Constance L. Cepko - US grants

We will be developing a lineage map of the chick retina using a new technology based on retrovirus vectors. The retrovirus vectors are ideal markers for lineage mapping in that they reproducibly integrate into virtually any cell type in most mammalian and avian species. This integration feature allows for the marking of a cell and all of its progeny. We will construct vectors which encode histochemical markers in order to provide a facile means to identify infected cells in tissue sections, and develop a methodology for infecting and assaying chick embryos. The chick retina will be the primary organ which will be analyzed using this new technique. The retina is well described anatomically and functionally and should provide for easy access in both early and late development stages. We wish to assess the lineage relationships among the various neuronal and glial cell types within the retina. We will also investigate questions of lineage within the retinal ganglion cells of the retina in order to assess the role of lineage in the establishment of the retino-tectal map. If we are successful in these studies, the vectors and methods described here will allow for lineage mapping in virtually any tissue in most mammalian and avian species.

Fundamental parameters of cortical and cerebellar development will be Studied using retrovirus-mediated gene transfer. Through introduction of a histochemical marker gene carried by a retrovirus vector, cells will be tagged with an innocuous and stable genetic marker, enabling an analysis of fate within the developing brain of rats and mice. Patterns of proliferation, migration and lineage will be examined. The role of lineage in the determination of cytoarchitectonic boundaries will also be investigated. The protocols employed in this proposal will build upon the previous methodology developed in the prior grant period. Improvements to the existing technology will be made by way of the development of novel histochemical marker viruses.

DESCRIPTION (provided by applicant): An understanding of the development of the vertebrate retina is of fundamental importance to our understanding of disease and to the generation of effective therapies for blinding illnesses. For example, replacement of degenerating photoreceptors will rely upon manipulation of the progenitor cells that make photoreceptors. A clear description of the number and types of progenitor cells that make up the retina will provide a platform from which mechanistic studies can be done. We and others showed many years ago that retinal progenitor cells are multipotent throughout development (progenitor cells are herein defined as cycling cells). We subsequently showed that even though they are multipotent, they are not equivalent throughout development. Over the last grant period we examined these cells further with respect to their cell cycle behavior. These studies provided additional evidence that retinal progenitor cells differ as they express different cell cycle regulatory proteins and use them to control cell cycle exit in different populations. We now wish to further define the number and types of retinal progenitor cells. These data will rule in or out several different models of retinal development. Further, we wish to use these data to generate strains of mice in which different types of progenitor cells will be differentially marked with reporter genes. These strains will be used to further our understanding of how retinal progenitor cells transition from one state to the next, and whether different types of progenitors vary in their proliferative behavior and/or production of different cell types.

One of the salient features of the vertebrate central nervous system (CNS) is the incredible complexity of cell types. The developmental processes that generate such diversity are as yet quite unexplored. A model system in which these processes can be studied in vitro and in vivo would greatly aid efforts aimed at understanding the molecular mechanisms that govern determination of cell fate. The retina offers such an opportunity. It is relatively simple, well-characterized, and classically has served as a model for CNS development and function. We have devised an in vitro system that allows the generation of postnatal retinal cell types, including rod photoreceptors, bipolar cells, muller glial cells, and amacrine cells. Mitosis of uncommitted progenitors, commitment to cell fate, and at least partial differentiation occur within dissociated primary cells explanted from rat retina and cultured in serum-free, defined medium. We propose to study these processes in vitro, focussing on the role of intrinsic and extrinsic cues. The developmental regulation of progenitor responsiveness, and of production of factors that are required for these events, will be studied. As the assay system is robust (in vivo rates of generation of rods, bipolars, and muller glia have been achieved), we also plan to isolate and identify proteins and/or genes encoding these activities. Our initial efforts will be on the rod photoreceptor pathway. Information concerning de novo rod generation may be relevant to rod survival, rod regeneration, or rod replacement and thus useful in the understanding and/or treatment of degenerative diseases such as retinitus pigmentosa.

