Diseases that cause photoreceptor degeneration such as age-related macular degeneration (AMD) and retinitis pigmentosa (RP) affect millions of people and lead to blindness. The degeneration of photoreceptor cells in AMD or cone photoreceptors in RP is caused by stress in their environment and/or lack of trophic support, not by direct genetic mutation targeting of the photoreceptor cells. Therefore, providing effective trophic support can delay or prevent photoreceptor degeneration. It has been known for more than 20 years that natural trophic factors exist in the retina that protect photoreceptor cells, but these factors have never been identified. Recent gene therapy studies have revealed that targeting genetic mutations is not sufficient and that there is an urgent need to identify trophic factors that can maintain photoreceptor survival in the treatment of photoreceptor degeneration. Identification and mechanistic study of trophic factors naturally present in the retina will lead to a better understanding of the photoreceptor protection mechanism and help to develop new therapeutic strategies. My thesis project aims to identify new natural trophic factors that protect photoreceptor cells and to test them using in vitro and in vivo models.

We found that natural trophic factors in the retina strongly protect photoreceptor cells from retinal-mediated light damage or chemically induced oxidative damage. Using classic techniques combined with advanced mass spectrometry and genome-wide database searching, we identified several candidate photoreceptor protective factors. These factors effectively protect primary cone photoreceptor cells from light damage and oxidative damage in vitro. We cloned candidate genes into lentiviral vectors and delivered them to the retina and RPE (retinal pigment epithelium) in vivo by subretinal injection. Using two mouse models of retinal degeneration: a light-induced retina degeneration model and a genetic retina degeneration model, we found that one factor almost completely protect retina from light-induced damage in vivo. In summary, we have identified a novel factor that can prevent photoreceptor cell degeneration in vitro and in vivo. This factor has potential therapeutic values in treating retina degenerative diseases.

Pigment Epithelium-Derived Factor (PEDF) is a natural factor with surprisingly diverse therapeutic functions. Since its identification more than 20 years ago, PEDF has been recognized as a neurotrophic factor, a stem cell niche factor, an anti-inflammatory factor, an anti-angiogenic factor, a tumor inhibitor, and a protein with declined expression in aging. This secreted factor has been demonstrated to have diverse therapeutic value in inhibiting the pathogenesis of several major diseases such as diverse cancer types including melanoma, neuroblastoma, osteosarcoma, hepatoblastoma, Lewis lung carcinoma, chondrosarcoma, gastric carcinoma, glioma, Wilm's tumor, prostate cancer, and pancreatic cancer and several major blinding diseases including ischemia-induced retinopathy, diabetic retinopathy, retinitis pigmentosa, and age-related macular degeneration. Although this multifunctional factor has drawn increasing attention, its therapeutic potential is greatly hampered by the lack of knowledge on PEDF’s signaling mechanism. Despite considerable efforts in academia and industry, the cell-surface receptors that transduce PEDF signal eluded identification since the 1990s.

To transform the high therapeutic value of PEDF to potent and efficient therapeutic agents to treat human diseases, we aimed to achieve three general goals. The first is to identify and characterize the cell-surface receptors for PEDF. The second is to investigate the mechanism of PEDF signaling through these receptors. The third is to screen small molecules that mimic PEDF functions by targeting its receptors. Receptors are ideal therapeutic targets due to their specificity in changing cell behaviors and small molecule drugs targeting receptors account for the largest fraction of drugs used clinically in treating human diseases.

After many years of effort and trying many strategies, we identified the long-sought PEDF cell-surface receptors as PLXDC1 and PLXDC2, two single transmembrane domain proteins that do not belong to well-known receptor families. We found that these two homologous transmembrane receptors not only confer cell-surface binding to PEDF but also mediate diverse PEDF functions such as promoting IL-10 secretion in macrophage, inducing endothelial cell death and protecting neuronal cell in a cell-type specific manner. We also found unique mechanisms of interaction between PEDF and its receptors, such as receptor-induced PEDF dimerization through a disulfide bond. To identify compounds that specifically target these receptors, we have developed a novel cell-based, fluorescence-based and high-throughput screening strategy that color-codes receptor-expressing cells. Using this strategy, we have identified potent lead compounds that mimic PEDF actions such as causing cell death in a receptor-specific manner. These compounds are potential first-in-class drugs in treating diseases.

In summary, identification of PEDF receptors allows further investigation of the molecular mechanism of PEDF signaling and provides novel therapeutic targets. By developing a novel cell-based screening technique, we have identified chemical compounds that specifically and potently target the PEDF receptors and mimic PEDF actions.

