Bruce A. Berkowitz - US grants

The proliferation of new preretinal blood vessels into the vitreous is a major cause of visual loss in a number of common eye diseases. Although certain endogenous growth factors play a role in this neovascularization (NV), recent evidence indicates these growth factors are constantly available in the retina. In addition, NV is often associated with focal retinal areas of ischemia and it is presumed that a metabolic trigger (for example hypoxia, elevated glucose utilization and lactate concentration, or increased blood flow) activates the ambient growth factors and stimulates the initiation of NV. It has not been possible however, to establish the actual role of a specific metabolic factor in the trigger and regulation of NV since these factors are interdependent processes. We hypothesize that the stimulus of NV is an ischemia-related trigger, such as a significant perturbation (either singly or in combination) of retinal oxygen level, glucose utilization, lactate concentration, and retinal blood flow, which causes the ambient growth factors to become efficacious. To date, direct evaluation of the quantitative contributions of these interdependent factors has been lacking because it has not been possible to perform experiments in which various metabolic signals can be noninvasively assessed over long periods of time following controlled induction of NV. We have developed multinuclear nuclear magnetic resonance (NMR) techniques that circumvent these technical problems and can simultaneously record from intact tissue the quantitative contributions of oxygen level, glucose utilization, and lactate concentration prior to and during the development of NV in the same animal. We propose to employ these newly-developed NMR methods in a rat pup model of NV to answer the following question during the onset and progression of NV: What are the quantitative contributions of metabolic factors (preretinal oxygen level, lactate concentration, glucose utilization, and retinal blood flow) to the development and progression of NV? We will manipulate the occurrence of NV in our model and measure changes in the metabolic factors in relationship to: a) the spatial location of NV on retina; b) the occurrence and severity of the NV; c) changes in the occurrence of NV over time. These experiments will demonstrate the quantitative contributions of these metabolic factors and whether or not that are predictive of the occurrence of NV in this model.

DESCRIPTION (Adapted from applicant's abstract): The pathogenesis of retinal neovascularization in ROP is incompletely understood, although hypoxia is thought to be critical for its development. The long-term objective of this research is to understand better the role of hypoxia in retinal neovascularization so that more effective preventative, diagnostic and therapeutic strategies may be developed. During the previous funding period a novel MRI method was developed to non-invasively map the oxygenation rate along the two-dimensional extent of the retina. This method showed that an unusually slow retinal oxygenation rate was present in retinas before the development of ROP. This slow oxygenation rate is consistent with hypoxia, but it cannot yet be interpreted unambiguously as a measure of hypoxia. The present proposal seeks to confirm two hypotheses: (1) that abnormally slow oxygenation provides an indirect, yet specific measure of hypoxia and (2) that this MRI method has potential clinical utility for directing supplemental oxygen therapy. Three specific aims will be addressed using this MRI method and a newborn rat model of ROP: (1) to determine the association between the preretinal vitreous retinal oxygen tension and oxygenation rate under normal and pathological conditions; (2) to determine if the change in retinal oxygenation that occurs during supplemental oxygen treatment is linear with supplemental dose and evolves during neovascularization; and (3) to determine the association between the retinal oxygenation rate measured before neovascularization and the level of supplemental oxygen needed to prevent vascular endothelial growth factor (VEGF) upregulation and neovascularization. The proposed studies will lay the groundwork for eventual clinical evaluation of similar MRI experiments in human ROP. This work may have broad impact beyond ROP since abnormal blood vessel growth is a common pathobiologic event for other blinding disorders, such as diabetic retinopathy.

[unreadable] DESCRIPTION (provided by applicant): Current treatments for diabetic retinopathy are not entirely successful. We reason that therapies could be more rapidly developed and evaluated than is currently possible if a patient's risk of developing retinopathy could be predicted early in the course of the disease. We have developed a unique functional magnetic resonance imaging (fMRI) technique that measures the change in partial pressure of oxygen (deltaPO2) in the vitreous humor while the subject breathes carbogen (95% O2: 5% CO2). Using this acute retinal "stress" test, we have shown that carbogen breathing in rats and mice at 3-4 months of diabetes (before the appearance of retinal lesions) produces a significantly reduced retinal deltaPO2 compared to normals. A subnormal deltaPO2 is consistent with the presence of hypoxia but it cannot yet be unambiguously interpreted as a measure of hypoxia. In addition, in 3 month diabetic rats, aminoguanidine (AMG) treatment prevented the decrease in retinal deltaPO2. AMG is known to prevent retinal lesion formation in diabetic rats and inhibit the activity of both the inducible form of nitric oxide synthase (iNOS) and protein kinase C (PKC) (among other actions). Hypothesis: Subnormal retinal deltaPO2 in experimental diabetes is correlated with retinal hypoxia (Aim 1), and increased iNOS and PKC activity (Aims 2 and 3) but precedes the appearance of histopathology. Specific Aim 1) To compare retinal PO2 and deltaPO2 before, and at 2 wks, and 3 and 9 months of diabetes in rats. fMRI and a carbogen challenge will be used to measure the retinal deltaPO2. To measure PO2, a perfluorocarbon droplet will be injected into the preretinal vitreous over superior or inferior retina and 19F magnetic resonance spectroscopy performed. Specific Aim 2) To compare the retinal deltaPO2 (at 3 months) and histology (at 3 and 15 months) in normal and diabetic rats with and without treatment with a selective iNOS inhibitor (L-NIL (L-N(6)-(1-iminoethyl)lysine)), and in normal, diabetic, and diabetic iNOS knockout mice. Specific Aim 3) To compare the retinal deltaPO2 (at 3 months) and histology (at 3 and 15 months) in normal and diabetic rats with and without treatment with a selective PKC inhibitor (LY333531). In addition, we will compare retinal deltaPO2 in normal mice, transgenic mice that overexpress PKC beta II, diabetic mice, and diabetic PKC beta II-knockout mice. The results of the proposed studies will prove whether or not changes in retinal deltaPO2 correlate with the development of retinal hypoxia or multiple biochemical abnormalities that are associated with diabetic retinopathy but cannot be measured in vivo by existing techniques. The results of these studies could lead to the development of a non-invasive real time method for the early evaluation of diabetic retinopathy and its treatment.