New technology has allowed us to envisage a time when we will routinely define the gene expression profile of individual cells. This will be particularly important in the study of the central nervous system (CNS), where overwhelming complexity clouds our understanding of its development and function. This is particularly true for our understanding of disease processes. If we could define gene expression differences among the normal and pathological states, we would much closer to an understanding of the disease process. Two crucial steps towards this goal are the preparation of representative nucleic acid probes from individual cells and the development of a high throughput method for identifying the sequences within such probes. Methods for the preparation of probes from individual cells currently exist, but have been difficult to test for representation as all of the genes expressed by any individual cell are not known. Similarly, methods exist that might allow the determination of the presence of known genes and expressed sequence tags (ESTs) in such probe preparations. We propose to test several methods for the preparation of representative probes from individual, identified cells, and compare gene profiling methods using such probes. Our strategy is to compare profiles from individual CNS cells that are nearly identical. This will allow us to score reproducibility, or "noise", to determine which methods allow for an adequate representation of gene expression in individual cells. We then plan to use multiple methods to isolate several different types of CNS cells and profile those cells. Electrophysiological recordings from selected cell types will be carried out in order to establish a correlation between RNA expression and protein, as well as to explore the benefits to an electrophysiologist inherent in a knowledge of the repertoire of receptors and channels expressed in a single cell.

New technology has allowed us to envisage a time when we will routinely define the gene expression profile of individual cells. This will be particularly important in the study of the central nervous system (CNS), where overwhelming complexity clouds our understanding of its development and function. This is particularly true for our understanding of disease processes. If we could define gene expression differences among the normal and pathological states, we would much closer to an understanding of the disease process. Two crucial steps towards this goal are the preparation of representative nucleic acid probes from individual cells and the development of a high throughput method for identifying the sequences within such probes. Methods for the preparation of probes from individual cells currently exist, but have been difficult to test for representation as all of the genes expressed by any individual cell are not known. Similarly, methods exist that might allow the determination of the presence of known genes and expressed sequence tags (ESTs) in such probe preparations. We propose to test several methods for the preparation of representative probes from individual, identified cells, and compare gene profiling methods using such probes. Our strategy is to compare profiles from individual CNS cells that are nearly identical. This will allow us to score reproducibility, or "noise", to determine which methods allow for an adequate representation of gene expression in individual cells. We then plan to use multiple methods to isolate several different types of CNS cells and profile those cells. Electrophysiological recordings from selected cell types will be carried out in order to establish a correlation between RNA expression and protein, as well as to explore the benefits to an electrophysiologist inherent in a knowledge of the repertoire of receptors and channels expressed in a single cell.

DESCRIPTION (provided by applicant): Photoreceptors are a highly specialized type of neuron with extraordinary properties. Rod photoreceptors are exquisitely sensitive to light due to their precise and unique morphology as well as their highly efficient phototransduction mechanism. They are also the cell type that seems most vulnerable to degeneration in that many genetic and environmental lesions lead to their degeneration. We propose to study the mechanisms of genesis and differentiation of murine rod photoreceptors. Over the last grant period, we characterized the response of retinal cells to two extrinsic cues that negatively and positively regulate rod development, ciliary neurotrophic factor (CNTF) and taurine, respectively. We also characterized a role for the retinal anterior homeobox gene, Rax, in photoreceptor development. We have been developing reagents to probe the molecular consequences of the action of these extrinsic factors and the Rax gene. In addition, we discovered that a number of newly characterized genes are expressed during retinal development in a manner that is highly suggestive of a role in rod development. We pioneered the application of two new techniques, electroporation and RNAi, for rapid manipulation of gene expression levels in the retina in vivo. These new methods, along with the knowledge of many new genes that serve as markers of the rod pathway, will greatly aid in the study of gene function in the development of rod photoreceptors. Our goal is to create a molecular description of the pathway of rod development, including the series of gene expression changes that occur in the pathway, and place the action of the regulators of rod development within this pathway.