Pigment epithelial derived factor (PEDF) is a natural ligand with a wide range of therapeutic functions. It was first characterized as a neurotrophic factor secreted by the retinal pigment epithelium (RPE) but later found to be a potent anti-angiogenic factor that acts independent of vascular endothelial growth factor (VEGF). As a result, PEDF garnered significant attention from industry and academia but progress in leveraging PEDF's ability was hampered by the lack of insight in its signaling mechanism. After 20 years from its initial identification, the cell surface receptor responsible for PEDF's anti-angiogenic effect was found to be tumor endothelial marker 7 (TEM7) or plexin-domain containing 1 (PLXDC1). This receptor was found to be highly expressed in the endothelium of a wide range of solid tumors (e.g. pancreatic and colon cancer) and vision disease (e.g. diabetic retinopathy and age-related macular degeneration).

To transform the power of PEDF into an anti-angiogenic therapy that can treat human disease, we focused on two main goals. The first was to elucidate how PLXDC1 receptor activation causes endothelial cell death. The second is to develop antibodies that can diagnose endogenous PLXDC1 expression and that can kill pathogenic blood vessels through PLXDC1 receptor activation. Receptors are ideal therapeutic targets due to their accessibility and ability to regulate cell outcomes. Although no receptor activating antibody is currently used clinically, antibodies are an ideal drug as they are highly specific with minimal off-target toxicity.

With the use of a small molecule PLXDC1 agonist, we found that the mechanism of PLXDC1-mediated endothelial cell death is consistent with anoikis, a detachment induced cell death pathway. We observed that a small molecule PLXDC1 agonist inhibited cell adhesion, which caused cell detachment, re-localization of beta-catenin, de-phosphorylation of focal adhesion kinase, and ultimately cell death. Independently, integrin beta-1 (ITGB1), a transmembrane protein involved in attaching endothelial cells to the extracellular matrix, was identified as a binding partner to PLXDC1 by unbiased affinity purification/mass spectrometry. We also discovered that ITGB1 not only bound to PLXDC1, but at higher levels in the activated form of PLXDC1. This was functionally confirmed by enhanced PEDF mediated receptor activation with co-expression of ITGB1 and inhibition of PEDF induced endothelial cell death with knockdown of ITGB1.

Separately, we produced monoclonal and fragment antigen binding (Fab) antibodies that detect and kill pathogenic blood vessels through hybridoma and phage display strategies. Amongst the 96 mouse monoclonal antibodies, we identified three antibodies that reliably detected PLXDC1 in a wide range of human cancers. We then screened over 200 billion human Fabs for antibodies that could preferentially bind the activated receptor. Amongst those, we developed 30 antibodies that can inhibited cell adhesion, induced cell detachment, and caused endothelial cell death in vitro and ex vivo. Lastly, we demonstrate that a few of those antibodies showed ability to inhibit laser induced choroidal neovascularization progression in vivo with intravitreal injections.

In summary, we have identified the mechanism of PLXDC1 mediated endothelial cell death and developed diagnostic and therapeutic antibodies targeting PLXDC1. Combined this provides further insight into how we can translate PEDF's natural anti-angiogenic potential into a therapy to treat human cancer and vision disease.

In this work, we study singular solutions and pattern formation in aggregation equations and more general active scalar problems.

We derive a generalization of the Birkhoff-Rott equation to the case of active scalar problems with both gradient and divergence free structures. We present numerical simulations of this model demonstrating how the gradient part and the divergence free part of K influence each other and cause some nonlinear effects. Examples include superfluids, classical fluids and swarming models.

The rest of this thesis focuses on aggregation models with gradient flow structure. The discrete version of the continuum aggregation equation is the kinematic equation. For both discrete and continuum versions, we use linear stability analysis of a ring equilibrium to classify the morphology of patterns in two dimensions. Conditions are identified that assure the linear well-posedness of the ring. In addition, weakly nonlinear theory and numerical simulations demonstrate how a ring can bifurcate to more complex equilibria. Moreover, linear stability analysis of clusters equilibrium patterns are also investigated in both two-dimensional and higher-dimensional cases.

We then apply our stability results of ring patterns and clusters patterns to a family of exact collapsing similarity solutions to the aggregation equation with pairwise potential U(r) = r^ν/ν. It was previously observed that radially symmetric solutions are attracted to a self-similar collapsing shell profile in infinite time for ν > 2 in all dimensions. The stability analysis for ring patterns and clusters patterns shows that the collapsing shell solution is stable for 2 < ν < 4, while always unstable and destabilizes into clusters that form a simplex for ν > 4. This holds in all spatial dimensions.