David R. Williams - US grants

DESCRIPTION (provided by applicant): The prevailing view about the organization of human foveal vision is that optic blur, photoreceptor sampling, and neural blur are all matched to each other. This project will scrutinize this view with laser interferometry, wave-front sensing, and adaptive optics. These technologies when combined allow the first quantitative measures of the eye's optics, cone mosaic, and neural response in the same individuals. Moreover, the use of adaptive optics allows any visual stimulus to be imaged on the retina at higher resolution than has previously been possible. We will explore the possibility that the neural visual system is not equipped to take advantage of all of the improvement in retinal image quality afforded by adaptive optics. We will test the hypothesis that improving the eye's optics leads to a degradation of performance on certain tasks such as vernier acuity and the identification of the color of small, brief flashes of light. Images of the trichromatic cone mosaic in living human eyes will be compared with maps of the color appearance of tiny flashes that stimulate single cones. These experiments will clarify the fundamental limits on spatial and color vision.

Though it has been known that color vision is mediated by three types of photoreceptors for some 200 years, there is little information about how the spacing and packing geometry of the short (S), middle (M), and long (L) wavelength sensitive cones. The spatial organization of the S cones is the best known, while almost nothing is known about M and L cone distribution. Even the ratio of L to M cones has proven elusive. The goal of this proposal is to determine the spacing and packing geometry of these three interleaved cone submosaics in the primate retina. In order to resolve individual photoreceptors using the light returned following a double pass through foveal cones: 1) A patch of retina will be removed from the eye and maintained in vitro. 2) Water immersion optics will be used to image the photoreceptor mosaic on a high resolution, high sensitivity CCD camera. The camera takes advantage of recent advances in low light imaging borrowed from astronomy. By comparing the amount of light reflected from the retina at three wavelengths chosen to produce the highest contrast between the cone types, first with the photopigment unbleached and then again following bleaching, the type of photopigment contained by each cone will be determined. The resulting maps showing the organization of the three types of photoreceptors across the retina will make it possible to 1) study how trichromacy is incorporated into the retina with minimal cost for spatial vision, 2) explore the contributions of the three cone types to the post- receptoral channels that are known from psychophysical experiments, 3) determine the chromatic organization of retinal receptive fields as a first step towards understanding the retinal circuitry that implements chromatic opponency, and 4) determine whether there are mechanisms responsible for forming regular patterns of cones during development.

The Center for Visual Science (CVS) is an interdisciplinary institute whose 18 investigators are visual scientists from six different departments at the University of Rochester (Computer Science, Neurobiol- ogy and Anatomy, Neurology, Ophthalmology, Physiology and Psychology). The CVS faculty, with expertise in psychophysical, physiological, computational, anatomical, and clinical approaches to vision research, are bound together by a shared interest in characterizing visual phenomena and linking them to their neural substrates. CVS provides a rich collaborative environment for vision research that is greatly facilitated by its Core Grant. This competing renewal application seeks support for three modules: 1) An Administrative Module to provide partial support for the PI and the CVS secretary. The role of the Administrative Module is to coordinate and facilitate interactions between vision scientists working within the Center and with scientists outside it. 2) A Computing Module to improve the centralized computing resources of CVS and to provide support for the systems analyst, whose expertise serves all members of the Center. 3) A Technical Services Module to modernize the aging CVS machine shop and to provide partial support for the CVS machinist. The Module will also provide partial support for the CVS electronics engineer, whose services are shared among CVS investigators.

vision; biomedical facility;

14 investigators from the Center for Visual Science (CVS) at the University of Rochester will participate in the Core Grant. These investigators have primary appointments in five departments: Brain and Cognitive Sciences, Computer Science, Neurobiology and Anatomy, Neurology, and Ophthalmology. They are bound together by research interests in three domains: 1) the relationship between visual experience and its neural substrate, 2) the role of vision in guiding behavior, and 3)mechanisms of retinal disease. The proposed Core will provide three modules, each of which will benefit all three of these research domains: 1) A computing Module that will provide expertise in the generation of complex visual stimuli for psychophysical and physiological experiments, real-time experimental control, data analysis, and modelling. 2) An Instrumental Module that will provide expertise in optical, electronic, and mechanical engineering to design and assemble novel instrumentation such as that required for eye, head, and hand tracking, multielectrode recording in primate cortex, and high resolution fundus imaging. 3) A new Histology Module that will provide a shared, staffed facility equipped to process both ocular and brain tissue.