[unreadable] DESCRIPTION (provided by applicant): Many diseases that ultimately lead to blindness are caused by the degeneration of photoreceptor (PR) cells. Cones typically die after rods, with kinetics somewhat dependent upon the particular disease. Rod loss followed by cone loss is also seen in cases where a genetic etiology has not been established, as in some forms of macular degeneration. While humans are able to function quite well without rods, the loss of cone-mediated vision is devastating. The reason(s) that cones die in these cases is unknown. However, since cone loss can be initiated by events that are not intrinsic to cones, these events must include some type of cell-cell interaction, perhaps including the action of a secreted molecule(s). Such a process may be susceptible to interruption through the application of a pharmacological or a cell based therapy. In addition to progressive diseases such as retinitis pigmentosa, there is a mouse model, the cyclin D1 knock-out (KO) mouse, in which degeneration is arrested. The cause of PR death, as well the cause of the arrest, are unknown. This model may provide some insight into how degeneration can be arrested in progressive diseases. We are seeking to use retinal microarrays to define the gene expression changes that accompany PR death in mice, with an emphasis on the events that lead to cone death. In addition, we will characterize the gene expression changes that accompany the arrest of PR degeneration in the cyclin D1 mutant. We further plan to characterize the expression patterns of such genes in normal and pathological tissue. Finally, we will explore the function of some of these genes using genetic approaches in mice. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): The mapping of connections among central nervous system (CNS) neurons is a longstanding goal of neuroscientists. Knowledge of the connections will inform our understanding of the function of the nervous system, in both health and disease. Dye labeling, physiological recordings, anatomy at the level of electron microscopy, and the transfer of proteins have all been used to track synaptic connections. Neurotropic viruses have also been used, taking advantage of their ability to spread among neurons. Viral tracers offer the promise of higher throughput, and can expand the repertoire of assays beyond mapping connections, via their gene transfer capabilities. We have recently developed vesicular stomatitis virus (VSV) as a transsynaptic tracer. This virus has several advantages over the previously developed transsynaptic viruses, rabies virus (RABV) and pseudorabies virus (PRV). RABV is a lethal virus, which limits its use in the laboratory. PRV has a large and complex genome, making it difficult to engineer in a straightforward manner. VSV has a relatively simple genome, very similar to that of RABV. It is very well characterized as it rapidly grows to high titr and has a long track record for safety in the laboratory. We found that VSV can transmit transsynaptically in mice, monkeys, and multiple other species, including birds, amphibians, and fish. Moreover, we found that we could create VSV vectors that could transmit specifically retrogradely or specifically anterogradely. The directionality was entirely dependent upon the viral envelope glycoprotein, encoded by a viral G gene. We propose to create a series of transsynaptic viral tracers that will be useful to neuroscientists working in different organisms and different areas of the nervous system. We will first explore the use of VSV with different viral G proteins, applying what we learn to the design of specific anterograde and retrograde tracers. We hope to expand the repertoire of G proteins that can be used not only with VSV, but with other tracers as well. We also propose to develop an alternative non-toxic virus for use as a tracer, using some of the tools that we have already developed for VSV. Finally, we propose to develop a tracer that, although based upon a virus, is actually not a virus, but a vehicle to move Cre recombinase across synapses to map connected neurons. These alternative strategies offer the promise of reduced toxicity, relative to VSV, RABV, or PRV, as well as other features that can be combined with engineered lines of mice, or other model organisms.

