Geoffrey K. Aguirre, MD, PhD - US grants

DESCRIPTION (provided by applicant): The remarkable human capacity for visual face recognition is a cornerstone of social interaction and communication. Numerous psychiatric and neurological disorders, however, impair this critical ability. The behavioral signature of normative face perception is "holistic" processing (the whole is greater than the sum of the features) and "face-space" representation (a metaphorical coordinate system that relates the similarity of one face to another). Neuroimaging studies have identified cortical brain regions that respond to aspects of face stimuli, but it is not clear how the fundamental behavioral properties of face perception arise from the computations of neural systems. The proposed experiments will unite quantitative measures of face processing behavior with regional measures of neural activity using fMRI to provide this understanding. In Aim 1, the neural correlates of an "information processing" model of face perception are identified. The selectivity of holistic mechanisms for face stimuli, the parallel processing of visual features and the ability of the system to paradoxically improve performance with an increased workload are all behavioral features that will be described in terms of neural processes. In Aim 2, a visual adaptation method is used to test the idea that populations of neurons represent faces explicitly as deviations from an average face, and that neural representations are sharpened to allow discrimination between typical-appearing faces. In Aim 3, the effect of training and expertise is studied in terms of these neural mechanisms. Is it the case that training with an artificial stimulus set can induce face-like neural processing? The results of these studies have implications for rehabilitation and training to improve visual function, and the methods developed can be used to characterize the underlying information processing impairment in patients who have difficulty with face perception.

DESCRIPTION (provided by applicant): Early blindness leads to structural and functional alteration of the brain, as demonstrated with non-invasive, magnetic resonance imaging (MRI) in human participants. While our preliminary studies replicate the finding that blind participants develop cross-modal responses within their visual cortex (i.e., activation to an auditory stimulus), and have altered cortical structure (i.e., atrophy of the visual cortex white matter and disruption of white matter fiber coherence along the optic radiations), we also find enormous variability across the blind in these measures. This variability is systematic, with a strong correlation across blind subjects within functional measures (i.e., resting blood flow in the occipital lobe covaries with cross-modal responses) and within structural measures (optic radiation disruption correlates with visual cortex atrophy), but a low correlation between structural and functional measures. Different alterations in the visual pathway may therefore depend upon individual clinical features, including age at onset of blindness, severity of blindness, rapidity of visual loss, and development of compensatory abilities (e.g., Braille reading). Understanding these forms of neural plasticity could guide selection of patients most likely to regain useful vision following ophthalmologic treatment, similar to cochlear implant experience for treatment of deafness (HJ Lee et al., 2007). Recent therapeutic developments aim to reverse blindness that is congenital (i.e., targeted gene therapy for Leber's congenital amaurosis 2, LCA; Cideciyan et al., 2008; Maguire et al., 2008) and acquired (i.e., implanted retinal chip for age-related macular degeneration; Thanos et al., 2007). Recently, we have demonstrated that gene therapy in a canine model of LCA leads to increased cortical responses to visual stimuli after treatment (Aguirre et al., 2007). An important translational question is whether specific alterations of brain structure and function can predict restoration of cortical responses and useful vision following ophthalmologic treatment. We will obtain several MRI measures from a diverse population of both completely and incompletely blind individuals. These measures will be both structural (cortical gray matter thickness and white matter volume; white matter coherence by diffusion tensor imaging) and functional (resting cerebral perfusion; cross-modal activation to auditory stimulation; resting-state connectivity; activation to a standardized luminance modulation). We will test the hypothesis that certain alterations cluster together across subjects, and that these alterations are in turn related to individual features of the clinical history of vision loss in each patient. In a population of patients with blindness from RPE65-LCA, we will further determine which of these measures are modified by successful retinal gene therapy to restore vision, and which measures are predictive of functional outcome. The studies will systemically elucidate brain plasticity in vision loss by identifying the determinant factors both of clinical history and from the interplay of regional connectivity and function in the cortex. Our results will have direct translational value for the treatment of blindness, both in the development of compensatory strategies for the blind and for the guidance of clinical trials for ophthalmologic therapy.