DESCRIPTION (provided by applicant): The two goals of this Bioengineering Research Partnership are (1) to design and construct a new generation of instruments for noninvasive imaging of the mammalian retina with 3-D resolution superior to existing technology and capable of resolving single cells in vivo. (2) to explore the value of this technology through application to human retinal disease and retinal surgery. These instruments will combine adaptive optics, a technology borrowed from astronomy that automatically corrects all the eye's aberrations, with confocal microscopy, a technology for optically sectioning the retina. The lead institution will be the University of Rochester, and partners include Lawrence Livermore National Laboratory, the Doheny Eye Institute at USC, the University of Houston, the University of California at Berkeley, and the Schepens Eye Research Institute. By the end of year 1, a device will be operational at each of four clinical sites: USC, Rochester, Houston, and Schepens. In years 2-5, these devices will provide high resolution imaging of neovascularization in age related macular degeneration and diabetic retinopathy, photoreceptors in retinal degenerative disease such as retinitis pigmentosa, ganglion cell bodies in glaucoma, individual retinal pigment epithelial cells, and blood flow in the smallest retinal capilliaries. In year 3, a new surgical microscope equipped with adaptive optics will be constructed by LLNL. Retinal surgeons at USC will evaluate this device in years 4-5. Based on its experience with earlier instruments, the BRP will design and build a sixth instrument in year 4 that will be portable, compact, and user friendly. This device will be available to investigators outside the BRP. The BRP brings together optical engineers, basic vision scientists, and clinical vision researchers. This will allow engineers to design instrumentation informed by the specific needs of clinical research, allowing them to translate adaptive optics technology directly into clinical application. LLNL brings to the partnership expertise in optical engineering and adaptive optics from the fields of astronomy and laser fusion. Rochester and Houston will contribute experience in adaptive optics applied to retinal imaging. Rochester first applied adaptive optics to high resolution retinal imaging and Houston has recently demonstrated a prototype adaptive optics system that is the precursor for the devices proposed here. Schepens brings international leadership in scanning laser ophthalmoscopy. UC Berkeley provides expertise in the study of retinal degenerative diseases. USC, with its innovative approaches to retinal disease and retinal surgery, will join Rochester, Houston, and Schepens in providing clinical sites for the evaluation of confocal adaptive optics technology.

[unreadable] DESCRIPTION (provided by applicant): This Bioengineering Research Partnership is a consortium of 6 laboratories that are building adaptive optics scanning laser ophthalmoscopes (AOSLOs) and applying them to microscopic examination of the living normal and diseased retina. The principal investigator is David Williams (University of Rochester) who introduced the first successful adaptive optics instruments to vision science. Other lead investigators include: Steve Burns (Indiana University) an international leader in laser scanning ophthalmoscopy, John Flannery (UC, Berkeley) an expert in retinal degeneration and the development of retinal biomarkers, and Austin Roorda (UC, Berkeley) who designed the first adaptive optics scanning laser ophthalmoscopes, David Arathorn (Montana State University) who brings strong mathematical skills and software development tools for tracking the eye in AOSLOs, and R. Daniel Ferguson (Physical Sciences, Inc.) whose expertise is in the optical engineering of innovative eye tracking systems. During years 1-5 of the previous funding period, the partnership designed and built four AOSLO instruments and two more instruments are under construction. These devices have produced the first images ever of numerous microscopic structures in the living eye including the RPE cell mosaic, single leucocytes flowing in the smallest retinal capillaries, and fluorescently-labelled ganglion cell dendrites, axons and cell bodies. In addition, technical challenges for imaging eyes ranging in size from human to rodent have been overcome. The partnership is now proposing continued funding for years 6-10 to develop new capabilities for these instruments such as a combined hardware and software approach to reduce the effects of eye motion on high resolution retinal imagery. We also will develop a new generation of instruments with special capabilities, such as the ability to image ganglion cells in the living human eye without the use of fluorescent dyes, and the ability to optically record neural responses from specific retinal cells. PUBLIC HEALTH RELEVANCE: This application will develop a technology, adaptive optics scanning laser ophthalmoscopy, for taking extremely sharp pictures of the inside of the living eye, so sharp that individual cells can be seen. This technology will be used to study diseases such as age-related macular degeneration and glaucoma. It may allow the earlier detection of retinal disease, better tracking of disease progression, and the efficacy of therapies for retinal disease. [unreadable] [unreadable] [unreadable]