[unreadable] DESCRIPTION (provided by applicant): Energy metabolism in all mammalian retinal cells is linked with demand for ions such as calcium. This metabolic-ion demand axis plays a central role in normal retinal function and hence healthy vision. Loss of vision and blindness are associated with the appearance of retinal neovascularization (NV) in diseases such as retinopathy of prematurity (ROP) and diabetic retinopathy. The pathophysiology associated with retinal NV is not well understood, although neuronal dysfunction and perturbed ion homeostasis have both been suggested as important factors. It is not yet known if abnormal ion demand occurs before the appearance of retinal NV and in retinal regions that give rise to retinal NV. These temporal and spatial knowledge gaps can not be addressed at present because current methods lack either spatial specificity (e.g., electroretinogram) or the ability to provide functional metrics of ion demand in vivo (histology). We propose a novel method to non-invasively measure retinal layer-specific ion demand that can also be applied in experimental rodent models of NV. This method, manganese-enhanced MRI (MEMRI), takes advantage of the facts that manganese (Mn2+) ion is a surrogate biomarker for various ions including calcium, and is a strong MRI contrast agent. We have validated that known retinal layer-specific changes in neuronal function / ion demand during light and dark adaptation can be robustly measured by high resolution MEMRI following systemic administration of a non-toxic amount of MnCl2 to awake rodents. Our overlying hypothesis is that the appearance of retinal NV will be temporally and spatially linked with abnormal neuronal function, as assessed by perturbed manganese uptake. Aim 1: To test the prediction that retinal NV in rat and mouse models is linked with abnormal Mn2+ uptake (indicative of perturbed ion demand) at the border of vascular and avascular retina (i.e., the site of retinal NV). The results of this innovative research will help clarify whether or not abnormal ion demand plays an important role in the development of preretinal NV. The novel methods in this application will also contribute to the advancement of functional MRI for the study of retinal diseases. The results of this innovative research will help clarify whether or not abnormal ion demand plays an important role in the development of preretinal neovascularization, a major cause of vision loss and blindness in premature births and diabetes. The novel methods in this application will also contribute to the advancement of functional MRI for the study of retinal diseases. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): There remains an urgent need to prevent vision loss from diabetic retinopathy (DR), a common and significant problem in patients with diabetes. Preclinical studies have highlighted retinal oxidative stress in the pathogenesis of DR. We reason that identifying the retinal mechanisms mediating pathogenic oxidative stress in diabetes will be important in the development of new and effective treatments. In neurons, oxidative stress and L-type calcium channels (LTCCs) activity are closely linked: increased LTCC activity elevates intracellular calcium content which must then be removed against a steep inter/extracellular calcium concentration gradient via ATP-dependent mechanisms which generate oxygen free radicals. We have discovered that diabetes profoundly changes the overall activity of photoreceptor LTCCs. Remarkably, a non-anti-oxidant and photoreceptor-specific treatment that corrected the abnormal outer retinal LTCC phenotype in vivo concurrently eliminated diabetic retinal oxidative stress. These experiments, using a validated and analytical tool (MEMRI), provide the first evidence for an unexpected and potentially very important discovery linking abnormal photoreceptor LTCCs and oxidative stress in diabetes. In the retina there are 3 major LTCC subtypes (Cav): Cav1.2 (located in inner retina), Cav1.3 (in inner retina, photoreceptor, and retinal pigment epithelium layers), and Cav1.4 (only in photoreceptors. How diabetes alters photoreceptor subtype channels are not yet known. We hypothesized that diabetes induces reproducible abnormalities in photoreceptor Cav1.3 and 1.4 expression and activity, and these changes actively contribute to oxidative stress and diabetic retinopathy. To begin to evaluate our new working hypothesis, we propose the following specific aim, which will take full advantage of our expertise in blending high resolution and analytical in vivo imaging of photoreceptor pathophysiology (manganese-enhanced MRI) with biochemistry (reactive oxygen species, western blot analysis (channel expression)) and histopathology (trypsin digest) in models of DR. SA 1: Test that diabetic alterations in photoreceptor Cav1.3, and 1.4 LTCC activity are required for oxidative stress and DR. The proposed experiments will open up a new line of scientific inquiry that is expected to improve our understanding about the origins of oxidative stress in DR. Because Cav1.3 and 1.4 subtypes are relatively insensitive to commonly used calcium channel blockers, our specific genetic and pharmacological approaches in concert with a novel imaging technique make these studies highly significant. These experiments will be crucial in the identification of new molecular targets for prevention and treatment of DR. The proposed research is highly innovative because it investigates a previously unsuspected but key role of photoreceptor LTCC subtype abnormalities in diabetes-induced oxidative stress, and will thus establish a firm scientific basis for changing current thinking regarding the origins of, and therapeutic options for, pathogenic oxidative stress in DR.

