We continued working on the characterization of a PEDF receptor termed PEDF-R. PEDF-R is a patatin-like phospholipase PNPLA2 with high affinity for PEDF and lipase activity. Optimized conditions for in vitro expression of full-length PEDF-R and a C-terminal truncated polypeptide PEDF-R4 using a nanolipoprotein particle were obtained. The recombinant proteins partitioned as soluble proteins, were purified and their phospholipase activities assayed. The phospholipase activities of both recombinant proteins were stimulated by PEDF in a concentration-dependent fashion, suggesting that the C-terminal region of PEDF-R was dispensable for PLA activity and PEDF-mediated stimulation. Several synthetic peptides designed from PEDF-R were assayed for binding to PEDF by ligand blot, surface plasmon resonance and affinity chromatography. Peptides derived from the longest PEDF-R extracellular loop exhibited PEDF-binding affinity. We obtained structural models of PEDF-R and PEDF to envision the ligand-receptor interface. Given that heparin can increase the binding affinity of PEDF for cell-surface receptors and can induce conformational changes in the PEDF protein, we examined its effects on PEDF binding to PEDF-R-derived peptides. PEDF was chemically modified by fluorescein conjugation (Fl-PEDF) and peptide mass fingerprinting showed that critical residues for heparin binding were conjugated with fluorescein. Synthetic peptides derived from the PEDF-binding region of PEDF-R were used to prepare affinity resins for PEDF-binding. PEDF bound to heparin-affinity resin and the binding was competed with soluble heparin. Fl-PEDF lacked heparin binding, but bound the PEDF-R peptides with high affinity. Unmodified PEDF had modest binding affinity for PEDF-R peptides, which increased significantly with additions of soluble heparin. We performed western blots, immunocytochemistry and enzymatic assays and detected PEDF-R in cell membrane fractions of retina R28 and ARPE-19 cells, all of which exhibited PLA activity that was stimulated by PEDF. Commercially available silencing lentivirus was used to silence the PEDF-R gene, while dexamethasone or fatty acids were used to induce it. Expression was followed by RT-PCR and western blots. We performed lipoxygenase activity assays using soybean lipoxygenase V. We discovered that PEDF-R-derived peptides E5b and P1 inhibited the enzymatic activity of the lipoxygenase in vitro in a concentration dependent manner. We investigated the binding of lipoxygenase to peptide-affinity resins. We optimized the binding using a variety of buffer and pH conditions and found that the lipoxygenase specifically bound the peptides. We performed western blots of ARPE-19 cell extracts with antibody to lipoxygenase, and co-immunoprecipitations of PEDF-R and lipoxygenase from ARPE-19 cell extracts. We treated ARPE-19 cells with hydrogen peroxide and TNF alpha to induce oxidative damage. We found that additions of E5b or P1 decreased the number of pyknotic nuclei in oxidative damaged ARPE-19 cells. To begin understanding the mechanisms that confer the multimodal functionality to PEDF, we investigated biological activities of biochemically distinct PEDF protein forms. Recombinant human PEDF was purified. Anion-exchange column chromatography revealed two PEDF protein peaks, PEDF1 eluting with lower ionic strength, and PEDF2 eluting with higher ionic strength. The proteins were analyzed by SDS-PAGE, isoelectric focusing, mass spectroscopy and by limited proteolysis using chymotrypsin, and examined for potential post-translational modifications, i.e., phosphorylation, N- and O-glycosylation. We used tumor and retina cell lines known to respond to stimulation by PEDF. Cell viability assays were performed. Real-time impedance of cells was followed with a microelectronic system. In collaboration with the laboratory of Dr. Larry Benowitz (Childrens Hospital, Boston/Harvard U), the proteins were also assayed for axon growth and survival of retinal ganglion cells (RGCs). We found that PEDF2 was potent in lowering cell viability and impedance in all tumor cell lines tested, while PEDF1 affected tumor cell viability only modestly. Both versions protected retina cells against death due to serum-starvation and promoted RGC axon growth. We completed the studies on the interactions of PEDF with ectopic F1 ATP synthase/ATPase. Protein fingerprinting suggested a match of a 60 kDa PEDF-binding protein in bovine retina with Bos taurus F1-ATP synthase beta-subunit. We examined the direct binding of PEDF to F1 by size-exclusion ultrafiltration and surface plasmon resonance. Recombinant human PEDF formed a complex with recombinant yeast F1. They interacted specifically and reversibly with high binding affinity, in agreement with PEDF affinities for endothelial cell surfaces. PEDF blocked interactions between F1 and angiostatin, another antiangiogenic factor, suggesting overlapping PEDF-binding and angiostatin-binding sites on F1. Endothelial cell surfaces contained PEDF-binding proteins of 60 kDa, which were captured specifically by antibodies to F1 beta-subunit. Extracellular ATP synthesis activity of endothelial cells was examined in the presence of PEDF. PEDF significantly decreased the amount of extracellular ATP produced by endothelial cells, in agreement with direct interactions between cell-surface ATP synthase and PEDF. We investigated the presence of F1/F0 ATP synthase on cell surfaces of breast and prostate tumor cell lines using biochemical and immunocytochemistry methodologies. The effects of PEDF and its derived peptides 34-mer and 44-mer were evaluated using a real-time cell impedence microsensor system and cell viability assays. We assayed extracellular ATP production and found that PEDF and 34-mer decreased cell viability and inhibited extracellular ATP synthesis activity. We also completed the studies on structure-function of PEDF as it relates to inhibition of neovascularization. The effects of PEDF protein and PEDF-derived peptide 34-mer on vessel sprouting and choroidal neovasculariztion (CNV) following subconjunctival administration were evaluated. Full-length PEDF, cleaved PEDF at its serpin exposed loop or peptide 34-mer exhibited ex-vivo antiangiogenic activity, while peptide 44-mer was inefficient. PEDF immunostaining around CNV lesions diminished after laser injury. Subconjunctival administration of PEDF or 34-mer decreased CNV lesion volumes, while those of 44-mer were similar to vehicle injections. PEDF decreased fully-developed CNV complex volumes relative to vehicle injections. A study on the evaluation of potential matrixes for ocular protein delivery was also completed. Fluorescein conjugated ovalbumin, a serpin closely related to PEDF, was mixed with a biodegradable, biocompatible thermal-gel and injected in the subconjunctival space of rats in collaboration with the laboratory of Dr. Robert Lutz (DDKR, NIBIB). Protein diffusion toward the choroid/RPE and retina was assessed. We also completed studies on the effects of doxycycline on CNV. PEDF is a novel substrate of MMP-2/-9 that can inhibit CNV. Given that doxycycline, a broad-spectrum antibiotic with certain antiangiogenic properties, can inhibit MMPs/gelatinases, we investigated the effects of doxycycline on CNV and on regulation of MMP-2 and -9 and PEDF. CNV complex lesion volumes decreased with doxycycline in a dose-response relationship. Doxycycline elevated PEDF levels in plasma and did not affect the active and pro-enzymes MMP-2 and MMP-9 levels. However, the in vitro enzymatic activities of purified MMP-2 and MMP-9 declined significantly with doxycycline. MMP-2, MMP-9, and gelatinolytic activities in situ increased early in CNV lesion development. Doxycycline inhibited gelatinolytic activities in CNV lesions.
Showing the most recent 10 out of 27 publications