Administrator; Advisory Committees; Attention; CCSG; Cancer Center Support Grant; Communication; Conflict; Conflict (Psychology); Consultations; Core Grant; Ensure; Equipment; Event; Faculty; Funding; Grant; Individual; Institution; Instrumentation, Other; Investigators; Left; Mediating; Monitor; Operation; Operative Procedures; Operative Surgical Procedures; P30 Grant; Policies; Principal Investigator; R01 Mechanism; R01 Program; RPG; Reporting; Research; Research Grants; Research Personnel; Research Project Grants; Research Projects; Research Projects, R-Series; Research Resources; Researchers; Resource Allocation; Resources; Review Committee; Role; SCHED; Schedule; Science; Scientist; Sight; Study Section; Surgical; Surgical Interventions; Surgical Procedure; Task Forces; Technology; Time; Training; Vision; Vision research; Visit; Writing; adaptive optics; experience; instrumentation; member; social role; surgery

DESCRIPTION (provided by applicant): A fundamental goal of neuroscience is to understand how neural circuitry carries out the computations that transform raw sensory signals into behavior. Current understanding is limited by available techniques for recording neural activity in awake, behaving organisms. At one end of the range of spatial scales, microelectrode arrays provide excellent temporal resolution of single neuron activity, but can only sparsely sample the local neuronal population. At the other end, functional magnetic resonance imaging has poor resolution in both space and time, and records neuronal activity only indirectly through blood flow. Optical imaging promises to revolutionize neuroscience by bridging this gap and offer minimally-invasive, direct recording of neural activity at single cell resolution in the intact working brain. Optical imaging of neural activity, however, has yet to deliver the ultimate prize of recording the activity of many individual neurons in real time throughout the depth of a brain structure such as the cerebral cortex. Achieving such a goal will require the coordinated effort of experts from disparate backgrounds, including neuroscientists, optical engineers, biochemists and molecular biologists. The obvious need to bring such experts together in dialogue is the central motivation for the 27th Symposium of the Center for Visual Science (CVS), entitled "Photons and Neurons", scheduled for June 4th- 6th, 2010 at the University of Rochester. The goal is to bring together scientists who develop and use imaging methods and provide a platform to exchange ideas, identify pressing neurobiological questions, discuss current limitations, and develop possible solutions. Individual sessions will cover the following topics: (1,4) Imaging CNS circuits, (2) Advances in Optical imaging, (3) Imaging the retina, (5) Controlling neurons with light. We expect that the Symposium will attract researchers representing multidisciplinary approaches ranging from neural coding and primate neurophysiology to optical engineering and molecular biology. We are particularly interested in attracting younger neuroscientists and optical engineers. To that end, we plan to award ten travel fellowships to graduate students interested in attending the meeting and provide them with an opportunity to present their work in a poster session. PUBLIC HEALTH RELEVANCE: Optical imaging of neural activity has yet to deliver the ultimate prize of recording the activity of many individual neurons in real time throughout the depth of a brain structure such as the cerebral cortex. Achieving such a goal will require the coordinated effort of experts from disparate backgrounds, including neuroscientists, optical engineers, biochemists and molecular biologists. The obvious need to bring such experts together in dialogue is the central motivation for the 27th Symposium of the Center for Visual Science (CVS), entitled "Photons and Neurons."

PROJECT SUMMARY/ABSTRACT A highly collaborative team at the University of Rochester led by Jennifer Hunter and David Williams will translate in vivo two-photon excited fluorescence ophthalmoscopy (TPEFO) from monkey to human. This technology has the potential to provide new information on microscopic retinal morphology and to serve as a superior, objective measure to assess retinal function, a rare capability among existing imaging technologies. Effects of IR autofluorescence reduction, currently the only noticeable, consequence of TPEFO will be characterized. Additionally, we will establish thresholds for retinal damage. Monitoring retinal health in cortically blind human subjects after repeated TPEF imaging will establish whether TPEFO can be applied safely to normal human eyes.