Instrumentation - Project Summary The VRC Instrumentation Core Module supports the design and construction of specialized laboratory instrumentation. It consists of a machine shop, electronics shop, and their associated personnel. The module provides design consultation, construction of precision mechanical and electronic apparatus, modifications to commercial instruments to meet specialized vision research needs, and instrument repair and maintenance. The module has been used heavily by VRC investigators over the past 5 years, and we expect that this will continue.

? DESCRIPTION (provided by applicant): A symptom of many ophthalmologic and neurologic disorders is photophobia: discomfort and pain from flickering and bright lights. More specifically, photophobia is a key symptom in patients with migraine, both during headache and also in the headache-free inter-ictal period. This clinical observation has been confirmed by systematic behavioral studies demonstrating lower discomfort thresholds for visual stimulation in such patients. There is also evidence that some forms of visual discomfort may be related to signals from intrinsically photosensitive retinal ganglion cells (ipRGCs) that contain the photopigment melanopsin. The three aims of this proposal will test the hypothesis that photophobia associated with migraine headache is related to altered responses to stimulation of melanopsin, as well as characterize the photoreceptor mechanisms that mediate the documented elevated neural response to light observed in migraine with aura. All three aims will employ a digital light integrator to produce spectral modulations that selectively and robustly stimulate individual photopigment classes. We will measure the effect of stimulation directed separately at melanopsin and the cone photoreceptors, as well as interactions between melanopsin and cone signals. Aim 1 will examine direct effects of melanopsin stimulation in healthy human subjects, using three distinct but complementary response measures: behavioral reports of visual discomfort and perception of brightness, fMRI and the pupillary light response. We will examine the specific role of melanopsin in behaviorally-assessed visual discomfort and brightness perception, measure sustained brain responses to direct melanopsin stimulation, and use the pupillary light response to assess individual differences in melanopsin responsivity as well as the stability of these differences over time. Aim 2 will also study healthy control subjects and characterize whether and how melanopsin signals interact with signals from cones, to regulate the response to cone-mediated light flicker. We will measure psychophysical thresholds for detection of cone-mediated flicker and assess how these are affected by changes in the melanopic component of an adapting background light, use fMRI to measure neural correlates of the psychophysical effects, and employ a novel paradigm that allows us to use the sluggish pupillary light response to test the hypothesis that melanopsin signals regulate the response to cone-mediated flicker at an early site along the visual pathways. Aim 3 will build on the results of Aims 1 and 2 to characterize the photoreceptor mechanism of the enhanced neural response to light observed in migraine with aura. We will also measure whether the migraineurs demonstrate systematic differences with controls in either the direct or interactive effects of melanopsin, and whether such differences are related to the enhanced light sensitivity of this patient population.

? DESCRIPTION (provided by applicant): Worldwide, over 285 million people are visually impaired, and 39 million of those are blind. Visual impairments (blindness and low vision) severely and negatively impact the quality of life. New sight-restoration treatments have arrived for some blinding diseases. An electronic retinal prosthesis is approved for sight restoration in advanced retinitis pigmentosa. Gene therapy trials have shown promise for Leber's congenital amaurosis. Stem cell treatments for age-related macular degeneration are starting clinical trials. In some cases, remarkable results have been demonstrated, suggesting that at least partial vision restoration is possible after prolonged blindness. However, psychophysical data from clinical trials often show large variance in outcomes. Preconditions in brain structure and function associated with the central visual pathway (CVP) may underlie some of the variance. The effect of complete blindness upon the brain is well studied, but most ophthalmologic disease produces effects that vary from point-to-point across the retina. How does the precise topography of eye disease map onto changes in brain? Is acquired retinal damage related to local effects upon the structure and function of the CVP? Are any of these effects of vision loss on the CVP reversible? We propose to answer these questions, which we believe are crucial to the on-going efforts in developing sight- restoration treatments, by combining advanced retinal imaging with the brain-mapping techniques developed in the Human Connectome Project (HCP), using novel yet robust analytical methods. We will reveal the relationships between retinal pathology and their downstream impact on the CVP in unprecedented detail over the natural courses of blinding diseases and their treatments. We will make available to the research community the raw and processed data on retinal pathology for a range of blinding diseases, along with neuroimaging data collected with the HCP protocol. We will provide the computer codes used to extract relevant features from the retinal and neural datasets and the statistical models that relate the two. The extensive datasets and the associated data-analysis tools will form a transformative foundation for studies of neural plasticity associated with visual impairments and sight restoration.

