We completed studies on the characterization of the PEDF area that interacts with a distinct ectodomain on PEDF-R to promote photoreceptor survival. Molecular docking studies suggested that the ligand binding site of PEDF-R interacts with the neurotrophic region of PEDF (44mer, positions 78-121). Binding assays demonstrated that PEDF-R bound the 44mer peptide. Peptide P1 from the PEDF-R ectodomain had affinity for the 44mer and a shorter fragment within it, 17mer (positions 98-114). Single residue substitutions to alanine along the 17mer sequence were designed and tested for binding and biological activity. Altered 17merR99A did not bind to the P1 peptide, while 17merH105A had higher affinity than the unmodified 17mer. Peptides 17mer, 17merH105A and 44mer, exhibited cytoprotective effects in cultured retina R28 cells. Intravitreal injections of these peptides and PEDF in the rd1 mouse model of retinal degeneration decreased the numbers of dying photoreceptors, 17merH105A being most effective (in collaboration with Dr. V. Marigo). The blocking peptide P1 hindered their protective effects both in retina cells and in vivo. In addition to demonstrating that the region comprised of positions 98-114 of PEDF contains critical residues for PEDF-R interaction that mediates survival effects, the findings reveal distinct small PEDF fragments with neurotrophic effects on photoreceptors. The cytoprotective activity of PEDF peptides 17mer and 17merH105A was evaluated in the presence of selective and competitive catalytic inhibitor of PEDF-R, atglistatin. No decrease in the TUNEL positive cells was observed with PEDF or peptides in the presence of atglistatin, demonstrating that PLA activity of PEDF-R is critical for PEDF peptide mediated antiapoptotic effect. Live R28 cells were labeled with FITC-PEDF peptides in increasing concentrations. Cells were lysed in RIPA buffer and samples were collected and the fluorescence intensity was measured using SpectraMAX plus. Expression vectors for full-length human PEDF with single alterations at R99A and H105A and a FLAG-tag at their N-end were constructed and transfected into BHK cells. The recombinant proteins were purified from conditioned media by a two-step cation- and anion-exchange column chromatography protocol. PEDF was followed by SDS-PAGE, western blots and ELISA, and purity reached >90% pure. The purified proteins were analyzed by circular dichroism spectroscopy to compare secondary structure and the mutant proteins share similar fold. They were tested for binding to the PEDF-R peptides P1 and E5b, and for biological activity in cell cultures and in vivo. FLAG-PEDFH105A had cytoprotective activity on R28 cells and on photoreceptors of the rd1 mouse model (collaboration with Dr. V Marigo), while PEDFR99A did not. We continued the studies on a novel inhibitor of lipoxygenase discovered from a PEDF-R region, namely P1 peptide, which we showed previously binds to 5-LOX and protects ARPE-19 cells from cell death due to oxidative stress. Lipoxygenases are enzymes responsible for the metabolism of arachidonic acid and other polyunsaturated fatty acids, thereby contributing to the generation of reactive oxygen species under oxidative stress. The effect of PNPLA2 genetic manipulation under oxidative stress was evaluated. Commercially purchased human PNPLA2 and 5-LOX siRNAs were tested in ARPE-19 cells. RT-PCR and western blots were performed to confirm decrease in the expression of PNPLA2 and 5-LOX genes and PEDF-R and 5-LOX proteins in transfected cells. An expression vector designed in our lab was used to transfect ARPE-19 and overexpression was confirmed by RT-PCR and westerns of ARPE-19 cells. Relative cell numbers were quantified using two different biomarkers for live and dead cells, intracellular ATP content and dead cell protease activity in cells subject to oxidative stress. The results from above experiments showed that both PEDF-R and peptide P1 increased the viability and decreased the number of dead cells under oxidative stress. Relative dead cell numbers were also quantitated in oxidative stress assay in the presence of scrambled peptide P1, obtained by random rearrangement of peptide P1. The effect of chemically synthesized small peptides spanning the PEDF-R ectodomain Leu159-Met325 on LOX activity was assessed. The scrambled peptide P1 was used as a control and P1 peptide was the only one with inhibitory activity against LOX. Binding of the mammalian full-length PEDF-R to 5-LOX was tested using pull-down experiments with differentially tagged proteins. Full-length PEDF-R tagged with N-terminal V5 tag and 5-LOX with N-terminal FLAG tag were co-expressed in ARPE-19 cells. Immunoprecipitation with FLAG antibody followed by probing with V5 antibody and vice versa were performed. In situ proximity ligation assay was performed to visualize protein-protein interactions in fixed cells. Given that LTB4 is a product of the 5-LOX enzymatic activity, the concentration of the secreted LTB4 was measured in the media from cells (ARPE-19 and pig RPE) by ELISA as a way of determining the 5-LOX activity in cells. We started to study another member of the serpin superfamily, protease nexin-1 (PN-1). PN-1 is encoded by the SERPINE2 gene and has neurotrophic effects in the brain and inhibits angiogenesis in the retina, resembling the activities of PEDF. The principal objective of this study is to compare PN-1 and PEDF as potential neuroprotectors for the retina. Murine RPE and retinal RNA was obtained from dissected C57BL/6N Crb1rd8 mouse eyes. Bovine protein lysates were obtained from bovine eye dissection. Serpine2 transcript and PN-1 protein levels were evaluated by RT-PCR and Western blotting, respectively. Recombinant human PN-1 and PEDF proteins were used. Biological activity in rat-derived retinal precursor (R28) cells was assessed with TUNEL cell death assays. PEDF-R was synthesized in vitro. Binding was evaluated through pulldown assays with full-length His6 Tag-PEDF-R and peptide affinity chromatography to peptide E5b derived from PEDF-R. Our results revealed that PN-1 is analogous in primary and tertiary structure to PEDF, and contains a region that shares homology with the neurotrophic active region of PEDF. SERPINE2 mRNA was detected in ARPE-19 cells, native human fetal RPE and retina, and murine RPE and retina. PN-1 protein was detected in ARPE-19 lysate, and extracted from the extracellular matrix (ECM), particularly bovine interphotoreceptor matrix (IPM), with increased NaCl. Low concentrations of PN-1 reduced TUNEL-positive nuclei in serum-starved R28 cells relative to untreated cells. Recessive null mutations in SERPINF1 cause type VI osteogenesis imperfecta, a heritable bone dysplasia characterized by high susceptibility to fracture, growth deficiency and defects in bone mineralization. Serum levels of PEDF are significantly decreased in type VI OI patients. Dominant mutations in IFITM5, encoding BRIL (bone-restricted ifitm-like protein), cause type V osteogenesis imperfecta. We continued the collaborative study investigating the role of PEDF in type V OI with clinical characteristics of type VI OI. RT-PCR and westerns of ARPE-19 cells, dissected mouse RPE and retina were performed and the results showed that BRIL is present in the RPE and the retina. Broncopulmonary dysplasia is a chronic lung disease of preterm infants characterized by arrested microvascularization and alveolarization. We completed the collaborative study evaluating the role of PEDF in lung vascular development in neonatal hypoxia. It was found that PEDF levels increase in hyperoxia compared to room air-exposed lungs. The levels of PEDF were positively correlated with reduced vasculogenesis and alveolarization in neonatal hyperoxia, implying that PEDF mediates impaired lung vascular development in neonatal hyperoxia.
Showing the most recent 10 out of 27 publications
