The Pnpla2 gene, coding for a lipase termed PEDF-R, mediates PEDF prosurvival and neurotrophic functions in the eye.
We aim ed to investigate the role of PEDF-R in the RPE and retina, using RPE and photoreceptor-specific Pnpla2 conditional knockout (KO) mice. The efficiency of deletion of the Pnpla2 gene mediated by the cyclization recombinase (Cre) in the RPE, was determined by correlating the expression of Pnpla2 with the percentage of Cre-expressing RPE cells. The Cre protein of RPE whole flat mounts from RPE-Pnpla2-KO was detected by immunofluorescence using a Cre-specific antibody. The actin cytoskeleton of the cells was visualized by staining with fluorescein-labelled phalloidin antibody. The expression of Pnpla2 in the RPE from one eye of RPE-Pnpla2-KO and littermate controls was measured by semiquantitative RT-PCR. The contralateral eyes were used to correlate Pnpla2 expression with the percentage of Cre+ cells and phenotypes observed in Pnpla2-deficient RPE, such as lipid accumulation by Oil Red O staining and electron microscopy (EM). The fundi of eyes of RPE-specific Pnpla2 KO mice and corresponding littermates were imaged to visualize accumulation of white dots as a sign of lipid accumulation in the RPE. Longitudinal sections of frozen fixed eyes (cryosections) of wild-type (WT) mice were immunolabeled with different antibodies for PEDF-R and an optimum antibody was identified. To determine the subcellular localization of PEDF-R in the mouse RPE, cryosections from eyes isolated from pigmented and albino mice were co-labeled with antibodies specific for markers of apical (Ezrin) or basal (MTC4) membranes. The distribution of PEDF-R protein in the RPE-Pnpla2-KO mice and control littermates was also assessed in cryosections or RPE whole flat mounts by immunofluorescence. The PEDF-R levels were determined by western blot of RPE and eyecup protein lysates isolated from RPE-Pnpla2-KO and control littermates. The morphology of the retina of RPE-Pnpla2-KO mice, evaluated by hematoxylin and eosin (H&E) staining of retina sections, was not affected. Given that phagocytosis is one of the main functions of the RPE, phagosomes in the RPE whole flat mounts were labelled with anti-rhodopsin specific antibody, detected by immunofluorescence and counted. Phagocytosis was also assessed in the eyecups of the contralateral eye by western blot with anti-rhodopsin and by semi quantitative RT-PCR of phagocytosis-associated genes. To characterize the lipase activity of PEDF-R, we used the state of the art technology of matrix assisted laser desorption imaging mass spectrometry (MALDI-IMS), as it allows for direct analysis of lipids from tissue sections. RPE whole flat mounts from rodents that lack Pnpla2 in the RPE cells were collected and processed for identification and distribution of lipid analytes, and images were constructed. To assess whether PEDF-R plays a role in retina function, electroretinograms of photoreceptor-specific Pnpla2-KO were compared with those of controls. The ultrastructure of the retina and RPE was visualized in photoreceptor-specific Pnpla2-KO by EM and compared to control mice. The levels and distribution of PEDF-R in the mouse retina (KO and control) were determined by western blot of retina lysates and by immunofluorescence on eye cryosections. To further explore the role of Pnpla2 in the entire eye, we begun to generate another mouse strain in which the gene has been constitutively knocked out using CRISPR-CAS-9 technology. We continued studying the effects of PEDF on the rd10 mouse, a model for retina degeneration. We determined the levels of Pnpla2 and Serpinf1 gene expression in the retina and RPE, respectively, in rd10 and WT mice at different ages. The levels of PEDF and PEDF-R proteins in the RPE and retina were also determined in these mice by western blot and ELISA. To explore how essential PEDF is for photoreceptors of rd10 mice, we used mice generated in our lab derived from crosses of rd10 with Serpinf1-null mice. The retinas of rd10/Serpinf1-null and rd10 mice were imaged by Optical Coherence Tomography at different postnatal ages. We compared the thickness of the retina of rd10/Serpinf1-null with those of rd10 and WT at the same age. The imaged eyes were then collected, fixed and embedded in optimal cutting temperature compound for TUNEL staining. The ultrastructure of the RPE and retina of Serpinf1-null mice and control littermates were visualized by EM for comparison. A method for isolation of RNA from mouse RPE was employed and optimized in preparation of transcriptome profiling of RPE from mice depleted of PEDF or PEDF-R. Depletion of PEDF proteins was confirmed by western blotting of serum from Serpinf1-null mice.
| Polato, Federica; Becerra, S Patricia (2016) Pigment Epithelium-Derived Factor, a Protective Factor for Photoreceptors in Vivo. Adv Exp Med Biol 854:699-706 |
| Kenealey, Jason; Subramanian, Preeti; Comitato, Antonella et al. (2015) Small Retinoprotective Peptides Reveal a Receptor Binding Region on Pigment Epithelium-derived Factor. J Biol Chem : |
| Amaral, Juan; Becerra, S Patricia (2010) Effects of human recombinant PEDF protein and PEDF-derived peptide 34-mer on choroidal neovascularization. Invest Ophthalmol Vis Sci 51:1318-26 |
| Samtani, Sonia; Amaral, Juan; Campos, Maria M et al. (2009) Doxycycline-mediated inhibition of choroidal neovascularization. Invest Ophthalmol Vis Sci 50:5098-106 |