DESCRIPTION (provided by applicant): Blindness is often caused by genetic lesions that directly affect photoreceptors. There are >200 disease genes in humans that lead to blindness (retnet:www.sph.uth.tmc.edu/Retnet). Addressing each genetic deficit by transduction of a gene specific to that disease would be a large and expensive undertaking. As an alternative, gene therapy can be used to attack a problem common to multiple genetic forms of blindness. One such approach is to preserve cone function in retinitis pigmentosa (RP). People with RP initially have poor night vision, as rods are dysfunctional. Rods are then lost, which is followed by loss of cone function, and then cones themselves. As cones do not express the disease gene in most cases, there must be a non-autonomous cause of cone death. If this cause can be identified, and combated, a more generic form of therapy can be developed. Using 4 mouse models of RP, and an unbiased microarray approach, we discovered that many genes involved in the regulation of metabolism were altered at the onset of cone death. We further showed that mTOR, a key regulator of metabolism, was not phosphorylated in RP cones. This is now the earliest sign of cone stress in RP that is known. As the disease progressed, we discovered that cones carried out chaperone- mediated autophagy. Injection of insulin into RP mice, which can lead to increased activity of mTOR, increased survival of cones. We have suggested a model wherein cones are dysfunctional and then die due to dysregulated metabolism. As rods are the major cell type in the ONL, cones experience a greatly altered environment following rod death. The cone OS collapse, they lose their intimate association with the RPE, and they are exposed to a hyperoxic environment. They show greater oxidation of their nucleic acids, proteins, and lipids. Fighting oxidation may cause cones to require more NADPH, which is generated from glucose via the pentose phosphate pathway (PPP). It is also produced by two cytosolic enzymes, malic enzyme and isocitrate dehydrogenase. If glucose is shuttled to the PPP, the glycolytic pathway would slow, which could lead to several metabolic outcomes, including reducing the surface area of cones, as well as reducing phototransduction and potentially the ATP levels. We wish to develop AAV-mediated gene therapy to combat metabolic stress in the cones in RP. As cone-mediated vision is of greatest importance to humans, preservation of cone function is critical to the quality of life among RP patients. If such therapies can be developed, it is also possible that these therapies can be extended to other diseases where cones are compromised, such as age-related macular degeneration (AMD).

? DESCRIPTION (provided by applicant): We propose to combine whole brain 2-photon imaging of neural activity in behaving larval zebrafish with detailed anatomical and connectivity information extracted from the same animals. The final goal is to generate quantitative models of brain wide neural circuits that explain the dynamic processing of sensory information as well as the generation of motor output by these circuits. Anatomical data will be generated by two complementary technologies: 1) whole brain EM data sets will be prepared from the same fish that were used for calcium imaging. Respective data sets will be registered to each other, functionally relevant neuronal ensembles will then be identified in the EM stacks and connectivity will be analyzed in these sub-networks via sparse reconstruction. 2) EM based connectivity information will be supplemented by trans-synaptic viral tracing technology. These two technologies for identifying synaptic connections have complementary strengths and weaknesses and are thus ideally suited for combination with in-vivo 2-photon calcium imaging studies. The specific power of this approach is that all three techniques, whole brain calcium imaging, viral tracing and EM reconstruction, can be done in the same animal. Functional, anatomical and behavioral data can then be analyzed in the context of the specific stimuli and quantified behavioral output and subsequently synthesized into a theoretical framework. To that end we will start with quantitative models of simple reflex behaviors, like the optomotor and optokinetic reflex, where the transformation of sensory input to motor output is relatively straightforward and well defined. These elementary models will serve as a scaffold that can be refined and complemented by additional data from structure function studies from fish performing in more sophisticated behavioral assays that involve more complex stimuli, different modalities and plastic changes. As such the process of building such a virtual fish will be an iterative, open ended process that requires continuous and bidirectional exchange of information between the theoretical and experimental groups of the research team.

PROJECT SUMMARY / ABSTRACT Oxidative damage and inflammation accumulate as we age and are prominent components of many genetic and age-related diseases. Alzheimer?s Disease (AD), and many other neurodegenerative diseases, show signs of oxidative damage and inflammation early in the disease progression. It is likely that the accumulation of a protein, the ?-amyloid precursor protein and its cleavage products, stimulates oxidation and inflammation in AD, beyond an organism?s ability to properly regulate the normally protective responses. Nrf2, a transcription factor, is part of an organism?s normal mechanism to activate a response to abnormal proteins. It regulates >100 genes that can be protective of oxidation, inflammation, xenobiotic stress, and autophagy. However, in AD, it appears that the activity of Nrf2 is not upregulated in response to the disease process. One approach is thus to augment the body?s natural response by delivering the Nrf2 gene using viral gene therapy. This approach has been successful in prolonging vision in animal models of neurodegeneration and acute nerve damage. We propose to test the ability of Nrf2 to reduce the neuronal death and behavioral symptoms in two mouse models of AD. The viral vector, AAV, that will be used for delivery directly to the brain, is being tested as this vector is emerging as a viable candidate for clinical applications. Two mouse models of AD with different genetic causes of AD will be tested to determine if the strategy is applicable across the disease spectrum. In addition, several different assays will be used to test different types of responses as indicators of efficacy. If effective, AAV-Nrf2 may prove to be useful in extending neuronal health and function, not only in AD, but perhaps in other neurodegenerative diseases that also exhibit inflammation and oxidaiton.

