system, but the extracellular signals that induce CNS synapse formation and function are poorly understood. We have been investigating the role of astrocytes in the formation and function of excitatory synapses in vitro and in vivo. Using retinal ganglion cells (RGCs) as a model CNS neuron, we recently found that astrocytes release several proteins that strongly enhance the formation and function of excitatory synapses onto RGCs. We identified thrombospondins as astrocyte-derived proteins that normally help to promote CNS synaptogenesis in vivo and are sufficient to induce ultrastructurally normal synapses in vitro. However we found that thrombospondin-induced synapses are postsynaptically silent, lacking AMPA glutamate receptors. The AMPA subtype of glutamate receptors mediates fast excitatory synaptic transmission, and regulated trafficking of AMPA receptors is an important mechanism for controlling synaptic strength. We have therefore used biochemistry to investigate the identity of the astrocyte-secreted signal that enhances the number of synaptic glutamate receptors, and in our preliminary results we have identified this signal. In this application, we will first test the hypothesis that the regulated release of this signal from astrocytes controls synaptic glutamate responsiveness by controlling AMPA receptor trafficking in vitro and in vivo. Second, we will test the hypothesis that the astrocyte-secreted factor acts through a candidate neuronal receptor. These studies are important for several reasons. First they have the potential to yield new insight into the mysterious roles of astrocytes in the development, function, and plasticity of the CNS. Second, these studies have the potential to yield new insight into the basic mechanisms that control synaptic plasticity and glutamate receptor responsiveness. Understanding the mechanisms that regulate synaptogenesis and synaptic plasticity are crucial to understanding the neural basis of learning and memory, Alzheimer?s disease, drug addiction, and epilepsy.

DESCRIPTION (provided by applicant): How are synapses eliminated? To achieve precise neural connectivity and form mature neural circuits, the nervous system needs to be remodeled extensively. These remodeling processes include elimination of excess axons, dendrites, synapses and their debris. Growing evidence suggests that elimination processes are essential in shaping neural circuits during development as well as in regulating synaptic plasticity in response to experience and memory in adulthood. Moreover, certain neurodegenerative diseases, such as Alzheimer's disease (AD), are associated with a profound loss of synapses early in the disease process, underscoring the importance of understanding the mechanisms controlling synapse loss. Studies in mammals, flies and worms have previously demonstrated that glial cells play central roles in clearing apoptotic neurons and degenerating axons through an engulfment process called phagocytosis. However, the cellular and molecular mechanisms that drive these phenomena in mammals are still poorly understood. Recently by gene profiling, we have found that several phagocytic pathways--the MEGF10/draper/ced1 pathway that mediates axon pruning in flies and the MERTK pathway that mediates outer segment clearance by retinal pigment epithelial cells--are specifically and highly expressed by mouse astrocytes in both the developing and adult CNS. Although it has been assumed that microglia in the mammalian CNS are largely responsible for clearing neural debris, our findings suggest that mammalian astrocytes may actively participate in eliminating excess axons, dendrites, synapses and their debris. In our preliminary studies, we have found that astrocytes have high phagocytic activity in clearing axonal debris in vitro. In this application, we will test the hypothesis that astrocytes phagocytose neural debris and synapses through the MEGF10 and/or MERTK phagocytic receptor pathways expressed by astrocytes in vitro and in vivo. In the first aim, we will test the roles of MEGF10 and MERTK in astrocytes in mediating phagocytosis of neural debris and synapses by using an in vitro engulfment assay. In the second aim, we will use astrocyte-specific conditional knockout mice to determine whether astrocytes mediate the elimination of synapses within the developing and adult CNS. These studies have the potential to shed new light onto how synapses are normally eliminated, the extent to which synapse elimination and turnover occurs in normal adult CNS during learning and memory, and on how synapse loss in neurological diseases can be prevented. PUBLIC HEALTH RELEVANCE: To achieve precise neural connectivity and form mature neural circuits, pre-formed axons and synapses need to be remodeled extensively but the mechanisms responsible are poorly understood. In this proposal we focus on the hypothesis that astrocytes actively phagocytose synapses in developing and adult brains by the MEGF10 and Mertk pathways. A better understanding of synapse sculpting may lead to improved understanding of how circuits are formed, the adult synaptic plasticity that underlies learning and memory, and why synapses are lost in neurodegenerative diseases such as Alzheimer's disease.