Clinician-scientists are uniquely positioned to ask new and insightful scientific questions inspired by patient observations, yet, they often lack the expertise to be able to translate their observations into carefully designed basic scientific and translational experiments. There are likely many reasons for this, but the most cited barriers are lack of specific training, mentoring, funding, and time. If these barriers could be removed, more clinician-scientists could pursue careers in laboratory-based translational research, thereby helping to reverse the current state of affairs in many neurological disorders, in which basic research is proceeding at an increasingly rapid pace but translational research is lagging, and most patients with neurological disorders are left without preventions, treatments or cures. Here we propose a research training program for MD-PhDs or MDs who have finished their clinical training in a neuroscience-related specialty and are highly motivated to pursue careers as physician-scientists in innovative, laboratory-based translational research in brain diseases. The ReConNecT-IT (Remapping Clinical Neurosciences through Translation and Innovation Training) program consists of intense research training under the close mentoring of 1-2 faculty mentors. Trainees design and conduct independent research projects that they can take with them when they transition to independent support, and upon which they will base their NIH K-award application. Research projects are directed toward the translation of the genetic, molecular and cellular pathophysiology of neurological diseases into strategies for prevention, treatment or cure. Trainees will be encouraged to pursue projects that are collaborative and cross-disciplinary, as this fosters their research development, and linking disciplines helps generate ideas that are novel. Trainees will have access to 20 core faculty and can collaborate with other groups. Trainees will work alongside PhD researchers and participate in journal clubs, lab meetings and basic science seminars. The curriculum includes formal training in experimental design, statistical methodology and quantitative literacy, and well as individualized training on statistical/quantitative methodology by our Director of Statistical Training. Trainees will gain an understanding of critical topics in translational research and how basic research is translated into clinical trials (patient-oriented research) via two specific, semester-long courses. They will gain professional skill and understand career opportunities by participating in workshops on developing a K-award application, grant and scientific writing, pursuing an academic career, job search skills, laboratory and project management, and responsible conduct of research. A unique feature of ReConNecT-IT is that prospective trainees can know of their acceptance before their clinical training ends, allowing them to schedule research into their remaining clinical time, thereby expanding the total amount of research experience they will have before writing a K-award application. The main expected short-term outcome for this program is application for an NIH K award or equivalent grants.

Scientific Transparency: Project Summary We introduce in this competing renewal the Scientific Transparency Core Module (STCM). This module places the activities of the prior Image Analysis Core Module within a next-generation data curation framework, to bring reproducible and shareable data analysis to VRC investigators. The STCM supports scientific rigor and reproducibility in image analysis for VRC investigators, and provides support from the inception of an innovative analysis approach, to its realization in code, to its application within a reproducible pipeline. Module personnel support the development of innovative analysis tools. This core element of support continues from the Image Analysis Core Module, which has been heavily used during the current funding cycle. We aim to create automated, open-source analysis algorithms that provide reproducible results. Module personnel will also support data curation and analysis pipelines. This module function is based around the commercial Flywheel platform (http://flywheel.io). Brain imaging data obtained by Participating Investigators are already being stored on this system; the STCM will underwrite the costs of placing retinal imaging data on this platform. Module personnel will further assist investigators in placing analysis code within analysis ?gears? that operate on data stored in Flywheel. These are software containers that allow open-source analysis code to be executed, without external dependencies, on virtually any computer hardware or operating system. These components create documented, reproducible analysis pipelines that may be shared between VRC investigators and with the larger vision science community.