? DESCRIPTION (provided by applicant): Berkowitz, Bruce A. Project Summary / Abstract: Decades of research on post-mortem tissue have suggested a pathogenic role of rod cell oxidative stress in blinding disorders, such as diabetic retinopathy (DR) and retinitis pigmentosa (RP). Confirming this hypothesis in vivo, and demonstrating clinical potential in experimental models, requires the currently unrealized ability to noninvasively measure rod cell oxidative stress using endogenous contrast mechanisms in vivo. Here, we demonstrate that 25 ?m axial resolution MRI 1/T1 transretinal mapping is sufficiently sensitive to measure in control mice continuous production of endogenous paramagnetic free radicals from rod photoreceptor cells in the dark compared to two quench conditions, light or pharmacologic suppression of the production of mitochondrial free radicals. In a disease linked with rod oxidative stress, diabetes, rod free radical productions is greater than normal and is associated with co-localized MRI measures of rod dysfunction in vivo. Our overriding hypothesis is that measuring both rod free radical production and several essential rod functions in vivo provides an index of the severity of rod oxidative stress over time in both DR and RP that will be predictive of progression of both diseases, as well as allow for assessment of the efficacy of anti-oxidant therapy on disease outcome. The results of the proposed experiments will directly and unambiguously measure rod oxidative stress burden in incipient DR and RP in vivo, and this will enable earlier evaluation of disease progression and anti-oxidant treatment efficacy than is currently possible. Most of our new assays of rod free radical production and function are based on endogenous contrast mechanisms which greatly facilitate their translation into patients with DR and RP, and other oxidative-stress-based retinal diseases.

Project Summary / Abstract: There is an urgent need for disease-modifying treatment of Alzheimer's disease (AD) starting at its very onset. This knowledge gap remains because conventional approaches cannot measure in vivo brain region-specific biomarkers of the earliest relevant dysfunction underlying abnormal behavior. Often, spatial disorientation is observed during prodromal AD, and its occurrence predicts later dementia. A brain region contributing to this spatial confusion is the CA1 subfield of hippocampus because of its essential role in encoding spatial information. HC oxidative stress is most commonly identified at the very start of AD, and in experimental models of AD. Yet, it has not been possible to prove that prodromal oxidative stress in the relevant CA1 subfield plays a pathogenic role in at-risk patients showing impaired spatial memory because conventional methods only measure oxidative stress from post-mortem tissue. Addressing this major knowledge gap requires a new paradigm that compares antioxidant treatment efficacy in HC CA1 subregions in vivo with improved spatial learning and memory in experimental models, and that can then be translated into patients. In this proposal, we present a transformative solution to this problem based on a novel method recently discovered by our lab: QUEnch-assiSTed MRI (QUEST MRI). QUEST MRI is a robust and sensitive tool that has been validated against ?gold standard? methods and maps in vivo excessive free radical production in, for example, murine dorsal CA1. The QUEST MRI index of abnormally high production of paramagnetic free radicals in specific brain regions is a greater- than-normal spin-lattice relaxation rate R1 (1/T1) that can be returned to baseline after acute antioxidant administration. Our QUEST MRI studies have confirmed dorsal HC CA1-specific oxidative stress in spontaneous and familial AD mouse models with declines in spatial learning and memory in conjunction with HC CA1 oxidative stress measured ex vivo. We also find downstream consequences of oxidative stress such as greater-than-normal amounts of the lipid peroxidation product 4-hydroxynonenal (HNE), dorsal HC CA1 calcium dysregulation and reductions in dorsal HC CA1 calcium-dependent afterhyperpolarization (AHP). To improve statistical power, this proposal is tightly focused on uniquely testing a specific working hypothesis that oxidative stress in dorsal CA1 in vivo causes deterioration of spatial memory in experimental models. Our highly innovative studies by an experienced team of experts will validate a new bridging tool for testing in vivo antioxidant therapeutic strategies to mitigate a clinically important early decline in spatial memory preceding later loss of personhood in AD.