DESCRIPTION (provided by applicant): We at the Stanford Department of Ophthalmology and Center for Vision and Prevention of Blindness are applying for a T32 Training Grant from the National Eye Institute to fund a the post-doctoral program designed to Integrate basic vision research and clinical ophthalmologic training. Our goals include: 1) intensive clinical ophthalmologic exposure, especially targeting non-clinically trained investigators, to facilitate future bench-to-bedside applications of basic vision research;2) facilitation of the transition of M.D. and M.D.-Ph.D.-trainees from clinical practice to rigorous vision research and fostering their successful application to the Career development (K) Award, an important funding mechanism for physician-scientists;and 3) the education and training of basic and clinical investigators from diverse backgrounds in molecular, cellular, synaptic, and systems level vision research to prepare them for academic careers in clinical and basic departments. With your help, we plan to train 4 M.D., Ph.D., or M.D.-Ph.D. fellows per year. An intensive course on basic vision research and ophthalmology (Ophthalmology 302A), weekly multi-disciplinary Stanford Center for Vision and Prevention of Blindness lectures and seminars, and other vision courses and training in epidemiology and statistics will be among the educational program tailored to each trainee's background and research interest, along with annual opportunities to present their work, teach, write grant applications, and attend scientific meetings. Each trainee will be assigned a basic science and a clinical mentor, chosen from Stanford vision investigators which collectively hold 16 R0l grants from the National Eye Institute and have published over 100 peer-reviewed publications in the last 2 years. These mentors help each trainee publish basic and clinical papers, serve as examples of rigorous scientific and clinical Investigations, and help develop a well-rounded program to interweave knowledge in basic areas such as developmental patterning of the eye and visual circuitry, synaptic plasticity and processing of the visual pathway, visual object recognition and reading, and visual-motor integration with a detailed understanding of the diagnosis and treatment of common causes of low vision and blindness and tools used to address them. An External Advisory Committee offers independent perspectives on the Program's operation. PUBLIC HEALTH RELEVANCE: Because our program is designed to provide each trainee with a comprehensive understanding of vision research and its applications to prevent vision loss, our trainees are poised to make significant research contributions to tackle the different causes of blindness and low vision, which affects 160 million people worldwide and costs the United States annually 35.4 billion dollars in financial burden when considering just the major adult visual disorders.

DESCRIPTION (provided by applicant): We will test the hypothesis that the superior cognitive abilities of humans compared to rodents is, at least in part, the result of an evolutionary increas in the ability of human astrocytes to control synapse formation and function compared to rodent astrocytes. Astrocytes are a major cell type in the brain that constitute at least one third of rodent and human brain cells. Long thought to be passive support cells, studies from many labs over the past 15 years have shown that astrocytes powerfully control the formation, function, and plasticity of synapses in the central nervous system (CNS). Could the superior cognitive abilities of humans be due to an evolutionary advance in astrocytic control of synaptic formation and/or function? We have developed new methods to purify both rodent and human astrocytes. These new methods will enable us to directly test our hypothesis. To address this hypothesis, we will take three different approaches. First we will use next generation RNA-Seq sequencing to determine and compare the transcriptomes of mouse and human astrocytes. A prediction of our hypothesis is that human astrocytes may secrete synaptic signals that are quantitatively or qualititavely different than those secreted by rodent astrocytes. The function of the human specific genes identified will therefore be further assessed for potential synaptic functions in Approach 2. Second, we will directly compare the ability of mouse and human astrocytes to stimulate synapse formation and function. We will also determine whether novel human astrocyte genes identified in approach 1 that encode for secreted proteins can stimulate synapse formation or function. Human genes that strongly control synapse formation or function will then be expressed in mouse astrocytes in transgenic mice to determine if they have enhanced cognitive abilities. In our final third approach, we will use modern metabolomics methods to elucidate the small chemical signaling molecules secreted by mouse and human astrocytes, with a focus on identifying novel astrocyte secreted chemicals that control synaptic function. The new molecular insight these studies provide about human astrocytes will also provide the foundation for investigations of pathological changes of astrocytes in human neurological disorders and reveal how malfunction of astrocytes leads to neurodevelopmental disorders such as autism and neuropsychiatric disorders with unique manifestations in human emotion and behavior, such as autism, anxiety disorder, and depression. PUBLIC HEALTH RELEVANCE: Our goal is to test the hypothesis that the superior cognitive abilities of humans compared with rodents is due to an evolutionary advance in the abilities of human astrocytes to control synapse formation and function. The proposed experiments will explore this using 3 different approaches (1) to use RNA---seq to determine and compare the rodent and human astrocyte transcriptomes, (2) to directly compare the effects of rodent and human astrocytes on synapse formation and function, and (3) to determine and compare the rodent and human astrocyte secreted metabolome. These experiments will advance our understanding of what it means to be human and lead to better understanding of the cause and treatment of human neurological and psychiatric diseases.