PROJECT SUMMARY / ABSTRACT AAV vectors have emerged as the leading vector for gene delivery to multiple tissues, proving to be both safe and efficacious in several clinical trials. However, they have not been fully explored for the limits of their safety and efficiency. As the therapeutic benefits of AAV mediated gene therapy will almost always increase with transduction of a greater number of cells, the safe delivery of high viral doses will likely provide a greater benefit to patients. However, high AAV doses in non-human primates (NHPs) and in other animal models have been associated with toxicity. If the mechanisms of these virus-induced problems can be elucidated, it may be possible to avoid them, so that a greater virus dose can be safely delivered. A mouse model allows for studies of mechanism that can then be investigated in larger animals. We have found that AAV vectors can harm ocular cells, in particular, cone photoreceptor cells and the retinal pigment epithelium (RPE) in mice, pigs, and dogs. This toxicity can lead to loss of retinal neurons and the RPE. Toxicity does not correlate with capsid type, the ?cleanliness? of the stock, the gene that is expressed, or the preparation method. Rather, toxicity tracks with dose and genome sequence. We propose to identify the sequences that cause, and/or protect, against this toxicity. We also plan to track the cellular response to these toxic genomic sequences, to identify the cell type(s) in which toxicity is initiated, as well as the cell types which may amplify the results of viral detection. To this end, we will explore the RNA changes in several ocular cell types over time following infection with toxic and non-toxic AAV preparations. We will also follow the mechanism(s) that are triggered by these gene expression changes, and seek ways to block them. These findings will be extended to the brain, using injections of toxic and non-toxic stocks into the cerebral cortex. We will assay changes in brain RNA and cell health, using as probes the changes that we find in the retina. We will also test whether mechanisms that alleviate toxicity in the retina and in the brain.

Photoreceptors (PRs) are highly specialized cells that transduce light energy into useful physiological signals. In vertebrates, the ciliary PRs, rods and cones, initiate vision. The classical definition of rods and cones is based upon the morphology of the outer segment, a membrane rich organelle specialized for the capture of light and its processing so as to signal other retinal neurons. The range of light intensities over which PRs operate is also a fundamental aspect of their definition, with rods operating in dim light and cones in brighter light. Cones initiate our color and high acuity vision, with cones further defined by the wavelength sensitivity of their opsin. Despite the high value we place on our vision, particularly our conemediated high acuity color vision, we know very little about how cones are generated during development. This information would aid in the efforts to replace them when they fail in many types of diseases leading to blindness. Current stem cell methods are very limited in cone production, principally due to a lack of knowledge regarding the normal developmental cues. New strategies are needed for these methods to produce enough cones to make this an effective therapy. In addition to the benefits for therapeutic applications, knowledge of the mechanisms of PR genesis would inform our understanding of the evolution and development of this important class of sensory neurons. Given the high value of vision, PRs are under heavy selective pressure. This has resulted in a wide variety of PR types, with a blurring of rod and cone definition using classically-defined morphology and physiology. In addition, there are dramatic differences among species in the distribution of rods and the different types of cones across the retina. For example, in humans, there is a small high acuity area, the fovea, where cones are the only PR type. An understanding of how rods and cones are determined would allow for an appreciation of some of the mechanisms that have driven these patterning, morphological, and physiological differences across species. The focus of this grant will be the question of how cones choose their fate during development, utilizing genomic data and modern methodologies. The starting point will be to define the transcription regulation of two key genes required for cone genesis, Otx2 and Oc1, and the downstream events under control of Notch1, the earliest known regulator of cone genesis, and then use these data to derive the cone gene regulatory network(s).