? DESCRIPTION (provided by applicant): The lack of methods to image functioning retinal circuitry in the living eye is a fundamental impediment to all efforts at vision restoration. The ability to assess the structure and function of retinal circuitry in vivo at multiple retinal layer simultaneously will not only transform our understanding of where different vision restoration strategies succeed and fail, the feedback these tools will provide will greatly shorten the development cycle for all approaches in which they are deployed. Investigators in the Advanced Retinal Imaging Alliance (ARIA) at the University of Rochester (Jennifer Hunter, Bill Merigan, and David Williams) will create a new retinal imaging tool designed to track changes in structure and function in both the inner and outer retina at cellular resolution in individual animals over time. This tool will combine two-photon, adaptive optics imaging of genetically-encoded calcium indicators to track the neuronal responses of individual photoreceptors and ganglion cells in two different animal models, mouse and monkey. We will assess the value of this imaging technology in three different approaches to vision restoration. Applying these tools to a gene therapy approach, the Rochester team will collaborate with Connie Cepko's laboratory at Harvard University to rescue cone function in retinitis pigmentosa through the up-regulation of anti-oxidants. We will demonstrate the value of these imaging tools for an optogenetics approach by expressing channel-rhodopsin in monkey cones in collaboration with Botond Roska at the Miescher Institute for Biomedical Research. This effort will also develop a primate model of retinal degeneration in which a viral vector is used to shave off photoreceptor outer segments. Finally, these imaging tools will be deployed to monitor the differentiation and neural connectivity of stem cells in mouse and then monkey in collaboration with David Gamm at the University of Wisconsin, Madison. This consortium will not only develop advanced imaging technology, it will translate it to the field of vision restoration, and create a collaborative environment for sharing best practices and combining different approaches.

ADMINISTRATIVE CORE SUMMARY The Administrative Core benefits from the expertise of the entire administrative staff of the Center for Visual Science, with the most significant responsibilities for administering the Core falling on the PI and his Administrative Assistant. Both interact with other members of the CVS Executive Committee, and the Core Directors and Associate Directors in particular, to ensure efficient, cost-effective, fair, and scientifically optimal use of the Technical Services of each of the three Cores.

The importance for vision of the tiny fovea has been established by centuries of investigation as well as observations of the devastating consequences of its damage through injury or disease. Though evidence suggests that the fovea contains the full complement of the two dozen or so classes of ganglion cells found in peripheral retina, we know little about the physiology of these foveal cells. This gap in our understanding is the result of challenges in obtaining electrophysiological recordings from this delicate and topographically-complex structure. These challenges have been overcome by a method developed in our laboratory that allows simultaneous calcium imaging of the fluorescence responses of hundreds of foveal retinal ganglion cells in response to visual stimuli. Because this technique allows recording from single cells without damage in the living eye, we can study the same cells for months or even years, offering the opportunity to characterize the performance of each cell more thoroughly than has been possible with any prior method. Since the first submission of the proposal, we have made significant improvements in the expression of calcium indicator, GCamMP6s, in ganglion cells that increases the extent of expression to greater eccentricities, the fluorescence signal from each cell, as well as reducing the loss of ganglion cells over time. Moreover, we have designed a new ophthalmoscope with two independent adaptive optics systems, one dedicated to high resolution stimulus delivery and a second dedicated to high resolution ganglion cell recording. We have also developed an extensive battery of visual stimuli to characterize the responses of each cell in space, time, and color. This battery will include a white noise stimulus capable of identifying the locations and classes of single cone inputs to the receptive fields of foveal ganglion cells. To assist in cell classification, these physiological observations will be supplemented with ex vivo and in vivo histological analysis of the morphology of ganglion cell dendritic arbors. Armed with these improvements, we will undertake a comprehensive survey of both the physiology and anatomy of the foveal ganglion cell classes that mediate primate foveal vision.