DESCRIPTION (provided by applicant): Microglia is myeloid-derived resident cells within the brain but their exact roles in health and disease are still poorly understood. The inability to reliably distinguish microglia from closely related myeloid cells called macrophages confounds many studies of microglial function. Therefore we screened for a new microglial specific gene that would allow the reliable identification, targeting, and characterization of microglia. In our preliminary studies, we identify TM119, a highly expressed transmembrane protein, as a highly expressed microglia-specific marker that is not expressed by macrophages or other peripheral immune cells. In this application, we will develop several new tools based on TM119 expression to selectively identify, isolate, and manipulate microglia. In the first aim, we will develop two antibodies against TM119 for the identification of microglia by immunostaining and isolation of pure microglia from whole brain tissue by immunopanning. In the second aim, we will develop two mouse lines, a TM119/Cre recombinase knock-in mouse and a TM119/CreERT2 BAC transgenic mouse. These mice will drive constitutive (knock-in) or inducible (BAC) Cre expression selectively within microglia. Crossing these mice with other mouse lines that express Cre- dependent genes will allow selective and specific manipulation of microglial genes. In the final aim, we will use the antibody and genetic tools developed in Aims 1 & 2 to acutely purify microglia from adult, developing, or inflamed mouse tissues to generate gene profiles of pure microglia by gene array and RNAseq. These profiles and all tools developed will be made immediately available upon publication to interested researchers. The tools developed in the proposed studies will enable investigators to better understand the roles of microglia, which may prove critical for a better understanding of the pathophysiology and treatment of human neurological diseases. PUBLIC HEALTH RELEVANCE: Microglia is myeloid-derived resident cells within the brain but their exact roles in health and disease are still poorly understood. We have recently identified a highly expressed gene called TMEM119 that is highly expressed specifically by microglia but not macrophages. In this application, we will exploit this discovery to develop new tools to selectively identify, isolate, and manipulate microglia. The tools will be made available to all researchers and will be helpful in elucidating the exact roles of microglia in CNS health and disease. !

DESCRIPTION (provided by applicant): We propose to investigate whether defects in astrocyte maturation and function contribute to the pathophysiology of human neurodevelopmental disorders (NDD) including autism and schizophrenia. Long thought to be primarily passive cells, in recent years, our laboratory and others have found that rodent astrocytes powerfully stimulate both excitatory and inhibitory synapse formation and function (Eroglu and Barres, 2010). Similarly, when human neurons are generated from embryonic or induced pluripotent stem cells (iPSCs), they form few synapses unless astrocytes are present. An emerging theme from recent research is that autism and schizophrenia are diseases of synapses. Could astrocyte defects contribute to the pathophysiology of common devastating NDD? In this application, we will take advantage of iPSC technology to study the development and function of astrocytes derived from iPSCs from patients who have autism and schizophrenia (Ricardo Dolmetsch, our Stanford colleague and collaborator in these studies, will provide these iPSCs). In our first aim, we will characterize and compare the molecular phenotype of human astrocytes generated by iPSCs by established methods to acutely isolated human fetal astrocytes, and then generate improved methods to more quickly generate human astrocytes from iPSCs that more closely resemble the gene profiles of acutely isolated fetal astrocytes from actual human brain tissue. In our second aim, we will characterize the phenotypes of astrocytes derived from iPSC cells from patients with NDD. In our 3rd aim, we will determine whether astrocytes from NDD patients are defective in promoting synapse formation and function. These studies have the potential to shed new light on the neural developmental basis of autism and schizophrenia in humans, have the potential to identify novel astrocyte genes that control synapse formation and function, and will generate new methods and drug testing platforms for human astrocyte generation from iPSCs.

? DESCRIPTION (provided by applicant): The decline of cognitive function has emerged as one of the greatest health threats of old age. The decline of cognitive function during normal aging is thought to be due to synaptic malfunction rather than loss of synapses or neurons. Although the mechanisms responsible are as yet unknown, the rate of synapse turnover is reduced in the aged brain, suggesting that senescent synapses accumulate with aging contributing to cognitive decline and placing the brain at higher risk for age-associated neurodegenerative diseases such as Alzheimer's disease. In this project, we propose to investigate the cellular and molecular mechanisms that underlie synaptic senescence of normal CNS aging. Specifically, we hypothesize that astrocytes, a major class of central nervous system glia, are critical mediators of synaptic health during aging. Recently, we discovered that astrocytes actively engulf and eliminate synapses in the developing and adult brain. Astrocytes appear to progressively engulf fewer synapses with normal aging. We will specifically investigate the hypothesis that reduced synaptic turnover by astrocytes with aging could lead to exponential accumulation of senescent synapses and cognitive decline and that enhancing this turnover mechanism could prevent or minimize synaptic senescence. First, we will characterize the rate of synapse engulfment by astrocytes in vivo over the lifetime of the brain to confirm that this engulfment rate progressivel declines with normal aging, and also determine whether astrocytes preferentially eliminate different synapse types or synapses in specific neural circuits. Second we will determine the molecular and cellular mechanisms that underlie this decline in engulfment of synapses by astrocytes by genetic profiling and in vitro assays with purified mammalian astrocytes. Finally, we will generate a transgenic mouse that has enhanced astrocyte phagocytosis of synapses to find out whether speeding up astrocyte synapse eating improves cognition during aging. These experiments have the potential to lead to a better understanding of why synaptic senescence occurs in the aging brain, and to lead to new therapies to lessen cognitive decline and vulnerability to Alzheimer's and other neurodegenerative diseases.